Anti-TSPAN8-anti-CD3 bispecific antibody and anti-TSPAN8 antibody

Abstract

The present invention provides an anti-TSPAN8-anti-CD3 bispecific antibody and an anti-TSPAN8 antibody that can be used for human therapy or prophylaxis. A bispecific antibody that binds to TSPAN8 and CD3, (a) a Fab region of an anti-TSPAN8 antibody consisting of a heavy chain fragment containing the heavy chain variable region of the anti-TSPAN8 antibody and a light chain containing the light chain variable region of the anti-TSPAN8 antibody; (b) an anti-CD3-scFv region comprising the heavy chain variable region and the light chain variable region of an anti-CD3 antibody; and (c) an Fc region consisting of a first Fc polypeptide linked to a heavy chain fragment of the Fab region of (a) and a second Fc polypeptide linked to the anti-CD3-scFv region of (b); The present invention provides a bispecific antibody comprising:

Description

[Technical Field]

[0001] The present invention relates to anti-TSPAN8-anti-CD3 bispecific antibodies and anti-TSPAN8 antibodies useful as active ingredients in pharmaceutical compositions for human therapy. [Background technology]

[0002] Metastatic gastric cancer is a condition in which primary gastric cancer invades deeper than the muscle layer, breaks through the serous membrane that surrounds the stomach wall, and spreads to various organs, such as lymph nodes, the peritoneal cavity, and even the blood and lymph. Peritoneal dissemination is observed in more than half of metastatic gastric cancer patients. Patients in the late stage of the disease are known to experience symptoms such as abdominal swelling, persistent bloating, pain, nausea, shortness of breath, insomnia, and fatigue due to the accumulation of ascites caused by peritoneal dissemination (World J. Gastroenterol., 2016, Vol. 22, pp. 6829-6840; Int. J. Cancer, 2010, Vol. 127, pp. 2209-2221). However, complete cure through surgery is difficult for patients with peritoneal dissemination of gastric cancer, and chemotherapy, the standard treatment for peritoneal dissemination of gastric cancer, is not sufficiently effective. As a result, the 5-year survival rate for these patients is approximately 2%, resulting in a very poor prognosis. Effective treatments for peritoneal dissemination of gastric cancer are desired. Furthermore, peritoneal dissemination is observed in many patients with primary cancers such as ovarian cancer, colorectal cancer, and pancreatic cancer (Int. J. Adv. Res., 2016, Vol. 4, pp. 735-748), and no established treatments have been established for these patients.

[0003] In the development of antibodies for cancer therapy, various methods have been attempted to identify tumor-associated antigens (TAA) that are selectively expressed in cancer cells. One method reported is to immunize animals with cancer cells to generate antibodies that bind to the TAA expressed in cancer cells (Biochem. Biophys. Res. Commun., 2018, Vol. 505, pp. 181e-186; FEBS Open Bio, 2017, Vol. 7, pp. 627-635).

[0004] Tetraspanin-8 (TSPAN8) is a four-transmembrane protein belonging to the tetraspanin family. It contains two extracellular loop regions, the small extracellular loop (SEL) and the large extracellular loop (LEL), and three cytoplasmic domains. It forms molecular clusters with a wide variety of transmembrane and cytoplasmic proteins as scaffolding proteins. TSPAN8 is known to be involved in cell adhesion, cell motility, cell activation, and proliferation. High expression of TSPAN8 has been observed in gastric, pancreatic, colon, and liver cancers, and its elevated expression has been reported to be associated with cancer progression or metastasis (Non-Patent Document 1). Research is currently being conducted on the use of anti-TSPAN8 antibodies for cancer diagnosis and treatment (Patent Documents 1-2 and Non-Patent Documents 1-2).

[0005] Cluster of differentiation 3 (CD3) is a protein that transmits activation signals to T cells by forming a complex with the T cell receptor (TCR) on the surface of T cells. CD3 is a complex consisting of five subunits: gamma (γ), delta (δ), epsilon (ε), zeta (ζ), and eta (η) chains. Each subunit forms three types of dimers: εγ, εδ, and ζζ. CD3 is expressed on both normal and neoplastic T cells and is therefore used as a T cell marker. Furthermore, various applications of bispecific antibodies, including various anti-TAA antibodies and anti-CD3 antibodies, as pharmaceuticals for cancer treatment have been reported (Non-Patent Document 3).

[0006] Bispecific T-cell-recruiting antibodies (bispecific T-cell-recruiting antibodies) consisting of various antibody formats have been reported as an innovative method for achieving selective cytotoxic activity against cancer cells at low antibody concentrations, and the effects of these antibodies on T-cell-mediated immunotherapy are currently being investigated (Non-Patent Document 4). Bispecific T-cell-recruiting antibodies are bispecific antibodies that contain an antibody against a TAA expressed on the surface of cancer cells and an antibody that binds to T cells. Anti-CD3 antibodies are often used as the antibody that binds to T cells. Bispecific T-cell-recruiting antibodies, which are molecules containing anti-TAA and anti-CD3 antibodies, reduce the physical distance between target cancer cells and cytotoxic T lymphocytes (CTLs), activate the CTLs via the anti-CD3 antibodies, and kill the cancer cells through the cytotoxic activity of the CTLs (redirected T cell cytotoxicity; RTCC). The anti-CD3-anti-epithelial cell adhesion molecule (EpCAM) bispecific antibody catumaxomab and the anti-CD3-anti-CD19 (Cluster of Differentiation 19) bispecific antibody blinatumomab have already been confirmed to be clinically effective ( Int. J. Cancer, 2010, Vol. 127, pp. 2209-2221; N. Engl. J. Med., 2017, Vol. 376, pp. 836-847). Research and development of bispecific T cell-recruiting antibodies for various TAAs is currently underway. However, no anti-TSPAN8-anti-CD3 bispecific antibodies are known to date. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] International Publication No. 2012 / 010696 [Patent Document 2] International Publication No. 2015 / 130115 [Non-patent literature]

[0008] [Non-Patent Document 1] Biomolecules, (Switzerland), 2020;10(3):p.383 [Non-patent document 2] Cancers, (Switzerland), 2019;11(2):p.179 [Non-patent document 3] Pharmacology and Therapeutics, (UK), 2018;182:p.161-175 [Non-patent document 4] mAbs, 2017:9(2):p.182-212 Summary of the Invention [Problem to be solved by the invention]

[0009] An object of the present invention is to provide an anti-TSPAN8-anti-CD3 bispecific antibody and an anti-TSPAN8 antibody that can be used for human therapy. [Means for solving the problem]

[0010] Aiming to create a therapeutic agent selective for cancer cells, the present inventors isolated anti-TSPAN8 antibodies 16B11 and 16B12 by immunizing human monoclonal antibody-producing mice with cancer peritoneal dissemination cells isolated from patients to obtain antibodies (Example 1). These antibodies bound more strongly to TSPAN8 expressed in cancer peritoneal dissemination cells than to TSPAN8 expressed in normal cells (Examples 1 to 5). Epitope analysis of 16B11 and 16B12 revealed that the antibodies recognize the region consisting of amino acid positions 126 to 155 of human TSPAN8 as an epitope, and furthermore, that the threonine at position 131 of human TSPAN8 is essential for binding to the antibodies (Example 4). Furthermore, we produced 16B11.1, a fully human antibody in which the Fc region of 16B11 was converted to a human sequence (Example 3), and found that this fully human antibody exhibited cytotoxic activity against 60As6-Luc / GFP cells (Example 6). Furthermore, to enhance the antigen-selective antitumor activity of T cells, an anti-TSPAN8-anti-CD3 bispecific antibody was produced (Example 7), which contained an anti-CD3-scFv region comprising a heavy-chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acids 1 to 121 of SEQ ID NO: 6 and a light-chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acids 1 to 107 of SEQ ID NO: 8, as well as a heavy-chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acids 1 to 125 of SEQ ID NO: 14 and a light-chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acids 146 to 254 of SEQ ID NO: 14. This bispecific antibody bound to TSPAN8 and CD3 (Example 8), exhibited cytotoxic activity against cancer cells expressing TSPAN8 on their cell surface (Examples 9, 10, 12-1, and 12-2), and was confirmed to prolong mouse survival and exert antitumor activity in vivo (Examples 11 and 12-3).

[0011] That is, the present invention relates to the following [1] to

[55] . [1] A bispecific antibody that binds to TSPAN8 and CD3, (a) a Fab region of an anti-TSPAN8 antibody consisting of a heavy chain fragment containing the heavy chain variable region of the anti-TSPAN8 antibody and a light chain containing the light chain variable region of the anti-TSPAN8 antibody; (b) an anti-CD3-scFv region comprising the heavy chain variable region and the light chain variable region of an anti-CD3 antibody; and (c) an Fc region consisting of a first Fc polypeptide linked to a heavy chain fragment of the Fab region of (a) and a second Fc polypeptide linked to the anti-CD3-scFv region of (b); A bispecific antibody comprising: [2] The bispecific antibody according to [1], the heavy chain variable region of the anti-TSPAN8 antibody comprises a CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 6, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 of SEQ ID NO: 6, and a CDR3 consisting of the amino acid sequence from amino acid numbers 99 to 110 of SEQ ID NO: 6, and the light chain variable region of the anti-TSPAN8 antibody comprises a CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 8, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 8, and a CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 8; or the heavy chain variable region of the anti-TSPAN8 antibody comprises a CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 10, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 of SEQ ID NO: 10, and a CDR3 consisting of the amino acid sequence from amino acid numbers 99 to 110 of SEQ ID NO: 10; and the light chain variable region of the anti-TSPAN8 antibody comprises a CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 12, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 12, and a CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 12; Bispecific antibodies. [3] The bispecific antibody according to [1], The heavy chain variable region of the anti-TSPAN8 antibody consists of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 6, and the light chain variable region of the anti-TSPAN8 antibody consists of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 8; or the heavy chain variable region of the anti-TSPAN8 antibody consists of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10, and the light chain variable region of the anti-TSPAN8 antibody consists of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12; Bispecific antibodies. [4] The bispecific antibody according to [1], The Fab region of the anti-TSPAN8 antibody consists of a heavy chain fragment consisting of the amino acid sequence from amino acid numbers 1 to 219 of SEQ ID NO: 6 and a light chain fragment consisting of the amino acid sequence of SEQ ID NO: 8; or the Fab region of the anti-TSPAN8 antibody consists of a heavy chain fragment consisting of the amino acid sequence from amino acid numbers 1 to 219 of SEQ ID NO: 10 and a light chain fragment consisting of the amino acid sequence of SEQ ID NO: 12; Bispecific antibodies. [5] The bispecific antibody according to any one of [1] to [4], wherein the heavy chain variable region of the anti-CD3 antibody comprises CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 101 to 114 of SEQ ID NO: 14, and the light chain variable region of the anti-CD3 antibody comprises CDR1 consisting of the amino acid sequence of amino acids 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 236 to 244 of SEQ ID NO: 14. [6] The bispecific antibody according to any one of [1] to [4], wherein the heavy chain variable region of the anti-CD3 antibody consists of the amino acid sequence from amino acid nos. 1 to 125 of SEQ ID NO: 14, and the light chain variable region of the anti-CD3 antibody consists of the amino acid sequence from amino acid nos. 146 to 254 of SEQ ID NO: 14. [7] The bispecific antibody according to any one of [1] to [4], wherein the anti-CD3-scFv region consists of the amino acid sequence from amino acid number 1 to amino acid number 254 of SEQ ID NO: 14. [8] The bispecific antibody according to [1], The bispecific antibody comprises a heavy chain of an anti-TSPAN8 antibody in which a first Fc polypeptide is linked to a heavy chain fragment of the anti-TSPAN8 antibody, the heavy chain fragment comprising a heavy chain variable region comprising CDR1 consisting of the amino acid sequence from 31 to 35 of SEQ ID NO: 6, CDR2 consisting of the amino acid sequence from 50 to 66 of SEQ ID NO: 6, and CDR3 consisting of the amino acid sequence from 99 to 110 of SEQ ID NO: 6; a light chain of an anti-TSPAN8 antibody comprising a light chain variable region comprising CDR1 consisting of the amino acid sequence from 24 to 34 of SEQ ID NO: 8, CDR2 consisting of the amino acid sequence from 50 to 56 of SEQ ID NO: 8, and CDR3 consisting of the amino acid sequence from 89 to 96 of SEQ ID NO: 8; and a polypeptide in which a second Fc polypeptide is linked to an anti-CD3-scFv region comprising a heavy chain variable region of an anti-CD3 antibody, the heavy chain variable region comprising CDR1 consisting of the amino acid sequence from amino acid number 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence from amino acid number 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence from amino acid number 101 to 114 of SEQ ID NO: 14, and a light chain variable region of an anti-CD3 antibody, the light chain variable region comprising CDR1 consisting of the amino acid sequence from amino acid number 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence from amino acid number 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence from amino acid number 236 to 244 of SEQ ID NO: 14; or The bispecific antibody comprises a heavy chain fragment of an anti-TSPAN8 antibody comprising a heavy chain variable region comprising CDR1 consisting of the amino acid sequence from 31 to 35 of SEQ ID NO: 10, CDR2 consisting of the amino acid sequence from 50 to 66 of SEQ ID NO: 10, and CDR3 consisting of the amino acid sequence from 99 to 110 of SEQ ID NO: 10, and a first Fc polypeptide linked to the heavy chain of an anti-TSPAN8 antibody; and an anti-TSPAN8 antibody comprising a light chain variable region comprising CDR1 consisting of the amino acid sequence from 24 to 34 of SEQ ID NO: 12, CDR2 consisting of the amino acid sequence from 50 to 56 of SEQ ID NO: 12, and CDR3 consisting of the amino acid sequence from 89 to 96 of SEQ ID NO: 12. and a heavy chain variable region of an anti-CD3 antibody comprising CDR1 consisting of the amino acid sequence from 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence from 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence from 101 to 114 of SEQ ID NO: 14, and a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked, the heavy chain variable region of an anti-CD3 antibody comprising CDR1 consisting of the amino acid sequence from 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence from 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence from 236 to 244 of SEQ ID NO: 14, Bispecific antibodies. [9] The bispecific antibody according to [1], The bispecific antibody comprises a heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody, which comprises a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 121 of SEQ ID NO: 6, and a first Fc polypeptide are linked, a light chain of an anti-TSPAN8 antibody in which a light chain variable region consisting of the amino acid sequence of amino acids 1 to 107 of SEQ ID NO: 8, and a polypeptide in which a second Fc polypeptide is linked to an anti-CD3-scFv region, which comprises a heavy chain variable region of an anti-CD3 antibody consisting of the amino acid sequence of amino acids 1 to 125 of SEQ ID NO: 14 and a light chain variable region of an anti-CD3 antibody consisting of the amino acid sequence of amino acids 146 to 254 of SEQ ID NO: 14; or The bispecific antibody comprises a heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody, which comprises a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 121 of SEQ ID NO: 10, and a first Fc polypeptide are linked; a light chain of an anti-TSPAN8 antibody in which a light chain variable region consisting of the amino acid sequence of amino acids 1 to 107 of SEQ ID NO: 12, and a polypeptide in which an anti-CD3-scFv region, which comprises a heavy chain variable region of an anti-CD3 antibody consisting of the amino acid sequence of amino acids 1 to 125 of SEQ ID NO: 14 and a light chain variable region of an anti-CD3 antibody consisting of the amino acid sequence of amino acids 146 to 254 of SEQ ID NO: 14, and a second Fc polypeptide are linked; Bispecific antibodies.

[10] The bispecific antibody according to any one of [1] to [9], comprising an Fc region containing LALA mutations (L234A and L235A (wherein the mutation positions are amino acid positions according to the EU index in the human Igγ1 constant region)).

[11] The bispecific antibody according to any one of [1] to

[10] , comprising an Fc region comprising an N297G mutation (wherein the mutation position is an amino acid position according to the EU index in a human Igγ1 constant region).

[12] The bispecific antibody according to any one of [1] to

[11] , which comprises an Fc region comprising a knobs-into-holes mutation.

[13] The bispecific antibody according to any one of [1] to

[12] , comprising an Fc region comprising a LALA mutation, an N297G mutation, and a knobs-into-holes mutation.

[14] A bispecific antibody according to

[12] or

[13] , wherein the knobs-into-holes mutations are a T366W mutation in one Fc polypeptide forming the Fc region, and T366S, L368A, and Y407V mutations in another Fc polypeptide forming the Fc region (wherein the mutation positions are amino acid positions according to the EU index in the human Igγ1 constant region).

[15] The bispecific antibody according to any one of [1] to

[14] , comprising an Fc region wherein the first Fc polypeptide consists of the amino acid sequence from 235 to 451 of SEQ ID NO: 6, and the second Fc polypeptide consists of the amino acid sequence from 270 to 486 of SEQ ID NO: 14.

[16] The bispecific antibody according to any one of [1] to

[15] , wherein the heavy chain fragment of the anti-TSPAN8 antibody and the first Fc polypeptide are linked via a hinge region, and the anti-CD3-scFv region and the second Fc polypeptide are linked via a hinge region.

[17] The bispecific antibody according to [1], A bispecific antibody comprising a heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody having the amino acid sequence of SEQ ID NO: 6 is linked to a first Fc polypeptide, a light chain of an anti-TSPAN8 antibody in which the heavy chain fragment has the amino acid sequence of SEQ ID NO: 8, and a polypeptide in which an anti-CD3-scFv region having the amino acid sequence of SEQ ID NO: 14 is linked to a second Fc polypeptide.

[18] The bispecific antibody according to any one of [2] to

[17] , which is post-translationally modified.

[19] The bispecific antibody according to

[18] , wherein the post-translational modification is pyroglutamylation of the N-terminus of the heavy chain variable region and / or deletion of a lysine at the C-terminus of the heavy chain.

[20] A polynucleotide selected from the group consisting of the following (a) to (e): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody comprising a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 6 and a first Fc polypeptide are linked; (b) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 8; (c) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody containing a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10 and a first Fc polypeptide are linked; (d) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12; (e) a polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region, which includes a heavy chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acid numbers 1 to 125 of SEQ ID NO: 14 and a light chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acid numbers 146 to 254 of SEQ ID NO: 14, and a second Fc polypeptide are linked.

[21] A polynucleotide selected from the group consisting of the following (a) to (c): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 6 and a first Fc polypeptide are linked; (b) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 8; (c) A polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region consisting of the amino acid sequence of SEQ ID NO: 14 and a second Fc polypeptide are linked together.

[22] An expression vector comprising the polynucleotide according to

[20] or

[21] .

[23] A host cell transformed with the expression vector described in

[22] .

[24] A host cell comprising: a polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody, comprising a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 6, and a first Fc polypeptide are linked; a polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody in which a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 8, and a polynucleotide comprising a base sequence encoding a polypeptide in which an anti-CD3-scFv region, comprising a heavy chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acid numbers 1 to 125 of SEQ ID NO: 14, and a light chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acid numbers 146 to 254 of SEQ ID NO: 14, and a second Fc polypeptide are linked.

[25] A host cell comprising a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody having the amino acid sequence of SEQ ID NO: 6 and a first Fc polypeptide are linked, a polynucleotide comprising a nucleotide sequence encoding the light chain of the anti-TSPAN8 antibody having the amino acid sequence of SEQ ID NO: 8, and a polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region having the amino acid sequence of SEQ ID NO: 14 and a second Fc polypeptide are linked.

[26] A method for producing a bispecific antibody that binds to TSPAN8 and CD3, the method comprising a step of culturing the host cell described in any one of

[23] to

[25] .

[27] A pharmaceutical composition comprising the bispecific antibody according to any one of [1] to

[19] and a pharmaceutically acceptable excipient.

[28] The bispecific antibody according to any one of [1] to

[19] , for use in treating cancer.

[29] The pharmaceutical composition according to

[27] for the treatment of cancer.

[30] A method for treating cancer, comprising the step of administering a therapeutically effective amount of the bispecific antibody according to any one of [1] to

[19] to a subject.

[31] Use of the bispecific antibody according to any one of [1] to

[19] in the manufacture of a pharmaceutical composition for the treatment of cancer.

[32] An anti-TSPAN8 antibody or its antigen-binding fragment that selectively binds to human TSPAN8-expressing cancer cells.

[33] An anti-TSPAN8 antibody or an antigen-binding fragment thereof that binds to at least one amino acid present in the region of human TSPAN8 from amino acid numbers 126 to 155 of SEQ ID NO: 2.

[34] An anti-TSPAN8 antibody or antigen-binding fragment thereof described in

[33] , which binds to at least amino acid number 131 of sequence number 2, which is present in the region of human TSPAN8.

[35] An anti-TSPAN8 antibody or antigen-binding fragment thereof selected from the following (a) and (b): (a) an anti-TSPAN8 antibody or an antigen-binding fragment thereof comprising: a heavy chain variable region comprising CDR1 consisting of the amino acid sequence from 31 to 35 of SEQ ID NO: 4, a CDR2 consisting of the amino acid sequence from 50 to 66 of SEQ ID NO: 4, and a CDR3 consisting of the amino acid sequence from 99 to 110 of SEQ ID NO: 4; and a light chain variable region comprising CDR1 consisting of the amino acid sequence from 24 to 34 of SEQ ID NO: 8, a CDR2 consisting of the amino acid sequence from 50 to 56 of SEQ ID NO: 8, and a CDR3 consisting of the amino acid sequence from 89 to 96 of SEQ ID NO: 8; (b) An anti-TSPAN8 antibody or an antigen-binding fragment thereof comprising: a heavy chain variable region comprising CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 10, CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 of SEQ ID NO: 10, and CDR3 consisting of the amino acid sequence from amino acid numbers 99 to 110 of SEQ ID NO: 10; and a light chain variable region comprising CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 12, CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 12, and CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 12.

[36] The anti-TSPAN8 antibody or antigen-binding fragment thereof according to

[35] , selected from the following (a) and (b): (a) an anti-TSPAN8 antibody or an antigen-binding fragment thereof, comprising a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 121 of SEQ ID NO: 4 and a light chain variable region consisting of the amino acid sequence of amino acids 1 to 107 of SEQ ID NO: 8; (b) An anti-TSPAN8 antibody or an antigen-binding fragment thereof, comprising a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10, and a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12.

[37] The anti-TSPAN8 antibody according to

[35] , selected from the following (a) and (b): (a) an anti-TSPAN8 antibody comprising a heavy chain consisting of the amino acid sequence of SEQ ID NO: 4 and a light chain consisting of the amino acid sequence of SEQ ID NO: 8; (b) An anti-TSPAN8 antibody comprising a heavy chain consisting of the amino acid sequence of SEQ ID NO: 10 and a light chain consisting of the amino acid sequence of SEQ ID NO: 12.

[38] An anti-TSPAN8 antibody or an antigen-binding fragment thereof that competes with the anti-TSPAN8 antibody or antigen-binding fragment thereof described in any one of

[33] to

[37] for binding to human TSPAN8-expressing cancer cells.

[39] An anti-TSPAN8 antibody or an antigen-binding fragment thereof described in any one of

[32] to

[38] , linked to an antibody or an antigen-binding fragment thereof against a surface antigen of T cells or NK cells.

[40] An anti-TSPAN8 antibody or antigen-binding fragment thereof according to

[39] , wherein the T cell surface antigen is CD3.

[41] An anti-TSPAN8 antibody or an antigen-binding fragment thereof according to

[40] , wherein the antigen-binding fragment for a T cell surface antigen is an scFv of an anti-CD3 antibody.

[42] The anti-TSPAN8 antibody or antigen-binding fragment thereof according to any one of

[32] to

[41] , which is post-translationally modified.

[43] An anti-TSPAN8 antibody or antigen-binding fragment thereof according to

[42] , wherein the post-translational modification is pyroglutamylation of the N-terminus of the heavy chain variable region and / or deletion of a lysine at the C-terminus of the heavy chain.

[44] A fusion or complex of an anti-TSPAN8 antibody or its antigen-binding fragment described in any one of

[32] to

[43] , or a cell expressing an anti-TSPAN8 antibody or its antigen-binding fragment described in any one of

[32] to

[43] on the cell surface.

[45] A polynucleotide selected from the group consisting of the following (a) to (d): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 4; (b) a polynucleotide comprising a nucleotide sequence encoding the light chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 8; (c) a polynucleotide comprising a nucleotide sequence encoding the heavy chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10; (d) A polynucleotide comprising a base sequence encoding the light chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12.

[46] A polynucleotide selected from the group consisting of the following (a) to (d): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 4; (b) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 8; (c) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 10; (d) A polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 12.

[47] An expression vector comprising the polynucleotide according to

[45] or

[46] .

[48] ​​A host cell transformed with the expression vector described in

[47] .

[49] A host cell selected from the following (a) or (b): (a) a host cell comprising a polynucleotide comprising a nucleotide sequence encoding the heavy chain variable region of the anti-TSPAN8 antibody described in

[45] and a polynucleotide comprising a nucleotide sequence encoding the light chain variable region of the anti-TSPAN8 antibody described in

[45] ; (b) A host cell comprising a polynucleotide comprising a base sequence encoding the heavy chain of the anti-TSPAN8 antibody described in

[46] and a polynucleotide comprising a base sequence encoding the light chain of the anti-TSPAN8 antibody described in

[46] .

[50] A method for producing an anti-TSPAN8 antibody or an antigen-binding fragment thereof, comprising culturing a host cell of either

[48] or

[49] .

[51] An anti-TSPAN8 antibody or antigen-binding fragment thereof described in any one of

[32] to

[43] , or a fusion, complex, or cell described in

[44] , for use in the treatment of cancer.

[52] A pharmaceutical composition comprising an anti-TSPAN8 antibody or antigen-binding fragment thereof described in any one of

[32] to

[43] , or a fusion, complex or cell described in

[44] , and further comprising a pharmaceutically acceptable excipient.

[53] The pharmaceutical composition according to

[52] for the treatment of cancer.

[54] A method for treating cancer, comprising the step of administering to a subject a therapeutically effective amount of an anti-TSPAN8 antibody or antigen-binding fragment thereof described in any one of

[32] to

[43] , or the step of administering to a subject a therapeutically effective amount of a fusion, complex, or cell described in

[44] .

[55] Use of an anti-TSPAN8 antibody or antigen-binding fragment thereof described in any one of

[32] to

[43] , or use of a fusion, complex, or cell described in

[44] , in the manufacture of a pharmaceutical composition for the treatment of cancer. [Effects of the Invention]

[0012] The anti-TSPAN8-anti-CD3 bispecific antibody of the present invention binds to both the cancer antigen TSPAN8 and the T cell surface molecule CD3, thereby shortening the physical distance between cancer cells and T cells and thereby enhancing the cancer cell-killing activity of T cells. Furthermore, the anti-TSPAN8 antibody of the present invention has the effect of killing cancer cells by binding to TSPAN8. The anti-TSPAN8-anti-CD3 bispecific antibody and anti-TSPAN8 antibody of the present invention, or pharmaceutical compositions containing the antibodies, can be used for the treatment of cancer. [Brief explanation of the drawings]

[0013] [Figure 1-1] Figure 1-1 shows the results of flow cytometry analysis of the binding of anti-TSPAN8 antibodies (16B11, 16B12, 9F6, and 18C10) to KM-291-As. 16B11, 16B12, 9F6, and 18C10 represent the names of the antibodies. The horizontal axis of the figure represents fluorescence intensity, and the vertical axis represents cell count. The open areas represent binding of the negative control antibody, and the dark gray areas represent binding of the anti-TSPAN8 antibody. [Figure 1-2] Figure 1-2 shows the results of flow cytometry analysis of the binding of anti-TSPAN8 antibodies (16B11, 9F6, and 18C10) to KM-555-As. 16B11, 9F6, and 18C10 in the figure represent the names of the antibodies. The horizontal axis of the figure represents fluorescence intensity, and the vertical axis represents cell count. The open areas in the figure represent binding of the negative control antibody, and the dark gray areas represent binding of the anti-TSPAN8 antibody. [Figure 1-3] Figure 1-3 shows the results of flow cytometry analysis of the binding of anti-TSPAN8 antibodies (16B11, 9F6, and 18C10) to KM-556-As. 16B11, 9F6, and 18C10 in the figure represent the names of the antibodies. The horizontal axis of the figure represents fluorescence intensity, and the vertical axis represents cell count. The open areas in the figure represent binding of the negative control antibody, and the dark gray areas represent binding of the anti-TSPAN8 antibody. [Figure 2-1]Figure 2-1 shows the results of flow cytometry analysis of the binding of anti-TSPAN8 antibodies (16B11, 16B12, 9F6, 5B7, 12C12, 13A9, 15D1, and TAL69) to cultured human peritoneal mesothelial cells. The horizontal axis of the figure represents fluorescence intensity, and the vertical axis represents cell count. The open areas in the figure represent binding of the negative control antibody, and the dark gray areas represent binding of the anti-TSPAN8 antibody. [Figure 2-2] Figure 2-2 shows the results of flow cytometry analysis of the binding of anti-TSPAN8 antibodies (18C10, 19E4, 21F7, and TAL69) to cultured human peritoneal mesothelial cells. The horizontal axis of the figure represents fluorescence intensity, and the vertical axis represents cell count. The open areas in the figure represent binding of the negative control antibody, and the dark gray areas represent binding of the anti-TSPAN8 antibody. [Figure 3-1] Figure 3-1 shows the results of flow cytometry analysis of the binding of anti-TSPAN8 antibodies (16B11, 16B12, 9F6, 18C10, and TAL69) to KM-501-As. The horizontal axis of the figure represents fluorescence intensity, and the vertical axis represents cell count. The open areas represent binding of the negative control antibody, and the dark gray areas represent binding of the anti-TSPAN8 antibody. [Figure 3-2] Figure 3-2 shows the results of flow cytometry analysis of the binding of anti-TSPAN8 antibodies (16B11, 16B12, 9F6, 18C10, and TAL69) to KM-503-As. The horizontal axis of the figure represents fluorescence intensity, and the vertical axis represents cell count. The open areas represent binding of the negative control antibody, and the dark gray areas represent binding of the anti-TSPAN8 antibody. [Figure 4] Figure 4 shows the results of flow cytometry analysis of the binding of anti-TSPAN8 antibodies (16B11, 16B12, 9F6, 5B7, 12C12, 13A9, 15D1, 18C10, 19E4, 21F7, and TAL69) to human umbilical vascular endothelial cells (donor 2). The horizontal axis of the figure represents fluorescence intensity, and the vertical axis represents cell count. The open areas in the figure represent binding of the negative control antibody, and the dark gray areas represent binding of the anti-TSPAN8 antibody. [Figure 5-1]Figure 5-1 shows the results of flow cytometry analysis of the binding of anti-TSPAN8 antibodies (16B11, 16B12, and TAL69) to CHO-K1 cells expressing human-mouse TSPAN8-GFP chimeric protein or human-rat TSPAN8-GFP chimeric protein (chimeric protein-expressing CHO-K1 cells). The horizontal axis of the figure represents fluorescence intensity, and the vertical axis represents cell count. In the figure, gray indicates binding of anti-TSPAN8 antibodies to wild-type human TSPAN8-expressing CHO-K1 cells, black indicates binding to chimeric protein-expressing CHO-K1 cells, and white indicates binding to mock cells. Experiments were performed in duplicate. [Figure 5-2] Figure 5-2 shows the homology of the sequences consisting of amino acids 126 to 155 of four types of TSPAN8 proteins from human, mouse, rat, and cynomolgus monkey. An asterisk indicates a perfect match, dots indicate that three of the four amino acids are identical, and spaces indicate that two or more amino acids are different. [Figure 5-3] Figure 5-3 shows the results of flow cytometry analysis of the binding of anti-TSPAN8 antibodies (16B11, 16B12, and TAL69) to the T131A and T131N mutants of human TSPAN8 protein. The horizontal axis of the figure represents fluorescence intensity, and the vertical axis represents cell count. The gray area in the figure represents binding of anti-TSPAN8 antibodies to wild-type human TSPAN8-expressing CHO-K1 cells, the black area represents binding of mutant-expressing CHO-K1 cells, and the open area represents binding of mock cells. Experiments were performed in duplicate. [Figure 6-1] Figure 6-1 shows the results of flow cytometry analysis of the competitive effect of other anti-TSPAN8 antibodies (Competitor (CPTR): 16B11, 9F6, 18C10, TAL69) on the binding of 16B11 to NSC-15CF. The horizontal axis of the figure represents fluorescence intensity, and the vertical axis represents cell count. The gray area in the figure represents binding of fluorescent-labeled 16B11 in the presence of a negative control antibody, the black area represents binding of fluorescent-labeled 16B11 in the presence of each CPTR, and the open area represents a histogram of unstained fluorescent-labeled 16B11. [Figure 6-2]Figure 6-2 shows the results of flow cytometry analysis of the competitive effect of 16B12 on the binding of fluorescently labeled anti-TSPAN8 antibodies (16B11, 9F6, and 18C10) to NSC-15CF. The horizontal axis of the figure represents fluorescence intensity, and the vertical axis represents cell count. The gray area in the figure represents binding of fluorescently labeled anti-TSPAN8 antibody in the presence of a negative control antibody, the black area represents binding of fluorescently labeled anti-TSPAN8 antibody in the presence of 16B12, and the open area represents a histogram of unstained fluorescently labeled antibody. [Figure 6-3] Figure 6-3 shows the results of flow cytometry analysis of the competitive effect of other anti-TSPAN8 antibodies (CPTR: 16B12, 16B11, 9F6, 18C10, and TAL69) on the binding of 16B12 to NSC-15CF. The horizontal axis of the figure represents fluorescence intensity, and the vertical axis represents cell count. The gray area in the figure represents binding of fluorescent-labeled 16B12 in the presence of a negative control antibody, the black area represents binding of fluorescent-labeled 16B12 in the presence of other anti-TSPAN8 antibodies (CPTR), and the open area represents a histogram of unstained fluorescent-labeled antibodies. [Figure 7] Figure 7 shows the cytotoxic activity of 16B11.1 in a co-culture system of 60As6-Luc / GFP cells and human NK cells. The horizontal axis shows antibody concentration, and the vertical axis shows cytotoxic activity calculated from luciferase activity produced by 60As6-Luc / GFP cells. ● and ▲ indicate the average cytotoxic activity at each concentration of the control antibody and 16B11.1, respectively. Error bars indicate standard deviation. [Figure 8] Figure 8 shows the binding activity of the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody to the TSPAN8 LEL region peptide and the CD3εδ complex protein. The horizontal axis represents antibody concentration, and the vertical axis represents the amount of antibody binding. Figures 8-1 and 8-2 show the average binding amounts of the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody to the TSPAN8 LEL region peptide and the CD3εδ complex protein, respectively. Error bars indicate standard deviation. [Figure 9]Figure 9 shows the cytotoxic activity of anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody in a coculture system of 60As6-Luc / GFP cells and human peripheral blood mononuclear cells. The horizontal axis shows antibody concentration, and the vertical axis shows cell proliferation (%) in the fluorescence area of ​​60As6-Luc / GFP cells 3 days after antibody addition, relative to the fluorescence area in the absence of antibody (100%). ● indicates the average cell proliferation at each concentration of anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody. Error bars indicate standard deviation. [Figure 10-1] Figure 10-1 shows the cytotoxic activity of anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody against gastric cancer cells in ascites cells from human gastric cancer patients. The horizontal axis shows antibody concentration, and the vertical axis shows the viable cell count (%) of gastric cancer cells 3 days after antibody addition, with the number of viable cells in the absence of antibody set at 100%. ● and ■ show the average viable cell count (%) for each concentration of control antibody and anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody, respectively. Error bars show the standard deviation. [Figure 10-2] Figure 10-2 shows the activation of CD4-positive T cells in ascites cells from a human gastric cancer patient by anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody, as measured by the induction of CD25 expression. The horizontal axis represents antibody concentration. The vertical axis represents the fold change in CD25 expression levels 3 days after addition of the antibody to CD4-positive T cells in ascites. ● and ■ represent the average fold change in CD25 expression levels at each concentration of control antibody and anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody, respectively. Error bars represent standard deviation. [Figure 10-3] Figure 10-3 shows the activation of CD8-positive T cells in ascites cells from a human gastric cancer patient by anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody, as measured by the induction of CD25 expression. The horizontal axis represents antibody concentration. The vertical axis represents the fold change in CD25 expression levels 3 days after addition of the antibody to CD8-positive T cells in ascites. ● and ■ represent the average fold change in CD25 expression levels at each concentration of control antibody and anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody, respectively. Error bars represent standard deviation. [Figure 11-1]Figure 11-1 shows the antitumor effect of the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody in a gastric cancer peritoneal dissemination model. The vertical axis shows the average luciferin luminescence intensity due to luciferase expressed by 60As6-Luc / GFP cells in the peritoneal cavity. Error bars indicate standard error. The horizontal axis shows the antibody dose. The significance probability P value was determined by comparing the luminescence intensity of the control group with that of the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody-treated group using Dunnett's multiple comparison test. The ** in the figure indicates groups with a P value less than the significance level of 0.01. [Figure 11-2] Figure 11-2 shows the effect of the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody on survival in a gastric cancer peritoneal dissemination model. The vertical axis represents the survival rate, and the horizontal axis represents the number of days after cancer cell transplantation. The anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody and expanded panT cells were administered 7 and 10 days after 60As6-Luc / GFP transplantation, as indicated by the triangle. [Figure 12] Figure 12 shows the results of flow cytometry analysis of the binding of 16B11 to various cancer cell lines. The horizontal axis of the figure represents fluorescence intensity, and the vertical axis represents cell count. The open areas in the figure represent binding of the negative control antibody, and the dark gray areas represent binding of 16B11. [Figure 13-1] Figure 13-1 shows the cytotoxic activity of anti-TSPAN8 (16B11)-anti-CD3 bispecific antibodies against various cancer cell lines in co-culture with human peripheral blood mononuclear cells. The horizontal axis represents antibody concentration, and the vertical axis represents the viable cell count (%) of each cancer cell line 3 days after antibody addition, with the count without antibody set at 100%. Each symbol represents the mean (quadruplicate) viable cell count (%) of each cancer cell line at each concentration of anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody. [Figure 13-2]Figure 13-2 shows the activation of CD4+ T cells in cocultures of human peripheral blood mononuclear cells and various cancer cell lines by anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody, as measured by the induction of CD25 expression. The horizontal axis represents the antibody concentration. The vertical axis represents the fold change in CD25 expression in CD4+ T cells 3 days after antibody addition. Each symbol represents the mean (quadruplicate) fold change in CD25 expression at each concentration of anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody. [Figure 13-3] Figure 13-3 shows the activation of CD8+ T cells in coculture of human peripheral blood mononuclear cells with various cancer cell lines by anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody, as measured by the induction of CD25 expression. The horizontal axis represents the antibody concentration. The vertical axis represents the fold change in CD25 expression in CD8+ T cells 3 days after antibody addition. Each symbol represents the mean (quadruplicate) fold change in CD25 expression at each concentration of anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody. [Figure 14] Figure 14 shows the antitumor effect of the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody in a human PBMC-transfected HT-29 cell-bearing subcutaneous tumor model. Figure 14-1 shows the mean tumor volume on days after the start of antibody administration, and the error bars indicate the standard error. Figure 14-2 shows the tumor volume value for each individual 11 days after the start of administration, with the horizontal lines indicating the mean and standard error, and the horizontal axis indicating the antibody dose. The significance probability P value was determined by comparing the tumor volume in the PBS-administered group with the tumor volume in the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody-administered group using Dunnett's multiple comparison test. ** in the figure indicates groups with a P value less than the significance level of 0.01. DETAILED DESCRIPTION OF THE INVENTION

[0014] The present invention will be described in detail below.

[0015] <Definition> Terms used herein are used in the sense commonly used by those skilled in the art unless otherwise defined below.

[0016] An antibody (or immunoglobulin) is a glycoprotein consisting of a four-chain structure with a symmetric Y-shaped configuration, consisting of two heavy chains with a single sequence and two light chains with a single sequence. There are five classes of antibodies: IgG, IgM, IgA, IgD, and IgE. The basic structure of antibody molecules is common to all classes: two heavy chains with a molecular weight of 50,000-70,000 and two light chains with a molecular weight of 20,000-30,000 are bound by disulfide bonds and non-covalent bonds to form a Y-shaped, four-chain antibody molecule with a molecular weight of 150,000-190,000. The heavy chains are typically polypeptide chains containing approximately 440 amino acids, and each class has a characteristic structure: Igγ, Igμ, Igα, Igδ, and Igε, corresponding to IgG, IgM, IgA, IgD, and IgE, respectively. IgG is further divided into subclasses: IgG1, IgG2, IgG3, and IgG4, with corresponding heavy chains called Igγ1, Igγ2, Igγ3, and Igγ4. Light chains typically consist of a polypeptide chain containing approximately 220 amino acids, and two types, L and K, are known, called Igλ and Igκ, respectively. These two types of light chains can pair with any type of heavy chain.

[0017] Antibody molecules have four intrachain disulfide bonds in heavy chains (five in Igμ and Igε) and two in light chains, forming a loop every 100–110 amino acid residues. The three-dimensional structures of these disulfide bonds are similar between each loop and are called structural units or domains. The domain located at the N-terminus of both heavy and light chains is called the variable region. It has diverse amino acid sequences even among antibodies of the same class (or subclass) produced by the same animal species, and is known to be involved in the specific binding between the antibody and the antigen. The amino acid sequence of the C-terminal domain downstream of the variable region is nearly constant for each class or subclass and is called the constant region. From the N-terminus to the C-terminus, the heavy chain contains a heavy chain variable region (VH) and a heavy chain constant region (CH). The CH is further divided into three domains from the N-terminus: CH1, CH2, and CH3. From the N-terminus to the C-terminus, the light chain contains a light chain variable region (VL) and a light chain constant region (CL).

[0018] The amino acid sequences of the three complementarity-determining regions (CDRs) present in VH and VL vary greatly, contributing to the variability of the variable regions. The CDRs are regions consisting of approximately 5 to 10 amino acid residues located in the order of CDR1, CDR2, and CDR3 at the N-terminus of each heavy chain and light chain, and form the antigen-binding site. Meanwhile, the portions of the variable regions other than the CDRs are called framework regions (FRs), which consist of FRs 1 to 4 and show relatively little variation in amino acid sequence.

[0019] When an antibody is treated with the protease papain, three antibody fragments are obtained. The two N-terminal fragments are called the Fab (Fragment, antigen binding) region. As used herein, the term "Fab region" refers to a region consisting of the VH and CH1 domains of the heavy chain and the light chain (VL and CL), and binds to an antigen at the antigen-binding site at the tip of the Fab region. As used herein, the term "heavy chain fragment" refers to a fragment consisting of the VH and CH1 domains of the heavy chain that make up the Fab region. The C-terminal fragment is referred to as the Fc (Fragment, crystallizable) region. As used herein, "Fc polypeptide" refers to a polypeptide consisting of the CH2 domain and CH3 domain of a heavy chain, and "Fc region" refers to a complex consisting of two Fc polypeptides. The heavy chain fragment and the Fc polypeptide are connected by a region called the hinge region, and the two heavy chains of an antibody are disulfide-bonded at the hinge region.

[0020] As used herein, the term "antigen" is used in its commonly used sense, particularly as a term referring to a molecule or a portion of a molecule to which an antigen-binding protein such as an antibody or an antigen-binding fragment can specifically bind. An antigen can be a molecule such as a protein or a nucleic acid. A single antigen may have one or more epitopes that can interact with different antibodies, etc.

[0021] As used herein, "epitope" or "antigenic determinant" refers to a specific structural unit of an antigen that is recognized and bound by an antigen-binding protein, including any determinant that can be bound by an antigen-binding protein such as an antibody or T cell receptor. Epitope determinants can include chemically active surface groups of molecules, such as amino acids, sugar side chains, phosphoryl groups, or sulfonyl groups, and can have specific three-dimensional structural characteristics and / or specific charge characteristics. When the antigen is a protein, they include specific amino acids that directly contact antibodies, etc. In general, antibodies specific for a particular target antigen preferentially recognize epitopes on the target antigen in a complex mixture of proteins and / or macromolecules. Epitopes often consist of surface-accessible amino acid residues and / or sugar side chains, typically consisting of a sequence of 6 to 10 amino acids or 5 to 8 monosaccharides. Epitopes may have specific three-dimensional structural characteristics and specific charge characteristics. Epitopes may include amino acid residues directly involved in binding and other amino acid residues not directly involved in binding. The epitope to which an antigen-binding protein binds can be identified using methods well known to those skilled in the art, such as mass spectrometry (e.g., hydrogen / deuterium exchange mass spectrometry; HDX-MS), alanine scanning mutagenesis, crystallography, peptide competition, etc.

[0022] As used herein, the terms "competition" or "competing" refer to the phenomenon in which, when two or more types of antibodies are added simultaneously or sequentially to a reaction solution, one antibody inhibits the binding of the other antibody to the antigen, thereby reducing the binding ability of the other antibody to the antigen.

[0023] As used herein, the term "antigen-binding fragment" refers to a molecule comprising at least one polypeptide chain possessing antigen-binding activity derived from an antibody. Representative antigen-binding fragments include single-chain variable region fragments (scFv), Fab fragments, Fab' fragments, and F(ab')2 fragments. scFv is a monovalent antigen-binding fragment consisting of a VH and VL linked by a linker. Fab fragments are monovalent antigen-binding fragments consisting of a light chain and a fragment containing the VH and CH1 domains of the heavy chain. Fab' fragments are monovalent antigen-binding fragments consisting of a fragment containing the light chain, the VH and CH1 domains of the heavy chain, and a portion of the hinge region, and this hinge region contains cysteine ​​residues that constituted the inter-heavy chain disulfide bonds. F(ab')2 fragments are divalent molecules in which Fab' fragments are linked by disulfide bonds. "Monovalent" means that the molecule contains one antigen-binding site, and "bivalent" means that the molecule contains two antigen-binding sites. As used herein, the term "scFv region" refers to a region containing a monovalent antigen-binding fragment comprising a VH and a VL linked by a linker.

[0024] A one-armed antibody is also a type of antigen-binding fragment and comprises one Fab region and one Fc region, with the heavy chain fragment of the Fab region linked to one of the two Fc polypeptides of the Fc region. In one embodiment, a one-armed antibody comprises one heavy chain (VH, CH1 domain, hinge region, Fc polypeptide (CH2 domain and CH3 domain)), one light chain (VL and CL), and an Fc polypeptide.

[0025] As used herein, the term "multispecific antibody" refers to an antibody that can specifically bind to two or more different antigens, and is called, for example, a bispecific antibody or a trispecific antibody depending on the number of antigens to which it binds. Multispecific antibodies include complexes of two or more antibodies and / or antigen-binding fragments, each capable of binding to a different antigen, and the term "antibody" as used herein includes multispecific antibodies unless otherwise limited by the context.

[0026] As used herein, the term "bispecific antibody" refers to an antibody that can specifically bind to two different antigens. The term "anti-TSPAN8-anti-CD3 bispecific antibody" refers to a bispecific antibody that has binding activity for TSPAN8 and binding activity for CD3.

[0027] As used herein, the term "human antibody" refers to an antibody having a human immunoglobulin amino acid sequence. As used herein, the term "humanized antibody" refers to an antibody in which some, most, or all of the amino acid residues other than the CDRs have been substituted with amino acid residues derived from human immunoglobulin molecules. The humanization method is not particularly limited, and humanized antibodies can be prepared by referring to, for example, U.S. Pat. No. 5,225,539 and U.S. Pat. No. 6,180,370.

[0028] The amino acid residue numbers of antibodies used herein can be specified according to the Kabat numbering system or the EU index (Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed., 1991, NIH Publication No. 91-3242) by designating them.

[0029] The terms "first" and "second" are used herein for convenience to distinguish between two or more of each type of moiety. The use of such terms is not intended to imply a particular order or meaning unless expressly stated.

[0030] As used herein, "linked" or "linked" means that multiple components (e.g., Fab regions and Fc polypeptides) are linked directly or via one or more intermediaries (e.g., peptide linkers). As used herein, "peptide linker" refers to any one or more amino acid residues that can be introduced by genetic engineering to link variable regions. The length of the peptide linker used in the present invention is not particularly limited and can be appropriately selected by those skilled in the art depending on the purpose.

[0031] As used herein, "identity" refers to the Identity value obtained using EMBOSS Needle (Nucleic Acids Res., 2015, Vol. 43, pW580-W584) with the default parameters. The parameters are as follows: Gap Open Penalty = 10 Gap Extend Penalty = 0.5 Matrix = EBLOSUM62 End Gap Penalty = false

[0032] As used herein, the term "subject" refers to a human or other animal in need of such prevention or treatment. In one embodiment, the subject is a human in need of such prevention or treatment.

[0033] <Anti-TSPAN8-anti-CD3 bispecific antibody of the present invention> The present invention provides the following bispecific antibodies that bind to TSPAN8 and CD3 (also referred to as "anti-TSPAN8-anti-CD3 bispecific antibodies"): A bispecific antibody that binds to TSPAN8 and CD3, (a) a Fab region of an anti-TSPAN8 antibody consisting of a heavy chain fragment containing the heavy chain variable region of the anti-TSPAN8 antibody and a light chain containing the light chain variable region of the anti-TSPAN8 antibody; (b) an anti-CD3-scFv region comprising the heavy chain variable region and the light chain variable region of an anti-CD3 antibody; and (c) an Fc region consisting of a first Fc polypeptide linked to a heavy chain fragment of the Fab region of (a) and a second Fc polypeptide linked to the anti-CD3-scFv region of (b); A bispecific antibody comprising:

[0034] The anti-TSPAN8-anti-CD3 bispecific antibody of the present invention has a structure comprising one Fab region of a first antibody, an scFv region of a second antibody, and one Fc region. Antibodies with such a structure are also known as "bottle-opener antibodies" (WO 2014 / 110601). In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention is a human antibody or a humanized antibody.

[0035] The anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises an anti-TSPAN8 antibody Fab region, which consists of a heavy chain fragment containing the heavy chain variable region of the anti-TSPAN8 antibody and a light chain containing the light chain variable region of the anti-TSPAN8 antibody.

[0036] In one embodiment, the heavy chain variable region of the anti-TSPAN8 antibody comprises a CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 6, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 of SEQ ID NO: 6, and a CDR3 consisting of the amino acid sequence from amino acid numbers 99 to 110 of SEQ ID NO: 6, and the light chain variable region of the anti-TSPAN8 antibody comprises a CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 8, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 8, and a CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 8.

[0037] In one embodiment, the heavy chain variable region of the anti-TSPAN8 antibody comprises a CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 10, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 of SEQ ID NO: 10, and a CDR3 consisting of the amino acid sequence from amino acid numbers 99 to 110 of SEQ ID NO: 10, and the light chain variable region of the anti-TSPAN8 antibody comprises a CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 12, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 12, and a CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 12.

[0038] In one embodiment, the heavy chain variable region of the anti-TSPAN8 antibody consists of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 6, and the light chain variable region of the anti-TSPAN8 antibody consists of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 8.

[0039] In one embodiment, the heavy chain variable region of the anti-TSPAN8 antibody consists of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10, and the light chain variable region of the anti-TSPAN8 antibody consists of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12.

[0040] The heavy chain constant region from which the CH1 domain of the heavy chain fragment of the anti-TSPAN8 antibody Fab region is derived can be any of Igγ, Igμ, Igα, Igδ, and Igε constant regions. Igγ can be selected from, for example, Igγ1, Igγ2, Igγ3, and Igγ4. In one embodiment, the heavy chain fragment of the anti-TSPAN8 antibody Fab region comprises a CH1 domain derived from a human Igγ1 constant region.

[0041] The CL of the light chain of the anti-TSPAN8 antibody Fab region can be selected from either an Igλ or Igκ constant region. In one embodiment, the anti-TSPAN8 antibody Fab region comprises a CL that is an Igκ constant region. In one embodiment, the light chain of the anti-TSPAN8 antibody Fab region comprises a CL that is a human Igκ constant region.

[0042] In one embodiment, the anti-TSPAN8 antibody Fab region consists of a heavy chain fragment consisting of the amino acid sequence of amino acids 1 to 219 of SEQ ID NO: 6 and a light chain fragment consisting of the amino acid sequence of SEQ ID NO: 8. In one embodiment, the anti-TSPAN8 antibody Fab region consists of a heavy chain fragment consisting of the amino acid sequence of amino acids 1 to 219 of SEQ ID NO: 10 and a light chain fragment consisting of the amino acid sequence of SEQ ID NO: 12.

[0043] The anti-TSPAN8-anti-CD3 bispecific antibodies of the present invention comprise an anti-CD3-scFv region comprising the heavy chain variable region and light chain variable region of an anti-CD3 antibody as the scFv region. The anti-TSPAN8-anti-CD3 bispecific antibodies of the present invention may use anti-CD3-scFv regions known in the art or scFv regions of anti-CD3 antibodies prepared based on the sequences of the heavy chain variable region and light chain variable region of anti-CD3 antibodies known in the art. Known anti-CD3 antibody clones include OKT3, UTCH1, L2K, and TR66, and their sequences have been used as bispecific antibodies (Pharmacol. Ther., 2018, Vol. 182, pp. 161-175).

[0044] In one embodiment, the heavy chain variable region of the anti-CD3 antibody comprises CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence from amino acid numbers 101 to 114 of SEQ ID NO: 14, and the light chain variable region of the anti-CD3 antibody comprises CDR1 consisting of the amino acid sequence from amino acid numbers 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence from amino acid numbers 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence from amino acid numbers 236 to 244 of SEQ ID NO: 14.

[0045] In one embodiment, the heavy chain variable region of the anti-CD3 antibody consists of the amino acid sequence from amino acid numbers 1 to 125 of SEQ ID NO: 14, and the light chain variable region of the anti-CD3 antibody consists of the amino acid sequence from amino acid numbers 146 to 254 of SEQ ID NO: 14.

[0046] In the anti-CD3-scFv region, the type and length of the peptide linker linking the heavy chain variable region and light chain variable region of the anti-CD3 antibody are not particularly limited and can be selected appropriately by those skilled in the art. However, the preferred length is 5 amino acids or more (the upper limit is not particularly limited, but is usually 30 amino acids or less, preferably 20 amino acids or less), and particularly preferably 15 amino acids. Examples of peptide linkers that can be used include a glycine-serine linker (GS linker) and a glycine-lysine-proline-glycine-serine linker (GKPGS linker). Examples of such linkers include the following: Ser Gly-Ser Gly-Gly-Ser Ser-Gly-Gly Gly-Gly-Gly-Ser (SEQ ID NO: 15) Ser-Gly-Gly-Gly (SEQ ID NO: 16) Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 17) Ser-Gly-Gly-Gly-Gly (SEQ ID NO: 18) Gly-Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 19) Ser-Gly-Gly-Gly-Gly-Gly (SEQ ID NO: 20) Gly-Gly-Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 21) Ser-Gly-Gly-Gly-Gly-Gly-Gly (SEQ ID NO: 22) (Gly-Gly-Gly-Gly-Ser)n (Ser-Gly-Gly-Gly-Gly)n Gly-Lys-Pro-Gly-Ser (SEQ ID NO: 23) (Gly-Lys-Pro-Gly-Ser)n The above n represents an integer equal to or greater than 1. The length and sequence of the peptide linker can be appropriately selected by those skilled in the art depending on the purpose.

[0047] In one embodiment, the anti-CD3-scFv region consists of the amino acid sequence from amino acid number 1 to amino acid number 254 of SEQ ID NO:14.

[0048] In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises a heavy chain fragment of an anti-TSPAN8 antibody comprising a heavy chain variable region comprising CDR1 consisting of the amino acid sequence from 31 to 35 of SEQ ID NO: 6, CDR2 consisting of the amino acid sequence from 50 to 66 of SEQ ID NO: 6, and CDR3 consisting of the amino acid sequence from 99 to 110 of SEQ ID NO: 6, and a heavy chain of an anti-TSPAN8 antibody linked to a first Fc polypeptide; and a light chain variable region comprising CDR1 consisting of the amino acid sequence from 24 to 34 of SEQ ID NO: 8, CDR2 consisting of the amino acid sequence from 50 to 56 of SEQ ID NO: 8, and CDR3 consisting of the amino acid sequence from 89 to 96 of SEQ ID NO: 8. and a heavy chain variable region of an anti-CD3 antibody comprising CDR1 consisting of the amino acid sequence from amino acid nos. 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence from amino acid nos. 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence from amino acid nos. 101 to 114 of SEQ ID NO: 14, and a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked, the light chain variable region of an anti-CD3 antibody comprising CDR1 consisting of the amino acid sequence from amino acid nos. 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence from amino acid nos. 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence from amino acid nos. 236 to 244 of SEQ ID NO: 14.In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises a heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody heavy chain variable region comprising CDR1 consisting of the amino acid sequence from 31 to 35 of SEQ ID NO: 10, CDR2 consisting of the amino acid sequence from 50 to 66 of SEQ ID NO: 10, and CDR3 consisting of the amino acid sequence from 99 to 110 of SEQ ID NO: 10 is linked to a first Fc polypeptide; and a light chain variable region comprising CDR1 consisting of the amino acid sequence from 24 to 34 of SEQ ID NO: 12, CDR2 consisting of the amino acid sequence from 50 to 56 of SEQ ID NO: 12, and CDR3 consisting of the amino acid sequence from 89 to 96 of SEQ ID NO: 12. the heavy chain variable region of an anti-CD3 antibody comprising CDR1 consisting of the amino acid sequence from 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence from 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence from 101 to 114 of SEQ ID NO: 14, and a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked, the light chain variable region of an anti-CD3 antibody comprising CDR1 consisting of the amino acid sequence from 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence from 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence from 236 to 244 of SEQ ID NO: 14.

[0049] In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises a heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody, comprising a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 121 of SEQ ID NO: 6, is linked to a first Fc polypeptide; a light chain of an anti-TSPAN8 antibody in which a light chain variable region consisting of the amino acid sequence of amino acids 1 to 107 of SEQ ID NO: 8, is linked to a second Fc polypeptide; and a polypeptide in which an anti-CD3-scFv region, comprising a heavy chain variable region of an anti-CD3 antibody consisting of the amino acid sequence of amino acids 1 to 125 of SEQ ID NO: 14 and a light chain variable region of an anti-CD3 antibody consisting of the amino acid sequence of amino acids 146 to 254 of SEQ ID NO: 14, is linked to a second Fc polypeptide. In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises a heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody, comprising a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 121 of SEQ ID NO: 10, is linked to a first Fc polypeptide; a light chain of an anti-TSPAN8 antibody in which a light chain variable region consisting of the amino acid sequence of amino acids 1 to 107 of SEQ ID NO: 12, is linked to a second Fc polypeptide; and a polypeptide in which an anti-CD3-scFv region, comprising a heavy chain variable region of an anti-CD3 antibody consisting of the amino acid sequence of amino acids 1 to 125 of SEQ ID NO: 14 and a light chain variable region of an anti-CD3 antibody consisting of the amino acid sequence of amino acids 146 to 254 of SEQ ID NO: 14, is linked to a second Fc polypeptide.

[0050] In the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention, the heavy chain constant region from which the first and second Fc polypeptides constituting the Fc region are derived can be any of Igγ, Igμ, Igα, Igδ, and Igε constant regions. Igγ can be selected from, for example, Igγ1, Igγ2, Igγ3, and Igγ4. In one embodiment, the first and second Fc polypeptides are Fc polypeptides derived from the human Igγ1 constant region.

[0051] The Fc region of the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention may contain mutations that reduce antibody-dependent cellular cytotoxicity (ADCC) or complement-dependent cytotoxicity (CDC). L234A is a substitution of leucine at amino acid position 234 according to the EU index in the human Igγ1 constant region with alanine. L235A is a substitution of leucine at amino acid position 235 according to the EU index in the human Igγ1 constant region with alanine. The amino acid mutations L234A and L235A in the human Igγ1 constant region are referred to as "LALA mutations." This mutation is known to reduce antibody-dependent cellular cytotoxicity and complement-dependent cytotoxicity of antibodies (Mol. Immunol., 1992, Vol. 29, pp. 633-639; J. Immunol., 2000, Vol. 164, pp. 4178-4184).

[0052] The Fc region of the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention may further comprise mutations based on other known techniques, such as the N297G mutation (Protein Cell, 2018, Vol. 9, pp. 63-73) or mutations based on the Knobs-into-holes technique (hereinafter also referred to as "Knobs-into-holes mutations"). Knobs-into-holes technology involves replacing the amino acid side chains in the CH3 region of one heavy chain with larger side chains (knobs) and the amino acid side chains in the CH3 region of the other heavy chain with smaller side chains (holes) so that the knobs are positioned within the holes, promoting heterodimerization of the heavy chains and enabling efficient production of the desired heterodimerized antibody molecules (Nature, 1994, Vol. 372, pp. 379-383; Nature Biotech., 1998, Vol. 16, pp. 677-681; J. Mol. Biol., 1997, Vol. 270, pp. 26-35; Proc. Natl. Acad. Sci. USA, 2013, Vol. 110, pp. E2987-E2996).

[0053] In one embodiment, an anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises an Fc region comprising amino acid mutations L234A and L235A (LALA mutations). In one embodiment, an anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises an Fc region comprising an N297G mutation. In one embodiment, an anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises an Fc region comprising one or more of amino acid mutations L234A and L235A (LALA mutations), an N297G mutation, and a knobs-into-holes mutation. In one embodiment, an anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises an Fc region comprising amino acid mutations L234A and L235A (LALA mutations), an N297G mutation, and a knobs-into-holes mutation. In one embodiment, the knobs-into-holes mutations contained in an anti-TSPAN8-anti-CD3 bispecific antibody of the invention are a T366W mutation in one Fc polypeptide forming its Fc region, and T366S, L368A, and Y407V mutations in another Fc polypeptide forming its Fc region (see WO 1998 / 050431).

[0054] In one embodiment, an anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises an Fc region consisting of a first Fc polypeptide consisting of the amino acid sequence of amino acids 235 to 451 of SEQ ID NO: 6 and a second Fc polypeptide consisting of the amino acid sequence of amino acids 270 to 486 of SEQ ID NO: 14.

[0055] In this specification, amino acid mutations such as the LALA mutation, N297G mutation, and knobs-into-holes mutation are described based on their amino acid positions in the human Igγ1 constant region according to the EU index. For example, as described above, L234A is a substitution of leucine with alanine at amino acid position 234 in the human Igγ1 constant region according to the EU index.

[0056] In the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention, the heavy chain fragment containing the heavy chain variable region of the anti-TSPAN8 antibody and the Fc polypeptide (first Fc polypeptide) may be linked via a hinge region to form the heavy chain of the anti-TSPAN8 antibody. Furthermore, in the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention, the anti-CD3-scFv region and the Fc polypeptide (second Fc polypeptide) may be linked via a hinge region.

[0057] In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises an anti-TSPAN8 antibody heavy chain in which a heavy chain fragment comprising the heavy chain variable region of the anti-TSPAN8 antibody and a first Fc polypeptide are linked via a hinge region. In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked via a hinge region. In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises an anti-TSPAN8 antibody heavy chain in which a heavy chain fragment comprising the heavy chain variable region of the anti-TSPAN8 antibody and a first Fc polypeptide are linked via a hinge region, and a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked via a hinge region.

[0058] In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises a heavy chain fragment of an anti-TSPAN8 antibody comprising a heavy chain variable region comprising CDR1 consisting of the amino acid sequence from 31 to 35 of SEQ ID NO: 6, CDR2 consisting of the amino acid sequence from 50 to 66 of SEQ ID NO: 6, and CDR3 consisting of the amino acid sequence from 99 to 110 of SEQ ID NO: 6, and a first Fc polypeptide linked via a hinge region; and a light chain variable region comprising CDR1 consisting of the amino acid sequence from 24 to 34 of SEQ ID NO: 8, CDR2 consisting of the amino acid sequence from 50 to 56 of SEQ ID NO: 8, and CDR3 consisting of the amino acid sequence from 89 to 96 of SEQ ID NO: 8. and a heavy chain variable region of an anti-CD3 antibody comprising CDR1 consisting of the amino acid sequence from amino acid nos. 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence from amino acid nos. 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence from amino acid nos. 101 to 114 of SEQ ID NO: 14, and a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked via a hinge region, the light chain variable region of an anti-CD3 antibody comprising CDR1 consisting of the amino acid sequence from amino acid nos. 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence from amino acid nos. 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence from amino acid nos. 236 to 244 of SEQ ID NO: 14.In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises a heavy chain of an anti-TSPAN8 antibody in which a first Fc polypeptide and a heavy chain fragment of the anti-TSPAN8 antibody are linked via a hinge region to each other, the heavy chain fragment comprising a heavy chain variable region comprising CDR1 consisting of the amino acid sequence from 31 to 35 of SEQ ID NO: 10, CDR2 consisting of the amino acid sequence from 50 to 66 of SEQ ID NO: 10, and CDR3 consisting of the amino acid sequence from 99 to 110 of SEQ ID NO: 10; and a light chain variable region comprising CDR1 consisting of the amino acid sequence from 24 to 34 of SEQ ID NO: 12, CDR2 consisting of the amino acid sequence from 50 to 56 of SEQ ID NO: 12, and CDR3 consisting of the amino acid sequence from 89 to 96 of SEQ ID NO: 12. the light chain of an anti-TSPAN8 antibody comprising a region; and a heavy chain variable region of an anti-CD3 antibody comprising a CDR1 consisting of the amino acid sequence from amino acid nos. 31 to 35 of SEQ ID NO: 14, a CDR2 consisting of the amino acid sequence from amino acid nos. 50 to 68 of SEQ ID NO: 14, and a CDR3 consisting of the amino acid sequence from amino acid nos. 101 to 114 of SEQ ID NO: 14, and a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked via a hinge region, the light chain variable region of an anti-CD3 antibody comprising a CDR1 consisting of the amino acid sequence from amino acid nos. 168 to 181 of SEQ ID NO: 14, a CDR2 consisting of the amino acid sequence from amino acid nos. 197 to 203 of SEQ ID NO: 14, and a CDR3 consisting of the amino acid sequence from amino acid nos. 236 to 244 of SEQ ID NO: 14.

[0059] In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises a heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody, comprising a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 121 of SEQ ID NO: 6, and a first Fc polypeptide are linked via a hinge region; a light chain of an anti-TSPAN8 antibody in which a light chain variable region consisting of the amino acid sequence of amino acids 1 to 107 of SEQ ID NO: 8, and a polypeptide in which an anti-CD3-scFv region, comprising a heavy chain variable region of an anti-CD3 antibody consisting of the amino acid sequence of amino acids 1 to 125 of SEQ ID NO: 14 and a light chain variable region of an anti-CD3 antibody consisting of the amino acid sequence of amino acids 146 to 254 of SEQ ID NO: 14, and a second Fc polypeptide are linked via a hinge region. In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises a heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody, comprising a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 121 of SEQ ID NO: 10, and a first Fc polypeptide are linked via a hinge region; a light chain of an anti-TSPAN8 antibody in which a light chain variable region consisting of the amino acid sequence of amino acids 1 to 107 of SEQ ID NO: 12, and a polypeptide in which an anti-CD3-scFv region, comprising a heavy chain variable region of an anti-CD3 antibody consisting of the amino acid sequence of amino acids 1 to 125 of SEQ ID NO: 14 and a light chain variable region of an anti-CD3 antibody consisting of the amino acid sequence of amino acids 146 to 254 of SEQ ID NO: 14, and a second Fc polypeptide are linked via a hinge region.

[0060] In one embodiment, the heavy chain of an anti-TSPAN8 antibody, in which a heavy chain fragment of an anti-TSPAN8 antibody comprising the heavy chain variable region of the anti-TSPAN8 antibody and a first Fc polypeptide are linked via a hinge region, consists of the amino acid sequence of SEQ ID NO: 6 or 10. In one embodiment, the heavy chain of an anti-TSPAN8 antibody, in which a heavy chain fragment of an anti-TSPAN8 antibody comprising the heavy chain variable region of the anti-TSPAN8 antibody and a first Fc polypeptide are linked via a hinge region, consists of the amino acid sequence of SEQ ID NO: 6. In one embodiment, the light chain of an anti-TSPAN8 antibody consists of the amino acid sequence of SEQ ID NO: 8 or 12. In one embodiment, the light chain of an anti-TSPAN8 antibody consists of the amino acid sequence of SEQ ID NO: 8. In one embodiment, the polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked via a hinge region consists of the amino acid sequence of SEQ ID NO: 14. In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises a heavy chain fragment of an anti-TSPAN8 antibody comprising the heavy chain variable region of the anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 6 or 10, an anti-TSPAN8 antibody heavy chain to which the anti-TSPAN8 antibody light chain consisting of the amino acid sequence of SEQ ID NO: 8 or 12 and a first Fc polypeptide are linked, and a polypeptide in which the anti-CD3-scFv region consisting of the amino acid sequence of SEQ ID NO: 14 and a second Fc polypeptide are linked. In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention comprises a heavy chain fragment of an anti-TSPAN8 antibody comprising the heavy chain variable region of the anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 6, an anti-TSPAN8 antibody light chain consisting of the amino acid sequence of SEQ ID NO: 8 or 12 and a heavy chain to which the anti-TSPAN8 antibody light chain consisting of the amino acid sequence of SEQ ID NO: 8 or 12 and a first Fc polypeptide are linked, and a polypeptide in which the anti-CD3-scFv region consisting of the amino acid sequence of SEQ ID NO: 14 and a second Fc polypeptide are linked.

[0061] In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention is a bispecific antibody comprising an anti-TSPAN8 antibody heavy chain in which a heavy chain fragment of the anti-TSPAN8 antibody having the amino acid sequence of SEQ ID NO: 6 is linked to a first Fc polypeptide, an anti-TSPAN8 antibody light chain in which the anti-TSPAN8 antibody heavy chain fragment has the amino acid sequence of SEQ ID NO: 8, and a polypeptide in which an anti-CD3-scFv region having the amino acid sequence of SEQ ID NO: 14 is linked to a second Fc polypeptide.

[0062] As used herein, the term "post-translational modification" refers to post-translational modification of an antibody when the antibody is expressed in a cell. Examples of post-translational modifications include pyroglutamylation, glycosylation, oxidation, deamidation, glycation, and other modifications of glutamine or glutamic acid at the N-terminus of the heavy chain, and lysine deletion due to cleavage of lysine at the C-terminus of the heavy chain by carboxypeptidase. Such post-translational modifications are known to occur in various antibodies (J. Pharm. Sci., 2008, Vol. 97, pp. 2426-2447).

[0063] In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention may be post-translationally modified. In one embodiment, the post-translation modification is N-terminal pyroglutamylation of the heavy chain variable region and / or C-terminal lysine deletion of the heavy chain. It is known in the art that post-translational modifications such as N-terminal pyroglutamylation or C-terminal lysine deletion do not affect antibody activity (Analytical Biochemistry, 2006, Vol. 348, pp. 24-39).

[0064] The anti-TSPAN8-anti-CD3 bispecific antibody of the present invention binds to human TSPAN8 (gene ID: NM_004616.2) and human CD3εδ complex protein (CD3ε gene ID: NM_000733.3, CD3δ gene ID: NM_000732.4, or NM_001040651.1). Binding to human TSPAN8 and human CD3εδ complex protein can be confirmed using known binding activity measurement methods. Examples of methods for measuring binding activity include enzyme-linked immunosorbent assay (ELISA) and flow cytometry. When using ELISA, the method described in Example 8 can be used, and when using flow cytometry, the method described in Example 1 can be used, for example.

[0065] The anti-TSPAN8-anti-CD3 bispecific antibodies of the present invention can be prepared by those skilled in the art using methods known in the art based on the sequence information of the heavy and light chain variable regions of the anti-TSPAN8 antibody and the anti-CD3-scFv region disclosed herein. Furthermore, the anti-CD3-scFv region of the anti-TSPAN8-anti-CD3 bispecific antibodies of the present invention can be prepared by those skilled in the art using methods known in the art based on the sequence information of the heavy and light chain variable regions of known anti-CD3 antibodies. In one embodiment, the anti-TSPAN8-anti-CD3 bispecific antibodies of the present invention are humanized or human. When preparing humanized antibodies, appropriate backmutations may be introduced using methods known to those skilled in the art (Bioinformatics, 2015, Vol. 31, pp. 434-435). The anti-TSPAN8-anti-CD3 bispecific antibodies of the present invention are not particularly limited and can be produced, for example, according to the method described below in "Methods for Producing Bispecific Antibodies of the Present Invention and Bispecific Antibodies of the Present Invention Produced by the Methods."

[0066] <Polynucleotides of bispecific antibodies of the present invention> The present invention also provides the following polynucleotides (also referred to as "polynucleotides of the bispecific antibodies of the present invention") that can be used to produce the anti-TSPAN8-anti-CD3 bispecific antibodies of the present invention: (1) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody, including a heavy chain fragment of the anti-TSPAN8 antibody and a first Fc polypeptide; (2) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody; (3) A polynucleotide comprising a nucleotide sequence encoding a polypeptide comprising an anti-CD3-scFv region and a polypeptide comprising a second Fc polypeptide.

[0067] In one embodiment of the polynucleotide of (1) above, the polynucleotide of the bispecific antibody of the present invention is a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody, comprising a heavy chain variable region including CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 6, CDR2 consisting of the amino acid sequence of amino acids 50 to 66 of SEQ ID NO: 6, and CDR3 consisting of the amino acid sequence of amino acids 99 to 110 of SEQ ID NO: 6, and a first Fc polypeptide is linked; (b) A polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody, comprising a heavy chain variable region including CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 10, CDR2 consisting of the amino acid sequence of amino acids 50 to 66 of SEQ ID NO: 10, and CDR3 consisting of the amino acid sequence of amino acids 99 to 110 of SEQ ID NO: 10, is linked to a first Fc polypeptide.

[0068] In one embodiment of the polynucleotide of (1) above, the polynucleotide of the bispecific antibody of the present invention is a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody containing a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 6 and a first Fc polypeptide are linked; (b) A polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody containing a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10 and a first Fc polypeptide are linked.

[0069] In one embodiment of the polynucleotide (1) above, the polynucleotide of the bispecific antibody of the present invention is a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which the heavy chain fragment of the anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 6 and the first Fc polypeptide are linked.

[0070] In one embodiment of the polynucleotide of (2) above, the polynucleotide of the bispecific antibody of the present invention is a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody, the light chain variable region comprising CDR1 consisting of the amino acid sequence of amino acids 24 to 34 of SEQ ID NO: 8, CDR2 consisting of the amino acid sequence of amino acids 50 to 56 of SEQ ID NO: 8, and CDR3 consisting of the amino acid sequence of amino acids 89 to 96 of SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region including CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 12, CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 8, and CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 12.

[0071] In one embodiment of the polynucleotide of (2) above, the polynucleotide of the bispecific antibody of the present invention is a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12.

[0072] In one embodiment of the polynucleotide (2) above, the polynucleotide of the bispecific antibody of the present invention is a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO:8.

[0073] In one embodiment of the polynucleotide of (3) above, the polynucleotide of the bispecific antibody of the present invention is the following polynucleotide: A polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked, the anti-CD3 antibody heavy chain variable region comprising: CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 101 to 114 of SEQ ID NO: 14; and an anti-CD3 antibody light chain variable region comprising: CDR1 consisting of the amino acid sequence of amino acids 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 236 to 244 of SEQ ID NO: 14.

[0074] In one embodiment of the polynucleotide of (3) above, the polynucleotide of the bispecific antibody of the present invention is the following polynucleotide: A polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region, comprising a heavy chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acid numbers 1 to 125 of SEQ ID NO: 14, and a light chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acid numbers 146 to 254 of SEQ ID NO: 14, and a second Fc polypeptide are linked.

[0075] In one embodiment of the polynucleotide of (3) above, the polynucleotide of a bispecific antibody of the present invention is a polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region consisting of the amino acid sequence of SEQ ID NO: 14 and a second Fc polypeptide are linked together.

[0076] Those skilled in the art can prepare polynucleotides of bispecific antibodies of the present invention based on their nucleotide sequences using methods known in the art. For example, polynucleotides of bispecific antibodies of the present invention can be synthesized using gene synthesis methods known in the art. Various methods known to those skilled in the art can be used for such gene synthesis, including the antibody gene synthesis method described in WO 90 / 07861.

[0077] <Expression vector for bispecific antibodies of the present invention> The present invention also provides expression vectors containing the polynucleotides of the bispecific antibodies of the present invention described in (1) to (3) below (also referred to as "expression vectors of the bispecific antibodies of the present invention"). These polynucleotides may be contained in separate vectors, or multiple polynucleotides may be contained in a single vector. (1) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which the heavy chain fragment of the anti-TSPAN8 antibody and a first Fc polypeptide are linked; (2) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody; (3) A polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region and a polypeptide comprising a second Fc polypeptide are linked together.

[0078] In one embodiment of the expression vector for a bispecific antibody of the present invention, comprising the polynucleotide (1) above, the expression vector comprises a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody, comprising a heavy chain variable region including CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 6, CDR2 consisting of the amino acid sequence of amino acids 50 to 66 of SEQ ID NO: 6, and CDR3 consisting of the amino acid sequence of amino acids 99 to 110 of SEQ ID NO: 6, and a first Fc polypeptide is linked; (b) A polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody, comprising a heavy chain variable region including CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 10, CDR2 consisting of the amino acid sequence of amino acids 50 to 66 of SEQ ID NO: 10, and CDR3 consisting of the amino acid sequence of amino acids 99 to 110 of SEQ ID NO: 10, is linked to a first Fc polypeptide.

[0079] In one embodiment of the expression vector for a bispecific antibody of the present invention, comprising the polynucleotide (1) above, the expression vector comprises a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody containing a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 6 and a first Fc polypeptide are linked; (b) A polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody containing a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10 and a first Fc polypeptide are linked.

[0080] In one embodiment of the expression vector for a bispecific antibody of the present invention comprising the polynucleotide (1) above, the expression vector comprises a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which the heavy chain fragment of the anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 6 and the first Fc polypeptide are linked.

[0081] In one embodiment of the expression vector for a bispecific antibody of the present invention, comprising the polynucleotide (2) above, the expression vector comprises a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody, the light chain variable region comprising CDR1 consisting of the amino acid sequence of amino acids 24 to 34 of SEQ ID NO: 8, CDR2 consisting of the amino acid sequence of amino acids 50 to 56 of SEQ ID NO: 8, and CDR3 consisting of the amino acid sequence of amino acids 89 to 96 of SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region including CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 12, CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 8, and CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 12.

[0082] In one embodiment of the expression vector for a bispecific antibody of the present invention, comprising the polynucleotide (2) above, the expression vector comprises a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12.

[0083] In one embodiment of the expression vector for a bispecific antibody of the present invention comprising the polynucleotide (2) above, the expression vector comprises a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 8.

[0084] In one embodiment of the expression vector for a bispecific antibody of the present invention comprising the polynucleotide (3) above, the expression vector comprises the following polynucleotide: A polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked, the anti-CD3 antibody heavy chain variable region comprising: CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 101 to 114 of SEQ ID NO: 14; and an anti-CD3 antibody light chain variable region comprising: CDR1 consisting of the amino acid sequence of amino acids 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 236 to 244 of SEQ ID NO: 14.

[0085] In one embodiment of the expression vector for a bispecific antibody of the present invention comprising the polynucleotide (3) above, the expression vector comprises the following polynucleotide: A polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region, comprising a heavy chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acid numbers 1 to 125 of SEQ ID NO: 14, and a light chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acid numbers 146 to 254 of SEQ ID NO: 14, and a second Fc polypeptide are linked.

[0086] In one embodiment of the expression vector for a bispecific antibody of the present invention comprising the polynucleotide (3) above, the expression vector comprises a polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region consisting of the amino acid sequence of SEQ ID NO: 14 and a second Fc polypeptide are linked together.

[0087] In one embodiment, the expression vector for a bispecific antibody of the present invention is an expression vector comprising one or more polynucleotides selected from the following (a) to (e): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody, comprising a heavy chain variable region including CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 6, CDR2 consisting of the amino acid sequence of amino acids 50 to 66 of SEQ ID NO: 6, and CDR3 consisting of the amino acid sequence of amino acids 99 to 110 of SEQ ID NO: 6, and a first Fc polypeptide is linked; (b) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody, the light chain variable region comprising CDR1 consisting of the amino acid sequence of amino acids 24 to 34 of SEQ ID NO: 8, CDR2 consisting of the amino acid sequence of amino acids 50 to 56 of SEQ ID NO: 8, and CDR3 consisting of the amino acid sequence of amino acids 89 to 96 of SEQ ID NO: 8; (c) A polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody, comprising a heavy chain variable region including CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 10, CDR2 consisting of the amino acid sequence of amino acids 50 to 66 of SEQ ID NO: 10, and CDR3 consisting of the amino acid sequence of amino acids 99 to 110 of SEQ ID NO: 10, is linked to a first Fc polypeptide. (d) A polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region including CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 12, CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 8, and CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 12. (e) a polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked, the anti-CD3 antibody heavy chain variable region comprising CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 101 to 114 of SEQ ID NO: 14, and an anti-CD3 antibody light chain variable region comprising CDR1 consisting of the amino acid sequence of amino acids 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 236 to 244 of SEQ ID NO: 14.

[0088] In one embodiment, the expression vector for a bispecific antibody of the present invention is an expression vector comprising one or more polynucleotides selected from the following (a) to (e): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody containing a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 6 and a first Fc polypeptide are linked; (b) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 8; (c) A polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody containing a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10 and a first Fc polypeptide are linked. (d) A polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12. (e) a polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region, which includes a heavy chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acid numbers 1 to 125 of SEQ ID NO: 14 and a light chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acid numbers 146 to 254 of SEQ ID NO: 14, and a second Fc polypeptide are linked.

[0089] In one embodiment, the expression vector for a bispecific antibody of the present invention is an expression vector comprising one or more polynucleotides selected from the following (a) to (c): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 6 and a first Fc polypeptide are linked; (b) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 8; (c) A polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region consisting of the amino acid sequence of SEQ ID NO: 14 and a second Fc polypeptide are linked together.

[0090] The expression vector for the bispecific antibody of the present invention is not particularly limited, as long as it is capable of producing the polynucleotide of the present invention in various host cells, such as eukaryotic cells (e.g., animal cells, insect cells, plant cells, yeast) and / or prokaryotic cells (e.g., Escherichia coli). Examples of such expression vectors include plasmid vectors and viral vectors. Examples of plasmid vectors that can be used include the pcDNA series (Thermo Fisher Scientific), pALTER®-MAX (Promega), pHEK293 Ultra Expression Vector (Takara Bio), pEE6.4, and pEE12.4 (Lonza Biologics). Examples of viral vectors that can be used include lentivirus, adenovirus, retrovirus, and adeno-associated virus. For example, when a lentivirus is used to introduce the polynucleotide of the present invention into cells, the lentivirus may be the pLVSIN-CMV / EF1α vector (Takara Bio), pLenti vector (Thermo Fisher Scientific), or the like. In one embodiment, the vector used for the expression vector for the bispecific antibody of the present invention is pcDNA. TM 3.4-TOPO® (Thermo Fisher Scientific) and pcDNA TM 3.1 (Thermo Fisher Scientific).

[0091] Expression vectors for bispecific antibodies of the present invention may contain a promoter operably linked to a bispecific antibody polynucleotide of the present invention. Examples of promoters for expressing bispecific antibody polynucleotides of the present invention in animal cells include promoters derived from viruses such as CMV, RSV, and SV40, actin promoters, elongation factor (EF) 1α promoters, and heat shock promoters. Examples of promoters for expressing bispecific antibody polynucleotides of the present invention in bacteria (e.g., Escherichia) include the trp promoter, lac promoter, λPL promoter, and tac promoter. Examples of promoters for expressing bispecific antibody polynucleotides of the present invention in yeast include the GAL1 promoter, GAL10 promoter, PH05 promoter, PGK promoter, GAP promoter, and ADH promoter.

[0092] When animal cells, insect cells, or yeast cells are used as host cells, expression vectors for bispecific antibodies of the present invention may contain an initiation codon and a stop codon. In this case, they may also contain an enhancer sequence, 5'- and 3'-untranslated regions of the gene encoding the antibody of the present invention or its heavy chain or light chain, a secretion signal sequence, a splicing junction, a polyadenylation site, or a replication unit. When Escherichia coli is used as a host cell, the expression vectors of the present invention may contain an initiation codon, a stop codon, a terminator region, and a replication unit. The expression vectors of the present invention may also contain a commonly used drug selection marker gene (e.g., a tetracycline resistance gene, an ampicillin resistance gene, a kanamycin resistance gene, a neomycin resistance gene, or a dihydrofolate reductase gene) depending on the purpose.

[0093] <Transformed host cells of the present invention> The present invention also provides host cells transformed with an expression vector for a bispecific antibody of the present invention (also referred to as "transformed host cells of the present invention"). The transformed host cells of the present invention may contain one or more of the polynucleotides of the bispecific antibody of the present invention shown in (1) to (3) below, upon transformation with the expression vector for a bispecific antibody of the present invention. In one embodiment, the transformed host cells of the present invention contain all of the following polynucleotides (1) to (3): (1) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which the heavy chain fragment of the anti-TSPAN8 antibody and a first Fc polypeptide are linked; (2) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody; (3) A polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region and a polypeptide comprising a second Fc polypeptide are linked together.

[0094] The host cell to be transformed is not particularly limited, as long as it is compatible with the expression vector used and can be transformed with the expression vector to express an antibody or fusion protein. Examples of host cells to be transformed include various cells, such as conventional cells commonly used in the technical field of the present invention or artificially established cells (e.g., animal cells (e.g., CHO-K1 cells, ExpiCHO-S (registered trademark) cells, CHOK1SV cells, CHO-DG44 cells, HEK293 cells, NS0 cells), insect cells (e.g., Sf9), bacteria (e.g., Escherichia), yeast (e.g., Saccharomyces, Pichia)). In one embodiment, the host cell of the present invention is a CHO-K1 cell or an ExpiCHO-S cell.

[0095] The method for transforming host cells is not particularly limited, and methods commonly used by those skilled in the art, such as the calcium phosphate method, electroporation, or lipofection, can be used.

[0096] Transformed host cells can be selected by methods commonly used by those skilled in the art, such as a drug selection method using a drug selection marker gene and a drug such as tetracycline, ampicillin, neomycin, or hygromycin, or a cell isolation method such as limiting dilution, single cell sorting, or colony picking.

[0097] In one embodiment of the transformed host cell of the present invention comprising the polynucleotide (1) above, the host cell comprises a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody, comprising a heavy chain variable region including CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 6, CDR2 consisting of the amino acid sequence of amino acids 50 to 66 of SEQ ID NO: 6, and CDR3 consisting of the amino acid sequence of amino acids 99 to 110 of SEQ ID NO: 6, and a first Fc polypeptide is linked; (b) A polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody, comprising a heavy chain variable region including CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 10, CDR2 consisting of the amino acid sequence of amino acids 50 to 66 of SEQ ID NO: 10, and CDR3 consisting of the amino acid sequence of amino acids 99 to 110 of SEQ ID NO: 10, is linked to a first Fc polypeptide.

[0098] In one embodiment of the transformed host cell of the present invention comprising the polynucleotide (1) above, the host cell comprises a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody containing a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 6 and a first Fc polypeptide are linked; (b) A polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody containing a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10 and a first Fc polypeptide are linked.

[0099] In one embodiment of the transformed host cell of the present invention comprising the polynucleotide (1) above, the host cell comprises a polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 6 and a first Fc polypeptide are linked.

[0100] In one embodiment of the transformed host cell of the present invention comprising the polynucleotide (2) above, the host cell comprises a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody, the light chain variable region comprising CDR1 consisting of the amino acid sequence of amino acids 24 to 34 of SEQ ID NO: 8, CDR2 consisting of the amino acid sequence of amino acids 50 to 56 of SEQ ID NO: 8, and CDR3 consisting of the amino acid sequence of amino acids 89 to 96 of SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region including CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 12, CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 8, and CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 12.

[0101] In one embodiment of the transformed host cell of the present invention comprising the polynucleotide (2) above, the host cell comprises a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding an anti-TSPAN8 antibody light chain comprising a light chain variable region consisting of the amino acid sequence of amino acid numbers 1 to 107 of SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12.

[0102] In one embodiment of the transformed host cell of the present invention comprising the polynucleotide (2) above, the host cell comprises a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 8.

[0103] In one embodiment of the transformed host cell of the present invention comprising the polynucleotide (3) above, the host cell comprises the following polynucleotide: A polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked, the anti-CD3 antibody heavy chain variable region comprising: CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 101 to 114 of SEQ ID NO: 14; and an anti-CD3 antibody light chain variable region comprising: CDR1 consisting of the amino acid sequence of amino acids 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 236 to 244 of SEQ ID NO: 14.

[0104] In one embodiment of the transformed host cell of the present invention comprising the polynucleotide (3) above, the host cell comprises the following polynucleotide: A polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region containing a heavy chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acid numbers 1 to 125 of SEQ ID NO: 14 and a light chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acid numbers 146 to 254 of SEQ ID NO: 14, and a second Fc polypeptide are linked.

[0105] In one embodiment of the transformed host cell of the present invention comprising the polynucleotide (3) above, the host cell comprises a polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv consisting of the amino acid sequence of SEQ ID NO: 14 and a second Fc polypeptide are linked together.

[0106] In one embodiment, the transformed host cell of the present invention is a host cell comprising one or more polynucleotides selected from the following (a) to (e): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody, comprising a heavy chain variable region including CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 6, CDR2 consisting of the amino acid sequence of amino acids 50 to 66 of SEQ ID NO: 6, and CDR3 consisting of the amino acid sequence of amino acids 99 to 110 of SEQ ID NO: 6, and a first Fc polypeptide is linked; (b) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody, the light chain variable region comprising CDR1 consisting of the amino acid sequence of amino acids 24 to 34 of SEQ ID NO: 8, CDR2 consisting of the amino acid sequence of amino acids 50 to 56 of SEQ ID NO: 8, and CDR3 consisting of the amino acid sequence of amino acids 89 to 96 of SEQ ID NO: 8; (c) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody, comprising a heavy chain variable region including CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 10, CDR2 consisting of the amino acid sequence of amino acids 50 to 66 of SEQ ID NO: 10, and CDR3 consisting of the amino acid sequence of amino acids 99 to 110 of SEQ ID NO: 10, and a first Fc polypeptide is linked; (d) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody, the light chain variable region comprising CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 12, CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 8, and CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 12; (e) a polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked, the anti-CD3 antibody heavy chain variable region comprising CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 101 to 114 of SEQ ID NO: 14, and an anti-CD3 antibody light chain variable region comprising CDR1 consisting of the amino acid sequence of amino acids 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 236 to 244 of SEQ ID NO: 14.

[0107] In one embodiment, the transformed host cell of the present invention is a host cell comprising one or more polynucleotides selected from the following (a) to (e): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody containing a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 6 and a first Fc polypeptide are linked; (b) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 8; (c) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of an anti-TSPAN8 antibody comprising a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10 and a first Fc polypeptide are linked; (d) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody, which comprises a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12; (e) a polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv containing a heavy chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acid numbers 1 to 125 of SEQ ID NO: 14 and a light chain variable region of an anti-CD3 antibody consisting of the amino acid sequence from amino acid numbers 146 to 254 of SEQ ID NO: 14 is linked to a second Fc polypeptide.

[0108] In one embodiment, the transformed host cell of the present invention is a host cell comprising a polynucleotide selected from the following (a) to (c): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 6 and a first Fc polypeptide are linked; (b) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 8; (c) A polynucleotide comprising a nucleotide sequence encoding a polypeptide in which an anti-CD3-scFv consisting of the amino acid sequence of SEQ ID NO: 14 and a second Fc polypeptide are linked together.

[0109] In one embodiment, the transformed host cell of the present invention comprises a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 6, a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence of SEQ ID NO: 8, and a polynucleotide comprising a nucleotide sequence encoding a polypeptide consisting of the amino acid sequence of SEQ ID NO: 14.

[0110] <Method for producing bispecific antibodies of the present invention> The present invention also provides methods for producing the anti-TSPAN8-anti-CD3 bispecific antibodies of the present invention (also referred to as "production methods of the present invention"). The production methods of the present invention may include, for example, culturing the transformed host cells described in <Transformed host cells of the present invention> and expressing the antibody in the cells or the culture supernatant, and methods for recovering, isolating, and purifying the antibody. However, the production methods of the present invention are not limited to these methods as long as the anti-TSPAN8-anti-CD3 bispecific antibodies of the present invention are produced.

[0111] The transformed host cells of the present invention can be cultured by known methods. Culture conditions, such as temperature, medium pH, and culture time, can be appropriately selected by those skilled in the art. When the host cells are animal cells, examples of media that can be used include MEM medium containing about 5 to 20% fetal bovine serum (Science, 1959, Vol. 130, pp. 432-437), DMEM medium (Virol., 1959, Vol. 8, p. 396), RPMI-1640 medium (J. Am. Med. Assoc., 1967, Vol. 199, p. 519), and 199 medium (Exp. Biol. Med., 1950, Vol. 73, pp. 1-8). The pH of the medium is, for example, about 6 to 8, and the culture is typically performed at about 30 to 40°C for about 15 to 336 hours, with aeration and stirring as necessary. When the host cells are insect cells, the culture medium can be, for example, Grace's medium containing fetal bovine serum (Proc. Natl. Acad. Sci. USA., 1985, Vol. 82, p. 8404). The pH of the medium is, for example, about 5 to 8, and the culture is typically performed at about 20 to 40°C for about 15 to 100 hours, with aeration or stirring as necessary. When the host cells are Escherichia coli or yeast, the culture medium can be, for example, a liquid medium containing nutrient sources. The nutrient medium contains, for example, a carbon source, inorganic nitrogen source, or organic nitrogen source necessary for the growth of the transformed host cells. Carbon sources include, for example, glucose, dextran, soluble starch, and sucrose. Inorganic and organic nitrogen sources include, for example, ammonium salts, nitrates, amino acids, corn steep liquor, peptone, casein, meat extract, soybean meal, and potato extract. If desired, the medium may contain other nutrients (for example, inorganic salts (e.g., calcium chloride, sodium dihydrogen phosphate, magnesium chloride), vitamins), antibiotics (e.g., tetracycline, neomycin, ampicillin, kanamycin), etc. The pH of the medium is, for example, about 5 to 8. When the host cell is Escherichia coli, the medium may be, for example, LB medium, M9 medium (Molecular Cloning, Cold Spring Harbor Laboratory, Vol. 3, A2.2), etc.Culturing is typically carried out at about 14 to 43°C for about 3 to 24 hours, with aeration and stirring as necessary. When the host cells are yeast, a medium such as Burkholder's minimal medium (Proc. Natl. Acad. Sci. USA., 1980, Vol. 77, p. 4505) can be used. Culturing is typically carried out at about 20 to 35°C for about 14 to 144 hours, with aeration and stirring as necessary. The anti-TSPAN8-anti-CD3 bispecific antibody of the present invention can be expressed by culturing as described above.

[0112] The production method of the present invention can include not only the step of culturing the transformed host cells of the present invention to express the anti-TSPAN8-anti-CD3 bispecific antibody, but also the step of recovering, isolating, or purifying the anti-TSPAN8-anti-CD3 bispecific antibody from the transformed host cells. Examples of isolation or purification methods include methods that utilize solubility (e.g., salting out or solvent precipitation); methods that utilize differences in molecular weight (e.g., dialysis, ultrafiltration, or gel filtration); methods that utilize charge (e.g., ion exchange chromatography or hydroxylapatite chromatography); methods that utilize specific affinity (e.g., affinity chromatography); methods that utilize differences in hydrophobicity (e.g., reverse-phase high-performance liquid chromatography); and methods that utilize differences in isoelectric point (e.g., isoelectric focusing). In one embodiment, the antibody secreted into the culture supernatant can be purified by various types of chromatography, for example, column chromatography using a protein A column or a protein G column.

[0113] The anti-TSPAN8-anti-CD3 bispecific antibody of the present invention also includes an anti-TSPAN8-anti-CD3 bispecific antibody produced by the production method of the present invention.

[0114] <Pharmaceutical compositions of bispecific antibodies of the present invention> Pharmaceutical compositions of the present invention include those containing an anti-TSPAN8-anti-CD3 bispecific antibody of the present invention and a pharmaceutically acceptable excipient. Pharmaceutical compositions of the present invention can be prepared by commonly used methods using excipients commonly used in the art, i.e., pharmaceutical excipients and pharmaceutical carriers. Examples of dosage forms of these pharmaceutical compositions include parenteral preparations such as injections and infusions, which can be administered intravenously, subcutaneously, or intraperitoneally. When formulating the compositions, excipients, carriers, additives, and the like appropriate for these dosage forms can be used within pharmaceutically acceptable limits.

[0115] The pharmaceutical compositions of the present invention may contain post-translationally modified forms of the anti-TSPAN8-anti-CD3 bispecific antibodies of the present invention. For example, pharmaceutical compositions containing antibodies that have undergone C-terminal lysine deletion and / or N-terminal pyroglutamylation are also included in the present invention.

[0116] In one embodiment, the pharmaceutical composition of the present invention comprises an anti-TSPAN8-anti-CD3 bispecific antibody of the present invention selected from the following (a) and (b) and / or a post-translationally modified form of the antibody: (a) a heavy chain of an anti-TSPAN8 antibody in which a first Fc polypeptide is linked to a heavy chain fragment of an anti-TSPAN8 antibody, the heavy chain fragment comprising a heavy chain variable region comprising CDR1 consisting of the amino acid sequence from 31 to 35 of SEQ ID NO: 6, CDR2 consisting of the amino acid sequence from 50 to 66 of SEQ ID NO: 6, and CDR3 consisting of the amino acid sequence from 99 to 110 of SEQ ID NO: 6; a light chain of an anti-TSPAN8 antibody comprising a light chain variable region comprising CDR1 consisting of the amino acid sequence from 24 to 34 of SEQ ID NO: 8, CDR2 consisting of the amino acid sequence from 50 to 56 of SEQ ID NO: 8, and CDR3 consisting of the amino acid sequence from 89 to 96 of SEQ ID NO: 8; and ... a bispecific antibody comprising a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked, the anti-CD3-scFv region comprising a heavy chain variable region of an anti-CD3 antibody comprising CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 101 to 114 of SEQ ID NO: 14, and a light chain variable region of an anti-CD3 antibody comprising CDR1 consisting of the amino acid sequence of amino acids 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 236 to 244 of SEQ ID NO: 14; (b) a heavy chain of an anti-TSPAN8 antibody in which a first Fc polypeptide is linked to a heavy chain fragment of an anti-TSPAN8 antibody, the heavy chain fragment comprising a heavy chain variable region comprising CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 10, CDR2 consisting of the amino acid sequence of amino acids 50 to 66 of SEQ ID NO: 10, and CDR3 consisting of the amino acid sequence of amino acids 99 to 110 of SEQ ID NO: 10; and a light chain of an anti-TSPAN8 antibody comprising a light chain variable region comprising CDR1 consisting of the amino acid sequence of amino acids 24 to 34 of SEQ ID NO: 12, CDR2 consisting of the amino acid sequence of amino acids 50 to 56 of SEQ ID NO: 12, and CDR3 consisting of the amino acid sequence of amino acids 89 to 96 of SEQ ID NO: 12; and a heavy chain variable region of an anti-CD3 antibody comprising CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 101 to 114 of SEQ ID NO: 14; and a light chain variable region of an anti-CD3 antibody comprising CDR1 consisting of the amino acid sequence of amino acids 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 236 to 244 of SEQ ID NO: 14, linked to a second Fc polypeptide.

[0117] In one embodiment, the pharmaceutical composition of the present invention comprises an anti-TSPAN8-anti-CD3 bispecific antibody of the present invention selected from the following (a) and (b) and / or a post-translationally modified form of the antibody: (a) a bispecific antibody comprising: a heavy chain of an anti-TSPAN8 antibody in which a first Fc polypeptide is linked to a heavy chain fragment of the anti-TSPAN8 antibody, the heavy chain fragment comprising a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 121 of SEQ ID NO: 6; a light chain of an anti-TSPAN8 antibody in which a light chain variable region consisting of the amino acid sequence of amino acids 1 to 107 of SEQ ID NO: 8; and a polypeptide in which a second Fc polypeptide is linked to an anti-CD3-scFv region in which a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 125 of SEQ ID NO: 14 and a light chain variable region consisting of the amino acid sequence of amino acids 146 to 254 of SEQ ID NO: 14 are linked to an anti-CD3-scFv region; (b) A bispecific antibody comprising: a heavy chain of an anti-TSPAN8 antibody in which a heavy chain fragment of the anti-TSPAN8 antibody, which comprises a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 121 of SEQ ID NO: 10, and a first Fc polypeptide are linked; a light chain of an anti-TSPAN8 antibody in which a light chain variable region consisting of the amino acid sequence of amino acids 1 to 107 of SEQ ID NO: 12, and a polypeptide in which an anti-CD3-scFv region, which comprises a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 125 of SEQ ID NO: 14 and a light chain variable region consisting of the amino acid sequence of amino acids 146 to 254 of SEQ ID NO: 14, and a second Fc polypeptide are linked.

[0118] In one embodiment, the pharmaceutical composition of the present invention is a pharmaceutical composition comprising an anti-TSPAN8-anti-CD3 bispecific antibody and / or a post-translationally modified form of the antibody, the anti-TSPAN8 antibody heavy chain comprising a heavy chain fragment of the anti-TSPAN8 antibody having the amino acid sequence of SEQ ID NO: 6 linked to a first Fc polypeptide, an anti-TSPAN8 antibody light chain comprising the amino acid sequence of SEQ ID NO: 8, and a polypeptide comprising an anti-CD3-scFv region having the amino acid sequence of SEQ ID NO: 14 linked to a second Fc polypeptide.

[0119] The amount of the anti-TSPAN8-anti-CD3 bispecific antibody or anti-TSPAN8 antibody of the present invention added in the formulation varies depending on the severity of symptoms and age of the patient, the dosage form of the formulation used, or the binding titer of the antibody, but can be, for example, approximately 0.001 mg / kg to 100 mg / kg.

[0120] <Medicinal uses of the anti-TSPAN8-anti-CD3 bispecific antibody of the present invention> The anti-TSPAN8-anti-CD3 bispecific antibodies of the present invention and pharmaceutical compositions containing them can be used to treat cancer. The present invention also includes a method for treating cancer, comprising administering a therapeutically effective amount of an anti-TSPAN8-anti-CD3 bispecific antibody of the present invention to a subject. The present invention also includes an anti-TSPAN8-anti-CD3 bispecific antibody of the present invention for use in cancer treatment. The present invention also includes use of an anti-TSPAN8-anti-CD3 bispecific antibody of the present invention in the manufacture of a pharmaceutical composition for cancer treatment. Cancers that can be treated by the present invention are not particularly limited, but include, for example, various peritoneal disseminated cancers, gastric cancer, lung cancer, blood cancers such as acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), Hodgkin's lymphoma, non-Hodgkin's lymphoma, B-cell lymphoma, multiple myeloma, and T-cell lymphoma, myelodysplastic syndrome, adenocarcinoma, squamous cell carcinoma, adenosquamous carcinoma, undifferentiated carcinoma, large cell carcinoma, non-small cell lung cancer, small cell lung cancer, mesothelioma, skin cancer, cutaneous T-cell lymphoma, breast cancer, prostate cancer, bladder cancer, vaginal cancer, cervical cancer, head and neck cancer, and other cancers. Examples of cancers include solid cancers such as cervical cancer, uterine cancer, cervical cancer, liver cancer, gallbladder cancer, bile duct cancer, kidney cancer, pancreatic cancer, colon cancer, colorectal cancer, rectal cancer, small intestine cancer, stomach cancer, esophageal cancer, testicular cancer, ovarian cancer, and brain tumors, as well as cancers of bone tissue, cartilage tissue, adipose tissue, muscle tissue, vascular tissue, and hematopoietic tissue, as well as sarcomas such as chondrosarcoma, Ewing's sarcoma, malignant hemangioendothelioma, malignant schwannoma, osteosarcoma, and soft tissue sarcoma, and blastomas such as glioblastoma, glioblastoma multiforme, hepatoblastoma, medulloblastoma, nephroblastoma, neuroblastoma, pancreatoblastoma, pleuropulmonary blastoma, and retinoblastoma.

[0121] <Anti-TSPAN8 antibody of the present invention> The present invention also provides novel anti-TSPAN8 antibodies or antigen-binding fragments thereof directed against human TSPAN8, as described below. The anti-TSPAN8 antibodies or antigen-binding fragments thereof provided by the present invention are collectively referred to as "anti-TSPAN8 antibodies or antigen-binding fragments thereof of the present invention."

[0122] The present invention provides anti-TSPAN8 antibodies or binding fragments thereof that selectively bind to human TSPAN8-expressing cancer cells.

[0123] As used herein, the phrase "selectively binds to human TSPAN8-expressing cancer cells" refers to the binding activity of an anti-TSPAN8 antibody to TSPAN8 expressed in human TSPAN8-expressing cancer cells, which is at least three times, preferably at least five times, and more preferably at least ten times, of the commercially available anti-TSPAN8 antibodies, when compared with those of commercially available anti-TSPAN8 antibodies (e.g., TAL69, REA443, etc.) or anti-TSPAN8 antibodies that exhibit a similar binding profile to the commercially available anti-TSPAN8 antibodies (e.g., 9F6, 18C10, etc.), and its binding strength to TSPAN8 expressed in normal cells is at most one-third, preferably at most one-fifth, and more preferably at most one-tenth. The binding strength of an antibody to cells can be calculated, for example, using the MFI (Mean Fluorescence Intensity) value obtained by flow cytometry as shown in Example 1, or the ΔMFI value calculated by subtracting the MFI of each antibody isotype from the MFI of each antibody. Furthermore, the binding strength of an antibody to a cell can also be measured and calculated by methods commonly used by those skilled in the art, such as ELISA using cancer cells and normal cells. Here, human TSPAN8-expressing cancer cells refer to cancer cells that express human TSPAN8 isolated from a cancer patient, and in addition to the patient-derived cancer peritoneal dissemination cells described in Example 1-1, cancer cell lines available from cell banks such as the American Type Culture Collection (ATCC) can be used. Examples of cancer cell lines that can be used include cell lines established from patient ascites using the method described in Example 1-1, as well as TSPAN8-expressing cell lines such as AGS, KATOIII, SNU5, SNU16, SNU520, ANU719, NCI-N87, HT-29, LoVo, GP2d, AsPC-1, OE19, Li-7Hs746, NUGC-4, OCUM1, and MNK45. Furthermore, normal cells refer to cells derived from normal tissues, and in addition to the patient-derived peritoneal mesothelial cells used in Examples 1-5, commercially available primary cultured cells or cell lines such as human peripheral blood mononuclear cells used in Examples 1-3 and cultured peritoneal mesothelial cells used in Examples 1-3 can be used. In one embodiment, the normal cells express TSPAN8. In one embodiment, the normal cells expressing TSPAN8 are patient-derived peritoneal mesothelial cells and the cultured peritoneal mesothelial cells used in Examples 1-3. In one embodiment, the normal cells are cells that do not express TSPAN8. In one embodiment, the normal cells that do not express TSPAN8 are the human peripheral blood mononuclear cells used in Examples 1-3.

[0124] The present invention also provides an anti-TSPAN8 antibody or antigen-binding fragment thereof that recognizes a portion of the TSPAN8 protein as an epitope. In one embodiment, the epitope is a structural unit consisting of an amino acid sequence contained in the LEL region of TSPAN8. In one embodiment, the epitope is a structural unit consisting of a portion of the TSPAN8 protein represented by the amino acid sequence from 126 to 155 of SEQ ID NO: 2. In one embodiment, the epitope is a structural unit consisting of one or more amino acid sequences contained in a portion of the TSPAN8 protein represented by the amino acid sequence from 126 to 155 of SEQ ID NO: 2. In one embodiment, the epitope is a structural unit comprising at least amino acid 131 of SEQ ID NO: 2.

[0125] In one embodiment, the anti-TSPAN8 antibody or antigen-binding fragment thereof of the present invention binds to at least one amino acid present in the region of human TSPAN8 from amino acid numbers 126 to 155 of SEQ ID NO: 2. In one embodiment, the anti-TSPAN8 antibody or antigen-binding fragment thereof of the present invention binds to at least one amino acid present in the region of human TSPAN8 from amino acid numbers 126 to 155 of SEQ ID NO: 2 and selectively binds to human TSPAN8-expressing cancer cells.

[0126] In one embodiment, the anti-TSPAN8 antibody or antigen-binding fragment thereof of the present invention binds to at least amino acid 131 of SEQ ID NO: 2, which is located in the region of human TSPAN8 from amino acid 126 to 155 of SEQ ID NO: 2. In one embodiment, the anti-TSPAN8 antibody or antigen-binding fragment thereof binds to at least amino acid 131 of SEQ ID NO: 2, which is located in the region of human TSPAN8 from amino acid 126 to 155 of SEQ ID NO: 2, and selectively binds to human TSPAN8-expressing cancer cells.

[0127] Whether an anti-TSPAN8 antibody or its antigen-binding fragment binds to amino acids present in the region of human TSPAN8 from amino acid numbers 126 to 155 of SEQ ID NO: 2 (e.g., amino acid number 131 of SEQ ID NO: 2) can be confirmed using the epitope identification method described in Examples 4-1 and 4-2 of the present application.

[0128] The present invention also provides anti-TSPAN8 antibodies or antigen-binding fragments thereof shown in (a) and (b) below: (a) an anti-TSPAN8 antibody or an antigen-binding fragment thereof comprising: a heavy chain variable region comprising CDR1 consisting of the amino acid sequence from 31 to 35 of SEQ ID NO: 4, a CDR2 consisting of the amino acid sequence from 50 to 66 of SEQ ID NO: 4, and a CDR3 consisting of the amino acid sequence from 99 to 110 of SEQ ID NO: 4; and a light chain variable region comprising CDR1 consisting of the amino acid sequence from 24 to 34 of SEQ ID NO: 8, a CDR2 consisting of the amino acid sequence from 50 to 56 of SEQ ID NO: 8, and a CDR3 consisting of the amino acid sequence from 89 to 96 of SEQ ID NO: 8; (b) An anti-TSPAN8 antibody or an antigen-binding fragment thereof comprising: a heavy chain variable region comprising CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 10, CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 of SEQ ID NO: 10, and CDR3 consisting of the amino acid sequence from amino acid numbers 99 to 110 of SEQ ID NO: 10; and a light chain variable region comprising CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 12, CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 12, and CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 12.

[0129] In one embodiment, the anti-TSPAN8 antibody or antigen-binding fragment thereof of the present invention is an anti-TSPAN8 antibody or antigen-binding fragment thereof selected from the following (a) and (b): (a) an anti-TSPAN8 antibody or an antigen-binding fragment thereof, comprising a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 121 of SEQ ID NO: 4 and a light chain variable region consisting of the amino acid sequence of amino acids 1 to 107 of SEQ ID NO: 8; (b) An anti-TSPAN8 antibody or an antigen-binding fragment thereof, comprising a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10, and a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12.

[0130] The heavy chain constant region of the anti-TSPAN8 antibody of the present invention can be any of Igγ, Igμ, Igα, Igδ, or Igε constant regions. Igγ can be selected from, for example, Igγ1, Igγ2, Igγ3, or Igγ4. In one embodiment, the heavy chain constant region is an Igγ1 constant region, for example, a human Igγ1 constant region. The heavy chain constant region of the anti-TSPAN8 antibody of the present invention may also contain amino acid mutations such as the LALA mutation to reduce ADCC or CDC. The light chain constant region of the anti-TSPAN8 antibody of the present invention can be any of Igλ or Igκ constant regions. In one embodiment, the light chain constant region is an Igκ constant region, for example, a human Igκ constant region.

[0131] In one embodiment, the antigen-binding fragment of an anti-TSPAN8 antibody of the invention is an scFv, Fab, Fab', F(ab')2, or single-arm antibody.

[0132] In one embodiment, the anti-TSPAN8 antibody of the present invention is an anti-TSPAN8 antibody selected from the following (a) and (b): (a) an anti-TSPAN8 antibody comprising a heavy chain consisting of the amino acid sequence of SEQ ID NO: 4 and a light chain consisting of the amino acid sequence of SEQ ID NO: 8; (b) An anti-TSPAN8 antibody comprising a heavy chain consisting of the amino acid sequence of SEQ ID NO: 10 and a light chain consisting of the amino acid sequence of SEQ ID NO: 12.

[0133] The present invention also provides anti-TSPAN8 antibodies or antigen-binding fragments thereof shown in (a) to (d) below (collectively, particularly also referred to as "competitive anti-TSPAN8 antibodies of the present invention"). (a) an anti-TSPAN8 antibody or an antigen-binding fragment thereof that competes with an anti-TSPAN8 antibody comprising a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 121 of SEQ ID NO: 4 and a light chain variable region consisting of the amino acid sequence of amino acids 1 to 107 of SEQ ID NO: 8 for binding to human TSPAN8-expressing cancer cells; (b) an anti-TSPAN8 antibody or an antigen-binding fragment thereof that competes with an anti-TSPAN8 antibody comprising a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 121 of SEQ ID NO: 10 and a light chain variable region consisting of the amino acid sequence of amino acids 1 to 107 of SEQ ID NO: 12 for binding to human TSPAN8-expressing cancer cells; (c) an anti-TSPAN8 antibody or an antigen-binding fragment thereof that competes with an anti-TSPAN8 antibody comprising a heavy chain variable region consisting of the amino acid sequence of SEQ ID NO: 34 and a light chain variable region consisting of the amino acid sequence of SEQ ID NO: 36 for binding to human TSPAN8-expressing cancer cells; (d) An anti-TSPAN8 antibody or its antigen-binding fragment that competes with an anti-TSPAN8 antibody comprising a heavy chain variable region consisting of the amino acid sequence of SEQ ID NO: 35 and a light chain variable region consisting of the amino acid sequence of SEQ ID NO: 37 for binding to human TSPAN8-expressing cancer cells.

[0134] The competitive anti-TSPAN8 antibodies of the present invention can be obtained by those skilled in the art, for example, by obtaining an antibody against human TSPAN8 using known antibody production techniques with human TSPAN8-expressing cells as antigens, and then conducting a competition test of the resulting antibody against a competing anti-TSPAN8 antibody for binding to human TSPAN8-expressing cells. The competition test can be performed using methods known to those skilled in the art, such as flow cytometry. For example, the competition test using human TSPAN8-expressing cancer cells described in Example 4-3 can be performed. The various cells described above can be used as human TSPAN8-expressing cancer cells for the competition test.

[0135] In one embodiment, the anti-TSPAN8 antibody or antigen-binding fragment thereof of the present invention is an anti-TSPAN8 antibody or antigen-binding fragment thereof selected from either (a) or (b) below: (a) an anti-TSPAN8 antibody or an antigen-binding fragment thereof that competes with an anti-TSPAN8 antibody comprising a heavy chain consisting of the amino acid sequence of SEQ ID NO: 4 and a light chain consisting of the amino acid sequence of SEQ ID NO: 8 for binding to human TSPAN8-expressing cancer cells; (b) An anti-TSPAN8 antibody or its antigen-binding fragment that competes with an anti-TSPAN8 antibody comprising a heavy chain consisting of the amino acid sequence of SEQ ID NO: 10 and a light chain consisting of the amino acid sequence of SEQ ID NO: 12 for binding to human TSPAN8-expressing cancer cells.

[0136] In one embodiment, the anti-TSPAN8 antibody or antigen-binding fragment thereof of the present invention is an anti-TSPAN8 antibody or antigen-binding fragment thereof selected from either (c) or (d) below: (c) an anti-TSPAN8 antibody or an antigen-binding fragment thereof that competes with an anti-TSPAN8 antibody comprising a heavy chain variable region consisting of the amino acid sequence of SEQ ID NO: 34 and a light chain variable region consisting of the amino acid sequence of SEQ ID NO: 36 for binding to human TSPAN8-expressing cancer cells; (d) An anti-TSPAN8 antibody or its antigen-binding fragment that competes with an anti-TSPAN8 antibody comprising a heavy chain variable region consisting of the amino acid sequence of SEQ ID NO: 35 and a light chain variable region consisting of the amino acid sequence of SEQ ID NO: 37 for binding to human TSPAN8-expressing cancer cells.

[0137] <Other bispecific antibodies of the present invention> The present invention provides bispecific antibodies comprising the anti-TSPAN8 antibody of the present invention or an antigen-binding fragment thereof linked to an antibody or antigen-binding fragment thereof that is directed against a surface antigen of T cells or natural killer (NK) cells. The shape of the bispecific antibody is not particularly limited, and can take any shape that is commonly used by those skilled in the art, such as the various antibody shapes described in Non-Patent Documents 3 and 4.

[0138] The present invention also provides the following bispecific antibodies: A bispecific antibody that binds to TSPAN8 and a surface antigen of a T cell or an NK cell, (a) a Fab region of an anti-TSPAN8 antibody, which comprises a heavy chain fragment containing the heavy chain variable region of the anti-TSPAN8 antibody of the present invention and a light chain fragment containing the light chain variable region of the anti-TSPAN8 antibody of the present invention; (b) an scFv region of an antibody against a surface antigen of a T cell or NK cell, comprising a heavy chain variable region and a light chain variable region of the antibody against the surface antigen of a T cell or NK cell; and (c) an Fc region consisting of a first Fc polypeptide linked to a heavy chain fragment of the Fab region of (a) and a second Fc polypeptide linked to the scFv region of (b); A bispecific antibody comprising:

[0139] Other bispecific antibodies of the present invention can be prepared by those skilled in the art using the method described in Non-Patent Document 4 or general methods with reference to the description in <Anti-TSPAN8-anti-CD3 bispecific antibodies of the present invention>. Many antibodies against surface antigens of T cells or NK cells are currently known (Current Opinion in Biotechnology, 2020, Vol. 65, pp. 9-16), and the sequence information of these antibodies can be used. Embodiments of other bispecific antibodies of the present invention are as described in <Anti-TSPAN8-anti-CD3 bispecific antibodies of the present invention>, except that the scFv region of an anti-CD3 antibody is used. Other bispecific antibodies of the present invention can also be used to treat cancer.

[0140] In one embodiment, the antibody or antigen-binding fragment thereof directed against a surface antigen on T cells or NK cells of another bispecific antibody of the present invention is an antibody or antigen-binding fragment thereof directed against a surface antigen on T cells or NK cells. In one embodiment, the antibody or antigen-binding fragment thereof directed against a surface antigen on T cells or NK cells is an anti-CD3 antibody, an anti-CD137 antibody, an anti-PD-1 (Programmed Cell Death-1) antibody, an anti-PD-L1 (Programmed Cell Death 1-Ligand 1) antibody, an anti-TIGIT (T cell immunoreceptor with Ig and ITIM domains) antibody, an anti-CD16 antibody, an anti-NKG2D (Natural Killer Group 2, member D) antibody, or an antigen-binding fragment thereof. In one embodiment, the antibody or antigen-binding fragment thereof directed against a surface antigen on T cells or NK cells is an anti-CD3 antibody or antigen-binding fragment thereof. In one embodiment, the antigen-binding fragment of the anti-CD3 antibody is an scFv of the anti-CD3 antibody.

[0141] <Fusions and conjugates of the present invention, and cells expressing the anti-TSPAN8 antibodies or antigen-binding fragments thereof on the cell surface> The present invention also provides anti-TSPAN8 antibodies or antigen-binding fragments thereof linked to proteins (including antibodies) or polypeptides other than TSPAN8 (also referred to as "fusions of the present invention"). The proteins or polypeptides used in the fusions of the present invention are not particularly limited, and examples include various antibodies, cytokines, chemokines, human serum albumin, various tag peptides, artificial helix motif peptides, maltose-binding protein, glutathione S-transferase, and other peptides or proteins that can promote multimerization. In one embodiment of the fusions of the present invention, a protein or polypeptide is linked to the anti-TSPAN8 antibody or antigen-binding fragment thereof. In one embodiment, the protein or polypeptide used in the fusions of the present invention may be, for example, an antibody or antigen-binding fragment thereof directed against a surface antigen of immune cells such as granulocytes or natural killer T (NKT) cells, or blood cells such as dendritic cells or macrophages, or a polypeptide that activates immune cells, such as various interleukins (e.g., IL-2, IL-7, IL-12, IL-15). In this case, the protein or polypeptide used in the fusion of the present invention may be directly linked to the anti-TSPAN8 antibody or its antigen-binding fragment of the present invention, or may be linked via any linker (e.g., a peptide linker).

[0142] The present invention also provides anti-TSPAN8 antibodies of the present invention or antigen-binding fragments thereof (also referred to as "conjugates of the present invention") conjugated with carbohydrates, lipids, metals (including radioisotopes), organic compounds (including toxins, near-infrared fluorescent dyes, chelating agents), etc. (also referred to as "modifying agents"). As used herein, "modifying agents" refer to non-peptide substances that are bound to antibodies or antigen-binding fragments thereof directly or via linkers, etc. The modifying agents used in the conjugates of the present invention are not particularly limited and include, for example, polyethylene glycol, sugar chains, phospholipids, radioisotopes (e.g., zirconium-89 ( 89 Zr), Yttrium-90( 90 Y), Indium-111( 111 In), astatine-211(211 At), Actinium-225( 225 Examples of the modifying agent include fluorophores, fluorophores (e.g., IRDye®), organic compounds, toxins, near-infrared fluorescent dyes (e.g., IRDye®), and chelating agents. The modifying agent used in the conjugate may be directly bound to the anti-TSPAN8 antibody or antigen-binding fragment thereof of the present invention, or may be bound via any linker. In one embodiment, the conjugate of the present invention is an antibody-drug conjugate (ADC) of the anti-TSPAN8 antibody or antigen-binding fragment thereof. The drug and linker used in the ADC can be selected from drugs and linkers commonly used by those skilled in the art. In one embodiment, the conjugate of the present invention is a radioisotope-labeled antibody in which a radioisotope is bound to the anti-TSPAN8 antibody or antigen-binding fragment thereof.

[0143] The present invention also provides cells (e.g., chimeric antigen receptor-T cells; CAR-T cells) expressing the anti-TSPAN8 antibody of the present invention or an antigen-binding fragment thereof on the cell surface. Such cells can be produced by those skilled in the art using a polynucleotide encoding the anti-TSPAN8 antibody of the present invention or an antigen-binding fragment thereof. Various immune cells (T cells, NK cells, NKT cells, etc.) can be used as cells expressing the anti-TSPAN8 antibody of the present invention or an antigen-binding fragment thereof.

[0144] The anti-TSPAN8 antibody or antigen-binding fragment thereof of the present invention, the fusion product of the present invention, the conjugate of the present invention, and cells expressing the anti-TSPAN8 antibody or antigen-binding fragment thereof on their cell surface bind to human TSPAN8 (gene ID: NM_004616.2). Binding to human TSPAN8 can be confirmed using known methods for measuring binding activity. Methods for measuring binding activity include, for example, ELISA and flow cytometry. When using ELISA, the method described in Example 8 can be used, and when using flow cytometry, the method described in Example 1 can be used, for example.

[0145] In one embodiment, the anti-TSPAN8 antibody or antigen-binding fragment thereof of the present invention, the fusion and conjugate of the present invention, and the antibody or antigen-binding fragment portion in a cell expressing the anti-TSPAN8 antibody or antigen-binding fragment thereof on its cell surface may be post-translationally modified. In one embodiment, the post-translation modification is pyroglutamylation of the N-terminus of the heavy chain variable region and / or deletion of a lysine at the C-terminus of the heavy chain.

[0146] The anti-TSPAN8 antibodies or antigen-binding fragments thereof of the present invention, the fusions of the present invention, and the conjugates of the present invention, as well as cells expressing the anti-TSPAN8 antibodies or antigen-binding fragments thereof on their cell surface, can be prepared by those skilled in the art using methods known in the art, based on the VH and VL sequence information of the anti-TSPAN8 antibodies or antigen-binding fragments thereof disclosed herein, other peptides or proteins (e.g., antibodies) used in the fusions of the present invention, and information on modifying agents used in the conjugates of the present invention. The anti-TSPAN8 antibodies or antigen-binding fragments thereof of the present invention can be produced, for example, by the methods described in the section "Methods for Producing Bispecific Antibodies of the Present Invention," without particular limitation.

[0147] <Polynucleotide, expression vector, host cell, and production method of the anti-TSPAN8 antibody of the present invention> The present invention also provides the following polynucleotides (1) to (4): (1) A polynucleotide comprising a nucleotide sequence encoding the heavy chain variable region of the anti-TSPAN8 antibody of the present invention or its antigen-binding fragment. (2) a polynucleotide comprising a nucleotide sequence encoding the light chain variable region of the anti-TSPAN8 antibody of the present invention or an antigen-binding fragment thereof; (3) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of the anti-TSPAN8 antibody of the present invention; (4) A polynucleotide comprising a nucleotide sequence encoding the light chain of the anti-TSPAN8 antibody of the present invention.

[0148] In one embodiment of the polynucleotide (1) above, the polynucleotide comprising a nucleotide sequence encoding the heavy chain variable region of the anti-TSPAN8 antibody or antigen-binding fragment thereof of the present invention is a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding a heavy chain variable region of an anti-TSPAN8 antibody, which comprises CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 4, CDR2 consisting of the amino acid sequence of amino acids 50 to 66 of SEQ ID NO: 4, and CDR3 consisting of the amino acid sequence of amino acids 99 to 110 of SEQ ID NO: 4; (b) A polynucleotide comprising a base sequence encoding the heavy chain variable region of an anti-TSPAN8 antibody, which includes a CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 10, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 of SEQ ID NO: 10, and a CDR3 consisting of the amino acid sequence from amino acid numbers 99 to 110 of SEQ ID NO: 10.

[0149] In one embodiment of the polynucleotide (1) above, the polynucleotide comprising a nucleotide sequence encoding the heavy chain variable region of the anti-TSPAN8 antibody or antigen-binding fragment thereof of the present invention is a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 4; (b) A polynucleotide comprising a base sequence encoding the heavy chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10.

[0150] In one embodiment of the polynucleotide (2) above, the polynucleotide comprising a nucleotide sequence encoding the light chain variable region of the anti-TSPAN8 antibody or antigen-binding fragment thereof of the present invention is a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding a light chain variable region of an anti-TSPAN8 antibody, which comprises a CDR1 consisting of the amino acid sequence of amino acids 24 to 34 of SEQ ID NO: 8, a CDR2 consisting of the amino acid sequence of amino acids 50 to 56 of SEQ ID NO: 8, and a CDR3 consisting of the amino acid sequence of amino acids 89 to 96 of SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain variable region of an anti-TSPAN8 antibody, which includes CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 12, CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 12, and CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 12.

[0151] In one embodiment of the polynucleotide (2) above, the polynucleotide comprising a nucleotide sequence encoding the light chain variable region of the anti-TSPAN8 antibody or antigen-binding fragment thereof of the present invention is a polynucleotide selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding the light chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12.

[0152] In one embodiment of the polynucleotide (3) above, the polynucleotide comprising a nucleotide sequence encoding the heavy chain of the anti-TSPAN8 antibody of the present invention is a polynucleotide selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 4; (b) A polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 10.

[0153] In one embodiment of the polynucleotide (4) above, the polynucleotide comprising a nucleotide sequence encoding the light chain of the anti-TSPAN8 antibody of the present invention is a polynucleotide selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 12.

[0154] The polynucleotides described in this section can be prepared by those skilled in the art based on the nucleotide sequences using methods known in the art.

[0155] The present invention also provides expression vectors (also referred to as "expression vectors for the anti-TSPAN8 antibodies of the present invention") containing one or more of the polynucleotides described in (1) to (4) below. Each expression vector may contain one or more of the respective polynucleotides. (1) a polynucleotide comprising a nucleotide sequence encoding the heavy chain variable region of the anti-TSPAN8 antibody of the present invention or an antigen-binding fragment thereof; (2) a polynucleotide comprising a nucleotide sequence encoding the light chain variable region of the anti-TSPAN8 antibody of the present invention or an antigen-binding fragment thereof; (3) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of the anti-TSPAN8 antibody of the present invention; (4) A polynucleotide comprising a nucleotide sequence encoding the light chain of the anti-TSPAN8 antibody of the present invention.

[0156] In one embodiment of the expression vector for the anti-TSPAN8 antibody of the present invention, which comprises the polynucleotide (1) above, the expression vector comprises a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding a heavy chain variable region of an anti-TSPAN8 antibody, which comprises CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 4, CDR2 consisting of the amino acid sequence of amino acids 50 to 66 of SEQ ID NO: 4, and CDR3 consisting of the amino acid sequence of amino acids 99 to 110 of SEQ ID NO: 4; (b) A polynucleotide comprising the heavy chain variable region of an anti-TSPAN8 antibody, which comprises a CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 10, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 of SEQ ID NO: 10, and a CDR3 consisting of the amino acid sequence from amino acid numbers 99 to 110 of SEQ ID NO: 10.

[0157] In one embodiment of the expression vector for the anti-TSPAN8 antibody of the present invention, which comprises the polynucleotide (1) above, the expression vector comprises a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 4; (b) A polynucleotide comprising a base sequence encoding the heavy chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10.

[0158] In one embodiment of the expression vector for the anti-TSPAN8 antibody of the present invention, which comprises the polynucleotide (2) above, the expression vector comprises a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding a light chain variable region of an anti-TSPAN8 antibody, which comprises a CDR1 consisting of the amino acid sequence of amino acids 24 to 34 of SEQ ID NO: 8, a CDR2 consisting of the amino acid sequence of amino acids 50 to 56 of SEQ ID NO: 8, and a CDR3 consisting of the amino acid sequence of amino acids 89 to 96 of SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain variable region of an anti-TSPAN8 antibody, which includes CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 12, CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 12, and CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 12.

[0159] In one embodiment of the expression vector for the anti-TSPAN8 antibody of the present invention, which comprises the polynucleotide (2) above, the expression vector comprises a polynucleotide selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding the light chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12.

[0160] In one embodiment of the expression vector for the anti-TSPAN8 antibody of the present invention, which comprises the polynucleotide (3) above, the expression vector comprises a polynucleotide selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 4; (b) A polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 10.

[0161] In one embodiment of the expression vector for the anti-TSPAN8 antibody of the present invention, which comprises the polynucleotide (4) above, the expression vector comprises a polynucleotide selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 12.

[0162] In one embodiment, the expression vector for the anti-TSPAN8 antibody of the present invention is an expression vector comprising a polynucleotide selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding a heavy chain variable region of an anti-TSPAN8 antibody, the heavy chain variable region comprising a CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 4, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 of SEQ ID NO: 4, and a CDR3 consisting of the amino acid sequence from amino acid numbers 99 to 110 of SEQ ID NO: 4, and a polynucleotide comprising a nucleotide sequence encoding a light chain variable region of an anti-TSPAN8 antibody, the light chain variable region comprising a CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 8, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 8, and a CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding a heavy chain variable region of an anti-TSPAN8 antibody, which comprises a CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 10, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 of SEQ ID NO: 10, and a CDR3 consisting of the amino acid sequence from amino acid numbers 99 to 110 of SEQ ID NO: 10, and a polynucleotide comprising a base sequence encoding a light chain variable region of an anti-TSPAN8 antibody, which comprises a CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 12, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 12, and a CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 12.

[0163] In one embodiment, the expression vector for the anti-TSPAN8 antibody of the present invention is an expression vector comprising a polynucleotide selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding a heavy chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence of amino acids 1 to 121 of SEQ ID NO: 4, and a polynucleotide comprising a nucleotide sequence encoding a light chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence of amino acids 1 to 107 of SEQ ID NO: 8; (b) a polynucleotide comprising a base sequence encoding the heavy chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10, and a polynucleotide comprising a base sequence encoding the light chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12.

[0164] In one embodiment, the expression vector for the anti-TSPAN8 antibody of the present invention is an expression vector comprising a polynucleotide selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 4, and a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 10, and a polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 12.

[0165] The expression vectors described in this section can be prepared by those skilled in the art using the methods described above in <Expression vectors for bispecific antibodies of the present invention>.

[0166] The present invention also provides host cells (also referred to as "host cells for the anti-TSPAN8 antibodies of the present invention") transformed with an expression vector containing the following polynucleotides (1) to (4): Host cells for the anti-TSPAN8 antibodies of the present invention may contain one or more of each of the following polynucleotides (1) to (4) by transformation with an expression vector for the anti-TSPAN8 antibody of the present invention. (1) A polynucleotide comprising a nucleotide sequence encoding the heavy chain variable region of the anti-TSPAN8 antibody of the present invention or its antigen-binding fragment. (2) a polynucleotide comprising a nucleotide sequence encoding the light chain variable region of the anti-TSPAN8 antibody of the present invention or an antigen-binding fragment thereof; (3) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of the anti-TSPAN8 antibody of the present invention; (4) A polynucleotide comprising a nucleotide sequence encoding the light chain of the anti-TSPAN8 antibody of the present invention.

[0167] In one embodiment of the host cell for the anti-TSPAN8 antibody of the present invention, which comprises the polynucleotide (1) above, the host cell comprises a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding a heavy chain variable region of an anti-TSPAN8 antibody, which comprises CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 4, CDR2 consisting of the amino acid sequence of amino acids 50 to 66 of SEQ ID NO: 4, and CDR3 consisting of the amino acid sequence of amino acids 99 to 110 of SEQ ID NO: 4; (b) A polynucleotide comprising a base sequence encoding the heavy chain variable region of an anti-TSPAN8 antibody, which includes a CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 10, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 of SEQ ID NO: 10, and a CDR3 consisting of the amino acid sequence from amino acid numbers 99 to 110 of SEQ ID NO: 10.

[0168] In one embodiment of the host cell for the anti-TSPAN8 antibody of the present invention, which comprises the polynucleotide (1) above, the host cell comprises a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 4; (b) A polynucleotide comprising a base sequence encoding the heavy chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10.

[0169] In one embodiment of the host cell for the anti-TSPAN8 antibody of the present invention, which comprises the polynucleotide (2) above, the host cell comprises a polynucleotide selected from the following (a) and (b): (a) a polynucleotide comprising a nucleotide sequence encoding a light chain variable region of an anti-TSPAN8 antibody, which comprises a CDR1 consisting of the amino acid sequence of amino acids 24 to 34 of SEQ ID NO: 8, a CDR2 consisting of the amino acid sequence of amino acids 50 to 56 of SEQ ID NO: 8, and a CDR3 consisting of the amino acid sequence of amino acids 89 to 96 of SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain variable region of an anti-TSPAN8 antibody, which includes CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 12, CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 12, and CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 12.

[0170] In one embodiment of the host cell for the anti-TSPAN8 antibody of the present invention, which comprises the polynucleotide (2) above, the host cell comprises a polynucleotide selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding the light chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12.

[0171] In one embodiment of the host cell for the anti-TSPAN8 antibody of the present invention, which comprises the polynucleotide (3) above, the host cell comprises a polynucleotide selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 4; (b) A polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 10.

[0172] In one embodiment of the host cell for use with the anti-TSPAN8 antibody of the present invention, the host cell comprises the polynucleotide (4) above, and the polynucleotide is selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 12.

[0173] In one embodiment, a host cell for an anti-TSPAN8 antibody of the invention comprises a polynucleotide selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding a heavy chain variable region of an anti-TSPAN8 antibody, the heavy chain variable region comprising a CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 4, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 of SEQ ID NO: 4, and a CDR3 consisting of the amino acid sequence from amino acid numbers 99 to 110 of SEQ ID NO: 4, and a polynucleotide comprising a nucleotide sequence encoding a light chain variable region of an anti-TSPAN8 antibody, the light chain variable region comprising a CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 8, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 8, and a CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding a heavy chain variable region of an anti-TSPAN8 antibody, which comprises a CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 10, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 of SEQ ID NO: 10, and a CDR3 consisting of the amino acid sequence from amino acid numbers 99 to 110 of SEQ ID NO: 10, and a polynucleotide comprising a base sequence encoding a light chain variable region of an anti-TSPAN8 antibody, which comprises a CDR1 consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 12, a CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 12, and a CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO: 12.

[0174] In one embodiment, the host cell for the anti-TSPAN8 antibody of the invention is a host cell comprising a polynucleotide selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding a heavy chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence of amino acids 1 to 121 of SEQ ID NO: 4, and a polynucleotide comprising a nucleotide sequence encoding a light chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence of amino acids 1 to 107 of SEQ ID NO: 8; (b) a polynucleotide comprising a base sequence encoding the heavy chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10, and a polynucleotide comprising a base sequence encoding the light chain variable region of an anti-TSPAN8 antibody consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO: 12.

[0175] In one embodiment, the host cell for the anti-TSPAN8 antibody of the invention is a host cell comprising a polynucleotide selected from the following (a) or (b): (a) a polynucleotide comprising a nucleotide sequence encoding the heavy chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 4, and a polynucleotide comprising a nucleotide sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 8; (b) A polynucleotide comprising a base sequence encoding the heavy chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 10, and a polynucleotide comprising a base sequence encoding the light chain of an anti-TSPAN8 antibody consisting of the amino acid sequence shown in SEQ ID NO: 12.

[0176] The host cells described in this section can be prepared by those skilled in the art according to the method described above in <Transformed host cells of the present invention>.

[0177] The present invention further provides a method for producing an anti-TSPAN8 antibody or an antigen-binding fragment thereof, comprising the step of culturing a host cell for the anti-TSPAN8 antibody of the present invention. This method can be carried out by those skilled in the art according to the above-mentioned <Method for producing a bispecific antibody of the present invention>.

[0178] <Medicinal uses of the anti-TSPAN8 antibody and the like of the present invention> The present invention also provides pharmaceutical compositions comprising an anti-TSPAN8 antibody or antigen-binding fragment thereof of the present invention, a fusion protein of the present invention, a conjugate of the present invention, and a cell expressing an anti-TSPAN8 antibody or antigen-binding fragment thereof on its cell surface (collectively referred to in this section as the "anti-TSPAN8 antibody, etc. of the present invention"), as well as a pharmaceutically acceptable excipient. These pharmaceutical compositions can be used for the treatment of cancer. The present invention also provides methods for treating cancer, including the step of administering a therapeutically effective amount of an anti-TSPAN8 antibody, etc. of the present invention to a subject; anti-TSPAN8 antibodies, etc. of the present invention for use in cancer treatment; and use of anti-TSPAN8 antibodies, etc. of the present invention in the manufacture of pharmaceutical compositions for cancer treatment. The pharmaceutical use of anti-TSPAN8 antibodies, etc. of the present invention can be carried out by those skilled in the art in accordance with the description in the above section "Pharmaceutical Compositions, etc. of Bispecific Antibodies of the Present Invention." Cancers that can be treated using the anti-TSPAN8 antibodies, etc. of the present invention include those described above in the section "Pharmaceutical Compositions, etc. of Bispecific Antibodies of the Present Invention."

[0179] <Anti-CD3 antibody of the present invention> The present invention also provides the following anti-CD3 antibodies or antigen-binding fragments thereof: An anti-CD3 antibody or antigen-binding fragment thereof, comprising: a heavy chain variable region comprising CDR1 consisting of the amino acid sequence of amino acids 31 to 35 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 50 to 68 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 101 to 114 of SEQ ID NO: 14; and a light chain variable region comprising CDR1 consisting of the amino acid sequence of amino acids 168 to 181 of SEQ ID NO: 14, CDR2 consisting of the amino acid sequence of amino acids 197 to 203 of SEQ ID NO: 14, and CDR3 consisting of the amino acid sequence of amino acids 236 to 244 of SEQ ID NO: 14.

[0180] In one embodiment, the anti-CD3 antibody or antigen-binding fragment thereof of the present invention is an anti-CD3 antibody or antigen-binding fragment thereof comprising a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 125 of SEQ ID NO: 14 and a light chain variable region consisting of the amino acid sequence of amino acids 146 to 254 of SEQ ID NO: 14.

[0181] In one embodiment, an antigen-binding fragment of an anti-CD3 antibody of the present invention is an scFv. In one embodiment, an antigen-binding fragment of an anti-CD3 antibody of the present invention is an anti-CD3 antibody scFv comprising a heavy chain variable region consisting of the amino acid sequence of amino acids 1 to 125 of SEQ ID NO: 14 and a light chain variable region consisting of the amino acid sequence of amino acids 146 to 254 of SEQ ID NO: 14. In one embodiment, an antigen-binding fragment of an anti-CD3 antibody of the present invention is an anti-CD3 antibody scFv consisting of the amino acid sequence of amino acids 1 to 254 of SEQ ID NO: 14.

[0182] The anti-CD3 antibodies or antigen-binding fragments thereof of the present invention can be prepared by those skilled in the art with reference to the descriptions herein, such as "Anti-TSPAN8-anti-CD3 bispecific antibodies of the present invention." The anti-CD3 antibodies or antigen-binding fragments thereof of the present invention can be confirmed using known binding activity measurement methods. The anti-CD3 antibodies or antigen-binding fragments thereof of the present invention can be used, for example, in bispecific antibodies with antibodies against tumor antigens, which are used in cancer treatment.

[0183] To provide a further understanding of the present invention, reference is now made to specific examples which are provided for purposes of illustration and not limitation. [Example]

[0184] Example 1: Obtaining antibodies that selectively bind to antigens expressed in peritoneal cancer dissemination cells [Example 1-1: Obtaining patient-derived cancer peritoneal disseminated cells] Cancer peritoneal dissemination cells were obtained from patients using the following method. They were obtained from patients according to the method described in "Establishment of Peritoneal Metastatic Cancer (Gastric, Pancreatic, Ovarian, etc.) Cell Lines" by Fumiko Senwaki and Hiroki Sasaki ("Practical Guide to Cancer Research Using Patient-Derived Cancer Models," edited by Hiroki Sasaki, Yodosha, 2019, pp. 28-37). Ascites collected from patients was dispensed into ProteoSave® SS 50 mL centrifuge tubes (Sumitomo Bakelite Co., Ltd., MS-52550, hereafter referred to as "50 mL centrifuge tubes") and centrifuged at 430 × g for 3 minutes at room temperature. After removing the supernatant, hemolysis buffer was added to the precipitate and allowed to hemolyze for 10–20 minutes at room temperature. The hemolysis buffer was prepared by filtering 17 mM Tris-HCl (pH 7.65) containing 0.75% ammonium chloride through a 0.22 μm pore size filter. After centrifugation, the supernatant was removed, and 50 mL of Dulbecco's PBS(-) (Nissui Pharmaceutical Co., Ltd., 05913, hereafter referred to as "PBS(-)") was added to wash the cells. The cells were then collected by centrifugation at 430 × g for 3 minutes at room temperature. All collected cells from the ascites were resuspended in RPMI-1640 (L-glutamine-containing) medium (Fujifilm Wako Pure Chemical Industries, Ltd., 189-02025) containing 10% FBS (Thermo Fisher Scientific, 10270-106) and ×1 Antibiotic-Antimycotic (Thermo Fisher Scientific, 15240062) (hereafter referred to as "RPMI-1640 medium"). 5 × 10 cells were placed in a 100 mm collagen-coated dish (hereafter referred to as "dish") (IWAKI, 4020-010). 6 ~1×10 7 The cells were seeded at 10 mL each and cultured at 37°C in a 5% CO2 incubator. Ascites contains both adherent and non-cancer cells. Adherent cells include not only cancer cells but also non-cancer cells (e.g., fibroblasts and peritoneal mesothelial cells). By utilizing the property that non-cancer cells detach more quickly than cancer cells, these cells were separated from the total cells in the ascites. Specifically, the dish in which the total cells in the ascites were cultured was washed with PBS(-), and then treated with 2 mL of 0.05% trypsin-EDTA (Thermo Fisher Scientific, 15400054) for several minutes to detach the non-cancer cells. The detached non-cancer cells were continuously cultured in a new dish and used in Examples 1-5. After removing non-cancer cells, the cancer cells were allowed to grow to approximately 80% confluence on the dish surface, and then half of the cells were subcultured onto a new dish. Cells that had been subcultured five or more times were designated as adherent cancer cells. For suspension cells, 5 mL of culture supernatant and 5 mL of RPMI-1640 medium were seeded onto a new 100 mm dish and subcultured. Cells that had been subcultured five or more times were designated as suspension cancer cells. When peritoneal cancer dissemination cells derived from a single patient contained both adherent and suspension cancer cells and continued to grow, they were designated as mixed cancer cells. In this specification, the adherent cancer cells, non-adherent cancer cells, and mixed cancer cells isolated from ascites collected from patients using the above-mentioned method are collectively referred to as "peritoneal cancer dissemination cells." The 12 cells obtained (NSC-7C, NSC-9C, NSC-10C, NSC-14C, NSC-15CF, NSC-16C, NSC-20C, NSC-22C, NSC-24C, NSC-32C, NSC-34C, and NSC-35C-1 (hereinafter also referred to as "12 types of peritoneal cancer dissemination cells")) were used in the following studies.

[0185] [Example 1-2: Preparation of anti-gastric cancer antigen antibody-producing hybridomas] Using human monoclonal antibody development technology "VelocImmune (registered trademark) antibody technology; Regeneron (US Patent No. 6,596,541)" and mice, we obtained anti-gastric cancer antigen antibodies that bind to cancer peritoneal disseminated cells. Of the cancer peritoneal dissemination cells obtained in Example 1-1, NSC-10C, NSC-35C-1, NSC-24C, NSC-7C, NSC-14C, and NSC-34C were mixed in groups of three and suspended in TiterMax® Gold ADJUVANT (MERCK, T2684) or PBS(-) to prepare a cancer peritoneal dissemination cell suspension. Velocimmune mice were immunized with this suspension, and hybridomas were generated according to standard methods. Single hybridoma colonies were isolated using an automated picking device to obtain monoclonal hybridoma cells (hereinafter referred to as "clones"). The isolated clones were cultured in an 8% CO2 incubator at 37°C. After 4 days of culture, the supernatant was collected in a 96-well plate and used in the following experiments.

[0186] [Example 1-3: Selection of anti-gastric cancer antigen antibodies that selectively bind to cancer peritoneal disseminated cells] 1. Confirmation of binding of anti-gastric cancer antigen antibodies to peritoneal cancer dissemination cells and EpCAM-expressing cells The cell supernatant of the clone obtained in Example 1-2 contains an antibody (hereinafter referred to as "antibody contained in the clone supernatant"). First, the binding of antibodies contained in the clone supernatants to the 12 types of cancer peritoneal dissemination cells obtained in Example 1-1 was measured by flow cytometry, and clones producing antibodies that strongly bound to cancer peritoneal dissemination cells were selected. BV421 Goat Anti-Mouse Ig (Becton, Dickinson and Company, 563846) was used for flow cytometry. Next, to exclude clones that produced antibodies binding to the cancer antigen EpCAM, the binding of antibodies contained in the clone supernatants to human EpCAM-Myc-DDK-expressing CHO-K1 cells was measured. Human EpCAM-Myc-DDK-expressing CHO-K1 cells were generated by transfecting CHO-K1 cells (ATCC, CCL-61) with EPCAM (Myc-DDK-tagged)-Human epithelial cell adhesion molecule (EPCAM) (ORIGENE, RC201989). The binding of antibodies contained in the clone supernatants to the cells was measured by flow cytometry. Flow cytometry was performed using BV421 Goat Anti-Mouse Ig. To select clones that provided supernatants that did not bind to EpCAM-expressing cells, clones that bound to these cells were excluded. CD326 (EpCAM) Monoclonal Antibody (1B7) (eBioscience, 14-9326) was used as a positive control. From the above experiments, clones were selected that provided antibodies that bound to 10 or more of 12 types of peritoneal cancer disseminated cells but did not bind to human EpCAM.

[0187] 2. Confirmation of antibody binding to human peripheral blood mononuclear cells Furthermore, in order to select clones that provide antibodies that selectively bind to cancer peritoneal disseminated cells, clones that provide antibodies that bind to human peripheral blood mononuclear cells were excluded. Flow cytometry was used to measure the binding of antibodies provided by each clone to human peripheral blood mononuclear cells (hMNC-PB), pooled, ultra-pure (PromoCell, C-12908). PE Goat Anti-Mouse Ig (Multiple Adsorption) (Becton, Dickinson and Company, 550589, hereafter referred to as "PE Goat Anti-Mouse Ig"), BV421 Mouse Anti-Human CD3 (Becton, Dickinson and Company, 562426), APC Mouse Anti-Human CD14 (Becton, Dickinson and Company, 555399), and BB515 Mouse Anti-Human CD19 (Becton, Dickinson and Company, 564456) were used.

[0188] 3. Antibody Purification from Hybridoma Supernatant The clones selected in steps 1 and 2 of Example 1-3 were cultured in CD Hybridoma Medium (Thermo Fisher Scientific, 11279023). Antibodies were purified from the culture supernatant using MabSelectSuRe (GE Healthcare, 17-5438-02) (hereinafter referred to as "purified antibodies"). Antibody purification was performed according to standard methods.

[0189] 4. Binding of Purified Antibodies to Cultured Human Peritoneal Mesothelial Cells To select antibodies that selectively bind to cancer peritoneal dissemination cells, antibodies that bind to cultured human peritoneal mesothelial cells were excluded from the purified antibodies obtained in Example 1-3, section 3. Human Mesothelial Cells (Zenbio, MES-F, Lot. MESM012916B) (referred to herein as "cultured human peritoneal mesothelial cells") were used as cultured human peritoneal mesothelial cells and cultured in Mesothelial Cell Growth Medium (Zenbio, MSO-1). Flow cytometry was used to measure the binding of purified antibodies to cultured human peritoneal mesothelial cells. PE Goat Anti-Mouse Ig was used for flow cytometry. Antibodies that did not bind to cultured human peritoneal mesothelial cells or showed weak binding were selected, and 14 purified antibodies were obtained (hereinafter referred to as "14 purified antibodies").

[0190] [Example 1-4: Identification of candidate antigen molecules recognized by the obtained antibodies] We identified candidate antigen molecules for 14 purified antibodies. As an example of the identification method, we will describe in detail the method for identifying candidate antigen molecules for 16B11, 16B12, and 21F7. As control antibodies in this experiment, 5D3, 9A1, and 21A3, which have binding patterns to peritoneal cancer disseminated cells different from those of 16B11, were used. A cell lysate was prepared from NSC-15CF cells. To the cell lysate, 16B11, 16B12, 21F7, and one of three control antibodies (5D3, 9A1, 21A3) were added. Dynabeads Protein G (Life Technologies, 10003D) was then added, stirred, and washed. Proteins bound to Dynabeads Protein G were digested with Tripsin / LysC (Promega, V5072) to obtain a peptide mixture. The resulting peptide mixture was subjected to LC-MS / MS analysis using an UltiMate 3000 RSnano column (Thermo Fisher Scientific) and an Orbitrap Fusion column (Thermo Fisher Scientific). The resulting LC-MS / MS data were analyzed using Progenesis QI for Proteomics (Waters) and Mascot (Matrix Science) software for comparative quantitative analysis and peptide / protein identification, and binding proteins were identified. The data for 16B11, 16B12, and 21F7 were compared with the data for each control antibody, and TSPAN8 was identified as a candidate antigen for 16B11 and 21F7. In this experiment, a candidate antigen for 16B12 could not be identified. Experiments were also performed using the same techniques as above with 5B7, 9F6, 12C12, 13A9, 15D1, 18C10, and 19E4, and TSPAN8 was identified as a candidate antigen. 24C7 was used as a control antibody in addition to 5D3, 9A1, and 21A3. Although 16B12 did not lead to the identification of a candidate antigen, it showed a binding profile similar to that of 16B11 in Examples 1-3. Therefore, TSPAN8 was presumed to be a candidate antigen and further investigations were carried out. Furthermore, to identify the antigen, we performed a binding experiment using human TSPAN8-Myc-DDK-expressing CHO-K1 cells. Human TSPAN8-Myc-DDK-expressing CHO-K1 cells were generated by transfecting CHO-K1 cells with TSPAN8 (Myc-DDK-tagged)-Human tetraspanin 8 (TSPAN8) (ORIGENE, RC202694) (SEQ ID NO: 2). Ten antibodies (16B11, 16B12, 5B7, 9F6, 12C12, 13A9, 15D1, 18C10, 19E4, and 21F7) (also referred to as the "ten anti-TSPAN8 antibodies") bound to human TSPAN8-Myc-DDK-expressing CHO-K1 cells and recognized TSPAN8 as an antigen.

[0191] Example 1-5: Confirmation of binding activity of anti-TSPAN8 antibodies to various cells 1. Confirmation and quantification of binding activity to peritoneal cancer dissemination cells The binding of seven types of peritoneal cancer dissemination cells (KM-291-As, KM-555-As, KM-556-As, P-249-As, KM-568-As, KM-570-As, and KM-577-As) isolated from the ascites of gastric cancer patients to four types of anti-TSPAN8 antibodies was measured by flow cytometry. KM-291-As was prepared from the patient's ascites using the same method as in Example 1-1. Since the prepared KM-291-As contained many blood cells expressing CD45, CD45-expressing cells were removed using a column to concentrate cancer cells. Specifically, 5 × 10 7 A cell suspension containing 100 cells was passed through Pre-Separation Columns (30 μm) (Miltenyi Biotec, 130-041-407) and Separation Columns (Miltenyi Biotec, 130-042-401) (hereinafter referred to as "Columns") using CD45 MicroBeads, human (Miltenyi Biotec, 130-045-801) according to standard procedures. The column flow-through was collected in a 50 mL centrifuge tube and centrifuged. 10 mL of buffer was added to the precipitate, and the cells were resuspended. 5 × 10 cells were extracted from this cell suspension. 6Cells were aliquoted into 1.5 mL microtubes (Watson, 131-7155C) and centrifuged to obtain a pellet. To the pellet, 950 μL of PBS (FCM buffer) containing 2% FBS and 100 μg / mL penicillin-streptomycin (Thermo Fisher Scientific, 15140-122) was added to prepare a cell suspension. 50 μL of FcR Blocking Reagent was added and incubated on ice for 10 minutes. This reaction mixture was dispensed into seven 1.5 mL microtubes (100 μL each), and the cells were stained as follows: 5 μL of mouse IgG1-PE antibody (Miltenyi Biotec, 130-092-212) was added to three of the seven microtubes, followed by the addition of 2.5 μL of Alexa Fluor 647-labeled control antibody to each microtube as an isotype control. As control antibodies, mouse IgG2a isotype control (Becton, Dickinson and Company, 558053), mouse IgG2b isotype control (Becton, Dickinson and Company, 558713), or mouse IgG3 isotype control (Becton, Dickinson and Company, 560803) was used. To the remaining four microtubes, 5 μL of CD326 (EpCAM)-PE (Miltenyi Biotec, 130-091-253) was added. Furthermore, 2.5 μL each of 16B11, 16B12, 9F6, and 18C10 (0.25 μg / tube) was added to each microtube to stain the cells. These four antibody samples were used for evaluation. After antibody addition, each microtube was incubated on ice for 30 minutes. 1 mL of FCM buffer was added and centrifuged, and 500 μL of FCM buffer was added to the resulting pellet to resuspend the cells.5 μL of 7-AAD (Becton, Dickinson and Company, 559925) was added, and the entire volume was transferred to a 5 mL round-bottom polystyrene tube with a cell strainer cap (Corning, 352235), and measurement was performed using a FACSVerse flow cytometer (Becton, Dickinson and Company). Data were acquired using BD FACSuite software (Becton, Dickinson and Company). KM-555-As, KM-556-As, P-249-As, KM-568-As, KM-570-As, and KM-577-As were prepared in the same manner as KM-291-As. For binding assays with KM-555-As and KM-556-As, 16B11, 9F6, and 18C10 were used as evaluation antibody samples. For binding assays with P-249-As, 16B11 and a commercially available anti-TSPAN8 antibody, TSPAN8 Antibody, Anti-Human, REAfinity (130-106-855, Miltenyi, herein referred to as "REA443"), were used as evaluation antibody samples. For binding assays with KM-568-As, KM-570-As, and KM-577-As, 16B11, 9F6, 18C10, and REA443 were used as evaluation antibody samples. In all experiments, an isotype control antibody, mouse IgG1 (130-113-196, Miltenyi Biotech), was used as an isotype control. Cells were stained with IgG2a-VioBlue antibody (Miltenyi Biotech, 130-113-277) and CD326 (EpCAM)-VioBlue antibody (Miltenyi Biotech, 131-113-266). Flow cytometry analysis of each peritoneal cancer dissemination cell line was performed using BD FACSuite software. Specifically, FSC-A (lin) / SSC-A (log) plots were created, and the resulting cell population was gated. The resulting cell population was then analyzed again using FSC-W (lin) / FSC-A (lin). Singlet populations were gated to create subsets for analysis. For analysis of KM-291-As cell lines, the subsets were analyzed using PE (log) / Alexa Fluor 647 (log). For analysis of KM-555-As, KM-556-As, P-249-As, KM-568-As, KM-570-As, and KM-577-As, data analyzed using VioBlue (log) / Alexa Fluor 647 (log) were used. For each sample, 1 × 10 4 Data were acquired for each cell subset. The acquired FCS files were analyzed using FlowJo (Becton, Dickinson and Company), and histograms for Alexa Fluor 647 were generated. The MFI of Alexa Fluor 647 in the positive population was calculated for each isotype and each antibody evaluated, and the ΔMFI was calculated by subtracting the MFI of each isotype from the MFI of each antibody (Table 1). As examples of the obtained histograms, histograms showing the binding of 16B11 and 16B12 in the binding measurement with KM-291-As, and 16B11, 9F6, and 18C10 in the binding measurement with KM-555-As and KM-556-As are shown in Figures 1-1 to 1-3, respectively.

[0192] 2. Confirmation and quantification of binding activity to cultured human peritoneal mesothelial cells The binding activity of the 10 anti-TSPAN8 antibodies identified in Examples 1-4 to cultured human peritoneal mesothelial cells was measured. Cultured human peritoneal mesothelial cells are normal cells. The binding of the 10 anti-TSPAN8 antibodies obtained in Examples 1-3 and a commercially available purified anti-human TSPAN8 antibody (BioLegend, 363702 Clone TAL69, herein referred to as "TAL69") to cultured human peritoneal mesothelial cells was measured by flow cytometry. PE Goat Anti-Mouse Ig was used as the secondary antibody. The resulting histograms are shown in Figures 2-1 and 2-2. Furthermore, the flow cytometry results were analyzed using FlowJo to calculate the MFI of PE for each cell population. Three negative control antibodies were used in the analysis: Ultra-LEAF Purified Mouse IgG1,κ Isotype Ctrl Antibody (BioLegend, 401408), Purified NA / LE Mouse IgG2a,κ Isotype Control (Becton, Dickinson and Company, 554645), and Purified NA / LE Mouse IgG2b,κ Isotype Control (Becton, Dickinson and Company, 559530). The ΔMFI values ​​for 16B11, 16B12, 9F6, 18C10, and TAL69 are listed in Table 1. The ΔMFI value of an antibody was calculated by subtracting the MFI value of the negative control antibody from the MFI value of the respective antibody.

[0193] 3. Confirmation and quantification of binding activity to peritoneal mesothelial cells derived from patients with peritoneal dissemination of gastric cancer KM-501-As and KM-503-As, which are peritoneal mesothelial cells isolated from the ascites of a human patient with peritoneal dissemination of gastric cancer (referred to herein as "patient-derived peritoneal mesothelial cells"), were obtained by the following method. During the cell establishment process in Example 1-1, mesothelial cells were observed growing in a paving-stone-like pattern all over the dish. These mesothelial cells were collected using 0.05% trypsin-EDTA and resuspended in 10 mL of D-MEM (high glucose) (Fujifilm Wako Pure Chemical Industries, Ltd., 044-29765) containing 10% FBS and 1× Antibiotic-Antimycotic. 4×10 5The cells were separated into a 1.5 mL microtube, centrifuged, and the supernatant was removed. 96 μL of FCM buffer was added to suspend the cells. 4 μL of FcR Blocking Reagent was added, and the reaction was allowed to proceed on ice for 10 minutes. 50 μL of this reaction mixture was dispensed into another 1.5 mL microtube, and 2 × 10 cells were added. 5 Two 50μL microtubes were used. One microtube contained 2μL of CD45-APC antibody (Miltenyi Biotec, 130-091-230) and 0.5μL of CD326 (EpCAM)-PE (Miltenyi Biotec, 130-113-264). The other tube contained 2μL of mouse IgG2a-APC antibody (Miltenyi Biotec, 130-091-836) and 0.5μL of mouse IgG1-PE antibody. Each microtube was incubated on ice for 30 minutes. 1mL of FCM buffer was added to each microtube, followed by centrifugation and removal of the supernatant. 500μL of FCM buffer was added to the precipitate to resuspend the cells, and analysis was performed using a FACS Verse. The cell population was gated using FSC-A (lin) and SSC-A (log), and the resulting subset was re-expanded using PE (log) and APC (log) to acquire data. The acquired FCS file was analyzed using FlowJo. Results confirmed that this cell population was CD45-negative and EpCAM-negative normal cells. These patient-derived peritoneal mesothelial cells are thought to be normal cells derived from the omentum and mesenteric tissue, which serve as a scaffold for cancer cells to take hold and grow during peritoneal dissemination. The isolated patient-derived peritoneal mesothelial cells, KM-501-As and KM-503-As, were used in the following experiments. The binding of 10 anti-TSPAN8 antibodies and TAL69 to KM-501-As and KM-503-As was measured using the same method as in Example 1-5, section 2, for the measurement of binding to cultured human peritoneal mesothelial cells. Histograms of 16B11, 16B12, 9F6, 18C10, and TAL69 against KM-501-As and KM-503-As are shown in Figures 3-1 and 3-2, and ΔMFI values ​​are listed in Table 1.

[0194] 4. Confirmation and quantification of binding activity to cultured human umbilical vascular endothelial cells The binding of 10 anti-TSPAN8 antibodies and TAL69 to cultured human umbilical vascular endothelial cells was measured by flow cytometry. Cultured human umbilical vascular endothelial cells (PromoCell, C-12200) were cultured using the Endothelial Cell Growth Medium 2 Kit (PromoCell, C-22111). The negative control antibodies used were Ultra-LEAF Purified Mouse IgG1,κ Isotype Control Antibody, Purified NA / LE Mouse IgG2a,κ Isotype Control, Ultra-LEAF Purified Mouse IgG2b,κ Isotype Control Antibody (BioLegend, 400348), and LEAF Purified Mouse IgG3,κ Isotype Control Antibody (BioLegend, 401310). PE Goat Anti-Mouse Ig was used as the secondary antibody. A histogram of the binding of the 10 anti-TSPAN8 antibodies and TAL69 is shown in Figure 4. The ΔMFI values ​​for 16B11, 16B12, 9F6, 18C10, and TAL69 are listed in Table 1. The ΔMFI values ​​were calculated by subtracting the MFI of the negative control antibody from the MFI of each antibody. The histograms in Figures 1 to 4 and the results in Table 1 indicate that 16B11 and 16B12 exhibited high binding to peritoneal disseminated cancer cells but low binding to normal cells (cultured human peritoneal mesothelial cells, patient-derived peritoneal mesothelial cells, and cultured human umbilical vascular endothelial cells). On the other hand, other anti-TSPAN8 antibodies, 9F6 and 18C10, and commercially available anti-TSPAN8 antibodies (TAL69 or REA443), exhibited lower binding to peritoneal disseminated cancer cells than 16B11 and 16B12, but higher binding to normal cells than 16B11 and 16B12. These results demonstrate that 16B11 and 16B12 have significantly different properties from other anti-TSPAN8 antibodies (9F6, 18C10, TAL69, etc.). [Table 1]

[0195] Example 2: Sequencing of 16B11 and 16B12 Genes encoding the heavy and light chains of 16B11 and 16B12 were cloned and the antibodies were sequenced according to standard methods. Velocimune technology is a technique for producing antibodies using transgenic mice in which the variable regions of endogenous immunoglobulin heavy and light chains are replaced with corresponding human variable regions. Therefore, antibodies obtained using Velocimune technology are antibodies possessing human antibody variable regions and mouse antibody constant regions (also referred to herein as "chimeric antibodies"). The amino acid sequence of the heavy chain variable region of the obtained 16B11 is shown in SEQ ID NO: 34, and the amino acid sequence of the light chain variable region of the antibody is shown in SEQ ID NO: 36. The amino acid sequence of the heavy chain variable region of the obtained 16B12 is shown in SEQ ID NO: 35, and the amino acid sequence of the light chain variable region of the antibody is shown in SEQ ID NO: 37.

[0196] Example 3: Preparation of fully human anti-TSPAN8 antibody Example 3-1: Construction of expression vector used to produce fully human anti-TSPAN8 antibody The fully human antibodies 16B11 and 16B12 were produced by linking the amino acid sequences of the human variable regions identified in Example 2 with the amino acid sequences of the human constant regions. Polypeptides were designed in which the amino acid sequence encoding the signal sequence set forth in SEQ ID NO: 38 was linked to the N-terminus of the heavy chain variable region of 16B11 and 16B12, and the amino acid sequence of the human IgG1 constant region (the sequence from amino acid numbers 122 to 451 of SEQ ID NO: 4 or 10) was linked to the C-terminus. Furthermore, a mutation was introduced into the 16th arginine (R) of the furin cleavage sequence (J. Biol. Chem., 1992, Vol. 267, pp. 16396-16402) consisting of amino acids 16 to 19 of the heavy chain variable region of the polypeptide, substituting glycine (G) for arginine (R). Polynucleotides encoding the designed polypeptides were introduced into the pcDNA3.4 TOPO® vector (Thermo Fisher Scientific). The constructed heavy chain vectors are designated pcDNA3.4-16B11.1_HC and pcDNA3.4-16B12.1_HC, respectively. Polypeptides were designed in which the amino acid sequence encoding the signal sequence set forth in SEQ ID NO: 39 was linked to the N-terminus of the light chain variable region of 16B11, and the amino acid sequence encoding the signal sequence set forth in SEQ ID NO: 40 was linked to the N-terminus of the light chain variable region of 16B12, and the amino acid sequence of the human κ chain constant region (the sequence from amino acid numbers 108 to 213 in SEQ ID NO: 8 or 12) was linked to the C-terminus of each antibody. Polynucleotides encoding the designed polypeptides were introduced into the pcDNA3.4 TOPO (registered trademark) vector. The constructed light chain vectors are designated pcDNA3.4-16B11_LC and pcDNA3.4-16B12_LC, respectively.

[0197] [Example 3-2: Preparation of fully human anti-TSPAN8 antibody] The 16B11.1 antibody was produced using pcDNA3.4-16B11.1_HC and pcDNA3.4-16B11_LC. Specifically, approximately 6.0 × 10 6ExpiCHO-S cells (Thermo Fisher Scientific, A29127) were cultured to a cell density of 16B11.1 / mL and transfected with pcDNA3.4-16B11.1_HC and pcDNA3.4-16B11_LC using the ExpiFectamine CHO Transfection Kit (Thermo Fisher Scientific, A29129) and cultured for 12 days. The culture supernatant was purified using MabSelectSuRe to obtain a purified fully human antibody. The resulting antibody is designated 16B11.1. The nucleotide sequence of the heavy chain of 16B11.1 is shown in SEQ ID NO: 3, the amino acid sequence encoded thereby is shown in SEQ ID NO: 4, the nucleotide sequence of the light chain of the antibody is shown in SEQ ID NO: 7, and the amino acid sequence encoded thereby is shown in SEQ ID NO: 8. 16B12.1 can be produced by the same method as above using pcDNA3.4-16B12.1_HC and pcDNA3.4-16B12_LC. The nucleotide sequence of the heavy chain of 16B12.1 is shown in SEQ ID NO: 9, the amino acid sequence encoded thereby is shown in SEQ ID NO: 10, the nucleotide sequence of the light chain of the antibody is shown in SEQ ID NO: 11, and the amino acid sequence encoded thereby is shown in SEQ ID NO: 12.

[0198] Example 4: Identification of epitope site on antigen to which antibody binds Example 4-1: Epitope mapping by hydrogen-deuterium exchange mass spectrometry To identify the epitope of 16B11.1, hydrogen-deuterium exchange mass spectrometry (HDX-MS) was performed. As a result, the human TSPAN8 region corresponding to amino acids 126-155 of SEQ ID NO: 2 was detected as a region in which the degree of deuterium exchange decreased in the presence of 16B11.1. Based on these results, it was estimated that the human TSPAN8 region corresponding to amino acids 126-155 of SEQ ID NO: 2 is the epitope of 16B11.1.

[0199] [Example 4-2: Narrowing down important epitopes by introducing mutations into human TSPAN8] To confirm whether the region predicted in Example 4-1 is the epitope of 16B11.1, chimeric proteins were prepared in which the region was replaced with the homologous region of mouse or rat TSPAN8, and binding was evaluated. Amino acids 126 to 155 of mouse TSPAN8, which correspond to amino acids 126 to 155 of human TSPAN8, are shown in SEQ ID NO: 41, and amino acids 126 to 155 of rat TSPAN8 are shown in SEQ ID NO: 42. To generate cells expressing a fusion protein of TSPAN8 and GFP, the human TSPAN8 sequence was excised from the TSPAN8 (Myc-DDK-tagged)-Human tetraspanin 8 (TSPAN8) (ORIGENE, RC202694) used in Examples 1-4 using restriction enzymes. The excised human TSPAN8 sequence was subcloned into the pCMV6-AC-GFP vector (ORIGENE, PS100010) (hereinafter referred to as the "human TSPAN8-GFP expression vector"). Furthermore, using the In-Fusion® HD Cloning Kit (Takara Bio, 639633), a vector was generated in which the sequence corresponding to amino acids 126 to 155 of the human TSPAN8-GFP expression vector set forth in SEQ ID NO: 2 was replaced with the sequence of amino acids 126 to 155 of mouse or rat TSPAN8. The resulting vectors were transfected into CHO-K1 cells to generate cells transiently expressing human TSPAN8-GFP protein, human-mouse TSPAN8-GFP chimeric protein, or human-rat TSPAN8-GFP chimeric protein. These cells are designated wild-type human TSPAN8-expressing CHO-K1 cells, human-mouse TSPAN8 chimeric protein-expressing CHO-K1 cells, or human-rat TSPAN8 chimeric protein-expressing CHO-K1 cells, respectively. Additionally, CHO-K1 cells transfected with the pCMV6-AC-GFP vector (referred to as mock cells) were generated. The binding of 16B11, 16B12, and TAL69 to GFP-positive cells in these cells was measured by flow cytometry. No reduction in TAL69 binding was observed in either human-rat or human-mouse TSPAN8 chimeric protein-expressing CHO-K1 cells. 16B11 and 16B12 showed binding to human-rat TSPAN8 chimeric protein-expressing CHO-K1 cells equivalent to that to wild-type human TSPAN8-expressing CHO-K1 cells, whereas binding to human-mouse TSPAN8 chimeric protein-expressing CHO-K1 cells was attenuated compared to wild-type human TSPAN8-expressing CHO-K1 cells (Figure 5-1). To identify the amino acid sequence responsible for the attenuation of binding activity to CHO-K1 cells expressing human-mouse TSPAN8 chimeric protein, we compared the amino acid sequences corresponding to amino acids 126 to 155 of the human TSPAN8 sequence for human, mouse, rat, and cynomolgus monkey TSPAN8 proteins (Figure 5-2). The sequences corresponding to amino acids 126 to 155 of the mouse, rat, and cynomolgus monkey TSPAN8 sequences are shown in SEQ ID NOs: 41, 42, and 43, respectively. As a result, the only amino acid sequence difference between the mouse TSPAN8 protein and the human TSPAN8 protein was the 131st amino acid. Based on this information, we inferred that the threonine (T) at position 131 of the human TSPAN8 protein shown in SEQ ID NO: 2 is important for binding to 16B11 or 16B12. Furthermore, to confirm whether the amino acid at position 131 of human TSPAN8 contributes to the binding of 16B11 or 16B12, we generated variants with this amino acid substitution. Specifically, we constructed vectors encoding human TSPAN8-GFP fusion proteins in which the threonine (T) at position 131 of the human TSPAN8 amino acid sequence of SEQ ID NO: 2 was replaced with alanine (A) or asparagine (N) (referred to as "human TSPAN8(T131A)-GFP" or "human TSPAN8(T131N)-GFP," respectively). We then transfected these vectors into CHO-K1 cells to construct cells transiently expressing human TSPAN8(T131A)-GFP or human TSPAN8(T131N)-GFP. These cells are referred to as human TSPAN8(T131A)-expressing CHO-K1 cells or human TSPAN8(T131N) cells. The binding of 16B11, 16B12, and TAL69 to GFP-positive cells in these cells was measured by flow cytometry (Figure 5-3). The binding of 16B11 and 16B12 to human TSPAN8 (T131A)-expressing CHO-K1 cells and human TSPAN8 (T131N)-expressing CHO-K1 cells was weakened compared to the binding to wild-type human TSPAN8-expressing CHO-K1 cells. On the other hand, TAL69 exhibited comparable binding activity to both cells. The results indicated that the threonine at position 131 in the region of human TSPAN8 corresponding to amino acids 126 to 155 of SEQ ID NO: 2, which was identified as the epitope, is essential for the binding of 16B11 and 16B12. The ΔMFI values ​​calculated by subtracting the MFI of binding to mock cells from the MFI of binding to each TSPAN8-expressing cell are shown in Table 2-1. In addition, the relative ΔMFI values ​​of binding to human-mouse TSPAN8 chimeric protein-expressing CHO-K1 cells, human-rat TSPAN8 chimeric protein-expressing CHO-K1 cells, and human TSPAN8 (T131A or T131N)-expressing CHO-K1 cells, when the ΔMFI of binding to wild-type human TSPAN8-expressing CHO-K1 cells in Table 2-1 is set to 100, are shown in Table 2-2. [Table 2-1] [Table 2-2]

[0200] [Example 4-3: Binding competition experiment] It was examined whether 16B11 or 16B12 competes with other anti-TSPAN8 antibodies (9F6, 18C10, or TAL69) for binding to TSPAN8. To examine the competition between 16B11 and other anti-TSPAN8 antibodies (9F6, 18C10, or TAL69), the binding of 16B11 to NSC-15CF cells was measured by flow cytometry (Experiment 1; Figure 6-1). 5 Alexa Fluor 647-labeled 16B11 (final concentration: 1 μg / mL) and one of the other anti-TSPAN8 antibodies (16B11, 9F6, 18C10, or TAL69) were added to NSC-15CF at a final concentration of 100 μg / mL, and the amount of 16B11 binding to NSC-15CF was measured using a flow cytometer. As negative control antibodies, Ultra-LEAF Purified Mouse IgG1,κ Isotype Control Antibody, Ultra-LEAF Purified Mouse IgG2a,κ Isotype Control Antibody (BioLegend, 400264), and Ultra-LEAF Purified Mouse IgG2b,κ Isotype Control Antibody were used. As a result, 16B11 binding was attenuated only when 16B11 itself was added. To examine the competition between 16B12 and other anti-TSPAN8 antibodies (16B11, 9F6, 18C10, or TAL69), the change in the amount of 16B11, 9F6, or 18C10 bound to NSC-15CF by 16B12 was measured by flow cytometry (Experiment 2; Figure 6-2). 5Alexa Fluor 647-labeled 16B11, 9F6, or 18C10 (final concentration: 0.25 μg / mL) and 16B12 (final concentration: 100 μg / mL) were added to NSC-15CF cells, and the amount of 16B11, 9F6, or 18C10 bound to NSC-15CF cells was measured using a flow cytometer. LEAF Purified Mouse IgG3,κ Isotype Ctrl Antibody was used as a negative control antibody. The results showed that 16B12 attenuated the binding of 16B11, but did not affect the binding of other antibodies. A similar competition experiment was performed using Alexa Fluor 647-labeled 16B12 and other anti-TSPAN8 antibodies (16B11, 16B12, 9F6, 18C10, or TAL69) (Experiment 3; Figure 6-3). The negative control antibodies used were LEAF Purified Mouse IgG3,κ Isotype Control Antibody, Ultra-LEAF Purified Mouse IgG2a,κ Isotype Control Antibody, Ultra-LEAF Purified Mouse IgG2b,κ Isotype Control Antibody, and Ultra-LEAF Purified Mouse IgG1,κ Isotype Control Antibody. The binding of 16B12 was attenuated only when 16B11 or 16B12 itself was added. Because the binding of 16B11 and 16B12 competed with each other, it was inferred that 16B11 and 16B12 recognize similar epitopes. The MFI of each labeled antibody binding was calculated by subtracting the MFI of unstained cells from the MFI of unstained cells. The value obtained when the isotype control was added as a competing antibody was set at 100. The relative values ​​obtained when the competing antibody was added are shown in Table 3. [Table 3]

[0201] [Example 5: Preparation of 60As6-Luc / GFP cells] 60As6-Luc / GFP cells were generated by expressing luciferase protein and green fluorescent protein (GFP) in 60As6 cells, a cell line derived from ascites of a gastric cancer patient, as follows. [Example 5-1: Preparation of virus solution containing Luc / GFP] Lentivirus was produced using L293T cells (Thermo Fisher Scientific, K4975-00) according to standard methods. Lentivirus was produced using MISSION® Lentiviral Packaging Mix (SIGMA, SHP001) and the pCDH-CMV-GL3-EF1a-GFP-T2A-puro modified vector (donated by Associate Professor Ryo Takahashi, Laboratory of Cellular and Molecular Biology, Graduate School of Biomedical Sciences, Hiroshima University (PLoS One, 2015, Vol. 10, e0123407; Front Biosci., 2008, Vol. 13, pp. 1619-1633)). Virus was filtered from the virus-containing cell culture supernatant using a 45 μm Millex®-HV filter (Merck Millipore, SLHV033RS) to obtain a virus solution, which was then frozen and stored at -80°C. [Example 5-2: Viral infection of 60As6 cells] 60As6 cells (gift from Dr. Ikichi Yanagihara, National Cancer Center) were infected with the virus according to standard methods. RPMI-1640 medium containing 10% FBS was used. Three days after infection, the culture medium was removed from the 60As6 cell plate and replaced with RPMI-1640 (selection medium) containing 10% FBS and 2 μg / mL Puromaycin (Thermo Fisher Scientific, A-11138-02). The cells were then repeatedly cultured and passaged in selection medium to remove uninfected cells and confirm complete virus removal. The established cells are referred to as 60As6-Luc / GFP cells. These cells expressed luciferase and GFP. Furthermore, high endogenous TSPAN8 expression was confirmed in 60As6-Luc / GFP cells by flow cytometry.

[0202] Example 6: Evaluation of ADCC activity of fully human antibody 16B11.1 The ADCC activity induced by the fully human anti-TSPAN8 antibody 16B11.1 was measured. ADCC activity can be measured by assessing the ability of effector cells to cytotoxicate target cells. In this example, NK cells (effector cells) were co-cultured with 60As6-Luc / GFP cells (target cells). NK cells were activated by 16B11.1, and the resulting cytotoxicity of 60As6-Luc / GFP cells due to ADCC activity was measured using luciferase as an indicator. NK cells were isolated from frozen human PBMCs (ePBMCs (registered trademark), Characterized Cryopreserved Human PBMCs, Cellular Technology Limited, CTL-CP1) using an NK Cell Isolation Kit (human NK Cell Isolation Kit, Miltenyi Biotec, 130-092-657) and cultured in NK cell medium (NK MACS Medium, Miltenyi Biotec, 130-114-429). 5 × 10 60As6-Luc / GFP cells were suspended in RPMI-1640 (SIGMA, R8758-500mL) medium containing 5% FBS (Hyclone, SH30084.03) in a round-bottom 96-well plate (Sumitomo Bakelite, MS-9096U). 3 cells / 25μL / well, 5×10 NK cells 4 16B11.1 or a negative control anti-KLH antibody (3G6) produced in-house was diluted to final concentrations of 1, 10, 100, 1,000, or 10,000 ng / mL and added at 25 μL / well. After 24 hours, the luciferase luminescence was measured using a luciferase quantification kit (ONE-Glo Luciferase Assay System, Promega, E6120). The luciferase luminescence indicates the viability of 60As6-Luc / GFP cells, and the decrease in luciferase luminescence can be used to measure cytotoxicity due to ADCC activity. The vertical axis in Figure 7 represents the relative value of the luciferase luminescence intensity for each sample, with the luciferase luminescence intensity measured in the medium alone set at 100% and the luciferase luminescence intensity in 60As6-Luc / GFP cells without antibody added set at 0%. The horizontal axis represents the concentration of antibody added to each well. As shown in Figure 7, target cells were damaged by ADCC activity only when 16B11.1 was added.

[0203] Example 7: Preparation of anti-TSPAN8-anti-CD3 bispecific antibody Example 7-1: Construction of bispecific antibody vector for anti-TSPAN8 antibody The heavy chain of 16B11.1 was modified with LALA mutations (L234A and L235A), substituting leucine (L) with alanine (A) at amino acid positions 238 and 239 (EU index: 234 and 235), respectively; knobs-into-holes mutations, substituting threonine (T) with serine (S), L with A, and tyrosine (Y) with valine (V) at amino acid positions 370, 372, and 411 (EU index: 366, 368, and 407), respectively; and an asparagine (N) with glycine (G) at amino acid position 301 (EU index: 297). The amino acid sequence of the designed 16B11.1 heavy chain is shown in SEQ ID NO: 6. The constructed vector is designated pcDNA3.4-16B11.1_HC_H.

[0204] Example 7-2: Construction of bispecific antibody vector for anti-human CD3 antibody The sequence of the humanized anti-CD3 antibody was designed based on the sequences of the heavy and light chain variable regions of the mouse anti-CD3 antibody described in Japanese Patent No. 5686953, according to the method described in the literature (Front Biosci., 2008, Vol. 13, pp. 1619-1633). Backmutations were introduced during this process. The three-dimensional structure information (PDB Code: 5FCS) was analyzed using the integrated computational chemistry system MOE provided by MOLSIS Inc., and the positions within the framework regions where backmutations should be introduced were determined. The humanized anti-CD3 antibody was designed to be arranged in the following order: heavy chain variable region (amino acids 1 to 125 of SEQ ID NO: 14), linker (amino acids 126 to 145 of SEQ ID NO: 14), light chain variable region (amino acids 146 to 254 of SEQ ID NO: 14), hinge (amino acids 255 to 269 of SEQ ID NO: 14), CH2 domain (amino acids 270 to 379 of SEQ ID NO: 14), and CH3 domain (amino acids 380 to 486 of SEQ ID NO: 14). Furthermore, SEQ ID NO: 14 contains the following mutations: (1) a cysteine ​​(C) substitution at amino acid positions 44 and 247; (2) a C-to-S substitution at amino acid position 259 (EU index: 220); (3) an LALA mutation substituting an L for an A at amino acids 273 and 274 (EU index: 234 and 235); (4) a knobs-into-holes mutation substituting a T for tryptophan (W) at amino acid position 405 (EU index: 366); and (5) an N-to-G substitution at amino acid position 336 (EU index: 297). To introduce these mutations, polynucleotides encoding the amino acid sequences containing each mutation were synthesized and inserted into the pcDNA3.1(+) vector (Thermo Fisher Scientific, V79020). The resulting vector is designated pcDNA3.1-m7_scFV_K. The nucleotide sequence of the prepared humanized anti-CD3 antibody is shown in SEQ ID NO: 13, and the amino acid sequence is shown in SEQ ID NO: 14.

[0205] Example 7-3: Preparation of anti-TSPAN8-anti-CD3 bispecific antibody To prepare a bispecific antibody composed of the Fab region of an anti-TSPAN8 antibody, the scFv region of an anti-CD3 antibody, and the Fc region, pcDNA3.4-16B11.1_HC_H, pcDNA3.4-16B11_LC, and pcDNA3.1-m7_scFV_K were transfected into ExpiCHO-S (registered trademark) cells in the same manner as in Example 3. The culture supernatant was analyzed by MabSelect The antibody was purified using SuRe and then further purified using a gel filtration column, HiLoad26 / 600 Superdex® 200 pg (GE Healthcare, 28-9893-36), to obtain a purified antibody with a purity of 95% or higher. The resulting antibody is referred to as the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody.

[0206] Example 8: Evaluation of binding activity of anti-TSPAN8-anti-CD3 bispecific antibodies The binding activity of the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody to TSPAN8 and CD3 was evaluated by ELISA using the TSPAN8 LEL protein or CD3εδ complex protein, respectively. Specifically, 1 μg / mL of TSPAN8 Protein, Human, Recombinant (Sino Biological, 15683-H07H) or Human CellExp CD3 epsilon & CD3 delta Heterodimer, Human Recombinant (BioVision, P1183-10) diluted with PBS was added at 30 μL / well to a 384-well white plate, MaxiSorp (Nunc, 460372). After incubation overnight at 4°C, the supernatant was removed, and 120 μL / well of Blocking One (Nacalai Tesque, 03953-95) was added. After incubation at room temperature for 1 hour, the supernatant was removed and the plate was washed twice with TBST buffer (Thermo Fisher Scientific, 28360). After that, 30 μL / well of anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody diluted in TBST containing 10% Blocking One was added and incubated at room temperature for 1 hour. The antibody solution was removed, washed twice with TBST buffer, and 30 μL / well of Goat Anti-Human Kappa, Mouse ads-HRP (SouthernBiotech, 2061-05) diluted 5000-fold in TBST containing 10% Blocking One was added and incubated at room temperature for 30 minutes. The antibody solution was removed, washed four times with TBST buffer, and 30 μL / well of BM Chemiluminescence ELISA Substrate (POD) (Roche, 11582950001) was added. After incubation at room temperature for 15 minutes, chemiluminescence was measured using an ARVO X3 (Perkin Elmer). The EC50 values ​​for TSPAN8 and CD3 were calculated to be 1.0 μg / mL (8.1 nM) and 4.6 μg / mL (36 nM), respectively (Figure 8).

[0207] Example 9: Evaluation of RTCC activity of anti-TSPAN8-anti-CD3 bispecific antibodies 500 mL of RPMI-1640 (Thermo Fisher Scientific, 11875-119) was mixed with 50 mL of FBS, 5 mL of MEM Non-essential Amino Acid (Merck, M7145), 5 mL of Sodium pyruvate (Merck, S8636), and GlutaMAX. The culture medium used was a mixture of 5 mL of 60As6-Luc / GFP cells (Thermo Fisher Scientific, 35050-061), 5 mL of penicillin-streptomycin (Thermo Fisher Scientific, 15070-063), and 5 mL of HEPES (Thermo Fisher Scientific, 15630-080) (hereinafter referred to as "culture medium" in this example). The 60As6-Luc / GFP cells prepared in Example 4 were cultured in the culture medium at a concentration of 2 × 10 5 The cell suspension, adjusted to cells / mL, was seeded in 50 μL aliquots into flat-bottom 96-well plates (IWAKI, 3860-096) and cultured at 37°C in a 5% CO2 incubator. After 3 hours, 1 × 10 cells were added to the culture medium. 6 Frozen human peripheral blood mononuclear cells (LP.CR.MNC 10M; AllCells LLC, 4W-270) prepared at a concentration of 100 cells / mL were seeded into 96-well plates in 100 μL aliquots. 50 μL of anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody (prepared to final concentrations of 0, 3, 10, 30, 100, 300, 1000, 3000, and 10,000 ng / mL) was added to each well. After 3 days of incubation at 37°C and 5% CO, the fluorescence (GFP) area of ​​each well was measured using an IncuCyte® ZOOM (Sartorius). The fluorescence area was used as an index of cell proliferation, and the cell proliferation curve is shown in Figure 9. The vertical axis of Figure 9 represents the relative fluorescence area of ​​60As6-Luc / GFP cells, with the fluorescence area of ​​wells containing medium alone set at 0% and the fluorescence area of ​​wells without antibody solution set at 100%. As a result, the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody exhibited cell growth inhibitory effects in vitro against TSPAN8-expressing gastric cancer cells, 60As6-Luc / GFP cells.

[0208] Example 10: Evaluation of efficacy of anti-TSPAN8-anti-CD3 bispecific antibody using patient ascites cells When the anti-TSPAN8-anti-CD3 bispecific antibody binds to cancer cells and immune cells contained in patient ascites, the immune cells are activated and the cancer cells are killed. The cancer cell cytotoxicity and immune cell activation effects of the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody were evaluated using the following methods. The patient ascites cells were cryopreserved after hemolysis of the patient ascites in the same manner as in Example 1-1. The thawed cells were cultured in the same culture medium as in Example 9 at 2 × 10 6The cell suspension was prepared to a concentration of 16B11.1 cells / mL, and 100 μL of the cell suspension was seeded into a flat-bottom 96-well plate. The test antibodies used were an anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody and a control antibody (a bispecific antibody consisting of anti-KLH (keyhole limpet hemocyanin) antibody, in which the Fab of 16B11.1 was replaced with the Fab of anti-KLH antibody, and anti-CD3 antibody). The test antibodies were diluted 10-fold from 10 μg / mL to 0.1 ng / mL. The test antibodies were added to the 96-well plate containing the seeded cells and then cultured at 37°C in a 5% CO2 incubator. Culture medium was used for the culture. After 3 days, the cells were harvested and seeded into a V-bottom microplate. Cells that had adhered to the plate at the time of harvesting were detached with Accutase (Innovative Cell Technologies, AT-104) and then added to the V-bottom microplate. After centrifugation at 720 × g for 2 minutes, the supernatant was removed, and 20 μL / well of a staining buffer (PBS containing 10% FBS, 0.09% NaN3, 2 mM EDTA) containing 1 / 40th the volume of Human BD Fc Block (Becton, Dickinson and Company, 564220) was added. FITC Mouse Anti-Human CD4 (Becton, Dickinson and Company, 550628), APC-H7 Mouse Anti-Human CD8 (Becton, Dickinson and Company, 560179), APC Mouse Anti-Human CD45 (Becton, Dickinson and Company, 555485), PE Mouse Anti-Human CD25 (Becton, Dickinson and Company, 555432), and Brilliant Violet 421 Anti-Human CD326 (EpCAM) Antibody (BioLegend, 324220) diluted in staining buffer were added to each well at 10 μL / well and incubated at 4°C for 50 minutes.After washing once with staining buffer, the cells were resuspended in staining buffer containing 1 / 200th the volume of 7-AAD solution. The binding of various antibodies to ascites cells was measured by flow cytometry using a CytoFLEX S (Beckman Coulter). Data analysis was performed using FlowJo. The 7-AAD-negative cell fraction, an indicator of viable cells, was developed using CD45, an indicator of immune cells, and EpCAM, an indicator of cancer cells. CD45-negative EpCAM-positive cancer cells were developed using Fsc(lin) and Ssc(lin), and the fragmented fraction was removed to determine the viable cancer cell count. Furthermore, the expression of the activation marker CD25 in CD4- or CD8-positive cells in the CD45-positive fraction was measured, and the MFI of anti-CD25-PE fluorescence intensity was calculated. Figure 10-1 shows the change in viable cancer cell count. The vertical axis indicates the relative cell count, with the number of cancer cells without antibody solution defined as 100%. 10-2 and 10-3 show the changes in CD25 expression levels on CD4-positive T cells and CD8-positive T cells due to test antibodies. The vertical axis shows the calculated relative MFI values ​​of anti-CD25-PE fluorescence intensity. As a result, as shown in Figure 10-1, the addition of anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody reduced the number of viable cancer cells in ascites, and as shown in Figures 10-2 and 10-3, activation of CD4-positive T cells and CD8-positive T cells in ascites was observed. These results suggest that the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody activates ascites CD4-positive T cells and CD8-positive T cells, leading to the killing of ascites cancer cells.

[0209] Example 11: In vivo antitumor evaluation of anti-TSPAN8-anti-CD3 bispecific antibodies The in vivo antitumor effect of the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody was evaluated using a gastric cancer peritoneal dissemination model. [Example 11-1: Preparation of expanded panT cells] Anti-CD3 antibody (BioLegend, 317315) dissolved in PBS at 3 μg / mL was added at 250 μL per well to a 24-well plate (IWAKI, 3820-024) and incubated at 4°C. The next day, the plate was washed twice with culture medium, after which culture medium was added and incubated at room temperature until cell seeding, as described below. PanT cells (including both CD4 and CD8 T cells) were isolated from HPBMC, human peripheral blood mononuclear cells, cryopreserved (LONZA, CC-2702). The PanT Cell Isolation Kit, human (Miltenyi Biotec, 130-096-535) was used for isolation, and the experiment was performed according to the attached protocol. The culture medium was removed from the 24-well plate, and 3 × 10 cells were added to the plate in culture medium. 6 500 μL of panT cells adjusted to cells / mL were seeded into each well. 500 μL of culture medium containing 20 ng / mL human IL-2 (PeproTech, 200-2) and 2 μg / mL anti-CD28 antibody (BioLegend, 302923) was added to each well, and the cells were cultured at 37°C in a 5% CO2 incubator. The cells were passaged onto new plates (IWAKI, 3810-006) 3, 5, and 7 days after the start of culture, and IL-2 was added to a final concentration of 10 ng / mL. Cells were collected 7 and 10 days after the start of culture and used in Example 10-2. The isolated and expanded cells are referred to as expanded panT cells.

[0210] [Example 11-2: Confirmation of drug efficacy in a gastric cancer peritoneal dissemination model] Seven 7-week-old C.B17 / Icr-scid / scidJcl female mice (CLEA Japan) were administered 1 × 10 intraperitoneally in each group. 6 60As6-Luc / GFP cells were transplanted at a concentration of 1 cell / 1mL / PBS. Six days after transplantation, the 60As6-Luc / GFP cells were divided into equal groups based on the amount of luminescence produced by luciferin, a substrate of luciferase introduced into the cells. Luciferin luminescence was used as an indicator of tumor volume. In detail, each individual received 3 mg of Luciferin (VivoGlo Luciferin (In Vivo Grade, Promega, P1043) dissolved in 0.5 mL of PBS was administered intraperitoneally, and the luminescence intensity was measured 10 minutes after administration using an IVIS Lumina II (PerkinElmer). Then, 1 × 10 cells were injected 7 and 10 days after the 60As6-Luc / GFP cell transplantation. 7 A suspension of 16 expanded panT cells in 0.5 mL of PBS and 0, 0.3, 1.0, or 3.0 mg / kg of the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody in 0.2 mL of PBS were intraperitoneally administered. Luciferin luminescence was measured 14 days after gastric cancer cell inoculation to assess tumor volume changes. Survival of peritoneal dissemination model mice was also monitored up to 34 days after 60As6-Luc / GFP cell inoculation. As shown in Figure 11-1, the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody group demonstrated significant tumor volume reduction. As shown in Figure 11-2, the 1.0 and 3.0 mg / kg groups demonstrated significant survival benefit. Table 4 shows median survival and assay results. The significance probability P values ​​in the table were determined by comparing the survival times of the control group (vehicle-treated group) and the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody-treated group using the log-rank test. ** indicates groups with P values ​​below the significance level of 0.01 / 3, corrected by the Bonferroni method. [Table 4]

[0211] Example 12: Effect of anti-TSPAN8-anti-CD3 bispecific antibodies on various carcinoma cell lines Example 12-1: Confirmation of binding activity of anti-TSPAN8 antibodies to various carcinoma cell lines Flow cytometry was used to measure the binding of Alexa Fluor 647-labeled 16B11 to gastric cancer cell lines (KATO III cells: Japanese Collection of Research Bioresources (JCRB), JCRB0611; NUGC-4 cells: RIKEN BioResource Research Center (BRC), RCB1939; 60As6-Luc / GFP cells), colon cancer cell lines (HT-29 cells: ATCC, HTB-38; LoVo cells: ATCC, CCL-229; GP2d cells: The European Collection of Authenticated Cell Cultures (ECACC), 95090714), pancreatic cancer cell line (AsPC-1 cells: ATCC, CRL-1682), esophageal cancer cell line (OE19 cells: ECACC, 96071721), and liver cancer cell line (Li-7 cells: RIKEN BRC, RCB1941). As a negative control antibody, we used an anti-KLH antibody (173A1) produced in-house and labeled with Alexa Fluor 647. Figure 12 shows a histogram of the binding of 16B11 to the negative control antibody in each carcinoma cell line.

[0212] Example 12-2: Evaluation of RTCC activity of anti-TSPAN8-anti-CD3 bispecific antibodies against various carcinoma cell lines KATOIII cells, NUGC-4 cells, HT-29 cells, LoVo cells, GP2d cells, AsPC-1 cells, OE19 cells, and Li-7 cells were cultured in culture medium at 2 × 10 5 The cells were prepared at a concentration of 1 × 10 cells / mL, and 50 μL of each was seeded into a flat-bottom 96-well plate (IWAKI, 3860-096) and cultured at 37°C in a 5% CO2 incubator. 6Frozen human peripheral blood mononuclear cells (LONZA, CC-2702) prepared at a concentration of 100 μL / mL were seeded into 96-well plates during incubation. Anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody was diluted with culture medium at a 2:1 common ratio, with a maximum concentration of 40 μg / mL or 20 μg / mL. 50 μL of the diluted anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody was added (maximum final concentration: 10 μg / mL or 5 μg / mL). The 96-well plates containing cancer cell lines, frozen human peripheral blood mononuclear cells, and antibodies were cultured at 37°C in a 5% CO2 incubator. After 3 days, the cells were harvested and seeded into V-bottom microplates. Cells that had adhered to the culture plate at the time of harvesting were detached with Accutase (Innovative Cell Technologies, AT104) and then added to the V-bottom microplates. After centrifugation at 720 × g for 2 minutes, the supernatant was removed, and 20 μL of staining buffer containing 1 / 40th volume of Human BD Fc Block was added to each well. Ten μL of each of APC Mouse Anti-Human CD4 (Becton, Dickinson and Company, 555349), APC-H7 Mouse Anti-Human CD8, Brilliant Violet 421 Mouse Anti-Human CD45 (Becton, Dickinson and Company, 563879), and PE Mouse Anti-Human CD25 diluted in staining buffer was added to each well and incubated at 4°C for 1 hour. After washing once with staining buffer, 1 / 200th volume of 7-AAD solution was added. The cells were resuspended in staining buffer, and antibody binding was measured by flow cytometry using a CytoFLEX S. Data analysis was performed using FlowJo. The number of CD45-negative cells in the 7-AAD-negative cell fraction, an indicator of viable cells, was taken as the number of viable cancer cells. The number without antibody solution was taken as 100%.Furthermore, the expression of the activation marker CD25 in CD4- or CD8-positive cells in the CD45-positive fraction was calculated as the fold change in MFI of anti-CD25-PE fluorescence intensity, relative to the value in the absence of antibody solution, which was set to 1. As shown in Figure 13-1, the addition of the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody reduced the number of viable TSPAN8-expressing cancer cells. Furthermore, as shown in Figures 13-2 and 13-3, activation of CD4-positive T cells and CD8-positive T cells, respectively, was observed.

[0213] [Example 12-3: Confirmation of drug efficacy in a human PBMC-transfected HT-29 cell subcutaneous tumor-bearing model] Normal human PBMC (Precision for Medicine, 33,000-10M) at 1.25 × 10 7 The cells were suspended in PBS at a concentration of 2.5 × 10 cells / mL and injected into 6-week-old NOD / Shi-scid, IL-2RγKO (NOG) female mice (In Vivo Science). 6 The cells were injected into the tail vein of the mice at a concentration of 5 × 10 cells / 200 μL. Ten days after the human PBMC transfer, HT-29 cells were added to the mice at a concentration of 5 × 10 cells / 200 μL. 7 Suspend in PBS to a concentration of 5 x 10 6 Mice were subcutaneously implanted with HT-29 cells at a concentration of 100 μL. Ten days after implantation, tumor volumes were measured using calipers and the groups were divided equally (n = 10). Administration of the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody began on the same day. The first day of administration was defined as day 0. On days 0, 4, and 7, mice were intravenously administered PBS or 0.3, 1, 3, or 10 mg / kg of the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody. Tumor volumes were measured on days 0, 4, 7, and 11 (Figure 14-1). Tumor volume [mm 3 ] was calculated using the following formula: (Length of tumor long axis [mm]) × (Length of tumor short axis [mm]) 2 ×0.5 As shown in Figure 14-2, the anti-TSPAN8 (16B11)-anti-CD3 bispecific antibody significantly inhibited the growth of HT-29 tumors at doses of 0.3, 1, 3, and 10 mg / kg. [Industrial Applicability]

[0214] The anti-TSPAN8 antibodies and fusions thereof, such as anti-TSPAN8-anti-CD3 bispecific antibodies, of the present invention are expected to be useful for cancer treatment. Furthermore, the polynucleotides, expression vectors, transformed host cells, and antibody production methods of the present invention are useful for producing the anti-TSPAN8 antibodies and fusions thereof. [Sequence List Free Text]

[0215] Numerical headings in the sequence listing below <223> The "Artificial Sequence" is described below. Specifically, SEQ ID NO: 2 represents the amino acid sequence of human TSPAN8-Myc-DDK, and the nucleotide sequence represented by SEQ ID NO: 1 encodes the amino acid sequence of human TSPAN8 represented by SEQ ID NO: 2. SEQ ID NO: 4, 6, or 10 represents the amino acid sequence of the heavy chain of an anti-TSPAN8 antibody, and the nucleotide sequence represented by SEQ ID NO: 3, 5, or 9 encodes the amino acid sequence of the heavy chain of the anti-TSPAN8 antibody represented by SEQ ID NO: 4, 6, or 10. SEQ ID NO: 8 or 12 represents the amino acid sequence of the light chain of an anti-TSPAN8 antibody, and the nucleotide sequence represented by SEQ ID NO: 7 or 11 encodes the amino acid sequence of the light chain of the anti-TSPAN8 antibody represented by SEQ ID NO: 8 or 12. SEQ ID NO: 14 represents the amino acid sequence of a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked, and the nucleotide sequence represented by SEQ ID NO: 13 encodes the amino acid sequence of a polypeptide in which an anti-CD3-scFv region and a second Fc polypeptide are linked, as represented by SEQ ID NO: 14. SEQ ID NOs: 15 to 23 are the amino acid sequences of various linkers described in the detailed description of the invention. SEQ ID NOs: 24 to 37 are the amino acid sequences of the CDRs and variable regions of 16B11 and 16B12. SEQ ID NOs: 38 to 40 are the amino acid sequences of signal sequences. SEQ ID NOs: 41 to 43 are the amino acid sequences of the regions of amino acids 126 to 155 of mouse, rat, and cynomolgus monkey TSPAN8, respectively.

Claims

1. An anti-TSPAN8 antibody or an antigen-binding fragment thereof selected from the following (a) and (b): to: (a) CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 4; CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 in sequence number 4, and heavy chain variable CDR3 comprising the amino acid sequence from amino acid numbers 99 to 110 of SEQ ID NO:4 region, and a CD consisting of the amino acid sequence from amino acid numbers 24 to 34 of SEQ ID NO: 8 R1, CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 8, and and a CDR3 consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO:

8. an anti-TSPAN8 antibody or antigen-binding fragment thereof, comprising a chain variable region; (b) CDR1 consisting of the amino acid sequence from amino acid numbers 31 to 35 of SEQ ID NO: 10; CDR2 consisting of the amino acid sequence from amino acid numbers 50 to 66 of SEQ ID NO: 10, and A CDR3 consisting of the amino acid sequence from amino acid numbers 99 to 110 of sequence number 10. The amino acid sequence of amino acid numbers 24 to 34 of SEQ ID NO: 12 is CDR1 consisting of the amino acid sequence from amino acid numbers 50 to 56 of SEQ ID NO: 12 DR2, and a CD consisting of the amino acid sequence from amino acid numbers 89 to 96 of SEQ ID NO:

12. An anti-TSPAN8 antibody or antigen-binding fragment thereof, comprising a light chain variable region comprising R3.

2. The anti-TSPAN8 antibody or its derivatives according to claim 1, which is selected from the following (a) and (b): Antigen-binding fragments: (a) a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 4; and a light chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO:

8. an anti-TSPAN8 antibody or antigen-binding fragment thereof comprising a region; (b) a heavy chain variable region consisting of the amino acid sequence from amino acid numbers 1 to 121 of SEQ ID NO: 10; and a light chain variant consisting of the amino acid sequence from amino acid numbers 1 to 107 of SEQ ID NO:

12. An anti-TSPAN8 antibody or antigen-binding fragment thereof, comprising a variable region.

3. The anti-TSPAN8 antibody of claim 2, selected from the following (a) and (b): (a) a heavy chain consisting of the amino acid sequence of SEQ ID NO: 4 and a light chain consisting of the amino acid sequence of SEQ ID NO: 8 an anti-TSPAN8 antibody comprising the nucleotide sequence of (b) a heavy chain consisting of the amino acid sequence of SEQ ID NO: 10 and a nucleic acid sequence consisting of the amino acid sequence of SEQ ID NO: 12 An anti-TSPAN8 antibody comprising a light chain comprising:

4. The anti-TSPAN8 antibody according to any one of claims 1 to 3, or a polypeptide thereof, which is post-translationally modified. Antigen-binding fragment.

5. Post-translational modifications include pyroglutamylation of the N-terminus of the heavy chain variable region and / or deletion of a lysine at the C-terminus of the heavy chain. The anti-TSPAN8 antibody or antigen-binding fragment thereof of claim 4, wherein the antibody or antigen-binding fragment is deficient.

6. The anti-TSPAN8 antibody or its antigen-binding flag according to any one of claims 1 to 5. fusion or conjugate of ment or anti-TSPAN8 according to any one of claims 1 to 5. A cell expressing an antibody or an antigen-binding fragment thereof on the cell surface.

7. An anti-TSPAN8 antibody according to any one of claims 1 to 5 for use in the treatment of cancer. or an antigen-binding fragment thereof, or the fusion, conjugate or cell of claim 6. Cell.

8. The anti-TSPAN8 antibody or its antigen-binding flag according to any one of claims 1 to 5. or the fusion, complex or cell of claim 6, further comprising a pharmaceutically acceptable carrier. A pharmaceutical composition comprising an excipient as described above.

9. The pharmaceutical composition according to claim 8 for the treatment of cancer.

10. A method according to any one of claims 1 to 5 in the manufacture of a pharmaceutical composition for the treatment of cancer. Use of an anti-TSPAN8 antibody or antigen-binding fragment thereof, or the fusion protein of claim 6. Use of the complex, composite or cell.

Images