HK40046552B - Antibodies specific for cd3 and uses thereof - Google Patents

Abstract

The present invention provides novel antibodies that specifically bind to CD3 and uses thereof. The invention also provides therapeutic and diagnostic methods utilizing the antibodies and antigen-binding fragments provided herein.

Description

CD3-specific antibodies and their uses

sequence list

This application was filed electronically via EFS-Web and includes a sequence list submitted electronically in .txt format. The .txt file contains a sequence list titled "PC72375-PRV2_SequenceListing_ST25_05132019.txt", created on May 13, 2019, and is 100 KB in size. The sequence list contained in this .txt file is part of this specification and is incorporated herein by reference in its entirety.

Invention Field

This invention relates to CD3-specific antibodies and their antigen-binding fragments, as well as methods of using them. The invention also relates to multispecific antibodies that bind to both CD3 and target antigens, and methods of using them.

Background of the Invention

Differentiation cluster 3 (CD3) is a homodimeric or heterodimeric antigen expressed on T cells that associates with the T cell receptor complex (TCR) and is required for T cell activation. T cell activation is a complex phenomenon that depends on the involvement of multiple cell surface molecules expressed on a population of reactive T cells. Functional CD3 is formed by the association of two of four different chains: ε, ζ, δ, and γ. For example, the antigen-specific T cell receptor (TCR) is composed of a disulfide-linked heterodimer containing two clonal, monolithic membrane glycoprotein chains, alpha and beta (α and β), or gamma and delta (γ and δ), non-covalently associated with a complex of a low molecular weight invariant protein (usually designated CD3). CD3 dimer arrangements also include γ/ε, δ/ε, and ζ/ζ.

Antibodies targeting CD3 have been shown to accumulate CD3 on T cells, thereby inducing T cell activation in a manner similar to the binding of MHC molecules loaded with peptides to the TCR. Therefore, CD3 antibodies can be used for therapeutic purposes involving T cell activation.

Invention Overview

This invention provides antibodies that bind to CD3 and compositions (such as pharmaceutical compositions) containing one or more CD3 antibodies. Antibodies according to this aspect of the invention are particularly useful for targeting CD3-expressing T cells and for stimulating T cell activation, for example, in cases where T cell-mediated killing is beneficial or desired. The CD3 antibodies of this invention may be included as part of a multispecific antibody (e.g., a bispecific antibody) that directs CD3-mediated T cell activation to specific cell types, such as tumor cells or infectious pathogens.

In one aspect, the present invention provides an antibody that specifically binds to CD3, wherein the antibody comprises: (a) a heavy chain variable (VH) region comprising VH complementarity-determining region 1 (VH CDR1), VH complementarity-determining region 2 (VH CDR2), and VH complementarity-determining region 3 (VH CDR3) of the VH sequence shown in SEQ ID NO: 2, 4, 5, 7, 10, 12, or 35; and/or (b) a light chain variable (VL) region comprising VL complementarity-determining region 1 (VL CDR1), VL complementarity-determining region 2 (VL CDR2), and VL complementarity-determining region 3 (VL CDR3) of the VL sequence shown in SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87, or 89.

In one embodiment of this aspect, the CD3 antibody comprises: (a) a VH region comprising (i) a VH CDR1 comprising the sequence of SEQ ID NO: 13, 14 or 15; (ii) a VH CDR2 comprising the sequence of SEQ ID NO: 16, 17, 19, 20, 21, 22, 23 or 24; and (iii) a VH CDR3 comprising the sequence of SEQ ID NO: 18 or 25; and/or (b) a VL region comprising (i) a VL CDR1 comprising the sequence of SEQ ID NO: 26, 29, 30, 32, 91 or 92; (ii) a VL CDR2 comprising the sequence of SEQ ID NO: 27 or 31; and (iii) a VL CDR3 comprising the sequence of SEQ ID NO: 28 or 33.

In some such embodiments, the present invention provides an antibody wherein: (a) the VH region contains a sequence of SEQ ID NO: 2, 4, 5, 7, 10, 12 or 35; and/or (b) the VL region contains a sequence of SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87 or 89.

In other embodiments, the antibody comprises a VL/VH amino acid sequence selected from the following: SEQ ID NO: 1 and 2; SEQ ID NO: 3 and 2; SEQ ID NO: 3 and 4; SEQ ID NO: 3 and 5; SEQ ID NO: 6 and 7; SEQ ID NO: 8 and 7; SEQ ID NO: 6 and 4; SEQ ID NO: 8 and 4; SEQ ID NO: 9 and 10; SEQ ID NO: 9 and 7; SEQ ID NO: 11 and 12; SEQ ID NO: 34 and 35; SEQ ID NO: 9 and 4; SEQ ID NO: 87 and 4; and SEQ ID NO: 89 and 4.

In some of the aforementioned embodiments, the antibody is selected from: Fab, Fab fragment, F(ab) ₂ fragment, Fv fragment, single-chain Fv fragment, disulfide-stabilized Fv fragment, single-chain antibody, monoclonal antibody, chimeric antibody, bispecific antibody, trispecific antibody, multispecific antibody, bispecific heterodimeric biantibody, bispecific heterodimeric IgG, polyclonal antibody, labeled antibody, humanized antibody, human antibody and fragment thereof.

In one specific embodiment, the antibody is a bispecific antibody. In some such embodiments, the bispecific antibody specifically binds to tumor antigens.

In some embodiments, the antibody of the present invention further comprises a human or humanized VH framework and a human or humanized VL framework. In some such embodiments, the antibody is a humanized antibody.

The present invention further provides pharmaceutical compositions comprising a CD3 antibody as further disclosed herein and a pharmaceutically acceptable carrier.

In another aspect, the present invention provides a method for modulating a T-cell-mediated immune response in a subject with such need, comprising administering an effective amount of an antibody or pharmaceutical composition as disclosed herein to the subject such that the immune response in the subject is modulated.

In one aspect, the present invention provides a method for inhibiting the growth of tumor cells in a subject, comprising administering to the subject, in an amount that effectively inhibits the growth of tumor cells, a bispecific antibody, a trispecific antibody, a multispecific antibody, a bispecific heterodimeric antibody, or a bispecific heterodimeric IgG as disclosed herein.

In some such embodiments, the tumor cells are cancer cells. In specific embodiments, the cancer is selected from: breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer, bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and cancers of the digestive system.

In one embodiment, the cancers of the digestive system are selected from: esophageal cancer, gastric cancer, small bowel cancer, colon cancer, rectal cancer, anal cancer, liver cancer, gallbladder cancer, appendix cancer, bile duct cancer, and pancreatic cancer.

In one embodiment, the present invention provides an antibody or a pharmaceutical composition comprising the antibody for use in a therapy.

In another embodiment, the present invention provides the use of the antibody of the present invention in the preparation of a medicament for use in a therapy. In some such embodiments, the therapy comprises activating a cytolytic T-cell response.

In one aspect, the present invention provides a polynucleotide comprising a nucleotide sequence encoding an antibody as disclosed herein. In another aspect, the present invention provides a vector comprising this polynucleotide.

In another aspect, the present invention provides host cells comprising vectors as disclosed herein. In some such embodiments, the host cells recombinantly produce antibodies as disclosed herein. In specific embodiments, the host cells are selected from bacterial cell lines, mammalian cell lines, insect cell lines, and yeast cell lines. In one specific embodiment, the mammalian cell line is a CHO cell line.

In one aspect, the present invention provides a method for generating antibodies as disclosed herein, comprising: culturing host cells under conditions that result in the generation of antibodies as disclosed herein, and purifying said antibodies from the culture supernatant.

In another aspect, the present invention provides the use of CD3 antibodies, polynucleotides, vectors or host cells as disclosed herein in the preparation of medicaments for treating diseases.

In one aspect, the present invention provides an antibody or pharmaceutical composition for use in a method of modulating a T-cell-mediated immune response in a subject with such a need. In a specific embodiment, the present invention provides an antibody or pharmaceutical composition for use in a method of inhibiting the growth of tumor cells in a subject.

In another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, optionally wherein said cancer is selected from: breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer, bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and cancers of the digestive system.

In yet another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, wherein a T-cell-mediated immune response is modulated or wherein the growth of tumor cells is inhibited.

Other embodiments will become apparent from the detailed summary. When aspects or embodiments of the invention are described in terms of the Markush group or other alternative groupings, the invention covers not only the entire group listed as a whole, but also each member of that group individually and all possible subgroups of the main group, as well as the main group in which one or more members are not present. The invention also contemplates the explicit exclusion of any one or more members of the claimed invention.

Brief description of the figure/attachment

Figure 1 provides a schematic diagram of the architecture of a bispecific antibody having a first heterodimer promoting domain and a second heterodimer promoting domain, the bispecific antibody comprising an Fc chain optimized for association via mortar association.

Figure 2 depicts the results of the differential scanning calorimetry temperature recording, showing the heat capacity of the GUCY2c-H2B4 and GUCY2c-2B5 bispecific antibodies as a function of temperature.

Figure 3 depicts the binding of the bispecific antibodies GUCY2c-H2B4 (GUCY2c-0247) and GUCY2c-2B5 (GUCY2c-0250) to naïve human T cells as determined by flow cytometry.

Figure 4 depicts the results of in vitro cytotoxicity mediated by the bispecific antibodies GUCY2c-H2B4 (GUCY2c-0247) and GUCY2c-2B5 (GUCY2c-0250) using cell survival data.

Figure 5 depicts the potential energy surface (PES) of the H2B4 Fv region from an X-ray crystal structure. This includes three views highlighting the VH region, the VH/VL interface, and the VL region. The CDR (Crystal Radiata Diffraction) with excess positive charge is indicated for reference. Dark black represents positive charge and white represents negative charge.

Figure 6 depicts the spatial aggregation tendency (SAP) surface of the Fv region of H2B4. Regions with concentrated hydrophobic residues are marked.

Figure 7 depicts the heterogeneity among bispecific antibodies containing CD3 antibody-binding domains H2B4, 2B5, and 2B5v6 using capillary gel electrophoresis.

Figure 8 depicts the temperature record using differential scanning calorimetry (DSC), showing the heat capacity of the GUCY2c-1608 (GUCY2c-2B5v6) bispecific antibody as a function of temperature.

Figure 9 depicts the binding of the GUCY2c-1608 (GUCY2c-2B5v6) bispecific antibody to naive T cells.

Figure 10 depicts cell survival data for in vitro cytotoxicity mediated by the bispecific antibody GUCY2c-1608 (GUCY2c-2B5v6).

Figures 11A to 11F depict in vitro cytokine release measurements of the GUCY2c-1608 (GUCY2c-2B5v6) bispecific antibody. The bispecific antibody recruited naïve human T cells to GUCY2c-expressing T84 cells to induce in vitro cytokine release. Luminex-based assays showed upregulation of human IFN-γ (Figure 11A), IL10 (Figure 11B), IL2 (Figure 11C), IL4 (Figure 11D), IL6 (Figure 11E), and TNF-α (Figure 11F).

Figure 12 illustrates the dose-dependent tumor growth inhibition of the bispecific antibody GUCY2c-1608 (GUCY2c-2B5v6) in the patient-derived xenograft PDX-CRX-11201 of colorectal cancer in the adoptive metastasis system.

Figure 13 illustrates the dose-dependent tumor growth inhibition of the bispecific antibody GUCY2C-1608 (GUCY2C-2B5v6) in the xenograft LS1034 colorectal cancer cell line in the adoptive metastasis system.

Figure 14A depicts the results of cell viability data, showing the in vitro cytotoxicity mediated by the FLT3-2B5v6 bispecific antibody using Flt3-expressing MV4-11 cells at different effector-to-target (E:T) ratios. Figure 14B depicts the results of cell viability data, showing the in vitro cytotoxicity mediated by the FLT3-2B5v6 bispecific antibody using Flt-3-expressing EOL-1 cells at different effector-to-target (E:T) ratios.

Figure 15 depicts the binding of the low-affinity 2B5v6 variant to Jurkat cells expressing CD3.

Figure 16 depicts a graph summarizing the findings, showing the cytotoxicity mediated by anti-GUCY2C bispecific T cells with anti-CD3 variants.

Detailed Explanation

The present invention disclosed herein provides antibodies that specifically bind to CD3 (e.g., human CD3), including full-length antibodies or antigen-binding fragments thereof. The invention also provides polynucleotides encoding these antibodies, compositions comprising these antibodies, and methods for preparing and using these antibodies. The invention further provides methods for treating conditions using antibodies as described herein.

General Technology

Unless otherwise indicated, the practice of this invention will employ conventional techniques of molecular biology (including recombinant techniques), microbiology, cell biology, biochemistry, and immunology, which are within the scope of the art. Such techniques are fully explained in the literature such as *Molecular Cloning: A Laboratory Manual, 2nd Edition* (Sambrook et al., 1989); *Cold Spring Harbor Press*; *Oligonucleotide Synthesis* (M.J. Gait, ed., 1984); *Methods in Molecular Biology*, *Humana Press*; *Cell Biology: A Laboratory Notebook* (J.E. Cellis, ed., 1998); *Academic Press*; *Animal Cell Culture* (R.I. Freshney, ed., 1987); *Introduction to Cell and Tissue Culture* (J.P.). Mather and P.E. Roberts, 1998) Plenum Press; Cell and Tissue Culture: LaboratoryProcedures (A.Doyle, J.B. Griffiths, and D.G. Newell, eds., 1993-1998) J. Wileyand Sons; Methods in Enzymolog y (Academic Press, Inc.); Handbook of Experimental Immunology (D.M. Weir and C.C. Blackwell, Eds.); Gene Transfer Vectors for Mammalian Cells (J.M. Miller and M.P. Calos, Eds., 1987); Current Protoc Mullis in Molecular Biology (F.M. Ausubel et al., eds., 1987); PCR: The Polymerase Chain Reaction (Mullis et al., eds., 1994); Current Protocols in Immunology (J.E. Colligan et al., eds., 1991); Short Protocols in Molecular Biology (Wiley and Sons, 1999); Immunobiology (C.A. Janeway and P. Travers, 1997); Antibodies (P. Finch, 1997); Antibodies: a practical approach (D. Catty., ed., IRL Press, 19...). 88-1989); Monoclonal Antibodies: A Practical Approach (P. Shepherd and C. Dean, eds., Oxford University Press, 2000); Using Antibodies: A Laboratory Manual (E. Harlow and D. Lane (Cold Spring Harbor Laboratory Press, 1999); The Antibodies (M. Zanetti and J.D. Capra, eds., Harwood Academic Publishers, 1995). In some cases, terms having their usual meaning are defined herein for clarity and/or for the purpose of preparing references, and the inclusion of such definitions herein is not necessarily construed as indicating a material difference from definitions commonly understood in the art.

The invention will now be described in detail by reference using the following definitions and examples. All patents and publications mentioned herein (including all sequences disclosed in such patents and publications) are expressly incorporated by reference.

definition

Generally, unless otherwise defined, all terms, symbols, and other scientific or terminological terms used herein are intended to have the meaning commonly understood by one of ordinary skill in the art to which this invention relates. For example, the term "and/or" as used herein, such as in phrases like "A and/or B," is intended to include both A and B; A or B; A (alone); and B (alone). Similarly, the term "and/or" as used herein, such as in phrases like "A, B, and/or C," is intended to cover each of the following embodiments: A, B, and C; A, B, or C; A or C; A or B; B or C; A and C; A and B; B and C; A (alone); B (alone); and C (alone).

As used herein, unless the context clearly indicates otherwise, the singular forms “a”, “an” and “the” include their respective multiple indicators.

As used in this article, the numerical range includes the numerical values that define the range.

The terms “about,” “approximately,” etc., when preceding a list of numerical values or ranges, independently refer to each individual value in that list or range as if each individual value immediately followed the term. The term implies that the value referred to by the term is the same as, close to, or similar to. For example, in some embodiments, “about” or “approximately” for a particular value may indicate 99%, 95%, or 90% of that value. As an example, the expression “about 100” includes 99 and 101 and all values between them (e.g., 99.1, 99.2, 99.3, 99.4, 99.5, etc.). As another example, if the temperature is 70°C, “about” or “approximately” 70°C may be equal to 69°C, 66°C, or 63°C. It should be understood that these are merely examples.

As used herein, nucleic acids are written from left to right in a 5' to 3' orientation; amino acid sequences are written from left to right in an amino-to-carboxyl orientation. For definitions and terminology in this art, practitioners particularly refer to Sambrook et al., 1989 and Ausubel FM et al., 1993. It should be understood that the invention is not limited to the specific methods, schemes, and reagents described, as these are subject to variation.

The terms “polypeptide,” “oligopeptide,” “peptide,” and “protein” are used interchangeably herein to refer to an amino acid chain of any length, preferably relatively short (e.g., 10-100 amino acids). This chain may be linear or branched, may contain modified amino acids, and/or may be interrupted by non-amino acid components. The term also covers amino acid chains that have been modified naturally or through intervention; for example, disulfide bond formation, glycosylation, esterification, acetylation, phosphorylation, or any other operation or modification, such as conjugation with a labeled component. This definition also includes, for example, polypeptides containing one or more amino acid analogs (including, for example, non-natural amino acids) and other modifications known in the art. It should be understood that polypeptides may exist as single chains or associated chains. Preferably, mammalian polypeptides (polypeptides originally derived from mammalian organisms) are used, more preferably those secreted directly into the culture medium.

An "antibody" is an immunoglobulin molecule that specifically binds to a target, such as carbohydrates, polynucleotides, lipids, peptides, etc., through at least one antigen recognition site located in the variable region of the immunoglobulin molecule. As used herein, the term encompasses polyclonal antibodies, monoclonal antibodies, chimeric antibodies, bispecific antibodies, dual-specific antibodies, bifunctional antibodies, trispecific antibodies, multispecific antibodies, bispecific heterodimeric biantibodies, bispecific heterodimeric IgG, labeled antibodies, humanized antibodies, human antibodies and their fragments (such as Fab, Fab', F(ab') ₂ , Fv), single-chain (ScFv) and domain antibodies (including, for example, antibodies from sharks and camels), fusion proteins containing antibodies, and any other modified conformation of an immunoglobulin molecule containing an antigen recognition site. Antibodies include any class of antibodies, such as IgG, IgA, or IgM (or their subclasses), and the antibody need not be of any particular class. Immunoglobulins can be designated as different classes depending on the amino acid sequence of their heavy chain constant region. There are five main classes of immunoglobulins: IgA, IgD, IgE, IgG, and IgM, and several of these can be further subdivided into subclasses (isotypes), such as IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2. The heavy chain constant regions corresponding to different classes of immunoglobulins are designated α, δ, ε, γ, and μ, respectively. The subunit structures and three-dimensional conformations of different classes of immunoglobulins are well known. This invention also includes "antibody analogs," other scaffolds based on non-antibody molecular proteins, such as fusion proteins and/or immunoconjugates using CDRs to provide specific antigen binding. The antibodies of this invention can be derived from any species, including but not limited to mice, humans, camels, llamas, fish, sharks, goats, rabbits, chickens, horses, and cattle.

The term "antibody" further includes immunoglobulin molecules comprising four polypeptide chains (two heavy chains (H) and two light chains (L)) linked by disulfide bonds, as well as their multimers (e.g., IgM). Each heavy chain comprises a heavy chain variable region (abbreviated herein as HC VR or VH) and a heavy chain constant region. The "variable region" of an antibody refers to the variable region of either the antibody light chain alone or in combination with the variable region of the antibody heavy chain. The heavy chain constant region comprises three domains: CH1, CH2, and CH3. The CH1 and CH2 domains are linked by hinge regions. Each light chain comprises a light chain variable region (abbreviated herein as LC VR or VL) and a light chain constant region. The light chain constant region comprises one domain (CL1). The VH and VL regions can be further subdivided into hypervariable regions (called complementarity-determining regions (CDRs)) which are interspersed with more conserved regions (called framework regions (FRs)). Each VH and VL consists of three CDRs and four FRs, arranged in the following order from the amino terminus to the carboxyl terminus: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. In different embodiments of the invention, the FRs of the CD3 antibody (or its antigen-binding portion) may be identical to the human germline sequence, or may be natural or artificially modified. The common amino acid sequence can be defined based on a side-by-side analysis of two or more CDRs. The CDRs in each chain are tightly held together by the FRs and, together with the CDRs from the other chain, contribute to the formation of the antigen-binding site of the antibody.

As used herein, the terms “antigen-binding fragment,” “antibody fragment,” or “antigen-binding portion” refer to one or more segments of an antibody that retain the ability to specifically bind to an antigen. Examples of binding fragments encompassed within the term "antigen-binding fragment" in the term antibody include (i) antibody heavy chain variable domains (VH) and/or antibody light chain variable domains (VL), or VH/VL pairs derived from full-length antibodies or antibody fragments (such as VH and/or VL domains); (ii) Fab fragments, i.e., monovalent fragments consisting of VL, VH, CL, and CH1 domains; (iii) Fab' fragments, which are essentially Fab fragments having a portion of a hinge region (e.g., Fundamental Immunology, edited by Paul, 3rd edition, 1993); (iv) F(ab') ₂ fragments, i.e., bivalent fragments containing two Fab fragments connected by a disulfide bridge at the hinge region; (v) Fd fragments, which consist of VH and CH1 domains; (vi) Fv fragments, which consist of the VL and VH domains of a single arm of the antibody; (vii) Single-chain Fv fragments (scFv), i.e., a single protein chain in which the VL and VH regions pair to form a monovalent molecule (e.g., Bird et al. (1988) Science 242:423-426; and Huston et al. (1988) Proc. Natl. Acad. Sci. USA 85:5879-5883); (viii) disulfide-stabilized Fv fragments (dsFv), i.e., Fv with engineered intermolecular disulfide bonds that stabilize the VH-VL pair; (ix) single variable domain antibody (sdAb or dAb) fragments (e.g., Ward et al. (1989) Nature 341:544-546), which consist of a variable domain of a heavy chain and have no light chain; (x) complementarity-determining regions (CDRs); and any derivative thereof.

As used herein, an antibody's "antigen-binding fragment" may contain at least one variable domain. The variable domain may have any size or amino acid composition and will typically contain at least one CDR adjacent to or framed with one or more framework sequences. As used herein, an "antigen-binding fragment" of an antibody may comprise a homodimer or heterodimer (or other multimer) of any variable region and constant domain configuration that is non-covalently associated with each other and/or non-covalently associated (e.g., via disulfide bonds) with one or more monomeric VH or VL regions. For example, the variable region may be dimer and contain VH-VH, VH-VL, or VL-VL dimers. The configurations of the variable and constant domains that can be found within the antigen-binding fragment of the antibody of the present invention include: VH-CH1; VH-CH2; VH-CH3; VH-CH1-CH2; VH-CH1-CH2-CH3; VH-CH2-CH3; VH-VL-CL, VH-VL-CH1, VH-VL-CH2; VH-CL; VL-CH1; VL-CH2; VL-CH3; VL-CH1-CH2; VL-CH1-CH2-CH3; VL-CH2-CH3; and VL-CL. The variable and constant domains can be directly connected to each other or can be connected via complete or partial hinge or linker regions. The hinge region can consist of at least two (e.g., five, ten, fifteen, twenty, forty, sixty, or more) amino acids, which form flexible or semi-flexible bonds between adjacent variable and/or constant domains in a single polypeptide molecule.

As used herein, a “binding domain” includes any region of a polypeptide (e.g., an antibody) responsible for selectively binding to a target molecule (e.g., an antigen, ligand, receptor, substrate, or inhibitor). Exemplary binding domains include antibody variable regions, receptor-binding domains, ligand-binding domains, and enzyme-catalyzing domains.

As used herein, the term "human receptor framework" is a framework comprising an amino acid sequence of a light chain variable (VL) framework or a heavy chain variable (VH) framework derived from the human immunoglobulin framework or the human common framework as defined below. A human receptor framework "derived from" the human immunoglobulin framework or the human common framework may contain the same amino acid sequence, or may contain amino acid sequence modifications. In some embodiments, the number of amino acid modifications is 10 or less, 9 or less, 8 or less, 7 or less, 6 or less, 5 or less, 4 or less, 3 or less, or 2 or less. In some embodiments, the sequence of the VL human receptor framework is identical to the sequence of the VL human immunoglobulin framework or the human common framework.

As used herein, an “affinity-matured” antibody is an antibody that has one or more modifications in one or more variable regions (including CDR and FR) compared to an unmodified parent antibody, wherein such modifications result in an increased affinity of the antibody for the antigen.

As used herein, the terms “Fc region,” “Fc domain,” “Fc chain,” or similar terms are used to define the C-terminal region of the IgG heavy chain. The Fc region of IgG contains two constant domains, CH2 and CH3. The CH2 domain of the human IgG Fc region typically extends from amino acid 231 to amino acid 340 according to the EU index numbering system, or from amino acid 244 to amino acid 360 according to the Kabat numbering system. The CH3 domain of the human IgG Fc region typically extends from amino acid 341 to amino acid 447 according to the EU index numbering system, or from amino acid 361 to amino acid 478 according to the Kabat numbering system. The CH2 domain of the human IgG Fc region (also known as the “Cγ2” domain) is unique because it does not pair closely with the other domain. Instead, two N-linked branched carbohydrate chains are inserted between the two CH2 domains of intact native IgG. The Fc region may contain native or variant Fc sequences. Although the boundaries of the Fc sequence of the immunoglobulin heavy chain may vary, the human IgG heavy chain Fc sequence is generally defined as an extension from an amino acid residue at approximately position Cys226 or approximately position Pro230 to the carboxyl terminus of said Fc sequence. Unless otherwise specified herein, the amino acid residues in the Fc region or constant region are numbered according to the EU numbering system, also known as the EU index, as described in Kabat et al., Sequences of Proteins of Immunological Interest, 5th edition, Public Health Service, National Institutes of Health, Bethesda, MD, 1991.

In some embodiments, the Fc chain begins in a hinge region immediately upstream of the papain cleavage site and terminates at the C-terminus of the antibody. Therefore, a complete Fc chain comprises at least a hinge domain, a CH2 domain, and a CH3 domain. In some embodiments, the Fc chain comprises at least one of the following: a hinge (e.g., upper, middle, and/or lower hinge region) domain, a CH2 domain, a CH3 domain, a CH4 domain, or a variant, portion, or fragment thereof. In some embodiments, the Fc domain comprises a complete Fc chain (i.e., a hinge domain, a CH2 domain, and a CH3 domain). In some embodiments, the Fc chain comprises a hinge domain (or a portion thereof) fused to a CH3 domain (or a portion thereof). In some embodiments, the Fc chain comprises a CH2 domain (or a portion thereof) fused to a CH3 domain (or a portion thereof). In some embodiments, the Fc chain consists of a CH3 domain or a portion thereof. In some embodiments, the Fc chain consists of a hinge domain (or a portion thereof) and a CH3 domain (or a portion thereof). In some embodiments, the Fc chain consists of a CH2 domain (or a portion thereof) and a CH3 domain. In some embodiments, the Fc chain consists of a hinge domain (or a portion thereof) and a CH2 domain (or a portion thereof). In some embodiments, the Fc chain lacks at least a portion of the CH2 domain (e.g., all or a portion of the CH2 domain). As used herein, Fc chain generally refers to a polypeptide comprising all or a portion of the Fc chain of an immunoglobulin heavy chain. This Fc chain includes, but is not limited to, polypeptides comprising the entire CHI, hinge, CH2, and/or CH3 domains, and fragments of such peptides comprising only, for example, the hinge, CH2, and CH3 domains. Fc chains can be derived from immunoglobulins of any species and/or any subtype, including but not limited to human IgG1, IgG2, IgG3, IgG4, IgD, IgA, IgE, or IgM antibodies. The Fc domain encompasses native Fc and Fc variant molecules. As with Fc variants and native Fc, the term Fc chain includes molecules in monomeric or polymeric form, whether derived from whole antibody digestion or otherwise. In some embodiments, the Fc chain comprises the carboxyl-terminal portions of two heavy chains held together by disulfide bonds. In some embodiments, the Fc chain consists of a CH2 domain and a CH3 domain.

As used in the art, “Fc receptor” and “FcR” describe receptors that bind to the Fc region of an antibody. Preferred FcRs are naturally occurring human FcRs. Furthermore, preferred FcRs are those that bind IgG antibodies (γ receptors) and include receptors of the FcγRI, FcγRII, and FcγRIII subclasses, including allelic variants and alternative splicing forms of these receptors. FcγRII receptors include FcγRIIA (“activating receptor”) and FcγRIIB (“inhibitory receptor”), which are distinguished primarily by similar amino acid sequences in their cytoplasmic domains. FcRs are reviewed in: Ravetch and Kinet, Ann. Rev. Immunol., 9:457-92, 1991; Capel et al., Immunomethods, 4:25-34, 1994; and de Haas et al., J. Lab. Clin. Med., 126:330-41, 1995. "FcR" also includes the neonatal receptor FcRn, which is responsible for the transfer of maternal IgG to the fetus (Guyer et al., J. Immunol., 117:587, 1976; and Kim et al., J. Immunol., 24:249, 1994).

The “natural sequence Fc region” or “wild-type Fc region” contains the same amino acid sequence as the Fc region found in nature. “Wild-type” human IgG Fc refers to the sequence of amino acids naturally present in the human population. Of course, just as the Fc sequence can vary slightly between individuals, one or more modifications can be made to the wild-type sequence while still remaining within the scope of this invention. For example, the Fc region may contain additional modifications unrelated to this invention, such as mutations in glycosylation sites, including non-natural amino acids, or “knobs-in-holes” mutations.

The “variant Fc region” or “variant Fc chain” comprises an amino acid sequence that differs from the amino acid sequence of the natural Fc region by means of at least one amino acid modification, but retains at least one effector function of the natural Fc region. In some embodiments, the variant Fc chain has at least one amino acid substitution compared to the natural Fc chain or the Fc region of the parent polypeptide, for example, about 1 to about 10 amino acid substitutions in the natural Fc chain or the Fc chain of the parent polypeptide, and preferably about 1 to about 5 amino acid substitutions. The variant Fc chain described herein will preferably have at least about 80% sequence identity with the natural Fc chain and/or the Fc chain of the parent polypeptide, and most preferably at least about 90% sequence identity, more preferably at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99% sequence identity.

As used herein, the term "effective function" refers to those biological activities induced by the Fc region of an antibody (the native sequence Fc chain or the Fc chain of an amino acid sequence variant), and varies with antibody isotype. Examples of antibody effector functions include: C1q binding and complement-dependent cytotoxicity; Fc receptor binding; antibody-dependent cell-mediated cytotoxicity (ADCC); phagocytosis; downregulation of cell surface receptors (e.g., B cell receptors); and B cell activation. Such effector functions generally require the combination of the Fc chain with a binding domain (e.g., the antibody variable region) and can be evaluated using various assays known in the art for assessing such antibody effector functions. Exemplary measurements of effector functions are via Fcγ3 and/or C1q binding.

The effector function of an antibody is determined by a sequence in the Fc chain; this chain is also the part recognized by the Fc receptor (FcR) found on certain types of cells.

In some embodiments, the Fc polypeptide contains part or all of the wild-type hinge sequence (typically at its N-terminus). In some embodiments, the Fc polypeptide does not contain a functional or wild-type hinge sequence.

As used herein, “hinge region,” “hinge sequence,” and their variations include meanings known in the art, as illustrated in, for example, Janeway et al., ImmunoBiology: the immune system in health and disease, Elsevier Science Ltd., NY (4th edition, 1999); Bloom et al., Protein Science, 6:407-415, 1997; and Humphreys et al., J. Immunol. Methods, 209:193-202, 1997.

As used herein, "immunoglobulin-like hinge region," "immunoglobulin-like hinge sequence," and variations thereof refer to the hinge region and hinge sequence of an immunoglobulin-like or antibody-like molecule (e.g., an immunoadhesin). In some embodiments, the immunoglobulin-like hinge region may be derived from or originate from any IgG1, IgG2, IgG3, or IgG4 subtype, or IgA, IgE, IgD, or IgM, including its chimeric forms, such as a chimeric IgG1/2 hinge region.

In some embodiments, the hinge region may be derived from a human IgG1 subtype extending from amino acid 216 to amino acid 230 according to the EU indexing system or from amino acid 226 to amino acid 243 according to the Kabat numbering system. In some embodiments, the sequence may be EPKSCDKTHTCPPCP (SEQ ID NO: 63). Those skilled in the art may have different understandings of the exact amino acids corresponding to the various domains of the IgG molecule. Therefore, the N-terminus or C-terminus of the domains outlined above may be extended or shortened by 1, 2, 3, 4, 5, 6, 7, 8, 9, or even 10 amino acids.

In some embodiments, the hinge region may be mutated by one or more amino acids. In some embodiments, the hinge region may be truncated and contain only a portion of the full hinge region. In some embodiments, the hinge region may contain only the last 5 amino acids of the hinge region, referred to herein as the "lower hinge" region. In some embodiments, the lower hinge region may contain amino acid CPPCP (SEQ ID NO: 64) or amino acids 226 to 230 according to the EU index numbering system or amino acids 239 to 243 according to the Kabat numbering system.

As used herein, the term "modification" refers to the substitution, insertion, and/or deletion of amino acids in a polypeptide sequence, alteration of the portion chemically linked to the protein, or modification of the function of a protein (e.g., an antibody). For example, a modification can be an alteration of antibody function or a change in the structure of a carbohydrate linked to the protein. As used herein, "amino acid modification" refers to a mutation (substitution), insertion (addition), or deletion of one or more amino acid residues in an antibody. The term "amino acid mutation" means that at least one existing amino acid residue is replaced by another different amino acid residue (e.g., a substitute amino acid residue). The term "amino acid deletion" means the removal of at least one amino acid residue at a predetermined position in the amino acid sequence. For example, mutation L234A indicates that the amino acid residue lysine at position 234 in the Fc region of the antibody is replaced by the amino acid residue alanine (replacing lysine with alanine) (numbered according to the EU index numbering system). "Natural amino acid residues" refers to amino acid residues from the following groups: alanine (three-letter code: Ala, single-letter code: A), arginine (Arg, R), asparagine (Asn, N), aspartic acid (Asp, D), cysteine (Cys, C), glutamine (Gin, Q), glutamic acid (Glu, E), glycine (Gly, G), histidine (His, H), isoleucine (Ile, I), leucine (Leu, L), lysine (Lys, K), methionine (Met, M), phenylalanine (Phe, F), proline (Pro, P), serine (Ser, S), threonine (Thr, T), tryptophan (Trp, W), tyrosine (Tyr, Y), and valine (Val, V).

The term "pharmaceutical" is used herein to mean a biological macromolecule, an extract derived from biological material, a mixture of biological macromolecules, a chemical compound, a mixture of chemical compounds, and/or a mixture of chemical compounds and biological macromolecules. The term "therapeutic agent" refers to a biologically active pharmaceutical agent.

As used herein, a “monoclonal antibody” refers to an antibody obtained from a substantially homogeneous population of antibodies, i.e., the individual antibodies constituting the population are identical except for the possibility of naturally occurring mutations present in small amounts. Monoclonal antibodies are highly specific against a single antigenic site. Furthermore, in contrast to polyclonal antibody formulations, which typically comprise different antibodies targeting different determinants (epitopes), each monoclonal antibody targets a single determinant on the antigen. The modifier “monoclonal” indicates the characteristics of antibodies obtained from a substantially homogeneous population of antibodies and should not be construed as requiring antibody production by any particular method. For example, a monoclonal antibody to be used according to the invention can be prepared by a hybridoma method first described by Kohler and Milstein, Nature 256:495, 1975, or by a recombinant DNA method such as that described in U.S. Patent No. 4,816,567. Monoclonal antibodies can also be isolated from phage libraries generated using, for example, techniques described in McCafferty et al., Nature 348:552-554, 1990.

The antibodies of this invention can be “humanized antibodies.” As used herein, a “humanized” antibody refers to a form of non-human (e.g., mouse, rat, rabbit, non-human primate, or other mammalian) antibody that is a chimeric immunoglobulin, immunoglobulin chain, or fragment thereof (such as Fv, Fab, Fab', F(ab') ₂ , or other antigen-binding sequences of the antibody) containing one or more amino acid residues introduced therefrom from a non-human source. These non-human amino acid residues are generally referred to as “input” residues and are typically derived from an “input” variable domain. The input residues, sequence, or antibody has desired affinity and/or specificity or other desired antibody biological activities as discussed herein.

Preferably, the humanized antibody is a human immunoglobulin (receptor antibody) in which residues from the complementarity-determining region (CDR) of the receptor are replaced with residues from a non-human species (donor antibody), such as mouse, rat, or rabbit, that have the desired specificity, affinity, and ability. In some cases, residues from the Fv framework region (FR) of the human immunoglobulin are replaced with corresponding non-human residues. Furthermore, the humanized antibody may contain residues not found in either the receptor antibody or the input CDR or framework sequence, but which are included to further refine and optimize antibody performance. Typically, the humanized antibody will contain at least one, and usually substantially all, of the two variable regions, wherein all or substantially all of the CDR regions correspond to those of non-human immunoglobulins, and all or substantially all of the FR regions are those of human immunoglobulin common sequences. The humanized antibody will also preferably contain at least a portion of the immunoglobulin constant region or structural domain (Fc), typically of the type of human immunoglobulin. Antibodies having Fc chains modified as described in WO 99/58572 are preferred. Other forms of humanized antibodies have one or more CDRs (CDR L1, CDR L2, CDR L3, CDR H1, CDR H2, or CDR H3) that are altered relative to the original antibody, and are also referred to as "derived from" one or more CDRs from the original antibody. As used herein, humanization intention includes deimmunization antibodies.

As used herein, "human antibody" means an antibody having an amino acid sequence corresponding to the amino acid sequence of an antibody produced by a human, and/or having been prepared using any technique known to those skilled in the art or disclosed herein for the preparation of human antibodies. Therefore, the term "human antibody" as used herein is intended to include antibodies having variable and constant regions derived from human immunoglobulin sequences. Human antibodies of the present invention may include amino acid residues not encoded by human immunoglobulin sequences (e.g., mutations introduced by in vitro random mutagenesis or site-specific mutagenesis or by in vivo somatic mutations), such as in the CDR, and specifically CDR3. This definition of human antibody includes antibodies comprising at least one human heavy chain polypeptide or at least one human light chain polypeptide. One such example is an antibody comprising a mouse light chain and a human heavy chain polypeptide. Various techniques known in the art can be used to produce human antibodies. In one embodiment, the human antibody is selected from a phage library expressing a human antibody (Vaughan et al., Nature Biotechnology, 14:309-314, 1996; Sheets et al., Proc. Natl. Acad. Sci. (USA) 95:6157-6162, 1998; Hoogenboom and Winter, J. Mol. Biol., 227:381, 1991; Marks et al., J. Mol. Biol., 222:581, 1991). The human antibody can also be prepared by immunizing an animal in which a human immunoglobulin locus has been transgenically introduced to replace an endogenous locus, such as in mice where the endogenous immunoglobulin gene has been partially or completely inactivated. This method is described in U.S. Patent Nos. 5,545,807; 5,545,806; 5,569,825; 5,625,126; 5,633,425; and 5,661,016. Alternatively, human antibodies can be prepared by immortalizing human B lymphocytes that produce antibodies against a target antigen (such B lymphocytes can be recovered from an individual or a single-cell clone of cDNA, or can have been immunized in vitro). See, for example, Cole et al. Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, p. 77, 1985; Boerner et al., J. Immunol., 147(1):86-95, 1991; and U.S. Patent No. 5,750,373.

The human antibodies of the present invention can exist in at least two forms related to hinge heterogeneity. For example, immunoglobulin molecules comprise a stable four-chain construct of approximately 150-160 kDa, wherein the dimers are held together by interchain heavy chain disulfide bonds. Alternatively, the dimers are not linked by interchain disulfide bonds and form a molecule of approximately 75-80 kDa (half-antibody) consisting of covalently coupled light and heavy chains.

As used herein, the term "recombinant human antibody" is intended to include all human antibodies prepared, expressed, generated, or isolated in a recombinant manner, such as antibodies expressed using a recombinant expression vector transfected into host cells (described further below), antibodies isolated from a library of recombinant human antibodies (described further below), antibodies isolated from animals transgenic with human immunoglobulin genes (e.g., mice) (see, for example, Taylor et al. (1992) Nucl. Acids Res. 20:6287-6295), or antibodies prepared, expressed, generated, or isolated by any other means involving splicing a human immunoglobulin gene sequence into another DNA sequence. Such recombinant human antibodies have variable and constant regions derived from human germline immunoglobulin sequences. However, in some implementations, such recombinant human antibodies are subjected to in vitro mutagenesis (or, when using animals transgenic with human Ig sequences, in vivo somatic cell mutagenesis), and thus the amino acid sequences of the VH and VL regions of the recombinant antibody are sequences that, although derived from and associated with human germline VH and VL sequences, may not naturally exist in the human antibody germline library.

The antibodies of this invention can be “chimeric antibodies” (where a portion of the heavy chain and/or light chain is identical or homologous to the corresponding sequence in an antibody derived from a specific species or belonging to a specific antibody class or subclass, while the remaining portion of the chain is identical or homologous to the corresponding sequence in an antibody derived from another species or belonging to another antibody class or subclass), and fragments of such antibodies, provided they exhibit the desired biological activity (US Patent 4,816,567; and Morrison et al., Proc. Natl. Acad. Sci. USA 81:6851-6855, 1984). The target chimeric antibodies herein include primate-derived antibodies comprising a variable region antigen-binding sequence derived from a non-human primate (e.g., Old World monkeys, apes, etc.) and a human constant region sequence.

A "monovalent antibody" contains one antigen-binding site/molecule (e.g., IgG or Fab). In some cases, a monovalent antibody may have more than one antigen-binding site, but the binding sites come from different antigens.

A "monospecific antibody" is an antibody or antibody preparation that contains two identical antigen-binding sites/molecules (e.g., IgG), such that the two binding sites bind to the same epitopes on the antigen. Therefore, they compete with each other when binding to a single antigen molecule. This term includes "monoclonal antibody" or "monoclonal antibody composition." Most antibodies found in nature are monospecific. In some cases, monospecific antibodies can also be monovalent antibodies (e.g., Fab).

The terms “mutation load,” “mutational load,” “mutation burden,” or “mutational burden” are used interchangeably in this article. Tumor mutational load is a measure of the number of mutations within a tumor genome, defined as the total number of mutations in each coding region of the tumor genome. Mutational load varies greatly across tumor types, ranging from just a few to several thousand mutations (Alexandrov LB et al., Nature 2013;500(7463):415-421; Lawrence MS et al., Nature 2013;499:214–218; Vogelstein B et al., Science, 2013;339:1546–1558).

"Original amino acid residue" refers to a residue replaced by an "input amino acid" residue whose side chain volume may be smaller or larger than the original residue. The input amino acid residue may be naturally occurring or non-naturally occurring amino acid residues, but is preferably the former. "Naturally occurring" amino acid residues are those residues encoded by the genetic code. "Non-naturally occurring" amino acid residues refer to residues that are not encoded by the genetic code but can covalently bind to neighboring amino acid residues in the polypeptide chain. Examples of non-naturally occurring amino acid residues are leucine, ornithine, valine, homoserine, and other amino acid residue analogs, such as those described, for example, in Ellman et al., Meth. Enzym. 202:301-336 (1991). In some embodiments, the method of the present invention involves replacing at least one original amino acid residue, but may replace more than one original residue. Typically, the total residues in no more than the first or second polypeptide interface may contain the replaced original amino acid residue.

As used herein with respect to the use of the term "competitive," it means that a first antibody or its antigen-binding fragment (or portion) binds to an epitope in a manner sufficiently similar to the binding of a second antibody or its antigen-binding portion, such that the binding of the first antibody to its homologous epitope is detectably reduced in the presence of the second antibody compared to the binding of the first antibody in the absence of the second antibody. Alternatives to this, where the binding of the second antibody to its epitope is also detectably reduced in the presence of the first antibody, are possible but not necessary. That is, the first antibody may inhibit the binding of the second antibody to its epitope, while the second antibody may not inhibit the binding of the first antibody to its corresponding epitope. However, when each antibody detectably inhibits the binding of another antibody to its homologous epitope or ligand, whether to the same, greater, or lesser extent, the antibodies are said to "cross-compete" with each other for binding to their respective epitopes. Both competitive antibodies and cross-competitive antibodies are covered by this invention. Regardless of the mechanism by which such competition or cross-competition occurs (e.g., steric hindrance, conformational change, or binding to a common epitope or a portion thereof), based on the teachings provided herein, those skilled in the art will understand that such competitive antibodies and/or cross-competitive antibodies are covered and can be used in the methods disclosed herein.

As used herein, “antibody-dependent cell-mediated cytotoxicity” or “ADCC” refers to a cell-mediated reaction in which nonspecific cytotoxic cells expressing Fc receptors (FcRs), such as natural killer (NK) cells, neutrophils, and macrophages, recognize binding antibodies on target cells and subsequently cause lysis of the target cells. The ADCC activity of a target molecule can be evaluated using in vitro ADCC assays, such as those described in U.S. Patent Nos. 5,500,362 or 5,821,337. Useful effector cells for such assays include peripheral blood mononuclear cells (PBMCs) and NK cells. Alternatively or additionally, the ADCC activity of a target molecule can be evaluated in vivo, for example, in animal models, such as those disclosed in Clynes et al., PNAS (USA), 95:652-656, 1998.

As used herein, “complement-dependent cytotoxicity” or “CDC” refers to target lysis in the presence of complement. The complement activation pathway is initiated by the binding of the first component (C1q) of the complement system to molecules (e.g., antibodies) that are complexed with homologous antigens. To evaluate complement activation, a CDC assay can be performed, for example, as described in Gazzano-Santoro et al., J. Immunol. Methods, 202: 163, 1996.

As used herein, the terms “immunospecific binding,” “immunospecific recognition,” “specific binding,” “specific recognition,” and similar terms refer to a molecule that specifically binds to an antigen (e.g., an epitope or immune complex) but does not specifically bind to another molecule (e.g., a binding domain). A molecule that specifically binds to an antigen may bind to other peptides or polypeptides with lower affinity, as determined by assays known in the art, such as immunoassays, BIACORE™, or other assays. Preferably, the molecule that specifically binds to the antigen does not cross-react with other proteins.

Suitable “moderately stringent conditions” include prewashing in a solution of 5X SSC, 0.5% SDS, and 1.0 mM EDTA (pH 8.0); hybridization overnight at 50–65°C with 5X SSC; followed by two washes at 65°C with 2X, 0.5X, and 0.2X SSC containing 0.1% SDS, each for 20 minutes.

As used herein, “highly stringent conditions” or “highly stringent conditions” are conditions in which: (1) low ionic strength and high temperature are used for washing, for example, 0.015 M sodium chloride/0.0015 M sodium citrate/0.1% sodium dodecyl sulfate at 50°C; (2) denaturing agents are used during hybridization at 42°C, such as formamide, for example, 50% (v/v) formamide with 0.1% bovine serum albumin/0.1% Ficoll/0.1% polyvinylpyrrolidone/50 mM sodium phosphate buffer with 750 mM sodium chloride and 75 mM sodium citrate at pH 6.5; Alternatively (3) wash at 42°C with 50% formamide, 5 x SSC (0.75 M NaCl, 0.075 M sodium citrate), 50 mM sodium phosphate (pH 6.8), 0.1% sodium pyrophosphate, 5 x Denhardt solution, sonicated salmon sperm DNA (50 µg/ml), 0.1% SDS, and 10% dextran sulfate, followed by washing at 42°C in 0.2 x SSC (sodium chloride/sodium citrate) and at 55°C in 50% formamide, followed by a high-rigidity wash at 55°C with 0.1 x SSC containing EDTA. Technicians will recognize how to adjust the temperature, ionic strength, etc., as needed to accommodate factors such as probe length.

The binding protein, binding domain, or CDR or antibody (as broadly defined herein) may be identified according to the accumulation, AbM, contact, North definition and/or conformational definition of Kabat, Chothia, Kabat and Chothia, or any method of CDR determination well known in the art. See, for example, Kabat et al., 1991, Sequences of Proteins of Immunological Interest, 5th edition (Hypervariate Region); Chothia et al., Nature 342:877-883, 1989 (Structural Ring Structure). The identity of the amino acid residues constituting the CDR in a particular antibody may be determined using methods well known in the art. The AbM definition of the CDR is a compromise between Kabat and Chothia and uses Oxford Molecular's AbM antibody modeling software (Accelrys®). The definition of “contact” for the CDR is based on observed antigen contact and is described in MacCallum et al., J. Mol. Biol., 262:732-745, 1996. The “conformation” definition of a CDR is based on residues that contribute enthalpy to antigen binding (see, for example, Makabe et al., J. Biol. Chem., 283:1156-1166, 2008). North has identified canonical CDR conformations using different preferred sets of CDR definitions (North et al., J. Mol. Biol. 406: 228-256, 2011). In another approach referred to herein as the “conformation definition” of a CDR, the position of the CDR can be identified as residues that contribute enthalpy to antigen binding (Makabe et al., J. Biol. Chem., 283:1156-1166, 2008). Other CDR boundary definitions may not strictly follow one of the above approaches but will still overlap with at least a portion of the Kabat CDR, although they may be shortened or lengthened based on predictions or experimental findings that specific residues or groups of residues or even the entire CDR do not significantly affect antigen binding. As used herein, a CDR can refer to a CDR defined by any method known in the art, including combinations of methods. The methods used herein can utilize a CDR defined according to any of these methods. For any given embodiment containing more than one CDR, a CDR (or other residues of the antibody) can be defined according to any of the following definitions: Kabat, Chothia, extension (a combination of Kabat and Chothia), North, extension, AbM, contact, and/or conformation.

Residues in the variable domain are numbered according to Kabat, a numbering system used for the heavy chain or light chain variable regions of the assembled antibody. See Kabat et al., Sequences of Proteins of Immunological Interest, 5th edition, Public Health Service, National Institutes of Health, Bethesda, MD, 1991. Using this numbering system, the actual linear amino acid sequence may contain fewer or additional shortened or inserted amino acids corresponding to the FR or CDR of the variable region. For example, the heavy chain variable region may include a single amino acid insertion after residue 52 of H2 (according to Kabat residue 52a) and inserted residues after heavy chain FR residue 82 (e.g., according to Kabat residues 82a, 82b, and 82c). The Kabat number of residues for a given antibody can be determined by comparing the homologous regions of the antibody sequence to a “standard” Kabat numbered sequence. Various algorithms for assigning Kabat numbers are available. This article uses the algorithm implemented in Abysis release 2.3.3 (www.abysis.org) to assign Kabat numbers to the variable regions VLCDR1, VL CDR2, VL CDR3, VHCDR1, VHCDR2, and VH CDR3. Specific amino acid residue positions within the antibody can also be numbered based on Kabat numbers.

As used herein, the term "phage display library" refers to a population of phages, each containing exogenous cDNA fused to a surface protein within a recombination frame. The phages display the exogenous protein encoded by the cDNA on their surface. After replication in a bacterial host (typically *E. coli*), phages containing the target exogenous cDNA are selected based on the expression of the exogenous protein on the phage surface.

The term "epitope" refers to a portion of a molecule that can be recognized, contacted, and/or bound by an antibody at one or more sites in an antigen-binding region called a complementary site. A single antigen may have more than one epitope. Thus, different antibodies can bind to different regions on the antigen and may have different biological effects. Epitopes are often composed of chemically active surface groups of a molecule (such as amino acid or sugar side chains) and have specific three-dimensional structural features and specific charge features. As used herein, epitopes can be conformational or linear. Conformational epitopes are generated by spatially juxtaposing amino acids from different segments of a linear polypeptide chain. Linear epitopes are epitopes generated from neighboring amino acid residues in a polypeptide chain. In some embodiments, an epitope may include a sugar, phosphoryl, or sulfonyl portion of the antigen.

As used herein, an "antigenic epitope" is defined as a portion of a polypeptide to which an antibody can specifically bind, as determined by any method well known in the art (e.g., by a routine immunoassay). A "nonlinear epitope" or "conformational epitope" comprises a non-neighboring polypeptide (or amino acid) within an antigenic protein to which an antibody specific for that epitope binds. Once a desired epitope on an antigen is identified, an antibody targeting that epitope can be generated, for example, using the techniques described herein.

When used to describe the interaction between antibodies and proteins or peptides, the terms "specific binding" or "specifically binding" refer to an interaction that depends on the presence of a specific structure on the protein (i.e., an antigenic determinant or epitope); in other words, the antibody recognizes and binds to a specific protein structure rather than the usual protein. For example, if an antibody is specific to epitope "A," the presence of a protein containing epitope A (or free unlabeled A) in a reaction containing labeled "A" and an antibody will reduce the amount of labeled A that binds to the antibody.

In some embodiments, "specific binding" means, for example, that an antibody binds to a protein with a KD of about 0.1 nM or less, but more typically less than about 1 μM. In some embodiments, "specific binding" means that an antibody sometimes binds to a target with a KD of at least about 0.1 μM or less, at other times with at least about 0.01 μM or less, and at other times with at least about 1 nM or less. Due to sequence identity between homologous proteins in different species, specific binding may include antibodies that recognize proteins in more than one species (e.g., human or mouse proteins). Similarly, due to homology within certain regions of the polypeptide sequences of different proteins, specific binding may include antibodies that recognize more than one protein. It should be understood that in some embodiments, an antibody or binding portion that specifically binds to a first target may or may not specifically bind to a second target. Therefore, "specific binding" does not necessarily require (although it may include) exclusive binding, i.e., binding to a single target. Therefore, in some embodiments, an antibody may specifically bind to more than one target. In some embodiments, the same antigen-binding site on an antibody may bind to multiple targets. For example, in some cases, an antibody may contain two identical antigen-binding sites, each specifically binding to the same epitope on two or more proteins. In some alternative embodiments, the antibody may be multispecific and contain at least two antigen-binding sites with different specificities. By way of non-limiting example, a bispecific antibody may contain an antigen-binding site that recognizes an epitope on one protein (e.g., human CD3) and further contain a second, different antigen-binding site that recognizes a different epitope on a second protein. Generally, but not necessarily, references to binding refer to specific binding.

Antibodies that specifically bind to antigens can bind to other peptides or polypeptides with lower affinity, as determined by assays known in the art (e.g., immunoassays, BIACORE®, or other assays). Preferably, antibodies that specifically bind to antigens do not cross-react with other proteins.

When used in relation to the interaction between antibodies and proteins or peptides, the terms “non-specific binding” or “background binding” refer to interactions that do not depend on the presence of a specific structure (i.e., antibody binding is typically protein rather than a specific structure, such as an epitope).

As used herein, the terms "k _on " or " _ka " refer to the rate constant relating to the association of antibody with antigen. Specifically, the rate constants (k _on / _ka and k _off /k _d ) and equilibrium dissociation constants are measured using intact antibodies (i.e., bivalent) and monomeric CD3 proteins.

As used in this article, the terms "k _off " or "k _d " refer to the rate constant of antibody dissociation from the antibody/antigen complex.

As used in this article, the term " _KD " refers to the equilibrium dissociation constant of antibody-antigen interactions.

As used herein, the term "binding affinity" generally refers to the total strength of the non-covalent interaction between a single binding site of a molecule (e.g., an antibody) and its binding partner (e.g., an antigen). Unless otherwise indicated, as used herein, "binding affinity" refers to the intrinsic binding affinity that reflects a 1:1 interaction between members of a binding pair (e.g., antibody and antigen). The affinity of molecule X for its partner Y is generally represented by the dissociation constant (KD). The affinity of molecule X for its partner Y is generally represented by the dissociation constant (Kd). For example, Kd can be about 200 nM, 150 nM, 100 nM, 60 nM, 50 nM, 40 nM, 30 nM, 20 nM, 10 nM, 8 nM, 6 nM, 4 nM, 2 nM, 1 nM, or stronger. Affinity can be measured by methods commonly known in the art, including those described herein. Low-affinity antibodies typically bind to antigens slowly and tend to dissociate easily, while high-affinity antibodies typically bind to antigens quickly and tend to remain bound for a longer period. Various methods for measuring binding affinity are known in the art, any of which may be used for the purposes of this invention. Specifically, the term "binding affinity" is intended to refer to the rate of dissociation of a particular antigen-antibody interaction. _KD is the ratio of the rate of dissociation (also called the "off-rate" (k _-off )) to the rate of association (k-on) (k- _on ). Therefore, _KD equals k _-off / k _-on and is expressed as a molar concentration (M). Thus, the smaller the _KD , the stronger the binding affinity. Therefore, a _KD of 1 μM indicates weak binding affinity compared to _{a KD} of 1 nM. The _KD value of an antibody can be determined using methods well established in the art. One method for determining the _KD of an antibody is by using surface plasmon resonance (SPR), typically employing a biosensor system (such as the BIACORE™ system). BIACORE™ kinetic analysis involves analyzing the binding and dissociation of antigens from chips with immobilized molecules (e.g., molecules containing epitope-binding domains) on their surfaces. Another method for determining the _KD of antibodies is through biolayer interferometry, typically using ^OCTET® technology (Octet QKe system, ForteBio).

Unless otherwise specified, with respect to the antibodies (such as antibodies, fragments or derivatives thereof) of the present invention, "bioactive", "bioactive" and "biocharacteristic" mean having the ability to bind biomolecules.

As used herein, the terms "nucleic acid" and "nucleotide sequence" include DNA molecules (e.g., cDNA or genomic DNA), RNA molecules (e.g., mRNA), combinations or hybrid DNA/RNA molecules, and analogs of DNA or RNA molecules. Such analogs can be generated using, for example, nucleotide analogs, including but not limited to inosine or triphenylmethylated bases. Such analogs may also comprise DNA or RNA molecules containing a modified backbone that imparts beneficial properties to the molecule, such as, for example, nuclease resistance or increased transmembrane translucency. Nucleic acid or nucleotide sequences can be single-stranded, double-stranded, contain both single-stranded and double-stranded portions, and may contain triple-stranded portions, but are preferably double-stranded DNA.

The present invention also includes a polynucleotide encoding the antibody of the present invention (including the polypeptide and binding region of the antibody). The polynucleotide encoding the molecule of the present invention can be obtained, and the nucleotide sequence of the polynucleotide can be determined by any method known in the art.

The polynucleotide encoding the antibody of the present invention may include: a variant-only coding sequence; a variant-coding sequence and additional coding sequences (such as functional polypeptides or signal sequences or secretion sequences or preprotein sequences); and antibody-coding and non-coding sequences (such as introns or non-coding sequences at the 5' and/or 3' of the antibody-coding sequence). The term "polynucleotide encoding antibody" encompasses polynucleotides including additional coding sequences of variants, and also encompasses polynucleotides including additional coding and/or non-coding sequences. It is known in the art that polynucleotide sequences optimized for specific host cells/expression systems can be readily obtained from the amino acid sequence of the desired protein (see ^GENEART® AG, Regensburg, Germany).

The antibodies of the present invention may have additional conserved or non-essential amino acid substitutions that do not substantially affect the function of the polypeptide. Whether a particular substitution will be tolerated (i.e. will not adversely affect the desired biological properties (such as binding activity)) can be determined as described in Bowie, JU et al., Science 247: 1306-1310, 1990 or Padlan et al., FASEBJ. 9: 133-139, 1995.

As used herein, "conservative amino acid substitution" is a substitution in which an amino acid residue is replaced by an amino acid residue having a similar side chain. Families of amino acid residues having similar side chains are defined in the art. These families include amino acids having the following side chains: basic side chains (e.g., lysine, arginine, histidine), acidic side chains (e.g., aspartic acid, glutamic acid), uncharged polar side chains (e.g., asparagine, glutamine, serine, threonine, tyrosine, cysteine), nonpolar side chains (e.g., glycine, alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine, tryptophan), β-branched side chains (e.g., threonine, valine, isoleucine), and aromatic side chains (e.g., tyrosine, phenylalanine, tryptophan, histidine).

As used herein, "non-essential" amino acid residues are residues that may differ from the wild-type sequence of a binding agent (e.g., an antibody) and do not eliminate or substantially alter biological activity, while "essential" amino acid residues cause this change. In antibodies, essential amino acid residues can be specificity-determining residues (SDRs).

As used herein, the term "isolated" refers to material removed from its original environment (e.g., the natural environment if it is naturally occurring). For example, naturally occurring polynucleotides or polypeptides present in living animals are not isolated, but the same polynucleotides or polypeptides isolated from some or all of the coexisting material in a natural system are isolated. This polynucleotide may be part of a carrier, and/or this polynucleotide or polypeptide may be part of a composition (e.g., a mixture, solution, or suspension) or contain isolated cells or cultured cells containing the polynucleotide or polypeptide, and is still isolated because the carrier or composition is not part of its natural environment. For example, a polypeptide synthesized chemically or synthesized in a cellular system different from the cell of natural origin of the polypeptide will be "isolated" from its natural associated components. Protein purification techniques well known in the art can also be used to isolate proteins to make them substantially free of natural associated components.

"Isolated" or "purified" antibodies are substantially free of cellular material or other contaminating proteins from cell or tissue sources or culture media from which the protein originates, or are substantially free of chemical precursors or other chemicals during chemical synthesis. The phrase "substantially free of cellular material" includes antibody formulations in which the polypeptide/protein is isolated from cellular components of the cells from which the polypeptide/protein is produced, either isolated or recombinantly. Therefore, antibodies substantially free of cellular material include antibody formulations having less than about 50%, 40%, 30%, 20%, 10%, 5%, 2.5%, or 1% (by dry weight) of contaminating proteins. When antibodies are produced by recombinant synthesis, they are also preferably substantially free of culture media, i.e., the culture medium constitutes less than about 50%, 40%, 30%, 20%, 10%, 5%, 2.5%, or 1% of the volume of the protein formulation. When antibodies are produced by chemical synthesis, they are preferably substantially free of chemical precursors or other chemicals and reagents, i.e., the target antibody is isolated from chemical precursors or other chemicals involved in protein synthesis. Therefore, this antibody preparation contains less than about 50%, 40%, 30%, 20%, 10%, 5%, or 1% (by dry weight) of a chemical precursor or compound in addition to the target antibody. In a preferred embodiment of the invention, the antibody is isolated or purified. For the purposes of this invention, an antibody that has been isolated or removed from at least one component of an organism or from tissues or cells in which the antibody is naturally present or produced is called an "isolated antibody".

As used herein, the term “recycling” refers to, for example, a process that uses protein purification techniques well known in the art to separate chemical substances (such as peptides) to make them substantially free of their natural associated components.

As used herein, the term "replicon" refers to any genetic element, such as a plasmid, chromosome, or virus, that acts as an autonomous polynucleotide replication unit within a cell.

As used herein, the term "operably ligated" refers to a situation where the relationship between said elements allows them to function in the intended manner. For example, "operably ligated" to a control sequence of a coding sequence in a manner that enables expression of the coding sequence under suitable or compatible conditions with that control sequence. Generally, "operably ligated" means that the ligated DNA sequences are adjacent, and in the case of a secretory leader sequence, adjacent and in the reading phase. However, enhancers do not require adjacency. Ligation is accomplished by ligating at a convenient restriction site. If such a site is not present, synthetic oligonucleotide adaptors or linkers are used according to conventional practice.

As used herein, “vector” means a construct capable of being delivered in a host cell and preferably expressing one or more target genes or sequences. Examples of vectors include, but are not limited to, viral vectors, naked DNA or RNA expression vectors, plasmids, granules or phage vectors, DNA or RNA expression vectors associated with cationic condensers, DNA or RNA expression vectors encapsulated in liposomes, and certain eukaryotic cells, such as production cells.

As used herein, the term "expression control sequence" or "control sequence" refers to a polynucleotide sequence necessary for the expression of the coding sequence it is linked to. The nature of this control sequence varies depending on the host organism. For example, in prokaryotes, such control sequences typically include a promoter, a ribosome binding site, a terminator, and (in some cases) an enhancer. Therefore, the term "control sequence" is intended to include at least all elements necessary for expression, but may also include additional elements that are advantageous for its presence, such as a leader sequence.

"Host cell" includes an individual cell or cell culture that may be or has been a vector recipient, the vector being used to incorporate a polynucleotide insert. Host cells include progeny of a single host cell, and the progeny may not be identical to the original parent cell (morphologically or in terms of genomic DNA complementation) due to natural, accidental, or intentional mutations. Host cells include cells transfected in vivo with the polynucleotides of the present invention.

As used herein, "mammalian cell" includes references to cells derived from mammals, including humans, rats, mice, guinea pigs, chimpanzees, or macaques. These cells may be cultured in vivo or in vitro.

As used herein, the term "purified product" refers to a product formulation that has been isolated from the cellular components to which the product is normally associated and/or from other cell types that may be present in the target sample.

As used herein, “substantially pure” means a material that is at least 50% pure (i.e., free of contaminants), more preferably at least 90% pure, more preferably at least 95% pure, even more preferably at least 98% pure, and most preferably at least 99% pure.

As used herein, the terms “treat,” “treating,” or “treatment” are methods for obtaining beneficial or desired clinical outcomes. For the purposes of this invention, treatment is defined as administering a CD3 antibody molecule (e.g., a monoclonal antibody, bispecific antibody) to a subject (e.g., a patient), or administering (e.g., by application) an isolated tissue or cell from the subject and returning it to said subject. The CD3 antibody molecule may be administered alone or in combination with one or more agents. Treatment can be curative, healing, alleviating, resolving, altering, remedying, improving, reducing, improving, or affecting a condition, its symptoms, or a predisposition to the condition (e.g., cancer).

As used herein, the term “subject” is intended to include any animal (e.g., a mammal), including but not limited to humans, non-human primates, rodents, etc., that is a recipient of a particular treatment. For example, a subject may be a patient with cancer (e.g., a human patient or a veterinary patient). Generally, the terms “subject,” “individual,” and “patient” are used interchangeably in this document when referring to human subjects.

Unless otherwise stated, the term "non-human animal" in this invention includes all non-human vertebrates, such as non-human mammals and non-mammals, such as non-human primates, sheep, dogs, cows, chickens, amphibians, reptiles, mice, rats, rabbits, or goats, etc.

As used herein, the term “pharmaceutical acceptable” means a product or compound that has been approved (or permitted to be approved) by a federal or state government regulatory agency or is listed in the United States Pharmacopeia or other recognized pharmacopoeia for use in animals (including humans).

As used herein, the terms "pharmaceutically acceptable excipient, carrier, or adjuvant" or "acceptable pharmaceutical carrier" mean an excipient, carrier, or adjuvant that can be administered to a subject together with at least one antibody of this disclosure without destroying the activity of the antibody. When administered together with an antibody at a dose sufficient to deliver a therapeutic effect, the excipient, carrier, or adjuvant shall be non-toxic.

As used herein, the term “improvement” means a reduction or improvement of one or more symptoms compared to the absence of the antibody molecules of the present invention. “Improvement” also includes a reduction or decrease in the duration of symptoms.

As used herein, the terms “prevent,” “preventing,” and “prevention” refer to the prevention of recurrence or onset of one or more symptoms of a disease in a subject by means of the administration of a prophylactic or therapeutic agent.

As used herein, “effective amount,” “therapeutic effective amount,” “therapeutic adequate amount,” or “effective dose” means the amount of a therapeutic agent (e.g., a CD3 antibody, alone or as part of a multispecific antibody (e.g., bispecific)) administered to a subject in a single or multiple doses that is effective or sufficient to prevent, heal, improve, treat, or manage a disease, condition, or side effect, or to reduce the rate of progression of a disease or condition, or to prolong healing, alleviate, reduce, or improve the condition of a subject with a condition as described herein. The term also includes, within its scope, amounts that effectively enhance normal physiological function. An effective amount can be considered in the context of administration of one or more therapeutic agents, and a single agent may be considered to have been administered in an effective amount if, in combination with one or more other agents, the desired outcome can be achieved.

As used herein, effective agents can target effector T cells to induce T cell-mediated cytotoxicity in cells associated with diseases such as cancer, autoimmune diseases, or infectious diseases. The therapeutic agents of the present invention induce a T cell-mediated immune response by modulating the CD3 antigen, specifically CD3, and more specifically CD3ε, on the surface of T cells.

Regarding cancer, the therapeutically effective dose refers to the amount of a therapeutic agent that inhibits the growth of a tumor or cancer by slowing, interrupting, stopping, or terminating its growth and/or metastasis.

Efficacy is a measure of the activity of a therapeutic agent, expressed as the amount required to produce an effect of a given strength. A highly potent agent elicits a greater response at low concentrations compared to a less potent agent that elicits a smaller response at low concentrations. Efficacy varies with affinity and potency. Potency refers to the ability of a therapeutic agent to produce a biological response upon binding to a target ligand and the quantitative magnitude of that response. As used herein, the term “half-maximum effective concentration ( _EC50 )” refers to the concentration of a therapeutic agent that elicits a response midway between baseline and maximum after a specified exposure time. Therapeutic agents can cause inhibition or stimulation. The _EC50 value is commonly used and is referred to herein as a measure of efficacy.

As used herein, “combination therapy” or “in combination with” refers to the use of more than one prophylactic and/or therapeutic agent. The use of the terms “combination therapy” or “combination” does not limit the order in which prophylactic and/or therapeutic agents are administered to a subject with the condition. In other words, combination therapy can be administered by treating with the therapeutic agents separately, sequentially, or simultaneously. In the case of “sequential administration,” the first administered agent may exert some physiological effect on the subject upon administration of the second agent or upon becoming active in the subject.

As used herein with respect to the administration of prophylactic and/or therapeutic agents, the term "simultaneous administration" means administration of the agents such that the individual agents are simultaneously present in the subject. Simultaneous administration can be achieved by molecules in a single composition or in separate compositions administered at the same or similar times. Sequential administration can be performed in any order as needed.

The term “differentiation cluster 3” or “CD3” refers to a polyprotein complex (historically known as the T3 complex) composed of four distinct polypeptide chains: epsilon (ε), gamma (γ), delta (δ), and zeta (ζ), which assemble and function as three pairs of dimers (εγ, εδ, ζζ). The CD3 complex acts as a T-cell co-receptor, non-covalently associating with the T-cell receptor (TCR) (Smith-Garvin et al. 2009) and generating activation signals in T lymphocytes. The CD3 protein complex is a defining characteristic of the T-cell lineage, and therefore CD3 antibodies can be effectively used as T-cell markers (Chetty and Gatter 1994). It is well known that CD3 antibodies induce the generation of cytotoxic T cells by activating endogenous lymphotropic mediators and can selectively kill tumor targets (Yun et al., Cancer Research, 49: 4770-4774 (1989)).

T cells play a central role in cell-mediated immunity in humans and animals. The recognition and binding of specific antigens are mediated by TCRs expressed on the surface of T cells. T cell TCRs can interact with immunogenic peptides (epitopes) that bind major histocompatibility complex (MHC) molecules and are presented on the surface of target cells. Specific binding of the TCR triggers a signaling cascade within the T cell, leading to proliferation and differentiation into mature effector T cells. More specifically, T cells express TCR complexes capable of inducing antigen-specific immune responses (Smith-Garvin et al., 2009). Antigens are peptides expressed by tumor cells and virus-infected cells capable of stimulating immune responses. Intracellularly expressed antigens bind to class I major histocompatibility (MHC) molecules and are transported to the surface, where they are exposed to T cells. If the binding affinity of the TCR to the class I MHC complexed with the antigen is sufficient, the formation of an immune synapse is initiated. Signaling via the immune synapse is mediated by CD3 co-receptors that form ε/δ, δ/γ, and ζ/ζ dimers. These dimers are associated with the TCR and generate activation signals in T lymphocytes. This signaling cascade guides T cell-mediated killing of antigen-expressing cells. Cytotoxicity is mediated by the release of granzyme B and perforin from T cells and their translocation to target cells.

As used herein, “activated T-cell antigen” refers to an antigenic determinant expressed on the surface of T lymphocytes, specifically cytotoxic T lymphocytes, which can induce T-cell activation upon interaction with antigen-binding molecules. Specifically, the interaction between antigen-binding molecules and activated T-cell antigen can induce T-cell activation by triggering a signaling cascade of the T-cell receptor complex. In one specific embodiment, the activated T-cell antigen is CD3, specifically the ε subunit of CD3 (for human sequences, see UniProt number P07766 (version 130), NCBI RefSeq number NP_000724.1, SEQ ID NO: 66). Typically, naturally occurring allele variants have an amino acid sequence that is at least 95%, 97%, or 99% identical to the protein described in Genbank accession number BAB71849.1.

As used herein, "T cell activation" refers to one or more cellular responses selected from the following: proliferation, differentiation, cytokine secretion, release of cytotoxic effector molecules, cytotoxic activity, and expression of activation markers by T lymphocytes, specifically cytotoxic T lymphocytes. The T cell activation bispecific antigen-binding molecule of the present invention can induce T cell activation. Assays suitable for measuring T cell activation are known in the art described herein.

As used herein, “antibody that binds CD3,” “antibody that recognizes CD3,” “anti-CD3 antibody,” or “CD3 antibody” includes an antibody that specifically recognizes a single CD3 subunit (e.g., ε, δ, γ, or ζ) and its antigen-binding fragment, and an antibody that specifically recognizes a dimer complex of two CD3 subunits (e.g., γ/ε, δ/ε, and ζ/ζ CD3 dimers) and its antigen-binding fragment. Human CD3ε is indicated in Genbank accession number NM_000733. The antibodies of the present invention can bind soluble CD3 and/or CD3 expressed on the cell surface. Soluble CD3 includes native CD3 protein and recombinant CD3 protein variants (such as, for example, monomeric and dimer CD3 constructs) lacking a transmembrane domain or otherwise not associated with the cell membrane.

Antibodies targeting CD3 can generate activation signals in T lymphocytes. Other T cell activation ligands, including but not limited to CD28, CD134, CD137, and CD27, can also be used.

This invention includes a single-arm antibody that binds to CD3. As used herein, "single-arm antibody" means an antigen-binding molecule comprising a single antibody heavy chain and a single antibody light chain. The single-arm antibody of this invention may comprise any of the VL/VH or CDR amino acid sequences as described in Tables 1, 2 and 3 herein.

The CD3 antibodies disclosed herein may contain one or more amino acid substitutions, insertions, and/or deletions in the framework and/or CDR regions of the heavy and light chain variable domains, compared to the corresponding germline sequences from which the antibodies are derived. Such mutations can be readily identified by comparing the amino acid sequences disclosed herein with germline sequences available, for example, from public antibody sequence databases. This invention includes antibodies derived from any of the amino acid sequences disclosed herein and their antigen-binding fragments, wherein one or more amino acids in one or more framework and/or CDR regions are mutated to corresponding residues of the germline sequence from which the antibody is derived, or mutated to corresponding residues of another human germline sequence, or mutated to conserved amino acid substitutions of the corresponding germline residues (such sequence changes are collectively referred to herein as “germline mutations”). Those skilled in the art can readily generate a variety of antibody and antigen-binding fragments containing one or more individual germline mutations or combinations thereof, starting with the heavy and light chain variable region sequences disclosed herein. In some embodiments, all framework and/or CDR residues within the VH and/or VL domains are mutated back to residues found in the original germline sequence from which the antibody is derived. In some embodiments, only certain residues are mutated back to the original germline sequence, such as mutated residues found only in the first 8 amino acids of FR1 or the last 8 amino acids of FR4, or mutated residues found only in CDR1, CDR2, or CDR3. In some further embodiments, one or more of the framework and/or CDR residues are mutated to corresponding residues in a different germline sequence (i.e., a germline sequence different from the germline sequence from which the antibody originally originated).

Furthermore, the antibodies of the present invention may contain any combination of two or more germline mutations within the framework and/or CDR region, for example, wherein certain individual residues are mutated to corresponding residues of a specific germline sequence, while certain other residues different from the original germline sequence are maintained or mutated to corresponding residues of a different germline sequence. Once obtained, one or more desired properties of antibodies containing one or more germline mutations can be readily tested, such as improved binding specificity, increased binding affinity, improved or enhanced antagonistic or agonistic biological properties (as applicable), reduced immunogenicity, etc. Antibodies obtained in this general manner are covered within the present invention.

The present invention also includes CD3 antibodies comprising variants of any of the disclosed VH, VL, and/or CDR amino acid sequences having one or more conserved substitutions. For example, the present invention includes CD3 antibodies having VH, VL, and/or CDR amino acid sequences having, for example, 10 or fewer, 8 or fewer, 6 or fewer, 4 or fewer, or other conserved amino acid substitutions relative to any of the VH, VL, and/or CDR amino acid sequences described in Table 1 herein.

As used herein, the term "complex" or "composite" refers to the association of two or more molecules that interact with each other via bonds and/or forces that are not peptide bonds (e.g., van der Waals forces, hydrophobic forces, hydrophilic forces). In one embodiment, the complex is heteropolymeric. It should be understood that the terms "protein complex" or "peptide complex" as used herein include complexes having non-protein entities (e.g., including but not limited to chemical molecules, such as toxins or detection agents) conjugated to proteins in the protein complex.

As used herein, "CD3 antibody" includes monovalent antibodies with single specificity, as well as "bispecific antibody," "two-specific antibody," "three-specific antibody," "bifunctional antibody," "heteropolymer," "heteropolymer complex," "bispecific heterodimeric biantibody," "heteropolypeptide," or bispecific heterodimeric IgG. In some embodiments of the present invention, the CD3 antibody of the present invention is a human or humanized antibody.

As used herein, a "bispecific antibody" is a molecule comprising at least a first polypeptide and a second polypeptide, wherein the second polypeptide differs from the first polypeptide in its amino acid sequence by at least one amino acid residue. In some cases, a bispecific antibody is an artificial hybrid antibody having two distinct heavy chain regions and light chain regions. Preferably, a bispecific antibody has binding specificity to at least two different ligands, antigens, or binding sites. Therefore, a bispecific antibody can bind to two different antigens simultaneously. The two antigen-binding sites of a bispecific antibody can reside on two different epitopes on the same or different protein targets (e.g., tumor targets).

As used herein, the first and second polypeptide chains of a “bispecific antibody” contain at least one antibody VL and an antibody VH region or a fragment thereof, wherein both antibody-binding domains are contained within a single polypeptide chain and wherein the VL and VH regions in each polypeptide chain are derived from different antibodies.

Bispecific antibodies can be derived from full-length antibodies or antibody fragments (e.g., F(ab')2 bispecific antibodies). Bispecific antibodies are more fully described, for example, in EP404,097; WO93/11161; and Hollinger et al., Proc. Natl. Acad. Sci. USA 90:6444-6448, 1993. A bispecific antibody is a heterodimer of two “crossed” sFv fragments, in which the VH and VL regions of the two antibodies are located on different polypeptide chains.

As used herein, the terms “link,” “fusion,” “fusion product,” “covalent bonding,” “covalent coupling,” and “gene fusion” are used interchangeably. These terms refer to the linking of two or more elements or components together by any means, including chemical conjugation or recombination. As used herein, the term “covalent bonding” means that the specified parts are directly covalently bonded to each other or indirectly covalently bonded to each other via one or more insert parts (such as linking peptides or parts). In a preferred embodiment, the parts are covalently fused. One type of covalent bonding is the peptide bond. Methods of chemical conjugation (e.g., using heterobifunctional crosslinking agents) are known in the art. The fusion part can also be a gene fusion. As used herein, the terms “gene fusion,” “gene link,” or “gene fusion” refer to the expression of a gene via a single polynucleotide molecule encoding two or more proteins, polypeptides, or fragments, which are collinearly, covalently linked, or connected via their individual peptide backbones. This genetic fusion results in the expression of a single adjacent gene sequence. Preferred gene fusions are in-frame fusions, meaning two or more open reading frames (ORFs) fuse in a manner that maintains the correct reading frame of the original ORF to form a continuously longer ORF. Therefore, the resulting recombinant fusion protein is a single polypeptide containing two or more protein segments corresponding to the polypeptides encoded by the original ORFs (segments that would not normally bind in this way naturally). In this case, the single polypeptide is cleaved during treatment to produce a dimer molecule containing two polypeptide chains.

As used herein, the terms "linked" or "links" refer to a direct peptide bond between first and second amino acid sequences, or a bond involving a third amino acid sequence peptide bonded to and between the first and second amino acid sequences. For example, a linker peptide is bonded to the C-terminus of one amino acid sequence and the N-terminus of another amino acid sequence.

As used herein, the term "linker" refers to an amino acid sequence of two or more amino acids in length. Linkers can consist of neutral polar or nonpolar amino acids. The length of a linker can be, for example, from 2 to 100 amino acids, such as between 2 and 50 amino acids, such as 3, 5, 10, 15, 20, 25, 30, 35, 40, 45, or 50 amino acids. Linkers can be "cleavable," for example, by self-cleavage, enzymatic cleavage, or chemical cleavage. Cleavage sites in amino acid sequences, and enzymes and chemicals that cleave at such sites, are well known in the art and are also described herein.

As used herein, the term "disulfide bond" or "cysteine-cysteine disulfide bond" refers to a covalent interaction between two cysteine residues, wherein the sulfur atom of the cysteine is oxidized to form a disulfide bond. The average bond energy of a disulfide bond is approximately 60 kcal/mol, compared to 1-2 kcal/mol for hydrogen bonds. In the context of this invention, the cysteine residues forming the disulfide bond are located within the framework region of a single-chain antibody and serve to stabilize the antibody's conformation. Cysteine residues can be introduced, for example, through site-directed mutagenesis, to generate stable disulfide bonds intramolecularly.

The term "the name of the pestle in the mortar" is similar to the term "protrusion and cavity" and they can be used interchangeably.

A "protrusion" or "pole" refers to at least one of the following amino acid side chains: extending from the interface of the first polypeptide and thus able to be positioned in a compensating cavity in a neighboring interface (i.e., the interface of the second polypeptide) to stabilize the heterodimer, thereby, for example, making heterodimer formation superior to homodimer formation. The protrusion may be present in the original interface or may be introduced synthetically (e.g., by altering the nucleic acid encoding the interface). Typically, the nucleic acid encoding the interface of the first polypeptide is modified to encode the protrusion. To achieve this, the nucleic acid encoding at least one "original" amino acid residue in the interface of the first polypeptide is replaced with an "input" amino acid residue encoding at least one side chain with a volume larger than the original amino acid residue. The upper limit for the number of original residues replaced is the total number of residues in the interface of the first polypeptide. Certain input residues used to form the protrusion are typically naturally occurring amino acid residues and are preferably selected from arginine (R), phenylalanine (F), tyrosine (Y), and tryptophan (W).

The protrusions or cavities are "localized" within cavities or sockets, meaning that the spatial location and size of the protrusions and cavities at the interface between the first and second peptides, respectively, allow the protrusions to be located within the cavities without significantly disrupting the normal association of the first and second peptides at the interface. Since protrusions, such as phenylalanine (F), tyrosine (Y), and tryptophan (W), typically do not extend perpendicularly from the interface axis, the comparison of protrusions to their corresponding cavities relies on modeling the protrusion/cavity pairs based on three-dimensional structures, such as those obtained by X-ray crystallography or nuclear magnetic resonance (NMR). This can be achieved using techniques well-established in the art.

A “cavity” or “mortise” refers to at least one amino acid side chain that recesses from the interface of the second polypeptide and thus accommodates a corresponding protrusion on the adjacent interface of the first polypeptide. The cavity may be present in the original interface or may be introduced synthetically (e.g., by altering the nucleic acid encoding the interface). Typically, the nucleic acid encoding the interface of the second polypeptide is altered to encode the cavity. To achieve this, the nucleic acid encoding at least one “original” amino acid residue in the interface of the second polypeptide is replaced with DNA encoding at least one “input” amino acid residue that has a side chain volume smaller than the original amino acid residue. The upper limit for the number of original residues replaced is the total number of residues in the interface of the second polypeptide. Some of the input residues used to form the cavity are typically naturally occurring amino acid residues and are preferably selected from alanine (A), serine (S), threonine (T), and valine (V).

As used herein, “target antigen,” “target cell antigen,” “tumor antigen,” or “tumor-specific antigen” refers to antigenic determinants that are presented on the surface of target cells (e.g., cells in a tumor, such as cancer cells) or cells in the tumor stroma).

As used herein, the term “cancer” or “cancerous” refers to or describes a physiological condition in mammals characterized by uncontrolled cell growth, growths or tumors of abnormal, uncontrolled growth originating from cells. In some respects, cancer refers to a malignant primary tumor that has not metastasized and remains localized. In other respects, cancer refers to a malignant tumor that has invaded and destroyed adjacent body structures and spread to distant sites. In some respects, cancer is associated with a specific cancer antigen. Examples of cancer include, but are not limited to, cancers of the following sites: mouth and pharynx, digestive system (e.g., esophagus, stomach, small intestine, colon, rectum, liver, gallbladder, bile ducts, and pancreas), respiratory system (e.g., larynx, lungs, and bronchi, including non-small cell lung cancer), bone and joints (e.g., bone metastases), soft tissue, skin (e.g., melanoma), breast, reproductive system (e.g., cervix, uterus, ovary, vulva, vagina, prostate, and testes), urinary system (e.g., bladder, kidneys, renal pelvis, and ureters), eye and orbit, brain and nervous system (e.g., glioma), head and neck, or endocrine system (e.g., thyroid). Cancer can also be lymphoma (such as Hodgkin's disease and non-Hodgkin's lymphoma), multiple myeloma, or leukemia (such as acute lymphoblastic leukemia, chronic lymphoblastic leukemia, acute myeloid leukemia, chronic myeloid leukemia, etc.). In some implementations, cancers of the digestive system are cancers of the esophagus, stomach, small intestine, colon, rectum, liver, gallbladder, bile ducts, and pancreas.

Regarding cancer, treatment is useful in any one or more of the following: (a) treating, preventing, or improving one or more symptoms of a condition associated with malignant cells in a subject (e.g., gastrointestinal cancers such as colorectal cancer (CRC)); (b) inhibiting tumor growth or progression in a subject with malignant tumors expressing tumor antigens; (c) inhibiting metastasis of cancer (malignant) cells expressing tumor antigens in a subject with one or more malignant cells expressing tumor antigens; (d) inducing regression (e.g., long-term regression) of tumors expressing tumor antigens; (e) exerting cytotoxic activity in malignant cells expressing tumor antigens; (f) increasing progression-free survival in a subject with a tumor antigen-associated condition; (g) increasing overall survival in a subject with a tumor antigen-associated condition; (h) reducing the use of additional chemotherapy agents or cytotoxic agents in a subject with a tumor antigen-associated condition; (i) reducing tumor burden in a subject with a tumor antigen-associated condition; or (j) blocking the interaction of tumor antigens with other unidentified factors. While not wishing to be bound by theory, the treatment is believed to have the ability to inhibit, ablate, or kill cells or otherwise reduce cell-mediated conditions (such as cancer) in vitro or in vivo.

As used in this article, the terms “malignant cell” or “malignant disease” refer to an aggressive and/or metastatic tumor or tumor cell, i.e., cancer cells.

Although any materials and methods similar to or equivalent to those described herein may be used in the practice or testing of this invention, preferred materials and methods are now described.

Materials and methods

Various techniques for antibody production have been described, including conventional hybridoma methods for preparing monoclonal antibodies, recombinant techniques for preparing antibodies (including chimeric antibodies, such as humanized antibodies), antibody production in transgenic animals, and the recently described phage display technique for preparing "fully human" antibodies. These techniques are briefly described below.

Polyclonal antibodies against a target antigen are typically generated in animals through multiple subcutaneous (sc) or intraperitoneal (ip) injections of the antigen and adjuvant. Using bifunctional reagents or derivatizing agents, such as maleimide benzoyl sulfosuccinimide (conjugated via cysteine residues) or N-hydroxysuccinimide (conjugated via lysine residues), the antigen (or a fragment containing the target amino acid sequence) can be conjugated to a protein (which is immunogenic in the species to be immunized, such as serum albumin, bovine thyroglobulin, or soybean trypsin inhibitors). Animals are immunized against the immunogenic conjugate or derivative, and booster immunizations are performed several weeks later via subcutaneous injections at multiple sites. Seven to 14 days later, blood is drawn from the animals, and serum antibody titers are measured. Booster immunizations are continued until a titer plateau is reached. Preferably, booster immunizations are performed using conjugates of the same antigen (but conjugated to different proteins and/or via different cross-linking agents). Conjugates can also be prepared as protein fusions in recombinant cell cultures. In addition, agglomerating agents (such as alum) are used to enhance the immune response.

Monoclonal antibodies can be obtained from a substantially homogeneous population of antibodies using the hybridoma method first described by Kohler and Milstein, Nature 256:495, 1975, or can be prepared via a recombinant DNA method (Cabilly et al., U.S. Patent No. 4,816,567). In the hybridoma method, immunization of mice or other suitable host animals, as described above, induces the production or ability to produce lymphocytes that will specifically bind to proteins used for immunization. Alternatively, immunization of lymphocytes can be performed in vitro. The lymphocytes are then fused with myeloma cells using a suitable fusion agent (such as polyethylene glycol) to form hybridoma cells (Goding, Monoclonal Antibodies: Principles and Practice, pp. 59-103 (Academic Press, 1986). The hybridoma cells thus prepared are seeded in a suitable culture medium, preferably containing one or more substances that inhibit the growth or survival of unfused parental myeloma cells, and allowed to grow therein. Alternatively, the hybridoma cells can grow in vivo as ascites tumors in animals. The monoclonal antibodies secreted by the subclones are appropriately separated from the culture medium, ascites, or serum using routine immunoglobulin purification procedures such as, for example, protein A-Sepharose, hydroxyapatite chromatography, gel electrophoresis, dialysis, or affinity chromatography.

The expression of antibodies, antigen-binding fragments of antibodies, or any antibody construct can be carried out in appropriate prokaryotic or eukaryotic host cells (such as CHO cells, NSO cells, SP2/0 cells, HEK293 cells, COS cells, PER.C6 cells, yeast or E. coli cells), and the secreted antibodies can be recovered and harvested from said cells (cell culture supernatant, regulated cell culture supernatant, cell lysate, or clarified bulk material). Common methods for generating antibodies are well known in the art and described in, for example, the following review articles: Makrides, S.C., Protein Expr. Purif. 17:183-202, 1999; Geisse, S., et al., Protein Expr. Purif. 8:271-282, 1986; Kaufman, R.J., MoI. Biotechnol. 16:151-160, 2000; Werner, R.G., Drug Res. 48:870-880, 1998. In one specific embodiment, the cell culture is a mammalian cell culture, such as a Chinese hamster ovary (CHO) cell culture.

In one embodiment, the antibody may be recovered or isolated from a naturally occurring source. In some such embodiments, the isolated or recovered antibody may undergo additional purification steps using conventional chromatographic methods known in the art. Specifically, the purification methods are contemplated, but are not limited to, affinity chromatography (e.g., protein A affinity chromatography), ion exchange chromatography (e.g., anion exchange chromatography or cation exchange chromatography), hydrophobic interaction chromatography, hydroxyapatite chromatography, gel filtration chromatography, and/or dialysis. Of these, protein A chromatography is the preferred purification method. The matrix for affinity ligand linkage is most often agarose, but other matrices may be used.

Other techniques for antibody purification are also known in the art, such as fractionation on ion exchange columns, ethanol precipitation, reversed-phase high-performance chromatography, reversed-phase HPLC, chromatography on silica, chromatography on heparin Sepharose™, chromatography on anion or cation exchange resins (such as polyaspartic acid columns), chromatographic focusing, electrophoresis using SDS-PAGE, and ammonium sulfate precipitation. The list of purification methods above is merely illustrative in nature and is not intended to be limiting.

Alternatively, transgenic animals (e.g., mice) can be produced that, upon immunization, generate a complete human antibody library in the absence of endogenous immunoglobulin production. For example, homozygous deletion of the antibody heavy chain linker (J.sub.H) gene in chimeric and germline mutant mice has been described as resulting in complete suppression of endogenous antibody production. Transplantation of human germline immunoglobulin gene arrays into such germline mutant mice will lead to the production of human antibodies upon antigen challenge. (See, for example, Jakobovits et al., Proc. Natl. Acad. Sci. USA 90:2551-255;1993 and Jakobovits et al., Nature 362:255-258, 1993).

In some embodiments, the antibodies of the present invention may be humanized, retaining high affinity for the antigen and other beneficial biological properties. Methods for humanizing nonhuman antibodies are well known in the art. Humanization can be essentially performed following the approach of Winter and collaborators (Jones et al., Nature 321:522-525, 1986; Riechmann et al., Nature 332:323-327, 1988; Verhoeyen et al., Science 239:1534-1536, 1988) by replacing the corresponding sequence of a human antibody with a rodent CDR or CDR sequence. Thus, such humanized antibodies are chimeric antibodies (Cabilly, ibid.), in which essentially less than one fully human variable domain has been replaced by a corresponding sequence from a non-human species. In practice, humanized antibodies are typically human antibodies, with some CDR residues and possibly some FR residues replaced by residues from similar sites in rodent antibodies. Importantly, the antibody is humanized, retaining its high affinity for the antigen and other beneficial biological properties. To achieve this, humanized antibodies are prepared according to a preferred method by analyzing the parental sequence and various conceptual humanized products using a three-dimensional model of the parental and humanized sequences. Three-dimensional immunoglobulin models are familiar to those skilled in the art. A computer program can be used to illustrate and demonstrate the possible three-dimensional conformational structures of selected candidate immunoglobulin sequences. Observing these displays allows analysis of the possible roles of residues in the function of the candidate immunoglobulin sequence, i.e., analysis of residues affecting the ability of the candidate immunoglobulin to bind its antigen. In this way, FR residues can be selected and merged from common and input sequences to achieve desired antibody properties, such as increased affinity for the target antigen. For further details, see WO92/22653, published December 23, 1992.

Various phage display methods known in the art can also be used to generate the antibodies of the present invention. In phage display methods, functional antibody domains are displayed on the surface of phage particles carrying a polynucleotide sequence encoding them. In particular, this phage can be used to display antigen-binding domains (such as Fab and Fv or disulfide-stable Fv) expressed from a library or combinatorial antibody library (e.g., human or mouse). Phages expressing antigen-binding domains that bind to target antigens can be selected or identified using antigens (e.g., using labeled antigens or antigens that bind to or capture onto solid surfaces or beads). The phages used in these methods are typically filamentous phages, including fd and M13. The antigen-binding domains are expressed as proteins recombinantly fused to phage gene III or gene VIII proteins. Examples of phage display methods that can be used to prepare the immunoglobulins or fragments thereof of the present invention include those disclosed in Brinkmann et al., "Phage Display Of Disulfide-Stabilized FvFragments," J. Immunol. Methods, 182:41-50, 1995.

The functional properties of various IgG isoforms and their domains are well known in the art. The amino acid sequences of IgG1, IgG2, IgG3, and IgG4 are known in the art. The selection and/or combination of two or more domains from a particular IgG isoform in the methods used in this invention can be based on any known parameters of the parental isoform (including affinity for FcγR). For example, using regions or domains from IgG isoforms (e.g., IgG2 or IgG4) that exhibit limited or no binding to FcγRIIB can be particularly useful in cases where it is desired to engineer bispecific antibodies to maximize binding to activating receptors and minimize binding to inhibitory receptors. Similarly, using Fc chains or domains from IgG isoforms (e.g., IgG3) that are known to preferentially bind to C1q or FcγRIIIA can be combined with Fc amino acid modifications known in the art to enhance antibody-dependent cell-mediated cytotoxicity (ADCC) and/or complement-dependent cytotoxicity (CDC) to engineer bispecific antibodies such that effector functional activity (e.g., complement activation or ADCC) is maximized. In a similar manner, mutations can be made in the Fc chain or domain of the IgG isotype to minimize or eliminate the effector function of the Fc chain.

During the discovery process, antibody generation and characterization can elucidate information about the desired epitope. Based on this information, antibodies can be competitively screened for binding to the same epitope. This is achieved by conducting competitive and cross-competitive studies to find antibodies that compete or cross-compete with each other (e.g., antibodies that competitively bind to antigens). One approach is to identify the epitope that the antibody binds to, or “epitope localization.” Many methods known in the art exist for mapping and characterizing the location of epitopes on proteins, including resolving the crystal structure of antibody-antigen complexes, competitive assays, gene fragment expression assays, and synthetic peptide-based assays, as described, for example, in Chapter 11 of Harlow and Lane, Using Antibodies, a Laboratory Manual, Cold SpringHarbor Laboratory Press, Cold SpringHarbor, N.Y., 1999. In an additional instance, epitope mapping can be used to determine the sequence to which the antibody binds. Epitope mapping is commercially available from various sources, such as Pepscan Systems (Edelhertweg 15, 8219 PH Lelystad, Netherlands). In an additional instance, mutagenesis of the antigen-binding domain, domain exchange experiments, and alanine partitioning mutagenesis can be performed to identify residues required, sufficient, and/or essential for epitope binding. Binding affinity and the dissociation rate of the antigen-binding domain interaction can be determined by a competitive binding assay. An example of a competitive binding assay is a radioimmunoassay, which involves incubating a labeled antigen and detecting the molecule binding to the labeled antigen. The affinity and binding-dissociation rate of the molecules of the present invention for the antigen can be determined from saturation data using Scatchard analysis.

The affinity and binding properties of the antibodies of the present invention to antigens can first be determined using in vitro assays (biochemical or immunoassays) known in the art for antigen-binding domains, including but not limited to enzyme-linked immunosorbent assays (ELISA), surface plasmon resonance (SPR) assays, biolayer interferometry, or immunoprecipitation assays. The molecules of the present invention may have similar binding properties in in vivo models (such as those described and disclosed herein) to those in in vitro-based assays. However, the present invention does not exclude molecules of the present invention that do not exhibit the desired phenotype in in vitro-based assays but do exhibit the desired phenotype in vivo.

Once the nucleic acid sequence encoding the molecule of the present invention (i.e., the binding domain) has been obtained, a vector for generating the molecule can be produced using recombinant DNA technology, a technique well known in the art.

The polynucleotide encoding the antibody-binding domain of the present invention may include an expression control polynucleotide sequence operatively linked to the antibody-coding sequence, including naturally associated or heterologous promoter regions known in the art. The expression control sequence may be a eukaryotic promoter system in a vector capable of transforming or transfecting eukaryotic host cells, but a prokaryotic host control sequence may also be used. Once the vector has been incorporated into a suitable host cell line, the host cells are allowed to proliferate under conditions suitable for expressing the nucleotide sequence and are used, as needed, to collect and purify the antibody. Eukaryotic cell lines include CHO cell lines, various COS cell lines, HeLa cells, myeloma cell lines, transformed B cells, or human embryonic kidney cell lines.

In some embodiments, the DNA encoding the antibodies of the present invention is isolated and sequenced using conventional procedures, e.g., by using oligonucleotide probes capable of specifically binding to genes encoding the heavy and light chains of mouse antibodies. Hybridoma cells of the present invention serve as a preferred source of this DNA. Once isolated, the DNA can be placed in an expression vector, which is then transfected into host cells that otherwise do not produce immunoglobulin proteins (such as simian COS cells, CHO cells, or myeloma cells) to obtain monoclonal antibody synthesis in recombinant host cells. The DNA can also be modified, for example, by replacing the homologous mouse sequences with coding sequences for the constant domains of the human heavy and light chains (Morrison et al., Proc. Nat. Acad. Sci. 81:6851, 1984). In this manner, chimeric antibodies having the binding specificity of the anti-antigen monoclonal antibodies described herein are prepared.

CD3 antibody

In one aspect, the present invention provides a reagent that specifically binds to CD3 (e.g., the full-length human CD3 ε subunit (e.g., accession number NP_000724 or SEQ ID NO: 66)).

In some implementations, the CD3 antibody specifically binds to human CD3.

Exemplary CD3 antibodies of the present invention are listed in Tables 1, 2, and 3 herein. Table 1 shows the amino acid sequence identifiers of the VH and VL regions. In some embodiments, antibodies are provided having any of the partial light chain sequences listed in Table 1 and/or any of the partial heavy chain sequences listed in Table 1.

Table 1

In Table 1, the underlined sequences are CDR sequences based on Kabat, and the bold sequences are CDR sequences based on Chothia.

This invention provides the CDR portion (including Chothia, Kabat CDR, and CDR contact region) of an antibody against CD3. Methods and techniques for identifying CDRs within VH and VL amino acid sequences are well known in the art and can be used to identify CDRs within the specified VH and/or VL amino acid sequences disclosed herein. Exemplary conventions for identifying CDR boundaries include, for example, the Kabat definition, the Chothia definition, and the AbM definition. Generally, the Kabat definition is based on sequence variability, the Chothia definition is based on the location of structural loop regions, and the AbM definition is a compromise between the Kabat and Chothia methods. See, for example, Kabat, "Sequences of Proteins of Immunological Interest," National Institutes of Health, Bethesda, Md. (1991); Al-Lazikani et al., J. Mol. Biol. 273:927-948 (1997); and Martin et al., Proc. Natl. Acad. Sci. USA 86:9268-9272 (1989). Public databases may also be used to identify CDR sequences within the antibody. It should be understood that in some embodiments, the CDR may be a combination of Kabat and Chothia CDRs (also referred to as a "combined CR" or "extended CDR"). In some embodiments, the CDR is the Kabat CDR. In some further embodiments, the CDR is the Chothia CDR. In other words, in an implementation having more than one CDR, the CDR can be any of Kabat, Chothia, combined CDRs, or a combination thereof. Tables 2 and 3 provide examples of the CDR sequences provided herein.

In one embodiment, the present invention provides an antibody comprising the amino acid sequences listed in Table 1 or substantially similar sequences having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with them.

In some such embodiments, the present invention provides an antibody comprising a VL region comprising an amino acid sequence of any of the VL region amino acid sequences listed in Table 1 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with it.

In some other embodiments of this invention, the present invention provides an antibody comprising a VH region comprising an amino acid sequence of any of the VH region amino acid sequences listed in Table 1 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with it.

In one embodiment, an antibody that specifically binds to CD3 is provided, wherein the antibody comprises: (a) a heavy chain variable (VH) region comprising VH complementarity-determining region 1 (VH CDR1), VH complementarity-determining region 2 (VH CDR2), and VH complementarity-determining region 3 (VH CDR3) of the VH sequence shown in SEQ ID NO: 2, 4, 5, 7, 10, 12, or 35; and/or (b) a light chain variable (VL) region comprising VL complementarity-determining region 1 (VL CDR1), VL complementarity-determining region 2 (VL CDR2), and VL complementarity-determining region 3 (VL CDR3) of the VL sequence shown in SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87, or 89.

In one embodiment, the present invention provides an antibody comprising a set of six CDRs (i.e., VL CDR1, VL CDR2, VLCDR3, VH CDR1, VH CDR2, VH CDR3), wherein the CDRs are contained within the VL/VH amino acid sequence pairs as defined by any of the exemplary CD3 antibodies listed in Table 1.

In specific embodiments, the present invention provides an antibody comprising a set of amino acid sequences of VL CDR1, VL CDR2, VL CDR3, VHCDR1, VHCDR2, or VH CDR3, wherein the set of amino acid sequences is contained within a VL/VH amino acid sequence pair selected from the following: SEQ ID NO: 1 and 2 (e.g., 2B5-0001 (2B5)); SEQ ID NO: 3 and 2 (2B5-0002); SEQ ID NO: 3 and 4 (e.g., 2B5-0006 (2B5v6)); SEQ ID NO: 3 and 5 (e.g., 2B5-0009); SEQ ID NO: 6 and 7 (e.g., 2B5-0517); SEQ ID NO: 8 and 7 (e.g., 2B5-0522); SE QID NO: 6 and 4 (e.g., 2B5-0533); SEQ ID NO: 8 and 4 (e.g., 2B5-0538); SEQ ID NO: 9 and 10 (e.g., 2B5-0598); SEQ ID NO: 9 and 7 (e.g., 2B5-0623); SEQ ID NO: 11 and 12 (e.g., 2B5-0707); SEQ ID NO: 34 and 35 (e.g., 2B5-0003); SEQ ID NO: 85 and 2 (e.g., H2B4); SEQ ID NO: 9 and 4 (e.g., 2B5-1038); SEQ ID NO: 87 and 4 (e.g., 2B5-1039); and SEQ ID NO: 89 and 4 (e.g., 2B5-1040).

Table 2 shows the VH CDR1, VH CDR2, and VH CDR3 regions of an exemplary CD3 antibody.

Table 2

In one embodiment, the present invention provides an antibody comprising VH CDR1, said VH CDR1 comprising an amino acid sequence of any of the VHCDR1 amino acid sequences listed in Table 2 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In another embodiment, the present invention provides an antibody comprising VH CDR2, said VH CDR2 comprising an amino acid sequence of any of the VHCDR2 amino acid sequences listed in Table 2 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In yet another embodiment, the present invention provides an antibody comprising VH CDR3, said VH CDR3 comprising an amino acid sequence of any of the VHCDR3 amino acid sequences listed in Table 2 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

Table 3 shows exemplary CD3 antibodies VL CDR1, VL CDR2, and VL CDR3.

Table 3

In one embodiment, the present invention provides an antibody comprising VL CDR1, said VL CDR1 comprising an amino acid sequence of any of the VLCDR1 amino acid sequences listed in Table 3 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In another embodiment, the present invention provides an antibody comprising VL CDR2, said VL CDR2 comprising an amino acid sequence of any of the VLCDR2 amino acid sequences listed in Table 3 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In yet another embodiment, the present invention provides an antibody comprising VL CDR3, said VL CDR3 comprising an amino acid sequence of any of the VLCDR3 amino acid sequences listed in Table 3 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In some embodiments, an antibody that specifically binds to CD3 is provided, wherein the antibody comprises: (a) a heavy chain (HC) variable (VH) region comprising VH complementarity-determining region 1 (CDR1), VH complementarity-determining region 2 (VH CDR2), and VH complementarity-determining region 3 (VH CDR3) of VH having an amino acid sequence selected from SEQ ID NO: 2, 4, 5, 7, 10, 12, and 35; and/or (b) a light chain (LC) variable (VL) region comprising VL complementarity-determining region 1 (VL CDR1), VL complementarity-determining region 2 (VL CDR2), and VL complementarity-determining region 3 (VLCDR3) of VL having an amino acid sequence selected from SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87, and 89.

In a specific embodiment, the antibody comprises: (a) a VH region comprising (i) a VH CDR1 comprising the sequence of SEQ ID NO: 13, 14 or 15; (ii) a VHCDR2 comprising the sequence of SEQ ID NO: 16, 17, 19, 20, 21, 22, 23 or 24; and (iii) a VH CDR3 comprising the sequence of SEQ ID NO: 18 or 25; and/or (b) a VL region comprising (i) a VL CDR1 comprising the sequence of SEQ ID NO: 26, 29, 30, 32, 91 or 92; (ii) a VL CDR2 comprising the sequence of SEQ ID NO: 27 or 31; and (iii) a VL CDR3 comprising the sequence of SEQ ID NO: 28 or 33.

In some embodiments, the present invention provides antibodies that bind to CD3 and compete with antibodies as described herein, including 2B5-0001 (2B5), 2B5-0002, 2B5-0006 (2B5v6), 2B5-0009, 2B5-0517, 2B5-0522, 2B5-0533, 2B5-0538, 2B5-0598, 2B5-0623, 2B5-0707, 2B5-0003, 2B5-1038, 2B5-1039, and 2B5-1040.

In some embodiments, the present invention also provides the CDR portion of a CD3 antibody based on a CDR contact region. A CDR contact region is an antibody region that imparts specificity to an antigen to the antibody. Typically, the CDR contact region comprises residue sites within the CDR and Vernier region, constrained to maintain a suitable loop structure for antibody binding to specific antigens. See, for example, Makabe et al., J. Biol. Chem., 283:1156-1166, 2007. Determination of the CDR contact region is entirely within the scope of the art.

Some of the CD3 antibodies described herein may be referred to by more than one name. For example, 2B5-0001 may be referred to as 2B5 or h2B5; and 2B5-0006 may be referred to as 2B5v6 or h2B5v6. Table 4 shows the forms of the antibodies and the CD3 antibody regions present in the antibodies disclosed herein.

Table 4

sequence	describe	form	CD3 antibody region
				2B5-0001 (2B5)	CD3 antibody	IgG	2B5-0001
2B5-0002	CD3 antibody	IgG	2B5-0002
				2B5-0003	CD3 antibody	IgG	2B5-0003
2B5-0006 (2B5v6)	CD3 antibody	IgG	2B5-0006
				2B5-0009	CD3 antibody	IgG	2B5-0009
2B5-0517	CD3 antibody	IgG	2B5-0517
				2B5-0522	CD3 antibody	IgG	2B5-0522
2B5-0533	CD3 antibody	IgG	2B5-0533
				2B5-0538	CD3 antibody	IgG	2B5-0538
2B5-0598	CD3 antibody	IgG	2B5-0598
				2B5-0623	CD3 antibody	IgG	2B5-0623
2B5-0707	CD3 antibody	IgG	2B5-0707
				2B5-1038	CD3 antibody	IgG	2B5-1038
2B5-1039	CD3 antibody	IgG	2B5-1039
				2B5-1040	CD3 antibody	IgG	2B5-1040
H2B4	CD3 antibody	IgG	H2B4
				GUCY2C-H2B4	CD3-GUCY2c bispecific antibody	CD3 bispecific	H2B4
GUCY2C-2B5	CD3-GUCY2c bispecific antibody	CD3 bispecific	2B5-0001
				GUCY2C-2B5v6	CD3-GUCY2c bispecific antibody	CD3 bispecific	2B5-0006
xFLT3-2B5v6	CD3-FLT3 bispecific antibody	CD3 bispecific	2B5-0006

The binding affinity ( _KD ) of the antibody described herein for CD3 (such as human CD3ε (e.g., SEQ ID NO:40) can be from about 0.001 nM to about 500 nM. In one embodiment, the binding affinity is about any of the following: 6500 nM, 6000 nM, 5500 nM, 4500 nM, 4000 nM, 3500 nM, 3000 nM, 2500 nM, 2000 nM, 1500 nM, 1000 nM, 750 nM, 500 nM, 400 nM, 300 nM, 250 nM, 200 nM, 150 nM, 100 nM, 75 nM, 50 nM, 45 nM, 40 nM, 35 nM, 30 nM, 25 nM, 20 nM, 19 nM, 17 nM, 16 nM, 15 nM, 10 nM, 8 nM, 7.5 nM, 7 nM, 6.5 nM, 6 nM, 5.5 nM, 5 nM, 4 nM, 3 nM, 2 nM, 1 nM, 0.5 nM, 0.3 nM, 0.1 nM, 0.01 nM, 0.002 nM, or 0.001 nM. In some embodiments, the binding affinity is at least about any one of the following: 6500 nM, 6000 nM, 5500 nM, 5000 nM, 4000 nM, 3000 nM, 2000 nM, 1000 nM, 900 nM, 800 nM, 500 nM, 250 nM, 200 nM, 100 nM, 50 nM, 30 nM, 20 nM, 10 nM, 7.5 nM, 7 nM, 6.5 nM, 6 The effective values are nM, 5 nM, 4.5 nM, 4 nM, 3.5 nM, 3 nM, 2.5 nM, 2 nM, 1.5 nM, 1 nM, 0.5 nM, 0.1 nM, 0.05 nM, 0.01 nM, or lower. In one specific embodiment, the antibody has a _KD of about 80 to about 200 pM. In another specific embodiment, the antibody has a _KD of about 10 nm to 2000 nm.

In one aspect, the present invention provides an antibody or pharmaceutical composition for use in a method of modulating a T-cell-mediated immune response in a subject with such a need. In a specific embodiment, the present invention provides an antibody or pharmaceutical composition for use in a method of inhibiting the growth of tumor cells in a subject.

In another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, optionally wherein said cancer is selected from: breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer, bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and cancers of the digestive system.

In yet another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, wherein a T-cell-mediated immune response is modulated or wherein the growth of tumor cells is inhibited.

The present invention also provides nucleic acid molecules encoding CD3 antibodies or portions thereof. For example, the present invention provides nucleic acid molecules encoding any VH or VL amino acid sequence listed in Table 1 or substantially similar sequences having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with them.

The present invention also provides nucleic acid molecules of any of the amino acid sequences of VH CDR1, VH CDR2 or VH CDR3 listed in Table 2 or substantially similar sequences having at least 90%, at least 95%, at least 98% or at least 99% sequence identity with them.

The present invention also provides nucleic acid molecules having a substantially similar sequence to any of the VL CDR1, VL CDR2 or VL CDR3 amino acid sequences listed in Table 3, or having at least 90%, at least 95%, at least 98% or at least 99% sequence identity with it.

The present invention also provides a nucleic acid molecule encoding HC VR, wherein the VH comprises a set of three CDRs (i.e., VH CDR1, VHCDR2, or VH CDR3), wherein the amino acid sequence sets of VH CDR1, VH CDR2, and VH CDR3 are as defined by any of the exemplary CD3 antibodies listed in Table 2.

The present invention also provides a nucleic acid molecule encoding LC VR, wherein the VL comprises a set of three CDRs (i.e., VL CDR1, VLCDR2, or VL CDR3), wherein the amino acid sequence sets of VL CDR1, VL CDR2, and VL CDR3 are as defined by any of the exemplary CD3 antibodies listed in Table 3.

The present invention also provides nucleic acid molecules encoding both VH and VL, wherein the VH comprises an amino acid sequence of any VH amino acid sequence listed in Table 1, and wherein the VL comprises an amino acid sequence of any VL amino acid sequence listed in Table 1.

The present invention also provides recombinant expression vectors capable of expressing polypeptides containing heavy or light chain variable regions of CD3 antibodies. For example, the present invention includes recombinant expression vectors comprising any of the nucleic acid molecules mentioned above (i.e., nucleic acid molecules encoding any of the VH and VL shown in Table 1 and/or CDR sequences as shown in Table 2 or 3). The scope of the present invention also includes host cells in which such vectors have been introduced, and methods for producing said antibodies or portions thereof by culturing said host cells under conditions allowing the production of antibodies or antibody fragments and recovering said antibodies and antibody fragments thus produced.

In some embodiments, the polynucleotide comprises a sequence encoding the heavy chain and/or light chain variable region of the following antibodies listed in Table 4: 2B5-0001 (2B5), 2B5-0002, 2B5-0006 (2B5v6), 2B5-0009, 2B5-0517, 2B5-0522, 2B5-0533, 2B5-0538, 2B5-0598, 2B5-0623, 2B5-0707, 2B5-0003, 2B5-1038, 2B5-1039, or 2B5-1040.

The sequence encoding the target antibody can be maintained in a vector within a host cell, and the host cell can then be amplified and frozen for future use. Vectors (including expression vectors) and host cells are further described herein.

Polynucleotides complementary to any such sequence are also covered by this invention. In one aspect, this invention provides a method for preparing any of the polynucleotides described herein. The polynucleotide may be single-stranded (coding or antisense) or double-stranded, and may be DNA (genomic, cDNA, or synthetic) or RNA molecules. RNA molecules include HnRNA molecules containing introns and corresponding one-to-one with DNA molecules; and mRNA molecules that do not contain introns. Additional coding or non-coding sequences may, but do not necessarily, be present in the polynucleotides of this invention, and the polynucleotides may, but do not necessarily, be linked to other molecules and/or supporting materials.

The polynucleotide may contain a native sequence (i.e., an endogenous sequence encoding an antibody or a portion thereof) or a variant thereof. Polynucleotide variants contain one or more substitutions, additions, deletions, and/or insertions such that the immunoreactivity of the encoded polypeptide is not reduced relative to the native immunoreactive molecule. The effect on the immunoreactivity of the encoded polypeptide can generally be evaluated as described herein. The variant preferably exhibits at least about 70% identity with the polynucleotide sequence encoding a native antibody or a portion thereof, more preferably at least about 80% identity, even more preferably at least about 90% identity, and most preferably at least about 95% identity.

The polynucleotides of this invention can be obtained using chemical synthesis, recombination methods, or PCR. Methods for the chemical synthesis of polynucleotides are well known in the art and need not be described in detail herein. Those skilled in the art can use the sequences provided herein and commercial DNA synthesizers to generate the desired DNA sequences.

To prepare polynucleotides using recombinant methods, a polynucleotide containing the desired sequence can be inserted into a suitable vector, which can then be introduced into a suitable host cell for replication and amplification, as discussed further herein. Polynucleotides can be inserted into host cells by any means known in the art. Cells are transformed by introducing exogenous polynucleotides through direct uptake, endocytosis, transfection, F-conjugation, or electroporation. Once introduced, the exogenous polynucleotide can be maintained within the cell as a non-integrating vector (such as a plasmid) or integrated into the host cell genome. The amplified polynucleotides can then be isolated from the host cell using methods well-known in the art. See, for example, Sambrook et al., 1989.

Alternatively, PCR allows for the replication of DNA sequences. PCR technology is well known in the art and is described in U.S. Patent Nos. 4,683,195, 4,800,159, 4,754,065, and 4,683,202, and PCR: The Polymerase Chain Reaction, edited by Mullis et al., Birkauswer Press, Boston, 1994.

RNA can be obtained by using DNA isolated in a suitable vector and inserted into a suitable host cell. Once the cell replicates and the DNA is transcribed into RNA, RNA can then be isolated using methods well known to those skilled in the art, such as Sambrook et al., 1989, as described above.

Suitable cloning vectors can be constructed according to standard techniques or selected from a large number of cloning vectors available in the art. While the chosen cloning vector can vary depending on the intended host cell, a useful cloning vector will generally be capable of self-replication, may have a single target for a specific restriction endonuclease, and/or may carry genes related to markers that can be used to select clones containing that vector. Suitable examples include plasmids and bacterial viruses such as pUC18, pUC19, Bluescript (e.g., pBS SK+) and its derivatives, mp18, mp19, pBR322, pMB9, ColE1, pCR1, RP4, phage DNA, and shuttle vectors such as pSA3 and pAT28. These and many other cloning vectors are available from commercial vendors such as BioRad, Strategene, and Invitrogen.

Expression vectors are typically reproducible polynucleotide constructs containing polynucleotides according to the invention. This implies that the expression vector must be replicable in the host cell as an episome or as an integral part of chromosomal DNA. Suitable expression vectors include, but are not limited to, plasmids, viral vectors including adenoviruses, adeno-associated viruses, retroviruses, kinases, and the expression vectors disclosed in PCT Publication No. WO 87/04462. Vector components may typically include, but are not limited to, one or more of the following: a signal sequence; an origin of replication; one or more marker genes; and suitable transcriptional control elements (such as promoters, enhancers, and terminators). For expression (i.e., translation), one or more translational control elements, such as ribosome binding sites, translation initiation sites, and stop codons, are also typically required.

Vectors containing the target polynucleotide can be introduced into host cells by any of a number of suitable means, including electroporation, transfection with calcium chloride, rubidium chloride, calcium phosphate, DEAE-glucan, or other substances; particle bombardment; lipid transfection; and infection (e.g., when the vector is an infectious pathogen such as vaccinia virus). The choice of vector or polynucleotide to be introduced will often depend on the characteristics of the host cell.

This invention also provides host cells comprising any of the polynucleotides described herein. Any host cell capable of overexpressing heterologous DNA can be used for the purpose of isolating genes encoding target antibodies, peptides, or proteins. Non-limiting examples of mammalian host cells include, but are not limited to, COS, HeLa, and CHO cells. See also PCT Publication WO 87/04462. Suitable non-mammalian host cells include prokaryotes (such as Escherichia coli or Bacillus subtilis ) and yeasts (such as Saccharomyces cerevisiae , Saccharomyces pombe , or Kluyveromyces lactis ). Preferably, the host cell expresses cDNA at a level about 5-fold higher, more preferably 10-fold higher, and even more preferably 20-fold higher than the corresponding endogenous antibody or protein CD3 or CD3 domain (e.g., domains 1-4), as determined by immunoassay or FACS. Cells overexpressing target antibodies or proteins can be identified.

This invention also covers scFvs of the antibodies of this invention. Single-chain variable region fragments are prepared by linking light and/or heavy chain variable regions using a short linker peptide (Bird et al., Science 242:423-426, 1988). Examples of linker peptides include the sequence of SEQ ID NO: 65. In one embodiment, the linker peptide bridges approximately 3.5 nm between the carboxyl terminus of one variable region and the amino terminus of another variable region. Linkers with other sequences have been designed and used (Bird et al., Science 242:423-426, 1988). Linkers should be short, flexible polypeptides and preferably contain fewer than about 20 amino acid residues. The linker can then be modified for additional functions, such as linking drugs or solid carriers. Single-chain variants can be generated recombinantly or synthetically. An automated synthesizer can be used to synthesize scFvs. To generate scFv through recombination, a suitable plasmid containing a polynucleotide encoding the scFv can be introduced into a suitable host cell, which can be eukaryotic (such as yeast, plant, insect, or mammalian cells) or prokaryotic (such as Escherichia coli). The polynucleotide encoding the target scFv can be prepared by conventional manipulation (such as linking the polynucleotide). The obtained scFv can be isolated using standard protein purification techniques known in the art.

In one aspect, polynucleotide sequences encoding any of the partial light chain sequences and/or any of the partial heavy chain sequences disclosed herein are provided. Exemplary CD3 polynucleotides encoding the VH and VL regions of the antibodies of the present invention are listed in Table 5.

Table 5

In one aspect, the present invention provides compositions (such as pharmaceutical compositions) comprising any of the polynucleotides of the present invention. In some embodiments, the composition may comprise any of the polynucleotides shown in Table 5. In a particular embodiment, the composition comprises an expression vector containing a polynucleotide encoding any of the antibodies described herein (e.g., any of the polynucleotides shown in Table 5).

This invention also covers polynucleotides complementary to any such sequence. Polynucleotides can be single-stranded (coding or antisense) or double-stranded, and can be DNA (genomic, cDNA, or synthetic) or RNA molecules. RNA molecules include mature and immature mRNAs, such as precursor mRNA (pre-mRNA) or heteronuclear mRNA (hnRNA) and mature mRNA. Additional coding or non-coding sequences may, but do not necessarily, be present within the polynucleotides of this invention, and the polynucleotides may, but do not necessarily, be linked to other molecules and/or supporting materials.

The polynucleotide may contain a native sequence (i.e., an endogenous sequence encoding an antibody or a portion thereof) or a variant thereof. Polynucleotide variants contain one or more substitutions, additions, deletions, and/or insertions such that the immunoreactivity of the encoded polypeptide is not reduced relative to the native immunoreactive molecule. The effect on the immunoreactivity of the encoded polypeptide can generally be evaluated as described herein. Variants preferably exhibit at least about 70% identity with the polynucleotide sequence encoding a native antibody or a portion thereof, more preferably at least about 80% identity, even more preferably at least about 90% identity, and most preferably at least about 95% identity.

Two polynucleotide or polypeptide sequences are said to be "identical" if the nucleotide or amino acid sequences in the two sequences are identical when compared for maximum correspondence as described below. Comparisons between two sequences are typically performed by comparing sequences on a comparison window to identify and compare local regions of sequence similarity. As used herein, a "comparison window" refers to a segment of at least about 20, typically 30 to about 75, or 40 to about 50 adjacent positions, where, after optimal alignment of the two sequences, the sequence can be compared to a reference sequence having the same number of adjacent positions.

Using default parameters, the Megalign procedure in the Lasergene Bioinformatics Software Suite (DNASTAR, Inc., Madison, WI) allows for optimal alignment of sequences for comparison. Preferably, the "percentage of sequence identity" is determined by comparing two optimally aligned sequences across a comparison window of at least 20 positions. The portion of the polynucleotide or polypeptide sequence in the comparison window may contain 20% or less, typically 5 to 15%, or 10 to 12%, of additions or deletions (i.e., gaps) compared to a reference sequence (which does not contain additions or deletions) for optimal alignment of the two sequences. The percentage is calculated by determining the number of positions in both sequences where the same nucleic acid base or amino acid residue appears, to obtain the number of matching positions; dividing the number of matching positions by the total number of positions in the reference sequence (i.e., the window size); and multiplying the result by 100 to obtain the percentage of sequence identity.

The variant can also, or alternatively, be substantially homologous to the natural gene or a portion thereof or its complement. This polynucleotide variant is capable of hybridizing with naturally occurring DNA sequences (or complementary sequences) encoding natural antibodies under moderately stringent conditions.

Those skilled in the art will understand that, due to the degeneracy of the genetic code, there are many nucleotide sequences encoding polypeptides as described herein. Some of these polynucleotides have minimal homology to the nucleotide sequences of any natural gene. However, the present invention specifically considers polynucleotides that vary due to differences in codon usage. Furthermore, alleles of genes comprising the polynucleotide sequences provided herein are within the scope of the invention. An allele is an endogenous gene altered due to one or more mutations (such as nucleotide deletions, additions, and/or substitutions). The resulting mRNA and protein may, but need not, have altered structure or function. Alleles can be identified using standard techniques (such as hybridization, amplification, and/or database sequence comparison).

In one aspect, the present invention provides a method for preparing any of the polynucleotides described herein. For example, the polynucleotides of the present invention can be obtained using chemical synthesis, recombination methods, or PCR. Methods for the chemical synthesis of polynucleotides are well known in the art and need not be described in detail herein. Those skilled in the art can use the sequences provided herein and commercial DNA synthesizers to generate desired DNA sequences.

To prepare polynucleotides using recombinant methods, a polynucleotide containing the desired sequence can be inserted into a suitable vector, which can then be introduced into a suitable host cell for replication and amplification, as discussed further herein. Polynucleotides can be inserted into host cells by any means known in the art. Cells are transformed by introducing exogenous polynucleotides through direct uptake, endocytosis, transfection, F-crossing, or electroporation. Once introduced, the exogenous polynucleotide can be maintained within the cell as a non-integrating vector (such as a plasmid) or integrated into the host cell genome. The amplified polynucleotides can then be isolated from the host cell using methods well-known in the art (e.g., Sambrook et al., 1989).

Alternatively, PCR allows for the replication of DNA sequences. PCR technology is well known in the art and is described in U.S. Patent Nos. 4,683,195, 4,800,159, 4,754,065, and 4,683,202, and PCR: The Polymerase Chain Reaction, edited by Mullis et al., Birkauswer Press, Boston, 1994.

RNA can be obtained by using DNA isolated in a suitable vector and inserted into a suitable host cell. Once the cell replicates and the DNA is transcribed into RNA, RNA can then be isolated using methods well known to those skilled in the art, such as Sambrook et al., 1989, as described above.

Suitable cloning vectors can be constructed using standard techniques or selected from a large number of cloning vectors available in the art. While the chosen cloning vector can vary depending on the intended host cell, useful cloning vectors typically possess self-replicating capabilities, may have a single target for a specific restriction endonuclease, and/or may carry genes related to markers that can be used to select clones containing that vector. Suitable examples include plasmids and bacterial viruses such as pUC18, pUC19, Bluescript (e.g., pBS SK+) and its derivatives, mp18, mp19, pBR322, pMB9, ColE1, pCR1, RP4, phage DNA, and shuttle vectors such as pSA3 and pAT28. These and many other cloning vectors are available from commercial vendors such as BioRad, Strategene, and Invitrogen.

Expression vectors are typically reproducible polynucleotide constructs containing polynucleotides according to the invention. This implies that the expression vector must be replicable in the host cell as an episome or as an integral part of chromosomal DNA. Suitable expression vectors include, but are not limited to, plasmids, viral vectors including adenoviruses, adeno-associated viruses, retroviruses, viscera, and the expression vectors disclosed in PCT Publication No. WO87/04462. Vector components may typically include, but are not limited to, one or more of the following: a signal sequence; an origin of replication; one or more marker genes; and suitable transcriptional control elements (such as promoters, enhancers, and terminators). For expression (i.e., translation), one or more translational control elements, such as ribosome binding sites, translation initiation sites, and stop codons, are also typically required.

Vectors containing the target polynucleotide can be introduced into host cells by any of a number of suitable means, including electroporation, transfection with calcium chloride, rubidium chloride, calcium phosphate, DEAE-glucan, or other substances; particle bombardment; lipid transfection; and infection (e.g., when the vector is an infectious pathogen such as vaccinia virus). The choice of vector or polynucleotide to be introduced will often depend on the characteristics of the host cell.

Any host cell capable of overexpressing heterologous DNA can be used for the purpose of isolating genes encoding target antibodies, peptides, or proteins. Non-limiting examples of mammalian host cells include, but are not limited to, COS, HeLa, and CHO cells. See also PCT Publication WO87/04462. Suitable non-mammalian host cells include prokaryotes (such as *Escherichia coli* or *Bacillus subtilis *) and yeasts (such as *Saccharomyces cerevisiae *, *Schizosaccharomyces pombe * , or *Kluyveromyces lactis *). Preferably, the host cell expresses cDNA at a level approximately 5-fold higher, more preferably 10-fold higher, and even more preferably 20-fold higher than the corresponding endogenous antibody or protein CD3 or CD3 domain (e.g., domains 1-4), as determined by immunoassay or FACS. Cells overexpressing the target antibody or protein can be identified.

In one respect, the CD3 antibody molecule will have an affinity for CD3, as measured, for example, by direct or competitive binding in the picomolar to micromolar affinity range, preferably in the picomolar to low nanomolar range.

Bispecific antibodies can be prepared using the sequences disclosed herein. Methods for preparing bispecific antibodies are known in the art (see, for example, Suresh et al., Methods in Enzymology 121:210, 1986). Traditionally, the recombinant generation of bispecific antibodies is based on the co-expression of two immunoglobulin heavy-light chain pairs, where the two heavy chains have different specificities (Millstein and Cuello, Nature 305, 537-539, 1983).

In another embodiment, the antibody as described herein comprises a full-length human antibody, wherein the antibody variable region of the heterodimeric protein is capable of recruiting human immune effector cells by specifically binding to effector antigens (e.g., CD3 antigens) located on human immune effector cells, and wherein a second antibody variable region of the heterodimeric protein is capable of specifically binding to the target antigen. In some embodiments, the human antibody has an IgG1, IgG2, IgG3, or IgG4 isotype. In some embodiments, the heterodimeric protein comprises an immune-inert Fc chain.

Human immune effector cells can be any variety of immune effector cells known in the art. For example, immune effector cells can be members of the human lymphoid cell lineage, including but not limited to T cells (e.g., cytotoxic T cells), B cells, and natural killer (NK) cells. Immune effector cells can also be, for example, but not limited to, members of the human myeloid lineage, including but not limited to monocytes, neutrophils, and dendritic cells. Such immune effector cells can have cytotoxic or apoptotic effects on target cells, or other desired effects, after activation by binding to effector antigens.

Effector antigens are antigens (e.g., proteins or peptides) expressed on human immune effector cells. Examples of effector antigens that can be bound by heterodimers (e.g., heterodimer antibodies or bispecific antibodies) include, but are not limited to, human CD3, CD16, NKG2D, NKp46, CD2, CD28, CD25, CD64, and CD89.

Target cells can be natural or foreign to humans. In natural target cells, the cell may have been transformed into a malignant cell or pathologically modified (e.g., a natural target cell infected by a virus, malaria parasite, or bacteria). In foreign target cells, the cell is an invading pathogen, such as bacteria, malaria parasite, or virus.

In some embodiments, the antibodies that can be used in this invention are Fab, Fab fragment, F(ab) ₂ fragment, Fv fragment, single-chain Fv fragment, disulfide-stabilized Fv fragment, single-chain antibody, monoclonal antibody, chimeric antibody, bispecific antibody, trispecific antibody, multispecific antibody, bispecific heterodimeric biantibody, bispecific heterodimeric IgG, polyclonal antibody, labeled antibody, humanized antibody, human antibody, or fragments thereof.

In some implementations, the CD3 antibody described herein is a monoclonal antibody. For example, the CD3 antibody is a humanized monoclonal antibody or a chimeric monoclonal antibody.

This invention covers antibodies comprising an Fc chain or a domain thereof, or a portion thereof. In some embodiments, the Fc chain or a portion thereof comprises one or more constant domains (e.g., CH2 or CH3 domains) of an Fc chain of IgG1, IgG2, IgG3, or IgG4. In another embodiment, this invention covers molecules comprising an Fc chain or a portion thereof, wherein the Fc chain or a portion thereof contains at least one amino acid modification (e.g., substitution) relative to an equivalent wild-type Fc chain or a portion thereof. Variant Fc regions are well known in the art and are primarily used to alter the phenotype of antibodies comprising variant Fc regions, as determined in any binding activity or effector function assay well known in the art (e.g., ELISA, SPR assay, or ADCC). Such variant Fc chains or portions thereof can prolong the plasma half-life and stability exhibited by bispecific antibodies comprising Fc chains or portions thereof of this invention. In another embodiment, this invention covers the use of any Fc variant known in the art.

In one embodiment, one or more amino acid modifications are made to the Fc chain to reduce the affinity and coercivity of the Fc chain, thereby reducing the affinity and coercivity of the bispecific antibody of the present invention for one or more FcγR receptors. In one specific embodiment, the present invention covers bispecific antibodies comprising a variant Fc chain or a portion thereof, wherein the variant Fc chain contains at least one amino acid modification relative to the wild-type Fc chain, and the variant Fc region binds only one FcγR, wherein the FcγR is FcγRIIIA. In another specific embodiment, the present invention covers bispecific antibodies comprising a variant Fc chain or a portion thereof, wherein the variant Fc chain contains at least one amino acid modification relative to the wild-type (WT) Fc chain, and the variant Fc region binds only one FcγR, wherein the FcγR is FcγRIIA. In yet another specific embodiment, the present invention covers bispecific antibodies comprising a variant Fc chain or a portion thereof, wherein the variant Fc chain contains at least one amino acid modification relative to the wild-type Fc chain, and the variant Fc chain binds only one FcγR, wherein the FcγR is FcγRIIB. In another embodiment, the invention covers molecules comprising a variant Fc chain, wherein, relative to molecules that do not contain an Fc chain or contain a wild-type Fc chain, the variant confers or mediates reduced ADCC activity (or other effector functions) and/or increased binding to FcγRIIB (CD32B), as measured using methods known to those skilled in the art and described herein.

This invention also covers the use of Fc regions comprising domains or regions from two or more IgG isoforms. As is known in the art, amino acid modifications to the Fc region can significantly affect Fc-mediated effector function and/or binding activity. However, when implemented in the context of a selected IgG isoform, these alterations to functional characteristics can be further improved and/or manipulated. Similarly, the native characteristics of isoform Fc can be manipulated through one or more amino acid modifications. Multiple IgG isoforms (i.e., IgG1, IgG2, IgG3, and IgG4) exhibit different physical and functional properties, including serum half-life, complement binding, FcγR binding affinity, and effector functional activity (e.g., ADCC, CDC), due to differences in the amino acid sequences of their hinges and/or Fc regions.

In one embodiment, amino acid modifications and IgG Fc regions are independently selected based on their respective individual binding and/or effector functional activities to engineer bispecific antibodies with desired characteristics. In one specific embodiment, the binding and/or effector functional activities of amino acid modifications and IgG hinge/Fc regions are separately measured in the context of IgG1, as described herein or known in the art. In one embodiment, amino acid modifications and IgG hinge/Fc regions exhibit similar functionality, such as reduced ADCC activity (or other effector functions) and/or increased binding to FcγRIIB in the context of bispecific antibodies or other Fc-containing molecules (e.g., immunoglobulins). In another embodiment, the invention covers variant Fc regions comprising combinations of amino acid modifications known in the art and selected IgG regions exhibiting novel properties that are undetectable when the modifications and/or regions are measured independently as described herein.

One method for determining the binding affinity of an antibody to CD3 is by measuring the binding affinity of a monofunctional Fab fragment of the antibody. To obtain a monofunctional Fab fragment, the antibody (e.g., IgG) can be cleaved with papain or recombinantly expressed. The affinity of the antibody's CD3 Fab fragment can be determined using a surface plasmon resonance (SPR) system (Biacore™ 3000™ SPR system, Biacore™, INC, Piscataway NJ) equipped with a pre-immobilized streptavidin sensor chip (SA) or anti-mouse Fc or anti-human Fc, using HBS-EP run buffer (0.01 M HEPES, pH 7.4, 0.15 M NaCl, 3 mM EDTA, 0.005% v/v surfactant P20). Biotinylated or Fc-fused human CD3 can be diluted in HBS-EP buffer to concentrations less than 0.5 μg/mL and injected across individual chip channels using variable contact times to achieve two ranges of antigen density: 50–200 reaction units (RU) for detailed kinetic studies, or 800–1,000 RU for screening assays. Regeneration studies have shown that 25 mM NaOH in 25% v/v ethanol effectively removes bound Fab while maintaining CD3 activity on the chip for more than 200 injections. Typically, serial dilutions of purified Fab samples (across concentrations of 0.1–10x estimated _KD ) are injected at 100 µL/min for 1 minute, allowing dissociation times up to 2 hours. Fab protein concentrations are determined by ELISA and/or SDS-PAGE electrophoresis using known concentrations of Fab (e.g., as determined by amino acid analysis) as standards. The data were fitted to a 1:1 Lammill binding model using the BIAevaluation procedure (Karlsson, R. Roos, H. Fagerstam, L. Petersson, B. (1994). Methods Enzymology 6. 99-110), simultaneously obtaining the kinetic binding rate (k _{_on} ) and dissociation rate (k_{_off} ). The equilibrium dissociation constant (K_{_D}) was calculated as k _{_off} /k _{_on} . This protocol is suitable for determining the binding affinity of antibodies to any CD3, including human CD3, CD3 from another mammal (such as mouse CD3, rat CD3, or primate CD3), and different forms of CD3 (e.g., glycosylated CD3). Antibody binding affinity is typically measured at 25°C, but can also be measured at 37°C.

Antibodies as described herein can be prepared by any method known in the art. For the generation of hybridoma cell lines, the host animal immunization pathway and timeline are generally consistent with established and routine techniques for antibody stimulation and production, as further described herein. General techniques for generating human and mouse antibodies are known in the art and/or described herein.

Consider manipulating any mammalian subject, including humans or antibody-producing cells derived therefrom, to serve as a basis for generating mammalian, including human, hybridoma cell lines. Typically, host animals are inoculated with a given amount of immunogen via intraperitoneal, intramuscular, oral, subcutaneous, plantar, and/or intradermal administration, including as described herein.

Hybridomas can be prepared from lymphocytes and immortalized myeloma cells using general somatic cell hybridization techniques, as described in Kohler, B. and Milstein, C., Nature 256:495-497, 1975, or modified techniques, as described in Buck, D. W. et al., In Vitro, 18:377-381, 1982. Available myeloma lines, including but not limited to X63-Ag8.653 and those from the Salk Institute, Cell Distribution Center, San Diego, Calif., USA, can be used in the hybridization. Typically, the technique involves fusing myeloma cells and lymphoid cells using a fusion agent, such as polyethylene glycol, or by electrical means well known to those skilled in the art. After fusion, the cells are separated from the fusion medium and grown in a selective growth medium (such as hypoxanthine-aminopterin-thymidine (HAT) medium) to eliminate unhybridized parental cells. Any culture medium, with or without serum supplementation, as described herein can be used to culture hybridomas that secrete monoclonal antibodies. As an alternative to cell fusion technology, EBV immortalized B cells can be used to generate the monoclonal antibodies of this invention. If necessary, the hybridoma is amplified and subcloned, and the antiimmunogen activity of the supernatant is determined using routine immunoassay procedures (e.g., radioimmunoassay, enzyme immunoassay, or fluorescence immunoassay).

Hybridomas that can be used as antibody sources encompass all derivatives of parental hybridomas that produce monoclonal antibodies specific to CD3 or a portion thereof, as well as progeny cells.

Hybridomas that produce such antibodies can be grown in vitro or in vivo using known procedures. If desired, monoclonal antibodies can be isolated from culture media or body fluids using routine immunoglobulin purification procedures such as ammonium sulfate precipitation, gel electrophoresis, dialysis, chromatography, and ultrafiltration. For example, unwanted activity (if present) can be removed by running the formulation on an adsorbent made of an immunogen attached to a solid phase and eluting or releasing the desired antibody from the immunogen. Using bifunctional reagents or derivatizing agents, such as maleimide benzoyl sulfosuccinimide (conjugated via cysteine residues), N-hydroxysuccinimide (conjugated via lysine residues), glutaraldehyde, succinic anhydride, _SOCl2 , or ^R1N =C=NR (where R and ^R1 are different alkyl groups), conjugated with human CD3 or fragments containing the target amino acid sequence in the host animal to be immunized, a population of antibodies (e.g., monoclonal antibodies) can be obtained.

If desired, the target antibody can be sequenced, and the polynucleotide sequence can then be cloned into a vector for expression or propagation. The sequence encoding the target antibody can be maintained in a vector within a host cell, which can then be expanded and frozen for future use. The generation of recombinant monoclonal antibodies in cell culture can be performed by means known in the art through cloning antibody genes from B cells. See, for example, Tiller et al., J. Immunol. Methods 329,112,2008; U.S. Patent No. 7,314,622.

In an alternative approach, polynucleotide sequences can be genetically manipulated to humanize antibodies or improve their affinity or other characteristics. For example, if an antibody is used in clinical trials and treatments in humans, the constant region can be engineered to more closely resemble the human constant region to avoid an immune response. It might be desirable to genetically manipulate antibody sequences to achieve greater affinity for CD3 and greater therapeutic efficacy.

In some embodiments, the antibody has a modified constant region that removes or reduces Fcγ receptor binding. For example, Fc may be human IgG2 containing the mutant D265, where the amino acid residues are numbered with reference to the wild-type IgG2 sequence. Therefore, in some embodiments, the constant region has a modified constant region having the sequence shown in SEQ ID NO: 80. In some embodiments, the antibody has a modified constant region having the sequence shown in SEQ ID NO: 81.

There are four general steps to humanizing a monoclonal antibody. These steps are: (1) determining the nucleotide and predicted amino acid sequences of the variable regions of the light and heavy chains of the starting antibody; (2) designing the humanized antibody, i.e., deciding which antibody framework region to use during the humanization process; (3) the actual humanization method/technique; and (4) the transfection and expression of the humanized antibody. See, for example, U.S. Patent Nos. 4,816,567; 5,807,715; 5,866,692; 6,331,415; 5,530,101; 5,693,761; 5,693,762; 5,585,089; and 6,180,370.

Many humanized antibody molecules containing antigen-binding sites derived from non-human immunoglobulins have been described, including chimeric antibodies with rodent or modified rodent V regions fused to human constant regions and their associated CDRs. See, for example, Winter et al. Nature 349:293-299, 1991; Lobuglio et al. Proc. Nat. Acad. Sci. USA 86:4220-4224, 1989; Shaw et al. J Immunol. 138:4534-4538, 1987; and Brown et al. Cancer Res. 47:3577-3583, 1987. Other references describe the transplantation of rodent CDRs into human support framework regions (FRs) prior to fusion with appropriate human antibody constant regions. See, for example, Riechmann et al., Nature 332:323-327, 1988; Verhoeyen et al., Science 239:1534-1536, 1988; and Jones et al., Nature 321:522-525, 1986. Another reference describes rodent CDRs supported by recombinantly engineered rodent framework regions. See, for example, European Publication No. EP0519596. These “humanized” molecules are designed to minimize unwanted immune responses to rodent anti-human antibody molecules, which limits the duration and effectiveness of therapeutic applications that are partially present in human receptors. For example, antibody constant regions can be engineered to be immunologically inert (e.g., not triggering complement cleavage). See, for example, PCT application No. PCT/GB99/01441; UK application No. 9809951.8. Other methods using humanized antibodies may also be disclosed by Daugherty et al., Nucl. Acids Res. 19:2471-2476, 1991, and in U.S. Patent Nos. 6,180,377; 6,054,297; 5,997,867; 5,866,692; 6,210,671; and 6,350,861; and PCT Publication No. WO01/27160.

The general principles discussed above regarding humanized antibodies also apply to custom antibodies, for example, for use in other mammals (e.g., humans, non-human primates, mice, donkeys, sheep, rabbits, goats, guinea pigs, camels, horses, or chickens). Furthermore, one or more aspects of the humanized antibodies described herein can be combined, such as CDR transplantation, scaffold mutation, and CDR mutation.

In one variant, fully human antibodies can be obtained using commercially available mice engineered to express specific human immunoglobulin proteins. Transgenic animals designed to produce a more desired (e.g., fully human antibody) or more robust immune response can also be used to generate humanized or human antibodies. Examples of this technology are Xenomouse™ from Abgenix, Inc. (Fremont, CA) and HuMAb-Mouse® and TC Mouse™ from Medarex, Inc. (Princeton, NJ).

In one alternative, the antibody can be recombinantly prepared and expressed using any method known in the art. In another alternative, the antibody can be recombinantly prepared using phage display technology. See, for example, U.S. Patent Nos. 5,565,332; 5,580,717; 5,733,743; and 6,265,150; and Winter et al., Annu. Rev. Immunol. 12:433-455, 1994. Alternatively, phage display technology (McCafferty et al., Nature 348:552-553, 1990) can be used to generate human antibodies and antibody fragments in vitro from a gene library of immunoglobulin variable (V) domains from non-immunized donors. According to this technology, the antibody V domain gene frame is cloned into the major or minor coat protein genes of filamentous phages, such as M13 or fd, and displayed as a functional antibody fragment on the surface of the phage particle. Because filamentous particles contain single-stranded DNA copies of the phage genome, selection based on antibody functional characteristics also leads to the selection of genes encoding antibodies that exhibit those characteristics. Thus, phages mimic some characteristics of B cells. Phage display can be performed in various forms; for reviews, see, for example, Johnson, Kevin S., and Chiswell, David J., Current Opinion in Structural Biology 3:564-571, 1993. Several sources of the V gene segment can be used for phage display. Clackson et al., Nature 352:624-628, 1991, isolated a wide variety of anti-oxazolone antibodies from a small, randomly combined library of the V gene derived from the spleen of immunized mice. Following the techniques described by Mark et al., J. Mol. Biol. 222:581-597, 1991, or Griffith et al., EMBO J.12:725-734, 1993, a V gene library from non-immunized human donors can be constructed, and antibodies against a wide variety of antigens, including self-antigens, can be isolated. In the innate immune response, antibody genes accumulate mutations at a high rate (somatic hypermutation). Some of the introduced changes will confer higher affinity, and B cells exhibiting high-affinity surface immunoglobulins will preferentially replicate and differentiate during subsequent antigen challenges. This natural process can be simulated using a technique called “chain shuffling” (Marks et al., Bio/Technol. 10:779-783, 1992). In this approach, the affinity of “primary” human antibodies obtained by phage display can be improved by sequentially replacing the heavy and light chain V region genes with a naturally occurring library (library) of variants of the V domain genes obtained from non-immunized donors. This technique allows for the production of antibodies and antibody fragments with affinities in the pM-nM range. A strategy for preparing very large phage antibody libraries (also known as “mother-of-all libraries”) has been described by Waterhouse et al., Nucl. Acids Res. 21:2265-2266, 1993. Gene shuffling can also be used to derive human antibodies from rodent antibodies, wherein the human antibodies have similar affinity and specificity to the starting rodent antibody. According to this method, also known as “epitope imprinting,” the heavy or light chain V domain genes of rodent antibodies obtained via phage display technology are replaced with a human V domain gene library, producing a rodent-human chimera. Selection on the antigen results in the isolation of the human variable region capable of restoring a functional antigen-binding site; that is, the epitope governs (marks) the selection of the partner. When this process is repeated to replace the remaining rodent V domains, human antibodies are obtained (see PCT Publication No. WO93/06213). Unlike traditional humanization of rodent antibodies via CDR transplantation, this technique provides fully human antibodies that do not contain rodent-derived frameworks or CDR residues.

Antibodies can be prepared recombinantly by first isolating the antibody and antibody-producing cells from a host animal, obtaining the gene sequence, and using the gene sequence to recombinantly express the antibody in host cells (e.g., CHO cells). Another approach is to express the antibody sequence in plants (e.g., tobacco) or transgenic milk. Methods for recombinantly expressing antibodies in plants or milk have been disclosed. See, for example, Peeters et al. Vaccine 19:2756, 2001; Lonberg, N. and D. Huszar Int. Rev. Immunol 13:65, 1995; and Pollock et al., J Immunol Methods 231:147, 1999. Methods for preparing antibody derivatives, such as humanized, single-chain, etc., are known in the art.

Immunoassays and flow cytometry sorting techniques, such as fluorescence activated cell sorting (FACS), can also be used to isolate antibodies specific to CD3 or target antigens.

Antibodies as described herein can bind to a variety of different solid supports or carriers. Such supports can be active and/or inert. Well-known supports include polypropylene, polystyrene, polyethylene, dextran, nylon, amylase, glass, natural and modified cellulose, polyacrylamide, agarose, and magnets. For the purposes of this invention, the nature of the support can be soluble or insoluble. Those skilled in the art will recognize other suitable supports for binding antibodies, or will be able to determine these using routine experiments. In some embodiments, the support comprises a portion targeting the myocardium.

DNA encoding monoclonal antibodies can be readily isolated and sequenced using standard procedures, such as by using oligonucleotide probes capable of specifically binding to genes encoding the heavy and light chains of monoclonal antibodies. Hybridoma cells serve as a preferred source of this DNA. Once isolated, the DNA can be placed in an expression vector (such as the one disclosed in PCT Publication No. WO87/04462) and then transfected into host cells that otherwise do not produce immunoglobulin proteins (such as E. coli cells, simian COS cells, CHO cells, or myeloma cells) to obtain monoclonal antibody synthesis in recombinant host cells. See, for example, PCT Publication No. WO87/04462. DNA can also be modified, for example, by replacing homologous mouse sequences with coding sequences of human heavy and light chain constant regions (Morrison et al., Proc. Nat. Acad. Sci. 81:6851, 1984), or by covalently binding all or part of the coding sequence of an immunoglobulin-coding sequence to the coding sequence of a non-immunoglobulin polypeptide. In this manner, "chimeric" or "hybrid" antibodies with the binding specificity of the monoclonal antibodies described herein are prepared.

Methods known in the art can be used to identify or characterize CD3 or other antigen antibodies as described herein, thereby detecting and/or measuring antigen expression levels. In some embodiments, CD3 antibodies are identified by incubating a candidate antibody with CD3 and monitoring the associated decrease in binding and/or CD3 expression levels. Binding assays can be performed using purified CD3 peptides, or cells naturally expressing or transfected to express CD3 peptides. In one embodiment, the binding assay is a competitive binding assay in which the ability of a candidate antibody to compete with a known CD3 antibody for CD3 binding is evaluated. This assay can be performed in various forms, including ELISA.

Following initial identification, the activity of candidate CD3 or other antigen-antibody pairs can be further confirmed and refined using bioassays known for testing targeted biological activity. Alternatively, bioassays can be used to directly screen candidates. Several methods for identifying and characterizing antibodies are described in detail in the examples.

CD3 or other antigen antibodies can be characterized using methods well-known in the art. For example, one method is to identify the epitope to which it binds, or "epitope mapping." Many methods are known in the art for mapping and characterizing the location of epitopes on proteins, including resolving the crystal structure of antibody-antigen complexes, competitive assays, gene fragment expression assays, and synthetic peptide-based assays, as described, for example, in Chapter 11 of Harlow and Lane, Using Antibodies, a Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 1999. In another instance, epitope mapping can be used to determine the sequence to which the antibody binds. Epitope mapping is commercially available from various sources, for example, Pepscan Systems (Edelhertweg 15, 8219 PH Lelystad, Netherlands). The epitope can be a linear epitope, i.e., contained in a single amino acid segment, or a conformational epitope formed by the three-dimensional interactions of amino acids, which may not necessarily be contained in a single segment. Peptides of varying lengths (e.g., at least 4-6 amino acids long) can be isolated or synthesized (e.g., recombinantly) and used for binding assays with CD3 or other antigens. In another example, epitopes to which CD3 or other antigens bind can be determined in systematic screening by using overlapping peptides derived from CD3 or other antigen sequences and measuring binding with CD3 or other antigens. Based on gene fragment expression assays, open reading frames encoding CD3 or other antigens are randomly or through specific genetic constructs fragmented, and the reactivity of expressed fragments of CD3 or other antigens with the antibody to be tested is determined. Gene fragments can be generated, for example, by PCR, and then transcribed and translated into proteins in vitro in the presence of radioactive amino acids. The binding of the antibody to the radiolabeled CD3 or other antigen fragment is then determined by immunoprecipitation and gel electrophoresis. Some epitopes can also be identified using large libraries (phage libraries) of random peptide sequences displayed on the surface of phage particles. Alternatively, the binding of a defined library of overlapping peptide fragments to a test antibody can be tested in a simple binding assay. In an additional instance, mutagenesis of antigen-binding domains, domain exchange assays, and alanine partitioning can be performed to identify residues required, sufficient, and/or essential for epitope binding. For example, domain exchange assays can be performed using mutant CD3 or other antigens, where various fragments of the CD3 or other antigenic protein have been replaced (exchanged) with sequences from antigens from another species (e.g., mice), or closely related but antigenically different proteins (e.g., Trop-1). By evaluating the binding of the antibody to mutant CD3 or other antigens, the importance of specific CD3 or other antigenic fragments to antibody binding can be assessed.

Another method that can be used to characterize CD3 or other antigen-antibodies is to use a competitive assay with other antibodies known to bind to the same antigen, i.e., various fragments on CD3 or other antigens, to determine whether the CD3 or other antigen-antibody binds to the same epitopes as the other antibodies. Competitive assays are well known to those skilled in the art.

Expression vectors can be used to direct the expression of CD3 or other antigens and antibodies. Those skilled in the art are familiar with the administration of expression vectors to obtain the expression of exogenous proteins in vivo. See, for example, U.S. Patent Nos. 6,436,908; 6,413,942; and 6,376,471. Administration of expression vectors includes local or systemic administration, including injection, oral administration, particle gun or catheter-directed administration, and local application. In another embodiment, the expression vector is administered directly to the sympathetic trunk or ganglia, or to the coronary arteries, atria, ventricles, or pericardium.

Targeted delivery can also be achieved using therapeutic compositions containing expression vectors or subgenomic polynucleotides. Receptor-mediated DNA delivery techniques are described, for example, in: Findeis et al., Trends Biotechnol., 11:202, 1993; Chiou et al., Gene Therapeutics: Methods And Applications of Direct Gene Transfer, J.A. Wolff, ed., 1994; Wu et al., J. Biol. Chem., 263:621, 1988; Wu et al., J. Biol. Chem., 269:542, 1994; Zenke et al., Proc. Natl. Acad. Sci. USA, 87:3655, 1990; and Wu et al., J. Biol. Chem., 266:338, 1991. Therapeutic compositions containing polynucleotides are administered topically in gene therapy regimens at concentrations ranging from about 100 ng to about 200 mg of DNA. Concentrations ranging from about 500 ng to about 50 mg, about 1 μg to about 2 mg, about 5 μg to about 500 μg, and about 20 µg to about 100 µg of DNA may also be used during gene therapy regimens. Gene delivery vehicles can be used to deliver therapeutic polynucleotides and peptides. Gene delivery vehicles can be of viral or non-viral origin (see generally, Jolly, Cancer Gene Therapy, 1:51, 1994; Kimura, Human Gene Therapy, 5:845, 1994; Connelly, Human Gene Therapy, 1:185, 1995; and Kaplitt, Nature Genetics, 6:148, 1994). Expression of such coding sequences can be induced using endogenous mammalian promoters or heterologous promoters. The expression of a coding sequence can be constitutive or regulated.

Virus-based vectors for delivering desired polynucleotides and expressing them in desired cells are well known in the art. Exemplary virus-based vectors include, but are not limited to, recombinant retroviruses (see, for example, PCT Publications WO90/07936; WO94/03622; WO93/25698; WO93/25234; WO93/11230; WO93/10218; WO91/02805; U.S. Patents 5,219,740 and 4,777,127; GB Patent 2,200,651; and EP Patent EP0 345242), alphavirus-based vectors (e.g., Sindbis virus vectors, Semliki Forest Virus (ATCC)). VR-67; ATCCVR-1247), Ross River virus (ATCC VR-373; ATCCVR-1246) and Venezuelan equine encephalitis virus (ATCC VR-923; ATCC VR-1250; ATCC VR 1249; ATCC VR-532), and adeno-associated virus (AAV) vectors (see, for example, PCT Publications WO94/12649, WO93/03769; WO93/19191; WO94/28938; WO95/11984 and WO95/00655). Alternatively, administration of DNA linked to a killed adenovirus may be used, as described in Curiel, Hum. Gene Ther., 3:147, 1992.

Non-viral delivery media and methods may also be used, including, but not limited to, polycationically condensed DNA linked to or unlinked by a single, killed adenovirus (see, for example, Curiel, Hum. Gene Ther., 3:147, 1992); ligand-linked DNA (see, for example, Wu, J. Biol. Chem., 264:16985, 1989); eukaryotic cell delivery media (see, for example, U.S. Patent No. 5,814,482; PCT Publications WO95/07994; WO96/17072; WO95/30763; and WO97/42338) and nucleic acid charge neutralization or fusion with the cell membrane. Naked DNA may also be used. Exemplary methods for introducing naked DNA are described in PCT Publication WO90/11092 and U.S. Patent No. 5,580,859. Liposomes that can act as gene delivery mediators are described in U.S. Patent No. 5,422,120; PCT Publications Nos. WO95/13796, WO94/23697, WO91/14445; and European Publication EP 0524968. Additional methods are described in Philip, Mol. Cell Biol., 14:2411, 1994 and Woffendin, Proc. Natl. Acad. Sci., 91:1581, 1994.

In some embodiments, the invention covers compositions including pharmaceutical compositions comprising antibodies of the invention as described herein or prepared by the methods described herein and having the characteristics described herein. As used herein, pharmaceutical compositions may comprise one or more antibodies binding to tumor antigens, one or more bispecific antibodies binding to CD3 and tumor antigens, and/or one or more polynucleotides encoding sequences of one or more of these antibodies. These compositions may further comprise suitable excipients, such as pharmaceutically acceptable excipients, including buffers, which are well known in the art.

In one embodiment, the invention is characterized by a composition that is a combination of a CD3 antibody and a second therapeutic agent. In one embodiment, the second therapeutic agent is any agent advantageously combined with a CD3 antibody. Exemplary agents that can be advantageously combined with a CD3 antibody include, but are not limited to, other agents that bind and/or activate CD3 signaling (including other antibodies or their antigen-binding fragments, etc.) and/or agents that do not directly bind CD3 but still activate or stimulate immune cell activation. The invention further includes additional combination therapies and co-formulations involving the CD3 antibodies of the invention. The invention also provides methods for preparing any of these antibodies. The antibodies of the invention can be prepared by procedures known in the art, including de novo protein synthesis and recombinant expression of nucleic acids encoding binding proteins. The desired nucleic acid sequence can be generated by recombinant methods (e.g., PCR mutagenesis of previously prepared variants of the desired polynucleotide) or by solid-phase DNA synthesis. Recombinant expression methods are typically used. In one aspect, the invention provides polynucleotides comprising sequences encoding CD3VH and/or VL. Due to the degeneracy of the genetic code, multiple nucleic acid sequences encode each immunoglobulin amino acid sequence, and the invention includes all nucleic acids encoding the binding proteins described herein.

Chimeric or hybrid antibodies can also be prepared in vitro using known methods of synthetic protein chemistry, including those involving cross-linking agents. For example, immunotoxins can be constructed using disulfide exchange reactions or by forming thioether bonds. Examples of suitable reagents for this purpose include iminothiolates and methyl-4-mercaptobutyrimidate.

In recombinant humanized antibodies, the Fcγ region can be modified to avoid interactions with the Fcγ receptor, as well as with complement and the immune system. Techniques for preparing such antibodies are described in PCT Publication WO99/58572. For example, if the antibody is to be used in clinical trials and treatments in humans, the constant region can be engineered to more closely resemble the human constant region to avoid an immune response. See, for example, U.S. Patent Nos. 5,997,867 and 5,866,692.

Compared to the corresponding germline sequences from which antibodies are derived, CD3 antibodies disclosed herein may contain one or more amino acid substitutions, insertions, and/or deletions in the framework and/or CDR regions of the heavy and light chain variable domains. Such mutations can be readily identified by comparing the amino acid sequences disclosed herein with germline sequences available from, for example, public antibody sequence databases. This invention includes antibodies and antigen-binding fragments thereof derived from any of the amino acid sequences disclosed herein, wherein one or more amino acids in one or more framework and/or CDR regions are mutated to one or more corresponding residues of the germline sequence from which the antibody is derived, or one or more corresponding residues of another human germline sequence, or conserved amino acid substitutions of one or more corresponding germline residues (such sequence changes are collectively referred to herein as “germline mutations”). Starting with the heavy and light chain variable region sequences disclosed herein, those skilled in the art can readily generate a variety of antibody and antigen-binding fragments containing one or more individual germline mutations or combinations thereof. In some embodiments, all framework and/or CDR residues within the VH and/or VL domains are mutated back to residues found in the original germline sequence from which the antibody is derived. In some embodiments, only certain residues are mutated back to the original germline sequence, for example, mutated residues found only in the first 8 amino acids of FR1 or the last 8 amino acids of FR4, or mutated residues found only in CDR1, CDR2, or CDR3. In some further embodiments, one or more framework and/or CDR residues are mutated to one or more corresponding residues of a different germline sequence (i.e., a germline sequence from which the antibody was originally derived).

Furthermore, the antibodies of the present invention may contain any combination of two or more germline mutations within the framework and/or CDR region, for example, wherein certain individual residues are mutated to corresponding residues of a specific germline sequence, while certain other residues different from the original germline sequence are maintained or mutated to corresponding residues of a different germline sequence. Once obtained, one or more desired properties of the antibody and antigen-binding fragment containing one or more germline mutations can be readily tested, such as increased binding specificity, increased binding affinity, improved or enhanced antagonistic or agonistic biological properties (as the case may be), reduced immunogenicity, etc. Antibody and antigen-binding fragments obtained in this general manner are included within the scope of the present invention.

The present invention also includes CD3 antibodies comprising variants of any of the HC VR, LC VR, and/or CDR amino acid sequences disclosed herein having one or more conserved substitutions. For example, the present invention includes CD3 antibodies having HC VR, LC VR, and/or CDR amino acid sequences having, for example, 10 or fewer, 8 or fewer, 6 or fewer, 4 or fewer, etc., conserved amino acid substitutions relative to any of the HC VR, LC VR, and/or CDR amino acid sequences, as disclosed herein.

Therefore, this invention covers modifications of the antibodies and peptides of this invention, variants as described herein, including functionally equivalent antibodies that do not significantly affect their properties, and variants with enhanced or reduced activity and/or affinity. For example, amino acid sequences can be mutated to obtain antibodies with desired binding affinity to tumor antigens and/or CD3. Peptide modification is a common practice in the art and need not be described in detail herein. Examples of modified peptides include peptides with conserved substitutions of amino acid residues, deletions or additions of one or more amino acids that do not significantly and harmfully alter the functional activity or affinity of the mature (enhancing) peptide for its ligands, or the use of chemical analogs.

Amino acid sequence insertions include fusion of the amino and/or carboxyl ends of peptides ranging in length from one residue to one hundred or more residues, as well as intra-sequence insertions of single or multiple amino acid residues. Examples of terminal insertions include antibodies with an N-terminal methionine residue or antibodies fused to an epitope tag. Other insertion variants of antibody molecules include fusion of an enzyme or a peptide that increases the antibody's half-life in the bloodstream with the N- or C-terminus of the antibody.

Substitution variants involve the removal of at least one amino acid residue from the antibody molecule and the insertion of a different residue at its position. Sites of most interest for substitution mutagenesis include hypervariable regions, but FR alterations are also expected. Conservative substitutions are shown under the heading “Conservative Substitutions” in Table 6. If such substitution results in an alteration of biological activity, more significant changes, designated “Exemplary Substitutions” in Table 6 or further described below regarding amino acid categories, can be introduced, and products screened.

Table 6

Significant modifications to the biological properties of antibodies are accomplished through selective substitution, which differs markedly in maintaining: (a) the polypeptide backbone structure in the substituted region, for example, as a sheet or helical conformation; (b) the charge or hydrophobicity of the molecule at the target site; or (c) the volume of the side chains. Naturally occurring amino acid residues are grouped based on common side chain properties:

(1) Nonpolar: Leucine, Met, Ala, Val, Leu, Ile;

(2) Neutral elements with no charge: Cys, Ser, Thr, Asn, Gln;

(3) Acidic (negatively charged): Asp, Glu;

(4) Alkaline (positively charged): Lys, Arg;

(5) Residues affecting chain orientation: Gly, Pro; and

(6) Aroma: Trp, Tyr, Phe, His.

Non-conservative substitutions are prepared by replacing members of one of these categories with members of another category.

Serine residues that are not involved in maintaining the correct conformation of the antibody can generally be replaced with serine to improve the oxidative stability of the molecule and prevent abnormal cross-linking. Conversely, one or more cysteine bonds can be added to the antibody to improve its stability, especially when the antibody is an antibody fragment, such as an Fv fragment.

Amino acid modifications can involve a complete redesign of a region (e.g., a variable region) by altering or modifying one or more amino acids. Changes in the variable region can alter binding affinity and/or specificity. In some embodiments, no more than one to five conserved amino acid substitutions are made within the CDR domain. In some embodiments, no more than one to three conserved amino acid substitutions are made within the CDR domain. In still other embodiments, the CDR domain is VH CDR3 and/or VL CDR3.

Modifications also include glycosylated and non-glycosylated peptides, as well as peptides with other post-translational modifications, such as, for example, glycosylation, acetylation, and phosphorylation with different sugars. Antibodies are glycosylated at conserved sites in their constant regions (Jefferis and Lund, Chem. Immunol. 65:111-128, 1997; Wright and Morrison, TibTECH 15:26-32, 1997). Oligosaccharide side chains of immunoglobulins influence protein function (Boyd et al., Mol. Immunol. 32:1311-1318, 1996; Wittwe and Howard, Biochem. 29:4175-4180, 1990), as well as intramolecular interactions between glycoprotein moieties, which can affect conformation and present the three-dimensional surface of glycoproteins (Jefferis and Lund, above; Wyss and Wagner, Current Opin. Biotech. 7:409-416, 1996). Oligosaccharides can also be used to target given glycoproteins to certain molecules based on specific recognition structures. Glycosylation of antibodies has also been reported to affect antibody-dependent cytotoxicity (ADCC). In particular, CHO cells with tetracycline-regulated β(1,4)-N-acetylglucosamine transferase III (GnTIII) (a glycosyltransferase that catalyzes the formation of aliquots of GlcNAc) expression have been reported to have improved ADCC activity (Umana et al., Mature Biotech. 17:176-180, 1999).

Antibody glycosylation is generally N-linked or O-linked. N-linked glycosylation refers to the attachment of the carbohydrate moiety to the side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine, asparagine-X-threonine, and asparagine-X-cysteine (where X is any amino acid except proline) are recognition sequences for the enzymatic attachment of the carbohydrate moiety to the asparagine side chain. Therefore, the presence of any of these tripeptide sequences in the polypeptide creates a potential glycosylation site. O-linked glycosylation refers to the attachment of one of the sugars N-acetylgalactosamine, galactose, or xylose to a hydroxy amino acid, most commonly serine or threonine, although 5-hydroxyproline or 5-hydroxylysine can also be used.

By altering the amino acid sequence to include one or more of the aforementioned tripeptide sequences (for N-glycosylation sites), the addition of glycosylation sites to the antibody can be conveniently accomplished. Alternatively, alterations can be made by adding or replacing one or more serine or threonine residues to the original antibody sequence (for O-glycosylation sites).

The glycosylation pattern of an antibody can also be altered without changing the underlying nucleotide sequence. Glycosylation is largely dependent on the host cell used to express the antibody. Since the cell types used to express recombinant glycoproteins such as antibodies as potential therapeutic agents are rarely native cells, variations in the antibody glycosylation pattern can be expected (see, for example, Hse et al., J. Biol. Chem. 272:9062-9070, 1997).

Besides host cell selection, factors influencing glycosylation during recombinant antibody production include growth mode, culture medium formulation, culture density, oxygenation, pH, purification protocol, and more. Various methods have been proposed to alter the glycosylation patterns achieved in specific host organisms, including the introduction or overexpression of certain enzymes associated with oligosaccharide production (US Patent Nos. 5,047,335; 5,510,261 and 5,278,299). For example, using endoglycosidase H (Endo H), N-glycosidase F, endoglycosidase F1, endoglycosidase F2, and endoglycosidase F3, glycosylation or certain types of glycosylation can be enzymatically removed from glycoproteins. Additionally, recombinant host cells can be genetically engineered to be defective in processing certain types of polysaccharides. These and similar techniques are well known in the art.

Other modification methods include the use of coupling techniques known in the art, including, but not limited to, enzymatic methods, oxidative substitution, and chelation. Modification can be used, for example, for label attachment for immunoassays. Modified peptides are prepared using procedures established in the art and can be screened using standard assays known in the art, some of which are described below and in the examples.

In some embodiments of the invention, the antibody comprises a modified constant region, such as a constant region that has increased affinity for the human Fcγ receptor and is immunologically inert or partially inert, for example, not triggering complement-mediated cleavage, not stimulating antibody-dependent cell-mediated cytotoxicity (ADCC), or not activating macrophages; or having reduced activity (compared to an unmodified antibody) in any one or more of the following: triggering complement-mediated cleavage, stimulating antibody-dependent cell-mediated cytotoxicity (ADCC), or activating microglia. Different modifications to the constant region can be used to achieve optimal levels and/or combinations of effector functions. See, for example, Morgan et al., Immunology 86:319-324, 1995; Lund et al., J. Immunology 157:4963-9 157:4963-4969, 1996; Idusogie et al., J. Immunology 164:4178-4184, 2000; Tao et al., J. Immunology 143: 2595-2601, 1989; and Jefferis et al., Immunological Reviews 163:59-76, 1998. In some embodiments, the constant region is modified as described in Eur. J. Immunol., 29:2613-2624, 1999; PCT application number PCT/GB99/01441; and/or UK application number 9809951.8. In still other embodiments, the constant region is aglycosylated against N-glycosylation. In some embodiments, the constant region is aglycosylated against N-glycosylation by mutating glycosylated amino acid residues or flanking residues in the constant region that are part of the N-glycosylation recognition sequence. For example, the N-glycosylation site N297 can be mutated to A, Q, K, or H. See Tao et al., J. Immunology 143: 2595-2601, 1989; and Jefferis et al., Immunological Reviews 163:59-76, 1998. In some implementations, the constant region is deglycosylated against N-glycosylation. The constant region may be deglycosylated enzymatically (e.g., by removing carbohydrates via the enzyme peptide-N-glycosidase (PNGase)) or by expression against N-glycosylation in a glycosylation-deficient host cell.

Other antibody modifications include antibodies modified as described in PCT Publication No. WO99/58572. In addition to the binding domain targeting the target molecule, these antibodies also contain an effector domain having an amino acid sequence substantially homologous to all or part of the constant region of the human immunoglobulin heavy chain. These antibodies are capable of binding the target molecule without triggering significant complement-dependent cleavage or cell-mediated target disruption. In some embodiments, the effector domain is capable of specifically binding FcRn and/or FcγRIIb. These are generally based on chimeric domains derived from two or more CH2 domains of the human immunoglobulin heavy chain. Antibodies modified in this manner are particularly suitable for use in long-term antibody therapy to avoid inflammation and other adverse reactions to conventional antibody therapy.

This invention includes embodiments of affinity maturation. For example, affinity-matured antibodies can be generated by procedures known in the art (Marks et al., Bio/Technology, 10:779-783, 1992; Barbas et al., ProcNat. Acad. Sci, USA 91:3809-3813, 1994; Schier et al., Gene, 169:147-155, 1995; Yelton et al., J. Immunol., 155:1994-2004, 1995; Jackson et al., J. Immunol., 154(7):3310-9, 1995; Hawkins et al., J. Mol. Biol., 226:889-896, 1992; and PCT Publication No. WO2004/058184).

The following methods can be used to modulate antibody affinity and characterize CDRs. One way to characterize the CDR of an antibody and/or alter (e.g., improve) the binding affinity of a peptide, such as an antibody, is called “library scan mutagenesis.” Typically, library scan mutagenesis is performed as described below. Using methods recognized in the art, one or more amino acid positions in the CDR are replaced with two or more amino acids (e.g., 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20). This produces small clonal libraries (in some embodiments, one clone for each amino acid position analyzed), each clone having the complexity of two or more members (if two or more amino acids are substituted at each position). Typically, the library also includes clones containing native (unsubstituted) amino acids. For a small number of clones from each library, e.g., approximately 20–80 clones (depending on library complexity), screening is performed for binding affinity to the target peptide (or other binding target), and candidates exhibiting increased binding, identical binding, decreased binding, or no binding are identified. Methods for determining binding affinity are well known in the art. Binding affinity can be determined using BIACORE™ surface plasmon resonance analysis, which detects differences in binding affinity of approximately 2-fold or greater. BIACORE™ is particularly useful when the starting antibody has already bound with relatively high affinity (e.g., approximately 10 nM or lower K _{_D} ). Screening using BIACORE™ surface plasmon resonance is described in the examples herein.

Binding affinity can be determined using Kinexa Biocensor, scintillation affinity assay, ELISA, ORIGEN immunoassay (IGEN), fluorescence quenching, fluorescence transfer, and/or yeast display. Binding affinity can also be screened using appropriate bioassays.

In some embodiments, using art-recognized mutagenesis methods (some of which are described herein), each amino acid position in the CDR is replaced (in some embodiments, one at a time) with all 20 natural amino acids. This produces small clonal libraries (in some embodiments, one clone for each amino acid position analyzed), each clone having a complexity of 20 members (if all 20 amino acids are replaced at each position).

In some implementations, the library to be screened contains substitutions at two or more positions, which may be in the same CDR or in two or more CDRs. Thus, the library may contain substitutions at two or more positions in a single CDR. The library may contain substitutions at two or more positions in two or more CDRs. The library may contain substitutions at three, four, five, or more positions, which are found in two, three, four, five, or six CDRs. Substitutions can be prepared using low-redundancy codons. See, for example, Table 2, Balint et al., Gene 137(1):109-18, 1993.

Candidates exhibiting improved binding can be sequenced to identify CDR substitution mutants that result in improved affinity (also known as "improved" substitutions). Binding candidates can also be sequenced to identify CDR substitutions that maintain binding.

Multiple rounds of screening can be performed. For example, candidates with improved binding (each candidate containing amino acid substitutions at one or more positions of one or more CDRs) can also be used to design a second library containing at least the original and substituted amino acids at each improved CDR position (i.e., the amino acid position in the CDR where the substitution mutant shows improved binding). The preparation and screening or selection of this library are discussed further below.

Library scanning mutagenesis also provides information about the importance of each amino acid position to the stability of the antibody-antigen complex because the frequency of clones exhibiting improved binding, identical binding, reduced binding, or no binding also provides information related to the stability of the antibody-antigen complex. For example, if a CDR position remains bound when changed to all 20 amino acids, that position is identified as unlikely to be required for antigen binding. Conversely, if a CDR position remains bound only with a small percentage of substitutions, that position is identified as important for CDR function. Therefore, library scanning mutagenesis yields information about positions in CDRs that can be changed to many different amino acids (including all 20 amino acids) and positions in CDRs that cannot be changed or can only be changed to a few amino acids.

Candidates with improved affinity can be combined in a second library comprising the improved amino acid, the original amino acid at that position, and may further include additional substitutions at that position, depending on the required complexity of the library or the complexity permitted by the desired screening or selection method. Additionally, if needed, adjacent amino acid positions can be randomized to at least two or more amino acids. Randomization of adjacent amino acids allows for additional conformational flexibility in the mutant CDR, which in turn allows or facilitates the introduction of a large number of improving mutations. The library may also contain substitutions at positions that did not exhibit improved affinity in the first round of screening.

Use any method known in the art to screen or select library members in the second library that have improved and/or altered binding affinity, including screening using BIACORE™ surface plasmon resonance analysis, and selection using any method known in the art for selection, including phage display, yeast display, and ribosome display.

The techniques used for isolating and preparing antibodies are as follows. However, it will be understood that the antibodies can be formed from or incorporated into other polypeptides using techniques known in the art. For example, nucleic acids encoding a target polypeptide (e.g., a ligand, receptor, or enzyme) can be isolated from a cDNA library prepared from a tissue believed to have polypeptide mRNA and express it at detectable levels. The library is screened using probes (e.g., antibodies or oligonucleotides of about 20-80 bases) designed to identify the target gene or the protein it encodes. Screening of cDNA or genomic libraries with selected probes can be performed using standard procedures such as those described in Chapters 10-12 of Sambrook et al., Molecular Cloning: A Laboratory Manual (New York: Cold Spring Harbor Laboratory Press, 1989).

CD3 antibodies used to treat autoimmune diseases

In one aspect, a method for treating an autoimmune disease in a subject is provided, comprising administering to the subject in need an effective amount of a composition comprising an antibody as described herein.

As used in this article, autoimmune diseases include, but are not limited to, inflammatory bowel disease, systemic lupus erythematosus, rheumatoid arthritis, diabetes mellitus (type I), multiple sclerosis, Addison’s disease, celiac disease, dermatomyositis, Graves’ disease, Hashimoto’s thyroiditis, Hashimoto’s encephalopathy, myasthenia gravis, reactive arthritis, Sjogren’s syndrome, atopic allergy, atopic dermatitis, autoimmune enteropathy, autoimmune hepatitis, autoimmune lymphoproliferative syndrome, autoimmune peripheral neuropathy, autoimmune pancreatitis, autoimmune polyendocrine syndrome, and autoimmune progesterone dermatitis. Autoimmune urticaria, autoimmune uveitis, Behcet’s disease, Castleman’s disease, cold agglutinin disease, Crohn’s disease, dermatomyositis, eosinophilic fasciitis, gastrointestinal pemphigoid, Goodpasture’s syndrome, Guillain-Barré syndrome, hidradenitis suppurativa, somnambulism, pemphigus vulgaris, polymyositis, relapsing polychondritis, rheumatic fever, transverse myelitis, ulcerative colitis, undifferentiated connective tissue disease, vasculitis, and Wegener’s granulomatosis.

Multispecific antibodies and their uses

The CD3 antibodies of the present invention may be specific to different epitopes of a single target polypeptide or may contain antigen-binding domains specific to more than one target polypeptide. See, for example, Tutt et al., 1991, J. Immunol. 147:60-69; Kufer et al., 2004, Trends Biotechnol. 22:238-244. The CD3 antibodies of the present invention may be ligated to or co-expressed with another functional molecule (e.g., another peptide or protein). For example, the antibody or a fragment thereof may be functionally linked (e.g., by chemical coupling, genetic fusion, non-covalent association, or other means) to one or more other molecular entities (e.g., another antibody or antibody fragment) to produce a multispecific antibody with a second binding specificity.

In one aspect, the antibody of the present invention is a multispecific antibody. In a specific embodiment, the CD3 antibody is a bispecific antibody that specifically binds to human CD3. In some such embodiments, the bispecific antibody further specifically binds to tumor antigens. In another such embodiment, the bispecific antibody of the present invention simultaneously targets T cells (CD3) and tumor cells, and successfully guides and activates T cells to exert cytotoxicity on tumor cells expressing tumor antigens.

In the respective further embodiments described above, the CD3-binding bispecific antibody is characterized by any one or more of the following properties: (a) treating, preventing, or improving one or more symptoms of a condition in a subject associated with malignant cells expressing a specific tumor antigen (e.g., B-cell-related cancers, such as multiple myeloma); (b) inhibiting tumor growth or progression in a subject (who has a malignant tumor expressing a specific tumor antigen); (c) inhibiting metastasis of cancer (malignant) cells expressing a specific tumor antigen in a subject (who has one or more malignant cells expressing a tumor antigen); (d) inducing regression (e.g., long-term regression) of tumors expressing tumor antigens; (e) exerting cytotoxic activity in malignant cells expressing tumor antigens; (f) prolonging progression-free survival in subjects with tumor-related conditions; (g) prolonging overall survival in subjects with tumor-related conditions; (h) reducing the use of additional chemotherapy or cytotoxic agents in subjects with tumor-related conditions; (i) reducing tumor burden in subjects with tumor-related conditions; or (j) blocking the interaction of tumor antigens with other unidentified factors.

As used herein, a bispecific antibody of the present invention refers to a complex of two or more polypeptide chains, each polypeptide chain comprising at least one antibody VL region and one antibody VH region or a fragment thereof, wherein the VL and VH regions in each polypeptide chain are derived from different antibodies. In a particular aspect, a bispecific antibody comprises a dimer or tetramer of polypeptide chains containing both VL and VH regions. The polypeptide chains constituting the multimeric protein may be covalently linked to at least one other peptide of the multimer via interchain disulfide bonds.

In some such embodiments, the bispecific antibody of the present invention comprises a first heterodimer-promoting domain on a first polypeptide chain and a second heterodimer-promoting domain on a second polypeptide chain (Figure 1). In summary, the first and second heterodimer-promoting domains drive heterodimerization and/or stabilize the bispecific antibody (e.g., through the interaction of pestle and mortar on complementary heterodimer-promoting domains) and/or serve to stabilize the bispecific antibody. In some embodiments, the first and second heterodimer-promoting domains each comprise a CH2 domain and a CH3 domain, wherein the amino acid sequences of each of the CH2 domains and/or the CH3 domains are modified to drive heterodimerization and/or stabilize the bispecific antibody.

In some embodiments, the first heterodimer promoting domain may comprise an Fc chain having CH2 and/or CH3 domains modified to comprise a club (protrusion) or a mortise (cavity). In some such embodiments, the amino acid sequences of the CH2 and/or CH3 domains comprise at least one amino acid modification, wherein: (a) the CH3 domain of the first heterodimer promoting domain forms a club; and (b) the CH3 domain of the second heterodimer promoting domain forms a hole. In another such embodiment, the CH3 domain of the first heterodimer promoting domain comprises the mutants Y349C and/or T366W; and the CH3 domain of the second heterodimer promoting domain comprises the mutants S354C, T366S, L368A, and/or Y407V (according to EU index number).

In one embodiment, if the second heterodimer promoting domain comprises a CH2 and/or CH3 domain modified to comprise a mortise (cavity), then the first heterodimer promoting domain may comprise a CH2 and/or CH3 domain modified to comprise a pestle (protrusion) containing the sequence of SEQ ID NO: 78. In another embodiment, if the second heterodimer promoting domain comprises a CH2 and/or CH3 domain modified to comprise a pestle (protrusion), then the first heterodimer promoting domain may comprise a mortise (cavity) containing the sequence of SEQ ID NO: 79.

Each polypeptide chain of a bispecific antibody contains a VL region and a VH region, which are covalently linked by a glycine-serine linker (linker 1 or linker 2) containing glycine and serine residues, thus confining the antibody-binding domain to prevent self-assembly. Furthermore, each polypeptide chain contains a heterodimerization domain, which promotes heterodimerization and/or stabilization of multiple polypeptide chains and reduces the likelihood of homodimerization between different polypeptide chains. The heterodimerization domain can be located at the N-terminus or C-terminus of the polypeptide chain. The heterodimerization domain can contain a cysteine linker (linker 3) of 1, 2, 3, 4, 5, 6, or more amino acid residues in length. The interaction of two polypeptide chains can produce two VL/VH pairs, thereby forming two epitope-binding domains, i.e., divalent molecules. Neither the VH nor the VL region is confined to any position within the polypeptide chain, i.e., restricted to the amino-terminus or carboxyl-terminus, nor is the region confined by their relative positions to each other; i.e., the VL region can be at the N-terminus of the VH region, and vice versa. The only limitation is the availability of complementary polypeptide chains to form functional bispecific antibodies. In cases where the VL and VH regions originate from antibodies specific to different antigens, the formation of functional bispecific antibodies requires the interaction of two distinct polypeptide chains, i.e., the formation of heterodimers. Conversely, in cases where two distinct polypeptide chains interact freely, for example in a recombinant expression system, one containing VLA and VHB (A being the first epitope and B the second epitope) and the other containing VLB and VHA, two distinct binding sites can be formed: VLA-VHA and VLB-VHB. For all bispecific antibody polypeptide chain pairs, misalignment or misbinding of the two chains is a possibility, for example, interactions between the VL-VL or VH-VH regions. However, the purification of functional bispecific antibodies can be readily controlled based on the immunospecificity of the correctly dimerized binding sites using any affinity-based method known in the art or exemplified herein, such as affinity chromatography.

In one embodiment, the polypeptide chain of the bispecific antibody may contain various linkers and peptides. The linkers and peptides may contain 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 or more amino acids.

In some embodiments, domain 1 is covalently bound to a first heterodimer via a cysteine linker to promote the domain, and domain 2 is covalently bound to a second heterodimer via a cysteine linker to promote the domain. Each cysteine linker includes at least one cysteine residue to allow intramolecular disulfide bonding. In a further embodiment of each of the foregoing embodiments, the cysteine linker (linker 3) comprises at least five amino acids.

In some embodiments, the first polypeptide chain is covalently bound to the second polypeptide chain via at least one disulfide bond. In some such embodiments, at least one disulfide bond is formed between the linker 3 of the first polypeptide chain and the linker 3 of the second polypeptide chain. In another such embodiment, at least one disulfide bond is formed between the first heterodimer promoting domain and the second heterodimer promoting domain. In a specific embodiment, each disulfide bond is formed by linking two cysteine residues. In one aspect of the invention, the bispecific antibody shown in FIG1 comprises a first polypeptide chain and a second polypeptide chain. In some such embodiments, the linker 3 may comprise a truncated lower hinge region of human IgG1 having the sequence SEQ ID NO: 64, which has at least one glycine residue preceding the lower hinge region.

The bispecific antibody of the present invention can simultaneously bind to two separate and distinct epitopes. In some embodiments, at least one epitope binding site is specific for a CD3 determinant expressed on immune effector cells, such as T lymphocytes. In one embodiment, the bispecific antibody molecule binds to an effector cell determinant and also activates the effector cell.

In one aspect, the present invention provides a first polypeptide chain and a second polypeptide chain. In one embodiment, the first polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 36. In another embodiment, the second polypeptide chain comprises the amino acid sequence shown in one or more of SEQ ID NO: 37 or 38 (Table 7). In a preferred embodiment, the first polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 36; and the second polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 37 or 38. In one embodiment, the second polypeptide chain is any polypeptide to be associated with the first polypeptide chain via an interface. Table 7 shows the sequences of the first and second polypeptide chains of the GUCY2c-H2B4 and GUCY2c-2B5 bispecific antibodies.

Table 7

In specific embodiments, the bispecific antibody of the present invention (a) binds to the extracellular domain of a human tumor antigen; (b) exhibits an extended serum and tumor half-life of 30 minutes to 100 days; and/or (c) exhibits a lower _EC50 value of 0.0001 nM to 100 nM in the presence of increased tumor expression levels or increased receptor density levels.

In one implementation, the epitope-binding domain is capable of binding to tumor-associated antigens associated with breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer (including but not limited to non-small cell lung cancer and small cell lung cancer), bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and digestive system cancers. Digestive system cancers include, but are not limited to, cancers of the esophagus, stomach, small intestine, colon, rectum, anus, liver, gallbladder, appendix, bile ducts, and pancreas. In a specific implementation, the therapy activates a cytolytic T-cell response.

Null effector mutations in human IgG1 CH2-CH3

Using standard primer-guided PCR mutagenesis, the Fc chain of human IgG1 was modified to introduce mutants L234A, L235A, and G237A (SEQ ID NO: 82, according to EU index number) to oblate effector functions attributed to binding to FcγRIII, thereby providing an effector-ineffective phenotype (Canfield et al., J. Exp. Med (1991) 173: 1483-1491; Shields et al., J. Biol. Chem. (2001) 276: 6591-604).

The mortise in human IgG1 CH2-CH3 is a mutation.

The 'mortar' configuration is a known efficient design strategy in the art for engineering antibody heavy chain homodimers for heterodimerization. In this method, 'mortar' variants are obtained by replacing smaller amino acids, such as Y349C and T366W (according to EU index numbers), with larger amino acids in one strand of the Fc chain of IgG1. The 'mortar' is designed to insert into the 'mortar' of the CH3 domain of the complementary chain of the Fc chain generated by replacing larger residues, such as S354C, T366S, L368A, and Y407V (according to EU index numbers), with smaller residues.

In some implementations, complementary mutations are introduced to achieve heterodimerization of the resulting Fc chains, such that each Fc chain carries a set of mutations, Y349C and T366W (SEQ ID NO: 78) for the club (or protrusion) Fc chain, or S354C, T366S, L368A, and Y407V (SEQ ID NO: 79) for the mortis (or cavity) Fc chain, as provided in Table 8. When co-transfected into a suitable mammalian host, the DNA encoding the amino acid sequences (e.g., SEQ ID NO: 78 and 79) generates Fc domains predominantly with a bispecific antibody associated with one club (or protrusion) Fc chain of one mortis (or cavity) Fc chain.

Table 8

The CD3-tumor antigen bispecific antibody of this invention is a thermostability-stable, efficient bispecific antibody-Fc fusion targeting both human CD3 and tumor antigens. The Fc domain in the mortise allows for improved expression and purification in CHO cells, resulting in the desired high purity of the heterodimer. The engineered mutation within the Fc domain cancels FcγR binding, thus potentially avoiding ADCC-mediated T cell exhaustion. Furthermore, as shown by differential scanning calorimetry (DSC), incorporating the Fc domain into the bispecific antibody enhances the molecule's stability.

If a bispecific antibody can be engineered to include at least one cysteine residue, it can interact with a corresponding cysteine residue on another polypeptide chain of the invention to form an interchain disulfide bond. The interchain disulfide bond can be used to stabilize the bispecific antibody, thereby improving expression and recovery in recombinant systems, resulting in stable and consistent formulations, and improved stability of isolated and/or purified products in vivo. One or more cysteine residues can be introduced into any part of the polypeptide chain as a single amino acid or as part of a larger amino acid sequence, such as a hinge region. In a specific aspect, at least one cysteine residue is engineered to appear at the C-terminus of the polypeptide chain.

This invention covers methods and compositions for treating, preventing, or managing cancer in a subject, including administering to the subject a therapeutically effective amount of an antibody engineered according to the invention, the molecule further binding to a cancer antigen. The antibodies of this invention are particularly useful for preventing, inhibiting, reducing the growth and/or regression of primary tumors, and for the metastasis of cancer cells. Although not intended to be limited to a specific mechanism of action, the antibodies of this invention can mediate effector functions that can lead to tumor clearance, tumor reduction, or a combination thereof.

In one aspect, the present invention provides a therapeutic method for stimulating T cell activation using the CD3 antibody of the present invention, wherein the therapeutic method comprises administering a therapeutically effective amount of a pharmaceutical composition comprising the antibody of the present invention to a subject in need. The treated condition is any disease or condition that is improved, alleviated, suppressed, or prevented by stimulating CD3 activity or signaling. In a specific embodiment, the present invention provides a bispecific antigen-binding molecule, such as a bispecific antibody that binds to both CD3 and a target antigen.

In one aspect, the present invention provides a method for inhibiting tumor growth, comprising contacting the tumor with an effective amount of an antibody that binds to CD3, said antibody comprising each of the antibodies described herein.

In another aspect, the present invention provides a method for inhibiting tumor growth in a subject, comprising administering to the subject a therapeutically effective amount of an antibody binding to CD3, including each of those antibodies described herein.

In another aspect, the present invention provides a method for modulating angiogenesis in a subject, comprising administering to the subject a therapeutically effective amount of an antibody binding to CD3, including each of those antibodies described herein.

In another aspect, the present invention provides a method for reducing the tumorigenicity of tumors in a subject, comprising administering to the subject a therapeutically effective amount of an antibody binding to CD3, including each of those antibodies described herein.

In one aspect, the present invention provides a method for treating conditions in subjects associated with tumor antigen expression. Therefore, the present invention provides a method for treating tumor antigen-related conditions by engineering bispecific antibodies to immune-specifically recognize tumor antigens and CD3 antigens on T cells. Bispecific antibodies engineered according to the present invention can be used for the prevention or treatment of cancer because they possess cytotoxic activity via CD3 antibody-induced activation of cytotoxic T cells.

In one specific embodiment, the present invention provides a method for treating cancer. In this embodiment, the cancer is breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer (including but not limited to non-small cell lung cancer and small cell lung cancer), bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, or cancer of the digestive system. In some embodiments, the cancer of the digestive system is selected from cancers of the esophagus, stomach, small intestine, colon, rectum, anus, liver, gallbladder, appendix, bile ducts, and pancreas.

In one aspect, the present invention provides antibodies, bispecific antibodies, or pharmaceutical compositions as disclosed herein for use in a therapy. In one specific embodiment, the present invention also provides a CD3 bispecific antibody for use in a method of treating cancer as defined herein. In another specific embodiment, the therapy activates a cytolytic T-cell response.

This invention further provides antibodies or bispecific antibodies as disclosed herein for the preparation of medicaments for use in therapies. In some embodiments, the therapy is a treatment for cancer. In some embodiments, the cancer is selected from breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer (including but not limited to non-small cell lung cancer and small cell lung cancer), bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, or cancers of the digestive system. In some embodiments, the cancers of the digestive system are selected from cancers of the esophagus, stomach, small intestine, colon, rectum, anus, liver, gallbladder, appendix, bile ducts, and pancreas.

In one aspect, the present invention provides polynucleotides encoding antibodies or bispecific antibodies as disclosed herein. In another embodiment, the present invention provides a vector comprising a polynucleotide as disclosed herein. In yet another embodiment, the present invention provides a host cell comprising a vector as disclosed herein. In some such embodiments, the host cell recombinants produce antibodies or bispecific antibodies as disclosed herein. In specific embodiments, the host cell is selected from bacterial cell lines, mammalian cell lines, insect cell lines, yeast cell lines, and in vitro cell-free protein synthesis systems. In a specific embodiment, the mammalian cell line is a CHO cell line.

In one aspect, a method for treating cancer in a subject with this need is provided, comprising: a) providing a bispecific antibody as described herein, and b) administering the bispecific antibody to the subject.

In some embodiments, a method is provided for inhibiting tumor growth or progression in a subject with malignant cells expressing tumor antigens, comprising administering to a subject in need an effective amount of a composition comprising antibodies as described herein. In some embodiments, a method is provided for inhibiting cell metastasis expressing tumor antigens in a subject, comprising administering to a subject in need an effective amount of a composition comprising antibodies as described herein. In some embodiments, a method is provided for inducing tumor regression in malignant cells in a subject, comprising administering to a subject in need an effective amount of a composition comprising antibodies as described herein.

In a specific aspect, relative to the growth of cancer cells in the absence of the antibodies or bispecific antibodies of the present invention, the antibodies of the present invention inhibit or reduce the growth of cancer cells by at least 99%, at least 95%, at least 90%, at least 85%, at least 80%, at least 75%, at least 70%, at least 60%, at least 50%, at least 45%, at least 40%, at least 45%, at least 35%, at least 30%, at least 25%, at least 20%, or at least 10%.

In specific respects, the antibody kills cells or inhibits or reduces the growth of cancer cells by at least 5%, at least 10%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, or at least 100% better than the antibody or bispecific antibody of the present invention.

In one aspect, the present invention provides an effective amount of a composition (e.g., a pharmaceutical composition) comprising a bispecific antibody as described herein for treating a condition (e.g., cancer) associated with tumor antigen expression in a subject in need of this treatment.

In another aspect, the present invention provides antibodies as described herein for treating conditions (e.g., cancer) associated with tumor antigen expression in subjects with this need. In some embodiments, antibodies as described herein are provided for inhibiting tumor growth or progression in subjects with malignant cells expressing tumor antigens. In some embodiments, antibodies as described herein are provided for inhibiting the metastasis of malignant cells expressing tumor antigens in subjects with this need. In some embodiments, antibodies as described herein are provided for inducing tumor regression in subjects with malignant cells expressing tumor antigens.

In another aspect, the present invention provides the use of antibodies as described herein in the preparation of medicaments for treating conditions (e.g., cancer) associated with tumor antigen expression. In some embodiments, the use of antibodies as described herein is provided in the preparation of medicaments for inhibiting tumor growth or progression. In some embodiments, the use of antibodies as described herein is provided in the preparation of medicaments for inhibiting the metastasis of malignant cells expressing tumor antigens. In some embodiments, the use of antibodies as described herein is provided in the preparation of medicaments for inducing tumor regression.

In some embodiments, the method described herein further includes the step of treating the subject with another form of therapy. In some embodiments, the other form of therapy is an additional anticancer therapy, including, but not limited to, chemotherapy, radiation, surgery, hormone therapy, and/or additional immunotherapy.

This invention further encompasses the administration of the molecules of the invention in combination with other therapies known to those skilled in the art for the treatment or prevention of cancer, including but not limited to current standard and experimental chemotherapy, biotherapy, immunotherapy, radiotherapy, or surgery. In some aspects, the molecules of the invention may be administered in combination with therapeutically or preventively effective amounts of one or more agents, therapeutic antibodies, or other agents known to those skilled in the art for the treatment and/or prevention of cancer.

Therefore, methods for treating cancer include administering an effective amount of the multispecific antibody (e.g., a bispecific antibody) of the present invention in combination with a chemotherapy agent to a subject in need. This combination therapy can be administered alone, sequentially, or simultaneously.

The dosages and frequencies of administration provided herein are covered by the terms therapeutically effective and prophylactically effective. Depending on the specific therapeutic or prophylactic agent administered, the severity and type of cancer, the route of administration, and the subject's age, weight, response, and past medical history, dosage and frequency may further vary based on factors specific to each subject. Those skilled in the art can select an appropriate regimen by taking such factors into account and by following the dosages recommended, for example, those reported in the literature and in Physician's Desk Reference (56th edition, 2002).

Therefore, in addition to redirecting T cells to tumor-specific antigens, bispecific T-cell conjugate molecules can also be used to deliver other diagnostic or therapeutic compounds to cells expressing tumors on their surface. Thus, bispecific T-cell conjugate molecules can be directly or indirectly (e.g., via a connector) linked to drugs, enabling direct delivery to tumor-bearing cells. Therapeutic agents include compounds such as nucleic acids, proteins, peptides, amino acids or derivatives, glycoproteins, radioisotopes, lipids, carbohydrates, or recombinant viruses. Therapeutic and diagnostic nucleic acid portions include antisense nucleic acids, derivatized oligonucleotides for covalent cross-linking with single- or double-stranded DNA, and oligonucleotides forming triple strands.

Pharmaceutical Composition

The present invention further provides compositions comprising a therapeutically effective amount of the antibody disclosed herein and a pharmaceutically acceptable carrier.

The antibodies of the present invention may be in the form of pharmaceutical compositions for administration, formulated to suit a chosen mode of administration, and containing pharmaceutically acceptable diluents or excipients, such as buffers, surfactants, preservatives, solubilizers, isotonic agents, stabilizers, carriers, etc. Remington's Pharmaceutical Sciences, Mack Publishing Co., Easton Pa., 18th edition, 1995 provides an outline of formulation techniques generally known to those skilled in the art.

These pharmaceutical compositions can be administered in any manner known in the art to achieve the generally intended purpose of treating cancer. Routes of administration can be parenteral, and are defined herein as modes of administration including, but not limited to, intravenous, intramuscular, intraperitoneal, subcutaneous, and intra-articular injections and infusions. The manner of administration and dosage of the molecules according to the invention (e.g., antibodies, pharmaceutical compositions) depends on the type of disease to be combated, its stage in appropriate circumstances, the antigen to be controlled, the type of coexisting treatment (if any), the frequency of treatment, the characteristics of the desired effect, and also the subject's weight, age, health status, diet, and sex. Therefore, the actual dosage regimens used can vary considerably and may therefore deviate from the dosage regimens stated herein.

Various formulations of the antibodies of this invention can be used for administration. In some embodiments, the antibodies can be administered in their pure form. In some embodiments, the antibodies and pharmaceutically acceptable excipients can be contained in various formulations. Pharmaceutically acceptable excipients are known in the art and are relatively inert substances that facilitate the administration of pharmacologically effective substances. For example, excipients can impart shape or consistency or act as diluents. Suitable excipients include, but are not limited to, stabilizers, wetting agents and emulsifiers, salts for altering molar osmotic pressure concentrations, encapsulation agents, buffers, and skin penetration enhancers. Excipients, as well as formulations for parenteral and non-parenteral drug delivery, are described in Remington, The Science and Practice of Pharmacy, 21st edition. Mack Publishing, 2005.

In some implementations, these agents are formulated for administration by injection (e.g., intraperitoneal, intravenous, subcutaneous, intramuscular, etc.). Therefore, these agents can be combined with pharmaceutically acceptable mediators, such as saline, Ringer's solution, glucose solution, etc. Specific dosing regimens, i.e., dosage, timing, and repetition, will depend on the specific individual and their medical history.

The antibodies described herein can be administered using any suitable method, including by injection (e.g., intraperitoneal, intravenous, subcutaneous, intramuscular, etc.). As described herein, antibodies, such as monoclonal or bispecific antibodies, can also be administered by inhalation. Generally, the dosage for administration of the antibodies of the present invention depends on the host being treated and the specific administration method. In one embodiment, the dosage range of the antibodies of the present invention will be from about 0.001 µg/kg body weight to about 20,000 µg/kg body weight. The term “body weight” may be used when treating a patient. When processing isolated cells, “body weight” as used herein refers to “total cell weight”. The term “total weight” can be used for both isolated cells and patient treatment. In this application, all concentrations and treatment levels are expressed as “body weight” or simply “kg”, and are also considered to include similar “total cell weight” and “total weight” concentrations. However, those skilled in the art will recognize the practicality of a variety of dosage ranges, such as 0.01 µg/kg body weight to 20,000 µg/kg body weight, 0.02 µg/kg body weight to 15,000 µg/kg body weight, 0.03 µg/kg body weight to 10,000 µg/kg body weight, 0.04 µg/kg body weight to 5,000 µg/kg body weight, 0.05 µg/kg body weight to 2,500 µg/kg body weight, 0.06 µg/kg body weight to 1,000 µg/kg body weight, 0.07 µg/kg body weight to 500 µg/kg body weight, 0.08 µg/kg body weight to 400 µg/kg body weight, 0.09 µg/kg body weight to 200 µg/kg body weight, or 0.1 µg/kg body weight to 100 µg/kg body weight. Furthermore, those skilled in the art will recognize that a variety of different dosage levels will be useful, for example, 0.0001 µg/kg, 0.0002 µg/kg, 0.0003 µg/kg, 0.0004 µg/kg, 0.005 µg/kg, 0.0007 µg/kg, 0.001 µg/kg, 0.1 µg/kg, 1.0 µg/kg, 1.5 µg/kg, 2.0 µg/kg, 5.0 µg/kg, 10.0 µg/kg, 15.0 µg/kg, 30.0 µg/kg, 50 µg/kg, 75 µg/kg, 80 µg/kg, 90 µg/kg, 100 µg/kg, 120 µg/kg, 140 µg/kg. g/kg, 150 µg/kg, 160 µg/kg, 180 µg/kg, 200 µg/kg, 225 µg/kg, 250 µg/kg, 2 75µg/kg, 300 µg/kg, 325 µg/kg, 350 µg/kg, 375µg/kg, 400 µg/kg, 450 µg/k g, 500 µg/kg, 550 µg/kg, 600 µg/kg, 700 µg/kg, 750 µg/kg, 800 µg/kg, 900 µg/kg, 1 µg/kg, 5 µg/kg, 10 µg/kg, 12 µg/kg, 15 mg/kg, 20 mg/kg and/or 30 mg/kg. All these dosages are exemplary, and any dosage between these points is also expected to be useful in this invention. Any of the above dosage ranges or dosage levels can be used with the antibodies of this invention. For repeated administration over a period of several days or longer, depending on the condition, treatment may continue until the desired symptom suppression is achieved or until a sufficient therapeutic level is reached, for example, to inhibit or delay tumor growth/progression or metastasis of cancer cells.

Depending on the desired pattern of pharmacokinetic decay, other dosing regimens may also be useful. In one embodiment, the antibody of the present invention is administered at an initial initiating dose followed by higher and/or continuous, substantially constant doses. In some embodiments, dosing is anticipated from once weekly to four times weekly. In some embodiments, dosing is anticipated from once monthly, every other month, or every three months. Progression of the therapy can be easily monitored using routine techniques and assays. The dosing regimen may vary over time.

For the purposes of this invention, the appropriate dosage of the antibody will depend on the antibody or a combination thereof used, the type and severity of the symptoms to be treated, whether the agent is administered for therapeutic purposes, prior therapy, the patient's clinical history and response to the agent, the patient's clearance rate of the administered agent, and the physician's judgment. Typically, the physician will administer the antibody until a dosage is reached to achieve the desired outcome. The dosage and/or frequency may vary during the treatment process. Empirical considerations such as half-life often aid in dosage determination. For example, antibodies compatible with the human immune system, such as humanized antibodies or fully human antibodies, can be used to prolong the antibody's half-life and prevent the antibody from being attacked by the host immune system. The frequency of administration can be determined and adjusted during the treatment process and is generally, but not necessarily, based on the treatment and/or suppression and/or improvement and/or delay of symptoms, such as tumor growth inhibition or delay. Alternatively, a sustained-release formulation of the antibody may be suitable. Various formulations and devices for achieving sustained release are known in the art.

In one implementation, the antibody dosage can be empirically determined in individuals who have received one or more doses of the antibody. Incremental doses of the antibody are administered to the individuals. Indicators of the disease can be tracked to assess efficacy.

In some implementations, the administration of the antibody results in at least one effect selected from the following: inhibition of tumor growth, tumor regression, reduction of tumor size, reduction of tumor cell number, delay of tumor growth, abscopal effect, inhibition of tumor metastasis, reduction of metastatic lesions over time, reduction of use of chemotherapy agents or cytotoxic agents, reduction of tumor burden, increase of progression-free survival, increase of overall survival, complete response, partial response, and stable disease.

Depending on factors such as the recipient's physiological condition, whether the purpose of administration is therapeutic or prophylactic, and other factors known to a skilled professional, the administration of antibodies according to the method of the invention can be continuous or intermittent. Antibody administration can be substantially continuous over a preselected time period, or it can be a series of doses spaced apart from each other.

In some implementations, more than one antibody may be present. At least one, at least two, at least three, at least four, at least five, or more different antibodies may be present. Typically, these antibodies may have complementary activities that do not harmfully interfere with each other. For example, one or more of the following antibodies may be used: a first CD3 antibody targeting one epitope on CD3 and a second CD3 antibody targeting a different epitope on CD3.

In some embodiments, the antibodies disclosed herein can be used, in particular, for parenteral administration, such as intravenous administration or administration into the cavities of body cavities or organs.

Therapeutic formulations of the antibodies used according to the invention are prepared for storage by mixing antibodies of desired purity with optional pharmaceutically acceptable carriers, excipients, or stabilizers (Remington, The Science and Practice of Pharmacy, 21st edition. Mack Publishing, 2005) in the form of lyophilized formulations or aqueous solutions. Acceptable carriers, excipients, or stabilizers are non-toxic to the recipient at the doses and concentrations used and may include buffers such as phosphates, citrates, and other organic acids; salts such as sodium chloride; antioxidants, including ascorbic acid and methionine; preservatives (e.g., octadecyl dimethyl benzyl ammonium chloride; hexane diammonium chloride; benzalkonium chloride, benzyl chloride; phenols, butanol, or benzyl alcohol; alkyl esters of p-hydroxybenzoate, such as methylparaben or propylparaben; catechol; resorcinol; cyclohexanol; 3-pentanol; and m-cresol); low molecular weight (small) The following are considered as salts: polypeptides (approximately 10 residues); proteins, such as serum albumin, gelatin, or immunoglobulins; hydrophilic polymers, such as polyvinylpyrrolidone; amino acids, such as glycine, glutamine, asparagine, histidine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates, including glucose, mannose, or dextrin; chelating agents, such as EDTA; sugars, such as sucrose, mannitol, trehalose, or sorbitol; salt-forming ions, such as sodium; metal complexes (e.g., Zn-protein complexes); and/or nonionic surfactants, such as TWEEN ^™ , PLURONICS ^™ , or polyethylene glycol (PEG).

Antibody-containing liposomes are prepared by methods known in the art, for example, as described in Epstein et al., Proc. Natl. Acad. Sci. USA 82:3688, 1985; Hwang et al., Proc. Natl. Acad. Sci. USA 77:4030, 1980; and U.S. Patent Nos. 4,485,045 and 4,544,545. Liposomes with extended cycle times are disclosed in U.S. Patent No. 5,013,556. Particularly useful liposomes can be produced by a reverse-phase evaporation method using a lipid composition comprising phosphatidylcholine, cholesterol, and PEG-derived phosphatidylethanolamine (PEG-PE). The liposomes are extruded through a filter having a defined pore size to produce liposomes with a desired diameter.

The active ingredient can also be encapsulated in microcapsules (e.g., hydroxymethyl cellulose or gelatin microcapsules and poly-(methyl methacrylate) microcapsules) prepared by agglomeration technology or interfacial polymerization, colloidal drug delivery systems (e.g., liposomes, albumin microspheres, microemulsions, nanoparticles and nanocapsules) or coarse droplet emulsions. This technology is disclosed in Remington, The Science and Practice of Pharmacy, 21st edition, Mack Publishing, 2005.

Sustained-release formulations can be prepared. Suitable examples of sustained-release formulations include semi-permeable matrices of solid hydrophobic polymers containing antibodies, said matrices being in the form of shaped articles, such as membranes or microcapsules. Examples of sustained-release matrices include polyesters, hydrogels (e.g., poly(2-hydroxyethyl methacrylate) or poly(vinyl alcohol)), polylactides (US Patent No. 3,773,919), copolymers of L-glutamic acid and L-glutamic acid 7-ethyl ester, non-degradable ethylene vinyl acetate, degradable lactic acid-glycolic acid copolymers, such as LUPRON DEPOT ^™ (injectable microspheres composed of lactic acid-glycolic acid copolymers and leuprolide acetate), sucrose isobutyrate acetate, and poly-D-(-)-3-hydroxybutyrate.

Preparations intended for in vivo administration must be sterile. This can be easily achieved, for example, by filtration through a sterile membrane. Therapeutic antibody (e.g., CD3 antibody) compositions are typically placed in containers with sterile inlets and outlets, such as intravenous solution bags or vials with stoppers that can be punctured by a subcutaneous needle.

The compositions according to the invention may be in unit dosage form, such as tablets, pills, capsules, powders, granules, solutions or suspensions, or suppositories, for oral, parenteral or rectal administration, or administration by inhalation or blowing.

To prepare solid compositions such as tablets, the main active ingredient is mixed with a pharmaceutical carrier (e.g., conventional tableting ingredients such as corn starch, lactose, sucrose, sorbitol, talc, stearic acid, magnesium stearate, dicalcium phosphate, or gum, and other pharmaceutical diluents, such as water) to form a solid preformulation composition containing a homogeneous mixture of the compound of the present invention or a pharmaceutically acceptable, non-toxic salt thereof. When these preformulation compositions are referred to as homogeneous, it means that the active ingredient is uniformly dispersed throughout the composition, making it easy to subdivide the composition into equivalent effective unit dose forms, such as tablets, pills, and capsules. The solid preformulation composition is then subdivided into unit dose forms of the type described above, containing 0.1 to about 500 mg of the active ingredient of the present invention. Tablets or pills of the novel composition can be coated or otherwise compounded to provide a dosage form that provides the advantage of prolonged action. For example, tablets or pills may contain an internal dose component and an external dose component, the latter being in the form of an outer coating on top of the former. These two components can be separated by an enteric layer, which resists breakdown in the stomach and allows the internal components to enter the duodenum intact or with delayed release. A variety of materials can be used for this enteric layer or coating, including many polymeric acids and mixtures of polymeric acids with materials such as shellac, cetyl alcohol, and cellulose acetate.

Suitable surfactants particularly include nonionic agents such as polyoxyethylene sorbitol (e.g., Tween ^™ 20, 40, 60, 80, or 85) and other sorbitols (e.g., Span ^™ 20, 40, 60, 80, or 85). Compositions containing surfactants will conveniently contain 0.05 to 5% surfactant, and may be 0.1 to 2.5%. It will be understood that, where necessary, other ingredients, such as mannitol or other pharmaceutically acceptable mediators, may be added.

The compositions within the scope of this invention include all compositions in which the antibody is present in an amount that effectively achieves the desired medical effect for treating cancer. Although individual needs may differ from one patient to another, the determination of the optimal range of effective amounts of all components is within the capabilities of a clinician with ordinary skills.

Examples of such compositions and how they are formulated have also been described in earlier sections and below. In some embodiments, the composition comprises one or more antibodies. In some embodiments, the antibody is a human antibody, a humanized antibody, or a chimeric antibody. In some embodiments, the antibody comprises a constant region capable of triggering a desired immune response (e.g., antibody-mediated cleavage or ADCC). In some embodiments, the antibody comprises a constant region that does not trigger unwanted or undesirable immune responses (e.g., antibody-mediated cleavage or ADCC).

It should be understood that the composition may contain more than one antibody, such as a mixture of CD3 antibodies that recognize CD3 or different epitopes of CD3 and tumor antigens. Other exemplary compositions contain more than one antibody that recognizes one or more of the same epitopes, or different kinds of antibodies, or different epitopes of CD3 and tumor antigens (e.g., human CD3).

In some embodiments, the antibody may be administered in combination with the administration of one or more additional therapeutic agents. These include, but are not limited to, the administration of biological and/or chemotherapeutic agents, such as, but not limited to, vaccines, CAR-T cell-based therapies, radiotherapy, cytokine therapy, CD3 bispecific antibodies, inhibitors of other immunosuppressive pathways, inhibitors of angiogenesis, T-cell activators, inhibitors of metabolic pathways, mTOR inhibitors, inhibitors of the adenosine pathway, tyrosine kinase inhibitors, including but not limited to Inlyta, ALK inhibitors and sunitinib, BRAF inhibitors, epigenetic modifying factors, IDO1 inhibitors, JAK inhibitors, STAT inhibitors, and cell cycle proteins. Leukocyte-dependent kinase inhibitors, biotherapeutic agents (including but not limited to antibodies against VEGF, VEGFR, EGFR, Her2/neu, other growth factor receptors, CD40, CD-40L, CTLA-4, OX-40, 4-1BB, TIGIT, and ICOS), immunogenic agents (e.g., attenuated cancer cells, tumor antigens, antigen-presenting cells such as dendritic cells bombarded with tumor-derived antigens or nucleic acids, immunostimulatory cytokines (e.g., IL-2, IFNα2, GM-CSF), and cells transfected with genes encoding immunostimulatory cytokines (e.g., but not limited to GM-CSF).

Examples of biotherapeutic agents include therapeutic antibodies, immunomodulators, and therapeutic immune cells.

Therapeutic antibodies may be specific to a variety of different antigens. For example, a therapeutic antibody may target a tumor-associated antigen, such that the binding of the antibody to the antigen promotes the death of cells expressing the antigen. In other instances, a therapeutic antibody may target an antigen on immune cells (e.g., PD-1), such that the binding of the antibody prevents the downregulation of the activity of cells expressing the antigen (and thereby promotes the activity of cells expressing the antigen). In some cases, a therapeutic antibody may act through a variety of different mechanisms (e.g., upon contact with cells expressing the antigen, it may both i) promote the death of cells expressing the antigen and ii) prevent the downregulation of immune cell activity caused by the antigen).

Therapeutic antibodies can target antigens such as those listed below. For some antigens, exemplary antibodies against said antigens are also included below (in parentheses after the antigen). The antigens listed below may also be referred to herein as "target antigens," etc. The target antigens used for therapeutic antibodies in this article include, for example: 4-1BB (e.g., utomilumab); 5T4; A33; α-folate receptor 1 (e.g., mirvetuximab soravtansine); Alk-1; BCMA [e.g., PF-06863135 (see US9969809)]; BTN1A1 (e.g., see WO2018222689); CA-125 (e.g., abagovomab); carbonic anhydrase IX; CCR2; CCR4 (e.g., mogamulizumab); CCR5 (e.g., leronlimab); CCR8; CD3 [e.g., blinatumomab (CD3/ CD19 bispecific), PF-06671008 (CD3/P-cadherin bispecific), PF-06863135 (CD3/BCMA bispecific), PF-07062119 (CD3/GUCY2c bispecific), CD19 (e.g., ibritumomab tiuxetan, MOR208); CD20 (e.g., ibritumomab tiuxetan, obinutuzumab, ofatumumab, rituximab, ublituximab); CD22 (inotuzumab ozogamicin, moxetumomab pasudotox); CD25; CD28; CD30 (e.g., brentuximab vedotin); CD33 (e.g., gemtuzumab ozogamicin); CD38 (e.g., daratumumab, isatuximab); CD40; CD-40L; CD44v6; CD47; CD52 (e.g., alemtuzumab); CD63; CD79 (e.g., polotuzumab vedotin); CD80; CD123; CD276/B7-H3 (e.g., omburtamab); CDH17; CEA; ClhCG; CTLA-4 (e.g., ipilimumab, tremel... imumab), CXCR4; desmosome core protein 4; DLL3 (e.g., rovalpituzumab tesirine); DLL4; E-cadherin; EDA; EDB; EFNA4; EGFR (e.g., cetuximab, depatuxizumab mafodotin, necitumumab, panitumumab); EGFRvIII; endothelial sialic acid protein; EpCAM (e.g., oportuzumab monatox); FAP; fetal acetylcholine receptor; FLT3 (e.g., see WO2018/220584); GD2 (e.g., dinutuximab, 3F8); GD3 ; GITR; GloboH; GM1; GM2; GUCY2C (e.g., PF-07062119); HER2/neu [e.g., margetuximab, pertuzumab, trastuzumab; ado-trastuzumab emtansine, trastuzumab duocarmazine, PF-06804103 (see US8828401)]; HER3; HER4; ICOS; IL-10; ITG-AvB6; LAG-3 (e.g., relatlimab); Lewis-Y; LG; Ly-6; M-CSF [e.g., PD-0360324 (see US8828401)]; HER3; HER4; ICOS; IL-10; ITG-AvB6; LAG-3 (e.g., relatlimab); Lewis-Y; LG; Ly-6; M-CSF [e.g., PD-0360324 (see US8828401)]; S7326414); MCSP; mesothelin; MUC1; MUC2; MUC3; MUC4; MUC5AC; MUC5B; MUC7; MUC16; Notch 1; Notch 3; Nectin-4 (e.g., enfortumab vedotin); OX40 [e.g., PF-04518600 (see US7960515)]; P-cadherein [e.g., PF-06671008 (see WO2016/001810)]; PCDHB2; PD-1 [e.g., BCD-100, camrelizumab, cemiplimab, genolimab] Nivolumab (CBT-501), MEDI0680 (CBT-501), nivolumab, pembrolizumab, RN888 (see WO2016/092419), sintilimab, spartalizumab, STI-A1110, tislelizumab, TSR-042; PD-L1 (e.g., atezolizumab, durvalumab, BMS-936559 (MDX-1105) or LY3300054); PDGFRA (e.g., olaratumab); plasma cell antigen; PolySA; PSCA; PSMA; PTK7 [e.g., PF-06647020 (see US9409995)]; Ror1; SAS; SCRx6; SLAMF7 (e.g., elotuzumab); SHH; SIRPa (e.g., ED9, Effi-DEM); STEAP; TGF-β; TIGIT; TIM-3; TMPRSS3; TNF-α precursor; TROP-2 (e.g., sacituzumabgovitecan); TSPAN8; VEGF (e.g., bevacizumab, brolucizumab); VEGFR1 (e.g., ranibizumab); VEGFR2 (e.g., ramucirumab, ranibizumab); Wue-1.

Therapeutic antibodies can take any suitable form. For example, therapeutic antibodies can take any form described elsewhere herein. In some embodiments, the therapeutic antibody may be a naked antibody. In some embodiments, the therapeutic antibody may be conjugated to a drug/pharmaceutical agent (also known as an "antibody-drug conjugate" (ADC)). In some embodiments, a therapeutic antibody targeting a specific antigen may be incorporated into a multispecific antibody (e.g., a bispecific antibody).

In some embodiments, antibodies against antigens may be conjugated to drugs/pharmaceuticals. The linked antibody-drug molecule is also referred to as an "antibody-drug conjugate" (ADC). The drug/pharmaceutical can be linked to the antibody directly or indirectly via a linker. Most commonly, toxic drugs are linked to antibodies such that the binding of the ADC to the corresponding antigen promotes the killing of cells expressing said antigen. For example, ADCs linked to toxic drugs are particularly useful for targeting tumor-associated antigens to promote the killing of tumor cells expressing tumor-associated antigens. In other embodiments, the agent that can be linked to the antibody may be, for example, an immunomodulator (e.g., to modulate the activity of immune cells near the ADC), an imaging agent (e.g., to facilitate imaging of the ADC in a subject or a biological sample from a subject), or an agent that increases the serum half-life or biological activity of the antibody.

Methods for conjugating cytotoxic agents or other therapeutic agents to antibodies have been described in various publications. For example, antibodies can be chemically modified to induce conjugation by lysine side-chain amines or by cysteine thiol groups activated by reduced interchain disulfide bonds. See, for example, Tanaka et al., FEBS Letters 579:2092-2096, 2005 and Gentle et al., Bioconjugate Chem. 15:658-663, 2004. Reactive cysteine residues engineered at specific sites on antibodies for conjugation to specific drugs at defined stoichiometric amounts have also been described. See, for example, Junutula et al., Nature Biotechnology, 26:925-932, 2008. International applications WO2012/059882 and WO2015015448 also describe the use of glutamine-containing tags with acyl donors or endogenous glutamine conjugated by peptide engineering to an active state (i.e., the ability to form covalent bonds as acyl donors) in the presence of transglutaminase and amines (e.g., cytotoxic agents containing or attached to active amines). In some embodiments, the ADC may have any of the features or characteristics of the ADCs provided in WO2016166629, which is incorporated herein by reference for all purposes.

Drugs/pharmaceuticals that can be linked to antibodies in the form of ADCs can include, for example, cytotoxic agents, immunomodulators, imaging agents, therapeutic proteins, biopolymers, or oligonucleotides.

Exemplary cytotoxic agents that can be incorporated into ADCs include anthracyclines, auristatin, dolastatin, compbretastatin, duocarmycin, pyrrolobenzodiazepine dimer, indoline-benzodiazepine dimer, enediyne, geldemycin, maytansine, puromycin, taxane, vinca alkaloids, camptothecin, tubulysin, hemiasterlin, spliceostatin, pladienolide, and their stereoisomers, isomers, analogs, or derivatives.

Exemplary immunomodulators that can be incorporated into ADCs include gancyclovier, etanercept, tacrolimus, sirolimus, voclosporin, cyclosporine, rapamycin, cyclophosphamide, azathioprine, mycophenolate mofetil, methotrexate, glucocorticoids and their analogues, cytokines, and stem cell production. Growth factors, lymphotoxin, tumor necrosis factor (TNF), hematopoietic factors, interleukins (e.g., interleukin-1 (IL-1), IL-2, IL-3, IL-6, IL-10, IL-12, IL-18, and IL-21), colony-stimulating factors (e.g., granulocyte colony-stimulating factor (G-CSF) and granulocyte-macrophage colony-stimulating factor (GM-CSF)), interferons (e.g., interferon-α, -β, and -γ), stem cell growth factors known as "S1 factors", erythropoietin, and thrombopoietin, or combinations thereof.

Exemplary imaging agents that may be included in an ADC include fluorescein, rhodamine, lanthanide phosphors and their derivatives, or radioisotopes bound to a chelating agent. Examples of fluorophores include, but are not limited to, fluorescein isothiocyanate (FITC) (e.g., 5-FITC), fluorescein phosphoramide (FAM) (e.g., 5-FAM), eosin, carboxyfluorescein, erythrosine, AlexaFluor.RTM. (e.g., Alexa 350, 405, 430, 488, 500, 514, 532, 546, 555, 568, 594, 610, 633, 647, 660, 680, 700, or 750), carboxytetramethylrhodamine (TAMRA) (e.g., 5-TAMRA), tetramethylrhodamine (TMR), and sulforhodamine (SR) (e.g., SR101). Examples of chelating agents include, but are not limited to, 1,4,7,10-tetraazacyclododecane-N,N',N'',N'''-tetraacetic acid (DOTA), 1,4,7-triazacyclononane-1,4,7-triacetic acid (NOTA), 1,4,7-triazacyclononane, 1-glutaric acid-4,7-acetic acid (deferroamine), diethylenetriaminepentaacetic acid (DTPA), and 1,2-bis(o-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA).

Exemplary therapeutic proteins that may be included in ADCs include toxins, hormones, enzymes, and growth factors.

Exemplary biocompatible polymers that can be incorporated into ADCs include water-soluble polymers, such as polyethylene glycol (PEG) or derivatives thereof, and biocompatible polymers containing zwitterions (e.g., polymers containing phosphocholine).

Exemplary biocompatible polymers that can be incorporated into ADCs include antisense oligonucleotides.

In some implementations, the antibodies against antigens provided herein can be incorporated into bispecific antibody molecules. A bispecific antibody is a monoclonal antibody that has binding specificity to at least two different antigens.

In some embodiments, the bispecific antibody comprises a first antibody variable domain and a second antibody variable domain, wherein the first antibody variable domain is capable of recruiting human immune effector cells by specifically binding to CD3 as provided herein, and wherein the second antibody variable domain is capable of specifically binding to a target antigen. In some embodiments, the antibody has an IgG1, IgG2, IgG3, or IgG4 isotype. In some embodiments, the antibody comprises an immune-inert Fc region. In some embodiments, the antibody is a human antibody or a humanized antibody.

Target antigens are typically expressed on target cells (e.g., cancer cells) in disease conditions. Examples of target antigens of particular interest for bispecific antibodies include, but are not limited to, BCMA, EpCAM (epithelial cell adhesion molecule), CCR5 (type 5 chemokine receptor), CD19, HER (human epidermal growth factor receptor)-2/neu, HER-3, HER-4, EGFR (epidermal growth factor receptor), PSMA, CEA, MUC-1 (mucin), MUC2, MUC3, MUC4, MUC5AC, MUC5B, MUC7, CIhCG, Lewis-Y, CD20, CD33, CD30, ganglioside GD3, 9-O-acetyl-GD3, GM2, Globo H, fucose GM1, Poly SA, GD2, carbonic anhydrase IX (MN/CA IX), CD44v6, and Shh (Sonic Hedgkinase). ehog), Wue-1, plasma cell antigen, (membrane-bound) IgE, MCSP (chondroitin sulfate proteoglycan for melanoma), CCR8, TNF-α precursor, STEAP, mesothelin, A33 antigen, PSCA (prostate stem cell antigen), Ly-6; desmosome core glycoprotein 4, E-cadherin epitope, fetal acetylcholine receptor, CD25, CA19-9 marker, CA-125 marker and MIS (Muellerian Inhibitory Substance) type II receptor, sTn (sialylated Tn antigen; TAG-72), FAP (fibroblast activation antigen), endothelial sialic acid protein, EGFRvIII, LG, SAS, PD-L1, CD47, SIRPa and CD63.

In some embodiments, the bispecific antibody comprises a full-length human antibody, wherein the first antibody variable domain of the bispecific antibody is capable of recruiting human immune effector cells by specifically binding to CD3, and wherein the second antibody variable domain of the heterodimeric protein is capable of specifically binding to a target antigen (e.g., CD20, EpCAM, or P-cadherin).

Methods for preparing bispecific antibodies are known in the art (see, for example, Suresh et al., Methods in Enzymology 121:210, 1986). Traditionally, recombinant production of bispecific antibodies is based on the co-expression of two immunoglobulin heavy-light chain pairs, where the two heavy chains have different specificities (Millstein and Cuello, Nature 305, 537-539, 1983).

According to a method for preparing bispecific antibodies, an antibody variable domain having the desired binding specificity (antibody-antigen combination site) is fused to an immunoglobulin constant region sequence. The fusion preferably has an immunoglobulin heavy chain constant region comprising at least a portion of the hinge, CH2, and CH3 regions. Preferably, it has a first heavy chain constant region (CH1) present in at least one of the fusions, containing the site necessary for light chain binding. DNA encoding the immunoglobulin heavy chain fusion and (if desired) the immunoglobulin light chain is inserted into separate expression vectors and co-transfected into a suitable host organism. This provides great flexibility in the embodiments to adjust the relative proportions of the three polypeptide fragments when unequal ratios of the three polypeptide chains used in the construction provide optimal yield. However, when expression of at least two polypeptide chains in equal ratios results in high yield, or when the ratios are not particularly significant, it is possible to insert coding sequences for two or all three polypeptide chains into a single expression vector.

In one method, the bispecific antibody comprises a hybrid immunoglobulin heavy chain having a first binding specificity in one arm and a hybrid immunoglobulin heavy chain-light chain pair (providing a second binding specificity) in the other arm. This asymmetric structure, with the immunoglobulin light chain in only half of the bispecific molecule, facilitates the separation of the desired bispecific compound from the unwanted combination of immunoglobulin chains. This method is described in PCT Publication No. WO 94/04690.

In another approach, the bispecific antibody is constructed by modifying an amino acid in a first hinge region of one arm, wherein the substituted/alternated amino acid in the first hinge region has an opposite charge to the corresponding amino acid in a second hinge region of the other arm. This approach is described in International Patent Application No. PCT/US2011/036419 (WO2011/143545).

In another approach, the formation of a desired heteromeric or heterodimeric protein (e.g., a bispecific antibody) is enhanced by altering or engineering the interface between the first and second immunoglobulin-like Fc regions (e.g., hinge regions and/or CH3 regions). In this approach, the bispecific antibody may be composed of a CH3 region containing a first and a second CH3 polypeptide that interact together to form a CH3 interface, wherein one or more amino acids within the CH3 interface destabilize homodimer formation and are electrostatically unfavorable to homodimer formation. This approach is described in International Patent Application No. PCT/US2011/036419 (WO2011/143545).

In another approach, bispecific antibodies can be generated in the presence of transglutaminase using a glutamine-containing peptide tag engineered for an antibody targeting an epitope (e.g., BCMA) in one arm, and another peptide tag engineered for a second antibody targeting a second epitope in the other arm (e.g., a Lys-containing peptide tag or reactive endogenous Lys). This approach is described in International Patent Application No. PCT/IB2011/054899 (WO2012/059882).

In some embodiments, the first and second antibody variable domains of the bispecific antibody include amino acid modifications at various positions, wherein the first and second antibody variable domains of the bispecific antibody include amino acid modifications at positions 223, 225, and 228 in the hinge region of human IgG2 (e.g., (C223E or C223R), (E225R), and (P228E or P228R)) and amino acid modifications at positions 409 or 368 in the CH3 region (e.g., K409R or L368E (EU numbering scheme)).

In some embodiments, the first and second antibody variable domains of the bispecific antibody include amino acid modifications at positions 221 and 228 in the hinge region of human IgG1 (e.g., (D221R or D221E) and (P228R or P228E)) and amino acid modifications at positions 409 or 368 in the CH3 region (e.g., K409R or L368E (EU numbering scheme)).

In some embodiments, the first and second antibody variable domains of the bispecific antibody include an amino acid modification at position 228 in the hinge region of human IgG4 (e.g., (P228E or P228R)) and an amino acid modification at position 409 or 368 in the CH3 region (e.g., R409 or L368E (EU numbering scheme)).

In some implementations, the bispecific antibody may have any of the features or properties of any bispecific antibody provided in WO2016166629, which is incorporated herein by reference for all purposes.

Immunomodulators include a variety of different molecular types that can stimulate immune responses in subjects, such as pattern recognition receptor (PRR) agonists, immunostimulatory cytokines, and cancer vaccines.

Pattern recognition receptors (PRRs) are receptors expressed by cells of the immune system that recognize a variety of molecules associated with pathogens and/or cell damage or death. PRRs are involved in both innate and adaptive immune responses. PRR agonists can be used to stimulate immune responses in subjects. Several classes of PRR molecules exist, including toll-like receptors (TLRs), RIG-I-like receptors (RLRs), nucleotide-binding oligomerization domain (NOD)-like receptors (NLRs), C-type lectin receptors (CLRs), and interferon gene stimulator (STING) proteins.

The terms “TLR” and “toll-like receptor” refer to any toll-like receptor. Toll-like receptors are receptors involved in activating immune responses. TLRs recognize, for example, pathogen-associated molecular patterns (PAMPs) expressed in microorganisms, and endogenous damage-associated molecular patterns (DAMPs) released from dying or dying cells.

Molecules that activate TLRs (and thus activate the immune response) are referred to herein as “TLR agonists.” TLR agonists can include, for example, small molecules (e.g., organic molecules with a molecular weight of less than about 1000 Daltons) as well as large molecules (e.g., oligonucleotides and proteins). Some TLR agonists are specific to a single type of TLR (e.g., TLR3 or TLR9), while others activate two or more types of TLRs (e.g., both TLR7 and TLR8).

The exemplary TLR agonists provided herein include agonists of TLR2, TLR3, TLR4, TLR5, TLR6, TLR7, TLR8, and TLR9.

Exemplary small molecule TLR agonists include, for example, U.S. Patent Nos. 4,689,338; 4,929,624; 5,266,575; 5,268,376; 5,346,905; 5,352,784; 5,389,640; 5,446,153; 5,482,936; 5,756,747; 6,110,929; 6,194,425; 6,331,539; 6,376,669; 6,451,810; 6,525,064; 6,541,485; 6,545,016; 6,545,017; 6,573,273; 6,656,938; 6,660,735; 6 660,747; 6,664,260; 6,664,264; 6,664,265; 6,667,312; 6,670,372; 6,677,347; 6,677,348; 6,677,349; 6,683,088; 6,756,382; 6,797,718; 6,818,650; and 7,7091,214; U.S. Patent Publications 2004/0091491, 2004/0176367 and 2006/0100229; and International Publications WO 2005/18551, WO2005/18556, WO 2005/20999, WO 2005/0 32484, WO 2005/048933, WO 2005/048945, WO 2005/051317, WO2005/051324, WO 2005/066169, WO 2005/066170, WO 2005/066172, WO 2005/076783, WO 2005/079195, WO2005/094531, WO 2005/123079, WO 2005/123080, WO 2006/009826, WO 2006/009832, WO 2006/026760, WO2006/028451 WO 2006/028545、WO 2006/028962、WO 2006/029115、WO 2006/038923、WO 2006/065280、WO2006/074003、WO 2006/083440、WO 2006/086449、WO 2006/091394, WO 2006/086633, WO 2006/086634, WO 2006/091567, WO 2006/091568, WO 2006/091647, WO 2006/093514 and WO 2006/098852.

Additional examples of small molecule TLR agonists include certain purine derivatives (such as those described in U.S. Patent Nos. 6,376,501 and 6,028,076), certain imidazoquinoline amide derivatives (such as those described in U.S. Patent No. 6,069,149), certain imidazopyridine derivatives (such as those described in U.S. Patent No. 6,518,265), and certain benzimidazole derivatives (such as those described in U.S. Patent No. 6,387,938). Certain derivatives of 4-aminopyrimidine fused with a five-membered nitrogen-containing heterocycle (e.g., adenine derivatives described in U.S. Patent Nos. 6,376,501, 6,028,076, and 6,329,381; and WO 02/08905) and certain 3-β-D-furanoribosylthiazo[4,5-d]pyrimidine derivatives (e.g., those described in U.S. Publication No. 2003/0199461) and certain small molecule immune enhancer compounds, such as those described in U.S. Patent Publication No. 2005/0136065.

Exemplary macromolecular TLR agonists comprise oligonucleotide sequences. Some TLR agonist oligonucleotide sequences contain cytosine-guanine dinucleotides (CpG) and are described, for example, in U.S. Patent Nos. 6,194,388; 6,207,646; 6,239,116; 6,339,068; and 6,406,705. Some CpG-containing oligonucleotides may include synthetic immunomodulatory structural motifs, such as those described, for example, in U.S. Patent Nos. 6,426,334 and 6,476,000. Other TLR agonist nucleotide sequences lack the CpG sequence and are described, for example, in International Patent Publication No. WO 00/75304. Other TLR agonist nucleotide sequences include single-stranded RNAs (ssRNAs) rich in guanosine and uridine, such as those described, for example, in Heil et al., Science, vol. 303, pp. 1526-1529, March 5, 2004.

Other TLR agonists include biomolecules such as aminoalkyl glucosinolate phosphates (AGPs), and are described, for example, in U.S. Patent Nos. 6,113,918; 6,303,347; 6,525,028; and 6,649,172.

TLR agonists also include inactivated pathogens or fractions thereof, which can activate a variety of different types of TLR receptors. Exemplary pathogen-derived TLR agonists include BCG, mycobacterium obuense extract, Talimogene laherparepvec (T-Vec) (derived from HSV-1), and Pexa-Vec (derived from vaccinia virus).

In some implementations, the TLR agonist may be an agonist antibody that specifically binds to a TLR.

The following provides a brief description of various TLRs and TLR agonists. The list of TLR agonists for a particular TLR below should not be interpreted as indicating that a given TLR agonist necessarily activates only that TLR (e.g., some molecules can activate multiple types of TLRs, or even multiple classes of PRRs). For example, some molecules provided below as exemplary TLR4 agonists could also be TLR5 agonists.

The terms “TLR1” and “toll-like receptor 1” refer to any form of TLR1 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR1 activity. Unless otherwise specified, such as by way of a specific reference to human TLR1, TLR1 includes the naturally occurring sequence TLR1 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR1 is provided under UniProt entry number Q15399.

As used herein, “TLR1 agonist” means any molecule that, upon binding to TLR1, (1) stimulates or activates TLR1, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR1, or (3) enhances, increases, promotes, or induces the expression of TLR1. TLR1 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, bacterial lipoproteins and their derivatives that bind to TLR1.

Examples of TLR1 agonists useful in the treatments, medicines, and uses of the present invention include, for example, bacterial lipoproteins and their derivatives, such as SPM-105 (derived from autoclaved mycobacteria), OM-174 (lipid A derivative), OmpS1 (porin from Salmonella typhi), OspA (from Borrelia burgdorferi), MALP-2 (mycoplasma macrophage-activated lipopeptide-2kD), STF (soluble tuberculosis factor), CU-T12-9, Diprovocim, and lipopeptides derived from cell wall components, such as PAM2CSK4, PAM3CSK4, and PAM3Cys.

TLR1 can form a heterodimer with TLR2, and therefore, many TLR1 agonists are also TLR2 agonists.

The terms “TLR2” and “toll-like receptor 2” refer to any form of TLR2 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR2 activity. Unless otherwise specified, such as by way of a specific reference to human TLR2, TLR2 includes naturally occurring sequence TLR2 in all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR2 is provided under UniProt entry number O60603.

As used herein, “TLR2 agonist” means any molecule that, upon binding to TLR2, (1) stimulates or activates TLR2, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR2, or (3) enhances, increases, promotes, or induces the expression of TLR2. TLR2 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, bacterial lipoproteins and their derivatives that bind to TLR2.

Examples of TLR2 agonists useful in the treatments, medicines, and uses of the present invention include, for example, bacterial lipoproteins (e.g., diacylated lipoproteins) and their derivatives, such as SPM-105 (derived from autoclaved mycobacteria), OM-174 (lipid A derivative), OmpS1 (porin from Salmonella typhi), OspA (from Borrelia brevicornuate), MALP-2 (mycoplasma macrophage-activated lipopeptide-2kD), STF (soluble tuberculosis factor), CU-T12-9, Diprovocim, Amplivant, and lipopeptides derived from cell wall components, such as PAM2CSK4, PAM3CSK4, and PAM3Cys.

TLR2 can form heterodimers with TLR1 or TLR6, and therefore, many TLR2 agonists are also TLR1 or TLR6 agonists.

The terms “TLR3” and “toll-like receptor 3” refer to any form of TLR3 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR3 activity. Unless otherwise specified, such as by way of a specific reference to human TLR3, TLR3 includes naturally occurring sequence TLR3 in all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR3 is provided under UniProt entry number O15455.

As used herein, “TLR3 agonist” means any molecule that, upon binding to TLR3, (1) stimulates or activates TLR3, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR3, or (3) enhances, increases, promotes, or induces the expression of TLR3. TLR3 agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, nucleic acid ligands that bind to TLR3.

Examples of TLR3 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include TLR3 ligands such as synthetic dsRNA, polyinosinic acid [“poly(I:C)”] (available from, for example, InvivoGen in high molecular weight (HMW) and low molecular weight (LMW) formulations), polyadenylic-polyuridylic acid [“poly(A:U)”] (available from, for example, InvivoGen), and polyICLC (see Levy et al., Journal of Infectious Disorders). (See Jasani et al., Vaccine, vol. 132, no. 4, pp. 434-439, 1975), Ampligen (see Jasani et al., Vaccine, vol. 27, no. 25-26, pp. 3401-3404, 2009), Hiltonol, Rintatolimod, and RGC100 (see Naumann et al., Clinical and Developmental Immunology, vol. 2013, article ID 283649).

The terms “TLR4” and “toll-like receptor 4” refer to any form of TLR4 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR4 activity. Unless otherwise specified, such as by way of a specific reference to human TLR4, TLR4 includes naturally occurring sequence TLR4 in all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR4 is provided under UniProt entry number O00206.

As used herein, “TLR4 agonist” means any molecule that, upon binding to TLR4, (1) stimulates or activates TLR4, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR4, or (3) enhances, increases, promotes, or induces the expression of TLR4. TLR4 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, bacterial lipopolysaccharide (LPS) and its derivatives that bind to TLR4.

Examples of TLR4 agonists useful in the treatments, medicines and uses of the present invention include, for example, bacterial lipopolysaccharide (LPS) and its derivatives, such as B:0111 (Sigma), monophospholipid A (MPLA), 3DMPL (3-O-deacylated MPL), GLA-AQ, G100, AS15, ASO2, GSK1572932A (GlaxoSmithKline, UK).

The terms “TLR5” and “toll-like receptor 5” refer to any form of TLR5 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR5 activity. Unless otherwise specified, such as by way of a specific reference to human TLR5, TLR5 includes naturally occurring sequence TLR5s from all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR5 is provided under UniProt entry number O60602.

As used herein, “TLR5 agonist” means any molecule that, upon binding to TLR5, (1) stimulates or activates TLR5, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR5, or (3) enhances, increases, promotes, or induces the expression of TLR5. TLR5 agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, bacterial flagellin and derivatives thereof that bind to TLR5.

Examples of TLR5 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include, for example, bacterial flagellin purified from Bacillus subtilis, flagellin purified from Pseudomonas aeruginosa, flagellin purified from Salmonella typhimurium, and recombinant flagellin (all available from InvivoGen), and entolimod (CBLB502; a pharmacologically optimized flagellin derivative).

The terms “TLR6” and “toll-like receptor 6” refer to any form of TLR6 receptor, as well as variants, isotypes, and species homologs that retain at least a portion of TLR6 activity. Unless otherwise specified, such as by way of a specific reference to human TLR6, TLR6 includes the naturally occurring sequence TLR6 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR6 is provided under UniProt entry number Q9Y2C9.

As used herein, “TLR6 agonist” means any molecule that, upon binding to TLR6, (1) stimulates or activates TLR6, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR6, or (3) enhances, increases, promotes, or induces the expression of TLR6. TLR6 agonists useful in any of the treatments, medicaments, and uses of this invention include, for example, bacterial lipopeptides and derivatives thereof that bind to TLR6.

Examples of TLR6 agonists useful in the treatments, medicines, and uses of the present invention include, for example, many of the TLR2 agonists provided above, because TLR2 and TLR6 can form heterodimers. TLR6 can also form heterodimers with TLR4, and TLR6 agonists can include the various TLR4 agonists provided above.

The terms “TLR7” and “toll-like receptor 7” refer to any form of TLR7 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR7 activity. Unless otherwise specified, such as by way of a specific reference to human TLR7, TLR7 includes the naturally occurring sequence TLR7 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR7 is provided under UniProt entry number Q9NYK1.

As used herein, “TLR7 agonist” means any molecule that, upon binding to TLR7, (1) stimulates or activates TLR7, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR7, or (3) enhances, increases, promotes, or induces the expression of TLR7. TLR7 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acid ligands that bind to TLR7.

Examples of TLR7 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include recombinant single-stranded (“ss”) RNA, imidazoquinoline compounds such as imiquimod (R837), gardiquimod, and resiquimod (R848); rozolidin (7-allyl-7,8-dihydro-8-oxoguanosine) and related compounds; 7-thia-8-oxoguanosine, 7-denitroguanosine, and related guanosine analogs; ANA975 (Anadys Pharmaceuticals) and related compounds; SM-360320 (Sumimoto); 3M-01, 3M-03, 3M-852, and 3M-S-34240 (3M Pharmaceuticals); GSK2245035 (GlaxoSmithKline; 8-oxoadenine molecule), AZD8848 (AstraZeneca; 8-oxoadenine molecule), MEDI9197 (Medimmune; formerly 3M-052), ssRNA40, and adenosine analogs, such as UC-1V150 (Jin et al., Bioorganic Medicinal Chem Lett (2006) 16:4559-4563, compound 4). Many TLR7 agonists are also TLR8 agonists.

The terms “TLR8” and “toll-like receptor 8” refer to any form of TLR8 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR8 activity. Unless otherwise specified, such as by way of a specific reference to human TLR8, TLR8 includes the naturally occurring sequence TLR8 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR8 is provided under UniProt entry number Q9NR97.

As used herein, “TLR8 agonist” means any molecule that, upon binding to TLR8, (1) stimulates or activates TLR8, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR8, or (3) enhances, increases, promotes, or induces the expression of TLR8. TLR8 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acid ligands that bind to TLR8.

Examples of TLR8 agonists useful in the treatments, medicines, and uses of the present invention include recombinant single-stranded ssRNA, imiquimod (R837), gademod, requimod (R848), 3M-01, 3M-03, 3M-852, and 3M-S-34240 (3M Pharmaceuticals); GSK2245035 (GlaxoSmithKline; 8-oxoadenine molecule), AZD8848 (AstraZeneca; 8-oxoadenine molecule), MEDI9197 (Medimmune; formerly 3M-052), Poly-G10, Motolimod, and various TLR7 agonists provided above (as previously noted, many TLR7 agonists are also TLR8 agonists).

The terms “TLR9” and “toll-like receptor 9” refer to any form of TLR9 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR9 activity. Unless otherwise specified, such as by way of a specific reference to human TLR9, TLR9 includes the naturally occurring sequence TLR9 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR9 is provided under UniProt entry number Q9NR96.

As used herein, “TLR9 agonist” means any molecule that, upon binding to TLR9, (1) stimulates or activates TLR9, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR9, or (3) enhances, increases, promotes, or induces the expression of TLR9. TLR9 agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, nucleic acid ligands that bind to TLR9.

Examples of TLR9 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include unmethylated CpG-containing DNA, immunostimulatory oligodeoxynucleotides (ODNs), such as CpG-containing ODNs, such as CpG24555, CpG10103, CpG7909 (PF-3512676/agatolimod), CpG1018, AZD1419, ODN2216, MGN1703, SD-101, 1018ISS, and CMP-001. TLR9 agonists also include a nucleotide sequence containing a synthetic cytosine-phosphate-2'-deoxy-7-dezoguanosine dinucleotide (CpR) (Hybridon, Inc.), dSLIM-30L1, and an immunoglobulin-DNA complex. Exemplary TLR9 agonists are disclosed in WO2003/015711, WO2004/016805, WO2009/022215, PCT/US95/01570, PCT/US97/19791 and U.S. Patent Nos. 8552165, 6194388 and 6239116, which are incorporated herein by reference for all purposes.

RLRs include various cytoplasmic PRRs that detect, for example, dsRNAs. Examples of RLRs include, for example, retinoic acid-inducible gene I (RIG-I), melanoma differentiation-associated gene 5 (MDA-5), and Laboratory of Genetics and Physiology 2 (LGP2).

As used herein, “RLR agonist” means any molecule that, upon binding to an RLR, (1) stimulates or activates the RLR, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of the RLR, or (3) enhances, increases, promotes, or induces the expression of the RLR. RLR agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acids that bind to RLRs and their derivatives, and agonistic monoclonal antibodies (mAbs) that specifically bind to RLRs.

Examples of RLRs agonists useful in the treatments, medicines, and uses of the present invention include, for example, short double-stranded RNAs having an uncapped 5' triphosphate (RIG-I agonists); poly I:C (MDA-5 agonists); and BO-112 (MDA-A agonists).

NLRs include various PRRs that detect molecules such as damage-related molecular patterns (DAMPs). NLRs include subfamilies NLRA-A, NLRB-B, NLRC-C, and NLRP-P. Examples of NLRs include, for example, NOD1, NOD2, NAIP, NLRC4, and NLRP3.

As used herein, “NLR agonist” means any molecule that, upon binding to an NLR, (1) stimulates or activates the NLR, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of the NLR, or (3) enhances, increases, promotes, or induces the expression of the NLR. NLR agonists useful in any of the treatments, medicines, and uses of this invention include, for example, DAMPs and their derivatives that bind to NLRs, and agonistic monoclonal antibodies (mAbs) that specifically bind to NLRs.

Examples of NLR agonists useful in the treatments, medicines, and uses of the present invention include, for example, liposomal muramyl tripeptide/mifamurtide (NOD2 agonist).

CLRs include various PRRs that detect, for example, carbohydrates and glycoproteins. CLRs include both transmembrane CLRs and secretory CLRs. Examples of CLRs include, for example, DEC-205/CD205, macrophage mannose receptor (MMR), Dectin-1, Dectin-2, mincle, DC-SIGN, DNGR-1, and mannose-binding lectin (MBL).

As used herein, “CLR agonist” means any molecule that, upon binding to a CLR, (1) stimulates or activates the CLR, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of the CLR, or (3) enhances, increases, promotes, or induces the expression of the CLR. CLR agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, carbohydrates and their derivatives that bind to CLRs, and agonistic monoclonal antibodies (mAbs) that specifically bind to CLRs.

Examples of CLR agonists useful in the treatments, pharmaceuticals, and uses of the present invention include, for example, MD fraction (a purified soluble β-glucan extract from Grifola frondosa) and imprime PGG (a yeast-derived β1,3/1,6-glucan PAMP).

STING proteins function as both cytoplasmic DNA sensors and adaptors in type 1 interferon signaling. The terms "STING" and "stimulator of interferon genes" refer to any form of STING protein, as well as variants, allotypes, and species homologs that retain at least a portion of STING activity. Unless otherwise specified, such as by way of a specific reference to human STING, STING includes the naturally occurring sequence of STING in all mammalian species, e.g., humans, monkeys, and mice. An exemplary human TLR9 is provided under UniProt entry number Q86WV6. STING is also known as TMEM173.

As used herein, “STING agonist” means any molecule that, upon binding to TLR9, (1) stimulates or activates STING, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of STING, or (3) enhances, increases, promotes, or induces the expression of STING. STING agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acid ligands that bind to STING.

Examples of STING agonists useful in the treatments, medicines, and uses of the present invention include various immunostimulatory nucleic acids, such as synthetic double-stranded DNA, cyclic di-GMP, cyclic GMP-AMP (cGAMP), synthetic cyclic dinucleotides (CDNs), such as MK-1454 and ADU-S100 (MIW815), and small molecules, such as PO-424.

Other PRRs include, for example, DNA-dependent Activator of IFN-regulatory Factors (DAI) and Absentin Melanoma 2 (AIM2).

Immunostimulatory cytokines include various signaling proteins that stimulate immune responses, such as interferons, interleukins, and hematopoietic growth factors.

Exemplary immunostimulatory cytokines include GM-CSF, G-CSF, IFN-α, IFN-γ; IL-2 (e.g., denileukin difitox), IL-6, IL-7, IL-11, IL-12, IL-15, IL-18, IL-21, and TNF-α.

Immunostimulatory cytokines can be in any suitable form. In some embodiments, the immunostimulatory cytokines can be recombinant forms of wild-type cytokines. In some embodiments, the immunostimulatory cytokines can be mutant proteins with one or more amino acid alterations compared to the corresponding wild-type cytokines. In some embodiments, the immunostimulatory cytokines can be incorporated into chimeric proteins containing cytokines and at least one other functional protein (e.g., antibodies). In some embodiments, the immunostimulatory cytokines can be covalently linked to a drug/pharmaceutical (e.g., any drug/pharmaceutical as described elsewhere herein as a possible ADC component).

Cancer vaccines include various compositions containing tumor-associated antigens (or that can be used to generate tumor-associated antigens in subjects) and thus can be used to induce an immune response in subjects against tumor cells containing tumor-associated antigens.

Examples of materials that may be included in a cancer vaccine include attenuated cancerous cells, tumor antigens, and antigen-presenting cells such as dendritic cells bombarded with nucleic acids encoding tumor-derived antigens or tumor-associated antigens. In some embodiments, the cancer vaccine may be prepared using the patient's own cancer cells. In some embodiments, the cancer vaccine may be prepared using biomaterials not derived from the patient's own cancer cells.

Cancer vaccines include, for example, sipuleucel-T and talimogene laherparepvec (T-VEC).

Immunotherapy involves treating patients with immune cells that can target cancer cells. Examples of immunotherapy include, for instance, tumor-infiltrating lymphocytes (TILs) and chimeric antigen receptor T cells (CAR-T cells).

Examples of chemotherapeutic agents include alkylating agents such as thiotepa and cyclophosphamide; alkyl sulfonates such as busulfan, dipropylamine sulfonate, and piperosulfan; aziridines such as benzodopa, carboquinone, meturedopa, and uredopa; ethyleneimines and methylamelamines, including hexamethylmelamine, tratamide, trietylenephosphoramide, ethyl thiophosphoric acid, and trimethylolomelamine; polyacetogenins (especially bullatacin and bullatacinone); camptothecin (including synthetic varieties). Analogs of topotecan; bryozoins; callystatin; CC-1065 (including its synthetic analogs of adolexin, carzelesin, and bizelesin); cryptophycins (especially cryptophycin 1 and cryptophycin 8); dolastatin; bacitracin (including synthetic analogs, KW-2189 and CBI-TMI); eleutherobin; pancratistatin; sarcodictyin; spongistatin; nitrogen mustard, such as chlorambucil, naphthalene, and chlorpyrifos. Cholophosphamide, estradiol, ifosfamide, nitrogen mustard, mechlorethamine, mechlorethamine oxide hydrochloride, melphalan, neonitrogen mustard, benzyl mustard cholesterol, pine oxychloride, tristyramine cyclophosphamide, uracil nitrogen mustard; nitrosoureas, such as carmustine, chloramphenicol, formazan, lomomia, nimustine, ranolamia; antibiotics, such as enediyne antibiotics (e.g., galicarmycin, especially galicarmycin γ1I and galicarmycin phiI1, see, e.g., Agnew, Chem. Intl. Ed. Engl., 33:183-186(19) 94); dynemicin, including dynemicin A; bisphosphonates, such as osteophosphine; esporamin; and new carcinogenic chromophores and related chromomophores, aclarubicin, actinomycin, autramycin, azoserine, bleomycin, cactinomycin C, carabicin, caminomycin, carcinomycin, chromomycin, daunorubicin, doxorubicin, 6-diazo-5-oxo-L-leucine, doxorubicin (including morpholino-doxorubicin, cyanomorpholino-doxorubicin, 2-pyrrolino-doxorubicin) (2-pyrrolino-doxorubicin and deoxydoxomyl), PEGylated liposomal doxorubicin, epirubicin, doxorubicin, idarubicin, mesorcinol, mitomycin such as mitomycin C, mycophenolic acid, nogaramycin, oligomycin, pemycin, potfiromycin, puromycin, triamcinolone acetonide, rodorubicin, streptomycin, streptozotocin, tuberculin, hydroxyphenylbutyryl leucine, neomycin, zorubicin; antimetabolites, such as methotrexate and 5-fluorouracil (5-FU); folic acid analogs, such as folate, methotrexate, pteroyltriglutamate, trimesat; purine analogs, such as fludarabine, 6-mercaptopurine, thioguanine, thioguanine; pyrimidine analogs, such as cyclocytidine, azathioprine, etc. Cytidine, 6-azouridine, carmoflurane, cytarabine, dideoxyuridine, docefluuridine, enoxabin, fluorouridine; androgens, such as carotestosterone, hydroxymethylandrostenone propionate, cyclothiosterol, methylcyclothiosterane, testrolide; anti-adrenergic drugs, such as aminoglutethimide, mitotane, and trilostertan; folic acid supplements, such as frolinic acid; aceglucuronolactone; aldophosphamide glycoside; aminolevulinic acid; eniluracil; acridine; bestrabucil; bethanycin; edatraxate; defofamine; dimethicone; diazinon; elfo-ornithine. rmithine); hydroxypicarazole acetate; epothilone; glycidyl ether; gallium nitrate; hydroxyurea; lentinan; chlordamine; maytansine alkaloids, such as maytansine and metansine; propiconazole; mitoxantrone; monpilamol; diammonium nitrate; pentostatin; benzylmethionine; pirarubicin; losoxantrone; podophyllin; 2-ethylhydrazide; methylbenzylhydrazine; propylimine; rhizomycin; sisofibril; spirogermanium; Alternaria spp. ketoacid; triazolam; 2,2',2''-trichlorotriethylamine; trichosporon toxins (especially T-2 toxin, verracurin A, baculosporin A, and anguidine); ursodeoxycholic acid; vinblastine; dacarbazin; mannitol; dibromomannitol; dibromodehydroxycervastatin Piperbromopropionate; gacytosine; cytarabine (“Ara-C”); cyclophosphamide; thiotepa; taxoids, such as paclitaxel and docetaxel; chlorambucil; gemcitabine; 6-thioguanine; mercaptopurine; methotrexate; platinum analogs, such as cisplatin and carboplatin; vincristine; platinum; etoposide (VP-16); ifosfamide; mitoxantrone; vincristine; vinblastine; nifedipine; teniposide; edaraxacum; daunorubicin; aminopterin; capecitabine; ibenphosphonate sodium; CPT-11; topoisomerase inhibitor RFS 2000; difluoromethylornithine (DMFO); retinoids, such as retinoic acid; capecitabine; and pharmaceutically acceptable salts, acids, or derivatives of any of the above. It also includes anti-hormonal agents used to regulate or inhibit the hormonal effects on tumors, such as anti-estrogens and selective estrogen receptor modulators (SERMs), including, for example, tamoxifen, raloxifene, droloxifene, 4-hydroxytamoxifen, treoxifene, naloxifene, LY117018, onasone, and toremifene (Fareston); aromatase inhibitors that inhibit aromatase, which regulate estrogen production in the adrenal glands, such as 4(5)-imidazole, aminoglutethimide, megestrol acetate, etc. Exemestane, formetastatin, fazodazole, vorazole, letrozole, and anastrozole; and anti-androgen substances such as flutamide, nilumid, bicalutamide, lipoprotein, and serotonin; KRAS inhibitors; MCT4 inhibitors; MAT2a inhibitors; tyrosine kinase inhibitors, such as sunitinib and axitinib; alk/c-Met/ROS inhibitors, such as crizotinib and lorlatinib. Intra-inib; mTOR inhibitors, such as temsirolimus and gedatolisib; src/abl inhibitors, such as bosutinib; cyclin-dependent kinase (CDK) inhibitors, such as palbociclib and PF-06873600; erb inhibitors, such as dacomitinib; PARP inhibitors, such as talazoparib; SMO inhibitors, such as glasdegib and PF-5274857; EGFR T790M inhibitors, such as PF-06747775; EZH2 inhibitors, such as PF-06821497; PRMT5 inhibitors, such as PF-06939999; TGFRβr1 inhibitors, such as PF-06952229; and pharmaceutically acceptable salts, acids, or derivatives of any of the above. In specific implementation schemes, this additional treatment agent is bevacizumab, cetuximab, sirolimus, panitumumab, 5-fluorouracil (5-FU), capecitabine, tevozanib, irinotecan, oxaliplatin, cisplatin, trifluorouridine, tilpyrimidine, leucovorin, gemcitabine, regorafinib, or erlotinib hydrochloride.

In some embodiments, the antibody is used in conjunction with one or more other therapeutic agents that target immune restriction point modulators or co-stimulators, such as, but not limited to, those targeting CTLA-4, LAG-3, B7-H3, B7-H4, B7-DC (PD-L2), B7-H5, B7-H6, B7-H8, B7-H2, B7-1, B7-2, ICOS, ICOS-L, TIGIT, CD2, CD47, CD80, CD86, CD48, C D58, CD226, CD155, CD112, LAIR1, 2B4, BTLA, CD160, TIM1, TIM-3, TIM4, VISTA(PD-H1), OX40, OX40L, GITR, GITRL, CD70, CD27, 4-1BB, 4-BBL, DR3, TL1A, CD40, CD40L, CD30, CD30L, LIGHT, HVEM, SLAM (SLAMF1, CD150 ), SLAMF2 (CD48), SLAMF3 (CD229), SLAMF4 (2B4, CD244), SLAMF5 (CD84), SLAMF6 (NTB-A), SLAMCF7 (CS 1), SLAMF8 (BLAME), SLAMF9 (CD2F), CD28, CEACAM1 (CD66a), CEACAM3, CEACAM4, CEACAM5, CEACAM6, CEAC AM7, CEACAM8, CEACAM1-3AS, CEACAM3C2, CEACAM1-15, PSG1-11, CEACAM1-4C1, CEACAM1-4S, CEACAM1-4L, IDO, TDO, CCR2, CD39–CD73–adenosine pathway (A2AR), BTKs, TIKs, CXCR2, CXCR4, CCR4, CCR8, CCR5, CSF-1, or agents that modulate the innate immune response.

In some implementations, the antibody is used in conjunction with, for example, the following: anti-CTLA-4 antagonist antibodies, such as ipilimumab; anti-LAG-3 antagonist antibodies, such as BMS-986016 and IMP701; anti-TIM-3 antagonist antibodies; anti-B7-H3 antagonist antibodies, such as MGA271; anti-VISTA antagonist antibodies; anti-TIGIT antagonist antibodies; antibodies; anti-CD80 antibodies; anti-CD86 antibodies; anti-B7-H4 antagonist antibodies; anti-ICOS agonist antibodies; anti-CD28 agonist antibodies; modulators of innate immune responses (e.g., TLRs, KIR, NKG2A); and IDO inhibitors.

In some embodiments, the antibody is used with an OX40 agonist, such as, for example, an anti-OX-40 agonist antibody. In some embodiments, the antibody is used with a GITR agonist, such as, for example, an anti-GITR agonist antibody, such as, but not limited to, TRX518. In some embodiments, the antibody is used with an IDO inhibitor. In some embodiments, the antibody is used with cytokine therapy (such as, for example, but not limited to, IL-15, CSF-1, MCSF-1, etc.).

In some implementations, the antibody is used in conjunction with one or more other therapeutic antibodies, such as, but not limited to, antibodies targeting CD19, CD22, CD40, CD52, or CCR4.

In some embodiments, the antibody composition comprises at least one additional pharmaceutical agent, such as bevacizumab, cetuximab, sirolimus, panitumumab, 5-fluorouracil (5-FU), capecitabine, tevozanib, irinotecan, oxaliplatin, cisplatin, trifluorouridine, tilpyridamole, leucovorin, gemcitabine, and erlotinib hydrochloride.

In some embodiments, the antibody therapy may be administered co-administered with other reagents or sequentially before or after other reagents at intervals ranging from minutes to weeks. In embodiments where other reagents and/or proteins or polynucleotides are administered alone, it will generally be ensured that a considerably long period of time does not expire between each delivery, such that the agent and the composition of the invention will still be able to produce a beneficial combined effect on the subject. In this case, it is anticipated that the two dosing regimens may be administered within approximately 12-24 hours of each other, and more preferably within approximately 6-12 hours of each other. However, in some cases, it may be desirable to significantly extend the time period for administration, in this regard, from several days (2, 3, 4, 5, 6, or 7) to several weeks (1, 2, 3, 4, 5, 6, 7, or 8) between each administration.

In some embodiments, the antibodies of the present invention are combined with treatment regimens that further include conventional therapies selected from: surgery, radiotherapy, chemotherapy, targeted therapy, immunotherapy, hormone therapy, angiogenesis inhibition, and palliative care.

The compositions used in this invention may further comprise pharmaceutically acceptable carriers, excipients, or stabilizers in the form of lyophilized formulations or aqueous solutions (Remington: The Science and Practice of Pharmacy, 21st edition, 2005, Lippincott Williams and Wilkins, Ed. K. E. Hoover). Acceptable carriers, excipients, or stabilizers are non-toxic to the recipient at the stated doses and concentrations and may include buffers such as phosphates, citrates, and other organic acids; antioxidants, including ascorbic acid and methionine; preservatives (e.g., octadecyl dimethyl benzyl ammonium chloride; hexane diammonium chloride; benzalkonium chloride, benzyl chloride; phenols, butanol, or benzyl alcohol; alkyl esters of p-hydroxybenzoate, such as methylparaben or propylparaben; catechol; resorcinol; cyclohexanol; 3-pentanol and m-cresol); low molecular weight (less than about 10 residues) polypeptides; proteins, such as blood... Albumin, gelatin, or immunoglobulins; hydrophilic polymers, such as polyvinylpyrrolidone; amino acids, such as glycine, glutamine, asparagine, histidine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates, including glucose, mannose, or dextran; chelating agents, such as EDTA; sugars, such as sucrose, mannitol, trehalose, or sorbitol; salt-forming ions, such as sodium; metal complexes (e.g., Zn-protein complexes); and/or nonionic surfactants, such as TWEEN™, PLURONICS™, or polyethylene glycol (PEG). Pharmaceutically acceptable excipients are further described herein.

In one aspect, the present invention provides a method for generating antibodies as disclosed herein, comprising: culturing host cells under conditions that result in the generation of antibodies as disclosed herein, and purifying the antibody or the bispecific antibody from the culture.

In another aspect, the present invention provides the use of antibodies, pharmaceutical compositions, polynucleotides, vectors or host cells as disclosed herein in the preparation of medicaments for the detection, diagnosis and/or treatment of tumor antigen-related conditions.

Diagnostic uses

In one aspect, for example for diagnostic purposes, the CD3 antibody of the present invention can be used to detect and/or measure CD3 or CD3-expressing cells in a sample. For example, the CD3 antibody can be used to diagnose conditions or diseases characterized by abnormal expression of CD3 (e.g., overexpression, underexpression, lack of expression, etc.). Exemplary diagnostic assays for CD3 may include, for example, contacting a sample obtained from a subject with the CD3 antibody of the present invention, wherein the CD3 antibody is labeled with a detectable marker or reporter molecule. Alternatively, an unlabeled CD3 antibody may be used in combination with a self-detectably labeled secondary antibody for diagnostic applications. The detectable marker or reporter molecule may be a radioisotope, such as ^14C , ^3H , ^125I , ^32P , or ^35S ; a fluorescent or chemiluminescent moiety, such as fluorescein isothiocyanate or rhodamine; or an enzyme, such as alkaline phosphatase, β-galactosidase, horseradish peroxidase, or luciferase. Specific exemplary assays that can be used to detect or measure CD3 in a sample include, but are not limited to, enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), fluorescence activated cell sorting (FACS), etc. Samples that can be used in the CD3 diagnostic assay according to the present invention include any tissue or body fluid sample obtainable from a patient, containing a detectable amount of CD3 protein or fragments thereof under normal or pathological conditions. Typically, the level of CD3 is measured in a specific sample obtained from a healthy patient (e.g., a patient without a disease or condition associated with abnormal CD3 levels or activity) to first establish a baseline or standard for CD3 levels. This baseline CD3 level can then be compared with the CD3 level measured in a sample obtained from an individual suspected of having a CD3-related disease or condition.

In another aspect, methods are provided for detecting, diagnosing, and/or monitoring conditions associated with tumor antigen expression. For example, multispecific antibodies as described herein can be labeled with detectable moieties (e.g., imaging agents and enzyme-substrate markers). Antibodies as described herein can also be used for in vivo diagnostic assays (e.g., in vivo imaging (e.g., PET or SPECT)) or staining reagents.

Reagent test kit

Another aspect of the invention is a kit comprising an antibody as disclosed above and instructions for use according to any method of the invention described herein. Typically, these instructions include a description of the administration of the antibody for the therapeutic treatment described above. The kit includes any pharmaceutical compositions disclosed herein. Pharmaceutical compositions and other reagents may be present in the kit in any convenient form, such as, for example, in solution or powder form.

In another aspect, the kit further includes one or more other preventative or therapeutic agents that can be used to treat cancer, contained in one or more containers. In one embodiment, said other preventative or therapeutic agent is a chemotherapeutic agent. In other aspects, said preventative or therapeutic agent is a biological or hormonal therapeutic agent.

Several aspects of the pharmaceutical compositions, preventive agents, or therapeutic agents of the present invention are preferably tested for desired therapeutic activity in vitro, in cell culture systems, and in animal model organisms (e.g., rodent model systems) before human use.

The toxicity and efficacy of the preventive and/or therapeutic regimens of the present invention can be determined by standard pharmaceutical procedures in cell cultures or laboratory animals, for example, for determining the _LD50 (the dose that is lethal to 50% of the population) and _ED50 (the dose that is therapeutically effective in 50% of the population). The dose ratio between toxicity and therapeutic effect is the therapeutic index, and it can be expressed as the ratio _LD50 / _ED50 . Prophylactic and/or therapeutic agents exhibiting a large therapeutic index are preferred.

Furthermore, any assays known to those skilled in the art can be used to evaluate the preventive and/or therapeutic efficacy of the therapies or combinations of therapies disclosed herein for the treatment or prevention of cancer.

Instructions for use with bispecific antibodies as described herein typically include information about the dosage, dosing regimen, and route of administration for the intended treatment. Containers may be unit doses, bulk packaging (e.g., multi-dose packs), or subunit doses. Instructions supplied with kits of the present invention are typically written instructions on a label or packaging insert (e.g., paper included in the kit), but machine-readable instructions (e.g., instructions carried on a disk or optical storage disc) are also acceptable.

The kit of the present invention is in suitable packaging. Suitable packaging includes, but is not limited to, vials, ampoules, tubes, bottles, wide-mouth bottles, flexible packaging (e.g., sealed polyester film or plastic bags), etc., for each pharmaceutical composition and other included reagents (e.g., buffers, balanced saline solutions, etc., for administering the pharmaceutical composition to a subject). Packaging for use in combination with specific devices such as inhalers, nasal delivery devices (e.g., nebulizers), or infusion devices such as micropumps is also contemplated. The kit may have a sterile inlet (e.g., the container may be an intravenous solution bag or vial with a stopper that can be punctured by a hypodermic needle). The container may also have a sterile inlet (e.g., the container may be an intravenous solution bag or vial with a stopper that can be punctured by a hypodermic needle). At least one active agent in the composition is an antibody. The container may further contain a second pharmaceutical active agent.

Typically, a kit includes a container and a label or packaging insert on or associated with the container.

Biological Preservation

Representative material of the present invention was deposited on February 13, 2018 at the American Center for Type Culture Collection (ATCC), 10801 University Boulevard, Manassas, Va. 20110-2209, USA. The vector GUCY2C-1608 chain A (VH-Synthesis SEQ ID NO: 94) with ATCC accession number PTA-124943 contains a DNA insert of the VH-Synthesis chain encoding the bispecific antibody GUCY2C-1608, which contains a polynucleotide (SEQ ID NO: 70) encoding the CD3 antibody light chain variable region (SEQ ID NO: 3); and the GUCY2C-1608 chain B (VL-Synthesis SEQ ID NO: 93) with ATCC accession number PTA-124944 contains a DNA insert of the VL-Synthesis chain encoding the bispecific antibody GUCY2C-1608, which contains a polynucleotide (SEQ ID NO: 71) encoding the CD3 antibody heavy chain variable region (SEQ ID NO: 4).

The deposit was made in accordance with the provisions of the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the Purpose of Patent Procedure and Regulations thereunder. This ensures that the live culture of the deposit remains intact for 30 years from the date of deposit. The deposit is available through the ATCC under the terms of the Budapest Treaty and complies with an agreement between Pfizer, Inc. and the ATCC that ensures the perpetual and unrestricted availability of the descendants of the deposited culture to the public at the time of publication of the relevant U.S. Patent or at the time of public disclosure of any U.S. or foreign patent application (whichever comes first), and ensures the availability of the descendants to persons authorized by the Director of the U.S. Patent and Trademark Office in accordance with 35 U.S.C. Part 122 and the Director's Rules established thereunder (including 37 C.F.R. Part 1.14, which specifically refers to 886 OG 638).

The assignee of this application has agreed that, should the culture of the deposited material die, be lost, or be destroyed under suitable conditions, the material will be promptly replaced with another identical copy upon notification. The availability of the deposited material shall not be construed as a violation of any government authority's right to practice the invention under its patent law.

The following embodiments for implementing specific aspects of the invention are provided for illustrative purposes only and are not intended to limit the scope of the invention in any way.

Example 1

Formation of soluble CD3ed

A protein called CD3ed, representing the extracellular domains of CD3 ε and δ subunits tandemly fused via a short linker (SEQ ID NO: 39), was generated using standard molecular biology, protein expression, and purification techniques. Using PEI as the transfection agent, plasmids carrying the gene encoding the amino acids shown in Table 9 (SEQ ID NO: 40) were transiently transfected into 20 μL of HEK293 cells. After incubation at 37°C for 7 days, 20 μL of the harvest was fractionally bound to 50 mL of His60 resin (equilibrated with PBS) overnight. The bound resin was then washed with PBS, followed by washing with PBS buffer (10⁻¹⁵ CV) containing 10 mM imidazole. Following washing, the protein was eluted with phosphate-buffered saline containing 200 mM imidazole. The eluent fractions were then analyzed by SDS gel electrophoresis. Fractions containing purified protein were pooled and concentrated using a 3 kDa Amicon ultra centricon. During concentration, 5% glycerol was maintained in the sample to stabilize the protein. The His60 pool was further loaded onto a size-restricted Superdex 200 column. Fractions containing CD3 ε/δ heterodimers were analyzed by SDS-PAGE, and the relevant fractions were pooled and filtered at 0.22 µm. Table 9 shows the sequences of the CD3 ε/δ subunits tandemly fused and used as antigens in all optimization efforts.

Table 9

Example 2

Optimization of the structure of CD3 antibody H2B4

The lead CD3e (CD3 ε) antibody H2B4 exhibited low thermostability when used in bispecific biantibody-Fc form. To improve stability, the light chain architecture was changed from VK4 to VK1 (DPK9), while the heavy chain architecture was retained as VH3 (DP54). VLCDR1 (SEQ ID NO: 26), VL CDR2 (SEQ ID NO: 27), and VL CDR3 (SEQ ID NO: 28) were transplanted into the DPK9 architecture and expressed in combination with heavy chains carrying CDRs (i.e., VH CDR1 (SEQ ID NO: 13), VHCDR2 (SEQ ID NO: 16), and VH CDR3 (SEQ ID NO: 18)). In addition to transplanting CDRs into the architecture, certain reversion mutations were tested. The reversion mutations introduced in both the heavy and light chain structures are shown in Table 10, where the residue number indicating the reversion mutation is indicated.

A cDNA containing the human receptor framework (VH3 for the heavy chain and VK1 for the light chain) and the associated CDR donor sequence was synthesized. The synthesized cDNA product was subcloned and fused to the human IgG1 constant region of the heavy chain and the human κ-frame of the light chain in a mammalian expression vector. To distinguish them from the H2B4 antibody, these variants will be referred to as 2B5 below, and all optimized molecules will be referred to as 2B5 and its variants. The competitive binding of all 2B5 mutants to Jurkat cells with endogenous CD3e expression was analyzed.

Competitive FACS was performed to examine whether humanized 2B5 maintained its binding epitope on the cell surface. Prior to the competitive FACS assay, chimeric cH2B4 was biotinylated. The EC50 of the binding of biotinylated cH2B4 to Jurkat cells was determined by direct binding FACS and was measured to be 0.8 nM. For competitive FACS, a 3-fold serially diluted antibody variant was mixed with a constant amount of biotin-cH2B4 representing the EC50 concentration and 1.0+E05 cells/well of Jurkat cells. The mixture was incubated on ice for 1 hour, followed by three washes with FACS buffer. A secondary antibody (i.e., PE (phycoerythrin)-conjugated streptavidin (PE-SA)) was added to the cells and incubated on ice for 30 minutes. Median fluorescence intensity (MFI) was then analyzed by FACS on FACS CANTO™ (BDBiosciences). Y-axis calculation: %Maximum binding = [MFI (serially diluted Ab wells) - MFI (secondary PE-SA)] / [MFI maximum (Jurkat cells with biotin-cH2B4 only) - MFI (secondary PE-SA)]. Table 10 shows the results of the _IC50 values for competitive binding when mutations were introduced into CD3 antibody H2B4 to optimize the scaffold. Based on these results, H2B4 1.0/1.0T showed similar binding characteristics to H2B4 and was designated as the CD3 antibody binding domain 2B5.

Table 10

Antibody	Heavy chain	Light chain	<![CDATA[Competitive combination of IC<sub>50</sub>]]>
				h2B4 1.0/1.0T (2B5)	CDR grafts	CDR graft, R24T	1.27
h2B4 1.0/1.1	vH1.0	Q13V, R24T, Y49S, L78V	1.68
				h2B4 1.0/1.2	vH1.0	Q13V, R24T, Y49S	Untested
h2B4 1.0/1.3	vH1.0	R24T, Y49S	1.309
				c2B4 (Chimeric 2B4)	mouse vH	mouse vK	0.752

Example 3

Bispecific biantibody-Fc molecules were generated and characterized using H2B4 and 2B5 sequences.

Bispecific biantibody-Fc molecules were generated using standard expression and purification techniques known in the art, using CD3 antibody 2B5 or H2B4 (Tables 2 and 3) and a GUCY2c antibody sequence in one of the configurations shown in Figure 1 (schematic diagram). The stability of the resulting bispecific antibodies (Table 7; SEQ.ID.NO.36-38) was tested using differential scanning calorimetry.

Example 4

The stability of bispecific antibodies was evaluated using differential scanning calorimetry.

The protein was diluted to 0.6 mg/mL in 400 µL of phosphate-buffered saline (PBS). PBS was used as a buffer blank in the reference chamber. PBS contained 137 mM NaCl, 2.7 mM KCl, 8.1 mM Na₂HPO₄, and 1.47 mM KH₂PO₄ at pH 7.2. Samples were dispensed into the sample trays of a MicroCal VP-Capillary DSC using an autosampler (Malvern Instruments Ltd, Malvern, UK). Samples were equilibrated at 10 °C for 5 min and then scanned at 100 °C per hour up to 110 °C. A 16-second filtration period was selected. The raw data were baseline corrected and the protein concentration was normalized. The data were fitted to the MN2-State model using Origin software 7.0 (OriginLab Corporation, Northampton, MA) with an appropriate number of transformations. As illustrated in Figure 2, the bispecific GUCY2c-2B5 exhibits an improved first melt transition at 5°C compared to the bispecific GUCY2c-H2B4. This indicates that bispecific antibodies containing 2B5 demonstrate better stability.

Using FACS, the binding of two bispecific antibodies to T cells was examined using standard procedures known in the art, and as shown in Figure 3, both bispecific antibodies showed similar binding to T cells, indicating that the binding was retained after H2B4 was converted to 2B5.

Both H2B4 and 2B5 bispecific antibodies were tested in a cytotoxicity assay to evaluate T cell retargeting activity. As shown in Figure 4, both bispecific antibodies showed similar cytotoxicity, indicating that 2B5 retained the same activity as H2B4.

Example 5

2B5-based structure-based rational mutation

The CD3 antibody H2B4 binding domain exhibits multireactivity (e.g., as evaluated by binding to DNA and insulin), shows a tendency for self-association (e.g., indicated by a high AC-SINS score), and exhibits a tendency for cleavage at the R53 position in CDR-H2 when expressed in CHO host cells. To reduce multireactivity, the tendency for self-association and cleavage in CDRH2, and mutations in the CDR of 2B5, were engineered to reduce these adverse factors while maintaining binding affinity and stability. For this effort, all predictions were made using the CD3 antibody H2B4 X-ray crystal structure, obtained in a complex with a peptide representing the binding epitope and derived from the N-terminus of the CD3 ε subunit. It is well known in the art that excess positive charge and/or hydrophobicity can play a major role in multireactivity. The electrostatic surface of the H2B4 Fv region from the crystal structure was determined using the DelPhi Poisson Boltzmann calculator in Discovery Studio 4.0 ^1. In Figure 5, an excess positive charge patch that is not disrupted by negative charge is observed. This indicates a correlation with high multi-reactivity. Additionally, the Spatial Aggregation Tendency (SAP) tool in Discovery Studio ¹ was used to identify excess hydrophobic patches (Figure 6). As shown in Figure 6, the hydrophobic patches appear small and insignificant. Based on this observation, the focus is on reducing the positive charge by removing positive charges or adding negative charges to disrupt the patches.

To determine the appropriate sites for mutagenesis, computational predictions of mutations for both affinity and stability changes were performed using Discovery Studio and FoldX. Tolerance mutations were those for which neither Discovery Studio nor FoldX methods predicted ΔΔG > 1 kcal/mol. From these predictions, a set of residues predicted to have minimal impact on binding and stability while still disrupting the altered patch were identified. Additionally, we favored hydrophilic mutations that did not increase hydrophobic surface area where possible. A list of possible CDR mutations introduced in CD3 antibody 2B5 to optimize biophysical properties is shown in Table 11. This includes mutations predicted to be tolerant at each site and reduce multireactivity and splicing.

Table 11

In addition, a set of multiple mutations (combinations) were designed for both the heavy and light chains. These include R52Q/R52bQ/R53Q, R52Q/R52bQ/R53Q/R96Q, R52bQ/R53Q/R96Q, and R52Q/R52bQ/R96Q for the heavy chain, and R29Q/K30Q/R54L/V27eE, R29Q_R54L_V27eE, and R29Q_K30Q_R54L for the light chain.

Example 6

Characterization of mutants

All mutants were generated as IgG molecules using standard expression and purification techniques well-known in the art, and their binding to CD3ed, multireactivity (binding to DNA and insulin), and AC-SINS were tested.

Combined assays to evaluate variants

Octet assays were designed and run to examine both the binding and dissociation rates of 10 μg/ml soluble human CD3ed antigen (SEQ ID NO: 40). A 10 μg/ml anti-CD3 antibody was captured using an anti-human IgG tip for 120 s, followed by a baseline in PBS for 30 s, then immersed in 10 μg/ml soluble human CD3ed antigen for 120 s, and finally dissociated in it for 300 s. Sensing patterns were examined to determine whether the antigen bound to the antibody, and Octet software was used to calculate kd(1/s) and kd error. Twenty-six viable mutations retained binding to the hCD3ed antigen. The binding kinetics of the 26 remaining antigen-binding CD3 antibody variants to soluble CD3 were then examined using Biacore analysis via SPR. CD3 antibodies were captured on anti-human Fc cells, and soluble CD3 antigen at concentrations of 0 nM, 3.7 nM, 11.1 nM, 33.3 nM, and 100 nM was injected onto the captured antibodies to determine binding kinetics. Twenty-three CD3 antibodies bound the soluble CD3 antigen with an affinity similar to that of the parental antibodies used in CDR transplantation.

DNA and insulin ELISA to measure multireactivity

384-well ELISA plates (Nunc Maxisorp) were coated overnight at 4°C with DNA (10 µg/ml) (Sigma-Aldrich, D1626) and insulin (5 µg/ml) (Sigma-Aldrich, I9278-5 mL) in PBS pH 7.5. The assay, modified from that described in Tiller et al. (17), was performed on a PerkinElmer Janus Automated Workstation liquid handling robot. Wells were washed with water, blocked for 1 hour at room temperature with 50 µl of multireactive ELISA buffer (PEB; PBS containing 0.05% Tween-20, 1 mM EDTA), and rinsed three times with water. Quadruple copies of serially diluted mAbs in 25 µl were added to the wells and incubated at room temperature for 1 hour. The plates were washed three times with water, and 25 µl of 10 ng/ml goat anti-human IgG (Face fragment specific) conjugated with horseradish peroxidase (Jackson ImmunoResearch, 109-035-008) was added to each well. The plates were incubated at room temperature for 1 hour, washed three times with 80 µl of water, and 25 µl of TMB substrate (Sigma Aldrich, T-0440) was added to each well. After approximately 7 minutes, the reaction was terminated by adding 25 µl of 0.18 M orthophosphate to each well, and the absorbance was read at 450 nm. The DNA and insulin binding score was calculated as the ratio of the ELISA signal of the 10 µg/ml antibody to the signal of the well containing buffer instead of the primary antibody. A score less than 8 was considered ideal and represented low multireactivity.

AC-SINS assay to measure self-association tendency

AC-SINS (Affinity Capture Self-interaction Nanoparticle Spectroscopy) assays were normalized in 384-well format on a Perkin-Elmer Janus liquid handling robot. 20 nm gold nanoparticles (Ted Pella, Inc., #15705) were coated with a mixture of 80% goat anti-human Fc (Jackson Immuno Research Laboratories, Inc. #109-005-098) and 20% nonspecific goat polyclonal antibody (Jackson Immuno Research Laboratories, Inc. #005-000-003), which was replaced with buffer at 20 mM sodium acetate pH 4.3 and diluted to 0.4 mg/mL. After incubation at room temperature for one hour, unoccupied sites on the gold nanoparticles were blocked with thiolized polyethylene glycol (2 kDa). The coated nanoparticles were then concentrated 10-fold using a syringe filter, and 10 µl was added to 100 µl of 0.05 mg/ml mAb in PBS pH 7.2. The coated nanoparticles and target antibody were incubated in 96-well polypropylene plates for 2 hours, and then transferred to 384-well polystyrene plates and read on a Tecan M1000 spectrophotometer. Absorbance was read from 450–650 nm in 2 nm increments, and a Microsoft Excel macro was used to identify the maximum absorbance, smooth the data, and fit the data using a second-order polynomial. The antibody AC-SINS score was determined by subtracting the smoothed maximum absorbance of the mean blank (PBS buffer only) from the smoothed maximum absorbance of the antibody sample. A score less than 8 was considered ideal and represented low self-association.

Evaluation of the lead CD3 antibody 2B5 variant in a bispecific Fc form

A comprehensive analysis of data from binding assays, AC-SINS, and multireactivity helped narrow down the binding variants to five. These five variants were reformatted into bispecific biantibody-Fc molecules using antitumor sequences and standard cloning strategies well-known in the art. These bispecific antibodies were transiently expressed and purified using standard techniques well-known in the art. Binding of the resulting bispecific antibodies to jurakat cells was tested, in vitro cytotoxicity was tested using a T-cell retargeting assay, multireactivity was tested, and thermostability was tested using DSC and AC-SINS. Of all five bispecific antibodies, only one, carrying the CD3 antibody 2B5 variant 2B5v6, showed potent binding and therefore potent cytotoxicity. This variant also showed reduced multireactivity, as evaluated by DNA and insulin binding and AC-SINS scores. The 2B5v6 bispecific antibody scored less than 8, compared to a score greater than 8 for the H2B4 bispecific antibody. Based on these results, 2B5v6 was selected as the lead optimized 2B5 variant for further analysis.

Example 7

Evaluation of shear in CDR-H2

When the bispecific antibodies were expressed in CHO cells but not in HEK293 cells, cleavage was observed in the VH CDR2 (SEQ ID NO: 16) at position R53 in the CD3 antibodies (H2B4 and 2B5). To mitigate the risk of this adverse effect and ensure better homogeneity of the product at manufacturing scale, care should be taken to select for mutations in the cleavage region, as described in previous examples, as well as other mutations to mitigate the risk of other adverse effects. To evaluate the cleavage, simply, bispecific antibody-Fc bispecific antibodies were generated using the variable domain sequence of the negative control antibody and CD3 antibody 2B5 or CD3 antibody H2B4 or CD3 antibody 2B5v6. All bispecific antibody-Fc molecules were prepared using a similar schematic diagram of expression and purification processes well known in the art. CHOSSI cell lines were generated using standard procedures. Two strands encoding a bispecific antibody were cloned into the mammalian expression vector pRY19-GA-Q, which contains a dual promoter and a recombinant site for unit site integration (SSI) (Zhang, L. et al. Biotechnology Progress. 2015; 31:1645-1656) into the CHO cell genome. The resulting plasmid, containing the target gene, was transfected into 10e9 CHO-K1 SV cells via electroporation using the pRY19 vector, followed by positive and negative selection pressures using hygromycin-B and ganciclovir. These CHO conjugates underwent a 3-week recovery period. When observed viability exceeded 90%, a cell library was generated for future work. The established CHO conjugates were then subjected to a 12-day fed-batch platform expression process in 1L of CD-CHO medium (Thermo Fisher Scientific, Waltham, MA). The day 12 culture harvest was captured on a Protein A column and eluted at low pH. The proA conjugate was immediately neutralized to pH 8.1. The purified protein was evaluated using capillary gel electrophoresis (cGE) according to standard protocols well-known in the art. As shown in Table 12 and Figure 7, the 2B5 bispecific antibody showed significantly reduced shearing compared to H2B4, and 2B5v6 showed no shearing while H2B4 showed significantly increased shearing or fragmentation. Table 12 shows the results of capillary gel electrophoresis demonstrating fragmentation of the CD3-GUCY2c bispecific antibody. POI% indicates the target peak.

Table 12

Reduced cGE sample	%POI	%Fragmentation
			Negative control - H2B4	88.8	11.2
Negative control - 2B5	98.9	1.1
			Negative control - 2B5v6	99.6	0.4

Example 8

GUCY2c-2B5v6 Bispecific Antibody: Expression and Purification of Chimeric CD3-GUCY2c Bispecific Antibody Using CD3 Antibody 2B5v6

Complementary construct pairs (12.5 µg per strand) were co-transfected into 25 mL of logarithmic-phase culture containing 1 million cells/mL HEK293 cells using the ExpiFectamine™ 293 Transfection Kit (Life Technologies). Twenty-four hours post-transfection, ExpiFectamine Transfection Enhancer was added, and cells were allowed to grow for an additional 4–5 days before harvest. The used culture was then collected, centrifuged to remove cell debris, and passed through a 20 µm filter.

The clarified conditioned medium containing the GUCY2c T cell bispecific antibody was then purified using protein A affinity chromatography. Samples were loaded onto 0.45 mL microcolumns pre-loaded with MabSelect SuRe protein A resin (GE Healthcare) using a liquid delivery device (Tecan). The bound protein was washed with PBS pH 7.2, eluted with 20 mM citric acid and 150 mM sodium chloride pH 3.5, and neutralized with 2 M tris at pH 8.0. The sample was then desalted to PBS pH 7.2 using G25 Sephadex drip columns (GE Healthcare) according to the manufacturer's instructions. The purity of the purified protein was analyzed on an Aglient 1200 HPLC system using analytical size exclusion chromatography (TOSOH) with a Mab HTP column, following the manufacturer's protocol. Concentrations were determined using a microspectrophotometer (Trinean) at OD 280 nm.

The stability of bispecific antibodies was evaluated using differential scanning calorimetry.

The protein was diluted to 0.6 mg/mL in phosphate-buffered saline (PBS) solution at a volume of 400 µL. PBS was used as a buffer blank in the reference chamber. PBS contained 137 mM NaCl, 2.7 mM KCl, 8.1 mM Na₂HPO₄, and 1.47 mM KH₂PO₄ at pH 7.2. Samples were dispensed into the sample trays of a MicroCal VP-Capillary DSC using an autosampler (Malvern Instruments Ltd, Malvern, UK). Samples were equilibrated at 10 °C for 5 min and then scanned at 100 °C per hour up to 110 °C. A 16-second filtration period was selected. The raw data were baseline corrected and the protein concentration was normalized. The data were fitted to the MN2-State model using Origin software 7.0 (OriginLab Corporation, Northampton, MA) with an appropriate number of transformations. As shown in Figure 8, the GUCY2c-2B5v6 bispecific antibody exhibits excellent thermal stability, with _Tm1 being 70℃.

Binding with naïve T cells

Using a BD LSRII Fortessa analyzer, purified CD3 bispecific antibody was titrated to bind to the cell surface of naïve human CD3 on virgin human T cells purified from fresh human peripheral blood mononuclear cells (PBMCs) using standard flow cytometry methods known in the art. As shown in Figure 9, the GUCY2c-2B5v6 bispecific antibody bound very well to T cells, with a binding EC50 of 35.44 nM.

Evaluation of T cell activity mediated by bispecificity

Human PBMCs were isolated from healthy donor blood using Histopaque-177 (Sigma). Naïve T cells (negatively selected T cells) were isolated from PBMCs using a T cell enrichment kit from Stem Cell Technologies. T84 human tumor cells expressing GUCY2c, transfected with a luciferase expression construct, were treated with serially diluted GUCY2c bispecific antibody or a negative control CD3 bispecific antibody, along with the T cells. The effector cell to target cell ratio (E:T) was 10:1 or 5:1. T cells and tumor cells were incubated at 37°C for 48 hours, followed by measurement of luciferase signal using neolite reagent (Perkin Elmer). EC50 values were calculated in Graphpad PRISM using a four-parameter nonlinear regression analysis. Target-negative HCT116 or HT29 cells did not show any GUCY2c bispecificity-mediated cytotoxicity. As shown in Figure 10, GUCY2c-2B5v6 bispecificity demonstrated potent cytotoxicity in the T cell retargeting assay.

T84 cells were treated with T cells (E:T ratio 5:1) and different doses of GUCY2c-1608 (GUCY2c-2B5v6). Supernatants were collected at different time points (24 h, 48 h, 96 h) and analyzed using multiplex Luminex assays according to the manufacturer's guidelines and read on a Luminex 200 with xPONENT software. The measured cytokines were human IFN-γ, IL10, IL2, IL4, IL6, and TNF-α. Figure 11 shows the upregulation of (11A) IFN-γ, (11B) IL10, (11C) IL2, (11D) IL4, (11E) IL6, and (11F) TNF-α cytokines after T cell activation with GUCY2c-1608.

In vivo assessment of bispecific activity mediated by GUCY2c-2B5v6

In vivo efficacy studies were conducted using an adoptive metastasis model. Tumor cell lines/patient-derived xenograft fragments were implanted into NSG mice and staged to approximately 200 ^mm³ . The bispecific compound was administered intravenously to the mice (unless otherwise noted). Twenty-four hours after bispecific antibody administration, two million activated and expanded T cells were administered intravenously (using a T cell expansion and activation kit from Miltenyi Biotec). The bispecific antibody was administered weekly. As shown in Figures 12 and 13, the GUCY2c-2B5v6 (GUCY2c-1608) bispecific antibody demonstrated excellent durable tumor regression in both the PDX CRX-11201 and LS1034 cell lines, representing colorectal cancer, in vivo models.

Example 9

Generation and cytotoxicity of FLT3-CD3 bispecific antibody and FLT3-2B5v6 bispecific IgG using an optimized CD3 antibody (2B5v6).

Human FLT3 antibody (xFLT3) and human CD3 antibody (2B5v6) are expressed as engineered human IgG2dA_D265A with EEEE on one arm and RRRR on the other arm, respectively, for bispecific exchange at positions 223, 225, and 228 (e.g., (C223E or C223R), (E225E or E225R), and (P228E or P228R)) in the hinge region of human IgG2 and at positions 409 or 368 (e.g., K409R or L368E (EU numbering scheme)) in the CH3 region. The antibodies also have a mutation from D to A at position 265 (EU numbering scheme).

Individual antibody arms were purified separately using Protein A resin. Then, 1 mg/mL of purified xFLT3 antibody was exchanged with 1 mg/mL of purified 2B5v6 antibody in the presence of 2 mM reduced glutathione. The exchange reaction was carried out at 37°C for 16 h. Then, 2 mM oxidized glutathione was added to the protein, and the mixture was incubated at 37°C for 30 min. The mixture was then diluted 1:5 to 20 mM MES pH 5.4 buffer without NaCl. The diluted sample was loaded onto a MonoS ion exchange column and washed with 20 mM MES pH 5.4 + 25 mM NaCl buffer. The bispecific antibody xFLT3-2B5v6 was then eluted with a 0–500 mM NaCl gradient in 20 mM MES pH 5.4 buffer. Finally, the bispecific antibody was dialyzed into phosphate-buffered saline (PBS) for cell-based assays. For the cytotoxicity assay, unstimulated human T cells and luciferase-expressing target cells (Eol-1 or MV411) were co-incubated in 96-well U-shaped plates in 200 μl RPMI + 10% FBS to determine the ratio. xFLT3-2B5v6 bispecific antibody was serially diluted in PBS and added to each well. After 24 hours, 100 μl of cell suspension was mixed with 100 μl of One-GLO reagent in 96-well white-walled plates, and luminescence was measured on an Envision photometer. The cytotoxicity value was determined using the following formula:

%Cytotoxicity = [1 - (RLU sample) / RLU target cells only)] × 100RLU: relative light units

As shown in Figures 14A and 14B, when tested on two cell lines expressing higher copies of FLT3 (EOL-1) and lower copies (MV4-11), the xFLT3-2B5v6 bispecific antibody containing the optimized CD3 antibody (2B5v6) appears to be highly potent when presented as an alternative to the bispecific IgG form.

Table 13 shows the light and heavy chain sequences of FLT3 (EOL-1) and the light and heavy chain sequences of the 2B5v6 bispecific antibody.

Table 13

Example 10

T-cell epitopes of the 2B5v6 bispecific antibody

Optimized analysis of potential T-cell epitopes in clone 2B5v6 was performed to de-risk the possibility of immunogenicity. This analysis was performed using the Epivax tool for predicting potential T-cell epitopes on a computer chip. This tool predicts the number of times a 9-mer framework peptide binds to any of eight different HLA alleles. Hit was defined as those in the top 5% with a Z-score in the top 1%. Hit to four or more alleles or one strong hit was considered a predicted T-cell epitope. Based on these criteria, three predicted non-germal T-cell epitopes in CDRH2, two predicted germal epitopes in CDRL1, and one predicted germal epitope in L2 were identified (Table 15). In addition to the scores of individual peptides, the overall risk of the sequence was defined by the total number of hits or the T-regitope-adjusted Epivax score, which had been optimized to correlate with the prevalence of clinical ADA. Here, for these predicted values, the lower the score, the lower the expected risk of ADA. The VH domain had 70 hits and a score of -42.82, while the VL domain had 60 hits and a score of -23.79.

To reduce the risk of immunogenicity, mutations that would decrease the overall score of this clone while removing hits and ideally predicted T-cell epitopes were identified. Any possible single mutant variations of CDR-L1, CDR-L2, and CDR-H2 were considered to identify mutations that would reduce hits, decrease the overall t-regitope-adjusted EpiMatrix score, and remove T-cell epitopes. Since these potential mutations could also affect binding and stability, care was taken to consider the predicted stability and affinity data described above, as well as to identify tolerant mutations. Tables 14A, 14B, and 14C show mutations designed to reduce the immunogenicity risk of 2B5v6. Additionally, mutations predicted as shown in SEQ ID NOs 44-47, 51-57, 60, and 62 reduced the number of T-cell epitopes.

Table 14A

Table 14B

Table 14C

Table 15 shows the regions of the variable domains of a CD3 antibody (2B5v6) that has T-cell epitopes and poses a potential risk of immunogenicity.

Table 15

All mutants were generated as monovalent IgG antibodies using standard transient HEK293 expression, followed by purification using a protein A column. The binding of the resulting clones to CD3-expressing Jurkat cells was tested using FACS, and the likelihood of self-association was tested using AC-SINS as described in Example 7 above. As shown in Table 16, five molecules with improved epivax scores and reduced T-cell epitopes, while retaining affinity for CD3, were selected for further testing as biantibody-Fc variants. The 2B5v6 variant with improved epivax scores was compared to the parental 2B5v6. Results showed comparable binding to CD3-expressing Jurkat cells and AC-SINS scores.

Table 16

Of the mutants tested, several exhibited lower binding to Jurkat cells compared to 2B5v6. Among these mutants, those with all T-cell epitopes eliminated and improved epivax scores were further advanced to generate bispecific antibody-Fc molecules in which the GUCY2c antibody serves as the other binding domain. Table 17 shows the epivax scores, and Figure 15 shows the reduced binding of these mutants to Jurkat cells. The low-affinity CD3 antibody 2B5v6 variant showed an improved epivax score.

Table 17

CD3 antibody variants	Epivax Score (tReg - Adjusted)
		2B5v6-0598	-65.15
2B5v6-0623	-56.74

Example 11

Three anti-CD3 variants showed improved affinity and cytotoxicity in the GUCY2C-CD3 bispecific antibody.

Using the standard cloning, expression, and purification techniques described above, three anti-CD3 variants derived from 2B5v6, 2B5-1038 (SEQ ID NO: 4 and 9), 2B5-1039 (SEQ ID NO: 4 and 87), and 2B5-1040 (SEQ ID NO: 4 and 89), were reformulated into bispecific biantibody-Fc molecules paired with the antitumor GUCY2c antibody sequence, one of the configurations shown in Figure 1. In vitro cytotoxicity of the resulting bispecific antibodies was assessed using T-cell retargeting assays as shown in Table 18, binding affinity was tested using surface plasmon resonance (SPR), and nonspecificity was tested using AC-SINs. Immunogenic risk was evaluated using the Epivax Tool, a computer-based assay for potential T-cell epitopes. Complete mass spectrometry data indicated that all bispecific antibodies were correctly paired and 100% heterodimer. Table 18 shows that all three bispecific antibodies (1) were approximately 2-fold more potent in in vitro cytotoxicity assays compared to the control bispecific GUCY2C-1608 paired with anti-CD3 2B5v6 (Fig. 6); (2) had binding affinity for recombinant human CD3 via SPR, consistent with cytotoxic activity; (3) had improved immunogenicity scores compared to the control; and (4) AC-SINs scores remained the same as the control bispecific GUCY2C-1608. These results indicate that anti-GUCY2c bispecific antibodies with different CD3 variants recruit naïve human T cells to induce cell killing in T84 tumor cells (Fig. 16). The three anti-CD3 variants derived from 2B5v6 showed approximately 2-fold greater potency compared to the control bispecific GUCY2C-1608.

Table 18

Bispecific

CD3 variant

Epivax Score (tReg - Adjusted)

MassSpec (Complete)

AC-SINs score

In vitro cytotoxicity (nM)

KD (nM, SPR)

GUCY2C-1608

2B5-0542

-33.7

100% heterodimer

3

0.772

127.88 ± 1.63

GUCY2C-1678

2B5-1038

-52.61

100% heterodimer

2

0.247

63.55±0.49

GUCY2C-1679

2B5-1040

-47.58

100% heterodimer

3

0.194

72.02 ± 0.36

GUCY2C-1680

2B5-1039

-53.51

100% heterodimer

3

0.385

78.57±0.9

The sequences of the bispecific antibodies are shown in Table 19.

Table 19

CD3-specific antibodies and their uses

sequence list

This application was filed electronically via EFS-Web and includes a sequence list submitted electronically in .txt format. The .txt file contains a sequence list titled "PC72375-PRV2_SequenceListing_ST25_05132019.txt", created on May 13, 2019, and is 100 KB in size. The sequence list contained in this .txt file is part of this specification and is incorporated herein by reference in its entirety.

Invention Field

This invention relates to CD3-specific antibodies and their antigen-binding fragments, as well as methods of using them. The invention also relates to multispecific antibodies that bind to both CD3 and target antigens, and methods of using them.

Background of the Invention

Differentiation cluster 3 (CD3) is a homodimeric or heterodimeric antigen expressed on T cells that associates with the T cell receptor complex (TCR) and is required for T cell activation. T cell activation is a complex phenomenon that depends on the involvement of multiple cell surface molecules expressed on a population of reactive T cells. Functional CD3 is formed by the association of two of four different chains: ε, ζ, δ, and γ. For example, the antigen-specific T cell receptor (TCR) is composed of a disulfide-linked heterodimer containing two clonal, monolithic membrane glycoprotein chains, alpha and beta (α and β), or gamma and delta (γ and δ), non-covalently associated with a complex of a low molecular weight invariant protein (usually designated CD3). CD3 dimer arrangements also include γ/ε, δ/ε, and ζ/ζ.

Antibodies targeting CD3 have been shown to accumulate CD3 on T cells, thereby inducing T cell activation in a manner similar to the binding of MHC molecules loaded with peptides to the TCR. Therefore, CD3 antibodies can be used for therapeutic purposes involving T cell activation.

Invention Overview

This invention provides antibodies that bind to CD3 and compositions (such as pharmaceutical compositions) containing one or more CD3 antibodies. Antibodies according to this aspect of the invention are particularly useful for targeting CD3-expressing T cells and for stimulating T cell activation, for example, in cases where T cell-mediated killing is beneficial or desired. The CD3 antibodies of this invention may be included as part of a multispecific antibody (e.g., a bispecific antibody) that directs CD3-mediated T cell activation to specific cell types, such as tumor cells or infectious pathogens.

In one aspect, the present invention provides an antibody that specifically binds to CD3, wherein the antibody comprises: (a) a heavy chain variable (VH) region comprising VH complementarity-determining region 1 (VH CDR1), VH complementarity-determining region 2 (VH CDR2), and VH complementarity-determining region 3 (VH CDR3) of the VH sequence shown in SEQ ID NO: 2, 4, 5, 7, 10, 12, or 35; and/or (b) a light chain variable (VL) region comprising VL complementarity-determining region 1 (VL CDR1), VL complementarity-determining region 2 (VL CDR2), and VL complementarity-determining region 3 (VL CDR3) of the VL sequence shown in SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87, or 89.

In one embodiment of this aspect, the CD3 antibody comprises: (a) a VH region comprising (i) a VH CDR1 comprising the sequence of SEQ ID NO: 13, 14 or 15; (ii) a VH CDR2 comprising the sequence of SEQ ID NO: 16, 17, 19, 20, 21, 22, 23 or 24; and (iii) a VH CDR3 comprising the sequence of SEQ ID NO: 18 or 25; and/or (b) a VL region comprising (i) a VL CDR1 comprising the sequence of SEQ ID NO: 26, 29, 30, 32, 91 or 92; (ii) a VL CDR2 comprising the sequence of SEQ ID NO: 27 or 31; and (iii) a VL CDR3 comprising the sequence of SEQ ID NO: 28 or 33.

In some such embodiments, the present invention provides an antibody wherein: (a) the VH region contains a sequence of SEQ ID NO: 2, 4, 5, 7, 10, 12 or 35; and/or (b) the VL region contains a sequence of SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87 or 89.

In other embodiments, the antibody comprises a VL/VH amino acid sequence selected from the following: SEQ ID NO: 1 and 2; SEQ ID NO: 3 and 2; SEQ ID NO: 3 and 4; SEQ ID NO: 3 and 5; SEQ ID NO: 6 and 7; SEQ ID NO: 8 and 7; SEQ ID NO: 6 and 4; SEQ ID NO: 8 and 4; SEQ ID NO: 9 and 10; SEQ ID NO: 9 and 7; SEQ ID NO: 11 and 12; SEQ ID NO: 34 and 35; SEQ ID NO: 9 and 4; SEQ ID NO: 87 and 4; and SEQ ID NO: 89 and 4.

In some of the aforementioned embodiments, the antibody is selected from: Fab, Fab fragment, F(ab) ₂ fragment, Fv fragment, single-chain Fv fragment, disulfide-stabilized Fv fragment, single-chain antibody, monoclonal antibody, chimeric antibody, bispecific antibody, trispecific antibody, multispecific antibody, bispecific heterodimeric biantibody, bispecific heterodimeric IgG, polyclonal antibody, labeled antibody, humanized antibody, human antibody and fragment thereof.

In one specific embodiment, the antibody is a bispecific antibody. In some such embodiments, the bispecific antibody specifically binds to tumor antigens.

In some embodiments, the antibody of the present invention further comprises a human or humanized VH framework and a human or humanized VL framework. In some such embodiments, the antibody is a humanized antibody.

The present invention further provides pharmaceutical compositions comprising a CD3 antibody as further disclosed herein and a pharmaceutically acceptable carrier.

In another aspect, the present invention provides a method for modulating a T-cell-mediated immune response in a subject with such need, comprising administering an effective amount of an antibody or pharmaceutical composition as disclosed herein to the subject such that the immune response in the subject is modulated.

In one aspect, the present invention provides a method for inhibiting the growth of tumor cells in a subject, comprising administering to the subject, in an amount that effectively inhibits the growth of tumor cells, a bispecific antibody, a trispecific antibody, a multispecific antibody, a bispecific heterodimeric antibody, or a bispecific heterodimeric IgG as disclosed herein.

In some such embodiments, the tumor cells are cancer cells. In specific embodiments, the cancer is selected from: breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer, bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and cancers of the digestive system.

In one embodiment, the cancers of the digestive system are selected from: esophageal cancer, gastric cancer, small bowel cancer, colon cancer, rectal cancer, anal cancer, liver cancer, gallbladder cancer, appendix cancer, bile duct cancer, and pancreatic cancer.

In one embodiment, the present invention provides an antibody or a pharmaceutical composition comprising the antibody for use in a therapy.

In another embodiment, the present invention provides the use of the antibody of the present invention in the preparation of a medicament for use in a therapy. In some such embodiments, the therapy comprises activating a cytolytic T-cell response.

In one aspect, the present invention provides a polynucleotide comprising a nucleotide sequence encoding an antibody as disclosed herein. In another aspect, the present invention provides a vector comprising this polynucleotide.

In another aspect, the present invention provides host cells comprising vectors as disclosed herein. In some such embodiments, the host cells recombinantly produce antibodies as disclosed herein. In specific embodiments, the host cells are selected from bacterial cell lines, mammalian cell lines, insect cell lines, and yeast cell lines. In one specific embodiment, the mammalian cell line is a CHO cell line.

In one aspect, the present invention provides a method for generating antibodies as disclosed herein, comprising: culturing host cells under conditions that result in the generation of antibodies as disclosed herein, and purifying said antibodies from the culture supernatant.

In another aspect, the present invention provides the use of CD3 antibodies, polynucleotides, vectors or host cells as disclosed herein in the preparation of medicaments for treating diseases.

In one aspect, the present invention provides an antibody or pharmaceutical composition for use in a method of modulating a T-cell-mediated immune response in a subject with such a need. In a specific embodiment, the present invention provides an antibody or pharmaceutical composition for use in a method of inhibiting the growth of tumor cells in a subject.

In another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, optionally wherein said cancer is selected from: breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer, bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and cancers of the digestive system.

In yet another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, wherein a T-cell-mediated immune response is modulated or wherein the growth of tumor cells is inhibited.

Other embodiments will become apparent from the detailed summary. When aspects or embodiments of the invention are described in terms of the Markush group or other alternative groupings, the invention covers not only the entire group listed as a whole, but also each member of that group individually and all possible subgroups of the main group, as well as the main group in which one or more members are not present. The invention also contemplates the explicit exclusion of any one or more members of the claimed invention.

Brief description of the figure/attachment

Figure 1 provides a schematic diagram of the architecture of a bispecific antibody having a first heterodimer promoting domain and a second heterodimer promoting domain, the bispecific antibody comprising an Fc chain optimized for association via mortar association.

Figure 2 depicts the results of the differential scanning calorimetry temperature recording, showing the heat capacity of the GUCY2c-H2B4 and GUCY2c-2B5 bispecific antibodies as a function of temperature.

Figure 3 depicts the binding of the bispecific antibodies GUCY2c-H2B4 (GUCY2c-0247) and GUCY2c-2B5 (GUCY2c-0250) to naïve human T cells as determined by flow cytometry.

Figure 4 depicts the results of in vitro cytotoxicity mediated by the bispecific antibodies GUCY2c-H2B4 (GUCY2c-0247) and GUCY2c-2B5 (GUCY2c-0250) using cell survival data.

Figure 5 depicts the potential energy surface (PES) of the H2B4 Fv region from an X-ray crystal structure. This includes three views highlighting the VH region, the VH/VL interface, and the VL region. The CDR (Crystal Radiata Diffraction) with excess positive charge is indicated for reference. Dark black represents positive charge and white represents negative charge.

Figure 6 depicts the spatial aggregation tendency (SAP) surface of the Fv region of H2B4. Regions with concentrated hydrophobic residues are marked.

Figure 7 depicts the heterogeneity among bispecific antibodies containing CD3 antibody-binding domains H2B4, 2B5, and 2B5v6 using capillary gel electrophoresis.

Figure 8 depicts the temperature record using differential scanning calorimetry (DSC), showing the heat capacity of the GUCY2c-1608 (GUCY2c-2B5v6) bispecific antibody as a function of temperature.

Figure 9 depicts the binding of the GUCY2c-1608 (GUCY2c-2B5v6) bispecific antibody to naive T cells.

Figure 10 depicts cell survival data for in vitro cytotoxicity mediated by the bispecific antibody GUCY2c-1608 (GUCY2c-2B5v6).

Figures 11A to 11F depict in vitro cytokine release measurements of the GUCY2c-1608 (GUCY2c-2B5v6) bispecific antibody. The bispecific antibody recruited naïve human T cells to GUCY2c-expressing T84 cells to induce in vitro cytokine release. Luminex-based assays showed upregulation of human IFN-γ (Figure 11A), IL10 (Figure 11B), IL2 (Figure 11C), IL4 (Figure 11D), IL6 (Figure 11E), and TNF-α (Figure 11F).

Figure 12 illustrates the dose-dependent tumor growth inhibition of the bispecific antibody GUCY2c-1608 (GUCY2c-2B5v6) in the patient-derived xenograft PDX-CRX-11201 of colorectal cancer in the adoptive metastasis system.

Figure 13 illustrates the dose-dependent tumor growth inhibition of the bispecific antibody GUCY2C-1608 (GUCY2C-2B5v6) in the xenograft LS1034 colorectal cancer cell line in the adoptive metastasis system.

Figure 14A depicts the results of cell viability data, showing the in vitro cytotoxicity mediated by the FLT3-2B5v6 bispecific antibody using Flt3-expressing MV4-11 cells at different effector-to-target (E:T) ratios. Figure 14B depicts the results of cell viability data, showing the in vitro cytotoxicity mediated by the FLT3-2B5v6 bispecific antibody using Flt-3-expressing EOL-1 cells at different effector-to-target (E:T) ratios.

Figure 15 depicts the binding of the low-affinity 2B5v6 variant to Jurkat cells expressing CD3.

Figure 16 depicts a graph summarizing the findings, showing the cytotoxicity mediated by anti-GUCY2C bispecific T cells with anti-CD3 variants.

Detailed Explanation

The present invention disclosed herein provides antibodies that specifically bind to CD3 (e.g., human CD3), including full-length antibodies or antigen-binding fragments thereof. The invention also provides polynucleotides encoding these antibodies, compositions comprising these antibodies, and methods for preparing and using these antibodies. The invention further provides methods for treating conditions using antibodies as described herein.

General Technology

Unless otherwise indicated, the practice of this invention will employ conventional techniques of molecular biology (including recombinant techniques), microbiology, cell biology, biochemistry, and immunology, which are within the scope of the art. Such techniques are fully explained in the literature such as *Molecular Cloning: A Laboratory Manual, 2nd Edition* (Sambrook et al., 1989); *Cold Spring Harbor Press*; *Oligonucleotide Synthesis* (M.J. Gait, ed., 1984); *Methods in Molecular Biology*, *Humana Press*; *Cell Biology: A Laboratory Notebook* (J.E. Cellis, ed., 1998); *Academic Press*; *Animal Cell Culture* (R.I. Freshney, ed., 1987); *Introduction to Cell and Tissue Culture* (J.P.). Mather and P.E. Roberts, 1998) Plenum Press; Cell and Tissue Culture: LaboratoryProcedures (A.Doyle, J.B. Griffiths, and D.G. Newell, eds., 1993-1998) J. Wileyand Sons; Methods in Enzymolog y (Academic Press, Inc.); Handbook of Experimental Immunology (D.M. Weir and C.C. Blackwell, Eds.); Gene Transfer Vectors for Mammalian Cells (J.M. Miller and M.P. Calos, Eds., 1987); Current Protoc Mullis in Molecular Biology (F.M. Ausubel et al., eds., 1987); PCR: The Polymerase Chain Reaction (Mullis et al., eds., 1994); Current Protocols in Immunology (J.E. Colligan et al., eds., 1991); Short Protocols in Molecular Biology (Wiley and Sons, 1999); Immunobiology (C.A. Janeway and P. Travers, 1997); Antibodies (P. Finch, 1997); Antibodies: a practical approach (D. Catty., ed., IRL Press, 19...). 88-1989); Monoclonal Antibodies: A Practical Approach (P. Shepherd and C. Dean, eds., Oxford University Press, 2000); Using Antibodies: A Laboratory Manual (E. Harlow and D. Lane (Cold Spring Harbor Laboratory Press, 1999); The Antibodies (M. Zanetti and J.D. Capra, eds., Harwood Academic Publishers, 1995). In some cases, terms having their usual meaning are defined herein for clarity and/or for the purpose of preparing references, and the inclusion of such definitions herein is not necessarily construed as indicating a material difference from definitions commonly understood in the art.

The invention will now be described in detail by reference using the following definitions and examples. All patents and publications mentioned herein (including all sequences disclosed in such patents and publications) are expressly incorporated by reference.

definition

Generally, unless otherwise defined, all terms, symbols, and other scientific or terminological terms used herein are intended to have the meaning commonly understood by one of ordinary skill in the art to which this invention relates. For example, the term "and/or" as used herein, such as in phrases like "A and/or B," is intended to include both A and B; A or B; A (alone); and B (alone). Similarly, the term "and/or" as used herein, such as in phrases like "A, B, and/or C," is intended to cover each of the following embodiments: A, B, and C; A, B, or C; A or C; A or B; B or C; A and C; A and B; B and C; A (alone); B (alone); and C (alone).

As used herein, unless the context clearly indicates otherwise, the singular forms “a”, “an” and “the” include their respective multiple indicators.

As used in this article, the numerical range includes the numerical values that define the range.

The terms “about,” “approximately,” etc., when preceding a list of numerical values or ranges, independently refer to each individual value in that list or range as if each individual value immediately followed the term. The term implies that the value referred to by the term is the same as, close to, or similar to. For example, in some embodiments, “about” or “approximately” for a particular value may indicate 99%, 95%, or 90% of that value. As an example, the expression “about 100” includes 99 and 101 and all values between them (e.g., 99.1, 99.2, 99.3, 99.4, 99.5, etc.). As another example, if the temperature is 70°C, “about” or “approximately” 70°C may be equal to 69°C, 66°C, or 63°C. It should be understood that these are merely examples.

As used herein, nucleic acids are written from left to right in a 5' to 3' orientation; amino acid sequences are written from left to right in an amino-to-carboxyl orientation. For definitions and terminology in this art, practitioners particularly refer to Sambrook et al., 1989 and Ausubel FM et al., 1993. It should be understood that the invention is not limited to the specific methods, schemes, and reagents described, as these are subject to variation.

The terms “polypeptide,” “oligopeptide,” “peptide,” and “protein” are used interchangeably herein to refer to an amino acid chain of any length, preferably relatively short (e.g., 10-100 amino acids). This chain may be linear or branched, may contain modified amino acids, and/or may be interrupted by non-amino acid components. The term also covers amino acid chains that have been modified naturally or through intervention; for example, disulfide bond formation, glycosylation, esterification, acetylation, phosphorylation, or any other operation or modification, such as conjugation with a labeled component. This definition also includes, for example, polypeptides containing one or more amino acid analogs (including, for example, non-natural amino acids) and other modifications known in the art. It should be understood that polypeptides may exist as single chains or associated chains. Preferably, mammalian polypeptides (polypeptides originally derived from mammalian organisms) are used, more preferably those secreted directly into the culture medium.

An "antibody" is an immunoglobulin molecule that specifically binds to a target, such as carbohydrates, polynucleotides, lipids, peptides, etc., through at least one antigen recognition site located in the variable region of the immunoglobulin molecule. As used herein, the term encompasses polyclonal antibodies, monoclonal antibodies, chimeric antibodies, bispecific antibodies, dual-specific antibodies, bifunctional antibodies, trispecific antibodies, multispecific antibodies, bispecific heterodimeric biantibodies, bispecific heterodimeric IgG, labeled antibodies, humanized antibodies, human antibodies and their fragments (such as Fab, Fab', F(ab') ₂ , Fv), single-chain (ScFv) and domain antibodies (including, for example, antibodies from sharks and camels), fusion proteins containing antibodies, and any other modified conformation of an immunoglobulin molecule containing an antigen recognition site. Antibodies include any class of antibodies, such as IgG, IgA, or IgM (or their subclasses), and the antibody need not be of any particular class. Immunoglobulins can be designated as different classes depending on the amino acid sequence of their heavy chain constant region. There are five main classes of immunoglobulins: IgA, IgD, IgE, IgG, and IgM, and several of these can be further subdivided into subclasses (isotypes), such as IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2. The heavy chain constant regions corresponding to different classes of immunoglobulins are designated α, δ, ε, γ, and μ, respectively. The subunit structures and three-dimensional conformations of different classes of immunoglobulins are well known. This invention also includes "antibody analogs," other scaffolds based on non-antibody molecular proteins, such as fusion proteins and/or immunoconjugates using CDRs to provide specific antigen binding. The antibodies of this invention can be derived from any species, including but not limited to mice, humans, camels, llamas, fish, sharks, goats, rabbits, chickens, horses, and cattle.

The term "antibody" further includes immunoglobulin molecules comprising four polypeptide chains (two heavy chains (H) and two light chains (L)) linked by disulfide bonds, as well as their multimers (e.g., IgM). Each heavy chain comprises a heavy chain variable region (abbreviated herein as HC VR or VH) and a heavy chain constant region. The "variable region" of an antibody refers to the variable region of either the antibody light chain alone or in combination with the variable region of the antibody heavy chain. The heavy chain constant region comprises three domains: CH1, CH2, and CH3. The CH1 and CH2 domains are linked by hinge regions. Each light chain comprises a light chain variable region (abbreviated herein as LC VR or VL) and a light chain constant region. The light chain constant region comprises one domain (CL1). The VH and VL regions can be further subdivided into hypervariable regions (called complementarity-determining regions (CDRs)) which are interspersed with more conserved regions (called framework regions (FRs)). Each VH and VL consists of three CDRs and four FRs, arranged in the following order from the amino terminus to the carboxyl terminus: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. In different embodiments of the invention, the FRs of the CD3 antibody (or its antigen-binding portion) may be identical to the human germline sequence, or may be natural or artificially modified. The common amino acid sequence can be defined based on a side-by-side analysis of two or more CDRs. The CDRs in each chain are tightly held together by the FRs and, together with the CDRs from the other chain, contribute to the formation of the antigen-binding site of the antibody.

As used herein, the terms “antigen-binding fragment,” “antibody fragment,” or “antigen-binding portion” refer to one or more segments of an antibody that retain the ability to specifically bind to an antigen. Examples of binding fragments encompassed within the term "antigen-binding fragment" in the term antibody include (i) antibody heavy chain variable domains (VH) and/or antibody light chain variable domains (VL), or VH/VL pairs derived from full-length antibodies or antibody fragments (such as VH and/or VL domains); (ii) Fab fragments, i.e., monovalent fragments consisting of VL, VH, CL, and CH1 domains; (iii) Fab' fragments, which are essentially Fab fragments having a portion of a hinge region (e.g., Fundamental Immunology, edited by Paul, 3rd edition, 1993); (iv) F(ab') ₂ fragments, i.e., bivalent fragments containing two Fab fragments connected by a disulfide bridge at the hinge region; (v) Fd fragments, which consist of VH and CH1 domains; (vi) Fv fragments, which consist of the VL and VH domains of a single arm of the antibody; (vii) Single-chain Fv fragments (scFv), i.e., a single protein chain in which the VL and VH regions pair to form a monovalent molecule (e.g., Bird et al. (1988) Science 242:423-426; and Huston et al. (1988) Proc. Natl. Acad. Sci. USA 85:5879-5883); (viii) disulfide-stabilized Fv fragments (dsFv), i.e., Fv with engineered intermolecular disulfide bonds that stabilize the VH-VL pair; (ix) single variable domain antibody (sdAb or dAb) fragments (e.g., Ward et al. (1989) Nature 341:544-546), which consist of a variable domain of a heavy chain and have no light chain; (x) complementarity-determining regions (CDRs); and any derivative thereof.

As used herein, an antibody's "antigen-binding fragment" may contain at least one variable domain. The variable domain may have any size or amino acid composition and will typically contain at least one CDR adjacent to or framed with one or more framework sequences. As used herein, an "antigen-binding fragment" of an antibody may comprise a homodimer or heterodimer (or other multimer) of any variable region and constant domain configuration that is non-covalently associated with each other and/or non-covalently associated (e.g., via disulfide bonds) with one or more monomeric VH or VL regions. For example, the variable region may be dimer and contain VH-VH, VH-VL, or VL-VL dimers. The configurations of the variable and constant domains that can be found within the antigen-binding fragment of the antibody of the present invention include: VH-CH1; VH-CH2; VH-CH3; VH-CH1-CH2; VH-CH1-CH2-CH3; VH-CH2-CH3; VH-VL-CL, VH-VL-CH1, VH-VL-CH2; VH-CL; VL-CH1; VL-CH2; VL-CH3; VL-CH1-CH2; VL-CH1-CH2-CH3; VL-CH2-CH3; and VL-CL. The variable and constant domains can be directly connected to each other or can be connected via complete or partial hinge or linker regions. The hinge region can consist of at least two (e.g., five, ten, fifteen, twenty, forty, sixty, or more) amino acids, which form flexible or semi-flexible bonds between adjacent variable and/or constant domains in a single polypeptide molecule.

As used herein, a “binding domain” includes any region of a polypeptide (e.g., an antibody) responsible for selectively binding to a target molecule (e.g., an antigen, ligand, receptor, substrate, or inhibitor). Exemplary binding domains include antibody variable regions, receptor-binding domains, ligand-binding domains, and enzyme-catalyzing domains.

As used herein, the term "human receptor framework" is a framework comprising an amino acid sequence of a light chain variable (VL) framework or a heavy chain variable (VH) framework derived from the human immunoglobulin framework or the human common framework as defined below. A human receptor framework "derived from" the human immunoglobulin framework or the human common framework may contain the same amino acid sequence, or may contain amino acid sequence modifications. In some embodiments, the number of amino acid modifications is 10 or less, 9 or less, 8 or less, 7 or less, 6 or less, 5 or less, 4 or less, 3 or less, or 2 or less. In some embodiments, the sequence of the VL human receptor framework is identical to the sequence of the VL human immunoglobulin framework or the human common framework.

As used herein, an “affinity-matured” antibody is an antibody that has one or more modifications in one or more variable regions (including CDR and FR) compared to an unmodified parent antibody, wherein such modifications result in an increased affinity of the antibody for the antigen.

As used herein, the terms “Fc region,” “Fc domain,” “Fc chain,” or similar terms are used to define the C-terminal region of the IgG heavy chain. The Fc region of IgG contains two constant domains, CH2 and CH3. The CH2 domain of the human IgG Fc region typically extends from amino acid 231 to amino acid 340 according to the EU index numbering system, or from amino acid 244 to amino acid 360 according to the Kabat numbering system. The CH3 domain of the human IgG Fc region typically extends from amino acid 341 to amino acid 447 according to the EU index numbering system, or from amino acid 361 to amino acid 478 according to the Kabat numbering system. The CH2 domain of the human IgG Fc region (also known as the “Cγ2” domain) is unique because it does not pair closely with the other domain. Instead, two N-linked branched carbohydrate chains are inserted between the two CH2 domains of intact native IgG. The Fc region may contain native or variant Fc sequences. Although the boundaries of the Fc sequence of the immunoglobulin heavy chain may vary, the human IgG heavy chain Fc sequence is generally defined as an extension from an amino acid residue at approximately position Cys226 or approximately position Pro230 to the carboxyl terminus of said Fc sequence. Unless otherwise specified herein, the amino acid residues in the Fc region or constant region are numbered according to the EU numbering system, also known as the EU index, as described in Kabat et al., Sequences of Proteins of Immunological Interest, 5th edition, Public Health Service, National Institutes of Health, Bethesda, MD, 1991.

In some embodiments, the Fc chain begins in a hinge region immediately upstream of the papain cleavage site and terminates at the C-terminus of the antibody. Therefore, a complete Fc chain comprises at least a hinge domain, a CH2 domain, and a CH3 domain. In some embodiments, the Fc chain comprises at least one of the following: a hinge (e.g., upper, middle, and/or lower hinge region) domain, a CH2 domain, a CH3 domain, a CH4 domain, or a variant, portion, or fragment thereof. In some embodiments, the Fc domain comprises a complete Fc chain (i.e., a hinge domain, a CH2 domain, and a CH3 domain). In some embodiments, the Fc chain comprises a hinge domain (or a portion thereof) fused to a CH3 domain (or a portion thereof). In some embodiments, the Fc chain comprises a CH2 domain (or a portion thereof) fused to a CH3 domain (or a portion thereof). In some embodiments, the Fc chain consists of a CH3 domain or a portion thereof. In some embodiments, the Fc chain consists of a hinge domain (or a portion thereof) and a CH3 domain (or a portion thereof). In some embodiments, the Fc chain consists of a CH2 domain (or a portion thereof) and a CH3 domain. In some embodiments, the Fc chain consists of a hinge domain (or a portion thereof) and a CH2 domain (or a portion thereof). In some embodiments, the Fc chain lacks at least a portion of the CH2 domain (e.g., all or a portion of the CH2 domain). As used herein, Fc chain generally refers to a polypeptide comprising all or a portion of the Fc chain of an immunoglobulin heavy chain. This Fc chain includes, but is not limited to, polypeptides comprising the entire CHI, hinge, CH2, and/or CH3 domains, and fragments of such peptides comprising only, for example, the hinge, CH2, and CH3 domains. Fc chains can be derived from immunoglobulins of any species and/or any subtype, including but not limited to human IgG1, IgG2, IgG3, IgG4, IgD, IgA, IgE, or IgM antibodies. The Fc domain encompasses native Fc and Fc variant molecules. As with Fc variants and native Fc, the term Fc chain includes molecules in monomeric or polymeric form, whether derived from whole antibody digestion or otherwise. In some embodiments, the Fc chain comprises the carboxyl-terminal portions of two heavy chains held together by disulfide bonds. In some embodiments, the Fc chain consists of a CH2 domain and a CH3 domain.

As used in the art, “Fc receptor” and “FcR” describe receptors that bind to the Fc region of an antibody. Preferred FcRs are naturally occurring human FcRs. Furthermore, preferred FcRs are those that bind IgG antibodies (γ receptors) and include receptors of the FcγRI, FcγRII, and FcγRIII subclasses, including allelic variants and alternative splicing forms of these receptors. FcγRII receptors include FcγRIIA (“activating receptor”) and FcγRIIB (“inhibitory receptor”), which are distinguished primarily by similar amino acid sequences in their cytoplasmic domains. FcRs are reviewed in: Ravetch and Kinet, Ann. Rev. Immunol., 9:457-92, 1991; Capel et al., Immunomethods, 4:25-34, 1994; and de Haas et al., J. Lab. Clin. Med., 126:330-41, 1995. "FcR" also includes the neonatal receptor FcRn, which is responsible for the transfer of maternal IgG to the fetus (Guyer et al., J. Immunol., 117:587, 1976; and Kim et al., J. Immunol., 24:249, 1994).

The “natural sequence Fc region” or “wild-type Fc region” contains the same amino acid sequence as the Fc region found in nature. “Wild-type” human IgG Fc refers to the sequence of amino acids naturally present in the human population. Of course, just as the Fc sequence can vary slightly between individuals, one or more modifications can be made to the wild-type sequence while still remaining within the scope of this invention. For example, the Fc region may contain additional modifications unrelated to this invention, such as mutations in glycosylation sites, including non-natural amino acids, or “knobs-in-holes” mutations.

The “variant Fc region” or “variant Fc chain” comprises an amino acid sequence that differs from the amino acid sequence of the natural Fc region by means of at least one amino acid modification, but retains at least one effector function of the natural Fc region. In some embodiments, the variant Fc chain has at least one amino acid substitution compared to the natural Fc chain or the Fc region of the parent polypeptide, for example, about 1 to about 10 amino acid substitutions in the natural Fc chain or the Fc chain of the parent polypeptide, and preferably about 1 to about 5 amino acid substitutions. The variant Fc chain described herein will preferably have at least about 80% sequence identity with the natural Fc chain and/or the Fc chain of the parent polypeptide, and most preferably at least about 90% sequence identity, more preferably at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99% sequence identity.

As used herein, the term "effective function" refers to those biological activities induced by the Fc region of an antibody (the native sequence Fc chain or the Fc chain of an amino acid sequence variant), and varies with antibody isotype. Examples of antibody effector functions include: C1q binding and complement-dependent cytotoxicity; Fc receptor binding; antibody-dependent cell-mediated cytotoxicity (ADCC); phagocytosis; downregulation of cell surface receptors (e.g., B cell receptors); and B cell activation. Such effector functions generally require the combination of the Fc chain with a binding domain (e.g., the antibody variable region) and can be evaluated using various assays known in the art for assessing such antibody effector functions. Exemplary measurements of effector functions are via Fcγ3 and/or C1q binding.

The effector function of an antibody is determined by a sequence in the Fc chain; this chain is also the part recognized by the Fc receptor (FcR) found on certain types of cells.

In some embodiments, the Fc polypeptide contains part or all of the wild-type hinge sequence (typically at its N-terminus). In some embodiments, the Fc polypeptide does not contain a functional or wild-type hinge sequence.

As used herein, “hinge region,” “hinge sequence,” and their variations include meanings known in the art, as illustrated in, for example, Janeway et al., ImmunoBiology: the immune system in health and disease, Elsevier Science Ltd., NY (4th edition, 1999); Bloom et al., Protein Science, 6:407-415, 1997; and Humphreys et al., J. Immunol. Methods, 209:193-202, 1997.

As used herein, "immunoglobulin-like hinge region," "immunoglobulin-like hinge sequence," and variations thereof refer to the hinge region and hinge sequence of an immunoglobulin-like or antibody-like molecule (e.g., an immunoadhesin). In some embodiments, the immunoglobulin-like hinge region may be derived from or originate from any IgG1, IgG2, IgG3, or IgG4 subtype, or IgA, IgE, IgD, or IgM, including its chimeric forms, such as a chimeric IgG1/2 hinge region.

In some embodiments, the hinge region may be derived from a human IgG1 subtype extending from amino acid 216 to amino acid 230 according to the EU indexing system or from amino acid 226 to amino acid 243 according to the Kabat numbering system. In some embodiments, the sequence may be EPKSCDKTHTCPPCP (SEQ ID NO: 63). Those skilled in the art may have different understandings of the exact amino acids corresponding to the various domains of the IgG molecule. Therefore, the N-terminus or C-terminus of the domains outlined above may be extended or shortened by 1, 2, 3, 4, 5, 6, 7, 8, 9, or even 10 amino acids.

In some embodiments, the hinge region may be mutated by one or more amino acids. In some embodiments, the hinge region may be truncated and contain only a portion of the full hinge region. In some embodiments, the hinge region may contain only the last 5 amino acids of the hinge region, referred to herein as the "lower hinge" region. In some embodiments, the lower hinge region may contain amino acid CPPCP (SEQ ID NO: 64) or amino acids 226 to 230 according to the EU index numbering system or amino acids 239 to 243 according to the Kabat numbering system.

As used herein, the term "modification" refers to the substitution, insertion, and/or deletion of amino acids in a polypeptide sequence, alteration of the portion chemically linked to the protein, or modification of the function of a protein (e.g., an antibody). For example, a modification can be an alteration of antibody function or a change in the structure of a carbohydrate linked to the protein. As used herein, "amino acid modification" refers to a mutation (substitution), insertion (addition), or deletion of one or more amino acid residues in an antibody. The term "amino acid mutation" means that at least one existing amino acid residue is replaced by another different amino acid residue (e.g., a substitute amino acid residue). The term "amino acid deletion" means the removal of at least one amino acid residue at a predetermined position in the amino acid sequence. For example, mutation L234A indicates that the amino acid residue lysine at position 234 in the Fc region of the antibody is replaced by the amino acid residue alanine (replacing lysine with alanine) (numbered according to the EU index numbering system). "Natural amino acid residues" refers to amino acid residues from the following groups: alanine (three-letter code: Ala, single-letter code: A), arginine (Arg, R), asparagine (Asn, N), aspartic acid (Asp, D), cysteine (Cys, C), glutamine (Gin, Q), glutamic acid (Glu, E), glycine (Gly, G), histidine (His, H), isoleucine (Ile, I), leucine (Leu, L), lysine (Lys, K), methionine (Met, M), phenylalanine (Phe, F), proline (Pro, P), serine (Ser, S), threonine (Thr, T), tryptophan (Trp, W), tyrosine (Tyr, Y), and valine (Val, V).

The term "pharmaceutical" is used herein to mean a biological macromolecule, an extract derived from biological material, a mixture of biological macromolecules, a chemical compound, a mixture of chemical compounds, and/or a mixture of chemical compounds and biological macromolecules. The term "therapeutic agent" refers to a biologically active pharmaceutical agent.

As used herein, a “monoclonal antibody” refers to an antibody obtained from a substantially homogeneous population of antibodies, i.e., the individual antibodies constituting the population are identical except for the possibility of naturally occurring mutations present in small amounts. Monoclonal antibodies are highly specific against a single antigenic site. Furthermore, in contrast to polyclonal antibody formulations, which typically comprise different antibodies targeting different determinants (epitopes), each monoclonal antibody targets a single determinant on the antigen. The modifier “monoclonal” indicates the characteristics of antibodies obtained from a substantially homogeneous population of antibodies and should not be construed as requiring antibody production by any particular method. For example, a monoclonal antibody to be used according to the invention can be prepared by a hybridoma method first described by Kohler and Milstein, Nature 256:495, 1975, or by a recombinant DNA method such as that described in U.S. Patent No. 4,816,567. Monoclonal antibodies can also be isolated from phage libraries generated using, for example, techniques described in McCafferty et al., Nature 348:552-554, 1990.

The antibodies of this invention can be “humanized antibodies.” As used herein, a “humanized” antibody refers to a form of non-human (e.g., mouse, rat, rabbit, non-human primate, or other mammalian) antibody that is a chimeric immunoglobulin, immunoglobulin chain, or fragment thereof (such as Fv, Fab, Fab', F(ab') ₂ , or other antigen-binding sequences of the antibody) containing one or more amino acid residues introduced therefrom from a non-human source. These non-human amino acid residues are generally referred to as “input” residues and are typically derived from an “input” variable domain. The input residues, sequence, or antibody has desired affinity and/or specificity or other desired antibody biological activities as discussed herein.

Preferably, the humanized antibody is a human immunoglobulin (receptor antibody) in which residues from the complementarity-determining region (CDR) of the receptor are replaced with residues from a non-human species (donor antibody), such as mouse, rat, or rabbit, that have the desired specificity, affinity, and ability. In some cases, residues from the Fv framework region (FR) of the human immunoglobulin are replaced with corresponding non-human residues. Furthermore, the humanized antibody may contain residues not found in either the receptor antibody or the input CDR or framework sequence, but which are included to further refine and optimize antibody performance. Typically, the humanized antibody will contain at least one, and usually substantially all, of the two variable regions, wherein all or substantially all of the CDR regions correspond to those of non-human immunoglobulins, and all or substantially all of the FR regions are those of human immunoglobulin common sequences. The humanized antibody will also preferably contain at least a portion of the immunoglobulin constant region or structural domain (Fc), typically of the type of human immunoglobulin. Antibodies having Fc chains modified as described in WO 99/58572 are preferred. Other forms of humanized antibodies have one or more CDRs (CDR L1, CDR L2, CDR L3, CDR H1, CDR H2, or CDR H3) that are altered relative to the original antibody, and are also referred to as "derived from" one or more CDRs from the original antibody. As used herein, humanization intention includes deimmunization antibodies.

As used herein, "human antibody" means an antibody having an amino acid sequence corresponding to the amino acid sequence of an antibody produced by a human, and/or having been prepared using any technique known to those skilled in the art or disclosed herein for the preparation of human antibodies. Therefore, the term "human antibody" as used herein is intended to include antibodies having variable and constant regions derived from human immunoglobulin sequences. Human antibodies of the present invention may include amino acid residues not encoded by human immunoglobulin sequences (e.g., mutations introduced by in vitro random mutagenesis or site-specific mutagenesis or by in vivo somatic mutations), such as in the CDR, and specifically CDR3. This definition of human antibody includes antibodies comprising at least one human heavy chain polypeptide or at least one human light chain polypeptide. One such example is an antibody comprising a mouse light chain and a human heavy chain polypeptide. Various techniques known in the art can be used to produce human antibodies. In one embodiment, the human antibody is selected from a phage library expressing a human antibody (Vaughan et al., Nature Biotechnology, 14:309-314, 1996; Sheets et al., Proc. Natl. Acad. Sci. (USA) 95:6157-6162, 1998; Hoogenboom and Winter, J. Mol. Biol., 227:381, 1991; Marks et al., J. Mol. Biol., 222:581, 1991). The human antibody can also be prepared by immunizing an animal in which a human immunoglobulin locus has been transgenically introduced to replace an endogenous locus, such as in mice where the endogenous immunoglobulin gene has been partially or completely inactivated. This method is described in U.S. Patent Nos. 5,545,807; 5,545,806; 5,569,825; 5,625,126; 5,633,425; and 5,661,016. Alternatively, human antibodies can be prepared by immortalizing human B lymphocytes that produce antibodies against a target antigen (such B lymphocytes can be recovered from an individual or a single-cell clone of cDNA, or can have been immunized in vitro). See, for example, Cole et al. Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, p. 77, 1985; Boerner et al., J. Immunol., 147(1):86-95, 1991; and U.S. Patent No. 5,750,373.

The human antibodies of the present invention can exist in at least two forms related to hinge heterogeneity. For example, immunoglobulin molecules comprise a stable four-chain construct of approximately 150-160 kDa, wherein the dimers are held together by interchain heavy chain disulfide bonds. Alternatively, the dimers are not linked by interchain disulfide bonds and form a molecule of approximately 75-80 kDa (half-antibody) consisting of covalently coupled light and heavy chains.

As used herein, the term "recombinant human antibody" is intended to include all human antibodies prepared, expressed, generated, or isolated in a recombinant manner, such as antibodies expressed using a recombinant expression vector transfected into host cells (described further below), antibodies isolated from a library of recombinant human antibodies (described further below), antibodies isolated from animals transgenic with human immunoglobulin genes (e.g., mice) (see, for example, Taylor et al. (1992) Nucl. Acids Res. 20:6287-6295), or antibodies prepared, expressed, generated, or isolated by any other means involving splicing a human immunoglobulin gene sequence into another DNA sequence. Such recombinant human antibodies have variable and constant regions derived from human germline immunoglobulin sequences. However, in some implementations, such recombinant human antibodies are subjected to in vitro mutagenesis (or, when using animals transgenic with human Ig sequences, in vivo somatic cell mutagenesis), and thus the amino acid sequences of the VH and VL regions of the recombinant antibody are sequences that, although derived from and associated with human germline VH and VL sequences, may not naturally exist in the human antibody germline library.

The antibodies of this invention can be “chimeric antibodies” (where a portion of the heavy chain and/or light chain is identical or homologous to the corresponding sequence in an antibody derived from a specific species or belonging to a specific antibody class or subclass, while the remaining portion of the chain is identical or homologous to the corresponding sequence in an antibody derived from another species or belonging to another antibody class or subclass), and fragments of such antibodies, provided they exhibit the desired biological activity (US Patent 4,816,567; and Morrison et al., Proc. Natl. Acad. Sci. USA 81:6851-6855, 1984). The target chimeric antibodies herein include primate-derived antibodies comprising a variable region antigen-binding sequence derived from a non-human primate (e.g., Old World monkeys, apes, etc.) and a human constant region sequence.

A "monovalent antibody" contains one antigen-binding site/molecule (e.g., IgG or Fab). In some cases, a monovalent antibody may have more than one antigen-binding site, but the binding sites come from different antigens.

A "monospecific antibody" is an antibody or antibody preparation that contains two identical antigen-binding sites/molecules (e.g., IgG), such that the two binding sites bind to the same epitopes on the antigen. Therefore, they compete with each other when binding to a single antigen molecule. This term includes "monoclonal antibody" or "monoclonal antibody composition." Most antibodies found in nature are monospecific. In some cases, monospecific antibodies can also be monovalent antibodies (e.g., Fab).

The terms “mutation load,” “mutational load,” “mutation burden,” or “mutational burden” are used interchangeably in this article. Tumor mutational load is a measure of the number of mutations within a tumor genome, defined as the total number of mutations in each coding region of the tumor genome. Mutational load varies greatly across tumor types, ranging from just a few to several thousand mutations (Alexandrov LB et al., Nature 2013;500(7463):415-421; Lawrence MS et al., Nature 2013;499:214–218; Vogelstein B et al., Science, 2013;339:1546–1558).

"Original amino acid residue" refers to a residue replaced by an "input amino acid" residue whose side chain volume may be smaller or larger than the original residue. The input amino acid residue may be naturally occurring or non-naturally occurring amino acid residues, but is preferably the former. "Naturally occurring" amino acid residues are those residues encoded by the genetic code. "Non-naturally occurring" amino acid residues refer to residues that are not encoded by the genetic code but can covalently bind to neighboring amino acid residues in the polypeptide chain. Examples of non-naturally occurring amino acid residues are leucine, ornithine, valine, homoserine, and other amino acid residue analogs, such as those described, for example, in Ellman et al., Meth. Enzym. 202:301-336 (1991). In some embodiments, the method of the present invention involves replacing at least one original amino acid residue, but may replace more than one original residue. Typically, the total residues in no more than the first or second polypeptide interface may contain the replaced original amino acid residue.

As used herein with respect to the use of the term "competitive," it means that a first antibody or its antigen-binding fragment (or portion) binds to an epitope in a manner sufficiently similar to the binding of a second antibody or its antigen-binding portion, such that the binding of the first antibody to its homologous epitope is detectably reduced in the presence of the second antibody compared to the binding of the first antibody in the absence of the second antibody. Alternatives to this, where the binding of the second antibody to its epitope is also detectably reduced in the presence of the first antibody, are possible but not necessary. That is, the first antibody may inhibit the binding of the second antibody to its epitope, while the second antibody may not inhibit the binding of the first antibody to its corresponding epitope. However, when each antibody detectably inhibits the binding of another antibody to its homologous epitope or ligand, whether to the same, greater, or lesser extent, the antibodies are said to "cross-compete" with each other for binding to their respective epitopes. Both competitive antibodies and cross-competitive antibodies are covered by this invention. Regardless of the mechanism by which such competition or cross-competition occurs (e.g., steric hindrance, conformational change, or binding to a common epitope or a portion thereof), based on the teachings provided herein, those skilled in the art will understand that such competitive antibodies and/or cross-competitive antibodies are covered and can be used in the methods disclosed herein.

As used herein, “antibody-dependent cell-mediated cytotoxicity” or “ADCC” refers to a cell-mediated reaction in which nonspecific cytotoxic cells expressing Fc receptors (FcRs), such as natural killer (NK) cells, neutrophils, and macrophages, recognize binding antibodies on target cells and subsequently cause lysis of the target cells. The ADCC activity of a target molecule can be evaluated using in vitro ADCC assays, such as those described in U.S. Patent Nos. 5,500,362 or 5,821,337. Useful effector cells for such assays include peripheral blood mononuclear cells (PBMCs) and NK cells. Alternatively or additionally, the ADCC activity of a target molecule can be evaluated in vivo, for example, in animal models, such as those disclosed in Clynes et al., PNAS (USA), 95:652-656, 1998.

As used herein, “complement-dependent cytotoxicity” or “CDC” refers to target lysis in the presence of complement. The complement activation pathway is initiated by the binding of the first component (C1q) of the complement system to molecules (e.g., antibodies) that are complexed with homologous antigens. To evaluate complement activation, a CDC assay can be performed, for example, as described in Gazzano-Santoro et al., J. Immunol. Methods, 202: 163, 1996.

As used herein, the terms “immunospecific binding,” “immunospecific recognition,” “specific binding,” “specific recognition,” and similar terms refer to a molecule that specifically binds to an antigen (e.g., an epitope or immune complex) but does not specifically bind to another molecule (e.g., a binding domain). A molecule that specifically binds to an antigen may bind to other peptides or polypeptides with lower affinity, as determined by assays known in the art, such as immunoassays, BIACORE™, or other assays. Preferably, the molecule that specifically binds to the antigen does not cross-react with other proteins.

Suitable “moderately stringent conditions” include prewashing in a solution of 5X SSC, 0.5% SDS, and 1.0 mM EDTA (pH 8.0); hybridization overnight at 50–65°C with 5X SSC; followed by two washes at 65°C with 2X, 0.5X, and 0.2X SSC containing 0.1% SDS, each for 20 minutes.

As used herein, “highly stringent conditions” or “highly stringent conditions” are conditions in which: (1) low ionic strength and high temperature are used for washing, for example, 0.015 M sodium chloride/0.0015 M sodium citrate/0.1% sodium dodecyl sulfate at 50°C; (2) denaturing agents are used during hybridization at 42°C, such as formamide, for example, 50% (v/v) formamide with 0.1% bovine serum albumin/0.1% Ficoll/0.1% polyvinylpyrrolidone/50 mM sodium phosphate buffer with 750 mM sodium chloride and 75 mM sodium citrate at pH 6.5; Alternatively (3) wash at 42°C with 50% formamide, 5 x SSC (0.75 M NaCl, 0.075 M sodium citrate), 50 mM sodium phosphate (pH 6.8), 0.1% sodium pyrophosphate, 5 x Denhardt solution, sonicated salmon sperm DNA (50 µg/ml), 0.1% SDS, and 10% dextran sulfate, followed by washing at 42°C in 0.2 x SSC (sodium chloride/sodium citrate) and at 55°C in 50% formamide, followed by a high-rigidity wash at 55°C with 0.1 x SSC containing EDTA. Technicians will recognize how to adjust the temperature, ionic strength, etc., as needed to accommodate factors such as probe length.

The binding protein, binding domain, or CDR or antibody (as broadly defined herein) may be identified according to the accumulation, AbM, contact, North definition and/or conformational definition of Kabat, Chothia, Kabat and Chothia, or any method of CDR determination well known in the art. See, for example, Kabat et al., 1991, Sequences of Proteins of Immunological Interest, 5th edition (Hypervariate Region); Chothia et al., Nature 342:877-883, 1989 (Structural Ring Structure). The identity of the amino acid residues constituting the CDR in a particular antibody may be determined using methods well known in the art. The AbM definition of the CDR is a compromise between Kabat and Chothia and uses Oxford Molecular's AbM antibody modeling software (Accelrys®). The definition of “contact” for the CDR is based on observed antigen contact and is described in MacCallum et al., J. Mol. Biol., 262:732-745, 1996. The “conformation” definition of a CDR is based on residues that contribute enthalpy to antigen binding (see, for example, Makabe et al., J. Biol. Chem., 283:1156-1166, 2008). North has identified canonical CDR conformations using different preferred sets of CDR definitions (North et al., J. Mol. Biol. 406: 228-256, 2011). In another approach referred to herein as the “conformation definition” of a CDR, the position of the CDR can be identified as residues that contribute enthalpy to antigen binding (Makabe et al., J. Biol. Chem., 283:1156-1166, 2008). Other CDR boundary definitions may not strictly follow one of the above approaches but will still overlap with at least a portion of the Kabat CDR, although they may be shortened or lengthened based on predictions or experimental findings that specific residues or groups of residues or even the entire CDR do not significantly affect antigen binding. As used herein, a CDR can refer to a CDR defined by any method known in the art, including combinations of methods. The methods used herein can utilize a CDR defined according to any of these methods. For any given embodiment containing more than one CDR, a CDR (or other residues of the antibody) can be defined according to any of the following definitions: Kabat, Chothia, extension (a combination of Kabat and Chothia), North, extension, AbM, contact, and/or conformation.

Residues in the variable domain are numbered according to Kabat, a numbering system used for the heavy chain or light chain variable regions of the assembled antibody. See Kabat et al., Sequences of Proteins of Immunological Interest, 5th edition, Public Health Service, National Institutes of Health, Bethesda, MD, 1991. Using this numbering system, the actual linear amino acid sequence may contain fewer or additional shortened or inserted amino acids corresponding to the FR or CDR of the variable region. For example, the heavy chain variable region may include a single amino acid insertion after residue 52 of H2 (according to Kabat residue 52a) and inserted residues after heavy chain FR residue 82 (e.g., according to Kabat residues 82a, 82b, and 82c). The Kabat number of residues for a given antibody can be determined by comparing the homologous regions of the antibody sequence to a “standard” Kabat numbered sequence. Various algorithms for assigning Kabat numbers are available. This article uses the algorithm implemented in Abysis release 2.3.3 (www.abysis.org) to assign Kabat numbers to the variable regions VLCDR1, VL CDR2, VL CDR3, VHCDR1, VHCDR2, and VH CDR3. Specific amino acid residue positions within the antibody can also be numbered based on Kabat numbers.

As used herein, the term "phage display library" refers to a population of phages, each containing exogenous cDNA fused to a surface protein within a recombination frame. The phages display the exogenous protein encoded by the cDNA on their surface. After replication in a bacterial host (typically *E. coli*), phages containing the target exogenous cDNA are selected based on the expression of the exogenous protein on the phage surface.

The term "epitope" refers to a portion of a molecule that can be recognized, contacted, and/or bound by an antibody at one or more sites in an antigen-binding region called a complementary site. A single antigen may have more than one epitope. Thus, different antibodies can bind to different regions on the antigen and may have different biological effects. Epitopes are often composed of chemically active surface groups of a molecule (such as amino acid or sugar side chains) and have specific three-dimensional structural features and specific charge features. As used herein, epitopes can be conformational or linear. Conformational epitopes are generated by spatially juxtaposing amino acids from different segments of a linear polypeptide chain. Linear epitopes are epitopes generated from neighboring amino acid residues in a polypeptide chain. In some embodiments, an epitope may include a sugar, phosphoryl, or sulfonyl portion of the antigen.

As used herein, an "antigenic epitope" is defined as a portion of a polypeptide to which an antibody can specifically bind, as determined by any method well known in the art (e.g., by a routine immunoassay). A "nonlinear epitope" or "conformational epitope" comprises a non-neighboring polypeptide (or amino acid) within an antigenic protein to which an antibody specific for that epitope binds. Once a desired epitope on an antigen is identified, an antibody targeting that epitope can be generated, for example, using the techniques described herein.

When used to describe the interaction between antibodies and proteins or peptides, the terms "specific binding" or "specifically binding" refer to an interaction that depends on the presence of a specific structure on the protein (i.e., an antigenic determinant or epitope); in other words, the antibody recognizes and binds to a specific protein structure rather than the usual protein. For example, if an antibody is specific to epitope "A," the presence of a protein containing epitope A (or free unlabeled A) in a reaction containing labeled "A" and an antibody will reduce the amount of labeled A that binds to the antibody.

In some embodiments, "specific binding" means, for example, that an antibody binds to a protein with a KD of about 0.1 nM or less, but more typically less than about 1 μM. In some embodiments, "specific binding" means that an antibody sometimes binds to a target with a KD of at least about 0.1 μM or less, at other times with at least about 0.01 μM or less, and at other times with at least about 1 nM or less. Due to sequence identity between homologous proteins in different species, specific binding may include antibodies that recognize proteins in more than one species (e.g., human or mouse proteins). Similarly, due to homology within certain regions of the polypeptide sequences of different proteins, specific binding may include antibodies that recognize more than one protein. It should be understood that in some embodiments, an antibody or binding portion that specifically binds to a first target may or may not specifically bind to a second target. Therefore, "specific binding" does not necessarily require (although it may include) exclusive binding, i.e., binding to a single target. Therefore, in some embodiments, an antibody may specifically bind to more than one target. In some embodiments, the same antigen-binding site on an antibody may bind to multiple targets. For example, in some cases, an antibody may contain two identical antigen-binding sites, each specifically binding to the same epitope on two or more proteins. In some alternative embodiments, the antibody may be multispecific and contain at least two antigen-binding sites with different specificities. By way of non-limiting example, a bispecific antibody may contain an antigen-binding site that recognizes an epitope on one protein (e.g., human CD3) and further contain a second, different antigen-binding site that recognizes a different epitope on a second protein. Generally, but not necessarily, references to binding refer to specific binding.

Antibodies that specifically bind to antigens can bind to other peptides or polypeptides with lower affinity, as determined by assays known in the art (e.g., immunoassays, BIACORE®, or other assays). Preferably, antibodies that specifically bind to antigens do not cross-react with other proteins.

When used in relation to the interaction between antibodies and proteins or peptides, the terms “non-specific binding” or “background binding” refer to interactions that do not depend on the presence of a specific structure (i.e., antibody binding is typically protein rather than a specific structure, such as an epitope).

As used herein, the terms "k _on " or " _ka " refer to the rate constant relating to the association of antibody with antigen. Specifically, the rate constants (k _on / _ka and k _off /k _d ) and equilibrium dissociation constants are measured using intact antibodies (i.e., bivalent) and monomeric CD3 proteins.

As used in this article, the terms "k _off " or "k _d " refer to the rate constant of antibody dissociation from the antibody/antigen complex.

As used in this article, the term " _KD " refers to the equilibrium dissociation constant of antibody-antigen interactions.

As used herein, the term "binding affinity" generally refers to the total strength of the non-covalent interaction between a single binding site of a molecule (e.g., an antibody) and its binding partner (e.g., an antigen). Unless otherwise indicated, as used herein, "binding affinity" refers to the intrinsic binding affinity that reflects a 1:1 interaction between members of a binding pair (e.g., antibody and antigen). The affinity of molecule X for its partner Y is generally represented by the dissociation constant (KD). The affinity of molecule X for its partner Y is generally represented by the dissociation constant (Kd). For example, Kd can be about 200 nM, 150 nM, 100 nM, 60 nM, 50 nM, 40 nM, 30 nM, 20 nM, 10 nM, 8 nM, 6 nM, 4 nM, 2 nM, 1 nM, or stronger. Affinity can be measured by methods commonly known in the art, including those described herein. Low-affinity antibodies typically bind to antigens slowly and tend to dissociate easily, while high-affinity antibodies typically bind to antigens quickly and tend to remain bound for a longer period. Various methods for measuring binding affinity are known in the art, any of which may be used for the purposes of this invention. Specifically, the term "binding affinity" is intended to refer to the rate of dissociation of a particular antigen-antibody interaction. _KD is the ratio of the rate of dissociation (also called the "off-rate" (k _-off )) to the rate of association (k-on) (k- _on ). Therefore, _KD equals k _-off / k _-on and is expressed as a molar concentration (M). Thus, the smaller the _KD , the stronger the binding affinity. Therefore, a _KD of 1 μM indicates weak binding affinity compared to _{a KD} of 1 nM. The _KD value of an antibody can be determined using methods well established in the art. One method for determining the _KD of an antibody is by using surface plasmon resonance (SPR), typically employing a biosensor system (such as the BIACORE™ system). BIACORE™ kinetic analysis involves analyzing the binding and dissociation of antigens from chips with immobilized molecules (e.g., molecules containing epitope-binding domains) on their surfaces. Another method for determining the _KD of antibodies is through biolayer interferometry, typically using ^OCTET® technology (Octet QKe system, ForteBio).

Unless otherwise specified, with respect to the antibodies (such as antibodies, fragments or derivatives thereof) of the present invention, "bioactive", "bioactive" and "biocharacteristic" mean having the ability to bind biomolecules.

As used herein, the terms "nucleic acid" and "nucleotide sequence" include DNA molecules (e.g., cDNA or genomic DNA), RNA molecules (e.g., mRNA), combinations or hybrid DNA/RNA molecules, and analogs of DNA or RNA molecules. Such analogs can be generated using, for example, nucleotide analogs, including but not limited to inosine or triphenylmethylated bases. Such analogs may also comprise DNA or RNA molecules containing a modified backbone that imparts beneficial properties to the molecule, such as, for example, nuclease resistance or increased transmembrane translucency. Nucleic acid or nucleotide sequences can be single-stranded, double-stranded, contain both single-stranded and double-stranded portions, and may contain triple-stranded portions, but are preferably double-stranded DNA.

The present invention also includes a polynucleotide encoding the antibody of the present invention (including the polypeptide and binding region of the antibody). The polynucleotide encoding the molecule of the present invention can be obtained, and the nucleotide sequence of the polynucleotide can be determined by any method known in the art.

The polynucleotide encoding the antibody of the present invention may include: a variant-only coding sequence; a variant-coding sequence and additional coding sequences (such as functional polypeptides or signal sequences or secretion sequences or preprotein sequences); and antibody-coding and non-coding sequences (such as introns or non-coding sequences at the 5' and/or 3' of the antibody-coding sequence). The term "polynucleotide encoding antibody" encompasses polynucleotides including additional coding sequences of variants, and also encompasses polynucleotides including additional coding and/or non-coding sequences. It is known in the art that polynucleotide sequences optimized for specific host cells/expression systems can be readily obtained from the amino acid sequence of the desired protein (see ^GENEART® AG, Regensburg, Germany).

The antibodies of the present invention may have additional conserved or non-essential amino acid substitutions that do not substantially affect the function of the polypeptide. Whether a particular substitution will be tolerated (i.e. will not adversely affect the desired biological properties (such as binding activity)) can be determined as described in Bowie, JU et al., Science 247: 1306-1310, 1990 or Padlan et al., FASEBJ. 9: 133-139, 1995.

As used herein, "conservative amino acid substitution" is a substitution in which an amino acid residue is replaced by an amino acid residue having a similar side chain. Families of amino acid residues having similar side chains are defined in the art. These families include amino acids having the following side chains: basic side chains (e.g., lysine, arginine, histidine), acidic side chains (e.g., aspartic acid, glutamic acid), uncharged polar side chains (e.g., asparagine, glutamine, serine, threonine, tyrosine, cysteine), nonpolar side chains (e.g., glycine, alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine, tryptophan), β-branched side chains (e.g., threonine, valine, isoleucine), and aromatic side chains (e.g., tyrosine, phenylalanine, tryptophan, histidine).

As used herein, "non-essential" amino acid residues are residues that may differ from the wild-type sequence of a binding agent (e.g., an antibody) and do not eliminate or substantially alter biological activity, while "essential" amino acid residues cause this change. In antibodies, essential amino acid residues can be specificity-determining residues (SDRs).

As used herein, the term "isolated" refers to material removed from its original environment (e.g., the natural environment if it is naturally occurring). For example, naturally occurring polynucleotides or polypeptides present in living animals are not isolated, but the same polynucleotides or polypeptides isolated from some or all of the coexisting material in a natural system are isolated. This polynucleotide may be part of a carrier, and/or this polynucleotide or polypeptide may be part of a composition (e.g., a mixture, solution, or suspension) or contain isolated cells or cultured cells containing the polynucleotide or polypeptide, and is still isolated because the carrier or composition is not part of its natural environment. For example, a polypeptide synthesized chemically or synthesized in a cellular system different from the cell of natural origin of the polypeptide will be "isolated" from its natural associated components. Protein purification techniques well known in the art can also be used to isolate proteins to make them substantially free of natural associated components.

"Isolated" or "purified" antibodies are substantially free of cellular material or other contaminating proteins from cell or tissue sources or culture media from which the protein originates, or are substantially free of chemical precursors or other chemicals during chemical synthesis. The phrase "substantially free of cellular material" includes antibody formulations in which the polypeptide/protein is isolated from cellular components of the cells from which the polypeptide/protein is produced, either isolated or recombinantly. Therefore, antibodies substantially free of cellular material include antibody formulations having less than about 50%, 40%, 30%, 20%, 10%, 5%, 2.5%, or 1% (by dry weight) of contaminating proteins. When antibodies are produced by recombinant synthesis, they are also preferably substantially free of culture media, i.e., the culture medium constitutes less than about 50%, 40%, 30%, 20%, 10%, 5%, 2.5%, or 1% of the volume of the protein formulation. When antibodies are produced by chemical synthesis, they are preferably substantially free of chemical precursors or other chemicals and reagents, i.e., the target antibody is isolated from chemical precursors or other chemicals involved in protein synthesis. Therefore, this antibody preparation contains less than about 50%, 40%, 30%, 20%, 10%, 5%, or 1% (by dry weight) of a chemical precursor or compound in addition to the target antibody. In a preferred embodiment of the invention, the antibody is isolated or purified. For the purposes of this invention, an antibody that has been isolated or removed from at least one component of an organism or from tissues or cells in which the antibody is naturally present or produced is called an "isolated antibody".

As used herein, the term “recycling” refers to, for example, a process that uses protein purification techniques well known in the art to separate chemical substances (such as peptides) to make them substantially free of their natural associated components.

As used herein, the term "replicon" refers to any genetic element, such as a plasmid, chromosome, or virus, that acts as an autonomous polynucleotide replication unit within a cell.

As used herein, the term "operably ligated" refers to a situation where the relationship between said elements allows them to function in the intended manner. For example, "operably ligated" to a control sequence of a coding sequence in a manner that enables expression of the coding sequence under suitable or compatible conditions with that control sequence. Generally, "operably ligated" means that the ligated DNA sequences are adjacent, and in the case of a secretory leader sequence, adjacent and in the reading phase. However, enhancers do not require adjacency. Ligation is accomplished by ligating at a convenient restriction site. If such a site is not present, synthetic oligonucleotide adaptors or linkers are used according to conventional practice.

As used herein, “vector” means a construct capable of being delivered in a host cell and preferably expressing one or more target genes or sequences. Examples of vectors include, but are not limited to, viral vectors, naked DNA or RNA expression vectors, plasmids, granules or phage vectors, DNA or RNA expression vectors associated with cationic condensers, DNA or RNA expression vectors encapsulated in liposomes, and certain eukaryotic cells, such as production cells.

As used herein, the term "expression control sequence" or "control sequence" refers to a polynucleotide sequence necessary for the expression of the coding sequence it is linked to. The nature of this control sequence varies depending on the host organism. For example, in prokaryotes, such control sequences typically include a promoter, a ribosome binding site, a terminator, and (in some cases) an enhancer. Therefore, the term "control sequence" is intended to include at least all elements necessary for expression, but may also include additional elements that are advantageous for its presence, such as a leader sequence.

"Host cell" includes an individual cell or cell culture that may be or has been a vector recipient, the vector being used to incorporate a polynucleotide insert. Host cells include progeny of a single host cell, and the progeny may not be identical to the original parent cell (morphologically or in terms of genomic DNA complementation) due to natural, accidental, or intentional mutations. Host cells include cells transfected in vivo with the polynucleotides of the present invention.

As used herein, "mammalian cell" includes references to cells derived from mammals, including humans, rats, mice, guinea pigs, chimpanzees, or macaques. These cells may be cultured in vivo or in vitro.

As used herein, the term "purified product" refers to a product formulation that has been isolated from the cellular components to which the product is normally associated and/or from other cell types that may be present in the target sample.

As used herein, “substantially pure” means a material that is at least 50% pure (i.e., free of contaminants), more preferably at least 90% pure, more preferably at least 95% pure, even more preferably at least 98% pure, and most preferably at least 99% pure.

As used herein, the terms “treat,” “treating,” or “treatment” are methods for obtaining beneficial or desired clinical outcomes. For the purposes of this invention, treatment is defined as administering a CD3 antibody molecule (e.g., a monoclonal antibody, bispecific antibody) to a subject (e.g., a patient), or administering (e.g., by application) an isolated tissue or cell from the subject and returning it to said subject. The CD3 antibody molecule may be administered alone or in combination with one or more agents. Treatment can be curative, healing, alleviating, resolving, altering, remedying, improving, reducing, improving, or affecting a condition, its symptoms, or a predisposition to the condition (e.g., cancer).

As used herein, the term “subject” is intended to include any animal (e.g., a mammal), including but not limited to humans, non-human primates, rodents, etc., that is a recipient of a particular treatment. For example, a subject may be a patient with cancer (e.g., a human patient or a veterinary patient). Generally, the terms “subject,” “individual,” and “patient” are used interchangeably in this document when referring to human subjects.

Unless otherwise stated, the term "non-human animal" in this invention includes all non-human vertebrates, such as non-human mammals and non-mammals, such as non-human primates, sheep, dogs, cows, chickens, amphibians, reptiles, mice, rats, rabbits, or goats, etc.

As used herein, the term “pharmaceutical acceptable” means a product or compound that has been approved (or permitted to be approved) by a federal or state government regulatory agency or is listed in the United States Pharmacopeia or other recognized pharmacopoeia for use in animals (including humans).

As used herein, the terms "pharmaceutically acceptable excipient, carrier, or adjuvant" or "acceptable pharmaceutical carrier" mean an excipient, carrier, or adjuvant that can be administered to a subject together with at least one antibody of this disclosure without destroying the activity of the antibody. When administered together with an antibody at a dose sufficient to deliver a therapeutic effect, the excipient, carrier, or adjuvant shall be non-toxic.

As used herein, the term “improvement” means a reduction or improvement of one or more symptoms compared to the absence of the antibody molecules of the present invention. “Improvement” also includes a reduction or decrease in the duration of symptoms.

As used herein, the terms “prevent,” “preventing,” and “prevention” refer to the prevention of recurrence or onset of one or more symptoms of a disease in a subject by means of the administration of a prophylactic or therapeutic agent.

As used herein, “effective amount,” “therapeutic effective amount,” “therapeutic adequate amount,” or “effective dose” means the amount of a therapeutic agent (e.g., a CD3 antibody, alone or as part of a multispecific antibody (e.g., bispecific)) administered to a subject in a single or multiple doses that is effective or sufficient to prevent, heal, improve, treat, or manage a disease, condition, or side effect, or to reduce the rate of progression of a disease or condition, or to prolong healing, alleviate, reduce, or improve the condition of a subject with a condition as described herein. The term also includes, within its scope, amounts that effectively enhance normal physiological function. An effective amount can be considered in the context of administration of one or more therapeutic agents, and a single agent may be considered to have been administered in an effective amount if, in combination with one or more other agents, the desired outcome can be achieved.

As used herein, effective agents can target effector T cells to induce T cell-mediated cytotoxicity in cells associated with diseases such as cancer, autoimmune diseases, or infectious diseases. The therapeutic agents of the present invention induce a T cell-mediated immune response by modulating the CD3 antigen, specifically CD3, and more specifically CD3ε, on the surface of T cells.

Regarding cancer, the therapeutically effective dose refers to the amount of a therapeutic agent that inhibits the growth of a tumor or cancer by slowing, interrupting, stopping, or terminating its growth and/or metastasis.

Efficacy is a measure of the activity of a therapeutic agent, expressed as the amount required to produce an effect of a given strength. A highly potent agent elicits a greater response at low concentrations compared to a less potent agent that elicits a smaller response at low concentrations. Efficacy varies with affinity and potency. Potency refers to the ability of a therapeutic agent to produce a biological response upon binding to a target ligand and the quantitative magnitude of that response. As used herein, the term “half-maximum effective concentration ( _EC50 )” refers to the concentration of a therapeutic agent that elicits a response midway between baseline and maximum after a specified exposure time. Therapeutic agents can cause inhibition or stimulation. The _EC50 value is commonly used and is referred to herein as a measure of efficacy.

As used herein, “combination therapy” or “in combination with” refers to the use of more than one prophylactic and/or therapeutic agent. The use of the terms “combination therapy” or “combination” does not limit the order in which prophylactic and/or therapeutic agents are administered to a subject with the condition. In other words, combination therapy can be administered by treating with the therapeutic agents separately, sequentially, or simultaneously. In the case of “sequential administration,” the first administered agent may exert some physiological effect on the subject upon administration of the second agent or upon becoming active in the subject.

As used herein with respect to the administration of prophylactic and/or therapeutic agents, the term "simultaneous administration" means administration of the agents such that the individual agents are simultaneously present in the subject. Simultaneous administration can be achieved by molecules in a single composition or in separate compositions administered at the same or similar times. Sequential administration can be performed in any order as needed.

The term “differentiation cluster 3” or “CD3” refers to a polyprotein complex (historically known as the T3 complex) composed of four distinct polypeptide chains: epsilon (ε), gamma (γ), delta (δ), and zeta (ζ), which assemble and function as three pairs of dimers (εγ, εδ, ζζ). The CD3 complex acts as a T-cell co-receptor, non-covalently associating with the T-cell receptor (TCR) (Smith-Garvin et al. 2009) and generating activation signals in T lymphocytes. The CD3 protein complex is a defining characteristic of the T-cell lineage, and therefore CD3 antibodies can be effectively used as T-cell markers (Chetty and Gatter 1994). It is well known that CD3 antibodies induce the generation of cytotoxic T cells by activating endogenous lymphotropic mediators and can selectively kill tumor targets (Yun et al., Cancer Research, 49: 4770-4774 (1989)).

T cells play a central role in cell-mediated immunity in humans and animals. The recognition and binding of specific antigens are mediated by TCRs expressed on the surface of T cells. T cell TCRs can interact with immunogenic peptides (epitopes) that bind major histocompatibility complex (MHC) molecules and are presented on the surface of target cells. Specific binding of the TCR triggers a signaling cascade within the T cell, leading to proliferation and differentiation into mature effector T cells. More specifically, T cells express TCR complexes capable of inducing antigen-specific immune responses (Smith-Garvin et al., 2009). Antigens are peptides expressed by tumor cells and virus-infected cells capable of stimulating immune responses. Intracellularly expressed antigens bind to class I major histocompatibility (MHC) molecules and are transported to the surface, where they are exposed to T cells. If the binding affinity of the TCR to the class I MHC complexed with the antigen is sufficient, the formation of an immune synapse is initiated. Signaling via the immune synapse is mediated by CD3 co-receptors that form ε/δ, δ/γ, and ζ/ζ dimers. These dimers are associated with the TCR and generate activation signals in T lymphocytes. This signaling cascade guides T cell-mediated killing of antigen-expressing cells. Cytotoxicity is mediated by the release of granzyme B and perforin from T cells and their translocation to target cells.

As used herein, “activated T-cell antigen” refers to an antigenic determinant expressed on the surface of T lymphocytes, specifically cytotoxic T lymphocytes, which can induce T-cell activation upon interaction with antigen-binding molecules. Specifically, the interaction between antigen-binding molecules and activated T-cell antigen can induce T-cell activation by triggering a signaling cascade of the T-cell receptor complex. In one specific embodiment, the activated T-cell antigen is CD3, specifically the ε subunit of CD3 (for human sequences, see UniProt number P07766 (version 130), NCBI RefSeq number NP_000724.1, SEQ ID NO: 66). Typically, naturally occurring allele variants have an amino acid sequence that is at least 95%, 97%, or 99% identical to the protein described in Genbank accession number BAB71849.1.

As used herein, "T cell activation" refers to one or more cellular responses selected from the following: proliferation, differentiation, cytokine secretion, release of cytotoxic effector molecules, cytotoxic activity, and expression of activation markers by T lymphocytes, specifically cytotoxic T lymphocytes. The T cell activation bispecific antigen-binding molecule of the present invention can induce T cell activation. Assays suitable for measuring T cell activation are known in the art described herein.

As used herein, “antibody that binds CD3,” “antibody that recognizes CD3,” “anti-CD3 antibody,” or “CD3 antibody” includes an antibody that specifically recognizes a single CD3 subunit (e.g., ε, δ, γ, or ζ) and its antigen-binding fragment, and an antibody that specifically recognizes a dimer complex of two CD3 subunits (e.g., γ/ε, δ/ε, and ζ/ζ CD3 dimers) and its antigen-binding fragment. Human CD3ε is indicated in Genbank accession number NM_000733. The antibodies of the present invention can bind soluble CD3 and/or CD3 expressed on the cell surface. Soluble CD3 includes native CD3 protein and recombinant CD3 protein variants (such as, for example, monomeric and dimer CD3 constructs) lacking a transmembrane domain or otherwise not associated with the cell membrane.

Antibodies targeting CD3 can generate activation signals in T lymphocytes. Other T cell activation ligands, including but not limited to CD28, CD134, CD137, and CD27, can also be used.

This invention includes a single-arm antibody that binds to CD3. As used herein, "single-arm antibody" means an antigen-binding molecule comprising a single antibody heavy chain and a single antibody light chain. The single-arm antibody of this invention may comprise any of the VL/VH or CDR amino acid sequences as described in Tables 1, 2 and 3 herein.

The CD3 antibodies disclosed herein may contain one or more amino acid substitutions, insertions, and/or deletions in the framework and/or CDR regions of the heavy and light chain variable domains, compared to the corresponding germline sequences from which the antibodies are derived. Such mutations can be readily identified by comparing the amino acid sequences disclosed herein with germline sequences available, for example, from public antibody sequence databases. This invention includes antibodies derived from any of the amino acid sequences disclosed herein and their antigen-binding fragments, wherein one or more amino acids in one or more framework and/or CDR regions are mutated to corresponding residues of the germline sequence from which the antibody is derived, or mutated to corresponding residues of another human germline sequence, or mutated to conserved amino acid substitutions of the corresponding germline residues (such sequence changes are collectively referred to herein as “germline mutations”). Those skilled in the art can readily generate a variety of antibody and antigen-binding fragments containing one or more individual germline mutations or combinations thereof, starting with the heavy and light chain variable region sequences disclosed herein. In some embodiments, all framework and/or CDR residues within the VH and/or VL domains are mutated back to residues found in the original germline sequence from which the antibody is derived. In some embodiments, only certain residues are mutated back to the original germline sequence, such as mutated residues found only in the first 8 amino acids of FR1 or the last 8 amino acids of FR4, or mutated residues found only in CDR1, CDR2, or CDR3. In some further embodiments, one or more of the framework and/or CDR residues are mutated to corresponding residues in a different germline sequence (i.e., a germline sequence different from the germline sequence from which the antibody originally originated).

Furthermore, the antibodies of the present invention may contain any combination of two or more germline mutations within the framework and/or CDR region, for example, wherein certain individual residues are mutated to corresponding residues of a specific germline sequence, while certain other residues different from the original germline sequence are maintained or mutated to corresponding residues of a different germline sequence. Once obtained, one or more desired properties of antibodies containing one or more germline mutations can be readily tested, such as improved binding specificity, increased binding affinity, improved or enhanced antagonistic or agonistic biological properties (as applicable), reduced immunogenicity, etc. Antibodies obtained in this general manner are covered within the present invention.

The present invention also includes CD3 antibodies comprising variants of any of the disclosed VH, VL, and/or CDR amino acid sequences having one or more conserved substitutions. For example, the present invention includes CD3 antibodies having VH, VL, and/or CDR amino acid sequences having, for example, 10 or fewer, 8 or fewer, 6 or fewer, 4 or fewer, or other conserved amino acid substitutions relative to any of the VH, VL, and/or CDR amino acid sequences described in Table 1 herein.

As used herein, the term "complex" or "composite" refers to the association of two or more molecules that interact with each other via bonds and/or forces that are not peptide bonds (e.g., van der Waals forces, hydrophobic forces, hydrophilic forces). In one embodiment, the complex is heteropolymeric. It should be understood that the terms "protein complex" or "peptide complex" as used herein include complexes having non-protein entities (e.g., including but not limited to chemical molecules, such as toxins or detection agents) conjugated to proteins in the protein complex.

As used herein, "CD3 antibody" includes monovalent antibodies with single specificity, as well as "bispecific antibody," "two-specific antibody," "three-specific antibody," "bifunctional antibody," "heteropolymer," "heteropolymer complex," "bispecific heterodimeric biantibody," "heteropolypeptide," or bispecific heterodimeric IgG. In some embodiments of the present invention, the CD3 antibody of the present invention is a human or humanized antibody.

As used herein, a "bispecific antibody" is a molecule comprising at least a first polypeptide and a second polypeptide, wherein the second polypeptide differs from the first polypeptide in its amino acid sequence by at least one amino acid residue. In some cases, a bispecific antibody is an artificial hybrid antibody having two distinct heavy chain regions and light chain regions. Preferably, a bispecific antibody has binding specificity to at least two different ligands, antigens, or binding sites. Therefore, a bispecific antibody can bind to two different antigens simultaneously. The two antigen-binding sites of a bispecific antibody can reside on two different epitopes on the same or different protein targets (e.g., tumor targets).

As used herein, the first and second polypeptide chains of a “bispecific antibody” contain at least one antibody VL and an antibody VH region or a fragment thereof, wherein both antibody-binding domains are contained within a single polypeptide chain and wherein the VL and VH regions in each polypeptide chain are derived from different antibodies.

Bispecific antibodies can be derived from full-length antibodies or antibody fragments (e.g., F(ab')2 bispecific antibodies). Bispecific antibodies are more fully described, for example, in EP404,097; WO93/11161; and Hollinger et al., Proc. Natl. Acad. Sci. USA 90:6444-6448, 1993. A bispecific antibody is a heterodimer of two “crossed” sFv fragments, in which the VH and VL regions of the two antibodies are located on different polypeptide chains.

As used herein, the terms “link,” “fusion,” “fusion product,” “covalent bonding,” “covalent coupling,” and “gene fusion” are used interchangeably. These terms refer to the linking of two or more elements or components together by any means, including chemical conjugation or recombination. As used herein, the term “covalent bonding” means that the specified parts are directly covalently bonded to each other or indirectly covalently bonded to each other via one or more insert parts (such as linking peptides or parts). In a preferred embodiment, the parts are covalently fused. One type of covalent bonding is the peptide bond. Methods of chemical conjugation (e.g., using heterobifunctional crosslinking agents) are known in the art. The fusion part can also be a gene fusion. As used herein, the terms “gene fusion,” “gene link,” or “gene fusion” refer to the expression of a gene via a single polynucleotide molecule encoding two or more proteins, polypeptides, or fragments, which are collinearly, covalently linked, or connected via their individual peptide backbones. This genetic fusion results in the expression of a single adjacent gene sequence. Preferred gene fusions are in-frame fusions, meaning two or more open reading frames (ORFs) fuse in a manner that maintains the correct reading frame of the original ORF to form a continuously longer ORF. Therefore, the resulting recombinant fusion protein is a single polypeptide containing two or more protein segments corresponding to the polypeptides encoded by the original ORFs (segments that would not normally bind in this way naturally). In this case, the single polypeptide is cleaved during treatment to produce a dimer molecule containing two polypeptide chains.

As used herein, the terms "linked" or "links" refer to a direct peptide bond between first and second amino acid sequences, or a bond involving a third amino acid sequence peptide bonded to and between the first and second amino acid sequences. For example, a linker peptide is bonded to the C-terminus of one amino acid sequence and the N-terminus of another amino acid sequence.

As used herein, the term "linker" refers to an amino acid sequence of two or more amino acids in length. Linkers can consist of neutral polar or nonpolar amino acids. The length of a linker can be, for example, from 2 to 100 amino acids, such as between 2 and 50 amino acids, such as 3, 5, 10, 15, 20, 25, 30, 35, 40, 45, or 50 amino acids. Linkers can be "cleavable," for example, by self-cleavage, enzymatic cleavage, or chemical cleavage. Cleavage sites in amino acid sequences, and enzymes and chemicals that cleave at such sites, are well known in the art and are also described herein.

As used herein, the term "disulfide bond" or "cysteine-cysteine disulfide bond" refers to a covalent interaction between two cysteine residues, wherein the sulfur atom of the cysteine is oxidized to form a disulfide bond. The average bond energy of a disulfide bond is approximately 60 kcal/mol, compared to 1-2 kcal/mol for hydrogen bonds. In the context of this invention, the cysteine residues forming the disulfide bond are located within the framework region of a single-chain antibody and serve to stabilize the antibody's conformation. Cysteine residues can be introduced, for example, through site-directed mutagenesis, to generate stable disulfide bonds intramolecularly.

The term "the name of the pestle in the mortar" is similar to the term "protrusion and cavity" and they can be used interchangeably.

A "protrusion" or "pole" refers to at least one of the following amino acid side chains: extending from the interface of the first polypeptide and thus able to be positioned in a compensating cavity in a neighboring interface (i.e., the interface of the second polypeptide) to stabilize the heterodimer, thereby, for example, making heterodimer formation superior to homodimer formation. The protrusion may be present in the original interface or may be introduced synthetically (e.g., by altering the nucleic acid encoding the interface). Typically, the nucleic acid encoding the interface of the first polypeptide is modified to encode the protrusion. To achieve this, the nucleic acid encoding at least one "original" amino acid residue in the interface of the first polypeptide is replaced with an "input" amino acid residue encoding at least one side chain with a volume larger than the original amino acid residue. The upper limit for the number of original residues replaced is the total number of residues in the interface of the first polypeptide. Certain input residues used to form the protrusion are typically naturally occurring amino acid residues and are preferably selected from arginine (R), phenylalanine (F), tyrosine (Y), and tryptophan (W).

The protrusions or cavities are "localized" within cavities or sockets, meaning that the spatial location and size of the protrusions and cavities at the interface between the first and second peptides, respectively, allow the protrusions to be located within the cavities without significantly disrupting the normal association of the first and second peptides at the interface. Since protrusions, such as phenylalanine (F), tyrosine (Y), and tryptophan (W), typically do not extend perpendicularly from the interface axis, the comparison of protrusions to their corresponding cavities relies on modeling the protrusion/cavity pairs based on three-dimensional structures, such as those obtained by X-ray crystallography or nuclear magnetic resonance (NMR). This can be achieved using techniques well-established in the art.

A “cavity” or “mortise” refers to at least one amino acid side chain that recesses from the interface of the second polypeptide and thus accommodates a corresponding protrusion on the adjacent interface of the first polypeptide. The cavity may be present in the original interface or may be introduced synthetically (e.g., by altering the nucleic acid encoding the interface). Typically, the nucleic acid encoding the interface of the second polypeptide is altered to encode the cavity. To achieve this, the nucleic acid encoding at least one “original” amino acid residue in the interface of the second polypeptide is replaced with DNA encoding at least one “input” amino acid residue that has a side chain volume smaller than the original amino acid residue. The upper limit for the number of original residues replaced is the total number of residues in the interface of the second polypeptide. Some of the input residues used to form the cavity are typically naturally occurring amino acid residues and are preferably selected from alanine (A), serine (S), threonine (T), and valine (V).

As used herein, “target antigen,” “target cell antigen,” “tumor antigen,” or “tumor-specific antigen” refers to antigenic determinants that are presented on the surface of target cells (e.g., cells in a tumor, such as cancer cells) or cells in the tumor stroma).

As used herein, the term “cancer” or “cancerous” refers to or describes a physiological condition in mammals characterized by uncontrolled cell growth, growths or tumors of abnormal, uncontrolled growth originating from cells. In some respects, cancer refers to a malignant primary tumor that has not metastasized and remains localized. In other respects, cancer refers to a malignant tumor that has invaded and destroyed adjacent body structures and spread to distant sites. In some respects, cancer is associated with a specific cancer antigen. Examples of cancer include, but are not limited to, cancers of the following sites: mouth and pharynx, digestive system (e.g., esophagus, stomach, small intestine, colon, rectum, liver, gallbladder, bile ducts, and pancreas), respiratory system (e.g., larynx, lungs, and bronchi, including non-small cell lung cancer), bone and joints (e.g., bone metastases), soft tissue, skin (e.g., melanoma), breast, reproductive system (e.g., cervix, uterus, ovary, vulva, vagina, prostate, and testes), urinary system (e.g., bladder, kidneys, renal pelvis, and ureters), eye and orbit, brain and nervous system (e.g., glioma), head and neck, or endocrine system (e.g., thyroid). Cancer can also be lymphoma (such as Hodgkin's disease and non-Hodgkin's lymphoma), multiple myeloma, or leukemia (such as acute lymphoblastic leukemia, chronic lymphoblastic leukemia, acute myeloid leukemia, chronic myeloid leukemia, etc.). In some implementations, cancers of the digestive system are cancers of the esophagus, stomach, small intestine, colon, rectum, liver, gallbladder, bile ducts, and pancreas.

Regarding cancer, treatment is useful in any one or more of the following: (a) treating, preventing, or improving one or more symptoms of a condition associated with malignant cells in a subject (e.g., gastrointestinal cancers such as colorectal cancer (CRC)); (b) inhibiting tumor growth or progression in a subject with malignant tumors expressing tumor antigens; (c) inhibiting metastasis of cancer (malignant) cells expressing tumor antigens in a subject with one or more malignant cells expressing tumor antigens; (d) inducing regression (e.g., long-term regression) of tumors expressing tumor antigens; (e) exerting cytotoxic activity in malignant cells expressing tumor antigens; (f) increasing progression-free survival in a subject with a tumor antigen-associated condition; (g) increasing overall survival in a subject with a tumor antigen-associated condition; (h) reducing the use of additional chemotherapy agents or cytotoxic agents in a subject with a tumor antigen-associated condition; (i) reducing tumor burden in a subject with a tumor antigen-associated condition; or (j) blocking the interaction of tumor antigens with other unidentified factors. While not wishing to be bound by theory, the treatment is believed to have the ability to inhibit, ablate, or kill cells or otherwise reduce cell-mediated conditions (such as cancer) in vitro or in vivo.

As used in this article, the terms “malignant cell” or “malignant disease” refer to an aggressive and/or metastatic tumor or tumor cell, i.e., cancer cells.

Although any materials and methods similar to or equivalent to those described herein may be used in the practice or testing of this invention, preferred materials and methods are now described.

Materials and methods

Various techniques for antibody production have been described, including conventional hybridoma methods for preparing monoclonal antibodies, recombinant techniques for preparing antibodies (including chimeric antibodies, such as humanized antibodies), antibody production in transgenic animals, and the recently described phage display technique for preparing "fully human" antibodies. These techniques are briefly described below.

Polyclonal antibodies against a target antigen are typically generated in animals through multiple subcutaneous (sc) or intraperitoneal (ip) injections of the antigen and adjuvant. Using bifunctional reagents or derivatizing agents, such as maleimide benzoyl sulfosuccinimide (conjugated via cysteine residues) or N-hydroxysuccinimide (conjugated via lysine residues), the antigen (or a fragment containing the target amino acid sequence) can be conjugated to a protein (which is immunogenic in the species to be immunized, such as serum albumin, bovine thyroglobulin, or soybean trypsin inhibitors). Animals are immunized against the immunogenic conjugate or derivative, and booster immunizations are performed several weeks later via subcutaneous injections at multiple sites. Seven to 14 days later, blood is drawn from the animals, and serum antibody titers are measured. Booster immunizations are continued until a titer plateau is reached. Preferably, booster immunizations are performed using conjugates of the same antigen (but conjugated to different proteins and/or via different cross-linking agents). Conjugates can also be prepared as protein fusions in recombinant cell cultures. In addition, agglomerating agents (such as alum) are used to enhance the immune response.

Monoclonal antibodies can be obtained from a substantially homogeneous population of antibodies using the hybridoma method first described by Kohler and Milstein, Nature 256:495, 1975, or can be prepared via a recombinant DNA method (Cabilly et al., U.S. Patent No. 4,816,567). In the hybridoma method, immunization of mice or other suitable host animals, as described above, induces the production or ability to produce lymphocytes that will specifically bind to proteins used for immunization. Alternatively, immunization of lymphocytes can be performed in vitro. The lymphocytes are then fused with myeloma cells using a suitable fusion agent (such as polyethylene glycol) to form hybridoma cells (Goding, Monoclonal Antibodies: Principles and Practice, pp. 59-103 (Academic Press, 1986). The hybridoma cells thus prepared are seeded in a suitable culture medium, preferably containing one or more substances that inhibit the growth or survival of unfused parental myeloma cells, and allowed to grow therein. Alternatively, the hybridoma cells can grow in vivo as ascites tumors in animals. The monoclonal antibodies secreted by the subclones are appropriately separated from the culture medium, ascites, or serum using routine immunoglobulin purification procedures such as, for example, protein A-Sepharose, hydroxyapatite chromatography, gel electrophoresis, dialysis, or affinity chromatography.

The expression of antibodies, antigen-binding fragments of antibodies, or any antibody construct can be carried out in appropriate prokaryotic or eukaryotic host cells (such as CHO cells, NSO cells, SP2/0 cells, HEK293 cells, COS cells, PER.C6 cells, yeast or E. coli cells), and the secreted antibodies can be recovered and harvested from said cells (cell culture supernatant, regulated cell culture supernatant, cell lysate, or clarified bulk material). Common methods for generating antibodies are well known in the art and described in, for example, the following review articles: Makrides, S.C., Protein Expr. Purif. 17:183-202, 1999; Geisse, S., et al., Protein Expr. Purif. 8:271-282, 1986; Kaufman, R.J., MoI. Biotechnol. 16:151-160, 2000; Werner, R.G., Drug Res. 48:870-880, 1998. In one specific embodiment, the cell culture is a mammalian cell culture, such as a Chinese hamster ovary (CHO) cell culture.

In one embodiment, the antibody may be recovered or isolated from a naturally occurring source. In some such embodiments, the isolated or recovered antibody may undergo additional purification steps using conventional chromatographic methods known in the art. Specifically, the purification methods are contemplated, but are not limited to, affinity chromatography (e.g., protein A affinity chromatography), ion exchange chromatography (e.g., anion exchange chromatography or cation exchange chromatography), hydrophobic interaction chromatography, hydroxyapatite chromatography, gel filtration chromatography, and/or dialysis. Of these, protein A chromatography is the preferred purification method. The matrix for affinity ligand linkage is most often agarose, but other matrices may be used.

Other techniques for antibody purification are also known in the art, such as fractionation on ion exchange columns, ethanol precipitation, reversed-phase high-performance chromatography, reversed-phase HPLC, chromatography on silica, chromatography on heparin Sepharose™, chromatography on anion or cation exchange resins (such as polyaspartic acid columns), chromatographic focusing, electrophoresis using SDS-PAGE, and ammonium sulfate precipitation. The list of purification methods above is merely illustrative in nature and is not intended to be limiting.

Alternatively, transgenic animals (e.g., mice) can be produced that, upon immunization, generate a complete human antibody library in the absence of endogenous immunoglobulin production. For example, homozygous deletion of the antibody heavy chain linker (J.sub.H) gene in chimeric and germline mutant mice has been described as resulting in complete suppression of endogenous antibody production. Transplantation of human germline immunoglobulin gene arrays into such germline mutant mice will lead to the production of human antibodies upon antigen challenge. (See, for example, Jakobovits et al., Proc. Natl. Acad. Sci. USA 90:2551-255;1993 and Jakobovits et al., Nature 362:255-258, 1993).

In some embodiments, the antibodies of the present invention may be humanized, retaining high affinity for the antigen and other beneficial biological properties. Methods for humanizing nonhuman antibodies are well known in the art. Humanization can be essentially performed following the approach of Winter and collaborators (Jones et al., Nature 321:522-525, 1986; Riechmann et al., Nature 332:323-327, 1988; Verhoeyen et al., Science 239:1534-1536, 1988) by replacing the corresponding sequence of a human antibody with a rodent CDR or CDR sequence. Thus, such humanized antibodies are chimeric antibodies (Cabilly, ibid.), in which essentially less than one fully human variable domain has been replaced by a corresponding sequence from a non-human species. In practice, humanized antibodies are typically human antibodies, with some CDR residues and possibly some FR residues replaced by residues from similar sites in rodent antibodies. Importantly, the antibody is humanized, retaining its high affinity for the antigen and other beneficial biological properties. To achieve this, humanized antibodies are prepared according to a preferred method by analyzing the parental sequence and various conceptual humanized products using a three-dimensional model of the parental and humanized sequences. Three-dimensional immunoglobulin models are familiar to those skilled in the art. A computer program can be used to illustrate and demonstrate the possible three-dimensional conformational structures of selected candidate immunoglobulin sequences. Observing these displays allows analysis of the possible roles of residues in the function of the candidate immunoglobulin sequence, i.e., analysis of residues affecting the ability of the candidate immunoglobulin to bind its antigen. In this way, FR residues can be selected and merged from common and input sequences to achieve desired antibody properties, such as increased affinity for the target antigen. For further details, see WO92/22653, published December 23, 1992.

Various phage display methods known in the art can also be used to generate the antibodies of the present invention. In phage display methods, functional antibody domains are displayed on the surface of phage particles carrying a polynucleotide sequence encoding them. In particular, this phage can be used to display antigen-binding domains (such as Fab and Fv or disulfide-stable Fv) expressed from a library or combinatorial antibody library (e.g., human or mouse). Phages expressing antigen-binding domains that bind to target antigens can be selected or identified using antigens (e.g., using labeled antigens or antigens that bind to or capture onto solid surfaces or beads). The phages used in these methods are typically filamentous phages, including fd and M13. The antigen-binding domains are expressed as proteins recombinantly fused to phage gene III or gene VIII proteins. Examples of phage display methods that can be used to prepare the immunoglobulins or fragments thereof of the present invention include those disclosed in Brinkmann et al., "Phage Display Of Disulfide-Stabilized FvFragments," J. Immunol. Methods, 182:41-50, 1995.

The functional properties of various IgG isoforms and their domains are well known in the art. The amino acid sequences of IgG1, IgG2, IgG3, and IgG4 are known in the art. The selection and/or combination of two or more domains from a particular IgG isoform in the methods used in this invention can be based on any known parameters of the parental isoform (including affinity for FcγR). For example, using regions or domains from IgG isoforms (e.g., IgG2 or IgG4) that exhibit limited or no binding to FcγRIIB can be particularly useful in cases where it is desired to engineer bispecific antibodies to maximize binding to activating receptors and minimize binding to inhibitory receptors. Similarly, using Fc chains or domains from IgG isoforms (e.g., IgG3) that are known to preferentially bind to C1q or FcγRIIIA can be combined with Fc amino acid modifications known in the art to enhance antibody-dependent cell-mediated cytotoxicity (ADCC) and/or complement-dependent cytotoxicity (CDC) to engineer bispecific antibodies such that effector functional activity (e.g., complement activation or ADCC) is maximized. In a similar manner, mutations can be made in the Fc chain or domain of the IgG isotype to minimize or eliminate the effector function of the Fc chain.

During the discovery process, antibody generation and characterization can elucidate information about the desired epitope. Based on this information, antibodies can be competitively screened for binding to the same epitope. This is achieved by conducting competitive and cross-competitive studies to find antibodies that compete or cross-compete with each other (e.g., antibodies that competitively bind to antigens). One approach is to identify the epitope that the antibody binds to, or “epitope localization.” Many methods known in the art exist for mapping and characterizing the location of epitopes on proteins, including resolving the crystal structure of antibody-antigen complexes, competitive assays, gene fragment expression assays, and synthetic peptide-based assays, as described, for example, in Chapter 11 of Harlow and Lane, Using Antibodies, a Laboratory Manual, Cold SpringHarbor Laboratory Press, Cold SpringHarbor, N.Y., 1999. In an additional instance, epitope mapping can be used to determine the sequence to which the antibody binds. Epitope mapping is commercially available from various sources, such as Pepscan Systems (Edelhertweg 15, 8219 PH Lelystad, Netherlands). In an additional instance, mutagenesis of the antigen-binding domain, domain exchange experiments, and alanine partitioning mutagenesis can be performed to identify residues required, sufficient, and/or essential for epitope binding. Binding affinity and the dissociation rate of the antigen-binding domain interaction can be determined by a competitive binding assay. An example of a competitive binding assay is a radioimmunoassay, which involves incubating a labeled antigen and detecting the molecule binding to the labeled antigen. The affinity and binding-dissociation rate of the molecules of the present invention for the antigen can be determined from saturation data using Scatchard analysis.

The affinity and binding properties of the antibodies of the present invention to antigens can first be determined using in vitro assays (biochemical or immunoassays) known in the art for antigen-binding domains, including but not limited to enzyme-linked immunosorbent assays (ELISA), surface plasmon resonance (SPR) assays, biolayer interferometry, or immunoprecipitation assays. The molecules of the present invention may have similar binding properties in in vivo models (such as those described and disclosed herein) to those in in vitro-based assays. However, the present invention does not exclude molecules of the present invention that do not exhibit the desired phenotype in in vitro-based assays but do exhibit the desired phenotype in vivo.

Once the nucleic acid sequence encoding the molecule of the present invention (i.e., the binding domain) has been obtained, a vector for generating the molecule can be produced using recombinant DNA technology, a technique well known in the art.

The polynucleotide encoding the antibody-binding domain of the present invention may include an expression control polynucleotide sequence operatively linked to the antibody-coding sequence, including naturally associated or heterologous promoter regions known in the art. The expression control sequence may be a eukaryotic promoter system in a vector capable of transforming or transfecting eukaryotic host cells, but a prokaryotic host control sequence may also be used. Once the vector has been incorporated into a suitable host cell line, the host cells are allowed to proliferate under conditions suitable for expressing the nucleotide sequence and are used, as needed, to collect and purify the antibody. Eukaryotic cell lines include CHO cell lines, various COS cell lines, HeLa cells, myeloma cell lines, transformed B cells, or human embryonic kidney cell lines.

In some embodiments, the DNA encoding the antibodies of the present invention is isolated and sequenced using conventional procedures, e.g., by using oligonucleotide probes capable of specifically binding to genes encoding the heavy and light chains of mouse antibodies. Hybridoma cells of the present invention serve as a preferred source of this DNA. Once isolated, the DNA can be placed in an expression vector, which is then transfected into host cells that otherwise do not produce immunoglobulin proteins (such as simian COS cells, CHO cells, or myeloma cells) to obtain monoclonal antibody synthesis in recombinant host cells. The DNA can also be modified, for example, by replacing the homologous mouse sequences with coding sequences for the constant domains of the human heavy and light chains (Morrison et al., Proc. Nat. Acad. Sci. 81:6851, 1984). In this manner, chimeric antibodies having the binding specificity of the anti-antigen monoclonal antibodies described herein are prepared.

CD3 antibody

In one aspect, the present invention provides a reagent that specifically binds to CD3 (e.g., the full-length human CD3 ε subunit (e.g., accession number NP_000724 or SEQ ID NO: 66)).

In some implementations, the CD3 antibody specifically binds to human CD3.

Exemplary CD3 antibodies of the present invention are listed in Tables 1, 2, and 3 herein. Table 1 shows the amino acid sequence identifiers of the VH and VL regions. In some embodiments, antibodies are provided having any of the partial light chain sequences listed in Table 1 and/or any of the partial heavy chain sequences listed in Table 1.

Table 1

In Table 1, the underlined sequences are CDR sequences based on Kabat, and the bold sequences are CDR sequences based on Chothia.

This invention provides the CDR portion (including Chothia, Kabat CDR, and CDR contact region) of an antibody against CD3. Methods and techniques for identifying CDRs within VH and VL amino acid sequences are well known in the art and can be used to identify CDRs within the specified VH and/or VL amino acid sequences disclosed herein. Exemplary conventions for identifying CDR boundaries include, for example, the Kabat definition, the Chothia definition, and the AbM definition. Generally, the Kabat definition is based on sequence variability, the Chothia definition is based on the location of structural loop regions, and the AbM definition is a compromise between the Kabat and Chothia methods. See, for example, Kabat, "Sequences of Proteins of Immunological Interest," National Institutes of Health, Bethesda, Md. (1991); Al-Lazikani et al., J. Mol. Biol. 273:927-948 (1997); and Martin et al., Proc. Natl. Acad. Sci. USA 86:9268-9272 (1989). Public databases may also be used to identify CDR sequences within the antibody. It should be understood that in some embodiments, the CDR may be a combination of Kabat and Chothia CDRs (also referred to as a "combined CR" or "extended CDR"). In some embodiments, the CDR is the Kabat CDR. In some further embodiments, the CDR is the Chothia CDR. In other words, in an implementation having more than one CDR, the CDR can be any of Kabat, Chothia, combined CDRs, or a combination thereof. Tables 2 and 3 provide examples of the CDR sequences provided herein.

In one embodiment, the present invention provides an antibody comprising the amino acid sequences listed in Table 1 or substantially similar sequences having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with them.

In some such embodiments, the present invention provides an antibody comprising a VL region comprising an amino acid sequence of any of the VL region amino acid sequences listed in Table 1 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with it.

In some other embodiments of this invention, the present invention provides an antibody comprising a VH region comprising an amino acid sequence of any of the VH region amino acid sequences listed in Table 1 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with it.

In one embodiment, an antibody that specifically binds to CD3 is provided, wherein the antibody comprises: (a) a heavy chain variable (VH) region comprising VH complementarity-determining region 1 (VH CDR1), VH complementarity-determining region 2 (VH CDR2), and VH complementarity-determining region 3 (VH CDR3) of the VH sequence shown in SEQ ID NO: 2, 4, 5, 7, 10, 12, or 35; and/or (b) a light chain variable (VL) region comprising VL complementarity-determining region 1 (VL CDR1), VL complementarity-determining region 2 (VL CDR2), and VL complementarity-determining region 3 (VL CDR3) of the VL sequence shown in SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87, or 89.

In one embodiment, the present invention provides an antibody comprising a set of six CDRs (i.e., VL CDR1, VL CDR2, VLCDR3, VH CDR1, VH CDR2, VH CDR3), wherein the CDRs are contained within the VL/VH amino acid sequence pairs as defined by any of the exemplary CD3 antibodies listed in Table 1.

In specific embodiments, the present invention provides an antibody comprising a set of amino acid sequences of VL CDR1, VL CDR2, VL CDR3, VHCDR1, VHCDR2, or VH CDR3, wherein the set of amino acid sequences is contained within a VL/VH amino acid sequence pair selected from the following: SEQ ID NO: 1 and 2 (e.g., 2B5-0001 (2B5)); SEQ ID NO: 3 and 2 (2B5-0002); SEQ ID NO: 3 and 4 (e.g., 2B5-0006 (2B5v6)); SEQ ID NO: 3 and 5 (e.g., 2B5-0009); SEQ ID NO: 6 and 7 (e.g., 2B5-0517); SEQ ID NO: 8 and 7 (e.g., 2B5-0522); SE QID NO: 6 and 4 (e.g., 2B5-0533); SEQ ID NO: 8 and 4 (e.g., 2B5-0538); SEQ ID NO: 9 and 10 (e.g., 2B5-0598); SEQ ID NO: 9 and 7 (e.g., 2B5-0623); SEQ ID NO: 11 and 12 (e.g., 2B5-0707); SEQ ID NO: 34 and 35 (e.g., 2B5-0003); SEQ ID NO: 85 and 2 (e.g., H2B4); SEQ ID NO: 9 and 4 (e.g., 2B5-1038); SEQ ID NO: 87 and 4 (e.g., 2B5-1039); and SEQ ID NO: 89 and 4 (e.g., 2B5-1040).

Table 2 shows the VH CDR1, VH CDR2, and VH CDR3 regions of an exemplary CD3 antibody.

Table 2

In one embodiment, the present invention provides an antibody comprising VH CDR1, said VH CDR1 comprising an amino acid sequence of any of the VHCDR1 amino acid sequences listed in Table 2 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In another embodiment, the present invention provides an antibody comprising VH CDR2, said VH CDR2 comprising an amino acid sequence of any of the VHCDR2 amino acid sequences listed in Table 2 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In yet another embodiment, the present invention provides an antibody comprising VH CDR3, said VH CDR3 comprising an amino acid sequence of any of the VHCDR3 amino acid sequences listed in Table 2 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

Table 3 shows exemplary CD3 antibodies VL CDR1, VL CDR2, and VL CDR3.

Table 3

In one embodiment, the present invention provides an antibody comprising VL CDR1, said VL CDR1 comprising an amino acid sequence of any of the VLCDR1 amino acid sequences listed in Table 3 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In another embodiment, the present invention provides an antibody comprising VL CDR2, said VL CDR2 comprising an amino acid sequence of any of the VLCDR2 amino acid sequences listed in Table 3 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In yet another embodiment, the present invention provides an antibody comprising VL CDR3, said VL CDR3 comprising an amino acid sequence of any of the VLCDR3 amino acid sequences listed in Table 3 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In some embodiments, an antibody that specifically binds to CD3 is provided, wherein the antibody comprises: (a) a heavy chain (HC) variable (VH) region comprising VH complementarity-determining region 1 (CDR1), VH complementarity-determining region 2 (VH CDR2), and VH complementarity-determining region 3 (VH CDR3) of VH having an amino acid sequence selected from SEQ ID NO: 2, 4, 5, 7, 10, 12, and 35; and/or (b) a light chain (LC) variable (VL) region comprising VL complementarity-determining region 1 (VL CDR1), VL complementarity-determining region 2 (VL CDR2), and VL complementarity-determining region 3 (VLCDR3) of VL having an amino acid sequence selected from SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87, and 89.

In a specific embodiment, the antibody comprises: (a) a VH region comprising (i) a VH CDR1 comprising the sequence of SEQ ID NO: 13, 14 or 15; (ii) a VHCDR2 comprising the sequence of SEQ ID NO: 16, 17, 19, 20, 21, 22, 23 or 24; and (iii) a VH CDR3 comprising the sequence of SEQ ID NO: 18 or 25; and/or (b) a VL region comprising (i) a VL CDR1 comprising the sequence of SEQ ID NO: 26, 29, 30, 32, 91 or 92; (ii) a VL CDR2 comprising the sequence of SEQ ID NO: 27 or 31; and (iii) a VL CDR3 comprising the sequence of SEQ ID NO: 28 or 33.

In some embodiments, the present invention provides antibodies that bind to CD3 and compete with antibodies as described herein, including 2B5-0001 (2B5), 2B5-0002, 2B5-0006 (2B5v6), 2B5-0009, 2B5-0517, 2B5-0522, 2B5-0533, 2B5-0538, 2B5-0598, 2B5-0623, 2B5-0707, 2B5-0003, 2B5-1038, 2B5-1039, and 2B5-1040.

In some embodiments, the present invention also provides the CDR portion of a CD3 antibody based on a CDR contact region. A CDR contact region is an antibody region that imparts specificity to an antigen to the antibody. Typically, the CDR contact region comprises residue sites within the CDR and Vernier region, constrained to maintain a suitable loop structure for antibody binding to specific antigens. See, for example, Makabe et al., J. Biol. Chem., 283:1156-1166, 2007. Determination of the CDR contact region is entirely within the scope of the art.

Some of the CD3 antibodies described herein may be referred to by more than one name. For example, 2B5-0001 may be referred to as 2B5 or h2B5; and 2B5-0006 may be referred to as 2B5v6 or h2B5v6. Table 4 shows the forms of the antibodies and the CD3 antibody regions present in the antibodies disclosed herein.

Table 4

sequence	describe	form	CD3 antibody region
				2B5-0001 (2B5)	CD3 antibody	IgG	2B5-0001
2B5-0002	CD3 antibody	IgG	2B5-0002
				2B5-0003	CD3 antibody	IgG	2B5-0003
2B5-0006 (2B5v6)	CD3 antibody	IgG	2B5-0006
				2B5-0009	CD3 antibody	IgG	2B5-0009
2B5-0517	CD3 antibody	IgG	2B5-0517
				2B5-0522	CD3 antibody	IgG	2B5-0522
2B5-0533	CD3 antibody	IgG	2B5-0533
				2B5-0538	CD3 antibody	IgG	2B5-0538
2B5-0598	CD3 antibody	IgG	2B5-0598
				2B5-0623	CD3 antibody	IgG	2B5-0623
2B5-0707	CD3 antibody	IgG	2B5-0707
				2B5-1038	CD3 antibody	IgG	2B5-1038
2B5-1039	CD3 antibody	IgG	2B5-1039
				2B5-1040	CD3 antibody	IgG	2B5-1040
H2B4	CD3 antibody	IgG	H2B4
				GUCY2C-H2B4	CD3-GUCY2c bispecific antibody	CD3 bispecific	H2B4
GUCY2C-2B5	CD3-GUCY2c bispecific antibody	CD3 bispecific	2B5-0001
				GUCY2C-2B5v6	CD3-GUCY2c bispecific antibody	CD3 bispecific	2B5-0006
xFLT3-2B5v6	CD3-FLT3 bispecific antibody	CD3 bispecific	2B5-0006

The binding affinity ( _KD ) of the antibody described herein for CD3 (such as human CD3ε (e.g., SEQ ID NO:40) can be from about 0.001 nM to about 500 nM. In one embodiment, the binding affinity is about any of the following: 6500 nM, 6000 nM, 5500 nM, 4500 nM, 4000 nM, 3500 nM, 3000 nM, 2500 nM, 2000 nM, 1500 nM, 1000 nM, 750 nM, 500 nM, 400 nM, 300 nM, 250 nM, 200 nM, 150 nM, 100 nM, 75 nM, 50 nM, 45 nM, 40 nM, 35 nM, 30 nM, 25 nM, 20 nM, 19 nM, 17 nM, 16 nM, 15 nM, 10 nM, 8 nM, 7.5 nM, 7 nM, 6.5 nM, 6 nM, 5.5 nM, 5 nM, 4 nM, 3 nM, 2 nM, 1 nM, 0.5 nM, 0.3 nM, 0.1 nM, 0.01 nM, 0.002 nM, or 0.001 nM. In some embodiments, the binding affinity is at least about any one of the following: 6500 nM, 6000 nM, 5500 nM, 5000 nM, 4000 nM, 3000 nM, 2000 nM, 1000 nM, 900 nM, 800 nM, 500 nM, 250 nM, 200 nM, 100 nM, 50 nM, 30 nM, 20 nM, 10 nM, 7.5 nM, 7 nM, 6.5 nM, 6 The effective values are nM, 5 nM, 4.5 nM, 4 nM, 3.5 nM, 3 nM, 2.5 nM, 2 nM, 1.5 nM, 1 nM, 0.5 nM, 0.1 nM, 0.05 nM, 0.01 nM, or lower. In one specific embodiment, the antibody has a _KD of about 80 to about 200 pM. In another specific embodiment, the antibody has a _KD of about 10 nm to 2000 nm.

In one aspect, the present invention provides an antibody or pharmaceutical composition for use in a method of modulating a T-cell-mediated immune response in a subject with such a need. In a specific embodiment, the present invention provides an antibody or pharmaceutical composition for use in a method of inhibiting the growth of tumor cells in a subject.

In another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, optionally wherein said cancer is selected from: breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer, bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and cancers of the digestive system.

In yet another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, wherein a T-cell-mediated immune response is modulated or wherein the growth of tumor cells is inhibited.

The present invention also provides nucleic acid molecules encoding CD3 antibodies or portions thereof. For example, the present invention provides nucleic acid molecules encoding any VH or VL amino acid sequence listed in Table 1 or substantially similar sequences having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with them.

The present invention also provides nucleic acid molecules of any of the amino acid sequences of VH CDR1, VH CDR2 or VH CDR3 listed in Table 2 or substantially similar sequences having at least 90%, at least 95%, at least 98% or at least 99% sequence identity with them.

The present invention also provides nucleic acid molecules having a substantially similar sequence to any of the VL CDR1, VL CDR2 or VL CDR3 amino acid sequences listed in Table 3, or having at least 90%, at least 95%, at least 98% or at least 99% sequence identity with it.

The present invention also provides a nucleic acid molecule encoding HC VR, wherein the VH comprises a set of three CDRs (i.e., VH CDR1, VHCDR2, or VH CDR3), wherein the amino acid sequence sets of VH CDR1, VH CDR2, and VH CDR3 are as defined by any of the exemplary CD3 antibodies listed in Table 2.

The present invention also provides a nucleic acid molecule encoding LC VR, wherein the VL comprises a set of three CDRs (i.e., VL CDR1, VLCDR2, or VL CDR3), wherein the amino acid sequence sets of VL CDR1, VL CDR2, and VL CDR3 are as defined by any of the exemplary CD3 antibodies listed in Table 3.

The present invention also provides nucleic acid molecules encoding both VH and VL, wherein the VH comprises an amino acid sequence of any VH amino acid sequence listed in Table 1, and wherein the VL comprises an amino acid sequence of any VL amino acid sequence listed in Table 1.

The present invention also provides recombinant expression vectors capable of expressing polypeptides containing heavy or light chain variable regions of CD3 antibodies. For example, the present invention includes recombinant expression vectors comprising any of the nucleic acid molecules mentioned above (i.e., nucleic acid molecules encoding any of the VH and VL shown in Table 1 and/or CDR sequences as shown in Table 2 or 3). The scope of the present invention also includes host cells in which such vectors have been introduced, and methods for producing said antibodies or portions thereof by culturing said host cells under conditions allowing the production of antibodies or antibody fragments and recovering said antibodies and antibody fragments thus produced.

In some embodiments, the polynucleotide comprises a sequence encoding the heavy chain and/or light chain variable region of the following antibodies listed in Table 4: 2B5-0001 (2B5), 2B5-0002, 2B5-0006 (2B5v6), 2B5-0009, 2B5-0517, 2B5-0522, 2B5-0533, 2B5-0538, 2B5-0598, 2B5-0623, 2B5-0707, 2B5-0003, 2B5-1038, 2B5-1039, or 2B5-1040.

The sequence encoding the target antibody can be maintained in a vector within a host cell, and the host cell can then be amplified and frozen for future use. Vectors (including expression vectors) and host cells are further described herein.

Polynucleotides complementary to any such sequence are also covered by this invention. In one aspect, this invention provides a method for preparing any of the polynucleotides described herein. The polynucleotide may be single-stranded (coding or antisense) or double-stranded, and may be DNA (genomic, cDNA, or synthetic) or RNA molecules. RNA molecules include HnRNA molecules containing introns and corresponding one-to-one with DNA molecules; and mRNA molecules that do not contain introns. Additional coding or non-coding sequences may, but do not necessarily, be present in the polynucleotides of this invention, and the polynucleotides may, but do not necessarily, be linked to other molecules and/or supporting materials.

The polynucleotide may contain a native sequence (i.e., an endogenous sequence encoding an antibody or a portion thereof) or a variant thereof. Polynucleotide variants contain one or more substitutions, additions, deletions, and/or insertions such that the immunoreactivity of the encoded polypeptide is not reduced relative to the native immunoreactive molecule. The effect on the immunoreactivity of the encoded polypeptide can generally be evaluated as described herein. The variant preferably exhibits at least about 70% identity with the polynucleotide sequence encoding a native antibody or a portion thereof, more preferably at least about 80% identity, even more preferably at least about 90% identity, and most preferably at least about 95% identity.

The polynucleotides of this invention can be obtained using chemical synthesis, recombination methods, or PCR. Methods for the chemical synthesis of polynucleotides are well known in the art and need not be described in detail herein. Those skilled in the art can use the sequences provided herein and commercial DNA synthesizers to generate the desired DNA sequences.

To prepare polynucleotides using recombinant methods, a polynucleotide containing the desired sequence can be inserted into a suitable vector, which can then be introduced into a suitable host cell for replication and amplification, as discussed further herein. Polynucleotides can be inserted into host cells by any means known in the art. Cells are transformed by introducing exogenous polynucleotides through direct uptake, endocytosis, transfection, F-conjugation, or electroporation. Once introduced, the exogenous polynucleotide can be maintained within the cell as a non-integrating vector (such as a plasmid) or integrated into the host cell genome. The amplified polynucleotides can then be isolated from the host cell using methods well-known in the art. See, for example, Sambrook et al., 1989.

Alternatively, PCR allows for the replication of DNA sequences. PCR technology is well known in the art and is described in U.S. Patent Nos. 4,683,195, 4,800,159, 4,754,065, and 4,683,202, and PCR: The Polymerase Chain Reaction, edited by Mullis et al., Birkauswer Press, Boston, 1994.

RNA can be obtained by using DNA isolated in a suitable vector and inserted into a suitable host cell. Once the cell replicates and the DNA is transcribed into RNA, RNA can then be isolated using methods well known to those skilled in the art, such as Sambrook et al., 1989, as described above.

Suitable cloning vectors can be constructed according to standard techniques or selected from a large number of cloning vectors available in the art. While the chosen cloning vector can vary depending on the intended host cell, a useful cloning vector will generally be capable of self-replication, may have a single target for a specific restriction endonuclease, and/or may carry genes related to markers that can be used to select clones containing that vector. Suitable examples include plasmids and bacterial viruses such as pUC18, pUC19, Bluescript (e.g., pBS SK+) and its derivatives, mp18, mp19, pBR322, pMB9, ColE1, pCR1, RP4, phage DNA, and shuttle vectors such as pSA3 and pAT28. These and many other cloning vectors are available from commercial vendors such as BioRad, Strategene, and Invitrogen.

Expression vectors are typically reproducible polynucleotide constructs containing polynucleotides according to the invention. This implies that the expression vector must be replicable in the host cell as an episome or as an integral part of chromosomal DNA. Suitable expression vectors include, but are not limited to, plasmids, viral vectors including adenoviruses, adeno-associated viruses, retroviruses, kinases, and the expression vectors disclosed in PCT Publication No. WO 87/04462. Vector components may typically include, but are not limited to, one or more of the following: a signal sequence; an origin of replication; one or more marker genes; and suitable transcriptional control elements (such as promoters, enhancers, and terminators). For expression (i.e., translation), one or more translational control elements, such as ribosome binding sites, translation initiation sites, and stop codons, are also typically required.

Vectors containing the target polynucleotide can be introduced into host cells by any of a number of suitable means, including electroporation, transfection with calcium chloride, rubidium chloride, calcium phosphate, DEAE-glucan, or other substances; particle bombardment; lipid transfection; and infection (e.g., when the vector is an infectious pathogen such as vaccinia virus). The choice of vector or polynucleotide to be introduced will often depend on the characteristics of the host cell.

This invention also provides host cells comprising any of the polynucleotides described herein. Any host cell capable of overexpressing heterologous DNA can be used for the purpose of isolating genes encoding target antibodies, peptides, or proteins. Non-limiting examples of mammalian host cells include, but are not limited to, COS, HeLa, and CHO cells. See also PCT Publication WO 87/04462. Suitable non-mammalian host cells include prokaryotes (such as Escherichia coli or Bacillus subtilis ) and yeasts (such as Saccharomyces cerevisiae , Saccharomyces pombe , or Kluyveromyces lactis ). Preferably, the host cell expresses cDNA at a level about 5-fold higher, more preferably 10-fold higher, and even more preferably 20-fold higher than the corresponding endogenous antibody or protein CD3 or CD3 domain (e.g., domains 1-4), as determined by immunoassay or FACS. Cells overexpressing target antibodies or proteins can be identified.

This invention also covers scFvs of the antibodies of this invention. Single-chain variable region fragments are prepared by linking light and/or heavy chain variable regions using a short linker peptide (Bird et al., Science 242:423-426, 1988). Examples of linker peptides include the sequence of SEQ ID NO: 65. In one embodiment, the linker peptide bridges approximately 3.5 nm between the carboxyl terminus of one variable region and the amino terminus of another variable region. Linkers with other sequences have been designed and used (Bird et al., Science 242:423-426, 1988). Linkers should be short, flexible polypeptides and preferably contain fewer than about 20 amino acid residues. The linker can then be modified for additional functions, such as linking drugs or solid carriers. Single-chain variants can be generated recombinantly or synthetically. An automated synthesizer can be used to synthesize scFvs. To generate scFv through recombination, a suitable plasmid containing a polynucleotide encoding the scFv can be introduced into a suitable host cell, which can be eukaryotic (such as yeast, plant, insect, or mammalian cells) or prokaryotic (such as Escherichia coli). The polynucleotide encoding the target scFv can be prepared by conventional manipulation (such as linking the polynucleotide). The obtained scFv can be isolated using standard protein purification techniques known in the art.

In one aspect, polynucleotide sequences encoding any of the partial light chain sequences and/or any of the partial heavy chain sequences disclosed herein are provided. Exemplary CD3 polynucleotides encoding the VH and VL regions of the antibodies of the present invention are listed in Table 5.

Table 5

In one aspect, the present invention provides compositions (such as pharmaceutical compositions) comprising any of the polynucleotides of the present invention. In some embodiments, the composition may comprise any of the polynucleotides shown in Table 5. In a particular embodiment, the composition comprises an expression vector containing a polynucleotide encoding any of the antibodies described herein (e.g., any of the polynucleotides shown in Table 5).

This invention also covers polynucleotides complementary to any such sequence. Polynucleotides can be single-stranded (coding or antisense) or double-stranded, and can be DNA (genomic, cDNA, or synthetic) or RNA molecules. RNA molecules include mature and immature mRNAs, such as precursor mRNA (pre-mRNA) or heteronuclear mRNA (hnRNA) and mature mRNA. Additional coding or non-coding sequences may, but do not necessarily, be present within the polynucleotides of this invention, and the polynucleotides may, but do not necessarily, be linked to other molecules and/or supporting materials.

The polynucleotide may contain a native sequence (i.e., an endogenous sequence encoding an antibody or a portion thereof) or a variant thereof. Polynucleotide variants contain one or more substitutions, additions, deletions, and/or insertions such that the immunoreactivity of the encoded polypeptide is not reduced relative to the native immunoreactive molecule. The effect on the immunoreactivity of the encoded polypeptide can generally be evaluated as described herein. Variants preferably exhibit at least about 70% identity with the polynucleotide sequence encoding a native antibody or a portion thereof, more preferably at least about 80% identity, even more preferably at least about 90% identity, and most preferably at least about 95% identity.

Two polynucleotide or polypeptide sequences are said to be "identical" if the nucleotide or amino acid sequences in the two sequences are identical when compared for maximum correspondence as described below. Comparisons between two sequences are typically performed by comparing sequences on a comparison window to identify and compare local regions of sequence similarity. As used herein, a "comparison window" refers to a segment of at least about 20, typically 30 to about 75, or 40 to about 50 adjacent positions, where, after optimal alignment of the two sequences, the sequence can be compared to a reference sequence having the same number of adjacent positions.

Using default parameters, the Megalign procedure in the Lasergene Bioinformatics Software Suite (DNASTAR, Inc., Madison, WI) allows for optimal alignment of sequences for comparison. Preferably, the "percentage of sequence identity" is determined by comparing two optimally aligned sequences across a comparison window of at least 20 positions. The portion of the polynucleotide or polypeptide sequence in the comparison window may contain 20% or less, typically 5 to 15%, or 10 to 12%, of additions or deletions (i.e., gaps) compared to a reference sequence (which does not contain additions or deletions) for optimal alignment of the two sequences. The percentage is calculated by determining the number of positions in both sequences where the same nucleic acid base or amino acid residue appears, to obtain the number of matching positions; dividing the number of matching positions by the total number of positions in the reference sequence (i.e., the window size); and multiplying the result by 100 to obtain the percentage of sequence identity.

The variant can also, or alternatively, be substantially homologous to the natural gene or a portion thereof or its complement. This polynucleotide variant is capable of hybridizing with naturally occurring DNA sequences (or complementary sequences) encoding natural antibodies under moderately stringent conditions.

Those skilled in the art will understand that, due to the degeneracy of the genetic code, there are many nucleotide sequences encoding polypeptides as described herein. Some of these polynucleotides have minimal homology to the nucleotide sequences of any natural gene. However, the present invention specifically considers polynucleotides that vary due to differences in codon usage. Furthermore, alleles of genes comprising the polynucleotide sequences provided herein are within the scope of the invention. An allele is an endogenous gene altered due to one or more mutations (such as nucleotide deletions, additions, and/or substitutions). The resulting mRNA and protein may, but need not, have altered structure or function. Alleles can be identified using standard techniques (such as hybridization, amplification, and/or database sequence comparison).

In one aspect, the present invention provides a method for preparing any of the polynucleotides described herein. For example, the polynucleotides of the present invention can be obtained using chemical synthesis, recombination methods, or PCR. Methods for the chemical synthesis of polynucleotides are well known in the art and need not be described in detail herein. Those skilled in the art can use the sequences provided herein and commercial DNA synthesizers to generate desired DNA sequences.

To prepare polynucleotides using recombinant methods, a polynucleotide containing the desired sequence can be inserted into a suitable vector, which can then be introduced into a suitable host cell for replication and amplification, as discussed further herein. Polynucleotides can be inserted into host cells by any means known in the art. Cells are transformed by introducing exogenous polynucleotides through direct uptake, endocytosis, transfection, F-crossing, or electroporation. Once introduced, the exogenous polynucleotide can be maintained within the cell as a non-integrating vector (such as a plasmid) or integrated into the host cell genome. The amplified polynucleotides can then be isolated from the host cell using methods well-known in the art (e.g., Sambrook et al., 1989).

Alternatively, PCR allows for the replication of DNA sequences. PCR technology is well known in the art and is described in U.S. Patent Nos. 4,683,195, 4,800,159, 4,754,065, and 4,683,202, and PCR: The Polymerase Chain Reaction, edited by Mullis et al., Birkauswer Press, Boston, 1994.

RNA can be obtained by using DNA isolated in a suitable vector and inserted into a suitable host cell. Once the cell replicates and the DNA is transcribed into RNA, RNA can then be isolated using methods well known to those skilled in the art, such as Sambrook et al., 1989, as described above.

Suitable cloning vectors can be constructed using standard techniques or selected from a large number of cloning vectors available in the art. While the chosen cloning vector can vary depending on the intended host cell, useful cloning vectors typically possess self-replicating capabilities, may have a single target for a specific restriction endonuclease, and/or may carry genes related to markers that can be used to select clones containing that vector. Suitable examples include plasmids and bacterial viruses such as pUC18, pUC19, Bluescript (e.g., pBS SK+) and its derivatives, mp18, mp19, pBR322, pMB9, ColE1, pCR1, RP4, phage DNA, and shuttle vectors such as pSA3 and pAT28. These and many other cloning vectors are available from commercial vendors such as BioRad, Strategene, and Invitrogen.

Expression vectors are typically reproducible polynucleotide constructs containing polynucleotides according to the invention. This implies that the expression vector must be replicable in the host cell as an episome or as an integral part of chromosomal DNA. Suitable expression vectors include, but are not limited to, plasmids, viral vectors including adenoviruses, adeno-associated viruses, retroviruses, viscera, and the expression vectors disclosed in PCT Publication No. WO87/04462. Vector components may typically include, but are not limited to, one or more of the following: a signal sequence; an origin of replication; one or more marker genes; and suitable transcriptional control elements (such as promoters, enhancers, and terminators). For expression (i.e., translation), one or more translational control elements, such as ribosome binding sites, translation initiation sites, and stop codons, are also typically required.

Vectors containing the target polynucleotide can be introduced into host cells by any of a number of suitable means, including electroporation, transfection with calcium chloride, rubidium chloride, calcium phosphate, DEAE-glucan, or other substances; particle bombardment; lipid transfection; and infection (e.g., when the vector is an infectious pathogen such as vaccinia virus). The choice of vector or polynucleotide to be introduced will often depend on the characteristics of the host cell.

Any host cell capable of overexpressing heterologous DNA can be used for the purpose of isolating genes encoding target antibodies, peptides, or proteins. Non-limiting examples of mammalian host cells include, but are not limited to, COS, HeLa, and CHO cells. See also PCT Publication WO87/04462. Suitable non-mammalian host cells include prokaryotes (such as *Escherichia coli* or *Bacillus subtilis *) and yeasts (such as *Saccharomyces cerevisiae *, *Schizosaccharomyces pombe * , or *Kluyveromyces lactis *). Preferably, the host cell expresses cDNA at a level approximately 5-fold higher, more preferably 10-fold higher, and even more preferably 20-fold higher than the corresponding endogenous antibody or protein CD3 or CD3 domain (e.g., domains 1-4), as determined by immunoassay or FACS. Cells overexpressing the target antibody or protein can be identified.

In one respect, the CD3 antibody molecule will have an affinity for CD3, as measured, for example, by direct or competitive binding in the picomolar to micromolar affinity range, preferably in the picomolar to low nanomolar range.

Bispecific antibodies can be prepared using the sequences disclosed herein. Methods for preparing bispecific antibodies are known in the art (see, for example, Suresh et al., Methods in Enzymology 121:210, 1986). Traditionally, the recombinant generation of bispecific antibodies is based on the co-expression of two immunoglobulin heavy-light chain pairs, where the two heavy chains have different specificities (Millstein and Cuello, Nature 305, 537-539, 1983).

In another embodiment, the antibody as described herein comprises a full-length human antibody, wherein the antibody variable region of the heterodimeric protein is capable of recruiting human immune effector cells by specifically binding to effector antigens (e.g., CD3 antigens) located on human immune effector cells, and wherein a second antibody variable region of the heterodimeric protein is capable of specifically binding to the target antigen. In some embodiments, the human antibody has an IgG1, IgG2, IgG3, or IgG4 isotype. In some embodiments, the heterodimeric protein comprises an immune-inert Fc chain.

Human immune effector cells can be any variety of immune effector cells known in the art. For example, immune effector cells can be members of the human lymphoid cell lineage, including but not limited to T cells (e.g., cytotoxic T cells), B cells, and natural killer (NK) cells. Immune effector cells can also be, for example, but not limited to, members of the human myeloid lineage, including but not limited to monocytes, neutrophils, and dendritic cells. Such immune effector cells can have cytotoxic or apoptotic effects on target cells, or other desired effects, after activation by binding to effector antigens.

Effector antigens are antigens (e.g., proteins or peptides) expressed on human immune effector cells. Examples of effector antigens that can be bound by heterodimers (e.g., heterodimer antibodies or bispecific antibodies) include, but are not limited to, human CD3, CD16, NKG2D, NKp46, CD2, CD28, CD25, CD64, and CD89.

Target cells can be natural or foreign to humans. In natural target cells, the cell may have been transformed into a malignant cell or pathologically modified (e.g., a natural target cell infected by a virus, malaria parasite, or bacteria). In foreign target cells, the cell is an invading pathogen, such as bacteria, malaria parasite, or virus.

In some embodiments, the antibodies that can be used in this invention are Fab, Fab fragment, F(ab) ₂ fragment, Fv fragment, single-chain Fv fragment, disulfide-stabilized Fv fragment, single-chain antibody, monoclonal antibody, chimeric antibody, bispecific antibody, trispecific antibody, multispecific antibody, bispecific heterodimeric biantibody, bispecific heterodimeric IgG, polyclonal antibody, labeled antibody, humanized antibody, human antibody, or fragments thereof.

In some implementations, the CD3 antibody described herein is a monoclonal antibody. For example, the CD3 antibody is a humanized monoclonal antibody or a chimeric monoclonal antibody.

This invention covers antibodies comprising an Fc chain or a domain thereof, or a portion thereof. In some embodiments, the Fc chain or a portion thereof comprises one or more constant domains (e.g., CH2 or CH3 domains) of an Fc chain of IgG1, IgG2, IgG3, or IgG4. In another embodiment, this invention covers molecules comprising an Fc chain or a portion thereof, wherein the Fc chain or a portion thereof contains at least one amino acid modification (e.g., substitution) relative to an equivalent wild-type Fc chain or a portion thereof. Variant Fc regions are well known in the art and are primarily used to alter the phenotype of antibodies comprising variant Fc regions, as determined in any binding activity or effector function assay well known in the art (e.g., ELISA, SPR assay, or ADCC). Such variant Fc chains or portions thereof can prolong the plasma half-life and stability exhibited by bispecific antibodies comprising Fc chains or portions thereof of this invention. In another embodiment, this invention covers the use of any Fc variant known in the art.

In one embodiment, one or more amino acid modifications are made to the Fc chain to reduce the affinity and coercivity of the Fc chain, thereby reducing the affinity and coercivity of the bispecific antibody of the present invention for one or more FcγR receptors. In one specific embodiment, the present invention covers bispecific antibodies comprising a variant Fc chain or a portion thereof, wherein the variant Fc chain contains at least one amino acid modification relative to the wild-type Fc chain, and the variant Fc region binds only one FcγR, wherein the FcγR is FcγRIIIA. In another specific embodiment, the present invention covers bispecific antibodies comprising a variant Fc chain or a portion thereof, wherein the variant Fc chain contains at least one amino acid modification relative to the wild-type (WT) Fc chain, and the variant Fc region binds only one FcγR, wherein the FcγR is FcγRIIA. In yet another specific embodiment, the present invention covers bispecific antibodies comprising a variant Fc chain or a portion thereof, wherein the variant Fc chain contains at least one amino acid modification relative to the wild-type Fc chain, and the variant Fc chain binds only one FcγR, wherein the FcγR is FcγRIIB. In another embodiment, the invention covers molecules comprising a variant Fc chain, wherein, relative to molecules that do not contain an Fc chain or contain a wild-type Fc chain, the variant confers or mediates reduced ADCC activity (or other effector functions) and/or increased binding to FcγRIIB (CD32B), as measured using methods known to those skilled in the art and described herein.

This invention also covers the use of Fc regions comprising domains or regions from two or more IgG isoforms. As is known in the art, amino acid modifications to the Fc region can significantly affect Fc-mediated effector function and/or binding activity. However, when implemented in the context of a selected IgG isoform, these alterations to functional characteristics can be further improved and/or manipulated. Similarly, the native characteristics of isoform Fc can be manipulated through one or more amino acid modifications. Multiple IgG isoforms (i.e., IgG1, IgG2, IgG3, and IgG4) exhibit different physical and functional properties, including serum half-life, complement binding, FcγR binding affinity, and effector functional activity (e.g., ADCC, CDC), due to differences in the amino acid sequences of their hinges and/or Fc regions.

In one embodiment, amino acid modifications and IgG Fc regions are independently selected based on their respective individual binding and/or effector functional activities to engineer bispecific antibodies with desired characteristics. In one specific embodiment, the binding and/or effector functional activities of amino acid modifications and IgG hinge/Fc regions are separately measured in the context of IgG1, as described herein or known in the art. In one embodiment, amino acid modifications and IgG hinge/Fc regions exhibit similar functionality, such as reduced ADCC activity (or other effector functions) and/or increased binding to FcγRIIB in the context of bispecific antibodies or other Fc-containing molecules (e.g., immunoglobulins). In another embodiment, the invention covers variant Fc regions comprising combinations of amino acid modifications known in the art and selected IgG regions exhibiting novel properties that are undetectable when the modifications and/or regions are measured independently as described herein.

One method for determining the binding affinity of an antibody to CD3 is by measuring the binding affinity of a monofunctional Fab fragment of the antibody. To obtain a monofunctional Fab fragment, the antibody (e.g., IgG) can be cleaved with papain or recombinantly expressed. The affinity of the antibody's CD3 Fab fragment can be determined using a surface plasmon resonance (SPR) system (Biacore™ 3000™ SPR system, Biacore™, INC, Piscataway NJ) equipped with a pre-immobilized streptavidin sensor chip (SA) or anti-mouse Fc or anti-human Fc, using HBS-EP run buffer (0.01 M HEPES, pH 7.4, 0.15 M NaCl, 3 mM EDTA, 0.005% v/v surfactant P20). Biotinylated or Fc-fused human CD3 can be diluted in HBS-EP buffer to concentrations less than 0.5 μg/mL and injected across individual chip channels using variable contact times to achieve two ranges of antigen density: 50–200 reaction units (RU) for detailed kinetic studies, or 800–1,000 RU for screening assays. Regeneration studies have shown that 25 mM NaOH in 25% v/v ethanol effectively removes bound Fab while maintaining CD3 activity on the chip for more than 200 injections. Typically, serial dilutions of purified Fab samples (across concentrations of 0.1–10x estimated _KD ) are injected at 100 µL/min for 1 minute, allowing dissociation times up to 2 hours. Fab protein concentrations are determined by ELISA and/or SDS-PAGE electrophoresis using known concentrations of Fab (e.g., as determined by amino acid analysis) as standards. The data were fitted to a 1:1 Lammill binding model using the BIAevaluation procedure (Karlsson, R. Roos, H. Fagerstam, L. Petersson, B. (1994). Methods Enzymology 6. 99-110), simultaneously obtaining the kinetic binding rate (k _{_on} ) and dissociation rate (k_{_off} ). The equilibrium dissociation constant (K_{_D}) was calculated as k _{_off} /k _{_on} . This protocol is suitable for determining the binding affinity of antibodies to any CD3, including human CD3, CD3 from another mammal (such as mouse CD3, rat CD3, or primate CD3), and different forms of CD3 (e.g., glycosylated CD3). Antibody binding affinity is typically measured at 25°C, but can also be measured at 37°C.

Antibodies as described herein can be prepared by any method known in the art. For the generation of hybridoma cell lines, the host animal immunization pathway and timeline are generally consistent with established and routine techniques for antibody stimulation and production, as further described herein. General techniques for generating human and mouse antibodies are known in the art and/or described herein.

Consider manipulating any mammalian subject, including humans or antibody-producing cells derived therefrom, to serve as a basis for generating mammalian, including human, hybridoma cell lines. Typically, host animals are inoculated with a given amount of immunogen via intraperitoneal, intramuscular, oral, subcutaneous, plantar, and/or intradermal administration, including as described herein.

Hybridomas can be prepared from lymphocytes and immortalized myeloma cells using general somatic cell hybridization techniques, as described in Kohler, B. and Milstein, C., Nature 256:495-497, 1975, or modified techniques, as described in Buck, D. W. et al., In Vitro, 18:377-381, 1982. Available myeloma lines, including but not limited to X63-Ag8.653 and those from the Salk Institute, Cell Distribution Center, San Diego, Calif., USA, can be used in the hybridization. Typically, the technique involves fusing myeloma cells and lymphoid cells using a fusion agent, such as polyethylene glycol, or by electrical means well known to those skilled in the art. After fusion, the cells are separated from the fusion medium and grown in a selective growth medium (such as hypoxanthine-aminopterin-thymidine (HAT) medium) to eliminate unhybridized parental cells. Any culture medium, with or without serum supplementation, as described herein can be used to culture hybridomas that secrete monoclonal antibodies. As an alternative to cell fusion technology, EBV immortalized B cells can be used to generate the monoclonal antibodies of this invention. If necessary, the hybridoma is amplified and subcloned, and the antiimmunogen activity of the supernatant is determined using routine immunoassay procedures (e.g., radioimmunoassay, enzyme immunoassay, or fluorescence immunoassay).

Hybridomas that can be used as antibody sources encompass all derivatives of parental hybridomas that produce monoclonal antibodies specific to CD3 or a portion thereof, as well as progeny cells.

Hybridomas that produce such antibodies can be grown in vitro or in vivo using known procedures. If desired, monoclonal antibodies can be isolated from culture media or body fluids using routine immunoglobulin purification procedures such as ammonium sulfate precipitation, gel electrophoresis, dialysis, chromatography, and ultrafiltration. For example, unwanted activity (if present) can be removed by running the formulation on an adsorbent made of an immunogen attached to a solid phase and eluting or releasing the desired antibody from the immunogen. Using bifunctional reagents or derivatizing agents, such as maleimide benzoyl sulfosuccinimide (conjugated via cysteine residues), N-hydroxysuccinimide (conjugated via lysine residues), glutaraldehyde, succinic anhydride, _SOCl2 , or ^R1N =C=NR (where R and ^R1 are different alkyl groups), conjugated with human CD3 or fragments containing the target amino acid sequence in the host animal to be immunized, a population of antibodies (e.g., monoclonal antibodies) can be obtained.

If desired, the target antibody can be sequenced, and the polynucleotide sequence can then be cloned into a vector for expression or propagation. The sequence encoding the target antibody can be maintained in a vector within a host cell, which can then be expanded and frozen for future use. The generation of recombinant monoclonal antibodies in cell culture can be performed by means known in the art through cloning antibody genes from B cells. See, for example, Tiller et al., J. Immunol. Methods 329,112,2008; U.S. Patent No. 7,314,622.

In an alternative approach, polynucleotide sequences can be genetically manipulated to humanize antibodies or improve their affinity or other characteristics. For example, if an antibody is used in clinical trials and treatments in humans, the constant region can be engineered to more closely resemble the human constant region to avoid an immune response. It might be desirable to genetically manipulate antibody sequences to achieve greater affinity for CD3 and greater therapeutic efficacy.

In some embodiments, the antibody has a modified constant region that removes or reduces Fcγ receptor binding. For example, Fc may be human IgG2 containing the mutant D265, where the amino acid residues are numbered with reference to the wild-type IgG2 sequence. Therefore, in some embodiments, the constant region has a modified constant region having the sequence shown in SEQ ID NO: 80. In some embodiments, the antibody has a modified constant region having the sequence shown in SEQ ID NO: 81.

There are four general steps to humanizing a monoclonal antibody. These steps are: (1) determining the nucleotide and predicted amino acid sequences of the variable regions of the light and heavy chains of the starting antibody; (2) designing the humanized antibody, i.e., deciding which antibody framework region to use during the humanization process; (3) the actual humanization method/technique; and (4) the transfection and expression of the humanized antibody. See, for example, U.S. Patent Nos. 4,816,567; 5,807,715; 5,866,692; 6,331,415; 5,530,101; 5,693,761; 5,693,762; 5,585,089; and 6,180,370.

Many humanized antibody molecules containing antigen-binding sites derived from non-human immunoglobulins have been described, including chimeric antibodies with rodent or modified rodent V regions fused to human constant regions and their associated CDRs. See, for example, Winter et al. Nature 349:293-299, 1991; Lobuglio et al. Proc. Nat. Acad. Sci. USA 86:4220-4224, 1989; Shaw et al. J Immunol. 138:4534-4538, 1987; and Brown et al. Cancer Res. 47:3577-3583, 1987. Other references describe the transplantation of rodent CDRs into human support framework regions (FRs) prior to fusion with appropriate human antibody constant regions. See, for example, Riechmann et al., Nature 332:323-327, 1988; Verhoeyen et al., Science 239:1534-1536, 1988; and Jones et al., Nature 321:522-525, 1986. Another reference describes rodent CDRs supported by recombinantly engineered rodent framework regions. See, for example, European Publication No. EP0519596. These “humanized” molecules are designed to minimize unwanted immune responses to rodent anti-human antibody molecules, which limits the duration and effectiveness of therapeutic applications that are partially present in human receptors. For example, antibody constant regions can be engineered to be immunologically inert (e.g., not triggering complement cleavage). See, for example, PCT application No. PCT/GB99/01441; UK application No. 9809951.8. Other methods using humanized antibodies may also be disclosed by Daugherty et al., Nucl. Acids Res. 19:2471-2476, 1991, and in U.S. Patent Nos. 6,180,377; 6,054,297; 5,997,867; 5,866,692; 6,210,671; and 6,350,861; and PCT Publication No. WO01/27160.

The general principles discussed above regarding humanized antibodies also apply to custom antibodies, for example, for use in other mammals (e.g., humans, non-human primates, mice, donkeys, sheep, rabbits, goats, guinea pigs, camels, horses, or chickens). Furthermore, one or more aspects of the humanized antibodies described herein can be combined, such as CDR transplantation, scaffold mutation, and CDR mutation.

In one variant, fully human antibodies can be obtained using commercially available mice engineered to express specific human immunoglobulin proteins. Transgenic animals designed to produce a more desired (e.g., fully human antibody) or more robust immune response can also be used to generate humanized or human antibodies. Examples of this technology are Xenomouse™ from Abgenix, Inc. (Fremont, CA) and HuMAb-Mouse® and TC Mouse™ from Medarex, Inc. (Princeton, NJ).

In one alternative, the antibody can be recombinantly prepared and expressed using any method known in the art. In another alternative, the antibody can be recombinantly prepared using phage display technology. See, for example, U.S. Patent Nos. 5,565,332; 5,580,717; 5,733,743; and 6,265,150; and Winter et al., Annu. Rev. Immunol. 12:433-455, 1994. Alternatively, phage display technology (McCafferty et al., Nature 348:552-553, 1990) can be used to generate human antibodies and antibody fragments in vitro from a gene library of immunoglobulin variable (V) domains from non-immunized donors. According to this technology, the antibody V domain gene frame is cloned into the major or minor coat protein genes of filamentous phages, such as M13 or fd, and displayed as a functional antibody fragment on the surface of the phage particle. Because filamentous particles contain single-stranded DNA copies of the phage genome, selection based on antibody functional characteristics also leads to the selection of genes encoding antibodies that exhibit those characteristics. Thus, phages mimic some characteristics of B cells. Phage display can be performed in various forms; for reviews, see, for example, Johnson, Kevin S., and Chiswell, David J., Current Opinion in Structural Biology 3:564-571, 1993. Several sources of the V gene segment can be used for phage display. Clackson et al., Nature 352:624-628, 1991, isolated a wide variety of anti-oxazolone antibodies from a small, randomly combined library of the V gene derived from the spleen of immunized mice. Following the techniques described by Mark et al., J. Mol. Biol. 222:581-597, 1991, or Griffith et al., EMBO J.12:725-734, 1993, a V gene library from non-immunized human donors can be constructed, and antibodies against a wide variety of antigens, including self-antigens, can be isolated. In the innate immune response, antibody genes accumulate mutations at a high rate (somatic hypermutation). Some of the introduced changes will confer higher affinity, and B cells exhibiting high-affinity surface immunoglobulins will preferentially replicate and differentiate during subsequent antigen challenges. This natural process can be simulated using a technique called “chain shuffling” (Marks et al., Bio/Technol. 10:779-783, 1992). In this approach, the affinity of “primary” human antibodies obtained by phage display can be improved by sequentially replacing the heavy and light chain V region genes with a naturally occurring library (library) of variants of the V domain genes obtained from non-immunized donors. This technique allows for the production of antibodies and antibody fragments with affinities in the pM-nM range. A strategy for preparing very large phage antibody libraries (also known as “mother-of-all libraries”) has been described by Waterhouse et al., Nucl. Acids Res. 21:2265-2266, 1993. Gene shuffling can also be used to derive human antibodies from rodent antibodies, wherein the human antibodies have similar affinity and specificity to the starting rodent antibody. According to this method, also known as “epitope imprinting,” the heavy or light chain V domain genes of rodent antibodies obtained via phage display technology are replaced with a human V domain gene library, producing a rodent-human chimera. Selection on the antigen results in the isolation of the human variable region capable of restoring a functional antigen-binding site; that is, the epitope governs (marks) the selection of the partner. When this process is repeated to replace the remaining rodent V domains, human antibodies are obtained (see PCT Publication No. WO93/06213). Unlike traditional humanization of rodent antibodies via CDR transplantation, this technique provides fully human antibodies that do not contain rodent-derived frameworks or CDR residues.

Antibodies can be prepared recombinantly by first isolating the antibody and antibody-producing cells from a host animal, obtaining the gene sequence, and using the gene sequence to recombinantly express the antibody in host cells (e.g., CHO cells). Another approach is to express the antibody sequence in plants (e.g., tobacco) or transgenic milk. Methods for recombinantly expressing antibodies in plants or milk have been disclosed. See, for example, Peeters et al. Vaccine 19:2756, 2001; Lonberg, N. and D. Huszar Int. Rev. Immunol 13:65, 1995; and Pollock et al., J Immunol Methods 231:147, 1999. Methods for preparing antibody derivatives, such as humanized, single-chain, etc., are known in the art.

Immunoassays and flow cytometry sorting techniques, such as fluorescence activated cell sorting (FACS), can also be used to isolate antibodies specific to CD3 or target antigens.

Antibodies as described herein can bind to a variety of different solid supports or carriers. Such supports can be active and/or inert. Well-known supports include polypropylene, polystyrene, polyethylene, dextran, nylon, amylase, glass, natural and modified cellulose, polyacrylamide, agarose, and magnets. For the purposes of this invention, the nature of the support can be soluble or insoluble. Those skilled in the art will recognize other suitable supports for binding antibodies, or will be able to determine these using routine experiments. In some embodiments, the support comprises a portion targeting the myocardium.

DNA encoding monoclonal antibodies can be readily isolated and sequenced using standard procedures, such as by using oligonucleotide probes capable of specifically binding to genes encoding the heavy and light chains of monoclonal antibodies. Hybridoma cells serve as a preferred source of this DNA. Once isolated, the DNA can be placed in an expression vector (such as the one disclosed in PCT Publication No. WO87/04462) and then transfected into host cells that otherwise do not produce immunoglobulin proteins (such as E. coli cells, simian COS cells, CHO cells, or myeloma cells) to obtain monoclonal antibody synthesis in recombinant host cells. See, for example, PCT Publication No. WO87/04462. DNA can also be modified, for example, by replacing homologous mouse sequences with coding sequences of human heavy and light chain constant regions (Morrison et al., Proc. Nat. Acad. Sci. 81:6851, 1984), or by covalently binding all or part of the coding sequence of an immunoglobulin-coding sequence to the coding sequence of a non-immunoglobulin polypeptide. In this manner, "chimeric" or "hybrid" antibodies with the binding specificity of the monoclonal antibodies described herein are prepared.

Methods known in the art can be used to identify or characterize CD3 or other antigen antibodies as described herein, thereby detecting and/or measuring antigen expression levels. In some embodiments, CD3 antibodies are identified by incubating a candidate antibody with CD3 and monitoring the associated decrease in binding and/or CD3 expression levels. Binding assays can be performed using purified CD3 peptides, or cells naturally expressing or transfected to express CD3 peptides. In one embodiment, the binding assay is a competitive binding assay in which the ability of a candidate antibody to compete with a known CD3 antibody for CD3 binding is evaluated. This assay can be performed in various forms, including ELISA.

Following initial identification, the activity of candidate CD3 or other antigen-antibody pairs can be further confirmed and refined using bioassays known for testing targeted biological activity. Alternatively, bioassays can be used to directly screen candidates. Several methods for identifying and characterizing antibodies are described in detail in the examples.

CD3 or other antigen antibodies can be characterized using methods well-known in the art. For example, one method is to identify the epitope to which it binds, or "epitope mapping." Many methods are known in the art for mapping and characterizing the location of epitopes on proteins, including resolving the crystal structure of antibody-antigen complexes, competitive assays, gene fragment expression assays, and synthetic peptide-based assays, as described, for example, in Chapter 11 of Harlow and Lane, Using Antibodies, a Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 1999. In another instance, epitope mapping can be used to determine the sequence to which the antibody binds. Epitope mapping is commercially available from various sources, for example, Pepscan Systems (Edelhertweg 15, 8219 PH Lelystad, Netherlands). The epitope can be a linear epitope, i.e., contained in a single amino acid segment, or a conformational epitope formed by the three-dimensional interactions of amino acids, which may not necessarily be contained in a single segment. Peptides of varying lengths (e.g., at least 4-6 amino acids long) can be isolated or synthesized (e.g., recombinantly) and used for binding assays with CD3 or other antigens. In another example, epitopes to which CD3 or other antigens bind can be determined in systematic screening by using overlapping peptides derived from CD3 or other antigen sequences and measuring binding with CD3 or other antigens. Based on gene fragment expression assays, open reading frames encoding CD3 or other antigens are randomly or through specific genetic constructs fragmented, and the reactivity of expressed fragments of CD3 or other antigens with the antibody to be tested is determined. Gene fragments can be generated, for example, by PCR, and then transcribed and translated into proteins in vitro in the presence of radioactive amino acids. The binding of the antibody to the radiolabeled CD3 or other antigen fragment is then determined by immunoprecipitation and gel electrophoresis. Some epitopes can also be identified using large libraries (phage libraries) of random peptide sequences displayed on the surface of phage particles. Alternatively, the binding of a defined library of overlapping peptide fragments to a test antibody can be tested in a simple binding assay. In an additional instance, mutagenesis of antigen-binding domains, domain exchange assays, and alanine partitioning can be performed to identify residues required, sufficient, and/or essential for epitope binding. For example, domain exchange assays can be performed using mutant CD3 or other antigens, where various fragments of the CD3 or other antigenic protein have been replaced (exchanged) with sequences from antigens from another species (e.g., mice), or closely related but antigenically different proteins (e.g., Trop-1). By evaluating the binding of the antibody to mutant CD3 or other antigens, the importance of specific CD3 or other antigenic fragments to antibody binding can be assessed.

Another method that can be used to characterize CD3 or other antigen-antibodies is to use a competitive assay with other antibodies known to bind to the same antigen, i.e., various fragments on CD3 or other antigens, to determine whether the CD3 or other antigen-antibody binds to the same epitopes as the other antibodies. Competitive assays are well known to those skilled in the art.

Expression vectors can be used to direct the expression of CD3 or other antigens and antibodies. Those skilled in the art are familiar with the administration of expression vectors to obtain the expression of exogenous proteins in vivo. See, for example, U.S. Patent Nos. 6,436,908; 6,413,942; and 6,376,471. Administration of expression vectors includes local or systemic administration, including injection, oral administration, particle gun or catheter-directed administration, and local application. In another embodiment, the expression vector is administered directly to the sympathetic trunk or ganglia, or to the coronary arteries, atria, ventricles, or pericardium.

Targeted delivery can also be achieved using therapeutic compositions containing expression vectors or subgenomic polynucleotides. Receptor-mediated DNA delivery techniques are described, for example, in: Findeis et al., Trends Biotechnol., 11:202, 1993; Chiou et al., Gene Therapeutics: Methods And Applications of Direct Gene Transfer, J.A. Wolff, ed., 1994; Wu et al., J. Biol. Chem., 263:621, 1988; Wu et al., J. Biol. Chem., 269:542, 1994; Zenke et al., Proc. Natl. Acad. Sci. USA, 87:3655, 1990; and Wu et al., J. Biol. Chem., 266:338, 1991. Therapeutic compositions containing polynucleotides are administered topically in gene therapy regimens at concentrations ranging from about 100 ng to about 200 mg of DNA. Concentrations ranging from about 500 ng to about 50 mg, about 1 μg to about 2 mg, about 5 μg to about 500 μg, and about 20 µg to about 100 µg of DNA may also be used during gene therapy regimens. Gene delivery vehicles can be used to deliver therapeutic polynucleotides and peptides. Gene delivery vehicles can be of viral or non-viral origin (see generally, Jolly, Cancer Gene Therapy, 1:51, 1994; Kimura, Human Gene Therapy, 5:845, 1994; Connelly, Human Gene Therapy, 1:185, 1995; and Kaplitt, Nature Genetics, 6:148, 1994). Expression of such coding sequences can be induced using endogenous mammalian promoters or heterologous promoters. The expression of a coding sequence can be constitutive or regulated.

Virus-based vectors for delivering desired polynucleotides and expressing them in desired cells are well known in the art. Exemplary virus-based vectors include, but are not limited to, recombinant retroviruses (see, for example, PCT Publications WO90/07936; WO94/03622; WO93/25698; WO93/25234; WO93/11230; WO93/10218; WO91/02805; U.S. Patents 5,219,740 and 4,777,127; GB Patent 2,200,651; and EP Patent EP0 345242), alphavirus-based vectors (e.g., Sindbis virus vectors, Semliki Forest Virus (ATCC)). VR-67; ATCCVR-1247), Ross River virus (ATCC VR-373; ATCCVR-1246) and Venezuelan equine encephalitis virus (ATCC VR-923; ATCC VR-1250; ATCC VR 1249; ATCC VR-532), and adeno-associated virus (AAV) vectors (see, for example, PCT Publications WO94/12649, WO93/03769; WO93/19191; WO94/28938; WO95/11984 and WO95/00655). Alternatively, administration of DNA linked to a killed adenovirus may be used, as described in Curiel, Hum. Gene Ther., 3:147, 1992.

Non-viral delivery media and methods may also be used, including, but not limited to, polycationically condensed DNA linked to or unlinked by a single, killed adenovirus (see, for example, Curiel, Hum. Gene Ther., 3:147, 1992); ligand-linked DNA (see, for example, Wu, J. Biol. Chem., 264:16985, 1989); eukaryotic cell delivery media (see, for example, U.S. Patent No. 5,814,482; PCT Publications WO95/07994; WO96/17072; WO95/30763; and WO97/42338) and nucleic acid charge neutralization or fusion with the cell membrane. Naked DNA may also be used. Exemplary methods for introducing naked DNA are described in PCT Publication WO90/11092 and U.S. Patent No. 5,580,859. Liposomes that can act as gene delivery mediators are described in U.S. Patent No. 5,422,120; PCT Publications Nos. WO95/13796, WO94/23697, WO91/14445; and European Publication EP 0524968. Additional methods are described in Philip, Mol. Cell Biol., 14:2411, 1994 and Woffendin, Proc. Natl. Acad. Sci., 91:1581, 1994.

In some embodiments, the invention covers compositions including pharmaceutical compositions comprising antibodies of the invention as described herein or prepared by the methods described herein and having the characteristics described herein. As used herein, pharmaceutical compositions may comprise one or more antibodies binding to tumor antigens, one or more bispecific antibodies binding to CD3 and tumor antigens, and/or one or more polynucleotides encoding sequences of one or more of these antibodies. These compositions may further comprise suitable excipients, such as pharmaceutically acceptable excipients, including buffers, which are well known in the art.

In one embodiment, the invention is characterized by a composition that is a combination of a CD3 antibody and a second therapeutic agent. In one embodiment, the second therapeutic agent is any agent advantageously combined with a CD3 antibody. Exemplary agents that can be advantageously combined with a CD3 antibody include, but are not limited to, other agents that bind and/or activate CD3 signaling (including other antibodies or their antigen-binding fragments, etc.) and/or agents that do not directly bind CD3 but still activate or stimulate immune cell activation. The invention further includes additional combination therapies and co-formulations involving the CD3 antibodies of the invention. The invention also provides methods for preparing any of these antibodies. The antibodies of the invention can be prepared by procedures known in the art, including de novo protein synthesis and recombinant expression of nucleic acids encoding binding proteins. The desired nucleic acid sequence can be generated by recombinant methods (e.g., PCR mutagenesis of previously prepared variants of the desired polynucleotide) or by solid-phase DNA synthesis. Recombinant expression methods are typically used. In one aspect, the invention provides polynucleotides comprising sequences encoding CD3VH and/or VL. Due to the degeneracy of the genetic code, multiple nucleic acid sequences encode each immunoglobulin amino acid sequence, and the invention includes all nucleic acids encoding the binding proteins described herein.

Chimeric or hybrid antibodies can also be prepared in vitro using known methods of synthetic protein chemistry, including those involving cross-linking agents. For example, immunotoxins can be constructed using disulfide exchange reactions or by forming thioether bonds. Examples of suitable reagents for this purpose include iminothiolates and methyl-4-mercaptobutyrimidate.

In recombinant humanized antibodies, the Fcγ region can be modified to avoid interactions with the Fcγ receptor, as well as with complement and the immune system. Techniques for preparing such antibodies are described in PCT Publication WO99/58572. For example, if the antibody is to be used in clinical trials and treatments in humans, the constant region can be engineered to more closely resemble the human constant region to avoid an immune response. See, for example, U.S. Patent Nos. 5,997,867 and 5,866,692.

Compared to the corresponding germline sequences from which antibodies are derived, CD3 antibodies disclosed herein may contain one or more amino acid substitutions, insertions, and/or deletions in the framework and/or CDR regions of the heavy and light chain variable domains. Such mutations can be readily identified by comparing the amino acid sequences disclosed herein with germline sequences available from, for example, public antibody sequence databases. This invention includes antibodies and antigen-binding fragments thereof derived from any of the amino acid sequences disclosed herein, wherein one or more amino acids in one or more framework and/or CDR regions are mutated to one or more corresponding residues of the germline sequence from which the antibody is derived, or one or more corresponding residues of another human germline sequence, or conserved amino acid substitutions of one or more corresponding germline residues (such sequence changes are collectively referred to herein as “germline mutations”). Starting with the heavy and light chain variable region sequences disclosed herein, those skilled in the art can readily generate a variety of antibody and antigen-binding fragments containing one or more individual germline mutations or combinations thereof. In some embodiments, all framework and/or CDR residues within the VH and/or VL domains are mutated back to residues found in the original germline sequence from which the antibody is derived. In some embodiments, only certain residues are mutated back to the original germline sequence, for example, mutated residues found only in the first 8 amino acids of FR1 or the last 8 amino acids of FR4, or mutated residues found only in CDR1, CDR2, or CDR3. In some further embodiments, one or more framework and/or CDR residues are mutated to one or more corresponding residues of a different germline sequence (i.e., a germline sequence from which the antibody was originally derived).

Furthermore, the antibodies of the present invention may contain any combination of two or more germline mutations within the framework and/or CDR region, for example, wherein certain individual residues are mutated to corresponding residues of a specific germline sequence, while certain other residues different from the original germline sequence are maintained or mutated to corresponding residues of a different germline sequence. Once obtained, one or more desired properties of the antibody and antigen-binding fragment containing one or more germline mutations can be readily tested, such as increased binding specificity, increased binding affinity, improved or enhanced antagonistic or agonistic biological properties (as the case may be), reduced immunogenicity, etc. Antibody and antigen-binding fragments obtained in this general manner are included within the scope of the present invention.

The present invention also includes CD3 antibodies comprising variants of any of the HC VR, LC VR, and/or CDR amino acid sequences disclosed herein having one or more conserved substitutions. For example, the present invention includes CD3 antibodies having HC VR, LC VR, and/or CDR amino acid sequences having, for example, 10 or fewer, 8 or fewer, 6 or fewer, 4 or fewer, etc., conserved amino acid substitutions relative to any of the HC VR, LC VR, and/or CDR amino acid sequences, as disclosed herein.

Therefore, this invention covers modifications of the antibodies and peptides of this invention, variants as described herein, including functionally equivalent antibodies that do not significantly affect their properties, and variants with enhanced or reduced activity and/or affinity. For example, amino acid sequences can be mutated to obtain antibodies with desired binding affinity to tumor antigens and/or CD3. Peptide modification is a common practice in the art and need not be described in detail herein. Examples of modified peptides include peptides with conserved substitutions of amino acid residues, deletions or additions of one or more amino acids that do not significantly and harmfully alter the functional activity or affinity of the mature (enhancing) peptide for its ligands, or the use of chemical analogs.

Amino acid sequence insertions include fusion of the amino and/or carboxyl ends of peptides ranging in length from one residue to one hundred or more residues, as well as intra-sequence insertions of single or multiple amino acid residues. Examples of terminal insertions include antibodies with an N-terminal methionine residue or antibodies fused to an epitope tag. Other insertion variants of antibody molecules include fusion of an enzyme or a peptide that increases the antibody's half-life in the bloodstream with the N- or C-terminus of the antibody.

Substitution variants involve the removal of at least one amino acid residue from the antibody molecule and the insertion of a different residue at its position. Sites of most interest for substitution mutagenesis include hypervariable regions, but FR alterations are also expected. Conservative substitutions are shown under the heading “Conservative Substitutions” in Table 6. If such substitution results in an alteration of biological activity, more significant changes, designated “Exemplary Substitutions” in Table 6 or further described below regarding amino acid categories, can be introduced, and products screened.

Table 6

Significant modifications to the biological properties of antibodies are accomplished through selective substitution, which differs markedly in maintaining: (a) the polypeptide backbone structure in the substituted region, for example, as a sheet or helical conformation; (b) the charge or hydrophobicity of the molecule at the target site; or (c) the volume of the side chains. Naturally occurring amino acid residues are grouped based on common side chain properties:

(1) Nonpolar: Leucine, Met, Ala, Val, Leu, Ile;

(2) Neutral elements with no charge: Cys, Ser, Thr, Asn, Gln;

(3) Acidic (negatively charged): Asp, Glu;

(4) Alkaline (positively charged): Lys, Arg;

(5) Residues affecting chain orientation: Gly, Pro; and

(6) Aroma: Trp, Tyr, Phe, His.

Non-conservative substitutions are prepared by replacing members of one of these categories with members of another category.

Serine residues that are not involved in maintaining the correct conformation of the antibody can generally be replaced with serine to improve the oxidative stability of the molecule and prevent abnormal cross-linking. Conversely, one or more cysteine bonds can be added to the antibody to improve its stability, especially when the antibody is an antibody fragment, such as an Fv fragment.

Amino acid modifications can involve a complete redesign of a region (e.g., a variable region) by altering or modifying one or more amino acids. Changes in the variable region can alter binding affinity and/or specificity. In some embodiments, no more than one to five conserved amino acid substitutions are made within the CDR domain. In some embodiments, no more than one to three conserved amino acid substitutions are made within the CDR domain. In still other embodiments, the CDR domain is VH CDR3 and/or VL CDR3.

Modifications also include glycosylated and non-glycosylated peptides, as well as peptides with other post-translational modifications, such as, for example, glycosylation, acetylation, and phosphorylation with different sugars. Antibodies are glycosylated at conserved sites in their constant regions (Jefferis and Lund, Chem. Immunol. 65:111-128, 1997; Wright and Morrison, TibTECH 15:26-32, 1997). Oligosaccharide side chains of immunoglobulins influence protein function (Boyd et al., Mol. Immunol. 32:1311-1318, 1996; Wittwe and Howard, Biochem. 29:4175-4180, 1990), as well as intramolecular interactions between glycoprotein moieties, which can affect conformation and present the three-dimensional surface of glycoproteins (Jefferis and Lund, above; Wyss and Wagner, Current Opin. Biotech. 7:409-416, 1996). Oligosaccharides can also be used to target given glycoproteins to certain molecules based on specific recognition structures. Glycosylation of antibodies has also been reported to affect antibody-dependent cytotoxicity (ADCC). In particular, CHO cells with tetracycline-regulated β(1,4)-N-acetylglucosamine transferase III (GnTIII) (a glycosyltransferase that catalyzes the formation of aliquots of GlcNAc) expression have been reported to have improved ADCC activity (Umana et al., Mature Biotech. 17:176-180, 1999).

Antibody glycosylation is generally N-linked or O-linked. N-linked glycosylation refers to the attachment of the carbohydrate moiety to the side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine, asparagine-X-threonine, and asparagine-X-cysteine (where X is any amino acid except proline) are recognition sequences for the enzymatic attachment of the carbohydrate moiety to the asparagine side chain. Therefore, the presence of any of these tripeptide sequences in the polypeptide creates a potential glycosylation site. O-linked glycosylation refers to the attachment of one of the sugars N-acetylgalactosamine, galactose, or xylose to a hydroxy amino acid, most commonly serine or threonine, although 5-hydroxyproline or 5-hydroxylysine can also be used.

By altering the amino acid sequence to include one or more of the aforementioned tripeptide sequences (for N-glycosylation sites), the addition of glycosylation sites to the antibody can be conveniently accomplished. Alternatively, alterations can be made by adding or replacing one or more serine or threonine residues to the original antibody sequence (for O-glycosylation sites).

The glycosylation pattern of an antibody can also be altered without changing the underlying nucleotide sequence. Glycosylation is largely dependent on the host cell used to express the antibody. Since the cell types used to express recombinant glycoproteins such as antibodies as potential therapeutic agents are rarely native cells, variations in the antibody glycosylation pattern can be expected (see, for example, Hse et al., J. Biol. Chem. 272:9062-9070, 1997).

Besides host cell selection, factors influencing glycosylation during recombinant antibody production include growth mode, culture medium formulation, culture density, oxygenation, pH, purification protocol, and more. Various methods have been proposed to alter the glycosylation patterns achieved in specific host organisms, including the introduction or overexpression of certain enzymes associated with oligosaccharide production (US Patent Nos. 5,047,335; 5,510,261 and 5,278,299). For example, using endoglycosidase H (Endo H), N-glycosidase F, endoglycosidase F1, endoglycosidase F2, and endoglycosidase F3, glycosylation or certain types of glycosylation can be enzymatically removed from glycoproteins. Additionally, recombinant host cells can be genetically engineered to be defective in processing certain types of polysaccharides. These and similar techniques are well known in the art.

Other modification methods include the use of coupling techniques known in the art, including, but not limited to, enzymatic methods, oxidative substitution, and chelation. Modification can be used, for example, for label attachment for immunoassays. Modified peptides are prepared using procedures established in the art and can be screened using standard assays known in the art, some of which are described below and in the examples.

In some embodiments of the invention, the antibody comprises a modified constant region, such as a constant region that has increased affinity for the human Fcγ receptor and is immunologically inert or partially inert, for example, not triggering complement-mediated cleavage, not stimulating antibody-dependent cell-mediated cytotoxicity (ADCC), or not activating macrophages; or having reduced activity (compared to an unmodified antibody) in any one or more of the following: triggering complement-mediated cleavage, stimulating antibody-dependent cell-mediated cytotoxicity (ADCC), or activating microglia. Different modifications to the constant region can be used to achieve optimal levels and/or combinations of effector functions. See, for example, Morgan et al., Immunology 86:319-324, 1995; Lund et al., J. Immunology 157:4963-9 157:4963-4969, 1996; Idusogie et al., J. Immunology 164:4178-4184, 2000; Tao et al., J. Immunology 143: 2595-2601, 1989; and Jefferis et al., Immunological Reviews 163:59-76, 1998. In some embodiments, the constant region is modified as described in Eur. J. Immunol., 29:2613-2624, 1999; PCT application number PCT/GB99/01441; and/or UK application number 9809951.8. In still other embodiments, the constant region is aglycosylated against N-glycosylation. In some embodiments, the constant region is aglycosylated against N-glycosylation by mutating glycosylated amino acid residues or flanking residues in the constant region that are part of the N-glycosylation recognition sequence. For example, the N-glycosylation site N297 can be mutated to A, Q, K, or H. See Tao et al., J. Immunology 143: 2595-2601, 1989; and Jefferis et al., Immunological Reviews 163:59-76, 1998. In some implementations, the constant region is deglycosylated against N-glycosylation. The constant region may be deglycosylated enzymatically (e.g., by removing carbohydrates via the enzyme peptide-N-glycosidase (PNGase)) or by expression against N-glycosylation in a glycosylation-deficient host cell.

Other antibody modifications include antibodies modified as described in PCT Publication No. WO99/58572. In addition to the binding domain targeting the target molecule, these antibodies also contain an effector domain having an amino acid sequence substantially homologous to all or part of the constant region of the human immunoglobulin heavy chain. These antibodies are capable of binding the target molecule without triggering significant complement-dependent cleavage or cell-mediated target disruption. In some embodiments, the effector domain is capable of specifically binding FcRn and/or FcγRIIb. These are generally based on chimeric domains derived from two or more CH2 domains of the human immunoglobulin heavy chain. Antibodies modified in this manner are particularly suitable for use in long-term antibody therapy to avoid inflammation and other adverse reactions to conventional antibody therapy.

This invention includes embodiments of affinity maturation. For example, affinity-matured antibodies can be generated by procedures known in the art (Marks et al., Bio/Technology, 10:779-783, 1992; Barbas et al., ProcNat. Acad. Sci, USA 91:3809-3813, 1994; Schier et al., Gene, 169:147-155, 1995; Yelton et al., J. Immunol., 155:1994-2004, 1995; Jackson et al., J. Immunol., 154(7):3310-9, 1995; Hawkins et al., J. Mol. Biol., 226:889-896, 1992; and PCT Publication No. WO2004/058184).

The following methods can be used to modulate antibody affinity and characterize CDRs. One way to characterize the CDR of an antibody and/or alter (e.g., improve) the binding affinity of a peptide, such as an antibody, is called “library scan mutagenesis.” Typically, library scan mutagenesis is performed as described below. Using methods recognized in the art, one or more amino acid positions in the CDR are replaced with two or more amino acids (e.g., 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20). This produces small clonal libraries (in some embodiments, one clone for each amino acid position analyzed), each clone having the complexity of two or more members (if two or more amino acids are substituted at each position). Typically, the library also includes clones containing native (unsubstituted) amino acids. For a small number of clones from each library, e.g., approximately 20–80 clones (depending on library complexity), screening is performed for binding affinity to the target peptide (or other binding target), and candidates exhibiting increased binding, identical binding, decreased binding, or no binding are identified. Methods for determining binding affinity are well known in the art. Binding affinity can be determined using BIACORE™ surface plasmon resonance analysis, which detects differences in binding affinity of approximately 2-fold or greater. BIACORE™ is particularly useful when the starting antibody has already bound with relatively high affinity (e.g., approximately 10 nM or lower K _{_D} ). Screening using BIACORE™ surface plasmon resonance is described in the examples herein.

Binding affinity can be determined using Kinexa Biocensor, scintillation affinity assay, ELISA, ORIGEN immunoassay (IGEN), fluorescence quenching, fluorescence transfer, and/or yeast display. Binding affinity can also be screened using appropriate bioassays.

In some embodiments, using art-recognized mutagenesis methods (some of which are described herein), each amino acid position in the CDR is replaced (in some embodiments, one at a time) with all 20 natural amino acids. This produces small clonal libraries (in some embodiments, one clone for each amino acid position analyzed), each clone having a complexity of 20 members (if all 20 amino acids are replaced at each position).

In some implementations, the library to be screened contains substitutions at two or more positions, which may be in the same CDR or in two or more CDRs. Thus, the library may contain substitutions at two or more positions in a single CDR. The library may contain substitutions at two or more positions in two or more CDRs. The library may contain substitutions at three, four, five, or more positions, which are found in two, three, four, five, or six CDRs. Substitutions can be prepared using low-redundancy codons. See, for example, Table 2, Balint et al., Gene 137(1):109-18, 1993.

Candidates exhibiting improved binding can be sequenced to identify CDR substitution mutants that result in improved affinity (also known as "improved" substitutions). Binding candidates can also be sequenced to identify CDR substitutions that maintain binding.

Multiple rounds of screening can be performed. For example, candidates with improved binding (each candidate containing amino acid substitutions at one or more positions of one or more CDRs) can also be used to design a second library containing at least the original and substituted amino acids at each improved CDR position (i.e., the amino acid position in the CDR where the substitution mutant shows improved binding). The preparation and screening or selection of this library are discussed further below.

Library scanning mutagenesis also provides information about the importance of each amino acid position to the stability of the antibody-antigen complex because the frequency of clones exhibiting improved binding, identical binding, reduced binding, or no binding also provides information related to the stability of the antibody-antigen complex. For example, if a CDR position remains bound when changed to all 20 amino acids, that position is identified as unlikely to be required for antigen binding. Conversely, if a CDR position remains bound only with a small percentage of substitutions, that position is identified as important for CDR function. Therefore, library scanning mutagenesis yields information about positions in CDRs that can be changed to many different amino acids (including all 20 amino acids) and positions in CDRs that cannot be changed or can only be changed to a few amino acids.

Candidates with improved affinity can be combined in a second library comprising the improved amino acid, the original amino acid at that position, and may further include additional substitutions at that position, depending on the required complexity of the library or the complexity permitted by the desired screening or selection method. Additionally, if needed, adjacent amino acid positions can be randomized to at least two or more amino acids. Randomization of adjacent amino acids allows for additional conformational flexibility in the mutant CDR, which in turn allows or facilitates the introduction of a large number of improving mutations. The library may also contain substitutions at positions that did not exhibit improved affinity in the first round of screening.

Use any method known in the art to screen or select library members in the second library that have improved and/or altered binding affinity, including screening using BIACORE™ surface plasmon resonance analysis, and selection using any method known in the art for selection, including phage display, yeast display, and ribosome display.

The techniques used for isolating and preparing antibodies are as follows. However, it will be understood that the antibodies can be formed from or incorporated into other polypeptides using techniques known in the art. For example, nucleic acids encoding a target polypeptide (e.g., a ligand, receptor, or enzyme) can be isolated from a cDNA library prepared from a tissue believed to have polypeptide mRNA and express it at detectable levels. The library is screened using probes (e.g., antibodies or oligonucleotides of about 20-80 bases) designed to identify the target gene or the protein it encodes. Screening of cDNA or genomic libraries with selected probes can be performed using standard procedures such as those described in Chapters 10-12 of Sambrook et al., Molecular Cloning: A Laboratory Manual (New York: Cold Spring Harbor Laboratory Press, 1989).

CD3 antibodies used to treat autoimmune diseases

In one aspect, a method for treating an autoimmune disease in a subject is provided, comprising administering to the subject in need an effective amount of a composition comprising an antibody as described herein.

As used in this article, autoimmune diseases include, but are not limited to, inflammatory bowel disease, systemic lupus erythematosus, rheumatoid arthritis, diabetes mellitus (type I), multiple sclerosis, Addison’s disease, celiac disease, dermatomyositis, Graves’ disease, Hashimoto’s thyroiditis, Hashimoto’s encephalopathy, myasthenia gravis, reactive arthritis, Sjogren’s syndrome, atopic allergy, atopic dermatitis, autoimmune enteropathy, autoimmune hepatitis, autoimmune lymphoproliferative syndrome, autoimmune peripheral neuropathy, autoimmune pancreatitis, autoimmune polyendocrine syndrome, and autoimmune progesterone dermatitis. Autoimmune urticaria, autoimmune uveitis, Behcet’s disease, Castleman’s disease, cold agglutinin disease, Crohn’s disease, dermatomyositis, eosinophilic fasciitis, gastrointestinal pemphigoid, Goodpasture’s syndrome, Guillain-Barré syndrome, hidradenitis suppurativa, somnambulism, pemphigus vulgaris, polymyositis, relapsing polychondritis, rheumatic fever, transverse myelitis, ulcerative colitis, undifferentiated connective tissue disease, vasculitis, and Wegener’s granulomatosis.

Multispecific antibodies and their uses

The CD3 antibodies of the present invention may be specific to different epitopes of a single target polypeptide or may contain antigen-binding domains specific to more than one target polypeptide. See, for example, Tutt et al., 1991, J. Immunol. 147:60-69; Kufer et al., 2004, Trends Biotechnol. 22:238-244. The CD3 antibodies of the present invention may be ligated to or co-expressed with another functional molecule (e.g., another peptide or protein). For example, the antibody or a fragment thereof may be functionally linked (e.g., by chemical coupling, genetic fusion, non-covalent association, or other means) to one or more other molecular entities (e.g., another antibody or antibody fragment) to produce a multispecific antibody with a second binding specificity.

In one aspect, the antibody of the present invention is a multispecific antibody. In a specific embodiment, the CD3 antibody is a bispecific antibody that specifically binds to human CD3. In some such embodiments, the bispecific antibody further specifically binds to tumor antigens. In another such embodiment, the bispecific antibody of the present invention simultaneously targets T cells (CD3) and tumor cells, and successfully guides and activates T cells to exert cytotoxicity on tumor cells expressing tumor antigens.

In the respective further embodiments described above, the CD3-binding bispecific antibody is characterized by any one or more of the following properties: (a) treating, preventing, or improving one or more symptoms of a condition in a subject associated with malignant cells expressing a specific tumor antigen (e.g., B-cell-related cancers, such as multiple myeloma); (b) inhibiting tumor growth or progression in a subject (who has a malignant tumor expressing a specific tumor antigen); (c) inhibiting metastasis of cancer (malignant) cells expressing a specific tumor antigen in a subject (who has one or more malignant cells expressing a tumor antigen); (d) inducing regression (e.g., long-term regression) of tumors expressing tumor antigens; (e) exerting cytotoxic activity in malignant cells expressing tumor antigens; (f) prolonging progression-free survival in subjects with tumor-related conditions; (g) prolonging overall survival in subjects with tumor-related conditions; (h) reducing the use of additional chemotherapy or cytotoxic agents in subjects with tumor-related conditions; (i) reducing tumor burden in subjects with tumor-related conditions; or (j) blocking the interaction of tumor antigens with other unidentified factors.

As used herein, a bispecific antibody of the present invention refers to a complex of two or more polypeptide chains, each polypeptide chain comprising at least one antibody VL region and one antibody VH region or a fragment thereof, wherein the VL and VH regions in each polypeptide chain are derived from different antibodies. In a particular aspect, a bispecific antibody comprises a dimer or tetramer of polypeptide chains containing both VL and VH regions. The polypeptide chains constituting the multimeric protein may be covalently linked to at least one other peptide of the multimer via interchain disulfide bonds.

In some such embodiments, the bispecific antibody of the present invention comprises a first heterodimer-promoting domain on a first polypeptide chain and a second heterodimer-promoting domain on a second polypeptide chain (Figure 1). In summary, the first and second heterodimer-promoting domains drive heterodimerization and/or stabilize the bispecific antibody (e.g., through the interaction of pestle and mortar on complementary heterodimer-promoting domains) and/or serve to stabilize the bispecific antibody. In some embodiments, the first and second heterodimer-promoting domains each comprise a CH2 domain and a CH3 domain, wherein the amino acid sequences of each of the CH2 domains and/or the CH3 domains are modified to drive heterodimerization and/or stabilize the bispecific antibody.

In some embodiments, the first heterodimer promoting domain may comprise an Fc chain having CH2 and/or CH3 domains modified to comprise a club (protrusion) or a mortise (cavity). In some such embodiments, the amino acid sequences of the CH2 and/or CH3 domains comprise at least one amino acid modification, wherein: (a) the CH3 domain of the first heterodimer promoting domain forms a club; and (b) the CH3 domain of the second heterodimer promoting domain forms a hole. In another such embodiment, the CH3 domain of the first heterodimer promoting domain comprises the mutants Y349C and/or T366W; and the CH3 domain of the second heterodimer promoting domain comprises the mutants S354C, T366S, L368A, and/or Y407V (according to EU index number).

In one embodiment, if the second heterodimer promoting domain comprises a CH2 and/or CH3 domain modified to comprise a mortise (cavity), then the first heterodimer promoting domain may comprise a CH2 and/or CH3 domain modified to comprise a pestle (protrusion) containing the sequence of SEQ ID NO: 78. In another embodiment, if the second heterodimer promoting domain comprises a CH2 and/or CH3 domain modified to comprise a pestle (protrusion), then the first heterodimer promoting domain may comprise a mortise (cavity) containing the sequence of SEQ ID NO: 79.

Each polypeptide chain of a bispecific antibody contains a VL region and a VH region, which are covalently linked by a glycine-serine linker (linker 1 or linker 2) containing glycine and serine residues, thus confining the antibody-binding domain to prevent self-assembly. Furthermore, each polypeptide chain contains a heterodimerization domain, which promotes heterodimerization and/or stabilization of multiple polypeptide chains and reduces the likelihood of homodimerization between different polypeptide chains. The heterodimerization domain can be located at the N-terminus or C-terminus of the polypeptide chain. The heterodimerization domain can contain a cysteine linker (linker 3) of 1, 2, 3, 4, 5, 6, or more amino acid residues in length. The interaction of two polypeptide chains can produce two VL/VH pairs, thereby forming two epitope-binding domains, i.e., divalent molecules. Neither the VH nor the VL region is confined to any position within the polypeptide chain, i.e., restricted to the amino-terminus or carboxyl-terminus, nor is the region confined by their relative positions to each other; i.e., the VL region can be at the N-terminus of the VH region, and vice versa. The only limitation is the availability of complementary polypeptide chains to form functional bispecific antibodies. In cases where the VL and VH regions originate from antibodies specific to different antigens, the formation of functional bispecific antibodies requires the interaction of two distinct polypeptide chains, i.e., the formation of heterodimers. Conversely, in cases where two distinct polypeptide chains interact freely, for example in a recombinant expression system, one containing VLA and VHB (A being the first epitope and B the second epitope) and the other containing VLB and VHA, two distinct binding sites can be formed: VLA-VHA and VLB-VHB. For all bispecific antibody polypeptide chain pairs, misalignment or misbinding of the two chains is a possibility, for example, interactions between the VL-VL or VH-VH regions. However, the purification of functional bispecific antibodies can be readily controlled based on the immunospecificity of the correctly dimerized binding sites using any affinity-based method known in the art or exemplified herein, such as affinity chromatography.

In one embodiment, the polypeptide chain of the bispecific antibody may contain various linkers and peptides. The linkers and peptides may contain 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 or more amino acids.

In some embodiments, domain 1 is covalently bound to a first heterodimer via a cysteine linker to promote the domain, and domain 2 is covalently bound to a second heterodimer via a cysteine linker to promote the domain. Each cysteine linker includes at least one cysteine residue to allow intramolecular disulfide bonding. In a further embodiment of each of the foregoing embodiments, the cysteine linker (linker 3) comprises at least five amino acids.

In some embodiments, the first polypeptide chain is covalently bound to the second polypeptide chain via at least one disulfide bond. In some such embodiments, at least one disulfide bond is formed between the linker 3 of the first polypeptide chain and the linker 3 of the second polypeptide chain. In another such embodiment, at least one disulfide bond is formed between the first heterodimer promoting domain and the second heterodimer promoting domain. In a specific embodiment, each disulfide bond is formed by linking two cysteine residues. In one aspect of the invention, the bispecific antibody shown in FIG1 comprises a first polypeptide chain and a second polypeptide chain. In some such embodiments, the linker 3 may comprise a truncated lower hinge region of human IgG1 having the sequence SEQ ID NO: 64, which has at least one glycine residue preceding the lower hinge region.

The bispecific antibody of the present invention can simultaneously bind to two separate and distinct epitopes. In some embodiments, at least one epitope binding site is specific for a CD3 determinant expressed on immune effector cells, such as T lymphocytes. In one embodiment, the bispecific antibody molecule binds to an effector cell determinant and also activates the effector cell.

In one aspect, the present invention provides a first polypeptide chain and a second polypeptide chain. In one embodiment, the first polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 36. In another embodiment, the second polypeptide chain comprises the amino acid sequence shown in one or more of SEQ ID NO: 37 or 38 (Table 7). In a preferred embodiment, the first polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 36; and the second polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 37 or 38. In one embodiment, the second polypeptide chain is any polypeptide to be associated with the first polypeptide chain via an interface. Table 7 shows the sequences of the first and second polypeptide chains of the GUCY2c-H2B4 and GUCY2c-2B5 bispecific antibodies.

Table 7

In specific embodiments, the bispecific antibody of the present invention (a) binds to the extracellular domain of a human tumor antigen; (b) exhibits an extended serum and tumor half-life of 30 minutes to 100 days; and/or (c) exhibits a lower _EC50 value of 0.0001 nM to 100 nM in the presence of increased tumor expression levels or increased receptor density levels.

In one implementation, the epitope-binding domain is capable of binding to tumor-associated antigens associated with breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer (including but not limited to non-small cell lung cancer and small cell lung cancer), bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and digestive system cancers. Digestive system cancers include, but are not limited to, cancers of the esophagus, stomach, small intestine, colon, rectum, anus, liver, gallbladder, appendix, bile ducts, and pancreas. In a specific implementation, the therapy activates a cytolytic T-cell response.

Null effector mutations in human IgG1 CH2-CH3

Using standard primer-guided PCR mutagenesis, the Fc chain of human IgG1 was modified to introduce mutants L234A, L235A, and G237A (SEQ ID NO: 82, according to EU index number) to oblate effector functions attributed to binding to FcγRIII, thereby providing an effector-ineffective phenotype (Canfield et al., J. Exp. Med (1991) 173: 1483-1491; Shields et al., J. Biol. Chem. (2001) 276: 6591-604).

The mortise in human IgG1 CH2-CH3 is a mutation.

The 'mortar' configuration is a known efficient design strategy in the art for engineering antibody heavy chain homodimers for heterodimerization. In this method, 'mortar' variants are obtained by replacing smaller amino acids, such as Y349C and T366W (according to EU index numbers), with larger amino acids in one strand of the Fc chain of IgG1. The 'mortar' is designed to insert into the 'mortar' of the CH3 domain of the complementary chain of the Fc chain generated by replacing larger residues, such as S354C, T366S, L368A, and Y407V (according to EU index numbers), with smaller residues.

In some implementations, complementary mutations are introduced to achieve heterodimerization of the resulting Fc chains, such that each Fc chain carries a set of mutations, Y349C and T366W (SEQ ID NO: 78) for the club (or protrusion) Fc chain, or S354C, T366S, L368A, and Y407V (SEQ ID NO: 79) for the mortis (or cavity) Fc chain, as provided in Table 8. When co-transfected into a suitable mammalian host, the DNA encoding the amino acid sequences (e.g., SEQ ID NO: 78 and 79) generates Fc domains predominantly with a bispecific antibody associated with one club (or protrusion) Fc chain of one mortis (or cavity) Fc chain.

Table 8

The CD3-tumor antigen bispecific antibody of this invention is a thermostability-stable, efficient bispecific antibody-Fc fusion targeting both human CD3 and tumor antigens. The Fc domain in the mortise allows for improved expression and purification in CHO cells, resulting in the desired high purity of the heterodimer. The engineered mutation within the Fc domain cancels FcγR binding, thus potentially avoiding ADCC-mediated T cell exhaustion. Furthermore, as shown by differential scanning calorimetry (DSC), incorporating the Fc domain into the bispecific antibody enhances the molecule's stability.

If a bispecific antibody can be engineered to include at least one cysteine residue, it can interact with a corresponding cysteine residue on another polypeptide chain of the invention to form an interchain disulfide bond. The interchain disulfide bond can be used to stabilize the bispecific antibody, thereby improving expression and recovery in recombinant systems, resulting in stable and consistent formulations, and improved stability of isolated and/or purified products in vivo. One or more cysteine residues can be introduced into any part of the polypeptide chain as a single amino acid or as part of a larger amino acid sequence, such as a hinge region. In a specific aspect, at least one cysteine residue is engineered to appear at the C-terminus of the polypeptide chain.

This invention covers methods and compositions for treating, preventing, or managing cancer in a subject, including administering to the subject a therapeutically effective amount of an antibody engineered according to the invention, the molecule further binding to a cancer antigen. The antibodies of this invention are particularly useful for preventing, inhibiting, reducing the growth and/or regression of primary tumors, and for the metastasis of cancer cells. Although not intended to be limited to a specific mechanism of action, the antibodies of this invention can mediate effector functions that can lead to tumor clearance, tumor reduction, or a combination thereof.

In one aspect, the present invention provides a therapeutic method for stimulating T cell activation using the CD3 antibody of the present invention, wherein the therapeutic method comprises administering a therapeutically effective amount of a pharmaceutical composition comprising the antibody of the present invention to a subject in need. The treated condition is any disease or condition that is improved, alleviated, suppressed, or prevented by stimulating CD3 activity or signaling. In a specific embodiment, the present invention provides a bispecific antigen-binding molecule, such as a bispecific antibody that binds to both CD3 and a target antigen.

In one aspect, the present invention provides a method for inhibiting tumor growth, comprising contacting the tumor with an effective amount of an antibody that binds to CD3, said antibody comprising each of the antibodies described herein.

In another aspect, the present invention provides a method for inhibiting tumor growth in a subject, comprising administering to the subject a therapeutically effective amount of an antibody binding to CD3, including each of those antibodies described herein.

In another aspect, the present invention provides a method for modulating angiogenesis in a subject, comprising administering to the subject a therapeutically effective amount of an antibody binding to CD3, including each of those antibodies described herein.

In another aspect, the present invention provides a method for reducing the tumorigenicity of tumors in a subject, comprising administering to the subject a therapeutically effective amount of an antibody binding to CD3, including each of those antibodies described herein.

In one aspect, the present invention provides a method for treating conditions in subjects associated with tumor antigen expression. Therefore, the present invention provides a method for treating tumor antigen-related conditions by engineering bispecific antibodies to immune-specifically recognize tumor antigens and CD3 antigens on T cells. Bispecific antibodies engineered according to the present invention can be used for the prevention or treatment of cancer because they possess cytotoxic activity via CD3 antibody-induced activation of cytotoxic T cells.

In one specific embodiment, the present invention provides a method for treating cancer. In this embodiment, the cancer is breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer (including but not limited to non-small cell lung cancer and small cell lung cancer), bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, or cancer of the digestive system. In some embodiments, the cancer of the digestive system is selected from cancers of the esophagus, stomach, small intestine, colon, rectum, anus, liver, gallbladder, appendix, bile ducts, and pancreas.

In one aspect, the present invention provides antibodies, bispecific antibodies, or pharmaceutical compositions as disclosed herein for use in a therapy. In one specific embodiment, the present invention also provides a CD3 bispecific antibody for use in a method of treating cancer as defined herein. In another specific embodiment, the therapy activates a cytolytic T-cell response.

This invention further provides antibodies or bispecific antibodies as disclosed herein for the preparation of medicaments for use in therapies. In some embodiments, the therapy is a treatment for cancer. In some embodiments, the cancer is selected from breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer (including but not limited to non-small cell lung cancer and small cell lung cancer), bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, or cancers of the digestive system. In some embodiments, the cancers of the digestive system are selected from cancers of the esophagus, stomach, small intestine, colon, rectum, anus, liver, gallbladder, appendix, bile ducts, and pancreas.

In one aspect, the present invention provides polynucleotides encoding antibodies or bispecific antibodies as disclosed herein. In another embodiment, the present invention provides a vector comprising a polynucleotide as disclosed herein. In yet another embodiment, the present invention provides a host cell comprising a vector as disclosed herein. In some such embodiments, the host cell recombinants produce antibodies or bispecific antibodies as disclosed herein. In specific embodiments, the host cell is selected from bacterial cell lines, mammalian cell lines, insect cell lines, yeast cell lines, and in vitro cell-free protein synthesis systems. In a specific embodiment, the mammalian cell line is a CHO cell line.

In one aspect, a method for treating cancer in a subject with this need is provided, comprising: a) providing a bispecific antibody as described herein, and b) administering the bispecific antibody to the subject.

In some embodiments, a method is provided for inhibiting tumor growth or progression in a subject with malignant cells expressing tumor antigens, comprising administering to a subject in need an effective amount of a composition comprising antibodies as described herein. In some embodiments, a method is provided for inhibiting cell metastasis expressing tumor antigens in a subject, comprising administering to a subject in need an effective amount of a composition comprising antibodies as described herein. In some embodiments, a method is provided for inducing tumor regression in malignant cells in a subject, comprising administering to a subject in need an effective amount of a composition comprising antibodies as described herein.

In a specific aspect, relative to the growth of cancer cells in the absence of the antibodies or bispecific antibodies of the present invention, the antibodies of the present invention inhibit or reduce the growth of cancer cells by at least 99%, at least 95%, at least 90%, at least 85%, at least 80%, at least 75%, at least 70%, at least 60%, at least 50%, at least 45%, at least 40%, at least 45%, at least 35%, at least 30%, at least 25%, at least 20%, or at least 10%.

In specific respects, the antibody kills cells or inhibits or reduces the growth of cancer cells by at least 5%, at least 10%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, or at least 100% better than the antibody or bispecific antibody of the present invention.

In one aspect, the present invention provides an effective amount of a composition (e.g., a pharmaceutical composition) comprising a bispecific antibody as described herein for treating a condition (e.g., cancer) associated with tumor antigen expression in a subject in need of this treatment.

In another aspect, the present invention provides antibodies as described herein for treating conditions (e.g., cancer) associated with tumor antigen expression in subjects with this need. In some embodiments, antibodies as described herein are provided for inhibiting tumor growth or progression in subjects with malignant cells expressing tumor antigens. In some embodiments, antibodies as described herein are provided for inhibiting the metastasis of malignant cells expressing tumor antigens in subjects with this need. In some embodiments, antibodies as described herein are provided for inducing tumor regression in subjects with malignant cells expressing tumor antigens.

In another aspect, the present invention provides the use of antibodies as described herein in the preparation of medicaments for treating conditions (e.g., cancer) associated with tumor antigen expression. In some embodiments, the use of antibodies as described herein is provided in the preparation of medicaments for inhibiting tumor growth or progression. In some embodiments, the use of antibodies as described herein is provided in the preparation of medicaments for inhibiting the metastasis of malignant cells expressing tumor antigens. In some embodiments, the use of antibodies as described herein is provided in the preparation of medicaments for inducing tumor regression.

In some embodiments, the method described herein further includes the step of treating the subject with another form of therapy. In some embodiments, the other form of therapy is an additional anticancer therapy, including, but not limited to, chemotherapy, radiation, surgery, hormone therapy, and/or additional immunotherapy.

This invention further encompasses the administration of the molecules of the invention in combination with other therapies known to those skilled in the art for the treatment or prevention of cancer, including but not limited to current standard and experimental chemotherapy, biotherapy, immunotherapy, radiotherapy, or surgery. In some aspects, the molecules of the invention may be administered in combination with therapeutically or preventively effective amounts of one or more agents, therapeutic antibodies, or other agents known to those skilled in the art for the treatment and/or prevention of cancer.

Therefore, methods for treating cancer include administering an effective amount of the multispecific antibody (e.g., a bispecific antibody) of the present invention in combination with a chemotherapy agent to a subject in need. This combination therapy can be administered alone, sequentially, or simultaneously.

The dosages and frequencies of administration provided herein are covered by the terms therapeutically effective and prophylactically effective. Depending on the specific therapeutic or prophylactic agent administered, the severity and type of cancer, the route of administration, and the subject's age, weight, response, and past medical history, dosage and frequency may further vary based on factors specific to each subject. Those skilled in the art can select an appropriate regimen by taking such factors into account and by following the dosages recommended, for example, those reported in the literature and in Physician's Desk Reference (56th edition, 2002).

Therefore, in addition to redirecting T cells to tumor-specific antigens, bispecific T-cell conjugate molecules can also be used to deliver other diagnostic or therapeutic compounds to cells expressing tumors on their surface. Thus, bispecific T-cell conjugate molecules can be directly or indirectly (e.g., via a connector) linked to drugs, enabling direct delivery to tumor-bearing cells. Therapeutic agents include compounds such as nucleic acids, proteins, peptides, amino acids or derivatives, glycoproteins, radioisotopes, lipids, carbohydrates, or recombinant viruses. Therapeutic and diagnostic nucleic acid portions include antisense nucleic acids, derivatized oligonucleotides for covalent cross-linking with single- or double-stranded DNA, and oligonucleotides forming triple strands.

Pharmaceutical Composition

The present invention further provides compositions comprising a therapeutically effective amount of the antibody disclosed herein and a pharmaceutically acceptable carrier.

The antibodies of the present invention may be in the form of pharmaceutical compositions for administration, formulated to suit a chosen mode of administration, and containing pharmaceutically acceptable diluents or excipients, such as buffers, surfactants, preservatives, solubilizers, isotonic agents, stabilizers, carriers, etc. Remington's Pharmaceutical Sciences, Mack Publishing Co., Easton Pa., 18th edition, 1995 provides an outline of formulation techniques generally known to those skilled in the art.

These pharmaceutical compositions can be administered in any manner known in the art to achieve the generally intended purpose of treating cancer. Routes of administration can be parenteral, and are defined herein as modes of administration including, but not limited to, intravenous, intramuscular, intraperitoneal, subcutaneous, and intra-articular injections and infusions. The manner of administration and dosage of the molecules according to the invention (e.g., antibodies, pharmaceutical compositions) depends on the type of disease to be combated, its stage in appropriate circumstances, the antigen to be controlled, the type of coexisting treatment (if any), the frequency of treatment, the characteristics of the desired effect, and also the subject's weight, age, health status, diet, and sex. Therefore, the actual dosage regimens used can vary considerably and may therefore deviate from the dosage regimens stated herein.

Various formulations of the antibodies of this invention can be used for administration. In some embodiments, the antibodies can be administered in their pure form. In some embodiments, the antibodies and pharmaceutically acceptable excipients can be contained in various formulations. Pharmaceutically acceptable excipients are known in the art and are relatively inert substances that facilitate the administration of pharmacologically effective substances. For example, excipients can impart shape or consistency or act as diluents. Suitable excipients include, but are not limited to, stabilizers, wetting agents and emulsifiers, salts for altering molar osmotic pressure concentrations, encapsulation agents, buffers, and skin penetration enhancers. Excipients, as well as formulations for parenteral and non-parenteral drug delivery, are described in Remington, The Science and Practice of Pharmacy, 21st edition. Mack Publishing, 2005.

In some implementations, these agents are formulated for administration by injection (e.g., intraperitoneal, intravenous, subcutaneous, intramuscular, etc.). Therefore, these agents can be combined with pharmaceutically acceptable mediators, such as saline, Ringer's solution, glucose solution, etc. Specific dosing regimens, i.e., dosage, timing, and repetition, will depend on the specific individual and their medical history.

The antibodies described herein can be administered using any suitable method, including by injection (e.g., intraperitoneal, intravenous, subcutaneous, intramuscular, etc.). As described herein, antibodies, such as monoclonal or bispecific antibodies, can also be administered by inhalation. Generally, the dosage for administration of the antibodies of the present invention depends on the host being treated and the specific administration method. In one embodiment, the dosage range of the antibodies of the present invention will be from about 0.001 µg/kg body weight to about 20,000 µg/kg body weight. The term “body weight” may be used when treating a patient. When processing isolated cells, “body weight” as used herein refers to “total cell weight”. The term “total weight” can be used for both isolated cells and patient treatment. In this application, all concentrations and treatment levels are expressed as “body weight” or simply “kg”, and are also considered to include similar “total cell weight” and “total weight” concentrations. However, those skilled in the art will recognize the practicality of a variety of dosage ranges, such as 0.01 µg/kg body weight to 20,000 µg/kg body weight, 0.02 µg/kg body weight to 15,000 µg/kg body weight, 0.03 µg/kg body weight to 10,000 µg/kg body weight, 0.04 µg/kg body weight to 5,000 µg/kg body weight, 0.05 µg/kg body weight to 2,500 µg/kg body weight, 0.06 µg/kg body weight to 1,000 µg/kg body weight, 0.07 µg/kg body weight to 500 µg/kg body weight, 0.08 µg/kg body weight to 400 µg/kg body weight, 0.09 µg/kg body weight to 200 µg/kg body weight, or 0.1 µg/kg body weight to 100 µg/kg body weight. Furthermore, those skilled in the art will recognize that a variety of different dosage levels will be useful, for example, 0.0001 µg/kg, 0.0002 µg/kg, 0.0003 µg/kg, 0.0004 µg/kg, 0.005 µg/kg, 0.0007 µg/kg, 0.001 µg/kg, 0.1 µg/kg, 1.0 µg/kg, 1.5 µg/kg, 2.0 µg/kg, 5.0 µg/kg, 10.0 µg/kg, 15.0 µg/kg, 30.0 µg/kg, 50 µg/kg, 75 µg/kg, 80 µg/kg, 90 µg/kg, 100 µg/kg, 120 µg/kg, 140 µg/kg. g/kg, 150 µg/kg, 160 µg/kg, 180 µg/kg, 200 µg/kg, 225 µg/kg, 250 µg/kg, 2 75µg/kg, 300 µg/kg, 325 µg/kg, 350 µg/kg, 375µg/kg, 400 µg/kg, 450 µg/k g, 500 µg/kg, 550 µg/kg, 600 µg/kg, 700 µg/kg, 750 µg/kg, 800 µg/kg, 900 µg/kg, 1 µg/kg, 5 µg/kg, 10 µg/kg, 12 µg/kg, 15 mg/kg, 20 mg/kg and/or 30 mg/kg. All these dosages are exemplary, and any dosage between these points is also expected to be useful in this invention. Any of the above dosage ranges or dosage levels can be used with the antibodies of this invention. For repeated administration over a period of several days or longer, depending on the condition, treatment may continue until the desired symptom suppression is achieved or until a sufficient therapeutic level is reached, for example, to inhibit or delay tumor growth/progression or metastasis of cancer cells.

Depending on the desired pattern of pharmacokinetic decay, other dosing regimens may also be useful. In one embodiment, the antibody of the present invention is administered at an initial initiating dose followed by higher and/or continuous, substantially constant doses. In some embodiments, dosing is anticipated from once weekly to four times weekly. In some embodiments, dosing is anticipated from once monthly, every other month, or every three months. Progression of the therapy can be easily monitored using routine techniques and assays. The dosing regimen may vary over time.

For the purposes of this invention, the appropriate dosage of the antibody will depend on the antibody or a combination thereof used, the type and severity of the symptoms to be treated, whether the agent is administered for therapeutic purposes, prior therapy, the patient's clinical history and response to the agent, the patient's clearance rate of the administered agent, and the physician's judgment. Typically, the physician will administer the antibody until a dosage is reached to achieve the desired outcome. The dosage and/or frequency may vary during the treatment process. Empirical considerations such as half-life often aid in dosage determination. For example, antibodies compatible with the human immune system, such as humanized antibodies or fully human antibodies, can be used to prolong the antibody's half-life and prevent the antibody from being attacked by the host immune system. The frequency of administration can be determined and adjusted during the treatment process and is generally, but not necessarily, based on the treatment and/or suppression and/or improvement and/or delay of symptoms, such as tumor growth inhibition or delay. Alternatively, a sustained-release formulation of the antibody may be suitable. Various formulations and devices for achieving sustained release are known in the art.

In one implementation, the antibody dosage can be empirically determined in individuals who have received one or more doses of the antibody. Incremental doses of the antibody are administered to the individuals. Indicators of the disease can be tracked to assess efficacy.

In some implementations, the administration of the antibody results in at least one effect selected from the following: inhibition of tumor growth, tumor regression, reduction of tumor size, reduction of tumor cell number, delay of tumor growth, abscopal effect, inhibition of tumor metastasis, reduction of metastatic lesions over time, reduction of use of chemotherapy agents or cytotoxic agents, reduction of tumor burden, increase of progression-free survival, increase of overall survival, complete response, partial response, and stable disease.

Depending on factors such as the recipient's physiological condition, whether the purpose of administration is therapeutic or prophylactic, and other factors known to a skilled professional, the administration of antibodies according to the method of the invention can be continuous or intermittent. Antibody administration can be substantially continuous over a preselected time period, or it can be a series of doses spaced apart from each other.

In some implementations, more than one antibody may be present. At least one, at least two, at least three, at least four, at least five, or more different antibodies may be present. Typically, these antibodies may have complementary activities that do not harmfully interfere with each other. For example, one or more of the following antibodies may be used: a first CD3 antibody targeting one epitope on CD3 and a second CD3 antibody targeting a different epitope on CD3.

In some embodiments, the antibodies disclosed herein can be used, in particular, for parenteral administration, such as intravenous administration or administration into the cavities of body cavities or organs.

Therapeutic formulations of the antibodies used according to the invention are prepared for storage by mixing antibodies of desired purity with optional pharmaceutically acceptable carriers, excipients, or stabilizers (Remington, The Science and Practice of Pharmacy, 21st edition. Mack Publishing, 2005) in the form of lyophilized formulations or aqueous solutions. Acceptable carriers, excipients, or stabilizers are non-toxic to the recipient at the doses and concentrations used and may include buffers such as phosphates, citrates, and other organic acids; salts such as sodium chloride; antioxidants, including ascorbic acid and methionine; preservatives (e.g., octadecyl dimethyl benzyl ammonium chloride; hexane diammonium chloride; benzalkonium chloride, benzyl chloride; phenols, butanol, or benzyl alcohol; alkyl esters of p-hydroxybenzoate, such as methylparaben or propylparaben; catechol; resorcinol; cyclohexanol; 3-pentanol; and m-cresol); low molecular weight (small) The following are considered as salts: polypeptides (approximately 10 residues); proteins, such as serum albumin, gelatin, or immunoglobulins; hydrophilic polymers, such as polyvinylpyrrolidone; amino acids, such as glycine, glutamine, asparagine, histidine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates, including glucose, mannose, or dextrin; chelating agents, such as EDTA; sugars, such as sucrose, mannitol, trehalose, or sorbitol; salt-forming ions, such as sodium; metal complexes (e.g., Zn-protein complexes); and/or nonionic surfactants, such as TWEEN ^™ , PLURONICS ^™ , or polyethylene glycol (PEG).

Antibody-containing liposomes are prepared by methods known in the art, for example, as described in Epstein et al., Proc. Natl. Acad. Sci. USA 82:3688, 1985; Hwang et al., Proc. Natl. Acad. Sci. USA 77:4030, 1980; and U.S. Patent Nos. 4,485,045 and 4,544,545. Liposomes with extended cycle times are disclosed in U.S. Patent No. 5,013,556. Particularly useful liposomes can be produced by a reverse-phase evaporation method using a lipid composition comprising phosphatidylcholine, cholesterol, and PEG-derived phosphatidylethanolamine (PEG-PE). The liposomes are extruded through a filter having a defined pore size to produce liposomes with a desired diameter.

The active ingredient can also be encapsulated in microcapsules (e.g., hydroxymethyl cellulose or gelatin microcapsules and poly-(methyl methacrylate) microcapsules) prepared by agglomeration technology or interfacial polymerization, colloidal drug delivery systems (e.g., liposomes, albumin microspheres, microemulsions, nanoparticles and nanocapsules) or coarse droplet emulsions. This technology is disclosed in Remington, The Science and Practice of Pharmacy, 21st edition, Mack Publishing, 2005.

Sustained-release formulations can be prepared. Suitable examples of sustained-release formulations include semi-permeable matrices of solid hydrophobic polymers containing antibodies, said matrices being in the form of shaped articles, such as membranes or microcapsules. Examples of sustained-release matrices include polyesters, hydrogels (e.g., poly(2-hydroxyethyl methacrylate) or poly(vinyl alcohol)), polylactides (US Patent No. 3,773,919), copolymers of L-glutamic acid and L-glutamic acid 7-ethyl ester, non-degradable ethylene vinyl acetate, degradable lactic acid-glycolic acid copolymers, such as LUPRON DEPOT ^™ (injectable microspheres composed of lactic acid-glycolic acid copolymers and leuprolide acetate), sucrose isobutyrate acetate, and poly-D-(-)-3-hydroxybutyrate.

Preparations intended for in vivo administration must be sterile. This can be easily achieved, for example, by filtration through a sterile membrane. Therapeutic antibody (e.g., CD3 antibody) compositions are typically placed in containers with sterile inlets and outlets, such as intravenous solution bags or vials with stoppers that can be punctured by a subcutaneous needle.

The compositions according to the invention may be in unit dosage form, such as tablets, pills, capsules, powders, granules, solutions or suspensions, or suppositories, for oral, parenteral or rectal administration, or administration by inhalation or blowing.

To prepare solid compositions such as tablets, the main active ingredient is mixed with a pharmaceutical carrier (e.g., conventional tableting ingredients such as corn starch, lactose, sucrose, sorbitol, talc, stearic acid, magnesium stearate, dicalcium phosphate, or gum, and other pharmaceutical diluents, such as water) to form a solid preformulation composition containing a homogeneous mixture of the compound of the present invention or a pharmaceutically acceptable, non-toxic salt thereof. When these preformulation compositions are referred to as homogeneous, it means that the active ingredient is uniformly dispersed throughout the composition, making it easy to subdivide the composition into equivalent effective unit dose forms, such as tablets, pills, and capsules. The solid preformulation composition is then subdivided into unit dose forms of the type described above, containing 0.1 to about 500 mg of the active ingredient of the present invention. Tablets or pills of the novel composition can be coated or otherwise compounded to provide a dosage form that provides the advantage of prolonged action. For example, tablets or pills may contain an internal dose component and an external dose component, the latter being in the form of an outer coating on top of the former. These two components can be separated by an enteric layer, which resists breakdown in the stomach and allows the internal components to enter the duodenum intact or with delayed release. A variety of materials can be used for this enteric layer or coating, including many polymeric acids and mixtures of polymeric acids with materials such as shellac, cetyl alcohol, and cellulose acetate.

Suitable surfactants particularly include nonionic agents such as polyoxyethylene sorbitol (e.g., Tween ^™ 20, 40, 60, 80, or 85) and other sorbitols (e.g., Span ^™ 20, 40, 60, 80, or 85). Compositions containing surfactants will conveniently contain 0.05 to 5% surfactant, and may be 0.1 to 2.5%. It will be understood that, where necessary, other ingredients, such as mannitol or other pharmaceutically acceptable mediators, may be added.

The compositions within the scope of this invention include all compositions in which the antibody is present in an amount that effectively achieves the desired medical effect for treating cancer. Although individual needs may differ from one patient to another, the determination of the optimal range of effective amounts of all components is within the capabilities of a clinician with ordinary skills.

Examples of such compositions and how they are formulated have also been described in earlier sections and below. In some embodiments, the composition comprises one or more antibodies. In some embodiments, the antibody is a human antibody, a humanized antibody, or a chimeric antibody. In some embodiments, the antibody comprises a constant region capable of triggering a desired immune response (e.g., antibody-mediated cleavage or ADCC). In some embodiments, the antibody comprises a constant region that does not trigger unwanted or undesirable immune responses (e.g., antibody-mediated cleavage or ADCC).

It should be understood that the composition may contain more than one antibody, such as a mixture of CD3 antibodies that recognize CD3 or different epitopes of CD3 and tumor antigens. Other exemplary compositions contain more than one antibody that recognizes one or more of the same epitopes, or different kinds of antibodies, or different epitopes of CD3 and tumor antigens (e.g., human CD3).

In some embodiments, the antibody may be administered in combination with the administration of one or more additional therapeutic agents. These include, but are not limited to, the administration of biological and/or chemotherapeutic agents, such as, but not limited to, vaccines, CAR-T cell-based therapies, radiotherapy, cytokine therapy, CD3 bispecific antibodies, inhibitors of other immunosuppressive pathways, inhibitors of angiogenesis, T-cell activators, inhibitors of metabolic pathways, mTOR inhibitors, inhibitors of the adenosine pathway, tyrosine kinase inhibitors, including but not limited to Inlyta, ALK inhibitors and sunitinib, BRAF inhibitors, epigenetic modifying factors, IDO1 inhibitors, JAK inhibitors, STAT inhibitors, and cell cycle proteins. Leukocyte-dependent kinase inhibitors, biotherapeutic agents (including but not limited to antibodies against VEGF, VEGFR, EGFR, Her2/neu, other growth factor receptors, CD40, CD-40L, CTLA-4, OX-40, 4-1BB, TIGIT, and ICOS), immunogenic agents (e.g., attenuated cancer cells, tumor antigens, antigen-presenting cells such as dendritic cells bombarded with tumor-derived antigens or nucleic acids, immunostimulatory cytokines (e.g., IL-2, IFNα2, GM-CSF), and cells transfected with genes encoding immunostimulatory cytokines (e.g., but not limited to GM-CSF).

Examples of biotherapeutic agents include therapeutic antibodies, immunomodulators, and therapeutic immune cells.

Therapeutic antibodies may be specific to a variety of different antigens. For example, a therapeutic antibody may target a tumor-associated antigen, such that the binding of the antibody to the antigen promotes the death of cells expressing the antigen. In other instances, a therapeutic antibody may target an antigen on immune cells (e.g., PD-1), such that the binding of the antibody prevents the downregulation of the activity of cells expressing the antigen (and thereby promotes the activity of cells expressing the antigen). In some cases, a therapeutic antibody may act through a variety of different mechanisms (e.g., upon contact with cells expressing the antigen, it may both i) promote the death of cells expressing the antigen and ii) prevent the downregulation of immune cell activity caused by the antigen).

Therapeutic antibodies can target antigens such as those listed below. For some antigens, exemplary antibodies against said antigens are also included below (in parentheses after the antigen). The antigens listed below may also be referred to herein as "target antigens," etc. The target antigens used for therapeutic antibodies in this article include, for example: 4-1BB (e.g., utomilumab); 5T4; A33; α-folate receptor 1 (e.g., mirvetuximab soravtansine); Alk-1; BCMA [e.g., PF-06863135 (see US9969809)]; BTN1A1 (e.g., see WO2018222689); CA-125 (e.g., abagovomab); carbonic anhydrase IX; CCR2; CCR4 (e.g., mogamulizumab); CCR5 (e.g., leronlimab); CCR8; CD3 [e.g., blinatumomab (CD3/ CD19 bispecific), PF-06671008 (CD3/P-cadherin bispecific), PF-06863135 (CD3/BCMA bispecific), PF-07062119 (CD3/GUCY2c bispecific), CD19 (e.g., ibritumomab tiuxetan, MOR208); CD20 (e.g., ibritumomab tiuxetan, obinutuzumab, ofatumumab, rituximab, ublituximab); CD22 (inotuzumab ozogamicin, moxetumomab pasudotox); CD25; CD28; CD30 (e.g., brentuximab vedotin); CD33 (e.g., gemtuzumab ozogamicin); CD38 (e.g., daratumumab, isatuximab); CD40; CD-40L; CD44v6; CD47; CD52 (e.g., alemtuzumab); CD63; CD79 (e.g., polotuzumab vedotin); CD80; CD123; CD276/B7-H3 (e.g., omburtamab); CDH17; CEA; ClhCG; CTLA-4 (e.g., ipilimumab, tremel... imumab), CXCR4; desmosome core protein 4; DLL3 (e.g., rovalpituzumab tesirine); DLL4; E-cadherin; EDA; EDB; EFNA4; EGFR (e.g., cetuximab, depatuxizumab mafodotin, necitumumab, panitumumab); EGFRvIII; endothelial sialic acid protein; EpCAM (e.g., oportuzumab monatox); FAP; fetal acetylcholine receptor; FLT3 (e.g., see WO2018/220584); GD2 (e.g., dinutuximab, 3F8); GD3 ; GITR; GloboH; GM1; GM2; GUCY2C (e.g., PF-07062119); HER2/neu [e.g., margetuximab, pertuzumab, trastuzumab; ado-trastuzumab emtansine, trastuzumab duocarmazine, PF-06804103 (see US8828401)]; HER3; HER4; ICOS; IL-10; ITG-AvB6; LAG-3 (e.g., relatlimab); Lewis-Y; LG; Ly-6; M-CSF [e.g., PD-0360324 (see US8828401)]; HER3; HER4; ICOS; IL-10; ITG-AvB6; LAG-3 (e.g., relatlimab); Lewis-Y; LG; Ly-6; M-CSF [e.g., PD-0360324 (see US8828401)]; S7326414); MCSP; mesothelin; MUC1; MUC2; MUC3; MUC4; MUC5AC; MUC5B; MUC7; MUC16; Notch 1; Notch 3; Nectin-4 (e.g., enfortumab vedotin); OX40 [e.g., PF-04518600 (see US7960515)]; P-cadherein [e.g., PF-06671008 (see WO2016/001810)]; PCDHB2; PD-1 [e.g., BCD-100, camrelizumab, cemiplimab, genolimab] Nivolumab (CBT-501), MEDI0680 (CBT-501), nivolumab, pembrolizumab, RN888 (see WO2016/092419), sintilimab, spartalizumab, STI-A1110, tislelizumab, TSR-042; PD-L1 (e.g., atezolizumab, durvalumab, BMS-936559 (MDX-1105) or LY3300054); PDGFRA (e.g., olaratumab); plasma cell antigen; PolySA; PSCA; PSMA; PTK7 [e.g., PF-06647020 (see US9409995)]; Ror1; SAS; SCRx6; SLAMF7 (e.g., elotuzumab); SHH; SIRPa (e.g., ED9, Effi-DEM); STEAP; TGF-β; TIGIT; TIM-3; TMPRSS3; TNF-α precursor; TROP-2 (e.g., sacituzumabgovitecan); TSPAN8; VEGF (e.g., bevacizumab, brolucizumab); VEGFR1 (e.g., ranibizumab); VEGFR2 (e.g., ramucirumab, ranibizumab); Wue-1.

Therapeutic antibodies can take any suitable form. For example, therapeutic antibodies can take any form described elsewhere herein. In some embodiments, the therapeutic antibody may be a naked antibody. In some embodiments, the therapeutic antibody may be conjugated to a drug/pharmaceutical agent (also known as an "antibody-drug conjugate" (ADC)). In some embodiments, a therapeutic antibody targeting a specific antigen may be incorporated into a multispecific antibody (e.g., a bispecific antibody).

In some embodiments, antibodies against antigens may be conjugated to drugs/pharmaceuticals. The linked antibody-drug molecule is also referred to as an "antibody-drug conjugate" (ADC). The drug/pharmaceutical can be linked to the antibody directly or indirectly via a linker. Most commonly, toxic drugs are linked to antibodies such that the binding of the ADC to the corresponding antigen promotes the killing of cells expressing said antigen. For example, ADCs linked to toxic drugs are particularly useful for targeting tumor-associated antigens to promote the killing of tumor cells expressing tumor-associated antigens. In other embodiments, the agent that can be linked to the antibody may be, for example, an immunomodulator (e.g., to modulate the activity of immune cells near the ADC), an imaging agent (e.g., to facilitate imaging of the ADC in a subject or a biological sample from a subject), or an agent that increases the serum half-life or biological activity of the antibody.

Methods for conjugating cytotoxic agents or other therapeutic agents to antibodies have been described in various publications. For example, antibodies can be chemically modified to induce conjugation by lysine side-chain amines or by cysteine thiol groups activated by reduced interchain disulfide bonds. See, for example, Tanaka et al., FEBS Letters 579:2092-2096, 2005 and Gentle et al., Bioconjugate Chem. 15:658-663, 2004. Reactive cysteine residues engineered at specific sites on antibodies for conjugation to specific drugs at defined stoichiometric amounts have also been described. See, for example, Junutula et al., Nature Biotechnology, 26:925-932, 2008. International applications WO2012/059882 and WO2015015448 also describe the use of glutamine-containing tags with acyl donors or endogenous glutamine conjugated by peptide engineering to an active state (i.e., the ability to form covalent bonds as acyl donors) in the presence of transglutaminase and amines (e.g., cytotoxic agents containing or attached to active amines). In some embodiments, the ADC may have any of the features or characteristics of the ADCs provided in WO2016166629, which is incorporated herein by reference for all purposes.

Drugs/pharmaceuticals that can be linked to antibodies in the form of ADCs can include, for example, cytotoxic agents, immunomodulators, imaging agents, therapeutic proteins, biopolymers, or oligonucleotides.

Exemplary cytotoxic agents that can be incorporated into ADCs include anthracyclines, auristatin, dolastatin, compbretastatin, duocarmycin, pyrrolobenzodiazepine dimer, indoline-benzodiazepine dimer, enediyne, geldemycin, maytansine, puromycin, taxane, vinca alkaloids, camptothecin, tubulysin, hemiasterlin, spliceostatin, pladienolide, and their stereoisomers, isomers, analogs, or derivatives.

Exemplary immunomodulators that can be incorporated into ADCs include gancyclovier, etanercept, tacrolimus, sirolimus, voclosporin, cyclosporine, rapamycin, cyclophosphamide, azathioprine, mycophenolate mofetil, methotrexate, glucocorticoids and their analogues, cytokines, and stem cell production. Growth factors, lymphotoxin, tumor necrosis factor (TNF), hematopoietic factors, interleukins (e.g., interleukin-1 (IL-1), IL-2, IL-3, IL-6, IL-10, IL-12, IL-18, and IL-21), colony-stimulating factors (e.g., granulocyte colony-stimulating factor (G-CSF) and granulocyte-macrophage colony-stimulating factor (GM-CSF)), interferons (e.g., interferon-α, -β, and -γ), stem cell growth factors known as "S1 factors", erythropoietin, and thrombopoietin, or combinations thereof.

Exemplary imaging agents that may be included in an ADC include fluorescein, rhodamine, lanthanide phosphors and their derivatives, or radioisotopes bound to a chelating agent. Examples of fluorophores include, but are not limited to, fluorescein isothiocyanate (FITC) (e.g., 5-FITC), fluorescein phosphoramide (FAM) (e.g., 5-FAM), eosin, carboxyfluorescein, erythrosine, AlexaFluor.RTM. (e.g., Alexa 350, 405, 430, 488, 500, 514, 532, 546, 555, 568, 594, 610, 633, 647, 660, 680, 700, or 750), carboxytetramethylrhodamine (TAMRA) (e.g., 5-TAMRA), tetramethylrhodamine (TMR), and sulforhodamine (SR) (e.g., SR101). Examples of chelating agents include, but are not limited to, 1,4,7,10-tetraazacyclododecane-N,N',N'',N'''-tetraacetic acid (DOTA), 1,4,7-triazacyclononane-1,4,7-triacetic acid (NOTA), 1,4,7-triazacyclononane, 1-glutaric acid-4,7-acetic acid (deferroamine), diethylenetriaminepentaacetic acid (DTPA), and 1,2-bis(o-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA).

Exemplary therapeutic proteins that may be included in ADCs include toxins, hormones, enzymes, and growth factors.

Exemplary biocompatible polymers that can be incorporated into ADCs include water-soluble polymers, such as polyethylene glycol (PEG) or derivatives thereof, and biocompatible polymers containing zwitterions (e.g., polymers containing phosphocholine).

Exemplary biocompatible polymers that can be incorporated into ADCs include antisense oligonucleotides.

In some implementations, the antibodies against antigens provided herein can be incorporated into bispecific antibody molecules. A bispecific antibody is a monoclonal antibody that has binding specificity to at least two different antigens.

In some embodiments, the bispecific antibody comprises a first antibody variable domain and a second antibody variable domain, wherein the first antibody variable domain is capable of recruiting human immune effector cells by specifically binding to CD3 as provided herein, and wherein the second antibody variable domain is capable of specifically binding to a target antigen. In some embodiments, the antibody has an IgG1, IgG2, IgG3, or IgG4 isotype. In some embodiments, the antibody comprises an immune-inert Fc region. In some embodiments, the antibody is a human antibody or a humanized antibody.

Target antigens are typically expressed on target cells (e.g., cancer cells) in disease conditions. Examples of target antigens of particular interest for bispecific antibodies include, but are not limited to, BCMA, EpCAM (epithelial cell adhesion molecule), CCR5 (type 5 chemokine receptor), CD19, HER (human epidermal growth factor receptor)-2/neu, HER-3, HER-4, EGFR (epidermal growth factor receptor), PSMA, CEA, MUC-1 (mucin), MUC2, MUC3, MUC4, MUC5AC, MUC5B, MUC7, CIhCG, Lewis-Y, CD20, CD33, CD30, ganglioside GD3, 9-O-acetyl-GD3, GM2, Globo H, fucose GM1, Poly SA, GD2, carbonic anhydrase IX (MN/CA IX), CD44v6, and Shh (Sonic Hedgkinase). ehog), Wue-1, plasma cell antigen, (membrane-bound) IgE, MCSP (chondroitin sulfate proteoglycan for melanoma), CCR8, TNF-α precursor, STEAP, mesothelin, A33 antigen, PSCA (prostate stem cell antigen), Ly-6; desmosome core glycoprotein 4, E-cadherin epitope, fetal acetylcholine receptor, CD25, CA19-9 marker, CA-125 marker and MIS (Muellerian Inhibitory Substance) type II receptor, sTn (sialylated Tn antigen; TAG-72), FAP (fibroblast activation antigen), endothelial sialic acid protein, EGFRvIII, LG, SAS, PD-L1, CD47, SIRPa and CD63.

In some embodiments, the bispecific antibody comprises a full-length human antibody, wherein the first antibody variable domain of the bispecific antibody is capable of recruiting human immune effector cells by specifically binding to CD3, and wherein the second antibody variable domain of the heterodimeric protein is capable of specifically binding to a target antigen (e.g., CD20, EpCAM, or P-cadherin).

Methods for preparing bispecific antibodies are known in the art (see, for example, Suresh et al., Methods in Enzymology 121:210, 1986). Traditionally, recombinant production of bispecific antibodies is based on the co-expression of two immunoglobulin heavy-light chain pairs, where the two heavy chains have different specificities (Millstein and Cuello, Nature 305, 537-539, 1983).

According to a method for preparing bispecific antibodies, an antibody variable domain having the desired binding specificity (antibody-antigen combination site) is fused to an immunoglobulin constant region sequence. The fusion preferably has an immunoglobulin heavy chain constant region comprising at least a portion of the hinge, CH2, and CH3 regions. Preferably, it has a first heavy chain constant region (CH1) present in at least one of the fusions, containing the site necessary for light chain binding. DNA encoding the immunoglobulin heavy chain fusion and (if desired) the immunoglobulin light chain is inserted into separate expression vectors and co-transfected into a suitable host organism. This provides great flexibility in the embodiments to adjust the relative proportions of the three polypeptide fragments when unequal ratios of the three polypeptide chains used in the construction provide optimal yield. However, when expression of at least two polypeptide chains in equal ratios results in high yield, or when the ratios are not particularly significant, it is possible to insert coding sequences for two or all three polypeptide chains into a single expression vector.

In one method, the bispecific antibody comprises a hybrid immunoglobulin heavy chain having a first binding specificity in one arm and a hybrid immunoglobulin heavy chain-light chain pair (providing a second binding specificity) in the other arm. This asymmetric structure, with the immunoglobulin light chain in only half of the bispecific molecule, facilitates the separation of the desired bispecific compound from the unwanted combination of immunoglobulin chains. This method is described in PCT Publication No. WO 94/04690.

In another approach, the bispecific antibody is constructed by modifying an amino acid in a first hinge region of one arm, wherein the substituted/alternated amino acid in the first hinge region has an opposite charge to the corresponding amino acid in a second hinge region of the other arm. This approach is described in International Patent Application No. PCT/US2011/036419 (WO2011/143545).

In another approach, the formation of a desired heteromeric or heterodimeric protein (e.g., a bispecific antibody) is enhanced by altering or engineering the interface between the first and second immunoglobulin-like Fc regions (e.g., hinge regions and/or CH3 regions). In this approach, the bispecific antibody may be composed of a CH3 region containing a first and a second CH3 polypeptide that interact together to form a CH3 interface, wherein one or more amino acids within the CH3 interface destabilize homodimer formation and are electrostatically unfavorable to homodimer formation. This approach is described in International Patent Application No. PCT/US2011/036419 (WO2011/143545).

In another approach, bispecific antibodies can be generated in the presence of transglutaminase using a glutamine-containing peptide tag engineered for an antibody targeting an epitope (e.g., BCMA) in one arm, and another peptide tag engineered for a second antibody targeting a second epitope in the other arm (e.g., a Lys-containing peptide tag or reactive endogenous Lys). This approach is described in International Patent Application No. PCT/IB2011/054899 (WO2012/059882).

In some embodiments, the first and second antibody variable domains of the bispecific antibody include amino acid modifications at various positions, wherein the first and second antibody variable domains of the bispecific antibody include amino acid modifications at positions 223, 225, and 228 in the hinge region of human IgG2 (e.g., (C223E or C223R), (E225R), and (P228E or P228R)) and amino acid modifications at positions 409 or 368 in the CH3 region (e.g., K409R or L368E (EU numbering scheme)).

In some embodiments, the first and second antibody variable domains of the bispecific antibody include amino acid modifications at positions 221 and 228 in the hinge region of human IgG1 (e.g., (D221R or D221E) and (P228R or P228E)) and amino acid modifications at positions 409 or 368 in the CH3 region (e.g., K409R or L368E (EU numbering scheme)).

In some embodiments, the first and second antibody variable domains of the bispecific antibody include an amino acid modification at position 228 in the hinge region of human IgG4 (e.g., (P228E or P228R)) and an amino acid modification at position 409 or 368 in the CH3 region (e.g., R409 or L368E (EU numbering scheme)).

In some implementations, the bispecific antibody may have any of the features or properties of any bispecific antibody provided in WO2016166629, which is incorporated herein by reference for all purposes.

Immunomodulators include a variety of different molecular types that can stimulate immune responses in subjects, such as pattern recognition receptor (PRR) agonists, immunostimulatory cytokines, and cancer vaccines.

Pattern recognition receptors (PRRs) are receptors expressed by cells of the immune system that recognize a variety of molecules associated with pathogens and/or cell damage or death. PRRs are involved in both innate and adaptive immune responses. PRR agonists can be used to stimulate immune responses in subjects. Several classes of PRR molecules exist, including toll-like receptors (TLRs), RIG-I-like receptors (RLRs), nucleotide-binding oligomerization domain (NOD)-like receptors (NLRs), C-type lectin receptors (CLRs), and interferon gene stimulator (STING) proteins.

The terms “TLR” and “toll-like receptor” refer to any toll-like receptor. Toll-like receptors are receptors involved in activating immune responses. TLRs recognize, for example, pathogen-associated molecular patterns (PAMPs) expressed in microorganisms, and endogenous damage-associated molecular patterns (DAMPs) released from dying or dying cells.

Molecules that activate TLRs (and thus activate the immune response) are referred to herein as “TLR agonists.” TLR agonists can include, for example, small molecules (e.g., organic molecules with a molecular weight of less than about 1000 Daltons) as well as large molecules (e.g., oligonucleotides and proteins). Some TLR agonists are specific to a single type of TLR (e.g., TLR3 or TLR9), while others activate two or more types of TLRs (e.g., both TLR7 and TLR8).

The exemplary TLR agonists provided herein include agonists of TLR2, TLR3, TLR4, TLR5, TLR6, TLR7, TLR8, and TLR9.

Exemplary small molecule TLR agonists include, for example, U.S. Patent Nos. 4,689,338; 4,929,624; 5,266,575; 5,268,376; 5,346,905; 5,352,784; 5,389,640; 5,446,153; 5,482,936; 5,756,747; 6,110,929; 6,194,425; 6,331,539; 6,376,669; 6,451,810; 6,525,064; 6,541,485; 6,545,016; 6,545,017; 6,573,273; 6,656,938; 6,660,735; 6 660,747; 6,664,260; 6,664,264; 6,664,265; 6,667,312; 6,670,372; 6,677,347; 6,677,348; 6,677,349; 6,683,088; 6,756,382; 6,797,718; 6,818,650; and 7,7091,214; U.S. Patent Publications 2004/0091491, 2004/0176367 and 2006/0100229; and International Publications WO 2005/18551, WO2005/18556, WO 2005/20999, WO 2005/0 32484, WO 2005/048933, WO 2005/048945, WO 2005/051317, WO2005/051324, WO 2005/066169, WO 2005/066170, WO 2005/066172, WO 2005/076783, WO 2005/079195, WO2005/094531, WO 2005/123079, WO 2005/123080, WO 2006/009826, WO 2006/009832, WO 2006/026760, WO2006/028451 WO 2006/028545、WO 2006/028962、WO 2006/029115、WO 2006/038923、WO 2006/065280、WO2006/074003、WO 2006/083440、WO 2006/086449、WO 2006/091394, WO 2006/086633, WO 2006/086634, WO 2006/091567, WO 2006/091568, WO 2006/091647, WO 2006/093514 and WO 2006/098852.

Additional examples of small molecule TLR agonists include certain purine derivatives (such as those described in U.S. Patent Nos. 6,376,501 and 6,028,076), certain imidazoquinoline amide derivatives (such as those described in U.S. Patent No. 6,069,149), certain imidazopyridine derivatives (such as those described in U.S. Patent No. 6,518,265), and certain benzimidazole derivatives (such as those described in U.S. Patent No. 6,387,938). Certain derivatives of 4-aminopyrimidine fused with a five-membered nitrogen-containing heterocycle (e.g., adenine derivatives described in U.S. Patent Nos. 6,376,501, 6,028,076, and 6,329,381; and WO 02/08905) and certain 3-β-D-furanoribosylthiazo[4,5-d]pyrimidine derivatives (e.g., those described in U.S. Publication No. 2003/0199461) and certain small molecule immune enhancer compounds, such as those described in U.S. Patent Publication No. 2005/0136065.

Exemplary macromolecular TLR agonists comprise oligonucleotide sequences. Some TLR agonist oligonucleotide sequences contain cytosine-guanine dinucleotides (CpG) and are described, for example, in U.S. Patent Nos. 6,194,388; 6,207,646; 6,239,116; 6,339,068; and 6,406,705. Some CpG-containing oligonucleotides may include synthetic immunomodulatory structural motifs, such as those described, for example, in U.S. Patent Nos. 6,426,334 and 6,476,000. Other TLR agonist nucleotide sequences lack the CpG sequence and are described, for example, in International Patent Publication No. WO 00/75304. Other TLR agonist nucleotide sequences include single-stranded RNAs (ssRNAs) rich in guanosine and uridine, such as those described, for example, in Heil et al., Science, vol. 303, pp. 1526-1529, March 5, 2004.

Other TLR agonists include biomolecules such as aminoalkyl glucosinolate phosphates (AGPs), and are described, for example, in U.S. Patent Nos. 6,113,918; 6,303,347; 6,525,028; and 6,649,172.

TLR agonists also include inactivated pathogens or fractions thereof, which can activate a variety of different types of TLR receptors. Exemplary pathogen-derived TLR agonists include BCG, mycobacterium obuense extract, Talimogene laherparepvec (T-Vec) (derived from HSV-1), and Pexa-Vec (derived from vaccinia virus).

In some implementations, the TLR agonist may be an agonist antibody that specifically binds to a TLR.

The following provides a brief description of various TLRs and TLR agonists. The list of TLR agonists for a particular TLR below should not be interpreted as indicating that a given TLR agonist necessarily activates only that TLR (e.g., some molecules can activate multiple types of TLRs, or even multiple classes of PRRs). For example, some molecules provided below as exemplary TLR4 agonists could also be TLR5 agonists.

The terms “TLR1” and “toll-like receptor 1” refer to any form of TLR1 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR1 activity. Unless otherwise specified, such as by way of a specific reference to human TLR1, TLR1 includes the naturally occurring sequence TLR1 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR1 is provided under UniProt entry number Q15399.

As used herein, “TLR1 agonist” means any molecule that, upon binding to TLR1, (1) stimulates or activates TLR1, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR1, or (3) enhances, increases, promotes, or induces the expression of TLR1. TLR1 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, bacterial lipoproteins and their derivatives that bind to TLR1.

Examples of TLR1 agonists useful in the treatments, medicines, and uses of the present invention include, for example, bacterial lipoproteins and their derivatives, such as SPM-105 (derived from autoclaved mycobacteria), OM-174 (lipid A derivative), OmpS1 (porin from Salmonella typhi), OspA (from Borrelia burgdorferi), MALP-2 (mycoplasma macrophage-activated lipopeptide-2kD), STF (soluble tuberculosis factor), CU-T12-9, Diprovocim, and lipopeptides derived from cell wall components, such as PAM2CSK4, PAM3CSK4, and PAM3Cys.

TLR1 can form a heterodimer with TLR2, and therefore, many TLR1 agonists are also TLR2 agonists.

The terms “TLR2” and “toll-like receptor 2” refer to any form of TLR2 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR2 activity. Unless otherwise specified, such as by way of a specific reference to human TLR2, TLR2 includes naturally occurring sequence TLR2 in all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR2 is provided under UniProt entry number O60603.

As used herein, “TLR2 agonist” means any molecule that, upon binding to TLR2, (1) stimulates or activates TLR2, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR2, or (3) enhances, increases, promotes, or induces the expression of TLR2. TLR2 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, bacterial lipoproteins and their derivatives that bind to TLR2.

Examples of TLR2 agonists useful in the treatments, medicines, and uses of the present invention include, for example, bacterial lipoproteins (e.g., diacylated lipoproteins) and their derivatives, such as SPM-105 (derived from autoclaved mycobacteria), OM-174 (lipid A derivative), OmpS1 (porin from Salmonella typhi), OspA (from Borrelia brevicornuate), MALP-2 (mycoplasma macrophage-activated lipopeptide-2kD), STF (soluble tuberculosis factor), CU-T12-9, Diprovocim, Amplivant, and lipopeptides derived from cell wall components, such as PAM2CSK4, PAM3CSK4, and PAM3Cys.

TLR2 can form heterodimers with TLR1 or TLR6, and therefore, many TLR2 agonists are also TLR1 or TLR6 agonists.

The terms “TLR3” and “toll-like receptor 3” refer to any form of TLR3 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR3 activity. Unless otherwise specified, such as by way of a specific reference to human TLR3, TLR3 includes naturally occurring sequence TLR3 in all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR3 is provided under UniProt entry number O15455.

As used herein, “TLR3 agonist” means any molecule that, upon binding to TLR3, (1) stimulates or activates TLR3, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR3, or (3) enhances, increases, promotes, or induces the expression of TLR3. TLR3 agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, nucleic acid ligands that bind to TLR3.

Examples of TLR3 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include TLR3 ligands such as synthetic dsRNA, polyinosinic acid [“poly(I:C)”] (available from, for example, InvivoGen in high molecular weight (HMW) and low molecular weight (LMW) formulations), polyadenylic-polyuridylic acid [“poly(A:U)”] (available from, for example, InvivoGen), and polyICLC (see Levy et al., Journal of Infectious Disorders). (See Jasani et al., Vaccine, vol. 132, no. 4, pp. 434-439, 1975), Ampligen (see Jasani et al., Vaccine, vol. 27, no. 25-26, pp. 3401-3404, 2009), Hiltonol, Rintatolimod, and RGC100 (see Naumann et al., Clinical and Developmental Immunology, vol. 2013, article ID 283649).

The terms “TLR4” and “toll-like receptor 4” refer to any form of TLR4 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR4 activity. Unless otherwise specified, such as by way of a specific reference to human TLR4, TLR4 includes naturally occurring sequence TLR4 in all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR4 is provided under UniProt entry number O00206.

As used herein, “TLR4 agonist” means any molecule that, upon binding to TLR4, (1) stimulates or activates TLR4, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR4, or (3) enhances, increases, promotes, or induces the expression of TLR4. TLR4 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, bacterial lipopolysaccharide (LPS) and its derivatives that bind to TLR4.

Examples of TLR4 agonists useful in the treatments, medicines and uses of the present invention include, for example, bacterial lipopolysaccharide (LPS) and its derivatives, such as B:0111 (Sigma), monophospholipid A (MPLA), 3DMPL (3-O-deacylated MPL), GLA-AQ, G100, AS15, ASO2, GSK1572932A (GlaxoSmithKline, UK).

The terms “TLR5” and “toll-like receptor 5” refer to any form of TLR5 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR5 activity. Unless otherwise specified, such as by way of a specific reference to human TLR5, TLR5 includes naturally occurring sequence TLR5s from all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR5 is provided under UniProt entry number O60602.

As used herein, “TLR5 agonist” means any molecule that, upon binding to TLR5, (1) stimulates or activates TLR5, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR5, or (3) enhances, increases, promotes, or induces the expression of TLR5. TLR5 agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, bacterial flagellin and derivatives thereof that bind to TLR5.

Examples of TLR5 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include, for example, bacterial flagellin purified from Bacillus subtilis, flagellin purified from Pseudomonas aeruginosa, flagellin purified from Salmonella typhimurium, and recombinant flagellin (all available from InvivoGen), and entolimod (CBLB502; a pharmacologically optimized flagellin derivative).

The terms “TLR6” and “toll-like receptor 6” refer to any form of TLR6 receptor, as well as variants, isotypes, and species homologs that retain at least a portion of TLR6 activity. Unless otherwise specified, such as by way of a specific reference to human TLR6, TLR6 includes the naturally occurring sequence TLR6 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR6 is provided under UniProt entry number Q9Y2C9.

As used herein, “TLR6 agonist” means any molecule that, upon binding to TLR6, (1) stimulates or activates TLR6, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR6, or (3) enhances, increases, promotes, or induces the expression of TLR6. TLR6 agonists useful in any of the treatments, medicaments, and uses of this invention include, for example, bacterial lipopeptides and derivatives thereof that bind to TLR6.

Examples of TLR6 agonists useful in the treatments, medicines, and uses of the present invention include, for example, many of the TLR2 agonists provided above, because TLR2 and TLR6 can form heterodimers. TLR6 can also form heterodimers with TLR4, and TLR6 agonists can include the various TLR4 agonists provided above.

The terms “TLR7” and “toll-like receptor 7” refer to any form of TLR7 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR7 activity. Unless otherwise specified, such as by way of a specific reference to human TLR7, TLR7 includes the naturally occurring sequence TLR7 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR7 is provided under UniProt entry number Q9NYK1.

As used herein, “TLR7 agonist” means any molecule that, upon binding to TLR7, (1) stimulates or activates TLR7, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR7, or (3) enhances, increases, promotes, or induces the expression of TLR7. TLR7 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acid ligands that bind to TLR7.

Examples of TLR7 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include recombinant single-stranded (“ss”) RNA, imidazoquinoline compounds such as imiquimod (R837), gardiquimod, and resiquimod (R848); rozolidin (7-allyl-7,8-dihydro-8-oxoguanosine) and related compounds; 7-thia-8-oxoguanosine, 7-denitroguanosine, and related guanosine analogs; ANA975 (Anadys Pharmaceuticals) and related compounds; SM-360320 (Sumimoto); 3M-01, 3M-03, 3M-852, and 3M-S-34240 (3M Pharmaceuticals); GSK2245035 (GlaxoSmithKline; 8-oxoadenine molecule), AZD8848 (AstraZeneca; 8-oxoadenine molecule), MEDI9197 (Medimmune; formerly 3M-052), ssRNA40, and adenosine analogs, such as UC-1V150 (Jin et al., Bioorganic Medicinal Chem Lett (2006) 16:4559-4563, compound 4). Many TLR7 agonists are also TLR8 agonists.

The terms “TLR8” and “toll-like receptor 8” refer to any form of TLR8 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR8 activity. Unless otherwise specified, such as by way of a specific reference to human TLR8, TLR8 includes the naturally occurring sequence TLR8 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR8 is provided under UniProt entry number Q9NR97.

As used herein, “TLR8 agonist” means any molecule that, upon binding to TLR8, (1) stimulates or activates TLR8, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR8, or (3) enhances, increases, promotes, or induces the expression of TLR8. TLR8 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acid ligands that bind to TLR8.

Examples of TLR8 agonists useful in the treatments, medicines, and uses of the present invention include recombinant single-stranded ssRNA, imiquimod (R837), gademod, requimod (R848), 3M-01, 3M-03, 3M-852, and 3M-S-34240 (3M Pharmaceuticals); GSK2245035 (GlaxoSmithKline; 8-oxoadenine molecule), AZD8848 (AstraZeneca; 8-oxoadenine molecule), MEDI9197 (Medimmune; formerly 3M-052), Poly-G10, Motolimod, and various TLR7 agonists provided above (as previously noted, many TLR7 agonists are also TLR8 agonists).

The terms “TLR9” and “toll-like receptor 9” refer to any form of TLR9 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR9 activity. Unless otherwise specified, such as by way of a specific reference to human TLR9, TLR9 includes the naturally occurring sequence TLR9 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR9 is provided under UniProt entry number Q9NR96.

As used herein, “TLR9 agonist” means any molecule that, upon binding to TLR9, (1) stimulates or activates TLR9, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR9, or (3) enhances, increases, promotes, or induces the expression of TLR9. TLR9 agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, nucleic acid ligands that bind to TLR9.

Examples of TLR9 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include unmethylated CpG-containing DNA, immunostimulatory oligodeoxynucleotides (ODNs), such as CpG-containing ODNs, such as CpG24555, CpG10103, CpG7909 (PF-3512676/agatolimod), CpG1018, AZD1419, ODN2216, MGN1703, SD-101, 1018ISS, and CMP-001. TLR9 agonists also include a nucleotide sequence containing a synthetic cytosine-phosphate-2'-deoxy-7-dezoguanosine dinucleotide (CpR) (Hybridon, Inc.), dSLIM-30L1, and an immunoglobulin-DNA complex. Exemplary TLR9 agonists are disclosed in WO2003/015711, WO2004/016805, WO2009/022215, PCT/US95/01570, PCT/US97/19791 and U.S. Patent Nos. 8552165, 6194388 and 6239116, which are incorporated herein by reference for all purposes.

RLRs include various cytoplasmic PRRs that detect, for example, dsRNAs. Examples of RLRs include, for example, retinoic acid-inducible gene I (RIG-I), melanoma differentiation-associated gene 5 (MDA-5), and Laboratory of Genetics and Physiology 2 (LGP2).

As used herein, “RLR agonist” means any molecule that, upon binding to an RLR, (1) stimulates or activates the RLR, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of the RLR, or (3) enhances, increases, promotes, or induces the expression of the RLR. RLR agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acids that bind to RLRs and their derivatives, and agonistic monoclonal antibodies (mAbs) that specifically bind to RLRs.

Examples of RLRs agonists useful in the treatments, medicines, and uses of the present invention include, for example, short double-stranded RNAs having an uncapped 5' triphosphate (RIG-I agonists); poly I:C (MDA-5 agonists); and BO-112 (MDA-A agonists).

NLRs include various PRRs that detect molecules such as damage-related molecular patterns (DAMPs). NLRs include subfamilies NLRA-A, NLRB-B, NLRC-C, and NLRP-P. Examples of NLRs include, for example, NOD1, NOD2, NAIP, NLRC4, and NLRP3.

As used herein, “NLR agonist” means any molecule that, upon binding to an NLR, (1) stimulates or activates the NLR, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of the NLR, or (3) enhances, increases, promotes, or induces the expression of the NLR. NLR agonists useful in any of the treatments, medicines, and uses of this invention include, for example, DAMPs and their derivatives that bind to NLRs, and agonistic monoclonal antibodies (mAbs) that specifically bind to NLRs.

Examples of NLR agonists useful in the treatments, medicines, and uses of the present invention include, for example, liposomal muramyl tripeptide/mifamurtide (NOD2 agonist).

CLRs include various PRRs that detect, for example, carbohydrates and glycoproteins. CLRs include both transmembrane CLRs and secretory CLRs. Examples of CLRs include, for example, DEC-205/CD205, macrophage mannose receptor (MMR), Dectin-1, Dectin-2, mincle, DC-SIGN, DNGR-1, and mannose-binding lectin (MBL).

As used herein, “CLR agonist” means any molecule that, upon binding to a CLR, (1) stimulates or activates the CLR, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of the CLR, or (3) enhances, increases, promotes, or induces the expression of the CLR. CLR agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, carbohydrates and their derivatives that bind to CLRs, and agonistic monoclonal antibodies (mAbs) that specifically bind to CLRs.

Examples of CLR agonists useful in the treatments, pharmaceuticals, and uses of the present invention include, for example, MD fraction (a purified soluble β-glucan extract from Grifola frondosa) and imprime PGG (a yeast-derived β1,3/1,6-glucan PAMP).

STING proteins function as both cytoplasmic DNA sensors and adaptors in type 1 interferon signaling. The terms "STING" and "stimulator of interferon genes" refer to any form of STING protein, as well as variants, allotypes, and species homologs that retain at least a portion of STING activity. Unless otherwise specified, such as by way of a specific reference to human STING, STING includes the naturally occurring sequence of STING in all mammalian species, e.g., humans, monkeys, and mice. An exemplary human TLR9 is provided under UniProt entry number Q86WV6. STING is also known as TMEM173.

As used herein, “STING agonist” means any molecule that, upon binding to TLR9, (1) stimulates or activates STING, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of STING, or (3) enhances, increases, promotes, or induces the expression of STING. STING agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acid ligands that bind to STING.

Examples of STING agonists useful in the treatments, medicines, and uses of the present invention include various immunostimulatory nucleic acids, such as synthetic double-stranded DNA, cyclic di-GMP, cyclic GMP-AMP (cGAMP), synthetic cyclic dinucleotides (CDNs), such as MK-1454 and ADU-S100 (MIW815), and small molecules, such as PO-424.

Other PRRs include, for example, DNA-dependent Activator of IFN-regulatory Factors (DAI) and Absentin Melanoma 2 (AIM2).

Immunostimulatory cytokines include various signaling proteins that stimulate immune responses, such as interferons, interleukins, and hematopoietic growth factors.

Exemplary immunostimulatory cytokines include GM-CSF, G-CSF, IFN-α, IFN-γ; IL-2 (e.g., denileukin difitox), IL-6, IL-7, IL-11, IL-12, IL-15, IL-18, IL-21, and TNF-α.

Immunostimulatory cytokines can be in any suitable form. In some embodiments, the immunostimulatory cytokines can be recombinant forms of wild-type cytokines. In some embodiments, the immunostimulatory cytokines can be mutant proteins with one or more amino acid alterations compared to the corresponding wild-type cytokines. In some embodiments, the immunostimulatory cytokines can be incorporated into chimeric proteins containing cytokines and at least one other functional protein (e.g., antibodies). In some embodiments, the immunostimulatory cytokines can be covalently linked to a drug/pharmaceutical (e.g., any drug/pharmaceutical as described elsewhere herein as a possible ADC component).

Cancer vaccines include various compositions containing tumor-associated antigens (or that can be used to generate tumor-associated antigens in subjects) and thus can be used to induce an immune response in subjects against tumor cells containing tumor-associated antigens.

Examples of materials that may be included in a cancer vaccine include attenuated cancerous cells, tumor antigens, and antigen-presenting cells such as dendritic cells bombarded with nucleic acids encoding tumor-derived antigens or tumor-associated antigens. In some embodiments, the cancer vaccine may be prepared using the patient's own cancer cells. In some embodiments, the cancer vaccine may be prepared using biomaterials not derived from the patient's own cancer cells.

Cancer vaccines include, for example, sipuleucel-T and talimogene laherparepvec (T-VEC).

Immunotherapy involves treating patients with immune cells that can target cancer cells. Examples of immunotherapy include, for instance, tumor-infiltrating lymphocytes (TILs) and chimeric antigen receptor T cells (CAR-T cells).

Examples of chemotherapeutic agents include alkylating agents such as thiotepa and cyclophosphamide; alkyl sulfonates such as busulfan, dipropylamine sulfonate, and piperosulfan; aziridines such as benzodopa, carboquinone, meturedopa, and uredopa; ethyleneimines and methylamelamines, including hexamethylmelamine, tratamide, trietylenephosphoramide, ethyl thiophosphoric acid, and trimethylolomelamine; polyacetogenins (especially bullatacin and bullatacinone); camptothecin (including synthetic varieties). Analogs of topotecan; bryozoins; callystatin; CC-1065 (including its synthetic analogs of adolexin, carzelesin, and bizelesin); cryptophycins (especially cryptophycin 1 and cryptophycin 8); dolastatin; bacitracin (including synthetic analogs, KW-2189 and CBI-TMI); eleutherobin; pancratistatin; sarcodictyin; spongistatin; nitrogen mustard, such as chlorambucil, naphthalene, and chlorpyrifos. Cholophosphamide, estradiol, ifosfamide, nitrogen mustard, mechlorethamine, mechlorethamine oxide hydrochloride, melphalan, neonitrogen mustard, benzyl mustard cholesterol, pine oxychloride, tristyramine cyclophosphamide, uracil nitrogen mustard; nitrosoureas, such as carmustine, chloramphenicol, formazan, lomomia, nimustine, ranolamia; antibiotics, such as enediyne antibiotics (e.g., galicarmycin, especially galicarmycin γ1I and galicarmycin phiI1, see, e.g., Agnew, Chem. Intl. Ed. Engl., 33:183-186(19) 94); dynemicin, including dynemicin A; bisphosphonates, such as osteophosphine; esporamin; and new carcinogenic chromophores and related chromomophores, aclarubicin, actinomycin, autramycin, azoserine, bleomycin, cactinomycin C, carabicin, caminomycin, carcinomycin, chromomycin, daunorubicin, doxorubicin, 6-diazo-5-oxo-L-leucine, doxorubicin (including morpholino-doxorubicin, cyanomorpholino-doxorubicin, 2-pyrrolino-doxorubicin) (2-pyrrolino-doxorubicin and deoxydoxomyl), PEGylated liposomal doxorubicin, epirubicin, doxorubicin, idarubicin, mesorcinol, mitomycin such as mitomycin C, mycophenolic acid, nogaramycin, oligomycin, pemycin, potfiromycin, puromycin, triamcinolone acetonide, rodorubicin, streptomycin, streptozotocin, tuberculin, hydroxyphenylbutyryl leucine, neomycin, zorubicin; antimetabolites, such as methotrexate and 5-fluorouracil (5-FU); folic acid analogs, such as folate, methotrexate, pteroyltriglutamate, trimesat; purine analogs, such as fludarabine, 6-mercaptopurine, thioguanine, thioguanine; pyrimidine analogs, such as cyclocytidine, azathioprine, etc. Cytidine, 6-azouridine, carmoflurane, cytarabine, dideoxyuridine, docefluuridine, enoxabin, fluorouridine; androgens, such as carotestosterone, hydroxymethylandrostenone propionate, cyclothiosterol, methylcyclothiosterane, testrolide; anti-adrenergic drugs, such as aminoglutethimide, mitotane, and trilostertan; folic acid supplements, such as frolinic acid; aceglucuronolactone; aldophosphamide glycoside; aminolevulinic acid; eniluracil; acridine; bestrabucil; bethanycin; edatraxate; defofamine; dimethicone; diazinon; elfo-ornithine. rmithine); hydroxypicarazole acetate; epothilone; glycidyl ether; gallium nitrate; hydroxyurea; lentinan; chlordamine; maytansine alkaloids, such as maytansine and metansine; propiconazole; mitoxantrone; monpilamol; diammonium nitrate; pentostatin; benzylmethionine; pirarubicin; losoxantrone; podophyllin; 2-ethylhydrazide; methylbenzylhydrazine; propylimine; rhizomycin; sisofibril; spirogermanium; Alternaria spp. ketoacid; triazolam; 2,2',2''-trichlorotriethylamine; trichosporon toxins (especially T-2 toxin, verracurin A, baculosporin A, and anguidine); ursodeoxycholic acid; vinblastine; dacarbazin; mannitol; dibromomannitol; dibromodehydroxycervastatin Piperbromopropionate; gacytosine; cytarabine (“Ara-C”); cyclophosphamide; thiotepa; taxoids, such as paclitaxel and docetaxel; chlorambucil; gemcitabine; 6-thioguanine; mercaptopurine; methotrexate; platinum analogs, such as cisplatin and carboplatin; vincristine; platinum; etoposide (VP-16); ifosfamide; mitoxantrone; vincristine; vinblastine; nifedipine; teniposide; edaraxacum; daunorubicin; aminopterin; capecitabine; ibenphosphonate sodium; CPT-11; topoisomerase inhibitor RFS 2000; difluoromethylornithine (DMFO); retinoids, such as retinoic acid; capecitabine; and pharmaceutically acceptable salts, acids, or derivatives of any of the above. It also includes anti-hormonal agents used to regulate or inhibit the hormonal effects on tumors, such as anti-estrogens and selective estrogen receptor modulators (SERMs), including, for example, tamoxifen, raloxifene, droloxifene, 4-hydroxytamoxifen, treoxifene, naloxifene, LY117018, onasone, and toremifene (Fareston); aromatase inhibitors that inhibit aromatase, which regulate estrogen production in the adrenal glands, such as 4(5)-imidazole, aminoglutethimide, megestrol acetate, etc. Exemestane, formetastatin, fazodazole, vorazole, letrozole, and anastrozole; and anti-androgen substances such as flutamide, nilumid, bicalutamide, lipoprotein, and serotonin; KRAS inhibitors; MCT4 inhibitors; MAT2a inhibitors; tyrosine kinase inhibitors, such as sunitinib and axitinib; alk/c-Met/ROS inhibitors, such as crizotinib and lorlatinib. Intra-inib; mTOR inhibitors, such as temsirolimus and gedatolisib; src/abl inhibitors, such as bosutinib; cyclin-dependent kinase (CDK) inhibitors, such as palbociclib and PF-06873600; erb inhibitors, such as dacomitinib; PARP inhibitors, such as talazoparib; SMO inhibitors, such as glasdegib and PF-5274857; EGFR T790M inhibitors, such as PF-06747775; EZH2 inhibitors, such as PF-06821497; PRMT5 inhibitors, such as PF-06939999; TGFRβr1 inhibitors, such as PF-06952229; and pharmaceutically acceptable salts, acids, or derivatives of any of the above. In specific implementation schemes, this additional treatment agent is bevacizumab, cetuximab, sirolimus, panitumumab, 5-fluorouracil (5-FU), capecitabine, tevozanib, irinotecan, oxaliplatin, cisplatin, trifluorouridine, tilpyrimidine, leucovorin, gemcitabine, regorafinib, or erlotinib hydrochloride.

In some embodiments, the antibody is used in conjunction with one or more other therapeutic agents that target immune restriction point modulators or co-stimulators, such as, but not limited to, those targeting CTLA-4, LAG-3, B7-H3, B7-H4, B7-DC (PD-L2), B7-H5, B7-H6, B7-H8, B7-H2, B7-1, B7-2, ICOS, ICOS-L, TIGIT, CD2, CD47, CD80, CD86, CD48, C D58, CD226, CD155, CD112, LAIR1, 2B4, BTLA, CD160, TIM1, TIM-3, TIM4, VISTA(PD-H1), OX40, OX40L, GITR, GITRL, CD70, CD27, 4-1BB, 4-BBL, DR3, TL1A, CD40, CD40L, CD30, CD30L, LIGHT, HVEM, SLAM (SLAMF1, CD150 ), SLAMF2 (CD48), SLAMF3 (CD229), SLAMF4 (2B4, CD244), SLAMF5 (CD84), SLAMF6 (NTB-A), SLAMCF7 (CS 1), SLAMF8 (BLAME), SLAMF9 (CD2F), CD28, CEACAM1 (CD66a), CEACAM3, CEACAM4, CEACAM5, CEACAM6, CEAC AM7, CEACAM8, CEACAM1-3AS, CEACAM3C2, CEACAM1-15, PSG1-11, CEACAM1-4C1, CEACAM1-4S, CEACAM1-4L, IDO, TDO, CCR2, CD39–CD73–adenosine pathway (A2AR), BTKs, TIKs, CXCR2, CXCR4, CCR4, CCR8, CCR5, CSF-1, or agents that modulate the innate immune response.

In some implementations, the antibody is used in conjunction with, for example, the following: anti-CTLA-4 antagonist antibodies, such as ipilimumab; anti-LAG-3 antagonist antibodies, such as BMS-986016 and IMP701; anti-TIM-3 antagonist antibodies; anti-B7-H3 antagonist antibodies, such as MGA271; anti-VISTA antagonist antibodies; anti-TIGIT antagonist antibodies; antibodies; anti-CD80 antibodies; anti-CD86 antibodies; anti-B7-H4 antagonist antibodies; anti-ICOS agonist antibodies; anti-CD28 agonist antibodies; modulators of innate immune responses (e.g., TLRs, KIR, NKG2A); and IDO inhibitors.

In some embodiments, the antibody is used with an OX40 agonist, such as, for example, an anti-OX-40 agonist antibody. In some embodiments, the antibody is used with a GITR agonist, such as, for example, an anti-GITR agonist antibody, such as, but not limited to, TRX518. In some embodiments, the antibody is used with an IDO inhibitor. In some embodiments, the antibody is used with cytokine therapy (such as, for example, but not limited to, IL-15, CSF-1, MCSF-1, etc.).

In some implementations, the antibody is used in conjunction with one or more other therapeutic antibodies, such as, but not limited to, antibodies targeting CD19, CD22, CD40, CD52, or CCR4.

In some embodiments, the antibody composition comprises at least one additional pharmaceutical agent, such as bevacizumab, cetuximab, sirolimus, panitumumab, 5-fluorouracil (5-FU), capecitabine, tevozanib, irinotecan, oxaliplatin, cisplatin, trifluorouridine, tilpyridamole, leucovorin, gemcitabine, and erlotinib hydrochloride.

In some embodiments, the antibody therapy may be administered co-administered with other reagents or sequentially before or after other reagents at intervals ranging from minutes to weeks. In embodiments where other reagents and/or proteins or polynucleotides are administered alone, it will generally be ensured that a considerably long period of time does not expire between each delivery, such that the agent and the composition of the invention will still be able to produce a beneficial combined effect on the subject. In this case, it is anticipated that the two dosing regimens may be administered within approximately 12-24 hours of each other, and more preferably within approximately 6-12 hours of each other. However, in some cases, it may be desirable to significantly extend the time period for administration, in this regard, from several days (2, 3, 4, 5, 6, or 7) to several weeks (1, 2, 3, 4, 5, 6, 7, or 8) between each administration.

In some embodiments, the antibodies of the present invention are combined with treatment regimens that further include conventional therapies selected from: surgery, radiotherapy, chemotherapy, targeted therapy, immunotherapy, hormone therapy, angiogenesis inhibition, and palliative care.

The compositions used in this invention may further comprise pharmaceutically acceptable carriers, excipients, or stabilizers in the form of lyophilized formulations or aqueous solutions (Remington: The Science and Practice of Pharmacy, 21st edition, 2005, Lippincott Williams and Wilkins, Ed. K. E. Hoover). Acceptable carriers, excipients, or stabilizers are non-toxic to the recipient at the stated doses and concentrations and may include buffers such as phosphates, citrates, and other organic acids; antioxidants, including ascorbic acid and methionine; preservatives (e.g., octadecyl dimethyl benzyl ammonium chloride; hexane diammonium chloride; benzalkonium chloride, benzyl chloride; phenols, butanol, or benzyl alcohol; alkyl esters of p-hydroxybenzoate, such as methylparaben or propylparaben; catechol; resorcinol; cyclohexanol; 3-pentanol and m-cresol); low molecular weight (less than about 10 residues) polypeptides; proteins, such as blood... Albumin, gelatin, or immunoglobulins; hydrophilic polymers, such as polyvinylpyrrolidone; amino acids, such as glycine, glutamine, asparagine, histidine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates, including glucose, mannose, or dextran; chelating agents, such as EDTA; sugars, such as sucrose, mannitol, trehalose, or sorbitol; salt-forming ions, such as sodium; metal complexes (e.g., Zn-protein complexes); and/or nonionic surfactants, such as TWEEN™, PLURONICS™, or polyethylene glycol (PEG). Pharmaceutically acceptable excipients are further described herein.

In one aspect, the present invention provides a method for generating antibodies as disclosed herein, comprising: culturing host cells under conditions that result in the generation of antibodies as disclosed herein, and purifying the antibody or the bispecific antibody from the culture.

In another aspect, the present invention provides the use of antibodies, pharmaceutical compositions, polynucleotides, vectors or host cells as disclosed herein in the preparation of medicaments for the detection, diagnosis and/or treatment of tumor antigen-related conditions.

Diagnostic uses

In one aspect, for example for diagnostic purposes, the CD3 antibody of the present invention can be used to detect and/or measure CD3 or CD3-expressing cells in a sample. For example, the CD3 antibody can be used to diagnose conditions or diseases characterized by abnormal expression of CD3 (e.g., overexpression, underexpression, lack of expression, etc.). Exemplary diagnostic assays for CD3 may include, for example, contacting a sample obtained from a subject with the CD3 antibody of the present invention, wherein the CD3 antibody is labeled with a detectable marker or reporter molecule. Alternatively, an unlabeled CD3 antibody may be used in combination with a self-detectably labeled secondary antibody for diagnostic applications. The detectable marker or reporter molecule may be a radioisotope, such as ^14C , ^3H , ^125I , ^32P , or ^35S ; a fluorescent or chemiluminescent moiety, such as fluorescein isothiocyanate or rhodamine; or an enzyme, such as alkaline phosphatase, β-galactosidase, horseradish peroxidase, or luciferase. Specific exemplary assays that can be used to detect or measure CD3 in a sample include, but are not limited to, enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), fluorescence activated cell sorting (FACS), etc. Samples that can be used in the CD3 diagnostic assay according to the present invention include any tissue or body fluid sample obtainable from a patient, containing a detectable amount of CD3 protein or fragments thereof under normal or pathological conditions. Typically, the level of CD3 is measured in a specific sample obtained from a healthy patient (e.g., a patient without a disease or condition associated with abnormal CD3 levels or activity) to first establish a baseline or standard for CD3 levels. This baseline CD3 level can then be compared with the CD3 level measured in a sample obtained from an individual suspected of having a CD3-related disease or condition.

In another aspect, methods are provided for detecting, diagnosing, and/or monitoring conditions associated with tumor antigen expression. For example, multispecific antibodies as described herein can be labeled with detectable moieties (e.g., imaging agents and enzyme-substrate markers). Antibodies as described herein can also be used for in vivo diagnostic assays (e.g., in vivo imaging (e.g., PET or SPECT)) or staining reagents.

Reagent test kit

Another aspect of the invention is a kit comprising an antibody as disclosed above and instructions for use according to any method of the invention described herein. Typically, these instructions include a description of the administration of the antibody for the therapeutic treatment described above. The kit includes any pharmaceutical compositions disclosed herein. Pharmaceutical compositions and other reagents may be present in the kit in any convenient form, such as, for example, in solution or powder form.

In another aspect, the kit further includes one or more other preventative or therapeutic agents that can be used to treat cancer, contained in one or more containers. In one embodiment, said other preventative or therapeutic agent is a chemotherapeutic agent. In other aspects, said preventative or therapeutic agent is a biological or hormonal therapeutic agent.

Several aspects of the pharmaceutical compositions, preventive agents, or therapeutic agents of the present invention are preferably tested for desired therapeutic activity in vitro, in cell culture systems, and in animal model organisms (e.g., rodent model systems) before human use.

The toxicity and efficacy of the preventive and/or therapeutic regimens of the present invention can be determined by standard pharmaceutical procedures in cell cultures or laboratory animals, for example, for determining the _LD50 (the dose that is lethal to 50% of the population) and _ED50 (the dose that is therapeutically effective in 50% of the population). The dose ratio between toxicity and therapeutic effect is the therapeutic index, and it can be expressed as the ratio _LD50 / _ED50 . Prophylactic and/or therapeutic agents exhibiting a large therapeutic index are preferred.

Furthermore, any assays known to those skilled in the art can be used to evaluate the preventive and/or therapeutic efficacy of the therapies or combinations of therapies disclosed herein for the treatment or prevention of cancer.

Instructions for use with bispecific antibodies as described herein typically include information about the dosage, dosing regimen, and route of administration for the intended treatment. Containers may be unit doses, bulk packaging (e.g., multi-dose packs), or subunit doses. Instructions supplied with kits of the present invention are typically written instructions on a label or packaging insert (e.g., paper included in the kit), but machine-readable instructions (e.g., instructions carried on a disk or optical storage disc) are also acceptable.

The kit of the present invention is in suitable packaging. Suitable packaging includes, but is not limited to, vials, ampoules, tubes, bottles, wide-mouth bottles, flexible packaging (e.g., sealed polyester film or plastic bags), etc., for each pharmaceutical composition and other included reagents (e.g., buffers, balanced saline solutions, etc., for administering the pharmaceutical composition to a subject). Packaging for use in combination with specific devices such as inhalers, nasal delivery devices (e.g., nebulizers), or infusion devices such as micropumps is also contemplated. The kit may have a sterile inlet (e.g., the container may be an intravenous solution bag or vial with a stopper that can be punctured by a hypodermic needle). The container may also have a sterile inlet (e.g., the container may be an intravenous solution bag or vial with a stopper that can be punctured by a hypodermic needle). At least one active agent in the composition is an antibody. The container may further contain a second pharmaceutical active agent.

Typically, a kit includes a container and a label or packaging insert on or associated with the container.

Biological Preservation

Representative material of the present invention was deposited on February 13, 2018 at the American Center for Type Culture Collection (ATCC), 10801 University Boulevard, Manassas, Va. 20110-2209, USA. The vector GUCY2C-1608 chain A (VH-Synthesis SEQ ID NO: 94) with ATCC accession number PTA-124943 contains a DNA insert of the VH-Synthesis chain encoding the bispecific antibody GUCY2C-1608, which contains a polynucleotide (SEQ ID NO: 70) encoding the CD3 antibody light chain variable region (SEQ ID NO: 3); and the GUCY2C-1608 chain B (VL-Synthesis SEQ ID NO: 93) with ATCC accession number PTA-124944 contains a DNA insert of the VL-Synthesis chain encoding the bispecific antibody GUCY2C-1608, which contains a polynucleotide (SEQ ID NO: 71) encoding the CD3 antibody heavy chain variable region (SEQ ID NO: 4).

The deposit was made in accordance with the provisions of the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the Purpose of Patent Procedure and Regulations thereunder. This ensures that the live culture of the deposit remains intact for 30 years from the date of deposit. The deposit is available through the ATCC under the terms of the Budapest Treaty and complies with an agreement between Pfizer, Inc. and the ATCC that ensures the perpetual and unrestricted availability of the descendants of the deposited culture to the public at the time of publication of the relevant U.S. Patent or at the time of public disclosure of any U.S. or foreign patent application (whichever comes first), and ensures the availability of the descendants to persons authorized by the Director of the U.S. Patent and Trademark Office in accordance with 35 U.S.C. Part 122 and the Director's Rules established thereunder (including 37 C.F.R. Part 1.14, which specifically refers to 886 OG 638).

The assignee of this application has agreed that, should the culture of the deposited material die, be lost, or be destroyed under suitable conditions, the material will be promptly replaced with another identical copy upon notification. The availability of the deposited material shall not be construed as a violation of any government authority's right to practice the invention under its patent law.

The following embodiments for implementing specific aspects of the invention are provided for illustrative purposes only and are not intended to limit the scope of the invention in any way.

Example 1

Formation of soluble CD3ed

A protein called CD3ed, representing the extracellular domains of CD3 ε and δ subunits tandemly fused via a short linker (SEQ ID NO: 39), was generated using standard molecular biology, protein expression, and purification techniques. Using PEI as the transfection agent, plasmids carrying the gene encoding the amino acids shown in Table 9 (SEQ ID NO: 40) were transiently transfected into 20 μL of HEK293 cells. After incubation at 37°C for 7 days, 20 μL of the harvest was fractionally bound to 50 mL of His60 resin (equilibrated with PBS) overnight. The bound resin was then washed with PBS, followed by washing with PBS buffer (10⁻¹⁵ CV) containing 10 mM imidazole. Following washing, the protein was eluted with phosphate-buffered saline containing 200 mM imidazole. The eluent fractions were then analyzed by SDS gel electrophoresis. Fractions containing purified protein were pooled and concentrated using a 3 kDa Amicon ultra centricon. During concentration, 5% glycerol was maintained in the sample to stabilize the protein. The His60 pool was further loaded onto a size-restricted Superdex 200 column. Fractions containing CD3 ε/δ heterodimers were analyzed by SDS-PAGE, and the relevant fractions were pooled and filtered at 0.22 µm. Table 9 shows the sequences of the CD3 ε/δ subunits tandemly fused and used as antigens in all optimization efforts.

Table 9

Example 2

Optimization of the structure of CD3 antibody H2B4

The lead CD3e (CD3 ε) antibody H2B4 exhibited low thermostability when used in bispecific biantibody-Fc form. To improve stability, the light chain architecture was changed from VK4 to VK1 (DPK9), while the heavy chain architecture was retained as VH3 (DP54). VLCDR1 (SEQ ID NO: 26), VL CDR2 (SEQ ID NO: 27), and VL CDR3 (SEQ ID NO: 28) were transplanted into the DPK9 architecture and expressed in combination with heavy chains carrying CDRs (i.e., VH CDR1 (SEQ ID NO: 13), VHCDR2 (SEQ ID NO: 16), and VH CDR3 (SEQ ID NO: 18)). In addition to transplanting CDRs into the architecture, certain reversion mutations were tested. The reversion mutations introduced in both the heavy and light chain structures are shown in Table 10, where the residue number indicating the reversion mutation is indicated.

A cDNA containing the human receptor framework (VH3 for the heavy chain and VK1 for the light chain) and the associated CDR donor sequence was synthesized. The synthesized cDNA product was subcloned and fused to the human IgG1 constant region of the heavy chain and the human κ-frame of the light chain in a mammalian expression vector. To distinguish them from the H2B4 antibody, these variants will be referred to as 2B5 below, and all optimized molecules will be referred to as 2B5 and its variants. The competitive binding of all 2B5 mutants to Jurkat cells with endogenous CD3e expression was analyzed.

Competitive FACS was performed to examine whether humanized 2B5 maintained its binding epitope on the cell surface. Prior to the competitive FACS assay, chimeric cH2B4 was biotinylated. The EC50 of the binding of biotinylated cH2B4 to Jurkat cells was determined by direct binding FACS and was measured to be 0.8 nM. For competitive FACS, a 3-fold serially diluted antibody variant was mixed with a constant amount of biotin-cH2B4 representing the EC50 concentration and 1.0+E05 cells/well of Jurkat cells. The mixture was incubated on ice for 1 hour, followed by three washes with FACS buffer. A secondary antibody (i.e., PE (phycoerythrin)-conjugated streptavidin (PE-SA)) was added to the cells and incubated on ice for 30 minutes. Median fluorescence intensity (MFI) was then analyzed by FACS on FACS CANTO™ (BDBiosciences). Y-axis calculation: %Maximum binding = [MFI (serially diluted Ab wells) - MFI (secondary PE-SA)] / [MFI maximum (Jurkat cells with biotin-cH2B4 only) - MFI (secondary PE-SA)]. Table 10 shows the results of the _IC50 values for competitive binding when mutations were introduced into CD3 antibody H2B4 to optimize the scaffold. Based on these results, H2B4 1.0/1.0T showed similar binding characteristics to H2B4 and was designated as the CD3 antibody binding domain 2B5.

Table 10

Antibody	Heavy chain	Light chain	<![CDATA[Competitive combination of IC<sub>50</sub>]]>
				h2B4 1.0/1.0T (2B5)	CDR grafts	CDR graft, R24T	1.27
h2B4 1.0/1.1	vH1.0	Q13V, R24T, Y49S, L78V	1.68
				h2B4 1.0/1.2	vH1.0	Q13V, R24T, Y49S	Untested
h2B4 1.0/1.3	vH1.0	R24T, Y49S	1.309
				c2B4 (Chimeric 2B4)	mouse vH	mouse vK	0.752

Example 3

Bispecific biantibody-Fc molecules were generated and characterized using H2B4 and 2B5 sequences.

Bispecific biantibody-Fc molecules were generated using standard expression and purification techniques known in the art, using CD3 antibody 2B5 or H2B4 (Tables 2 and 3) and a GUCY2c antibody sequence in one of the configurations shown in Figure 1 (schematic diagram). The stability of the resulting bispecific antibodies (Table 7; SEQ.ID.NO.36-38) was tested using differential scanning calorimetry.

Example 4

The stability of bispecific antibodies was evaluated using differential scanning calorimetry.

The protein was diluted to 0.6 mg/mL in 400 µL of phosphate-buffered saline (PBS). PBS was used as a buffer blank in the reference chamber. PBS contained 137 mM NaCl, 2.7 mM KCl, 8.1 mM Na₂HPO₄, and 1.47 mM KH₂PO₄ at pH 7.2. Samples were dispensed into the sample trays of a MicroCal VP-Capillary DSC using an autosampler (Malvern Instruments Ltd, Malvern, UK). Samples were equilibrated at 10 °C for 5 min and then scanned at 100 °C per hour up to 110 °C. A 16-second filtration period was selected. The raw data were baseline corrected and the protein concentration was normalized. The data were fitted to the MN2-State model using Origin software 7.0 (OriginLab Corporation, Northampton, MA) with an appropriate number of transformations. As illustrated in Figure 2, the bispecific GUCY2c-2B5 exhibits an improved first melt transition at 5°C compared to the bispecific GUCY2c-H2B4. This indicates that bispecific antibodies containing 2B5 demonstrate better stability.

Using FACS, the binding of two bispecific antibodies to T cells was examined using standard procedures known in the art, and as shown in Figure 3, both bispecific antibodies showed similar binding to T cells, indicating that the binding was retained after H2B4 was converted to 2B5.

Both H2B4 and 2B5 bispecific antibodies were tested in a cytotoxicity assay to evaluate T cell retargeting activity. As shown in Figure 4, both bispecific antibodies showed similar cytotoxicity, indicating that 2B5 retained the same activity as H2B4.

Example 5

2B5-based structure-based rational mutation

The CD3 antibody H2B4 binding domain exhibits multireactivity (e.g., as evaluated by binding to DNA and insulin), shows a tendency for self-association (e.g., indicated by a high AC-SINS score), and exhibits a tendency for cleavage at the R53 position in CDR-H2 when expressed in CHO host cells. To reduce multireactivity, the tendency for self-association and cleavage in CDRH2, and mutations in the CDR of 2B5, were engineered to reduce these adverse factors while maintaining binding affinity and stability. For this effort, all predictions were made using the CD3 antibody H2B4 X-ray crystal structure, obtained in a complex with a peptide representing the binding epitope and derived from the N-terminus of the CD3 ε subunit. It is well known in the art that excess positive charge and/or hydrophobicity can play a major role in multireactivity. The electrostatic surface of the H2B4 Fv region from the crystal structure was determined using the DelPhi Poisson Boltzmann calculator in Discovery Studio 4.0 ^1. In Figure 5, an excess positive charge patch that is not disrupted by negative charge is observed. This indicates a correlation with high multi-reactivity. Additionally, the Spatial Aggregation Tendency (SAP) tool in Discovery Studio ¹ was used to identify excess hydrophobic patches (Figure 6). As shown in Figure 6, the hydrophobic patches appear small and insignificant. Based on this observation, the focus is on reducing the positive charge by removing positive charges or adding negative charges to disrupt the patches.

To determine the appropriate sites for mutagenesis, computational predictions of mutations for both affinity and stability changes were performed using Discovery Studio and FoldX. Tolerance mutations were those for which neither Discovery Studio nor FoldX methods predicted ΔΔG > 1 kcal/mol. From these predictions, a set of residues predicted to have minimal impact on binding and stability while still disrupting the altered patch were identified. Additionally, we favored hydrophilic mutations that did not increase hydrophobic surface area where possible. A list of possible CDR mutations introduced in CD3 antibody 2B5 to optimize biophysical properties is shown in Table 11. This includes mutations predicted to be tolerant at each site and reduce multireactivity and splicing.

Table 11

In addition, a set of multiple mutations (combinations) were designed for both the heavy and light chains. These include R52Q/R52bQ/R53Q, R52Q/R52bQ/R53Q/R96Q, R52bQ/R53Q/R96Q, and R52Q/R52bQ/R96Q for the heavy chain, and R29Q/K30Q/R54L/V27eE, R29Q_R54L_V27eE, and R29Q_K30Q_R54L for the light chain.

Example 6

Characterization of mutants

All mutants were generated as IgG molecules using standard expression and purification techniques well-known in the art, and their binding to CD3ed, multireactivity (binding to DNA and insulin), and AC-SINS were tested.

Combined assays to evaluate variants

Octet assays were designed and run to examine both the binding and dissociation rates of 10 μg/ml soluble human CD3ed antigen (SEQ ID NO: 40). A 10 μg/ml anti-CD3 antibody was captured using an anti-human IgG tip for 120 s, followed by a baseline in PBS for 30 s, then immersed in 10 μg/ml soluble human CD3ed antigen for 120 s, and finally dissociated in it for 300 s. Sensing patterns were examined to determine whether the antigen bound to the antibody, and Octet software was used to calculate kd(1/s) and kd error. Twenty-six viable mutations retained binding to the hCD3ed antigen. The binding kinetics of the 26 remaining antigen-binding CD3 antibody variants to soluble CD3 were then examined using Biacore analysis via SPR. CD3 antibodies were captured on anti-human Fc cells, and soluble CD3 antigen at concentrations of 0 nM, 3.7 nM, 11.1 nM, 33.3 nM, and 100 nM was injected onto the captured antibodies to determine binding kinetics. Twenty-three CD3 antibodies bound the soluble CD3 antigen with an affinity similar to that of the parental antibodies used in CDR transplantation.

DNA and insulin ELISA to measure multireactivity

384-well ELISA plates (Nunc Maxisorp) were coated overnight at 4°C with DNA (10 µg/ml) (Sigma-Aldrich, D1626) and insulin (5 µg/ml) (Sigma-Aldrich, I9278-5 mL) in PBS pH 7.5. The assay, modified from that described in Tiller et al. (17), was performed on a PerkinElmer Janus Automated Workstation liquid handling robot. Wells were washed with water, blocked for 1 hour at room temperature with 50 µl of multireactive ELISA buffer (PEB; PBS containing 0.05% Tween-20, 1 mM EDTA), and rinsed three times with water. Quadruple copies of serially diluted mAbs in 25 µl were added to the wells and incubated at room temperature for 1 hour. The plates were washed three times with water, and 25 µl of 10 ng/ml goat anti-human IgG (Face fragment specific) conjugated with horseradish peroxidase (Jackson ImmunoResearch, 109-035-008) was added to each well. The plates were incubated at room temperature for 1 hour, washed three times with 80 µl of water, and 25 µl of TMB substrate (Sigma Aldrich, T-0440) was added to each well. After approximately 7 minutes, the reaction was terminated by adding 25 µl of 0.18 M orthophosphate to each well, and the absorbance was read at 450 nm. The DNA and insulin binding score was calculated as the ratio of the ELISA signal of the 10 µg/ml antibody to the signal of the well containing buffer instead of the primary antibody. A score less than 8 was considered ideal and represented low multireactivity.

AC-SINS assay to measure self-association tendency

AC-SINS (Affinity Capture Self-interaction Nanoparticle Spectroscopy) assays were normalized in 384-well format on a Perkin-Elmer Janus liquid handling robot. 20 nm gold nanoparticles (Ted Pella, Inc., #15705) were coated with a mixture of 80% goat anti-human Fc (Jackson Immuno Research Laboratories, Inc. #109-005-098) and 20% nonspecific goat polyclonal antibody (Jackson Immuno Research Laboratories, Inc. #005-000-003), which was replaced with buffer at 20 mM sodium acetate pH 4.3 and diluted to 0.4 mg/mL. After incubation at room temperature for one hour, unoccupied sites on the gold nanoparticles were blocked with thiolized polyethylene glycol (2 kDa). The coated nanoparticles were then concentrated 10-fold using a syringe filter, and 10 µl was added to 100 µl of 0.05 mg/ml mAb in PBS pH 7.2. The coated nanoparticles and target antibody were incubated in 96-well polypropylene plates for 2 hours, and then transferred to 384-well polystyrene plates and read on a Tecan M1000 spectrophotometer. Absorbance was read from 450–650 nm in 2 nm increments, and a Microsoft Excel macro was used to identify the maximum absorbance, smooth the data, and fit the data using a second-order polynomial. The antibody AC-SINS score was determined by subtracting the smoothed maximum absorbance of the mean blank (PBS buffer only) from the smoothed maximum absorbance of the antibody sample. A score less than 8 was considered ideal and represented low self-association.

Evaluation of the lead CD3 antibody 2B5 variant in a bispecific Fc form

A comprehensive analysis of data from binding assays, AC-SINS, and multireactivity helped narrow down the binding variants to five. These five variants were reformatted into bispecific biantibody-Fc molecules using antitumor sequences and standard cloning strategies well-known in the art. These bispecific antibodies were transiently expressed and purified using standard techniques well-known in the art. Binding of the resulting bispecific antibodies to jurakat cells was tested, in vitro cytotoxicity was tested using a T-cell retargeting assay, multireactivity was tested, and thermostability was tested using DSC and AC-SINS. Of all five bispecific antibodies, only one, carrying the CD3 antibody 2B5 variant 2B5v6, showed potent binding and therefore potent cytotoxicity. This variant also showed reduced multireactivity, as evaluated by DNA and insulin binding and AC-SINS scores. The 2B5v6 bispecific antibody scored less than 8, compared to a score greater than 8 for the H2B4 bispecific antibody. Based on these results, 2B5v6 was selected as the lead optimized 2B5 variant for further analysis.

Example 7

Evaluation of shear in CDR-H2

When the bispecific antibodies were expressed in CHO cells but not in HEK293 cells, cleavage was observed in the VH CDR2 (SEQ ID NO: 16) at position R53 in the CD3 antibodies (H2B4 and 2B5). To mitigate the risk of this adverse effect and ensure better homogeneity of the product at manufacturing scale, care should be taken to select for mutations in the cleavage region, as described in previous examples, as well as other mutations to mitigate the risk of other adverse effects. To evaluate the cleavage, simply, bispecific antibody-Fc bispecific antibodies were generated using the variable domain sequence of the negative control antibody and CD3 antibody 2B5 or CD3 antibody H2B4 or CD3 antibody 2B5v6. All bispecific antibody-Fc molecules were prepared using a similar schematic diagram of expression and purification processes well known in the art. CHOSSI cell lines were generated using standard procedures. Two strands encoding a bispecific antibody were cloned into the mammalian expression vector pRY19-GA-Q, which contains a dual promoter and a recombinant site for unit site integration (SSI) (Zhang, L. et al. Biotechnology Progress. 2015; 31:1645-1656) into the CHO cell genome. The resulting plasmid, containing the target gene, was transfected into 10e9 CHO-K1 SV cells via electroporation using the pRY19 vector, followed by positive and negative selection pressures using hygromycin-B and ganciclovir. These CHO conjugates underwent a 3-week recovery period. When observed viability exceeded 90%, a cell library was generated for future work. The established CHO conjugates were then subjected to a 12-day fed-batch platform expression process in 1L of CD-CHO medium (Thermo Fisher Scientific, Waltham, MA). The day 12 culture harvest was captured on a Protein A column and eluted at low pH. The proA conjugate was immediately neutralized to pH 8.1. The purified protein was evaluated using capillary gel electrophoresis (cGE) according to standard protocols well-known in the art. As shown in Table 12 and Figure 7, the 2B5 bispecific antibody showed significantly reduced shearing compared to H2B4, and 2B5v6 showed no shearing while H2B4 showed significantly increased shearing or fragmentation. Table 12 shows the results of capillary gel electrophoresis demonstrating fragmentation of the CD3-GUCY2c bispecific antibody. POI% indicates the target peak.

Table 12

Reduced cGE sample	%POI	%Fragmentation
			Negative control - H2B4	88.8	11.2
Negative control - 2B5	98.9	1.1
			Negative control - 2B5v6	99.6	0.4

Example 8

GUCY2c-2B5v6 Bispecific Antibody: Expression and Purification of Chimeric CD3-GUCY2c Bispecific Antibody Using CD3 Antibody 2B5v6

Complementary construct pairs (12.5 µg per strand) were co-transfected into 25 mL of logarithmic-phase culture containing 1 million cells/mL HEK293 cells using the ExpiFectamine™ 293 Transfection Kit (Life Technologies). Twenty-four hours post-transfection, ExpiFectamine Transfection Enhancer was added, and cells were allowed to grow for an additional 4–5 days before harvest. The used culture was then collected, centrifuged to remove cell debris, and passed through a 20 µm filter.

The clarified conditioned medium containing the GUCY2c T cell bispecific antibody was then purified using protein A affinity chromatography. Samples were loaded onto 0.45 mL microcolumns pre-loaded with MabSelect SuRe protein A resin (GE Healthcare) using a liquid delivery device (Tecan). The bound protein was washed with PBS pH 7.2, eluted with 20 mM citric acid and 150 mM sodium chloride pH 3.5, and neutralized with 2 M tris at pH 8.0. The sample was then desalted to PBS pH 7.2 using G25 Sephadex drip columns (GE Healthcare) according to the manufacturer's instructions. The purity of the purified protein was analyzed on an Aglient 1200 HPLC system using analytical size exclusion chromatography (TOSOH) with a Mab HTP column, following the manufacturer's protocol. Concentrations were determined using a microspectrophotometer (Trinean) at OD 280 nm.

The stability of bispecific antibodies was evaluated using differential scanning calorimetry.

The protein was diluted to 0.6 mg/mL in phosphate-buffered saline (PBS) solution at a volume of 400 µL. PBS was used as a buffer blank in the reference chamber. PBS contained 137 mM NaCl, 2.7 mM KCl, 8.1 mM Na₂HPO₄, and 1.47 mM KH₂PO₄ at pH 7.2. Samples were dispensed into the sample trays of a MicroCal VP-Capillary DSC using an autosampler (Malvern Instruments Ltd, Malvern, UK). Samples were equilibrated at 10 °C for 5 min and then scanned at 100 °C per hour up to 110 °C. A 16-second filtration period was selected. The raw data were baseline corrected and the protein concentration was normalized. The data were fitted to the MN2-State model using Origin software 7.0 (OriginLab Corporation, Northampton, MA) with an appropriate number of transformations. As shown in Figure 8, the GUCY2c-2B5v6 bispecific antibody exhibits excellent thermal stability, with _Tm1 being 70℃.

Binding with naïve T cells

Using a BD LSRII Fortessa analyzer, purified CD3 bispecific antibody was titrated to bind to the cell surface of naïve human CD3 on virgin human T cells purified from fresh human peripheral blood mononuclear cells (PBMCs) using standard flow cytometry methods known in the art. As shown in Figure 9, the GUCY2c-2B5v6 bispecific antibody bound very well to T cells, with a binding EC50 of 35.44 nM.

Evaluation of T cell activity mediated by bispecificity

Human PBMCs were isolated from healthy donor blood using Histopaque-177 (Sigma). Naïve T cells (negatively selected T cells) were isolated from PBMCs using a T cell enrichment kit from Stem Cell Technologies. T84 human tumor cells expressing GUCY2c, transfected with a luciferase expression construct, were treated with serially diluted GUCY2c bispecific antibody or a negative control CD3 bispecific antibody, along with the T cells. The effector cell to target cell ratio (E:T) was 10:1 or 5:1. T cells and tumor cells were incubated at 37°C for 48 hours, followed by measurement of luciferase signal using neolite reagent (Perkin Elmer). EC50 values were calculated in Graphpad PRISM using a four-parameter nonlinear regression analysis. Target-negative HCT116 or HT29 cells did not show any GUCY2c bispecificity-mediated cytotoxicity. As shown in Figure 10, GUCY2c-2B5v6 bispecificity demonstrated potent cytotoxicity in the T cell retargeting assay.

T84 cells were treated with T cells (E:T ratio 5:1) and different doses of GUCY2c-1608 (GUCY2c-2B5v6). Supernatants were collected at different time points (24 h, 48 h, 96 h) and analyzed using multiplex Luminex assays according to the manufacturer's guidelines and read on a Luminex 200 with xPONENT software. The measured cytokines were human IFN-γ, IL10, IL2, IL4, IL6, and TNF-α. Figure 11 shows the upregulation of (11A) IFN-γ, (11B) IL10, (11C) IL2, (11D) IL4, (11E) IL6, and (11F) TNF-α cytokines after T cell activation with GUCY2c-1608.

In vivo assessment of bispecific activity mediated by GUCY2c-2B5v6

In vivo efficacy studies were conducted using an adoptive metastasis model. Tumor cell lines/patient-derived xenograft fragments were implanted into NSG mice and staged to approximately 200 ^mm³ . The bispecific compound was administered intravenously to the mice (unless otherwise noted). Twenty-four hours after bispecific antibody administration, two million activated and expanded T cells were administered intravenously (using a T cell expansion and activation kit from Miltenyi Biotec). The bispecific antibody was administered weekly. As shown in Figures 12 and 13, the GUCY2c-2B5v6 (GUCY2c-1608) bispecific antibody demonstrated excellent durable tumor regression in both the PDX CRX-11201 and LS1034 cell lines, representing colorectal cancer, in vivo models.

Example 9

Generation and cytotoxicity of FLT3-CD3 bispecific antibody and FLT3-2B5v6 bispecific IgG using an optimized CD3 antibody (2B5v6).

Human FLT3 antibody (xFLT3) and human CD3 antibody (2B5v6) are expressed as engineered human IgG2dA_D265A with EEEE on one arm and RRRR on the other arm, respectively, for bispecific exchange at positions 223, 225, and 228 (e.g., (C223E or C223R), (E225E or E225R), and (P228E or P228R)) in the hinge region of human IgG2 and at positions 409 or 368 (e.g., K409R or L368E (EU numbering scheme)) in the CH3 region. The antibodies also have a mutation from D to A at position 265 (EU numbering scheme).

Individual antibody arms were purified separately using Protein A resin. Then, 1 mg/mL of purified xFLT3 antibody was exchanged with 1 mg/mL of purified 2B5v6 antibody in the presence of 2 mM reduced glutathione. The exchange reaction was carried out at 37°C for 16 h. Then, 2 mM oxidized glutathione was added to the protein, and the mixture was incubated at 37°C for 30 min. The mixture was then diluted 1:5 to 20 mM MES pH 5.4 buffer without NaCl. The diluted sample was loaded onto a MonoS ion exchange column and washed with 20 mM MES pH 5.4 + 25 mM NaCl buffer. The bispecific antibody xFLT3-2B5v6 was then eluted with a 0–500 mM NaCl gradient in 20 mM MES pH 5.4 buffer. Finally, the bispecific antibody was dialyzed into phosphate-buffered saline (PBS) for cell-based assays. For the cytotoxicity assay, unstimulated human T cells and luciferase-expressing target cells (Eol-1 or MV411) were co-incubated in 96-well U-shaped plates in 200 μl RPMI + 10% FBS to determine the ratio. xFLT3-2B5v6 bispecific antibody was serially diluted in PBS and added to each well. After 24 hours, 100 μl of cell suspension was mixed with 100 μl of One-GLO reagent in 96-well white-walled plates, and luminescence was measured on an Envision photometer. The cytotoxicity value was determined using the following formula:

%Cytotoxicity = [1 - (RLU sample) / RLU target cells only)] × 100RLU: relative light units

As shown in Figures 14A and 14B, when tested on two cell lines expressing higher copies of FLT3 (EOL-1) and lower copies (MV4-11), the xFLT3-2B5v6 bispecific antibody containing the optimized CD3 antibody (2B5v6) appears to be highly potent when presented as an alternative to the bispecific IgG form.

Table 13 shows the light and heavy chain sequences of FLT3 (EOL-1) and the light and heavy chain sequences of the 2B5v6 bispecific antibody.

Table 13

Example 10

T-cell epitopes of the 2B5v6 bispecific antibody

Optimized analysis of potential T-cell epitopes in clone 2B5v6 was performed to de-risk the possibility of immunogenicity. This analysis was performed using the Epivax tool for predicting potential T-cell epitopes on a computer chip. This tool predicts the number of times a 9-mer framework peptide binds to any of eight different HLA alleles. Hit was defined as those in the top 5% with a Z-score in the top 1%. Hit to four or more alleles or one strong hit was considered a predicted T-cell epitope. Based on these criteria, three predicted non-germal T-cell epitopes in CDRH2, two predicted germal epitopes in CDRL1, and one predicted germal epitope in L2 were identified (Table 15). In addition to the scores of individual peptides, the overall risk of the sequence was defined by the total number of hits or the T-regitope-adjusted Epivax score, which had been optimized to correlate with the prevalence of clinical ADA. Here, for these predicted values, the lower the score, the lower the expected risk of ADA. The VH domain had 70 hits and a score of -42.82, while the VL domain had 60 hits and a score of -23.79.

To reduce the risk of immunogenicity, mutations that would decrease the overall score of this clone while removing hits and ideally predicted T-cell epitopes were identified. Any possible single mutant variations of CDR-L1, CDR-L2, and CDR-H2 were considered to identify mutations that would reduce hits, decrease the overall t-regitope-adjusted EpiMatrix score, and remove T-cell epitopes. Since these potential mutations could also affect binding and stability, care was taken to consider the predicted stability and affinity data described above, as well as to identify tolerant mutations. Tables 14A, 14B, and 14C show mutations designed to reduce the immunogenicity risk of 2B5v6. Additionally, mutations predicted as shown in SEQ ID NOs 44-47, 51-57, 60, and 62 reduced the number of T-cell epitopes.

Table 14A

Table 14B

Table 14C

Table 15 shows the regions of the variable domains of a CD3 antibody (2B5v6) that has T-cell epitopes and poses a potential risk of immunogenicity.

Table 15

All mutants were generated as monovalent IgG antibodies using standard transient HEK293 expression, followed by purification using a protein A column. The binding of the resulting clones to CD3-expressing Jurkat cells was tested using FACS, and the likelihood of self-association was tested using AC-SINS as described in Example 7 above. As shown in Table 16, five molecules with improved epivax scores and reduced T-cell epitopes, while retaining affinity for CD3, were selected for further testing as biantibody-Fc variants. The 2B5v6 variant with improved epivax scores was compared to the parental 2B5v6. Results showed comparable binding to CD3-expressing Jurkat cells and AC-SINS scores.

Table 16

Of the mutants tested, several exhibited lower binding to Jurkat cells compared to 2B5v6. Among these mutants, those with all T-cell epitopes eliminated and improved epivax scores were further advanced to generate bispecific antibody-Fc molecules in which the GUCY2c antibody serves as the other binding domain. Table 17 shows the epivax scores, and Figure 15 shows the reduced binding of these mutants to Jurkat cells. The low-affinity CD3 antibody 2B5v6 variant showed an improved epivax score.

Table 17

CD3 antibody variants	Epivax Score (tReg - Adjusted)
		2B5v6-0598	-65.15
2B5v6-0623	-56.74

Example 11

Three anti-CD3 variants showed improved affinity and cytotoxicity in the GUCY2C-CD3 bispecific antibody.

Using the standard cloning, expression, and purification techniques described above, three anti-CD3 variants derived from 2B5v6, 2B5-1038 (SEQ ID NO: 4 and 9), 2B5-1039 (SEQ ID NO: 4 and 87), and 2B5-1040 (SEQ ID NO: 4 and 89), were reformulated into bispecific biantibody-Fc molecules paired with the antitumor GUCY2c antibody sequence, one of the configurations shown in Figure 1. In vitro cytotoxicity of the resulting bispecific antibodies was assessed using T-cell retargeting assays as shown in Table 18, binding affinity was tested using surface plasmon resonance (SPR), and nonspecificity was tested using AC-SINs. Immunogenic risk was evaluated using the Epivax Tool, a computer-based assay for potential T-cell epitopes. Complete mass spectrometry data indicated that all bispecific antibodies were correctly paired and 100% heterodimer. Table 18 shows that all three bispecific antibodies (1) were approximately 2-fold more potent in in vitro cytotoxicity assays compared to the control bispecific GUCY2C-1608 paired with anti-CD3 2B5v6 (Fig. 6); (2) had binding affinity for recombinant human CD3 via SPR, consistent with cytotoxic activity; (3) had improved immunogenicity scores compared to the control; and (4) AC-SINs scores remained the same as the control bispecific GUCY2C-1608. These results indicate that anti-GUCY2c bispecific antibodies with different CD3 variants recruit naïve human T cells to induce cell killing in T84 tumor cells (Fig. 16). The three anti-CD3 variants derived from 2B5v6 showed approximately 2-fold greater potency compared to the control bispecific GUCY2C-1608.

Table 18

Bispecific

CD3 variant

Epivax Score (tReg - Adjusted)

MassSpec (Complete)

AC-SINs score

In vitro cytotoxicity (nM)

KD (nM, SPR)

GUCY2C-1608

2B5-0542

-33.7

100% heterodimer

3

0.772

127.88 ± 1.63

GUCY2C-1678

2B5-1038

-52.61

100% heterodimer

2

0.247

63.55±0.49

GUCY2C-1679

2B5-1040

-47.58

100% heterodimer

3

0.194

72.02 ± 0.36

GUCY2C-1680

2B5-1039

-53.51

100% heterodimer

3

0.385

78.57±0.9

The sequences of the bispecific antibodies are shown in Table 19.

Table 19

sequence list

This application was filed electronically via EFS-Web and includes a sequence list submitted electronically in .txt format. The .txt file contains a sequence list titled "PC72375-PRV2_SequenceListing_ST25_05132019.txt", created on May 13, 2019, and is 100 KB in size. The sequence list contained in this .txt file is part of this specification and is incorporated herein by reference in its entirety.

Invention Field

This invention relates to CD3-specific antibodies and their antigen-binding fragments, as well as methods of using them. The invention also relates to multispecific antibodies that bind to both CD3 and target antigens, and methods of using them.

Background of the Invention

Differentiation cluster 3 (CD3) is a homodimeric or heterodimeric antigen expressed on T cells that associates with the T cell receptor complex (TCR) and is required for T cell activation. T cell activation is a complex phenomenon that depends on the involvement of multiple cell surface molecules expressed on a population of reactive T cells. Functional CD3 is formed by the association of two of four different chains: ε, ζ, δ, and γ. For example, the antigen-specific T cell receptor (TCR) is composed of a disulfide-linked heterodimer containing two clonal, monolithic membrane glycoprotein chains, alpha and beta (α and β), or gamma and delta (γ and δ), non-covalently associated with a complex of a low molecular weight invariant protein (usually designated CD3). CD3 dimer arrangements also include γ/ε, δ/ε, and ζ/ζ.

Antibodies targeting CD3 have been shown to accumulate CD3 on T cells, thereby inducing T cell activation in a manner similar to the binding of MHC molecules loaded with peptides to the TCR. Therefore, CD3 antibodies can be used for therapeutic purposes involving T cell activation.

Invention Overview

This invention provides antibodies that bind to CD3 and compositions (such as pharmaceutical compositions) containing one or more CD3 antibodies. Antibodies according to this aspect of the invention are particularly useful for targeting CD3-expressing T cells and for stimulating T cell activation, for example, in cases where T cell-mediated killing is beneficial or desired. The CD3 antibodies of this invention may be included as part of a multispecific antibody (e.g., a bispecific antibody) that directs CD3-mediated T cell activation to specific cell types, such as tumor cells or infectious pathogens.

In one aspect, the present invention provides an antibody that specifically binds to CD3, wherein the antibody comprises: (a) a heavy chain variable (VH) region comprising VH complementarity-determining region 1 (VH CDR1), VH complementarity-determining region 2 (VH CDR2), and VH complementarity-determining region 3 (VH CDR3) of the VH sequence shown in SEQ ID NO: 2, 4, 5, 7, 10, 12, or 35; and/or (b) a light chain variable (VL) region comprising VL complementarity-determining region 1 (VL CDR1), VL complementarity-determining region 2 (VL CDR2), and VL complementarity-determining region 3 (VL CDR3) of the VL sequence shown in SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87, or 89.

In one embodiment of this aspect, the CD3 antibody comprises: (a) a VH region comprising (i) a VH CDR1 comprising the sequence of SEQ ID NO: 13, 14 or 15; (ii) a VH CDR2 comprising the sequence of SEQ ID NO: 16, 17, 19, 20, 21, 22, 23 or 24; and (iii) a VH CDR3 comprising the sequence of SEQ ID NO: 18 or 25; and/or (b) a VL region comprising (i) a VL CDR1 comprising the sequence of SEQ ID NO: 26, 29, 30, 32, 91 or 92; (ii) a VL CDR2 comprising the sequence of SEQ ID NO: 27 or 31; and (iii) a VL CDR3 comprising the sequence of SEQ ID NO: 28 or 33.

In some such embodiments, the present invention provides an antibody wherein: (a) the VH region contains a sequence of SEQ ID NO: 2, 4, 5, 7, 10, 12 or 35; and/or (b) the VL region contains a sequence of SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87 or 89.

In other embodiments, the antibody comprises a VL/VH amino acid sequence selected from the following: SEQ ID NO: 1 and 2; SEQ ID NO: 3 and 2; SEQ ID NO: 3 and 4; SEQ ID NO: 3 and 5; SEQ ID NO: 6 and 7; SEQ ID NO: 8 and 7; SEQ ID NO: 6 and 4; SEQ ID NO: 8 and 4; SEQ ID NO: 9 and 10; SEQ ID NO: 9 and 7; SEQ ID NO: 11 and 12; SEQ ID NO: 34 and 35; SEQ ID NO: 9 and 4; SEQ ID NO: 87 and 4; and SEQ ID NO: 89 and 4.

In some of the aforementioned embodiments, the antibody is selected from: Fab, Fab fragment, F(ab) ₂ fragment, Fv fragment, single-chain Fv fragment, disulfide-stabilized Fv fragment, single-chain antibody, monoclonal antibody, chimeric antibody, bispecific antibody, trispecific antibody, multispecific antibody, bispecific heterodimeric biantibody, bispecific heterodimeric IgG, polyclonal antibody, labeled antibody, humanized antibody, human antibody and fragment thereof.

In one specific embodiment, the antibody is a bispecific antibody. In some such embodiments, the bispecific antibody specifically binds to tumor antigens.

In some embodiments, the antibody of the present invention further comprises a human or humanized VH framework and a human or humanized VL framework. In some such embodiments, the antibody is a humanized antibody.

The present invention further provides pharmaceutical compositions comprising a CD3 antibody as further disclosed herein and a pharmaceutically acceptable carrier.

In another aspect, the present invention provides a method for modulating a T-cell-mediated immune response in a subject with such need, comprising administering an effective amount of an antibody or pharmaceutical composition as disclosed herein to the subject such that the immune response in the subject is modulated.

In one aspect, the present invention provides a method for inhibiting the growth of tumor cells in a subject, comprising administering to the subject, in an amount that effectively inhibits the growth of tumor cells, a bispecific antibody, a trispecific antibody, a multispecific antibody, a bispecific heterodimeric antibody, or a bispecific heterodimeric IgG as disclosed herein.

In some such embodiments, the tumor cells are cancer cells. In specific embodiments, the cancer is selected from: breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer, bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and cancers of the digestive system.

In one embodiment, the cancers of the digestive system are selected from: esophageal cancer, gastric cancer, small bowel cancer, colon cancer, rectal cancer, anal cancer, liver cancer, gallbladder cancer, appendix cancer, bile duct cancer, and pancreatic cancer.

In one embodiment, the present invention provides an antibody or a pharmaceutical composition comprising the antibody for use in a therapy.

In another embodiment, the present invention provides the use of the antibody of the present invention in the preparation of a medicament for use in a therapy. In some such embodiments, the therapy comprises activating a cytolytic T-cell response.

In one aspect, the present invention provides a polynucleotide comprising a nucleotide sequence encoding an antibody as disclosed herein. In another aspect, the present invention provides a vector comprising this polynucleotide.

In another aspect, the present invention provides host cells comprising vectors as disclosed herein. In some such embodiments, the host cells recombinantly produce antibodies as disclosed herein. In specific embodiments, the host cells are selected from bacterial cell lines, mammalian cell lines, insect cell lines, and yeast cell lines. In one specific embodiment, the mammalian cell line is a CHO cell line.

In one aspect, the present invention provides a method for generating antibodies as disclosed herein, comprising: culturing host cells under conditions that result in the generation of antibodies as disclosed herein, and purifying said antibodies from the culture supernatant.

In another aspect, the present invention provides the use of CD3 antibodies, polynucleotides, vectors or host cells as disclosed herein in the preparation of medicaments for treating diseases.

In one aspect, the present invention provides an antibody or pharmaceutical composition for use in a method of modulating a T-cell-mediated immune response in a subject with such a need. In a specific embodiment, the present invention provides an antibody or pharmaceutical composition for use in a method of inhibiting the growth of tumor cells in a subject.

In another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, optionally wherein said cancer is selected from: breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer, bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and cancers of the digestive system.

In yet another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, wherein a T-cell-mediated immune response is modulated or wherein the growth of tumor cells is inhibited.

Other embodiments will become apparent from the detailed summary. When aspects or embodiments of the invention are described in terms of the Markush group or other alternative groupings, the invention covers not only the entire group listed as a whole, but also each member of that group individually and all possible subgroups of the main group, as well as the main group in which one or more members are not present. The invention also contemplates the explicit exclusion of any one or more members of the claimed invention.

Brief description of the figure/attachment

Figure 1 provides a schematic diagram of the architecture of a bispecific antibody having a first heterodimer promoting domain and a second heterodimer promoting domain, the bispecific antibody comprising an Fc chain optimized for association via mortar association.

Figure 2 depicts the results of the differential scanning calorimetry temperature recording, showing the heat capacity of the GUCY2c-H2B4 and GUCY2c-2B5 bispecific antibodies as a function of temperature.

Figure 3 depicts the binding of the bispecific antibodies GUCY2c-H2B4 (GUCY2c-0247) and GUCY2c-2B5 (GUCY2c-0250) to naïve human T cells as determined by flow cytometry.

Figure 4 depicts the results of in vitro cytotoxicity mediated by the bispecific antibodies GUCY2c-H2B4 (GUCY2c-0247) and GUCY2c-2B5 (GUCY2c-0250) using cell survival data.

Figure 5 depicts the potential energy surface (PES) of the H2B4 Fv region from an X-ray crystal structure. This includes three views highlighting the VH region, the VH/VL interface, and the VL region. The CDR (Crystal Radiata Diffraction) with excess positive charge is indicated for reference. Dark black represents positive charge and white represents negative charge.

Figure 6 depicts the spatial aggregation tendency (SAP) surface of the Fv region of H2B4. Regions with concentrated hydrophobic residues are marked.

Figure 7 depicts the heterogeneity among bispecific antibodies containing CD3 antibody-binding domains H2B4, 2B5, and 2B5v6 using capillary gel electrophoresis.

Figure 8 depicts the temperature record using differential scanning calorimetry (DSC), showing the heat capacity of the GUCY2c-1608 (GUCY2c-2B5v6) bispecific antibody as a function of temperature.

Figure 9 depicts the binding of the GUCY2c-1608 (GUCY2c-2B5v6) bispecific antibody to naive T cells.

Figure 10 depicts cell survival data for in vitro cytotoxicity mediated by the bispecific antibody GUCY2c-1608 (GUCY2c-2B5v6).

Figures 11A to 11F depict in vitro cytokine release measurements of the GUCY2c-1608 (GUCY2c-2B5v6) bispecific antibody. The bispecific antibody recruited naïve human T cells to GUCY2c-expressing T84 cells to induce in vitro cytokine release. Luminex-based assays showed upregulation of human IFN-γ (Figure 11A), IL10 (Figure 11B), IL2 (Figure 11C), IL4 (Figure 11D), IL6 (Figure 11E), and TNF-α (Figure 11F).

Figure 12 illustrates the dose-dependent tumor growth inhibition of the bispecific antibody GUCY2c-1608 (GUCY2c-2B5v6) in the patient-derived xenograft PDX-CRX-11201 of colorectal cancer in the adoptive metastasis system.

Figure 13 illustrates the dose-dependent tumor growth inhibition of the bispecific antibody GUCY2C-1608 (GUCY2C-2B5v6) in the xenograft LS1034 colorectal cancer cell line in the adoptive metastasis system.

Figure 14A depicts the results of cell viability data, showing the in vitro cytotoxicity mediated by the FLT3-2B5v6 bispecific antibody using Flt3-expressing MV4-11 cells at different effector-to-target (E:T) ratios. Figure 14B depicts the results of cell viability data, showing the in vitro cytotoxicity mediated by the FLT3-2B5v6 bispecific antibody using Flt-3-expressing EOL-1 cells at different effector-to-target (E:T) ratios.

Figure 15 depicts the binding of the low-affinity 2B5v6 variant to Jurkat cells expressing CD3.

Figure 16 depicts a graph summarizing the findings, showing the cytotoxicity mediated by anti-GUCY2C bispecific T cells with anti-CD3 variants.

Detailed Explanation

The present invention disclosed herein provides antibodies that specifically bind to CD3 (e.g., human CD3), including full-length antibodies or antigen-binding fragments thereof. The invention also provides polynucleotides encoding these antibodies, compositions comprising these antibodies, and methods for preparing and using these antibodies. The invention further provides methods for treating conditions using antibodies as described herein.

General Technology

Unless otherwise indicated, the practice of this invention will employ conventional techniques of molecular biology (including recombinant techniques), microbiology, cell biology, biochemistry, and immunology, which are within the scope of the art. Such techniques are fully explained in the literature such as *Molecular Cloning: A Laboratory Manual, 2nd Edition* (Sambrook et al., 1989); *Cold Spring Harbor Press*; *Oligonucleotide Synthesis* (M.J. Gait, ed., 1984); *Methods in Molecular Biology*, *Humana Press*; *Cell Biology: A Laboratory Notebook* (J.E. Cellis, ed., 1998); *Academic Press*; *Animal Cell Culture* (R.I. Freshney, ed., 1987); *Introduction to Cell and Tissue Culture* (J.P.). Mather and P.E. Roberts, 1998) Plenum Press; Cell and Tissue Culture: LaboratoryProcedures (A.Doyle, J.B. Griffiths, and D.G. Newell, eds., 1993-1998) J. Wileyand Sons; Methods in Enzymolog y (Academic Press, Inc.); Handbook of Experimental Immunology (D.M. Weir and C.C. Blackwell, Eds.); Gene Transfer Vectors for Mammalian Cells (J.M. Miller and M.P. Calos, Eds., 1987); Current Protoc Mullis in Molecular Biology (F.M. Ausubel et al., eds., 1987); PCR: The Polymerase Chain Reaction (Mullis et al., eds., 1994); Current Protocols in Immunology (J.E. Colligan et al., eds., 1991); Short Protocols in Molecular Biology (Wiley and Sons, 1999); Immunobiology (C.A. Janeway and P. Travers, 1997); Antibodies (P. Finch, 1997); Antibodies: a practical approach (D. Catty., ed., IRL Press, 19...). 88-1989); Monoclonal Antibodies: A Practical Approach (P. Shepherd and C. Dean, eds., Oxford University Press, 2000); Using Antibodies: A Laboratory Manual (E. Harlow and D. Lane (Cold Spring Harbor Laboratory Press, 1999); The Antibodies (M. Zanetti and J.D. Capra, eds., Harwood Academic Publishers, 1995). In some cases, terms having their usual meaning are defined herein for clarity and/or for the purpose of preparing references, and the inclusion of such definitions herein is not necessarily construed as indicating a material difference from definitions commonly understood in the art.

The invention will now be described in detail by reference using the following definitions and examples. All patents and publications mentioned herein (including all sequences disclosed in such patents and publications) are expressly incorporated by reference.

definition

Generally, unless otherwise defined, all terms, symbols, and other scientific or terminological terms used herein are intended to have the meaning commonly understood by one of ordinary skill in the art to which this invention relates. For example, the term "and/or" as used herein, such as in phrases like "A and/or B," is intended to include both A and B; A or B; A (alone); and B (alone). Similarly, the term "and/or" as used herein, such as in phrases like "A, B, and/or C," is intended to cover each of the following embodiments: A, B, and C; A, B, or C; A or C; A or B; B or C; A and C; A and B; B and C; A (alone); B (alone); and C (alone).

As used herein, unless the context clearly indicates otherwise, the singular forms “a”, “an” and “the” include their respective multiple indicators.

As used in this article, the numerical range includes the numerical values that define the range.

The terms “about,” “approximately,” etc., when preceding a list of numerical values or ranges, independently refer to each individual value in that list or range as if each individual value immediately followed the term. The term implies that the value referred to by the term is the same as, close to, or similar to. For example, in some embodiments, “about” or “approximately” for a particular value may indicate 99%, 95%, or 90% of that value. As an example, the expression “about 100” includes 99 and 101 and all values between them (e.g., 99.1, 99.2, 99.3, 99.4, 99.5, etc.). As another example, if the temperature is 70°C, “about” or “approximately” 70°C may be equal to 69°C, 66°C, or 63°C. It should be understood that these are merely examples.

As used herein, nucleic acids are written from left to right in a 5' to 3' orientation; amino acid sequences are written from left to right in an amino-to-carboxyl orientation. For definitions and terminology in this art, practitioners particularly refer to Sambrook et al., 1989 and Ausubel FM et al., 1993. It should be understood that the invention is not limited to the specific methods, schemes, and reagents described, as these are subject to variation.

The terms “polypeptide,” “oligopeptide,” “peptide,” and “protein” are used interchangeably herein to refer to an amino acid chain of any length, preferably relatively short (e.g., 10-100 amino acids). This chain may be linear or branched, may contain modified amino acids, and/or may be interrupted by non-amino acid components. The term also covers amino acid chains that have been modified naturally or through intervention; for example, disulfide bond formation, glycosylation, esterification, acetylation, phosphorylation, or any other operation or modification, such as conjugation with a labeled component. This definition also includes, for example, polypeptides containing one or more amino acid analogs (including, for example, non-natural amino acids) and other modifications known in the art. It should be understood that polypeptides may exist as single chains or associated chains. Preferably, mammalian polypeptides (polypeptides originally derived from mammalian organisms) are used, more preferably those secreted directly into the culture medium.

An "antibody" is an immunoglobulin molecule that specifically binds to a target, such as carbohydrates, polynucleotides, lipids, peptides, etc., through at least one antigen recognition site located in the variable region of the immunoglobulin molecule. As used herein, the term encompasses polyclonal antibodies, monoclonal antibodies, chimeric antibodies, bispecific antibodies, dual-specific antibodies, bifunctional antibodies, trispecific antibodies, multispecific antibodies, bispecific heterodimeric biantibodies, bispecific heterodimeric IgG, labeled antibodies, humanized antibodies, human antibodies and their fragments (such as Fab, Fab', F(ab') ₂ , Fv), single-chain (ScFv) and domain antibodies (including, for example, antibodies from sharks and camels), fusion proteins containing antibodies, and any other modified conformation of an immunoglobulin molecule containing an antigen recognition site. Antibodies include any class of antibodies, such as IgG, IgA, or IgM (or their subclasses), and the antibody need not be of any particular class. Immunoglobulins can be designated as different classes depending on the amino acid sequence of their heavy chain constant region. There are five main classes of immunoglobulins: IgA, IgD, IgE, IgG, and IgM, and several of these can be further subdivided into subclasses (isotypes), such as IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2. The heavy chain constant regions corresponding to different classes of immunoglobulins are designated α, δ, ε, γ, and μ, respectively. The subunit structures and three-dimensional conformations of different classes of immunoglobulins are well known. This invention also includes "antibody analogs," other scaffolds based on non-antibody molecular proteins, such as fusion proteins and/or immunoconjugates using CDRs to provide specific antigen binding. The antibodies of this invention can be derived from any species, including but not limited to mice, humans, camels, llamas, fish, sharks, goats, rabbits, chickens, horses, and cattle.

The term "antibody" further includes immunoglobulin molecules comprising four polypeptide chains (two heavy chains (H) and two light chains (L)) linked by disulfide bonds, as well as their multimers (e.g., IgM). Each heavy chain comprises a heavy chain variable region (abbreviated herein as HC VR or VH) and a heavy chain constant region. The "variable region" of an antibody refers to the variable region of either the antibody light chain alone or in combination with the variable region of the antibody heavy chain. The heavy chain constant region comprises three domains: CH1, CH2, and CH3. The CH1 and CH2 domains are linked by hinge regions. Each light chain comprises a light chain variable region (abbreviated herein as LC VR or VL) and a light chain constant region. The light chain constant region comprises one domain (CL1). The VH and VL regions can be further subdivided into hypervariable regions (called complementarity-determining regions (CDRs)) which are interspersed with more conserved regions (called framework regions (FRs)). Each VH and VL consists of three CDRs and four FRs, arranged in the following order from the amino terminus to the carboxyl terminus: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. In different embodiments of the invention, the FRs of the CD3 antibody (or its antigen-binding portion) may be identical to the human germline sequence, or may be natural or artificially modified. The common amino acid sequence can be defined based on a side-by-side analysis of two or more CDRs. The CDRs in each chain are tightly held together by the FRs and, together with the CDRs from the other chain, contribute to the formation of the antigen-binding site of the antibody.

As used herein, the terms “antigen-binding fragment,” “antibody fragment,” or “antigen-binding portion” refer to one or more segments of an antibody that retain the ability to specifically bind to an antigen. Examples of binding fragments encompassed within the term "antigen-binding fragment" in the term antibody include (i) antibody heavy chain variable domains (VH) and/or antibody light chain variable domains (VL), or VH/VL pairs derived from full-length antibodies or antibody fragments (such as VH and/or VL domains); (ii) Fab fragments, i.e., monovalent fragments consisting of VL, VH, CL, and CH1 domains; (iii) Fab' fragments, which are essentially Fab fragments having a portion of a hinge region (e.g., Fundamental Immunology, edited by Paul, 3rd edition, 1993); (iv) F(ab') ₂ fragments, i.e., bivalent fragments containing two Fab fragments connected by a disulfide bridge at the hinge region; (v) Fd fragments, which consist of VH and CH1 domains; (vi) Fv fragments, which consist of the VL and VH domains of a single arm of the antibody; (vii) Single-chain Fv fragments (scFv), i.e., a single protein chain in which the VL and VH regions pair to form a monovalent molecule (e.g., Bird et al. (1988) Science 242:423-426; and Huston et al. (1988) Proc. Natl. Acad. Sci. USA 85:5879-5883); (viii) disulfide-stabilized Fv fragments (dsFv), i.e., Fv with engineered intermolecular disulfide bonds that stabilize the VH-VL pair; (ix) single variable domain antibody (sdAb or dAb) fragments (e.g., Ward et al. (1989) Nature 341:544-546), which consist of a variable domain of a heavy chain and have no light chain; (x) complementarity-determining regions (CDRs); and any derivative thereof.

As used herein, an antibody's "antigen-binding fragment" may contain at least one variable domain. The variable domain may have any size or amino acid composition and will typically contain at least one CDR adjacent to or framed with one or more framework sequences. As used herein, an "antigen-binding fragment" of an antibody may comprise a homodimer or heterodimer (or other multimer) of any variable region and constant domain configuration that is non-covalently associated with each other and/or non-covalently associated (e.g., via disulfide bonds) with one or more monomeric VH or VL regions. For example, the variable region may be dimer and contain VH-VH, VH-VL, or VL-VL dimers. The configurations of the variable and constant domains that can be found within the antigen-binding fragment of the antibody of the present invention include: VH-CH1; VH-CH2; VH-CH3; VH-CH1-CH2; VH-CH1-CH2-CH3; VH-CH2-CH3; VH-VL-CL, VH-VL-CH1, VH-VL-CH2; VH-CL; VL-CH1; VL-CH2; VL-CH3; VL-CH1-CH2; VL-CH1-CH2-CH3; VL-CH2-CH3; and VL-CL. The variable and constant domains can be directly connected to each other or can be connected via complete or partial hinge or linker regions. The hinge region can consist of at least two (e.g., five, ten, fifteen, twenty, forty, sixty, or more) amino acids, which form flexible or semi-flexible bonds between adjacent variable and/or constant domains in a single polypeptide molecule.

As used herein, a “binding domain” includes any region of a polypeptide (e.g., an antibody) responsible for selectively binding to a target molecule (e.g., an antigen, ligand, receptor, substrate, or inhibitor). Exemplary binding domains include antibody variable regions, receptor-binding domains, ligand-binding domains, and enzyme-catalyzing domains.

As used herein, the term "human receptor framework" is a framework comprising an amino acid sequence of a light chain variable (VL) framework or a heavy chain variable (VH) framework derived from the human immunoglobulin framework or the human common framework as defined below. A human receptor framework "derived from" the human immunoglobulin framework or the human common framework may contain the same amino acid sequence, or may contain amino acid sequence modifications. In some embodiments, the number of amino acid modifications is 10 or less, 9 or less, 8 or less, 7 or less, 6 or less, 5 or less, 4 or less, 3 or less, or 2 or less. In some embodiments, the sequence of the VL human receptor framework is identical to the sequence of the VL human immunoglobulin framework or the human common framework.

As used herein, an “affinity-matured” antibody is an antibody that has one or more modifications in one or more variable regions (including CDR and FR) compared to an unmodified parent antibody, wherein such modifications result in an increased affinity of the antibody for the antigen.

As used herein, the terms “Fc region,” “Fc domain,” “Fc chain,” or similar terms are used to define the C-terminal region of the IgG heavy chain. The Fc region of IgG contains two constant domains, CH2 and CH3. The CH2 domain of the human IgG Fc region typically extends from amino acid 231 to amino acid 340 according to the EU index numbering system, or from amino acid 244 to amino acid 360 according to the Kabat numbering system. The CH3 domain of the human IgG Fc region typically extends from amino acid 341 to amino acid 447 according to the EU index numbering system, or from amino acid 361 to amino acid 478 according to the Kabat numbering system. The CH2 domain of the human IgG Fc region (also known as the “Cγ2” domain) is unique because it does not pair closely with the other domain. Instead, two N-linked branched carbohydrate chains are inserted between the two CH2 domains of intact native IgG. The Fc region may contain native or variant Fc sequences. Although the boundaries of the Fc sequence of the immunoglobulin heavy chain may vary, the human IgG heavy chain Fc sequence is generally defined as an extension from an amino acid residue at approximately position Cys226 or approximately position Pro230 to the carboxyl terminus of said Fc sequence. Unless otherwise specified herein, the amino acid residues in the Fc region or constant region are numbered according to the EU numbering system, also known as the EU index, as described in Kabat et al., Sequences of Proteins of Immunological Interest, 5th edition, Public Health Service, National Institutes of Health, Bethesda, MD, 1991.

In some embodiments, the Fc chain begins in a hinge region immediately upstream of the papain cleavage site and terminates at the C-terminus of the antibody. Therefore, a complete Fc chain comprises at least a hinge domain, a CH2 domain, and a CH3 domain. In some embodiments, the Fc chain comprises at least one of the following: a hinge (e.g., upper, middle, and/or lower hinge region) domain, a CH2 domain, a CH3 domain, a CH4 domain, or a variant, portion, or fragment thereof. In some embodiments, the Fc domain comprises a complete Fc chain (i.e., a hinge domain, a CH2 domain, and a CH3 domain). In some embodiments, the Fc chain comprises a hinge domain (or a portion thereof) fused to a CH3 domain (or a portion thereof). In some embodiments, the Fc chain comprises a CH2 domain (or a portion thereof) fused to a CH3 domain (or a portion thereof). In some embodiments, the Fc chain consists of a CH3 domain or a portion thereof. In some embodiments, the Fc chain consists of a hinge domain (or a portion thereof) and a CH3 domain (or a portion thereof). In some embodiments, the Fc chain consists of a CH2 domain (or a portion thereof) and a CH3 domain. In some embodiments, the Fc chain consists of a hinge domain (or a portion thereof) and a CH2 domain (or a portion thereof). In some embodiments, the Fc chain lacks at least a portion of the CH2 domain (e.g., all or a portion of the CH2 domain). As used herein, Fc chain generally refers to a polypeptide comprising all or a portion of the Fc chain of an immunoglobulin heavy chain. This Fc chain includes, but is not limited to, polypeptides comprising the entire CHI, hinge, CH2, and/or CH3 domains, and fragments of such peptides comprising only, for example, the hinge, CH2, and CH3 domains. Fc chains can be derived from immunoglobulins of any species and/or any subtype, including but not limited to human IgG1, IgG2, IgG3, IgG4, IgD, IgA, IgE, or IgM antibodies. The Fc domain encompasses native Fc and Fc variant molecules. As with Fc variants and native Fc, the term Fc chain includes molecules in monomeric or polymeric form, whether derived from whole antibody digestion or otherwise. In some embodiments, the Fc chain comprises the carboxyl-terminal portions of two heavy chains held together by disulfide bonds. In some embodiments, the Fc chain consists of a CH2 domain and a CH3 domain.

As used in the art, “Fc receptor” and “FcR” describe receptors that bind to the Fc region of an antibody. Preferred FcRs are naturally occurring human FcRs. Furthermore, preferred FcRs are those that bind IgG antibodies (γ receptors) and include receptors of the FcγRI, FcγRII, and FcγRIII subclasses, including allelic variants and alternative splicing forms of these receptors. FcγRII receptors include FcγRIIA (“activating receptor”) and FcγRIIB (“inhibitory receptor”), which are distinguished primarily by similar amino acid sequences in their cytoplasmic domains. FcRs are reviewed in: Ravetch and Kinet, Ann. Rev. Immunol., 9:457-92, 1991; Capel et al., Immunomethods, 4:25-34, 1994; and de Haas et al., J. Lab. Clin. Med., 126:330-41, 1995. "FcR" also includes the neonatal receptor FcRn, which is responsible for the transfer of maternal IgG to the fetus (Guyer et al., J. Immunol., 117:587, 1976; and Kim et al., J. Immunol., 24:249, 1994).

The “natural sequence Fc region” or “wild-type Fc region” contains the same amino acid sequence as the Fc region found in nature. “Wild-type” human IgG Fc refers to the sequence of amino acids naturally present in the human population. Of course, just as the Fc sequence can vary slightly between individuals, one or more modifications can be made to the wild-type sequence while still remaining within the scope of this invention. For example, the Fc region may contain additional modifications unrelated to this invention, such as mutations in glycosylation sites, including non-natural amino acids, or “knobs-in-holes” mutations.

The “variant Fc region” or “variant Fc chain” comprises an amino acid sequence that differs from the amino acid sequence of the natural Fc region by means of at least one amino acid modification, but retains at least one effector function of the natural Fc region. In some embodiments, the variant Fc chain has at least one amino acid substitution compared to the natural Fc chain or the Fc region of the parent polypeptide, for example, about 1 to about 10 amino acid substitutions in the natural Fc chain or the Fc chain of the parent polypeptide, and preferably about 1 to about 5 amino acid substitutions. The variant Fc chain described herein will preferably have at least about 80% sequence identity with the natural Fc chain and/or the Fc chain of the parent polypeptide, and most preferably at least about 90% sequence identity, more preferably at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99% sequence identity.

As used herein, the term "effective function" refers to those biological activities induced by the Fc region of an antibody (the native sequence Fc chain or the Fc chain of an amino acid sequence variant), and varies with antibody isotype. Examples of antibody effector functions include: C1q binding and complement-dependent cytotoxicity; Fc receptor binding; antibody-dependent cell-mediated cytotoxicity (ADCC); phagocytosis; downregulation of cell surface receptors (e.g., B cell receptors); and B cell activation. Such effector functions generally require the combination of the Fc chain with a binding domain (e.g., the antibody variable region) and can be evaluated using various assays known in the art for assessing such antibody effector functions. Exemplary measurements of effector functions are via Fcγ3 and/or C1q binding.

The effector function of an antibody is determined by a sequence in the Fc chain; this chain is also the part recognized by the Fc receptor (FcR) found on certain types of cells.

In some embodiments, the Fc polypeptide contains part or all of the wild-type hinge sequence (typically at its N-terminus). In some embodiments, the Fc polypeptide does not contain a functional or wild-type hinge sequence.

As used herein, “hinge region,” “hinge sequence,” and their variations include meanings known in the art, as illustrated in, for example, Janeway et al., ImmunoBiology: the immune system in health and disease, Elsevier Science Ltd., NY (4th edition, 1999); Bloom et al., Protein Science, 6:407-415, 1997; and Humphreys et al., J. Immunol. Methods, 209:193-202, 1997.

As used herein, "immunoglobulin-like hinge region," "immunoglobulin-like hinge sequence," and variations thereof refer to the hinge region and hinge sequence of an immunoglobulin-like or antibody-like molecule (e.g., an immunoadhesin). In some embodiments, the immunoglobulin-like hinge region may be derived from or originate from any IgG1, IgG2, IgG3, or IgG4 subtype, or IgA, IgE, IgD, or IgM, including its chimeric forms, such as a chimeric IgG1/2 hinge region.

In some embodiments, the hinge region may be derived from a human IgG1 subtype extending from amino acid 216 to amino acid 230 according to the EU indexing system or from amino acid 226 to amino acid 243 according to the Kabat numbering system. In some embodiments, the sequence may be EPKSCDKTHTCPPCP (SEQ ID NO: 63). Those skilled in the art may have different understandings of the exact amino acids corresponding to the various domains of the IgG molecule. Therefore, the N-terminus or C-terminus of the domains outlined above may be extended or shortened by 1, 2, 3, 4, 5, 6, 7, 8, 9, or even 10 amino acids.

In some embodiments, the hinge region may be mutated by one or more amino acids. In some embodiments, the hinge region may be truncated and contain only a portion of the full hinge region. In some embodiments, the hinge region may contain only the last 5 amino acids of the hinge region, referred to herein as the "lower hinge" region. In some embodiments, the lower hinge region may contain amino acid CPPCP (SEQ ID NO: 64) or amino acids 226 to 230 according to the EU index numbering system or amino acids 239 to 243 according to the Kabat numbering system.

As used herein, the term "modification" refers to the substitution, insertion, and/or deletion of amino acids in a polypeptide sequence, alteration of the portion chemically linked to the protein, or modification of the function of a protein (e.g., an antibody). For example, a modification can be an alteration of antibody function or a change in the structure of a carbohydrate linked to the protein. As used herein, "amino acid modification" refers to a mutation (substitution), insertion (addition), or deletion of one or more amino acid residues in an antibody. The term "amino acid mutation" means that at least one existing amino acid residue is replaced by another different amino acid residue (e.g., a substitute amino acid residue). The term "amino acid deletion" means the removal of at least one amino acid residue at a predetermined position in the amino acid sequence. For example, mutation L234A indicates that the amino acid residue lysine at position 234 in the Fc region of the antibody is replaced by the amino acid residue alanine (replacing lysine with alanine) (numbered according to the EU index numbering system). "Natural amino acid residues" refers to amino acid residues from the following groups: alanine (three-letter code: Ala, single-letter code: A), arginine (Arg, R), asparagine (Asn, N), aspartic acid (Asp, D), cysteine (Cys, C), glutamine (Gin, Q), glutamic acid (Glu, E), glycine (Gly, G), histidine (His, H), isoleucine (Ile, I), leucine (Leu, L), lysine (Lys, K), methionine (Met, M), phenylalanine (Phe, F), proline (Pro, P), serine (Ser, S), threonine (Thr, T), tryptophan (Trp, W), tyrosine (Tyr, Y), and valine (Val, V).

The term "pharmaceutical" is used herein to mean a biological macromolecule, an extract derived from biological material, a mixture of biological macromolecules, a chemical compound, a mixture of chemical compounds, and/or a mixture of chemical compounds and biological macromolecules. The term "therapeutic agent" refers to a biologically active pharmaceutical agent.

As used herein, a “monoclonal antibody” refers to an antibody obtained from a substantially homogeneous population of antibodies, i.e., the individual antibodies constituting the population are identical except for the possibility of naturally occurring mutations present in small amounts. Monoclonal antibodies are highly specific against a single antigenic site. Furthermore, in contrast to polyclonal antibody formulations, which typically comprise different antibodies targeting different determinants (epitopes), each monoclonal antibody targets a single determinant on the antigen. The modifier “monoclonal” indicates the characteristics of antibodies obtained from a substantially homogeneous population of antibodies and should not be construed as requiring antibody production by any particular method. For example, a monoclonal antibody to be used according to the invention can be prepared by a hybridoma method first described by Kohler and Milstein, Nature 256:495, 1975, or by a recombinant DNA method such as that described in U.S. Patent No. 4,816,567. Monoclonal antibodies can also be isolated from phage libraries generated using, for example, techniques described in McCafferty et al., Nature 348:552-554, 1990.

The antibodies of this invention can be “humanized antibodies.” As used herein, a “humanized” antibody refers to a form of non-human (e.g., mouse, rat, rabbit, non-human primate, or other mammalian) antibody that is a chimeric immunoglobulin, immunoglobulin chain, or fragment thereof (such as Fv, Fab, Fab', F(ab') ₂ , or other antigen-binding sequences of the antibody) containing one or more amino acid residues introduced therefrom from a non-human source. These non-human amino acid residues are generally referred to as “input” residues and are typically derived from an “input” variable domain. The input residues, sequence, or antibody has desired affinity and/or specificity or other desired antibody biological activities as discussed herein.

Preferably, the humanized antibody is a human immunoglobulin (receptor antibody) in which residues from the complementarity-determining region (CDR) of the receptor are replaced with residues from a non-human species (donor antibody), such as mouse, rat, or rabbit, that have the desired specificity, affinity, and ability. In some cases, residues from the Fv framework region (FR) of the human immunoglobulin are replaced with corresponding non-human residues. Furthermore, the humanized antibody may contain residues not found in either the receptor antibody or the input CDR or framework sequence, but which are included to further refine and optimize antibody performance. Typically, the humanized antibody will contain at least one, and usually substantially all, of the two variable regions, wherein all or substantially all of the CDR regions correspond to those of non-human immunoglobulins, and all or substantially all of the FR regions are those of human immunoglobulin common sequences. The humanized antibody will also preferably contain at least a portion of the immunoglobulin constant region or structural domain (Fc), typically of the type of human immunoglobulin. Antibodies having Fc chains modified as described in WO 99/58572 are preferred. Other forms of humanized antibodies have one or more CDRs (CDR L1, CDR L2, CDR L3, CDR H1, CDR H2, or CDR H3) that are altered relative to the original antibody, and are also referred to as "derived from" one or more CDRs from the original antibody. As used herein, humanization intention includes deimmunization antibodies.

As used herein, "human antibody" means an antibody having an amino acid sequence corresponding to the amino acid sequence of an antibody produced by a human, and/or having been prepared using any technique known to those skilled in the art or disclosed herein for the preparation of human antibodies. Therefore, the term "human antibody" as used herein is intended to include antibodies having variable and constant regions derived from human immunoglobulin sequences. Human antibodies of the present invention may include amino acid residues not encoded by human immunoglobulin sequences (e.g., mutations introduced by in vitro random mutagenesis or site-specific mutagenesis or by in vivo somatic mutations), such as in the CDR, and specifically CDR3. This definition of human antibody includes antibodies comprising at least one human heavy chain polypeptide or at least one human light chain polypeptide. One such example is an antibody comprising a mouse light chain and a human heavy chain polypeptide. Various techniques known in the art can be used to produce human antibodies. In one embodiment, the human antibody is selected from a phage library expressing a human antibody (Vaughan et al., Nature Biotechnology, 14:309-314, 1996; Sheets et al., Proc. Natl. Acad. Sci. (USA) 95:6157-6162, 1998; Hoogenboom and Winter, J. Mol. Biol., 227:381, 1991; Marks et al., J. Mol. Biol., 222:581, 1991). The human antibody can also be prepared by immunizing an animal in which a human immunoglobulin locus has been transgenically introduced to replace an endogenous locus, such as in mice where the endogenous immunoglobulin gene has been partially or completely inactivated. This method is described in U.S. Patent Nos. 5,545,807; 5,545,806; 5,569,825; 5,625,126; 5,633,425; and 5,661,016. Alternatively, human antibodies can be prepared by immortalizing human B lymphocytes that produce antibodies against a target antigen (such B lymphocytes can be recovered from an individual or a single-cell clone of cDNA, or can have been immunized in vitro). See, for example, Cole et al. Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, p. 77, 1985; Boerner et al., J. Immunol., 147(1):86-95, 1991; and U.S. Patent No. 5,750,373.

The human antibodies of the present invention can exist in at least two forms related to hinge heterogeneity. For example, immunoglobulin molecules comprise a stable four-chain construct of approximately 150-160 kDa, wherein the dimers are held together by interchain heavy chain disulfide bonds. Alternatively, the dimers are not linked by interchain disulfide bonds and form a molecule of approximately 75-80 kDa (half-antibody) consisting of covalently coupled light and heavy chains.

As used herein, the term "recombinant human antibody" is intended to include all human antibodies prepared, expressed, generated, or isolated in a recombinant manner, such as antibodies expressed using a recombinant expression vector transfected into host cells (described further below), antibodies isolated from a library of recombinant human antibodies (described further below), antibodies isolated from animals transgenic with human immunoglobulin genes (e.g., mice) (see, for example, Taylor et al. (1992) Nucl. Acids Res. 20:6287-6295), or antibodies prepared, expressed, generated, or isolated by any other means involving splicing a human immunoglobulin gene sequence into another DNA sequence. Such recombinant human antibodies have variable and constant regions derived from human germline immunoglobulin sequences. However, in some implementations, such recombinant human antibodies are subjected to in vitro mutagenesis (or, when using animals transgenic with human Ig sequences, in vivo somatic cell mutagenesis), and thus the amino acid sequences of the VH and VL regions of the recombinant antibody are sequences that, although derived from and associated with human germline VH and VL sequences, may not naturally exist in the human antibody germline library.

The antibodies of this invention can be “chimeric antibodies” (where a portion of the heavy chain and/or light chain is identical or homologous to the corresponding sequence in an antibody derived from a specific species or belonging to a specific antibody class or subclass, while the remaining portion of the chain is identical or homologous to the corresponding sequence in an antibody derived from another species or belonging to another antibody class or subclass), and fragments of such antibodies, provided they exhibit the desired biological activity (US Patent 4,816,567; and Morrison et al., Proc. Natl. Acad. Sci. USA 81:6851-6855, 1984). The target chimeric antibodies herein include primate-derived antibodies comprising a variable region antigen-binding sequence derived from a non-human primate (e.g., Old World monkeys, apes, etc.) and a human constant region sequence.

A "monovalent antibody" contains one antigen-binding site/molecule (e.g., IgG or Fab). In some cases, a monovalent antibody may have more than one antigen-binding site, but the binding sites come from different antigens.

A "monospecific antibody" is an antibody or antibody preparation that contains two identical antigen-binding sites/molecules (e.g., IgG), such that the two binding sites bind to the same epitopes on the antigen. Therefore, they compete with each other when binding to a single antigen molecule. This term includes "monoclonal antibody" or "monoclonal antibody composition." Most antibodies found in nature are monospecific. In some cases, monospecific antibodies can also be monovalent antibodies (e.g., Fab).

The terms “mutation load,” “mutational load,” “mutation burden,” or “mutational burden” are used interchangeably in this article. Tumor mutational load is a measure of the number of mutations within a tumor genome, defined as the total number of mutations in each coding region of the tumor genome. Mutational load varies greatly across tumor types, ranging from just a few to several thousand mutations (Alexandrov LB et al., Nature 2013;500(7463):415-421; Lawrence MS et al., Nature 2013;499:214–218; Vogelstein B et al., Science, 2013;339:1546–1558).

"Original amino acid residue" refers to a residue replaced by an "input amino acid" residue whose side chain volume may be smaller or larger than the original residue. The input amino acid residue may be naturally occurring or non-naturally occurring amino acid residues, but is preferably the former. "Naturally occurring" amino acid residues are those residues encoded by the genetic code. "Non-naturally occurring" amino acid residues refer to residues that are not encoded by the genetic code but can covalently bind to neighboring amino acid residues in the polypeptide chain. Examples of non-naturally occurring amino acid residues are leucine, ornithine, valine, homoserine, and other amino acid residue analogs, such as those described, for example, in Ellman et al., Meth. Enzym. 202:301-336 (1991). In some embodiments, the method of the present invention involves replacing at least one original amino acid residue, but may replace more than one original residue. Typically, the total residues in no more than the first or second polypeptide interface may contain the replaced original amino acid residue.

As used herein with respect to the use of the term "competitive," it means that a first antibody or its antigen-binding fragment (or portion) binds to an epitope in a manner sufficiently similar to the binding of a second antibody or its antigen-binding portion, such that the binding of the first antibody to its homologous epitope is detectably reduced in the presence of the second antibody compared to the binding of the first antibody in the absence of the second antibody. Alternatives to this, where the binding of the second antibody to its epitope is also detectably reduced in the presence of the first antibody, are possible but not necessary. That is, the first antibody may inhibit the binding of the second antibody to its epitope, while the second antibody may not inhibit the binding of the first antibody to its corresponding epitope. However, when each antibody detectably inhibits the binding of another antibody to its homologous epitope or ligand, whether to the same, greater, or lesser extent, the antibodies are said to "cross-compete" with each other for binding to their respective epitopes. Both competitive antibodies and cross-competitive antibodies are covered by this invention. Regardless of the mechanism by which such competition or cross-competition occurs (e.g., steric hindrance, conformational change, or binding to a common epitope or a portion thereof), based on the teachings provided herein, those skilled in the art will understand that such competitive antibodies and/or cross-competitive antibodies are covered and can be used in the methods disclosed herein.

As used herein, “antibody-dependent cell-mediated cytotoxicity” or “ADCC” refers to a cell-mediated reaction in which nonspecific cytotoxic cells expressing Fc receptors (FcRs), such as natural killer (NK) cells, neutrophils, and macrophages, recognize binding antibodies on target cells and subsequently cause lysis of the target cells. The ADCC activity of a target molecule can be evaluated using in vitro ADCC assays, such as those described in U.S. Patent Nos. 5,500,362 or 5,821,337. Useful effector cells for such assays include peripheral blood mononuclear cells (PBMCs) and NK cells. Alternatively or additionally, the ADCC activity of a target molecule can be evaluated in vivo, for example, in animal models, such as those disclosed in Clynes et al., PNAS (USA), 95:652-656, 1998.

As used herein, “complement-dependent cytotoxicity” or “CDC” refers to target lysis in the presence of complement. The complement activation pathway is initiated by the binding of the first component (C1q) of the complement system to molecules (e.g., antibodies) that are complexed with homologous antigens. To evaluate complement activation, a CDC assay can be performed, for example, as described in Gazzano-Santoro et al., J. Immunol. Methods, 202: 163, 1996.

As used herein, the terms “immunospecific binding,” “immunospecific recognition,” “specific binding,” “specific recognition,” and similar terms refer to a molecule that specifically binds to an antigen (e.g., an epitope or immune complex) but does not specifically bind to another molecule (e.g., a binding domain). A molecule that specifically binds to an antigen may bind to other peptides or polypeptides with lower affinity, as determined by assays known in the art, such as immunoassays, BIACORE™, or other assays. Preferably, the molecule that specifically binds to the antigen does not cross-react with other proteins.

Suitable “moderately stringent conditions” include prewashing in a solution of 5X SSC, 0.5% SDS, and 1.0 mM EDTA (pH 8.0); hybridization overnight at 50–65°C with 5X SSC; followed by two washes at 65°C with 2X, 0.5X, and 0.2X SSC containing 0.1% SDS, each for 20 minutes.

As used herein, “highly stringent conditions” or “highly stringent conditions” are conditions in which: (1) low ionic strength and high temperature are used for washing, for example, 0.015 M sodium chloride/0.0015 M sodium citrate/0.1% sodium dodecyl sulfate at 50°C; (2) denaturing agents are used during hybridization at 42°C, such as formamide, for example, 50% (v/v) formamide with 0.1% bovine serum albumin/0.1% Ficoll/0.1% polyvinylpyrrolidone/50 mM sodium phosphate buffer with 750 mM sodium chloride and 75 mM sodium citrate at pH 6.5; Alternatively (3) wash at 42°C with 50% formamide, 5 x SSC (0.75 M NaCl, 0.075 M sodium citrate), 50 mM sodium phosphate (pH 6.8), 0.1% sodium pyrophosphate, 5 x Denhardt solution, sonicated salmon sperm DNA (50 µg/ml), 0.1% SDS, and 10% dextran sulfate, followed by washing at 42°C in 0.2 x SSC (sodium chloride/sodium citrate) and at 55°C in 50% formamide, followed by a high-rigidity wash at 55°C with 0.1 x SSC containing EDTA. Technicians will recognize how to adjust the temperature, ionic strength, etc., as needed to accommodate factors such as probe length.

The binding protein, binding domain, or CDR or antibody (as broadly defined herein) may be identified according to the accumulation, AbM, contact, North definition and/or conformational definition of Kabat, Chothia, Kabat and Chothia, or any method of CDR determination well known in the art. See, for example, Kabat et al., 1991, Sequences of Proteins of Immunological Interest, 5th edition (Hypervariate Region); Chothia et al., Nature 342:877-883, 1989 (Structural Ring Structure). The identity of the amino acid residues constituting the CDR in a particular antibody may be determined using methods well known in the art. The AbM definition of the CDR is a compromise between Kabat and Chothia and uses Oxford Molecular's AbM antibody modeling software (Accelrys®). The definition of “contact” for the CDR is based on observed antigen contact and is described in MacCallum et al., J. Mol. Biol., 262:732-745, 1996. The “conformation” definition of a CDR is based on residues that contribute enthalpy to antigen binding (see, for example, Makabe et al., J. Biol. Chem., 283:1156-1166, 2008). North has identified canonical CDR conformations using different preferred sets of CDR definitions (North et al., J. Mol. Biol. 406: 228-256, 2011). In another approach referred to herein as the “conformation definition” of a CDR, the position of the CDR can be identified as residues that contribute enthalpy to antigen binding (Makabe et al., J. Biol. Chem., 283:1156-1166, 2008). Other CDR boundary definitions may not strictly follow one of the above approaches but will still overlap with at least a portion of the Kabat CDR, although they may be shortened or lengthened based on predictions or experimental findings that specific residues or groups of residues or even the entire CDR do not significantly affect antigen binding. As used herein, a CDR can refer to a CDR defined by any method known in the art, including combinations of methods. The methods used herein can utilize a CDR defined according to any of these methods. For any given embodiment containing more than one CDR, a CDR (or other residues of the antibody) can be defined according to any of the following definitions: Kabat, Chothia, extension (a combination of Kabat and Chothia), North, extension, AbM, contact, and/or conformation.

Residues in the variable domain are numbered according to Kabat, a numbering system used for the heavy chain or light chain variable regions of the assembled antibody. See Kabat et al., Sequences of Proteins of Immunological Interest, 5th edition, Public Health Service, National Institutes of Health, Bethesda, MD, 1991. Using this numbering system, the actual linear amino acid sequence may contain fewer or additional shortened or inserted amino acids corresponding to the FR or CDR of the variable region. For example, the heavy chain variable region may include a single amino acid insertion after residue 52 of H2 (according to Kabat residue 52a) and inserted residues after heavy chain FR residue 82 (e.g., according to Kabat residues 82a, 82b, and 82c). The Kabat number of residues for a given antibody can be determined by comparing the homologous regions of the antibody sequence to a “standard” Kabat numbered sequence. Various algorithms for assigning Kabat numbers are available. This article uses the algorithm implemented in Abysis release 2.3.3 (www.abysis.org) to assign Kabat numbers to the variable regions VLCDR1, VL CDR2, VL CDR3, VHCDR1, VHCDR2, and VH CDR3. Specific amino acid residue positions within the antibody can also be numbered based on Kabat numbers.

As used herein, the term "phage display library" refers to a population of phages, each containing exogenous cDNA fused to a surface protein within a recombination frame. The phages display the exogenous protein encoded by the cDNA on their surface. After replication in a bacterial host (typically *E. coli*), phages containing the target exogenous cDNA are selected based on the expression of the exogenous protein on the phage surface.

The term "epitope" refers to a portion of a molecule that can be recognized, contacted, and/or bound by an antibody at one or more sites in an antigen-binding region called a complementary site. A single antigen may have more than one epitope. Thus, different antibodies can bind to different regions on the antigen and may have different biological effects. Epitopes are often composed of chemically active surface groups of a molecule (such as amino acid or sugar side chains) and have specific three-dimensional structural features and specific charge features. As used herein, epitopes can be conformational or linear. Conformational epitopes are generated by spatially juxtaposing amino acids from different segments of a linear polypeptide chain. Linear epitopes are epitopes generated from neighboring amino acid residues in a polypeptide chain. In some embodiments, an epitope may include a sugar, phosphoryl, or sulfonyl portion of the antigen.

As used herein, an "antigenic epitope" is defined as a portion of a polypeptide to which an antibody can specifically bind, as determined by any method well known in the art (e.g., by a routine immunoassay). A "nonlinear epitope" or "conformational epitope" comprises a non-neighboring polypeptide (or amino acid) within an antigenic protein to which an antibody specific for that epitope binds. Once a desired epitope on an antigen is identified, an antibody targeting that epitope can be generated, for example, using the techniques described herein.

When used to describe the interaction between antibodies and proteins or peptides, the terms "specific binding" or "specifically binding" refer to an interaction that depends on the presence of a specific structure on the protein (i.e., an antigenic determinant or epitope); in other words, the antibody recognizes and binds to a specific protein structure rather than the usual protein. For example, if an antibody is specific to epitope "A," the presence of a protein containing epitope A (or free unlabeled A) in a reaction containing labeled "A" and an antibody will reduce the amount of labeled A that binds to the antibody.

In some embodiments, "specific binding" means, for example, that an antibody binds to a protein with a KD of about 0.1 nM or less, but more typically less than about 1 μM. In some embodiments, "specific binding" means that an antibody sometimes binds to a target with a KD of at least about 0.1 μM or less, at other times with at least about 0.01 μM or less, and at other times with at least about 1 nM or less. Due to sequence identity between homologous proteins in different species, specific binding may include antibodies that recognize proteins in more than one species (e.g., human or mouse proteins). Similarly, due to homology within certain regions of the polypeptide sequences of different proteins, specific binding may include antibodies that recognize more than one protein. It should be understood that in some embodiments, an antibody or binding portion that specifically binds to a first target may or may not specifically bind to a second target. Therefore, "specific binding" does not necessarily require (although it may include) exclusive binding, i.e., binding to a single target. Therefore, in some embodiments, an antibody may specifically bind to more than one target. In some embodiments, the same antigen-binding site on an antibody may bind to multiple targets. For example, in some cases, an antibody may contain two identical antigen-binding sites, each specifically binding to the same epitope on two or more proteins. In some alternative embodiments, the antibody may be multispecific and contain at least two antigen-binding sites with different specificities. By way of non-limiting example, a bispecific antibody may contain an antigen-binding site that recognizes an epitope on one protein (e.g., human CD3) and further contain a second, different antigen-binding site that recognizes a different epitope on a second protein. Generally, but not necessarily, references to binding refer to specific binding.

Antibodies that specifically bind to antigens can bind to other peptides or polypeptides with lower affinity, as determined by assays known in the art (e.g., immunoassays, BIACORE®, or other assays). Preferably, antibodies that specifically bind to antigens do not cross-react with other proteins.

When used in relation to the interaction between antibodies and proteins or peptides, the terms “non-specific binding” or “background binding” refer to interactions that do not depend on the presence of a specific structure (i.e., antibody binding is typically protein rather than a specific structure, such as an epitope).

As used herein, the terms "k _on " or " _ka " refer to the rate constant relating to the association of antibody with antigen. Specifically, the rate constants (k _on / _ka and k _off /k _d ) and equilibrium dissociation constants are measured using intact antibodies (i.e., bivalent) and monomeric CD3 proteins.

As used in this article, the terms "k _off " or "k _d " refer to the rate constant of antibody dissociation from the antibody/antigen complex.

As used in this article, the term " _KD " refers to the equilibrium dissociation constant of antibody-antigen interactions.

As used herein, the term "binding affinity" generally refers to the total strength of the non-covalent interaction between a single binding site of a molecule (e.g., an antibody) and its binding partner (e.g., an antigen). Unless otherwise indicated, as used herein, "binding affinity" refers to the intrinsic binding affinity that reflects a 1:1 interaction between members of a binding pair (e.g., antibody and antigen). The affinity of molecule X for its partner Y is generally represented by the dissociation constant (KD). The affinity of molecule X for its partner Y is generally represented by the dissociation constant (Kd). For example, Kd can be about 200 nM, 150 nM, 100 nM, 60 nM, 50 nM, 40 nM, 30 nM, 20 nM, 10 nM, 8 nM, 6 nM, 4 nM, 2 nM, 1 nM, or stronger. Affinity can be measured by methods commonly known in the art, including those described herein. Low-affinity antibodies typically bind to antigens slowly and tend to dissociate easily, while high-affinity antibodies typically bind to antigens quickly and tend to remain bound for a longer period. Various methods for measuring binding affinity are known in the art, any of which may be used for the purposes of this invention. Specifically, the term "binding affinity" is intended to refer to the rate of dissociation of a particular antigen-antibody interaction. _KD is the ratio of the rate of dissociation (also called the "off-rate" (k _-off )) to the rate of association (k-on) (k- _on ). Therefore, _KD equals k _-off / k _-on and is expressed as a molar concentration (M). Thus, the smaller the _KD , the stronger the binding affinity. Therefore, a _KD of 1 μM indicates weak binding affinity compared to _{a KD} of 1 nM. The _KD value of an antibody can be determined using methods well established in the art. One method for determining the _KD of an antibody is by using surface plasmon resonance (SPR), typically employing a biosensor system (such as the BIACORE™ system). BIACORE™ kinetic analysis involves analyzing the binding and dissociation of antigens from chips with immobilized molecules (e.g., molecules containing epitope-binding domains) on their surfaces. Another method for determining the _KD of antibodies is through biolayer interferometry, typically using ^OCTET® technology (Octet QKe system, ForteBio).

Unless otherwise specified, with respect to the antibodies (such as antibodies, fragments or derivatives thereof) of the present invention, "bioactive", "bioactive" and "biocharacteristic" mean having the ability to bind biomolecules.

As used herein, the terms "nucleic acid" and "nucleotide sequence" include DNA molecules (e.g., cDNA or genomic DNA), RNA molecules (e.g., mRNA), combinations or hybrid DNA/RNA molecules, and analogs of DNA or RNA molecules. Such analogs can be generated using, for example, nucleotide analogs, including but not limited to inosine or triphenylmethylated bases. Such analogs may also comprise DNA or RNA molecules containing a modified backbone that imparts beneficial properties to the molecule, such as, for example, nuclease resistance or increased transmembrane translucency. Nucleic acid or nucleotide sequences can be single-stranded, double-stranded, contain both single-stranded and double-stranded portions, and may contain triple-stranded portions, but are preferably double-stranded DNA.

The present invention also includes a polynucleotide encoding the antibody of the present invention (including the polypeptide and binding region of the antibody). The polynucleotide encoding the molecule of the present invention can be obtained, and the nucleotide sequence of the polynucleotide can be determined by any method known in the art.

The polynucleotide encoding the antibody of the present invention may include: a variant-only coding sequence; a variant-coding sequence and additional coding sequences (such as functional polypeptides or signal sequences or secretion sequences or preprotein sequences); and antibody-coding and non-coding sequences (such as introns or non-coding sequences at the 5' and/or 3' of the antibody-coding sequence). The term "polynucleotide encoding antibody" encompasses polynucleotides including additional coding sequences of variants, and also encompasses polynucleotides including additional coding and/or non-coding sequences. It is known in the art that polynucleotide sequences optimized for specific host cells/expression systems can be readily obtained from the amino acid sequence of the desired protein (see ^GENEART® AG, Regensburg, Germany).

The antibodies of the present invention may have additional conserved or non-essential amino acid substitutions that do not substantially affect the function of the polypeptide. Whether a particular substitution will be tolerated (i.e. will not adversely affect the desired biological properties (such as binding activity)) can be determined as described in Bowie, JU et al., Science 247: 1306-1310, 1990 or Padlan et al., FASEBJ. 9: 133-139, 1995.

As used herein, "conservative amino acid substitution" is a substitution in which an amino acid residue is replaced by an amino acid residue having a similar side chain. Families of amino acid residues having similar side chains are defined in the art. These families include amino acids having the following side chains: basic side chains (e.g., lysine, arginine, histidine), acidic side chains (e.g., aspartic acid, glutamic acid), uncharged polar side chains (e.g., asparagine, glutamine, serine, threonine, tyrosine, cysteine), nonpolar side chains (e.g., glycine, alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine, tryptophan), β-branched side chains (e.g., threonine, valine, isoleucine), and aromatic side chains (e.g., tyrosine, phenylalanine, tryptophan, histidine).

As used herein, "non-essential" amino acid residues are residues that may differ from the wild-type sequence of a binding agent (e.g., an antibody) and do not eliminate or substantially alter biological activity, while "essential" amino acid residues cause this change. In antibodies, essential amino acid residues can be specificity-determining residues (SDRs).

As used herein, the term "isolated" refers to material removed from its original environment (e.g., the natural environment if it is naturally occurring). For example, naturally occurring polynucleotides or polypeptides present in living animals are not isolated, but the same polynucleotides or polypeptides isolated from some or all of the coexisting material in a natural system are isolated. This polynucleotide may be part of a carrier, and/or this polynucleotide or polypeptide may be part of a composition (e.g., a mixture, solution, or suspension) or contain isolated cells or cultured cells containing the polynucleotide or polypeptide, and is still isolated because the carrier or composition is not part of its natural environment. For example, a polypeptide synthesized chemically or synthesized in a cellular system different from the cell of natural origin of the polypeptide will be "isolated" from its natural associated components. Protein purification techniques well known in the art can also be used to isolate proteins to make them substantially free of natural associated components.

"Isolated" or "purified" antibodies are substantially free of cellular material or other contaminating proteins from cell or tissue sources or culture media from which the protein originates, or are substantially free of chemical precursors or other chemicals during chemical synthesis. The phrase "substantially free of cellular material" includes antibody formulations in which the polypeptide/protein is isolated from cellular components of the cells from which the polypeptide/protein is produced, either isolated or recombinantly. Therefore, antibodies substantially free of cellular material include antibody formulations having less than about 50%, 40%, 30%, 20%, 10%, 5%, 2.5%, or 1% (by dry weight) of contaminating proteins. When antibodies are produced by recombinant synthesis, they are also preferably substantially free of culture media, i.e., the culture medium constitutes less than about 50%, 40%, 30%, 20%, 10%, 5%, 2.5%, or 1% of the volume of the protein formulation. When antibodies are produced by chemical synthesis, they are preferably substantially free of chemical precursors or other chemicals and reagents, i.e., the target antibody is isolated from chemical precursors or other chemicals involved in protein synthesis. Therefore, this antibody preparation contains less than about 50%, 40%, 30%, 20%, 10%, 5%, or 1% (by dry weight) of a chemical precursor or compound in addition to the target antibody. In a preferred embodiment of the invention, the antibody is isolated or purified. For the purposes of this invention, an antibody that has been isolated or removed from at least one component of an organism or from tissues or cells in which the antibody is naturally present or produced is called an "isolated antibody".

As used herein, the term “recycling” refers to, for example, a process that uses protein purification techniques well known in the art to separate chemical substances (such as peptides) to make them substantially free of their natural associated components.

As used herein, the term "replicon" refers to any genetic element, such as a plasmid, chromosome, or virus, that acts as an autonomous polynucleotide replication unit within a cell.

As used herein, the term "operably ligated" refers to a situation where the relationship between said elements allows them to function in the intended manner. For example, "operably ligated" to a control sequence of a coding sequence in a manner that enables expression of the coding sequence under suitable or compatible conditions with that control sequence. Generally, "operably ligated" means that the ligated DNA sequences are adjacent, and in the case of a secretory leader sequence, adjacent and in the reading phase. However, enhancers do not require adjacency. Ligation is accomplished by ligating at a convenient restriction site. If such a site is not present, synthetic oligonucleotide adaptors or linkers are used according to conventional practice.

As used herein, “vector” means a construct capable of being delivered in a host cell and preferably expressing one or more target genes or sequences. Examples of vectors include, but are not limited to, viral vectors, naked DNA or RNA expression vectors, plasmids, granules or phage vectors, DNA or RNA expression vectors associated with cationic condensers, DNA or RNA expression vectors encapsulated in liposomes, and certain eukaryotic cells, such as production cells.

As used herein, the term "expression control sequence" or "control sequence" refers to a polynucleotide sequence necessary for the expression of the coding sequence it is linked to. The nature of this control sequence varies depending on the host organism. For example, in prokaryotes, such control sequences typically include a promoter, a ribosome binding site, a terminator, and (in some cases) an enhancer. Therefore, the term "control sequence" is intended to include at least all elements necessary for expression, but may also include additional elements that are advantageous for its presence, such as a leader sequence.

"Host cell" includes an individual cell or cell culture that may be or has been a vector recipient, the vector being used to incorporate a polynucleotide insert. Host cells include progeny of a single host cell, and the progeny may not be identical to the original parent cell (morphologically or in terms of genomic DNA complementation) due to natural, accidental, or intentional mutations. Host cells include cells transfected in vivo with the polynucleotides of the present invention.

As used herein, "mammalian cell" includes references to cells derived from mammals, including humans, rats, mice, guinea pigs, chimpanzees, or macaques. These cells may be cultured in vivo or in vitro.

As used herein, the term "purified product" refers to a product formulation that has been isolated from the cellular components to which the product is normally associated and/or from other cell types that may be present in the target sample.

As used herein, “substantially pure” means a material that is at least 50% pure (i.e., free of contaminants), more preferably at least 90% pure, more preferably at least 95% pure, even more preferably at least 98% pure, and most preferably at least 99% pure.

As used herein, the terms “treat,” “treating,” or “treatment” are methods for obtaining beneficial or desired clinical outcomes. For the purposes of this invention, treatment is defined as administering a CD3 antibody molecule (e.g., a monoclonal antibody, bispecific antibody) to a subject (e.g., a patient), or administering (e.g., by application) an isolated tissue or cell from the subject and returning it to said subject. The CD3 antibody molecule may be administered alone or in combination with one or more agents. Treatment can be curative, healing, alleviating, resolving, altering, remedying, improving, reducing, improving, or affecting a condition, its symptoms, or a predisposition to the condition (e.g., cancer).

As used herein, the term “subject” is intended to include any animal (e.g., a mammal), including but not limited to humans, non-human primates, rodents, etc., that is a recipient of a particular treatment. For example, a subject may be a patient with cancer (e.g., a human patient or a veterinary patient). Generally, the terms “subject,” “individual,” and “patient” are used interchangeably in this document when referring to human subjects.

Unless otherwise stated, the term "non-human animal" in this invention includes all non-human vertebrates, such as non-human mammals and non-mammals, such as non-human primates, sheep, dogs, cows, chickens, amphibians, reptiles, mice, rats, rabbits, or goats, etc.

As used herein, the term “pharmaceutical acceptable” means a product or compound that has been approved (or permitted to be approved) by a federal or state government regulatory agency or is listed in the United States Pharmacopeia or other recognized pharmacopoeia for use in animals (including humans).

As used herein, the terms "pharmaceutically acceptable excipient, carrier, or adjuvant" or "acceptable pharmaceutical carrier" mean an excipient, carrier, or adjuvant that can be administered to a subject together with at least one antibody of this disclosure without destroying the activity of the antibody. When administered together with an antibody at a dose sufficient to deliver a therapeutic effect, the excipient, carrier, or adjuvant shall be non-toxic.

As used herein, the term “improvement” means a reduction or improvement of one or more symptoms compared to the absence of the antibody molecules of the present invention. “Improvement” also includes a reduction or decrease in the duration of symptoms.

As used herein, the terms “prevent,” “preventing,” and “prevention” refer to the prevention of recurrence or onset of one or more symptoms of a disease in a subject by means of the administration of a prophylactic or therapeutic agent.

As used herein, “effective amount,” “therapeutic effective amount,” “therapeutic adequate amount,” or “effective dose” means the amount of a therapeutic agent (e.g., a CD3 antibody, alone or as part of a multispecific antibody (e.g., bispecific)) administered to a subject in a single or multiple doses that is effective or sufficient to prevent, heal, improve, treat, or manage a disease, condition, or side effect, or to reduce the rate of progression of a disease or condition, or to prolong healing, alleviate, reduce, or improve the condition of a subject with a condition as described herein. The term also includes, within its scope, amounts that effectively enhance normal physiological function. An effective amount can be considered in the context of administration of one or more therapeutic agents, and a single agent may be considered to have been administered in an effective amount if, in combination with one or more other agents, the desired outcome can be achieved.

As used herein, effective agents can target effector T cells to induce T cell-mediated cytotoxicity in cells associated with diseases such as cancer, autoimmune diseases, or infectious diseases. The therapeutic agents of the present invention induce a T cell-mediated immune response by modulating the CD3 antigen, specifically CD3, and more specifically CD3ε, on the surface of T cells.

Regarding cancer, the therapeutically effective dose refers to the amount of a therapeutic agent that inhibits the growth of a tumor or cancer by slowing, interrupting, stopping, or terminating its growth and/or metastasis.

Efficacy is a measure of the activity of a therapeutic agent, expressed as the amount required to produce an effect of a given strength. A highly potent agent elicits a greater response at low concentrations compared to a less potent agent that elicits a smaller response at low concentrations. Efficacy varies with affinity and potency. Potency refers to the ability of a therapeutic agent to produce a biological response upon binding to a target ligand and the quantitative magnitude of that response. As used herein, the term “half-maximum effective concentration ( _EC50 )” refers to the concentration of a therapeutic agent that elicits a response midway between baseline and maximum after a specified exposure time. Therapeutic agents can cause inhibition or stimulation. The _EC50 value is commonly used and is referred to herein as a measure of efficacy.

As used herein, “combination therapy” or “in combination with” refers to the use of more than one prophylactic and/or therapeutic agent. The use of the terms “combination therapy” or “combination” does not limit the order in which prophylactic and/or therapeutic agents are administered to a subject with the condition. In other words, combination therapy can be administered by treating with the therapeutic agents separately, sequentially, or simultaneously. In the case of “sequential administration,” the first administered agent may exert some physiological effect on the subject upon administration of the second agent or upon becoming active in the subject.

As used herein with respect to the administration of prophylactic and/or therapeutic agents, the term "simultaneous administration" means administration of the agents such that the individual agents are simultaneously present in the subject. Simultaneous administration can be achieved by molecules in a single composition or in separate compositions administered at the same or similar times. Sequential administration can be performed in any order as needed.

The term “differentiation cluster 3” or “CD3” refers to a polyprotein complex (historically known as the T3 complex) composed of four distinct polypeptide chains: epsilon (ε), gamma (γ), delta (δ), and zeta (ζ), which assemble and function as three pairs of dimers (εγ, εδ, ζζ). The CD3 complex acts as a T-cell co-receptor, non-covalently associating with the T-cell receptor (TCR) (Smith-Garvin et al. 2009) and generating activation signals in T lymphocytes. The CD3 protein complex is a defining characteristic of the T-cell lineage, and therefore CD3 antibodies can be effectively used as T-cell markers (Chetty and Gatter 1994). It is well known that CD3 antibodies induce the generation of cytotoxic T cells by activating endogenous lymphotropic mediators and can selectively kill tumor targets (Yun et al., Cancer Research, 49: 4770-4774 (1989)).

T cells play a central role in cell-mediated immunity in humans and animals. The recognition and binding of specific antigens are mediated by TCRs expressed on the surface of T cells. T cell TCRs can interact with immunogenic peptides (epitopes) that bind major histocompatibility complex (MHC) molecules and are presented on the surface of target cells. Specific binding of the TCR triggers a signaling cascade within the T cell, leading to proliferation and differentiation into mature effector T cells. More specifically, T cells express TCR complexes capable of inducing antigen-specific immune responses (Smith-Garvin et al., 2009). Antigens are peptides expressed by tumor cells and virus-infected cells capable of stimulating immune responses. Intracellularly expressed antigens bind to class I major histocompatibility (MHC) molecules and are transported to the surface, where they are exposed to T cells. If the binding affinity of the TCR to the class I MHC complexed with the antigen is sufficient, the formation of an immune synapse is initiated. Signaling via the immune synapse is mediated by CD3 co-receptors that form ε/δ, δ/γ, and ζ/ζ dimers. These dimers are associated with the TCR and generate activation signals in T lymphocytes. This signaling cascade guides T cell-mediated killing of antigen-expressing cells. Cytotoxicity is mediated by the release of granzyme B and perforin from T cells and their translocation to target cells.

As used herein, “activated T-cell antigen” refers to an antigenic determinant expressed on the surface of T lymphocytes, specifically cytotoxic T lymphocytes, which can induce T-cell activation upon interaction with antigen-binding molecules. Specifically, the interaction between antigen-binding molecules and activated T-cell antigen can induce T-cell activation by triggering a signaling cascade of the T-cell receptor complex. In one specific embodiment, the activated T-cell antigen is CD3, specifically the ε subunit of CD3 (for human sequences, see UniProt number P07766 (version 130), NCBI RefSeq number NP_000724.1, SEQ ID NO: 66). Typically, naturally occurring allele variants have an amino acid sequence that is at least 95%, 97%, or 99% identical to the protein described in Genbank accession number BAB71849.1.

As used herein, "T cell activation" refers to one or more cellular responses selected from the following: proliferation, differentiation, cytokine secretion, release of cytotoxic effector molecules, cytotoxic activity, and expression of activation markers by T lymphocytes, specifically cytotoxic T lymphocytes. The T cell activation bispecific antigen-binding molecule of the present invention can induce T cell activation. Assays suitable for measuring T cell activation are known in the art described herein.

As used herein, “antibody that binds CD3,” “antibody that recognizes CD3,” “anti-CD3 antibody,” or “CD3 antibody” includes an antibody that specifically recognizes a single CD3 subunit (e.g., ε, δ, γ, or ζ) and its antigen-binding fragment, and an antibody that specifically recognizes a dimer complex of two CD3 subunits (e.g., γ/ε, δ/ε, and ζ/ζ CD3 dimers) and its antigen-binding fragment. Human CD3ε is indicated in Genbank accession number NM_000733. The antibodies of the present invention can bind soluble CD3 and/or CD3 expressed on the cell surface. Soluble CD3 includes native CD3 protein and recombinant CD3 protein variants (such as, for example, monomeric and dimer CD3 constructs) lacking a transmembrane domain or otherwise not associated with the cell membrane.

Antibodies targeting CD3 can generate activation signals in T lymphocytes. Other T cell activation ligands, including but not limited to CD28, CD134, CD137, and CD27, can also be used.

This invention includes a single-arm antibody that binds to CD3. As used herein, "single-arm antibody" means an antigen-binding molecule comprising a single antibody heavy chain and a single antibody light chain. The single-arm antibody of this invention may comprise any of the VL/VH or CDR amino acid sequences as described in Tables 1, 2 and 3 herein.

The CD3 antibodies disclosed herein may contain one or more amino acid substitutions, insertions, and/or deletions in the framework and/or CDR regions of the heavy and light chain variable domains, compared to the corresponding germline sequences from which the antibodies are derived. Such mutations can be readily identified by comparing the amino acid sequences disclosed herein with germline sequences available, for example, from public antibody sequence databases. This invention includes antibodies derived from any of the amino acid sequences disclosed herein and their antigen-binding fragments, wherein one or more amino acids in one or more framework and/or CDR regions are mutated to corresponding residues of the germline sequence from which the antibody is derived, or mutated to corresponding residues of another human germline sequence, or mutated to conserved amino acid substitutions of the corresponding germline residues (such sequence changes are collectively referred to herein as “germline mutations”). Those skilled in the art can readily generate a variety of antibody and antigen-binding fragments containing one or more individual germline mutations or combinations thereof, starting with the heavy and light chain variable region sequences disclosed herein. In some embodiments, all framework and/or CDR residues within the VH and/or VL domains are mutated back to residues found in the original germline sequence from which the antibody is derived. In some embodiments, only certain residues are mutated back to the original germline sequence, such as mutated residues found only in the first 8 amino acids of FR1 or the last 8 amino acids of FR4, or mutated residues found only in CDR1, CDR2, or CDR3. In some further embodiments, one or more of the framework and/or CDR residues are mutated to corresponding residues in a different germline sequence (i.e., a germline sequence different from the germline sequence from which the antibody originally originated).

Furthermore, the antibodies of the present invention may contain any combination of two or more germline mutations within the framework and/or CDR region, for example, wherein certain individual residues are mutated to corresponding residues of a specific germline sequence, while certain other residues different from the original germline sequence are maintained or mutated to corresponding residues of a different germline sequence. Once obtained, one or more desired properties of antibodies containing one or more germline mutations can be readily tested, such as improved binding specificity, increased binding affinity, improved or enhanced antagonistic or agonistic biological properties (as applicable), reduced immunogenicity, etc. Antibodies obtained in this general manner are covered within the present invention.

The present invention also includes CD3 antibodies comprising variants of any of the disclosed VH, VL, and/or CDR amino acid sequences having one or more conserved substitutions. For example, the present invention includes CD3 antibodies having VH, VL, and/or CDR amino acid sequences having, for example, 10 or fewer, 8 or fewer, 6 or fewer, 4 or fewer, or other conserved amino acid substitutions relative to any of the VH, VL, and/or CDR amino acid sequences described in Table 1 herein.

As used herein, the term "complex" or "composite" refers to the association of two or more molecules that interact with each other via bonds and/or forces that are not peptide bonds (e.g., van der Waals forces, hydrophobic forces, hydrophilic forces). In one embodiment, the complex is heteropolymeric. It should be understood that the terms "protein complex" or "peptide complex" as used herein include complexes having non-protein entities (e.g., including but not limited to chemical molecules, such as toxins or detection agents) conjugated to proteins in the protein complex.

As used herein, "CD3 antibody" includes monovalent antibodies with single specificity, as well as "bispecific antibody," "two-specific antibody," "three-specific antibody," "bifunctional antibody," "heteropolymer," "heteropolymer complex," "bispecific heterodimeric biantibody," "heteropolypeptide," or bispecific heterodimeric IgG. In some embodiments of the present invention, the CD3 antibody of the present invention is a human or humanized antibody.

As used herein, a "bispecific antibody" is a molecule comprising at least a first polypeptide and a second polypeptide, wherein the second polypeptide differs from the first polypeptide in its amino acid sequence by at least one amino acid residue. In some cases, a bispecific antibody is an artificial hybrid antibody having two distinct heavy chain regions and light chain regions. Preferably, a bispecific antibody has binding specificity to at least two different ligands, antigens, or binding sites. Therefore, a bispecific antibody can bind to two different antigens simultaneously. The two antigen-binding sites of a bispecific antibody can reside on two different epitopes on the same or different protein targets (e.g., tumor targets).

As used herein, the first and second polypeptide chains of a “bispecific antibody” contain at least one antibody VL and an antibody VH region or a fragment thereof, wherein both antibody-binding domains are contained within a single polypeptide chain and wherein the VL and VH regions in each polypeptide chain are derived from different antibodies.

Bispecific antibodies can be derived from full-length antibodies or antibody fragments (e.g., F(ab')2 bispecific antibodies). Bispecific antibodies are more fully described, for example, in EP404,097; WO93/11161; and Hollinger et al., Proc. Natl. Acad. Sci. USA 90:6444-6448, 1993. A bispecific antibody is a heterodimer of two “crossed” sFv fragments, in which the VH and VL regions of the two antibodies are located on different polypeptide chains.

As used herein, the terms “link,” “fusion,” “fusion product,” “covalent bonding,” “covalent coupling,” and “gene fusion” are used interchangeably. These terms refer to the linking of two or more elements or components together by any means, including chemical conjugation or recombination. As used herein, the term “covalent bonding” means that the specified parts are directly covalently bonded to each other or indirectly covalently bonded to each other via one or more insert parts (such as linking peptides or parts). In a preferred embodiment, the parts are covalently fused. One type of covalent bonding is the peptide bond. Methods of chemical conjugation (e.g., using heterobifunctional crosslinking agents) are known in the art. The fusion part can also be a gene fusion. As used herein, the terms “gene fusion,” “gene link,” or “gene fusion” refer to the expression of a gene via a single polynucleotide molecule encoding two or more proteins, polypeptides, or fragments, which are collinearly, covalently linked, or connected via their individual peptide backbones. This genetic fusion results in the expression of a single adjacent gene sequence. Preferred gene fusions are in-frame fusions, meaning two or more open reading frames (ORFs) fuse in a manner that maintains the correct reading frame of the original ORF to form a continuously longer ORF. Therefore, the resulting recombinant fusion protein is a single polypeptide containing two or more protein segments corresponding to the polypeptides encoded by the original ORFs (segments that would not normally bind in this way naturally). In this case, the single polypeptide is cleaved during treatment to produce a dimer molecule containing two polypeptide chains.

As used herein, the terms "linked" or "links" refer to a direct peptide bond between first and second amino acid sequences, or a bond involving a third amino acid sequence peptide bonded to and between the first and second amino acid sequences. For example, a linker peptide is bonded to the C-terminus of one amino acid sequence and the N-terminus of another amino acid sequence.

As used herein, the term "linker" refers to an amino acid sequence of two or more amino acids in length. Linkers can consist of neutral polar or nonpolar amino acids. The length of a linker can be, for example, from 2 to 100 amino acids, such as between 2 and 50 amino acids, such as 3, 5, 10, 15, 20, 25, 30, 35, 40, 45, or 50 amino acids. Linkers can be "cleavable," for example, by self-cleavage, enzymatic cleavage, or chemical cleavage. Cleavage sites in amino acid sequences, and enzymes and chemicals that cleave at such sites, are well known in the art and are also described herein.

As used herein, the term "disulfide bond" or "cysteine-cysteine disulfide bond" refers to a covalent interaction between two cysteine residues, wherein the sulfur atom of the cysteine is oxidized to form a disulfide bond. The average bond energy of a disulfide bond is approximately 60 kcal/mol, compared to 1-2 kcal/mol for hydrogen bonds. In the context of this invention, the cysteine residues forming the disulfide bond are located within the framework region of a single-chain antibody and serve to stabilize the antibody's conformation. Cysteine residues can be introduced, for example, through site-directed mutagenesis, to generate stable disulfide bonds intramolecularly.

The term "the name of the pestle in the mortar" is similar to the term "protrusion and cavity" and they can be used interchangeably.

A "protrusion" or "pole" refers to at least one of the following amino acid side chains: extending from the interface of the first polypeptide and thus able to be positioned in a compensating cavity in a neighboring interface (i.e., the interface of the second polypeptide) to stabilize the heterodimer, thereby, for example, making heterodimer formation superior to homodimer formation. The protrusion may be present in the original interface or may be introduced synthetically (e.g., by altering the nucleic acid encoding the interface). Typically, the nucleic acid encoding the interface of the first polypeptide is modified to encode the protrusion. To achieve this, the nucleic acid encoding at least one "original" amino acid residue in the interface of the first polypeptide is replaced with an "input" amino acid residue encoding at least one side chain with a volume larger than the original amino acid residue. The upper limit for the number of original residues replaced is the total number of residues in the interface of the first polypeptide. Certain input residues used to form the protrusion are typically naturally occurring amino acid residues and are preferably selected from arginine (R), phenylalanine (F), tyrosine (Y), and tryptophan (W).

The protrusions or cavities are "localized" within cavities or sockets, meaning that the spatial location and size of the protrusions and cavities at the interface between the first and second peptides, respectively, allow the protrusions to be located within the cavities without significantly disrupting the normal association of the first and second peptides at the interface. Since protrusions, such as phenylalanine (F), tyrosine (Y), and tryptophan (W), typically do not extend perpendicularly from the interface axis, the comparison of protrusions to their corresponding cavities relies on modeling the protrusion/cavity pairs based on three-dimensional structures, such as those obtained by X-ray crystallography or nuclear magnetic resonance (NMR). This can be achieved using techniques well-established in the art.

A “cavity” or “mortise” refers to at least one amino acid side chain that recesses from the interface of the second polypeptide and thus accommodates a corresponding protrusion on the adjacent interface of the first polypeptide. The cavity may be present in the original interface or may be introduced synthetically (e.g., by altering the nucleic acid encoding the interface). Typically, the nucleic acid encoding the interface of the second polypeptide is altered to encode the cavity. To achieve this, the nucleic acid encoding at least one “original” amino acid residue in the interface of the second polypeptide is replaced with DNA encoding at least one “input” amino acid residue that has a side chain volume smaller than the original amino acid residue. The upper limit for the number of original residues replaced is the total number of residues in the interface of the second polypeptide. Some of the input residues used to form the cavity are typically naturally occurring amino acid residues and are preferably selected from alanine (A), serine (S), threonine (T), and valine (V).

As used herein, “target antigen,” “target cell antigen,” “tumor antigen,” or “tumor-specific antigen” refers to antigenic determinants that are presented on the surface of target cells (e.g., cells in a tumor, such as cancer cells) or cells in the tumor stroma).

As used herein, the term “cancer” or “cancerous” refers to or describes a physiological condition in mammals characterized by uncontrolled cell growth, growths or tumors of abnormal, uncontrolled growth originating from cells. In some respects, cancer refers to a malignant primary tumor that has not metastasized and remains localized. In other respects, cancer refers to a malignant tumor that has invaded and destroyed adjacent body structures and spread to distant sites. In some respects, cancer is associated with a specific cancer antigen. Examples of cancer include, but are not limited to, cancers of the following sites: mouth and pharynx, digestive system (e.g., esophagus, stomach, small intestine, colon, rectum, liver, gallbladder, bile ducts, and pancreas), respiratory system (e.g., larynx, lungs, and bronchi, including non-small cell lung cancer), bone and joints (e.g., bone metastases), soft tissue, skin (e.g., melanoma), breast, reproductive system (e.g., cervix, uterus, ovary, vulva, vagina, prostate, and testes), urinary system (e.g., bladder, kidneys, renal pelvis, and ureters), eye and orbit, brain and nervous system (e.g., glioma), head and neck, or endocrine system (e.g., thyroid). Cancer can also be lymphoma (such as Hodgkin's disease and non-Hodgkin's lymphoma), multiple myeloma, or leukemia (such as acute lymphoblastic leukemia, chronic lymphoblastic leukemia, acute myeloid leukemia, chronic myeloid leukemia, etc.). In some implementations, cancers of the digestive system are cancers of the esophagus, stomach, small intestine, colon, rectum, liver, gallbladder, bile ducts, and pancreas.

Regarding cancer, treatment is useful in any one or more of the following: (a) treating, preventing, or improving one or more symptoms of a condition associated with malignant cells in a subject (e.g., gastrointestinal cancers such as colorectal cancer (CRC)); (b) inhibiting tumor growth or progression in a subject with malignant tumors expressing tumor antigens; (c) inhibiting metastasis of cancer (malignant) cells expressing tumor antigens in a subject with one or more malignant cells expressing tumor antigens; (d) inducing regression (e.g., long-term regression) of tumors expressing tumor antigens; (e) exerting cytotoxic activity in malignant cells expressing tumor antigens; (f) increasing progression-free survival in a subject with a tumor antigen-associated condition; (g) increasing overall survival in a subject with a tumor antigen-associated condition; (h) reducing the use of additional chemotherapy agents or cytotoxic agents in a subject with a tumor antigen-associated condition; (i) reducing tumor burden in a subject with a tumor antigen-associated condition; or (j) blocking the interaction of tumor antigens with other unidentified factors. While not wishing to be bound by theory, the treatment is believed to have the ability to inhibit, ablate, or kill cells or otherwise reduce cell-mediated conditions (such as cancer) in vitro or in vivo.

As used in this article, the terms “malignant cell” or “malignant disease” refer to an aggressive and/or metastatic tumor or tumor cell, i.e., cancer cells.

Although any materials and methods similar to or equivalent to those described herein may be used in the practice or testing of this invention, preferred materials and methods are now described.

Materials and methods

Various techniques for antibody production have been described, including conventional hybridoma methods for preparing monoclonal antibodies, recombinant techniques for preparing antibodies (including chimeric antibodies, such as humanized antibodies), antibody production in transgenic animals, and the recently described phage display technique for preparing "fully human" antibodies. These techniques are briefly described below.

Polyclonal antibodies against a target antigen are typically generated in animals through multiple subcutaneous (sc) or intraperitoneal (ip) injections of the antigen and adjuvant. Using bifunctional reagents or derivatizing agents, such as maleimide benzoyl sulfosuccinimide (conjugated via cysteine residues) or N-hydroxysuccinimide (conjugated via lysine residues), the antigen (or a fragment containing the target amino acid sequence) can be conjugated to a protein (which is immunogenic in the species to be immunized, such as serum albumin, bovine thyroglobulin, or soybean trypsin inhibitors). Animals are immunized against the immunogenic conjugate or derivative, and booster immunizations are performed several weeks later via subcutaneous injections at multiple sites. Seven to 14 days later, blood is drawn from the animals, and serum antibody titers are measured. Booster immunizations are continued until a titer plateau is reached. Preferably, booster immunizations are performed using conjugates of the same antigen (but conjugated to different proteins and/or via different cross-linking agents). Conjugates can also be prepared as protein fusions in recombinant cell cultures. In addition, agglomerating agents (such as alum) are used to enhance the immune response.

Monoclonal antibodies can be obtained from a substantially homogeneous population of antibodies using the hybridoma method first described by Kohler and Milstein, Nature 256:495, 1975, or can be prepared via a recombinant DNA method (Cabilly et al., U.S. Patent No. 4,816,567). In the hybridoma method, immunization of mice or other suitable host animals, as described above, induces the production or ability to produce lymphocytes that will specifically bind to proteins used for immunization. Alternatively, immunization of lymphocytes can be performed in vitro. The lymphocytes are then fused with myeloma cells using a suitable fusion agent (such as polyethylene glycol) to form hybridoma cells (Goding, Monoclonal Antibodies: Principles and Practice, pp. 59-103 (Academic Press, 1986). The hybridoma cells thus prepared are seeded in a suitable culture medium, preferably containing one or more substances that inhibit the growth or survival of unfused parental myeloma cells, and allowed to grow therein. Alternatively, the hybridoma cells can grow in vivo as ascites tumors in animals. The monoclonal antibodies secreted by the subclones are appropriately separated from the culture medium, ascites, or serum using routine immunoglobulin purification procedures such as, for example, protein A-Sepharose, hydroxyapatite chromatography, gel electrophoresis, dialysis, or affinity chromatography.

The expression of antibodies, antigen-binding fragments of antibodies, or any antibody construct can be carried out in appropriate prokaryotic or eukaryotic host cells (such as CHO cells, NSO cells, SP2/0 cells, HEK293 cells, COS cells, PER.C6 cells, yeast or E. coli cells), and the secreted antibodies can be recovered and harvested from said cells (cell culture supernatant, regulated cell culture supernatant, cell lysate, or clarified bulk material). Common methods for generating antibodies are well known in the art and described in, for example, the following review articles: Makrides, S.C., Protein Expr. Purif. 17:183-202, 1999; Geisse, S., et al., Protein Expr. Purif. 8:271-282, 1986; Kaufman, R.J., MoI. Biotechnol. 16:151-160, 2000; Werner, R.G., Drug Res. 48:870-880, 1998. In one specific embodiment, the cell culture is a mammalian cell culture, such as a Chinese hamster ovary (CHO) cell culture.

In one embodiment, the antibody may be recovered or isolated from a naturally occurring source. In some such embodiments, the isolated or recovered antibody may undergo additional purification steps using conventional chromatographic methods known in the art. Specifically, the purification methods are contemplated, but are not limited to, affinity chromatography (e.g., protein A affinity chromatography), ion exchange chromatography (e.g., anion exchange chromatography or cation exchange chromatography), hydrophobic interaction chromatography, hydroxyapatite chromatography, gel filtration chromatography, and/or dialysis. Of these, protein A chromatography is the preferred purification method. The matrix for affinity ligand linkage is most often agarose, but other matrices may be used.

Other techniques for antibody purification are also known in the art, such as fractionation on ion exchange columns, ethanol precipitation, reversed-phase high-performance chromatography, reversed-phase HPLC, chromatography on silica, chromatography on heparin Sepharose™, chromatography on anion or cation exchange resins (such as polyaspartic acid columns), chromatographic focusing, electrophoresis using SDS-PAGE, and ammonium sulfate precipitation. The list of purification methods above is merely illustrative in nature and is not intended to be limiting.

Alternatively, transgenic animals (e.g., mice) can be produced that, upon immunization, generate a complete human antibody library in the absence of endogenous immunoglobulin production. For example, homozygous deletion of the antibody heavy chain linker (J.sub.H) gene in chimeric and germline mutant mice has been described as resulting in complete suppression of endogenous antibody production. Transplantation of human germline immunoglobulin gene arrays into such germline mutant mice will lead to the production of human antibodies upon antigen challenge. (See, for example, Jakobovits et al., Proc. Natl. Acad. Sci. USA 90:2551-255;1993 and Jakobovits et al., Nature 362:255-258, 1993).

In some embodiments, the antibodies of the present invention may be humanized, retaining high affinity for the antigen and other beneficial biological properties. Methods for humanizing nonhuman antibodies are well known in the art. Humanization can be essentially performed following the approach of Winter and collaborators (Jones et al., Nature 321:522-525, 1986; Riechmann et al., Nature 332:323-327, 1988; Verhoeyen et al., Science 239:1534-1536, 1988) by replacing the corresponding sequence of a human antibody with a rodent CDR or CDR sequence. Thus, such humanized antibodies are chimeric antibodies (Cabilly, ibid.), in which essentially less than one fully human variable domain has been replaced by a corresponding sequence from a non-human species. In practice, humanized antibodies are typically human antibodies, with some CDR residues and possibly some FR residues replaced by residues from similar sites in rodent antibodies. Importantly, the antibody is humanized, retaining its high affinity for the antigen and other beneficial biological properties. To achieve this, humanized antibodies are prepared according to a preferred method by analyzing the parental sequence and various conceptual humanized products using a three-dimensional model of the parental and humanized sequences. Three-dimensional immunoglobulin models are familiar to those skilled in the art. A computer program can be used to illustrate and demonstrate the possible three-dimensional conformational structures of selected candidate immunoglobulin sequences. Observing these displays allows analysis of the possible roles of residues in the function of the candidate immunoglobulin sequence, i.e., analysis of residues affecting the ability of the candidate immunoglobulin to bind its antigen. In this way, FR residues can be selected and merged from common and input sequences to achieve desired antibody properties, such as increased affinity for the target antigen. For further details, see WO92/22653, published December 23, 1992.

Various phage display methods known in the art can also be used to generate the antibodies of the present invention. In phage display methods, functional antibody domains are displayed on the surface of phage particles carrying a polynucleotide sequence encoding them. In particular, this phage can be used to display antigen-binding domains (such as Fab and Fv or disulfide-stable Fv) expressed from a library or combinatorial antibody library (e.g., human or mouse). Phages expressing antigen-binding domains that bind to target antigens can be selected or identified using antigens (e.g., using labeled antigens or antigens that bind to or capture onto solid surfaces or beads). The phages used in these methods are typically filamentous phages, including fd and M13. The antigen-binding domains are expressed as proteins recombinantly fused to phage gene III or gene VIII proteins. Examples of phage display methods that can be used to prepare the immunoglobulins or fragments thereof of the present invention include those disclosed in Brinkmann et al., "Phage Display Of Disulfide-Stabilized FvFragments," J. Immunol. Methods, 182:41-50, 1995.

The functional properties of various IgG isoforms and their domains are well known in the art. The amino acid sequences of IgG1, IgG2, IgG3, and IgG4 are known in the art. The selection and/or combination of two or more domains from a particular IgG isoform in the methods used in this invention can be based on any known parameters of the parental isoform (including affinity for FcγR). For example, using regions or domains from IgG isoforms (e.g., IgG2 or IgG4) that exhibit limited or no binding to FcγRIIB can be particularly useful in cases where it is desired to engineer bispecific antibodies to maximize binding to activating receptors and minimize binding to inhibitory receptors. Similarly, using Fc chains or domains from IgG isoforms (e.g., IgG3) that are known to preferentially bind to C1q or FcγRIIIA can be combined with Fc amino acid modifications known in the art to enhance antibody-dependent cell-mediated cytotoxicity (ADCC) and/or complement-dependent cytotoxicity (CDC) to engineer bispecific antibodies such that effector functional activity (e.g., complement activation or ADCC) is maximized. In a similar manner, mutations can be made in the Fc chain or domain of the IgG isotype to minimize or eliminate the effector function of the Fc chain.

During the discovery process, antibody generation and characterization can elucidate information about the desired epitope. Based on this information, antibodies can be competitively screened for binding to the same epitope. This is achieved by conducting competitive and cross-competitive studies to find antibodies that compete or cross-compete with each other (e.g., antibodies that competitively bind to antigens). One approach is to identify the epitope that the antibody binds to, or “epitope localization.” Many methods known in the art exist for mapping and characterizing the location of epitopes on proteins, including resolving the crystal structure of antibody-antigen complexes, competitive assays, gene fragment expression assays, and synthetic peptide-based assays, as described, for example, in Chapter 11 of Harlow and Lane, Using Antibodies, a Laboratory Manual, Cold SpringHarbor Laboratory Press, Cold SpringHarbor, N.Y., 1999. In an additional instance, epitope mapping can be used to determine the sequence to which the antibody binds. Epitope mapping is commercially available from various sources, such as Pepscan Systems (Edelhertweg 15, 8219 PH Lelystad, Netherlands). In an additional instance, mutagenesis of the antigen-binding domain, domain exchange experiments, and alanine partitioning mutagenesis can be performed to identify residues required, sufficient, and/or essential for epitope binding. Binding affinity and the dissociation rate of the antigen-binding domain interaction can be determined by a competitive binding assay. An example of a competitive binding assay is a radioimmunoassay, which involves incubating a labeled antigen and detecting the molecule binding to the labeled antigen. The affinity and binding-dissociation rate of the molecules of the present invention for the antigen can be determined from saturation data using Scatchard analysis.

The affinity and binding properties of the antibodies of the present invention to antigens can first be determined using in vitro assays (biochemical or immunoassays) known in the art for antigen-binding domains, including but not limited to enzyme-linked immunosorbent assays (ELISA), surface plasmon resonance (SPR) assays, biolayer interferometry, or immunoprecipitation assays. The molecules of the present invention may have similar binding properties in in vivo models (such as those described and disclosed herein) to those in in vitro-based assays. However, the present invention does not exclude molecules of the present invention that do not exhibit the desired phenotype in in vitro-based assays but do exhibit the desired phenotype in vivo.

Once the nucleic acid sequence encoding the molecule of the present invention (i.e., the binding domain) has been obtained, a vector for generating the molecule can be produced using recombinant DNA technology, a technique well known in the art.

The polynucleotide encoding the antibody-binding domain of the present invention may include an expression control polynucleotide sequence operatively linked to the antibody-coding sequence, including naturally associated or heterologous promoter regions known in the art. The expression control sequence may be a eukaryotic promoter system in a vector capable of transforming or transfecting eukaryotic host cells, but a prokaryotic host control sequence may also be used. Once the vector has been incorporated into a suitable host cell line, the host cells are allowed to proliferate under conditions suitable for expressing the nucleotide sequence and are used, as needed, to collect and purify the antibody. Eukaryotic cell lines include CHO cell lines, various COS cell lines, HeLa cells, myeloma cell lines, transformed B cells, or human embryonic kidney cell lines.

In some embodiments, the DNA encoding the antibodies of the present invention is isolated and sequenced using conventional procedures, e.g., by using oligonucleotide probes capable of specifically binding to genes encoding the heavy and light chains of mouse antibodies. Hybridoma cells of the present invention serve as a preferred source of this DNA. Once isolated, the DNA can be placed in an expression vector, which is then transfected into host cells that otherwise do not produce immunoglobulin proteins (such as simian COS cells, CHO cells, or myeloma cells) to obtain monoclonal antibody synthesis in recombinant host cells. The DNA can also be modified, for example, by replacing the homologous mouse sequences with coding sequences for the constant domains of the human heavy and light chains (Morrison et al., Proc. Nat. Acad. Sci. 81:6851, 1984). In this manner, chimeric antibodies having the binding specificity of the anti-antigen monoclonal antibodies described herein are prepared.

CD3 antibody

In one aspect, the present invention provides a reagent that specifically binds to CD3 (e.g., the full-length human CD3 ε subunit (e.g., accession number NP_000724 or SEQ ID NO: 66)).

In some implementations, the CD3 antibody specifically binds to human CD3.

Exemplary CD3 antibodies of the present invention are listed in Tables 1, 2, and 3 herein. Table 1 shows the amino acid sequence identifiers of the VH and VL regions. In some embodiments, antibodies are provided having any of the partial light chain sequences listed in Table 1 and/or any of the partial heavy chain sequences listed in Table 1.

Table 1

In Table 1, the underlined sequences are CDR sequences based on Kabat, and the bold sequences are CDR sequences based on Chothia.

This invention provides the CDR portion (including Chothia, Kabat CDR, and CDR contact region) of an antibody against CD3. Methods and techniques for identifying CDRs within VH and VL amino acid sequences are well known in the art and can be used to identify CDRs within the specified VH and/or VL amino acid sequences disclosed herein. Exemplary conventions for identifying CDR boundaries include, for example, the Kabat definition, the Chothia definition, and the AbM definition. Generally, the Kabat definition is based on sequence variability, the Chothia definition is based on the location of structural loop regions, and the AbM definition is a compromise between the Kabat and Chothia methods. See, for example, Kabat, "Sequences of Proteins of Immunological Interest," National Institutes of Health, Bethesda, Md. (1991); Al-Lazikani et al., J. Mol. Biol. 273:927-948 (1997); and Martin et al., Proc. Natl. Acad. Sci. USA 86:9268-9272 (1989). Public databases may also be used to identify CDR sequences within the antibody. It should be understood that in some embodiments, the CDR may be a combination of Kabat and Chothia CDRs (also referred to as a "combined CR" or "extended CDR"). In some embodiments, the CDR is the Kabat CDR. In some further embodiments, the CDR is the Chothia CDR. In other words, in an implementation having more than one CDR, the CDR can be any of Kabat, Chothia, combined CDRs, or a combination thereof. Tables 2 and 3 provide examples of the CDR sequences provided herein.

In one embodiment, the present invention provides an antibody comprising the amino acid sequences listed in Table 1 or substantially similar sequences having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with them.

In some such embodiments, the present invention provides an antibody comprising a VL region comprising an amino acid sequence of any of the VL region amino acid sequences listed in Table 1 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with it.

In some other embodiments of this invention, the present invention provides an antibody comprising a VH region comprising an amino acid sequence of any of the VH region amino acid sequences listed in Table 1 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with it.

In one embodiment, an antibody that specifically binds to CD3 is provided, wherein the antibody comprises: (a) a heavy chain variable (VH) region comprising VH complementarity-determining region 1 (VH CDR1), VH complementarity-determining region 2 (VH CDR2), and VH complementarity-determining region 3 (VH CDR3) of the VH sequence shown in SEQ ID NO: 2, 4, 5, 7, 10, 12, or 35; and/or (b) a light chain variable (VL) region comprising VL complementarity-determining region 1 (VL CDR1), VL complementarity-determining region 2 (VL CDR2), and VL complementarity-determining region 3 (VL CDR3) of the VL sequence shown in SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87, or 89.

In one embodiment, the present invention provides an antibody comprising a set of six CDRs (i.e., VL CDR1, VL CDR2, VLCDR3, VH CDR1, VH CDR2, VH CDR3), wherein the CDRs are contained within the VL/VH amino acid sequence pairs as defined by any of the exemplary CD3 antibodies listed in Table 1.

In specific embodiments, the present invention provides an antibody comprising a set of amino acid sequences of VL CDR1, VL CDR2, VL CDR3, VHCDR1, VHCDR2, or VH CDR3, wherein the set of amino acid sequences is contained within a VL/VH amino acid sequence pair selected from the following: SEQ ID NO: 1 and 2 (e.g., 2B5-0001 (2B5)); SEQ ID NO: 3 and 2 (2B5-0002); SEQ ID NO: 3 and 4 (e.g., 2B5-0006 (2B5v6)); SEQ ID NO: 3 and 5 (e.g., 2B5-0009); SEQ ID NO: 6 and 7 (e.g., 2B5-0517); SEQ ID NO: 8 and 7 (e.g., 2B5-0522); SE QID NO: 6 and 4 (e.g., 2B5-0533); SEQ ID NO: 8 and 4 (e.g., 2B5-0538); SEQ ID NO: 9 and 10 (e.g., 2B5-0598); SEQ ID NO: 9 and 7 (e.g., 2B5-0623); SEQ ID NO: 11 and 12 (e.g., 2B5-0707); SEQ ID NO: 34 and 35 (e.g., 2B5-0003); SEQ ID NO: 85 and 2 (e.g., H2B4); SEQ ID NO: 9 and 4 (e.g., 2B5-1038); SEQ ID NO: 87 and 4 (e.g., 2B5-1039); and SEQ ID NO: 89 and 4 (e.g., 2B5-1040).

Table 2 shows the VH CDR1, VH CDR2, and VH CDR3 regions of an exemplary CD3 antibody.

Table 2

In one embodiment, the present invention provides an antibody comprising VH CDR1, said VH CDR1 comprising an amino acid sequence of any of the VHCDR1 amino acid sequences listed in Table 2 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In another embodiment, the present invention provides an antibody comprising VH CDR2, said VH CDR2 comprising an amino acid sequence of any of the VHCDR2 amino acid sequences listed in Table 2 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In yet another embodiment, the present invention provides an antibody comprising VH CDR3, said VH CDR3 comprising an amino acid sequence of any of the VHCDR3 amino acid sequences listed in Table 2 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

Table 3 shows exemplary CD3 antibodies VL CDR1, VL CDR2, and VL CDR3.

Table 3

In one embodiment, the present invention provides an antibody comprising VL CDR1, said VL CDR1 comprising an amino acid sequence of any of the VLCDR1 amino acid sequences listed in Table 3 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In another embodiment, the present invention provides an antibody comprising VL CDR2, said VL CDR2 comprising an amino acid sequence of any of the VLCDR2 amino acid sequences listed in Table 3 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In yet another embodiment, the present invention provides an antibody comprising VL CDR3, said VL CDR3 comprising an amino acid sequence of any of the VLCDR3 amino acid sequences listed in Table 3 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In some embodiments, an antibody that specifically binds to CD3 is provided, wherein the antibody comprises: (a) a heavy chain (HC) variable (VH) region comprising VH complementarity-determining region 1 (CDR1), VH complementarity-determining region 2 (VH CDR2), and VH complementarity-determining region 3 (VH CDR3) of VH having an amino acid sequence selected from SEQ ID NO: 2, 4, 5, 7, 10, 12, and 35; and/or (b) a light chain (LC) variable (VL) region comprising VL complementarity-determining region 1 (VL CDR1), VL complementarity-determining region 2 (VL CDR2), and VL complementarity-determining region 3 (VLCDR3) of VL having an amino acid sequence selected from SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87, and 89.

In a specific embodiment, the antibody comprises: (a) a VH region comprising (i) a VH CDR1 comprising the sequence of SEQ ID NO: 13, 14 or 15; (ii) a VHCDR2 comprising the sequence of SEQ ID NO: 16, 17, 19, 20, 21, 22, 23 or 24; and (iii) a VH CDR3 comprising the sequence of SEQ ID NO: 18 or 25; and/or (b) a VL region comprising (i) a VL CDR1 comprising the sequence of SEQ ID NO: 26, 29, 30, 32, 91 or 92; (ii) a VL CDR2 comprising the sequence of SEQ ID NO: 27 or 31; and (iii) a VL CDR3 comprising the sequence of SEQ ID NO: 28 or 33.

In some embodiments, the present invention provides antibodies that bind to CD3 and compete with antibodies as described herein, including 2B5-0001 (2B5), 2B5-0002, 2B5-0006 (2B5v6), 2B5-0009, 2B5-0517, 2B5-0522, 2B5-0533, 2B5-0538, 2B5-0598, 2B5-0623, 2B5-0707, 2B5-0003, 2B5-1038, 2B5-1039, and 2B5-1040.

In some embodiments, the present invention also provides the CDR portion of a CD3 antibody based on a CDR contact region. A CDR contact region is an antibody region that imparts specificity to an antigen to the antibody. Typically, the CDR contact region comprises residue sites within the CDR and Vernier region, constrained to maintain a suitable loop structure for antibody binding to specific antigens. See, for example, Makabe et al., J. Biol. Chem., 283:1156-1166, 2007. Determination of the CDR contact region is entirely within the scope of the art.

Some of the CD3 antibodies described herein may be referred to by more than one name. For example, 2B5-0001 may be referred to as 2B5 or h2B5; and 2B5-0006 may be referred to as 2B5v6 or h2B5v6. Table 4 shows the forms of the antibodies and the CD3 antibody regions present in the antibodies disclosed herein.

Table 4

sequence	describe	form	CD3 antibody region
				2B5-0001 (2B5)	CD3 antibody	IgG	2B5-0001
2B5-0002	CD3 antibody	IgG	2B5-0002
				2B5-0003	CD3 antibody	IgG	2B5-0003
2B5-0006 (2B5v6)	CD3 antibody	IgG	2B5-0006
				2B5-0009	CD3 antibody	IgG	2B5-0009
2B5-0517	CD3 antibody	IgG	2B5-0517
				2B5-0522	CD3 antibody	IgG	2B5-0522
2B5-0533	CD3 antibody	IgG	2B5-0533
				2B5-0538	CD3 antibody	IgG	2B5-0538
2B5-0598	CD3 antibody	IgG	2B5-0598
				2B5-0623	CD3 antibody	IgG	2B5-0623
2B5-0707	CD3 antibody	IgG	2B5-0707
				2B5-1038	CD3 antibody	IgG	2B5-1038
2B5-1039	CD3 antibody	IgG	2B5-1039
				2B5-1040	CD3 antibody	IgG	2B5-1040
H2B4	CD3 antibody	IgG	H2B4
				GUCY2C-H2B4	CD3-GUCY2c bispecific antibody	CD3 bispecific	H2B4
GUCY2C-2B5	CD3-GUCY2c bispecific antibody	CD3 bispecific	2B5-0001
				GUCY2C-2B5v6	CD3-GUCY2c bispecific antibody	CD3 bispecific	2B5-0006
xFLT3-2B5v6	CD3-FLT3 bispecific antibody	CD3 bispecific	2B5-0006

The binding affinity ( _KD ) of the antibody described herein for CD3 (such as human CD3ε (e.g., SEQ ID NO:40) can be from about 0.001 nM to about 500 nM. In one embodiment, the binding affinity is about any of the following: 6500 nM, 6000 nM, 5500 nM, 4500 nM, 4000 nM, 3500 nM, 3000 nM, 2500 nM, 2000 nM, 1500 nM, 1000 nM, 750 nM, 500 nM, 400 nM, 300 nM, 250 nM, 200 nM, 150 nM, 100 nM, 75 nM, 50 nM, 45 nM, 40 nM, 35 nM, 30 nM, 25 nM, 20 nM, 19 nM, 17 nM, 16 nM, 15 nM, 10 nM, 8 nM, 7.5 nM, 7 nM, 6.5 nM, 6 nM, 5.5 nM, 5 nM, 4 nM, 3 nM, 2 nM, 1 nM, 0.5 nM, 0.3 nM, 0.1 nM, 0.01 nM, 0.002 nM, or 0.001 nM. In some embodiments, the binding affinity is at least about any one of the following: 6500 nM, 6000 nM, 5500 nM, 5000 nM, 4000 nM, 3000 nM, 2000 nM, 1000 nM, 900 nM, 800 nM, 500 nM, 250 nM, 200 nM, 100 nM, 50 nM, 30 nM, 20 nM, 10 nM, 7.5 nM, 7 nM, 6.5 nM, 6 The effective values are nM, 5 nM, 4.5 nM, 4 nM, 3.5 nM, 3 nM, 2.5 nM, 2 nM, 1.5 nM, 1 nM, 0.5 nM, 0.1 nM, 0.05 nM, 0.01 nM, or lower. In one specific embodiment, the antibody has a _KD of about 80 to about 200 pM. In another specific embodiment, the antibody has a _KD of about 10 nm to 2000 nm.

In one aspect, the present invention provides an antibody or pharmaceutical composition for use in a method of modulating a T-cell-mediated immune response in a subject with such a need. In a specific embodiment, the present invention provides an antibody or pharmaceutical composition for use in a method of inhibiting the growth of tumor cells in a subject.

In another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, optionally wherein said cancer is selected from: breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer, bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and cancers of the digestive system.

In yet another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, wherein a T-cell-mediated immune response is modulated or wherein the growth of tumor cells is inhibited.

The present invention also provides nucleic acid molecules encoding CD3 antibodies or portions thereof. For example, the present invention provides nucleic acid molecules encoding any VH or VL amino acid sequence listed in Table 1 or substantially similar sequences having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with them.

The present invention also provides nucleic acid molecules of any of the amino acid sequences of VH CDR1, VH CDR2 or VH CDR3 listed in Table 2 or substantially similar sequences having at least 90%, at least 95%, at least 98% or at least 99% sequence identity with them.

The present invention also provides nucleic acid molecules having a substantially similar sequence to any of the VL CDR1, VL CDR2 or VL CDR3 amino acid sequences listed in Table 3, or having at least 90%, at least 95%, at least 98% or at least 99% sequence identity with it.

The present invention also provides a nucleic acid molecule encoding HC VR, wherein the VH comprises a set of three CDRs (i.e., VH CDR1, VHCDR2, or VH CDR3), wherein the amino acid sequence sets of VH CDR1, VH CDR2, and VH CDR3 are as defined by any of the exemplary CD3 antibodies listed in Table 2.

The present invention also provides a nucleic acid molecule encoding LC VR, wherein the VL comprises a set of three CDRs (i.e., VL CDR1, VLCDR2, or VL CDR3), wherein the amino acid sequence sets of VL CDR1, VL CDR2, and VL CDR3 are as defined by any of the exemplary CD3 antibodies listed in Table 3.

The present invention also provides nucleic acid molecules encoding both VH and VL, wherein the VH comprises an amino acid sequence of any VH amino acid sequence listed in Table 1, and wherein the VL comprises an amino acid sequence of any VL amino acid sequence listed in Table 1.

The present invention also provides recombinant expression vectors capable of expressing polypeptides containing heavy or light chain variable regions of CD3 antibodies. For example, the present invention includes recombinant expression vectors comprising any of the nucleic acid molecules mentioned above (i.e., nucleic acid molecules encoding any of the VH and VL shown in Table 1 and/or CDR sequences as shown in Table 2 or 3). The scope of the present invention also includes host cells in which such vectors have been introduced, and methods for producing said antibodies or portions thereof by culturing said host cells under conditions allowing the production of antibodies or antibody fragments and recovering said antibodies and antibody fragments thus produced.

In some embodiments, the polynucleotide comprises a sequence encoding the heavy chain and/or light chain variable region of the following antibodies listed in Table 4: 2B5-0001 (2B5), 2B5-0002, 2B5-0006 (2B5v6), 2B5-0009, 2B5-0517, 2B5-0522, 2B5-0533, 2B5-0538, 2B5-0598, 2B5-0623, 2B5-0707, 2B5-0003, 2B5-1038, 2B5-1039, or 2B5-1040.

The sequence encoding the target antibody can be maintained in a vector within a host cell, and the host cell can then be amplified and frozen for future use. Vectors (including expression vectors) and host cells are further described herein.

Polynucleotides complementary to any such sequence are also covered by this invention. In one aspect, this invention provides a method for preparing any of the polynucleotides described herein. The polynucleotide may be single-stranded (coding or antisense) or double-stranded, and may be DNA (genomic, cDNA, or synthetic) or RNA molecules. RNA molecules include HnRNA molecules containing introns and corresponding one-to-one with DNA molecules; and mRNA molecules that do not contain introns. Additional coding or non-coding sequences may, but do not necessarily, be present in the polynucleotides of this invention, and the polynucleotides may, but do not necessarily, be linked to other molecules and/or supporting materials.

The polynucleotide may contain a native sequence (i.e., an endogenous sequence encoding an antibody or a portion thereof) or a variant thereof. Polynucleotide variants contain one or more substitutions, additions, deletions, and/or insertions such that the immunoreactivity of the encoded polypeptide is not reduced relative to the native immunoreactive molecule. The effect on the immunoreactivity of the encoded polypeptide can generally be evaluated as described herein. The variant preferably exhibits at least about 70% identity with the polynucleotide sequence encoding a native antibody or a portion thereof, more preferably at least about 80% identity, even more preferably at least about 90% identity, and most preferably at least about 95% identity.

The polynucleotides of this invention can be obtained using chemical synthesis, recombination methods, or PCR. Methods for the chemical synthesis of polynucleotides are well known in the art and need not be described in detail herein. Those skilled in the art can use the sequences provided herein and commercial DNA synthesizers to generate the desired DNA sequences.

To prepare polynucleotides using recombinant methods, a polynucleotide containing the desired sequence can be inserted into a suitable vector, which can then be introduced into a suitable host cell for replication and amplification, as discussed further herein. Polynucleotides can be inserted into host cells by any means known in the art. Cells are transformed by introducing exogenous polynucleotides through direct uptake, endocytosis, transfection, F-conjugation, or electroporation. Once introduced, the exogenous polynucleotide can be maintained within the cell as a non-integrating vector (such as a plasmid) or integrated into the host cell genome. The amplified polynucleotides can then be isolated from the host cell using methods well-known in the art. See, for example, Sambrook et al., 1989.

Alternatively, PCR allows for the replication of DNA sequences. PCR technology is well known in the art and is described in U.S. Patent Nos. 4,683,195, 4,800,159, 4,754,065, and 4,683,202, and PCR: The Polymerase Chain Reaction, edited by Mullis et al., Birkauswer Press, Boston, 1994.

RNA can be obtained by using DNA isolated in a suitable vector and inserted into a suitable host cell. Once the cell replicates and the DNA is transcribed into RNA, RNA can then be isolated using methods well known to those skilled in the art, such as Sambrook et al., 1989, as described above.

Suitable cloning vectors can be constructed according to standard techniques or selected from a large number of cloning vectors available in the art. While the chosen cloning vector can vary depending on the intended host cell, a useful cloning vector will generally be capable of self-replication, may have a single target for a specific restriction endonuclease, and/or may carry genes related to markers that can be used to select clones containing that vector. Suitable examples include plasmids and bacterial viruses such as pUC18, pUC19, Bluescript (e.g., pBS SK+) and its derivatives, mp18, mp19, pBR322, pMB9, ColE1, pCR1, RP4, phage DNA, and shuttle vectors such as pSA3 and pAT28. These and many other cloning vectors are available from commercial vendors such as BioRad, Strategene, and Invitrogen.

Expression vectors are typically reproducible polynucleotide constructs containing polynucleotides according to the invention. This implies that the expression vector must be replicable in the host cell as an episome or as an integral part of chromosomal DNA. Suitable expression vectors include, but are not limited to, plasmids, viral vectors including adenoviruses, adeno-associated viruses, retroviruses, kinases, and the expression vectors disclosed in PCT Publication No. WO 87/04462. Vector components may typically include, but are not limited to, one or more of the following: a signal sequence; an origin of replication; one or more marker genes; and suitable transcriptional control elements (such as promoters, enhancers, and terminators). For expression (i.e., translation), one or more translational control elements, such as ribosome binding sites, translation initiation sites, and stop codons, are also typically required.

Vectors containing the target polynucleotide can be introduced into host cells by any of a number of suitable means, including electroporation, transfection with calcium chloride, rubidium chloride, calcium phosphate, DEAE-glucan, or other substances; particle bombardment; lipid transfection; and infection (e.g., when the vector is an infectious pathogen such as vaccinia virus). The choice of vector or polynucleotide to be introduced will often depend on the characteristics of the host cell.

This invention also provides host cells comprising any of the polynucleotides described herein. Any host cell capable of overexpressing heterologous DNA can be used for the purpose of isolating genes encoding target antibodies, peptides, or proteins. Non-limiting examples of mammalian host cells include, but are not limited to, COS, HeLa, and CHO cells. See also PCT Publication WO 87/04462. Suitable non-mammalian host cells include prokaryotes (such as Escherichia coli or Bacillus subtilis ) and yeasts (such as Saccharomyces cerevisiae , Saccharomyces pombe , or Kluyveromyces lactis ). Preferably, the host cell expresses cDNA at a level about 5-fold higher, more preferably 10-fold higher, and even more preferably 20-fold higher than the corresponding endogenous antibody or protein CD3 or CD3 domain (e.g., domains 1-4), as determined by immunoassay or FACS. Cells overexpressing target antibodies or proteins can be identified.

This invention also covers scFvs of the antibodies of this invention. Single-chain variable region fragments are prepared by linking light and/or heavy chain variable regions using a short linker peptide (Bird et al., Science 242:423-426, 1988). Examples of linker peptides include the sequence of SEQ ID NO: 65. In one embodiment, the linker peptide bridges approximately 3.5 nm between the carboxyl terminus of one variable region and the amino terminus of another variable region. Linkers with other sequences have been designed and used (Bird et al., Science 242:423-426, 1988). Linkers should be short, flexible polypeptides and preferably contain fewer than about 20 amino acid residues. The linker can then be modified for additional functions, such as linking drugs or solid carriers. Single-chain variants can be generated recombinantly or synthetically. An automated synthesizer can be used to synthesize scFvs. To generate scFv through recombination, a suitable plasmid containing a polynucleotide encoding the scFv can be introduced into a suitable host cell, which can be eukaryotic (such as yeast, plant, insect, or mammalian cells) or prokaryotic (such as Escherichia coli). The polynucleotide encoding the target scFv can be prepared by conventional manipulation (such as linking the polynucleotide). The obtained scFv can be isolated using standard protein purification techniques known in the art.

In one aspect, polynucleotide sequences encoding any of the partial light chain sequences and/or any of the partial heavy chain sequences disclosed herein are provided. Exemplary CD3 polynucleotides encoding the VH and VL regions of the antibodies of the present invention are listed in Table 5.

Table 5

In one aspect, the present invention provides compositions (such as pharmaceutical compositions) comprising any of the polynucleotides of the present invention. In some embodiments, the composition may comprise any of the polynucleotides shown in Table 5. In a particular embodiment, the composition comprises an expression vector containing a polynucleotide encoding any of the antibodies described herein (e.g., any of the polynucleotides shown in Table 5).

This invention also covers polynucleotides complementary to any such sequence. Polynucleotides can be single-stranded (coding or antisense) or double-stranded, and can be DNA (genomic, cDNA, or synthetic) or RNA molecules. RNA molecules include mature and immature mRNAs, such as precursor mRNA (pre-mRNA) or heteronuclear mRNA (hnRNA) and mature mRNA. Additional coding or non-coding sequences may, but do not necessarily, be present within the polynucleotides of this invention, and the polynucleotides may, but do not necessarily, be linked to other molecules and/or supporting materials.

The polynucleotide may contain a native sequence (i.e., an endogenous sequence encoding an antibody or a portion thereof) or a variant thereof. Polynucleotide variants contain one or more substitutions, additions, deletions, and/or insertions such that the immunoreactivity of the encoded polypeptide is not reduced relative to the native immunoreactive molecule. The effect on the immunoreactivity of the encoded polypeptide can generally be evaluated as described herein. Variants preferably exhibit at least about 70% identity with the polynucleotide sequence encoding a native antibody or a portion thereof, more preferably at least about 80% identity, even more preferably at least about 90% identity, and most preferably at least about 95% identity.

Two polynucleotide or polypeptide sequences are said to be "identical" if the nucleotide or amino acid sequences in the two sequences are identical when compared for maximum correspondence as described below. Comparisons between two sequences are typically performed by comparing sequences on a comparison window to identify and compare local regions of sequence similarity. As used herein, a "comparison window" refers to a segment of at least about 20, typically 30 to about 75, or 40 to about 50 adjacent positions, where, after optimal alignment of the two sequences, the sequence can be compared to a reference sequence having the same number of adjacent positions.

Using default parameters, the Megalign procedure in the Lasergene Bioinformatics Software Suite (DNASTAR, Inc., Madison, WI) allows for optimal alignment of sequences for comparison. Preferably, the "percentage of sequence identity" is determined by comparing two optimally aligned sequences across a comparison window of at least 20 positions. The portion of the polynucleotide or polypeptide sequence in the comparison window may contain 20% or less, typically 5 to 15%, or 10 to 12%, of additions or deletions (i.e., gaps) compared to a reference sequence (which does not contain additions or deletions) for optimal alignment of the two sequences. The percentage is calculated by determining the number of positions in both sequences where the same nucleic acid base or amino acid residue appears, to obtain the number of matching positions; dividing the number of matching positions by the total number of positions in the reference sequence (i.e., the window size); and multiplying the result by 100 to obtain the percentage of sequence identity.

The variant can also, or alternatively, be substantially homologous to the natural gene or a portion thereof or its complement. This polynucleotide variant is capable of hybridizing with naturally occurring DNA sequences (or complementary sequences) encoding natural antibodies under moderately stringent conditions.

Those skilled in the art will understand that, due to the degeneracy of the genetic code, there are many nucleotide sequences encoding polypeptides as described herein. Some of these polynucleotides have minimal homology to the nucleotide sequences of any natural gene. However, the present invention specifically considers polynucleotides that vary due to differences in codon usage. Furthermore, alleles of genes comprising the polynucleotide sequences provided herein are within the scope of the invention. An allele is an endogenous gene altered due to one or more mutations (such as nucleotide deletions, additions, and/or substitutions). The resulting mRNA and protein may, but need not, have altered structure or function. Alleles can be identified using standard techniques (such as hybridization, amplification, and/or database sequence comparison).

In one aspect, the present invention provides a method for preparing any of the polynucleotides described herein. For example, the polynucleotides of the present invention can be obtained using chemical synthesis, recombination methods, or PCR. Methods for the chemical synthesis of polynucleotides are well known in the art and need not be described in detail herein. Those skilled in the art can use the sequences provided herein and commercial DNA synthesizers to generate desired DNA sequences.

To prepare polynucleotides using recombinant methods, a polynucleotide containing the desired sequence can be inserted into a suitable vector, which can then be introduced into a suitable host cell for replication and amplification, as discussed further herein. Polynucleotides can be inserted into host cells by any means known in the art. Cells are transformed by introducing exogenous polynucleotides through direct uptake, endocytosis, transfection, F-crossing, or electroporation. Once introduced, the exogenous polynucleotide can be maintained within the cell as a non-integrating vector (such as a plasmid) or integrated into the host cell genome. The amplified polynucleotides can then be isolated from the host cell using methods well-known in the art (e.g., Sambrook et al., 1989).

Alternatively, PCR allows for the replication of DNA sequences. PCR technology is well known in the art and is described in U.S. Patent Nos. 4,683,195, 4,800,159, 4,754,065, and 4,683,202, and PCR: The Polymerase Chain Reaction, edited by Mullis et al., Birkauswer Press, Boston, 1994.

RNA can be obtained by using DNA isolated in a suitable vector and inserted into a suitable host cell. Once the cell replicates and the DNA is transcribed into RNA, RNA can then be isolated using methods well known to those skilled in the art, such as Sambrook et al., 1989, as described above.

Suitable cloning vectors can be constructed using standard techniques or selected from a large number of cloning vectors available in the art. While the chosen cloning vector can vary depending on the intended host cell, useful cloning vectors typically possess self-replicating capabilities, may have a single target for a specific restriction endonuclease, and/or may carry genes related to markers that can be used to select clones containing that vector. Suitable examples include plasmids and bacterial viruses such as pUC18, pUC19, Bluescript (e.g., pBS SK+) and its derivatives, mp18, mp19, pBR322, pMB9, ColE1, pCR1, RP4, phage DNA, and shuttle vectors such as pSA3 and pAT28. These and many other cloning vectors are available from commercial vendors such as BioRad, Strategene, and Invitrogen.

Expression vectors are typically reproducible polynucleotide constructs containing polynucleotides according to the invention. This implies that the expression vector must be replicable in the host cell as an episome or as an integral part of chromosomal DNA. Suitable expression vectors include, but are not limited to, plasmids, viral vectors including adenoviruses, adeno-associated viruses, retroviruses, viscera, and the expression vectors disclosed in PCT Publication No. WO87/04462. Vector components may typically include, but are not limited to, one or more of the following: a signal sequence; an origin of replication; one or more marker genes; and suitable transcriptional control elements (such as promoters, enhancers, and terminators). For expression (i.e., translation), one or more translational control elements, such as ribosome binding sites, translation initiation sites, and stop codons, are also typically required.

Vectors containing the target polynucleotide can be introduced into host cells by any of a number of suitable means, including electroporation, transfection with calcium chloride, rubidium chloride, calcium phosphate, DEAE-glucan, or other substances; particle bombardment; lipid transfection; and infection (e.g., when the vector is an infectious pathogen such as vaccinia virus). The choice of vector or polynucleotide to be introduced will often depend on the characteristics of the host cell.

Any host cell capable of overexpressing heterologous DNA can be used for the purpose of isolating genes encoding target antibodies, peptides, or proteins. Non-limiting examples of mammalian host cells include, but are not limited to, COS, HeLa, and CHO cells. See also PCT Publication WO87/04462. Suitable non-mammalian host cells include prokaryotes (such as *Escherichia coli* or *Bacillus subtilis *) and yeasts (such as *Saccharomyces cerevisiae *, *Schizosaccharomyces pombe * , or *Kluyveromyces lactis *). Preferably, the host cell expresses cDNA at a level approximately 5-fold higher, more preferably 10-fold higher, and even more preferably 20-fold higher than the corresponding endogenous antibody or protein CD3 or CD3 domain (e.g., domains 1-4), as determined by immunoassay or FACS. Cells overexpressing the target antibody or protein can be identified.

In one respect, the CD3 antibody molecule will have an affinity for CD3, as measured, for example, by direct or competitive binding in the picomolar to micromolar affinity range, preferably in the picomolar to low nanomolar range.

Bispecific antibodies can be prepared using the sequences disclosed herein. Methods for preparing bispecific antibodies are known in the art (see, for example, Suresh et al., Methods in Enzymology 121:210, 1986). Traditionally, the recombinant generation of bispecific antibodies is based on the co-expression of two immunoglobulin heavy-light chain pairs, where the two heavy chains have different specificities (Millstein and Cuello, Nature 305, 537-539, 1983).

In another embodiment, the antibody as described herein comprises a full-length human antibody, wherein the antibody variable region of the heterodimeric protein is capable of recruiting human immune effector cells by specifically binding to effector antigens (e.g., CD3 antigens) located on human immune effector cells, and wherein a second antibody variable region of the heterodimeric protein is capable of specifically binding to the target antigen. In some embodiments, the human antibody has an IgG1, IgG2, IgG3, or IgG4 isotype. In some embodiments, the heterodimeric protein comprises an immune-inert Fc chain.

Human immune effector cells can be any variety of immune effector cells known in the art. For example, immune effector cells can be members of the human lymphoid cell lineage, including but not limited to T cells (e.g., cytotoxic T cells), B cells, and natural killer (NK) cells. Immune effector cells can also be, for example, but not limited to, members of the human myeloid lineage, including but not limited to monocytes, neutrophils, and dendritic cells. Such immune effector cells can have cytotoxic or apoptotic effects on target cells, or other desired effects, after activation by binding to effector antigens.

Effector antigens are antigens (e.g., proteins or peptides) expressed on human immune effector cells. Examples of effector antigens that can be bound by heterodimers (e.g., heterodimer antibodies or bispecific antibodies) include, but are not limited to, human CD3, CD16, NKG2D, NKp46, CD2, CD28, CD25, CD64, and CD89.

Target cells can be natural or foreign to humans. In natural target cells, the cell may have been transformed into a malignant cell or pathologically modified (e.g., a natural target cell infected by a virus, malaria parasite, or bacteria). In foreign target cells, the cell is an invading pathogen, such as bacteria, malaria parasite, or virus.

In some embodiments, the antibodies that can be used in this invention are Fab, Fab fragment, F(ab) ₂ fragment, Fv fragment, single-chain Fv fragment, disulfide-stabilized Fv fragment, single-chain antibody, monoclonal antibody, chimeric antibody, bispecific antibody, trispecific antibody, multispecific antibody, bispecific heterodimeric biantibody, bispecific heterodimeric IgG, polyclonal antibody, labeled antibody, humanized antibody, human antibody, or fragments thereof.

In some implementations, the CD3 antibody described herein is a monoclonal antibody. For example, the CD3 antibody is a humanized monoclonal antibody or a chimeric monoclonal antibody.

This invention covers antibodies comprising an Fc chain or a domain thereof, or a portion thereof. In some embodiments, the Fc chain or a portion thereof comprises one or more constant domains (e.g., CH2 or CH3 domains) of an Fc chain of IgG1, IgG2, IgG3, or IgG4. In another embodiment, this invention covers molecules comprising an Fc chain or a portion thereof, wherein the Fc chain or a portion thereof contains at least one amino acid modification (e.g., substitution) relative to an equivalent wild-type Fc chain or a portion thereof. Variant Fc regions are well known in the art and are primarily used to alter the phenotype of antibodies comprising variant Fc regions, as determined in any binding activity or effector function assay well known in the art (e.g., ELISA, SPR assay, or ADCC). Such variant Fc chains or portions thereof can prolong the plasma half-life and stability exhibited by bispecific antibodies comprising Fc chains or portions thereof of this invention. In another embodiment, this invention covers the use of any Fc variant known in the art.

In one embodiment, one or more amino acid modifications are made to the Fc chain to reduce the affinity and coercivity of the Fc chain, thereby reducing the affinity and coercivity of the bispecific antibody of the present invention for one or more FcγR receptors. In one specific embodiment, the present invention covers bispecific antibodies comprising a variant Fc chain or a portion thereof, wherein the variant Fc chain contains at least one amino acid modification relative to the wild-type Fc chain, and the variant Fc region binds only one FcγR, wherein the FcγR is FcγRIIIA. In another specific embodiment, the present invention covers bispecific antibodies comprising a variant Fc chain or a portion thereof, wherein the variant Fc chain contains at least one amino acid modification relative to the wild-type (WT) Fc chain, and the variant Fc region binds only one FcγR, wherein the FcγR is FcγRIIA. In yet another specific embodiment, the present invention covers bispecific antibodies comprising a variant Fc chain or a portion thereof, wherein the variant Fc chain contains at least one amino acid modification relative to the wild-type Fc chain, and the variant Fc chain binds only one FcγR, wherein the FcγR is FcγRIIB. In another embodiment, the invention covers molecules comprising a variant Fc chain, wherein, relative to molecules that do not contain an Fc chain or contain a wild-type Fc chain, the variant confers or mediates reduced ADCC activity (or other effector functions) and/or increased binding to FcγRIIB (CD32B), as measured using methods known to those skilled in the art and described herein.

This invention also covers the use of Fc regions comprising domains or regions from two or more IgG isoforms. As is known in the art, amino acid modifications to the Fc region can significantly affect Fc-mediated effector function and/or binding activity. However, when implemented in the context of a selected IgG isoform, these alterations to functional characteristics can be further improved and/or manipulated. Similarly, the native characteristics of isoform Fc can be manipulated through one or more amino acid modifications. Multiple IgG isoforms (i.e., IgG1, IgG2, IgG3, and IgG4) exhibit different physical and functional properties, including serum half-life, complement binding, FcγR binding affinity, and effector functional activity (e.g., ADCC, CDC), due to differences in the amino acid sequences of their hinges and/or Fc regions.

In one embodiment, amino acid modifications and IgG Fc regions are independently selected based on their respective individual binding and/or effector functional activities to engineer bispecific antibodies with desired characteristics. In one specific embodiment, the binding and/or effector functional activities of amino acid modifications and IgG hinge/Fc regions are separately measured in the context of IgG1, as described herein or known in the art. In one embodiment, amino acid modifications and IgG hinge/Fc regions exhibit similar functionality, such as reduced ADCC activity (or other effector functions) and/or increased binding to FcγRIIB in the context of bispecific antibodies or other Fc-containing molecules (e.g., immunoglobulins). In another embodiment, the invention covers variant Fc regions comprising combinations of amino acid modifications known in the art and selected IgG regions exhibiting novel properties that are undetectable when the modifications and/or regions are measured independently as described herein.

One method for determining the binding affinity of an antibody to CD3 is by measuring the binding affinity of a monofunctional Fab fragment of the antibody. To obtain a monofunctional Fab fragment, the antibody (e.g., IgG) can be cleaved with papain or recombinantly expressed. The affinity of the antibody's CD3 Fab fragment can be determined using a surface plasmon resonance (SPR) system (Biacore™ 3000™ SPR system, Biacore™, INC, Piscataway NJ) equipped with a pre-immobilized streptavidin sensor chip (SA) or anti-mouse Fc or anti-human Fc, using HBS-EP run buffer (0.01 M HEPES, pH 7.4, 0.15 M NaCl, 3 mM EDTA, 0.005% v/v surfactant P20). Biotinylated or Fc-fused human CD3 can be diluted in HBS-EP buffer to concentrations less than 0.5 μg/mL and injected across individual chip channels using variable contact times to achieve two ranges of antigen density: 50–200 reaction units (RU) for detailed kinetic studies, or 800–1,000 RU for screening assays. Regeneration studies have shown that 25 mM NaOH in 25% v/v ethanol effectively removes bound Fab while maintaining CD3 activity on the chip for more than 200 injections. Typically, serial dilutions of purified Fab samples (across concentrations of 0.1–10x estimated _KD ) are injected at 100 µL/min for 1 minute, allowing dissociation times up to 2 hours. Fab protein concentrations are determined by ELISA and/or SDS-PAGE electrophoresis using known concentrations of Fab (e.g., as determined by amino acid analysis) as standards. The data were fitted to a 1:1 Lammill binding model using the BIAevaluation procedure (Karlsson, R. Roos, H. Fagerstam, L. Petersson, B. (1994). Methods Enzymology 6. 99-110), simultaneously obtaining the kinetic binding rate (k _{_on} ) and dissociation rate (k_{_off} ). The equilibrium dissociation constant (K_{_D}) was calculated as k _{_off} /k _{_on} . This protocol is suitable for determining the binding affinity of antibodies to any CD3, including human CD3, CD3 from another mammal (such as mouse CD3, rat CD3, or primate CD3), and different forms of CD3 (e.g., glycosylated CD3). Antibody binding affinity is typically measured at 25°C, but can also be measured at 37°C.

Antibodies as described herein can be prepared by any method known in the art. For the generation of hybridoma cell lines, the host animal immunization pathway and timeline are generally consistent with established and routine techniques for antibody stimulation and production, as further described herein. General techniques for generating human and mouse antibodies are known in the art and/or described herein.

Consider manipulating any mammalian subject, including humans or antibody-producing cells derived therefrom, to serve as a basis for generating mammalian, including human, hybridoma cell lines. Typically, host animals are inoculated with a given amount of immunogen via intraperitoneal, intramuscular, oral, subcutaneous, plantar, and/or intradermal administration, including as described herein.

Hybridomas can be prepared from lymphocytes and immortalized myeloma cells using general somatic cell hybridization techniques, as described in Kohler, B. and Milstein, C., Nature 256:495-497, 1975, or modified techniques, as described in Buck, D. W. et al., In Vitro, 18:377-381, 1982. Available myeloma lines, including but not limited to X63-Ag8.653 and those from the Salk Institute, Cell Distribution Center, San Diego, Calif., USA, can be used in the hybridization. Typically, the technique involves fusing myeloma cells and lymphoid cells using a fusion agent, such as polyethylene glycol, or by electrical means well known to those skilled in the art. After fusion, the cells are separated from the fusion medium and grown in a selective growth medium (such as hypoxanthine-aminopterin-thymidine (HAT) medium) to eliminate unhybridized parental cells. Any culture medium, with or without serum supplementation, as described herein can be used to culture hybridomas that secrete monoclonal antibodies. As an alternative to cell fusion technology, EBV immortalized B cells can be used to generate the monoclonal antibodies of this invention. If necessary, the hybridoma is amplified and subcloned, and the antiimmunogen activity of the supernatant is determined using routine immunoassay procedures (e.g., radioimmunoassay, enzyme immunoassay, or fluorescence immunoassay).

Hybridomas that can be used as antibody sources encompass all derivatives of parental hybridomas that produce monoclonal antibodies specific to CD3 or a portion thereof, as well as progeny cells.

Hybridomas that produce such antibodies can be grown in vitro or in vivo using known procedures. If desired, monoclonal antibodies can be isolated from culture media or body fluids using routine immunoglobulin purification procedures such as ammonium sulfate precipitation, gel electrophoresis, dialysis, chromatography, and ultrafiltration. For example, unwanted activity (if present) can be removed by running the formulation on an adsorbent made of an immunogen attached to a solid phase and eluting or releasing the desired antibody from the immunogen. Using bifunctional reagents or derivatizing agents, such as maleimide benzoyl sulfosuccinimide (conjugated via cysteine residues), N-hydroxysuccinimide (conjugated via lysine residues), glutaraldehyde, succinic anhydride, _SOCl2 , or ^R1N =C=NR (where R and ^R1 are different alkyl groups), conjugated with human CD3 or fragments containing the target amino acid sequence in the host animal to be immunized, a population of antibodies (e.g., monoclonal antibodies) can be obtained.

If desired, the target antibody can be sequenced, and the polynucleotide sequence can then be cloned into a vector for expression or propagation. The sequence encoding the target antibody can be maintained in a vector within a host cell, which can then be expanded and frozen for future use. The generation of recombinant monoclonal antibodies in cell culture can be performed by means known in the art through cloning antibody genes from B cells. See, for example, Tiller et al., J. Immunol. Methods 329,112,2008; U.S. Patent No. 7,314,622.

In an alternative approach, polynucleotide sequences can be genetically manipulated to humanize antibodies or improve their affinity or other characteristics. For example, if an antibody is used in clinical trials and treatments in humans, the constant region can be engineered to more closely resemble the human constant region to avoid an immune response. It might be desirable to genetically manipulate antibody sequences to achieve greater affinity for CD3 and greater therapeutic efficacy.

In some embodiments, the antibody has a modified constant region that removes or reduces Fcγ receptor binding. For example, Fc may be human IgG2 containing the mutant D265, where the amino acid residues are numbered with reference to the wild-type IgG2 sequence. Therefore, in some embodiments, the constant region has a modified constant region having the sequence shown in SEQ ID NO: 80. In some embodiments, the antibody has a modified constant region having the sequence shown in SEQ ID NO: 81.

There are four general steps to humanizing a monoclonal antibody. These steps are: (1) determining the nucleotide and predicted amino acid sequences of the variable regions of the light and heavy chains of the starting antibody; (2) designing the humanized antibody, i.e., deciding which antibody framework region to use during the humanization process; (3) the actual humanization method/technique; and (4) the transfection and expression of the humanized antibody. See, for example, U.S. Patent Nos. 4,816,567; 5,807,715; 5,866,692; 6,331,415; 5,530,101; 5,693,761; 5,693,762; 5,585,089; and 6,180,370.

Many humanized antibody molecules containing antigen-binding sites derived from non-human immunoglobulins have been described, including chimeric antibodies with rodent or modified rodent V regions fused to human constant regions and their associated CDRs. See, for example, Winter et al. Nature 349:293-299, 1991; Lobuglio et al. Proc. Nat. Acad. Sci. USA 86:4220-4224, 1989; Shaw et al. J Immunol. 138:4534-4538, 1987; and Brown et al. Cancer Res. 47:3577-3583, 1987. Other references describe the transplantation of rodent CDRs into human support framework regions (FRs) prior to fusion with appropriate human antibody constant regions. See, for example, Riechmann et al., Nature 332:323-327, 1988; Verhoeyen et al., Science 239:1534-1536, 1988; and Jones et al., Nature 321:522-525, 1986. Another reference describes rodent CDRs supported by recombinantly engineered rodent framework regions. See, for example, European Publication No. EP0519596. These “humanized” molecules are designed to minimize unwanted immune responses to rodent anti-human antibody molecules, which limits the duration and effectiveness of therapeutic applications that are partially present in human receptors. For example, antibody constant regions can be engineered to be immunologically inert (e.g., not triggering complement cleavage). See, for example, PCT application No. PCT/GB99/01441; UK application No. 9809951.8. Other methods using humanized antibodies may also be disclosed by Daugherty et al., Nucl. Acids Res. 19:2471-2476, 1991, and in U.S. Patent Nos. 6,180,377; 6,054,297; 5,997,867; 5,866,692; 6,210,671; and 6,350,861; and PCT Publication No. WO01/27160.

The general principles discussed above regarding humanized antibodies also apply to custom antibodies, for example, for use in other mammals (e.g., humans, non-human primates, mice, donkeys, sheep, rabbits, goats, guinea pigs, camels, horses, or chickens). Furthermore, one or more aspects of the humanized antibodies described herein can be combined, such as CDR transplantation, scaffold mutation, and CDR mutation.

In one variant, fully human antibodies can be obtained using commercially available mice engineered to express specific human immunoglobulin proteins. Transgenic animals designed to produce a more desired (e.g., fully human antibody) or more robust immune response can also be used to generate humanized or human antibodies. Examples of this technology are Xenomouse™ from Abgenix, Inc. (Fremont, CA) and HuMAb-Mouse® and TC Mouse™ from Medarex, Inc. (Princeton, NJ).

In one alternative, the antibody can be recombinantly prepared and expressed using any method known in the art. In another alternative, the antibody can be recombinantly prepared using phage display technology. See, for example, U.S. Patent Nos. 5,565,332; 5,580,717; 5,733,743; and 6,265,150; and Winter et al., Annu. Rev. Immunol. 12:433-455, 1994. Alternatively, phage display technology (McCafferty et al., Nature 348:552-553, 1990) can be used to generate human antibodies and antibody fragments in vitro from a gene library of immunoglobulin variable (V) domains from non-immunized donors. According to this technology, the antibody V domain gene frame is cloned into the major or minor coat protein genes of filamentous phages, such as M13 or fd, and displayed as a functional antibody fragment on the surface of the phage particle. Because filamentous particles contain single-stranded DNA copies of the phage genome, selection based on antibody functional characteristics also leads to the selection of genes encoding antibodies that exhibit those characteristics. Thus, phages mimic some characteristics of B cells. Phage display can be performed in various forms; for reviews, see, for example, Johnson, Kevin S., and Chiswell, David J., Current Opinion in Structural Biology 3:564-571, 1993. Several sources of the V gene segment can be used for phage display. Clackson et al., Nature 352:624-628, 1991, isolated a wide variety of anti-oxazolone antibodies from a small, randomly combined library of the V gene derived from the spleen of immunized mice. Following the techniques described by Mark et al., J. Mol. Biol. 222:581-597, 1991, or Griffith et al., EMBO J.12:725-734, 1993, a V gene library from non-immunized human donors can be constructed, and antibodies against a wide variety of antigens, including self-antigens, can be isolated. In the innate immune response, antibody genes accumulate mutations at a high rate (somatic hypermutation). Some of the introduced changes will confer higher affinity, and B cells exhibiting high-affinity surface immunoglobulins will preferentially replicate and differentiate during subsequent antigen challenges. This natural process can be simulated using a technique called “chain shuffling” (Marks et al., Bio/Technol. 10:779-783, 1992). In this approach, the affinity of “primary” human antibodies obtained by phage display can be improved by sequentially replacing the heavy and light chain V region genes with a naturally occurring library (library) of variants of the V domain genes obtained from non-immunized donors. This technique allows for the production of antibodies and antibody fragments with affinities in the pM-nM range. A strategy for preparing very large phage antibody libraries (also known as “mother-of-all libraries”) has been described by Waterhouse et al., Nucl. Acids Res. 21:2265-2266, 1993. Gene shuffling can also be used to derive human antibodies from rodent antibodies, wherein the human antibodies have similar affinity and specificity to the starting rodent antibody. According to this method, also known as “epitope imprinting,” the heavy or light chain V domain genes of rodent antibodies obtained via phage display technology are replaced with a human V domain gene library, producing a rodent-human chimera. Selection on the antigen results in the isolation of the human variable region capable of restoring a functional antigen-binding site; that is, the epitope governs (marks) the selection of the partner. When this process is repeated to replace the remaining rodent V domains, human antibodies are obtained (see PCT Publication No. WO93/06213). Unlike traditional humanization of rodent antibodies via CDR transplantation, this technique provides fully human antibodies that do not contain rodent-derived frameworks or CDR residues.

Antibodies can be prepared recombinantly by first isolating the antibody and antibody-producing cells from a host animal, obtaining the gene sequence, and using the gene sequence to recombinantly express the antibody in host cells (e.g., CHO cells). Another approach is to express the antibody sequence in plants (e.g., tobacco) or transgenic milk. Methods for recombinantly expressing antibodies in plants or milk have been disclosed. See, for example, Peeters et al. Vaccine 19:2756, 2001; Lonberg, N. and D. Huszar Int. Rev. Immunol 13:65, 1995; and Pollock et al., J Immunol Methods 231:147, 1999. Methods for preparing antibody derivatives, such as humanized, single-chain, etc., are known in the art.

Immunoassays and flow cytometry sorting techniques, such as fluorescence activated cell sorting (FACS), can also be used to isolate antibodies specific to CD3 or target antigens.

Antibodies as described herein can bind to a variety of different solid supports or carriers. Such supports can be active and/or inert. Well-known supports include polypropylene, polystyrene, polyethylene, dextran, nylon, amylase, glass, natural and modified cellulose, polyacrylamide, agarose, and magnets. For the purposes of this invention, the nature of the support can be soluble or insoluble. Those skilled in the art will recognize other suitable supports for binding antibodies, or will be able to determine these using routine experiments. In some embodiments, the support comprises a portion targeting the myocardium.

DNA encoding monoclonal antibodies can be readily isolated and sequenced using standard procedures, such as by using oligonucleotide probes capable of specifically binding to genes encoding the heavy and light chains of monoclonal antibodies. Hybridoma cells serve as a preferred source of this DNA. Once isolated, the DNA can be placed in an expression vector (such as the one disclosed in PCT Publication No. WO87/04462) and then transfected into host cells that otherwise do not produce immunoglobulin proteins (such as E. coli cells, simian COS cells, CHO cells, or myeloma cells) to obtain monoclonal antibody synthesis in recombinant host cells. See, for example, PCT Publication No. WO87/04462. DNA can also be modified, for example, by replacing homologous mouse sequences with coding sequences of human heavy and light chain constant regions (Morrison et al., Proc. Nat. Acad. Sci. 81:6851, 1984), or by covalently binding all or part of the coding sequence of an immunoglobulin-coding sequence to the coding sequence of a non-immunoglobulin polypeptide. In this manner, "chimeric" or "hybrid" antibodies with the binding specificity of the monoclonal antibodies described herein are prepared.

Methods known in the art can be used to identify or characterize CD3 or other antigen antibodies as described herein, thereby detecting and/or measuring antigen expression levels. In some embodiments, CD3 antibodies are identified by incubating a candidate antibody with CD3 and monitoring the associated decrease in binding and/or CD3 expression levels. Binding assays can be performed using purified CD3 peptides, or cells naturally expressing or transfected to express CD3 peptides. In one embodiment, the binding assay is a competitive binding assay in which the ability of a candidate antibody to compete with a known CD3 antibody for CD3 binding is evaluated. This assay can be performed in various forms, including ELISA.

Following initial identification, the activity of candidate CD3 or other antigen-antibody pairs can be further confirmed and refined using bioassays known for testing targeted biological activity. Alternatively, bioassays can be used to directly screen candidates. Several methods for identifying and characterizing antibodies are described in detail in the examples.

CD3 or other antigen antibodies can be characterized using methods well-known in the art. For example, one method is to identify the epitope to which it binds, or "epitope mapping." Many methods are known in the art for mapping and characterizing the location of epitopes on proteins, including resolving the crystal structure of antibody-antigen complexes, competitive assays, gene fragment expression assays, and synthetic peptide-based assays, as described, for example, in Chapter 11 of Harlow and Lane, Using Antibodies, a Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 1999. In another instance, epitope mapping can be used to determine the sequence to which the antibody binds. Epitope mapping is commercially available from various sources, for example, Pepscan Systems (Edelhertweg 15, 8219 PH Lelystad, Netherlands). The epitope can be a linear epitope, i.e., contained in a single amino acid segment, or a conformational epitope formed by the three-dimensional interactions of amino acids, which may not necessarily be contained in a single segment. Peptides of varying lengths (e.g., at least 4-6 amino acids long) can be isolated or synthesized (e.g., recombinantly) and used for binding assays with CD3 or other antigens. In another example, epitopes to which CD3 or other antigens bind can be determined in systematic screening by using overlapping peptides derived from CD3 or other antigen sequences and measuring binding with CD3 or other antigens. Based on gene fragment expression assays, open reading frames encoding CD3 or other antigens are randomly or through specific genetic constructs fragmented, and the reactivity of expressed fragments of CD3 or other antigens with the antibody to be tested is determined. Gene fragments can be generated, for example, by PCR, and then transcribed and translated into proteins in vitro in the presence of radioactive amino acids. The binding of the antibody to the radiolabeled CD3 or other antigen fragment is then determined by immunoprecipitation and gel electrophoresis. Some epitopes can also be identified using large libraries (phage libraries) of random peptide sequences displayed on the surface of phage particles. Alternatively, the binding of a defined library of overlapping peptide fragments to a test antibody can be tested in a simple binding assay. In an additional instance, mutagenesis of antigen-binding domains, domain exchange assays, and alanine partitioning can be performed to identify residues required, sufficient, and/or essential for epitope binding. For example, domain exchange assays can be performed using mutant CD3 or other antigens, where various fragments of the CD3 or other antigenic protein have been replaced (exchanged) with sequences from antigens from another species (e.g., mice), or closely related but antigenically different proteins (e.g., Trop-1). By evaluating the binding of the antibody to mutant CD3 or other antigens, the importance of specific CD3 or other antigenic fragments to antibody binding can be assessed.

Another method that can be used to characterize CD3 or other antigen-antibodies is to use a competitive assay with other antibodies known to bind to the same antigen, i.e., various fragments on CD3 or other antigens, to determine whether the CD3 or other antigen-antibody binds to the same epitopes as the other antibodies. Competitive assays are well known to those skilled in the art.

Expression vectors can be used to direct the expression of CD3 or other antigens and antibodies. Those skilled in the art are familiar with the administration of expression vectors to obtain the expression of exogenous proteins in vivo. See, for example, U.S. Patent Nos. 6,436,908; 6,413,942; and 6,376,471. Administration of expression vectors includes local or systemic administration, including injection, oral administration, particle gun or catheter-directed administration, and local application. In another embodiment, the expression vector is administered directly to the sympathetic trunk or ganglia, or to the coronary arteries, atria, ventricles, or pericardium.

Targeted delivery can also be achieved using therapeutic compositions containing expression vectors or subgenomic polynucleotides. Receptor-mediated DNA delivery techniques are described, for example, in: Findeis et al., Trends Biotechnol., 11:202, 1993; Chiou et al., Gene Therapeutics: Methods And Applications of Direct Gene Transfer, J.A. Wolff, ed., 1994; Wu et al., J. Biol. Chem., 263:621, 1988; Wu et al., J. Biol. Chem., 269:542, 1994; Zenke et al., Proc. Natl. Acad. Sci. USA, 87:3655, 1990; and Wu et al., J. Biol. Chem., 266:338, 1991. Therapeutic compositions containing polynucleotides are administered topically in gene therapy regimens at concentrations ranging from about 100 ng to about 200 mg of DNA. Concentrations ranging from about 500 ng to about 50 mg, about 1 μg to about 2 mg, about 5 μg to about 500 μg, and about 20 µg to about 100 µg of DNA may also be used during gene therapy regimens. Gene delivery vehicles can be used to deliver therapeutic polynucleotides and peptides. Gene delivery vehicles can be of viral or non-viral origin (see generally, Jolly, Cancer Gene Therapy, 1:51, 1994; Kimura, Human Gene Therapy, 5:845, 1994; Connelly, Human Gene Therapy, 1:185, 1995; and Kaplitt, Nature Genetics, 6:148, 1994). Expression of such coding sequences can be induced using endogenous mammalian promoters or heterologous promoters. The expression of a coding sequence can be constitutive or regulated.

Virus-based vectors for delivering desired polynucleotides and expressing them in desired cells are well known in the art. Exemplary virus-based vectors include, but are not limited to, recombinant retroviruses (see, for example, PCT Publications WO90/07936; WO94/03622; WO93/25698; WO93/25234; WO93/11230; WO93/10218; WO91/02805; U.S. Patents 5,219,740 and 4,777,127; GB Patent 2,200,651; and EP Patent EP0 345242), alphavirus-based vectors (e.g., Sindbis virus vectors, Semliki Forest Virus (ATCC)). VR-67; ATCCVR-1247), Ross River virus (ATCC VR-373; ATCCVR-1246) and Venezuelan equine encephalitis virus (ATCC VR-923; ATCC VR-1250; ATCC VR 1249; ATCC VR-532), and adeno-associated virus (AAV) vectors (see, for example, PCT Publications WO94/12649, WO93/03769; WO93/19191; WO94/28938; WO95/11984 and WO95/00655). Alternatively, administration of DNA linked to a killed adenovirus may be used, as described in Curiel, Hum. Gene Ther., 3:147, 1992.

Non-viral delivery media and methods may also be used, including, but not limited to, polycationically condensed DNA linked to or unlinked by a single, killed adenovirus (see, for example, Curiel, Hum. Gene Ther., 3:147, 1992); ligand-linked DNA (see, for example, Wu, J. Biol. Chem., 264:16985, 1989); eukaryotic cell delivery media (see, for example, U.S. Patent No. 5,814,482; PCT Publications WO95/07994; WO96/17072; WO95/30763; and WO97/42338) and nucleic acid charge neutralization or fusion with the cell membrane. Naked DNA may also be used. Exemplary methods for introducing naked DNA are described in PCT Publication WO90/11092 and U.S. Patent No. 5,580,859. Liposomes that can act as gene delivery mediators are described in U.S. Patent No. 5,422,120; PCT Publications Nos. WO95/13796, WO94/23697, WO91/14445; and European Publication EP 0524968. Additional methods are described in Philip, Mol. Cell Biol., 14:2411, 1994 and Woffendin, Proc. Natl. Acad. Sci., 91:1581, 1994.

In some embodiments, the invention covers compositions including pharmaceutical compositions comprising antibodies of the invention as described herein or prepared by the methods described herein and having the characteristics described herein. As used herein, pharmaceutical compositions may comprise one or more antibodies binding to tumor antigens, one or more bispecific antibodies binding to CD3 and tumor antigens, and/or one or more polynucleotides encoding sequences of one or more of these antibodies. These compositions may further comprise suitable excipients, such as pharmaceutically acceptable excipients, including buffers, which are well known in the art.

In one embodiment, the invention is characterized by a composition that is a combination of a CD3 antibody and a second therapeutic agent. In one embodiment, the second therapeutic agent is any agent advantageously combined with a CD3 antibody. Exemplary agents that can be advantageously combined with a CD3 antibody include, but are not limited to, other agents that bind and/or activate CD3 signaling (including other antibodies or their antigen-binding fragments, etc.) and/or agents that do not directly bind CD3 but still activate or stimulate immune cell activation. The invention further includes additional combination therapies and co-formulations involving the CD3 antibodies of the invention. The invention also provides methods for preparing any of these antibodies. The antibodies of the invention can be prepared by procedures known in the art, including de novo protein synthesis and recombinant expression of nucleic acids encoding binding proteins. The desired nucleic acid sequence can be generated by recombinant methods (e.g., PCR mutagenesis of previously prepared variants of the desired polynucleotide) or by solid-phase DNA synthesis. Recombinant expression methods are typically used. In one aspect, the invention provides polynucleotides comprising sequences encoding CD3VH and/or VL. Due to the degeneracy of the genetic code, multiple nucleic acid sequences encode each immunoglobulin amino acid sequence, and the invention includes all nucleic acids encoding the binding proteins described herein.

Chimeric or hybrid antibodies can also be prepared in vitro using known methods of synthetic protein chemistry, including those involving cross-linking agents. For example, immunotoxins can be constructed using disulfide exchange reactions or by forming thioether bonds. Examples of suitable reagents for this purpose include iminothiolates and methyl-4-mercaptobutyrimidate.

In recombinant humanized antibodies, the Fcγ region can be modified to avoid interactions with the Fcγ receptor, as well as with complement and the immune system. Techniques for preparing such antibodies are described in PCT Publication WO99/58572. For example, if the antibody is to be used in clinical trials and treatments in humans, the constant region can be engineered to more closely resemble the human constant region to avoid an immune response. See, for example, U.S. Patent Nos. 5,997,867 and 5,866,692.

Compared to the corresponding germline sequences from which antibodies are derived, CD3 antibodies disclosed herein may contain one or more amino acid substitutions, insertions, and/or deletions in the framework and/or CDR regions of the heavy and light chain variable domains. Such mutations can be readily identified by comparing the amino acid sequences disclosed herein with germline sequences available from, for example, public antibody sequence databases. This invention includes antibodies and antigen-binding fragments thereof derived from any of the amino acid sequences disclosed herein, wherein one or more amino acids in one or more framework and/or CDR regions are mutated to one or more corresponding residues of the germline sequence from which the antibody is derived, or one or more corresponding residues of another human germline sequence, or conserved amino acid substitutions of one or more corresponding germline residues (such sequence changes are collectively referred to herein as “germline mutations”). Starting with the heavy and light chain variable region sequences disclosed herein, those skilled in the art can readily generate a variety of antibody and antigen-binding fragments containing one or more individual germline mutations or combinations thereof. In some embodiments, all framework and/or CDR residues within the VH and/or VL domains are mutated back to residues found in the original germline sequence from which the antibody is derived. In some embodiments, only certain residues are mutated back to the original germline sequence, for example, mutated residues found only in the first 8 amino acids of FR1 or the last 8 amino acids of FR4, or mutated residues found only in CDR1, CDR2, or CDR3. In some further embodiments, one or more framework and/or CDR residues are mutated to one or more corresponding residues of a different germline sequence (i.e., a germline sequence from which the antibody was originally derived).

Furthermore, the antibodies of the present invention may contain any combination of two or more germline mutations within the framework and/or CDR region, for example, wherein certain individual residues are mutated to corresponding residues of a specific germline sequence, while certain other residues different from the original germline sequence are maintained or mutated to corresponding residues of a different germline sequence. Once obtained, one or more desired properties of the antibody and antigen-binding fragment containing one or more germline mutations can be readily tested, such as increased binding specificity, increased binding affinity, improved or enhanced antagonistic or agonistic biological properties (as the case may be), reduced immunogenicity, etc. Antibody and antigen-binding fragments obtained in this general manner are included within the scope of the present invention.

The present invention also includes CD3 antibodies comprising variants of any of the HC VR, LC VR, and/or CDR amino acid sequences disclosed herein having one or more conserved substitutions. For example, the present invention includes CD3 antibodies having HC VR, LC VR, and/or CDR amino acid sequences having, for example, 10 or fewer, 8 or fewer, 6 or fewer, 4 or fewer, etc., conserved amino acid substitutions relative to any of the HC VR, LC VR, and/or CDR amino acid sequences, as disclosed herein.

Therefore, this invention covers modifications of the antibodies and peptides of this invention, variants as described herein, including functionally equivalent antibodies that do not significantly affect their properties, and variants with enhanced or reduced activity and/or affinity. For example, amino acid sequences can be mutated to obtain antibodies with desired binding affinity to tumor antigens and/or CD3. Peptide modification is a common practice in the art and need not be described in detail herein. Examples of modified peptides include peptides with conserved substitutions of amino acid residues, deletions or additions of one or more amino acids that do not significantly and harmfully alter the functional activity or affinity of the mature (enhancing) peptide for its ligands, or the use of chemical analogs.

Amino acid sequence insertions include fusion of the amino and/or carboxyl ends of peptides ranging in length from one residue to one hundred or more residues, as well as intra-sequence insertions of single or multiple amino acid residues. Examples of terminal insertions include antibodies with an N-terminal methionine residue or antibodies fused to an epitope tag. Other insertion variants of antibody molecules include fusion of an enzyme or a peptide that increases the antibody's half-life in the bloodstream with the N- or C-terminus of the antibody.

Substitution variants involve the removal of at least one amino acid residue from the antibody molecule and the insertion of a different residue at its position. Sites of most interest for substitution mutagenesis include hypervariable regions, but FR alterations are also expected. Conservative substitutions are shown under the heading “Conservative Substitutions” in Table 6. If such substitution results in an alteration of biological activity, more significant changes, designated “Exemplary Substitutions” in Table 6 or further described below regarding amino acid categories, can be introduced, and products screened.

Table 6

Significant modifications to the biological properties of antibodies are accomplished through selective substitution, which differs markedly in maintaining: (a) the polypeptide backbone structure in the substituted region, for example, as a sheet or helical conformation; (b) the charge or hydrophobicity of the molecule at the target site; or (c) the volume of the side chains. Naturally occurring amino acid residues are grouped based on common side chain properties:

(1) Nonpolar: Leucine, Met, Ala, Val, Leu, Ile;

(2) Neutral elements with no charge: Cys, Ser, Thr, Asn, Gln;

(3) Acidic (negatively charged): Asp, Glu;

(4) Alkaline (positively charged): Lys, Arg;

(5) Residues affecting chain orientation: Gly, Pro; and

(6) Aroma: Trp, Tyr, Phe, His.

Non-conservative substitutions are prepared by replacing members of one of these categories with members of another category.

Serine residues that are not involved in maintaining the correct conformation of the antibody can generally be replaced with serine to improve the oxidative stability of the molecule and prevent abnormal cross-linking. Conversely, one or more cysteine bonds can be added to the antibody to improve its stability, especially when the antibody is an antibody fragment, such as an Fv fragment.

Amino acid modifications can involve a complete redesign of a region (e.g., a variable region) by altering or modifying one or more amino acids. Changes in the variable region can alter binding affinity and/or specificity. In some embodiments, no more than one to five conserved amino acid substitutions are made within the CDR domain. In some embodiments, no more than one to three conserved amino acid substitutions are made within the CDR domain. In still other embodiments, the CDR domain is VH CDR3 and/or VL CDR3.

Modifications also include glycosylated and non-glycosylated peptides, as well as peptides with other post-translational modifications, such as, for example, glycosylation, acetylation, and phosphorylation with different sugars. Antibodies are glycosylated at conserved sites in their constant regions (Jefferis and Lund, Chem. Immunol. 65:111-128, 1997; Wright and Morrison, TibTECH 15:26-32, 1997). Oligosaccharide side chains of immunoglobulins influence protein function (Boyd et al., Mol. Immunol. 32:1311-1318, 1996; Wittwe and Howard, Biochem. 29:4175-4180, 1990), as well as intramolecular interactions between glycoprotein moieties, which can affect conformation and present the three-dimensional surface of glycoproteins (Jefferis and Lund, above; Wyss and Wagner, Current Opin. Biotech. 7:409-416, 1996). Oligosaccharides can also be used to target given glycoproteins to certain molecules based on specific recognition structures. Glycosylation of antibodies has also been reported to affect antibody-dependent cytotoxicity (ADCC). In particular, CHO cells with tetracycline-regulated β(1,4)-N-acetylglucosamine transferase III (GnTIII) (a glycosyltransferase that catalyzes the formation of aliquots of GlcNAc) expression have been reported to have improved ADCC activity (Umana et al., Mature Biotech. 17:176-180, 1999).

Antibody glycosylation is generally N-linked or O-linked. N-linked glycosylation refers to the attachment of the carbohydrate moiety to the side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine, asparagine-X-threonine, and asparagine-X-cysteine (where X is any amino acid except proline) are recognition sequences for the enzymatic attachment of the carbohydrate moiety to the asparagine side chain. Therefore, the presence of any of these tripeptide sequences in the polypeptide creates a potential glycosylation site. O-linked glycosylation refers to the attachment of one of the sugars N-acetylgalactosamine, galactose, or xylose to a hydroxy amino acid, most commonly serine or threonine, although 5-hydroxyproline or 5-hydroxylysine can also be used.

By altering the amino acid sequence to include one or more of the aforementioned tripeptide sequences (for N-glycosylation sites), the addition of glycosylation sites to the antibody can be conveniently accomplished. Alternatively, alterations can be made by adding or replacing one or more serine or threonine residues to the original antibody sequence (for O-glycosylation sites).

The glycosylation pattern of an antibody can also be altered without changing the underlying nucleotide sequence. Glycosylation is largely dependent on the host cell used to express the antibody. Since the cell types used to express recombinant glycoproteins such as antibodies as potential therapeutic agents are rarely native cells, variations in the antibody glycosylation pattern can be expected (see, for example, Hse et al., J. Biol. Chem. 272:9062-9070, 1997).

Besides host cell selection, factors influencing glycosylation during recombinant antibody production include growth mode, culture medium formulation, culture density, oxygenation, pH, purification protocol, and more. Various methods have been proposed to alter the glycosylation patterns achieved in specific host organisms, including the introduction or overexpression of certain enzymes associated with oligosaccharide production (US Patent Nos. 5,047,335; 5,510,261 and 5,278,299). For example, using endoglycosidase H (Endo H), N-glycosidase F, endoglycosidase F1, endoglycosidase F2, and endoglycosidase F3, glycosylation or certain types of glycosylation can be enzymatically removed from glycoproteins. Additionally, recombinant host cells can be genetically engineered to be defective in processing certain types of polysaccharides. These and similar techniques are well known in the art.

Other modification methods include the use of coupling techniques known in the art, including, but not limited to, enzymatic methods, oxidative substitution, and chelation. Modification can be used, for example, for label attachment for immunoassays. Modified peptides are prepared using procedures established in the art and can be screened using standard assays known in the art, some of which are described below and in the examples.

In some embodiments of the invention, the antibody comprises a modified constant region, such as a constant region that has increased affinity for the human Fcγ receptor and is immunologically inert or partially inert, for example, not triggering complement-mediated cleavage, not stimulating antibody-dependent cell-mediated cytotoxicity (ADCC), or not activating macrophages; or having reduced activity (compared to an unmodified antibody) in any one or more of the following: triggering complement-mediated cleavage, stimulating antibody-dependent cell-mediated cytotoxicity (ADCC), or activating microglia. Different modifications to the constant region can be used to achieve optimal levels and/or combinations of effector functions. See, for example, Morgan et al., Immunology 86:319-324, 1995; Lund et al., J. Immunology 157:4963-9 157:4963-4969, 1996; Idusogie et al., J. Immunology 164:4178-4184, 2000; Tao et al., J. Immunology 143: 2595-2601, 1989; and Jefferis et al., Immunological Reviews 163:59-76, 1998. In some embodiments, the constant region is modified as described in Eur. J. Immunol., 29:2613-2624, 1999; PCT application number PCT/GB99/01441; and/or UK application number 9809951.8. In still other embodiments, the constant region is aglycosylated against N-glycosylation. In some embodiments, the constant region is aglycosylated against N-glycosylation by mutating glycosylated amino acid residues or flanking residues in the constant region that are part of the N-glycosylation recognition sequence. For example, the N-glycosylation site N297 can be mutated to A, Q, K, or H. See Tao et al., J. Immunology 143: 2595-2601, 1989; and Jefferis et al., Immunological Reviews 163:59-76, 1998. In some implementations, the constant region is deglycosylated against N-glycosylation. The constant region may be deglycosylated enzymatically (e.g., by removing carbohydrates via the enzyme peptide-N-glycosidase (PNGase)) or by expression against N-glycosylation in a glycosylation-deficient host cell.

Other antibody modifications include antibodies modified as described in PCT Publication No. WO99/58572. In addition to the binding domain targeting the target molecule, these antibodies also contain an effector domain having an amino acid sequence substantially homologous to all or part of the constant region of the human immunoglobulin heavy chain. These antibodies are capable of binding the target molecule without triggering significant complement-dependent cleavage or cell-mediated target disruption. In some embodiments, the effector domain is capable of specifically binding FcRn and/or FcγRIIb. These are generally based on chimeric domains derived from two or more CH2 domains of the human immunoglobulin heavy chain. Antibodies modified in this manner are particularly suitable for use in long-term antibody therapy to avoid inflammation and other adverse reactions to conventional antibody therapy.

This invention includes embodiments of affinity maturation. For example, affinity-matured antibodies can be generated by procedures known in the art (Marks et al., Bio/Technology, 10:779-783, 1992; Barbas et al., ProcNat. Acad. Sci, USA 91:3809-3813, 1994; Schier et al., Gene, 169:147-155, 1995; Yelton et al., J. Immunol., 155:1994-2004, 1995; Jackson et al., J. Immunol., 154(7):3310-9, 1995; Hawkins et al., J. Mol. Biol., 226:889-896, 1992; and PCT Publication No. WO2004/058184).

The following methods can be used to modulate antibody affinity and characterize CDRs. One way to characterize the CDR of an antibody and/or alter (e.g., improve) the binding affinity of a peptide, such as an antibody, is called “library scan mutagenesis.” Typically, library scan mutagenesis is performed as described below. Using methods recognized in the art, one or more amino acid positions in the CDR are replaced with two or more amino acids (e.g., 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20). This produces small clonal libraries (in some embodiments, one clone for each amino acid position analyzed), each clone having the complexity of two or more members (if two or more amino acids are substituted at each position). Typically, the library also includes clones containing native (unsubstituted) amino acids. For a small number of clones from each library, e.g., approximately 20–80 clones (depending on library complexity), screening is performed for binding affinity to the target peptide (or other binding target), and candidates exhibiting increased binding, identical binding, decreased binding, or no binding are identified. Methods for determining binding affinity are well known in the art. Binding affinity can be determined using BIACORE™ surface plasmon resonance analysis, which detects differences in binding affinity of approximately 2-fold or greater. BIACORE™ is particularly useful when the starting antibody has already bound with relatively high affinity (e.g., approximately 10 nM or lower K _{_D} ). Screening using BIACORE™ surface plasmon resonance is described in the examples herein.

Binding affinity can be determined using Kinexa Biocensor, scintillation affinity assay, ELISA, ORIGEN immunoassay (IGEN), fluorescence quenching, fluorescence transfer, and/or yeast display. Binding affinity can also be screened using appropriate bioassays.

In some embodiments, using art-recognized mutagenesis methods (some of which are described herein), each amino acid position in the CDR is replaced (in some embodiments, one at a time) with all 20 natural amino acids. This produces small clonal libraries (in some embodiments, one clone for each amino acid position analyzed), each clone having a complexity of 20 members (if all 20 amino acids are replaced at each position).

In some implementations, the library to be screened contains substitutions at two or more positions, which may be in the same CDR or in two or more CDRs. Thus, the library may contain substitutions at two or more positions in a single CDR. The library may contain substitutions at two or more positions in two or more CDRs. The library may contain substitutions at three, four, five, or more positions, which are found in two, three, four, five, or six CDRs. Substitutions can be prepared using low-redundancy codons. See, for example, Table 2, Balint et al., Gene 137(1):109-18, 1993.

Candidates exhibiting improved binding can be sequenced to identify CDR substitution mutants that result in improved affinity (also known as "improved" substitutions). Binding candidates can also be sequenced to identify CDR substitutions that maintain binding.

Multiple rounds of screening can be performed. For example, candidates with improved binding (each candidate containing amino acid substitutions at one or more positions of one or more CDRs) can also be used to design a second library containing at least the original and substituted amino acids at each improved CDR position (i.e., the amino acid position in the CDR where the substitution mutant shows improved binding). The preparation and screening or selection of this library are discussed further below.

Library scanning mutagenesis also provides information about the importance of each amino acid position to the stability of the antibody-antigen complex because the frequency of clones exhibiting improved binding, identical binding, reduced binding, or no binding also provides information related to the stability of the antibody-antigen complex. For example, if a CDR position remains bound when changed to all 20 amino acids, that position is identified as unlikely to be required for antigen binding. Conversely, if a CDR position remains bound only with a small percentage of substitutions, that position is identified as important for CDR function. Therefore, library scanning mutagenesis yields information about positions in CDRs that can be changed to many different amino acids (including all 20 amino acids) and positions in CDRs that cannot be changed or can only be changed to a few amino acids.

Candidates with improved affinity can be combined in a second library comprising the improved amino acid, the original amino acid at that position, and may further include additional substitutions at that position, depending on the required complexity of the library or the complexity permitted by the desired screening or selection method. Additionally, if needed, adjacent amino acid positions can be randomized to at least two or more amino acids. Randomization of adjacent amino acids allows for additional conformational flexibility in the mutant CDR, which in turn allows or facilitates the introduction of a large number of improving mutations. The library may also contain substitutions at positions that did not exhibit improved affinity in the first round of screening.

Use any method known in the art to screen or select library members in the second library that have improved and/or altered binding affinity, including screening using BIACORE™ surface plasmon resonance analysis, and selection using any method known in the art for selection, including phage display, yeast display, and ribosome display.

The techniques used for isolating and preparing antibodies are as follows. However, it will be understood that the antibodies can be formed from or incorporated into other polypeptides using techniques known in the art. For example, nucleic acids encoding a target polypeptide (e.g., a ligand, receptor, or enzyme) can be isolated from a cDNA library prepared from a tissue believed to have polypeptide mRNA and express it at detectable levels. The library is screened using probes (e.g., antibodies or oligonucleotides of about 20-80 bases) designed to identify the target gene or the protein it encodes. Screening of cDNA or genomic libraries with selected probes can be performed using standard procedures such as those described in Chapters 10-12 of Sambrook et al., Molecular Cloning: A Laboratory Manual (New York: Cold Spring Harbor Laboratory Press, 1989).

CD3 antibodies used to treat autoimmune diseases

In one aspect, a method for treating an autoimmune disease in a subject is provided, comprising administering to the subject in need an effective amount of a composition comprising an antibody as described herein.

As used in this article, autoimmune diseases include, but are not limited to, inflammatory bowel disease, systemic lupus erythematosus, rheumatoid arthritis, diabetes mellitus (type I), multiple sclerosis, Addison’s disease, celiac disease, dermatomyositis, Graves’ disease, Hashimoto’s thyroiditis, Hashimoto’s encephalopathy, myasthenia gravis, reactive arthritis, Sjogren’s syndrome, atopic allergy, atopic dermatitis, autoimmune enteropathy, autoimmune hepatitis, autoimmune lymphoproliferative syndrome, autoimmune peripheral neuropathy, autoimmune pancreatitis, autoimmune polyendocrine syndrome, and autoimmune progesterone dermatitis. Autoimmune urticaria, autoimmune uveitis, Behcet’s disease, Castleman’s disease, cold agglutinin disease, Crohn’s disease, dermatomyositis, eosinophilic fasciitis, gastrointestinal pemphigoid, Goodpasture’s syndrome, Guillain-Barré syndrome, hidradenitis suppurativa, somnambulism, pemphigus vulgaris, polymyositis, relapsing polychondritis, rheumatic fever, transverse myelitis, ulcerative colitis, undifferentiated connective tissue disease, vasculitis, and Wegener’s granulomatosis.

Multispecific antibodies and their uses

The CD3 antibodies of the present invention may be specific to different epitopes of a single target polypeptide or may contain antigen-binding domains specific to more than one target polypeptide. See, for example, Tutt et al., 1991, J. Immunol. 147:60-69; Kufer et al., 2004, Trends Biotechnol. 22:238-244. The CD3 antibodies of the present invention may be ligated to or co-expressed with another functional molecule (e.g., another peptide or protein). For example, the antibody or a fragment thereof may be functionally linked (e.g., by chemical coupling, genetic fusion, non-covalent association, or other means) to one or more other molecular entities (e.g., another antibody or antibody fragment) to produce a multispecific antibody with a second binding specificity.

In one aspect, the antibody of the present invention is a multispecific antibody. In a specific embodiment, the CD3 antibody is a bispecific antibody that specifically binds to human CD3. In some such embodiments, the bispecific antibody further specifically binds to tumor antigens. In another such embodiment, the bispecific antibody of the present invention simultaneously targets T cells (CD3) and tumor cells, and successfully guides and activates T cells to exert cytotoxicity on tumor cells expressing tumor antigens.

In the respective further embodiments described above, the CD3-binding bispecific antibody is characterized by any one or more of the following properties: (a) treating, preventing, or improving one or more symptoms of a condition in a subject associated with malignant cells expressing a specific tumor antigen (e.g., B-cell-related cancers, such as multiple myeloma); (b) inhibiting tumor growth or progression in a subject (who has a malignant tumor expressing a specific tumor antigen); (c) inhibiting metastasis of cancer (malignant) cells expressing a specific tumor antigen in a subject (who has one or more malignant cells expressing a tumor antigen); (d) inducing regression (e.g., long-term regression) of tumors expressing tumor antigens; (e) exerting cytotoxic activity in malignant cells expressing tumor antigens; (f) prolonging progression-free survival in subjects with tumor-related conditions; (g) prolonging overall survival in subjects with tumor-related conditions; (h) reducing the use of additional chemotherapy or cytotoxic agents in subjects with tumor-related conditions; (i) reducing tumor burden in subjects with tumor-related conditions; or (j) blocking the interaction of tumor antigens with other unidentified factors.

As used herein, a bispecific antibody of the present invention refers to a complex of two or more polypeptide chains, each polypeptide chain comprising at least one antibody VL region and one antibody VH region or a fragment thereof, wherein the VL and VH regions in each polypeptide chain are derived from different antibodies. In a particular aspect, a bispecific antibody comprises a dimer or tetramer of polypeptide chains containing both VL and VH regions. The polypeptide chains constituting the multimeric protein may be covalently linked to at least one other peptide of the multimer via interchain disulfide bonds.

In some such embodiments, the bispecific antibody of the present invention comprises a first heterodimer-promoting domain on a first polypeptide chain and a second heterodimer-promoting domain on a second polypeptide chain (Figure 1). In summary, the first and second heterodimer-promoting domains drive heterodimerization and/or stabilize the bispecific antibody (e.g., through the interaction of pestle and mortar on complementary heterodimer-promoting domains) and/or serve to stabilize the bispecific antibody. In some embodiments, the first and second heterodimer-promoting domains each comprise a CH2 domain and a CH3 domain, wherein the amino acid sequences of each of the CH2 domains and/or the CH3 domains are modified to drive heterodimerization and/or stabilize the bispecific antibody.

In some embodiments, the first heterodimer promoting domain may comprise an Fc chain having CH2 and/or CH3 domains modified to comprise a club (protrusion) or a mortise (cavity). In some such embodiments, the amino acid sequences of the CH2 and/or CH3 domains comprise at least one amino acid modification, wherein: (a) the CH3 domain of the first heterodimer promoting domain forms a club; and (b) the CH3 domain of the second heterodimer promoting domain forms a hole. In another such embodiment, the CH3 domain of the first heterodimer promoting domain comprises the mutants Y349C and/or T366W; and the CH3 domain of the second heterodimer promoting domain comprises the mutants S354C, T366S, L368A, and/or Y407V (according to EU index number).

In one embodiment, if the second heterodimer promoting domain comprises a CH2 and/or CH3 domain modified to comprise a mortise (cavity), then the first heterodimer promoting domain may comprise a CH2 and/or CH3 domain modified to comprise a pestle (protrusion) containing the sequence of SEQ ID NO: 78. In another embodiment, if the second heterodimer promoting domain comprises a CH2 and/or CH3 domain modified to comprise a pestle (protrusion), then the first heterodimer promoting domain may comprise a mortise (cavity) containing the sequence of SEQ ID NO: 79.

Each polypeptide chain of a bispecific antibody contains a VL region and a VH region, which are covalently linked by a glycine-serine linker (linker 1 or linker 2) containing glycine and serine residues, thus confining the antibody-binding domain to prevent self-assembly. Furthermore, each polypeptide chain contains a heterodimerization domain, which promotes heterodimerization and/or stabilization of multiple polypeptide chains and reduces the likelihood of homodimerization between different polypeptide chains. The heterodimerization domain can be located at the N-terminus or C-terminus of the polypeptide chain. The heterodimerization domain can contain a cysteine linker (linker 3) of 1, 2, 3, 4, 5, 6, or more amino acid residues in length. The interaction of two polypeptide chains can produce two VL/VH pairs, thereby forming two epitope-binding domains, i.e., divalent molecules. Neither the VH nor the VL region is confined to any position within the polypeptide chain, i.e., restricted to the amino-terminus or carboxyl-terminus, nor is the region confined by their relative positions to each other; i.e., the VL region can be at the N-terminus of the VH region, and vice versa. The only limitation is the availability of complementary polypeptide chains to form functional bispecific antibodies. In cases where the VL and VH regions originate from antibodies specific to different antigens, the formation of functional bispecific antibodies requires the interaction of two distinct polypeptide chains, i.e., the formation of heterodimers. Conversely, in cases where two distinct polypeptide chains interact freely, for example in a recombinant expression system, one containing VLA and VHB (A being the first epitope and B the second epitope) and the other containing VLB and VHA, two distinct binding sites can be formed: VLA-VHA and VLB-VHB. For all bispecific antibody polypeptide chain pairs, misalignment or misbinding of the two chains is a possibility, for example, interactions between the VL-VL or VH-VH regions. However, the purification of functional bispecific antibodies can be readily controlled based on the immunospecificity of the correctly dimerized binding sites using any affinity-based method known in the art or exemplified herein, such as affinity chromatography.

In one embodiment, the polypeptide chain of the bispecific antibody may contain various linkers and peptides. The linkers and peptides may contain 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 or more amino acids.

In some embodiments, domain 1 is covalently bound to a first heterodimer via a cysteine linker to promote the domain, and domain 2 is covalently bound to a second heterodimer via a cysteine linker to promote the domain. Each cysteine linker includes at least one cysteine residue to allow intramolecular disulfide bonding. In a further embodiment of each of the foregoing embodiments, the cysteine linker (linker 3) comprises at least five amino acids.

In some embodiments, the first polypeptide chain is covalently bound to the second polypeptide chain via at least one disulfide bond. In some such embodiments, at least one disulfide bond is formed between the linker 3 of the first polypeptide chain and the linker 3 of the second polypeptide chain. In another such embodiment, at least one disulfide bond is formed between the first heterodimer promoting domain and the second heterodimer promoting domain. In a specific embodiment, each disulfide bond is formed by linking two cysteine residues. In one aspect of the invention, the bispecific antibody shown in FIG1 comprises a first polypeptide chain and a second polypeptide chain. In some such embodiments, the linker 3 may comprise a truncated lower hinge region of human IgG1 having the sequence SEQ ID NO: 64, which has at least one glycine residue preceding the lower hinge region.

The bispecific antibody of the present invention can simultaneously bind to two separate and distinct epitopes. In some embodiments, at least one epitope binding site is specific for a CD3 determinant expressed on immune effector cells, such as T lymphocytes. In one embodiment, the bispecific antibody molecule binds to an effector cell determinant and also activates the effector cell.

In one aspect, the present invention provides a first polypeptide chain and a second polypeptide chain. In one embodiment, the first polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 36. In another embodiment, the second polypeptide chain comprises the amino acid sequence shown in one or more of SEQ ID NO: 37 or 38 (Table 7). In a preferred embodiment, the first polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 36; and the second polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 37 or 38. In one embodiment, the second polypeptide chain is any polypeptide to be associated with the first polypeptide chain via an interface. Table 7 shows the sequences of the first and second polypeptide chains of the GUCY2c-H2B4 and GUCY2c-2B5 bispecific antibodies.

Table 7

In specific embodiments, the bispecific antibody of the present invention (a) binds to the extracellular domain of a human tumor antigen; (b) exhibits an extended serum and tumor half-life of 30 minutes to 100 days; and/or (c) exhibits a lower _EC50 value of 0.0001 nM to 100 nM in the presence of increased tumor expression levels or increased receptor density levels.

In one implementation, the epitope-binding domain is capable of binding to tumor-associated antigens associated with breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer (including but not limited to non-small cell lung cancer and small cell lung cancer), bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and digestive system cancers. Digestive system cancers include, but are not limited to, cancers of the esophagus, stomach, small intestine, colon, rectum, anus, liver, gallbladder, appendix, bile ducts, and pancreas. In a specific implementation, the therapy activates a cytolytic T-cell response.

Null effector mutations in human IgG1 CH2-CH3

Using standard primer-guided PCR mutagenesis, the Fc chain of human IgG1 was modified to introduce mutants L234A, L235A, and G237A (SEQ ID NO: 82, according to EU index number) to oblate effector functions attributed to binding to FcγRIII, thereby providing an effector-ineffective phenotype (Canfield et al., J. Exp. Med (1991) 173: 1483-1491; Shields et al., J. Biol. Chem. (2001) 276: 6591-604).

The mortise in human IgG1 CH2-CH3 is a mutation.

The 'mortar' configuration is a known efficient design strategy in the art for engineering antibody heavy chain homodimers for heterodimerization. In this method, 'mortar' variants are obtained by replacing smaller amino acids, such as Y349C and T366W (according to EU index numbers), with larger amino acids in one strand of the Fc chain of IgG1. The 'mortar' is designed to insert into the 'mortar' of the CH3 domain of the complementary chain of the Fc chain generated by replacing larger residues, such as S354C, T366S, L368A, and Y407V (according to EU index numbers), with smaller residues.

In some implementations, complementary mutations are introduced to achieve heterodimerization of the resulting Fc chains, such that each Fc chain carries a set of mutations, Y349C and T366W (SEQ ID NO: 78) for the club (or protrusion) Fc chain, or S354C, T366S, L368A, and Y407V (SEQ ID NO: 79) for the mortis (or cavity) Fc chain, as provided in Table 8. When co-transfected into a suitable mammalian host, the DNA encoding the amino acid sequences (e.g., SEQ ID NO: 78 and 79) generates Fc domains predominantly with a bispecific antibody associated with one club (or protrusion) Fc chain of one mortis (or cavity) Fc chain.

Table 8

The CD3-tumor antigen bispecific antibody of this invention is a thermostability-stable, efficient bispecific antibody-Fc fusion targeting both human CD3 and tumor antigens. The Fc domain in the mortise allows for improved expression and purification in CHO cells, resulting in the desired high purity of the heterodimer. The engineered mutation within the Fc domain cancels FcγR binding, thus potentially avoiding ADCC-mediated T cell exhaustion. Furthermore, as shown by differential scanning calorimetry (DSC), incorporating the Fc domain into the bispecific antibody enhances the molecule's stability.

If a bispecific antibody can be engineered to include at least one cysteine residue, it can interact with a corresponding cysteine residue on another polypeptide chain of the invention to form an interchain disulfide bond. The interchain disulfide bond can be used to stabilize the bispecific antibody, thereby improving expression and recovery in recombinant systems, resulting in stable and consistent formulations, and improved stability of isolated and/or purified products in vivo. One or more cysteine residues can be introduced into any part of the polypeptide chain as a single amino acid or as part of a larger amino acid sequence, such as a hinge region. In a specific aspect, at least one cysteine residue is engineered to appear at the C-terminus of the polypeptide chain.

This invention covers methods and compositions for treating, preventing, or managing cancer in a subject, including administering to the subject a therapeutically effective amount of an antibody engineered according to the invention, the molecule further binding to a cancer antigen. The antibodies of this invention are particularly useful for preventing, inhibiting, reducing the growth and/or regression of primary tumors, and for the metastasis of cancer cells. Although not intended to be limited to a specific mechanism of action, the antibodies of this invention can mediate effector functions that can lead to tumor clearance, tumor reduction, or a combination thereof.

In one aspect, the present invention provides a therapeutic method for stimulating T cell activation using the CD3 antibody of the present invention, wherein the therapeutic method comprises administering a therapeutically effective amount of a pharmaceutical composition comprising the antibody of the present invention to a subject in need. The treated condition is any disease or condition that is improved, alleviated, suppressed, or prevented by stimulating CD3 activity or signaling. In a specific embodiment, the present invention provides a bispecific antigen-binding molecule, such as a bispecific antibody that binds to both CD3 and a target antigen.

In one aspect, the present invention provides a method for inhibiting tumor growth, comprising contacting the tumor with an effective amount of an antibody that binds to CD3, said antibody comprising each of the antibodies described herein.

In another aspect, the present invention provides a method for inhibiting tumor growth in a subject, comprising administering to the subject a therapeutically effective amount of an antibody binding to CD3, including each of those antibodies described herein.

In another aspect, the present invention provides a method for modulating angiogenesis in a subject, comprising administering to the subject a therapeutically effective amount of an antibody binding to CD3, including each of those antibodies described herein.

In another aspect, the present invention provides a method for reducing the tumorigenicity of tumors in a subject, comprising administering to the subject a therapeutically effective amount of an antibody binding to CD3, including each of those antibodies described herein.

In one aspect, the present invention provides a method for treating conditions in subjects associated with tumor antigen expression. Therefore, the present invention provides a method for treating tumor antigen-related conditions by engineering bispecific antibodies to immune-specifically recognize tumor antigens and CD3 antigens on T cells. Bispecific antibodies engineered according to the present invention can be used for the prevention or treatment of cancer because they possess cytotoxic activity via CD3 antibody-induced activation of cytotoxic T cells.

In one specific embodiment, the present invention provides a method for treating cancer. In this embodiment, the cancer is breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer (including but not limited to non-small cell lung cancer and small cell lung cancer), bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, or cancer of the digestive system. In some embodiments, the cancer of the digestive system is selected from cancers of the esophagus, stomach, small intestine, colon, rectum, anus, liver, gallbladder, appendix, bile ducts, and pancreas.

In one aspect, the present invention provides antibodies, bispecific antibodies, or pharmaceutical compositions as disclosed herein for use in a therapy. In one specific embodiment, the present invention also provides a CD3 bispecific antibody for use in a method of treating cancer as defined herein. In another specific embodiment, the therapy activates a cytolytic T-cell response.

This invention further provides antibodies or bispecific antibodies as disclosed herein for the preparation of medicaments for use in therapies. In some embodiments, the therapy is a treatment for cancer. In some embodiments, the cancer is selected from breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer (including but not limited to non-small cell lung cancer and small cell lung cancer), bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, or cancers of the digestive system. In some embodiments, the cancers of the digestive system are selected from cancers of the esophagus, stomach, small intestine, colon, rectum, anus, liver, gallbladder, appendix, bile ducts, and pancreas.

In one aspect, the present invention provides polynucleotides encoding antibodies or bispecific antibodies as disclosed herein. In another embodiment, the present invention provides a vector comprising a polynucleotide as disclosed herein. In yet another embodiment, the present invention provides a host cell comprising a vector as disclosed herein. In some such embodiments, the host cell recombinants produce antibodies or bispecific antibodies as disclosed herein. In specific embodiments, the host cell is selected from bacterial cell lines, mammalian cell lines, insect cell lines, yeast cell lines, and in vitro cell-free protein synthesis systems. In a specific embodiment, the mammalian cell line is a CHO cell line.

In one aspect, a method for treating cancer in a subject with this need is provided, comprising: a) providing a bispecific antibody as described herein, and b) administering the bispecific antibody to the subject.

In some embodiments, a method is provided for inhibiting tumor growth or progression in a subject with malignant cells expressing tumor antigens, comprising administering to a subject in need an effective amount of a composition comprising antibodies as described herein. In some embodiments, a method is provided for inhibiting cell metastasis expressing tumor antigens in a subject, comprising administering to a subject in need an effective amount of a composition comprising antibodies as described herein. In some embodiments, a method is provided for inducing tumor regression in malignant cells in a subject, comprising administering to a subject in need an effective amount of a composition comprising antibodies as described herein.

In a specific aspect, relative to the growth of cancer cells in the absence of the antibodies or bispecific antibodies of the present invention, the antibodies of the present invention inhibit or reduce the growth of cancer cells by at least 99%, at least 95%, at least 90%, at least 85%, at least 80%, at least 75%, at least 70%, at least 60%, at least 50%, at least 45%, at least 40%, at least 45%, at least 35%, at least 30%, at least 25%, at least 20%, or at least 10%.

In specific respects, the antibody kills cells or inhibits or reduces the growth of cancer cells by at least 5%, at least 10%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, or at least 100% better than the antibody or bispecific antibody of the present invention.

In one aspect, the present invention provides an effective amount of a composition (e.g., a pharmaceutical composition) comprising a bispecific antibody as described herein for treating a condition (e.g., cancer) associated with tumor antigen expression in a subject in need of this treatment.

In another aspect, the present invention provides antibodies as described herein for treating conditions (e.g., cancer) associated with tumor antigen expression in subjects with this need. In some embodiments, antibodies as described herein are provided for inhibiting tumor growth or progression in subjects with malignant cells expressing tumor antigens. In some embodiments, antibodies as described herein are provided for inhibiting the metastasis of malignant cells expressing tumor antigens in subjects with this need. In some embodiments, antibodies as described herein are provided for inducing tumor regression in subjects with malignant cells expressing tumor antigens.

In another aspect, the present invention provides the use of antibodies as described herein in the preparation of medicaments for treating conditions (e.g., cancer) associated with tumor antigen expression. In some embodiments, the use of antibodies as described herein is provided in the preparation of medicaments for inhibiting tumor growth or progression. In some embodiments, the use of antibodies as described herein is provided in the preparation of medicaments for inhibiting the metastasis of malignant cells expressing tumor antigens. In some embodiments, the use of antibodies as described herein is provided in the preparation of medicaments for inducing tumor regression.

In some embodiments, the method described herein further includes the step of treating the subject with another form of therapy. In some embodiments, the other form of therapy is an additional anticancer therapy, including, but not limited to, chemotherapy, radiation, surgery, hormone therapy, and/or additional immunotherapy.

This invention further encompasses the administration of the molecules of the invention in combination with other therapies known to those skilled in the art for the treatment or prevention of cancer, including but not limited to current standard and experimental chemotherapy, biotherapy, immunotherapy, radiotherapy, or surgery. In some aspects, the molecules of the invention may be administered in combination with therapeutically or preventively effective amounts of one or more agents, therapeutic antibodies, or other agents known to those skilled in the art for the treatment and/or prevention of cancer.

Therefore, methods for treating cancer include administering an effective amount of the multispecific antibody (e.g., a bispecific antibody) of the present invention in combination with a chemotherapy agent to a subject in need. This combination therapy can be administered alone, sequentially, or simultaneously.

The dosages and frequencies of administration provided herein are covered by the terms therapeutically effective and prophylactically effective. Depending on the specific therapeutic or prophylactic agent administered, the severity and type of cancer, the route of administration, and the subject's age, weight, response, and past medical history, dosage and frequency may further vary based on factors specific to each subject. Those skilled in the art can select an appropriate regimen by taking such factors into account and by following the dosages recommended, for example, those reported in the literature and in Physician's Desk Reference (56th edition, 2002).

Therefore, in addition to redirecting T cells to tumor-specific antigens, bispecific T-cell conjugate molecules can also be used to deliver other diagnostic or therapeutic compounds to cells expressing tumors on their surface. Thus, bispecific T-cell conjugate molecules can be directly or indirectly (e.g., via a connector) linked to drugs, enabling direct delivery to tumor-bearing cells. Therapeutic agents include compounds such as nucleic acids, proteins, peptides, amino acids or derivatives, glycoproteins, radioisotopes, lipids, carbohydrates, or recombinant viruses. Therapeutic and diagnostic nucleic acid portions include antisense nucleic acids, derivatized oligonucleotides for covalent cross-linking with single- or double-stranded DNA, and oligonucleotides forming triple strands.

Pharmaceutical Composition

The present invention further provides compositions comprising a therapeutically effective amount of the antibody disclosed herein and a pharmaceutically acceptable carrier.

The antibodies of the present invention may be in the form of pharmaceutical compositions for administration, formulated to suit a chosen mode of administration, and containing pharmaceutically acceptable diluents or excipients, such as buffers, surfactants, preservatives, solubilizers, isotonic agents, stabilizers, carriers, etc. Remington's Pharmaceutical Sciences, Mack Publishing Co., Easton Pa., 18th edition, 1995 provides an outline of formulation techniques generally known to those skilled in the art.

These pharmaceutical compositions can be administered in any manner known in the art to achieve the generally intended purpose of treating cancer. Routes of administration can be parenteral, and are defined herein as modes of administration including, but not limited to, intravenous, intramuscular, intraperitoneal, subcutaneous, and intra-articular injections and infusions. The manner of administration and dosage of the molecules according to the invention (e.g., antibodies, pharmaceutical compositions) depends on the type of disease to be combated, its stage in appropriate circumstances, the antigen to be controlled, the type of coexisting treatment (if any), the frequency of treatment, the characteristics of the desired effect, and also the subject's weight, age, health status, diet, and sex. Therefore, the actual dosage regimens used can vary considerably and may therefore deviate from the dosage regimens stated herein.

Various formulations of the antibodies of this invention can be used for administration. In some embodiments, the antibodies can be administered in their pure form. In some embodiments, the antibodies and pharmaceutically acceptable excipients can be contained in various formulations. Pharmaceutically acceptable excipients are known in the art and are relatively inert substances that facilitate the administration of pharmacologically effective substances. For example, excipients can impart shape or consistency or act as diluents. Suitable excipients include, but are not limited to, stabilizers, wetting agents and emulsifiers, salts for altering molar osmotic pressure concentrations, encapsulation agents, buffers, and skin penetration enhancers. Excipients, as well as formulations for parenteral and non-parenteral drug delivery, are described in Remington, The Science and Practice of Pharmacy, 21st edition. Mack Publishing, 2005.

In some implementations, these agents are formulated for administration by injection (e.g., intraperitoneal, intravenous, subcutaneous, intramuscular, etc.). Therefore, these agents can be combined with pharmaceutically acceptable mediators, such as saline, Ringer's solution, glucose solution, etc. Specific dosing regimens, i.e., dosage, timing, and repetition, will depend on the specific individual and their medical history.

The antibodies described herein can be administered using any suitable method, including by injection (e.g., intraperitoneal, intravenous, subcutaneous, intramuscular, etc.). As described herein, antibodies, such as monoclonal or bispecific antibodies, can also be administered by inhalation. Generally, the dosage for administration of the antibodies of the present invention depends on the host being treated and the specific administration method. In one embodiment, the dosage range of the antibodies of the present invention will be from about 0.001 µg/kg body weight to about 20,000 µg/kg body weight. The term “body weight” may be used when treating a patient. When processing isolated cells, “body weight” as used herein refers to “total cell weight”. The term “total weight” can be used for both isolated cells and patient treatment. In this application, all concentrations and treatment levels are expressed as “body weight” or simply “kg”, and are also considered to include similar “total cell weight” and “total weight” concentrations. However, those skilled in the art will recognize the practicality of a variety of dosage ranges, such as 0.01 µg/kg body weight to 20,000 µg/kg body weight, 0.02 µg/kg body weight to 15,000 µg/kg body weight, 0.03 µg/kg body weight to 10,000 µg/kg body weight, 0.04 µg/kg body weight to 5,000 µg/kg body weight, 0.05 µg/kg body weight to 2,500 µg/kg body weight, 0.06 µg/kg body weight to 1,000 µg/kg body weight, 0.07 µg/kg body weight to 500 µg/kg body weight, 0.08 µg/kg body weight to 400 µg/kg body weight, 0.09 µg/kg body weight to 200 µg/kg body weight, or 0.1 µg/kg body weight to 100 µg/kg body weight. Furthermore, those skilled in the art will recognize that a variety of different dosage levels will be useful, for example, 0.0001 µg/kg, 0.0002 µg/kg, 0.0003 µg/kg, 0.0004 µg/kg, 0.005 µg/kg, 0.0007 µg/kg, 0.001 µg/kg, 0.1 µg/kg, 1.0 µg/kg, 1.5 µg/kg, 2.0 µg/kg, 5.0 µg/kg, 10.0 µg/kg, 15.0 µg/kg, 30.0 µg/kg, 50 µg/kg, 75 µg/kg, 80 µg/kg, 90 µg/kg, 100 µg/kg, 120 µg/kg, 140 µg/kg. g/kg, 150 µg/kg, 160 µg/kg, 180 µg/kg, 200 µg/kg, 225 µg/kg, 250 µg/kg, 2 75µg/kg, 300 µg/kg, 325 µg/kg, 350 µg/kg, 375µg/kg, 400 µg/kg, 450 µg/k g, 500 µg/kg, 550 µg/kg, 600 µg/kg, 700 µg/kg, 750 µg/kg, 800 µg/kg, 900 µg/kg, 1 µg/kg, 5 µg/kg, 10 µg/kg, 12 µg/kg, 15 mg/kg, 20 mg/kg and/or 30 mg/kg. All these dosages are exemplary, and any dosage between these points is also expected to be useful in this invention. Any of the above dosage ranges or dosage levels can be used with the antibodies of this invention. For repeated administration over a period of several days or longer, depending on the condition, treatment may continue until the desired symptom suppression is achieved or until a sufficient therapeutic level is reached, for example, to inhibit or delay tumor growth/progression or metastasis of cancer cells.

Depending on the desired pattern of pharmacokinetic decay, other dosing regimens may also be useful. In one embodiment, the antibody of the present invention is administered at an initial initiating dose followed by higher and/or continuous, substantially constant doses. In some embodiments, dosing is anticipated from once weekly to four times weekly. In some embodiments, dosing is anticipated from once monthly, every other month, or every three months. Progression of the therapy can be easily monitored using routine techniques and assays. The dosing regimen may vary over time.

For the purposes of this invention, the appropriate dosage of the antibody will depend on the antibody or a combination thereof used, the type and severity of the symptoms to be treated, whether the agent is administered for therapeutic purposes, prior therapy, the patient's clinical history and response to the agent, the patient's clearance rate of the administered agent, and the physician's judgment. Typically, the physician will administer the antibody until a dosage is reached to achieve the desired outcome. The dosage and/or frequency may vary during the treatment process. Empirical considerations such as half-life often aid in dosage determination. For example, antibodies compatible with the human immune system, such as humanized antibodies or fully human antibodies, can be used to prolong the antibody's half-life and prevent the antibody from being attacked by the host immune system. The frequency of administration can be determined and adjusted during the treatment process and is generally, but not necessarily, based on the treatment and/or suppression and/or improvement and/or delay of symptoms, such as tumor growth inhibition or delay. Alternatively, a sustained-release formulation of the antibody may be suitable. Various formulations and devices for achieving sustained release are known in the art.

In one implementation, the antibody dosage can be empirically determined in individuals who have received one or more doses of the antibody. Incremental doses of the antibody are administered to the individuals. Indicators of the disease can be tracked to assess efficacy.

In some implementations, the administration of the antibody results in at least one effect selected from the following: inhibition of tumor growth, tumor regression, reduction of tumor size, reduction of tumor cell number, delay of tumor growth, abscopal effect, inhibition of tumor metastasis, reduction of metastatic lesions over time, reduction of use of chemotherapy agents or cytotoxic agents, reduction of tumor burden, increase of progression-free survival, increase of overall survival, complete response, partial response, and stable disease.

Depending on factors such as the recipient's physiological condition, whether the purpose of administration is therapeutic or prophylactic, and other factors known to a skilled professional, the administration of antibodies according to the method of the invention can be continuous or intermittent. Antibody administration can be substantially continuous over a preselected time period, or it can be a series of doses spaced apart from each other.

In some implementations, more than one antibody may be present. At least one, at least two, at least three, at least four, at least five, or more different antibodies may be present. Typically, these antibodies may have complementary activities that do not harmfully interfere with each other. For example, one or more of the following antibodies may be used: a first CD3 antibody targeting one epitope on CD3 and a second CD3 antibody targeting a different epitope on CD3.

In some embodiments, the antibodies disclosed herein can be used, in particular, for parenteral administration, such as intravenous administration or administration into the cavities of body cavities or organs.

Therapeutic formulations of the antibodies used according to the invention are prepared for storage by mixing antibodies of desired purity with optional pharmaceutically acceptable carriers, excipients, or stabilizers (Remington, The Science and Practice of Pharmacy, 21st edition. Mack Publishing, 2005) in the form of lyophilized formulations or aqueous solutions. Acceptable carriers, excipients, or stabilizers are non-toxic to the recipient at the doses and concentrations used and may include buffers such as phosphates, citrates, and other organic acids; salts such as sodium chloride; antioxidants, including ascorbic acid and methionine; preservatives (e.g., octadecyl dimethyl benzyl ammonium chloride; hexane diammonium chloride; benzalkonium chloride, benzyl chloride; phenols, butanol, or benzyl alcohol; alkyl esters of p-hydroxybenzoate, such as methylparaben or propylparaben; catechol; resorcinol; cyclohexanol; 3-pentanol; and m-cresol); low molecular weight (small) The following are considered as salts: polypeptides (approximately 10 residues); proteins, such as serum albumin, gelatin, or immunoglobulins; hydrophilic polymers, such as polyvinylpyrrolidone; amino acids, such as glycine, glutamine, asparagine, histidine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates, including glucose, mannose, or dextrin; chelating agents, such as EDTA; sugars, such as sucrose, mannitol, trehalose, or sorbitol; salt-forming ions, such as sodium; metal complexes (e.g., Zn-protein complexes); and/or nonionic surfactants, such as TWEEN ^™ , PLURONICS ^™ , or polyethylene glycol (PEG).

Antibody-containing liposomes are prepared by methods known in the art, for example, as described in Epstein et al., Proc. Natl. Acad. Sci. USA 82:3688, 1985; Hwang et al., Proc. Natl. Acad. Sci. USA 77:4030, 1980; and U.S. Patent Nos. 4,485,045 and 4,544,545. Liposomes with extended cycle times are disclosed in U.S. Patent No. 5,013,556. Particularly useful liposomes can be produced by a reverse-phase evaporation method using a lipid composition comprising phosphatidylcholine, cholesterol, and PEG-derived phosphatidylethanolamine (PEG-PE). The liposomes are extruded through a filter having a defined pore size to produce liposomes with a desired diameter.

The active ingredient can also be encapsulated in microcapsules (e.g., hydroxymethyl cellulose or gelatin microcapsules and poly-(methyl methacrylate) microcapsules) prepared by agglomeration technology or interfacial polymerization, colloidal drug delivery systems (e.g., liposomes, albumin microspheres, microemulsions, nanoparticles and nanocapsules) or coarse droplet emulsions. This technology is disclosed in Remington, The Science and Practice of Pharmacy, 21st edition, Mack Publishing, 2005.

Sustained-release formulations can be prepared. Suitable examples of sustained-release formulations include semi-permeable matrices of solid hydrophobic polymers containing antibodies, said matrices being in the form of shaped articles, such as membranes or microcapsules. Examples of sustained-release matrices include polyesters, hydrogels (e.g., poly(2-hydroxyethyl methacrylate) or poly(vinyl alcohol)), polylactides (US Patent No. 3,773,919), copolymers of L-glutamic acid and L-glutamic acid 7-ethyl ester, non-degradable ethylene vinyl acetate, degradable lactic acid-glycolic acid copolymers, such as LUPRON DEPOT ^™ (injectable microspheres composed of lactic acid-glycolic acid copolymers and leuprolide acetate), sucrose isobutyrate acetate, and poly-D-(-)-3-hydroxybutyrate.

Preparations intended for in vivo administration must be sterile. This can be easily achieved, for example, by filtration through a sterile membrane. Therapeutic antibody (e.g., CD3 antibody) compositions are typically placed in containers with sterile inlets and outlets, such as intravenous solution bags or vials with stoppers that can be punctured by a subcutaneous needle.

The compositions according to the invention may be in unit dosage form, such as tablets, pills, capsules, powders, granules, solutions or suspensions, or suppositories, for oral, parenteral or rectal administration, or administration by inhalation or blowing.

To prepare solid compositions such as tablets, the main active ingredient is mixed with a pharmaceutical carrier (e.g., conventional tableting ingredients such as corn starch, lactose, sucrose, sorbitol, talc, stearic acid, magnesium stearate, dicalcium phosphate, or gum, and other pharmaceutical diluents, such as water) to form a solid preformulation composition containing a homogeneous mixture of the compound of the present invention or a pharmaceutically acceptable, non-toxic salt thereof. When these preformulation compositions are referred to as homogeneous, it means that the active ingredient is uniformly dispersed throughout the composition, making it easy to subdivide the composition into equivalent effective unit dose forms, such as tablets, pills, and capsules. The solid preformulation composition is then subdivided into unit dose forms of the type described above, containing 0.1 to about 500 mg of the active ingredient of the present invention. Tablets or pills of the novel composition can be coated or otherwise compounded to provide a dosage form that provides the advantage of prolonged action. For example, tablets or pills may contain an internal dose component and an external dose component, the latter being in the form of an outer coating on top of the former. These two components can be separated by an enteric layer, which resists breakdown in the stomach and allows the internal components to enter the duodenum intact or with delayed release. A variety of materials can be used for this enteric layer or coating, including many polymeric acids and mixtures of polymeric acids with materials such as shellac, cetyl alcohol, and cellulose acetate.

Suitable surfactants particularly include nonionic agents such as polyoxyethylene sorbitol (e.g., Tween ^™ 20, 40, 60, 80, or 85) and other sorbitols (e.g., Span ^™ 20, 40, 60, 80, or 85). Compositions containing surfactants will conveniently contain 0.05 to 5% surfactant, and may be 0.1 to 2.5%. It will be understood that, where necessary, other ingredients, such as mannitol or other pharmaceutically acceptable mediators, may be added.

The compositions within the scope of this invention include all compositions in which the antibody is present in an amount that effectively achieves the desired medical effect for treating cancer. Although individual needs may differ from one patient to another, the determination of the optimal range of effective amounts of all components is within the capabilities of a clinician with ordinary skills.

Examples of such compositions and how they are formulated have also been described in earlier sections and below. In some embodiments, the composition comprises one or more antibodies. In some embodiments, the antibody is a human antibody, a humanized antibody, or a chimeric antibody. In some embodiments, the antibody comprises a constant region capable of triggering a desired immune response (e.g., antibody-mediated cleavage or ADCC). In some embodiments, the antibody comprises a constant region that does not trigger unwanted or undesirable immune responses (e.g., antibody-mediated cleavage or ADCC).

It should be understood that the composition may contain more than one antibody, such as a mixture of CD3 antibodies that recognize CD3 or different epitopes of CD3 and tumor antigens. Other exemplary compositions contain more than one antibody that recognizes one or more of the same epitopes, or different kinds of antibodies, or different epitopes of CD3 and tumor antigens (e.g., human CD3).

In some embodiments, the antibody may be administered in combination with the administration of one or more additional therapeutic agents. These include, but are not limited to, the administration of biological and/or chemotherapeutic agents, such as, but not limited to, vaccines, CAR-T cell-based therapies, radiotherapy, cytokine therapy, CD3 bispecific antibodies, inhibitors of other immunosuppressive pathways, inhibitors of angiogenesis, T-cell activators, inhibitors of metabolic pathways, mTOR inhibitors, inhibitors of the adenosine pathway, tyrosine kinase inhibitors, including but not limited to Inlyta, ALK inhibitors and sunitinib, BRAF inhibitors, epigenetic modifying factors, IDO1 inhibitors, JAK inhibitors, STAT inhibitors, and cell cycle proteins. Leukocyte-dependent kinase inhibitors, biotherapeutic agents (including but not limited to antibodies against VEGF, VEGFR, EGFR, Her2/neu, other growth factor receptors, CD40, CD-40L, CTLA-4, OX-40, 4-1BB, TIGIT, and ICOS), immunogenic agents (e.g., attenuated cancer cells, tumor antigens, antigen-presenting cells such as dendritic cells bombarded with tumor-derived antigens or nucleic acids, immunostimulatory cytokines (e.g., IL-2, IFNα2, GM-CSF), and cells transfected with genes encoding immunostimulatory cytokines (e.g., but not limited to GM-CSF).

Examples of biotherapeutic agents include therapeutic antibodies, immunomodulators, and therapeutic immune cells.

Therapeutic antibodies may be specific to a variety of different antigens. For example, a therapeutic antibody may target a tumor-associated antigen, such that the binding of the antibody to the antigen promotes the death of cells expressing the antigen. In other instances, a therapeutic antibody may target an antigen on immune cells (e.g., PD-1), such that the binding of the antibody prevents the downregulation of the activity of cells expressing the antigen (and thereby promotes the activity of cells expressing the antigen). In some cases, a therapeutic antibody may act through a variety of different mechanisms (e.g., upon contact with cells expressing the antigen, it may both i) promote the death of cells expressing the antigen and ii) prevent the downregulation of immune cell activity caused by the antigen).

Therapeutic antibodies can target antigens such as those listed below. For some antigens, exemplary antibodies against said antigens are also included below (in parentheses after the antigen). The antigens listed below may also be referred to herein as "target antigens," etc. The target antigens used for therapeutic antibodies in this article include, for example: 4-1BB (e.g., utomilumab); 5T4; A33; α-folate receptor 1 (e.g., mirvetuximab soravtansine); Alk-1; BCMA [e.g., PF-06863135 (see US9969809)]; BTN1A1 (e.g., see WO2018222689); CA-125 (e.g., abagovomab); carbonic anhydrase IX; CCR2; CCR4 (e.g., mogamulizumab); CCR5 (e.g., leronlimab); CCR8; CD3 [e.g., blinatumomab (CD3/ CD19 bispecific), PF-06671008 (CD3/P-cadherin bispecific), PF-06863135 (CD3/BCMA bispecific), PF-07062119 (CD3/GUCY2c bispecific), CD19 (e.g., ibritumomab tiuxetan, MOR208); CD20 (e.g., ibritumomab tiuxetan, obinutuzumab, ofatumumab, rituximab, ublituximab); CD22 (inotuzumab ozogamicin, moxetumomab pasudotox); CD25; CD28; CD30 (e.g., brentuximab vedotin); CD33 (e.g., gemtuzumab ozogamicin); CD38 (e.g., daratumumab, isatuximab); CD40; CD-40L; CD44v6; CD47; CD52 (e.g., alemtuzumab); CD63; CD79 (e.g., polotuzumab vedotin); CD80; CD123; CD276/B7-H3 (e.g., omburtamab); CDH17; CEA; ClhCG; CTLA-4 (e.g., ipilimumab, tremel... imumab), CXCR4; desmosome core protein 4; DLL3 (e.g., rovalpituzumab tesirine); DLL4; E-cadherin; EDA; EDB; EFNA4; EGFR (e.g., cetuximab, depatuxizumab mafodotin, necitumumab, panitumumab); EGFRvIII; endothelial sialic acid protein; EpCAM (e.g., oportuzumab monatox); FAP; fetal acetylcholine receptor; FLT3 (e.g., see WO2018/220584); GD2 (e.g., dinutuximab, 3F8); GD3 ; GITR; GloboH; GM1; GM2; GUCY2C (e.g., PF-07062119); HER2/neu [e.g., margetuximab, pertuzumab, trastuzumab; ado-trastuzumab emtansine, trastuzumab duocarmazine, PF-06804103 (see US8828401)]; HER3; HER4; ICOS; IL-10; ITG-AvB6; LAG-3 (e.g., relatlimab); Lewis-Y; LG; Ly-6; M-CSF [e.g., PD-0360324 (see US8828401)]; HER3; HER4; ICOS; IL-10; ITG-AvB6; LAG-3 (e.g., relatlimab); Lewis-Y; LG; Ly-6; M-CSF [e.g., PD-0360324 (see US8828401)]; S7326414); MCSP; mesothelin; MUC1; MUC2; MUC3; MUC4; MUC5AC; MUC5B; MUC7; MUC16; Notch 1; Notch 3; Nectin-4 (e.g., enfortumab vedotin); OX40 [e.g., PF-04518600 (see US7960515)]; P-cadherein [e.g., PF-06671008 (see WO2016/001810)]; PCDHB2; PD-1 [e.g., BCD-100, camrelizumab, cemiplimab, genolimab] Nivolumab (CBT-501), MEDI0680 (CBT-501), nivolumab, pembrolizumab, RN888 (see WO2016/092419), sintilimab, spartalizumab, STI-A1110, tislelizumab, TSR-042; PD-L1 (e.g., atezolizumab, durvalumab, BMS-936559 (MDX-1105) or LY3300054); PDGFRA (e.g., olaratumab); plasma cell antigen; PolySA; PSCA; PSMA; PTK7 [e.g., PF-06647020 (see US9409995)]; Ror1; SAS; SCRx6; SLAMF7 (e.g., elotuzumab); SHH; SIRPa (e.g., ED9, Effi-DEM); STEAP; TGF-β; TIGIT; TIM-3; TMPRSS3; TNF-α precursor; TROP-2 (e.g., sacituzumabgovitecan); TSPAN8; VEGF (e.g., bevacizumab, brolucizumab); VEGFR1 (e.g., ranibizumab); VEGFR2 (e.g., ramucirumab, ranibizumab); Wue-1.

Therapeutic antibodies can take any suitable form. For example, therapeutic antibodies can take any form described elsewhere herein. In some embodiments, the therapeutic antibody may be a naked antibody. In some embodiments, the therapeutic antibody may be conjugated to a drug/pharmaceutical agent (also known as an "antibody-drug conjugate" (ADC)). In some embodiments, a therapeutic antibody targeting a specific antigen may be incorporated into a multispecific antibody (e.g., a bispecific antibody).

In some embodiments, antibodies against antigens may be conjugated to drugs/pharmaceuticals. The linked antibody-drug molecule is also referred to as an "antibody-drug conjugate" (ADC). The drug/pharmaceutical can be linked to the antibody directly or indirectly via a linker. Most commonly, toxic drugs are linked to antibodies such that the binding of the ADC to the corresponding antigen promotes the killing of cells expressing said antigen. For example, ADCs linked to toxic drugs are particularly useful for targeting tumor-associated antigens to promote the killing of tumor cells expressing tumor-associated antigens. In other embodiments, the agent that can be linked to the antibody may be, for example, an immunomodulator (e.g., to modulate the activity of immune cells near the ADC), an imaging agent (e.g., to facilitate imaging of the ADC in a subject or a biological sample from a subject), or an agent that increases the serum half-life or biological activity of the antibody.

Methods for conjugating cytotoxic agents or other therapeutic agents to antibodies have been described in various publications. For example, antibodies can be chemically modified to induce conjugation by lysine side-chain amines or by cysteine thiol groups activated by reduced interchain disulfide bonds. See, for example, Tanaka et al., FEBS Letters 579:2092-2096, 2005 and Gentle et al., Bioconjugate Chem. 15:658-663, 2004. Reactive cysteine residues engineered at specific sites on antibodies for conjugation to specific drugs at defined stoichiometric amounts have also been described. See, for example, Junutula et al., Nature Biotechnology, 26:925-932, 2008. International applications WO2012/059882 and WO2015015448 also describe the use of glutamine-containing tags with acyl donors or endogenous glutamine conjugated by peptide engineering to an active state (i.e., the ability to form covalent bonds as acyl donors) in the presence of transglutaminase and amines (e.g., cytotoxic agents containing or attached to active amines). In some embodiments, the ADC may have any of the features or characteristics of the ADCs provided in WO2016166629, which is incorporated herein by reference for all purposes.

Drugs/pharmaceuticals that can be linked to antibodies in the form of ADCs can include, for example, cytotoxic agents, immunomodulators, imaging agents, therapeutic proteins, biopolymers, or oligonucleotides.

Exemplary cytotoxic agents that can be incorporated into ADCs include anthracyclines, auristatin, dolastatin, compbretastatin, duocarmycin, pyrrolobenzodiazepine dimer, indoline-benzodiazepine dimer, enediyne, geldemycin, maytansine, puromycin, taxane, vinca alkaloids, camptothecin, tubulysin, hemiasterlin, spliceostatin, pladienolide, and their stereoisomers, isomers, analogs, or derivatives.

Exemplary immunomodulators that can be incorporated into ADCs include gancyclovier, etanercept, tacrolimus, sirolimus, voclosporin, cyclosporine, rapamycin, cyclophosphamide, azathioprine, mycophenolate mofetil, methotrexate, glucocorticoids and their analogues, cytokines, and stem cell production. Growth factors, lymphotoxin, tumor necrosis factor (TNF), hematopoietic factors, interleukins (e.g., interleukin-1 (IL-1), IL-2, IL-3, IL-6, IL-10, IL-12, IL-18, and IL-21), colony-stimulating factors (e.g., granulocyte colony-stimulating factor (G-CSF) and granulocyte-macrophage colony-stimulating factor (GM-CSF)), interferons (e.g., interferon-α, -β, and -γ), stem cell growth factors known as "S1 factors", erythropoietin, and thrombopoietin, or combinations thereof.

Exemplary imaging agents that may be included in an ADC include fluorescein, rhodamine, lanthanide phosphors and their derivatives, or radioisotopes bound to a chelating agent. Examples of fluorophores include, but are not limited to, fluorescein isothiocyanate (FITC) (e.g., 5-FITC), fluorescein phosphoramide (FAM) (e.g., 5-FAM), eosin, carboxyfluorescein, erythrosine, AlexaFluor.RTM. (e.g., Alexa 350, 405, 430, 488, 500, 514, 532, 546, 555, 568, 594, 610, 633, 647, 660, 680, 700, or 750), carboxytetramethylrhodamine (TAMRA) (e.g., 5-TAMRA), tetramethylrhodamine (TMR), and sulforhodamine (SR) (e.g., SR101). Examples of chelating agents include, but are not limited to, 1,4,7,10-tetraazacyclododecane-N,N',N'',N'''-tetraacetic acid (DOTA), 1,4,7-triazacyclononane-1,4,7-triacetic acid (NOTA), 1,4,7-triazacyclononane, 1-glutaric acid-4,7-acetic acid (deferroamine), diethylenetriaminepentaacetic acid (DTPA), and 1,2-bis(o-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA).

Exemplary therapeutic proteins that may be included in ADCs include toxins, hormones, enzymes, and growth factors.

Exemplary biocompatible polymers that can be incorporated into ADCs include water-soluble polymers, such as polyethylene glycol (PEG) or derivatives thereof, and biocompatible polymers containing zwitterions (e.g., polymers containing phosphocholine).

Exemplary biocompatible polymers that can be incorporated into ADCs include antisense oligonucleotides.

In some implementations, the antibodies against antigens provided herein can be incorporated into bispecific antibody molecules. A bispecific antibody is a monoclonal antibody that has binding specificity to at least two different antigens.

In some embodiments, the bispecific antibody comprises a first antibody variable domain and a second antibody variable domain, wherein the first antibody variable domain is capable of recruiting human immune effector cells by specifically binding to CD3 as provided herein, and wherein the second antibody variable domain is capable of specifically binding to a target antigen. In some embodiments, the antibody has an IgG1, IgG2, IgG3, or IgG4 isotype. In some embodiments, the antibody comprises an immune-inert Fc region. In some embodiments, the antibody is a human antibody or a humanized antibody.

Target antigens are typically expressed on target cells (e.g., cancer cells) in disease conditions. Examples of target antigens of particular interest for bispecific antibodies include, but are not limited to, BCMA, EpCAM (epithelial cell adhesion molecule), CCR5 (type 5 chemokine receptor), CD19, HER (human epidermal growth factor receptor)-2/neu, HER-3, HER-4, EGFR (epidermal growth factor receptor), PSMA, CEA, MUC-1 (mucin), MUC2, MUC3, MUC4, MUC5AC, MUC5B, MUC7, CIhCG, Lewis-Y, CD20, CD33, CD30, ganglioside GD3, 9-O-acetyl-GD3, GM2, Globo H, fucose GM1, Poly SA, GD2, carbonic anhydrase IX (MN/CA IX), CD44v6, and Shh (Sonic Hedgkinase). ehog), Wue-1, plasma cell antigen, (membrane-bound) IgE, MCSP (chondroitin sulfate proteoglycan for melanoma), CCR8, TNF-α precursor, STEAP, mesothelin, A33 antigen, PSCA (prostate stem cell antigen), Ly-6; desmosome core glycoprotein 4, E-cadherin epitope, fetal acetylcholine receptor, CD25, CA19-9 marker, CA-125 marker and MIS (Muellerian Inhibitory Substance) type II receptor, sTn (sialylated Tn antigen; TAG-72), FAP (fibroblast activation antigen), endothelial sialic acid protein, EGFRvIII, LG, SAS, PD-L1, CD47, SIRPa and CD63.

In some embodiments, the bispecific antibody comprises a full-length human antibody, wherein the first antibody variable domain of the bispecific antibody is capable of recruiting human immune effector cells by specifically binding to CD3, and wherein the second antibody variable domain of the heterodimeric protein is capable of specifically binding to a target antigen (e.g., CD20, EpCAM, or P-cadherin).

Methods for preparing bispecific antibodies are known in the art (see, for example, Suresh et al., Methods in Enzymology 121:210, 1986). Traditionally, recombinant production of bispecific antibodies is based on the co-expression of two immunoglobulin heavy-light chain pairs, where the two heavy chains have different specificities (Millstein and Cuello, Nature 305, 537-539, 1983).

According to a method for preparing bispecific antibodies, an antibody variable domain having the desired binding specificity (antibody-antigen combination site) is fused to an immunoglobulin constant region sequence. The fusion preferably has an immunoglobulin heavy chain constant region comprising at least a portion of the hinge, CH2, and CH3 regions. Preferably, it has a first heavy chain constant region (CH1) present in at least one of the fusions, containing the site necessary for light chain binding. DNA encoding the immunoglobulin heavy chain fusion and (if desired) the immunoglobulin light chain is inserted into separate expression vectors and co-transfected into a suitable host organism. This provides great flexibility in the embodiments to adjust the relative proportions of the three polypeptide fragments when unequal ratios of the three polypeptide chains used in the construction provide optimal yield. However, when expression of at least two polypeptide chains in equal ratios results in high yield, or when the ratios are not particularly significant, it is possible to insert coding sequences for two or all three polypeptide chains into a single expression vector.

In one method, the bispecific antibody comprises a hybrid immunoglobulin heavy chain having a first binding specificity in one arm and a hybrid immunoglobulin heavy chain-light chain pair (providing a second binding specificity) in the other arm. This asymmetric structure, with the immunoglobulin light chain in only half of the bispecific molecule, facilitates the separation of the desired bispecific compound from the unwanted combination of immunoglobulin chains. This method is described in PCT Publication No. WO 94/04690.

In another approach, the bispecific antibody is constructed by modifying an amino acid in a first hinge region of one arm, wherein the substituted/alternated amino acid in the first hinge region has an opposite charge to the corresponding amino acid in a second hinge region of the other arm. This approach is described in International Patent Application No. PCT/US2011/036419 (WO2011/143545).

In another approach, the formation of a desired heteromeric or heterodimeric protein (e.g., a bispecific antibody) is enhanced by altering or engineering the interface between the first and second immunoglobulin-like Fc regions (e.g., hinge regions and/or CH3 regions). In this approach, the bispecific antibody may be composed of a CH3 region containing a first and a second CH3 polypeptide that interact together to form a CH3 interface, wherein one or more amino acids within the CH3 interface destabilize homodimer formation and are electrostatically unfavorable to homodimer formation. This approach is described in International Patent Application No. PCT/US2011/036419 (WO2011/143545).

In another approach, bispecific antibodies can be generated in the presence of transglutaminase using a glutamine-containing peptide tag engineered for an antibody targeting an epitope (e.g., BCMA) in one arm, and another peptide tag engineered for a second antibody targeting a second epitope in the other arm (e.g., a Lys-containing peptide tag or reactive endogenous Lys). This approach is described in International Patent Application No. PCT/IB2011/054899 (WO2012/059882).

In some embodiments, the first and second antibody variable domains of the bispecific antibody include amino acid modifications at various positions, wherein the first and second antibody variable domains of the bispecific antibody include amino acid modifications at positions 223, 225, and 228 in the hinge region of human IgG2 (e.g., (C223E or C223R), (E225R), and (P228E or P228R)) and amino acid modifications at positions 409 or 368 in the CH3 region (e.g., K409R or L368E (EU numbering scheme)).

In some embodiments, the first and second antibody variable domains of the bispecific antibody include amino acid modifications at positions 221 and 228 in the hinge region of human IgG1 (e.g., (D221R or D221E) and (P228R or P228E)) and amino acid modifications at positions 409 or 368 in the CH3 region (e.g., K409R or L368E (EU numbering scheme)).

In some embodiments, the first and second antibody variable domains of the bispecific antibody include an amino acid modification at position 228 in the hinge region of human IgG4 (e.g., (P228E or P228R)) and an amino acid modification at position 409 or 368 in the CH3 region (e.g., R409 or L368E (EU numbering scheme)).

In some implementations, the bispecific antibody may have any of the features or properties of any bispecific antibody provided in WO2016166629, which is incorporated herein by reference for all purposes.

Immunomodulators include a variety of different molecular types that can stimulate immune responses in subjects, such as pattern recognition receptor (PRR) agonists, immunostimulatory cytokines, and cancer vaccines.

Pattern recognition receptors (PRRs) are receptors expressed by cells of the immune system that recognize a variety of molecules associated with pathogens and/or cell damage or death. PRRs are involved in both innate and adaptive immune responses. PRR agonists can be used to stimulate immune responses in subjects. Several classes of PRR molecules exist, including toll-like receptors (TLRs), RIG-I-like receptors (RLRs), nucleotide-binding oligomerization domain (NOD)-like receptors (NLRs), C-type lectin receptors (CLRs), and interferon gene stimulator (STING) proteins.

The terms “TLR” and “toll-like receptor” refer to any toll-like receptor. Toll-like receptors are receptors involved in activating immune responses. TLRs recognize, for example, pathogen-associated molecular patterns (PAMPs) expressed in microorganisms, and endogenous damage-associated molecular patterns (DAMPs) released from dying or dying cells.

Molecules that activate TLRs (and thus activate the immune response) are referred to herein as “TLR agonists.” TLR agonists can include, for example, small molecules (e.g., organic molecules with a molecular weight of less than about 1000 Daltons) as well as large molecules (e.g., oligonucleotides and proteins). Some TLR agonists are specific to a single type of TLR (e.g., TLR3 or TLR9), while others activate two or more types of TLRs (e.g., both TLR7 and TLR8).

The exemplary TLR agonists provided herein include agonists of TLR2, TLR3, TLR4, TLR5, TLR6, TLR7, TLR8, and TLR9.

Exemplary small molecule TLR agonists include, for example, U.S. Patent Nos. 4,689,338; 4,929,624; 5,266,575; 5,268,376; 5,346,905; 5,352,784; 5,389,640; 5,446,153; 5,482,936; 5,756,747; 6,110,929; 6,194,425; 6,331,539; 6,376,669; 6,451,810; 6,525,064; 6,541,485; 6,545,016; 6,545,017; 6,573,273; 6,656,938; 6,660,735; 6 660,747; 6,664,260; 6,664,264; 6,664,265; 6,667,312; 6,670,372; 6,677,347; 6,677,348; 6,677,349; 6,683,088; 6,756,382; 6,797,718; 6,818,650; and 7,7091,214; U.S. Patent Publications 2004/0091491, 2004/0176367 and 2006/0100229; and International Publications WO 2005/18551, WO2005/18556, WO 2005/20999, WO 2005/0 32484, WO 2005/048933, WO 2005/048945, WO 2005/051317, WO2005/051324, WO 2005/066169, WO 2005/066170, WO 2005/066172, WO 2005/076783, WO 2005/079195, WO2005/094531, WO 2005/123079, WO 2005/123080, WO 2006/009826, WO 2006/009832, WO 2006/026760, WO2006/028451 WO 2006/028545、WO 2006/028962、WO 2006/029115、WO 2006/038923、WO 2006/065280、WO2006/074003、WO 2006/083440、WO 2006/086449、WO 2006/091394, WO 2006/086633, WO 2006/086634, WO 2006/091567, WO 2006/091568, WO 2006/091647, WO 2006/093514 and WO 2006/098852.

Additional examples of small molecule TLR agonists include certain purine derivatives (such as those described in U.S. Patent Nos. 6,376,501 and 6,028,076), certain imidazoquinoline amide derivatives (such as those described in U.S. Patent No. 6,069,149), certain imidazopyridine derivatives (such as those described in U.S. Patent No. 6,518,265), and certain benzimidazole derivatives (such as those described in U.S. Patent No. 6,387,938). Certain derivatives of 4-aminopyrimidine fused with a five-membered nitrogen-containing heterocycle (e.g., adenine derivatives described in U.S. Patent Nos. 6,376,501, 6,028,076, and 6,329,381; and WO 02/08905) and certain 3-β-D-furanoribosylthiazo[4,5-d]pyrimidine derivatives (e.g., those described in U.S. Publication No. 2003/0199461) and certain small molecule immune enhancer compounds, such as those described in U.S. Patent Publication No. 2005/0136065.

Exemplary macromolecular TLR agonists comprise oligonucleotide sequences. Some TLR agonist oligonucleotide sequences contain cytosine-guanine dinucleotides (CpG) and are described, for example, in U.S. Patent Nos. 6,194,388; 6,207,646; 6,239,116; 6,339,068; and 6,406,705. Some CpG-containing oligonucleotides may include synthetic immunomodulatory structural motifs, such as those described, for example, in U.S. Patent Nos. 6,426,334 and 6,476,000. Other TLR agonist nucleotide sequences lack the CpG sequence and are described, for example, in International Patent Publication No. WO 00/75304. Other TLR agonist nucleotide sequences include single-stranded RNAs (ssRNAs) rich in guanosine and uridine, such as those described, for example, in Heil et al., Science, vol. 303, pp. 1526-1529, March 5, 2004.

Other TLR agonists include biomolecules such as aminoalkyl glucosinolate phosphates (AGPs), and are described, for example, in U.S. Patent Nos. 6,113,918; 6,303,347; 6,525,028; and 6,649,172.

TLR agonists also include inactivated pathogens or fractions thereof, which can activate a variety of different types of TLR receptors. Exemplary pathogen-derived TLR agonists include BCG, mycobacterium obuense extract, Talimogene laherparepvec (T-Vec) (derived from HSV-1), and Pexa-Vec (derived from vaccinia virus).

In some implementations, the TLR agonist may be an agonist antibody that specifically binds to a TLR.

The following provides a brief description of various TLRs and TLR agonists. The list of TLR agonists for a particular TLR below should not be interpreted as indicating that a given TLR agonist necessarily activates only that TLR (e.g., some molecules can activate multiple types of TLRs, or even multiple classes of PRRs). For example, some molecules provided below as exemplary TLR4 agonists could also be TLR5 agonists.

The terms “TLR1” and “toll-like receptor 1” refer to any form of TLR1 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR1 activity. Unless otherwise specified, such as by way of a specific reference to human TLR1, TLR1 includes the naturally occurring sequence TLR1 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR1 is provided under UniProt entry number Q15399.

As used herein, “TLR1 agonist” means any molecule that, upon binding to TLR1, (1) stimulates or activates TLR1, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR1, or (3) enhances, increases, promotes, or induces the expression of TLR1. TLR1 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, bacterial lipoproteins and their derivatives that bind to TLR1.

Examples of TLR1 agonists useful in the treatments, medicines, and uses of the present invention include, for example, bacterial lipoproteins and their derivatives, such as SPM-105 (derived from autoclaved mycobacteria), OM-174 (lipid A derivative), OmpS1 (porin from Salmonella typhi), OspA (from Borrelia burgdorferi), MALP-2 (mycoplasma macrophage-activated lipopeptide-2kD), STF (soluble tuberculosis factor), CU-T12-9, Diprovocim, and lipopeptides derived from cell wall components, such as PAM2CSK4, PAM3CSK4, and PAM3Cys.

TLR1 can form a heterodimer with TLR2, and therefore, many TLR1 agonists are also TLR2 agonists.

The terms “TLR2” and “toll-like receptor 2” refer to any form of TLR2 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR2 activity. Unless otherwise specified, such as by way of a specific reference to human TLR2, TLR2 includes naturally occurring sequence TLR2 in all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR2 is provided under UniProt entry number O60603.

As used herein, “TLR2 agonist” means any molecule that, upon binding to TLR2, (1) stimulates or activates TLR2, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR2, or (3) enhances, increases, promotes, or induces the expression of TLR2. TLR2 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, bacterial lipoproteins and their derivatives that bind to TLR2.

Examples of TLR2 agonists useful in the treatments, medicines, and uses of the present invention include, for example, bacterial lipoproteins (e.g., diacylated lipoproteins) and their derivatives, such as SPM-105 (derived from autoclaved mycobacteria), OM-174 (lipid A derivative), OmpS1 (porin from Salmonella typhi), OspA (from Borrelia brevicornuate), MALP-2 (mycoplasma macrophage-activated lipopeptide-2kD), STF (soluble tuberculosis factor), CU-T12-9, Diprovocim, Amplivant, and lipopeptides derived from cell wall components, such as PAM2CSK4, PAM3CSK4, and PAM3Cys.

TLR2 can form heterodimers with TLR1 or TLR6, and therefore, many TLR2 agonists are also TLR1 or TLR6 agonists.

The terms “TLR3” and “toll-like receptor 3” refer to any form of TLR3 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR3 activity. Unless otherwise specified, such as by way of a specific reference to human TLR3, TLR3 includes naturally occurring sequence TLR3 in all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR3 is provided under UniProt entry number O15455.

As used herein, “TLR3 agonist” means any molecule that, upon binding to TLR3, (1) stimulates or activates TLR3, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR3, or (3) enhances, increases, promotes, or induces the expression of TLR3. TLR3 agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, nucleic acid ligands that bind to TLR3.

Examples of TLR3 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include TLR3 ligands such as synthetic dsRNA, polyinosinic acid [“poly(I:C)”] (available from, for example, InvivoGen in high molecular weight (HMW) and low molecular weight (LMW) formulations), polyadenylic-polyuridylic acid [“poly(A:U)”] (available from, for example, InvivoGen), and polyICLC (see Levy et al., Journal of Infectious Disorders). (See Jasani et al., Vaccine, vol. 132, no. 4, pp. 434-439, 1975), Ampligen (see Jasani et al., Vaccine, vol. 27, no. 25-26, pp. 3401-3404, 2009), Hiltonol, Rintatolimod, and RGC100 (see Naumann et al., Clinical and Developmental Immunology, vol. 2013, article ID 283649).

The terms “TLR4” and “toll-like receptor 4” refer to any form of TLR4 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR4 activity. Unless otherwise specified, such as by way of a specific reference to human TLR4, TLR4 includes naturally occurring sequence TLR4 in all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR4 is provided under UniProt entry number O00206.

As used herein, “TLR4 agonist” means any molecule that, upon binding to TLR4, (1) stimulates or activates TLR4, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR4, or (3) enhances, increases, promotes, or induces the expression of TLR4. TLR4 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, bacterial lipopolysaccharide (LPS) and its derivatives that bind to TLR4.

Examples of TLR4 agonists useful in the treatments, medicines and uses of the present invention include, for example, bacterial lipopolysaccharide (LPS) and its derivatives, such as B:0111 (Sigma), monophospholipid A (MPLA), 3DMPL (3-O-deacylated MPL), GLA-AQ, G100, AS15, ASO2, GSK1572932A (GlaxoSmithKline, UK).

The terms “TLR5” and “toll-like receptor 5” refer to any form of TLR5 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR5 activity. Unless otherwise specified, such as by way of a specific reference to human TLR5, TLR5 includes naturally occurring sequence TLR5s from all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR5 is provided under UniProt entry number O60602.

As used herein, “TLR5 agonist” means any molecule that, upon binding to TLR5, (1) stimulates or activates TLR5, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR5, or (3) enhances, increases, promotes, or induces the expression of TLR5. TLR5 agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, bacterial flagellin and derivatives thereof that bind to TLR5.

Examples of TLR5 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include, for example, bacterial flagellin purified from Bacillus subtilis, flagellin purified from Pseudomonas aeruginosa, flagellin purified from Salmonella typhimurium, and recombinant flagellin (all available from InvivoGen), and entolimod (CBLB502; a pharmacologically optimized flagellin derivative).

The terms “TLR6” and “toll-like receptor 6” refer to any form of TLR6 receptor, as well as variants, isotypes, and species homologs that retain at least a portion of TLR6 activity. Unless otherwise specified, such as by way of a specific reference to human TLR6, TLR6 includes the naturally occurring sequence TLR6 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR6 is provided under UniProt entry number Q9Y2C9.

As used herein, “TLR6 agonist” means any molecule that, upon binding to TLR6, (1) stimulates or activates TLR6, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR6, or (3) enhances, increases, promotes, or induces the expression of TLR6. TLR6 agonists useful in any of the treatments, medicaments, and uses of this invention include, for example, bacterial lipopeptides and derivatives thereof that bind to TLR6.

Examples of TLR6 agonists useful in the treatments, medicines, and uses of the present invention include, for example, many of the TLR2 agonists provided above, because TLR2 and TLR6 can form heterodimers. TLR6 can also form heterodimers with TLR4, and TLR6 agonists can include the various TLR4 agonists provided above.

The terms “TLR7” and “toll-like receptor 7” refer to any form of TLR7 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR7 activity. Unless otherwise specified, such as by way of a specific reference to human TLR7, TLR7 includes the naturally occurring sequence TLR7 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR7 is provided under UniProt entry number Q9NYK1.

As used herein, “TLR7 agonist” means any molecule that, upon binding to TLR7, (1) stimulates or activates TLR7, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR7, or (3) enhances, increases, promotes, or induces the expression of TLR7. TLR7 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acid ligands that bind to TLR7.

Examples of TLR7 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include recombinant single-stranded (“ss”) RNA, imidazoquinoline compounds such as imiquimod (R837), gardiquimod, and resiquimod (R848); rozolidin (7-allyl-7,8-dihydro-8-oxoguanosine) and related compounds; 7-thia-8-oxoguanosine, 7-denitroguanosine, and related guanosine analogs; ANA975 (Anadys Pharmaceuticals) and related compounds; SM-360320 (Sumimoto); 3M-01, 3M-03, 3M-852, and 3M-S-34240 (3M Pharmaceuticals); GSK2245035 (GlaxoSmithKline; 8-oxoadenine molecule), AZD8848 (AstraZeneca; 8-oxoadenine molecule), MEDI9197 (Medimmune; formerly 3M-052), ssRNA40, and adenosine analogs, such as UC-1V150 (Jin et al., Bioorganic Medicinal Chem Lett (2006) 16:4559-4563, compound 4). Many TLR7 agonists are also TLR8 agonists.

The terms “TLR8” and “toll-like receptor 8” refer to any form of TLR8 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR8 activity. Unless otherwise specified, such as by way of a specific reference to human TLR8, TLR8 includes the naturally occurring sequence TLR8 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR8 is provided under UniProt entry number Q9NR97.

As used herein, “TLR8 agonist” means any molecule that, upon binding to TLR8, (1) stimulates or activates TLR8, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR8, or (3) enhances, increases, promotes, or induces the expression of TLR8. TLR8 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acid ligands that bind to TLR8.

Examples of TLR8 agonists useful in the treatments, medicines, and uses of the present invention include recombinant single-stranded ssRNA, imiquimod (R837), gademod, requimod (R848), 3M-01, 3M-03, 3M-852, and 3M-S-34240 (3M Pharmaceuticals); GSK2245035 (GlaxoSmithKline; 8-oxoadenine molecule), AZD8848 (AstraZeneca; 8-oxoadenine molecule), MEDI9197 (Medimmune; formerly 3M-052), Poly-G10, Motolimod, and various TLR7 agonists provided above (as previously noted, many TLR7 agonists are also TLR8 agonists).

The terms “TLR9” and “toll-like receptor 9” refer to any form of TLR9 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR9 activity. Unless otherwise specified, such as by way of a specific reference to human TLR9, TLR9 includes the naturally occurring sequence TLR9 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR9 is provided under UniProt entry number Q9NR96.

As used herein, “TLR9 agonist” means any molecule that, upon binding to TLR9, (1) stimulates or activates TLR9, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR9, or (3) enhances, increases, promotes, or induces the expression of TLR9. TLR9 agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, nucleic acid ligands that bind to TLR9.

Examples of TLR9 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include unmethylated CpG-containing DNA, immunostimulatory oligodeoxynucleotides (ODNs), such as CpG-containing ODNs, such as CpG24555, CpG10103, CpG7909 (PF-3512676/agatolimod), CpG1018, AZD1419, ODN2216, MGN1703, SD-101, 1018ISS, and CMP-001. TLR9 agonists also include a nucleotide sequence containing a synthetic cytosine-phosphate-2'-deoxy-7-dezoguanosine dinucleotide (CpR) (Hybridon, Inc.), dSLIM-30L1, and an immunoglobulin-DNA complex. Exemplary TLR9 agonists are disclosed in WO2003/015711, WO2004/016805, WO2009/022215, PCT/US95/01570, PCT/US97/19791 and U.S. Patent Nos. 8552165, 6194388 and 6239116, which are incorporated herein by reference for all purposes.

RLRs include various cytoplasmic PRRs that detect, for example, dsRNAs. Examples of RLRs include, for example, retinoic acid-inducible gene I (RIG-I), melanoma differentiation-associated gene 5 (MDA-5), and Laboratory of Genetics and Physiology 2 (LGP2).

As used herein, “RLR agonist” means any molecule that, upon binding to an RLR, (1) stimulates or activates the RLR, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of the RLR, or (3) enhances, increases, promotes, or induces the expression of the RLR. RLR agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acids that bind to RLRs and their derivatives, and agonistic monoclonal antibodies (mAbs) that specifically bind to RLRs.

Examples of RLRs agonists useful in the treatments, medicines, and uses of the present invention include, for example, short double-stranded RNAs having an uncapped 5' triphosphate (RIG-I agonists); poly I:C (MDA-5 agonists); and BO-112 (MDA-A agonists).

NLRs include various PRRs that detect molecules such as damage-related molecular patterns (DAMPs). NLRs include subfamilies NLRA-A, NLRB-B, NLRC-C, and NLRP-P. Examples of NLRs include, for example, NOD1, NOD2, NAIP, NLRC4, and NLRP3.

As used herein, “NLR agonist” means any molecule that, upon binding to an NLR, (1) stimulates or activates the NLR, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of the NLR, or (3) enhances, increases, promotes, or induces the expression of the NLR. NLR agonists useful in any of the treatments, medicines, and uses of this invention include, for example, DAMPs and their derivatives that bind to NLRs, and agonistic monoclonal antibodies (mAbs) that specifically bind to NLRs.

Examples of NLR agonists useful in the treatments, medicines, and uses of the present invention include, for example, liposomal muramyl tripeptide/mifamurtide (NOD2 agonist).

CLRs include various PRRs that detect, for example, carbohydrates and glycoproteins. CLRs include both transmembrane CLRs and secretory CLRs. Examples of CLRs include, for example, DEC-205/CD205, macrophage mannose receptor (MMR), Dectin-1, Dectin-2, mincle, DC-SIGN, DNGR-1, and mannose-binding lectin (MBL).

As used herein, “CLR agonist” means any molecule that, upon binding to a CLR, (1) stimulates or activates the CLR, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of the CLR, or (3) enhances, increases, promotes, or induces the expression of the CLR. CLR agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, carbohydrates and their derivatives that bind to CLRs, and agonistic monoclonal antibodies (mAbs) that specifically bind to CLRs.

Examples of CLR agonists useful in the treatments, pharmaceuticals, and uses of the present invention include, for example, MD fraction (a purified soluble β-glucan extract from Grifola frondosa) and imprime PGG (a yeast-derived β1,3/1,6-glucan PAMP).

STING proteins function as both cytoplasmic DNA sensors and adaptors in type 1 interferon signaling. The terms "STING" and "stimulator of interferon genes" refer to any form of STING protein, as well as variants, allotypes, and species homologs that retain at least a portion of STING activity. Unless otherwise specified, such as by way of a specific reference to human STING, STING includes the naturally occurring sequence of STING in all mammalian species, e.g., humans, monkeys, and mice. An exemplary human TLR9 is provided under UniProt entry number Q86WV6. STING is also known as TMEM173.

As used herein, “STING agonist” means any molecule that, upon binding to TLR9, (1) stimulates or activates STING, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of STING, or (3) enhances, increases, promotes, or induces the expression of STING. STING agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acid ligands that bind to STING.

Examples of STING agonists useful in the treatments, medicines, and uses of the present invention include various immunostimulatory nucleic acids, such as synthetic double-stranded DNA, cyclic di-GMP, cyclic GMP-AMP (cGAMP), synthetic cyclic dinucleotides (CDNs), such as MK-1454 and ADU-S100 (MIW815), and small molecules, such as PO-424.

Other PRRs include, for example, DNA-dependent Activator of IFN-regulatory Factors (DAI) and Absentin Melanoma 2 (AIM2).

Immunostimulatory cytokines include various signaling proteins that stimulate immune responses, such as interferons, interleukins, and hematopoietic growth factors.

Exemplary immunostimulatory cytokines include GM-CSF, G-CSF, IFN-α, IFN-γ; IL-2 (e.g., denileukin difitox), IL-6, IL-7, IL-11, IL-12, IL-15, IL-18, IL-21, and TNF-α.

Immunostimulatory cytokines can be in any suitable form. In some embodiments, the immunostimulatory cytokines can be recombinant forms of wild-type cytokines. In some embodiments, the immunostimulatory cytokines can be mutant proteins with one or more amino acid alterations compared to the corresponding wild-type cytokines. In some embodiments, the immunostimulatory cytokines can be incorporated into chimeric proteins containing cytokines and at least one other functional protein (e.g., antibodies). In some embodiments, the immunostimulatory cytokines can be covalently linked to a drug/pharmaceutical (e.g., any drug/pharmaceutical as described elsewhere herein as a possible ADC component).

Cancer vaccines include various compositions containing tumor-associated antigens (or that can be used to generate tumor-associated antigens in subjects) and thus can be used to induce an immune response in subjects against tumor cells containing tumor-associated antigens.

Examples of materials that may be included in a cancer vaccine include attenuated cancerous cells, tumor antigens, and antigen-presenting cells such as dendritic cells bombarded with nucleic acids encoding tumor-derived antigens or tumor-associated antigens. In some embodiments, the cancer vaccine may be prepared using the patient's own cancer cells. In some embodiments, the cancer vaccine may be prepared using biomaterials not derived from the patient's own cancer cells.

Cancer vaccines include, for example, sipuleucel-T and talimogene laherparepvec (T-VEC).

Immunotherapy involves treating patients with immune cells that can target cancer cells. Examples of immunotherapy include, for instance, tumor-infiltrating lymphocytes (TILs) and chimeric antigen receptor T cells (CAR-T cells).

Examples of chemotherapeutic agents include alkylating agents such as thiotepa and cyclophosphamide; alkyl sulfonates such as busulfan, dipropylamine sulfonate, and piperosulfan; aziridines such as benzodopa, carboquinone, meturedopa, and uredopa; ethyleneimines and methylamelamines, including hexamethylmelamine, tratamide, trietylenephosphoramide, ethyl thiophosphoric acid, and trimethylolomelamine; polyacetogenins (especially bullatacin and bullatacinone); camptothecin (including synthetic varieties). Analogs of topotecan; bryozoins; callystatin; CC-1065 (including its synthetic analogs of adolexin, carzelesin, and bizelesin); cryptophycins (especially cryptophycin 1 and cryptophycin 8); dolastatin; bacitracin (including synthetic analogs, KW-2189 and CBI-TMI); eleutherobin; pancratistatin; sarcodictyin; spongistatin; nitrogen mustard, such as chlorambucil, naphthalene, and chlorpyrifos. Cholophosphamide, estradiol, ifosfamide, nitrogen mustard, mechlorethamine, mechlorethamine oxide hydrochloride, melphalan, neonitrogen mustard, benzyl mustard cholesterol, pine oxychloride, tristyramine cyclophosphamide, uracil nitrogen mustard; nitrosoureas, such as carmustine, chloramphenicol, formazan, lomomia, nimustine, ranolamia; antibiotics, such as enediyne antibiotics (e.g., galicarmycin, especially galicarmycin γ1I and galicarmycin phiI1, see, e.g., Agnew, Chem. Intl. Ed. Engl., 33:183-186(19) 94); dynemicin, including dynemicin A; bisphosphonates, such as osteophosphine; esporamin; and new carcinogenic chromophores and related chromomophores, aclarubicin, actinomycin, autramycin, azoserine, bleomycin, cactinomycin C, carabicin, caminomycin, carcinomycin, chromomycin, daunorubicin, doxorubicin, 6-diazo-5-oxo-L-leucine, doxorubicin (including morpholino-doxorubicin, cyanomorpholino-doxorubicin, 2-pyrrolino-doxorubicin) (2-pyrrolino-doxorubicin and deoxydoxomyl), PEGylated liposomal doxorubicin, epirubicin, doxorubicin, idarubicin, mesorcinol, mitomycin such as mitomycin C, mycophenolic acid, nogaramycin, oligomycin, pemycin, potfiromycin, puromycin, triamcinolone acetonide, rodorubicin, streptomycin, streptozotocin, tuberculin, hydroxyphenylbutyryl leucine, neomycin, zorubicin; antimetabolites, such as methotrexate and 5-fluorouracil (5-FU); folic acid analogs, such as folate, methotrexate, pteroyltriglutamate, trimesat; purine analogs, such as fludarabine, 6-mercaptopurine, thioguanine, thioguanine; pyrimidine analogs, such as cyclocytidine, azathioprine, etc. Cytidine, 6-azouridine, carmoflurane, cytarabine, dideoxyuridine, docefluuridine, enoxabin, fluorouridine; androgens, such as carotestosterone, hydroxymethylandrostenone propionate, cyclothiosterol, methylcyclothiosterane, testrolide; anti-adrenergic drugs, such as aminoglutethimide, mitotane, and trilostertan; folic acid supplements, such as frolinic acid; aceglucuronolactone; aldophosphamide glycoside; aminolevulinic acid; eniluracil; acridine; bestrabucil; bethanycin; edatraxate; defofamine; dimethicone; diazinon; elfo-ornithine. rmithine); hydroxypicarazole acetate; epothilone; glycidyl ether; gallium nitrate; hydroxyurea; lentinan; chlordamine; maytansine alkaloids, such as maytansine and metansine; propiconazole; mitoxantrone; monpilamol; diammonium nitrate; pentostatin; benzylmethionine; pirarubicin; losoxantrone; podophyllin; 2-ethylhydrazide; methylbenzylhydrazine; propylimine; rhizomycin; sisofibril; spirogermanium; Alternaria spp. ketoacid; triazolam; 2,2',2''-trichlorotriethylamine; trichosporon toxins (especially T-2 toxin, verracurin A, baculosporin A, and anguidine); ursodeoxycholic acid; vinblastine; dacarbazin; mannitol; dibromomannitol; dibromodehydroxycervastatin Piperbromopropionate; gacytosine; cytarabine (“Ara-C”); cyclophosphamide; thiotepa; taxoids, such as paclitaxel and docetaxel; chlorambucil; gemcitabine; 6-thioguanine; mercaptopurine; methotrexate; platinum analogs, such as cisplatin and carboplatin; vincristine; platinum; etoposide (VP-16); ifosfamide; mitoxantrone; vincristine; vinblastine; nifedipine; teniposide; edaraxacum; daunorubicin; aminopterin; capecitabine; ibenphosphonate sodium; CPT-11; topoisomerase inhibitor RFS 2000; difluoromethylornithine (DMFO); retinoids, such as retinoic acid; capecitabine; and pharmaceutically acceptable salts, acids, or derivatives of any of the above. It also includes anti-hormonal agents used to regulate or inhibit the hormonal effects on tumors, such as anti-estrogens and selective estrogen receptor modulators (SERMs), including, for example, tamoxifen, raloxifene, droloxifene, 4-hydroxytamoxifen, treoxifene, naloxifene, LY117018, onasone, and toremifene (Fareston); aromatase inhibitors that inhibit aromatase, which regulate estrogen production in the adrenal glands, such as 4(5)-imidazole, aminoglutethimide, megestrol acetate, etc. Exemestane, formetastatin, fazodazole, vorazole, letrozole, and anastrozole; and anti-androgen substances such as flutamide, nilumid, bicalutamide, lipoprotein, and serotonin; KRAS inhibitors; MCT4 inhibitors; MAT2a inhibitors; tyrosine kinase inhibitors, such as sunitinib and axitinib; alk/c-Met/ROS inhibitors, such as crizotinib and lorlatinib. Intra-inib; mTOR inhibitors, such as temsirolimus and gedatolisib; src/abl inhibitors, such as bosutinib; cyclin-dependent kinase (CDK) inhibitors, such as palbociclib and PF-06873600; erb inhibitors, such as dacomitinib; PARP inhibitors, such as talazoparib; SMO inhibitors, such as glasdegib and PF-5274857; EGFR T790M inhibitors, such as PF-06747775; EZH2 inhibitors, such as PF-06821497; PRMT5 inhibitors, such as PF-06939999; TGFRβr1 inhibitors, such as PF-06952229; and pharmaceutically acceptable salts, acids, or derivatives of any of the above. In specific implementation schemes, this additional treatment agent is bevacizumab, cetuximab, sirolimus, panitumumab, 5-fluorouracil (5-FU), capecitabine, tevozanib, irinotecan, oxaliplatin, cisplatin, trifluorouridine, tilpyrimidine, leucovorin, gemcitabine, regorafinib, or erlotinib hydrochloride.

In some embodiments, the antibody is used in conjunction with one or more other therapeutic agents that target immune restriction point modulators or co-stimulators, such as, but not limited to, those targeting CTLA-4, LAG-3, B7-H3, B7-H4, B7-DC (PD-L2), B7-H5, B7-H6, B7-H8, B7-H2, B7-1, B7-2, ICOS, ICOS-L, TIGIT, CD2, CD47, CD80, CD86, CD48, C D58, CD226, CD155, CD112, LAIR1, 2B4, BTLA, CD160, TIM1, TIM-3, TIM4, VISTA(PD-H1), OX40, OX40L, GITR, GITRL, CD70, CD27, 4-1BB, 4-BBL, DR3, TL1A, CD40, CD40L, CD30, CD30L, LIGHT, HVEM, SLAM (SLAMF1, CD150 ), SLAMF2 (CD48), SLAMF3 (CD229), SLAMF4 (2B4, CD244), SLAMF5 (CD84), SLAMF6 (NTB-A), SLAMCF7 (CS 1), SLAMF8 (BLAME), SLAMF9 (CD2F), CD28, CEACAM1 (CD66a), CEACAM3, CEACAM4, CEACAM5, CEACAM6, CEAC AM7, CEACAM8, CEACAM1-3AS, CEACAM3C2, CEACAM1-15, PSG1-11, CEACAM1-4C1, CEACAM1-4S, CEACAM1-4L, IDO, TDO, CCR2, CD39–CD73–adenosine pathway (A2AR), BTKs, TIKs, CXCR2, CXCR4, CCR4, CCR8, CCR5, CSF-1, or agents that modulate the innate immune response.

In some implementations, the antibody is used in conjunction with, for example, the following: anti-CTLA-4 antagonist antibodies, such as ipilimumab; anti-LAG-3 antagonist antibodies, such as BMS-986016 and IMP701; anti-TIM-3 antagonist antibodies; anti-B7-H3 antagonist antibodies, such as MGA271; anti-VISTA antagonist antibodies; anti-TIGIT antagonist antibodies; antibodies; anti-CD80 antibodies; anti-CD86 antibodies; anti-B7-H4 antagonist antibodies; anti-ICOS agonist antibodies; anti-CD28 agonist antibodies; modulators of innate immune responses (e.g., TLRs, KIR, NKG2A); and IDO inhibitors.

In some embodiments, the antibody is used with an OX40 agonist, such as, for example, an anti-OX-40 agonist antibody. In some embodiments, the antibody is used with a GITR agonist, such as, for example, an anti-GITR agonist antibody, such as, but not limited to, TRX518. In some embodiments, the antibody is used with an IDO inhibitor. In some embodiments, the antibody is used with cytokine therapy (such as, for example, but not limited to, IL-15, CSF-1, MCSF-1, etc.).

In some implementations, the antibody is used in conjunction with one or more other therapeutic antibodies, such as, but not limited to, antibodies targeting CD19, CD22, CD40, CD52, or CCR4.

In some embodiments, the antibody composition comprises at least one additional pharmaceutical agent, such as bevacizumab, cetuximab, sirolimus, panitumumab, 5-fluorouracil (5-FU), capecitabine, tevozanib, irinotecan, oxaliplatin, cisplatin, trifluorouridine, tilpyridamole, leucovorin, gemcitabine, and erlotinib hydrochloride.

In some embodiments, the antibody therapy may be administered co-administered with other reagents or sequentially before or after other reagents at intervals ranging from minutes to weeks. In embodiments where other reagents and/or proteins or polynucleotides are administered alone, it will generally be ensured that a considerably long period of time does not expire between each delivery, such that the agent and the composition of the invention will still be able to produce a beneficial combined effect on the subject. In this case, it is anticipated that the two dosing regimens may be administered within approximately 12-24 hours of each other, and more preferably within approximately 6-12 hours of each other. However, in some cases, it may be desirable to significantly extend the time period for administration, in this regard, from several days (2, 3, 4, 5, 6, or 7) to several weeks (1, 2, 3, 4, 5, 6, 7, or 8) between each administration.

In some embodiments, the antibodies of the present invention are combined with treatment regimens that further include conventional therapies selected from: surgery, radiotherapy, chemotherapy, targeted therapy, immunotherapy, hormone therapy, angiogenesis inhibition, and palliative care.

The compositions used in this invention may further comprise pharmaceutically acceptable carriers, excipients, or stabilizers in the form of lyophilized formulations or aqueous solutions (Remington: The Science and Practice of Pharmacy, 21st edition, 2005, Lippincott Williams and Wilkins, Ed. K. E. Hoover). Acceptable carriers, excipients, or stabilizers are non-toxic to the recipient at the stated doses and concentrations and may include buffers such as phosphates, citrates, and other organic acids; antioxidants, including ascorbic acid and methionine; preservatives (e.g., octadecyl dimethyl benzyl ammonium chloride; hexane diammonium chloride; benzalkonium chloride, benzyl chloride; phenols, butanol, or benzyl alcohol; alkyl esters of p-hydroxybenzoate, such as methylparaben or propylparaben; catechol; resorcinol; cyclohexanol; 3-pentanol and m-cresol); low molecular weight (less than about 10 residues) polypeptides; proteins, such as blood... Albumin, gelatin, or immunoglobulins; hydrophilic polymers, such as polyvinylpyrrolidone; amino acids, such as glycine, glutamine, asparagine, histidine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates, including glucose, mannose, or dextran; chelating agents, such as EDTA; sugars, such as sucrose, mannitol, trehalose, or sorbitol; salt-forming ions, such as sodium; metal complexes (e.g., Zn-protein complexes); and/or nonionic surfactants, such as TWEEN™, PLURONICS™, or polyethylene glycol (PEG). Pharmaceutically acceptable excipients are further described herein.

In one aspect, the present invention provides a method for generating antibodies as disclosed herein, comprising: culturing host cells under conditions that result in the generation of antibodies as disclosed herein, and purifying the antibody or the bispecific antibody from the culture.

In another aspect, the present invention provides the use of antibodies, pharmaceutical compositions, polynucleotides, vectors or host cells as disclosed herein in the preparation of medicaments for the detection, diagnosis and/or treatment of tumor antigen-related conditions.

Diagnostic uses

In one aspect, for example for diagnostic purposes, the CD3 antibody of the present invention can be used to detect and/or measure CD3 or CD3-expressing cells in a sample. For example, the CD3 antibody can be used to diagnose conditions or diseases characterized by abnormal expression of CD3 (e.g., overexpression, underexpression, lack of expression, etc.). Exemplary diagnostic assays for CD3 may include, for example, contacting a sample obtained from a subject with the CD3 antibody of the present invention, wherein the CD3 antibody is labeled with a detectable marker or reporter molecule. Alternatively, an unlabeled CD3 antibody may be used in combination with a self-detectably labeled secondary antibody for diagnostic applications. The detectable marker or reporter molecule may be a radioisotope, such as ^14C , ^3H , ^125I , ^32P , or ^35S ; a fluorescent or chemiluminescent moiety, such as fluorescein isothiocyanate or rhodamine; or an enzyme, such as alkaline phosphatase, β-galactosidase, horseradish peroxidase, or luciferase. Specific exemplary assays that can be used to detect or measure CD3 in a sample include, but are not limited to, enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), fluorescence activated cell sorting (FACS), etc. Samples that can be used in the CD3 diagnostic assay according to the present invention include any tissue or body fluid sample obtainable from a patient, containing a detectable amount of CD3 protein or fragments thereof under normal or pathological conditions. Typically, the level of CD3 is measured in a specific sample obtained from a healthy patient (e.g., a patient without a disease or condition associated with abnormal CD3 levels or activity) to first establish a baseline or standard for CD3 levels. This baseline CD3 level can then be compared with the CD3 level measured in a sample obtained from an individual suspected of having a CD3-related disease or condition.

In another aspect, methods are provided for detecting, diagnosing, and/or monitoring conditions associated with tumor antigen expression. For example, multispecific antibodies as described herein can be labeled with detectable moieties (e.g., imaging agents and enzyme-substrate markers). Antibodies as described herein can also be used for in vivo diagnostic assays (e.g., in vivo imaging (e.g., PET or SPECT)) or staining reagents.

Reagent test kit

Another aspect of the invention is a kit comprising an antibody as disclosed above and instructions for use according to any method of the invention described herein. Typically, these instructions include a description of the administration of the antibody for the therapeutic treatment described above. The kit includes any pharmaceutical compositions disclosed herein. Pharmaceutical compositions and other reagents may be present in the kit in any convenient form, such as, for example, in solution or powder form.

In another aspect, the kit further includes one or more other preventative or therapeutic agents that can be used to treat cancer, contained in one or more containers. In one embodiment, said other preventative or therapeutic agent is a chemotherapeutic agent. In other aspects, said preventative or therapeutic agent is a biological or hormonal therapeutic agent.

Several aspects of the pharmaceutical compositions, preventive agents, or therapeutic agents of the present invention are preferably tested for desired therapeutic activity in vitro, in cell culture systems, and in animal model organisms (e.g., rodent model systems) before human use.

The toxicity and efficacy of the preventive and/or therapeutic regimens of the present invention can be determined by standard pharmaceutical procedures in cell cultures or laboratory animals, for example, for determining the _LD50 (the dose that is lethal to 50% of the population) and _ED50 (the dose that is therapeutically effective in 50% of the population). The dose ratio between toxicity and therapeutic effect is the therapeutic index, and it can be expressed as the ratio _LD50 / _ED50 . Prophylactic and/or therapeutic agents exhibiting a large therapeutic index are preferred.

Furthermore, any assays known to those skilled in the art can be used to evaluate the preventive and/or therapeutic efficacy of the therapies or combinations of therapies disclosed herein for the treatment or prevention of cancer.

Instructions for use with bispecific antibodies as described herein typically include information about the dosage, dosing regimen, and route of administration for the intended treatment. Containers may be unit doses, bulk packaging (e.g., multi-dose packs), or subunit doses. Instructions supplied with kits of the present invention are typically written instructions on a label or packaging insert (e.g., paper included in the kit), but machine-readable instructions (e.g., instructions carried on a disk or optical storage disc) are also acceptable.

The kit of the present invention is in suitable packaging. Suitable packaging includes, but is not limited to, vials, ampoules, tubes, bottles, wide-mouth bottles, flexible packaging (e.g., sealed polyester film or plastic bags), etc., for each pharmaceutical composition and other included reagents (e.g., buffers, balanced saline solutions, etc., for administering the pharmaceutical composition to a subject). Packaging for use in combination with specific devices such as inhalers, nasal delivery devices (e.g., nebulizers), or infusion devices such as micropumps is also contemplated. The kit may have a sterile inlet (e.g., the container may be an intravenous solution bag or vial with a stopper that can be punctured by a hypodermic needle). The container may also have a sterile inlet (e.g., the container may be an intravenous solution bag or vial with a stopper that can be punctured by a hypodermic needle). At least one active agent in the composition is an antibody. The container may further contain a second pharmaceutical active agent.

Typically, a kit includes a container and a label or packaging insert on or associated with the container.

Biological Preservation

Representative material of the present invention was deposited on February 13, 2018 at the American Center for Type Culture Collection (ATCC), 10801 University Boulevard, Manassas, Va. 20110-2209, USA. The vector GUCY2C-1608 chain A (VH-Synthesis SEQ ID NO: 94) with ATCC accession number PTA-124943 contains a DNA insert of the VH-Synthesis chain encoding the bispecific antibody GUCY2C-1608, which contains a polynucleotide (SEQ ID NO: 70) encoding the CD3 antibody light chain variable region (SEQ ID NO: 3); and the GUCY2C-1608 chain B (VL-Synthesis SEQ ID NO: 93) with ATCC accession number PTA-124944 contains a DNA insert of the VL-Synthesis chain encoding the bispecific antibody GUCY2C-1608, which contains a polynucleotide (SEQ ID NO: 71) encoding the CD3 antibody heavy chain variable region (SEQ ID NO: 4).

The deposit was made in accordance with the provisions of the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the Purpose of Patent Procedure and Regulations thereunder. This ensures that the live culture of the deposit remains intact for 30 years from the date of deposit. The deposit is available through the ATCC under the terms of the Budapest Treaty and complies with an agreement between Pfizer, Inc. and the ATCC that ensures the perpetual and unrestricted availability of the descendants of the deposited culture to the public at the time of publication of the relevant U.S. Patent or at the time of public disclosure of any U.S. or foreign patent application (whichever comes first), and ensures the availability of the descendants to persons authorized by the Director of the U.S. Patent and Trademark Office in accordance with 35 U.S.C. Part 122 and the Director's Rules established thereunder (including 37 C.F.R. Part 1.14, which specifically refers to 886 OG 638).

The assignee of this application has agreed that, should the culture of the deposited material die, be lost, or be destroyed under suitable conditions, the material will be promptly replaced with another identical copy upon notification. The availability of the deposited material shall not be construed as a violation of any government authority's right to practice the invention under its patent law.

The following embodiments for implementing specific aspects of the invention are provided for illustrative purposes only and are not intended to limit the scope of the invention in any way.

Example 1

Formation of soluble CD3ed

A protein called CD3ed, representing the extracellular domains of CD3 ε and δ subunits tandemly fused via a short linker (SEQ ID NO: 39), was generated using standard molecular biology, protein expression, and purification techniques. Using PEI as the transfection agent, plasmids carrying the gene encoding the amino acids shown in Table 9 (SEQ ID NO: 40) were transiently transfected into 20 μL of HEK293 cells. After incubation at 37°C for 7 days, 20 μL of the harvest was fractionally bound to 50 mL of His60 resin (equilibrated with PBS) overnight. The bound resin was then washed with PBS, followed by washing with PBS buffer (10⁻¹⁵ CV) containing 10 mM imidazole. Following washing, the protein was eluted with phosphate-buffered saline containing 200 mM imidazole. The eluent fractions were then analyzed by SDS gel electrophoresis. Fractions containing purified protein were pooled and concentrated using a 3 kDa Amicon ultra centricon. During concentration, 5% glycerol was maintained in the sample to stabilize the protein. The His60 pool was further loaded onto a size-restricted Superdex 200 column. Fractions containing CD3 ε/δ heterodimers were analyzed by SDS-PAGE, and the relevant fractions were pooled and filtered at 0.22 µm. Table 9 shows the sequences of the CD3 ε/δ subunits tandemly fused and used as antigens in all optimization efforts.

Table 9

Example 2

Optimization of the structure of CD3 antibody H2B4

The lead CD3e (CD3 ε) antibody H2B4 exhibited low thermostability when used in bispecific biantibody-Fc form. To improve stability, the light chain architecture was changed from VK4 to VK1 (DPK9), while the heavy chain architecture was retained as VH3 (DP54). VLCDR1 (SEQ ID NO: 26), VL CDR2 (SEQ ID NO: 27), and VL CDR3 (SEQ ID NO: 28) were transplanted into the DPK9 architecture and expressed in combination with heavy chains carrying CDRs (i.e., VH CDR1 (SEQ ID NO: 13), VHCDR2 (SEQ ID NO: 16), and VH CDR3 (SEQ ID NO: 18)). In addition to transplanting CDRs into the architecture, certain reversion mutations were tested. The reversion mutations introduced in both the heavy and light chain structures are shown in Table 10, where the residue number indicating the reversion mutation is indicated.

A cDNA containing the human receptor framework (VH3 for the heavy chain and VK1 for the light chain) and the associated CDR donor sequence was synthesized. The synthesized cDNA product was subcloned and fused to the human IgG1 constant region of the heavy chain and the human κ-frame of the light chain in a mammalian expression vector. To distinguish them from the H2B4 antibody, these variants will be referred to as 2B5 below, and all optimized molecules will be referred to as 2B5 and its variants. The competitive binding of all 2B5 mutants to Jurkat cells with endogenous CD3e expression was analyzed.

Competitive FACS was performed to examine whether humanized 2B5 maintained its binding epitope on the cell surface. Prior to the competitive FACS assay, chimeric cH2B4 was biotinylated. The EC50 of the binding of biotinylated cH2B4 to Jurkat cells was determined by direct binding FACS and was measured to be 0.8 nM. For competitive FACS, a 3-fold serially diluted antibody variant was mixed with a constant amount of biotin-cH2B4 representing the EC50 concentration and 1.0+E05 cells/well of Jurkat cells. The mixture was incubated on ice for 1 hour, followed by three washes with FACS buffer. A secondary antibody (i.e., PE (phycoerythrin)-conjugated streptavidin (PE-SA)) was added to the cells and incubated on ice for 30 minutes. Median fluorescence intensity (MFI) was then analyzed by FACS on FACS CANTO™ (BDBiosciences). Y-axis calculation: %Maximum binding = [MFI (serially diluted Ab wells) - MFI (secondary PE-SA)] / [MFI maximum (Jurkat cells with biotin-cH2B4 only) - MFI (secondary PE-SA)]. Table 10 shows the results of the _IC50 values for competitive binding when mutations were introduced into CD3 antibody H2B4 to optimize the scaffold. Based on these results, H2B4 1.0/1.0T showed similar binding characteristics to H2B4 and was designated as the CD3 antibody binding domain 2B5.

Table 10

Antibody	Heavy chain	Light chain	<![CDATA[Competitive combination of IC<sub>50</sub>]]>
				h2B4 1.0/1.0T (2B5)	CDR grafts	CDR graft, R24T	1.27
h2B4 1.0/1.1	vH1.0	Q13V, R24T, Y49S, L78V	1.68
				h2B4 1.0/1.2	vH1.0	Q13V, R24T, Y49S	Untested
h2B4 1.0/1.3	vH1.0	R24T, Y49S	1.309
				c2B4 (Chimeric 2B4)	mouse vH	mouse vK	0.752

Example 3

Bispecific biantibody-Fc molecules were generated and characterized using H2B4 and 2B5 sequences.

Bispecific biantibody-Fc molecules were generated using standard expression and purification techniques known in the art, using CD3 antibody 2B5 or H2B4 (Tables 2 and 3) and a GUCY2c antibody sequence in one of the configurations shown in Figure 1 (schematic diagram). The stability of the resulting bispecific antibodies (Table 7; SEQ.ID.NO.36-38) was tested using differential scanning calorimetry.

Example 4

The stability of bispecific antibodies was evaluated using differential scanning calorimetry.

The protein was diluted to 0.6 mg/mL in 400 µL of phosphate-buffered saline (PBS). PBS was used as a buffer blank in the reference chamber. PBS contained 137 mM NaCl, 2.7 mM KCl, 8.1 mM Na₂HPO₄, and 1.47 mM KH₂PO₄ at pH 7.2. Samples were dispensed into the sample trays of a MicroCal VP-Capillary DSC using an autosampler (Malvern Instruments Ltd, Malvern, UK). Samples were equilibrated at 10 °C for 5 min and then scanned at 100 °C per hour up to 110 °C. A 16-second filtration period was selected. The raw data were baseline corrected and the protein concentration was normalized. The data were fitted to the MN2-State model using Origin software 7.0 (OriginLab Corporation, Northampton, MA) with an appropriate number of transformations. As illustrated in Figure 2, the bispecific GUCY2c-2B5 exhibits an improved first melt transition at 5°C compared to the bispecific GUCY2c-H2B4. This indicates that bispecific antibodies containing 2B5 demonstrate better stability.

Using FACS, the binding of two bispecific antibodies to T cells was examined using standard procedures known in the art, and as shown in Figure 3, both bispecific antibodies showed similar binding to T cells, indicating that the binding was retained after H2B4 was converted to 2B5.

Both H2B4 and 2B5 bispecific antibodies were tested in a cytotoxicity assay to evaluate T cell retargeting activity. As shown in Figure 4, both bispecific antibodies showed similar cytotoxicity, indicating that 2B5 retained the same activity as H2B4.

Example 5

2B5-based structure-based rational mutation

The CD3 antibody H2B4 binding domain exhibits multireactivity (e.g., as evaluated by binding to DNA and insulin), shows a tendency for self-association (e.g., indicated by a high AC-SINS score), and exhibits a tendency for cleavage at the R53 position in CDR-H2 when expressed in CHO host cells. To reduce multireactivity, the tendency for self-association and cleavage in CDRH2, and mutations in the CDR of 2B5, were engineered to reduce these adverse factors while maintaining binding affinity and stability. For this effort, all predictions were made using the CD3 antibody H2B4 X-ray crystal structure, obtained in a complex with a peptide representing the binding epitope and derived from the N-terminus of the CD3 ε subunit. It is well known in the art that excess positive charge and/or hydrophobicity can play a major role in multireactivity. The electrostatic surface of the H2B4 Fv region from the crystal structure was determined using the DelPhi Poisson Boltzmann calculator in Discovery Studio 4.0 ^1. In Figure 5, an excess positive charge patch that is not disrupted by negative charge is observed. This indicates a correlation with high multi-reactivity. Additionally, the Spatial Aggregation Tendency (SAP) tool in Discovery Studio ¹ was used to identify excess hydrophobic patches (Figure 6). As shown in Figure 6, the hydrophobic patches appear small and insignificant. Based on this observation, the focus is on reducing the positive charge by removing positive charges or adding negative charges to disrupt the patches.

To determine the appropriate sites for mutagenesis, computational predictions of mutations for both affinity and stability changes were performed using Discovery Studio and FoldX. Tolerance mutations were those for which neither Discovery Studio nor FoldX methods predicted ΔΔG > 1 kcal/mol. From these predictions, a set of residues predicted to have minimal impact on binding and stability while still disrupting the altered patch were identified. Additionally, we favored hydrophilic mutations that did not increase hydrophobic surface area where possible. A list of possible CDR mutations introduced in CD3 antibody 2B5 to optimize biophysical properties is shown in Table 11. This includes mutations predicted to be tolerant at each site and reduce multireactivity and splicing.

Table 11

In addition, a set of multiple mutations (combinations) were designed for both the heavy and light chains. These include R52Q/R52bQ/R53Q, R52Q/R52bQ/R53Q/R96Q, R52bQ/R53Q/R96Q, and R52Q/R52bQ/R96Q for the heavy chain, and R29Q/K30Q/R54L/V27eE, R29Q_R54L_V27eE, and R29Q_K30Q_R54L for the light chain.

Example 6

Characterization of mutants

All mutants were generated as IgG molecules using standard expression and purification techniques well-known in the art, and their binding to CD3ed, multireactivity (binding to DNA and insulin), and AC-SINS were tested.

Combined assays to evaluate variants

Octet assays were designed and run to examine both the binding and dissociation rates of 10 μg/ml soluble human CD3ed antigen (SEQ ID NO: 40). A 10 μg/ml anti-CD3 antibody was captured using an anti-human IgG tip for 120 s, followed by a baseline in PBS for 30 s, then immersed in 10 μg/ml soluble human CD3ed antigen for 120 s, and finally dissociated in it for 300 s. Sensing patterns were examined to determine whether the antigen bound to the antibody, and Octet software was used to calculate kd(1/s) and kd error. Twenty-six viable mutations retained binding to the hCD3ed antigen. The binding kinetics of the 26 remaining antigen-binding CD3 antibody variants to soluble CD3 were then examined using Biacore analysis via SPR. CD3 antibodies were captured on anti-human Fc cells, and soluble CD3 antigen at concentrations of 0 nM, 3.7 nM, 11.1 nM, 33.3 nM, and 100 nM was injected onto the captured antibodies to determine binding kinetics. Twenty-three CD3 antibodies bound the soluble CD3 antigen with an affinity similar to that of the parental antibodies used in CDR transplantation.

DNA and insulin ELISA to measure multireactivity

384-well ELISA plates (Nunc Maxisorp) were coated overnight at 4°C with DNA (10 µg/ml) (Sigma-Aldrich, D1626) and insulin (5 µg/ml) (Sigma-Aldrich, I9278-5 mL) in PBS pH 7.5. The assay, modified from that described in Tiller et al. (17), was performed on a PerkinElmer Janus Automated Workstation liquid handling robot. Wells were washed with water, blocked for 1 hour at room temperature with 50 µl of multireactive ELISA buffer (PEB; PBS containing 0.05% Tween-20, 1 mM EDTA), and rinsed three times with water. Quadruple copies of serially diluted mAbs in 25 µl were added to the wells and incubated at room temperature for 1 hour. The plates were washed three times with water, and 25 µl of 10 ng/ml goat anti-human IgG (Face fragment specific) conjugated with horseradish peroxidase (Jackson ImmunoResearch, 109-035-008) was added to each well. The plates were incubated at room temperature for 1 hour, washed three times with 80 µl of water, and 25 µl of TMB substrate (Sigma Aldrich, T-0440) was added to each well. After approximately 7 minutes, the reaction was terminated by adding 25 µl of 0.18 M orthophosphate to each well, and the absorbance was read at 450 nm. The DNA and insulin binding score was calculated as the ratio of the ELISA signal of the 10 µg/ml antibody to the signal of the well containing buffer instead of the primary antibody. A score less than 8 was considered ideal and represented low multireactivity.

AC-SINS assay to measure self-association tendency

AC-SINS (Affinity Capture Self-interaction Nanoparticle Spectroscopy) assays were normalized in 384-well format on a Perkin-Elmer Janus liquid handling robot. 20 nm gold nanoparticles (Ted Pella, Inc., #15705) were coated with a mixture of 80% goat anti-human Fc (Jackson Immuno Research Laboratories, Inc. #109-005-098) and 20% nonspecific goat polyclonal antibody (Jackson Immuno Research Laboratories, Inc. #005-000-003), which was replaced with buffer at 20 mM sodium acetate pH 4.3 and diluted to 0.4 mg/mL. After incubation at room temperature for one hour, unoccupied sites on the gold nanoparticles were blocked with thiolized polyethylene glycol (2 kDa). The coated nanoparticles were then concentrated 10-fold using a syringe filter, and 10 µl was added to 100 µl of 0.05 mg/ml mAb in PBS pH 7.2. The coated nanoparticles and target antibody were incubated in 96-well polypropylene plates for 2 hours, and then transferred to 384-well polystyrene plates and read on a Tecan M1000 spectrophotometer. Absorbance was read from 450–650 nm in 2 nm increments, and a Microsoft Excel macro was used to identify the maximum absorbance, smooth the data, and fit the data using a second-order polynomial. The antibody AC-SINS score was determined by subtracting the smoothed maximum absorbance of the mean blank (PBS buffer only) from the smoothed maximum absorbance of the antibody sample. A score less than 8 was considered ideal and represented low self-association.

Evaluation of the lead CD3 antibody 2B5 variant in a bispecific Fc form

A comprehensive analysis of data from binding assays, AC-SINS, and multireactivity helped narrow down the binding variants to five. These five variants were reformatted into bispecific biantibody-Fc molecules using antitumor sequences and standard cloning strategies well-known in the art. These bispecific antibodies were transiently expressed and purified using standard techniques well-known in the art. Binding of the resulting bispecific antibodies to jurakat cells was tested, in vitro cytotoxicity was tested using a T-cell retargeting assay, multireactivity was tested, and thermostability was tested using DSC and AC-SINS. Of all five bispecific antibodies, only one, carrying the CD3 antibody 2B5 variant 2B5v6, showed potent binding and therefore potent cytotoxicity. This variant also showed reduced multireactivity, as evaluated by DNA and insulin binding and AC-SINS scores. The 2B5v6 bispecific antibody scored less than 8, compared to a score greater than 8 for the H2B4 bispecific antibody. Based on these results, 2B5v6 was selected as the lead optimized 2B5 variant for further analysis.

Example 7

Evaluation of shear in CDR-H2

When the bispecific antibodies were expressed in CHO cells but not in HEK293 cells, cleavage was observed in the VH CDR2 (SEQ ID NO: 16) at position R53 in the CD3 antibodies (H2B4 and 2B5). To mitigate the risk of this adverse effect and ensure better homogeneity of the product at manufacturing scale, care should be taken to select for mutations in the cleavage region, as described in previous examples, as well as other mutations to mitigate the risk of other adverse effects. To evaluate the cleavage, simply, bispecific antibody-Fc bispecific antibodies were generated using the variable domain sequence of the negative control antibody and CD3 antibody 2B5 or CD3 antibody H2B4 or CD3 antibody 2B5v6. All bispecific antibody-Fc molecules were prepared using a similar schematic diagram of expression and purification processes well known in the art. CHOSSI cell lines were generated using standard procedures. Two strands encoding a bispecific antibody were cloned into the mammalian expression vector pRY19-GA-Q, which contains a dual promoter and a recombinant site for unit site integration (SSI) (Zhang, L. et al. Biotechnology Progress. 2015; 31:1645-1656) into the CHO cell genome. The resulting plasmid, containing the target gene, was transfected into 10e9 CHO-K1 SV cells via electroporation using the pRY19 vector, followed by positive and negative selection pressures using hygromycin-B and ganciclovir. These CHO conjugates underwent a 3-week recovery period. When observed viability exceeded 90%, a cell library was generated for future work. The established CHO conjugates were then subjected to a 12-day fed-batch platform expression process in 1L of CD-CHO medium (Thermo Fisher Scientific, Waltham, MA). The day 12 culture harvest was captured on a Protein A column and eluted at low pH. The proA conjugate was immediately neutralized to pH 8.1. The purified protein was evaluated using capillary gel electrophoresis (cGE) according to standard protocols well-known in the art. As shown in Table 12 and Figure 7, the 2B5 bispecific antibody showed significantly reduced shearing compared to H2B4, and 2B5v6 showed no shearing while H2B4 showed significantly increased shearing or fragmentation. Table 12 shows the results of capillary gel electrophoresis demonstrating fragmentation of the CD3-GUCY2c bispecific antibody. POI% indicates the target peak.

Table 12

Reduced cGE sample	%POI	%Fragmentation
			Negative control - H2B4	88.8	11.2
Negative control - 2B5	98.9	1.1
			Negative control - 2B5v6	99.6	0.4

Example 8

GUCY2c-2B5v6 Bispecific Antibody: Expression and Purification of Chimeric CD3-GUCY2c Bispecific Antibody Using CD3 Antibody 2B5v6

Complementary construct pairs (12.5 µg per strand) were co-transfected into 25 mL of logarithmic-phase culture containing 1 million cells/mL HEK293 cells using the ExpiFectamine™ 293 Transfection Kit (Life Technologies). Twenty-four hours post-transfection, ExpiFectamine Transfection Enhancer was added, and cells were allowed to grow for an additional 4–5 days before harvest. The used culture was then collected, centrifuged to remove cell debris, and passed through a 20 µm filter.

The clarified conditioned medium containing the GUCY2c T cell bispecific antibody was then purified using protein A affinity chromatography. Samples were loaded onto 0.45 mL microcolumns pre-loaded with MabSelect SuRe protein A resin (GE Healthcare) using a liquid delivery device (Tecan). The bound protein was washed with PBS pH 7.2, eluted with 20 mM citric acid and 150 mM sodium chloride pH 3.5, and neutralized with 2 M tris at pH 8.0. The sample was then desalted to PBS pH 7.2 using G25 Sephadex drip columns (GE Healthcare) according to the manufacturer's instructions. The purity of the purified protein was analyzed on an Aglient 1200 HPLC system using analytical size exclusion chromatography (TOSOH) with a Mab HTP column, following the manufacturer's protocol. Concentrations were determined using a microspectrophotometer (Trinean) at OD 280 nm.

The stability of bispecific antibodies was evaluated using differential scanning calorimetry.

The protein was diluted to 0.6 mg/mL in phosphate-buffered saline (PBS) solution at a volume of 400 µL. PBS was used as a buffer blank in the reference chamber. PBS contained 137 mM NaCl, 2.7 mM KCl, 8.1 mM Na₂HPO₄, and 1.47 mM KH₂PO₄ at pH 7.2. Samples were dispensed into the sample trays of a MicroCal VP-Capillary DSC using an autosampler (Malvern Instruments Ltd, Malvern, UK). Samples were equilibrated at 10 °C for 5 min and then scanned at 100 °C per hour up to 110 °C. A 16-second filtration period was selected. The raw data were baseline corrected and the protein concentration was normalized. The data were fitted to the MN2-State model using Origin software 7.0 (OriginLab Corporation, Northampton, MA) with an appropriate number of transformations. As shown in Figure 8, the GUCY2c-2B5v6 bispecific antibody exhibits excellent thermal stability, with _Tm1 being 70℃.

Binding with naïve T cells

Using a BD LSRII Fortessa analyzer, purified CD3 bispecific antibody was titrated to bind to the cell surface of naïve human CD3 on virgin human T cells purified from fresh human peripheral blood mononuclear cells (PBMCs) using standard flow cytometry methods known in the art. As shown in Figure 9, the GUCY2c-2B5v6 bispecific antibody bound very well to T cells, with a binding EC50 of 35.44 nM.

Evaluation of T cell activity mediated by bispecificity

Human PBMCs were isolated from healthy donor blood using Histopaque-177 (Sigma). Naïve T cells (negatively selected T cells) were isolated from PBMCs using a T cell enrichment kit from Stem Cell Technologies. T84 human tumor cells expressing GUCY2c, transfected with a luciferase expression construct, were treated with serially diluted GUCY2c bispecific antibody or a negative control CD3 bispecific antibody, along with the T cells. The effector cell to target cell ratio (E:T) was 10:1 or 5:1. T cells and tumor cells were incubated at 37°C for 48 hours, followed by measurement of luciferase signal using neolite reagent (Perkin Elmer). EC50 values were calculated in Graphpad PRISM using a four-parameter nonlinear regression analysis. Target-negative HCT116 or HT29 cells did not show any GUCY2c bispecificity-mediated cytotoxicity. As shown in Figure 10, GUCY2c-2B5v6 bispecificity demonstrated potent cytotoxicity in the T cell retargeting assay.

T84 cells were treated with T cells (E:T ratio 5:1) and different doses of GUCY2c-1608 (GUCY2c-2B5v6). Supernatants were collected at different time points (24 h, 48 h, 96 h) and analyzed using multiplex Luminex assays according to the manufacturer's guidelines and read on a Luminex 200 with xPONENT software. The measured cytokines were human IFN-γ, IL10, IL2, IL4, IL6, and TNF-α. Figure 11 shows the upregulation of (11A) IFN-γ, (11B) IL10, (11C) IL2, (11D) IL4, (11E) IL6, and (11F) TNF-α cytokines after T cell activation with GUCY2c-1608.

In vivo assessment of bispecific activity mediated by GUCY2c-2B5v6

In vivo efficacy studies were conducted using an adoptive metastasis model. Tumor cell lines/patient-derived xenograft fragments were implanted into NSG mice and staged to approximately 200 ^mm³ . The bispecific compound was administered intravenously to the mice (unless otherwise noted). Twenty-four hours after bispecific antibody administration, two million activated and expanded T cells were administered intravenously (using a T cell expansion and activation kit from Miltenyi Biotec). The bispecific antibody was administered weekly. As shown in Figures 12 and 13, the GUCY2c-2B5v6 (GUCY2c-1608) bispecific antibody demonstrated excellent durable tumor regression in both the PDX CRX-11201 and LS1034 cell lines, representing colorectal cancer, in vivo models.

Example 9

Generation and cytotoxicity of FLT3-CD3 bispecific antibody and FLT3-2B5v6 bispecific IgG using an optimized CD3 antibody (2B5v6).

Human FLT3 antibody (xFLT3) and human CD3 antibody (2B5v6) are expressed as engineered human IgG2dA_D265A with EEEE on one arm and RRRR on the other arm, respectively, for bispecific exchange at positions 223, 225, and 228 (e.g., (C223E or C223R), (E225E or E225R), and (P228E or P228R)) in the hinge region of human IgG2 and at positions 409 or 368 (e.g., K409R or L368E (EU numbering scheme)) in the CH3 region. The antibodies also have a mutation from D to A at position 265 (EU numbering scheme).

Individual antibody arms were purified separately using Protein A resin. Then, 1 mg/mL of purified xFLT3 antibody was exchanged with 1 mg/mL of purified 2B5v6 antibody in the presence of 2 mM reduced glutathione. The exchange reaction was carried out at 37°C for 16 h. Then, 2 mM oxidized glutathione was added to the protein, and the mixture was incubated at 37°C for 30 min. The mixture was then diluted 1:5 to 20 mM MES pH 5.4 buffer without NaCl. The diluted sample was loaded onto a MonoS ion exchange column and washed with 20 mM MES pH 5.4 + 25 mM NaCl buffer. The bispecific antibody xFLT3-2B5v6 was then eluted with a 0–500 mM NaCl gradient in 20 mM MES pH 5.4 buffer. Finally, the bispecific antibody was dialyzed into phosphate-buffered saline (PBS) for cell-based assays. For the cytotoxicity assay, unstimulated human T cells and luciferase-expressing target cells (Eol-1 or MV411) were co-incubated in 96-well U-shaped plates in 200 μl RPMI + 10% FBS to determine the ratio. xFLT3-2B5v6 bispecific antibody was serially diluted in PBS and added to each well. After 24 hours, 100 μl of cell suspension was mixed with 100 μl of One-GLO reagent in 96-well white-walled plates, and luminescence was measured on an Envision photometer. The cytotoxicity value was determined using the following formula:

%Cytotoxicity = [1 - (RLU sample) / RLU target cells only)] × 100RLU: relative light units

As shown in Figures 14A and 14B, when tested on two cell lines expressing higher copies of FLT3 (EOL-1) and lower copies (MV4-11), the xFLT3-2B5v6 bispecific antibody containing the optimized CD3 antibody (2B5v6) appears to be highly potent when presented as an alternative to the bispecific IgG form.

Table 13 shows the light and heavy chain sequences of FLT3 (EOL-1) and the light and heavy chain sequences of the 2B5v6 bispecific antibody.

Table 13

Example 10

T-cell epitopes of the 2B5v6 bispecific antibody

Optimized analysis of potential T-cell epitopes in clone 2B5v6 was performed to de-risk the possibility of immunogenicity. This analysis was performed using the Epivax tool for predicting potential T-cell epitopes on a computer chip. This tool predicts the number of times a 9-mer framework peptide binds to any of eight different HLA alleles. Hit was defined as those in the top 5% with a Z-score in the top 1%. Hit to four or more alleles or one strong hit was considered a predicted T-cell epitope. Based on these criteria, three predicted non-germal T-cell epitopes in CDRH2, two predicted germal epitopes in CDRL1, and one predicted germal epitope in L2 were identified (Table 15). In addition to the scores of individual peptides, the overall risk of the sequence was defined by the total number of hits or the T-regitope-adjusted Epivax score, which had been optimized to correlate with the prevalence of clinical ADA. Here, for these predicted values, the lower the score, the lower the expected risk of ADA. The VH domain had 70 hits and a score of -42.82, while the VL domain had 60 hits and a score of -23.79.

To reduce the risk of immunogenicity, mutations that would decrease the overall score of this clone while removing hits and ideally predicted T-cell epitopes were identified. Any possible single mutant variations of CDR-L1, CDR-L2, and CDR-H2 were considered to identify mutations that would reduce hits, decrease the overall t-regitope-adjusted EpiMatrix score, and remove T-cell epitopes. Since these potential mutations could also affect binding and stability, care was taken to consider the predicted stability and affinity data described above, as well as to identify tolerant mutations. Tables 14A, 14B, and 14C show mutations designed to reduce the immunogenicity risk of 2B5v6. Additionally, mutations predicted as shown in SEQ ID NOs 44-47, 51-57, 60, and 62 reduced the number of T-cell epitopes.

Table 14A

Table 14B

Table 14C

Table 15 shows the regions of the variable domains of a CD3 antibody (2B5v6) that has T-cell epitopes and poses a potential risk of immunogenicity.

Table 15

All mutants were generated as monovalent IgG antibodies using standard transient HEK293 expression, followed by purification using a protein A column. The binding of the resulting clones to CD3-expressing Jurkat cells was tested using FACS, and the likelihood of self-association was tested using AC-SINS as described in Example 7 above. As shown in Table 16, five molecules with improved epivax scores and reduced T-cell epitopes, while retaining affinity for CD3, were selected for further testing as biantibody-Fc variants. The 2B5v6 variant with improved epivax scores was compared to the parental 2B5v6. Results showed comparable binding to CD3-expressing Jurkat cells and AC-SINS scores.

Table 16

Of the mutants tested, several exhibited lower binding to Jurkat cells compared to 2B5v6. Among these mutants, those with all T-cell epitopes eliminated and improved epivax scores were further advanced to generate bispecific antibody-Fc molecules in which the GUCY2c antibody serves as the other binding domain. Table 17 shows the epivax scores, and Figure 15 shows the reduced binding of these mutants to Jurkat cells. The low-affinity CD3 antibody 2B5v6 variant showed an improved epivax score.

Table 17

CD3 antibody variants	Epivax Score (tReg - Adjusted)
		2B5v6-0598	-65.15
2B5v6-0623	-56.74

Example 11

Three anti-CD3 variants showed improved affinity and cytotoxicity in the GUCY2C-CD3 bispecific antibody.

Using the standard cloning, expression, and purification techniques described above, three anti-CD3 variants derived from 2B5v6, 2B5-1038 (SEQ ID NO: 4 and 9), 2B5-1039 (SEQ ID NO: 4 and 87), and 2B5-1040 (SEQ ID NO: 4 and 89), were reformulated into bispecific biantibody-Fc molecules paired with the antitumor GUCY2c antibody sequence, one of the configurations shown in Figure 1. In vitro cytotoxicity of the resulting bispecific antibodies was assessed using T-cell retargeting assays as shown in Table 18, binding affinity was tested using surface plasmon resonance (SPR), and nonspecificity was tested using AC-SINs. Immunogenic risk was evaluated using the Epivax Tool, a computer-based assay for potential T-cell epitopes. Complete mass spectrometry data indicated that all bispecific antibodies were correctly paired and 100% heterodimer. Table 18 shows that all three bispecific antibodies (1) were approximately 2-fold more potent in in vitro cytotoxicity assays compared to the control bispecific GUCY2C-1608 paired with anti-CD3 2B5v6 (Fig. 6); (2) had binding affinity for recombinant human CD3 via SPR, consistent with cytotoxic activity; (3) had improved immunogenicity scores compared to the control; and (4) AC-SINs scores remained the same as the control bispecific GUCY2C-1608. These results indicate that anti-GUCY2c bispecific antibodies with different CD3 variants recruit naïve human T cells to induce cell killing in T84 tumor cells (Fig. 16). The three anti-CD3 variants derived from 2B5v6 showed approximately 2-fold greater potency compared to the control bispecific GUCY2C-1608.

Table 18

Bispecific

CD3 variant

Epivax Score (tReg - Adjusted)

MassSpec (Complete)

AC-SINs score

In vitro cytotoxicity (nM)

KD (nM, SPR)

GUCY2C-1608

2B5-0542

-33.7

100% heterodimer

3

0.772

127.88 ± 1.63

GUCY2C-1678

2B5-1038

-52.61

100% heterodimer

2

0.247

63.55±0.49

GUCY2C-1679

2B5-1040

-47.58

100% heterodimer

3

0.194

72.02 ± 0.36

GUCY2C-1680

2B5-1039

-53.51

100% heterodimer

3

0.385

78.57±0.9

The sequences of the bispecific antibodies are shown in Table 19.

Table 19

sequence list

This application was filed electronically via EFS-Web and includes a sequence list submitted electronically in .txt format. The .txt file contains a sequence list titled "PC72375-PRV2_SequenceListing_ST25_05132019.txt", created on May 13, 2019, and is 100 KB in size. The sequence list contained in this .txt file is part of this specification and is incorporated herein by reference in its entirety.

Invention Field

This invention relates to CD3-specific antibodies and their antigen-binding fragments, as well as methods of using them. The invention also relates to multispecific antibodies that bind to both CD3 and target antigens, and methods of using them.

Background of the Invention

Differentiation cluster 3 (CD3) is a homodimeric or heterodimeric antigen expressed on T cells that associates with the T cell receptor complex (TCR) and is required for T cell activation. T cell activation is a complex phenomenon that depends on the involvement of multiple cell surface molecules expressed on a population of reactive T cells. Functional CD3 is formed by the association of two of four different chains: ε, ζ, δ, and γ. For example, the antigen-specific T cell receptor (TCR) is composed of a disulfide-linked heterodimer containing two clonal, monolithic membrane glycoprotein chains, alpha and beta (α and β), or gamma and delta (γ and δ), non-covalently associated with a complex of a low molecular weight invariant protein (usually designated CD3). CD3 dimer arrangements also include γ/ε, δ/ε, and ζ/ζ.

Antibodies targeting CD3 have been shown to accumulate CD3 on T cells, thereby inducing T cell activation in a manner similar to the binding of MHC molecules loaded with peptides to the TCR. Therefore, CD3 antibodies can be used for therapeutic purposes involving T cell activation.

Invention Overview

This invention provides antibodies that bind to CD3 and compositions (such as pharmaceutical compositions) containing one or more CD3 antibodies. Antibodies according to this aspect of the invention are particularly useful for targeting CD3-expressing T cells and for stimulating T cell activation, for example, in cases where T cell-mediated killing is beneficial or desired. The CD3 antibodies of this invention may be included as part of a multispecific antibody (e.g., a bispecific antibody) that directs CD3-mediated T cell activation to specific cell types, such as tumor cells or infectious pathogens.

In one aspect, the present invention provides an antibody that specifically binds to CD3, wherein the antibody comprises: (a) a heavy chain variable (VH) region comprising VH complementarity-determining region 1 (VH CDR1), VH complementarity-determining region 2 (VH CDR2), and VH complementarity-determining region 3 (VH CDR3) of the VH sequence shown in SEQ ID NO: 2, 4, 5, 7, 10, 12, or 35; and/or (b) a light chain variable (VL) region comprising VL complementarity-determining region 1 (VL CDR1), VL complementarity-determining region 2 (VL CDR2), and VL complementarity-determining region 3 (VL CDR3) of the VL sequence shown in SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87, or 89.

In one embodiment of this aspect, the CD3 antibody comprises: (a) a VH region comprising (i) a VH CDR1 comprising the sequence of SEQ ID NO: 13, 14 or 15; (ii) a VH CDR2 comprising the sequence of SEQ ID NO: 16, 17, 19, 20, 21, 22, 23 or 24; and (iii) a VH CDR3 comprising the sequence of SEQ ID NO: 18 or 25; and/or (b) a VL region comprising (i) a VL CDR1 comprising the sequence of SEQ ID NO: 26, 29, 30, 32, 91 or 92; (ii) a VL CDR2 comprising the sequence of SEQ ID NO: 27 or 31; and (iii) a VL CDR3 comprising the sequence of SEQ ID NO: 28 or 33.

In some such embodiments, the present invention provides an antibody wherein: (a) the VH region contains a sequence of SEQ ID NO: 2, 4, 5, 7, 10, 12 or 35; and/or (b) the VL region contains a sequence of SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87 or 89.

In other embodiments, the antibody comprises a VL/VH amino acid sequence selected from the following: SEQ ID NO: 1 and 2; SEQ ID NO: 3 and 2; SEQ ID NO: 3 and 4; SEQ ID NO: 3 and 5; SEQ ID NO: 6 and 7; SEQ ID NO: 8 and 7; SEQ ID NO: 6 and 4; SEQ ID NO: 8 and 4; SEQ ID NO: 9 and 10; SEQ ID NO: 9 and 7; SEQ ID NO: 11 and 12; SEQ ID NO: 34 and 35; SEQ ID NO: 9 and 4; SEQ ID NO: 87 and 4; and SEQ ID NO: 89 and 4.

In some of the aforementioned embodiments, the antibody is selected from: Fab, Fab fragment, F(ab) ₂ fragment, Fv fragment, single-chain Fv fragment, disulfide-stabilized Fv fragment, single-chain antibody, monoclonal antibody, chimeric antibody, bispecific antibody, trispecific antibody, multispecific antibody, bispecific heterodimeric biantibody, bispecific heterodimeric IgG, polyclonal antibody, labeled antibody, humanized antibody, human antibody and fragment thereof.

In one specific embodiment, the antibody is a bispecific antibody. In some such embodiments, the bispecific antibody specifically binds to tumor antigens.

In some embodiments, the antibody of the present invention further comprises a human or humanized VH framework and a human or humanized VL framework. In some such embodiments, the antibody is a humanized antibody.

The present invention further provides pharmaceutical compositions comprising a CD3 antibody as further disclosed herein and a pharmaceutically acceptable carrier.

In another aspect, the present invention provides a method for modulating a T-cell-mediated immune response in a subject with such need, comprising administering an effective amount of an antibody or pharmaceutical composition as disclosed herein to the subject such that the immune response in the subject is modulated.

In one aspect, the present invention provides a method for inhibiting the growth of tumor cells in a subject, comprising administering to the subject, in an amount that effectively inhibits the growth of tumor cells, a bispecific antibody, a trispecific antibody, a multispecific antibody, a bispecific heterodimeric antibody, or a bispecific heterodimeric IgG as disclosed herein.

In some such embodiments, the tumor cells are cancer cells. In specific embodiments, the cancer is selected from: breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer, bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and cancers of the digestive system.

In one embodiment, the cancers of the digestive system are selected from: esophageal cancer, gastric cancer, small bowel cancer, colon cancer, rectal cancer, anal cancer, liver cancer, gallbladder cancer, appendix cancer, bile duct cancer, and pancreatic cancer.

In one embodiment, the present invention provides an antibody or a pharmaceutical composition comprising the antibody for use in a therapy.

In another embodiment, the present invention provides the use of the antibody of the present invention in the preparation of a medicament for use in a therapy. In some such embodiments, the therapy comprises activating a cytolytic T-cell response.

In one aspect, the present invention provides a polynucleotide comprising a nucleotide sequence encoding an antibody as disclosed herein. In another aspect, the present invention provides a vector comprising this polynucleotide.

In another aspect, the present invention provides host cells comprising vectors as disclosed herein. In some such embodiments, the host cells recombinantly produce antibodies as disclosed herein. In specific embodiments, the host cells are selected from bacterial cell lines, mammalian cell lines, insect cell lines, and yeast cell lines. In one specific embodiment, the mammalian cell line is a CHO cell line.

In one aspect, the present invention provides a method for generating antibodies as disclosed herein, comprising: culturing host cells under conditions that result in the generation of antibodies as disclosed herein, and purifying said antibodies from the culture supernatant.

In another aspect, the present invention provides the use of CD3 antibodies, polynucleotides, vectors or host cells as disclosed herein in the preparation of medicaments for treating diseases.

In one aspect, the present invention provides an antibody or pharmaceutical composition for use in a method of modulating a T-cell-mediated immune response in a subject with such a need. In a specific embodiment, the present invention provides an antibody or pharmaceutical composition for use in a method of inhibiting the growth of tumor cells in a subject.

In another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, optionally wherein said cancer is selected from: breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer, bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and cancers of the digestive system.

In yet another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, wherein a T-cell-mediated immune response is modulated or wherein the growth of tumor cells is inhibited.

Other embodiments will become apparent from the detailed summary. When aspects or embodiments of the invention are described in terms of the Markush group or other alternative groupings, the invention covers not only the entire group listed as a whole, but also each member of that group individually and all possible subgroups of the main group, as well as the main group in which one or more members are not present. The invention also contemplates the explicit exclusion of any one or more members of the claimed invention.

Brief description of the figure/attachment

Figure 1 provides a schematic diagram of the architecture of a bispecific antibody having a first heterodimer promoting domain and a second heterodimer promoting domain, the bispecific antibody comprising an Fc chain optimized for association via mortar association.

Figure 2 depicts the results of the differential scanning calorimetry temperature recording, showing the heat capacity of the GUCY2c-H2B4 and GUCY2c-2B5 bispecific antibodies as a function of temperature.

Figure 3 depicts the binding of the bispecific antibodies GUCY2c-H2B4 (GUCY2c-0247) and GUCY2c-2B5 (GUCY2c-0250) to naïve human T cells as determined by flow cytometry.

Figure 4 depicts the results of in vitro cytotoxicity mediated by the bispecific antibodies GUCY2c-H2B4 (GUCY2c-0247) and GUCY2c-2B5 (GUCY2c-0250) using cell survival data.

Figure 5 depicts the potential energy surface (PES) of the H2B4 Fv region from an X-ray crystal structure. This includes three views highlighting the VH region, the VH/VL interface, and the VL region. The CDR (Crystal Radiata Diffraction) with excess positive charge is indicated for reference. Dark black represents positive charge and white represents negative charge.

Figure 6 depicts the spatial aggregation tendency (SAP) surface of the Fv region of H2B4. Regions with concentrated hydrophobic residues are marked.

Figure 7 depicts the heterogeneity among bispecific antibodies containing CD3 antibody-binding domains H2B4, 2B5, and 2B5v6 using capillary gel electrophoresis.

Figure 8 depicts the temperature record using differential scanning calorimetry (DSC), showing the heat capacity of the GUCY2c-1608 (GUCY2c-2B5v6) bispecific antibody as a function of temperature.

Figure 9 depicts the binding of the GUCY2c-1608 (GUCY2c-2B5v6) bispecific antibody to naive T cells.

Figure 10 depicts cell survival data for in vitro cytotoxicity mediated by the bispecific antibody GUCY2c-1608 (GUCY2c-2B5v6).

Figures 11A to 11F depict in vitro cytokine release measurements of the GUCY2c-1608 (GUCY2c-2B5v6) bispecific antibody. The bispecific antibody recruited naïve human T cells to GUCY2c-expressing T84 cells to induce in vitro cytokine release. Luminex-based assays showed upregulation of human IFN-γ (Figure 11A), IL10 (Figure 11B), IL2 (Figure 11C), IL4 (Figure 11D), IL6 (Figure 11E), and TNF-α (Figure 11F).

Figure 12 illustrates the dose-dependent tumor growth inhibition of the bispecific antibody GUCY2c-1608 (GUCY2c-2B5v6) in the patient-derived xenograft PDX-CRX-11201 of colorectal cancer in the adoptive metastasis system.

Figure 13 illustrates the dose-dependent tumor growth inhibition of the bispecific antibody GUCY2C-1608 (GUCY2C-2B5v6) in the xenograft LS1034 colorectal cancer cell line in the adoptive metastasis system.

Figure 14A depicts the results of cell viability data, showing the in vitro cytotoxicity mediated by the FLT3-2B5v6 bispecific antibody using Flt3-expressing MV4-11 cells at different effector-to-target (E:T) ratios. Figure 14B depicts the results of cell viability data, showing the in vitro cytotoxicity mediated by the FLT3-2B5v6 bispecific antibody using Flt-3-expressing EOL-1 cells at different effector-to-target (E:T) ratios.

Figure 15 depicts the binding of the low-affinity 2B5v6 variant to Jurkat cells expressing CD3.

Figure 16 depicts a graph summarizing the findings, showing the cytotoxicity mediated by anti-GUCY2C bispecific T cells with anti-CD3 variants.

Detailed Explanation

The present invention disclosed herein provides antibodies that specifically bind to CD3 (e.g., human CD3), including full-length antibodies or antigen-binding fragments thereof. The invention also provides polynucleotides encoding these antibodies, compositions comprising these antibodies, and methods for preparing and using these antibodies. The invention further provides methods for treating conditions using antibodies as described herein.

General Technology

Unless otherwise indicated, the practice of this invention will employ conventional techniques of molecular biology (including recombinant techniques), microbiology, cell biology, biochemistry, and immunology, which are within the scope of the art. Such techniques are fully explained in the literature such as *Molecular Cloning: A Laboratory Manual, 2nd Edition* (Sambrook et al., 1989); *Cold Spring Harbor Press*; *Oligonucleotide Synthesis* (M.J. Gait, ed., 1984); *Methods in Molecular Biology*, *Humana Press*; *Cell Biology: A Laboratory Notebook* (J.E. Cellis, ed., 1998); *Academic Press*; *Animal Cell Culture* (R.I. Freshney, ed., 1987); *Introduction to Cell and Tissue Culture* (J.P.). Mather and P.E. Roberts, 1998) Plenum Press; Cell and Tissue Culture: LaboratoryProcedures (A.Doyle, J.B. Griffiths, and D.G. Newell, eds., 1993-1998) J. Wileyand Sons; Methods in Enzymolog y (Academic Press, Inc.); Handbook of Experimental Immunology (D.M. Weir and C.C. Blackwell, Eds.); Gene Transfer Vectors for Mammalian Cells (J.M. Miller and M.P. Calos, Eds., 1987); Current Protoc Mullis in Molecular Biology (F.M. Ausubel et al., eds., 1987); PCR: The Polymerase Chain Reaction (Mullis et al., eds., 1994); Current Protocols in Immunology (J.E. Colligan et al., eds., 1991); Short Protocols in Molecular Biology (Wiley and Sons, 1999); Immunobiology (C.A. Janeway and P. Travers, 1997); Antibodies (P. Finch, 1997); Antibodies: a practical approach (D. Catty., ed., IRL Press, 19...). 88-1989); Monoclonal Antibodies: A Practical Approach (P. Shepherd and C. Dean, eds., Oxford University Press, 2000); Using Antibodies: A Laboratory Manual (E. Harlow and D. Lane (Cold Spring Harbor Laboratory Press, 1999); The Antibodies (M. Zanetti and J.D. Capra, eds., Harwood Academic Publishers, 1995). In some cases, terms having their usual meaning are defined herein for clarity and/or for the purpose of preparing references, and the inclusion of such definitions herein is not necessarily construed as indicating a material difference from definitions commonly understood in the art.

The invention will now be described in detail by reference using the following definitions and examples. All patents and publications mentioned herein (including all sequences disclosed in such patents and publications) are expressly incorporated by reference.

definition

Generally, unless otherwise defined, all terms, symbols, and other scientific or terminological terms used herein are intended to have the meaning commonly understood by one of ordinary skill in the art to which this invention relates. For example, the term "and/or" as used herein, such as in phrases like "A and/or B," is intended to include both A and B; A or B; A (alone); and B (alone). Similarly, the term "and/or" as used herein, such as in phrases like "A, B, and/or C," is intended to cover each of the following embodiments: A, B, and C; A, B, or C; A or C; A or B; B or C; A and C; A and B; B and C; A (alone); B (alone); and C (alone).

As used herein, unless the context clearly indicates otherwise, the singular forms “a”, “an” and “the” include their respective multiple indicators.

As used in this article, the numerical range includes the numerical values that define the range.

The terms “about,” “approximately,” etc., when preceding a list of numerical values or ranges, independently refer to each individual value in that list or range as if each individual value immediately followed the term. The term implies that the value referred to by the term is the same as, close to, or similar to. For example, in some embodiments, “about” or “approximately” for a particular value may indicate 99%, 95%, or 90% of that value. As an example, the expression “about 100” includes 99 and 101 and all values between them (e.g., 99.1, 99.2, 99.3, 99.4, 99.5, etc.). As another example, if the temperature is 70°C, “about” or “approximately” 70°C may be equal to 69°C, 66°C, or 63°C. It should be understood that these are merely examples.

As used herein, nucleic acids are written from left to right in a 5' to 3' orientation; amino acid sequences are written from left to right in an amino-to-carboxyl orientation. For definitions and terminology in this art, practitioners particularly refer to Sambrook et al., 1989 and Ausubel FM et al., 1993. It should be understood that the invention is not limited to the specific methods, schemes, and reagents described, as these are subject to variation.

The terms “polypeptide,” “oligopeptide,” “peptide,” and “protein” are used interchangeably herein to refer to an amino acid chain of any length, preferably relatively short (e.g., 10-100 amino acids). This chain may be linear or branched, may contain modified amino acids, and/or may be interrupted by non-amino acid components. The term also covers amino acid chains that have been modified naturally or through intervention; for example, disulfide bond formation, glycosylation, esterification, acetylation, phosphorylation, or any other operation or modification, such as conjugation with a labeled component. This definition also includes, for example, polypeptides containing one or more amino acid analogs (including, for example, non-natural amino acids) and other modifications known in the art. It should be understood that polypeptides may exist as single chains or associated chains. Preferably, mammalian polypeptides (polypeptides originally derived from mammalian organisms) are used, more preferably those secreted directly into the culture medium.

An "antibody" is an immunoglobulin molecule that specifically binds to a target, such as carbohydrates, polynucleotides, lipids, peptides, etc., through at least one antigen recognition site located in the variable region of the immunoglobulin molecule. As used herein, the term encompasses polyclonal antibodies, monoclonal antibodies, chimeric antibodies, bispecific antibodies, dual-specific antibodies, bifunctional antibodies, trispecific antibodies, multispecific antibodies, bispecific heterodimeric biantibodies, bispecific heterodimeric IgG, labeled antibodies, humanized antibodies, human antibodies and their fragments (such as Fab, Fab', F(ab') ₂ , Fv), single-chain (ScFv) and domain antibodies (including, for example, antibodies from sharks and camels), fusion proteins containing antibodies, and any other modified conformation of an immunoglobulin molecule containing an antigen recognition site. Antibodies include any class of antibodies, such as IgG, IgA, or IgM (or their subclasses), and the antibody need not be of any particular class. Immunoglobulins can be designated as different classes depending on the amino acid sequence of their heavy chain constant region. There are five main classes of immunoglobulins: IgA, IgD, IgE, IgG, and IgM, and several of these can be further subdivided into subclasses (isotypes), such as IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2. The heavy chain constant regions corresponding to different classes of immunoglobulins are designated α, δ, ε, γ, and μ, respectively. The subunit structures and three-dimensional conformations of different classes of immunoglobulins are well known. This invention also includes "antibody analogs," other scaffolds based on non-antibody molecular proteins, such as fusion proteins and/or immunoconjugates using CDRs to provide specific antigen binding. The antibodies of this invention can be derived from any species, including but not limited to mice, humans, camels, llamas, fish, sharks, goats, rabbits, chickens, horses, and cattle.

The term "antibody" further includes immunoglobulin molecules comprising four polypeptide chains (two heavy chains (H) and two light chains (L)) linked by disulfide bonds, as well as their multimers (e.g., IgM). Each heavy chain comprises a heavy chain variable region (abbreviated herein as HC VR or VH) and a heavy chain constant region. The "variable region" of an antibody refers to the variable region of either the antibody light chain alone or in combination with the variable region of the antibody heavy chain. The heavy chain constant region comprises three domains: CH1, CH2, and CH3. The CH1 and CH2 domains are linked by hinge regions. Each light chain comprises a light chain variable region (abbreviated herein as LC VR or VL) and a light chain constant region. The light chain constant region comprises one domain (CL1). The VH and VL regions can be further subdivided into hypervariable regions (called complementarity-determining regions (CDRs)) which are interspersed with more conserved regions (called framework regions (FRs)). Each VH and VL consists of three CDRs and four FRs, arranged in the following order from the amino terminus to the carboxyl terminus: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. In different embodiments of the invention, the FRs of the CD3 antibody (or its antigen-binding portion) may be identical to the human germline sequence, or may be natural or artificially modified. The common amino acid sequence can be defined based on a side-by-side analysis of two or more CDRs. The CDRs in each chain are tightly held together by the FRs and, together with the CDRs from the other chain, contribute to the formation of the antigen-binding site of the antibody.

As used herein, the terms “antigen-binding fragment,” “antibody fragment,” or “antigen-binding portion” refer to one or more segments of an antibody that retain the ability to specifically bind to an antigen. Examples of binding fragments encompassed within the term "antigen-binding fragment" in the term antibody include (i) antibody heavy chain variable domains (VH) and/or antibody light chain variable domains (VL), or VH/VL pairs derived from full-length antibodies or antibody fragments (such as VH and/or VL domains); (ii) Fab fragments, i.e., monovalent fragments consisting of VL, VH, CL, and CH1 domains; (iii) Fab' fragments, which are essentially Fab fragments having a portion of a hinge region (e.g., Fundamental Immunology, edited by Paul, 3rd edition, 1993); (iv) F(ab') ₂ fragments, i.e., bivalent fragments containing two Fab fragments connected by a disulfide bridge at the hinge region; (v) Fd fragments, which consist of VH and CH1 domains; (vi) Fv fragments, which consist of the VL and VH domains of a single arm of the antibody; (vii) Single-chain Fv fragments (scFv), i.e., a single protein chain in which the VL and VH regions pair to form a monovalent molecule (e.g., Bird et al. (1988) Science 242:423-426; and Huston et al. (1988) Proc. Natl. Acad. Sci. USA 85:5879-5883); (viii) disulfide-stabilized Fv fragments (dsFv), i.e., Fv with engineered intermolecular disulfide bonds that stabilize the VH-VL pair; (ix) single variable domain antibody (sdAb or dAb) fragments (e.g., Ward et al. (1989) Nature 341:544-546), which consist of a variable domain of a heavy chain and have no light chain; (x) complementarity-determining regions (CDRs); and any derivative thereof.

As used herein, an antibody's "antigen-binding fragment" may contain at least one variable domain. The variable domain may have any size or amino acid composition and will typically contain at least one CDR adjacent to or framed with one or more framework sequences. As used herein, an "antigen-binding fragment" of an antibody may comprise a homodimer or heterodimer (or other multimer) of any variable region and constant domain configuration that is non-covalently associated with each other and/or non-covalently associated (e.g., via disulfide bonds) with one or more monomeric VH or VL regions. For example, the variable region may be dimer and contain VH-VH, VH-VL, or VL-VL dimers. The configurations of the variable and constant domains that can be found within the antigen-binding fragment of the antibody of the present invention include: VH-CH1; VH-CH2; VH-CH3; VH-CH1-CH2; VH-CH1-CH2-CH3; VH-CH2-CH3; VH-VL-CL, VH-VL-CH1, VH-VL-CH2; VH-CL; VL-CH1; VL-CH2; VL-CH3; VL-CH1-CH2; VL-CH1-CH2-CH3; VL-CH2-CH3; and VL-CL. The variable and constant domains can be directly connected to each other or can be connected via complete or partial hinge or linker regions. The hinge region can consist of at least two (e.g., five, ten, fifteen, twenty, forty, sixty, or more) amino acids, which form flexible or semi-flexible bonds between adjacent variable and/or constant domains in a single polypeptide molecule.

As used herein, a “binding domain” includes any region of a polypeptide (e.g., an antibody) responsible for selectively binding to a target molecule (e.g., an antigen, ligand, receptor, substrate, or inhibitor). Exemplary binding domains include antibody variable regions, receptor-binding domains, ligand-binding domains, and enzyme-catalyzing domains.

As used herein, the term "human receptor framework" is a framework comprising an amino acid sequence of a light chain variable (VL) framework or a heavy chain variable (VH) framework derived from the human immunoglobulin framework or the human common framework as defined below. A human receptor framework "derived from" the human immunoglobulin framework or the human common framework may contain the same amino acid sequence, or may contain amino acid sequence modifications. In some embodiments, the number of amino acid modifications is 10 or less, 9 or less, 8 or less, 7 or less, 6 or less, 5 or less, 4 or less, 3 or less, or 2 or less. In some embodiments, the sequence of the VL human receptor framework is identical to the sequence of the VL human immunoglobulin framework or the human common framework.

As used herein, an “affinity-matured” antibody is an antibody that has one or more modifications in one or more variable regions (including CDR and FR) compared to an unmodified parent antibody, wherein such modifications result in an increased affinity of the antibody for the antigen.

As used herein, the terms “Fc region,” “Fc domain,” “Fc chain,” or similar terms are used to define the C-terminal region of the IgG heavy chain. The Fc region of IgG contains two constant domains, CH2 and CH3. The CH2 domain of the human IgG Fc region typically extends from amino acid 231 to amino acid 340 according to the EU index numbering system, or from amino acid 244 to amino acid 360 according to the Kabat numbering system. The CH3 domain of the human IgG Fc region typically extends from amino acid 341 to amino acid 447 according to the EU index numbering system, or from amino acid 361 to amino acid 478 according to the Kabat numbering system. The CH2 domain of the human IgG Fc region (also known as the “Cγ2” domain) is unique because it does not pair closely with the other domain. Instead, two N-linked branched carbohydrate chains are inserted between the two CH2 domains of intact native IgG. The Fc region may contain native or variant Fc sequences. Although the boundaries of the Fc sequence of the immunoglobulin heavy chain may vary, the human IgG heavy chain Fc sequence is generally defined as an extension from an amino acid residue at approximately position Cys226 or approximately position Pro230 to the carboxyl terminus of said Fc sequence. Unless otherwise specified herein, the amino acid residues in the Fc region or constant region are numbered according to the EU numbering system, also known as the EU index, as described in Kabat et al., Sequences of Proteins of Immunological Interest, 5th edition, Public Health Service, National Institutes of Health, Bethesda, MD, 1991.

In some embodiments, the Fc chain begins in a hinge region immediately upstream of the papain cleavage site and terminates at the C-terminus of the antibody. Therefore, a complete Fc chain comprises at least a hinge domain, a CH2 domain, and a CH3 domain. In some embodiments, the Fc chain comprises at least one of the following: a hinge (e.g., upper, middle, and/or lower hinge region) domain, a CH2 domain, a CH3 domain, a CH4 domain, or a variant, portion, or fragment thereof. In some embodiments, the Fc domain comprises a complete Fc chain (i.e., a hinge domain, a CH2 domain, and a CH3 domain). In some embodiments, the Fc chain comprises a hinge domain (or a portion thereof) fused to a CH3 domain (or a portion thereof). In some embodiments, the Fc chain comprises a CH2 domain (or a portion thereof) fused to a CH3 domain (or a portion thereof). In some embodiments, the Fc chain consists of a CH3 domain or a portion thereof. In some embodiments, the Fc chain consists of a hinge domain (or a portion thereof) and a CH3 domain (or a portion thereof). In some embodiments, the Fc chain consists of a CH2 domain (or a portion thereof) and a CH3 domain. In some embodiments, the Fc chain consists of a hinge domain (or a portion thereof) and a CH2 domain (or a portion thereof). In some embodiments, the Fc chain lacks at least a portion of the CH2 domain (e.g., all or a portion of the CH2 domain). As used herein, Fc chain generally refers to a polypeptide comprising all or a portion of the Fc chain of an immunoglobulin heavy chain. This Fc chain includes, but is not limited to, polypeptides comprising the entire CHI, hinge, CH2, and/or CH3 domains, and fragments of such peptides comprising only, for example, the hinge, CH2, and CH3 domains. Fc chains can be derived from immunoglobulins of any species and/or any subtype, including but not limited to human IgG1, IgG2, IgG3, IgG4, IgD, IgA, IgE, or IgM antibodies. The Fc domain encompasses native Fc and Fc variant molecules. As with Fc variants and native Fc, the term Fc chain includes molecules in monomeric or polymeric form, whether derived from whole antibody digestion or otherwise. In some embodiments, the Fc chain comprises the carboxyl-terminal portions of two heavy chains held together by disulfide bonds. In some embodiments, the Fc chain consists of a CH2 domain and a CH3 domain.

As used in the art, “Fc receptor” and “FcR” describe receptors that bind to the Fc region of an antibody. Preferred FcRs are naturally occurring human FcRs. Furthermore, preferred FcRs are those that bind IgG antibodies (γ receptors) and include receptors of the FcγRI, FcγRII, and FcγRIII subclasses, including allelic variants and alternative splicing forms of these receptors. FcγRII receptors include FcγRIIA (“activating receptor”) and FcγRIIB (“inhibitory receptor”), which are distinguished primarily by similar amino acid sequences in their cytoplasmic domains. FcRs are reviewed in: Ravetch and Kinet, Ann. Rev. Immunol., 9:457-92, 1991; Capel et al., Immunomethods, 4:25-34, 1994; and de Haas et al., J. Lab. Clin. Med., 126:330-41, 1995. "FcR" also includes the neonatal receptor FcRn, which is responsible for the transfer of maternal IgG to the fetus (Guyer et al., J. Immunol., 117:587, 1976; and Kim et al., J. Immunol., 24:249, 1994).

The “natural sequence Fc region” or “wild-type Fc region” contains the same amino acid sequence as the Fc region found in nature. “Wild-type” human IgG Fc refers to the sequence of amino acids naturally present in the human population. Of course, just as the Fc sequence can vary slightly between individuals, one or more modifications can be made to the wild-type sequence while still remaining within the scope of this invention. For example, the Fc region may contain additional modifications unrelated to this invention, such as mutations in glycosylation sites, including non-natural amino acids, or “knobs-in-holes” mutations.

The “variant Fc region” or “variant Fc chain” comprises an amino acid sequence that differs from the amino acid sequence of the natural Fc region by means of at least one amino acid modification, but retains at least one effector function of the natural Fc region. In some embodiments, the variant Fc chain has at least one amino acid substitution compared to the natural Fc chain or the Fc region of the parent polypeptide, for example, about 1 to about 10 amino acid substitutions in the natural Fc chain or the Fc chain of the parent polypeptide, and preferably about 1 to about 5 amino acid substitutions. The variant Fc chain described herein will preferably have at least about 80% sequence identity with the natural Fc chain and/or the Fc chain of the parent polypeptide, and most preferably at least about 90% sequence identity, more preferably at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99% sequence identity.

As used herein, the term "effective function" refers to those biological activities induced by the Fc region of an antibody (the native sequence Fc chain or the Fc chain of an amino acid sequence variant), and varies with antibody isotype. Examples of antibody effector functions include: C1q binding and complement-dependent cytotoxicity; Fc receptor binding; antibody-dependent cell-mediated cytotoxicity (ADCC); phagocytosis; downregulation of cell surface receptors (e.g., B cell receptors); and B cell activation. Such effector functions generally require the combination of the Fc chain with a binding domain (e.g., the antibody variable region) and can be evaluated using various assays known in the art for assessing such antibody effector functions. Exemplary measurements of effector functions are via Fcγ3 and/or C1q binding.

The effector function of an antibody is determined by a sequence in the Fc chain; this chain is also the part recognized by the Fc receptor (FcR) found on certain types of cells.

In some embodiments, the Fc polypeptide contains part or all of the wild-type hinge sequence (typically at its N-terminus). In some embodiments, the Fc polypeptide does not contain a functional or wild-type hinge sequence.

As used herein, “hinge region,” “hinge sequence,” and their variations include meanings known in the art, as illustrated in, for example, Janeway et al., ImmunoBiology: the immune system in health and disease, Elsevier Science Ltd., NY (4th edition, 1999); Bloom et al., Protein Science, 6:407-415, 1997; and Humphreys et al., J. Immunol. Methods, 209:193-202, 1997.

As used herein, "immunoglobulin-like hinge region," "immunoglobulin-like hinge sequence," and variations thereof refer to the hinge region and hinge sequence of an immunoglobulin-like or antibody-like molecule (e.g., an immunoadhesin). In some embodiments, the immunoglobulin-like hinge region may be derived from or originate from any IgG1, IgG2, IgG3, or IgG4 subtype, or IgA, IgE, IgD, or IgM, including its chimeric forms, such as a chimeric IgG1/2 hinge region.

In some embodiments, the hinge region may be derived from a human IgG1 subtype extending from amino acid 216 to amino acid 230 according to the EU indexing system or from amino acid 226 to amino acid 243 according to the Kabat numbering system. In some embodiments, the sequence may be EPKSCDKTHTCPPCP (SEQ ID NO: 63). Those skilled in the art may have different understandings of the exact amino acids corresponding to the various domains of the IgG molecule. Therefore, the N-terminus or C-terminus of the domains outlined above may be extended or shortened by 1, 2, 3, 4, 5, 6, 7, 8, 9, or even 10 amino acids.

In some embodiments, the hinge region may be mutated by one or more amino acids. In some embodiments, the hinge region may be truncated and contain only a portion of the full hinge region. In some embodiments, the hinge region may contain only the last 5 amino acids of the hinge region, referred to herein as the "lower hinge" region. In some embodiments, the lower hinge region may contain amino acid CPPCP (SEQ ID NO: 64) or amino acids 226 to 230 according to the EU index numbering system or amino acids 239 to 243 according to the Kabat numbering system.

As used herein, the term "modification" refers to the substitution, insertion, and/or deletion of amino acids in a polypeptide sequence, alteration of the portion chemically linked to the protein, or modification of the function of a protein (e.g., an antibody). For example, a modification can be an alteration of antibody function or a change in the structure of a carbohydrate linked to the protein. As used herein, "amino acid modification" refers to a mutation (substitution), insertion (addition), or deletion of one or more amino acid residues in an antibody. The term "amino acid mutation" means that at least one existing amino acid residue is replaced by another different amino acid residue (e.g., a substitute amino acid residue). The term "amino acid deletion" means the removal of at least one amino acid residue at a predetermined position in the amino acid sequence. For example, mutation L234A indicates that the amino acid residue lysine at position 234 in the Fc region of the antibody is replaced by the amino acid residue alanine (replacing lysine with alanine) (numbered according to the EU index numbering system). "Natural amino acid residues" refers to amino acid residues from the following groups: alanine (three-letter code: Ala, single-letter code: A), arginine (Arg, R), asparagine (Asn, N), aspartic acid (Asp, D), cysteine (Cys, C), glutamine (Gin, Q), glutamic acid (Glu, E), glycine (Gly, G), histidine (His, H), isoleucine (Ile, I), leucine (Leu, L), lysine (Lys, K), methionine (Met, M), phenylalanine (Phe, F), proline (Pro, P), serine (Ser, S), threonine (Thr, T), tryptophan (Trp, W), tyrosine (Tyr, Y), and valine (Val, V).

The term "pharmaceutical" is used herein to mean a biological macromolecule, an extract derived from biological material, a mixture of biological macromolecules, a chemical compound, a mixture of chemical compounds, and/or a mixture of chemical compounds and biological macromolecules. The term "therapeutic agent" refers to a biologically active pharmaceutical agent.

As used herein, a “monoclonal antibody” refers to an antibody obtained from a substantially homogeneous population of antibodies, i.e., the individual antibodies constituting the population are identical except for the possibility of naturally occurring mutations present in small amounts. Monoclonal antibodies are highly specific against a single antigenic site. Furthermore, in contrast to polyclonal antibody formulations, which typically comprise different antibodies targeting different determinants (epitopes), each monoclonal antibody targets a single determinant on the antigen. The modifier “monoclonal” indicates the characteristics of antibodies obtained from a substantially homogeneous population of antibodies and should not be construed as requiring antibody production by any particular method. For example, a monoclonal antibody to be used according to the invention can be prepared by a hybridoma method first described by Kohler and Milstein, Nature 256:495, 1975, or by a recombinant DNA method such as that described in U.S. Patent No. 4,816,567. Monoclonal antibodies can also be isolated from phage libraries generated using, for example, techniques described in McCafferty et al., Nature 348:552-554, 1990.

The antibodies of this invention can be “humanized antibodies.” As used herein, a “humanized” antibody refers to a form of non-human (e.g., mouse, rat, rabbit, non-human primate, or other mammalian) antibody that is a chimeric immunoglobulin, immunoglobulin chain, or fragment thereof (such as Fv, Fab, Fab', F(ab') ₂ , or other antigen-binding sequences of the antibody) containing one or more amino acid residues introduced therefrom from a non-human source. These non-human amino acid residues are generally referred to as “input” residues and are typically derived from an “input” variable domain. The input residues, sequence, or antibody has desired affinity and/or specificity or other desired antibody biological activities as discussed herein.

Preferably, the humanized antibody is a human immunoglobulin (receptor antibody) in which residues from the complementarity-determining region (CDR) of the receptor are replaced with residues from a non-human species (donor antibody), such as mouse, rat, or rabbit, that have the desired specificity, affinity, and ability. In some cases, residues from the Fv framework region (FR) of the human immunoglobulin are replaced with corresponding non-human residues. Furthermore, the humanized antibody may contain residues not found in either the receptor antibody or the input CDR or framework sequence, but which are included to further refine and optimize antibody performance. Typically, the humanized antibody will contain at least one, and usually substantially all, of the two variable regions, wherein all or substantially all of the CDR regions correspond to those of non-human immunoglobulins, and all or substantially all of the FR regions are those of human immunoglobulin common sequences. The humanized antibody will also preferably contain at least a portion of the immunoglobulin constant region or structural domain (Fc), typically of the type of human immunoglobulin. Antibodies having Fc chains modified as described in WO 99/58572 are preferred. Other forms of humanized antibodies have one or more CDRs (CDR L1, CDR L2, CDR L3, CDR H1, CDR H2, or CDR H3) that are altered relative to the original antibody, and are also referred to as "derived from" one or more CDRs from the original antibody. As used herein, humanization intention includes deimmunization antibodies.

As used herein, "human antibody" means an antibody having an amino acid sequence corresponding to the amino acid sequence of an antibody produced by a human, and/or having been prepared using any technique known to those skilled in the art or disclosed herein for the preparation of human antibodies. Therefore, the term "human antibody" as used herein is intended to include antibodies having variable and constant regions derived from human immunoglobulin sequences. Human antibodies of the present invention may include amino acid residues not encoded by human immunoglobulin sequences (e.g., mutations introduced by in vitro random mutagenesis or site-specific mutagenesis or by in vivo somatic mutations), such as in the CDR, and specifically CDR3. This definition of human antibody includes antibodies comprising at least one human heavy chain polypeptide or at least one human light chain polypeptide. One such example is an antibody comprising a mouse light chain and a human heavy chain polypeptide. Various techniques known in the art can be used to produce human antibodies. In one embodiment, the human antibody is selected from a phage library expressing a human antibody (Vaughan et al., Nature Biotechnology, 14:309-314, 1996; Sheets et al., Proc. Natl. Acad. Sci. (USA) 95:6157-6162, 1998; Hoogenboom and Winter, J. Mol. Biol., 227:381, 1991; Marks et al., J. Mol. Biol., 222:581, 1991). The human antibody can also be prepared by immunizing an animal in which a human immunoglobulin locus has been transgenically introduced to replace an endogenous locus, such as in mice where the endogenous immunoglobulin gene has been partially or completely inactivated. This method is described in U.S. Patent Nos. 5,545,807; 5,545,806; 5,569,825; 5,625,126; 5,633,425; and 5,661,016. Alternatively, human antibodies can be prepared by immortalizing human B lymphocytes that produce antibodies against a target antigen (such B lymphocytes can be recovered from an individual or a single-cell clone of cDNA, or can have been immunized in vitro). See, for example, Cole et al. Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, p. 77, 1985; Boerner et al., J. Immunol., 147(1):86-95, 1991; and U.S. Patent No. 5,750,373.

The human antibodies of the present invention can exist in at least two forms related to hinge heterogeneity. For example, immunoglobulin molecules comprise a stable four-chain construct of approximately 150-160 kDa, wherein the dimers are held together by interchain heavy chain disulfide bonds. Alternatively, the dimers are not linked by interchain disulfide bonds and form a molecule of approximately 75-80 kDa (half-antibody) consisting of covalently coupled light and heavy chains.

As used herein, the term "recombinant human antibody" is intended to include all human antibodies prepared, expressed, generated, or isolated in a recombinant manner, such as antibodies expressed using a recombinant expression vector transfected into host cells (described further below), antibodies isolated from a library of recombinant human antibodies (described further below), antibodies isolated from animals transgenic with human immunoglobulin genes (e.g., mice) (see, for example, Taylor et al. (1992) Nucl. Acids Res. 20:6287-6295), or antibodies prepared, expressed, generated, or isolated by any other means involving splicing a human immunoglobulin gene sequence into another DNA sequence. Such recombinant human antibodies have variable and constant regions derived from human germline immunoglobulin sequences. However, in some implementations, such recombinant human antibodies are subjected to in vitro mutagenesis (or, when using animals transgenic with human Ig sequences, in vivo somatic cell mutagenesis), and thus the amino acid sequences of the VH and VL regions of the recombinant antibody are sequences that, although derived from and associated with human germline VH and VL sequences, may not naturally exist in the human antibody germline library.

The antibodies of this invention can be “chimeric antibodies” (where a portion of the heavy chain and/or light chain is identical or homologous to the corresponding sequence in an antibody derived from a specific species or belonging to a specific antibody class or subclass, while the remaining portion of the chain is identical or homologous to the corresponding sequence in an antibody derived from another species or belonging to another antibody class or subclass), and fragments of such antibodies, provided they exhibit the desired biological activity (US Patent 4,816,567; and Morrison et al., Proc. Natl. Acad. Sci. USA 81:6851-6855, 1984). The target chimeric antibodies herein include primate-derived antibodies comprising a variable region antigen-binding sequence derived from a non-human primate (e.g., Old World monkeys, apes, etc.) and a human constant region sequence.

A "monovalent antibody" contains one antigen-binding site/molecule (e.g., IgG or Fab). In some cases, a monovalent antibody may have more than one antigen-binding site, but the binding sites come from different antigens.

A "monospecific antibody" is an antibody or antibody preparation that contains two identical antigen-binding sites/molecules (e.g., IgG), such that the two binding sites bind to the same epitopes on the antigen. Therefore, they compete with each other when binding to a single antigen molecule. This term includes "monoclonal antibody" or "monoclonal antibody composition." Most antibodies found in nature are monospecific. In some cases, monospecific antibodies can also be monovalent antibodies (e.g., Fab).

The terms “mutation load,” “mutational load,” “mutation burden,” or “mutational burden” are used interchangeably in this article. Tumor mutational load is a measure of the number of mutations within a tumor genome, defined as the total number of mutations in each coding region of the tumor genome. Mutational load varies greatly across tumor types, ranging from just a few to several thousand mutations (Alexandrov LB et al., Nature 2013;500(7463):415-421; Lawrence MS et al., Nature 2013;499:214–218; Vogelstein B et al., Science, 2013;339:1546–1558).

"Original amino acid residue" refers to a residue replaced by an "input amino acid" residue whose side chain volume may be smaller or larger than the original residue. The input amino acid residue may be naturally occurring or non-naturally occurring amino acid residues, but is preferably the former. "Naturally occurring" amino acid residues are those residues encoded by the genetic code. "Non-naturally occurring" amino acid residues refer to residues that are not encoded by the genetic code but can covalently bind to neighboring amino acid residues in the polypeptide chain. Examples of non-naturally occurring amino acid residues are leucine, ornithine, valine, homoserine, and other amino acid residue analogs, such as those described, for example, in Ellman et al., Meth. Enzym. 202:301-336 (1991). In some embodiments, the method of the present invention involves replacing at least one original amino acid residue, but may replace more than one original residue. Typically, the total residues in no more than the first or second polypeptide interface may contain the replaced original amino acid residue.

As used herein with respect to the use of the term "competitive," it means that a first antibody or its antigen-binding fragment (or portion) binds to an epitope in a manner sufficiently similar to the binding of a second antibody or its antigen-binding portion, such that the binding of the first antibody to its homologous epitope is detectably reduced in the presence of the second antibody compared to the binding of the first antibody in the absence of the second antibody. Alternatives to this, where the binding of the second antibody to its epitope is also detectably reduced in the presence of the first antibody, are possible but not necessary. That is, the first antibody may inhibit the binding of the second antibody to its epitope, while the second antibody may not inhibit the binding of the first antibody to its corresponding epitope. However, when each antibody detectably inhibits the binding of another antibody to its homologous epitope or ligand, whether to the same, greater, or lesser extent, the antibodies are said to "cross-compete" with each other for binding to their respective epitopes. Both competitive antibodies and cross-competitive antibodies are covered by this invention. Regardless of the mechanism by which such competition or cross-competition occurs (e.g., steric hindrance, conformational change, or binding to a common epitope or a portion thereof), based on the teachings provided herein, those skilled in the art will understand that such competitive antibodies and/or cross-competitive antibodies are covered and can be used in the methods disclosed herein.

As used herein, “antibody-dependent cell-mediated cytotoxicity” or “ADCC” refers to a cell-mediated reaction in which nonspecific cytotoxic cells expressing Fc receptors (FcRs), such as natural killer (NK) cells, neutrophils, and macrophages, recognize binding antibodies on target cells and subsequently cause lysis of the target cells. The ADCC activity of a target molecule can be evaluated using in vitro ADCC assays, such as those described in U.S. Patent Nos. 5,500,362 or 5,821,337. Useful effector cells for such assays include peripheral blood mononuclear cells (PBMCs) and NK cells. Alternatively or additionally, the ADCC activity of a target molecule can be evaluated in vivo, for example, in animal models, such as those disclosed in Clynes et al., PNAS (USA), 95:652-656, 1998.

As used herein, “complement-dependent cytotoxicity” or “CDC” refers to target lysis in the presence of complement. The complement activation pathway is initiated by the binding of the first component (C1q) of the complement system to molecules (e.g., antibodies) that are complexed with homologous antigens. To evaluate complement activation, a CDC assay can be performed, for example, as described in Gazzano-Santoro et al., J. Immunol. Methods, 202: 163, 1996.

As used herein, the terms “immunospecific binding,” “immunospecific recognition,” “specific binding,” “specific recognition,” and similar terms refer to a molecule that specifically binds to an antigen (e.g., an epitope or immune complex) but does not specifically bind to another molecule (e.g., a binding domain). A molecule that specifically binds to an antigen may bind to other peptides or polypeptides with lower affinity, as determined by assays known in the art, such as immunoassays, BIACORE™, or other assays. Preferably, the molecule that specifically binds to the antigen does not cross-react with other proteins.

Suitable “moderately stringent conditions” include prewashing in a solution of 5X SSC, 0.5% SDS, and 1.0 mM EDTA (pH 8.0); hybridization overnight at 50–65°C with 5X SSC; followed by two washes at 65°C with 2X, 0.5X, and 0.2X SSC containing 0.1% SDS, each for 20 minutes.

As used herein, “highly stringent conditions” or “highly stringent conditions” are conditions in which: (1) low ionic strength and high temperature are used for washing, for example, 0.015 M sodium chloride/0.0015 M sodium citrate/0.1% sodium dodecyl sulfate at 50°C; (2) denaturing agents are used during hybridization at 42°C, such as formamide, for example, 50% (v/v) formamide with 0.1% bovine serum albumin/0.1% Ficoll/0.1% polyvinylpyrrolidone/50 mM sodium phosphate buffer with 750 mM sodium chloride and 75 mM sodium citrate at pH 6.5; Alternatively (3) wash at 42°C with 50% formamide, 5 x SSC (0.75 M NaCl, 0.075 M sodium citrate), 50 mM sodium phosphate (pH 6.8), 0.1% sodium pyrophosphate, 5 x Denhardt solution, sonicated salmon sperm DNA (50 µg/ml), 0.1% SDS, and 10% dextran sulfate, followed by washing at 42°C in 0.2 x SSC (sodium chloride/sodium citrate) and at 55°C in 50% formamide, followed by a high-rigidity wash at 55°C with 0.1 x SSC containing EDTA. Technicians will recognize how to adjust the temperature, ionic strength, etc., as needed to accommodate factors such as probe length.

The binding protein, binding domain, or CDR or antibody (as broadly defined herein) may be identified according to the accumulation, AbM, contact, North definition and/or conformational definition of Kabat, Chothia, Kabat and Chothia, or any method of CDR determination well known in the art. See, for example, Kabat et al., 1991, Sequences of Proteins of Immunological Interest, 5th edition (Hypervariate Region); Chothia et al., Nature 342:877-883, 1989 (Structural Ring Structure). The identity of the amino acid residues constituting the CDR in a particular antibody may be determined using methods well known in the art. The AbM definition of the CDR is a compromise between Kabat and Chothia and uses Oxford Molecular's AbM antibody modeling software (Accelrys®). The definition of “contact” for the CDR is based on observed antigen contact and is described in MacCallum et al., J. Mol. Biol., 262:732-745, 1996. The “conformation” definition of a CDR is based on residues that contribute enthalpy to antigen binding (see, for example, Makabe et al., J. Biol. Chem., 283:1156-1166, 2008). North has identified canonical CDR conformations using different preferred sets of CDR definitions (North et al., J. Mol. Biol. 406: 228-256, 2011). In another approach referred to herein as the “conformation definition” of a CDR, the position of the CDR can be identified as residues that contribute enthalpy to antigen binding (Makabe et al., J. Biol. Chem., 283:1156-1166, 2008). Other CDR boundary definitions may not strictly follow one of the above approaches but will still overlap with at least a portion of the Kabat CDR, although they may be shortened or lengthened based on predictions or experimental findings that specific residues or groups of residues or even the entire CDR do not significantly affect antigen binding. As used herein, a CDR can refer to a CDR defined by any method known in the art, including combinations of methods. The methods used herein can utilize a CDR defined according to any of these methods. For any given embodiment containing more than one CDR, a CDR (or other residues of the antibody) can be defined according to any of the following definitions: Kabat, Chothia, extension (a combination of Kabat and Chothia), North, extension, AbM, contact, and/or conformation.

Residues in the variable domain are numbered according to Kabat, a numbering system used for the heavy chain or light chain variable regions of the assembled antibody. See Kabat et al., Sequences of Proteins of Immunological Interest, 5th edition, Public Health Service, National Institutes of Health, Bethesda, MD, 1991. Using this numbering system, the actual linear amino acid sequence may contain fewer or additional shortened or inserted amino acids corresponding to the FR or CDR of the variable region. For example, the heavy chain variable region may include a single amino acid insertion after residue 52 of H2 (according to Kabat residue 52a) and inserted residues after heavy chain FR residue 82 (e.g., according to Kabat residues 82a, 82b, and 82c). The Kabat number of residues for a given antibody can be determined by comparing the homologous regions of the antibody sequence to a “standard” Kabat numbered sequence. Various algorithms for assigning Kabat numbers are available. This article uses the algorithm implemented in Abysis release 2.3.3 (www.abysis.org) to assign Kabat numbers to the variable regions VLCDR1, VL CDR2, VL CDR3, VHCDR1, VHCDR2, and VH CDR3. Specific amino acid residue positions within the antibody can also be numbered based on Kabat numbers.

As used herein, the term "phage display library" refers to a population of phages, each containing exogenous cDNA fused to a surface protein within a recombination frame. The phages display the exogenous protein encoded by the cDNA on their surface. After replication in a bacterial host (typically *E. coli*), phages containing the target exogenous cDNA are selected based on the expression of the exogenous protein on the phage surface.

The term "epitope" refers to a portion of a molecule that can be recognized, contacted, and/or bound by an antibody at one or more sites in an antigen-binding region called a complementary site. A single antigen may have more than one epitope. Thus, different antibodies can bind to different regions on the antigen and may have different biological effects. Epitopes are often composed of chemically active surface groups of a molecule (such as amino acid or sugar side chains) and have specific three-dimensional structural features and specific charge features. As used herein, epitopes can be conformational or linear. Conformational epitopes are generated by spatially juxtaposing amino acids from different segments of a linear polypeptide chain. Linear epitopes are epitopes generated from neighboring amino acid residues in a polypeptide chain. In some embodiments, an epitope may include a sugar, phosphoryl, or sulfonyl portion of the antigen.

As used herein, an "antigenic epitope" is defined as a portion of a polypeptide to which an antibody can specifically bind, as determined by any method well known in the art (e.g., by a routine immunoassay). A "nonlinear epitope" or "conformational epitope" comprises a non-neighboring polypeptide (or amino acid) within an antigenic protein to which an antibody specific for that epitope binds. Once a desired epitope on an antigen is identified, an antibody targeting that epitope can be generated, for example, using the techniques described herein.

When used to describe the interaction between antibodies and proteins or peptides, the terms "specific binding" or "specifically binding" refer to an interaction that depends on the presence of a specific structure on the protein (i.e., an antigenic determinant or epitope); in other words, the antibody recognizes and binds to a specific protein structure rather than the usual protein. For example, if an antibody is specific to epitope "A," the presence of a protein containing epitope A (or free unlabeled A) in a reaction containing labeled "A" and an antibody will reduce the amount of labeled A that binds to the antibody.

In some embodiments, "specific binding" means, for example, that an antibody binds to a protein with a KD of about 0.1 nM or less, but more typically less than about 1 μM. In some embodiments, "specific binding" means that an antibody sometimes binds to a target with a KD of at least about 0.1 μM or less, at other times with at least about 0.01 μM or less, and at other times with at least about 1 nM or less. Due to sequence identity between homologous proteins in different species, specific binding may include antibodies that recognize proteins in more than one species (e.g., human or mouse proteins). Similarly, due to homology within certain regions of the polypeptide sequences of different proteins, specific binding may include antibodies that recognize more than one protein. It should be understood that in some embodiments, an antibody or binding portion that specifically binds to a first target may or may not specifically bind to a second target. Therefore, "specific binding" does not necessarily require (although it may include) exclusive binding, i.e., binding to a single target. Therefore, in some embodiments, an antibody may specifically bind to more than one target. In some embodiments, the same antigen-binding site on an antibody may bind to multiple targets. For example, in some cases, an antibody may contain two identical antigen-binding sites, each specifically binding to the same epitope on two or more proteins. In some alternative embodiments, the antibody may be multispecific and contain at least two antigen-binding sites with different specificities. By way of non-limiting example, a bispecific antibody may contain an antigen-binding site that recognizes an epitope on one protein (e.g., human CD3) and further contain a second, different antigen-binding site that recognizes a different epitope on a second protein. Generally, but not necessarily, references to binding refer to specific binding.

Antibodies that specifically bind to antigens can bind to other peptides or polypeptides with lower affinity, as determined by assays known in the art (e.g., immunoassays, BIACORE®, or other assays). Preferably, antibodies that specifically bind to antigens do not cross-react with other proteins.

When used in relation to the interaction between antibodies and proteins or peptides, the terms “non-specific binding” or “background binding” refer to interactions that do not depend on the presence of a specific structure (i.e., antibody binding is typically protein rather than a specific structure, such as an epitope).

As used herein, the terms "k _on " or " _ka " refer to the rate constant relating to the association of antibody with antigen. Specifically, the rate constants (k _on / _ka and k _off /k _d ) and equilibrium dissociation constants are measured using intact antibodies (i.e., bivalent) and monomeric CD3 proteins.

As used in this article, the terms "k _off " or "k _d " refer to the rate constant of antibody dissociation from the antibody/antigen complex.

As used in this article, the term " _KD " refers to the equilibrium dissociation constant of antibody-antigen interactions.

As used herein, the term "binding affinity" generally refers to the total strength of the non-covalent interaction between a single binding site of a molecule (e.g., an antibody) and its binding partner (e.g., an antigen). Unless otherwise indicated, as used herein, "binding affinity" refers to the intrinsic binding affinity that reflects a 1:1 interaction between members of a binding pair (e.g., antibody and antigen). The affinity of molecule X for its partner Y is generally represented by the dissociation constant (KD). The affinity of molecule X for its partner Y is generally represented by the dissociation constant (Kd). For example, Kd can be about 200 nM, 150 nM, 100 nM, 60 nM, 50 nM, 40 nM, 30 nM, 20 nM, 10 nM, 8 nM, 6 nM, 4 nM, 2 nM, 1 nM, or stronger. Affinity can be measured by methods commonly known in the art, including those described herein. Low-affinity antibodies typically bind to antigens slowly and tend to dissociate easily, while high-affinity antibodies typically bind to antigens quickly and tend to remain bound for a longer period. Various methods for measuring binding affinity are known in the art, any of which may be used for the purposes of this invention. Specifically, the term "binding affinity" is intended to refer to the rate of dissociation of a particular antigen-antibody interaction. _KD is the ratio of the rate of dissociation (also called the "off-rate" (k _-off )) to the rate of association (k-on) (k- _on ). Therefore, _KD equals k _-off / k _-on and is expressed as a molar concentration (M). Thus, the smaller the _KD , the stronger the binding affinity. Therefore, a _KD of 1 μM indicates weak binding affinity compared to _{a KD} of 1 nM. The _KD value of an antibody can be determined using methods well established in the art. One method for determining the _KD of an antibody is by using surface plasmon resonance (SPR), typically employing a biosensor system (such as the BIACORE™ system). BIACORE™ kinetic analysis involves analyzing the binding and dissociation of antigens from chips with immobilized molecules (e.g., molecules containing epitope-binding domains) on their surfaces. Another method for determining the _KD of antibodies is through biolayer interferometry, typically using ^OCTET® technology (Octet QKe system, ForteBio).

Unless otherwise specified, with respect to the antibodies (such as antibodies, fragments or derivatives thereof) of the present invention, "bioactive", "bioactive" and "biocharacteristic" mean having the ability to bind biomolecules.

As used herein, the terms "nucleic acid" and "nucleotide sequence" include DNA molecules (e.g., cDNA or genomic DNA), RNA molecules (e.g., mRNA), combinations or hybrid DNA/RNA molecules, and analogs of DNA or RNA molecules. Such analogs can be generated using, for example, nucleotide analogs, including but not limited to inosine or triphenylmethylated bases. Such analogs may also comprise DNA or RNA molecules containing a modified backbone that imparts beneficial properties to the molecule, such as, for example, nuclease resistance or increased transmembrane translucency. Nucleic acid or nucleotide sequences can be single-stranded, double-stranded, contain both single-stranded and double-stranded portions, and may contain triple-stranded portions, but are preferably double-stranded DNA.

The present invention also includes a polynucleotide encoding the antibody of the present invention (including the polypeptide and binding region of the antibody). The polynucleotide encoding the molecule of the present invention can be obtained, and the nucleotide sequence of the polynucleotide can be determined by any method known in the art.

The polynucleotide encoding the antibody of the present invention may include: a variant-only coding sequence; a variant-coding sequence and additional coding sequences (such as functional polypeptides or signal sequences or secretion sequences or preprotein sequences); and antibody-coding and non-coding sequences (such as introns or non-coding sequences at the 5' and/or 3' of the antibody-coding sequence). The term "polynucleotide encoding antibody" encompasses polynucleotides including additional coding sequences of variants, and also encompasses polynucleotides including additional coding and/or non-coding sequences. It is known in the art that polynucleotide sequences optimized for specific host cells/expression systems can be readily obtained from the amino acid sequence of the desired protein (see ^GENEART® AG, Regensburg, Germany).

The antibodies of the present invention may have additional conserved or non-essential amino acid substitutions that do not substantially affect the function of the polypeptide. Whether a particular substitution will be tolerated (i.e. will not adversely affect the desired biological properties (such as binding activity)) can be determined as described in Bowie, JU et al., Science 247: 1306-1310, 1990 or Padlan et al., FASEBJ. 9: 133-139, 1995.

As used herein, "conservative amino acid substitution" is a substitution in which an amino acid residue is replaced by an amino acid residue having a similar side chain. Families of amino acid residues having similar side chains are defined in the art. These families include amino acids having the following side chains: basic side chains (e.g., lysine, arginine, histidine), acidic side chains (e.g., aspartic acid, glutamic acid), uncharged polar side chains (e.g., asparagine, glutamine, serine, threonine, tyrosine, cysteine), nonpolar side chains (e.g., glycine, alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine, tryptophan), β-branched side chains (e.g., threonine, valine, isoleucine), and aromatic side chains (e.g., tyrosine, phenylalanine, tryptophan, histidine).

As used herein, "non-essential" amino acid residues are residues that may differ from the wild-type sequence of a binding agent (e.g., an antibody) and do not eliminate or substantially alter biological activity, while "essential" amino acid residues cause this change. In antibodies, essential amino acid residues can be specificity-determining residues (SDRs).

As used herein, the term "isolated" refers to material removed from its original environment (e.g., the natural environment if it is naturally occurring). For example, naturally occurring polynucleotides or polypeptides present in living animals are not isolated, but the same polynucleotides or polypeptides isolated from some or all of the coexisting material in a natural system are isolated. This polynucleotide may be part of a carrier, and/or this polynucleotide or polypeptide may be part of a composition (e.g., a mixture, solution, or suspension) or contain isolated cells or cultured cells containing the polynucleotide or polypeptide, and is still isolated because the carrier or composition is not part of its natural environment. For example, a polypeptide synthesized chemically or synthesized in a cellular system different from the cell of natural origin of the polypeptide will be "isolated" from its natural associated components. Protein purification techniques well known in the art can also be used to isolate proteins to make them substantially free of natural associated components.

"Isolated" or "purified" antibodies are substantially free of cellular material or other contaminating proteins from cell or tissue sources or culture media from which the protein originates, or are substantially free of chemical precursors or other chemicals during chemical synthesis. The phrase "substantially free of cellular material" includes antibody formulations in which the polypeptide/protein is isolated from cellular components of the cells from which the polypeptide/protein is produced, either isolated or recombinantly. Therefore, antibodies substantially free of cellular material include antibody formulations having less than about 50%, 40%, 30%, 20%, 10%, 5%, 2.5%, or 1% (by dry weight) of contaminating proteins. When antibodies are produced by recombinant synthesis, they are also preferably substantially free of culture media, i.e., the culture medium constitutes less than about 50%, 40%, 30%, 20%, 10%, 5%, 2.5%, or 1% of the volume of the protein formulation. When antibodies are produced by chemical synthesis, they are preferably substantially free of chemical precursors or other chemicals and reagents, i.e., the target antibody is isolated from chemical precursors or other chemicals involved in protein synthesis. Therefore, this antibody preparation contains less than about 50%, 40%, 30%, 20%, 10%, 5%, or 1% (by dry weight) of a chemical precursor or compound in addition to the target antibody. In a preferred embodiment of the invention, the antibody is isolated or purified. For the purposes of this invention, an antibody that has been isolated or removed from at least one component of an organism or from tissues or cells in which the antibody is naturally present or produced is called an "isolated antibody".

As used herein, the term “recycling” refers to, for example, a process that uses protein purification techniques well known in the art to separate chemical substances (such as peptides) to make them substantially free of their natural associated components.

As used herein, the term "replicon" refers to any genetic element, such as a plasmid, chromosome, or virus, that acts as an autonomous polynucleotide replication unit within a cell.

As used herein, the term "operably ligated" refers to a situation where the relationship between said elements allows them to function in the intended manner. For example, "operably ligated" to a control sequence of a coding sequence in a manner that enables expression of the coding sequence under suitable or compatible conditions with that control sequence. Generally, "operably ligated" means that the ligated DNA sequences are adjacent, and in the case of a secretory leader sequence, adjacent and in the reading phase. However, enhancers do not require adjacency. Ligation is accomplished by ligating at a convenient restriction site. If such a site is not present, synthetic oligonucleotide adaptors or linkers are used according to conventional practice.

As used herein, “vector” means a construct capable of being delivered in a host cell and preferably expressing one or more target genes or sequences. Examples of vectors include, but are not limited to, viral vectors, naked DNA or RNA expression vectors, plasmids, granules or phage vectors, DNA or RNA expression vectors associated with cationic condensers, DNA or RNA expression vectors encapsulated in liposomes, and certain eukaryotic cells, such as production cells.

As used herein, the term "expression control sequence" or "control sequence" refers to a polynucleotide sequence necessary for the expression of the coding sequence it is linked to. The nature of this control sequence varies depending on the host organism. For example, in prokaryotes, such control sequences typically include a promoter, a ribosome binding site, a terminator, and (in some cases) an enhancer. Therefore, the term "control sequence" is intended to include at least all elements necessary for expression, but may also include additional elements that are advantageous for its presence, such as a leader sequence.

"Host cell" includes an individual cell or cell culture that may be or has been a vector recipient, the vector being used to incorporate a polynucleotide insert. Host cells include progeny of a single host cell, and the progeny may not be identical to the original parent cell (morphologically or in terms of genomic DNA complementation) due to natural, accidental, or intentional mutations. Host cells include cells transfected in vivo with the polynucleotides of the present invention.

As used herein, "mammalian cell" includes references to cells derived from mammals, including humans, rats, mice, guinea pigs, chimpanzees, or macaques. These cells may be cultured in vivo or in vitro.

As used herein, the term "purified product" refers to a product formulation that has been isolated from the cellular components to which the product is normally associated and/or from other cell types that may be present in the target sample.

As used herein, “substantially pure” means a material that is at least 50% pure (i.e., free of contaminants), more preferably at least 90% pure, more preferably at least 95% pure, even more preferably at least 98% pure, and most preferably at least 99% pure.

As used herein, the terms “treat,” “treating,” or “treatment” are methods for obtaining beneficial or desired clinical outcomes. For the purposes of this invention, treatment is defined as administering a CD3 antibody molecule (e.g., a monoclonal antibody, bispecific antibody) to a subject (e.g., a patient), or administering (e.g., by application) an isolated tissue or cell from the subject and returning it to said subject. The CD3 antibody molecule may be administered alone or in combination with one or more agents. Treatment can be curative, healing, alleviating, resolving, altering, remedying, improving, reducing, improving, or affecting a condition, its symptoms, or a predisposition to the condition (e.g., cancer).

As used herein, the term “subject” is intended to include any animal (e.g., a mammal), including but not limited to humans, non-human primates, rodents, etc., that is a recipient of a particular treatment. For example, a subject may be a patient with cancer (e.g., a human patient or a veterinary patient). Generally, the terms “subject,” “individual,” and “patient” are used interchangeably in this document when referring to human subjects.

Unless otherwise stated, the term "non-human animal" in this invention includes all non-human vertebrates, such as non-human mammals and non-mammals, such as non-human primates, sheep, dogs, cows, chickens, amphibians, reptiles, mice, rats, rabbits, or goats, etc.

As used herein, the term “pharmaceutical acceptable” means a product or compound that has been approved (or permitted to be approved) by a federal or state government regulatory agency or is listed in the United States Pharmacopeia or other recognized pharmacopoeia for use in animals (including humans).

As used herein, the terms "pharmaceutically acceptable excipient, carrier, or adjuvant" or "acceptable pharmaceutical carrier" mean an excipient, carrier, or adjuvant that can be administered to a subject together with at least one antibody of this disclosure without destroying the activity of the antibody. When administered together with an antibody at a dose sufficient to deliver a therapeutic effect, the excipient, carrier, or adjuvant shall be non-toxic.

As used herein, the term “improvement” means a reduction or improvement of one or more symptoms compared to the absence of the antibody molecules of the present invention. “Improvement” also includes a reduction or decrease in the duration of symptoms.

As used herein, the terms “prevent,” “preventing,” and “prevention” refer to the prevention of recurrence or onset of one or more symptoms of a disease in a subject by means of the administration of a prophylactic or therapeutic agent.

As used herein, “effective amount,” “therapeutic effective amount,” “therapeutic adequate amount,” or “effective dose” means the amount of a therapeutic agent (e.g., a CD3 antibody, alone or as part of a multispecific antibody (e.g., bispecific)) administered to a subject in a single or multiple doses that is effective or sufficient to prevent, heal, improve, treat, or manage a disease, condition, or side effect, or to reduce the rate of progression of a disease or condition, or to prolong healing, alleviate, reduce, or improve the condition of a subject with a condition as described herein. The term also includes, within its scope, amounts that effectively enhance normal physiological function. An effective amount can be considered in the context of administration of one or more therapeutic agents, and a single agent may be considered to have been administered in an effective amount if, in combination with one or more other agents, the desired outcome can be achieved.

As used herein, effective agents can target effector T cells to induce T cell-mediated cytotoxicity in cells associated with diseases such as cancer, autoimmune diseases, or infectious diseases. The therapeutic agents of the present invention induce a T cell-mediated immune response by modulating the CD3 antigen, specifically CD3, and more specifically CD3ε, on the surface of T cells.

Regarding cancer, the therapeutically effective dose refers to the amount of a therapeutic agent that inhibits the growth of a tumor or cancer by slowing, interrupting, stopping, or terminating its growth and/or metastasis.

Efficacy is a measure of the activity of a therapeutic agent, expressed as the amount required to produce an effect of a given strength. A highly potent agent elicits a greater response at low concentrations compared to a less potent agent that elicits a smaller response at low concentrations. Efficacy varies with affinity and potency. Potency refers to the ability of a therapeutic agent to produce a biological response upon binding to a target ligand and the quantitative magnitude of that response. As used herein, the term “half-maximum effective concentration ( _EC50 )” refers to the concentration of a therapeutic agent that elicits a response midway between baseline and maximum after a specified exposure time. Therapeutic agents can cause inhibition or stimulation. The _EC50 value is commonly used and is referred to herein as a measure of efficacy.

As used herein, “combination therapy” or “in combination with” refers to the use of more than one prophylactic and/or therapeutic agent. The use of the terms “combination therapy” or “combination” does not limit the order in which prophylactic and/or therapeutic agents are administered to a subject with the condition. In other words, combination therapy can be administered by treating with the therapeutic agents separately, sequentially, or simultaneously. In the case of “sequential administration,” the first administered agent may exert some physiological effect on the subject upon administration of the second agent or upon becoming active in the subject.

As used herein with respect to the administration of prophylactic and/or therapeutic agents, the term "simultaneous administration" means administration of the agents such that the individual agents are simultaneously present in the subject. Simultaneous administration can be achieved by molecules in a single composition or in separate compositions administered at the same or similar times. Sequential administration can be performed in any order as needed.

The term “differentiation cluster 3” or “CD3” refers to a polyprotein complex (historically known as the T3 complex) composed of four distinct polypeptide chains: epsilon (ε), gamma (γ), delta (δ), and zeta (ζ), which assemble and function as three pairs of dimers (εγ, εδ, ζζ). The CD3 complex acts as a T-cell co-receptor, non-covalently associating with the T-cell receptor (TCR) (Smith-Garvin et al. 2009) and generating activation signals in T lymphocytes. The CD3 protein complex is a defining characteristic of the T-cell lineage, and therefore CD3 antibodies can be effectively used as T-cell markers (Chetty and Gatter 1994). It is well known that CD3 antibodies induce the generation of cytotoxic T cells by activating endogenous lymphotropic mediators and can selectively kill tumor targets (Yun et al., Cancer Research, 49: 4770-4774 (1989)).

T cells play a central role in cell-mediated immunity in humans and animals. The recognition and binding of specific antigens are mediated by TCRs expressed on the surface of T cells. T cell TCRs can interact with immunogenic peptides (epitopes) that bind major histocompatibility complex (MHC) molecules and are presented on the surface of target cells. Specific binding of the TCR triggers a signaling cascade within the T cell, leading to proliferation and differentiation into mature effector T cells. More specifically, T cells express TCR complexes capable of inducing antigen-specific immune responses (Smith-Garvin et al., 2009). Antigens are peptides expressed by tumor cells and virus-infected cells capable of stimulating immune responses. Intracellularly expressed antigens bind to class I major histocompatibility (MHC) molecules and are transported to the surface, where they are exposed to T cells. If the binding affinity of the TCR to the class I MHC complexed with the antigen is sufficient, the formation of an immune synapse is initiated. Signaling via the immune synapse is mediated by CD3 co-receptors that form ε/δ, δ/γ, and ζ/ζ dimers. These dimers are associated with the TCR and generate activation signals in T lymphocytes. This signaling cascade guides T cell-mediated killing of antigen-expressing cells. Cytotoxicity is mediated by the release of granzyme B and perforin from T cells and their translocation to target cells.

As used herein, “activated T-cell antigen” refers to an antigenic determinant expressed on the surface of T lymphocytes, specifically cytotoxic T lymphocytes, which can induce T-cell activation upon interaction with antigen-binding molecules. Specifically, the interaction between antigen-binding molecules and activated T-cell antigen can induce T-cell activation by triggering a signaling cascade of the T-cell receptor complex. In one specific embodiment, the activated T-cell antigen is CD3, specifically the ε subunit of CD3 (for human sequences, see UniProt number P07766 (version 130), NCBI RefSeq number NP_000724.1, SEQ ID NO: 66). Typically, naturally occurring allele variants have an amino acid sequence that is at least 95%, 97%, or 99% identical to the protein described in Genbank accession number BAB71849.1.

As used herein, "T cell activation" refers to one or more cellular responses selected from the following: proliferation, differentiation, cytokine secretion, release of cytotoxic effector molecules, cytotoxic activity, and expression of activation markers by T lymphocytes, specifically cytotoxic T lymphocytes. The T cell activation bispecific antigen-binding molecule of the present invention can induce T cell activation. Assays suitable for measuring T cell activation are known in the art described herein.

As used herein, “antibody that binds CD3,” “antibody that recognizes CD3,” “anti-CD3 antibody,” or “CD3 antibody” includes an antibody that specifically recognizes a single CD3 subunit (e.g., ε, δ, γ, or ζ) and its antigen-binding fragment, and an antibody that specifically recognizes a dimer complex of two CD3 subunits (e.g., γ/ε, δ/ε, and ζ/ζ CD3 dimers) and its antigen-binding fragment. Human CD3ε is indicated in Genbank accession number NM_000733. The antibodies of the present invention can bind soluble CD3 and/or CD3 expressed on the cell surface. Soluble CD3 includes native CD3 protein and recombinant CD3 protein variants (such as, for example, monomeric and dimer CD3 constructs) lacking a transmembrane domain or otherwise not associated with the cell membrane.

Antibodies targeting CD3 can generate activation signals in T lymphocytes. Other T cell activation ligands, including but not limited to CD28, CD134, CD137, and CD27, can also be used.

This invention includes a single-arm antibody that binds to CD3. As used herein, "single-arm antibody" means an antigen-binding molecule comprising a single antibody heavy chain and a single antibody light chain. The single-arm antibody of this invention may comprise any of the VL/VH or CDR amino acid sequences as described in Tables 1, 2 and 3 herein.

The CD3 antibodies disclosed herein may contain one or more amino acid substitutions, insertions, and/or deletions in the framework and/or CDR regions of the heavy and light chain variable domains, compared to the corresponding germline sequences from which the antibodies are derived. Such mutations can be readily identified by comparing the amino acid sequences disclosed herein with germline sequences available, for example, from public antibody sequence databases. This invention includes antibodies derived from any of the amino acid sequences disclosed herein and their antigen-binding fragments, wherein one or more amino acids in one or more framework and/or CDR regions are mutated to corresponding residues of the germline sequence from which the antibody is derived, or mutated to corresponding residues of another human germline sequence, or mutated to conserved amino acid substitutions of the corresponding germline residues (such sequence changes are collectively referred to herein as “germline mutations”). Those skilled in the art can readily generate a variety of antibody and antigen-binding fragments containing one or more individual germline mutations or combinations thereof, starting with the heavy and light chain variable region sequences disclosed herein. In some embodiments, all framework and/or CDR residues within the VH and/or VL domains are mutated back to residues found in the original germline sequence from which the antibody is derived. In some embodiments, only certain residues are mutated back to the original germline sequence, such as mutated residues found only in the first 8 amino acids of FR1 or the last 8 amino acids of FR4, or mutated residues found only in CDR1, CDR2, or CDR3. In some further embodiments, one or more of the framework and/or CDR residues are mutated to corresponding residues in a different germline sequence (i.e., a germline sequence different from the germline sequence from which the antibody originally originated).

Furthermore, the antibodies of the present invention may contain any combination of two or more germline mutations within the framework and/or CDR region, for example, wherein certain individual residues are mutated to corresponding residues of a specific germline sequence, while certain other residues different from the original germline sequence are maintained or mutated to corresponding residues of a different germline sequence. Once obtained, one or more desired properties of antibodies containing one or more germline mutations can be readily tested, such as improved binding specificity, increased binding affinity, improved or enhanced antagonistic or agonistic biological properties (as applicable), reduced immunogenicity, etc. Antibodies obtained in this general manner are covered within the present invention.

The present invention also includes CD3 antibodies comprising variants of any of the disclosed VH, VL, and/or CDR amino acid sequences having one or more conserved substitutions. For example, the present invention includes CD3 antibodies having VH, VL, and/or CDR amino acid sequences having, for example, 10 or fewer, 8 or fewer, 6 or fewer, 4 or fewer, or other conserved amino acid substitutions relative to any of the VH, VL, and/or CDR amino acid sequences described in Table 1 herein.

As used herein, the term "complex" or "composite" refers to the association of two or more molecules that interact with each other via bonds and/or forces that are not peptide bonds (e.g., van der Waals forces, hydrophobic forces, hydrophilic forces). In one embodiment, the complex is heteropolymeric. It should be understood that the terms "protein complex" or "peptide complex" as used herein include complexes having non-protein entities (e.g., including but not limited to chemical molecules, such as toxins or detection agents) conjugated to proteins in the protein complex.

As used herein, "CD3 antibody" includes monovalent antibodies with single specificity, as well as "bispecific antibody," "two-specific antibody," "three-specific antibody," "bifunctional antibody," "heteropolymer," "heteropolymer complex," "bispecific heterodimeric biantibody," "heteropolypeptide," or bispecific heterodimeric IgG. In some embodiments of the present invention, the CD3 antibody of the present invention is a human or humanized antibody.

As used herein, a "bispecific antibody" is a molecule comprising at least a first polypeptide and a second polypeptide, wherein the second polypeptide differs from the first polypeptide in its amino acid sequence by at least one amino acid residue. In some cases, a bispecific antibody is an artificial hybrid antibody having two distinct heavy chain regions and light chain regions. Preferably, a bispecific antibody has binding specificity to at least two different ligands, antigens, or binding sites. Therefore, a bispecific antibody can bind to two different antigens simultaneously. The two antigen-binding sites of a bispecific antibody can reside on two different epitopes on the same or different protein targets (e.g., tumor targets).

As used herein, the first and second polypeptide chains of a “bispecific antibody” contain at least one antibody VL and an antibody VH region or a fragment thereof, wherein both antibody-binding domains are contained within a single polypeptide chain and wherein the VL and VH regions in each polypeptide chain are derived from different antibodies.

Bispecific antibodies can be derived from full-length antibodies or antibody fragments (e.g., F(ab')2 bispecific antibodies). Bispecific antibodies are more fully described, for example, in EP404,097; WO93/11161; and Hollinger et al., Proc. Natl. Acad. Sci. USA 90:6444-6448, 1993. A bispecific antibody is a heterodimer of two “crossed” sFv fragments, in which the VH and VL regions of the two antibodies are located on different polypeptide chains.

As used herein, the terms “link,” “fusion,” “fusion product,” “covalent bonding,” “covalent coupling,” and “gene fusion” are used interchangeably. These terms refer to the linking of two or more elements or components together by any means, including chemical conjugation or recombination. As used herein, the term “covalent bonding” means that the specified parts are directly covalently bonded to each other or indirectly covalently bonded to each other via one or more insert parts (such as linking peptides or parts). In a preferred embodiment, the parts are covalently fused. One type of covalent bonding is the peptide bond. Methods of chemical conjugation (e.g., using heterobifunctional crosslinking agents) are known in the art. The fusion part can also be a gene fusion. As used herein, the terms “gene fusion,” “gene link,” or “gene fusion” refer to the expression of a gene via a single polynucleotide molecule encoding two or more proteins, polypeptides, or fragments, which are collinearly, covalently linked, or connected via their individual peptide backbones. This genetic fusion results in the expression of a single adjacent gene sequence. Preferred gene fusions are in-frame fusions, meaning two or more open reading frames (ORFs) fuse in a manner that maintains the correct reading frame of the original ORF to form a continuously longer ORF. Therefore, the resulting recombinant fusion protein is a single polypeptide containing two or more protein segments corresponding to the polypeptides encoded by the original ORFs (segments that would not normally bind in this way naturally). In this case, the single polypeptide is cleaved during treatment to produce a dimer molecule containing two polypeptide chains.

As used herein, the terms "linked" or "links" refer to a direct peptide bond between first and second amino acid sequences, or a bond involving a third amino acid sequence peptide bonded to and between the first and second amino acid sequences. For example, a linker peptide is bonded to the C-terminus of one amino acid sequence and the N-terminus of another amino acid sequence.

As used herein, the term "linker" refers to an amino acid sequence of two or more amino acids in length. Linkers can consist of neutral polar or nonpolar amino acids. The length of a linker can be, for example, from 2 to 100 amino acids, such as between 2 and 50 amino acids, such as 3, 5, 10, 15, 20, 25, 30, 35, 40, 45, or 50 amino acids. Linkers can be "cleavable," for example, by self-cleavage, enzymatic cleavage, or chemical cleavage. Cleavage sites in amino acid sequences, and enzymes and chemicals that cleave at such sites, are well known in the art and are also described herein.

As used herein, the term "disulfide bond" or "cysteine-cysteine disulfide bond" refers to a covalent interaction between two cysteine residues, wherein the sulfur atom of the cysteine is oxidized to form a disulfide bond. The average bond energy of a disulfide bond is approximately 60 kcal/mol, compared to 1-2 kcal/mol for hydrogen bonds. In the context of this invention, the cysteine residues forming the disulfide bond are located within the framework region of a single-chain antibody and serve to stabilize the antibody's conformation. Cysteine residues can be introduced, for example, through site-directed mutagenesis, to generate stable disulfide bonds intramolecularly.

The term "the name of the pestle in the mortar" is similar to the term "protrusion and cavity" and they can be used interchangeably.

A "protrusion" or "pole" refers to at least one of the following amino acid side chains: extending from the interface of the first polypeptide and thus able to be positioned in a compensating cavity in a neighboring interface (i.e., the interface of the second polypeptide) to stabilize the heterodimer, thereby, for example, making heterodimer formation superior to homodimer formation. The protrusion may be present in the original interface or may be introduced synthetically (e.g., by altering the nucleic acid encoding the interface). Typically, the nucleic acid encoding the interface of the first polypeptide is modified to encode the protrusion. To achieve this, the nucleic acid encoding at least one "original" amino acid residue in the interface of the first polypeptide is replaced with an "input" amino acid residue encoding at least one side chain with a volume larger than the original amino acid residue. The upper limit for the number of original residues replaced is the total number of residues in the interface of the first polypeptide. Certain input residues used to form the protrusion are typically naturally occurring amino acid residues and are preferably selected from arginine (R), phenylalanine (F), tyrosine (Y), and tryptophan (W).

The protrusions or cavities are "localized" within cavities or sockets, meaning that the spatial location and size of the protrusions and cavities at the interface between the first and second peptides, respectively, allow the protrusions to be located within the cavities without significantly disrupting the normal association of the first and second peptides at the interface. Since protrusions, such as phenylalanine (F), tyrosine (Y), and tryptophan (W), typically do not extend perpendicularly from the interface axis, the comparison of protrusions to their corresponding cavities relies on modeling the protrusion/cavity pairs based on three-dimensional structures, such as those obtained by X-ray crystallography or nuclear magnetic resonance (NMR). This can be achieved using techniques well-established in the art.

A “cavity” or “mortise” refers to at least one amino acid side chain that recesses from the interface of the second polypeptide and thus accommodates a corresponding protrusion on the adjacent interface of the first polypeptide. The cavity may be present in the original interface or may be introduced synthetically (e.g., by altering the nucleic acid encoding the interface). Typically, the nucleic acid encoding the interface of the second polypeptide is altered to encode the cavity. To achieve this, the nucleic acid encoding at least one “original” amino acid residue in the interface of the second polypeptide is replaced with DNA encoding at least one “input” amino acid residue that has a side chain volume smaller than the original amino acid residue. The upper limit for the number of original residues replaced is the total number of residues in the interface of the second polypeptide. Some of the input residues used to form the cavity are typically naturally occurring amino acid residues and are preferably selected from alanine (A), serine (S), threonine (T), and valine (V).

As used herein, “target antigen,” “target cell antigen,” “tumor antigen,” or “tumor-specific antigen” refers to antigenic determinants that are presented on the surface of target cells (e.g., cells in a tumor, such as cancer cells) or cells in the tumor stroma).

As used herein, the term “cancer” or “cancerous” refers to or describes a physiological condition in mammals characterized by uncontrolled cell growth, growths or tumors of abnormal, uncontrolled growth originating from cells. In some respects, cancer refers to a malignant primary tumor that has not metastasized and remains localized. In other respects, cancer refers to a malignant tumor that has invaded and destroyed adjacent body structures and spread to distant sites. In some respects, cancer is associated with a specific cancer antigen. Examples of cancer include, but are not limited to, cancers of the following sites: mouth and pharynx, digestive system (e.g., esophagus, stomach, small intestine, colon, rectum, liver, gallbladder, bile ducts, and pancreas), respiratory system (e.g., larynx, lungs, and bronchi, including non-small cell lung cancer), bone and joints (e.g., bone metastases), soft tissue, skin (e.g., melanoma), breast, reproductive system (e.g., cervix, uterus, ovary, vulva, vagina, prostate, and testes), urinary system (e.g., bladder, kidneys, renal pelvis, and ureters), eye and orbit, brain and nervous system (e.g., glioma), head and neck, or endocrine system (e.g., thyroid). Cancer can also be lymphoma (such as Hodgkin's disease and non-Hodgkin's lymphoma), multiple myeloma, or leukemia (such as acute lymphoblastic leukemia, chronic lymphoblastic leukemia, acute myeloid leukemia, chronic myeloid leukemia, etc.). In some implementations, cancers of the digestive system are cancers of the esophagus, stomach, small intestine, colon, rectum, liver, gallbladder, bile ducts, and pancreas.

Regarding cancer, treatment is useful in any one or more of the following: (a) treating, preventing, or improving one or more symptoms of a condition associated with malignant cells in a subject (e.g., gastrointestinal cancers such as colorectal cancer (CRC)); (b) inhibiting tumor growth or progression in a subject with malignant tumors expressing tumor antigens; (c) inhibiting metastasis of cancer (malignant) cells expressing tumor antigens in a subject with one or more malignant cells expressing tumor antigens; (d) inducing regression (e.g., long-term regression) of tumors expressing tumor antigens; (e) exerting cytotoxic activity in malignant cells expressing tumor antigens; (f) increasing progression-free survival in a subject with a tumor antigen-associated condition; (g) increasing overall survival in a subject with a tumor antigen-associated condition; (h) reducing the use of additional chemotherapy agents or cytotoxic agents in a subject with a tumor antigen-associated condition; (i) reducing tumor burden in a subject with a tumor antigen-associated condition; or (j) blocking the interaction of tumor antigens with other unidentified factors. While not wishing to be bound by theory, the treatment is believed to have the ability to inhibit, ablate, or kill cells or otherwise reduce cell-mediated conditions (such as cancer) in vitro or in vivo.

As used in this article, the terms “malignant cell” or “malignant disease” refer to an aggressive and/or metastatic tumor or tumor cell, i.e., cancer cells.

Although any materials and methods similar to or equivalent to those described herein may be used in the practice or testing of this invention, preferred materials and methods are now described.

Materials and methods

Various techniques for antibody production have been described, including conventional hybridoma methods for preparing monoclonal antibodies, recombinant techniques for preparing antibodies (including chimeric antibodies, such as humanized antibodies), antibody production in transgenic animals, and the recently described phage display technique for preparing "fully human" antibodies. These techniques are briefly described below.

Polyclonal antibodies against a target antigen are typically generated in animals through multiple subcutaneous (sc) or intraperitoneal (ip) injections of the antigen and adjuvant. Using bifunctional reagents or derivatizing agents, such as maleimide benzoyl sulfosuccinimide (conjugated via cysteine residues) or N-hydroxysuccinimide (conjugated via lysine residues), the antigen (or a fragment containing the target amino acid sequence) can be conjugated to a protein (which is immunogenic in the species to be immunized, such as serum albumin, bovine thyroglobulin, or soybean trypsin inhibitors). Animals are immunized against the immunogenic conjugate or derivative, and booster immunizations are performed several weeks later via subcutaneous injections at multiple sites. Seven to 14 days later, blood is drawn from the animals, and serum antibody titers are measured. Booster immunizations are continued until a titer plateau is reached. Preferably, booster immunizations are performed using conjugates of the same antigen (but conjugated to different proteins and/or via different cross-linking agents). Conjugates can also be prepared as protein fusions in recombinant cell cultures. In addition, agglomerating agents (such as alum) are used to enhance the immune response.

Monoclonal antibodies can be obtained from a substantially homogeneous population of antibodies using the hybridoma method first described by Kohler and Milstein, Nature 256:495, 1975, or can be prepared via a recombinant DNA method (Cabilly et al., U.S. Patent No. 4,816,567). In the hybridoma method, immunization of mice or other suitable host animals, as described above, induces the production or ability to produce lymphocytes that will specifically bind to proteins used for immunization. Alternatively, immunization of lymphocytes can be performed in vitro. The lymphocytes are then fused with myeloma cells using a suitable fusion agent (such as polyethylene glycol) to form hybridoma cells (Goding, Monoclonal Antibodies: Principles and Practice, pp. 59-103 (Academic Press, 1986). The hybridoma cells thus prepared are seeded in a suitable culture medium, preferably containing one or more substances that inhibit the growth or survival of unfused parental myeloma cells, and allowed to grow therein. Alternatively, the hybridoma cells can grow in vivo as ascites tumors in animals. The monoclonal antibodies secreted by the subclones are appropriately separated from the culture medium, ascites, or serum using routine immunoglobulin purification procedures such as, for example, protein A-Sepharose, hydroxyapatite chromatography, gel electrophoresis, dialysis, or affinity chromatography.

The expression of antibodies, antigen-binding fragments of antibodies, or any antibody construct can be carried out in appropriate prokaryotic or eukaryotic host cells (such as CHO cells, NSO cells, SP2/0 cells, HEK293 cells, COS cells, PER.C6 cells, yeast or E. coli cells), and the secreted antibodies can be recovered and harvested from said cells (cell culture supernatant, regulated cell culture supernatant, cell lysate, or clarified bulk material). Common methods for generating antibodies are well known in the art and described in, for example, the following review articles: Makrides, S.C., Protein Expr. Purif. 17:183-202, 1999; Geisse, S., et al., Protein Expr. Purif. 8:271-282, 1986; Kaufman, R.J., MoI. Biotechnol. 16:151-160, 2000; Werner, R.G., Drug Res. 48:870-880, 1998. In one specific embodiment, the cell culture is a mammalian cell culture, such as a Chinese hamster ovary (CHO) cell culture.

In one embodiment, the antibody may be recovered or isolated from a naturally occurring source. In some such embodiments, the isolated or recovered antibody may undergo additional purification steps using conventional chromatographic methods known in the art. Specifically, the purification methods are contemplated, but are not limited to, affinity chromatography (e.g., protein A affinity chromatography), ion exchange chromatography (e.g., anion exchange chromatography or cation exchange chromatography), hydrophobic interaction chromatography, hydroxyapatite chromatography, gel filtration chromatography, and/or dialysis. Of these, protein A chromatography is the preferred purification method. The matrix for affinity ligand linkage is most often agarose, but other matrices may be used.

Other techniques for antibody purification are also known in the art, such as fractionation on ion exchange columns, ethanol precipitation, reversed-phase high-performance chromatography, reversed-phase HPLC, chromatography on silica, chromatography on heparin Sepharose™, chromatography on anion or cation exchange resins (such as polyaspartic acid columns), chromatographic focusing, electrophoresis using SDS-PAGE, and ammonium sulfate precipitation. The list of purification methods above is merely illustrative in nature and is not intended to be limiting.

Alternatively, transgenic animals (e.g., mice) can be produced that, upon immunization, generate a complete human antibody library in the absence of endogenous immunoglobulin production. For example, homozygous deletion of the antibody heavy chain linker (J.sub.H) gene in chimeric and germline mutant mice has been described as resulting in complete suppression of endogenous antibody production. Transplantation of human germline immunoglobulin gene arrays into such germline mutant mice will lead to the production of human antibodies upon antigen challenge. (See, for example, Jakobovits et al., Proc. Natl. Acad. Sci. USA 90:2551-255;1993 and Jakobovits et al., Nature 362:255-258, 1993).

In some embodiments, the antibodies of the present invention may be humanized, retaining high affinity for the antigen and other beneficial biological properties. Methods for humanizing nonhuman antibodies are well known in the art. Humanization can be essentially performed following the approach of Winter and collaborators (Jones et al., Nature 321:522-525, 1986; Riechmann et al., Nature 332:323-327, 1988; Verhoeyen et al., Science 239:1534-1536, 1988) by replacing the corresponding sequence of a human antibody with a rodent CDR or CDR sequence. Thus, such humanized antibodies are chimeric antibodies (Cabilly, ibid.), in which essentially less than one fully human variable domain has been replaced by a corresponding sequence from a non-human species. In practice, humanized antibodies are typically human antibodies, with some CDR residues and possibly some FR residues replaced by residues from similar sites in rodent antibodies. Importantly, the antibody is humanized, retaining its high affinity for the antigen and other beneficial biological properties. To achieve this, humanized antibodies are prepared according to a preferred method by analyzing the parental sequence and various conceptual humanized products using a three-dimensional model of the parental and humanized sequences. Three-dimensional immunoglobulin models are familiar to those skilled in the art. A computer program can be used to illustrate and demonstrate the possible three-dimensional conformational structures of selected candidate immunoglobulin sequences. Observing these displays allows analysis of the possible roles of residues in the function of the candidate immunoglobulin sequence, i.e., analysis of residues affecting the ability of the candidate immunoglobulin to bind its antigen. In this way, FR residues can be selected and merged from common and input sequences to achieve desired antibody properties, such as increased affinity for the target antigen. For further details, see WO92/22653, published December 23, 1992.

Various phage display methods known in the art can also be used to generate the antibodies of the present invention. In phage display methods, functional antibody domains are displayed on the surface of phage particles carrying a polynucleotide sequence encoding them. In particular, this phage can be used to display antigen-binding domains (such as Fab and Fv or disulfide-stable Fv) expressed from a library or combinatorial antibody library (e.g., human or mouse). Phages expressing antigen-binding domains that bind to target antigens can be selected or identified using antigens (e.g., using labeled antigens or antigens that bind to or capture onto solid surfaces or beads). The phages used in these methods are typically filamentous phages, including fd and M13. The antigen-binding domains are expressed as proteins recombinantly fused to phage gene III or gene VIII proteins. Examples of phage display methods that can be used to prepare the immunoglobulins or fragments thereof of the present invention include those disclosed in Brinkmann et al., "Phage Display Of Disulfide-Stabilized FvFragments," J. Immunol. Methods, 182:41-50, 1995.

The functional properties of various IgG isoforms and their domains are well known in the art. The amino acid sequences of IgG1, IgG2, IgG3, and IgG4 are known in the art. The selection and/or combination of two or more domains from a particular IgG isoform in the methods used in this invention can be based on any known parameters of the parental isoform (including affinity for FcγR). For example, using regions or domains from IgG isoforms (e.g., IgG2 or IgG4) that exhibit limited or no binding to FcγRIIB can be particularly useful in cases where it is desired to engineer bispecific antibodies to maximize binding to activating receptors and minimize binding to inhibitory receptors. Similarly, using Fc chains or domains from IgG isoforms (e.g., IgG3) that are known to preferentially bind to C1q or FcγRIIIA can be combined with Fc amino acid modifications known in the art to enhance antibody-dependent cell-mediated cytotoxicity (ADCC) and/or complement-dependent cytotoxicity (CDC) to engineer bispecific antibodies such that effector functional activity (e.g., complement activation or ADCC) is maximized. In a similar manner, mutations can be made in the Fc chain or domain of the IgG isotype to minimize or eliminate the effector function of the Fc chain.

During the discovery process, antibody generation and characterization can elucidate information about the desired epitope. Based on this information, antibodies can be competitively screened for binding to the same epitope. This is achieved by conducting competitive and cross-competitive studies to find antibodies that compete or cross-compete with each other (e.g., antibodies that competitively bind to antigens). One approach is to identify the epitope that the antibody binds to, or “epitope localization.” Many methods known in the art exist for mapping and characterizing the location of epitopes on proteins, including resolving the crystal structure of antibody-antigen complexes, competitive assays, gene fragment expression assays, and synthetic peptide-based assays, as described, for example, in Chapter 11 of Harlow and Lane, Using Antibodies, a Laboratory Manual, Cold SpringHarbor Laboratory Press, Cold SpringHarbor, N.Y., 1999. In an additional instance, epitope mapping can be used to determine the sequence to which the antibody binds. Epitope mapping is commercially available from various sources, such as Pepscan Systems (Edelhertweg 15, 8219 PH Lelystad, Netherlands). In an additional instance, mutagenesis of the antigen-binding domain, domain exchange experiments, and alanine partitioning mutagenesis can be performed to identify residues required, sufficient, and/or essential for epitope binding. Binding affinity and the dissociation rate of the antigen-binding domain interaction can be determined by a competitive binding assay. An example of a competitive binding assay is a radioimmunoassay, which involves incubating a labeled antigen and detecting the molecule binding to the labeled antigen. The affinity and binding-dissociation rate of the molecules of the present invention for the antigen can be determined from saturation data using Scatchard analysis.

The affinity and binding properties of the antibodies of the present invention to antigens can first be determined using in vitro assays (biochemical or immunoassays) known in the art for antigen-binding domains, including but not limited to enzyme-linked immunosorbent assays (ELISA), surface plasmon resonance (SPR) assays, biolayer interferometry, or immunoprecipitation assays. The molecules of the present invention may have similar binding properties in in vivo models (such as those described and disclosed herein) to those in in vitro-based assays. However, the present invention does not exclude molecules of the present invention that do not exhibit the desired phenotype in in vitro-based assays but do exhibit the desired phenotype in vivo.

Once the nucleic acid sequence encoding the molecule of the present invention (i.e., the binding domain) has been obtained, a vector for generating the molecule can be produced using recombinant DNA technology, a technique well known in the art.

The polynucleotide encoding the antibody-binding domain of the present invention may include an expression control polynucleotide sequence operatively linked to the antibody-coding sequence, including naturally associated or heterologous promoter regions known in the art. The expression control sequence may be a eukaryotic promoter system in a vector capable of transforming or transfecting eukaryotic host cells, but a prokaryotic host control sequence may also be used. Once the vector has been incorporated into a suitable host cell line, the host cells are allowed to proliferate under conditions suitable for expressing the nucleotide sequence and are used, as needed, to collect and purify the antibody. Eukaryotic cell lines include CHO cell lines, various COS cell lines, HeLa cells, myeloma cell lines, transformed B cells, or human embryonic kidney cell lines.

In some embodiments, the DNA encoding the antibodies of the present invention is isolated and sequenced using conventional procedures, e.g., by using oligonucleotide probes capable of specifically binding to genes encoding the heavy and light chains of mouse antibodies. Hybridoma cells of the present invention serve as a preferred source of this DNA. Once isolated, the DNA can be placed in an expression vector, which is then transfected into host cells that otherwise do not produce immunoglobulin proteins (such as simian COS cells, CHO cells, or myeloma cells) to obtain monoclonal antibody synthesis in recombinant host cells. The DNA can also be modified, for example, by replacing the homologous mouse sequences with coding sequences for the constant domains of the human heavy and light chains (Morrison et al., Proc. Nat. Acad. Sci. 81:6851, 1984). In this manner, chimeric antibodies having the binding specificity of the anti-antigen monoclonal antibodies described herein are prepared.

CD3 antibody

In one aspect, the present invention provides a reagent that specifically binds to CD3 (e.g., the full-length human CD3 ε subunit (e.g., accession number NP_000724 or SEQ ID NO: 66)).

In some implementations, the CD3 antibody specifically binds to human CD3.

Exemplary CD3 antibodies of the present invention are listed in Tables 1, 2, and 3 herein. Table 1 shows the amino acid sequence identifiers of the VH and VL regions. In some embodiments, antibodies are provided having any of the partial light chain sequences listed in Table 1 and/or any of the partial heavy chain sequences listed in Table 1.

Table 1

In Table 1, the underlined sequences are CDR sequences based on Kabat, and the bold sequences are CDR sequences based on Chothia.

This invention provides the CDR portion (including Chothia, Kabat CDR, and CDR contact region) of an antibody against CD3. Methods and techniques for identifying CDRs within VH and VL amino acid sequences are well known in the art and can be used to identify CDRs within the specified VH and/or VL amino acid sequences disclosed herein. Exemplary conventions for identifying CDR boundaries include, for example, the Kabat definition, the Chothia definition, and the AbM definition. Generally, the Kabat definition is based on sequence variability, the Chothia definition is based on the location of structural loop regions, and the AbM definition is a compromise between the Kabat and Chothia methods. See, for example, Kabat, "Sequences of Proteins of Immunological Interest," National Institutes of Health, Bethesda, Md. (1991); Al-Lazikani et al., J. Mol. Biol. 273:927-948 (1997); and Martin et al., Proc. Natl. Acad. Sci. USA 86:9268-9272 (1989). Public databases may also be used to identify CDR sequences within the antibody. It should be understood that in some embodiments, the CDR may be a combination of Kabat and Chothia CDRs (also referred to as a "combined CR" or "extended CDR"). In some embodiments, the CDR is the Kabat CDR. In some further embodiments, the CDR is the Chothia CDR. In other words, in an implementation having more than one CDR, the CDR can be any of Kabat, Chothia, combined CDRs, or a combination thereof. Tables 2 and 3 provide examples of the CDR sequences provided herein.

In one embodiment, the present invention provides an antibody comprising the amino acid sequences listed in Table 1 or substantially similar sequences having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with them.

In some such embodiments, the present invention provides an antibody comprising a VL region comprising an amino acid sequence of any of the VL region amino acid sequences listed in Table 1 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with it.

In some other embodiments of this invention, the present invention provides an antibody comprising a VH region comprising an amino acid sequence of any of the VH region amino acid sequences listed in Table 1 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with it.

In one embodiment, an antibody that specifically binds to CD3 is provided, wherein the antibody comprises: (a) a heavy chain variable (VH) region comprising VH complementarity-determining region 1 (VH CDR1), VH complementarity-determining region 2 (VH CDR2), and VH complementarity-determining region 3 (VH CDR3) of the VH sequence shown in SEQ ID NO: 2, 4, 5, 7, 10, 12, or 35; and/or (b) a light chain variable (VL) region comprising VL complementarity-determining region 1 (VL CDR1), VL complementarity-determining region 2 (VL CDR2), and VL complementarity-determining region 3 (VL CDR3) of the VL sequence shown in SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87, or 89.

In one embodiment, the present invention provides an antibody comprising a set of six CDRs (i.e., VL CDR1, VL CDR2, VLCDR3, VH CDR1, VH CDR2, VH CDR3), wherein the CDRs are contained within the VL/VH amino acid sequence pairs as defined by any of the exemplary CD3 antibodies listed in Table 1.

In specific embodiments, the present invention provides an antibody comprising a set of amino acid sequences of VL CDR1, VL CDR2, VL CDR3, VHCDR1, VHCDR2, or VH CDR3, wherein the set of amino acid sequences is contained within a VL/VH amino acid sequence pair selected from the following: SEQ ID NO: 1 and 2 (e.g., 2B5-0001 (2B5)); SEQ ID NO: 3 and 2 (2B5-0002); SEQ ID NO: 3 and 4 (e.g., 2B5-0006 (2B5v6)); SEQ ID NO: 3 and 5 (e.g., 2B5-0009); SEQ ID NO: 6 and 7 (e.g., 2B5-0517); SEQ ID NO: 8 and 7 (e.g., 2B5-0522); SE QID NO: 6 and 4 (e.g., 2B5-0533); SEQ ID NO: 8 and 4 (e.g., 2B5-0538); SEQ ID NO: 9 and 10 (e.g., 2B5-0598); SEQ ID NO: 9 and 7 (e.g., 2B5-0623); SEQ ID NO: 11 and 12 (e.g., 2B5-0707); SEQ ID NO: 34 and 35 (e.g., 2B5-0003); SEQ ID NO: 85 and 2 (e.g., H2B4); SEQ ID NO: 9 and 4 (e.g., 2B5-1038); SEQ ID NO: 87 and 4 (e.g., 2B5-1039); and SEQ ID NO: 89 and 4 (e.g., 2B5-1040).

Table 2 shows the VH CDR1, VH CDR2, and VH CDR3 regions of an exemplary CD3 antibody.

Table 2

In one embodiment, the present invention provides an antibody comprising VH CDR1, said VH CDR1 comprising an amino acid sequence of any of the VHCDR1 amino acid sequences listed in Table 2 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In another embodiment, the present invention provides an antibody comprising VH CDR2, said VH CDR2 comprising an amino acid sequence of any of the VHCDR2 amino acid sequences listed in Table 2 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In yet another embodiment, the present invention provides an antibody comprising VH CDR3, said VH CDR3 comprising an amino acid sequence of any of the VHCDR3 amino acid sequences listed in Table 2 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

Table 3 shows exemplary CD3 antibodies VL CDR1, VL CDR2, and VL CDR3.

Table 3

In one embodiment, the present invention provides an antibody comprising VL CDR1, said VL CDR1 comprising an amino acid sequence of any of the VLCDR1 amino acid sequences listed in Table 3 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In another embodiment, the present invention provides an antibody comprising VL CDR2, said VL CDR2 comprising an amino acid sequence of any of the VLCDR2 amino acid sequences listed in Table 3 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In yet another embodiment, the present invention provides an antibody comprising VL CDR3, said VL CDR3 comprising an amino acid sequence of any of the VLCDR3 amino acid sequences listed in Table 3 or a substantially similar sequence having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

In some embodiments, an antibody that specifically binds to CD3 is provided, wherein the antibody comprises: (a) a heavy chain (HC) variable (VH) region comprising VH complementarity-determining region 1 (CDR1), VH complementarity-determining region 2 (VH CDR2), and VH complementarity-determining region 3 (VH CDR3) of VH having an amino acid sequence selected from SEQ ID NO: 2, 4, 5, 7, 10, 12, and 35; and/or (b) a light chain (LC) variable (VL) region comprising VL complementarity-determining region 1 (VL CDR1), VL complementarity-determining region 2 (VL CDR2), and VL complementarity-determining region 3 (VLCDR3) of VL having an amino acid sequence selected from SEQ ID NO: 1, 3, 6, 8, 9, 11, 34, 87, and 89.

In a specific embodiment, the antibody comprises: (a) a VH region comprising (i) a VH CDR1 comprising the sequence of SEQ ID NO: 13, 14 or 15; (ii) a VHCDR2 comprising the sequence of SEQ ID NO: 16, 17, 19, 20, 21, 22, 23 or 24; and (iii) a VH CDR3 comprising the sequence of SEQ ID NO: 18 or 25; and/or (b) a VL region comprising (i) a VL CDR1 comprising the sequence of SEQ ID NO: 26, 29, 30, 32, 91 or 92; (ii) a VL CDR2 comprising the sequence of SEQ ID NO: 27 or 31; and (iii) a VL CDR3 comprising the sequence of SEQ ID NO: 28 or 33.

In some embodiments, the present invention provides antibodies that bind to CD3 and compete with antibodies as described herein, including 2B5-0001 (2B5), 2B5-0002, 2B5-0006 (2B5v6), 2B5-0009, 2B5-0517, 2B5-0522, 2B5-0533, 2B5-0538, 2B5-0598, 2B5-0623, 2B5-0707, 2B5-0003, 2B5-1038, 2B5-1039, and 2B5-1040.

In some embodiments, the present invention also provides the CDR portion of a CD3 antibody based on a CDR contact region. A CDR contact region is an antibody region that imparts specificity to an antigen to the antibody. Typically, the CDR contact region comprises residue sites within the CDR and Vernier region, constrained to maintain a suitable loop structure for antibody binding to specific antigens. See, for example, Makabe et al., J. Biol. Chem., 283:1156-1166, 2007. Determination of the CDR contact region is entirely within the scope of the art.

Some of the CD3 antibodies described herein may be referred to by more than one name. For example, 2B5-0001 may be referred to as 2B5 or h2B5; and 2B5-0006 may be referred to as 2B5v6 or h2B5v6. Table 4 shows the forms of the antibodies and the CD3 antibody regions present in the antibodies disclosed herein.

Table 4

sequence	describe	form	CD3 antibody region
				2B5-0001 (2B5)	CD3 antibody	IgG	2B5-0001
2B5-0002	CD3 antibody	IgG	2B5-0002
				2B5-0003	CD3 antibody	IgG	2B5-0003
2B5-0006 (2B5v6)	CD3 antibody	IgG	2B5-0006
				2B5-0009	CD3 antibody	IgG	2B5-0009
2B5-0517	CD3 antibody	IgG	2B5-0517
				2B5-0522	CD3 antibody	IgG	2B5-0522
2B5-0533	CD3 antibody	IgG	2B5-0533
				2B5-0538	CD3 antibody	IgG	2B5-0538
2B5-0598	CD3 antibody	IgG	2B5-0598
				2B5-0623	CD3 antibody	IgG	2B5-0623
2B5-0707	CD3 antibody	IgG	2B5-0707
				2B5-1038	CD3 antibody	IgG	2B5-1038
2B5-1039	CD3 antibody	IgG	2B5-1039
				2B5-1040	CD3 antibody	IgG	2B5-1040
H2B4	CD3 antibody	IgG	H2B4
				GUCY2C-H2B4	CD3-GUCY2c bispecific antibody	CD3 bispecific	H2B4
GUCY2C-2B5	CD3-GUCY2c bispecific antibody	CD3 bispecific	2B5-0001
				GUCY2C-2B5v6	CD3-GUCY2c bispecific antibody	CD3 bispecific	2B5-0006
xFLT3-2B5v6	CD3-FLT3 bispecific antibody	CD3 bispecific	2B5-0006

The binding affinity ( _KD ) of the antibody described herein for CD3 (such as human CD3ε (e.g., SEQ ID NO:40) can be from about 0.001 nM to about 500 nM. In one embodiment, the binding affinity is about any of the following: 6500 nM, 6000 nM, 5500 nM, 4500 nM, 4000 nM, 3500 nM, 3000 nM, 2500 nM, 2000 nM, 1500 nM, 1000 nM, 750 nM, 500 nM, 400 nM, 300 nM, 250 nM, 200 nM, 150 nM, 100 nM, 75 nM, 50 nM, 45 nM, 40 nM, 35 nM, 30 nM, 25 nM, 20 nM, 19 nM, 17 nM, 16 nM, 15 nM, 10 nM, 8 nM, 7.5 nM, 7 nM, 6.5 nM, 6 nM, 5.5 nM, 5 nM, 4 nM, 3 nM, 2 nM, 1 nM, 0.5 nM, 0.3 nM, 0.1 nM, 0.01 nM, 0.002 nM, or 0.001 nM. In some embodiments, the binding affinity is at least about any one of the following: 6500 nM, 6000 nM, 5500 nM, 5000 nM, 4000 nM, 3000 nM, 2000 nM, 1000 nM, 900 nM, 800 nM, 500 nM, 250 nM, 200 nM, 100 nM, 50 nM, 30 nM, 20 nM, 10 nM, 7.5 nM, 7 nM, 6.5 nM, 6 The effective values are nM, 5 nM, 4.5 nM, 4 nM, 3.5 nM, 3 nM, 2.5 nM, 2 nM, 1.5 nM, 1 nM, 0.5 nM, 0.1 nM, 0.05 nM, 0.01 nM, or lower. In one specific embodiment, the antibody has a _KD of about 80 to about 200 pM. In another specific embodiment, the antibody has a _KD of about 10 nm to 2000 nm.

In one aspect, the present invention provides an antibody or pharmaceutical composition for use in a method of modulating a T-cell-mediated immune response in a subject with such a need. In a specific embodiment, the present invention provides an antibody or pharmaceutical composition for use in a method of inhibiting the growth of tumor cells in a subject.

In another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, optionally wherein said cancer is selected from: breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer, bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and cancers of the digestive system.

In yet another aspect, the present invention provides antibody or pharmaceutical compositions for treating cancer, wherein a T-cell-mediated immune response is modulated or wherein the growth of tumor cells is inhibited.

The present invention also provides nucleic acid molecules encoding CD3 antibodies or portions thereof. For example, the present invention provides nucleic acid molecules encoding any VH or VL amino acid sequence listed in Table 1 or substantially similar sequences having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity with them.

The present invention also provides nucleic acid molecules of any of the amino acid sequences of VH CDR1, VH CDR2 or VH CDR3 listed in Table 2 or substantially similar sequences having at least 90%, at least 95%, at least 98% or at least 99% sequence identity with them.

The present invention also provides nucleic acid molecules having a substantially similar sequence to any of the VL CDR1, VL CDR2 or VL CDR3 amino acid sequences listed in Table 3, or having at least 90%, at least 95%, at least 98% or at least 99% sequence identity with it.

The present invention also provides a nucleic acid molecule encoding HC VR, wherein the VH comprises a set of three CDRs (i.e., VH CDR1, VHCDR2, or VH CDR3), wherein the amino acid sequence sets of VH CDR1, VH CDR2, and VH CDR3 are as defined by any of the exemplary CD3 antibodies listed in Table 2.

The present invention also provides a nucleic acid molecule encoding LC VR, wherein the VL comprises a set of three CDRs (i.e., VL CDR1, VLCDR2, or VL CDR3), wherein the amino acid sequence sets of VL CDR1, VL CDR2, and VL CDR3 are as defined by any of the exemplary CD3 antibodies listed in Table 3.

The present invention also provides nucleic acid molecules encoding both VH and VL, wherein the VH comprises an amino acid sequence of any VH amino acid sequence listed in Table 1, and wherein the VL comprises an amino acid sequence of any VL amino acid sequence listed in Table 1.

The present invention also provides recombinant expression vectors capable of expressing polypeptides containing heavy or light chain variable regions of CD3 antibodies. For example, the present invention includes recombinant expression vectors comprising any of the nucleic acid molecules mentioned above (i.e., nucleic acid molecules encoding any of the VH and VL shown in Table 1 and/or CDR sequences as shown in Table 2 or 3). The scope of the present invention also includes host cells in which such vectors have been introduced, and methods for producing said antibodies or portions thereof by culturing said host cells under conditions allowing the production of antibodies or antibody fragments and recovering said antibodies and antibody fragments thus produced.

In some embodiments, the polynucleotide comprises a sequence encoding the heavy chain and/or light chain variable region of the following antibodies listed in Table 4: 2B5-0001 (2B5), 2B5-0002, 2B5-0006 (2B5v6), 2B5-0009, 2B5-0517, 2B5-0522, 2B5-0533, 2B5-0538, 2B5-0598, 2B5-0623, 2B5-0707, 2B5-0003, 2B5-1038, 2B5-1039, or 2B5-1040.

The sequence encoding the target antibody can be maintained in a vector within a host cell, and the host cell can then be amplified and frozen for future use. Vectors (including expression vectors) and host cells are further described herein.

Polynucleotides complementary to any such sequence are also covered by this invention. In one aspect, this invention provides a method for preparing any of the polynucleotides described herein. The polynucleotide may be single-stranded (coding or antisense) or double-stranded, and may be DNA (genomic, cDNA, or synthetic) or RNA molecules. RNA molecules include HnRNA molecules containing introns and corresponding one-to-one with DNA molecules; and mRNA molecules that do not contain introns. Additional coding or non-coding sequences may, but do not necessarily, be present in the polynucleotides of this invention, and the polynucleotides may, but do not necessarily, be linked to other molecules and/or supporting materials.

The polynucleotide may contain a native sequence (i.e., an endogenous sequence encoding an antibody or a portion thereof) or a variant thereof. Polynucleotide variants contain one or more substitutions, additions, deletions, and/or insertions such that the immunoreactivity of the encoded polypeptide is not reduced relative to the native immunoreactive molecule. The effect on the immunoreactivity of the encoded polypeptide can generally be evaluated as described herein. The variant preferably exhibits at least about 70% identity with the polynucleotide sequence encoding a native antibody or a portion thereof, more preferably at least about 80% identity, even more preferably at least about 90% identity, and most preferably at least about 95% identity.

The polynucleotides of this invention can be obtained using chemical synthesis, recombination methods, or PCR. Methods for the chemical synthesis of polynucleotides are well known in the art and need not be described in detail herein. Those skilled in the art can use the sequences provided herein and commercial DNA synthesizers to generate the desired DNA sequences.

To prepare polynucleotides using recombinant methods, a polynucleotide containing the desired sequence can be inserted into a suitable vector, which can then be introduced into a suitable host cell for replication and amplification, as discussed further herein. Polynucleotides can be inserted into host cells by any means known in the art. Cells are transformed by introducing exogenous polynucleotides through direct uptake, endocytosis, transfection, F-conjugation, or electroporation. Once introduced, the exogenous polynucleotide can be maintained within the cell as a non-integrating vector (such as a plasmid) or integrated into the host cell genome. The amplified polynucleotides can then be isolated from the host cell using methods well-known in the art. See, for example, Sambrook et al., 1989.

Alternatively, PCR allows for the replication of DNA sequences. PCR technology is well known in the art and is described in U.S. Patent Nos. 4,683,195, 4,800,159, 4,754,065, and 4,683,202, and PCR: The Polymerase Chain Reaction, edited by Mullis et al., Birkauswer Press, Boston, 1994.

RNA can be obtained by using DNA isolated in a suitable vector and inserted into a suitable host cell. Once the cell replicates and the DNA is transcribed into RNA, RNA can then be isolated using methods well known to those skilled in the art, such as Sambrook et al., 1989, as described above.

Suitable cloning vectors can be constructed according to standard techniques or selected from a large number of cloning vectors available in the art. While the chosen cloning vector can vary depending on the intended host cell, a useful cloning vector will generally be capable of self-replication, may have a single target for a specific restriction endonuclease, and/or may carry genes related to markers that can be used to select clones containing that vector. Suitable examples include plasmids and bacterial viruses such as pUC18, pUC19, Bluescript (e.g., pBS SK+) and its derivatives, mp18, mp19, pBR322, pMB9, ColE1, pCR1, RP4, phage DNA, and shuttle vectors such as pSA3 and pAT28. These and many other cloning vectors are available from commercial vendors such as BioRad, Strategene, and Invitrogen.

Expression vectors are typically reproducible polynucleotide constructs containing polynucleotides according to the invention. This implies that the expression vector must be replicable in the host cell as an episome or as an integral part of chromosomal DNA. Suitable expression vectors include, but are not limited to, plasmids, viral vectors including adenoviruses, adeno-associated viruses, retroviruses, kinases, and the expression vectors disclosed in PCT Publication No. WO 87/04462. Vector components may typically include, but are not limited to, one or more of the following: a signal sequence; an origin of replication; one or more marker genes; and suitable transcriptional control elements (such as promoters, enhancers, and terminators). For expression (i.e., translation), one or more translational control elements, such as ribosome binding sites, translation initiation sites, and stop codons, are also typically required.

Vectors containing the target polynucleotide can be introduced into host cells by any of a number of suitable means, including electroporation, transfection with calcium chloride, rubidium chloride, calcium phosphate, DEAE-glucan, or other substances; particle bombardment; lipid transfection; and infection (e.g., when the vector is an infectious pathogen such as vaccinia virus). The choice of vector or polynucleotide to be introduced will often depend on the characteristics of the host cell.

This invention also provides host cells comprising any of the polynucleotides described herein. Any host cell capable of overexpressing heterologous DNA can be used for the purpose of isolating genes encoding target antibodies, peptides, or proteins. Non-limiting examples of mammalian host cells include, but are not limited to, COS, HeLa, and CHO cells. See also PCT Publication WO 87/04462. Suitable non-mammalian host cells include prokaryotes (such as Escherichia coli or Bacillus subtilis ) and yeasts (such as Saccharomyces cerevisiae , Saccharomyces pombe , or Kluyveromyces lactis ). Preferably, the host cell expresses cDNA at a level about 5-fold higher, more preferably 10-fold higher, and even more preferably 20-fold higher than the corresponding endogenous antibody or protein CD3 or CD3 domain (e.g., domains 1-4), as determined by immunoassay or FACS. Cells overexpressing target antibodies or proteins can be identified.

This invention also covers scFvs of the antibodies of this invention. Single-chain variable region fragments are prepared by linking light and/or heavy chain variable regions using a short linker peptide (Bird et al., Science 242:423-426, 1988). Examples of linker peptides include the sequence of SEQ ID NO: 65. In one embodiment, the linker peptide bridges approximately 3.5 nm between the carboxyl terminus of one variable region and the amino terminus of another variable region. Linkers with other sequences have been designed and used (Bird et al., Science 242:423-426, 1988). Linkers should be short, flexible polypeptides and preferably contain fewer than about 20 amino acid residues. The linker can then be modified for additional functions, such as linking drugs or solid carriers. Single-chain variants can be generated recombinantly or synthetically. An automated synthesizer can be used to synthesize scFvs. To generate scFv through recombination, a suitable plasmid containing a polynucleotide encoding the scFv can be introduced into a suitable host cell, which can be eukaryotic (such as yeast, plant, insect, or mammalian cells) or prokaryotic (such as Escherichia coli). The polynucleotide encoding the target scFv can be prepared by conventional manipulation (such as linking the polynucleotide). The obtained scFv can be isolated using standard protein purification techniques known in the art.

In one aspect, polynucleotide sequences encoding any of the partial light chain sequences and/or any of the partial heavy chain sequences disclosed herein are provided. Exemplary CD3 polynucleotides encoding the VH and VL regions of the antibodies of the present invention are listed in Table 5.

Table 5

In one aspect, the present invention provides compositions (such as pharmaceutical compositions) comprising any of the polynucleotides of the present invention. In some embodiments, the composition may comprise any of the polynucleotides shown in Table 5. In a particular embodiment, the composition comprises an expression vector containing a polynucleotide encoding any of the antibodies described herein (e.g., any of the polynucleotides shown in Table 5).

This invention also covers polynucleotides complementary to any such sequence. Polynucleotides can be single-stranded (coding or antisense) or double-stranded, and can be DNA (genomic, cDNA, or synthetic) or RNA molecules. RNA molecules include mature and immature mRNAs, such as precursor mRNA (pre-mRNA) or heteronuclear mRNA (hnRNA) and mature mRNA. Additional coding or non-coding sequences may, but do not necessarily, be present within the polynucleotides of this invention, and the polynucleotides may, but do not necessarily, be linked to other molecules and/or supporting materials.

The polynucleotide may contain a native sequence (i.e., an endogenous sequence encoding an antibody or a portion thereof) or a variant thereof. Polynucleotide variants contain one or more substitutions, additions, deletions, and/or insertions such that the immunoreactivity of the encoded polypeptide is not reduced relative to the native immunoreactive molecule. The effect on the immunoreactivity of the encoded polypeptide can generally be evaluated as described herein. Variants preferably exhibit at least about 70% identity with the polynucleotide sequence encoding a native antibody or a portion thereof, more preferably at least about 80% identity, even more preferably at least about 90% identity, and most preferably at least about 95% identity.

Two polynucleotide or polypeptide sequences are said to be "identical" if the nucleotide or amino acid sequences in the two sequences are identical when compared for maximum correspondence as described below. Comparisons between two sequences are typically performed by comparing sequences on a comparison window to identify and compare local regions of sequence similarity. As used herein, a "comparison window" refers to a segment of at least about 20, typically 30 to about 75, or 40 to about 50 adjacent positions, where, after optimal alignment of the two sequences, the sequence can be compared to a reference sequence having the same number of adjacent positions.

Using default parameters, the Megalign procedure in the Lasergene Bioinformatics Software Suite (DNASTAR, Inc., Madison, WI) allows for optimal alignment of sequences for comparison. Preferably, the "percentage of sequence identity" is determined by comparing two optimally aligned sequences across a comparison window of at least 20 positions. The portion of the polynucleotide or polypeptide sequence in the comparison window may contain 20% or less, typically 5 to 15%, or 10 to 12%, of additions or deletions (i.e., gaps) compared to a reference sequence (which does not contain additions or deletions) for optimal alignment of the two sequences. The percentage is calculated by determining the number of positions in both sequences where the same nucleic acid base or amino acid residue appears, to obtain the number of matching positions; dividing the number of matching positions by the total number of positions in the reference sequence (i.e., the window size); and multiplying the result by 100 to obtain the percentage of sequence identity.

The variant can also, or alternatively, be substantially homologous to the natural gene or a portion thereof or its complement. This polynucleotide variant is capable of hybridizing with naturally occurring DNA sequences (or complementary sequences) encoding natural antibodies under moderately stringent conditions.

Those skilled in the art will understand that, due to the degeneracy of the genetic code, there are many nucleotide sequences encoding polypeptides as described herein. Some of these polynucleotides have minimal homology to the nucleotide sequences of any natural gene. However, the present invention specifically considers polynucleotides that vary due to differences in codon usage. Furthermore, alleles of genes comprising the polynucleotide sequences provided herein are within the scope of the invention. An allele is an endogenous gene altered due to one or more mutations (such as nucleotide deletions, additions, and/or substitutions). The resulting mRNA and protein may, but need not, have altered structure or function. Alleles can be identified using standard techniques (such as hybridization, amplification, and/or database sequence comparison).

In one aspect, the present invention provides a method for preparing any of the polynucleotides described herein. For example, the polynucleotides of the present invention can be obtained using chemical synthesis, recombination methods, or PCR. Methods for the chemical synthesis of polynucleotides are well known in the art and need not be described in detail herein. Those skilled in the art can use the sequences provided herein and commercial DNA synthesizers to generate desired DNA sequences.

To prepare polynucleotides using recombinant methods, a polynucleotide containing the desired sequence can be inserted into a suitable vector, which can then be introduced into a suitable host cell for replication and amplification, as discussed further herein. Polynucleotides can be inserted into host cells by any means known in the art. Cells are transformed by introducing exogenous polynucleotides through direct uptake, endocytosis, transfection, F-crossing, or electroporation. Once introduced, the exogenous polynucleotide can be maintained within the cell as a non-integrating vector (such as a plasmid) or integrated into the host cell genome. The amplified polynucleotides can then be isolated from the host cell using methods well-known in the art (e.g., Sambrook et al., 1989).

Alternatively, PCR allows for the replication of DNA sequences. PCR technology is well known in the art and is described in U.S. Patent Nos. 4,683,195, 4,800,159, 4,754,065, and 4,683,202, and PCR: The Polymerase Chain Reaction, edited by Mullis et al., Birkauswer Press, Boston, 1994.

RNA can be obtained by using DNA isolated in a suitable vector and inserted into a suitable host cell. Once the cell replicates and the DNA is transcribed into RNA, RNA can then be isolated using methods well known to those skilled in the art, such as Sambrook et al., 1989, as described above.

Suitable cloning vectors can be constructed using standard techniques or selected from a large number of cloning vectors available in the art. While the chosen cloning vector can vary depending on the intended host cell, useful cloning vectors typically possess self-replicating capabilities, may have a single target for a specific restriction endonuclease, and/or may carry genes related to markers that can be used to select clones containing that vector. Suitable examples include plasmids and bacterial viruses such as pUC18, pUC19, Bluescript (e.g., pBS SK+) and its derivatives, mp18, mp19, pBR322, pMB9, ColE1, pCR1, RP4, phage DNA, and shuttle vectors such as pSA3 and pAT28. These and many other cloning vectors are available from commercial vendors such as BioRad, Strategene, and Invitrogen.

Expression vectors are typically reproducible polynucleotide constructs containing polynucleotides according to the invention. This implies that the expression vector must be replicable in the host cell as an episome or as an integral part of chromosomal DNA. Suitable expression vectors include, but are not limited to, plasmids, viral vectors including adenoviruses, adeno-associated viruses, retroviruses, viscera, and the expression vectors disclosed in PCT Publication No. WO87/04462. Vector components may typically include, but are not limited to, one or more of the following: a signal sequence; an origin of replication; one or more marker genes; and suitable transcriptional control elements (such as promoters, enhancers, and terminators). For expression (i.e., translation), one or more translational control elements, such as ribosome binding sites, translation initiation sites, and stop codons, are also typically required.

Vectors containing the target polynucleotide can be introduced into host cells by any of a number of suitable means, including electroporation, transfection with calcium chloride, rubidium chloride, calcium phosphate, DEAE-glucan, or other substances; particle bombardment; lipid transfection; and infection (e.g., when the vector is an infectious pathogen such as vaccinia virus). The choice of vector or polynucleotide to be introduced will often depend on the characteristics of the host cell.

Any host cell capable of overexpressing heterologous DNA can be used for the purpose of isolating genes encoding target antibodies, peptides, or proteins. Non-limiting examples of mammalian host cells include, but are not limited to, COS, HeLa, and CHO cells. See also PCT Publication WO87/04462. Suitable non-mammalian host cells include prokaryotes (such as *Escherichia coli* or *Bacillus subtilis *) and yeasts (such as *Saccharomyces cerevisiae *, *Schizosaccharomyces pombe * , or *Kluyveromyces lactis *). Preferably, the host cell expresses cDNA at a level approximately 5-fold higher, more preferably 10-fold higher, and even more preferably 20-fold higher than the corresponding endogenous antibody or protein CD3 or CD3 domain (e.g., domains 1-4), as determined by immunoassay or FACS. Cells overexpressing the target antibody or protein can be identified.

In one respect, the CD3 antibody molecule will have an affinity for CD3, as measured, for example, by direct or competitive binding in the picomolar to micromolar affinity range, preferably in the picomolar to low nanomolar range.

Bispecific antibodies can be prepared using the sequences disclosed herein. Methods for preparing bispecific antibodies are known in the art (see, for example, Suresh et al., Methods in Enzymology 121:210, 1986). Traditionally, the recombinant generation of bispecific antibodies is based on the co-expression of two immunoglobulin heavy-light chain pairs, where the two heavy chains have different specificities (Millstein and Cuello, Nature 305, 537-539, 1983).

In another embodiment, the antibody as described herein comprises a full-length human antibody, wherein the antibody variable region of the heterodimeric protein is capable of recruiting human immune effector cells by specifically binding to effector antigens (e.g., CD3 antigens) located on human immune effector cells, and wherein a second antibody variable region of the heterodimeric protein is capable of specifically binding to the target antigen. In some embodiments, the human antibody has an IgG1, IgG2, IgG3, or IgG4 isotype. In some embodiments, the heterodimeric protein comprises an immune-inert Fc chain.

Human immune effector cells can be any variety of immune effector cells known in the art. For example, immune effector cells can be members of the human lymphoid cell lineage, including but not limited to T cells (e.g., cytotoxic T cells), B cells, and natural killer (NK) cells. Immune effector cells can also be, for example, but not limited to, members of the human myeloid lineage, including but not limited to monocytes, neutrophils, and dendritic cells. Such immune effector cells can have cytotoxic or apoptotic effects on target cells, or other desired effects, after activation by binding to effector antigens.

Effector antigens are antigens (e.g., proteins or peptides) expressed on human immune effector cells. Examples of effector antigens that can be bound by heterodimers (e.g., heterodimer antibodies or bispecific antibodies) include, but are not limited to, human CD3, CD16, NKG2D, NKp46, CD2, CD28, CD25, CD64, and CD89.

Target cells can be natural or foreign to humans. In natural target cells, the cell may have been transformed into a malignant cell or pathologically modified (e.g., a natural target cell infected by a virus, malaria parasite, or bacteria). In foreign target cells, the cell is an invading pathogen, such as bacteria, malaria parasite, or virus.

In some embodiments, the antibodies that can be used in this invention are Fab, Fab fragment, F(ab) ₂ fragment, Fv fragment, single-chain Fv fragment, disulfide-stabilized Fv fragment, single-chain antibody, monoclonal antibody, chimeric antibody, bispecific antibody, trispecific antibody, multispecific antibody, bispecific heterodimeric biantibody, bispecific heterodimeric IgG, polyclonal antibody, labeled antibody, humanized antibody, human antibody, or fragments thereof.

In some implementations, the CD3 antibody described herein is a monoclonal antibody. For example, the CD3 antibody is a humanized monoclonal antibody or a chimeric monoclonal antibody.

This invention covers antibodies comprising an Fc chain or a domain thereof, or a portion thereof. In some embodiments, the Fc chain or a portion thereof comprises one or more constant domains (e.g., CH2 or CH3 domains) of an Fc chain of IgG1, IgG2, IgG3, or IgG4. In another embodiment, this invention covers molecules comprising an Fc chain or a portion thereof, wherein the Fc chain or a portion thereof contains at least one amino acid modification (e.g., substitution) relative to an equivalent wild-type Fc chain or a portion thereof. Variant Fc regions are well known in the art and are primarily used to alter the phenotype of antibodies comprising variant Fc regions, as determined in any binding activity or effector function assay well known in the art (e.g., ELISA, SPR assay, or ADCC). Such variant Fc chains or portions thereof can prolong the plasma half-life and stability exhibited by bispecific antibodies comprising Fc chains or portions thereof of this invention. In another embodiment, this invention covers the use of any Fc variant known in the art.

In one embodiment, one or more amino acid modifications are made to the Fc chain to reduce the affinity and coercivity of the Fc chain, thereby reducing the affinity and coercivity of the bispecific antibody of the present invention for one or more FcγR receptors. In one specific embodiment, the present invention covers bispecific antibodies comprising a variant Fc chain or a portion thereof, wherein the variant Fc chain contains at least one amino acid modification relative to the wild-type Fc chain, and the variant Fc region binds only one FcγR, wherein the FcγR is FcγRIIIA. In another specific embodiment, the present invention covers bispecific antibodies comprising a variant Fc chain or a portion thereof, wherein the variant Fc chain contains at least one amino acid modification relative to the wild-type (WT) Fc chain, and the variant Fc region binds only one FcγR, wherein the FcγR is FcγRIIA. In yet another specific embodiment, the present invention covers bispecific antibodies comprising a variant Fc chain or a portion thereof, wherein the variant Fc chain contains at least one amino acid modification relative to the wild-type Fc chain, and the variant Fc chain binds only one FcγR, wherein the FcγR is FcγRIIB. In another embodiment, the invention covers molecules comprising a variant Fc chain, wherein, relative to molecules that do not contain an Fc chain or contain a wild-type Fc chain, the variant confers or mediates reduced ADCC activity (or other effector functions) and/or increased binding to FcγRIIB (CD32B), as measured using methods known to those skilled in the art and described herein.

This invention also covers the use of Fc regions comprising domains or regions from two or more IgG isoforms. As is known in the art, amino acid modifications to the Fc region can significantly affect Fc-mediated effector function and/or binding activity. However, when implemented in the context of a selected IgG isoform, these alterations to functional characteristics can be further improved and/or manipulated. Similarly, the native characteristics of isoform Fc can be manipulated through one or more amino acid modifications. Multiple IgG isoforms (i.e., IgG1, IgG2, IgG3, and IgG4) exhibit different physical and functional properties, including serum half-life, complement binding, FcγR binding affinity, and effector functional activity (e.g., ADCC, CDC), due to differences in the amino acid sequences of their hinges and/or Fc regions.

In one embodiment, amino acid modifications and IgG Fc regions are independently selected based on their respective individual binding and/or effector functional activities to engineer bispecific antibodies with desired characteristics. In one specific embodiment, the binding and/or effector functional activities of amino acid modifications and IgG hinge/Fc regions are separately measured in the context of IgG1, as described herein or known in the art. In one embodiment, amino acid modifications and IgG hinge/Fc regions exhibit similar functionality, such as reduced ADCC activity (or other effector functions) and/or increased binding to FcγRIIB in the context of bispecific antibodies or other Fc-containing molecules (e.g., immunoglobulins). In another embodiment, the invention covers variant Fc regions comprising combinations of amino acid modifications known in the art and selected IgG regions exhibiting novel properties that are undetectable when the modifications and/or regions are measured independently as described herein.

One method for determining the binding affinity of an antibody to CD3 is by measuring the binding affinity of a monofunctional Fab fragment of the antibody. To obtain a monofunctional Fab fragment, the antibody (e.g., IgG) can be cleaved with papain or recombinantly expressed. The affinity of the antibody's CD3 Fab fragment can be determined using a surface plasmon resonance (SPR) system (Biacore™ 3000™ SPR system, Biacore™, INC, Piscataway NJ) equipped with a pre-immobilized streptavidin sensor chip (SA) or anti-mouse Fc or anti-human Fc, using HBS-EP run buffer (0.01 M HEPES, pH 7.4, 0.15 M NaCl, 3 mM EDTA, 0.005% v/v surfactant P20). Biotinylated or Fc-fused human CD3 can be diluted in HBS-EP buffer to concentrations less than 0.5 μg/mL and injected across individual chip channels using variable contact times to achieve two ranges of antigen density: 50–200 reaction units (RU) for detailed kinetic studies, or 800–1,000 RU for screening assays. Regeneration studies have shown that 25 mM NaOH in 25% v/v ethanol effectively removes bound Fab while maintaining CD3 activity on the chip for more than 200 injections. Typically, serial dilutions of purified Fab samples (across concentrations of 0.1–10x estimated _KD ) are injected at 100 µL/min for 1 minute, allowing dissociation times up to 2 hours. Fab protein concentrations are determined by ELISA and/or SDS-PAGE electrophoresis using known concentrations of Fab (e.g., as determined by amino acid analysis) as standards. The data were fitted to a 1:1 Lammill binding model using the BIAevaluation procedure (Karlsson, R. Roos, H. Fagerstam, L. Petersson, B. (1994). Methods Enzymology 6. 99-110), simultaneously obtaining the kinetic binding rate (k _{_on} ) and dissociation rate (k_{_off} ). The equilibrium dissociation constant (K_{_D}) was calculated as k _{_off} /k _{_on} . This protocol is suitable for determining the binding affinity of antibodies to any CD3, including human CD3, CD3 from another mammal (such as mouse CD3, rat CD3, or primate CD3), and different forms of CD3 (e.g., glycosylated CD3). Antibody binding affinity is typically measured at 25°C, but can also be measured at 37°C.

Antibodies as described herein can be prepared by any method known in the art. For the generation of hybridoma cell lines, the host animal immunization pathway and timeline are generally consistent with established and routine techniques for antibody stimulation and production, as further described herein. General techniques for generating human and mouse antibodies are known in the art and/or described herein.

Consider manipulating any mammalian subject, including humans or antibody-producing cells derived therefrom, to serve as a basis for generating mammalian, including human, hybridoma cell lines. Typically, host animals are inoculated with a given amount of immunogen via intraperitoneal, intramuscular, oral, subcutaneous, plantar, and/or intradermal administration, including as described herein.

Hybridomas can be prepared from lymphocytes and immortalized myeloma cells using general somatic cell hybridization techniques, as described in Kohler, B. and Milstein, C., Nature 256:495-497, 1975, or modified techniques, as described in Buck, D. W. et al., In Vitro, 18:377-381, 1982. Available myeloma lines, including but not limited to X63-Ag8.653 and those from the Salk Institute, Cell Distribution Center, San Diego, Calif., USA, can be used in the hybridization. Typically, the technique involves fusing myeloma cells and lymphoid cells using a fusion agent, such as polyethylene glycol, or by electrical means well known to those skilled in the art. After fusion, the cells are separated from the fusion medium and grown in a selective growth medium (such as hypoxanthine-aminopterin-thymidine (HAT) medium) to eliminate unhybridized parental cells. Any culture medium, with or without serum supplementation, as described herein can be used to culture hybridomas that secrete monoclonal antibodies. As an alternative to cell fusion technology, EBV immortalized B cells can be used to generate the monoclonal antibodies of this invention. If necessary, the hybridoma is amplified and subcloned, and the antiimmunogen activity of the supernatant is determined using routine immunoassay procedures (e.g., radioimmunoassay, enzyme immunoassay, or fluorescence immunoassay).

Hybridomas that can be used as antibody sources encompass all derivatives of parental hybridomas that produce monoclonal antibodies specific to CD3 or a portion thereof, as well as progeny cells.

Hybridomas that produce such antibodies can be grown in vitro or in vivo using known procedures. If desired, monoclonal antibodies can be isolated from culture media or body fluids using routine immunoglobulin purification procedures such as ammonium sulfate precipitation, gel electrophoresis, dialysis, chromatography, and ultrafiltration. For example, unwanted activity (if present) can be removed by running the formulation on an adsorbent made of an immunogen attached to a solid phase and eluting or releasing the desired antibody from the immunogen. Using bifunctional reagents or derivatizing agents, such as maleimide benzoyl sulfosuccinimide (conjugated via cysteine residues), N-hydroxysuccinimide (conjugated via lysine residues), glutaraldehyde, succinic anhydride, _SOCl2 , or ^R1N =C=NR (where R and ^R1 are different alkyl groups), conjugated with human CD3 or fragments containing the target amino acid sequence in the host animal to be immunized, a population of antibodies (e.g., monoclonal antibodies) can be obtained.

If desired, the target antibody can be sequenced, and the polynucleotide sequence can then be cloned into a vector for expression or propagation. The sequence encoding the target antibody can be maintained in a vector within a host cell, which can then be expanded and frozen for future use. The generation of recombinant monoclonal antibodies in cell culture can be performed by means known in the art through cloning antibody genes from B cells. See, for example, Tiller et al., J. Immunol. Methods 329,112,2008; U.S. Patent No. 7,314,622.

In an alternative approach, polynucleotide sequences can be genetically manipulated to humanize antibodies or improve their affinity or other characteristics. For example, if an antibody is used in clinical trials and treatments in humans, the constant region can be engineered to more closely resemble the human constant region to avoid an immune response. It might be desirable to genetically manipulate antibody sequences to achieve greater affinity for CD3 and greater therapeutic efficacy.

In some embodiments, the antibody has a modified constant region that removes or reduces Fcγ receptor binding. For example, Fc may be human IgG2 containing the mutant D265, where the amino acid residues are numbered with reference to the wild-type IgG2 sequence. Therefore, in some embodiments, the constant region has a modified constant region having the sequence shown in SEQ ID NO: 80. In some embodiments, the antibody has a modified constant region having the sequence shown in SEQ ID NO: 81.

There are four general steps to humanizing a monoclonal antibody. These steps are: (1) determining the nucleotide and predicted amino acid sequences of the variable regions of the light and heavy chains of the starting antibody; (2) designing the humanized antibody, i.e., deciding which antibody framework region to use during the humanization process; (3) the actual humanization method/technique; and (4) the transfection and expression of the humanized antibody. See, for example, U.S. Patent Nos. 4,816,567; 5,807,715; 5,866,692; 6,331,415; 5,530,101; 5,693,761; 5,693,762; 5,585,089; and 6,180,370.

Many humanized antibody molecules containing antigen-binding sites derived from non-human immunoglobulins have been described, including chimeric antibodies with rodent or modified rodent V regions fused to human constant regions and their associated CDRs. See, for example, Winter et al. Nature 349:293-299, 1991; Lobuglio et al. Proc. Nat. Acad. Sci. USA 86:4220-4224, 1989; Shaw et al. J Immunol. 138:4534-4538, 1987; and Brown et al. Cancer Res. 47:3577-3583, 1987. Other references describe the transplantation of rodent CDRs into human support framework regions (FRs) prior to fusion with appropriate human antibody constant regions. See, for example, Riechmann et al., Nature 332:323-327, 1988; Verhoeyen et al., Science 239:1534-1536, 1988; and Jones et al., Nature 321:522-525, 1986. Another reference describes rodent CDRs supported by recombinantly engineered rodent framework regions. See, for example, European Publication No. EP0519596. These “humanized” molecules are designed to minimize unwanted immune responses to rodent anti-human antibody molecules, which limits the duration and effectiveness of therapeutic applications that are partially present in human receptors. For example, antibody constant regions can be engineered to be immunologically inert (e.g., not triggering complement cleavage). See, for example, PCT application No. PCT/GB99/01441; UK application No. 9809951.8. Other methods using humanized antibodies may also be disclosed by Daugherty et al., Nucl. Acids Res. 19:2471-2476, 1991, and in U.S. Patent Nos. 6,180,377; 6,054,297; 5,997,867; 5,866,692; 6,210,671; and 6,350,861; and PCT Publication No. WO01/27160.

The general principles discussed above regarding humanized antibodies also apply to custom antibodies, for example, for use in other mammals (e.g., humans, non-human primates, mice, donkeys, sheep, rabbits, goats, guinea pigs, camels, horses, or chickens). Furthermore, one or more aspects of the humanized antibodies described herein can be combined, such as CDR transplantation, scaffold mutation, and CDR mutation.

In one variant, fully human antibodies can be obtained using commercially available mice engineered to express specific human immunoglobulin proteins. Transgenic animals designed to produce a more desired (e.g., fully human antibody) or more robust immune response can also be used to generate humanized or human antibodies. Examples of this technology are Xenomouse™ from Abgenix, Inc. (Fremont, CA) and HuMAb-Mouse® and TC Mouse™ from Medarex, Inc. (Princeton, NJ).

In one alternative, the antibody can be recombinantly prepared and expressed using any method known in the art. In another alternative, the antibody can be recombinantly prepared using phage display technology. See, for example, U.S. Patent Nos. 5,565,332; 5,580,717; 5,733,743; and 6,265,150; and Winter et al., Annu. Rev. Immunol. 12:433-455, 1994. Alternatively, phage display technology (McCafferty et al., Nature 348:552-553, 1990) can be used to generate human antibodies and antibody fragments in vitro from a gene library of immunoglobulin variable (V) domains from non-immunized donors. According to this technology, the antibody V domain gene frame is cloned into the major or minor coat protein genes of filamentous phages, such as M13 or fd, and displayed as a functional antibody fragment on the surface of the phage particle. Because filamentous particles contain single-stranded DNA copies of the phage genome, selection based on antibody functional characteristics also leads to the selection of genes encoding antibodies that exhibit those characteristics. Thus, phages mimic some characteristics of B cells. Phage display can be performed in various forms; for reviews, see, for example, Johnson, Kevin S., and Chiswell, David J., Current Opinion in Structural Biology 3:564-571, 1993. Several sources of the V gene segment can be used for phage display. Clackson et al., Nature 352:624-628, 1991, isolated a wide variety of anti-oxazolone antibodies from a small, randomly combined library of the V gene derived from the spleen of immunized mice. Following the techniques described by Mark et al., J. Mol. Biol. 222:581-597, 1991, or Griffith et al., EMBO J.12:725-734, 1993, a V gene library from non-immunized human donors can be constructed, and antibodies against a wide variety of antigens, including self-antigens, can be isolated. In the innate immune response, antibody genes accumulate mutations at a high rate (somatic hypermutation). Some of the introduced changes will confer higher affinity, and B cells exhibiting high-affinity surface immunoglobulins will preferentially replicate and differentiate during subsequent antigen challenges. This natural process can be simulated using a technique called “chain shuffling” (Marks et al., Bio/Technol. 10:779-783, 1992). In this approach, the affinity of “primary” human antibodies obtained by phage display can be improved by sequentially replacing the heavy and light chain V region genes with a naturally occurring library (library) of variants of the V domain genes obtained from non-immunized donors. This technique allows for the production of antibodies and antibody fragments with affinities in the pM-nM range. A strategy for preparing very large phage antibody libraries (also known as “mother-of-all libraries”) has been described by Waterhouse et al., Nucl. Acids Res. 21:2265-2266, 1993. Gene shuffling can also be used to derive human antibodies from rodent antibodies, wherein the human antibodies have similar affinity and specificity to the starting rodent antibody. According to this method, also known as “epitope imprinting,” the heavy or light chain V domain genes of rodent antibodies obtained via phage display technology are replaced with a human V domain gene library, producing a rodent-human chimera. Selection on the antigen results in the isolation of the human variable region capable of restoring a functional antigen-binding site; that is, the epitope governs (marks) the selection of the partner. When this process is repeated to replace the remaining rodent V domains, human antibodies are obtained (see PCT Publication No. WO93/06213). Unlike traditional humanization of rodent antibodies via CDR transplantation, this technique provides fully human antibodies that do not contain rodent-derived frameworks or CDR residues.

Antibodies can be prepared recombinantly by first isolating the antibody and antibody-producing cells from a host animal, obtaining the gene sequence, and using the gene sequence to recombinantly express the antibody in host cells (e.g., CHO cells). Another approach is to express the antibody sequence in plants (e.g., tobacco) or transgenic milk. Methods for recombinantly expressing antibodies in plants or milk have been disclosed. See, for example, Peeters et al. Vaccine 19:2756, 2001; Lonberg, N. and D. Huszar Int. Rev. Immunol 13:65, 1995; and Pollock et al., J Immunol Methods 231:147, 1999. Methods for preparing antibody derivatives, such as humanized, single-chain, etc., are known in the art.

Immunoassays and flow cytometry sorting techniques, such as fluorescence activated cell sorting (FACS), can also be used to isolate antibodies specific to CD3 or target antigens.

Antibodies as described herein can bind to a variety of different solid supports or carriers. Such supports can be active and/or inert. Well-known supports include polypropylene, polystyrene, polyethylene, dextran, nylon, amylase, glass, natural and modified cellulose, polyacrylamide, agarose, and magnets. For the purposes of this invention, the nature of the support can be soluble or insoluble. Those skilled in the art will recognize other suitable supports for binding antibodies, or will be able to determine these using routine experiments. In some embodiments, the support comprises a portion targeting the myocardium.

DNA encoding monoclonal antibodies can be readily isolated and sequenced using standard procedures, such as by using oligonucleotide probes capable of specifically binding to genes encoding the heavy and light chains of monoclonal antibodies. Hybridoma cells serve as a preferred source of this DNA. Once isolated, the DNA can be placed in an expression vector (such as the one disclosed in PCT Publication No. WO87/04462) and then transfected into host cells that otherwise do not produce immunoglobulin proteins (such as E. coli cells, simian COS cells, CHO cells, or myeloma cells) to obtain monoclonal antibody synthesis in recombinant host cells. See, for example, PCT Publication No. WO87/04462. DNA can also be modified, for example, by replacing homologous mouse sequences with coding sequences of human heavy and light chain constant regions (Morrison et al., Proc. Nat. Acad. Sci. 81:6851, 1984), or by covalently binding all or part of the coding sequence of an immunoglobulin-coding sequence to the coding sequence of a non-immunoglobulin polypeptide. In this manner, "chimeric" or "hybrid" antibodies with the binding specificity of the monoclonal antibodies described herein are prepared.

Methods known in the art can be used to identify or characterize CD3 or other antigen antibodies as described herein, thereby detecting and/or measuring antigen expression levels. In some embodiments, CD3 antibodies are identified by incubating a candidate antibody with CD3 and monitoring the associated decrease in binding and/or CD3 expression levels. Binding assays can be performed using purified CD3 peptides, or cells naturally expressing or transfected to express CD3 peptides. In one embodiment, the binding assay is a competitive binding assay in which the ability of a candidate antibody to compete with a known CD3 antibody for CD3 binding is evaluated. This assay can be performed in various forms, including ELISA.

Following initial identification, the activity of candidate CD3 or other antigen-antibody pairs can be further confirmed and refined using bioassays known for testing targeted biological activity. Alternatively, bioassays can be used to directly screen candidates. Several methods for identifying and characterizing antibodies are described in detail in the examples.

CD3 or other antigen antibodies can be characterized using methods well-known in the art. For example, one method is to identify the epitope to which it binds, or "epitope mapping." Many methods are known in the art for mapping and characterizing the location of epitopes on proteins, including resolving the crystal structure of antibody-antigen complexes, competitive assays, gene fragment expression assays, and synthetic peptide-based assays, as described, for example, in Chapter 11 of Harlow and Lane, Using Antibodies, a Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 1999. In another instance, epitope mapping can be used to determine the sequence to which the antibody binds. Epitope mapping is commercially available from various sources, for example, Pepscan Systems (Edelhertweg 15, 8219 PH Lelystad, Netherlands). The epitope can be a linear epitope, i.e., contained in a single amino acid segment, or a conformational epitope formed by the three-dimensional interactions of amino acids, which may not necessarily be contained in a single segment. Peptides of varying lengths (e.g., at least 4-6 amino acids long) can be isolated or synthesized (e.g., recombinantly) and used for binding assays with CD3 or other antigens. In another example, epitopes to which CD3 or other antigens bind can be determined in systematic screening by using overlapping peptides derived from CD3 or other antigen sequences and measuring binding with CD3 or other antigens. Based on gene fragment expression assays, open reading frames encoding CD3 or other antigens are randomly or through specific genetic constructs fragmented, and the reactivity of expressed fragments of CD3 or other antigens with the antibody to be tested is determined. Gene fragments can be generated, for example, by PCR, and then transcribed and translated into proteins in vitro in the presence of radioactive amino acids. The binding of the antibody to the radiolabeled CD3 or other antigen fragment is then determined by immunoprecipitation and gel electrophoresis. Some epitopes can also be identified using large libraries (phage libraries) of random peptide sequences displayed on the surface of phage particles. Alternatively, the binding of a defined library of overlapping peptide fragments to a test antibody can be tested in a simple binding assay. In an additional instance, mutagenesis of antigen-binding domains, domain exchange assays, and alanine partitioning can be performed to identify residues required, sufficient, and/or essential for epitope binding. For example, domain exchange assays can be performed using mutant CD3 or other antigens, where various fragments of the CD3 or other antigenic protein have been replaced (exchanged) with sequences from antigens from another species (e.g., mice), or closely related but antigenically different proteins (e.g., Trop-1). By evaluating the binding of the antibody to mutant CD3 or other antigens, the importance of specific CD3 or other antigenic fragments to antibody binding can be assessed.

Another method that can be used to characterize CD3 or other antigen-antibodies is to use a competitive assay with other antibodies known to bind to the same antigen, i.e., various fragments on CD3 or other antigens, to determine whether the CD3 or other antigen-antibody binds to the same epitopes as the other antibodies. Competitive assays are well known to those skilled in the art.

Expression vectors can be used to direct the expression of CD3 or other antigens and antibodies. Those skilled in the art are familiar with the administration of expression vectors to obtain the expression of exogenous proteins in vivo. See, for example, U.S. Patent Nos. 6,436,908; 6,413,942; and 6,376,471. Administration of expression vectors includes local or systemic administration, including injection, oral administration, particle gun or catheter-directed administration, and local application. In another embodiment, the expression vector is administered directly to the sympathetic trunk or ganglia, or to the coronary arteries, atria, ventricles, or pericardium.

Targeted delivery can also be achieved using therapeutic compositions containing expression vectors or subgenomic polynucleotides. Receptor-mediated DNA delivery techniques are described, for example, in: Findeis et al., Trends Biotechnol., 11:202, 1993; Chiou et al., Gene Therapeutics: Methods And Applications of Direct Gene Transfer, J.A. Wolff, ed., 1994; Wu et al., J. Biol. Chem., 263:621, 1988; Wu et al., J. Biol. Chem., 269:542, 1994; Zenke et al., Proc. Natl. Acad. Sci. USA, 87:3655, 1990; and Wu et al., J. Biol. Chem., 266:338, 1991. Therapeutic compositions containing polynucleotides are administered topically in gene therapy regimens at concentrations ranging from about 100 ng to about 200 mg of DNA. Concentrations ranging from about 500 ng to about 50 mg, about 1 μg to about 2 mg, about 5 μg to about 500 μg, and about 20 µg to about 100 µg of DNA may also be used during gene therapy regimens. Gene delivery vehicles can be used to deliver therapeutic polynucleotides and peptides. Gene delivery vehicles can be of viral or non-viral origin (see generally, Jolly, Cancer Gene Therapy, 1:51, 1994; Kimura, Human Gene Therapy, 5:845, 1994; Connelly, Human Gene Therapy, 1:185, 1995; and Kaplitt, Nature Genetics, 6:148, 1994). Expression of such coding sequences can be induced using endogenous mammalian promoters or heterologous promoters. The expression of a coding sequence can be constitutive or regulated.

Virus-based vectors for delivering desired polynucleotides and expressing them in desired cells are well known in the art. Exemplary virus-based vectors include, but are not limited to, recombinant retroviruses (see, for example, PCT Publications WO90/07936; WO94/03622; WO93/25698; WO93/25234; WO93/11230; WO93/10218; WO91/02805; U.S. Patents 5,219,740 and 4,777,127; GB Patent 2,200,651; and EP Patent EP0 345242), alphavirus-based vectors (e.g., Sindbis virus vectors, Semliki Forest Virus (ATCC)). VR-67; ATCCVR-1247), Ross River virus (ATCC VR-373; ATCCVR-1246) and Venezuelan equine encephalitis virus (ATCC VR-923; ATCC VR-1250; ATCC VR 1249; ATCC VR-532), and adeno-associated virus (AAV) vectors (see, for example, PCT Publications WO94/12649, WO93/03769; WO93/19191; WO94/28938; WO95/11984 and WO95/00655). Alternatively, administration of DNA linked to a killed adenovirus may be used, as described in Curiel, Hum. Gene Ther., 3:147, 1992.

Non-viral delivery media and methods may also be used, including, but not limited to, polycationically condensed DNA linked to or unlinked by a single, killed adenovirus (see, for example, Curiel, Hum. Gene Ther., 3:147, 1992); ligand-linked DNA (see, for example, Wu, J. Biol. Chem., 264:16985, 1989); eukaryotic cell delivery media (see, for example, U.S. Patent No. 5,814,482; PCT Publications WO95/07994; WO96/17072; WO95/30763; and WO97/42338) and nucleic acid charge neutralization or fusion with the cell membrane. Naked DNA may also be used. Exemplary methods for introducing naked DNA are described in PCT Publication WO90/11092 and U.S. Patent No. 5,580,859. Liposomes that can act as gene delivery mediators are described in U.S. Patent No. 5,422,120; PCT Publications Nos. WO95/13796, WO94/23697, WO91/14445; and European Publication EP 0524968. Additional methods are described in Philip, Mol. Cell Biol., 14:2411, 1994 and Woffendin, Proc. Natl. Acad. Sci., 91:1581, 1994.

In some embodiments, the invention covers compositions including pharmaceutical compositions comprising antibodies of the invention as described herein or prepared by the methods described herein and having the characteristics described herein. As used herein, pharmaceutical compositions may comprise one or more antibodies binding to tumor antigens, one or more bispecific antibodies binding to CD3 and tumor antigens, and/or one or more polynucleotides encoding sequences of one or more of these antibodies. These compositions may further comprise suitable excipients, such as pharmaceutically acceptable excipients, including buffers, which are well known in the art.

In one embodiment, the invention is characterized by a composition that is a combination of a CD3 antibody and a second therapeutic agent. In one embodiment, the second therapeutic agent is any agent advantageously combined with a CD3 antibody. Exemplary agents that can be advantageously combined with a CD3 antibody include, but are not limited to, other agents that bind and/or activate CD3 signaling (including other antibodies or their antigen-binding fragments, etc.) and/or agents that do not directly bind CD3 but still activate or stimulate immune cell activation. The invention further includes additional combination therapies and co-formulations involving the CD3 antibodies of the invention. The invention also provides methods for preparing any of these antibodies. The antibodies of the invention can be prepared by procedures known in the art, including de novo protein synthesis and recombinant expression of nucleic acids encoding binding proteins. The desired nucleic acid sequence can be generated by recombinant methods (e.g., PCR mutagenesis of previously prepared variants of the desired polynucleotide) or by solid-phase DNA synthesis. Recombinant expression methods are typically used. In one aspect, the invention provides polynucleotides comprising sequences encoding CD3VH and/or VL. Due to the degeneracy of the genetic code, multiple nucleic acid sequences encode each immunoglobulin amino acid sequence, and the invention includes all nucleic acids encoding the binding proteins described herein.

Chimeric or hybrid antibodies can also be prepared in vitro using known methods of synthetic protein chemistry, including those involving cross-linking agents. For example, immunotoxins can be constructed using disulfide exchange reactions or by forming thioether bonds. Examples of suitable reagents for this purpose include iminothiolates and methyl-4-mercaptobutyrimidate.

In recombinant humanized antibodies, the Fcγ region can be modified to avoid interactions with the Fcγ receptor, as well as with complement and the immune system. Techniques for preparing such antibodies are described in PCT Publication WO99/58572. For example, if the antibody is to be used in clinical trials and treatments in humans, the constant region can be engineered to more closely resemble the human constant region to avoid an immune response. See, for example, U.S. Patent Nos. 5,997,867 and 5,866,692.

Compared to the corresponding germline sequences from which antibodies are derived, CD3 antibodies disclosed herein may contain one or more amino acid substitutions, insertions, and/or deletions in the framework and/or CDR regions of the heavy and light chain variable domains. Such mutations can be readily identified by comparing the amino acid sequences disclosed herein with germline sequences available from, for example, public antibody sequence databases. This invention includes antibodies and antigen-binding fragments thereof derived from any of the amino acid sequences disclosed herein, wherein one or more amino acids in one or more framework and/or CDR regions are mutated to one or more corresponding residues of the germline sequence from which the antibody is derived, or one or more corresponding residues of another human germline sequence, or conserved amino acid substitutions of one or more corresponding germline residues (such sequence changes are collectively referred to herein as “germline mutations”). Starting with the heavy and light chain variable region sequences disclosed herein, those skilled in the art can readily generate a variety of antibody and antigen-binding fragments containing one or more individual germline mutations or combinations thereof. In some embodiments, all framework and/or CDR residues within the VH and/or VL domains are mutated back to residues found in the original germline sequence from which the antibody is derived. In some embodiments, only certain residues are mutated back to the original germline sequence, for example, mutated residues found only in the first 8 amino acids of FR1 or the last 8 amino acids of FR4, or mutated residues found only in CDR1, CDR2, or CDR3. In some further embodiments, one or more framework and/or CDR residues are mutated to one or more corresponding residues of a different germline sequence (i.e., a germline sequence from which the antibody was originally derived).

Furthermore, the antibodies of the present invention may contain any combination of two or more germline mutations within the framework and/or CDR region, for example, wherein certain individual residues are mutated to corresponding residues of a specific germline sequence, while certain other residues different from the original germline sequence are maintained or mutated to corresponding residues of a different germline sequence. Once obtained, one or more desired properties of the antibody and antigen-binding fragment containing one or more germline mutations can be readily tested, such as increased binding specificity, increased binding affinity, improved or enhanced antagonistic or agonistic biological properties (as the case may be), reduced immunogenicity, etc. Antibody and antigen-binding fragments obtained in this general manner are included within the scope of the present invention.

The present invention also includes CD3 antibodies comprising variants of any of the HC VR, LC VR, and/or CDR amino acid sequences disclosed herein having one or more conserved substitutions. For example, the present invention includes CD3 antibodies having HC VR, LC VR, and/or CDR amino acid sequences having, for example, 10 or fewer, 8 or fewer, 6 or fewer, 4 or fewer, etc., conserved amino acid substitutions relative to any of the HC VR, LC VR, and/or CDR amino acid sequences, as disclosed herein.

Therefore, this invention covers modifications of the antibodies and peptides of this invention, variants as described herein, including functionally equivalent antibodies that do not significantly affect their properties, and variants with enhanced or reduced activity and/or affinity. For example, amino acid sequences can be mutated to obtain antibodies with desired binding affinity to tumor antigens and/or CD3. Peptide modification is a common practice in the art and need not be described in detail herein. Examples of modified peptides include peptides with conserved substitutions of amino acid residues, deletions or additions of one or more amino acids that do not significantly and harmfully alter the functional activity or affinity of the mature (enhancing) peptide for its ligands, or the use of chemical analogs.

Amino acid sequence insertions include fusion of the amino and/or carboxyl ends of peptides ranging in length from one residue to one hundred or more residues, as well as intra-sequence insertions of single or multiple amino acid residues. Examples of terminal insertions include antibodies with an N-terminal methionine residue or antibodies fused to an epitope tag. Other insertion variants of antibody molecules include fusion of an enzyme or a peptide that increases the antibody's half-life in the bloodstream with the N- or C-terminus of the antibody.

Substitution variants involve the removal of at least one amino acid residue from the antibody molecule and the insertion of a different residue at its position. Sites of most interest for substitution mutagenesis include hypervariable regions, but FR alterations are also expected. Conservative substitutions are shown under the heading “Conservative Substitutions” in Table 6. If such substitution results in an alteration of biological activity, more significant changes, designated “Exemplary Substitutions” in Table 6 or further described below regarding amino acid categories, can be introduced, and products screened.

Table 6

Significant modifications to the biological properties of antibodies are accomplished through selective substitution, which differs markedly in maintaining: (a) the polypeptide backbone structure in the substituted region, for example, as a sheet or helical conformation; (b) the charge or hydrophobicity of the molecule at the target site; or (c) the volume of the side chains. Naturally occurring amino acid residues are grouped based on common side chain properties:

(1) Nonpolar: Leucine, Met, Ala, Val, Leu, Ile;

(2) Neutral elements with no charge: Cys, Ser, Thr, Asn, Gln;

(3) Acidic (negatively charged): Asp, Glu;

(4) Alkaline (positively charged): Lys, Arg;

(5) Residues affecting chain orientation: Gly, Pro; and

(6) Aroma: Trp, Tyr, Phe, His.

Non-conservative substitutions are prepared by replacing members of one of these categories with members of another category.

Serine residues that are not involved in maintaining the correct conformation of the antibody can generally be replaced with serine to improve the oxidative stability of the molecule and prevent abnormal cross-linking. Conversely, one or more cysteine bonds can be added to the antibody to improve its stability, especially when the antibody is an antibody fragment, such as an Fv fragment.

Amino acid modifications can involve a complete redesign of a region (e.g., a variable region) by altering or modifying one or more amino acids. Changes in the variable region can alter binding affinity and/or specificity. In some embodiments, no more than one to five conserved amino acid substitutions are made within the CDR domain. In some embodiments, no more than one to three conserved amino acid substitutions are made within the CDR domain. In still other embodiments, the CDR domain is VH CDR3 and/or VL CDR3.

Modifications also include glycosylated and non-glycosylated peptides, as well as peptides with other post-translational modifications, such as, for example, glycosylation, acetylation, and phosphorylation with different sugars. Antibodies are glycosylated at conserved sites in their constant regions (Jefferis and Lund, Chem. Immunol. 65:111-128, 1997; Wright and Morrison, TibTECH 15:26-32, 1997). Oligosaccharide side chains of immunoglobulins influence protein function (Boyd et al., Mol. Immunol. 32:1311-1318, 1996; Wittwe and Howard, Biochem. 29:4175-4180, 1990), as well as intramolecular interactions between glycoprotein moieties, which can affect conformation and present the three-dimensional surface of glycoproteins (Jefferis and Lund, above; Wyss and Wagner, Current Opin. Biotech. 7:409-416, 1996). Oligosaccharides can also be used to target given glycoproteins to certain molecules based on specific recognition structures. Glycosylation of antibodies has also been reported to affect antibody-dependent cytotoxicity (ADCC). In particular, CHO cells with tetracycline-regulated β(1,4)-N-acetylglucosamine transferase III (GnTIII) (a glycosyltransferase that catalyzes the formation of aliquots of GlcNAc) expression have been reported to have improved ADCC activity (Umana et al., Mature Biotech. 17:176-180, 1999).

Antibody glycosylation is generally N-linked or O-linked. N-linked glycosylation refers to the attachment of the carbohydrate moiety to the side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine, asparagine-X-threonine, and asparagine-X-cysteine (where X is any amino acid except proline) are recognition sequences for the enzymatic attachment of the carbohydrate moiety to the asparagine side chain. Therefore, the presence of any of these tripeptide sequences in the polypeptide creates a potential glycosylation site. O-linked glycosylation refers to the attachment of one of the sugars N-acetylgalactosamine, galactose, or xylose to a hydroxy amino acid, most commonly serine or threonine, although 5-hydroxyproline or 5-hydroxylysine can also be used.

By altering the amino acid sequence to include one or more of the aforementioned tripeptide sequences (for N-glycosylation sites), the addition of glycosylation sites to the antibody can be conveniently accomplished. Alternatively, alterations can be made by adding or replacing one or more serine or threonine residues to the original antibody sequence (for O-glycosylation sites).

The glycosylation pattern of an antibody can also be altered without changing the underlying nucleotide sequence. Glycosylation is largely dependent on the host cell used to express the antibody. Since the cell types used to express recombinant glycoproteins such as antibodies as potential therapeutic agents are rarely native cells, variations in the antibody glycosylation pattern can be expected (see, for example, Hse et al., J. Biol. Chem. 272:9062-9070, 1997).

Besides host cell selection, factors influencing glycosylation during recombinant antibody production include growth mode, culture medium formulation, culture density, oxygenation, pH, purification protocol, and more. Various methods have been proposed to alter the glycosylation patterns achieved in specific host organisms, including the introduction or overexpression of certain enzymes associated with oligosaccharide production (US Patent Nos. 5,047,335; 5,510,261 and 5,278,299). For example, using endoglycosidase H (Endo H), N-glycosidase F, endoglycosidase F1, endoglycosidase F2, and endoglycosidase F3, glycosylation or certain types of glycosylation can be enzymatically removed from glycoproteins. Additionally, recombinant host cells can be genetically engineered to be defective in processing certain types of polysaccharides. These and similar techniques are well known in the art.

Other modification methods include the use of coupling techniques known in the art, including, but not limited to, enzymatic methods, oxidative substitution, and chelation. Modification can be used, for example, for label attachment for immunoassays. Modified peptides are prepared using procedures established in the art and can be screened using standard assays known in the art, some of which are described below and in the examples.

In some embodiments of the invention, the antibody comprises a modified constant region, such as a constant region that has increased affinity for the human Fcγ receptor and is immunologically inert or partially inert, for example, not triggering complement-mediated cleavage, not stimulating antibody-dependent cell-mediated cytotoxicity (ADCC), or not activating macrophages; or having reduced activity (compared to an unmodified antibody) in any one or more of the following: triggering complement-mediated cleavage, stimulating antibody-dependent cell-mediated cytotoxicity (ADCC), or activating microglia. Different modifications to the constant region can be used to achieve optimal levels and/or combinations of effector functions. See, for example, Morgan et al., Immunology 86:319-324, 1995; Lund et al., J. Immunology 157:4963-9 157:4963-4969, 1996; Idusogie et al., J. Immunology 164:4178-4184, 2000; Tao et al., J. Immunology 143: 2595-2601, 1989; and Jefferis et al., Immunological Reviews 163:59-76, 1998. In some embodiments, the constant region is modified as described in Eur. J. Immunol., 29:2613-2624, 1999; PCT application number PCT/GB99/01441; and/or UK application number 9809951.8. In still other embodiments, the constant region is aglycosylated against N-glycosylation. In some embodiments, the constant region is aglycosylated against N-glycosylation by mutating glycosylated amino acid residues or flanking residues in the constant region that are part of the N-glycosylation recognition sequence. For example, the N-glycosylation site N297 can be mutated to A, Q, K, or H. See Tao et al., J. Immunology 143: 2595-2601, 1989; and Jefferis et al., Immunological Reviews 163:59-76, 1998. In some implementations, the constant region is deglycosylated against N-glycosylation. The constant region may be deglycosylated enzymatically (e.g., by removing carbohydrates via the enzyme peptide-N-glycosidase (PNGase)) or by expression against N-glycosylation in a glycosylation-deficient host cell.

Other antibody modifications include antibodies modified as described in PCT Publication No. WO99/58572. In addition to the binding domain targeting the target molecule, these antibodies also contain an effector domain having an amino acid sequence substantially homologous to all or part of the constant region of the human immunoglobulin heavy chain. These antibodies are capable of binding the target molecule without triggering significant complement-dependent cleavage or cell-mediated target disruption. In some embodiments, the effector domain is capable of specifically binding FcRn and/or FcγRIIb. These are generally based on chimeric domains derived from two or more CH2 domains of the human immunoglobulin heavy chain. Antibodies modified in this manner are particularly suitable for use in long-term antibody therapy to avoid inflammation and other adverse reactions to conventional antibody therapy.

This invention includes embodiments of affinity maturation. For example, affinity-matured antibodies can be generated by procedures known in the art (Marks et al., Bio/Technology, 10:779-783, 1992; Barbas et al., ProcNat. Acad. Sci, USA 91:3809-3813, 1994; Schier et al., Gene, 169:147-155, 1995; Yelton et al., J. Immunol., 155:1994-2004, 1995; Jackson et al., J. Immunol., 154(7):3310-9, 1995; Hawkins et al., J. Mol. Biol., 226:889-896, 1992; and PCT Publication No. WO2004/058184).

The following methods can be used to modulate antibody affinity and characterize CDRs. One way to characterize the CDR of an antibody and/or alter (e.g., improve) the binding affinity of a peptide, such as an antibody, is called “library scan mutagenesis.” Typically, library scan mutagenesis is performed as described below. Using methods recognized in the art, one or more amino acid positions in the CDR are replaced with two or more amino acids (e.g., 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20). This produces small clonal libraries (in some embodiments, one clone for each amino acid position analyzed), each clone having the complexity of two or more members (if two or more amino acids are substituted at each position). Typically, the library also includes clones containing native (unsubstituted) amino acids. For a small number of clones from each library, e.g., approximately 20–80 clones (depending on library complexity), screening is performed for binding affinity to the target peptide (or other binding target), and candidates exhibiting increased binding, identical binding, decreased binding, or no binding are identified. Methods for determining binding affinity are well known in the art. Binding affinity can be determined using BIACORE™ surface plasmon resonance analysis, which detects differences in binding affinity of approximately 2-fold or greater. BIACORE™ is particularly useful when the starting antibody has already bound with relatively high affinity (e.g., approximately 10 nM or lower K _{_D} ). Screening using BIACORE™ surface plasmon resonance is described in the examples herein.

Binding affinity can be determined using Kinexa Biocensor, scintillation affinity assay, ELISA, ORIGEN immunoassay (IGEN), fluorescence quenching, fluorescence transfer, and/or yeast display. Binding affinity can also be screened using appropriate bioassays.

In some embodiments, using art-recognized mutagenesis methods (some of which are described herein), each amino acid position in the CDR is replaced (in some embodiments, one at a time) with all 20 natural amino acids. This produces small clonal libraries (in some embodiments, one clone for each amino acid position analyzed), each clone having a complexity of 20 members (if all 20 amino acids are replaced at each position).

In some implementations, the library to be screened contains substitutions at two or more positions, which may be in the same CDR or in two or more CDRs. Thus, the library may contain substitutions at two or more positions in a single CDR. The library may contain substitutions at two or more positions in two or more CDRs. The library may contain substitutions at three, four, five, or more positions, which are found in two, three, four, five, or six CDRs. Substitutions can be prepared using low-redundancy codons. See, for example, Table 2, Balint et al., Gene 137(1):109-18, 1993.

Candidates exhibiting improved binding can be sequenced to identify CDR substitution mutants that result in improved affinity (also known as "improved" substitutions). Binding candidates can also be sequenced to identify CDR substitutions that maintain binding.

Multiple rounds of screening can be performed. For example, candidates with improved binding (each candidate containing amino acid substitutions at one or more positions of one or more CDRs) can also be used to design a second library containing at least the original and substituted amino acids at each improved CDR position (i.e., the amino acid position in the CDR where the substitution mutant shows improved binding). The preparation and screening or selection of this library are discussed further below.

Library scanning mutagenesis also provides information about the importance of each amino acid position to the stability of the antibody-antigen complex because the frequency of clones exhibiting improved binding, identical binding, reduced binding, or no binding also provides information related to the stability of the antibody-antigen complex. For example, if a CDR position remains bound when changed to all 20 amino acids, that position is identified as unlikely to be required for antigen binding. Conversely, if a CDR position remains bound only with a small percentage of substitutions, that position is identified as important for CDR function. Therefore, library scanning mutagenesis yields information about positions in CDRs that can be changed to many different amino acids (including all 20 amino acids) and positions in CDRs that cannot be changed or can only be changed to a few amino acids.

Candidates with improved affinity can be combined in a second library comprising the improved amino acid, the original amino acid at that position, and may further include additional substitutions at that position, depending on the required complexity of the library or the complexity permitted by the desired screening or selection method. Additionally, if needed, adjacent amino acid positions can be randomized to at least two or more amino acids. Randomization of adjacent amino acids allows for additional conformational flexibility in the mutant CDR, which in turn allows or facilitates the introduction of a large number of improving mutations. The library may also contain substitutions at positions that did not exhibit improved affinity in the first round of screening.

Use any method known in the art to screen or select library members in the second library that have improved and/or altered binding affinity, including screening using BIACORE™ surface plasmon resonance analysis, and selection using any method known in the art for selection, including phage display, yeast display, and ribosome display.

The techniques used for isolating and preparing antibodies are as follows. However, it will be understood that the antibodies can be formed from or incorporated into other polypeptides using techniques known in the art. For example, nucleic acids encoding a target polypeptide (e.g., a ligand, receptor, or enzyme) can be isolated from a cDNA library prepared from a tissue believed to have polypeptide mRNA and express it at detectable levels. The library is screened using probes (e.g., antibodies or oligonucleotides of about 20-80 bases) designed to identify the target gene or the protein it encodes. Screening of cDNA or genomic libraries with selected probes can be performed using standard procedures such as those described in Chapters 10-12 of Sambrook et al., Molecular Cloning: A Laboratory Manual (New York: Cold Spring Harbor Laboratory Press, 1989).

CD3 antibodies used to treat autoimmune diseases

In one aspect, a method for treating an autoimmune disease in a subject is provided, comprising administering to the subject in need an effective amount of a composition comprising an antibody as described herein.

As used in this article, autoimmune diseases include, but are not limited to, inflammatory bowel disease, systemic lupus erythematosus, rheumatoid arthritis, diabetes mellitus (type I), multiple sclerosis, Addison’s disease, celiac disease, dermatomyositis, Graves’ disease, Hashimoto’s thyroiditis, Hashimoto’s encephalopathy, myasthenia gravis, reactive arthritis, Sjogren’s syndrome, atopic allergy, atopic dermatitis, autoimmune enteropathy, autoimmune hepatitis, autoimmune lymphoproliferative syndrome, autoimmune peripheral neuropathy, autoimmune pancreatitis, autoimmune polyendocrine syndrome, and autoimmune progesterone dermatitis. Autoimmune urticaria, autoimmune uveitis, Behcet’s disease, Castleman’s disease, cold agglutinin disease, Crohn’s disease, dermatomyositis, eosinophilic fasciitis, gastrointestinal pemphigoid, Goodpasture’s syndrome, Guillain-Barré syndrome, hidradenitis suppurativa, somnambulism, pemphigus vulgaris, polymyositis, relapsing polychondritis, rheumatic fever, transverse myelitis, ulcerative colitis, undifferentiated connective tissue disease, vasculitis, and Wegener’s granulomatosis.

Multispecific antibodies and their uses

The CD3 antibodies of the present invention may be specific to different epitopes of a single target polypeptide or may contain antigen-binding domains specific to more than one target polypeptide. See, for example, Tutt et al., 1991, J. Immunol. 147:60-69; Kufer et al., 2004, Trends Biotechnol. 22:238-244. The CD3 antibodies of the present invention may be ligated to or co-expressed with another functional molecule (e.g., another peptide or protein). For example, the antibody or a fragment thereof may be functionally linked (e.g., by chemical coupling, genetic fusion, non-covalent association, or other means) to one or more other molecular entities (e.g., another antibody or antibody fragment) to produce a multispecific antibody with a second binding specificity.

In one aspect, the antibody of the present invention is a multispecific antibody. In a specific embodiment, the CD3 antibody is a bispecific antibody that specifically binds to human CD3. In some such embodiments, the bispecific antibody further specifically binds to tumor antigens. In another such embodiment, the bispecific antibody of the present invention simultaneously targets T cells (CD3) and tumor cells, and successfully guides and activates T cells to exert cytotoxicity on tumor cells expressing tumor antigens.

In the respective further embodiments described above, the CD3-binding bispecific antibody is characterized by any one or more of the following properties: (a) treating, preventing, or improving one or more symptoms of a condition in a subject associated with malignant cells expressing a specific tumor antigen (e.g., B-cell-related cancers, such as multiple myeloma); (b) inhibiting tumor growth or progression in a subject (who has a malignant tumor expressing a specific tumor antigen); (c) inhibiting metastasis of cancer (malignant) cells expressing a specific tumor antigen in a subject (who has one or more malignant cells expressing a tumor antigen); (d) inducing regression (e.g., long-term regression) of tumors expressing tumor antigens; (e) exerting cytotoxic activity in malignant cells expressing tumor antigens; (f) prolonging progression-free survival in subjects with tumor-related conditions; (g) prolonging overall survival in subjects with tumor-related conditions; (h) reducing the use of additional chemotherapy or cytotoxic agents in subjects with tumor-related conditions; (i) reducing tumor burden in subjects with tumor-related conditions; or (j) blocking the interaction of tumor antigens with other unidentified factors.

As used herein, a bispecific antibody of the present invention refers to a complex of two or more polypeptide chains, each polypeptide chain comprising at least one antibody VL region and one antibody VH region or a fragment thereof, wherein the VL and VH regions in each polypeptide chain are derived from different antibodies. In a particular aspect, a bispecific antibody comprises a dimer or tetramer of polypeptide chains containing both VL and VH regions. The polypeptide chains constituting the multimeric protein may be covalently linked to at least one other peptide of the multimer via interchain disulfide bonds.

In some such embodiments, the bispecific antibody of the present invention comprises a first heterodimer-promoting domain on a first polypeptide chain and a second heterodimer-promoting domain on a second polypeptide chain (Figure 1). In summary, the first and second heterodimer-promoting domains drive heterodimerization and/or stabilize the bispecific antibody (e.g., through the interaction of pestle and mortar on complementary heterodimer-promoting domains) and/or serve to stabilize the bispecific antibody. In some embodiments, the first and second heterodimer-promoting domains each comprise a CH2 domain and a CH3 domain, wherein the amino acid sequences of each of the CH2 domains and/or the CH3 domains are modified to drive heterodimerization and/or stabilize the bispecific antibody.

In some embodiments, the first heterodimer promoting domain may comprise an Fc chain having CH2 and/or CH3 domains modified to comprise a club (protrusion) or a mortise (cavity). In some such embodiments, the amino acid sequences of the CH2 and/or CH3 domains comprise at least one amino acid modification, wherein: (a) the CH3 domain of the first heterodimer promoting domain forms a club; and (b) the CH3 domain of the second heterodimer promoting domain forms a hole. In another such embodiment, the CH3 domain of the first heterodimer promoting domain comprises the mutants Y349C and/or T366W; and the CH3 domain of the second heterodimer promoting domain comprises the mutants S354C, T366S, L368A, and/or Y407V (according to EU index number).

In one embodiment, if the second heterodimer promoting domain comprises a CH2 and/or CH3 domain modified to comprise a mortise (cavity), then the first heterodimer promoting domain may comprise a CH2 and/or CH3 domain modified to comprise a pestle (protrusion) containing the sequence of SEQ ID NO: 78. In another embodiment, if the second heterodimer promoting domain comprises a CH2 and/or CH3 domain modified to comprise a pestle (protrusion), then the first heterodimer promoting domain may comprise a mortise (cavity) containing the sequence of SEQ ID NO: 79.

Each polypeptide chain of a bispecific antibody contains a VL region and a VH region, which are covalently linked by a glycine-serine linker (linker 1 or linker 2) containing glycine and serine residues, thus confining the antibody-binding domain to prevent self-assembly. Furthermore, each polypeptide chain contains a heterodimerization domain, which promotes heterodimerization and/or stabilization of multiple polypeptide chains and reduces the likelihood of homodimerization between different polypeptide chains. The heterodimerization domain can be located at the N-terminus or C-terminus of the polypeptide chain. The heterodimerization domain can contain a cysteine linker (linker 3) of 1, 2, 3, 4, 5, 6, or more amino acid residues in length. The interaction of two polypeptide chains can produce two VL/VH pairs, thereby forming two epitope-binding domains, i.e., divalent molecules. Neither the VH nor the VL region is confined to any position within the polypeptide chain, i.e., restricted to the amino-terminus or carboxyl-terminus, nor is the region confined by their relative positions to each other; i.e., the VL region can be at the N-terminus of the VH region, and vice versa. The only limitation is the availability of complementary polypeptide chains to form functional bispecific antibodies. In cases where the VL and VH regions originate from antibodies specific to different antigens, the formation of functional bispecific antibodies requires the interaction of two distinct polypeptide chains, i.e., the formation of heterodimers. Conversely, in cases where two distinct polypeptide chains interact freely, for example in a recombinant expression system, one containing VLA and VHB (A being the first epitope and B the second epitope) and the other containing VLB and VHA, two distinct binding sites can be formed: VLA-VHA and VLB-VHB. For all bispecific antibody polypeptide chain pairs, misalignment or misbinding of the two chains is a possibility, for example, interactions between the VL-VL or VH-VH regions. However, the purification of functional bispecific antibodies can be readily controlled based on the immunospecificity of the correctly dimerized binding sites using any affinity-based method known in the art or exemplified herein, such as affinity chromatography.

In one embodiment, the polypeptide chain of the bispecific antibody may contain various linkers and peptides. The linkers and peptides may contain 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 or more amino acids.

In some embodiments, domain 1 is covalently bound to a first heterodimer via a cysteine linker to promote the domain, and domain 2 is covalently bound to a second heterodimer via a cysteine linker to promote the domain. Each cysteine linker includes at least one cysteine residue to allow intramolecular disulfide bonding. In a further embodiment of each of the foregoing embodiments, the cysteine linker (linker 3) comprises at least five amino acids.

In some embodiments, the first polypeptide chain is covalently bound to the second polypeptide chain via at least one disulfide bond. In some such embodiments, at least one disulfide bond is formed between the linker 3 of the first polypeptide chain and the linker 3 of the second polypeptide chain. In another such embodiment, at least one disulfide bond is formed between the first heterodimer promoting domain and the second heterodimer promoting domain. In a specific embodiment, each disulfide bond is formed by linking two cysteine residues. In one aspect of the invention, the bispecific antibody shown in FIG1 comprises a first polypeptide chain and a second polypeptide chain. In some such embodiments, the linker 3 may comprise a truncated lower hinge region of human IgG1 having the sequence SEQ ID NO: 64, which has at least one glycine residue preceding the lower hinge region.

The bispecific antibody of the present invention can simultaneously bind to two separate and distinct epitopes. In some embodiments, at least one epitope binding site is specific for a CD3 determinant expressed on immune effector cells, such as T lymphocytes. In one embodiment, the bispecific antibody molecule binds to an effector cell determinant and also activates the effector cell.

In one aspect, the present invention provides a first polypeptide chain and a second polypeptide chain. In one embodiment, the first polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 36. In another embodiment, the second polypeptide chain comprises the amino acid sequence shown in one or more of SEQ ID NO: 37 or 38 (Table 7). In a preferred embodiment, the first polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 36; and the second polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 37 or 38. In one embodiment, the second polypeptide chain is any polypeptide to be associated with the first polypeptide chain via an interface. Table 7 shows the sequences of the first and second polypeptide chains of the GUCY2c-H2B4 and GUCY2c-2B5 bispecific antibodies.

Table 7

In specific embodiments, the bispecific antibody of the present invention (a) binds to the extracellular domain of a human tumor antigen; (b) exhibits an extended serum and tumor half-life of 30 minutes to 100 days; and/or (c) exhibits a lower _EC50 value of 0.0001 nM to 100 nM in the presence of increased tumor expression levels or increased receptor density levels.

In one implementation, the epitope-binding domain is capable of binding to tumor-associated antigens associated with breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer (including but not limited to non-small cell lung cancer and small cell lung cancer), bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, and digestive system cancers. Digestive system cancers include, but are not limited to, cancers of the esophagus, stomach, small intestine, colon, rectum, anus, liver, gallbladder, appendix, bile ducts, and pancreas. In a specific implementation, the therapy activates a cytolytic T-cell response.

Null effector mutations in human IgG1 CH2-CH3

Using standard primer-guided PCR mutagenesis, the Fc chain of human IgG1 was modified to introduce mutants L234A, L235A, and G237A (SEQ ID NO: 82, according to EU index number) to oblate effector functions attributed to binding to FcγRIII, thereby providing an effector-ineffective phenotype (Canfield et al., J. Exp. Med (1991) 173: 1483-1491; Shields et al., J. Biol. Chem. (2001) 276: 6591-604).

The mortise in human IgG1 CH2-CH3 is a mutation.

The 'mortar' configuration is a known efficient design strategy in the art for engineering antibody heavy chain homodimers for heterodimerization. In this method, 'mortar' variants are obtained by replacing smaller amino acids, such as Y349C and T366W (according to EU index numbers), with larger amino acids in one strand of the Fc chain of IgG1. The 'mortar' is designed to insert into the 'mortar' of the CH3 domain of the complementary chain of the Fc chain generated by replacing larger residues, such as S354C, T366S, L368A, and Y407V (according to EU index numbers), with smaller residues.

In some implementations, complementary mutations are introduced to achieve heterodimerization of the resulting Fc chains, such that each Fc chain carries a set of mutations, Y349C and T366W (SEQ ID NO: 78) for the club (or protrusion) Fc chain, or S354C, T366S, L368A, and Y407V (SEQ ID NO: 79) for the mortis (or cavity) Fc chain, as provided in Table 8. When co-transfected into a suitable mammalian host, the DNA encoding the amino acid sequences (e.g., SEQ ID NO: 78 and 79) generates Fc domains predominantly with a bispecific antibody associated with one club (or protrusion) Fc chain of one mortis (or cavity) Fc chain.

Table 8

The CD3-tumor antigen bispecific antibody of this invention is a thermostability-stable, efficient bispecific antibody-Fc fusion targeting both human CD3 and tumor antigens. The Fc domain in the mortise allows for improved expression and purification in CHO cells, resulting in the desired high purity of the heterodimer. The engineered mutation within the Fc domain cancels FcγR binding, thus potentially avoiding ADCC-mediated T cell exhaustion. Furthermore, as shown by differential scanning calorimetry (DSC), incorporating the Fc domain into the bispecific antibody enhances the molecule's stability.

If a bispecific antibody can be engineered to include at least one cysteine residue, it can interact with a corresponding cysteine residue on another polypeptide chain of the invention to form an interchain disulfide bond. The interchain disulfide bond can be used to stabilize the bispecific antibody, thereby improving expression and recovery in recombinant systems, resulting in stable and consistent formulations, and improved stability of isolated and/or purified products in vivo. One or more cysteine residues can be introduced into any part of the polypeptide chain as a single amino acid or as part of a larger amino acid sequence, such as a hinge region. In a specific aspect, at least one cysteine residue is engineered to appear at the C-terminus of the polypeptide chain.

This invention covers methods and compositions for treating, preventing, or managing cancer in a subject, including administering to the subject a therapeutically effective amount of an antibody engineered according to the invention, the molecule further binding to a cancer antigen. The antibodies of this invention are particularly useful for preventing, inhibiting, reducing the growth and/or regression of primary tumors, and for the metastasis of cancer cells. Although not intended to be limited to a specific mechanism of action, the antibodies of this invention can mediate effector functions that can lead to tumor clearance, tumor reduction, or a combination thereof.

In one aspect, the present invention provides a therapeutic method for stimulating T cell activation using the CD3 antibody of the present invention, wherein the therapeutic method comprises administering a therapeutically effective amount of a pharmaceutical composition comprising the antibody of the present invention to a subject in need. The treated condition is any disease or condition that is improved, alleviated, suppressed, or prevented by stimulating CD3 activity or signaling. In a specific embodiment, the present invention provides a bispecific antigen-binding molecule, such as a bispecific antibody that binds to both CD3 and a target antigen.

In one aspect, the present invention provides a method for inhibiting tumor growth, comprising contacting the tumor with an effective amount of an antibody that binds to CD3, said antibody comprising each of the antibodies described herein.

In another aspect, the present invention provides a method for inhibiting tumor growth in a subject, comprising administering to the subject a therapeutically effective amount of an antibody binding to CD3, including each of those antibodies described herein.

In another aspect, the present invention provides a method for modulating angiogenesis in a subject, comprising administering to the subject a therapeutically effective amount of an antibody binding to CD3, including each of those antibodies described herein.

In another aspect, the present invention provides a method for reducing the tumorigenicity of tumors in a subject, comprising administering to the subject a therapeutically effective amount of an antibody binding to CD3, including each of those antibodies described herein.

In one aspect, the present invention provides a method for treating conditions in subjects associated with tumor antigen expression. Therefore, the present invention provides a method for treating tumor antigen-related conditions by engineering bispecific antibodies to immune-specifically recognize tumor antigens and CD3 antigens on T cells. Bispecific antibodies engineered according to the present invention can be used for the prevention or treatment of cancer because they possess cytotoxic activity via CD3 antibody-induced activation of cytotoxic T cells.

In one specific embodiment, the present invention provides a method for treating cancer. In this embodiment, the cancer is breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer (including but not limited to non-small cell lung cancer and small cell lung cancer), bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, or cancer of the digestive system. In some embodiments, the cancer of the digestive system is selected from cancers of the esophagus, stomach, small intestine, colon, rectum, anus, liver, gallbladder, appendix, bile ducts, and pancreas.

In one aspect, the present invention provides antibodies, bispecific antibodies, or pharmaceutical compositions as disclosed herein for use in a therapy. In one specific embodiment, the present invention also provides a CD3 bispecific antibody for use in a method of treating cancer as defined herein. In another specific embodiment, the therapy activates a cytolytic T-cell response.

This invention further provides antibodies or bispecific antibodies as disclosed herein for the preparation of medicaments for use in therapies. In some embodiments, the therapy is a treatment for cancer. In some embodiments, the cancer is selected from breast cancer, ovarian cancer, thyroid cancer, prostate cancer, cervical cancer, lung cancer (including but not limited to non-small cell lung cancer and small cell lung cancer), bladder cancer, endometrial cancer, head and neck cancer, testicular cancer, glioblastoma, or cancers of the digestive system. In some embodiments, the cancers of the digestive system are selected from cancers of the esophagus, stomach, small intestine, colon, rectum, anus, liver, gallbladder, appendix, bile ducts, and pancreas.

In one aspect, the present invention provides polynucleotides encoding antibodies or bispecific antibodies as disclosed herein. In another embodiment, the present invention provides a vector comprising a polynucleotide as disclosed herein. In yet another embodiment, the present invention provides a host cell comprising a vector as disclosed herein. In some such embodiments, the host cell recombinants produce antibodies or bispecific antibodies as disclosed herein. In specific embodiments, the host cell is selected from bacterial cell lines, mammalian cell lines, insect cell lines, yeast cell lines, and in vitro cell-free protein synthesis systems. In a specific embodiment, the mammalian cell line is a CHO cell line.

In one aspect, a method for treating cancer in a subject with this need is provided, comprising: a) providing a bispecific antibody as described herein, and b) administering the bispecific antibody to the subject.

In some embodiments, a method is provided for inhibiting tumor growth or progression in a subject with malignant cells expressing tumor antigens, comprising administering to a subject in need an effective amount of a composition comprising antibodies as described herein. In some embodiments, a method is provided for inhibiting cell metastasis expressing tumor antigens in a subject, comprising administering to a subject in need an effective amount of a composition comprising antibodies as described herein. In some embodiments, a method is provided for inducing tumor regression in malignant cells in a subject, comprising administering to a subject in need an effective amount of a composition comprising antibodies as described herein.

In a specific aspect, relative to the growth of cancer cells in the absence of the antibodies or bispecific antibodies of the present invention, the antibodies of the present invention inhibit or reduce the growth of cancer cells by at least 99%, at least 95%, at least 90%, at least 85%, at least 80%, at least 75%, at least 70%, at least 60%, at least 50%, at least 45%, at least 40%, at least 45%, at least 35%, at least 30%, at least 25%, at least 20%, or at least 10%.

In specific respects, the antibody kills cells or inhibits or reduces the growth of cancer cells by at least 5%, at least 10%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, or at least 100% better than the antibody or bispecific antibody of the present invention.

In one aspect, the present invention provides an effective amount of a composition (e.g., a pharmaceutical composition) comprising a bispecific antibody as described herein for treating a condition (e.g., cancer) associated with tumor antigen expression in a subject in need of this treatment.

In another aspect, the present invention provides antibodies as described herein for treating conditions (e.g., cancer) associated with tumor antigen expression in subjects with this need. In some embodiments, antibodies as described herein are provided for inhibiting tumor growth or progression in subjects with malignant cells expressing tumor antigens. In some embodiments, antibodies as described herein are provided for inhibiting the metastasis of malignant cells expressing tumor antigens in subjects with this need. In some embodiments, antibodies as described herein are provided for inducing tumor regression in subjects with malignant cells expressing tumor antigens.

In another aspect, the present invention provides the use of antibodies as described herein in the preparation of medicaments for treating conditions (e.g., cancer) associated with tumor antigen expression. In some embodiments, the use of antibodies as described herein is provided in the preparation of medicaments for inhibiting tumor growth or progression. In some embodiments, the use of antibodies as described herein is provided in the preparation of medicaments for inhibiting the metastasis of malignant cells expressing tumor antigens. In some embodiments, the use of antibodies as described herein is provided in the preparation of medicaments for inducing tumor regression.

In some embodiments, the method described herein further includes the step of treating the subject with another form of therapy. In some embodiments, the other form of therapy is an additional anticancer therapy, including, but not limited to, chemotherapy, radiation, surgery, hormone therapy, and/or additional immunotherapy.

This invention further encompasses the administration of the molecules of the invention in combination with other therapies known to those skilled in the art for the treatment or prevention of cancer, including but not limited to current standard and experimental chemotherapy, biotherapy, immunotherapy, radiotherapy, or surgery. In some aspects, the molecules of the invention may be administered in combination with therapeutically or preventively effective amounts of one or more agents, therapeutic antibodies, or other agents known to those skilled in the art for the treatment and/or prevention of cancer.

Therefore, methods for treating cancer include administering an effective amount of the multispecific antibody (e.g., a bispecific antibody) of the present invention in combination with a chemotherapy agent to a subject in need. This combination therapy can be administered alone, sequentially, or simultaneously.

The dosages and frequencies of administration provided herein are covered by the terms therapeutically effective and prophylactically effective. Depending on the specific therapeutic or prophylactic agent administered, the severity and type of cancer, the route of administration, and the subject's age, weight, response, and past medical history, dosage and frequency may further vary based on factors specific to each subject. Those skilled in the art can select an appropriate regimen by taking such factors into account and by following the dosages recommended, for example, those reported in the literature and in Physician's Desk Reference (56th edition, 2002).

Therefore, in addition to redirecting T cells to tumor-specific antigens, bispecific T-cell conjugate molecules can also be used to deliver other diagnostic or therapeutic compounds to cells expressing tumors on their surface. Thus, bispecific T-cell conjugate molecules can be directly or indirectly (e.g., via a connector) linked to drugs, enabling direct delivery to tumor-bearing cells. Therapeutic agents include compounds such as nucleic acids, proteins, peptides, amino acids or derivatives, glycoproteins, radioisotopes, lipids, carbohydrates, or recombinant viruses. Therapeutic and diagnostic nucleic acid portions include antisense nucleic acids, derivatized oligonucleotides for covalent cross-linking with single- or double-stranded DNA, and oligonucleotides forming triple strands.

Pharmaceutical Composition

The present invention further provides compositions comprising a therapeutically effective amount of the antibody disclosed herein and a pharmaceutically acceptable carrier.

The antibodies of the present invention may be in the form of pharmaceutical compositions for administration, formulated to suit a chosen mode of administration, and containing pharmaceutically acceptable diluents or excipients, such as buffers, surfactants, preservatives, solubilizers, isotonic agents, stabilizers, carriers, etc. Remington's Pharmaceutical Sciences, Mack Publishing Co., Easton Pa., 18th edition, 1995 provides an outline of formulation techniques generally known to those skilled in the art.

These pharmaceutical compositions can be administered in any manner known in the art to achieve the generally intended purpose of treating cancer. Routes of administration can be parenteral, and are defined herein as modes of administration including, but not limited to, intravenous, intramuscular, intraperitoneal, subcutaneous, and intra-articular injections and infusions. The manner of administration and dosage of the molecules according to the invention (e.g., antibodies, pharmaceutical compositions) depends on the type of disease to be combated, its stage in appropriate circumstances, the antigen to be controlled, the type of coexisting treatment (if any), the frequency of treatment, the characteristics of the desired effect, and also the subject's weight, age, health status, diet, and sex. Therefore, the actual dosage regimens used can vary considerably and may therefore deviate from the dosage regimens stated herein.

Various formulations of the antibodies of this invention can be used for administration. In some embodiments, the antibodies can be administered in their pure form. In some embodiments, the antibodies and pharmaceutically acceptable excipients can be contained in various formulations. Pharmaceutically acceptable excipients are known in the art and are relatively inert substances that facilitate the administration of pharmacologically effective substances. For example, excipients can impart shape or consistency or act as diluents. Suitable excipients include, but are not limited to, stabilizers, wetting agents and emulsifiers, salts for altering molar osmotic pressure concentrations, encapsulation agents, buffers, and skin penetration enhancers. Excipients, as well as formulations for parenteral and non-parenteral drug delivery, are described in Remington, The Science and Practice of Pharmacy, 21st edition. Mack Publishing, 2005.

In some implementations, these agents are formulated for administration by injection (e.g., intraperitoneal, intravenous, subcutaneous, intramuscular, etc.). Therefore, these agents can be combined with pharmaceutically acceptable mediators, such as saline, Ringer's solution, glucose solution, etc. Specific dosing regimens, i.e., dosage, timing, and repetition, will depend on the specific individual and their medical history.

The antibodies described herein can be administered using any suitable method, including by injection (e.g., intraperitoneal, intravenous, subcutaneous, intramuscular, etc.). As described herein, antibodies, such as monoclonal or bispecific antibodies, can also be administered by inhalation. Generally, the dosage for administration of the antibodies of the present invention depends on the host being treated and the specific administration method. In one embodiment, the dosage range of the antibodies of the present invention will be from about 0.001 µg/kg body weight to about 20,000 µg/kg body weight. The term “body weight” may be used when treating a patient. When processing isolated cells, “body weight” as used herein refers to “total cell weight”. The term “total weight” can be used for both isolated cells and patient treatment. In this application, all concentrations and treatment levels are expressed as “body weight” or simply “kg”, and are also considered to include similar “total cell weight” and “total weight” concentrations. However, those skilled in the art will recognize the practicality of a variety of dosage ranges, such as 0.01 µg/kg body weight to 20,000 µg/kg body weight, 0.02 µg/kg body weight to 15,000 µg/kg body weight, 0.03 µg/kg body weight to 10,000 µg/kg body weight, 0.04 µg/kg body weight to 5,000 µg/kg body weight, 0.05 µg/kg body weight to 2,500 µg/kg body weight, 0.06 µg/kg body weight to 1,000 µg/kg body weight, 0.07 µg/kg body weight to 500 µg/kg body weight, 0.08 µg/kg body weight to 400 µg/kg body weight, 0.09 µg/kg body weight to 200 µg/kg body weight, or 0.1 µg/kg body weight to 100 µg/kg body weight. Furthermore, those skilled in the art will recognize that a variety of different dosage levels will be useful, for example, 0.0001 µg/kg, 0.0002 µg/kg, 0.0003 µg/kg, 0.0004 µg/kg, 0.005 µg/kg, 0.0007 µg/kg, 0.001 µg/kg, 0.1 µg/kg, 1.0 µg/kg, 1.5 µg/kg, 2.0 µg/kg, 5.0 µg/kg, 10.0 µg/kg, 15.0 µg/kg, 30.0 µg/kg, 50 µg/kg, 75 µg/kg, 80 µg/kg, 90 µg/kg, 100 µg/kg, 120 µg/kg, 140 µg/kg. g/kg, 150 µg/kg, 160 µg/kg, 180 µg/kg, 200 µg/kg, 225 µg/kg, 250 µg/kg, 2 75µg/kg, 300 µg/kg, 325 µg/kg, 350 µg/kg, 375µg/kg, 400 µg/kg, 450 µg/k g, 500 µg/kg, 550 µg/kg, 600 µg/kg, 700 µg/kg, 750 µg/kg, 800 µg/kg, 900 µg/kg, 1 µg/kg, 5 µg/kg, 10 µg/kg, 12 µg/kg, 15 mg/kg, 20 mg/kg and/or 30 mg/kg. All these dosages are exemplary, and any dosage between these points is also expected to be useful in this invention. Any of the above dosage ranges or dosage levels can be used with the antibodies of this invention. For repeated administration over a period of several days or longer, depending on the condition, treatment may continue until the desired symptom suppression is achieved or until a sufficient therapeutic level is reached, for example, to inhibit or delay tumor growth/progression or metastasis of cancer cells.

Depending on the desired pattern of pharmacokinetic decay, other dosing regimens may also be useful. In one embodiment, the antibody of the present invention is administered at an initial initiating dose followed by higher and/or continuous, substantially constant doses. In some embodiments, dosing is anticipated from once weekly to four times weekly. In some embodiments, dosing is anticipated from once monthly, every other month, or every three months. Progression of the therapy can be easily monitored using routine techniques and assays. The dosing regimen may vary over time.

For the purposes of this invention, the appropriate dosage of the antibody will depend on the antibody or a combination thereof used, the type and severity of the symptoms to be treated, whether the agent is administered for therapeutic purposes, prior therapy, the patient's clinical history and response to the agent, the patient's clearance rate of the administered agent, and the physician's judgment. Typically, the physician will administer the antibody until a dosage is reached to achieve the desired outcome. The dosage and/or frequency may vary during the treatment process. Empirical considerations such as half-life often aid in dosage determination. For example, antibodies compatible with the human immune system, such as humanized antibodies or fully human antibodies, can be used to prolong the antibody's half-life and prevent the antibody from being attacked by the host immune system. The frequency of administration can be determined and adjusted during the treatment process and is generally, but not necessarily, based on the treatment and/or suppression and/or improvement and/or delay of symptoms, such as tumor growth inhibition or delay. Alternatively, a sustained-release formulation of the antibody may be suitable. Various formulations and devices for achieving sustained release are known in the art.

In one implementation, the antibody dosage can be empirically determined in individuals who have received one or more doses of the antibody. Incremental doses of the antibody are administered to the individuals. Indicators of the disease can be tracked to assess efficacy.

In some implementations, the administration of the antibody results in at least one effect selected from the following: inhibition of tumor growth, tumor regression, reduction of tumor size, reduction of tumor cell number, delay of tumor growth, abscopal effect, inhibition of tumor metastasis, reduction of metastatic lesions over time, reduction of use of chemotherapy agents or cytotoxic agents, reduction of tumor burden, increase of progression-free survival, increase of overall survival, complete response, partial response, and stable disease.

Depending on factors such as the recipient's physiological condition, whether the purpose of administration is therapeutic or prophylactic, and other factors known to a skilled professional, the administration of antibodies according to the method of the invention can be continuous or intermittent. Antibody administration can be substantially continuous over a preselected time period, or it can be a series of doses spaced apart from each other.

In some implementations, more than one antibody may be present. At least one, at least two, at least three, at least four, at least five, or more different antibodies may be present. Typically, these antibodies may have complementary activities that do not harmfully interfere with each other. For example, one or more of the following antibodies may be used: a first CD3 antibody targeting one epitope on CD3 and a second CD3 antibody targeting a different epitope on CD3.

In some embodiments, the antibodies disclosed herein can be used, in particular, for parenteral administration, such as intravenous administration or administration into the cavities of body cavities or organs.

Therapeutic formulations of the antibodies used according to the invention are prepared for storage by mixing antibodies of desired purity with optional pharmaceutically acceptable carriers, excipients, or stabilizers (Remington, The Science and Practice of Pharmacy, 21st edition. Mack Publishing, 2005) in the form of lyophilized formulations or aqueous solutions. Acceptable carriers, excipients, or stabilizers are non-toxic to the recipient at the doses and concentrations used and may include buffers such as phosphates, citrates, and other organic acids; salts such as sodium chloride; antioxidants, including ascorbic acid and methionine; preservatives (e.g., octadecyl dimethyl benzyl ammonium chloride; hexane diammonium chloride; benzalkonium chloride, benzyl chloride; phenols, butanol, or benzyl alcohol; alkyl esters of p-hydroxybenzoate, such as methylparaben or propylparaben; catechol; resorcinol; cyclohexanol; 3-pentanol; and m-cresol); low molecular weight (small) The following are considered as salts: polypeptides (approximately 10 residues); proteins, such as serum albumin, gelatin, or immunoglobulins; hydrophilic polymers, such as polyvinylpyrrolidone; amino acids, such as glycine, glutamine, asparagine, histidine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates, including glucose, mannose, or dextrin; chelating agents, such as EDTA; sugars, such as sucrose, mannitol, trehalose, or sorbitol; salt-forming ions, such as sodium; metal complexes (e.g., Zn-protein complexes); and/or nonionic surfactants, such as TWEEN ^™ , PLURONICS ^™ , or polyethylene glycol (PEG).

Antibody-containing liposomes are prepared by methods known in the art, for example, as described in Epstein et al., Proc. Natl. Acad. Sci. USA 82:3688, 1985; Hwang et al., Proc. Natl. Acad. Sci. USA 77:4030, 1980; and U.S. Patent Nos. 4,485,045 and 4,544,545. Liposomes with extended cycle times are disclosed in U.S. Patent No. 5,013,556. Particularly useful liposomes can be produced by a reverse-phase evaporation method using a lipid composition comprising phosphatidylcholine, cholesterol, and PEG-derived phosphatidylethanolamine (PEG-PE). The liposomes are extruded through a filter having a defined pore size to produce liposomes with a desired diameter.

The active ingredient can also be encapsulated in microcapsules (e.g., hydroxymethyl cellulose or gelatin microcapsules and poly-(methyl methacrylate) microcapsules) prepared by agglomeration technology or interfacial polymerization, colloidal drug delivery systems (e.g., liposomes, albumin microspheres, microemulsions, nanoparticles and nanocapsules) or coarse droplet emulsions. This technology is disclosed in Remington, The Science and Practice of Pharmacy, 21st edition, Mack Publishing, 2005.

Sustained-release formulations can be prepared. Suitable examples of sustained-release formulations include semi-permeable matrices of solid hydrophobic polymers containing antibodies, said matrices being in the form of shaped articles, such as membranes or microcapsules. Examples of sustained-release matrices include polyesters, hydrogels (e.g., poly(2-hydroxyethyl methacrylate) or poly(vinyl alcohol)), polylactides (US Patent No. 3,773,919), copolymers of L-glutamic acid and L-glutamic acid 7-ethyl ester, non-degradable ethylene vinyl acetate, degradable lactic acid-glycolic acid copolymers, such as LUPRON DEPOT ^™ (injectable microspheres composed of lactic acid-glycolic acid copolymers and leuprolide acetate), sucrose isobutyrate acetate, and poly-D-(-)-3-hydroxybutyrate.

Preparations intended for in vivo administration must be sterile. This can be easily achieved, for example, by filtration through a sterile membrane. Therapeutic antibody (e.g., CD3 antibody) compositions are typically placed in containers with sterile inlets and outlets, such as intravenous solution bags or vials with stoppers that can be punctured by a subcutaneous needle.

The compositions according to the invention may be in unit dosage form, such as tablets, pills, capsules, powders, granules, solutions or suspensions, or suppositories, for oral, parenteral or rectal administration, or administration by inhalation or blowing.

To prepare solid compositions such as tablets, the main active ingredient is mixed with a pharmaceutical carrier (e.g., conventional tableting ingredients such as corn starch, lactose, sucrose, sorbitol, talc, stearic acid, magnesium stearate, dicalcium phosphate, or gum, and other pharmaceutical diluents, such as water) to form a solid preformulation composition containing a homogeneous mixture of the compound of the present invention or a pharmaceutically acceptable, non-toxic salt thereof. When these preformulation compositions are referred to as homogeneous, it means that the active ingredient is uniformly dispersed throughout the composition, making it easy to subdivide the composition into equivalent effective unit dose forms, such as tablets, pills, and capsules. The solid preformulation composition is then subdivided into unit dose forms of the type described above, containing 0.1 to about 500 mg of the active ingredient of the present invention. Tablets or pills of the novel composition can be coated or otherwise compounded to provide a dosage form that provides the advantage of prolonged action. For example, tablets or pills may contain an internal dose component and an external dose component, the latter being in the form of an outer coating on top of the former. These two components can be separated by an enteric layer, which resists breakdown in the stomach and allows the internal components to enter the duodenum intact or with delayed release. A variety of materials can be used for this enteric layer or coating, including many polymeric acids and mixtures of polymeric acids with materials such as shellac, cetyl alcohol, and cellulose acetate.

Suitable surfactants particularly include nonionic agents such as polyoxyethylene sorbitol (e.g., Tween ^™ 20, 40, 60, 80, or 85) and other sorbitols (e.g., Span ^™ 20, 40, 60, 80, or 85). Compositions containing surfactants will conveniently contain 0.05 to 5% surfactant, and may be 0.1 to 2.5%. It will be understood that, where necessary, other ingredients, such as mannitol or other pharmaceutically acceptable mediators, may be added.

The compositions within the scope of this invention include all compositions in which the antibody is present in an amount that effectively achieves the desired medical effect for treating cancer. Although individual needs may differ from one patient to another, the determination of the optimal range of effective amounts of all components is within the capabilities of a clinician with ordinary skills.

Examples of such compositions and how they are formulated have also been described in earlier sections and below. In some embodiments, the composition comprises one or more antibodies. In some embodiments, the antibody is a human antibody, a humanized antibody, or a chimeric antibody. In some embodiments, the antibody comprises a constant region capable of triggering a desired immune response (e.g., antibody-mediated cleavage or ADCC). In some embodiments, the antibody comprises a constant region that does not trigger unwanted or undesirable immune responses (e.g., antibody-mediated cleavage or ADCC).

It should be understood that the composition may contain more than one antibody, such as a mixture of CD3 antibodies that recognize CD3 or different epitopes of CD3 and tumor antigens. Other exemplary compositions contain more than one antibody that recognizes one or more of the same epitopes, or different kinds of antibodies, or different epitopes of CD3 and tumor antigens (e.g., human CD3).

In some embodiments, the antibody may be administered in combination with the administration of one or more additional therapeutic agents. These include, but are not limited to, the administration of biological and/or chemotherapeutic agents, such as, but not limited to, vaccines, CAR-T cell-based therapies, radiotherapy, cytokine therapy, CD3 bispecific antibodies, inhibitors of other immunosuppressive pathways, inhibitors of angiogenesis, T-cell activators, inhibitors of metabolic pathways, mTOR inhibitors, inhibitors of the adenosine pathway, tyrosine kinase inhibitors, including but not limited to Inlyta, ALK inhibitors and sunitinib, BRAF inhibitors, epigenetic modifying factors, IDO1 inhibitors, JAK inhibitors, STAT inhibitors, and cell cycle proteins. Leukocyte-dependent kinase inhibitors, biotherapeutic agents (including but not limited to antibodies against VEGF, VEGFR, EGFR, Her2/neu, other growth factor receptors, CD40, CD-40L, CTLA-4, OX-40, 4-1BB, TIGIT, and ICOS), immunogenic agents (e.g., attenuated cancer cells, tumor antigens, antigen-presenting cells such as dendritic cells bombarded with tumor-derived antigens or nucleic acids, immunostimulatory cytokines (e.g., IL-2, IFNα2, GM-CSF), and cells transfected with genes encoding immunostimulatory cytokines (e.g., but not limited to GM-CSF).

Examples of biotherapeutic agents include therapeutic antibodies, immunomodulators, and therapeutic immune cells.

Therapeutic antibodies may be specific to a variety of different antigens. For example, a therapeutic antibody may target a tumor-associated antigen, such that the binding of the antibody to the antigen promotes the death of cells expressing the antigen. In other instances, a therapeutic antibody may target an antigen on immune cells (e.g., PD-1), such that the binding of the antibody prevents the downregulation of the activity of cells expressing the antigen (and thereby promotes the activity of cells expressing the antigen). In some cases, a therapeutic antibody may act through a variety of different mechanisms (e.g., upon contact with cells expressing the antigen, it may both i) promote the death of cells expressing the antigen and ii) prevent the downregulation of immune cell activity caused by the antigen).

Therapeutic antibodies can target antigens such as those listed below. For some antigens, exemplary antibodies against said antigens are also included below (in parentheses after the antigen). The antigens listed below may also be referred to herein as "target antigens," etc. The target antigens used for therapeutic antibodies in this article include, for example: 4-1BB (e.g., utomilumab); 5T4; A33; α-folate receptor 1 (e.g., mirvetuximab soravtansine); Alk-1; BCMA [e.g., PF-06863135 (see US9969809)]; BTN1A1 (e.g., see WO2018222689); CA-125 (e.g., abagovomab); carbonic anhydrase IX; CCR2; CCR4 (e.g., mogamulizumab); CCR5 (e.g., leronlimab); CCR8; CD3 [e.g., blinatumomab (CD3/ CD19 bispecific), PF-06671008 (CD3/P-cadherin bispecific), PF-06863135 (CD3/BCMA bispecific), PF-07062119 (CD3/GUCY2c bispecific), CD19 (e.g., ibritumomab tiuxetan, MOR208); CD20 (e.g., ibritumomab tiuxetan, obinutuzumab, ofatumumab, rituximab, ublituximab); CD22 (inotuzumab ozogamicin, moxetumomab pasudotox); CD25; CD28; CD30 (e.g., brentuximab vedotin); CD33 (e.g., gemtuzumab ozogamicin); CD38 (e.g., daratumumab, isatuximab); CD40; CD-40L; CD44v6; CD47; CD52 (e.g., alemtuzumab); CD63; CD79 (e.g., polotuzumab vedotin); CD80; CD123; CD276/B7-H3 (e.g., omburtamab); CDH17; CEA; ClhCG; CTLA-4 (e.g., ipilimumab, tremel... imumab), CXCR4; desmosome core protein 4; DLL3 (e.g., rovalpituzumab tesirine); DLL4; E-cadherin; EDA; EDB; EFNA4; EGFR (e.g., cetuximab, depatuxizumab mafodotin, necitumumab, panitumumab); EGFRvIII; endothelial sialic acid protein; EpCAM (e.g., oportuzumab monatox); FAP; fetal acetylcholine receptor; FLT3 (e.g., see WO2018/220584); GD2 (e.g., dinutuximab, 3F8); GD3 ; GITR; GloboH; GM1; GM2; GUCY2C (e.g., PF-07062119); HER2/neu [e.g., margetuximab, pertuzumab, trastuzumab; ado-trastuzumab emtansine, trastuzumab duocarmazine, PF-06804103 (see US8828401)]; HER3; HER4; ICOS; IL-10; ITG-AvB6; LAG-3 (e.g., relatlimab); Lewis-Y; LG; Ly-6; M-CSF [e.g., PD-0360324 (see US8828401)]; HER3; HER4; ICOS; IL-10; ITG-AvB6; LAG-3 (e.g., relatlimab); Lewis-Y; LG; Ly-6; M-CSF [e.g., PD-0360324 (see US8828401)]; S7326414); MCSP; mesothelin; MUC1; MUC2; MUC3; MUC4; MUC5AC; MUC5B; MUC7; MUC16; Notch 1; Notch 3; Nectin-4 (e.g., enfortumab vedotin); OX40 [e.g., PF-04518600 (see US7960515)]; P-cadherein [e.g., PF-06671008 (see WO2016/001810)]; PCDHB2; PD-1 [e.g., BCD-100, camrelizumab, cemiplimab, genolimab] Nivolumab (CBT-501), MEDI0680 (CBT-501), nivolumab, pembrolizumab, RN888 (see WO2016/092419), sintilimab, spartalizumab, STI-A1110, tislelizumab, TSR-042; PD-L1 (e.g., atezolizumab, durvalumab, BMS-936559 (MDX-1105) or LY3300054); PDGFRA (e.g., olaratumab); plasma cell antigen; PolySA; PSCA; PSMA; PTK7 [e.g., PF-06647020 (see US9409995)]; Ror1; SAS; SCRx6; SLAMF7 (e.g., elotuzumab); SHH; SIRPa (e.g., ED9, Effi-DEM); STEAP; TGF-β; TIGIT; TIM-3; TMPRSS3; TNF-α precursor; TROP-2 (e.g., sacituzumabgovitecan); TSPAN8; VEGF (e.g., bevacizumab, brolucizumab); VEGFR1 (e.g., ranibizumab); VEGFR2 (e.g., ramucirumab, ranibizumab); Wue-1.

Therapeutic antibodies can take any suitable form. For example, therapeutic antibodies can take any form described elsewhere herein. In some embodiments, the therapeutic antibody may be a naked antibody. In some embodiments, the therapeutic antibody may be conjugated to a drug/pharmaceutical agent (also known as an "antibody-drug conjugate" (ADC)). In some embodiments, a therapeutic antibody targeting a specific antigen may be incorporated into a multispecific antibody (e.g., a bispecific antibody).

In some embodiments, antibodies against antigens may be conjugated to drugs/pharmaceuticals. The linked antibody-drug molecule is also referred to as an "antibody-drug conjugate" (ADC). The drug/pharmaceutical can be linked to the antibody directly or indirectly via a linker. Most commonly, toxic drugs are linked to antibodies such that the binding of the ADC to the corresponding antigen promotes the killing of cells expressing said antigen. For example, ADCs linked to toxic drugs are particularly useful for targeting tumor-associated antigens to promote the killing of tumor cells expressing tumor-associated antigens. In other embodiments, the agent that can be linked to the antibody may be, for example, an immunomodulator (e.g., to modulate the activity of immune cells near the ADC), an imaging agent (e.g., to facilitate imaging of the ADC in a subject or a biological sample from a subject), or an agent that increases the serum half-life or biological activity of the antibody.

Methods for conjugating cytotoxic agents or other therapeutic agents to antibodies have been described in various publications. For example, antibodies can be chemically modified to induce conjugation by lysine side-chain amines or by cysteine thiol groups activated by reduced interchain disulfide bonds. See, for example, Tanaka et al., FEBS Letters 579:2092-2096, 2005 and Gentle et al., Bioconjugate Chem. 15:658-663, 2004. Reactive cysteine residues engineered at specific sites on antibodies for conjugation to specific drugs at defined stoichiometric amounts have also been described. See, for example, Junutula et al., Nature Biotechnology, 26:925-932, 2008. International applications WO2012/059882 and WO2015015448 also describe the use of glutamine-containing tags with acyl donors or endogenous glutamine conjugated by peptide engineering to an active state (i.e., the ability to form covalent bonds as acyl donors) in the presence of transglutaminase and amines (e.g., cytotoxic agents containing or attached to active amines). In some embodiments, the ADC may have any of the features or characteristics of the ADCs provided in WO2016166629, which is incorporated herein by reference for all purposes.

Drugs/pharmaceuticals that can be linked to antibodies in the form of ADCs can include, for example, cytotoxic agents, immunomodulators, imaging agents, therapeutic proteins, biopolymers, or oligonucleotides.

Exemplary cytotoxic agents that can be incorporated into ADCs include anthracyclines, auristatin, dolastatin, compbretastatin, duocarmycin, pyrrolobenzodiazepine dimer, indoline-benzodiazepine dimer, enediyne, geldemycin, maytansine, puromycin, taxane, vinca alkaloids, camptothecin, tubulysin, hemiasterlin, spliceostatin, pladienolide, and their stereoisomers, isomers, analogs, or derivatives.

Exemplary immunomodulators that can be incorporated into ADCs include gancyclovier, etanercept, tacrolimus, sirolimus, voclosporin, cyclosporine, rapamycin, cyclophosphamide, azathioprine, mycophenolate mofetil, methotrexate, glucocorticoids and their analogues, cytokines, and stem cell production. Growth factors, lymphotoxin, tumor necrosis factor (TNF), hematopoietic factors, interleukins (e.g., interleukin-1 (IL-1), IL-2, IL-3, IL-6, IL-10, IL-12, IL-18, and IL-21), colony-stimulating factors (e.g., granulocyte colony-stimulating factor (G-CSF) and granulocyte-macrophage colony-stimulating factor (GM-CSF)), interferons (e.g., interferon-α, -β, and -γ), stem cell growth factors known as "S1 factors", erythropoietin, and thrombopoietin, or combinations thereof.

Exemplary imaging agents that may be included in an ADC include fluorescein, rhodamine, lanthanide phosphors and their derivatives, or radioisotopes bound to a chelating agent. Examples of fluorophores include, but are not limited to, fluorescein isothiocyanate (FITC) (e.g., 5-FITC), fluorescein phosphoramide (FAM) (e.g., 5-FAM), eosin, carboxyfluorescein, erythrosine, AlexaFluor.RTM. (e.g., Alexa 350, 405, 430, 488, 500, 514, 532, 546, 555, 568, 594, 610, 633, 647, 660, 680, 700, or 750), carboxytetramethylrhodamine (TAMRA) (e.g., 5-TAMRA), tetramethylrhodamine (TMR), and sulforhodamine (SR) (e.g., SR101). Examples of chelating agents include, but are not limited to, 1,4,7,10-tetraazacyclododecane-N,N',N'',N'''-tetraacetic acid (DOTA), 1,4,7-triazacyclononane-1,4,7-triacetic acid (NOTA), 1,4,7-triazacyclononane, 1-glutaric acid-4,7-acetic acid (deferroamine), diethylenetriaminepentaacetic acid (DTPA), and 1,2-bis(o-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA).

Exemplary therapeutic proteins that may be included in ADCs include toxins, hormones, enzymes, and growth factors.

Exemplary biocompatible polymers that can be incorporated into ADCs include water-soluble polymers, such as polyethylene glycol (PEG) or derivatives thereof, and biocompatible polymers containing zwitterions (e.g., polymers containing phosphocholine).

Exemplary biocompatible polymers that can be incorporated into ADCs include antisense oligonucleotides.

In some implementations, the antibodies against antigens provided herein can be incorporated into bispecific antibody molecules. A bispecific antibody is a monoclonal antibody that has binding specificity to at least two different antigens.

In some embodiments, the bispecific antibody comprises a first antibody variable domain and a second antibody variable domain, wherein the first antibody variable domain is capable of recruiting human immune effector cells by specifically binding to CD3 as provided herein, and wherein the second antibody variable domain is capable of specifically binding to a target antigen. In some embodiments, the antibody has an IgG1, IgG2, IgG3, or IgG4 isotype. In some embodiments, the antibody comprises an immune-inert Fc region. In some embodiments, the antibody is a human antibody or a humanized antibody.

Target antigens are typically expressed on target cells (e.g., cancer cells) in disease conditions. Examples of target antigens of particular interest for bispecific antibodies include, but are not limited to, BCMA, EpCAM (epithelial cell adhesion molecule), CCR5 (type 5 chemokine receptor), CD19, HER (human epidermal growth factor receptor)-2/neu, HER-3, HER-4, EGFR (epidermal growth factor receptor), PSMA, CEA, MUC-1 (mucin), MUC2, MUC3, MUC4, MUC5AC, MUC5B, MUC7, CIhCG, Lewis-Y, CD20, CD33, CD30, ganglioside GD3, 9-O-acetyl-GD3, GM2, Globo H, fucose GM1, Poly SA, GD2, carbonic anhydrase IX (MN/CA IX), CD44v6, and Shh (Sonic Hedgkinase). ehog), Wue-1, plasma cell antigen, (membrane-bound) IgE, MCSP (chondroitin sulfate proteoglycan for melanoma), CCR8, TNF-α precursor, STEAP, mesothelin, A33 antigen, PSCA (prostate stem cell antigen), Ly-6; desmosome core glycoprotein 4, E-cadherin epitope, fetal acetylcholine receptor, CD25, CA19-9 marker, CA-125 marker and MIS (Muellerian Inhibitory Substance) type II receptor, sTn (sialylated Tn antigen; TAG-72), FAP (fibroblast activation antigen), endothelial sialic acid protein, EGFRvIII, LG, SAS, PD-L1, CD47, SIRPa and CD63.

In some embodiments, the bispecific antibody comprises a full-length human antibody, wherein the first antibody variable domain of the bispecific antibody is capable of recruiting human immune effector cells by specifically binding to CD3, and wherein the second antibody variable domain of the heterodimeric protein is capable of specifically binding to a target antigen (e.g., CD20, EpCAM, or P-cadherin).

Methods for preparing bispecific antibodies are known in the art (see, for example, Suresh et al., Methods in Enzymology 121:210, 1986). Traditionally, recombinant production of bispecific antibodies is based on the co-expression of two immunoglobulin heavy-light chain pairs, where the two heavy chains have different specificities (Millstein and Cuello, Nature 305, 537-539, 1983).

According to a method for preparing bispecific antibodies, an antibody variable domain having the desired binding specificity (antibody-antigen combination site) is fused to an immunoglobulin constant region sequence. The fusion preferably has an immunoglobulin heavy chain constant region comprising at least a portion of the hinge, CH2, and CH3 regions. Preferably, it has a first heavy chain constant region (CH1) present in at least one of the fusions, containing the site necessary for light chain binding. DNA encoding the immunoglobulin heavy chain fusion and (if desired) the immunoglobulin light chain is inserted into separate expression vectors and co-transfected into a suitable host organism. This provides great flexibility in the embodiments to adjust the relative proportions of the three polypeptide fragments when unequal ratios of the three polypeptide chains used in the construction provide optimal yield. However, when expression of at least two polypeptide chains in equal ratios results in high yield, or when the ratios are not particularly significant, it is possible to insert coding sequences for two or all three polypeptide chains into a single expression vector.

In one method, the bispecific antibody comprises a hybrid immunoglobulin heavy chain having a first binding specificity in one arm and a hybrid immunoglobulin heavy chain-light chain pair (providing a second binding specificity) in the other arm. This asymmetric structure, with the immunoglobulin light chain in only half of the bispecific molecule, facilitates the separation of the desired bispecific compound from the unwanted combination of immunoglobulin chains. This method is described in PCT Publication No. WO 94/04690.

In another approach, the bispecific antibody is constructed by modifying an amino acid in a first hinge region of one arm, wherein the substituted/alternated amino acid in the first hinge region has an opposite charge to the corresponding amino acid in a second hinge region of the other arm. This approach is described in International Patent Application No. PCT/US2011/036419 (WO2011/143545).

In another approach, the formation of a desired heteromeric or heterodimeric protein (e.g., a bispecific antibody) is enhanced by altering or engineering the interface between the first and second immunoglobulin-like Fc regions (e.g., hinge regions and/or CH3 regions). In this approach, the bispecific antibody may be composed of a CH3 region containing a first and a second CH3 polypeptide that interact together to form a CH3 interface, wherein one or more amino acids within the CH3 interface destabilize homodimer formation and are electrostatically unfavorable to homodimer formation. This approach is described in International Patent Application No. PCT/US2011/036419 (WO2011/143545).

In another approach, bispecific antibodies can be generated in the presence of transglutaminase using a glutamine-containing peptide tag engineered for an antibody targeting an epitope (e.g., BCMA) in one arm, and another peptide tag engineered for a second antibody targeting a second epitope in the other arm (e.g., a Lys-containing peptide tag or reactive endogenous Lys). This approach is described in International Patent Application No. PCT/IB2011/054899 (WO2012/059882).

In some embodiments, the first and second antibody variable domains of the bispecific antibody include amino acid modifications at various positions, wherein the first and second antibody variable domains of the bispecific antibody include amino acid modifications at positions 223, 225, and 228 in the hinge region of human IgG2 (e.g., (C223E or C223R), (E225R), and (P228E or P228R)) and amino acid modifications at positions 409 or 368 in the CH3 region (e.g., K409R or L368E (EU numbering scheme)).

In some embodiments, the first and second antibody variable domains of the bispecific antibody include amino acid modifications at positions 221 and 228 in the hinge region of human IgG1 (e.g., (D221R or D221E) and (P228R or P228E)) and amino acid modifications at positions 409 or 368 in the CH3 region (e.g., K409R or L368E (EU numbering scheme)).

In some embodiments, the first and second antibody variable domains of the bispecific antibody include an amino acid modification at position 228 in the hinge region of human IgG4 (e.g., (P228E or P228R)) and an amino acid modification at position 409 or 368 in the CH3 region (e.g., R409 or L368E (EU numbering scheme)).

In some implementations, the bispecific antibody may have any of the features or properties of any bispecific antibody provided in WO2016166629, which is incorporated herein by reference for all purposes.

Immunomodulators include a variety of different molecular types that can stimulate immune responses in subjects, such as pattern recognition receptor (PRR) agonists, immunostimulatory cytokines, and cancer vaccines.

Pattern recognition receptors (PRRs) are receptors expressed by cells of the immune system that recognize a variety of molecules associated with pathogens and/or cell damage or death. PRRs are involved in both innate and adaptive immune responses. PRR agonists can be used to stimulate immune responses in subjects. Several classes of PRR molecules exist, including toll-like receptors (TLRs), RIG-I-like receptors (RLRs), nucleotide-binding oligomerization domain (NOD)-like receptors (NLRs), C-type lectin receptors (CLRs), and interferon gene stimulator (STING) proteins.

The terms “TLR” and “toll-like receptor” refer to any toll-like receptor. Toll-like receptors are receptors involved in activating immune responses. TLRs recognize, for example, pathogen-associated molecular patterns (PAMPs) expressed in microorganisms, and endogenous damage-associated molecular patterns (DAMPs) released from dying or dying cells.

Molecules that activate TLRs (and thus activate the immune response) are referred to herein as “TLR agonists.” TLR agonists can include, for example, small molecules (e.g., organic molecules with a molecular weight of less than about 1000 Daltons) as well as large molecules (e.g., oligonucleotides and proteins). Some TLR agonists are specific to a single type of TLR (e.g., TLR3 or TLR9), while others activate two or more types of TLRs (e.g., both TLR7 and TLR8).

The exemplary TLR agonists provided herein include agonists of TLR2, TLR3, TLR4, TLR5, TLR6, TLR7, TLR8, and TLR9.

Exemplary small molecule TLR agonists include, for example, U.S. Patent Nos. 4,689,338; 4,929,624; 5,266,575; 5,268,376; 5,346,905; 5,352,784; 5,389,640; 5,446,153; 5,482,936; 5,756,747; 6,110,929; 6,194,425; 6,331,539; 6,376,669; 6,451,810; 6,525,064; 6,541,485; 6,545,016; 6,545,017; 6,573,273; 6,656,938; 6,660,735; 6 660,747; 6,664,260; 6,664,264; 6,664,265; 6,667,312; 6,670,372; 6,677,347; 6,677,348; 6,677,349; 6,683,088; 6,756,382; 6,797,718; 6,818,650; and 7,7091,214; U.S. Patent Publications 2004/0091491, 2004/0176367 and 2006/0100229; and International Publications WO 2005/18551, WO2005/18556, WO 2005/20999, WO 2005/0 32484, WO 2005/048933, WO 2005/048945, WO 2005/051317, WO2005/051324, WO 2005/066169, WO 2005/066170, WO 2005/066172, WO 2005/076783, WO 2005/079195, WO2005/094531, WO 2005/123079, WO 2005/123080, WO 2006/009826, WO 2006/009832, WO 2006/026760, WO2006/028451 WO 2006/028545、WO 2006/028962、WO 2006/029115、WO 2006/038923、WO 2006/065280、WO2006/074003、WO 2006/083440、WO 2006/086449、WO 2006/091394, WO 2006/086633, WO 2006/086634, WO 2006/091567, WO 2006/091568, WO 2006/091647, WO 2006/093514 and WO 2006/098852.

Additional examples of small molecule TLR agonists include certain purine derivatives (such as those described in U.S. Patent Nos. 6,376,501 and 6,028,076), certain imidazoquinoline amide derivatives (such as those described in U.S. Patent No. 6,069,149), certain imidazopyridine derivatives (such as those described in U.S. Patent No. 6,518,265), and certain benzimidazole derivatives (such as those described in U.S. Patent No. 6,387,938). Certain derivatives of 4-aminopyrimidine fused with a five-membered nitrogen-containing heterocycle (e.g., adenine derivatives described in U.S. Patent Nos. 6,376,501, 6,028,076, and 6,329,381; and WO 02/08905) and certain 3-β-D-furanoribosylthiazo[4,5-d]pyrimidine derivatives (e.g., those described in U.S. Publication No. 2003/0199461) and certain small molecule immune enhancer compounds, such as those described in U.S. Patent Publication No. 2005/0136065.

Exemplary macromolecular TLR agonists comprise oligonucleotide sequences. Some TLR agonist oligonucleotide sequences contain cytosine-guanine dinucleotides (CpG) and are described, for example, in U.S. Patent Nos. 6,194,388; 6,207,646; 6,239,116; 6,339,068; and 6,406,705. Some CpG-containing oligonucleotides may include synthetic immunomodulatory structural motifs, such as those described, for example, in U.S. Patent Nos. 6,426,334 and 6,476,000. Other TLR agonist nucleotide sequences lack the CpG sequence and are described, for example, in International Patent Publication No. WO 00/75304. Other TLR agonist nucleotide sequences include single-stranded RNAs (ssRNAs) rich in guanosine and uridine, such as those described, for example, in Heil et al., Science, vol. 303, pp. 1526-1529, March 5, 2004.

Other TLR agonists include biomolecules such as aminoalkyl glucosinolate phosphates (AGPs), and are described, for example, in U.S. Patent Nos. 6,113,918; 6,303,347; 6,525,028; and 6,649,172.

TLR agonists also include inactivated pathogens or fractions thereof, which can activate a variety of different types of TLR receptors. Exemplary pathogen-derived TLR agonists include BCG, mycobacterium obuense extract, Talimogene laherparepvec (T-Vec) (derived from HSV-1), and Pexa-Vec (derived from vaccinia virus).

In some implementations, the TLR agonist may be an agonist antibody that specifically binds to a TLR.

The following provides a brief description of various TLRs and TLR agonists. The list of TLR agonists for a particular TLR below should not be interpreted as indicating that a given TLR agonist necessarily activates only that TLR (e.g., some molecules can activate multiple types of TLRs, or even multiple classes of PRRs). For example, some molecules provided below as exemplary TLR4 agonists could also be TLR5 agonists.

The terms “TLR1” and “toll-like receptor 1” refer to any form of TLR1 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR1 activity. Unless otherwise specified, such as by way of a specific reference to human TLR1, TLR1 includes the naturally occurring sequence TLR1 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR1 is provided under UniProt entry number Q15399.

As used herein, “TLR1 agonist” means any molecule that, upon binding to TLR1, (1) stimulates or activates TLR1, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR1, or (3) enhances, increases, promotes, or induces the expression of TLR1. TLR1 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, bacterial lipoproteins and their derivatives that bind to TLR1.

Examples of TLR1 agonists useful in the treatments, medicines, and uses of the present invention include, for example, bacterial lipoproteins and their derivatives, such as SPM-105 (derived from autoclaved mycobacteria), OM-174 (lipid A derivative), OmpS1 (porin from Salmonella typhi), OspA (from Borrelia burgdorferi), MALP-2 (mycoplasma macrophage-activated lipopeptide-2kD), STF (soluble tuberculosis factor), CU-T12-9, Diprovocim, and lipopeptides derived from cell wall components, such as PAM2CSK4, PAM3CSK4, and PAM3Cys.

TLR1 can form a heterodimer with TLR2, and therefore, many TLR1 agonists are also TLR2 agonists.

The terms “TLR2” and “toll-like receptor 2” refer to any form of TLR2 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR2 activity. Unless otherwise specified, such as by way of a specific reference to human TLR2, TLR2 includes naturally occurring sequence TLR2 in all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR2 is provided under UniProt entry number O60603.

As used herein, “TLR2 agonist” means any molecule that, upon binding to TLR2, (1) stimulates or activates TLR2, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR2, or (3) enhances, increases, promotes, or induces the expression of TLR2. TLR2 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, bacterial lipoproteins and their derivatives that bind to TLR2.

Examples of TLR2 agonists useful in the treatments, medicines, and uses of the present invention include, for example, bacterial lipoproteins (e.g., diacylated lipoproteins) and their derivatives, such as SPM-105 (derived from autoclaved mycobacteria), OM-174 (lipid A derivative), OmpS1 (porin from Salmonella typhi), OspA (from Borrelia brevicornuate), MALP-2 (mycoplasma macrophage-activated lipopeptide-2kD), STF (soluble tuberculosis factor), CU-T12-9, Diprovocim, Amplivant, and lipopeptides derived from cell wall components, such as PAM2CSK4, PAM3CSK4, and PAM3Cys.

TLR2 can form heterodimers with TLR1 or TLR6, and therefore, many TLR2 agonists are also TLR1 or TLR6 agonists.

The terms “TLR3” and “toll-like receptor 3” refer to any form of TLR3 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR3 activity. Unless otherwise specified, such as by way of a specific reference to human TLR3, TLR3 includes naturally occurring sequence TLR3 in all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR3 is provided under UniProt entry number O15455.

As used herein, “TLR3 agonist” means any molecule that, upon binding to TLR3, (1) stimulates or activates TLR3, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR3, or (3) enhances, increases, promotes, or induces the expression of TLR3. TLR3 agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, nucleic acid ligands that bind to TLR3.

Examples of TLR3 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include TLR3 ligands such as synthetic dsRNA, polyinosinic acid [“poly(I:C)”] (available from, for example, InvivoGen in high molecular weight (HMW) and low molecular weight (LMW) formulations), polyadenylic-polyuridylic acid [“poly(A:U)”] (available from, for example, InvivoGen), and polyICLC (see Levy et al., Journal of Infectious Disorders). (See Jasani et al., Vaccine, vol. 132, no. 4, pp. 434-439, 1975), Ampligen (see Jasani et al., Vaccine, vol. 27, no. 25-26, pp. 3401-3404, 2009), Hiltonol, Rintatolimod, and RGC100 (see Naumann et al., Clinical and Developmental Immunology, vol. 2013, article ID 283649).

The terms “TLR4” and “toll-like receptor 4” refer to any form of TLR4 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR4 activity. Unless otherwise specified, such as by way of a specific reference to human TLR4, TLR4 includes naturally occurring sequence TLR4 in all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR4 is provided under UniProt entry number O00206.

As used herein, “TLR4 agonist” means any molecule that, upon binding to TLR4, (1) stimulates or activates TLR4, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR4, or (3) enhances, increases, promotes, or induces the expression of TLR4. TLR4 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, bacterial lipopolysaccharide (LPS) and its derivatives that bind to TLR4.

Examples of TLR4 agonists useful in the treatments, medicines and uses of the present invention include, for example, bacterial lipopolysaccharide (LPS) and its derivatives, such as B:0111 (Sigma), monophospholipid A (MPLA), 3DMPL (3-O-deacylated MPL), GLA-AQ, G100, AS15, ASO2, GSK1572932A (GlaxoSmithKline, UK).

The terms “TLR5” and “toll-like receptor 5” refer to any form of TLR5 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR5 activity. Unless otherwise specified, such as by way of a specific reference to human TLR5, TLR5 includes naturally occurring sequence TLR5s from all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR5 is provided under UniProt entry number O60602.

As used herein, “TLR5 agonist” means any molecule that, upon binding to TLR5, (1) stimulates or activates TLR5, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR5, or (3) enhances, increases, promotes, or induces the expression of TLR5. TLR5 agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, bacterial flagellin and derivatives thereof that bind to TLR5.

Examples of TLR5 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include, for example, bacterial flagellin purified from Bacillus subtilis, flagellin purified from Pseudomonas aeruginosa, flagellin purified from Salmonella typhimurium, and recombinant flagellin (all available from InvivoGen), and entolimod (CBLB502; a pharmacologically optimized flagellin derivative).

The terms “TLR6” and “toll-like receptor 6” refer to any form of TLR6 receptor, as well as variants, isotypes, and species homologs that retain at least a portion of TLR6 activity. Unless otherwise specified, such as by way of a specific reference to human TLR6, TLR6 includes the naturally occurring sequence TLR6 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR6 is provided under UniProt entry number Q9Y2C9.

As used herein, “TLR6 agonist” means any molecule that, upon binding to TLR6, (1) stimulates or activates TLR6, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR6, or (3) enhances, increases, promotes, or induces the expression of TLR6. TLR6 agonists useful in any of the treatments, medicaments, and uses of this invention include, for example, bacterial lipopeptides and derivatives thereof that bind to TLR6.

Examples of TLR6 agonists useful in the treatments, medicines, and uses of the present invention include, for example, many of the TLR2 agonists provided above, because TLR2 and TLR6 can form heterodimers. TLR6 can also form heterodimers with TLR4, and TLR6 agonists can include the various TLR4 agonists provided above.

The terms “TLR7” and “toll-like receptor 7” refer to any form of TLR7 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR7 activity. Unless otherwise specified, such as by way of a specific reference to human TLR7, TLR7 includes the naturally occurring sequence TLR7 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR7 is provided under UniProt entry number Q9NYK1.

As used herein, “TLR7 agonist” means any molecule that, upon binding to TLR7, (1) stimulates or activates TLR7, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR7, or (3) enhances, increases, promotes, or induces the expression of TLR7. TLR7 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acid ligands that bind to TLR7.

Examples of TLR7 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include recombinant single-stranded (“ss”) RNA, imidazoquinoline compounds such as imiquimod (R837), gardiquimod, and resiquimod (R848); rozolidin (7-allyl-7,8-dihydro-8-oxoguanosine) and related compounds; 7-thia-8-oxoguanosine, 7-denitroguanosine, and related guanosine analogs; ANA975 (Anadys Pharmaceuticals) and related compounds; SM-360320 (Sumimoto); 3M-01, 3M-03, 3M-852, and 3M-S-34240 (3M Pharmaceuticals); GSK2245035 (GlaxoSmithKline; 8-oxoadenine molecule), AZD8848 (AstraZeneca; 8-oxoadenine molecule), MEDI9197 (Medimmune; formerly 3M-052), ssRNA40, and adenosine analogs, such as UC-1V150 (Jin et al., Bioorganic Medicinal Chem Lett (2006) 16:4559-4563, compound 4). Many TLR7 agonists are also TLR8 agonists.

The terms “TLR8” and “toll-like receptor 8” refer to any form of TLR8 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR8 activity. Unless otherwise specified, such as by way of a specific reference to human TLR8, TLR8 includes the naturally occurring sequence TLR8 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR8 is provided under UniProt entry number Q9NR97.

As used herein, “TLR8 agonist” means any molecule that, upon binding to TLR8, (1) stimulates or activates TLR8, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR8, or (3) enhances, increases, promotes, or induces the expression of TLR8. TLR8 agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acid ligands that bind to TLR8.

Examples of TLR8 agonists useful in the treatments, medicines, and uses of the present invention include recombinant single-stranded ssRNA, imiquimod (R837), gademod, requimod (R848), 3M-01, 3M-03, 3M-852, and 3M-S-34240 (3M Pharmaceuticals); GSK2245035 (GlaxoSmithKline; 8-oxoadenine molecule), AZD8848 (AstraZeneca; 8-oxoadenine molecule), MEDI9197 (Medimmune; formerly 3M-052), Poly-G10, Motolimod, and various TLR7 agonists provided above (as previously noted, many TLR7 agonists are also TLR8 agonists).

The terms “TLR9” and “toll-like receptor 9” refer to any form of TLR9 receptor, as well as variants, allotypes, and species homologs that retain at least a portion of TLR9 activity. Unless otherwise specified, such as by way of a specific reference to human TLR9, TLR9 includes the naturally occurring sequence TLR9 of all mammalian species, such as humans, monkeys, and mice. An exemplary human TLR9 is provided under UniProt entry number Q9NR96.

As used herein, “TLR9 agonist” means any molecule that, upon binding to TLR9, (1) stimulates or activates TLR9, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of TLR9, or (3) enhances, increases, promotes, or induces the expression of TLR9. TLR9 agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, nucleic acid ligands that bind to TLR9.

Examples of TLR9 agonists useful in the treatments, pharmaceuticals, and uses of the present invention include unmethylated CpG-containing DNA, immunostimulatory oligodeoxynucleotides (ODNs), such as CpG-containing ODNs, such as CpG24555, CpG10103, CpG7909 (PF-3512676/agatolimod), CpG1018, AZD1419, ODN2216, MGN1703, SD-101, 1018ISS, and CMP-001. TLR9 agonists also include a nucleotide sequence containing a synthetic cytosine-phosphate-2'-deoxy-7-dezoguanosine dinucleotide (CpR) (Hybridon, Inc.), dSLIM-30L1, and an immunoglobulin-DNA complex. Exemplary TLR9 agonists are disclosed in WO2003/015711, WO2004/016805, WO2009/022215, PCT/US95/01570, PCT/US97/19791 and U.S. Patent Nos. 8552165, 6194388 and 6239116, which are incorporated herein by reference for all purposes.

RLRs include various cytoplasmic PRRs that detect, for example, dsRNAs. Examples of RLRs include, for example, retinoic acid-inducible gene I (RIG-I), melanoma differentiation-associated gene 5 (MDA-5), and Laboratory of Genetics and Physiology 2 (LGP2).

As used herein, “RLR agonist” means any molecule that, upon binding to an RLR, (1) stimulates or activates the RLR, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of the RLR, or (3) enhances, increases, promotes, or induces the expression of the RLR. RLR agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acids that bind to RLRs and their derivatives, and agonistic monoclonal antibodies (mAbs) that specifically bind to RLRs.

Examples of RLRs agonists useful in the treatments, medicines, and uses of the present invention include, for example, short double-stranded RNAs having an uncapped 5' triphosphate (RIG-I agonists); poly I:C (MDA-5 agonists); and BO-112 (MDA-A agonists).

NLRs include various PRRs that detect molecules such as damage-related molecular patterns (DAMPs). NLRs include subfamilies NLRA-A, NLRB-B, NLRC-C, and NLRP-P. Examples of NLRs include, for example, NOD1, NOD2, NAIP, NLRC4, and NLRP3.

As used herein, “NLR agonist” means any molecule that, upon binding to an NLR, (1) stimulates or activates the NLR, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of the NLR, or (3) enhances, increases, promotes, or induces the expression of the NLR. NLR agonists useful in any of the treatments, medicines, and uses of this invention include, for example, DAMPs and their derivatives that bind to NLRs, and agonistic monoclonal antibodies (mAbs) that specifically bind to NLRs.

Examples of NLR agonists useful in the treatments, medicines, and uses of the present invention include, for example, liposomal muramyl tripeptide/mifamurtide (NOD2 agonist).

CLRs include various PRRs that detect, for example, carbohydrates and glycoproteins. CLRs include both transmembrane CLRs and secretory CLRs. Examples of CLRs include, for example, DEC-205/CD205, macrophage mannose receptor (MMR), Dectin-1, Dectin-2, mincle, DC-SIGN, DNGR-1, and mannose-binding lectin (MBL).

As used herein, “CLR agonist” means any molecule that, upon binding to a CLR, (1) stimulates or activates the CLR, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of the CLR, or (3) enhances, increases, promotes, or induces the expression of the CLR. CLR agonists useful in any of the treatments, medicines, and uses of the present invention include, for example, carbohydrates and their derivatives that bind to CLRs, and agonistic monoclonal antibodies (mAbs) that specifically bind to CLRs.

Examples of CLR agonists useful in the treatments, pharmaceuticals, and uses of the present invention include, for example, MD fraction (a purified soluble β-glucan extract from Grifola frondosa) and imprime PGG (a yeast-derived β1,3/1,6-glucan PAMP).

STING proteins function as both cytoplasmic DNA sensors and adaptors in type 1 interferon signaling. The terms "STING" and "stimulator of interferon genes" refer to any form of STING protein, as well as variants, allotypes, and species homologs that retain at least a portion of STING activity. Unless otherwise specified, such as by way of a specific reference to human STING, STING includes the naturally occurring sequence of STING in all mammalian species, e.g., humans, monkeys, and mice. An exemplary human TLR9 is provided under UniProt entry number Q86WV6. STING is also known as TMEM173.

As used herein, “STING agonist” means any molecule that, upon binding to TLR9, (1) stimulates or activates STING, (2) enhances, increases, promotes, induces, or prolongs the activity, function, or presence of STING, or (3) enhances, increases, promotes, or induces the expression of STING. STING agonists useful in any of the treatments, medicines, and uses of this invention include, for example, nucleic acid ligands that bind to STING.

Examples of STING agonists useful in the treatments, medicines, and uses of the present invention include various immunostimulatory nucleic acids, such as synthetic double-stranded DNA, cyclic di-GMP, cyclic GMP-AMP (cGAMP), synthetic cyclic dinucleotides (CDNs), such as MK-1454 and ADU-S100 (MIW815), and small molecules, such as PO-424.

Other PRRs include, for example, DNA-dependent Activator of IFN-regulatory Factors (DAI) and Absentin Melanoma 2 (AIM2).

Immunostimulatory cytokines include various signaling proteins that stimulate immune responses, such as interferons, interleukins, and hematopoietic growth factors.

Exemplary immunostimulatory cytokines include GM-CSF, G-CSF, IFN-α, IFN-γ; IL-2 (e.g., denileukin difitox), IL-6, IL-7, IL-11, IL-12, IL-15, IL-18, IL-21, and TNF-α.

Immunostimulatory cytokines can be in any suitable form. In some embodiments, the immunostimulatory cytokines can be recombinant forms of wild-type cytokines. In some embodiments, the immunostimulatory cytokines can be mutant proteins with one or more amino acid alterations compared to the corresponding wild-type cytokines. In some embodiments, the immunostimulatory cytokines can be incorporated into chimeric proteins containing cytokines and at least one other functional protein (e.g., antibodies). In some embodiments, the immunostimulatory cytokines can be covalently linked to a drug/pharmaceutical (e.g., any drug/pharmaceutical as described elsewhere herein as a possible ADC component).

Cancer vaccines include various compositions containing tumor-associated antigens (or that can be used to generate tumor-associated antigens in subjects) and thus can be used to induce an immune response in subjects against tumor cells containing tumor-associated antigens.

Examples of materials that may be included in a cancer vaccine include attenuated cancerous cells, tumor antigens, and antigen-presenting cells such as dendritic cells bombarded with nucleic acids encoding tumor-derived antigens or tumor-associated antigens. In some embodiments, the cancer vaccine may be prepared using the patient's own cancer cells. In some embodiments, the cancer vaccine may be prepared using biomaterials not derived from the patient's own cancer cells.

Cancer vaccines include, for example, sipuleucel-T and talimogene laherparepvec (T-VEC).

Immunotherapy involves treating patients with immune cells that can target cancer cells. Examples of immunotherapy include, for instance, tumor-infiltrating lymphocytes (TILs) and chimeric antigen receptor T cells (CAR-T cells).

Examples of chemotherapeutic agents include alkylating agents such as thiotepa and cyclophosphamide; alkyl sulfonates such as busulfan, dipropylamine sulfonate, and piperosulfan; aziridines such as benzodopa, carboquinone, meturedopa, and uredopa; ethyleneimines and methylamelamines, including hexamethylmelamine, tratamide, trietylenephosphoramide, ethyl thiophosphoric acid, and trimethylolomelamine; polyacetogenins (especially bullatacin and bullatacinone); camptothecin (including synthetic varieties). Analogs of topotecan; bryozoins; callystatin; CC-1065 (including its synthetic analogs of adolexin, carzelesin, and bizelesin); cryptophycins (especially cryptophycin 1 and cryptophycin 8); dolastatin; bacitracin (including synthetic analogs, KW-2189 and CBI-TMI); eleutherobin; pancratistatin; sarcodictyin; spongistatin; nitrogen mustard, such as chlorambucil, naphthalene, and chlorpyrifos. Cholophosphamide, estradiol, ifosfamide, nitrogen mustard, mechlorethamine, mechlorethamine oxide hydrochloride, melphalan, neonitrogen mustard, benzyl mustard cholesterol, pine oxychloride, tristyramine cyclophosphamide, uracil nitrogen mustard; nitrosoureas, such as carmustine, chloramphenicol, formazan, lomomia, nimustine, ranolamia; antibiotics, such as enediyne antibiotics (e.g., galicarmycin, especially galicarmycin γ1I and galicarmycin phiI1, see, e.g., Agnew, Chem. Intl. Ed. Engl., 33:183-186(19) 94); dynemicin, including dynemicin A; bisphosphonates, such as osteophosphine; esporamin; and new carcinogenic chromophores and related chromomophores, aclarubicin, actinomycin, autramycin, azoserine, bleomycin, cactinomycin C, carabicin, caminomycin, carcinomycin, chromomycin, daunorubicin, doxorubicin, 6-diazo-5-oxo-L-leucine, doxorubicin (including morpholino-doxorubicin, cyanomorpholino-doxorubicin, 2-pyrrolino-doxorubicin) (2-pyrrolino-doxorubicin and deoxydoxomyl), PEGylated liposomal doxorubicin, epirubicin, doxorubicin, idarubicin, mesorcinol, mitomycin such as mitomycin C, mycophenolic acid, nogaramycin, oligomycin, pemycin, potfiromycin, puromycin, triamcinolone acetonide, rodorubicin, streptomycin, streptozotocin, tuberculin, hydroxyphenylbutyryl leucine, neomycin, zorubicin; antimetabolites, such as methotrexate and 5-fluorouracil (5-FU); folic acid analogs, such as folate, methotrexate, pteroyltriglutamate, trimesat; purine analogs, such as fludarabine, 6-mercaptopurine, thioguanine, thioguanine; pyrimidine analogs, such as cyclocytidine, azathioprine, etc. Cytidine, 6-azouridine, carmoflurane, cytarabine, dideoxyuridine, docefluuridine, enoxabin, fluorouridine; androgens, such as carotestosterone, hydroxymethylandrostenone propionate, cyclothiosterol, methylcyclothiosterane, testrolide; anti-adrenergic drugs, such as aminoglutethimide, mitotane, and trilostertan; folic acid supplements, such as frolinic acid; aceglucuronolactone; aldophosphamide glycoside; aminolevulinic acid; eniluracil; acridine; bestrabucil; bethanycin; edatraxate; defofamine; dimethicone; diazinon; elfo-ornithine. rmithine); hydroxypicarazole acetate; epothilone; glycidyl ether; gallium nitrate; hydroxyurea; lentinan; chlordamine; maytansine alkaloids, such as maytansine and metansine; propiconazole; mitoxantrone; monpilamol; diammonium nitrate; pentostatin; benzylmethionine; pirarubicin; losoxantrone; podophyllin; 2-ethylhydrazide; methylbenzylhydrazine; propylimine; rhizomycin; sisofibril; spirogermanium; Alternaria spp. ketoacid; triazolam; 2,2',2''-trichlorotriethylamine; trichosporon toxins (especially T-2 toxin, verracurin A, baculosporin A, and anguidine); ursodeoxycholic acid; vinblastine; dacarbazin; mannitol; dibromomannitol; dibromodehydroxycervastatin Piperbromopropionate; gacytosine; cytarabine (“Ara-C”); cyclophosphamide; thiotepa; taxoids, such as paclitaxel and docetaxel; chlorambucil; gemcitabine; 6-thioguanine; mercaptopurine; methotrexate; platinum analogs, such as cisplatin and carboplatin; vincristine; platinum; etoposide (VP-16); ifosfamide; mitoxantrone; vincristine; vinblastine; nifedipine; teniposide; edaraxacum; daunorubicin; aminopterin; capecitabine; ibenphosphonate sodium; CPT-11; topoisomerase inhibitor RFS 2000; difluoromethylornithine (DMFO); retinoids, such as retinoic acid; capecitabine; and pharmaceutically acceptable salts, acids, or derivatives of any of the above. It also includes anti-hormonal agents used to regulate or inhibit the hormonal effects on tumors, such as anti-estrogens and selective estrogen receptor modulators (SERMs), including, for example, tamoxifen, raloxifene, droloxifene, 4-hydroxytamoxifen, treoxifene, naloxifene, LY117018, onasone, and toremifene (Fareston); aromatase inhibitors that inhibit aromatase, which regulate estrogen production in the adrenal glands, such as 4(5)-imidazole, aminoglutethimide, megestrol acetate, etc. Exemestane, formetastatin, fazodazole, vorazole, letrozole, and anastrozole; and anti-androgen substances such as flutamide, nilumid, bicalutamide, lipoprotein, and serotonin; KRAS inhibitors; MCT4 inhibitors; MAT2a inhibitors; tyrosine kinase inhibitors, such as sunitinib and axitinib; alk/c-Met/ROS inhibitors, such as crizotinib and lorlatinib. Intra-inib; mTOR inhibitors, such as temsirolimus and gedatolisib; src/abl inhibitors, such as bosutinib; cyclin-dependent kinase (CDK) inhibitors, such as palbociclib and PF-06873600; erb inhibitors, such as dacomitinib; PARP inhibitors, such as talazoparib; SMO inhibitors, such as glasdegib and PF-5274857; EGFR T790M inhibitors, such as PF-06747775; EZH2 inhibitors, such as PF-06821497; PRMT5 inhibitors, such as PF-06939999; TGFRβr1 inhibitors, such as PF-06952229; and pharmaceutically acceptable salts, acids, or derivatives of any of the above. In specific implementation schemes, this additional treatment agent is bevacizumab, cetuximab, sirolimus, panitumumab, 5-fluorouracil (5-FU), capecitabine, tevozanib, irinotecan, oxaliplatin, cisplatin, trifluorouridine, tilpyrimidine, leucovorin, gemcitabine, regorafinib, or erlotinib hydrochloride.

In some embodiments, the antibody is used in conjunction with one or more other therapeutic agents that target immune restriction point modulators or co-stimulators, such as, but not limited to, those targeting CTLA-4, LAG-3, B7-H3, B7-H4, B7-DC (PD-L2), B7-H5, B7-H6, B7-H8, B7-H2, B7-1, B7-2, ICOS, ICOS-L, TIGIT, CD2, CD47, CD80, CD86, CD48, C D58, CD226, CD155, CD112, LAIR1, 2B4, BTLA, CD160, TIM1, TIM-3, TIM4, VISTA(PD-H1), OX40, OX40L, GITR, GITRL, CD70, CD27, 4-1BB, 4-BBL, DR3, TL1A, CD40, CD40L, CD30, CD30L, LIGHT, HVEM, SLAM (SLAMF1, CD150 ), SLAMF2 (CD48), SLAMF3 (CD229), SLAMF4 (2B4, CD244), SLAMF5 (CD84), SLAMF6 (NTB-A), SLAMCF7 (CS 1), SLAMF8 (BLAME), SLAMF9 (CD2F), CD28, CEACAM1 (CD66a), CEACAM3, CEACAM4, CEACAM5, CEACAM6, CEAC AM7, CEACAM8, CEACAM1-3AS, CEACAM3C2, CEACAM1-15, PSG1-11, CEACAM1-4C1, CEACAM1-4S, CEACAM1-4L, IDO, TDO, CCR2, CD39–CD73–adenosine pathway (A2AR), BTKs, TIKs, CXCR2, CXCR4, CCR4, CCR8, CCR5, CSF-1, or agents that modulate the innate immune response.

In some implementations, the antibody is used in conjunction with, for example, the following: anti-CTLA-4 antagonist antibodies, such as ipilimumab; anti-LAG-3 antagonist antibodies, such as BMS-986016 and IMP701; anti-TIM-3 antagonist antibodies; anti-B7-H3 antagonist antibodies, such as MGA271; anti-VISTA antagonist antibodies; anti-TIGIT antagonist antibodies; antibodies; anti-CD80 antibodies; anti-CD86 antibodies; anti-B7-H4 antagonist antibodies; anti-ICOS agonist antibodies; anti-CD28 agonist antibodies; modulators of innate immune responses (e.g., TLRs, KIR, NKG2A); and IDO inhibitors.

In some embodiments, the antibody is used with an OX40 agonist, such as, for example, an anti-OX-40 agonist antibody. In some embodiments, the antibody is used with a GITR agonist, such as, for example, an anti-GITR agonist antibody, such as, but not limited to, TRX518. In some embodiments, the antibody is used with an IDO inhibitor. In some embodiments, the antibody is used with cytokine therapy (such as, for example, but not limited to, IL-15, CSF-1, MCSF-1, etc.).

In some implementations, the antibody is used in conjunction with one or more other therapeutic antibodies, such as, but not limited to, antibodies targeting CD19, CD22, CD40, CD52, or CCR4.

In some embodiments, the antibody composition comprises at least one additional pharmaceutical agent, such as bevacizumab, cetuximab, sirolimus, panitumumab, 5-fluorouracil (5-FU), capecitabine, tevozanib, irinotecan, oxaliplatin, cisplatin, trifluorouridine, tilpyridamole, leucovorin, gemcitabine, and erlotinib hydrochloride.

In some embodiments, the antibody therapy may be administered co-administered with other reagents or sequentially before or after other reagents at intervals ranging from minutes to weeks. In embodiments where other reagents and/or proteins or polynucleotides are administered alone, it will generally be ensured that a considerably long period of time does not expire between each delivery, such that the agent and the composition of the invention will still be able to produce a beneficial combined effect on the subject. In this case, it is anticipated that the two dosing regimens may be administered within approximately 12-24 hours of each other, and more preferably within approximately 6-12 hours of each other. However, in some cases, it may be desirable to significantly extend the time period for administration, in this regard, from several days (2, 3, 4, 5, 6, or 7) to several weeks (1, 2, 3, 4, 5, 6, 7, or 8) between each administration.

In some embodiments, the antibodies of the present invention are combined with treatment regimens that further include conventional therapies selected from: surgery, radiotherapy, chemotherapy, targeted therapy, immunotherapy, hormone therapy, angiogenesis inhibition, and palliative care.

The compositions used in this invention may further comprise pharmaceutically acceptable carriers, excipients, or stabilizers in the form of lyophilized formulations or aqueous solutions (Remington: The Science and Practice of Pharmacy, 21st edition, 2005, Lippincott Williams and Wilkins, Ed. K. E. Hoover). Acceptable carriers, excipients, or stabilizers are non-toxic to the recipient at the stated doses and concentrations and may include buffers such as phosphates, citrates, and other organic acids; antioxidants, including ascorbic acid and methionine; preservatives (e.g., octadecyl dimethyl benzyl ammonium chloride; hexane diammonium chloride; benzalkonium chloride, benzyl chloride; phenols, butanol, or benzyl alcohol; alkyl esters of p-hydroxybenzoate, such as methylparaben or propylparaben; catechol; resorcinol; cyclohexanol; 3-pentanol and m-cresol); low molecular weight (less than about 10 residues) polypeptides; proteins, such as blood... Albumin, gelatin, or immunoglobulins; hydrophilic polymers, such as polyvinylpyrrolidone; amino acids, such as glycine, glutamine, asparagine, histidine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates, including glucose, mannose, or dextran; chelating agents, such as EDTA; sugars, such as sucrose, mannitol, trehalose, or sorbitol; salt-forming ions, such as sodium; metal complexes (e.g., Zn-protein complexes); and/or nonionic surfactants, such as TWEEN™, PLURONICS™, or polyethylene glycol (PEG). Pharmaceutically acceptable excipients are further described herein.

In one aspect, the present invention provides a method for generating antibodies as disclosed herein, comprising: culturing host cells under conditions that result in the generation of antibodies as disclosed herein, and purifying the antibody or the bispecific antibody from the culture.

In another aspect, the present invention provides the use of antibodies, pharmaceutical compositions, polynucleotides, vectors or host cells as disclosed herein in the preparation of medicaments for the detection, diagnosis and/or treatment of tumor antigen-related conditions.

Diagnostic uses

In one aspect, for example for diagnostic purposes, the CD3 antibody of the present invention can be used to detect and/or measure CD3 or CD3-expressing cells in a sample. For example, the CD3 antibody can be used to diagnose conditions or diseases characterized by abnormal expression of CD3 (e.g., overexpression, underexpression, lack of expression, etc.). Exemplary diagnostic assays for CD3 may include, for example, contacting a sample obtained from a subject with the CD3 antibody of the present invention, wherein the CD3 antibody is labeled with a detectable marker or reporter molecule. Alternatively, an unlabeled CD3 antibody may be used in combination with a self-detectably labeled secondary antibody for diagnostic applications. The detectable marker or reporter molecule may be a radioisotope, such as ^14C , ^3H , ^125I , ^32P , or ^35S ; a fluorescent or chemiluminescent moiety, such as fluorescein isothiocyanate or rhodamine; or an enzyme, such as alkaline phosphatase, β-galactosidase, horseradish peroxidase, or luciferase. Specific exemplary assays that can be used to detect or measure CD3 in a sample include, but are not limited to, enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), fluorescence activated cell sorting (FACS), etc. Samples that can be used in the CD3 diagnostic assay according to the present invention include any tissue or body fluid sample obtainable from a patient, containing a detectable amount of CD3 protein or fragments thereof under normal or pathological conditions. Typically, the level of CD3 is measured in a specific sample obtained from a healthy patient (e.g., a patient without a disease or condition associated with abnormal CD3 levels or activity) to first establish a baseline or standard for CD3 levels. This baseline CD3 level can then be compared with the CD3 level measured in a sample obtained from an individual suspected of having a CD3-related disease or condition.

In another aspect, methods are provided for detecting, diagnosing, and/or monitoring conditions associated with tumor antigen expression. For example, multispecific antibodies as described herein can be labeled with detectable moieties (e.g., imaging agents and enzyme-substrate markers). Antibodies as described herein can also be used for in vivo diagnostic assays (e.g., in vivo imaging (e.g., PET or SPECT)) or staining reagents.

Reagent test kit

Another aspect of the invention is a kit comprising an antibody as disclosed above and instructions for use according to any method of the invention described herein. Typically, these instructions include a description of the administration of the antibody for the therapeutic treatment described above. The kit includes any pharmaceutical compositions disclosed herein. Pharmaceutical compositions and other reagents may be present in the kit in any convenient form, such as, for example, in solution or powder form.

In another aspect, the kit further includes one or more other preventative or therapeutic agents that can be used to treat cancer, contained in one or more containers. In one embodiment, said other preventative or therapeutic agent is a chemotherapeutic agent. In other aspects, said preventative or therapeutic agent is a biological or hormonal therapeutic agent.

Several aspects of the pharmaceutical compositions, preventive agents, or therapeutic agents of the present invention are preferably tested for desired therapeutic activity in vitro, in cell culture systems, and in animal model organisms (e.g., rodent model systems) before human use.

The toxicity and efficacy of the preventive and/or therapeutic regimens of the present invention can be determined by standard pharmaceutical procedures in cell cultures or laboratory animals, for example, for determining the _LD50 (the dose that is lethal to 50% of the population) and _ED50 (the dose that is therapeutically effective in 50% of the population). The dose ratio between toxicity and therapeutic effect is the therapeutic index, and it can be expressed as the ratio _LD50 / _ED50 . Prophylactic and/or therapeutic agents exhibiting a large therapeutic index are preferred.

Furthermore, any assays known to those skilled in the art can be used to evaluate the preventive and/or therapeutic efficacy of the therapies or combinations of therapies disclosed herein for the treatment or prevention of cancer.

Instructions for use with bispecific antibodies as described herein typically include information about the dosage, dosing regimen, and route of administration for the intended treatment. Containers may be unit doses, bulk packaging (e.g., multi-dose packs), or subunit doses. Instructions supplied with kits of the present invention are typically written instructions on a label or packaging insert (e.g., paper included in the kit), but machine-readable instructions (e.g., instructions carried on a disk or optical storage disc) are also acceptable.

The kit of the present invention is in suitable packaging. Suitable packaging includes, but is not limited to, vials, ampoules, tubes, bottles, wide-mouth bottles, flexible packaging (e.g., sealed polyester film or plastic bags), etc., for each pharmaceutical composition and other included reagents (e.g., buffers, balanced saline solutions, etc., for administering the pharmaceutical composition to a subject). Packaging for use in combination with specific devices such as inhalers, nasal delivery devices (e.g., nebulizers), or infusion devices such as micropumps is also contemplated. The kit may have a sterile inlet (e.g., the container may be an intravenous solution bag or vial with a stopper that can be punctured by a hypodermic needle). The container may also have a sterile inlet (e.g., the container may be an intravenous solution bag or vial with a stopper that can be punctured by a hypodermic needle). At least one active agent in the composition is an antibody. The container may further contain a second pharmaceutical active agent.

Typically, a kit includes a container and a label or packaging insert on or associated with the container.

Biological Preservation

Representative material of the present invention was deposited on February 13, 2018 at the American Center for Type Culture Collection (ATCC), 10801 University Boulevard, Manassas, Va. 20110-2209, USA. The vector GUCY2C-1608 chain A (VH-Synthesis SEQ ID NO: 94) with ATCC accession number PTA-124943 contains a DNA insert of the VH-Synthesis chain encoding the bispecific antibody GUCY2C-1608, which contains a polynucleotide (SEQ ID NO: 70) encoding the CD3 antibody light chain variable region (SEQ ID NO: 3); and the GUCY2C-1608 chain B (VL-Synthesis SEQ ID NO: 93) with ATCC accession number PTA-124944 contains a DNA insert of the VL-Synthesis chain encoding the bispecific antibody GUCY2C-1608, which contains a polynucleotide (SEQ ID NO: 71) encoding the CD3 antibody heavy chain variable region (SEQ ID NO: 4).

The deposit was made in accordance with the provisions of the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the Purpose of Patent Procedure and Regulations thereunder. This ensures that the live culture of the deposit remains intact for 30 years from the date of deposit. The deposit is available through the ATCC under the terms of the Budapest Treaty and complies with an agreement between Pfizer, Inc. and the ATCC that ensures the perpetual and unrestricted availability of the descendants of the deposited culture to the public at the time of publication of the relevant U.S. Patent or at the time of public disclosure of any U.S. or foreign patent application (whichever comes first), and ensures the availability of the descendants to persons authorized by the Director of the U.S. Patent and Trademark Office in accordance with 35 U.S.C. Part 122 and the Director's Rules established thereunder (including 37 C.F.R. Part 1.14, which specifically refers to 886 OG 638).

The assignee of this application has agreed that, should the culture of the deposited material die, be lost, or be destroyed under suitable conditions, the material will be promptly replaced with another identical copy upon notification. The availability of the deposited material shall not be construed as a violation of any government authority's right to practice the invention under its patent law.

The following embodiments for implementing specific aspects of the invention are provided for illustrative purposes only and are not intended to limit the scope of the invention in any way.

Example 1

Formation of soluble CD3ed

A protein called CD3ed, representing the extracellular domains of CD3 ε and δ subunits tandemly fused via a short linker (SEQ ID NO: 39), was generated using standard molecular biology, protein expression, and purification techniques. Using PEI as the transfection agent, plasmids carrying the gene encoding the amino acids shown in Table 9 (SEQ ID NO: 40) were transiently transfected into 20 μL of HEK293 cells. After incubation at 37°C for 7 days, 20 μL of the harvest was fractionally bound to 50 mL of His60 resin (equilibrated with PBS) overnight. The bound resin was then washed with PBS, followed by washing with PBS buffer (10⁻¹⁵ CV) containing 10 mM imidazole. Following washing, the protein was eluted with phosphate-buffered saline containing 200 mM imidazole. The eluent fractions were then analyzed by SDS gel electrophoresis. Fractions containing purified protein were pooled and concentrated using a 3 kDa Amicon ultra centricon. During concentration, 5% glycerol was maintained in the sample to stabilize the protein. The His60 pool was further loaded onto a size-restricted Superdex 200 column. Fractions containing CD3 ε/δ heterodimers were analyzed by SDS-PAGE, and the relevant fractions were pooled and filtered at 0.22 µm. Table 9 shows the sequences of the CD3 ε/δ subunits tandemly fused and used as antigens in all optimization efforts.

Table 9

Example 2

Optimization of the structure of CD3 antibody H2B4

The lead CD3e (CD3 ε) antibody H2B4 exhibited low thermostability when used in bispecific biantibody-Fc form. To improve stability, the light chain architecture was changed from VK4 to VK1 (DPK9), while the heavy chain architecture was retained as VH3 (DP54). VLCDR1 (SEQ ID NO: 26), VL CDR2 (SEQ ID NO: 27), and VL CDR3 (SEQ ID NO: 28) were transplanted into the DPK9 architecture and expressed in combination with heavy chains carrying CDRs (i.e., VH CDR1 (SEQ ID NO: 13), VHCDR2 (SEQ ID NO: 16), and VH CDR3 (SEQ ID NO: 18)). In addition to transplanting CDRs into the architecture, certain reversion mutations were tested. The reversion mutations introduced in both the heavy and light chain structures are shown in Table 10, where the residue number indicating the reversion mutation is indicated.

A cDNA containing the human receptor framework (VH3 for the heavy chain and VK1 for the light chain) and the associated CDR donor sequence was synthesized. The synthesized cDNA product was subcloned and fused to the human IgG1 constant region of the heavy chain and the human κ-frame of the light chain in a mammalian expression vector. To distinguish them from the H2B4 antibody, these variants will be referred to as 2B5 below, and all optimized molecules will be referred to as 2B5 and its variants. The competitive binding of all 2B5 mutants to Jurkat cells with endogenous CD3e expression was analyzed.

Competitive FACS was performed to examine whether humanized 2B5 maintained its binding epitope on the cell surface. Prior to the competitive FACS assay, chimeric cH2B4 was biotinylated. The EC50 of the binding of biotinylated cH2B4 to Jurkat cells was determined by direct binding FACS and was measured to be 0.8 nM. For competitive FACS, a 3-fold serially diluted antibody variant was mixed with a constant amount of biotin-cH2B4 representing the EC50 concentration and 1.0+E05 cells/well of Jurkat cells. The mixture was incubated on ice for 1 hour, followed by three washes with FACS buffer. A secondary antibody (i.e., PE (phycoerythrin)-conjugated streptavidin (PE-SA)) was added to the cells and incubated on ice for 30 minutes. Median fluorescence intensity (MFI) was then analyzed by FACS on FACS CANTO™ (BDBiosciences). Y-axis calculation: %Maximum binding = [MFI (serially diluted Ab wells) - MFI (secondary PE-SA)] / [MFI maximum (Jurkat cells with biotin-cH2B4 only) - MFI (secondary PE-SA)]. Table 10 shows the results of the _IC50 values for competitive binding when mutations were introduced into CD3 antibody H2B4 to optimize the scaffold. Based on these results, H2B4 1.0/1.0T showed similar binding characteristics to H2B4 and was designated as the CD3 antibody binding domain 2B5.

Table 10

Antibody	Heavy chain	Light chain	<![CDATA[Competitive combination of IC<sub>50</sub>]]>
				h2B4 1.0/1.0T (2B5)	CDR grafts	CDR graft, R24T	1.27
h2B4 1.0/1.1	vH1.0	Q13V, R24T, Y49S, L78V	1.68
				h2B4 1.0/1.2	vH1.0	Q13V, R24T, Y49S	Untested
h2B4 1.0/1.3	vH1.0	R24T, Y49S	1.309
				c2B4 (Chimeric 2B4)	mouse vH	mouse vK	0.752

Example 3

Bispecific biantibody-Fc molecules were generated and characterized using H2B4 and 2B5 sequences.

Bispecific biantibody-Fc molecules were generated using standard expression and purification techniques known in the art, using CD3 antibody 2B5 or H2B4 (Tables 2 and 3) and a GUCY2c antibody sequence in one of the configurations shown in Figure 1 (schematic diagram). The stability of the resulting bispecific antibodies (Table 7; SEQ.ID.NO.36-38) was tested using differential scanning calorimetry.

Example 4

The stability of bispecific antibodies was evaluated using differential scanning calorimetry.

The protein was diluted to 0.6 mg/mL in 400 µL of phosphate-buffered saline (PBS). PBS was used as a buffer blank in the reference chamber. PBS contained 137 mM NaCl, 2.7 mM KCl, 8.1 mM Na₂HPO₄, and 1.47 mM KH₂PO₄ at pH 7.2. Samples were dispensed into the sample trays of a MicroCal VP-Capillary DSC using an autosampler (Malvern Instruments Ltd, Malvern, UK). Samples were equilibrated at 10 °C for 5 min and then scanned at 100 °C per hour up to 110 °C. A 16-second filtration period was selected. The raw data were baseline corrected and the protein concentration was normalized. The data were fitted to the MN2-State model using Origin software 7.0 (OriginLab Corporation, Northampton, MA) with an appropriate number of transformations. As illustrated in Figure 2, the bispecific GUCY2c-2B5 exhibits an improved first melt transition at 5°C compared to the bispecific GUCY2c-H2B4. This indicates that bispecific antibodies containing 2B5 demonstrate better stability.

Using FACS, the binding of two bispecific antibodies to T cells was examined using standard procedures known in the art, and as shown in Figure 3, both bispecific antibodies showed similar binding to T cells, indicating that the binding was retained after H2B4 was converted to 2B5.

Both H2B4 and 2B5 bispecific antibodies were tested in a cytotoxicity assay to evaluate T cell retargeting activity. As shown in Figure 4, both bispecific antibodies showed similar cytotoxicity, indicating that 2B5 retained the same activity as H2B4.

Example 5

2B5-based structure-based rational mutation

The CD3 antibody H2B4 binding domain exhibits multireactivity (e.g., as evaluated by binding to DNA and insulin), shows a tendency for self-association (e.g., indicated by a high AC-SINS score), and exhibits a tendency for cleavage at the R53 position in CDR-H2 when expressed in CHO host cells. To reduce multireactivity, the tendency for self-association and cleavage in CDRH2, and mutations in the CDR of 2B5, were engineered to reduce these adverse factors while maintaining binding affinity and stability. For this effort, all predictions were made using the CD3 antibody H2B4 X-ray crystal structure, obtained in a complex with a peptide representing the binding epitope and derived from the N-terminus of the CD3 ε subunit. It is well known in the art that excess positive charge and/or hydrophobicity can play a major role in multireactivity. The electrostatic surface of the H2B4 Fv region from the crystal structure was determined using the DelPhi Poisson Boltzmann calculator in Discovery Studio 4.0 ^1. In Figure 5, an excess positive charge patch that is not disrupted by negative charge is observed. This indicates a correlation with high multi-reactivity. Additionally, the Spatial Aggregation Tendency (SAP) tool in Discovery Studio ¹ was used to identify excess hydrophobic patches (Figure 6). As shown in Figure 6, the hydrophobic patches appear small and insignificant. Based on this observation, the focus is on reducing the positive charge by removing positive charges or adding negative charges to disrupt the patches.

To determine the appropriate sites for mutagenesis, computational predictions of mutations for both affinity and stability changes were performed using Discovery Studio and FoldX. Tolerance mutations were those for which neither Discovery Studio nor FoldX methods predicted ΔΔG > 1 kcal/mol. From these predictions, a set of residues predicted to have minimal impact on binding and stability while still disrupting the altered patch were identified. Additionally, we favored hydrophilic mutations that did not increase hydrophobic surface area where possible. A list of possible CDR mutations introduced in CD3 antibody 2B5 to optimize biophysical properties is shown in Table 11. This includes mutations predicted to be tolerant at each site and reduce multireactivity and splicing.

Table 11

In addition, a set of multiple mutations (combinations) were designed for both the heavy and light chains. These include R52Q/R52bQ/R53Q, R52Q/R52bQ/R53Q/R96Q, R52bQ/R53Q/R96Q, and R52Q/R52bQ/R96Q for the heavy chain, and R29Q/K30Q/R54L/V27eE, R29Q_R54L_V27eE, and R29Q_K30Q_R54L for the light chain.

Example 6

Characterization of mutants

All mutants were generated as IgG molecules using standard expression and purification techniques well-known in the art, and their binding to CD3ed, multireactivity (binding to DNA and insulin), and AC-SINS were tested.

Combined assays to evaluate variants

Octet assays were designed and run to examine both the binding and dissociation rates of 10 μg/ml soluble human CD3ed antigen (SEQ ID NO: 40). A 10 μg/ml anti-CD3 antibody was captured using an anti-human IgG tip for 120 s, followed by a baseline in PBS for 30 s, then immersed in 10 μg/ml soluble human CD3ed antigen for 120 s, and finally dissociated in it for 300 s. Sensing patterns were examined to determine whether the antigen bound to the antibody, and Octet software was used to calculate kd(1/s) and kd error. Twenty-six viable mutations retained binding to the hCD3ed antigen. The binding kinetics of the 26 remaining antigen-binding CD3 antibody variants to soluble CD3 were then examined using Biacore analysis via SPR. CD3 antibodies were captured on anti-human Fc cells, and soluble CD3 antigen at concentrations of 0 nM, 3.7 nM, 11.1 nM, 33.3 nM, and 100 nM was injected onto the captured antibodies to determine binding kinetics. Twenty-three CD3 antibodies bound the soluble CD3 antigen with an affinity similar to that of the parental antibodies used in CDR transplantation.

DNA and insulin ELISA to measure multireactivity

384-well ELISA plates (Nunc Maxisorp) were coated overnight at 4°C with DNA (10 µg/ml) (Sigma-Aldrich, D1626) and insulin (5 µg/ml) (Sigma-Aldrich, I9278-5 mL) in PBS pH 7.5. The assay, modified from that described in Tiller et al. (17), was performed on a PerkinElmer Janus Automated Workstation liquid handling robot. Wells were washed with water, blocked for 1 hour at room temperature with 50 µl of multireactive ELISA buffer (PEB; PBS containing 0.05% Tween-20, 1 mM EDTA), and rinsed three times with water. Quadruple copies of serially diluted mAbs in 25 µl were added to the wells and incubated at room temperature for 1 hour. The plates were washed three times with water, and 25 µl of 10 ng/ml goat anti-human IgG (Face fragment specific) conjugated with horseradish peroxidase (Jackson ImmunoResearch, 109-035-008) was added to each well. The plates were incubated at room temperature for 1 hour, washed three times with 80 µl of water, and 25 µl of TMB substrate (Sigma Aldrich, T-0440) was added to each well. After approximately 7 minutes, the reaction was terminated by adding 25 µl of 0.18 M orthophosphate to each well, and the absorbance was read at 450 nm. The DNA and insulin binding score was calculated as the ratio of the ELISA signal of the 10 µg/ml antibody to the signal of the well containing buffer instead of the primary antibody. A score less than 8 was considered ideal and represented low multireactivity.

AC-SINS assay to measure self-association tendency

AC-SINS (Affinity Capture Self-interaction Nanoparticle Spectroscopy) assays were normalized in 384-well format on a Perkin-Elmer Janus liquid handling robot. 20 nm gold nanoparticles (Ted Pella, Inc., #15705) were coated with a mixture of 80% goat anti-human Fc (Jackson Immuno Research Laboratories, Inc. #109-005-098) and 20% nonspecific goat polyclonal antibody (Jackson Immuno Research Laboratories, Inc. #005-000-003), which was replaced with buffer at 20 mM sodium acetate pH 4.3 and diluted to 0.4 mg/mL. After incubation at room temperature for one hour, unoccupied sites on the gold nanoparticles were blocked with thiolized polyethylene glycol (2 kDa). The coated nanoparticles were then concentrated 10-fold using a syringe filter, and 10 µl was added to 100 µl of 0.05 mg/ml mAb in PBS pH 7.2. The coated nanoparticles and target antibody were incubated in 96-well polypropylene plates for 2 hours, and then transferred to 384-well polystyrene plates and read on a Tecan M1000 spectrophotometer. Absorbance was read from 450–650 nm in 2 nm increments, and a Microsoft Excel macro was used to identify the maximum absorbance, smooth the data, and fit the data using a second-order polynomial. The antibody AC-SINS score was determined by subtracting the smoothed maximum absorbance of the mean blank (PBS buffer only) from the smoothed maximum absorbance of the antibody sample. A score less than 8 was considered ideal and represented low self-association.

Evaluation of the lead CD3 antibody 2B5 variant in a bispecific Fc form

A comprehensive analysis of data from binding assays, AC-SINS, and multireactivity helped narrow down the binding variants to five. These five variants were reformatted into bispecific biantibody-Fc molecules using antitumor sequences and standard cloning strategies well-known in the art. These bispecific antibodies were transiently expressed and purified using standard techniques well-known in the art. Binding of the resulting bispecific antibodies to jurakat cells was tested, in vitro cytotoxicity was tested using a T-cell retargeting assay, multireactivity was tested, and thermostability was tested using DSC and AC-SINS. Of all five bispecific antibodies, only one, carrying the CD3 antibody 2B5 variant 2B5v6, showed potent binding and therefore potent cytotoxicity. This variant also showed reduced multireactivity, as evaluated by DNA and insulin binding and AC-SINS scores. The 2B5v6 bispecific antibody scored less than 8, compared to a score greater than 8 for the H2B4 bispecific antibody. Based on these results, 2B5v6 was selected as the lead optimized 2B5 variant for further analysis.

Example 7

Evaluation of shear in CDR-H2

When the bispecific antibodies were expressed in CHO cells but not in HEK293 cells, cleavage was observed in the VH CDR2 (SEQ ID NO: 16) at position R53 in the CD3 antibodies (H2B4 and 2B5). To mitigate the risk of this adverse effect and ensure better homogeneity of the product at manufacturing scale, care should be taken to select for mutations in the cleavage region, as described in previous examples, as well as other mutations to mitigate the risk of other adverse effects. To evaluate the cleavage, simply, bispecific antibody-Fc bispecific antibodies were generated using the variable domain sequence of the negative control antibody and CD3 antibody 2B5 or CD3 antibody H2B4 or CD3 antibody 2B5v6. All bispecific antibody-Fc molecules were prepared using a similar schematic diagram of expression and purification processes well known in the art. CHOSSI cell lines were generated using standard procedures. Two strands encoding a bispecific antibody were cloned into the mammalian expression vector pRY19-GA-Q, which contains a dual promoter and a recombinant site for unit site integration (SSI) (Zhang, L. et al. Biotechnology Progress. 2015; 31:1645-1656) into the CHO cell genome. The resulting plasmid, containing the target gene, was transfected into 10e9 CHO-K1 SV cells via electroporation using the pRY19 vector, followed by positive and negative selection pressures using hygromycin-B and ganciclovir. These CHO conjugates underwent a 3-week recovery period. When observed viability exceeded 90%, a cell library was generated for future work. The established CHO conjugates were then subjected to a 12-day fed-batch platform expression process in 1L of CD-CHO medium (Thermo Fisher Scientific, Waltham, MA). The day 12 culture harvest was captured on a Protein A column and eluted at low pH. The proA conjugate was immediately neutralized to pH 8.1. The purified protein was evaluated using capillary gel electrophoresis (cGE) according to standard protocols well-known in the art. As shown in Table 12 and Figure 7, the 2B5 bispecific antibody showed significantly reduced shearing compared to H2B4, and 2B5v6 showed no shearing while H2B4 showed significantly increased shearing or fragmentation. Table 12 shows the results of capillary gel electrophoresis demonstrating fragmentation of the CD3-GUCY2c bispecific antibody. POI% indicates the target peak.

Table 12

Reduced cGE sample	%POI	%Fragmentation
			Negative control - H2B4	88.8	11.2
Negative control - 2B5	98.9	1.1
			Negative control - 2B5v6	99.6	0.4

Example 8

GUCY2c-2B5v6 Bispecific Antibody: Expression and Purification of Chimeric CD3-GUCY2c Bispecific Antibody Using CD3 Antibody 2B5v6

Complementary construct pairs (12.5 µg per strand) were co-transfected into 25 mL of logarithmic-phase culture containing 1 million cells/mL HEK293 cells using the ExpiFectamine™ 293 Transfection Kit (Life Technologies). Twenty-four hours post-transfection, ExpiFectamine Transfection Enhancer was added, and cells were allowed to grow for an additional 4–5 days before harvest. The used culture was then collected, centrifuged to remove cell debris, and passed through a 20 µm filter.

The clarified conditioned medium containing the GUCY2c T cell bispecific antibody was then purified using protein A affinity chromatography. Samples were loaded onto 0.45 mL microcolumns pre-loaded with MabSelect SuRe protein A resin (GE Healthcare) using a liquid delivery device (Tecan). The bound protein was washed with PBS pH 7.2, eluted with 20 mM citric acid and 150 mM sodium chloride pH 3.5, and neutralized with 2 M tris at pH 8.0. The sample was then desalted to PBS pH 7.2 using G25 Sephadex drip columns (GE Healthcare) according to the manufacturer's instructions. The purity of the purified protein was analyzed on an Aglient 1200 HPLC system using analytical size exclusion chromatography (TOSOH) with a Mab HTP column, following the manufacturer's protocol. Concentrations were determined using a microspectrophotometer (Trinean) at OD 280 nm.

The stability of bispecific antibodies was evaluated using differential scanning calorimetry.

The protein was diluted to 0.6 mg/mL in phosphate-buffered saline (PBS) solution at a volume of 400 µL. PBS was used as a buffer blank in the reference chamber. PBS contained 137 mM NaCl, 2.7 mM KCl, 8.1 mM Na₂HPO₄, and 1.47 mM KH₂PO₄ at pH 7.2. Samples were dispensed into the sample trays of a MicroCal VP-Capillary DSC using an autosampler (Malvern Instruments Ltd, Malvern, UK). Samples were equilibrated at 10 °C for 5 min and then scanned at 100 °C per hour up to 110 °C. A 16-second filtration period was selected. The raw data were baseline corrected and the protein concentration was normalized. The data were fitted to the MN2-State model using Origin software 7.0 (OriginLab Corporation, Northampton, MA) with an appropriate number of transformations. As shown in Figure 8, the GUCY2c-2B5v6 bispecific antibody exhibits excellent thermal stability, with _Tm1 being 70℃.

Binding with naïve T cells

Using a BD LSRII Fortessa analyzer, purified CD3 bispecific antibody was titrated to bind to the cell surface of naïve human CD3 on virgin human T cells purified from fresh human peripheral blood mononuclear cells (PBMCs) using standard flow cytometry methods known in the art. As shown in Figure 9, the GUCY2c-2B5v6 bispecific antibody bound very well to T cells, with a binding EC50 of 35.44 nM.

Evaluation of T cell activity mediated by bispecificity

Human PBMCs were isolated from healthy donor blood using Histopaque-177 (Sigma). Naïve T cells (negatively selected T cells) were isolated from PBMCs using a T cell enrichment kit from Stem Cell Technologies. T84 human tumor cells expressing GUCY2c, transfected with a luciferase expression construct, were treated with serially diluted GUCY2c bispecific antibody or a negative control CD3 bispecific antibody, along with the T cells. The effector cell to target cell ratio (E:T) was 10:1 or 5:1. T cells and tumor cells were incubated at 37°C for 48 hours, followed by measurement of luciferase signal using neolite reagent (Perkin Elmer). EC50 values were calculated in Graphpad PRISM using a four-parameter nonlinear regression analysis. Target-negative HCT116 or HT29 cells did not show any GUCY2c bispecificity-mediated cytotoxicity. As shown in Figure 10, GUCY2c-2B5v6 bispecificity demonstrated potent cytotoxicity in the T cell retargeting assay.

T84 cells were treated with T cells (E:T ratio 5:1) and different doses of GUCY2c-1608 (GUCY2c-2B5v6). Supernatants were collected at different time points (24 h, 48 h, 96 h) and analyzed using multiplex Luminex assays according to the manufacturer's guidelines and read on a Luminex 200 with xPONENT software. The measured cytokines were human IFN-γ, IL10, IL2, IL4, IL6, and TNF-α. Figure 11 shows the upregulation of (11A) IFN-γ, (11B) IL10, (11C) IL2, (11D) IL4, (11E) IL6, and (11F) TNF-α cytokines after T cell activation with GUCY2c-1608.

In vivo assessment of bispecific activity mediated by GUCY2c-2B5v6

In vivo efficacy studies were conducted using an adoptive metastasis model. Tumor cell lines/patient-derived xenograft fragments were implanted into NSG mice and staged to approximately 200 ^mm³ . The bispecific compound was administered intravenously to the mice (unless otherwise noted). Twenty-four hours after bispecific antibody administration, two million activated and expanded T cells were administered intravenously (using a T cell expansion and activation kit from Miltenyi Biotec). The bispecific antibody was administered weekly. As shown in Figures 12 and 13, the GUCY2c-2B5v6 (GUCY2c-1608) bispecific antibody demonstrated excellent durable tumor regression in both the PDX CRX-11201 and LS1034 cell lines, representing colorectal cancer, in vivo models.

Example 9

Generation and cytotoxicity of FLT3-CD3 bispecific antibody and FLT3-2B5v6 bispecific IgG using an optimized CD3 antibody (2B5v6).

Human FLT3 antibody (xFLT3) and human CD3 antibody (2B5v6) are expressed as engineered human IgG2dA_D265A with EEEE on one arm and RRRR on the other arm, respectively, for bispecific exchange at positions 223, 225, and 228 (e.g., (C223E or C223R), (E225E or E225R), and (P228E or P228R)) in the hinge region of human IgG2 and at positions 409 or 368 (e.g., K409R or L368E (EU numbering scheme)) in the CH3 region. The antibodies also have a mutation from D to A at position 265 (EU numbering scheme).

Individual antibody arms were purified separately using Protein A resin. Then, 1 mg/mL of purified xFLT3 antibody was exchanged with 1 mg/mL of purified 2B5v6 antibody in the presence of 2 mM reduced glutathione. The exchange reaction was carried out at 37°C for 16 h. Then, 2 mM oxidized glutathione was added to the protein, and the mixture was incubated at 37°C for 30 min. The mixture was then diluted 1:5 to 20 mM MES pH 5.4 buffer without NaCl. The diluted sample was loaded onto a MonoS ion exchange column and washed with 20 mM MES pH 5.4 + 25 mM NaCl buffer. The bispecific antibody xFLT3-2B5v6 was then eluted with a 0–500 mM NaCl gradient in 20 mM MES pH 5.4 buffer. Finally, the bispecific antibody was dialyzed into phosphate-buffered saline (PBS) for cell-based assays. For the cytotoxicity assay, unstimulated human T cells and luciferase-expressing target cells (Eol-1 or MV411) were co-incubated in 96-well U-shaped plates in 200 μl RPMI + 10% FBS to determine the ratio. xFLT3-2B5v6 bispecific antibody was serially diluted in PBS and added to each well. After 24 hours, 100 μl of cell suspension was mixed with 100 μl of One-GLO reagent in 96-well white-walled plates, and luminescence was measured on an Envision photometer. The cytotoxicity value was determined using the following formula:

%Cytotoxicity = [1 - (RLU sample) / RLU target cells only)] × 100RLU: relative light units

As shown in Figures 14A and 14B, when tested on two cell lines expressing higher copies of FLT3 (EOL-1) and lower copies (MV4-11), the xFLT3-2B5v6 bispecific antibody containing the optimized CD3 antibody (2B5v6) appears to be highly potent when presented as an alternative to the bispecific IgG form.

Table 13 shows the light and heavy chain sequences of FLT3 (EOL-1) and the light and heavy chain sequences of the 2B5v6 bispecific antibody.

Table 13

Example 10

T-cell epitopes of the 2B5v6 bispecific antibody

Optimized analysis of potential T-cell epitopes in clone 2B5v6 was performed to de-risk the possibility of immunogenicity. This analysis was performed using the Epivax tool for predicting potential T-cell epitopes on a computer chip. This tool predicts the number of times a 9-mer framework peptide binds to any of eight different HLA alleles. Hit was defined as those in the top 5% with a Z-score in the top 1%. Hit to four or more alleles or one strong hit was considered a predicted T-cell epitope. Based on these criteria, three predicted non-germal T-cell epitopes in CDRH2, two predicted germal epitopes in CDRL1, and one predicted germal epitope in L2 were identified (Table 15). In addition to the scores of individual peptides, the overall risk of the sequence was defined by the total number of hits or the T-regitope-adjusted Epivax score, which had been optimized to correlate with the prevalence of clinical ADA. Here, for these predicted values, the lower the score, the lower the expected risk of ADA. The VH domain had 70 hits and a score of -42.82, while the VL domain had 60 hits and a score of -23.79.

To reduce the risk of immunogenicity, mutations that would decrease the overall score of this clone while removing hits and ideally predicted T-cell epitopes were identified. Any possible single mutant variations of CDR-L1, CDR-L2, and CDR-H2 were considered to identify mutations that would reduce hits, decrease the overall t-regitope-adjusted EpiMatrix score, and remove T-cell epitopes. Since these potential mutations could also affect binding and stability, care was taken to consider the predicted stability and affinity data described above, as well as to identify tolerant mutations. Tables 14A, 14B, and 14C show mutations designed to reduce the immunogenicity risk of 2B5v6. Additionally, mutations predicted as shown in SEQ ID NOs 44-47, 51-57, 60, and 62 reduced the number of T-cell epitopes.

Table 14A

Table 14B

Table 14C

Table 15 shows the regions of the variable domains of a CD3 antibody (2B5v6) that has T-cell epitopes and poses a potential risk of immunogenicity.

Table 15

All mutants were generated as monovalent IgG antibodies using standard transient HEK293 expression, followed by purification using a protein A column. The binding of the resulting clones to CD3-expressing Jurkat cells was tested using FACS, and the likelihood of self-association was tested using AC-SINS as described in Example 7 above. As shown in Table 16, five molecules with improved epivax scores and reduced T-cell epitopes, while retaining affinity for CD3, were selected for further testing as biantibody-Fc variants. The 2B5v6 variant with improved epivax scores was compared to the parental 2B5v6. Results showed comparable binding to CD3-expressing Jurkat cells and AC-SINS scores.

Table 16

Of the mutants tested, several exhibited lower binding to Jurkat cells compared to 2B5v6. Among these mutants, those with all T-cell epitopes eliminated and improved epivax scores were further advanced to generate bispecific antibody-Fc molecules in which the GUCY2c antibody serves as the other binding domain. Table 17 shows the epivax scores, and Figure 15 shows the reduced binding of these mutants to Jurkat cells. The low-affinity CD3 antibody 2B5v6 variant showed an improved epivax score.

Table 17

CD3 antibody variants	Epivax Score (tReg - Adjusted)
		2B5v6-0598	-65.15
2B5v6-0623	-56.74

Example 11

Three anti-CD3 variants showed improved affinity and cytotoxicity in the GUCY2C-CD3 bispecific antibody.

Using the standard cloning, expression, and purification techniques described above, three anti-CD3 variants derived from 2B5v6, 2B5-1038 (SEQ ID NO: 4 and 9), 2B5-1039 (SEQ ID NO: 4 and 87), and 2B5-1040 (SEQ ID NO: 4 and 89), were reformulated into bispecific biantibody-Fc molecules paired with the antitumor GUCY2c antibody sequence, one of the configurations shown in Figure 1. In vitro cytotoxicity of the resulting bispecific antibodies was assessed using T-cell retargeting assays as shown in Table 18, binding affinity was tested using surface plasmon resonance (SPR), and nonspecificity was tested using AC-SINs. Immunogenic risk was evaluated using the Epivax Tool, a computer-based assay for potential T-cell epitopes. Complete mass spectrometry data indicated that all bispecific antibodies were correctly paired and 100% heterodimer. Table 18 shows that all three bispecific antibodies (1) were approximately 2-fold more potent in in vitro cytotoxicity assays compared to the control bispecific GUCY2C-1608 paired with anti-CD3 2B5v6 (Fig. 6); (2) had binding affinity for recombinant human CD3 via SPR, consistent with cytotoxic activity; (3) had improved immunogenicity scores compared to the control; and (4) AC-SINs scores remained the same as the control bispecific GUCY2C-1608. These results indicate that anti-GUCY2c bispecific antibodies with different CD3 variants recruit naïve human T cells to induce cell killing in T84 tumor cells (Fig. 16). The three anti-CD3 variants derived from 2B5v6 showed approximately 2-fold greater potency compared to the control bispecific GUCY2C-1608.

Table 18

Bispecific

CD3 variant

Epivax Score (tReg - Adjusted)

MassSpec (Complete)

AC-SINs score

In vitro cytotoxicity (nM)

KD (nM, SPR)

GUCY2C-1608

2B5-0542

-33.7

100% heterodimer

3

0.772

127.88 ± 1.63

GUCY2C-1678

2B5-1038

-52.61

100% heterodimer

2

0.247

63.55±0.49

GUCY2C-1679

2B5-1040

-47.58

100% heterodimer

3

0.194

72.02 ± 0.36

GUCY2C-1680

2B5-1039

-53.51

100% heterodimer

3

0.385

78.57±0.9

The sequences of the bispecific antibodies are shown in Table 19.

Table 19

Claims (19)

1. An antibody that specifically binds to CD3, wherein the antibody comprises:

VH region, comprising (i) VH CDR1 consisting of a sequence of SEQ ID NO:13 according to the Kabat numbering system, SEQ ID NO:14 according to the Chothia numbering system, or SEQ ID NO:15 according to an extended numbering system; (ii) VH CDR2 consisting of a sequence of SEQ ID NO:19 according to the Kabat numbering system or SEQ ID NO:20 according to the Chothia numbering system; and (iii) VH CDR3 consisting of a sequence of SEQ ID NO:18; and VL CDR1 consisting of a sequence of SEQ ID NO:29; (ii) VL CDR2 consisting of a sequence of SEQ ID NO:27; and (iii) VL CDR3 consisting of a sequence of SEQ ID NO:28.

2. The antibody of claim 1, wherein the antibody comprises:

The VH region consists of the sequence of SEQ ID NO:4, and the VL region consists of the sequence of SEQ ID NO:3.

3. The antibody of claim 1, wherein the antibody is selected from: Fab, F(ab) ₂ fragment, Fv fragment, monoclonal antibody, chimeric antibody, multispecific antibody, labeled antibody, humanized antibody, human antibody and fragment thereof.

4. The antibody of claim 1, wherein the antibody is selected from: Fab fragments, single-chain Fv fragments, bispecific antibodies, and trispecific antibodies.

5. The antibody of claim 1, wherein the antibody is selected from: disulfide-stabilized Fv fragments, bispecific heterodimeric biantibodies, and bispecific heterodimeric IgG.

6. The antibody of claim 1, wherein the antibody is a single-chain antibody.

7. The antibody of claim 4, wherein the antibody is a bispecific antibody.

8. The antibody of claim 7, which specifically binds to tumor antigens.

9. The antibody of claim 1, further comprising a human or humanized VH framework and a human or humanized VL framework.

10. The antibody of claim 1, wherein the antibody is a humanized antibody.

11. A pharmaceutical composition comprising an antibody of any one of claims 1 to 10 and a pharmaceutically acceptable carrier.

12. A polynucleotide comprising a nucleotide sequence encoding an antibody of any one of claims 1 to 10.

13. A vector comprising the polynucleotide of claim 12.

14. A host cell comprising the vector of claim 13.

15. The host cell of claim 14, wherein recombination produces an antibody of any one of claims 1 to 10.

16. The host cell of claim 14 or 15, wherein the host cell is selected from bacterial cell lines, mammalian cell lines, insect cell lines, and yeast cell lines.

17. The host cell of claim 16, wherein the mammalian cell line is a CHO cell line.

18. A method for producing an antibody, comprising culturing a host cell of any one of claims 14 to 17 under conditions that result in the production of an antibody of any one of claims 1 to 10, and purifying the antibody from the culture supernatant.

19. Use of the antibody of any one of claims 1 to 10, the pharmaceutical composition of claim 11, the polynucleotide of claim 12, the carrier of claim 13, or the host cell of any one of claims 14 to 17 in the preparation of a medicament for treating colorectal cancer.