Anti-CD3 multispecific antibody and method of use thereof

Abstract

This disclosure provides antibodies that bind to human CD3 and antigen-binding fragments thereof, pharmaceutical compositions comprising the antibody or its antigen-binding fragment, and the use of the antibody or its antigen-binding fragment or the composition for treating diseases such as cancer.

Description

[Technical Field]

[0001] Cross-reference of related applications This application claims priority to the international application PCT / CN2023 / 079817, filed on 6 March 2023, which is incorporated herein by reference in its entirety.

[0002] This specification discloses antibodies or antigen-binding fragments that bind to human differentiation cluster 3 (human CD3), multispecific antibodies or antigen-binding fragments that bind to human claudin 6 (CLDN6) and human CD3, and methods for producing them. Specifically, this disclosure provides, among other things, pharmaceutical compositions comprising antibodies or antigen-binding fragments, and methods for treating cancer. [Background technology]

[0003] The following background information on this technology is provided to assist in understanding this technology and is not intended to describe or constitute prior art.

[0004] CD3 bispecific antibodies (BsAbs) are a newly emerging treatment in the field of cancer immunotherapy. CD3 BsAbs act by simultaneously binding to tumor-associated antigens expressed on tumor cells and CD3 expressed on T cells. By cross-linking these two cell types, CD3 BsAbs enable the formation of immunological synapses independent of MHC restriction, leading to T cell activation and subsequent anti-tumor immune responses (Kamakura et al., Pharmaceuticals (Basel). 2021). Currently, CD3-BsAbs show great potential for hematological malignancies, and promising early clinical data have been reported for solid tumors. One of the main obstacles to the development of CD3 BsAbs for solid tumors is the identification of cell surface targets with high cancer-specific expression that would enable efficient tumor eradication and a low risk of on-target off-tumor toxicity (Middelburg et al., Cancers (Basel). 2021; Singh et al., Br J Cancer. 2021; Baeuerle et al., Current Opinion in Oncology. 2022). [Overview of the project]

[0005] This disclosure provides anti-CD3 antibodies and antigen-binding fragments thereof, including humanized anti-CD3 antibodies with improved binding affinity to human CD3 and antigen-binding fragments thereof. This disclosure encompasses the following embodiments.

[0006] In some embodiments, the disclosure provides an antibody or an antigen-binding fragment thereof, comprising an antigen-binding domain that specifically binds to human differentiation cluster 3 (CD3).

[0007] In some embodiments, the antigen-binding domain that specifically binds to human CD3 includes a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 48, (b) HCDR2 of SEQ ID NO: 71, and (c) HCDR3 of SEQ ID NO: 50, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 51, (e) LCDR2 of SEQ ID NO: 52, and (f) LCDR3 of SEQ ID NO: 53; or a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 48, (b) HCDR2 of SEQ ID NO: 71, and (c) HCDR3 of SEQ ID NO: 75, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 51, (e) LCDR2 of SEQ ID NO: 52, and (f) LCDR3 of SEQ ID NO: 53.

[0008] In some embodiments, the antigen-binding domain is: a heavy chain variable region (VH) containing an amino acid sequence having at least 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% sequence identity to SEQ ID NO: 58; and a light chain variable region (VL) containing an amino acid sequence having at least 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% sequence identity to SEQ ID NO: 59; and at least 90, 91, 92, 9 A heavy chain variable region (VH) containing an amino acid sequence having 3, 94, 95, 96, 97, 98, or 99% sequence identity, and a light chain variable region (VL) containing an amino acid sequence having at least 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% sequence identity to SEQ ID NO: 63; a heavy chain variable region containing an amino acid sequence having at least 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% sequence identity to SEQ ID NO: 62. Light chain variable region (VL) containing an amino acid sequence having at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% sequence identity with SEQ ID NO: 68; Heavy chain variable region (VH) containing an amino acid sequence having at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% sequence identity with SEQ ID NO: 72, and at least 90, 91, 92, 93, 94, 95, 96, 97, 9 8. A light chain variable region (VL) containing an amino acid sequence having 99% sequence identity; or a heavy chain variable region (VH) containing an amino acid sequence having at least 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% sequence identity to SEQ ID NO: 76, and a light chain variable region (VL) containing an amino acid sequence having at least 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% sequence identity to SEQ ID NO: 68.

[0009] In some embodiments, one, two, three, four, five, six, seven, eight, nine, or ten amino acids in sequence numbers 62, 63, 68, 72, or 76 are inserted, deleted, or substituted.

[0010] In some embodiments, the antigen-binding domain includes: a heavy chain variable region (VH) containing the sequence of SEQ ID NO: 58 and a light chain variable region (VL) containing the sequence of SEQ ID NO: 59; a heavy chain variable region (VH) containing the sequence of SEQ ID NO: 62 and a light chain variable region (VL) containing the sequence of SEQ ID NO: 63; a heavy chain variable region (VH) containing the sequence of SEQ ID NO: 62 and a light chain variable region (VL) containing the sequence of SEQ ID NO: 68; a heavy chain variable region (VH) containing the sequence of SEQ ID NO: 72 and a light chain variable region (VL) containing the sequence of SEQ ID NO: 68; or a heavy chain variable region (VH) containing the sequence of SEQ ID NO: 76 and a light chain variable region (VL) containing the sequence of SEQ ID NO: 68.

[0011] In some embodiments, the antigen-binding domain includes: a single-stranded variable fragment (scFv) containing the sequence of SEQ ID NO: 66; a single-stranded variable fragment (scFv) containing the sequence of SEQ ID NO: 69; a single-stranded variable fragment (scFv) containing the sequence of SEQ ID NO: 73; or a single-stranded variable fragment (scFv) containing the sequence of SEQ ID NO: 77.

[0012] In some embodiments, the antibody or antigen-binding fragment is a monoclonal antibody, a chimeric antibody, a humanized antibody, a human-modified antibody, a single-chain antibody (scFv), a Fab fragment, a Fab' fragment, an F(ab')2 fragment, or a multispecific antibody.

[0013] In some embodiments, the antibody is a bispecific antibody.

[0014] In some embodiments, the antibody is BG143P and has a sequence including SEQ ID NO: 80, SEQ ID NO: 82, or SEQ ID NO: 84.

[0015] In some embodiments, the antibody or its antigen-binding fragment has antibody-dependent cell-mediated cytotoxicity (ADCC) or complement-dependent cell-mediated cytotoxicity (CDC).

[0016] In some embodiments, the antibody or its antigen-binding fragment is either deglycosylated, not glycosylated, or defucosylated.

[0017] In some embodiments, the antibody or antigen-binding fragment thereof comprises an increase in a bisecting GlcNac structure.

[0018] In some embodiments, the Fc domain is IgG1 with reduced effector function.

[0019] In some embodiments, the Fc domain is IgG4.

[0020] In some aspects, the present disclosure provides a pharmaceutical composition comprising the antibody or antigen-binding fragment disclosed herein according to any one of claims 1 to 14.

[0021] In some embodiments, the pharmaceutical composition comprises a pharmaceutically acceptable carrier.

[0022] In some embodiments, the pharmaceutical composition further comprises histidine / histidine HCl, trehalose dihydrate, and / or polysorbate 20.

[0023] In some aspects, the present disclosure provides a method of treating cancer, comprising administering to a patient in need thereof an effective amount of the antibody or antigen-binding fragment disclosed herein.

[0024] In some embodiments, the cancer is a solid tumor.

[0025] In some embodiments, the cancer is selected from gastric cancer, colon cancer, pancreatic cancer, breast cancer, head and neck cancer, kidney cancer, liver cancer, lung cancer, small cell lung cancer, non-small cell lung cancer, ovarian cancer, skin cancer, mesothelioma, lymphoma, leukemia, myeloma, sarcoma, brain cancer, colorectal cancer, prostate cancer, cervical cancer, testicular cancer, endometrial cancer, bladder cancer, rhabdomyosarcoma, and / or glioma.

[0026] In some embodiments, the antibody or antigen-binding fragment thereof is administered in combination with one or more additional therapeutic agents.

[0027] In some embodiments, one or more therapeutic agents are selected from paclitaxel or paclitaxel preparations, docetaxel, carboplatin, topotecan, cisplatin, irinotecan, doxorubicin, lenalidomide, or 5-azacitidine.

[0028] In some embodiments, one or more therapeutic agents are paclitaxel, lenalidomide, or 5-azacitidine.

[0029] In some embodiments, at least one of the one or more therapeutic agents is an anti-PD1 antibody or an anti-PDL1 antibody.

[0030] In some embodiments, the anti-PD-1 antibody is tislerizumab.

[0031] In some embodiments, the disclosure provides isolated nucleic acids encoding antibody or antigen-binding fragments disclosed herein.

[0032] In some embodiments, the disclosure provides vectors comprising nucleic acids disclosed herein.

[0033] In some embodiments, the disclosure provides a host cell comprising a nucleic acid or vector disclosed herein.

[0034] In some embodiments, the Disclosure provides a process for producing an antibody or antigen-binding fragment as disclosed herein, comprising culturing a host cell as disclosed herein and recovering an antibody or antigen-binding fragment from the culture.

[0035] In some embodiments, antibodies or antigen-binding fragments are used in methods for treating cancer.

[0036] In some embodiments, antibodies or antigen-binding fragments are used in the manufacture of pharmaceuticals for treating cancer.

[0037] In some embodiments, the pharmaceutical composition is used in a method for treating cancer.

[0038] An antibody or its antigen-binding fragment, comprising an antigen-binding domain that specifically binds to human CD3.

[0039] The antibody or antigen-binding fragment disclosed herein, wherein the antigen-binding domain specifically binds to human CD3, is: (i) A heavy chain variable region including (a) HCDR1 of SEQ ID NO: 48, (b) HCDR2 of SEQ ID NO: 71, (c) HCDR3 of SEQ ID NO: 50, and a light chain variable region including (d) LCDR1 of SEQ ID NO: 51, (e) LCDR2 of SEQ ID NO: 52, and (f) LCDR3 of SEQ ID NO: 53; or (ii) A heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 48, (b) HCDR2 of SEQ ID NO: 71, and (c) HCDR3 of SEQ ID NO: 75, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 51, (e) LCDR2 of SEQ ID NO: 52, and (f) LCDR3 of SEQ ID NO: 53.

[0040] An antibody or antigen-binding fragment disclosed herein, wherein the antigen-binding domain is: (i) A heavy chain variable region (VH) containing at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequences to SEQ ID NO: 58, and a light chain variable region (VL) containing at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequences to SEQ ID NO: 59; (ii) A heavy chain variable region (VH) containing at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequences to SEQ ID NO: 62, and a light chain variable region (VL) containing at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequences to SEQ ID NO: 63; (iii) A heavy chain variable region (VH) containing at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequences to SEQ ID NO: 62, and a light chain variable region (VL) containing at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequences to SEQ ID NO: 68; (iv) A heavy chain variable region (VH) containing at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequences to SEQ ID NO: 72, and a light chain variable region (VL) containing at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequences to SEQ ID NO: 68; or (v) A heavy chain variable region (VH) comprising an amino acid sequence that is at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical to SEQ ID NO: 76, and a light chain variable region (VL) comprising an amino acid sequence that is at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical to SEQ ID NO: 68.

[0041] An antibody or antigen-binding fragment disclosed herein, wherein 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acids in SEQ ID NOs. 62, 63, 68, 72, or 76 are inserted, deleted, or substituted.

[0042] An antibody or antigen-binding fragment disclosed herein, wherein the antigen-binding domain is: (i) A heavy chain variable region (VH) containing sequence number 58, and a light chain variable region (VL) containing sequence number 59; (ii) A heavy chain variable region (VH) containing sequence number 62, and a light chain variable region (VL) containing sequence number 63; (iii) A heavy chain variable region (VH) containing sequence number 62, and a light chain variable region (VL) containing sequence number 68; (iv) A heavy chain variable region (VH) containing sequence number 72 and a light chain variable region (VL) containing sequence number 68; or (v) A heavy chain variable region (VH) containing sequence number 76 and a light chain variable region (VL) containing sequence number 68.

[0043] An antibody or antigen-binding fragment disclosed herein, wherein the antigen-binding domain is: (i) A single-stranded variable fragment (scFv) containing sequence number 66, (ii) Single-stranded variable fragment (scFv) containing sequence number 69, (iii) Single-stranded variable fragment (scFv) containing sequence number 73, (iv) A single-stranded variable fragment (scFv) containing sequence number 77.

[0044] The antibodies or antigen-binding fragments disclosed herein include monoclonal antibodies, chimeric antibodies, humanized antibodies, human-modified antibodies, single-chain antibodies (scFv), Fab fragments, Fab' fragments, F(ab')2 fragments, or multispecific antibodies.

[0045] The antibodies disclosed herein are bispecific antibodies.

[0046] The antibody according to claim 8, wherein the antibody is BG143P (SEQ ID NO: 80, SEQ ID NO: 82, and SEQ ID NO: 84).

[0047] Antibodies or antigen-binding fragments disclosed herein, wherein the antibody or antigen-binding fragment has antibody-dependent cell-mediated cytotoxicity (ADCC) or complement-dependent cell-mediated cytotoxicity (CDC).

[0048] Antibodies or antigen-binding fragments disclosed herein, wherein the antibody or antigen-binding fragment is reduced in glycosylation, not glycosylated, or low in fucosylation.

[0049] An antibody or antigen-binding fragment disclosed herein, wherein the antibody or antigen-binding fragment comprises an increase in the bisecting GlcNac structure.

[0050] An antibody or antigen-binding fragment disclosed herein, wherein the Fc domain is IgG1 with reduced effector function.

[0051] An antibody or antigen-binding fragment disclosed herein, wherein the Fc domain is IgG4.

[0052] A pharmaceutical composition comprising an antibody or antigen-binding fragment disclosed herein, further comprising a pharmaceutically acceptable carrier.

[0053] The pharmaceutical composition according to claim 15, further comprising histidine / histidine HCl, trehalose dihydrate, and polysorbate 20.

[0054] A method for treating cancer, comprising administering an effective amount of an antibody or antigen-binding fragment disclosed herein to a patient in need.

[0055] Methods disclosed herein, wherein cancer is gastric cancer, colon cancer, pancreatic cancer, breast cancer, head and neck cancer, kidney cancer, liver cancer, lung cancer, small cell lung cancer, non-small cell lung cancer, ovarian cancer, skin cancer, mesothelioma, lymphoma, leukemia, myeloma, and sarcoma.

[0056] A method disclosed herein involves administering an antibody or antigen-binding fragment in combination with another therapeutic agent.

[0057] A method disclosed herein, wherein the therapeutic agent is paclitaxel or a paclitaxel preparation, docetaxel, carboplatin, topotecan, cisplatin, irinotecan, doxorubicin, lenalidomide, or 5-azacitidine.

[0058] A method disclosed herein, wherein the therapeutic agent is paclitaxel, lenalidomide, or 5-azacitidine.

[0059] A method disclosed herein, wherein the therapeutic agent is an anti-PD1 antibody or an anti-PDL1 antibody.

[0060] A method disclosed herein, wherein the anti-PD1 antibody is tislerizumab.

[0061] Isolated nucleic acids encoding an antibody or antigen-binding fragment disclosed herein.

[0062] A vector comprising nucleic acids disclosed herein.

[0063] A host cell containing a nucleic acid or vector disclosed herein.

[0064] A process for producing an antibody or an antigen-binding fragment thereof, comprising culturing host cells as disclosed herein and recovering the antibody or antigen-binding fragment from the culture.

[0065] In another embodiment, the multispecific antibody or its antigen-binding fragment includes a heavy chain variable region containing three complementarity-determining regions (HCDRs), HCDR includes HCDR1 containing the amino acid sequence of SEQ ID NO: 47; HCDR2 containing the amino acid sequence of SEQ ID NO: 71; and HCDR3 containing the amino acid sequence of SEQ ID NO: 50; or These are HCDR1 containing the amino acid sequence of SEQ ID NO: 48; HCDR2 containing the amino acid sequence of SEQ ID NO: 71; and HCDR3 containing the amino acid sequence of SEQ ID NO: 75.

[0066] In another embodiment, the multispecific antibody or antigen-binding fragment is: A heavy chain variable region including (a) HCDR1 of SEQ ID NO: 48, (b) HCDR2 of SEQ ID NO: 71, (c) HCDR3 of SEQ ID NO: 50, and a light chain variable region including (d) LCDR1 of SEQ ID NO: 51, (e) LCDR2 of SEQ ID NO: 52, and (f) LCDR3 of SEQ ID NO: 53; or The device includes a heavy chain variable region comprising (a) HCDR1 of sequence number 48, (b) HCDR2 of sequence number 71, and (c) HCDR3 of sequence number 75, and a light chain variable region comprising (d) LCDR1 of sequence number 51, (e) LCDR2 of sequence number 52, and (f) LCDR3 of sequence number 53.

[0067] In one embodiment, the multispecific antibody of the present disclosure is of the isotype IgG1, IgG2, IgG3, or IgG4. In a more specific embodiment, the antibody of the present disclosure comprises the Fc domain of wild-type human IgG1 (also referred to as human IgG1wt or huIgG1) or IgG2.

[0068] In one embodiment, the multispecific antibody of the present disclosure binds to CD3 with a binding affinity (K -6 ) of 1×10 -10 M to 1×10 D M. In another embodiment, the antibody of the present disclosure binds to CD3 with a binding affinity (K -6 ) of about 1×10 -7 M, about 1×10 -8 M, about 1×10 -9 M, about 1×10 -10 M, or about 1×10 D M.

[0069] In another embodiment, the anti-human CD3 multispecific antibody of the present disclosure exhibits cross-species binding activity against cynomolgus CD3.

[0070] In one embodiment, the antibody of the present disclosure has strong Fc-mediated effector functions. The antibody mediates antibody-dependent cell cytotoxicity (ADCC) against target cells expressing CD3.

[0071] In another aspect, the present disclosure relates to a pharmaceutical composition comprising an anti-CD3 antibody or an antigen-binding fragment thereof, and optionally a pharmaceutically acceptable excipient.

[0072] In yet another aspect, the present disclosure relates to a method of treating a disease in a subject, comprising administering to the subject a therapeutically effective amount of an anti-CD3 antibody or an antigen-binding fragment thereof, or an anti-CD3 antibody pharmaceutical composition. In another embodiment, the disease treated by the antibody or antigen-binding fragment is cancer.

[0073] The present disclosure relates to the use of an anti-CD3 antibody or an antigen-binding fragment thereof, or an anti-CD3 antibody pharmaceutical composition, for treating diseases such as cancer. [Brief explanation of the drawing]

[0074] [Figure 1A] This shows the cell binding activity of modified chBG87P mutants. It shows the cell binding activity of the first round of BG87P humanized revertant variants (BG87P-z0, BG87P-Bz0, BG87P-Bz1, BG87P-Bz2, BG87P-Bz3, BG87P-Bz4, BG87P-Bz5, BG87P-Bz6, BG87P-Bz7, and BG87P-Bz8) against HEK293T / human CLDN6 compared to anti-CLDN6 chimeric BG87P (chBG87P). [Figure 1B] This shows the cell-binding activity of modified chBG87P mutants. It also shows the cell-binding activity of combined humanized variants (BG87P-21, BG87P-22, BG87P-23, and BG87P-24) compared to anti-CLDN6 chimeric BG87P (chBG87P) against HEK293T / human CLDN6. [Figure 1C] This shows the cell-binding activity of modified chBG87P mutants. It also shows the cell-binding activity of combined humanized variants (BG87P-25, BG87P-26, and BG87P-27) compared to anti-CLDN6 chimeric BG87P (chBG87P) against HEK293T / human CLDN6. [Figure 1D] This shows the cell-binding activity of modified chBG87P mutants. It also shows the cell-binding activity of combined humanized variants (BG87P-21, BG87P-22, BG87P-23, and BG87P-24) compared to anti-CLDN6 chimeric BG87P (chBG87P) against the cancer cell line PA-1. [Figure 1E] This shows the cell-binding activity of modified chBG87P mutants. It also shows the cell-binding activity of combined humanized variants (BG87P-25, BG87P-26, and BG87P-27) compared to anti-CLDN6 chimeric BG87P (chBG87P) against the cancer cell line PA-1. [Figure 1F]This shows the cell binding activity of modified chBG87P mutants. It also shows the cell binding activity of post-translational modification (PTM) removal variants (BG87P-m1, BG87P-m2, BG87P-m3, BG87P-m4, BG87P-m5, BG87P-m6, BG87P-m7, and BG87P-m8) compared to anti-CLDN6 chimeric BG87P (chBG87P) and BG87P-Bz0 against HEK293T / human CLDN6. [Figure 1G] This shows the cell-binding activity of modified chBG87P mutants. It also shows the cell-binding activity of BG87P soluble variants (BG87P-21, BG87P-34, and BG87P-33) compared to anti-CLDN6 chimeric BG87P (chBG87P) against HEK293T / human CLDN6. [Figure 1H] This shows the cell binding activity of modified chBG87P mutants. It also shows the nonspecific binding activity of soluble variants (BG87P-21, BG87P-34, and BG87P-33) to HEK293T-human CLDN9 compared to anti-CLDN6 chimeric BG87P (chBG87P). [Figure 1I] This shows the cell-binding activity of modified chBG87P mutants. It also shows the cross-reactivity of humanized variants (BG87P-21, BG87P-34, and BG87P-33) compared to anti-CLDN6 chimeric BG87P (chBG87P) against CHOK1-cyno CLDN6. [Figure 1J] This shows the cell-binding activity of modified chBG87P mutants. It also shows the cross-reactivity of humanized variants (BG87P-21, BG87P-34, and BG87P-33) compared to anti-CLDN6 chimeric BG87P (chBG87P) against CHOK1-mouse CLDN6. [Figure 2] The expected hydrophobic patch in Schroedinger's homology model of chimeric BG87P is illustrated. I97-Y98-Y100-V100a of HCDR3 is expected to form an exposed hydrophobic patch together with Y49-W50 of HCDR2 (Y49 is the last residue of FR2 in the variable region of the light chain, and W50 is the first residue of HCDR2). [Figure 3]The hydrophobicity of the selected humanized BG87P variants (BG87-33, BG87-34, BG87P-21) after manipulation, as determined by HIC-HPLC, is shown. [Figure 4A] This shows a comparison of binding activity between chimeric sp34 and humanized sp34 in Hut78 cells. It also shows a comparison of binding affinity between chimeric sp34 (ch-sp34) and humanized sp34 BG53P (BG53P) measured by melt-flow index (MFI) in Hut78 cells. [Figure 4B] This shows a comparison of binding activity between chimeric sp34 and humanized sp34 in Hut78 cells. It also shows a comparison of binding affinity between chimeric sp34 (ch-sp34), humanized sp34 BG53P (BG53P), and BG56P. [Figure 5] This shows a comparison of the binding activity between humanized sp34 BG56P (BG56P) and humanized sp34 scFv BG561p (BG561P). [Figure 6A] This shows a comparison of the binding affinity of humanized sp34 scFv in Hut78 cells. This shows a comparison of the binding affinity between humanized sp34 scFv BG561p (BG561P) and humanized scFv BG562P (BG562P) in Hut78 cells. [Figure 6B] This shows a comparison of the binding affinity of humanized sp34 scFv in Hut78 cells. This also shows a comparison of the binding affinity between humanized scFv BG562P (BG562P) and humanized scFv BG563P (BG563P) in Hut78 cells. [Figure 6C] This shows a comparison of the binding affinity of humanized sp34 scFv in Hut78 cells. This also shows a comparison of the binding affinity between humanized scFv BG563P (BG563P) and humanized scFv BG564P (BG564P) in Hut78 cells. [Figure 7] A schematic diagram of CLDN6×CD3 BsAb BG143P is shown. [Figure 8A] This shows the target binding activity of CLDN6×CD3 BsAb BG143P. It also shows the CD3 binding activity of BG143P in CD3-expressing Jurkat cells. [Figure 8B] This shows the target binding activity of CLDN6 × CD3 BsAb BG143P. This shows the CLDN6 binding activity of BG143P in CLDN6-expressing PA-1 cells. [Figure 9A-1] This study demonstrates the on-target functional activity of CLDN6×CD3 BsAb BG143P in tumor cell lines with differing CLDN6 expression. Cell lysis assays show the cytotoxicity of BG143P in re-induced T cells of CLDN6-expressing PA-1 cells, Hutu80 cells, AGS cells, and NCI-H1299 cells. [Figure 9A-2] This study demonstrates the on-target functional activity of CLDN6×CD3 BsAb BG143P in tumor cell lines with differing CLDN6 expression. Cell lysis assays show the cytotoxicity of BG143P in re-induced T cells of CLDN6-expressing PA-1 cells, Hutu80 cells, AGS cells, and NCI-H1299 cells. [Figure 9B-1] This study demonstrates the on-target functional activity of CLDN6×CD3 BsAb BG143P in tumor cell lines with differing CLDN6 expression. It also shows the IFN-γ induction activity of BG143P in CLDN6-expressing PA-1 cells, Hutu80 cells, AGS cells, and NCI-H1299 cells. [Figure 9B-2] This study demonstrates the on-target functional activity of CLDN6×CD3 BsAb BG143P in tumor cell lines with differing CLDN6 expression. It also shows the IFN-γ induction activity of BG143P in CLDN6-expressing PA-1 cells, Hutu80 cells, AGS cells, and NCI-H1299 cells. [Figure 9C-1] This study demonstrates the on-target functional activity of CLDN6×CD3 BsAb BG143P in tumor cell lines with differing CLDN6 expression. It also shows the IL-2 inducing activity of BG143P in CLDN6-expressing PA-1 cells, Hutu80 cells, AGS cells, and NCI-H1299 cells. [Figure 9C-2]This study demonstrates the on-target functional activity of CLDN6×CD3 BsAb BG143P in tumor cell lines with differing CLDN6 expression. It also shows the IL-2 inducing activity of BG143P in CLDN6-expressing PA-1 cells, Hutu80 cells, AGS cells, and NCI-H1299 cells. [Figure 10A] This shows the functional specificity of CLDN6×CD3 BsAb BG143P to human CLDN6 and CLDN9. It also shows the binding specificity of BG143P to human CLDN6 (left graph) and CLDN9 (right graph) in NCI-H1299 cells. [Figure 10B] This shows the functional specificity of CLDN6×CD3 BsAb BG143P against human CLDN6 and CLDN9. It also shows the killing specificity (cytolytic activity) of BG143P against human CLDN6 (left graph) and CLDN9 (right graph) in NCI-H1299 cells. [Figure 10C] This demonstrates the functional specificity of CLDN6×CD3 BsAb BG143P to human CLDN6 and CLDN9. It also shows the cytokine (IFN-γ) induction by BG143P to human CLDN6 (left graph) and CLDN9 (right graph) in NCI-H1299 cells. [Figure 11A] This study demonstrates the in vivo efficacy of CLDN6×CD3 BsAb BG143P in an OV-90 xenograft model using PBMC-humanized mice. Tumor volume over time is shown. Mice were left untreated (no PBMC), treated with PBS (PBS ip QW), treated with 0.01 mg / kg BG143P (BG143P-0.01 mg / kg, ip), treated with 0.03 mg / kg BG143P (BG143P-0.01 mg / kg, ip), or treated with 0.1 mg / kg BG143P (BG143P-0.1 mg / kg, ip). Treatment was administered weekly and is represented by triangles on the x-axis. [Figure 11B]This study demonstrates the in vivo efficacy of CLDN6×CD3 BsAb BG143P in an OV-90 xenograft model using humanized PBMC mice. The percentage of hCD45+ cells in the peripheral blood of mice treated with 0.01 mg / kg BG143P (PBS ip QW), 0.01 mg / kg BG143P (BG143P-0.01 mg / kg, ip), 0.03 mg / kg BG143P (BG143P-0.01 mg / kg, ip), or 0.1 mg / kg BG143P (BG143P-0.1 mg / kg, ip) at 13, 21, and 27 days after PBMC injection is shown, indicating the reconstitution of human PBMCs. [Figure 12A] This study demonstrates the in vivo efficacy of CLDN6×CD3 BsAb BG143P in an aB16F10 / human CLDN6 syngeneic model in hCD3EDG transgenic mice. Tumor volume over time is shown. Mice were treated with PBS (PBS ip QW), 0.01 mg / kg BG143P (BG143P-0.01 mg / kg, ip), 0.03 mg / kg BG143P (BG143P-0.01 mg / kg, ip), or 0.1 mg / kg BG143P (BG143P-0.1 mg / kg, ip). Treatment was administered weekly and is represented by a triangle on the x-axis. [Figure 12B] This study demonstrates the in vivo efficacy of CLDN6×CD3 BsAb BG143P in an aB16F10 / human CLDN6 syngeneic model in hCD3EDG transgenic mice. As an indicator of mouse tolerance to the antibody, body weight is shown for mice treated with PBS (PBS ip QW), 0.01 mg / kg BG143P (BG143P-0.01 mg / kg, ip), 0.03 mg / kg BG143P (BG143P-0.01 mg / kg, ip), or 0.1 mg / kg BG143P (BG143P-0.1 mg / kg, ip) from day 11 to day 27 after inoculation.

[0075] definition Unless otherwise defined elsewhere in this document, all other technical and scientific terms used herein have the meanings generally understood by those skilled in the art.

[0076] As used herein, including in the appended claims, singular words such as "a," "an," and "the" refer to multiple corresponding subjects unless specifically indicated by the context.

[0077] As used herein, the term "or" means the term "and / or" unless the context explicitly indicates otherwise, and is used interchangeably. Also as used herein, "and / or" means and encompasses all possible combinations of one or more related enumerations, and, if interpreted as an alternative ("or"), the absence of any combination.

[0078] In this specification, "approximately" used with a number means both the stated number and a range of plus or minus 10% of that number. For example, "approximately 10" should be understood as both "10" and "9 to 11".

[0079] As used herein, the “A / B” form or the “A and / or B” form means (A), (B), or (A and B), and the form “at least one of A, B, and C” means (A), (B), (C), (A and B), (A and C), (B and C), or (A, B, and C).

[0080] As used herein, the term “anti-cancer agent” refers to any agent that can be used to treat cell proliferation disorders such as cancer, and includes, but is not limited to, cytotoxic agents, chemotherapeutic agents, radiotherapy and radiotherapeutic agents, targeted anti-cancer agents, and immunotherapeutic agents.

[0081] The term "Claudin 6" or "CLDN6" refers to a member of the CLDN family. CLDN6 has a molecular weight of 23 kDa. CLDN6 has four transmembrane domains and a PDZ-binding domain at the cytoplasmic carboxyl terminus. The amino acid sequence of human CLDN6 can be found at UniPort ID P56747. An exemplary human CLDN6 sequence is Sequence ID No. 87.

[0082] The term "Claudin 9" or "CLDN9" refers to another member of the CLDN family. CLDN9 has a molecular weight of 23 kDa, and its amino acid sequence can be found at UniPort ID O95484. An exemplary human CLDN9 sequence is Sequence ID No. 88.

[0083] The terms “differentiation cluster 3” or “CD3,” as used herein, refer to any native CD3 from any vertebrate source, including mammals such as primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise indicated, and include, for example, CD3ε, CD3γ, CD3α, and CD3β chains. The term encompasses “full-length” unprocessed CD3 (e.g., unprocessed or unmodified CD3ε or CD3γ), as well as any form of CD3 obtained from intracellular processing. The term also encompasses spontaneously occurring variants of CD3, including, for example, splice variants or allele variants. CD3 includes, for example, the human CD3ε protein with a length of 207 amino acids (NCBI reference sequence number NP_000724) and the human CD3γ protein with a length of 182 amino acids (NCBI reference sequence number NP_000064).

[0084] As used herein, the terms “administer,” “administer,” “treat,” and “treat” mean, when applied to animals, humans, subjects, cells, tissues, organs, or biological fluids, the contact of an exogenous medicinal, therapeutic, or diagnostic agent or composition to such animals, humans, subjects, cells, tissues, organs, or biological fluids. Treatment of cells includes the contact of a reagent with cells and, if the fluid is in contact with cells, the contact of a reagent with such fluid. The terms “administer” and “treat” also mean in vitro and ex vivo treatment of cells, for example, with a reagent, diagnostic agent, conjugate compound, or another cell. The term “subject” as used herein includes any living organism. A non-limiting example is an animal. In any embodiment, an animal is a mammal (e.g., primates, higher primates, humans, rats, mice, dogs, cats, rabbits). In any embodiment, a mammal is a human. In any embodiment, a subject is a patient who has or is at risk of having the disorder described herein. In any embodiment, treating any disease or disorder means improving the disease or disorder (i.e., delaying, preventing, or reducing the onset of at least one of the disease or its clinical symptoms). In another embodiment, “treating,” “treating,” or “treatment” means alleviating or improving at least one physical parameter, including those that may not be identifiable by the patient. In yet another embodiment, “treating,” “treating,” or “treatment” means regulating a disease or disorder physically (e.g., stabilizing identifiable symptoms), physiologically (e.g., stabilizing physical parameters), or both. In yet another embodiment, “treating,” “treating,” or “treatment” means preventing or delaying the onset, development, or progression of a disease or disorder. In some embodiment, the terms “preventing,” “preventing,” or “prevention” as used herein in relation to cancer mean eliminating or reducing the risk of developing cancer. Prevention may also mean preventing recurrence or secondary cancer after initial cancer has been treated or cured.

[0085] The terms “individual,” “subject,” and “patient” are used interchangeably herein and refer to any individual mammalian subject, e.g., a cow, a dog, a cat, a horse, or a human. In certain embodiments, the subject, individual, or patient is a human.

[0086] As used herein, the term "affinity" refers to the strength of the interaction between an antibody and an antigen. Within the antigen, the variable region of the antibody interacts with the antigen at numerous sites via forces other than covalent bonds. Generally, the more interactions there are, the stronger the affinity.

[0087] As used herein, the term “antibody” refers to a polypeptide of the immunoglobulin family that can bind to a corresponding antigen in a more reversible and specific manner than covalent bonding. For example, naturally occurring IgG antibodies are tetramers containing at least two heavy (H) chains and two light (L) chains interconnected by disulfide bonds. Each heavy chain consists of a heavy chain variable region (abbreviated herein as VH) and a heavy chain constant region. The heavy chain constant region consists of three domains CH1, CH2, and CH3. Each light chain consists of a light chain variable region (abbreviated herein as VL or Vκ) and a light chain constant region. The light chain constant region consists of one domain CL. The VH and VL regions can be further subdivided into hypervariable regions called complementarity-determining regions (CDRs), and interspersed with more conserved regions called framework regions (FRs). Each VH and VL consists of three CDRs and four framework regions (FRs), arranged in the order FR1, CDR1, FR2, CDR2, FR3, CDR3, and FR4 from the amino terminus to the carboxyl terminus. The variable regions of the heavy and light chains contain binding domains that interact with the antigen. The constant region of the antibody can mediate the binding of immunoglobulins to host tissues or factors, including various cells of the immune system (e.g., effector cells) and the first component of the classical complement system (C1q).

[0088] The term "antibody" includes, but is not limited to, monoclonal antibodies, human antibodies, humanized antibodies, chimeric antibodies, and anti-idiotype (anti-Id) antibodies. Antibodies can be of any isotype / class (e.g., IgG, IgE, IgM, IgD, IgA, and IgY) or subclass (e.g., IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2).

[0089] The term "chimeric" antibody refers to an antibody in which a portion of the heavy chain and / or light chain originates from a specific source or species, while the rest of the heavy chain and / or light chain originates from a different source or species.

[0090] The terms “full-length antibody,” “intact antibody,” and “whole antibody” are used interchangeably herein to refer to antibodies having a structure substantially similar to that of a natural antibody or having a heavy chain containing an Fc region.

[0091] In some embodiments, the anti-CD3 antibody comprises at least one antigen-binding site and at least a variable region. In some embodiments, the anti-CD3 antibody comprises an antigen-binding fragment derived from a CD3 antibody described herein. In some embodiments, the anti-CD3 antibody is isolated or recombinant.

[0092] In this specification, the terms “monoclonal antibody” or “mAb” or “Mab” refer to a substantially homogeneous population of antibodies, i.e., the antibody molecules in that population have identical amino acid sequences, except for spontaneously occurring mutations that may be present in small amounts. In contrast, conventional (polyclonal) antibody preparations typically contain a number of different antibodies having different amino acid sequences within variable domains, particularly complementarity-determining regions (CDRs), which are often specific to different epitopes. The modifier “monoclonal” characterizes the antibody as being obtained from a substantially homogeneous population of antibodies and is not to be interpreted as requiring antibody production by any particular method. Monoclonal antibodies (mAbs) can be obtained by methods known to those skilled in the art. See, for example, Kohler et al., Nature 1975 256:495-497, U.S. Patent No. 4,376,110, Ausubel et al., CURRENT PROTOCOLS IN MOLECULAR BIOLOGY 1992, Harlow et al., ANTIBODIES: A LABORATORY MANUAL, Cold Spring Harbor Laboratory 1988, and Colligan et al., CURRENT PROTOCOLS IN IMMUNOLOGY 1993. The antibodies disclosed herein may be any immunoglobulin class such as IgG, IgM, IgD, IgE, IgA, and any subclass thereof, e.g., IgG1, IgG2, IgG3, IgG4. Hybridomas producing monoclonal antibodies can be cultured in vitro or in vivo. High-titer monoclonal antibodies can be obtained through in vivo production. In this case, cells from individual hybridomas are intraperitoneally injected into mice, such as Balb / c mice, stimulated with pristine, to induce the production of ascites containing high concentrations of the desired antibody. From such ascites, or from the culture supernatant, isotype IgM or IgG monoclonal antibodies can be purified using column chromatography methods well known to those skilled in the art.

[0093] Generally, the basic structural unit of an antibody is a tetramer. Each tetramer contains two identical polypeptide chains, each pair having one "light chain" (approximately 25 kDa) and one "heavy chain" (approximately 50-70 kDa). The amino-terminus of each chain contains a variable region of approximately 100-110 or more amino acids, primarily involved in antigen recognition. The carboxyl-terminus of the heavy chain can define a constant region, primarily involved in effector function. Typically, human light chains are classified into kappa and lambda light chains. Furthermore, human heavy chains are usually classified as α, δ, ε, γ, or μ, and the isotypes of the antibody are defined as IgA, IgD, IgE, IgG, and IgM, respectively. Within the light and heavy chains, the variable and constant regions are linked by a "J" region of approximately 12 or more amino acids, and the heavy chain also contains a "D" region of approximately 10 amino acids.

[0094] The variable region of each light chain / heavy chain (VL / VH) pair forms the antibody binding site. Therefore, generally, intact antibodies have two binding sites. Except for bifunctional or bispecific antibodies, the two binding sites generally have the same primary sequence.

[0095] Typically, both the heavy and light chain variable domains contain three hypervariable regions, also called "complementarity-determining regions" or "CDRs," which are located between relatively conserved framework regions (FRs). CDRs are usually aligned by framework regions, enabling binding to specific epitopes. Generally, from the N-terminus to the C-terminus, both the light and heavy chain variable domains contain FR-1 (or FR1), CDR-1 (or CDR1), FR-2 (FR2), CDR-2 (CDR2), FR-3 (or FR3), CDR-3 (CDR3), and FR-4 (or FR4). The location of the CDR and framework region can be determined using various definitions well known in the art, such as Kabat, Chothia, AbM, and IMGT (e.g., Johnson et al., Nucleic Acids Res., 29:205-206 (2001); Chothia and Lesk, J.Mol.Biol., 196:901-917 (1987); Chothia et al., Nature, 342:877-883 (1989); Chothia et al., J.Mol.Biol., 227:799-817 (1992); Al-Lazikani et al., J.Mol.Biol., 273:927-748 (1997); ImMunoGenTics (IMGT) numbering (Lefranc, M.-P., The Immunologist, 7, 132-136 (1999); Lefranc, M.-P. et al., Dev. Comp. Immunol., 27, 55-77 (2003) (see "IMGT" numbering procedure).The definition of antigen-binding sites is also described in the following: Ruiz et al., Nucleic Acids Res., 28:219-221 (2000); and Lefranc, MP, Nucleic Acids Res., 29:207-209 (2001); MacCallum et al., J.Mol.Biol., 262:732-745 (1996); and Martin et al., Proc.Natl.Acad.Sci.USA, 86:9268-9272 (1989); Martin et al., Methods Enzymol., 203:121-153 (1991); and Rees et al., In Sternberg MJE (ed.), Protein Structure Prediction, Oxford University Press, Oxford, 141-172 (1996). For example, in Kabat, the CDR amino acid residues in the heavy chain variable domain (VH) are numbered 31-35 (HCDR1), 50-65 (HCDR2), and 95-102 (HCDR3), while the CDR amino acid residues in the light chain variable domain (VL) are numbered 24-34 (LCDR1), 50-56 (LCDR2), and 89-97 (LCDR3). In Chothia, the CDR amino acids in the VH are numbered 26-32 (HCDR1), 52-56 (HCDR2), and 95-102 (HCDR3), while the amino acid residues in the VL are numbered 26-32 (LCDR1), 50-52 (LCDR2), and 91-96 (LCDR3). Combining the definitions of CDRs by Kabat and Chothia, the CDR consists of amino acid residues 26-35 (HCDR1), 50-65 (HCDR2), and 95-102 (HCDR3) in human VH, and amino acid residues 24-34 (LCDR1), 50-56 (LCDR2), and 89-97 (LCDR3) in human VL.In IMGT, the CDR amino acid residues in the VH region are numbered approximately 26-35 (HCDR1), 51-57 (HCDR2), and 93-102 (HCDR3), while the CDR amino acid residues in the VL region are numbered approximately 27-32 (LCDR1), 50-52 (LCDR2), and 89-97 (LCDR3) (numbering according to Kabat). In IMGT, the CDR region of an antibody can be determined using the program IMGT / DomainGap Align.

[0096] The term "hypervariable region" refers to the amino acid residues of an antibody that are involved in antigen binding. The hypervariable region includes amino acid residues from the "CDR" (e.g., LCDR1, LCDR2, and LCDR3 of the light chain variable domain, and HCDR1, HCDR2, and HCDR3 of the heavy chain variable domain). See Kabat et al., (1991) Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, Md. (which defines the antibody CDR region by sequence). Also see Chothia and Lesk (1987) J. Mol. Biol. 196:901-917 (which defines the antibody CDR region by structure). The term "framework" or "FR" residues refers to variable domain residues other than the hypervariable region residues defined herein as CDR residues.

[0097] Unless otherwise specified, “antigen-binding fragment” means an antigen-binding fragment of an antibody, i.e., an antibody fragment that retains the ability to specifically bind to an antigen to which it is bound by a full-length antibody, for example, a fragment that retains one or more CDR regions. Examples of antigen-binding fragments include, but are not limited to, Fab, Fab', F(ab')2, and Fv fragments; diabodies; linear antibodies; single-chain antibody molecules, e.g., single-chain Fv(ScFv); nanobodies, and multispecific antibodies formed from antibody fragments.

[0098] As used herein, "specifically binds" an antibody to a target protein means that such an antibody exhibits selective binding to its target compared to binding to other proteins, but this specificity does not necessarily require absolute binding specificity. "Specifically binding" or "selectively binding" of an antibody is used in the context of describing an interaction between an antigen (e.g., a protein) and an antibody or antigen-binding antibody fragment, referring to a binding reaction that determines the presence of an antigen in a heterogeneous population of proteins and other biologics, e.g., a biological sample, blood, serum, plasma, or tissue sample. Therefore, under specific, designated immunoassay conditions, the antibody or its antigen-binding fragment binds specifically to a particular antigen at least twice as much as the background level, and does not specifically bind to other antigens present in the sample in significant amounts. In one embodiment, under designated immunoassay conditions, the antibody or its antigen-binding fragment binds specifically to a particular antigen at least ten times as much as the background level of binding, and does not specifically bind to other antigens present in the sample in significant amounts.

[0099] As used herein, an "antigen-binding domain" comprises at least three CDRs and specifically binds to an epitope. The "antigen-binding domain" of a multispecific antibody (e.g., a bispecific antibody) comprises a first antigen-binding domain that specifically binds to a first epitope, and a second antigen-binding domain that also comprises at least three CDRs that specifically bind to a second epitope. A multispecific antibody may have antigen-binding domains directed to each specific epitope, and may be bispecific, triplicate, quadruplicate, etc. A multispecific antibody may be polyvalent (e.g., a bispecific quadruple antibody) and contain multiple antigen-binding domains, for example, two, three, four, or more antigen-binding domains that specifically bind to a first epitope, and two, three, four, or more antigen-binding domains that specifically bind to a second epitope.

[0100] In this specification, the term "human antibody" means an antibody containing only human immunoglobulin protein sequences. Human antibodies may contain mouse glycans if they are produced in mice, mouse cells, or mouse cell-derived hybridomas. Similarly, "mouse antibody" or "rat antibody" means an antibody containing only mouse immunoglobulin protein sequences or an antibody containing only rat immunoglobulin protein sequences, respectively.

[0101] The terms "humanized" or "humanized antibody" refer to a form of antibody that contains sequences derived from non-human (e.g., mouse) antibodies and human antibodies. Such antibodies contain minimal sequences derived from non-human immunoglobulins. Generally, humanized antibodies contain substantially all of at least one, typically two, variable domains, where all or substantially all hypervariable loops in the variable domains correspond to the hypervariable loops of non-human immunoglobulins, and all or substantially all FR regions are FR regions of human immunoglobulin sequences. Humanized antibodies also optionally contain at least a portion of the immunoglobulin constant region (Fc), typically at least a portion of the human immunoglobulin constant region (Fc). When it is necessary to distinguish humanized antibodies from rodent-parent antibodies, the prefix "hum," "hu," "Hu," or "h" is added to the name of the antibody clone. Humanized forms of rodent antibodies generally contain the same CDR sequence as the parent rodent antibody, but may include certain amino acid substitutions to increase affinity, improve the stability of the humanized antibody, remove post-translational modifications, or for other reasons.

[0102] The term "epitope" refers to a specific site on an antigen to which an antibody binds. This specific site on the antigen can be determined, for example, by crystal structure analysis. Methods such as hydroxyl radical protein footprinting and alanine scanning mutagenesis can also be used, although they may have low resolution.

[0103] The term "monospecific antibody" refers to an antibody that specifically binds to only one antigen. A monospecific antibody may bind to only one epitope of an antigen, or it may bind to two or more epitopes of an antigen. A monospecific antibody that binds to two or more epitopes of an antigen is called a monospecific polyepitope antibody.

[0104] The term "multispecific antibody" refers to an antibody that specifically binds to two or more antigens (e.g., bispecific antibody, tripspecific antibody, etc.). Non-exclusive examples of multispecific antibodies include, but are not limited to, antibodies containing a heavy chain variable domain (VH) and a light chain variable domain (VL) in which the VH / VL unit has polyepitope specificity, antibodies having two or more VL and VH domains in which each VH / VL unit binds to a different epitope, antibodies having two or more single variable domains in which each single variable domain binds to a different epitope, diabodies, triabodies, and full-length antibodies and / or antibody fragments linked covalently or noncovalently.

[0105] The terms "polyepitope antibody" and "antibody having polyepitope specificity" are used interchangeably herein and refer to antibodies that bind to two or more epitopes on the same or different antigens.

[0106] The term “Fc region” is used herein to define the C-terminal region of an immunoglobulin heavy chain, including the natural sequence Fc region and mutant Fc regions. While the boundaries of the Fc region of an immunoglobulin heavy chain can vary, the human IgG heavy chain Fc region is typically defined as extending from the amino acid residue at position Cys226 or from Pro230 to its carboxyl terminus. The C-terminal lysine of the Fc region (residue 447 according to the Eu numbering system) may be removed, for example, during antibody production or purification, or by recombination of the nucleic acid encoding the antibody heavy chain. Thus, a composition of intact antibodies may include an antibody population from which the entire Lys447 residue has been removed, an antibody population without the removed Lys447 residue, and an antibody population having a mixture of antibodies with and without the Lys447 residue.

[0107] A "functional Fc region" possesses effector function from a naturally occurring Fc region. Exemplary effector functions include C1q binding; complement-dependent cytotoxicity (CDC); Fc receptor binding; antibody-dependent cell-mediated cytotoxicity (ADCC); phagocytosis; and downregulation of cell surface receptors (e.g., B cell receptors; BCRs). Such effector functions generally require an Fc region combined with a binding domain (e.g., an antibody-variable domain) and can be evaluated using various assays disclosed herein or known in the art. Functional Fc regions may have substantially similar effector function to wild-type IgG, reduced effector function compared to wild-type IgG, or enhanced effector function compared to wild-type IgG. In the case of antibodies containing human Fc regions, comparison is typically made with wild-type human IgG1.

[0108] The "natural sequence Fc region" contains amino acid sequences identical to those of naturally occurring Fc regions. Natural sequence human Fc regions include the natural sequence human IgG1 Fc region (non-A and A allotypes), the natural sequence human IgG2 Fc region, the natural sequence human IgG3 Fc region, and the natural sequence human IgG4 Fc region, as well as their naturally occurring variants.

[0109] The "mutant Fc region" contains an amino acid sequence different from that of the natural sequence Fc region by at least one amino acid modification (e.g., about 1 to about 10 amino acid modifications, and in some embodiments, about 1 to about 5 amino acid modifications), preferably one or more amino acid substitutions. The mutant Fc region as described herein preferably has at least about 80% homology to the natural sequence Fc region and / or the Fc region of the parent polypeptide, preferably at least about 90% homology, or preferably at least about 95% homology. In some embodiments, the mutant Fc region may have reduced or enhanced effector function compared to wild-type IgG. In the case of an antibody containing a human Fc region, it is typically compared to wild-type human IgG1.

[0110] As used herein, the term "Fc component" refers to the hinge region, CH2 domain, or CH3 domain of the Fc region.

[0111] The term "hinge region" is generally defined as extending to approximately IgG residues 216–230 (Eu numbering), IgG residues 226–243 (Kabat numbering), or IgG residues 1–15 (IMGT's unique numbering).

[0112] The term "antibody fragment" refers to molecules other than the intact antibody, including a portion of the intact antibody that binds to the antigen to which the intact antibody binds. Examples of antigen-binding fragments include diabody, Fab, Fab', F(ab')2, and F(ab). cExamples of binding fragments include, but are not limited to, Fv fragments, disulfide-stabilized Fv fragments (dsFv), (dsFv)2, bispecific dsFv (dsFv-dsFv'), disulfide-stabilized diabodies (dsdiabodies), triabodies, tetrabodies, single-chain antibodies, scFv, scFv dimers, single-domain antibodies, and multivalent domain antibodies. Typically, the binding fragment competes with the intact antibody from which it originates for specific binding. Binding fragments can be produced by recombinant DNA techniques or by enzymatic or chemical separation of intact immunoglobulins.

[0113] The term "Fab" refers to the portion of an antibody consisting of a single light chain (both variable and constant regions) attached by disulfide bonds to the variable region and first constant region of a single heavy chain.

[0114] The term "Fab" refers to a Fab fragment that includes part of the hinge region.

[0115] The term "F(ab')2" refers to the dimer of Fab'. The F(ab')2 antibody fragment was originally produced as a pair with a Fab' fragment that had a hinge cysteine ​​in between. Other chemical couplings of antibody fragments are also known.

[0116] The term "Fv" refers to the smallest fragment of an antibody that has a complete antigen-binding site. An Fv fragment consists of a single variable region of a light chain bound to a variable region of a single heavy chain.

[0117] The term "single-chain antibody" refers to an antibody consisting of a heavy-chain variable region and a light-chain variable region linked by a linker. In most cases, though not all, the linker can be a peptide. The length of the linker varies depending on the type of single-chain antibody. Higher-order forms can be obtained by linking two or more single-chain antibodies covalently or by means other than covalent bonds. Examples of single-chain antibodies and their higher-order forms include, but are not limited to, single-domain antibodies, multivalent-domain antibodies, single-chain variant fragments (scFv), divalent scFv (di-scFv), trivalent scFv (tri-scFv), tetravalent scFv (tetra-scFv), diabodies, triabodies, and tetrabodies.

[0118] The terms “single-chain Fv antibody” and “scFv” are used interchangeably herein and refer to a single-chain antibody consisting of a heavy-chain variable region and a light-chain variable region linked by a linker. In most, though not all, cases, the linker can be a peptide. The linker peptide is preferably about 5–30 amino acids long, or about 10–25 amino acids long. Typically, the linker allows for the stabilization of the variable domain without interfering with proper folding and the formation of the active binding site. In preferred embodiments, the linker peptide is rich in serine or threonine, as well as glycine. By linking two or more scFvs covalently or noncovalently, higher-order forms such as di-scFv, tri-scFv, and tetra-scFv can be obtained. The antigen-binding site of each scFv in a higher-order form can target the same or different antigens or epitopes.

[0119] The terms "single-chain Fv-Fc antibody" or "scFv-Fc" refer to a full-length antibody consisting of scFv attached to an Fc region.

[0120] A "diabody" is a higher-order variant of a single-chain antibody, consisting of two single-chain antibodies. For each single-chain antibody, a linker too short to allow pairing between two domains on the same chain is used, creating two antigen-binding sites by pairing the domains with complementary domains on the other chain. In most, though not all, cases, the linker can be a peptide. The antigen-binding sites can target the same or different antigens or epitopes. Triabodies (three single-chain antibodies assembled to form three antigen-binding sites), tetrabodies (four single-chain antibodies assembled to form four antigen-binding sites), and higher-order variants can similarly be produced. See, for example, Holliger P. et al., Proc Natl Acad Sci USA. July 15;90(14):6444-8 (1993);EP404097;WO93 / 11161.

[0121] A "single-domain antibody" refers to an antibody fragment that contains only the variable region of the heavy chain or the variable region of the light chain. In certain cases, two or more V H The domains covalently bond to the peptide linker to create a multivalent domain antibody. Two or more Vs of the multivalent domain antibody H Domains can target the same or different antigens or epitopes.

[0122] The term "heavy-chain antibody" refers to an antibody consisting of two heavy chains. Heavy-chain antibodies can be IgG-like antibodies from camels, llamas, alpacas, sharks, etc., or IgNARs from cartilaginous fish. See, for example, Riechmann L. and Muyldermans S., J Immunol Methods. December 10;231(1-2): 25-38 (1999); Muyldermans S., J Biotechnol. June;74(4):277-302 (2001); WO94 / 04678; WO94 / 25591; or U.S. Patent No. 6,005,079. Heavy-chain antibodies originally originated from the camelid family (camels, dromedaries, and llamas). Camelized antibodies lack a light chain but possess a genuine antigen-binding repertoire (Hamers-Csterman C. et al., Nature. June 3;363(6428):446-8 (1993); Nguyen VK et al. “Heavy-chain antibodies in Camelidae; a case of evolutionary innovation,” Immunogenetics. April;54(1):39-47 (2002); Nguyen VK et al. Immunology. May;109(1):93-101 (2003)). The variable domain (VHH domain) of heavy-chain antibodies represents the smallest known antigen-binding unit generated by the adaptive immune response (Koch-Nolte F. et al., FASEB J. November;21(13):3490-8. Epub 2007 Jun. 15 (2007)).

[0123] The term “corresponding human germline sequence” refers to a nucleic acid sequence encoding a human variable region amino acid sequence or subsequence that shares the highest determined amino acid sequence identity with a reference variable region amino acid sequence or subsequence compared to all other known variable region amino acid sequences encoded by a human germline immunoglobulin variable region sequence. The corresponding human germline sequence may also refer to a human variable region amino acid sequence or subsequence that has the highest amino acid sequence identity with a reference variable region amino acid sequence or subsequence compared to all other evaluated variable region amino acid sequences. The corresponding human germline sequence may be a sequence or subsequence containing only the framework region, only the complementarity-determining region, the framework and complementarity-determining region, a variable segment (as defined above), or any other combination of sequences or subsequences containing the variable region. Sequence identity can be determined using the methods described herein, e.g., aligning two sequences using BLAST, ALIGN, or another alignment algorithm known in the art. The corresponding human germline nucleic acid or amino acid sequence may have at least approximately 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity with the nucleic acid or amino acid sequence of the reference variable region. Furthermore, if the antibody contains a constant region, the constant region may also be derived from such a human sequence, e.g., a human germline sequence, or a variant version of a human germline sequence, or from an antibody containing a consensus framework sequence derived from human framework sequence analysis, as described, for example, in Knappik et al., J.Mol.Biol.296:57-86, 2000.

[0124] The term “equilibrium dissociation constant (KD, M)” refers to the dissociation rate constant (kd, time-1) divided by the association rate constant (ka, time-1, Ml). The equilibrium dissociation constant can be measured using any method known in the art. The antibodies of this disclosure generally have an equilibrium dissociation constant less than about 10⁻⁷ or less than 10⁻⁸ M, for example, less than about 10⁻⁹ M or less than 10⁻¹⁰ M, and in some embodiments, less than about 10⁻¹¹ M, less than 10⁻¹² M or less than 10⁻¹³ M.

[0125] In this specification, the terms “cancer” or “tumor” have the broadest meaning as understood in the art and refer to a physiological condition in mammals typically characterized by uncontrolled cell proliferation. In the context of this disclosure, cancer is not limited to any particular type or location.

[0126] In the context of this disclosure, when referring to an amino acid sequence, the term “conservative substitution” means the substitution of an original amino acid with a new amino acid that does not substantially alter the chemical, physical, and / or functional properties of the antibody or fragment, such as its binding affinity to CD3. Specifically, common conservation transformations of amino acids are well known in the art.

[0127] As used herein, the term “knob-into-hole” technology refers to an amino acid that directs the pairing of two polypeptides together, either in vitro or in vivo, by introducing a spatial bulge (knob) into one polypeptide and a socket or cavity (hole) into another polypeptide (at the interface where they interact). For example, knob-into-holes are introduced at the Fc:Fc binding interface, CL:CHI interface, or VH / VL interface of an antibody (see, e.g., US2011 / 0287009, US2007 / 0178552, WO96 / 027011, WO98 / 050431, and Zhu et al., 1997, Protein Science 6:781-788). In some embodiments, knob-into-holes ensure the correct pairing of two different heavy chains during the production of multispecific antibodies. For example, multispecific antibodies having knob-into-hole amino acids within their Fc regions may further contain a single variable domain linked to each Fc region, or different heavy chain variable domains that pair with similar or different light chain variable domains. The knob-into-hole technique can also be used in VH or VL regions to ensure correct pairing.

[0128] As used herein, the term “knob” refers, in the context of “knob-into-hole” technology, to an amino acid change that introduces a bulge (knob) within a polypeptide at an interface where one polypeptide interacts with another polypeptide. In some embodiments, the other polypeptide has a hole mutation.

[0129] As used herein, the term "hole" in the context of "knob-into-hole" refers to an amino acid change that introduces a socket or cavity (hole) within a polypeptide at an interface where one polypeptide interacts with another polypeptide. In some embodiments, the other polypeptide has a knob mutation.

[0130] Examples of algorithms suitable for determining percent sequence identity and sequence similarity include the BLAST algorithm, described in Altschul et al., Nuc. Acids Res. 25:3389-3402, 1977; and Altschul et al., J. Mol. Biol. 215:403-410, 1990, respectively. Software for performing BLAST analysis is publicly available through the National Center for Biotechnology Information. This algorithm first identifies high-scoring sequence pairs (HSPs) by identifying short words of length W within the query sequence, which are words that match or satisfy a threshold score T that has a positive value when aligned with words of the same length in the database sequence. T is called the neighbor word score threshold. These first hit neighbor words serve as a value to initiate a search for longer HSPs that contain them. Word hits are extended toward both ends of each sequence as long as the cumulative alignment score can be increased. For nucleotide sequences, the cumulative score is calculated using parameters M (reward score for a matching pair of residues, always >0) and N (penalty score for mismatched residues, always <0). For amino acid sequences, the cumulative score is calculated using a score matrix. Word hits are stopped in each direction if the cumulative alignment score falls by X from the maximum attainable value; if the cumulative score becomes zero or less due to the accumulation of one or more negative score residue alignments; or if either end of the sequence is reached. The BLAST algorithm parameters W, T, and X determine the sensitivity and speed of alignment. The BLASTN program (for nucleotide sequences) uses 11 word lengths (W), 10 expected values ​​(E), M=5, N=-4, and double-strand comparison as defaults.For amino acid sequences, the BLAST program uses a word length of 3, an expected value of 10 (E), and a BLOSUM62 score matrix of 50 (see Henikoff and Henikoff, (1989) Proc. Natl. Acad. Sci. USA 89:10915), alignment (B), expected value of 10 (E), M=5, N=-4, and double-strand comparison as defaults.

[0131] The BLAST algorithm also performs a statistical analysis of the similarity between two sequences (see, e.g., Karlin and Altschul, Proc. Natl. Acad. Sci. USA 90:5873-5787, 1993). One measure of similarity provided by the BLAST algorithm is the smallest sum probability (P(N)), which provides an indicator of the probability that a match between two sequences of nucleotides or amino acids occurs by chance. For example, a nucleic acid is considered similar to a reference sequence if the smallest sum probability when comparing the test nucleic acid with the reference nucleic acid is less than about 0.2, more preferably less than about 0.01, and most preferably less than about 0.001.

[0132] The percentage of identity between two amino acid sequences can also be determined using the algorithm of E. Meyers and W. Miller, Comput. Appl. Biosci. 4:11-17, (1988). This algorithm is incorporated into the ALIGN program (version 2.0) using a PAM120 weighted remainder table, a gap length penalty of 12, and a gap penalty of 4. Furthermore, the percentage of identity between two amino acid sequences can also be determined using the algorithm of Needleman and Wunsch, J. Mol. Biol. 48:444-453, (1970), incorporated into the GAP program of the GCG software package, using either a BLOSUM62 matrix or a PAM250 matrix, as well as gap weights of 16, 14, 12, 10, 8, 6, or 4 and length weights of 1, 2, 3, 4, 5, or 6.

[0133] The term “nucleic acid” is used herein in the same sense as the term “polynucleotide” and refers to deoxyribonucleotides or ribonucleotides in either single-stranded or double-stranded form and polymers thereof. This term encompasses nucleic acids containing known nucleotide analogs or modified skeletal residues or bonds, which include synthetic nucleic acids, naturally occurring nucleic acids, and naturally occurring nucleic acids, which have similar binding properties to the reference nucleic acid and are metabolized in a similar manner to the reference nucleotide. Examples of such analogs include, but are not limited to, phosphorothioates, phosphoramidates, methylphosphonates, chiral-methylphosphonates, 2-O-methylribonucleotides, and peptide nucleic acids (PNAs).

[0134] In relation to nucleic acids, the term "functionally linked" refers to a functional relationship between two or more polynucleotide (e.g., DNA) segments. Typically, this refers to a functional relationship between a transcriptional regulatory sequence and a transcription sequence. For example, a promoter or enhancer sequence is functionally linked to a coding sequence if it stimulates or modulates the transcription of that coding sequence in a suitable host cell or other expression system. Generally, promoters and transcriptional regulatory sequences that are functionally linked to a transcription sequence are physically contiguous to the transcription sequence; i.e., they are cis-acting. However, some transcriptional regulatory sequences, such as enhancers, do not need to be physically contiguous or located in close proximity to the coding sequence in which they enhance transcription.

[0135] In some embodiments, the Disclosure provides compositions comprising the anti-CD3 multispecific antibodies described herein, for example, pharmaceutically acceptable compositions, formulated with at least one pharmaceutically acceptable excipient. As used herein, the term “pharmaceutically acceptable excipient” includes all physiologically compatible solvents, dispersions, isotonic agents, and absorption retarders, etc. The excipient may be suitable for intravenous, intramuscular, subcutaneous, parenteral, rectal, spinal, or epidermal administration (e.g., by injection or infusion).

[0136] The compositions disclosed herein may be in a variety of forms. These include, for example, liquid solutions (e.g., injection and infusion solutions), dispersions or suspensions, liposomes, and liquid, semi-solid, and solid dosage forms such as suppositories. The appropriate form depends on the intended method of administration and therapeutic use. Typical appropriate compositions are in the form of injection and infusion solutions. One appropriate method of administration is parenteral administration (e.g., intravenous, subcutaneous, intraperitoneal, intramuscular). In some embodiments, antibodies are administered by intravenous infusion or injection. In certain embodiments, antibodies are administered by intramuscular or subcutaneous injection.

[0137] As used herein, the term “therapeutic dose” means the amount of antibody sufficient to produce such treatment for a disease, disorder, or symptom when administered to a subject to treat the disease, disorder, or symptom, or to treat at least one of the clinical symptoms of the disease or disorder. “Therapeutic dose” may vary depending on the antibody, the disease, disorder, and / or the symptoms of the disease or disorder, the severity of the symptoms of the disease, disorder, and / or the age of the subject being treated, and / or the weight of the subject being treated. An appropriate dose in any given case may be obvious to those skilled in the art and may also be determined by routine experimentation. In the case of combination therapy, “therapeutic dose” means the total amount of the combined subject to effectively treat the disease, disorder, or condition.

[0138] The term “combination therapy” refers to the administration of two or more therapeutic agents for treating a therapeutic condition or disorder described herein. Such administrations include the simultaneous administration of these therapeutic agents in substantially simultaneous mode. Such administrations also include the simultaneous administration of each active ingredient in multiple containers or in separate containers (e.g., capsules, powders, and liquids). Powders and / or liquids may be reconstituted or diluted to a desired dose before administration. Furthermore, such administrations also include the use of various therapeutic agents in a sequential manner, either at approximately the same time or at different times. In any case, the treatment regimen provides the beneficial effects of drug combination in the treatment of the conditions or disorders described herein.

[0139] As used herein, the phrase "in combination with" means that the anti-CD3 multispecific antibody is administered to the subject simultaneously with, immediately before, or immediately after, the administration of an additional therapeutic agent. In certain embodiments, the anti-CD3 multispecific antibody is administered as a combination with an additional therapeutic agent. [Modes for carrying out the invention]

[0140] This disclosure provides antibodies, antigen-binding fragments, and anti-CD3 multispecific antibodies. Furthermore, this disclosure provides antibodies that have desirable pharmacokinetic properties and other desirable attributes and are therefore usable for reducing the likelihood of cancer or for treating cancer. This disclosure further provides pharmaceutical compositions comprising antibodies for the prevention and treatment of cancer and related disorders, as well as methods for manufacturing and using such pharmaceutical compositions.

[0141] Anti-CD3 antibody This disclosure provides antibodies or antigen-binding fragments thereof that specifically bind to CD3. The antibodies or antigen-binding fragments of this disclosure include, but are not limited to, antibodies or antigen-binding fragments thereof that are produced as described below.

[0142] This disclosure provides an antibody or antigen-binding fragment that specifically binds to CD3, wherein the antibody or antibody fragment (e.g., an antigen-binding fragment) comprises a VH domain having the amino acid sequences listed in Table 2. This disclosure also provides an antibody or antigen-binding fragment that specifically binds to CD3, wherein the antibody or antigen-binding fragment comprises an HCDR having any one of the HCDRs listed in Table 2. In one embodiment, this disclosure also provides an antibody or antigen-binding fragment that specifically binds to CD3, wherein the antibody comprises (or, otherwise, consists of) one, two, three, or more HCDRs having any of the HCDRs listed in Table 2.

[0143] Other antibodies or antigen-binding fragments of this disclosure contain mutated amino acids that have at least 60%, 70%, 80%, 90%, 95%, or 99% identity in the CDR region with those disclosed in Table 2. In some embodiments, the amino acid sequence has at least 90%, 91, 92, 93, 94, 95, 96, 97, 98, or 99% sequence identity. In some embodiments, the mutated amino acid sequence contains one, two, three, four, or five or fewer amino acids mutated in the CDR region compared to the CDR region shown in the sequences listed in Table 2.

[0144] Other antibodies in this disclosure include antibodies in which an amino acid or nucleic acid encoding an amino acid is mutated but which have at least 60%, 70%, 80%, 90%, 95%, or 99% identity with the sequences listed in Table 2. In some embodiments, the amino acid sequences have at least 90%, 91, 92, 93, 94, 95, 96, 97, 98, or 99% sequence identity. In some embodiments, the amino acid sequence mutations include mutations in which one, two, three, four, or five or fewer amino acids are mutated in the variable region compared to the variable region shown in the sequences listed in Table 2, while maintaining substantially the same therapeutic activity.

[0145] This disclosure also provides nucleic acid sequences encoding the VH, VL, full-length heavy chain, and full-length light chain of an antibody that specifically binds to CD3. Such nucleic acid sequences can be optimized for expression in mammalian cells.

[0146] This disclosure provides antibodies and antigen-binding fragments that bind to the epitopes of human CD3. In certain embodiments, the antibody and antigen-binding fragment can bind to the same epitope of CD3.

[0147] This disclosure also provides antibodies and their antigen-binding fragments that bind to the same epitopes as the anti-CD3 antibodies listed in Table 2. Therefore, additional antibodies and their antigen-binding fragments can be identified based on their ability to cross-compete with other antibodies in a binding assay (e.g., competitively inhibit binding in a statistically significant manner). The ability of a test antibody to inhibit the binding of the antibodies and their antigen-binding fragments of this disclosure to CD3 demonstrates that the test antibody can compete with that antibody or its antigen-binding fragment for binding to CD3. Such antibodies can bind to the same or related (e.g., structurally similar or spatially proximal) epitopes on CD3 as those of the competing antibody or its antigen-binding fragment, without being constrained by any one theory. In certain embodiments, antibodies that bind to the same epitopes on CD3 as those of the antibodies or their antigen-binding fragments of this disclosure are human or humanized monoclonal antibodies. Such human or humanized monoclonal antibodies can be prepared and isolated as described herein.

[0148] In one embodiment, the anti-CD3 antibody disclosed herein is an anti-CD3 multispecific antibody. The antibody molecule is a multispecific antibody molecule, for example, comprising multiple antigen-binding domains, wherein at least one antigen-binding domain sequence specifically binds to CD3 as a first epitope, and further antigen-binding domain sequences(s) specifically bind to other epitopes(s). In one embodiment, the multispecific antibody comprises a third, fourth, or fifth antigen-binding domain. In one embodiment, the multispecific antibody is a bispecific, triplicate, or quadruplicate antibody. In each example, the multispecific antibody comprises at least one anti-CD3 antigen-binding domain.

[0149] In one embodiment, a multispecific antibody is a bispecific antibody. As used herein, a bispecific antibody specifically binds to only two antigens. A bispecific antibody comprises a first antigen-binding domain that specifically binds to CD3 and a second antigen-binding domain that specifically binds to another epitope. This includes a bispecific antibody comprising a heavy-chain variable domain and a light-chain variable domain that specifically binds to CLDN6 as the first epitope, and a heavy-chain variable domain that specifically binds to CD3 as the second epitope. In a bispecific antibody comprising an antigen-binding fragment, the antigen-binding fragment may be Fab, F(ab')2, Fv, or single-chain Fv(ScFv) or scFv.

[0150] Previous experiments (Coloma and Morrison Nature Biotech. 15:159-163 (1997)) described a tetravalent bispecific antibody recombinant by fusing DNA encoding a single-chain anti-dansyl antibody Fv (scFv) after the C-terminus (CH3-scFv) or after the hinge (hinge-scFv) of IgG3 of an anti-dansyl antibody. This disclosure provides a polyvalent antibody (e.g., a tetravalent antibody) having at least two antigen-binding domains, which can be readily produced by recombinant expression of the nucleic acid encoding the polypeptide chain of the antibody. The polyvalent antibodies herein comprise 3 to 8, preferably 4, antigen-binding domains that specifically bind to at least two antigens.

[0151] Linker It is also understood that the domains and / or regions of the polypeptide chain of a bispecific tetravalent antibody can be separated by linker regions of varying lengths. In some embodiments, the antigen-binding domains are separated from each other by linker regions from the entire CL, CH1, hinge, CH2, CH3, or Fc region. For example, VL1-CL-(linker)VH2-CH1, VH-linker-VL. Such linker regions may contain a random classification of amino acids or a limited set of amino acids. Such linker regions may be flexible or rigid (see US2009 / 0155275).

[0152] Multispecific antibodies can be dimerized via leucine zippers (Kostelny et al., J.Immunol. 1992 148:1547-53; de Kruifetal J.Biol.Chem. 1996 271:7630-4) and Ig C / CH1 domains (Muller et al., FEBS Lett. 422:259-64), with or without the use of flexible linkers; via dimerization mechanisms such as diabodies (Holliger et al., (1993) Proc.Nat.Acad.Sci.USA. 1998 90:6444-8; Zhu et al., Bio / Technology (NY) 1996 14:192-6); Fab-scFv fusion (Schoonjans et al., J.Immunol. 2000 165:7050-7); and mini-antibody formats (Packet). It is constructed by genetically fusing two single-stranded Fv (scFv) or Fab fragments, as reported by al., Biochemistry 1992.31:1579-84; Packet al., Bio / Technology 1993 11:1271-7) (Mallender et al., J. Biol. Chem. 1994 269:199-206; Packet et al., Proc. Natl. Acad. Sci. USA. 1995 92:7021-5; Zapata et al., Protein Eng. 1995 8.1057-62).

[0153] The bispecific tetravalent antibodies disclosed herein include a linker region between one or more of the antigen-binding domain, CL domain, CH1 domain, hinge region, CH2 domain, CH3 domain, or Fc region, comprising at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, or more amino acid residues. In some embodiments, the amino acids glycine and serine constitute the amino acids within the linker region. In another embodiment, the linker is GS, GGS, GSG, SGG, GGG, GGGS, SGGG, GGGGS, GGGGSGS, GGGGSGS, GGGGSGS, GGGGSGGGGS, GGGGSGGGGSGGGGS, AKTTPKLEEGEFSEAR, AKTTPKLEEGEFSEARV, AKTTPKLGG, SAKTTPKLGG, AKTTPKLEEGEFSEARV, SAKTTP, SAKTTPKLGG, RADAAP, RADAAPTVS, RADAAAAGGPGS, RADAAAA(G4S)4, SAKTTP , SAKTTPKLGG, SAKTTPKLEEGEFSEARV, ADAAP, ADAAPTVSIFPP, TVAAP, TVAAPSVFIFPP, QPKAAP, QPKAAPSVTLFPP, AKTTPP, AKTTPPSVTPLAP, AKTTAP, AKTTAPSVYPLAP, ASTKGP, ASTKGPSVFPLAP, GENKVEYAPALMALS, GPAKELTPLKEAKVS, and GHEAAAVMQVQYPAS or any combination thereof (see WO2007 / 024715).

[0154] Dimerization-specific amino acids In one embodiment, a polyvalent antibody contains at least one dimerization-specific amino acid change. This dimerization-specific amino acid change results in a "knobs into holes" interaction, increasing the construction of the correct polyvalent antibody. The dimerization-specific amino acid may be within a CH1 domain, a CL domain, or a combination thereof. Dimerization-specific amino acids are used to pair a CH1 domain with another CH1 domain (CH1-CH1) and a CL domain with another CL domain (CL-CL), and can be found in at least the disclosures WO2014082179, WO2015181805 family, and WO2017059551. The dimerization-specific amino acid may be within an Fc domain and may be combined with a dimerization-specific amino acid within a CH1 or CL domain. In one embodiment, the disclosure provides a bispecific antibody containing at least one dimerization-specific amino acid pair.

[0155] Further modifications to the FC domain framework In yet another embodiment, the Fc region is modified by substituting at least one amino acid residue with a different amino acid residue to alter the effector function of the antibody. For example, one or more amino acids can be substituted with different amino acid residues, resulting in an antibody that has a modified affinity for the effector ligand but retains the antigen-binding ability of the parent antibody. The effector ligand with modified affinity may be, for example, the Fc receptor or the C1 component of complement. This approach is described, for example, in U.S. Patent Nos. 5,624,821 and 5,648,260, both by Winter et al.

[0156] In another embodiment, one or more amino acid residues may be substituted with one or more different amino acid residues so that the antibody has modified C1q binding and / or reduced or absent complement-dependent cytotoxicity (CDC). This technique is described, for example, in U.S. Patent No. 6,194,551 by Idusogie et al.

[0157] In yet another embodiment, one or more amino acid residues are modified to alter the antibody's ability to immobilize complement. This technique is described, for example, in Publication WO94 / 29351 by Bodmer et al. In certain embodiments, one or more amino acids of the antibody or its antigen-binding fragment of the Disclosure are replaced with one or more allotype amino acid residues for the IgG1 subclass and kappa isotype. Allotype amino acid residues include, but are not limited to, the constant regions of the heavy chains of the IgG1, IgG2, and IgG3 subclasses and the constant regions of the light chains of kappa isotypes, as described by Jefferis et al., MAbs.1:332-338 (2009).

[0158] In another embodiment, the Fc region is modified by modifying one or more amino acids to enhance the antibody's ability to mediate antibody-dependent cytotoxicity (ADCC) and / or to increase the antibody's affinity for the Fcγ receptor. This approach is described, for example, in Presta's publication WO00 / 42072. Furthermore, the binding sites of FcγRI, FcγRII, FcγRIII, and FcRn on human IgG1 have been mapped, and variants with improved binding affinity have been described (see Shields et al., J. Biol. Chem. 276:6591-6604, 2001).

[0159] In yet another embodiment, the glycosylation of multispecific antibodies is modified. For example, non-glycosylated antibodies (i.e., antibodies lacking or reduced glycosylation) can be produced. By modifying glycosylation, for example, the affinity of the antibody for an "antigen" can be increased. Such carbohydrate modification can be achieved, for example, by altering one or more glycosylation sites in the antibody sequence. For example, one or more amino acid substitutions can be made, thereby removing one or more variable region framework glycosylation sites, and thus removing glycosylation at those sites. Such deglycosylation can increase the affinity of the antibody for an antigen. Such approaches are described, for example, in U.S. Patents 5,714,350 and 6,350,861 by Co et al.

[0160] Additionally or alternatively, antibodies with altered glycosylation types can be produced (e.g., low-fucosylated antibodies with reduced amounts of fucosyl residues, or antibodies with increased bisecting GlcNAc structures). Such modified glycosylation patterns have been shown to enhance the ADCC activity of antibodies. Such glycosylation modifications can be achieved, for example, by expressing antibodies in host cells with modified glycosylation pathways. Cells with modified glycosylation pathways have been described in the art and can be used as host cells to express recombinant antibodies, thereby producing antibodies with modified glycosylation. For example, Hang et al. in EP1,176,195 describe cell lines having a functionally disrupted FUT8 gene encoding a fucosyltransferase such that antibodies expressed in such cell lines exhibit low fucosylation. Presta's publication WO03 / 035835 describes Lecl3 cells, a mutant CHO cell line with reduced ability to bind fucose to Asn(297)-linked carbohydrates, which also results in decreased fucosylation of antibodies expressed in its host cells (see also Shields et al., (2002) J. Biol. Chem. 277:26733-26740). Umana et al.'s WO99 / 54342 describes a cell line engineered to express glycoprotein-modified glycosyltransferases (e.g., beta(1,4)-N-acetylglucosaminyltransferase III (GnTIII)) such that antibodies expressed in the engineered cell line show increased bifidative GlcNac structure, resulting in increased ADCC activity of the antibodies (see also Umana et al., Nat. Biotech. 17:176-180, 1999).

[0161] In another embodiment, when reduction of ADCC is desired, human antibody subclass IgG4 has been shown in many previous reports to possess only moderate ADCC and little CDC effector function (Moore GL, et al., 2010 MAbs, 2:181-189). However, natural IgG4 has been found to be less stable in acidic buffers or under stress conditions such as elevated temperature (Angal, S. 1993 Mol Immunol, 30:105-108, Dall'Acqua, W. et al, 1998 Biochemistry, 37:9266-9273, Aalberse et al. 2002 Immunol, 105:9-19). Reduced ADCC can be achieved by operably binding the antibody to IgG4 Fc manipulated by a combination of modifications that reduce FcγR binding or C1q binding activity, thereby reducing or eliminating ADCC and CDC effector function. Considering the physicochemical properties of antibodies as biological drugs, one of the less desirable intrinsic properties of IgG4 is its ability to dynamically separate its two heavy chains in solution to form half an antibody, which in vivo generates a bispecific antibody via a process called "Fab arm exchange" (Van der Neut Kolfschoten M, et al., 2007 Science, 317:1554-157). A serine-to-proline mutation at position 228 (EU numbering system) appeared to inhibit the separation of the IgG4 heavy chain (Angal, S. 1993 Mol Immunol, 30:105-108, Aalberse et al., 2002 Immunol, 105:9-19).Some amino acid residues in the hinge and γFc region have been reported to affect antibody interactions with the Fcγ receptor (Chappel SM, et al., 1991 Proc. Natl. Acad. Sci. USA, 88:9036-9040, Mukherjee, J. et al., 1995 FASEB J, 9:115-119, Armour, K. et al., 1999 Eur J Immunol, 29:2613-2624, Clynes, RA et al, 2000 Nature Medicine, 6:443-446, Arnold JN, 2007 Annu Rev immunol, 25:21-50). Furthermore, some IgG4 isoforms that occur rarely in the human population can also produce different physicochemical properties (Brusco, A. et al., 1998 Eur J Immunogenet, 25:349-55; Aalberse et al., 2002 Immunol, 105:9-19). To produce multispecific antibodies with good stability but low ADCC and CDC, it is possible to modify the hinge and Fc region of human IgG4 and introduce several modifications. These modified IgG4 Fc molecules can be found in U.S. Patent No. 8,735,553 by Li et al., which is referenced by reference.

[0162] Antibody production Antibodies and their antigen-binding fragments can be produced by any means known in the art, including but not limited to recombinant expression of antibody tetramers, chemosynthesis, and enzymatic digestion, while full-length monoclonal antibodies can be obtained, for example, by hybridoma or recombinant production. Recombinant expression may be derived from any suitable host cell known in the art, such as mammalian host cells, bacterial host cells, yeast host cells, insect host cells, etc.

[0163] This disclosure further provides polynucleotides encoding antibodies as described herein, for example, polynucleotides encoding segments including heavy chain or light chain variable regions or complementarity-determining regions as described herein. In some embodiments, the polynucleotide encoding the heavy chain variable region has at least 85%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% nucleic acid sequence identity with a polynucleotide selected from the group consisting of SEQ ID NOs: 9, 56, 60, or 64. In some embodiments, the polynucleotide encoding the light chain variable region has at least 85%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% nucleic acid sequence identity with a polynucleotide selected from the group consisting of SEQ ID NOs: 10, 57, 61, or 65.

[0164] The polynucleotides of this disclosure can encode variable region sequences of anti-CD3 antibodies. They can also encode both the variable and constant regions of antibodies. Some of the polynucleotide sequences encode polypeptides containing variable regions of both the heavy and light chains of the exemplified anti-CD3 antibodies.

[0165] This disclosure also provides expression vectors and host cells for producing anti-CD3 antibodies. The choice of expression vector depends on the host cells in which the vector is intended to be expressed. Typically, an expression vector contains a promoter and other regulatory sequences (e.g., enhancers) operably ligated to a polynucleotide encoding an anti-CD3 antibody chain or antigen-binding fragment. In some embodiments, an inducible promoter is used to prevent the expression of the insertion sequence outside of controlled induction conditions. Inducible promoters include, for example, arabinose, lacZ, metallothionein promoters, or heat shock promoters. Culture of transformed organisms can be grown under non-inducible conditions without biasing the population towards encoding sequences in which the expression product is better tolerated by the host cells. In addition to promoters, other regulatory elements may also be required or desired for the efficient expression of anti-CD3 antibodies or antigen-binding fragments. These elements typically include the ATG start codon and adjacent ribosome binding sites or other sequences. Furthermore, the efficiency of expression can be enhanced by incorporating the appropriate enhancer into the cell line being used (see, for example, Scharf et al., Results Probl. Cell Differ. 20:125, 1994, and Bittner et al., Meth. Enzymol., 153:516, 1987). For example, an SV40 enhancer or a CMV enhancer can be used to increase expression in mammalian host cells.

[0166] The host cells for possessing and expressing anti-CD3 antibody chains may be prokaryotic or eukaryotic. E. coli is one prokaryotic host useful for cloning and expressing the polynucleotides of this disclosure. Other suitable microbial hosts include rods such as Bacillus subtilis, and other Enterobacteriaceae such as Salmonella, Serratia, and various Pseudomonas species. In these prokaryotic hosts, it is also possible to create expression vectors containing expression regulatory sequences (e.g., origins of replication) that are adapted to the host cell. Furthermore, there are various well-known promoters, such as lactose promoter systems, tryptophan (trp) promoter systems, beta-lactamase promoter systems, or phage-lambda-derived promoter systems. Promoters typically contain (optionally, operator sequences) ribosome-binding site sequences for controlling expression and initiating and completing transcription and translation. Other microorganisms such as yeast can also be used to express anti-CD3 antibodies. Insect cells can also be used in combination with baculovirus vectors. In other embodiments, mammalian host cells are used to express and produce the anti-CD3 antibodies of this disclosure. For example, these may be hybridoma cell lines expressing endogenous immunoglobulin genes or mammalian cell lines having exogenous expression vectors. These include any normally lethal, or normal or abnormally immortal, animal or human cells. For example, several suitable host cell lines capable of secreting intact immunoglobulins have been developed, including CHO cell lines, various COS cell lines, HEK293 cells, myeloma cell lines, transformed B cells, and hybridomas. The use of mammalian tissue cell cultures for polypeptide expression is generally discussed, for example, in Winnacker, From Genes to Clones, VCH Publishers, NY, NY, 1987.Mammalian host cell expression vectors may contain expression regulatory sequences such as origins of replication, promoters, and enhancers (see, e.g., Queen et al., Immunol. Rev. 89:49-68, 1986), as well as necessary processing information sites such as ribosome binding sites, RNA splice sites, polyadenylation sites, and transcription terminator sequences. These expression vectors typically contain promoters derived from mammalian genes or mammalian viruses. Appropriate promoters may be constitutive, cell type-specific, stage-specific, and / or tunable or modifiable. Useful promoters include, but are not limited to, the metallothionein promoter, the constitutive adenovirus major late promoter, the dexamethasone-inducible MMTV promoter, the SV40 promoter, the MRP polIII promoter, the constitutive MPSV promoter, the tetracycline-inducible CMV promoter (such as the human early CMV promoter), the constitutive CMV promoter, and promoter-enhancer combinations known in the art.

[0167] Production of bispecific antibodies The current standard for manipulated heterodimer antibody Fc domains is a knob-into-hole (KiH) design, which introduces a mutation at the core CH3 domain interface. The resulting heterodimers exhibit a reduced CH3 melting temperature (below 69°C). In contrast, the ZW heterodimer Fc design has a thermal stability of 81.5°C, which is comparable to that of the wild-type CH3 domain.

[0168] Detection and diagnostic methods The antibodies or antigen-binding fragments of this disclosure are useful in a variety of applications, including but not limited to methods for detecting CD3. In one embodiment, the antibodies or antigen-binding fragments are useful for detecting the presence of CD3 in a biological sample. As used herein, the term “detecting” includes quantitative or qualitative detection. In certain embodiments, the biological sample includes cells or tissue. In other embodiments, such tissue includes normal tissue and / or cancerous tissue that express CD3 at a higher level compared to other tissues.

[0169] In one embodiment, the present disclosure provides a method for detecting the presence of CD3 in a biological sample. In a particular embodiment, the method includes contacting the biological sample with an anti-CD3 antibody under conditions in which the binding of the antibody to the antigen is permitted, and detecting whether a complex is formed between the antibody and the antigen. The biological sample may include, but is not limited to, a urine, tissue, sputum, or blood sample.

[0170] The method also includes a method for diagnosing disorders related to CD3 expression. In certain embodiments, the method includes contacting test cells with an anti-CD3 antibody; determining the level of CD3 expression by the test cells (quantitatively or qualitatively) by detecting the binding of the anti-CD3 antibody to the CD3 polypeptide; and comparing the expression level by the test cells with the CD3 expression level in control cells (e.g., normal or non-CD3 expressing cells of the same tissue origin as the test cells), wherein a higher level of CD3 expression in the test cells compared to the control cells indicates the presence of a disorder related to CD3 expression.

[0171] Treatment method The antibodies or antigen-binding fragments of this disclosure are useful in a variety of applications, including but not limited to methods for treating CD3-related disorders or diseases. In one embodiment, the CD3-related disorder or disease is cancer.

[0172] In one embodiment, the present disclosure provides a method for treating cancer. In a particular embodiment, the method comprises administering an effective amount of anti-CD3 antibody or antigen-binding fragment to a patient who requires it. In some embodiments, the cancer is a solid tumor. Examples of cancer include, but are not limited to, gastric cancer, colon cancer, pancreatic cancer, breast cancer, head and neck cancer, kidney cancer, liver cancer, small cell lung cancer, non-small cell lung cancer, ovarian cancer, skin cancer, mesothelioma, lymphoma, leukemia, myeloma, sarcoma, brain cancer, colorectal cancer, prostate cancer, cervical cancer, testicular cancer, endometrial cancer, bladder cancer, rhabdoid tumors and / or gliomas.

[0173] The antibodies or antigen-binding fragments disclosed herein may be administered by any suitable means, for example, parenteral, intrapulmonary, and intranasal administration, and, if local treatment is desired, intra-focal administration. Parenteral infusions include intramuscular, intravenous, intra-arterial, intraperitoneal, or subcutaneous administration. Dosage may be by any suitable route, such as intravenous or subcutaneous injection, depending in part on whether the administration is short-term or long-term. Various dosing schedules, including single or multiple doses, bolus administration, and pulse infusion at various time points, are contemplated herein, but are not limited to these.

[0174] The antibodies or antigen-binding fragments of this disclosure can be formulated, administered, and given in accordance with appropriate medical practice. Factors to consider in this regard include the specific disorder being treated, the specific mammal being treated, the individual patient's clinical condition, the cause of the disorder, the site of drug delivery, the method of administration, the plan of administration, and other factors known to the healthcare professional. The antibodies may, but are not necessarily, be formulated together with one or more agents currently used to prevent or treat the disorder in question. The effective dose of such other agents will vary depending on the amount of antibody present in the formulation, the type of disorder or treatment, and the other factors mentioned above. These are generally used in the same doses and routes of administration as described herein, or at about 1–99% of the doses described herein, or in any dose and route that is empirically / clinically determined to be appropriate.

[0175] For the prevention or treatment of disease, the appropriate dose of the antibody or antigen-binding fragment of this disclosure will vary depending on the type of disease being treated, the type of antibody, the severity and course of the disease, whether the antibody is administered for preventive or therapeutic purposes, prior treatment, the patient's clinical history and response to the antibody, and the judgment of the attending physician. The antibody is preferably administered to the patient in a single dose or over a series of treatments. Depending on the type and severity of the disease, approximately 1 μg / kg to 100 mg / kg of antibody may be an initial candidate dose for administration to the patient, for example, by one or more individual doses or by continuous infusion. A typical daily dose may range from approximately 1 μg / kg to 100 mg / kg or more, depending on the factors mentioned above. Depending on the condition, for repeated administrations over several days or longer periods, the treatment is generally continued until the desired suppression of disease symptoms occurs. Such doses may be administered intermittently, for example, weekly or every three weeks (for example, so that the patient receives approximately 2 to approximately 20 doses, or for example, approximately 6 doses of antibody). A higher loading dose can be administered first, followed by one or more lower doses. However, other drug regimens may be useful. The progression of this treatment can be easily monitored using conventional techniques and assays.

[0176] Combination therapy In one embodiment, the anti-CD3 antibody of this disclosure can be used in combination with other therapeutic agents. Other therapeutic agents that can be used in combination with the anti-CD3 antibody of this disclosure include chemotherapeutic agents (e.g., paclitaxel or paclitaxel preparations; (e.g., Abraxane®), docetaxel; carboplatin; topotecan; cisplatin; irinotecan, doxorubicin, lenalidomide, 5-azacitidine, ifosfamide, oxaliplatin, pemetrexed disodium, cyclophosphamide, etoposide, decitabine, fludarabine, vincristine, bendamustine, chlorambucil, busulfan, gemcitabine, melphalan, pentostatin, mitoxantrone, pemetrexed disodium), tyrosine kinase inhibitors (e.g., EGFR inhibitors (e.g., erlotinib), multi-kinase inhibitors (e.g., MGCD265, RGB-286638), CD20 targeted agents (e.g., rituximab, Ofa (Tumumab, RO5072759, LFB-R603), CD52 targeted drugs (e.g., alemtuzumab), prednisolone, darbepoetin alfa, lenalidomide, Bcl-2 inhibitors (e.g., oblimersen sodium), aurora kinase inhibitors (e.g., MLN8237, TAK-901), proteasome inhibitors (e.g., bortezomib), CD-19 targeted drugs (e.g., MEDI-551, MOR20) 8) Examples include, but are not limited to, MEK inhibitors (e.g., ABT-348), JAK-2 inhibitors (e.g., INCB018424), mTOR inhibitors (e.g., temsirolimus, everolimus), BCR / ABL inhibitors (e.g., imatinib), ET-A receptor antagonists (e.g., ZD4054), TRAIL receptor 2 (TR-2) agonists (e.g., CS-1008), EGEN-001, and polo-like kinase 1 inhibitors (e.g., BI 672).

[0177] The anti-CD3 antibody disclosed herein can be used in combination with other therapeutic agents, such as immune checkpoint antibodies. Such immune checkpoint antibodies may include anti-PD1 antibodies. Examples of anti-PD1 antibodies include, but are not limited to, tislerizumab, pembrolizumab, or nivolumab. Tislerizumab is disclosed in US8,735,553. Pembrolizumab (formerly MK-3475), disclosed in US8,354,509 and US8,900,587, is a humanized IgG4-K immunoglobulin that targets the PD1 receptor and inhibits the binding of PD1 receptor ligands PD-L1 and PD-L2. Pembrolizumab is approved for the treatment of metastatic melanoma and metastatic non-small cell lung cancer (NSCLC) and is being clinically studied for the treatment of head and neck squamous cell carcinoma (HNSCC) and refractory Hodgkin lymphoma (cHL). Nivolumab (disclosed by Bristol-Meyers Squibb) is a fully human IgG4-K monoclonal antibody. Nivolumab (clone 5C4) is disclosed in U.S. patents US8,008,449 and WO2006 / 121168. Nivolumab is approved for the treatment of melanoma, lung cancer, renal cancer, and Hodgkin lymphoma.

[0178] Other immune checkpoint antibodies for use in combination with anti-CD3 antibodies may include anti-TIGIT antibodies. Such anti-TIGIT antibodies may include, but are not limited to, the anti-TIGIT antibodies disclosed in WO2019 / 129261.

[0179] Other immune checkpoint antibodies for use in combination with anti-CD3 antibodies may include anti-OX40 antibodies. Such anti-OX40 antibodies may include, but are not limited to, the anti-OX40 antibodies disclosed in WO2019 / 223733.

[0180] Other immune checkpoint antibodies for use in combination with anti-CD3 antibodies may include anti-TIM3 antibodies. Such anti-TIM3 antibodies may include, but are not limited to, the anti-TIM3 antibodies disclosed in WO2018 / 036561.

[0181] Pharmaceutical compositions and preparations Compositions such as pharmaceutical formulations are also provided, comprising an anti-CD3 antibody or its antigen-binding fragment, or a polynucleotide comprising a sequence encoding an anti-CD3 antibody or antigen-binding fragment. In certain embodiments, the composition comprises one or more anti-CD3 antibodies or antigen-binding fragments, or one or more polynucleotides comprising a sequence encoding one or more anti-CD3 antibodies or antigen-binding fragments. These compositions may further contain suitable carriers, such as pharmaceutically acceptable excipients such as buffers, which are well known in the art.

[0182] The pharmaceutical formulations of anti-CD3 antibodies or antigen-binding fragments described herein are prepared in the form of lyophilized formulations or aqueous solutions by mixing such antibodies or antigen-binding fragments having a desired degree of purity with one or more optionally selected pharmaceutically acceptable carriers (Remington's Pharmaceutical Sciences 16th edition, Osol, A. Ed. (1980)). Medicinally acceptable carriers are generally non-toxic to the recipient at the dosage and concentration used, and include buffers such as phosphates, citrates, and other organic acids, antioxidants including ascorbic acid and methionine, preservatives (octadecyldimethylbenzylammonium chloride, hexamethonium chloride, benzalkonium chloride, benzethonium chloride, phenol, butyl or benzyl alcohol, alkylparabens such as methyl or propylparaben, catechol, resorcinol, cyclohexanol, 3-pentanol, and m-cresol, etc.), low molecular weight (less than approximately 10 residues) polypeptides, and proteins. Examples of pharmaceutically acceptable carriers herein include, but are not limited to, serum albumin, gelatin, or immunoglobulin; hydrophilic polymers, such as polyvinylpyrrolidone; amino acids, such as glycine, glutamine, asparagine, histidine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates including glucose, mannose, or dextrin; chelating agents, such as EDTA; sugars, such as sucrose, mannitol, trehalose, or sorbitol; salt-forming counterions, such as sodium; metal complexes (e.g., Zn-protein complexes); and / or nonionic surfactants, such as polyethylene glycol (PEG). Further examples of pharmaceutically acceptable carriers herein include interstitial drug dispersants such as soluble neutral-active hyaluronidase glycoproteins (sHASEGP), such as human soluble PH-20 hyaluronidase glycoproteins such as rHuPH20 (HYLENEX®, Baxter International, Inc.). Certain exemplary sHASEGPs and uses, including rHuPH20, are described in U.S. Patents US7,871,607 and US2006 / 0104968.In one embodiment, sHASEGP is combined with one or more additional glycosaminoglycanases, such as chondroitinase.

[0183] In one embodiment, the formulation consists of L-histidine / L-histidine hydrochloride monohydrate, trehalose, and polysorbate 20. In another embodiment, the concentration of the anti-CD3 antibody drug is an isotonic solution consisting of 10 mg / mL anti-CD3 antibody, 20 mM histidine / histidine HCl, 240 mM trehalose dihydrate, and 0.02% polysorbate 20, after being prepared with sterile water for injection, with a pH of approximately 5.5.

[0184] An exemplary lyophilized antibody preparation is described in U.S. Patent No. 6,267,958. Aqueous antibody preparations include those described in U.S. Patent No. 6,171,586 and WO2006 / 044908, the latter of which contains a histidine-acetate buffer.

[0185] Sustained-release formulations can be prepared. Suitable examples of sustained-release preparations include a semipermeable matrix of a solid hydrophobic polymer containing an antibody, where these matrices are in the form of molded articles, such as films or microcapsules.

[0186] Preparations used for in vivo administration are generally sterile. Sterility can be easily achieved, for example, by filtration through a sterile filtration membrane.

[0187] Sequence List The sequence listings for this disclosure are shown in Tables 1-3 below. [Table 1-1] [Table 1-2] [Table 1-3] [Table 1-4] Table 1-5 Table 1-6 Table 1-7 Table 1-8 Table 1-9 Table 1-10 Table 1-11 Table 2-1 Table 2-2 Table 2-3 Table 2-4 Table 2-5 Table 2-6 Table 2-7 Table 2-8 Table 2-9 Table 2-10 [Table 2-11] [Table 3-1] [Table 3-2] [Table 3-3] [Table 3-4] [Table 3-5] [Table 3-6] [Examples]

[0188] Example 1. Generation of mouse anti-CLDN6 antibody To generate antibodies against CLDN6, a cohort of 20–25 BALB / C,SJL strain inbred mice was immunized with human CLDN6 overexpressing cells (L929 / human CLDN6, internally prepared). Each cohort underwent an immunization strategy including unique combinations of CLDN6 antigen, dose, injection route, adjuvant, and timing of immunization. A total of 5 animals were immunized across 4–5 cohorts. Animals were immunized over various periods between 0 and 56 days. Titrated serum was screened by FACS to monitor the immune response, typically after 2–4 immunizations over 21–56 days. Serum was screened for antibodies binding to CLDN6 overexpressing cell CHOK1 / human CLDN6. In each animal, the response of CLDN6-specific antibodies was measured, and animals with sufficient anti-CLDN6 Ig titers were selected for a final 4-day boost.

[0189] Lymphoid organs, including the spleen and lymph nodes, were isolated from mice immunized as described above. Hybridomas were generated by fusion with immortalized mouse myeloma cells derived from SP2 / 0 using PEG-based fusion. The resulting cells were plated into 96-well cell culture plates using standard 1640 medium supplemented with HAT for hybridoma selection.

[0190] Example 2. Screening and selection of anti-CLDN6 antibodies Hybridomas were generated as described in Example 1. After 10–13 days of culture and growth medium changes, hybridoma culture supernatant was collected from individual wells and screened to identify wells that secreted CLDN6-specific antibodies. All supernatants were first screened against at least two overexpression cell lines, including CHOK1 / human CLDN6 and CHOK1 / human CLDN9 (internal preparation). Antibody binding in the overexpression cell lines was measured by FACS. CLDN6 antibodies were screened using supernatants from over 20,000 culture wells in four hybridoma fusions. Briefly, 100 μL of hybridoma culture supernatant was co-incubated with CLDN6-expressing cancer cell lines (such as PA-1 or CHOK1 / human CLDN6 stable cell lines) or control cells (such as parental CHOK1) for 30–60 minutes, washed, and then incubated with APC-labeled anti-mouse IgG Fc secondary antibody. After incubation and washing, fluorescence was measured by flow cytometry.

[0191] Hybridomas derived from positive wells were transferred to a 24-well plate in fresh culture medium and grown for 2-3 days. They were then screened again by flow cytometry to identify antibodies binding to cynomolgus monkey CLDN6-overexpressing cell lines and human CLDN6-positive cancer cell lines (PA-1). Antibody binding to cynomolgus monkey CLDN6-overexpressing cell lines and human CLDN6-positive cancer cell lines (PA-1) was measured by flow cytometry. Briefly, 100 μL of hybridoma culture supernatant was co-incubated with CLDN-expressing cancer cell lines (PA-1 or CHOK1 / human CLDN6 stable cell lines, and CHOK1 / human CLDN9 stable cell lines, etc.) or control cells (parent CHOK1, etc.) for 30-60 minutes. After washing, the cells were incubated with APC-labeled anti-mouse IgG Fc secondary antibody. Fluorescence was measured by flow cytometry after incubation and washing.

[0192] Example 3. Subcloning of selected CLDN6 antibody-secreting hybridomas Selected CLDN6 antibody-secreting hybridomas were subcloned once or twice to ensure monoclonality. Briefly, approximately 80-100 viable hybridoma cells were plated in 3 mL of semi-solid methylcellulose medium (Stem Cell Technologies) in a 6-well plate. After 7-10 days, hybridoma colonies that had emerged from single cells as visible clones were collected in a 96-well plate and cultured further in fresh medium for 2-4 days. The culture supernatant was screened by ELISA and flow cytometry as described above to confirm human and cynomolgus monkey CLDN6 binding. Stable hybridoma subclones were cultured in vitro, cells were cryopreserved, antibody production was carried out, and gene cloning and sequencing of antibodies VH and VL were performed.

[0193] Example 4. Determination of CLDN6-bound EC50 value of mouse anti-CLDN6 antibody. After subcloning, the selected anti-CLDN6 antibody-secreting hybridomas were plated into T75 flasks containing 40 ml of fresh 1640 medium supplemented with 2% FBS for antibody production. After 7–10 days of incubation, the supernatant of the hybridomas was collected for antibody purification using a protein A column. The binding activity of the mouse anti-CLDN6 antibody to CLDN6-positive cells was then characterized using flow cytometry. The EC50 values ​​of clone BG87P are shown in Tables 4–6. These data indicate that clone BG87P binds to human CLDN6 but not to human CLDN9. Furthermore, BG87P binds to mouse CLDN6 and cynomolgus monkey CLDN6. [Table 4] [Table 5] [Table 6]

[0194] Example 5. Gene cloning and sequencing of antibodies VH and VL After removing the supernatant, the CLDN6 antibody-secreting hybridomas selected after subcloning were lysed in 100 mL of RLT buffer in a 96-well round-bottom plate. The mRNA-containing lysates were then transferred to a 96-well deep-well plate for mRNA isolation, cDNA synthesis, and DNA sequencing using standard sequencing methods (Sanger sequencing and next-generation sequencing). Generally, total RNA from cell lysates was prepared using a total RNA isolation kit according to the manufacturer's instructions. cDNA was generated by reverse transcription of mRNA using Super Script III first-strand synthesis SuperMix (Invitrogen®) according to the manufacturer's instructions. The nucleic acid and amino acid sequences of BG87P are shown in Table 1 (SEQ ID NOs: 1-10).

[0195] Generation of chimeric BG87P antibody (chBG87P) The ChBG87P antibody was generated by subcloning the variable region (SEQ ID NOs: 7 and 8) of mouse BG87P into a proprietary expression vector containing human wild-type IgG1 and the constant region of the kappa chain. The antibody was expressed by co-transfection of the two constructs into HEK293T cells and purified using a Protein A column (catalog: 17-5438-02, GE Life Sciences®). The purified chimeric antibody was concentrated to 0.5–10 mg / mL in PBS, divided into aliquots, and stored in a freezer at -80°C.

[0196] Example 6. Humanization of anti-CLDN6 chimeric BG87P antibody (chBG87P) Humanization approach To humanize chBG87P, we searched for human germline IgG genes that share high homology with the protein sequence of the chBG87P variable region by sequence comparison in the IMGT human immunoglobulin gene database. Human IGHV and IGKV genes, which are frequently present in the human antibody repertoire and have high homology with chBG87P, were selected as templates for humanization.

[0197] Design of humanized variants Humanization was performed by CDR transplantation, followed by the incorporation of key revertant mutations. Humanized antibodies were engineered in human IgG1 wild-type format using a proprietary expression vector. In the initial first round of humanization, mutations from the mouse variable region to human amino acid residues in the framework region were induced by 3D structural analysis, and structurally important mouse framework residues for maintaining the standard structure of the CDR were preserved in the first round of humanization design. Five revertant mutations on the heavy chain and three mutations on the light chain were selected, and single-point mutations were performed to search for significant revertant mutations: BG87P-Bz1 (VH SEQ ID NO: 15 and VL SEQ ID NO: 14), BG87P-Bz2 (VH SEQ ID NO: 16 and VL SEQ ID NO: 14), BG87P-Bz3 (VH SEQ ID NO: 17 and VL SEQ ID NO: 14), BG87P-Bz4 (VH SEQ ID NO: 18 and VL SEQ ID NO: 14), BG87P-Bz5 (VH SEQ ID NO: 19 and VL SEQ ID NO: 14), BG87P-Bz6 (VH SEQ ID NO: 13 and VL SEQ ID NO: 20), BG87P-Bz7 (VH SEQ ID NO: 13 and VL SEQ ID NO: 20), and BG87P-Bz8 (VH SEQ ID NO: 13 and VL SEQ ID NO: 22). BG87P-Bz0 (VH SEQ ID NO: 13 and VL SEQ ID NO: 14) is a variant that incorporates all theoretically possible revertant mutations, and its binding capacity should be comparable to that of the parent chBG87P. Comparison of binding data will reveal which revertant mutations significantly affect binding. Specifically, the LCDR (SEQ ID NOs: 4-6) of chBG87P, along with the mouse framework residues A43S, L78V, and Y87F, will be compared with the human germline variable genes IGKV1-5 and 01-IGKJ4. *It was transplanted into the 01 framework (obtained as SEQ ID NO: 14). The HCDR of chBG87P (SEQ ID NOs: 1-3) was transplanted into the framework of the human germline variable genes IGHV1-3 and 01-JH6c, retaining the mouse framework residues V2I, T28S, I69L, R71V, and Y91F (obtained as SEQ ID NO: 13); BG87P-z0 (VH SEQ ID NO: 11 and VL SEQ ID NO: 12) is a resulting humanized variant in which the above HCDR and LCDR are transplanted, but there are no revertant mutations from the mouse VH and VL frameworks.

[0198] Expression and purification of chBG87P and humanized antibodies All of the first round of BG87P humanized variants (BG87P-z0, BG87P-Bz0, BG87P-Bz1, BG87P-Bz2, BG87P-Bz3, BG87P-Bz4, BG87P-Bz5, BG87P-Bz6, BG87P-Bz7, and BG87P-Bz8) were constructed as humanized full-length antibodies using a proprietary expression vector containing the constant regions of human wild-type IgG1 and kappa chain, respectively, and featuring easily adaptable subcloning sites. All humanized variants were expressed by co-transfection of the two constructs described above into HEK293T cells and purified using a Protein A column (catalog: 17-5438-02, GE Life Sciences). The purified antibodies were concentrated in PBS to 0.5–10 mg / mL, divided into aliquots, and stored in a freezer at -80°C.

[0199] Measurement of cell binding activity in the first round of humanized BG87P variant (hBG87P) and PTM-removed variants. To measure affinity, the binding activity of BG87P-related modified variants was evaluated using CLDN6-overexpressing HEK293T cells expressing high levels of human CLDN6 and the cancer cell line PA-1. Live cells were seeded in 96-well plates and incubated with a series of dilutions of chBG87P and its modified variants. Goat anti-human IgG was used as a secondary antibody to detect antibody binding to the cell surface. The EC50 values ​​of dose-dependent binding to CLDN6-expressing cell lines were determined by fitting dose-response data to a 4-parameter logistic model using GraphPad Prism. The cell binding activity of the first round of BG87P humanized variants to HEK293T / human CLDN6 was compared with chBG87P and is shown in Figure 1A. The cell binding affinity (EC50) and Emax (MFI) of the first round of humanized variants were normalized against chBG87P for direct comparison and ranking (Table 7).

[0200] Starting with the chBG87P antibody and BG87P-Bz0, we made several additional amino acid modifications in the CDR regions of both the VH and VL regions to further improve their biophysical properties for use in human therapy. These considerations include removing post-translational modifications (PTMs) while maintaining binding activity and improving thermal stability (Tm), resulting in variants such as BG87P-m1 (VH SEQ ID NO: 31 and VL SEQ ID NO: 8), BG87P-m2 (VH SEQ ID NO: 32 and VL SEQ ID NO: 8), BG87P-m3 (VH SEQ ID NO: 33 and VL SEQ ID NO: 8), and BG87P-m4 (VH SEQ ID NO: 34 and VL SEQ ID NO: 8), as well as BG87P-m5 (VH SEQ ID NO: 35 and VL SEQ ID NO: 14), BG87P-m6 (VH SEQ ID NO: 36 and VL SEQ ID NO: 14), BG87P-m7 (VH SEQ ID NO: 37 and VL SEQ ID NO: 14), and BG87P-m8 (VH SEQ ID NO: 38 and VL SEQ ID NO: 14). The cell binding activity of chBG87P and BG87P-Bz0-related PTM removal variants to HEK293T / human CLDN6 was compared with that of chBG87P and BG87-Bz0, respectively (Figure 1F). Cell binding affinity (EC50) and Emax (MFI) were normalized relative to chBG87P for direct comparison and ranking (Table 7). The results showed that, with the exception of the H33A mutation resulting in BG87P-m4 and BG87P-m8, other substitutions of potentially harmful residues maintained the binding ability of their respective parent residues. [Table 7] [Table 8]

[0201] Determination of cell-binding activity of critical reverse mutations in combination with PTM removal sites in the second round. After a comprehensive analysis of EC50 and Emax data from the first round of humanized cell binding (Table 7), four critical revertant sites, VH:V2I;VH:T28S;VH:I69L;VH:Y91F, were identified and combined with PTM removal sites for the second round of validation and determination of the final humanization candidates. The PTM removal mutation VH:V65G is a site with a high frequency of G occurrence in the human germline (G62%, V<1%), demonstrating a potential advantage in antibody framework stability. It was included in variant BG87P-m3 and showed improved Emax and EC50 compared to chBG87P, as shown in Figure 1F and Table 7. This was incorporated into the second round of combinations for further validation. Table 1 shows the VH and VL sequences of the second-round humanized variants BG87P-21, BG87P-22, BG87P-23, BG87P-24, BG87P-25, BG87P-26, and BG87P-27 obtained.

[0202] After confirming the cell-binding activity of the humanized combination variant, as shown in Figures 1B and 1C in HEK293T / human CLDN6 cells and Figures 1D and 1E in PA-1 cells, BG87P-21 was selected as a top humanized candidate for further investigation (VH and VL amino acid sequences are SEQ ID NOs. 24 and 12, respectively). BG87P-21 contains the significant revertant site VH:T28S and the PTM site VH:V65G, which demonstrated comparable cell-binding affinity compared to chBG87P. Emax was reduced by 22% in HEK293T / human CLDN6 and by 40% in PA-1 (Tables 7 and 8).

[0203] Feasibility assessment of humanized anti-CLDN6 antibodies Biophysical properties were profiled to identify top-performing humanized anti-CLDN6 antibodies. The data showed that BG87P-21 exhibited moderate to high hydrophobicity risk, followed by self-interaction risk in PBS buffer, as indicated by AC-SINS, B22KD, and CIC readings (Tables 9-11). [Table 9]

[0204] For hydrophobicity evaluation, 50 μg of the sample was diluted with mobile phase A solution (1.5 M ammonium sulfate, 50 mM sodium phosphate, pH 7.0) at 1 mg / ml to achieve a final ammonium sulfate concentration of approximately 1 M before analysis. A MABPac HIC-10 column was used with a linear gradient of mobile phase A and mobile phase B solutions (50 mM sodium phosphate, pH 7.0) over 29 minutes at a flow rate of 0.5 ml / min. Peak retention times were monitored at A280 absorbance. As shown in Table 9, both chBG87P and BG87P-21 showed higher hydrophobicity performance, exceeding the internal standard of 21.1 minutes in the IgG format.

[0205] For thermal stability evaluation, the thermal stability of BG87P-related modified variants was represented by the midpoint Tm (°C) of the thermal unfolding transition measured by extrinsic fluorescence. Tm was measured using an Applied Biosystems QuantStudio™ 6 Flex System. 20 μL of a 1 mg / ml sample was mixed with 20 μL of 40X SYPRO orange. The plate was scanned from 25 °C to 95 °C at a rate of 0.9 °C / min. Tm was assigned using the first derivative of the raw data from the QuantStudio™ 6 Flex System Analysis software. The results were summarized in Table 9. This suggests that both chBG87P and the humanized variant BG87P-21 showed good thermal stability.

[0206] To determine the aggregation tendency of modified variants related to BG87P, static light scattering intensity was measured using the Uncle system (Unchained Labs). During measurement, approximately 8.8 μL of protein sample at 1 mg / ml was loaded into a cuvette. The sample was held at 25°C for 120 seconds, and then heated to 95°C at a rate of 0.3°C / min. Scattering data was collected at a 90° angle using a 266 nm laser wavelength. Tagg (aggregation temperature) was analyzed and calculated using Uncle analysis software. The results are summarized in Table 9. Acceptable Tagg values ​​were shown for both chBG87P and humanized variants.

[0207] CIC is a technique for identifying antibody candidates with low solubility or nonspecific binding. Human serum-derived IgG or other ligands were chemically coupled to an NHS-activated chromatography resin. To evaluate protein solubility, the retention time of the protein in this resin was tested using HPLC. After coupling human serum-derived IgG to the column, the antibody sample and sample buffer were diluted to 0.1 mg / mL with mobile phase (PBS). The diluted sample and buffer were transferred to HPLC vials for LC-MS analysis. The results summarized in Table 9 suggest that both chBG87P and BG87P-21 exhibit acceptable nonspecific interactions with human IgG.

[0208] General description and purpose of the B22 and KD test methods. This method is used to study weak protein-protein interactions, predict aggregation tendencies, clarify the effect of formulation components on intermolecular interactions, and assist in the selection of formulation buffers. Antibodies were diluted to 1 mg / mL with buffer-exchanged samples and centrifuged at 14000 rpm for 30 minutes, after which Tm, Tagg, and DLS were checked. Samples were packed into Uni at 9 μL / well. Each sample was set up with a duplicate hole. Instrument parameters were set according to Uncle's guidance and the experiment was performed. In this experiment, the B22 & Kd mode was used. Execution information: Temperature (°C): 25. Incubation time (seconds): 120. Number of acquisitions: 4. Acquisition time (seconds): 5. Attenuator control: Automatic. Laser. Control: Automatic execution. kd: For the diffusion interaction parameter, if protein interactions increase (they are attracted to each other) as the concentration increases, the proteins behave as if they were larger and the diffusion coefficient (KD) decreases (negative gradient). B22: For the second virial coefficient, if protein interactions increase as concentration increases (they are attracted to each other), the proteins behave as if they become larger and 1 / R90 decreases (negative gradient). These data showed that in PBS, both chBG87P and the humanized variant attracted each other and tended to aggregate under these conditions (Table 10).

[0209] AC-SINS is an assay for obtaining self-interactions of a sample and predicting the likelihood of aggregation. It is based on concentrating antibodies from a diluted solution around gold nanoparticles pre-coated with polyclonal capture. Interactions between immobilized antibodies reduce the interparticle distance and increase the plasmon wavelength (wavelength of maximum absorbance), which can be easily measured by optical means. The antibodies were diluted to 0.05 mg / mL each in the supply buffer. After the preparation of the gold nanoparticles, the gold nanoparticle solution and the coating solution were mixed using a volume ratio of 9:1. After incubation at room temperature for 1 hour, empty sites of the AuNPs were blocked using tyrolized PEG (final concentration 0.1 μM). Then, incubation was performed at room temperature for a further 1 hour. The particle solution was then centrifuged at 15000 rpm for 6 minutes. The upper solution was discarded. The particles were redissolved using 1 / 10 of the starting volume of storage buffer. Ten μL of concentrated coated particles were incubated with 100 μL of test antibody solution in a polypropylene plate at room temperature for 2 hours, and then 90 μL of the resulting solution was transferred to a polystyrene UV transparent plate. These data suggested that both chBG87P and BG87P-21 exhibited suboptimal self-interaction tendencies (Table 9). [Table 10] [Table 11]

[0210] Due to the hydrophobic patch, the HIC retention time for chBG87P exceeded 25 minutes, and for humanized BG87P-21 it was 21.9 minutes, both higher than the acceptable threshold of 21.1 minutes for IgG form. The underlying cause is the hydrophobic patch of HCDR3, particularly the I97-Y98-Y100-V100a portion, which combines the FR2 (framework region 2) of the light chain and Y49-W50 (mainly W50) located at the end of LCDR2 (Figure 2A). Automated antibody pipeline analysis by Schrodinger regarding BG87P aggregation also indicated a high risk of agglutination (Table 12). [Table 12]

[0211] Example 7. Modification of the solubility of humanized anti-CLDN6 antibody Overall strategy for modifying the solubility of BG87P-21 As described above, chBG87P was engineered into a humanized antibody, and the inventors identified BG87P-21 as the final superior clone. However, the potential development risk of the HCDR3-driven hydrophobic patch of chBG87P remains unresolved (Figure 2). Given that chBG87P exhibits promising binding activity and excellent CLDN6 selectivity (Figures 1A, 1B, 1C, 1D, 1E, 1F, 1G, and 1H), further manipulation of BG87P-21 is being performed to remove the hydrophobic patch for optimal manufacturability and to mitigate potential ADA risks.

[0212] To address the solubility issues of BG87P-21, two main strategies were employed: single-point mutation and framework exchange. [Table 13-1] [Table 13-2]

[0213] Numerous single-point mutations were designed based on two theoretical considerations: one involving the substitution of hydrophobic amino acids with more hydrophilic ones, and the other involving the substitution of rare amino acids with more common ones at the same Kabat position within the human antibody repertoire. In the first round of screening, 57 variants were tested, and in the second round, 104 variants were tested. It was found that seven positions could be replaced with other more hydrophilic amino acids that had binding affinity comparable to the parent BG87P-21 and slightly improved hydrophilicity (Table 14). The top variants selected in the first and second rounds of screening were combined to generate 56 variants for further validation. The combined variants BG87P-31 and BG87P-32 (with VH and VL amino acid sequences of SEQ ID NOs. 46 and 42, respectively) were selected as top candidates because they possessed binding affinity comparable to BG87P-21 and improved HIC retention times of 17.4 minutes for BG87P-31 and 18.49 minutes for BG87P-32 (both superior to the parent BG87P-21's 22.3 minutes) (Table 14 and Figure 33). [Table 14]

[0214] However, even when hydrophobicity was reduced by a single-point mutation approach to solubility modification, the self-association risk, as determined by AC-SINS, still showed moderate to high levels. The reason for this unresolved problem is primarily the hydrophobic risk, which is reflected in the amount of hydrophobic patch actually exhibited on the antibody surface, while the reasons for inducing self-interactions involve isoelectric point issues, uniform charge distribution, and even unknown specific interactions (Doi.org / 10.1021 / mp200566k). Therefore, the self-interaction risk caused by the above reasons cannot be reduced solely by substituting hydrophobic residues with hydrophilic residues. Furthermore, we found that the parent chBG87P exhibited a lower tendency for self-interaction, with an AC-SINS value of around 12.85 nm in PBS buffer, while having a longer HIC retention time of 25 minutes (Table 14). Another finding is that the calculated net charge of chBG87P is even lower than that of BG87P-21, at 2.9 vs. 8.8. Therefore, the inventors hypothesized that the framework or net charge could influence the self-association effect.

[0215] Additional frameworks for IGHV3-23 and IGKV1-39 were tested. Both revertant mutations based on the new pair framework and selected top point mutations by soluble modification of BG87P-21 were incorporated (Table 13). The final top candidate, BG87P-34, showed no signs of problems in all routine biophysical properties (Table 14; VH and VL amino acid sequences are SEQ ID NOs. 43 and 44, respectively).

[0216] Furthermore, after the framework was switched to IGHV3-23 and IGKV1-39, the lost Emax of BG87P-21 was restored in the original humanization procedure (Figure 1G). The principle for the identification of the critical back mutations used in these two humanization procedures is the same, and thus the possibility that any back mutations related to the Emax problem in the previous round of humanization was excluded was ruled out. One explanation for the restoration of Emax in cell binding by framework exchange is that the VH-VL angles of different pairs of frameworks can vary, while a specific VH-VL angle may help maintain the Emax of cell binding. The final lead clone BG87P-34 shows good cross-reactivity in different species (Figure 1I and Figure 1J). Non-specific binding to human CLDN9 was performed using HEK293T / human CLDN9, and the data showed that BG87P-34 has better selectivity for human CLDN6 than for CLDN9 (Figure 1H).

[0217] Example 8. Humanization of anti-human CD3 antibody sp34 and modification of scFv The well-reported mouse clone sp34 (Blumberg 1990 PNAS 87(18):7220-24) was an optimal clone for developing anti-CD3-based therapeutic agents due to its cynomolgus CD3 cross-reactivity. For the humanization of sp34, sequences sharing a high degree of homology with the protein sequences of the variable regions of sp34 (SEQ ID NOs: 48-57) were searched for human germline IgG genes by blasting the human immunoglobulin gene databases of IMGT (http: / / www(dot)imgt(dot)org / IMGT_vquest / share / textes / index(dot)html) and NCBI (http: / / www(dot)ncbi(dot)nlm(dot)nih(dot)gov / igblast / ). Human IGVH genes and IGVK genes that are present at high frequencies in the human antibody repertoire (Glanville et al., 2009 PNAS 106:20216-20221) and are homologous to sp34 were selected as humanization templates.

[0218] Humanization was performed using the CDR transplantation method (Methods in Molecular Biology, Vol 248: Antibody Engineering, Methods and Protocols, Humana Press), and the humanized antibody (hu-sp34) was manipulated in human IgG1 format using a proprietary expression vector. In the first round of humanization, mutations from mouse amino acid residues to human amino acid residues in the framework region were induced by a simulated 3D structure, and structurally important mouse framework residues for maintaining the standard structure of the CDR were retained in the first version of the humanized antibody sp34. Specifically, the CDRs of sp34 VL (SEQ ID NOs. 51-53) were transplanted into the framework of the human germline variable region gene IGVκ3-15 while retaining several mouse framework residues (Q1, A2, V4, V36, E38, L43, F44, T45, G46, G49, L66, D69, A71, I85, and F87). The CDRs (SEQ ID NOs. 48-50) of sp34 VH were transplanted into the framework of the human germline variable region gene IGVH3-7 while retaining several mouse framework residues (D73, S76, M89, V93).

[0219] Humanized sp34 (hu-sp34) and chimeric sp34 (ch-sp34) were constructed as full-length human antibody formats using proprietary expression vectors containing the constant regions of human IgG1 and kappa chains, respectively, which have easily adaptable subcloning sites. Expression and preparation of humanized sp34 and chimeric sp34 antibodies were achieved by co-transfection of heavy chain constructs and corresponding light chain constructs into 293G cells (proprietary development) and purification using a Protein A column. The purified antibodies were concentrated to 0.5-5 mg / mL in PBS and aliquoted for use in the following assays, and stored in a -80°C freezer.

[0220] For affinity measurements, antibodies were captured on an anti-human Fc surface and used in affinity assays based on surface plasmon resonance (SPR) technology. The binding activity of humanized sp34 to native CD3 on live cells was evaluated using HuT78 cells in a FACS-based assay. Live HuT78 cells were seeded in 96-well plates and incubated with a series of dilutions of chimeric or humanized sp34. Mouse anti-human IgG was used as a secondary antibody to detect antibody binding to the cell surface. The EC50 values ​​for dose-dependent binding to human native CD3 were determined by fitting dose-response data to a 4-parameter logistic model using GraphPad Prism. Humanized sp34 BG53P (SEQ ID NOs. 48-53 and 58-61) showed comparable binding affinity to ch-sp34 in both SPR and FACS assays (Table 15 and Figure 4A). [Table 15]

[0221] Based on the humanized sp34 BG53P template, several single mutations were generated that converted retained mouse residues within the framework region to corresponding human germline residues, including four retained mouse residues in VH (D73, S76, M89, V93) and fifteen retained mouse residues in VL (Q1, A2, V4, V36, E38, L43, F44, T45, G46, G49, L66, D69, A71, I85, and F87). All humanized mutations were generated using primers containing mutations at specific locations and a site-directed mutagenesis kit (catalog number FM111-02, TransGen, Beijing, China). Desired mutations were validated by sequencing analysis. These hu-sp34 variant antibodies were tested in the aforementioned binding assays. Compared to hu-sp34-1A-1f, mutations in VK V36Y, G46L, and G49Y (Kabat numbered) significantly impaired the binding affinity of the humanized variant, while the remaining versions of the hu-sp34 humanized variant exhibited binding activity comparable to hu-sp34-1A-1f. D73N in VH significantly reduced expression levels (data not shown).

[0222] In summary, BG56P (SEQ ID NOs. 70-77 and 72-86), a well-engineered version of the humanized monoclonal antibody, was obtained from the mutation process described above and characterized in detail (Table 16 and Figure 4B). [Table 16]

[0223] Example 9: ScFv modification of humanized sp34 To generate a plug-and-play bispecific format and avoid light-heavy chain mispairing, the BG56P antibody was reformatted into a single-stranded fragment variable (scFv) format containing a 3xG4S linker between VH and VK. The reformatted scFv was fused to the N-terminus of the human IgG1 Fc region to form the scFv-Fc format using a proprietary expression vector containing an easily adaptable subcloning site. Expression and preparation of the parental and redesigned hu-sp34 scFv-Fc were achieved by transfection of the scFv-Fc construct into 293G cells (proprietary) and purification using a Protein A column. The purified scFv-Fc antibody was concentrated to 0.5–5 mg / mL in PBS and aliquoted for use in the following assays, and stored in a -80°C freezer. scFv-modified BG56P (referred to as BG561P, sequence numbers 48-53 and 62-65) showed comparable binding affinity to the antibody version of BG56P in SPR and FACS (Table 17 and Figure 5). [Table 17]

[0224] Based on BG561P, several mutations were made to the framework and CDR to remove potential PTM sites and improve thermal and colloidal stability for therapeutic use in humans. The L4V mutation in VL (the resulting humanized scFv is called BG562P, SEQ ID NOs. 48-53 and 69-70) showed a 5-degree improvement in the aggregation temperature (Tagg). The combination of L4V in VL and A49G and D65G in VH (the resulting humanized scFv is called BG563P) (SEQ ID NOs. 48, 71, 50, 51-53, 73, and 74) showed improved thermal and colloidal stability compared to BG561P, and a slight improvement in binding affinity to human CD3 in FACS assays. Potential PTM sites include the potential deamidation site N30 (NT) in the junction region between FR1 and HCDR1 (Kabat's definition of CDR), and N100 (NS) in HCDR3. Each N was mutated to S to eliminate potential deamidation sites. All mutations were generated using primers containing mutations at specific locations and a site-directed mutagenesis kit (catalog number FM111-02, TransGen, Beijing, China). In summary, BG564P (sequence IDs 48, 71, 75, 51-53, 77, and 78), which are well-manipulated versions of humanized scFv, were obtained from the above mutation process and characterized in detail. These results showed that humanized scFv BG564P retained its binding affinity to CD3 (Tables 18-20 and Figure 6) and exhibited improved biophysical stability compared to humanized scFv BG561P (Table 20). [Table 18] [Table 19] [Table 20]

[0225] The melting temperature (Tm) was measured using a high-throughput MicroCal® VP-Capillary DSC (Malvern Instruments, Northampton, MA). Thermograms of each protein (350 μL at 0.5 mg / mL) were acquired at 20°C to 100°C using a scanning rate of 90°C / hour. Thermograms of the buffer solution alone were subtracted from each protein sample. The results show the midpoint of the transition temperature (Tm) and the calorimetric enthalpy (ΔH) of the samples, suggesting that the Tm of BG564P was improved compared to BG561P (Table 20).

[0226] The aggregation temperature Tag (°C) represents the colloidal stability of the sample and was obtained by monitoring the onset of aggregation using SLS266 with UNCLE® (Unchained lab, Pleasanton, CA). The sample was packed into the Uni and the temperature was increased from 15°C to 95°C. The back-reflected optics could not detect near-ultraviolet light scattering by protein aggregates, and therefore only unscattered light reached the detector. Thus, the reduction in back-reflected light is a direct measure of aggregation in the sample, suggesting that the Tag of BG564P was improved compared to BG561P (Table 20).

[0227] Example 10. Production of CLDN6×CD3 BsAb BG143P Agonist anti-CD3 antibodies have demonstrated toxicity in clinical settings, which may indicate that systemic FcγR crosslinking is not ideal for CD3 activation. The objective was to achieve tumor-specific and potent CD3 stimulation for a wide range of cancers without systemic CD3 activation. To overcome the dependence on FcγR crosslinking, we generated CLDN6xCD3 BsAb BG143P with the following features shown in Figure 7. This particular construct BG143P includes an IgG fusion-like multispecific antibody format with a module ratio of 1:1, a well-engineered Fab fragment BG87P-34 that binds to CLDN6, and scFv of BG564P that binds to the CD3 fusion at the N-terminus of CH2, as well as an Fc-null version of huIgG1 that lacks FcγR binding but retains FcRn binding. Knob into Hole (KIH) was also introduced to Fc to increase heterodimerization. Sequence information for BG143P is listed in SEQ ID NOs. 79-84.

[0228] Example 11. Target binding activity of CLDN6×CD3 BsAb BG143P The binding reaction rate of CLDN6×CD3 BsAb BG143P was measured using SPR. The on-rate constant (k) of the antibody against recombinant CDεγ was measured using SPR. a ) and off-speed constant (k d ) is measured, and then the affinity constant (K D The results showed that CLDN6×CD3 BsAb has a strong binding affinity to human CDεγ, as shown in Table 21. [Table 21]

[0229] FACS results further confirmed the binding activity of BG143P to CD3 and CLDN6. BsAb was 6.98 nM EC 50 It showed dose-dependent strong binding activity to CD3-expressing Jurkat (Figure 8A). Similarly, BG143P showed dose-dependent strong binding activity to CLDN6-expressing PA-1, and EC 50The value was 81.26 nM (Figure 8B).

[0230] Example 12. In vitro functional activity of CLDN6 × CD3 antibody On-target T cells redirect cytotoxicity and cytokine release. The T-cell redirection cytotoxicity of BG143P against PA-1 (a cancer cell line with high CLDN6 expression), Hutu80 (a cancer cell line with moderate CLDN6 expression), AGS (a cancer cell line with low and heterogeneous CLDN6 expression), and NCI-H1299 (a cancer cell line negative for CLDN6 expression) was evaluated using human PBMCs as effector cells. To measure cytotoxicity, target cancer cell lines were modified to express nanoluciferase. Approximately 10,000 target cells and 25,000 human PBMCs (E / T=2.5) were seeded in each well of a 96-well U-bottom plate and incubated with various antibody concentrations at 37°C and 5% CO2 for 48 hours. The supernatant was collected for cytokine detection. Target cell killing was measured using the Nano-Glo detection kit (Promega). The cytotoxic activity (%) of the antibody was calculated using the following formula: Cytotoxic activity (%) = (AB) / (AC) * 100%. "A" represents the average luminescence signal of wells containing only untreated target cells, "B" represents the average luminescence signal of wells containing antibodies and PBMCs, and "C" represents the average luminescence signal of wells containing target cells completely lysed with Triton-X100. IFN-γ and IL-2 were detected in the supernatant using an HTRF kit (Cisbio).

[0231] As shown in Figure 9, BG143P demonstrated dose-dependent potent T cell redirection-killing and cytokine release-inducing efficacy at the pM EC50 level.

[0232] Functional specificity for human CLDN6 and CLDN9 The amino acid sequences of human CLDN6 and CLDN9 are highly conserved, differing in only three amino acids in the extracellular domain. Since CLDN9 is widely expressed in normal human tissues, the binding specificity between CLDN6 and CLDN9 is important and can be investigated by FACS analysis.

[0233] Human CLDN6 and CLDN9 expression vectors were established by inserting synthetic cDNA encoding the corresponding sequences into mammalian expression vectors. Stable NCI-H1299 cells expressing human CLDN6 and CLDN9 were generated by transfection with the corresponding plasmids. Cells were then placed in FACS buffer (2% FBS, 1×PBS) at a rate of 1×10⁶ cells. 6 The cells were suspended at their concentration, and the cell suspension was dispensed into a U-bottom 96-well plate (100 μL / well). The antibody was added to the cells at a final peak concentration of 100 nM, diluted 11 times by 2-fold dilution, mixed with the cells, and incubated at 4°C for 1 hour. After centrifugation, the reaction solution was removed, and the cells were washed twice with 200 μL / well of FACS buffer. Next, APC-anti-human Fcγ was diluted 500-fold with FACS buffer and added to the cells as a secondary antibody. The cells were incubated at 4°C for 30 minutes, then washed twice as described above, and suspended in 100 μL of FACS buffer. The cell suspension was subjected to flow cytometry.

[0234] The killing of NCI-H1299-CLDN6 / CLDN9 cells by BG143P was evaluated using a Nano-Glo assay with human PBMCs. Approximately 10,000 target cells and 25,000 human PBMCs (E / T=2.5) were seeded in each well of a 96-well U-bottom plate and incubated with various antibody concentrations at 37°C and 5% CO2 for 48 hours. The supernatant was collected for cytokine detection. Target cell killing was measured using a Nano-Glo detection kit (Promega). The cytotoxic activity (%) of the antibody was calculated using the following formula: Cytotoxic activity (%) = (AB) / (AC) *100%. "A" represents the average luminescence signal of wells containing only untreated target cells, "B" represents the average luminescence signal of wells containing antibodies and PBMCs, and "C" represents the average luminescence signal of wells containing target cells completely lysed with Triton-X100. IFN-γ was detected using an HTRF kit (Cisbio).

[0235] As shown in Figure 10, BG143P was an antibody that specifically bound to human CLDN6 (Figure 10A), killed cells (e.g., lysed) (Figure 10B), and had IFN-γ inducing activity (Figure 10B), but did not have this activity against human CLDN9.

[0236] Example 13. In vivo efficacy of CLDN6×CD3 BsAb BG143P in an OV90 xenograft model. The in vivo antitumor effect of CLDN6×CD3 BsAb BG143P was evaluated in a xenograft model in PBMC humanized mice. Human ovarian cancer cell line OV-90 (ATCC) expressing human CLDN6 was used in NCG(NOD / ShiLtJGpt-Prkdc em26Cd52 Il2rg em26Cd22 (Gpt) Subcutaneous inoculation was performed into mice, and human PBMCs were intravenously injected into the mice the following day. Tumor volume was approximately 200 mm². 3 When tumor-bearing mice reached a certain stage, they were randomized to a treatment group and administered either antibody or a control medium (PBS). Antibody / medium was administered once a week. The length (L) and width (W) of the tumor mass, as well as body weight, were measured three times a week for each mouse. Tumor volume (TV) was also calculated as TV = (L × W). 2 The calculation was performed using ) / 2. Figure 11A shows the in vivo antitumor effect of BG143P, which showed strong efficacy at 0.03 mg / kg and 0.1 mg / kg, with TGI% (tumor growth inhibition ratio, %) of 115.43% and 125.92%.

[0237] hPBMC rearrangement in mice was confirmed at weeks 2, 3, and 4 after PBMC injection. Among the viable cells in peripheral blood, hCD45+ cells increased from 20% at week 2 to 60% at week 4. Figure 11B shows the rearrangement of hPBMCs.

[0238] Example 14. In vivo efficacy of CLDN6×CD3 BsAb BG143P in a B16F10- / hCLDN6 syngeneic model. Another type of efficacy model was performed to evaluate the in vivo efficacy of CLDN6×CD3 BsAb BG143P. A human CLDN6 expression plasmid was constructed and stably transfected into the B16F10 cell line. The resulting B16F10 / human CLDN6 cell line could be proliferated in human CD3EDG transgenic mice, and it was confirmed that hCLDN6 expression was maintained even after tumor formation. To establish this model, B16F10 / human CLDN6 cells were subcutaneously inoculated into hCD3EDG transgenic mice, and the mouse CD3 gene was replaced with a human counterpart. The tumor volume was approximately 100 mm². 3 Mice were randomized after reaching a certain stage. The test substance or PBS was injected intraperitoneally into the mice weekly. The length (L) and width (W) of the tumor mass, as well as body weight, were measured three times a week for each mouse. Tumor volume (TV) was calculated as TV = (L × W²) / 2. BG143P at 0.1 mg / kg showed potent efficacy with a TGI% of 93.54%, as shown in Figure 12A. No significant weight loss was observed in this study, as shown in Figure 12B.

[0239] literature 1. Krause, G., et al., Structure and function of claudins. Biochimica et Biophysica Acta (BBA)-Biomembranes, 2008.1778(3):p. 631-645. 2. Guenzel, D. and AS Yu, Claudins and the modulation of tight junction permeability. Physiol Rev,2013. 93(2): p. 525-69. 3. Tsukita, S., H. Tanaka, and A. Tamura, The Claudins: From Tight Junctions to Biological Systems. Trends in Biochemical Sciences,2019.44(2):p. 141-152。 4. Matter, K. and M.S. Balda, Signalling to and from tight junctions. Nat Rev Mol Cell Biol,2003. 4(3): p. 225-36。 5. Singh, A.B., S.B. Uppada, and P. Dhawan, Claudin proteins, outside-in signaling, and carcinogenesis. Pflugers Arch, 2017. 469(1): p. 69-75。 6. Turksen, K. and T.-C. Troy, Claudin-6: A novel tight junction molecule is developmentally regulated in mouse embryonic epithelium. Developmental Dynamics, 2001. 222(2): p. 292-300。 7. Du, H., et al., Claudin 6: Therapeutic prospects for tumours, and mechanisms of expression and regulation (Review). Mol Med Rep,2021.24(3). 8. Ben-David, U., N. Nudel, and N. Benvenisty, Immunologic and chemical targeting of the tight-junction protein Claudin-6 eliminates tumorigenic human pluripotent stem cells. Nat Commun, 2013. 4: p. 1992。 9. Reinhard, K., et al., An RNA vaccine drives expansion and efficacy of claudin-CAR-T cells against solid tumors. Science, 2020. 367(6476): p. 446-453。 10. Abuazza, G., et al., Claudins 6, 9, and 13 are developmentally expressed renal tight junction proteins. Am J Physiol Renal Physiol, 2006. 291(6): p. F1132-41。 11. Hashizume, A., et al., Expression patterns of claudin family of tight junction membrane proteins in developing mouse submandibular gland. Dev Dyn, 2004. 231(2): p. 425-31。 12. Kohmoto, T., et al., Claudin-6 is a single prognostic marker and functions as a tumor-promoting gene in a subgroup of intestinal type gastric cancer. Gastric Cancer,2020.23(3):p. 403-417。 13. Kojima, M., et al., Prognostic Significance of Aberrant Claudin-6 Expression in Endometrial Cancer. Cancers (Basel), 2020. 12(10)。 14. Micke, P., et al., Aberrantly activated claudin 6 and 18.2 as potential therapy targets in non-small-cell lung cancer. Int J Cancer, 2014. 135(9): p. 2206-14。 15. Sullivan, L.M., et al., Claudin-6 is a nonspecific marker for malignant rhabdoid and other pediatric tumors. Am J Surg Pathol, 2012. 36(1): p. 73-80。 16. Ushiku, T., et al., Distinct expression pattern of claudin-6, a primitive phenotypic tight junction molecule, in germ cell tumours and visceral carcinomas. Histopathology,2012.61(6):p. 1043-56。 17. Kamakura, D., R. Asano, and M. Yasunaga, T Cell Bispecific Antibodies: An Antibody-Based Delivery System for Inducing Antitumor Immunity. Pharmaceuticals (Basel), 2021. 14(11)。 18. Middelburg, J., et al., Overcoming Challenges for CD3-Bispecific Antibody Therapy in Solid Tumors. Cancers (Basel), 2021. 13(2)。 19. Singh, A., S. Dees, and I.S. Grewal, Overcoming the challenges associated with CD3+ T-cell redirection in cancer. Br J Cancer, 2021. 124(6): p. 1037-1048。 20. Baeuerle, P.A. and H. Wesche, T-cell-engaging antibodies for the treatment of solid tumors: challenges and opportunities. Current Opinion in Oncology,2022.34(5)。 21. Stadler, C.R., et al., Characterization of the first-in-class T-cell-engaging bispecific single-chain antibody for targeted immunotherapy of solid tumors expressing the oncofetal protein claudin 6. Oncoimmunology, 2016. 5(3): p. e1091555。 22. Stadler, C.R., et al., Elimination of large tumors in mice by mRNA-encoded bispecific antibodies. Nat Med, 2017. 23(7): p. 815-817。

Claims

1. An antibody or its antigen-binding fragment, comprising an antigen-binding domain that specifically binds to human differentiation cluster 3 (CD3).

2. The antigen-binding domain that specifically binds to human CD3 is: (a) A heavy chain variable region comprising (i) a heavy chain complementarity-determining region (HCDR) 1 having the amino acid sequence of SEQ ID NO: 48, (ii) an HCDR 2 having the amino acid sequence of SEQ ID NO: 71, (iii) an HCDR 3 having the amino acid sequence of SEQ ID NO: 75, and (iv) a light chain variable region comprising a light chain complementarity-determining region (LCDR) 1 having the amino acid sequence of SEQ ID NO: 51, (v) an LCDR 2 having the amino acid sequence of SEQ ID NO: 52, and (vi) an LCDR 3 having the amino acid sequence of SEQ ID NO: 53; or (b) A heavy chain variable region comprising (i) HCDR1 having the amino acid sequence of SEQ ID NO: 48, (ii) HCDR2 having the amino acid sequence of SEQ ID NO: 71, (iii) HCDR3 having the amino acid sequence of SEQ ID NO: 50, and (iv) a light chain variable region comprising LCDR1 having the amino acid sequence of SEQ ID NO: 51, (v) LCDR2 having the amino acid sequence of SEQ ID NO: 52, and (vi) LCDR3 having the amino acid sequence of SEQ ID NO:

53.

3. The aforementioned antigen-binding domain is: (i) A heavy chain variable region having at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequence to SEQ ID NO: 76, and a light chain variable region having at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequence to SEQ ID NO: 68; (ii) A heavy chain variable region having at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequence to SEQ ID NO: 58, and a light chain variable region having at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequence to SEQ ID NO: 59; (iii) A heavy chain variable region having at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequence to SEQ ID NO: 62, and a light chain variable region having at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequence to SEQ ID NO: 63; (iv) A heavy chain variable region having at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequence to SEQ ID NO: 62, and a light chain variable region having at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequence to SEQ ID NO: 68; (v) The antibody or antigen-binding fragment according to any one of the prior claims, comprising a heavy chain variable region having at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequence to SEQ ID NO: 72, and a light chain variable region having at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identical amino acid sequence to SEQ ID NO:

68.

4. An antibody or antigen-binding fragment according to any one of the prior claims, wherein one, two, three, four, five, six, seven, eight, nine, or ten amino acids in SEQ ID NO: 62, 63, 68, 72, or 76 are inserted, deleted, or substituted.

5. The aforementioned antigen-binding domain is: (i) A heavy chain variable region having an amino acid sequence containing SEQ ID NO: 76, and a light chain variable region having an amino acid sequence containing SEQ ID NO: 68; (ii) A heavy chain variable region having an amino acid sequence containing SEQ ID NO: 58, and a light chain variable region having an amino acid sequence containing SEQ ID NO: 59; (iii) A heavy chain variable region having an amino acid sequence containing SEQ ID NO: 62, and a light chain variable region having an amino acid sequence containing SEQ ID NO: 63; (iv) A heavy chain variable region having an amino acid sequence containing SEQ ID NO: 62, and a light chain variable region having an amino acid sequence containing SEQ ID NO: 68; or (v) The antibody or antigen-binding fragment according to any one of the prior claims, comprising a heavy chain variable region having an amino acid sequence including SEQ ID NO: 72, and a light chain variable region (VL) including SEQ ID NO:

68.

6. The aforementioned antigen-binding domain is: (i) A single-stranded variable fragment (scFv) having the amino acid sequence containing SEQ ID NO: 77, (ii) A single-stranded variable fragment (scFv) having the amino acid sequence containing SEQ ID NO: 66, (iii). Single-stranded variable fragment (scFv) having an amino acid sequence including sequence number 69, (iv) The antibody or antigen-binding fragment according to any one of the prior claims, comprising a single-stranded variable fragment (scFv) having an amino acid sequence including SEQ ID NO:

73.

7. An antibody or antigen-binding fragment according to any one of the prior claims, which is a monoclonal antibody, a chimeric antibody, a humanized antibody, a human-modified antibody, a single-chain antibody (scFv), a Fab fragment, a Fab' fragment, an F(ab')2 fragment, or a multispecific antibody.

8. The antibody or antigen-binding fragment according to any one of the prior claims, wherein the antibody is a bispecific antibody.

9. The antibody or antigen-binding fragment according to claim 8, wherein the antibody is BG143P.

10. The antibody or antigen-binding fragment thereof has antibody-dependent cell-mediated cytotoxicity (ADCC) or complement-dependent cell-mediated cytotoxicity (CDC), as described in any one of the prior claims.

11. The antibody or antigen-binding fragment according to any one of the prior claims, wherein the antibody or antigen-binding fragment has reduced glycosylation, is not glycosylated, or is low-fucosylated.

12. The antibody or antigen-binding fragment according to any one of the prior claims, comprising an increased bisecting GlcNac structure.

13. The antibody or antigen-binding fragment according to any one of the prior claims, wherein the Fc domain is IgG1 with reduced effector function.

14. The antibody or antigen-binding fragment according to any one of the prior claims, wherein the Fc domain is IgG4.

15. A pharmaceutical composition comprising an antibody or antigen-binding fragment according to any one of claims 1 to 14, further comprising a pharmaceutically acceptable carrier.

16. The pharmaceutical composition according to claim 15, further comprising histidine / histidine HCl, trehalose dihydrate, and / or polysorbate 20.

17. A method for treating cancer, comprising administering an effective amount of the antibody or antigen-binding fragment described in claims 1 to 14 to a patient in need.

18. The method according to claim 17, wherein the cancer is a solid tumor.

19. The method according to claim 17, wherein the cancer is selected from gastric cancer, colon cancer, pancreatic cancer, breast cancer, head and neck cancer, kidney cancer, liver cancer, lung cancer, small cell lung cancer, non-small cell lung cancer, ovarian cancer, skin cancer, mesothelioma, lymphoma, leukemia, myeloma, sarcoma, brain cancer, colorectal cancer, prostate cancer, cervical cancer, testicular cancer, endometrial cancer, bladder cancer, rhabdoid tumor and / or glioma.

20. The method according to claim 19, wherein the antibody or the antigen-binding fragment is administered in combination with one or more further therapeutic agents.

21. The method according to claim 20, wherein the one or more therapeutic agents are selected from paclitaxel or a paclitaxel preparation, docetaxel, carboplatin, topotecan, cisplatin, irinotecan, doxorubicin, lenalidomide, or 5-azacitidine.

22. The method according to claim 21, wherein the one or more therapeutic agents are paclitaxel, lenalidomide, or 5-azacitidine.

23. The method according to claim 20, wherein at least one of the one or more therapeutic agents is an anti-PD1 or anti-PDL1 antibody.

24. The method according to claim 23, wherein the anti-PD1 antibody is tislerizumab.

25. An isolated nucleic acid encoding an antibody or antigen-binding fragment according to any one of claims 1 to 14.

26. A vector comprising the nucleic acid described in claim 25.

27. A host cell comprising the nucleic acid described in claim 25 or the vector described in claim 26.

28. A step for producing an antibody or an antigen-binding fragment thereof, comprising culturing the host cells described in claim 27, and recovering the antibody or the antigen-binding fragment from the culture.

29. An antibody or antigen-binding fragment according to any one of claims 1 to 14, or a pharmaceutical composition according to claim 15 or 16, for use in pharmaceuticals or therapeutics.

30. An antibody or antigen-binding fragment according to any one of claims 1 to 14, or a pharmaceutical composition according to claim 15 or 16, for use in a method of treating cancer.

31. Use of an antibody or antigen-binding fragment according to any one of claims 1 to 14 in the manufacture of a pharmaceutical product for treating cancer.

Images