EA041572B1 - ANTIBODIES TO CD3, ANTIBODIES TO CD123 AND BI-SPECIFIC ANTIBODIES THAT BIND SPECIFICALLY TO CD3 AND/OR CD123 - Google Patents

Description

The invention extends to an antibody-like binding protein that specifically binds to CD3 and specifically binds to at least one additional antigen, such as CD123. The present invention further extends to anti-CD3 antibodies and anti-CD123 antibodies. The present invention also extends to an antibody-like binding protein that specifically binds to CD123 and specifically binds to at least one additional antigen. The present invention also relates to pharmaceutical compositions containing the antibody-like binding protein, anti-CD3 antibodies or anti-CD123 antibodies of the present invention, and their use in the treatment of cancer. The present invention further relates to isolated nucleic acids, vectors and host cells containing a sequence encoding said antibody-like binding protein, an anti-CD3 antibody, or an anti-CD123 antibody, and to the use of said anti-CD123 antibody as a diagnostic tool.

The first generation of bispecific antibodies was developed over 20 years ago. Since then, a number of clinical studies have tested bispecific antibodies designed to target the surface antigens of cancer cells. This group of anti-cancer fusion proteins contains two or more functional domains that locate immunological effector cells near target cancer cells to provide anti-cancer activity.

As bispecific antibody technology has advanced, another group of fusion proteins called bispecific T cell activators (BiTEs) have been generated by joining only two single-chain antibody variable regions (scFv) (without including Fc amino acid segments) with a flexible linker, with one scFv binding to the target cells, and the other binds CD3 on the surface of T cells. One BiTE, blinatumomab, characterized by a bispecific CD19xCD3 binding activity, showed encouraging results for patients in phase II clinical trials with minimal residual disease in B-cell acute lymphoblastic leukemia.

CD123 (interleukin-3 receptor alpha chain, IL-3Ra) is a tumor antigen that is overexpressed in various hematological neoplasms. Most blast cells in AML express CD123 on the surface, and this expression does not vary between AML subtypes. Higher expression of CD123 in AML at diagnosis has been reported to be associated with poorer prognosis. CD123 has been reported to be expressed on leukemic stem cells (LSC). A growing body of evidence suggests that AML arises from these leukemic stem cells (LSCs), which have been shown to be dormant and relatively resistant to DNA-damaging chemotherapy.

Thus, the increased expression of CD123 on LSCs compared to hematopoietic stem cells (HSCs) provides an opportunity for therapeutic targeting of AML-LSCs.

As previously shown, the monoclonal antibody (MAb) 7G3, raised against CD123, inhibits IL-3 mediated proliferation and activation of both leukemic cell lines and primary cells (US Pat. No. 6,177,078). However, it remained unclear whether CD123 targeting could lead to functional impairment in AML-LSC.

The use of an antibody-like binding protein to CD123xCD3 leads to the cytolysis of tumor cells, as shown in this document by the authors of the present invention.

The idea of obtaining a bispecific antibody-like binding protein with bispecific CD123xCD3 binding activity has already been proposed and described in international patent application WO 2013/173820.

In addition, a dual-affinity bispecific retargeting antibody (DART) molecule for CD123xCD3 from MacroGenics entered phase I clinical trials in 2014.

However, as shown by the present inventors, the MacroGenics CD123xCD3 bispecific retargeting antibody (DART) molecule, for example, is characterized by activation of 82% of CD4+ expressing T cells and 83% of CD8+ expressing T cells in the absence of target cells. Inappropriate T cell activation can lead to severe side effects such as cytokine release syndrome. Cytokine release syndrome refers to the release of cytokines by activated T cells, resulting in a type of systemic inflammatory response similar to that seen in severe infections and characterized by hypotension, hyperthermia, and shivering. Deaths due to cytokine release syndrome have been reported with OKT3.

Therefore, despite these advances in bispecific antibody technology, there remains a need for additional anti-cancer therapeutics, in particular those that effectively target and kill cancer cells, either directly or indirectly.

The present inventors have been successful in generating, screening and selecting rat-specific anti-CD3 antibodies that exhibit high affinity for both human and Macac fascicularis CD3 protein.

The present inventors have developed antibody-like binding proteins having biological and immunological specificity for the CD3 antigen and at least one

- 1 041572 additional target antigen. In one example, to demonstrate the use of these anti-CD3 antibodies in the production of bespecifically antibody-like binding proteins, the present inventors prepared anti-CD3/CD123 antibody-like binding proteins and demonstrated their therapeutic use. Such bispecific anti-CD3/CD123 antibody-like binding proteins are characterized by low T-cell activation, as observed with anti-CD3 antibody alone. However, if CD123-expressing target cells, such as THP-1 cells, are present, the anti-CD3/CD123 bespecifically antibody-like binding proteins exhibit high T-cell activation. Accordingly, the anti-CD3 antibody of the present invention as defined above is particularly useful for preparing the antibody-like binding proteins of the present invention.

Definitions.

Throughout this application, the term and/or is a grammatical conjunction, which should be interpreted to mean that one or more of the cases it connects can occur. For example, the formulation of such native sequence proteins can be produced using standard recombinant and/or synthetic methods indicates that native sequence proteins can be produced using standard recombinant and synthetic methods, or native sequence proteins can be produced using standard recombinant methods. , or native sequence proteins can be made using synthetic methods.

In addition, throughout this application, the term containing should be interpreted as covering all the features specifically mentioned, as well as optional, additional, unspecified. As used herein, the use of the term comprising also discloses an embodiment in which no features are present except for the features specifically mentioned (ie, consisting of). In addition, the use of the singular form does not exclude the plural. The fact that certain measures are set forth in mutually exclusive dependent claims does not in itself indicate that a combination of these measures cannot be used to an advantage.

The term gene means a DNA sequence that codes for or corresponds to a particular sequence of amino acids that is all or part of one or more proteins or enzymes, and may or may not include DNA regulatory sequences such as promoter sequences, for example, which determine the conditions under which which the gene is expressed. Some genes that are not structural genes can be transcribed from DNA to RNA but are not translated into an amino acid sequence. Other genes may function as regulators of structural genes or as regulators of DNA transcription. In particular, the term gene may refer to a genomic sequence encoding a protein, ie, a sequence containing the sequences of the regulator, promoter, introns and exons.

A sequence that is at least 85% identical to a reference sequence is one that has 85% or more identity over its entire length, such as 90%, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identity. sequence with the full length of the reference sequence.

In the context of the present application, percent identity is calculated using a global pairwise alignment (i.e., two sequences are compared by their full length). Methods for comparing the identity of two or more sequences are well known in the art. For example, you can use the needle program, which uses the Needleman and Wunsch global sequence alignment algorithm (Needleman and Wunsch, 1970 J. Mol. Biol. 48:443-453) to find the optimal alignment (gapped) of two sequences over their full length. The Needle program, for example, is available on the Internet at ebi.ac.uk. According to the present invention, percent identity between two polypeptides is calculated using the EMBOSS: needle (global alignment) program with a gap opening penalty of 10.0, a gap extension penalty of 0.5, and a Blosum62 matrix.

Proteins consisting of an amino acid sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98% or 99% identical to the reference sequence may contain mutations such as deletions, insertions and/or substitutions compared to the reference sequence. In the case of substitutions, a protein consisting of an amino acid sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, or 99% identical to the reference sequence may correspond to a homologous sequence derived from a species other than that from which received the reference sequence.

Amino acid substitutions may be conservative or non-conservative. Preferably, the substitutions are conservative substitutions in which one amino acid is substituted for another amino acid with similar structural and/or chemical properties. The substitution preferably corresponds to the conservative substitution indicated in the table below.

- 2 041572

Conservative substitutions Amino acid type Ala, Vai, Leu, lie, Met, Pro, Phe, Trp Amino acids with aliphatic hydrophobic side chains Ser, Tyr, Asn, Gin, Cys Amino acids with uncharged but polar side chains Asp, Glu Amino acids with acidic side chains Lys, Arg, His Amino acids with basic side chains gly Neutral side chain

An antibody, also referred to as an immunoglobulin, may be a natural or conventional antibody in which two heavy chains are linked to each other by disulfide bonds and each heavy chain is linked to a light chain by a disulfide bond. There are two types of light chains, lambda (l) and kappa (k). There are five major classes (or isotypes) of heavy chains that determine the functional activity of an antibody molecule: IgM, IgD, IgG, IgA, and IgE. Each strand contains a different sequence of domains. The light chain contains two domains or regions, a variable domain (VL) and a constant domain (CL).

The heavy chain contains four domains, a variable domain (VH) and three constant domains (CH1, CH2 and CH3, collectively referred to as CH). The variable regions of both the light (VL) and heavy (VH) chains determine binding recognition and antigen specificity. The constant region domains of the light (CL) and heavy (CH) chains confer important biological properties such as antibody chain assembly, secretion, placental movement, complement fixation, and binding to Fc receptors (FcRs). The Fv fragment is the N-terminal portion of the Fab fragment of an immunoglobulin and consists of the variable portions of one light chain and one heavy chain. The specificity of an antibody lies in the structural complementarity between the paratope of the antibody and the antigenic determinant. Antibody paratopes are composed of residues primarily of hypervariable or complementarity determining regions (CDRs). Residues from non-hypervariable or framework regions (FRs) sometimes affect the overall structure of the domain and hence the paratope. Complementarity-determining regions or CDRs refer to amino acid sequences that together determine the binding affinity and specificity of the natural Fv binding region of a native immunoglobulin. Each of the immunoglobulin light and heavy chains has three CDRs, designated as CDR1-L, CDR2-L, CDR3-L and CDR1-H, CDR2-H, CDR3-H, respectively. The conventional antigen binding site of an antibody thus contains six CDRs, which include a set of CDRs from each of the heavy and light chain V regions.

In the context of the present invention, an antibody or immunoglobulin is IgM, IgD, IgG, IgA and IgE.

Framework regions (FRs) are the amino acid sequences located between the CDRs, i.e. those portions of the variable regions of an immunoglobulin light and heavy chain that are relatively conserved among different immunoglobulins in the same species. Each of the immunoglobulin light and heavy chains has four FRs, designated FR1-L, FR2-L, FR3-L, FR4-L, and FR1-H, FR2-H, FR3-H, FR4-H, respectively. Accordingly, the light chain variable domain can thus be designated as (FR1-L)-(CDR1-L)-(FR2-L)-(CDR2-L)-(FR3-L)-(CDR3-L)-(FR4- L), and the heavy chain variable domain can thus be referred to as (FR1-H)-(CDR1-H)-(FR2-H)(CDR2-H)-(FR3-H)-(CDR3-H)-(FR4 -H).

Knowing the amino acid sequence of the CDR, one skilled in the art will readily be able to determine the framework regions of FR1-L, FR2-L, FR3-L, FR4-L and/or FR1-H, FR2-H, FR3-H, FR4-H.

As used herein, a human framework is a framework that is substantially identical (about 85% or more, in particular 90%, 95%, 97%, 99%, or 100%) to a naturally occurring human antibody framework.

In the context of the present invention, the definition of CDR/FR in the light or heavy chain of an immunoglobulin should be given based on the determination by IMGT (Lefranc et al. Dev. Comp. Immunol., 2003, 27(1):55-77; www.imgt.org) .

As used herein, the term antibody refers to conventional antibodies and fragments thereof, as well as single domain antibodies and fragments thereof, in particular the heavy chain variable region of single domain antibodies, and chimeric, humanized, bispecific or polyspecific antibodies.

As used herein, an antibody or immunoglobulin also includes single-domain antibodies, which have been described more recently, which are antibodies in which complementarity-determining regions are part of a single-domain polypeptide. Examples of single domain antibodies include heavy chain-only antibodies, antibodies

- 3 041572 natural light chain-free, single-domain antibodies derived from conventional four-chain antibodies, engineered single-domain antibodies. Single domain antibodies can be derived from any species including, but not limited to, mouse, human, camel, llama, goat, rabbit, bovine. Single domain antibodies can be naturally occurring single domain antibodies known as heavy chain only antibodies with no light chains. In particular, species of the family Camelidae, such as the camel, dromedary, llama, alpaca, and guanaco, produce antibodies consisting only of heavy chains, which do not naturally have light chains. Camelid antibodies, consisting only of heavy chains, also lack the CH1 domain.

The heavy chain variable region of these light chain-free single domain antibodies is known in the art as VHH or nanobody. Like traditional VH domains, VHHs contain four FRs and three CDRs. Nanobodies have advantages over traditional antibodies: they are approximately ten times smaller than IgG molecules, and as a result, correctly folded functional nanobodies can be obtained by in vitro expression in high yields. In addition, nanobodies are very stable and resistant to proteases. A review of the properties and preparation of nanobodies has been given in Harmsen and De Haard HJ (Appl. Microbiol. Biotechnol. 2007 Nov;77(1):13-22).

As used herein, the term monoclonal antibody or mAb refers to an antibody molecule with a single amino acid composition that is directed against a specific antigen and should not be construed as requiring the production of an antibody by any particular method. The monoclonal antibody may be derived from a single B cell clone or hybridoma, but may also be recombinant, ie. obtained using protein engineering.

The term chimeric antibody refers to an engineered antibody that, in its broadest sense, contains one or more regions from one antibody and one or more regions from one or more other antibodies. In particular, a chimeric antibody contains the VH domain and VL domain of an antibody derived from a non-human animal, together with the CH domain and CL domain of another antibody, in particular a human antibody. As the non-human animal, any animal such as mouse, rat, hamster, rabbit, and the like can be used. A chimeric antibody can also mean a polyspecific antibody having specificity for at least two different antigens.

The term humanized antibody refers to an antibody that is wholly or partly non-human and has been modified to replace certain amino acids, in particular in the framework regions of the heavy and light chains, in order to avoid or minimize the immune response in humans. The constant domains of a humanized antibody are in most cases human CH and CL domains.

Numerous methods for humanizing an antibody sequence are known in the art; see for example the review in Almagro & Fransson (2008) Front Biosci. 13:1619-1633. One commonly used method is CDR grafting, or antibody rearrangement, which involves grafting CDR sequences from a donor antibody, typically a mouse antibody, into a backbone from a human antibody framework region of differing specificity. Because CDR grafting can reduce the binding specificity and affinity, and thus the biological activity, of a non-human CDR-grafted antibody, backmutations can be introduced at selected positions in the CDR-grafted antibody to retain the binding specificity and binding affinity of the parent antibody. The identification of positions for possible backmutations can be performed using information available in the literature and in the antibody database. Amino acid residues that are candidates for backmutation are typically those that are located on the surface of the antibody molecule, while residues that are immersed or that are low on the surface will usually not be changed. An alternative humanization technique to CDR grafting and backmutation is resurfacing in which non-surface non-human residues are left while surface residues are changed to human residues. Another alternative technique is known as guided selection (Jespers et al. (1994) Biotechnology 12, 899) and can be used, for example, to generate a fully human antibody from a mouse or rat antibody while maintaining the epitope and binding characteristics of the parent antibody. An additional humanization method is the so-called 4D humanization. A 4D humanization protocol is described in US 20110027266 A1 (WO 2009032661 A1) and is exemplified below by using 4D humanization to humanize the light (VL) and heavy (VH) rat antibody chain variable domains. In one example, a homologous rat antibody model was generated primarily using MOE software (v. 2011.10 - Chemical Computing Group, Quebec, Canada) using PDB structures (Berman et al., Nucleic Acids Research, 2000, 28:235-242 ) as templates and then subjected to energy minimization using standard procedures implemented in MOE. Then, on a minimized 3D homologous model (performed using software

rat antibodies were modeled by molecular dynamics (MD) and compared, for example, with 49 human antibody models derived from seven representative light chains (vk1, vk2, vk3, vk4, vlambda1, vlambda2, vlambda3) and seven representative heavy chains (vh1a, vh1b, vh2, vh3, vh4, vh5, vh6) designated by LGCR/SDI and available from MOE. For example, one chain pair model (Vkx-Vhx) with the best in both hydrophobic, electrostatic components and sequence identity beyond the CDR was selected for 4D humanization. For pairwise comparison of the variable domain of the rat antibody and the selected model, the sequences were aligned mainly based on the optimal 3D overlap of alpha carbons in the respective homologous models. Undesirable motifs were then identified and mutated. Finally, the resulting humanized sequences were subjected to sequence similarity Blast, e.g. to the IEDB database (http://www.immuneepitope.org; version 2012/01/30 available locally) to ensure that none of the sequences does not contain any known B - or T-cell epitope given in it.

In the case of chimeric antibodies, humanization typically involves modification of the framework regions of the variable region sequences.

Amino acid residues that are part of a CDR will generally not be changed due to humanization, although in some cases it may be necessary to change individual CDR amino acid residues, for example, to remove a glycosylation site, a deamidation site, or an unwanted cysteine residue. N-linked glycosylation occurs by attaching an oligosaccharide chain to an aspartic residue in the Asn-X-Ser or Asn-X-Thr tripeptide sequence, where X can be any amino acid except Pro. Removal of the N-glycosylation site can be achieved by mutating either the Asn or the Ser/Thr residue to another residue, in particular by conservative substitution. Deamidation of aspartic and glutamine residues can occur depending on factors such as pH and surface exposure. Aspartic residues are particularly susceptible to deamidation, primarily when found in the Asn-Gly sequence, and to a lesser extent in other dipeptide sequences such as Asn-Ala. When such a deamidation site, particularly Asn-Gly, is present in a CDR sequence, then it may be desirable to remove that site, usually by making a conservative substitution to remove one of the residues involved. It is understood that a substitution in the CDR sequence to remove one of the residues involved is also covered by the present invention.

Fragments of (traditional) antibodies contain a portion of an intact antibody, in particular the antigen-binding region or variable region of an intact antibody. Examples of antibody fragments include Fv, Fab, F(ab')2, Fab', dsFv, (dsFv)2, scFv, sc(Fv)2, diabodies, bispecific and multispecific antibodies formed from antibody fragments. The conventional antibody fragment can also be a single domain antibody, such as a heavy chain-only or VHH antibody.

The term Fab denotes an antibody fragment with a molecular weight of approximately 50,000 and antigen-binding activity from among the fragments obtained by treatment of IgG with the protease papain, in which approximately half of the N-terminal part of the H chain and the entire L chain are linked to each other by a disulfide connections.

The term F(ab')2 refers to an antibody fragment with a molecular weight of approximately 100,000 and antigen-binding activity from among the fragments obtained by treating an IgG protease with pepsin, which is slightly larger than Fab, linked by a hinge disulfide bond.

The term Fab refers to an antibody fragment with a molecular weight of approximately 50,000 and antigen-binding activity, which is obtained by cutting the disulfide bond of the F(ab')2 hinge region.

A single chain Fv polypeptide (scFv) is a covalently linked VH::VL heterodimer that is typically expressed by gene fusion containing the genes encoding VH and VL connected by a linker encoding the peptide. The human scFv fragment of the present invention contains CDRs maintained in the proper conformation, in particular by the use of gene recombination techniques. Bivalent and polyvalent antibody fragments can either be formed spontaneously by combining monovalent scFvs, or they can be obtained by connecting monovalent scFvs with a peptide linker, such as in the case of bivalent sc(Fv)2. dsFv is a VH::VL heterodimer stabilized with a disulfide bond. (dsFv) ₂ denotes two dsFv connected by a peptide linker.

The term bispecific antibody or BsAb generally refers to an antibody in which the antigen-binding sites of two antibodies are combined within a single molecule. Thus, BsAbs are able to bind to two different antigens simultaneously. For the development, modification and production of antibodies or antibody derivatives with the required set of binding properties and effector

- 5 041572 functions are increasingly using genetic engineering, as described, for example, in EP 2050764 A1.

The term polyspecific antibody means an antibody in which the antigen-binding sites of two or more antibodies are combined in one molecule.

The term diabodies refers to small fragments of antibodies with two antigen-binding sites, while these fragments contain a heavy chain variable domain (VH) connected to a light chain variable domain (VL) as part of a single polypeptide chain (VH-VL). Due to the use of a linker too short to allow pairing between two domains on the same strand, the domains are forced to pair with complementary domains on the other strand and form two antigen-binding sites.

The term hybridoma refers to a cell which is obtained by fusing a B cell obtained by immunizing a non-human mammal with an antigen with a myeloma cell obtained from a mouse, etc., which produces a desired monoclonal antibody having specificity for the antigen.

By purified and isolated is meant, when referring to a polypeptide (ie, an antibody of the present invention) or nucleotide sequence, that said molecule is present in the virtual absence of other biological macromolecules of the same type. The term "purified" as used herein specifically means that at least 75%, 85%, 95%, or 98% by weight of one type of biological macromolecule is present. An isolated nucleic acid molecule that codes for a particular polypeptide is a nucleic acid molecule that is substantially free of other nucleic acid molecules that do not code for that polypeptide; however, the molecule may include some additional bases or moieties that do not adversely affect the essential characteristics of the composition.

The term antigen or target antigen as used herein refers to a molecule or portion of a molecule capable of being bound by an antibody or antibody-like binding protein. The term also refers to a molecule or portion of a molecule that can be used in an animal to produce antibodies that are capable of binding to an epitope of such an antigen. The target antigen may have one or more epitopes. Given that each target antigen is recognized by an antibody or antibody-like binding protein, the antibody-like binding protein is able to compete with an intact antibody that recognizes the target antigen.

Theoretically, affinity is determined by the equilibrium relationship between the whole antibody and antigen. Affinity can be expressed, for example, as half maximum effective concentration (EC ₅₀ ) or equilibrium dissociation constant (KD).

The half-maximal effective concentration, also referred to as EC ₅₀ , is the concentration of a drug, antibody, or toxicant that induces a response that is midway between the initial response and the maximum response after a specified exposure time. EC ₅₀ and affinity are inversely proportional to each other, and the lower the EC ₅₀ value, the higher the affinity of the antibody.

KD is the equilibrium dissociation constant, koff/kon, between an antibody and its antigen. KD and affinity are inversely proportional to each other. The KD value refers to the concentration of the antibody, and the lower the KD value, the higher the affinity of the antibody. Affinity can be measured experimentally using various known methods, such as by measuring the rate of association and dissociation using surface plasmon resonance or measuring the EU ₅₀ in an immunoassay (ELISA, FACS). An enzyme-linked immunosorbent assay (ELISA) is a biochemical assay that uses a solid phase enzyme immunoassay to detect the presence of a substance, usually an antigen, in a liquid sample or wet sample. Antigens from the sample are attached to the surface. An additional specific antibody is then applied to the surface so that it can bind to the antigen. This antibody is bound to the enzyme and, as a final step, a substance containing the substrate of the enzyme is added. The subsequent reaction produces a detectable signal, most often a color change of the substrate. Fluorescence activated cell sorting (FACS) provides a method for sorting a heterogeneous mixture of biological cells into two or more containers, one cell at a time, based on the specific light scattering and fluorescence characteristics of each cell. In these assays, the EC ₅₀ is the concentration of antibody that induces a response that is midway between the response at the initial time point and the maximum response after a certain time of exposure to a certain concentration of antigen in ELISA (enzyme immunosorbent assay) or expression of antigen by a cell in FACS (cell sorting with fluorescence activation). Surface plasmon resonance is a labelless method that directly measures the binding of a molecule in the soluble phase (analyte) to a ligand molecule immobilized on the sensor surface. In a sensor device, ligand binding is monitored by an optical phenomenon called surface plasmon. In particular, when the analyte molecule dissociates from the ligand molecule, there is a decrease in the SPR signal (expressed in resonance units, RU). The rates of association (association rate, k _a ) and dissociation (dissociation rate, kd) are obtained

- 6 041572 on the basis of the signal obtained during the association and dissociation, and from them it is possible to calculate the equilibrium dissociation constant (binding constant, K _D ). The signal given in resonance units (RU) depends on the size of the ligand present in the analyte, however, in the case of the same experimental conditions, i.e. the ligand is the same molecule under the same conditions, the value obtained in RU can measure affinity, the higher the signal obtained in RU, the higher the binding.

Binding of a monoclonal antibody to antigen 1 (Ag1) is cross-reactive to antigen 2 (Ag2) when the EC _50s are in the same range for both antigens.

In the present application, the binding of a monoclonal antibody to Ag1 is cross-reactive to Ag2 when the ratio of affinity for Ag2 and affinity for Ag1 is 10 or less (in particular 5, 2, 1 or 0.5), while affinities are measured using one method for both antigens.

Binding of a monoclonal antibody to Ag1 is not significantly cross-reactive to Ag2 when the affinities for the two antigens are very different. Affinity for Ag2 may not be measurable if the binding reaction is too weak. In the present application, the binding of a monoclonal antibody to Ag1 is not significantly cross-reactive to Ag2 when the binding reaction of the monoclonal antibody to Ag2 is less than 5% of the binding reaction of the same monoclonal antibody to Ag1 under the same experimental conditions and at the same concentration of antibody. In practice, the concentration of antibody used may be EC ₅₀ or the concentration required to reach the saturation plateau observed with Ag1.

As used herein, specificity refers to the ability of an antibody to distinguish the target peptide sequence to which it binds (epitope) from closely related, highly homologous peptide sequences.

A monoclonal antibody specifically binds Ag1 when it is not significantly cross-reactive with Ag2.

A domain can be any region of a protein, usually defined on the basis of sequence homology and conceived associated with a certain structural or functional unity.

A recombinant molecule is a molecule that has been produced, expressed, created or isolated by recombinant methods.

As used herein, the term subject means a mammal, such as a rodent, a feline, a canine, and a primate. In particular, the subject of the present invention is a human.

Antibodies to CD3.

CD3 refers to an antigen that is expressed on T cells as part of a multimolecular complex of T cell receptors and that consists of at least three different chains, CD3ε, CD3δ and CD3y. CD3δ and CD3y have low sequence identity and/or similarity to human CD3ε (less than 20% similarity and identity). CD3ε and CDR3δ together may form a complex, referred to herein as the CD3ε/δ complex. CD3ε also forms a complex with CDR3y, the so-called CD3ε/γ complex. Clustering of CD3 on T cells, for example by immobilized anti-CD3 antibodies, results in T cell activation similar to T cell receptor involvement, but independent of the specificity typical of its clone. CD3ε contains three domains, an intracellular domain, a transmembrane domain, and an extracellular domain.

Most of the prior art anti-CD3 antibodies recognized the CD3ε chain. One such anti-CD3 antibody of the prior art is OKT3. In the prior art, examples have been given of T cell activation phenomena when using antibody molecules, for example when using an OKT3 antibody molecule. An anti-CD3 antibody and a variant thereof have been described in the prior art (US 4361549; US 4361549; US 5885573; US 5929212 and WO 98/52975 or US 5955358). OKT3 has additionally been used as an effective immunosuppressive agent in clinical transplantation for the treatment of allograft rejection (Thistlethwaite 1984, Transplantation 38, 695-701; Woodle 1991, Transplantation 51, 1207-1212; Choi 2001, Eur. J. Immunol. 31(1), 94 -106).

The main disadvantage of this therapy is the activation of T cells, manifested in the release of cytokines, due to cross-linking of T cells and FcyR-bearing cells and response with human anti-mouse antibody (HAMA). Some publications have described changes such as humanizing OKT3 to reduce these side effects: US 5929212; US 5885573 and others. On the other hand, OKT3 or other anti-CD3 antibodies can be used as immunostimulatory agents to stimulate T cell activation and proliferation (US 6406696 from Bluestone; US 6143297 from Bluestone; US 6113901 from Bluestone; Yannelly 1990, J. Immunol. Meth. 1 , 91-100). Anti-CD3 antibodies have also been described as agents used in combination with anti-CD28 antibodies to induce T cell proliferation (US 6352 694). OKT3 was additionally used on its own or as a component

- 7 041572 bispecific antibodies for targeting cytotoxic T cells to tumor cells or virus-infected cells (Nitta 1990, Lancet 335, 368-376; Sanna 1995, Bio/Technology 13, 12211224; WO 99/54440).

Until now, approaches using antibodies as a means of recruiting T cells have been held back by some data. First, naturally occurring or engineered antibodies with high binding affinity for T cells often do not activate the T cells they bind to. Second, naturally occurring or engineered antibodies with low T cell binding affinity are also often ineffective due to their ability to trigger T cell mediated cell lysis.

The reference sequence of the full-length human CD3e protein, including the signal peptide, is available from the Uniprot database under accession number P07766 and is disclosed herein under SEQ ID NO: 1 (as available on December 12, 2014).

The reference sequence of the full-length Macac fascicularis CD3e protein, including the signal peptide, is available from the Uniprot database under accession number Q95LI5 and is disclosed herein under SEQ ID NO: 2 (as available on December 12, 2014).

The sequence of fusion proteins containing Fc and His-tagged mature human CD3e cloned by the present inventors from genomic DNA is disclosed under SEQ ID NO: 3. Said fusion protein containing Fc and His-tagged mature human CD3e contains amino acids 23-126 of full-length human CD3e protein and thus contains the extracellular domain of human CD3e.

The sequence of a fusion protein containing Fc and mature Macaca fascicularis CD3e cloned by the present inventors from genomic DNA is disclosed under SEQ ID NO: 4. Said fusion protein containing Fc and mature Macaca fascicularis CD3e contains amino acids 23-117 from the full-length Macaca CD3e protein fascicularis and thus contains an extracellular domain of human or Macaca fascicularis CD3e containing one alanine to valine substitution at amino acid position 35 compared to amino acid position 57 of the wild-type sequence.

The domain organization of human CD3e and Macaca fascicularis is as follows (based on the P07766 sequence in Uniprot (human) and the Q95LI5 sequence in Uniprot (Macaca fascicularis)):

CD3s domains Provisions in SEQ NO: 1 (person) ID Positions in SEQ ID NO: 2 (Masasa fascicularis) Extracellular 23-126 22-117 transmembrane domain 127-152 118-138 cytoplasmic 153-207 139-198

Accordingly, the extracellular domain of human CD3e consists of amino acids at positions 23-126 of SEQ ID NO: 1, and the extracellular domain of Macaca fascicularis CD3e consists of amino acids at positions 22-117 of SEQ ID NO: 2.

The authors of the present invention have been successful in obtaining, screening and selection of mouse and rat specific antibodies to CD3. These anti-CD3 antibodies show high affinity for both human and Macaca fascicularis CD3 protein and at the same time have low T-cell activation in the absence of target cells.

The present inventors determined the sequence of the heavy and light chain variable regions of such monoclonal antibodies, anti-CD3 antibodies, referred to as 20G6-F3, 4B4-D7, 4E7C9, 18F5-H10, 12D2-E5, 11D7-C3, 11H3-E5, 13H2-C2, 13C1-F6, 18H11-F10, 1E6-C9, 10F4-C10, 10E6-G6, 18G9-H11, 11F3-B9, 12G3-E8, 5B1-G2, 16F8-A7, 11F9-F8, 3G5-E10, 9D7- F3, 8C2-F7, 20E5-F10, 20E5-F10, 6C9-C9, 3E8-G1, 3H6-D2 and 8H2.

The anti-CD3 antibody, called 20G6-F3, contains:

heavy chain variable domain, consisting of the sequence

EVQLVETGGSLVQPGKSLKLTCATSGFTFTKAWMHWVRQSPEKQLEWVAQIKDKSNSYA

TYYAESVKGRFTISRDDSKSTIYLOMNSLKEEDTAIYYCRGVYYALSPFDYWGOGVMVTVSS (SEQ ID NO: 5, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 6, CDR2-H with the sequence of SEQ ID NO: 7, and CDR3-H with the sequence under SEQ ID NO: 8, and a light chain variable domain consisting of the sequence

DWMTQTPVSLSVSLGGQVSISCRSSQSLVHNNGNTYLSWYLQKPGQSPQSLIYKVSNRFSGFS

DRFSGSGSGTDFTLKISRRVDPDDLGVYYCGQGTQYPFTFGSGTKLEIK

- 8 041572 (SEQ ID NO: 9, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence of SEQ ID NO: 10, a CDR2-L consisting of the KVS sequence, and a CDR3-L with the sequence under SEQ ID NO: 11.

The anti-CD3 antibody, called 4B4-D7, contains a heavy chain variable domain consisting of the sequence

EVQLVETGGRLVQPGRSLKLTSATSGFTFSNAWMHWVROSPEKOLEWVAQIKARSNNYATYYAE

SVKGRFTISRDDSKSTIYLOMNSLKEEDTAIYYCRGTYYASKPFDYWGOGVMVTVSS (SEQ ID NO: 12, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 13, CDR2-H with the sequence of SEQ ID NO: 14, and CDR3-H with the sequence under SEQ ID NO: 15, and a light chain variable domain consisting of the sequence

DVVMTOTPVSLSVSLGGOVSISCR5SQSLVHDNGNTYLSWSLORPGOSPOVL IYKVSNRFSGTS

DRFTGSGSGTDFTLKISRVEPDDLGVYYCGQGTQYPFTFGSGTKLEIK (SEQ ID NO: 16, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 17, a CDR2-L consisting of the sequence under KVS, and a CDR3-L with the sequence under SEQ ID NO: 11.

The anti-CD3 antibody, called 4E7-C9, contains a heavy chain variable domain consisting of the sequence

EVOWETGGSLVOPGKSLKLTCATSGFTFSNAWMHWVROSPEKOLEWVAQIKDKSNNYATYYAE

SLKGRFTISRDDPKRSIYLOMNSLREEDTAIYYCRYVHYGIGYAMDAWGOGTSVTVSS (SEQ ID NO: 18, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 13, CDR2-H with the sequence of SEQ ID NO: 19, and CDR3-H with the sequence under SEQ ID NO: 20, and a light chain variable domain consisting of the sequence

DVWTOTPVSLSVSLGGOVSISCRSSQSLEHNNGNTYLSWYLOKPGOSPOPLIYKVSNRFSGIS

DRFSGSGSGTDFTLKISRVEPDDLGVYYCGQGTQYPFTFGPGTKLELK (SEQ ID NO: 21, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 22, a CDR2-L consisting of the sequence under KVS, and a CDR3-L with the sequence under SEQ ID NO: 11.

The anti-CD3 antibody, called 18F5-H10, contains a heavy chain variable domain consisting of the sequence

EVOWETGGSLVOPGKSLKLTCATSGFTFTNAWMHWVRRSPEKOLEWVAQIKDKSNNYATYYAE

SVKGRFTISRDDSKSSIYLOMNSLKEEDTAIYYCRYVHYRFAYALDAWGRGTSVSVSS (SEQ ID NO: 23, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 24, CDR2-H with the sequence of SEQ ID NO: 19, and CDR3-H with the sequence under SEQ ID NO: 25, and a light chain variable domain consisting of the sequence

DVLMTOTPVSLSVSLGGOVSISCRSSQSLVHTNGNTYLSWYLOKPGOSPOLLIYKVSNRLSGIS

DRFSGSGSGTDFTLKISRVEPDDLGVYYCGQGTHYPFTFGAGTKLELK (SEQ ID NO: 26, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 27, a CDR2-L consisting of the sequence under KVS, and a CDR3-L with the sequence under SEQ ID NO: 28.

The anti-CD3 antibody, called 12D2-E5, contains a heavy chain variable domain consisting of the sequence

EVKLVESGGGLVQPGRSLRLSCAASGFNFYAYWMGWVROAPGKGLEWIGEIKKDGTTINYTPSL

KDRFTISRDNAONTLYLOMTKLGSEDTALYYCAREERDGYFDYWGOGVMVTVSS (SEQ ID NO: 29, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 30, CDR2-H with the sequence of SEQ ID NO: 31, and CDR3-H with the sequence under SEQ ID NO: 32, and a light chain variable domain consisting of the sequence

OFVLTOPNSVSTNLGSTVKLSCKRSTGNIGSNYVNWYOOHEGRSPTTMIYRDDKRPDGVPDRFS

GSIDRSSNSALLTINNVOTEDEADYFCQSYSSGIVFGGGTKLTVL (SEQ ID NO: 33, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 34, a CDR2-L consisting of the sequence RDD, and a CDR3-L with the sequence under SEQ ID NO: 35.

The anti-CD3 antibody, called 11D7-C3, contains a heavy chain variable domain consisting of the sequence

EVQFVETGGSLVQPGKSLKLTCATSGFTFSNAWMHWVROSPEKOLEWVAQIKAKSNNYATYYAE

SVKGRFTISRDDSKS SVYLQMNSLKEEDTATYYCRGLYYGLSPSDYWGQGVMVTVS S

- 9 041572 (SEQ ID NO: 36, with CDRs shown in bold and underlined), containing CDR1-H with the sequence under SEQ ID NO: 13, CDR2-H with the sequence under

SEQ ID NO: 37 and CDR3-H with the sequence under SEQ ID NO: 38, and a light chain variable domain consisting of the sequence

DVWTOTPVSLSVSLGGOVSISCRSSQSLVHNNGNTYLSWYLOKPGOSPOLLIYKV'SNRFSGIS

DRFSGSGSGTDFTLKISRVEPDDLGVYYCGQGTHYPFTFGSGTKLEIK (SEQ ID NO: 39, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 10, a CDR2-L consisting of the KVS sequence, and a CDR3-L with the sequence under SEQ ID NO: 28.

The anti-CD3 antibody, called 11H3-E5, contains a heavy chain variable domain consisting of the sequence

EVQLVEТGGSLVQPGKSLKLTCATSGFTFSNAWMHWVROSREKQLEWVAQIKAKSNNYATYYAE

SVKGRFTISRDDSKSSVYLOMNSL KEEDTAIYYCRGTYYAYKPFDYWGOGVMVTVSS (SEQ ID NO: 40, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 13, CDR2-H with the sequence of SEQ ID NO: 37, and CDR3-H with the sequence of SEQ ID NO: 41, and a light chain variable domain consisting of the sequence

DWMTOTPVSLSVSLGGOVSISCRSSQSLVHDNGNTYLSWSLOKPGOSPOVLIYKVSNRFSGTS

DRFTGSGSGTDFTLKISRVEPDDLGVYYCGQGTQYPFTFGSGTKLEIK (SEQ ID NO: 42, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence of SEQ ID NO: 17, a CDR2-L consisting of the sequence of KVS, and a CDR3-L with the sequence of SEQ ID NO: 11.

The anti-CD3 antibody, called 13H2-C2, contains:

heavy chain variable domain, consisting of the sequence

EEELVETGGSLVOPGKSLKLTCATSGFTFSNAWMHWVROSPDKOLEWVAQIKAKSNNYATYYAE

SVKGRFTISRDDSKSSVYLQMNNLKEEDTAIYYCRYVHYGLAPMDAWG0GTSVTVS5 (SEQ ID NO: 43, with CDRs shown in bold and underlined), containing CDR1-H with the sequence of SEQ ID NO: 13, CDR2-H with the sequence of SEQ ID NO: 37, and CDR3-H with the sequence under SEQ ID NO: 44, and a light chain variable domain consisting of the sequence

DWMTQTPVSLSVSLGGQVSISCRSSQSLVHNNGNTYLSWYLQKAGQSPQLLIYKVSNRFSGIS

DRFSGSGSGTDFILKISRVEPDDLGVFYCGQGTQYPFTFGAGTKLELK (SEQ ID NO: 45, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 10, a CDR2-L consisting of the sequence under KVS, and a CDR3-L with the sequence under SEQ ID NO: 11.

The anti-CD3 antibody, called 13C1-F6, contains a heavy chain variable domain consisting of the sequence

EVOLVETGGTLVOPGKSLKLTCATSGFTFSNAWMHWVROSPEKOLEWVAQIKAKSNNYATYYAE

SVKGRFTISRDDSKTSVYLOLNSLREEDTAIYYCRGTQYGYNPFDYWGOGVMVTVSS (SEQ ID NO: 46, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 13, CDR2-H with the sequence of SEQ ID NO: 37, and CDR3-H with the sequence under SEQ ID NO: 47, and a light chain variable domain consisting of the sequence

DVVMTOSPVSLSVSLGGOVSISCRSSQSLVHNNGNTYLSWYLORSGOSPOLLIYKV’SNRLSGIS

DRFSGSGSGTDFTLKISRIEPDDLGVYYCGQGTQYPFTFGSGTRLEIK (SEQ ID NO: 48, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence of SEQ ID NO: 10, a CDR2-L consisting of the sequence of KVS, and a CDR3-L with the sequence of SEQ ID NO: 11.

An anti-CD3 antibody called 18H11-F10 contains a heavy chain variable domain consisting of the sequence

EVOLVESGGGLVOPGRSLKLSCAASGFIFSDYYMAWVRQAPKKGLEWVATISISGSRTYYPDSV

KGRFTVSRDNAKSSLYLQMNSLKSEDTATYYCATNNPGGWFVYWGOGTLVTVSS (SEQ ID NO: 49, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 50, CDR2-H with the sequence of SEQ ID NO: 51, and CDR3-H with the sequence under SEQ ID NO: 52, and a light chain variable domain consisting of the sequence

NIOMTOSPSLLSASVGDRVTLSCKAGQNINNDLAWYOOKLGEAPRLLIYNANSLOTGIPSRFSG

SGSGADFTLTISSLOPEDVATYFCOQYSSGDTFGAGTKLELK

- 10 041572 (SEQ ID NO: 53, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence of SEQ ID NO: 54, a CDR2-L consisting of the NAN sequence, and a CDR3-L with the sequence under SEQ ID NO: 55.

The anti-CD3 antibody, called 1E6-C9, contains a heavy chain variable domain consisting of the sequence

EVQLVE TGGS LVQPGKS LKL T CAT SGFTFSYAWMHWVROSPDKQLQWVAQIKAKSNNYAT Y YAE

SVEGRFTISRDDSKSSVYLOMNSLKEEDTAIYYCRGVYYGLLGLDAWGOGTSVTVSS (SEQ ID NO: 56, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 57, CDR2-H with the sequence of SEQ ID NO: 37, and CDR3-H with the sequence under SEQ ID NO: 58, and a light chain variable domain consisting of the sequence

DWMTOTPVSLSVRLGGOVSISCRSSQSLVHNNGNTYLSWFLOKPGOSPOLLIYKVSNRFSGIS

DRFSGSASGTDFTLKISRVEPDDLGVYYCGQGTHYPFTFGSGTKLEIK (SEQ ID NO: 59, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 10, a CDR2-L consisting of the KVS sequence, and a CDR3-L with the sequence under SEQ ID NO: 28.

The anti-CD3 antibody, called 10F4-C10, contains a heavy chain variable domain consisting of the sequence

EVQLVEТGGSLVQPGKSLKITSATSGFTFSNAWMHWVRQSPEKQLEWVAQIKAKSNNYATYYAE

SVKGRFTISRDDSKSSIYLQMNSLKEEDTAIYYCRAVNYGNYPLDYWGQGVMVTVS S (SEQ ID NO: 60, with CDRs shown in bold and underlined), containing CDR1-H with the sequence of SEQ ID NO: 13, CDR2-H with the sequence of SEQ ID NO: 37, and CDR3-H with sequence under SEQ ID NO: 61, and a light chain variable domain consisting of the sequence

DWMTOTPVSLSVSLGGOVSISCRSSQSLVHNNGNTYLSWCLOKPGOSPOLLIYKVSNRFSGIS

DRFSGSGSGTDFTLKISRVEPDDLGVYYCGQGTQYPFTFGAGTKLELK (SEQ ID NO: 62, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence of SEQ ID NO: 10, a CDR2-L consisting of the sequence of KVS, and a CDR3-L with the sequence of SEQ ID NO: 11.

The anti-CD3 antibody, called 10E6-G6, contains a heavy chain variable domain consisting of the sequence

EVQLVE T GGGLVQS GKS LKL T CAT SGFTVTNAWMHWVROS PEKQLEWVAQIKAKSNNYE TYYAE

SVKGRFTISRDDSKSSVYLOMNSLKEEDTAIYYCRGTQYGYNPFDYWGOGVMVTVSS (SEQ ID NO: 63, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 64, CDR2-H with the sequence of SEQ ID NO: 65, and CDR3-L with the sequence under SEQ ID NO: 47, and a light chain variable domain consisting of the sequence

DWMTOSPVSLSVSLGGOVSISCRSSQSLVHNNGYTYLSWYLOKPGOSPOVFIYKVSNRFSGIS

DRFSGSGSGTDFTLKISRIEPDDLGVYYCGQGTHYPFTFGSGTKLEIK (SEQ ID NO: 66, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 67, a CDR2-L consisting of the sequence KVS, and a CDR3-L with the sequence under SEQ ID NO: 28.

The anti-CD3 antibody, called 18G9-H11, contains a heavy chain variable domain consisting of the sequence

EVOLVETGGSLVOPGKSLKLTCATSGFTFSNAWIOWVROSPEKOLEWVAQIKAKSNNYATYYAE

SVKGRFTISRDDSKSSVYLOMNSLKEEDTALYYCTWRHYYSSHTMDAWGOGTLVTVSS (SEQ ID NO: 68, with CDRs shown in bold and underlined), containing CDR1-H with the sequence of SEQ ID NO: 13, CDR2-H with the sequence of SEQ ID NO: 37, and CDR3-H with the sequence under SEQ ID NO: 69, and a light chain variable domain consisting of the sequence

DWMTOTPVSLSVSLGGOVSISCRSSQSLVHNNGNTYLSWYLQKPGOSPOLLIYKVSNRFSGIS

DRFSGSGSGTDFTLKISRVAPTDLGVYYCGQGSQYPFTFGAGTKLELK (SEQ ID NO: 70, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 10, a CDR2-L consisting of the sequence under KVS, and a CDR3-L with the sequence under SEQ ID NO: 71.

The anti-CD3 antibody, called 11F3-B9, contains a heavy chain variable domain consisting of the sequence

- 11 041572

EVOLVETGGSLVOPGKSLTLTCATSGFTFSNAWMHWVROSPEKOLEWVAQIKAKSNNYATYYAE

SVKGRFTISRDDSKRSVYLOMNSLKEEDTAIYYCRYVNYGLAPMDVWGOGTSVTVSS (SEQ ID NO: 72, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 13, CDR2-H with the sequence of SEQ ID NO: 37, and CDR3-H with the sequence under SEQ ID NO: 84, and a light chain variable domain consisting of the sequence

DWMTOTPVSLSVSLGGQVSISCRSSQSLVHDNGNTYLSWYLOKPGOSPOLLIYKVSNRFSGFS

DRFSGSGSGTDFTLKISRVEPDDLGVYYCGQGTQYPFTFGAGTKLELK (SEQ ID NO: 73, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 17, a CDR2-L consisting of the sequence under KVS, and a CDR3-L with the sequence under SEQ ID NO: 11.

The anti-CD3 antibody, called 12G3-E8, contains a heavy chain variable domain consisting of the sequence

EVRWE T GGS LVQ R GKS LKL T CAT SGFTFSLAWMHWVROSPEKKLEWVAQIKDKANNYATYYAE

SVKGRFTISRDDSKRSVYLQMNRLKEEDTAIYYCRGVYYGFSMTPFDYWGOGVMVTVSS (SEQ ID NO: 74, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 75, CDR2-H with the sequence of SEQ ID NO: 76, and CDR3-H with the sequence under SEQ ID NO: 77, and a light chain variable domain consisting of the sequence

DVAMTQTPVSLSVSLGQVSISCRSSQSLVHNNGNTYLSWYLQKPGQSPQVLIYKVSNRFSGIS

DRFSGSGSGADFTLKISRVEPDDLGVYYCGQGT^YPFTFGAGTKLELK (SEQ ID NO: 78, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence of SEQ ID NO: 10, a CDR2-L consisting of the KVS sequence, and a CDR3-L with the sequence under SEQ ID NO: 11.

The anti-CD3 antibody, called 5B1-G2, contains a heavy chain variable domain consisting of the sequence

EVOWETGGSLVOPGKSLKLTCATSGFSFSNAWMHWVROSPEKOLEWVAQIKDKANNYATYYAE

SVKGRFTISRDDSKGSIYLOMNSLKEEDTAVYYCRGLYYGLFPSDYWGOGVMVTVSS (SEQ ID NO: 79, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 80, CDR2-H with the sequence of SEQ ID NO: 76, and CDR3-H with the sequence under SEQ ID NO: 81, and a light chain variable domain consisting of the sequence

DVVMTOTPVSLSVSL GGOVSISCRSSQSLVHNNGNTYL SWYLLKPGOSPOL L IYKVSNRFSGIS

DRFSGGGSGTDFTLKISRLEPDDLGIYYCGQGTQYPFTFGSGTKLEIK (SEQ ID NO: 82, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 10, a CDR2-L consisting of the sequence KVS, and a CDR3-L with the sequence under SEQ ID NO: 11.

An anti-CD3 antibody called 16F8-A7 contains a heavy chain variable domain containing the sequence

VETGGNLVOPGKSLKLTCATSGFTFSNAWMHWVROSPEKOLEWVAQIKAKSNNYATYYAESVKG

RFTISRDDSKSSVYLOMNSLKEEDTAIYYCRYVNYGLAPMDVWGOGTSVTVSS (SEQ ID NO: 83, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 13, CDR2-H with the sequence of SEQ ID NO: 37, and CDR3-H with the sequence under SEQ ID NO: 84, and a light chain variable domain consisting of the sequence

DVVMTOTPVSLSVSLGGOVSISCRSSQSLVHNNGNTYLSWYL OKPGOSPOL L IYKVSNRFSGFS

DRFSGSGSGTDFTLKISRVEPDDLGVYYCGQGTQYPFTFGAGTKLELK with CDRs shown in bold and underlined) containing a CDR1-L with the sequence of SEQ ID NO: 10, a CDR2-L consisting of the sequence of KVS, and a CDR3-L with the sequence of SEQ ID NO: 11.

The anti-CD3 antibody, called 11F9-F8, contains a heavy chain variable domain consisting of the sequence

EVQLVEТGGTLVQPGKSLKLTCATSGFTFSNAWMHWVRQSPEKQLEWVAQIKAKSNNYATYYAE

SVKGRFTISRDDSKTSVYLOLNSLREEDTAIYYCRGTQYGYNPFDYWGOGVMVTVSS (SEQ ID NO: 46, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 13, CDR2-H with the sequence of SEQ ID NO: 37, and CDR3-H with the sequence under SEQ ID NO: 47, and a light chain variable domain consisting of the sequence

- 12 041572

DWLTOTPVSLSVSLGGOVSISCRSSQSLVHNNGNTYLSWSLOKPGOSPOVLIYKVSNRFSGIS

NRFSGSGSGTDFTLKISRVEPDDLGVYYCGQGAHYPFTFGSGTKLEIK (SEQ ID NO: 87, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 10, a CDR2-L consisting of the sequence KVS, and a CDR3-L with the sequence under SEQ ID NO: 88.

An anti-CD3 antibody called 3G5-E10 contains: a heavy chain variable domain consisting of the sequence

EVKLVE S GGGLVQ PGRS LKL 3 CAASGFNFNVYWMGWVROAPGKGLEWIGEIKKDSNSINYTP3 L

KEKFTISRDNAQNTLYLOVNKLGSEDTAIYYCAREERDGYFDYWGOGVMVTVSS (SEQ ID NO: 89, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 90, CDR2-H with the sequence of SEQ ID NO: 91, and CDR3-H with the sequence under SEQ ID NO: 32, and a light chain variable domain consisting of the sequence

NIVMTOSPKSMSMSVGERVTLTCKASENWTYVSWYOOKPEQSPKLLIYGASNRYTGVPDRFTG

SGSATDFTLT13 SVQAEDLADYHCGQGYSYPYTFGGGTKLEIK (SEQ ID NO: 92, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 93, a CDR2-L consisting of the GAS sequence, and a CDR3-L with the sequence under SEQ ID NO: 94.

The anti-CD3 antibody, called 9D7-F3, contains a heavy chain variable domain consisting of the sequence

AVQLVE 3 GGGLVQPKE 3 LKI3 SAAZGFTFSNAAMYWVRQAPGKGLEWVARIRTKPNNYATYYAD

SVTGRFIISRDDSRSMVYLQMDNLOTEDTAMYYCTALISTAMAAWGOGTSVTVSS (SEQ ID NO: 95, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 96, CDR2-H with the sequence of SEQ ID NO: 97, and CDR3-H with the sequence under SEQ ID NO: 98, and a light chain variable domain consisting of the sequence

DIQMTQS PS FLSASVGDRVTINCKASQNINKYLNWYHQMLGEAPKLVISNTNNLQAGIPSRFSG

SGSGTDFTLTISSLQPEDVATYFCLQHRSGYTFGLGTKLELK (SEQ ID NO: 99, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 100, a CDR2-L consisting of the NTN sequence, and a CDR3-L with the sequence under SEQ ID NO: 101.

The anti-CD3 antibody, called 8C2-F7, contains a heavy chain variable domain consisting of the sequence

OIOLVQSGPELKKPGESVKISCKASGYTFFTDFAMNWVKOAPGNGLKWMGWINTQTGKPTYADGF

KORFVFSLETSASTIYLQINNLNIEDTATYFCTRGALASVGOGVLVTVSS (SEQ ID NO: 102, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 103, CDR2-H with the sequence of SEQ ID NO: 104, and CDR3-H with the sequence under SEQ ID NO: 105, and a light chain variable domain consisting of the sequence

DWMTQTPVSLSVSLGSHVSISCRSSQSLVHNNGNTYLSWYLQKPGOSPQPLIYKVSNRFSGIS

DRFSGSGSGTDFTLEINRVEPDDLGVYYCGQGAQYPFTFGSGTKLEIK (SEQ ID NO: 106, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 10, a CDR2-L consisting of the sequence under KVS, and a CDR3-L with the sequence under SEQ ID NO: 11.

An anti-CD3 antibody called 20E5-F10 contains a heavy chain variable domain consisting of the sequence

EVQLVE T GENLVQP GKS LRL T CAT 3GFSFSNAWMHWIRQS PEKQLEWVAQIKDKSNNYATYYAE

SVNGRFTISRDDSKSSIYLHMDNLKEEDSAIYYCRYVHYGVRFFYTMDVWGQGTSVTVSS (SEQ ID NO: 107, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 80, CDR2-H with the sequence of SEQ ID NO: 19, and CDR3-H with the sequence under SEQ ID NO: 108, and a light chain variable domain consisting of the sequence

DWMTOTPVSLSVSLGDOVSISCRPSQSLVHNNGNTYLSWYLOKPGOSPHPLIYKVSNRFSGIS

DRFSGSGSGTDFTLKISRVEPDDLGVYYCGQGTQYPFTFGSGTKLEIK (SEQ ID NO: 109, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 10, a CDR2-L consisting of the sequence under KVS, and a CDR3-L with the sequence under SEQ ID NO: 11.

An anti-CD3 antibody called 20B5-F10 contains a heavy chain variable domain consisting of the sequence

- 13 041572

EVOLVETGGSLVOPGKSLKLTCATSGFTFSNAWMHWVROSPEKQLEWVAQIKAKSNNYATYYAE

SVKGRFTISRDDSIS SVYLOMNNLKEEDTAIYYCRGVYYGFLGMDAWGOGTSVTVSS (SEQ ID NO: 110, with CDRs shown in bold and underlined), containing CDR1-H with the sequence of SEQ ID NO: 13, CDR2-H with the sequence of SEQ ID NO: 37, and CDR3-H with sequence under SEQ ID NO: 111, and a light chain variable domain consisting of the sequence

DWMTOTPVSLSVSLGGOVSISCRSSQRLVHNNGNTYLSWYLOKPGOSPOLLVYKVSNRFSGIS

DRFSGSGSGTDFTLKISRVEPDDLGVYYCGQGTEYPFTFGSGTKLEIK (SEQ ID NO: 112, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 113, a CDR2-L consisting of the sequence KVS, and a CDR3-L with the sequence under SEQ ID NO: 114.

The anti-CD3 antibody, called 6C9-C9, contains a heavy chain variable domain consisting of the sequence

AVQLVE S GGGLVRPKE S LKIS CAASGFTFRNAAMYWVROAPGKGLEWVARIRTQPNNYAKYYAD

SVKDRFTISRDDSKSMVYLQMDNLKTEDTAMYYCTGLVVTAMDAWGOGTSVTVSS (SEQ ID NO: 115, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 116, CDR2-H with the sequence of SEQ ID NO: 117, and CDR3-H with the sequence under SEQ ID NO: 118, and a light chain variable domain consisting of the sequence

DIOMTOSPSFLSASVGDRVTINCKASQNINKYLNWYOQKLGEAPKLLIYVTNNLQTGIPSRFSG

SGSGTDYTLTISSLOPEDVATYFCLQHRSMYTFGTGTKLELK (SEQ ID NO: 119, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence of SEQ ID NO: 100, a CDR2-L consisting of the sequence of VTN, and a CDR3-L with the sequence of SEQ ID NO: 120.

The anti-CD3 antibody, called 3E8-G1, contains a heavy chain variable domain consisting of the sequence

EVOWESGGGLVOPGRSLKLSCAASGFTFSNYYMDWVROAPKKGLEWVATITASGSRIYYPDSV

KGRFTISRDNAKSSLYLLMNSLKSEDTATYYCARERTDAYFDYWGOGVMVTVS5 (SEQ ID NO: 121, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 122, CDR2-H with the sequence of SEQ ID NO: 123, and CDR3-H with the sequence under SEQ ID NO: 124, and a light chain variable domain consisting of the sequence

QFILTOPNSVSTILGSTVKLSCKRSTGNIGTNYVSWYQHHEGRSPTTMIYRDDKRPDGVPDRFS

GSIDRSSNSALLTINNVQTEDEADYFCQSYISGLNPVFGGGSKLTVL (SEQ ID NO: 125, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 126, a CDR2-L consisting of the sequence under RDD, and a CDR3-L with the sequence under SEQ ID NO: 127.

The anti-CD3 antibody, called 3H6-D2, contains a heavy chain variable domain consisting of the sequence

EVQLVEТGGRLVQPGKSLKLTCATSGFTFSNAWMHWVROSPEKOLEWVAQIKDKSNNYATYYAE

SVKGRFTISRDDSKSIIYLQMNSLKEEDTAIYYCRALTYYGYKRDAMDGWGHGTSVTVSS (SEQ ID NO: 128, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 13, CDR2-H with the sequence of SEQ ID NO: 19, and CDR3-H with the sequence under SEQ ID NO: 129, and a light chain variable domain consisting of the sequence

DWMTOTPVSLSVSLGGOVSISCRSSQSLVHNNGNTYLSWYLQKPGOSPOLLIYKVSNRFSGIS

DRFSGSGSGTDFTLKISRVEPDDLGVYYCGQGTQYPFTFGAGTKLELK (SEQ ID NO: 130, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 10, a CDR2-L consisting of the sequence under KVS, and a CDR3-L with the sequence under SEQ ID NO: 11.

The anti-CD3 antibody, called 8H2, contains a heavy chain variable domain consisting of the sequence

QIQLVQSGPELKKPGESVKISCKASGYTFTFAMNWVKOAPGNGLKWMGWINTQTGKPTYADGE

KQRFVFS LE T SAS TIYLQINNLNIEDTATY FCTRGALASVGOGVMVTVSS (SEQ ID NO: 131, with CDRs shown in bold and underlined) containing CDR1-H with the sequence of SEQ ID NO: 103, CDR2-H with the sequence of SEQ ID NO: 104, and CDR3-H with the sequence of SEQ ID NO: 105, and a light chain variable domain consisting of the sequence

- 14 041572

DWMTQTPVSLSVAIGQPASISCKSSQSLVGTSGKTYLNWLLQRPGOSPKRLIYLVSKLDSGIP

DRFSGSGSETDFTLKISRVETDDLGVYYCLQGSHFPLTFGSGTKLEIK (SEQ ID NO: 132, with CDRs shown in bold and underlined) containing a CDR1-L with the sequence under SEQ ID NO: 133, a CDR2-L consisting of the sequence LVS, and a CDR3-L with the sequence under SEQ ID NO: 134.

In an embodiment, an anti-CD3 antibody of the present invention binds to human CD3. In another embodiment, an anti-CD3 antibody of the present invention further binds to Macac fascicularis CD3. In particular, the anti-CD3 antibody of the present invention binds to the extracellular domain of human CD3, or both human and Macac fascicularis CD3. More specifically, the antibody binds to CD3ε. More specifically, the anti-CD3 antibody binds to the extracellular domain of human CD3ε and Macac fascicularis. An anti-CD3 antibody binds to CD3ε when it is present in the form of a complex, such as a CD3ε/δ complex, or when it is present as a single protein, whether it is expressed in an isolated form or present as a soluble extracellular domain or a full-length anchored in the CD3ε membrane, present, for example, in T cells.

The anti-CD3 antibody of the present invention is specific for the human CD3 surface protein, or for both human and Macac fascicularis CD3 proteins, in particular for CD3ε.

In an embodiment, an anti-CD3 antibody of the present invention has a ratio of Macaca fascicularis CD3 affinity to human CD3 affinity (KD(Macaca fascicularis)/KD(human)) that is <10, such as <6, <5, <4, <3, such as <2, <1, or <0.5. Such a polypeptide of the present invention can be used in toxicology studies in monkeys because the toxicity profile observed in monkeys is consistent with the expected potential adverse effects in humans.

In particular, the anti-CD3 antibody of the present invention does not bind to or is not significantly cross-reactive with CD3y and/or CD35 protein(s).

In particular, the antibody does not bind to the extracellular domain of the aforementioned human CD3y and/or CD35 protein(s) and Macac fascicularis or is not significantly cross-reactive with them.

The sequence of the full-length human CD35 protein is available from the Uniprot database under accession number P04234 (SEQ ID NO: 86, as of December 14, 2014). The extracellular domain of human CD35 consists of amino acids at positions 22-105 of SEQ ID NO: 86.

The sequence of the full-length human CD3y protein is available from the Uniprot database under accession number P09693 (SEQ ID NO: 185, as accessed on December 14, 2014). The extracellular domain of human CD3y consists of amino acids at positions 23-116 of SEQ ID NO: 185.

In addition, the anti-CD3 antibody of the present invention has an affinity (KD) for human CD3 or Macac fascicularis CD3, or both, which is <90 nM, <50 nM, or <30 nM, e.g., <20 nM, <10 nM, <8 nM, <6 nM, <4 nM or <2 nM, for example with an affinity of 0.1 to 10 nM, in particular 0.1 to 8 nM or 0.1 to 4 nM.

Affinity for human CD3 or Macac fascicularis CD3 is determined as a KD value by surface plasmon resonance using a soluble recombinant CD3ε/δ complex of human proteins and Macac fascicularis as the capture antigen.

In one example, the binding affinities of an antibody to CD3 are measured by surface plasmon resonance (SPR) using, for example, the Biacore3000 instrument (GE Healthcare). The assay buffer is, for example, HBS-EP (BR-1001-88, GE Healthcare). As an antigen, for example, constructs based on the extracellular domain of human CD3ε and human CD3δ subunits, including a signal peptide, in the form of Fc-containing fusion proteins, as described in the examples, can be used. Alternatively, constructs based on the extracellular domain of Macac fascicularis CD3ε and Macac fascicularis CD3δ subunits, including a signal peptide, in the form of Fc-containing fusion proteins, as described in the examples, can be used as an antigen. Capture of fusion proteins containing human or Macac fascicularis Fc and CD3ε/δ is achieved using, for example, a human antibody capture kit (GE Healthcare). For example, a capture antibody can be coupled to CM5 chips (BR-1001-88, GE Healthcare), for example, up to about 12,000 RU using, for example, an amine coupling kit (BR-100-50, GE Healthcare). CD3e/5Fc fusion proteins are captured at 10 µl/min. up to approximately 70 RU with Rmax values of 30 RU. The binding kinetics for an anti-CD3 antibody can be measured, for example, at 30 μl/min for 240 and 600 s for the association and dissociation phases, respectively. For example, 2-fold dilutions of the anti-CD3 antibody from 3 to 400 nM in assay buffer can be used. Regeneration of the capture surface can be carried out, for example, with a 1 minute injection of, for example, a 3M MgCl ₂ solution at 30 μl/min. For data analysis, for example, the software BIAevaluation-15 041572tion v.4.1 (GE Healthcare) can be used. The data can be approximated as a whole using a 1:1 Langmuir model with mass transfer.

In an embodiment, an anti-CD3 antibody of the present invention also has an apparent EC50, e.g., as determined by a FACS assay on human T cells, that is <60 nM, e.g., <50 nM, <40 nM, <30 nM, <20 nM, or <15 nM. Typically, the apparent EC50 is in the range of 1-60 nM, in particular 1-30 nM, eg 1-20 nM.

In one embodiment, an anti-CD3 antibody of the present invention has a T cell activation that is less than 10%, less than 8%, less than 6%, less than 4%, less than 2%, less than 1%, e.g., less than 0.5% in lack of target cells.

The term T cell activation is used herein to refer to the initiation of CD3 signaling, which entails the fusion of cytotoxic granules, the transient release of cytokines, and proliferation. The antibody-like binding protein and anti-CD3 antibody of the present invention targets CD3ε and activates T cells in the presence of target cells; this activity is also referred to as the T cell recruitment effect. The effect of recruiting T cells induces cytotoxicity against the target cell.

As known to those skilled in the art, T cell activation induces the expression of a surface marker such as CD69 and CD25. T cell activation can thus be measured by detecting and measuring the expression of CD4+/CD25+, CD4+/CD69+, CD8+/CD25+ or CD8+/CD69+ T cells. Methods for measuring T cell activation are known to those skilled in the art.

A method for measuring T cell activation is further disclosed in the examples section (Example 3.3). Accordingly, in the context of the present invention, T cell activation is measured either as the percentage of cells expressing CD69 as % of total cells, or as the percentage of cells expressing CD4 and CD69 as % of total cells, or as the percentage of cells expressing CD8 and CD69, in % of the total number of cells.

Low T cell activation in the context of an anti-CD3 antibody of the present invention refers to less than 10%, less than 8%, less than 6%, less than 4%, less than 2%, less than 1%, e.g., less than 0, 5% T cell activation in the absence of target cells.

Sequence alignment of the VH and VL regions from antibodies to CD3, 20G6-F3, 4B4-D7, 4E7-C9, 18F5-H10, 12D2-E5, 11D7-C3, 11H3-B5, 13H2-C2, 13C1-F6, 18H1Sch10, 1E6-C9, 10F4-C10, 10E6-G6, 18G9-H11, 11F3-B9, 12G3-E8, 5B1-G2, 16F8-A7, 11F9-F8, 3G5-E10, 9D7-F3, 8C2-F7, 20E5- F10, 20B5-F10, 6C9-C9, 3E8-G1, 3H6-D2 and 8H2. Comparison of the CDR-H and CDR-L sequences in most cases indicates that 20G6-F3, 4B4D7, 4E7-C9, 18F5-H10, 11D7-C3, 11H3-E5, 13H2-C2, 13C1-F6, 1E6-C9 , 10F4-C10, 10E6G6, 18G9-H11, 11F3-B9, 12G3-E8, 5B1-G2, 16F8-A7, 11F9-F8, 20E5-F10, 20B5-F10 , and these antibodies are likely to bind to the same epitope. Sequence comparison of CDR-H and CDR-L from these related antibodies is shown in FIG. 1 and 2, respectively. Positions of the CDRs were identified that were strongly conserved across antibodies and thus presumably important in terms of specificity, while the remaining positions could contain substitutions.

Accordingly, the antibody of the present invention contains a heavy chain variable domain containing a CDR1-H, consisting of the sequence GFX1X2X3X4AW (SEQ ID NO: 331), where X1 is T or S, X ₂ is F or V, X ₃ is S or T and X ₄ is N, K, L or Y, or any combination thereof; And

CDR2-H consisting of the sequence IKX1X ₂ X ₃ NX ₄ YX ₅ T (SEQ ID NO: 332), where X1 is A or D, X ₂ is K or R, X ₃ is S or A, X ₄ is N or S and X ₅ is A or E, or any combination thereof; And

CDR3-H consisting of the sequence TWRHYYSSHTMDA (SEQ ID NO: 69) or RALTYYGYKRDAMDG (SEQ ID NO: 129) or RX1X2X3YX4X5X6X7X8X9X10X11DX12 (SEQ ID NO: 333), where X ₁ is Y, G or A, X2 is V, T or L, X3 is H, N, Y or Q, X4 is G, R or

A, X ₅ is F or V or no amino acid, X ₆ is R or no amino acid, X ₇ is F, S, or I or no amino acid, X ₈ is F, L, N, M , Y, S, A or G, X ₉ is Y, A, K, S, N, T, F or L, X ₁₀ is A, P, G or T, X11 is M, L, F or S, and X ₁₂ is A, V, or Y, or any combination thereof, and a light chain variable domain containing CDR1-L, consisting of the sequence QX ₁ LX ₂ HX ₃ NGX ₄ TY (SEQ ID NO: 334), where X ₁ is R or S, X ₂ is V or E, X3 is N, D or T, and X4 is N or Y, or any combination thereof; And

CDR2-L, consisting of the KVS sequence; And

CDR3-L, consisting of the sequence GQGX1X2YPFT (SEQ ID NO: 335), where X1 is T, A or S and X2 is H, E or Q, or any combination thereof.

- 16 041572

In an embodiment, an anti-CD3 antibody of the present invention comprises the CDR sequences from the heavy and/or light chains of one of 28 anti-CD3 antibodies referred to as 20G6-F3, 4B4-D7, 4E7-C9, 18F5-H10, 12D2-E5, 11D7-C3 . , 3G5-E10, 9D7-F3, 8C2-F7, 20E5-F10, 20B5-F10, 6C9-C9, 3E8-G1, 3H6-D2 and 8H2 listed above.

Therefore, the present invention relates to an anti-CD3 antibody that contains:

a) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 6 or a sequence different from SEQ ID NO: 6 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 7 or a sequence different from SEQ ID NO: 7 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 8 or a sequence different from SEQ ID NO: 8 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence under SEQ ID NO: 10 or SEQ ID NO: 142 or a sequence different from SEQ ID NO: 10 or SEQ ID NO: 142 by one amino acid substitution; CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 11 or a sequence different from SEQ ID NO: 11 by one amino acid substitution; or

b) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 13 or a sequence different from SEQ ID NO: 13 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 14 or a sequence different from SEQ ID NO: 14 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 15 or a sequence different from SEQ ID NO: 15 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence under SEQ ID NO: 17 or SEQ ID NO: 184 or a sequence different from SEQ ID NO: 17 or SEQ ID NO: 184 by one amino acid substitution; CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 11 or a sequence different from SEQ ID NO: 11 by one amino acid substitution; or

c) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 13 or a sequence different from SEQ ID NO: 13 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 19 or a sequence different from SEQ ID NO: 19 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 20 or a sequence different from SEQ ID NO: 20 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence under SEQ ID NO: 22 or a sequence different from SEQ ID NO: 22 by one amino acid substitution; CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 11 or a sequence different from SEQ ID NO: 11 by one amino acid substitution; or

d) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 24 or a sequence different from SEQ ID NO: 24 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 19 or a sequence different from SEQ ID NO: 19 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 25 or a sequence different from SEQ ID NO: 25 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence under SEQ ID NO: 27 or a sequence different from SEQ ID NO: 27 by one amino acid substitution;

CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 28 or a sequence different from SEQ ID NO: 28 by one amino acid substitution; or

e) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 30 or a sequence different from SEQ ID NO: 30 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 31 or a sequence different from SEQ ID NO: 31 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 32 or a sequence different from SEQ ID NO: 32 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence under SEQ ID NO: 34 or a sequence different from SEQ ID NO: 34 by one amino acid substitution;

CDR2-L with the RDD sequence or a sequence different from the RDD sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 35 or a sequence different from SEQ ID NO: 35 by one amino acid substitution; or

f) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 13 or a sequence different from SEQ ID NO: 13 by one amino acid substitution;

- 17 041572

CDR2-H with a sequence under SEQ ID NO: 37 or a sequence different from SEQ ID

NO: 37 with one or more amino acid substitutions; CDR3-H with sequence under SEQ

ID NO: 38 or a sequence different from SEQ ID NO: 38 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 10 or a sequence different from SEQ ID NO: 10 by one amino acid substitution;

CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 28 or a sequence different from SEQ ID NO: 28 by one amino acid substitution; or

g) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 13 or a sequence different from SEQ ID NO: 13 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 37 or a sequence different from SEQ ID NO: 37 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 41 or a sequence different from SEQ ID NO: 41 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence under SEQ ID NO: 17 or a sequence different from SEQ ID NO: 17 by one amino acid substitution;

CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 11 or a sequence different from SEQ ID NO: 11 by one amino acid substitution; or

h) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 13 or a sequence different from SEQ ID NO: 13 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 37 or a sequence different from SEQ ID NO: 37 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 44 or a sequence different from SEQ ID NO: 44 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 10 or a sequence different from SEQ ID NO: 10 by one amino acid substitution;

CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 11 or a sequence different from SEQ ID NO: 11 by one amino acid substitution; or

i) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 13 or a sequence different from SEQ ID NO: 13 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 37 or a sequence different from SEQ ID NO: 37 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 47 or a sequence different from SEQ ID NO: 47 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 10 or a sequence different from SEQ ID NO: 10 by one amino acid substitution; CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 11 or a sequence different from SEQ ID NO: 11 by one amino acid substitution; or

j) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 50 or a sequence different from SEQ ID NO: 50 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 51 or a sequence different from SEQ ID NO: 51 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 52 or a sequence different from SEQ ID NO: 52 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence under SEQ ID NO: 54 or a sequence different from SEQ ID NO: 54 by one amino acid substitution;

CDR2-L with the NAN sequence or a sequence different from the NAN sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 55 or a sequence different from SEQ ID NO: 55 by one amino acid substitution; or

k) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 57 or a sequence different from SEQ ID NO: 57 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 37 or a sequence different from SEQ ID NO: 37 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 58 or a sequence different from SEQ ID NO: 58 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 10 or a sequence different from SEQ ID NO: 10 by one amino acid substitution; CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 28 or a sequence different from SEQ ID NO: 28 by one amino acid substitution; or

l) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 13 or a sequence different from SEQ ID NO: 13 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 37 or a sequence different from SEQ ID NO: 37 of one or

- 18 041572 with several amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 61 or a sequence different from SEQ ID NO: 61 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 10 or a sequence different from SEQ ID NO: 10 by one amino acid substitution; CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 11 or a sequence different from SEQ ID NO: 11 by one amino acid substitution; or

m) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 64 or a sequence different from SEQ ID NO: 64 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 65 or a sequence different from SEQ ID NO: 65 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 47 or a sequence different from SEQ ID NO: 47 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence under SEQ ID NO: 67 or a sequence different from SEQ ID NO: 67 by one amino acid substitution; CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 28 or a sequence different from SEQ ID NO: 28 by one amino acid substitution; or

n) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 13 or a sequence different from SEQ ID NO: 13 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 37 or a sequence different from SEQ ID NO: 37 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 69 or a sequence different from SEQ ID NO: 69 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 10 or a sequence different from SEQ ID NO: 10 by one amino acid substitution; CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 71 or a sequence different from SEQ ID NO: 71 by one amino acid substitution; or

o) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 13 or a sequence different from SEQ ID NO: 13 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 37 or a sequence different from SEQ ID NO: 37 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 84 or a sequence different from SEQ ID NO: 84 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence under SEQ ID NO: 17 or a sequence different from SEQ ID NO: 17 by one amino acid substitution;

CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 11 or a sequence different from SEQ ID NO: 11 by one amino acid substitution; or

p) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 75 or a sequence different from SEQ ID NO: 75 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 76 or a sequence different from SEQ ID NO: 76 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 77 or a sequence different from SEQ ID NO: 77 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 10 or a sequence different from SEQ ID NO: 10 by one amino acid substitution;

CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 11 or a sequence different from SEQ ID NO: 11 by one amino acid substitution; or

q) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 80 or a sequence different from SEQ ID NO: 80 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 76 or a sequence different from SEQ ID NO: 76 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 81 or a sequence different from SEQ ID NO: 81 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 10 or a sequence different from SEQ ID NO: 10 by one amino acid substitution;

CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 11 or a sequence different from SEQ ID NO: 11 by one amino acid substitution; or

r) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 13 or a sequence different from SEQ ID NO: 13 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 37 or a sequence different from SEQ ID NO: 37 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 84 or a sequence different from SEQ ID NO: 84 by one amino acid substitution; And

- 19 041572 a light chain variable domain containing a CDR1-L with a sequence under SEQ ID NO: 10 or a sequence different from SEQ ID NO: 10 by one amino acid substitution;

CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 11 or a sequence different from SEQ ID NO: 11 by one amino acid substitution; or

s) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 13 or a sequence different from SEQ ID NO: 13 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 37 or a sequence different from SEQ ID NO: 37 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 47 or a sequence different from SEQ ID NO: 47 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 10 or a sequence different from SEQ ID NO: 10 by one amino acid substitution;

CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 88 or a sequence different from SEQ ID NO: 88 by one amino acid substitution; or

t) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 90 or a sequence different from SEQ ID NO: 90 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 91 or a sequence different from SEQ ID NO: 91 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 32 or a sequence different from SEQ ID NO: 32 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence under SEQ ID NO: 93 or a sequence different from SEQ ID NO: 93 by one amino acid substitution; CDR2-L with a GAS sequence or a sequence different from the GAS sequence by one amino acid substitution, and CDR3-L with a sequence under SEQ ID NO: 94 or a sequence different from SEQ ID NO: 94 by one amino acid substitution; or

u) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 96 or a sequence different from SEQ ID NO: 96 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 97 or a sequence different from SEQ ID NO: 97 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 98 or a sequence different from SEQ ID NO: 98 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with a sequence under SEQ ID NO: 100 or a sequence different from SEQ ID NO: 100 by one amino acid substitution;

CDR2-L with the NTN sequence or a sequence different from the NTN sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 101 or a sequence different from SEQ ID NO: 101 by one amino acid substitution; or

v) a heavy chain variable domain containing a CDR1-H with the sequence of SEQ ID NO: 103 or a sequence different from SEQ ID NO: 103 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 104 or a sequence different from SEQ ID NO: 104 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 105 or a sequence different from SEQ ID NO: 105 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 10 or a sequence different from SEQ ID NO: 10 by one amino acid substitution; CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 11 or a sequence different from SEQ ID NO: 11 by one amino acid substitution; or

w) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 80 or a sequence different from SEQ ID NO: 8 0 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 19 or a sequence different from SEQ ID NO: 19 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 108 or a sequence different from SEQ ID NO: 108 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 10 or a sequence different from SEQ ID NO: 10 by one amino acid substitution; CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 11 or a sequence different from SEQ ID NO: 11 by one amino acid substitution; or

x) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 13 or a sequence different from SEQ ID NO: 13 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 37 or a sequence different from SEQ ID NO: 37 by one or more amino acid substitutions; CDR3-H with sequence under SEQ

- 20 041572

ID NO: 111 or a sequence different from SEQ ID NO: 111 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 113 or a sequence different from SEQ ID NO: 113 by one amino acid substitution; CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 114 or a sequence different from SEQ ID NO: 114 by one amino acid substitution; or

y) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 116 or a sequence different from SEQ ID NO: 116 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 117 or a sequence different from SEQ ID NO: 117 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 118 or a sequence different from SEQ ID NO: 118 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with a sequence under SEQ ID NO: 100 or a sequence different from SEQ ID NO: 100 by one amino acid substitution; CDR2-L with the VIN' sequence or a sequence different from the VTN' sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 120 or a sequence different from SEQ ID NO: 120 by one amino acid substitution; or

z) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 122 or a sequence different from SEQ ID NO: 122 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 123 or a sequence different from SEQ ID NO: 123 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 124 or a sequence different from SEQ ID NO: 124 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence under SEQ ID NO: 126 or a sequence different from SEQ ID NO: 126 by one amino acid substitution; CDR2-L with the RDD sequence or a sequence different from the RDD sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 127 or a sequence different from SEQ ID NO: 127 by one amino acid substitution; or aa) a heavy chain variable domain comprising a CDR1-H with the sequence of SEQ ID NO: 13 or a sequence different from SEQ ID NO: 13 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 19 or a sequence different from SEQ ID NO: 19 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 129 or a sequence different from SEQ ID NO: 12 by 9 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 10 or a sequence different from SEQ ID NO: 10 by one amino acid substitution; CDR2-L with the KVS sequence or a sequence different from the KVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 11 or a sequence different from SEQ ID NO: 11 by one amino acid substitution; or bb) a heavy chain variable domain containing a CDR1-H with the sequence of SEQ ID NO: 103 or a sequence different from SEQ ID NO: 103 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 104 or a sequence different from SEQ ID NO: 104 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 105 or a sequence different from SEQ ID NO: 105 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 133 or a sequence different from SEQ ID NO: 133 by one amino acid substitution;

CDR2-L with the LVS sequence or a sequence differing from the LVS sequence by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 134 or a sequence differing from SEQ ID NO: 134 by one amino acid substitution.

One or more individual amino acids can be changed by introducing a substitution, in particular a conservative substitution, into one or more of the above CDR sequences. Such a change may involve, for example, the removal of a glycosylation site or a deamidation site in combination with humanization of the antibody.

Based on sequence alignment of VH and VL regions from 20G6-F3, 4B4-D7, 4E7C9, 18F5-H10, 11D7-C3, 11H3-E5, 13H2-C2, 13C1-F6, 1E6-C9, 10F4-C10, 10E6-G6 , 18G9H11, 11F3-B9, 12G3-E8, 5B1-G2, 16F8-A7, 11F9-F8, 20E5-F10, 20B5-F10 and 3H6-D2, various amino acid substitutions have been identified. Therefore, in one embodiment, the amino acid is substituted:

in CDR1-H, in one or more of positions 3-6, such as position 3, 5 or 6 of CDR1-H

- 21 041572 with the sequence GFTFSNAW (SEQ ID NO: 13), or, for example, at positions 4 and 5 in CDR1-H with the sequence GFTFSNAW (SEQ ID NO: 13); and/or in CDR2-H, in one or more of positions 3-5, 7 and 9, for example, positions 3, 4, 5, 7 or 9 of CDR2-H with the sequence IKAKSNNYAT (SEQ ID NO: 37), or, for example, at positions 3 and 5 or 3 and 7 in CDR2-H with the sequence IKAKSNNYAT (SEQ ID NO: 37); and/or in CDR3-H, in one or more of positions 2-4, 7-10 and 12, such as positions 2, 3, 4, 6, 7, 8, 9 and 10, or positions 2, 4 , 6, 8, 10, and 12, or at positions 2, 3, 7, and 8 of CDR3-H with the sequence RGVYYALSPFDY (SEQ ID NO: 8), or at positions 7 and 8 of CDR3-H with the sequence RGLYYGLSPSDY (SEQ ID NO: 38), or at positions 2, 3, 4, 6, 7, 8, 9 and 10 in CDR3-H with the sequence RGLYYGLSPSDY (SEQ ID NO: 38); and/or in CDR1-L, at one or more of positions 2, 4, 6 and 9, e.g. at position 6 of CDR1-L with the sequence QSLVHDNGNTY (SEQ ID NO: 17) or QSLVHTNGNTY (SEQ ID NO: 27) , or at position 2 in CDR1-L with the sequence QSLVHNNGNTY (SEQ ID NO: 10) or QRLVHNNGNTY (SEQ ID NO: 113), or at positions 4, 6 and 9 in CDR1-L with the sequence QSLVHNNGNTY (SEQ ID NO: 10 ); and/or in CDR3-L, at one or more of positions 4 and 5, for example at position 4 of CDR3-L with the sequence GQGSQYPFT (SEQ ID NO: 71) or GQGTQYPFT (SEQ ID NO: 11), or at position 4 and 5 in CDR3-L with the sequence GQGAHYPFT (SEQ ID NO: 88) or at position 5 in CDR3-L with the sequence GQGTHYPFT (SEQ ID NO: 28) or GQGTEYPFT (SEQ ID NO: 114).

The anti-CD3 antibody of the present invention is particularly a conventional antibody, such as a conventional monoclonal antibody or antibody fragment, a bispecific or polyspecific antibody.

The anti-CD3 antibody of the present invention specifically comprises or consists of IgG or a fragment thereof.

According to a further embodiment, the present invention provides an anti-CD3 antibody which comprises:

a) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 6, a CDR2-H of the sequence of SEQ ID NO: 7, a CDR3-H of the sequence of SEQ ID NO: 8, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 10 or SEQ ID NO: 142, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

b) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 13, a CDR2-H of the sequence of SEQ ID NO: 14, a CDR3-H of the sequence of SEQ ID NO: 15, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 17 or SEQ ID NO: 184, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

c) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 13, a CDR2-H of the sequence of SEQ ID NO: 19, a CDR3-H of the sequence of SEQ ID NO: 20, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 22, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

d) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 24, a CDR2-H of the sequence of SEQ ID NO: 19, a CDR3-H of the sequence of SEQ ID NO: 25, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 27, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 28; or

e) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 30, a CDR2-H of the sequence of SEQ ID NO: 31, a CDR3-H of the sequence of SEQ ID NO: 32, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 34, CDR2-L with the sequence RDD and CDR3-L with the sequence under SEQ ID NO: 35; or

f) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 13, a CDR2-H of the sequence of SEQ ID NO: 37, a CDR3-H of the sequence of SEQ ID NO: 38, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 10, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 28; or

g) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 13, a CDR2-H of the sequence of SEQ ID NO: 37, a CDR3-H of the sequence of SEQ ID NO: 41, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 17, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

- 22 041572

h) a heavy chain variable domain containing a CDR1-H with the sequence of SEQ ID

NO: 13, CDR2-H with sequence under SEQ ID NO: 37, CDR3-H with sequence under SEQ

ID NO: 44, and a light chain variable domain containing CDR1-L with the sequence under SEQ ID NO: 10,

CDR2-L with the sequence KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

i) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 13, a CDR2-H of the sequence of SEQ ID NO: 37, a CDR3-H of the sequence of SEQ ID NO: 47, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 10, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

j) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 50, a CDR2-H of the sequence of SEQ ID NO: 51, a CDR3-H of the sequence of SEQ ID NO: 52, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 54, CDR2-L with the sequence NAN and CDR3-L with the sequence under SEQ ID NO: 55; or

k) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 57, a CDR2-H of the sequence of SEQ ID NO: 37, a CDR3-H of the sequence of SEQ ID NO: 58, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 10, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 28; or

l) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 13, a CDR2-H of the sequence of SEQ ID NO: 37, a CDR3-H of the sequence of SEQ ID NO: 61, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 10, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

m) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 64, a CDR2-H of the sequence of SEQ ID NO: 65, a CDR3-H of the sequence of SEQ ID NO: 47, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 67, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 28; or

n) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 13, a CDR2-H of the sequence of SEQ ID NO: 37, a CDR3-H of the sequence of SEQ ID NO: 69, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 10, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 71; or

o) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 13, a CDR2-H of the sequence of SEQ ID NO: 37, a CDR3-H of the sequence of SEQ ID NO: 84, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 17, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

p) a heavy chain variable domain containing a CDR1-H with the sequence of SEQ ID NO: 75, a CDR2-H with the sequence of SEQ ID NO: 76, a CDR3-H with the sequence of SEQ ID NO: 77, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 10, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

q) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 80, a CDR2-H of the sequence of SEQ ID NO: 76, a CDR3-H of the sequence of SEQ ID NO: 81, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 10, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

r) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 13, a CDR2-H of the sequence of SEQ ID NO: 37, a CDR3-H of the sequence of SEQ ID NO: 84, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 10, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

s) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 13, a CDR2-H of the sequence of SEQ ID NO: 37, a CDR3-H of the sequence of SEQ ID NO: 47, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 10, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 88; or

t) heavy chain variable domain containing CDR1-H with the sequence under SEQ ID NO: 90, CDR2-H with the sequence under SEQ ID NO: 91, CDR3-H with the sequence under SEQ

- 23 041572

ID NO: 32, and a light chain variable domain containing CDR1-L with the sequence under SEQ ID

NO: 93, CDR2-L with the sequence GAS and CDR3-L with the sequence under SEQ ID NO: 94;

or

u) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 96, a CDR2-H of the sequence of SEQ ID NO: 97, a CDR3-H of the sequence of SEQ ID NO: 98, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 100, CDR2-L with the sequence of NTN and CDR3-L with the sequence under SEQ ID NO: 101; or

v) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 103, a CDR2-H of the sequence of SEQ ID NO: 104, a CDR3-H of the sequence of SEQ ID NO: 105, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 10, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

w) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 80, a CDR2-H of the sequence of SEQ ID NO: 19, a CDR3-H of the sequence of SEQ ID NO: 108, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 10, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

x) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 13, a CDR2-H of the sequence of SEQ ID NO: 37, a CDR3-H of the sequence of SEQ ID NO: 111, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 113, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 114; or

y) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 116, a CDR2-H of the sequence of SEQ ID NO: 117, a CDR3-H of the sequence of SEQ ID NO: 118, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 100, CDR2-L with the sequence 'VTN' and CDR3-L with the sequence under SEQ ID NO: 120; or

z) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 122, a CDR2-H of the sequence of SEQ ID NO: 123, a CDR3-H of the sequence of SEQ ID NO: 124, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 126, CDR2-L with the sequence RDD and CDR3-L with the sequence under SEQ ID NO: 127; or aa) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 13, a CDR2-H of the sequence of SEQ ID NO: 19, a CDR3-H of the sequence of SEQ ID NO: 129, and a light chain variable domain containing CDR1-L with the sequence under SEQ ID NO: 10, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or bb) a heavy chain variable domain comprising CDR1-H of SEQ ID NO: 103, CDR2-H of SEQ ID NO: 104, CDR3-H of SEQ ID NO: 105; and a light chain variable domain comprising CDR1-L with the sequence of SEQ ID NO: 133, CDR2-L with the sequence of LVS, and CDR3-L with the sequence of SEQ ID NO: 134.

The present invention also provides an anti-CD3 antibody comprising at least a heavy chain variable domain and/or a light chain variable domain of one of the anti-CD3 antibodies listed above.

Thus, the present invention relates in particular to an anti-CD3 antibody which contains:

a) a heavy chain variable domain with a sequence of SEQ ID NO: 5 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 9 or a sequence that is at least 85% identical to her; or

b) a heavy chain variable domain with a sequence of SEQ ID NO: 12 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 16 or a sequence that is at least 85% identical to her; or

c) a heavy chain variable domain with a sequence of SEQ ID NO: 18 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 21 or a sequence that is at least 85% identical to her; or

d) a heavy chain variable domain with the sequence of SEQ ID NO: 23 or a sequence that is at least 85% identical thereto and/or a light chain variable domain with

- 24 041572 sequence under SEQ ID NO: 26 or a sequence that is at least 85% identical to it; or

e) a heavy chain variable domain with a sequence of SEQ ID NO: 29 or a sequence that is at least 85% identical to it, and/or a light chain variable domain with a sequence of SEQ ID NO: 33 or a sequence that is at least 85% identical to her; or

f) a heavy chain variable domain with a sequence of SEQ ID NO: 36 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 39 or a sequence that is at least 85% identical to her; or

g) a heavy chain variable domain with a sequence of SEQ ID NO: 40 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 42 or a sequence that is at least 85% identical to her; or

h) a heavy chain variable domain with a sequence of SEQ ID NO: 43 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 45 or a sequence that is at least 85% identical to her; or

i) a heavy chain variable domain with a sequence of SEQ ID NO: 46 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 48 or a sequence that is at least 85% identical to her; or

j) a heavy chain variable domain with a sequence of SEQ ID NO: 49 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 53 or a sequence that is at least 85% identical to her; or

k) a heavy chain variable domain with a sequence of SEQ ID NO: 56 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 59 or a sequence that is at least 85% identical to her; or

l) a heavy chain variable domain with a sequence of SEQ ID NO: 60 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 62 or a sequence that is at least 85% identical to her; or

m) a heavy chain variable domain with a sequence of SEQ ID NO: 63 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 66 or a sequence that is at least 85% identical to her; or

n) a heavy chain variable domain with a sequence of SEQ ID NO: 68 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 70 or a sequence that is at least 85% identical to her; or

o) a heavy chain variable domain with a sequence of SEQ ID NO: 72 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 73 or a sequence that is at least 85% identical to her; or

p) a heavy chain variable domain with a sequence of SEQ ID NO: 74 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 78 or a sequence that is at least 85% identical to her; or

q) a heavy chain variable domain with a sequence of SEQ ID NO: 79 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 82 or a sequence that is at least 85% identical to her; or

r) a heavy chain variable domain comprising a sequence of SEQ ID NO: 83 or a sequence that is at least 85% identical thereto and/or a light chain variable domain having a sequence of SEQ ID NO: 85 or a sequence that is at least 85% identical to her; or

s) a heavy chain variable domain with a sequence of SEQ ID NO: 46 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 87 or a sequence that is at least 85% identical to her; or

- 25 041572

t) a heavy chain variable domain with a sequence of SEQ ID NO: 89 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 92 or a sequence that is at least 85% identical to her; or

u) a heavy chain variable domain with a sequence of SEQ ID NO: 95 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 99 or a sequence that is at least 85% identical to her; or

v) a heavy chain variable domain with a sequence of SEQ ID NO: 102 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 106 or a sequence that is at least 85% identical thereto. 85% identical to her; or

w) a heavy chain variable domain with a sequence of SEQ ID NO: 107 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 109 or a sequence that is at least 85% identical to her; or

x) a heavy chain variable domain with a sequence of SEQ ID NO: 110 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 112 or a sequence that is at least 85% identical to her; or

s) a heavy chain variable domain with a sequence of SEQ ID NO: 115 or a sequence that is at least 85% identical to it, and/or a light chain variable domain with a sequence of SEQ ID NO: 119 or a sequence that is at least 85% identical to her; or

z) a heavy chain variable domain with a sequence of SEQ ID NO: 121 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 125 or a sequence that is at least 85% identical to her; or aa) a heavy chain variable domain with the sequence of SEQ ID NO: 128 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with the sequence of SEQ ID NO: 130 or a sequence that is at least 85% identical to her; or bb) a heavy chain variable domain having the sequence of SEQ ID NO: 131 or a sequence that is at least 85% identical thereto and/or a light chain variable domain having the sequence of SEQ ID NO: 132 or a sequence that is at least 85% identical to her.

For example, the heavy or light chain variable domain sequence may differ from the reference sequence under SEQ ID NOs: 5, 9, 12, 16, 18, 21, 23, 26, 29, 33, 36, 39, 40, 42, 43, 45 , 46, 48, 49, 53, 56, 59, 60, 62, 63, 66, 68, 70, 72, 73, 74, 78, 79, 82, 83, 85, 87, 89, 92, 95, 99 , 102, 106, 107, 109, 110, 112, 115, 119, 121, 125, 128, 130, 131, 132, optionally with one or more amino acid substitutions, in particular one or more conservative amino acid substitutions and/or replacements for canonical remainders. In particular, the sequence of the heavy or light chain variable domain may differ from the reference sequence under SEQ ID NOs: 5, 9, 12, 16, 18, 21, 23, 26, 29, 33, 36, 39, 40, 42, 43, 45, 46, 48, 49, 53, 56, 59, 60, 62, 63, 66, 68, 70, 72, 73, 74, 78, 79, 82, 83, 85, 87, 89, 92, 95, 99, 102, 106, 107, 109, 110, 112, 115, 119, 121, 125, 128, 130, 131, 132 only conservative amino acid substitution(s).

Sequence changes compared to the sequence under SEQ ID NO: 5, 9, 12, 16, 18, 21, 23, 26, 29, 33, 36, 39, 40, 42, 43, 45, 46, 48, 49, 53 , 56, 59, 60, 62, 63, 66, 68, 70, 72, 73, 74, 78, 79, 82, 83, 85, 87, 89, 92, 95, 99, 102, 106, 107, 109 , 110, 112, 115, 119, 121, 125, 128, 130, 131, 132 in particular will be present primarily in one or more framework regions, FR1-L, FR2-L, FR3-L, FR4-L and/or FR1-H, FR2-H, FR3-H, FR4-H.

However, amino acid substitutions in one or more CDRs are also possible. In particular, the sequence of the light chain variable domain may differ from the sequence under SEQ ID NO: 9 by at least replacing S with R at position 28 of SEQ ID NO: 9 (in CDR1-L), and/or at least replacing V with E at position 30 of SEQ ID NO: 9 (in CDR1-L), and/or at least replacing N with D or T at position 33 of SEQ ID NO: 9 (in CDR1-L), and/or at least by changing N to Y at position 35 of SEQ ID NO: 9 (in CDR1-L), and/or the sequence of the light chain variable domain may differ from the sequence of SEQ ID NO: 9 by at least changing T to S or A at position 97 of SEQ ID NO: 9 (in CDR3-L), and/or at least replacing Q with H or E at position 98 of SEQ ID NO: 9 (in CDR3-L), and/or the heavy chain variable domain may differ from the sequence under SEQ ID NO: 5 by at least replacing T with N or S at position 28 of SEQ ID NO: 5 (in CDR1-H), and/or at least replacing F with V at position 29 of SEQ ID NO: 5 (in CDR1-H), or according

- 26 041572 at least replacing T with N or Y at position 30 of SEQ ID NO: 5 (in CDR1-H), or at least replacing K with L or Y at position 31 of SEQ ID NO: 5 (in CDR1- H), and/or the heavy chain variable domain may differ from the sequence under SEQ ID NO: 5 by at least changing D to A at position 53 of SEQ ID NO: 5 (in CDR2-H), or at least changing K to R at position 54 of SEQ ID NO: 5 (in CDR2-H), or at least a substitution of S for A at position 55 of SEQ ID NO: 5 (in CDR2-H), or at least a substitution of S for N in position 57 of SEQ ID NO: 5 (in CDR2-H), or at least a substitution of A for E at position 59 of SEQ ID NO: 5 (in CDR2-H), and/or the heavy chain variable domain may differ from the sequence under SEQ ID NO: 5 by at least replacing G with A at position 100 of SEQ ID NO: 5 (in CDR3-H) and/or at least replacing T with V at position 101 of SEQ ID NO: 5 (in CDR3 -H) by at least replacing Q with Y at position 102 of SEQ ID NO: 5 (in CDR3-H).

In one embodiment, an anti-CD3 antibody of the present invention and a fragment thereof are a rat antibody and a rat antibody fragment, respectively.

In one aspect of the present invention, the anti-CD3 antibody of the present invention may also be a chimeric antibody, and in particular a rat/human antibody, for example an antibody comprising rat heavy and light chain variable domains and a CH domain and a CL domain from a human antibody.

In a further aspect of the present invention, an anti-CD3 antibody can also be a humanized antibody or a humanized antibody fragment obtained, for example, by CDR grafting or by a 4D method (US 20110027266).

Accordingly, in one embodiment, an anti-CD3 antibody of the present invention is a humanized antibody comprising:

a) a heavy chain variable domain comprising an amino acid sequence selected from the group consisting of SEQ ID NO: 135, SEQ ID NO: 136, SEQ ID NO: 137, SEQ ID NO: 138, SEQ ID NO: 144, SEQ ID NO : 145, SEQ ID NO: 146, SEQ ID NO: 147, SEQ ID NO: 152, SEQ ID NO: 153, and a light chain variable domain containing an amino acid sequence selected from the group consisting of SEQ ID NO: 139, SEQ ID NO: 140, SEQ ID NO: 141, SEQ ID NO: 143, SEQ ID NO: 148, SEQ ID NO: 149, SEQ ID NO: 150, SEQ ID NO: 151, SEQ ID NO: 154, SEQ ID NO : 155, SEQ ID NO: 156, SEQ ID NO: 157, or

b) a heavy chain variable domain comprising an amino acid sequence selected from the group consisting of SEQ ID NO: 171, SEQ ID NO: 172, SEQ ID NO: 173, SEQ ID NO: 174, SEQ ID NO: 175, SEQ ID NO : 176, SEQ ID NO: 177, SEQ ID NO: 178, SEQ ID NO: 179, SEQ ID NO: 180, SEQ ID NO: 181, SEQ ID NO: 183, and a light chain variable domain containing the amino acid sequence selected from the group consisting of SEQ ID NO: 158, SEQ ID NO: 159, SEQ ID NO: 160, SEQ ID NO: 161, SEQ ID NO: 162, SEQ ID NO: 163, SEQ ID NO: 164, SEQ ID NO : 165, SEQ ID NO: 166, SEQ ID NO: 167, SEQ ID NO: 168, SEQ ID NO: 169, SEQ ID NO: 170, SEQ ID NO: 182.

In a further embodiment, an anti-CD3 antibody of the present invention is a humanized antibody comprising:

a) a heavy chain variable domain with the sequence of SEQ ID NO: 138 and/or a light chain variable domain with the sequence of SEQ ID NO: 143; or

b) a heavy chain variable domain with the sequence of SEQ ID NO: 171 and/or a light chain variable domain with the sequence of SEQ ID NO: 158; or

c) a heavy chain variable domain of the sequence of SEQ ID NO: 176 and/or a light chain variable domain of the sequence of SEQ ID NO: 164.

In one embodiment, an anti-CD3 antibody of the present invention comprises three CDR sequences or a heavy chain variable domain, or six CDR sequences or heavy and light chain variable domains of one of the anti-CD3 antibodies listed above.

The present invention further relates to a humanized anti-CD3 antibody fragment as defined above. In one embodiment, the humanized anti-CD3 antibody described above is a chimeric antibody.

An anti-CD3 antibody of the present invention may also be a single domain antibody or a fragment thereof. In particular, a fragment of a single domain antibody may consist of a heavy chain variable region (VHH) that contains the CDR1-H, CDR2-H and CDR3-H of one of the antibodies described above. The anti-CD3 antibody can also be a heavy chain-only antibody, ie. an antibody having no light chain, which may or may not contain a CH1 domain.

The single domain antibody or fragment thereof may also comprise framework regions from the camelid single domain antibody and optionally a constant domain from the camelid single domain antibody.

An anti-CD3 antibody of the present invention may also be an antibody fragment, in particular a humanized antibody fragment, selected from the group consisting of Fv, Fab, F(ab')2, Fab', dsFv, (dsFv)2, scFv, sc(Fv )2 and diatel.

- 27 041572

Accordingly, an anti-CD3 antibody of the present invention is a Fab containing or consisting of:

a) the heavy chain amino acid sequence of SEQ ID NO: 186 and/or the light chain amino acid sequence of SEQ ID NO: 187; or

b) the heavy chain amino acid sequence of SEQ ID NO: 188 and/or the light chain amino acid sequence of SEQ ID NO: 189; or

c) the heavy chain amino acid sequence of SEQ ID NO: 190 and/or the light chain amino acid sequence of SEQ ID NO: 191; or

d) the heavy chain amino acid sequence of SEQ ID NO: 192 and/or the light chain amino acid sequence of SEQ ID NO: 193.

In an embodiment, an anti-CD3 antibody is a bispecific or multispecific antibody formed from at least one antibody fragment or at least one variable domain of an anti-CD3 antibody of the present invention. Multispecific antibodies are polyvalent protein complexes, as described, for example, in EP 2050764 A1 or US 2005/0003403 A1.

The bispecific or polyspecific anti-CD3 antibodies of the present invention can be characterized by specificity for (a) the extracellular domain of human or human CD3 and Macac fascicularis targeted by one of the anti-CD3 antibodies described above, and (b) at least one other antigen.

In one embodiment, the other antigen is CD123 and, accordingly, the resulting bispecific antibody is an anti-CD3/CD123 bispecific antibody. Traditional bispecific antibodies can be obtained using techniques that are known to the person skilled in the art.

Anti-CD3 antibodies and fragments of the present invention can be used as isolated (eg, purified) molecules or contained in a carrier such as a membrane or lipid vesicle (eg, liposome).

In one additional embodiment, an anti-CD3 antibody of the present invention is for use in making an antibody-like binding protein of the present invention, further defined in the Antibody-Like Binding Protein section.

Any combination of the above embodiments forms part of the present invention.

Antibodies to CD123.

CD123 (differentiation cluster 123) is also known as interleukin 3 receptor alpha (IL3RA) or IL3R, IL3RX, IL3RY, IL3RAY, hIL-3Ra and denotes an interleukin 3 specific heterodimeric cytokine receptor subunit. The functional interleukin 3 receptor is a heterodimer that contains a specific alpha chain (IL-3A; CD123) and a beta chain of the IL-3 receptor (β ₀ ; CD 131), which is shared with receptors for granulocyte-macrophage colony-stimulating factor (GM-CSF) and interleukin 5 (IL-5). CD123 is an integral type I transmembrane protein with a predicted molecular weight of approximately 43 kDa, which contains an extracellular domain involved in IL3 binding, a transmembrane domain, and a short cytoplasmic tail of approximately 50 amino acids. The extracellular domain consists of two regions: an N-terminal region of approximately 100 amino acids, the sequence of which shows similarity to the equivalent regions of the alpha chains of the GM-CSF and IL-5 receptors; and a region near the transmembrane domain that contains four conserved cysteine residues and a WSXWS motif common to other members of this cytokine receptor family. The IL-3 binding domain contains cytokine receptor motifs (CRMs) of approximately 200 amino acid residues, which are composed of two Ig-like folded domains. The extracellular domain of CD123 is highly glycosylated, with N-glycosylation required for both ligand binding and receptor signaling. The family of proteins includes three representatives: IL3RA (CD123A), CSF2RA and IL5RA. The overall structure of the three members is substantially conserved, but the degrees of sequence homology are very low. So far, one isoform of CD123 has been discovered, 300 amino acids long, but only at the RNA level, which is available in the Getentry database under accession number ACM24116.1.

US Pat. No. 6,177,078 discloses the monoclonal antibody 7G3 to the IL-3 receptor alpha chain (IL-3Ra, CD123) and the ability of 7G3 to bind to the N-terminal domain, namely amino acid residues 19-49, of IL-3Ra. US Pat. No. 6,733,743 discloses a method of attenuating a hemablastosis progenitor cell that expresses CD123 but does not significantly express CD131 by contacting the cell with a composition of an antibody and a cytotoxic agent (selected from a chemotherapeutic agent, a toxin, or an alpha particle emitting radioisotope), wherein the composition selectively binds to CD123 in an amount effective to cause cell death. However, it remained unclear whether CD123 targeting could lead to functional impairment in AML-LSC.

The reference sequence of the full-length human CD123 protein, including the signal peptide, is available from the NCBI database under accession number NP_002174.1 and from Uniprot under accession number P26951, and is also disclosed herein under SEQ ID NO: 194 (as available on 14 December 2014). The reference sequence of the full-length Macac fascicularis CD123 protein, including the signal peptide, is available from the GenBank database under accession number EH61867.1 and from Uniprot under accession number G8F3K3, and is also disclosed herein under SEQ ID NO: 195 (as available at 14 December 2014).

The sequence of a fusion protein containing Fc and mature human CD123 labeled with Strep-II cloned by the present inventors from genomic DNA is disclosed under SEQ ID NO: 196. Said fusion protein containing Fc and mature human CD123 contains amino acids 19-305 from the full human CD123 protein and thus contains the extracellular domain of human CD123.

The sequence of a fusion protein containing Fc and mature Macac fascicularis CD123 labeled with Strep-II cloned by the present inventors from cDNA is disclosed under SEQ ID NO: 197. Said fusion protein containing Fc and mature Macac fascicularis CD123 contains amino acids 19-305 of full-length Macac fascicularis CD123 protein and thus contains the extracellular domain of Macac fascicularis CD123.

The domain organization of human CD123 and Macac fascicularis is as follows (based on the sequence of human CD123 available from the NCBI database under accession number NP_002174.1 (SEQ ID NO: 194) and based on the CD123 sequence of Macac fascicularis available from the Uniprot database under accession number G8F3K3, SEQ ID NO: 195).

Positions in SEQ ID Positions in SEQ ID

CD123 domains

NO: 194 NO: 195 human(s) (Masasa fascicularis)

Extracellular 19-305 19-305 transmembrane 306-325 306-325 domain cytoplasmic 326-378 326-378

Accordingly, the extracellular domain of human CD123 consists of amino acids at positions 19305 of SEQ ID NO: 194.

CD123 (interleukin-3 receptor alpha chain, IL-3Ra) is a tumor antigen that is overexpressed in various hematological neoplasms. Most blast cells in AML express CD123 on the surface, and this expression does not vary between AML subtypes. Higher expression of CD123 in AML at diagnosis has been reported to be associated with poorer prognosis. CD123 expression has been reported in other hematologic malignancies, including myelodysplasia, systemic mastocytosis, blast plasmacytoid dendritic cell neoplasm (BPDCN), ALL, and malignant reticuloendotheliosis.

CD123 is expressed on AML leukemia stem cells, and a growing body of evidence suggests that AML arises from these LSCs, which have been shown to be dormant and relatively resistant to DNA-damaging chemotherapy. It has been hypothesized that LSC persistence underlies relapse after primary remission, and thus eradication of LSCs can be considered a requirement for cure and an important therapeutic challenge.

Leukemic stem cells (LSC) are cancer cells that have the characteristics of normal stem cells, namely, the property of self-renewal and the ability to develop into multiple lines. It is assumed that such cells persist in hemoblastoses, such as AML, as separate populations. The LCS present in AML patients are the so-called AML-LCS.

Acute myeloid leukemia (AML) is a clonal disorder that manifests clinically as an increased proliferation of heterogeneous and undifferentiated myeloid blast cells. The hierarchical structure of leukemia is maintained by a small population of LSCs (AMLLCS), which are characterized by a pronounced ability to self-renewal and are able to differentiate into leukemia progenitor cells. These progenitor cells form huge numbers of leukemia blast cells that are easily detected in patients at diagnosis and relapse, which eventually leads to death. AML-LSCs have generally been reported to be quiescent cells, as opposed to rapidly dividing clonogenic progenitor cells. This property of AML-LSC makes traditional chemotherapy drugs that target dividing cells less effective, possibly explaining the current evidence

- 29 041572 In the second, a large proportion of AML patients go into complete remission, but relapse almost always occurs, with <30% of adult patients surviving more than 4 years. In addition, the appearance of minimal residual disease and a low survival rate were characteristic of a high incidence of LSC at the time of diagnosis in patients with AML. Therefore, for the long-term management of AML (and also the other aforementioned hemablastosis conditions), it is necessary that new therapies be developed that specifically kill LSCs.

CD123 overexpression has been reported on AML blast cells and on CD34+/CD38 AML-LSC compared to normal hematopoietic cells.

Thus, CD123 represents an important therapeutic target for cancer therapy, in particular for cancer therapy in patients with a poor prognosis.

The present inventors have been successful in generating, screening and selecting mouse and rat specific anti-CD123 antibodies that exhibit a high degree of affinity for both human and Macac fascicularis CD123 protein and that are not highly cross-reactive with human CSF2RA and IL5RA proteins and CD3 proteins of Macaca fascicularis.

The present inventors determined the variable region sequence of the heavy and light chains of such monoclonal antibodies, anti-CD123 antibodies, called 3E3-D3, 1E1-G5, 2B8F3, 2F8-D6, 3B10-E6, 5A5-B4, 6B10-E4, 6C10-C4, 6D6-B8, 8B11-B7, 9B8-G6, 9D7-C8 and 9F6-G3.

An anti-CD123 antibody called 3E3-D3 contains a heavy chain variable domain consisting of the sequence

OVOLOESGPGLVOPSOTLSLTCTVSGFSLTTYDVHWVROPPGKGLEWMGRIQNGGITDYNSALK

SRLIISRDTSKSOVFLKMNSVOTEDTAMYFCAKTGSYFYAFDHWGOGTLVTVSS (SEQ ID NO: 226, with CDRs shown in bold characters) containing CDR1-H with the sequence under SEQ ID NO: 227, CDR2-H with the sequence under SEQ ID NO: 228 and CDR3-H with the sequence under SEQ ID NO: 229, or a heavy chain variable domain consisting of the sequence

OVOLOESGPGLVOPSOTLSLTCTVSGFSLTTYDVHWVROPPGKGLEWMGRIQNAGITDYNSALK

SRLIISRDTSKSOVFLKMNSVOTEDTAMYFCAKTGSYFYAFDHWGOGTLVTVSS (SEQ ID NO: 277, with CDRs shown in bold characters) containing CDR1-H with the sequence of SEQ ID NO: 227, CDR2-H with the sequence of SEQ ID NO:353, and CDR3-H with the sequence of SEQ ID NO: 229, or a heavy chain variable domain consisting of the sequence

OVOLOESGPGLVOPSOTL SL TCTVSGFSLTTYDVHWVROPPGKGL EWMGRIODGGITDYNSALK

SRLIISRDTSKSOVFLKMNSVOTEDTAMYFCAKTGSYFYAFDHWGOGTLVTVSS (SEQ ID NO: 278, with CDRs shown in bold characters) containing CDR1-H with the sequence of SEQ ID NO: 227, CDR2-H with the sequence of SEQ ID NO: 279, and CDR3-H with the sequence of SEQ ID NO: 229, and a light chain variable domain consisting of the sequence

OFVLTQPNSVSTNLGSTVKLSCKRNTGNIGSNYVNWYOOHEGRSPTTMIYRDDKRPDGVPDRFS

GSIDRSSNSALLTINNVQTEDEADYFCQSYSSGINIIFGGGTKLTVL (SEQ ID NO: 230, with CDRs shown in bold characters) containing a CDR1-L with the sequence of SEQ ID NO: 231, a CDR2-L consisting of the sequence of RDD, and a CDR3-L with the sequence of SEQ ID NO : 232.

The anti-CD123 antibody, called 1E1-G5, contains a heavy chain variable domain consisting of the sequence

QVQLQESGPTLVKPGDSVKMSCKAFGYTFTDHIIHWVKQSHGKSLEWIGYINPYSGGTNYNEKF

KSKATLTVDKSSSTAYMEFSRLTSEDSAICYCALNYGSYYAMDAWGOGTSVTVSS (SEQ ID NO: 198, with CDRs shown in bold characters) containing CDR1-H with the sequence of SEQ ID NO: 199, CDR2-H with the sequence of SEQ ID NO: 200, and CDR3-H with the sequence of SEQ ID NO: 201, and a light chain variable domain consisting of the sequence

DIOMTOSPASLSASLGOTVTIECRPSEDIYSNLAWFOQKPGSSPQLLIYDANNLADGVPSRFSG

SGSGTOYSLKINSLOSEDVASYFCOQYNKYPYTFGTGTKLELK (SEQ ID NO: 202, with CDRs shown in bold characters) containing a CDR1-L with the sequence of SEQ ID NO: 203, a CDR2-L consisting of the sequence of DAN, and a CDR3-L with the sequence of SEQ ID NO : 204.

The anti-CD123 antibody, called 2B8-F3, contains a heavy chain variable domain consisting of the sequence

- 30 041572

EVOLVESGGGL VOPGRSLKL SCAASGFTFSDYNMAWVROAPKKGL EWVATILYDGGRTYYRGSV

KGRFTISRDNAKSTLYLRMDSLRSEDTATYYCATHSRGTDYFDYWGOGVMVTVSS (SEQ ID NO: 205, with CDRs shown in bold characters) containing CDR1-H with the sequence of SEQ ID NO: 206, CDR2-H with the sequence of SEQ ID NO: 207, and CDR3-H with the sequence of SEQ ID NO: 208, and a light chain variable domain consisting of the sequence

EIVLTQSPTSMTASPGEQVTITCRASSSINYMHWYQQKPGASPRPWIYETSKLASGVPDRFSGS

ASGTSYSLTINNMEAEDAATYYCQQWNYPSWTFGGGTKLELK (SEQ ID NO: 209, with CDRs shown in bold characters) containing a CDR1-L with the sequence of SEQ ID NO: 210, a CDR2-L consisting of the sequence of ETS, and a CDR3-L with the sequence of SEQ ID NO : 211.

The anti-CD123 antibody, called 2F8-D6, contains a heavy chain variable domain consisting of the sequence

OVOLKESGPGLVOPSOTLSLTCTVSGFSLTSYGVSWVRQPPGKGLEWIATISSAGSTYYDLVLK

SRLSITRDTSKSOVFLKVHSLQTEDTAIYLCARDAPVFNYGSYNAMDSWGOGTSVTVSS (SEQ ID NO: 212, with CDRs shown in bold characters) containing CDR1-H with the sequence of SEQ ID NO: 213, CDR2-H with the sequence of SEQ ID NO: 214, and CDR3-H with the sequence of SEQ ID NO: 215, and a light chain variable domain consisting of the sequence

DIQMTOSPSFLSATVGDRVTINCKASQNINKYLNWYOQKLGEAPKRLIYNTNSLQTGIPSRFSG

SGSGTDYTLTISSLQPEDVATYFCLOHKSGLTFGSGTKLEIK (SEQ ID NO: 216, with CDRs shown in bold characters) containing a CDR1-L with the sequence of SEQ ID NO: 217, a CDR2-L consisting of the sequence of NTN, and a CDR3-L with the sequence of SEQ ID NO : 218.

The anti-CD123 antibody, called 3B10-E6, contains a heavy chain variable domain consisting of the sequence

OVOLQOSGPELRRPGSSVRLSCKASGYRIKDFLIHWIKNRPEHGLEWIGWIDPEDGETKYAQKF

QTKATL TADT SSNTAYMQL S S L T S E DTATY FCARWGDVYYGLMRGHVMDAWGOGTSVTVSS (SEQ ID NO: 219, CDRs shown in bold) containing CDR1-H with the sequence of SEQ ID NO: 220, CDR2-H with the sequence of SEQ ID NO: 221 and CDR3-H with the sequence of SEQ ID NO: 222, and a light chain variable domain consisting of the sequence

DVLMTOTPVSLPVSLGGOVSISCRSSQSLVHSDGDTYLHWYLQKPGOSPOLLIYRVSNRFSGVP

DRFSGSGSGTDFTLKISRVEPEDLGLYYCLQTTHFPPWTFGGGTKLEMK (SEQ ID NO: 223, with CDRs shown in bold characters) containing a CDR1-L with the sequence of SEQ ID NO: 224, a CDR2-L consisting of the sequence of RVS, and a CDR3-L with the sequence of SEQ ID NO : 225.

The anti-CD123 antibody, called 5A5-B4, contains a heavy chain variable domain consisting of the sequence

QIQIVQSGSDVTSТCNGCGTСYFSGFSLSTTGICVSWIRQPSGKGQEWLADFCWDDGKGYNPSL

KNRLSISKDTSNNQVFLKITSVDTADTATYYCARRRVYYGIYFDYWGOGVMVTVSS (SEQ ID NO: 233, with CDRs shown in bold characters) containing CDR1-H with the sequence of SEQ ID NO: 234, CDR2-H with the sequence of SEQ ID NO: 235, and CDR3-H with the sequence of SEQ ID NO: 236, and a light chain variable domain consisting of the sequence

DIVMTOSPALAVSPGERVSISCRASNSVSTRMHWYOOKPGOOPKLLIYGASNLESGVPARFSGS

GSGTDDFTLTIPDVEADDIATYFCQQSWNDPLTFGSGTKLEIK (SEQ ID NO: 237, with CDRs shown in bold characters) containing a CDR1-L with the sequence of SEQ ID NO: 238, a CDR2-L consisting of the sequence of GAS, and a CDR3-L with the sequence of SEQ ID NO : 239.

The anti-CD123 antibody, called 6B10-E4, contains a heavy chain variable domain consisting of the sequence

EVOLVESGGGLVOPGRSLKLSCAASGFTFSHYNMAWVROAPKKGLEWVATITYDDHSTYYRDSV

KGRFTISRDTAKSTLYLOMDSLRSEDTATYYCARLVNYAFAYWGOGTLVTVSS (SEQ ID NO: 240, with CDRs shown in bold characters) containing CDR1-H with the sequence of SEQ ID NO: 241, CDR2-H with the sequence of SEQ ID NO: 242, and CDR3-H with the sequence of SEQ ID NO: 243, and a light chain variable domain consisting of the sequence

- 31 041572

NIVMTQSPKSMSISVGDRVTMNCKASQTVGNNIAWYQQKPGLSPQLLIDYASNRYTGVPNRFTG

GGYGTDFILTINSVQAEDAAFYYCQRMYNSPTFGGGTKLELK (SEQ ID NO: 244, with CDRs shown in bold characters) containing a CDR1-L with the sequence of SEQ ID NO: 245, a CDR2-L consisting of the sequence of YAS, and a CDR3-L with the sequence of SEQ ID NO : 246.

The anti-CD123 antibody, called 6C10-C4, contains a heavy chain variable domain consisting of the sequence

EVKLQESGPSLVQPSETLSLTCTVSGFSLTSYSVHWVRQHSGKSLEWMGRMWNDGDTSYNSAFT

SRLSISRDTSKGOVFLKMNSLOTEDTGTYYCARGHRTPFDYWGOGVMVTVSS (SEQ ID NO: 247, with CDRs shown in bold characters) containing CDR1-H with the sequence of SEQ ID NO: 248, CDR2-H with the sequence of SEQ ID NO: 249, and CDR3-H with the sequence of SEQ ID NO: 250, and a light chain variable domain consisting of the sequence

DIVMTQSPSSLAVSAGETVTINCKSSQSFLSSGDERNYVAWYQHKPGOSPKLLIYWASTRHSGV

PDRFIGSGSGTDFTLTISSVQAEDLAIYYCQQYYDTPLTFGSGTKLEIK (SEQ ID NO: 251, with CDRs shown in bold characters) containing a CDR1-L with the sequence of SEQ ID NO: 252, a CDR2-L consisting of the sequence of WAS, and a CDR3-L with the sequence of SEQ ID NO : 253.

An anti-CD123 antibody called 6D6-B8 contains a heavy chain variable domain consisting of the sequence

QVQLQESGPTLVKPGDSVKMSCKASAYTFDTDNIIHWVKQSHGKSLDWIGYINPYSGGTNYNGWF

RSKATLTVDKSSSTAYMEFSRLTSDDSAIYYCALNYGSYYAMDAWGOGTSVTVSS (SEQ ID NO: 254, with CDRs shown in bold) containing CDR1-H with the sequence of SEQ ID NO: 255, CDR2-H with the sequence of SEQ ID NO: 200, and CDR3-H with the sequence of SEQ ID NO: 201, and a light chain variable domain consisting of the sequence

DIQMT QS PAS LSAS LGETVTIDCRPSEDIFNNLAWYOQKPGNSPQLLIYDANSLADGVPSRFS G

SGSGTQYSLMIIRLQSEDVASYFCHQYNIYPYTFGAGTKLELK (SEQ ID NO: 256, with CDRs shown in bold characters) containing a CDR1-L with the sequence of SEQ ID NO: 257, a CDR2-L consisting of the sequence of DAN, and a CDR3-L with the sequence of SEQ ID NO : 258.

The anti-CD123 antibody, called 8B11-B7, contains a heavy chain variable domain consisting of the sequence

QVQLQESGPTLVNPGDSVKMSCKASGYTFTDHIIHWVKQSHGKSLEWIGYINPYSGGANYNGKF

KSKATLTIDKSSSTAYMEFSRLTSGDSAIYYCALNYGSYYAMDAWGOGTSVTVSS (SEQ ID NO: 259, with CDRs shown in bold characters) containing CDR1-H with the sequence of SEQ ID NO: 199, CDR2-H with the sequence of SEQ ID NO: 260, and CDR3-H with the sequence of SEQ ID NO: 201, and a light chain variable domain consisting of the sequence

DIQMTQSPASLSASLGETVTIECRTSKDIYSNLAWFQQEPGNSPQLLIYDASNLADGVPSRFSG

SGSGTQYSLOINNLQSEDVASYFCHQYNNYPYTFGTGTKLELK (SEQ ID NO: 261, with CDRs shown in bold characters) containing a CDR1-L with the sequence of SEQ ID NO: 262, a CDR2-L consisting of the sequence of 'DAS', and a CDR3-L with the sequence of SEQ ID NO: 263.

An anti-CD123 antibody called 9B8-G6 contains a heavy chain variable domain consisting of the sequence

EVKLOESGPSLVOSSOTLSLTCTVSGFSLTSYHIHWVRQPPGKGLEWMGVMWSDGDTSYSSALK

SRLSISRDTSQSQVFLKMNSLQTEDTATYYCARGDYSSYIYLWFAYWGOGTLVTVSS (SEQ ID NO: 264, with CDRs shown in bold characters) containing CDR1-H with the sequence of SEQ ID NO: 265, CDR2-H with the sequence of SEQ ID NO: 266, and CDR3-H with the sequence of SEQ ID NO: 267, and a light chain variable domain consisting of the sequence

DIVMTQSPSSLAVSAGETVTINCKSSQSFLSSGDERNYVAWYQHKPGOSPKLLIYWASTRHSGV

PDRFIGSGSGTDFTLTISSVOAEDLAIYYCOQYYDTPLTFGSGTKLEIK (SEQ ID NO: 251, with CDRs shown in bold characters) containing a CDR1-L with the sequence of SEQ ID NO: 252, a CDR2-L consisting of the sequence of WAS, and a CDR3-L with the sequence of SEQ ID NO : 253.

An anti-CD123 antibody called 9D7-C8 contains

- 32 041572 heavy chain variable domain, consisting of the sequence

EVKLOESGPSLVOSSOTLSLTCTVSGFSLTSYHIHWVROTPGKGLEWMGVMWSDGDTSYNSALK

SRLSISRDTSOSOVFLKMNSLOTEDTATYYCARGYYSSYLYLWFAYWGOGTLVTVSS (SEQ ID NO: 268, with CDRs shown in bold characters) containing CDR1-H with the sequence of SEQ ID NO: 265, CDR2-H with the sequence of SEQ ID NO: 266, and CDR3-H with the sequence of SEQ ID NO: 269, and a light chain variable domain consisting of the sequence

DIVMTOSPSSLAVSEGETVTINCKSSQSFLSSGDGKNYVAWYOYKPGOSPKLLIYWASTRQSGV

PDRFIGSGSGTDFTLTISTVOAEDLAIYYCOQYYDTPLTFGSGTKLEIK (SEQ ID NO: 270, with CDRs shown in bold characters) containing a CDR1-L with the sequence of SEQ ID NO: 271, a CDR2-L consisting of the sequence of WAS, and a CDR3-L with the sequence of SEQ ID NO : 253.

An anti-CD123 antibody called 9F6-G3 contains a heavy chain variable domain consisting of the sequence

OVOLOESGPTLVKPGDSVKMSCKASGYTFTDYIIHWVKQSHGKSLEWIGYINPYSDGTNYNEKF

KSCATLTVDKSTAYMEFSRLTSEDSAIYFCALNYGSYYAMDAWGOGTSVTVSS (SEQ ID NO: 272, with CDRs shown in bold characters) containing CDR1-H with the sequence of SEQ ID NO: 273, CDR2-H with the sequence of SEQ ID NO: 274, and CDR3-H with the sequence of SEQ ID NO: 201, and a light chain variable domain consisting of the sequence

DIQMTQSPASLSASLGETVTIECRPSEDIHSNVAWYQQKPGNSPQLLIYDASNLADGVPSRFSG

5G5GTOYSLKINSLO5EDVASYFCHQYNIYPYTFGSGTKLELK (SEQ ID NO: 275, with CDRs shown in bold characters) containing a CDR1-L with the sequence of SEQ ID NO: 276, a CDR2-L consisting of the sequence of DAS, and a CDR3-L with the sequence of SEQ ID NO : 258.

In one aspect of the present invention, an anti-CD123 antibody binds to human CD123. In another embodiment, the anti-CD123 antibody further binds to Macaca fascicularis CD123. In particular, the anti-CD123 antibody of the present invention binds to the extracellular domain of human CD123 or both human and Macac fascicularis CD123. More specifically, an anti-CD123 antibody binds to a distal fragment of CD12 3, such as the amino acids at position 19-49 of human CD123 with the amino acid sequence of SEQ ID NO: 194. An anti-CD123 antibody binds to CD123 whether it is expressed as an isolated form or present as a soluble extracellular domain or full-length membrane-anchored CD123 present in cells expressing CD123, such as AML cells or cells transfected with CD123. The anti-CD123 antibody of the present invention is specific for cells that express human and Macac fascicularis CD123 proteins on their surface, such as CD123-expressing cancer cells.

The anti-CD123 antibody of the present invention is characterized by an affinity ratio for Macaca fascicularis CD123 and an affinity for human CD123 (KD(Macaca fascicularis)/KD(human)) which is <10, in particular <6, <5, <4, <3 , <2, <1 or <0.5. Thus, the polypeptide of the present invention can be used in toxicology studies in monkeys because the toxicity profile observed in monkeys is consistent with the expected potential adverse effects in humans.

In particular, the anti-CD123 antibody of the present invention does not bind to or is not significantly cross-reactive with CSF2RA and/or IL5RA proteins.

In particular, the antibody does not bind to the extracellular domain of the aforementioned human and Macac fascicularis CSF2RA and/or IL5RA protein(s) or is not significantly cross-reactive with them.

In addition, the anti-CD123 antibody of the present invention has an affinity (KD) for human CD123 or Macac fascicularis CD123, or both, which is <50 nM, <40 nM, <30 nM, e.g., <20 nM, <15 nM, < 10 nM, <8 nM, <6 nM, <4 nM, <2 nM, <1 nM, or <0.5 nM, e.g., affinity is 0.1 nM to 20 nM, e.g., 0.1 nM to 10 nM, in particular from 0.1 nM to 2 nM or from 0.1 nM to 1 nM.

In one example, affinity for human CD3 or Macac fascicularis CD3 is determined as a KD value using surface plasmon resonance using recombinant CD123 protein from human and Macac fascicularis as the capture antigen, e.g. using a fusion protein containing Fc and CD123 of human and Macac fascicularis.

As an antigen, for example, the extracellular domain of human CD123 or Macac fascicularis can be used, including a signal sequence corresponding to the amino acid sequence from position M1 to R305 from the wild-type protein (SEQ ID NO: 194 (human), SEQ ID NO: 195 (Ma- 33 041572 caca fascicularis)). The resulting amino acid sequence of the mature human CD123 protein or

Maca fascicularis is listed under SEQ ID NO: 196 and SEQ ID NO: 197, respectively. Biacore measurements are known to the person skilled in the art. In this example, the measurement with

Biacore can be performed as described in the anti-CD3 antibody section above.

An anti-CD123 antibody of the present invention can also have an apparent affinity constant (apparent KD), which is, for example, determined by FACS analysis on cells expressing CD123, such as HEK293 cells transfected with CD123, using a crude anti-CD123 antibody in the supernatant. a hybridoma that is <20 nM, such as <15 nM, <10 nM, <6 nM, <5 nM, <4 nM, <3 nM, <2 nM, or <1 nM. Typically, the apparent KD is in the range of 0.1-20 nM, in particular 0.1-10 nM, eg 0.1-5 nM.

Sequence alignments of the VH and VL regions of anti-CD123 antibodies referred to as 3E3-D3, 1E1-G5, 2B8-F3, 2F8-D6, 3B10-E6, 5A5-B4, 6B10-E4, 6C10-C4, 6D6-B8, 8B11-B7 , 9B8G6, 9D7-C8, 9F6-G3. Comparison of the CDR-H and CDR-L sequences in most cases indicates that antibodies to CD123, 1E1-G5, 6D6-B8, 8B11-B7 and 9F6-G3, on the one hand, and antibodies to CD123, 6C10-C4, 9B8-G6, 9D7-C8, on the other hand, are closely related in terms of structure, with these antibodies likely to bind to the same epitope. The alignment of anti-CD123 antibodies named 1E1-G5, 6D6-B8, 8B11-B7 and 9F6-G3 and anti-CD123 antibodies named 6C10-C4, 9B8-G6, 9D7-C8 is shown in FIG. 3 and 4, respectively. Comparing the sequences of CDR-H and CDR-L additionally identified CDR positions that were strongly conserved between the two antibody groups and thus presumably important in terms of specificity, while the remaining positions could contain substitutions.

Accordingly, an antibody of the present invention contains:

a) a heavy chain variable domain containing a CDR1-H consisting of the sequence X ₁ YTFDDX ₂ I (SEQ ID NO: 336), where X ₁ is G or A and X ₂ is H, Y or N, or any their combination; And

CDR2-H, consisting of the sequence INPYSX ₁ GX ₂ (SEQ ID NO: 337), where X ₁ is G or D and X ₂ is T or A, or any combination thereof; And

CDR3-H consisting of the sequence ALNYGSYYAMDA (SEQ ID NO: 201) and a light chain variable domain containing CDR1-L consisting of the sequence X ₁ DIX ₂ X ₃ N (SEQ ID NO: 338), where X ₁ is E or K, X ₂ is F, H or Y, and X ₃ is N or S, or any combination thereof; And

CDR2-L, consisting of a DAN or DAS sequence; And

CDR3-L, consisting of the sequence X ₁ QYNX ₂ YPYT (SEQ ID NO: 339), where X ₁ represents H or Q, and X ₂ represents I, K or N, or any combination thereof; or

b) a heavy chain variable domain containing a CDR1-H, consisting of the sequence GFSLTSYX1 (SEQ ID NO: 340), where X1 is H or S; And

CDR2-H, consisting of the sequence MWX^GDT (SEQ ID NO: 341), where X ₁ represents S or N; And

CDR3-H consisting of the sequence ARGX ₁ X ₂ X ₃ X4X ₅ X ₆ X ₇ X ₈ X ₉ FX ₁₀ Y (SEQ ID NO: 342), where X ₁ is D, Y or H, X ₂ is Y or R, X ₃ is S or T, X ₄ is S or P, X ₅ is Y or no amino acid, X ₆ is L, I or no amino acid, X ₇ is Y or no amino acid, X ₈ is L or the absence of an amino acid, X ₉ is W or the absence of an amino acid, X ₁₀ is A or D, or any combination thereof, and a light chain variable domain containing a CDR1-L consisting of the sequence QSFLSSGDX ₁ X ₂ NY (SEQ ID NO: 343), where X ₁ is E or G and X ₂ is R or K, or any combination thereof; And

CDR2-L, consisting of the WAS sequence; And

CDR3-L consisting of the sequence QQYYDTPLT (SEQ ID NO: 253).

In an embodiment, an anti-CD123 antibody of the present invention comprises the CDR sequences from the heavy and/or light chains of one of 13 anti-CD123 antibodies referred to as 3E3-D3, 1E1-G5, 2B8-F3, 2F8-D6, 3B10-E6, 5A5-B4 , 6B10-E4, 6C10-C4, 6D6-B8, 8B11B7, 9B8-G6, 9D7-C8 and 9F6-G3 listed above.

Therefore, the present invention provides an anti-CD123 antibody which contains:

a) a heavy chain variable domain containing a CDR1-H with the sequence of SEQ ID NO: 227 or a sequence different from SEQ ID NO: 227 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 228 or SEQ ID NO: 353 or SEQ ID NO: 279 or a sequence other than SEQ ID NO: 228 or SEQ ID NO: 353 or SEQ ID NO: 279 one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 229 or a sequence different from SEQ ID NO: 229 by one amino acid substitution; and a light chain variable domain containing CDR1-L with the sequence of SEQ ID NO:

- 34 041572

231 or a sequence different from SEQ ID NO: 231 by one amino acid substitution;

CDR2-L with the RDD sequence or a sequence different from RDD by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 232 or a sequence different from SEQ ID NO: 232 by one amino acid substitution; or

b) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 199 or a sequence different from SEQ ID NO: 199 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 200 or a sequence different from SEQ ID NO: 200 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 201 or a sequence different from SEQ ID NO: 201 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 203 or a sequence different from SEQ ID NO: 203 by one amino acid substitution; CDR2-L with the DAN sequence or a sequence different from DAN by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 204 or a sequence different from SEQ ID NO: 204 by one amino acid substitution; or

c) a heavy chain variable domain containing a CDR1-H with the sequence of SEQ ID NO: 206 or a sequence different from SEQ ID NO: 206 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 207 or a sequence different from SEQ ID NO: 207 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 208 or a sequence different from SEQ ID NO: 208 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence under SEQ ID NO: 210 or a sequence different from SEQ ID NO: 210 by one amino acid substitution; CDR2-L with the ETS sequence or a sequence different from ETS by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 211 or a sequence different from SEQ ID NO: 211 by one amino acid substitution; or

d) a heavy chain variable domain containing a CDR1-H with the sequence of SEQ ID NO: 213 or a sequence different from SEQ ID NO: 213 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 214 or a sequence different from SEQ ID NO: 214 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 215 or a sequence different from SEQ ID NO: 215 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 217 or a sequence different from SEQ ID NO: 217 by one amino acid substitution;

CDR2-L with the sequence of NTN or a sequence different from NTN by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 218 or a sequence different from SEQ ID NO: 218 by one amino acid substitution; or

e) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 220 or a sequence different from SEQ ID NO: 220 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 221 or a sequence different from SEQ ID NO: 221 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 222 or a sequence different from SEQ ID NO: 222 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 224 or a sequence different from SEQ ID NO: 224 by one amino acid substitution;

CDR2-L with the sequence of RVS or a sequence different from RVS by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 225 or a sequence different from SEQ ID NO: 225 by one amino acid substitution; or

f) a heavy chain variable domain containing a CDR1-H with the sequence of SEQ ID NO: 234 or a sequence different from SEQ ID NO: 234 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 235 or a sequence different from SEQ ID NO: 235 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 236 or a sequence different from SEQ ID NO: 236 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 238 or a sequence different from SEQ ID NO: 238 by one amino acid substitution;

CDR2-L with a GAS sequence or a sequence differing from GAS by one amino acid substitution, and CDR3-L with a sequence of SEQ ID NO: 239 or a sequence differing from SEQ ID NO: 239 by one amino acid substitution; or

g) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 241 or a sequence different from SEQ ID NO: 241 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 242 or a sequence different from SEQ

- 35 041572

ID NO: 242 with one or more amino acid substitutions; CDR3-H with sequence under

SEQ ID NO: 243 or a sequence different from SEQ ID NO: 243 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 245 or a sequence different from SEQ ID NO: 245 by one amino acid substitution; CDR2-L with a YAS sequence or a sequence different from YAS by one amino acid substitution, and CDR3-L with a sequence under SEQ ID NO: 246 or a sequence different from SEQ ID NO: 246 by one amino acid substitution; or

h) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 248 or a sequence different from SEQ ID NO: 248 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 249 or a sequence different from SEQ ID NO: 249 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 250 or a sequence different from SEQ ID NO: 250 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 252 or a sequence different from SEQ ID NO: 252 by one amino acid substitution; CDR2-L with the sequence of WAS or a sequence different from WAS by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 253 or a sequence different from SEQ ID NO: 253 by one amino acid substitution; or

i) a heavy chain variable domain containing a CDR1-H with the sequence of SEQ ID NO: 255 or a sequence different from SEQ ID NO: 255 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 200 or a sequence different from SEQ ID NO: 200 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 201 or a sequence different from SEQ ID NO: 201 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence under SEQ ID NO: 257 or a sequence different from SEQ ID NO: 257 by one amino acid substitution; CDR2-L with the DAN sequence or a sequence different from DAN by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 258 or a sequence different from SEQ ID NO: 258 by one amino acid substitution; or

j) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 199 or a sequence different from SEQ ID NO: 199 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 260 or a sequence different from SEQ ID NO: 260 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 201 or a sequence different from SEQ ID NO: 201 by one amino acid substitution; and a light chain variable domain comprising a CDR1-L with the sequence of SEQ ID NO: 262 or a sequence different from SEQ ID NO: 262 by one amino acid substitution; CDR2-L with a DAS sequence or a sequence different from DAS by one amino acid substitution, and CDR3-L with a sequence under SEQ ID NO: 263 or a sequence different from SEQ ID NO: 263 by one amino acid substitution; or

k) a heavy chain variable domain containing a CDR1-H with the sequence under SEQ ID NO: 265 or a sequence different from SEQ ID NO: 265 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 266 or a sequence different from SEQ ID NO: 266 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 267 or a sequence different from SEQ ID NO: 267 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 252 or a sequence different from SEQ ID NO: 252 by one amino acid substitution;

CDR2-L with the sequence of WAS or a sequence different from WAS by one amino acid substitution, and CDR3-L with the sequence of SEQ ID NO: 253 or a sequence different from SEQ ID NO: 253 by one amino acid substitution; or

l) a heavy chain variable domain containing a CDR1-H with the sequence of SEQ ID NO: 265 or a sequence different from SEQ ID NO: 265 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 266 or a sequence different from SEQ ID NO: 266 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 269 or a sequence different from SEQ ID NO: 269 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence of SEQ ID NO: 271 or a sequence different from SEQ ID NO: 271 by one amino acid substitution;

CDR2-L with a WAS sequence or a sequence differing from WAS by one amino acid substitution, and a CDR3-L with a sequence of SEQ ID NO: 253 or a sequence differing from SEQ ID NO: 253 by one amino acid substitution; or

m) a heavy chain variable domain containing a CDR1-H with the sequence of SEQ ID NO: 273 or a sequence different from SEQ ID NO: 273 by one amino acid substitution; CDR2-H with a sequence under SEQ ID NO: 274 or a sequence different from SEQ ID NO: 274 by one or more amino acid substitutions; CDR3-H with a sequence under SEQ ID NO: 201 or a sequence different from SEQ ID NO: 201 by one amino acid substitution; and a light chain variable domain containing a CDR1-L with the sequence under SEQ ID NO: 276 or a sequence different from SEQ ID NO: 276 by one amino acid substitution;

CDR2-L with a DAS sequence or a sequence different from DAS by one amino acid substitution, and CDR3-L with a sequence under SEQ ID NO: 258 or a sequence different from SEQ ID NO: 258 by one amino acid substitution.

One or more individual amino acids can be changed by introducing a substitution, in particular a conservative substitution, into one or more of the above CDR sequences. Such a change may involve, for example, the removal of a glycosylation site or a deamidation site in combination with humanization of the antibody.

Based on sequence alignment of the VH and VL regions from antibodies to CD123,1E1-G5, 6D6-B8, 8B11-B7, 9F6-G3, various amino acid substitutions were identified. Therefore, in one embodiment, the amino acid may be substituted:

in CDR1-H, at one or more of positions 1 and 7, for example, at position 7 of CDR1-H with the sequence GYTFTDHI (SEQ ID NO:199) or GYTFTDYI (SEQ ID NO:273), for example, at position 1 and 5 from CDR1-H with the sequence AYTFTDNI (SEQ ID NO: 255); and/or in CDR2-H, at one or more of positions b and 8, for example, at position 6 and/or 8 of CDR2H with the sequence INPYSGGT (SEQ ID NO: 200), or INPYSDGT (SEQ ID NO: 274), or INPYSGGA (SEQ ID NO: 260); and/or in CDR1-L, at one or more of positions 1, 4 and 5, for example at position 1 or 1, 4 and 5 of CDR1-L with the sequence KDIYSN (SEQ ID NO: 262) or EDIFNN (SEQ ID NO : 257); and/or at CDR2-L, position 3 of the DAN or DAS sequence; and/or in CDR3-L, at one or more of positions 1 and 5, for example, at position 1 and/or 5 of CDR3L with the sequence HQYNIYPYT (SEQ ID NO: 258), or QQYNKYPYT (SEQ ID NO: 204), or HQYNNYPYT (SEQ ID NO: 263).

In an additional embodiment, the amino acid is substituted:

in CDR1-H, at position 8, for example, at position 8 of CDR1-H with the sequence GFSLTSYH (SEQ ID NO: 265) or GFSLTSYS (SEQ ID NO: 248); and/or in CDR2-H, at position 3, for example at position 3 of CDR2-H with the sequence MWSDGD (SEQ ID NO: 265) or MWNDGD (SEQ ID NO: 248); and/or in CDR3-H, at positions 4 and 9, for example, at positions 4 and 9 of CDR3-H with the sequence ARGDYSSYIYLWFAY (SEQ ID NO: 267) or ARGYYSSYLYLWFAY (SEQ ID NO: 269); and/or in CDR1-L, at one or more of positions 9 and 10, for example at positions 9 and 10 of CDR1-L with the sequence QSFLSSGDERNY (SEQ ID NO: 252) or QSFLSSGDGKNY (SEQ ID NO: 271).

In an embodiment, the antibody comprises:

a) heavy chain variable domain containing CDR1-H with the sequence under SEQ ID NO: 227, CDR2-H with the sequence under SEQ ID NO: 228, SEQ ID NO:353, SEQ ID NO:279, CDR3-H with the sequence under SEQ ID NO: 229, and a light chain variable domain containing CDR1-L with the sequence of SEQ ID NO: 231, CDR2-L with the sequence of RDD and CDR3-L with the sequence of SEQ ID NO: 232; or

n) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 199, a CDR2-H of the sequence of SEQ ID NO: 200, a CDR3-H of the sequence of SEQ ID NO: 201, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 203, CDR2-L with the sequence DAN and CDR3-L with the sequence under SEQ ID NO: 204; or

o) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 206, a CDR2-H of the sequence of SEQ ID NO: 207, a CDR3-H of the sequence of SEQ ID NO: 208, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 210, CDR2-L with the sequence of ETS and CDR3-L with the sequence under SEQ ID NO: 211; or

p) a heavy chain variable domain containing a CDR1-H with the sequence of SEQ ID NO: 213, a CDR2-H with the sequence of SEQ ID NO: 214, a CDR3-H with the sequence of SEQ ID NO: 215, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 217, CDR2-L with the sequence of NTN and CDR3-L with the sequence under SEQ ID NO: 218; or

q) heavy chain variable domain containing CDR1-H with the sequence under SEQ ID

- 37 041572

NO: 220, CDR2-H with the sequence under SEQ ID NO: 221, CDR3-H with the sequence under SEQ

ID NO: 222, and a light chain variable domain containing CDR1-L with the sequence under SEQ ID NO:

224, CDR2-L with the sequence RVS and CDR3-L with the sequence under SEQ ID NO: 225; or

r) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 234, a CDR2-H of the sequence of SEQ ID NO: 235, a CDR3-H of the sequence of SEQ ID NO: 236, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 238, CDR2-L with the sequence of GAS and CDR3-L with the sequence under SEQ ID NO: 239; or

s) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 241, a CDR2-H of the sequence of SEQ ID NO: 242, a CDR3-H of the sequence of SEQ ID NO: 243, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 245, CDR2-L with the sequence YAS and CDR3-L with the sequence under SEQ ID NO: 246; or

t) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 248, a CDR2-H of the sequence of SEQ ID NO: 249, a CDR3-H of the sequence of SEQ ID NO: 250, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 252, CDR2-L with the sequence of WAS and CDR3-L with the sequence under SEQ ID NO: 253; or

u) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 255, a CDR2-H of the sequence of SEQ ID NO: 200, a CDR3-H of the sequence of SEQ ID NO: 201, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 257, CDR2-L with the sequence DAN and CDR3-L with the sequence under SEQ ID NO: 258; or

v) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 199, a CDR2-H of the sequence of SEQ ID NO: 260, a CDR3-H of the sequence of SEQ ID NO: 201, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 262, CDR2-L with the sequence of DAS and CDR3-L with the sequence under SEQ ID NO: 263; or

w) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 265, a CDR2-H of the sequence of SEQ ID NO: 266, a CDR3-H of the sequence of SEQ ID NO: 267, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 252, CDR2-L with the sequence of WAS and CDR3-L with the sequence under SEQ ID NO: 253; or

x) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 265, a CDR2-H of the sequence of SEQ ID NO: 266, a CDR3-H of the sequence of SEQ ID NO: 269, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 271, CDR2-L with the sequence of WAS and CDR3-L with the sequence under SEQ ID NO: 253; or

s) a heavy chain variable domain containing CDR1-H with the sequence of SEQ ID NO: 273, CDR2-H with the sequence of SEQ ID NO: 274, CDR3-H with the sequence of SEQ ID NO: 201, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 276, CDR2-L with the sequence of DAS and CDR3-L with the sequence under SEQ ID NO: 258.

An anti-CD123 antibody of the present invention is particularly a conventional antibody, such as a conventional monoclonal antibody or antibody fragment, a bispecific or polyspecific antibody.

The anti-CD123 antibody of the present invention specifically comprises or consists of IgG or a fragment thereof.

The present invention also provides an anti-CD123 antibody as defined above, further comprising at least a heavy chain variable domain and/or a light chain variable domain of one of the anti-CD123 antibodies named above.

Thus, the present invention relates in particular to an anti-CD123 antibody which contains:

a) a heavy chain variable domain with the sequence under SEQ ID NO: 226, or SEQ ID NO: 277, or SEQ ID NO: 278, or a sequence that is at least 85% identical to it, and/or a light chain variable domain with the sequence under SEQ ID NO: 230 or a sequence that is at least 85% identical to it; or

b) a heavy chain variable domain with a sequence of SEQ ID NO: 198 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 202 or a sequence that is at least 85% identical, or

- 38 041572

c) a heavy chain variable domain with a sequence of SEQ ID NO: 205 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 209 or a sequence that is at least 85% identical to her; or

d) a heavy chain variable domain with a sequence of SEQ ID NO: 212 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 216 or a sequence that is at least 85% identical to her; or

e) a heavy chain variable domain with a sequence of SEQ ID NO: 219 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 223 or a sequence that is at least 85% identical to her; or

f) a heavy chain variable domain with a sequence of SEQ ID NO: 233 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 237 or a sequence that is at least 85% identical to her; or

g) a heavy chain variable domain with a sequence of SEQ ID NO: 240 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 244 or a sequence that is at least 85% identical to her; or

h) a heavy chain variable domain with a sequence of SEQ ID NO: 247 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 251 or a sequence that is at least 85% identical to her; or

i) a heavy chain variable domain with a sequence of SEQ ID NO: 254 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 256 or a sequence that is at least 85% identical to her; or

j) a heavy chain variable domain with a sequence of SEQ ID NO: 259 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 261 or a sequence that is at least 85% identical thereto. 85% identical to her; or

k) a heavy chain variable domain with a sequence of SEQ ID NO: 264 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 251 or a sequence that is at least 85% identical to her; or

l) a heavy chain variable domain with a sequence of SEQ ID NO: 268 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 270 or a sequence that is at least 85% identical to her; or

m) a heavy chain variable domain with a sequence of SEQ ID NO: 272 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 275 or a sequence that is at least 85% identical to her.

For example, the heavy or light chain variable domain sequence may differ from the reference sequence under SEQ ID NOs: 226, 277, 278, 230, 198, 202, 205, 209, 212, 216, 219, 223, 233, 237, 240, 244 , 247, 251, 254, 256, 259, 261, 264, 251, 268, 270, 272 or 275, optionally with one or more amino acid substitutions, in particular one or more conservative amino acid substitutions and/or substitutions for canonical residues . In particular, the sequence of the heavy or light chain variable domain may differ from the reference sequence under SEQ ID NOs: 226, 277, 278, 230, 198, 202, 205, 209, 212, 216, 219, 223, 233, 237, 240, 244, 247, 251, 254, 256, 259, 261, 264, 251, 268, 270, 272 or 275 with conservative amino acid substitution(s) only.

Sequence changes compared to the sequence under SEQ ID NOs: 226, 277, 278, 230, 198, 202, 205, 209, 212, 216, 219, 223, 233, 237, 240, 244, 247, 251, 254, 256 , 259, 261, 264, 251, 268, 270, 272, or 275, in particular, will be present primarily in one or more framework regions, FR1-L, FR2-L, FR3-L, FR4-L and/or FR1 -H, FR2-H, FR3-H, FR4-H.

In one embodiment, an anti-CD123 antibody of the present invention and a fragment thereof are a rat antibody and a rat antibody fragment, respectively.

The anti-CD123 antibody of the present invention may also be a chimeric antibody, and in particular a rat/human antibody, for example, an antibody comprising rat heavy and light chain variable domains and a CH domain and a CL domain from a human antibody. The polypeptide may be a fragment of such an antibody. The anti-CD123 antibody can also be a humanized antibody or a humanized antibody fragment obtained by CDR grafting or by the 4D method (US 20110027266).

Accordingly, in one embodiment, an anti-CD123 antibody of the present invention is a humanized antibody comprising:

a) a heavy chain variable domain comprising an amino acid sequence selected from the group consisting of SEQ ID NO: 280, SEQ ID NO: 281, SEQ ID NO: 282, SEQ ID NO: 283, SEQ ID NO: 284, SEQ ID NO : 285, SEQ ID NO: 286, SEQ ID NO: 287, SEQ ID NO: 288, SEQ ID NO: 289, SEQ ID NO: 290, SEQ ID NO: 291, SEQ ID NO: 301 and SEQ ID NO: 302 ; And

b) a light chain variable domain comprising an amino acid sequence selected from the group consisting of SEQ ID NO: 292, SEQ ID NO: 293, SEQ ID NO: 294, SEQ ID NO: 295, SEQ ID NO: 296, SEQ ID NO : 297, SEQ ID NO: 298, SEQ ID NO: 299, SEQ ID NO: 300, SEQ ID NO: 303, SEQ ID NO: 304 and SEQ ID NO: 305.

In one embodiment, an anti-CD123 antibody of the present invention comprises three CDR sequences or a heavy chain variable domain, or six CDR sequences or heavy and light chain variable domains of one of the anti-CD123 antibodies listed above.

The present invention further relates to a humanized anti-CD123 antibody fragment as defined above. In one embodiment, the humanized anti-CD123 antibody described above is a chimeric antibody.

An anti-CD123 antibody of the present invention may also be a single domain antibody or a fragment thereof. In particular, a fragment of a single domain antibody may consist of a heavy chain variable region (VHH) that contains the CDR1-H, CDR2-H and CDR3-H of one of the antibodies described above. The anti-CD123 antibody can also be a heavy chain-only antibody, ie. an antibody having no light chain, which may or may not contain a CH1 domain.

The single domain antibody or fragment thereof may also comprise framework regions from the camelid single domain antibody and optionally a constant domain from the camelid single domain antibody.

An anti-CD123 antibody of the present invention may also be an antibody fragment, in particular a humanized antibody fragment selected from the group consisting of Fv, Fab, F(ab')2, Fab', dsFv, (dsFv)2, scFv, sc( Fv)2 and diatel.

An anti-CD123 antibody can also be a bispecific or polyspecific antibody formed from at least one antibody fragment or at least one variable domain of an anti-CD123 antibody of the present invention.

Multispecific antibodies are polyvalent protein complexes, as described, for example, in EP 2050764 A1 or US 2005/0003403 A1.

The anti-CD123 bispecific or polyspecific antibodies of the present invention can be characterized by specificity for (a) the extracellular domain of human or human CD123 and Macac fascicularis targeted by one of the anti-CD123 antibodies described above, and (b) at least one other antigen.

In a specific embodiment, the other antigen is CD3 and accordingly the resulting bispecific antibody is an anti-CD3/CD123 bispecific antibody. Traditional bispecific antibodies can be obtained using techniques that are known to the person skilled in the art.

Antibodies and fragments thereof according to the present invention can be used as isolated (eg, purified) molecules or contained in a carrier such as a membrane or lipid vesicle (eg, liposome).

In one additional embodiment, an anti-CD123 antibody of the present invention is used to produce an antibody-like binding protein of the present invention, further defined in the Antibody-Like Binding Protein section.

Any combination of the above embodiments forms part of the present invention.

Antibody-like binding proteins.

The present inventors have made several antibody-like binding proteins, namely antibody-like binding proteins, called 7G3x20G6, 7G3x4E7, 7G3x4B4, 7G3x18F5, hz20G6x7G3, 7G3xhz4B4, hz4B4x3E3 hz20G6x7G3-TL4 and hz20G6xhz7G3, where the term hz20G6xhz7G3 denotes humanized antibodies. These antibody-like binding proteins are characterized by the CODV construct, in particular the CODV-Fab or CODV-Ig construct.

In the context of the present invention, the CODV format refers to the dual variable region (CODV) crossover configuration of bispecific antibodies or polyspecific antibodies.

- 40 041572

The CODV format allows variable domain interchangeability with fold preservation and highest binding affinity.

The CODV ball format was previously described in international patent application WO 2012/135345. Accordingly, in one embodiment, the antibody-like binding protein of the present invention is in the CODV format previously described in international patent application WO 2012/135345, which is incorporated herein by reference.

In one embodiment, the present invention provides an antibody-like binding protein in CODV-Fab format. Accordingly, in one embodiment, the present invention relates to an antibody-like binding protein containing two polypeptide chains that form two antigen-binding sites, where the first polypeptide has the structure represented by the formula [I] ^V D1 ^-L 1 ^-V D2 ^-L 2 ^-C L ^[ 1], and the second polypeptide has the structure represented by the formula [II] ^V d3 ^-L 3 ^-V D4 ^-L 4 ^-C H1 [II], where V _D1 represents the variable domain of the heavy or light chain of the first immunoglobulin;

V _D2 is the heavy or light chain variable domain of the second immunoglobulin;

V _D3 is the heavy or light chain variable domain of said second immunoglobulin;

V _D4 is the heavy or light chain variable domain of said first immunoglobulin;

CL is an immunoglobulin light chain constant domain;

C _H1 is the C _H1 constant domain of an immunoglobulin heavy chain;

L ₁ , L2, L3 and L4 are amino acid linkers;

and wherein the first and second polypeptides form a light chain-heavy chain crossover pair, wherein both V _D1 and V _D2 are either light chain variable domains or heavy chain variable domains, and both V _D3 and V _D4 are variable domains heavy chains if V _D1 and V _D2 are light chain variable domains, or both V _D3 and V _D4 are light chain variable domains if V _D1 and V _D2 are heavy chain variable domains.

The addition of the Fc domain to the ABP in the CODV-Fab further stabilizes the ABP. More specifically, the addition of an Fc domain to a polypeptide of formula (II) of an antibody-like binding protein in a CODV-Fab increases the half-life of the antibody-like binding protein and thus improves the pharmacokinetic profile of the antibody-like binding protein. The addition of a single Fc region to a CODV-Fab dimerizes the polypeptides containing the Fc domain, and the resulting antibody-like binding protein is an antibody-like binding protein in CODV-Ig format. Therefore, the present invention further relates to an antibody-like binding protein in CODV-Ig format.

Therefore, the present invention further relates to an antibody-like binding protein containing four polypeptide chains that form four antigen-binding sites, where two polypeptide chains have the structure represented by the formula [I] ^V D1 ^-L 1 ^-V D2 ^-L 2 ^-C L ^[ 1 ^] , and two polypeptide chains have the structure represented by the formula [III]

Vd3-L3-VD4-L4-CH1-FC [III], where V _D1 represents the variable domain of the heavy or light chain of the first immunoglobulin;

VD2 is the heavy or light chain variable domain of the second immunoglobulin;

V _D3 is the heavy or light chain variable domain of said second immunoglobulin;

V _D4 is the heavy or light chain variable domain of said first immunoglobulin;

C _L is an immunoglobulin light chain constant domain;

C _H1 is the C _H1 constant domain of an immunoglobulin heavy chain;

FC represents the immunoglobulin hinge region and the CH ₂ , CH ₃ constant domains of the immunoglobulin heavy chain from immunoglobulin;

L ₁ , L ₂ , L3 and L4 are amino acid linkers;

and wherein the polypeptides of formula I and the polypeptides of formula III form a light chain-heavy chain crossover pair, wherein both VD1 and VD2 are either light chain variable domains or heavy chain variable domains, and both V _D3 and V _D4 are variable heavy chain domains if V _D1 and V _D2 are light chain variable domains, or both V _D3 and V _D4 are light chain variable domains if V _D1 and V _D2 are variable

- 41 041572 heavy chain domains.

In this CODV-Ig format, two polypeptide chains having the structure represented by formula [III] form a dimer with their Fc domains.

In a further embodiment, a first Fc domain is added to the CODV-Fab antibody-like binding protein formula [II] polypeptide and a second _Fc domain (referred to as _Fc2 ) is added to the CODV-Fab antibody-like binding protein formula [I] polypeptide. In addition, in the same embodiment, an L ₅ linker is present between the C _L and P _c2 domain of the polypeptide chains of formula [I], resulting in the polypeptide chains of formula [IV].

Accordingly, the present invention further relates to an antibody-like binding protein comprising two polypeptide chains that form two antigen-binding sites, wherein one polypeptide chain has the structure represented by the formula [IV] ^V D1 ^-L 1 ^-V D2 ^-L 2 ^-C L ^-L 5 ^-F c2 [IV], and one polypeptide chain has the structure represented by the formula [III]

Vd3-L3-Vd4-L4-Chi-Fc [III], where V _D1 represents the variable domain of the heavy or light chain of the first immunoglobulin;

V _D2 is the heavy or light chain variable domain of the second immunoglobulin;

V _D3 is the heavy or light chain variable domain of said second immunoglobulin;

V _d4 is the heavy or light chain variable domain of said first immunoglobulin;

CL is an immunoglobulin light chain constant domain;

CH1 is the CH1 constant domain of an immunoglobulin heavy chain;

F _c represents the immunoglobulin hinge region and the CH2, _CH3 constant domains of the immunoglobulin heavy chain from immunoglobulin;

F _c2 is the immunoglobulin hinge region and the CH2, _CH3 constant domains of the immunoglobulin heavy chain from immunoglobulin;

L1, L2, L ₃ , L4 and L ₅ are amino acid linkers;

wherein the polypeptide of formula [IV] and the polypeptide of formula [III] form a light chain-heavy chain crossover pair, wherein both V _D1 and V _D2 are either light chain variable domains or heavy chain variable domains, and both V _D3 and V _d4 are heavy chain variable domains if V _D1 and V _D2 are light chain variable domains, or both V _D3 and V _d4 are light chain variable domains if V _D1 and V _D2 are heavy chain variable domains.

This CODV format, in which the polypeptide chains represented by formulas [III] and [IV] form a dimer with their respective F _c2 and F _c regions, is referred to herein as CODV-Fab-TL.

In another embodiment, the CODV-Fab first Fc domain is added to the polypeptide chain represented by formula [II] (resulting in formula [III]) and the antibody-like binding protein contains a third polypeptide chain containing or consisting of a second Fc domain (referred to as Fc3).

The present invention further relates to an antibody-like binding protein that contains three polypeptide chains that form two antigen-binding sites, where the first polypeptide has the structure represented by the formula [I] ^V D1 ^-L 1 ^-V D2 ^-L 2 ^-C L [I], the second polypeptide has the structure represented by the formula [III]

Vd3-L3-Vd4-L4-Chi-Fc [III], the third F _c3 polypeptide (also called Fc-stump) is the immunoglobulin hinge region and constant domains CH ₂ , CH ₃ of the immunoglobulin heavy chain from immunoglobulin;

where V _D1 represents the variable domain of the heavy or light chain of the first immunoglobulin;

V _D2 is the heavy or light chain variable domain of the second immunoglobulin;

V _D3 is the heavy or light chain variable domain of said second immunoglobulin;

V _d4 is the heavy or light chain variable domain of said first immunoglobulin;

CL is an immunoglobulin light chain constant domain;

CH1 is the CH1 constant domain of an immunoglobulin heavy chain;

Fc is the immunoglobulin hinge region and the CH2, CH3 constant domains of the immunoglobulin heavy chain from immunoglobulin;

L1, L2, L ₃ and L4 are amino acid linkers;

- 42 041572 and at the same time the polypeptide of formula [I] and the polypeptide of formula [III] form a light chain-heavy chain crossover pair, and both V _D1 and V _D2 are either light chain variable domains or heavy chain variable domains, and both V _D3 and V _D4 are heavy chain variable domains if V _D1 and V _D2 are light chain variable domains, or both V _D3 and V _D4 are light chain variable domains if V _D1 and V _D2 are variable domains heavy chains;

and wherein the polypeptide of formula [III] forms a heterodimer with the third polypeptide via its Fc domain.

Accordingly, in this embodiment, the so-called Fc-stump (F _c3 ) forms a heterodimer with the Fc region of the polypeptide according to formula III. This CODV format is referred to in this document as CODV-Fab-OL. This design prevents the formation of CODV-Fab aggregates.

In one embodiment of the CODV-Fab-OL, both _Fc and _Fc3 are immunoglobulin variants in which the CH3 domain has been modified: each of the _Fc and _Fc3 has been genetically engineered at the CH3-CH3 interface to promote heteropolymer formation according to the so-called ridge-to-bottom technology, which has been described, in particular, in patents US5731168 and US8216805 and which is incorporated herein by reference.

Accordingly, in an embodiment, the CH3 domain of one of _Fc and _Fc3 contains mutations Y349C, T366S, L368A, and Y407V, while the CH3 domain of the other of _Fc and _Fc3 contains mutations S354C and T366W (the amino acid position is indicated relative to the IgG1 sequence). ).

Examples of suitable pairs of F _c and F _c3 include the pair of SEQ ID NO: 396 (F _c ) and SEQ ID NO: 397 (F _c3 ) and the pair of SEQ ID NO: 394 (Fc) and SEQ ID NO: 398 (Fc ₃ ).

In one embodiment of the present invention, the first immunoglobulin or the second immunoglobulin is a single anti-CD3 antibody as defined in the Anti-CD3 Antibodies section above.

In another embodiment of the present invention, the first immunoglobulin or the second immunoglobulin is a single anti-CD123 antibody as defined in the Anti-CD123 section above.

According to one embodiment of the present invention, both V _D1 and V _D2 from a polypeptide of formula I or formula [IV] are either light chain variable domains or heavy chain variable domains, and both V _D3 and V _D4 from a polypeptide II or III are variable heavy chain or light chain domains. This interchangeability is also referred to as exchangeability and thus defines the double variable region (CODV) crossover configuration of the antibody-like binding proteins of the present invention.

According to the above definition, VD1 and VD4 are the heavy or light chain variable domains of the first immunoglobulin, and VD2 and VD3 are the heavy or light chain variable domains of the second immunoglobulin, therefore, VD1 and VD4 are considered as related domains, as well as V _D2 and V _D3 .

Accordingly, the term crossover refers to an interchanged alignment of V _D1 or V _D2 from a formula [I] or formula [IV] polypeptide relative to its cognate variable domain V _D4 or V _D3 from a formula [II] or formula [III] polypeptide.

In one particular embodiment, V _D1 and V _D2 are light chain variable domains and V _D3 and V _D4 are heavy chain variable domains.

The antibody-like binding proteins of the present invention can be made using domains or sequences derived from or derived from any non-human human or animal antibody, including, for example, a human antibody, a rat antibody, or a humanized antibody.

In one embodiment, the immunoglobulin is an IgG immunoglobulin.

Accordingly, in one embodiment, C _L is an IgG immunoglobulin light chain constant domain. In a further embodiment, C _H1 is the C _H1 heavy chain constant domain of an IgG immunoglobulin.

In one embodiment, an antibody-like binding protein of the present invention can be obtained using the domains or sequences from an anti-CD3 antibody and an anti-CD123 antibody described herein.

The term linker, as used herein, refers to one or more amino acid residues inserted between the immunoglobulin domains to provide sufficient mobility for the light and heavy chain domains to fold into immunoglobulins with a dual variable region crossover region. In some embodiments, the implementation of the linker consists of 0 amino acids, which means that the linker is missing. The linker is inserted into the transition zone between variable domains or between variable and constant domains, respectively, at the sequence level. The transition zone between the domains can be identified because the approximate size of the immunoglobulin domains is well known. The exact location of the transition zone of domains can be determined by determining the position of peptide stretches that do not form elements with secondary structure, such as beta sheets or alpha helices, as shown by experimental data or as suggested by modeling techniques or secondary structure predictions. Linkers described in the context of the present invention are linkers L1, L ₂ , L ₃ , L ₄ and L ₅ . L1 is located between the N-terminus of the V _D1 domain and the V _D2 domain; L2 is located between the V _D2 domain and the C-terminus of the CL domain. Linkers L ₃ and L ₄ are located in the polypeptide, defined according to the formula II or III of antibody-like proteins. More precisely, L ₃ is located between the N-terminus of the V _D3 domain and the V _D4 domain, and L ₄ is located between the V _D4 domain and the C-terminus of the C _H1 Fc domain. L ₅ is located between CL and the N-terminus of F _c2 . Linkers L1, L ₂ , L ₃ , L ₄ and L ₅ are independent, but in some embodiments, they have the same sequence and/or length.

In some antibody-like binding proteins of the present invention, the length of L ₃ is at least twice the length of L1. In other antibody-like binding proteins of the present invention, the length of L4 is at least twice the length of L2. In some antibody-like binding proteins of the present invention, the L ₁ length is at least twice the L ₃ length. In other antibody-like binding proteins of the present invention, the length of L ₂ is at least twice the length of L ₄ .

In one embodiment, the linkers L1, L ₂ , L ₃ and L ₄ contain 0-20 amino acids. In one embodiment, L ₅ contains 0-10 amino acids.

In some antibody-like binding proteins of the present invention, L ₁ is 3-12 amino acid residues long, L2 is 3-14 amino acid residues long, L ₃ is 1-8 amino acid residues long, and L ₄ is 1-3 amino acid residues long. In other antibody-like binding proteins, Li is 5-10 amino acid residues long, L2 is 5-8 amino acid residues long, L ₃ is 1-5 amino acid residues long, and L ₄ is 1-2 amino acid residues long. In a preferred antibody-like binding protein, L1 is 7 amino acid residues long, L2 is 5 amino acid residues long, L ₃ is 1 amino acid residue long and L ₄ is 2 amino acid residues long.

In some antibody-like binding proteins of the present invention, L ₁ is 1-3 amino acid residues long, L ₂ is 1-4 amino acid residues long, L ₃ is 2-15 amino acid residues long, and L ₄ is 2-15 amino acid residues long. In other antibody-like binding proteins, L1 is 1-2 amino acid residues long, L2 is 1-2 amino acid residues long, L3 is 4-12 amino acid residues long, and L4 is 2-12 amino acid residues long. In a preferred antibody-like binding protein, L1 is 1 amino acid residue long, L2 is 2 amino acid residue long, L3 is 7 amino acid residue long and L4 is 5 amino acid residue long.

In some antibody-like binding proteins of the present invention, the length of L1, L ₃ or L ₄ may be zero. However, in antibody-like binding proteins in which the L ₃ or L ₄ length is zero, the corresponding length of the transient linker between the variable region and the constant region, or between two variable domains in the other chain, cannot be zero. In some embodiments, the length of L ₁ is zero and L ₃ is 2 or more amino acid residues, the length of L ₃ is zero and the length of L ₁ is 1 or more amino acid residues, or the length of L ₄ is 0 and L ₂ is 3 or more amino acid residues.

In some antibody-like binding proteins of the present invention, at least one of the linkers selected from the group consisting of L2, L3 and L4 contains at least one cysteine residue.

Examples of suitable linkers include a single glycine, threonine or serine residue; a dipeptide such as a diglycine peptide, a histidine-threonine peptide, or a glycine serine dipeptide; a tripeptide of three glycine residues, a Thr-His-Thr tripeptide, a Gly-Gly-Ser tripeptide; a peptide of four glycine residues; a peptide of five glycine residues; a peptide of six glycine residues; a peptide of seven glycine residues; peptide of eight glycine residues. Other combinations of amino acid residues can be used, such as Gly-Gly-Gly-Ser peptide (SEQ ID NO: 354), Gly-Gly-Gly-Gly-Ser peptide (SEQ ID NO: 344), Ser-Gly-Gly- Gly-Gly-Ser (SEQ ID NO: 355), Gly-Ser-Gly-Gly-Gly-Gly-Ser peptide (SEQ ID NO: 356), Gly-Gly-Ser-Gly-Gly-Gly-Gly- peptide Ser (SEQ ID NO: 357), peptide Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser (SEQ ID nO: 358), and peptide Gly-Gly-GlyGly-Ser-Gly-Gly-Gly- Gly-Ser (SEQ ID NO: 345). Other suitable linkers include a single Ser and Val residue; dipeptide Arg-Thr, Gin-Pro, Ser-Ser, Thr-Lys, and Ser-Leu; Lys-Thr-His-Thr (SEQ ID NO: 359); Lys-Thr-His-Thr-Ser (SEQ ID NO: 360); Asp-Lys-Thr-His-Thr-Ser (SEQ ID NO: 361); Asp-Lys-Thr-His-ThrSer-Pro (SEQ ID NO: 362); Ser-Asp-Lys-Thr-His-Thr-Ser-Pro (SEQ ID NO: 363); Ser-Asp-Lys-Thr-His-ThrSer-Pro-Pro (SEQ ID NO: 364); Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser (SEQ ID NO: 365); Pro-LysSer-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser (SEQ ID NO: 366); Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-ProSer-Pro (SEQ ID NO: 367); Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro (SEQ ID NO: 368); Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-Gly (SEQ ID NO: 369); Gly-Glu-Pro-Lys-Ser-Asp

- 44 041572

Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-Gly (SEQ ID NO: 370); Gly-Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-SerPro-Pro-Ser-Pro-Gly-Gly (SEQ ID NO: 371); Gly-Gly-Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-SerPro-Gly-Gly (SEQ ID NO: 372); Gly-Gly-Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-GlyGly-Gly (SEQ ID NO: 373); Gly-Gly-Gly-Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-GlyGly-Gly (SEQ ID NO: 374); Thr-Val-Ala-Ala-Pro (SEQ ID NO: 346), Gln-Pro-Lys-Ala-Ala (SEQ ID NO: 347), Gln-Arg-Ile-Glu-Gly (SEQ ID NO: 348) ; Ala-Ser-Thr-Lys-Gly-Pro-Ser (SEQ ID NO: 349), Arg-Thr-Val-AlaAla-Pro-Ser (SEQ ID NO: 350), Gly-Gln-Pro-Lys-Ala- Ala-Pro (SEQ ID NO: 307), Thr-Lys-Gly-Pro-Ser (SEQ ID NO: 309), His-Ile-Asp-Ser-Pro-Asn-Lys (SEQ ID NO: 351), and Gly-Gly-Ser-Gly-Ser-Ser-Gly-Ser-Gly-Gly (SEQ ID NO: 389). The examples listed above are not intended to limit the scope of the present invention in any way, and it has been shown that linkers containing randomly selected amino acids selected from the group consisting of valine, leucine, isoleucine, serine, threonine, lysine, arginine, histidine , aspartate, glutamate, asparagine, glutamine, glycine, and proline are suitable for the antibody-like binding proteins of the present invention.

The composition and sequence of amino acid residues in the linker may vary depending on the type of element of the secondary structure, which must be achieved using the linker. For example, glycine, serine and alanine are the most suitable linkers for maximum flexibility. Some combinations of glycine, proline, threonine and serine are useful if a stiffer and longer linker is needed. If necessary, any amino acid residue can be considered as a linker in combination with other amino acid residues to construct longer peptide linkers, depending on the desired properties.

In one embodiment, the L1 linker has the sequence Gly-Gln-Pro-Lys-Ala-Ala-Pro (SEQ ID NO: 307), the L ₂ linker has the sequence Thr-Lys-Gly-Pro-Ser (SEQ ID NO: 309) , linker L ₃ has the sequence S, and linker L ₄ has the sequence RT.

In a further embodiment, the linker sequences L1, L ₂ , L ₃ and L4 are selected from the group consisting of threonine; a dipeptide such as histidine-theronine peptide; tripeptide Thr-His-Thr, Lys-Thr-His-Thr (SEQ ID NO: 359); Lys-Thr-His-Thr-Ser (SEQ ID NO: 360); Asp-Lys-Thr-HisThr-Ser (SEQ ID NO: 361); Asp-Lys-Thr-His-Thr-Ser-Pro (SEQ ID NO: 362); Ser-Asp-Lys-Thr-His-Thr-SerPro (SeQ ID NO: 363); Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro (SEQ ID NO: 364); Lys-Ser-Asp-Lys-Thr-HisThr-Ser-Pro-Pro-Ser (SEQ ID NO: 365); Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser (SEQ ID NO: 366); Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro (SEQ ID NO: 367); Glu-Pro-Lys-Ser-Asp-LysThr-His-Thr-Ser-Pro-Pro-Ser-Pro (SEQ ID NO: 368); Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-SerPro-Gly (SEQ ID NO: 369); Gly-Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-Gly (SEQ ID NO: 370); Gly-Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-Gly-Gly (SEQ ID NO: 371); Gly-GlyGlu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-Gly-Gly (SEQ ID NO: 372); Gly-Gly-Glu-ProLys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-Gly-Gly-Gly (SEQ ID NO: 373) and Gly-Gly-Gly-Glu -ProLys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-Gly-Gly-Gly (SEQ ID NO: 374). In one embodiment, the L ₅ linker sequence is selected from the group consisting of a single serine residue, a dipeptide such as a glycine-serine dipeptide; Gly-Gly-Ser tripeptide, GlyGly-Gly-Ser peptide (SEQ ID NO: 354), Gly-Gly-Gly-Gly-Ser peptide (SEQ ID NO: 344), Ser-Gly-Gly-GlyGly-Ser peptide ( SEQ ID NO: 355), Gly-Ser-Gly-Gly-Gly-Gly-Ser peptide (SEQ ID NO: 356), Gly-Gly-SerGly-Gly-Gly-Gly-Ser peptide (SEQ ID NO: 357) , Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser peptide (SEQ ID NO: 358), Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser peptide (SEQ ID NO: 345) and the Gly-Gly-Ser-Gly-Ser-SerGly-Ser-Gly-Gly peptide (SEQ ID NO: 389).

The term Fc domain as used herein encompasses native Fc as well as Fc variants and sequences as defined above. With regard to Fc variants and native Fc molecules, the term Fc domain includes molecules in monomeric or multimeric form, whether formed by cleavage of the whole antibody or obtained by other means.

The term native Fc, as used herein, refers to a molecule containing a sequence that is not an antigen-binding fragment, which is obtained by cleavage of an antibody or obtained by other methods, regardless of whether it is in monomeric or multimeric form, and it may contain hinge area. The original immunoglobulin source of the native Fc is particularly human-derived and may be any of the immunoglobulins, although IgG1 and IgG2 are preferred. Native Fc molecules are composed of monomeric polypeptides that can be linked into dimeric or multimeric forms via covalent (ie disulfide bonds) and non-covalent bonds. The number of intermolecular disulfide bonds between the monomeric subunits of native Fc molecules varies from 1 to 4, depending on the class (eg IgG, IgA and IgE) or subclass (eg IgGl, IgG2, IgG3, IgA1 and IgGA2). One example of a native Fc is a disulfide dimer resulting from papain cleavage of IgG. The term native Fc, as used herein, is generic to monomeric, dimeric, and multimeric forms.

The term Fc variant, as used herein, refers to a molecule or sequence that is modified from the native Fc but still contains the binding site

- 45 041572 for the recycling receptor, FcRn (neonatal Fc receptor). Exemplary Fc variants and their interaction with the recycling receptor are known in the art. Thus, the term Fc variant may include a molecule or sequence that is humanized based on a native Fc that is derived from a non-human molecule. In addition, native Fc contains regions that can be removed because they provide structural features or biological activity that are not required for the antibody-like binding proteins of the present invention. Thus, the term Fc variant encompasses a molecule or sequence that lacks one or more native Fc sites or residues or that has been modified with one or more Fc sites or residues that affect or are involved in: (1) disulfide bond formation, (2 ) incompatibility with the selected host cell, (3) N-terminal heterogeneity when expressed in the selected host cell, (4) glycosylation, (5) interaction with complement, (6) binding to an Fc receptor other than a salvage receptor, or (7) antibody-dependent cellular cytotoxicity (ADCC).

In some embodiments, when the antibody-like binding protein contains two Fc domains, i.e. in CODV-Ig (F _c and F _c2 ), CODV-Fab-TL (two Fc domains), and CODV-Fab-OL (F _c and F _c3 ), the two Fc domains belong to the same immunoglobulin isotype or isotype subclass. Accordingly, in some embodiments, both _Fc and _Fc2 from CODV-Ig, or both Fc domains from CODV-Fab-TL, or both _Fc and _Fc3 from CODV-Fab-OL belong to the IgG1 subclass, or the IgG2 subclass , or an IgG3 subclass, or an IgG4 subclass.

All antibody-like binding proteins described herein do not have an effector function. This means that when an antibody-like binding protein contains one or more Fc domains (i.e., _Fc in formula [III], _Fc2 in formula [IV] and/or _Fc3 ) belonging to the IgG1 subclass, then said one or several IgG1 backbone Fc domains contain a L234A and L235A double mutation (called the LALA mutation) that disrupts F _c effector function. Fc with double mutation L234A and L235A does not bind FcyR or C1q, and both ADCC and CDC functions of the IgGl subclass Fc domain are impaired (Hezareh, M. et al., J Virol. 2001 Dec; 75(24): 12161-12168).

In one example, the Fc region contains the amino acid sequence under SEQ ID NO: 330, SEQ ID NO: 394, or SEQ ID NO: 396. In one embodiment, the E _c2 region contains two amino acid substitutions within the CH3 domain, H435R and Y436F, as described in Jendeberg, L. et al. (1997, J. Immunological Meth. 201: 25-34). Accordingly, in one embodiment, the F _c2 region contains the amino acid sequence of SEQ ID NO: 327. In another embodiment, the F _c2 region contains the amino acid sequence of SEQ ID NO: 392.

The antibody-like binding protein, called CODV-Fab 7G3x20G6, contains one polypeptide according to formula I with the amino acid sequence

DWMTQTPVSLSVSLGGQVSISCRSSQSLVHNNGNTYLSWYLQKPGQSPQSLIYKVSNRFSGFS

DRFSGSGSGTDFTLKISRRVDPDDLGVYYCGOGTOYPFTFGSGTKLEIKGQPKAAPDFVMTOSPS

SLTVTAGEKVTMSCKSSQSLLNSGNQKNYLTWYLQKPGQPPKLLIYWASTRESGVPDRFTGSGS GTDFTLTISSVOAEDLAVYYCONDYSYPYTFGGGTKLEIKTKGPSRTVAAPSVFIFPPSDEOLK SGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHK VYACEVTHQGLSSPVTKSFNRGEC (SEQ ID NO: 306, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D1 с последовательностью под SEQ ID NO: 9, L ₁ с последовательностью под SEQ ID NO: 307, V _D2 с последовательностью under SEQ ID NO: 308, L2 with the sequence under SEQ ID NO: 309 and C _L with the sequence under SEQ ID NO: 310, and one polypeptide according to formula II with the amino acid sequence

EVQLQQSGPELVKPGASVKMSCKASGYTFTDYYMKWVKQSHGKSLEWIGDIIPSNGATFYNQKF

KGKATLTVDRSSSTAYMHLNSLTSEDSAVYYCTRSHLLRASWFAYWGQGTLVTVSASEVQLVET GGSLVQPGKSLKLTCATSGFTFTKAWMHWVRQSPEKQLEWVAQIKDKSNSYATYYAESVKGRFT ISRDDSKSTIYLQMNSLKEEDTAIYYCRGVYYALSPFDYWGQGVMVTVSSRTASTKGPSVFPLA PSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSWTVPSSSL GTQTYICNVNHKPSNTKVDKKV (SEQ ID NO: 311, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D3 с последовательностью под SEQ ID NO: 312, L ₃ с аминокислотной последовательностью S, V _D4 с последовательностью под SEQ ID NO: 5, L ₄ with the amino acid sequence RT and C _H1 with the sequence under SEQ ID NO: 313.

The antibody-like binding protein, called CODV-Fab 7G3x4E7, contains one polypeptide according to formula I with the amino acid sequence

- 46 041572

DWMTQTPVSLSVSLGGQVSISCRSSQSLEHNNGNTYLSWYLQKPGQSPQPLIYKVSNRFSGIS DRFSGSGSGTDFTLKISRVEPDDLGVYYCGOGTQYPFTFGPGTKLELKGQPKAAPDFVMTOSPS SLTVTAGEKVTMSCKSSQSLLNSGNQKNYLTWYLQKPGQPPKLLIYWASTRESGVPDRFTGSGS GTDFTLTISSVOAEDLAVYYCONDYSYPYTFGGGTKLEIKTKGPSRTVAAPSVFIFPPSDEOLK

SGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHK

VYACEVTHQGLS S PVTKS FNRGEC (SEQ ID NO: 314, with linkers shown in bold and underlined) containing V _D1 with the sequence of SEQ ID NO: 21, L ₁ with the sequence of SEQ ID NO: 307, V _D2 with the sequence under SEQ ID NO: 308, L ₂ with the sequence under SEQ ID NO: 309 and CL with the sequence under SEQ ID NO: 310, and one polypeptide according to formula II with the amino acid sequence

EVQLQQSGPELVKPGASVKMSCKASGYTFTDYYMKWVKQSHGKSLEWIGDIIPSNGATFYNQKF KGKATLTVDRSSSTAYMHLNSLTSEDSAVYYCTRSHLLRASWFAYWGQGTLVTVSASEVQWET GGSLVQPGKSLKLTCATSGFTFSNAWMHWVRQSPEKQLEWVAQIKDKSNNYATYYAESLKGRFT ISRDDPKRSIYLQMNSLREEDTAIYYCRYVHYGIGYAMDAWGQGTSVTVSSRTASTKGPSVFPL APSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSWTVPSSS LGTQTYICNVNHKPSNTKVDKKV (SEQ ID NO: 315, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D3 с последовательностью под SEQ ID NO: 312, L ₃ с аминокислотной последовательностью S, V _D4 с последовательностью под SEQ ID NO: 18, L4 with the amino acid sequence RT and C _H1 with the sequence under SEQ ID NO: 313.

The antibody-like binding protein, called CODV-Fab 7G3x4B4, contains: one polypeptide according to formula I with the amino acid sequence

DWMTQTPVSLSVSLGGQVSISCRSSQSLVHDNGNTYLSWSLQRPGQSPQVLIYKVSNRFSGTS DRFTGSGSGTDFTLKISRVEPDDLGVYYCGOGTQYPFTFGSGTKLEIKGQPKAAPDFVMTOSPS SLTVTAGEKVTMSCKSSQSLLNSGNQKNYLTWYLQKPGQPPKLLIYWASTRESGVPDRFTGSGS GTDFTLTISSVOAEDLAVYYCONDYSYPYTFGGGTKLEIKTKGPSRTVAAPSVFIFPPSDEOLK SGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHK VYACEVTHQGLSSPVTKSFNRGEC (SEQ ID NO: 316, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D1 с последовательностью под SEQ ID NO: 16, L ₁ с последовательностью под SEQ ID NO: 307, V _D2 with the sequence under SEQ ID NO: 308, L ₂ with the sequence under SEQ ID NO: 309 and CL with the sequence under SEQ ID NO: 310, and one polypeptide according to formula II with the amino acid sequence

EVQLQQSGPELVKPGASVKMSCKASGYTFTDYYMKWVKQSHGKSLEWIGDIIPSNGATFYNQKF KGKATLTVDRSSSTAYMHLNSLTSEDSAVYYCTRSHLLRASWFAYWGQGTLVTVSASEVQLVET GGRLVQPGRSLKLTCATSGFTFSNAWMHWVRQSPEKQLEWVAQIKARSNNYATYYAESVKGRFT ISRDDSKSTIYLQMNSLKEEDTAIYYCRGTYYASKPFDYWGQGVMVTVSSRTASTKGPSVFPLA PSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSWTVPSSSL GTQTYICNVNHKPSNTKVDKKV (SEQ ID NO: 317, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D3 с последовательностью под SEQ ID NO: 312, L ₃ с аминокислотной последовательностью S, V _D4 с последовательностью под SEQ ID NO: 12, L4 with the amino acid sequence RT and C _H1 with the sequence under SEQ ID NO: 313.

The antibody-like binding protein, called CODV-Fab 7G3x18F5, contains one polypeptide according to formula I with the amino acid sequence

DVLMTQTPVSLSVSLGQVSISCRSSQSLVHTNGNTYLSWYLQKPGQSPQLLIYKVSNRLSGIS

DRFSGSGSGTDFTLKISRVEPDDLGVYYCGOGTHYPFTFGAGTKLELKGQPKAAPDFVMTQSPS SLTVTAGEKVTMSCKSSQSLLNSGNQKNYLTWYLQKPGQPPKLLIYWASTRESGVPDRFTGSGS GTDFTLTISSVOAEDLAVYYCONDYSYPYTFGGGTKLEIKTKGPSRTVAAPSVFIFPPSDEQLK SGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHK VYACEVTHQGLSSPVTKSFNRGEC (SEQ ID NO: 318, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D1 с последовательностью под SEQ ID NO: 26, L ₁ с последовательностью под SEQ ID

- 47 041572

NO: 307, V _D2 with the sequence under SEQ ID NO: 308, L ₂ with the sequence under SEQ ID NO: 309 and _CL with the sequence under SEQ ID NO: 310, and one polypeptide according to formula II with the amino acid sequence

EVQLQQSGPELVKPGASVKMSCKASGYTFTDYYMKWVKQSHGKSLEWIGDIIPSNGATFYNQKF KGKATLTVDRSSSTAYMHLNSLTSEDSAVYYCTRSHLLRASWFAYWGQGTLVTVSASEVQWET GGSLVQPGKSLKLTCATSGFTFTNAWMHWVRRSPEKQLEWVAQIKDKSNNYATYYAESVKGRFT ISRDDSKSSIYLQMNSLKEEDTAIYYCRYVHYRFAYALDAWGRGTSVSVSSRTASTKGPSVFPL APSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSWTVPSSS LGTQTYICNVNHKPSNTKVDKKV (SEQ ID NO: 319, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D3 с последовательностью под SEQ ID NO: 312, L ₃ с аминокислотной последовательностью S, V _D4 с последовательностью под SEQ ID NO: 23, L ₄ with the amino acid sequence RT and C _H1 with the sequence under SEQ ID NO: 313.

The antibody-like binding protein, called CODV-Fab hz20G6x7G3, contains one polypeptide according to formula I with the amino acid sequence

DFVMTQSPSSLTVTAGEKVTMSCKSSQSLLNSGNQKNYLTWYLQKPGQPPKLLIYWASTRESGV PDRFTGSGSGTDFTLTISSVOAEDLAVYYCONDYSYPYTFGGGTKLEIKGQPKAAPDIVMTOTP LSLSVTPGQPASISCKSSQSLVHNNGNTYLSWYLQKPGQSPQSLIYKVSNRFSGVPDRFSGSGS GTDFTLKISRVEAEDVGVYYCGOGTOYPFTFGSGTKVEIKTKGPSRTVAAPSVFIFPPSDEQLK SGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHK VYACEVTHQGLSSPVTKSFNRGEC (SEQ ID NO: 320, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D1 с последовательностью под SEQ ID NO: 308, L1 с последовательностью под SEQ ID NO: 307, V _D2 с sequence under SEQ ID NO: 143, L2 with the sequence under SEQ ID NO: 309 and C _L with the sequence under SEQ ID NO: 310, and one polypeptide according to formula II with the amino acid sequence

QVQLVESGGGWQPGRSLRLSCAASGFTFTKAWMHWVRQAPGKQLEWVAQIKDKSNSYATYYAD SVKGRFTISRDDSKNTLYLQMNSLRAEDTAVYYCRGVYYALSPFDYWGQGTLVTVSSSEVQLQQ SGPELVKPGASVKMSCKASGYTFTDYYMKWVKQSHGKSLEWIGDIIPSNGATFYNQKFKGKATL TVDRSSSTAYMHLNSLTSEDSAVYYCTRSHLLRASWFAYWGQGTLVTVSARTASTKGPSVFPLA PSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSWTVPSSSL GTQTYICNVNHKPSNTKVDKKV (SEQ ID NO: 321, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D3 с последовательностью под SEQ ID NO: 138, L ₃ с аминокислотной последовательностью S, V _D4 с последовательностью под SEQ ID NO: 312, L ₄ with the amino acid sequence RT and C _H1 with the sequence under SEQ ID NO: 313.

The antibody-like binding protein, called CODV-Fab 7G3xhz4B4, contains one polypeptide according to formula I with the amino acid sequence

DWMTQTPVSLSVSVGGRVSISCRSSQSLVHDNGNTYLSWSLQKPGKSPKVLIYKVSNRFSGVS SRFTGSGSGTDFTLKISSVOPDDLGVYYCGQGTOYPFTFGSGTKLEIKGQPKAAPDFVMTOSPS SLTVTAGEKVTMSCKSSQSLLNSGNQKNYLTWYLQKPGQPPKLLIYWASTRESGVPDRFTGSGS GTDFTLTISSVOAEDLAVYYCONDYSYPYTFGGGTKLEIKTKGPSRTVAAPSVFIFPPSDEOLK SGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHK VYACEVTHQGLSSPVTKSFNRGEC (SEQ ID NO: 322, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D1 с последовательностью под SEQ ID NO: 158, L1 с последовательностью под SEQ ID NO: 307, VD2 с последовательностью under SEQ ID NO: 308, L2 with the sequence under SEQ ID NO: 309 and C _L with the sequence under SEQ ID NO: 310, and one polypeptide according to formula II with the amino acid sequence

- 48 041572

EVQLQQSGPELVKPGASVKMSCKASGYTFTDYYMKWVKQSHGKSLEWIGDIIPSNGATFYNQKF KGKATLTVDRSSSTAYMHLNSLTSEDSAVYYCTRSHLLRASWFAYWGQGTLVTVSASQVQLVET GGGLVKPGQSLKLTCATSGFTFSNAWMHWVRQSPEKGLEWVAQIKARSNNYATYYAESVKGRFT ISRDDSKSTIYLQMNSLTPEDTAIYYCRGTYYASKPFDYWGQGVMVTVSSRTASTKGPSVFPLA PSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSWTVPSSSL GTQTYICNVNHKPSNTKVDKKV (SEQ ID NO: 323, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D3 с последовательностью под SEQ ID NO: 312, L ₃ с аминокислотной последовательностью S, V _D4 с последовательностью под SEQ ID NO: 171, L ₄ with the amino acid sequence RT and C _H1 with the sequence under SEQ ID NO: 313.

The antibody-like binding protein, called CODV-Fab hz4B4x3E3, contains one polypeptide according to formula I with the amino acid sequence

QFVLTQPNSVSTNLGSTVKLSCKRNTGNIGSNYVNWYQQHEGRSPTTMIYRDDKRPDGVPDRFS GSIDRSSNSALLTINNVQTEDEADYFCQSYSSGINIIFGGGTKLTVLGQPKAAPDWMTQTPVS LSVSVGGRVSISCRSSQSLVHDNGNTYLSWSLQKPGKSPKVLIYKVSNRFSGVSSRFTGSGSGT DFTLKISSVOPDDLGVYYCGOGTOYPFTFGSGTKLEIKTKGPSRTVAAPSVFIFPPSDEOLKSG TASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVY ACEVTHQGLS S PVTKS FNRGEC (SEQ ID NO: 32 4, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D1 с последовательностью под SEQ ID NO: 230, L1 с последовательностью под SEQ ID NO: 307 , V _D2 with the sequence under SEQ ID NO: 158, L2 with the sequence under SEQ ID NO: 309 and _CL with the sequence under SEQ ID NO: 310, and one polypeptide according to formula II with the amino acid sequence

QVQLVETGGGLVKPGQSLKLTCATSGFTFSNAWMHWVRQSPEKGLEWVAQIKARSNNYATYYAE SVKGRFTISRDDSKSTIYLQMNSLTPEDTAIYYCRGTYYASKPFDYWGQGVMVTVSSSQVQLQE SGPGLVQPSQTLSLTCTVSGFSLTTYDVHWVRQPPGKGLEWMGRIQNGGITDYNSALKSRLIIS RDTSKSQVFLKMNSVQTEDTAMYFCAKTGSYFYAFDHWGQGTLVTVSSRTASTKGPSVFPLAPS SKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSWTVPSSSLGT QTYICNVNHKPSNTKVDKKV (SEQ ID NO: 325, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D3 с последовательностью под SEQ ID NO: 171, L ₃ с аминокислотной последовательностью S, V _D4 с последовательностью под SEQ ID NO: 226, L ₄ with the amino acid sequence RT and C _H1 with the sequence under SEQ ID NO: 313.

The antibody-like binding protein, called CODV-Fab hz20G6xhz7G3, contains one polypeptide according to formula [I], actually consisting of the amino acid sequence

DIVMTQSPDSLAVSLGERATINCESSQSLLNSGNQKNYLTWYQQKPGQPPKPLIYWAST RESGVPDRFSGSGSGTDFTLTISSLOAEDVAVYYCONDYSYPYTFGOGTKLEIKGGSGSSGSGG DIVMTQTPLSLSVTPGQPASISCKSSQSLVHNNANTYLSWYLQKPGQSPQSLIYKVSNRFSGVP DRFSGSGSGTDFTLKISRVEAEDVGVYYCGOGTOYPFTFGSGTKVEIKGGSGSSGSGGRTVAAP SVFIFPPSDEQLKSGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSS TLTLSKADYEKHKVYACEVTHQGLS S PVTKS FNRGEC (SEQ ID NO: 388, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D1 с последовательностью под SEQ ID NO: 385, L1 с последовательностью под SEQ ID NO: 389, V _D2 with the sequence under SEQ ID NO: 141, L2 with the sequence under SEQ ID NO: 389 and _CL with the sequence under SEQ ID NO: 310, and one polypeptide according to formula [II], actually consisting of the amino acid sequence

QVQLVESGGGVVQPGRSLRLSCAASGFTFTKAWMHWVRQAPGKQLEWVAQIKDKSNSYA TYYADSVKGRFTISRDDSKNTLYLQMNSLRAEDTAVYYCRGVYYALSPFDYWGQGTLVTVSSEV QLVQSGAEVKKPGESLKISCKGSGYSFTDYYMKWARQMPGKGLEWMGDIIPSSGATFYNQKFKG QVTISADKSISTTYLQWSSLKASDTAMYYCARSHLLRASWFAYWGQGTMVTVSSASYRGPSYFP LAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSS SLGTQTYICNVNHKPSNTKVDKKV

- 49 041572 (SEQ ID NO: 390), containing V _D3 with the sequence under SEQ ID NO: 138, and L ₃ represents 0 amino acids, V _D4 with the sequence under SEQ ID NO: 383 (italicized and underlined), and L ₄ is 0 amino acids, and C _H1 with the sequence under SEQ ID

NO: 313.

In another example, the Form II polypeptides of the antibody-like binding proteins termed CODV-Fab 7G3x20G6, 7G3x4E7, 7G3x4B4, 7G3x18F5, hz20G6x7G3, 7G3xhz4B4, hz4B4x3E3 and hz20G6xhz7G3, further comprise the sequence EPKSCDKTHTHHHHHH, which further corresponds to the sequence ID: 35 His-tags, for example, for cleaning.

An antibody-like binding protein called CODV-Fab hz20G6x7G3-TL4 (also called CODV-Fab-TL4 hz20G6x7G3) contains:

one polypeptide according to the formula [IV], actually consisting of the amino acid sequence

DFVMTQSPSSLTVTAGEKVTMSCKSSQSLLNSGNQKNYLTWYLQKPGQPPKLLIYWASTRESGV

PDRFTGSGSGTDFTLTISSVOAEDLAVYYCONDYSYPYTFGGGTKLEIKGQPKAAPDIVMTOTP

LSLSVTPGQPASISCKSSQSLVHNNGNTYLSWYLQKPGQSPQSLIYKVSNRFSGVPDRFSGSGS

GTDFTLKISRVEAEDVGVYYCGOGTOYPFTFGSGTKVEIKTKGPSRTVAAPSVFIFPPSDEOLK

SGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHK

VYACEVTHQGLS S PVTKS FNRGECESKYGPPCPPCPAPEFEGGPSVFLFPPKPKDTLMISRTPE

VTCVWDVSQEDPEVQFNWYVDGVEVHNAKTKPREEQFNSTYRWSVLTVLHQDWLNGKEYKCK

VSNKGLPSSIEKTISKAKGQPREPQVYTLPPSQEEMTKNQVSLTCLVKGFYPSDIAVEWE5NGQ

PENNYKTTPPVLDSDGSFFLYSRLTVDKSRWOEGNVFSCSVMHEALHNRFTQKSLSLSLG (SEQ ID NO: 326, with linkers shown in bold and underlined) containing V _D1 with the sequence of SEQ ID NO: 308, L ₁ with the sequence of SEQ ID NO: 307, V _D2 with the sequence of SEQ ID NO: 143, L2 with the sequence under SEQ ID NO: 309, _CL with the sequence under SEQ ID NO: 310 and F _c2 (underlined) with the sequence under SEQ ID NO: 327, and one polypeptide according to formula [III] with the amino acid sequence

QVQLVESGGGVVQPGRSLRLSCAASGFTFTKAWMHWVRQAPGKQLEWVAQIKDKSNSYATYYAD

SVKGRFTISRDDSKNTLYLQMNSLRAEDTAVYYCRGVYYALSPFDYWGQGTLVTVSSSEVQLQQ

SGPELVKPGASVKMSCKASGYTFTDYYMKWVKQSHGKSLEWIGDIIPSNGATFYNQKFKGKATL

TVDRSSSTAYMHLNSLTSEDSAVYYCTRSHLLRASWFAYWGQGTLVTVSARTASTKGPSVFPLA

PCSRSTSESTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSWTVPSSSL

GTKTYTCNVDHKPSNTKVDKRVESKYGPPCPPCPAPEFEGGPSVFLFPPKPKDTLMISRTPEVT

CVVVDVSQEDPEVQFNWYVDGVEVHNAKTKPREEQFNSTYRVVSVLTVLHQDWLNGKEYKCKVS

NKGLPSSIEKTISKAKGQPREPQVYTLPPSQEEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPE

NNYKTTPPVLDSDGSFFLYSRLTVDKSRWQEGNVFSCSVMHEALHNHYTQKSLSLSLG (SEQ ID NO: 328, with linkers shown in bold and underlined) containing V _D3 with the sequence of SEQ ID NO: 138, L ₃ with the amino acid sequence of S, V _D4 with the sequence of SEQ ID NO: 312, L ₄ with the amino acid sequence RT, as well as C _H1 with the sequence under SEQ ID NO: 329 and F _c with the sequence under SEQ ID NO: 330.

In the above antibody-like binding protein CODV-Fab-TL4 hz20G6x7G3 Fc with the sequence of SEQ ID NO: 330 and _Fc2 with the sequence of SEQ ID NO: 327 are derived from the backbone of IgG4. Said antibody-like binding protein is in CODV-Fab-TL format and contains or consists of one formula III polypeptide and one formula IV polypeptide.

The antibody-like binding protein, called CODV-Fab-TL1 hz20G6xhz7G3, contains one polypeptide according to formula IV with the amino acid sequence

- 50 041572

DIVMTQSPDSLAVSLGERATINCESSQSLLNSGNQKNYLTWYQQKPGQPPKPLIYWAST RESGVPDRFSGSGSGTDFTLTISSLOAEDVAVYYCONDYSYPYTFGOGTKLEIKGGSGSSGSGG DIVMTQTPLSLSVTPGQPASISCKSSQSLVHNNANTYLSWYLQKPGQSPQSLIYKVSNRFSGVP DRFSGSGSGTDFTLKISRVEAEDVGVYYCGOGTOYPFTFGSGTKVEIKGGSGSSGSGGRTVAAP SVFIFPPSDEQLKSGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSS TLTLS KADYEKHKVYACEVTHQGL SS PVTKS FNRGE CDKTHTCPPCPAPEAAGGPSVFLFPPKP KDTLMISRTPEVTCWVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRWSVLTVLHO DWLNGKEYKCKVSNKALPAPIEKTISKAKGOPREPOVYTLPPSRDELTKNQVSLTCLVKGFYPS DIAVEWESNGOPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWOOGNVFSCSVMHEALHNRFTQK SLSLSPG (SEQ ID NO: 391, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D1 с последовательностью под SEQ ID NO: 385, L ₁ с последовательностью under SEQ ID NO: 389, V _D2 with the sequence under SEQ ID NO: 141, L2 with the sequence under SEQ ID NO: 389, C _L with the sequence under SEQ ID NO: 310, L ₅ which contains 0 amino acids, and F _c2 with sequence under SEQ ID NO: 392; and one polypeptide according to formula III with the amino acid sequence

QVQLVE S GGGWQPGRS LRL S CAAS G FT FTKAWMHWVRQAPGKQLEWVAQIKDKS NS YA

TYYADSVKGRFTISRDDSKNTLYLQMNSLRAEDTAVYYCRGVYYALSPFDYWGQGTLVTVSSEV QLVQSGAEVKKPGESLKISCKGSGYSFTDYYMKWARQMPGKGLEWMGDIIPSSGATFYNQKFKG QVTISADKSISTTYLQWSSLKASDTAMYYCARSHLLRASWFAYWGQGTMVTVSSASYRGPSYFP LAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSWTVPSS SLGTQTYICNVNHKP SNTKVDKKVEPKSCDKTHTCPPCPAPEAAGGPSVFLFPPKPKDTLMISR TPEVTCWVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEOYNSTYRWSVLTVLHQDWLNGKEY KCKVSNKALPAPIEKTISKAKGQPREPQVYTLPPSRDELTKNQVSLTCLVKGFYPSDIAVEWES NGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPG (SEQ ID NO: 393), содержащий V _D3 с последовательностью под SEQ ID NO: 138, причем L ₃ представляет собой 0 аминокислот, V _D4 с последовательностью под SEQ ID NO: 383 (выделено курсивом ), where L4 is 0 amino acids, C _H1 with the sequence under SEQ ID NO: 313 and Fc (underlined) with the sequence under SEQ ID NO: 394.

In the antibody-like binding protein CODV-Fab-TL1 hz20G6x7G3 Fc with the sequence of SEQ ID NO: 394 and _Fc2 with the sequence of SEQ ID NO: 392 are derived from the backbone of IgG1. Said antibody-like binding protein is in CODV-Fab-TL format. It contains or consists of one formula IV polypeptide and one formula III polypeptide.

The antibody-like binding protein, called CODV-Fab-OL1 hz20G6xhz7G3, contains one polypeptide according to formula I with the amino acid sequence

DIVMTQSPDSLAVSLGERATINCESSQSLLNSGNQKNYLTWYQQKPGQPPKPLIYWAST RESGVPDRFSGSGSGTDFTLTISSLOAEDVAVYYCONDYSYPYTFGOGTKLEIKGGSGSSGSGG DIVMTQTPLSLSVTPGQPASISCKSSQSLVHNNANTYLSWYLQKPGQSPQSLIYKVSNRFSGVP DRFSGSGSGTDFTLKISRVEAEDVGVYYCGOGTOYPFTFGSGTKVEIKGGSGSSGSGGRTVAAP SVFIFPPSDEQLKSGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSS TLTLSKADYEKHKVYACEVTHQGLS S PVTKS FNRGEC (SEQ ID NO: 388, при этом линкеры показаны символами с полужирным начертанием и подчеркнуты), содержащий V _D1 с последовательностью под SEQ ID NO: 385, L ₁ с последовательностью под SEQ ID NO: 389 , V _D2 with the sequence under SEQ ID NO: 141, L2 with the sequence under SEQ ID NO: 389 and C _L with the sequence under SEQ ID NO: 310; and one polypeptide according to formula III with the amino acid sequence

- 51 041572

QVQLVESGGGWQPGRSLRLSCAASGFTFTKAWMHWVRQAPGKQLEWVAQIKDKSNSYA TYYADSVKGRFTISRDDSKNTLYLQMNSLRAEDTAVYYCRGVYYALSPFDYWGQGTLVTVSSEV QLVQSGAEVKKPGESLKISCKGSGYSFTDYYMKWARQMPGKGLEWMGDIIPSSGATFYNQKFKG QVTISADKSISTTYLQWSSLKASDTAMYYCARSHLLRASWFAYWGQGTMVTVSSASYKGPSVFP LAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSWTVPSS SLGTQTYIC NVNH К PSNT КVD KKVEPKSCDKTHTCPPCPAPEAAGGPSVFLFPPKPKDTLMISR TPEVTCVWDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRWSVL·TVL·HQDWLNGKEY KCKVSNKALPAPIEKTISKAKGQPREPQVYTLPPCRDELTKNQVSLWCLVKGFYPSDIAVEWES NGOPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWOOGNVFSCSVMHEALHNRFTQKSLSLSPG (SEQ ID NO: 395), содержащий V _D3 с последовательностью под SEQ ID NO: 138, причем L ₃ представляет собой 0 аминокислот, V _D4 с последовательностью under SEQ ID NO: 383 (italicized and underlined), with L4 representing 0 amino acids, C _H1 with the sequence under SEQ ID NO: 313 and F _c (underlined) with the sequence under SEQ ID NO: 396;

and wherein the antibody-like binding protein termed CODV-Fab-OL1 hz20G6xhz7G3 further comprises an Fc-stump (F _c3 ) with the amino acid sequence

GSDKTHTCPPCPAPEAAGGPSVFLFPPKPKDTLMISRTPEVTCVWDVSHEDPEVKFNW

YVDGVEVHNAKTKPREEQYNSTYRWSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKG QPREPQVCTLPPSRDELTKNQVSLSCAVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFF LVSKLTVDKSRWQGNVFSCSVMHEALHNHYTQKSLSLSPG (SEQ ID NO: 397) and which forms a heterodimer according to formula III.

Said antibody-like binding protein is in CODV-Fab-OL format, i. e. it contains or consists of one formula I polypeptide, one formula III polypeptide and one Fcstump. Its F _c and F _c3 sequences were designed according to the ledge-in-trough technique and additionally contain the double mutation L234A and L235A.

The F _c sequence of SEQ ID NO: 396 was designed to contain RF residues at positions 200-221 (characters in bold above), instead of HY residues that would otherwise be present in this Fc region. The HY > RF mutation (i.e. H435R and Y436F in the CH3 domain as described in Jendeberg, L. et al. 1997, J. Immunological Meth., 201: 25-34) is advantageous for purification purposes because protein A binding. In the case of CODV-FabOL1 hz20G6xhz7G3 F _c -stump with the sequence of SEQ ID NO: 397 contains HY residues at positions 217-218 (symbols in bold above).

The antibody-like binding protein, called CODV-Fab-OL1a hz20G6xhz7G3, contains one polypeptide according to formula I with the amino acid sequence

DIVMTQSPDSLAVSLGERATINCESSQSLLNSGNQKNYLTWYQQKPGQPPKPLIYWAST RESGVPDRFSGSGSGTDFTLTISSLQAEDVAVYYCQNDYSYPYTFGQGTKLEIKGGSGSSGSGG DIVMTQTPLSLSVTPGQPASISCKSSQSLVHNNANTYLSWYLQKPGQSPQSLIYKVSNRFSGVP DRFSGSGSGTDFTLKISRVEAEDVGVYYCGQGTQYPFTFGSGTKVEIKGGSGSSGSGGRTVAAP SVFIFPPSDEQLKSGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSS TLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC (SEQ ID NO: 388), который содержит V _D1 с последовательностью под SEQ ID NO: 385, Li с последовательностью под SEQ ID NO: 389, V _D2 с последовательностью под SEQ ID NO: 141, L2 с the sequence under SEQ ID NO: 389 and C _L with the sequence under SEQ ID NO: 310;

one polypeptide according to formula III with the amino acid sequence

QVQLVESGGGWQPGRSLRLSCAASGFTFTKAWMHWVRQAPGKQLEWVAQIKDKSNSYA TYYADSVKGRFTISRDDSKNTLYLQMNSLRAEDTAVYYCRGVYYALSPFDYWGQGTLVTVSSEV QLVQSGAEVKKPGESLKISCKGSGYSFTDYYMKWARQMPGKGLEWMGDIIPSSGATFYNQKFKG QVTISADKSISTTYLQWSSLKASDTAMYYCARSHLLRASWFAYWGOGTMVTVSSASTKG Р SVFP LAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSWTVPSS SLGTQTYICNVNHKPSNTKVDKKVEPKSCDKTHTCPPCPAPEAAGGPSVFLFPPKPKDTLMISR TPEVTCWVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEOYNSTYRWSVLTVLHQDWLNGKEY KCKVSNKALPAPIEKTISKAKGOPREPOVYTLPPCRDEL·TKNOVSL·WCLVKGFYPSDIAVEWES NGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWOOGNVFSCSVMHEALHNHYTQKSLSLSPG (SEQ ID NO: 399), содержащий VD3 с последовательностью под SEQ ID NO: 138, причем L3 представляет собой 0 аминокислот, V _D4 с последовательностью под SEQ ID NO: 383 (italicized and underlined), with L ₄ representing 0 amino acids, C _H1 with the sequence under SEQ ID NO:

- 52 041572

313 and Fc (underlined) with the sequence under SEQ ID NO: 400;

and wherein the antibody-like binding protein termed CODV-Fab-OL1a hz20G6xhz7G3 further comprises an Fc-stump (F _c3 ) with the amino acid sequence

GSDKTHTCPPCPAPEAAGGPSVFLFPPKPKDTLMISRTPEVTCVWDVSHEDPEVKFNW

YVDGVEVHNAKTKPREEQYNSTYRWSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKG

QPREPQVCTLPPSRDELTKNQVSLSCAVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFF

LVSKLTVDKSRWQQGNVFSCSVMHEALHNRFTQKSLSLSPG (SEQ ID NO: 398), which forms a heterodimer with the Fc region of a polypeptide according to formula III.

F _c with the sequence under SEQ ID NO: 400 contains HY residues at positions 200-221 (characters in bold above), while F _c -stump with the sequence under SEQ ID NO: 398 contains RF residues at positions 217-218 (characters in bold above).

Said antibody-like binding protein is in CODV-Fab-OL format, i. e. it contains or consists of one formula I polypeptide, one formula III polypeptide and one Fcstump. Its F _c and F _c3 sequences were designed according to the ledge-in-trough technique and contain the double mutation L234A and L235A.

In one embodiment, the first immunoglobulin or the second immunoglobulin is a single anti-CD123 antibody selected from the anti-CD123 antibodies referred to as 3E3-D3, 1E1-G5, 2B8-F3, 2F8-D6, 3B10-E6, 5A5-B4, 6B10-E4 , 6C10-C4, 6D6-B8, 8B11-B7, 9B8-G6, 9D7C8, and 9F6-G3, or a humanized form thereof, or an anti-CD123 7G3 antibody described herein below, for example, an anti-CD123 antibody 3E3-D3 or 7G3 or their humanized form.

In one embodiment, the first immunoglobulin or the second immunoglobulin is a single anti-CD3 antibody selected from the anti-CD3 antibodies referred to as 20G6-F3, 4B4-D7, 4E7C9, 18F5-H10, 12D2-E5, 11D7-C3, 11H3-E5, 13H2 -C2, 13C1-F6, 18H11-F10, 1E6-C9, 10F4-C10, 10E6-G6, 18G9-H11, 11F3-B9, 12G3-E8, 5B1-G2, 16F8-A7, 11F9-F8, 3G5-E10 , 9D7-F3, 8C2-F7, 20E5-F10, 20B5-F10, 6C9-C9, 3E8-G1, 3H6-D2, and 8H2, or a humanized form thereof, such as anti-CD3 antibodies called 20G6-F3, 4B4- D7, 4E7-C9, 18F5-H10 and hz20G6, such as anti-CD3 antibodies called 20G6-F3, 4B4-D7.

Accordingly, V _D1 and V _D4 or V _D2 and V _D3 are the heavy or light chain variable domains of an anti-CD3 antibody, wherein said anti-CD3 antibody comprises:

a) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 6, a CDR2-H of the sequence of SEQ ID NO: 7, a CDR3-H of the sequence of SEQ ID NO: 8, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 10 or SEQ ID NO: 142, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

b) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 13, a CDR2-H of the sequence of SEQ ID NO: 14, a CDR3-H of the sequence of SEQ ID NO: 15, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 17 or SEQ ID NO: 184, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

c) a heavy chain variable domain comprising a CDR1-H of the sequence of SEQ ID NO: 13, a CDR2-H of the sequence of SEQ ID NO: 19, a CDR3-H of the sequence of SEQ ID NO: 20, and a light chain variable domain, containing CDR1-L with the sequence under SEQ ID NO: 22, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11; or

d) a heavy chain variable domain containing a CDR1-H with the sequence of SEQ ID NO: 24, a CDR2-H with the sequence of SEQ ID NO: 19, a CDR3-H with the sequence of SEQ ID NO: 25 and a light chain variable domain containing CDR1-L with the sequence of SEQ ID NO: 27, CDR2-L with the sequence of KVS and CDR3-L with the sequence of SEQ ID NO: 28, and where V _D1 is a heavy chain variable domain as defined above, if V _D4 is is a light chain variable domain, or V _D1 is a light chain variable domain as defined above if V _D4 is a heavy chain variable domain, or

VD2 is a heavy chain variable domain as defined above if VD3 is a light chain variable domain, or VD2 is a light chain variable domain as defined above if V _D3 is a heavy chain variable domain.

In a further embodiment, V _D1 and V _D4 or V _D2 and V _D3 are the heavy or light chain variable domains of an anti-CD3 antibody, wherein said anti-CD3 antibody is a humanized antibody and contains:

a) a heavy chain variable domain with a sequence of SEQ ID NO: 138 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 143 or a sequence that is at least 85% identical to her; or

- 53 041572

b) a heavy chain variable domain with a sequence of SEQ ID NO: 171 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with a sequence of SEQ ID NO: 158 or a sequence that is at least 85% identical to her; or

c) a heavy chain variable domain with the sequence of SEQ ID NO: 176 or a sequence that is at least 85% identical thereto, and/or a light chain variable domain with the sequence of SEQ ID NO: 164 or a sequence that is at least 85% identical to her; or where V _D1 is a heavy chain variable domain as defined above if V _D4 is a light chain variable domain, or V _D1 is a light chain variable domain as defined above if V _D4 is a heavy chain variable domain, or

V _D2 is a heavy chain variable domain as defined above if V _D3 is a light chain variable domain, or V _D2 is a light chain variable domain as defined above if V _D3 is a heavy chain variable domain.

In the specified sequence, which is at least 85% identical to the sequence under SEQ ID NO: 138, SEQ ID NO: 143, SEQ ID NO: 171, SEQ ID NO: 158, SEQ ID NO: 176 or SEQ ID NO: 164, the 6 CDR sequences are unchanged from the 6 CDRs present in the reference sequence under SEQ ID NO: 138, SEQ ID NO: 143, SEQ ID NO: 171, SEQ ID NO: 158, SEQ ID NO: 176, or SEQ ID NO: 164.

In an embodiment, the antibody-like binding protein of the present invention binds to human CD3. In another embodiment, the antibody-like binding protein of the present invention additionally binds to Macac fascicularis CD3. In particular, the antibody-like binding protein of the present invention binds to the extracellular domain of human CD3 or both human and Macac fascicularis CD3. More specifically, the antibody binds to CD3ε. More specifically, the antibody-like binding protein binds to the extracellular domain of human or human CD3ε and Macac fascicularis. An antibody-like binding protein binds to CD3ε when it is present in the form of a complex, such as a CD3ε/δ complex, or when it is present as a single protein, whether it is expressed in an isolated form or present as a soluble extracellular domain or a full-length anchored in the CD3ε membrane, present, for example, in T cells. The antibody-like binding protein of the present invention is specific for human CD3 surface protein or both human and Macac fascicularis CD3 proteins, in particular CD3ε.

The antibody-like binding protein of the present invention is characterized by a ratio of Macaca fascicularis CD3 affinity and human CD3 affinity KD(Macaca fascicularis)/KD (human) which is <10, in particular <6, <5, <4, <3, <2 , <1 or <0.5. Thus, the antibody-like binding protein of the present invention can be used in toxicology studies in monkeys because the toxicity profile observed in monkeys is consistent with the expected potential adverse effects in humans.

In addition, the antibody-like binding protein of the present invention has an affinity (KD) for human CD3 or Macac fascicularis CD3, or both, which is <50 nM, <40 nM, or <30 nM, such as <20 nM, such as affinity between 0 .1 to 30 nM, in particular, is from 0.4 to 20 nM or is from 0.4 to 15 nM.

In one embodiment, the antibody-like binding protein of the present invention has a T cell activation that is less than 20%, less than 18%, less than 16%, less than 14%, less than 12%, less than 10% in the absence of target cells.

In one embodiment, the antibody-like binding protein of the present invention has a T cell activation that is greater than 55%, greater than 60%, greater than 62%, greater than 64%, greater than 66%, greater than 68%, greater than 70% in the presence of target cells.

Low T cell activation in the context of the antibody-like binding proteins of the present invention refers to T cell activation that is less than 20%, less than 18%, less than 16%, less than 14%, less than 12%, less than 10%.

Target cells herein refer to cells that express the second antigen, in one example, target cells herein refer to cells expressing CD123, such as THP-1 cells.

High T cell activation is herein referred to as T cell activation that is greater than 50%, greater than 55%, greater than 60%, greater than 62%, greater than 64%, greater than 66%, greater than 68%, greater than 70%.

In a further embodiment, the present invention provides an antibody-like binding protein having biological and immunological specificity for at least one additional target antigen.

Therefore, in one aspect of the present invention, the antibody-like binding protein

- 54 041572 of the present invention binds additionally to at least one target antigen.

Accordingly, in one embodiment, the antibody-like binding protein of the present invention is bispecific and capable of binding two different target antigens or epitopes.

Accordingly, in one embodiment, the first immunoglobulin is an immunoglobulin directed against at least one additional target, if the second immunoglobulin is a single anti-CD123 antibody as defined in the Anti-CD123 section above, or the second immunoglobulin is an immunoglobulin directed against at least one additional target if the first immunoglobulin is a single anti-CD123 antibody as defined in the Anti-CD123 section above.

In one further embodiment, the first immunoglobulin is an immunoglobulin directed against at least one additional target, if the second immunoglobulin is a single anti-CD3 antibody as defined in the Anti-CD3 antibodies section above, or the second immunoglobulin is an immunoglobulin directed against at least at least one additional target if the first immunoglobulin is a single anti-CD3 antibody as defined in the Anti-CD3 Antibodies section above.

The antibody-like binding protein of the present invention has a T-cell recruiting effect. This T cell recruiting effect induces cytotoxicity to the target cell. In one embodiment, the target cell is a cell expressing CD123, such as a cancer cell expressing CD123, such as THP-1 or TF-1.

Accordingly, in one embodiment, the antibody-like binding protein of the present invention is capable of recruiting primary T cells and leading to lysis of target cells in vitro, wherein (EC ₅₀ ) is <40 pM, <35 pM, e.g., <30 pM.

Cytotoxicity is herein referred to as the quality of a compound, such as an antibody-like binding protein or an anti-CD123 antibody of the present invention, to be toxic to cells. Cytotoxicity can be induced by various mechanisms of action and thus can be classified into cell-mediated cytotoxicity, apoptosis, antibody-dependent cell-mediated cytotoxicity (ADCC), or complement-dependent cytotoxicity (CDC).

Antibody-dependent cell-mediated cytotoxicity or ADCC refers to a cell-mediated immune defense mechanism in which an effector cell of the immune system actively lyses a target cell whose membrane surface antigens have been bound by specific antibodies.

Complement-dependent cytotoxicity or CDC, in the context of the present invention, refers to the lysis of a target cell in the presence of proteins of the complement system.

Cell mediated cytotoxicity refers to the cytolysis of the target cell by effector lymphocytes such as cytotoxic T lymphocytes or natural killer cells, and thus can be divided into T cell mediated cytotoxicity and NK cell cytotoxicity.

In one embodiment, cytotoxicity is herein referred to as cell-mediated cytotoxicity, such as T-cell mediated cytotoxicity.

In addition, in one embodiment, cell-mediated cytotoxicity refers to cell-mediated cytotoxicity by T cells.

Accordingly, the antibody-like binding protein of the present invention induces cell-mediated cytotoxicity in the target cell, which is mediated by T cells.

Methods for measuring cytotoxicity are known to one of skill in the art and include the use of a 51-chromium (Cr) release assay, staining of target cells for live/dead cells, including staining with propidium iodide, 7-AAD and other dyes that are known to one of skill in the art, detecting lytic molecules released by T cells, including granzyme and perforin, by flow cytometry or ELISA, detection of lactate dehydrogenase (LDH) released into the medium from damaged cells as a biomarker of cellular cytotoxicity and cytolysis, detection of CD107a transfer to the cell surface, staining with using annexin V (calcium-dependent phospholipid-binding proteins) of apoptotic target cells and, for example, detection of activated caspase-3 (CASP3). In addition, one skilled in the art will be able to distinguish between different mechanisms of cytotoxicity based on the test chosen and on the basis of the experimental setup.

In one example, cell-mediated cytotoxicity, for example, can be measured using CFSE to label target cells and 7-AAD to label dead cells, as described, for example, in Example 3.2.

In a further embodiment, the antibody-like binding protein is capable of binding to CD3 and at least one additional target antigen, such as CD123.

In one embodiment, the antibody-like binding protein is capable of inhibiting

- 55 041572 the functioning of this additional target antigen, for example CD 123.

In one aspect of the present invention, the antibody-like binding protein binds to human CD123. In another embodiment, the antibody-like binding protein further binds to Macaca fascicularis CD123. In particular, the antibody-like binding protein of the present invention binds to the extracellular domain of human CD123 or both human and Macac fascicularis CD123. More specifically, the antibody-like binding protein binds to a distal fragment of CD123, such as amino acids ranging from position 19 to 49 of human CD123 with the amino acid sequence of SEQ ID NO: 104. The antibody-like binding protein binds to CD123 whether it is expressed as an isolated form or present as a soluble extracellular domain or full-length membrane-anchored CD123 present in cells expressing CD123, such as AML cells or cells transfected with CD123. The antibody-like binding protein of the present invention is specific for cells that express human or human and Macac fascicularis CD123 proteins on their surface, such as CD123-expressing cancer cells.

Accordingly, the antibody-like binding protein of the present invention has an affinity (KD) for human CD123 or Macac fascicularis CD123, or both, which is <20 nM, <15 nM, or <10 nM, e.g. 01 to 5 nM, in particular is from 0.1 to 5 nM.

Accordingly, in one embodiment, the first immunoglobulin is a single anti-CD3 antibody selected from the anti-CD3 antibodies referred to as 20G6-F3, 4B4-D7, 4E7-C9, 18F5H10, 12D2-E5, 11D7-C3, 11H3-E5, 13H2- C2, 13C1-E6, 18H11-F10, 1E6-C9, 10F4-C10, 10E6-G6, 18G9-H11, 11F3-B9, 12G3-E8, 5B1-G2, 16F8-A7, 11F9-F8, 3G5-E10, 9D7-F3, 8C2-F7, 20E5-F10, 20B5-F10, 6C9-C9, 3E8-G1, 3H6-D2, and 8H2, or their humanized form, such as anti-CD3 antibodies called 20G6-F3, 4B4-D7, 4E7-C9, 18F5-H10, hz4B4 and hz20G6, and the second immunoglobulin is an anti-CD123 antibody selected from the anti-CD123 antibodies referred to as 3E3-D3, 1E1-G5, 2B8-F3, 2F8-D6, 3B10-E6, 5A5- B4, 6B10-E4, 6C10C4, 6D6-B8, 8B11-B7, 9B8-G6, 9D7-C8, and 9F6-G3.

Accordingly, in a further embodiment, the second immunoglobulin is a single anti-CD3 antibody selected from the anti-CD3 antibodies referred to as 20G6-F3, 4B4-D7, 4E7-C9, 18F5-H10, 12D2-E5, 11D7-C3, 11H3-E5, 13H2-C2, 13C1-F6, 18H11-F10, 1E6-C9, 10F4C10, 10E6-G6, 18G9-H11, 11F3-B9, 12G3-E8, 5B1-G2, 16F8-A7, 11F9-F8, 3G5-E10, 9D7F3, 8C2-F7, 20E5-F10, 20B5-F10, 6C9-C9, 3E8-G1, 3H6-D2, and 8H2, or their humanized form, such as anti-CD3 antibodies called 20G6-F3, 4B4-D7, 4E7- C9, 18F5-H10, hz4B4 and hz20G6, and the first immunoglobulin is a single anti-CD123 antibody selected from the anti-CD123 antibodies referred to as 3E3-D3, 1E1-G5, 2B8-F3, 2F8-D6, 3B10-E6, 5A5-B4 , 6B10-E4, 6C10C4, 6D6-B8, 8B11-B7, 9B8-G6, 9D7-C8, and 9F6-G3.

Accordingly, in one embodiment, V _D1 and VD4 or V _D2 and V _D3 comprise a heavy chain variable domain and a light chain variable domain, each defined by either three CDR sequences or heavy and light chain variable domain sequences from one of 13 antibodies to CD123, called 3E3-D3, 1E1-G5, 2B8-F3, 2F8-D6, 3B10-E6, 5A5-B4, 6B10-E4, 6C10-C4, 6D6-B8, 8B11-B7, 9B8-G6, 9D7- C8, and 9F6-G3 as defined above, wherein both V _D1 and V _D4 contain three CDR sequences from the heavy and light chain variable domain sequences of a single anti-CD123 antibody as defined above, if both V _D2 and V _D3 contain three CDR sequences from heavy and light chain variable domain sequences of one of the anti-CD3 antibodies as defined above, or where both VD2 and VD3 contain three CDRs from the heavy and light chain variable domain sequences of a single anti-CD123 antibody as defined above, if VD1 and VD4 contain the CDR sequences and 3 heavy and light chain variable domain sequences from one of the anti-CD3 antibodies defined above.

Accordingly, in one embodiment, VD1 and VD4 are the heavy or light chain variable domain of an anti-CD3 antibody, wherein said anti-CD3 antibody comprises a heavy chain variable domain comprising a CDR1-H with the sequence of SEQ ID NO: 6, a CDR2-H with the sequence under SEQ ID NO: 7, CDR3-H with the sequence under SEQ ID NO: 8, and a light chain variable domain containing CDR1-L with the sequence under SEQ ID NO: 10 or SEQ ID NO: 142, CDR2-L with the sequence KVS and CDR3-L with the sequence of SEQ ID NO: 11, and V _D2 and V _D3 are the heavy or light chain variable domain of an anti-CD123 antibody selected from the group consisting of antibodies to CD123, 3E3-D3, 1E1-G5, 2B8 -F3, 2F8-D6, 3B10-E6, 5A5B4, 6B10-E4, 6C10-C4, 6D6-B8, 8B11-B7, 9B8-G6, 9D7-C8, 9F6-G3, as described above under Antibodies to CD123, while both V _D1 and VD2 are light chain variable domains, if both VD3 and VD4 are heavy chain variable domains , or both VD1 and VD2 are heavy chain variable domains if both VD3 and VD4 are light chain variable domains.

Accordingly, in a further embodiment, V _D2 and V _D3 are a variation

- 56 041572 a heavy or light chain white domain of an anti-CD3 antibody, wherein said anti-CD3 antibody comprises a heavy chain variable domain comprising CDR1-H with the sequence of SEQ ID NO: 6, CDR2-H with the sequence of SEQ ID NO: 7, CDR3 -H with the sequence of SEQ ID NO: 8, and a light chain variable domain containing CDR1-L with the sequence of SEQ ID NO: 10 or SEQ ID NO: 142, CDR2-L with the sequence of KVS and CDR3-L with the sequence of SEQ ID NO: 11, and VDi and V _D4 are the heavy or light chain variable domain of an anti-CD123 antibody selected from the group consisting of anti-CD123, 3E3-D3, 1E1-G5, 2B8-F3, 2F8-D6, 3B10E6, 5A5-B4, 6B10-E4, 6C10-C4, 6D6-B8, 8B11-B7, 9B8-G6, 9D7-C8, 9F6-G3, as described above under Antibodies to CD123, while both V _D1 and V _D2 represent are light chain variable domains if both V _D3 and V _D4 are heavy chain variable domains, or both V _D1 and V _D2 are heavy chain variable domains chains if both V _D3 and V _D4 are light chain variable domains.

Accordingly, in a further embodiment, VD1 and VD4 are the heavy or light chain variable domain of an anti-CD3 antibody, wherein said anti-CD3 antibody comprises a heavy chain variable domain comprising a CDR1-H with the sequence of SEQ ID NO: 13, a CDR2-H with the sequence under SEQ ID NO: 14, CDR3-H with the sequence under SEQ ID NO: 15, and a light chain variable domain containing CDR1-L with the sequence under SEQ ID NO: 17 or SEQ ID NO: 184, CDR2-L with the sequence KVS and CDR3-L with the sequence of SEQ ID NO: 11, and V _D2 and V _D3 are the heavy or light chain variable domain of an anti-CD123 antibody selected from the group consisting of antibodies to CD123, 3E3-D3, 1E1-G5, 2B8 -F3, 2F8-D6, 3B10-E6, 5A5-B4, 6B10-E4, 6C10-C4, 6D6-B8, 8B11-B7, 9B8-G6, '^7^8, 9F6-G3, as described above in section Anti-CD123 antibodies where both _VD1 and _VD2 are light chain variable domains if both VD3 and VD4 are t variable domains heavy chains, or both VD1 and VD2 are heavy chain variable domains if both VD3 and VD4 are light chain variable domains.

Accordingly, in a further embodiment, V _D1 and V _D4 are the heavy or light chain variable domain of an anti-CD3 antibody, wherein said anti-CD3 antibody comprises a heavy chain variable domain comprising a CDR1-H with the sequence of SEQ ID NO: 13, CDR2-H with the sequence under SEQ ID NO: 14, CDR3-H with the sequence under SEQ ID NO: 15, and a light chain variable domain containing CDR1-L with the sequence under SEQ ID NO: 17 or SEQ ID NO: 184, CDR2-L with the sequence of KVS and CDR3-L with the sequence under SEQ ID NO: 11, and VD2 and VD3 represent the variable domain of the heavy or light chain of an anti-CD123 antibody selected from the group consisting of antibodies to CD123, 3E3-D3, 1E1-G5, 2B8 -F3, 2F8-D6, 3B10-E6, 5A5-B4, 6B10-E4, 6C10-C4, 6D6-B8, 8B11-B7, 9B8-G6, 9D7-C8, 9F6-G3, as described above under Antibodies to CD123, where both V _D1 and V _D2 are light chain variable domains, if both VD3 and VD4 are variable domains heavy chains, or both VD1 and VD2 are heavy chain variable domains if both VD3 and VD4 are light chain variable domains.

In a further embodiment, V _D1 and V _D4 are the heavy or light chain variable domain of a humanized anti-CD3 antibody, wherein said anti-CD3 antibody comprises a heavy chain variable domain with the sequence of SEQ ID NO: 138 and/or a light chain variable domain with the sequence of SEQ ID NO: 143, and VD ₂ and VD ₃ are the heavy or light chain variable domain of an anti-CD123 antibody selected from the group consisting of anti-CD123, 3E3-D3, 1E1-G5, 2B8-F3, 2F8-D6, 3B10-E6, 5A5-B4, 6B10-E4, 6C10-C4, 6D6-B8, 8B11-B7, 9B8-G6, 9D7-C8, 9F6-G3, as described above under Antibodies to CD123, with both VD1 and VD2 are light chain variable domains if both VD3 and VD4 are heavy chain variable domains, or both VD1 and VD2 are heavy chain variable domains if both VD3 and VD4 are light chain variable domains.

In a further embodiment, V _D2 and V _D3 are the heavy or light chain variable domain of a humanized anti-CD3 antibody, wherein said anti-CD3 antibody comprises a heavy chain variable domain with the sequence of SEQ ID NO: 138 and/or a light chain variable domain with the sequence of SEQ ID NO: 143, and V _D1 and V _D4 are the heavy or light chain variable domain of an anti-CD123 antibody selected from the group consisting of anti-CD123, 3E3-D3, 1E1-G5, 2B8-F3, 2F8-D6, 3B10-E6, 5A5-B4, 6B10-E4, 6C10-C4, 6D6-B8, 8B11-B7, 9B8-G6, 9D7-C8, 9F6-G3, as described above under Antibodies to CD123, with both VD1 and VD2 are light chain variable domains if both VD3 and VD4 are heavy chain variable domains, or both VD1 and VD2 are heavy chain variable domains if both VD3 and VD4 are light chain variable domains.

In a further embodiment, VD1 and VD4 are the heavy or light chain variable domain of a humanized anti-CD3 antibody, wherein said anti-CD3 antibody comprises a heavy chain variable domain of the sequence of SEQ ID NO: 171 and/or a light chain variable domain of the sequence of SEQ ID NO: 158, and V _D2 and V _D3 are variable

- 57 041572 heavy or light chain domain of an anti-CD123 antibody selected from the group consisting of anti-CD123 antibodies

CD123, 3E3-D3, 1E1-G5, 2B8-F3, 2F8-D6, 3B10-E6, 5A5-B4, 6B10-E4, 6C10-C4, 6D6-B8,

8B11-B7, 9B8-G6, 9D7-C8, 9F6-G3 as described above under Antibodies to CD123, where both V _D1 and V _D2 are light chain variable domains if both V _D3 and V _D4 are variable heavy chain domains, or both V _D1 and V _D2 are heavy chain variable domains if both V _D3 and V _D4 are light chain variable domains.

In a further embodiment, V _D2 and V _D3 are the heavy or light chain variable domain of a humanized anti-CD3 antibody, wherein said anti-CD3 antibody comprises a heavy chain variable domain with the sequence of SEQ ID NO: 171 and/or a light chain variable domain with the sequence of SEQ ID NO: 158, and V _D1 and V _D4 are the heavy or light chain variable domain of an anti-CD123 antibody selected from the group consisting of anti-CD123, 3E3-D3, 1E1-G5, 2B8-F3, 2F8-D6, 3B10-E6, 5A5-B4, 6B10-E4, 6C10-C4, 6D6-B8, 8B11-B7, 9B8-G6, 9D7-C8, 9F6-G3, as described above under Antibodies to CD123, both V _D1 and V _D2 are light chain variable domains if both V _D3 and V _D4 are heavy chain variable domains, or both VD1 and VD2 are heavy chain variable domains if both V _D3 and V _D4 are light chain variable domains.

In a further embodiment, V _D1 and V _D4 are the heavy or light chain variable domain of a humanized anti-CD3 antibody, wherein said anti-CD3 antibody comprises a heavy chain variable domain with the sequence of SEQ ID NO: 176 and/or a light chain variable domain with the sequence of SEQ ID NO: 164, and V _D2 and V _D3 are the heavy or light chain variable domain of an anti-CD123 antibody selected from the group consisting of anti-CD123, 3E3-D3, 1E1-G5, 2B8-F3, 2F8-D6, 3B10-E6, 5A5-B4, 6B10-E4, 6C10-C4, 6D6-B8, 8B11-B7, 9B8-G6, 9D7-C8, 9F6-G3, as described above under Antibodies to CD123, both V _D1 and V _D2 are light chain variable domains if both V _D3 and V _D4 are heavy chain variable domains, or both V _D1 and V _D2 are heavy chain variable domains if both V _D3 and V _D4 are light chain variable domains chains.

In a further embodiment, V _D2 and V _D3 are the heavy or light chain variable domain of a humanized anti-CD3 antibody, wherein said anti-CD3 antibody comprises a heavy chain variable domain with the sequence of SEQ ID NO: 176 and/or a light chain variable domain with the sequence of SEQ ID NO: 164, and V _D1 and V _D4 are the heavy or light chain variable domain of an anti-CD123 antibody selected from the group consisting of anti-CD123, 3E3-D3, 1E1-G5, 2B8-F3, 2F8-D6, 3B10-E6, 5A5-B4, 6B10-E4, 6C10-C4, 6D6-B8, 8B11-B7, 9B8-G6, 9D7-C8, 9F6-G3, as described above under Antibodies to CD123, both V _D1 and V _D2 are light chain variable domains if both V _D3 and V _D4 are heavy chain variable domains, or both V _D1 and V _D2 are heavy chain variable domains if both V _D3 and V _D4 are light chain variable domains chains.

According to a further aspect of the present invention, the first or second immunoglobulin is an anti-CD123, 7G3 antibody. Therefore, in one embodiment, V _D1 and V _D4 or VD2 and VD3 comprise a heavy chain variable domain and a light chain variable domain, as determined by the CDR sequences from the heavy and light chain variable domain sequences of the 7G3 antibody, as defined herein below. In one embodiment, V _D1 and V _D4 or V _D2 and V _D3 comprise the heavy chain variable domain and the light chain variable domain of the 7G3 antibody described in patent application WO2013/173820, which is incorporated herein by reference.

Accordingly, the anti-CD123 antibody, referred to herein as 7G3, contains a heavy chain variable domain consisting of the sequence

EVQLQQSGPELVKPGASVKMSCKASGYTFTDYYMKWVKQSHGKSLEWIGDIIPSNGATFYNQKF

KGKATLTVDRSSSTAYMHLNSLTSEDSAVYYCTRSHLLRASWFAYWGQGTLVTVSA (SEQ ID NO: 312, with CDRs shown in bold characters) containing CDR1-H with the sequence of SEQ ID NO: 375, CDR2-H with the sequence of SEQ ID NO: 376, and CDR3-H with the sequence of SEQ ID NO: 377, and a light chain variable domain consisting of the sequence

DFVMTQSPSSLTVTAGEKVTMSCKSSQSLLNSGNQKNYLTWYLQKPGQPPKLLIYWASTRESGV

PDRFTGSGSGTDFTLTISSVQAEDLAVYYCQNDYSYPYTFGGGTKLEIK (SEQ ID NO: 308, with CDRs shown in bold characters) containing CDR1-L with the sequence of SEQ ID NO: 378, CDR2-L with the sequence of WAS, and CDR3-L with the sequence of SEQ ID NO: 379.

In a further aspect of the present invention, the 7G3 antibody may also be a humanized antibody or fragment of a humanized antibody. Accordingly, in one embodiment, the 7G3 antibody of the present invention is a humanized anti-body containing a heavy chain variable domain consisting of the sequence

EVQLVQSGAEVKKPGESLKISCKGSGYSFTDYYMKWARQMPGKGLEWMGDIIPSNGATFYNQKF

KGQVTISADKSISTTYLQWSSLKASDTAMYYCARSHLLRASWFAYWGQGTMVTVSS (SEQ ID NO: 380, with CDRs shown in bold) containing CDR1-H with the sequence of SEQ ID NO: 381, CDR2-H with the sequence of SEQ ID NO: 377, and CDR3-H with the sequence of SEQ ID NO: 382, or a heavy chain variable domain consisting of the sequence

EVQLVQSGAEVKKPGESLKISCKGSGYSFTDYYMKWARQMPGKGLEWMGDIIPSSGATFYNQKF

KGQVTISADKSISTTYLQWSSLKASDTAMYYCARSHLLRASWFAYWGQGTMVTVSS (SEQ ID NO: 383, with CDRs shown in bold characters) containing CDR1-H with the sequence of SEQ ID NO: 381, CDR2-H with the sequence of SEQ ID NO: 384, and CDR3-H with the sequence of SEQ ID NO: 382, or a light chain variable domain consisting of the sequence

DIVMTQSPDSLAVSLGERATINCESSQSLLNSGNQKNYLTWYQQKPGQPPKPLIYWASTRESGV

PDRFSGSGSGTDFTLTISSLQAEDVAVYYCQNDYSYPYTFGQGTKLEIK (SEQ ID NO: 385, with CDRs shown in bold characters) containing CDR1-L with the sequence of SEQ ID NO: 378, CDR2-L with the sequence of WAS, and CDR3-L with the sequence of SEQ ID NO: 379.

In one embodiment, an antibody-like binding protein that specifically binds to human CD3ε and human CD123 comprises:

a) the light chain variable domain of the first immunoglobulin (V _D1 ) consisting of the amino acid sequence under SEQ ID NO: 9 or a sequence that is at least 85% identical to it, the light chain variable domain of the second immunoglobulin (V _D2 ) consisting of the amino acid sequence under SEQ ID NO: 308 or a sequence that is at least 85% identical to it, the variable domain of the heavy chain of the second immunoglobulin (V _D3 ) consisting of the amino acid sequence under SEQ ID NO: 312 or a sequence that is at least 85 % is identical to it, and the first immunoglobulin heavy chain variable domain (V _D4 ) consisting of the amino acid sequence under SEQ ID NO: 5 or a sequence that is at least 85% identical to it, or

b) a first immunoglobulin light chain variable domain (V _D1 ) consisting of the amino acid sequence of SEQ ID NO: 21 or a sequence that is at least 85% identical thereto, a second immunoglobulin light chain variable domain (V _D2 ) consisting of the amino acid sequence under SEQ ID NO: 308 or a sequence that is at least 85% identical to it, the variable domain of the heavy chain of the second immunoglobulin (V _D3 ) consisting of the amino acid sequence under SEQ ID NO: 312 or a sequence that is at least 85 % is identical to it, and the variable domain of the first immunoglobulin heavy chain (V _D4 ) consisting of the amino acid sequence under SEQ ID NO: 18 or a sequence that is at least 85% identical to it, or

c) a first immunoglobulin light chain variable domain (V _D1 ) consisting of the amino acid sequence of SEQ ID NO: 16 or a sequence that is at least 85% identical thereto, a second immunoglobulin light chain variable domain (V _D2 ) consisting of the amino acid sequence under SEQ ID NO: 308 or a sequence that is at least 85% identical to it, the variable domain of the heavy chain of the second immunoglobulin (V _D3 ) consisting of the amino acid sequence under SEQ ID NO: 312 or a sequence that is at least 85 % is identical to it, and the variable domain of the first immunoglobulin heavy chain (V _D4 ) consisting of the amino acid sequence under SEQ ID NO: 12 or a sequence that is at least 85% identical to it, or

d) a first immunoglobulin light chain variable domain (V _D1 ) consisting of the amino acid sequence of SEQ ID NO: 26 or a sequence that is at least 85% identical thereto, a second immunoglobulin light chain variable domain (VD2) consisting of the amino acid sequence under SEQ ID NO: 308 or a sequence that is at least 85% identical to it, the variable domain of the heavy chain of the second immunoglobulin (V _D3 ), consisting of the amino acid sequence under SEQ ID NO: 312 or a sequence that is at least 85% identical thereto, and a first immunoglobulin heavy chain variable domain (V _D4 ) consisting of the amino acid sequence of SEQ ID NO: 23 or a sequence that is at least 85% identical thereto, or

e) a first immunoglobulin light chain variable domain (V _D1 ) consisting of the amino acid sequence of SEQ ID NO: 308 or a sequence that is at least 85% identical thereto, a second immunoglobulin light chain variable domain (VD2) consisting of the amino acid sequence under SEQ ID NO: 143 or a sequence that is at least 85% identical to it, the variable domain of the heavy chain of the second immunoglobulin (V _D3 ), consisting of

- 59 041572 amino acid sequence under SEQ ID NO: 138 or a sequence that is at least 85% identical to it, and the variable domain of the heavy chain of the first immunoglobulin (V _D4 ) consisting of the amino acid sequence under SEQ ID NO: 312 or a sequence that at least 85% identical to it, or

f) a first immunoglobulin light chain variable domain (V _D1 ) consisting of the amino acid sequence of SEQ ID NO: 158 or a sequence that is at least 85% identical thereto, a second immunoglobulin light chain variable domain (V _D2 ) consisting of the amino acid sequence under SEQ ID NO: 308 or a sequence that is at least 85% identical to it, the variable domain of the heavy chain of the second immunoglobulin (V _D3 ) consisting of the amino acid sequence under SEQ ID NO: 312 or a sequence that is at least 85 % is identical to it, and the variable domain of the heavy chain of the first immunoglobulin (V _D4 ), consisting of the amino acid sequence under SEQ ID NO: 171 or a sequence that is at least 85% identical to it,

g) a first immunoglobulin light chain variable domain (V _D1 ) consisting of the amino acid sequence of SEQ ID NO: 230 or a sequence that is at least 85% identical thereto, a second immunoglobulin light chain variable domain (V _D2 ) consisting of the amino acid sequence under SEQ ID NO: 158 or a sequence that is at least 85% identical to it, the variable domain of the heavy chain of the second immunoglobulin (V _D3 ) consisting of the amino acid sequence under SEQ ID NO: 171 or a sequence that is at least 85 % is identical to it, and the variable domain of the heavy chain of the first immunoglobulin (V _D4 ), consisting of the amino acid sequence under SEQ ID NO: 226 or a sequence that is at least 85% identical to it,

h) a first immunoglobulin light chain variable domain (V _D1 ) consisting of the amino acid sequence of SEQ ID NO: 385 or a sequence that is at least 85% identical thereto, a second immunoglobulin light chain variable domain (VD2) consisting of the amino acid sequence under SEQ ID NO: 141 or a sequence that is at least 85% identical to it, the variable domain of the heavy chain of the second immunoglobulin (V _D3 ), consisting of the amino acid sequence under SEQ ID NO: 138 or a sequence that is at least 85% identical to it, and the variable domain of the heavy chain of the first immunoglobulin (VD4), consisting of the amino acid sequence under SEQ ID NO: 383 or a sequence that is at least 85% identical to it.

In the above sequence that is at least 85% identical to a reference sequence (e.g., a sequence that is at least 85% identical to the sequence under SEQ ID NO: 383 or SEQ ID NO: 385), the 6 CDR sequences are unchanged from with 6 CDRs present in the reference sequence.

In one embodiment, the antibody-like binding protein according to any of definitions a) - g) further comprises a linker L1 with the sequence under SEQ ID NO: 307, L2 with the sequence under SEQ ID NO: 309, L ₃ with the amino acid sequence 'S', L ₄ with the amino acid sequence 'RT' and C _H1 with the sequence under SEQ ID NO: 313.

In one embodiment, the antibody-like binding protein according to any of the definitions of α)-g) further comprises F _c2 with the sequence of SEQ ID NO: 327.

In one embodiment, the antibody-like binding protein according to definition h) further comprises a linker L ₁ with the sequence under SEQ ID NO: 389, L2 with the sequence under SEQ ID NO: 389, L ₃ and L ₄ consisting of 0 amino acids, and CH1 with the sequence under SEQ ID NO: 313.

In one embodiment, the antibody-like binding protein as defined in h) further comprises an Fc2 with the sequence of SEQ ID NO: 392.

In one embodiment, L ₅ from an antibody-like binding protein according to any of definitions a)-h) contains 0 amino acids.

In an additional embodiment, the antibody-like binding protein according to any of the definitions a)-g) further comprises a linker L1 with the sequence under SEQ ID NO: 307, L2 with the sequence under SEQ ID NO: 309, L ₃ with the amino acid sequence S, L ₄ with the amino acid the sequence RT, C _H1 with the sequence under SEQ ID NO: 329 and F _c with the sequence under SEQ ID NO: 330.

In an additional embodiment, the antibody-like binding protein according to definition h) further comprises a linker L1 with the sequence under SEQ ID NO: 389, L2 with the sequence under SEQ ID NO: 389, L ₃ and L ₄ consisting of O amino acids, C _H i with the sequence under SEQ ID NO: 313 and F _c with the sequence under SEQ ID NO: 394.

In an additional embodiment, the antibody-like binding protein according to definition h) further comprises an Fc-stump with the sequence of SEQ ID NO: 397 or SEQ ID NO: 398, or a sequence that is at least 85% identical to the sequence of SEQ ID NO: 397 or SEQ ID NO: 398.

- 60 041572

In an embodiment, the antibody-like binding protein that specifically binds to human CD3ε and human CD123 contains or actually consists of:

a) one polypeptide of formula [I], consisting of the sequence under SEQ ID NO: 388 (V _D1 with the sequence under SEQ ID NO: 385, L1 with the sequence under SEQ ID NO: 389, V _D2 with the sequence under SEQ ID NO: 141 , L2 with the sequence under SEQ ID NO: 389 and C _L with the sequence under SEQ ID NO: 310), or a sequence that is at least 85% identical to the sequence under SEQ ID NO: 388, in which 3 CDRs from the sequence under SEQ ID NO: 378, WAS sequences and sequences under SEQ ID NO: 379 from hz7G3 light chain variable domain (VD1 with sequence under SEQ ID NO: 385) and 3 CDRs with sequence under SEQ ID NO: 142, KVS sequence and sequence under SEQ ID NO: 11 from the hz20G6 light chain variable domain (V _D2 with the sequence under SEQ ID NO: 141) are unchanged; And

b) one polypeptide of formula [II], consisting of the sequence under SEQ ID NO: 390 (V _D3 with the sequence under SEQ ID NO: 138, L ₃ is 0 amino acids, VD4 with the sequence under SEQ ID NO: 383, L4 is 0 amino acids, and C _H1 with the sequence under SEQ ID NO: 313), or a sequence that is at least 85% identical to the sequence under SEQ ID NO: 390, in which 3 CDRs from the sequences under SEQ ID NO: 381, SEQ ID NO: 384 and SEQ ID NO: 382 from the hz7G3 heavy chain variable domain (V _D4 with the sequence under SEQ ID NO: 383) and 3 CDRs with the sequences under SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8 from the hz20G6 heavy chain variable domain (V _D3 with the sequence of SEQ ID NO: 138) are unchanged;

wherein the polypeptide of formula [I] and the polypeptide of formula [II] form a light chain-heavy chain crossover pair.

In an embodiment, the antibody-like binding protein that specifically binds to human CD3ε and human CD123 contains or actually consists of:

a) one polypeptide according to formula [IV] with the amino acid sequence under SEQ ID NO: 391 (V _D1 with the sequence under SEQ ID NO: 385, L1 with the sequence under SEQ ID NO: 389, V _D2 with the sequence under SEQ ID NO: 141 , L2 with the sequence under SEQ ID NO: 389, CL with the sequence under SEQ ID NO: 310, L ₅ which contains 0 amino acids, and F _c2 with the sequence under SEQ ID NO: 392), or a sequence that is at least 85% identical to the sequence under SEQ ID NO: 391, in which 3 CDRs from the sequence under SEQ ID NO: 378, the WAS sequence and the sequence under SEQ ID NO: 379 from the hz7G3 light chain variable domain (V _D1 with the sequence under SEQ ID NO: 385) and 3 CDRs with the sequence of SEQ ID NO: 142, the KVS sequence and the sequence of SEQ ID NO: 11 from the hz20G6 light chain variable domain (VD2 with the sequence of SEQ ID NO: 141) are unchanged; And

b) one polypeptide according to formula [III] with the amino acid sequence under SEQ ID NO: 393 (V _D3 with the sequence under SEQ ID NO: 138, and L ₃ is 0 amino acids, V _D4 with the sequence under SEQ ID NO: 383, and L4 is 0 amino acids, C _H1 with the sequence of SEQ ID NO: 313 and Fc with the sequence of SEQ ID NO: 394), or a sequence that is at least 85% identical to the sequence of SEQ ID NO: 393, in which 3 CDRs from the sequences of SEQ ID NO: 381, SEQ ID NO: 384 and SEQ ID NO: 382 from the hz7G3 heavy chain variable domain (VD ₄ with the sequence of SEQ ID NO: 383) and 3 CDRs of the sequences of SEQ ID NO: 6 , SEQ ID NO: 7, SEQ ID NO: 8 of hz20G6 heavy chain variable domain (V _D3 with the sequence under SEQ ID NO: 138) are unchanged;

wherein the polypeptide of formula [IV] and the polypeptide of formula [III] form a light chain-heavy chain crossover pair.

In said antibody-like binding protein, the polypeptide chains represented by formulas [III] and [IV] form a dimer with their respective F _c2 and F _c regions.

In an embodiment, the antibody-like binding protein that specifically binds to human CD3ε and human CD123 contains or actually consists of:

a) one polypeptide according to formula [I], consisting of the amino acid sequence under SEQ ID NO: 388 (V _D1 with the sequence under SEQ ID NO: 385, L1 with the sequence under SEQ ID NO: 389, VD 2 with the sequence under SEQ ID NO: 389, VD ₂ with the sequence under SEQ ID NO: : 141, L2 with the sequence of SEQ ID NO: 389 and C _L with the sequence of SEQ ID NO: 310); or a sequence that is at least 85% identical to the sequence of SEQ ID NO: 388, in which 3 CDRs from the sequence of SEQ ID NO: 378, the WAS sequence and the sequence of SEQ ID NO: 379 from the hz7G3 light chain variable domain (V _D1 with the sequence of SEQ ID NO: 385) and 3 CDRs with the sequence of SEQ ID NO: 142, the KVS sequence and the sequence of SEQ ID NO: 11 from hz20G6 light chain variable domain (VD2 with the sequence of SEQ ID NO: 141) are unchanged; And

- 61 041572

b) one polypeptide according to the formula [III] with the amino acid sequence under SEQ ID

NO: 395 (VD ₃ with the sequence under SEQ ID NO: 138, and L ₃ represents 0 amino acids,

V _D4 with the sequence under SEQ ID NO: 383, and L ₄ is 0 amino acids, C _H1 with the sequence under SEQ ID NO: 313 and F _c with the sequence under SEQ ID NO: 396), or a sequence that is at least 85% identical to the sequence under SEQ ID NO: 395, in which 3 CDRs from the sequences under SEQ ID NO: 381, SEQ ID NO: 384 and SEQ ID NO: 382 from the hz7G3 heavy chain variable domain (V _D4 with the sequence under SEQ ID NO: : 383) and 3 CDRs with the sequences under SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8 from the hz20G6 heavy chain variable domain (V _D3 with the sequence under SEQ ID NO: 138) are unchanged; And

c) one Fc-stump (F _c3 polypeptide) with the amino acid sequence under SEQ ID NO: 397 or a sequence that is at least 85% identical to it, where the specified F _c3 -stump or a sequence that is at least 85% identical him, forms a heterodimer with the Fc region of the polypeptide according to the formula [III];

wherein the polypeptide of formula [I] and the polypeptide of formula [III] form a light chain-heavy chain crossover pair.

In an embodiment, the antibody-like binding protein that specifically binds to human CD3ε and human CD123 contains or actually consists of:

a) one polypeptide according to formula [I], consisting of the amino acid sequence under SEQ ID NO: 388 (V _D1 with the sequence under SEQ ID NO: 385, L1 with the sequence under SEQ ID NO: 389, V _D2 with the sequence under SEQ ID NO: : 141, L2 with the sequence under SEQ ID NO: 389 and C _L with the sequence under SEQ ID NO: 310) or a sequence that is at least 85% identical to the sequence under SEQ ID NO: 388, in which 3 CDRs from the sequence under SEQ ID NO: 378, WAS sequences and sequences under SEQ ID NO: 379 from hz7G3 light chain variable domain (V _D1 with sequence under SEQ ID NO: 385) and 3 CDRs with sequence under SEQ ID NO: 142, KVS sequence and sequence under SEQ ID NO: 11 from the hz20G6 light chain variable domain (V _D2 with the sequence under SEQ ID NO: 141) are unchanged;

b) a polypeptide according to formula III with the amino acid sequence under SEQ ID NO: 399 (V _D3 with the sequence under SEQ ID NO: 138, and L ₃ is 0 amino acids, V _D4 with the sequence under SEQ ID NO: 383, and L ₄ is is 0 amino acids, C _H1 with the sequence under SEQ ID NO: 313 and F _c with the sequence under SEQ ID NO: 400), or a sequence that is at least 85% identical to it, in which 3 CDRs with sequences under SEQ ID NO : 381, SEQ ID NO: 384 and SEQ ID NO: 382 from hz7G3 heavy chain variable domain (VD4 with sequence under SEQ ID NO: 383) and 3 CDRs with sequences under SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8 from hz20G6 heavy chain variable domain (VD3 with the sequence under SEQ ID NO: 138) are unchanged; And

c) one Fc-stump (Fc3 polypeptide) with the amino acid sequence under SEQ ID NO: 398 or a sequence that is at least 85% identical to it, where the specified Fc3-stump or a sequence that is at least 85% identical to it, forms a heterodimer with the Fc region of the polypeptide according to the formula [III];

wherein the polypeptide of formula [I] and the polypeptide of formula [III] form a light chain-heavy chain crossover pair.

Immunoconjugates.

In one embodiment, an anti-CD123 antibody of the present invention is conjugated or linked to a growth inhibitory agent, a cytotoxic agent, or a prodrug activating enzyme. In particular, the anti-CD123 antibodies of the present invention are indeed useful for targeting said growth inhibitor, cytotoxic agent, or prodrug to cancer cells expressing or overexpressing CD123 on their surface.

Nucleic acids, vectors and recombinant host cells.

A further object of the present invention relates to a nucleic acid sequence comprising or consisting of a sequence encoding an anti-CD3 antibody, an anti-CD123 antibody, or an antibody-like binding protein as defined above.

Typically, said nucleic acid is a DNA or RNA molecule that can be incorporated into any suitable vector such as a plasmid, cosmid, episome, artificial chromosome, phage, or viral vector.

The terms vector, cloning vector, and expression vector refer to a vehicle by which a DNA or RNA sequence (eg, a foreign gene) can be introduced into a host cell in order to transform the host and stimulate expression (eg, transcription and translation) of the introduced sequence.

Thus, a further object of the present invention relates to a vector containing the nucleic acid of the present invention.

- 62 041572

Such vectors may contain regulatory elements, such as a promoter, enhancer, terminator, and the like, that cause or direct the expression of said polypeptide upon administration to a subject. Examples of promoters and enhancers used in an animal cell expression vector include the SV40 early promoter and enhancer (Mizukami T. et al. 1987), the Moloney murine leukemia virus LTR promoter and enhancer (Kuwana Y. et al. 1987), the ( Mason J.O. et al. 1985) and enhancer (Gillies S.D. et al. 1983) immunoglobulin H chain, and the like.

Any expression vector for an animal cell can be used, as long as the gene encoding the C region of a human antibody can be inserted and expressed. Examples of suitable vectors include pAGE107 (Miyaji H. et al. 1990), pAGE103 (Mizukami T. et al. 1987), pHSG274 (Brady G. et al. 1984), pKCR (O'Hare K. et al. 1981), pSG1 beta-d2-4- (Miyαji H. et al. 1990) and the like. Other examples of plasmids include replicating plasmids containing an origin of replication or integrative plasmids such as, for example, pUC, pcDNA, pBR, and the like.

Other examples of viral vectors include adenovirus, retrovirus, herpesvirus and AAV vectors. Such recombinant viruses can be obtained using techniques known in the art, such as by transfection into packaging cells or by transient transfection with helper plasmids or viruses. Representative examples of virus packaging cells include PA317 cells, PsiCRIP cells, GPenv+ cells, 293 cells, and so on. Detailed protocols for the production of such recombinant replication-deficient viruses can be found, for example, in WO 95/14785, WO 96/22378, US 5882877, US 6013516, US 4861719, US 5278056 and WO 94/19478.

A further object of the present invention relates to a cell that has been transfected, infected or transformed with a nucleic acid and/or a vector according to the present invention.

The term transformation means the introduction of a foreign (i.e. foreign) gene, DNA or RNA sequence into a host cell in such a way that the host cell expresses the introduced gene or sequence to obtain the desired substance, usually a protein or enzyme encoded by the introduced gene or sequence. . A host cell that has received and expresses the introduced DNA or RNA becomes transformed.

The nucleic acids of the present invention can be used to produce a recombinant antibody of the present invention in a suitable expression system. The term expression system means a host cell and a compatible vector under conditions suitable, for example, for the expression of a protein encoded by foreign DNA that is in the vector and introduced into the host cell.

Standard expression systems include E. coli host cells and plasmid vectors, insect host cells and baculovirus vectors, and mammalian host cells and expression vectors. Other examples of host cells include, without limitation, prokaryotic cells (such as bacteria) and eukaryotic cells (such as yeast cells, mammalian cells, insect cells, plant cells, etc.). Specific examples include E. coli, Kluyveromyces or Saccharomyces yeast, mammalian cell lines (e.g., Vero cells, CHO cells, 3T3 cells, COS cells, etc.), and primary or stable mammalian cell cultures (e.g., derived from lymphoblasts). , fibroblasts, embryonic cells, epithelial cells, nerve cells, adipocytes, etc.). Examples also include mouse SP2/0-Ag14 cell (ATCC CRL1581), mouse P3X63-Ag8.653 cell (ATcC CRL1580), CHO cell in which the dihydrofolate reductase gene (hereinafter referred to as the DHFR gene) is defective (Urlaub G et al; 1980) , a YB2/3HL.P2.G11.16Ag.20 rat cell (ATCC CRL1662, hereinafter referred to as a YB2/0 cell), and the like. The YB2/0 cell is preferred because the ADCC activity of chimeric or humanized antibodies is enhanced when expressed in this cell.

In particular, for the expression of a humanized antibody or antibody-like binding protein, the expression vector may be either of the type in which the gene encoding the antibody heavy chain and the gene encoding the antibody light chain are located on separate vectors, or of the type in which both genes are located. on one vector (tandem type). Due to the simpler construction of the expression vector for the humanized antibody and the antibody-like binding protein, the easier introduction into animal cells, and the balance between expression levels of the H- and L-chains of the antibody in animal cells, a tandem type of expression vector for the humanized antibody is preferred (Shitara K. et al J Immunol Methods 1994 Jan 3;167(1-2):271-8). Examples of a tandem type expression vector for a humanized antibody include pKANTEX93 (WO 97/10354), pEE18, and the like.

The present invention also relates to a method for producing a recombinant host cell expressing an anti-CD3 antibody, an anti-CD123 antibody, or an antibody-like binding protein of the present invention, said method comprising the steps consisting of: (i) in vitro or ex vivo administering the recombinant nucleic acid, or the vector described above into a competent host cell, (ii) in vitro or ex vivo cultivation of the resulting recombinant host cell, and (iii) optionally selecting cells that express and/or secrete said antibody.

- 63 041572

Such recombinant host cells can be used to produce an anti-CD3 antibody, at least one anti-CD123 antibody, or at least one antibody-like binding protein of the present invention.

Methods for producing an antibody-like binding protein of the present invention.

In one embodiment, the present invention provides a method for producing an antibody-like binding protein containing two polypeptide chains that form two antigen-binding sites, where the first polypeptide has the structure represented by formula [I]:

VD1 ^-L 1 ^-V D2 ^- L2 ^-C L SC, and the second polypeptide has the structure represented by the formula [II]

VD3 ^-L 3 ^-V D4 ^-L 4 ^- CH1 And, where V _D1 represents the variable domain of the heavy or light chain of the first immunoglobulin;

V _D2 is the heavy or light chain variable domain of the second immunoglobulin;

V _D3 is the heavy or light chain variable domain of said second immunoglobulin;

V _D4 is the heavy or light chain variable domain of said first immunoglobulin;

C _L is an immunoglobulin light chain constant domain;

C _Hi is the C _Hi constant domain of an immunoglobulin heavy chain;

L1, L ₂ , L ₃ and L ₄ are amino acid linkers;

and wherein the first and second polypeptides form a light chain-heavy chain crossover pair, wherein both V _D1 and V _D2 are either light chain variable domains or heavy chain variable domains, and both V _D3 and V _D4 are variable domains heavy chains if V _D1 and V _D2 are light chain variable domains, or both V _D3 and V _D4 are light chain variable domains if V _D1 and V _D2 are heavy chain variable domains.

In a further embodiment, the present invention provides a method for producing an antibody-like binding protein comprising four polypeptide chains that form four antigen-binding sites, wherein the two polypeptide chains have the structure represented by formula [I]:

^V D1 ^-L 1 ^-V D2 ^-L 2 ^-C L ^[ 1], and two polypeptide chains have the structure represented by formula (III)

Vd3-L3-Vd4-L4-Ch1-Fc [III], where V _D1 represents the variable domain of the heavy or light chain of the first immunoglobulin;

V _D2 is the heavy or light chain variable domain of the second immunoglobulin;

V _D3 is the heavy or light chain variable domain of said second immunoglobulin;

V _D4 is the heavy or light chain variable domain of said first immunoglobulin;

C _l is an immunoglobulin light chain constant domain;

C _h1 is the C _h1 constant domain of an immunoglobulin heavy chain;

F _c represents the hinge region of the immunoglobulin and constant domains CH ₂ , CH ₃ of the immunoglobulin heavy chain from immunoglobulin;

L ₁ , L ₂ , L ₃ and L ₄ are amino acid linkers;

and wherein the formula I polypeptides and the formula III polypeptides form a light chain-heavy chain crossover pair, wherein both V _D1 and V _D2 are either light chain variable domains or heavy chain variable domains, and both V _D3 and V _D4 are are heavy chain variable domains if V _D1 and V _D2 are light chain variable domains, or both V _D3 and V _D4 are light chain variable domains if V _D1 and V _D2 are heavy chain variable domains.

In a further embodiment, the present invention provides a method for producing an antibody-like binding protein comprising four polypeptide chains that form four antigen-binding sites, wherein the two polypeptide chains have the structure represented by the formula [IV] ^V D1 ^-L 1 ^-V D2 ^-L 2 ^-C L ^-L 5 ^-F c2 ^[IV] , and two polypeptide chains have the structure represented by the formula [III]

Vd3-L3-Vd4-L4-Ch1-Fc [III], where V _D1 represents the variable domain of the heavy or light chain of the first immunoglobulin;

V _D2 is the heavy or light chain variable domain of the second immunoglobulin;

- 64 041572

V _D3 is the heavy or light chain variable domain of said second immunoglobulin;

V _D4 is the heavy or light chain variable domain of said first immunoglobulin;

C _L is an immunoglobulin light chain constant domain;

C _H1 is the C _H1 constant domain of an immunoglobulin heavy chain;

F _c represents the immunoglobulin hinge region and the CH2, _CH3 constant domains of the immunoglobulin heavy chain from immunoglobulin;

F _c2 represents the immunoglobulin hinge region and the CH2, _CH3 constant domains of the immunoglobulin heavy chain from immunoglobulin;

L1 _; L2, L ₃ , L ₄ and L ₅ are amino acid linkers;

and wherein the formula I polypeptides and the formula III polypeptides form a light chain-heavy chain crossover pair, wherein both V _D1 and V _D2 are either light chain variable domains or heavy chain variable domains, and both V _D3 and V _D4 are are heavy chain variable domains if V _D1 and V _D2 are light chain variable domains, or both V _D3 and V _D4 are light chain variable domains if V _D1 and V _D2 are heavy chain variable domains.

In one embodiment of the present invention, the first immunoglobulin or the second immunoglobulin is a single anti-CD3 antibody as defined in the Anti-CD3 Antibodies section above.

In another embodiment of the present invention, the first immunoglobulin or the second immunoglobulin is a single anti-CD123 antibody as defined in the Anti-CD123 section above.

An anti-CD3 antibody, an anti-CD123 antibody, and/or antibody-like binding proteins of the present invention can be prepared using any technique known in the art, such as, without limitation, any chemical, biological, genetic, or enzymatic technique, either alone or in combination.

Knowing the amino acid sequence of the desired sequence, one skilled in the art can easily prepare said antibodies or immunoglobulin chains using standard techniques for producing polypeptides. For example, they can be synthesized using a well known solid phase method, in particular using a commercially available peptide synthesis device (such as those manufactured by Applied Biosystems, Foster City, CA) and following the manufacturer's instructions. Alternatively, the antibodies, immunoglobulin chains, and antibody-like binding proteins of the present invention can be synthesized using recombinant DNA techniques well known in the art. For example, these fragments can be obtained as DNA expression products after introducing DNA sequences encoding the desired (poly)peptide into expression vectors and introducing such vectors into suitable eukaryotic or prokaryotic hosts that will express the desired polypeptide, from which they can later be isolated with using well known techniques.

In particular, the present invention further relates to a method for producing anti-CD3 antibodies, anti-CD123 antibodies and/or antibody-like binding proteins of the present invention, said method comprising the steps consisting of: (i) culturing a transformed host cell according to the present invention; (ii) allowing expression of said antibody or polypeptide; and (iii) recovering the expressed antibody or polypeptide.

Anti-CD3 antibodies, anti-CD123 antibodies and/or antibody-like binding proteins of the present invention are suitably separated from the culture medium by conventional immunoglobulin purification procedures such as, for example, protein A-sepharose, hydroxyapatite chromatography, gel electrophoresis, dialysis or affinity chromatography.

In a specific embodiment, a humanized chimeric anti-CD3 antibody and/or an anti-CD123 antibody of the present invention can be obtained by obtaining nucleic sequences encoding VL and VH domains humanized as previously described, constructing an expression vector for the human chimeric antibody by inserting them into the vector expression for an animal cell that carries the genes encoding the human CH antibody and the human CL antibody, and causing the expression of the coding sequence by introducing the expression vector into the animal cell. Similarly, a humanized antibody-like binding protein can be prepared by using instead of the heavy or light chain variable domain of the first humanized immunoglobulin (VD1), instead of the heavy or light chain variable domain of the second humanized immunoglobulin (VD2), instead of the heavy or light chain variable domain of said second immunoglobulin (VD3) and instead of the heavy or light chain variable domain of said first immunoglobulin, the heavy and light chain variable domains of two humanized antibodies.

The CH domain of the human chimeric antibody or the CH domain of the antibody-like binding protein of the present invention may be any region that belongs to the heavy chains of human immunoglobulin, but those of the IgG class are suitable, and those of any of subclasses belonging to the IgG class such as IgG1, IgG2, IgG3 and IgG4. Also, as the CL of the human chimeric antibody or the CL of the antibody-like binding protein of the present invention, any region that belongs to human immunoglobulin light chains can be used, and chains belonging to the kappa or lambda class can be used.

Methods for producing humanized or chimeric antibodies include conventional recombinant DNA and gene transfection techniques known in the art (see Morrison S. L. et al. (1984) and US 5202238 and US 5204244).

Methods for producing humanized antibodies based on traditional techniques of recombinant DNA and gene transfection are well known in the art (see, for example, Riechmann L. et al. 1988; Neuberger M.S. et al. 1985). Antibodies can be humanized using a variety of techniques known in the art, including, for example, the technique disclosed in WO 2009/032661, CDR grafting (EP 239400; PCT publications WO 91/09967; US Pat. Nos. 5,225,539; 5,530,101 and 5585089), masking of surface residues or surface changes (EP 592106; EP 519596; Padlan E.A. (1991); Studnicka G.M. et al. (1994); Roguska M.A. et al. (1994)) and chain shuffling (US patent No. 5565332). The general recombinant DNA technology for producing such antibodies is also known (see European patent application EP 125023 and international patent application WO 96/02576).

The Fab of the present invention can be obtained by treating an antibody that specifically reacts with CD3 or CD123 with a protease such as papain. The Fab can also be generated by inserting DNA sequences encoding both Fab chains from the antibody into a vector for expression in a eukaryotic cell or for expression in a eukaryotic cell and introducing the vector into prokaryotic or eukaryotic cells (as appropriate) to express the Fab.

The F(ab')2 of the present invention can be produced by treating an antibody that specifically reacts with CD3 or CD123 with a protease such as pepsin. Also, F(ab')2 can be obtained by linking Fab', described below, with a thioether bond or a disulfide bond.

The Fab' of the present invention can be obtained by treating F(ab')2 which specifically reacts with CD3 or CD123, a reducing agent such as dithiothreitol. A Fab' can also be generated by inserting DNA sequences encoding Fab' chains from an antibody into a eukaryotic cell expression vector or eukaryotic cell expression vector and introducing the vector into prokaryotic or eukaryotic cells (as appropriate) to effect its expression.

The scFv of the present invention can be obtained by determining the CDR or VH and VL domain sequences described previously, constructing a DNA encoding the scFv fragment, inserting the DNA into an expression vector for a prokaryotic or eukaryotic cell, and then introducing the expression vector into prokaryotic or eukaryotic cells ( where applicable) for scFv expression. To obtain a humanized scFv fragment, a well-known technique called CDR grafting can be used, which involves selecting complementarity determining regions (CDRs) of the present invention and grafting them onto a human scFv fragment scaffold with a known three-dimensional structure (see, e.g., WO 98/45322; WO 87/02671; US 5859205; US 5585089; US 4816567; EP 0173494).

Modification of the antibodies of the present invention.

Provided(s) modification(s) of the amino acid sequence of antibodies or antibody-like binding proteins described in this document. For example, it may be desirable to enhance the binding affinity and/or other biological properties of an antibody or antibody-like binding protein. For example, it is known that when a humanized antibody is produced by simply grafting only the CDRs from the VH and VL of an antibody derived from a non-human animal into the FR from the VH and VL of a human antibody, the antigen-binding activity may be reduced compared to that of the parent antibody derived from a non-human animal. from man to animal. It is believed that certain amino acid residues VH and VL of a non-human antibody, not only in the CDR but also in the FR, may be directly or indirectly associated with antigen-binding activity. Therefore, substitution of these amino acid residues with other amino acid residues derived from the FR from the VH and VL of the human antibody will reduce the binding activity. To solve this problem regarding human antibodies grafted with CDRs from non-human antibodies, attempts have been made to identify in the amino acid sequences of FRs from VH and VL human antibodies an amino acid residue that is directly associated with antibody binding, or that interacts with the amino acid residue of the CDR, or which maintains the three-dimensional structure of the antibody and which is directly associated with antigen binding. Reduced antigen-binding activity can be increased by substituting the identified amino acids with amino acid residues from the parent antibody derived from a non-human animal. The antibody-like binding protein of the present invention may contain the variable regions of a humanized antibody and therefore the considerations mentioned in this

- 66 041572 document, are equally applicable to the antibody-like binding proteins of the present invention.

Modifications and changes can be made in the structure of the antibodies of the present invention and in the DNA sequences encoding them, and still obtain a functional antibody, antibody-like binding protein or polypeptide with the desired characteristics.

When making changes to the amino acid sequences of a polypeptide, the hydropathic index of the amino acids can be taken into account. The importance of the hydropathic index of amino acids in providing the biological function of protein interaction is generally understood in the art. It is accepted that the feature of the relative hydropathicity of the amino acid contributes to the secondary structure of the resulting protein, which, in turn, determines the interaction of the protein with other molecules, for example, enzymes, substrates, receptors, DNA, antibodies, antigens, etc. Each amino acid has been assigned a hydropathic index based on their hydrophobicity and charge characteristics, which is: isoleucine (+4.5); valine (+4.2); leucine (+3.8); phenylalanine (+2.8); cysteine/cystine (+2.5); methionine (+1.9); alanine (+1.8); glycine (-0.4); threonine (-0.7); serine (-0.8); tryptophan (-0.9); tyrosine (-1.3); proline (-1.6); histidine (-3.2); glutamate(-3.5); glutamine (-3.5); aspartate (-3.5); asparagine (-3.5); lysine (-3.9) and arginine (-4.5).

A further object of the present invention also encompasses conserved function variants of the anti-CD3 antibody, anti-CD123 antibody and antibody-like binding protein polypeptides of the present invention.

For example, certain amino acids can be replaced by other amino acids in the protein structure without significant loss of activity. Since the interoperability and nature of a protein determine its biological functional activity, it is possible to make certain amino acid substitutions in the protein sequence and, of course, in its DNA coding sequence, and still obtain a protein with similar properties. Thus, it is understood that it is possible to make various changes in the sequences of the antibodies of the present invention or the corresponding DNA sequences that encode these polypeptides without significant loss of their biological activity.

It is known in the art that certain amino acids can be replaced with other amino acids having a similar hydropathic index or score and still produce a protein with similar biological activity, i.e. still receive a protein with an equivalent biological function. It is also possible to use well known techniques, such as alanine scanning approaches, to identify in an antibody or polypeptide of the present invention all amino acids that can be substituted without significant loss of antigen binding. Such residues can be characterized as neutral because they are not involved in antigen binding or in maintaining the structure of the antibody. One or more of these neutral positions can be substituted for an alanine or other amino acid without changing the essential characteristics of the antibody or polypeptide of the present invention.

As stated above, amino acid substitutions are therefore generally based on the relative similarity of the substituents in the amino acid side chain, such as their hydrophobicity, hydrophilicity, charge, size, and the like. Illustrative substitutions that take into account various of the above features are well known to those skilled in the art and include: arginine and lysine; glutamate and aspartate; serine and threonine; glutamine and asparagine; as well as valine, leucine and isoleucine.

It may also be desirable to modify the anti-CD3 antibody, anti-CD123 antibody, and antibody-like binding protein of the present invention for effector function, for example, to increase or decrease the antigen-dependent cell-mediated cytotoxicity (ADCC) and/or complement-dependent cytotoxicity (CDC) of the antibody. . This can be achieved by introducing one or more amino acid substitutions in the Fc region of an antibody, herein resulting in so-called Fc variants in the context of the antibody-like binding proteins of the present invention. Alternatively or additionally, a cysteine residue(s) can be introduced into the Fc region, thereby allowing the formation of a disulfide bond between chains in the region. The homodimeric antibody thus obtained may have increased or decreased internalization capacity and/or increased complement-mediated cytolysis and/or antibody-dependent cellular cytotoxicity (ADCC) (Caron P.C. et al. 1992; and Shopes B. 1992).

Another type of amino acid modification in an anti-CD3 antibody, an anti-CD123 antibody, and an antibody-like binding protein of the present invention may be useful to change the original glycosylation pattern of an anti-CD3 antibody, an anti-CD123 antibody, and an antibody-like binding protein, i. by removing one or more carbohydrate moieties found in the anti-CD3 antibody, anti-CD123 antibody, and antibody-like binding protein, and/or by adding one or more glycosylation sites that were not present in the anti-CD3 antibody, anti-CD123 antibody, and antibody-like binding protein. The presence of either the asparagine-Xserine and asparagine-X-threonine tripeptide sequences, in which X is any amino acid except proline, creates a potential glycosylation site. Adding glycosylation sites to

The anti-CD3 antibody, the anti-CD123 antibody, and the antibody-like binding protein or removal from them is conveniently accomplished by altering the amino acid sequence to contain one or more of the tripeptide sequences described above (in the case of N-linked glycosylation sites).

Another type of modification involves the removal of sequences identified, either in silico or experimentally, as potentially resulting in degradation products in anti-CD3 antibody, anti-CD123 antibody, and antibody-like binding protein preparations or their heterogeneity. As examples, deamidation of aspartic and glutamine residues can occur depending on factors such as pH and surface location. Aspartic residues are particularly susceptible to deamidation, primarily when found in the Asn-Gly sequence, and to a lesser extent in other dipeptide sequences such as Asn-Ala. When such a deamidation site, in particular Asn-Gly, is present in an anti-CD3 antibody, an anti-CD123 antibody, and an antibody-like binding protein or polypeptide of the present invention, it may therefore be necessary to remove the site, usually by making a conservative substitution to remove one of the involved leftovers. It is intended that such substitutions in sequence to remove one or more of the residues involved are also covered by the present invention.

Another type of covalent modification involves the chemical or enzymatic binding of glycosides to an anti-CD3 antibody, an anti-CD123 antibody, and an antibody-like binding protein. These procedures have the advantage that they do not require the production of an anti-CD3 antibody, an anti-CD123 antibody, or an antibody-like binding protein in the host cell that is glycosylated for N- or O-linked glycosylation. Depending on the binding mode employed, the sugar(s) can be attached to (a) arginine and histidine, (b) free carboxyl groups, (c) free sulfhydryl groups such as cysteine groups, (d) free hydroxyl groups such as serine groups , threonine or hydroxyproline, (e) aromatic residues such as phenylalanine, tyrosine or tryptophan, or (f) glutamine amide group. For example, such methods are described in WO 87/05330.

Removal of any carbohydrate moieties present on an anti-CD3 antibody, an anti-CD123 antibody, or an antibody-like binding protein can be accomplished chemically or enzymatically. Chemical deglycosylation requires exposure of an anti-CD3 antibody, an anti-CD123 antibody, or an antibody-like binding protein to a trifluoromethanesulfonic acid compound or equivalent. This treatment removes most or all of the sugars except for the crosslinking sugars (N-acetylglucosamine or N-acetylgalactosamine) while leaving the antibody intact. Chemical deglycosylation is described in Sojahr H. et al. (1987) and in Edge A.S. et al. (1981). Enzymatic cleavage of carbohydrate fragments of antibodies can be achieved using various endo- and exoglycosidases described in Thotakura N.R. et al. (1987).

Another type of covalent modification of an anti-CD3 antibody, an anti-CD123 antibody, or an antibody-like binding protein includes binding the antibody to one of a variety of non-protein polymers, such as polyethylene glycol, polypropylene glycol, or polyoxyalkylenes, using the methods set forth in US Pat. Nos. 4,640,835; 4496689; 4301144; 4670417; 4791192 or 4179337.

pharmaceutical compositions.

An anti-CD3 antibody, an anti-CD123 antibody and/or an antibody-like binding protein of the present invention can be combined with pharmaceutically acceptable excipients and optionally sustained release matrices such as biodegradable polymers to form therapeutic compositions.

Thus, another object of the present invention relates to a pharmaceutical composition comprising an anti-CD3 antibody, an anti-CD123 antibody or an antibody-like binding protein of the present invention and a pharmaceutically acceptable carrier.

The present invention also relates to an anti-CD3 antibody, an anti-CD123 antibody or an antibody-like binding protein of the present invention for use as a drug. The present invention also relates to the pharmaceutical composition of the present invention for use as a drug.

The terms pharmaceutical composition or therapeutic composition as used herein refer to a compound or composition capable of inducing the desired therapeutic effect when properly administered to a patient.

Such therapeutic or pharmaceutical compositions may contain a therapeutically effective amount of an anti-CD3 antibody, an anti-CD123 antibody, or an antibody-like binding protein or drug conjugates thereof, in admixture with a pharmaceutically or physiologically acceptable formulation agent selected according to the mode of administration.

Pharmaceutical or pharmaceutically acceptable refers to molecular entities and compositions that do not cause an adverse, allergic or other undesirable reaction when administered to a mammal, in particular a human, as appropriate. A pharmaceutically acceptable carrier or excipient refers to a non-toxic solid, semi-solid, or liquid filler, diluent, encapsulating material, or excipient for any type of formulation.

As used herein, pharmaceutically acceptable carriers include all possible solvents, dispersion media, coatings, antibacterial and antifungal agents, and the like, which are physiologically compatible. Examples of suitable carriers, diluents and/or excipients include one or more of water, amino acids, saline, phosphate buffered saline, dextrose, glycerol, ethanol, and the like, as well as combinations thereof. In many cases it will be preferred to include isotonic agents such as sugars, polyhydric alcohols or sodium chloride in the composition, and the composition may also contain an antioxidant such as tryptamine and a stabilizing agent such as Tween 20.

The form of pharmaceutical compositions, route of administration, dosage and regimen, of course, depend on the condition to be treated, the severity of the disease, the age, weight and sex of the patient, and so on.

Pharmaceutical compositions of the present invention can be formulated for topical, oral, parenteral, intranasal, intravenous, intramuscular, subcutaneous or intraocular administration, and the like.

In particular, pharmaceutical compositions contain bases that are pharmaceutically acceptable for the formulation to be injected. They may be, in particular, isotonic sterile saline solutions (monosodium or disodium phosphate, sodium, potassium, calcium or magnesium chloride, etc., or mixtures of such salts) or dry, in particular freeze-dried, compositions which, when added , depending on the case, sterilized water or saline provide injectable solutions.

The doses used for administration can be adapted depending on various parameters and, in particular, depending on the mode of administration used, on the pathology concerned or, alternatively, on the required duration of treatment.

To prepare pharmaceutical compositions, an effective amount of an antibody or immunoconjugate of the present invention may be dissolved or dispersed in a pharmaceutically acceptable carrier or aqueous medium.

Pharmaceutical forms suitable for injectable use include sterile aqueous solutions or dispersions, formulations including sesame oil, peanut oil, or aqueous propylene glycol; as well as sterile powders for the preparation of sterile injectable solutions or dispersions for immediate administration. In all cases, the form must be sterile and must be fluid to the extent that it can be easily injected through a syringe. It must be stable under the conditions of manufacture and storage, and it must be protected from the contaminating action of microorganisms such as bacteria and fungi.

Solutions of the active compounds as free base or pharmacologically acceptable salts can be prepared in water by appropriate mixing with a surfactant such as hydroxypropyl cellulose. Dispersions can also be obtained in glycerol, liquid polyethylene glycols and mixtures thereof, as well as in oils. Under normal conditions of storage and use, these preparations contain a preservative to prevent the growth of microorganisms.

An anti-CD3 antibody, an anti-CD123 antibody, or an antibody-like binding protein of the present invention can be formulated in neutral or salt form. Pharmaceutically acceptable salts include acid addition salts (formed with the free amino groups of the protein) and they are formed with inorganic acids such as, for example, hydrochloric or phosphoric acid, or organic acids such as acetic, oxalic, tartaric, mandelic, etc. P. Salts formed with free carboxyl groups can also be prepared with inorganic bases such as, for example, sodium, potassium, ammonium, calcium or iron hydroxides, and organic bases such as isopropylamine, trimethylamine, glycine, histidine, procaine and etc.

The carrier may also be a solvent or dispersion medium containing, for example, water, ethanol, a polyhydric alcohol (eg, glycerol, propylene glycol and liquid polyethylene glycol, etc.), suitable mixtures thereof, and vegetable oils. Proper fluidity can be maintained, for example, by applying a coating such as lecithin, by maintaining the desired particle size in the case of a dispersion, and by using surfactants. Prevention of the action of microorganisms can be achieved with various antibacterial and antifungal agents, for example, parabens, chlorobutanol, phenol, sorbic acid, thimerosal, and the like. In many cases it will be preferable to include isotonic agents such as sugars or sodium chloride. Sustained absorption of injectable compositions can be achieved by the use of absorption delaying agents in the compositions, such as aluminum monostearate and gelatin.

Sterile injectable solutions are prepared by incorporating the active compounds in the required amount in an appropriate solvent together with various other ingredients listed above, if necessary, followed by sterilization by filtration. Generally, dispersions are prepared by incorporating the various sterilized active ingredients into a sterile base which contains the main dispersion medium and other desired ingredients from

- 69 041572 listed above. In the case of sterile powders for the preparation of sterile injectable solutions, the preferred methods of preparation are vacuum drying and freeze-drying techniques, which provide a powder form of the active ingredient with any additional necessary ingredient from its solution, previously subjected to sterilization by filtration.

A formulation with more concentrated or highly concentrated solutions for direct injection is also contemplated, and the use of DMSO as a solvent is believed to result in very rapid penetration, delivering high concentrations of active agents to a small tumor site.

Once formulated, the solutions will be administered in a manner compatible with the dosage formulation and in such amount as is therapeutically effective. The formulations are readily administered in a variety of dosage forms, such as the types of injectable solutions described above, but drug-release capsules and the like. can also be used.

For parenteral administration as an aqueous solution, for example, if necessary, the solution should be suitably buffered and the liquid diluent first made isotonic with sufficient saline or glucose. These particular aqueous solutions are particularly suitable for intravenous, intramuscular, subcutaneous and intraperitoneal administration. As such, sterile aqueous media that can be used will be known to those skilled in the art in light of the present disclosure. For example, one dose can be dissolved in 1 ml of isotonic NaCl solution and either added to 1000 ml of hypodermolysis fluid or injected at the intended infusion site (see, for example, Remington's Pharmaceutical Sciences, 15th edition, pp. 1035-1038 and 1570 -1580). There will inevitably be some dose variation depending on the condition of the subject being treated. The person responsible for administration will in any event determine the appropriate dose for the individual subject.

In one embodiment, an anti-CD3 antibody, an anti-CD123 antibody, or an antibody-like binding protein of the present invention is formulated into a therapeutic mixture containing about 0.01-100 mg per dose or so.

In addition to an anti-CD3 antibody, an anti-CD123 antibody, or an antibody-like binding protein formulated for parenteral administration, such as intravenous or intramuscular injection, other pharmaceutically acceptable forms include, for example, tablets or other solids for oral administration; sustained release capsules; and any other form currently in use.

In certain embodiments, the use of liposomes and/or nanoparticles is contemplated for introducing polypeptides, such as an anti-CD3 antibody, an anti-CD123 antibody, or an antibody-like binding protein, into host cells. The composition and use of liposomes and/or nanoparticles are known to those skilled in the art.

Nanocapsules can usually hold compounds in a stable and reproducible way. To avoid side effects due to intracellular polymer overload, such ultrafine particles (about 0.1 μm in size) are usually designed using polymers that can degrade in vivo. Biodegradable polyalkyl cyanoacrylate nanoparticles that meet these requirements are provided for use in the present invention, and such particles can be easily obtained.

Liposomes are formed from phospholipids that are dispersed in an aqueous medium and spontaneously form multilayer concentric bilayer vesicles (also called multilayer vesicles (MLV)). MLVs typically range in diameter from 25 nm to 4 µm. Sonication of the MLV results in the formation of small unilamellar vesicles (SUVs) with a diameter in the range of 200-500 A containing an aqueous solution in the core. The physical characteristics of liposomes depend on pH, ionic strength and the presence of divalent cations.

Once the pharmaceutical composition has been formulated, it can be stored in sterile vials as a solution, suspension, gel, emulsion, solid, or as a dehydrated or lyophilized powder. Such formulations can be stored either in a ready-to-use form or in a form (eg, lyophilized) requiring reconstitution prior to administration.

Therapeutic methods and methods of application.

The present inventors have shown for some of the bispecific compounds of the present invention, such as hz20G6x7G3, 7G3xhz4B4, hz4B4x3E3 and hz20G6x7G3-TL4, T cell mediated cytotoxicity in a CD 123 positive cancer cell line model. The bispecific compounds of the present invention activate T cells in the presence of target cells, resulting in cytotoxicity against tumor cells. The present inventors also demonstrated low or no T cell activation in the absence of target cells.

It is well known that anti-CD123 therapeutic monoclonal antibodies can deplete the population of cells bearing an antigen specifically recognized by the antibody. This population depletion can be mediated by at least three mechanisms: antibody-dependent cellular cytotoxicity (ADCC), complement-dependent cytotoxicity (CDC), and direct suppression of tumor growth by signals transmitted through the antigen targeted by the antibody. In one embodiment, the anti-CD123 antibodies of the present invention induce cytotoxicity in cells expressing CD123 via antibody dependent cellular cytotoxicity (ADCC), complement dependent cytotoxicity (CDC).

Complement-dependent cytotoxicity or CDC refers to the lysis of a target cell in the presence of complement. Activation of the classical complement pathway is initiated by the binding of the first component of the complement system to antibodies that have bound to their respective antigen. To assess complement activation, a CDC assay can be performed, for example, as described in GazzanoSantoro et al. (1997).

Antibody-dependent cell-mediated cytotoxicity, or ADCC, is a form of cytotoxicity in which secreted antibodies bound to Fc receptors (FcRs) present on certain cytotoxic cells (eg, natural killer (NK) cells, neutrophils, and macrophages) allow these cytotoxic effector cells to specifically bind to a target cell carrying an antigen and then kill the target cell. An in vitro ADCC assay, such as described in US Pat. No. 5,500,362 or US Pat.

As described above under Anti-CD3 Antibodies, the anti-CD3 antibodies of the present invention have low T-cell activation, thus they have therapeutic potential in a subject for use as an immunosuppressive agent.

In addition, the antibody-like binding protein as defined above aims to enhance a patient's immune response against tumors by targeting T cells to tumor cells. In one embodiment, an antibody-like binding protein as defined above is targeted to the CD3ε subunit of the T cell receptor (TCR) on the surface of the T cell and the other arm is targeted to a CD123 expressing cancer cell such that the T cell and the cancer cell are co-engaged using a bispecific construct leads to the formation of a cytolytic synapse, which induces the activation of T cells and leads to the cytolysis of tumor cells. Cytolysis of tumor cells can be mediated by at least two mechanisms: perforin/granzyme-mediated cytolysis and FasL/Fas-mediated cytolysis, eg, perforin/granzyme-mediated cytolysis.

Therefore, in one embodiment, the present invention provides a method of treating or preventing a disease or disorder, comprising administering to a subject in need thereof a therapeutically effective amount of an anti-CD3 antibody, an anti-CD123 antibody, an antibody-like binding protein, or a pharmaceutical composition of the present invention as defined above in section Pharmaceutical composition.

The present invention further relates to the use of an anti-CD3 antibody, an anti-CD123 antibody, an antibody-like binding protein or a pharmaceutical composition of the present invention for the preparation of a medicament for the treatment or prevention of a disease or disorder in a subject. In one embodiment, the present invention relates to the use of an anti-CD3 antibody, an anti-CD123 antibody, an antibody-like binding protein, or a pharmaceutical composition for the treatment or prevention of a disease or disorder in a subject.

The terms subject or individual are used interchangeably and may refer to, for example, a human or a non-human mammal. For example, the subject is a bat; ferret; a rabbit; feline representative (cat); canine representative (dog); primate (monkey), equine representative (horse); person, including man, woman and child. In one embodiment, the subject is a human.

In the context of the present invention, the term treatment or treatment refers to therapeutic use (i.e., in a subject having a given disease), and it means the reversal, alleviation, suppression of the progression of one or more symptoms of such a disorder or condition. Therefore, a treatment refers not only to a treatment that leads to a complete cure of a disease, but also to treatments that slow down the progression of the disease and/or prolong the survival of the subject.

By prevention is meant prophylactic use (ie, a subject susceptible to the development of a given disease).

In one embodiment, the disease or disorder is any condition that would benefit from treatment with an anti-CD123 antibody or antibody-like binding protein of the present invention. In one embodiment, it includes chronic or acute disorders or diseases, including those pathological conditions that predispose the subject to the disorders in question. The term in need of treatment refers to a subject who already has a disease, as well as those in whom the disorder should be prevented.

In one embodiment, the disease or disorder is any condition that would benefit from treatment with an anti-CD3 antibody of the present invention.

Therefore, in one embodiment, it includes diseases or disorders characterized by abnormal immune responses.

In the context of the present invention, an abnormal immune response is an inflammatory immune response.

In one embodiment, the disease characterized by an abnormal immune response is an autoimmune disease, a disease associated with transplantation, or a disease associated with inflammation.

The autoimmune disease is, for example, Crohn's disease, ulcerative colitis, and type 1 diabetes, or a transplant-associated disease such as graft-versus-host disease (GVHD).

Therefore, in one embodiment, the subject has been diagnosed as suffering from a disease or disorder characterized by an abnormal immune response.

In one embodiment, the subject has been diagnosed as suffering from an autoimmune disease.

In another embodiment, the disorder is referred to as cancer.

In an additional embodiment, cancer refers to hemoblastosis, in particular hemoblastosis associated with the expression of CD123.

In one embodiment, expression of CD123 by cancer cells is readily assayed, for example, using an anti-CD123 antibody of the present invention, as described in the following section, Diagnostic Applications.

Hemoblastoses associated with CD123 expression include types of leukemia (such as acute myeloid leukemia, chronic myeloid leukemia, acute lymphoblastic leukemia, chronic lymphoblastic leukemia, malignant reticuloendotheliosis, and myelodysplastic syndrome) and malignant lymphoproliferative conditions, blast plasmacytoid dendritic cell neoplasm (BPDCN), systemic mastocytosis, including types of lymphomas (such as multiple myeloma, non-Hodgkin's lymphoma, Burkitt's lymphoma, and small and large cell follicular lymphoma).

As described above under Anti-CD123 Antibodies, LSCs express CD123.

Thus, in a related embodiment, cancer refers to hemoblastosis associated with leukemic stem cells.

Leukemic stem cell (LSC) associated hemoblastosis conditions that can be treated in accordance with the present invention include types of leukemias (such as acute myeloid leukemia, chronic myeloid leukemia, acute lymphoblastic leukemia, chronic lymphoblastic leukemia, and myelodysplastic syndrome) and malignant lymphoproliferative conditions. , including types of lymphomas (such as multiple myeloma, non-Hodgkin's lymphoma, Burkitt's lymphoma, and small cell and large cell follicular lymphoma).

In one aspect of the present invention, the hemoblastosis is acute myeloid leukemia (AML).

In one embodiment, the subject has been diagnosed as suffering from AML.

In a further embodiment, the subject has already been treated with chemotherapy to complete remission but has relapsed.

Relapse is defined as a recurrence of AML after complete remission.

Complete remission or CR is defined as follows: normal neutrophil count (>1.0*10 ⁹ /l), hemoglobin level 10 g/dl and platelet count (>100*10 ⁹ /l) and no need for red blood cell transfusion; less than 5% blast cells, no clusters or clusters of blast cells, and no Auer bodies on bone marrow examination; as well as normal maturation of blood cells (morphology; myelogram) and absence of extramedullary leukemia.

In one embodiment, an anti-CD3 antibody, an anti-CD123 antibody, or an antibody-like binding protein of the present invention is used alone or in combination with any suitable growth inhibitor.

By a therapeutically effective amount of a polypeptide of the present invention is meant an amount of the polypeptide sufficient to treat said cancer at an acceptable benefit/risk ratio applicable to any therapeutic treatment. It will be understood, however, that the total daily amount of polypeptides and compositions of the present invention to be used will be determined by the attending physician within the bounds of sound medical judgment. The specific therapeutically effective dose level for any particular patient will depend on a number of factors including the disorder being treated and the severity of the disorder; the activity of the particular polypeptide used; the specific composition used, age, body weight, general health, sex and diet of the patient; the time of administration, route of administration, and rate of excretion of the particular polypeptide used; duration of treatment; drugs used in combination or simultaneously with the specific polypeptide used; and similar factors well known in the medical field. For example, it is well known within the skill in the art to start with doses of the compound at levels lower than those required to achieve the desired therapeutic effect and gradually increase the dosage until the desired effect is achieved.

In one embodiment, the effectiveness of treatment with an anti-CD3 antibody, an anti-CD123 antibody, or an antibody-like binding protein of the present invention or an immunoconjugate of the present invention is readily analyzed in vivo, for example, in a mouse model of cancer, and by measuring, for example, changes in tumor volume between groups treatment and control group.

Diagnostic applications

It has been reported that CD123 is overexpressed on the surface of various hematological neoplasms.

Therefore, CD123 is a marker of cancer and therefore has the potential to be used to measure the effectiveness of anti-cancer therapy or detect disease recurrence.

In an embodiment, an anti-CD123 antibody of the present invention is used as a component of an assay in the context of therapy targeting tumors expressing CD123 to determine a patient's responsiveness to a therapeutic agent, monitor the effectiveness of an anti-cancer therapy, or detect disease recurrence after treatment. In particular, the same anti-CD123 antibody of the present invention is used both as a component of a therapeutic agent and as a component of a diagnostic assay.

Thus, a further object of the present invention relates to an anti-CD123 antibody of the present invention for use in in vivo detection of CD123 expression in a subject or for use in ex vivo detection of CD123 expression in a biological sample from a subject. In one embodiment, the specified detection is intended, in particular, for:

a) diagnosing the presence of cancer in the subject, or

b) determining the susceptibility of a patient having cancer to a CD123-targeting therapeutic agent, in particular an anti-CD123 antibody or an antibody-like binding protein of the present invention, or

c) monitoring the effectiveness of CD123-targeted cancer therapy or detecting cancer recurrence after CD123-targeted cancer therapy, in particular in the case of therapy with an anti-CD123 antibody or an antibody-like binding protein of the present invention; by detecting CD123 surface protein expression on tumor cells.

In an embodiment, the antibody is provided for in vitro or ex vivo use. For example, CD123 is detected in vitro or ex vivo in a biological sample obtained from a subject using an anti-CD123 antibody of the present invention. The use according to the present invention may also be an in vivo use. For example, an anti-CD123 antibody or an antibody-like binding protein of the present invention is administered to a subject and antibody cell complexes are detected and/or quantified, the detection of these complexes being indicative of cancer.

The present invention further relates to an in vitro or ex vivo method for detecting the presence of cancer in a subject, comprising the steps consisting of:

(a) bringing a biological sample from the subject into contact with the antibody to CD123 according to the present invention, in particular, under conditions suitable for the antibody to form complexes with the specified biological sample;

(b) measuring the level of antibody bound to said biological sample, (c) detecting the presence of cancer by comparing the measured level of bound antibody to a control, wherein an increased level of bound antibody compared to the control is indicative of cancer.

The present invention also relates to an in vitro or ex vivo method for determining the susceptibility of a patient having cancer to a therapeutic agent targeting CD123, in particular an anti-CD123 antibody or an antibody-like binding protein according to the present invention, the method comprising the steps consisting of:

(a) contacting a biological sample from a patient with cancer with an anti-CD123 antibody of the present invention, in particular under conditions suitable for the antibody to form complexes with said biological sample;

(b) measuring the level of antibody bound to said biological sample, (c) comparing the measured level of antibody bound to said biological sample with the level of antibody bound to a control;

- 73 041572 where an increased level of antibody bound to the specified biological sample, compared with the control is an indication that the patient is susceptible to a therapeutic agent that targets at least CD123.

In the above methods, said control may be a normal, cancer-free biological sample of the same type, or a reference value, a determined typical value of the level of antibody binding in a normal biological sample of the same type.

In an embodiment, the anti-CD123 antibody of the present invention is useful in diagnosing hematological malignancies associated with CD123 expression, including types of leukemias (such as acute myeloid leukemia, chronic myeloid leukemia, acute lymphoblastic leukemia, chronic lymphoblastic leukemia, malignant reticuloendotheliosis, and myelodysplastic syndrome) and malignant lymphoproliferative diseases. conditions, blast plasmacytoid dendritic cell neoplasm (BPDCN), systemic mastocytosis, including types of lymphoma (such as multiple myeloma, non-Hodgkin's lymphoma, Burkitt's lymphoma, and small and large cell follicular lymphoma).

The present invention further relates to an in vitro or ex vivo method for monitoring the effectiveness of CD123-targeted cancer therapy, comprising the steps consisting of:

(a) contacting a biological sample from a subject undergoing CD123-targeted cancer therapy with an anti-CD123 antibody or antibody-like binding protein of the present invention, in particular under conditions suitable for the antibody to form complexes with said biological sample;

(b) measuring the level of antibody bound to said biological sample, (c) comparing the measured level of bound antibody with the level of antibody bound to the control;

wherein a reduced level of antibody bound to said biological sample compared to a control is indicative of the efficacy of said CD123-targeted cancer therapy.

In said method, an increased level of antibody bound to said biological sample compared to a control is indicative of failure of said CD123-targeted cancer therapy.

The specified control is, in particular, a biological sample of the same type as the biological sample provided for analysis, but which was obtained from the subject earlier in time, during cancer therapy aimed at CD123.

The present invention further relates to an in vitro or ex vivo method for detecting cancer recurrence after cancer therapy directed at CD123, comprising the steps consisting of:

(a) contacting a biological sample from a subject who has completed CD123-targeted cancer therapy with an anti-CD123 antibody or antibody-like binding protein of the present invention, in particular under conditions suitable for the antibody to form complexes with said biological sample;

(b) measuring the level of antibody bound to said biological sample, (c) comparing the measured level of bound antibody with the level of antibody bound to the control;

where an increased level of antibody bound to the specified biological sample, compared with the control is indicative of cancer recurrence after cancer therapy directed to CD123.

The specified control is, in particular, a biological sample of the same type as the biological sample provided for analysis, but which was obtained from the subject earlier in time, at the time of completion or after completion of cancer therapy directed to CD123.

Said cancer therapy directed to CD123 is, in particular, therapy using an anti-CD123 antibody or an antibody-like binding protein or an immunoconjugate according to the present invention. Said CD123-targeted cancer therapy targets CD123-expressing cancer, in particular CD123-expressing hemoblastosis, including types of leukemia (such as acute myeloid leukemia, chronic myeloid leukemia, acute lymphoblastic leukemia, chronic lymphoblastic leukemia, malignant reticuloendotheliosis, and myelodysplastic syndrome) and malignant lymphoproliferative conditions, blast plasmacytoid dendritic cell neoplasm (BPDCN), systemic mastocytosis, including types of lymphoma (such as multiple myeloma, non-Hodgkin's lymphoma, Burkitt's lymphoma, and small and large cell follicular lymphoma).

In an embodiment, an anti-CD123 antibody of the present invention is labeled with a detectable molecule or substance, such as a fluorescent molecule, a radioactive molecule, or any other label known to provide (either directly or indirectly) a signal.

As used herein, the term labeled, in relation to an anti-CD123 antibody of the present invention, is meant to encompass direct labeling of an anti-CD123 antibody by

- 74 041572 binding (i.e. a physical connection) of a detectable substance, such as a radioactive agent or a fluorophore (for example, fluorescein isothiocyanate (FITC), or phycoerythrin (PE), or indocyanine (Cy5)) with a polypeptide, as well as indirect labeling of the polypeptide due to ability to react with the substance to be detected.

In a further embodiment, an anti-CD123 antibody of the present invention is labeled with a radioactive molecule using any method known in the art. For example, radioactive molecules include, without limitation, radioactive atoms for scintigraphy studies, such as I ¹²³ , I ¹²⁴ , In ¹¹¹ , Re ¹⁸⁶ , Re ¹⁸⁸ , Tc ⁹⁹ . In one example, the polypeptides of the present invention are also labeled with a spin label for nuclear magnetic resonance (NMR) imaging (also known as magnetic resonance imaging, MRI), such as iodine-123, indium-111, fluorine-19, carbon -13, nitrogen-15, oxygen-17, gadolinium, manganese or iron.

A biological sample covers many types of samples obtained from a subject and may be used in a diagnostic or monitoring assay. Biological samples include, without limitation, blood and other liquid samples of biological origin, solid tissue samples such as biopsies or tissue cultures, or cells derived therefrom, and progeny thereof. Therefore, biological samples include clinical samples, cells in culture, cell supernatants, cell lysates, blood serum, blood plasma, biological fluids, and tissue samples, in particular a tumor sample.

The present invention also relates to an in vivo method for detecting the presence of cancer in a subject, comprising the steps consisting of:

a) administering an antibody of the present invention, which is labeled with a detectable label, to a patient;

b) detecting the localization of said detectably labeled antibody in the patient by imaging.

In one embodiment, the antibodies of the present invention are useful for cancer staging (eg, radioisotope imaging). They are used, for example, alone or in combination with other cancer markers.

The terms detection or detected as used herein include qualitative and/or quantitative detection (measurement of levels) with or without reference to a control.

In the context of the present invention, the term diagnosis, as used herein, means determining the nature of a medical condition, intended to identify a pathology that adversely affects the subject based on a set of collected data.

In said method, the cancer is a cancer that expresses CD123 as defined above.

Sets.

Finally, the present invention also provides kits containing at least one anti-CD3 antibody, at least one anti-CD123 antibody, or at least one antibody-like binding protein of the present invention. Kits containing anti-CD123 antibodies or anti-CD3 antibodies of the present invention find use in detection of CD123 or CD3 surface protein or in therapeutic or diagnostic assays. The kits of the present invention may comprise an anti-CD3 polypeptide or antibody, at least one anti-CD123 antibody, or at least one antibody-like binding protein bound to a solid support, such as a tissue culture plate, or beads (eg, Sepharose beads). Kits may be provided that contain antibodies to detect and quantify CD123 or CD3 surface protein in vitro, for example in an ELISA assay or Western blot. In one embodiment, said antibody is useful for detection and is provided with a label, such as a fluorescent or radioactive label.

In one embodiment, the present invention encompasses kits for producing a single dose unit. Each of the kits may contain both the first container with the dried protein and the second container with the aqueous composition. Also included within the scope of the present invention are kits containing single and multi-chamber pre-filled syringes (eg, liquid syringes and lyophilisate syringes).

Throughout this application, the term containing should be interpreted to encompass all of the features specifically mentioned, as well as optional, additional, unspecified features. As used herein, the use of the term comprising also discloses an embodiment in which no features are present except for the features specifically mentioned (ie, consisting of). Moreover, the use of the singular form does not preclude the plural. The fact that certain measures are set forth in mutually exclusive dependent claims does not in itself indicate that a combination of these measures cannot be used to an advantage.

- 75 041572

The present invention will now be described in more detail with reference to the following figures and examples. All publications and patent documents cited herein are hereby incorporated by reference. While the present invention has been illustrated and described in detail in the foregoing description, the examples are to be considered illustrative or exemplary and not limiting.

Brief description of the sequences

SEQ ID NO: 1 shows the amino acid sequence of the full length human CD3ε protein, including the signal peptide, available from the Uniprot database under accession number P07766.

Under SEQ ID NO: 2 shows the amino acid sequence of the full-length Macaca fascicularis CD3ε protein, including the signal peptide, available from the Uniprot database under accession number Q95LI5.

SEQ ID NO: 3 shows the amino acid sequence of a fusion containing Fc and a His-tagged mature human CD3ε that contains amino acids 23-126 from the full-length wild-type human CD3ε protein.

SEQ ID NO: 4 shows the amino acid sequence of a fusion containing Fc and a mature Macaca fascicularis CD3ε that contains amino acids 23-117 of the full-length Macaca fascicularis CD3ε protein (SEQ ID NO:2) containing one Ala to Val substitution at amino acid position 35 compared to amino acid position 57 of the wild-type sequence.

Under SEQ ID NO: 5 shows the amino acid sequence of the variable domain of the heavy chain of the antibody, called 20G6-F3.

Under SEQ ID NO: 6, 7 and 8 shows the amino acid sequences of CDR1-H, CDR2-H and CDR3H of an antibody called 20G6-F3.

Under SEQ ID NO: 9 shows the amino acid sequence of the variable domain of the light chain of the antibody, called 20G6-F3.

Under SEQ ID NO: 10 shows the amino acid sequence of CDR1-L antibodies referred to as 20G6-F3, 11D7-C3, 13H2-C2, 13C1-F6, 1E6-C9, 10F4-C10, 18G9-H11, 12G3-E8, 5B1-G2 , 16F8-A7, 11F9-F8, 8C2-F7, 20E5-F10 and 3H6-D2.

Under SEQ ID NO: 11 shows the amino acid sequence of CDR3-L antibodies called 20G6-F3, 4B4-O7, 11H3-E5, 13H2-C2, 13C1-F6, 10F4-C10, 4E7-C9, 11F3-B9, 12G3-E8 , 5B1-G2, 16F8-A7, 20E5-F10 and 3H6-D2.

Under SEQ ID NO: 12 shows the amino acid sequence of the variable domain of the heavy chain of the antibody, called 4B4-D7.

Under SEQ ID NO: 13 shows the amino acid sequence of the CDR1-H antibodies referred to as 4B4-D7, 11D7-C3, 11H3-E5, 13H2-C2, 13C1-F6, 10F4-C10, 18G9-H11, 4E7-C9, 11F3-B9 , 16F8-A7, 11F9-F8, 20B5-F10 and 3H6-D2.

Under SEQ ID NO: 14 and 15 shows the amino acid sequences of CDR2-H and CDR3-H of an antibody called 4B4-O7.

Under SEQ ID NO: 16 shows the amino acid sequence of the variable domain of the light chain of the antibody, called 4B4-D7.

Under SEQ ID NO: 17 shows the amino acid sequence of the CDR1-L antibodies called 4B4-D7, 11H3-E5 and 11F3-B9.

Under SEQ ID NO: 18 shows the amino acid sequence of the variable domain of the heavy chain of the antibody, called 4E7-C9.

Under SEQ ID NO: 19 shows the amino acid sequence of the CDR2-H antibodies, called 4E7-C9, 18F5-H10, 20E5-F10 and 3H6-D2.

Under SEQ ID NO: 20 shows the amino acid sequence of the CDR3-H of the antibody called 4E7-C9.

Under SEQ ID NO: 21 shows the amino acid sequence of the variable domain of the light chain of the antibody, called 4E7-C9.

Under SEQ ID NO: 22 shows the amino acid sequence of the CDR1-H of the antibody called 4E7-C9.

Under SEQ ID NO: 23 shows the amino acid sequence of the variable domain of the heavy chain of the antibody, called 18F5-H10.

Under SEQ ID NO: 24 shows the amino acid sequence of the CDRl-H of the antibody called 18F5-H10.

Under SEQ ID NO: 25 shows the amino acid sequence of the CDR3-H of an antibody called 18F5-H10.

Under SEQ ID NO: 26 shows the amino acid sequence of the variable domain of the light chain of the antibody, called 18F5-H10.

Shown under SEQ ID NO: 27 is the amino acid sequence of the CDRl-L of an anti-CD3 antibody called 18F5-H10.

- 76 041572

Under SEQ ID NO: 28 shows the amino acid sequence of the CDR3-L antibodies to CD3, called 18F5-H10, 11D7-C3, 1E6-C9 and 10E6-G6.

Under SEQ ID NO: 29 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 12D2-E5.

Shown under SEQ ID NOs: 30 and 31 are the amino acid sequences of CDR1-H and CDR2-H of an anti-CD3 antibody called 12D2-E5.

Under SEQ ID NO: 32 shows the amino acid sequence of the CDR3-H antibodies to CD3, called 12D2-E5 and 3G5-E10.

SEQ ID NO: 33 shows the amino acid sequence of the light chain variable domain of an anti-CD3 antibody called 12D2-E5.

Shown under SEQ ID NOs: 34 and 35 are the amino acid sequences of the CDR1-H and CDR3-H of an anti-CD3 antibody called 12D2-E5.

Under SEQ ID NO: 36 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 11D7-C3.

SEQ ID NO: 37 shows the amino acid sequence of the CDR2-H of anti-CD3 antibodies referred to as 11D7-C3, 11H3-E5, 13H2-C2, 13C1-F6, 1E6-C9, 10F4-C10, 18O9-H11, 11F3-B9, 16F8 -A7, 11F9-F8 and 20B5-F10.

Under SEQ ID NO: 38 shows the amino acid sequence of the CDR3-H of an anti-CD3 antibody called 11D7-C3.

Under SEQ ID NO: 39 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 11D7-C3.

Under SEQ ID NO: 40 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 11H3-E5.

Under SEQ ID NO: 41 shows the amino acid sequence of the CDR3-H of an anti-CD3 antibody called 11H3-E5.

Under SEQ ID NO: 42 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 11H3-E5.

Under SEQ ID NO: 43 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 13H2-C2.

SEQ ID NO: 44 shows the amino acid sequence of the CDR3-H of an anti-CD3 antibody called 13H2-C2.

Under SEQ ID NO: 45 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 13H2-C2.

Under SEQ ID NO: 46 shows the amino acid sequence of the heavy chain variable domain of antibodies to CD3, called 13C1-F6 and 11F9-F8.

Under SEQ ID NO: 47 shows the amino acid sequence of the CDR3-H antibodies to CD3, called 13C1-F6, 10E6-G6 and 11F9-F8.

Under SEQ ID NO: 48 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 13H2-C2.

Under SEQ ID NO: 49 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 18H11-F10.

Shown under SEQ ID NOs: 50, 51 and 52 are the amino acid sequences of CDR1-H, CDR2-H and CDR3-H of an anti-CD3 antibody called 18H11-F10.

Under SEQ ID NO: 53 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 18H11-F10.

Shown under SEQ ID NOs: 54 and 55 are the amino acid sequences of CDR1-L and CDR3-L of an anti-CD3 antibody called 18H11-F10.

Under SEQ ID NO: 56 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 1E6-C9.

Shown under SEQ ID NOs: 57 and 58 are the amino acid sequences of CDR1-H and CDR3-H of an anti-CD3 antibody called 1E6-C9.

Under SEQ ID NO: 59 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 1E6-C9.

Under SEQ ID NO: 60 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 10F4-C10.

Shown under SEQ ID NO: 61 is the amino acid sequence of the CDR3-H of an anti-CD3 antibody called ^4-00.

Under SEQ ID NO: 62 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 10F4-C10.

Under SEQ ID NO: 63 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 10E6-G6.

Shown under SEQ ID NOs: 64 and 65 are the amino acid sequences of CDR1-H and CDR2-H of an anti-CD3 body called 10E6-G6.

Under SEQ ID NO: 66 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 10E6-G6.

Under SEQ ID NO: 67 shows the amino acid sequence of the CDR1-L of an anti-CD3 antibody called 10E6-G6.

Under SEQ ID NO: 68 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 18G9-H11.

SEQ ID NO: 69 shows the amino acid sequence of the CDR3-H of an anti-CD3 antibody called 18O9-H11.

Under SEQ ID NO: 70 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 18G9-H11.

SEQ ID NO: 71 shows the amino acid sequence of the CDR3-L of an anti-CD3 antibody called 18G9-H11.

Under SEQ ID NO: 72 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 11F3-B9.

Under SEQ ID NO: 73 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 11F3-B9.

Under SEQ ID NO: 74 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 12G3-E8.

Shown under SEQ ID NOs: 75, 76 and 77 are the amino acid sequences of CDR1-H, CDR2-H and CDR3-H of an anti-CD3 antibody called 12G3-E8.

Under SEQ ID NO: 78 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 12G3-E8.

Under SEQ ID NO: 79 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 5B1-G2.

Shown under SEQ ID NOs: 80 and 81 are the amino acid sequences of CDR1-H and CDR3-H of an anti-CD3 antibody called 5B1-G2.

Under SEQ ID NO: 82 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 5B1-G2.

Under SEQ ID NO: 83 shows the amino acid sequence of part of the variable domain of the heavy chain of the anti-CD3 antibody, called 16F8-A7.

Under SEQ ID NO: 84 shows the amino acid sequence of the CDR3-H antibodies to CD3, called 16F8-A7 and 11F3-B9.

Under SEQ ID NO: 85 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 16F8-A7.

SEQ ID NO: 86 shows the amino acid sequence of the full length human CD3δ protein, including the signal peptide, available from the Uniprot database under accession number P04234.

Under SEQ ID NO: 87 shows the amino acid sequence of the variable domain of the light chain of the antibody to CD3, called 11F9-F8.

SEQ ID NO: 88 shows the amino acid sequence of the CDR3-L of an anti-CD3 antibody called 11F9-F8.

Under SEQ ID NO: 89 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 3G5-E10.

Shown under SEQ ID NOs: 90 and 91 are the amino acid sequences of CDR1-H and CDR2-H of an anti-CD3 antibody called 3G5-E10.

Under SEQ ID NO: 92 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 3G5-E10.

Shown under SEQ ID NOs: 93 and 94 are the CDR1-L and CDR3-L sequences of an anti-CD3 antibody called 3G5-E10.

Under SEQ ID NO: 95 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 9D7-F3.

Shown under SEQ ID NOs: 96, 97 and 98 are the amino acid sequences of CDR1-H, CDR2-H and CDR3-H of an anti-CD3 antibody called 9D7-F3.

Under SEQ ID NO: 99 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 9D7-F3.

Shown under SEQ ID NOs: 100 and 101 are the amino acid sequences of CDR1-L and CDR3-L of anti-CD3 antibodies referred to as 9D7-F3 and 6C9-C9.

Under SEQ ID NO: 102 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 8C2-F7.

Shown under SEQ ID NOs: 103, 104 and 105 are the amino acid sequences of CDR1-H, CDR2-H and CDR3-H of an anti-CD3 antibody called 8C2-F7.

Under SEQ ID NO: 106 shows the amino acid sequence of the variable domain of light

- 78 041572 chain of an anti-CD3 antibody called 8C2-F7.

Under SEQ ID NO: 107 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 20E5-F10.

SEQ ID NO: 108 shows the amino acid sequences of the CDR3-H of an anti-CD3 antibody called 2OE5-F1O.

Under SEQ ID NO: 109 shows the amino acid sequence of the variable domain of the light chain of the antibody to CD3, called 20E5-F10.

Under SEQ ID NO: 110 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 20B5-F10.

SEQ ID NO: 111 shows the amino acid sequence of the CDR3-H of an anti-CD3 antibody called 2OB5-F1O.

Under SEQ ID NO: 112 shows the amino acid sequence of the variable domain of the light chain of the antibody to CD3, called 20B5-F10.

Shown under SEQ ID NOs: 113 and 114 are the amino acid sequences of CDR1-L and CDR3-L of an anti-CD3 antibody called 20B5-F10.

Under SEQ ID NO: 115 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 6C9-C9.

Under SEQ ID NO: 116, 117 and 118 shows the amino acid sequences of CDR1-H, CDR2-H and CDR3-H of an anti-CD3 antibody called 6C9-C9.

Under SEQ ID NO: 119 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 6C9-C9.

SEQ ID NO: 120 shows the amino acid sequence of the CDR3-L of an anti-CD3 antibody called 6C9-C9.

Under SEQ ID NO: 121 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 3E8-G1.

Under SEQ ID NO: 122, 123 and 124 shows the amino acid sequences of CDR1-H, CDR2-H and CDR3-H of an anti-CD3 antibody called 3E8-G1.

Under SEQ ID NO: 125 shows the amino acid sequence of the light chain variable domain of the anti-CD3 antibody, called 3E8-G1.

Shown under SEQ ID NOs: 126 and 127 are the amino acid sequences of CDR1-L and CDR3-L of an anti-CD3 antibody called 3E8-G1.

Under SEQ ID NO: 128 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 3H6-D2.

SEQ ID NO: 129 shows the amino acid sequence of the CDR3-H of an anti-CD3 antibody called 3H6-D2.

SEQ ID NO: 130 shows the amino acid sequence of the light chain variable domain of an anti-CD3 antibody called 3H6-D2.

Under SEQ ID NO: 131 shows the amino acid sequence of the heavy chain variable domain of the anti-CD3 antibody, called 8H2.

SEQ ID NO: 132 shows the amino acid sequence of the light chain variable domain of an anti-CD3 antibody called 8H2.

Shown under SEQ ID NOs: 133 and 134 are the amino acid sequences of CDR1-L and CDR3-L of an anti-CD3 antibody called 8H2.

SEQ ID NO: 135 shows the amino acid sequence of the VH variant, VH1a, of the 20G6 humanized anti-CD3 antibody.

SEQ ID NO: 136 shows the amino acid sequence of the VH variant, VH1b, of the humanized anti-CD3 antibody 20G6.

SEQ ID NO: 137 shows the amino acid sequence of the VH variant, VH1c, of the humanized anti-CD3 antibody 20G6.

SEQ ID NO: 138 shows the amino acid sequence of the VH variant, VH1d, of the humanized anti-CD3 antibody 20G6.

SEQ ID NO: 139 shows the amino acid sequence of the VL variant, VL1a, of the humanized anti-CD3 antibody 20G6.

SEQ ID NO: 140 shows the amino acid sequence of the VL variant, VL1b, of the humanized anti-CD3 antibody 20G6.

SEQ ID NO: 141 shows the amino acid sequence of the VL variant, VL1c, of the humanized anti-CD3 antibody 20G6.

SEQ ID NO: 142 shows the amino acid sequence of the CDR1-L variant VL1c of the humanized anti-CD3 antibody 20G6.

SEQ ID NO: 143 shows the amino acid sequence of the VL variant, VL1d, of the humanized anti-CD3 antibody 20G6.

SEQ ID NO: 144 shows the amino acid sequence of the VH variant, VH2a, of the humanized anti-CD3 antibody 20G6.

SEQ ID NO: 145 shows the amino acid of the anti-CD3 antibody 20G6.

SEQ ID NO: 146 shows the amino acid of the anti-CD3 antibody 20G6.

SEQ ID NO: 147 shows the amino acid of the anti-CD3 antibody 20G6.

SEQ ID NO: 148 shows the amino acid of the anti-CD3 antibody 20G6.

SEQ ID NO: 149 shows the amino acid of the anti-CD3 antibody 20G6.

SEQ ID NO: 150 shows the amino acid of the anti-CD3 antibody 20G6.

SEQ ID NO: 151 shows the amino acid of the anti-CD3 antibody 20G6.

SEQ ID NO: 152 shows the amino acid of the anti-CD3 antibody 20G6.

SEQ ID NO: 153 shows the amino acid of the anti-CD3 antibody 20G6.

SEQ ID NO: 154 shows the amino acid of the anti-CD3 antibody 20G6.

SEQ ID NO: 155 shows the amino acid of the anti-CD3 antibody 20G6.

SEQ ID NO: 156 shows the amino acid of the anti-CD3 antibody 20G6.

SEQ ID NO: 157 shows the amino acid of the anti-CD3 antibody 20G6.

SEQ ID NO: 158 shows the amino acid of the anti-CD3 antibody 4B4.

SEQ ID NO: 159 shows the amino acid of the anti-CD3 antibody 4B4.

SEQ ID NO: 160 shows the amino acid of the anti-CD3 antibody 4B4.

SEQ ID NO: 161 shows the amino acid of the anti-CD3 antibody 4B4.

SEQ ID NO: 162 shows the amino acid of the manized anti-CD3 antibody 4B4.

Under SEQ ID NO: 163 shows the amino acid manized antibodies to CD3 4B4.

SEQ ID NO: 164 shows the amino acid of the manized anti-CD3 antibody 4B4.

SEQ ID NO: 165 shows the amino acid of the manized anti-CD3 antibody 4B4.

SEQ ID NO: 166 shows the amino acid of the manized anti-CD3 antibody 4B4.

SEQ ID NO: 167 shows the amino acid of the manized anti-CD3 antibody 4B4.

SEQ ID NO: 168 shows the amino acid of the anti-CD3 antibody 4B4.

SEQ ID NO: 169 shows the amino acid of the anti-CD3 antibody 4B4.

SEQ ID NO: 170 shows the amino acid of the anti-CD3 antibody 4B4.

SEQ ID NO: 171 shows the amino acid of the anti-CD3 antibody 4B4.

SEQ ID NO: 172 shows the amino acid of the anti-CD3 antibody 4B4.

SEQ ID NO: 173 shows the amino acid of the anti-CD3 antibody 4B4.

SEQ ID NO: 174 shows the amino acid of the anti-CD3 antibody 4B4.

SEQ ID NO: 175 shows the amino acid sequence of VH variant, VH2b, humanization sequence of variant VH, VH2c, humanization sequence of variant VH, VH2d, humanization sequence of variant VL, VL2a, humanization sequence of variant VL, VL2b, humanization sequence of variant VL, VL2c, humanization sequence of variant VL, VL2d, VH variant humanization sequence, VH3a, VH variant humanization sequence, VH3b, VL variant humanization sequence, VL3a, VL variant humanization sequence, VL3b, VL variant humanization sequence, VL3c, VL variant humanization sequence, VL3d, VL variant humanization sequence, VL1A, VL variant humanization sequence, VL1B, variant humanization sequence VL, VL2A, VL variant humanizing sequence, VL2B, VL variant humanizing sequence, VL1Cmodif1, VL variant hyping sequence, VL1Cmodif2, VL variant hyping sequence, VL1Cmodif3, VL variant hyping sequence, VL1Am odif1, VL variant husequence, VL1Amodif2, VL variant husequence, VL1Amodif3, VL variant husequence, VL2C, VL variant humanizationsequence, VL2D, VL variant humanizationsequence, VL2F, VH variant humanizationsequence, VH1A, VH variant humanizationsequence, VH1B, VH variant humanizationsequence, VH2A, the humanization sequence of the VH variant, VH2B, the humanization sequence of the VH variant, VH6Bmodif1, humanized anti-CD3 antibody 4B4.

SEQ ID NO: 176 shows the amino acid sequence of the VH variant, VH6Bmodif2, a humanized anti-CD3 antibody 4B4.

SEQ ID NO: 177 shows the amino acid sequence of the VH variant, VH6Amodif1, a humanized anti-CD3 antibody 4B4.

SEQ ID NO: 178 shows the amino acid sequence of the VH variant, VH6Amodif2, a humanized anti-CD3 antibody 4B4.

SEQ ID NO: 179 shows the amino acid sequence of the VH variant, VH6Amodif3, a humanized anti-CD3 antibody 4B4.

SEQ ID NO: 180 shows the amino acid sequence of the VH variant, VH6C, of the humanized anti-CD3 antibody 4B4.

SEQ ID NO: 181 shows the amino acid sequence of the VH variant, VH6D, of the humanized anti-CD3 antibody 4B4.

SEQ ID NO: 182 shows the amino acid sequence of the VL variant, D7-VK3mut, a humanized anti-CD3 antibody 4B4.

SEQ ID NO: 183 shows the amino acid sequence of the VH variant, D7-VH1mut, of the humanized anti-CD3 antibody 4B4.

SEQ ID NO: 184 shows the amino acid sequence of the CDR1-L variants VL1B, VL2B, VL1Cmodif3 and VL2F of the humanized anti-CD3 antibody 4B4.

SEQ ID NO: 185 shows the amino acid sequence of the full length human CD3y protein, including the signal peptide, available from the Uniprot database under accession number P09693.

SEQ ID NO: 186 shows the amino acid sequence of the Fab heavy chain from an anti-CD3 antibody called 20G6-F3.

SEQ ID NO: 187 shows the light chain amino acid sequence of Fab from an anti-CD3 antibody called 20G6-F3.

SEQ ID NO: 188 shows the amino acid sequence of the Fab heavy chain from an anti-CD3 antibody called 4E7-C9.

SEQ ID NO: 189 shows the light chain amino acid sequence of Fab from an anti-CD3 antibody called 4E7-C9.

SEQ ID NO: 190 shows the amino acid sequence of the Fab heavy chain from an anti-CD3 antibody called 4B4-D7.

SEQ ID NO: 191 shows the amino acid sequence of the Fab light chain from an anti-CD3 antibody called 4B4-D7.

SEQ ID NO: 192 shows the amino acid sequence of the Fab heavy chain from an anti-CD3 antibody called 18F5-H10.

SEQ ID NO: 193 shows the light chain amino acid sequence of Fab from an anti-CD3 antibody called 18F5-H10.

SEQ ID NO: 194 shows the amino acid sequence of the full-length human CD123 protein, including the signal peptide, available from the NCBI database under accession number NP_002174.1 and from the Uniprot database under accession number P26951.

SEQ ID NO: 195 shows the amino acid sequence of the full-length Macac fascicularis CD123 protein, including the signal peptide, available from the GenBank database under EH61867.1 and the Uniprot database under G8F3K3.

SEQ ID NO: 196 shows the amino acid sequence of a fusion containing Fc and mature human CD123 labeled Strep-II, which contains amino acids 22-305 of the full-length human CD123 protein (SEQ ID NO: 194).

SEQ ID NO: 197 shows the amino acid sequence of a fusion containing Fc and mature Macac fascicularis CD123 tagged Strep-II, which contains amino acids 22-305 of the full-length Macac fascicularis CD123 protein (SEQ ID NO: 195).

Under SEQ ID NO: 198 shows the amino acid sequence of the heavy chain variable domain of the antibody to CD123, called 1E1-G5.

Under SEQ ID NO: 199 shows the amino acid sequences of the CDR1-H antibodies to CD123, called 1E1-G5 and 8B11-B7.

Under SEQ ID NO: 200 shows the amino acid sequences of the CDR2-H antibodies to CD123, called 1E1-G5 and 6D6-B8.

SEQ ID NO: 201 shows the amino acid sequences of the CDR3-H of the anti-CD123 antibodies referred to as 1E1-G5, 6D6-B8, 8B11-B7 and 9F6-G3.

Under SEQ ID NO: 202 shows the amino acid sequence of the variable domain of the light chain of the antibody to CD123, called 1E1-G5.

Shown under SEQ ID NOs: 203 and 204 are the amino acid sequences of CDR1-L and CDR3-L of an anti-CD123 antibody called 1E1-G5.

Under SEQ ID NO: 205 shows the amino acid sequence of the variable domain of severe

- 81 041572 chain of an anti-CD123 antibody called 2B8-F3.

Under SEQ ID NO: 206, 207 and 208 shows the amino acid sequences of CDR1-H, CDR2-H and

CDR3-H of an anti-CD123 antibody called 2B8-F3.

SEQ ID NO: 209 shows the amino acid sequence of the light chain variable domain of an anti-CD123 antibody called 2B8-F3.

Shown under SEQ ID NOs: 210 and 211 are the amino acid sequences of CDR1-L and CDR3-L of an anti-CD123 antibody called 2B8-F3.

Under SEQ ID NO: 212 shows the amino acid sequence of the heavy chain variable domain of the anti-CD123 antibody, called 2F8-D6.

Shown under SEQ ID NOs: 213, 214 and 215 are the amino acid sequences of CDR1-H, CDR2-H and CDR3-H of an anti-CD123 antibody called 2F8-D6.

Under SEQ ID NO: 216 shows the amino acid sequence of the variable domain of the light chain of the antibody to CD123, called 2F8-D6.

Shown under SEQ ID NOs: 217 and 218 are the amino acid sequences of CDR1-L and CDR3-L of an anti-CD123 antibody called 2F8-D6.

Under SEQ ID NO: 219 shows the amino acid sequence of the heavy chain variable domain of the antibody to CD123, called 3B10-E6.

Shown under SEQ ID NOs: 220, 221 and 222 are the amino acid sequences of CDR1-H, CDR2-H and CDR3-H of an anti-CD123 antibody called 3B10-E6.

Under SEQ ID NO: 223 shows the amino acid sequence of the light chain variable domain of the anti-CD123 antibody, called 3B10-E6.

Shown under SEQ ID NOs: 224 and 225 are the amino acid sequences of the CDR1-L and CDR3-L of an anti-CD123 antibody called 3B10-E6.

Under SEQ ID NO: 226 shows the amino acid sequence of the heavy chain variable domain of the anti-CD123 antibody, called 3E3-D3.

Shown under SEQ ID NOs: 227, 228 and 229 are the amino acid sequences of CDR1-H, CDR2-H and CDR3-H of an anti-CD123 antibody called 3E3-D3.

SEQ ID NO: 230 shows the amino acid sequence of the light chain variable domain of an anti-CD123 antibody called 3E3-D3.

Shown under SEQ ID NOs: 231 and 232 are the amino acid sequences of CDR1-L and CDR3-L of an anti-CD123 antibody called 3E3-D3.

Under SEQ ID NO: 233 shows the amino acid sequence of the heavy chain variable domain of the anti-CD123 antibody, called 5A5-B4.

Shown under SEQ ID NOs: 234, 235 and 236 are the amino acid sequences of CDR1-H, CDR2-H and CDR3-H of an anti-CD123 antibody called 5A5-B4.

Under SEQ ID NO: 237 shows the amino acid sequence of the variable domain of the light chain of the antibody to CD123, called 5A5-B4.

Shown under SEQ ID NOs: 238 and 239 are the amino acid sequences of CDR1-L and CDR3-L of an anti-CD123 antibody called 5A5-B4.

Under SEQ ID NO: 240 shows the amino acid sequence of the heavy chain variable domain of the anti-CD123 antibody, called 6B10-E4.

Shown under SEQ ID NOs: 241, 242 and 243 are the amino acid sequences of CDR1-H, CDR2-H and CDR3-H of an anti-CD123 antibody called 6B10-E4.

Under SEQ ID NO: 244 shows the amino acid sequence of the light chain variable domain of the anti-CD123 antibody, called 6B10-E4.

Shown under SEQ ID NOs: 245 and 246 are the amino acid sequences of the CDR1-L and CDR3-L of an anti-CD123 antibody called 6B10-E4.

Under SEQ ID NO: 247 shows the amino acid sequence of the heavy chain variable domain of the antibody to CD123, called 6C10-C4.

Shown under SEQ ID NOs: 248, 249 and 250 are the amino acid sequences of CDR1-H, CDR2-H and CDR3-H of an anti-CD123 antibody called 6C10-C4.

SEQ ID NO: 251 shows the amino acid sequence of the light chain variable domain of the anti-CD123 antibodies referred to as 6C10-C4 and 9B8-G6.

Under SEQ ID NO: 252 shows the amino acid sequence of the CDR1-L antibodies to CD123, called 6C10-C4 and 9B8-G6.

Under SEQ ID NO: 253 shows the amino acid sequence of the CDR3-L antibodies to CD123, called 6C10-C4, 9B8-G6 and 9D7-G3.

Under SEQ ID NO: 254 shows the amino acid sequence of the heavy chain variable domain of the anti-CD123 antibody, called 6D6-B8.

SEQ ID NO: 255 shows the amino acid sequence of the CDR1-H of an anti-CD123 antibody called 6D6-B8.

Under SEQ ID NO: 256 shows the amino acid sequence of the variable domain of light

- 82 041572 chain of an anti-CD123 antibody called 6D6-B8.

Shown under SEQ ID NOs: 257 and 258 are the amino acid sequences of CDR1-L and CDR3-L of an anti-CD123 antibody called 6D6-B8.

Under SEQ ID NO: 259 shows the amino acid sequence of the heavy chain variable domain of the anti-CD123 antibody, called 8B11-B7.

Shown under SEQ ID NO: 260 is the amino acid sequence of the CDR2-H of an anti-CD123 antibody called 8B11-B7.

SEQ ID NO: 261 shows the amino acid sequence of the light chain variable domain of the anti-CD123 antibody, referred to as 8B11-B7.

Shown under SEQ ID NOs: 262 and 263 are the amino acid sequences of the CDR1-L and CDR3-L of an anti-CD123 antibody called 8B11-B7.

Under SEQ ID NO: 264 shows the amino acid sequence of the heavy chain variable domain of the anti-CD123 antibody, called 9B8-G6.

Shown under SEQ ID NOs: 25 and 266 are the amino acid sequences of the CDR1-H and CDR2-H of the anti-CD123 antibodies referred to as 9B8-G6 and 9D7-C8.

SEQ ID NO: 267 shows the amino acid sequences of the CDR3-H of an anti-CD123 antibody called 9B8-G6.

Under SEQ ID NO: 268 shows the amino acid sequence of the heavy chain variable domain of the anti-CD123 antibody, called 9D7-C8.

SEQ ID NO: 269 shows the amino acid sequences of the CDR3-H of an anti-CD123 antibody called 9D7-C8.

SEQ ID NO: 270 shows the amino acid sequence of the light chain variable domain of an anti-CD123 antibody called 9D7-C8.

SEQ ID NO: 271 shows the amino acid sequences of the CDR1-L of an anti-CD123 antibody called 9D7-C8.

Under SEQ ID NO: 272 shows the amino acid sequence of the heavy chain variable domain of the anti-CD123 antibody, called 9F6-G3.

Shown under SEQ ID NOs: 273 and 274 are the amino acid sequences of the CDR1-H and CDR2-H of an anti-CD123 antibody called 9F6-G3.

SEQ ID NO: 275 shows the amino acid sequence of the light chain variable domain of an anti-CD123 antibody called 9F6-G3.

SEQ ID NO: 276 shows the amino acid sequence of the CDR1-L of an anti-CD123 antibody called 9F6-G3.

SEQ ID NO: 277 shows the amino acid sequence of the VH variant, VH_G45A, of the 3E3 anti-CD123 antibody.

SEQ ID NO: 278 shows the amino acid sequence of the VH variant, VHmDG, of the 3E3 anti-CD123 antibody.

SEQ ID NO: 279 shows the amino acid sequence of the CDR2-H of the VHmDG variant of the anti-CD123 antibody 3E3 and the CDR2-H of the VH1Fm2DG variant of the humanized anti-CD123 antibody 3E3.

SEQ ID NO: 280 shows the amino acid sequence of the VH variant, VH1A, of the humanized anti-CD123 antibody 3E3.

SEQ ID NO: 281 shows the amino acid sequence of the VH variant, VH1B, of the humanized anti-CD123 antibody 3E3.

SEQ ID NO: 282 shows the amino acid sequence of the VH variant, VH1C, of the humanized anti-CD123 3E3 antibody.

SEQ ID NO: 283 shows the amino acid sequence of the VH variant, VH1D, of the humanized anti-CD123 antibody 3E3.

SEQ ID NO: 284 shows the amino acid sequence of the VH variant, VH1E, of the humanized anti-CD123 antibody 3E3.

Under SEQ ID NO: 285 shows the amino acid sequence of the variant VH, VH1F, humanized antibodies to CD123 3E3.

SEQ ID NO: 286 shows the amino acid sequence of the VH variant, VH1G, of the humanized anti-CD123 3E3 antibody.

SEQ ID NO: 287 shows the amino acid sequence of the VH variant, VH1Fm1, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 288 shows the amino acid sequence of the VH variant, VH1Fm2, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 289 shows the amino acid sequence of the VH variant, VH1Fm2DG, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 290 shows the amino acid sequence of the VH variant, VH1Dm1, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 291 shows the amino acid sequence of the VH variant, VH1Em1, of the humanized anti-CD123 antibody 3E3.

SEQ ID NO: 292 shows the amino acid sequence of the VL variant, VL1A, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 293 shows the amino acid sequence of the VL variant, VL1B, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 294 shows the amino acid sequence of the VL variant, VL1C, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 295 shows the amino acid sequence of the VL variant, VL1D, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 296 shows the amino acid sequence of the VL variant, VL1E, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 297 shows the amino acid sequence of the VL variant, VL1F, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 298 shows the amino acid sequence of the VL variant, VL1G, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 299 shows the amino acid sequence of the VL variant, VL1Fm1, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 300 shows the amino acid sequence of the VL variant, VL1Fm2, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 301 shows the amino acid sequence of the VH variant, VH2A, of the humanized anti-CD123 antibody 3E3.

SEQ ID NO: 302 shows the amino acid sequence of the VH variant, VH3A, of the humanized anti-CD123 antibody 3E3.

SEQ ID NO: 303 shows the amino acid sequence of the VL variant, VL2A, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 304 shows the amino acid sequence of the VL variant, VL2Am1, a humanized anti-CD123 3E3 antibody.

SEQ ID NO: 305 shows the amino acid sequence of the VL variant, VL2Am2, a humanized anti-CD123 3E3 antibody.

Under SEQ ID NO: 306 shows the amino acid sequence of the polypeptide according to formula I from an antibody-like binding protein called CODV-Fab 7G3x20G6.

Under SEQ ID NO: 307 shows the amino acid sequence of the L1 linker from the antibody-like binding proteins called CODV-Fab 7G3x20G6, 7G3x4E7, 7G3x4B4, 7G3x18F5, hz20G6x7G3, 7G3xhz4B4, hz4B4x3E3 and CODV-Fab hz20G6x7G3-TL4.

Under SEQ ID NO: 308 shows the amino acid sequence of the 7G3 light chain variable domain, which is the V _D1 or V _D2 domain of the antibody-like binding proteins called CODV-Fab 7G3x20G6, 7G3x4E7, 7G3x4B4, 7G3x18F5, hz20G6x7G3, 7G3xhz4B4 and CODV-Fab hz20 .

Under SEQ ID NO: 309 shows the amino acid sequence of the L2 linker from the antibody-like binding proteins called CODV-Fab 7G3x20G6, 7G3x4E7, 7G3x4B4, 7G3x18F5, hz20G6x7G3, 7G3xhz4B4, hz4B4x3E3 and CODV-Fab hz20G6x7G3-TL4.

Under SEQ ID NO: 310 shows the amino acid sequence of CL from antibody-like binding proteins called CODV-Fab 7G3x20G6, 7G3x4E7, 7G3x4B4, 7G3x18F5, hz20G6x7G3, 7G3xhz4B4 and hz4B4x3E3.

Under SEQ ID NO: 311 shows the amino acid sequence of the polypeptide according to formula II from an antibody-like binding protein called CODV-Fab 7G3x20G6.

SEQ ID NO: 312 shows the amino acid sequence of the 7G3 heavy chain variable domain, which is the V _H1 or VH2 domain of the antibody-like binding proteins called CODV-Fab 7G3x20G6, 7G3x4E7, 7G3x4B4, 7G3x18F5, hz20G6x7G3, 7G3xhz4B4 and CODV-Fab hz20.

Under SEQ ID NO: 313 shows the amino acid sequence of C _H1 from antibody-like binding proteins called CODV-Fab 7G3x20G6, 7G3x4E7, 7G3x4B4, 7G3x18F5, hz20G6x7G3, 7G3xhz4B4, hz4B4x3E3.

Under SEQ ID NO: 314 shows the amino acid sequence of the polypeptide according to formula I from an antibody-like binding protein called CODV-Fab 7G3x4E7.

Under SEQ ID NO: 315 shows the amino acid sequence of the polypeptide according to formula II from an antibody-like binding protein called CODV-Fab 7G3x4E7.

Under SEQ ID NO: 316 shows the amino acid sequence of the polypeptide according to formula I from an antibody-like binding protein called CODV-Fab 7G3x4B4.

Under SEQ ID NO: 317 shows the amino acid sequence of the polypeptide according to the formula

- 84 041572

II from an antibody-like binding protein called CODV-Fab 7G3x4B4.

Under SEQ ID NO: 318 shows the amino acid sequence of the polypeptide according to the formula

I from an antibody-like binding protein called CODV-Fab 7G3x18F5.

Under SEQ ID NO: 319 shows the amino acid sequence of the polypeptide according to the formula

II from an antibody-like binding protein called CODV-Fab 7G3x18F5.

Under SEQ ID NO: 320 shows the amino acid sequence of the polypeptide according to formula I from an antibody-like binding protein called CODV-Fab hz20G6x7G3.

Under SEQ ID NO: 321 shows the amino acid sequence of the polypeptide according to formula II from an antibody-like binding protein called CODV-Fab hz20G6x7G3.

Under SEQ ID NO: 322 shows the amino acid sequence of the polypeptide according to formula I from an antibody-like binding protein called CODV-Fab 7G3xhz4B4.

Under SEQ ID NO: 323 shows the amino acid sequence of the polypeptide according to formula II from an antibody-like binding protein called CODV-Fab 7G3xhz4B4.

Under SEQ ID NO: 324 shows the amino acid sequence of the polypeptide according to formula I from an antibody-like binding protein called CODV-Fab hz4B4x3E3.

Under SEQ ID NO: 325 shows the amino acid sequence of the polypeptide according to formula II from an antibody-like binding protein called CODV-Fab hz4B4x3E3.

Under SEQ ID NO: 326 shows the amino acid sequence of the polypeptide according to formula I from an antibody-like binding protein called CODV-Fab hz20G6x7G3-TL4.

SEQ ID NO: 327 shows the amino acid sequence of F _c2 from an antibody-like binding protein called CODV-Fab hz20G6x7G3-TL4.

Under SEQ ID NO: 328 shows the amino acid sequence of the polypeptide according to formula III from an antibody-like binding protein called CODV-Fab hz20G6x7G3-TL4.

SEQ ID NO: 329 shows the amino acid sequence of CH1 from an antibody-like binding protein called CODV-Fab hz20G6x7G3-TL4.

Under SEQ ID NO: 330 shows the amino acid sequence of F _c from an antibody-like binding protein called CODV-Fab hz20G6x7G3-TL4.

Under SEQ ID NO: 331 shows the consensus sequence of the CDR1-H antibodies to CD3, called 20G6-F3, 4B4-D7, 4E7-C9, 18F5-H10, 11D7-C3, 11H3-E5, 13H2-C2, 13C1-F6, 1E6C9 , 10F4-C10, 10E6-G6, 18G9-H11, 11F3-B9, 12G3-E8, 5B1-G2, 16F8-A7, 11F9-F8, 20E5F10, 20B5-F10, 3H6-D2, based on sequence alignment.

Under SEQ ID NO: 332 shows the consensus sequence of the CDR2-H antibodies to CD3, called 20G6-F3, 4B4-D7, 4E7-C9, 18F5-H10, 11D7-C3, 11H3-E5, 13H2-C2, 13C1-F6, 1E6C9 , 10F4-C10, 10E6-G6, 18G9-H11, 11F3-B9, 12G3-E8, 5B1-G2, 16F8-A7, 11F9-F8, 20E5F10, 20B5-F10, 3H6-D2, based on sequence alignment.

Under SEQ ID NO: 333 shows the consensus sequence of the CDR3-H antibodies to CD3, called 20G6-F3, 4B4-D7, 4E7-C9, 18F5-H10, 11D7-C3, 11H3-E5, 13H2-C2, 13C1-F6, 1E6C9 , 10F4-C10, 10E6-G6, 18G9-H11, 11F3-B9, 12G3-E8, 5B1-G2, 16F8-A7, 11F9-F8, 20E5F10, 20B5-F10, 3H6-D2, based on sequence alignment.

Under SEQ ID NO: 334 shows the consensus sequence of the CDR1-L antibodies to CD3, called 20G6-F3, 4B4-D7, 4E7-C9, 18F5-H10, 11D7-C3, 11H3-E5, 13H2-C2, 13C1-F6, 1E6C9 , 10F4-C10, 10E6-G6, 18G9-H11, 11F3-B9, 12G3-E8, 5B1-G2, 16F8-A7, 11F9-F8, 20E5-F10, 20B5-F10, 3H6-D2, based on sequence alignment.

Under SEQ ID NO: 335 shows the consensus sequence of the CDR3-L antibodies to CD3, called 20G6-F3, 4B4-D7, 4E7-C9, 18F5-H10, 11D7-C3, 11H3-E5, 13H2-C2, 13C1-F6, 1E6C9 , 10F4-C10, 10E6-G6, 18G9-H11, 11F3-B9, 12G3-E8, 5B1-G2, 16F8-A7, 11F9-F8, 20E5F10, 20B5-F10, 3H6-D2, based on sequence alignment.

Under SEQ ID NO: 336 shows the consensus sequence of the CDR1-H antibodies to CD123, called 1E1-G5, 6D6-B8, 8B11-B7, 9F6-G3, based on sequence alignment.

Under SEQ ID NO: 337 shows the consensus sequence of the CDR2-H antibodies to CD123, called 1E1-G5, 6D6-B8, 8B11-B7, 9F6-G3, based on sequence alignment.

Under SEQ ID NO: 338 shows the consensus sequence of the CDR1-L antibodies to CD123, called 1E1-G5, 6D6-B8, 8B11-B7, 9F6-G3, based on sequence alignment.

Under SEQ ID NO: 339 shows the consensus sequence of the CDR3-L antibodies to CD123, called 1E1-G5, 6D6-B8, 8B11-B7, 9F6-G3, based on sequence alignment.

Under SEQ ID NO: 340 shows the consensus sequence of the CDR1-H antibodies to CD123, called 6C10-C4, 9B8-G6, 9D7-C8, based on sequence alignment.

Under SEQ ID NO: 341 shows the consensus sequence of the CDR2-H antibodies to CD123, called 1E1-G5, 6D6-B8, 8B11-B7, 9F6-G3, based on sequence alignment.

Under SEQ ID NO: 342 shows the consensus sequence of the CDR3-H antibodies to CD123, called 1E1-G5, 6D6-B8, 8B11-B7, 9F6-G3, based on sequence alignment.

- 85 041572

SEQ ID NO: 343 shows the CDR1-L consensus sequence of anti-CD123, 1E1G5, 6D6-B8, 8B11-B7, 9F6-G3 antibodies based on sequence alignment.

Under SEQ ID NO: 344 shows the amino acid sequence of the linker sequence (Gly-Gly-Gly-Gly-Ser).

Under SEQ ID NO: 345 shows the amino acid sequence of the linker sequence (Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser).

Under SEQ ID NO: 346 shows the amino acid sequence of the linker sequence (Thr-Val-Ala-Ala-Pro).

Under SEQ ID NO: 347 shows the amino acid sequence of the linker sequence (Gln-Pro-Lys-Ala-Ala).

Under SEQ ID NO: 348 shows the amino acid sequence of the linker sequence (Gln-Arg-Ile-Glu-Gly).

Under SEQ ID NO: 349 shows the amino acid sequence of the linker sequence (Ala-Ser-Thr-Lys-Gly-Pro-Ser).

Under SEQ ID NO: 350 shows the amino acid sequence of the linker sequence (Ala-Ser-Thr-Lys-Gly-Pro-Ser).

Under SEQ ID NO: 351 shows the amino acid sequence of the linker sequence (His-Ile-Asp-Ser-Pro-Asn-Lys).

SEQ ID NO: 352 shows the amino acid sequence of a linker sequence and a His tag sequence added at the C-terminus of a polypeptide according to formula II from antibody-like binding proteins called CODV-Fab 7G3x20G6, 7G3x4E7, 7G3x4B4, 7G3x18F5, hz20G6x7G3, 7G3xhz4B4 and hz4B4x3, corresponding to the sequences hinge and His mark, used for cleaning, for example.

SEQ ID NO: 353 shows the amino acid sequence of the CDR2-H of an anti-CD123 antibody variant called 3E3.

Under SEQ ID NO: 354 shows the amino acid sequence of the linker sequence (Gly-Gly-Gly-Ser).

Under SEQ ID NO: 355 shows the amino acid sequence of the linker sequence (Ser-Gly-Gly-Gly-Ser).

Under SEQ ID NO: 356 shows the amino acid sequence of the linker sequence (Gly-Ser-Gly-Gly-Gly-Gly-Ser).

Under SEQ ID NO: 357 shows the amino acid sequence of the linker sequence (Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser).

Under SEQ ID NO: 358 shows the amino acid sequence of the linker sequence (Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser).

Under SEQ ID NO: 359 shows the amino acid sequence of the linker sequence (Lys-Thr-His-Thr).

Under SEQ ID NO: 360 shows the amino acid sequence of the linker sequence (Lys-Thr-His-Thr-Ser).

Under SEQ ID NO: 361 shows the amino acid sequence of the linker sequence (Asp-Lys-Thr-His-Thr-Ser).

Under SEQ ID NO: 362 shows the amino acid sequence of the linker sequence (Asp-Lys-Thr-His-Thr-Ser-Pro).

Under SEQ ID NO: 363 shows the amino acid sequence of the linker sequence (Ser-Asp-Lys-Thr-His-Thr-Ser-Pro).

Under SEQ ID NO: 364 shows the amino acid sequence of the linker sequence (Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro).

Under SEQ ID NO: 365 shows the amino acid sequence of the linker sequence (Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser).

Under SEQ ID NO: 366 shows the amino acid sequence of the linker sequence (Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser).

Under SEQ ID NO: 367 shows the amino acid sequence of the linker sequence (Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro).

Under SEQ ID NO: 368 shows the amino acid sequence of the linker sequence (Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro).

Under SEQ ID NO: 369 shows the amino acid sequence of the linker sequence (Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-Gly).

Under SEQ ID NO: 370 shows the amino acid sequence of the linker sequence (Gly-Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-Gly).

Under SEQ ID NO: 371 shows the amino acid sequence of the linker sequence (Gly-Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-Gly-Gly).

- 86 041572

Under SEQ ID NO: 372 shows the amino acid sequence of the linker sequence (Gly-Gly-Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-Gly-Gly).

Under SEQ ID NO: 373 shows the amino acid sequence of the linker sequence (Gly-Gly-Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-Gly-Gly-Gly) .

Under SEQ ID NO: 374 shows the amino acid sequence of the linker sequence (Gly-Gly-Gly-Glu-Pro-Lys-Ser-Asp-Lys-Thr-His-Thr-Ser-Pro-Pro-Ser-Pro-Gly-Gly- gly).

Under SEQ ID NO: 375, 376 and 377 shows the amino acid sequences of CDR1-H, CDR2-H and CDR3-H of an antibody called 7G3.

Under SEQ ID NO: 378 and 379 shows the amino acid sequence of CDR1-L and CDR3-L of an antibody called 7G3.

SEQ ID NO: 380 shows the amino acid sequence of a humanized antibody heavy chain variable domain variant called 7G3.

Shown under SEQ ID NOs: 381 and 382 are the amino acid sequences of CDR1-H and CDR3-H of a humanized antibody called 7G3.

Under SEQ ID NO: 383 shows the amino acid sequence of an additional variant heavy chain variable domain of the humanized antibody, called 7G3.

SEQ ID NO: 384 shows the amino acid sequences of the CDR2-H of one of the humanized antibodies called 7G3.

Under SEQ ID NO: 385 shows the amino acid sequence of the light chain variable domain of the humanized antibody, called 7G3.

Shown under SEQ ID NO: 386 is the amino acid sequence of SEQ ID NO: 1 as shown in WO 2015026892.

Shown under SEQ ID NO: 387 is the amino acid sequence of SEQ ID NO: 3 as shown in WO 2015026892.

Under SEQ ID NO: 388 shows the amino acid sequence of the polypeptide according to formula I of the antibody-like binding proteins called Fab hz20G6xhz7G3, CODV-Fab-OL1 hz20G6xhz7G3 and CODV-Fab-OL1a hz20G6xhz7G3.

Under SEQ ID NO: 389 shows the amino acid sequence of the linker sequence (Gly-Gly-Ser-Gly-Ser-Ser-Gly-Ser-Gly-Gly).

Under SEQ ID NO: 390 shows the amino acid sequence of the polypeptide according to formula II from an antibody-like binding protein called CODV-Fab hz20G6xhz7G3.

Under SEQ ID NO: 391 shows the amino acid sequence of the polypeptide according to formula IV from an antibody-like binding protein called CODV-Fab-TL1 hz20G6xhz7G3.

SEQ ID NO: 392 shows the amino acid sequence of the F _c2 region from an antibody-like binding protein called CODV-Fab-TL1 hz20G6xhz7G3.

Under SEQ ID NO: 393 shows the amino acid sequence of the polypeptide according to formula IV from an antibody-like binding protein called CODV-Fab-TL1 hz20G6xhz7G3.

SEQ ID NO: 394 shows the amino acid sequence of the Fc region from an antibody-like binding protein called CODV-Fab-TL1 hz20G6xhz7G3. Under SEQ ID NO: 395 shows the amino acid sequence of the polypeptide according to formula II from an antibody-like binding protein called CODV-Fab-OL1 hz20G6xhz7G3.

SEQ ID NO: 396 shows the amino acid sequence of the Fc region from an antibody-like binding protein called CODV-Fab-OL1 hz20G6xhz7G3.

SEQ ID NO: 397 shows the amino acid sequence of Fc-stump (Fc ₃ ) from an antibody-like binding protein called CODV-Fab-OL1 hz20G6xhz7G3.

SEQ ID NO: 398 shows the amino acid sequence of Fc-stump (Fc ₃ ) from an antibody-like binding protein called CODV-Fab-OL1a hz20G6xhz7G3.

SEQ ID NO: 399 shows the amino acid sequence of a polypeptide according to formula II from an antibody-like binding protein called CODV-Fab-OL1a hz20G6xhz7G3.

SEQ ID NO: 400 shows the amino acid sequence of Fc-stump (Fc3) from an antibody-like binding protein called CODV-Fab-OL1a hz20G6xhz7G3.

figures

Fig. 1 - sequence alignments of the VH regions of antibodies to CD3, called 20G6-F3, 4B4-D7, 4E7-C9, 18F5-H10, 11D7-C3, 11H3-E5, 13H2-C2, 13C1-F6, 1E6-C9, 10F4-C10 , 10E6-G6, 18G9-H11, 11F3-B9, 12G3-E8, 5B1-G2, 16F8-A7, 11F9-F8, 20E5-F10, 20B5-F10, 3H6-D2.

Fig. 2 - sequence alignments of the VL regions of antibodies to CD3, called 20G6-F3, 4B4-D7, 4E7-C9, 18F5-H10, 11D7-C3, 11H3-E5, 13H2-C2, 13C1-F6, 1E6-C9, 10F4-C10 , 10E6-G6, 18G9-H11, 11F3-B9, 12G3-E8, 5B1-G2, 16F8-A7, 11F9-F8, 20E5-F10, 20B5-F10, 3H6-D2.

Fig. 3 - Sequence alignments of the VH and VL regions of the anti-CD123 antibodies referred to as 1E1- 87 041572

G5, 6D6-B8, 8B11-B7, 9F6-G3.

Fig. 4 - Sequence alignments of the VH and VL regions of antibodies to CD123, called

6C10-C4, 9B8-G6, 9D7-C8.

Fig. 5 and 7 - IV administration of fully human CODV-Fab-TL1 hz20G6xhz7G3, every 3 days, in the presence of human T-cells suppresses growth of Molm13 tumor throughout the body at all tested doses.

Fig. 6 and 8 - IV administration of fully human CODV-Fab-TL1 hz20G6xhz7G3 every 3 days in the presence of human T cells is associated with tumor regression in long bones at all tested doses.

Fig. 9 and 11 IV administration of the fully human CODV-Fab hz20G6xhz7G3 in the presence of human T cells inhibits tumor growth.

Fig. 10 and 12 - IV administration of the fully human CODV-Fab hz20G6xhz7G3 in the presence of human T cells is associated with tumor regression in long bones.

Fig. 13 and 15 - CIP administration of fully human CODV-Fab hz20G6xhz7G3 in the presence of human T cells inhibits tumor growth throughout the body at a dosage of 0.13 nmol/kg/day or higher.

Fig. 14 and 16 - CIP administration of fully human CODV-Fab hz20G6xhz7G3 in the presence of human T cells inhibits tumor growth in long bones at a dosage of 0.13 nmol/kg/day or higher.

Fig. 17 is a schematic representation of the structure of CODV-Ig, CODV-Fab-TL, and CODV-Fab-OL (further showing LALA mutations (if Fc from the IgG1 backbone is used) and lip-to-bottom mutations).

Examples

Example 1. Obtaining antibodies.

1.1. Construction of an expression plasmid for the hCD3ε/δ-hFc fusion (CD3ed-Fc).

Using cDNA-containing plasmids as a template, fusion proteins were generated containing CD3ε and CDδ of human and Macaca fascicularis, as detailed herein below, in frame with the heavy chain constant region, including the hinge region, CH2 and CH3 domains of human IgG immunoglobulin. additionally labeled with 8 x His or Strep-II for optional tandem cleaning.

Using human genomic DNA as a template, the extracellular domains of the human CD3ε and human CDδ subunits, including the signal sequence, were amplified. The resulting amplified, digested and purified PCR products were pooled by ligation PCR and ligated into the mammalian expression vector pXL by InFusion using the NheI and HindIII sites. Each subunit was cloned into one plasmid. The sequence of the resulting fusion protein containing Fc and His-tagged mature human CD3ε is disclosed herein under SEQ ID NO:3. disclosed herein under SEQ ID NO: 1, available from the Uniprot database under accession number P07766) and thus the extracellular domain of human CD3ε.

Using cynomolgus monkey genomic DNA as a template, the extracellular domains of CD3ε and CD3δ of Macac fascicularis were amplified, including the signal sequence. The resulting amplified, digested and purified PCR products were pooled by ligation PCR and ligated into the pXL mammalian expression vector using the InFusion method using NheI and HindIII. Each subunit was cloned into one plasmid. The resulting fusion protein sequences containing Fc and mature Macac fascicularis CD3ε are disclosed under SEQ ID NO:4. wild-type Macaca fascicularis full-length CD3ε domain (disclosed herein under SEQ ID NO: 2, available from the Uniprot database under accession number Q95LI5). The cloned fusion protein further contains one alanine to valine substitution at amino acid position 35 compared to amino acid position 57 of the wild-type sequence.

1.2. Expression and purification of human and cynomolgus CD3ed-Fc.

Freestyle HEK293 cells grown as a suspension culture in F17 serum-free medium (Life) were transiently transfected with the expression plasmid. Co-transfection with both plasmids representing the extracellular domain (ECD) subunit of CD3ε and CD3δ was performed using the Cellfectin (Life) transfection reagent. Cells were cultured at 37°C for 7 days. The culture supernatant containing the recombinant protein was collected by centrifugation and purified from impurities by filtration (0.22 μm).

For purification, Fc-containing fusion protein variants were captured on a Protein A (GE) matrix and

- 88 041572 was eluted with a pH gradient. After further purification of the protein by size exclusion chromatography (SEC) using Superdex 200 (GE) and a final concentration step by ultrafiltration, the protein was used for further analyses.

The human protein heterodimer was additionally applied to a His-Trap (GE) column after being captured on Protein A and desalted. The eluted protein was loaded onto a streptavidin (GE) column and eluted with d-destiobiotin before final SEC post-purification using Superdex 200 (GE). This strategy was used to separate heterodimers from homodimers.

1.3. Generation of antibodies with cross-reactivity to human CD3/cynomolgus monkey cDNA of human CD3ε and CD3δ and Macaca fascicularis were cloned into proprietary immunization vectors from Aldevron (pB8 and VV8, respectively) and used for gene immunization of rats. Rats in the MR12-266 (CD3-cyno) immunization group were first immunized with human CD3ε and CD3δ cDNA, followed by two additional immunizations with a mixture of human CD3ε and CD3δ cDNA and Macaca fascicularis. Immune sera were collected on day 24 of the immunization protocol, after 4 applications of gene immunization (IS24d-4). Serum diluted in PBS with 1% BSA was tested by flow cytometry using mammalian cells transiently transfected with the target cDNA in co-transfection experiments to generate TCR complexes from human CD3ε and CD3δ and Macaca fascicularis. In addition, immune sera were tested on the following cell lines: Jurkat E6-1 (expressing human TCR), Jurkat-RT-T3.5 (negative for TCR), and cynomolgus HSC-F (expressing cynomolgus TCR); however, no cell line was available that was negative for the cynomolgus TCR. Goat anti-rat IgG conjugate with R-phycoerythrin (Southern Biotech, no. 3030-09) was used as a secondary antibody at a concentration of 10 μg/ml.

Specific immune sera reactivity, in particular against cells transfected with CD3ε and CD3δ cDNA combinations, could be detected in immunized animals compared to cells transfected with irrelevant cDNA. The same was valid in the test with the TCR-positive Jurkat cell line (E6-1) compared to the TCR-negative cell line (RTT3.5), with significantly weaker signals, but still significant signals, found in the HSC cell line -F of the cynomolgus macaque (see Table 1).

Table 1. Detection of specific reactivity of immune sera against the TCR complex in each animal by flow cytometry. Data are presented as geometric means of relative fluorescence intensity (rfu)

pXL-CD3ehuman+pXLCOZc1-human pXL-CD3emaca as a beetle+pXLC03c1-maca as a beetle pB1_ irrelevant Jurkat- E6-1 Jurkat- RT-T3.5 HSC-F monkey eater without primary antibodies 5 5 4 7 10 6 OKTZ - - - 858 25 - FN18 - - 4 7 10 1256 Antibody mice to ON 6 6 - - - - Antibody rats to tus 9 7 1033 10 14 9 rat No. 1 1:1000 290 273 5 186 36 55 1:5000 79 86 5 65 21 20 rat no. 2 1:1000 499 401 5 390 55 75 1:5000 147 110 4 126 28 25

- 89 041572

rat no. 3 1:1000 239 247 5 108 thirty 31 1:5000 89 92 4 32 17 9 rat no. 4 1:1000 280 283 5 176 37 73 1:5000 97 110 4 55 22 23 rat no. 5 1:1000 208 137 4 99 31 eleven 1:5000 61 65 4 47 19 7

Rats in the MR12-265 (human CD3) immunization group were co-immunized with human CD3b and CD3δ cDNA cloned into their respective expression vectors. Immune sera were collected on day 24 of the immunization protocol, after 4 applications of gene immunization (IS24d-4). Serum diluted in PBS c1% BSA was tested by flow cytometry using mammalian cells transiently transfected with the above target cDNA in co-transfection experiments to generate TCR complexes of human CD3s and CD35 and Macaca fascicularis. Goat anti-rat IgG conjugate with R-phycoerythrin (Southern Biotech, no. 3030-09) was used as a secondary antibody at a concentration of 10 μg/ml. In addition, immune sera were tested on the following cell lines: Jurkat E6-1 (expressing human TCR), Jurkat-RT-T3.5 (negative for TCR), and cynomolgus HSC-F (expressing cynomolgus TCR); however, no cell line was available that was negative for the cynomolgus TCR. Goat anti-rat IgG conjugate with R-phycoerythrin (Southern Biotech, no. 3030-09) was used as a secondary antibody at a concentration of 10 μg/ml.

Specific reactivity of immune sera, in particular against cells transfected with combinations of CD3s and CD3δ cDNA, could be detected in immunized animals when compared with cells transfected with irrelevant cDNA. The same was valid in the test with the TCR-positive Jurkat cell line (E6-1) compared to the TCR-negative cell line (RT-T3.5), but significantly weaker signals (but still significant in positive rats) were found in cynomolgus monkey HSC-F cell lines (Table 2).

Table 2. Detection of specific reactivity of immune sera against the TCR complex in each animal by flow cytometry. Data are presented as geometric means of relative fluorescence intensity (rfu)

pXL-CD3ehuman+pXLCREAT-human pXL-CD3ecrabeater monkey+pXLCREATION-crabeater monkey pB1_ irrelevant Jurkat- E6-1 Jurkat- RT-T3.5 HSC-F monkey eater without primary antibodies 5 5 4 7 10 6 OKTZ - - - 858 25 - FN18 - - 4 7 10 1256 Antibody MOUSE to ON 6 6 - - - - Antibody rats to tus 9 7 1033 10 14 9 rat 1:1000 230 132 5 170 33 50 No. 1 1:5000 78 40 4 54 19 15 rat No. 2 1:1000 216 136 6 216 48 32 1:5000 75 40 4 66 24 eleven

- 90 041572

rat No. 3 1:1000 73 54 5 56 25 7 1:5000 25 20 4 23 15 6 rat No. 4 1:1000 70 56 5 41 25 8 1:5000 25 18 4 15 15 6 rat No. 5 1:1000 65 37 5 42 20 7 1:5000 24 17 4 17 14 6

Sera positive rats were sacrificed and B cells fused with mouse myeloma cells. The resulting hybridomas were screened on HEK293 cells transfected with human or cynomolgus CD3ε and CD3δ expression plasmids, on Jurkat E6.1 (CD3+) and Jurkat T3.5 (CD3-) cells by flow cytometry. The supernatant of the hybridoma clones was used to evaluate the kinetics of the reaction with the human CD3ε/δ complex and Macac fascicularis using the single point method by fixing the analyte at 25 nM using surface plasmon resonance (data are presented in Table 3).

Table 3. Detection of specific reactivity of supernatants of various hybridoma clones by flow cytometry and Biacore analysis. Supernatants were tested on different cell lines (Jurkat E6-1, Jurkat-RT-T3.5 and HSC-F cynomolgus monkey). Biacore analysis was performed on human and cynomolgus CD3e/d complexes, respectively.

Sample ID Biacore for human CD3e/6 [RU] at 25 nM Biacore for Cynomolgus CD3e/6 [RU] at 25 nM 3G5-E10 27 3 18G9-H11 7 -7 12D2-E5 27 5 6S9-S9 26 32 11F9-F8 11.5 16 8H2-F3 32.5 40 4B4-D7 eleven 14 4E7-C9 2.5 10 13C1-F6 17 15.5 11D7-C3 17 10 13H2-C3 16 7 10E6-G6 10.5 0 12G3-E8 24 17 11F3-B9 22 17 20EZ-B5 eleven 5

- 91 041572

10F4-C10 22 27 11НЗ-Е5 7 9 18F5-H10 η. A. η. A. 18H11-F10 35 55 20E5-F10 12 8 Sample ID Biacore for human CD3e/6 [RU] at 2 5 nM Biacore for ΟΌ3ε/δ cynomolgus [RU] at 25 nM 3H6-D2 9.5 7 8C2-F7 4.5 3 5B1-G2 1 0.5 7B7-F10 28 42 20B5-F10 1 9.5 16D3-E4 -2 -2 1E6-B8 η. A. η. A. 16F8-A7 17.5 22.5 11A8-D6 3E8-G1 20G6-F3 9D7-B5 η. A. η. A. 9G5-G10 17A11-D3

Positive clones were expanded and the corresponding heavy and light chain variable domain cDNAs were isolated by RT-PCR. The VH and VL sequences were cloned into expression vectors in fusion with one of the CH1, IGHG1 backbone or human kappa chain to allow expression of Fab fragments as well as complete IgG.

1.4. Expression of IgG and Fab fragments.

Expression plasmids encoding IgG heavy and light chains and Fab fragments were propagated in E. coli NEB 10beta (derived from DH10B). Plasmids used for transfection were obtained from E. coli using the Plasmid Plus kit from QIAGEN (cat no.: 12991).

HEK293-FS cells grown in Freestyle medium (Invitrogen) were transfected with the indicated LC and HC plasmids encoding heavy chains and light chains using the 293fectin transfection reagent (Invitrogen) as described by the manufacturer. Cells were cultured at 37°C in a Kuhner ISF1-X shaker-incubator at 110 rpm. with 8% CO2. After 7 days of cultivation, the cells were removed by centrifugation, 10% v/v was added. 1M Tris-HCl, pH 8.0 and the supernatant were filtered through a 0.2 µm vial neck filter to remove particles. The CODV-IgG1 constructs were purified by affinity chromatography on Protein A columns (HP columns HiTrap Protein A, GE Life Sciences). After eluting from the column with 0.1 M citrate, pH 3.0, the CODV-IgG1 constructs were desalted using HiPrep 26/10 desalting columns, then formulated in PBS (Gibco 14190-136).

The bispecific CODV-Fab constructs were purified using HisTrap high performance columns (GE Healthcare, Cat #: 17-5248-02). After elution from the column (elution buffer: 20 mM sodium phosphate, 0.5 M NaCl, 500 mM imidazole, pH 7.4), protein containing fractions were pooled and desalted using HiPrep 26/10 desalting columns, then compiled in PBS ( Gibco 14190136).

To separate monomers from aggregates, a high resolution fractionation step in PBS (Gibco 14190-136) was performed on both constructs, CODV-IgG and CODV-Fab fragment, using a 320 ml HiLoad Superdex 200 26/60 column (GE Healthcare, cat # .: 29-9893-36). Monomer fractions were pooled and concentrated to 1 mg/mL using a Vivaspin 20 centrifuge column (VS2002, Sartorius Stedim biotech) and filtered using a 0.22 µm membrane (Millex® SLGV033RS syringe filters).

The protein concentration was determined by measuring the absorbance at 280 nm. Each batch was analyzed by SDS-PAGE under reducing and non-reducing conditions to determine the purity and molecular weight of each subunit and monomer.

1.5. Assessment of the affinity of antibodies to CD3.

- 92 041572

1.5.1. Affinity assessment for both human CD3ε/δ and Maca fascicularis.

The binding affinities of CD3 binding Fab or CODV-Fab were measured by surface plasmon resonance (SPR) using the Biacore3000 instrument (GE Healthcare). The assay buffer was HBS-EP (BR-1001-88, GE Healthcare). Capture of CD3ε/δ-Fc fusion proteins was achieved using a human antibody capture kit (GE Healthcare). The capture antibody was bound on CM5 chips (BR-1001-88, GE Healthcare) to about 12,000 RU using an amine coupling kit (BR-100-50, GE Healthcare). CD3εδ-Fc fusion proteins were captured at 10 ml/min up to about 70 RU to give Rmax values of 30 RU. The kinetics of binding to Fab or CODV-Fab to CD3 was measured at 30 μl/min for 240 and 600 s for the association and dissociation phase, respectively. Two-fold dilutions of Fab from 3 to 400 nM in assay buffer were used. All Fab concentrations were run in duplicate with buffer control in duplicate to provide a dual control. Regeneration of the capture surface was provided with a 1 minute injection of a 3M MgCl ₂ solution at a rate of 30 μl/min. BIAevaluation v.4.1 software (GE Healthcare) was used for data analysis. The data were approximated as a whole using a 1:1 Langmuir model with mass transfer.

Measurement of the binding affinity of IgG and CODV-Fc proteins for CD3 was performed similarly to the binding assay for Fab and CODV-Fab except for the capture antibody. In this case, the His Capture Kit (28-9950-56, GE Healthcare) was used to capture human CD3-Fc protein via a His tag. For the CD3-Fc binding assay of Macaca fascicularis, the classical antibody Strep-MAB (2-1507-001, IBA) was used as the capture antibody. In this case, the regeneration solution was 10 mM glycine buffer, pH 2.0.

Table 4. Affinities of selected anti-CD3 antibodies for human CD3ε/δ complexes and Macac fascicularis measured with Biacore

CD3εδ Clone (Fabfragment) ka (1/ms) kd (1/s) Rmax (RU) KD (M) Chi ² Ratio of KD (macaque crab-eater) / KD (human) Human 12D2-E5 5, 4E+ 06 6.2E-03 35 1,2Ε- 09 0.34 4 macaques- crabeater 12D2-E5 equilibrium state equilibrium state 8 4, 8Ε- 09 0.129 Human 10E6-G6 1.4E+ 05 2.0E-03 37 1,4Ε- 08 0.218 1 macaques- crabeater 10E6-G6 1.9E+ 05 2.4E-03 18 1, SE- 08 0.218 Human 4E7-C9 8.9E+ 04 1.6E-03 29 1.7Ε- 08 0.244 1 macaques- crabeater 4E7-C9 9, 8E+ 04 1.5E-03 20 1.5Ε- 08 0.232 Human 12G3-E8 1, 1E+ 05 1.1E-03 thirty 9, SE- 09 0.307 1 macaques- crabeater 12G3-E8 1.7E+ 05 1.8E-03 15 1, 0E- 08 0.349 Human 5B1-G2 6.0E+ 04 3.9E-04 36 6, SE- 09 0.218 1 macaques- crabeater 5B1-G2 3.4E+ 04 1.5E-04 20 4, SE- 09 2.1 Human 18F5- H10 2.0E+ 05 1.8E-04 28 8.9E- 10 0.235 1 macaques- crabeater 18F5- H10 1.4E+ 05 1,ZE-04 15 8.9E- 10 0.224

- 93 041572

Human 18G9- H11 2.5E+ 05 1.1E-03 31 4.4E- 09 0.975 6 macaques- crabeater 18G9- H11 5, 1E+ 04 1.2E-03 18 2.4E- 08 0.617 Human 4B4-D7 1, 1E+ 05 2.9E-03 22 2,ZE- 08 0.147 1 macaques- crabeater 4B4-D7 1.1E+0 5 2.7E-03 12 2.4E- 08 0.188 Human 1E6-B8 9, WE+ 04 3.9E-03 26 4, 1E- 08 0.157 1 macaques- crabeater 1E6-B8 1, 0E+ 05 4.2E-03 15 4.2E- 08 0.122 Human 13H2-C3 1.7E+ 05 4.9E-03 20 2.9E- 08 0.317 1 macaques- crabeater 13H2-C3 1, 8E+ 05 5.9E-03 12 3.2E- 08 0.166 Human 20G6-F3 3.5E+ 04 2.7E-04 21 3, SE- 09 0.208 1 macaques- crabeater 20G6-F3 2.7E+ 04 2.2E-04 15 8.2E- 09 0.184 Human 11F9-F8 7.7E+ 04 4.0E-04 22 5, 2E- 09 0.137 1 macaques- crabeater 11F9-F8 8, WE+ 04 3.6E-04 13 4.4E- 09 0.109 Human 20E5- F10 9, 9E+ 04 4,ZE-04 25 4, SE- 09 0.125 1 macaques- crabeater 20E5- F10 1.2E+ 05 3.4E-04 14 2.9E- 09 0.104 Human SP-34 1, 1E+ 06 3.0E-03 34 2, SE- 09 2, SE- 01 5 macaques- crabeater SP-34 2.2E+ 05 2.8E-03 34 1, SE- 08 4, SE- 01

SP-34 was used as IgG.

1.5.2. Binding of anti-CD3 antibodies to huCD3ε, huCD3δ, and huCD3ε/δ expressed on the surface of HEK293F cells by flow cytometry.

To analyze antibody binding to human CD3ε and human CD35 expressed on the cell surface, HEK293F cells were transfected with both constructs, either alone or co-transfected, and signals were measured by flow cytometry. For the transfection procedure, FuGENE HD transfection reagent (Promega, #E2311) was used according to the manufacturer's protocol.

HEK293F cells were seeded in Freestyle293 medium (Gibco) at a density of 6E6 cells per tube in 50 ml Cellstar Cellreactor tubes with filter (Greiner bio-one). Transfections were performed according to the FuGENE protocol. Complexation was performed in OptiMEM without phenol red (Gibco) at a ratio of 3:1 (protocol for transfection of 293F cells grown in 8000 μl of medium in T-25 flasks using a FuGENE® HD:DNA ratio of 3.0:1, http ://www.promega.com/techserv/tools/FugeneHdTool/default.aspx).

Cells were incubated on a shaker at 37°C and 5% CO ₂ . One to three days after transfection, cells were harvested and antibody binding was analyzed by flow cytometry.

Antibodies for staining were distributed at 1 μg in 50 μl of FBS staining buffer (BD Pharmingen) per well in 96-well U-bottomed suspension culture plates (Greiner bio-one). The harvested transfected cells were resuspended in FBS staining buffer and added to the antibodies at 50 µl per well. Cells were incubated at 4°C in the dark for 30 min and washed twice. 0.5 µg of goat anti-human IgG F(ab')2 secondary antibody FITC conjugated (Beckman Coulter, no. 732598) or goat F(ab')2 anti-human kappa chain conjugated to PE ( Southern Biotech, no. 206209) respectively in combination with 0.5 µg 7AAD in 100 µl FBS staining buffer. Cells were incubated at 4°C in the dark for 15 min and washed twice. For measurement, cells were resuspended in 200 µl of FBS staining buffer. Cells were measured using a MACSQuant (Miltenyi Biotec) or an LSRII flow cytometer.

- 94 041572 (BD) respectively. Additional data analyzes were performed using FlowJo software (Tree Star, Inc.). The results were the percentage of 7-AAD-negative individual cells positive for antibody staining (data are presented in Table 5).

1.5.3. Fab binding to CD3 with huCD3ε/δ and huCD3ε/γ as measured by SPR.

Binding was tested by SPR using the BIAcore3000 instrument used with HBS-EP buffer.

Recombinant huCD3 proteins (ε/δ (PB01226), ε/γ (PB01225)) were captured at 10 μl/min via Fc labeling with the human Fc capture antibody MAB1302 (Millipore) immobilized on a CM5 sensor chip. Anti-CD3 Fab were used as analytes at 100 nM with association and dissociation times of 240 and 300 s, respectively, at 30 μl/min. After each cycle, surfaces were regenerated by bubbling for 2 min with 10 mM glycine buffer, pH 2.5.

When only huCD3δ was expressed on the HEK293F cell surface, no signals could be detected by flow cytometry. In contrast, almost all antibodies could bind to cells transfected exclusively with huCD3c, or when co-transfected with huCD3δ, indicating that huCD3E is required as an epitope. In Biacore assays, binding to huCD3E was shown to be independent of whether the δ or γ chain was used for the recombinant protein, suggesting that huCD3E is sufficient as an antigen. Exceptionally, the 12D2 antibody only binds to huCD3E if a co-chain is present. This may be due to an indirect effect of co-expression of the g or d chain on the conformational structure of the protein required to expose the epitope for this antibody. The same effect was shown for the published OKT3 antibody. This antibody has been described to interact with a conformational epitope formed upon association of huCD3E with huCD3δ or γ, respectively (Salmeron et al., 1991, The Journal of Immunology). It has also been shown to bind exclusively to the huCD3E subunit (Kjer-Nielsen et al., 2004, PNAS). Due to the similar behavior of 12D2 and OKT3, interaction with huCD3E is allowed for 12D2. All things considered, huCD3E appears to be the antigenic structure for all analyzed antibodies (data are presented in Table 5).

Table 5 Affinities of selected CD3 antibodies for huCD3δ, huCD3E, huCD3E/δ and huCD3ε/γ complexes measured by flow cytometry and Biacore

flow cytometry Biacore % of transfected cells positive for antibody (n=3-6) RU for Fab binding at 100 nM Clone huCD35 huCD38 (+/-SEM) huCD3s/5 (+/-SEM) huCD3s /δ huCD 3ε/γ 10E6 n.b. 96.3 +/- 2.2 93.3 +/- 4.7 eleven 8 11F9 n.b. 97.0 +/- 1.5 96.0 +/- 2.5 10 7 12D2 n.b. n.b. 96.3 +/- 2.2 eleven 5 12G3 n.b. 96.3 +/- 2.2 96.7 +/- 1.9 10 7 13H2 n.b. 93.3 +/- 3.3 93.0 +/- 4.2 9 6 18F5 n.b. 70.6 +/- 11.9 70.4 +/- 9.9 13 7 18G9 n.b. 95.0 +/- 2.5 97.0 +/- 1.5 10 3 1E6 n.b. 87.7 +/- 4.4 91.7 +/- 4.9 10 7 20E5 n.b. 96.7 +/- 1.9 95.3 +/- 3.2 eleven 8 20G6 n.b. 68.6 +/- 12.1 67.4 +/- 10.4 7 4 4B4 n.b. 71.5 +/- 10.8 69.5 +/- 8.7 eleven 3 4E7 n.b. 69.0 +/- 12.3 66.4 +/- 10.0 8 6 5B1 n.b. 97.0 +/- 1.5 97.0 +/- 1.5 10 7 OKTZ n.b. n.b. 70.2 +/- 9.6 4 3

n.b. - no binding (<20% huCD3e+huCD3d binding).

1.6. Binding of CD3-Fab to human T cells.

The ability to bind CD3-Fab was determined using flow cytometry. Primary human T cells were used as target cells. In this regard, peripheral blood mononuclear cells (PBMCs) were isolated from 200 ml of EDTA-treated peripheral blood of healthy donors using ficoll density gradient centrifugation. 15 ml Histopaque (Sigma-Aldrich) was preloaded into a 50 ml Leucosep tube (Greiner bio-one). Blood was diluted with autoMACS+1% BSA wash buffer (Miltenyi Biotec) and loaded onto the membrane in a total of ten prepared tubes. The tubes were centrifuged without braking for 10 min at 1000 xg. PBMCs were harvested and washed with autoMACS+1% BSA wash buffer three times. Finally, the PBMCs were resuspended in autoMACS run buffer (Miltenyi Biotec) to isolate T lymphocytes using autoMACSpro technology using an all T cell isolation kit.

- 95 041572 (Miltenyi Biotec) in accordance with the manufacturer's instructions. The purity of the isolated T cells was analyzed by MACSQuant flow cytometry using a 7-color human immunophenotyping kit (Miltenyi Biotec). The isolated T cells were resuspended in FBS staining buffer (BD Pharmingen) and plated at 1E5 cells in 100 μl per well in 96-well U-bottomed suspension culture plates (Greiner bio-one). Fab antibodies were serially diluted 1:3 in PBS (Invitrogen) and 5 μl of each was added to cells at a final maximum concentration of 30,000 ng/ml. The analyzed mixture was incubated for 45 min at 4°C. Cells were washed twice with FBS staining buffer and 1 μg of FITC-conjugated goat anti-human kappa chain F(ab')2 secondary antibody F(ab')2 (Beckman Coulter, no. 732621) in 100 μl of FBS staining buffer was added to each well. to the hole The analyzed mixture was incubated for 20 min at 4°C and then washed twice. Cells were resuspended in 150 μl of FBS staining buffer per well and measured using a MACSQuant (Miltenyi Biotec) or LSRII (BD) flow cytometer. Additional data analyzes were performed using FlowJo software (Tree Star, Inc.). The results were the percentage of cells positive for antibody binding. Cells treated with secondary antibody only and not with primary antibody were used for gating. EC50 curves were calculated using XLfit (Algorithm 205), EC50 values were calculated as the inflection point of the curve slope (data shown in Table 6).

Table 6 CD3-Fab affinities for human T cells measured by flow cytometry. Shown are average EC50 values calculated from curves

Antibody Human T cell binding, mean EC50 [nM] +/- SEM 12D2-E5-Fab 3, 8 +/- 0.2 4B4-D7-Fab 15, 3 +/- thirty 1E6-C9-Fab 18.7 +/- 3.5 10E6-G6-Fab 4.7 +/- 0.9 4E7-C9-Fab 6, 2 +/- 0.8 12G3-E8-Fab 4.4 +/- 0.3 5B1-G2-Fab 11, 1 +/- 3, 6 18F5-H10-Fab 4.1 +/- 0, 0 18G9-Hll-Fab 1.7 +/- 0.5 13H2-C2-Fab 14.3 +/- 0.9 20G6-F3-Fab 14.0 (n=1) llF9-F8-huFab 11, 6 (n=1) 20E5-F10-huFab 8.2 (n=1)

1.7. Safety assessment of CD3-Fab.

.7.1. CD3-Fab safety as measured by CD25+ and CD69+ expression on human T cells.

The effect of anti-CD3 Fab antibodies on T cell activation status as a safety outcome was analyzed by flow cytometry-based detection of CD25 and CD69 activation marker expression on the surface of primary human T cells.

Isolated primary human T lymphocytes were resuspended in RPMI+GlutaMAX I (Gibco)+10% FCS (Invitrogen) and 2.5E5 cells were seeded in 96-well U-bottomed suspension culture plates (Greiner bio-one) in 100 µl to the hole

μl of anti-CD3 Fab was added to the cells at a final concentration of 30,000 ng/ml. The analyzed mixture was incubated for 20 h at 37°C in 5% CO ₂ .

After an incubation period, cells were pelleted and stained for 15 min at 4° C. in 100 μl of FBS staining buffer (BD Pharmingen) per well with the following labeled antibodies: CD25-V450, CD69-APC.

Cells were washed twice after staining, resuspended in 150 μl of FBS staining buffer, and 5000 cells were measured using an LSRII flow cytometer (BD). Additional data analyzes were performed using FlowJo software (Tree Star, Inc.). The results were the percentage of CD25pos and CD69pos T cells (Table 7).

- 96 041572

Table 7 Safety of CD3-Fab measured by expression of CD25+ and CD69+ on human T cells

Antibody CD25+ % activation of human T cells normalized by PBS C=30,000 ng/mL n=1 CD69+ % activation of human T-cells normalized by PBS С=30000 ng/ml n=1 11H3-E5-Fab 0.7 0.5 12D2-E5-Fab-SEC 3.2 11.1 4B4-D7-Fab 2.5 5, 4 1E6-B8/B9-C9-Fab 1.8 3.2 10E6-G6-Fab 0.8 2, 6 4E7-C9-Fab 0.4 0, 1 12G3-E8-Fab 0, 0 1.5 5B1-G2-Fab 0.4 4.2 6F4-D10-Fab 2.1 6, 4 18F5-H10-Fab 0, 0 0, 1 18G9-Hll-Fab 0.7 2.2 13H2-C2-Fab 0.5 0.4 13Cl-F6-Fab 0.4 0.2 1E6-B8/B9-C9-Fab 0.6 1.7 20G6-F3-Fab 0.2 3.2 1E6-B8/B9-C9-Fab 2.5 7.8 1E6-B8/B9-C9-Fab 0, 1 1.0 HF9-F8-huFab 1.0 0.8 20E5-F10-huFab 1.5 1.3

1.7.2. CD3-Fab safety as measured by expression of CD4+/CD69+, CD4+/CD25+, CD8+/CD69+ and CD8+/CD25+ on human T cells.

The effect of anti-CD3 Fab antibodies on T cell activation status as a safety outcome was analyzed by flow cytometry-based detection of CD25 and CD69 activation marker expression on the surface of primary human T cells. Isolated primary human T lymphocytes were resuspended in RPMI+GlutaMAX I (Gibco)+10% FCS (Invitrogen) and 2.5E5 cells were plated in 96-well U-bottom suspension culture plates (Greiner bio-one) at 50 µl per well . Either only T cells were tested and wells filled with 50 µl RPMI+GlutaMAX I+10% FCS, or target cells (i.e. THP-1 cell line) were added at a rate of 2.5E4 cells per well in 50 µl RPMI+GlutaMAX I +10% FCS. The bispecific antibodies were serially diluted 1:3 in PBS (Invitrogen) and 5 μl of each was added to the cells at a final maximum concentration of 30,000 ng/ml. The analyzed mixture was incubated for 20 h at 37°C in 5% CO ₂ . After the incubation period, the cells were pelleted and stained for 15 min at 4°C in 100 µl of FBS staining buffer (BD Pharmingen) per well with the following labeled antibodies: CD4-PE, CD8-APC-Cy7, CD25-APC, CD69 -PE-Cy7. As a multi-antibody combination control without one (Fluorescence Minus One (FMO)) activated T cells were stained as described above, but CD25 was replaced with its isotype (isotypic APC-IG1k) in one tube, and CD69 was replaced with its isotype (isotypic PE-Cy7 -IG1k) in the second tube. Cells were washed twice after staining, resuspended in 150 μl of FBS staining buffer, and 5000 cells were measured using an LSRII flow cytometer (BD). Additional data analyzes were performed using FlowJo software (Tree Star, Inc.). The results were the percentage of CD4posCD25pos, CD4posCD69pos, CD8posCD25pos and CD8posCD69pos T cells. Gating was performed according to FMO controls (see Table 8).

Table 8 Safety of CD3-Fab measured by expression of CD4+/CD69+, CD4+/CD25+, CD8+/CD69+ and CD8+/CD25+ on human T cells

Antibody CD4+/CD69+ % activation PBS normalized C=100 nM CD8+/CD69+ % activation PBS normalized C=100 nM CD4+/CD25+ % activation PBS normalized C=100 nM CD8+/CD25+ % activation normalized by PBS

- 97 041572

mean + /SEM mean + /SEM mean + /SEM C=100 nM mean +/- SEM 20G6-F3Fab oh 1 +/ oh 1 0.0 +/- 0.0 0.4 +/- oh 1 0.0 +/ 0.1 4B4-D7Fab 3.2 +/ 1.1 0.3 +/- 0.3 1.3 +/- 0.6 0.0 +/ 0.0

Example 2 CD123 sequences.

2.1. Construction of expression plasmids for CD123(IL3RA)-hFc(CD123-Fc) fusion.

Using cDNA-containing plasmids as a template, fusion proteins were generated containing human CD123 and Macaca fascicularis in frame with a heavy chain constant region, including the GS linker (used in the Macaca protein), the hinge region, the CH2 and CH3 domains of human IgG immunoglobulin, additionally labeled with Strep-II (only in the human protein version).

Using human genomic DNA as a template, the extracellular domain of human CD123 (IL3RA) was amplified, including the signal sequence. The resulting amplified, digested and purified PCR products were pooled by ligation PCR and ligated into the mammalian expression vector pXL by InFusion using the NheI and HindIII sites. The sequence of the resulting fusion protein containing Fc and mature human CD123 labeled Strep-II is disclosed herein under SEQ ID NO: 196. under SEQ ID NO: 194 available from the NCBI database under accession number NP_002174.1) and thus the extracellular domain of human CD123.

For cloning CD123 Macac fascicularis, cDNA was obtained from the blood of the Macac fascicularis population. Using this isolated cDNA as a template, the extracellular domain of CD123 Macaca (IL3ra) was amplified, including the signal sequence. The resulting amplified, digested and purified PCR products were pooled by ligation PCR and ligated into the pXL mammalian expression vector using the InFusion method using NheI and HindIII. The sequence of the resulting fusion protein containing Fc and mature human CD123 tagged Strep-II is disclosed herein under SEQ ID NO:197. document under SEQ ID NO: 195, available from the NCBI database under accession number NP 002174.1) and thus the extracellular domain of human CD123.

2.2. Expression and purification of human CD123-FC and Macac fascicularis.

Freestyle HEK293 cells grown as a suspension culture in F17 serum-free medium (Life) were transiently transfected with the expression plasmid. Transfection was performed using Cellfectin (Life) transfection reagent. Cells were cultured at 37°C for 7 days. The culture supernatant containing the recombinant protein was collected by centrifugation and purified from impurities by filtration (0.22 μm).

For purification, Fc-containing fusion protein variants were captured on a Protein A (GE) matrix and eluted with a pH gradient. After purification of the protein by SEC in PBS using Superdex 200 (GE) and a final concentration step by ultrafiltration, the protein was used for further analyses.

2.3. Assessing the affinity of rat IgG from hybridoma for both human CD123 and Macac fascicularis.

Rat IgG screening for CD123 for binding affinity to human CD123 and cross-reactivity to cynomolgus CD123 was performed using hybridoma supernatants using Proteon XPR36 (Biorad) in a single point reaction kinetics approach. The capture assay was performed using goat anti-rat IgG (112-005-071, Jackson Immuno Research). Capture antibody was applied to GLC chips (176-5011, Biorad) up to about 8000 RU in the vertical direction using an amine coupling kit (176-2410, Biorad). Uptake of rat IgG up to about 200 RU in the vertical direction resulted in Rmax values of up to 100 RU for CD123-FC. Binding kinetics to the fusion protein containing human and cynomolgus Fc and CD123 were measured at 100 μl/min horizontally for 120 and 600 s for association and dissociation, respectively. CD123-FC proteins were used in two-fold dilutions from 6 to 100 nM. Buffer PBSET (176-2730, Biorad) was used as assay buffer. Regeneration was achieved by injection of 10 mM glycine buffer, pH 1.5, for 18 seconds at 30 μl/min. Data processing and analysis were carried out using ProteonManager v3.0 software. The sensograms were approximated using a 1:1 Langmuir model. Clones were selected based on affinity for human CD123 with KD < 1 nM and cross-reactivity with cynomolgus CD123.

Fab and CODV Fab.

The binding affinities of CD123-binding Fab or CODV-Fab were measured using the instrument

- 98 041572 Biacore3000 (GE Healthcare). The assay buffer was HBS-EP (BR-1001-88, GE Healthcare). Capture of CD123-FC fusion proteins was achieved using a human antibody capture kit (GE Healthcare). The capture antibody was bound on CM5 chips (BR-1001-88, GE Healthcare) to about 12,000 RU using an amine coupling kit (BR-100-50, GE Healthcare). CD123-Fc fusion proteins were captured at 10 ml/min up to about 70 RU to give Rmax values of 30 RU. The kinetics of binding to Fab or CODV-Fab to CD123 was measured at 30 μl/min for 240 and 600 s for the association and dissociation phase, respectively. Two-fold dilutions of Fab from 3 to 200 nM in assay buffer were used. All Fab concentrations were run in duplicate with buffer control in duplicate to provide a dual control. The capture surface was regenerated with a 1-minute injection of a 3M MgCl ₂ solution at a rate of 30 μl/min. BIAevaluation v.4.1 software (GE Healthcare) was used for data analysis. The data were approximated as a whole using a 1:1 Langmuir model with mass transfer.

IgG and CODV-Fc proteins.

Measurement of the binding affinity of IgG and CODV-Fc proteins for CD123 was carried out similarly to the binding assay for Fab and CODV-Fab except for the capture antibody. In this case, the classic Strep-MAB antibody (2-1507-001, IBA) was used to capture human CD123-Fc via its Strep-II tag. In this case, the regeneration solution was 10 mM glycine buffer, pH 2.0.

2.4. Obtaining antibodies with cross-reactivity to human CD123 and Macac fascicularis.

Human CD123 and Macac fascicularis cDNA were cloned into proprietary immunization vectors from Aldevron (pB8 and W8), respectively. Three rats in the MR13-296 immunization group were immunized with the IL3RA-hum.-ECD immunization vector (aa 19-305). Immune sera were collected on day 24 of the immunization protocol, after 4 applications of gene immunization (IS24d-4). Sera diluted in PBS with 3% FBS were tested by flow cytometry using mammalian cells transiently transfected with human and cynomolgus IL3RA cDNA variants, IL3RA-hum.ECD and IL3RA-hum.D3.

Specific reactivity of immune sera against cells transfected with pB1IL3RA-hum.ECD as well as IL3RA-cyno (pFF1262) and THP-1 cells could be detected in all immunized animals when compared with cells transfected with irrelevant cDNA.

Sera positive rats were sacrificed and B cells fused with mouse myeloma cells. The resulting hybridomas were screened in HEK293 cells transfected with human or cynomolgus CD123 expression plasmids in various CD123 expressing cell lines by flow cytometry (data shown in Table 10).

Target cells were seeded at a rate of 5E4 cells in 50 μl of FBS staining buffer (BD Pharmingen) per well in 96-well U-bottomed suspension culture plates (Greiner bio-one).

Hybridoma supernatants were serially diluted 1:3 in PBS (Invitrogen) and 50 μl of each was added to cells at a final maximum concentration of 1 μg/ml. The analyzed mixture was incubated for 45 min at 4°C.

Cells were washed twice with FBS staining buffer and 1 µg of goat anti-rat IgG (H+L) secondary antibody conjugated to Alexa Fluor 488 (Invitrogen-Life Technologies, #MH10520) was added to 100 µl of FBS staining buffer per well. The analyzed mixture was incubated for 15 min at 4°C and then washed twice.

Cells were resuspended in 200 μl of FBS staining buffer per well and measured using a MACSQuant (Miltenyi Biotec) or LSRII (BD) flow cytometer. Additional data analyzes were performed using FlowJo software (Tree Star, Inc.). The results were the percentage of cells positive for antibody binding. Cells treated with secondary antibody only and not with primary antibody were used for gating. Curves were calculated using XLfit (Algorithm 205).

Specific clone binding to CD123 could be shown on the surface of transfected HEK293 compared to non-transfected HEK293 cells in which the signal could not be detected (data not shown). Antibody binding was concentration dependent, with EC50 ranging from 0.4 to 17.7 ng/mL (Table 9).

- 99 041572

Table 9 Specific Binding of CD123 by Rat IgG Clones in Hybridoma Supernatants Detected by Flow Cytometry

Concentration-dependent antibody binding was measured using CD123-transfected HEK293 as target cells and secondary goat anti-rat IgG (H+L) conjugated to Alexa Fluor 488. EC50 values calculated from the curve are shown.

Table 10 Anti-CD123 antibody binding data to recombinant CD123 and cells expressing CD 123

Clone ka (1/ms) kd (1/s) KD (nM) Human KD (nM) Macaccrabeater Molm-13 TNR-1 OCI-AML3 SP2 Basophils Blocking IL3 1E1 4.6E+05 1.1E-04 0.23 0.72 + + + + + - 2B8 6.6E+05 1.2E-05 0.018 0.77 + + + + + - 2F8 1,ZE+05 6.6E-05 0.53 0.51 + + + - + + SG10 2.2E+05 0.5E-05 0.21 0.27 + + + - + + ZEZ 1.6E+05 6.4Е-05 0.4 0.12 + + + - + + 5A5 1,ZE+05 1.2E-04 1.1 0.05 + + + + + - 6B10 2,WE+05 1.5E-05 0.15 0, 1 + + + + + - 6S10 1.8E+05 1.4Е-04 0.75 0.79 + + + + + - 8B11 4.7E+05 6.2E-06 0.01 0, 1 + + + + + - 9B8 1.5E+06 2.8E-04 0.19 0.17 + + + + + + 9D7 1.0E+06 3.2E-04 0.3 0.07 + + + + + + 9F6 1.8E+05 6.7E-06 0.04 0.5 + + + + + -

Binding data showing the affinities of anti-CD123 antibodies for recombinant human and cynomolgus monkey CD123 protein. Cell surface binding was detected by flow cytometry. SP2 refers to cells expressing the N-terminal truncated (D1 region) variant of CD123. Antibodies were tested with Proteon XPR36 for their ability to compete with IL3 binding to CD123.

2.5. Humanization of rat anti-CD123 antibody sequences.

- 100 041572

Rat antibodies were humanized by CDR grafting or by the 4D method (US 20110027266). For the rat anti-CD3 antibody, 3E3, the closest identified Rattus germline sequences were IGHV2S48*01 and IGHJ3*01 for the heavy chain variable region and IGLV3S2*01 and IGKJ3*01 for the light chain variable region. The calculated identity with the rat germline sequence (framework sequences only) is 94.51% for VH and 98.9% for VL.

Potential available problematic residues were checked and one residue in CDRH2 was modified.

Using the grafting method, a plurality of humanized variants were generated based on the closest human germline sequences identified: IGHV459*05 and IGHJ4*01 for VH (framework germline sequence identity: 75.82%); IGLV6-57*01 and IGLJ3*01 for VL (Identity rate with germline sequence in framework regions: 72.22%).

In addition to CDR grafting, a 4D humanization protocol (US20110027266) was used to humanize the light chain (VL) and heavy chain (VH) variable domains of the rat anti-CD123 antibody, 3E3. Molecular dynamics (MD) modeling was performed on a minimized 3D homologous model (performed with MOE; PDB: 1FLR used) of rat anti-CD123, 3E3 antibodies and compared with 49 human antibody models derived from seven representative light chains (vk1, vk2, vk3, vk4, vlambda1, vlambda2, vlambda3) and seven representative heavy chains (vh1a, vh1b, vh2, vh3, vh4, vh5, vh6) identified by LGCR/SDI and available from MOE.

Two models were chosen for 4D humanization: Vl3-vh4 and VL3-VH2 with the best both hydrophobic and electrostatic components and sequence identity beyond the CDR. For pairwise comparison of the variable domain of the rat anti-CD123 antibody, 3E3, and the two selected models, the sequences were aligned based on the optimal 3D overlap of alpha carbons in the respective homologous models.

Example 3: Antibodies in CODV-Fab/.1 bispecific format. Cloning of selected CD3 sequences in combination with an anti-CD123 mAb, 7G3, as a bispecific CODV-Fab format to test their T cell recruitment activity.

Selected anti-CD3 antibody sequences such as I2C, mAb2 (Macrogenics) and referred to as 20G6-F3, 4E7-C9, 4B4-D7 and 18F5-H10 were expressed as monospecific anti-CD3 Fabs. These selected sequences were also expressed as anti-CD3xCD123 bispecific CODV-Fabs using sequences from the 7G3 monoclonal antibody, resulting in the CODV-Fab I2Cx7G3 constructs termed 7G3x20G6, 7G3x4E7, 7G3x4B4, and 7G3xl8F5, as further described herein above under Antibody-Like Binding Proteins. . Purified proteins were used in the Biacore assay to compare affinity for CD3ε/δ complexes (data are presented in Table 11). No change in affinity could be detected by the Biacore assay when the CD3 sequences were administered in the CODV-Fab bispecific format.

Table 11. Comparison of the affinities of Fab fragments and CODV-Fab for human and cynomolgus CD3ε/δ complexes

Molecule hu kon 1/ms hu koff 1/s hu KD nM su KD nM Fab fragments I2C 1.1E+07 1,ZE-OZ 0.13 0.5 rnAb2 3.1E+05 4,ZE-OZ 13.7 16, 3 BDW-20G6-F3 3.5E+04 2.7E-04 8 8 BDX-4E7-C9 8.9E+04 1.5E-03 17 15 BDW-4B4-D7 1.1E+05 2.9E-03 27 24 BDX-18F5- H10 2.0E+05 1.8E-04 1 1 CODV Fab 7G3XI2C 4.0E+06 2.0E-03 0.5 0.4 7G3X20G6 7.2E+04 5.1E-04 7 7 7G3X4E7 1.2E+05 1.9E-03 16 10 7G3X4B4 2.7E+05 4.4E-03 17 18

3.2. Bispecific CODV-Fabs directed against CD123 and CD3 mediate strong retargeting cytolysis by T cells.

Such bispecific CODV-Fabs have the ability to localize a T cell (by associating such a T cell with the CD3 binding portion of a CD3 binding CODV-Fab) at the location of a tumor cell (by associating such a cancer cell with a CD123 binding portion of the CODV-Fab) . The localized T cell can then mediate tumor cell cytolysis in a process referred to herein as redirected cytolysis. Bispecific CD123 and CD3 specific CODVFabs were constructed with the variable domains of an anti-CD123 monoclonal antibody, 7G3, and the variable domains of anti-CD3 antibodies selected from the anti-CD3 antibodies prepared in Example 1.

In this regard, peripheral blood mononuclear cells (PBMCs) were isolated from 200 ml of EDTA-treated peripheral blood of healthy donors using ficoll density gradient centrifugation.

ml Histopaque (Sigma-Aldrich) was preloaded into a 50 ml Leucosep tube (Greiner bio-one). Blood was diluted with autoMACS+1% BSA wash buffer (Miltenyi Biotec) and loaded onto the membrane in a total of ten prepared tubes. The tubes were centrifuged without braking for 10 min at 1000 x g. The PBMCs were harvested and washed with autoMACS+1% BSA wash buffer three times. Finally, the PBMCs were resuspended in autoMACS run buffer (Miltenyi Biotec) for T lymphocyte isolation using autoMACSpro technology using the Total T Cell Isolation Kit (Miltenyi Biotec) according to the manufacturer's instructions. The purity of the isolated T cells was analyzed by MACSQuant flow cytometry using a 7-color human immunophenotyping kit (Miltenyi Biotec).

The effect of recruiting T cells under the action of bispecific antibodies was analyzed using analysis based on flow cytometry. Target cells (ie THP-1 cell line) were stained for 15 min at 37°C with 1 μM CFSE in 1 ml RPMI+GlutaMAX I (Gibco) per 1E7 cells. The cells were then washed twice and resuspended in RPMI+GlutaMAX 1+10% FCS (Invitrogen). 2.5E4 target cells were plated in 96-well U-bottomed suspension culture plates (Greiner bio-one) in 50 μl of medium per well.

Isolated primary human T lymphocytes were resuspended in RPMI+GlutaMAX I+10% FCS and added at the indicated effector-target ratio at 50 µl per well to target cells (in most cases E:T=10:1).

The bispecific antibodies were serially diluted 1:3 in PBS (Invitrogen) and 5 μl of each was added to the cells at a final maximum concentration of 3000 ng/ml.

The analyzed mixture was incubated for 20 h at 37°C in 5% CO ₂ .

To detect dead target cells, all cells were stained with 7-AAD. Therefore, 5 μg/ml of 7-AAD diluted in FBS staining buffer (BD Pharmingen) was added to each well and incubated for 15 min at 4° C. in the dark. Cells were measured using MACSQuant (Miltenyi Biotec) or LSRII flow cytometer (BD), respectively. Additional data analyzes were performed using FlowJo software (Tree Star, Inc.). The results were the percentage of double positive cells for CFSE and 7-AAD.

The results of these studies, shown, for example, in table. 20-22 demonstrate the ability of the anti-CD123xCD3 CODV-Fab to mediate redirected tumor cell cytolysis.

3.3. Evaluation of the safety of redirected cytolysis under the influence of T-cells.

The effect of bispecific antibodies on T cell activation status as a safety outcome was analyzed by flow cytometry-based detection of CD25 and CD69 activation marker expression on the surface of primary human T cells.

Isolated primary human T lymphocytes were resuspended in RPMI+GlutaMAX I (Gibco)+10% FCS (Invitrogen) and 2.5E5 cells were plated in 96-well U-bottom suspension culture plates (Greiner bio-one) in 50 µl to the hole

Either only T cells were tested and wells filled with 50 µl RPMI + GlutaMAX I + 10% FCS, or target cells (i.e. THP-1 cell line) were added at a rate of 2.5E4 cells per well in 50 µl RPMI+GlutaMAX I +10% FCS.

The bispecific antibodies were serially diluted 1:3 in PBS (Invitrogen) and 5 μl of each was added to the cells at a final maximum concentration of 30,000 ng/ml.

The analyzed mixture was incubated for 20 h at 37°C in 5% CO ₂ .

After the incubation period, the cells were pelleted and stained for 15 min at 4°C in 100 µl of FBS staining buffer (BD Pharmingen) per well with the following labeled antibodies: CD4PE, CD8-APC-Cy7, CD25-APC, CD69-PE -Cy7.

As a multi-antibody combination control without one (Fluorescence Minus One (FMO)) activated T cells were stained as described above, but CD25 was replaced by its isotype (isotypic APC-IG1k) in one tube, and CD69 was replaced by its isotype (isotypic PE- Cy7-IGlk) in the second tube.

- 102 041572

Cells were washed twice after staining, resuspended in 150 μl of FBS staining buffer, and 5000 cells were measured using an LSRII flow cytometer (BD). Additional data analyzes were performed using FlowJo software (Tree Star, Inc.). The results were the percentage of CD4posCD25pos, CD4posCD69pos, CD8posCD25pos and CD8posCD69pos T cells. Gating was performed according to FMO controls (see Table 12).

Table 12 Anti-CD123xCD3 activity of CODV-Fab as measured by cytotoxicity assay and safety as measured by CD69 expression

Molecule Activity EC50 [ng/ml] Safety Expression of CD69 % cells/% max. 7G3 x I2C 20+9 25.8 + 7.6 7G3 x OKTZ 16+5 25.5 + 8.0 7G3x20G6 18+5 8.7 + 0.8 7G3x4E7 12+3 12.3+2.5 7G3x4B4 11+2 11.6 + 0.8 7G3X18F5 61+15 9.8+1.1

3.4. Humanization of rat anti-CD3 antibody and bispecific antibody sequences.

Humanization of rat antibodies was carried out by CDR grafting or by the 4D method (US 20110027266).

For the rat anti-CD3 antibody, 20G6, the closest identified Rattus germline sequences were IGHV6S17*01 and IGHJ2*01 (for the heavy chain variable region) and IGKV1S21*O1 and IGKJ4*01 (for the light chain variable region). The calculated rate of identity with the rat germline sequence (framework sequences only) is 97.80% for VH and 95.5% for VL.

Created a lot of humanized variants using the grafting method based on:

1) closest human germline sequences identified (IGHV3-30-01 IGHJ4-01 for VH with 77% germline identity in framework regions; IGK2D-29-02 IGKJ4-01 for VL with sequence identity germinal type in the frame regions, constituting 80%), or

2) based on the closest germline sequence identified with lower pI IGVH3-48*02-IGHJ4-01 for VH with a germline sequence identity in the framework regions of 75%; IGKV2-28*01-IGKJ4-01 for VL with 77.5% identity with the germline sequence in the framework regions, or

3) based on more distant human germline sequence (germline clade change) (IGVH1-46*01-IGHJ4*01 for VH with 56% germline identity in framework regions; IGKV4*01-IGKJ4*01 for VL with a germline sequence identity in the framework regions of 67.5%).

The humanized sequences were then formatted as a CODV-Fab molecule called CODV-Fab 7G3x20G6 in combination with the anti-CD123 sequence from the 7G3 antibody described previously. Purified anti-CD123xCD3 CODV-Fabs were used in the Biacore assay to assess affinity for CD3ε/δ (see Table 13).

- 103 041572

Table 13. Affinity of selected humanized variants of the anti-CD3 antibody, 20G6, for the recombinant CD3ε/δ complex

Combinations of humanized sequences used in CODVFab 7G3X20G6 ka (1/Mc) kd (1/c) Rmax (RU) KD(M) Source sequence 2.17E+04 5.71E-04 46 2.63E-08 VL1A_VH1A 2.86E+04 4.34E-03 46 1.52E-07 VL1B_VH1A 2.99E+04 4.14E-03 45 1.38E-07 VL1C_VH1D 2.30E+04 1.01E-03 58 4.40E-08 VL1D_VH1B 1.86E+04 1.09E-03 62 5.87E-08 VL1D_VH1C 2.22E+04 7.13E-04 71 3.21E-08 VL1D_VH1D 2.40E+04 7.46E-04 59 3.10E-08

For the rat anti-CD3 antibody, 4B4-D7, the closest identified Rattus germline sequences were IGHV6S17*01 (93% identity) and IGHJ2*01 (87.5% identity) for the heavy chain variable region and IGKV1S21*01 (93% identity ) and IGKJ4*01 (100% identity) for the light chain variable region.

The calculated percentage identity of the identified Rattus V sequences to the human germline sequence (framework sequences only) is 79% for VH and 77.53% for VL.

A number of pairs of humanized variants for VH and VL were generated by grafting an additional engineered sequence using the closest human germline sequences (IGHV3-30*01_IGHJ6*02; IGKV2-30*02/IGKV2D-39*02_IGKJ2*01). The calculated percentages of identity with the human germline sequence (frame 4 IMGT sequences only) for the humanized V sequences are listed in Table 1. 14.

Table 14. Percentage identity with human germline sequence for humanized V sequences grafted from an antibody termed B4-D7

Grafted variants 4B4-D7 Index of identity with the human germ type sequence, % VLlCmodifl 92.1 VLlCmodif2 91.0 VLlCmodif3 91.0 VLIAmodif1 97.8 VLlAmodif2 96.6 VLlAmodif3 95.5 VL2C 95.5 VL2D 96.7 VL2F 96.7 VH6Bmodifl 92.3 VH6Bmodif2 90.1 VH6Amodifl 94.5 VH6Amodif2 92.3 VH6Amodif3 91.2 VH6C 95.5 VH6D 93.5

Percentages were calculated based on 4 IMGT framework sequences only.

In addition to CDR grafting to humanize light chain (VL) and heavy chain variable domains

- 104 041572 rat anti-CD3 antibody VH, 4B4-D7, used the 4D humanization protocol described in US Patent Application US 20110027266. Molecular dynamics (MD) modeling was performed on a minimized 3D homologous model (performed with MOE; used PDB: 1FLR) rat anti-CD3 antibody, 4B4-D7, and compared with 49 human antibody models derived from seven representative light chains (vkl, vk2, vk3, vk4, vlambdal, vlambda2, vlambda3) and seven representative heavy chains ( vh1a, vh1b, vh2, vh3, vh4, vh5, vh6) labeled with LGCR/SDI and available from MOE.

Two models were chosen for 4D humanization. vk1-vh6 with the highest 4D similarity for both hydrophobic and electrostatic components. vk2-vh3 with the highest sequence identity outside the CDR. For pairwise comparison of the variable domain of the rat anti-CD3 antibody, 4B4-D7, and the two selected models, the sequences were aligned based on the optimal 3D overlap of alpha carbons in the respective homologous models. A number of other pairs of humanized variants for VH and VL were further optimized.

The calculated percentages of identity with the human germline sequence (frame 4 IMGT sequences only) for the humanized V sequences are listed in Table 1. 15.

Table 15. Percentage identity with human germline sequence for humanized V sequences obtained

Percentages were calculated based on 4 IMGT framework sequences only.

The humanized sequences were expressed as Fab fragments and purified followed by Biacore analysis to assess affinity for CD3ε/δ (data shown in Table 16).

Table 16. Affinity of selected humanized anti-CD3 antibody variants,

4B4-D7, in relation to the recombinant CD3ε/δ complex

fab VL vh Ligan D ka kd KD 4B4 huCD3 3.26E+05 2.85E-03 8.79E-09 cyCD3 1.90E+05 2.37E-03 l,26E-08 4В4_1 Vlla 75% VH1B 80% huCD3 6.28E+05 1.74E-03 2.77E-09 cyCD3 5.06E+05 1.58E-03 3.13E-09 4B4_2 Vllb 75% VH1B 80% huCD3 2.86E+05 2.14E-03 7.48E-09 cyCD3 2.77E+05 2.10E-03 7.57E-09 4B4_5 VL2C 95.5% VH6D 93.5% huCD3 2.33E+05 4.23E-03 l,82E-08 cyCD3 2.53E+05 4.20E-03 l,66E-08 4B4_6 Vlla 75% VH1A 80% huCD3 1.50E+06 1.85E-03 l,24E-09

- 105 041572

cyCD3 1.49E+06 1.77E-03 l,19E-09 4B4_7 Vllb 75% VH1A 80% huCD3 3.06E+05 2.48E-03 8.10E-09 cyCD3 3.01E+05 2.19E-03 7.28E-09 4B4_9 VL2B 83% VH2A 80% huCD3 2.18E+05 2.91E-03 l,34E-08 cyCD3 2.85E+05 3.44E-03 l,21E-08 4В4_1 0 VL1_CM2 91% VH6_BM2 90% huCD3 4.34E+05 3.99E-03 9.19E-09 cyCD3 4.30E+05 3.84E-03 8.94E-09 4В4_1 1 VL1_CM3 91% VH6_BM2 90% huCD3 5.66E+05 7.74E-03 l,37E-08 cyCD3 2.38E+05 4.86E-03 2.04E-08 4В4_1 7 VL2A 83% VH2B 84.5% huCD3 3.10E+05 2.76E-03 8.91E-09 cyCD3 2.98E+05 2.70E-03 9.07E-09 4В4_2 0 VL2F 96.5% VH6C 95.5% huCD3 5.09E+04 1.71E-03 3.35E-08 cyCD3 4.56E+04 1.98E-03 4.34E-08

3.5. Binding of CODV hu-Fab to CD123xCD3 with THP-1 and TF-1 cells.

Selected anti-CD123 antibody sequences were cloned into CODV-Fab format in combination with a CD3 binding sequence, and the proteins were expressed and purified.

Their ability to bind to cells naturally expressing CD123 was determined by flow cytometry. The THP-1 cell line or the TF-1 cell line was used as target cells.

Target cells were blocked with an FcR blocker (Sigma). In this regard, target cells were resuspended in FBS staining buffer (BD Pharmingen) and blocked with 100 μl of blocking reagent per ml for 1 hour at 4°C. Cells were flooded with FBS staining buffer and 1E5 cells at 50 μl per well were plated in 96-well U-bottomed suspension culture plates (Greiner bio-one).

Antibodies were added at the rate of 3 μg in 50 μl of FBS staining buffer per well. The analyzed mixture was incubated for 30 min at 4°C.

Cells were washed twice with FBS staining buffer and 1 μg of FITC-conjugated goat anti-human kappa chain F(ab')2 secondary antibody F(ab')2 (Beckman Coulter, no. 732621) in 100 μl of FBS staining buffer was added to each well. to the hole The analyzed mixture was incubated for 20 min at 4°C and then washed twice.

Cells were resuspended in 150 μl of FBS staining buffer per well and measured using a MACSQuant (Miltenyi Biotec) or LSRII (BD) flow cytometer. Additional data analyzes were performed using FlowJo software (Tree Star, Inc.). The results were the percentage of cells positive for antibody binding. Cells treated with secondary antibody only and not with primary antibody were used for gating.

CD123xCD3 CODV-Fab binding to CD123 was shown with two different cell lines expressing CD123, either co-expressed with CD131 on a TF-1 cell line or on the surface of THP-1 cells in the absence of CD131 expression. As an example, five different clones are shown binding to target cells. Anti-CD19xCD3 CODV-Fab was used as a negative control (specificity control) and a reference CD123xCD3 CODV-Fab as a positive control (Table 17).

Table 17 Specific binding of CD123 targeting sequences cloned into the backbone of CODV-Fab in combination with CD3 binding sequence to THP-1 and TF-1 cell lines

% FITC-positive cells Antibody THP-1 cells TF-1 cells Environmental control 0.10 0.00 Control with secondary AV 0.00 0.20 Specificity control with CD19 x CD3 0.10 0.20 CODV-Fab CD123 x CD3 99, 90 82.50 CODV-Fab BFX-2F8-D6 x CD3 24.00 9.30 CODV-Fab BFX-3E3-D3 x CD3 99.00 31, 60 CODV-Fab BFX-9B8-G6 x CD3 48.80 31, 10 CODV-Fab BFX-9D7-C8 x CD3 88.70 50.20 CODV-Fab BFX-9F6-G3 x CD3 99, 70 60.70

Antibody binding was detected using THP-1 and TF-1 expressing CD123 as target cells. 3 μg of antibody was added and detected with a goat anti-human kappa chain secondary antibody conjugated to FITC. The percentages of antibody-positive target cells are shown.

3.6. Cytotoxic effect on THP-1 cells mediated by CODV-Fab to

CD123xCD3.

The effects of T cell recruitment by bispecific antibodies consisting of the newly generated anti-CD123 sequence and the same CD3 binding sequence were analyzed by flow cytometry based cytotoxicity assay. The effector cells were primary T cells isolated from whole blood of healthy donors. THP-1 cells were used as target cells expressing CD123.

Peripheral blood mononuclear cells (PBMCs) were isolated from 200 ml of EDTA-treated peripheral blood of healthy donors using ficoll density gradient centrifugation. 15 ml Histopaque (Sigma-Aldrich) was preloaded into a 50 ml Leucosep tube (Greiner bio-one). Blood was diluted with autoMACS+1% BSA wash buffer (Miltenyi Biotec) and loaded onto the membrane in a total of ten prepared tubes. The tubes were centrifuged without braking for 10 min at 1000xg. The PBMCs were harvested and washed with autoMACS+1% BSA wash buffer three times. Finally, the PBMCs were resuspended in autoMACS run buffer (Miltenyi Biotec) for T lymphocyte isolation using autoMACSpro technology using the Total T Cell Isolation Kit (Miltenyi Biotec) according to the manufacturer's instructions. The purity of the isolated T cells was analyzed by MACSQuant flow cytometry using a 7-color human immunophenotyping kit (Miltenyi Biotec).

Target cells (ie THP-1 cell line) were stained for 15 min at 37°C with 1 μM CFSE in 1 ml RPMI+GlutaMAX 1+10% FCS (Invitrogen). 2.5E4 target cells were plated in 96-well U-bottomed suspension culture plates (Greiner bio-one) in 50 μl of medium per well.

Isolated primary human T lymphocytes were resuspended in RPMI+GlutaMAX 1+10% FCS and added at the indicated effector-target ratio at 50 µl per well to target cells (in most cases E:T=10:1).

The bispecific antibodies were serially diluted 1:3 in PBS (Invitrogen) and 5 μl of each was added to the cells at a final maximum concentration of 3000 ng/ml.

The analyzed mixture was incubated for 20 h at 37°C in 5% CO ₂ .

To detect dead target cells, all cells were stained with 7-AAD. Therefore, 5 μg/ml of 7-AAD diluted in FBS staining buffer (BD Pharmingen) was added to each well and incubated for 15 min at 4° C. in the dark. Cells were measured using MACSQuant (Miltenyi Biotec) or LSRII flow cytometer (BD), respectively. Additional data analyzes were performed using FlowJo software (Tree Star, Inc.). The results were the percentage of double positive cells for CFSE and 7-AAD. Curves were calculated using XLfit (Algorithm 205).

As shown as an example in table. 18, the bispecific antibodies were able to recruit primary T cells and lyse target THP-1 cells in vitro. An antibody concentration-dependent increase in dead target cells could be detected after 20 hours of co-incubation. For the antibodies shown here, the calculated EC50 values ranged from 12.2 to 429.3 ng/mL.

- 107 041572

Table 18 T cell recruitment effect of anti-CD123xCD3 bispecific CODV-Fab detected in flow cytometry-based cytotoxicity assays

Antibody EC50 [ng/ml] geometric mean +/- SEM EC50 [nM] geometric mean +/- SEM (n=4) η (G v=4) CODV-Fab x CD3 BFX-2B8-F3 106.9 +/- 33, 9 1420.5 +/- 450.2 4 CODV-Fab x CD3 BFX-3E3-D3 12.2 +/- 5, 9 163.0 +/- 80.1 4 CODV Fab E4 x CD3 BFX-6B10- 39.7 +/- 17.9 529.1 +/- 237.7 4 CODV-Fab x CD3 BFX-9B8-G6 106.1 +/- 36, 1 1410.2 +/- 479.8 4 CODV-Fab x CD3 BFX-9D7-C8 86.3 +/- 29, 3 1148.6 +/- 390.5 4 CODV-Fab x CD3 BFX-9F6-G3 13, 9 +/- 7.7 185.9 +/- 104.1 3 CODV Fab E6 x CD3 BFX-3B10- 429.3 +/- 82.6 5709.3 +/- 1099.0 2

Average EC50 values calculated from the curves are shown.

Anti-CD123 clone 3E3 was combined with a humanized version of the anti-CD3 antibody, 4B4, in CODV-Fab format. Their effect on T cell recruitment and their ability to activate T cells in vitro were analyzed.

Cytotoxicity assays were performed as described above. In table. 20 shows the lytic effect of primary human T cells on the target THP-1 cell mediated by these constructs, eg CODV-Fab hz4B4(4D_A)x3E3.

Cytotoxic activity could be reliably induced with a concentration-dependent effect using T cells isolated from seven different healthy donors (Table 19).

Table 19 T cell recruitment effect of bispecific CODV-Fab Fz4B4(4D_A)x3E3 detected in flow cytometry-based cytotoxicity assay

CODV Fab Cytotoxicity (THP-1) Cytotoxicity (THP-1) geometric mean EC50 [nM] +/- SEM n=7 geometric mean EC50 [ng/ml] + /SEM n=7 hz4B4(4D_A) x ZEZ CODV-Fab 26.5 +/- 5, 9 2.0 +/- 0.5

Average EC50 values calculated from the curves are shown.

3.7. Effect of T cell activation by CODV-Fab to CD123xCD3 or DART.

The effect of bispecific antibodies on T cell activation status as a safety outcome was analyzed by flow cytometry-based detection of CD25 and CD69 activation marker expression on the surface of primary human T cells as described above. The comparison included a DART format CD123xCD3 bispecific diabody single chain (herein referred to as MGD006) which was described in WO 2015026892 containing a first polypeptide chain with the sequence under SEQ ID NO: 386 (which is SEQ ID NO: 1 shown in WO 2015026892) and a second polypeptide chain with the sequence under SEQ ID NO: 387 (which is SEQ ID NO: 3 shown in WO 2015026892) covalently linked to each other via a disulfide bond.

When CODV-Fab was incubated with isolated T cells alone, no significant increase in expression of the late activation marker CD25 could be detected on the surface of CD4 positive and CD8 positive T cells (data not shown). Equally, no concentration-dependent increase in expression of the early activation marker CD69 was observed on both T cell subgroups (Table 20). As a result, the construct was scored as inactive (NA). In contrast, a significant increase in the expression level of both markers was measurable when target THP-1 cells were added (data for CD25 not shown, data for CD69 are shown in Table 21).

- 108 041572

Table 20. Effect of anti-CD123xCD3 or DART bispecific CODV-Fabs on T cell activation status as detected by CD69 expression level in flow cytometry-based analysis

CODV Fab Safety - T cell activation without target cells n=3 CD4+/CD69+% activation, DBS normalized C=100 nM mean +/- SEM Minimum effective concentration [pM] for CD4+/CD69+ mean +/- SEM CD8+/CD69+ activation normalized to BBS C=100 nM mean +/- SEM Minimum effective concentration [pM] for CD8+/CD69+ mean +/- SEM hz4B4(4D_A)x3E 3 CODV-Fab 14 + / 2 NA 13 + / 2 NA hz2 0G6xhz7G3 CODV-Fab 22 + / 2 22 + / 2 hz2 0G6xhz7G3 CODV-Fab-TLl 18 + / 4 15 + / 2 hz20G6xhz7G3 CODV-Fab-OLl 6 + / 2 9 + / 2 Single chain antibody in DART format, MGD006 82 + / 9 83 + / 4

Shown are the mean percentages of activated CD8 and CD4 T cells at 100 nM antibody and the lowest effective concentration in T cell-only assays.

The results shown in table. 20 indicate that the single chain antibody (DART) causes significantly more T cell activation in the absence of target cells under the conditions tested.

Table 21. Effect of bispecific CODV-Fab Hz4B4(4D_A)x3E3 on T cell activation state as detected by CD69 expression level in flow cytometry-based analysis

CODV Fab Safety - T cell activation with target THP-1 cells n=3 CD4+/CD69+% activation normalized to PBS C=mean cytotoxicity at EC50 mean +/- SEM CD4+/CD69+ activation normalized to PBS Cmax mean +/- SEM CD8+/CD69+% activation normalized by PBS C=mean cytotoxicity at EC50 mean +/- SEM CD8+/CD69+ % activation normalized by PBS Cmax mean +/- SEM hz4B4(4D_A)x ZEZ CODV-Fab 71 +/- 4 84 +/- 2 69 +/- 6 82 +/- 4

Shown are the mean percentages of activated CD8 and CD4 T cells in the cytotoxicity assay at maximum antibody concentration (Cmax) and at EC50 antibody concentration. The assays were performed by co-incubation of target THP-1 cells and T cells.

- 109 041572

Table 22 Effect of Bispecific, Fully Humanized 7G3 Containing Molecules

CODV and single-stranded Dart on the state of T-cell activation, detected by the level of CD69 expression in the analysis based on flow cytometry

CODV Fab Safety - T cell activation with targeted THP-1 cells CD4+ T cells (% CD69+ cells) EC50 (pM) n=1-6 CD8+ T cells (% CD69+ cells) EC50 (pM) n=1-6 hz20G6xhz7G3 CODV-Fab 1.4 3, 1 hz20G6xhz7G3 CODV-Fab-TLl 3.2 9, 9 hz20G6xhz7G3 CODV-OL1 10 3.3 Single chain antibody DART format, MGD006 10 3.5

Shown are EC50 values for representative assays of activated CD8 and CD4 T cells in a cytotoxicity assay. The assays were performed by co-incubation of target THP-1 cells and T cells.

To evaluate the cytotoxic effects of novel anti-CD123 antibodies with humanized anti-CD3 moieties, CODV-Fab hz20G6x7G3, 7G3xhz4B4, hz4B4x3E3 containing various combinations of Fv were generated. One Fc-containing variant, CODV-Fab hz20G6x7G3-TL4, was also made, labeled on the light chain Fc to form Fc heterodimers with the corresponding heavy chain (variant TL4). The affinities of the bispecific construct for the CD3ε/δ complex and CD123 were measured using Biacore. In addition, a cytotoxicity assay was performed as described above, and CD4+ activation and CD8+ activation were measured.

Table 23. Affinities and activities of bispecific CODV molecules for CD123xCD3

Bispecific molecule KD (CD3e/d) [hM] KD (CD123) [HM] Cytotoxicity assay (ΤΗΡ cells) EC50 [pM] Expression of CD69 with EC50 cytotoxicity % activation of CD4+ % activation of CD8+ hz20G6x7G3 CODV-Fab 50 0.6 30.9+3.6 73 66 7G3 x hz4B4 CODV-Fab 5, 4 o, 1 26.7 + 2.9 65 55 hz4B4x3E3 CODV-Fab 7.0 4.8 26.5+5.9 71 69 Fc-labeled variant TL4 : hz20G6x7G3 CODV-Fab- TL4 13, 9 1.7 16.7+10.1 73 74

The cytotoxic effects of CODV-Fab hz20G6xhz7G3, CODV-Fab-TL1 hz20G6xhz7G3, CODV-Fab-OL1 hz20G6xhz7G3 and single chain Dart MGD006 were also evaluated. The affinities of each bispecific construct for the CD3ε/δ complex and CD123 were measured using Biacore. In addition, a cytotoxicity assay was performed as described above, and CD4+ activation and CD8+ activation were measured.

Table 24. Affinities and activities of bispecific CODV molecules for CD123xCD3 and DART (MGD006)

Bispecific molecule KD (CD3e/d) [hM] KD (CD123) [hM] Cytotoxicity assay (ΤΗΡ cells) EC50 [pM] η= 3 hz2 0G6xhz7G3 CODV-Fab 9 0.2 2.1+/-0.2

- 110 041572

hz2 0G6xhz7G3 CODV-Fab-TLl eleven 0.2 1.0+/-0.1 hz20G6xhz7G3 CODV-Fab-OLl 15 0.4 0.9+/-0.1 Single chain antibody DART format, MGD006 9 0.2 0.3+/-0.04

A novel assay was used to assess the potential of molecules to trigger T cell activation in the presence (desirable) and absence (desirable) of target cells. NFAT-RE-luc2 Jurkat cells (cells from Promega no. CS176403) were incubated with freshly isolated human T cells at a 1:1 effector-target ratio at 37°C and 5% CO ₂ in RPMI 1640 with 2 g/L (11 mM) glucose, with GlutaMAX, with 25 mM HEPES in 386-well plates. After 5 hours, the incubation was stopped and luminescence measured using a Bio-Glo luciferase assay system, Promega No. G7 940, in an HTS luminescence microplate reader.

Table 25. T cell activation measured as activation induced by CODV molecules to CD123xCD3 and MGD006 measured in cell line with Jurkat-NFAT-Luc reporter gene

Bispecific molecule With THP1 cells, EC50 (pM) n= 3 Without target cells Activation at Cmax relative to maximum activation in the assay with target cells (%) n= 3 hz2 0G6xhz7G3 CODV-Fab 561+0.2 0.2+0.1 hz2 0G6xhz7G3 CODV-Fab-TLl 444+0.2 0.4+0.3 hz20G6xhz7G3 CODV-Fab-OLl 320+0.2 0.4+0.3 Single chain antibody DART format, MGD006 370±0.2 25.1±9.9

The results shown in table. 25 indicate that all antibodies induce the activation of reporter gene cells with EC50 values below nM in the presence of target cells. For T cell recruitment approaches, T cell activation should only occur in the presence of target cells. This is observed in the case of CODV molecules, since there is no significant luminescence signal in the absence of target cells. In contrast, a single-stranded DART molecule induces higher activation of a cell line with a reporter gene in the absence of target cells. These results are consistent with those obtained with primary T cells.

3.8. In vivo antitumor activity of CD123xCD3 bispecific CODV-Fab-TL1 and CD123xCD3 bispecific CODV-Fab.

Materials and methods

Isolation of PBMCs and Human T Cells from Whole Blood PBMCs were isolated from whole blood of healthy human donors by Ficoll gradient centrifugation. Whole blood was diluted 1:1 in sterile phosphate buffered saline (PBS). Then two volumes of thirty-five ml of the diluted blood were placed in two 50-ml Falcon tubes in the presence of 15 ml of Ficoll-Pak. The tubes were centrifuged at 200 g for 40 min at room temperature without braking. Two buffy coats were removed and placed in six 50 ml Falcon tubes with 45 ml sterile PBS and centrifuged three times (between each centrifugation, the supernatant was decanted and 45 ml PBS was added) at 100 g for ten minutes at room temperature without braking. After the last centrifugation, the two pellets were pooled in a final volume of 50 ml supplemented with PBS in a 50 ml Falcon tube. The total number of viable PBMCs was determined by counting in Vicell. The pellet was then reconstituted in Automacs run buffer from Myltenyi Biotech (130091-221) and T cells were separated from PBMCs using a negative selection kit from Miltenyi Biotech (130-091-156) and Automacs according to the manufacturer's instructions. Purified T cells were recovered and placed in culture in Xvivo-15 medium, 5% HIS + penicillin-streptomycin 1X at a concentration of 2.5 x 10E + 6 cells/ml.

Reproduction of human T-cells.

The T cell-rich population of human cells was activated and expanded in vitro for 14 days using a T cell activation/expansion kit from Miltenyi Biotech (130-091441).

Preparation of human T cells for in vivo administration.

Cells and cell culture medium were centrifuged for 10 min at 400 g. sediment uprising

- 111 041572 was infused at a concentration of 2x10E+7 cells/ml in sterile PBS. Removal of activating beads from expanded T cells was performed using a MACsiMAG separator from Myltenyi Biotech (130092-168) according to the manufacturer's instructions. The number of cells in the T-cell enriched cell populations was determined by Vicell counting and reconstituted in 25 ml sterile PBS in a 50 ml Falcon tube. After a centrifugation step at 400 g for 10 min at room temperature, the cell pellet was reconstituted in an appropriate volume of sterile PBS to obtain a final concentration of 5x10E+7 cells/ml.

tumor model.

Molm-13 human acute myeloid leukemia cells expressing CD123 were obtained from DSMZ - German Collection of Microorganisms and Cell Cultures at the Leibniz Institute (DSMZ, Braunschweig, Germany). Cells were grown in culture (37° C., 5% CO ₂ , 95% humidity) in RPMI1640 medium from Glutamax (supplemented with 20% fetal bovine serum). Molm-13 cells were infected with a vector containing a luciferase gene (SV40-PGL4-Puro - i.e. a vector containing a luciferase gene consisting of the simian virus 40 promoter linked to a luciferase 2 sequence and a puromycin resistance cassette sequence) carried by a non-replicative lentivirus .

Molm13-luc+ tumor cells were injected intravenously (IV) in NOD.Cg-Prkdcscid Il2rgtm1WjI/SzJ NSG mice (10E+6 cells per animal in 200 μl suspension in PBS). Twenty-four hours later, the same mice were injected intraperitoneally (IP) with 10E+7 human T cells in a volume of 0.2 ml of sterile PBS.

Initial bioluminescent imaging on the third day after tumor implantation was performed using an IVIS100 imaging device (PerkinElmer, Waltham, MA, USA) with Living Image 3.2 data acquisition software (Perkin-Elmer, Waltham, MA, USA). 15 min prior to imaging, IP animals were injected with a 120 mg/kg solution of Beetle luciferin potassium salt (batch #316019, Promega, Lyon, France) in PBS. Mice were anesthetized with Ketamine®/Xylazine® (120 mg/kg; 6 mg/kg IM, 5 ml/kg) 5 min prior to imaging.

Treatment with CODV-Fab-TL1 hz2 0G6xhz7G3, CODV-Fab hz2 0G6xhz7G3, anti-CD123xCD3 DART bispecific competitive antibody (single chain antibody in DART format MGD006 or a close analogue, referred to herein as DART agent) or PBS via intravenous (IV) or continuous intraperitoneal infusion (CIP) was started on day four after tumor implantation on established tumors that were already present in the bones, as indicated in Table 1. 26 (CODV-Fab-TL1 hz20G6xhz7G3), tab. 27 (CODV-Fab hz20G6xhz7G3 IV) and tab. 28 (CODV-Fab hz20G6xhz7G3 CIP).

Table 26 Study design to evaluate intravenous (IV) administration of anti-CD123xCD3 bispecific CODV-Fab-TL1

Treatment group Dose (nmol/kg) Volume / route of administration Scheme Quantity about animals Control - - - 7 CODV-Fab-TLl hz20G6xhz7G3 Π 1.3 0.2ml IV Q3d(4,7,10) 7 CODV-Fab-TLl hz20G6xhz7G3 Π 0.13 0.2ml IV Q3d(4,7,10) 8 CODV-Fab-TLl hz20G6xhz7G3 Π 0.013 0.2ml IV Q3d(4,7,10) 6 CODV-Fab hz20G6xhz7G3 Π 1.3 0.2ml IV Qd (4-13) 8 Single-stranded e antibody in DART format, MGD006 1.3 0.2ml IV Qd (4-13) 7

- 112 041572

Table 27. Study design to evaluate intravenous (IV) administration of anti-CD123xCD3 bispecific CODV-Fab

Treatment group Dose (nmol/kg) Volume / route of administration Scheme Number of animals Control - - - 5 CODV-Fab hz20G6xhz7G3 1.3 0.2ml IV Qd (4-13) 8 CODV-Fab hz20G6xhz7G3 0.13 0.2ml IV Qd (4-13) 5 Single chain antibody DART format, DART tool 1.3 0.2ml IV Qd (4-13) 8 Single chain antibody DART format, DART tool 0.13 0.2ml IV Qd (4-13) 7 CODV-Fab hz20G6x7G3 1.3 0.2ml IV Qd (4-13) 8

Table 28. Study design to evaluate continuous intraperitoneal infusion (CIP) of anti-CD123xCD3 bispecific CODV-Fab

Treatment group Dose (nmol/kg/day) Volume / day / route of administration Scheme Number of animals Media carrier - 6 µl/day, ip, Alzet 1002 Continuous infusion (414) 9 CODV-Fab hz2 0G6xhz7G 3 3, 9 6 µl/day, ip, Alzet 1002 Continuous infusion (414) 9 CODV-Fab hz20G6xhz7G 3 1.3 6 µl/day, ip, Alzet 1002 Continuous infusion (414) 10 Single chain oe antibody in DART format, MGD006 3, 9 6 µl/day, ip, Alzet 1002 Continuous infusion (414) 9 Single chain oe antibody in DART format, MGD006 1.3 6 µl/day, ip, Alzet 1002 Continuous infusion (414) 9 CODV-Fab hz20G6xhz7G 3 1.3 0.2 ml IV Qd (4-13) 10 Media carrier - 6 µl/day, ip, Alzet 1002 Continuous infusion (414) 10 CODV-Fab hz2 0G6xhz7G 3 0.13 6 µl/day, ip, Alzet 1002 Continuous infusion (414) 8 CODV-Fab hz20G6xhz7G 3 0.013 6 µl/day, ip, Alzet 1002 Continuous infusion (414) 8 Single chain oe antibody in DART format, MGD006 0.13 6 µl/day, ip, Alzet 1002 Continuous infusion (414) 8 Single chain oe antibody in DART format, MGD006 0.013 6 µl/day, ip, Alzet 1002 Continuous infusion (414) 9 CODV-Fab hz2 0G6xhz7G 3 1.3 0.2 ml IV Qd (4-13) 10

Data collection and performance criteria.

The body weight of the animals was monitored from day 3 until the end of the analysis to track the effect of therapy. A dosage resulting in 20% weight loss or 15% weight loss for 3 consecutive days or 10% or more drug-related deaths was considered an overly toxic dosage. The body weight of the animals included the mass of the tumor.

Tumor burden was monitored using non-invasive bioluminescent imaging (BLI). The initial BLI was performed on the third day after tumor implantation, 24 hours before the start of treatment. Animals were divided into different groups based on the whole body bioluminescence signal. Tumor growth was monitored throughout the body and long bones of the hind legs by measuring BLI signals on days 7, 10 and 14 post tumor implantation. Long bone signal was measured by segmentation and could be influenced by nearby locoregional signal (eg residual soft tissue signal at late time points). The treatment groups were compared with control animals carrying disseminated Molm13-luc+ tumor and human T cells.

The primary efficacy end points were the ratio of tumor signal change from baseline between treatment and control groups (dT/dC), number of partial tumor regressions (PR), and number of complete tumor regressions (CR).

Tumor growth based on bioluminescence signal (expressed in photon/s) versus time curves was monitored for each animal in each treatment group and presented as a median ± MAD curve for both whole body and segmented bone signals. Changes in tumor bioluminescence signal were calculated for each control (C) or treated (T) animal and for each day by subtracting the tumor signal on the first day of treatment (staging day) from the tumor signal on the designated observation day. Median T was calculated for the treatment group and median C was calculated for the control group.

The T/C ratio was then calculated and expressed as a percentage:

dT / dC \u003d [(median T, day of observation - median T, day 3) / (median C, day of observation - median C, day 3)] x 100.

A dose was considered therapeutically active when dT/dC at the end of the experiment (day 14) was less than 42% and very active when dT/dC was less than 10%.

Percent tumor regression was defined as the % reduction in tumor signal in the treatment group on the designated day of observation compared to its signal on the first day of treatment. At a specific time point and for each animal, % regression was calculated as % Signal _t0 - Signal _t regression (at time t)=----Signal------x 100

Considering the risk of signal fluctuations associated with the kinetics of the luciferin conversion reaction and the possible omission of an IP injection, signal regression in an animal was considered true tumor regression only if it occurred at least two consecutive time points.

Partial regression (PR). Regressions were defined as partial if the tumor signal decreased compared to the signal at the start of treatment for two consecutive time points, while the signal was still greater than 50% of the initial signal.

Complete regression (CR). Regressions were defined as complete if the tumor signal was reduced by more than 50% compared to the signal at the start of treatment for two consecutive time points.

Statistical analysis.

Evaluation of compounds administered by the IV route.

The individual bioluminescence signal of each treatment group was compared to the others using a Holm-Bonferroni adjustment for multiple pairwise comparisons followed by two-way repeated measures daily ANOVA: p > 0.05: NS, 0.05 > p > 0.01: * , p<0.01: **. Statistical analysis was performed for both whole body bioluminescence signals and long bone bioluminescence signals.

Evaluation of Compounds Injected with CIP

Evaluation of CODV-Fab hz20G6xhz7G3 administered by CIP resulted in a pooling of data from two independent studies (1st study for high compound dosages, 2nd study for low dosages, both studies including a vehicle control group , and a positive control group treated with CODV-Fab hz20G6xhz7G3, 1.3 nmol/kg, IV, Qd). Statistical analysis of the bioluminescence signal in each mouse each day was performed after normalizing the data to the mean value of the bioluminescence signal in the vehicle medium group on the same day of the same experiment (pooled vehicle controls, n=19; pooled positive controls, n=20). The individual normalized bioluminescence signal of each treatment group was compared to that of the other treatment groups using a Holm-Bonferroni adjustment for multiple pairwise comparisons followed by two-way repeated measures daily ANOVA: p > 0.05: NS, 0.05 > p > 0, 01: *, p<0.01: **. Statistical analysis was performed for both whole body bioluminescence signals and long bone bioluminescence signals.

results

IV introduction of bispecific CODV-Fab-TL1 to CD123xCD3 hz20G6xhz7G3.

Fully human CODV-Fab-TL1 hz20G6xhz7G3 administered IV, Q3d in the presence of human T cells inhibited Molm13 tumor growth at all doses tested (1.3, 0.13 and 0.013 nmol/kg Q3d) with dT/dC of 20.14 and 38%, respectively, for the whole body (Fig. 5 and 7), and was associated with tumor regression in long bones at all tested doses with 4/7 CR, 6/8 CR and 2/6 CR co- 114 -

Claims (15)

responsibly (FIGS. 6 and 8). The maximum response for fully human CODV-Fab-TL1 hz20G6xhz7G3 was obtained in the whole body and bones at a dose of 0.13 nmol/kg, Q3d. At this dose, activity was not statistically different from DART administered at 1.3 nmol/kg, IV, Qd (whole body dT/dC 29% with 1/7CR and 1/7PR in long bones) and was equivalent to activity CODV-Fab hz20G6xhz7G3 administered at 1.3 nmol/kg, IV, Qd (whole body dT/dC 23% with tumor regression in long bones 1/8CR and 1/8PR). Data were confirmed by final histopathological analysis (not shown). The differences observed between whole body and long bone data are due to residual tumor growth in the ovaries and abdominal fat due to extramedullary tumor dissemination after IV injection. IV introduction of a bispecific CODV-Fab to CD123xCD3 hz20G6xhz7G3. Fully human CODV-Fab hz20G6xhz7G3 administered IV in the presence of human T cells inhibited tumor growth at all doses tested (1.3 and 0.13 nmol/kg in Qd4-13), with dT/dC being 14 and 39%, respectively (FIGS. 9 and 11) was associated with tumor regression at 1.3 nmol/kg in long bones with 5/8 CR (FIGS. 10 and 12). DART given at 1.3 nmol/kg, IV, Qd4-13, suppressed tumor growth at 29% whole body dT/dC and 3/8 CR tumor regression in long bones, not significantly different from CODV-Fab values hz20G6xhz7G3 administered at a dose of 1.3 nmol/kg, IV. DART was inactive in suppressing whole body tumor signal when administered at 0.13 nmol/kg, IV, Qd4-13 (dT/dC 62%) despite regression of 1/7PR tumor in long bones. No significant difference was observed at the same dosage of fully human CODV-Fab hz20G6xhz7G3 at the end of the study. No statistically significant difference was noted between the fully human CODV-Fab hz20G6xhz7G3 and the partially humanized CODV-Fab hz20G6x7G3 compound when administered at the same dose of 1.3 nmol/kg, IV, QD4-13: at this dose, CODV-Fab hz20G6x7G3 inhibited whole body tumor growth with a dT/dC of 34% was associated with regression of tumors in long bones (1/8CR and 1/8PR). The differences observed between whole body and long bone data are due to residual tumor growth in the ovaries and abdominal fat due to extramedullary tumor dissemination after IV injection. CIP introduction of a bispecific CODV-Fab to CD123xCD3. Fully human CODV-Fab hz20G6xhz7G3 administered by CIP, in the presence of human T cells, inhibited whole body tumor growth at 3.9, 1.3, and 0.13 nmol/kg/day, CIP4-14, and was inactive at 0.013 nmol/kg/day with corresponding dT/dC of 2, 3, 21 and 57% www (FIGS. 13 and 15). It was associated with tumor regression at 3.9, 1.3, and 0.13 nmol/kg in long bones with 5/9CR 3/9PR, 6/10CR 1/10PR, and 7/8CR, respectively (Fig. 14 and 16). DART given at 3.9, 1.3 and 0.13 nmol/kg/day, CIP4-14, but not at 0.013 nmol/kg/day, inhibited whole body tumor growth with a dT/dC of 21 , 5, 21, and 46%, respectively (FIG. 13), inducing tumor regression in long bones at 3.9, 1.3, and 0.13 nmol/kg with 3/9 CR 3/9PR; 8/9CR and 4/8CR 2/8PR, respectively (Fig. 14). Fully human CODV-Fab hz20G6xhz7G3, administered at 1.3 nmol/kg, IV, Qd4-13, inhibited whole body tumor growth with dT/dC of 4% and 5% (studies 1 and 2, respectively) was associated with tumor regression in long bones (8/10CR 1/10PR versus 8/10 CR in the 1st and 2nd studies, respectively). CLAIM 1. An antibody-like binding protein that specifically binds to human CD3ε and at least one additional target antigen, which is CD123, containing two polypeptide chains that form two antigen-binding sites, where the first polypeptide has the structure represented by the formula [ I] VD1-L1-VD2-L2-CL[I]; and the second polypeptide has the structure represented by the formula [II] ^V d3 ^-L 3 ^-V D4 ^-L 4 ^-C H1 [And], where V _D1 represents the variable domain of the heavy or light chain of the first immunoglobulin; V _D2 is the heavy or light chain variable domain of the second immunoglobulin; V _D3 is the heavy or light chain variable domain of said second immunoglobulin; V _D4 is the heavy or light chain variable domain of said first immunoglobulin; C L is an immunoglobulin light chain constant domain; - 115 041572 Chi is the C _h1 constant domain of an immunoglobulin heavy chain; L1, L ₂ , L ₃ and L ₄ are amino acid linkers; and where the first and second polypeptides form a light chain-heavy chain crossover pair, and where both V _D1 and V _D2 are either light chain variable domains or heavy chain variable domains, and both V _D3 and V _D4 are heavy chain variable domains if V _D1 and V _D2 are light chain variable domains, or both V _D3 and V _D4 are light chain variable domains if V _D1 and V _D2 are heavy chain variable domains; where the first or second immunoglobulin is an antibody that binds to the extracellular domain of the human CD3ε protein and which contains a heavy chain variable domain containing CDR1-H with the amino acid sequence of SEQ ID NO: 6, CDR2-H with the amino acid sequence of SEQ ID NO: 7, CDR3-H with the amino acid sequence of SEQ ID NO: 8 and a light chain variable domain comprising CDR1-L with the amino acid sequence of SEQ ID NO: 142, CDR2-L with the amino acid sequence of KVS and CDR3-L with the amino acid sequence of SEQ ID NO: 11; and wherein the other of the first or second immunoglobulins is an anti-CD123 antibody that contains a heavy chain variable domain comprising CDR1-H with the amino acid sequence of SEQ ID NO: 381, CDR2-H with the amino acid sequence of SEQ ID NO: 384, CDR3-H with amino acid sequence of SEQ ID NO: 382 and a light chain variable domain containing CDR1-L with the amino acid sequence of SEQ ID NO: 378, CDR2-L with the amino acid sequence of WAS and CDR3-L with the amino acid sequence of SEQ ID NO: 379. 2. The antibody-like binding protein of claim 1, wherein the polypeptide of formula [II] further comprises an Fc domain. 3. Antibody-like binding protein according to claim 1 or 2, containing two polypeptide chains that form two antigen-binding sites, where one polypeptide chain has the structure represented by the formula [I] ^V D1 ^-L 1 ^-V D2 ^-L 2 ^-C L Hi and the second polypeptide has the structure represented by the formula [III] ^V d3 ^-L 3 ^-V D4 ^{-L 4} -CH1 ^-F c ^[ W], where V _D1 represents the variable domain of the heavy or light chain of the first immunoglobulin; V _D2 is the heavy or light chain variable domain of the second immunoglobulin; V D3 is the heavy or light chain variable domain of said second immunoglobulin; V D4 is the heavy or light chain variable domain of said first immunoglobulin; C _l is an immunoglobulin light chain constant domain; C _h1 is the C _h1 constant domain of an immunoglobulin heavy chain; F _c is the immunoglobulin hinge region and the CH2, _CH3 constant domains of the immunoglobulin heavy chain; L1, L ₂ , L ₃ and L4 are amino acid linkers; and wherein the polypeptide of formula [I] and the polypeptide of formula [III] form a light chain-heavy chain crossover pair. 4. The antibody-like binding protein according to claim 2 or 3, wherein the polypeptide of formula [I] further comprises an Fc domain (F _c2 ). 5. Antibody-like binding protein according to claim 2 or 4, containing two polypeptide chains that form two antigen-binding sites, where one polypeptide chain has the structure represented by the formula [IV] Vdi-Li-Vd2-L ₂ -Cl- L5-Fc2 [IV]; and the first polynucleotide chain has the structure represented by the formula [III] ^V D3- ^L 3 ^-V D4 ^-L 4 ^-C H1 ^-F c ^[ W ^] , where V _D1 represents the variable domain of the heavy or light chain of the first immunoglobulin; V _{d 2} is the heavy or light chain variable domain of the second immunoglobulin; VD 3 is the heavy or light chain variable domain of said second immunoglobulin; VD 4 is the heavy or light chain variable domain of said first immunoglobulin; C _l is an immunoglobulin light chain constant domain; C _h1 is the C _h1 constant domain of an immunoglobulin heavy chain; _Fc is the immunoglobulin hinge region and CH2, _CH3 constant domains of the immunoglobulin heavy chain; F _c2 represents the immunoglobulin hinge region and the CH2, _CH3 constant domains of the immunoglobulin heavy chain; L ₁ , L ₂ , L ₃ , L4 and L ₅ are amino acid linkers; and wherein the polypeptide of formula [IV] and the polypeptide of formula [III] form a light chain-heavy chain crossover pair. 6. An antibody-like binding protein according to claim 2 or 3, which contains a third polypeptide chain containing an Fc domain (F _c3 ). 7. Antibody-like binding protein according to any one of claims 2, 3 or 6, containing three polypeptide chains that form two antigen-binding sites, where the first polypeptide has the structure of formula [I] Vdi-Li-Vd2-L2-Cl [I]; the second polypeptide has the structure represented by the formula [III] Vd3-L3-Vd4-L4-Chi-Fc [III]; the third polypeptide F _c3 represents the hinge region of the immunoglobulin, and CH ₂ , CH ₃ represent the constant domains of the heavy chain of the immunoglobulin, where V _Di represents the variable domain of the heavy or light chain of the first immunoglobulin; V _D2 is the heavy or light chain variable domain of the second immunoglobulin; V _D3 is the heavy or light chain variable domain of said second immunoglobulin; V _d4 is the heavy or light chain variable domain of said first immunoglobulin; CL is an immunoglobulin light chain constant domain; CHi is the CHi constant domain of an immunoglobulin heavy chain; F _c represents the hinge region of the immunoglobulin, and constant domains CH ₂ , CH ₃ of the immunoglobulin heavy chain; L ₁ , L ₂ , L ₃ and L4 are amino acid linkers, where the polypeptide of formula [I] and the polypeptide of formula [III] form a light chain-heavy chain crossover pair; and where the polypeptide of formula [III] heterodimerizes with the third polypeptide through its Fc domain. 8. An antibody-like binding protein according to any one of claims 1 to 7, wherein V _D1 and V _D2 are light chain variable domains and V _D3 and V _d4 are heavy chain variable domains. 9. Antibody-like binding protein according to any one of claims 1-8, where V D1 contains CDR1-L with the amino acid sequence of SEQ ID NO: 378, CDR2-L with the amino acid sequence WAS, and CDR3-L with the amino acid sequence of SEQ ID NO: 379; V _D2 contains CDR4-L with the amino acid sequence of SEQ ID NO: 142, CDR5-L with the amino acid sequence KVS, and CDR6-L with the amino acid sequence of SEQ ID NO: 11; V D3 contains CDR1-H with the amino acid sequence of SEQ ID NO: 6, CDR2-H with the amino acid sequence of SEQ ID NO: 7, and CDR3-H with the amino acid sequence of SEQ ID NO: 8; And V _d4 contains CDR4-H with the amino acid sequence of SEQ ID NO: 381, CDR5-H with the amino acid sequence of SEQ ID NO: 384, and CDR6-H with the amino acid sequence of SEQ ID NO: 382. 10. The antibody-like binding protein of claim 9, wherein the antibody-like binding protein comprises V _D1 consisting of the amino acid sequence of SEQ ID NO: 385 or an amino acid sequence that is at least 85% identical to it, VD2, consisting of the amino acid sequence of SEQ ID NO: 141 or an amino acid sequence that is at least 85% identical to it, V _D3 consisting of the amino acid sequence of SEQ ID NO: 138 or an amino acid sequence that is at least 85% identical to it, and V _d4 consisting of the amino acid sequence of SEQ ID NO: 383 or an amino acid sequence that is at least 85% identical to it. 11. An antibody-like binding protein according to claim 10, which contains a linker L ₁ with the sequence of SEQ ID NO: 389, L2 with the sequence of SEQ ID NO: 389, L ₃ and L ₄ , consisting of 0 amino acids - - 117 041572 lot, and C _H1 with the sequence of SEQ ID NO: 313. 12. An antibody-like binding protein according to claim 11, which further comprises an L ₅ linker consisting of 0 amino acids, an Fc with the sequence of SEQ ID NO:394, an Fc ₂ with the sequence of SEQ ID NO: 392. 13. An antibody-like binding protein according to any one of claims 1, 2, 4 or 6-12, which contains: a) one polypeptide of formula [I], consisting of the amino acid sequence of SEQ ID NO: 388, which contains V _D1 with the amino acid sequence SEQ ID NO: 385, L ₁ with the amino acid sequence of SEQ ID NO: 389, V _D2 with the amino acid sequence of SEQ ID NO: 141, L2 with the amino acid sequence of SEQ ID NO: 389, and C _L with the amino acid sequence of SEQ ID NO: 310, or from a sequence that is at least 85% identical to the amino acid sequence of SEQ ID NO: 388, which contains CDR1-L with the amino acid sequence of SEQ ID NO: 378, CDR2-L with the amino acid sequence WAS, and CDR3-L with the amino acid sequence of SEQ ID NO: 379, CDR4-L with the amino acid sequence of SEQ ID NO: 142, CDR5-L with the amino acid sequence KVS, and CDR6-L with the amino acid sequence of SEQ ID NO: 11; And b) one polypeptide of formula [II], consisting of the amino acid sequence of SEQ ID NO: 390, which contains V _D3 with the amino acid sequence of SEQ ID NO: 138, L ₃ which has 0 amino acids, VD4 with the amino acid sequence of SEQ ID NO: 383, L4, which has 0 amino acids, and C _H1 with the amino acid sequence of SEQ ID NO: 313, or from a sequence that is at least 85% identical to the amino acid sequence of SEQ ID NO: 390, which contains CDR1-H with the amino acid sequence of SEQ ID NO: 381, CDR2-H with the amino acid sequence of SEQ ID NO: 384, CDR3-H with the amino acid sequence of SEQ ID NO: 382, and CDR4-H with the amino acid sequence of SEQ ID NO: 6, CDR5 -H with the amino acid sequence of SEQ ID NO: 7, and CDR6-H with the amino acid sequence of SEQ ID NO: 8; and wherein the polypeptide of formula [I] and the polypeptide of formula [II] form a light chain-heavy chain crossover pair. 14. An antibody-like binding protein according to claim 13, which contains one polypeptide of formula [I], consisting of the amino acid sequence of SEQ ID NO: 388, and one polypeptide of formula [II], consisting of the amino acid sequence of SEQ ID NO: 390 . 15. An antibody-like binding protein according to claim 5, which contains: a) one polypeptide with the formula [IV], consisting of the amino acid sequence of SEQ ID NO: 391, which contains V _D1 with the amino acid sequence SEQ ID NO: 385, L1 with the amino acid sequence SEQ ID NO: 389, VD2 with the amino acid sequence SEQ ID NO: 141, L2 with the amino acid sequence SEQ ID NO: 389, CL with the amino acid sequence SEQ ID NO: 310, L5, which has 0 amino acids, and F _c2 with the amino acid sequence of SEQ ID NO: 392, or from a sequence that is at least 85% identical to the amino acid sequence of SEQ ID NO: 391, which contains CDR1-L with the amino acid sequence of SEQ ID NO: 378, CDR2-L with the amino acid sequence of WAS, CDR3-L with the amino acid sequence of SEQ ID NO: 379, CDR4-L with the amino acid sequence of SEQ ID NO: 142, CDR5-L with the amino acid sequence KVS, and CDR6-L with the amino acid sequence of SEQ ID NO: 11; And b) one polypeptide of formula [III], consisting of the amino acid sequence of SEQ ID NO: 393, where Vd ₃ with the amino acid sequence of SEQ ID NO: 138, L3, which has 0 amino acids, VD4 with the amino acid sequence SEQ ID NO: 383, L4, which has 0 amino acids, C _H1 with the amino acid sequence of SEQ ID NO: 313, and F _c with the amino acid sequence of SEQ ID NO: 394, - 118 -