Multispecific antibodies and uses thereof

Abstract

The present disclosure relates to multispecific antibodies (e.g., bispecific antibodies or trispecific antibodies) or antigen binding fragments thereof. In one aspect, the multispecific antibody, or antigen-binding fragment thereof, may bind to a T cell antigen (e.g., CD3) and / or one or two tumor-associated antigens (e.g., CEA-CAM6, EGFR, HER2), or a combination thereof.

Description

Technical Field

[0001] The present disclosure relates to multispecific antibodies or antigen-binding fragments thereof. Background Art

[0002] Naturally occurring antibodies usually target only one antigen. Multispecific antibodies can be manufactured in different structural forms so that they can bind two or more different epitopes simultaneously. These epitopes can be in the same antigen or in different antigens. This opens up a wide range of applications, including redirecting T cells to tumor cells, blocking two different signal transduction pathways simultaneously, dual targeting different disease mediators, and delivering payloads to target sites.

[0003] Multispecific antibodies have a variety of applications. However, in some cases, multispecific antibodies may not have the desired efficacy and may be difficult to express and purify. There is a need to continue to develop various multispecific antibody-based therapies. Summary of the invention

[0004] The present disclosure relates to multispecific antibodies or antigen-binding fragments thereof, wherein the multispecific antibodies or antigen-binding fragments thereof bind to a T cell antigen (eg, CD3) and / or one or more tumor-associated antigens or a combination thereof.

[0005] In one aspect, the present disclosure relates to an antigen binding protein comprising

[0006] (a) Fc;

[0007] (b) a Fab fragment (Fab) that specifically binds to a T cell antigen; and

[0008] (c) a first single domain antibody variable domain (VHH) that specifically binds to a first tumor-associated antigen;

[0009] (d) a second single domain antibody variable domain (VHH) that specifically binds to a second tumor-associated antigen.

[0010] In some embodiments, the Fab, the first VHH, and the second VHH are linked to the Fc.

[0011] In some embodiments, the Fab comprises or consists of a light chain variable domain (VL), a light chain constant domain (CL), a heavy chain variable domain (VH) and a heavy chain first constant domain (CH1).

[0012] In some embodiments, the Fab can activate T cells upon binding to a T cell antigen.

[0013] In some embodiments, the T cell antigen is cluster of differentiation 3 (CD3).

[0014] In some embodiments, the first tumor-associated antigen and the second tumor-associated antigen are independently selected from the group consisting of cluster of differentiation 20 (CD20), carcinoembryonic antigen (CEA), prostate specific antigen (PSA), prostate stem cell antigen (PSCA), programmed death ligand 1 (PD-L1), human epidermal growth factor receptor 2 (HER2), human epidermal growth factor receptor 3 (Her3), human epidermal growth factor receptor (Her1), β-catenin, cluster of differentiation 19 (CD19), epidermal growth factor receptor (EGFR), tyrosine protein kinase Met (c-Met), epithelial cell adhesion molecule (EPCAM), prostate specific membrane antigen (PSMA), cluster of differentiation 40 (CD40), cell surface associated mucin 1 (Mucin 1, Cell Surface Associated) (MUC1), insulin-like growth factor 1 receptor (IGF1R) and carcinoembryonic antigen cell adhesion molecule 6 (CEA-CAM6).

[0015] In some embodiments, the Fc is human IgG4 Fc.

[0016] In some embodiments, the CH1 domain of the Fab is linked to the CH2 domain of the Fc, optionally via a hinge region.

[0017] In some embodiments, the hinge region is a human IgG4 hinge region, optionally with an S228P mutation according to EU numbering.

[0018] In some embodiments, the first VHH is linked to the CH2 domain in the Fc, optionally via a hinge region.

[0019] In some embodiments, the first VHH is linked to the CH3 domain in the Fc, optionally via a linker sequence.

[0020] In some embodiments, the second VHH is linked to the CH3 domain in the Fc, optionally via a linker sequence.

[0021] In some embodiments, the second VHH is linked to the Fab.

[0022] In some embodiments, the second VHH is linked to the C-terminus of the CL domain in the Fab.

[0023] In some embodiments, the second VHH is linked to the N-terminus of the VH domain in the Fab.

[0024] In some embodiments, the second VHH is linked to the N-terminus of the VL domain in the Fab.

[0025] In one aspect, the present disclosure relates to a protein complex comprising:

[0026] (a) a first polypeptide comprising, in the N-terminal to C-terminal direction: a VH, a CH1 domain, optionally a first hinge region, a first CH2 domain and a first CH3 domain;

[0027] (b) a second polypeptide comprising, in the N-terminal to C-terminal direction: a first VHH, optionally a second hinge region, a second CH2 domain, and a second CH3 domain;

[0028] (c) a third polypeptide comprising, in the N-terminal to C-terminal direction: a VL and a CL domain.

[0029] In some embodiments, the VH and the VL bind to each other to form an antigen binding site of the Fab that specifically binds to a T cell antigen. In some embodiments, the first VHH specifically binds to a first tumor associated antigen.

[0030] In some embodiments, the third polypeptide comprises, in the N-terminal to C-terminal direction: a VL, a CL domain, and a second VHH. In some embodiments, the second VHH specifically binds to a second tumor-associated antigen.

[0031] In some embodiments, the third polypeptide comprises, in the N-terminal to C-terminal direction: a second VHH, a VL, and a CL domain. In some embodiments, the second VHH specifically binds to a second tumor-associated antigen.

[0032] In some embodiments, the first polypeptide comprises, in the N-terminal to C-terminal direction: a second VHH, a VH, a CH1 domain, an optional first hinge region, a first CH2 domain, and a first CH3 domain. In some embodiments, the second VHH specifically binds to a second tumor-associated antigen.

[0033] In some embodiments, the second polypeptide comprises, in the N-terminal to C-terminal direction: a first VHH, a second VHH, an optional second hinge region, a second CH2 domain, and a second CH3 domain. In some embodiments, the second VHH specifically binds to a second tumor-associated antigen.

[0034] In some embodiments, the second polypeptide comprises, in the N-terminal to C-terminal direction: a first VHH, an optional second hinge region, a second CH2 domain, a second CH3 domain, and a second VHH. In some embodiments, the second VHH specifically binds to a second tumor-associated antigen.

[0035] In some embodiments, the first polypeptide comprises, in the N-terminal to C-terminal direction: a first VHH, a second VHH, an optional second hinge region, a second CH2 domain, a second CH3 domain, and a second VHH. In some embodiments, the second VHH specifically binds to a second tumor-associated antigen.

[0036] In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 1; the second polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 2; and the third polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 3.

[0037] In some embodiments, the T cell antigen is CD3, and the first tumor-associated antigen is selected from the group consisting of: CEA, CEA-CAM6, HER2, and EGFR.

[0038] In some embodiments, the T cell antigen is CD3, and the second tumor-associated antigen is selected from the group consisting of: CEA, CEA-CAM6, HER2, and EGFR.

[0039] In one aspect, the present disclosure relates to a protein complex comprising:

[0040] (a) a first polypeptide comprising, in the N-terminal to C-terminal direction: a first VHH, optionally a first hinge region, a first CH2 domain and a first CH3 domain;

[0041] (b) a second polypeptide comprising, in the N-terminal to C-terminal direction: a VH, a CH1 domain, optionally a second hinge region, a second CH2 domain, and a second CH3 domain;

[0042] (c) a third polypeptide comprising, in the N-terminal to C-terminal direction: a VL and a CL domain.

[0043] In some embodiments, the VH and the VL bind to each other to form an antigen binding site of the Fab that specifically binds to a T cell antigen. In some embodiments, the first VHH specifically binds to a first tumor associated antigen.

[0044] In some embodiments, the third polypeptide comprises, in the N-terminal to C-terminal direction: a VL, a CL domain, and a second VHH. In some embodiments, the second VHH specifically binds to a second tumor-associated antigen.

[0045] In some embodiments, the third polypeptide comprises, in the N-terminal to C-terminal direction: a second VHH, a VL, and a CL domain. In some embodiments, the second VHH specifically binds to a second tumor-associated antigen.

[0046] In some embodiments, the second polypeptide comprises, in the N-terminal to C-terminal direction: a second VHH, a VH, a CH1 domain, an optional first hinge region, a first CH2 domain, and a first CH3 domain. In some embodiments, the second VHH specifically binds to a second tumor-associated antigen.

[0047] In some embodiments, the first polypeptide comprises, in the N-terminal to C-terminal direction: a first VHH, a second VHH, an optional second hinge region, a second CH2 domain, and a second CH3 domain. In some embodiments, the second VHH specifically binds to a second tumor-associated antigen.

[0048] In some embodiments, the second polypeptide comprises, in the N-terminal to C-terminal direction: a VH, a CH1 domain, an optional first hinge region, a first CH2 domain, a first CH3 domain, and a second VHH. In some embodiments, the second VHH specifically binds to a second tumor-associated antigen.

[0049] In some embodiments, the first polypeptide comprises, in the N-terminal to C-terminal direction: a first VHH, an optional second hinge region, a second CH2 domain, a second CH3 domain, and a second VHH. In some embodiments, the second VHH specifically binds to a second tumor-associated antigen.

[0050] In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 1; the second polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 2; and the third polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 3.

[0051] In some embodiments, the T cell antigen is CD3, and the first tumor-associated antigen is selected from the group consisting of: CEA, CEA-CAM6, HER2, and EGFR.

[0052] In some embodiments, the T cell antigen is CD3, and the second tumor-associated antigen is selected from the group consisting of: CEA, CEA-CAM6, HER2, and EGFR.

[0053] In one aspect, the present disclosure relates to a protein complex comprising

[0054] (a) a first polypeptide comprising, in the N-terminal to C-terminal direction: a VH, a CH1 domain, optionally a first hinge region, a first CH2 domain, a first CH3 domain and a first VHH;

[0055] (b) a second polypeptide comprising, in the N-terminal to C-terminal direction: optionally a second hinge region, a second CH2 domain, a second CH3 domain, and a second VHH; and

[0056] (c) a third polypeptide comprising, in the N-terminal to C-terminal direction: a VL and a CL domain.

[0057] In some embodiments, the VH and the VL bind to each other to form an antigen binding site of the Fab that specifically binds to a T cell antigen. In some embodiments, the first VHH specifically binds to a first tumor-associated antigen and the second VHH specifically binds to a second tumor-associated antigen.

[0058] In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 1; the second polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 2; and the third polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 3.

[0059] In some embodiments, the T cell antigen is CD3, and the first tumor-associated antigen is selected from the group consisting of: CEA, CEA-CAM6, HER2, and EGFR.

[0060] In some embodiments, the T cell antigen is CD3, and the second tumor-associated antigen is selected from the group consisting of: CEA, CEA-CAM6, HER2, and EGFR.

[0061] In one aspect, the present disclosure relates to a protein complex comprising

[0062] (a) a first polypeptide comprising, in the N-terminal to C-terminal direction: an optional first hinge region, a first CH2 domain and a first CH3 domain, and a first VHH;

[0063] (b) a second polypeptide comprising, in the N-terminal to C-terminal direction: a VH, a CH1 domain, optionally a second hinge region, a second CH2 domain, a second CH3 domain, and a second VHH; and

[0064] (c) a third polypeptide comprising, in the N-terminal to C-terminal direction: a VL and a CL domain.

[0065] In some embodiments, the VH and the VL bind to each other to form an antigen binding site of the Fab that specifically binds to a T cell antigen. In some embodiments, the first VHH specifically binds to a first tumor-associated antigen and the second VHH specifically binds to a second tumor-associated antigen.

[0066] In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 1; the second polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 2; and the third polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 3.

[0067] In some embodiments, the T cell antigen is CD3, and the first tumor-associated antigen is selected from the group consisting of: CEA, CEA-CAM6, HER2, and EGFR.

[0068] In some embodiments, the T cell antigen is CD3, and the second tumor-associated antigen is selected from the group consisting of: CEA, CEA-CAM6, HER2, and EGFR.

[0069] In some embodiments, the first CH3 domain comprises one or more knob mutations and the second CH3 domain comprises one or more hole mutations.

[0070] In some embodiments, the VH comprises a sequence at least 80%, 90%, 95%, or 100% identical to SEQ ID NO:5, and the VL comprises a sequence at least 80%, 90%, 95%, or 100% identical to SEQ ID NO:6.

[0071] In some embodiments, the VHH comprises a sequence that is at least 80%, 90%, 95%, or 100% identical to any one of SEQ ID NOs: 7-10.

[0072] In some embodiments, the VHH is connected to the CH3 domain via a linker peptide.

[0073] In one aspect, the disclosure relates to a nucleic acid comprising a polynucleotide encoding an antigen binding protein described herein or a protein complex described herein.

[0074] In some embodiments, the nucleic acid is DNA (eg, cDNA) or RNA (eg, mRNA).

[0075] In one aspect, the disclosure relates to a vector comprising one or more of the nucleic acids described herein.

[0076] In one aspect, the disclosure relates to a cell comprising a vector described herein.

[0077] In some embodiments, the cell is a HEK293F cell or a CHO cell.

[0078] In one aspect, the disclosure relates to a cell comprising one or more of the nucleic acids described herein.

[0079] In one aspect, the disclosure relates to a method of producing an antigen binding protein or protein complex, the method comprising culturing a cell described herein under conditions sufficient for the cell to produce the antigen binding protein or protein complex; and collecting the antigen binding protein or protein complex produced by the cell.

[0080] In one aspect, the present disclosure relates to an antibody-drug conjugate comprising an antigen binding protein described herein or a protein complex described herein covalently bound to a therapeutic agent.

[0081] In some embodiments, the therapeutic agent is a cytotoxic or cytostatic agent.

[0082] In one aspect, the disclosure relates to a method of treating a subject having cancer, the method comprising administering to the subject a therapeutically effective amount of a composition comprising an antigen binding protein described herein, a protein complex described herein, or an antibody-drug conjugate described herein.

[0083] In some embodiments, the subject has a cancer that expresses CEA-CAM6.

[0084] In some embodiments, the cancer is lung cancer, colorectal cancer, head and neck cancer, gastric cancer, pancreatic cancer, urothelial cancer, breast cancer, cervical cancer, or endometrial cancer.

[0085] In one aspect, the disclosure relates to a method of reducing tumor growth rate, the method comprising contacting tumor cells with an effective amount of a composition comprising an antigen binding protein described herein, a protein complex described herein, or an antibody-drug conjugate described herein.

[0086] In one aspect, the present disclosure relates to a method of killing tumor cells, the method comprising contacting the tumor cells with an effective amount of a composition comprising an antigen binding protein described herein, a protein complex described herein, or an antibody-drug conjugate described herein.

[0087] In one aspect, the present disclosure relates to a pharmaceutical composition comprising an antigen binding protein described herein or a protein complex described herein, and a pharmaceutically acceptable carrier.

[0088] In one aspect, the present disclosure provides an antigen binding protein comprising (a) Fc; (b) a Fab fragment (Fab) that specifically binds to a T cell antigen; and (c) a single domain antibody variable domain (VHH) that specifically binds to a tumor-associated antigen.

[0089] In some embodiments, the Fab and the VHH are connected to the Fc. In some embodiments, the Fab comprises or consists of a light chain variable domain (VL), a light chain constant domain (CL), a heavy chain variable domain (VH) and a heavy chain first constant domain (CH1). In some embodiments, the Fab can activate T cells after binding to a T cell antigen. In some embodiments, the T cell antigen is a differentiation cluster 3 (CD3).

[0090] In some embodiments, the tumor-associated antigen is cluster of differentiation 20 (CD20), carcinoembryonic antigen (CEA), prostate specific antigen (PSA), prostate stem cell antigen (PSCA), programmed death ligand 1 (PD-L1), human epidermal growth factor receptor 2 (HER2), human epidermal growth factor receptor 3 (Her3), human epidermal growth factor receptor (Her1), β-catenin, cluster of differentiation 19 (CD19), epidermal growth factor receptor (EGFR), tyrosine protein kinase Met (c-Met), epithelial cell adhesion molecule (EPCAM), prostate specific membrane antigen (PSMA), cluster of differentiation 40 (CD40), cell surface-associated mucin 1 (MUC1), insulin-like growth factor 1 receptor (IGF1R) and carcinoembryonic antigen cell adhesion molecule 6 (CEA-CAM6).

[0091] In some embodiments, the Fc is a human IgG4 Fc. In some embodiments, the CH1 domain of the Fab is connected to the CH2 domain of the Fc, optionally through a hinge region. In some embodiments, the hinge region is a human IgG4 hinge region, optionally with an S228P mutation according to EU numbering. In some embodiments, the Fab is connected to the C-terminus of the CH3 domain in the Fc. In some embodiments, the Fab is connected to the CH3 domain through a linker peptide. In some embodiments, the VHH is connected to the CH2 domain in the Fc, optionally through a hinge region. In some embodiments, the hinge region is a human IgG4 hinge region, optionally with an S228P mutation according to EU numbering. In some embodiments, the VHH is connected to the C-terminus of the CH3 domain in the Fc. In some embodiments, the VHH is connected to the CH3 domain through a linker peptide. In some embodiments, the Fc comprises a first polypeptide and a second polypeptide. In some embodiments, each polypeptide comprises one or more "knobs-into-hole" mutations.

[0092] As used herein, the term "antibody" refers to any antigen binding molecule containing at least one (e.g., one, two, three, four, five or six) complementary determining regions (CDRs) (e.g., any one of the three CDRs of an immunoglobulin light chain or any one of the three CDRs of an immunoglobulin heavy chain) and capable of specifically binding to an antigen epitope. Non-limiting examples of antibodies include: monoclonal antibodies, polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies), single-chain antibodies, single variable domain (VHH) antibodies, chimeric antibodies, human antibodies, and humanized antibodies. In some embodiments, the antibody may include the Fc region of a human antibody. The term antibody also includes derivatives, such as multispecific antibodies, bispecific antibodies, trispecific antibodies, single-chain antibodies, diabodies, linear antibodies and antigen-binding protein constructs formed by these antibodies or antibody fragments.

[0093] As used herein, the term "antigen binding fragment" refers to a portion of a full-length antibody, wherein the antibody portion is capable of specifically binding to an antigen. In some embodiments, the antigen binding fragment comprises at least one variable domain (e.g., a heavy chain variable domain, a light chain variable domain, or VHH). Non-limiting examples of antibody fragments include, for example, Fab, Fab', F(ab')2 and Fv fragments, scFv and VHH.

[0094] As used herein, the terms "subject" and "patient" are used interchangeably throughout the specification to describe an animal, human or non-human being to whom treatment is provided according to the methods of the present invention. Veterinary and non-veterinary applications are contemplated in the present disclosure. Human patients can be adults or adolescents (e.g., people under 18 years of age). In addition to humans, patients include, but are not limited to, mice, rats, hamsters, guinea pigs, rabbits, ferrets, cats, dogs and primates. Including, for example, non-human primates (e.g., monkeys, chimpanzees, gorillas, etc.), rodents (e.g., rats, mice, gerbils, hamsters, ferrets, rabbits), lagomorphs, swine (e.g., pigs, miniature pigs), horses, dogs, cats, cattle and other domestic, farm and zoo animals.

[0095] As used herein, when referring to an antibody or antigen-binding fragment, the terms "specific binding" and "specific binding" mean that the antibody or antigen-binding fragment is more inclined to interact with its target molecule than other molecules because this interaction depends on the presence of a specific structure (i.e., antigenic determinant or epitope) on the target molecule; in other words, the reactant generally recognizes and binds to molecules containing a specific structure, rather than all molecules. An antibody that specifically binds to a target molecule may be referred to as a target-specific antibody. For example, an antibody that specifically binds to CEA-CAM6 may be referred to as a CEA-CAM6-specific antibody or an anti-CEA-CAM6 antibody.

[0096] As used herein, the term "bispecific antibody" refers to an antibody that binds to two different epitopes. The epitopes can be on the same antigen or on different antigens.

[0097] As used herein, the term "trispecific antibody" refers to an antibody that binds to three different epitopes. The epitopes can be on the same antigen or on different antigens.

[0098] As used herein, the term "multispecific antibody" refers to an antibody that binds to two or more different epitopes. The epitopes can be on the same antigen or on different antigens. The multispecific antibody can be, for example, a bispecific antibody or a trispecific antibody. In some embodiments, the multispecific antibody binds to two, three, four, five or six different epitopes.

[0099] As used herein, "VHH" refers to the variable domain of a heavy chain antibody. In some embodiments, the VHH is a humanized VHH. In some embodiments, the VHH is a single domain antibody (sdAb).

[0100] As used herein, the terms "polypeptide," "peptide," and "protein" are used interchangeably to refer to amino acid polymers of any length of at least two amino acids.

[0101] As used herein, the terms "polynucleotide," "nucleic acid molecule," and "nucleic acid sequence" are used interchangeably to refer to nucleotide polymers of any length of at least two nucleotides, including but not limited to DNA, RNA, DNA / RNA hybrids, and modifications thereof.

[0102] As used herein, a "single-sided" multispecific antibody refers to a multispecific antibody in which all antigen binding domains are located on the same side of the Fc region (the N-terminus of the Fc region or the C-terminus of the Fc region). In some embodiments, the single-sided multispecific antibody is a single-sided bispecific antibody or a single-sided trispecific antibody.

[0103] As used herein, a "non-unilateral" multispecific antibody refers to a multispecific antibody in which not all antigen binding domains are located on the same side of the Fc region (the N-terminus of the Fc region or the C-terminus of the Fc region). In some embodiments, the non-unilateral multispecific antibody is a non-unilateral bispecific antibody or a non-unilateral trispecific antibody.

[0104] Unless otherwise indicated, all technical and scientific terms used herein have the meaning commonly understood by one of ordinary skill in the art to which the invention belongs. Methods and materials for use in the present invention are described herein; other suitable methods and materials known in the art may also be used. The materials, methods, and examples are illustrative only and are not limiting. All publications, patent applications, patents, sequences, database entries, and other references described herein are incorporated herein by reference in their entirety. In the event of a conflict, the present specification (including definitions) shall prevail.

[0105] Other features and advantages of the invention will be apparent from the following detailed description and drawings, and from the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0106] Figures 1A-1D Schematic structures of several bispecific antibody constructs (CEA-BsAb, EGFR-BsAb, HER2-BsAb, and CEACAM6-BsAb) are shown, which include a tumor-associated antigen targeting moiety in the form of a Fab (T cell activator) and an sdAb.

[0107] Figure 2A-2L Schematic structures of several non-unilateral trispecific antibody constructs (301-312) are shown, which include (1) Fab (T cell activator), (2) a tumor-associated antigen targeting moiety in the form of an sdAb, and (3) another tumor-associated antigen targeting moiety in the form of an sdAb.

[0108] Figure 3A-3BThe SDS-PAGE (sodium dodecyl sulfate-polyacrylamide gel electrophoresis) result graph of the trispecific antibody (301-312) is shown. Each trispecific antibody sample was analyzed under non-reducing or reducing conditions. MK is the molecular weight (MW) marker. The name of the trispecific antibody is marked on each gel image. The band size is also marked on the left side of each gel image.

[0109] Figures 4A-4F The non-unilateral CD3 / CEA / EGFR trispecific antibody-induced T cell-mediated cell killing effect on target tumor cells was shown. The following tumor target cells were tested: NCI-H157 cells ( Figure 4A-4B )、NCI-H1299 cells( Figure 4C-4D ) and NCI-H1395 cells ( Figures 4E-4F ).

[0110] Figures 5A-5N Cytokine secretion (IL-2, IL-6, IFN-γ and TNF-α) following T cell-mediated killing of tumor cells (NCI-H157) induced by CD3 / CEA / EGFR trispecific antibodies (301-312) is shown.

[0111] Figure 6A-6N Cytokine secretion (IL-2, IL-6, IFN-γ and TNF-α) following T cell-mediated killing of tumor cells (NCI-H1299) induced by CD3 / CEA / EGFR trispecific antibodies (301-312) is shown.

[0112] Figures 7A-7N Cytokine secretion (IL-2, IL-6, IFN-γ and TNF-α) following T cell-mediated killing of tumor cells (NCI-H1395) induced by CD3 / CEA / EGFR trispecific antibodies (301-312) is shown.

[0113] Figures 8A-8P show the schematic structures of several single-sided trispecific antibody constructs (401-416), which include (1) Fab (T cell activator), (2) a tumor-associated antigen targeting portion in the form of an sdAb, and (3) another tumor-associated antigen targeting portion in the form of an sdAb.

[0114] Figures 9A-9B The SDS-PAGE results of the trispecific antibody (401-416) are shown. Each trispecific antibody sample was analyzed under non-reducing or reducing conditions. MK is the molecular weight (MW) marker. The name of the trispecific antibody is marked on each gel image. The band size is also marked on the left side of each gel image.

[0115] Figures 10A-10F The single-sided trispecific antibody (401-416) induced T cell-mediated cell killing effect on target tumor cells. The following tumor target cells were tested: NCI-H157 cells ( Figures 10A-10B )、NCI-H1299 cells( Figures 10C-10D )、NCI-H1395 cells( Figures 10E-10F )、NCI-H1573( Figure 10G-10H )、NCI-H157-21#( Fig.10I ) and SK-BR-3( Fig.10J ).

[0116] Figures 11A-11R Schematic structures of several single-sided trispecific antibody constructs (501, 503, 509-512, 601, 603, 609-612, 701, 703, 709-712) are shown, which contain (1) Fab (T cell activator), (2) a tumor-associated antigen targeting portion in the form of an sdAb, and (3) another tumor-associated antigen targeting portion in the form of an sdAb.

[0117] Figures 12A-12B The SDS-PAGE results of the trispecific antibodies (501, 503, 509-512, 601, 603, 609-612, 701, 703, 709-712) are shown. Each trispecific antibody sample was analyzed under non-reducing or reducing conditions. MK is a molecular weight (MW) marker. The name of the trispecific antibody is marked on each gel image. The band size is also marked on the left side of each gel image.

[0118] Figures 13A-13F The unilateral trispecific antibodies (501, 503, 509-512, 601, 603, 609-612, 701, 703, 709-712) induced T cell-mediated cell killing effect on target tumor cells. The following tumor target cells were tested: NCI-H1975 cells ( Fig.13A )、MDA-MB-468 cells( Fig. 13B )、NCI-H157-21# cells( Fig. 13C ), SK-OV-3 cells ( Figures 13D-13F )、NCI-H1573 cells( Figures 13G-13I )、MCF-7 cells( Figures 13J-13K )、MKN-45 cells( Figures 13L-13N ), SK-BR-3 cells ( Figures 13O-13Q ) and MDA-MB-453 cells ( Figure 13R-13T ).

[0119] Fig.14The schematic structure of a control bispecific antibody is shown.

[0120] Figures 15A-15L Schematic structures of several single-sided trispecific antibody constructs (801-809, 813-815) are shown, comprising (1) Fab (T cell activator), (2) a tumor-associated antigen targeting moiety in the form of an sdAb, and (3) another tumor-associated antigen targeting moiety in the form of an sdAb.

[0121] Figures 16A-16B The SDS-PAGE results of the trispecific antibodies (801-809, 813-815) are shown. Each trispecific antibody sample was analyzed under non-reducing or reducing conditions. M is the molecular weight (MW) marker. The name of the trispecific antibody is marked on each gel image. The band size is also marked on the left side of each gel image.

[0122] Figures 17A-17H The anti-tumor effect of the single-sided trispecific antibody (801-809, 813-815) in mice was shown. Figures 17A-17D Tumor size data are shown. Figures 17E-17H The body weights of mice are shown. DETAILED DESCRIPTION

[0123] The present disclosure relates to multispecific antibodies (e.g., trispecific antibodies) or antigen-binding proteins thereof. On the one hand, the multispecific antibodies or antigen-binding proteins thereof can be combined with T cell antigens (e.g., CD3) and / or one or more tumor-associated antigens (e.g., CEA-CAM6, EGFR, HER2) or a combination thereof.

[0124] Cluster of differentiation 3 (CD3) is considered in the art to be a multiprotein complex composed of six chains (see Abbas and Lichtman, 2003; Janeway et al., pp. 172 and 178, 1999). In mammals, the complex consists of a homodimer of one CD3γ chain, one CD3δ chain, two CD3ε chains, and one CD3ζ chain. The CD3γ, CD3δ, and CD3ε chains are highly related cell surface proteins of the immunoglobulin superfamily that contain a single immunoglobulin domain. The transmembrane regions of the CD3γ, CD3δ, and CD3ε chains are negatively charged, a property that allows these chains to bind to the positively charged T cell receptor chains. The intracellular tails of the CD3γ, CD3δ, and CD3ε chains each contain a single conserved motif, called an immunoreceptor tyrosine-based activation motif, or ITAM, while each CD3ζ chain has three. ITAMs are thought to be important for the signaling ability of the TCR complex.

[0125] Carcinoembryonic antigen (CEA) is a substance present on the surface of some cells. It is a glycoprotein produced by cells of the digestive tract during embryonic development. It is produced in very small amounts after birth. Therefore, unless some disease, including some forms of cancer, is present, the level of CEA in the bloodstream is relatively low. In the past, CEA was considered to be a carcinoembryonic antigen that is expressed during embryonic development and only re-expressed in cancer patients. In fact, CEA is also expressed in normal adults, but it is produced in the colon and then disappears in the feces. This cell surface glycoprotein plays a role in cell adhesion and intracellular signaling. In colon cancer, tumor cells lose their polarity and CEA is distributed around the cell surface and eventually enters the blood. Compared to the well-known glucose oxidase with a molecular weight of 160kDa, CEA is a complex, highly glycosylated large molecule (50% carbohydrate) with a molecular weight of ~200kDa. In fact, CEA has a broad band with an average molecular weight of 200kDa and a narrow band with an average molecular weight of 130kDa in normal colon tissue. However, in cancer cells, only a single band of 170 kDa or less is observed, partly due to the modification of the glycosylation pattern of CEA by N-glycanase. As a stable molecule, CEA is one of the most widely used tumor biomarkers and a prognostic indicator in clinical testing. The level of CEA in the blood of healthy adults should be extremely low, while abnormal CEA levels may be a sign of cancer, especially colon and rectal cancer. The level of CEA in the serum of patients with colorectal cancer is usually higher than that of healthy people, at ~2.5 μg / L. CEA may also be present in patients with pancreatic cancer, liver cancer, breast cancer, ovarian cancer, or lung cancer. Therefore, CEA levels measured before and after surgery indicate the success of the surgery and the prognosis of the patient's recovery. CEA levels can also be tested during chemotherapy to assess treatment progress and outcome.

[0126] The epidermal growth factor receptor (EGFR) belongs to the ErbB family of receptor tyrosine kinases (RTKs) and plays an important role in epithelial cell physiology (Schlessinger, 2014). It is frequently mutated and / or overexpressed in different types of human cancers and is a target for a variety of cancer therapies currently used in clinical practice (Yarden and Pines, 2012). The physiological function of the epidermal growth factor receptor (EGFR) is to regulate the development and homeostasis of epithelial tissues. In pathological settings, mainly in lung cancer, breast cancer and glioblastoma, EGFR is a driver of tumorigenesis. Inappropriate activation of EGFR in cancer mainly stems from amplification and point mutations of genomic loci, but transcriptional upregulation or ligand overproduction due to autocrine / paracrine mechanisms have also been described. In addition, EGFR is increasingly being considered as a biomarker of tumor resistance, as it has been found to be amplified or mutated secondary to drug stress. In addition to the important functions played by this receptor in normal epithelium, this evidence has prompted in-depth research on the role of EGFR at both physiological and pathological levels.

[0127] Human epidermal growth factor receptor 2 (HER2) is a transmembrane growth factor receptor present in normal and malignant breast epithelial cells. Phosphorylation by intracellular tyrosine kinases results in intracellular signaling and activation of genes involved in cell growth. In early breast cancer, overexpression of HER2 has independent prognostic significance and also predicts response to hormonal and cytotoxic therapy, although the latter role has been less studied. In current clinical trials, prospective classification of HER2 status may more accurately characterize these effects. In patients with metastatic HER2-overexpressing breast cancer, anti-HER2 therapy with a humanized monoclonal antibody improves survival when given with chemotherapy compared with chemotherapy alone. A potential limitation to its use in the adjuvant setting is the increased incidence of cardiotoxicity in patients who have concurrent or prior anthracycline therapy; well-designed prospective adjuvant trials are currently underway. HER2 is a relatively new prognostic marker that holds promise for predicting response to multiple therapies and for target-specific therapies.

[0128] Carcinoembryonic antigen (CEA)-related cell adhesion molecule 6 (CEACAM6) is a glycophosphoinositide-anchored glycoprotein belonging to the CEA family. A large body of evidence has demonstrated that CEACAM6 is upregulated in gastrointestinal tumors and promotes the progression of gastrointestinal cancers through multiple mechanisms. However, the role of CEACAM6 in non-gastrointestinal cancer cells remains controversial. For example, CEACAM6 is used as a marker for the isolation of cancer stem cells in colorectal cancer, while a subpopulation of CEACAM6-positive breast cancer cells exhibits less sphere formation and tumorigenesis in nude mice compared with CEACAM6-negative breast cancer cells. This discrepancy is currently unresolved, but it is suspected that post-translational glycosylation of CEACAM6 may be key, as glycosylation is a cell type-dependent process. The CEACAM6 polypeptide contains 344 amino acids with a predicted molecular weight (MW) of 37 kDa, but multiple glycosylation isoforms have been reported in different types of tissues and at different developmental stages. CEACAM6 is a heavily glycosylated tumor-associated molecule with a MW range of 75–100 kDa in pancreatic and colorectal cancer. In addition, biochemical studies have shown that N-glycosylation of the immunoglobulin V (IgV) domain of CEACAM affects its oligomerization or dimerization. CEACAM5 glycans in colorectal cancer cells interact with antigen-presenting cells through glycan-binding proteins such as DC-SIGN (human C-type lectin). 17 These results suggest that glycosylation can regulate the protein structure and biological function of CEACAM.

[0129] In general, multispecific antibodies (e.g., trispecific antibodies) include two or more antigen binding sites for different antigens or different epitopes of the same antigen. Therefore, compared with monospecific antibodies, multispecific antibodies (e.g., trispecific antibodies) can have more functions. For example, these functions include, but are not limited to: stronger binding to antigens through avidity effects; co-localization of bound antigens on the cell surface and their effects; extending the serum half-life of an antibody fragment by linking it to a second antibody fragment that binds to a protein with a longer serum half-life (e.g., albumin or transferrin); and bringing two cells into proximity by binding to antigens on each cell.

[0130] Among the purposes of multispecific antibodies (e.g., trispecific antibodies), one class of molecules, T cell engagers (TCEs), has received more attention. TCEs are multispecific antibodies (e.g., trispecific antibodies) that can simultaneously bind to an antigen on a T cell and an antigen on another cell. CD3 is usually selected as the antigen on a T cell. As discussed above, cancer cells or tumor cells are usually selected as another cell type. By binding to CD3 on T cells and tumor associated antigens (TAA) on cancer cells, TCEs can induce T cells to activate and kill the latter after binding to cancer cells.

[0131] However, there are still some difficulties to overcome in order to generate the desired homogeneous multispecific antibodies (e.g., trispecific antibodies). The first difficulty is the mispairing of heavy chains that bind to the same target (e.g., antigen or epitope). For example, to generate the desired form of multispecific antibodies, heavy chains targeting different targets should ideally form heterodimers. However, the percentage of desired multispecific antibodies varies greatly in different constructs. Mutations that induce the formation of a "knob-and-hole" structure between two heavy chains can be used to reduce the formation of homodimers of heavy chains that bind to the same target. Exemplary amino acid sequences of knob-chain and hole-chain Fc that promote heterodimer formation are shown in SEQ ID NO: 1 and SEQ ID NO: 2, respectively.

[0132] The second difficulty to overcome is the mismatch between the heavy chain variable region (VH) and the light chain variable region (VL). Monoclonal antibodies have two identical Fab fragments, each with a pair of VH and VL. In contrast, multispecific antibodies usually have several different heavy chain variable regions and several different light chain variable regions. Therefore, each VH may bind to multiple VLs, and each VL may bind to multiple VHs. Therefore, if this mismatch problem is not solved, only some of the multispecific antibodies formed will be functional.

[0133] Several strategies have been designed to eliminate this mismatch phenomenon. One solution is to design an antibody with a universal light chain. Specifically, a light chain, or more precisely a VL, is selected that can form dimers with all VHs. This design eliminates the need to match VHs and VLs with the same target binding specificity. The disadvantage of this strategy is that the contribution of VL to target binding is greatly reduced, making it difficult to find the optimal VH, because VH is largely responsible for target binding, even if it is not completely responsible.

[0134] Another solution is to use CrossMAb technology, in which the VL is fused to the heavy chain constant domain 1 (CH1) to become part of the heavy chain. At the same time, the corresponding VH is fused to the light chain constant region (CL) to become part of the light chain.

[0135] The present disclosure provides a different strategy. The reason for adopting the technology of, for example, CrossMAb or universal light chain is that conventional antibodies have and need heavy chain and light chain to play a role and / or remain stable. However, it is not necessary to have both heavy chain and light chain at the same time to bind to the target. For example, the Fab fragment (Fab) comprises the variable domain and constant domain of the light chain, as well as the variable domain and the first constant domain (CH1) of the heavy chain, and has a complete antigen binding function. The variable domain of the heavy chain antibody (VHH), for example, derived from camelids such as llamas, camels or alpacas, has a complete antigen binding function. By fusing antibody fragments (e.g., Fab or VHH) with human Fc with heterodimer preference, multispecific antibodies can be produced without considering the mismatch between the heavy chain and the light chain.

[0136] In the present disclosure, a Fab that can bind to human CD3 and a VHH that can bind to one or more tumor-associated antigens (e.g., HER2, CEA-CAM6, or EGFR) are provided. One of the antibody fragments is located at the four ends of the Fc (the N-terminus or C-terminus of the knob chain or the hole chain), and the other antibody fragment is located at the other three available ends of the Fc. This design produces 4 different bispecific antibody molecules (see Figures 1A-1D ) and 12 different non-unilateral trispecific antibodies (see Figure 2A-2L , corresponding to 301-312). Further, 16 different unilateral trispecific antibodies were constructed, in which the CD3 Fab and two VHHs were located on the same side of the Fc region (see Figures 8A-8P, corresponding to 401-416). Using different VHHs, another 18 unilateral trispecific antibodies were constructed (see Figures 11A-11R , corresponding to 501, 503, 509-512, 601, 603, 609-612, 701, 703, 709-712). Using different VHHs, another 12 unilateral trispecific antibodies were constructed (see Figures 15A-15L , corresponding to 801-809, 813-815). These molecules are provided, characterized for their ability to form heterodimers, and tested for their ability to induce T cell-mediated tumor cell killing in the presence of human PBMCs.

[0137] In some embodiments, the multispecific antibody or antigen-binding protein may include 1, 2, 3, 4 or more than 4 Fab fragments. In some embodiments, the multispecific antibody or antigen-binding protein may include 1, 2, 3, 4, 5 or more than 5 VHHs. In some embodiments, the Fab may target CD3 or another tumor-associated antigen. In some embodiments, the VHH may target CD3 or another tumor-associated antigen. The Fab, the VHH and various forms of multispecific antibodies or antigen-binding proteins will be described in detail below.

[0138] Fab fragment (Fab)

[0139] The Fab fragment (Fab) contains the variable and constant domains of the light chain and the variable domain and the first constant domain (CH1) of the heavy chain.

[0140] The present disclosure provides anti-CD3 antibodies, modified antibodies thereof, chimeric antibodies thereof, and humanized antibodies thereof, etc. The present disclosure also provides Fab fragments targeting CD3. The Fab can be used in the various multispecific antibody constructs described herein.

[0141] In some embodiments, the Fab may have a VH that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 5 (TA1). In some embodiments, the Fab may have a VL that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 6 (TA1).

[0142] In some embodiments, the Fab, antibody or antigen-binding fragment described herein may have a heavy chain variable region (VH) comprising a VH CDR1, VH CDR2, and VH CDR3 identical to the VH CDR1, VH CDR2, and VH CDR3 of any VH described herein (e.g., SEQ ID NO: 5); and a light chain variable region (VL) comprising a VL CDR1, VL CDR2, and VL CDR3 identical to the VL CDR1, VL CDR2, and VL CDR3 of a VL described herein (e.g., SEQ ID NO: 6).

[0143] In some embodiments, the Fab, antibody or antigen-binding fragment described herein may comprise a VH comprising a VH CDR1 having zero, one or two amino acid insertions, deletions or substitutions, a VH CDR2 having zero, one or two amino acid insertions, deletions or substitutions, a VH CDR3 having zero, one or two amino acid insertions, deletions or substitutions; and a VL comprising one, two or three of the following: a VL CDR1 having zero, one or two amino acid insertions, deletions or substitutions; a VL CDR2 having zero, one or two amino acid insertions, deletions or substitutions; a VLCDR3 having zero, one or two amino acid insertions, deletions or substitutions, wherein the VH CDR is selected from any VH described herein and the VL CDR is selected from any VL described herein.

[0144] Heavy chain antibody variable domain (VHH)

[0145] Monoclonal antibodies and recombinant antibodies are important tools in medicine and biotechnology. Like all mammals, camelids (e.g., llamas) can produce conventional antibodies (e.g., IgG1) composed of two heavy chains and two light chains bound by disulfide bonds into a Y shape. However, they also produce two unique IgG subclasses: IgG2 and IgG3, also known as heavy chain antibodies. These antibodies consist of only two heavy chains, lack the CH1 region, but still have an antigen binding domain at their N-terminus, called VHH (or nanobody). Conventional Ig requires the variable regions on both the heavy and light chains to be combined to allow a high diversity of antigen-antibody interactions. Although isolated heavy and light chains still have this ability, their affinity is very low compared to paired heavy and light chains. The unique feature of heavy chain antibodies is that their monomeric antigen binding region is able to bind antigens with specificity, affinity, and especially diversity comparable to conventional antibodies without matching other regions. This feature is mainly due to several major variations in the amino acid sequences of the two heavy chain variable regions, which cause deep conformational changes compared to conventional Ig. Major substitutions in the variable region prevent the light chain from binding to the heavy chain, but also prevent the unbound heavy chain from being retrieved by the Immunoglobulin Binding Protein.

[0146] The single variable domain of these antibodies (called VHH, sdAb, nanobody or heavy chain antibody variable domain) is the smallest antigen binding domain produced by the adaptive immune system. The third complementary determining region (CDR3) of these antibody variable regions is usually twice as long as that of conventional antibodies. This leads to an increase in the interaction surface with the antigen and an increase in the diversity of antigen-antibody interactions, thereby compensating for the lack of light chains. Due to the longer complementary determining region 3 (CDR3), VHH can extend into protein crevices that conventional antibodies cannot enter, including functionally interesting sites, such as the active site of an enzyme or the receptor binding canyon on the surface of a virus. In addition, the additional cysteine ​​residues make the structure more stable, thereby enhancing the strength of the interaction.

[0147] Compared with conventional antibodies with conventional antibody variable domains (VH and VL), VHH provides many other advantages, including higher stability, solubility, expression yield and refolding ability, and better in vivo tissue penetration. In addition, unlike the VH domain of conventional antibodies, VHH does not show an intrinsic tendency to bind to the light chain. This makes it easier to induce heavy chain antibodies in the presence of functional light chain sites. In addition, since VHH does not bind to the VL domain, it is much easier to reconstruct VHH into a multispecific antibody construct than a construct containing a conventional VH-VL pair or a single domain based on the VH domain.

[0148] The present disclosure provides anti-CEA-CAM6 antibodies, modified antibodies thereof, chimeric antibodies thereof, and humanized antibodies thereof, etc. The present disclosure also provides VHHs of these antibodies. These VHHs can be used in the various multispecific antibody constructs described herein.

[0149] Also provided are the amino acid sequences of various VHHs. In some embodiments, the VHH domain is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to any one of SEQ ID NOs: 7-10.

[0150] In some embodiments, the antibodies or antigen-binding fragments described herein may comprise a heavy chain antibody variable domain (VHH) comprising one, two or three of the following: VH CDR1 having zero, one or two amino acid insertions, deletions or substitutions; VH CDR2 having zero, one or two amino acid insertions, deletions or substitutions; VH CDR3 having zero, one or two amino acid insertions, deletions or substitutions, wherein VHH CDR1, VHH CDR2 and VHH CDR3 are selected from the CDRs of any one of SEQ ID NOs: 7-10.

[0151] The insertion, deletion and substitution can be within the CDR sequence or at one or both ends of the CDR sequence. In some embodiments, the CDR is determined based on the Kabat numbering scheme. In some embodiments, the CDR is determined based on the Chothia numbering scheme. In some embodiments, the CDR is determined based on a combined numbering scheme.

[0152] In order to determine the percentage identity of two amino acid sequences or two nucleic acid sequences, the sequences are compared to achieve the best comparison purpose (for example, a gap can be introduced in one or both of the first and second amino acid or nucleic acid sequences to achieve the best comparison, and non-homologous sequences can be ignored for comparison purposes). The amino acid residues or nucleotides at the corresponding amino acid positions or nucleotide positions are then compared. When a certain position of the first sequence is occupied by the same amino acid residue or nucleotide as the corresponding position of the second sequence, the molecules at that position are identical. The percentage identity between the two sequences is a function of the number of identical positions shared between the two sequences, taking into account the number of gaps and the length of each gap required to be introduced when the two sequences are optimally compared. For example, the comparison of sequences between two sequences and the determination of percentage identity can be achieved, for example, using a Blossum 62 scoring matrix with a gap penalty of 12, a gap extension penalty of 4, and a frameshift gap penalty of 5.

[0153] The present disclosure also provides nucleic acids comprising polynucleotides encoding polypeptides comprising immunoglobulin heavy chain antibody variable domains (VHH).

[0154] In some embodiments, the antibody or antigen-binding fragment thereof comprises an Fc domain, which may be derived from a variety of types (e.g., IgG, IgE, IgM, IgD, IgA, and IgY), classes (e.g., IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2), or subclasses. In some embodiments, the Fc domain is derived from an IgG antibody or antigen-binding fragment thereof. In some embodiments, the Fc domain comprises one, two, three, four, or more heavy chain constant regions.

[0155] The structure of multispecific antibodies

[0156] Multispecific antibodies (e.g., bispecific antibodies or trispecific antibodies) can be designed to include one or more antigen binding sites targeting T cell antigens (e.g., CD3, CD4, and CD8), and one or more antigen binding sites targeting tumor-associated antigens (e.g., CEA, CEA-CAM6, HER2, EGFR). Antigen binding sites may include Fab, scFv, VHH, etc. In some embodiments, one or more antigen binding sites targeting T cell antigens (e.g., CD3) may include Fab. In some embodiments, one or more antigen binding sites targeting tumor-associated antigens may include VHH. Tumor-associated antigens refer to antigens specifically expressed on the surface of tumor cells. These antigens can be used to identify tumor cells. Normal cells rarely express these tumor-associated antigens. Some exemplary tumor-associated antigens include, for example, CEA, CD20, PSA, PSCA, PD-L1, HER2, Her3, Her1, β-catenin, CD19, CEA-CAM6, EGFR, c-Met, EPCAM, PSMA, CD40, MUC1, and IGF1R, among others.

[0157] In some embodiments, the multispecific antibodies (e.g., bispecific antibodies or trispecific antibodies) or antigen-binding fragments thereof described herein include Fabs that specifically bind to T cell antigens. In some embodiments, the T cell antigen is CD3 (e.g., human CD3). In some embodiments, the T cell antigen is CD28. In some embodiments, the T cell antigen is CD27. In some embodiments, the T cell antigen is CD137. In some embodiments, the T cell antigen is OX40. In some embodiments, the T cell antigen is PD1. In some embodiments, the T cell antigen is CTLA-4. In some embodiments, the T cell antigen is Tim3. In some embodiments, the T cell antigen is LAG-3. In some embodiments, the multispecific antibody or antigen-binding fragment thereof can activate T cells after binding to T cell antigens.

[0158] In some embodiments, the multispecific antibodies (e.g., bispecific antibodies or trispecific antibodies) or antigen-binding fragments thereof described herein include VHHs that specifically bind to tumor-associated antigens. In some embodiments, the tumor-associated antigen is CEA-CAM6 (e.g., human CEA-CAM6). In some embodiments, the tumor-associated antigen is CEA. In some embodiments, the tumor-associated antigen is EGFR or HER2. In some embodiments, the tumor-associated antigen is claudin 18.2. In some embodiments, the tumor-associated antigen is CD166. In some embodiments, the tumor-associated antigen is glypican-3. These are just examples of tumor-associated antigens. Enumerating these antigens does not mean limiting the use of these antigens.

[0159] The present disclosure provides antigen binding protein constructs described herein in various forms. Although not intending to be bound by any theory, it is assumed that in the presence of target cells (e.g., cancer cells) and T cells, the protein constructs can effectively activate T cells.

[0160] In some embodiments, the multispecific antibody (e.g., bispecific antibody or trispecific antibody) is designed to include a Fab targeting CD3. In some embodiments, the multispecific antibody (e.g., bispecific antibody or trispecific antibody) is designed to include a VHH targeting CEA-CAM6. The multispecific antibody is as described below.

[0161] CD3 (Cluster of Differentiation 3) is a protein complex and T cell co-receptor that is involved in activating cytotoxic T cells (CD8+ naive T cells) and helper T cells (CD4+ naive T cells). It consists of four different chains. In mammals, the complex contains a CD3γ chain, a CD3δ chain, and two CD3ε chains. These chains bind to the T cell receptor (TCR) and CD3-zeta (ζ chain) to generate activation signals in T lymphocytes. TCR, CD3-zeta and other CD3 molecules together constitute the TCR complex. In some embodiments, the multispecific antibody targets CD3ε.

[0162] CEA-CAM6 (carcinoembryonic antigen-related cell adhesion molecule 6) is a cell surface glycoprotein and a member of the carcinoembryonic antigen (CEA) protein family. Members of this family play a role in cell adhesion and are widely used as tumor markers in cancer serum immunoassays.

[0163] The present disclosure provides multispecific antibodies (e.g., bispecific antibodies or trispecific antibodies) that bind to both T cell antigens (e.g., CD3) and tumor-associated antigens. Multispecific antibodies can be used to treat tumor-associated antigen-positive cancers in subjects (e.g., human patients). In some embodiments, the tumor-associated antigen-positive cancer is CEA-CAM6 positive (e.g., ovarian cancer, colon cancer, breast cancer, or non-small cell lung cancer).

[0164] In general, a multispecific antibody (e.g., a bispecific antibody or a trispecific antibody) described herein can be prepared, comprising: (a) a first polypeptide comprising a first Fc region (e.g., a CH2 domain and a CH3 domain); and (b) a second polypeptide comprising a second Fc region (e.g., a CH2 domain and a CH3 domain). In some embodiments, the first Fc region and / or the second Fc region are derived from human IgG4. In some embodiments, the first Fc region and / or the second Fc region comprise one or more knob-to-hole mutations. For example, according to EU numbering, the first Fc region (e.g., a CH3 domain in the Fc region) may comprise a tryptophan (Trp) at position 366; according to EU numbering, the second Fc region (e.g., a CH3 domain in the Fc region) may comprise one or more of a serine (Ser) at position 366, an alanine (Ala) at position 368, and / or a valine (Val) at position 407.

[0165] In some embodiments, the Fc region is derived from the Fc of any antibody described herein (e.g., IgG1, IgG2, IgG3, and IgG4). In some embodiments, the Fc region is human IgG1, IgG2, or IgG4 (e.g., human IgG4). In some embodiments, relative to the Fc region of wild-type human IgG (e.g., IgG4), the first Fc region and / or the second Fc region include additional mutations. For example, according to EU numbering, the first Fc region and / or the second Fc region may include a proline (Pro) at position 228 to reduce the chain exchange of multispecific antibodies. According to EU numbering, the first Fc region and / or the second Fc region may also include an alanine (Ala) at position 234 to reduce the ADCC effect of multispecific antibodies. According to EU numbering, the first Fc region may include a cysteine ​​(Cys) at position 354, and the second Fc region may further include a cysteine ​​(Cys) at position 349 to stabilize the multispecific antibody. According to EU numbering, the second Fc region may include lysine (Lys) at position 435 and / or phenylalanine (Phe) at position 436 to reduce the binding of the second polypeptide to protein A. In addition, in order to improve the stability of the antibody, glycine (Gly) at position 446 and / or lysine (Lys) at position 447 of the first Fc region and / or the second Fc region may be deleted. Although not intending to be bound by any theory, it will be appreciated by those skilled in the art that the mutations and deletions described herein may be introduced in the first Fc region or the second Fc region.

[0166] On the one hand, the present disclosure relates to an antigen binding protein comprising (a) Fc; (b) a first antigen binding site comprising a Fab that specifically binds to CD3; and (c) a second antigen binding site comprising a single-chain antibody variable domain (VHH) that specifically binds to a tumor-associated antigen, wherein in some embodiments, the first antigen binding site and the second antigen binding site are connected to the Fc.

[0167] In some embodiments, the first antigen binding site comprises a Fab fragment (Fab), which comprises a variable domain and a constant domain of a light chain and a variable domain and a first constant domain (CH1) of a heavy chain. In some embodiments, the Fab can activate T cells after binding to CD3. In some embodiments, the Fab is a human IgG4 Fab.

[0168] In some embodiments, the Fab is connected to the CH2 domain in the Fc, optionally via a hinge region. In some embodiments, the hinge region is a human IgG4 hinge region, optionally with an S228P mutation according to EU numbering.

[0169] In some embodiments, the Fab is connected to the C-terminus of the CH3 domain in the Fc. In some embodiments, the Fab is connected to the CH3 domain via a linker peptide.

[0170] In some embodiments, the VHH is connected to the CH2 domain in the Fc, optionally via a hinge region. In some embodiments, the hinge region is a human IgG4 hinge region, optionally with an S228P mutation according to EU numbering.

[0171] In some embodiments, the VHH is connected to the C-terminus of the CH3 domain in the Fc. In some embodiments, the VHH is connected to the CH3 domain via a linker peptide.

[0172] In some embodiments, the Fc comprises a first polypeptide chain and a second polypeptide chain, and in some embodiments, each chain comprises one or more knob-in-hole mutations.

[0173] In some embodiments, the VH comprises a sequence at least 80%, 90%, 95%, or 100% identical to SEQ ID NO:5, and the VL comprises a sequence at least 80%, 90%, 95%, or 100% identical to SEQ ID NO:6.

[0174] In some embodiments, the Fc region comprises a sequence that is at least 80%, 90%, 95%, or 100% identical to SEQ ID NO: 1 or 2.

[0175] In some embodiments, the constant domain of the light chain comprises a sequence that is at least 80%, 90%, 95%, or 100% identical to SEQ ID NO:3.

[0176] In some embodiments, the CH1 domain comprises a sequence that is at least 80%, 90%, 95%, or 100% identical to SEQ ID NO:4.

[0177] In some embodiments, the linker peptide comprises a sequence that is at least 80%, 90%, 95%, or 100% identical to SEQ ID NO:11.

[0178] Multispecific antibodies having various structures are described below.

[0179] 1.Fab-Knob+VHH-Hole

[0180] like Figures 1A-1DAs shown, a multispecific antibody (e.g., a bispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: VH, CH1 domain, optional first hinge region and first Fc region (e.g., CH2 domain and CH3 domain); (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: a single-chain antibody variable domain (VHH), an optional second hinge region and a second Fc region (e.g., CH2 domain and CH3 domain); and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: VL and CL. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0181] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0182] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0183] 2.Fab-Knob+VHH-Hole-VHH(301-302)

[0184] like Figure 2A-2B As shown in (301-302), a multispecific antibody (e.g., a trispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: VH, CH1 domain, optional first hinge region and first Fc region (e.g., CH2 domain and CH3 domain); (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: a first single-domain antibody variable domain (VHH), an optional second hinge region, a second Fc region (e.g., CH2 domain and CH3 domain), an optional linker peptide and a second VHH; and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: VL and CL. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0185] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0186] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0187] 3.Fab-Knob-VHH+VHH-Hole(303-304)

[0188] like Figure 2C-2DAs shown in (303-304), a multispecific antibody (e.g., a trispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: VH, CH1 domain, optional first hinge region, first Fc region (e.g., CH2 domain and CH3 domain), optional linker peptide and first VHH; (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: a second VHH, optional second hinge region, second Fc region (e.g., CH2 domain and CH3 domain); and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: VL and CL. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0189] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0190] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0191] 4.Fab-Knob-VHH+Hole-VHH(305-306)

[0192] like Figure 2E-2F As shown in (305-306), a multispecific antibody (e.g., a trispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: VH, CH1 domain, optional first hinge region, first Fc region (e.g., CH2 domain and CH3 domain), optional linker peptide and first VHH; (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: optional second hinge region, second Fc region (e.g., CH2 domain and CH3 domain), optional linker peptide and second VHH; and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: VL and CL. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0193] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0194] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0195] 5.VHH-Knob+Fab-Hole-VHH(307-308)

[0196] like Figure 2G-2HAs shown in (307-308), a multispecific antibody (e.g., a trispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: a first VHH, an optional first hinge region, a first Fc region (e.g., a CH2 domain and a CH3 domain); (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: a VH, a CH1 domain, an optional second hinge region, a second Fc region (e.g., a CH2 domain and a CH3 domain), an optional linker peptide, and a second VHH; and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: a VL and a CL. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0197] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0198] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0199] 6.VHH-Knob-VHH+Fab-Hole(309-310)

[0200] like Figure 2I-2G As shown in (309-310), a multispecific antibody (e.g., a trispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: a first VHH, an optional first hinge region, a first Fc region (e.g., a CH2 domain and a CH3 domain), an optional linker peptide, and a second VHH; (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: a VH, a CH1 domain, an optional second hinge region, a second Fc region (e.g., a CH2 domain and a CH3 domain); and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: a VL and a CL. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0201] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0202] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0203] 7.Knob-VHH+Fab-Hole-VHH(311-312)

[0204] like Figure 2K-2LAs shown in (311-312), a multispecific antibody (e.g., a trispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: an optional first hinge region, a first Fc region (e.g., a CH2 domain and a CH3 domain), an optional linker peptide, and a first VHH; (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: a VH, a CH1 domain, an optional second hinge region, a second Fc region (e.g., a CH2 domain and a CH3 domain), an optional linker peptide, and a second VHH; and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: a VL and a CL. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0205] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0206] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0207] 8.VHH-Fab-Knob+VHH-Hole(401, 407, 501, 601, 701, 801, 807)

[0208] As shown in Figures 8A (401), 8G (407), 11A (501), 11G (601), 11M (701), 15A (801) and 15G (807), a multispecific antibody (e.g., a trispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: VH, CH1 domain, optional first hinge region, first Fc region (e.g., CH2 domain and CH3 domain); (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: a first VHH, an optional second hinge region, a second Fc region (e.g., CH2 domain and CH3 domain); and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: VL, CL and a second VHH. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0209] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0210] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0211] 9.VHH-Knob+VHH-Fab-Hole(402, 408, 802, 808)

[0212] As shown in Figures 8B (402), 8H (408), 15B (802) and 15H (808), a multispecific antibody (e.g., a trispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: a first VHH, an optional first hinge region, a first Fc region (e.g., a CH2 domain and a CH3 domain); (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: a VH, a CH1 domain, an optional second hinge region, a second Fc region (e.g., a CH2 domain and a CH3 domain); and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: a VL, a CL and a second VHH. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0213] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0214] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0215] 10.VHH-Fab-Knob+VHH-Hole(403, 409, 503, 509, 603, 609, 703, 709, 803, 809)

[0216] As shown in Figures 8C (403), 8I (409), 11B (503), 11C (509), 11H (603), 11I (609), 11N (703), 11O (709), 15C (803) and 15I (809), a multispecific antibody (e.g., a trispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: VH, CH1 domain, optional first hinge region, first Fc region (e.g., CH2 domain and CH3 domain); (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: a first VHH, an optional second hinge region, a second Fc region (e.g., CH2 domain and CH3 domain); and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: a second VHH, VL and CL. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0217] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0218] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0219] 11.VHH-Knob+VHH-Fab-Hole(404, 410, 510, 610, 710, 804)

[0220] As shown in Figures 8D (404), 8J (410), 11D (510), 11J (610), 11P (710), and 15D (804), a multispecific antibody (e.g., a trispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: a first VHH, an optional first hinge region, a first Fc region (e.g., a CH2 domain and a CH3 domain); (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: a VH, a CH1 domain, an optional second hinge region, a second Fc region (e.g., a CH2 domain and a CH3 domain); and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: a second VHH, a VL, and a CL. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0221] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0222] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0223] 12.VHH-Fab-Knob+VHH-Hole(405, 411, 511, 611, 711, 805)

[0224] As shown in Figures 8E(405), 8K(411), 11E(511), 11K(611), 11Q(711) and 15E(805), a multispecific antibody (e.g., a trispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: a first VHH, VH, a CH1 domain, an optional first hinge region, a first Fc region (e.g., a CH2 domain and a CH3 domain); (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: a second VHH, an optional second hinge region, a second Fc region (e.g., a CH2 domain and a CH3 domain); and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: VL and CL. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0225] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0226] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0227] 13.VHH-Knob+VHH-Fab-Hole(406, 412, 512, 612, 712, 806)

[0228] As shown in Figures 8F (406), 8L (412), 11F (512), 11L (612), 11R (712) and 15F (806), a multispecific antibody (e.g., a trispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: a first VHH, an optional first hinge region, a first Fc region (e.g., a CH2 domain and a CH3 domain); (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: a second VHH, a VH, a CH1 domain, an optional second hinge region, a second Fc region (e.g., a CH2 domain and a CH3 domain); and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: VL and CL. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0229] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0230] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0231] 14.Fab-Knob+VHH-VHH-Hole(413, 415, 813, 815)

[0232] As shown in Figures 8M (413), 8O (415), 15G (813) and 15L (815), a multispecific antibody (e.g., a trispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: VH, CH1 domain, optional first hinge region, first Fc region (e.g., CH2 domain and CH3 domain); (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: a first VHH, a second VHH, an optional second hinge region, a second Fc region (e.g., CH2 domain and CH3 domain); and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: VL and CL. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0233] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0234] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0235] 15.VHH-VHH-Knob+Fab-Hole(414, 416, 814)

[0236] As shown in Figures 8N (414), 8P (416) and 15K (814), a multispecific antibody (e.g., a trispecific antibody) can be prepared, which comprises: (a) a first polypeptide, preferably comprising from N-terminus to C-terminus: a first VHH, a second VHH, an optional first hinge region, a first Fc region (e.g., a CH2 domain and a CH3 domain); (b) a second polypeptide, preferably comprising from N-terminus to C-terminus: a VH, a CH1 domain, an optional second hinge region, a second Fc region (e.g., a CH2 domain and a CH3 domain); and (c) a third polypeptide, preferably comprising from N-terminus to C-terminus: a VL and a CL. In some embodiments, the VH and the VL bind to each other to form an antigen binding site of a Fab that specifically binds to a T cell antigen. In some embodiments, the VHH specifically binds to a tumor-associated antigen.

[0237] In some embodiments, the first Fc region comprises one or more knob mutations. In some embodiments, the second Fc region comprises one or more hole mutations. In some embodiments, the first polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 1. In some embodiments, the second polypeptide comprises a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 2.

[0238] In some embodiments, the Fab can target CD3 (e.g., human CD3). In some embodiments, the VH comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 5, and the VL comprises a sequence having at least 80%, 90%, 95%, or 100% identity to SEQ ID NO: 6. In some embodiments, the VHH can target CEA and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 7. In some embodiments, the VHH can target EGFR and comprise a sequence having at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100% identity to SEQ ID NO: 8. In some embodiments, the VHH can target HER2 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 9. In some embodiments, the VHH can target CEA-CAM6 and comprise a sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or 100% identical to SEQ ID NO: 10. In some embodiments, the first hinge region and / or the second hinge region is derived from the hinge region of human IgG4.

[0239] Antigen binding protein construct characteristics

[0240] The anti-CD3, anti-TAA (tumor-associated antigen) or anti-CD3 / TAA antigen binding protein construct (e.g., an antibody, a bispecific antibody, a trispecific antibody, a multispecific antibody or an antibody fragment thereof) may comprise an antigen binding site derived from any anti-CD3 antibody, anti-TAA antibody (e.g., anti-CEA, anti-CEA-CAM6, anti-EGFR or anti-HER2) or any antigen binding fragment thereof described herein.

[0241] In some embodiments, the antibody or antigen binding fragment thereof described herein is a CEA antagonist. In some embodiments, the antibody or antigen binding fragment thereof is a CEA agonist. In some embodiments, the antibody or antigen binding fragment thereof described herein is a CEA-CAM6 antagonist. In some embodiments, the antibody or antigen binding fragment thereof is a CEA-CAM6 agonist. In some embodiments, the antibody or antigen binding fragment thereof described herein is a CD3 antagonist. In some embodiments, the antibody or antigen binding fragment thereof is a CD3 agonist. In some embodiments, the antibody or antigen binding fragment thereof described herein is an EGFR antagonist. In some embodiments, the antibody or antigen binding fragment thereof is an EGFR agonist. In some embodiments, the antibody or antigen binding fragment thereof described herein is a HER2 antagonist. In some embodiments, the antibody or antigen binding fragment thereof is a HER2 agonist.

[0242] In some embodiments, the antibodies or antigen-binding fragments thereof described herein can bind to CD3 and a target tumor-associated antigen (e.g., CEA, CEA-CAM6, EGFR, or HER2), thereby bridging T cells with target cells; activating T cells; and inducing T cells to directly kill cancer cells.

[0243] In some embodiments, the antibody (or antigen-binding fragment thereof) specifically binds to a target antigen (e.g., CD3, CEA, CEA-CAM6, EGFR, or HER2) with an off-rate (koff, kdis) of less than 0.1 s -1 Less than 0.01s -1 , less than 0.001s -1 , less than 0.0001s -1 , or less than 0.00001s -1 In some embodiments, the dissociation rate (koff) is greater than 0.01 s -1 , greater than 0.001s -1 , greater than 0.0001s -1 , greater than 0.00001s -1 , or greater than 0.000001s -1 In some embodiments, the kinetic binding rate (kon, ka) is greater than 1×10 2 M -1 s -1 , greater than 1×10 3 M -1 s -1 , greater than 1×10 4 M -1 s -1 , greater than 1×10 5 M-1 s -1 , greater than 1×10 6 M -1 s -1 In some embodiments, the kinetic binding rate (kon) is less than 1×10 5 M -1 s -1 , less than 1×10 6 M -1 s -1 or less than 1×10 7 M -1 s -1 .

[0244] Affinity can be derived from the quotient of the kinetic rate constants (KD = koff / kon). In some embodiments, KD is less than 1×10 -4 M, less than 1×10 -5 M, less than 1×10 -6 M, less than 1×10 -7 M, less than 1×10 -8 M, less than 1×10 - 9 M or less than 1×10 -10 In some embodiments, the KD is less than 50 nM, 30 nM, 20 nM, 15 nM, 10 nM, 9 nM, 8 nM, 7 nM, 6 nM, 5 nM, 4 nM, 3 nM, 2 nM, 1 nM, 0.9 nM, 0.8 nM, 0.7 nM, 0.6 nM, 0.5 nM, 0.4 nM, 0.3 nM, 0.2 nM, or 0.1 nM. In some embodiments, the KD is greater than 1×10 -4 M, greater than 1×10 -5 M, greater than 1×10 -6 M, greater than 1×10 -7 M, greater than 1×10 -8 M, greater than 1×10 -9 M, greater than 1×10 -10 M, greater than 1×10 -11 M or greater than 1×10 -12 M. In addition, Ka can be derived from KD by the formula Ka = 1 / KD.

[0245] Conventional techniques for measuring the affinity of an antibody for an antigen include, for example, ELISA, RIA, and surface plasmon resonance (SPR).

[0246] In some embodiments, the expression level of an antibody, antigen-binding fragment thereof, or antigen-binding protein construct described herein as determined using the methods described herein is at least or about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 50, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000, 20000, 50000, or 100000 mg / L. In some embodiments, the percentage of the multispecific antibody (e.g., bispecific antibody or trispecific antibody) formed as a percentage of the total protein levels as determined by size-exclusion chromatography described herein is at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, or at least 97%.

[0247] In some embodiments, the T cell-mediated cell killing EC50 of the antibodies, antigen-binding fragments thereof, or antigen-binding protein constructs described herein determined using the methods described herein is less than 0.1 nM, less than 0.2 nM, less than 0.5 nM, less than 0.8 nM, less than 1 nM, less than 2 nM, less than 5 nM, less than 8 nM, less than 10 nM, less than 20 nM, less than 50 nM, less than 100 nM, less than 200 nM, less than 500 nM, or less than 800 nM. In some embodiments, the T cell-mediated cell killing EC50 of the antibodies, antigen-binding fragments thereof, or antigen-binding protein constructs described herein, determined using the methods described herein, is greater than 0.1 nM, greater than 0.2 nM, greater than 0.5 nM, greater than 0.8 nM, greater than 1 nM, greater than 2 nM, greater than 5 nM, greater than 8 nM, greater than 10 nM, greater than 20 nM, greater than 50 nM, greater than 100 nM, greater than 200 nM, greater than 500 nM, or greater than 800 nM. In some embodiments, the T cell-mediated cell killing EC50 value of the antibody, antigen-binding fragment thereof, or antigen-binding protein construct described herein is 0.1-1nM, 0.2-2nM, 0.5-5nM, 0.8-8nM, 1-10nM, 2-20nM, 5-50nM, 8-80nM, 10-100nM, 20-200nM, 50-500nM, or 80-800nM. In some embodiments, the T cell-mediated cell killing EC50 value of the antibody, antigen-binding fragment thereof, or antigen-binding protein construct described herein is less than or about 50%, less than or about 30%, less than or about 20%, less than or about 10%, less than or about 5%, less than or about 1% compared to the isotype control antibody.

[0248] In some embodiments, the cellular binding EC50 of an antibody, antigen-binding fragment thereof, antigen-binding protein construct, or protein complex described herein, determined using the methods described herein, is less than or about 0.1 nm, 0.25 nM, 0.5 nM, 0.75 nM, 1 nM, 1.25 nM, 1.5 nM, 2 nM, 2.5 nM, 5 nM, 7.5 nM, 10 nM, or 20 nM.

[0249] In some embodiments, the melting temperature (Tm) value of the antibodies, antigen-binding fragments thereof, antigen-binding protein constructs, or protein complexes described herein, as determined using the methods described herein, is greater than 60°C, 60.5°C, 61°C, 61.5°C, 62°C, 62.5°C, 63°C, 63.5°C, 64°C, 65°C, 66°C, 67°C, 68°C, 69°C, 70°C, 71°C, 72°C, 73°C, or 74°C.

[0250] In some embodiments, the tumor growth inhibition percentage (TGI) of the antibodies, antigen-binding fragments thereof, antigen-binding protein constructs, or protein complexes described herein is TV In some embodiments, the antibody has an inhibition rate of less than 60%, 70%, 80%, 90%, 100%, 110%, 120%, 130%, 140%, 150%, 160%, 170%, 180%, 190%, or 200%. TGI can be measured, for example, on day 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, or 30 after the start of treatment, or on month 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 after the start of treatment. TV As used herein, tumor growth inhibition percentage (TGI TV %) is calculated according to the following formula: TGI TV (%) = [1-(Ti-T0) / (Vi-V0)]×100 (Ti is the average tumor volume of the treatment group on day i, T0 is the average tumor volume of the treatment group on day 0. Vi is the average tumor volume of the control group on day i, V0 is the average tumor volume of the control group on day 0).

[0251] In some embodiments, the antibodies, antigen-binding fragments thereof, antigen-binding protein constructs or protein complexes described herein have a serum stability of greater than 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% cell binding after storage in human serum for 1-10 days as determined using the methods described herein.

[0252] Thermal stability can also be determined. The Tm of the antibody, its antigen binding fragment or antigen binding protein construct (e.g., bispecific antibody) described herein can be greater than 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94 or 95°C. Since IgG can be described as a multi-domain protein, two transitions sometimes occur in the melting curve at the first denaturation temperature Tm1 and the second denaturation temperature Tm2. The appearance of these two peaks usually represents the denaturation of the Fc domain (Tm1) and the Fab domain (Tm2), respectively. When two peaks occur, Tm usually refers to Tm2. Thus, in some embodiments, the Tm1 of an antibody or antigen-binding fragment described herein is greater than 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, or 95°C. In some embodiments, the Tm2 of the antibodies or antigen binding fragments described herein is greater than 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, or 95°C. In some embodiments, Tm, Tm1, Tm2 is less than 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, or 95°C.

[0253] In some embodiments, the self-interaction signal of the antibody, its antigen binding fragment or antigen binding protein construct ranges from 0.037nm to 0.121nm. In some embodiments, the self-interaction signal of the antibody, its antigen binding fragment or antigen binding protein construct is less than 0.01nm, less than 0.02nm, less than 0.03nm, less than 0.04nm, less than 0.05nm, less than 0.06nm, less than 0.07nm, less than 0.08nm, less than 0.09nm, less than 0.1nm, less than 0.11nm, less than 0.12nm or less than 0.13nm. In some embodiments, the self-interaction signal of the antibody, antigen-binding fragment thereof, or antigen-binding protein construct is greater than 0.01 nm, greater than 0.02 nm, greater than 0.03 nm, greater than 0.04 nm, greater than 0.05 nm, greater than 0.06 nm, greater than 0.07 nm, greater than 0.08 nm, greater than 0.09 nm, greater than 0.1 nm, greater than 0.11 nm, greater than 0.12 nm, or greater than 0.13 nm. In some embodiments, the self-interaction signal of the antibody, antigen-binding fragment thereof, or antigen-binding protein construct is 0.01-0.05 nm, 0.02-0.06 nm, 0.03-0.07 nm, 0.04-0.08 nm, 0.05-0.09 nm, 0.06-0.1 nm, 0.07-1.1 nm, 0.08-1.2 nm, or 0.09-1.3 nm.

[0254] In some embodiments, the antibody, its antigen binding fragment or antigen binding protein construct has a functional Fc region. In some embodiments, the effector function of the functional Fc region is antibody-dependent cell-mediated cytotoxicity (ADCC) and / or complement-dependent cytotoxicity (CDC). In some embodiments, the Fc region is human IgG1, human IgG2, human IgG3 or human IgG4.

[0255] In some embodiments, the antibody, its antigen binding fragment or antigen binding protein construct does not have a functional Fc region. For example, the antibody or antigen binding fragment is a Fab, Fab', F(ab')2 and Fv fragment. In some embodiments, the Fc region of the antibody, antigen binding fragment or antigen binding protein construct includes one or more mutations to reduce effector function. In some embodiments, the antibody, its antigen binding fragment or antigen binding protein construct described herein does not have antibody-dependent cell-mediated cytotoxicity (ADCC) or complement-dependent cytotoxicity (CDC).

[0256] In some embodiments, the antibody or its antigen-binding fragment is a humanized antibody.Humanization percentage refers to the percentage identity of heavy chain or light chain variable region sequence compared with the human antibody sequence in the International Immunogenetics Information System (IMGT) database. In some embodiments, the humanization percentage is greater than 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94% or 95%. The specific details of how to determine the humanization percentage and how to determine the highest hit (top hit) are known in the art, and description can be found in, for example, Jones, et al. "The INNs and outs ofantibody nonproprietary names." MAbs. Vol. 8. No. 1. Taylor & Francis, 2016, the full text of the article is incorporated herein by reference. A high percentage of humanization generally has many advantages, such as being safer in humans, more effective, more likely to be tolerated by human subjects, and / or less likely to have side effects.

[0257] In some embodiments, a multispecific antibody including a trispecific antibody described herein (e.g., a CD3 / CEA-CAM6 / EGFR trispecific antibody) has an asymmetric structure comprising: 2, 3, 4, 5 or 6 antigen binding sites. In some embodiments, the multispecific antibody described herein comprises 2, 3, 4, 5 or 6 antigen binding sites targeting CEA-CAM6 (e.g., antigen binding scFv domains, Fab or VHH). In some embodiments, the CEA-CAM6 binding Fab domains comprise the same variable domain sequence. In some embodiments, the CEA-CAM6 binding Fab domains comprise different variable domain sequences.

[0258] The present disclosure also provides an antibody or an antigen-binding fragment thereof, which cross-competes with any antibody or antigen-binding fragment described herein. Cross-competition assays are known in the art and are described in, for example, Moore et al., "Antibody cross-competition analysis of the human immunodeficiency virustype 1gp120 exterior envelope glycoprotein." Journal of Virology 70.3 (1996): 1863-1872, which is incorporated herein by reference in its entirety. On the one hand, the present disclosure also provides an antibody or an antigen-binding fragment thereof, which binds to the same epitope or region as any antibody or antigen-binding fragment described herein. Epitope binning assays are known in the art and are described in, for example, Estep et al. "High throughput solution-based measurement of antibody-antigen affinity and epitope binning." MAbs. Vol. 5. No. 2. Taylor & Francis, 2013, which is incorporated herein by reference in its entirety.

[0259] Antibodies and antigen-binding fragments

[0260] In general, antibodies (also known as immunoglobulins) consist of two types of polypeptide chains (light chains and heavy chains). A non-limiting antibody of the present disclosure may be a complete tetra-immunoglobulin chain antibody comprising two heavy chains and two light chains. The heavy chain of the antibody may be of any isotype, including IgM, IgG, IgE, IgA or IgD, or a subisotype, including IgG1, IgG2, IgG2a, IgG2b, IgG3, IgG4, IgE1, IgE2, etc. The light chain may be a kappa (kappa) light chain or a lambda (lambda) light chain. The antibody may comprise two identical copies of a light chain and / or two identical copies of a heavy chain. Each heavy chain comprises a variable domain (or variable region, VH) and multiple constant domains (or constant regions), which are bound to each other by disulfide bonds within their constant domains to form the "stem" of the antibody. Each light chain contains a variable domain (or variable region, VL) and a constant domain (or constant region), and each light chain is bound to a heavy chain through a disulfide bond. The variable region of each light chain is aligned with the variable region of the heavy chain to which it is bound. The variable regions of both light and heavy chains contain three hypervariable regions sandwiched between more conserved framework regions (framework regions, FR).

[0261] These hypervariable regions are called complementary determining regions (CDRs) and form loops that contain the main antigen-binding surface of the antibody. The four framework regions mainly adopt a β-sheet conformation, and the CDRs form loops that connect the β-sheet structure and in some cases form part of the β-sheet structure. The CDRs on each chain are tightly connected by the framework regions and contribute to the formation of the antigen-binding region together with the CDRs from the other chain.

[0262] The method of determining the CDR region of an antibody by analyzing the amino acid sequence of the antibody is well known, and there are many commonly used definitions of CDR. Kabat's definition is based on sequence variability, while Chothia's definition is based on the position of the structural loop region. These methods and definitions are described in, for example, Martin, "Protein sequence and structure analysis of antibody variable domains," Antibody engineering, Springer Berlin Heidelberg, 2001. 422-439; Abhinandan, et al. "Analysis and improvements to Kabat and structurally correct numbering of antibody variable domains," Molecular immunology 45.14 (2008): 3832-3839; Wu, TT and Kabat, EA (1970) J. Exp. Med. 132: 211-250; Martin et al., Methods Enzymol. 203: 121-53 (1991); Morea et al., Biophys Chem. 68 (1-3): 9-16 (Oct. 1997); Morea et al., J Mol Biol. 275 (2): 269-94 (Jan. 1998); Chothia et al. al., Nature 342(6252):877-83(Dec.1989); Ponomarenko and Bourne, BMC Structural Biology 7:64(2007); Kontermann, R., & Dübel, S. (Eds.). (2010). Antibody engineering: Volume 2. Springer; each of which is incorporated herein by reference in its entirety. In some embodiments, the CDR is based on the Kabat definition. In some embodiments, the CDR is based on the Chothia definition. In some embodiments, the CDR is the longest CDR sequence determined by the Kabat, Chothia, ABM, IMGT or contact definitions.

[0263] CDR is very important for identifying antigenic epitopes. As used herein, "epitope" is the smallest part of a target molecule that can be specifically bound by the antigen binding domain of an antibody. The minimum size of an epitope can be about 3, 4, 5, 6 or 7 amino acids, but these amino acids are not necessarily within a continuous linear sequence of the primary structure of the antigen, because the epitope may depend on the three-dimensional configuration of the antigen based on the secondary and tertiary structures of the antigen.

[0264] In some embodiments, the antibody or antigen binding protein may include a complete immunoglobulin molecule (e.g., IgG1, IgG2a, IgG2b, IgG3, IgM, IgD, IgE, IgA) or a fragment thereof. The IgG subclasses (IgG1, IgG2, IgG3, and IgG4) are highly conserved, with differences in their constant regions, particularly in their hinge and upper CH2 domains. The sequences and differences of the IgG subclasses are known in the art and are described in, for example, Vidarsson, et al, "IgG subclasses and allotypes: from structure to effector functions." Frontiers in immunology 5 (2014); Irani, et al. "Molecular properties of human IgG subclasses and their implications for designing therapeutic monoclonal antibodies against infectious diseases." Molecular immunology 67.2 (2015): 171-182; Shakib, Farouk, ed. The human IgG subclasses: molecular analysis of structure, function and regulation. Elsevier, 2016; each of which is incorporated herein by reference in its entirety.

[0265] The antibody or antigen-binding protein can also be an immunoglobulin molecule derived from any species (e.g., humans, rodents, mice, rats, camelids). Antibodies described herein also include, but are not limited to, polyclonal, monoclonal, monospecific, multispecific antibodies and chimeric antibodies, which contain an immunoglobulin binding domain fused to another polypeptide. The term "antigen binding domain" or "antigen binding fragment" is an antibody portion that retains the specific binding activity of a complete antibody, i.e., any portion of an antibody that can specifically bind to an epitope on a complete antibody target molecule. It includes, for example, Fab, Fab', F(ab')2 and variants of these fragments. Therefore, in some embodiments, an antibody or its antigen-binding fragment can be, for example, scFv, Fv, Fd, dAb, bispecific antibodies, trispecific antibodies, bispecific scFv, biantibodies, linear antibodies, single-chain antibody molecules, multispecific antibodies formed by antibody fragments, and any polypeptides containing binding domains that are or are homologous to antibody binding domains. Non-limiting examples of antigen binding domains include, e.g., heavy and / or light chain CDRs of an intact antibody, heavy and / or light chain variable regions of an intact antibody, the full length heavy or light chain of an intact antibody, or a single CDR from a heavy or light chain of an intact antibody.

[0266] In some embodiments, the antibody or antigen-binding fragment thereof can bind to two different antigens or two different epitopes. In some embodiments, the antibody or antigen-binding fragment thereof can bind to three different antigens or three different epitopes.

[0267] The Fab fragment comprises the variable and constant domains of the light chain, and the variable domain and the first constant domain (CH1) of the heavy chain. The F(ab')2 antibody fragment comprises a pair of Fab fragments, which are usually covalently linked near their carboxyl termini by a hinge cysteine ​​between them. Other chemical couplings of antibody fragments are also known in the art.

[0268] In some embodiments, the Fc region can be further modified to increase or decrease effector function and serum half-life.

[0269] Any antibody, antigen-binding fragment thereof, or antigen-binding protein described herein can be conjugated to a stabilizing molecule (e.g., a molecule that increases the half-life of the antibody or its antigen-binding fragment in a subject or solution). Non-limiting examples of stabilizing molecules include: polymers (e.g., polyethylene glycol) or proteins (e.g., serum albumin, such as human serum albumin). Conjugation of stabilizing molecules can increase the half-life of the antibody, its antigen-binding fragment, or antigen-binding protein or extend its in vitro (e.g., in tissue culture or when stored as a pharmaceutical composition) or in vivo (e.g., in humans) biological activity.

[0270] In some embodiments, the antibodies, antigen-binding fragments thereof, or antigen-binding proteins (e.g., multispecific antibodies) described herein may be conjugated to a therapeutic agent. Antibody-drug conjugates comprising the antibodies, antigen-binding fragments thereof, or antigen-binding proteins may be covalently or non-covalently bound to a therapeutic agent. In some embodiments, the therapeutic agent is a cytotoxic or cytostatic agent (e.g., cytochalasin B, gramicidin D, ethidium bromide, emetine, mitomycin, etoposide, teniposide, vincristine, vinblastine, colchicine, doxorubicin, daunorubicin, dihydroxy anthracin, maytansine such as DM-1 and DM-4, ​​diketones, mitoxantrone, milamimycin, actinomycin D, 1-dehydrotestosterone, glucocorticoids, procaine, tetracaine, lidocaine, propranolol, puromycin, epirubicin, and cyclophosphamide and the like).

[0271] In some embodiments, the binding affinity of the multispecific antibody or antigen-binding fragment thereof described herein (e.g., CD3 / CEA-CAM6 / EGFR multispecific antibody) to CD3 (e.g., human CD3) is about 50%, about 60%, about 70%, about 80%, about 90%, about 100%, about 110%, about 120%, about 130%, about 140%, about 150% or about 200% of that of an antibody (e.g., anti-CD3 antibody) comprising the same antigen-binding region (e.g., Fab, scFv or VHH) of the multispecific antibody.

[0272] In some embodiments, the binding affinity of the multispecific antibody or antigen-binding fragment thereof described herein (e.g., CD3 / CEA-CAM6 / EGFR multispecific antibody) to CEA-CAM6 is about 50%, about 60%, about 70%, about 80%, about 90%, about 100%, about 110%, about 120%, about 130%, about 140%, about 150% or about 200% of that of an antibody (e.g., anti-CEA-CAM6 antibody) comprising the same antigen-binding region (e.g., Fab, scFv or VHH) of the multispecific antibody.

[0273] Recombinant vector

[0274] The present disclosure also provides recombinant vectors (e.g., expression vectors) comprising the isolated polynucleotides described herein (e.g., polynucleotides encoding the polypeptides described herein), host cells into which the recombinant vectors are introduced (i.e., so that the host cells contain the polynucleotides and / or vectors comprising the polynucleotides), and recombinant antibody polypeptides or fragments thereof or antigen-binding protein constructs produced by recombinant technology.

[0275] As used herein, "vector" is any construct that can deliver one or more polynucleotides of interest to the host cell when the vector is introduced into the host cell. "Expression vector" can deliver and express the one or more polynucleotides of interest in the form of a coded polypeptide in the host cell introduced into the expression vector. Therefore, in the expression vector, the polynucleotide of interest is operably connected to be positioned in the vector for expression by a regulatory element (e.g., a promoter, an enhancer and / or a poly-A tail), either in the vector, or in the host cell genome near or flanking the integration site of the polynucleotide of interest, so that the polynucleotide of interest will be translated in the host cell introduced into the expression vector.

[0276] The vector can be introduced into the host cell by methods known in the art, such as electroporation, chemical transfection (e.g., DEAE-dextran), transformation, transfection, infection and / or transduction (e.g., by recombinant virus). Therefore, non-limiting examples of vectors include viral vectors (which can be used to generate recombinant viruses), naked DNA or RNA, plasmids, cosmids, phage vectors, and DNA or RNA expression vectors associated with cationic condensing agents.

[0277] In some embodiments, a polynucleotide described herein (e.g., a polynucleotide encoding a polypeptide described herein) is introduced using a viral expression system (e.g., vaccinia virus or other poxvirus, retrovirus, or adenovirus), which may involve the use of non-pathogenic (defective), replication-competent viruses, or replication-defective viruses. In the latter case, viral propagation generally occurs only in complementing virus packaging cells.

[0278] For ease of expression, the DNA insert comprising the antibody encoding or polypeptide encoding polynucleotide described herein can be operably linked to a suitable promoter (e.g., a heterologous promoter), such as the bacteriophage λPL promoter, E. coli (E. coli) lac, trp and tac promoters, SV40 early and late promoters, and promoters of retroviral LTR. Other suitable promoters are known to those skilled in the art. The expression construct may further include a transcription initiation site, a termination site, and a ribosome binding site for translation in the transcribed region. The coding portion of the mature transcript expressed by the construct may include a translation initiation site at the start and a termination codon (UAA, UGA or UAG) appropriately located at the end of the polypeptide to be translated.

[0279] Representative examples of suitable hosts include, but are not limited to, bacterial cells, such as Escherichia coli, Streptomyces, and Salmonella typhimurium cells; fungal cells, such as yeast cells; insect cells, such as Drosophila S2 and Spodoptera Sf9 cells; animal cells, such as CHO, COS, Bowes melanoma, and HEK293 cells; and plant cells. Suitable culture media and conditions for the host cells described herein are known in the art.

[0280] The construct can be introduced into the host cell by calcium phosphate transfection, DEAE-dextran mediated transfection, cationic lipid mediated transfection, electroporation, transduction, infection or other methods. These methods are described in many standard laboratory manuals, such as Davis et al., Basic Methods In Molecular Biology (1986), which is incorporated herein by reference in its entirety.

[0281] In order to secrete the translated protein into the lumen of the endoplasmic reticulum, the periplasmic space or the extracellular environment, appropriate secretion signals may be incorporated into the expressed polypeptide. The signals may be endogenous signals of the polypeptide, or they may be heterologous signals.

[0282] The polypeptide (e.g., antibody) can be expressed in a modified form, such as a fusion protein (e.g., GST-fusion) or carrying a histidine tag, and can include not only a secretion signal, but also an additional heterologous functional region. For example, an additional amino acid region, particularly a charged amino acid, can be added to the N-terminus of the polypeptide to improve its stability and durability in a host cell, in a purification process, or in subsequent treatment and storage. Peptide moieties can also be added to the polypeptide to facilitate purification. Before the final preparation of the polypeptide, these regions can be removed. It is all familiar and conventional technology in the art to add peptide moieties to produce secretion or excretion, improve stability, and facilitate purification, etc. in the polypeptide.

[0283] The present disclosure also provides nucleic acid sequences having at least 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% identity to any of the nucleotide sequences described herein. The invention relates to amino acid sequences, and amino acid sequences that are at least 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% identical to any of the amino acid sequences described herein.

[0284] The present disclosure also provides nucleic acids having at least 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% homology to any of the nucleotide sequences described herein. The invention relates to amino acid sequences, and amino acid sequences having at least 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% homology to any amino acid sequence described herein.

[0285] In some embodiments, the disclosure relates to a nucleotide sequence encoding any polypeptide described herein, or any amino acid sequence encoded by any nucleotide sequence described herein. In some embodiments, the nucleic acid sequence is less than 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 150, 200, 250, 300, 350, 400, 500, or 600 nucleotides. In some embodiments, the amino acid sequence is less than 5, 6, 7, 8, 9, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200, 250, 300, 350, or 400 amino acid residues.

[0286] In some embodiments, the amino acid sequence (i) comprises an amino acid sequence; or (ii) consists of an amino acid sequence, wherein the amino acid sequence is any one of the sequences described herein.

[0287] In some embodiments, the nucleic acid sequence (i) comprises a nucleic acid sequence; or (ii) consists of a nucleic acid sequence, wherein the nucleic acid sequence is any sequence described herein.

[0288] Sequence homology (e.g., amino acid sequence homology or nucleic acid homology) percentage can also be determined. How to determine the sequence homology percentage is known in the art. In some embodiments, amino acid residues with similar physicochemical properties (percent homology), such as leucine and isoleucine, can be used to measure sequence similarity. Families of amino acid residues with similar physicochemical properties have been defined in the art. These families include, for example, amino acids with basic side chains (e.g., lysine, arginine, histidine), amino acids with acidic side chains (e.g., aspartic acid, glutamic acid), amino acids with uncharged polar side chains (e.g., glycine, asparagine, glutamine, serine, threonine, tyrosine, cysteine), amino acids with non-polar side chains (e.g., alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine, tryptophan), amino acids with β-branched side chains (e.g., threonine, valine, isoleucine) and amino acids with aromatic side chains (e.g., tyrosine, phenylalanine, tryptophan, histidine). In many cases, the percent homology is higher than the percent identity.

[0289] Method for making antibody or antigen binding protein construct

[0290] Isolated human protein fragments (e.g., CD3 or CEA-CAM6) can be used as immunogens to produce antibodies using standard polyclonal and monoclonal antibody preparation techniques. Polyclonal antibodies can be produced in animals by multiple injections (e.g., subcutaneous or intraperitoneal injections) of antigenic peptides or proteins. In some embodiments, the antigenic peptides or proteins are injected together with at least one adjuvant. In some embodiments, the antigenic peptides or proteins can be conjugated with an immunogenic preparation in the species to be immunized. Animals can be injected with more than one antigenic peptide or protein (e.g., twice, three times, or four times).

[0291] Full-length polypeptides or proteins can be used, and alternatively, antigenic peptide fragments thereof can be used as immunogens. The antigenic peptide of a protein comprises at least 8 (e.g., at least 10, 15, 20, or 30) amino acid residues in the amino acid sequence of the protein, and covers the epitope of the protein, so that antibodies produced against the peptide form specific immune complexes with the protein.

[0292] Immunogens are typically used to prepare antibodies by immunizing a suitable subject (e.g., a human or transgenic animal expressing at least one human immunoglobulin site). Suitable immunogen preparations may include, for example, recombinantly expressed or chemically synthesized polypeptides. The preparations may further include an adjuvant, such as Freund's complete or incomplete adjuvant, or a similar immunostimulant.

[0293] Polyclonal antibodies can be prepared as described above by using a polypeptide or its antigenic peptide (e.g., a portion of a protein) as an immunogen to immunize a suitable subject. Antibody titers in the immunized subject can be monitored over time by standard techniques, such as an enzyme-linked immunosorbent assay (ELISA) using an immobilized polypeptide or peptide. If desired, antibody molecules can be separated from mammals (e.g., from blood) and further purified by known techniques, such as protein A by protein G chromatography, to obtain IgG fractions. At an appropriate time after immunization, for example, when specific antibody titers are highest, antibody-producing cells can be obtained from the subject and monoclonal antibodies can be prepared by standard techniques, such as the hybridoma technique originally described by Kohler et al. (Nature 256: 495-497, 1975), human B cell hybridoma technique (Kozbor et al., Immunol. Today 4: 72, 1983), EBV hybridoma technique (Cole et al., Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, Inc., pp. 77-96, 1985) or trioma technique. Techniques for producing hybridomas are well known (see generally Current Protocols in Immunology, 1994, Coligan et al. (Eds.), John Wiley & Sons, Inc., New York, NY). Hybridoma cells producing monoclonal antibodies are detected by screening the hybridoma culture supernatants for antibodies that bind to the polypeptide or epitope of interest, for example, using a standard ELISA assay.

[0294] VHH can also be obtained from an original or designed synthetic llama VHH library. PBMC can be obtained from a llama, and RNA can be isolated to generate cDNA by reverse transcription. Then, the VHH gene is amplified by PCR and cloned into a phage display vector to construct an original VHH library. Synthetic (e.g., humanized) VHH libraries can be prepared by combining the recombinant (shuffled) VHH CDR 1, 2, and 3 produced by overlapping PCR with modified human VH framework regions to enhance diversity and maintain low immunogenicity. Then, the VHH library can be panned for antigens to obtain VHH with desired binding affinity.

[0295] Variants of the antibodies, antigen-binding fragments or antigen-binding protein constructs described herein can be prepared by introducing appropriate nucleotide changes into DNA or by peptide synthesis, wherein the DNA encodes a human antibody, a humanized antibody or a chimeric antibody or its antigen-binding fragment described herein. Such variants include, for example, residue deletions, insertions or substitutions in the amino acid sequence constituting the antigen-binding domain or the antibody antigen-binding site. In this group of variants, the affinity of some antibodies or antigen-binding fragments for the target protein will be improved. Deletions, insertions and / or combinations can be combined in any manner to obtain antibodies or antigen-binding fragments thereof with improved binding affinity for the target. The amino acid changes introduced in the antibody or antigen-binding fragment can also change or introduce new post-translational modifications in the antibody or antigen-binding fragment, such as changing (e.g., increasing or decreasing) the number of glycosylation sites, changing the type of glycosylation sites (e.g., changing the amino acid sequence so that the enzyme present in the cell is connected to different sugars) or introducing new glycosylation sites.

[0296] The antibodies disclosed herein can be derived from any species of animal, including mammals. Non-limiting examples of natural antibodies include antibodies derived from humans, primates (e.g., monkeys and apes), cattle, pigs, horses, sheep, camelids (e.g., camels and llamas), chickens, goats, and rodents (e.g., rats, mice, hamsters, and rabbits), including transgenic rodents that have been genetically engineered to produce human antibodies.

[0297] Human antibodies and humanized antibodies include antibodies with variable and constant regions derived from human germline immunoglobulin sequences (or with the same amino acid sequence as that derived from human germline immunoglobulin sequences). Human antibodies may include amino acid residues not encoded by human germline immunoglobulin sequences (e.g., mutations introduced by random or site-specific mutagenesis in vitro or somatic mutations in vivo), such as in CDRs.

[0298] Humanized antibodies generally have a human framework (FR) spliced ​​with non-human CDRs. Therefore, humanized antibodies have one or more amino acid sequences introduced from non-human sources. These non-human amino acid residues are generally referred to as "import" residues, which are generally taken from the "import" variable domain. Humanization can be achieved by, for example, replacing the corresponding sequence of a rodent CDR or CDR sequence with a human antibody. These methods are described in Jones et al., Nature, 321: 522-525 (1986); Riechmann et al., Nature, 332: 323-327 (1988); Verhoeyen et al., Science, 239: 1534-1536 (1988); each of which is incorporated herein by reference in its entirety. Therefore, a "humanized" antibody is a chimeric antibody in which a substantially incomplete human V domain is replaced with a corresponding sequence of a non-human species. In practice, humanized antibodies are typically mouse antibodies in which some CDR residues and some FR residues are substituted by residues from analogous sites in human antibodies.

[0299] More importantly, the antibody is humanized while retaining high specificity and affinity to antigen and other favorable biological properties. In order to achieve this goal, humanized antibodies can be prepared by using the three-dimensional model of the parent and humanized sequence to analyze the process of parent sequence and multiple conceptual humanized products. The three-dimensional immunoglobulin model is usually available and is familiar to those skilled in the art. Computer programs are available, which illustrate and display the possible three-dimensional conformational structure of the selected candidate immunoglobulin sequence. The inspection of these displays allows the analysis of the possible role of residues in the function of the candidate immunoglobulin sequence, that is, the analysis affects the residues of the ability of the candidate immunoglobulin to bind its antigen. In this way, FR residues can be selected and combined from receptors and input sequences to achieve desired antibody characteristics, such as affinity to the target antigen. The identity or homology with the original sequence generally refers to comparing the sequence and introducing a gap to reach the maximum percentage of sequence identity when necessary, and without considering any conservative substitution as a part of sequence identity, the percentage of the amino acid residues present in the candidate sequence is the same as the sequence present in human antibodies, humanized antibodies or chimeric antibodies or fragments.

[0300] In some embodiments, the antibody or its antigen-binding fragment can be covalently modified. These covalent modifications can be performed by chemical or enzymatic synthesis, or by enzymatic or chemical cleavage. Other types of covalent modifications of antibodies or antibody fragments are introduced into the molecule by reacting the targeted amino acid residues of the antibody or fragment with an organic derivatizing agent that is capable of reacting with selected side chains or N- or C-terminal residues.

[0301] In some embodiments, antibody variants having a carbohydrate structure lacking fucose connection (directly or indirectly) to the Fc region are provided. For example, the amount of fucose in such antibodies may be 1% to 80%, 1% to 65%, 5% to 65%, or 20% to 40%. The amount of fucose is determined by calculating the average amount of fucose in the sugar chain at Asn29 relative to the sum of all sugar structures (e.g., complex, hybrid, and high mannose structures) connected to Asn297, as measured by MALDI-TOF mass spectrometry as described in WO2008 / 077546. Asn297 refers to an asparagine residue located at approximately position 297 in the Fc region (EU numbering of Fc region residues; or position 314 in Kabat numbering); however, due to minor sequence variations in antibodies, Asn297 may also be located approximately ±3 amino acids upstream or downstream of position 297, i.e., between positions 294 and 300. Such fucosylated variants may have improved ADCC function. In some embodiments, to reduce glycan heterogeneity, the Fc region of the antibody can be further engineered to replace the asparagine at position 297 with alanine (N297A).

[0302] In some embodiments, in order to promote production efficiency by avoiding Fab arm exchange, the Fc region of the antibody is further engineered to replace the serine at position 228 (EU numbering) of IgG4 with proline (S228P). The S228 mutation is described in detail in "The S228P mutation prevents in vivo and in vitro IgG4 Fab-arm exchange as demonstrated using a combination of novel quantitative immunoassays and physiological matrix preparation." Journal of Biological Chemistry 290.9 (2015): 5462-5469 by Silva et al., which is incorporated herein by reference in its entirety.

[0303] In some embodiments, the methods described herein are designed to prepare bispecific antibodies. In some embodiments, the methods described herein are designed to prepare trispecific antibodies. Bispecific or trispecific antibodies can be prepared by engineering the interface between different antibody molecules to maximize the percentage of heterodimers recovered from recombinant cell culture. For example, the interface can comprise at least a portion of the CH3 domain of an antibody constant domain. In this method, one or more small amino acid side chains from the interface of a first antibody molecule are replaced with larger side chains (e.g., tyrosine or tryptophan). A compensatory "cavity" of the same or similar size to the large side chain can be created at the interface of a second antibody molecule by replacing the large amino acid side chain with a smaller amino acid side chain (e.g., alanine or threonine). This provides a mechanism for increasing heterodimer yield over other unwanted end products (e.g., homodimers). The method is described in detail, for example, in WO96 / 27011, which is incorporated herein by reference in its entirety.

[0304] In some embodiments, one or more amino acid residues in the CH3 portion of IgG are replaced. In some embodiments, one heavy chain has one or more of the following T366W substitutions. The other heavy chain can have one or more of the following T366S, L368A, and Y407V substitutions. In addition, substitutions can also be introduced into the hinge region of the two substituted IgGs (-ppcpScp-->-ppcpPcp-).

[0305] In addition, anion exchange chromatography can be used to purify the antibodies described herein. Anion exchange chromatography is a method of separating substances according to their charge using an ion exchange resin containing a positively charged group such as diethylaminoethyl (DEAE). In solution, the resin is coated with positively charged counterions (cations). The anion exchange resin will bind to negatively charged molecules, displacing the counterions. Anion exchange chromatography can be used to purify proteins based on their isoelectric point (pI). The isoelectric point is defined as the pH at which a protein has no net charge. When pH > pI, the protein has a net negative charge, and when pH < pI, the protein has a net positive charge. Thus, in some embodiments, different amino acid substitutions can be introduced into the two heavy chains such that the pI of the homodimer containing two arm A's and the pI of the homodimer containing two arm B's are different. The pI of the bispecific or trispecific antibody having arm A and arm B will be somewhere between the two pI's of the homodimers. Thus, the two homodimers and the bispecific or trispecific antibody can be eluted under different pH conditions. The present disclosure shows that some amino acid residue substitutions can be introduced into the heavy chain to modulate the pI.

[0306] Therapeutic methods

[0307] Methods described herein include methods for treating cancer-related diseases. In general, these methods include administering a therapeutically effective amount of an engineered multispecific antibody (e.g., a bispecific antibody or a trispecific antibody) or antigen-binding protein construct described herein to a subject in need or determined to be in need of such treatment.

[0308] As used herein, "treatment" refers to improving at least one symptom of a cancer-related disease. Generally speaking, cancer leads to death; therefore, treatment can lead to an increase in life expectancy (e.g., at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 months, or at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 years). Administering a therapeutically effective amount of a formulation described herein (e.g., an antigen binding protein construct) to treat a cancer-related condition will result in a reduction in the number of cancer cells and / or relief of symptoms.

[0309] As used herein, the term "cancer" refers to cells with autonomous growth capacity, that is, an abnormal state or condition characterized by rapidly proliferating cell growth. The term is intended to include all types of cancerous growth or carcinogenic processes, metastatic tissues or malignantly transformed cells, tissues or organs, regardless of histopathological type or invasion stage. The term "tumor" used herein refers to cancerous cells, for example, a mass of cancer cells. Cancers that can be treated or diagnosed using the methods described herein include malignant tumors of various organ systems, such as malignant tumors affecting the lungs, breasts, thyroids, lymphs, gastrointestinal tracts and urogenital tracts, and adenocarcinomas, which include malignant tumors, such as most colon cancers, renal cell carcinomas, prostate cancers and / or testicular tumors, non-small cell carcinomas of the lungs, small intestinal cancers and esophageal cancers. In some embodiments, the agents described herein are designed to treat or diagnose cancer in a subject. The term "cancer" is recognized in the art and refers to malignant tumors of epithelial or endocrine tissues, including respiratory cancer, gastrointestinal cancer, urogenital cancer, testicular cancer, breast cancer, prostate cancer, endocrine system cancer and melanoma. In some embodiments, the cancer is renal cancer or melanoma. Exemplary cancers include those formed by tissues of the cervix, lung, prostate, breast, head and neck, colon, and ovary. The term also includes carcinomatous sarcomas, for example, which include malignant tumors composed of carcinomatous and sarcomatous tissues. "Adenocarcinoma" refers to a cancer derived from glandular tissue or in which tumor cells form a recognizable glandular structure. The term "sarcoma" is recognized in the art and refers to a malignant tumor of mesenchymal origin.

[0310] In one aspect, the disclosure also provides methods for treating cancer in a subject, methods for reducing the rate at which tumor volume increases over time in a subject, methods for reducing the risk of metastasis, or methods for reducing the risk of additional metastasis in a subject. In some embodiments, treatment can stop, slow, delay, or inhibit the progression of cancer. In some embodiments, treatment can result in a reduction in the number, severity, and / or duration of one or more symptoms of cancer in a subject.

[0311] In one aspect, the disclosure features methods comprising administering a therapeutically effective amount of an antibody or antigen-binding fragment thereof, antigen-binding protein construct, or antibody drug conjugate disclosed herein to a subject in need thereof, e.g., a subject having, or identified or diagnosed as having, a cancer, e.g., breast cancer (e.g., triple-negative breast cancer), carcinoid, cervical cancer, endometrial cancer, glioma, head and neck cancer, liver cancer, lung cancer, small cell lung cancer, lymphoma, melanoma, ovarian cancer, pancreatic cancer, prostate cancer, renal cancer, colorectal cancer, gastric cancer, testicular cancer, thyroid cancer, bladder cancer, urethral cancer, or a hematological malignancy.

[0312] As used herein, the terms "subject" and "patient" are used interchangeably throughout the description and describe an animal (human or non-human) to which treatment is provided according to the methods of the present invention. Veterinary and non-veterinary applications are contemplated by the present invention. Human patients can be adults or young people (e.g., people under the age of 18). In addition to humans, patients include, but are not limited to, mice, rats, hamsters, guinea pigs, rabbits, ferrets, cats, dogs, and primates. Including, for example, non-human primates (e.g., monkeys, chimpanzees, gorillas, etc.), rodents (e.g., rats, mice, gerbils, hamsters, ferrets, rabbits), lagomorphs, pigs (e.g., pigs, miniature pigs), horses, dogs, cats, cattle, and other domestic, farm, and zoo animals.

[0313] In some embodiments, the cancer is a CEA-CAM6 expressing cancer.

[0314] In some embodiments, the cancer is lung cancer, colorectal cancer, head and neck cancer, gastric cancer, pancreatic cancer, urethral cancer, breast cancer, cervical cancer, or endometrial cancer.

[0315] In some embodiments, the cancer cells described herein are cell lines, such as H1395 cells. In some embodiments, the CEA-CAM6 level of the cancer cells is elevated, such as at least 10%, at least 20%, at least 30%, at least 40%, at least 50% higher than that of non-cancerous cells.

[0316] In some embodiments, the compositions and methods described herein can be used to treat patients at risk for cancer.Cancer patients can be identified by a variety of methods known in the art.

[0317] As used herein, an "effective amount" refers to an amount or dosage sufficient to achieve a beneficial or desired result, including stopping, slowing, delaying or inhibiting the progression of a disease, e.g., cancer. The effective amount will depend on, e.g., the age and weight of the subject to whom the antibody, antigen binding fragment, antigen binding protein construct, antibody-drug conjugate, polynucleotide encoding the antibody, vector comprising the polynucleotide, and / or composition thereof is to be administered, the severity of the symptoms, and the route of administration, and thus administration can be determined on an individual basis.

[0318] The effective amount can be applied in one or more administrations. For example, the effective amount of an antibody, an antigen binding fragment, an antigen binding protein construct, or an antibody-drug conjugate is enough to improve, stop, stabilize, reverse, inhibit, slow down and / or delay the amount of cancer progression in a patient, or is enough to improve, stop, stabilize, reverse, slow down and / or delay the amount of a cell (e.g., a biopsy cell, any cancer cell described herein, or a cell line (e.g., a cancer cell line)) in vitro proliferation. As understood in the art, the effective amount can vary, particularly depending on the patient's medical history and other factors, such as the type (and / or dosage) of the agent used.

[0319] Effective amounts and schedules for administering an antibody, antigen binding protein construct, antibody encoding polynucleotide, antibody-drug conjugate, and / or composition disclosed herein can be determined empirically, and making such determinations is within the skill in the art.

[0320] The typical dosage of an effective amount of an antibody or antigen-binding protein construct is 0.01 mg / kg to 100 mg / kg. In some embodiments, the dosage may be less than 100 mg / kg, 10 mg / kg, 9 mg / kg, 8 mg / kg, 7 mg / kg, 6 mg / kg, 5 mg / kg, 4 mg / kg, 3 mg / kg, 2 mg / kg, 1 mg / kg, 0.5 mg / kg or 0.1 mg / kg. In some embodiments, the dosage may be greater than 10 mg / kg, 9 mg / kg, 8 mg / kg, 7 mg / kg, 6 mg / kg, 5 mg / kg, 4 mg / kg, 3 mg / kg, 2 mg / kg, 1 mg / kg, 0.5 mg / kg, 0.1 mg / kg, 0.05 mg / kg or 0.01 mg / kg. In some embodiments, the dosage is about 10 mg / kg, 9 mg / kg, 8 mg / kg, 7 mg / kg, 6 mg / kg, 5 mg / kg, 4 mg / kg, 3 mg / kg, 2 mg / kg, 1 mg / kg, 0.9 mg / kg, 0.8 mg / kg, 0.7 mg / kg, 0.6 mg / kg, 0.5 mg / kg, 0.4 mg / kg, 0.3 mg / kg, 0.2 mg / kg, or 0.1 mg / kg.

[0321] In any of the methods described herein, the at least one antibody, antigen-binding fragment thereof, antigen-binding protein construct, antibody-drug conjugate, or pharmaceutical composition (e.g., any antibody, antigen-binding fragment, antigen-binding protein construct, antibody-drug conjugate, or pharmaceutical composition described herein) and optionally at least one additional therapeutic agent can be administered to the subject at least once a week (e.g., once a week, twice a week, three times a week, four times a week, once a day, twice a day, or three times a day).

[0322] In some embodiments, one or more additional therapeutic agents may be administered to a subject. The additional therapeutic agent may include one or more selected from B-Raf inhibitors, EGFR inhibitors, MEK inhibitors, ERK inhibitors, K-Ras inhibitors, c-Met inhibitors, anaplastic lymphoma kinase (ALK) inhibitors, phosphatidylinositol 3-kinase (PI3K) inhibitors, Akt inhibitors, mTOR inhibitors, dual PI3K / mTOR inhibitors, Bruton's tyrosine kinase (BTK) inhibitors and isocitrate dehydrogenase 1 (IDH1) and / or isocitrate dehydrogenase 2 (IDH2) inhibitors. In some embodiments, the additional therapeutic agent is an inhibitor of indoleamine 2,3-dioxygenase-1 (IDO1) (e.g., icadolstat).

[0323] In some embodiments, the additional therapeutic agent may comprise one or more inhibitors selected from the group consisting of inhibitors of HER3, inhibitors of LSD1, inhibitors of MDM2, inhibitors of BCL2, inhibitors of CHK1, inhibitors of the activated hedgehog signaling pathway, and agents that selectively degrade estrogen receptors.

[0324] In some embodiments, the additional therapeutic agent may include one or more therapeutic agents selected from the group consisting of trabectedin, nab-paclitaxel, trabantib, pazopanib, cediranib, palbociclib, everolimus, fluoropyrimidine, IFL, regorafenib, Reolysin, Alimta, Zykadia, sunitinib, temsirolimus, axitinib, everolimus, sorafenib, Vbtrient, pazopanib, IMA-901, AGS-003, cabozantinib, vinflunine, Hsp90 inhibitors, Ad-GM-CSF, temozolomide, IL-2, IFNa, vinblastine, thalidomide, dacarbazine, cyclophosphamide, lenalidomide, azacytidine, lenalidomide, bortezomib, amrubicin, carfilzomib, pralatrexate, and enzalumim.

[0325] In some embodiments, the additional therapeutic agents may include one or more therapeutic agents selected from the following: adjuvants, TLR agonists, tumor necrosis factor (TNF) α, IL-1, HMGB1, IL-10 antagonists, IL-4 antagonists, IL-13 antagonists, IL-17 antagonists, HVEM antagonists, ICOS agonists, therapeutic agents targeting CX3CL1, therapeutic agents targeting CXCL9, therapeutic agents targeting CXCL10, therapeutic agents targeting CCL5, LFA-1 agonists, ICAM1 agonists and selectin agonists.

[0326] In some embodiments, the subject is administered carboplatin, nab-paclitaxel, paclitaxel, cisplatin, pemetrexed, gemcitabine, FOLFOX, or FOLFIRI.

[0327] In some embodiments, the additional therapeutic agent is an anti-OX40 antibody, an anti-PD-1 antibody, an anti-PD-L1 antibody, an anti-PD-L2 antibody, an anti-LAG-3 antibody, an anti-TIGIT antibody, an anti-BTLA antibody, an anti-CTLA-4 antibody, or an anti-GITR antibody.

[0328] Pharmaceutical compositions and routes of administration

[0329] The pharmaceutical composition is formulated to meet its intended route of administration (e.g., intravenous, intraarterial, intramuscular, intradermal, subcutaneous or intraperitoneal). The composition may include a sterile diluent (e.g., sterile water or saline), a fixed oil, polyethylene glycol, glycerol, propylene glycol or other synthetic solvents, antibacterial or antifungal agents, such as benzyl alcohol or methyl paraben, chlorobutanol, phenol, ascorbic acid, thimerosal, etc.; antioxidants, such as ascorbic acid or sodium bisulfite; chelating agents, such as ethylenediaminetetraacetic acid; buffers, such as acetates, citrates or phosphates; and isotonic agents, such as sugars (e.g., glucose), polyols (e.g., mannitol or sorbitol), or salts (e.g., sodium chloride) or any combination thereof. Liposomal suspensions may also be used as pharmaceutically acceptable carriers (see, e.g., U.S. Patent No. 4,522,811). The preparation of the composition may be formulated and packaged in ampoules, disposable syringes or multi-dose vials. Where necessary (e.g., in, for example, injectable formulations), suitable fluidity can be maintained, for example, by the use of a coating (e.g., lecithin) or a surfactant. The absorption of the antibody, its antigen-binding fragment, or antigen-binding protein construct can be extended by including agents that delay absorption (e.g., aluminum monostearate and gelatin). Alternatively, controlled release can be achieved by implants and microencapsulated delivery systems that can include biodegradable, biocompatible polymers (e.g., ethylene vinyl acetate, polyanhydrides, polyglycolic acid, collagen, polyorthoesters, and polylactic acid; Alza Corporation and Nova Pharmaceutical, Inc.).

[0330] Compositions containing one or more of any of the antibodies, antigen-binding fragments, antigen-binding protein constructs, antigen-binding proteins, antibody-drug conjugates described herein can be formulated for parenteral (e.g., intravenous, intraarterial, intramuscular, intradermal, subcutaneous or intraperitoneal) administration in dosage unit form (i.e., physically discrete units containing predetermined quantities of active compound to facilitate administration and uniformity of dosage).

[0331] The toxicity and therapeutic efficacy of the composition can be determined by standard pharmaceutical procedures in cell culture or experimental animals (e.g., monkeys). The LD50 (the dose lethal to 50% of the population) and the ED50 (the dose therapeutically effective in 50% of the population) can be determined: the therapeutic index is the ratio of LD50:ED50. Agents that exhibit high therapeutic indices are preferred. When an agent exhibits undesirable side effects, care should be taken to minimize potential damage (i.e., reduce unwanted side effects). Toxicity and therapeutic efficacy can be determined by other standard pharmaceutical procedures.

[0332] The data obtained from cell culture assays and animal studies can be used to formulate the appropriate dose of any given medicament for an object (e.g., a person). One or more (e.g., one, two, three or four) antibodies, antigen binding fragments thereof, or antigen binding protein constructs (e.g., any antibody, antibody fragment or antigen binding protein construct described herein) of a therapeutically effective amount will be a treatment object (e.g., a human subject identified as having cancer) or an object (e.g., previously having cancer but now cured) identified as being at risk of the disease (e.g., killing cancer cells), reducing the severity, frequency and / or duration of one or more symptoms of the disease in the object (e.g., a person). The effectiveness and dosage of any antibody, antigen binding fragment or antigen binding protein construct described herein can be determined by a health care professional or veterinary professional using methods known in the art and by observing one or more disease symptoms in an object (e.g., a person). Certain factors may affect the dosage and time (e.g., the severity of the disease or illness, previous treatment, the general health status of the object and / or the presence of age and other diseases) expected by the effective treatment object.

[0333] Exemplary dosages include milligram or microgram amounts of any antibody or antigen-binding fragment described herein, antigen-binding protein constructs, or antibody-drug conjugates per kilogram of subject body weight (e.g., about 1 μg / kg to about 500 mg / kg; about 100 μg / kg to about 500 mg / kg; about 100 μg / kg to about 50 mg / kg; about 10 μg / kg to about 5 mg / kg; about 10 μg / kg to about 0.5 mg / kg; or about 1 μg / kg to about 50 μg / kg). Although these dosages cover a wide range, it will be appreciated by those of ordinary skill in the art that therapeutic agents (including antibodies and antigen-binding fragments thereof) vary in potency, and effective amounts can be determined by methods known in the art. Typically, a relatively low ridge is first administered, and the attending health care professional or veterinary professional (in the case of therapeutic applications) or researcher (when still working in the development phase) can then gradually increase the dosage until an appropriate response is obtained. Furthermore, it should be understood that the specific dosage level for any particular subject will depend on a variety of factors, including the activity of the specific compound used, the age, weight, general health, sex and diet of the subject, the time of administration, the route of administration, the rate of excretion and the in vivo half-life of the antibody, antibody fragment, or antigen binding protein construct.

[0334] The pharmaceutical composition may be contained in a container, pack, or dispenser together with instructions for administration.The present disclosure also provides methods of making antibodies, antigen-binding fragments thereof, or antigen-binding protein constructs for the various uses as described herein.

[0335] Example

[0336] The present invention is further described by the following examples, which are not intended to limit the scope of the invention as described in the claims.

[0337] Materials and methods

[0338] Preparation of human peripheral blood mononuclear cells (PBMC)

[0339] PBMCs were prepared by density gradient centrifugation from blood samples obtained from blood banks or healthy human donors.

[0340] Blood from healthy donors was stored in anticoagulant tubes containing EDTA and allowed to stand for 10 minutes. An equal volume of 2% PBS (pH 7.4) was added and mixed thoroughly. 15 ml of Ficoll Paque PLUS low-density centrifugal solution (GE) was added to a 50 ml Falcon tube, 30 ml of diluted fresh blood was drawn, and the upper layer of the density gradient solution was slowly added along the side wall of the Falcon tube. The centrifuge was set to stop, the test tube was loaded into the centrifuge, and centrifuged at 1450 rpm for 45 minutes at room temperature. Gently remove the test tube and visually ensure that the blood in the centrifuge tube is divided into three layers (the upper layer is the serum layer, the middle white part is the peripheral blood mononuclear cells, and the bottom layer is the red blood cells). Carefully aspirate the upper yellow transparent serum layer, leaving only the middle layer, and transfer it to a new 15 ml centrifuge tube. Add an equal volume of gradient solution and centrifuge the diluted middle layer at 1450 rpm for 30 minutes at room temperature. Remove the upper suspension and retain the bottom cell pellet. Resuspend the cells and wash 3 times until the upper layer becomes clear. 1 ml of 2% FBS-PBS buffer was added to resuspend the cells. After counting the PBMCs, the cells were resuspended in RPMI1640 medium containing 10% FBS (GIBCO) and 1% L-glutamine (GIBCO), and cultured in a CO2 incubator at 37°C and 5% CO2.

[0341] DNA Sequencing and Synthesis

[0342] The DNA sequence was determined by double-stranded DNA sequencing, and the sequencing was performed by GENEWIZ (Suzhou Genewiz Biotechnology Co., Ltd.).

[0343] All genes involved in the experiments were synthesized by GenScript (Nanjing GenScript Biotechnology Co., Ltd.).

[0344] Example 1: Construction of CD3 / CEA / EGFR trispecific antibody

[0345] Based on the mouse CD3 antibody, humanized CD3 antibody TA1 was obtained through mutation, library construction, humanization transformation and multiple screening.

[0346] CEA and EGFR single domain antibodies (VHH) were obtained by immunizing llamas. Adult alpacas were subcutaneously immunized with 250 μg of recombinant human tumor antigen CEA (CEACAM-5 / CD66e, ACRO system, CE5-H5226) or human EGFR protein (ACRO system, EGR-H5222), once every month for a total of 5 times. PBMC cells were separated, RNA was extracted, and a phage antibody library was constructed. After the library was packaged to form phage particles, liquid phase method was used for panning. The phage was combined with biotinylated CEA antigen solution or EGFR antigen solution and then separated with streptavidin-labeled magnetic beads. Phages were screened using a variety of methods (such as ELISA and engineered cells overexpressing CEA or EGFR) to obtain CEA and EGFR single domain antibodies, which are referred to as CEA and EGFR.

[0347] Design of non-unilateral CD3 / CEA / EGFR trispecific antibody structure: 12 trispecific antibody structures (TA1-Fab CD3 monovalent, CEA monovalent and EGFR monovalent) were designed according to Figure 2. TA1-Fab is located at the N-terminus of Fc, and the three antigen binding domains (TA1-Fab, EGFR single domain antibody and CEA single domain antibody) are not located at the N-terminus or C-terminus of Fc at the same time. When the single domain antibody VHH is located at the C-terminus of Fc, GGGGSGGGS (SEQ ID NO: 11) serves as a linker connecting the two fragments. When the single domain antibody VHH is located at the N-terminus of Fc, the Fc hinge region (Hinge) (SEQ ID NO: 12) serves as a linker connecting the two fragments. At the same time, based on Figures 1A-1D Bispecific control antibodies (CEA-BsAb and EGFR-BsAb) containing TA1-Fab against CEA or EGFR were designed. "TA1F" represents the Fab region of the anti-CD3 antibody TA1. TA1 VH is linked to CH1-Hinge-Fc (IgG4), and TA1 VL is linked to CL (κ).

[0348] In some embodiments, the structure of the trispecific antibody is as follows Figure 2A-2L As shown. A double-chain asymmetric structure design containing Fc is adopted, which has 1 CEA antigen binding domain, 1 EGFR antigen binding domain and 1 CD3 antigen binding domain. CEA and EGFR are in the form of single domain antibodies (VHH), and CD3 is in the form of Fab. The Fc region gives the antibody a longer half-life and good stability. At the same time, the knob into hole (KIH) design of the two heavy chains greatly reduces the chance of mismatching and improves the yield and uniformity. The structures (forms) of specific bispecific antibodies and trispecific antibodies are shown as follows. Figures 1A-1D and Figure 2A-2L shown.

[0349] The two heavy chain IgG4 Fcs are knob-Fc and hole-Fc, respectively. The sequences of knob-Fc and hole-Fc are preferably those shown in the following table. The light chain is a universal kappa light chain, and CH1 is derived from IgG4.

[0350] Table 1 Universal sequences

[0351]

[0352] The variable region sequences of CD3 antibody TA1 and the heavy chain sequences of CEA and EGFR single domain antibodies are shown in the following table.

[0353] Table 2: Sequence listing of CD3 variable regions, CEA and EGFR single domain antibodies

[0354]

[0355] Plasmid construction: All genes involved in this application were synthesized by GenScript (Nanjing GenScript Biotechnology Co., Ltd.) and then ligated to the mammalian expression vector pEE12.4 by enzyme digestion. The variable regions of the heavy and light chain DNA sequences were cloned into pEE12.4 with a pre-inserted human IgG4 heavy chain constant region or a kappa light chain constant region in frame. The plasmids were extracted using the OMEGA plasmid extraction kit and stored at -80°C.

[0356] Example 2: Preparation of non-unilateral CD3 / CEA / EGFR trispecific antibodies

[0357] Expi 293F was transfected with the above plasmid using polyethyleneimine (PEI, POLYETHYLENEIMINE'MAX; polysciences, 24765-2). TM Cells (ThermoFisher) were transfected with the corresponding expression plasmids at a ratio of 1:2:1 [“heavy chain Fc (hole)”: “light chain”: “heavy chain Fc (knob)”].

[0358] Expi 293F TM Cells in CD OptiCHO TM The cells were suspended and cultured in the culture medium at 37°C (5% CO2, 135 rpm). One day before transfection, 293F cells were subcultured into 1L gas permeable conical flasks (Corning) at a seeding density of 1.0×10 6 cells / mL, the cell volume is 200mL. The cell density on the day of transfection is estimated to be 1.8-2.0×10 6cells / mL. The cell suspension was centrifuged at 1000 rpm for 5 minutes at room temperature and washed once with Expi293 medium. The cells were collected and resuspended in 200 mL of Expi293 medium. 400 μg of plasmid was diluted with 5 mL of Opti-MEM medium and vortexed for 15 seconds. 1.2 mg of PEI was diluted with 5 mL of Opti-MEM medium and vortexed for 15 seconds. The PEI-containing solution was added dropwise to the DNA-containing solution, mixed gently, and incubated at room temperature for 15 minutes. The plasmid / PEI mixture was added to the cell suspension and incubated in an incubator at 37°C, 5% CO2, and 85 rpm for 4 hours. After 4 hours, 200 mL of EX-CELLTM293 medium and 2 mM glutamine (Gibco) were added, the speed was adjusted to 135 rpm, and the culture was continued for 24 hours. After 24 hours, 3.8 mM VPA (a cell proliferation inhibitor) was added. After 72 hours, 40 mL of medium D was added. After 7 days, the cell suspension was centrifuged at 18,000 rpm for 30 minutes and the supernatant was collected. The supernatant was sterile filtered using a 0.22 μm filter and sodium azide was added to a final concentration of 0.01% w / v. The supernatant was stored at 4°C for purification.

[0359] The target protein was purified by protein A chromatography. The collected supernatant was loaded into a MabSelect Prism A FF (GE; 17-5498-01) balanced with 20 mL of equilibration buffer (25 mMTris, 150 mM NaCl, pH 7.5). Unbound protein was removed with at least 10 column volumes of wash buffer (25 mM Tris, 150 mM NaCl, pH 7.5). The target protein was eluted with 5 column volumes of elution buffer (20 mM sodium citrate, pH 3.5). The protein solution was neutralized by adding 1 / 10 volume of 1 M Tris, pH 9.5.

[0360] Using Zeba TM The target protein was transferred to the desired buffer using a desalting spin column (ThermoFisher) or ultrafiltration tube (Millipore). The purity and concentration of the protein were determined by SDS-PAGE electrophoresis (load: 2-3 μg) and NanoDrop2000.

[0361] The target protein was concentrated, filtered, and then loaded into a gel filtration column (Gel Filtration, HiLoad Superdex 200, GE) equilibrated with an equilibrium buffer (20 mM histidine, 140 mM sodium chloride, pH 6.0). The molecular weight, purity, and aggregation of the sample were analyzed at room temperature using an analytical buffer (PBS, pH 7.2) at a flow rate of 0.5 mL / min. The obtained high-purity monomeric protein was aliquoted and stored at -80°C.

[0362] In an alternative embodiment, protein A affinity chromatography (MabSelect SuRe, GE) is used to purify the protein from the cell supernatant. The protein eluate is then subjected to cation exchange chromatography (HiTrap SP HP, GE) and subsequently analyzed using a gel filtration column. The target protein purity obtained using this purification method is greater than 90%.

[0363] Protein structure design Figures 1A-1D and Figure 2A-2L The SDS-PAGE results are shown in Figure 3A-3B The yield of the third antibody was 13.73-22.83 mg / L, and the purity was >92%. See the table below for details.

[0364] Table 3: Yield and purity of non-unilateral CD3 / CEA / EGFR trispecific antibodies

[0365] sample Yield (mg / L) purity% 301 20.28 96.82 302 15.65 95.89 303 19.86 94.21 304 13.73 93.22 305 17.42 94.25 306 15.60 95.21 307 18.46 93.33 308 16.90 92.53 309 22.83 95.82 310 17.16 92.63 311 21.58 92.45 312 18.32 95.23

[0366] Example 3: Determination of affinity of non-unilateral CD3 / CEA / EGFR trispecific antibodies by biofilm interferometry (BLI)

[0367] Biomembrane interferometry (BLI) experiments were performed on an OCTECT RED96e (ForteBio) at 30°C using 0.02% PBST (10 mmol / L Na2HPO4; 1.75 mmol / L KH2PO4; 137 mmol / L NaCl; 2.65 mmol / L KCl; pH 7.2-7.4, 0.02% surfactant Tween 20) as running buffer.

[0368] The antibody was first captured onto an anti-human IgG Fc AHC biosensor (anti-hIgG Fc, Sartorius, 18-5060). The antibody was then diluted to 5 μg / mL and 1.5 nM protein was coupled to the AHC sensor surface in 0.02% PBST.

[0369] The analytes, recombinant human CD3ε / δ protein (ACRO system, CDD-H52W1), recombinant human tumor antigen CEA (CEACAM-5 / CD66e, ACRO system, CE5-H5226) and recombinant human EGFR (Bipsi system, EGR-5222) were diluted in 0.02% PBST to 200nM, 100nM, 50nM, 25nM, 12.5nM, 6.25nM and 3.125nM, respectively.

[0370] After the sensor captured the antibody, it was used to bind and dissociate with different concentrations of analyte to determine the KD value of the interaction. The experimental parameters were as follows: baseline 1: 60 seconds, loading: 240 seconds, baseline 2: 180 seconds, binding: 240 seconds, dissociation: 600 seconds, high sensitivity kinetics: 2Hz. The kinetic constants were derived according to the 1:1 Langmuir binding rate equation to obtain the equilibrium dissociation constant (KD). The binding and dissociation curves were fitted simultaneously using Data Analysis HT 12 software (Sartorius, 50-5029) to calculate K on (ka) and K off (kdis). The affinity KD is calculated according to the following formula: KD = K off / K on .

[0371] The experimental results show that the affinity of the non-unilateral trispecific antibody to CEA is 0.4nM-1.2nM, the affinity to EGFR is 5.8nM-26.7nM, and the affinity to CD3 is 0.5nM-5.9nM. The fitted affinity KD data are shown in the following table.

[0372] Table 4: Affinity of non-unilateral CD3 / CEA / EGFR trispecific antibodies for CEA (T = 30°C)

[0373]

[0374] Table 5: Affinity of non-unilateral CD3 / CEA / EGFR trispecific antibodies to EGFR (T = 30°C)

[0375]

[0376] Table 6: Affinity of non-unilateral CD3 / CEA / EGFR trispecific antibodies for CD3ε / δ (T=30°C)

[0377]

[0378]

[0379] Example 4: Non-unilateral CD3 / CEA / EGFR trispecific antibody induces T cell-mediated killing of EGFR-positive tumor target cells

[0380] The trispecific antibody-induced T cell-mediated killing of target cells was evaluated in human tumor cells NCI-H1299 (CEA negative; EGFR medium level), NCI-H157 (CEA negative; EGFR medium level) and NCI-H1395 (CEA medium level; EGFR medium level). Human PBMCs were used as effectors and killing was detected after 72 hours of incubation with antibodies.

[0381] Briefly, tumor cells were digested with trypsin / EDTA, washed once with pre-cooled PBS, resuspended in RPMI 1640 medium containing 10% FBS, and plated at a density of 5,000 cells / well in a flat-bottom 96-well plate (Corning 3599). After 4 hours of culture, 50 μL of serially diluted antibody solution was added to each well. Each concentration was added to three replicate wells. The antibody was incubated with tumor cells for more than 30 minutes to allow the protein to fully attach to the cells. PBMC cells (Reid Bio, 1521) were quickly thawed, revived, added to 10 mL of RPMI 1640 medium containing 10% FBS, centrifuged at 1000 rpm for 5 minutes, and the supernatant was discarded and resuspended. The cell density was adjusted and 50 μL PBMC was added to each well of target cells to achieve a final E:T ratio of 10:1. The 96-well plate was placed in a 37°C, 5% CO2 incubator for 3 days.

[0382] After 3 days of culture, carefully remove the cell culture supernatant. Pat the plate dry with a paper towel. Add serum-free DMEM containing 10% CCK8 detection solution (Dojindo, CK04-20) to the 96-well plate and incubate at 37°C for 2-4 hours. After 4 hours, use a microplate reader (BioTech, Synergy LX) to detect the absorbance at 450nm. Calculate cell viability according to the following formula: Cell viability = [(As-Ab) / (Ac-Ab)] × 100%.

[0383] As: experimental wells (culture medium containing cells, CCK-8, and tested antibodies);

[0384] Ac: control wells (containing cells, CCK-8, and culture medium without the tested antibody);

[0385] Ab: blank well (culture medium without cells, test antibodies, or CCK-8).

[0386] GraphPad Prism software was used to perform a four-parameter fit on cell activity according to the following formula: Y = (AD) / [1+(X / C)B]+D.

[0387] A: estimated value of the asymptote above the curve;

[0388] D: estimate of the asymptote below the curve;

[0389] B: slope of the curve;

[0390] C: The dose corresponding to half of the maximum binding (EC50);

[0391] Y is the detected cell activity;

[0392] X is the drug concentration.

[0393] The results showed that the trispecific antibody induced PBMC to kill tumor antigen-positive cells ( Figures 4A-4F ). The EC50 values ​​related to cell killing ability calculated by GraphPad Prism are shown in the following table. It can be seen that the cell killing effect of the non-unilateral CD3 / CEA / EGFR trispecific antibody is significantly weaker than that of the anti-EGFR-BsAb bispecific antibody, but its cell killing effect is significantly stronger than that of the CEA-BsAb bispecific antibody.

[0394] Table 7: Cytotoxic effect of non-unilateral CD3 / CEA / EGFR trispecific antibodies on EGFR-positive tumor cells (EC50, nM)

[0395] sample NCI-H157 NCI-H1299 NCI-H1395 301 11.33 17.82 8.211 302 1.779 26.06 6.478 303 1.462 1.745 1.148 304 0.7299 0.8946 0.6343 305 0.7595 2.598 0.9622 306 2.677 5.17 4.39 307 0.9831 1.561 5.559 308 0.2087 0.9503 1.082 309 0.9552 5.159 0.6949 310 0.5263 3.372 2.709 311 6.857 8.322 2.496 312 3.207 3.015 5.34 CEA-BsAb >100 >100 43.14 EGFR-BsAb 0.02 0.03177 0.0108

[0396] Example 5: Cytokine detection after non-unilateral CD3 / CEA / EGFR trispecific antibody induces T cells to kill tumor cells

[0397] After CD3 / CEA / EGFR trispecific antibody induced T cell-mediated cytotoxicity of tumor cells, cytokine secretion was assessed by ELISA analysis of cell supernatants.

[0398] After 72 hours of incubation, the CCK8 assay was performed with an E:T ratio of 10:1 as described above (Example 4). Subsequently, the 96-well plate was centrifuged at 2000 rpm for 10 minutes, and the supernatant was transferred to a new 96-well plate and stored at -20°C for subsequent analysis. The amount of TNFα, IFN-γ, IL-2 and IL-6 in the cell supernatant was determined using R&D detection kits, including: human IL-2 DuoSet ELISA (R&D, #DY202-05), human IL-6 DuoSet ELISA (R&D, #DY206-05), human IFN-γ DuoSet ELISA (R&D, #DY285B-05), and human TNF-α DuoSet ELISA (R&D, #DY210).

[0399] The ELISA test steps are as follows: Dilute the capture antibody 120 times with PBS to its working concentration. Add 100 μl of the diluted antibody to each well of the 96-well sample plate. Seal the 96-well plate and incubate overnight at room temperature. Carefully aspirate the liquid from the 96-well plate. Wash the 96-well plate three times with 300 μl of wash buffer and pat dry with a paper towel several times to absorb the liquid. Add 300 μl of Reagent Diluent to each well and incubate at room temperature for 1 hour after sealing. Prepare samples and standards during the incubation period. Dilute the standards 2-fold with Reagent Diluent for a total of 7 dilutions to a maximum concentration of 1000 pg / ml. Dilute the supernatant of the cells to be tested 2.5 times with Reagent Diluent. After blocking, wash the plate three times with 300 μl of wash buffer and pat dry with a paper towel several times to absorb the liquid. Add 100 μl of standard or sample to each well and incubate for 2 hours at room temperature. Then wash the wells three times with 300 μl of wash buffer and pat dry. Dilute the detection antibody 60-fold to its working concentration with reagent diluent. Add 100 μl of the diluted detection antibody to each well and incubate for 2 hours at room temperature. Wash the wells three times with 300 μL of wash buffer and pat dry. Dilute streptavidin-HRP 40-fold to its working concentration with reagent diluent. Add 100 μl of diluted streptavidin-HRP to each well and incubate for 20 minutes at room temperature. Wash the wells three times with 300 μl of wash buffer and pat dry. Add 100 μl of matrix solution to each well. Incubate at room temperature in the dark for 20 minutes. Add 50 μl of stop solution to each well. Tap the 96-well plate to mix the liquid, place it in a microplate reader, and read the optical density at 450 nm and 540 nm respectively. The OD450 reading minus the OD540 reading is the final reading. Use GraphPad software to fit the standard curve and calculate the concentration of the corresponding cytokine in the sample.

[0400] The cells were incubated with antibodies for 72 hours, and the cytokine content in the cell supernatant was detected (see Figures 5-7). The results showed that the antibody significantly induced the release of cytokines during the antibody-induced T cell-mediated cell killing process. The amount of cytokines released was ranked as follows: INF-γ>IL-6>TNFα. The secretion of IFN-γ increased significantly, which is closely related to cell killing. In the case of high antibody concentrations, the cells secreted significantly increased amounts of TNFα and IL-6, indicating that adverse reactions (cytokine storms) may have occurred. The content of IL-2 did not change significantly throughout the culture process.

[0401] In summary, these examples show that our CD3 / CEA / EGFR trispecific antibody has a significant T cell-mediated cell killing effect. IFN-γ secretion was induced during cell killing, and the levels of TNFα and IL-6 were significantly increased, indicating that the safety of non-unilateral trispecific antibodies should be monitored.

[0402] According to the above results, the cell killing effect of the non-unilateral CD3 / CEA / EGFR trispecific antibody is significantly weaker than that of EGFR-BsAb, but stronger than that of CEA-BsAb.

[0403] Example 6: Construction of a unilateral CD3 / CEA / EGFR trispecific antibody

[0404] According to Figure 8, 16 unilateral trispecific antibody structures based on CD3 (TA1-Fab), CEA and EGFR monovalent binding domains were designed. In these structures, TA1-Fab, EGFR single domain antibody and CEA single domain antibody are simultaneously located at the N-terminus of Fc.

[0405] When the single domain antibody is located at the C-terminus or N-terminus of the TA1 light chain, a linker (SEQ ID NO: 11) is used to connect the two fragments. When the single domain antibody is located at the C-terminus of the TA1 heavy chain, a linker (SEQ ID NO: 11) is used to connect the two fragments. When the EGFR and CEA single domain antibodies are located at the knob-Fc or hole-Fc at the same time (Figure 8M-8P), a linker (SEQ ID NO: 11) is used to connect the two single domain antibodies, and the Fc hinge region (Hinge) is used as a linker to connect the Fc and the single domain antibody close to the Fc.

[0406] Plasmid construction: All genes involved in this application were synthesized by GenScript (Nanjing GenScript Biotechnology Co., Ltd.) and then inserted into the mammalian expression vector pEE12.4 by enzyme digestion. The variable regions of the heavy and light chain DNA sequences were cloned into pEE12.4 with a pre-inserted human IgG4 heavy chain constant region or kappa light chain constant region in the frame. The plasmids were extracted with the OMEGA plasmid extraction kit and stored at -80°C.

[0407] Example 7: Preparation of unilateral CD3 / CEA / EGFR trispecific antibody

[0408] The unilateral CD3 / CEA / EGFR trispecific antibody was expressed by the above plasmid through polyethyleneimine (PEI, POLYETHYLENEIMINE'MAX; Polysciences, 24765-2)-mediated Expi 293F TM Cells were co-transfected with the corresponding expression plasmids at a ratio of 1:2:1 [“heavy chain Fc (hole)”: “light chain”: “heavy chain Fc (knob)”].

[0409] Expi 293F TM Cells in CD OptiCHO TM The cells were suspended and cultured in the culture medium at 37°C (5% CO2, 135 rpm). One day before transfection, 293F cells were subcultured into 1L gas permeable conical flasks (Corning) at a seeding density of 1.0×10 6 cells / mL in a volume of 200 mL. The cell density on the day of transfection was estimated to be 1.8-2.0×10 6 cells / mL. The cell suspension was centrifuged at 1000 rpm for 5 minutes at room temperature and washed once with Expi293 medium. The cells were collected and resuspended in 200 mL of Expi293 medium. 400 μg of plasmid was diluted with 5 mL of Opti-MEM medium and vortexed for 15 seconds. 1.2 mg of PEI was diluted with 5 mL of Opti-MEM medium and vortexed for 15 seconds. The PEI-containing solution was added dropwise to the DNA-containing solution, mixed gently, and incubated at room temperature for 15 minutes. The plasmid / PEI mixture was added to the cell suspension. The cell suspension was placed in an incubator at 37°C, 5% CO2, and 85 rpm for 4 hours. After 4 hours, 200 mL of EX-CELLTM293 medium and 2 mM glutamine (Gibco) were added, and the speed was adjusted to 135 rpm. After 24 hours, 3.8 mM VPA (a cell proliferation inhibitor) was added. After 72 hours, 40 mL of medium D was added. After 7 days, the supernatant was collected by centrifugation at 18,000 rpm for 30 minutes. The supernatant was filtered through a 0.22 μm sterile filter, sodium azide was added to a final concentration of 0.01% w / v, and the supernatant was stored at 4°C for purification.

[0410] The target protein was purified using protein A. The cell culture supernatant was loaded onto a MabSelect Prism A FF (GE; 17-5498-01) resin balanced with 20 mL of equilibration buffer (25 mM Tris, 150 mM NaCl, pH 7.5). Unbound protein was removed with at least 10 column volumes of wash buffer (25 mM Tris, 150 mM NaCl, pH 7.5). The target protein was eluted with 5 column volumes of elution buffer (20 mM sodium citrate, pH 3.5). The protein solution was neutralized by adding 1 / 10 volume of neutralization buffer (1 M Tris, pH 9.5).

[0411] Using Zeba TM The target protein was transferred to the desired buffer using a desalting centrifugal column (ThermoFisher) or ultrafiltration tube (Millipore). The purity and concentration of the protein were determined by SDS-PAGE electrophoresis and NanoDrop 2000. 2-3 μg of sample was used for SDS-PAGE electrophoresis.

[0412] The target protein was concentrated and filtered, and then used for a gel filtration column (Gel Filtration, HiLoad Superdex 200, GE) equilibrated with an equilibration buffer (20 mM histidine, 140 mM sodium chloride, pH 6.0). The sample (molecular weight, purity, aggregation, etc.) was analyzed at room temperature using a PBS buffer (pH 7.2) at a flow rate of 0.5 mL / min. The resulting high-purity monomeric protein was aliquoted and stored at -80°C.

[0413] In an alternative purification method, protein A affinity chromatography (MabSelect SuRe, GE) was used to purify the protein from the cell supernatant. The protein eluate was then subjected to cation exchange chromatography (HiTrap SP HP, GE) and subsequently analyzed by gel filtration columns. The target protein purity obtained using this purification method was greater than 90%.

[0414] The protein SDS-PAGE results are shown in Figure 9. Protein 404 was not successfully expressed. The yield of the trispecific antibodies was 13.68-35.12 mg / L, and the purity was >92%. See the table below for details.

[0415] Table 8: Yield and purity of unilateral CD3 / CEA / EGFR trispecific antibodies

[0416]

[0417]

[0418] Example 8: Determination of the affinity of a unilateral CD3 / CEA / EGFR trispecific antibody by biofilm interferometry (BLI)

[0419] Biomembrane interferometry (BLI) experiments were performed on an OCTECT RED96e (ForteBio) at 30°C using 0.02% PBST (10 mmol / L Na2HPO4; 1.75 mmol / L KH2PO4; 137 mmol / L NaCl; 2.65 mmol / L KCl; pH 7.2-7.4, 0.02% surfactant Tween 20) as running buffer.

[0420] The antibody was first captured onto an AHC biosensor coated with anti-human IgG Fc (anti-hIgG Fc, Sartorius, 18-5060). The antibody was diluted to 5 μg / mL and 1.5 nM protein was coupled to the AHC sensor surface in 0.02% PBST.

[0421] The analytes including (1) recombinant human CD3ε / δ protein (ACRO system, CDD-H52W1); (2) recombinant human tumor antigen CEA (CEACAM-5 / CD66e, ACRO system, CE5-H5226) and (3) recombinant human EGFR protein (ACRO system, EGR-H5222) were diluted to 200nM, 100nM, 50nM, 25nM, 12.5nM, 6.25nM and 3.125nM in 0.02% PBST.

[0422] After the antibody was captured by the sensor, it was used to bind and dissociate with different concentrations of analyte to obtain the KD value of the interaction. The experimental parameters were as follows: baseline 1: 60 seconds, loading: 240 seconds, baseline 2: 180 seconds, binding: 240 seconds, dissociation: 600 seconds, high sensitivity kinetics: 2 Hz. The kinetic constants were derived according to the 1:1 Langmuir binding rate equation to obtain the equilibrium dissociation constant (KD). The binding and dissociation curves were fitted simultaneously using Data Analysis HT 12 software (Sartorius, 50-5029) to calculate the KD. on and K off The affinity KD is calculated according to the following formula: KD = K off / K on .

[0423] The experimental results show that the affinity of the unilateral trispecific antibody to CEA is 0.12nM-0.90nM, the affinity to EGFR is 6.87nM-25.3nM, and the affinity to CD3 is 0.013nM-3.52nM. The KD values ​​between the trispecific antibody and different antigens are shown in the following table.

[0424] Table 9: Affinity of unilateral CD3 / CEA / EGFR trispecific antibodies for CEA (T=30°C)

[0425]

[0426]

[0427] Table 10: Affinity of unilateral CD3 / CEA / EGFR trispecific antibodies to EGFR (T=30°C)

[0428]

[0429]

[0430] Table 11: Affinity of single-sided CD3 / CEA / EGFR trispecific antibodies for CD3ε / δ (T=30°C)

[0431]

[0432] Example 9: Unilateral CD3 / CEA / EGFR trispecific antibody induces T cell-mediated killing of EGFR-positive tumor target cells

[0433] The trispecific antibody-induced T cell-mediated killing of target cells was tested in human tumor cells NCI-H1299 (CEA negative; EGFR medium level), NCI-H157 (CEA negative; EGFR medium level), NCI-H1395 (CEA medium level; EGFR medium level) and NCI-H1573 (CEA high level; EFGR high level). Human PBMCs were used as effectors and killing was detected after 72 hours of incubation with antibodies.

[0434] Briefly, tumor cells were digested with trypsin / EDTA, washed once with pre-cooled PBS, resuspended in RPMI 1640 medium containing 10% FBS, and plated at a density of 5,000 cells / well in a flat-bottom 96-well plate (Corning3599). After 4 hours of culture, 50 μL of serially diluted antibody solution was added to each well. Each concentration was added to three replicate wells. The antibody was incubated with tumor cells for more than 30 minutes to allow the protein to fully attach to the cells. PBMC cells (Reid Bio, 1521) were quickly thawed, revived, added to 10 mL of RPMI 1640 medium containing 10% FBS, centrifuged at 1000 rpm for 5 minutes, and the supernatant was discarded and resuspended. The cell density was adjusted and 50 μL PBMC was added to each well of target cells to achieve a final E:T ratio of 10:1. The 96-well plate was placed in a 37°C, 5% CO2 incubator for 3 days.

[0435] After 3 days of culture, carefully remove the cell culture supernatant. Pat the plate dry with a paper towel. Add serum-free DMEM containing 10% CCK8 detection solution (Dojindo, CK04-20) to the 96-well plate and incubate at 37°C for 2-4 hours. After 4 hours, use a microplate reader (BioTech, Synergy LX) to detect the absorbance at 450nm. Calculate cell activity according to the following formula: Cell activity = [(As-Ab) / (Ac-Ab)] × 100%.

[0436] The results showed that the unilateral trispecific antibody induced PBMC to kill tumor antigen EGFR positive cells (Figure 10). The EC50 values ​​related to cell killing ability calculated by GraphPad Prism are shown in the following table.

[0437] It can be seen that the killing effect of the unilateral trispecific antibody on the tumor cells NCI-H157 and NCI-H1299 is significantly stronger than that of the CEA-BsAb bispecific antibody. The killing effect of 409-412 is stronger than or close to that of the EGFR-BsAb bispecific antibody. Except for 410-414, the killing effect of the unilateral trispecific antibody on the tumor cell NCI-H1395 is significantly stronger than or close to that of CEA-BsAb. The killing effect of 409 is significantly stronger than that of EGFR-BsAb. The killing effect of the unilateral trispecific antibodies 401, 406, 408-411 and 416 on the tumor cell NCI-H1573 is significantly stronger than that of CEA-BsAb, and its killing effect is close to that of EGFR-BsAb.

[0438] Table 12: Cytotoxicity of EGFR-positive tumor cells induced by unilateral CD3 / CEA / EGFR trispecific antibodies (EC50, nM)

[0439]

[0440]

[0441] Example 10: Unilateral CD3 / CEA / EGFR trispecific antibody induces T cell-mediated killing of EGFR-positive tumor target cells with low EGFR expression

[0442] The ability of unilateral CD3 / CEA / EGFR trispecific antibodies 401, 403, 409, 410, 411 and 412 to induce T cell-mediated killing of tumor target cells with low EGFR expression was evaluated. The tumor cell lines used included: human breast cancer SK-BR-3 (EGFR negative), human non-small cell lung adenocarcinoma NCI-H157 monoclonal 21 cell line (NCI-H157-21#) (EGFR knockdown) (iGene Biotechnology, HSH117865-LVRU6GP).

[0443] Human PBMCs were used as effectors and killing was measured after 72 hours of incubation with antibodies. The EC50 values ​​related to cell killing ability calculated using GraphPad Prism are shown in the table below.

[0444] The results showed that the unilateral CD3 / CEA / EGFR trispecific antibodies 410-412 significantly enhanced the killing of T cells against tumor cells with low EGFR expression (Figure 10), and their killing effect was significantly stronger than that of the CEA-BsAb and EGFR-BsAb bispecific control antibodies. In addition, the killing effect of 409 on SK-BR-3 cells was also significantly stronger than that of CEA-BsAb and EGFR-BsAb. The killing effects of 401 and 403 on NCI-H157-21# cells were significantly stronger than those of CEA-BsAb and EGFR-BsAb. Since SK-BR-3 and NCI-H157-21# are both cells with low CEA expression, the unilateral CD3 / CEA / EGFR trispecific antibodies 401, 403, 409, 410, 411 and 412 have potential application value in cells with low CEA and EGFR expression.

[0445] Table 13: Cytotoxic effect of unilateral CD3 / CEA / EGFR trispecific antibody on tumor cells with low EGFR expression (EC50, nM)

[0446] sample SK-BR-3 NCI-H157-21# 401 >100 8.966 403 >100 11.35 409 0.0884 >100 410 0.7919 2.096 411 0.1146 2.74 412 0.7869 47.84 CEA-BsAb >100 60.83 EGFR-BsAb >100 79.13

[0447] Example 11: Preparation of unilateral CD3 / CEA-CAM6 / EGFR, CD3 / CEA / HER2 and CD3 / CEA-CAM6 / HER2 trispecific antibodies

[0448] HER2 and CEA-CAM6 single domain antibodies were obtained by immunizing llamas. Following Example 6, based on the structures of 401, 403, 409-412, we designed unilateral CD3 / CEA-CAM6 / EGFR, CD3 / CEA / HER2 and CD3 / CEA-CAM6 / HER2 trispecific antibodies, which were named 501, 503, 509-512; 601, 603, 609-612; 701, 703, 709-712 (Figure 11). The sequences of HER2 and CEA-CAM6 single domain antibodies are shown in the following table.

[0449] Table 14: HER2, CEA-CAM6 single domain antibody sequence, linker sequence and Fc hinge region sequence, linker sequence and Fc hinge region sequence

[0450]

[0451] Plasmid construction: All genes involved in this application were synthesized by GenScript (Nanjing GenScript Biotechnology Co., Ltd.) and then inserted into the mammalian expression vector pEE12.4 by enzyme digestion. The variable regions of the heavy and light chain DNA sequences were cloned into pEE12.4 with a pre-inserted human IgG4 heavy chain constant region or kappa light chain constant region in the frame. The plasmids were extracted using the OMEGA plasmid extraction kit and stored at -80°C.

[0452] Following Examples 2 and 7, the above plasmids were transfected into Expi 293F using PEI. TM cells and use a variety of purification methods to produce antibodies.

[0453] The protein SDS-PAGE results are shown in Figure 12. The yield of the trispecific antibody was 13.18-45.1 mg / L, and the purity was greater than 92%. See the table below for details.

[0454] Table 15: Yield and purity of single-sided CD3 / CEA-CAM6 / EGFR, CD3 / CEA / HER2 and CD3 / CEA-CAM6 / HER2 trispecific antibodies

[0455]

[0456] Example 12: Determination of the affinity of single-sided CD3 / CEA-CAM6 / EGFR, CD3 / CEA / HER2 and CD3 / CEA-CAM6 / HER2 trispecific antibodies by BLI

[0457] Biomembrane interferometry (BLI) experiments were performed on an OCTECT RED96e (ForteBio) at 30°C using 0.02% PBST (10 mmol / L Na2HPO4, 1.75 mmol / L KH2PO4, 137 mmol / L NaCl, 2.65 mmol / L KCl, pH 7.2-7.4, 0.02% surfactant Tween 20) as running buffer.

[0458] The antibody was first captured onto an AHC biosensor coated with anti-human IgG Fc (anti-hIgG Fc, Sartorius, 18-5060). The antibody was diluted to 5 μg / mL and 1.5 nM protein was coupled to the AHC sensor surface in 0.02% PBST.

[0459] The analytes including (1) recombinant human CD3ε / δ protein (ACRO system, CDD-H52W1), (2) recombinant human tumor antigen CEA (CEACAM-5 / CD66e, ACRO system, CE5-H5226), (3) recombinant human EGFR protein (ACRO system, EGR-H5222), (4) recombinant human tumor antigen HER2 (ERBB2, ACRO system, HE2-H5212) and (5) recombinant human tumor antigen CEACAM-6 (ACRO system, CE6-H5223) were diluted to 200nM, 100nM, 50nM, 25nM, 12.5nM, 6.25nM and 3.125nM in 0.02% PBST.

[0460] After the antibody was captured by the sensor, it was allowed to bind and dissociate with different concentrations of analyte to obtain the KD value of the interaction. The experimental parameters were as follows: baseline 1: 60 seconds, loading: 240 seconds, baseline 2: 180 seconds, binding: 240 seconds, dissociation: 600 seconds, high sensitivity kinetics: 2Hz. The kinetic constants were derived according to the 1:1 Langmuir binding rate equation to obtain the equilibrium dissociation constant (KD). The binding and dissociation curves were fitted simultaneously using Data Analysis HT 12 software (Sartorius, 50-5029) to calculate the KD. on and K off The affinity KD is calculated according to the following formula: KD = K off / K on .

[0461] The experimental results show that the affinity of the unilateral CD3 / CEA-CAM6 / EGFR trispecific antibody to CEA-CAM6 is 0.634nM-2.01nM, the affinity to EGFR is 2.65nM-11.3nM, and the affinity to CD3 is 6.09nM-7.47nM. The KD values ​​of different antigens are shown in the following table.

[0462] Table 16: Affinity of the unilateral CD3 / CEA-CAM6 / EGFR trispecific antibody for CEA-CAM6 (T=30°C)

[0463]

[0464] Table 17: Affinity of unilateral CD3 / CEA-CAM6 / EGFR trispecific antibodies to EGFR (T=30°C)

[0465]

[0466]

[0467] Table 18: Affinity of the single-sided CD3 / CEA-CAM6 / EGFR trispecific antibody for CD3ε / δ (T=30°C)

[0468]

[0469] The experimental results show that the affinity of the unilateral CD3 / CEA / HER2 trispecific antibody to CEA is 0.107nM-0.195nM, the affinity to HER2 is 48.4nM-108nM, and the affinity to CD3 is 0.608nM-1.77nM. The KD data are shown in the following table.

[0470] Table 19: Affinity of the unilateral CD3 / CEA-CAM6 / EGFR trispecific antibody for CEA (T=30°C)

[0471]

[0472] Table 20: Affinity of unilateral CD3 / CEA-CAM6 / EGFR trispecific antibodies for HER2 (T=30°C)

[0473]

[0474]

[0475] Table 21: Affinity of the unilateral CD3 / CEA-CAM6 / EGFR trispecific antibody for CD3ε / δ (T=30°C)

[0476]

[0477] The experimental results show that the affinity of the unilateral CD3 / CEA-CAM6 / HER2 trispecific antibody to CEA-CAM6 is 4.16nM-13.3nM, the affinity to HER2 is 15.4nM-38.9nM, and the affinity to CD3 is 0.351nM-1.80nM. The KD data for binding to different antigens are shown in the following table.

[0478] Table 22: Affinity of the unilateral CD3 / CEA-CAM6 / EGFR trispecific antibody for CEA-CAM6 (T=30°C)

[0479]

[0480] Table 23: Affinity of unilateral CD3 / CEA-CAM6 / EGFR trispecific antibodies for HER2 (T=30°C)

[0481]

[0482] Table 24: Affinity of the unilateral CD3 / CEA-CAM6 / EGFR trispecific antibody for CD3ε / δ (T=30°C)

[0483]

[0484] Example 13: Unilateral CD3 / CEA-CAM6 / EGFR, CD3 / CEA / HER2 and CD3 / CEA-CAM6 / HER2 trispecific antibodies induce T cell-mediated tumor cell killing

[0485] T cell-mediated killing of target cells induced by unilateral CD3 / CEA-CAM6 / EGFR trispecific antibodies was evaluated in human tumor cells NCI-H157-21# (high expression level of CEA-CAM6, low expression level of EGFR), MDA-MB-453 (low expression level of CEA-CAM6, negative EGFR), SK-OV-3 (high expression level of CEA-CAM6, medium expression level of EGFR), NCI-H1573 (high expression level of CEA-CAM6, high expression level of EGFR), SK-BR-3 (low expression level of CEA-CAM6, low expression level of EGFR), NCI-H1975 (high expression level of CEA-CAM6, medium expression level of EGFR), MDA-MB-468 (high expression level of CEA-CAM6, medium expression level of EGFR) and MKN-45 (high expression level of CEA-CAM6, medium expression level of EGFR).

[0486] T cell-mediated killing of target cells induced by unilateral CD3 / CEA / HER2 trispecific antibodies was evaluated in the human tumor cells MDA-MB-453 (moderate expression level of CEA, moderate expression level of HER2), SK-OV-3 (negative CEA, high expression level of HER2), NCI-H1573 (high expression level of CEA, negative HER2), MCF-7 (low expression level of CEA, low expression level of HER2), SK-BR-3 (low expression level of CEA, moderate expression level of HER2) and MKN-45 (high expression level of CEA, negative HER2).

[0487] T cell-mediated killing of target cells induced by unilateral CD3 / CEA-CAM6 / HER2 trispecific antibodies was evaluated in the human tumor cells MDA-MB-453 (intermediate expression level of CEA-CAM6, intermediate expression level of HER2), SK-OV-3 (high expression level of CEA-CAM6, high expression level of HER2), NCI-H1573 (high expression level of CEA-CAM6, negative for HER2), MCF-7 (low expression level of CEA-CAM6, low expression level of HER2), SK-BR-3 (low expression level of CEA-CAM6, intermediate expression level of HER2) and MKN-45 (high expression level of CEA-CAM6, negative for HER2).

[0488] Human PBMCs were used as effectors and killing was measured after 72 h of incubation with antibodies.

[0489] Briefly, tumor cells were digested with trypsin / EDTA, washed once with pre-cooled PBS, resuspended in RPMI 1640 medium containing 10% FBS, and plated at a density of 5,000 cells / well in a flat-bottom 96-well plate (Corning3599). After 4 hours of culture, 50 μL of serially diluted antibody solution was added to each well. Each concentration was added to three replicate wells. The antibody was incubated with tumor cells for more than 30 minutes to allow the protein to fully attach to the cells. PBMC cells (Reid Bio, 1521) were quickly thawed, revived, added to 10 mL of RPMI 1640 medium containing 10% FBS, centrifuged at 1000 rpm for 5 minutes, and the supernatant was discarded and resuspended. The cell density was adjusted and 50 μL PBMC was added to each well of target cells to achieve a final E:T ratio of 10:1. The 96-well plate was placed in a 37°C, 5% CO2 incubator for 3 days.

[0490] After 3 days of culture, carefully remove the cell culture supernatant. Pat the plate dry with a paper towel. Add serum-free DMEM containing 10% CCK8 detection solution (Dojindo, CK04-20) to the 96-well plate and incubate at 37°C for 2-4 hours. After 4 hours, use a microplate reader (BioTech, Synergy LX) to detect the absorbance at 450nm. Calculate cell activity according to the following formula: Cell activity = [(As-Ab) / (Ac-Ab)] × 100%.

[0491] The results showed that the single-sided trispecific antibody induced PBMC to kill tumor antigen-positive cells (Figure 13). The EC50 values ​​related to cell killing ability calculated by GraphPadPrism are shown in the following table.

[0492] For cell lines with high expression of CEA-CAM6 or EGFR (SK-OV-3, NCI-H1573, MDA-MB-468 and MKN-45), the cell killing effect of the unilateral CD3 / CEA-CAM6 / EGFR trispecific antibody was significantly stronger than that of the bispecific control antibodies (anti-EGFR-BsAb and CEA-CAM6-BsAb).

[0493] In the cell line NCI-H1975 with high expression of CEA-CAM6 and EGFR, the cell killing effect of the unilateral trispecific antibody was significantly stronger than that of the bispecific control antibody anti-CEA-CAM6-BsAb, and was close to (511, 512) or stronger than (501, 503, 509, 510) the bispecific control antibody anti-EGFR-BsAb.

[0494] In the cell line NCI-H157-21# with low EGFR expression and high CEA-CAM6 expression, the killing effect of the unilateral trispecific antibody was stronger than that of EGFR-BsAb, and the killing effects of 509 and 510 were stronger than those of CEA-CAM6-BsAb.

[0495] In the cell line MDA-MB-453 with low EGFR expression and high CEA-CAM6 expression, the killing effect of the unilateral trispecific antibody was stronger than that of EGFR-BsAb, while the killing effects of some unilateral trispecific antibodies (503, 509, 510) were similar to those of CEA-CAM6-BsAb.

[0496] In the cell line SK-BR-3 with low expression of EGFR and CEA-CAM6, the unilateral trispecific antibody showed a stronger killing effect than EGFR-BsAb and CEA-CAM6-BsAb.

[0497] The above results show that the unilateral CD3 / CEA-CAM6 / EGFR trispecific antibody has a significant killing effect on cell lines with different CEA-CAM6 / EGFR expressions. The results are shown in Figure 13 and the table below.

[0498] Table 25: Cytotoxicity of unilateral CD3 / CEA-CAM6 / EGFR trispecific antibodies to tumor cells (EC50: nM)

[0499] sample NCI-H157-21# MDA-MB-453 sK-Ov-3 NCI-H1573 sK-BR-3 NCI-H1975 MDA-MB-468 MKN-45 501 7.12E-02 2.498 2.47E-04 4.94E-04 0.2661 1.03E-04 1.65E-06 1.72E-05 503 8.10E-02 0.8603 3.31E-05 1.42E-03 0.2499 8.68E-04 1.46E-06 1.74E-05 509 7.90E-04 0.8595 1.71E-05 3.36E-04 0.02512 4.37E-04 1.96E-06 2.98E-05 510 4.53E-03 0.717 9.69E-06 2.44E-04 0.05413 4.19E-03 1.57E-06 2.61E-05 511 2.16E-02 1.084 2.27E-05 2.57E-04 0.07084 1.20E-02 2.40E-06 2.02E-05 512 9.00E-02 2.876 9.02E-06 1.22E-04 0.1234 1.95E-02 2.08E-06 2.09E-05 EGFR-BsAb 7.516 >100 1.10E-03 7.48E-03 7.399 5.57E-03 1.34E-03 5.72E-02 CEA-CAM6-BsAb 4.65E-03 0.1145 1.14E-02 6.28E-03 4.195 5.63E-02 0.1082 3.04E-04

[0500] In the cell line MDA-MB-453, which moderately expresses CEA and HER2, the killing effect of the unilateral CD3 / CEA / HER2 trispecific antibodies was stronger than that of CEA-BsAb and HER2-BsAb, except for 610 and 612.

[0501] In the CEA-negative cell line SK-OV-3 with high HER2 expression, the unilateral CD3 / CEA / HER2 trispecific antibody showed a stronger killing effect than CEA-BsAb, and some antibodies showed similar (601, 603, 609, 610, 611) or stronger (612) killing effects compared with HER2-BsAb.

[0502] In the cell line MCF-7 with low expression of CEA and HER2, the unilateral trispecific antibodies (609, 610, 611, 612) showed a stronger killing effect than CEA-BsAb and HER2-BsAb.

[0503] In the cell line SK-BR-3 with low CEA expression and moderate HER2 expression, the killing effect of the unilateral trispecific antibody (603, 609, 610) was stronger than that of CEA-BsAb and HER2-BsAb.

[0504] In the HER2-negative cell line NCI-H1573 with high CEA expression, the killing effect of the unilateral CD3 / CEA / HER2 trispecific antibody was stronger than that of HER2-BsAb, and the killing effect was similar (603, 609, 610) or stronger (601, 611, 612) compared with CEA-BsAb.

[0505] In the HER2-negative cell line MKN-45 with high CEA expression, the killing effect of the unilateral CD3 / CEA / HER2 trispecific antibody was stronger than that of HER2-BsAb and HER2-BsAb.

[0506] The results showed that the six unilateral CD3 / CEA / HER2 trispecific antibodies had significant killing effects on cell lines with different CEA / HER2 expressions. The results are shown in Figure 13 and the table below.

[0507] Table 26: Cytotoxicity of unilateral CD3 / CEA / HER2 trispecific antibodies against tumor cells (EC50: nM)

[0508] sample MDA-MB-453 SK-OV-3 NCI-H1573 MCF-7 SK-BR-3 MKN-45 601 0.816 1.26E-03 5.60E-03 >100 >100 1.25E-04 603 0.1305 1.01E-02 2.48E-02 >100 8.447 5.34E-04 609 0.6294 6.66E-04 8.40E-03 10.15 6.031 3.23E-05 610 >100 3.26E-03 1.36E-02 5.97 1.709 2.52E-05 611 0.667 1.36E-03 5.23E-03 10.58 >100 2.94E-04 612 >100 9.66E-05 7.45E-03 10.94 >100 3.53E-04 CEA-BsAb 5.565 >100 7.55E-03 >100 80.76 5.72E-02 HER2-BsAb 0.9762 1.97E-04 >100 11.43 3.5 >100

[0509] In the cell line MDA-MB-453, which moderately expresses CEA-CAM6 and HER2, the unilateral CD3 / CEA-CAM6 / HER2 trispecific antibody showed stronger (701, 703, 709) or similar (710, 711, 712) killing effects than CEA-BsAb and HER2-BsAb.

[0510] In the SK-OV-3 cell line with high expression of CEA-CAM6 and HER2, the unilateral CD3 / CEA-CAM6 / HER2 trispecific antibody showed a stronger killing effect than CEA-BsAb and HER2-BsAb.

[0511] In the HER2-negative cell lines NCI-H1573 and MKN-45, which highly express CEA-CAM6, the unilateral CD3 / CEA-CAM6 / HER2 trispecific antibodies showed stronger cell killing effects than HER2-BsAb. In NCI-H1573, the unilateral trispecific antibodies (703, 709, 710, 711, 712) showed similar killing effects as CEA-CAM6-BsAb. In MKN-45, the killing effects of the unilateral trispecific antibodies were similar to (701) or stronger (703, 709, 710, 711, 712) than those of CEA-CAM6-BsAb.

[0512] In the cell lines MCF-7 and SK-BR-3 with low CEA-CAM6 expression and low / moderate HER2 expression, the killing effect of the unilateral CD3 / CEA-CAM6 / HER2 trispecific antibody was similar to or stronger than that of CEA-BsAb and HER2-BsAb.

[0513] The results showed that the six unilateral CD3 / CEA-CAM6 / HER2 trispecific antibodies had significant killing effects on cell lines with different CEA-CAM6 / HER2 expressions. The results are shown in Figure 13 and the table below.

[0514] Table 27: Cytotoxic effect of unilateral CD3 / CEA-CAM6 / HER2 trispecific antibody on tumor cells (EC50: nM)

[0515]

[0516]

[0517] Example 14: DSF method to detect the thermal stability of trispecific antibodies

[0518] Differential scanning fluorimetry (DSF) is a method that slowly heats samples on a fluorescence quantitative PCR instrument and detects the amount of fluorescent dye bound to proteins whose structures change during the heating process to evaluate the thermal stability of proteins.

[0519] The thermal stability of the trispecific antibody was monitored by DSF. After 19 μl of 5 μM protein sample was thoroughly mixed with 1 μl of 200×sypro orange (ThermoFisher, S6650), the mixture was added to a 96-well plate (Applied Biosystems, N8010560) in triplicate. After incubation at 25°C for 30 seconds, the temperature was raised from 25°C to 95°C at a rate of 0.05°C / min, and the fluorescence signal intensity was collected using a real-time fluorescence quantitative PCR instrument (Applied Biosystems, QuantStudio 5System). PBS was used as a blank control. After the experiment, the Tm value was analyzed using Protein Thermal Shift software.

[0520] The Tm values ​​were measured three times in triplicate wells. The experimental results are shown in the following table. The Tm1 values ​​of the single-sided trispecific antibody were 59.4°C-63.2°C. The Tm2 values ​​of the single-sided trispecific antibody were 69.5°C-74.5°C.

[0521] Table 28: Tm values ​​of single-sided trispecific antibodies detected by DSF method

[0522]

[0523] Example 15: Detection of the multispecificity of a unilateral trispecific antibody by ELISA

[0524] The experimental steps are as follows

[0525] (1) Antigen coating: Cardiolipin (Sigma, cat. C0563), keyhole limpethaemocyanin (KLH, Sigma, H8283), LPS (Sigma, L6529), ssDNA (Sigma, D8899), dsDNA (Sigma, D4522) and insulin (abs42019847) were plated in 96-well plates (Thermo Nunc, 475094) at 50 μg / mL, 5 μg / mL, 10 μg / mL, 1 μg / mL, 1 μg / mL and 5 μg / mL per well and incubated overnight at 4°C;

[0526] (2) The next morning, the plates were washed three times with 0.05% PBST (Biotek, 4052S Microplate Washer), blocked in PBS solution containing 3% skim milk at room temperature for 1 hour, and then washed three times with 0.05% PBST;

[0527] (3) Add 50 μl of 100 nM antibody or blank control PBS to each well (repeat 3 wells for each sample) and incubate at room temperature for 1 hour; wash the plate 3 times with 0.05% PBST;

[0528] (4) Add 50 μl of anti-human IgG-HRP conjugate (Sigma, AP113P, 1:8000 dilution) to each well; incubate the plate at room temperature for 1 hour and wash 6 times with 0.05% PBST;

[0529] (5) Add 50 μl of TMB colorimetric solution (Biopanda, TMB-S-001) to each well and incubate at room temperature for 10-15 minutes to develop color;

[0530] (6) Add 50 μl of 1 M HCl to each well to stop color development, and measure the absorbance at 450 nm using a microplate reader (Biotek, SYNERGY LX);

[0531] Data analysis: If the ratio of the antibody absorbance to the blank PBS absorbance is ≥ 3, there is excessive nonspecific binding.

[0532] The experimental results are shown in the table below. From the results, it can be seen that none of the three specific antibodies tested showed excessive non-specific binding, indicating that these three specific antibodies had no non-specific binding with the six antigens tested.

[0533] Table 29: Absorbance ratio of trispecific antibodies bound to various substances

[0534]

[0535]

[0536] Example 16: Detection of self-interaction of trispecific antibodies by biolayer interferometry (BLI) assay

[0537] Biomembrane interferometry (BLI) experiments were performed on an OCTECT RED96e (ForteBio) at 30°C using PBS (10 mmol / L Na2HPO4, 1.75 mmol / L KH2PO4, 137 mmol / L NaCl, 2.65 mmol / L KCl, pH 7.2-7.4) as running buffer.

[0538] First, the trispecific antibody and the control antibody (e.g., ofatumumab) were captured on the chip surface of the AHQ sensor (anti-hIgG Fc, sartorius, 18-5001) immobilized with anti-human Fc antibody. The bispecific control antibody was diluted to 1 μM and the protein was coupled to the AHQ sensor chip surface in PBS, with a signal value of about ~0.8 nm. Subsequently, the sites on the AHQ sensor that were not bound to the test antibody were completely blocked with human IgG antibody, and the self-binding signal of the trispecific antibody or the control antibody was analyzed. The experimental parameters were as follows: baseline 1: 60 s, loading 180 s, loading response: 0.8 nm, baseline 2: 180 s, binding: 240 s, high sensitivity kinetics: 2 Hz. The binding curve was obtained by Date Analysis HT 12 software (sartorius, 50-5029). If the self-binding signal of the test antibody was higher than the self-binding signal of the control antibody by more than 0.1 nm, the test antibody was considered to have self-interaction; otherwise, it was considered that there was no self-interaction. The experimental results are shown in the following table.

[0539] The self-interaction signal of the positive control ofatumumab was 0.11 nm, and the self-interaction signal of the trispecific antibody ranged from 0.037 nm to 0.121 nm, which was lower than 0.21 nm (0.11 nm + 0.1 nm), indicating that the self-interaction was weak, indicating that the trispecific antibody had good solubility.

[0540] Table 30: Trispecific antibody self-interaction signals detected by BLI

[0541] sample BLI response (nm) sample BLI response (nm) 401 0.127 610 0.088 403 0.092 611 0.07 409 0.106 612 0.082 410 0.121 701 0.094 411 0.101 703 0.108 412 0.089 709 0.083 501 0.104 710 0.09 503 0.072 711 0.104 509 0.092 712 0.073 510 0.137 CEA-BsAb 0.103 511 0.084 EGFR-BsAb 0.037 512 0.07 HER2-BsAb 0.066 601 0.096 CEA-CAM6-BsAb 0.045 603 0.084 Ofatumumab 0.11 609 0.098

[0542] Example 17: Construction and preparation of CD3 / c-MET / EGFR trispecific antibody

[0543] c-Met single domain antibody and EGFR single domain antibody (VHH) were obtained by immunizing llamas. In the schematic structure, they are referred to as c-Met and EGFR. According to the method of Example 6, based on the structures of 401-409 and 413-415, we designed CD3 / c-Met / EGFR trispecific antibodies, which were named 801-809 and 813-815, respectively (see Figures 15A-15L ). Meanwhile, a bispecific control antibody (c-Met BsAb and EGFR BsAb) containing the anti-CD3 fab TA1F was also designed (see Fig.14 ). TA1F represents the Fab region of the anti-CD3 antibody clone TA1. Anti-CD3 VH is linked to CH1-Hinge-Fc (IgG4), and anti-CD3 VL is linked to CL (κ). Table 1 lists the common sequences. Table 2 lists the sequences of EGFR single domain antibodies. The following table lists the sequences of c-Met single domain antibodies.

[0544] Table 31: Sequence Listing

[0545]

[0546] Based on the methods in Examples 2 and 7, the above plasmids were transfected into Expi 293F by PEI. TM cells, and used a variety of purification methods to prepare antibodies. Protein SDS-PAGE results are shown in Figures 16A-16B The yield of the trispecific antibody was 12.7-45.92 mg / L, and the purity of the trispecific antibody was ≥90%. The yield and purity data are listed in the table below.

[0547] Table 32: Yield and purity of CD3 / C-MET / EGFR trispecific antibodies

[0548] sample Yield (mg / L) purity(%) 801 39.24 92 802 45.92 90 803 17.7 90 804 52.5 90 805 21.8 92 806 25.6 90 807 12.7 92 808 16.3 92 809 13.9 90 813 31.4 92 814 15.0 92 815 34.9 92

[0549] Example 18: Determination of the affinity of CD3 / C-MET / EGFR trispecific antibodies by biofilm interferometry (BLI)

[0550] Biomembrane interferometry (BLI) experiments were performed on an OCTECT RED96e (ForteBio) at 30°C using 0.02% PBST (10 mmol / L Na2HPO4, 1.75 mmol / L KH2PO4, 137 mmol / L NaCl, 2.65 mmol / L KCl, pH 7.2-7.4, 0.02% surfactant Tween 20) as running buffer.

[0551] The antibody was first captured onto an AHC biosensor coated with anti-human IgG Fc (anti-hIgG Fc, Sartorius, 18-5060). The antibody was diluted to 5 μg / mL and 1.5 nM protein was coupled to the AHC sensor surface in 0.02% PBST.

[0552] Analytes including (1) recombinant human CD3ε / δ protein (ACRO System, CDD-H52W1), (2) recombinant human c-Met protein (ACRO System, MET-H5227) and (3) recombinant human EGFR protein (ACRO System, EGR-H5222) were diluted to 200 nM, 100 nM, 50 nM, 25 nM, 12.5 nM, 6.25 nM and 3.125 nM in 0.02% PBST.

[0553] After the antibody was captured by the sensor, it was used to bind and dissociate with different concentrations of analyte to obtain the KD value of the interaction. The experimental parameters were as follows: baseline 1: 60 seconds, loading: 240 seconds, baseline 2: 180 seconds, binding: 240 seconds, dissociation: 600 seconds, high sensitivity kinetics: 2 Hz. The kinetic constants were derived according to the 1:1 Langmuir binding rate equation to obtain the equilibrium dissociation constant (KD). The binding and dissociation curves were fitted simultaneously using Data Analysis HT 12 software (Sartorius, 50-5029) to calculate the KD. on and K off The affinity KD is calculated according to the following formula: KD = K off / K on .

[0554] The experimental results show that the affinity of the single-sided trispecific antibody to human CD3 is 0.17nM-3.14nM, the affinity to EGFR is 0.6nM-17nM, and the affinity to c-Met is 0.47nM-1.93nM. The KD values ​​between the trispecific antibody and different antigens are shown in the following table.

[0555] Table 33: Affinity of the single-sided CD3 / C-MET / EGFR trispecific antibody to EGFR (T=30°C)

[0556]

[0557]

[0558] Table 34: Affinity of single-sided CD3 / C-MET / EGFR trispecific antibodies for c-Met (T=30°C)

[0559]

[0560] Table 35: Affinity of single-sided CD3 / C-MET / EGFR trispecific antibodies to CD3 (T=30°C)

[0561]

[0562]

[0563] Example 19: Unilateral CD3 / C-MET / EGFR trispecific antibody induces T cell-mediated killing of EGFR-positive tumor target cells

[0564] The trispecific antibody-induced T cell-mediated killing of target cells was tested in human tumor cells NCI-H157(-21) (c-Met negative; EGFR negative), OSRC2 (c-Met low; EGFR high), UT44 (c-Met medium; EGFR high), UT33A (c-Met high; EGFR high), HT-29 (c-Met low; EGFR low), H-1975 (c-Met low; EGFR low). Human PBMCs were used as effectors and killing was measured after 72 hours of incubation with antibodies.

[0565] Briefly, tumor cells were digested with trypsin / EDTA, washed once with pre-cooled PBS, resuspended in RPMI 1640 medium containing 10% FBS, and plated at a density of 5,000 cells / well in a flat-bottom 96-well plate (Corning3599). After 4 hours of culture, 50 μL of serially diluted antibody solution was added to each well. Each concentration was added to three replicate wells. The antibody was incubated with tumor cells for more than 30 minutes to allow the protein to fully attach to the cells. PBMC cells (Reid Bio, 1521) were quickly thawed, revived, added to 10 mL of RPMI 1640 medium containing 10% FBS, centrifuged at 1000 rpm for 5 minutes, and the supernatant was discarded and resuspended. The cell density was adjusted and 50 μL PBMC was added to each well of target cells to achieve a final E:T ratio of 10:1. The 96-well plate was placed in a 37°C, 5% CO2 incubator for 3 days.

[0566] After 3 days of culture, carefully remove the cell culture supernatant. Pat the plate dry with a paper towel. Add serum-free DMEM containing 10% CCK8 detection solution (Dojindo, CK04-20) to the 96-well plate and incubate at 37°C for 2-4 hours. After 4 hours, use a microplate reader (BioTech, Synergy LX) to detect the absorbance at 450nm. Calculate cell activity according to the following formula: Cell activity = [(As-Ab) / (Ac-Ab)] × 100%.

[0567] As: experimental wells (culture medium containing cells, CCK-8, and tested antibodies);

[0568] Ac: control wells (containing cells, CCK-8, and culture medium without the tested antibody);

[0569] Ab: blank well (culture medium without cells, test antibodies, or CCK-8).

[0570] GraphPad Prism software was used to perform a four-parameter fit on cell activity according to the following formula: Y = (AD) / [1+(X / C)B]+D.

[0571] A: estimated value of the asymptote above the curve;

[0572] D: estimate of the asymptote below the curve;

[0573] B: slope of the curve;

[0574] C: The dose corresponding to half of the maximum binding capacity (EC50);

[0575] Y is the detected cell activity;

[0576] X is the drug concentration.

[0577] The results showed that the trispecific antibody induced PBMC to kill tumor antigen-positive cells. The EC50 values ​​related to cell killing ability calculated by GraphPad Prism are shown in the following table ("-" indicates negative, "+" indicates low expression, "++" indicates medium expression, and "+++" indicates high expression). It can be seen that the cell killing effect of the unilateral CD3 / C-MET / EGFR trispecific antibody is significantly stronger than that of the EGFR-BsAb and c-Met-BsAb bispecific antibodies.

[0578] Table 36: Cytotoxicity of CD3 / C-MET / EGFR trispecific antibodies against EGFR-positive tumor cells (EC50, nM)

[0579]

[0580] Example 20: In vivo antitumor effect of CD3 / C-MET / EGFR trispecific antibody on HT-29 human colon cancer xenografts

[0581] Female NCG mice (NOD / ShiLtJGpt-Prkdc em26Cd52 Iq em26Cd22 / Gpt, GemPharmatech, strain NO.T001475) (n=5) were subcutaneously injected with 1×10 6HT-29 cells (E:T ratio 1:1, total volume 200μl) were added to evaluate the therapeutic effects of the control bispecific antibody and trispecific antibody. One hour after the subcutaneous co-transplantation of tumor cells / PBMCs in mice, 0.0325mg / kg of bispecific antibody was injected intravenously twice a week, and the solvent control group (solvent control) was given PBS buffer (a total of 3 times). Tumor volume and body weight data were measured and recorded weekly with a digital caliper.

[0582] The relative change of body weight (RCB) was calculated according to the following formula: RCB (%) = [1-(Bi / B0) × 100%] (Bi: average body weight of mice on the i-th day of administration, B0: average body weight of mice on the 0th day of administration). At the same time, the tumor volume was calculated according to the following formula: long diameter × short diameter 2 / 2. Tumor growth inhibition rate (TGI TV ) is calculated according to the following formula: TGI TV (%) = [1-(Ti-T0) / (Vi-V0)] × 100% (Ti: average tumor volume of the treatment group on the i-th day of administration; T0: average tumor volume of the treatment group that can be measured for the first time; Vi: average tumor volume of the solvent control group on the i-th day of administration; V0: average tumor volume of the solvent control group that can be measured for the first time after administration). Two-way ANOVA analysis was performed, and P < 0.05 means statistically significant.

[0583] On the 23rd day of administration, the CD3 / C-MET / EGFR trispecific antibody showed a significant inhibitory effect on tumor volume compared with the vehicle control group, which was statistically significant (P<0.001). Figures 17A-17D And the following table. At a dose of 0.0325 mg / kg, the tumor growth inhibition rate (TGI) of 801, 802, 807, 808 and 813 TV ) were 85.33%, 78.11%, 96.33%, 76.91% and 99.67%, respectively, indicating that the above five trispecific antibodies have a strong tumor inhibitory effect. The TGI of the control bispecific antibody EGFR-BsAb at a dose of 0.3 mg / kg TV The TGI of the control bispecific antibody EGFR-BsAb at a dose of 0.1 mg / kg was 96.24%. TV The TGI of 7.5mg / kg and 1.5mg / kg c-Met-BsAb in PBMC / HT-29 mixed inoculation mouse model TV were 60.30% and 43.39% respectively, indicating that its tumor inhibition effect was poor (data not shown).

[0584] Table 37: Effect of CD3 / C-MET / EGFR trispecific antibodies on tumor volume in NCG mice inoculated with PBMC / HT-29 mixture

[0585]

[0586] During the administration period, the activity and feeding status of NCG mice were good, and the relative weight change rate was within ±5%. On the 23rd day of administration, there was no significant weight loss in the mice in the administration group. The above results show that the mice tolerated the administration of this frequency and dose, and the drug safety was good. Figures 17E-17H and as shown in the table below.

[0587] Table 38: Effect of antibodies on body weight of NCG mice inoculated with PBMC / HT-29

[0588]

[0589]

[0590] Other Implementations

[0591] It should be understood that although the present invention has been described in conjunction with its detailed description, the above description is intended to illustrate rather than limit the scope of the present invention, and the scope of the present invention is defined by the scope of the appended claims. Other aspects, advantages and modifications are within the scope of the appended claims.

Claims

1. An antigen binding protein comprising (a) Fc; (b) a Fab fragment (Fab) that specifically binds to a T cell antigen; and (c) a first single domain antibody variable domain (VHH) that specifically binds to a first tumor-associated antigen; (d) a second single domain antibody variable domain (VHH) that specifically binds to a second tumor-associated antigen; in, The Fab, the first VHH and the second VHH are linked to the Fc.

2. The antigen-binding protein of claim 1, wherein the Fab comprises or consists of a light chain variable domain (VL), a light chain constant domain (CL), a heavy chain variable domain (VH) and a heavy chain first constant domain (CH1).

3. The antigen binding protein according to claim 2, wherein the Fab can activate T cells after binding to a T cell antigen.

4. The antigen binding protein according to any one of claims 1 to 3, wherein the T cell antigen is cluster of differentiation 3 (CD3).

5. The antigen binding protein according to any one of claims 1 to 4, wherein the first tumor-associated antigen and the second tumor-associated antigen are independently selected from: cluster of differentiation 20 (CD20), carcinoembryonic antigen (CEA), prostate specific antigen (PSA), prostate stem cell antigen (PSCA), programmed death ligand 1 (PD-L1), human epidermal growth factor receptor 2 (HER2), human epidermal growth factor receptor 3 (Her3), human epidermal growth factor receptor (Her1), β-catenin, cluster of differentiation 19 (CD19), epidermal growth factor receptor (EGFR), tyrosine protein kinase Met (c-Met), epithelial cell adhesion molecule (EPCAM), prostate specific membrane antigen (PSMA), cluster of differentiation 40 (CD40), cell surface-associated mucin 1 (MUC1), insulin-like growth factor 1 receptor (IGF1R) and carcinoembryonic antigen cell adhesion molecule 6 (CEA-CAM6).

6. The antigen binding protein according to any one of claims 1-5, wherein the Fc is human IgG4 Fc.

7. The antigen binding protein according to any one of claims 2 to 6, wherein the CH1 domain of the Fab is connected to the CH2 domain of the Fc, optionally via a hinge region.

8. The antigen binding protein of claim 7, wherein the hinge region is a human IgG4 hinge region, optionally with an S228P mutation according to EU numbering.

9. The antigen binding protein of any one of claims 1-8, wherein the first VHH is connected to the CH2 domain in the Fc, optionally via a hinge region.

10. The antigen binding protein of any one of claims 1-8, wherein the first VHH is linked to the CH3 domain in the Fc, optionally via a linker sequence.

11. The antigen binding protein of any one of claims 1-10, wherein the second VHH is linked to the CH3 domain in the Fc, optionally via a linker sequence.

12. The antigen binding protein of any one of claims 1-10, wherein the second VHH is linked to the Fab.

13. The antigen binding protein of claim 12, wherein the second VHH is linked to the C-terminus of the CL domain in the Fab. The antigen binding protein of claim 12 , wherein the second VHH is linked to the N-terminus of the VH domain in the Fab.

15. The antigen binding protein of claim 12, wherein the second VHH is linked to the N-terminus of the VL domain in the Fab.

16. A protein complex comprising: (a) a first polypeptide comprising, in the N-terminal to C-terminal direction: a VH, a CH1 domain, optionally a first hinge region, a first CH2 domain and a first CH3 domain; (b) a second polypeptide comprising, in the N-terminal to C-terminal direction: a first VHH, optionally a second hinge region, a second CH2 domain, and a second CH3 domain; (c) a third polypeptide comprising, in the N-terminal to C-terminal direction: a VL and a CL domain; wherein the VH and the VL bind to each other to form an antigen binding site of the Fab that specifically binds to a T cell antigen, wherein the first VHH specifically binds to a first tumor-associated antigen.

17. The protein complex according to claim 16, wherein the third polypeptide comprises in the N-terminal to C-terminal direction: a VL, a CL domain and a second VHH, wherein the second VHH specifically binds to a second tumor-associated antigen.

18. The protein complex of claim 16, wherein the third polypeptide comprises, in the N-terminal to C-terminal direction: a second VHH, a VL and a CL domain, wherein the second VHH specifically binds to a second tumor-associated antigen.

19. The protein complex of claim 16, wherein the first polypeptide comprises, in the N-terminal to C-terminal direction: a second VHH, a VH, a CH1 domain, optionally a first hinge region, a first CH2 domain, and a first CH3 domain, wherein the second VHH specifically binds to a second tumor-associated antigen.

20. The protein complex of claim 16, wherein the second polypeptide comprises, in the N-terminal to C-terminal direction: a first VHH, a second VHH, optionally a second hinge region, a second CH2 domain, and a second CH3 domain, wherein the second VHH specifically binds to a second tumor-associated antigen.

21. The protein complex of claim 16, wherein the second polypeptide comprises, in the N-terminal to C-terminal direction: a first VHH, optionally a second hinge region, a second CH2 domain, a second CH3 domain, and a second VHH, wherein the second VHH specifically binds to a second tumor-associated antigen.

22. The protein complex of claim 16, wherein the first polypeptide comprises, in the N-terminal to C-terminal direction: a first VHH, a second VHH, optionally a second hinge region, a second CH2 domain, a second CH3 domain, and a second VHH, wherein the second VHH specifically binds to a second tumor-associated antigen.

23. A protein complex according to any one of claims 16-22, wherein the first polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 1; the second polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 2; and the third polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO:

3.

24. The protein complex according to any one of claims 16-23, wherein the T cell antigen is CD3, and the first tumor-associated antigen is selected from the group consisting of: CEA, CEA-CAM6, HER2 and EGFR.

25. The protein complex according to any one of claims 16-24, wherein the T cell antigen is CD3, and the second tumor-associated antigen is selected from the group consisting of: CEA, CEA-CAM6, HER2 and EGFR.

26. A protein complex comprising: (a) a first polypeptide comprising, in the N-terminal to C-terminal direction: a first VHH, optionally a first hinge region, a first CH2 domain and a first CH3 domain; (b) a second polypeptide comprising, in the N-terminal to C-terminal direction: a VH, a CH1 domain, optionally a second hinge region, a second CH2 domain, and a second CH3 domain; (c) a third polypeptide comprising, in the N-terminal to C-terminal direction: a VL and a CL domain; wherein the VH and the VL bind to each other to form an antigen binding site of the Fab that specifically binds to a T cell antigen, wherein the first VHH specifically binds to a first tumor-associated antigen.

27. The protein complex of claim 26, wherein the third polypeptide comprises in the N-terminal to C-terminal direction: a VL, a CL domain and a second VHH, wherein the second VHH specifically binds to a second tumor-associated antigen.

28. The protein complex of claim 26, wherein the third polypeptide comprises in the N-terminal to C-terminal direction: a second VHH, a VL and a CL domain, wherein the second VHH specifically binds to a second tumor-associated antigen.

29. The protein complex of claim 26, wherein the second polypeptide comprises, in the N-terminal to C-terminal direction: a second VHH, a VH, a CH1 domain, optionally a first hinge region, a first CH2 domain, and a first CH3 domain, wherein the second VHH specifically binds to a second tumor-associated antigen.

30. The protein complex of claim 26, wherein the first polypeptide comprises, in the N-terminal to C-terminal direction: a first VHH, a second VHH, optionally a second hinge region, a second CH2 domain, and a second CH3 domain, wherein the second VHH specifically binds to a second tumor-associated antigen.

31. The protein complex of claim 26, wherein the second polypeptide comprises, in the N-terminal to C-terminal direction: a VH, a CH1 domain, optionally a first hinge region, a first CH2 domain, a first CH3 domain, and a second VHH, wherein the second VHH specifically binds to a second tumor-associated antigen.

32. The protein complex of claim 26, wherein the first polypeptide comprises, in the N-terminal to C-terminal direction: a first VHH, optionally a second hinge region, a second CH2 domain, a second CH3 domain, and a second VHH, wherein the second VHH specifically binds to a second tumor-associated antigen.

33. A protein complex according to any one of claims 26-32, wherein the first polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 1; the second polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 2; and the third polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO:

3.

34. The protein complex according to any one of claims 26-33, wherein the T cell antigen is CD3, and the first tumor-associated antigen is selected from: CEA, CEA-CAM6, HER2 and EGFR.

35. The protein complex according to any one of claims 26-34, wherein the T cell antigen is CD3, and the second tumor-associated antigen is selected from the group consisting of: CEA, CEA-CAM6, HER2 and EGFR.

36. A protein complex comprising (a) a first polypeptide comprising, in the N-terminal to C-terminal direction: a VH, a CH1 domain, optionally a first hinge region, a first CH2 domain, a first CH3 domain and a first VHH; (b) a second polypeptide comprising, in the N-terminal to C-terminal direction: optionally a second hinge region, a second CH2 domain, a second CH3 domain, and a second VHH; and (c) a third polypeptide comprising, in the N-terminal to C-terminal direction: a VL and a CL domain, The VH and the VL bind to each other to form an antigen binding site of the Fab that specifically binds to a T cell antigen, wherein the first VHH specifically binds to a first tumor-associated antigen, and the second VHH specifically binds to a second tumor-associated antigen.

37. A protein complex according to claim 36, wherein the first polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 1; the second polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 2; and the third polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO:

3.

38. The protein complex according to any one of claims 36-37, wherein the T cell antigen is CD3, and the first tumor-associated antigen is selected from: CEA, CEA-CAM6, HER2 and EGFR.

39. The protein complex according to any one of claims 36-38, wherein the T cell antigen is CD3, and the second tumor-associated antigen is selected from: CEA, CEA-CAM6, HER2 and EGFR.

40. A protein complex comprising (a) a first polypeptide comprising, in the N-terminal to C-terminal direction: an optional first hinge region, a first CH2 domain and a first CH3 domain, and a first VHH; (b) a second polypeptide comprising, in the N-terminal to C-terminal direction: a VH, a CH1 domain, optionally a second hinge region, a second CH2 domain, a second CH3 domain, and a second VHH; and (c) a third polypeptide comprising, in the N-terminal to C-terminal direction: a VL and a CL domain, The VH and the VL bind to each other to form an antigen binding site of the Fab that specifically binds to a T cell antigen, wherein the first VHH specifically binds to a first tumor-associated antigen, and the second VHH specifically binds to a second tumor-associated antigen.

41. A protein complex according to claim 40, wherein the first polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 1; the second polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO: 2; and the third polypeptide comprises a sequence that is at least 80%, 90%, 95% or 100% identical to SEQ ID NO:

3.

42. The protein complex according to any one of claims 40-41, wherein the T cell antigen is CD3, and the first tumor-associated antigen is selected from: CEA, CEA-CAM6, HER2 and EGFR.

43. The protein complex according to any one of claims 40-42, wherein the T cell antigen is CD3, and the second tumor-associated antigen is selected from: CEA, CEA-CAM6, HER2 and EGFR.

44. The protein complex of any one of claims 16-43, wherein the first CH3 domain comprises one or more knob mutations and the second CH3 domain comprises one or more hole mutations.

45. The protein complex of any one of claims 16-44, wherein the VH comprises a sequence at least 80%, 90%, 95% or 100% identical to SEQ ID NO:5, and the VL comprises a sequence at least 80%, 90%, 95% or 100% identical to SEQ ID NO:

6.

46. ​​The protein complex of any one of claims 16-45, wherein the VHH comprises a sequence that is at least 80%, 90%, 95% or 100% identical to any one of SEQ ID NOs: 7-10.

47. The protein complex of any one of claims 16-46, wherein the VHH is connected to the CH3 domain via a linker peptide.

48. A nucleic acid comprising a polynucleotide encoding the antigen binding protein of any one of claims 1-15 or the protein complex of any one of claims 16-47.

49. The nucleic acid of claim 48, wherein the nucleic acid is DNA (eg, cDNA) or RNA (eg, mRNA).

50. A vector comprising one or more of the nucleic acids of claim 48 or 49.

51. A cell comprising the vector of claim 50.

52. The cell according to claim 51, wherein the cell is a HEK293F cell or a CHO cell.

53. A cell comprising one or more of the nucleic acids of claim 48 or 49.

54. A method for producing an antigen binding protein or protein complex, the method comprising (a) culturing the cell of any one of claims 51-53 under conditions sufficient for the cell to produce the antigen binding protein or protein complex; and (b) collecting the antigen-binding protein or protein complex produced by the cells.

55. An antibody-drug conjugate comprising the antigen binding protein of any one of claims 1-15 or the protein complex of any one of claims 16-47 covalently bound to a therapeutic agent.

56. The antibody-drug conjugate of claim 55, wherein the therapeutic agent is a cytotoxic agent or a cytostatic agent.

57. A method of treating a subject having cancer, the method comprising administering to the subject a therapeutically effective amount of a composition comprising the antigen binding protein of any one of claims 1-15, the protein complex of any one of claims 16-47, or the antibody-drug conjugate of claim 55 or 56.

58. The method of claim 57, wherein the subject has a cancer that expresses CEA-CAM6.

59. The method of claim 57 or 58, wherein the cancer is lung cancer, colorectal cancer, head and neck cancer, gastric cancer, pancreatic cancer, urothelial cancer, breast cancer, cervical cancer, or endometrial cancer.

60. A method for reducing tumor growth rate, the method comprising The tumor cells are contacted with an effective amount of a composition comprising the antigen binding protein of any one of claims 1-15, the protein complex of any one of claims 16-47, or the antibody-drug conjugate of claim 55 or 56.

61. A method for killing tumor cells, the method comprising The tumor cells are contacted with an effective amount of a composition comprising the antigen binding protein of any one of claims 1-15, the protein complex of any one of claims 16-47, or the antibody-drug conjugate of claim 55 or 56.

62. A pharmaceutical composition comprising the antigen binding protein of any one of claims 1-15 or the protein complex of any one of claims 16-47, and a pharmaceutically acceptable carrier.

Images