Anti-cd137 antibodies and methods of use

Abstract

The present disclosure provides antibodies or antigen-binding fragments thereof that specifically bind to human CD137, multispecific antibodies and antigen-binding fragments thereof that specifically bind to human GPC3 and CD137, pharmaceutical compositions comprising the antibodies, and uses of the antibodies, multispecific antibodies, or the compositions for treating diseases, e.g., cancer.

Description

[0001] This application is a divisional application of the invention patent application filed on November 20, 2023, with Chinese application number 202380079710.5 and entitled "Anti-CD137 Antibody and Method of Use". Technical Field

[0002] This article discloses antibodies that specifically bind to human CD137 (TNF receptor superfamily member 9 (TNFRSF9)), multispecific antibodies or antigen-binding fragments thereof that bind to human CD137, compositions comprising said antibodies, and methods or uses for treating cancer. Background Technology

[0003] Phosphatidylinositol proteoglycan 3 (GPC3) belongs to the heparan sulfate proteoglycan (HSPG) family and includes a 60-70 kDa core protein that is attached to the cell membrane surface via a glycosyl phosphatidylinositol anchor (GPI). The carboxyl terminus of GPC3 is modified with a heparan sulfate side chain (Filmus J et al., J. Clin. Inv. [Journal of Clinical Research] 2001; 108: 497-501).

[0004] The specific expression of GPC3 in tumor cells has attracted much attention. GPC3 is expressed in hepatocellular carcinoma (HCC), the most common type of liver cancer. Notably, its expression has not been detected in non-malignant tissues. Overexpression of GPC3 has also been reported in hepatoblastoma, squamous cell carcinoma of the lung (LSCC), and other cancers. Therefore, GPC3 is suitable as a tumor antigen for targeted therapy. (Li N et al., Trends Cancer. 2018; 4: 741-54; Ho M et al., Eur J Cancer. 2011; 47: 333-8; Moek et al., Am. J. Pathol. 2018; 188 (9): 1973-1981).

[0005] CD137 (also known as TNFRSF9 / 41BB) is a costimulatory molecule belonging to the TNFRSF family. It was discovered through T-cytokine screening of mouse helper cells and cytotoxic cells stimulated by concanavalin A. It was identified in 1989 as an inducible gene that is expressed on antigen-induced T cells but not on resting T cells (Kwon et al., Proc. Natl. Acad. Sci. USA. [Proceedings of the National Academy of Sciences] 1989; 86:1963–1967). In addition, it is known to be expressed in the following cells: dendritic cells (DCs), natural killer (NK) cells (Vinay et al., Mol. Cancer Ther. [Molecular Cancer Therapy] 2012; 11:1062–1070), activated CD4+ and CD8+ T lymphocytes, eosinophils, natural killer T cells (NKTs), and mast cells (Kwon et al., 1989 ibid.; Vinay D., Int. J. Hematol. [International Journal of Hematology] 2006; 83:23–28). CD137 maintains and enhances immune effector function by inducing the production of Th1 cytokines (Bartkowiak et al., Front Oncol. [Frontiers in Oncology] 2015;5:117; Shuford et al., J Exp Med. [Journal of Experimental Medicine] 1997; 186:47-55). Upon binding to its only ligand (CD137L, 4-1BBL, or TNFSF9), CD137 signaling is activated via the NF-κB pathway, leading to increased expression of pro-survival molecules (Wang et al., Immunol Rev. 2009; 229:192-215).

[0006] The anti-CD137 antibody urrelumab (BMS-663513), which binds to CRD I of CD137, and utomilumab (PF-05082566), which binds to CRD III and IV of CD137, have shown potential as cancer therapeutics due to their ability to activate cytotoxic T cells and increase the production of interferon-γ (IFN-γ). The mechanism by which these antibodies regress tumors is their enhancement of the cancer immune cell response. In particular, anti-CD137 antibodies stimulate and activate effector T lymphocytes (e.g., by stimulating CD8+ T lymphocytes to produce INF-γ) and enhance the production of NKT and APCs (e.g., macrophages).

[0007] Urelumab showed promising results in preclinical trials and early clinical studies (Sznol et al., Clin. Oncol. [Clinical Oncology] 2008; 26(Supplement 15)). However, in later studies, urrelumab exhibited hepatotoxicity, leading to a pause in the antibody's development until February 2012 (Segal et al., Clin. Cancer Res. [Clinical Cancer Research] 2017; 23:1929–1936). The hepatotoxicity was primarily due to the S100A4 protein secreted by tumor and stromal cells. Studies that limited the dosage of urrelumab to 8 mg or 0.1 mg / kg per patient every 3 weeks rekindled interest in the antibody (Segal et al., Clin. Cancer Res. [Clinical Cancer Research] 2017; 23:1929–1936).

[0008] Compared to urogenumab, urogenumab has shown a better safety profile, and preliminary studies have shown no hepatotoxicity or other dose-limiting factors (Segal et al., J. Clin. Oncol. [Journal of Clinical Oncology] 2014;32(Supplement 15)). Reports from a phase I trial of urogenumab as monotherapy indicate a favorable safety profile (Segal et al., Clin. Cancer Res. [Clinical Cancer Research] 2018; 24:1816–1823). It is speculated that the difference between these two antibodies is due to their different binding sites on the CD137 receptor.

[0009] There are currently no approved therapeutic antibodies targeting CD137, and the medical need for CD137-targeted therapeutics remains unmet. Furthermore, anti-TAAxCD137 multispecific antibodies, which recruit immune cells to cancers expressing tumor-associated antigens (TAAs), could be used to treat cancer. Summary of the Invention

[0010] This disclosure contains antibodies and antigen-binding fragments that specifically bind to human CD137. Furthermore, the CD137 VHH domain fragment disclosed herein can be used to construct multispecific antibodies in combination with other forms, such as TAAs, immune checkpoints, or immunostimulatory factors. CD137 antibodies, alone or in combination with other forms, have the potential to treat or prevent cancer, autoimmune diseases, or infectious diseases.

[0011] In addition, this disclosure relates to a multispecific anti-GPC3xCD137 antibody and its antigen-binding fragment.

[0012] This disclosure covers the following embodiments.

[0013] Example 1: An antibody or its antigen-binding fragment that specifically binds to human CD137, wherein: (1) The antibody or its antigen-binding fragment specifically binds to the following epitopes, which contain or are composed of amino acid residues Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64, Gln67 of human CD137 (SEQ ID NO:35). (2) The antibody or its antigen-binding fragment specifically binds to the following epitope, which comprises, or is composed of, the amino acid residues Ser55, Ala56, Arg75, Glu85, Ala97, and Gly98 of human CD137 (SEQ ID NO:35); or (3) The antibody or its antigen-binding fragment specifically binds to a human CD137 dimer, which comprises a first human CD137 monomer and a second human CD137 monomer, or is composed of the latter; wherein the antibody or its antigen-binding fragment specifically binds to an epitope of the first human CD137 monomer (SEQ ID NO:35), which comprises amino acid residues Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64, Gln67, or is composed of the latter; and the antibody or its antigen-binding fragment specifically binds to an epitope of the second human CD137 monomer (SEQ ID NO:35), which comprises amino acid residues Ser55, Ala56, Arg75, Glu85, Ala97, and Gly98, or is composed of the latter; and / or the antibody or its antigen-binding fragment binds to the human CD137 dimer and promotes human CD137 aggregation.

[0014] Example 2. An antibody or antigen-binding fragment thereof that specifically binds to human CD137, the antibody or antigen-binding fragment thereof comprising: (i) a heavy chain variable region (VH) comprising (a) HCDR1 (heavy chain complementarity determination region 1) of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 2, and (c) HCDR3 of SEQ ID NO: 3; or (ii) a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 10, and (c) HCDR3 of SEQ ID NO: 3.

[0015] Example 3. An antibody or antigen-binding fragment thereof according to any one of Examples 1-2, wherein the antibody or antigen-binding fragment comprises: (i) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 17; (ii) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 11; (iii) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 13; (iv) A heavy chain variable region (VH) comprising at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the same amino acid sequence as SEQ ID NO: 15; or (v) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 4.

[0016] Example 4. The antibody or antigen-binding fragment thereof according to Example 3, wherein one, two, three, four, five, six, seven, eight, nine or ten amino acids have been inserted, deleted or substituted in SEQ ID NO: 17, SEQ ID NO: 11, SEQ ID NO: 13, SEQ ID NO: 15 or SEQ ID NO: 4.

[0017] Example 5. An antibody or antigen-binding fragment thereof according to any one of the foregoing examples, wherein the antibody or antigen-binding fragment comprises: (i) Heavy chain variable region (VH) containing SEQ ID NO: 17; (ii) Heavy chain variable region (VH) containing SEQ ID NO: 11; (iii) Heavy chain variable region (VH) containing SEQ ID NO: 13; (iv) A heavy chain variable region (VH) containing SEQ ID NO: 15; or (v) Heavy chain variable region (VH) containing SEQ ID NO: 4.

[0018] Example 6. The antibody or antigen-binding fragment thereof according to any one of the foregoing examples, wherein the antibody or antigen-binding fragment thereof is a monoclonal antibody, a chimeric antibody, a humanized antibody, a human-engineered antibody, a single-chain antibody (scFv), a Fab fragment, a Fab' fragment, or an F(ab')2 fragment.

[0019] Example 7. An antibody or antigen-binding fragment thereof according to any one of the foregoing examples, wherein the antibody or antigen-binding fragment thereof comprises a heavy chain constant region of IgG1, IgG2, IgG3 or IgG4 subclass and / or a light chain constant region of type κ or λ.

[0020] Example 8. An antibody or antigen-binding fragment thereof according to any one of the foregoing examples, wherein the antibody or antigen-binding fragment thereof has antibody-dependent cytotoxicity (ADCC) or complement-dependent cytotoxicity (CDC).

[0021] Example 9. An antibody or antigen-binding fragment thereof according to any one of the foregoing examples, wherein the antibody or antigen-binding fragment thereof has reduced glycosylation or no glycosylation or low fucosylation.

[0022] Example 10. An antibody or antigen-binding fragment thereof according to any one of the foregoing examples, wherein the antibody or antigen-binding fragment thereof comprises an increased bipartite GlcNac structure.

[0023] Example 11. An antibody or antigen-binding fragment thereof according to any one of the foregoing examples, wherein the antibody or antigen-binding fragment thereof comprises an Fc domain, and wherein the Fc domain is an IgG1Fc with reduced effector function, optionally comprising the amino acid sequence of SEQ ID NO: 19 or SEQ ID NO: 53.

[0024] Example 12. An antibody or antigen-binding fragment thereof according to any one of the preceding examples, wherein the antibody or antigen-binding fragment thereof comprises an Fc domain, and wherein the Fc domain is an IgG1 Fc having reduced effector function and / or an extended half-life, optionally comprising the amino acid sequence of SEQ ID NO: 20.

[0025] Example 13. A pharmaceutical composition comprising an antibody or an antigen-binding fragment thereof according to any one of the foregoing examples, and a pharmaceutically acceptable carrier.

[0026] Example 14. A method for treating cancer, comprising administering to a patient in need a therapeutically effective amount of an antibody or antigen-binding fragment thereof according to any one of Examples 1-12, or a pharmaceutical composition according to Example 13.

[0027] Example 15. The method according to Example 14, wherein the 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.

[0028] Example 16. The method according to any one of Examples 14-15, wherein the antibody or its antigen-binding fragment is administered in combination with another therapeutic agent.

[0029] Example 17. The method according to Example 16, wherein the therapeutic agent is an anti-PD-1 antibody.

[0030] Example 18. The method according to Example 17, wherein the anti-PD1 antibody is tislelizumab.

[0031] Example 19: A multispecific antibody or antigen-binding fragment thereof, the multispecific antibody or antigen-binding fragment comprising at least a first antigen-binding domain specifically binding to human tumor-associated antigens (TAAs), and At least a second antigen-binding domain that specifically binds to human CD137, wherein the second antigen-binding domain is: (1) An antibody or an antigen-binding fragment thereof that specifically binds to the following epitope, which contains or is composed of the amino acid residues Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64, Gln67 of human CD137 (SEQ ID NO:35); (2) An antibody or its antigen-binding fragment that specifically binds to the following epitope, wherein the epitope comprises, or is composed of, the amino acid residues Ser55, Ala56, Arg75, Glu85, Ala97, and Gly98 of human CD137 (SEQ ID NO:35); or (3) An antibody or its antigen-binding fragment that specifically binds to a human CD137 dimer, wherein the human CD137 dimer comprises a first human CD137 monomer and a second human CD137 monomer, or is composed of the latter; wherein the antibody or its antigen-binding fragment specifically binds to an epitope of the first human CD137 monomer (SEQ ID NO:35), the epitope comprising amino acid residues Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64, Gln67, or is composed of the latter; and the antibody or its antigen-binding fragment specifically binds to an epitope of the second human CD137 monomer (SEQ ID NO:35), the epitope comprising amino acid residues Ser55, Ala56, Arg75, Glu85, Ala97, and Gly98, or is composed of the latter; and / or the antibody or its antigen-binding fragment binds to the human CD137 dimer and promotes human CD137 aggregation.

[0032] Example 20. A multispecific antibody or an antigen-binding fragment thereof, the multispecific antibody or antigen-binding fragment comprising at least a first antigen-binding domain specifically binding to human tumor-associated antigen (TAA), and at least a second antigen-binding domain specifically binding to human CD137, wherein the second antigen-binding domain comprises: (i) a heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 2, and (c) HCDR3 of SEQ ID NO: 3; or (ii) Heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 10, and (c) HCDR3 of SEQ ID NO: 3.

[0033] Example 21. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-20, wherein the second antigen-binding domain that specifically binds to human CD137 comprises: (i) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 17; (ii) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 11; (iii) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 13; (iv) A heavy chain variable region (VH) comprising at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the same amino acid sequence as SEQ ID NO: 15; or (v) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 4.

[0034] Example 22. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-21, wherein the second antigen-binding domain that specifically binds to human CD137 comprises: (i) Heavy chain variable region (VH) containing SEQ ID NO: 17; (ii) Heavy chain variable region (VH) containing SEQ ID NO: 11; (iii) Heavy chain variable region (VH) containing SEQ ID NO: 13; (iv) A heavy chain variable region (VH) containing SEQ ID NO: 15; or (v) Heavy chain variable region (VH) containing SEQ ID NO: 4.

[0035] Example 23. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-22, wherein the TAA is GPC3.

[0036] Example 24. A multispecific antibody or antigen-binding fragment thereof, the multispecific antibody or antigen-binding fragment thereof comprising a first antigen-binding domain specifically binding to human phosphatidylinositol proteoglycan 3 (GPC3) and a second antigen-binding domain specifically binding to human CD137.

[0037] Example 25: The multispecific antibody or its antigen-binding fragment according to Example 24, wherein the second antigen-binding domain that specifically binds to human CD137 is: (1) An antibody or an antigen-binding fragment thereof that specifically binds to the following epitope, which contains or is composed of the amino acid residues Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64, Gln67 of human CD137 (SEQ ID NO:35); (2) An antibody or its antigen-binding fragment that specifically binds to the following epitope, wherein the epitope comprises, or is composed of, the amino acid residues Ser55, Ala56, Arg75, Glu85, Ala97, and Gly98 of human CD137 (SEQ ID NO:35); or (3) An antibody or its antigen-binding fragment that specifically binds to a human CD137 dimer, wherein the human CD137 dimer comprises a first human CD137 monomer and a second human CD137 monomer, or is composed of the latter; wherein the antibody or its antigen-binding fragment specifically binds to an epitope of the first human CD137 monomer (SEQ ID NO:35), the epitope comprising amino acid residues Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64, Gln67, or is composed of the latter; and the antibody or its antigen-binding fragment specifically binds to an epitope of the second human CD137 monomer (SEQ ID NO:35), the epitope comprising amino acid residues Ser55, Ala56, Arg75, Glu85, Ala97, and Gly98, or is composed of the latter; and / or the antibody or its antigen-binding fragment binds to the human CD137 dimer and promotes human CD137 aggregation.

[0038] Example 26. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 24-25, wherein the second antigen-binding domain that specifically binds to human CD137 comprises: (i) a heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 2, and (c) HCDR3 of SEQ ID NO: 3; or (ii) Heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 10, and (c) HCDR3 of SEQ ID NO: 3.

[0039] Example 27. The multispecific antibody or its antigen-binding fragment according to Example 26, wherein the second antigen-binding domain that specifically binds to human CD137 comprises: (i) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 17; (ii) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 11; (iii) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 13; (iv) A heavy chain variable region (VH) comprising at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the same amino acid sequence as SEQ ID NO: 15; or (v) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 4.

[0040] Example 28. The multispecific antibody or its antigen-binding fragment according to Example 27, wherein one, two, three, four, five, six, seven, eight, nine or ten amino acids have been inserted, deleted or substituted in SEQ ID NO: 17, SEQ ID NO: 11, SEQ ID NO: 13, SEQ ID NO: 15 or SEQ ID NO: 4.

[0041] Example 29. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 24-28, wherein the second antigen-binding domain that specifically binds to human CD137 comprises: (i) Heavy chain variable region (VH) containing SEQ ID NO: 17; (ii) Heavy chain variable region (VH) containing SEQ ID NO: 11; (iii) Heavy chain variable region (VH) containing SEQ ID NO: 13; (iv) A heavy chain variable region (VH) containing SEQ ID NO: 15; or (v) Heavy chain variable region (VH) containing SEQ ID NO: 4.

[0042] Example 30. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-29, wherein the first antigen-binding domain that specifically binds to human GPC3 comprises: The heavy chain variable region (VH) comprises (a) HCDR1 of SEQ ID NO: 45, (b) HCDR2 of SEQ ID NO: 46, and (c) HCDR3 of SEQ ID NO: 47; and The light chain variable region (VL) comprises (d) LCDR1 of SEQ ID NO: 48, (e) LCDR2 of SEQ ID NO: 49, and (f) LCDR3 of SEQ ID NO: 50.

[0043] Example 31. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-30, wherein the first antigen-binding domain that specifically binds to human GPC3 comprises: The heavy chain variable region (VH) contains at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the same amino acid sequence as SEQ ID NO: 41, and the light chain variable region (VL) contains at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the same amino acid sequence as SEQ ID NO: 43.

[0044] Example 32. The multispecific antibody or its antigen-binding fragment according to Example 31, wherein one, two, three, four, five, six, seven, eight, nine or ten amino acids have been inserted, deleted or substituted in SEQ ID NO: 41 or SEQ ID NO: 43.

[0045] Example 33. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-32, wherein the first antigen-binding domain that specifically binds to human GPC3 comprises: a heavy chain variable region (VH) comprising SEQ ID NO: 41, and a light chain variable region (VL) comprising SEQ ID NO: 43.

[0046] Example 34. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-33, wherein the first antigen-binding domain that specifically binds to human GPC3 comprises The heavy chain variable region (VH) comprises (a) HCDR1 of SEQ ID NO: 45, (b) HCDR2 of SEQ ID NO: 46, and (c) HCDR3 of SEQ ID NO: 47; and The light chain variable region (VL) comprises (d) LCDR1 of SEQ ID NO: 48, (e) LCDR2 of SEQ ID NO: 49, and (f) LCDR3 of SEQ ID NO: 50, and The second antigen-binding domain that specifically binds to human CD137 includes: (i) a heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 2, and (c) HCDR3 of SEQ ID NO: 3; or (ii) Heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 10, and (c) HCDR3 of SEQ ID NO: 3.

[0047] Example 35. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-34, wherein the first antigen-binding domain that specifically binds to human GPC3 comprises: It includes the heavy chain variable region (VH) of SEQ ID NO: 41 and the light chain variable region (VL) of SEQ ID NO: 43; and The second antigen-binding domain that specifically binds to human CD137 includes: (i) Heavy chain variable region (VH) containing SEQ ID NO: 4; (ii) Heavy chain variable region (VH) containing SEQ ID NO: 11; (iii) Heavy chain variable region (VH) containing SEQ ID NO: 13; (iv) A heavy chain variable region (VH) containing SEQ ID NO: 15; or (v) Heavy chain variable region (VH) containing SEQ ID NO: 17.

[0048] Example 36. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-35, wherein the multispecific antibody or antigen-binding fragment thereof is a monoclonal antibody, a chimeric antibody, a humanized antibody, a human-engineered antibody, a single-chain antibody (scFv), a Fab fragment, a Fab' fragment, or an F(ab')2 fragment.

[0049] Example 37. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-36, wherein the first antigen-binding domain that specifically binds to human GPC3 is a monoclonal antibody, chimeric antibody, humanized antibody, human engineered antibody, single-chain antibody (scFv), single-domain antibody, Fab fragment, Fab' fragment, or F(ab')2 fragment, and The second antigen-binding domain that specifically binds to human CD137 is a monoclonal antibody, chimeric antibody, humanized antibody, human engineered antibody, single-chain antibody (scFv), single-domain antibody, Fab fragment, Fab' fragment, or F(ab')2 fragment.

[0050] Example 38. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-37, wherein the multispecific antibody or antigen-binding fragment thereof is a bispecific antibody.

[0051] Example 39. A multispecific antibody or antigen-binding fragment thereof according to Example 38, wherein the bispecific antibody is in a 2+2 form.

[0052] Example 40. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-39, wherein the multispecific antibody or antigen-binding fragment thereof contains a linker of SEQ ID NO: 60 to SEQ ID NO: 101.

[0053] Example 41. The multispecific antibody or its antigen-binding fragment according to Example 40, wherein the adapter is SEQ ID NO: 62.

[0054] Example 42. The multispecific antibody or its antigen-binding fragment according to Example 40, wherein the adapter is SEQ ID NO: 67.

[0055] Example 43. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-42, wherein the multispecific antibody or antigen-binding fragment comprises a heavy chain constant region of an IgG1, IgG2, IgG3, or IgG4 subclass and / or a light chain constant region of type κ or λ, and The heavy chain constant region contains CH1 and / or Fc domains.

[0056] Example 44. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-43, wherein the multispecific antibody or antigen-binding fragment thereof has antibody-dependent cytotoxicity (ADCC) or complement-dependent cytotoxicity (CDC).

[0057] Example 45. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-44, wherein the multispecific antibody or antigen-binding fragment thereof has reduced glycosylation or no glycosylation or low fucosylation.

[0058] Example 46. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-45, wherein the multispecific antibody or antigen-binding fragment thereof comprises an increased bipartite GlcNac structure.

[0059] Example 47. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-46, wherein the multispecific antibody or antigen-binding fragment thereof comprises an Fc domain, and wherein the Fc domain is an IgG1 Fc with reduced effector function, optionally comprising the amino acid sequence of SEQ ID NO: 53.

[0060] Example 48. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-47, wherein the multispecific antibody or antigen-binding fragment thereof comprises an Fc domain, and wherein the Fc domain is an IgG1 Fc having reduced effector function and / or an extended half-life, optionally comprising the amino acid sequence of SEQ ID NO: 20.

[0061] Example 49. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-48, wherein the multispecific antibody or antigen-binding fragment thereof comprises an Fc domain, and wherein the Fc domain is IgG4 Fc.

[0062] Example 50. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-49, wherein: a) The heavy chain variable region (VH) of the first antigen-binding domain that specifically binds to human GPC3, the CH1 domain, the Fc domain, and the heavy chain variable region (VH) of the second antigen-binding domain that specifically binds to human CD137 are arranged in the first polypeptide from the N-terminus to the C-terminus. Optionally, the C-terminus of the Fc domain is connected via a linker to the N-terminus of the heavy chain variable region (VH) of the second antigen-binding domain; and b) The light chain variable region (VL) and the first light chain constant region of the first antigen-binding domain that specifically binds to human GPC3 are arranged in the second polypeptide in the direction from the N-terminus to the C-terminus.

[0063] Example 51. A multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-50, wherein the multispecific antibody or antigen-binding fragment comprises (i) The first polypeptide of SEQ ID NO: 25 and the second polypeptide of SEQ ID NO: 23; (ii) The first polypeptide of SEQ ID NO: 21 and the second polypeptide of SEQ ID NO: 23; (iii) The first polypeptide of SEQ ID NO: 33 and the second polypeptide of SEQ ID NO: 23; (iv) The first polypeptide of SEQ ID NO: 27 and the second polypeptide of SEQ ID NO: 23; (v) The first polypeptide of SEQ ID NO: 29 and the second polypeptide of SEQ ID NO: 23; or (vi) The first polypeptide of SEQ ID NO: 31 and the second polypeptide of SEQ ID NO: 23.

[0064] Example 52. A pharmaceutical composition comprising a multispecific antibody or an antigen-binding fragment thereof according to any one of Examples 19-51, and a pharmaceutically acceptable carrier.

[0065] Example 53. A method of treating cancer, comprising administering to a patient in need a therapeutically effective amount of a multispecific antibody or antigen-binding fragment thereof according to any one of Examples 19-51, or a pharmaceutical composition according to Example 52.

[0066] Example 54. The method according to Example 53, wherein the cancer is an advanced or metastatic solid tumor.

[0067] Example 55. The method according to any one of Examples 53-54, wherein the cancer expresses GPC3.

[0068] Example 56. The method according to any one of Examples 53-55, wherein the cancer is liver cancer, lung cancer, gastric cancer, germ cell tumor, thyroid cancer, pancreatic cancer, ovarian cancer, skin cancer, kidney cancer, esophageal cancer, atypical teratoid rhabdomyosarcoma of the brain, or undifferentiated synovial sarcoma.

[0069] Example 57. The method according to Example 56, wherein the liver cancer is hepatoblastoma or hepatocellular carcinoma (HCC).

[0070] Example 58. The method according to Example 56, wherein the lung cancer is non-small cell lung cancer (NSCLC) or small cell lung cancer (SCLC).

[0071] Example 59. The method according to Example 58, wherein the non-small cell lung cancer is squamous non-small cell lung cancer.

[0072] Example 60. The method according to Example 58, wherein the non-small cell lung cancer is GPC3+ squamous non-small cell lung cancer.

[0073] Example 61. The method according to Example 56, wherein the gastric cancer is alpha-fetoprotein (AFP+) gastric cancer.

[0074] Example 62. The method according to Example 56, wherein the renal cell carcinoma is a nephroblastoma.

[0075] Example 63. The method according to Example 56, wherein the esophageal cancer is esophageal squamous cell carcinoma.

[0076] Example 64. The method according to Example 56, wherein the esophageal cancer is GPC3+ esophageal squamous cell carcinoma.

[0077] Example 65. The method according to Example 56, wherein the germ cell tumor is a yolk sac tumor or a non-dysgerminoma.

[0078] Example 66. The method according to any one of Examples 53-65, wherein the multispecific antibody or its antigen-binding fragment or pharmaceutical composition is administered in combination with another therapeutic agent.

[0079] Example 67. The method according to Example 66, wherein the therapeutic agent is an anti-PD1 or anti-PDL1 antibody.

[0080] Example 68. The method according to Example 67, wherein the anti-PD1 antibody is tislelizumab.

[0081] Example 69. An isolated nucleic acid that encodes an antibody, multispecific antibody or antigen-binding fragment thereof according to any one of Examples 1-12 and 19-51.

[0082] Example 70. A vector comprising the nucleic acid as described in Example 69.

[0083] Example 71. A host cell comprising the nucleic acid described in Example 69 or the vector described in Example 70.

[0084] Example 72. A process for producing a multispecific antibody or an antigen-binding fragment thereof, the process comprising culturing a host cell as described in Example 71 and recovering the antibody or an antigen-binding fragment thereof from the culture.

[0085] In some embodiments, this disclosure provides anti-CD137 antibodies or antigen-binding fragments thereof that exhibit specific binding and high affinity for human CD137 and cynomolgus monkey CD137, exhibit superior overall biophysical properties such as Tm or Tagg, exhibit superior pharmacokinetics, and / or are able to bind to CD137 dimers and promote CD137 aggregation.

[0086] In some embodiments, this disclosure provides an anti-CD137 antibody or an antigen-binding fragment thereof having at least one or more of the following features: (1) It showed specific binding and high affinity to human CD137 and cynomolgus monkey CD137; (2) It has superior pharmacokinetic properties; (3) Possesses superior overall biophysical properties, such as Tm or Tagg, and / or superior overall stability; (4) Humanized antibodies with low immunogenicity risk to humans; and (5) It can bind to CD137 dimers and promote CD137 aggregation.

[0087] In some embodiments, this disclosure provides anti-CD137 antibodies or antigen-binding fragments thereof that are humanized antibodies with low immunogenicity risk to humans while maintaining specific binding and high affinity for human CD137 and cynomolgus monkey CD137 and exhibiting superior overall biophysical properties and / or stability. In some embodiments, this disclosure provides anti-CD137 antibodies or antigen-binding fragments thereof that exhibit specific binding and high affinity for human CD137 and cynomolgus monkey CD137 and / or are able to bind to CD137 dimers and promote CD137 aggregation (e.g., via CDR residues).

[0088] In some embodiments, this disclosure provides a GPC3xCD137 multispecific antibody or an antigen-binding fragment thereof having at least one or more of the following features: (1) It showed specific binding and high affinity to human CD137 and cynomolgus monkey CD137; (2) It has specific binding and high affinity for human GPC3 and cynomolgus monkey GPC3, and shows high affinity for a wide range of GPC expression (low to high expression); (3) Inducing T cell activation in a GPC3-dependent manner, including cytokine release (e.g., IFN-γ or IL-2) and T cell killing activity, and reducing T cell activation or T cell killing activity in the absence of cells expressing GPC3. (4) Induces T cell activation and effective T cell killing activity in a wide range of GPC3-expressing cells (low, intermediate, and high expression); (5) When used alone, it effectively inhibits tumor growth; (7) When administered in combination with anti-PD-1 antibodies, it induces synergistic effects (e.g., tumor growth inhibition and / or tumor-free rate). (8) It has excellent pharmacokinetic properties; (9) Possesses excellent overall biophysical properties, such as Tm or Tagg, and / or stability; and (10) It can bind to CD137 dimers and promote CD137 aggregation. Attached Figure Description

[0089] Figure 1 ELISA demonstrated that BGA-9612 binds to huCD137 and partially competes with 20 µg / ml CD137L compared to urirolimumab (BMS-663513).

[0090] Figure 2 The comparison of FACS binding affinity between BGA-9612 and other humanized VHHs in human CD137-overexpressing HuT78 cells is shown.

[0091] Figure 3 This is a schematic diagram of the design of a multispecific antibody against tumor GPC3xCD137.

[0092] Figure 4A The binding assay of BE-774 to Hut78 / CD137 cells overexpressing CD137, performed by flow cytometry, demonstrates the binding of BE-774 to native huCD137 expressed on the cell surface. Figure 4B The binding assay of BE-774 to HepG2 cells expressing GPC3, performed by flow cytometry, demonstrates the binding of BE-774 to native huGPC3 expressed on the cell surface.

[0093] Figures 5A-5BThis study demonstrated that BE-774 and BE-653 induced T cell activation when co-cultured with GPC3-positive HepG2 tumor cells. Figure 5A is a schematic diagram illustrating the activation of CD137 (41BB) in huPBMCs by co-stimulating them with BE-774 or BE-653 and an OS8-expressing hepatocellular carcinoma (HCC) cell line. Figure 5B shows that BE-774 and BE-653 induced dose-dependent cytokine release in PBMCs co-cultured with HepG2 cells, but not in PBMCs co-cultured with GPC3-negative cells.

[0094] Figures 6A-6B This study demonstrated that BE-774 and BE-653 enhance T-cell killing activity against GPC3-positive HepG2 tumor cells. Figure 6A is a schematic diagram illustrating the activation of CD137 (4-1BB) by co-stimulating huPBMCs with BE-774 or BE-653 in combination with EpCAM / CD3 bispecific T-cell conjugate (BiTE), which provides the first signal for T-cell activation. Figure 6B shows that BE-774 and BE-653 dose-dependently enhance T-cell killing activity against GPC3-expressing cells, but do not enhance killing activity against GPC3-negative cells.

[0095] Figure 7 The PK characteristics of BE-933 and BE-774 in cynomolgus monkeys are shown.

[0096] Figure 8 The PK characteristics of BE-933 and BE-774 in the hFcRn mouse model are shown.

[0097] Figures 9A-9B The binding of BE-915 with human CD137 overexpressed on Hut78 (Fig. 9B) and human GPC3 overexpressed on HepG2 (Fig. 9A) is shown.

[0098] Figure 10 The binding specificity of BE-915 to CD137 and other TNFRSF members was demonstrated.

[0099] Figure 11A-11B The results show that BE-915 and CD137L cross-competitively bind to human CD137. CD137L blocks the binding of BE-915 to CD137 expressed on HuT78 (Figure 11A). BE-915 blocks the binding of CD137L to CD137 expressed on HuT78 (Figure 11B).

[0100] Figures 12A-12CThe figures show BE-915 inducing the release of IL-2 and IFN-γ from human PBMCs. Figure 12A is a schematic diagram of CD137 activation in huPBMCs via co-stimulation with BE-915 and an OS8-expressing hepatocellular carcinoma (HCC) cell line. Figures 12B-12C show BE-915 inducing dose-dependent cytokine release from PBMCs in a GPC3-dependent manner. PBMCs from two donors were tested.

[0101] Figures 13A-13C The diagram illustrates the T-cell killing activity induced by BE-915 in human PBMCs. Figure 13A is a schematic diagram of CD137 activation via co-stimulation of huPBMCs with BE-915 in combination with EpCAM / CD3 bispecific T-cell conjugate (BiTE), which provides the first signal for T-cell activation. Figures 13B-C show that BE-915 dose-dependently enhances T-cell killing activity against GPC3-expressing cells, but does not enhance killing activity against GPC3-negative cells. PBMCs from two donors were tested.

[0102] Figure 14 The pharmacokinetic characteristics of BE-915 in cynomolgus monkeys after intravenous infusion (5 mg / kg, N = 2) are shown.

[0103] Figure 15 This study demonstrated the efficacy of BE-915 monotherapy in the MC38 / hGPC3 model of humanized CD137 knock-in mice.

[0104] Figure 16 The efficacy of the combination of BE-915 and anti-PD-1 antibody in the humanized CD137 knock-in mouse LL / 2 / hGPC3 model was demonstrated.

[0105] Figure 17 A schematic diagram of the partially competitive binding of VHH(BGA-2524) to CD137L and CD137 is shown. The crystal structure of VHH(BGA-2524) / CD137 overlaps with the CD137L / CD137 complex (PDB:6MGP) through the CD137 CRD1 and CRD2 domains. CD137, CD137L, and VHH(BGA-2524) are colored black, white, and gray, respectively.

[0106] Figure 18The image shows VHH (BGA-2524) bound to a CD137 dimer. Crystal structure analysis reveals that VHH (BGA-2524) has the ability to bind to the CD137 dimer to promote CD137 aggregation. Each monomer of the CD137 dimer is shown in white or gray, while VHH (BGA-2524) on the surface is shown as a black cartoon (left). The epitopes of VHH (BGA-2524) are shown in black on the surface of the CD137 dimer (right, BGA-2524 removed).

[0107] Figure 19 The atomic interactions on the binding surface of the VHH(BGA-2524) / CD137 complex are shown. The binding interface between VHH(BGA-2524) and CD137 identifies certain key residues of VHH(BGA-2524) (para residues, underlined amino acids) and CD137 (epitope residues). Each monomer of the CD137 dimer is shown in white or gray cartoon form with a transparent surface, and the CRD1, CRD2, and CRD3 domains are indicated by lines. Para residues are shown with black lines, and amino acids are underlined (most of the framework region has been removed). Detailed Implementation

[0108] This disclosure provides anti-CD137 antibodies and their antigen-binding fragments, as well as multispecific antibodies or their antigen-binding fragments that recognize CD137 as one antigen and at least one tumor-associated antigen (TAA) as another antigen. This disclosure also provides anti-GPC3xCD137 multispecific antibodies and their antigen-binding fragments. Furthermore, this disclosure provides antibodies having desired pharmacokinetic properties, desired biophysical properties, and other desired attributes, and thus being usable for reducing the likelihood of cancer or treating cancer. This disclosure further provides pharmaceutical compositions comprising antibodies and methods for preparing and using such pharmaceutical compositions for the prevention and treatment of cancer and related disorders.

[0109] I. Anti-GPC3 antibody This disclosure provides antibodies or antigen-binding fragments thereof that specifically bind to human GPC3. In one embodiment, an anti-GPC3 antibody or antigen-binding fragment thereof is prepared at a concentration of 1 × 10⁻⁶. -6 M to 1×10 -10 Binding affinity of M (K) D ( ) specifically binds to human GPC3. In another embodiment, the anti-GPC3 antibody or its antigen-binding fragment binds at approximately 1 × 10 -6 M, approximately 1×10 -7 M, approximately 1×10 - 8 M, approximately 1×10 -9 M or approximately 1×10 -10 Binding affinity of M (K)D It combines with human GPC3.

[0110] In one embodiment, the anti-GPC3 antibody or its antigen-binding fragment comprises: a heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 45, (b) HCDR2 of SEQ ID NO: 46, and (c) HCDR3 of SEQ ID NO: 47; and a light chain variable region (VL) comprising (d) LCDR1 of SEQ ID NO: 48, (e) LCDR2 of SEQ ID NO: 49, and (f) LCDR3 of SEQ ID NO: 50, according to the Kabat number.

[0111] In another embodiment, the anti-GPC3 antibody or its antigen-binding fragment comprises: HCDR1, HCDR2 and HCDR3 from the heavy chain variable region (VH) shown in SEQ ID NO: 41; and LCDR1, LCDR2 and LCDR3 from the light chain variable region (VL) shown in SEQ ID NO: 43.

[0112] In another embodiment, the anti-GPC3 antibody or its antigen-binding fragment further comprises no more than one, two, three, four or five amino acid deletions, insertions or substitutions in the CDR, preferably conserved amino acid substitutions that maintain binding specificity and affinity.

[0113] In another embodiment, the anti-GPC3 antibody or its antigen-binding fragment comprises a heavy chain variable region (VH) containing at least 70%, 71%, 72%, 73%, 74%, 75%, 76%, 77%, 78%, 79%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the same amino acid sequence as SEQ ID NO: 41, and a light chain variable region (VL) containing the same amino acid sequence as SEQ ID NO: 41. SEQ ID NO: 43 contains at least 70%, 71%, 72%, 73%, 74%, 75%, 76%, 77%, 78%, 79%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical amino acid sequences. In another embodiment, SEQ ID NO: 41 or SEQ ID NO: 43 contains inserted, deleted, or substituted (optionally conserved amino acid substitutions) 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30 amino acids. In another embodiment, such variations are within the frame region of the variable region. In another embodiment, anti-GPC3 antibodies or their antigen-binding fragments with such variations maintain binding specificity and affinity.

[0114] In another embodiment, the anti-GPC3 antibody or its antigen-binding fragment comprises a heavy chain variable region (VH) containing SEQ ID NO: 41 and a light chain variable region (VL) containing SEQ ID NO: 43.

[0115] In another embodiment, the anti-human GPC3 antibody or its antigen-binding fragment exhibits cross-species binding activity against cynomolgus monkey GPC3.

[0116] II. Anti-CD137 antibody Table 1: Sequence of anti-CD137 antibody

[0117] This disclosure provides antibodies or antigen-binding fragments thereof that specifically bind to human CD137. The antibodies or antigen-binding fragments disclosed herein include, but are not limited to, antibodies or antigen-binding fragments thereof generated as described below.

[0118] This disclosure provides an antibody or antigen-binding fragment thereof that specifically binds to human CD137, wherein the antibody or antigen-binding fragment thereof specifically binds to an epitope comprising or consisting of amino acid residues Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64, and Gln67 of human CD137 (SEQ ID NO: 35), and optionally the epitope is determined by X-ray diffraction.

[0119] This disclosure provides an antibody or antigen-binding fragment thereof that specifically binds to human CD137, wherein the antibody or antigen-binding fragment thereof specifically binds to an epitope comprising, or consisting of, amino acid residues Ser55, Ala56, Arg75, Glu85, Ala97 and Gly98 of human CD137 (SEQ ID NO: 35), optionally determined by X-ray diffraction.

[0120] This disclosure provides an antibody or antigen-binding fragment thereof that specifically binds to human CD137, wherein the antibody or antigen-binding fragment thereof specifically binds to a human CD137 dimer comprising, or consisting of, a first human CD137 monomer and a second human CD137 monomer; wherein the antibody or antigen-binding fragment thereof specifically binds to an epitope of the first human CD137 monomer (SEQ ID NO: 35), the epitope comprising, substantially consisting of, or consisting of, amino acid residues Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64, and Gln67; and the antibody or antigen-binding fragment thereof binds to, a second human CD137 monomer (SEQ ID NO: 35); The antibody (NO:35) specifically binds to an epitope comprising, substantially composed of, or consisting of, amino acid residues Ser55, Ala56, Arg75, Glu85, Ala97, and Gly98, optionally determined by X-ray diffraction, and optionally the antibody or its antigen-binding fragment binds to human CD137 dimers and promotes human CD137 aggregation.

[0121] This disclosure provides an antibody or antigen-binding fragment thereof that specifically binds to human CD137, wherein the antibody or antigen-binding fragment thereof specifically binds to an epitope comprising one or more amino acid residues selected from the group consisting of Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64 and Gln67 of human CD137 (SEQ ID NO: 35).

[0122] This disclosure provides an antibody or antigen-binding fragment thereof that specifically binds to human CD137, wherein the antibody or antigen-binding fragment thereof specifically binds to an epitope comprising one or more amino acid residues selected from the group consisting of Ser55, Ala56, Arg75, Glu85, Ala97 and Gly98 of human CD137 (SEQ ID NO: 35).

[0123] This disclosure provides an antibody or antigen-binding fragment thereof that specifically binds to human CD137, wherein the antibody or antigen-binding fragment thereof specifically binds to a human CD137 dimer comprising, or composed of, a first human CD137 monomer and a second human CD137 monomer, wherein the antibody or antigen-binding fragment thereof specifically binds to an epitope of the first human CD137 monomer (SEQ ID NO: 35), the epitope comprising one or more amino acid residues selected from the group consisting of Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64, and Gln67, and the antibody or antigen-binding fragment thereof binds to the second human CD137 monomer (SEQ ID NO: 35). The antibody specifically binds to the epitope NO:35, which contains one or more amino acid residues selected from the group consisting of Ser55, Ala56, Arg75, Glu85, Ala97 and Gly98. Optionally, the antibody or its antigen-binding fragment binds to the human CD137 dimer and promotes human CD137 aggregation.

[0124] This disclosure provides an antibody or antigen-binding fragment thereof that specifically binds to human CD137, wherein the antibody or antigen-binding fragment thereof specifically binds to an epitope comprising one or more amino acid residues selected from the group consisting of Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64, Gln67, Ser55, Ala56, Arg75, Glu85, Ala97, and Gly98 of human CD137 (SEQ ID NO: 35).

[0125] This disclosure provides an antibody or antigen-binding fragment thereof that specifically binds to human CD137, wherein the antibody or antigen-binding fragment thereof specifically binds to the following epitope, which is composed of the amino acid residues Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64, and Gln67 of human CD137 (SEQ ID NO: 35).

[0126] This disclosure provides an antibody or antigen-binding fragment thereof that specifically binds to human CD137, wherein the antibody or antigen-binding fragment thereof specifically binds to an epitope consisting of the amino acid residues Ser55, Ala56, Arg75, Glu85, Ala97 and Gly98 of human CD137 (SEQ ID NO: 35).

[0127] This disclosure provides an antibody or antigen-binding fragment thereof that specifically binds to human CD137, wherein the antibody or antigen-binding fragment thereof specifically binds to a human CD137 dimer comprising, or composed of, a first human CD137 monomer and a second human CD137 monomer, wherein the antibody or antigen-binding fragment thereof specifically binds to an epitope of the first human CD137 (SEQ ID NO:35) monomer, the epitope being composed of amino acid residues Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64, and Gln67, and the antibody or antigen-binding fragment thereof specifically binds to an epitope of the second human CD137 monomer (SEQ ID NO:35), the epitope being composed of amino acid residues Ser55, Ala56, Arg75, Glu85, Ala97, and Gly98, wherein the antibody or antigen-binding fragment thereof binds to the human CD137 dimer and promotes human CD137 aggregation.

[0128] This disclosure provides antibodies or antigen-binding fragments thereof that specifically bind to human CD137, wherein the antibody or antigen-binding fragment comprises para positions of one or more amino acid residues selected from the group consisting of: Asn31, Tyr32, Ala33, Trp52, Ser53, Tyr55, His57, Leu98, Lys99, Tyr100, Pro101, Thr104, Thr106, Tyr109 (natural sequence order) or Asn31, Tyr32, Ala33, Trp52, Ser54, Tyr56, His58, Leu96, Lys97, Tyr98, Pro99, Thr100B, Thr100D, Tyr102 (Kabat nomenclature).

[0129] In one embodiment, the antibody or its antigen-binding fragment that specifically binds to human CD137 binds primarily to the side surface of the (e.g., human) CD137 CRD2 domain via CDR residues (e.g., Asn31, Tyr32, Ala33, Trp52, Ser54, Tyr56, His58, Leu96, Lys97, Tyr98, Pro99, Thr100B, Thr100D, Tyr102 of human CD137 VHH (Kabat nomenclature)).

[0130] In some embodiments, the epitopes of human CD137 bound by an antibody or antigen-binding fragment thereof that specifically binds to human CD137 disclosed herein are determined by X-ray diffraction.

[0131] This disclosure provides antibody or antigen-binding fragments that specifically bind to human CD137, wherein said antibody or antibody fragment (e.g., antigen-binding fragment) comprises a VH domain having an amino acid sequence of SEQ ID NO: 4, SEQ ID NO: 8, SEQ ID NO: 6, SEQ ID NO: 11, SEQ ID NO: 13, SEQ ID NO: 15, or SEQ ID NO: 17 (Table 1). This disclosure also provides antibody or antigen-binding fragments that specifically bind to human CD137, wherein said antibody or antigen-binding fragment comprises an HCDR having an amino acid sequence of any one of the HCDRs listed in Table 1. In one aspect, this disclosure provides antibody or antigen-binding fragments that specifically bind to human CD137, wherein said antibody comprises (or alternatively constitutes) one, two, three, or more HCDRs having an amino acid sequence of any one of the HCDRs listed in Table 1.

[0132] In one embodiment, the antibody or its antigen-binding fragment comprises one or more complementarity-determining regions (CDRs) comprising an amino acid sequence selected from SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 10; or selected from SEQ ID NO: 54, SEQ ID NO: 55, SEQ ID NO: 56; or selected from SEQ ID NO: 57, SEQ ID NO: 58, SEQ ID NO: 59.

[0133] In one embodiment, the anti-CD137 antibody or its antigen-binding fragment comprises: (i) HCDR1 (heavy chain complementarity-determining region 1), HCDR2, and HCDR3 from the heavy chain variable region (VH) shown in SEQ ID NO: 4; (ii) HCDR1, HCDR2, and HCDR3 from the heavy chain variable region (VH) shown in SEQ ID NO: 8; (iii) HCDR1, HCDR2, and HCDR3 from the heavy chain variable region (VH) shown in SEQ ID NO: 6; (iv) HCDR1, HCDR2, and HCDR3 from the heavy chain variable region (VH) shown in SEQ ID NO: 11; (v) HCDR1, HCDR2, and HCDR3 from the heavy chain variable region (VH) shown in SEQ ID NO: 13; (vi) HCDR1, HCDR2, and HCDR3 from the heavy chain variable region (VH) shown in SEQ ID NO: 15; or (vii) from SEQ ID NO: HCDR1, HCDR2 and HCDR3 of the heavy chain variable region (VH) shown in Figure 17.

[0134] In one embodiment, the anti-CD137 antibody or its antigen-binding fragment comprises: (i) a heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 2, and (c) HCDR3 of SEQ ID NO: 3; or (ii) a heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 10, and (c) HCDR3 of SEQ ID NO: 3, according to the Kabat number.

[0135] Other antibodies or antigen-binding fragments thereof disclosed herein include amino acids that have been altered but have at least 60%, 70%, 80%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% amino acid identity in the CDR region as disclosed in Table 1. In some aspects, this includes amino acid alterations (insertions, deletions, or substitutions, optionally conserved amino acid substitutions), wherein no more than 1, 2, 3, 4, or 5 amino acids are altered in the CDR region when compared to the CDR region depicted in the sequences described in Table 1, while maintaining binding specificity and affinity.

[0136] Other antibodies disclosed herein include those in which the amino acids or nucleic acids encoding the amino acids have been altered, but which have at least 60%, 70%, 71%, 72%, 73%, 74%, 75%, 76%, 77%, 78%, 79%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identity percentage with the sequences described in Table 1, optionally with the corresponding sequence of the CDR remaining unchanged. In some respects, it includes changes in the amino acid sequence, wherein no more than 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29 or 30 amino acids are changed when compared with the variable region depicted in the sequence described in Table 1, while the therapeutic activity / binding specificity / affinity is preserved, and optionally the corresponding sequence of the CDR is not changed. In some respects, it includes changes to the amino acid sequence, wherein, when compared to the variable region depicted in the sequence described in Table 1, one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteenth ...

[0137] In some embodiments, this disclosure provides antibodies or antigen-binding fragments thereof that specifically bind to human CD137, the antibodies or antigen-binding fragments thereof comprising a heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 2, and (c) HCDR3 of SEQ ID NO: 3; wherein amino acids F37, Y47, G49, and I94 (Kabat number) in the frame region are retained.

[0138] In some embodiments, this disclosure provides antibodies or antigen-binding fragments thereof that specifically bind to human CD137, the antibodies or antigen-binding fragments thereof comprising a heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 10, and (c) HCDR3 of SEQ ID NO: 3; wherein amino acids F37, Y47, G49, and I94 (Kabat number) in the frame region are retained.

[0139] In some embodiments, this disclosure provides antibodies or antigen-binding fragments thereof that specifically bind to human CD137, the antibodies or antigen-binding fragments thereof comprising a VH domain having the amino acid sequences or variants thereof described in Table 1, wherein HCDR1, HCDR2, and HCDR3 are unchanged, and amino acids F37, Y47, G49, and I94 (Kabat numbers) in the frame region are retained.

[0140] In some embodiments, this disclosure provides antibodies or antigen-binding fragments thereof that specifically bind to human CD137, the antibodies or antigen-binding fragments thereof comprising a heavy chain variable region containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the same amino acid sequence as SEQ ID NO: 4, SEQ ID NO: 8, SEQ ID NO: 6, SEQ ID NO: 11, SEQ ID NO: 13, SEQ ID NO: 15, or SEQ ID NO: 17, wherein HCDR1, HCDR2, and HCDR3 are unchanged, and amino acids F37, Y47, G49, and I94 (Kabat number) in the frame region are retained.

[0141] In another embodiment, this disclosure provides antibodies or antigen-binding fragments thereof, which are arranged in a 1×10-1 ratio. -6 M to 1×10 -10 Binding affinity of M (K) D ( ) specifically binds to human CD137. In another embodiment, the anti-CD137 antibody or its antigen-binding fragment binds at approximately 1 × 10 -6 M, approximately 1×10 -7 M, approximately 1×10 -8 M, approximately 1×10 -9 M or approximately 1×10 -10 Binding affinity of M (K) D It binds to human CD137.

[0142] This disclosure also provides the VH and full-length heavy chain nucleic acid sequences encoding an antibody that specifically binds to human CD137. Such nucleic acid sequences can be optimized for expression in mammalian cells.

[0143] This disclosure also provides antibodies and their antigen-binding fragments that bind to the same epitopes as the anti-CD137 antibodies described in Table 1. Therefore, other antibodies and their antigen-binding fragments can be identified based on their ability to cross-compete with other antibodies in binding assays (e.g., competitively inhibiting their binding in a statistically significant manner). The ability of the test antibody to inhibit the binding of the antibodies and their antigen-binding fragments disclosed herein to CD137 demonstrates that the test antibody can compete with that antibody or its antigen-binding fragment for binding to CD137. Without being bound by any single theory, such antibodies can bind to epitopes on CD137 that are the same as or related to (e.g., structurally similar or spatially proximate) the competing antibodies or their antigen-binding fragments. In some respects, antibodies that bind to the same epitopes on CD137 as the antibodies or their antigen-binding fragments disclosed herein are human or humanized monoclonal antibodies. Such human or humanized monoclonal antibodies can be prepared and isolated as described herein.

[0144] In some embodiments, the anti-CD137 antibody includes at least one antigen-binding site and at least one variable region. In some embodiments, the anti-CD137 antibody includes an antigen-binding fragment from the CD137 antibody described herein. In some embodiments, the anti-CD137 antibody is isolated or recombinant. In some embodiments, the anti-CD137 antibody also encompasses multispecific antibodies that target CD137 as at least one arm and target one or more other antigens as additional one or more arms.

[0145] III. Anti-CD137 multispecific antibody In one embodiment, the anti-CD137 antibody disclosed herein can be used to construct multispecific antibodies together with other forms such as human tumor-associated antigens (TAAs), immune checkpoints, or immune stimulating factors.

[0146] In one embodiment, the anti-CD137 antibody disclosed herein may be incorporated into an anti-CD137 x TAA multispecific antibody, wherein the TAA is an antibody or fragment thereof targeting any human tumor-associated antigen. The antibody molecule is a multispecific antibody molecule, for example, it comprises multiple antigen-binding domains, wherein at least one antigen-binding domain sequence specifically binds to a human TAA as a first antigen / epitope, and a second antigen-binding domain sequence specifically binds to a human CD137 as a second antigen / epitope. In one embodiment, the multispecific antibody comprises a third, fourth, or fifth antigen-binding domain. In one embodiment, the multispecific antibody is a bispecific antibody, a trispecific antibody, or a tetraspecific antibody. In each example, the multispecific antibody comprises at least one anti-TAA antigen-binding domain and at least one anti-CD137 antigen-binding domain.

[0147] In one embodiment, a multispecific antibody is a bispecific antibody. As used herein, a bispecific antibody binds specifically to only two antigens. A bispecific antibody comprises a first antigen-binding domain that specifically binds to a TAA and a second antigen-binding domain that specifically binds to human CD137. This includes bispecific antibodies comprising a heavy chain variable domain and a light chain variable domain that specifically bind to a TAA, and a heavy chain variable domain that specifically binds to human CD137. In some embodiments, a bispecific antibody comprises an antigen-binding fragment, wherein the antigen-binding fragment may be Fab, F(ab')2, Fv, a single-chain Fv (scFv), or a single-domain antibody.

[0148] In some embodiments, the second antigen-binding domain that specifically binds to human CD137 includes the anti-CD137 antibody disclosed in Section II.

[0149] In one embodiment, the multispecific antibody disclosed herein is in the form of 1×10 -6 M to 1×10 -10 Binding affinity of M (K) D It binds to human TAA and / or human CD137. In another embodiment, the multispecific antibody disclosed herein binds at approximately 1 × 10⁻⁶. -6 M, approximately 1×10 -7 M, approximately 1×10 -8 M, approximately 1×10 -9 M or approximately 1×10 -10 Binding affinity of M (K) D It binds to human TAA and / or human CD137.

[0150] In one embodiment, this disclosure provides a multispecific antibody or an antigen-binding fragment thereof, wherein a first antigen-binding domain specifically binds to human TAA and a second antigen-binding domain specifically binds to human CD137 comprises: (i) a heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 2, and (c) HCDR3 of SEQ ID NO: 3; or (ii) a heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 10, and (c) HCDR3 of SEQ ID NO: 3.

[0151] In another embodiment, this disclosure provides a multispecific antibody or an antigen-binding fragment thereof, wherein a first antigen-binding domain specifically binds to human TAA, and a second antigen-binding domain specifically binds to human CD137 comprises: (i) a heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the amino acid sequence identical to that of SEQ ID NO: 17; (ii) a heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the amino acid sequence identical to that of SEQ ID NO: 11; (iii) a heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the amino acid sequence identical to that of SEQ ID NO: 13; (iv) The heavy chain variable region (VH) contains at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the same amino acid sequence as SEQ ID NO: 15; or (v) the heavy chain variable region (VH) contains at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the same amino acid sequence as SEQ ID NO: 4.

[0152] In another embodiment, this disclosure provides a multispecific antibody or an antigen-binding fragment thereof, wherein a first antigen-binding domain specifically binds to human TAA and a second antigen-binding domain specifically binds to human CD137 comprises: (i) a heavy chain variable region (VH) comprising SEQ ID NO: 4; (ii) a heavy chain variable region (VH) comprising SEQ ID NO: 11; (iii) a heavy chain variable region (VH) comprising SEQ ID NO: 13; (iv) a heavy chain variable region (VH) comprising SEQ ID NO: 15; or (v) a heavy chain variable region (VH) comprising SEQ ID NO: 17.

[0153] In one embodiment, the TAA is human GPC3. The first antigen-binding domain that specifically binds to human GPC3 includes the anti-GPC3 antibody disclosed in Section I.

[0154] This disclosure provides multivalent antibodies (e.g., tetravalent antibodies) having at least two antigen-binding domains, which can be readily generated through recombinant expression of nucleic acids encoding antibody polypeptide chains. The multivalent antibodies described herein contain three to eight, but preferably four, antigen-binding domains that specifically bind to at least two antigens.

[0155] IV. Anti-GPC3xCD137 multispecific antibody In one embodiment, anti-GPC3 and anti-CD137 antibodies, as disclosed herein, can be incorporated into an anti-GPC3xCD137 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 GPC3 as a first epitope / antigen, and a second antigen-binding domain sequence specifically binds to CD137 as a second epitope / antigen. In one embodiment, the multispecific antibody comprises a third, fourth, or fifth antigen-binding domain. In one embodiment, the multispecific antibody is a bispecific antibody, a trispecific antibody, or a tetraspecific antibody. In each example, the multispecific antibody comprises at least one anti-GPC3 antigen-binding domain and at least one anti-CD137 antigen-binding domain.

[0156] 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 human GPC3 and a second antigen-binding domain that specifically binds to human CD137. This includes bispecific antibodies comprising a heavy chain variable domain and a light chain variable domain that specifically bind to human GPC3 as a first epitope / antigen, and a heavy chain variable domain that specifically binds to human CD137 as a second epitope / antigen. In some embodiments, a bispecific antibody comprises an antigen-binding fragment, wherein the antigen-binding fragment may be Fab, F(ab')2, Fv, single-chain Fv (scFv), or a single-domain antibody.

[0157] This disclosure provides a multispecific antibody or antigen-binding fragment thereof comprising a first antigen-binding domain that specifically binds to human phosphatidylinositol proteoglycan 3 (GPC3) and a second antigen-binding domain that specifically binds to human CD137.

[0158] The first antigen-binding domain that specifically binds to human phosphatidylinositol proteoglycan 3 (GPC3) includes the anti-GPC3 antibody described in Section I. The second antigen-binding domain that specifically binds to human CD137 includes the anti-CD137 antibody disclosed in Section II.

[0159] In one embodiment, the multispecific antibody disclosed herein is in the form of 1×10 -6M to 1×10 -10 Binding affinity of M (K) D It binds to human GPC3 and / or human CD137. In another embodiment, the multispecific antibody disclosed herein binds at approximately 1 × 10⁻⁶. - 6 M, approximately 1×10 -7 M, approximately 1×10 -8 M, approximately 1×10 -9 M or approximately 1×10 -10 Binding affinity of M (K) D It binds to human GPC3 and / or human CD137.

[0160] In one embodiment, the disclosed multispecific antibody specifically binds to human GPC3 and exhibits high affinity for both human and monkey GPC3. In another embodiment, the disclosed multispecific antibody specifically binds to human CD137. In yet another embodiment, the disclosed multispecific antibody exhibits high affinity for both human and monkey CD137.

[0161] In one embodiment, this disclosure provides a multispecific antibody or an antigen-binding fragment thereof, wherein a first antigen-binding domain specifically binding to human GPC3 comprises: a heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 45, (b) HCDR2 of SEQ ID NO: 46, and (c) HCDR3 of SEQ ID NO: 47; and a light chain variable region (VL) comprising (d) LCDR1 of SEQ ID NO: 48, (e) LCDR2 of SEQ ID NO: 49, and (f) LCDR3 of SEQ ID NO: 50, according to the Kabat number; and wherein a second antigen-binding domain specifically binding to human CD137 comprises: (i) a heavy chain variable region (VH) comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 2, and (c) HCDR3 of SEQ ID NO: 3; or (ii) The heavy chain variable region (VH) comprises (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 10, and (c) HCDR3 of SEQ ID NO: 3, according to the Kabat number.

[0162] In another embodiment, this disclosure provides a multispecific antibody or an antigen-binding fragment thereof, wherein a first antigen-binding domain specifically binding to human GPC3 comprises: a heavy chain variable region (VH) comprising SEQ ID NO: 41, and a light chain variable region (VL) comprising SEQ ID NO: 43; and wherein a second antigen-binding domain specifically binding to human CD137 comprises: (i) a heavy chain variable region (VH) comprising SEQ ID NO: 4; (ii) a heavy chain variable region (VH) comprising SEQ ID NO: 11; (iii) a heavy chain variable region (VH) comprising SEQ ID NO: 13; (iv) a heavy chain variable region (VH) comprising SEQ ID NO: 15; or (v) a heavy chain variable region (VH) comprising SEQ ID NO: 17.

[0163] In another embodiment, this disclosure provides a multispecific antibody or antigen-binding fragment thereof, wherein the multispecific antibody or antigen-binding fragment is (i) BE-933, comprising a first polypeptide of SEQ ID NO: 27 and a second polypeptide of SEQ ID NO: 23; (ii) BE-774, comprising a first polypeptide of SEQ ID NO: 25 and a second polypeptide of SEQ ID NO: 23; (iii) BE-653, comprising a first polypeptide of SEQ ID NO: 29 and a second polypeptide of SEQ ID NO: 23; (iv) BE-915, comprising a first polypeptide of SEQ ID NO: 21 and a second polypeptide of SEQ ID NO: 23; (v) BE-647, comprising a first polypeptide of SEQ ID NO: 31 and a second polypeptide of SEQ ID NO: 23; or (vi) BE-621, comprising a first polypeptide of SEQ ID NO: 33 and a second polypeptide of SEQ ID NO: 23.

[0164] Other multispecific antibodies or antigen-binding fragments thereof disclosed herein include those in which amino acids or nucleic acids encoding amino acids have been altered, but have at least 60%, 70%, 71%, 72%, 73%, 74%, 75%, 76%, 77%, 78%, 79%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identity percentage with the sequence described herein (e.g., CDR is not altered). In some respects, it includes changes in the amino acid sequence, wherein no more than 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29 or 30 amino acids are altered when compared to the variable regions described herein, while the therapeutic activity / binding specificity / affinity is preserved.

[0165] V. Other Form and module ratio In one embodiment, the anti-CD137 antibody disclosed herein can be used to construct multispecific antibodies in conjunction with other forms, such as TAAs, immune checkpoints, or immune stimulating factors. A TAA (e.g., GPC3) is used as an example in the forms and ratios described below. The description of GPC3 in the following examples can also be applied to other TAAs.

[0166] The multispecific antibodies disclosed herein may be in different forms. In one embodiment, the multispecific antibody disclosed herein has the following forms, including: (1) Form A provides a symmetrical IgG-like multispecific molecule having a Fab × VH configuration. An anti-huCD137 VH domain antibody is fused to the C-terminus of the Fc (CH3 domain) of an anti-GPC3 antibody, with a linker between them, such as Figure 3 As shown in the diagram; (2) Form B also provides a symmetrical IgG-like multispecific molecule with a Fab × VH configuration. The anti-huCD137 VH domain antibody is fused to the C-terminus of the light chain (Cκ) of the anti-GPC3 antibody, with a linker between them; (3) Form C provides a symmetrical VH antibody-like multispecific molecule with a Fab × VH configuration. The Fab region of the anti-GPC3 antibody is fused to the N-terminus of the VH of the anti-huCD137 VH domain Ab, with a linker between them; and (4) Form D also provides a symmetrical IgG-like multispecific molecule with a Fab × VH configuration. The anti-huCD137 VH domain antibody is fused to the N-terminus of the heavy chain (VH) of the anti-GPC3 antibody, with a linker between them. In one embodiment, the multispecific antibody is form A, as shown in the diagram. Figure 3As shown in the image.

[0167] The multispecific antibodies disclosed herein can be constructed using different module ratios, such as 1:1. In one embodiment, an inert Fc can be used for the multispecific antibody, and the Azymetric™ platform from Zymeworks can be used to assemble the Fab×VH configuration, in which a ZW1 mutation (A chain: T350V / L351Y / F405A / Y407V; B chain: T350V / T366L / K392L / T394W) can be introduced into the CH3 domain of the heavy chain to allow efficient heterodimer formation (Von Kreudenstein et al., (2013) Mabs [Monoclonal Antibodies] 5(5):646-54, incorporated herein by reference in its entirety). In one aspect, a specific ratio activates CD137 in a GPC3-dependent manner, while CD137 is not activated in the absence of GPC3.

[0168] In one embodiment, a multispecific antibody or its antigen-binding fragment comprises: a) a first polypeptide comprising, from its N-terminus to its C-terminus: a first heavy chain variable region (e.g., a single first heavy chain variable region); a CH1 domain, an Fc domain, and a second heavy chain variable region (e.g., a single second heavy chain variable region); optionally, the C-terminus of the Fc domain is connected to the N-terminus of the second heavy chain variable region via a linker; and b) a second polypeptide comprising, from its N-terminus to its C-terminus: a first light chain variable region (e.g., a single first light chain variable region); and a first light chain constant region; wherein the first heavy chain variable region and the first light chain variable region form a first antigen-binding domain that specifically binds to human GPC3, and the second heavy chain variable region forms a second antigen-binding domain that specifically binds to human CD137. In another embodiment, a multispecific antibody or its antigen-binding fragment comprises two first polypeptides and two second polypeptides.

[0169] connector It should also be understood that the domains and / or regions of the polypeptide chain of a bispecific antibody can be separated by linkers of various lengths. In some embodiments, the antigen-binding domains are separated from each other by linker regions CL, CH1, hinge, CH2, CH3, or the entire Fc. For example, VL1-CL-(linker)VH2-CH1. Such linker regions can contain randomly sorted amino acids or a restricted set of amino acids. Such linker regions can be flexible or rigid (see, for example, US 2009 / 0155275, incorporated herein by reference in its entirety).

[0170] Multispecific antibodies have been constructed using either flexible adapter gene fusion of two single-stranded Fv (scFv) or Fab fragments (Mallender et al., J. Biol. Chem. 1994 269:199-206; Mack et al., Proc. Natl. Acad. Sci. USA. 1995; 92:7021-5; Zapata et al., Protein Eng. 1995; 8:1057-62), via dimerization devices such as 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., FEBSLett). [European Federation of Biochemical Societies Letters] 1998; 422:259-64); via biantibodies (Holliger et al., Proc. Nat. Acad. Sci. USA. [Proceedings of the National Academy of Sciences of the United States of America] 1993; 90:6444-8; Zhu et al., Bio / Technology (NY) 1996; 14:192-6); Fab-scFv fusion (Schoonjans et al., J. Immunol. [Journal of Immunology] 2000; 165:7050-7); and microantibody forms (Pack et al., Biochemistry 1992; 31:1579-84; Pack et al., Bio / Technology 1993; 11:1271-7). Each reference mentioned in this paragraph is incorporated in its entirety by way of citation.

[0171] The multispecific antibodies disclosed herein include a linker 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 between one or more of their antigen-binding domains, CL domains, CH1 domains, hinge regions, CH2 domains, CH3 domains or Fc regions. In some embodiments, the amino acids glycine and serine are included within the linker region. In another embodiment, the connector may be GS (SEQ ID NO: 97), GGS (SEQ ID NO: 98), GSG (SEQ ID NO: 99), SGG (SEQ ID NO: 100), GGG (SEQ ID NO: 101), GGGS (SEQ ID NO: 60), SGGG (SEQ ID NO: 61), GGGGS (SEQ ID NO: 62), GGGGSGS (SEQ ID NO: 63), GGGGSGS (SEQ ID NO: 64), GGGGSGGS (SEQ ID NO: 65), GGGGSGGGGS (SEQ ID NO: 66), GGGGSGGGGSGGGGS (SEQ ID NO: 67), AKTTPKLEEGEFSEAR (SEQ ID NO: 68), AKTTPKLEEGEFSEARV (SEQ ID NO: 69), AKTTPKLGG (SEQ ID NO: 70), SAKTTPKLGG (SEQ ID NO: 70), or SAKTTPKLGG (SEQ ID NO: 70). 71), AKTTPKLEEGEFSEARV (SEQ ID NO: 72), SAKTTP (SEQ ID NO: 73), SAKTTPKLGG (SEQ ID NO: 74), RADAAP (SEQ ID NO: 75), RADAAPTVS (SEQ ID NO: 76), RADAAAAGGPGS (SEQ ID NO: 77), RADAAAA(G4S)4 (SEQ ID NO: 78), SAKTTP (SEQ ID NO: 79), SAKTTPKLGG (SEQ ID NO: 80), SAKTTPKLEEGEFSEARV (SEQ ID NO: 81), ADAAP (SEQ ID NO: 82), ADAAPTVSIFPP (SEQ ID NO: 83), TVAAP (SEQ ID NO: 84), TVAAPSVFIFPP (SEQ ID NO: 85), QPKAAP (SEQ ID NO: 86), QPKAAPSVTLFPP (SEQSEQ ID NO: 87), AKTTPP (SEQ ID NO: 88), AKTTPPSVTPLAP (SEQ ID NO: 89), AKTTAP (SEQ ID NO: 90), AKTTAPSVYPLAP (SEQ ID NO: 91), ASTKGP (SEQ ID NO: 92), ASTKGPSVFPLAP (SEQ ID NO: 93), GENKVEYAPALMALS (SEQ ID NO: 94), GPAKELTPLKEAKVS (SEQ ID NO: 95), and GHEAAAVMQVQYPAS (SEQ ID NO: 96) or any combination thereof (see WO 2007 / 024715, which is incorporated herein by reference in its entirety).

[0172] Dimerization-specific amino acids In one embodiment, the multivalent antibody comprises at least one dimerization-specific amino acid alteration. The dimerization-specific amino acid alteration results in a “protrusion into the pore” interaction and increases the assembly of the correct multivalent antibody. The dimerization-specific amino acid may be within the CH1 domain or the CL domain, or a combination thereof. Dimerization-specific amino acids used for pairing the CH1 domain with other CH1 domains (CH1-CH1) and pairing the CL domain with other CL domains (CL-CL) can be found at least in the disclosures of WO 2014082179, the WO 2015181805 family, and WO 2017059551, all of which are incorporated herein by reference in their entirety. The dimerization-specific amino acid may also be within the Fc domain and may be combined with dimerization-specific amino acids within the CH1 or CL domains. In one embodiment, this disclosure provides a bispecific antibody comprising at least one dimerization-specific amino acid pair.

[0173] Changes in the Fc area The Fc region (if present) can be a wild-type Fc region of the IgG1, IgG2, IgG3, or IgG4 subclass.

[0174] In one embodiment, the antibody, multispecific antibody, or antigen-binding fragment thereof comprises an Fc domain of IgG1 or IgG4 having reduced effector function. In another embodiment, the Fc domain comprises the amino acid sequence of SEQ ID NO: 53 or SEQ ID NO: 19. In yet another embodiment, the IgG1 Fc comprises mutants E233P, L234A, L235A, G236del, and P329A.

[0175] In one embodiment, the multispecific antibody or its antigen-binding fragment comprises an Fc domain having an extended half-life. In another embodiment, the multispecific antibody or its antigen-binding fragment comprises the Fc domain of IgG1, wherein a YTE mutation (M252Y / S254T / T256E, EU number, as described in US7658921, which is incorporated herein by reference in its entirety) is introduced at CH2 of the IgG Fc region.

[0176] In one embodiment, the multispecific antibody or its antigen-binding fragment comprises an Fc domain having reduced effector function and an extended half-life. In another embodiment, the Fc domain comprises the amino acid sequence of SEQ ID NO: 20.

[0177] In another embodiment, the antibody disclosed herein has a strong Fc-mediated effector function and the antibody mediates antibody-dependent cytotoxicity (ADCC) against target cells expressing TAA (e.g., GPC3).

[0178] In other respects, the effector function of an antibody can be altered by substituting at least one amino acid residue into the Fc region with different amino acid residues. For example, one or more amino acids can be substituted with different amino acid residues to give the antibody altered affinity for the effector ligand while retaining the antigen-binding ability of the parent antibody. The effector ligand with altered affinity can be, for example, an Fc receptor or the C1 component of complement. This method is described, for example, in U.S. Patents 5,624,821 and 5,648,260 to Winter et al., both of which are incorporated herein by reference in their entirety.

[0179] On the other hand, one or more amino acid residues can be substituted with one or more different amino acid residues to give the antibody altered C1q binding and / or reduced or eliminated complement-dependent cytotoxicity (CDC). This method is described, for example, in U.S. Patent No. 6,194,551 to Idusogie et al., which is incorporated herein by reference in its entirety.

[0180] On the other hand, altering one or more amino acid residues can change the antibody's ability to fix complement. This method is described, for example, in publication WO 94 / 29351 by Bodmer et al., which is incorporated herein by reference in its entirety. In a particular aspect, one or more amino acids of the antibody or its antigen-binding fragment disclosed herein are replaced by one or more allotropic amino acid residues of the IgG1 subclass and the κ isotype. The allotropic amino acid residues also include, but are not limited to, the heavy chain constant regions of the IgG1, IgG2, and IgG3 subclasses and the light chain constant regions of the κ isotype, as described by Jefferis et al., Mabs. [Monoclonal Antibodies] 2009; 1:332-338, which is incorporated herein by reference in its entirety.

[0181] On the other hand, the Fc region can be modified by modifying one or more amino acids to increase the ability of antibody-mediated antibody-dependent cytotoxicity (ADCC) and / or increase the antibody affinity for the Fcγ receptor. This method is described, for example, in Presta Publication WO 00 / 42072, which is incorporated herein by reference in its entirety. Furthermore, binding sites on human IgG1 for FcγRI, FcγRII, FcγRIII, and FcRn have been mapped, and variants with improved binding have been described (see Shields et al., J. Biol. Chem. 2001; 276:6591-6604), which is incorporated herein by reference in its entirety.

[0182] On the other hand, the glycosylation of multispecific antibodies is modified. For example, glycosylated-free antibodies (i.e., antibodies lacking or having reduced glycosylation) can be prepared. For example, glycosylation can be altered to increase the antibody's affinity for an "antigen." Such carbohydrate modification can be achieved, for example, by altering one or more glycosylation sites within the antibody sequence. For example, one or more amino acid substitutions can be performed, resulting in the elimination of one or more variable region framework glycosylation sites, thereby eliminating the glycosylation at that site. Such glycosylation can increase the antibody's affinity for the antigen. This method is described, for example, in U.S. Patent Nos. 5,714,350 and 6,350,861 to Co et al., both of which are incorporated herein by reference in their entirety.

[0183] Alternatively or concurrently, antibodies with altered glycosylation patterns can be prepared, such as hypofucosylated antibodies with reduced amounts of fucosylated residues or antibodies with increased bipartite GlcNac structures. Such altered glycosylation patterns have been shown to increase the ADCC capacity of antibodies. Such carbohydrate modifications can be achieved, for example, by expressing antibodies in host cells with altered glycosylation pathways. Cells with altered glycosylation pathways have been described in the art and can be used as host cells in which recombinant antibodies are expressed to produce antibodies with altered glycosylation. For example, Hang et al.'s EP 1,176,195 (incorporated in its entirety by reference) describes a cell line with a dysfunctional FUT8 gene encoding a fucosylated transferase, such that antibodies expressed in such cell lines exhibit hypofucosylation. Presta’s publication WO 03 / 035835 (incorporated in full by reference) describes a variant CHO cell line, Lecl3, which has a reduced ability to link fucose to Asn(297)-linked carbohydrates, which also leads to low fucosylation of antibodies expressed in this host cell (see also Shields et al., J. Biol. Chem. 2002; 277:26733-26740, in full by reference). Umana et al.'s WO 99 / 54342 (included in full by reference) describes cell lines engineered to express glycoprotein-modified glycosyltransferases (e.g., β(1,4)-N-acetylglucosamine transferase III (GnTIII)), resulting in antibodies expressed in engineered cell lines exhibiting an increased bipartite GlcNac structure, which leads to increased ADCC activity of the antibodies (see also Umana et al., Nat. Biotech. [Nature Biotechnology] 1999; 17:176-180, in full by reference).

[0184] On the other hand, if a reduction in ADCC is expected, many previous reports have shown that the human antibody subclass IgG4 has only a modest ADCC and almost no CDC effector function (Moore GL, et al., MAbs. [Monoclonal Antibodies] 2010;2:181-189, cited in full). However, native IgG4 has been found to be less stable under stress conditions, such as in acidic buffers or at elevated temperatures (Angal, S. Mol Immunol. [Molecular Immunology] 1993; 30:105-108; Dall'Acqua, W. et al., 1998 Biochemistry, 37:9266-9273; Aalberse et al., Immunol. [Immunology] 2002; 105:9-19, all cited in full). Reduced ADCC can be achieved by operatively linking antibodies to modified IgG4 Fc with altered combinations that reduce FcγR binding or C1q binding activity, thereby reducing or eliminating ADCC and CDC effector functions. Given the physicochemical properties of antibodies as biopharmaceuticals, one of the less desirable inherent properties of IgG4 is the dynamic separation of its two heavy chains in solution to form a half-antibody, which leads to the in vivo generation of bispecific antibodies via a process known as “Fab arm exchange” (Van der NeutKolfschoten M, et al., Science. [Science] 2007; 317:1554-157, incorporated herein by reference in its entirety). A serine mutation at position 228 (EU numbering system) to proline exhibits an inhibitory effect on IgG4 heavy chain segregation (Angal, S. Mol Immunol. [Molecular Immunology] 1993; 30:105-108, cited in full; Aalberse et al., Immunol. [Immunology] 2002; 105:9-19, cited in full).It has been reported that some amino acid residues in the hinge and γFc region influence the interaction between antibodies and Fcγ receptors (Chappel SM, et al., Proc. Natl. Acad. Sci. USA. [Proceedings of the National Academy of Sciences] 1991; 88:9036-9040; Mukherjee, J. et al., FASEB J. [Journal of the Federation of American Societies for Experimental Biology] 1995; 9:115-119; Armour, KL et al., Eur J Immunol. [European Journal of Immunology] 1999; 29:2613-2624; Clynes, RA et al., 2000 Nature Medicine, 6:443-446; Arnold JN, Annu Rev Immunol. [Annals of Immunology] 2007; 25:21-50, all of which are incorporated herein by reference in their full text). Furthermore, some rare IgG4 isotypes in the human population can also induce different physicochemical properties (Brusco, A. et al., Eur J Immunogenet. [European Journal of Immunogenetics] 1998; 25:349-55; Aalberse et al., Immunol. [Immunology] 2002; 105:9-19, both incorporated herein by reference in their entirety). To generate multispecific antibodies with low ADCC and CDC but good stability, the hinge and Fc regions of human IgG4 can be modified to introduce numerous alterations. These modified IgG4 Fc molecules can be found in SEQ ID NO: 83-88 of U.S. Patent No. 8,735,553 to Li et al., which is incorporated herein by reference in its entirety.

[0185] In another embodiment, the antibody disclosed herein comprises the Fc domain of human IgG4 having S228P and / or R409K substitutions (according to the EU numbering system).

[0186] Antibody production Antibodies and their antigen-binding fragments can be produced by any method known in the art, including but not limited to recombinant expression of antibody tetramers, chemical synthesis, and enzymatic digestion, while full-length monoclonal antibodies can be obtained, for example, through hybridoma or recombinant generation. Recombinant expression can 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.

[0187] This disclosure further provides polynucleotides encoding the antibodies described herein, such as polynucleotides encoding heavy or light chain variable regions or segments containing the complementarity-determining regions described herein. In some aspects, the polynucleotides encoding the heavy or light chain variable regions have at least 85%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% nucleic acid sequence identity with the polynucleotides selected from the group consisting of SEQ ID NO: 5, SEQ ID NO: 9, SEQ ID NO: 7, SEQ ID NO: 12, SEQ ID NO: 14, SEQ ID NO: 16, SEQ ID NO: 18, SEQ ID NO: 42, or SEQ ID NO: 44.

[0188] The polynucleotides disclosed herein can encode variable region sequences of anti-TAA (e.g., GPC3) xCD137 antibodies. They can also encode both variable and constant regions of the antibody. Some polynucleotide sequences encode polypeptides containing variable regions of the heavy and light chains of example anti-TAA (e.g., GPC3) xCD137 antibodies.

[0189] This disclosure further provides a polynucleotide encoding the anti-GPC3xCD137 antibody described herein. In some aspects, the polynucleotide encoding the first or second polypeptide of the anti-GPC3xCD137 antibody has at least 85%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% nucleic acid sequence identity with the polynucleotide selected from the group consisting of SEQ ID NO: 22, SEQ ID NO: 26, SEQ ID NO: 28, SEQ ID NO: 30, SEQ ID NO: 32, SEQ ID NO: 34, or SEQ ID NO: 24. In some embodiments, the polynucleotide described herein may be codon-optimized for expression in host cells (e.g., eukaryotic cells, more particularly, mammalian cells (e.g., CHO cells)).

[0190] This disclosure also provides expression vectors and host cells for generating the antibodies described herein, such as anti-CD137 antibodies and anti-TAA (e.g., GPC3) xCD137 antibodies. The choice of expression vector depends on the intended host cells for the expression vector. Typically, the expression vector contains a promoter and other regulatory sequences (e.g., enhancers) operably linked to a polynucleotide encoding an antibody chain or antigen-binding fragment. In some respects, inducible promoters are used to prevent the expression of the inserted sequence, in addition to being controlled under induction conditions. Inducible promoters include, for example, arabinose, lacZ, metallothionein promoters, or heat shock promoters. The culture of transformed organisms can be scaled up under non-inducible conditions without bias towards a population of host cells that better tolerate the coding sequence of their expression product. In addition to promoters, other regulatory elements may also be required or desired for the efficient expression of antibodies or antigen-binding fragments. These elements typically include an ATG start codon and an adjacent ribosome binding site or other sequences. Furthermore, expression efficiency can be improved by including enhancers suitable for the cell system in use (see, for example, Scharf et al., Results Probl. Cell Differ. 1994; 20:125; and Bittner et al., Meth. Enzymol. 1987; 153:516, both incorporated herein by reference in their entirety). For example, the SV40 enhancer or CMV enhancer can be used to increase expression in mammalian host cells.

[0191] The host cells used to carry and express the antibody chain can be prokaryotic or eukaryotic. *Escherichia coli* is a prokaryotic host suitable for cloning and expressing the disclosed polynucleotides. Other suitable microbial hosts include bacilli, such as *Bacillus subtilis*, and other Enterobacteriaceae, such as *Salmonella*, *Serratia*, and various *Pseudomonas* species. Expression vectors can also be prepared from these prokaryotic hosts, typically containing expression control sequences (e.g., origin of replication) compatible with the host cell. Furthermore, any number of well-known promoters will be available, such as the lactose promoter system, the tryptophan (trp) promoter system, the β-lactamase promoter system, or promoter systems derived from bacteriophage λ. Promoters are typically optionally controlled by operon sequences and have ribosome binding site sequences, etc., for initiating and completing transcription and translation. Other microorganisms such as yeast can also be used for antibody expression. Combinations of insect cells with baculovirus vectors can also be used. In other respects, mammalian host cells are used to express and produce the antibodies disclosed herein. For example, they can be hybridoma cell lines expressing endogenous immunoglobulin genes or mammalian cell lines carrying exogenous expression vectors. These include any normally dead or normally or abnormally immortalized animal or human cells. Several suitable host cell lines capable of secreting intact immunoglobulins have been developed, for example, including CHO cell lines, various COS cell lines, HEK 293 cells, myeloma cell lines, transformed B cells, and hybridomas. The use of mammalian tissue cell cultures to express peptides is generally discussed, for example, in the following: Winnacker, From Genes to Clones, VCH Publishers, New York, New York City, 1987, incorporated herein by reference in its entirety. Expression vectors for mammalian host cells may include expression control sequences, such as origin of replication, promoters, and enhancers (see, for example, Queen et al., Immunol. Rev. [Immunology Review] 1986; 89:49-68, incorporated herein by reference in its entirety), as well as necessary processing information sites, such as ribosome binding sites, RNA splicing sites, polyadenylation sites, and transcription terminator sequences. These expression vectors typically contain promoters derived from mammalian genes or mammalian viruses. Suitable promoters can be constitutive, cell type-specific, stage-specific, and / or tunable or modulotropic.Useful promoters include, but are not limited to, metallothionein promoters, constitutive adenovirus major late promoters, dexamethasone-inducible MMTV promoters, SV40 promoters, MRP polIII promoters, constitutive MPSV promoters, tetracycline-inducible CMV promoters (such as human immediate early CMV promoters), constitutive CMV promoters, and promoter-enhancer combinations known in the art.

[0192] Production of bispecific antibodies The current standard for engineered heterodimeric antibody Fc domains is the KiH design, which introduces a mutation at the core CH3 domain interface. The resulting heterodimer exhibits a reduced CH3 melting temperature (69°C or lower). In contrast, the Zymeworks Azymetric™ platform (as above) provides heterodimeric Fc designs with thermal stability of 81.5°C, comparable to the wild-type CH3 domain.

[0193] Pharmaceutical Composition Compositions comprising an antibody described herein or an antigen-binding fragment thereof, or comprising a polynucleotide encoding a sequence of an antibody or antigen-binding fragment thereof, are also provided, including pharmaceutical formulations. In some embodiments, the composition comprises one or more antibodies or antigen-binding fragments described herein, or comprises one or more polynucleotides encoding a sequence of one or more antibodies or antigen-binding fragments described herein. These compositions may also comprise suitable carriers, such as pharmaceutically acceptable excipients well known in the art, including buffers.

[0194] The compositions disclosed herein can be in various forms. These include, for example, liquid, semi-solid, and solid dosage forms, such as liquid solutions (e.g., injectable and infusion solutions), dispersions or suspensions, liposomes, and suppositories. A suitable form depends on the intended route of administration and therapeutic application. A typical suitable composition is in the form of an injectable or infusion solution. A suitable route of administration is parenteral (e.g., intravenous, subcutaneous, intraperitoneal, intramuscular). In some embodiments, the antibody is administered by intravenous infusion or injection. In some embodiments, the antibody is administered by intramuscular or subcutaneous injection.

[0195] Detection and Diagnostic Methods The antibody or antigen-binding fragments disclosed herein can be used in a variety of applications, including but not limited to methods for detecting CD137 or GPC3. In one aspect, the antibody or antigen-binding fragment can be used to detect the presence of CD137 or GPC3 in biological samples. As used herein, the term "detection" includes quantitative or qualitative detection. In some aspects, biological samples include cells or tissues. In other aspects, such tissues include normal and / or cancerous tissues that express CD137 or GPC3 at higher levels relative to other tissues.

[0196] In one aspect, this disclosure provides a method for detecting the presence of CD137 or GPC3 in a biological sample. In some aspects, the method includes contacting the biological sample with the antibody described herein under conditions that allow antibody-antigen binding, and detecting whether a complex is formed between the antibody and the antigen. The biological sample may include, but is not limited to, urine, tissue, sputum, or blood samples.

[0197] The method also includes methods for diagnosing impairments related to GPC3 expression. In some aspects, the method includes contacting test cells with an anti-GPC3xCD137 antibody; determining the expression level (quantitative or qualitative) of GPC3 expressed in the test cells by detecting the binding of the anti-GPC3xCD137 antibody to the GPC3 peptide; and comparing the expression level of the test cells with the GPC3 expression level in control cells (e.g., normal cells from the same tissue source as the test cells or cells that do not express GPC3), wherein a higher level of GPC3 expression in the test cells compared to the control cells indicates the presence of impairments related to GPC3 expression.

[0198] VI. Treatment methods Anti-CD137 antibody The antibody or antigen-binding fragments disclosed herein can be used for a variety of applications, including but not limited to methods for treating CD137-related disorders or diseases. In one aspect, CD137-related disorders or diseases are cancer. In the case of CD137 x TAA multispecific antibodies, the cancer can be TAA-specific, where CD137 is used to recruit immune cells to tumors expressing TAA.

[0199] In one aspect, this disclosure provides methods for treating cancer. In some aspects, the method includes administering to a patient in need a therapeutically effective amount of an anti-CD137 antibody or antigen-binding fragment or a multispecific antibody containing CD137, or a pharmaceutical composition thereof. In another aspect, this disclosure provides an anti-CD137 antibody or antigen-binding fragment or a multispecific antibody, or a pharmaceutical composition thereof, for use in the treatment of cancer. In yet another aspect, this disclosure provides the use of an anti-CD137 antibody or antigen-binding fragment, a multispecific antibody, or an antigen-binding fragment thereof or a pharmaceutical composition thereof in the manufacture of a medicament for the treatment of cancer.

[0200] Cancer can include, but is not limited to, stomach 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.

[0201] Anti-GPC3xCD137 multispecific antibody The antibody or antigen-binding fragments disclosed herein can be used in a variety of applications, including but not limited to methods for treating GPC3-related disorders or diseases. In one aspect, GPC3-related disorders or diseases are cancer.

[0202] In one aspect, this disclosure provides methods for treating cancer. In some aspects, the method includes administering a therapeutically effective amount of an anti-GPC3xCD137 antibody or antigen-binding fragment, or a pharmaceutical composition thereof, to a patient in need. In another aspect, this disclosure provides a multispecific antibody or antigen-binding fragment thereof, or a pharmaceutical composition thereof, for use in the treatment of cancer. In yet another aspect, this disclosure provides the use of a multispecific antibody or antigen-binding fragment thereof, or a pharmaceutical composition thereof, in the manufacture of a medicament for the treatment of cancer.

[0203] In one embodiment, the cancer expresses GPC3. In one embodiment, the cancer is an advanced or metastatic solid tumor.

[0204] Cancer may include, but is not limited to, any one or more of the following: liver cancer, lung cancer, gastric cancer, germ cell tumors, thyroid cancer, pancreatic cancer, ovarian cancer, skin cancer, kidney cancer (e.g., nephroblastoma), esophageal cancer, atypical teratoid rhabdomyosarcoma of the brain, or undifferentiated synovial sarcoma. In one embodiment, liver cancer is hepatoblastoma or hepatocellular carcinoma (HCC). In another embodiment, lung cancer is non-small cell lung cancer (NSCLC) or small cell lung cancer (SCLC). In another embodiment, non-small cell lung cancer is squamous non-small cell lung cancer. In another embodiment, non-small cell lung cancer is GPC3+ squamous non-small cell lung cancer. In another embodiment, gastric cancer is alpha-fetoprotein-positive (AFP+) gastric cancer. In another embodiment, kidney cancer is nephroblastoma. In another embodiment, esophageal cancer is esophageal squamous cell carcinoma. In another embodiment, esophageal cancer is GPC3+ esophageal squamous cell carcinoma. In another embodiment, germ cell tumors are yolk sac tumors or non-dysgerminomas.

[0205] other The antibodies or antigen-binding fragments disclosed herein can be administered by any suitable route, including parenteral, intrapulmonary, and intranasal administration, and, if intended for local treatment, intralesional administration. Parenteral infusion includes intramuscular, intravenous, intra-arterial, intraperitoneal, or subcutaneous administration. Administration can be made via any suitable route, such as by injection, like intravenous or subcutaneous injection, depending in part on whether the administration is transient or long-term. This document considers a variety of dosing regimens, including but not limited to single or multiple administrations at different time points, bolus administration, and pulsatile infusion.

[0206] The antibody or antigen-binding fragment disclosed herein can be formulated, administered, and applied in accordance with good medical practice. Factors to be considered in this regard include the specific barrier to treatment, the specific mammal to be treated, the individual patient's clinical condition, the cause of the barrier, the delivery site of the agent, the method of administration, the administration regimen, and other factors known to the medical practitioner. The antibody does not need to be formulated, but optionally may be, with one or more agents currently used for the prevention or treatment of the barrier under investigation. The effective amount of such other agents depends on the amount of antibody present in the formulation, the type of barrier or treatment, and other factors discussed above.

[0207] For the prevention or treatment of disease, the appropriate dosage of the antibody or antigen-binding fragment disclosed herein will depend on the type of disease to be treated, the type of antibody, the severity and course of the disease, whether the antibody is administered for prevention or treatment purposes, prior therapy, the patient's clinical history and response to the antibody, and the judgment of the attending physician.

[0208] VII. Combination Therapy On the one hand, the anti-CD137 antibody or multispecific antibody containing anti-CD137 disclosed herein, such as anti-CD137xTAA antibody or anti-GPC3xCD137 antibody, can be used in combination with other therapeutic agents.

[0209] Other therapeutic agents include, for example, other immune checkpoint antibodies. Such immune checkpoint antibodies may include anti-PD1 antibodies. Anti-PD1 antibodies may include, but are not limited to, tislelizumab, pembrolizumab, or nivolumab. Tislelizumab is disclosed in US 8,735,553. Pembrolizumab (formerly known as MK-3475) is disclosed in US 8,354,509 and US 8,900,587 and is a humanized IgG4-K immunoglobulin that targets the PD1 receptor and inhibits the binding of PD1 receptor ligands PD-L1 and PD-L2. Nivolumab (as disclosed by Bristol-Meyers Squibb) is a fully human IgG4-K monoclonal antibody. Nivolumab (clone 5C4) is disclosed in US patents US 8,008,449 and WO 2006 / 121168.

[0210] Other immune checkpoint antibodies that may be combined with the anti-CD137 antibody or a multispecific antibody containing anti-CD137 disclosed herein may include anti-TIGIT antibodies. Such anti-TIGIT antibodies may include, but are not limited to, the anti-TIGIT antibodies disclosed in WO 2019 / 129261.

[0211] In one embodiment, this disclosure provides the use of the disclosed anti-CD137 antibody or a combination of an anti-CD137 multispecific antibody (e.g., an anti-CD137xTAA antibody or an anti-GPC3xCD137 antibody) and an anti-PD-1 antibody (e.g., tislelizumab or other anti-PD-1 antibodies described above) in the manufacture of a medicament for treating cancer (e.g., the cancer described above). In another embodiment, this disclosure provides the use of the disclosed anti-CD137 antibody or a combination of an anti-CD137 multispecific antibody (e.g., an anti-CD137xTAA antibody or an anti-GPC3xCD137 antibody) and an anti-PD-1 antibody (e.g., tislelizumab or other anti-PD-1 antibodies described above) in the treatment of cancer (e.g., the cancer described above).

[0212] Combination therapy may refer to and include any one of the following: - When such a combination therapy is administered simultaneously to a patient requiring treatment, the components are formulated together into a single dosage form that releases the components to the patient substantially simultaneously. - Such a combination is administered to patients requiring treatment substantially simultaneously, wherein the components are formulated separately into individual dosage forms for the patients to take at substantially the same time, thereby releasing the components to the patients at substantially the same time. - Such a combination therapy is administered sequentially to a patient requiring treatment, wherein the components are formulated separately into individual dosage forms for the patient to take at consecutive times, with significant time intervals between each administration, thereby releasing the components to the patient at substantially different times; and - Such a combination is administered sequentially to a patient in need of treatment, wherein the components are formulated together into a single dosage form for controlled release of the components, thereby releasing them to the patient simultaneously, sequentially and / or overlappingly at the same and / or different times, wherein each part may be administered via the same or different routes.

[0213] definition Unless specifically defined elsewhere in this document, all other technical and scientific terms used herein have the meanings commonly understood by one of ordinary skill in the art.

[0214] As used herein, including the appended claims, unless the context clearly indicates otherwise, the singular forms of “a,” “an,” and “the” include their respective plural references.

[0215] Unless the context clearly indicates otherwise, the term “or” means the term “and / or” and is used interchangeably with the term “and / or”.

[0216] As used herein, the term "anticancer agent" refers to any agent that can be used to treat cell proliferation disorders such as cancer, including but not limited to cytotoxic agents, chemotherapeutic agents, radiotherapy and radiotherapy agents, targeted anticancer agents, and immunotherapy agents.

[0217] The terms “CD137” or “TNFRSF9”, “ILA” or “41BB” or “4-1BB” refer to co-stimulatory molecules belonging to the TNFRSF family. The nucleic acid sequence of human CD137 is shown in SEQ ID NO: 36, based on GenBank accession number: NM_001561.4. The amino acid sequence of human CD137 is shown in SEQ ID NO: 35.

[0218] The term "phosphatidylinositol proteoglycan 3" (GPC3) is also known as DGSX, GTR2-2, MXR7, OCI-5, SDYS, SGB, SGBS, and SGBS1. The amino acid sequence of human GPC3 (SEQ ID NO: 51) can also be found in NCBI reference sequence: NP_004475.1. The nucleic acid sequence of human GPC3 is shown in SEQ ID NO: 52.

[0219] As used herein, the terms “administration” and “treatment” when applied to an animal, human, experimental subject, cell, tissue, organ, or biological fluid mean contact between an exogenous drug, therapeutic agent, diagnostic agent, or composition and that animal, human, subject, cell, tissue, organ, or biological fluid. Cellular treatment encompasses contact between a reagent and a cell, as well as contact between a reagent and a fluid, wherein the fluid contacts the cell. The terms “administration” and “treatment” also mean in vitro and ex vivo treatment, such as treatment of cells by a reagent, diagnostic agent, conjugated compound, or by another cell. The term “subject” herein includes any organism, preferably an animal, more preferably a mammal (e.g., rat, mouse, dog, cat, rabbit), and most preferably a human. In one aspect, treating any disease or disorder means improving that disease or disorder (i.e., slowing or halting or reducing the development of the disease or at least one of its clinical symptoms). In the other aspect, “treatment” means alleviating or improving at least one bodily parameter, including those that the patient may not be able to discern. On another level, “treatment” refers to regulating a disease or disorder in a physical (e.g., stabilization of identifiable symptoms), physiological (e.g., stabilization of bodily parameters), or both. On yet another level, “treatment” refers to preventing or delaying the onset, development, or progression of a disease or disorder.

[0220] In the context of this disclosure, the term “subject” refers to a mammal, such as a primate, preferably a higher primate, such as a human (e.g., a patient who has the disorder described herein or is at risk of developing the disorder described herein).

[0221] As used in this article, the term "affinity" refers to the strength of the interaction between an antibody and an antigen. Within an antigen, the variable region of the antibody interacts with the antigen at many sites through non-covalent forces. Generally, the more interactions, the stronger the affinity.

[0222] As used herein, the term "antibody" refers to a polypeptide of the immunoglobulin family that can bind to a corresponding antigen nonvalently, reversibly, and specifically. For example, naturally occurring IgG antibodies are tetramers comprising at least two heavy (H) chains and two light (L) chains linked together by disulfide bonds. Each heavy chain consists of a heavy chain variable region (abbreviated 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 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), 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 following order from the amino terminus to the carboxyl terminus: FR1, CDR1, FR2, CDR2, FR3, CDR3, and FR4. The variable regions of the heavy and light chains contain binding domains that interact with antigens. The constant regions of antibodies 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 (Clq) of the classical complement system.

[0223] The term "antibody" includes, but is not limited to, monoclonal antibodies, human antibodies, humanized antibodies, chimeric antibodies, anti-idiotype (anti-Id) antibodies, human-engineered antibodies, single-chain antibodies (scFv), single-domain antibodies, Fab fragments, Fab' fragments, or F(ab')2 fragments. Antibodies can be any isotype / class (e.g., IgG, IgE, IgM, IgD, IgA, and IgY) or subclass (e.g., IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2). Furthermore, antibodies include their derivatives, such as those formed directly or indirectly by linking to another agent (e.g., other drugs) or by forming a complex with another agent. The term "antibody" in this document includes monospecific antibodies, bispecific antibodies, and multispecific antibodies.

[0224] The term "chimeric antibody" refers to a molecule composed of domains from different species, that is, the fusion of the variable domain of an antibody from a host species (such as a mouse, rabbit, or llama) with the constant domain of an antibody from a different species (such as a human).

[0225] In some embodiments, the anti-GPC3 antibody includes at least one antigen-binding site and at least one variable region. In some embodiments, the anti-GPC3 antibody includes an antigen-binding fragment from the GPC3 antibody described herein. In some embodiments, the anti-GPC3 antibody is isolated or recombinant.

[0226] In some embodiments, the anti-CD137 antibody includes at least one antigen-binding site and at least one variable region. In some embodiments, the anti-CD137 antibody includes an antigen-binding fragment from the CD137 antibody described herein. In some embodiments, the anti-CD137 antibody is isolated or recombinant.

[0227] The terms “monoclonal antibody” or “mAb” or “Mab” used herein refer to a group of substantially homogeneous antibodies, meaning that the antibody molecules contained in this group are identical in amino acid sequence except for a small number of potentially naturally occurring mutations. In contrast, conventional (polyclonal) antibody formulations typically comprise a variety of different antibodies with different amino acid sequences in their variable domains, particularly their complementarity-determining regions (CDRs), which are generally specific to different epitopes. The modifier “monoclonal” indicates the characteristic of antibodies obtained from a substantially homogeneous group of antibodies and should not be construed as requiring the antibody to be produced 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. [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, Coldspring Harbor Laboratory, 1988; and Colligan et al., CURRENTPROTOCOLS IN IMMUNOLOGY, 1993. The antibodies disclosed herein can be any class of immunoglobulins (including IgG, IgM, IgD, IgE, IgA) and any of their subclasses (e.g., IgG1, IgG2, IgG3, IgG4). Hybridomas that produce monoclonal antibodies can be cultured in vitro or in vivo. High-titer monoclonal antibodies can be obtained through in vivo production, wherein cells from a single hybridoma are injected intraperitoneally into mice, such as primitive-initiated Balb / c mice, to produce ascites containing high concentrations of the desired antibody. Monoclonal antibodies of the same type IgM or IgG can be purified from such ascites or from culture supernatant using column chromatography methods well known to those skilled in the art.

[0228] Typically, the basic antibody structural unit comprises a tetramer. Each tetramer consists of two pairs of identical polypeptide chains, each pair having a "light chain" (approximately 25 kDa) and a "heavy chain" (approximately 50–70 kDa). The amino-terminal portion of each chain includes a variable region of approximately 100 to 110 or more amino acids primarily responsible for antigen recognition. The carboxyl-terminal portion of the heavy chain can be defined as a constant region primarily responsible for effector function. Typically, human light chains are classified as κ and λ light chains. Furthermore, human heavy chains are typically classified as α, δ, ε, γ, or μ, and antibody isotypes are defined as IgA, IgD, IgE, IgG, and IgM, respectively. Within both 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 includes a "D" region of approximately 10 or more amino acids.

[0229] Each light chain / heavy chain (VL / VH) pair has a variable region that forms an antibody binding site. Therefore, in general, an intact antibody has two binding sites. Except in bifunctional or bispecific antibodies, the two binding sites are usually identical in the primary sequence.

[0230] Typically, the variable domains of both heavy and light chains contain three hypervariable regions, also known as "complementarity-determining regions (CDRs)," which lie between relatively conserved frame regions (FRs). CDRs are usually aligned with the frame regions to enable the binding of specific epitopes. Generally, from the N-terminus to the C-terminus, both 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 locations of the CDR and framework regions can be determined using a variety of definitions well-known in the art, such as those of Kabat, Chothia, AbM, and IMGT (see, for example, Johnson et al., Nucleic Acids Res. 2001; 29:205-206; Chothia and Lesk, J. Mol. Biol. 1987; 196:901-917; Chothia et al., Nature. 1989; 342:877-883; Chothia et al., J. Mol. Biol. 1992; 227:799-817; Al-Lazikani et al., J. Mol. Biol. 1997; 273:927-748; ImMunoGenTics (IMGT) number (Lefranc, M.-P., The Immunologist). [Immunologist] 1999; 7, 132-136; Lefranc, M.-P. et al., Dev. Comp. Immunol. [Developmental and Comparative Immunology] 27, 55-77 (2003) (“IMGT” numbering scheme)).The definition of antigen-binding sites is also described in the following literature: 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. In Oxford, 141-172 (1996). For example, according to 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); and the CDR amino acid residues in the light chain variable domain (VL) are numbered 24-34 (LCDR1), 50-56 (LCDR2), and 89-97 (LCDR3). According to Chothia, the CDR amino acids in VH are numbered 26-32 (HCDR1), 52-56 (HCDR2), and 95-102 (HCDR3); and the amino acid residues in VL are numbered 26-32 (LCDR1), 50-52 (LCDR2), and 91-96 (LCDR3). By combining the CDR definitions from Kabat and Chothia, the CDR is composed 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. According to IMGT, the CDR amino acid residues in VH are numbered approximately 26-35 (HCDR1), 51-57 (HCDR2), and 93-102 (HCDR3), and the CDR amino acid residues in VL are numbered approximately 27-32 (LCDR1), 50-52 (LCDR2), and 89-97 (LCDR3) (numbering according to Kabat). The CDR region of an antibody can be determined using the IMGT / DomainGap Align procedure according to IMGT.

[0231] The term “hypervariant region” refers to the amino acid residues in an antibody responsible for antigen binding. Hypervariant regions contain amino acid residues from “CDRs” (e.g., LCDR1, LCDR2, and LCDR3 in the light chain variable domain and HCDR1, HCDR2, and HCDR3 in the heavy chain variable domain). See Kabat et al., (1991) Sequences of Proteins of Immunological Interest, 5th edition, Public Health Service, National Institutes of Health, Bethesda, Maryland (Defining CDR regions of antibodies by sequence); also see Chothia and Lesk, J. Mol. Biol. 1987; 196: 901-917 (Defining CDR regions of antibodies by structure). The terms “frame” or “FR” residues refer to those variable domain residues other than the hypervariant residues defined herein as CDR residues.

[0232] Unless otherwise stated, "antigen-binding fragment" refers to an antigen-binding fragment of an antibody, i.e., an antibody fragment that retains the ability to specifically bind to an antigen that binds to a full-length antibody, such as a fragment retaining one or more CDR regions. Examples of antigen-binding fragments include, but are not limited to, Fab, Fab', F(ab')2, and Fv fragments; biantibodies; linear antibodies; single-chain antibody molecules (e.g., single-chain Fv(ScFv)); nanobodies; and multispecific antibodies formed from antibody fragments.

[0233] As used herein, “specific binding” of an antibody to a target protein means that the antibody preferentially binds to the target protein compared to other proteins, but this specificity does not require absolute binding specificity. In the context of describing the interaction between an antigen (e.g., a protein) and an antibody or antigen-binding antibody fragment, “specific binding” or “selective binding” of an antibody refers to a binding reaction that determines the presence of an antigen in a heterogeneous population of proteins and other biological agents (e.g., in biological samples, blood, serum, plasma, or tissue samples). Thus, under certain specified immunoassay conditions, an antibody or its antigen-binding fragment binds specifically to a particular antigen at least twice the background level and does not bind specifically to other antigens present in the sample in significant amounts. On the other hand, under specified immunoassay conditions, an antibody or its antigen-binding fragment binds specifically to a particular antigen at least ten (10) times the background binding level and does not bind specifically to other antigens present in the sample in significant amounts.

[0234] As used herein, "antigen-binding domain" refers to the portion of an antibody that specifically binds to an antigen. In some embodiments, it comprises at least six CDRs and binds specifically to an epitope (or three CDRs, in the case of a single-domain antibody). An "antigen-binding fragment" 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 specifically binds to a second epitope. Multispecific antibodies can be bispecific, trispecific, tetraspecific, etc., having an antigen-binding domain for each specific epitope. Multispecific antibodies can be multivalent antibodies (e.g., bispecific tetravalent antibodies) that comprise multiple antigen-binding domains, such as 2, 3, 4, or more antigen-binding domains that specifically bind to a first epitope and 2, 3, 4, or more antigen-binding domains that specifically bind to a second epitope.

[0235] The term "human antibody" in this article refers to an antibody that contains only the sequence of human immunoglobulin proteins. Human antibodies may contain mouse carbohydrate chains if generated in mice, mouse cells, or hybridomas derived from mouse cells. Similarly, "mouse antibody" or "rat antibody" refers to an antibody that contains only the sequence of mouse or rat immunoglobulin proteins, respectively.

[0236] The term "humanized" or "humanized antibody" refers to an antibody form containing sequences derived from non-human (e.g., mouse, rabbit, llama, etc.) antibodies as well as human antibodies. Such antibodies contain minimal sequences derived from non-human immunoglobulins. Typically, humanized antibodies will contain essentially at least one, and typically both, variable domains, whose hypervariable loops all or substantially all correspond to those of non-human immunoglobulins, and whose FRs are all or substantially all of those sequences of human immunoglobulins. Humanized antibodies will also optionally contain at least a portion of the immunoglobulin constant region (Fc), typically at least a portion of a human immunoglobulin. When it is necessary to distinguish humanized antibodies from parental rodent antibodies, the prefix "hum," "hu," "Hu," or "h" is added to the antibody clone name. Humanized forms of rodent / cameloid antibodies typically contain the same CDR sequence as the parental rodent antibody, but may include certain amino acid substitutions to increase affinity, increase the stability of the humanized antibody, remove post-translational modifications, or for other reasons.

[0237] The term "corresponding phylogenetic sequence" refers to a nucleic acid sequence encoding a human variable region amino acid sequence or subsequence that has 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 human immunoglobulin variable region sequences. A corresponding phylogenetic sequence can 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. A corresponding phylogenetic sequence can be a frame region only, a complementarity-determining region only, a frame and a complementarity-determining region, a variable region (as defined above), or other combinations of sequences or subsequences containing a variable region. Sequence identity can be determined using the methods described herein, such as aligning two sequences using BLAST, ALIGN, or another alignment algorithm known in the art. A corresponding phylogenetic nucleic acid or amino acid sequence can have at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity with a reference variable region nucleic acid or amino acid sequence. Furthermore, if the antibody contains a constant region, the constant region is also derived from such human sequences, such as human germline sequences, or mutant forms of human germline sequences, or antibodies containing common frame sequences derived from human frame sequence analysis, as described by Knappik et al., J. Mol. Biol. [Journal of Molecular Biology] 296:57-86, 2000.

[0238] The “2+2 format” refers to a bispecific antibody that targets two different antigens or two different epitopes, in which the antibody contains two first antigen-binding domains that specifically bind to the first antigen or the first epitope, and two second antibody-binding domains that specifically bind to the second antigen or the second epitope.

[0239] The term "equilibrium dissociation constant (K)" D "M" refers to the dissociation rate constant (kd, time). -1 Divide by the association rate constant (ka, time) -1 M -l The equilibrium dissociation constant can be measured using any method known in the art. The antibodies disclosed herein will typically have a dissociation constant of less than about 10. -7 Or 10 -8 M, for example, less than about 10 -9 M or 10 -10 M, in some respects, is less than approximately 10. -11 M, 10 -12 M or 10 -13 The equilibrium dissociation constant of M.

[0240] The terms “cancer” or “tumor” used herein have the broadest meaning as understood in the art, and refer to a physiological condition in mammals characterized by uncontrolled cell growth. In the context of this disclosure, cancer is not limited to a particular type or location.

[0241] In the context of this disclosure, when referring to amino acid sequences, the term "conservative substitution" means replacing an original amino acid with a new amino acid that substantially does not alter the chemical, physical, and / or functional properties of the antibody or fragment, such as its binding affinity to GPC3 or CD137. In particular, common conserved substitutions of amino acids are well known in the art.

[0242] As used herein, the term "knob-into-hole" technique refers to the process by which amino acids, in vitro or in vivo, guide the pairing of two polypeptides by introducing a spatial protrusion (knob) into one polypeptide and a cavity (pothole) into another polypeptide at their interacting interface. For example, knob-into-hole techniques have been introduced into the Fc:Fc binding interface of antibodies, C... L :C H I-interface, or V-interface H / V L Interface (see, for example, US 2011 / 0287009, US 2007 / 0178552, WO 96 / 027011, WO98 / 050431, and Zhu et al., Protein Science. 1997; 6:781-788). In some embodiments, the mortis ensures that two different heavy chains are correctly paired together during the manufacture of multispecific antibodies. For example, multispecific antibodies having mortis amino acids in their Fc regions may also include a single variable domain linked to each Fc region, or further include different heavy chain variable domains paired with similar or different light chain variable domains. The mortis technique may also be used in VH or VL regions to ensure correct pairing.

[0243] As used herein in the context of the "mortar and pestle" technique, the term "mortar" refers to an amino acid change in which a polypeptide introduces a protrusion (mortar) at the interface where it interacts with another polypeptide. In some embodiments, the other polypeptide has a mortar mutation.

[0244] As used herein in the context of "mortar and pestle," the term "mortar" refers to an amino acid alteration that introduces a pocket or cavity (mortar) into the interface where the polypeptide interacts with another polypeptide. In some embodiments, the other polypeptide has a pestle mutation.

[0245] An example of an algorithm suitable for determining percentage sequence identity and sequence similarity is the BLAST algorithm, described in Altschul et al., Nuc. Acids Res. 25:3389-3402, 1977; and Altschul et al., J. Mol. Biol. 1990; 215:403-410. Software for performing BLAST analysis is available from the National Center for Biotechnology Information. This algorithm involves first identifying high-scoring sequence pairs (HSPs) of short word length W in the query sequence, which match or satisfy some positive threshold score T when compared to sequences of the same word length in the database. T is called the neighborhood word score threshold. These initial neighborhood word hits serve as the starting point for a search to find values ​​of longer HSPs containing them. Word hits extend in both directions along each sequence until the cumulative alignment score can be increased. For nucleotide sequences, the cumulative score is calculated using parameters M (reward score for a pair of matching residues; always >0) and N (penalty score for mismatched residues; always <0). For amino acid sequences, a score matrix is ​​used to calculate the cumulative score. A stop word is hit in each direction of extension if: the cumulative alignment score decreases by an amount X from its maximum realized value; the cumulative score tends to zero or lower due to the accumulation of one or more negative score residue alignments; or the end of either sequence is reached. The BLAST algorithm parameters W, T, and X determine the sensitivity and speed of the alignment. The BLASTN program (for nucleotide sequences) defaults to a word length (W) of 11, an expected value (E) of 10, M = 5, N = -4, and compares two strands. For amino acid sequences, the BLAST program defaults to a word length of 3, an expected value (E) of 10, and a BLOSUM62 score matrix (see Henikoff and Henikoff, Proc. Natl. Acad. Sci. USA. [Proceedings of the National Academy of Sciences] 1989; 89: 10915) with an alignment (B) of 50, an expected value (E) of 10, M = 5, N = -4, and comparison of the two strands.

[0246] The BLAST algorithm also performs statistical analysis on the similarity between two sequences (see, for example, Karlin and Altschul, Proc. Natl. Acad. Sci. USA [Proceedings of the National Academy of Sciences] 90:5873-5787, 1993). One similarity measure provided by the BLAST algorithm is the minimum sum probability (P(N)), which indicates the probability that a match will occur by chance between two nucleotide or amino acid sequences. For example, if the minimum sum probability in a comparison of the test nucleic acid with a reference nucleic acid is less than about 0.2, more preferably less than about 0.01, and most preferably less than about 0.001, then the nucleic acid is considered similar to the reference sequence.

[0247] The percentage of identity between two amino acid sequences can also be determined using the algorithm incorporated into the ALIGN algorithm (version 2.0) by E. Meyers and W. Miller, Comput. Appl. Biosci., 1988; 4:11-17, using a PAM120 weighted residue table, a vacancy length penalty of 12, and a vacancy penalty of 4. Alternatively, the percentage of identity between two amino acid sequences can be determined using the algorithm by Needleman and Wunsch, J. Mol. Biol. 48:444-453 (1970), which is incorporated into the GAP program in the GCG software package, using a BLOSUM62 or PAM250 matrix with vacancy weights of 16, 14, 12, 10, 8, 6, or 4, and length weights of 1, 2, 3, 4, 5, or 6.

[0248] The term "nucleic acid" is used interchangeably with the term "polynucleotide" herein and refers to deoxyribonucleotides or ribonucleotides in single-stranded or double-stranded form and their polymers. This term encompasses nucleic acids containing known nucleotide analogs or modified backbone residues or linkages, which are synthetic, naturally occurring, and non-natural, have similar binding properties to a reference nucleic acid, and are metabolized in a similar manner to a reference nucleotide. Examples of such analogs include, but are not limited to, phosphate thioesters, aminophosphate esters, methylphosphonates, chiral methylphosphonates, 2-O-methylribonucleotides, and peptide-nucleic acids (PNAs).

[0249] In the context of nucleic acids, the term "operably linked" refers to a functional relationship between two or more segments of a polynucleotide (e.g., DNA). Typically, it refers to the functional relationship between a transcriptional regulatory sequence and a transcriptional sequence. For example, a promoter or enhancer sequence is operably linked to a coding sequence if it stimulates or regulates transcription of the coding sequence in a suitable host cell or other expression system. Usually, promoters that are operably linked to transcriptional sequences are physically adjacent to the transcriptional sequence; that is, they act in cis. However, some transcriptional regulatory sequences (such as enhancers) do not need to be physically adjacent to or immediately adjacent to the coding sequence that enhances their transcription.

[0250] As used herein, the term "pharmaceuticalally acceptable excipient" includes any and all physiologically compatible solvents, dispersion media, isotonic agents, and absorption delay agents. Excipients may be suitable for intravenous, intramuscular, subcutaneous, parenteral, rectal, spinal, or epidermal administration (e.g., by injection or infusion).

[0251] As used herein, the term "therapeutic effective amount" refers to the amount of antibody sufficient to affect the treatment of a disease, disorder, or symptom when administered to a subject to treat at least one clinical symptom of that disease, disorder, or symptom. "Therapeutic effective amount" can vary depending on the antibody, the disease, disorder, and / or the symptom of the disease or disorder, the severity of the disease, disorder, and / or symptom of the disease or disorder, the age of the subject being treated, and / or the weight of the subject being treated. The appropriate amount in any given situation will be obvious to those skilled in the art or can be determined by routine experiments. In the case of combination therapy, "therapeutic effective amount" refers to the total amount of the combination of subjects used to effectively treat the disease, disorder, or symptom.

[0252] The term "combination therapy" refers to the administration of two or more therapeutic agents to treat the conditions or disorders described in this disclosure. Such administration encompasses the combined administration of these therapeutic agents in a substantially simultaneous manner. Such administration also encompasses combined administration in multiple containers or in separate containers (e.g., capsules, powders, and liquids) for each active ingredient. Powders and / or liquids may be reconstituted or diluted to the desired dose prior to administration. Furthermore, such administration also encompasses the sequential use of each type of therapeutic agent at substantially the same time or at different times. In any case, the treatment regimen will provide the beneficial effect of the combination of drugs in treating the conditions or disorders described herein.

[0253] As used herein, the phrase "in combination with" means administering the antibody described herein to a subject concurrently with, before, or after administration of another therapeutic agent. In some embodiments, the antibody described herein is administered as a co-preparation with another therapeutic agent.

[0254] equivalent It should be understood that although the invention has been described in conjunction with a detailed description, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.

[0255] It should be understood that one, some, any, or all of the features of the various embodiments described herein may be combined to form other embodiments of this disclosure. These and other aspects of this disclosure will become apparent to those skilled in the art.

[0256] Example Example 1. Generation of single-chain anti-huCD137 VHH antibody CD137 recombinant protein for phage activity and binding assays To discover VHH antibodies against human CD137, several recombinant proteins were designed and expressed for phage panning and screening. The full-length human CD137 (huCD137) cDNA coding region was sequenced based on the CD137 GenBank sequence (accession number: NM_001561.4, available from Sinobio, catalog number: HG10041-M, cited in this article as SEQ ID NO: 36). The human CD137 ligand (TNFSF9) was sequenced based on the CD137 ligand GenBank sequence (accession number: NM_003811.3, available from Sinobio, catalog number: HG15693-G). In summary, PCR amplification was performed on the coding regions of the extracellular domain (ECD) of huCD137 (SEQ ID NO: 35) composed of amino acids (AA) 24-183 and the ECD of the huCD137 ligand (SEQ ID NO: 37) composed of AA 71-254, respectively. PCR amplification was also performed on the coding region of mIgG2a Fc (SEQ ID NO: 39), followed by conjugation with the ECD of human CD137 or the ECD of the human CD137 ligand via overlap PCR to prepare the mIgG2a Fc fusion protein. The PCR products were then cloned into a pcDNA3.1-based expression vector (Invitrogen, Carlsbad, California, USA) to generate two recombinant mIgG2a Fc fusion protein expression plasmids: human CD137 ECD-mIgG2a and human CD137 ligand ECD-mIgG2a. Alternatively, the ECD coding region of huCD137 (SEQ ID NO:35), consisting of AA 24-183, was also cloned into a pcDNA3.1-based expression vector (Ingenie Biosciences, Carlsbad, California, USA), with its C-terminus fused to a 6xHis tag, thus generating human CD137-ECD-his. To generate the recombinant fusion protein, the plasmid was transiently transfected into a HEK293-based mammalian cell expression system (in-house developed) and cultured for 5–7 days in a CO2 incubator equipped with a rotary shaker. The supernatant containing the recombinant protein was collected and centrifuged for clarification. The recombinant protein was purified using a Protein A column (catalog number: 17127901, General Life Sciences) or Ni-NTA agarose gel (catalog number: R90115, Ingenie Biosciences). All recombinant proteins were dialyzed against phosphate-buffered saline (PBS) and stored in aliquots at -80°C.

[0257] Construction of llama immune and phage libraries One llama was immunized with human CD137 ECD-mIgG2a. Two weeks after the fourth immunization, llama PBMCs were collected using standard techniques for RNA extraction (Chomczynski et al., Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction, Analytical. Biochem. 1987; 162(1):156-159).

[0258] Phage libraries were constructed using reverse transcription and splicing overlap extension PCR. The PCR products were double-digested with NcoI / NotI and ligated into the phage vector pCANTAB-5E. The library was then transformed into *E. coli* TG1 bacteria and validated by Sanger sequencing of randomly cloned DNA (analysis of >96 clones). After a phage rescue step using KM13 helper phages, the phages were purified by two precipitations directly from the culture supernatant using PEG / NaCl. After transformation into *E. coli* bacteria, phages with sizes >10 were obtained. 7 The library.

[0259] Phage display screening and selection Phage display selection was performed using a standard protocol (Silacci et al., (2005) Proteomics, 5, 2340-50; Zhao et al., (2014) PLoS One, 9, e111339). Briefly, in rounds 1 and 2, 10 µg / ml immobilized human CD137 ECD-mIgG2a (catalog number 470319, Thermo Fisher Scientific) was used. In round 3, selection was performed using Hut78 / huCD137 cells. Immunotubes were blocked for 1 hour with PBS solution of 5% milk powder (w / v) supplemented with 1% Tween 20 (MPBST). After washing with PBST (PBS buffer supplemented with 0.05% Tween 20), 5 × 10⁶ cells were selected from each sub-library. 12 One (round 1) or 5 × 10 11Two (round 2) phages were depleted for 1 hour in MPBST using human CD40 ECD-mIgG2a, and then incubated with the antigen for 1 hour. For the third round of selection, Hut78 / huCD137 cells and HEK293 (ATCC, CRL-1573) cells were used as depleted cells for cell panning. After washing with PBST, the bound phages were eluted with 100 mM triethylamine (Sigma-Aldrich). The eluted phages were used to infect mid-log Escherichia coli TG1 bacteria and inoculated onto TYE agar plates supplemented with 2% glucose and 100 µg / ml ampicillin. After three rounds of selection, single clones were picked, and a phage-containing supernatant was prepared using a standard protocol. Phage ELISA was used to screen for anti-huCD137 VHH antibodies.

[0260] For phage ELISA, Maxisorp immunoassay plates were coated with antigen and blocked with 5% milk powder (w / v) PBS buffer. Phage supernatant was blocked with MPBST for 30 minutes and then added to the wells of the ELISA plate for 1 hour. After washing with PBST, bound phages were detected using HRP-conjugated anti-M13 antibody (GE Healthcare) and 3,3',5,5'-tetramethylbenzidine substrate (catalog number: 00-4201-56, eBioscience, USA). Cells expressing CD137 (10... 5 Cells / well were incubated with ELISA-positive phage supernatant and then bound to Alexa Fluro-647-labeled anti-M13 antibody (GE Healthcare). Cell fluorescence was quantified using a flow cytometer (Guava easyCyte™ 8HT, Merck-Millipore, USA).

[0261] Expression and purification of Fc fusion VHH antibody Then, using an internally developed expression vector, the anti-huCD137 VHH antibody was constructed into a human Fc fusion VHH antibody form (VHH-Fc). The VHH domain antibody was fused to the N-terminus of the human Fc with the G4S linker (SEQ ID NO: 62). An inactive form of human IgG1 Fc (an inert Fc without binding to FcγR, SEQ ID NO: 19) was used. Expression and preparation of the Fc fusion VHH antibody were achieved by transfection into 293G cells and purification using a protein A column (catalog number 17543802, General Life Sciences). The purified antibody was concentrated in PBS to 0.5–5 mg / mL and aliquoted and stored at -80°C.

[0262] Example 2. Characterization of purified anti-huCD137 VHH antibody For antigen ELISA, Maxisorp immunoassay plates were coated with antigen and blocked with 3% BSA (w / v) in PBS buffer (blocking buffer). Monoclonal VHH antibodies were blocked with blocking buffer for 30 minutes and then added to the wells of the ELISA plate for 1 hour. After washing with PBST, bound antibodies were detected using HRP-conjugated anti-human IgG antibody (Sigma, A0170) and 3,3',5,5'-tetramethylbenzidine substrate (catalog number: 00-4201-56, E. Biotech, USA). Ligand competition was also applied to the ELISA assay. ELISA analysis and ligand competition results for a representative clone, BGA-9612, are shown in [the table / link]. Figure 1 The results show that BGA-9612 (SEQ ID NO: 1-5) can bind to human CD137 with good affinity, but not to mouse CD137. The binding of human CD137 to BGA-9612 can be reduced due to competition with the huCD137 ligand.

[0263] Example 3. Humanization of anti-human CD137 VHH BGA-9612 For the humanization of BGA-9612, blasting was performed on the human immunoglobulin gene databases on the IMGT and NCBI websites to search for cDNA sequences in human germline IgG genes that were highly homologous to the variable region of BGA-9612. The human IGVH gene, which is frequently found in human antibody libraries (Glanville 2009 PNAS [Proceedings of the National Academy of Sciences] 106:20216-20221) and is highly homologous to BGA-9612, was selected as the humanization template.

[0264] Humanization was performed via CDR transplantation (Methods in Molecular Biology, Vol 248: Antibody Engineering, Methods and Protocols, Humana Press), and humanized VHH from BGA-9612 was engineered into Fc-VHH form using internally developed expression vectors. Using internally developed expression vectors containing an Fc variant of human IgG1 (SEQ ID NO: 19), humanized VHH from BGA-9612 was fused to the C-terminus of Fc in Fc-VHH form with a G4S adapter (SEQ ID NO: 62). These vectors have easily adaptable subcloning sites. Expression and preparation of humanized VHH from BGA-9612 were achieved by transfecting the construct into ExpiCHO™ cells and purifying using a protein A column. The purified antibody was concentrated in PBS to 0.5–5 mg / mL and aliquoted and stored at -80°C.

[0265] Frame exchange In the first round of humanization, mutations in camel-like framework residues in the frame region were guided by simulated 3D structures, preserving structurally important camel-like framework residues crucial for maintaining the classical structure of the CDR, including amino acid residues R27, F37, E44, R45, Q46, Y47, G49, V78, I94, and Q103 (Kabat numbers). Specifically, the CDR of BGA-9612 VHH (SEQ ID NO: 1-3) was transplanted into the frame of the human germline variable gene IGVH3-23, preserving several camel-like framework residues, resulting in BGA-6582 (SEQ ID NO: 8-9). The binding affinity of BGA-9612 and BGA-6582 via SPR is shown in Table 2.

[0266] In this example, the Kabat number and Kabat definition are used for sequences in CDR and VH / VL. The Fc sequence uses the EU number.

[0267] Table 2. Binding affinity of BGA-9612 and BGA-6582 via SPR

[0268] Based on BGA-6582, we performed several single mutations (converting cameloid residues retained in the frame region to corresponding human residues), as well as combinations of single mutations, as shown in Table 3. All humanization mutations were performed using primers containing mutations at specific sites and a site-directed mutagenesis kit (catalog number FM111-02, TransGen, Beijing, China). The desired mutations were validated by sequencing analysis. These further humanized VHHs from BGA-6582 were tested in SPR binding assays, as shown in Table 3. BGA-3726 (SEQ ID NOs: 1-3 and 6-7), including the BGA-6582-based amino acid Q46E, was selected for further engineering. Amino acid residues F37, Y47, G49, and I94 in the frame region are critical for binding CD137 (Kabat numbers).

[0269] Table 3. Binding affinity of variants with a single amino acid mutation based on BGA-6582

[0270] Improved biophysical properties Humanized VHH BGA-3726 did not exhibit superior overall biophysical properties (e.g., Tm or Tagg) compared to the camel-dwelling VHH BGA-9612 (data not shown). Therefore, BGA-3726 was further engineered by introducing mutations in the CDR and frame regions to improve its biophysical properties for human therapeutic use.

[0271] Amino acid N73 in frame region 3 (FR3) of BGA-3726 was identified as a deamination hotspot. To mitigate the risk of post-translational modification (PTM), the subsequent amino acid S74 was mutated to alanine (reverting to a camel-like amino acid residue).

[0272] In summary, based on BGA-3726, the following engineered humanized VHHs were derived from the mutation process described above: (1) BGA-3544, containing amino acids N64K and N65G in HCDR2, and L5V, S82bN, A84P, and L108Q (Kabat numbers) in the frame (SEQ ID NO: 1, 10, 3, and 11-12); (2) BGA-7031, containing L5V, S82bN, A84P, and L108Q (Kabat numbers) in the frame (SEQ ID NO: 1-3 and 15-16); (3) BGA-9502, containing amino acids N64K and N65G in HCDR2, and L5V, S74A, S82bN, A84P, and L108Q (Kabat numbers) in the frame (SEQ ID NO: 13-14); (4) BGA-2524 contains L5V, S74A, S82bN, A84P and L108Q (Kabat number) in the frame (SEQ ID NO: 17-18).

[0273] For affinity assays, antibodies were captured on the anti-human Fc surface and used for affinity determination based on surface plasmon resonance (SPR) technology. Table 4 summarizes the results of the binding characteristics of the anti-CD137 antibodies in the SPR assay. BGA-3544 and BGA-7031 showed similar binding affinities, with dissociation constants of 17.6 nM and 11.7 nM, respectively, which are comparable to the dissociation constant of BGA-9612 (15.2 nM). BGA-6582 also exhibited binding affinities comparable to BGA-9612.

[0274] Table 4. Comparison of binding affinity between BGA-9612 and humanized VHH via SPR

[0275] The biophysical properties of the chimeric and humanized Fc-VHHs were tested. The tested biophysical properties included melting temperature by DSC, aggregation temperature by SLS266, hydrophobicity by HIC-HPLC, and self-association tendency by AC-SINS (see detailed description below). BGA-9612 exhibited the best thermal stability by Tm and Tagg, and good colloidal stability by AC-SINS. As shown in Table 5, the humanized VHHs BGA-3544 and BGA-7031 showed overall biophysical properties comparable to BGA-9612 (chimeric). Furthermore, BGA-3544 and BGA-7031 showed improved Tm compared to BGA-9612 (chimeric).

[0276] Table 5. Summary of biophysical characteristics against CD137 VHH

[0277] Melting temperatures (Tm) were determined using a high-throughput MicroCal™ VP-Capillary DSC (Malvern Instruments, Northampton, MA). Thermal maps of each protein (350 μL, 0.5 mg / mL) from 20°C to 100°C were obtained using a scan rate of 60°C / hr. Thermal maps of individual buffer solutions were subtracted from each protein sample. The results will display the midpoint of the transition temperature (Tm) and the enthalpy (ΔH) of the sample.

[0278] Aggregation temperature Tagg (°C) represents the colloidal stability of the sample and is obtained by monitoring the onset of aggregation using an UNCLE™ (Unchainedlab, Pleasanton, CA) via an SLS266. The sample was loaded into the Uni, and the temperature was increased from 15°C to 95°C. Back-reflective optics cannot detect near-UV light scattering from protein aggregates; therefore, only non-scattered light reaches the detector. Thus, the reduction in back-reflected light is a direct measure of aggregation in the sample.

[0279] To determine the hydrophobicity of a given VHH using HPLC e2695 (Waters Corporation, Milford, MA), protein samples were diluted in mobile phase A (50 mM sodium phosphate, 1.5 M ammonium sulfate, pH 7.0) and then filtered before being loaded onto a MAbPac™ HIC-10 column (Thermo Fisher Scientific, Waltham, MA) equilibrated in mobile phase A. Samples were eluted using a reverse gradient from mobile phase A to mobile phase B (50 mM sodium phosphate, pH 7.0). After elution, the A280 nm fraction was recorded over time, and the data were exported and analyzed using Empower™ software. The retention time for each sample was compared to a reference, and the hydrophobicity of the VHH was determined, with longer elution times correlated with higher hydrophobicity.

[0280] AC-SINS assays measure protein self-interactions by capturing VHH on a gold colloid surface that exhibits surface resonance oscillations at visible frequencies. When the immobilized antibody self-interacts, the colloid aggregates, altering its oscillation frequency to absorb longer wavelengths. Gold nanoparticles were incubated with an 80 / 20 (v / v) mixture of capturing and non-capturing antibodies. The coated gold nanoparticles were then concentrated 10x to PBS. The sample was pre-diluted in PBS to a concentration of 50 μg / mL. 10 μL of the 10x concentrated AuNP was incubated with 100 μL of the sample in the dark for 2 hours at room temperature in a 384-well plate. The absorbance spectrum from 510 to 570 nm was then read from each well using a BMG ClarioStar™ (BMG Labtech, Offenburg, Germany). The redshift of the maximum absorption peak and its intensity indicated the self-interaction tendency of the tested VHH sample.

[0281] Example 4. Binding activity of humanized VHH to natural CD137 To evaluate the binding activity of anti-CD137 antibodies to native CD137 on live cells, HuT78 cells were engineered to overexpress human CD137. Live HuT78 / CD137 cells were seeded in 96-well plates and incubated with a series of diluted anti-CD137 antibodies. Goat anti-human IgG was used as a secondary antibody to detect antibody binding to the cell surface. ECs showed dose-dependent binding to native human CD137. 50 The values ​​were determined by fitting the dose-response data to a four-parameter logistic model using GraphPad Prism™. The data was... Figure 2 As shown in Table 6, humanized VHH retains sub-nanomolar binding affinity to natural CD137.

[0282] Table 6. Combinations of chimeric and humanized VHH with HuT78 / CD137

[0283] Example 5. Anti-GPC3xCD137 multispecific antibody Activating anti-huCD137 antibodies have shown toxicity in clinical settings, which may indicate systemic Fc R-crosslinking is not ideal for CD137 activation. The goal is to achieve effective CD137 stimulation specifically at the tumor site, without requiring systemic CD137 activation across a wide range of cancers. To overcome Fc... Based on the R-crosslinking dependence, we generated a multispecific antibody against GPC3xCD137 with the following characteristics: Figure 3As shown in the figure. This specific construct comprises an IgG fusion-like multispecific antibody in a module ratio of 2:2, a bivalent F(ab')2 fragment binding to human GPC3, a VH domain fragment fused to the C-terminus of CH3 (which binds huCD137), and an Fc-invalid form of huIgG1 (IgG1mf Fc, SEQ ID NO: 53), which lacks FcγR binding but retains FcRn binding. The amino acid and DNA sequences of the GPC3 antibody are shown in Table 7.

[0284] Table 7. Amino acid and DNA sequences of GPC3 antibodies

[0285] To further increase the in vivo half-life of the GPC3xCD137 multispecific antibody, a YTE mutation (M252Y / S254T / T256E, EU number) located at CH2 in the IgG Fc region was introduced into IgG1mf Fc to generate IgG1mf Fc-YTE (SEQ ID NO: 20).

[0286] The following GPC3xCD137 multispecific antibody was generated.

[0287] BE-933: GPC3 antibody (SEQ ID No: 41-50) was combined with CD137 VHH BGA-9612 (SEQ ID NO: 1-5) to produce the construct BE-933 (VL is SEQ ID NO: 23-24, VH is SEQ ID NO: 27-28).

[0288] BE-774: The YTE mutation in Fc was introduced into BE-933 to generate BE-774 (VL is SEQ ID NO: 23-24, VH is SEQ ID NO: 25-26).

[0289] BE-653: GPC3 antibody (SEQ ID No: 41-50) was combined with CD137 VHH BGA-7031 (SEQ ID No: 1-3 and 15-16) to produce the construct BE-653 (VL is SEQ ID NO: 23-24, VH is SEQ ID NO: 29-30).

[0290] BE-915: The construct BE-915 (VL is SEQ ID NO: 23-24, VH is SEQ ID NO: 21-22) is generated by combining GPC3 antibody (SEQ ID No: 41-50) with CD137 VHH BGA-2524 (SEQ ID No: 1-3 and 17-18) and YTE mutation in Fc.

[0291] BE-647: GPC3 antibody (SEQ ID No: 41-50) was combined with CD137 VHH BGA-9502 (SEQ ID No: 1, 10, 3, 13-14) to produce the construct BE-647 (VL is SEQ ID NO: 23-24, VH is SEQ ID NO: 31-32).

[0292] BE-621: The construct BE-621 (VL is SEQ ID NO: 23-24, VH is SEQ ID NO: 33-34) is generated by combining GPC3 antibody (SEQ ID No: 41-50), CD137 VHH BGA-9502 (SEQ ID No: 1, 10, 3, 13-14) and YTE mutation in Fc.

[0293] Example 6. Target binding activity of anti-GPC3xCD137 bispecific antibody The binding kinetics of BE-774 were characterized by SPR assay using a BIAcore™ T-200 (General Life Sciences). In short, anti-κ antibodies were immobilized on an activated CM5 biosensor chip (catalog number: BR100839, General Life Sciences). BE-774 was allowed to flow across the chip surface and captured by the anti-κ antibody. Then, serially diluted solutions (6.0 nM to 2150 nM) of human CD137 ECD-mIgG2a or huGPC3-His were allowed to flow across the chip surface, and changes in surface plasmon resonance signal were analyzed using a one-to-one Langmuir binding model (BIA assessment software, General Life Sciences) to calculate the association rate (kJ / g). on ) and dissociation rate (k off The equilibrium dissociation constant ( K D ) Calculated as a ratio k off / k onThe results showed that BE-774 exhibited binding to both huCD137 and huGPC3, as shown in Table 8 below. To evaluate the binding activity of the BE-774 bispecific antibody to native huCD137 on live cells, Hut78 cells were transfected to overexpress human CD137. Live cells expressing Hut78 / huCD137 were seeded in 96-well plates and incubated with serially diluted BE-774. Goat anti-human IgG was used as a secondary antibody to detect antibody binding to the cell surface. EC50 showed dose-dependent binding to native human CD137. 50 The values ​​were determined by fitting the dose-response data to a four-parameter logistic model using GraphPadPrism™. Figure 4A As shown, BE-774 exhibited dose-responsive specific binding to native CD137 on live cells, with an EC50 of 0.7 nM. To assess binding to huGPC3 on live cells, HepG2 cells expressing huGPC3 were seeded in 96-well plates and incubated with serially diluted BE-774. Goat anti-human IgG was used as a secondary antibody to detect antibody binding to the cell surface. The dose-dependent binding EC50 to native human GPC3 was... 50 The values ​​were determined by fitting the dose-response data to a four-parameter logistic model using GraphPad Prism™. Figure 4B As shown, BE-774 exhibits specific binding to native GPC3 on living cells in a dose-responsive manner, with an EC50 of 0.9 nM.

[0294] Table 8. SPR affinity of BE-774 with huGPC3 and huCD137

[0295] Example 7. Anti-GPC3xCD137-induced T cell activation co-cultured with GPC3-positive tumor cells The functional activity of the anti-GPC3xCD137 bispecific antibody was evaluated in an in vitro co-culture assay using human peripheral blood mononuclear cells (PBMCs) and a hepatocellular carcinoma (HCC) cell line expressing OS8. OS8 is a single-chain variable fragment (scFv) of the anti-human CD3 antibody OKT3, fused to the C-terminal domain (113-220 aa) of mouse CD8α, which includes a hinge domain, a transmembrane domain, and a cytoplasmic domain. When expressed on target cells, OS8 can provide a signal for T cell activation (Fig. 5A). GPC3-overexpressing HepG2 cells were selected to assess the functional activity of the GPC3xCD137 bispecific antibody, while SK-HEP-1 (GPC3-negative) was used as a negative control cell line.

[0296] Frozen human PBMCs (Auscelles) were thawed in RPMI 1640 medium and incubated overnight at 37°C. Target cells expressing OS8 were seeded into 384-well plates and allowed to attach for 16 hours. The next day, PBMCs were added to the 384-well plates at an effector cell to target cell ratio (E:T) of 2:1. The co-cultured cells were then treated with serially diluted BE-774 or BE-653 at 37°C for 48 hours. The culture supernatant was collected for subsequent measurement of IFN-γ and IL-2 concentrations using the TR-FRET-based method as described in the manufacturer's manual (Cisbio) (Degorce, François, et al. Current chemicalgenomics. [Contemporary Chemical Genomics] 2009, 3: 22). The results showed that anti-GPC3xCD137 bispecific antibodies (including BE-774 and BE-653) induced dose-dependent cytokine release in PBMCs co-cultured with HepG2 cells but not GPC3-negative cells (Figure 5B). PBMCs from both donors were tested, and the results are shown in Figure 5B.

[0297] Example 8. Anti-GPC3xCD137 enhances T-cell killing activity against GPC3-positive tumor cells. The ability of the anti-GPC3xCD137 bispecific antibody to induce T-cell killing activity was assessed by impedance measurement using an xCELLigence™ RTCA MP instrument (Agilent Technologies). Frozen human PBMCs (Auscells) were thawed in RPMI 1640 medium and incubated overnight at 37°C. Target cells were seeded into 96-well E-plates (Agilent Technologies) and allowed to adhere for 16 hours. The next day, PBMCs were added to the 96-well E-plates at an effector cell to target cell ratio (E:T) of 5:1. The co-cultured cells were then treated with a combination of serially diluted BE-774 or BE-653 and an EpCAM / CD3 bispecific T-cell conjugate (BiTE), which provides a signal for T-cell activation (Figure 6A). The experiment was allowed to proceed for 4 days, and electrode impedance was measured using live adherent target cells. The results showed that both BE-774 and BE-653 dose-dependently enhanced the killing activity of T cells against GPC3-expressing HepG2 cells, but did not enhance the killing activity against GPC3-negative SK-OV-3 cells (Figure 6B).

[0298] Example 9. Pharmacokinetics of BE-774 and BE-933 in cynomolgus monkeys Blood samples were collected from cynomolgus monkeys at 0, 0.00694 (10 min), 0.0417 (1 h), 0.167 (4 h), 0.333 (8 h), 1, 4, 7, 10, 14, 21, and 28 days after intravenous administration of 5 mg / kg of BE-774 or BE-933, followed by centrifugation (4 °C, 3500 × g, 2 min) to separate serum. The concentrations of BE-774 or BE-933 were measured using an in-house developed ELISA. Briefly, labeled GPC3 protein (catalog number: C414, Novoprotein, China) was used as the capture reagent, and biotin-labeled CD137 antigen (catalog number: 41B-H82E6, ACRO, China) was used as the assay reagent for BE-774 or BE-933. The pharmacokinetic (PK) characteristics of BE-774 and BE-933 at the 5 mg / kg dose level were... Figure 7 The pharmacokinetic parameters of BE-774 and BE-933 at a dose level of 5 mg / kg are shown in Table 9. Compared with BE-933, BE-774, which has a YTE mutation in Fc, shows significantly improved pharmacokinetic characteristics.

[0299] Table 9. PK parameters of BE-774 and BE-933 in cynomolgus monkeys

[0300] Example 10. Pharmacokinetics of BE-774 and BE-933 in the hFcRn mouse model Blood samples were collected in hFcRn mice at 0, 0.0833 (2 h), 1, 3, 7, 10, 14, 21, and 28 days after intravenous administration of 3 mg / kg of BE-774 or BE-933, followed by centrifugation (4 °C, 3500 × g, 2 min) to separate serum. The concentrations of BE-774 or BE-933 were measured using an in-house developed ELISA. Briefly, labeled GPC3 protein (catalog number: C414, Nearshore Protein Technology Co., Ltd., China) was used as the capture reagent, and biotin-labeled CD137 antigen (catalog number: 41B-H82E6, ACRO, China) was used as the detection reagent for BE-774 or BE-933 measurement. The pharmacokinetic (PK) characteristics of BE-774 and BE-933 at the 3 mg / kg dose level were... Figure 8 The pharmacokinetic parameters of BE-774 and BE-933 at a dose level of 3 mg / kg are shown in Table 10. BE-933 exhibits favorable pharmacokinetic characteristics in hFcRn mice, and BE-774 with the YTE mutation in Fc shows significantly improved pharmacokinetic characteristics compared to BE-933.

[0301] Table 10. PK parameters of BE-774 and BE-933 in hFcRn mice

[0302] Example 11. BE-915 combined with CD137 and GPC3 SPR Antibody binding kinetics were measured using surface plasmon resonance (SPR). The association rate constants (K0) between the antibody and recombinant CD137 and GPC3 proteins were measured using SPR. a ) and dissociation rate constant (K d Then, the affinity constant (K) was determined. D The results showed that BE-915 had good binding affinity for both human CD137 and human GPC3 (Tables 11 and 12).

[0303] To test the binding specificity of BE-915 to CD137 from different species, SPR binding studies were conducted using human CD137 (catalog number: 41B-H5256, Acrobio, China) and cynomolgus monkey CD137 ECD (SEQ ID NO: 102) as bait proteins. BE-915 showed a high binding affinity to human CD137, with its K... D The value is approximately 3.6 nM, which is similar to the binding affinity for CD137 in cynomolgus monkeys (K). D The concentration is approximately 4.4 nM, as shown in Table 11. This also indicates that the anti-CD137 antibody BGA-2524 used in BE-915 has a high binding affinity for human CD137 and cynomolgus monkey CD137.

[0304] To test the binding specificity of BE-915 to GPC3 from different species, SPR binding studies were conducted using human GPC3 (catalog number: 10088-H08H, Sino Biological, China) and cynomolgus monkey GPC3 (catalog number: GP3-C5225, Acrobio, China) as bait proteins. BE-915 showed a binding affinity (K0) to human GPC3. D (approximately 2.4 nM) and binding affinity to GPC3 in cynomolgus monkeys (K). D (approximately 0.53 nM), as shown in Table 12.

[0305] Table 11. SPR binding of BE-915 to human and cynomolgus monkey CD137

[0306] Table 12. BE-915 binding to SPR in human and cynomolgus monkey GPC3

[0307] Example 12. Binding of BE-915 with natural human CD137 and natural human GPC3 To verify the binding of BE-915 to expressed human CD137 and human GPC3 on cells, HuT78 cells overexpressing human CD137 (HuT78 / CD137) and HepG2 cells expressing human GPC3 were used to assess the binding of BE-915. Fluorescence activated cell sorting (FACS) showed that BE-915 had strong binding activity to human CD137 and human GPC3 in a dose-dependent manner, with EC50 values ​​of 0.69 nM (Figure 9B) and 1.09 nM (Figure 9A), respectively, as shown in Figure 9. The isotype control huIgG (catalog number: 02-7102, Thermo, USA) showed no binding activity to either HuT78 / CD137 or HepG2 cells.

[0308] Example 13. Binding specificity of BE-915 with other TNFRSF members To test the binding specificity of BE-915 with other TNFRSF members, ELISA was performed by coating plates with TNFRSF4 / OX40 (catalog number: OXO-H5252, Acrobio, China), TNFRSF7 / CD27 (catalog number: CD7-H5257, Acrobio, China), TNFRSF14 / HVEM (catalog number: CD7-H522b, Acrobio, China), TNFRSF8 / CD30 (catalog number: HVM-H5258, Acrobio, China) and human CD137 (catalog number: 41B-H5256, Acrobio, China), and then binding them with BE-915. Figure 10 As shown, BE-915 specifically binds to human CD137, but not to other TNFRSF members. This also demonstrates that the anti-CD137 antibody BGA-2524 used in BE-915 specifically binds to human CD137.

[0309] Example 14. BE-915 competes with human CD137L for binding to human CD137. To accurately assess the blocking effect of BE-915 on CD137-CD137L binding at the cellular level, a cell-based blocking assay using HuT78 / CD137 was established. The competitive blocking effect of CD137L on the interaction between CD137 and BE-915 was measured by detecting the binding of BE-915 (starting at a concentration of 15 μg / mL, followed by 3-fold serial dilutions) to human CD137 expressed on HuT78 in the presence of 10 µg / mL, 1 µg / mL, 0.1 µg / mL, or 0 μg / mL CD137L (catalog number: 41L-H52D4, Acrobio, China). Binding signal was detected using the secondary antibody anti-hFc 647 (catalog number: 109-605-098, Jackson, USA). huIgG (catalog number: 02-7102, Thermo Fisher Scientific, USA) was used as an isotype control (Figure 11A). The binding of BE-915 to CD137 decreased with increasing CD137L concentration (Figure 11A). The competitive blocking of the CD137-CD137L interaction by BE-915 was measured by detecting the binding of CD137L (starting at 15 μg / mL, followed by 3-fold serial dilutions) to CD137 expressed on HuT78 in the presence of BE-915 at concentrations of 1 µg / mL, 0.1 µg / mL, or 0 μg / mL. Binding signals were detected using the secondary antibody anti-his 647 (catalog number: A01802, Genscript, China). huIgG (catalog number: 02-7102, Thermo Fisher Scientific, USA) was used as an isotype control (Figure 11B). The binding of CD137L to CD137 decreased with increasing BE-915 concentration (Figure 11B). These data indicate that BE-915 and CD137L cross-competitively bind to human CD137. This also indicates that the anti-CD137 antibody BGA-2524 used in BE-915 cross-competitively binds to human CD137 with CD137L.

[0310] Example 15. BE-915-induced T cell activation co-cultured with GPC3-positive tumor cells The functional activity of the GPC3 x CD137 bispecific antibody BE-915 was evaluated in an in vitro co-culture assay using human peripheral blood mononuclear cells (PBMCs) and OS8-expressing hepatocellular carcinoma (HCC) cell lines (Figure 12A). Three HCC cell lines—HepG2, Huh7, and Hep3B—with high to low GPC3 expression were selected based on FACS analysis (Figure 12B) to assess the effect of GPC3 levels on the functional activity of BE-915. SK-HEP-1, which does not express GPC3, was used as a negative control cell line.

[0311] Frozen human PBMCs (OriBiotech) were thawed in RPMI 1640 medium and incubated overnight at 37°C. Target cells expressing OS8 were seeded into 384-well plates and allowed to attach for 16 hours. The next day, PBMCs were added to 384-well plates at an effector cell to target cell ratio (E:T) of 2:1. The co-cultured cells were then treated with serially diluted BE-915 at 37°C for 48 hours. The culture supernatant was collected for subsequent measurement of IFN-γ and IL-2 concentrations using a TR-FRET-based method as described in the manufacturer's manual (Cisbio). (Degorce, François, et al. Current Chemical Genomics. 2009, 3: 22). The results showed that BE-915 induced dose-dependent cytokine release in PBMCs from two independent donors co-cultured with GPC3-expressing cells, but not GPC3-negative cells (Figure 12C).

[0312] Example 16. Enhanced PBMC-based cell killing by BE-915 and GPC3 positive tumor cell culture. In co-culture experiments, the BE-915-regulated T-cell cytotoxic activity was assessed by impedance measurement using an xCELLigence RTCA MP instrument (Agilent Technologies). Frozen human PBMCs (OriBiotech) were thawed in RPMI 1640 medium and incubated overnight at 37°C. Target cells were seeded into 96-well E-plates (Agilent Technologies) and allowed to attach for 16 hours. The next day, PBMCs were added to the 96-well E-plates at an effector cell to target cell ratio (E:T) of 5:1. The co-cultured cells were then treated with a combination of serially diluted BE-915 and EpCAM / CD3 bispecific T-cell conjugate (BiTE), which provides a signal for T-cell activation (Fig. 13A). Three HCC cell lines, HepG2, Huh7, and Hep3B, with high to low GPC3 expression (Fig. 13B), were selected based on FACS analysis to assess the effect of GPC3 levels on the functional activity of BE-915. SK-OV-3 cells that do not express GPC3 were used as a negative control cell line.

[0313] The experiment was allowed to run for 4 days, during which electrode impedance was measured using live adherent target cells. Consistent with cytokine production assays, BE-915 dose-dependently enhanced the cytotoxic activity of T cells against GPC3-expressing cells, but did not enhance the cytotoxic activity against GPC3-negative cells (Figure 13C). PBMCs from two donors were used in this experiment.

[0314] Example 17. Pharmacokinetic characteristics of BE-915 in cynomolgus monkeys Blood samples were collected from cynomolgus monkeys at 0, 0.167 h, 1 h, 4 h, 8 h, 1, 3, 6, 9, 13, 20, and 27 days after intravenous infusion of 5 mg / kg BE-915, followed by centrifugation (4 °C, 3000 × g, 15 min) to separate serum. BE-915 concentrations were measured using an internally developed ELISA ligand binding method. Briefly, labeled GPC3 antigen (catalog number: C414, Nearshore Protein Technology Co., Ltd., China) was used as the capture reagent, and biotin-labeled CD137 antigen (catalog number: 41B-H82E6, ACRO, China) was used as the detection reagent for BE-915. The obtained pharmacokinetic characteristics and parameters were obtained at [dates to be inserted here]. Figure 14 As shown in Table 13. In the 5 mg / kg dosing group, BE-915 was below the lower limit of quantitation (0.0391 μg / mL) on day 13 post-dosing. Antidrug antibodies (ADA) were detected in serum from day 9 in the 5 mg / kg dosing group, indicating a potential impact on the pharmacokinetic profile. PK parameters were calculated after the concentration values ​​at the removal time points. The clearance of BE-915 was 10.4 mL / day / kg, and the antibody-like half-life was 3.2 days.

[0315] Table 13. iv Pharmacokinetic parameters of BE-915 in cynomolgus monkeys after infusion

[0316] Note: A non-compartmental model was used to calculate pharmacokinetic parameters. Example 18. Efficacy of BE-915 monotherapy in the humanized CD137 knock-in mouse MC38 / hGPC3 model. The in vivo efficacy of BE-915 was examined in a humanized CD137 knock-in mouse MC38 / hGPC3 colorectal cancer model. MC38 / hGPC3 cells were subcutaneously implanted into the right abdomen of recipient mice. Seven days after cell inoculation, mice were randomly assigned to four groups based on tumor volume. BE-915 was administered intraperitoneally on day 1, once weekly for 18 days. BE-915 (0.1, 0.5, and 3.0 mg / kg, once weekly) effectively inhibited tumor growth. Tumor volume was significantly reduced at the study endpoint (D18). Furthermore, on day 18, the tumor-free rates in the 0.1, 0.5, and 3.0 mg / kg groups were 0%, 0%, and 10%, respectively. Figure 15 (See Table 14). Throughout the study, no treatment group had a significant effect on animal body weight.

[0317] Table 14. Efficacy of BE-915 in the MC38 / hGPC3 syngeneic tumor model of humanized CD137 knock-in mice

[0318] Abbreviations: hGPC3, human phosphatidylinositol proteoglycan 3; n, number of animals; NA, not applicable; QW, once a week; SEM, standard deviation of the mean; TGI, tumor growth inhibition.

[0319] Note: The TGI rate is calculated using the following formula: %TGI = [1 - (Tt of treatment - T0 of treatment) / (Tt of medium - T0 of medium)] × 100%. Tt of treatment = mean tumor volume in the treatment group on day t; T0 of treatment = mean tumor volume in the treatment group on day 0; Tt of medium = mean tumor volume in the medium group on day t; and T0 of medium = mean tumor volume in the medium group on day 0.

[0320] Example 19. Efficacy of the combination of BE-915 and anti-PD-1 antibody in the humanized 4-1BB knock-in mouse LL / 2 / hGPC3 model. The antitumor activity of the combination of BE-915 and anti-mouse PD-1 antibody was investigated in a humanized CD137 knock-in mouse LL / 2 / hGPC3 syngeneic model (lung cancer). LL / 2 / hGPC3 cells were implanted into mice. Seven days after cell inoculation, mice were randomly divided into four groups based on tumor volume. Mice treated with the combination of BE-915 (10.0 mg / kg, once weekly) and anti-mouse PD-1 antibody (10.0 mg / kg, once weekly) showed synergistic tumor growth inhibition. On day 13, the tumor growth inhibition rate in the combination group was 74.7%, significantly higher than that in the groups treated with BE-915 alone (34.1%) or anti-PD-1 alone (38.0%). Figure 16 (See Table 15). No significant effect on animal body weight was observed in any treatment group throughout the study period.

[0321] Table 15. Antitumor effects of BE-915 and anti-mouse PD-1 antibody in the LL / 2 / hGPC3 syngeneic model of humanized CD137 knock-in mice.

[0322] Abbreviations: hGPC3, human phosphatidylinositol proteoglycan 3; n, number of animals; NA, not applicable; QW, once a week; SEM, standard deviation of the mean; TGI, tumor growth inhibition.

[0323] Note: The TGI rate is calculated using the following formula: %TGI = [1 - (Tt of treatment - T0 of treatment) / (Tt of medium - T0 of medium)] × 100%. Tt of treatment = mean tumor volume of the treatment group on day t; T0 of treatment = mean tumor volume of the treatment group on day 0; Tt of medium = mean tumor volume of the medium group on day t; and T0 of medium = mean tumor volume of the medium group on day 0.

[0324] Example 20. Biophysical properties of BE-915 The biophysical properties of BE-774 (using camel CD137 VHH BGA-9612) and BE-915 (using humanized CD137VHH BGA-2524) were tested. The tested biophysical properties included melting temperature, aggregation temperature, hydrophobicity by HIC-HPLC, and self-association tendency by AC-SINS (see detailed description below). BE-915 exhibited the best thermal stability by Tm and Tagg, and good colloidal stability by AC-SINS. As shown in Table 16, BE-915 showed overall biophysical properties comparable to BE-774. This also indicates that the humanized CD137VHH BGA-2524 used in BE-915 has comparable overall biophysical properties to the camel CD137 VHH BGA-9612 used in BE-774.

[0325] Table 16. Summary of biophysical properties of BE-915 and BE-774

[0326] Melting temperature (Tm) and aggregation temperature (Tagg) (°C) were determined by UNCLE™ (Unchained Labs, Pleasanton, CA) (an instrument that simultaneously measures intrinsic fluorescence and static light scattering). During the measurement, 9 μL of protein sample in PBS buffer (1 mg / mL) was loaded into a cuvette; the sample was held at 20°C for 120 s, and then heated to 95°C at a rate of 0.3°C / min. After excitation at 266 nm, fluorescence and static light scattering (at 266 nm) were collected.

[0327] To determine the hydrophobicity of a given antibody using an HPLC system, 50 μg of sample (1 mg / mL) was diluted with mobile phase A (1.5 M ammonium sulfate, 50 mM sodium phosphate, pH 7.0) prior to analysis to achieve a final ammonium sulfate concentration of approximately 1 M. A linear gradient was performed using a MABPac HIC-10 column with mobile phases A and B (50 mM sodium phosphate, pH 7.0) at a flow rate of 0.5 mg / min over 29 minutes. Peak retention time was monitored at A280 absorbance.

[0328] AC-SINS assays measure protein self-interactions by capturing antibodies on the surface of a gold colloid that exhibits surface resonance oscillations at visible light frequencies. When the immobilized antibodies self-interact, the colloid aggregates, altering its oscillation frequency to absorb longer wavelengths. Gold nanoparticles were incubated with an 80 / 20 (v / v) mixture of capturing and non-capturing antibodies. The coated gold nanoparticles were then centrifuged and resuspended in PBS. Samples were diluted to 0.05 mg / ml (to conjugation buffer), and 45 μl of each dilution was loaded into a 384-well plate. Then, 5 μl of the previously prepared gold nanoparticles were added to each well of a plate containing both mAb and buffer controls. The plate was then capped with an aluminum cap and incubated at room temperature for 2 hours, followed by rapid centrifugation at 3000 rpm before reading the absorption spectrum of each well from 450 to 650 nm using a plate reader. The spectra of each sample were recorded, and the redshift of the maximum absorption peak compared to the buffer was analyzed. The redshift and its intensity indicate the self-interaction tendency of the tested mAb sample.

[0329] The amino acid and DNA sequences of the anti-CD137 VHH and anti-GPC3xCD137 bispecific antibodies are shown in Table 17 below.

[0330] Table 17. Amino acid and DNA sequences of anti-CD137 VHH and anti-GPC3xCD137 bispecific antibodies

[0331] Example 21. Structural and functional CD137 epitope plotting To better understand how the anti-CD137 single-domain antibody arm can have a high affinity for CD137 and is a potent agonist of the CD137 / CD137L interaction, the crystal structure of the VHH (BGA-2524) CD137 complex was determined.

[0332] A. Expression, purification, and crystallization of CD137 and VHH (BGA-2524) The extracellular domain of human CD137 containing portions of the three CRDs (CRD1-3; SEQ ID NO: 35 (human CD137-full length) amino acid 24-105) was expressed in HEK293G cells. The cDNA encoding CD137 was cloned into a pMAX vector with an N-terminal secretion sequence and a C-terminal TEV cleavage site followed by an Fc tag. Culture supernatant containing the secreted CD137-Fc fusion protein was mixed with Mab Select Sure. TMResin (GE Healthcare LifeSciences) was mixed at 4°C for 3 hours. Proteins were washed with a buffer containing 20 mM Tris-HCl pH 8.0 and 150 mM NaCl, followed by elution with 50 mM acetic acid (pH adjusted to 3.5 with 5 M NaOH), and finally neutralized with 1 / 10 CV 1.0 M Tris-HCl pH 8.0. The eluted proteins were mixed with TEV protease (10:1 molar ratio) and dialyzed overnight at 4°C with buffer (20 mM Tris-HCl, pH 8.0, 100 mM NaCl). The mixture was loaded onto a Ni-NTA column (Qiagen) and Mab Select Sure. TM The resin was coated to remove TEV protease and Fc tags, and then HiLoad 16 / 600 Superdex was used. TM The eluent was further purified by size exclusion chromatography in buffer (20 mM Tris pH 8.0, 100 mM NaCl) using a 75 pg column (GE Healthcare Life Sciences).

[0333] The DNA sequence encoding VHH (BGA-2524) was cloned into the PET21a vector with an N-terminal HIS-MBP tag followed by a TEV protease site. Protein expression in Shuffle T7 cells was induced for 16 h at 18°C ​​with 1 mM IPTG at an OD600 of 0.6–1.0. Cells were harvested by centrifugation at 7,000 g for 10 min. The cell pellet was resuspended in lysis buffer (50 mM Tris-HCl pH 8.0, 300 mM NaCl) and lysed by sonication on ice. The lysates were then centrifuged at 48,000 g for 30 min at 4°C. The supernatant was mixed with Talon resin and processed in batches at 4°C for 3 h. The resin was washed with lysis buffer containing 5 mM imidazole, and proteins were eluted with lysis buffer containing an additional 100 mM imidazole. The eluent was mixed with TEV protease (10:1 molar ratio) and dialyzed overnight at 4°C with buffer (20 mM Tris-HCl, pH 8.0, 100 mM NaCl). The mixture was loaded onto a Talon column to remove the TEV protease and HIS-MBP tag, and then analyzed using a HiLoad16 / 600 Superdex column. TM The eluent was further purified by size exclusion chromatography in buffer (20 mM Tris pH 8.0, 100 mM NaCl) using a 75 pg column (GE Healthcare Life Sciences).

[0334] Excess purified CD137 was mixed with purified VHH (BGA-2524) (1.2:1 molar ratio) to generate the CD137 / VHH (BGA-2524) complex. The complex was then further purified by gel filtration in buffer (20 mM Tris pH 8.0, 100 mM NaCl) using a Superdex™ 75 Increase 10 / 300 column (GE Healthcare Life Sciences). The CD137 / VHH (BGA-2524) complex (10 mg / ml) was crystallized in 18% PEG 4000, 0.1 M Tris pH 8.7, and 0.2 M Li₂SO₄. The crystals, cryoprotected with 20% PEG 4000, 0.1 M Tris pH 8.7, 0.2 M Li₂SO₄, and 10% glycerol, were rapidly frozen in liquid nitrogen. X-ray diffraction data were collected at beamline BL02U1 of the Shanghai Synchrotron Radiation Facility (Shanghai, China).

[0335] B. Data Collection and Structure Analysis X-ray diffraction data were collected at beamline BL02U1 of the Shanghai Synchrotron Radiation Facility (Shanghai, China) under cryogenic cooling conditions of 100 Kelvin. Diffraction images were processed using integrated data processing software XDS (Kabsch, W., Xds. Acta Crystallogr D Biol Crystallogr [Acta Crystallographica D: Structural Biology], 2010. 66(Pt 2): 125-32). The structures of human CD137 (PDB: 6MGP) and the internal VHH model were used as search models. Initial solutions were found using the molecular substitution procedure PHASER (McCoy, AJ, et al., Phaser crystallographic software. [Phaser crystallography software] J Appl Crystallogr [Journal of Applied Crystallography], 2007. 40(Pt 4): 658-674). The model was then manually iteratively built using the program COOT (Emsley, P. and K. Cowtan, Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr, 2004. 60(Pt 12 Pt 1): 2126-32), and refined using PHENIX (Adams, PD, et al., PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr D Biol Crystallogr, 2010. 66(Pt 2): 213-21). The final model was refined to acceptable R and R free values, as well as Ramachandran conformation statistics (calculated via Molprobability). Data processing and refinement statistics are available in Table 18.

[0336] Table 18. Data Collection and Detailed Statistics

[0337] a The value in parentheses is the value of the highest resolution shell.

[0338] b Calculated based on approximately 5% of reflection left during the refinement process. c RMSD, Root Mean Square Deviation C. Structure of VHH (BGA-2524) bound to human CD137 dimer VHH (BGA-2524) complexed with CD137 crystallized in space group P21, with both complexes in an asymmetric unit, and diffracted to 1.71 Å. The structure of VHH (BGA-2524) bound to human CD137 indicates that VHH (BGA-2524) and CD137L are partially spatially connected. Figure 17 The embedded surface area between VHH (BGA-2524) and CD137 is approximately 863 Å. 2 VHH (BGA-2524) interactions cluster around the CD137 CRD2 domain. VHH (BGA-2524) primarily binds to the side surfaces of the CD137 CRD2 domain via CDR residues. All CDRs in VHH (BGA-2524) participate in CD137 dimer binding, with CDR3 contributing the greatest potential. CDR1 and CDR3 bind both monomers of the CD137 dimer, while CDR2 binds only one monomer. The curved CDR3 ring always covers the hydrophobic patches of the VHH framework, which may lead to better biophysical properties. Figure 18VHH(BGA-2524) CDR1 Tyr32 contacts one monomer of the CD137 dimer at residue Gly98, while CDR1 Asn31 and Ala33 contact the other monomer of the CD137 dimer at residues Ile64 and Gln67. VHH(BGA-2524) CDR2 Trp52, Ser53, Tyr55, and His57 contact only one monomer of the CD137 dimer at residues Asp38, Pro49, Pro50, Asn51, and Ile64. VHH(BGA-2524) CDR3 residues Leu98, Thr104, Thr106, and Tyr109 contact one monomer of the CD137 dimer at residues Ser55, Ala56, Arg75, Glu85, and Ala97, while CDR3 residues Leu98, Lys99, Tyr100, and Pro101 contact the other monomer of the CD137 dimer at residues Phe36, Pro49, Thr61, Cys62, Asp63, and Ile64. VHH(BGA-2524) interacts with CD137 using a combination of hydrogen bonds and salt bridges, as well as hydrophobic interactions. For example, VHH(BGA-2524) CDR2 residue His57 forms two salt bridges with CD137 residue Asp38. The CDR3 residue Lys99 of VHH(BGA-2524) forms two salt bridges with the CD137 residue Asp63. The residues Tyr32, Ser53, His57, Leu98, Lys99, Pro101, and Thr106 of VHH(BGA-2524) form hydrogen bonds with the CD137 residues Gly98, Asn51, Asp38, Ile64, Asp63, Thr61, and Ser55, respectively (i.e., Tyr32-Gly98, Ser53-Asn51, His57-Asp38, Leu98-Ile64, Lys99-Asp63, Pro101-Thr61, and Thr106-Ser55). The residue Trp52 of VHH(BGA-2524) forms two hydrogen bonds with the CD137 residues Pro50 and Asn51. The VHH (BGA-2524) residue Tyr109 forms two hydrogen bonds with CD137 residues Arg75 and Glu85. The VHH (BGA-2524) residue Tyr100 forms two hydrogen bonds with CD137 residue Cys62. Figure 19 ).

[0339] Based on the crystal structure of the VHH(BGA-2524) / CD137 complex, the CD137 residues contacted by VHH(BGA-2524) (i.e., the epitope residues of CD137 bound by VHH(BGA-2524)) and the VHH(BGA-2524) residues contacted by CD137 (i.e., the para-site residues of VHH(BGA-2524) contacted by CD137) were determined. Tables 19 and 20 below show the CD137 and VHH(BGA-2524) residues in contact with them, which, as assessed using a contact distance strictness of 3.7 Å, represent the points where van der Waals (nonpolar) interactions are strongest.

[0340] Table 19. Epitope residues of CD137 and their corresponding para residues of VHH (BGA-2524)

[0341] Table 20. Para residues of VHH (BGA-2524) and their corresponding CD137 epitope residues

Claims

1. An antibody or antigen-binding fragment thereof that specifically binds to human CD137, wherein... (1) The antibody or its antigen-binding fragment specifically binds to the following epitopes, which contain or are composed of amino acid residues Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64, Gln67 of human CD137 (SEQ ID NO:35); (2) The antibody or its antigen-binding fragment specifically binds to the following epitope, which comprises, or is composed of, the amino acid residues Ser55, Ala56, Arg75, Glu85, Ala97, and Gly98 of human CD137 (SEQ ID NO: 35); or (3) The antibody or its antigen-binding fragment specifically binds to a human CD137 dimer, which comprises a first human CD137 monomer and a second human CD137 monomer, or is composed of the latter; wherein the antibody or its antigen-binding fragment specifically binds to an epitope of the first human CD137 monomer (SEQ ID NO:35), which comprises amino acid residues Phe36, Asp38, Pro49, Pro50, Asn51, Thr61, Cys62, Asp63, Ile64, Gln67, or is composed of the latter; and the antibody or its antigen-binding fragment specifically binds to an epitope of the second human CD137 monomer (SEQ ID NO:35), which comprises amino acid residues Ser55, Ala56, Arg75, Glu85, Ala97, and Gly98, or is composed of the latter; and / or the antibody or its antigen-binding fragment binds to the human CD137 dimer and promotes human CD137 aggregation.

2. An antibody or antigen-binding fragment thereof that specifically binds to human CD137, the antibody or antigen-binding fragment comprising: (i) a heavy chain variable region (VH) comprising (a) HCDR1 (heavy chain complementarity determination region 1) of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 2, and (c) HCDR3 of SEQ ID NO: 3; or (ii) a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 1, (b) HCDR2 of SEQ ID NO: 10, and (c) HCDR3 of SEQ ID NO:

3.

3. The antibody or antigen-binding fragment thereof according to any one of claims 1-2, wherein the antibody or antigen-binding fragment thereof comprises: (i) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 17; (ii) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 11; (iii) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO: 13; (iv) A heavy chain variable region (VH) comprising at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the same amino acid sequence as SEQ ID NO: 15; or (v) Heavy chain variable region (VH) containing at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the same amino acid sequence as SEQ ID NO:

4.

4. The antibody or antigen-binding fragment thereof according to claim 3, wherein one, two, three, four, five, six, seven, eight, nine or ten amino acids have been inserted, deleted or substituted in SEQ ID NO: 17, SEQ ID NO: 11, SEQ ID NO: 13, SEQ ID NO: 15 or SEQ ID NO:

4.

5. The antibody or antigen-binding fragment thereof according to any one of the preceding claims, wherein the antibody or antigen-binding fragment comprises: (i) Heavy chain variable region (VH) containing SEQ ID NO: 17; (ii) Heavy chain variable region (VH) containing SEQ ID NO: 11; (iii) Heavy chain variable region (VH) containing SEQ ID NO: 13; (iv) A heavy chain variable region (VH) containing SEQ ID NO: 15; or (v) Heavy chain variable region (VH) containing SEQ ID NO:

4.

6. The antibody or antigen-binding fragment thereof according to any one of the preceding claims, wherein the antibody or antigen-binding fragment thereof is a monoclonal antibody, a chimeric antibody, a humanized antibody, a human-engineered antibody, a single-chain antibody (scFv), a Fab fragment, a Fab' fragment, or an F(ab')2 fragment.

7. The antibody or antigen-binding fragment thereof according to any one of the preceding claims, wherein the antibody or antigen-binding fragment thereof comprises a heavy chain constant region of IgG1, IgG2, IgG3 or IgG4 subclass and / or a light chain constant region of type κ or λ.

8. The antibody or antigen-binding fragment thereof according to any one of the preceding claims, wherein the antibody or antigen-binding fragment thereof has antibody-dependent cytotoxicity (ADCC) or complement-dependent cytotoxicity (CDC).

9. The antibody or antigen-binding fragment thereof according to any one of the preceding claims, wherein the antibody or antigen-binding fragment thereof has reduced glycosylation or no glycosylation or low fucosylation.

10. The antibody or antigen-binding fragment thereof according to any one of the preceding claims, wherein the antibody or antigen-binding fragment thereof comprises an increased bipartite GlcNac structure.

Images