WO2020102647A1 - Constructions d'anticorps agonistes à agrégation de récepteurs multivalentes et protéines de liaison à l'antigène - Google Patents

Constructions d'anticorps agonistes à agrégation de récepteurs multivalentes et protéines de liaison à l'antigène Download PDF

Info

Publication number
WO2020102647A1
WO2020102647A1 PCT/US2019/061660 US2019061660W WO2020102647A1 WO 2020102647 A1 WO2020102647 A1 WO 2020102647A1 US 2019061660 W US2019061660 W US 2019061660W WO 2020102647 A1 WO2020102647 A1 WO 2020102647A1
Authority
WO
WIPO (PCT)
Prior art keywords
domain
abp
seq
amino acid
antigen binding
Prior art date
Application number
PCT/US2019/061660
Other languages
English (en)
Inventor
Mandar BAWADEKAR
Matthew Joseph BISSEN
Bryan Glaser
Roland Green
Li QUFEI
Lucas Bailey
Original Assignee
Invenra Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Invenra Inc. filed Critical Invenra Inc.
Publication of WO2020102647A1 publication Critical patent/WO2020102647A1/fr

Links

Classifications

    • C—CHEMISTRY; METALLURGY
    • C07—ORGANIC CHEMISTRY
    • C07K—PEPTIDES
    • C07K16/00—Immunoglobulins [IGs], e.g. monoclonal or polyclonal antibodies
    • C07K16/18—Immunoglobulins [IGs], e.g. monoclonal or polyclonal antibodies against material from animals or humans
    • C07K16/28—Immunoglobulins [IGs], e.g. monoclonal or polyclonal antibodies against material from animals or humans against receptors, cell surface antigens or cell surface determinants
    • C07K16/2878—Immunoglobulins [IGs], e.g. monoclonal or polyclonal antibodies against material from animals or humans against receptors, cell surface antigens or cell surface determinants against the NGF-receptor/TNF-receptor superfamily, e.g. CD27, CD30, CD40, CD95
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61P—SPECIFIC THERAPEUTIC ACTIVITY OF CHEMICAL COMPOUNDS OR MEDICINAL PREPARATIONS
    • A61P35/00—Antineoplastic agents
    • C—CHEMISTRY; METALLURGY
    • C07—ORGANIC CHEMISTRY
    • C07K—PEPTIDES
    • C07K16/00—Immunoglobulins [IGs], e.g. monoclonal or polyclonal antibodies
    • C07K16/18—Immunoglobulins [IGs], e.g. monoclonal or polyclonal antibodies against material from animals or humans
    • C07K16/28—Immunoglobulins [IGs], e.g. monoclonal or polyclonal antibodies against material from animals or humans against receptors, cell surface antigens or cell surface determinants
    • C07K16/2803—Immunoglobulins [IGs], e.g. monoclonal or polyclonal antibodies against material from animals or humans against receptors, cell surface antigens or cell surface determinants against the immunoglobulin superfamily
    • C07K16/2809—Immunoglobulins [IGs], e.g. monoclonal or polyclonal antibodies against material from animals or humans against receptors, cell surface antigens or cell surface determinants against the immunoglobulin superfamily against the T-cell receptor (TcR)-CD3 complex
    • C—CHEMISTRY; METALLURGY
    • C07—ORGANIC CHEMISTRY
    • C07K—PEPTIDES
    • C07K2317/00—Immunoglobulins specific features
    • C07K2317/30—Immunoglobulins specific features characterized by aspects of specificity or valency
    • C07K2317/31—Immunoglobulins specific features characterized by aspects of specificity or valency multispecific
    • C—CHEMISTRY; METALLURGY
    • C07—ORGANIC CHEMISTRY
    • C07K—PEPTIDES
    • C07K2317/00—Immunoglobulins specific features
    • C07K2317/30—Immunoglobulins specific features characterized by aspects of specificity or valency
    • C07K2317/33—Crossreactivity, e.g. for species or epitope, or lack of said crossreactivity
    • C—CHEMISTRY; METALLURGY
    • C07—ORGANIC CHEMISTRY
    • C07K—PEPTIDES
    • C07K2317/00—Immunoglobulins specific features
    • C07K2317/30—Immunoglobulins specific features characterized by aspects of specificity or valency
    • C07K2317/35—Valency
    • C—CHEMISTRY; METALLURGY
    • C07—ORGANIC CHEMISTRY
    • C07K—PEPTIDES
    • C07K2317/00—Immunoglobulins specific features
    • C07K2317/50—Immunoglobulins specific features characterized by immunoglobulin fragments
    • C07K2317/55—Fab or Fab'
    • C—CHEMISTRY; METALLURGY
    • C07—ORGANIC CHEMISTRY
    • C07K—PEPTIDES
    • C07K2317/00—Immunoglobulins specific features
    • C07K2317/70—Immunoglobulins specific features characterized by effect upon binding to a cell or to an antigen
    • C07K2317/74—Inducing cell proliferation
    • C—CHEMISTRY; METALLURGY
    • C07—ORGANIC CHEMISTRY
    • C07K—PEPTIDES
    • C07K2317/00—Immunoglobulins specific features
    • C07K2317/70—Immunoglobulins specific features characterized by effect upon binding to a cell or to an antigen
    • C07K2317/75—Agonist effect on antigen

Definitions

  • TNFR tumor necrosis factor receptor family
  • etanercept Enbrel
  • anti-TNFa antibodies such as infliximab (Remicade)
  • adalimumab Humira
  • golimumab Simponi
  • pegylated Fab constructs, such as Certolizumab pegol (Cimzia).
  • Agonists of the TNF/TNFR signaling axis also have therapeutic potential. Croft et al., Nature Rev. Drug Discovery 12:147-168 (2013). However, agonists of the TNFR have been much less successful: effective TNFR activation is much more difficult to achieve than blocking of the TNF and TNFR interactions, because activation of TNFR generally requires specific oligomerization (clustering the receptor trimers) and immobilization.
  • Fusion proteins comprising two ligand trimers (a pseudo-hexamer) can be effective oligomerizing agonists, but the ligands are small cysteine-rich domains and their oligomers generally possess poor biophysical properties, making them inferior drugs compared to antibodies.
  • Conventional TNFR mAbs which are bivalent and monospecific for a single TNFR epitope– typically possess low-to moderate TNFR agonist activity. Accordingly, additional crosslinking of the TNFR mAbs is required to potentiate agonistic activity.
  • agents that induce MET clustering results in downregulation of MET oncogenic pathways. See, e.g., Li, Wenjing et al.“Induction of MET Receptor Tyrosine Kinase Down-regulation through Antibody-mediated Receptor Clustering” Scientific reports vol.9,11988.13 Feb.2019, doi:10.1038/s41598-018-36963-3, which is hereby incorporated by reference in its entirety.
  • an antibody construct capable of (i) binding a cell surface receptor target, and (ii) clustering the receptor target on the cell surface in the absence of an independent cross-linking agent.
  • antibody constructs having these characteristics and that are also capable of high level expression, such as bivalent and trivalent constructs, with high fidelity pairing of cognate heavy chain pairs and cognate heavy and light chain pairs and that can be readily purified.
  • an antibody construct capable of (i) binding a cell surface receptor target that requires clustering for agonist activity, and (ii) clustering the receptor target on the cell surface in the absence of an independent cross-linking agent.
  • antibody constructs having these characteristics and that are also capable of high level expression, such as bivalent and trivalent constructs, with high fidelity pairing of cognate heavy chain pairs and cognate heavy and light chain pairs and that can be readily purified.
  • Bispecific antibodies have been developed to redirect T-cells to cancer cells.
  • the CD19/CD3 BiTE blinatumomab is approved for treatment of B-cell acute lymphoblastic leukemia.
  • the promise of redirecting bispecifics has not been fully realized. It has been reported that many patients fail blinatumomab despite CD19 antigen expression on their cancer cells, for reasons that are not fully understood.
  • some studies suggest that lack of T-cell costimulation activity may contribute to poor efficacy.
  • ABSPs antigen binding proteins
  • a multivalent antibody construct wherein the construct is capable of (i) binding a cell surface receptor target, and (ii) clustering the receptor target on the cell surface in the absence of an independent cross-linking agent or one or more Fc mutations that drive hexamer formation, and wherein each of the target receptor-binding antigen binding sites of the construct is contributed by antibody variable region binding domains.
  • a multivalent ABP wherein the construct is capable of (i) binding a cell surface receptor target, optionally wherein the cell surface receptor target requires clustering for agonist activity, and (ii) clustering the receptor target on the cell surface in the absence of an independent cross-linking agent or one or more Fc mutations that drive hexamer formation, and wherein each of the target receptor-binding antigen binding sites of the construct is contributed by antibody variable region binding domains.
  • the construct is monospecific.
  • the construct is multispecific.
  • the construct comprises a first antigen binding site specific for a first epitope of the target receptor, and a second antigen binding site specific for a second antigenic target.
  • the second antigenic target is a second epitope of the target receptor, optionally wherein the first epitope and the second epitope are non-overlapping epitopes.
  • the second antigenic target is an epitope of a second protein.
  • the second protein is a second cell surface receptor.
  • the target cell surface receptor and the second cell surface receptor are commonly expressed on the surface of at least some mammalian cells.
  • the target receptor is a TNF Receptor superfamily (TNFRSF) member.
  • TNFRSF TNF Receptor superfamily
  • the target receptor is OX40 (TNFRSF4), CD40 (TNFRSF5), or 4-1BB (TNFRSF9).
  • the target receptor is a human TNFRSF.
  • the target receptor is human OX40, human CD40, or human 4- 1BB.
  • the target receptor is human OX40.
  • the construct is bivalent.
  • the bivalent construct is a bivalent (1x1) construct.
  • the construct is monospecific.
  • the construct is bispecific.
  • the second antigenic target is a second epitope of the target receptor, optionally wherein the first epitope and the second epitope are non-overlapping epitopes.
  • the second antigenic target is an epitope of a second protein.
  • the second protein is a second cell surface receptor.
  • the antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the construct is trivalent.
  • the construct is a trivalent (2x1) construct.
  • the construct is monospecific. [0041] In some embodiments, the construct is bispecific.
  • the construct contains one copy of the antigen binding site (ABS) specific for a first epitope of the target receptor.
  • ABS antigen binding site
  • the antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the construct contains two copies of the antigen binding site specific for a first epitope of the target receptor.
  • a first antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site and a second antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site and a second antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site and a second antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the second antigenic target is a second epitope of the target receptor, optionally wherein the first epitope and the second epitope are non-overlapping epitopes.
  • the second antigenic target is an epitope of a second protein.
  • the second protein is a second cell surface receptor.
  • the construct is trispecific.
  • the antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the second antigenic target is a second epitope of the target receptor, optionally wherein the first epitope and the second epitope are non-overlapping epitopes.
  • the second antigenic target is a first epitope of a second protein.
  • the third antigenic target is a third epitope of the target receptor.
  • the third antigenic target is a second epitope of a second protein, optionally wherein the first epitope of the second protein and the second epitope of the second protein are non-overlapping epitopes.
  • the third antigenic target is a first epitope of a third protein.
  • the second protein or third protein is a second or third cell surface receptor.
  • the construct is a trivalent (1x2) construct.
  • the construct is monospecific.
  • the construct is bispecific.
  • the construct contains one copy of the antigen binding site (ABS) specific for a first epitope of the target receptor.
  • ABS antigen binding site
  • the antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an R:T antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the construct contains two copies of the antigen binding site specific for a first epitope of the target receptor.
  • a first antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an R:T antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site and a second antigen binding site specific for a first epitope of the target receptor is an R:T antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site and a second antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an R:T antigen binding site and a second antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the second antigenic target is a second epitope of the target receptor, optionally the first epitope and the second epitope are non-overlapping epitopes.
  • the second antigenic target is an epitope of a second protein.
  • the second protein is a second cell surface receptor.
  • the construct is trispecific.
  • the antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an R:T antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the second antigenic target is a second epitope of the target receptor, optionally the first epitope and the second epitope are non-overlapping epitopes.
  • the second antigenic target is a first epitope of a second protein.
  • the third antigenic target is a third epitope of the target receptor.
  • the third antigenic target is a second epitope of a second protein, optionally the first epitope of the second protein and the second epitope of the second protein are non-overlapping epitopes.
  • the third antigenic target is a first epitope of a third protein.
  • the second protein or third protein is a second or third cell surface receptor.
  • the presence of an independent cross-linking agent does not increase agonist activity in vitro above that achieved in the absence of the independent cross-linking agent.
  • the presence of an independent cross-linking agent increases agonist activity in vitro above that achieved in the absence of the independent cross-linking agent.
  • the presence of an independent cross-linking agent increases agonist in vitro activity 50% above activity observed in the absence of the independent cross-linking agent.
  • the construct comprises: a first, second, third, and fourth polypeptide chain, wherein: (a) the first polypeptide chain comprises a domain A, a domain B, a domain D, and a domain E, wherein the domains are arranged, from N-terminus to C-terminus, in a A-B-D-E orientation, and domain A has a VL amino acid sequence, domain B has a CH3 amino acid sequence, domain D has a CH2 amino acid sequence, and domain E has a constant region domain amino acid sequence; (b) the second polypeptide chain comprises a domain F and a domain G, wherein the domains are arranged, from N-terminus to C-terminus, in a F-G orientation, and wherein domain F has a VH amino acid sequence and domain G has a CH3 amino acid sequence; (c) the third polypeptide chain comprises a domain H, a domain I, a domain J, and a domain K,
  • the presence of an independent cross- linking agent does not increase agonist activity in vitro above that achieved in the absence of the independent cross-linking agent.
  • the presence of an independent cross- linking agent increases agonist activity in vitro above that achieved in the absence of the independent cross-linking agent.
  • the presence of an independent cross- linking agent increases agonist in vitro activity 50% above activity observed in the absence of the independent cross-linking agent.
  • the construct comprises: a first, second, third, and fourth polypeptide chain, wherein: (a) the first polypeptide chain comprises a domain A, a domain B, a domain D, and a domain E, wherein the domains are arranged, from N-terminus to C-terminus, in a A-B-D-E orientation, and domain A has a VL amino acid sequence, domain B has a CH3 amino acid sequence, domain D has a CH2 amino acid sequence, and domain E has a constant region domain amino acid sequence; (b) the second polypeptide chain comprises a domain F and a domain G, wherein the domains are arranged, from N-terminus to C-terminus, in a F-G orientation, and wherein domain F has a VH amino acid sequence and domain G has a CH3 amino acid sequence; (c) the third polypeptide chain comprises a domain H, a domain I, a domain J, and a domain K,
  • polypeptides are associated through an interaction between the D and the J domains and an interaction between the E and the K domains to form the binding molecule.
  • amino acid sequences of the B and the G domains are identical, wherein the sequence is an endogenous CH3 sequence.
  • the amino acid sequences of the B and the G domains are different and separately comprise respectively orthogonal modifications in an endogenous CH3 sequence, wherein the B domain interacts with the G domain, and wherein neither the B domain nor the G domain interacts with a CH3 domain lacking the orthogonal modification.
  • the orthogonal modifications comprise mutations that generate engineered disulfide bridges between domain B and G.
  • the mutations that generate engineered disulfide bridges are a S354C mutation in one of the B domain and G domain, and a 349C in the other domain.
  • the orthogonal modifications comprise knob-in-hole mutations.
  • the knob-in hole mutations are a T366W mutation in one of the B domain and G domain, and a T366S, L368A, and aY407V mutation in the other domain.
  • the orthogonal modifications comprise charge-pair mutations.
  • the charge-pair mutations are a T366K mutation in one of the B domain and G domain, and a L351D mutation in the other domain.
  • the domain E has a CH3 amino acid sequence.
  • amino acid sequences of the E and K domains are identical, wherein the sequence is an endogenous CH3 sequence.
  • amino acid sequences of the E and K domains are different.
  • the different sequences separately comprise
  • the orthogonal modifications comprise mutations that generate engineered disulfide bridges between domain E and K.
  • the mutations that generate engineered disulfide bridges are a S354C mutation in one of the E domain and K domain, and a 349C in the other domain.
  • the orthogonal modifications in the E and K domains comprise knob-in-hole mutations.
  • knob-in hole mutations are a T366W mutation in one of the E domain or K domain and a T366S, L368A, and aY407V mutation in the other domain.
  • the orthogonal modifications comprise charge-pair mutations.
  • the charge-pair mutations are a T366K mutation in one of the E domain or K domain and a corresponding L351D mutation in the other domain.
  • the amino acid sequences of the E domain and the K domain are endogenous sequences of two different antibody domains, the domains selected to have a specific interaction that promotes the specific association between the first and the third polypeptides.
  • the two different amino acid sequences are a CH1 sequence and a CL sequence.
  • domain I has a CL sequence and domain M has a CH1 sequence.
  • domain H has a VL sequence and domain L has a VH sequence.
  • domain H has a VL amino acid sequence
  • domain I has a CL amino acid sequence
  • domain K has a CH3 amino acid sequence
  • domain L has a VH amino acid sequence
  • domain M has a CH1 amino acid sequence
  • the construct further comprises: a sixth polypeptide chain, wherein: (a) the third polypeptide chain further comprises a domain R and a domain S, wherein the domains are arranged, from N-terminus to C-terminus, in a R-S-H-I-J-K orientation, and wherein domain R has a VL amino acid sequence and domain S has a constant domain amino acid sequence; (b) the binding molecule further comprises a sixth polypeptide chain, comprising: a domain T and a domain U, wherein the domains are arranged, from N-terminus to C-terminus, in a T-U orientation, and wherein domain T has a VH amino acid sequence and domain U has a constant domain amino acid sequence; and (c) the third and the sixth polypeptides are associated through an interaction between the R and the T domains and an interaction between the S and the U domains to form the binding molecule.
  • the amino acid sequences of domain R and domain A are identical, the amino acid sequences of domain H is different from domain R and A, the amino acid sequences of domain S and domain B are identical, the amino acid sequences of domain I is different from domain S and B, the amino acid sequences of domain T and domain F are identical, the amino acid sequences of domain L is different from domain T and F, the amino acid sequences of domain U and domain G are identical, the amino acid sequences of domain M is different from domain U and G and (b) the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen, the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigen, and the domain R and domain T form a third antigen binding site specific for the first antigen.
  • the amino acid sequences of domain R and domain H are identical, the amino acid sequences of domain A is different from domain R and H, the amino acid sequences of domain S and domain I are identical, the amino acid sequences of domain B is different from domain S and I, the amino acid sequences of domain T and domain L are identical, the amino acid sequences of domain F is different from domain T and L, the amino acid sequences of domain U and domain M are identical, the amino acid sequences of domain G is different from domain U and M; and (b) the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen, the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigen, and the domain R and domain T form a third antigen binding site specific for the second antigen.
  • the amino acid sequences of domain R, domain A, and domain H are different, the amino acid sequences of domain S, domain B, and domain I are different, the amino acid sequences of domain T, domain F, and domain L are different, and the amino acid sequences of domain U, domain G, and domain M are different; and (b) the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen, the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigen, and the domain R and domain T form a third antigen binding site specific for a third antigen.
  • the construct further comprises: a fifth polypeptide chain, wherein: (a) the first polypeptide chain further comprises a domain N and a domain O, wherein the domains are arranged, from N-terminus to C-terminus, in a N-O-A-B-D-E orientation, and wherein domain N has a VL amino acid sequence, domain O has a CH3 amino acid sequence; (b) the binding molecule further comprises a fifth polypeptide chain, comprising: a domain P and a domain Q, wherein the domains are arranged, from N-terminus to C-terminus, in a P-Q orientation, and wherein domain P has a VH amino acid sequence and domain Q has a CH3 amino acid sequence; and (c) the first and the fifth polypeptides are associated through an interaction between the N and the P domains and an interaction between the O and the Q domains to form the binding molecule.
  • the amino acid sequences of domain N and domain A are identical, the amino acid sequences of domain H is different from domains N and A, the amino acid sequences of domain O and domain B are identical, the amino acid sequences of domain I is different from domains O and B, the amino acid sequences of domain P and domain F are identical, the amino acid sequences of domain L is different from domains P and F, the amino acid sequences of domain Q and domain G are identical, the amino acid sequences of domain M is different from domains Q and G; and (b) the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen, the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigen, and the domain N and domain P form a third antigen binding site specific for the first antigen.
  • the amino acid sequences of domain N, domain A, and domain H are different, the amino acid sequences of domain O, domain B, and domain I are different, the amino acid sequences of domain P, domain F, and domain L are different, and the amino acid sequences of domain Q, domain G, and domain M are different; and (b) the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen, the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigen, and the domain N and domain P form a third antigen binding site specific for a third antigen.
  • sequence that links the A domain and the B domain is IKRTPREP (SEQ ID NO: 36) or IKRTVREP (SEQ ID NO: 37).
  • sequence that links the F domain and the G domain is SSASPREP (SEQ ID NO: 42).
  • At least one CH3 amino acid sequence has a C-terminal tripeptide insertion linking the CH3 amino acid sequence to a hinge amino acid sequence, wherein the tripeptide insertion is selected from the group consisting of PGK, KSC, and GEC.
  • sequences are human sequences.
  • at least one CH3 amino acid sequence is an IgG sequence.
  • the IgG sequences are IgG1 sequences.
  • At least one CH3 amino acid sequence has one or more isoallotype mutations.
  • the isoallotype mutations are D356E and L358M.
  • the CL amino acid sequence is a Ckappa sequence.
  • an OX40 binding molecule comprising: a first antigen binding site specific for an OX40 antigen, wherein the first antigen binding site comprises: A) a CDR1, a CDR2, and a CDR3 amino acid sequences of a specific light chain variable region (VL), wherein the CDR1, CDR2, and CDR3 VL sequences are selected from Table 4 corresponding to a specific OX40 antigen binding site (ABS); and B) a CDR1, a CDR2, and a CDR3 amino acid sequences of a specific heavy chain variable region (VH), wherein the CDR1, CDR2, and CDR3 VH sequences are selected from Table 3 corresponding to the specific OX40 ABS.
  • VL light chain variable region
  • VH specific heavy chain variable region
  • the first antigen binding site is specific for a first epitope of the OX40 antigen.
  • the OX40 antigen comprises an OX40 domain selected from the group consisting of: OX40 amino acids 2-214, OX40 amino acids 66-214, OX40 amino acids 108-214, and OX40 amino acids 127-214.
  • the OX40 antigen comprises a human OX40 antigen.
  • the first antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:203, and a VH CDR1 comprising SEQ ID NO:83, a VH CDR2 comprising SEQ ID NO:123, and a VH CDR3 comprising SEQ ID NO:163.
  • the first antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:234, and a VH CDR1 comprising SEQ ID NO:83, a VH CDR2 comprising SEQ ID NO:123, and a VH CDR3 comprising SEQ ID NO:163.
  • the first antigen binding site comprises a VH sequence comprising SEQ ID NO: 227.
  • the first antigen binding site comprises a VL sequence comprising SEQ ID NO: 228 or 229.
  • the first antigen binding site comprises a VH sequence comprising SEQ ID NO: 227 and a VL sequence comprising SEQ ID NO: 228.
  • the first antigen binding site comprises a VH sequence comprising SEQ ID NO: 227 and a VL sequence comprising SEQ ID NO: 229.
  • the first antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:190, and a VH CDR1 comprising SEQ ID NO:70, a VH CDR2 comprising SEQ ID NO:110, and a VH CDR3 comprising SEQ ID NO:150.
  • the first antigen binding site comprises a VH sequence comprising SEQ ID NO: 230.
  • the first antigen binding site comprises a VL sequence comprising SEQ ID NO: 231.
  • the first antigen binding site comprises a VH sequence comprising SEQ ID NO: 230 and a VL sequence comprising SEQ ID NO: 231.
  • the OX40 antigen binding molecule further comprises a second antigen binding site.
  • the second antigen binding site is specific for the OX40 antigen.
  • the second antigen binding site is specific for the first epitope of the OX40 antigen.
  • the first and second antigen binding sites comprise a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:190, and a VH CDR1 comprising SEQ ID NO:70, a VH CDR2 comprising SEQ ID NO:110, and a VH CDR3 comprising SEQ ID NO:150.
  • the second antigen binding site is specific for a second epitope of the OX40 antigen.
  • the first epitope and the second epitope are non- overlapping epitopes.
  • the first antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:203, and a VH CDR1 comprising SEQ ID NO:83, a VH CDR2 comprising SEQ ID NO:123, and a VH CDR3 comprising SEQ ID NO:163; and the second antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:190, and a VH CDR1 comprising SEQ ID NO:70, a VH CDR2 comprising SEQ ID NO:110, and a VH CDR3 comprising SEQ ID NO:150.
  • the first antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:234, and a VH CDR1 comprising SEQ ID NO:83, a VH CDR2 comprising SEQ ID NO:123, and a VH CDR3 comprising SEQ ID NO:163; and the second antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:190, and a VH CDR1 comprising SEQ ID NO:70, a VH CDR2 comprising SEQ ID NO:110, and a VH CDR3 comprising SEQ ID NO:150.
  • the first antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:190, and a VH CDR1 comprising SEQ ID NO:70, a VH CDR2 comprising SEQ ID NO:110, and a VH CDR3 comprising SEQ ID NO:150; and the second antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:234, and a VH CDR1 comprising SEQ ID NO:83, a VH CDR2 comprising SEQ ID NO:123, and a VH CDR3 comprising SEQ ID NO:163.
  • the first antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:190, and a VH CDR1 comprising SEQ ID NO:70, a VH CDR2 comprising SEQ ID NO:110, and a VH CDR3 comprising SEQ ID NO:150; and the second antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:203, and a VH CDR1 comprising SEQ ID NO:83, a VH CDR2 comprising SEQ ID NO:123, and a VH CDR3 comprising SEQ ID NO:163.
  • the second antigen binding site is specific for a second antigen different from the OX40 antigen.
  • the second antigen is a second cell surface receptor.
  • the OX40 antigen binding molecule comprises an antibody format selected from the group consisting of: full-length antibodies, Fab fragments, Fvs, scFvs, tandem scFvs, Diabodies, scDiabodies, DARTs, tandAbs, and minibodies.
  • the OX40 antigen binding molecule comprises: a first and a second polypeptide chain, wherein: (a) the first polypeptide chain comprises a domain A, a domain B, a domain D, and a domain E, wherein the domains are arranged, from N- terminus to C-terminus, in a A-B-D-E orientation, wherein domain A has a variable region domain amino acid sequence, and wherein domain B, domain D, and domain E have a constant region domain amino acid sequence; (b) the second polypeptide chain comprises a domain F and a domain G, wherein the domains are arranged, from N-terminus to C- terminus, in a F-G orientation, and wherein domain F has a variable region domain amino acid sequence and domain G has a constant region domain amino acid sequence c) the first and the second polypeptides are associated through an interaction between the A and the F domain and an interaction between the B domain and the G domain to form the OX40 antigen binding molecule,
  • the OX40 antigen binding molecule further comprises: a third and a fourth polypeptide chain, wherein: (a) the third polypeptide chain comprises a domain H, a domain I, a domain J, and a domain K, wherein the domains are arranged, from N-terminus to C-terminus, in a H-I-J-K orientation, and wherein domain H has a variable region domain amino acid sequence, and domains I, J, and K have a constant region domain amino acid sequence; (b) the fourth polypeptide chain comprises a domain L and a domain M, wherein the domains are arranged, from N-terminus to C-terminus, in a L-M orientation, and wherein domain L has a variable region domain amino acid sequence and domain M has a constant region amino acid sequence; (c) the third and the fourth polypeptides are associated through an interaction between the H and the L domains and an interaction between the I and the M domains; and (d) the first and the third polypeptide chain
  • the first antigen binding site is specific for the OX40 antigen.
  • the second antigen binding site is specific for the OX40 antigen.
  • the first antigen binding site is specific for a first epitope of the OX40 antigen and the second antigen binding site is specific for a second epitope of the OX40 antigen.
  • the first epitope and the second epitope are non- overlapping epitopes.
  • domain B and domain G have a CH3 amino acid sequence.
  • amino acid sequences of the B domain and the G domain are identical, wherein the sequence is an endogenous CH3 sequence.
  • the amino acid sequences of the B domain and the G domain are different and separately comprise respectively orthogonal modifications in an endogenous CH3 sequence, the B domain interacts with the G domain, and neither the B domain nor the G domain significantly interacts with a CH3 domain lacking the orthogonal modification.
  • the orthogonal modifications of the B domain and the G domain comprise mutations that generate engineered disulfide bridges between the B domain and the G domain.
  • the mutations of the B domain and the G domain that generate engineered disulfide bridges are a S354C mutation in one of the B domain and G domain, and a 349C in the other domain.
  • the orthogonal modifications of the B domain and the G domain comprise knob-in-hole mutations.
  • the knob-in hole mutations of the B domain and the G domain are a T366W mutation in one of the B domain and G domain, and a T366S, L368A, and aY407V mutation in the other domain.
  • the orthogonal modifications of the B domain and the G domain comprise charge-pair mutations.
  • the charge-pair mutations of the B domain and the G domain are a T366K mutation in one of the B domain and G domain, and a L351D mutation in the other domain.
  • domain B and domain G have an IgM CH2 amino acid sequence or an IgE CH2 amino acid sequence.
  • the IgM CH2 amino acid sequence or the IgE CH2 amino acid sequence comprise orthogonal modifications.
  • domain I has a CL sequence and domain M has a CH1 sequence. In some embodiments, domain I has a CH1 sequence and domain M has a CL sequence. In some embodiments, the CH1 sequence and the CL sequence each comprise one or more orthogonal modifications, wherein a domain having the CH1 sequence does not significantly interact with a domain having a CL sequence lacking the orthogonal
  • the orthogonal modifications in the CH1 sequence and the CL sequence comprise mutations that generate engineered disulfide bridges between the at least one CH1 domain and a CL domain, the mutations selected from the group consisting of: an engineered cysteine at position 138 of the CH1 sequence and position 116 of the CL sequence; an engineered cysteine at position 128 of the CH1 sequence and position 119 of the CL sequence, and an engineered cysteine at position 129 of the CH1 sequence and position 210 of the CL sequence.
  • the orthogonal modifications in the CH1 sequence and the CL sequence comprise mutations that generate engineered disulfide bridges between the at least one CH1 domain and a CL domain, wherein the mutations comprise and engineered cysteines at position 128 of the CH1 sequence and position 118 of a CL Kappa sequence.
  • the orthogonal modifications in the CH1 sequence and the CL sequence comprise mutations that generate engineered disulfide bridges between the at least one CH1 domain and a CL domain, the mutations selected from the group consisting of: a F118C mutation in the CL sequence with a corresponding A141C in the CH1 sequence; a F118C mutation in the CL sequence with a corresponding L128C in the CH1 sequence; and a S162C mutations in the CL sequence with a corresponding P171C mutation in the CH1 sequence.
  • the orthogonal modifications in the CH1 sequence and the CL sequence comprise charge-pair mutations between the at least one CH1 domain and a CL domain, the charge-pair mutations selected from the group consisting of: a F118S mutation in the CL sequence with a corresponding A141L in the CH1 sequence; a F118A mutation in the CL sequence with a corresponding A141L in the CH1 sequence; a F118V mutation in the CL sequence with a corresponding A141L in the CH1 sequence; and a T129R mutation in the CL sequence with a corresponding K147D in the CH1 sequence.
  • the orthogonal modifications in the CH1 sequence and the CL sequence comprise charge-pair mutations between the at least one CH1 domain and a CL domain, the charge-pair mutations selected from the group consisting of: a N138K mutation in the CL sequence with a corresponding G166D in the CH1 sequence, and a N138D mutation in the CL sequence with a corresponding G166K in the CH1 sequence.
  • domain A has a VL amino acid sequence and domain F has a VH amino acid sequence.
  • domain A has a VH amino acid sequence and domain F has a VL amino acid sequence.
  • domain H has a VL amino acid sequence and domain L has a VH amino acid sequence.
  • domain H has a VH amino acid sequence and domain L has a VL amino acid sequence.
  • domain D and domain J have a CH2 amino acid sequence.
  • the E domain has a CH3 amino acid sequence.
  • amino acid sequences of the E domain and the K domain are identical, wherein the sequence is an endogenous CH3 sequence.
  • the amino acid sequences of the E domain and the K domain are different.
  • the different sequences separately comprise respectively orthogonal modifications in an endogenous CH3 sequence, wherein the E domain interacts with the K domain, and wherein neither the E domain nor the K domain significantly interacts with a CH3 domain lacking the orthogonal modification.
  • the orthogonal modifications comprise mutations that generate engineered disulfide bridges between the E domain and the K domain.
  • the mutations that generate engineered disulfide bridges are a S354C mutation in one of the E domain and the K domain, and a 349C in the other domain.
  • the orthogonal modifications in the E domain and the K domain comprise knob-in-hole mutations.
  • the knob-in hole mutations are a T366W mutation in one of the E domain or the K domain and a T366S, L368A, and aY407V mutation in the other domain.
  • the orthogonal modifications in the E domain and the K domain comprise charge-pair mutations.
  • the charge-pair mutations are a T366K mutation in one of the E domain or the K domain and a corresponding L351D mutation in the other domain.
  • the amino acid sequences of the E domain and the K domain are endogenous sequences of two different antibody domains, the domains selected to have a specific interaction that promotes the specific association between the first and the third polypeptides.
  • the two different amino acid sequences are a CH1 sequence and a CL sequence.
  • the OX40 antigen binding molecule further comprises a third antigen binding site.
  • the third antigen binding site is specific for an OX40 antigen.
  • the first antigen binding site and the third antigen binding site are specific for the same OX40 antigen.
  • the first antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:203, and a VH CDR1 comprising SEQ ID NO:83, a VH CDR2 comprising SEQ ID NO:123, and a VH CDR3 comprising SEQ ID NO:163.
  • the first antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:234, and a VH CDR1 comprising SEQ ID NO:83, a VH CDR2 comprising SEQ ID NO:123, and a VH CDR3 comprising SEQ ID NO:163.
  • the first antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:190, and a VH CDR1 comprising SEQ ID NO:70, a VH CDR2 comprising SEQ ID NO:110, and a VH CDR3 comprising SEQ ID NO:150.
  • the first antigen binding site comprises a VH sequence comprising SEQ ID NO: 230. In some embodiments, the first antigen binding site comprises a VL sequence comprising SEQ ID NO: 231. In some embodiments, the first antigen binding site comprises a VH sequence comprising SEQ ID NO: 230 and a VL sequence comprising SEQ ID NO: 231.
  • the second antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:234, and a VH CDR1 comprising SEQ ID NO:83, a VH CDR2 comprising SEQ ID NO:123, and a VH CDR3 comprising SEQ ID NO:163.
  • the second antigen binding site comprises a VH sequence comprising SEQ ID NO: 227. In some embodiments, the second antigen binding site comprises a VL sequence comprising SEQ ID NO: 228. In some embodiments, the second antigen binding site comprises a VL sequence comprising SEQ ID NO: 229. In some embodiments, the second antigen binding site comprises a VH sequence comprising SEQ ID NO: 227 and a VL sequence comprising SEQ ID NO: 228. In some embodiments, the second antigen binding site comprises a VH sequence comprising SEQ ID NO: 227 and a VL sequence comprising SEQ ID NO: 229.
  • the second antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:203, and a VH CDR1 comprising SEQ ID NO:83, a VH CDR2 comprising SEQ ID NO:123, and a VH CDR3 comprising SEQ ID NO:163.
  • the first antigen binding site and the third antigen binding site are specific for different OX40 antigens.
  • the first antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:203, and a VH CDR1 comprising SEQ ID NO:83, a VH CDR2 comprising SEQ ID NO:123, and a VH CDR3 comprising SEQ ID NO:163; and the third antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:190, and a VH CDR1 comprising SEQ ID NO:70, a VH CDR2 comprising SEQ ID NO:110, and a VH CDR3 comprising SEQ ID NO:150.
  • the third antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:203, and a VH CDR1 comprising SEQ ID NO:83, a VH CDR2 comprising SEQ ID NO:123, and a VH CDR3 comprising SEQ ID NO:163; and the first antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:190, and a VH CDR1 comprising SEQ ID NO:70, a VH CDR2 comprising SEQ ID NO:110, and a VH CDR3 comprising SEQ ID NO:150.
  • the first antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:234, and a VH CDR1 comprising SEQ ID NO:83, a VH CDR2 comprising SEQ ID NO:123, and a VH CDR3 comprising SEQ ID NO:163; and the third antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:190, and a VH CDR1 comprising SEQ ID NO:70, a VH CDR2 comprising SEQ ID NO:110, and a VH CDR3 comprising SEQ ID NO:150.
  • the third antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:234, and a VH CDR1 comprising SEQ ID NO:83, a VH CDR2 comprising SEQ ID NO:123, and a VH CDR3 comprising SEQ ID NO:163; and the first antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:190, and a VH CDR1 comprising SEQ ID NO:70, a VH CDR2 comprising SEQ ID NO:110, and a VH CDR3 comprising SEQ ID NO:150.
  • the second antigen binding site comprises a VL CDR1 comprising SEQ ID NO:220, a VL CDR2 comprising SEQ ID NO:221, and a VL CDR3 comprising SEQ ID NO:190, and a VH CDR1 comprising SEQ ID NO:70, a VH CDR2 comprising SEQ ID NO:110, and a VH CDR3 comprising SEQ ID NO:150.
  • the first and second antigen binding sites comprise a VH sequence comprising SEQ ID NO: 230. In some embodiments, the first and second antigen binding sites comprise a VL sequence comprising SEQ ID NO: 231. In some embodiments, the first and second antigen binding site comprises a VH sequence comprising SEQ ID NO: 230 and a VL sequence comprising SEQ ID NO: 231. In some embodiments, the third antigen binding site comprises a VH sequence comprising SEQ ID NO: 227. In some embodiments, the third antigen binding site comprises a VL sequence comprising SEQ ID NO: 228. In some embodiments, the third antigen binding site comprises a VL sequence comprising SEQ ID NO: 229.
  • the third antigen binding site comprises a VH sequence comprising SEQ ID NO: 227 and a VL sequence comprising SEQ ID NO: 228. In some embodiments, the third antigen binding site comprises a VH sequence comprising SEQ ID NO: 227 and a VL sequence comprising SEQ ID NO: 229.
  • the OX40 antigen binding molecule comprises a fifth polypeptide chain, wherein (a) the first polypeptide chain further comprises a domain N and a domain O, wherein the domains are arranged, from N-terminus to C-terminus, in a N-O-A-B- D-E orientation, and wherein domain N has a variable region domain amino acid sequence, domain O has a constant region amino acid sequence; (b) the fifth polypeptide chain comprises a domain P and a domain Q, wherein the domains are arranged, from N-terminus to C-terminus, in a P-Q orientation, and wherein domain P has a variable region domain amino acid sequence and domain Q has a constant region amino acid sequence; and (c) the first and the fifth polypeptides are associated through an interaction between the N and the P domains and an interaction between the O and the Q domains to form the OX40 antigen binding molecule.
  • the amino acid sequences of domain N and domain A are identical
  • the amino acid sequences of domain H is different from the sequence of domain N and domain A
  • the amino acid sequences of domain O and domain B are identical
  • the amino acid sequences of domain I is different from the sequence of domain O and domain B
  • the amino acid sequences of domain P and domain F are identical
  • the amino acid sequences of domain L is different from the sequence of domain P and domain F
  • the amino acid sequences of domain Q and domain G are identical
  • the amino acid sequences of domain M is different from the sequence of domain Q and domain G
  • the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen
  • the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigen
  • the interaction between the N domain and the P domain form a third antigen binding site specific for the first antigen.
  • the amino acid sequences of domain A and H are identical
  • the amino acid sequence of domain N is different from the sequence of domain A and domain H
  • the amino acid sequences of domain O and domain B are identical
  • the amino acid sequences of domain I is different from the sequence of domain O and domain B
  • the amino acid sequences of domains L and F are identical
  • the amino acid sequence of domain P is different from the sequence of domain L and domain F
  • the amino acid sequences of domain Q and domain G are identical
  • the amino acid sequences of domain M is different from the sequence of domain Q and domain G
  • the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen
  • the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigen
  • the interaction between the N domain and the P domain form a third antigen binding site specific for the first antigen.
  • the first antigen is a first epitope of the OX40 antigen.
  • the second antigen is a second epitope of the OX40 antigen.
  • the first epitope and the second epitope are non- overlapping epitopes.
  • the first and second antigens are identical, the first antigen is a first epitope of the OX40 antigen, and the third antigen is a second epitope of the OX40 antigen.
  • the amino acid sequences of domain N, domain A, and domain H are different, the amino acid sequences of domain O, domain B, and domain I are different, the amino acid sequences of domain P, domain F, and domain L are different, and the amino acid sequences of domain Q, domain G, and domain M are different; and (b) the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen, the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigen, and the interaction between the N domain and the P domain form a third antigen binding site specific for a third antigen.
  • the OX40 antigen binding molecule comprises a sixth polypeptide chain, wherein: (a) the third polypeptide chain further comprises a domain R and a domain S, wherein the domains are arranged, from N-terminus to C-terminus, in a R-S-H-I- J-K orientation, and wherein domain R has a variable region amino acid sequence and domain S has a constant domain amino acid sequence; (b) the sixth polypeptide chain comprises: a domain T and a domain U, wherein the domains are arranged, from N-terminus to C-terminus, in a T-U orientation, and wherein domain T has a variable region amino acid sequence and domain U has a constant domain amino acid sequence; and (c) the third and the sixth polypeptides are associated through an interaction between the R and the T domains and an interaction between the S and the U domains to form the OX40 antigen binding molecule.
  • the amino acid sequences of domain R and domain A are identical, the amino acid sequences of domain H is different from the sequence of domain R and domain A, the amino acid sequences of domain S and domain B are identical, the amino acid sequences of domain I is different from the sequence of domain S and domain B, the amino acid sequences of domain T and domain F are identical, the amino acid sequences of domain L is different from the sequence of domain T and domain F, the amino acid sequences of domain U and domain G are identical, the amino acid sequences of domain M is different from the sequence of domain U and domain G, and (b) the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen, the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigen, and the interaction between the R domain and the T domain form a third antigen binding site specific for the first antigen.
  • the first antigen is a first epitope of the OX40 antigen.
  • the second antigen is a second epitope of the OX40 antigen.
  • the first epitope and the second epitope are non- overlapping epitopes.
  • the amino acid sequences of domain R and domain H are identical, the amino acid sequences of domain A is different from the sequence of domain R and domain H, the amino acid sequences of domain S and domain I are identical, the amino acid sequences of domain B is different from the sequence of domain S and domain I, the amino acid sequences of domain T and domain L are identical, the amino acid sequences of domain F is different from the sequence of domain T and domain L, the amino acid sequences of domain U and domain M are identical, the amino acid sequences of domain G is different from the sequence of domain U and domain M, and (b) the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen, the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigen, and the interaction between the R domain and the T domain form a third antigen binding site specific for the second antigen.
  • the second antigen is a first epitope of the OX40 antigen.
  • the first antigen is a second epitope of the OX40 antigen.
  • the first epitope and the second epitope are non- overlapping epitopes.
  • the amino acid sequences of domain R, domain A, and domain H are different, the amino acid sequences of domain S, domain B, and domain I are different, the amino acid sequences of domain T, domain F, and domain L are different, and the amino acid sequences of domain U, domain G, and domain M are different; and (b) the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen, the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigen, and the interaction between the R domain and the T domain form a third antigen binding site specific for a third antigen.
  • Also provided herein is a purified binding molecule comprising a multivalent antibody construct described herein or the OX40 antigen binding molecule described herein.
  • Also provided herein is a purified binding molecule comprising a multivalent ABP described herein or the OX40 antigen binding molecule described herein.
  • the purified binding molecule is purified by a purification method comprising a CH1 affinity purification step.
  • the purified binding molecule is purified by a single- step purification method.
  • a multivalent antibody construct described herein, a OX40 antigen binding molecule described herein, or a purified binding molecule described herein wherein the multivalent antibody construct, the OX40 antigen binding molecule, or the purified binding molecule comprises a biophysical property selected from the group consisting of high yield, high purity, homogeneity, stability, long-term stability, acid stability, thermostability, low antibody cross-interaction, low antibody self-interaction, low
  • the target receptor requires clustering for agonist activity
  • the target receptor is not a TNFRSF member.
  • the target receptor is selected from the group consisting of: CD28 family receptors, LFA1, CD2, Siglecs family receptors, NKp46, NKp44, KIR family of receptors, NKG2 family receptors, CD36, CD14, CR3, ITAM/ITIM, Fc alpha, gamma, and epsilon receptor family.
  • the target receptor is CD20, MET, HER2, or CD38.
  • ABSP antigen binding protein
  • ABP that binds to the same epitope or set of epitopes as a multivalent antibody construct, an OX40 antigen binding molecule, or a purified binding molecule disclosed herein.
  • composition comprising a multivalent antibody construct, an OX40 antigen binding molecule, or a purified binding molecule disclosed herein; and a pharmaceutically acceptable diluent.
  • Also provided herein is a method of treating cancer, comprising administering a therapeutically effective amount of a pharmaceutical composition disclosed herein to a patient in need thereof.
  • Also provided herein is an isolated polynucleotide encoding an amino acid sequence comprising a multivalent antibody construct, an OX40 antigen binding molecule, or a purified binding molecule disclosed herein.
  • a vector comprising an isolated polynucleotide disclosed herein.
  • a host cell comprising a vector disclosed herein.
  • a multivalent ABP described herein, a OX40 antigen binding molecule described herein, or a purified binding molecule described herein wherein the multivalent ABP, the OX40 antigen binding molecule, or the purified binding molecule comprises a biophysical property selected from the group consisting of high yield, high purity, homogeneity, stability, long-term stability, acid stability, thermostability, low antibody cross-interaction, low antibody self-interaction, low hydrophobic binding, and cyno crossreactivity.
  • ABP that competes for binding with an ABP described herein.
  • ABP that binds to the same epitope or set of epitopes as an ABP described herein.
  • Also provided herein is a method of treating a proliferative disease in a subject in need thereof, comprising co-administering to the subject (a) a first antigen binding protein (ABP), wherein the first ABP is capable of (i) binding a cell surface receptor target, optionally wherein the cell surface receptor target requires clustering for agonist activity, and (ii) clustering the receptor target on the cell surface in the absence of an independent cross- linking agent or one or more Fc mutations that drive hexamer formation, and (b) a second ABP, wherein the second ABP is capable of binding (i) a cell surface protein present on a tumor cell and (ii) a cell surface antigen present on an immune cell.
  • ABP antigen binding protein
  • a pharmaceutical composition comprising: (a) a first antigen binding protein (ABP), wherein the first ABP is capable of (i) binding a cell surface receptor target, optionally wherein the cell surface receptor target requires clustering for agonist activity, and (ii) clustering the receptor target on the cell surface in the absence of an independent cross-linking agent or one or more Fc mutations that drive hexamer formation; (b) a second ABP, wherein the second ABP is capable of binding (i) a cell surface protein present on a tumor cell and (ii) a cell surface antigen present on an immune cell, and (c) a pharmaceutically acceptable carrier.
  • ABSP antigen binding protein
  • kits comprising: (a) a first antigen binding protein (ABP), wherein the first ABP is capable of (i) binding a cell surface receptor target, optionally wherein the cell surface receptor target requires clustering for agonist activity, and (ii) clustering the receptor target on the cell surface in the absence of an independent cross- linking agent or one or more Fc mutations that drive hexamer formation; (b) a second ABP, wherein the second ABP is capable of binding (i) a cell surface protein present on a tumor cell and (ii) a cell surface antigen present on an immune cell, and (c) instructions for use according to a method described herein.
  • ABSP antigen binding protein
  • the first cell surface receptor target is a TNF Receptor superfamily (TNFRSF) member.
  • the TNFRSF member is selected from the group consisting of TNFR1 (also known as CD120a and TNFRSF1A), TNFR2 (also known as CD120b and TNFRSF1B), TNFRSF3 (also known as LTbR), TNFRSF4 (also known as OX40 and CD134), TNFRSF5 (also known as CD40), TNFRSF6 (also known as FAS and CD95), TNFRSF6B (also known as DCR3), TNFRSF7 (also known as CD27), TNFRSF8 (also known as CD30), TNFRSF9 (also known as 4-1BB), TNFRSF10A (also known as TRAILR1, DR4, and CD26), TNFRSF10B (also known as TRAILR2, DR5, and CD262), TNFRSF10C (also known as TRAFRSF10C (also known as TRAF10
  • the first cell surface receptor target is not a TNFRSF member.
  • the first cell surface receptor target is selected from the group consisting of: CD28 family receptors, LFA1, CD2, Siglecs family receptors, NKp46, NKp44, KIR family of receptors, NKG2 family receptors, CD36, CD14, CR3, ITAM/ITIM, Fc alpha, gamma, and epsilon receptor family.
  • the cell surface protein present on the tumor cell is selected from ADP-Ribosyl Cyclase/Cyclic ADP-Ribose Hydrolase 1 (CD38), B-Cell Receptor CD22, B-Lymphocyte Antigen CD20 (CD20), B7-H3, C-Type Lectin Domain Family 12 member A (CLL-1), C-X-C Chemokine Receptor Type 5 (CXCR5), Cadherin-17 (CDH17), Cadherin-3, Carcinoembryonic Antigen-Related Cell Adhesion Molecule 5, CD123, CD133, CD155, CD19, CD37 Antigen, CD44 Antigen, CD79b, CEA Family members, Chondroitin Sulfate Proteoglycan 4 (NG2), Claudin-6, CLDN18.2, CLEC12A, Cytotoxic T-Lymphocyte Protein 4 (CTLA-4; CD152), Dipeptidyl Peptidase 4 (DPP), DPP
  • GPRC5D GD2, Glutamate Carboxypeptidase 2 (PSMA), Glypican-3, gp120, gpA33, GPC3, GPRC5D, Hepatitis A Virus Cellular Receptor 2 (TIM-3), Hepatocyte Growth Factor Receptor (HGFR; c-MET), HER-2, HIV, HLA-A2, IL-13R, IL-23, IL-23 Receptor (IL23R), IL-23 Receptor Complex, IL-3 Receptor alpha (CD123), IL-5 Receptor alpha, Insulin-Like Growth Factor 1 Receptor (IGF-1R), Integrin alpha4beta1 Receptor (VLA-4), Interleukin-13 Receptor subunit alpha-2 (IL-13RA2), Large Envelope Protein (L-HBsAg) (HBV),
  • Macrophage-Stimulating Protein Receptor (MSPR), Mesothelin, Metalloreductase STEAP1, MHC- Peptide Complexes, MUC-1, MUC16, Mucin-16 (CA125; MUC-16), Myeloid Cell Surface Antigen CD33, p-Cadherin, Programmed Cell Death 1 (PD-1), Prominin-1 (CD133), Prostate Stem Cell Antigen (PSCA), Prostate-Specific Antigen (PSA; Kallikrein-3), PSMA, Receptor Tyrosine-Protein Kinase erbB-2 (HER2), Receptor Tyrosine-Protein Kinase erbB-3 (HER3), Receptor-Type Tyrosine-Protein Kinase FLT3 (FLT-3), Roundabout Homolog 1 (ROBO1), S antigen (HBV), Somatostatin Receptor Type 2 (SS2R), SSTR2, T-Cell-Specific Surface Glycoprotein CD28, Tropho
  • the cell surface protein present on the tumor cell is selected from BCMA, CD123, CD19, CD20, CEA, CLDN18.2, CLEC12A, CLL-1, EGFR, EpCAM, FCRH5, FLT3, gpA33, GPC3, GPRC5D, HER-2, HIV, HLA-A2, MHC- Peptide Complexes, MUC-1, MUC16, PSMA, 5T4, B7-H3, B7-H4, CD123, CD133, CD155, CD79b, DLL3, EGFRviii, EMP2, EphA2, FOLR-alpha, gp120,IL-13R, IL-23, Mesothelin, p-Cadherin, PSMA, and SSTR2.
  • the cell surface protein present on the tumor cell is BCMA.
  • the tumor cell is a lymphoma cell, a leukemia cell, a multiple myeloma cell or a hematological cell.
  • the tumor indication is lymphoma, leukemia, multiple myeloma or a hematological malignancy.
  • the immune cell is either a T cell or NK cell.
  • the T cell is an effector T cell, optionally wherein the effector T cell is a CD8+ killer T cell.
  • the cell surface antigen present on the T cell is CD3.
  • the CD3 is CD3e.
  • the immune cell is an NK cell.
  • the cell surface antigen present on the NK cell is CD16.
  • the first ABP is an antigen clustering ABP described herein.
  • composition comprising the multivalent ABP described herein, the OX40 ABP described herein or a purified ABP described herein, and a pharmaceutically acceptable diluent.
  • Also provided herein is a method of treating cancer, comprising administering a therapeutically effective amount of a pharmaceutical composition described herein to a patient in need thereof.
  • Also provided herein is an isolated polynucleotide encoding any one of the ABPs described herein.
  • Also provided herein is a vector comprising an isolated polynucleotide described herein.
  • Also provided herein is a host cell comprising a vector described herein.
  • FIG.1 presents schematic architectures, with respective naming conventions, for various binding molecules (also called antibody constructs or ABPs) described herein.
  • FIGs.2A-E present higher resolution schematics of polypeptide chains and their domains for the bivalent (1x1) antibody constructs described herein.
  • FIG.2A presents a higher resolution schematic of polypeptide chains and their domains, with respective naming conventions, for the bivalent (1x1) antibody constructs described herein.
  • FIG.2B presents a higher resolution schematic of polypeptide chains and their domains for the“BC1” bivalent (1x1) format.
  • FIG.2C presents a higher resolution schematic of polypeptide chains and their domains for the“BC6” bivalent (1x1) format.
  • FIG.2D presents a higher resolution schematic of polypeptide chains and their domains for the“BC28” bivalent (1x1) format.
  • FIG.2E presents a higher resolution schematic of polypeptide chains and their domains for the “BC44” bivalent (1x1) format.
  • FIGs.3A-C present higher resolution schematics of polypeptide chains and their domains for the trivalent (2x1) antibody constructs described herein.
  • FIG.3A presents a schematic of polypeptide chains and their domains, with respective naming conventions, for the trivalent (2x1) antibody constructs described herein.
  • FIG.3B presents a higher resolution schematic of polypeptide chains and their domains for the“BC1 (2x1)” trivalent (2x1) format.
  • FIG.3C presents a higher resolution schematic of polypeptide chains and their domains for the“TB111” trivalent (1x1) format.
  • FIGs.4A-C present higher resolution schematics of polypeptide chains and their domains for the trivalent (1x2) antibody constructs described herein.
  • FIG.4A presents a schematic of polypeptide chains and their domains, with respective naming conventions, for the trivalent (1x2) antibody constructs described herein.
  • FIG.4B illustrates features of an exemplary trivalent 1x2 construct“CTLA4-4 x Nivo x CTLA4-4.”
  • FIG.4C illustrates features of an exemplary trivalent 1x2 trispecific construct,“BC28-1x1x1a.”
  • FIGs.4D-F present higher resolution schematics of polypeptide chains and their domains for the tetravalent (2x2) antibody constructs described herein.
  • FIG.4E illustrates certain salient features of the exemplary tetravalent 2x2 construct,“BC22-2x2.”
  • FIG.4F illustrates certain salient features of another exemplary tetravalent 2x2 construct.
  • FIGs.5A-B illustrate schematically functional differences between two antibody-mediated strategies for receptor clustering.
  • FIG.5A illustrates clustering by crosslinking antibody agonists that require an independent crosslinking agent (“First
  • FIG.5B illustrates clustering by multispecific/multivalent antibodies capable of driving receptor clustering without the use of independent crosslinking agent, such as those described herein.
  • FIG.6 shows epitope binning data for 17 unique OX40 binders obtained from a single phage display screening campaign.
  • FIG.7 shows the setup, in 96 well format, of 96 bispecific bivalent (1x1) B- Body constructs. Each construct has two anti-OX40 specificities. Numerical numbers represent unique OX40 binders.
  • FIG.8 tabulates concentrations in mg/mL of the respective bivalent 1x1 B- Body constructs after one-step purification. The average concentration was 950 +/- 500 mg/mL.
  • FIG.9 shows NFkB activation by the 96 bispecific bivalent 1x1 B-Body constructs.
  • Black column 6 nM bispecific bivalent (1x1) B-Body.
  • Open column 6 nM of the respective 1x1 B-Body with 20 nM goat-anti-human (GAH) antibody added as an independent crosslinking agent.
  • Data are normalized, with activation by crosslinked 6 nM OX40L-Fc ligand set to 1.
  • FIG.10 shows NFkB activation by 96 trivalent (2x1) anti-OX40 B-Body constructs.
  • Black column 6 nM trivalent (2x1) B-Body.
  • Open column 6 nM of the respective (2x1) B-Body with 20 nM goat-anti-human (GAH) antibody added as an independent crosslinking agent.
  • Data are normalized, with activation by crosslinked 6 nM OX40L-Fc ligand set to 1.
  • FIG.11 compares agonist activity of three clinical OX40 agonists to activation by crosslinked natural ligand (OX40L-FC + GAH), with FIG.11A showing the activity of the mAbs in the absence of the independent crosslinking agent, GAH (goat anti- human Fc antibody), and FIG.11B showing the activity of the mAbs in the presence of the independent crosslinking agent, GAH.
  • OX40L-FC + GAH crosslinked natural ligand
  • FIG.12 compares the three anti-OX40 clinical mAbs in the absence of GAH crosslinking to a bispecific bivalent (1x1) construct from our first campaign,“10x9”, a monospecific trivalent construct from our first campaign,“2x2x2”, and crosslinked antigen. Both constructs are seen to possess activity comparable to the crosslinked natural ligand, OX40L-Fc, in the absence of an independent cross-linking agent, and to be far superior as agonists as compared to the three known clinical anti-OX40 mAbs.
  • FIG.13 shows the results of a high throughput screen for greater than 900 combinations of B-body candidate OX40 agonists tested in the HEK 293-NFkb-GFP/Luc- OX40 covering a wide range of affinity, epitope, and antibody construct geometry combinations.
  • Arrows indicate clinical OX40 candidates used as controls (arrows from left to right: Pogalizumab, Tavolixizumab, and GSK3174998), each demonstrating activity below the 100% agonism by the OX40L-Fc fusion protein.
  • FIG.14 shows agonist activity of three bivalent OX40 agonists in the absence and presence (+GAH) of the goat-anti-human (GAH) antibody crosslinking agent, as well as agonist activity of the control, crosslinked natural ligand-Fc fusion (OX40L-Fc), in the absence and presence of GAH (OX40L-Fc + GAH).
  • FIG.15 shows dose response curves for a subset of bispecific OX40 agonists using both bivalent and trivalent formats identified during the high throughput screen.
  • FIG.16A illustrates OX40 and OX40L bound in trimer from a top view (left panel) and side view (right panel).
  • the extracellular domain of OX40 consists of four cysteine rich domains (CRD) with boundaries for each CRD noted.
  • FIGs.16B-G show binding of the indicated monospecific antibodies to different OX40 fragments having a series of truncations from the N-terminus (AA 2-214, AA 66-214, AA 108-214, and AA 127-214), with the specific epitope region determined listed next to each monospecific antibody or ligand.
  • Fig.16B shows binding of candidate“2x2” for the different OX40 truncations.
  • Fig.16C shows binding of candidate“8x8” for the different OX40 truncations.
  • Fig.16D shows binding of the OX40 ligand“OX40L” for the different OX40 truncations.
  • FIG.16E shows binding of clinical antibody“GSK3174998” for the different OX40 truncations.
  • Fig.16F shows binding of clinical antibody“Pogalizumab” for the different OX40 truncations.
  • FIG.16G shows binding of clinical antibody
  • FIG.17 shows simultaneous binding of OX40 by different combinations of candidate antigen binding sites.
  • FIG.18 shows the result summary from testing non-overlapping epitope binding for all possible combinations of the panel of the 40 antigen binding sites identified in the screen.
  • FIG.19 shows T cell activation by plate bound (“coated”) and soluble OX40 agonists OX402-2x8 and GSK3174998 (“clinical”). Left panel shows T cell proliferation and right panel shows IL-2 secretion.
  • FIG.20 shows a summary of primary CD4+ na ⁇ ve T cell stimulatory activity for different multispecific multivalent candidate OX40 agonists.
  • the X-axis represents the IL2 secretion, while the Y-axis is the CD4+/CD45RA+/CD25- T cell proliferation stimulated by each candidate.
  • the shaded circle provides a cutoff identifying those agonists considered the most potent.
  • FIG.21 shows the kinetics of T cell activation monitored by microscopy and charted using cell size measurement using the IncuCyte system to track the growth and proliferation of T cell clusters.
  • FIG.22 shows a non-reducing SDS-PAGE analysis of two-step purified candidate OX40 agonists and two clinical monoclonal antibodies.
  • FIG.23A-C shows dose response curves for activation, as monitored by cytokine secretion, of T cells using OX40 candidates OX40:24-11x11 and OX40:24-24x11 in a soluble 2x1 format, as well as by soluble GSK3174998“GSK”, plate-bound GSK3174998 “GSK-Coated), and cross-linked GSK3174998 (“GSK+GAH”).
  • FIG.23A shows activation as monitored by TNFa secretion.
  • FIG.23B shows activation as monitored by IL-2 secretion.
  • FIG.23C shows activation as monitored by IFNg secretion.
  • FIGs.24A-C show dose response curves for activation, as monitored by cytokine secretion, of T cells using OX40 candidates OX40:24-11x11, OX40:24-24x11, and OX40:24-24(WEE)x11 in a soluble 2x1 format, and OX40 candidates OX40:24x11 and OX40:11x24 in a soluble 1x1 B-body format, as well as by cross-linked GSK3174998 (“GSK+GAH”).
  • FIG.24A shows activation as monitored by TNFa secretion.
  • FIG.24B shows activation as monitored by IL-2 secretion.
  • FIG.24C shows activation as monitored by IFNg secretion.
  • FIGs.25A-C show dose response curves for activation, as monitored by cytokine secretion, of T cells at Day 3 using OX40 candidates OX40:24-11x11, OX40:24- 24x11, OX40:24-24(WEE)x11, and OX40:24(WEE)-11x11 in a soluble 2x1 format.
  • FIG. 25A shows activation as monitored by TNFa secretion.
  • FIG.25B shows activation as monitored by IL-2 secretion.
  • FIG.25C shows activation as monitored by IFNg secretion.
  • FIGs.26A-C show dose response curves for activation, as monitored by cytokine secretion, of T cells at Day 4 using OX40 candidates OX40:24-11x11, OX40:24- 24x11, OX40:24-24(WEE)x11, and OX40:24(WEE)-11x11 in a soluble 2x1 format.
  • FIG. 26A shows activation as monitored by TNFa secretion.
  • FIG.26B shows activation as monitored by IL-2 secretion.
  • FIG.26C shows activation as monitored by IFNg secretion.
  • FIGS.27A-C show dose response curves for activation of T cells at Day 5 using OX40 candidates OX40:24-11x11, OX40:24-24x11, OX40:24-24(WEE)x11, and OX40:24(WEE)-11x11 in a soluble 2x1 format monitored by cytokine secretion.
  • FIG.27A shows activation as monitored by TNFa secretion.
  • FIG.27B shows activation as monitored by IL-2 secretion.
  • FIG.27C shows activation as monitored by IFNg secretion.
  • FIGS.28A-D show dose response curves for kinetics of the activation of T cells using OX40 candidates OX40:24-11x11, OX40:24-24x11, OX40:24-24(WEE)x11, and OX40:24(WEE)-11x11 in a soluble 2x1 format monitored by proliferation.
  • FIG.28A shows proliferation across Days 1-5 using the candidate OX40:24-24x11.
  • FIG.28B shows proliferation across Days 1-5 using the candidate OX40:24-24(WEE)x11.
  • FIG.28C shows proliferation across Days 1-5 using the candidate OX40:24-11x11.
  • FIG.28B shows proliferation across Days 1-5 using the candidate OX40:24(WEE)-11x11.
  • FIG.29 shows dose response curves for activation of T cells using OX40 candidates OX40:24-24x11 and OX40:24-24(WEE)x11 in a soluble 2x1 format, OX40 candidate OX40:24x11 in a soluble 1x1 B-body format, and soluble and cross-linked (“+GAH”) OX40 candidate OX40:24-11x11 as monitored by TNFa secretion, as well as activation by soluble and cross-linked GSK3174998 (“GSK+GAH”).
  • FIGS.30A-B show activation, as monitored by cytokine secretion, of T cells using OX40 candidates OX40:24-11x11, OX40:24-24x11, OX40:24-24(WEE)x11, and OX40:24-24x38 in a soluble 2x1 format, OX40 candidate OX40:11x24 in a soluble 1x1 B- body format, OX40 candidates OX40:11 and OX40:24 in a native IgG format, a combination of both OX40 candidates OX40:11 and OX40:24 in a native IgG format, by soluble and cross-linked GSK3174998 (“GSK+GAH”), as well as an anti-CD3 antibody or untreated (“no anti-CD3”) conditions.
  • FIG.27A shows activation as monitored by TNFa secretion.
  • FIG. 27B shows activation as monitored by IL-2 secretion.
  • FIG.31 shows dose response curves for activation of T cells as monitored in an NFkB Luc2 OX40 Jurkat T cell stimulation assay of OX40 candidates OX40:24-11x11 and OX40:24-24x11 in a soluble 2x1 format, and OX40 candidates OX40:24x11 and OX40:11x24 in a soluble 1x1 B-body format, as well as by soluble (“GSK”) and cross-linked (“GSK+GAH”) GSK3174998.
  • GSK soluble
  • GSK+GAH cross-linked
  • FIG.32 depicts an exemplary photomicrograph of T cell clusters formed by Day 5 of treatment with either the clinical mAB or bispecific B-Body OX40:2x8.
  • FIG.33 depicts affinity results for OX40-11 and OX40-24 candidates.
  • FIG.34 depicts cell binding data for the OX40-11 and OX40-24 candidates.
  • FIG.35 depicts effects of OX40 candidate OX40:24(WEE)-11X11 on IL-10 secretion in iTregs, cultured alone or co-cultured with na ⁇ ve activated CD4+ T-cells.
  • FIG.36 depicts effects of OX40 candidate OX40:24(WEE)-11X11 on IL-10 secretion in M2A cells.
  • FIG.37 depicts effects of OX40 candidate OX4024(WEE)-11X11 on rhesus T-cell proliferation and TNF-a secretion.
  • FIG.38 depicts exemplary photomicrographs of T-cell clusters for either the CPG alone, INV24111 alone, or CPG+INV241111 treatment conditions.
  • FIG.39 depicts T-cell confluence results for either the CPG alone, INV24111 alone, or CPG+INV241111 treatment conditions.
  • FIG.40 depicts TNF-a secretion results for cells treated with varying doses of CPG, alone or in combination with a fixed dose of INV241111.
  • FIG.41A depicts next generation sequencing results indicating that hundreds of OX-40 specific clones were enriched during panning.
  • FIG.41B depicts isoaffinity plots from 40 selected OX40 clones.
  • FIG.42 shows exemplary simultaneous binding results from three different OX40 candidates.
  • FIG.43A depicts Zenix column affinity results from the antibody discovery campaign.
  • FIG.43B depicts cross-interaction chromatography results from the antibody discovery campaign.
  • FIG.44 depicts simultaneous binding of OX40:11 and OX40:24 binding molecule candidates to OX40.
  • FIG.45 depicts BLI results demonstrating that each of the OX40:11 and OX40:24 antigen binding sites bind to a constrained region of OX40 that interacts with OX40L.
  • FIG.46 depicts a framework for the clustering potential of antibody-based agonists.
  • FIG.47 depicts binding affinity of the top two clones to BCMA as determined by Octet (Pall ForteBio) biolayer interferometry analysis.
  • FIGS.48A, 48B, and 48C depict results of a Jurkat co-culture assay for BCMAxCD3 B-body TM antibodies.
  • FIGS.49A and 49B depict results from an in vitro cytotoxicity assay, testing different BCMAxCD3 bispecific constructs.
  • FIG.50 depicts results from an APRIL Competition assay, testing the 2x1 BCMA-1xCD3 B-BodyTM construct.
  • FIG.51A depicts results from a Jurkat co-culture assay, testing the effect of various OX40 clustering agonists in combination with the BCMAxCD3 redirecting bispecific antibody.
  • FIG.51B depicts Jurkat co-culture assay results from a BCMAxCD3 dose- response study, administered alone or in combination with two doses of the OX40 clustering agonist antibody INV241111.
  • “antigen binding site” is meant a region of a binding molecule, that specifically recognizes or binds to a given antigen or epitope.
  • “B-Body,” as used herein and with reference to FIGs.2A, 3A, 4A and 4D refers to binding molecules comprising the features of a first and a second polypeptide chain, wherein: (a) the first polypeptide chain comprises a domain A, a domain B, a domain D, and a domain E, wherein the domains are arranged, from N-terminus to C-terminus, in a A-B-D-E orientation, and wherein domain A has a VL amino acid sequence, domain B has a CH3 amino acid sequence, domain D has a CH2 amino acid sequence, and domain E has a constant region domain amino acid sequence; (b) the second polypeptide chain comprises a domain F and a domain G, wherein the domains are arranged, from N-terminus to C-terminus, in a F-G orientation
  • the terms “treat” or “treatment” refer to both therapeutic treatment and prophylactic or preventative measures, wherein the object is to prevent or slow down (lessen) an undesired physiological change or disorder, such as the progression of multiple sclerosis, arthritis, or cancer.
  • Beneficial or desired clinical results include, but are not limited to, alleviation of symptoms, diminishment of extent of disease, stabilized (i.e., not worsening) state of disease, delay or slowing of disease progression, amelioration or palliation of the disease state, and remission (whether partial or total), whether detectable or undetectable.
  • Treatment can also mean prolonging survival as compared to expected survival if not receiving treatment.
  • Those in need of treatment include those already with the condition or disorder as well as those prone to have the condition or disorder or those in which the condition or disorder is to be prevented.
  • subject or “individual” or “animal” or “patient” or “mammal,” is meant any subject, particularly a mammalian subject, for whom diagnosis, prognosis, or therapy is desired.
  • Mammalian subjects include humans, domestic animals, farm animals, and zoo, sports, or pet animals such as dogs, cats, guinea pigs, rabbits, rats, mice, horses, cattle, cows, and so on.
  • the term“sufficient amount” means an amount sufficient to produce a desired effect, e.g., an amount sufficient to modulate protein aggregation in a cell.
  • therapeutically effective amount is an amount that is effective to ameliorate a symptom of a disease.
  • a therapeutically effective amount can be a
  • prophylaxis can be considered therapy.
  • Other interpretational conventions [00326] Unless otherwise specified, all references to sequences herein are to amino acid sequences.
  • antibody constant region residue numbering is according to the Eu index as described at
  • endogenous sequence or“native sequence” is meant any sequence, including both nucleic acid and amino acid sequences, which originates from an organism, tissue, or cell and has not been artificially modified or mutated.
  • Polypeptide chain numbers e.g., a“first” polypeptide chains, a“second” polypeptide chain. etc. or polypeptide“chain 1,”“chain 2,” etc. are used herein as a unique identifier for specific polypeptide chains that form a binding molecule and is not intended to connote order or quantity of the different polypeptide chains within the binding molecule.
  • Ranges provided herein are understood to be shorthand for all of the values within the range, inclusive of the recited endpoints.
  • a range of 1 to 50 is understood to include any number, combination of numbers, or sub-range from the group consisting of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, and 50.
  • multivalent antibody constructs are provided.
  • the construct is capable of (i) binding a cell surface receptor target, and (ii) clustering the receptor target on the cell surface in the absence of an independent cross-linking agent.
  • Each of the target receptor-binding antigen binding sites of the construct is contributed by antibody variable region binding domains.
  • ABSPs multivalent antigen binding proteins
  • clustering of the cell surface receptor target enhances receptor-activated pathways (e.g., has agonizing effects on the receptor-mediated pathways).
  • the cell surface receptor target requires clustering for agonist activity.
  • clustering of the cell surface receptor target suppresses or inhibits a disease-related signal or pathway.
  • clustering of the cell surface receptor target may suppress or inhibit an unwanted immunosuppressive signal, e.g., in a tumor microenvironment.
  • clustering of the cell surface receptor target may suppress an oncogenic signal or pathway.
  • a proliferative disease in a subject in need thereof comprising administering to the subject (a) a first antigen binding protein (ABP), wherein the first ABP is capable of (i) binding a cell surface receptor target, which cell surface receptor target requires clustering for agonist activity, and (ii) clustering the receptor target on the cell surface in the absence of an independent cross-linking agent or one or more Fc mutations that drive hexamer formation; and (b) a second ABP, wherein the second ABP is capable of binding (i) a cell surface protein present on a tumor cell and (ii) a cell surface antigen present on an immune cell.
  • ABP antigen binding protein
  • the multivalent antibody or ABP construct is monospecific.
  • the construct is multispecific.
  • the construct comprises a first antigen binding site specific for a first epitope of the target receptor, and a second antigen binding site specific for a second antigenic target.
  • the second antigenic target is a second epitope of the target receptor.
  • the first epitope and the second epitope are non-overlapping epitopes.
  • the second antigenic target is an epitope of a second protein.
  • the second antigenic target is an epitope of a second protein, wherein the second protein is a second cell surface receptor.
  • the target cell surface receptor and the second cell surface receptor are commonly expressed on the surface of at least some mammalian cells. 6.4.1.
  • Target receptors [00346] In some embodiments, the target receptor is expressed by an immune cell. In some embodiments, the target receptor is expressed on a T cell. In some embodiments, the target receptor is expressed by an activated T cell. In some embodiments, the target receptor activates a costimulatory signaling pathway in the T cell.
  • the target receptor is expressed by a regulatory T cell.
  • the regulatory T cell is located in a tumor microenvironment.
  • the target receptor is a TNF Receptor superfamily (TNFRSF) member.
  • TNFRSF TNF Receptor superfamily
  • the TNFRSF member is TNFR1 (also known as CD120a and TNFRSF1A), TNFR2 (also known as CD120b and TNFRSF1B), TNFRSF3 (also known as LTbR), TNFRSF4 (also known as OX40 and CD134), TNFRSF5 (also known as CD40), TNFRSF6 (also known as FAS and CD95), TNFRSF6B (also known as DCR3), TNFRSF7 (also known as CD27), TNFRSF8 (also known as CD30), TNFRSF9 (also known as 4-1BB), TNFRSF10A (also known as TRAILR1, DR4, and CD26), TNFRSF10B (also known as TRAILR2, DR5, and CD262), TNFRSF10C (also known as TRAILR3, DCR1, CD263), TNFRSF10D (also known as TRAILR4, DCR2, and CD264), TNFR1 (also known as CD120a and
  • the target receptor is OX40 (TNFRSF4), CD40 (TNFRSF5), or 4-1BB (TNFRSF9).
  • the target receptor is OX40.
  • the target receptor is CD40.
  • the target receptor is 4-1BB.
  • the target receptor is a human TNFRSF.
  • the target receptor is human OX40, human CD40, or human 4-1BB.
  • the target receptor is human OX40.
  • the target receptor is human CD40.
  • the target receptor is human 4- 1BB.
  • the target receptor is not a TNFRSF member.
  • the target receptor is selected from the group consisting of: CD28 family receptors, LFA1, CD2, Siglecs family receptors, NKp46, NKp44, KIR family of receptors, NKG2 family receptors, CD36, CD14, CR3, ITAM/ITIM, Fc alpha, gamma, and epsilon receptor family.
  • the target receptor is a human CD28 family receptor, LFA1, CD2, Siglecs family receptor, NKp46, NKp44, KIR family receptor, NKG2 family receptor, CD36, CD14, CR3, ITAM/ITIM, Fc alpha, gamma, or epsilon receptor family member.
  • the target receptor is CD20.
  • the target receptor is human CD20.
  • the target receptor is expressed by a cell other than an immune cell. In some embodiments, the target receptor is expressed by a tumor cell.
  • the target receptor expressed by the tumor cell is CD20. See, e.g., Chu, Te-Wei et al.“A Two-Step Pretargeted Nanotherapy for CD20 Crosslinking May Achieve Superior Anti-Lymphoma Efficacy to Rituximab” Theranostics vol.5,8834-46.26 Apr.2015, doi:10.7150/thno.12040, which is hereby incorporated by reference in its entirety. See, e.g., Rossi EA et al., Cancer Res.2008 Oct 15;68(20):8384-92, which is hereby incorporated by reference in its entirety.
  • the target receptor expressed by the tumor cell is the proto-oncoprotein MET. See, e.g., Li, Wenjing et al.“Induction of MET Receptor Tyrosine Kinase Down-regulation through Antibody- mediated Receptor Clustering” Scientific reports vol.9,11988.13 Feb.2019,
  • the target receptor expressed by the tumor cell is Her2. See, e.g., Brack S. et al., Mol Cancer Ther.2014 Aug;13(8):2030-9. doi: 10.1158/1535-7163.MCT-14- 0046-T, which is hereby incorporated by reference in its entirety.
  • the target receptor expressed by the tumor cell is CD38. Studies show that a clinical phase anti- CD38 antibody daratumumab, known to induce apoptosis of tumor cells, clusters CD38 on the plasma membrane of multiple myeloma cells. See, e.g., Jansen JHM et al., Blood 2012 120:2974; and Horenstein, Alberto L et al.“NAD+-Metabolizing Ectoenzymes in
  • the target receptor is CD20. In a particular embodiment, the target receptor is human CD20. 6.4.2. Receptor clustering activity [00357] The capability of the ABP to cluster the receptor target in the absence of an independent cross-linking reagent may be assessed by any suitable means.
  • the capability of the ABP to cluster the receptor target in the absence of an independent cross-linking reagent is visualized by any suitable means.
  • the clustering activity of the ABP in the absence of an independent cross-linking reagent may be assessed by visualizing receptor clusters or cluster size by, e.g., microscopy.
  • Suitable microscopy techniques for visualizing receptor clustering include, but are not limited to, confocal microscopy, atomic force microscopy (described in Li M. et al., J Immunol Methods.2016 Sep;436:41-9.
  • presence of receptor clusters are assayed by determining size or molecular weight of receptor monomers and oligomers.
  • receptor clusters are assayed by non-reducing SDS- PAGE and western blotting.
  • non-reducing SDS-PAGE and western blotting may reveal the presence of higher-order molecular weight bands
  • receptor clusters are assayed by size exclusion chromatography.
  • An exemplary method for determining receptor clustering by size exclusion chromatography is described in US20180057598, which is hereby incorporated by reference in its entirety.
  • the capability of the ABP to cluster the receptor target is indicated by the ability of the ABP to activate or enhance receptor-mediated pathway activity.
  • the ABP may induce or increase agonist activity of the receptor, in the absence of a cross-linking reagent.
  • the presence of an independent cross-linking agent does not increase agonist activity above that achieved in the absence of the independent cross-linking agent. In certain embodiments, the presence of an independent cross-linking agent does not increase agonist activity above that achieved in the absence of the independent cross-linking agent when tested in vitro. In particular embodiments in which the independent cross-linking agent is cross-linked natural ligand of the target receptor, the presence of cross- linked ligand for the target receptor does not increase in vitro agonist activity above that achieved in the absence of the cross-linked ligand. In certain embodiments, the presence of an independent cross-linking agent does not increase agonist activity above that achieved in the absence of the independent cross-linking agent when the multivalent antibody construct is administered in vivo.
  • the presence of an independent cross-linking agent increases agonist activity above that achieved in the absence of the independent cross-linking agent.
  • the presence of an independent cross-linking agent increases agonist activity above that achieved in the absence of the independent cross-linking agent when tested in vitro.
  • the independent cross-linking agent is cross-linked natural ligand of the target receptor
  • the presence of cross-linked ligand for the target receptor increases in vitro agonist activity above that achieved in the absence of the cross-linked ligand.
  • cross-linked target receptor ligand increases in vitro agonist activity 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, 200%, 300% or even 400% above that achieved in the absence of the cross-linked ligand.
  • cross-linked target receptor ligand increases in vitro agonist activity more than 2-fold above that achieved in the absence of the cross-linked ligand.
  • cross-linked target receptor ligand increases in vitro agonist activity more than 3-fold above that achieved in the absence of the cross-linked ligand [00365]
  • the presence of an independent cross-linking agent increases agonist activity above that achieved in the absence of the independent cross-linking agent when the multivalent construct is administered in vivo.
  • the agonist activity is indicated by increased T cell effector activity.
  • T cell effector activity include, but are not limited to cytokine secretion, e.g., TNFa secretion, IL-2 secretion, and/or IFNg secretion.
  • the agonist activity is indicated by increased T cell proliferation.
  • the agonist activity is indicated by proliferating T-cell cluster size.
  • the agonist activity is indicated by a reporter assay, e.g., a Jurkat T cell stimulation reporter assay. 6.4.3.
  • Bivalent constructs [00367] In some embodiments, the multivalent construct is bivalent.
  • the bivalent construct is a bivalent (1x1) construct.
  • the basic architecture of bivalent (1x1) constructs is included among the architectures schematized in FIG.1, and is specifically shown in greater detail in FIG.2A.
  • the binding molecules comprise a first, a second, a third, and a fourth polypeptide chain, wherein: a) the first polypeptide chain comprises a domain A, a domain B, a domain D, and a domain E, wherein the domains are arranged, from N-terminus to C-terminus, in a A-B-D-E orientation, wherein domain A has a variable region domain amino acid sequence, and wherein domain B, domain D, and domain E have a constant region domain amino acid sequence; (b) the second polypeptide chain comprises a domain F and a domain G, wherein the domains are arranged, from N-terminus to C-terminus, in a F-G orientation, and wherein domain F has a variable region domain amino acid sequence and domain G has a constant region domain amino acid sequence; (c) the third polypeptide chain comprises a domain H, a domain I, a domain J, and a domain K,
  • the bivalent binding molecules comprise a native antibody architecture, wherein the binding molecule is structured as described in Section 6.4.3 (see“Bivalent constructs”) wherein domains A and H comprise VH amino acid sequences, domains F and L comprise VL amino acid sequences, domains B and I comprise CH1, domains G and M comprise CL, domains D and J comprise CH2, and domains E and K comprise CH3.
  • the binding molecule is a B-Body TM .
  • B-Body TM binding molecules are described in International Patent Application No.
  • the binding molecule is structured as described in Section 6.4.3 (see“Bivalent constructs”) wherein domains A and H comprise VL, domains B and G comprise CH3, domain I comprises CL or CH1, domain M comprises CH1 or CL, domains D and J comprise CH2, and domains E and K comprise CH3.
  • domains A and H comprise VL
  • domains B and G comprise CH3
  • domain I comprises CL or CH1
  • domain M comprises CH1 or CL
  • domains D and J comprise CH2
  • domains E and K comprise CH3.
  • domain I comprises CL and domain M comprises CH1.
  • domain M comprises CH1.
  • domain I comprises CH1 and domain M comprises CL.
  • the binding molecule is a CrossMab TM .
  • CrossMab TM antibodies are described in U.S. Patent Nos.8,242,247; 9,266,967; and 8,227,577, U.S. Patent Application Pub. No.20120237506, U.S. Patent Application Pub. No. US20090162359, WO2016016299, WO2015052230.
  • the binding molecule is a bivalent, bispecific antibody, comprising: a) the light chain and heavy chain of an antibody specifically binding to a first antigen; and b) the light chain and heavy chain of an antibody specifically binding to a second antigen, wherein constant domains CL and CH1 from the antibody specifically binding to a second antigen are replaced by each other.
  • the binding molecule is structured as described in Section 6.4.3 wherein A is VH, B is CH1, D is CH2, E is CH3, F is VL, G is CL, H is VL or VH, I is CL, J is CH2, K is CH3, L is VH or VL, and M is CH1.
  • the binding molecule is an antibody having a general architecture described in U.S. Patent No.8,871,912 and WO2016087650.
  • the binding molecule is a domain-exchanged antibody comprising a light chain (LC) composed of VL- CH3, and a heavy chain (HC) comprising VH-CH3-CH2-CH3, wherein the VL-CH3 of the LC dimerizes with the VH-CH3 of the HC thereby forming a domain-exchanged LC/HC dimer comprising a CH3LC/CH3HC domain pair.
  • LC light chain
  • HC heavy chain
  • the binding molecule is structured as described in Section 6.4.3 (see“Bivalent constructs”) wherein A is VH, B is CH3, D is CH2, E is CH3, F is VL, G is CH3, H is VH, I is CH1, J is CH2, K is CH3, L is VL, and M is CL.
  • the binding molecule is as described in
  • the binding molecule is structured as described in herein (see“Bivalent constructs”) and Section 6.4.3 wherein A is VH or VL, B is CH2 from IgM or IgE, D is CH2, E is CH3, F is VL or VH, G is CH2 from IgM or IgE, H is VH, I is CH1, J is CH2, K is CH3, L is VL, and M is CL.
  • the binding molecule is as described in
  • the binding molecule is structured as described in herein (see“Bivalent constructs”) and Section 6.4.3 wherein A is VH, B is CH1, D is CH2, E is CH3, F is VL, G is CL, H is VL, I is CL or CH1, J is CH2, K is CH3, L is VH, and M is CH1 or CL.
  • the first and third polypeptide chains are identical in sequence to one another, and the second and fourth polypeptide are identical in sequence to one another.
  • association of the first and third polypeptide chains through interactions between domains E & K form a bivalent monospecific antibody construct.
  • first and third polypeptide chains are non-identical in sequence to one another, and the second and fourth polypeptide are non-identical in sequence to one another.
  • association of the first and third polypeptide chains through interactions between domains E & K is capable of forming a bivalent bispecific antibody construct.
  • domain A has a variable region domain amino acid sequence.
  • Variable region domain amino acid sequences as described herein, are variable region domain amino acid sequences of an antibody including VL and VH antibody domain sequences. VL and VH sequences are described in greater detail in Sections 6.4.3.2.1 and 6.4.3.2.2, respectively.
  • domain A has a VL antibody domain sequence and domain F has a VH antibody domain sequence.
  • domain A has a VH antibody domain sequence and domain F has a VL antibody domain sequence. 6.4.3.2.1.
  • VL amino acid sequences in the binding molecules described herein are typically sequences of a native antibody light chain variable domain.
  • a specific VL amino acid sequence associates with a specific VH amino acid sequence to form an antigen-binding site.
  • the VL amino acid sequences are mammalian sequences, including human sequences, synthesized sequences, or combinations of human, non-human mammalian, mammalian, and/or synthesized sequences, as described in further detail herein and in Sections 6.4.3.2.3 and 6.4.3.2.4.
  • the VL amino acid sequences are human antibody light chain sequences.
  • the VL amino acid sequences are lambda (l) light chain variable domain sequences.
  • the VL amino acid sequences are kappa (k) light chain variable domain sequences.
  • VL amino acid sequences are mutated sequences of naturally occurring (e.g.,“native”) sequences.
  • the C-terminus of domain A is connected to the N-terminus of domain B.
  • domain A has a VL amino acid sequence that is mutated at its C-terminus at the junction between domain A and domain B, as described in greater detail herein and in Section 6.4.6. 6.4.3.2.2.
  • VH Regions [00383]
  • the VH amino acid sequences in the binding molecules described herein are typically sequences of a native antibody heavy chain variable domain. In a typical antibody arrangement in both nature and in the binding molecules described herein, a specific VH amino acid sequence associates with a specific VL amino acid sequence to form an antigen- binding site.
  • VH amino acid sequences are mammalian sequences, including human sequences, synthesized sequences, or combinations of non-human mammalian, mammalian, and/or synthesized sequences, as described in further detail in Sections 6.4.3.2.3 and 6.4.3.2.4.
  • VH amino acid sequences are mutated sequences of naturally occurring (e.g.,“native”) sequences. 6.4.3.2.3.
  • Complementarity Determining Regions may comprise highly variable sequences termed“complementarity determining regions” (CDRs), typically three CDRs (CDR1, CD2, and CDR3).
  • the CDRs are mammalian sequences, including, but not limited to, mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences. In a preferred embodiment, the CDRs are human sequences. In various embodiments, the CDRs are naturally occurring sequences. In various embodiments, the CDRs are naturally occurring sequences that have been mutated to alter the binding affinity of the antigen-binding site for a particular antigen or epitope. In certain embodiments, the naturally occurring CDRs have been mutated in an in vivo host through affinity maturation and somatic hypermutation.
  • the CDRs have been mutated in vitro through methods including, but not limited to, PCR-mutagenesis and chemical mutagenesis.
  • the CDRs are synthesized sequences including, but not limited to, CDRs obtained from random sequence CDR libraries and rationally designed CDR libraries. 6.4.3.2.4. Framework Regions and CDR Grafting [00385]
  • VH and VL amino acid sequences may comprise“framework region” (FR) sequences.
  • FRs are generally conserved sequence regions that act as a scaffold for interspersed CDRs (see Section 6.4.3.2.3), typically in a FR1-CDR1-FR2-CDR2-FR3-CDR3- FR4 arrangement (from N-terminus to C-terminus).
  • the FRs are mammalian sequences, including, but not limited to mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences.
  • the FRs are human sequences.
  • the FRs are naturally occurring sequences.
  • the FRs are human FR sequences.
  • the FRs are synthesized sequences including, but not limited, rationally designed sequences.
  • the FRs and the CDRs are both from the same naturally occurring variable domain sequence.
  • the FRs and the CDRs are from different variable domain sequences, wherein the CDRs are grafted onto the FR scaffold with the CDRs providing specificity for a particular antigen.
  • the grafted CDRs are all derived from the same naturally occurring variable domain sequence.
  • the grafted CDRs are derived from different variable domain sequences.
  • the grafted CDRs are synthesized sequences including, but not limited to, CDRs obtained from random sequence CDR libraries and rationally designed CDR libraries.
  • the grafted CDRs and the FRs are from the same species. In certain embodiments, the grafted CDRs and the FRs are from different species.
  • an antibody is“humanized”, wherein the grafted CDRs are non-human mammalian sequences including, but not limited to, mouse, rat, hamster, rabbit, camel, donkey, and goat sequences, and the FRs are human sequences. Humanized antibodies are discussed in more detail in U.S. Pat. No.6,407,213, the entirety of which is hereby incorporated by reference for all it teaches. In various
  • portions or specific sequences of FRs from one species are used to replace portions or specific sequences of another species’ FRs.
  • domain B has a constant region domain sequence. Constant region domain amino acid sequences, as described herein, are typically sequences of a constant region domain of a native antibody.
  • the constant region sequences are mammalian sequences, including, but not limited to, mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences. In a preferred embodiment, the constant region sequences are human sequences. In certain embodiments, the constant region sequences are from an antibody light chain. In particular embodiments, the constant region sequences are from a lambda or kappa light chain. In certain embodiments, the constant region sequences are from an antibody heavy chain. In particular embodiments, the constant region sequences are an antibody heavy chain sequence that is an IgA1, IgA2, IgD, IgE, IgG1, IgG2, IgG3, IgG4, or IgM isotype.
  • the constant region sequences are from an IgG isotype. In a preferred embodiment, the constant region sequences are from an IgG1 isotype. In preferred specific embodiments, the constant region sequence is a CH3 sequence. CH3 sequences are described in greater detail in Section 6.4.3.3.1. In other preferred embodiments, the constant region sequence is an orthologous CH2 sequence. Orthologous CH2 sequences are described in greater detail in Section 6.4.3.3.2.
  • domain B has a CH1 sequence. In some embodiments, domain B has a CH2 sequence from IgE. In some embodiments, domain B has a CH2 sequence from IgM.
  • the constant region sequence has been mutated to include one or more orthogonal mutations.
  • domain B has a constant region sequence that is a CH3 sequence comprising knob-hole (synonymously, “knob-in-hole,”“KIH”) orthogonal mutations, as described in greater detail in Section 6.4.3.15.2, and either a S354C or a Y349C mutation that forms an engineered disulfide bridge with a CH3 domain containing an orthogonal mutation, as described in in greater detail in Section 6.4.3.15.1.
  • the knob-hole orthogonal mutation is a T366W mutation. 6.4.3.3.1.
  • CH3 Regions [00391] CH3 amino acid sequences, as described herein, are typically sequences of the C-terminal domain of a native antibody heavy chain.
  • the CH3 sequences are mammalian sequences, including, but not limited to, mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences. In a preferred embodiment, the CH3 sequences are human sequences. In certain embodiments, the CH3 sequences are from an IgA1, IgA 2 , IgD, IgE, IgM, IgG 1 , IgG 2 , IgG 3 , IgG4 isotype or CH4 sequences from an IgE or IgM isotype. In a specific embodiment, the CH3 sequences are from an IgG isotype. In a preferred embodiment, the CH3 sequences are from an IgG 1 isotype.
  • the CH3 sequences are endogenous sequences.
  • the CH3 sequence is UniProt accession number P01857 amino acids 224-330.
  • a CH3 sequence is a segment of an endogenous CH3 sequence.
  • a CH3 sequence has an endogenous CH3 sequence that lacks the N- terminal amino acids G224 and Q225.
  • a CH3 sequence has an endogenous CH3 sequence that lacks the C-terminal amino acids P328, G329, and K330.
  • a CH3 sequence has an endogenous CH3 sequence that lacks both the N-terminal amino acids G224 and Q225 and the C-terminal amino acids P328, G329, and K330.
  • a binding molecule has multiple domains that have CH3 sequences, wherein a CH3 sequence can refer to both a full endogenous CH3 sequence as well as a CH3 sequence that lacks N-terminal amino acids, C-terminal amino acids, or both.
  • the CH3 sequences are endogenous sequences that have one or more mutations.
  • the mutations are one or more orthogonal mutations that are introduced into an endogenous CH3 sequence to guide specific pairing of specific CH3 sequences, as described in more detail in Sections 6.4.3.15.1-6.4.3.15.3.
  • the CH3 sequences are engineered to reduce
  • isoallotype mutations are replaced.
  • isoallotype mutations D356E and L358M are made in the CH3 sequence.
  • domain B has a human IgG1 CH3 amino acid sequence with the following mutational changes: P343V; Y349C; and a tripeptide insertion, 445P, 446G, 447K.
  • domain B has a human IgG1 CH3 sequence with the following mutational changes: T366K; and a tripeptide insertion, 445K, 446S, 447C.
  • domain B has a human IgG1 CH3 sequence with the following mutational changes: Y349C and a tripeptide insertion, 445P, 446G, 447K.
  • domain B has a human IgG1 CH3 sequence with a 447C mutation incorporated into an otherwise endogenous CH3 sequence.
  • domain B In the bivalent (1x1) binding molecules described herein, the N-terminus of domain B is connected to the C-terminus of domain A. In certain embodiments, domain B has a CH3 amino acid sequence that is mutated at its N-terminus at the junction between domain A and domain B, as described in greater detail in herein (see“Bivalent constructs”) and in Section 6.4.6.1.
  • domain B has a CH3 amino acid sequence that is extended at the C-terminus at the junction between domain B and domain D, as described in greater detail herein (see“Bivalent constructs”) and in Section 6.4.6.3. 6.4.3.3.2. Orthologous CH2 Regions [00400]
  • CH2 amino acid sequences, as described herein, are typically sequences of the third domain of a native antibody heavy chain, with reference from the N-terminus to C- terminus. CH2 amino acid sequences, in general, are discussed in more detail in Section 6.4.3.4.
  • a binding molecule has more than one paired set of CH2 domains that have CH2 sequences, wherein a first set has CH2 amino acid sequences from a first isotype and one or more orthologous sets of CH2 amino acid sequences from another isotype.
  • the orthologous CH2 amino acid sequences as described herein, are able to interact with CH2 amino acid sequences from a shared isotype, but not significantly interact with the CH2 amino acid sequences from another isotype present in the binding molecule.
  • all sets of CH2 amino acid sequences are from the same species.
  • all sets of CH2 amino acid sequences are human CH2 amino acid sequences.
  • the sets of CH2 amino acid sequences are from different species.
  • the first set of CH2 amino acid sequences is from the same isotype as the other non-CH2 domains in the binding molecule.
  • the first set has CH2 amino acid sequences from an IgG isotype and the one or more orthologous sets have CH2 amino acid sequences from an IgM or IgE isotype.
  • one or more of the sets of CH2 amino acid sequences are endogenous CH2 sequences.
  • one or more of the sets of CH2 amino acid sequences are endogenous CH2 sequences that have one or more mutations.
  • the one or more mutations are orthogonal knob-hole mutations, orthogonal charge-pair mutations, or orthogonal hydrophobic mutations. Orthologous CH2 amino acid sequences useful for the binding molecules are described in more detail in international PCT
  • domain D (Constant Region)
  • domain D has a constant region amino acid sequence. Constant region amino acid sequences are described in more detail herein. For example, also see Section 6.4.3.3.
  • domain D has a CH2 amino acid sequence.
  • CH2 amino acid sequences as described herein, are typically sequences of the third domain of a native antibody heavy chain, with reference from the N-terminus to C-terminus.
  • the CH2 sequences are mammalian sequences, including but not limited to mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences.
  • the CH2 sequences are human sequences.
  • the CH2 sequences are from a IgA1, IgA2, IgD, IgE, IgG1, IgG2, IgG3, IgG4, or IgM isotype.
  • the CH2 sequences are from an IgG1 isotype.
  • the CH2 sequences are endogenous sequences.
  • the sequence is Uniprot accession number P01857 amino acids 111-223.
  • the CH2 sequences have an N-terminal hinge region peptide that connects the N-terminal variable domain-constant domain segment to the CH2 domain, as discussed in more detail in herein and in Sections 6.4.6.3 and 6.4.6.4.
  • the CH2 sequence comprises one or more mutations that reduce effector function, as discussed in more detail herein (see“Bivalent constructs”) and in Section 6.6.4.
  • domain D is connected to the C-terminus of domain B.
  • domain B has a CH3 amino acid sequence that is extended at the C-terminus at the junction between domain D and domain B, as described in greater detail in Section 6.4.6.3.
  • domain D is connected to the N-terminus of domain E.
  • domain D is connected to the N-terminus of domain E that has a CH1 amino acid sequence or CL amino acid sequence, as described in greater detail herein (see“Bivalent constructs”) and in Section 6.4.6.5.
  • domain E (Constant Region)
  • domain E has a constant region domain amino acid sequence. Constant region amino acid sequences are described in greater detail in herein and in Section 6.4.3.3.
  • the constant region sequence is a CH3 sequence. CH3 sequences are described in greater detail in Section 6.4.3.3.1.
  • the constant region sequence has been mutated to include one or more orthogonal mutations.
  • domain E has a constant region sequence that is a CH3 sequence comprising knob-hole (synonymously,“knob-in- hole,”“KIH”) orthogonal mutations, as described in greater detail in Section 6.4.3.15.2, and either a S354C or a Y349C mutation that forms an engineered disulfide bridge with a CH3 domain containing an orthogonal mutation, as described in in greater detail in Section 6.4.3.15.1.
  • the knob-hole orthogonal mutation is a T366W mutation.
  • the constant region domain sequence is a CH1 sequence.
  • CH1 sequences are described in greater detail in Section 6.4.3.9.1.
  • the N-terminus of the CH1 domain is connected to the C-terminus of a CH2 domain, as described in greater detail herein (see“Bivalent constructs”) and in Section 6.4.6.5.
  • the constant region sequence is a CL sequence.
  • CL sequences are described in greater detail in Section 6.4.3.9.2.
  • the N- terminus of the CL domain is connected to the C-terminus of a CH2 domain, as described in greater detail herein (see“Bivalent constructs”) and in Section 6.4.6.5.
  • domain F has a variable region domain amino acid sequence.
  • Variable region domain amino acid sequences as discussed in greater detail in Section 6.4.3.2, are variable region domain amino acid sequences of an antibody including VL and VH antibody domain sequences. VL and VH sequences are described in greater detail in Sections 6.4.3.2.1 and 6.4.3.2.2, respectively.
  • domain F has a VH antibody domain sequence.
  • domain F has a VL antibody domain sequence.
  • domain G (Constant Region)
  • Constant region amino acid sequences are described in greater detail in Section 6.4.3.3.
  • domain G has a CH3 amino acid sequence.
  • CH3 sequences are described in greater detail in Section 6.4.3.3.1.
  • the constant region sequence is an orthologous CH2 sequence. Orthologous CH2 sequences are described in greater detail in Section 6.4.3.3.2.
  • domain G has a human IgG1 CH3 sequence with the following mutational changes: S354C; and a tripeptide insertion, 445P, 446G, 447K.
  • domain G has a human IgG1 CH3 sequence with the following mutational changes: S354C; and 445P, 446G, 447K tripeptide insertion.
  • domain G has a human IgG1 CH3 sequence with the following changes:
  • domain H has a variable region domain amino acid sequence.
  • Variable region domain amino acid sequences discussed in greater detail in Section 6.4.3.2, are variable region domain amino acid sequences of an antibody including VL and VH antibody domain sequences. VL and VH sequences are described in greater detail in Sections 6.4.3.2.1 and 6.4.3.2.2, respectively.
  • domain H has a VL antibody domain sequence.
  • domain H has a VH antibody domain sequence.
  • domain I (Constant Region)
  • domain I has a constant region domain amino acid sequence. Constant region amino acid sequences are described in greater detail in Section 6.4.3.3.
  • domain I has a CL amino acid sequence.
  • domain I has a CH1 amino acid sequence. CH1 and CL amino acid sequences are described in further detail in Sections 6.4.3.9.1 and 6.4.3.9.2, respectively. 6.4.3.9.1.
  • CH1 Domains [00417] CH1 amino acid sequences, as described herein, are typically sequences of the second domain of a native antibody heavy chain, with reference from the N-terminus to C- terminus.
  • the CH1 sequences are endogenous sequences.
  • the CH1 sequences are mammalian sequences, including, but not limited to mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences.
  • the CH1 sequences are human sequences.
  • the CH1 sequences are from an IgA1, IgA2, IgD, IgE, IgG1, IgG2, IgG3, IgG4, or IgM isotype.
  • the CH1 sequences are from an IgG1 isotype.
  • the CH1 sequence is Uniprot accession number P01857 amino acids 1-98. 6.4.3.9.2.
  • CL amino acid sequences are typically sequences of the second domain of a native antibody light chain, with reference from the N-terminus to C-terminus.
  • the CL sequences are endogenous sequences.
  • the CL sequences are mammalian sequences, including, but not limited to mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences.
  • CL sequences are human sequences.
  • the CL amino acid sequences are lambda (l) light chain constant domain sequences.
  • the CL amino acid sequences are human lambda light chain constant domain sequences.
  • the lambda (l) light chain sequence is UniProt accession number P0CG04.
  • the CL amino acid sequences are kappa (k) light chain constant domain sequences.
  • the CL amino acid sequences are human kappa (k) light chain constant domain sequences.
  • the kappa light chain sequence is UniProt accession number P01834.
  • domain J has a constant region amino acid sequence. Constant region amino acid sequences are described in more detail herein, for example in Section 6.4.3.3. In a preferred series of embodiments, domain J has a CH2 amino acid sequence. CH2 amino acid sequences are described in greater detail in Section 6.4.3.4. In a preferred embodiment, the CH2 amino acid sequence has an N-terminal hinge region that connects domain J to domain I, as described in greater detail herein (see “Bivalent constructs”) and in Section 6.4.6.4.
  • domain J is connected to the N-terminus of domain K.
  • domain J is connected to the N-terminus of domain K that has a CH1 amino acid sequence or CL amino acid sequence, as described in greater detail herein (see“Bivalent constructs”) and in Section 6.4.6.5.
  • domain K has a constant region domain amino acid sequence. Constant region domain amino acid sequences are described in greater detail in Section 6.4.3.3. In certain embodiments, the constant region sequence is a CH3 sequence. CH3 sequences are described in greater detail in Section 6.4.3.3.1. In a preferred
  • domain K has a constant region sequence that is a CH3 sequence comprising knob-hole orthogonal mutations, as described in greater detail herein (see“Bivalent constructs”) and in Section 6.4.3.15.2, isoallotype mutations, as described in more detail in 6.4.3.3.1.,and either a S354C or a Y349C mutation that forms an engineered disulfide bridge with a CH3 domain containing an orthogonal mutation, as described in in greater detail in Section 6.4.3.15.1.
  • the knob-hole orthogonal mutations combined with isoallotype mutations are the following mutational changes: D356E, L358M, T366S, L368A, and Y407V.
  • the constant region domain sequence is a CH1 sequence.
  • the N-terminus of the CH1 domain is connected to the C-terminus of a CH2 domain, as described in greater detail in Section 6.4.6.5.
  • the constant region sequence is a CL sequence.
  • the N-terminus of the CL domain is connected to the C-terminus of a CH2 domain, as described in greater detail in Section 6.4.6.5.
  • CH1 and CL amino acid sequences are described in further detail in Sections 6.4.3.9.1 and 6.4.3.9.2, respectively.
  • domain L has a variable region domain amino acid sequence.
  • Variable region domain amino acid sequences discussed in greater detail in Section 6.4.3.2, are variable region domain amino acid sequences of an antibody including VL and VH antibody domain sequences. VL and VH sequences are described in greater detail in Sections 6.4.3.2.1 and 6.4.3.2.2, respectively.
  • domain L has a VH antibody domain sequence.
  • domain L has a VL antibody domain sequence.
  • domain M has a constant region domain amino acid sequence. Constant region amino acid sequences are described in greater detail in Section 6.4.3.3.
  • domain I has a CH1 amino acid sequence and domain M has a CL amino acid sequence.
  • domain I has a CL amino acid sequence and domain M has a CH1 amino acid sequence. CH1 and CL amino acid sequences are described in further detail in Sections 6.4.3.9.1 and 6.4.3.9.2, respectively. 6.4.3.14.
  • a domain A VL or VH amino acid sequence and a cognate domain F VH or VL amino acid sequence are associated and form an antigen binding site (ABS).
  • the A:F antigen binding site (ABS) is capable of specifically binding an epitope of an antigen. Antigen binding by an ABS is described in greater detail in Section 6.4.3.14.1.
  • the ABS formed by domains A and F is identical in sequence to one or more other ABSs within the binding molecule and therefore has the same recognition specificity as the one or more other sequence-identical ABSs within the binding molecule.
  • the A:F ABS is non-identical in sequence to one or more other ABSs within the binding molecule.
  • the A:F ABS has a recognition specificity different from that of one or more other sequence-non-identical ABSs in the binding molecule.
  • the A:F ABS recognizes a different antigen from that recognized by at least one other sequence-non-identical ABS in the binding molecule.
  • the A:F ABS recognizes a different epitope of an antigen that is also recognized by at least one other sequence-non-identical ABS in the binding molecule.
  • the ABS formed by domains A and F recognizes an epitope of antigen, wherein one or more other ABSs within the binding molecule recognizes the same antigen but not the same epitope.
  • 6.4.3.14.1. Binding of Antigen by ABS [00430] An ABS, and the binding molecule comprising such ABS, is said to“recognize” the epitope (or more generally, the antigen) to which the ABS specifically binds, and the epitope (or more generally, the antigen) is said to be the“recognition specificity” or“binding specificity” of the ABS.
  • ABS is said to bind to its specific antigen or epitope with a particular affinity.
  • affinity refers to the strength of interaction of non-covalent intermolecular forces between one molecule and another.
  • the affinity i.e. the strength of the interaction, can be expressed as a dissociation equilibrium constant (KD), wherein a lower KD value refers to a stronger interaction between molecules.
  • KD values of antibody constructs are measured by methods well known in the art including, but not limited to, bio-layer interferometry (e.g., Octet/FORTEBIO®), surface plasmon resonance (SPR) technology (e.g. Biacore®), and cell binding assays.
  • affinities are dissociation equilibrium constants measured by bio-layer interferometry using Octet/FORTEBIO®.
  • “Specific binding,” as used herein, refers to an affinity between an ABS and its cognate antigen or epitope in which the KD value is below 10-6M, 10-7M, 10-8M, 10-9M, or 10-10M.
  • ABSs in a binding molecule as described herein defines the “valency” of the binding molecule.
  • a binding molecule having a single ABS is“monovalent.”
  • a binding molecule having a plurality of ABSs is said to be “multivalent.”
  • a multivalent binding molecule having two ABSs is“bivalent.”
  • a multivalent binding molecule having three ABSs is“trivalent.”
  • a multivalent binding molecule having four ABSs is“tetravalent.”
  • all of the plurality of ABSs have the same recognition specificity.
  • a binding molecule is a
  • “monospecific”“multivalent” binding construct In other multivalent embodiments, at least two of the plurality of ABSs have different recognition specificities. Such binding molecules are multivalent and“multispecific”. In multivalent embodiments in which the ABSs collectively have two recognition specificities, the binding molecule is“bispecific.” In multivalent embodiments in which the ABSs collectively have three recognition specificities, the binding molecule is“trispecific.”
  • the binding molecule is“multiparatopic.”
  • Multivalent embodiments in which the ABSs collectively recognize two epitopes on the same antigen are“biparatopic.”
  • multivalency of the binding molecule improves the avidity of the binding molecule for a specific target.
  • avidity refers to the overall strength of interaction between two or more molecules, e.g., a multivalent binding molecule for a specific target, wherein the avidity is the cumulative strength of interaction provided by the affinities of multiple ABSs. Avidity can be measured by the same methods as those used to determine affinity, as described above.
  • the avidity of a binding molecule for a specific target is such that the interaction is a specific binding interaction, wherein the avidity between two molecules has a KD value below 10 -6 M, 10 -7 M, 10 -8 M, 10 -9 M, or 10 -10 M.
  • the avidity of a binding molecule for a specific target has a K D value such that the interaction is a specific binding interaction, wherein the one or more affinities of individual ABSs do not have has a KD value that qualifies as specifically binding their respective antigens or epitopes on their own.
  • the avidity is the cumulative strength of interaction provided by the affinities of multiple ABSs for separate antigens on a shared specific target or complex, such as separate antigens found on an individual cell. In certain embodiments, the avidity is the cumulative strength of interaction provided by the affinities of multiple ABSs for separate epitopes on a shared individual antigen.
  • domain B and domain G have CH3 amino acid sequences.
  • CH3 sequences are described in greater detail in Section 6.4.3.3.1.
  • the sequence may be a CH3 sequence from human IgG1.
  • amino acid sequences of the B and the G domains are identical.
  • the sequence is an endogenous CH3 sequence.
  • the amino acid sequences of the B and the G domains are different, and separately comprise respectively orthogonal modifications in an endogenous CH3 sequence, wherein the B domain interacts with the G domain, and wherein neither the B domain nor the G domain significantly interacts with a CH3 domain lacking the orthogonal modification.
  • “Orthogonal modifications” or synonymously“orthogonal mutations” as described herein are one or more engineered mutations in an amino acid sequence of an antibody domain that alter the affinity of binding of a first domain having orthogonal modification for a second domain having a complementary orthogonal modification, as compared to binding of the first and second domains in the absence of the orthogonal modifications.
  • the orthogonal modifications decrease the affinity of binding of the first domain having the orthogonal modification for the second domain having the complementary orthogonal modification, as compared to binding of the first and second domains in the absence of the orthogonal modifications.
  • the orthogonal modifications increase the affinity of binding of the first domain having the orthogonal modification for the second domain having the complementary orthogonal modification, as compared to binding of the first and second domains in the absence of the orthogonal modifications. In certain preferred embodiments, the orthogonal modifications decrease the affinity of a domain having the orthogonal modifications for a domain lacking the complementary orthogonal modifications.
  • orthogonal modifications are mutations in an endogenous antibody domain sequence.
  • orthogonal modifications are modifications of the N-terminus or C-terminus of an endogenous antibody domain sequence including, but not limited to, amino acid additions or deletions.
  • orthogonal modifications include, but are not limited to, engineered disulfide bridges, knob- in-hole mutations, and charge-pair mutations, as described in greater detail in Sections 6.4.3.15.1-6.4.3.15.3.
  • orthogonal modifications include a combination of orthogonal modifications selected from, but not limited to, engineered disulfide bridges, knob-in-hole mutations, and charge-pair mutations.
  • the orthogonal modifications can be combined with amino acid substitutions that reduce immunogenicity, such as isoallotype mutations, as described in greater detail in Section 6.4.3.3.1. 6.4.3.15.1. Orthogonal Engineered Disulfide Bridges in CH3 [00444]
  • the orthogonal modifications comprise mutations that generate engineered disulfide bridges between a first and a second domain.
  • “engineered disulfide bridges” are mutations that provide non-endogenous cysteine amino acids in two or more domains such that a non-native disulfide bond forms when the two or more domains associate.
  • Engineered disulfide bridges are described in greater detail in Merchant et al. (Nature Biotech (1998) 16:677-681), the entirety of which is hereby incorporated by reference for all it teaches.
  • engineered disulfide bridges improve orthogonal association between specific domains.
  • a particular embodiments of the orthogonal modifications improve orthogonal association between specific domains.
  • the mutations that generate engineered disulfide bridges are a K392C mutation in one of a first or second CH3 domains, and a D399C in the other CH3 domain.
  • the mutations that generate engineered disulfide bridges are a S354C mutation in one of a first or second CH3 domains, and a Y349C in the other CH3 domain.
  • the mutations that generate engineered disulfide bridges are a 447C mutation in both the first and second CH3 domains that are provided by extension of the C-terminus of a CH3 domain incorporating a KSC tripeptide sequence.
  • orthogonal modifications comprise knob-hole
  • knob-hole mutations are mutations that change the steric features of a first domain’s surface such that the first domain will preferentially associate with a second domain having complementary steric mutations relative to association with domains without the complementary steric mutations. Knob-hole mutations are described in greater detail in U.S. Pat. No.5,821,333 and U.S. Pat. No.
  • knob-hole mutations are combined with engineered disulfide bridges, as described in greater detail in Merchant et al. (Nature Biotech (1998) 16:677-681)), incorporated herein by reference in its entirety.
  • knob-hole mutations, isoallotype mutations, and engineered disulfide mutations are combined.
  • the knob-in-hole mutations are a T366Y mutation in a first domain, and a Y407T mutation in a second domain.
  • the knob-in-hole mutations are a F405A in a first domain, and a T394W in a second domain.
  • the knob-in-hole mutations are a T366Y mutation and a F405A in a first domain, and a T394W and a Y407T in a second domain.
  • the knob- in-hole mutations are a T366W mutation in a first domain, and a Y407A in a second domain.
  • the combined knob-in-hole mutations and engineered disulfide mutations are a S354C and T366W mutations in a first domain, and a Y349C, T366S, L368A, and aY407V mutation in a second domain.
  • the combined knob-in-hole mutations, isoallotype mutations, and engineered disulfide mutations are a S354C and T366W mutations in a first domain, and a Y349C, D356E, L358M, T366S, L368A, and aY407V mutation in a second domain.
  • orthogonal modifications are charge-pair mutations.
  • “charge-pair mutations” are mutations that affect the charge of an amino acid in a domain’s surface such that the domain will preferentially associate with a second domain having complementary charge-pair mutations relative to association with domains without the complementary charge-pair mutations.
  • charge-pair mutations improve orthogonal association between specific domains. Charge-pair mutations are described in greater detail in U.S. Pat. No.8,592,562, U.S. Pat. No.9,248,182, and U.S. Pat.
  • charge-pair mutations improve stability between specific domains.
  • the charge-pair mutations are a T366K mutation in a first domain, and a L351D mutation in the other domain.
  • the E domain has a CH3 amino acid sequence.
  • the K domain has a CH3 amino acid sequence.
  • the amino acid sequences of the E and K domains are identical, wherein the sequence is an endogenous CH3 sequence.
  • CH3 sequences are described in Section 6.4.3.3.1.
  • the CH3 sequences of domains E and K are IgG-CH3 sequences.
  • the sequences of the E and K domains are different.
  • the different sequences separately comprise respectively orthogonal modifications in an endogenous CH3 sequence, wherein the E domain interacts with the K domain, and wherein neither the E domain nor the K domain significantly interacts with a CH3 domain lacking the orthogonal modification.
  • the orthogonal modifications include, but are not limited to, engineered disulfide bridges, knob-in-hole mutations, and charge-pair mutations, as described in greater detail in sections 6.4.3.15.1- 6.4.3.15.3.
  • orthogonal modifications include a combination of orthogonal modifications selected from, but not limited to, engineered disulfide bridges, knob-in-hole mutations, and charge-pair mutations.
  • the orthogonal modifications can be combined with amino acid substitutions that reduce immunogenicity, such as isoallotype mutations, as described in greater detail in Section 6.4.3.3.1.
  • the amino acid sequences of the E domain and the K domain are endogenous sequences of two different antibody domains, the domains selected to have a specific interaction that promotes the specific association between the first and the third polypeptides.
  • the two different amino acid sequences are a CH1 sequence and a CL sequence.
  • CH1 sequences and CL sequences are described in greater detail in Sections 6.4.3.9.1 and 6.4.3.9.2, respectively.
  • Use of CH1 and CL sequences at the C-terminus of a heavy chain to promote specific heavy chain association is described in U.S. Pat. No.8,242,247, the entirety of which is hereby incorporated by reference for all it teaches.
  • the CH1 sequence and the CL sequences are both endogenous sequences.
  • the CH1 sequence and the CL sequences separately comprise respectively orthogonal modifications in endogenous CH1 and CL sequences.
  • the orthogonal modifications in endogenous CH1 and CL sequences are an engineered disulfide bridge selected from engineered cysteines at position 138 of the CH1 sequence and position 116 of the CL sequence, at position 128 of the CH1 sequence and position 119 of the CL sequence, or at position 129 of the CH1 sequence and position 210 of the CL sequence, as numbered and discussed in more detail in U.S. Pat. No.8,053,562 and U.S. Pat. No.9,527,927, each incorporated herein by reference in its entirety.
  • the engineered cysteines are at position 128 of the CH1 sequence and position 118 of the CL Kappa sequence, as numbered by the Eu index.
  • domain I has a CL sequence and domain M has a CH1 sequence.
  • domain H has a VL sequence and domain L has a VH sequence.
  • domain H has a VL amino acid sequence
  • domain I has a CL amino acid sequence
  • domain L has a VH amino acid sequence
  • domain M has a CH1 amino acid sequence.
  • domain H has a VL amino acid sequence
  • domain I has a CL amino acid sequence
  • domain L has a VH amino acid sequence
  • domain M has a CH1 amino acid sequence.
  • domain H has a VL amino acid sequence
  • domain I has a CL amino acid sequence
  • domain L has a VH amino acid sequence
  • domain M has a CH1 amino acid sequence
  • domain K has a CH3 amino acid sequence.
  • the amino acid sequences of the I domain and the M domain separately comprise respectively orthogonal modifications in an endogenous sequence, wherein the I domain interacts with the M domain, and wherein neither the I domain nor the M domain significantly interacts with a domain lacking the orthogonal modification.
  • the orthogonal mutations in the I domain are in a CL sequence and the orthogonal mutations in the M domain are in CH1 sequence. Orthogonal mutations are described in more detail in Sections 6.4.3.15.1-6.4.3.15.3.
  • the orthogonal mutations in the CL sequence and the CH1 sequence are charge-pair mutations.
  • the charge-pair mutations are a F118S, F118A or F118V mutation in the CL sequence with a corresponding A141L in the CH1 sequence, or a T129R mutation in the CL sequence with a corresponding K147D in the CH1 sequence, as numbered by the Eu index and described in greater detail in Bonisch et al.
  • the charge-pair mutations are a N138K mutation in the CL sequence with a corresponding G166D in the CH1 sequence, or a N138D mutation in the CL sequence with a corresponding G166K in the CH1 sequence, as numbered by the Eu index.
  • the orthogonal mutations in the CL sequence and the CH1 sequence generate an engineered disulfide bridge.
  • the mutations that provide non-endogenous cysteine amino acids are a F118C mutation in the CL sequence with a corresponding A141C in the CH1 sequence, or a F118C mutation in the CL sequence with a corresponding L128C in the CH1 sequence, or a S162C mutations in the CL sequence with a corresponding P171C mutation in the CH1 sequence, as numbered by the Eu index.
  • the amino acid sequences of the H domain and the L domain separately comprise respectively orthogonal modifications in an endogenous sequence, wherein the H domain interacts with the L domain, and wherein neither the H domain nor the L domain significantly interacts with a domain lacking the orthogonal modification.
  • the orthogonal mutations in the H domain are in a VL sequence and the orthogonal mutations in the L domain are in VH sequence.
  • the orthogonal mutations are charge-pair mutations at the VH/VL interface.
  • the charge-pair mutations at the VH/VL interface are a Q39E in VH with a corresponding Q38K in VL, or a Q39K in VH with a corresponding Q38E in VL, as described in greater detail in Igawa et al. (Protein Eng. Des. Sel., 2010, vol.23, 667–677), herein incorporated by reference for all it teaches.
  • the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen
  • the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigen
  • the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen
  • the interaction between the H domain and the L domain form a second antigen binding site specific for the first antigen. 6.4.3.18.
  • Bivalent Specificity [00458] In various embodiments, the bivalent construct is monospecific. In these
  • the bivalent construct comprises two copies of a first antigen binding site specific for a first epitope of the target receptor.
  • the bivalent construct is bispecific.
  • the construct comprises a first antigen binding site specific for a first epitope of the target receptor, and a second antigen binding site specific for a second antigenic target.
  • the second antigenic target is a second epitope of the target receptor.
  • the first epitope and the second epitope are non-overlapping epitopes.
  • the second antigenic target is an epitope of a second protein.
  • the antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site. In some embodiments, the antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site. 6.4.4. Trivalent constructs [00461] In another series of embodiments, the binding molecules have three antigen binding sites and are therefore termed“trivalent.”
  • the binding molecules further comprise a fifth polypeptide chain, wherein (a) the first polypeptide chain further comprises a domain N and a domain O, wherein the domains are arranged, from N-terminus to C-terminus, in a N-O-A-B-D-E orientation, and wherein domain N has a variable region amino acid sequence, domain O has a constant region amino acid sequence; (b) the binding molecule further comprises a fifth polypeptide chain, comprising: a domain P and a domain Q, wherein the domains are arranged, from N-terminus to C-terminus, in a P-Q orientation, and wherein domain P has a variable region amino acid sequence and domain Q has a constant region amino acid sequence; and (c) the first and the fifth polypeptides are associated through an interaction between the N and the P domains and an interaction between the O and the Q domains to form the binding molecule.
  • these trivalent embodiments are associated through an interaction between the N and the P domains and an interaction between the O and the Q domain
  • the binding molecules further comprise a sixth polypeptide chain, wherein (a) the third polypeptide chain further comprises a domain R and a domain S, wherein the domains are arranged, from N- terminus to C-terminus, in a R-S-H-I-J-K orientation, and wherein domain R has a variable region amino acid sequence and domain S has a constant domain amino acid sequence; (b) the binding molecule further comprises a sixth polypeptide chain, comprising: a domain T and a domain U, wherein the domains are arranged, from N-terminus to C-terminus, in a T-U orientation, and wherein domain T has a variable region amino acid sequence and domain U has a constant domain amino acid sequence; and (c) the third and the sixth polypeptides are associated through an interaction between the R and the T domains and an interaction between the S and the U domains to form the binding molecule.
  • the third and the sixth polypeptides are associated through an interaction between the R and the T domains and an interaction between the S
  • the domain O is connected to domain A through a peptide linker.
  • the domain S is connected to domain H through a peptide linker.
  • the peptide linker connecting either domain O to domain A or connecting domain S to domain H is a 6 amino acid GSGSGS peptide sequence (SEQ ID NO: 232), as described in more detail in Section 6.4.6.6.
  • the amino acid sequences of domain N and domain A are identical, the amino acid sequences of domain H is different from domains N and A; the amino acid sequences of domain O and domain B are identical, the amino acid sequences of domain I is different from domains O and B; the amino acid sequences of domain P and domain F are identical, the amino acid sequences of domain L is different from domains P and F; the amino acid sequences of domain Q and domain G are identical, the amino acid sequences of domain M is different from domains Q and G; and the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigenic epitope, the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigenic epitope, and the domain N and domain P form a third antigen binding site specific for the first antigenic epitope
  • the construct contains one copy of the antigen binding site (ABS) specific for a first epitope of the target receptor.
  • the antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the construct contains two copies of the antigen binding site specific for a first epitope of the target receptor.
  • a first antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site. In some embodiments, a first antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site. In some embodiments, a first antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site. In certain embodiments, a first antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site and a second antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site.
  • the second antigenic epitope is a second epitope of the target receptor.
  • the first epitope and the second epitope are non-overlapping epitopes.
  • the second antigenic epitope is an epitope of a second protein.
  • the second protein is a second cell surface receptor.
  • the amino acid sequences of domain N and domain H are identical, the amino acid sequences of domain A is different from domains N and H, the amino acid sequences of domain O and domain I are identical, the amino acid sequences of domain B is different from domains O and I, the amino acid sequences of domain P and domain L are identical, the amino acid sequences of domain F is different from domains P and L, the amino acid sequences of domain Q and domain M are identical, the amino acid sequences of domain G is different from domains Q and M; and the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigenic epitope, the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigenic epitope, and the domain N and domain P form a third antigen binding site specific for the second antigenic epitope
  • the construct contains one copy of the antigen binding site (ABS) specific for a first epitope of the target receptor.
  • the antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the construct contains two copies of the antigen binding site specific for a first epitope of the target receptor.
  • a first antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site and a second antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the second antigenic epitope is a second epitope of the target receptor.
  • the first epitope and the second epitope are non-overlapping epitopes.
  • the second antigenic epitope is an epitope of a second protein.
  • the second protein is a second cell surface receptor.
  • the amino acid sequences of domain A and domain H are identical, the amino acid sequences of domain N is different from domains A and H, the amino acid sequences of domain B and domain I are identical, the amino acid sequences of domain O is different from domains B and I, the amino acid sequences of domain F and domain L are identical, the amino acid sequences of domain P is different from domains F and L, the amino acid sequences of domain G and domain M are identical, the amino acid sequences of domain Q is different from domains G and M; and the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigenic epitope, the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigenic epitope, and the domain N and domain P form a third antigen binding site specific for the second antigenic epitope
  • the construct contains one copy of the antigen binding site (ABS) specific for a first epitope of the target receptor.
  • the antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the construct contains two copies of the antigen binding site specific for a first epitope of the target receptor.
  • a first antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site and a second antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the second antigenic epitope is a second epitope of the target receptor.
  • the first epitope and the second epitope are non-overlapping epitopes.
  • the second antigenic epitope is an epitope of a second protein.
  • the second protein is a second cell surface receptor.
  • Trivalent 2x1 Trispecific Constructs [2(A-B)x1(C)] [00476]
  • the amino acid sequences of domain N, domain A, and domain H are different
  • the amino acid sequences of domain O, domain B, and domain I are different
  • the amino acid sequences of domain P, domain F, and domain L are different
  • the amino acid sequences of domain Q, domain G, and domain M are different
  • the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigenic epitope
  • the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigenic epitope
  • the domain N and domain P form a third antigen binding site specific for a third antigenic epitope.
  • domain O has a constant region sequence that is a CL from a kappa light chain and domain Q has a constant region sequence that is a CH1 from an IgG1 isotype, as discussed in more detail in Sections 6.4.3.9.2 and 6.4.3.9.1, respectively.
  • domain O and domain Q have CH3 sequences such that they specifically associate with each other, as discussed in more detail in Section 6.4.3.15.
  • the antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site. In some embodiments, the antigen binding site specific for a first epitope of the target receptor is an N:P antigen binding site. In some embodiments, the antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the second antigenic target is a second epitope of the target receptor. In some aspects, the first epitope and the second epitope are non-overlapping epitopes. In some embodiments, the second antigenic target is a first epitope of a second protein. In some embodiments, the third antigenic target is a third epitope of the target receptor. In some embodiments, the third antigenic target is a second epitope of a second protein. In some aspects, the first epitope of the second protein and the second epitope of the second protein are non-overlapping epitopes. In some embodiments, the third antigenic target is a first epitope of a third protein. In some embodiments, the second protein or third protein is a second cell surface receptor or third cell surface receptor.
  • the amino acid sequences of domain N, domain A, and domain H are identical, the amino acid sequences of domain O and domain B are identical; the amino acid sequences of domain P, domain F, and domain L are identical; and the amino acid sequences of domain Q and domain G are identical; and the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigenic epitope, the interaction between the H domain and the L domain form a second antigen binding site specific for the first antigenic epitope, and the domain N and domain P form a third antigen binding site specific for the first antigenic epitope.
  • These trivalent constructs are monospecific; all three antigen binding sites are specific for a first epitope of the target receptor.
  • the amino acid sequences of domain N, domain A, and domain H are identical, the amino acid sequences of domain O, domain B, and domain I are identical, the amino acid sequences of domain P, domain F, and domain L are identical, and the amino acid sequences of domain Q, domain G, and domain M are identical; and the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigenic epitope, the interaction between the H domain and the L domain form a second antigen binding site specific for the first antigenic epitope, and the domain N and domain P form a third antigen binding site specific for the first antigenic epitope.
  • These trivalent constructs are monospecific; all three antigen binding sites are specific for a first epitope of the target receptor.
  • the amino acid sequences of domain R and domain A are identical, the amino acid sequences of domain H is different from domain R and A, the amino acid sequences of domain S and domain B are identical, the amino acid sequences of domain I is different from domain S and B, the amino acid sequences of domain T and domain F are identical, the amino acid sequences of domain L is different from domain T and F, the amino acid sequences of domain U and domain G are identical, the amino acid sequences of domain M is different from domain U and G and the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigenic epitope, the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigenic epitope, and the domain R and domain T form a third antigen binding site specific for the first antigenic epitope.
  • the trivalent construct contains one copy of the antigen binding site (ABS) specific for a first epitope of the target receptor.
  • the antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an R:T antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the bispecific trivalent construct contains two copies of the antigen binding site specific for a first epitope of the target receptor.
  • a first antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an R:T antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site and a second antigen binding site specific for a first epitope of the target receptor is an R:T antigen binding site.
  • the second antigenic target is a second epitope of the target receptor.
  • the first epitope and the second epitope are non-overlapping epitopes.
  • the second antigenic target is an epitope of a second protein.
  • the second protein is a second cell surface receptor.
  • the amino acid sequences of domain R and domain H are identical, the amino acid sequences of domain A is different from domain R and H, the amino acid sequences of domain S and domain I are identical, the amino acid sequences of domain B is different from domain S and I, the amino acid sequences of domain T and domain L are identical, the amino acid sequences of domain F is different from domain T and L, the amino acid sequences of domain U and domain M are identical, the amino acid sequences of domain G is different from domain U and M and the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigenic epitope, the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigenic epitope, and the domain R and domain T form a third antigen binding site specific for the second antigenic epitope.
  • the trivalent construct contains one copy of the antigen binding site (ABS) specific for a first epitope of the target receptor.
  • the antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an R:T antigen binding site.
  • the antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the bispecific trivalent construct contains two copies of the antigen binding site specific for a first epitope of the target receptor.
  • a first antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an R:T antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • a first antigen binding site specific for a first epitope of the target receptor is an R:T antigen binding site and a second antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the second antigenic target is a second epitope of the target receptor.
  • the first epitope and the second epitope are non-overlapping epitopes.
  • the second antigenic target is an epitope of a second protein.
  • the second protein is a second cell surface receptor.
  • Trivalent 1x2 Trispecific Constructs [1(A)x2(B-C)] [00492]
  • the amino acid sequences of domain R, domain A, and domain H are different
  • the amino acid sequences of domain S, domain B, and domain I are different
  • the amino acid sequences of domain T, domain F, and domain L are different
  • the amino acid sequences of domain U, domain G, and domain M are different
  • the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigenic epitope
  • the interaction between the H domain and the L domain form a second antigen binding site specific for a second antigenic epitope
  • the domain R and domain T form a third antigen binding site specific for a third antigenic epitope.
  • domain S has a constant region sequence that is a CL from a kappa light chain and domain U has a constant region sequence that is a CH1 from an IgG1 isotype, as discussed in more detail in Sections 6.4.3.9.2 and 6.4.3.9.1, respectively.
  • domain S and domain U have CH3 sequences such that they specifically associate with each other, as discussed in more detail in Section 6.4.3.15.
  • the antigen binding site specific for a first epitope of the target receptor is an A:F antigen binding site. In various embodiments, the antigen binding site specific for a first epitope of the target receptor is an R:T antigen binding site. In various embodiments, the antigen binding site specific for a first epitope of the target receptor is an H:L antigen binding site.
  • the second antigenic target is a second epitope of the target receptor. In some aspects, the first epitope and the second epitope are non-overlapping epitopes. In some embodiments, the second antigenic target is a first epitope of a second protein. In some embodiments, the third antigenic target is a third epitope of the target receptor. In some embodiments, the third antigenic target is a second epitope of a second protein. In some aspects, the first epitope of the second protein and the second epitope of the second protein are non-overlapping epitopes. In some embodiments, the third antigenic target is a first epitope of a third protein.
  • the second protein or third protein is a second or third cell surface receptor.
  • 6.4.4.1. Trivalent 1x2 Monospecific Constructs [00497] With reference to FIG.4A, in a variety of embodiments, the amino acid sequences of domain R, domain A, and domain H are identical, the amino acid sequences of domain S and domain B are identical, the amino acid sequences of domain T, domain F, and domain L are identical, and the amino acid sequences of domain U and domain G are identical; and the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigenic epitope, the interaction between the H domain and the L domain form a second antigen binding site specific for the first antigenic epitope, and the domain R and domain T form a third antigen binding site specific for the first antigenic epitope. These trivalent constructs are monospecific; all three antigen binding sites are specific for a first epitope of the target receptor. 6.4.5. Tetravalent 2x2 Binding Mol
  • the binding molecules have 4 antigen binding sites and are therefore termed“tetravalent.”
  • the binding molecules further comprise a fifth and a sixth polypeptide chain, wherein (a) the first polypeptide chain further comprises a domain N and a domain O, wherein the domains are arranged, from N-terminus to C-terminus, in a N-O-A-B-D-E orientation; (b) the third polypeptide chain further comprises a domain R and a domain S, wherein the domains are arranged, from N-terminus to C-terminus, in a R-S-H-I-J-K orientation; (c) the binding molecule further comprises a fifth and a sixth polypeptide chain, wherein the fifth and a sixth polypeptide chain, wherein the fifth polypeptide chain.
  • polypeptide chain comprises a domain P and a domain Q, wherein the domains are arranged, from N-terminus to C-terminus, in a P-Q orientation
  • the sixth polypeptide chain comprises a domain T and a domain U, wherein the domains are arranged, from N-terminus to C-terminus, in a T-U orientation
  • the first and the fifth polypeptides are associated through an interaction between the N and the P domains and an interaction between the O and the Q domains
  • the third and the sixth polypeptides are associated through an interaction between the R and the T domains and an interaction between the S and the U domains to form the binding molecule.
  • the domain O is connected to domain A through a peptide linker and the domain S is connected to domain H through a peptide linker.
  • the peptide linker connecting domain O to domain A and connecting domain S to domain H is a 6 amino acid GSGSGS peptide sequence (SEQ ID NO: 232), as described in more detail in Section 6.4.6.6. 6.4.5.1.Tetravalent 2x2 Bispecific Constructs
  • the amino acid sequences of domain N and domain A are identical, the amino acid sequences of domain H and domain R are identical, the amino acid sequences of domain O and domain B are identical, the amino acid sequences of domain I and domain S are identical, the amino acid sequences of domain P and domain F are identical, the amino acid sequences of domain L and domain T are identical, the amino acid sequences of domain Q and domain G are identical, the amino acid sequences of domain M and domain U are identical; and wherein the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen, the domain N and domain P form a second antigen binding site specific for the first antigen, the interaction between the H domain and the L domain form a third antigen binding site specific for a second antigen, and the interaction between the R domain and the T domain form a fourth antigen binding site specific for the second antigen.
  • the amino acid sequences of domain H and domain A are identical, the amino acid sequences of domain N and domain R are identical, the amino acid sequences of domain I and domain B are identical, the amino acid sequences of domain O and domain S are identical, the amino acid sequences of domain L and domain F are identical, the amino acid sequences of domain P and domain T are identical, the amino acid sequences of domain M and domain G are identical, the amino acid sequences of domain Q and domain U are identical; and wherein the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen, the domain N and domain P form a second antigen binding site specific for a second antigen, the interaction between the H domain and the L domain form a third antigen binding site specific for the first antigen, and the interaction between the R domain and the T domain form a fourth antigen binding site specific for the second antigen.
  • FIG.4E shows the overall architecture of a 2x2 tetravalent bispecific construct“BC22 -2x2”.
  • the 2x2 tetravalent bispecific represents a “BC1” scaffold, as described in greater detail herein and in Section 6.4.7.1, with duplications of each variable domain-constant domain segment.
  • the amino acid sequences of domain N, domain A, domain H and domain R are identical, the amino acid sequences of domain O and domain B are identical, the amino acid sequences of domain I and domain S are identical, the amino acid sequences of domain P, domain F, domain L, and domain T are identical, the amino acid sequences of domain Q and domain G are identical, the amino acid sequences of domain M and domain U are identical; and wherein the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen, the domain N and domain P form a second antigen binding site specific for the first antigen, the interaction between the H domain and the L domain form a third antigen binding site specific for the first antigen, and the interaction between the R domain and the T domain form a fourth antigen binding site specific for the first antigen.
  • the amino acid sequences of domain N, domain A, domain H and domain R are identical, the amino acid sequences of domain I and domain B are identical, the amino acid sequences of domain O and domain S are identical, the amino acid sequences of domain P, domain F, domain L, and domain T are identical, the amino acid sequences of domain M and domain G are identical, the amino acid sequences of domain Q and domain U are identical; and wherein the interaction between the A domain and the F domain form a first antigen binding site specific for a first antigen, the domain N and domain P form a second antigen binding site specific for the first antigen, the interaction between the H domain and the L domain form a third antigen binding site specific for the first antigen, and the interaction between the R domain and the T domain form a fourth antigen binding site specific for the first antigen.
  • the amino acid sequence that forms a junction between the C-terminus of a VL domain and the N-terminus of a CH3 domain is an engineered sequence.
  • one or more amino acids are deleted or added in the C- terminus of the VL domain.
  • the junction connecting the C-terminus of a VL domain and the N-terminus of a CH3 domain is one of the sequences described in Table 1 below.
  • A111 is deleted in the C-terminus of the VL domain.
  • one or more amino acids are deleted or added in the N- terminus of the CH3 domain.
  • P343 is deleted in the N-terminus of the CH3 domain.
  • P343 and R344 are deleted in the N-terminus of the CH3 domain.
  • one or more amino acids are deleted or added to both the C-terminus of the VL domain and the N-terminus of the CH3 domain.
  • A111 is deleted in the C-terminus of the VL domain and P343 is deleted in the N-terminus of the CH3 domain.
  • A111 and V110 are deleted in the C-terminus of the VL domain.
  • A111 and V110 are deleted in the C-terminus of the VL domain and the N-terminus of the CH3 domain has a P343V mutation.
  • the amino acid sequence that forms a junction between the C-terminus of a VH domain and the N-terminus of a CH3 domain is an engineered sequence.
  • one or more amino acids are deleted or added in the C- terminus of the VH domain.
  • the junction connecting the C-terminus of a VH domain and the N-terminus of the CH3 domain is one of the sequences described in Table 2 below.
  • K117 and G118 are deleted in the C-terminus of the VH domain.
  • one or more amino acids are deleted or added in the N-terminus of the CH3 domain.
  • P343 is deleted in the N-terminus of the CH3 domain.
  • P343 and R344 are deleted in the N-terminus of the CH3 domain.
  • P343, R344, and E345 are deleted in the N- terminus of the CH3 domain.
  • one or more amino acids are deleted or added to both the C-terminus of the VH domain and the N-terminus of the CH3 domain.
  • T116, K117, and G118 are deleted in the C-terminus of the VH domain.
  • the N-terminus of the CH2 domain has a “hinge” region amino acid sequence.
  • hinge regions are sequences of an antibody heavy chain that link the N-terminal variable domain-constant domain segment of an antibody (e.g., the segment corresponding to domain A connected to domain B) and a CH2 domain of an antibody.
  • the hinge region typically provides both flexibility between the N-terminal variable domain-constant domain segment and CH2 domain, as well as amino acid sequence motifs that form disulfide bridges between heavy chains (e.g. the first and the third polypeptide chains).
  • the hinge region amino acid sequence is SEQ ID NO:18.
  • the constant region domain is a CH3 amino acid sequence
  • the CH3 amino acid sequence is extended at the C-terminus at the junction between the C- terminus of the CH3 domain and the N-terminus of a CH2 domain.
  • a CH3 amino acid sequence is extended at the C-terminus at the junction between the C- terminus of the CH3 domain and a hinge region, which in turn is connected to the N-terminus of a CH2 domain.
  • the CH3 amino acid sequence is extended by inserting a PGK tripeptide sequence followed by the DKTHT motif (SEQ ID NO: 233) of an IgG1 hinge region.
  • the extension at the C-terminus of the CH3 domain incorporates amino acid sequences that can form a disulfide bond with orthogonal C-terminal extension of another CH3 domain.
  • the extension at the C- terminus of the CH3 domain incorporates a KSC tripeptide sequence that is followed by the DKTHT motif (SEQ ID NO: 233) of an IgG1 hinge region that forms a disulfide bond with orthogonal C-terminal extension of another CH3 domain that incorporates a GEC motif of a kappa light chain.
  • a CL amino acid sequence is connected through its C- terminus to a hinge region, which in turn is connected to the N-terminus of a CH2 domain. Hinge region sequences are described in greater detail herein (see“Bivalent constructs”) and in Section 6.4.6.
  • the hinge region amino acid sequence is SEQ ID NO:18. 6.4.6.5. Junctions Connecting CH2 C-terminus to Constant
  • a CH2 amino acid sequence is connected through its C- terminus to the N-terminus of a constant region domain. Constant regions are described in more detail herein (see“Bivalent constructs”) and in Section 6.4.3.5.
  • the CH2 sequence is connected to a CH3 sequence via its endogenous sequence.
  • the CH2 sequence is connected to a CH1 or CL sequence. Examples discussing connecting a CH2 sequence to a CH1 or CL sequence are described in more detail in US Pat. No.8,242,247, which is hereby incorporated in its entirety.
  • heavy chains of antibodies are extended at their N-terminus to include additional domains that provide additional ABSs.
  • the C-terminus of the constant region domain amino acid sequence of a domain O and/or a domain S is connected to the N-terminus of the variable region domain amino acid sequence of a domain A and/or a domain H, respectively.
  • the constant region domain is a CH3 amino acid sequence and the variable region domain is a VL amino acid sequence.
  • the constant region domain is a CL amino acid sequence and the variable region domain is a VL amino acid sequence.
  • the constant region domain is connected to the variable region domain through a peptide linker.
  • the peptide linker is a 6 amino acid GSGSGS peptide sequence (SEQ ID NO: 232).
  • light chains of antibodies are extended at their N-terminus to include additional variable domain- constant domain segments of an antibody.
  • the constant region domain is a CH1 amino acid sequence and the variable region domain is a VH amino acid sequence. 6.4.7.
  • the bivalent binding molecule has a first, second, third, and fourth polypeptide chain, wherein (a) the first polypeptide chain comprises a domain A, a domain B, a domain D, and a domain E, wherein the domains are arranged, from N-terminus to C-terminus, in a A-B-D-E orientation, and domain A has a first VL amino acid sequence, domain B has a human IgG1 CH3 amino acid sequence with a T366K mutation and a C-terminal extension incorporating a KSC tripeptide sequence that is followed by the DKTHT motif (SEQ ID NO: 233) of an IgG1 hinge region, domain D has a human IgG1 CH2 amino acid sequence, and domain E has human IgG1 CH3 amino acid with a S354C and T366W mutation; (b) the second polypeptide
  • domain A and domain F form a first antigen binding site specific for a first antigen
  • domain H and domain L form a second antigen binding site specific for a second antigen.
  • domain A and domain F form a first antigen binding site specific for a first antigen
  • domain H and domain L form a second antigen binding site specific for the first antigen.
  • the first polypeptide chain has a scaffold sequence SEQ ID NO:23
  • the second polypeptide chain has a scaffold sequence SEQ ID NO:24
  • the third polypeptide chain has a scaffold sequence SEQ ID NO:25
  • the fourth polypeptide chain has a scaffold sequence SEQ ID NO:26, as described in more detail herein and in Section 6.10.2.1.
  • the bivalent binding molecule has a first, second, third, and fourth polypeptide chain, wherein (a) the first polypeptide chain comprises a domain A, a domain B, a domain D, and a domain E, wherein the domains are arranged, from N-terminus to C-terminus, in a A-B-D-E orientation, and domain A has a first VL amino acid sequence, domain B has a human IgG1 CH3 amino acid sequence with a C-terminal extension incorporating a KSC tripeptide sequence that is followed by the DKTHT motif (SEQ ID NO: 233) of an IgG1 hinge region, domain D has a human IgG1 CH2 amino acid sequence, and domain E has human IgG1 CH3 amino acid with a S354C and a T366W mutation; (b) the second polypeptide chain
  • domain A and domain F form a first antigen binding site specific for a first antigen
  • domain H and domain L form a second antigen binding site specific for a second antigen
  • domain A and domain F form a first antigen binding site specific for a first antigen
  • domain H and domain L form a second antigen binding site specific for the first antigen
  • the bivalent (1x1) binding molecule has a first, second, third, and fourth polypeptide chain, wherein (a) the first polypeptide chain comprises a domain A, a domain B, a domain D, and a domain E, wherein the domains are arranged, from N-terminus to C-terminus, in a A-B-D-E orientation, and domain A has a first VL amino acid sequence, domain B has a human IgG1 CH3 amino acid sequence with a Y349C mutation and a C-terminal extension incorporating a PGK tripeptide sequence that is followed by the DKTHT motif (SEQ ID NO: 233) of an IgG1 hinge region, domain D has a human IgG1 CH2 amino acid sequence, and domain E has a human IgG1 CH3 amino acid with a S354C and a
  • domain A and domain F form a first antigen binding site specific for a first antigen
  • domain H and domain L form a second antigen binding site specific for a second antigen
  • domain A and domain F form a first antigen binding site specific for a first antigen
  • domain H and domain L form a second antigen binding site specific for the first antigen
  • the bivalent (1x1) binding molecule has a first, second, third, and fourth polypeptide chain, wherein (a) the first polypeptide chain comprises a domain A, a domain B, a domain D, and a domain E, wherein the domains are arranged, from N-terminus to C-terminus, in a A-B-D-E orientation, and domain A has a first VL amino acid sequence, domain B has a human IgG1 CH3 amino acid sequence with a Y349C mutation, a P343V mutation, and a C-terminal extension incorporating a PGK tripeptide sequence that is followed by the DKTHT motif (SEQ ID NO: 233) of an IgG1 hinge region, domain D has a human IgG1 CH2 amino acid sequence, and domain E has human IgG1 CH3 amino acid with a SEQ ID NO: 233) of an IgG1 hinge region, domain D has a human IgG1 CH2 amino acid sequence, and domain E has human IgG1
  • domain A and domain F form a first antigen binding site specific for a first antigen
  • domain H and domain L form a second antigen binding site specific for a second antigen
  • domain A and domain F form a first antigen binding site specific for a first antigen
  • domain H and domain L form a second antigen binding site specific for the first antigen.
  • Exemplary Trivalent Binding Molecules 6.4.8.1. Trivalent 2x1 Bispecific B-Body“BC1-2x1”
  • FIG.3B illustrates the salient features of a trivalent 2x1 bispecific B-Body binding molecule further comprising a fifth polypeptide chain and as described below:
  • the binding molecules further comprise a sixth polypeptide chain, wherein (a) the third polypeptide chain further comprises a domain R and a domain S, wherein the domains are arranged, from N-terminus to C-terminus, in a R-S-H-I-J-K orientation, and wherein domain R has the first VL amino acid sequence and domain S has a human IgG1 CH3 amino acid sequence with a Y349C mutation and a C-terminal extension incorporating a PGK tripeptide sequence that is followed by GSGSGS linker peptide (SEQ ID NO: 232) connecting domain S to domain H; (b) the binding molecule further comprises a sixth polypeptide chain, comprising: a domain T and a domain U, wherein the domains are arranged, from N-terminus to C-
  • binding molecule platforms are not limiting.
  • the antigen binding sites described herein, including specific CDR subsets, can be formatted into any binding molecule platform including, but not limited to, full-length antibodies, Fab fragments, Fvs, scFvs, tandem scFvs, Diabodies, scDiabodies, DARTs, tandAbs, minibodies, camelid VHH, and other antibody fragments or formats known to those skilled in the art.
  • Exemplary antibody and antibody fragment formats are described in detail in Brinkmann et al. (MABS, 2017, Vol.9, No.2, 182–212), herein incorporated by reference for all that it teaches. 6.6. Further modifications
  • binding molecules described herein have additional modifications. 6.6.1. Antibody-Drug Conjugates
  • the binding molecule is conjugated to a therapeutic agent (i.e. drug) to form a binding molecule-drug conjugate.
  • therapeutic agents include, but are not limited to, chemotherapeutic agents, imaging agents (e.g., radioisotopes), immune modulators (e.g., cytokines, chemokines, or checkpoint inhibitors), and toxins (e.g., cytotoxic agents).
  • the therapeutic agents are attached to the binding molecule through a linker peptide, as discussed in more detail in herein and in Section 6.6.3.
  • ADCs antibody-drug conjugates
  • the binding molecule has modifications that comprise one or more additional binding moieties.
  • the binding moieties are antibody fragments or antibody formats including, but not limited to, full-length antibodies, Fab fragments, Fvs, scFvs, tandem scFvs, Diabodies, scDiabodies, DARTs, tandAbs, minibodies, camelid VHH, and other antibody fragments or formats known to those skilled in the art. Exemplary antibody and antibody fragment formats are described in detail in
  • the one or more additional binding moieties are attached to the C-terminus of the first or third polypeptide chain. In particular embodiments, the one or more additional binding moieties are attached to the C-terminus of both the first and third polypeptide chain. In particular embodiments, the one or more additional binding moieties are attached to the C-terminus of both the first and third polypeptide chains. In certain embodiments, individual portions of the one or more additional binding moieties are separately attached to the C-terminus of the first and third polypeptide chains such that the portions form the functional binding moiety.
  • the one or more additional binding moieties are attached to the N-terminus of any of the polypeptide chains (e.g., the first, second, third, fourth, fifth, or sixth polypeptide chains).
  • individual portions of the additional binding moieties are separately attached to the N-terminus of different polypeptide chains such that the portions form the functional binding moiety.
  • the one or more additional binding moieties are specific for a different antigen or epitope of the ABSs within the binding molecule.
  • the one or more additional binding moieties are specific for the same antigen or epitope of the ABSs within the binding molecule. In certain embodiments, wherein the modification is two or more additional binding moieties, the additional binding moieties are specific for the same antigen or epitope. In certain embodiments, wherein the modification is two or more additional binding moieties, the additional binding moieties are specific for different antigens or epitopes.
  • the one or more additional binding moieties are attached to the binding molecule using in vitro methods including, but not limited to, reactive chemistry and affinity tagging systems, as discussed in more detail herein and in Section 6.6.3.
  • the one or more additional binding moieties are attached to the binding molecule through Fc-mediated binding (e.g. Protein A/G).
  • the one or more additional binding moieties are attached to the binding molecule using recombinant DNA techniques, such as encoding the nucleotide sequence of the fusion product between the binding molecule and the additional binding moieties on the same expression vector (e.g. plasmid). 6.6.3. Functional/Reactive Groups
  • the binding molecule has modifications that comprise functional groups or chemically reactive groups that can be used in downstream processes, such as linking to additional moieties (e.g., drug conjugates and additional binding moieties, as discussed in more detail herein and in Sections 6.6.1. and 6.6.2.) and downstream purification processes.
  • additional moieties e.g., drug conjugates and additional binding moieties, as discussed in more detail herein and in Sections 6.6.1. and 6.6.2.
  • the modifications are chemically reactive groups including, but not limited to, reactive thiols (e.g., maleimide based reactive groups), reactive amines (e.g., N-hydroxysuccinimide based reactive groups),“click chemistry” groups (e.g., reactive alkyne groups), and aldehydes bearing formylglycine (FGly).
  • the modifications are functional groups including, but not limited to, affinity peptide sequences (e.g., HA, HIS, FLAG, GST, MBP, and Strep systems etc.).
  • the functional groups or chemically reactive groups have a cleavable peptide sequence.
  • the cleavable peptide is cleaved by means including, but not limited to, photocleavage, chemical cleavage, protease cleavage, reducing conditions, and pH conditions.
  • protease cleavage is carried out by intracellular proteases.
  • protease cleavage is carried out by extracellular or membrane associated proteases.
  • ADC therapies adopting protease cleavage are described in more detail in Choi et al. (Theranostics, 2012; 2(2): 156–178.), the entirety of which is hereby incorporated by reference for all it teaches. 6.6.4. Reduced Effector Function
  • the binding molecule has one or more engineered mutations in an amino acid sequence of an antibody domain that reduce the effector functions naturally associated with antibody binding.
  • Effector functions include, but are not limited to, cellular functions that result from an Fc receptor binding to an Fc portion of an antibody, such as antibody- dependent cellular cytotoxicity (ADCC, also referred to as antibody-dependent cell-mediated cytotoxicity), complement fixation (e.g. C1q binding), antibody dependent cellular-mediated phagocytosis (ADCP), and opsonization.
  • ADCC antibody- dependent cellular cytotoxicity
  • complement fixation e.g. C1q binding
  • ADCP antibody dependent cellular-mediated phagocytosis
  • opsonization e.g. C1q binding
  • Engineered mutations that reduce the effector functions are described in more detail in U.S. Pub. No.2017/0137530, Armour, et al. (Eur. J.
  • the binding molecule has one or more engineered mutations in an amino acid sequence of an antibody domain that reduce binding of an Fc portion of the binding molecule by FcR receptors.
  • the FcR receptors are FcRg receptors.
  • the FcR receptors are FcgRIIa and/or FcgRIIIA receptors.
  • the one or more engineered mutations that reduce effector function are mutations in a CH2 domain of an antibody.
  • the one or more engineered mutations comprise a mutation at position L234 of the CH2 domain.
  • the mutation at position L234 is L234A.
  • the mutation at position L234 is L234G.
  • the one or more engineered mutations comprise a mutation at position L235 of the CH2 domain.
  • the mutation at position L235 is L235A.
  • the mutation at position L235 is L235G.
  • the one or more engineered mutations comprise mutations at positions L234 and L235 of the CH2 domain.
  • the mutations at positions L234 and L235 of the CH2 domain are L234A and L235A.
  • the mutations at positions L234 and L235 of the CH2 domain are L234G and L235G.
  • the one or more engineered mutations comprise a mutation at position P329 of the CH2 domain.
  • the mutation at position P329 of the CH2 domain is P329A.
  • the mutation at position P329 of the CH2 domain is P329G.
  • the mutation at position P329 of the CH2 domain is P329K.
  • the one or more engineered mutations are at positions L234, L235, and P329 of the CH2 domain.
  • the one or more engineered mutations are L234A, L235A, and P329A of the CH2 domain.
  • the one or more engineered mutations are L234A, L235A, and P329G of the CH2 domain.
  • the one or more engineered mutations are L234A, L235A, and P329K of the CH2 domain.
  • the one or more engineered mutations are L234G, L235G, and P329A of the CH2 domain.
  • the one or more engineered mutations are L234G, L235G, and P329G of the CH2 domain. In particular embodiments, the one or more engineered mutations are L234G, L235G, and P329K of the CH2 domain.
  • the one or more engineered mutations are as provided in the Table 8 below.
  • compositions that comprise a binding molecule as described herein and a pharmaceutically acceptable carrier or diluent.
  • the pharmaceutical composition is sterile.
  • the pharmaceutical composition comprises the binding molecule at a concentration of 0.1 mg/ml– 100 mg/ml.
  • the pharmaceutical composition comprises the binding molecule at a concentration of 0.5 mg/ml, 1 mg/ml, 1.5 mg/ml, 2 mg/ml, 2.5 mg/ml, 5 mg/ml, 7.5 mg/ml, or 10 mg/ml.
  • the pharmaceutical composition comprises the binding molecule at a concentration of more than 10 mg/ml.
  • the binding molecule is present at a concentration of 20 mg/ml, 25 mg/ml, 30 mg/ml, 35 mg/ml, 40 mg/ml, 45 mg/ml, or even 50 mg/ml or higher. In particular embodiments, the binding molecule is present at a concentration of more than 50 mg/ml.
  • the pharmaceutical compositions are described in more detail in U.S. Pat No.8,961,964, U.S. Pat No.8,945,865, U.S. Pat No.8,420,081, U.S. Pat No.6,685,940, U.S. Pat No.6,171,586, U.S. Pat No.8,821,865, U.S. Pat No.9,216,219, US application 10/813,483, WO 2014/066468, WO 2011/104381, and WO 2016/180941, each of which is incorporated herein in its entirety. 6.8. Methods of Manufacturing [00550]
  • the binding molecules described herein can readily be manufactured by expression using standard cell free translation, transient transfection, and stable transfection approaches currently used for antibody manufacture.
  • Expi293 cells are described in more detail in U.S. Pat No.8,961,964, U.S. Pat No.8,945,865, U.S. Pat No.8,420,081, U.S. Pat No.6,685,940, U.S.
  • ThermoFisher can be used for production of the binding molecules using protocols and reagents from ThermoFisher, such as ExpiFectamine, or other reagents known to those skilled in the art, such as polyethylenimine as described in detail in Fang et al. (Biological Procedures Online, 2017, 19:11), herein incorporated by reference for all it teaches.
  • the expressed proteins can be readily purified using a CH1 affinity resin, such as the CaptureSelect CH1 resin and provided protocol from ThermoFisher. Further purification can be effected using ion exchange chromatography as is routinely used in the art. 6.9. Methods of Treatment [00552] In another aspect, methods of treatment are provided. The methods comprise administering a therapeutically effective amount of the pharmaceutical compositions described herein to a patient in need thereof. [00553] In some embodiments, the method of treatment is a method of treating a proliferative disease in a subject in need thereof. The disease may be cancer.
  • the target receptor is a member of the TNFRSF, and the disease to be treated is cancer.
  • the target receptor is OX40
  • TNFRSF4 TNFRSF4
  • CD40 TNFRSF5
  • 4-1BB TNFRSF9
  • the antibody constructs or ABPs provided herein can be used in a combination therapy regimen for the treatment or prevention of a proliferative disease in a subject.
  • At least one antibody construct is administered with a T cell redirection therapy.
  • at least ABP is administered with a T cell redirection therapy (e.g., T-cell redirecting antibody or the like).
  • T-cell redirecting therapy redirects T cells to attack tumor such as a solid tumor or blood tumor. Any T-cell redirecting therapy can be used in the methods of the disclosure.
  • T-cell redirecting therapies that can be used with the methods include but are not limited to, T cell redirecting therapies with intracellular T-cell activation domains such as CD28, 4-1BB, OX40 and CD3z or to specific antigen-bearing cells such as antigens CD19, BCMA, CD123, Mesothelin, GD2, CD20, CD33, HER2, CD22, CD30, PSMA, EGFRvIII, EGFR, CD38, EpCAM, PSCA, CEA (CEACAM5), Glypican-3, Flt3, NKG2D ligands, Claudin, DLL3, CS1 (SLAMF7), MUC16, Lewis-Y, cMet, or a combination thereof.
  • T cell redirecting therapies with intracellular T-cell activation domains such as CD28, 4-1BB, OX40 and CD3z or to specific antigen-bearing cells such as antigens CD19, BCMA, CD123, Mesothelin, GD2, CD20, CD33,
  • a method of treating a proliferative disease in a subject in need thereof comprising co-administering to the subject (a) a first antigen binding protein (ABP), wherein the first ABP is capable of (i) binding a cell surface receptor target, which cell surface receptor target requires clustering for agonist activity, and (ii) clustering the receptor target on the cell surface in the absence of an independent cross-linking agent or one or more Fc mutations that drive hexamer formation, and (b) a second ABP, wherein the second ABP is capable of binding (i) a cell surface protein present on a tumor cell and (ii) a cell surface antigen present on an immune cell.
  • ABP antigen binding protein
  • “co-administering,”“co-administration,”“co-treating,” and“co- treatment” can include, without limitation, administering two or more therapeutic agents simultaneously, concurrently, or sequentially, and does not imply specific time limits or ordering of the sequential administrations unless otherwise indicated.
  • the first ABP may be administered simultaneously with the second ABP, or may be administered prior to administration of the second ABP, or may be administered after administration of the second ABP.
  • the first and second ABPs are administered to the subject within 5 minutes, 15 minutes, 30 minutes, 45 minutes, 1 hour, 2 hours, 4 hours, 6 hours, 12 hours, 24 hours, 48 hours, 72 hours, 96 hours, 1 week.2 weeks, 3 weeks, 4 weeks, 5 weeks, 6 weeks, 8 weeks, or 12 weeks of each other. In some embodiments, the first and second ABPs are administered to the subject within 1-3 days of each other. In some embodiments, the therapeutic agents are present in a cell of the subject at the same time. In some embodiments, the therapeutic agents are present in the subject’s body at the same time. In some embodiments, the therapeutic agents exert their biological effects on a cell of the subject or in the subject’s body at the same time. In some embodiments, the therapeutic agents are administered as a single pharmaceutical composition or unit dosage form. In some embodiments, the therapeutic agents are administered as separate pharmaceutical
  • compositions or unit dosage forms comprising the first ABP, pharmaceutical compositions comprising the second ABP, and pharmaceutical compositions comprising the first ABP and the second ABP.
  • kits comprising the first ABP, the second ABP, and instructions for use according to the treatment methods provided herein.
  • the first ABP is any of the receptor-clustering ABPs described herein.
  • the cell surface protein present on the tumor cell is selected from ADP-Ribosyl Cyclase/Cyclic ADP-Ribose Hydrolase 1 (CD38), B-Cell Receptor CD22, B-Lymphocyte Antigen CD20 (CD20), B7-H3, C-Type Lectin Domain Family 12 member A (CLL-1), C-X-C Chemokine Receptor Type 5 (CXCR5), Cadherin-17 (CDH17), Cadherin-3, Carcinoembryonic Antigen-Related Cell Adhesion Molecule 5, CD123, CD133, CD155, CD19, CD37 Antigen, CD44 Antigen, CD79b, CEA Family members, Chondroitin Sulfate Proteoglycan 4 (NG2), Claudin-6, CLDN18.2, CLEC12A, Cytotoxic T-Lymphocyte Protein 4 (CTLA-4; CD152), Dipeptidyl Pepti
  • Macrophage-Stimulating Protein Receptor (MSPR), Mesothelin, Metalloreductase STEAP1, MHC- Peptide Complexes, MUC-1, MUC16, Mucin-16 (CA125; MUC-16), Myeloid Cell Surface Antigen CD33, p-Cadherin, Programmed Cell Death 1 (PD-1), Prominin-1 (CD133), Prostate Stem Cell Antigen (PSCA), Prostate-Specific Antigen (PSA; Kallikrein-3), PSMA, Receptor Tyrosine-Protein Kinase erbB-2 (HER2), Receptor Tyrosine-Protein Kinase erbB-3 (HER3), Receptor-Type Tyrosine-Protein Kinase FLT3 (FLT-3), Roundabout Homolog 1 (ROBO1), S antigen (HBV), Somatostatin Receptor Type 2 (SS2R), SSTR2, T-Cell-Specific Surface Glycoprotein CD28, Tropho
  • the cell surface protein present on the tumor cell is selected from BCMA, CD123, CD19, CD20, CEA, CLDN18.2, CLEC12A, CLL-1, EGFR, EpCAM, FCRH5, FLT3, gpA33, GPC3, GPRC5D, HER-2, HIV, HLA-A2, MHC- Peptide Complexes, MUC-1, MUC16, PSMA, 5T4, B7-H3, B7-H4, CD123, CD133, CD155, CD79b, DLL3, EGFRviii, EMP2, EphA2, FOLR-alpha, gp120,IL-13R, IL-23, Mesothelin, p-Cadherin, PSMA, and SSTR2.
  • the cell surface protein present on the tumor cell is BCMA.
  • the second ABP is capable of redirecting an immune cell to the tumor cell. In some embodiments, the second ABP is capable of mediating immune-cell directed killing of the tumor cell.
  • the immune cell is a T cell, optionally an effector T cell. In some embodiments of the second ABP, the immune cell is a CD8+ killer T cell. In some embodiments of the second ABP, the cell surface antigen present on the T cell is CD3, optionally CD3e. In some embodiments of the second ABP, the immune cell is a NK cell. In some embodiments, the cell surface antigen present on the NK cell is CD16.
  • co-administration of the first ABP and the second ABP increases the antiproliferative or tumor-cell killing effect of the second ABP, as compared to administration of the second ABP alone. 6.10.
  • the following examples are provided by way of illustration, not limitation. 6.10.1. Methods [00568] Non-limiting, illustrative methods for the purification of the various antigen-binding proteins and their use in various assays are described in more detail below. 6.10.1.1. Expi293 Expression
  • the various antigen-binding proteins tested were expressed using the Expi293 transient transfection system according to manufacturer’s instructions. Briefly, four plasmids coding for four individual chains were mixed at 1:1:1:1 mass ratio, unless otherwise stated, and transfected with ExpiFectamine 293 transfection kit to Expi 293 cells. Cells were cultured at 37°C with 8% CO2, 100% humidity and shaking at 125 rpm. Transfected cells were fed once after 16-18 hours of transfections. The cells were harvested at day 5 by centrifugation at 2000 g for 10 munities. The supernatant was collected for affinity chromatography purification. 6.10.1.2. Protein A and Anti-CH1 Purification
  • the MonoS column was equilibrated with buffer A 10 mM MES pH 6.0. The samples were loaded onto the column at 2 ml/min. The sample was eluted using a 0-30% gradient with buffer B (10 mM MES pH 6.0, 1 M sodium chloride) over 6 CV. The elution was monitored by absorbance at 280 nm and the purity of the samples was calculated by peak integration to identify the abundance of the monomer peak and contaminants peaks. The monomer peak and contaminant peaks were separately pooled for analysis by SDS-PAGE as described above. 6.10.1.5. Analytical SEC Chromatography
  • Samples containing the various separated antigen-binding proteins were analyzed by mass spectrometry to confirm the correct species by molecular weight. All analysis was performed by a third-party research organization. Briefly, samples were treated with a cocktail of enzymes to remove glycosylation. Samples were both tested in the reduced format to specifically identify each chain by molecular weight. Samples were all tested under non-reducing conditions to identify the molecular weights of all complexes in the samples. Mass spec analysis was used to identify the number of unique products based on molecular weight. 6.10.1.7. NFkB Luc2 OX40 Jurkat T cell Stimulation Assay
  • the NFkB Luc2 OX40 Jurkat T cell Stimulation Assay (Promega, Cat# CS197704, CS197707) was performed according to manufacturer’s instructions. Briefly, the Thaw-and-Use Jurkat/OX40 cells were thawed at 37 degC and diluted in assay buffer as recommended. The Thaw-and-Use cells were dispensed into 96-well plates (50 uL/well) and incubated overnight in a CO2 incubator at 37 degC. The next day, serial dilutions of the OX40 ligand are made as a standard control at 3X of the final concentration.
  • T-cell stimulation was measured using multiple assays to follow the cytokine production as well as impact of T cell proliferation.
  • Measurement of T cell activation through measurement of cytokine (IL-2, TNFa, and IFNg) production was performed using the Cytokine Screen Opteia ELISA Kit (BD Cat# 555212, 555190, & 555142) according to the manufacturer’s instructions. Briefly, 96-well ELISA plates were coated with the specific capture antibody overnight using 100 uL/well according to instructions. The ELISA plates were blocked with 150 uL RPMI per well.
  • PrestoBlue reagent was added directly to the wells at 1/10 th the volume of media within the wells. The plates were then incubated at 37 0 C/5% CO2 for 10 min– overnight. The fluorescence of each well was determined using a Safire plate reader (Tecan) with an excitation wavelength of 560 nm and an emission wavelength of 590 nm. 6.10.1.9. IncuCyte System
  • T cell activation was monitored using the IncuCyte system (Sartorius). The kinetics of T cell activation were monitored for 3 to 6 days by microscopy in a controlled growth environment. T cell activation was charted using cell size measurement to track the growth and proliferation of T cell clusters. 6.10.2.
  • Example 1 Bivalent Anti-OX40 Agonist Antibodies 6.10.2.1. Human anti-OX40 antibody discovery by phage display [00579] Phage display of human Fab libraries was carried out using standard protocols. Biotinylated extracellular domain of human OX40 protein was purchased from Acro
  • Phage clones were screened for the ability to bind human OX40 by phage ELISA using standard protocols. Briefly, Fab-formatted phage libraries were constructed using expression vectors capable of replication and expression in phage (also referred to as a phagemid). Both the heavy chain and the light chain were encoded for in the same expression vector, where the heavy chain was fused to a truncated variant of the phage coat protein pIII. The light chain and heavy chain were expressed as separate polypeptides, and the light chain and heavy chain-pIII fusion assemble in the bacterial periplasm, where the redox potential enables disulfide bond formation, to form the antibody containing the candidate ABS
  • the library was created using sequences derived from a specific human heavy chain variable domain (VH3-23) and a specific human light chain variable domain (Vk-1). Light chain variable domains within the screened library were generated with diversity introduced into the VL CDR3 (L3) and where the light chain VL CDR1 (L1) and CDR2 (L2) remained the human germline sequence. For the screened library, all three CDRs of the VH domain were diversified to match the positional amino acid frequency by CDR length found in the human antibody repertoire.
  • phage display heavy chain (SEQ ID NO:19) and light chain (SEQ ID NO:20) scaffolds used in the library are listed below, where a lower case“x” represents CDR amino acids that were varied to create the library, and bold italic represents the CDR sequences that were constant.
  • Phage panning was performed using standard procedures. Briefly, the first round of phage panning was performed with target immobilized on streptavidin magnetic beads which were subjected to ⁇ 5x10 12 phage from the prepared library in a volume of 1 mL in PBST-2% BSA. After a one-hour incubation, the bead-bound phage were separated from the supernatant using a magnetic stand. Beads were washed three times to remove non- specifically bound phage and were then added to ER2738 cells (5 mL) at OD 600 ⁇ 0.6.
  • infected cells were sub-cultured in 25 mL 2xYT + Ampicillin and M13K07 helper phage and allowed to grow overnight at 37 ⁇ C with vigorous shaking.
  • phage were prepared using standard procedures by PEG precipitation. Pre-clearance of phage specific to SAV-coated beads was performed prior to panning. The second round of panning was performed using the KingFisher magnetic bead handler with 100 nM bead-immobilized antigen using standard procedures. In total, 3-4 rounds of phage panning were performed to enrich in phage displaying Fabs specific for the target antigen. Target-specific enrichment was confirmed using polyclonal and monoclonal phage ELISA. DNA sequencing was used to determine isolated Fab clones containing a candidate ABS.
  • VL and VH domains were formatted into a bivalent monospecific native human full-length IgG1 architecture and immobilized to a biosensor on an Octet (Pall ForteBio) biolayer interferometer. Soluble antigens were then added to the system and binding measured.
  • VL variable regions of individual clones were formatted into Domain A and/or H, and VH region into Domain F and/or L of a bivalent 1x1 B-Body“BC1” scaffold shown below and with reference to FIG. 2B.
  • VL and VH domains were formatted only into Domain H and L, respectively, and the constructs each contained the same A:F antigen binding site with a known expression profile for an unrelated target.
  • the sequence of the common first polypeptide and common second polypeptide chain are provided, respectively, in SEQ ID NO:1 and SEQ ID NO:2.
  • the VL and VH domains were formatted into a bivalent monospecific native IgG architecture.
  • Domain A Antigen 1 B-Body Domain A/H Scaffold (SEQ ID NO:21)
  • Domain B CH3 (T366K; 445K, 446S, 447C tripeptide insertion)
  • Domain D CH2
  • Domain F Antigen 1 B-Body Domain F/L Scaffold (SEQ ID NO:22)
  • Domain G CH3 (L351D; 445G, 446E, 447C tripeptide insertion) 3 rd polypeptide chain (SEQ ID NO:25):
  • Domain H Antigen 2 B-Body Domain A/H Scaffold (SEQ ID NO:21)
  • Domain I CL (Kappa)
  • Domain L Anitgen 2 B-Body Domain F/L Scaffold (SEQ ID NO:22)
  • Domain M CH1.
  • the variable domains were formatted into the 2(A-A)x1(B) format described herein and in Section 6.4.4.1, where the 1st polypeptide scaffold chain is SEQ ID NO:27.
  • the other BC12x1 chains are identical to the BC1 chains, with the 5th chain identical to the 2nd chain.
  • candidates using the variable domains formatted into the 2(B-A)x1(B) format as described below e.g., the OX40:24-11x11 described herein and in Section 6.10.13, and see Section 6.4.4.3.
  • the B-Body protein was purified in 96-well format using CaptureSelect CH1 affinity resin (ThermoFisher) and average yield was ⁇ 50 mg B-Body/mL culture.
  • the bispecific 1x1 B-Body proteins each containing one OX40 antigen binding site– were evaluated for overall yield, protein purity, affinity for OX40, and cell binding.
  • the clones were tested for cross-competition and sorted into epitope bins. Affinity was determined using biolayer interferometry (BLI, Octet/FORTEBIO®).
  • the first discovery campaign identified 17 clones that can be expressed in the Expi 293 system, bind to human OX40 on the cell surface with monovalent affinity in the range of 0.1-100 nM, and do not exhibit non-specific binding.
  • VL and VH domains were formatted into a bivalent monospecific native IgG architecture. As shown in FIG.17 and FIG.18, clones were tested for cross-competition and sorted into epitope bins. Affinity was determined using biolayer interferometry (BLI, Octet/FORTEBIO®).
  • Table 3 lists the VH CDR1/2/3 sequences from the 40 selected clones.
  • Table 4 lists the VL CDR3 sequences from the 40 selected clones, and the constant CDR1 and CDR2 sequences used in the screen.
  • Each construct was expressed and purified. The purity was normally >85% as estimated by SDS PAGE. The concentration of purified antibodies was ⁇ 1mg/mL on average after one-step affinity purification using CH1 affinity resin and neutralization. The proteins were directly used for activation assay at 1 mg/mL after ⁇ 1000X dilution in DMEM media.
  • FIG.8 tabulates concentrations (in mg/mL) of the respective bivalent 1x1 B-Body constructs after one-step purification. The average concentration was 950 +/- 500 mg/mL.
  • Luminescent-based reporter cell lines were generated to assay the NFkb pathway activation by OX40.
  • a plasmid coding the full-length human OX40 under a CMV promotor with hygromycin resistance was transfected into NFkB/293/GFP-Luc (catalog number: TR860A-1, SystemBio) cells. Selection was performed with 200 mg/mL
  • Hygromycin B for three weeks.
  • the pool was detached, labeled with anti-human OX40- phycoerythrin antibody, and sorted for PE positive and GFP negative cells by FACS.
  • the ⁇ 106 collected cells were expanded for two weeks under DMEM+200 mg/mL Hygromycin B and sorted again for GFP negative.
  • the 2nd sorted pool was annotated as NFkb/293/GFP- Luc-OX40 to assay NFkb activation.
  • NFkB activation by OX40 agonist For high throughput screening, activation assays were prepared in half-area 96-well plates containing 5x10 4 NFkb/293/GFP-Luc-OX40 cells, 6 nM B-Body antibodies, with or without 20 nM Goat-anti-human (GAH) antibody. After a 6 hr incubation at 37 °C, an equal volume of One-step BPS Luminescence Kit mix was added and the luminescence was measured. The luminescence intensity is proportional to agonist activity through OX40. An activation assay with an antibody titration (0.01– 100 nM) was performed with candidates showing top potency from high throughput single point activation.
  • GH goat-anti-human
  • Example 2 Trivalent Anti-OX40 Agonist Candidates [00605] We also cloned the variable regions of the initial 17 OX40 agonist candidates we identified into antigen binding sites in the trivalent 2x1 B-Body format (see FIG.3) and trivalent 1x2 format (see FIG.4). By selectively pairing variable regions, we created and screened in the range of 100 OX40 bispecific trivalent 2x1 B-Body constructs.
  • the trivalent B-Body constructs were expressed at 1.5 mL scale in 96-well deep well blocks and purified with CH1 affinity resin.
  • FIG.11 shows agonist activity of three clinical OX40 agonists, in comparison to crosslinked natural ligand (OX40L-Fc + GAH).
  • FIG.11A shows the activity of the mAbs in the absence of the independent crosslinking agent, GAH.
  • FIG.11B shows the activity of the mAbs in the presence of the independent crosslinking agent, GAH.
  • the agonist activities of the three clinical mAbs demonstrated minimal activity by themselves, but were comparable to the natural ligand, OX40L, in presence of cross-linking.
  • the activation assay was performed in the range of 0.01– 30 nM mAb.
  • the efficacy of OX40 agonist largely depends on the amount of crosslinker. Therefore, experimental conditions below 10 nM of mAb, such as 6 NM, were considered reliable for interpretation of observed efficacies.
  • FIG.12 compares the three anti-OX40 clinical mAbs in the absence of GAH crosslinking (black dashed lines) to our top bivalent construct“10x9” (blue solid line), top trivalent construct“2x2x2” (red solid line), and crosslinked antigen (black solid line). Both of our constructs were seen to possess activity comparable to the crosslinked natural ligand, O1X40L, in the absence of an independent cross-linking agent, and demonstrated increased agonism as compared to the three known clinical anti-OX40 mAbs.
  • Example 4 Expanded High Throughput Agonist Discovery [00613] An expanded screen, performed essentially as described above, increased the number of identified candidate B-body OX40 agonists from 17 to 40. Briefly, B-Body candidate agonists were transiently expressed and purified using the one-step CH1 purification scheme. Candidates were added to HEK 293-NFkb-GFP/Luc-OX40 in soluble form without additional cross-linker or immobilization, and luminescence was read as agonistic activity from NFkB activation through OX40. The natural OX40 ligand Fc fusion protein (“OX40L- Fc”) was used to establish 100% agonism. Three clinical anti-OX40 monoclonal antibodies were also tested (arrows from left to right: Pogalizumab, Tavolixizumab, and GSK3174998)
  • FIG.14 shows agonist activity of three bivalent OX40 agonists in the absence and presence (+GAH) of the goat-anti-human (GAH) antibody crosslinking agent, as well as agonist activity of the control, crosslinked natural ligand-Fc fusion (OX40L-Fc), in the absence and presence of goat-anti-mouse (GAM) antibody crosslinking agent (OX40L-Fc + GAM).
  • Fig.15 shows dose response curves for a subset of bispecific OX40 agonists identified during the high throughput screen. Agonist activity was tested using NFkB activation to identify potent agonists. Candidates were expressed and purified by one-step CH1 affinity chromatography. Dose response experiments were performed using the HEK 293-NFkb-GFP/Luc-OX40 reporter assay in a range of 0.03 nM to 30 nM. Multiple bispecific OX40 agonists were identified that are more potent than cross-linked OX40 ligand- Fc fusion (OX40L-Fc). 6.10.8.
  • Example 7 Epitope Mapping of OX40 Antigen Binding Sequences
  • OX40:2 and OX40:8 Two candidates with non-overlapping OX40 antigen binding sites (OX40:2 and OX40:8) identified in the screen as well as OX40L-Fc and clinical OX40 antibodies were investigated further to determine the specific OX40 epitope bound by each.
  • FIG.16A Two candidates with non-overlapping OX40 antigen binding sites (OX40:2 and OX40:8) identified in the screen as well as OX40L-Fc and clinical OX40 antibodies were investigated further to determine the specific OX40 epitope bound by each.
  • FIG.16B shows binding for the different monospecific antibodies to different OX40 fragments having a series of truncations from the N-terminus (AA 2-214, AA 66-214, AA 108-214, and AA 127-214).
  • the OX40 fragments were prepared as Fc fusion proteins, and also contained a signal peptide, an Avi-tag, a TEV cleavage site, and a HIS tag for purification.
  • the full length or truncated OX40-Fc fusion proteins were immobilized onto BLI sensor and the different monospecific antibodies included the clinical antibodies GSK3174998, Pogalizumab, and Tavolixizumab, as well as monospecific bivalent BC1 formatted candidates with either the OX40 antigen binding site OX40:2 (“2x2”) or the OX40 antigen binding site OX40:8 (“8x8”).
  • OX40:2-214 indicating that OX40L only bound the first CRD (amino acids 2-66).
  • the OX40:2 antigen binding site and the clinical antibody GSK3174998 demonstrated binding to the full length fragment (OX40:2-214) and partial binding to the first truncation (OX40:66- 214), indicating that both bound the first and second CRD (amino acids 2-108).
  • the other two clinical antibodies, Pogalizumab and Tavolixizumab demonstrated the strongest binding to the fragment OX40:108-214, while binding was no longer present in the OX40:127-214 truncation, indicating that both bound the third CRD (amino acids 108-127).
  • the OX40:8 antigen binding site demonstrated binding to all tested truncations of OX40, indicating binding to the fourth CRD (amino acids 127-214).
  • our OX40 screen identified antigen binding sites that bind epitopes that did not overlap (OX40:2 binding an epitope within amino acids 2-108 and OX40:8 binding an epitope within amino acids 127-214), as well as an antigen binding site that binds an epitope different from that bound by the tested clinical monoclonal antibodies (OX40:8 binding an epitope within amino acids 127-214). 6.10.9.
  • Example 8 Measuring Binding of Non-overlapping Epitopes
  • Candidate OX40 antigen binding sites identified in the screen were tested in combination for simultaneous binding to OX40. As shown in FIG.17, 100 nM biotinylated OX40 was immobilized through streptavidin to a BLI sensor (“+OX40”). After baseline equilibration, 100 nM of a first candidate antigen binding site formatted in a native monospecific IgG antibody conformation (top panel OX40:8; middle panel OX40:21; bottom panel OX40:35) was added as indicated, with each demonstrating binding to OX40.
  • FIG.42 shows exemplary results from another simultaneous binding experiment. Briefly, 100 nM biotinylated OX40 was immobilized through streptavidin to a BLI sensor (“+OX40”). After baseline equilibration, 100 nM of a first candidate antigen binding site formatted in a native monospecific IgG antibody conformation (“mAB1”) was added, followed by addition of a second candidate antigen binding site in native IgG format (“mAB2”), followed by addition of a third candidate antigen binding site in native IgG format (“mAB3”). As shown in FIG.42, additional binding was demonstrated with the addition of each successive mAB, indicating non-overlapping epitopes for each of the candidates.
  • mAB1 monospecific IgG antibody conformation
  • mAB2 native IgG format
  • mAB3 native IgG format
  • Fig.18 summarizes the results of the binding experiments for the panel of the 40 antigen binding sites in all possible combinations (i.e., a 40x40 matrix).
  • the level of the BLI response is based on the mass of the antibodies bound.
  • the expected BLI response level was predicted for complete binding by both a first and second OX40 candidate agonist.
  • Molecules with a higher percentage of expected binding indicated simultaneous binding of OX40 by both candidates, suggestive of non-overlapping epitopes.
  • the top row identifies the first antibody added to the sensor, and the first column identifies the second antibody. Shaded squares identify binding site combinations that demonstrated simultaneous binding to non-overlapping epitopes. 6.10.10.
  • Example 9 T cell Activation by Non-crosslinked OX40 Agonist
  • Candidate OX40 agonists were screened in CD4+/CD45RA+/CD25 naive T cell assays. Soluble candidates were directly applied to the primary cell assay and a clinical mAB GSK3174998 was applied in both soluble and plate-coated forms as controls. The T cell proliferation was assayed by PrestoBlue and IL-2 secretion was quantified by ELISA. As shown in FIG.19, GSK3173998 only stimulated T cell proliferation when bound to a plate, while no proliferation was detected when soluble GSK3173998 was added (left panel).
  • bispecific trivalent B-body OX40:2-2x8 stimulated similar levels of T cell proliferation regardless of being soluble or plate bound, suggesting receptor clustering activity in the absence of a crosslinking agent. Measuring IL-2 secretion also demonstrated activity of soluble OX40:2-2x8 but not soluble GSK3173998.
  • FIG.20 shows a summary of T cell stimulatory activity for different multispecific multivalent candidate OX40 agonists.
  • the X-axis represents the IL2 secretion, while the Y- axis is the CD4+/CD45RA+/CD25- T cell proliferation for each candidate.
  • the shaded circle provides a cutoff for those agonists considered potent, with those lying outside of the circle considered potent. All of the agonists lying outside the circle were bispecific trivalent B- bodies in a 2x1 format and had the highest potency.
  • Candidates of interest identified in the screen are:
  • Chain 2 VH (OX40:24)- CH3 (BC1) [SEQ ID NO:48]
  • Chain 3 VL (OX40:11) - CL -CH2 -CH3 (Hole, 349C) [SEQ ID NO:49]
  • Chain 4 VH (OX40:11) - CH1 [SEQ ID NO:50]
  • Chain 5 equivalent to chain 2
  • Chain 2 VH (OX40:24)- CH3 (BC1) [SEQ ID NO:48]
  • Chain 3 VL (OX40:10) - CL -CH2 -CH3 (Hole, 349C) [SEQ ID NO:51]
  • Chain 4 VH (OX40:10) - CH1 [SEQ ID NO:52]
  • Chain 5 equivalent to chain 2 OX40: 24-24x6
  • Chain 2 VH (OX40:24)- CH3 (BC1) [SEQ ID NO:48]
  • Chain 3 VL (OX40:6) - CL -CH2 -CH3 (Hole, 349C) [SEQ ID NO:53]
  • Chain 4 VH (OX40:6) - CH1 [SEQ ID NO:54]
  • Chain 5 equivalent to chain 2
  • Chain 2 VH (OX40:24)- CH3 (BC1) [SEQ ID NO:48]
  • Chain 3 VL (OX40:4) - CL -CH2 -CH3 (Hole, 349C) [SEQ ID NO:55]
  • Chain 4 VH (OX40:4) - CH1 [SEQ ID NO:56]
  • FIG. 32 depicts an exemplary photomicrograph of T cell clusters formed by Day 5 of treatment with either the clinical mAB or bispecific B-Body OX40:2x8.
  • the photomicrograph shows that treatment with OX40:2x8 significantly increased T cell proliferation as compared to the clinical mAB.
  • the screens described above identified several multivalent multispecific agonists capable of OX40 receptor clustering activity in primary cells. 6.10.12.
  • Example 11 Two step Purification of OX40 Agonist Candidates [00627] Candidate OX40 agonists and clinical monoclonal antibodies were purified using a two-step purification process.
  • OX40:2-2x8, OX40:3-3x25, and OX40:33x25 were purified by CH1 and anion exchange chromatography, while the clinical antibodies were purified by Protein A and anion exchange chromatography.
  • FIG.22 shows a non-reducing SDS-PAGE analysis of the two-step purified antibodies demonstrating a high level of purity. 6.10.13.
  • Example 12 T cell Activation by OX40 Agonist Candidates [00628] Activation of T cells using OX40 agonist candidates in a soluble 2x1 format was monitored by cytokine secretion (see FIGs.23-27). The OX40:24-24x11 is described above, and OX40:24-11x11 is described below.
  • OX40:24-11x11 and OX40:24-24x11 both stimulated T cells greater than cross-linked GSK3174998 (“GSK+GAH”) as measured by TNFa and IL-2 secretion (FIG.23A and FIG.23B, respectively), and comparable activity to cross-linked GSK3174998 by IFNg secretion (FIG.23C).
  • GSK+GAH cross-linked GSK3174998
  • IFNg secretion FIG.23C
  • both OX40:24-11x11 and OX40:24- 24x11 demonstrated activity at the lowest antibody concentration tested suggesting the constructs were active in the soluble format, while soluble GSK3174998 did not result in detectable activation and plate-bound (“coated”) GSK3174998 demonstrated activity only at higher antibody concentrations.
  • OX40:24-11x11 and OX40:24-24x11 both demonstrated activity in the sub-nanomolar range as measured by TNF, IL-2, and IFNg secretion (FIG. 24A-C, respectively).
  • OX40:24- 24(WEE)x11 also shown in FIG.24 is a modified OX40:24-24x11 construct, termed OX40:24- 24(WEE)x11, with two aspartic acids in each OX40 antigen binding site modified to glutamic acid residues to remove a potential proteolytic cleavage site in Chain 1 (see SEQ ID NO:59, all other chains equivalent).
  • OX40:24-24(WEE)x11 demonstrated agonist activity greater than the crosslinked GSK3174998 control, though slightly less than the unmodified version in this assay.
  • OX40:24-11x11 A modified version of OX40:24-11x11, termed“OX40:24(WEE)-11x11,” was also constructed (see SEQ ID NO:58 for Chain 1, all other chains equivalent).
  • Example 13 T cell Proliferation by OX40 Agonist Candidates [00634] Activation of T cells using OX40 agonist candidates in a soluble 2x1 format was also monitored by proliferation.
  • candidates OX40:24-24x11 (FIG.28A), OX40:24-24(WEE)x11 (FIG.28B), OX40:24-11x11 (FIG.28C), and OX40:24(WEE)- 11x11 (FIG.28D) all proliferated with similar kinetics.
  • proliferation for all the constructs was attenuated at higher antibody concentrations.
  • concentration of antibody that resulted in attenuated proliferation varied between constructs, with peak proliferation having resulted at 10nM for OX40:24-24x11 based constructs at concentrations below 10nM for OX40:24-11x11 based constructs.
  • Monospecific bivalent 1x1 candidates OX40:24 and OX40:11 formatted in a native IgG architecture were compared alone and in combination against a bispecific bivalent 1x1 candidate (OX40:11x24) and various bispecific trivalent 2x1 candidates.
  • native IgG formats of OX40:11 and OX40:24 candidates demonstrated minimal activity comparable to soluble anti-CD3 and soluble GSK3174998 as measured by TNFa and IL-2 secretion (FIG.30A and FIG.30B, respectively).
  • OX40:11 + OX40:24 A combination of both OX40:11 and OX40:24 candidates (“OX40:11 + OX40:24”) resulted in detectable activity, though significantly below the bispecific bivalent 1x1 candidate and bispecific trivalent 2x1 candidates.
  • bispecific formats both bivalent and trivalent resulted in significant agonist activity, while monospecific formats of combinations of monospecific native IgG formats did not.
  • the bispecific trivalent 2x1 candidates demonstrated increased agonist activity compared to the bispecific bivalent 1x1 candidate as measured by TNFa secretion (FIG.30A). 6.10.16.
  • Example 15 T cell Activation by Cross-Linking
  • Various B-body constructs using ABS OX40:24 and OX40:11 were tested for agonist activity in combination with cross-linking.
  • the clinical antibody GSK3174998 required addition of a soluble cross-linker (“GSK+GAH”) to demonstrate activity as compared to GSK3174998 in the absence of cross-linking (“GSK”) as measured by TNFa secretion.
  • bispecific trivalent OX40:24-24x11 and OX40:24-11x11 candidates demonstrated the greatest activity as measured by luciferase, with OX40:24-11x11 detectably above OX40:24- 24x11.
  • Bispecific bivalent OX40:24x11 and OX40:11x24 candidates also demonstrated activity, with both above the cross-linked clinical antibody GSK3174998.
  • a range of agonist activity was seen across B-body formats, with bispecific trivalent candidates having increased potency in comparison to bispecific bivalent candidates. 6.10.18.
  • Example 17 Biophysical Properties of OX40 Agonist Candidates
  • Bispecific trivalent OX40 agonist candidates OX40:24-24x11 and OX40:24- 11x11 were assessed for biophysical properties, such as those relevant for production of antibodies in a clinical setting or on an industrial scale.
  • the candidates OX40:24-24x11 and OX40:24-11x11 demonstrated biophysical properties useful in manufacturing for clinical and industrial settings. Properties were assessed using standard assays. Examples of biophysical properties and methods to assess the same are described in more detail in Jain et al. (Proc Natl Acad Sci U S A.2017 Jan 31;114(5):944-949.), herein incorporated by reference for all it teaches.
  • Properties assessed were yield, purity, homogeneity, stability, long-term stability, acid stability, thermostability, low antibody cross- interaction, low antibody self-interaction, low hydrophobic binding, and cyno crossreactivity.
  • Example 18 Further characterization of OX40:24 and OX40:11 candidates.
  • the heavy chain CDRs and light chain CDRs of ABS candidates 24, 24WEE, and 11 (also referred to as OX40:24, Ox40:24WEE, and OX40:11, respectively), included in Tables 6 and 7, are shown below.
  • VH and VL sequences for the OX40:24, OX40:24WEE, and OX40:11 candidates are provided as SEQ ID NOs: 227-231.
  • VH and VL regions of clones 24 and 11 from the discovery campaign were formatted into IgG. Affinity was determined using biolayer interferometry (BLI, Octet/FORTEBIO®) as described herein. Results are shown in FIG.33.
  • the Kd for OX40- 11, corresponding to ABS 11 as shown in Table 6 and Table 7, was 18 nM.
  • Bivalent affinities of OX40 candidate binding molecules were determined as follows. HEK293 cells stably expressing OX40 were tested for cell binding with a dilution series of candidate molecules. The binding molecules tested were bivalent, monospecific OX40:24 IgG (“OX24”), bivalent, monospecific OX40:11 IgG (“OX11”), bivalent, bispecific
  • Cells were incubated with the candidate molecules, followed by an AlexaFluor488 labeled goat anti-human Fab, and the mean fluorescence intensity (MFI) determined by flow cytometry.
  • MFI mean fluorescence intensity
  • OX40:11 and OX40:24 candidates were also determined using biolayer interferometry. Briefly, OX40 was immobilized on the sensor, followed by either the OX40:11 or OX40:24 candidate, then followed by OX40L. Results are shown in FIG.45, indicating that each of the OX40:11 and OX40:24 antigen binding sites, while binding to non-overlapping epitopes, both bind to a region of OX40 that interacts with OX40L. 6.10.20.
  • Example 19 Suppression of induced Treg immunosuppressive signals by OX40 agonist candidates
  • Tumor resident regulatory T-cells which represent a large subset of T-cells within the tumor microenvironment, promote immunosuppressive cytokines such as IL-10, thus downregulating effector T-cell function and promoting tumor growth.
  • immunosuppressive cytokines such as IL-10
  • Both iTregs and activated na ⁇ ve CD4+ T cells exhibit surface OX40 expression.
  • OX40 axis stimulation in iTregs downregulates Foxp3 phenotypes in Tregs, thereby reducing IL-10 secretion.
  • OX40 axis stimulation in activated na ⁇ ve CD4+ T-cells causes release of cytokines, e.g., IFN-g, TNF-a, and IL-2.
  • cytokines e.g., IFN-g, TNF-a, and IL-2.
  • Such cytokines promote suppression of IL-10 secretion by Tregs.
  • CD3+CD4+CD45RA+CD45RO- were purified from human PBMCs from a healthy donor using the EasySep Human Na ⁇ ve CD4+ T Cell Isolation Kit (Stemcell Technologies, Vancouver, Canada) and were activated with crosslinked anti-CD3 (1ug/ml) for 48 hr at 37°C in the ImmunoCult-XF T Cell Expansion Medium (Stemcell Technologies, Vancouver, Canada).
  • naive CD4 T-cells were purified from human PBMCs obtained from the same healthy donor and were in vitro differentiated into CD3+CD4+CD25+Foxp3+ induced regulatory T-cells (iTregs), using the ImmunoCult Human Treg Differentiation Supplement (Stemcell Technologies, Vancouver, Canada). These iTregs were then added at 3:1 ratio to the wells with or without activated na ⁇ ve CD4 T-cells and incubated for 4 days at 37°C in the ImmunoCult-XF T Cell Expansion Medium. Test antibodies (OX40:24(WEE)- 11X11, GSK3174998) at 10nM concentration were then added to the cultured cells. At day 3 supernatants were harvested and IL-10 secretion was measured by BD OptEIA - Human IL- 10 ELISA Set (BD Biosciences, San Jose, CA).
  • Results are shown in FIG.35.
  • OX40:24-11X11 outperformed clinical candidate GSK3174998 in reducing secretion of the immunosuppressive cytokine IL-10 in iTregs alone and in iTregs co-cultured with activated CD4+ T-cells. These results show that soluble OX40 agonist antibodies downregulate the immunosuppressive phenotype of iTregs.
  • TAMs Tumor associated macrophages
  • M2 macrophages have been shown to provide a favorable microenvironment for tumor growth, tumor survival, and angiogenesis.
  • monocytes CD45+CD14+
  • SF macrophage medium Stemcell Technologies, Vancouver, Canada
  • CD3+CD4+CD45RA+CD45RO- were purified from human PBMCs from the same healthy donor using the EasySep Human Na ⁇ ve CD4+ T Cell Isolation Kit (Stemcell Technologies, Vancouver, Canada) and were activated with crosslinked anti-CD3 (1ug/ml) for 48 hr at 37°C in the ImmunoCult-XF T Cell Expansion Medium (Stemcell Technologies, Vancouver, Canada). These activated CD4+ T-cells were then added to the M2a macrophage culture at the ratio of 1:3, along with 10nM of test antibodies (OX40:24(WEE)-11X11, GSK3174998), followed by incubation at 37°C. At day 3 supernatants were harvested and IL-10 secretion was measured by BD OptEIA - Human IL-10 ELISA Set (BD Biosciences, San Jose, CA).
  • Results are shown in FIG.36.
  • the clinical Mab candidate GSK had no significant effect on IL-10 secretion, as compared to untreated M2a cells.
  • both OX40:24- 11X11 and OX40:24-11X11 sF.c. outperformed GSK by robustly decreasing IL-10 secretion in M2A cells.
  • OX40 stimulation in activated CD4+ T cells reduces the immunosuppressive phenotype of adjacent M2a macrophages. 6.10.21.
  • Example 20 OX40 clinical candidates OX40:24(WEE)-11X11 and OX40:24(WEE)-11X11 s.Fc are cross-reactive to rhesus OX40.
  • the rhesus OX40 protein is 93% identical to its human counterpart, according to the NCBI database.
  • T-cell proliferation was analyzed by tracking the formation of T-cell clusters by the Incucyte Live Cell Imaging system (Sartorius AG, Göttingen, Germany). The formation of T-cell clusters were plotted as percent confluent clusters per well.
  • cell culture supernatants were harvested and TNF- ⁇ secretion was analyzed using rhesus validated BD OptEIA - Human TNF- ⁇ ELISA Set (BD Biosciences, San Jose, CA).
  • healthy donor human PBMC were incubated with cross linked anti-CD3 (1mg/ml) and pre-determined optimal concentration (50pM) of CPG class C ODN-2395 (Invivogen, San Diego, CA), with or without pre-determined optimal concentration (10 nM) of soluble OX40 agonist antibody (OX40:24(WEE)-11X11 (“INV241111”)).
  • T-cell proliferation was analyzed by tracking the formation of T-cell clusters (yellow-masked) by the Incucyte Live Cell Imaging system (Sartorius AG,
  • T-cell clusters were plotted as percent confluent clusters per well.
  • cell culture supernatants were harvested and TNF- ⁇ secretion was analyzed using BD OptEIA - Human TNF- ⁇ ELISA Set (BD Biosciences, San Jose, CA).
  • BD Biosciences San Jose, CA
  • a CPG-C dose response curve (16.6 nM, 5 nM, 1 nM, 500 pM,167 pM, 50 pM, 17 pM, 5 pM, 2 pM was tested in the presence or absence of the INV241111 candidate.
  • Photomicrographs of T cell clusters on Days 0 and 4 are shown in FIG.38. No clusters were apparent on Day 0 for either the CPG alone, INV24111 alone, or
  • INV241111 treatment group exhibited strikingly larger T cell clusters as compared to the CPG alone and INV241111 alone conditions.
  • T cell confluence results are shown in FIG.39. Without anti-CD3 treatment, T cell proliferation is minimal. Anti-CD3 treatment alone induced modest T cell proliferation. In the presence of anti-CD3, INV241111 treatment outperformed CPG in inducing T cell proliferation, and CPG + INV241111 in combination enhanced T cell proliferation to an even greater degree.
  • TNF-a secretion results are shown in FIG.40.
  • CPG + INV241111 vastly outperformed CPG alone in inducing TNF- a secretion.
  • OX40 mAbs require cross-linking to generate observable agonistic activities, as we demonstrated in our cellular assay with Tavolixizumab, Pagolizumab and GSK3174998.
  • the clinical trials of these known clinical-stage antibodies therefore rely on the Fc receptor engagement to effect agonist activity, which may contribute to the low response rate that has been observed so far.
  • the only clinical trial with significant efficacy (12/30 with tumor shrinkage) was conducted with a mouse anti-human OX40 antibody, and all responders were demonstrated to have significant amount of mouse-anti-human-antibody (MAHA), which likely act as cross-linkers, increasing in vivo efficacy of the OX40 antibody.
  • MAHA mouse-anti-human-antibody
  • the B-Body platform described in detail in U.S. Patent Application No.15/787,640, filed October 18, 2017, incorporated herein by reference, provides superior orthogonality, dramatically decreased incomplete pairing, and increased yield to ⁇ 100 mg antibody construct/mL cell culture in various valency formats. Together with standardized cloning protocols, high throughput protein expression, and single-step purification, the B-Body platform allowed us to perform high throughput cellular assay screening for multivalent agonist antibodies. [00662] Using this system, we successfully demonstrated that multivalent antibodies can be potent OX40 agonists by themselves, in the absence of an independent crosslinking agent, such as cellular FcgR or MAHA.
  • an independent crosslinking agent such as cellular FcgR or MAHA.
  • FIG.46 provides a framework for the clustering potential of antibody-based agonists.
  • Monospecific bivalent antibodies have limited clustering capability. Each arm of such antibody binds to the same epitope, thus limiting the clustering ability of the antibody to at most two OX40 molecules. See FIG.46, “Monospecific Antibody, single epitope.”
  • Biparatopic antibodies have two non-overlapping epitopes. In some cases, the location of the non-overlapping epitopes allows each paratope of the antibody to simultaneously bind to the same OX40 molecule, thus reducing clustering capability.
  • FIG.46 “Biparatopic antibody, two non-overlapping epitopes.”
  • the spacing of the epitopes on a single OX40 molecule constrains the ability of the paratopes of a single antibody to bind to a single OX40 molecule.
  • the biparatopic antibody binds to two different OX40 molecules at two different epitopes, thus enabling high density clustering of the OX40 molecules.
  • Phage display of human Fab libraries was carried out using standard protocols. Biotinylated-Fc Fusion of the extracellular domain of human BCMA was purchased from Acro Biosystems. Three rounds of phage panning were conducted including human IgG used at a final concentration of 0.3 mg/ml as a competitor to deselect for binding to the Fc fusion. Phage clones were screened for the ability to bind BCMA by phage ELISA using standard protocols.
  • the library included different antigen binding sites generated by pairing a select VH sequence containing a CDR1, a CDR2, and a CDR3 grafted into a heavy-chain framework with a select VL sequence containing a CDR3 grafted into a light-chain framework (CDR1 and CDR2 for VL remained constant in the panel tested). DNA sequencing was used to determine unique clones.
  • Table 9 depicts sequences of top clones identified by phage display.
  • the LCDR1 amino acid sequence was RASQSVSSAVA and the LCDR2 amino acid sequence was SASSLYS.
  • VL and VH domains were formatted into a bivalent monospecific native IgG1 architecture.
  • VL variable regions of individual clones were formatted into Domain A and/or H, and VH region into Domain F and/or L of a bivalent 1x1 B-Body“BC1” scaffold with reference to FIG.2B.
  • FIG.47 demonstrates binding affinity of the top two clones to BCMA as determined by Octet (Pall ForteBio) biolayer interferometry analysis.
  • CD3 binding arm variants based on a humanized version of the SP34 anti-CD3 antibody (SP34-89, SEQ ID NOs: 232 (VH) and 233 (VL)) (SP34-88, SED ID NOs 234 (VH) and 235 (VL)) were created.
  • BCMA binding sites BCMA-1 or BCMA-2
  • CD3 binding sites SP34-89 or SP34-88, described above
  • RPMI-8226, KMS-12-BM, or REH cells (35,000 cells per well), NFkB-GFP Jurkat cells (75,000 cells per well), anti-CD-28 antibody (1 ⁇ g/mL), and the indicated dilution series of a CD3xBCMA B-bodyTM were added to the wells of a black-walled, clear-bottom 96 well plate in assay media (RPMI + 10% FBS). The plate was incubated at 37oC and 5% CO2 for 6 hours. A background suppression dye (Solution C, Thermo Fisher) was added to each well. The fluorescence signal was determined with a plate reader. Results are depicted in FIGS. 48A, 48B, and 48C.
  • BCMA-1 was found to be more potent in activation of the NFkB pathway than BCMA-2 and the activity did not change when in the 1x1 or 2x1 format.
  • the 2x1 BCMA-1xCD3 B-BodyTM has been tested in a competition assay with a BCMA ligand– APRIL Biolayer interferometry analysis.
  • bispecific antibody was allowed to bind to biotinylated immobilized BCMA to saturation.
  • binding of APRIL as a competitor was analyzed. Results are depicted in FIG.50.
  • APRIL and B- BodyTM share the same epitope, as they compete for binding to BCMA.
  • the inset shows APRIL binding to BCMA in the absence of B-BodyTM. 6.10.25.
  • Example 23 Co-administration of redirecting bispecific with
  • OX40 clustering binding molecules enhances activity of the redirecting bispecific [00681] It was next tested if costimulatory molecules could improve the total activity of the BCMAxCD3 in the Jurkat co-culture assay.
  • RPMI-8226 cells were plated along with the OX40-NF-kB-Luc Jurkat line, plus 1 mg/mL anti-CD28 antibody in RPMI containing 10% FBS.
  • the BCMAxCD3 T-Cell redirecting antibody was added to this co-culture at EC 50 concentration of 586 pM.
  • the OX40 agonist antibodies (INV241111 or GSK) were then added at a final concentration of 10nM, 2.5nM, 0.625nM, 0.156nM, 39pM, 9.8pM, 2.4pM or 0.6pM. Cells were incubated with these antibodies for 6 hours and then luciferase reagent was added. In another instance, similar experiment was performed with 1nM or 10 nM OX40 agonist antibodies in combination with BCMAxCD3 T-cell redirecting antibody at a final concentration of 20 nM, 3.33 nM, 0.56 nM, 93 pM, 15 pM, 2.5 pM, 0.43 pM or 0.07 pM. Luciferase activity was then read on a PheraStar plate reader. Data is presented as total Relative Light Units (RLU).
  • RLU Relative Light Units
  • INV241111 exhibited an EC50 of 159 pM, as compared to GSK which exhibited an EC50 of 352.
  • OX40 agonist antibodies were kept constant at 1nM or 10nM and BCMAxCD3 T-cell redirecting antibody was tested at a final concentration of 20 nM, 3.33 nM, 0.56 nM, 93 pM, 15 pM, 2.5 pM, 0.43 pM or 0.07 pM
  • the EC 50 of BCMAxCD3 decreased from 586 pM to an EC 50 of 256 pM or 214 pM, respectively. 6.10.26.
  • Example 24 Testing combinations of clustering antibodies and redirecting bispecific antibodies [00683]
  • a tumor antigen-expressing target cell line is co-cultured with either: effector primary T-cells, NK cells, monocytes, macrophages, or engineered NF-kB-Luc reporter cells expressing either CD3 or a type of Fc-gamma receptor and the antigenic target of a clustering antibody described herein.
  • a clustering antibody and a bispecific antibody is then added to this co-culture and the resulting cytotoxicity in case of T-cells, NK cells, is measured by an LDH assay.
  • luciferase activity is measured in case of engineered reporter cells (e.g., engineered NF-kB-Luc reporter cells).
  • DKTHTCPPCP >Phage display heavy chain (SEQ ID NO:19):
  • TEV cleavage site ENLYFQ (SEQ ID NO:31)

Landscapes

  • Health & Medical Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Immunology (AREA)
  • Organic Chemistry (AREA)
  • Medicinal Chemistry (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Genetics & Genomics (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • Biochemistry (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • General Chemical & Material Sciences (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • Pharmacology & Pharmacy (AREA)
  • Animal Behavior & Ethology (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Peptides Or Proteins (AREA)
  • Micro-Organisms Or Cultivation Processes Thereof (AREA)

Abstract

L'invention concerne des constructions d'anticorps agonistes à agrégation de récepteurs multivalentes capables de (i) se lier à un récepteur cible de la surface cellulaire qui nécessite une agrégation pour une activité agoniste, et (ii) d'agréger le récepteur cible sur la surface cellulaire en l'absence d'un agent de réticulation indépendant. À chacun des sites de liaison à l'antigène se liant au récepteur cible de la construction participent des domaines de liaison à des régions variables d'anticorps. L'invention concerne également des compositions pharmaceutiques comprenant la construction d'anticorps, ainsi que des procédés de traitement de maladies, notamment du cancer, par administration de quantités thérapeutiquement efficaces de la composition pharmaceutique.
PCT/US2019/061660 2018-11-15 2019-11-15 Constructions d'anticorps agonistes à agrégation de récepteurs multivalentes et protéines de liaison à l'antigène WO2020102647A1 (fr)

Applications Claiming Priority (6)

Application Number Priority Date Filing Date Title
US201862768009P 2018-11-15 2018-11-15
US62/768,009 2018-11-15
US201962811929P 2019-02-28 2019-02-28
US201962812165P 2019-02-28 2019-02-28
US62/812,165 2019-02-28
US62/811,929 2019-02-28

Publications (1)

Publication Number Publication Date
WO2020102647A1 true WO2020102647A1 (fr) 2020-05-22

Family

ID=70731931

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2019/061660 WO2020102647A1 (fr) 2018-11-15 2019-11-15 Constructions d'anticorps agonistes à agrégation de récepteurs multivalentes et protéines de liaison à l'antigène

Country Status (1)

Country Link
WO (1) WO2020102647A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023039243A3 (fr) * 2021-09-13 2023-09-28 Achelois Biopharma, Inc. Compositions antivirus de l'hépatite b (anti-vhb) et leurs méthodes d'utilisation

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040123343A1 (en) * 2000-04-19 2004-06-24 La Rosa Thomas J. Rice nucleic acid molecules and other molecules associated with plants and uses thereof for plant improvement
US20070218069A1 (en) * 2005-12-15 2007-09-20 Gordon Nathaniel C Methods and compositions for targeting polyubiquitin
WO2010006059A1 (fr) * 2008-07-08 2010-01-14 Abbott Laboratories Protéines de liaison à la prostaglandine e2 et leurs utilisations
WO2017096179A1 (fr) * 2015-12-02 2017-06-08 Agenus Inc. Anticorps et leurs méthodes d'utilisation
US20170198051A1 (en) * 2016-01-11 2017-07-13 Inhibrx Lp Multivalent and multispecific ox40-binding fusion proteins

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040123343A1 (en) * 2000-04-19 2004-06-24 La Rosa Thomas J. Rice nucleic acid molecules and other molecules associated with plants and uses thereof for plant improvement
US20070218069A1 (en) * 2005-12-15 2007-09-20 Gordon Nathaniel C Methods and compositions for targeting polyubiquitin
WO2010006059A1 (fr) * 2008-07-08 2010-01-14 Abbott Laboratories Protéines de liaison à la prostaglandine e2 et leurs utilisations
WO2017096179A1 (fr) * 2015-12-02 2017-06-08 Agenus Inc. Anticorps et leurs méthodes d'utilisation
US20170198051A1 (en) * 2016-01-11 2017-07-13 Inhibrx Lp Multivalent and multispecific ox40-binding fusion proteins

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023039243A3 (fr) * 2021-09-13 2023-09-28 Achelois Biopharma, Inc. Compositions antivirus de l'hépatite b (anti-vhb) et leurs méthodes d'utilisation

Similar Documents

Publication Publication Date Title
TWI718206B (zh) Pd-l1抗體、其抗原結合片段及其醫藥用途
CN109651507B (zh) 一种激动型4-1bb单克隆抗体
KR102256152B1 (ko) Ｐｄ-1 항체, 이의 항원 결합 단편, 및 이의 의약 용도
JP7257971B2 (ja) 抗ｃｄ４０抗体、その抗原結合フラグメント、およびその医学的使用
JP2021524278A (ja) 抗メソテリン抗体
CN112969714B (zh) 抗cd40抗体、其抗原结合片段及其医药用途
KR20200143470A (ko) 항-ror 항체 구조체
JP2021508449A (ja) Ｃｈ３ドメイン中に挿入された特異的ｐｄ−ｌ１結合配列
CA3229113A1 (fr) Anticorps bispecifiques anti-b7-h4/anti-4-1bb et leurs utilisations
CA3094005A1 (fr) Anticorps anti-cd27, fragment de liaison a l'antigene de celui-ci et utilisation medicale associee
TWI685504B (zh) 抗gitr抗體、其抗原結合片段及其醫藥用途
JP2021521781A (ja) 三価三重特異性抗体構築物
US20210363265A1 (en) Multivalent receptor-clustering agonist antibody constructs
WO2020102647A1 (fr) Constructions d'anticorps agonistes à agrégation de récepteurs multivalentes et protéines de liaison à l'antigène
KR20230042090A (ko) 항-cd22 단일 도메인 항체 및 치료학적 작제물
CN108473586B (zh) 抗cd27抗体、其抗原结合片段及其医药用途
JP2021534204A (ja) 抗ＳＩＲＰｇ化合物
US20240190983A1 (en) Novel TNFR2 Binding Molecules
RU2779128C2 (ru) Антитело к cd40, его антигенсвязывающий фрагмент и его медицинское применение
TW202315889A (zh) 抗體
CN116554324A (zh) 一种特异性识别4-1bb的抗体、其制备方法及其用途

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19884383

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19884383

Country of ref document: EP

Kind code of ref document: A1