Antibodies against PD-l1

Abstract

To provide novel classes of antibodies capable of binding human PD-L1, and to provide bispecific antibodies capable of binding human PD-L1 and capable of binding human CD3.SOLUTION: Provided is an antibody comprising an antigen-binding region capable of binding to human PD-L1, in which the antibody inhibits the binding of human PD-L1 to human PD-1 and competes for binding to human PD-L1 with an antibody comprising a VH sequence and VL sequence as set forth in particular amino acid sequences but does not compete for binding to human PD-L1 with an antibody comprising a VH sequence and VL sequence as set forth in particular amino acid sequences.SELECTED DRAWING: None

Description

Technical Field

[0001] Field of the Invention The present invention relates to novel antibodies and their use in medicine. In particular, the present invention relates to bispecific antibodies that can bind to human PD-L1 and can bind to human CD3. A novel class of antibodies that can bind to human PD-L1 is also provided. The present invention further relates to the use of the antibodies of the present invention, and to methods, nucleic acid constructs, and host cells for producing the antibodies of the present invention.

Background Art

[0002] Background of the Invention Programmed death ligand 1 (PD-L1, PDL1, CD274, B7H1, B7-H1) is a 33 kDa single-pass transmembrane type I membrane protein. Three types of PD-L1 isoforms have been described based on alternative splicing. PD-L1 belongs to the immunoglobulin (Ig) superfamily and contains one Ig-like C2-type domain and one Ig-like V-type domain. Freshly isolated T cells and B cells express only very small amounts of PD-L1, and only a small part (about 16%) of CD14+ monocytes constitutively express PD-L1 (Rietz and Chen, 2004 Am J Transplant 4: 8-14 (Non-Patent Document 1)). Interferon-γ (IFN-γ) is known to upregulate PD-L1 on tumor cells (Abiko et al., 2015 Br J Cancer 112:1501-1509 (Non-Patent Document 2); Dong et al., 2002 Nature Medicine 8(8): 793-800 (Non-Patent Document 3)).

[0003] PD-L1 inhibits anti-tumor immunity by (1) tolerizing tumor-reactive T cells by binding to the receptor PD-1 (CD279) of PD-L1 on activated T cells (Dong et al., supra; Latchman et al., 2004 Proc Natl Acad Sci USA 101, 10691-6 (Non-Patent Document 4)); (2) rendering tumor cells resistant to lysis via CD8+ T cells and Fas ligand through PD-1 signaling mediated by PD-L1 expressed by tumor cells (Azuma et al., 2008 Blood 111, 3635-43 (Non-Patent Document 5)); (3) tolerizing T cells by reverse signaling through CD80 (B7.1) expressed by T cells (Butte et al., 2007 Immunity 27, 111-22 (Non-Patent Document 6); Park et al., 2010 Blood 116, 1291-8 (Non-Patent Document 7)); and (4) promoting the development and maintenance of induced regulatory T cells (Francisco et al., 2009 J Exp Med 206, 3015-29 (Non-Patent Document 8)). PD-L1 is expressed in many human cancers, including melanoma, ovarian cancer, lung cancer, and colon cancer (Dong et al., supra).

[0004] PD-L1 blocking antibodies have shown clinical activity in several types of cancers known to overexpress PD-L1, including melanoma and NSCLC. For example, atezolizumab is a humanized IgG1 monoclonal antibody against PD-L1. Atezolizumab is currently in clinical trials as an immunotherapy agent for several indications, including various types of solid tumors (see, e.g., Rittmeyer et al., 2017 Lancet 389:255-265 (Non-Patent Document 9)). Avelumab, a PD-L1 antibody (Kaufman et al Lancet Oncol. 2016;17(10):1374-1385 (Non-Patent Document 10)), has been approved by the FDA as a treatment for adults and pediatric patients 12 years of age and older with metastatic Merkel cell carcinoma and is currently in clinical trials for several cancer indications, including bladder cancer, gastric cancer, head and neck cancer, mesothelioma, NSCLC, ovarian cancer, and kidney cancer. Durvalumab, a PD-L1 antibody, is approved for locally advanced or metastatic urothelial cancer indications and is in clinical development for multiple types of solid tumors and blood cancers (see, e.g., Massard et al., 2016 J Clin Oncol. 34(26):3119-25 (Non-Patent Document 11)). Additional anti-PD-L1 antibodies are described in WO2004004771 (Patent Document 1), WO2007005874 (Patent Document 2), WO2010036959 (Patent Document 3), WO2010077634 (Patent Document 4), WO2013079174 (Patent Document 5), WO2013164694 (Patent Document 6), WO2013173223 (Patent Document 7), and WO2014022758 (Patent Document 8).

[0005] Although significant progress has been made in cancer eradication, further improvement in antibody-based cancer therapy is still needed.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

[0007] [Non-Patent Document 1] Rietz and Chen, 2004 Am J Transplant 4: 8-14 [Non-Patent Document 2] Abiko et al., 2015 Br J Cancer 112:1501-1509 [Non-Patent Document 3] Dong et al., 2002 Nature Medicine 8(8): 793-800 [Non-Patent Document 4] Latchman et al., 2004 Proc Natl Acad Sci USA 101, 10691-6 [Non-Patent Document 5] Azuma et al., 2008 Blood 111, 3635-43 [Non-Patent Document 6] Butte et al., 2007 Immunity 27, 111-22 [Non-Patent Document 7] Park et al., 2010 Blood 116, 1291-8 [Non-Patent Document 8] Francisco et al., 2009 J Exp Med 206, 3015-29

Non-Patent Document 9

Non-Patent Document 10

Non-Patent Document 11

Summary of the Invention

[0008] In a first aspect, the present invention provides a novel anti-PD-L1 antibody comprising an antigen-binding region capable of binding to human PD-L1. The antibody of this second aspect of the present invention may be monospecific or multispecific. If it is multispecific, the multispecific antibody may or may not include an antigen-binding region capable of binding to human CD3ε.

[0009] The present invention further provides a bispecific antibody comprising an antigen-binding region capable of binding to human PD-L1 and an antigen-binding region capable of binding to human CD3ε (epsilon). Such bispecific antibodies have the following two effects.

[0010] First, the antibody binds to tumor cells expressing PD-L1 via its PD-L1 binding region. At the same time, the antibody binds to T cells via its CD3 binding region. Therefore, the antibody attracts T cells close to the tumor cells, thereby facilitating tumor cell killing by T cells. Furthermore, without being limited to any particular theory, bringing PD-L1-expressing cells and effector T cells close to each other may initiate interferon-γ release, as a result of which PD-L1 on tumor cells is upregulated, and therefore, more antibodies may be recruited to the tumor and its killing may be further enhanced.

[0011] Second, the bispecific antibody of the present invention inhibits the binding of human PD-L1 and human PD-1, and thus prevents PD-L1 from interfering with anti-tumor immunity via PD-1.

[0012] The CD3×PD-L1 bispecific antibody is particularly useful in therapeutic settings where specific targeting and T cell-mediated killing of PD-L1-expressing cells are desired. The CD3×PD-L1 bispecific antibody is highly efficient in mediating the killing of PD-L1-expressing cells, including, in some embodiments, cells with low PD-L1 copy numbers.

[0013] The antibody of the present invention can bind to PD-L1-expressing cells, such as MDA-MB-231 cells, PC3 cells, or HELA cells. Furthermore, the antibody of the present invention can inhibit the interaction between PD-L1 and PD-1 and mediate the killing of MDA-MB-231 cells, PC3 cells, and / or HELA cells by purified T cells or PBMCs.

[0014] In a further aspect, the present invention relates to the use of the antibody of the present invention in medicine, particularly the use of the antibody of the present invention for treating cancer.

[0015] [Invention 1001] An antibody comprising an antigen-binding region capable of binding to human PD-L1, which inhibits the binding of human PD-L1 and human PD-1, and (i) competes for binding to human PD-L1 with an antibody

[0511] comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15, but does not compete for binding to human PD-L1 with an antibody

[0547] comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22, or (ii) With respect to binding to human PD-L1, it competes with the antibody

[0547] comprising the VH sequence shown in SEQ ID NO: 18 and the VL sequence shown in SEQ ID NO: 22, but does not compete with the antibody

[0511] comprising the VH sequence shown in SEQ ID NO: 8 and the VL sequence shown in SEQ ID NO: 15 with respect to binding to human PD-L1. The antibody described above. [Invention 1002] The antibody of Invention 1001 that competes with the antibody

[0338] comprising the VH sequence shown in SEQ ID NO: 1 and the VL sequence shown in SEQ ID NO: 5 with respect to binding to human PD-L1. [Invention 1003] Any of the antibodies of the present invention, wherein the binding of any of the antibodies of the present invention to human PD-L1 is not replaced by the antibody

[0476] comprising the VH sequence shown in SEQ ID NO: 53 and the VL sequence shown in SEQ ID NO: 57. [Invention 1004] Any of the antibodies of the present invention, wherein the binding of any of the antibodies of the present invention to human PD-L1 is not blocked by the binding of the antibody

[0625] comprising the VH sequence shown in SEQ ID NO: 106 and the VL sequence shown in SEQ ID NO: 110. [Invention 1005] Any of the antibodies of the present invention, wherein the binding of any of the antibodies of the present invention to human PD-L1 is blocked by the antibody

[0547] comprising the VH sequence shown in SEQ ID NO: 18 and the VL sequence shown in SEQ ID NO: 22. [Invention 1006] (i) It can bind to the same human PD-L1 epitope as the antibody

[0338] comprising the VH sequence shown in SEQ ID NO: 1 and the VL sequence shown in SEQ ID NO: 5, or (ii) It can bind to the same human PD-L1 epitope as the antibody

[0511] comprising the VH sequence shown in SEQ ID NO: 8 and the VL sequence shown in SEQ ID NO: 15, or (iii) An antibody comprising the VH sequence shown in SEQ ID NO: 18 and the VL sequence shown in SEQ ID NO: 22, which can bind to the same human PD-L1 epitope as the antibody of

[0547] The antibody of the present invention 1001. [The present invention 1007] The binding of any one of the antibodies of the present invention 1001 - 1004 and 1006 to a mutant PD-L1 in which any one or more of the amino acid residues (R113) corresponding to position 113 in SEQ ID NO: 94, the amino acid residue (Y123) corresponding to position 123, and the amino acid residue (R125) corresponding to position 125 are substituted with alanine is reduced compared to the binding to wild-type PD-L1 having the amino acid sequence shown in SEQ ID NO: 94. A reduction in binding is determined when the fold change in the binding of the antibody is less than the average of the fold changes in binding across all alanine mutants - 1.5×SD, where SD is the standard deviation of all calculated fold change values of the antibody with respect to the mutant PDL1, and the fold change in binding is calculated as shown in Example 13

[0338] . Any one of the antibodies of the present invention 1001 - 1004 and 1006. [The present invention 1008] Any one of the antibodies of the present invention 1001 - 1002 and 1006 binds to an epitope on PD-L1 (SEQ ID NO: 94), and the epitope comprises the amino acid residue (R113) at position 113 of SEQ ID NO: 94, the amino acid residue (Y123) at position 123, and / or the amino acid residue (R125) at position 125. Any one of the antibodies of the present invention 1001 - 1002 and 1006. [The present invention 1009] The binding of any one of the antibodies of the present invention 1001, 1003, 1004, and 1006 to a mutant PD-L1 in which any one or more of the amino acid residues (F19) at the position corresponding to position 19 in SEQ ID NO: 94, the amino acid residue (F42) at the position corresponding to position 42, the amino acid residue (E45) at the position corresponding to position 45, the amino acid residue (K46) at the position corresponding to position 46, the amino acid residue (L94) at the position corresponding to position 94, and the amino acid residue (I116) at the position corresponding to position 116 are substituted with alanine is reduced as compared with wild-type PD-L1 having the amino acid sequence shown in SEQ ID NO: 94, and when the fold change in the binding of the antibody is smaller than the average of the fold changes in binding across all alanine mutants - 1.5×SD, the reduction in binding is determined, where SD is the standard deviation of all the calculated fold change values of the antibody against mutant PDL1, and the fold change in binding is calculated as shown in Example 13

[0511] , any one of the antibodies of the present invention 1001, 1003, 1004, and 1006. [The present invention 1010] Any one of the antibodies of the present invention 1001, 1003, 1004, and 1006 binds to an epitope on PD-L1 (SEQ ID NO: 94), and the epitope comprises one or more amino acid residues selected from the group consisting of the amino acid residue (E45) at position 45 of SEQ ID NO: 94, the amino acid residue (K46) at position 46, and / or the amino acid residue (L94) at position 94, any one of the antibodies of the present invention 1001, 1003, 1004, and 1006. [The present invention 1011] The binding of any of the antibodies of the present invention 1001, 1005, and 1006 to a mutant PD-L1 in which one or more of the amino acid residues (E58) at the position corresponding to position 58 in SEQ ID NO: 94 and the amino acid residue (R113) at the position corresponding to position 113 are substituted with alanine is reduced as compared to wild-type PD-L1 having the amino acid sequence shown in SEQ ID NO: 94. A reduction in binding is determined when the fold change in the binding of the antibody is less than the average of the fold changes in binding across all alanine mutants - 1.5×SD, where SD is the standard deviation of all calculated fold change values of the antibody with respect to mutant PDL1, and the fold change in binding is calculated as shown in Example 13

[0547] , any of the antibodies of the present invention 1001, 1005, and 1006. [The present invention 1012] Any of the antibodies of the present invention 1001, 1005, and 1006 binds to an epitope on PD-L1 (SEQ ID NO: 94), and the epitope contains the amino acid residue (E58) at position 58 of SEQ ID NO: 94 and / or the amino acid residue (R113) at position 113, any of the antibodies of the present invention 1001, 1005, and 1006. [The present invention 1013] The antigen-binding region capable of binding to human PD-L1 includes a heavy-chain variable region (VH) containing a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, and a light-chain variable region (VL) containing a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, and the VH CDR3 sequence is selected from the group consisting of the sequences shown in SEQ ID NO: 4, SEQ ID NO: 11, and SEQ ID NO: 21, any of the antibodies of the present invention. [The present invention 1014] The antigen-binding region capable of binding to human PD-L1 is (i) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:2, the CDR2 sequence shown in SEQ ID NO:3, and the CDR3 sequence shown in SEQ ID NO:4, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:6, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:7

[0338] , or (ii) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:9, the CDR2 sequence shown in SEQ ID NO:10, and the CDR3 sequence shown in SEQ ID NO:11, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:16, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:17

[0511] , or (iii) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:19, the CDR2 sequence shown in SEQ ID NO:20, and the CDR3 sequence shown in SEQ ID NO:21, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:23, a CDR2 having the sequence DDN, and a CDR3 having the sequence shown in SEQ ID NO:24

[0547] Any antibody of the present invention comprising [Invention 1015] Any antibody of the present invention, wherein the antigen-binding region capable of binding to human PD-L1 comprises a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VH sequence selected from the group consisting of the sequences shown in SEQ ID NO:1, SEQ ID NO:8, and SEQ ID NO:18. [Invention 1016] The antibody of any one of the present invention, wherein the antigen-binding region capable of binding to human PD-L1 comprises a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VL sequence selected from the group consisting of the sequences shown in SEQ ID NO:5, SEQ ID NO:15, and SEQ ID NO:22. [Invention 1017] The antigen-binding region capable of binding to human PD-L1 is (i) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:1, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:5

[0338] , or (ii) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:8, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:15

[0511] , or (iii) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:18, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:22

[0547] comprises the antibody of any one of the present invention. [Invention 1018] The VH and VL sequences each contain three CDR sequences, namely CDR1, CDR2, and CDR3, and four framework sequences, namely FR1, FR2, FR3, and FR4, and each combined FR1, FR2, FR3, and FR4 framework sequence of VH has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with the respective combined FR1, FR2, FR3, and FR4 framework sequences of the VH sequence, and the VH CDR sequences are not mutated, and each combined FR1, FR2, FR3, and FR4 framework sequence of VL has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with the respective combined FR1, FR2, FR3, and FR4 framework sequences of the VL sequence, and the VL CDR sequences are not mutated, the antibody of the present invention 1017. [The present invention 1019] Any of the antibodies of the present invention capable of inducing dose-dependent lysis of epithelial cells of adenocarcinoma, such as dose-dependent lysis of MDA-MB-231, via antibody-dependent cell-mediated cytotoxicity (ADCC). [The present invention 1020] The antibody of the present invention 1019 capable of reducing the number of cells in the culture of the epithelial cells by at least 5%, such as at least 6%, 7%, 8%, 9%, or at least 10% as a result of cell lysis. [The present invention 1021] ADCC is 51 The antibody of the present invention 1019 or 1020 measured in vitro by the Cr release assay method, for example, the assay method disclosed in Example 14. [The present invention 1022] ADCC is the epithelial cell together with a composition containing the antibody of the present invention 1019 or 1020 and an effector cell, such as peripheral blood mononuclear cells (PBMC), at 37°C, 5% CO 2Measured in vitro by incubating for 4 hours, the amount of antibody in the composition is in the range of 0.1 - 1 μg / mL, and the ratio of effector cells to epithelial cells is 100:1, the antibody of the present invention 1019 or 1020. [The present invention 1023] The lysis of the epithelial cells is measured in vitro in a luciferase reporter assay method as an alternative to ADCC, for example, the luminescence ADCC reporter bioassay method disclosed in Example 14, the antibody of the present invention 1019 or 1020. [The present invention 1024] ADCC is (i) contacting the culture of the epithelial cells with a composition comprising the antibody of the present invention 1023 and Jurkat human T cells (effector cells) stably expressing FcγRIIIa (CD16) and firefly luciferase at an effector cell:epithelial cell ratio of 1:1, (ii) adjusting the culture of the epithelial cells and the effector cells to room temperature for 15 minutes, (iii) incubating the culture of the epithelial cells and the effector cells with a luciferase substrate, and (iv) measuring the luciferase production in the cell culture by which is measured in vitro, the amount of antibody in the composition is in the range of 0.5 - 250 ng / mL, and the ratio of effector cells to epithelial cells is 1:1, the antibody of the present invention 1023. [The present invention 1025] The ADCC of the epithelial cells is measured by a luciferase reporter assay method, for example, the reporter assay method defined in the present invention 1023 or 1024, and then the ADCC observed after incubating the culture of the epithelial cells with a test composition containing any of the antibodies of the present invention 1019, 1020, 1023, and 1024 is at least 1.5 times the ADCC observed after incubating the culture of the epithelial cells with a composition containing a reference antibody. The ADCC is measured as relative light units (RLU), the antibody concentration in the test composition is the same as the antibody concentration in the composition containing the reference antibody and is in the range of 20 to 250 ng / ml, and the reference antibody is (a) an antibody comprising the VH sequence shown in SEQ ID NO:74 and the VL sequence shown in SEQ ID NO:78; and (b) an antibody comprising the VH sequence shown in SEQ ID NO:81 and the VL sequence shown in SEQ ID NO:85 selected from any of the antibodies of the present invention 1019, 1020, 1023, and 1024. [The present invention 1026] The antigen-binding region capable of binding to human PD-L1 is (i) the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5

[0338] , or (ii) the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15

[0511] , or (iii) the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22

[0547] comprised in any of the antibodies of the present invention. [The present invention 1027] An antibody comprising an antigen-binding region capable of binding to human PD-L1, wherein (i)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:33, the CDR2 sequence shown in SEQ ID NO:34, and the CDR3 sequence shown in SEQ ID NO:35, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:37, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:38

[0321] , or (ii)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:47, the CDR2 sequence shown in SEQ ID NO:48, and the CDR3 sequence shown in SEQ ID NO:49, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:51, a CDR2 having the sequence DVI, and a CDR3 having the sequence shown in SEQ ID NO:52

[0421] , or (iii)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:54, the CDR2 sequence shown in SEQ ID NO:55, and the CDR3 sequence shown in SEQ ID NO:56, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:58, a CDR2 having the sequence RDS, and a CDR3 having the sequence shown in SEQ ID NO:59

[0476] , or (iv)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:61, the CDR2 sequence shown in SEQ ID NO:62, and the CDR3 sequence shown in SEQ ID NO:63, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:65, a CDR2 having the sequence DDS, and a CDR3 having the sequence shown in SEQ ID NO:66

[0516] , or (v)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:107, the CDR2 sequence shown in SEQ ID NO:108, and the CDR3 sequence shown in SEQ ID NO:109, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:111, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:113

[0625] , (vi)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:68, the CDR2 sequence shown in SEQ ID NO:69, and the CDR3 sequence shown in SEQ ID NO:70, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:72, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:73

[0632] The antibody comprising the above. [Invention 1028] (i) The VH sequence shown in SEQ ID NO:32 and the VL sequence shown in SEQ ID NO:36

[0321] , or (ii) The VH sequence shown in SEQ ID NO:46 and the VL sequence shown in SEQ ID NO:50

[0421] , or (iii) The VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57

[0476] , or (iv) The VH sequence shown in SEQ ID NO:60 and the VL sequence shown in SEQ ID NO:64

[0516] , or The VH sequence shown in SEQ ID NO:106 and the VL sequence shown in SEQ ID NO:110

[0625] , (v) The VH sequence shown in SEQ ID NO:67 and the VL sequence shown in SEQ ID NO:71

[0632] The antibody of Invention 1027 comprising the above. [Invention 1029] Any of the antibodies of the present invention that is monovalent. [Invention 1030] Any of the antibodies of the present invention, wherein the antibody is a bivalent antibody having two antigen-binding regions capable of binding to human PD-L1, and the two antigen-binding regions have the same variable region sequence. [Invention 1031] Any antibody of the present invention is a bivalent bispecific antibody and, in addition to the (first) antigen-binding region capable of binding to human PD-L1, includes a (second) antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, and the second antigen is not human CD3ε, any antibody of the present invention. [Invention 1032] A bispecific antibody comprising an antigen-binding region capable of binding to human PD-L1 and an antigen-binding region capable of binding to human CD3ε (epsilon), wherein the antigen-binding region capable of binding to human PD-L1 has the characteristics shown in any of the present invention, the bispecific antibody. [Invention 1033] The antigen-binding region capable of binding to human CD3ε includes a heavy-chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light-chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, the bispecific antibody of Invention 1032. [Invention 1034] (i) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:2, the CDR2 sequence shown in SEQ ID NO:3, and the CDR3 sequence shown in SEQ ID NO:4, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:6, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:7, which can bind to human PD-L1

[0338] , and (a) a heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, which can bind to human CD3ε, or (ii) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:9, the CDR2 sequence shown in SEQ ID NO:10, and the CDR3 sequence shown in SEQ ID NO:11, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:16, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:17, which can bind to human PD-L1

[0338] , and (a) a heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, which can bind to human CD3ε, or (iii) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:19, the CDR2 sequence shown in SEQ ID NO:20, and the CDR3 sequence shown in SEQ ID NO:21, and a light chain variable region (VL) comprising a CDR1 sequence having the sequence shown in SEQ ID NO:23, a CDR2 sequence having the sequence DDN, and a CDR3 sequence having the sequence shown in SEQ ID NO:24, an antigen-binding region capable of binding to human PD-L1

[0547] , and (a) a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, a heavy chain variable region (VH), and a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, a light chain variable region (VL), an antigen-binding region capable of binding to human CD3ε The bispecific antibody of the present invention 1032 or 1033 comprising [The present invention 1035] The bispecific antibody according to any one of the present inventions 1032 to 1034, wherein the antigen-binding region capable of binding to human CD3ε comprises the VH sequence shown in SEQ ID NO:25 and the VL sequence shown in SEQ ID NO:29. [The present invention 1036] (i) The bispecific antibody has a lower affinity for human CD3ε binding compared to an antibody having an antigen-binding region that can comprise the VH sequence shown in SEQ ID NO:25 and the VL sequence shown in SEQ ID NO:29, preferably, the affinity is at least 1 / 2-fold, for example, at least 1 / 5-fold, for example, at least 1 / 10-fold, for example, at least 1 / 25-fold, for example, at least 1 / 50-fold, and (ii) When the bispecific antibody is assayed as described in Example 11 herein, it can mediate concentration-dependent cytotoxicity of MDA-MB-231 cells, PC-3 cells, and / or HELA cells when using PBMC or purified T cells as effector cells. The bispecific antibody of the present invention 1032. [The present invention 1037] The antigen-binding region capable of binding to human CD3ε is (i) a heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:99, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (ii) a heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:100, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (iii) a heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:101, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (iv) a heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:102, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (v) A heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:103, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (vi) A heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:104, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (vii) A heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:105, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31 The bispecific antibody of the present invention 1036 comprising [The present invention 1038] The antigen-binding region capable of binding to human CD3ε is (i) The VH sequence shown in SEQ ID NO:39 and the VL sequence shown in SEQ ID NO:29, or (ii) The VH sequence shown in SEQ ID NO:40 and the VL sequence shown in SEQ ID NO:29, or (iii) The VH sequence shown in SEQ ID NO:41 and the VL sequence shown in SEQ ID NO:29, or (iv) The VH sequence shown in SEQ ID NO:42 and the VL sequence shown in SEQ ID NO:29, or (v) The VH sequence shown in SEQ ID NO: 43 and the VL sequence shown in SEQ ID NO: 29, or (vi) The VH sequence shown in SEQ ID NO: 44 and the VL sequence shown in SEQ ID NO: 29, or (vii) The VH sequence shown in SEQ ID NO: 45 and the VL sequence shown in SEQ ID NO: 29 The bispecific antibody of the present invention 1036 or 1037, comprising: [The present invention 1039] A multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, wherein the antigen-binding region capable of binding to human PD-L1 has the characteristics shown in any of the present inventions 1001 to 1031. The multispecific antibody. [The present invention 1040] The antigen-binding region capable of binding to human PD-L1 includes a heavy chain variable region (VH) comprising a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, and a light chain variable region (VL) comprising a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence. The multispecific antibody of the present invention 1039, wherein the VH CDR3 sequence is selected from the group consisting of the sequences shown in SEQ ID NO: 4

[0338] , SEQ ID NO: 11

[0511] , and SEQ ID NO: 21

[0547] . [The present invention 1041] The first antigen-binding region capable of binding to human PD-L1 is (i) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO: 2, the CDR2 sequence shown in SEQ ID NO: 3, and the CDR3 sequence shown in SEQ ID NO: 4, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO: 6, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO: 7

[0338] , or (ii) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:9, the CDR2 sequence shown in SEQ ID NO:10, and the CDR3 sequence shown in SEQ ID NO:11, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:16, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:17

[0511] , or (iii) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:19, the CDR2 sequence shown in SEQ ID NO:20, and the CDR3 sequence shown in SEQ ID NO:21, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:23, a CDR2 having the sequence DDN, and a CDR3 having the sequence shown in SEQ ID NO:24

[0547] The multispecific antibody of the present invention 1040, comprising [The present invention 1042] The first antigen-binding region capable of binding to human PD-L1 comprises a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VH sequence selected from the group consisting of the sequences shown in SEQ ID NO:1

[0338] , SEQ ID NO:8

[0511] , and SEQ ID NO:18

[0547] , the multispecific antibody of the present invention 1040 or 1041. [The present invention 1043] The first antigen-binding region capable of binding to human PD-L1 comprises a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VL sequence selected from the group consisting of the sequences shown in SEQ ID NO:5

[0338] , SEQ ID NO:15

[0511] , and SEQ ID NO:22

[0547] , the multispecific antibody of any one of the present inventions 1040 to 1042. [The present invention 1044] The first antigen-binding region capable of binding to human PD-L1 is (i) A VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:1, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:5

[0338] , or (ii) A VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:8, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:15

[0511] , or (iii) A VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:18, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:22

[0547] A multispecific antibody according to any one of 1040 to 1043 of the present invention, comprising: [1045 of the present invention] The VH and VL sequences each include three CDR sequences, namely CDR1, CDR2, and CDR3, and four framework sequences, namely FR1, FR2, FR3, and FR4, and each combined FR1, FR2, FR3, and FR4 framework sequence of VH has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with the corresponding combined FR1, FR2, FR3, and FR4 framework sequences of the VH sequence, and the VH CDR sequences are not mutated, and each combined FR1, FR2, FR3, and FR4 framework sequence of VL has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with the corresponding combined FR1, FR2, FR3, and FR4 framework sequences of the VL sequence, and the VL CDR sequences are not mutated, the multispecific antibody according to any one of the present invention 1040 to 1044. [The present invention 1046] The first antigen-binding region capable of binding to human PD-L1 is (i) the VH sequence shown in SEQ ID NO: 1 and the VL sequence shown in SEQ ID NO: 5

[0338] , or (ii) the VH sequence shown in SEQ ID NO: 8 and the VL sequence shown in SEQ ID NO: 15

[0511] , or (iii) the VH sequence shown in SEQ ID NO: 18 and the VL sequence shown in SEQ ID NO: 22

[0547] The multispecific antibody according to any one of the present invention 1040 to 1045, comprising [The present invention 1047] A multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, (i) With respect to binding to human PD-L1, it competes with an antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15, but does not compete with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22, or (ii) With respect to binding to human PD-L1, it competes with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22, but does not compete with an antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15, The multispecific antibody. [Inventive Item 1048] The multispecific antibody of Inventive Item 1047 that competes with an antibody comprising the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5 with respect to binding to human PD-L1. [Inventive Item 1049] A multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, The multispecific antibody, wherein the binding of the antibody to human PD-L1 is not replaced by an antibody comprising the VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57. [Inventive Item 1050] The multispecific antibody of Inventive Item 1049 that inhibits the binding between human PD-L1 and human PD-1. [Inventive Item 1051] The multispecific antibody of Inventive Item 1049 or 1050 that competes with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 with respect to binding to human PD-L1. [Inventive Item 1052] A multispecific antibody according to any one of the present inventions 1049 to 1051, wherein the binding of the multispecific antibody according to any one of the present inventions 1049 to 1051 to human PD-L1 is blocked by an antibody comprising the VH sequence shown in SEQ ID NO: 18 and the VL sequence shown in SEQ ID NO: 22. [The present invention 1053] A multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, wherein the first antigen-binding region (i) can bind to the same human PD-L1 epitope as the antibody

[0338] comprising the VH sequence shown in SEQ ID NO: 1 and the VL sequence shown in SEQ ID NO: 5, or (ii) can bind to the same human PD-L1 epitope as the antibody

[0511] comprising the VH sequence shown in SEQ ID NO: 8 and the VL sequence shown in SEQ ID NO: 15, or (iii) can bind to the same human PD-L1 epitope as the antibody

[0547] comprising the VH sequence shown in SEQ ID NO: 18 and the VL sequence shown in SEQ ID NO: 22, the multispecific antibody. [The present invention 1054] A multispecific antibody according to any one of the present inventions 1040 to 1053, which is bispecific. [The present invention 1055] The multispecific antibody of the present invention 1054, which is bivalent. [The present invention 1056] A multispecific antibody according to any one of the present inventions 1040 to 1055, wherein the multispecific antibody according to any one of the present inventions 1040 to 1055 can bind to a second antigen, and the second antigen is not human CD3ε. [The present invention 1057] Any antibody of the present invention, which is a full-length antibody. [The present invention 1058] The antibody of the present invention 1057, which is a full-length IgG1 antibody. [The present invention 1059] Any antibody of the present invention that is an antibody fragment. [Invention 1060] Any antibody of Inventions 1032 to 1059 contains two half-molecules each containing an antigen-binding region, (i) The half-molecule containing the antigen-binding region capable of binding to human PD-L1 is chimeric, and / or (ii) The half-molecule containing the antigen-binding region capable of binding to human CD3ε (epsilon) is chimeric if present. Any antibody of Inventions 1032 to 1059. [Invention 1061] (i) The antigen-binding region capable of binding to human PD-L1 is humanized, and / or (ii) The antigen-binding region capable of binding to human CD3ε (epsilon) is humanized if present. Any antibody of the present invention. [Invention 1062] (i) The antigen-binding region capable of binding to human PD-L1 is human, and / or (ii) The antigen-binding region capable of binding to human CD3ε (epsilon) is human if present. Any antibody of the present invention. [Invention 1063] Each of the antigen-binding regions contains a heavy-chain variable region (VH) and a light-chain variable region (VL), and each of the variable regions contains three CDR sequences, CDR1, CDR2, and CDR3, respectively, and four framework sequences, FR1, FR2, FR3, and FR4, respectively. Any antibody of the present invention. [Invention 1064] The antibody of Invention 1063 containing two heavy-chain constant regions (CH) and two light-chain constant regions (CL). [Invention 1065] Any antibody of the present invention includes a first heavy chain and a second heavy chain, each of the first heavy chain and the second heavy chain includes at least a hinge region, CH2, and CH3 regions, in the first heavy chain, at least one amino acid at a position corresponding to a position selected from the group consisting of T366, L368, K370, D399, F405, Y407, and K409 (according to EU numbering) is substituted, and in the second heavy chain, at least one amino acid at a position corresponding to a position selected from the group consisting of T366, L368, K370, D399, F405, Y407, and K409 (according to EU numbering) is substituted, and the first heavy chain and the second heavy chain are not substituted at the same position, any antibody of the present invention. [Invention 1066] (i) In the first heavy chain, the amino acid at the position corresponding to F405 (according to EU numbering) is L, and in the second heavy chain, the amino acid at the position corresponding to K409 (according to EU numbering) is R, or (ii) In the first heavy chain, the amino acid at the position corresponding to K409 (according to EU numbering) is R, and in the second heavy chain, the amino acid at the position corresponding to F405 (according to EU numbering) is L, the antibody of Invention 1065. [Invention 1067] Any antibody of the present invention includes a first heavy chain and a second heavy chain, and the antibody induces effector functions via Fc to a lesser extent compared to an antibody identical except that it includes unmodified first and second heavy chains, and one or both heavy chains are modified. Any antibody of the present invention. [Invention 1068] The effector function via Fc is measured by determining Fc-mediated CD69 expression, by binding to an Fcγ receptor, by binding to C1q, or by inducing FcR cross-linking via Fc, the antibody of Invention 1067. [Invention 1069] The antibody of the present invention 1067 or 1068 has a heavy chain constant region and a light chain constant region modified such that Fc-mediated CD69 expression is reduced by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99%, or 100% compared to the wild-type antibody, wherein the Fc-mediated CD69 expression is measured in a functional assay based on PBMC, The antibody of the present invention 1067 or 1068. [The present invention 1070] Any of the antibodies of the present invention as described above, wherein the antibody comprises a first heavy chain and a second heavy chain, and in at least one of the first heavy chain and the second heavy chain, one or more amino acids at positions corresponding to positions L234, L235, D265, N297, and P331 in a human IgG1 heavy chain according to EU numbering are not L, L, D, N, and P, respectively. [The present invention 1071] The antibody of the present invention 1070, wherein in the first heavy chain and the second heavy chain, the positions corresponding to positions L234 and L235 in a human IgG1 heavy chain according to EU numbering are F and E, respectively. [The present invention 1072] The antibody of the present invention 1071 is a bispecific antibody comprising a first heavy chain and a second heavy chain, and in both the first heavy chain and the second heavy chain, the positions corresponding to positions L234 and L235 in a human IgG1 heavy chain according to EU numbering are F and E, respectively, and (i) the position corresponding to F405 in a human IgG1 heavy chain according to EU numbering in the first heavy chain is L, and the position corresponding to K409 in a human IgG1 heavy chain according to EU numbering in the second heavy chain is R, or (ii) the position corresponding to K409 in a human IgG1 heavy chain according to EU numbering in the first heavy chain is R, and the position corresponding to F405 in a human IgG1 heavy chain according to EU numbering in the second heavy chain is L. [The present invention 1073] The antibody of the present invention 1070, wherein in the first heavy chain and the second heavy chain, the positions corresponding to positions L234, L235, and D265 in the human IgG1 heavy chain according to EU numbering are F, E, and A, respectively. [The present invention 1074] The antibody of the present invention 1073 is a bispecific antibody comprising a first heavy chain and a second heavy chain, and the positions corresponding to positions L234, L235, and D265 in the human IgG1 heavy chain according to EU numbering of both the first heavy chain and the second heavy chain are F, E, and A, respectively, and (i) the position corresponding to F405 in the human IgG1 heavy chain according to EU numbering of the first heavy chain is L, and the position corresponding to K409 in the human IgG1 heavy chain according to EU numbering of the second heavy chain is R, or (ii) the position corresponding to K409 in the human IgG1 heavy chain according to EU numbering of the first heavy chain is R, and the position corresponding to F405 in the human IgG1 heavy chain according to EU numbering of the second heavy chain is L, the antibody of the present invention 1073. [The present invention 1075] Any of the antibodies of the present invention that do not bind to human PD-L2. [The present invention 1076] When measured as described in Example 8 herein, about 10 -8 M or less, for example, about 10 -9 M or less, for example, about 10 -10 M or less of K D Any of the antibodies of the present invention that bind to human PD-L1 with. [The present invention 1077] Any of the antibodies of the present invention that mediate concentration-dependent cytotoxicity of MDA-MB-231 cells, PC-3 cells, and / or HELA cells when assayed as described in Example 11 herein using purified T cells as effector cells. [The present invention 1078] (i) A nucleic acid sequence encoding a heavy chain sequence of an antibody comprising an antigen-binding region capable of binding to human PD-L1 as defined in any of the present inventions 1001 to 1031, and / or A nucleic acid sequence encoding a light chain sequence of an antibody comprising an antigen-binding region capable of binding to human PD-L1 as defined in any one of 1001 to 1031 of the present invention A nucleic acid construct comprising the same. [1079 of the present invention] (i) A nucleic acid sequence encoding a heavy chain sequence of an antibody comprising an antigen-binding region capable of binding to human CD3ε as defined in any one of 1033 to 1038 of the present invention, and (ii) A nucleic acid sequence encoding a light chain sequence of an antibody comprising an antigen-binding region capable of binding to human CD3ε as defined in any one of 1033 to 1038 of the present invention The nucleic acid construct of 1073 of the present invention, further comprising the same. [1080 of the present invention] An expression vector comprising the nucleic acid construct defined in 1078 or 1079 of the present invention. [1081 of the present invention] A host cell comprising the nucleic acid construct defined in 1078 or 1079 of the present invention or the expression vector defined in 1080 of the present invention. [1082 of the present invention] The host cell of 1081 of the present invention, which is a mammalian cell such as Chinese hamster ovary cells. [1083 of the present invention] A pharmaceutical composition comprising an antibody according to any one of 1001 to 1077 of the present invention and a pharmaceutically acceptable carrier. [1084 of the present invention] An antibody according to any one of 1001 to 1077 of the present invention or the pharmaceutical composition of 1083 of the present invention for use as a medicament. [1085 of the present invention] An antibody according to any one of 1001 to 1080 of the present invention or the pharmaceutical composition of 1070 of the present invention for use in the treatment of cancer. [1086 of the present invention] An antibody according to any one of 1001 to 1077 of the present invention or the pharmaceutical composition of 1070 of the present invention for use in the treatment of a cancer disease characterized by the presence of a solid tumor. [1087 of the present invention] An antibody of any one of the present invention 1001 to 1078 or the pharmaceutical composition of the present invention 1070 for use in the treatment of cancer diseases selected from the group consisting of melanoma, ovarian cancer, lung cancer, colon cancer, and head and neck cancer. [The present invention 1088] A method for treating a disease, comprising the step of administering to a subject in need thereof an antibody of any one of the present invention 1001 to 1075 or the pharmaceutical composition of the present invention 1083. [The present invention 1089] Use of an antibody of any one of the present invention 1001 to 1077 for manufacturing a medicament such as a medicament for treating a cancer disease characterized by the presence of a cancer, for example, a solid tumor, or a cancer disease selected from the group consisting of melanoma, ovarian cancer, lung cancer, colon cancer, and head and neck cancer. [The present invention 1090] A combination with one or more additional therapeutic substances, for example, a combination with a chemotherapeutic agent, of any one of the methods or uses of the present invention 1083 to 1089. [The present invention 1091] (a) Culturing a host cell that produces a first antibody comprising an antigen-binding region capable of binding to human PD-L1 as defined in any one of the present invention 1001 to 1013, and purifying the first antibody from the culture; (b) Culturing a host cell that produces a second antibody comprising an antigen-binding region capable of binding to a different PD-L1 epitope or a different antigen, for example, a human CD3ε-binding region as defined in any one of the present invention 1014 to 1019, and purifying the second antibody from the culture; (c) Incubating the first antibody together with the second antibody under reducing conditions sufficient to allow the cysteine in the hinge region to undergo disulfide bond isomerization; and (d) Obtaining a bispecific antibody A method for producing an antibody of any one of the present invention 1001 to 1077, comprising the above steps. [The present invention 1092] An anti-idiotype antibody that binds to an antigen-binding region capable of binding to human PD-L1 as defined in any of 1001 to 1077 of the present invention. These and other aspects of the invention are described in more detail below.

Brief Description of the Drawings

[0016]

Figure 1A

Figure 1B

Figure 1C

Figure 1D

Figure 1E

Figure 1F

Figure 2A

Figure 2B

Figure 2C

Figure 2D

Figure 2E

Figure 2F

Figure 3

Figure 4

Figure 5A

Figure 5B

Figure 5C

Figure 5D

Figure 5E

Figure 5F

Figure 6A

Figure 6B

Figure 6C

Figure 7

Figure 8A

Figure 8B

Figure 9A

Figure 9B

Figure 10A

Figure 10B

Figure 11

Figure 12

Figure 13

Figure 14

Mode for Carrying Out the Invention

[0017] Detailed Description of the Invention Definition The term "immunoglobulin" refers to a class of structurally related glycoproteins consisting of two pairs of polypeptide chains, one pair of low molecular weight light chains (L) and one pair of heavy chains (H), all four of which are interconnected by disulfide bonds. The structure of immunoglobulins is well characterized. See, for example, Fundamental Immunology Ch. 7 (Paul, W., ed., 2nd ed. Raven Press, N.Y. (1989)). Briefly, each heavy chain typically consists of a heavy chain variable region (abbreviated herein as VH or VH) and a heavy chain constant region (abbreviated herein as CH or CH). The heavy chain constant region typically consists of three domains, CH1, CH2, and CH3. The hinge region is the region between the CH1 and CH2 domains of the heavy chain and is highly flexible. The disulfide bonds in the hinge region are part of the interaction between the two heavy chains in the IgG molecule. Each light chain typically consists of a light chain variable region (abbreviated herein as VL or VL) and a light chain constant region (abbreviated herein as CL or CL). The light chain constant region typically consists of one domain, CL. The VH and VL regions can be further subdivided into hypervariable regions (or regions where the sequence can vary significantly and / or can take the form of loops with defined structures), also called complementarity determining regions (CDRs), which are interspersed with regions called framework regions (FRs) that are more conserved than the hypervariable regions. Each VH and VL typically consists of three CDRs and four FRs, arranged in the following order from the amino terminus to the carboxy terminus: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4 (Chothia and Lesk J. Mol. Biol. 196, see also 901-917(1987)). Unless otherwise specified or contrary to the context, the CDR sequences in this specification are identified according to the IMGT rules using DomainGapAlign version 4.9.2 (2016-09-26) (Lefranc MP., Nucleic Acids Research 1999;27:209-212 and Ehrenmann F., Kaas Q. and Lefranc M.-P. Nucleic Acids Res., 38, D301-307 (2010); Internet http address http: / / www.imgt.org / , see also). Unless otherwise specified or contrary to the context, the description of the amino acid positions of the constant regions in the present invention follows EU numbering (Edelman et al., Proc Natl Acad Sci U S A. 1969 May;63(1):78-85; Kabat et al., Sequences of Proteins of Immunological Interest, Fifth Edition. 1991 NIH Publication No. 91-3242). For example, SEQ ID NO:93 in this specification shows amino acid positions 118-447 of the IgG1m(f) heavy chain constant region according to EU numbering.

[0018] As used herein, the term "amino acid corresponding to position..." refers to the amino acid position number of the human IgG1 heavy chain. The corresponding amino acid positions in other immunoglobulins may be found by aligning with human IgG1. Thus, an amino acid or segment in one sequence "corresponding to" an amino acid or segment in another sequence is an amino acid or segment that is aligned in a row with the other amino acid or segment using a standard sequence alignment program, such as ALIGN, ClustalW, or the like, typically using default settings, and having at least 50%, at least 80%, at least 90%, or at least 95% identity with the human IgG1 heavy chain. The technique of aligning sequences or segments within a sequence and thereby determining the positions within the sequence that correspond to the amino acid positions according to the present invention is considered well known in the art.

[0019] In the context of the present invention, the term "antibody" (Ab) refers to an immunoglobulin molecule, a fragment of an immunoglobulin molecule, or any derivative thereof, which, under representative physiological conditions, has a relatively long half-life, e.g., at least about 30 minutes, at least about 45 minutes, at least about 1 hour, at least about 2 hours, at least about 4 hours, at least about 8 hours, at least about 12 hours, about 24 hours or more, about 48 hours or more, about 3 days, 4 days, 5 days, 6 days, 7 days, or more, or any other relevant, function-defined period (e.g., a time sufficient to induce, promote, enhance, and / or modulate a physiological response associated with the binding of the antibody to an antigen, and / or a time sufficient for the antibody to enhance effector activity), and has the ability to specifically bind to an antigen. The variable regions of the heavy and light chains of an immunoglobulin molecule contain binding domains that interact with an antigen. As used herein, the term "antibody binding region" refers to a region that interacts with an antigen and includes both the VH region and the VL region. When the term antibody is used herein, it includes not only monospecific antibodies but also multispecific antibodies that contain multiple, e.g., two or more, e.g., three or more, different antigen-binding regions. The constant region of an antibody (Ab) can mediate the binding of the immunoglobulin to host tissues or host factors, including various cells of the immune system (e.g., effector cells) and components of the complement system, e.g., C1q, the first component of the classical pathway of complement activation. As noted above, as used herein, the term antibody includes, unless otherwise specified or clearly contrary to the context, antigen-binding fragments, i.e., antibody fragments that retain the ability to specifically bind to an antigen. It has been shown that the antigen-binding function of an antibody can be performed by fragments of a full-length antibody. Examples of antigen-binding fragments included within the term "antibody" include: (i) Fab' or Fab fragments, monovalent fragments consisting of the VL, VH, CL, and CH1 domains, or monovalent antibodies described in WO2007059782 (Genmab); (ii) F(ab') 2Fragment, a bivalent fragment in which two Fab fragments are linked by a disulfide bridge in the hinge region; (iii) an Fd fragment consisting essentially of VH and CH1 domains; (iv) an Fv fragment consisting essentially of the VL and VH domains of a single arm of an antibody, (v) a dAb fragment consisting essentially of a VH domain and also called a domain antibody (Holt et al; Trends Biotechnol. 2003 Nov; 21 (11):484-90), and (vi) a camelid or nanobody (Revets et al; Expert Opin Biol Ther. 2005 Jan; 341 (1):111-24), and (vii) isolated complementarity determining regions (CDRs). Further, the two domains of the Fv fragment, VL and VH, are encoded by separate genes, but the VL and VH regions pair to form a monovalent molecule (known as a single-chain antibody or single-chain Fv (scFv). For example, Bird et al., Science 5 (1):111-24), and (vii) isolated complementarity determining regions (CDRs). Further, the two domains of the Fv fragment, VL and VH, are encoded by separate genes, but the VL and VH regions pair to form a monovalent molecule (known as a single-chain antibody or single-chain Fv (scFv). For example, Bird et al., Science 242 , 423-426(1988) and Huston et al., PNAS USA 85, which may be connected using recombinant methods by a synthetic linker that allows it to be made as a single protein chain that forms (see, e.g., 5879-5883(1988)). Such single-chain antibodies are included within the term antibody unless otherwise specified or clearly indicated by context. Such fragments are generally included within the meaning of antibody, but taken together and each independently are unique features of the invention that exhibit different biological properties and utilities. These antibody fragments and other useful antibody fragments and bispecific forms of such fragments in the context of the present invention are further discussed herein. The term antibody, unless otherwise specified, includes polyclonal antibodies, monoclonal antibodies (mAbs), antibody-like polypeptides, such as chimeric and humanized antibodies, and antibody fragments (antigen-binding fragments) that retain the ability to specifically bind to an antigen provided by any known technique, such as enzymatic cleavage, peptide synthesis, and recombinant techniques. The antibodies produced may have any isotype. As used herein, the term "isotype" refers to an immunoglobulin class encoded by a heavy chain constant region gene (e.g., IgG1, IgG2, IgG3, IgG4, IgD, IgA, IgE, or IgM). When a particular isotype, e.g., IgG1, is referred to herein, the term is not limited to a particular isotype sequence, e.g., a particular IgG1 sequence, but is used to indicate that the sequence of the antibody is more similar to that isotype, e.g., IgG1, than to other isotypes. Thus, for example, an IgG1 antibody of the present invention may be a sequence variant of a native IgG1 antibody that includes changes in the constant region.

[0020] As used herein, the term "monoclonal antibody" refers to a preparation of antibody molecules of a single molecular composition. A monoclonal antibody composition exhibits only one binding specificity and affinity for a particular epitope. Thus, the term "human monoclonal antibody" refers to an antibody having only one binding specificity and having variable and constant regions derived from human germline immunoglobulin sequences. Human monoclonal antibodies can be produced by hybridomas in which B cells obtained from transgenic non-human animals or translocating non-human animals, e.g., transgenic mice having a genome comprising a human heavy chain transgene and a light chain transgene, are fused with immortalized cells.

[0021] In the context of the present invention, the term "bispecific antibody" or "bs" refers to an antibody having two different antigen-binding regions defined by different antibody sequences. In some embodiments, said different antigen-binding regions bind to different epitopes on the same antigen. However, in preferred embodiments, said different antigen-binding regions bind to different target antigens. A bispecific antibody may be any form of bispecific antibody including any of the bispecific antibody formats described hereinafter.

[0022] As used herein, the terms "half molecule", "Fab arm", and "arm" refer to one heavy chain-light chain pair.

[0023] When a bispecific antibody is said to comprise a first half - molecule antibody "derived from" a first antibody and a second half - molecule antibody "derived from" a second antibody, the term "derived from" indicates that the bispecific antibody has been produced by recombining the half - molecules derived from each of the first and second antibodies by any known method. In this context, "recombining" is not intended to be limited to a particular recombination method, and thus includes, for example, recombination by half - molecule exchange, as well as recombination at the nucleic acid level and / or recombination by co - expressing two types of half - molecules in the same cell, and all of the methods for producing bispecific antibodies described hereinafter in this specification.

[0024] The term "monovalent antibody" means, in the context of the present invention, that an antibody molecule can bind to only one type of antigen molecule and thus cannot cross - link antigens or cells.

[0025] The term "full - length", when used in the context of an antibody, indicates that the antibody is not a fragment and contains all of the domains of a particular isotype that are normally found naturally for that particular isotype, for example, in the case of an IgG1 antibody, the VH, CH1, CH2, CH3, hinge, VL, and CL domains.

[0026] As used herein, unless the context dictates otherwise, the term "Fc region" refers to the antibody region consisting of two Fc sequences of the heavy chain of an immunoglobulin, and the Fc sequences include at least the hinge region, CH2 domain, and CH3 domain.

[0027] As used herein, the term "heterodimeric interaction between a first CH3 region and a second CH3 region" refers to the interaction between the first CH3 region and the second CH3 region in a first CH3 / second CH3 heterodimeric protein.

[0028] As used herein, the term "homo-dimer interaction of the first CH3 region and the second CH3 region" refers to the interaction between the first CH3 region in the first CH3 / first CH3 homodimer protein and another first CH3 region, and the interaction between the second CH3 region in the second CH3 / second CH3 homodimer protein and another second CH3 region.

[0029] With respect to the binding of an antibody to a given antigen or epitope, the term "able to bind" or "bind" as used herein typically refers to, when measured using biolayer interferometry (BLI), for example, when measured using biolayer interferometry (BLI) as described in Example 8, or, for example, when measured in a BIAcore 3000 instrument using surface plasmon resonance (SPR) technology, using the antigen as a ligand and the antibody as an analyte, about 10 -6 M or less, for example, about 10 -7 M or less, for example, about 10 -8 M or less, for example, about 10 -9 M or less, about 10 -10 M or less, or about 10 -11 M, or even less K D corresponding to the binding at an affinity. The antibody binds to the given antigen with an affinity that is at least 1 / 10, for example, at least 1 / 100, for example, at least 1 / 1,000, for example, at least 1 / 10,000, for example, at least 1 / 100,000 of the affinity for binding to a non-specific antigen (e.g., BSA, casein) other than the given antigen or an antigen closely related thereto. D The antibody binds to the given antigen with an affinity corresponding to the amount at which the affinity decreases. The amount at which the affinity decreases depends on the K D of the antibody, and as a result, when the K D of the antibody is very low (i.e., the antibody is highly specific), the degree to which the affinity for the antigen is less than the affinity for the non-specific antigen may be at least 1 / 10,000.

[0030] As used herein, "k d」(sec -1 ) The term " ", where k is the dissociation rate constant of a particular antibody-antigen interaction. This value is also called the k off or k dis value.

[0031] As used herein, the term "K D " (M) refers to the dissociation equilibrium constant of a particular antibody-antigen interaction. This is obtained by dividing kd by ka.

[0032] As used herein, the term "k a " (M -1 ×sec -1 ) refers to the association rate constant of a particular antibody-antigen interaction. This value is also called the k on value or on-rate.

[0033] In a preferred embodiment, the antibody of the invention is isolated. As used herein, an "isolated antibody" is intended to mean an antibody that is substantially free of other antibodies having different antigen specificities. In a preferred embodiment, an isolated bispecific antibody that specifically binds to PD-L1 and a second target, e.g., CD3, is substantially free of monospecific antibodies that specifically bind to PD-L1 or the second target, e.g., CD3. In another preferred embodiment, the antibody, or a pharmaceutical composition comprising the antibody, is substantially free of naturally occurring antibodies that are unable to bind to PD-L1. In a further preferred embodiment, the antibody of the invention has an amino acid sequence structure that is altered compared to the structure of a natural anti-PD-L1 antibody, and this structural alteration results in the antibody exhibiting an altered functionality compared to the functionality exhibited by the natural anti-PD-L1 antibody, and the functionality is selected from the group consisting of (i) PD-L1 binding affinity, (ii) the ability to show binding between PD-L1 and PD-1, and (iii) the ability to induce effector functions via Fc.

[0034] As used herein, the term "PD-L1" refers to programmed death ligand 1 protein. PD-L1 has been found in humans and other species, and thus, the term "PD-L1" is not limited to human PD-L1, unless the context dictates otherwise. The human PD-L1 sequence, the cynomolgus PD-L1 sequence, and the mouse PD-L1 sequence are found by Genbank accession numbers NP_054862.1, XP_005581836, and NP_068693, respectively.

[0035] As used herein, the term "PD-L2" refers to human programmed death 1-ligand 2 protein (Genbank accession number NP_079515).

[0036] As used herein, the term "PD-1" refers to human programmed death-1 protein, also known as CD279.

[0037] As used herein, the term "CD3" refers to the human cluster of differentiation 3 protein, which is part of the T cell receptor protein complex and consists of four different chains. CD3 has also been found in other species, and thus the term "CD3" is not limited to human CD3 unless the context dictates otherwise. In mammals, this complex consists of one CD3γ (gamma) chain (human CD3γ chain UniProtKB / Swiss-Prot No P09693, or cynomolgus monkey CD3γ UniProtKB / Swiss-Prot No Q95LI7), one CD3δ (delta) chain (human CD3δ UniProtKB / Swiss-Prot No P04234, or cynomolgus monkey CD3δ UniProtKB / Swiss-Prot No Q95LI8), two CD3ε (epsilon) chains (human CD3ε UniProtKB / Swiss-Prot No P07766 (SEQ ID NO:95); cynomolgus monkey CD3ε UniProtKB / Swiss-Prot No Q95LI5; or rhesus monkey CD3ε UniProtKB / Swiss-Prot No G7NCB9), and one CD3ζ (zeta) chain (human CD3ζ UniProtKB / Swiss-Prot No P20963, cynomolgus monkey CD3ζ UniProtKB / Swiss-Prot No Q09TK0). These chains bind to a molecule known as the T cell receptor (TCR) and generate activation signals in T lymphocytes. The TCR and CD3 molecules together form the TCR complex.

[0038] A "PD-L1 antibody" or "anti-PD-L1 antibody" is an antibody as described above that specifically binds to the antigen PD-L1, particularly human PD-L1.

[0039] A "CD3 antibody" or "anti-CD3 antibody" is an antibody as described above that specifically binds to the antigen CD3, particularly human CD3ε (epsilon).

[0040] The "CD3×PD-L1 antibody", "anti-CD3×PD-L1 antibody", "PD-L1×CD3 antibody", or "anti-PD-L1×CD3 antibody" is a bispecific antibody that contains two different antigen-binding regions. One of the antigen-binding regions specifically binds to the antigen PD-L1, and one of the antigen-binding regions specifically binds to CD3.

[0041] The present invention also provides antibodies comprising functional variants of the VL region, VH region, or one or more CDRs of the antibodies of the examples. Even for functional variants of VL, VH, or CDRs used with respect to an antibody, the antibody still retains at least a substantial proportion (at least about 50%, 60%, 70%, 80%, 90%, 95%, or greater) of the affinity and / or specificity / selectivity of the "reference" or "parent" antibody, and in some cases, such antibodies may be associated with greater affinity, selectivity, and / or specificity than the parent antibody.

[0042] Such functional variants typically retain significant sequence identity with the parent antibody. The percent identity between two sequences is a function of the number of identical positions shared by these sequences, taking into account the number of gaps that need to be introduced for optimal alignment of the two sequences and the length of each gap (i.e., % homology = number of identical positions / total number of positions × 100). The percent identity between two nucleotide sequences or amino acid sequences can be determined, for example, using the algorithm of E. Meyers and W. Miller, Comput. Appl. Biosci 4, 11-17 (1988), incorporated in the ALIGN program (version 2.0), using a PAM120 weight residue table, a 12 gap length penalty, and a 4 gap penalty. Additionally, the percent identity between two amino acid sequences can be determined using the Needleman and Wunsch, J. Mol. Biol. 48, 444-453 (1970) algorithm.

[0043] Exemplary variants include variants that are different from the VH region and / or VL region and / or CDR region of the parental antibody sequence, mainly in terms of conservative substitutions. For example, 10 of the substitutions in the variant, such as 9, 8, 7, 6, 5, 4, 3, 2, or 1, are conservative amino acid residue exchanges.

[0044] In the context of the present invention, conservative substitutions can be defined by substitutions within the amino acid classes reflected in the following table.

[0045] Amino Acid Residue Classes for Conservative Substitutions TIFF2025081535000001.tif53139

[0046] In the context of the present invention, unless otherwise specified, the following notation is used to describe mutations. (i) The substitution of an amino acid at a particular position is written, for example, as K409R, which means that the lysine at position 409 is substituted with arginine. (ii) For a particular variant, a specific three-letter or one-letter notation containing the symbols Xaa and X, which indicate any amino acid residue, is used. Thus, the substitution of lysine with arginine at position 409 is indicated as K409R, and the substitution of lysine with any amino acid residue at position 409 is indicated as K409X. In the case of the deletion of lysine at position 409, this is indicated as K409 * as shown.

[0047] In the context of the present invention, "competition" refers to the tendency of a particular molecule to bind to a particular binding partner being significantly reduced in the presence of another molecule that binds to the binding partner. "Competition" can be referred to as "blocking" or "substitution". That is, a competing molecule can be either a blocking molecule or a substituting molecule. "Substitution" refers to a state in which a second antibody replaces the antigen from an (already formed) antigen-antibody complex and as a result can exchange with this antigen (Abdiche et al., 2017 Plos One 12(1): e0169535). Competition by two or more anti-PD-L1 antibodies in binding to PD-L1 can be measured by any suitable technique. In one embodiment, competition is measured as described in Example 9 herein.

[0048] Similarly, in the context of the present invention, "inhibition of the binding of PD-L1 to PD-1" refers to the binding of PD-L1 to PD-1 being detectably and significantly reduced in the presence of an antibody capable of binding to PD-L1. Typically, inhibition means at least about 10% reduction, such as at least about 15%, such as at least about 20%, such as at least 40% reduction in the binding between PD-L1 and PD-1 caused by the presence of an anti-PD-L1 antibody. Inhibition of the binding of PD-L1 to PD-1 can be measured by any suitable technique. In one embodiment, inhibition is measured as described in Example 10 herein.

[0049] The term "epitope" means a protein determinant that can specifically bind to an antibody. Epitopes usually consist of surface groups of molecules such as amino acids or sugar side chains and usually have specific three-dimensional structural characteristics as well as specific charge characteristics. Conformational epitopes and non-conformational epitopes are distinguished in that in the presence of a denaturing solvent, the binding to the former is lost, but the binding to the latter is not lost. Epitopes may include amino acid residues directly involved in the binding and other amino acid residues not directly involved in the binding, for example, amino acid residues that are effectively blocked or covered by a peptide that specifically binds to the antigen (in other words, this amino acid residue is within the footprint of the peptide that specifically binds to the antigen).

[0050] As used herein, the term "chimeric antibody" refers to an antibody in which the variable regions are derived from a non-human species (e.g., derived from a rodent) and the constant regions are derived from a different species such as a human. Chimeric monoclonal antibodies for therapeutic use have been developed to reduce antibody immunogenicity. The term "variable region" or "variable domain" as used with respect to chimeric antibodies refers to a region that includes the CDRs and framework regions of both the heavy and light chains of an immunoglobulin. Chimeric antibodies can be produced by using standard DNA techniques as described in Sambrook et al., 1989, Molecular Cloning: A laboratory Manual, New York: Cold Spring Harbor Laboratory Press, Ch.15. Chimeric antibodies may be genetically engineered recombinant antibodies or enzymatically engineered recombinant antibodies. Producing chimeric antibodies is within the knowledge of those skilled in the art, and thus, the production of chimeric antibodies according to the present invention can be carried out by methods other than those described herein.

[0051] As used herein, the term "humanized antibody" refers to a genetically engineered non-human antibody that includes human antibody constant domains and non-human variable domains that have been modified to include a high level of sequence homology to human variable domains. This can be achieved by connecting the six non-human antibody complementarity determining regions (CDRs) that together form the antigen binding site to homologous human acceptor framework regions (FRs) (see WO92 / 22653 and EP0629240). In order to fully reconstruct the binding affinity and specificity of the parental antibody, it may be necessary to substitute (reverse mutate) framework residues derived from the parental antibody (i.e., the non-human antibody) with human frameworks. Structural homology modeling may be useful in identifying amino acid residues within the framework regions that are important for the binding properties of the antibody. Thus, a humanized antibody may include non-human CDR sequences, a human framework region that optionally includes one or more amino acid reverse mutations to non-human amino acid sequences, and a fully human constant region. Optionally, additional amino acid modifications, which are not necessarily reverse mutations, may be applied to obtain a humanized antibody with desirable characteristics, such as affinity and biochemical properties.

[0052] As used herein, the term "human antibody" refers to an antibody having variable and constant regions derived from human germline immunoglobulin sequences. A human antibody may include amino acid residues not encoded by the human germline immunoglobulin sequences (e.g., mutations introduced by in vitro random mutagenesis or site-directed mutagenesis, or mutations introduced by somatic mutations in vivo). However, the term "human antibody" as used herein is not intended to include antibodies in which CDR sequences derived from the germline of another mammalian species, such as a mouse, are grafted onto human framework sequences. The human monoclonal antibodies of the present invention can be produced by a variety of techniques including conventional monoclonal antibody methods, e.g., the standard somatic cell hybridization technique of Kohler and Milstein, Nature 256: 495 (1975). In principle, the somatic cell hybridization method is preferred, but other techniques for producing monoclonal antibodies, e.g., viral transformation or immortalization of B lymphocytes or phage display techniques using human antibody gene libraries, can be used. A suitable animal system for preparing hybridomas that secrete human monoclonal antibodies is the mouse system. Hybridoma production in mice is a well-established procedure. Immunization protocols and methods for isolating immunized spleen cells for fusion are known in the art. Fusion partners (e.g., mouse myeloma cells) and fusion procedures are also known. Thus, human monoclonal antibodies can be produced, for example, using transgenic or transchromosomal mice or rats that have a part of the human immune system instead of a mouse or rat system. Thus, in one aspect, a human antibody is obtained from a transgenic animal, e.g., a mouse or rat, having human germline immunoglobulin sequences instead of animal immunoglobulin sequences. In such an aspect, the antibody is of origin in the human germline immunoglobulin sequences introduced into the animal, but the final antibody sequence is the result of further modification of the human germline immunoglobulin sequences by somatic hypermutation and affinity maturation by the endogenous animal antibody mechanism.See, for example, Mendez et al. 1997 Nat Genet.15(2):146-56. The term "reducing conditions" or "reducing environment" refers to conditions or environments where the substrate, here the cysteine residues in the hinge region of the antibody, are more likely to be reduced than oxidized.

[0053] As used herein, the term "recombinant host cell" (or simply "host cell") is intended to refer to a cell into which an expression vector, such as an expression vector encoding an antibody of the invention, has been introduced. Recombinant host cells include, for example, transfectomas, such as CHO, CHO-S, HEK, HEK293, HEK-293F, Expi293F, PER.C6, or NS0 cells, and lymphocyte cells.

[0054] The term "treating" refers to administering a therapeutically effective amount of an antibody of the invention for the purpose of alleviating, ameliorating, suppressing, or eradicating (curing) a symptom or disease state.

[0055] The term "effective amount" or "therapeutically effective amount" refers to an amount that exhibits an effective effect for obtaining a desired therapeutic result with the dosage and time necessary to obtain the desired therapeutic result. The therapeutically effective amount of an antibody may vary depending on factors such as the disease state, age, sex, and weight of the individual, as well as the ability of the antibody to induce a desired response in the individual. The therapeutically effective amount is also an amount where the beneficial effects of treating the antibody or antibody portion outweigh the toxic or detrimental effects.

[0056] The term "anti-idiotype antibody" refers to an antibody that recognizes unique determinants generally associated with the antigen-binding site of an antibody.

[0057] Further aspects and embodiments of the present invention As described above, in a first aspect, the present invention relates to a bispecific antibody comprising an antigen-binding region capable of binding to human PD-L1 and an antigen-binding region capable of binding to human CD3ε (epsilon), wherein the bispecific antibody inhibits the binding between human PD-L1 and human PD-1. Accordingly, such a bispecific antibody comprises two different antigen-binding regions, one antigen-binding region having binding specificity for PD-L1 and one antigen-binding region having binding specificity for CD3.

[0058] In one embodiment, the antigen-binding region capable of binding to human PD-L1 comprises a heavy chain variable region (VH) comprising a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, and a light chain variable region (VL) comprising a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, wherein the VH CDR3 sequence is selected from the group consisting of the sequences shown in SEQ ID NO:4, SEQ ID NO:11, and SEQ ID NO:21.

[0059] In a further embodiment, the antigen-binding region capable of binding to human PD-L1 is (i) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:2, the CDR2 sequence shown in SEQ ID NO:3, and the CDR3 sequence shown in SEQ ID NO:4, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:6, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:7, or (ii) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:9, the CDR2 sequence shown in SEQ ID NO:10, and the CDR3 sequence shown in SEQ ID NO:11, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:16, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:17, or (iii) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:19, the CDR2 sequence shown in SEQ ID NO:20, and the CDR3 sequence shown in SEQ ID NO:21, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:23, a CDR2 having the sequence DDN, and a CDR3 having the sequence shown in SEQ ID NO:24 comprising.

[0060] In another aspect, the antigen-binding region capable of binding to human PD-L1 comprises a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VH sequence selected from the group consisting of the sequences shown in SEQ ID NO:1, SEQ ID NO:8, and SEQ ID NO:18.

[0061] In another aspect, the antigen-binding region capable of binding to human PD-L1 comprises a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VL sequence selected from the group consisting of the sequences shown in SEQ ID NO:5, SEQ ID NO:15, and SEQ ID NO:22.

[0062] In a further aspect, the antigen-binding region capable of binding to human PD-L1 is (i) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:1, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:5, or (ii) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:8, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:15, or (iii) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:18, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:22 comprising.

[0063] Thus, for example, the antigen-binding region capable of binding to the human PD-L1 is a VH sequence having at least 95% amino acid sequence identity with the VH sequence shown in SEQ ID NO:1, and a VL sequence having at least 95% amino acid sequence identity with the VL sequence shown in SEQ ID NO:5, or a VH sequence having at least 97% amino acid sequence identity with the VH sequence shown in SEQ ID NO:1, and a VL sequence having at least 97% amino acid sequence identity with the VL sequence shown in SEQ ID NO:5, or a VH sequence having at least 99% amino acid sequence identity with the VH sequence shown in SEQ ID NO:1, and a VL sequence having at least 99% amino acid sequence identity with the VL sequence shown in SEQ ID NO:5, or a VH sequence having at least 95% amino acid sequence identity with the VH sequence shown in SEQ ID NO:8, and a VL sequence having at least 95% amino acid sequence identity with the VL sequence shown in SEQ ID NO:15, or A VH sequence having at least 97% amino acid sequence identity with the VH sequence shown in SEQ ID NO:8, and a VL sequence having at least 97% amino acid sequence identity with the VL sequence shown in SEQ ID NO:15, or A VH sequence having at least 99% amino acid sequence identity with the VH sequence shown in SEQ ID NO:8, and a VL sequence having at least 99% amino acid sequence identity with the VL sequence shown in SEQ ID NO:15, or A VH sequence having at least 95% amino acid sequence identity with the VH sequence shown in SEQ ID NO:18, and a VL sequence having at least 95% amino acid sequence identity with the VL sequence shown in SEQ ID NO:22, or A VH sequence having at least 97% amino acid sequence identity with the VH sequence shown in SEQ ID NO:18, and a VL sequence having at least 97% amino acid sequence identity with the VL sequence shown in SEQ ID NO:22, or A VH sequence having at least 99% amino acid sequence identity with the VH sequence shown in SEQ ID NO:18, and a VL sequence having at least 99% amino acid sequence identity with the VL sequence shown in SEQ ID NO:22 is included.

[0064] In a further aspect, each of the VH and VL sequences comprises three CDR sequences, CDR1, CDR2, and CDR3 respectively, and four framework sequences, FR1, FR2, FR3, and FR4 respectively, and each combined FR1, FR2, FR3, and FR4 framework sequence of VH has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with each combined FR1, FR2, FR3, and FR4 framework sequence of the VH sequence, and the VH CDR sequences are not mutated, and each combined FR1, FR2, FR3, and FR4 framework sequence of VL has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with each combined FR1, FR2, FR3, and FR4 framework sequence of the VL sequence, and the VL CDR sequences are not mutated. In relation to this aspect, the % identity refers to the percentage of identity obtained when there are no intervening CDR sequences and the framework sequences are combined as one continuous sequence.

[0065] In a preferred aspect of the antibody of the present invention, the antigen-binding region capable of binding to the human PD-L1 is (i) the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5, or (ii) the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15, or (iii) the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 and comprises.

[0066] Different antibodies that can bind to the same antigen, such as PD-L1, may bind to different regions of the antigen. In some cases, even if a certain PD-L1 antibody binds to PD-L1, a different PD-L1 antibody may still bind to PD-L1. However, in other cases, the binding of a certain PD-L1 antibody to PD-L1 may compete with the binding of a different PD-L1 antibody to PD-L1 (either blocking the binding of the different PD-L1 antibody to PD-L1 or replacing the binding of the different PD-L1 antibody to PD-L1). Therefore, from the competition experiment, information can be obtained regarding where on the target antigen the antibody binds and which one can affect the functional effect of the antibody binding.

[0067] In one aspect, the bispecific antibody of the present invention (i) competes with an antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15 for binding to human PD-L1, but does not compete with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 for binding to human PD-L1, or (ii) competes with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 for binding to human PD-L1, but does not compete with an antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15 for binding to human PD-L1.

[0068] In a further aspect of the present specification, the antibody competes with an antibody comprising the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5 for binding to human PD-L1.

[0069] Antibodies that compete for binding to a target antigen may bind to different epitopes on the antigen, and since these epitopes are very close to each other, a first antibody that binds to one epitope blocks the binding of a second antibody to the other epitope. However, in other situations, two different antibodies may bind to the same epitope on the antigen.

[0070] Thus, in one aspect, the antibodies of the present invention (i) can bind to the same human PD-L1 epitope as the antibody comprising the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5, or (ii) can bind to the same human PD-L1 epitope as the antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15, or (iii) can bind to the same human PD-L1 epitope as the antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22.

[0071] In a further aspect, the binding of the bispecific antibody to human PD-L1 is not replaced by an antibody comprising the VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57.

[0072] In a further aspect, the binding of the bispecific antibody to human PD-L1 is blocked by an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22. As used herein, "blocked" means that the antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 competes with the bispecific antibody but does not replace it.

[0073] As described above, the bispecific antibody of the first aspect of the present invention described above comprises an antigen-binding region that can bind to human CD3ε.

[0074] In one embodiment, the antigen-binding region capable of binding to human CD3ε comprises a heavy chain variable (VH) region CDR1 having the sequence shown in SEQ ID NO:26, a heavy chain variable (VH) region CDR2 having the sequence shown in SEQ ID NO:27, and a heavy chain variable (VH) region CDR3 having the sequence shown in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31.

[0075] The six CDR sequences defined above are derived from a mouse antibody designated SP34. A humanized version of this antibody has been prepared and this humanized antibody is designated huCD3 herein and is further disclosed in WO2015001085 (Genmab).

[0076] In a preferred embodiment of the bispecific antibody of the present invention, the bispecific antibody (i) an antigen-binding region capable of binding to human PD-L1, comprising a heavy chain variable region (VH) comprising a CDR1 sequence shown in SEQ ID NO:2, a CDR2 sequence shown in SEQ ID NO:3, and a CDR3 sequence shown in SEQ ID NO:4, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:6, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:7, and (a) a heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, which is an antigen-binding region capable of binding to human CD3ε, or (ii) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:9, the CDR2 sequence shown in SEQ ID NO:10, and the CDR3 sequence shown in SEQ ID NO:11, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:16, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:17, which can bind to human PD-L1, and (a) a heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, which can bind to human CD3ε, or (iii) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:19, the CDR2 sequence shown in SEQ ID NO:20, and the CDR3 sequence shown in SEQ ID NO:21, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:23, a CDR2 having the sequence DDN, and a CDR3 having the sequence shown in SEQ ID NO:24, which can bind to human PD-L1, and (a) a heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, which can bind to human CD3ε comprising.

[0077] In one aspect, the bispecific antibody comprises an antigen-binding region capable of binding to human CD3ε and comprising a heavy-chain variable region (VH), wherein the VH sequence has at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO: 25.

[0078] In another aspect, the bispecific antibody comprises an antigen-binding region capable of binding to human CD3ε and comprising a light-chain variable region (VL), wherein the VL sequence has at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO: 29.

[0079] In a preferred aspect, the antigen-binding region capable of binding to human CD3ε comprises the VH sequence shown in SEQ ID NO: 25 and the VL sequence shown in SEQ ID NO: 29.

[0080] In one aspect, the bispecific antibody according to the present invention may be modified to reduce the affinity of the antibody. This may be advantageous in some situations and may lead to an increase in efficacy. In particular, low affinity for binding to human CD3ε may affect the motility of T cells in circulation and at the tumor site and may thus lead to better binding between T cells and tumor cells. See Molhoj et al., Molecular Immunology 44 (2007).

[0081] Thus, in different aspects of the bispecific antibody of the present invention comprising an antigen-binding region capable of binding to human PD-L1 and an antigen-binding region capable of binding to human CD3ε, (i) The bispecific antibody has a lower affinity for human CD3ε binding compared to an antibody having an antigen-binding region that can comprise the VH sequence shown in SEQ ID NO:25 and the VL sequence shown in SEQ ID NO:29, preferably, the affinity is at least 1 / 2-fold, for example, at least 1 / 5-fold, for example, at least 1 / 10-fold, for example, at least 1 / 25-fold, for example, at least 1 / 50-fold, and (ii) The bispecific antibody can mediate concentration-dependent cytotoxicity of MDA-MB-231 cells, PC-3 cells, and / or HELA cells when using PBMC or purified T cells as effector cells.

[0082] Similarly, in one aspect, the antibody of the present invention comprises an antigen-binding region capable of binding to human CD3ε, (i) The antigen-binding region capable of binding to human CD3ε has a lower affinity for human CD3ε binding compared to an antibody having an antigen-binding region that can comprise the VH sequence shown in SEQ ID NO:25 and the VL sequence shown in SEQ ID NO:29, preferably, the affinity is at least 1 / 2-fold, for example, at least 1 / 5-fold, for example, at least 1 / 10-fold, for example, at least 1 / 25-fold, for example, at least 1 / 50-fold, and (ii) The antigen-binding region capable of binding to human CD3ε can mediate concentration-dependent cytotoxicity of MDA-MB-231 cells, PC-3 cells, and / or HELA cells when using PBMC or purified T cells as effector cells.

[0083] The affinity for human CD3ε may be measured using octet binding affinity measurement as described in Example 7 of WO2017009442.

[0084] The ability of an antibody to mediate cell cytotoxicity by PBMCs or purified T cells may be measured as described, for example, in Example 11 herein, i.e., MDA-MB-231, PC-3 cells, or HELA cells are seeded and cultured in wells. Serial dilutions of tumor cells, PBMCs, and the antibody are added, and after incubation, the tumor cells are stained to determine if they are viable.

[0085] As used herein, huCD3-H1L1 refers to an anti-CD3 antibody having the VH sequence shown in SEQ ID NO:25 and the VL sequence shown in SEQ ID NO:29. IgG1-huCD3-FEAL refers to a variant thereof that includes the substitutions L234F, L235E, D265A, and F405L (see also other parts herein).

[0086] Examples of variants of IgG1-huCD3-FEAL with reduced affinity for human CD3ε are described in WO2017009442 (Genmab). Example 7 (Table 6) of WO2017009442 discloses the affinities of IgG1-huCD3-FEAL and seven of its variants measured using octet binding affinity measurements. TIFF2025081535000002.tif60128

[0087] In one aspect of the bispecific antibodies of the invention, the antigen-binding region capable of binding to human CD3ε is (i) a heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:99, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (ii) A heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:100, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (iii) A heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:101, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (iv) A heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:102, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (v) A heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:103, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (vi) A heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:104, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (vii) A heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:105, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31 comprising.

[0088] In another aspect, an antigen-binding region capable of binding to human CD3ε is (i) The VH sequence shown in SEQ ID NO:39 and the VL sequence shown in SEQ ID NO:29, or (ii) The VH sequence shown in SEQ ID NO:40 and the VL sequence shown in SEQ ID NO:29, or (iii) The VH sequence shown in SEQ ID NO:41 and the VL sequence shown in SEQ ID NO:29, or (iv) The VH sequence shown in SEQ ID NO:42 and the VL sequence shown in SEQ ID NO:29, or (v) The VH sequence shown in SEQ ID NO:43 and the VL sequence shown in SEQ ID NO:29, or (vi) The VH sequence shown in SEQ ID NO:44 and the VL sequence shown in SEQ ID NO:29, or (vii) The VH sequence shown in SEQ ID NO:45 and the VL sequence shown in SEQ ID NO:29 comprising.

[0089] In a more preferred embodiment of the bispecific antibody of the present invention, each of the antigen-binding regions comprises a heavy-chain variable region (VH) and a light-chain variable region (VL), and each of said variable regions comprises three CDR sequences, namely CDR1, CDR2, and CDR3, and four framework sequences, namely FR1, FR2, FR3, and FR4, respectively.

[0090] In a more preferred embodiment of the bispecific antibody of the present invention, said antibody comprises two heavy-chain constant regions (CH) and two light-chain constant regions (CL).

[0091] In a preferred embodiment, the bispecific antibody comprises a first heavy chain and a second heavy chain, each of said first heavy chain and second heavy chain comprising at least a hinge region, CH2, and CH3 regions, wherein in said first heavy chain, at least one amino acid at a position corresponding to a position selected from the group consisting of T366, L368, K370, D399, F405, Y407, and K409 (according to EU numbering) is substituted, and in said second heavy chain, at least one amino acid at a position corresponding to a position selected from the group consisting of T366, L368, K370, D399, F405, Y407, and K409 (according to EU numbering) is substituted, and said first heavy chain and said second heavy chain are not substituted at the same position.

[0092] Most preferably, (i) in said first heavy chain, the amino acid at the position corresponding to F405 (according to EU numbering) is L, and in said second heavy chain, the amino acid at the position corresponding to K409 (according to EU numbering) is R, or (ii) in the first heavy chain, the amino acid at the position corresponding to K409 (according to EU numbering) is R, and in the second heavy chain, the amino acid at the position corresponding to F405 (according to EU numbering) is L.

[0093] In an even more particularly preferred embodiment, the antibody is a CD3×PD-L1 bispecific antibody comprising a first heavy chain and a second heavy chain, wherein the positions corresponding to positions L234 and L235 of the human IgG1 heavy chain according to EU numbering in both the first heavy chain and the second heavy chain are F and E, respectively, and (i) the position corresponding to F405 of the human IgG1 heavy chain according to EU numbering in the first heavy chain is L, and the position corresponding to K409 of the human IgG1 heavy chain according to EU numbering in the second heavy chain is R, or (ii) the position corresponding to K409 of the human IgG1 heavy chain according to EU numbering in the first heavy chain is R, and the position corresponding to F405 of the human IgG1 heavy chain according to EU numbering in the second heavy chain is L.

[0094] In an even more particularly preferred embodiment, the antibody is a CD3×PD-L1 bispecific antibody comprising a first heavy chain and a second heavy chain, wherein the positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain according to EU numbering in both the first heavy chain and the second heavy chain are F, E, and A, respectively, and (i) the position corresponding to F405 of the human IgG1 heavy chain according to EU numbering in the first heavy chain is L, and the position corresponding to K409 of the human IgG1 heavy chain according to EU numbering in the second heavy chain is R, or (ii) the position corresponding to K409 of the human IgG1 heavy chain according to EU numbering in the first heavy chain is R, and the position corresponding to F405 of the human IgG1 heavy chain according to EU numbering in the second heavy chain is L.

[0095] Novel class of PD-L1 antibodies In a further aspect, the present invention provides a novel anti-PD-L1 antibody comprising an antigen-binding region capable of binding to human PD-L1. The antibody of this aspect of the present invention may be monospecific or multispecific, and if multispecific, the multispecific antibody may or may not comprise an antigen-binding region capable of binding to human CD3ε.

[0096] In one aspect, the present invention provides an antibody comprising an antigen-binding region capable of binding to human PD-L1, wherein the antigen-binding region capable of binding to human PD-L1 has the characteristics defined above herein.

[0097] In one aspect, the present invention provides an antibody comprising an antigen-binding region capable of binding to human PD-L1, wherein the binding of the antibody to a mutant PD-L1 in which any one or more of the amino acid residues (R113) at the position corresponding to position 113 in SEQ ID NO: 94, the amino acid residue (Y123) at the position corresponding to position 123, and the amino acid residue (R125) at the position corresponding to position 125 are substituted with alanine is reduced as compared to the binding to wild-type PD-L1 having the amino acid sequence shown in SEQ ID NO: 94, and the reduction in binding is determined when the change factor of the antibody binding is smaller than the average of the change factors of the binding over all alanine mutants - 1.5×SD, where SD is the standard deviation of all the calculated change factors of the antibody with respect to the mutant PDL1, and the change factor of the binding is calculated as shown in Example 13.

[0098] The antibody may bind to an epitope on PD-L1 (SEQ ID NO: 94), and the epitope comprises the amino acid residue (R113) at position 113 of SEQ ID NO: 94, the amino acid residue (Y123) at position 123, and / or the amino acid residue (R125) at position 125.

[0099] In one aspect, the present invention provides an antibody comprising an antigen-binding region capable of binding to human PD-L1, wherein the binding of the antibody to a mutant PD-L1 in which any one or more of the amino acid residues corresponding to position 19 (F19), position 42 (F42), position 45 (E45), position 46 (K46), position 94 (L94), and position 116 (I116) in SEQ ID NO: 94 are substituted with alanine is reduced as compared to wild-type PD-L1 having the amino acid sequence shown in SEQ ID NO: 94, and the reduction in binding is determined when the change factor of the binding of the antibody is less than the average of the change factors of the binding across all alanine mutants - 1.5×SD, where SD is the standard deviation of all the calculated change factors of the antibody with respect to the mutant PDL1, and the change factor of the binding is calculated as shown in Example 13.

[0100] The antibody may bind to an epitope on PD-L1 (SEQ ID NO: 94), and the epitope comprises one or more amino acid residues selected from the group consisting of the amino acid residue at position 45 (E45), the amino acid residue at position 46 (K46), and / or the amino acid residue at position 94 (L94) in SEQ ID NO: 94.

[0101] In one aspect, the present invention provides an antibody comprising an antigen-binding region capable of binding to human PD-L1, wherein the binding of the antibody to a mutant PD-L1 in which one or more of the amino acid residues (E58) corresponding to position 58 in SEQ ID NO: 94 and the amino acid residues (R113) corresponding to position 113 are substituted with alanine is reduced as compared to wild-type PD-L1 having the amino acid sequence shown in SEQ ID NO: 94, and the reduction in binding is determined when the change factor of the binding of the antibody is less than the average of the change factors of the binding across all alanine mutants - 1.5×SD, where SD is the standard deviation of all the calculated change factors of the antibody with respect to mutant PDL1, and the change factor of the binding is calculated as shown in Example 13.

[0102] The antibody may bind to an epitope on PD-L1 (SEQ ID NO: 94), and the epitope comprises the amino acid residue (E58) at position 58 and / or the amino acid residue (R113) at position 113 of SEQ ID NO: 94.

[0103] In a further aspect, the antibody according to the present invention can induce dose-dependent lysis of epithelial cells of adenocarcinoma, such as dose-dependent lysis of MDA-MB-231, via antibody-dependent cell-mediated cytotoxicity (ADCC).

[0104] In a further aspect, the antibody according to the present invention can reduce the number of cells in the culture of the epithelial cells by at least 5%, such as at least 6%, 7%, 8%, 9%, or at least 10% as a result of cell lysis.

[0105] ADCC is 51 It may be measured in vitro by a Cr release assay method, such as the assay method disclosed in Example 14. In particular, ADCC is carried out by incubating the epithelial cells with a composition containing the antibody and effector cells, such as peripheral blood mononuclear cells (PBMC), at 37°C, 5% CO 2Measured in vitro by incubating for 4 hours, the amount of antibody in the composition is in the range of 0.1 to 1 μg / mL, and the ratio of effector cells to epithelial cells is 100:1.

[0106] Lysis of epithelial cells may be measured in vitro in a luciferase reporter assay as an alternative to ADCC, for example, in the luminescence ADCC reporter bioassay disclosed in Example 14.

[0107] ADCC is particularly (i) contacting the culture of the epithelial cells with a composition comprising the antibody and Jurkat human T cells (effector cells) stably expressing FcγRIIIa (CD16) and firefly luciferase at an effector cell:epithelial cell ratio of 1:1, (ii) adjusting the culture of epithelial cells and effector cells to room temperature for 15 minutes, (iii) incubating the culture of epithelial cells and effector cells with a luciferase substrate, and (iv) measuring luciferase production in the cell culture by which it may be measured in vitro, the amount of antibody in the composition is in the range of 0.5 to 250 ng / mL, and the ratio of effector cells to epithelial cells is 1:1.

[0108] ADCC of epithelial cells may be measured in a luciferase reporter assay, for example, in the reporter assay defined in claim 23 or 24, and then the ADCC observed after incubating the culture of epithelial cells with a test composition comprising the antibody is at least 1.5 times the ADCC observed after incubating the culture of epithelial cells with a composition comprising a reference antibody, ADCC is measured as relative light units (RLU), the antibody concentration in the test composition and the antibody concentration in the composition comprising the reference antibody are the same and in the range of 20 to 250 ng / ml, and the reference antibody is (a) An antibody comprising the VH sequence shown in SEQ ID NO:74 and the VL sequence shown in SEQ ID NO:78; and (b) An antibody comprising the VH sequence shown in SEQ ID NO:81 and the VL sequence shown in SEQ ID NO:85 may be selected.

[0109] In one aspect, an antibody is provided that comprises an antigen-binding region capable of binding to human PD-L1, wherein the antigen-binding region capable of binding to human PD-L1 comprises a heavy chain variable region (VH) comprising a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, and a light chain variable region (VL) comprising a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, and the VH CDR3 sequence is selected from the group consisting of the sequences shown in SEQ ID NO:4, SEQ ID NO:11, and SEQ ID NO:21.

[0110] In a further aspect, an antibody is provided that comprises an antigen-binding region capable of binding to human PD-L1, wherein the antigen-binding region capable of binding to human PD-L1 is (i) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:2, the CDR2 sequence shown in SEQ ID NO:3, and the CDR3 sequence shown in SEQ ID NO:4, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:6, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:7, or (ii) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:9, the CDR2 sequence shown in SEQ ID NO:10, and the CDR3 sequence shown in SEQ ID NO:11, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:16, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:17, or (iii)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:19, the CDR2 sequence shown in SEQ ID NO:20, and the CDR3 sequence shown in SEQ ID NO:21, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:23, a CDR2 having the sequence DDN, and a CDR3 having the sequence shown in SEQ ID NO:24 comprising.

[0111] In a further aspect, an antibody is provided that comprises an antigen-binding region capable of binding to human PD-L1, wherein the antigen-binding region capable of binding to human PD-L1 comprises a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VH sequence selected from the group consisting of the sequences shown in SEQ ID NO:1, SEQ ID NO:8, and SEQ ID NO:18.

[0112] In a further aspect, an antibody is provided that comprises an antigen-binding region capable of binding to human PD-L1, wherein the antigen-binding region capable of binding to human PD-L1 comprises a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VL sequence selected from the group consisting of the sequences shown in SEQ ID NO:52, SEQ ID NO:15, and SEQ ID NO:22.

[0113] In a further aspect, an antibody is provided that comprises an antigen-binding region capable of binding to human PD-L1, wherein the antigen-binding region capable of binding to human PD-L1 is (i) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:1, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:5, or (ii) A VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:8, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:15, or (iii) A VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:18, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:22 comprising.

[0114] In a further aspect, an antibody is provided that comprises an antigen-binding region capable of binding to human PD-L1, wherein the antigen-binding region capable of binding to human PD-L1 comprises a VH sequence and a VL sequence, the VH sequence and the VL sequence each comprise three CDR sequences, CDR1, CDR2, and CDR3 respectively, and four framework sequences, FR1, FR2, FR3, and FR4 respectively, and each combined FR1, FR2, FR3, and FR4 framework sequence of VH has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with each combined FR1, FR2, FR3, and FR4 framework sequence of the VH sequence, and the VH CDR sequences are not mutated, and each combined FR1, FR2, FR3, and FR4 framework sequence of VL has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with each combined FR1, FR2, FR3, and FR4 framework sequence of the VL sequence, and the VL CDR sequences are not mutated.

[0115] In a further aspect, there is provided an antibody comprising an antigen-binding region capable of binding to human PD-L1, wherein the antigen-binding region capable of binding to human PD-L1 is (i) the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5, or (ii) the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15, or (iii) the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 comprises.

[0116] In a further aspect, there is provided an antibody comprising an antigen-binding region capable of binding to human PD-L1, wherein the antibody (i) competes with an antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15 for binding to human PD-L1, but does not compete with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 for binding to human PD-L1, or (ii) competes with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 for binding to human PD-L1, but does not compete with an antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15 for binding to human PD-L1.

[0117] Preferably, the antibody competes with an antibody comprising the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5 for binding to human PD-L1.

[0118] In a further aspect, there is provided an antibody comprising an antigen-binding region capable of binding to human PD-L1, wherein the binding of the antibody to human PD-L1 is not replaced by an antibody comprising the VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57.

[0119] In a further aspect, an antibody is provided that comprises an antigen-binding region capable of binding to human PD-L1, and the binding of the antibody to human PD-L1 is not blocked by the binding of an antibody comprising the VH sequence shown in SEQ ID NO:106 and the VL sequence shown in SEQ ID NO:110.

[0120] In a further aspect, an antibody is provided that comprises an antigen-binding region capable of binding to human PD-L1, the binding of the antibody to human PD-L1 is not replaced by an antibody comprising the VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57, and inhibits the binding of human PD-L1 to human PD-1.

[0121] In a further aspect, an antibody is provided that comprises an antigen-binding region capable of binding to human PD-L1, the binding of the antibody to human PD-L1 is not replaced by an antibody comprising the VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57, and competes with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 for binding to human PD-L1.

[0122] In a further aspect, an antibody is provided that comprises an antigen-binding region capable of binding to human PD-L1, the binding of the antibody to human PD-L1 is not replaced by an antibody comprising the VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57, and the binding of the antibody to human PD-L1 is blocked by an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22. As used herein, "blocked" means that the antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 competes with, but does not replace, the bispecific antibody.

[0123] In a further aspect, an antibody is provided that comprises an antigen-binding region capable of binding to human PD-L1, wherein the antibody (i) Can it bind to the same human PD-L1 epitope as the antibody comprising the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5, or (ii) Can it bind to the same human PD-L1 epitope as the antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15, or (iii) Can it bind to the same human PD-L1 epitope as the antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22.

[0124] In a further aspect, the present invention relates to an antibody comprising an antigen-binding region capable of binding to human PD-L1, wherein the antibody (i) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:33, the CDR2 sequence shown in SEQ ID NO:34, and the CDR3 sequence shown in SEQ ID NO:35, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:37, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:38, or (ii) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:47, the CDR2 sequence shown in SEQ ID NO:48, and the CDR3 sequence shown in SEQ ID NO:49, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:51, a CDR2 having the sequence DVI, and a CDR3 having the sequence shown in SEQ ID NO:52, or (iii) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:54, the CDR2 sequence shown in SEQ ID NO:55, and the CDR3 sequence shown in SEQ ID NO:56, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:58, a CDR2 having the sequence RDS, and a CDR3 having the sequence shown in SEQ ID NO:59, or (iv)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:61, the CDR2 sequence shown in SEQ ID NO:62, and the CDR3 sequence shown in SEQ ID NO:63, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:65, a CDR2 having the sequence DDS, and a CDR3 having the sequence shown in SEQ ID NO:66, or (v)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:107, the CDR2 sequence shown in SEQ ID NO:108, and the CDR3 sequence shown in SEQ ID NO:109, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:111, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:113 (vi)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:68, the CDR2 sequence shown in SEQ ID NO:69, and the CDR3 sequence shown in SEQ ID NO:70, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:72, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:73 comprising.

[0125] In a further aspect, the antibody is (i)the VH sequence shown in SEQ ID NO:32 and the VL sequence shown in SEQ ID NO:36, or (ii)the VH sequence shown in SEQ ID NO:46 and the VL sequence shown in SEQ ID NO:50, or (iii)the VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57, or (iv)the VH sequence shown in SEQ ID NO:60 and the VL sequence shown in SEQ ID NO:64, or (v)the VH sequence shown in SEQ ID NO:106 and the VL sequence shown in SEQ ID NO:110, or (v) the VH sequence shown in SEQ ID NO:67 and the VL sequence shown in SEQ ID NO:71 comprises.

[0126] In one aspect, an antibody comprising an antigen-binding region capable of binding to human PD-L1 is monovalent.

[0127] In another aspect, an antibody comprising an antigen-binding region capable of binding to human PD-L1 is a monospecific antibody comprising two or more identical antigen-binding regions.

[0128] In a further aspect, an antibody comprising an antigen-binding region capable of binding to human PD-L1 is a bivalent antibody having two antigen-binding regions capable of binding to human PD-L1, and the two antigen-binding regions have the same variable region sequence.

[0129] In different aspects, an antibody comprising an antigen-binding region capable of binding to human PD-L1 is a bivalent bispecific antibody, and in addition to the (first) antigen-binding region capable of binding to the human PD-L1, it comprises a (second) antigen-binding region capable of binding to a second antigen, and the second antigen is human CD3ε.

[0130] In different aspects, an antibody comprising an antigen-binding region capable of binding to human PD-L1 is a bivalent bispecific antibody, and in addition to the (first) antigen-binding region capable of binding to the human PD-L1, it comprises a (second) antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, and the second antigen is not human CD3ε.

[0131] In a further aspect, the present invention relates to a multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, wherein the second antigen is optionally not human CD3ε, and the antigen-binding region capable of binding to human PD-L1 comprises a heavy-chain variable region (VH) comprising a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, and a light-chain variable region (VL) comprising a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, and the VH CDR3 sequence is selected from the group consisting of the sequences shown in SEQ ID NO:4, SEQ ID NO:11, SEQ ID NO:21, SEQ ID NO:35, SEQ ID NO:49, SEQ ID NO:56, SEQ ID NO:63, and SEQ ID NO:70. As used herein, "different epitope" means that the epitope to which the second antigen-binding region binds is different from the epitope to which the first antigen-binding region binds.

[0132] In one embodiment of the multispecific antibody, the antigen-binding region capable of binding to human PD-L1 is (i) a heavy-chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:2, the CDR2 sequence shown in SEQ ID NO:3, and the CDR3 sequence shown in SEQ ID NO:4, and a light-chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:6, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:7, or (ii) a heavy-chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:9, the CDR2 sequence shown in SEQ ID NO:10, and the CDR3 sequence shown in SEQ ID NO:11, and a light-chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:16, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:17, or (iii) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:19, the CDR2 sequence shown in SEQ ID NO:20, and the CDR3 sequence shown in SEQ ID NO:21, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:23, a CDR2 having the sequence DDN, and a CDR3 having the sequence shown in SEQ ID NO:24 (iv) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:33, the CDR2 sequence shown in SEQ ID NO:34, and the CDR3 sequence shown in SEQ ID NO:35, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:37, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:38, or (v) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:47, the CDR2 sequence shown in SEQ ID NO:48, and the CDR3 sequence shown in SEQ ID NO:49, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:51, a CDR2 having the sequence DVI, and a CDR3 having the sequence shown in SEQ ID NO:52, or (vi) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:54, the CDR2 sequence shown in SEQ ID NO:55, and the CDR3 sequence shown in SEQ ID NO:56, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:58, a CDR2 having the sequence RDS, and a CDR3 having the sequence shown in SEQ ID NO:59, or (vii) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:61, the CDR2 sequence shown in SEQ ID NO:62, and the CDR3 sequence shown in SEQ ID NO:63, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:65, a CDR2 having the sequence DDS, and a CDR3 having the sequence shown in SEQ ID NO:66, or (viii) A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:68, the CDR2 sequence shown in SEQ ID NO:69, and the CDR3 sequence shown in SEQ ID NO:70, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:72, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:73 comprising.

[0133] In another aspect of the multispecific antibody, the antigen-binding region capable of binding to human PD-L1 comprises a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VH sequence selected from the group consisting of the sequences shown in SEQ ID NO:1, SEQ ID NO:8, and SEQ ID NO:18.

[0134] In another aspect of the multispecific antibody, the antigen-binding region capable of binding to human PD-L1 comprises a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VL sequence selected from the group consisting of the sequences shown in SEQ ID NO:5, SEQ ID NO:15, and SEQ ID NO:22.

[0135] In another aspect of the multispecific antibody, the antigen-binding region capable of binding to human PD-L1 is (i) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:1, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:5, or (ii) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:8, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:15, or (iii) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:18, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:22 comprising.

[0136] In another aspect of the multispecific antibody, each of the VH and VL sequences comprises three CDR sequences, namely CDR1, CDR2, and CDR3, respectively, and four framework sequences, namely FR1, FR2, FR3, and FR4, respectively, and each combined FR1, FR2, FR3, and FR4 framework sequence of VH has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with each combined FR1, FR2, FR3, and FR4 framework sequence of the VH sequence, and the VH CDR sequences are not mutated, and each combined FR1, FR2, FR3, and FR4 framework sequence of VL has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with each combined FR1, FR2, FR3, and FR4 framework sequence of the VL sequence, and the VL CDR sequences are not mutated.

[0137] In another aspect of the multispecific antibody, the antigen-binding region capable of binding to human PD-L1 is (i) the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5, or (ii) the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15, or (iii) the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22, or (iv) the VH sequence shown in SEQ ID NO:32 and the VL sequence shown in SEQ ID NO:36, or (v) the VH sequence shown in SEQ ID NO:46 and the VL sequence shown in SEQ ID NO:50, or (vi) the VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57, or (v) the VH sequence shown in SEQ ID NO:106 and the VL sequence shown in SEQ ID NO:110, or (vii) the VH sequence shown in SEQ ID NO:60 and the VL sequence shown in SEQ ID NO:64, or (viii) the VH sequence shown in SEQ ID NO:67 and the VL sequence shown in SEQ ID NO:71 comprises.

[0138] In a further aspect, the invention relates to a multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, wherein the second antigen is optionally not human CD3ε, and the antibody (i) competes for binding to human PD-L1 with an antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15, but does not compete for binding to human PD-L1 with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22, or (ii) With respect to binding to human PD-L1, it competes with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22, but does not compete with an antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15 with respect to binding to human PD-L1.

[0139] In one embodiment of the multispecific antibody, the antibody competes with an antibody comprising the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5 with respect to binding to human PD-L1.

[0140] In a further aspect, the present invention relates to a multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, wherein the binding of the antibody to human PD-L1 is not replaced by an antibody comprising the VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57.

[0141] In one embodiment of the present specification, the antibody inhibits the binding between human PD-L1 and human PD-1.

[0142] In a further embodiment, the antibody competes with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 with respect to binding to human PD-L1.

[0143] In a further embodiment, the binding of the antibody to human PD-L1 is blocked by an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22. As used herein, "blocked" means that an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 competes with, but does not replace, the bispecific antibody.

[0144] In a further aspect, the present invention relates to a multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, wherein the second antigen is optionally not human CD3ε and the first antigen-binding region is (i) capable of binding to the same human PD-L1 epitope as an antibody comprising the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5, or (ii) capable of binding to the same human PD-L1 epitope as an antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15, or (iii) capable of binding to the same human PD-L1 epitope as an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22.

[0145] In a further aspect, the present invention relates to a bivalent bispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, wherein the second antigen is optionally not human CD3ε and the antigen-binding region capable of binding to the human PD-L1 comprises a heavy chain variable region (VH) comprising a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, and a light chain variable region (VL) comprising a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, and the VH CDR3 sequence is selected from the group consisting of the sequences shown in SEQ ID NO:4, SEQ ID NO:11, and SEQ ID NO:21.

[0146] In one embodiment of the bivalent bispecific antibody, the antigen-binding region capable of binding to the human PD-L1 is (i)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:2, the CDR2 sequence shown in SEQ ID NO:3, and the CDR3 sequence shown in SEQ ID NO:4, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:6, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:7, or (ii)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:9, the CDR2 sequence shown in SEQ ID NO:10, and the CDR3 sequence shown in SEQ ID NO:11, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:16, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:17, or (iii)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:19, the CDR2 sequence shown in SEQ ID NO:20, and the CDR3 sequence shown in SEQ ID NO:21, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:23, a CDR2 having the sequence DDN, and a CDR3 having the sequence shown in SEQ ID NO:24 comprising.

[0147] In another aspect of the bivalent bispecific antibody, the antigen-binding region capable of binding to human PD-L1 comprises a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VH sequence selected from the group consisting of the sequences shown in SEQ ID NO:1, SEQ ID NO:8, and SEQ ID NO:18.

[0148] In another aspect of the bivalent bispecific antibody, the antigen-binding region capable of binding to human PD-L1 comprises a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VL sequence selected from the group consisting of the sequences shown in SEQ ID NO:5, SEQ ID NO:15, and SEQ ID NO:22.

[0149] In another aspect of the bivalent bispecific antibody, the antigen-binding region capable of binding to human PD-L1 is (i) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:1, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:5, or (ii) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:8, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:15, or (iii) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:18, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:22 comprises.

[0150] In another aspect of the bivalent bispecific antibody, each of the VH and VL sequences comprises three CDR sequences, namely CDR1, CDR2, and CDR3, and four framework sequences, namely FR1, FR2, FR3, and FR4, and each combined FR1, FR2, FR3, and FR4 framework sequence of VH has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with the combined FR1, FR2, FR3, and FR4 framework sequences of the VH sequence, and the VH CDR sequences are not mutated, and each combined FR1, FR2, FR3, and FR4 framework sequence of VL has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with the combined FR1, FR2, FR3, and FR4 framework sequences of the VL sequence, and the VL CDR sequences are not mutated.

[0151] In another aspect of the bivalent bispecific antibody, the antigen-binding region capable of binding to human PD-L1 is (i) the VH sequence shown in SEQ ID NO: 1 and the VL sequence shown in SEQ ID NO: 5, or (ii) the VH sequence shown in SEQ ID NO: 8 and the VL sequence shown in SEQ ID NO: 15, or (iii) the VH sequence shown in SEQ ID NO: 18 and the VL sequence shown in SEQ ID NO: 22 and comprises.

[0152] In a further aspect, the present invention relates to a bivalent bispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, wherein the second antigen is optionally not human CD3ε, and the antibody is (i) With respect to binding to human PD-L1, it competes with an antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15, but does not compete with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22, or (ii) With respect to binding to human PD-L1, it competes with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22, but does not compete with an antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15.

[0153] In one aspect of the bivalent bispecific antibody, the antibody competes with an antibody comprising the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5 with respect to binding to human PD-L1.

[0154] In a further aspect, the present invention relates to a bivalent bispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, wherein the binding of the antibody to human PD-L1 is not replaced by an antibody comprising the VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57.

[0155] In one aspect of the present specification, the antibody inhibits the binding of human PD-L1 and human PD-1.

[0156] In a further aspect, the antibody competes with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 with respect to binding to human PD-L1.

[0157] In a further aspect, the binding of said antibody to human PD-L1 is blocked by an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22. As used herein, "blocked" means that an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 competes with, but does not replace, said bispecific antibody.

[0158] In a further aspect, the invention relates to a bivalent bispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, wherein said second antigen is optionally not human CD3ε and wherein said first antigen-binding region (i) is capable of binding to the same human PD-L1 epitope as an antibody comprising the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5, or (ii) is capable of binding to the same human PD-L1 epitope as an antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15, or (iii) is capable of binding to the same human PD-L1 epitope as an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22.

[0159] Further embodiments of the antibodies of the present invention In one aspect, the antibody according to the invention binds to human PD-L1 with a KD of about 10 -8 M or less, for example, about 10 -9 M or less, for example, about 10 -10 M or less, as measured as described in Example 8 herein.

[0160] In a further aspect, the antibody of the invention mediates concentration-dependent cytotoxicity of MDA-MB-231 cells, PC-3 cells, and / or HELA cells when assayed as described in Example 11 herein using purified T cells as effector cells.

[0161] In a preferred aspect, the antibody of the invention does not bind to human PD-L2.

[0162] Antibody format As noted above, various antibody formats have been described in the art. The antibodies of the invention can in principle be antibodies of any isotype. The choice of isotype is typically dictated by the requirements desired for Fc-mediated effector functions, e.g., ADCC induction, or for antibodies lacking Fc-mediated effector functions ( "inert" antibodies). Exemplary isotypes are IgG1, IgG2, IgG3, and IgG4. Either the human kappa or lambda light chain constant region can be used. The effector function of the antibodies of the invention can be altered by isotype switching, e.g., to IgG1, IgG2, IgG3, IgG4, IgD, IgA, IgE, or IgM antibodies, for various therapeutic uses. In one aspect, both heavy chains of the antibody of the invention are heavy chains of the IgG1 isotype, e.g., IgG1,kappa. Optionally, the heavy chains can be modified in the hinge and / or CH3 region as described elsewhere herein.

[0163] Preferably, each of the antigen-binding regions comprises a heavy chain variable region (VH) and a light chain variable region (VL), and each of said variable regions comprises three CDR sequences, CDR1, CDR2, and CDR3, respectively, and four framework sequences, FR1, FR2, FR3, and FR4, respectively. Further preferably, the antibody comprises two heavy chain constant regions (CH) and two light chain constant regions (CL).

[0164] In one embodiment, the antibody is a full-length antibody, for example, a full-length IgG1 antibody. In another embodiment, the antibody is a full-length IgG4 antibody, preferably a full-length IgG4 antibody having a stabilized hinge region. Modifications that stabilize the IgG4 hinge region, such as the S228P mutation in the core hinge, have been described in the art. See, for example, Labrijn et al., 2009 Nat Biotechnol.27(8):767-71.

[0165] In other embodiments, the antibodies of the invention are antibody fragments, such as Fab’ or Fab fragments, monovalent fragments consisting of VL, VH, CL, and CH1 domains, monovalent antibodies described in WO2007059782 (Genmab), F(ab') 2 fragments, Fd fragments, Fv fragments, dAb fragments, camelids or nanobodies, or isolated complementarity determining regions (CDRs).

[0166] The antibodies of the invention are preferably human, humanized, or chimeric. In embodiments where the antibody is a bispecific antibody, both half-molecules may be human, humanized, or chimeric. Or, the half-molecules may differ in the origin of their sequences.

[0167] For example, in one embodiment, the bispecific antibody comprises two half-molecules each comprising an antigen-binding region, (i) the half-molecule comprising an antigen-binding region capable of binding to human PD-L1 is chimeric and / or (ii) the half-molecule comprising an antigen-binding region capable of binding to human CD3ε (epsilon) is chimeric if present.

[0168] For example, in another embodiment, the bispecific antibody comprises two half-molecules each comprising an antigen-binding region, (i) the half-molecule comprising an antigen-binding region capable of binding to human PD-L1 is humanized and / or (ii) The half molecule containing an antigen-binding region capable of binding to human CD3ε (epsilon) is humanized if it exists.

[0169] For example, in a further aspect, the bispecific antibody comprises two half molecules each containing an antigen-binding region, (i) The half molecule containing an antigen-binding region capable of binding to human PD-L1 is human and / or (ii) The half molecule containing an antigen-binding region capable of binding to human CD3ε (epsilon) is human if it exists.

[0170] Thus, for example, in one aspect, the antigen-binding region capable of binding to human PD-L1 is humanized, and the antigen-binding region capable of binding to human CD3ε (epsilon) is humanized if it exists.

[0171] In different aspects, the antigen-binding region capable of binding to human PD-L1 is human, and the antigen-binding region capable of binding to human CD3ε (epsilon) is human if it exists.

[0172] In a further aspect, the antibody is a bispecific antibody comprising an antigen-binding region capable of binding to human PD-L1 and an antigen-binding region capable of binding to human CD3ε (epsilon), and the half molecule containing the antigen-binding region capable of binding to human PD-L1 is human, humanized, or chimeric, and the half molecule containing the antigen-binding region capable of binding to human CD3ε (epsilon) is humanized.

[0173] Preferably, the half molecule containing the antigen-binding region capable of binding to human PD-L1 is human, and the half molecule containing the antigen-binding region capable of binding to human CD3ε (epsilon) is humanized.

[0174] Bispecific antibody format Many different formats and uses of bispecific antibodies are known in the art and have been reviewed by Kontermann; Drug Discov Today, 2015 Jul;20(7):838-47 and MAbs, 2012 Mar-Apr;4(2):182-97.

[0175] The bispecific antibodies according to the present invention are not limited to any particular bispecific format or method for generating bispecific formats.

[0176] Examples of bispecific antibody molecules that can be used in the present invention include: (i) a single antibody having two arms containing different antigen-binding regions; (ii) a single-chain antibody having specificity for two different epitopes via, for example, two scFvs linked in series by an extra peptide linker; (iii) a dual-variable-domain antibody (DVD-Ig) in which each light chain and heavy chain contains two variable domains in series via a short peptide bond (Wu et al., Generation and Characterization of a Dual Variable Domain Immunoglobulin (DVD-Ig™) Molecule, In:Antibody Engineering, Springer Berlin Heidelberg(2010)); (iv) a chemically linked bispecific (Fab')2 fragment; (v) a Tandab, in which two single-chain diabodies are fused to form a tetravalent bispecific antibody having two binding sites for each of the target antigens; (vi) a Flexibody, in which an scFv and a diabody are combined to form a multivalent molecule; (vii) a so-called "dock and lock" molecule based on the "dimerization and docking domain" in protein kinase A. Applying this to Fab results in a trivalent bispecific binding protein in which two identical Fab fragments are linked to a different single Fab fragment; (viii) a so-called scorpion molecule. For example, a scorpion molecule in which two scFvs are fused to both ends of a human Fab arm; and (ix) a diabody.

[0177] In one aspect, the bispecific antibodies of the invention are diabodies, crossbodies, or bispecific antibodies obtained via engineered Fab-arm exchange (e.g., as described in WO2011131746 (Genmab)).

[0178] Examples of different classes of bispecific antibodies include: (i) IgG-like molecules having complementary CH3 domain molecules that enforce heterodimerization; (ii) recombinant IgG-like bispecific molecules, flanked on each side by at least two different Fab fragments or portions of Fab fragments of antibodies; (iii) IgG fusion molecules, where a full-length IgG antibody is fused to an extra Fab fragment or portion of a Fab fragment; (iv) Fc fusion molecules, where a single-chain Fv molecule or stabilized diabody is fused to a heavy chain constant domain, Fc region, or portion thereof; (v) Fab fusion molecules, where different Fab fragments are fused together and further fused to a heavy chain constant domain, Fc region, or portion thereof; and (vi) ScFv-based and diabody-based antibodies and heavy chain antibodies (e.g., domain antibodies, nanobodies), where different single-chain Fv molecules or different diabodies or different heavy chain antibodies (e.g., domain antibodies, nanobodies) are fused to each other or fused to another protein or carrier molecule fused to a heavy chain constant domain, Fc region, or portion thereof, but are not limited thereto.

[0179] Examples of IgG-like molecules with complementary CH3 domain molecules include Triomab / Quadroma molecules (Trion Pharma / Fresenius Biotech; Roche, WO2011069104), so-called Knob-into-Hole molecules (Genentech, WO9850431), CrossMAb (Roche, WO2011117329) and electrostatically-matched molecules (Amgen, EP1870459 and WO2009089004; Chugai, US201000155133; Oncomed, WO2010129304), LUZ-Y molecules (Genentech, Wranik et al. J. Biol. Chem. 2012, 287(52): 43331-9, doi: 10.1074 / jbc.M112.397869. Epub 2012 Nov 1), DIG body and PIG body molecules (Pharmabcine, WO2010134666, WO2014081202), Strand Exchange Engineered Domain body (SEEDbody) molecules (EMD Serono, WO2007110205), Biclonics molecules (Merus, WO2013157953), FcΔAdp molecules (Regeneron, WO201015792), bispecific IgG1 and IgG2 molecules (Pfizer / Rinat, WO11143545), Azymetric scaffold molecules (Zymeworks / Merck, WO2012058768), mAb-Fv molecules (Xencor, WO2011028952), bivalent bispecific antibodies (WO2009080254), and DuoBody® molecules (Genmab, WO2011131746), but are not limited thereto.

[0180] Examples of recombinant IgG-like bispecific molecules include, but are not limited to, Dual Targeting (DT)-Ig molecules (WO2009058383), Two-in-one Antibodies (Genentech; Bostrom, et al 2009. Science 323, 1610-1614.), Cross-linked Mabs (Karmanos Cancer Center), mAb2 (F-Star, WO2008003116), Zybody molecules (Zyngenia; LaFleur et al. MAbs. 2013 Mar-Apr;5(2):208-18), approaches using a common light chain (Crucell / Merus, US7,262,028), κλ bodies (NovImmune, WO2012023053), and CovX bodies (CovX / Pfizer; Doppalapudi, V.R., et al 2007. Bioorg. Med. Chem. Lett. 17,501-506.).

[0181] Examples of IgG fusion molecules include, but are not limited to, Dual Variable Domain (DVD)-Ig molecules (Abbott, US7,612,181), Dual domain double head antibodies (Unilever; Sanofi Aventis, WO20100226923), IgG-like bispecific molecules (ImClone / Eli Lilly, Lewis et al. Nat Biotechnol. 2014 Feb;32(2):191-8), Ts2Ab (MedImmune / AZ; Dimasi et al.J Mol Biol. 2009 Oct 30;393(3):672-92), and BsAb molecules (Zymogenetics, WO2010111625), HERCULES molecules (Biogen Idec, US007951918), scFv fusion molecules (Novartis), scFv fusion molecules (Changzhou Adam Biotech Inc, CN 102250246), and TvAb molecules (Roche, WO2012025525, WO2012025530).

[0182] Examples of Fc fusion molecules include, but are not limited to, ScFv / Fc fusions (Pearce et al., Biochem Mol Biol Int. 1997 Sep;42(6):1179-88), Scorpion molecules (Emergent BioSolutions / Trubion, Blankenship JW, et al. AACR 100th Annual meeting 2009 (Abstract # 5465); Zymogenetics / BMS, WO2010111625), Dual Affinity Retargeting Technology (Fc-DART) molecules (MacroGenics, WO2008157379, WO2010080538), and Dual (ScFv)2-Fab molecules (National Research Center for Antibody Medicine-China).

[0183] Examples of Fab-fusion bispecific antibodies include, but are not limited to, F(ab)2 molecules (Medarex / AMGEN; Deo et al J Immunol. 1998 Feb 15;160(4):1677-86.), Dual-Action or Bis-Fab molecules (Genentech, Bostrom, et al 2009. Science 323, 1610-1614.), Dock-and-Lock (DNL) molecules (ImmunoMedics, WO2003074569, WO2005004809), bivalent bispecific molecules (Biotecnol, Schoonjans, J Immunol. 2000 Dec 15;165(12):7050-7.), and Fab-Fv molecules (UCB-Celltech, WO2009040562A1).

[0184] Examples of ScFv antibodies, diabody-based antibodies, and domain antibodies include bispecific T cell engager (BiTE) molecules (Micromet, WO2005061547), tandem diabody molecules (TandAb) (Affimed) Le Gall et al., Protein Eng Des Sel. 2004 Apr;17(4):357-66.), dual affinity retargeting technology (DART) molecules (MacroGenics, WO2008157379, WO2010080538), single-chain diabody molecules (Lawrence, FEBS Lett. 1998 Apr 3;425(3):479-84), TCR-like antibodies (AIT, ReceptorLogics), human serum albumin ScFv fusion (Human Serum Albumin ScFv Fusion) (Merrimack, WO2010059315), and combody (COMBODY) molecules (Epigen Biotech, Zhu et al. Immunol Cell Biol. 2010 Aug;88(6):667-75.), dual targeting molecules (Ablynx, Hmila et al., FASEB J. 2010), and dual targeting heavy chain only domain antibodies, but are not limited thereto.

[0185] In one aspect, the bispecific antibody of the invention comprises a first Fc sequence comprising a first CH3 region and a second Fc sequence comprising a second CH3 region, wherein the sequences of the first CH3 region and the second CH3 region are different, and the heterodimeric interaction between the first CH3 region and the second CH3 region is stronger than each of the homodimeric interactions of the first CH3 region and the second CH3 region. Further details regarding these interactions and how they can be achieved are shown in WO2011131746 and WO2013060867 (Genmab), which are incorporated herein by reference.

[0186] As further described herein, stable bispecific antibodies, such as bispecific CD3×PD-L1 antibodies, contain a very small number of conservative asymmetric mutations in the CH3 region and can be obtained in high yield using a specific method based on one homodimeric starting PD-L1 antibody and one homodimeric starting antibody (e.g., a homodimeric starting CD3 antibody) that can bind to a different PD-L1 epitope or a different antigen. An asymmetric mutation means that the sequences of the first CH3 region and the second CH3 region contain amino acid substitutions at non-identical positions.

[0187] In one embodiment of the bispecific antibody as defined in any of the embodiments disclosed herein, the first CH3 region has an amino acid substitution at a position selected from the group consisting of positions 366, 368, 370, 399, 405, 407, and 409, the second CH3 region has an amino acid substitution at a position selected from the group consisting of positions 366, 368, 370, 399, 405, 407, and 409, and the first CH3 region and the second CH3 region are not substituted at the same position.

[0188] In one embodiment of the bispecific antibody as defined in any of the embodiments disclosed herein, the first CH3 region has an amino acid substitution at position 366, and the second CH3 region has an amino acid substitution at a position selected from the group consisting of positions 368, 370, 399, 405, 407, and 409. In one embodiment, the amino acid at position 366 is selected from Ala, Asp, Glu, His, Asn, Val, or Gln.

[0189] In one embodiment of the bispecific antibody as defined in any of the embodiments disclosed herein, the first CH3 region has an amino acid substitution at position 368, and the second CH3 region has an amino acid substitution at a position selected from the group consisting of positions 366, 370, 399, 405, 407, and 409.

[0190] In one embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region has an amino acid substitution at position 370, and the second CH3 region has an amino acid substitution at a position selected from the group consisting of positions 366, 368, 399, 405, 407, and 409.

[0191] In one embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region has an amino acid substitution at position 399, and the second CH3 region has an amino acid substitution at a position selected from the group consisting of positions 366, 368, 370, 405, 407, and 409.

[0192] In one embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region has an amino acid substitution at position 405, and the second CH3 region has an amino acid substitution at a position selected from the group consisting of positions 366, 368, 370, 399, 407, and 409.

[0193] In one embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region has an amino acid substitution at position 407, and the second CH3 region has an amino acid substitution at a position selected from the group consisting of positions 366, 368, 370, 399, 405, and 409.

[0194] In one embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region has an amino acid substitution at position 409, and the second CH3 region has an amino acid substitution at a position selected from the group consisting of positions 366, 368, 370, 399, 405, and 407.

[0195] Thus, in one embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the sequences of the first CH3 region and the second CH3 region have asymmetric mutations, i.e., mutations at different positions in the two CH3 regions. For example, one of the CH3 regions contains a mutation at position 405 and the other CH3 region contains a mutation at position 409.

[0196] In one embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region has an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys at position 409, and the second CH3 region has an amino acid substitution at a position selected from the group consisting of positions 366, 368, 370, 399, 405, and 407. In one such embodiment, the first CH3 region has an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys at position 409, and the second CH3 region has an amino acid other than Phe, such as Gly, Ala, Val, Ile, Ser, Thr, Lys, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, Cys, Lys, or Leu at position 405. In a further embodiment thereof, the first CH3 region has an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys at position 409, and the second CH3 region has an amino acid other than Phe, Arg, or Gly, such as Leu, Ala, Val, Ile, Ser, Thr, Met, Lys, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys at position 405.

[0197] In another embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region contains Phe at position 405 and, at position 409, an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys, and the second CH3 region contains, at position 405, an amino acid other than Phe, such as Gly, Ala, Val, Ile, Ser, Thr, Lys, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, Leu, Met, or Cys, and contains Lys at position 409. In a further embodiment thereof, the first CH3 region contains Phe at position 405 and, at position 409, an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys, and the second CH3 region contains, at position 405, an amino acid other than Phe, Arg, or Gly, such as Leu, Ala, Val, Ile, Ser, Thr, Met, Lys, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys, and contains Lys at position 409.

[0198] In another embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region contains Phe at position 405 and, at position 409, an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys, and the second CH3 region contains Leu at position 405 and Lys at position 409. In a further embodiment thereof, the first CH3 region contains Phe at position 405 and Arg at position 409, and the second CH3 region contains, at position 405, an amino acid other than Phe, Arg, or Gly, such as Leu, Ala, Val, Ile, Ser, Thr, Lys, Met, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys, and Lys at position 409. In another embodiment, the first CH3 region contains Phe at position 405 and Arg at position 409, and the second CH3 region contains Leu at position 405 and Lys at position 409.

[0199] In a further embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region contains, at position 409, an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys, the second CH3 region contains Lys at position 409, Thr at position 370, and Leu at position 405. In a further embodiment, the first CH3 region contains Arg at position 409, the second CH3 region contains Lys at position 409, Thr at position 370, and Leu at position 405.

[0200] In yet a further embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region contains Lys at position 370, Phe at position 405, and Arg at position 409, and the second CH3 region contains Lys at position 409, Thr at position 370, and Leu at position 405.

[0201] In another embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region contains an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys at position 409, the second CH3 region contains Lys at position 409, and (a) contains Ile at position 350 and Leu at position 405, or (b) contains Thr at position 370 and Leu at position 405.

[0202] In another embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region contains Arg at position 409, the second CH3 region contains Lys at position 409, and (a) contains Ile at position 350 and Leu at position 405, or (b) contains Thr at position 370 and Leu at position 405.

[0203] In another embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region contains Thr at position 350, Lys at position 370, Phe at position 405, and Arg at position 409, the second CH3 region contains Lys at position 409, and (a) contains Ile at position 350 and Leu at position 405, or (b) contains Thr at position 370 and Leu at position 405.

[0204] In another embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region contains Thr at position 350, Lys at position 370, Phe at position 405, and Arg at position 409, the second CH3 region contains Ile at position 350, Thr at position 370, Leu at position 405, and Lys at position 409.

[0205] In one embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region has an amino acid other than Lys, Leu, or Met at position 409, the second CH3 region has an amino acid other than Phe at position 405, for example, has an amino acid other than Phe, Arg, or Gly at position 405, or the first CH3 region has an amino acid other than Lys, Leu, or Met at position 409, and the second CH3 region has an amino acid other than Tyr, Asp, Glu, Phe, Lys, Gln, Arg, Ser, or Thr at position 407.

[0206] In one embodiment, the bispecific antibody defined in any of the embodiments disclosed herein comprises a first CH3 region having an amino acid other than Lys, Leu, or Met at position 409, and a second CH3 region having an amino acid other than Tyr, Asp, Glu, Phe, Lys, Gln, Arg, Ser, or Thr at position 407.

[0207] In one embodiment, the bispecific antibody defined in any of the embodiments disclosed herein comprises a first CH3 region having Tyr at position 407 and an amino acid other than Lys, Leu, or Met at position 409, and a second CH3 region having an amino acid other than Tyr, Asp, Glu, Phe, Lys, Gln, Arg, Ser, or Thr at position 407 and having Lys at position 409.

[0208] In one embodiment, the bispecific antibody defined in any of the embodiments disclosed herein comprises a first CH3 region having Tyr at position 407 and Arg at position 409, and a second CH3 region having an amino acid other than Tyr, Asp, Glu, Phe, Lys, Gln, Arg, Ser, or Thr at position 407 and having Lys at position 409.

[0209] In another aspect, the first CH3 region has, at position 409, an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys, and the second CH3 region has, at position 407, an amino acid other than Tyr, Asp, Glu, Phe, Lys, Gln, Arg, Ser, or Thr, such as Leu, Met, Gly, Ala, Val, Ile, His, Asn, Pro, Trp, or Cys. In another aspect, the first CH3 region has, at position 409, an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys, and the second CH3 region has, at position 407, Ala, Gly, His, Ile, Leu, Met, Asn, Val, or Trp.

[0210] In another aspect of the bispecific antibody as defined in any aspect disclosed herein, the first CH3 region has, at position 409, an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys, and the second CH3 region has, at position 407, Gly, Leu, Met, Asn, or Trp.

[0211] In another embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region has Tyr at position 407 and, at position 409, an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys, and the second CH3 region has, at position 407, an amino acid other than Tyr, Asp, Glu, Phe, Lys, Gln, Arg, Ser, or Thr, such as Leu, Met, Gly, Ala, Val, Ile, His, Asn, Pro, Trp, or Cys, and has Lys at position 409.

[0212] In another embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region has Tyr at position 407 and, at position 409, an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys, and the second CH3 region has Ala, Gly, His, Ile, Leu, Met, Asn, Val, or Trp at position 407 and has Lys at position 409.

[0213] In another embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region has Tyr at position 407 and, at position 409, an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys, and the second CH3 region has Gly, Leu, Met, Asn, or Trp at position 407 and has Lys at position 409.

[0214] In another embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region has Tyr at position 407 and Arg at position 409, and the second CH3 region has an amino acid other than Tyr, Asp, Glu, Phe, Lys, Gln, Arg, Ser, or Thr, such as Leu, Met, Gly, Ala, Val, Ile, His, Asn, Pro, Trp, or Cys at position 407 and Lys at position 409.

[0215] In another embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region has Tyr at position 407 and Arg at position 409, and the second CH3 region has Ala, Gly, His, Ile, Leu, Met, Asn, Val, or Trp at position 407 and Lys at position 409.

[0216] In another embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region has Tyr at position 407 and Arg at position 409, and the second CH3 region has Gly, Leu, Met, Asn, or Trp at position 407 and Lys at position 409.

[0217] In another embodiment of the bispecific antibody defined in any of the embodiments disclosed herein, the first CH3 region has an amino acid other than Lys, Leu, or Met, such as Gly, Ala, Val, Ile, Ser, Thr, Phe, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys at position 409, and the second CH3 region (i) has an amino acid other than Phe, Leu, and Met, such as Gly, Ala, Val, Ile, Ser, Thr, Lys, Arg, His, Asp, Asn, Glu, Gln, Pro, Trp, Tyr, or Cys at position 368, or (ii) has Trp at position 370, or (iii) having an amino acid other than Asp, Cys, Pro, Glu, or Gln, such as Phe, Leu, Met, Gly, Ala, Val, Ile, Ser, Thr, Lys, Arg, His, Asn, Trp, Tyr, or Cys, at position 399, or (iv) having an amino acid other than Lys, Arg, Ser, Thr, or Trp, such as Phe, Leu, Met, Ala, Val, Gly, Ile, Asn, His, Asp, Glu, Gln, Pro, Tyr, or Cys, at position 366.

[0218] In one embodiment, the first CH3 region has Arg, Ala, His, or Gly at position 409, and the second CH3 region (i) has Lys, Gln, Ala, Asp, Glu, Gly, His, Ile, Asn, Arg, Ser, Thr, Val, or Trp at position 368, or (ii) has Trp at position 370, or (iii) has Ala, Gly, Ile, Leu, Met, Asn, Ser, Thr, Trp, Phe, His, Lys, Arg, or Tyr at position 399, or (iv) has Ala, Asp, Glu, His, Asn, Val, Gln, Phe, Gly, Ile, Leu, Met, or Tyr at position 366.

[0219] In one embodiment, the first CH3 region has Arg at position 409, and the second CH3 region (i) has Asp, Glu, Gly, Asn, Arg, Ser, Thr, Val, or Trp at position 368, or (ii) has Trp at position 370, or (iii) has Phe, His, Lys, Arg, or Tyr at position 399, or (iv) has Ala, Asp, Glu, His, Asn, Val, Gln at position 366.

[0220] In a preferred embodiment, the bispecific antibody comprises a first heavy chain and a second heavy chain, each of the first heavy chain and the second heavy chain comprising at least a hinge region, CH2, and CH3 regions, wherein (i) in the first heavy chain, the amino acid at the position corresponding to F405 (according to EU numbering) is L, and in the second heavy chain, the amino acid at the position corresponding to K409 (according to EU numbering) is R, or (ii) in the first heavy chain, the amino acid at the position corresponding to K409 (according to EU numbering) is R, and in the second heavy chain, the amino acid at the position corresponding to F405 (according to EU numbering) is L.

[0221] In addition to the amino acid substitutions described above, the first heavy chain and the second heavy chain may further contain amino acid substitutions, deletions, or insertions compared to the wild-type heavy chain sequence.

[0222] In a further embodiment, the first Fab arm and the second Fab arm (or heavy chain constant domain) independently comprise a CH3 sequence selected from (IgG1m(a)) (SEQ ID NO:96), (IgG1m(f)) (SEQ ID NO:97), and (IgG1m(ax) (SEQ ID NO:98) other than the specified mutations.

[0223] In one embodiment, neither the first Fc sequence nor the second Fc sequence contains a Cys-Pro-Ser-Cys sequence in the (core) hinge region.

[0224] In a further embodiment, both the first Fc sequence and the second Fc sequence contain a Cys-Pro-Pro-Cys sequence in the (core) hinge region.

[0225] Method for preparing bispecific antibodies In the preparation of the bispecific antibody of the present invention, conventional methods such as hybrid hybridomas and chemical conjugation methods (Marvin and Zhu (2005) Acta Pharmacol Sin 26:649) can be used. When two types of antibodies consisting of different heavy and light chains are co-expressed in a host cell, in addition to the desired bispecific antibody, a mixture of possible antibody products may be generated. Subsequently, the desired bispecific antibody can be isolated by, for example, affinity chromatography or a similar method.

[0226] Strategies that are advantageously effective for the formation of functional bispecific products when different antibody constructs are co-expressed can also be used, for example, the method described in Lindhofer et al. (1995 J Immunol 155:219). Fusion of rat hybridomas and mouse hybridomas that produce different antibodies results in a limited number of heterodimeric proteins due to species-restricted preferential heavy chain / light chain pair formation. Another strategy that promotes heterodimer formation over homodimer formation is the "knob-into-hole" strategy. In this strategy, for the purpose of promoting heterodimer formation and preventing homodimer formation, a protrusion is introduced onto the first heavy chain polypeptide and a corresponding cavity is introduced into the second heavy chain polypeptide such that the protrusion can be placed within the cavity at the interface between these two heavy chains. The "protrusion" is constructed by replacing small amino acid side chains derived from the interface of the first polypeptide with larger side chains. A complementary "cavity" of the same or similar size as the protrusion is created at the interface of the second polypeptide by replacing large amino acid side chains with small amino acid side chains (U.S. Patent No. 5,731,168). EP1870459 (Chugai) and WO2009089004 (Amgen) describe other strategies that are advantageously effective for heterodimer formation when different antibody domains are co-expressed in a host cell. In these methods, one or more residues that constitute the CH3-CH3 interface in both CH3 domains are replaced with charged amino acids such that homodimer formation is electrostatically unfavorable and heterodimerization is electrostatically favorable. WO2007110205 (Merck) describes yet another strategy in which differences between IgA and IgG CH3 domains are exploited to promote heterodimerization.

[0227] Another in vitro method for producing bispecific antibodies is described in WO2008119353 (Genmab), where the bispecific antibody is formed by "Fab arm" exchange or "half-molecule" exchange (exchange of the heavy chain and the attached light chain) between two monospecific IgG4 antibodies or IgG4-like antibodies when incubated under reducing conditions. The resulting product is a bispecific antibody having two Fab arms that may contain different sequences.

[0228] A preferred method for preparing the bispecific antibodies of the invention, for example, bispecific CD3×PD-L1 antibodies, is (a) providing a first antibody comprising an Fc region, wherein the Fc region comprises a first CH3 region; (b) providing a second antibody comprising a second Fc region, wherein the Fc region comprises a second CH3 region, the first antibody is a PD-L1 antibody according to the invention, the second antibody is an antibody capable of binding to a different PD-L1 epitope or a different antigen, for example, human CD3, or vice versa, the sequences of the first CH3 region and the second CH3 region are different, and the heterodimeric interaction between the first CH3 region and the second CH3 region is stronger than each of the homodimeric interactions of the first CH3 region and the second CH3 region; (c) incubating the first antibody together with the second antibody under reducing conditions; and (d) obtaining the bispecific antibody, for example, bispecific PD-L1×CD3 antibody comprising the method described in WO2011131746 and WO2013060867 (Genmab).

[0229] Similarly, (a) culturing a host cell that produces a first antibody comprising an antigen-binding region capable of binding to human PD-L1 as defined herein, and purifying the first antibody from the culture; (b) Culturing a host cell that produces a second antibody that can bind to different PD-L1 epitopes or different antigens, for example, an antibody comprising a human CD3ε binding region as defined herein, and purifying the second antibody from the culture; (c) Incubating the first antibody together with the second antibody under reducing conditions sufficient to allow cysteines in the hinge region to undergo disulfide bond isomerization; and (d) Obtaining the bispecific antibody A method for producing an antibody according to the present invention is provided, which comprises the above steps.

[0230] In one embodiment, the first antibody and the second antibody are incubated under reducing conditions sufficient to allow cysteines in the hinge region to undergo disulfide bond isomerization, and the heterodimeric interaction between the first antibody and the second antibody in the resulting heterodimeric antibody is a heterodimeric interaction such that Fab arm exchange does not occur after 24 hours at 37 °C with 0.5 mM GSH.

[0231] Although not limited to theory, in step (c), the heavy chain disulfide bonds in the hinge region of the parental antibody are reduced, and the resulting cysteines can then form inter-heavy chain disulfide bonds with cysteine residues of another parental antibody molecule (which has different specificities from the start). In one embodiment of this method, the reducing conditions in step (c) include the addition of a reducing agent selected from the group consisting of a reducing agent such as 2-mercaptoethylamine (2-MEA), dithiothreitol (DTT), dithioerythritol (DTE), glutathione, tris(2-carboxyethyl)phosphine (TCEP), L-cysteine, and β-mercapto-ethanol, preferably a reducing agent selected from the group consisting of 2-mercaptoethylamine, dithiothreitol, and tris(2-carboxyethyl)phosphine. In a further embodiment, step (c) includes restoring the conditions to non-reducing or less reducing conditions, for example, by removing the reducing agent, for example, by desalting.

[0232] For this method, any of the antibodies of said antibody can be used, for example, said CD3 antibody and PD-L1 antibody, including the first and second CD3 antibodies and PD-L1 antibodies each containing a first Fc region and / or a second Fc region. Examples of such first Fc regions and second Fc regions may include any of the foregoing, including combinations of such first Fc regions and second Fc regions. In certain embodiments, the first antibody and the second antibody, for example, the CD3 antibody and the PD-L1 antibody respectively, may be selected to obtain a bispecific antibody as described herein.

[0233] In one embodiment of this method, the first antibody and / or the second antibody is a full-length antibody.

[0234] The Fc regions of the first antibody and the second antibody may be of any isotype including, but not limited to, IgG1, IgG2, IgG3, or IgG4. In one embodiment of this method, the Fc regions of both the first antibody and the second antibody are Fc regions of the IgG1 isotype. In another embodiment, one of the Fc regions of the antibody is an Fc region of the IgG1 isotype and the other is an Fc region of the IgG4 isotype. In the latter embodiment, the resulting bispecific antibody contains the Fc sequence of IgG1 and the Fc sequence of IgG4 and may thus have interesting intermediate properties with respect to activation of effector functions.

[0235] In a further embodiment, one of the antibody starting proteins is engineered to not bind to protein A, and thus it is possible to separate the heterodimeric protein from the homodimeric starting protein by passing the product over a protein A column.

[0236] As described above, the sequences of the first CH3 region and the second CH3 region of the homodimeric starting antibody are different, and the heterodimeric interaction between the first CH3 region and the second CH3 region is stronger than each of the homodimeric interactions of the first CH3 region and the second CH3 region. Further details regarding these interactions and how to achieve them are shown in WO2011131746 and WO2013060867 (Genmab). WO2011131746 and WO2013060867 (Genmab) are hereby incorporated by reference in their entirety.

[0237] In particular, stable bispecific antibodies, such as bispecific CD3×PD-L1 antibodies, can be obtained in high yield using the above-described method of the present invention based on two types of homodimer starting antibodies, a homodimer starting antibody that binds to PD-L1 and a homodimer starting antibody that binds to a different antigen or a different PD-L1 epitope, such as CD3, respectively, and contain a very small number of conservative asymmetric mutations in the CH3 region. The asymmetric mutation means that the sequences of the first CH3 region and the second CH3 region contain amino acid substitutions at non-identical positions.

[0238] The bispecific antibody of the present invention may also be obtained by co-expressing constructs encoding a first polypeptide and a second polypeptide in a single cell. Thus, in a further aspect, the present invention provides: (a) providing a first nucleic acid construct encoding a first polypeptide comprising a first Fc sequence and a first antigen-binding region of a first antibody heavy chain, wherein the first Fc sequence comprises a first CH3 region; (b) providing a second nucleic acid construct encoding a second polypeptide comprising a second Fc sequence and a second antigen-binding region of a second antibody heavy chain, wherein the second Fc sequence comprises a second CH3 region, wherein the sequences of the first CH3 region and the second CH3 region are different, and the heterodimer interaction between the first CH3 region and the second CH3 region is stronger than each of the homodimer interactions of the first CH3 region and the second CH3 region, the first homodimer protein has an amino acid other than Lys, Leu, or Met at position 409, and the second homodimer protein has an amino acid substitution at a position selected from the group consisting of positions 366, 368, 370, 399, 405, and 407; optionally, the first nucleic acid construct and the second nucleic acid construct encode the light chain sequences of the first antibody and the second antibody; (c) co-expressing the first nucleic acid construct and the second nucleic acid construct in a host cell; and (d) obtaining the heterodimer protein from the cell culture Relates to a method for producing a bispecific antibody, which comprises

[0239] Materials and methods for producing the antibodies of the present invention In a further aspect, the present invention relates to materials and methods for the recombinant production of the antibodies according to the present invention. Suitable expression vectors, including promoters, enhancers, etc., and suitable host cells for producing the antibodies are well known in the art.

[0240] Thus, in one aspect, (i) a nucleic acid sequence encoding a heavy chain sequence of an antibody comprising an antigen-binding region capable of binding to human PD-L1 as defined herein, and / or (ii) a nucleic acid sequence encoding a light chain sequence of an antibody comprising an antigen-binding region capable of binding to human PD-L1 as defined herein is provided with a nucleic acid construct.

[0241] In one embodiment, the nucleic acid construct (i) a nucleic acid sequence encoding a heavy chain sequence of an antibody comprising an antigen-binding region capable of binding to a different PD-L1 epitope or a different antigen, such as human CD3ε, as defined herein, and (ii) a nucleic acid sequence encoding a light chain sequence of an antibody comprising an antigen-binding region capable of binding to a different PD-L1 epitope or a different antigen, such as human CD3ε, as defined herein further comprises.

[0242] In still a further aspect, the present invention relates to an expression vector comprising the nucleic acid construct as defined above herein.

[0243] In the context of the present invention, an expression vector may be any suitable vector including chromosomal nucleic acid vectors, non-chromosomal nucleic acid vectors, and synthetic nucleic acid vectors (nucleic acid sequences containing an appropriate set of expression control elements). Examples of such vectors include derivatives of SV40, bacterial plasmids, phage DNA, baculoviruses, yeast plasmids, vectors derived from combinations of plasmids and phage DNA, and viral nucleic acid (RNA or DNA) vectors. In one embodiment, an antibody-encoding nucleic acid, e.g., a PD-L1-encoding nucleic acid or a CD3 antibody-encoding nucleic acid, is a naked DNA vector or RNA vector (e.g., as described in Sykes and Johnston, Nat Biotech 17, 355-59 (1997)) including, for example, linear expression elements, a compressed nucleic acid vector (e.g., as described in US6,077,835 and / or WO00 / 70087), a plasmid vector, e.g., pBR322, pUC19 / 18, or pUC118 / 119, a "midge" minimal size nucleic acid vector (e.g., as described in Schakowski et al., Mol Ther 3, 793-800 (2001)), or a precipitated nucleic acid vector construct, e.g., a construct precipitated by Ca3(P04)2 (e.g., as described in WO200046147, Benvenisty and Reshef, PNAS USA 83, 9551-55 (1986), Wigler et al., Cell 14, 725 (1978), and Coraro and Pearson, Somatic Cell Genetics 7, 603 (1981)). Such nucleic acid vectors and their use are well known in the art (see, for example, US5,589,466 and US5,973,972).

[0244] In one embodiment, the vector is suitable for the expression of the antibody, e.g., a PD-L1 antibody and / or a CD3 antibody, in bacterial cells. Examples of such vectors include expression vectors, e.g., BlueScript (Stratagene), pIN vectors (Van Heeke & Schuster, J Biol Chem264 , including 5503-5509 (1989), pET vectors (Novagen, Madison WI), etc.).

[0245] The expression vector may similarly or alternatively be a vector suitable for expression in a yeast system. Any vector suitable for expression in a yeast system can be used. Suitable vectors include, for example, vectors containing constitutive or inducible promoters, such as the α-factor, alcohol oxidase, and vectors containing PGH (F. Ausubel et al., ed. Current Protocols in Molecular Biology, Greene Publishing and Wiley InterScience New York (1987), and Grant et al., Methods in Enzymol 153 , overviewed in 516-544 (1987)).

[0246] The expression vector may similarly or alternatively be a vector suitable for expression in mammalian cells, such as a vector containing glutamine synthetase as a selectable marker, such as the vector described in Bebbington (1992) Biotechnology (NY) 10:169-175.

[0247] The nucleic acid and / or vector may also include a nucleic acid sequence encoding a secretion / localization sequence. The secretion / localization sequence can target a polypeptide, such as a nascent polypeptide chain, to the periplasmic space or into the cell culture medium. Such sequences are known in the art and include secretion leaders or signal peptides.

[0248] The expression vector may contain and may be ligated with any suitable promoter, enhancer, and other expression-promoting elements. Examples of such elements include strong expression promoters (e.g., human CMV IE promoter / enhancer and RSV, SV40, SL3-3, MMTV, and HIV LTR promoters), effective poly(A) termination sequences, origins of replication for plasmid products in E. coli, antibiotic resistance genes as selection markers, and / or convenient cloning sites (e.g., polylinkers). The nucleic acid may also include an inducible promoter relative to a constitutive promoter, e.g., CMV IE.

[0249] In one aspect, the antibody-encoding expression vector, e.g., a PD-L1 antibody-encoding expression vector and / or a CD3 antibody-encoding expression vector, may be placed into and / or delivered to a host cell or host animal via a viral vector.

[0250] In yet a further aspect, the present invention relates to a host cell comprising one or more of the nucleic acid constructs or expression vectors specified above herein.

[0251] Accordingly, the present invention also relates to a recombinant eukaryotic host cell or prokaryotic host cell that produces the antibody of the present invention, e.g., a transfectoma.

[0252] Examples of host cells include yeast, bacterial cells, plant cells, and mammalian cells, e.g., CHO, CHO-S, HEK, HEK293, HEK-293F, Expi293F, PER.C6, or NS0 cells, or lymphocyte cells. A preferred host cell is a CHO-K1 cell.

[0253] For example, in one aspect, the host cell may contain a first nucleic acid construct and a second nucleic acid construct stably integrated into the cell genome. In another aspect, the present invention provides a cell comprising a non-integrating nucleic acid, such as a plasmid, cosmid, phagemid, or linear expression element, containing the first nucleic acid construct and the second nucleic acid construct specified above.

[0254] In a further aspect, the present invention relates to a hybridoma producing a PD-L1 antibody as defined herein.

[0255] Fc region In some aspects, the antibody according to the present invention comprises, in addition to the antigen-binding region, an Fc region consisting of the Fc sequences of two heavy chains.

[0256] The first Fc sequence and the second Fc sequence may each be of any isotype including, but not limited to, IgG1, IgG2, IgG3, and IgG4, and may include one or more mutations or modifications. In one aspect, the first Fc sequence and the second Fc sequence may each be of the IgG4 isotype, or be derived therefrom, and optionally have one or more mutations or modifications. In another aspect, the first Fc sequence and the second Fc sequence may each be of the IgG1 isotype, or be derived therefrom, and optionally have one or more mutations or modifications. In another aspect, one of the Fc sequences is an Fc sequence of the IgG1 isotype and the other is an Fc sequence of the IgG4 isotype, or is derived from each such isotype and optionally has one or more mutations or modifications.

[0257] In one embodiment, one or both of the Fc sequences are deficient in effector function. For example, the Fc sequence may be an Fc sequence of the IgG4 isotype that is mutated such that the effector function, such as the ability to mediate ADCC, is reduced or even eliminated, or it may be an Fc sequence of a non-IgG4 type, such as IgG1, IgG2, or IgG3. Such mutations are described, for example, in Dall'Acqua WF et al., J Immunol. 177 (2):1129-1138 (2006) and Hezareh M, J Virol.; 75 (24):12161-12168 (2001). In another embodiment, one or both of the Fc sequences comprise an IgG1 wild-type sequence.

[0258] The antibodies according to the invention may comprise modifications in the Fc region. When the antibody comprises such a modification, it may become an inert or inactivated antibody. As used herein, the terms "inert", "inactive", or "inactivated" mean, at least, that it cannot bind to any Fcγ receptor, cannot induce FcR cross-linking via Fc, or cannot induce cross-linking of the target antigen via FcR via the two Fc regions of an individual antibody, or an Fc region that cannot bind to C1q. The inactivation of the Fc region of an antibody, such as a humanized or chimeric CD3 antibody, is advantageously tested using the antibody in a monospecific form.

[0259] For the development of therapeutic antibodies, several variants can be constructed to render the Fc region of the antibody inert to the interaction of Fcγ (gamma) receptor and C1q. Examples of such variants are described herein.

[0260] Thus, in one aspect of the antibody of the present invention, the antibody comprises a first heavy chain and a second heavy chain, and one or both of the heavy chains are modified to induce effector functions via Fc to a lesser extent compared to the same antibody except that the antibody comprises unmodified first and second heavy chains. The effector function via Fc may be measured by determining Fc-mediated CD69 expression, or may be measured by binding to Fcγ receptors, or may be measured by binding to C1q, or may be measured by inducing FcR cross-linking via Fc.

[0261] In such one aspect, the heavy chain constant sequence is modified such that the antibody reduces Fc-mediated CD69 expression by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99%, or 100% compared to a wild-type (unmodified) antibody, and the Fc-mediated CD69 expression is determined in a PBMC-based functional assay as described in Example 3 of WO2015001085.

[0262] In another such aspect, the heavy chain constant sequence and the light chain constant sequence are modified such that the binding of C1q to the antibody is reduced by at least 70%, at least 80%, at least 90%, at least 95%, at least 97%, or 100% compared to the unmodified antibody, and the C1q binding is measured by ELISA.

[0263] In another aspect, the antibody comprises an Fc region modified to mediate Fc-mediated T cell proliferation reduced by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99%, or 100% compared to the unmodified antibody, and the T cell proliferation is measured in a PBMC-based functional assay.

[0264] Thus, amino acids within the Fc region that play a central role in the interaction with C1q and Fcγ receptors can be modified.

[0265] For example, examples of amino acid positions that can be modified in an IgG1 isotype antibody include positions L234, L235, and P331. Combinations thereof, such as L234F / L235E / P331S, can significantly reduce binding to human CD64, CD32, CD16, and C1q.

[0266] Thus, in one aspect, the amino acids at at least one position corresponding to L234, L235, and P331 may be A, A, and S, respectively (Xu et al., 2000, Cell Immunol. 200(1):16 - 26; Oganesyan et al., 2008, Acta Cryst.(D64):700 - 4). Also, the L234F and L235E amino acid substitutions can result in an Fc region with suppressed interaction with Fcγ receptors and C1q (Canfield et al., 1991, J. Exp. Med. (173):1483 - 91; Duncan et al., 1988, Nature (332):738 - 40). Thus, in one aspect, the amino acids at the positions corresponding to L234 and L235 may be F and E, respectively. The D265A amino acid substitution can reduce binding to all Fcγ receptors and block ADCC (Shields et al., 2001, J. Biol. Chem. (276):6591 - 604). Thus, in one aspect, the amino acid at the position corresponding to D265 may be A. Binding to C1q can be inhibited by mutating positions D270, K322, P329, and P331. Mutating these positions to any of D270A or K322A or P329A or P331A can render the antibody deficient in CDC activity (Idusogie EE, et al., 2000, J Immunol. 164: 4178 - 84). Thus, in one aspect, the amino acids at at least one position corresponding to D270, K322, P329, and P331 may be A, A, A, and A, respectively.

[0267] An alternative approach to minimizing the interaction of the Fc region with Fcγ receptors and C1q is by removing the glycosylation site of the antibody. Mutating position N297 to, for example, Q, A, or E removes the glycosylation site that has significant meaning for IgG-Fcγ receptor interaction. Thus, in one aspect, the amino acid at the position corresponding to N297 may be G, Q, A, or E. (Leabman et al., 2013, MAbs; 5(6):896-903). Another alternative approach to minimizing the interaction between the Fc region and Fcγ receptors may be obtained by the following mutations: P238A, A327Q, P329A, or E233P / L234V / L235A / G236del (Shields et al., 2001, J. Biol. Chem. (276):6591-604).

[0268] Alternatively, human IgG2 and IgG4 subclasses are thought to have impaired interaction with C1q and Fcγ receptors naturally, but interaction with Fcγ receptors has been reported (Parren et al., 1992, J. Clin Invest. 90: 1537-1546; Bruhns et al., 2009, Blood 113: 3716-3725). In both isotypes, adding mutations that suppress these remaining interactions can reduce unwanted side effects associated with FcR binding. In the case of IgG2, these include L234A and G237A, and in the case of IgG4, L235E. Thus, in one aspect, the amino acids at the positions corresponding to L234 and G237 in the human IgG2 heavy chain may be A and A, respectively. In one aspect, the amino acid at the position corresponding to L235 in the human IgG4 heavy chain may be E.

[0269] Other approaches to further minimize the interaction of IgG2 antibodies with Fcγ receptors and C1q include the approaches described in WO2011066501 and Lightle, S., et al., 2010, Protein Science (19):753-62.

[0270] The hinge region of said antibody may also be important with respect to the interaction with Fcγ receptor and complement (Brekke et al., 2006, J Immunol 177:1129-1138; Dall’Acqua WF, et al., 2006, J Immunol 177:1129-1138). Therefore, mutations within the hinge region or deletions of the hinge region may affect the effector function of the antibody.

[0271] Thus, in one aspect, said antibody comprises a first immunoglobulin heavy chain and a second immunoglobulin heavy chain, and in at least one of the first immunoglobulin heavy chain and the second immunoglobulin heavy chain, one or more amino acids at positions corresponding to positions L234, L235, D265, N297, and P331 in the human IgG1 heavy chain are not L, L, D, N, and P, respectively.

[0272] In one aspect, in both the first heavy chain and the second heavy chain, one or more amino acids at positions corresponding to positions L234, L235, D265, N297, and P331 in the human IgG1 heavy chain are not L, L, D, N, and P, respectively.

[0273] In one aspect, in both the first heavy chain and the second heavy chain, the amino acid at the position corresponding to position D265 in the human IgG1 heavy chain is not D.

[0274] Thus, in one aspect, in both the first heavy chain and the second heavy chain, the amino acid at the position corresponding to position D265 in the human IgG1 heavy chain is selected from the group consisting of A and E.

[0275] In a further aspect, in at least one of the first heavy chain and the second heavy chain, the amino acids at positions corresponding to positions L234 and L235 in the human IgG1 heavy chain are not L and L, respectively.

[0276] In certain embodiments, in at least one of the first heavy chain and the second heavy chain, the amino acids at positions corresponding to positions L234 and L235 in the human IgG1 heavy chain are F and E, respectively.

[0277] In one embodiment, in both the first heavy chain and the second heavy chain, the amino acids at positions corresponding to positions L234 and L235 in the human IgG1 heavy chain are F and E, respectively.

[0278] In certain embodiments, in at least one of the first heavy chain and the second heavy chain, the amino acids at positions corresponding to positions L234, L235, and D265 in the human IgG1 heavy chain are F, E, and A, respectively.

[0279] In a particularly preferred embodiment, in both the first heavy chain and the second heavy chain, the amino acids at positions corresponding to positions L234, L235, and D265 in the human IgG1 heavy chain are F, E, and A, respectively.

[0280] In an even more particularly preferred embodiment, the antibody is a bispecific antibody comprising a first heavy chain and a second heavy chain, wherein the positions corresponding to positions L234 and L235 in the human IgG1 heavy chain according to EU numbering in both the first heavy chain and the second heavy chain are F and E, respectively, and (i) the position corresponding to F405 in the human IgG1 heavy chain according to EU numbering of the first heavy chain is L, and the position corresponding to K409 in the human IgG1 heavy chain according to EU numbering of the second heavy chain is R, or (ii) the position corresponding to K409 in the human IgG1 heavy chain according to EU numbering of the first heavy chain is R, and the position corresponding to F405 in the human IgG1 heavy chain according to EU numbering of the second heavy chain is L.

[0281] In a further particularly preferred embodiment, the antibody is a bispecific antibody comprising a first heavy chain and a second heavy chain, wherein the positions corresponding to L234, L235, and D265 in the human IgG1 heavy chain according to EU numbering are F, E, and A, respectively, for both the first heavy chain and the second heavy chain, and (i) the position corresponding to F405 in the human IgG1 heavy chain according to EU numbering of the first heavy chain is L, and the position corresponding to K409 in the human IgG1 heavy chain according to EU numbering of the second heavy chain is R, or (ii) the position corresponding to K409 in the human IgG1 heavy chain according to EU numbering of the first heavy chain is R, and the position corresponding to F405 in the human IgG1 heavy chain according to EU numbering of the second heavy chain is L.

[0282] Antibody variants having a combination of three amino acid substitutions L234F, L235E, and D265A and further having a K409R or F405L mutation are named herein with the suffix "FEAR" or "FEAL", respectively.

[0283] As used herein, huCD3-H1L1 refers to a humanized SP34 anti-CD3 antibody having the VH sequence shown in SEQ ID NO:25 and the VL sequence shown in SEQ ID NO:29.

[0284] In a preferred embodiment, the bispecific antibody of the present invention is (i) a half-molecule antibody derived from IgG1-huCD3-H1L1-FEAL and a half-molecule antibody derived from IgG1-PDL1-338-FEAR, IgG1-PDL1-511-FEAR, or IgG1-PDL1-547-FEAR, or (ii) a half-molecule antibody derived from IgG1-huCD3-H1L1-FEAR and a half-molecule antibody and a half-molecule antibody derived from IgG1-PDL1-338-FEAL, IgG1-PDL1-511-FEAL, or IgG1-PDL1-547-FEAL and comprises.

[0285] In a further aspect, the first heavy chain or half molecule comprises the sequence shown in SEQ ID NO:90 and the second heavy chain comprises the sequence shown in SEQ ID NO:89.

[0286] In a further aspect of the invention, to manipulate the serum half-life of the bispecific antibody, one or both antibody-forming portions of the bispecific antibody are engineered to reduce or increase binding to the neonatal Fc receptor (FcRn). Techniques for increasing or reducing serum half-life are well known in the art. See, for example, Dall’Acqua et al. 2006, J. Biol. Chem., 281:23514-24; Hinton et al. 2006, J. Immunol., 176:346-56; and Zalevsky et al. 2010 Nat. Biotechnol., 28:157-9.

[0287] Conjugate In a further aspect, the invention provides an antibody conjugated or fused to one or more therapeutic moieties such as cytokines, immunosuppressive agents, immunostimulatory molecules, and / or radioisotopes. Such conjugates are referred to herein as “immunoconjugates” or “drug conjugates”. An immunoconjugate that comprises one or more cytotoxins is called an “immunotoxin”.

[0288] In one aspect, the first Fc sequence and / or the second Fc sequence is conjugated to a drug or prodrug or contains an acceptor group for a drug or prodrug. Such acceptor groups may be, for example, unnatural amino acids.

[0289] Composition In a further aspect, the invention relates to a pharmaceutical composition comprising an antibody according to any one of the aspects disclosed herein and a pharmaceutically acceptable carrier.

[0290] The pharmaceutical composition of the present invention may contain one kind of antibody of the present invention or a combination of different antibodies of the present invention.

[0291] The pharmaceutical composition can be formulated according to conventional techniques, for example, the techniques disclosed in Remington: The Science and Practice of Pharmacy, 19th Edition, Gennaro, Ed., Mack Publishing Co., Easton, PA, 1995. The pharmaceutical composition of the present invention may contain, for example, diluents, extenders, salts, buffers, surfactants (e.g., nonionic surfactants, e.g., Tween-20 or Tween-80), stabilizers (e.g., amino acids without sugars or proteins), preservatives, tissue fixatives, solubilizers, and / or other materials suitable for inclusion in the pharmaceutical composition.

[0292] Pharmaceutically acceptable carriers include any and all suitable solvents, dispersion media, coatings, antibacterial and antifungal agents, isotonic agents, antioxidants, and absorption delaying agents that are physiologically compatible with the antibodies of the present invention. Examples of suitable aqueous and non-aqueous carriers that can be used in the pharmaceutical composition of the present invention include water, saline, phosphate buffered saline, ethanol, dextrose, polyols (e.g., glycerol, propylene glycol, polyethylene glycol, etc.), and suitable mixtures thereof, vegetable oils, carboxymethylcellulose colloidal solutions, tragacanth gum, and injectable organic esters, such as ethyl oleate, and / or various buffers. Pharmaceutically acceptable carriers include sterile aqueous solutions or dispersions, and sterile powders for the immediate preparation of sterile injectable solutions or dispersions. Suitable fluidity can be maintained, for example, by using coating materials, such as lecithin, in the case of dispersions, by maintaining the required particle size, and by using surfactants.

[0293] The pharmaceutical composition of the present invention may also contain pharmaceutically acceptable antioxidants, for example, (1) water-soluble antioxidants such as ascorbic acid, cysteine hydrochloride, sodium bisulfate, sodium metabisulfite, sodium sulfite, etc.; (2) oil-soluble antioxidants such as ascorbyl palmitate, butylated hydroxyanisole, butylated hydroxytoluene, lecithin, propyl gallate, α-tocopherol, etc.; and (3) metal chelating agents such as citric acid, ethylenediaminetetraacetic acid (EDTA), sorbitol, tartaric acid, phosphoric acid, etc.

[0294] The pharmaceutical composition of the present invention may also contain in the composition an isotonic agent such as sugar, polyhydric alcohol such as mannitol, sorbitol, glycerol, or sodium chloride.

[0295] The pharmaceutical composition of the present invention may also contain one or more adjuvants suitable for the selected route of administration that can enhance the shelf life or effectiveness of the pharmaceutical composition, such as preservatives, wetting agents, emulsifying agents, dispersing agents, preservatives, or buffers. The antibody of the present invention may be prepared using a sustained-release preparation including a carrier that protects the antibody from rapid release, such as a delivery system encapsulated in a graft, transdermal patch, and microcapsule. Such carriers may include gelatin, glyceryl monostearate, glyceryl distearate, biodegradable biocompatible polymers such as ethylene vinyl acetate, polyanhydrides, polyglycolic acid, collagen, polyorthoesters, and polylactic acid alone or in combination with wax, or other materials well known in the art. Methods for preparing such formulations are generally known to those skilled in the art.

[0296] A sterile injectable solution can be prepared by incorporating the required amount of the active compound, optionally with one or a combination of the ingredients enumerated above, into a suitable solvent and then performing sterile microfiltration. Generally, the dispersion is prepared by incorporating the active compound into a sterile vehicle containing a basic dispersion medium and the other ingredients required, for example, those enumerated above. In the case of a sterile powder for preparing a sterile injectable solution, an example of the preparation method is vacuum drying and freeze-drying (lyophilization) to yield a powder of the active ingredient + any additional desired ingredients from a pre-filter-sterilized solution of the active ingredient + any additional desired ingredients.

[0297] The actual dosage level of the active ingredient in the pharmaceutical composition can be varied so as to obtain an amount of the active ingredient that is effective to achieve the desired therapeutic response for a particular patient, composition, and method of administration without undue toxicity to the patient. The selected dosage level will depend on a variety of pharmacokinetic factors including the activity of the particular composition or its amide used in the invention, route of administration, time of administration, rate of excretion of the particular compound being used, duration of treatment, other drugs, compounds, and / or materials used in combination with the particular composition being used, the age, sex, weight, condition, general health, and prior medical history of the patient being treated, as well as other factors well known in the medical arts.

[0298] The pharmaceutical composition can be administered by any suitable route and method. In one embodiment, the pharmaceutical composition of the invention is administered parenterally. As used herein, "administered parenterally" means a route of administration other than enteral and topical administration, usually by injection, including injection and infusion into the epidermis, intravenous, intramuscular, intraarterial, intrathecal, intracapsular, intraorbital, intracardiac, intradermal, intraperitoneal, intratendinous, transtracheal, subcutaneous, subepidermal, intraarticular, subcapsular, intrathecal, intraspinal, intracranial, intrathoracic, epidural, and intrasternal.

[0299] In one embodiment, the pharmaceutical composition is administered by intravenous or subcutaneous injection or infusion.

[0300] Use In one aspect, the present invention relates to an antibody according to any one of the aspects disclosed herein or a pharmaceutical composition disclosed herein for use as a medicament.

[0301] In a further aspect, the present invention relates to an antibody according to any one of the aspects disclosed herein or a pharmaceutical composition disclosed herein for use in the treatment of diseases such as cancer.

[0302] In a further aspect, the present invention relates to a method for treating a disease, comprising the step of administering to a subject in need thereof an effective amount of an antibody according to any one of the aspects disclosed herein or a pharmaceutical composition disclosed herein.

[0303] In particular, the bispecific antibodies according to the present invention may be useful in therapeutic settings where specific targeting and killing of PD-L1-expressing cells by T cells is desirable, and may be more efficient in certain specific indications and situations compared to conventional anti-PD-L1 antibodies.

[0304] The antibodies of the present invention also have further utility in the therapy and diagnosis of various PD-L1-related diseases. For example, the antibodies can be used to induce one or more of the following biological activities in vivo or in vitro: inhibition of the proliferation and / or differentiation of PD-L1-expressing cells; killing of PD-L1-expressing cells; mediating phagocytosis or ADCC of PD-L1-expressing cells in the presence of human effector cells; mediating CDC of PD-L1-expressing cells in the presence of complement; mediating apoptosis of PD-L1-expressing cells; and / or inducing translocation to lipid rafts upon PD-L1 binding.

[0305] In one aspect, the present invention relates to an antibody according to any one of the aspects disclosed herein or a pharmaceutical composition disclosed herein for use in the treatment of cancer.

[0306] In a further aspect, the invention relates to an antibody according to any one of the aspects disclosed herein or a pharmaceutical composition disclosed herein for use in the treatment of cancer diseases characterized by the presence of solid tumors.

[0307] In a further aspect, the invention relates to an antibody according to any one of the aspects disclosed herein or a pharmaceutical composition disclosed herein for use in the treatment of cancer diseases selected from the group consisting of melanoma, ovarian cancer, lung cancer, colorectal cancer, head and neck cancer, gastric cancer, breast cancer, kidney cancer, bladder cancer, esophageal cancer, pancreatic cancer, liver cancer, thymoma and thymic cancer, brain cancer, glioma, adrenocortical cancer, thyroid cancer, other skin cancers, sarcoma, multiple myeloma, leukemia, lymphoma, myelodysplastic syndrome, ovarian cancer, endometrial cancer, prostate cancer, penile cancer, Hodgkin lymphoma, non-Hodgkin lymphoma, Merkel cell cancer, and mesothelioma.

[0308] In a further aspect, the invention relates to the use of an antibody according to any one of the aspects disclosed herein for the manufacture of a medicament, such as a medicament for the treatment of cancer, for example, cancer diseases characterized by the presence of solid tumors, or cancer diseases selected from the group consisting of melanoma, ovarian cancer, lung cancer, colon cancer, and head and neck cancer.

[0309] The invention also relates to a method for inhibiting the growth and / or proliferation of one or more PD-L1-expressing tumor cells, comprising administering an antibody of the invention to an individual in need thereof.

[0310] The invention also relates to (a) selecting a subject suffering from cancer comprising PD-L1-expressing tumor cells, and (b) administering to the subject an antibody of the invention or a pharmaceutical composition of the invention and relates to a method for treating cancer.

[0311] The dosing regimens in the aforementioned methods of treatment and uses are adjusted to provide an optimal desirable response (e.g., a therapeutic response). For example, a single bolus may be administered, or several divided doses may be administered over a period of time, and the dosage may be proportionally decreased or increased as indicated by the exigencies of the therapeutic situation. Parenteral compositions may be formulated in unit dosage form for ease of administration and uniformity of dosing.

[0312] The effective dosage and dosing regimen of the antibody depend on the disease or condition to be treated and can be determined by one of ordinary skill in the art. Exemplary and non-limiting ranges for a therapeutically effective amount of the compounds of the invention are from about 0.001 to 10 mg / kg, such as from about 0.001 to 5 mg / kg, such as from about 0.001 to 2 mg / kg, such as from about 0.001 to 1 mg / kg, such as about 0.001 mg / kg, about 0.01 mg / kg, about 0.1 mg / kg, about 1 mg / kg, or about 10 mg / kg. Another exemplary and non-limiting range for a therapeutically effective amount of the antibodies of the invention is from about 0.1 to 100 mg / kg, such as from about 0.1 to 50 mg / kg, such as from about 0.1 to 20 mg / kg, such as from about 0.1 to 10 mg / kg, such as about 0.5 mg / kg, such as about 0.3 mg / kg, about 1 mg / kg, about 3 mg / kg, about 5 mg / kg, or about 8 mg / kg.

[0313] A physician or veterinarian having ordinary skill in the art can readily determine and prescribe an effective amount of the pharmaceutical composition required. For example, a physician or veterinarian can start with a dosage of the antibody used in the pharmaceutical composition at a level lower than that required to achieve the desired therapeutic effect and gradually increase the dosage until the desired effect is achieved. Generally, the appropriate daily dose of the antibody of the invention is the amount of compound that is the minimum dose effective to produce a therapeutic effect. Administration may be, for example, parenteral, such as intravenous, intramuscular, or subcutaneous. In one embodiment, the antibody is mg / m 2It may be administered by injection at a weekly dosage calculated in units. Such a dosage may be based on the mg / kg dosage shown above, for example, according to dosage (mg / kg) × 70:1.8. Such administration may be repeated, for example, 1 to 8 times, for example, 3 to 5 times. Administration may be performed by continuous infusion over a period of 2 to 24 hours, for example, 2 to 12 hours. In one embodiment, in order to reduce toxic side effects, the antibody may be administered by slow continuous infusion over a long period, for example, exceeding 24 hours.

[0314] In one embodiment, when the antibody is administered once a week, it may be administered at a weekly dosage calculated as a certain dosage, up to 8 times, for example, 4 to 6 times. Such a regimen may be repeated one or more times as needed, for example, after 6 months or 12 months. Such a certain dosage may be based on the mg / kg dosage shown above, with an estimated body weight of 70 kg. The dosage may be determined or adjusted by measuring the amount of the antibody in the blood at the time of administration, for example, by collecting a biological sample and using an anti-idiotype antibody that targets the PD-L1 antigen-binding region of the antibody of the present invention.

[0315] In one embodiment, the antibody may be administered as maintenance therapy, for example, once a week for 6 months or more.

[0316] The antibody may also be administered prophylactically to reduce the risk of developing cancer, to delay the onset of events in cancer progression, and / or to reduce the risk of recurrence when the cancer is in remission.

[0317] The antibody of the present invention may also be administered in combination therapy, that is, in combination with other therapeutic substances related to the disease or condition to be treated. Thus, in one embodiment, the pharmaceutical containing the antibody is for use in combination with one or more additional therapeutic substances, for example, a cytotoxic agent, a chemotherapeutic agent, or an anti-angiogenic agent.

[0318] In a further aspect, the present invention relates to an anti-idiotype antibody that binds to a PD-L1 binding region defined in any one of the aspects disclosed herein.

[0319] Further items of the present disclosure: 1. An antibody comprising an antigen-binding region capable of binding to human PD-L1, which inhibits the binding of human PD-L1 to human PD-1, and (i) competes with an antibody

[0511] comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15 for binding to human PD-L1, but does not compete with an antibody

[0547] comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 for binding to human PD-L1, or (ii) competes with an antibody

[0547] comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 for binding to human PD-L1, but does not compete with an antibody

[0511] comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15 for binding to human PD-L1, said antibody. 2. The antibody according to item 1, which competes with an antibody

[0338] comprising the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5 for binding to human PD-L1. 3. The antibody according to any of the preceding items, wherein the binding of the antibody according to any of the preceding items to human PD-L1 is not replaced by an antibody

[0476] comprising the VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57. 4. The antibody according to any of the preceding items, wherein the binding of the antibody according to any of the preceding items to human PD-L1 is not blocked by the binding of an antibody

[0625] comprising the VH sequence shown in SEQ ID NO:106 and the VL sequence shown in SEQ ID NO:110. 5. The antibody according to any of the preceding items, wherein the binding between the antibody according to any of the preceding items and human PD-L1 is blocked by an antibody

[0547] comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22. 6. (i) An antibody that can bind to the same human PD-L1 epitope as the antibody

[0338] comprising the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5, or (ii) An antibody that can bind to the same human PD-L1 epitope as the antibody

[0511] comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15, or (iii) An antibody that can bind to the same human PD-L1 epitope as the antibody

[0547] comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22, The antibody according to item 1. 7. The antibody according to any of items 1 to 4 and 6, wherein the binding between the antibody according to any of items 1 to 4 and 6 and a mutant PD-L1 in which any one or more of the amino acid residues (R113) at the position corresponding to position 113 in SEQ ID NO:94, the amino acid residue (Y123) at the position corresponding to position 123, and the amino acid residue (R125) at the position corresponding to position 125 are substituted with alanine is reduced as compared to the binding to wild-type PD-L1 having the amino acid sequence shown in SEQ ID NO:94, and the reduction in binding is determined when the change ratio of the binding of the antibody is smaller than the average of the change ratios of the binding over all alanine mutants - 1.5×SD, where SD is the standard deviation of all the calculated change ratio values of the antibody with respect to the mutant PDL1, and the change ratio of the binding is calculated as shown in Example 13

[0338] , the antibody according to any of items 1 to 4 and 6. 8. The antibody according to any one of items 1 to 2 and 6 binds to an epitope on PD-L1 (SEQ ID NO: 94), and the epitope comprises the amino acid residue (R113) at position 113, the amino acid residue (Y123) at position 123, and / or the amino acid residue (R125) at position 125 of SEQ ID NO: 94. The antibody according to any one of items 1 to 2 and 6. 9. The binding of the antibody according to any one of items 1, 3, 4, and 6 to a mutant PD-L1 in which any one or more of the amino acid residues (F19) corresponding to position 19 in SEQ ID NO: 94, the amino acid residue (F42) corresponding to position 42, the amino acid residue (E45) corresponding to position 45, the amino acid residue (K46) corresponding to position 46, the amino acid residue (L94) corresponding to position 94, and the amino acid residue (I116) corresponding to position 116 are substituted with alanine is reduced as compared with wild-type PD-L1 having the amino acid sequence shown in SEQ ID NO: 94. When the change factor of the binding of the antibody is smaller than the average of the change factors of the binding over all alanine mutants - 1.5×SD, the reduction of the binding is determined, where SD is the standard deviation of all the calculated change factors of the antibody with respect to the mutant PDL1, and the change factor of the binding is calculated as shown in Example 13. The antibody according to any one of items 1, 3, 4, and 6. 10. The antibody according to any one of items 1, 3, 4, and 6 binds to an epitope on PD-L1 (SEQ ID NO: 94), and the epitope comprises one or more amino acid residues selected from the group consisting of the amino acid residue (E45) at position 45, the amino acid residue (K46) at position 46, and / or the amino acid residue (L94) at position 94 of SEQ ID NO: 94. The antibody according to any one of items 1, 3, 4, and 6. 11. The binding of the antibody according to any one of items 1, 5, and 6 to a mutant PD-L1 in which one or more of the amino acid residues (E58) at the position corresponding to position 58 in SEQ ID NO: 94 and the amino acid residues (R113) at the position corresponding to position 113 are substituted with alanine is reduced compared to wild-type PD-L1 having the amino acid sequence shown in SEQ ID NO: 94. When the change factor of the antibody binding is smaller than the average of the change factors of the binding over all alanine mutants - 1.5×SD, where SD is the standard deviation of all the calculated change factors of the antibody with respect to the mutant PDL1, and the change factor of the binding is calculated as shown in Example 13, the antibody according to any one of items 1, 5, and 6. 12. The antibody according to any one of items 1, 5, and 6, wherein the antibody binds to an epitope on PD-L1 (SEQ ID NO: 94), and the epitope comprises the amino acid residue (E58) at position 58 and / or the amino acid residue (R113) at position 113 of SEQ ID NO: 94. 13. The antigen-binding region capable of binding to human PD-L1 comprises a heavy-chain variable region (VH) comprising a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, and a light-chain variable region (VL) comprising a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, and the VH CDR3 sequence is selected from the group consisting of the sequences shown in SEQ ID NO: 4, SEQ ID NO: 11, and SEQ ID NO: 21, the antibody according to any one of the preceding items. 14. The antigen-binding region capable of binding to human PD-L1 is (i) a heavy-chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO: 2, the CDR2 sequence shown in SEQ ID NO: 3, and the CDR3 sequence shown in SEQ ID NO: 4, and a light-chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO: 6, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO: 7

[0338] , or (ii) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:9, the CDR2 sequence shown in SEQ ID NO:10, and the CDR3 sequence shown in SEQ ID NO:11, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:16, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:17

[0511] , or (iii) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:19, the CDR2 sequence shown in SEQ ID NO:20, and the CDR3 sequence shown in SEQ ID NO:21, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:23, a CDR2 having the sequence DDN, and a CDR3 having the sequence shown in SEQ ID NO:24

[0547] An antibody according to any of the preceding items, comprising 15. The antibody according to any of the preceding items, wherein the antigen-binding region capable of binding to human PD-L1 comprises a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VH sequence selected from the group consisting of the sequences shown in SEQ ID NO:1, SEQ ID NO:8, and SEQ ID NO:18. 16. The antibody according to any of the preceding items, wherein the antigen-binding region capable of binding to human PD-L1 comprises a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VL sequence selected from the group consisting of the sequences shown in SEQ ID NO:5, SEQ ID NO:15, and SEQ ID NO:22. 17. The antigen-binding region capable of binding to human PD-L1 is (i) A VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:1, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:5

[0338] , or (ii) A VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:8, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:15

[0511] , or (iii) A VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:18, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:22

[0547] The antibody according to any one of the above items, comprising 18. Each of the VH and VL sequences includes three CDR sequences, namely CDR1, CDR2, and CDR3, and four framework sequences, namely FR1, FR2, FR3, and FR4. Each combined FR1, FR2, FR3, and FR4 framework sequence of VH has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with the respective combined FR1, FR2, FR3, and FR4 framework sequences of the VH sequence, and the VH CDR sequences are not mutated. Each combined FR1, FR2, FR3, and FR4 framework sequence of VL has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with the respective combined FR1, FR2, FR3, and FR4 framework sequences of the VL sequence, and the VL CDR sequences are not mutated. The antibody according to item 17. 19. The antibody according to any of the preceding items, which can induce dose-dependent lysis of epithelial cells of adenocarcinoma, such as dose-dependent lysis of MDA-MB-231, via antibody-dependent cell-mediated cytotoxicity (ADCC). 20. The antibody according to item 19, which can reduce the number of cells in the culture of the epithelial cells by at least 5%, such as at least 6%, 7%, 8%, 9%, or at least 10% as a result of cell lysis. 21. ADCC is 51 Measured in vitro by the Cr release assay method, such as the assay method disclosed in Example 14. The antibody according to item 19 or 20. 22. ADCC is measured in vitro by incubating the epithelial cells with a composition containing the antibody according to item 19 or 20 and effector cells, such as peripheral blood mononuclear cells (PBMC), at 37°C and 5% CO 2 for 4 hours, wherein the amount of the antibody in the composition is in the range of 0.1 - 1 μg / mL, and the ratio of effector cells to epithelial cells is 100:1. The antibody according to item 19 or 20. 23. The lysis of the epithelial cells is measured in vitro in a luciferase reporter assay method as an alternative to ADCC, for example, the luminescence ADCC reporter bioassay method disclosed in Example 14, and the antibody according to item 19 or 20. 24. ADCC is (i) contacting a culture of the epithelial cells with a composition comprising the antibody according to item 23, FcγRIIIa (CD16), and Jurkat human T cells (effector cells) stably expressing firefly luciferase, at an effector cell:epithelial cell ratio of 1:1, (ii) adjusting the culture of the epithelial cells and the effector cells to room temperature for 15 minutes, (iii) incubating the culture of the epithelial cells and the effector cells with a luciferase substrate, and (iv) measuring luciferase production in the cell culture by which it is measured in vitro, the amount of the antibody in the composition is in the range of 0.5 to 250 ng / mL, and the ratio of effector cells to epithelial cells is 1:1, and the antibody according to item 23. 25. The ADCC of the epithelial cells is measured in a luciferase reporter assay method, for example, the reporter assay method defined in item 23 or 24, and then the ADCC observed after incubating the culture of the epithelial cells with a test composition comprising the antibody according to any one of items 19, 20, 23, and 24 is at least 1.5 times the ADCC observed after incubating the culture of the epithelial cells with a composition comprising a reference antibody. The ADCC is measured as relative light units (RLU), the antibody concentration in the test composition and the antibody concentration in the composition comprising the reference antibody are the same and in the range of 20 to 250 ng / ml, and the reference antibody is (a) an antibody comprising the VH sequence shown in SEQ ID NO:74 and the VL sequence shown in SEQ ID NO:78; and (b) an antibody comprising the VH sequence shown in SEQ ID NO:81 and the VL sequence shown in SEQ ID NO:85 An antibody according to any one of items 19, 20, 23, and 24, selected therefrom. 26. The antigen-binding region capable of binding to human PD-L1, (i) the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5

[0338] , or (ii) the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15

[0511] , or (iii) the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22

[0547] An antibody according to any one of the above items, comprising. 27. An antibody comprising an antigen-binding region capable of binding to human PD-L1, (i) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:33, the CDR2 sequence shown in SEQ ID NO:34, and the CDR3 sequence shown in SEQ ID NO:35, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:37, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:38

[0321] , or (ii) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:47, the CDR2 sequence shown in SEQ ID NO:48, and the CDR3 sequence shown in SEQ ID NO:49, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:51, a CDR2 having the sequence DVI, and a CDR3 having the sequence shown in SEQ ID NO:52

[0421] , or (iii) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:54, the CDR2 sequence shown in SEQ ID NO:55, and the CDR3 sequence shown in SEQ ID NO:56, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:58, a CDR2 having the sequence RDS, and a CDR3 having the sequence shown in SEQ ID NO:59

[0476] , or (iv)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:61, the CDR2 sequence shown in SEQ ID NO:62, and the CDR3 sequence shown in SEQ ID NO:63, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:65, a CDR2 having the sequence DDS, and a CDR3 having the sequence shown in SEQ ID NO:66

[0516] , or (v)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:107, the CDR2 sequence shown in SEQ ID NO:108, and the CDR3 sequence shown in SEQ ID NO:109, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:111, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:113

[0625] , or (vi)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:68, the CDR2 sequence shown in SEQ ID NO:69, and the CDR3 sequence shown in SEQ ID NO:70, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:72, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:73

[0632] The antibody comprising the above. 28. (i)The VH sequence shown in SEQ ID NO:32 and the VL sequence shown in SEQ ID NO:36

[0321] , or (ii)The VH sequence shown in SEQ ID NO:46 and the VL sequence shown in SEQ ID NO:50

[0421] , or (iii)The VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57

[0476] , or (iv)The VH sequence shown in SEQ ID NO:60 and the VL sequence shown in SEQ ID NO:64

[0516] , or (v)The VH sequence shown in SEQ ID NO:106 and the VL sequence shown in SEQ ID NO:110

[0625] , or (vi) The VH sequence shown in SEQ ID NO:67 and the VL sequence shown in SEQ ID NO:71

[0632] The antibody according to item 27, comprising the same. 29. The antibody according to any of the preceding items, which is monovalent. 30. The antibody according to any of the preceding items, wherein the antibody according to any of the preceding items is a bivalent antibody having two antigen-binding regions capable of binding to human PD-L1, and the two antigen-binding regions have the same variable region sequence. 31. The antibody according to any of the preceding items, wherein the antibody according to any of the preceding items is a bivalent bispecific antibody, and in addition to the (first) antigen-binding region capable of binding to human PD-L1, it comprises a (second) antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, and the second antigen is not human CD3ε. 32. A bispecific antibody comprising an antigen-binding region capable of binding to human PD-L1 and an antigen-binding region capable of binding to human CD3ε (epsilon), wherein the antigen-binding region capable of binding to human PD-L1 has the characteristics shown in any of the preceding items. 33. The bispecific antibody according to item 32, wherein the antigen-binding region capable of binding to human CD3ε comprises a heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31. 34. (i) An antigen-binding region capable of binding to human PD-L1, comprising a heavy-chain variable region (VH) containing the CDR1 sequence shown in SEQ ID NO:2, the CDR2 sequence shown in SEQ ID NO:3, and the CDR3 sequence shown in SEQ ID NO:4, and a light-chain variable region (VL) containing a CDR1 having the sequence shown in SEQ ID NO:6, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:7

[0338] , and (a) a heavy-chain variable region (VH) containing a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light-chain variable region (VL) containing a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, an antigen-binding region capable of binding to human CD3ε, or (ii) An antigen-binding region capable of binding to human PD-L1, comprising a heavy-chain variable region (VH) containing the CDR1 sequence shown in SEQ ID NO:9, the CDR2 sequence shown in SEQ ID NO:10, and the CDR3 sequence shown in SEQ ID NO:11, and a light-chain variable region (VL) containing a CDR1 having the sequence shown in SEQ ID NO:16, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:17

[0338] , and (a) a heavy-chain variable region (VH) containing a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light-chain variable region (VL) containing a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, an antigen-binding region capable of binding to human CD3ε, or (iii)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:19, the CDR2 sequence shown in SEQ ID NO:20, and the CDR3 sequence shown in SEQ ID NO:21, and a light chain variable region (VL) comprising a CDR1 sequence having the sequence shown in SEQ ID NO:23, a CDR2 sequence having the sequence DDN, and a CDR3 sequence having the sequence shown in SEQ ID NO:24, an antigen-binding region capable of binding to human PD-L1

[0547] , and (a) a heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, an antigen-binding region capable of binding to human CD3ε The bispecific antibody according to item 32 or 33, comprising 35. The bispecific antibody according to any one of items 32 to 34, wherein the antigen-binding region capable of binding to human CD3ε comprises the VH sequence shown in SEQ ID NO:25 and the VL sequence shown in SEQ ID NO:29. 36. (i) The bispecific antibody has a lower affinity for human CD3ε binding compared to an antibody having an antigen-binding region that can comprise the VH sequence shown in SEQ ID NO:25 and the VL sequence shown in SEQ ID NO:29, preferably, the affinity is at least 1 / 2-fold, for example, at least 1 / 5-fold, for example, at least 1 / 10-fold, for example, at least 1 / 25-fold, for example, at least 1 / 50-fold, and (ii) The bispecific antibody is capable of mediating concentration-dependent cytotoxicity of MDA-MB-231 cells, PC-3 cells, and / or HELA cells when PBMC or purified T cells are used as effector cells, for example, as assayed as described in Example 11 herein. The bispecific antibody according to item 32. 37. the antigen-binding region capable of binding to human CD3ε, (i) a heavy-chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:99, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light-chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (ii) a heavy-chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:100, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:28, and a light-chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (iii) a heavy-chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:101, and a light-chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (iv) a heavy-chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:102, and a light-chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (v)A heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:103, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (vi)A heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:104, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31, or (vii)A heavy chain variable region (VH) comprising a CDR1 having the sequence shown in SEQ ID NO:26, a CDR2 having the sequence shown in SEQ ID NO:27, and a CDR3 having the sequence shown in SEQ ID NO:105, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence shown in SEQ ID NO:31 The bispecific antibody according to item 36, comprising 38. The antigen-binding region capable of binding to human CD3ε is (i) The VH sequence shown in SEQ ID NO:39 and the VL sequence shown in SEQ ID NO:29, or (ii) The VH sequence shown in SEQ ID NO:40 and the VL sequence shown in SEQ ID NO:29, or (iii) The VH sequence shown in SEQ ID NO:41 and the VL sequence shown in SEQ ID NO:29, or (iv) The VH sequence shown in SEQ ID NO:42 and the VL sequence shown in SEQ ID NO:29, or (v) The VH sequence shown in SEQ ID NO:43 and the VL sequence shown in SEQ ID NO:29, or (vi) the VH sequence shown in SEQ ID NO:44 and the VL sequence shown in SEQ ID NO:29, or (vii) the VH sequence shown in SEQ ID NO:45 and the VL sequence shown in SEQ ID NO:29 The bispecific antibody according to item 36 or 37, comprising 39. A multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, wherein the antigen-binding region capable of binding to human PD-L1 has the characteristics shown in any of items 1 to 31. The multispecific antibody. 40. The antigen-binding region capable of binding to human PD-L1 comprises a heavy chain variable region (VH) comprising a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, and a light chain variable region (VL) comprising a CDR1 sequence, a CDR2 sequence, and a CDR3 sequence, and the VH CDR3 sequence is selected from the group consisting of the sequences shown in SEQ ID NO:4

[0338] , SEQ ID NO:11

[0511] , and SEQ ID NO:21

[0547] . The multispecific antibody according to item 39. 41. The first antigen-binding region capable of binding to human PD-L1 is (i) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:2, the CDR2 sequence shown in SEQ ID NO:3, and the CDR3 sequence shown in SEQ ID NO:4, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:6, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:7

[0338] , or (ii)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:9, the CDR2 sequence shown in SEQ ID NO:10, and the CDR3 sequence shown in SEQ ID NO:11, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:16, a CDR2 having sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:17

[0511] , or (iii)A heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:19, the CDR2 sequence shown in SEQ ID NO:20, and the CDR3 sequence shown in SEQ ID NO:21, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:23, a CDR2 having sequence DDN, and a CDR3 having the sequence shown in SEQ ID NO:24

[0547] The multispecific antibody according to item 40, comprising 42. The multispecific antibody according to item 40 or 41, wherein the first antigen-binding region capable of binding to human PD-L1 comprises a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VH sequence selected from the group consisting of the sequences shown in SEQ ID NO:1

[0338] , SEQ ID NO:8

[0511] , and SEQ ID NO:18

[0547] . 43. The multispecific antibody according to any one of items 40 to 42, wherein the first antigen-binding region capable of binding to human PD-L1 comprises a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with a VL sequence selected from the group consisting of the sequences shown in SEQ ID NO:5

[0338] , SEQ ID NO:15

[0511] , and SEQ ID NO:22

[0547] . 44. The first antigen-binding region capable of binding to human PD-L1 is (i) A VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:1, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:5

[0338] , or (ii) A VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:8, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:15

[0511] , or (iii) A VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VH sequence shown in SEQ ID NO:18, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity with the VL sequence shown in SEQ ID NO:22

[0547] The multispecific antibody according to any one of items 40 to 43, comprising 45. Each of the VH and VL sequences includes three CDR sequences, namely CDR1, CDR2, and CDR3, and four framework sequences, namely FR1, FR2, FR3, and FR4, and each combined FR1, FR2, FR3, and FR4 framework sequence of VH has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with the respective combined FR1, FR2, FR3, and FR4 framework sequences of the VH sequence, and the VH CDR sequences are not mutated, and each combined FR1, FR2, FR3, and FR4 framework sequence of VL has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with the respective combined FR1, FR2, FR3, and FR4 framework sequences of the VL sequence, and the VL CDR sequences are not mutated. The multispecific antibody according to any one of items 40 to 44. 46. The first antigen-binding region capable of binding to human PD-L1 is (i) the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5

[0338] , or (ii) the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15

[0511] , or (iii) the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22

[0547] The multispecific antibody according to any one of items 40 to 45, comprising. 47. A multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, (i) competes with an antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15 for binding to human PD-L1, but does not compete with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 for binding to human PD-L1, or (ii) With respect to binding to human PD-L1, it competes with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22, but does not compete with an antibody comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15 with respect to binding to human PD-L1, The multispecific antibody. 48. The multispecific antibody according to item 47, which competes with an antibody comprising the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5 with respect to binding to human PD-L1. 49. A multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, The multispecific antibody, wherein the binding of the antibody to human PD-L1 is not replaced by an antibody comprising the VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57. 50. The multispecific antibody according to item 49, which inhibits the binding between human PD-L1 and human PD-1. 51. The multispecific antibody according to item 49 or 50, which competes with an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22 with respect to binding to human PD-L1. 52. The multispecific antibody according to any one of items 49 to 51, wherein the binding between the multispecific antibody according to any one of items 49 to 51 and human PD-L1 is blocked by an antibody comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22. 53. A multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or a different human PD-L1 epitope, wherein the first antigen-binding region is (i) An antibody that can bind to the same human PD-L1 epitope as the antibody

[0338] comprising the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5, or (ii) An antibody that can bind to the same human PD-L1 epitope as the antibody

[0511] comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15, or (iii) An antibody that can bind to the same human PD-L1 epitope as the antibody

[0547] comprising the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22, Said multispecific antibody. 54. The multispecific antibody according to any one of items 40 to 53, which is bispecific. 55. The multispecific antibody according to item 54, which is bivalent. 56. The multispecific antibody according to any one of items 40 to 55, wherein the multispecific antibody according to any one of items 40 to 55 can bind to a second antigen and the second antigen is not human CD3ε. 57. The antibody according to any one of the preceding items, which is a full-length antibody. 58. The antibody according to item 57, which is a full-length IgG1 antibody. 59. The antibody according to any one of the preceding items, which is an antibody fragment. 60. The antibody according to any one of items 32 to 59 comprises two half-molecules each containing an antigen-binding region, (i) The half-molecule containing the antigen-binding region that can bind to human PD-L1 is chimeric, and / or (ii) The half-molecule containing the antigen-binding region that can bind to human CD3ε (epsilon) is chimeric if present, The antibody according to any one of items 32 to 59. 61. (i) The antigen-binding region that can bind to human PD-L1 is humanized, and / or (ii) The antigen-binding region that can bind to human CD3ε (epsilon) is humanized if present, The antibody according to any one of the preceding items. 62. (i) The antigen-binding region capable of binding to human PD-L1 is human and / or (ii) The antigen-binding region capable of binding to human CD3ε (epsilon) is human, if present. The antibody according to any one of the preceding items. 63. Each of the antigen-binding regions comprises a heavy-chain variable region (VH) and a light-chain variable region (VL), and each of the variable regions comprises three CDR sequences, CDR1, CDR2, and CDR3, respectively, and four framework sequences, FR1, FR2, FR3, and FR4, respectively, The antibody according to any one of the preceding items. 64. The antibody according to item 63, comprising two heavy-chain constant regions (CH) and two light-chain constant regions (CL). 65. The antibody according to any one of the preceding items, wherein the antibody comprises a first heavy chain and a second heavy chain, each of the first heavy chain and the second heavy chain comprising at least a hinge region, CH2, and a CH3 region, and in the first heavy chain, at least one amino acid at a position corresponding to a position selected from the group consisting of T366, L368, K370, D399, F405, Y407, and K409 (according to EU numbering) is substituted, and in the second heavy chain, at least one amino acid at a position corresponding to a position selected from the group consisting of T366, L368, K370, D399, F405, Y407, and K409 (according to EU numbering) is substituted, and the first heavy chain and the second heavy chain are not substituted at the same position. The antibody according to any one of the preceding items. 66. (i) In the first heavy chain, the amino acid at the position corresponding to F405 (according to EU numbering) is L, and in the second heavy chain, the amino acid at the position corresponding to K409 (according to EU numbering) is R, or (ii) in the first heavy chain, the amino acid at the position corresponding to K409 (according to EU numbering) is R, and in the second heavy chain, the amino acid at the position corresponding to F405 (according to EU numbering) is L. The antibody according to item 65. 67. The antibody according to any of the preceding items comprises a first heavy chain and a second heavy chain, and the antibody induces effector functions via Fc to a lesser extent compared to an antibody identical except that it comprises an unmodified first heavy chain and a second heavy chain, wherein one or both heavy chains are modified. The antibody according to any of the preceding items. 68. The antibody according to item 67, wherein the effector function via Fc is measured by determining Fc-mediated CD69 expression, by binding to an Fcγ receptor, by binding to C1q, or by inducing FcR cross-linking via Fc. 69. The antibody according to item 67 or 68, wherein the heavy chain constant region and the light chain constant region are modified such that Fc-mediated CD69 expression is reduced by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99%, or 100% when compared to the wild-type antibody, and the Fc-mediated CD69 expression is measured in a functional assay based on PBMC. The antibody according to item 67 or 68. 70. The antibody according to any of the preceding items comprises a first heavy chain and a second heavy chain, and in at least one of the first heavy chain and the second heavy chain, one or more amino acids at positions corresponding to positions L234, L235, D265, N297, and P331 in a human IgG1 heavy chain according to EU numbering are not L, L, D, N, and P, respectively, of the antibody according to any of the preceding items. 71. The antibody according to item 70, wherein the positions corresponding to positions L234 and L235 in a human IgG1 heavy chain according to EU numbering are F and E, respectively, in the first heavy chain and the second heavy chain. 72. The antibody according to item 71 is a bispecific antibody comprising a first heavy chain and a second heavy chain, and the positions corresponding to L234 and L235 in the human IgG1 heavy chain according to EU numbering in both the first heavy chain and the second heavy chain are F and E, respectively, and (i) the position corresponding to F405 in the human IgG1 heavy chain according to EU numbering in the first heavy chain is L, and the position corresponding to K409 in the human IgG1 heavy chain according to EU numbering in the second heavy chain is R, or (ii) the position corresponding to K409 in the human IgG1 heavy chain according to EU numbering in the first heavy chain is R, and the position corresponding to F405 in the human IgG1 heavy chain according to EU numbering in the second heavy chain is L, the antibody according to item 71. 73. In the first heavy chain and the second heavy chain, the positions corresponding to L234, L235, and D265 in the human IgG1 heavy chain according to EU numbering are F, E, and A, respectively, the antibody according to item 70. 74. The antibody according to item 73 is a bispecific antibody comprising a first heavy chain and a second heavy chain, and the positions corresponding to L234, L235, and D265 in the human IgG1 heavy chain according to EU numbering in both the first heavy chain and the second heavy chain are F, E, and A, respectively, and (i) the position corresponding to F405 in the human IgG1 heavy chain according to EU numbering in the first heavy chain is L, and the position corresponding to K409 in the human IgG1 heavy chain according to EU numbering in the second heavy chain is R, or (ii) the position corresponding to K409 in the human IgG1 heavy chain according to EU numbering in the first heavy chain is R, and the position corresponding to F405 in the human IgG1 heavy chain according to EU numbering in the second heavy chain is L, the antibody according to item 73. 75. The antibody according to any of the preceding items that does not bind to human PD-L2. 76. When measured as described in Example 8 herein, about 10 -8 M or less, for example, about 10 -9 M or less, for example, about 10 -10 M or less of K DAn antibody according to any of the preceding items that binds to human PD-L1. 77. An antibody according to any of the preceding items that mediates concentration-dependent cytotoxicity of MDA-MB-231 cells, PC-3 cells, and / or HELA cells when assayed as described in Example 11 herein, using purified T cells as effector cells. 78. (i) A nucleic acid sequence encoding a heavy chain sequence of an antibody comprising an antigen-binding region capable of binding to human PD-L1 as defined in any of items 1 to 31, and / or (ii) A nucleic acid sequence encoding a light chain sequence of an antibody comprising an antigen-binding region capable of binding to human PD-L1 as defined in any of items 1 to 31 A nucleic acid construct comprising. 79. (i) A nucleic acid sequence encoding a heavy chain sequence of an antibody comprising an antigen-binding region capable of binding to human CD3ε as defined in any of items 33 to 38, and (ii) A nucleic acid sequence encoding a light chain sequence of an antibody comprising an antigen-binding region capable of binding to human CD3ε as defined in any of items 33 to 38 The nucleic acid construct according to item 73, further comprising. 80. An expression vector comprising the nucleic acid construct defined in item 78 or 79. 81. A host cell comprising the nucleic acid construct defined in item 78 or 79 or the expression vector defined in item 80. 82. The host cell according to item 81, which is a mammalian cell such as Chinese hamster ovary cells. 83. A pharmaceutical composition comprising an antibody according to any of items 1 to 77 and a pharmaceutically acceptable carrier. 84. An antibody according to any of items 1 to 77 or a pharmaceutical composition according to item 83 for use as a medicament. 85. An antibody according to any of items 1 to 80 or a pharmaceutical composition according to item 70 for use in the treatment of cancer. 86. An antibody according to any one of items 1 to 77 or a pharmaceutical composition according to item 70 for use in the treatment of a cancer disease characterized by the presence of a solid tumor. 87. An antibody according to any one of items 1 to 78 or a pharmaceutical composition according to item 70 for use in the treatment of a cancer disease selected from the group consisting of melanoma, ovarian cancer, lung cancer, colon cancer, and head and neck cancer. 88. A method of treating a disease comprising the step of administering to a subject in need thereof an antibody according to any one of items 1 to 75 or a pharmaceutical composition according to item 83. 89. Use of an antibody according to any one of items 1 to 77 for the manufacture of a medicament, such as a medicament for the treatment of a cancer disease characterized by the presence of a solid tumor or a cancer disease selected from the group consisting of melanoma, ovarian cancer, lung cancer, colon cancer, and head and neck cancer. 90. A method or use according to any one of items 83 to 89, comprising a combination with one or more additional therapeutic substances, for example, a combination with a chemotherapeutic agent. 91. (a) Culturing a host cell that produces a first antibody comprising an antigen-binding region capable of binding to human PD-L1 as defined in any one of items 1 to 13, and purifying the first antibody from the culture; (b) Culturing a host cell that produces a second antibody comprising an antigen-binding region capable of binding to a different PD-L1 epitope or a different antigen, for example, a human CD3ε-binding region as defined in any one of items 14 to 19, and purifying the second antibody from the culture; (c) Incubating the first antibody together with the second antibody under reducing conditions sufficient to allow the cysteine in the hinge region to undergo disulfide bond isomerization; and (d) Obtaining a bispecific antibody A method for producing an antibody according to any one of items 1 to 77, comprising. 92. An anti-idiotype antibody that binds to an antigen-binding region capable of binding to human PD-L1 as defined in any one of items 1 to 77.

[0320] The present invention is further illustrated by the following examples. The following examples should not be construed as limiting the scope of the present invention.

Example

[0321] Example 1: Preparation of a PD-L1 antibody Immunization of OmniRat animals and hybridoma production Immunization and hybridoma production were performed at Aldevron GmbH (Freiburg, Germany). The cDNA encoding amino acids 19 - 238 of human PD-L1 was cloned into an expression plasmid with intellectual property rights held by Aldevron. OmniRat animal groups (transgenic rats expressing a diverse antibody repertoire with fully human idiotypes; Ligand Pharmaceuticals Inc., San Diego, USA) were immunized by intradermal application of DNA-coated gold particles using a portable device for particle bombardment ("gene gun"). Cell surface expression on transiently transfected HEK cells was confirmed using an anti-PD-L1 antibody based on Genentech's MPDL3280A. Serum samples were collected after a series of immunizations and tested by flow cytometry against HEK cells transiently transfected with the aforementioned expression plasmid. Antibody-producing cells were isolated and fused with mouse myeloma cells (Ag8) according to standard procedures. Hybridomas producing PD-L1-specific antibodies were identified by screening using the same assay method as described above. Cell pellets of positive hybridoma cells were prepared using an RNA protectant (RNAlater, ThermoFisher Scientific, catalog number AM7020) and further processed to sequence the variable domains of the antibody.

[0322] Sequence analysis of the PD-L1 antibody variable domain and cloning into expression vectors 0.2 to 5×10 6Total RNA was prepared from individual hybridoma cells, and 5'-RACE-complementary DNA (cDNA) was prepared from the total RNA using the SMART RACE cDNA Amplification kit (Clontech) according to the manufacturer's instructions. The VH coding region and the VL coding region were PCR amplified and cloned directly in-frame into the pOMTG1f-FEAR-LIC (human IgG1) and pEFC33D-κ (human κ) or pOMTL-LIC (human λ) expression vectors by ligation-independent cloning (Aslanidis, C. and P.J. de Jong, Nucleic Acids Res 1990;18(20): 6069-74). In these plasmids, the antibody sequences are expressed using the CMV promoter. For each antibody, 8 VL clones and 8 VH clones were sequenced. The CDR sequences were defined according to the IMGT definition [Lefranc MP. et al., Nucleic Acids Research, 27, 209-212, 1999; Brochet X. Nucl. Acids Res. 36, W503-508 (2008)]. Clones with the correct open reading frame (ORF) were selected for further study and expression. All combinations of heavy and light chain LEE PCR products found for each hybridoma culture were transiently co-expressed in Expi293F cells using ExpiFectamine. For each hybridoma, the HC / LC pair that showed the best binding in a homogeneous dose-response screen was selected as a lead candidate.

[0323] For further experiments, three PD-L1 antibodies numbered 338, 511, and 547 were selected. Their variable region sequences are shown in the Sequence Listing of this specification.

[0324] For the antibody IgG1-PDL1-511-FEAR, a variant: IgG1-PDL1-511-FEAR-LC33S with point mutations in the variable domain was generated to remove cysteine residues that could potentially give rise to unwanted disulfide bridges. This variant was generated by gene synthesis (Geneart).

[0325] LEE PCR A linear expression element (LEE) was generated by amplifying a fragment containing the CMV promoter and a polyA signal containing elements derived from the HC coding region or LC coding region and the expression plasmid. For this purpose, Accuprime Taq DNA polymerase (Life Technologies) and primers CMVPf(BsaI)2 and TkpA(BsaI)r were used, and 35 cycles of 45 seconds at 94 °C, 30 seconds at 55 °C, and 2 minutes (LC) or 3 minutes (HC) at 68 °C were performed, and the above region was amplified using a 50× diluted plasmid miniprep material as the DNA template.

[0326] Transient expression of the LEE fragment in Expi293F cells For the LEE expression of the Ab, 1.11 μl of the HC LEE PCR reaction mixture and 1.11 μl of the LC PCR reaction mixture were mixed, and in Expi293F cells, using ExpiFectamin 293 as the transfection reagent at a total volume of 125 μl, transfection was performed according to the manufacturer's instructions (Thermo Fisher Scientific, USA) using a 96-well plate as the container.

[0327] Antibody expression The antibody was expressed as IgG1,κ (in the case of 338) or IgG1,λ (in the cases of 511 and 547). A plasmid DNA mixture encoding both the heavy and light chains of the antibody was transiently expressed essentially as described by the manufacturer using the Expi293F expression platform (Thermo Fisher Scientific, USA).

[0328] Homogeneous binding assay method Antibodies were tested for binding in a homogeneous dose-response screening using CHO cells transfected with PDL1, PDL1mm, or PDL1Mf (see also Example 2). Un-transfected CHO cells were used as a negative control.

[0329] Cells (2.5×10 5 cells / ml) were mixed with goat anti-human IgG Alexa647, Fcγ fragment specific (0.2 μg / ml; Jackson ImmunoResearch Laboratories, 109-605-098). Serial dilutions of the test and control antibodies (in the range of 0.001 - 3 μg / mL in 2-fold dilution steps) were prepared and 2 μl of the antibody dilutions were added to 5 μl of the cell / conjugate mixture in a 1536-well plate (Greiner, 789866). The plate was incubated at room temperature for 9 hours, after which the fluorescence intensity was measured using an ImageXpress Velos Laser Scanning Cytometer (Molecular Devices).

[0330] Purification of antibodies The culture supernatant was filtered through a 0.2 μm dead-end filter, loaded onto a 5 mL MabSelect SuRe column (GE Healthcare), and eluted with 0.1 M sodium citrate - NaOH, pH 3. The eluate was immediately neutralized with 2 M Tris-HCl, pH 9, and 8.7 mM Na 2 HPO 4 、1.8 mM NaH 2 PO 4, dialyzed overnight against 140.3 mM NaCl, pH 7.4 (B. Braun or GE Healthcare). Alternatively, after purification, the eluate was loaded onto a HiPrep Desalting column and the antibody was exchanged with 8.7 mM Na 2 HPO 4 , 1.8 mM NaH 2 PO 4 , 140.3 mM NaCl, pH 7.4 (B. Braun or GE Healthcare) buffer. After dialysis or buffer exchange, the sample was sterile filtered through a 0.2 μm dead-end filter. Purity was determined by CE-SDS using a LabChip GXII (Caliper Life Sciences, MA), and IgG concentration was measured using a Nanodrop ND-1000 spectrophotometer (Isogen Life Science, Maarssen, The Netherlands). The purified antibody was stored at 4°C.

[0331] Example 2: Preparation of Screening Material Expression construct for PD-L1 The following codon-optimized constructs were made to express full-length PD-L1: Homo sapiens PD-L1 (Genbank accession number NP_054862), Macaca fascicularis PD-L1 (Genbank accession number XP_005581836), Mus musculus PD-L1 (Genbank accession number NP_068693).

[0332] Furthermore, for PD-L1 ECD, the following codon-optimized construct was made: the extracellular domain (ECD) (aa1-238) of human PD-L1 with a C-terminal His tag and a C-tag (PDLoneECDHisC tag).

[0333] The construct contained restriction sites suitable for cloning and an optimal Kozak (GCCGCCACC) sequence [Kozak et al. (1999) Gene 234: 187-208]. The construct was cloned into the mammalian expression vector pMA (Geneart).

[0334] Expression construct for PD-L2 Similarly, the following codon-optimized construct: human PD-L2 (Genbank accession number NP_079515) was generated to express full-length human PD-L2.

[0335] Expression in CHO-S cells CHO-S cells were transiently transfected with a pMA vector containing the coding sequence of full human PD-L1, a pMA vector containing the coding sequence of cynomolgus monkey, and a pMA vector containing the coding sequence of mouse.

[0336] Purification of His-tagged PD-L1 PDLoneECDHisC tag was expressed in HEK-293F cells. The presence of the His tag enables purification by immobilized metal affinity chromatography. In this process, the chelating agent immobilized on the chromatography resin is Co 2+ charged by cations. The supernatant containing the His-tagged protein is incubated with the resin in batch format (i.e., in solution). The His-tagged protein binds strongly to the resin beads, while other proteins present in the culture supernatant either do not bind or bind weakly compared to the His-tagged protein. After incubation, the beads are recovered from the supernatant and packed into a column. The column is washed to remove weakly bound proteins. Then, the strongly bound His-tagged protein is eluted using a buffer containing imidazole that competes with the binding of His and Co 2+ and. The eluent is removed from the protein by buffer exchange on a desalting column.

[0337] Example 3: Humanized CD3 Antibody for Producing CD3×PDL1 Bispecific Antibody The production of the humanized antibody IgG1-huCD3-H1L1 is described in Example 1 of WO2015001085. The antibody huCD3-H1L1-FEAL is a variant thereof having the following substitutions: L234F, L235E, D265A, and F405L, as described above herein.

[0338] Example 4: Production of Bispecific Antibody by 2-MEA-Induced Fab Arm Exchange The bispecific IgG1 antibody was produced by Fab arm exchange under controlled reducing conditions. The basis of this method is the use of complementary CH3 domains that promote heterodimer formation under specific assay conditions as described in WO2011 / 131746. To generate antibody pairs with complementary CH3 domains, the F405L and K409R (EU numbering) mutations were introduced into the relevant antibodies.

[0339] To produce the bispecific antibody, two parental complementary antibodies with a final concentration of 0.5 mg / mL each were incubated with 75 mM 2-mercaptoethylamine-HCl (2-MEA) at 31 °C for 5 hours in a total volume of 100 μL of TE. The reduction reaction was stopped by removing the reducing agent 2-MEA using a spin column (Microcon centrifugal filter, 30k, Millipore) according to the manufacturer's protocol.

[0340] The following antibodies were used in the examples.

[0341] CD3 antibody IgG1-huCD3-H1L1-FEAL (having the VH sequence shown in SEQ ID NO:25 and the VL sequence shown in SEQ ID NO:29) bsIgG1-huCD3-H1L1-FEALxb12-FEAR is a bispecific antibody that uses antibody b12 (Barbas, CF. J Mol Biol. 1993 Apr 5;230(3):812-23), a gp120-specific antibody, as the second arm.

[0342] PDL1 antibody and CD3×PDL1 bispecific antibody IgG1-338-FEAR (having the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:5) IgG1-338-F405L bsIgG1-huCD3-H1L1-FEALx338-FEAR bsIgG1-b12-FEALx338-FEAR

[0343] IgG1-511-LC33S-FEAR (having the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:15) IgG1-511-F405L-LC33S bsIgG1-huCD3-H1L1-FEALx511-LC33S-FEAR bsIgG1-b12-FEALx511-LC33S-FEAR

[0344] IgG1-547-FEAR (having the VH sequence shown in SEQ ID NO:18 and the VL sequence shown in SEQ ID NO:22) IgG1-547-F405L bsIgG1-huCD3-H1L1-FEALx547-FEAR bsIgG1-b12-FEALx547-FEAR

[0345] IgG1-321-FEAR (having the VH sequence shown in SEQ ID NO:32 and the VL sequence shown in SEQ ID NO:36)

[0346] IgG1-421-LC91S-FEAR (having the VH sequence shown in SEQ ID NO:46 and the VL sequence shown in SEQ ID NO:50)

[0347] IgG1-476-N101Q-LC33S-FEAR (having the VH sequence shown in SEQ ID NO:53 and the VL sequence shown in SEQ ID NO:57)

[0348] IgG1-625-FEAR (having the VH sequence shown in SEQ ID NO:106 and the VL sequence shown in SEQ ID NO:110)

[0349] IgG1-632-FEAR (having the VH sequence shown in SEQ ID NO:67 and the VL sequence shown in SEQ ID NO:71) IgG1-516-FEAR (having the VH sequence shown in SEQ ID NO:60 and the VL sequence shown in SEQ ID NO:64)

[0350] IgG1-MPDL3280A-FEAR (based on Genentech's PDL1 antibody MPDL3280A; having the VH sequence shown in SEQ ID NO:74 and the VL sequence shown in SEQ ID NO:78) IgG1-MPDL3280A-K409R

[0351] IgG1-MEDI4736-FEAR (based on MedImmune's PDL1 antibody MEDI4736; having the VH sequence shown in SEQ ID NO:81 and the VL sequence shown in SEQ ID NO:85) IgG1-MEDI4736-F405L

[0352] Example 5: Binding of a PD-L1 antibody or a CD3×PD-L1 bispecific antibody or a b12×PD-L1 bispecific antibody to tumor cells The binding of the PD-L1 antibody, as well as the CD3×PD-L1 bispecific antibody and the b12×PD-L1 bispecific antibody, to human tumor cell lines SK-MES-1 (lung squamous cell carcinoma; ATCC; catalog number HTB-58), MDA-MB-231 (breast adenocarcinoma; ATCC; catalog number HTB-26), PC-3 (prostate adenocarcinoma; ATCC; catalog number CRL-1435) and HELA (cervical adenocarcinoma; ATCC; catalog number CCL-2) was analyzed by flow cytometry.

[0353] Cells (3 - 5×10 4 cells / well) were incubated with serial dilutions of the antibody (in the range of 0.0001 - 10 μg / mL in 5-fold dilution steps) dissolved in 50 μL of PBS / 0.1% BSA / 0.02% azide (staining buffer) in a polystyrene 96-well round-bottom plate (Greiner bio-one, catalog number 650101) at 4°C for 30 minutes.

[0354] After washing twice with the staining buffer, the cells were incubated with 50 μL of the secondary antibody at 4°C for 30 minutes. As the secondary antibody, R-phycoerythrin (PE)-conjugated goat-anti-human IgG F(ab’) 2 (catalog number 109-116-098, Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) was used in all experiments. Next, the cells were washed twice with the staining buffer, resuspended in 20 μL of the staining buffer, and analyzed using an iQue screener (Intellicyt Corporation, USA). The binding curves were analyzed using GraphPad Prism V75.04 software (GraphPad Software, San Diego, CA, USA) with non-linear regression (sigmoid dose-response with changing slope).

[0355] To quantify target expression on the plasma membranes of MDA-MB-231 cells, PC-3 cells, and HELA cells, and to determine the number of bound PDL1 molecules, quantitative flow cytometry (QIFIKIT®, Dako; catalog number K0078) was performed as described (Poncelet and Carayon, 1985, J. Immunol. Meth. 85: 65-74). The following PD-L1 antigen densities (ABC, antibody binding capacity) were determined for the cell lines: · SK-MES-1: approximately 30,000 ABC / cell, · MDA-MB-231: approximately 21,000 ABC / cell, · PC-3: approximately 6,000 ABC / cell, · HELA cells: approximately 2,000 ABC / cell were determined to have.

[0356] Binding to MDA-MB-231 cells Figure 1 shows that bsIgG1-huCD3-H1L1-FEALx338-FEAR (A), bsIgG1-b12-FEALx338-FEAR (D), bsIgG1-huCD3-H1L1-FEALx547-FEAR (B), and bsIgG1-b12-FEALx547-FEAR (E) showed dose-dependent binding with higher maximum binding to MDA-MB-231 cells than the monospecific bivalent PD-L1 antibodies IgG1-338-FEAR and IgG1-547-FEAR. The maximum binding of bsIgG1-huCD3-H1L1-FEALx511-LC33S-FEAR (C) and bsIgG1-b12-FEALx511-LC33S-FEAR (F) was less than that of the bivalent monospecific PD-L1 antibody IgG1-511-LC33S-FEAR.

[0357] Binding to PC-3 cells Figure 2 shows that bsIgG1-huCD3-H1L1-FEALx338-FEAR (A), bsIgG1-b12-FEALx338-FEAR (D), bsIgG1-huCD3-H1L...

Claims

1. 1. An antibody comprising an antigen-binding region capable of binding to human PD-L1, Inhibits the binding of human PD-L1 to human PD-1, and (i) competes for binding to human PD-L1 with antibody [511] comprising the VH sequence set forth in SEQ ID NO:8 and the VL sequence set forth in SEQ ID NO:15, but does not compete for binding to human PD-L1 with antibody [547] comprising the VH sequence set forth in SEQ ID NO:18 and the VL sequence set forth in SEQ ID NO:22; or (ii) competes for binding to human PD-L1 with antibody [547] comprising the VH sequence set forth in SEQ ID NO:18 and the VL sequence set forth in SEQ ID NO:22, but does not compete for binding to human PD-L1 with antibody [511] comprising the VH sequence set forth in SEQ ID NO:8 and the VL sequence set forth in SEQ ID NO:15; The antibody.

2. The antibody of claim 1, which competes for binding to human PD-L1 with an antibody [338] comprising the VH sequence shown in SEQ ID NO:1 and the VL sequence shown in SEQ ID NO:

5.

3. The antibody of any one of the preceding claims, wherein the binding of the antibody of any one of the preceding claims to human PD-L1 is not displaced by an antibody comprising the VH sequence set forth in SEQ ID NO:53 and the VL sequence set forth in SEQ ID NO:57 [476].

4. The antibody of any one of the preceding claims, wherein the binding of the antibody of any one of the preceding claims to human PD-L1 is not blocked by binding of antibody [625] comprising the VH sequence set forth in SEQ ID NO:106 and the VL sequence set forth in SEQ ID NO:

110.

5. The antibody of any one of the preceding claims, wherein the binding of the antibody of any one of the preceding claims to human PD-L1 is blocked by antibody [547] comprising the VH sequence set forth in SEQ ID NO:18 and the VL sequence set forth in SEQ ID NO:

22.

6. (i) is capable of binding to the same epitope on human PD-L1 as antibody [338] comprising the VH sequence set forth in SEQ ID NO:1 and the VL sequence set forth in SEQ ID NO:5; or (ii) is capable of binding to the same epitope on human PD-L1 as antibody [511] comprising the VH sequence set forth in SEQ ID NO:8 and the VL sequence set forth in SEQ ID NO:15; or (iii) is capable of binding to the same human PD-L1 epitope as antibody [547] comprising the VH sequence set forth in SEQ ID NO:18 and the VL sequence set forth in SEQ ID NO:22; The antibody of claim 1.

7. 7. The antibody of any one of claims 1 to 4 and 6, wherein the binding of the antibody to a mutant PD-L1 having one or more of the amino acid residues at a position corresponding to position 113 (R113), 123 (Y123), and 125 (R125) in SEQ ID NO:94 substituted with alanine compared to binding to wild-type PD-L1 having the amino acid sequence as set forth in SEQ ID NO:94, wherein reduced binding is determined when the fold change in binding of the antibody is less than the average of the fold change in binding across all alanine mutants - 1.5 x SD, where SD is the standard deviation of all calculated fold change values ​​of the antibody to mutant PDL1, and the fold change in binding is calculated as set forth in Example 13[338].

8. The antibody of any one of claims 1-2 and 6, wherein the antibody binds to an epitope on PD-L1 (SEQ ID NO:94), the epitope comprising the amino acid residue at position 113 (R113), the amino acid residue at position 123 (Y123), and / or the amino acid residue at position 125 (R125) of SEQ ID NO:

94.

9. 94。 [511] The antibody of any one of claims 1, 3, 4 and 6, wherein the antibody has reduced binding to a mutant PD-L1 having an alanine substitution at one or more of the amino acid residues corresponding to position 19 (F19), 42 (F42), 45 (E45), 46 (K46), 94 (L94), and 116 (I116) in SEQ ID NO:94, compared to wild-type PD-L1 having the amino acid sequence set forth in SEQ ID NO:94, wherein reduced binding is determined when the fold change in binding of the antibody is less than the average of the fold change in binding across all alanine mutants - 1.5 × SD, where SD is the standard deviation of all calculated fold change values ​​of the antibody to mutant PDL1, and the fold change in binding is calculated as set forth in Example 13.

10. The antibody of any one of claims 1, 3, 4, and 6, wherein the antibody binds to an epitope on PD-L1 (SEQ ID NO:94), wherein the epitope comprises one or more amino acid residues selected from the group consisting of the amino acid residue at position 45 (E45), the amino acid residue at position 46 (K46), and / or the amino acid residue at position 94 (L94) of SEQ ID NO:

94.

11. 7. The antibody of any one of claims 1, 5 and 6, wherein the binding of the antibody to a mutant PD-L1 comprising one or more of the amino acid residues at a position corresponding to position 58 in SEQ ID NO:94 (E58) and the amino acid residue at a position corresponding to position 113 in SEQ ID NO:94 (R113) substituted with alanine compared to wild-type PD-L1 having the amino acid sequence set forth in SEQ ID NO:94, wherein reduced binding is determined when the fold change in binding of the antibody is less than the average of the fold change in binding across all alanine mutants - 1.5 x SD, where SD is the standard deviation of all calculated fold change values ​​of the antibody to mutant PDL1, and where fold change in binding is calculated as set forth in Example 13[547].

12. The antibody of any one of claims 1, 5, and 6, wherein the antibody binds to an epitope on PD-L1 (SEQ ID NO:94), the epitope comprising the amino acid residue at position 58 (E58) and / or the amino acid residue at position 113 (R113) of SEQ ID NO:

94.

13. The antibody of any one of the preceding claims, wherein the antigen-binding region capable of binding to human PD-L1 comprises a heavy chain variable region (VH) comprising a CDR1, CDR2, and CDR3 sequence, and a light chain variable region (VL) comprising a CDR1, CDR2, and CDR3 sequence, and the VH CDR3 sequence is selected from the group consisting of the sequences set forth in SEQ ID NO:4, SEQ ID NO:11, and SEQ ID NO:

21.

14. the antigen-binding region capable of binding to human PD-L1 being (i) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:2, the CDR2 sequence shown in SEQ ID NO:3, and the CDR3 sequence shown in SEQ ID NO:4, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:6, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:7 [338]; or (ii) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:9, the CDR2 sequence shown in SEQ ID NO:10, and the CDR3 sequence shown in SEQ ID NO:11, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:16, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:17 [511]; or (iii) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:19, the CDR2 sequence shown in SEQ ID NO:20, and the CDR3 sequence shown in SEQ ID NO:21, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:23, a CDR2 having the sequence DDN, and a CDR3 having the sequence shown in SEQ ID NO:24 [547] The antibody of any one of the preceding claims, comprising:

15. The antibody of any one of the preceding claims, wherein the antigen-binding region capable of binding to human PD-L1 comprises a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to a VH sequence selected from the group consisting of the sequences set forth in SEQ ID NO:1, SEQ ID NO:8, and SEQ ID NO:

18.

16. The antibody of any one of the preceding claims, wherein the antigen-binding region capable of binding to human PD-L1 comprises a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to a VL sequence selected from the group consisting of the sequences set forth in SEQ ID NO:5, SEQ ID NO:15, and SEQ ID NO:

22.

17. the antigen-binding region capable of binding to human PD-L1 being (i) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to the VH sequence set forth in SEQ ID NO:1, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to the VL sequence set forth in SEQ ID NO:5 [338]; or (ii) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to the VH sequence set forth in SEQ ID NO:8, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to the VL sequence set forth in SEQ ID NO:15 [511]; or (iii) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to the VH sequence set forth in SEQ ID NO: 18, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to the VL sequence set forth in SEQ ID NO: 22 [547] The antibody of any one of the preceding claims, comprising:

18. 18. The antibody of claim 17, wherein the VH sequence and the VL sequence each comprise three CDR sequences, CDR1, CDR2, and CDR3, respectively, and four framework sequences, FR1, FR2, FR3, and FR4, respectively, and each combined FR1, FR2, FR3, and FR4 framework sequence of the VH has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with each combined FR1, FR2, FR3, and FR4 framework sequence of the VH sequence, and the VH CDR sequence is not mutated, and each combined FR1, FR2, FR3, and FR4 framework sequence of the VL has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with each combined FR1, FR2, FR3, and FR4 framework sequence of the VL sequence, and the VL CDR sequence is not mutated.

19. 2. The antibody of any one of the preceding claims, capable of inducing dose-dependent lysis of adenocarcinoma epithelial cells, such as dose-dependent lysis of MDA-MB-231, via antibody-dependent cellular cytotoxicity (ADCC).

20. 20. The antibody of claim 19, which is capable of reducing cell number in a culture of said epithelial cells by at least 5%, such as at least 6%, 7%, 8%, 9%, or at least 10% as a result of cell lysis.

21. ADCC, 51 21. The antibody of claim 19 or 20, measured in vitro in a Cr release assay, e.g., the assay disclosed in Example 14.

22. ADCC is performed by incubating the epithelial cells with a composition comprising an antibody of claim 19 or 20 and an effector cell, e.g., a peripheral blood mononuclear cell (PBMC), at 37° C., 5% CO. 2 21. The antibody of claim 19 or 20, wherein the amount of antibody in the composition is in the range of 0.1-1 μg / mL and the ratio of effector cells to epithelial cells is 100:1 as measured in vitro by incubating with 500 mM NaCl for 4 hours.

23. 21. The antibody of claim 19 or 20, wherein the lysis of epithelial cells is measured in vitro in a luciferase reporter assay as a surrogate for ADCC, such as a luminescence ADCC reporter bioassay as disclosed in Example 14.

24. ADCC, (i) contacting the culture of epithelial cells with a composition comprising the antibody of claim 23 and Jurkat human T cells that stably express FcγRIIIa (CD16) and firefly luciferase (effector cells) at a 1:1 effector cell:epithelial cell ratio; (ii) allowing the epithelial cell culture and the effector cells to equilibrate to room temperature for 15 minutes; (iii) incubating the culture of epithelial cells and the effector cells with a luciferase substrate; and (iv) measuring luciferase production in the cell culture It is measured in vitro by 24. The antibody of claim 23, wherein the amount of antibody in the composition is in the range of 0.5 to 250 ng / mL and the ratio of effector cells to epithelial cells is 1:

1.

25. The ADCC of the epithelial cells is measured in a luciferase reporter assay, such as a reporter assay as defined in claim 23 or 24, and then the ADCC observed after incubating a culture of the epithelial cells with a test composition comprising an antibody according to any one of claims 19, 20, 23 and 24 is at least 1.5 times the ADCC observed after incubating a culture of the epithelial cells with a composition comprising a reference antibody, the ADCC being measured as relative light units (RLU), the antibody concentrations in the test composition and in the composition comprising a reference antibody are the same and are in the range of 20-250 ng / ml, and the reference antibody is (a) an antibody comprising the VH sequence set forth in SEQ ID NO:74 and the VL sequence set forth in SEQ ID NO:78; and (b) an antibody comprising the VH sequence shown in SEQ ID NO:81 and the VL sequence shown in SEQ ID NO:85 25. The antibody of any one of claims 19, 20, 23, and 24, selected from the group consisting of:

26. the antigen-binding region capable of binding to human PD-L1 being (i) the VH sequence set forth in SEQ ID NO:1 and the VL sequence set forth in SEQ ID NO:5 [338]; or (ii) the VH sequence set forth in SEQ ID NO:8 and the VL sequence set forth in SEQ ID NO:15 [511]; or (iii) the VH sequence shown in SEQ ID NO: 18 and the VL sequence shown in SEQ ID NO: 22 [547] The antibody of any one of the preceding claims, comprising:

27. 1. An antibody comprising an antigen-binding region capable of binding to human PD-L1, (i) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:33, the CDR2 sequence shown in SEQ ID NO:34, and the CDR3 sequence shown in SEQ ID NO:35, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:37, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:38 [321]; or (ii) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:47, the CDR2 sequence shown in SEQ ID NO:48, and the CDR3 sequence shown in SEQ ID NO:49, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:51, a CDR2 having the sequence DVI, and a CDR3 having the sequence shown in SEQ ID NO:52 [421]; or (iii) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:54, the CDR2 sequence shown in SEQ ID NO:55, and the CDR3 sequence shown in SEQ ID NO:56, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:58, a CDR2 having the sequence RDS, and a CDR3 having the sequence shown in SEQ ID NO:59 [476]; or (iv) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:61, the CDR2 sequence shown in SEQ ID NO:62, and the CDR3 sequence shown in SEQ ID NO:63, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:65, a CDR2 having the sequence DDS, and a CDR3 having the sequence shown in SEQ ID NO:66 [516]; or (v) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO: 107, the CDR2 sequence shown in SEQ ID NO: 108, and the CDR3 sequence shown in SEQ ID NO: 109, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO: 111, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO: 113 [625]; (vi) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:68, the CDR2 sequence shown in SEQ ID NO:69, and the CDR3 sequence shown in SEQ ID NO:70, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:72, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:73 [632] The antibody comprising:

28. (i) the VH sequence set forth in SEQ ID NO: 32 and the VL sequence set forth in SEQ ID NO: 36 [321]; or (ii) the VH sequence set forth in SEQ ID NO:46 and the VL sequence set forth in SEQ ID NO:50 [421]; or (iii) the VH sequence set forth in SEQ ID NO:53 and the VL sequence set forth in SEQ ID NO:57 [476]; or (iv) the VH sequence set forth in SEQ ID NO:60 and the VL sequence set forth in SEQ ID NO:64 [516]; or the VH sequence shown in SEQ ID NO: 106 and the VL sequence shown in SEQ ID NO: 110 [625]; (v) the VH sequence shown in SEQ ID NO: 67 and the VL sequence shown in SEQ ID NO: 71 [632] 28. The antibody of claim 27, comprising:

29. The antibody of any one of the preceding claims, which is monovalent.

30. The antibody of any one of the preceding claims, wherein the antibody is a bivalent antibody having two antigen-binding regions capable of binding to human PD-L1, and the two antigen-binding regions have identical variable region sequences.

31. The antibody of any one of the preceding claims, wherein the antibody is a bivalent, bispecific antibody and comprises, in addition to the (first) antigen-binding region capable of binding to human PD-L1, a (second) antigen-binding region capable of binding to a second antigen or to a different human PD-L1 epitope, wherein the second antigen is not human CD3ε.

32. A bispecific antibody comprising an antigen-binding region capable of binding to human PD-L1 and an antigen-binding region capable of binding to human CD3ε (epsilon), The bispecific antibody, wherein the antigen-binding region capable of binding to human PD-L1 has the characteristics as defined in any one of the preceding claims.

33. 33. The bispecific antibody of claim 32, wherein said antigen-binding region capable of binding to human CD3ε comprises: (a) a heavy chain variable region (VH) comprising CDR1 having the sequence shown in SEQ ID NO:26, CDR2 having the sequence shown in SEQ ID NO:27, and CDR3 having the sequence shown in SEQ ID NO:28, and a light chain variable region (VL) comprising CDR1 having the sequence shown in SEQ ID NO:30, CDR2 having the sequence GTN, and CDR3 having the sequence shown in SEQ ID NO:

31.

34. (i) an antigen-binding region capable of binding to human PD-L1 [338], comprising a heavy chain variable region (VH) comprising the CDR1 sequence set forth in SEQ ID NO:2, the CDR2 sequence set forth in SEQ ID NO:3, and the CDR3 sequence set forth in SEQ ID NO:4, and a light chain variable region (VL) comprising a CDR1 having the sequence set forth in SEQ ID NO:6, a CDR2 having the sequence KAS, and a CDR3 having the sequence set forth in SEQ ID NO:7; and (a) an antigen-binding region capable of binding to human CD3ε, comprising a heavy chain variable region (VH) comprising a CDR1 having the sequence set forth in SEQ ID NO:26, a CDR2 having the sequence set forth in SEQ ID NO:27, and a CDR3 having the sequence set forth in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence set forth in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence set forth in SEQ ID NO:31; or (ii) an antigen-binding region capable of binding to human PD-L1 [338], comprising a heavy chain variable region (VH) comprising the CDR1 sequence set forth in SEQ ID NO:9, the CDR2 sequence set forth in SEQ ID NO:10, and the CDR3 sequence set forth in SEQ ID NO:11, and a light chain variable region (VL) comprising a CDR1 having the sequence set forth in SEQ ID NO:16, a CDR2 having the sequence EDS, and a CDR3 having the sequence set forth in SEQ ID NO:17; and (a) an antigen-binding region capable of binding to human CD3ε, comprising a heavy chain variable region (VH) comprising a CDR1 having the sequence set forth in SEQ ID NO:26, a CDR2 having the sequence set forth in SEQ ID NO:27, and a CDR3 having the sequence set forth in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence set forth in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence set forth in SEQ ID NO:31; or (iii) an antigen-binding region capable of binding to human PD-L1 [547], comprising a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:19, the CDR2 sequence shown in SEQ ID NO:20, and the CDR3 sequence shown in SEQ ID NO:21, and a light chain variable region (VL) comprising the CDR1 sequence shown in SEQ ID NO:23, the CDR2 sequence shown in SEQ ID NO:27, and the CDR3 sequence shown in SEQ ID NO:28; and (a) an antigen-binding region capable of binding to human CD3ε, comprising a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:26, the CDR2 sequence shown in SEQ ID NO:27, and the CDR3 sequence shown in SEQ ID NO:28, and a light chain variable region (VL) comprising the CDR1 sequence shown in SEQ ID NO:30, the CDR2 having the sequence shown in SEQ ID NO:

31.

34. The bispecific antibody of claim 32 or 33, comprising:

35. 35. The bispecific antibody of any one of claims 32 to 34, wherein said antigen-binding region capable of binding to human CD3ε comprises the VH sequence shown in SEQ ID NO:25 and the VL sequence shown in SEQ ID NO:

29.

36. (i) said bispecific antibody has a lower affinity for human CD3ε binding compared to an antibody having an antigen-binding region that can comprise the VH sequence shown in SEQ ID NO:25 and the VL sequence shown in SEQ ID NO:29, preferably the affinity is at least 2-fold, such as at least 5-fold, such as at least 10-fold, such as at least 25-fold, such as at least 50-fold lower; and (ii) the bispecific antibody is capable of mediating concentration-dependent cytotoxicity of MDA-MB-231 cells, PC-3 cells, and / or HELA cells using PBMCs or purified T cells as effector cells, e.g., when assayed as described in Example 11 herein; 33. The bispecific antibody of claim 32.

37. the antigen-binding region capable of binding to human CD3ε is (i) a heavy chain variable region (VH) comprising a CDR1 having the sequence set forth in SEQ ID NO:99, a CDR2 having the sequence set forth in SEQ ID NO:27, and a CDR3 having the sequence set forth in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence set forth in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence set forth in SEQ ID NO:31; or (ii) a heavy chain variable region (VH) comprising a CDR1 having the sequence set forth in SEQ ID NO:100, a CDR2 having the sequence set forth in SEQ ID NO:27, and a CDR3 having the sequence set forth in SEQ ID NO:28, and a light chain variable region (VL) comprising a CDR1 having the sequence set forth in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence set forth in SEQ ID NO:31; or (iii) a heavy chain variable region (VH) comprising a CDR1 having the sequence set forth in SEQ ID NO:26, a CDR2 having the sequence set forth in SEQ ID NO:27, and a CDR3 having the sequence set forth in SEQ ID NO:101, and a light chain variable region (VL) comprising a CDR1 having the sequence set forth in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence set forth in SEQ ID NO:31; or (iv) a heavy chain variable region (VH) comprising a CDR1 having the sequence set forth in SEQ ID NO:26, a CDR2 having the sequence set forth in SEQ ID NO:27, and a CDR3 having the sequence set forth in SEQ ID NO:102, and a light chain variable region (VL) comprising a CDR1 having the sequence set forth in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence set forth in SEQ ID NO:31; or (v) a heavy chain variable region (VH) comprising a CDR1 having the sequence set forth in SEQ ID NO:26, a CDR2 having the sequence set forth in SEQ ID NO:27, and a CDR3 having the sequence set forth in SEQ ID NO:103, and a light chain variable region (VL) comprising a CDR1 having the sequence set forth in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence set forth in SEQ ID NO:31; or (vi) a heavy chain variable region (VH) comprising a CDR1 having the sequence set forth in SEQ ID NO:26, a CDR2 having the sequence set forth in SEQ ID NO:27, and a CDR3 having the sequence set forth in SEQ ID NO:104, and a light chain variable region (VL) comprising a CDR1 having the sequence set forth in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence set forth in SEQ ID NO:31; or (vii) a heavy chain variable region (VH) comprising a CDR1 having the sequence set forth in SEQ ID NO:26, a CDR2 having the sequence set forth in SEQ ID NO:27, and a CDR3 having the sequence set forth in SEQ ID NO:105, and a light chain variable region (VL) comprising a CDR1 having the sequence set forth in SEQ ID NO:30, a CDR2 having the sequence GTN, and a CDR3 having the sequence set forth in SEQ ID NO:

31.

37. The bispecific antibody of claim 36, comprising:

38. the antigen-binding region capable of binding to human CD3ε is (i) the VH sequence set forth in SEQ ID NO:39 and the VL sequence set forth in SEQ ID NO:29; or (ii) the VH sequence set forth in SEQ ID NO:40 and the VL sequence set forth in SEQ ID NO:29; or (iii) the VH sequence set forth in SEQ ID NO:41 and the VL sequence set forth in SEQ ID NO:29; or (iv) the VH sequence set forth in SEQ ID NO:42 and the VL sequence set forth in SEQ ID NO:29; or (v) the VH sequence set forth in SEQ ID NO: 43 and the VL sequence set forth in SEQ ID NO: 29; or (vi) the VH sequence set forth in SEQ ID NO:44 and the VL sequence set forth in SEQ ID NO:29; or (vii) the VH sequence shown in SEQ ID NO: 45 and the VL sequence shown in SEQ ID NO: 29 38. The bispecific antibody of claim 36 or 37, comprising:

39. A multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or to a different human PD-L1 epitope, 32. The multispecific antibody, wherein said antigen-binding region capable of binding to human PD-L1 has the characteristics as defined in any one of claims 1 to 31.

40. The multispecific antibody of claim 39, wherein the antigen-binding region capable of binding to human PD-L1 comprises a heavy chain variable region (VH) comprising CDR1, CDR2, and CDR3 sequences, and a light chain variable region (VL) comprising CDR1, CDR2, and CDR3 sequences, and the VH CDR3 sequence is selected from the group consisting of the sequences set forth in SEQ ID NO:4[338], SEQ ID NO:11[511], and SEQ ID NO:21[547].

41. the first antigen-binding region capable of binding to human PD-L1 comprises: (i) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:2, the CDR2 sequence shown in SEQ ID NO:3, and the CDR3 sequence shown in SEQ ID NO:4, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:6, a CDR2 having the sequence KAS, and a CDR3 having the sequence shown in SEQ ID NO:7 [338]; or (ii) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:9, the CDR2 sequence shown in SEQ ID NO:10, and the CDR3 sequence shown in SEQ ID NO:11, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:16, a CDR2 having the sequence EDS, and a CDR3 having the sequence shown in SEQ ID NO:17 [511]; or (iii) a heavy chain variable region (VH) comprising the CDR1 sequence shown in SEQ ID NO:19, the CDR2 sequence shown in SEQ ID NO:20, and the CDR3 sequence shown in SEQ ID NO:21, and a light chain variable region (VL) comprising a CDR1 having the sequence shown in SEQ ID NO:23, a CDR2 having the sequence DDN, and a CDR3 having the sequence shown in SEQ ID NO:24 [547] 41. The multispecific antibody of claim 40, comprising:

42. The multispecific antibody of claim 40 or 41, wherein the first antigen-binding region capable of binding to human PD-L1 comprises a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to a VH sequence selected from the group consisting of the sequences set forth in SEQ ID NO:1[338], SEQ ID NO:8[511], and SEQ ID NO:18[547].

43. The multispecific antibody of any one of claims 40 to 42, wherein the first antigen-binding region capable of binding to human PD-L1 comprises a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to a VL sequence selected from the group consisting of the sequences set forth in SEQ ID NO:5[338], SEQ ID NO:15[511], and SEQ ID NO:22[547].

44. the first antigen-binding region capable of binding to human PD-L1 comprises: (i) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to the VH sequence set forth in SEQ ID NO:1, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to the VL sequence set forth in SEQ ID NO:5 [338]; or (ii) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to the VH sequence set forth in SEQ ID NO:8, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to the VL sequence set forth in SEQ ID NO:15 [511]; or (iii) a VH sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to the VH sequence set forth in SEQ ID NO: 18, and a VL sequence having at least 90%, at least 95%, at least 97%, at least 99%, or 100% amino acid sequence identity to the VL sequence set forth in SEQ ID NO: 22 [547] The multispecific antibody according to any one of claims 40 to 43, comprising:

45. 45. The multispecific antibody of any one of claims 40 to 44, wherein the VH and VL sequences each comprise three CDR sequences, CDR1, CDR2 and CDR3, respectively, and four framework sequences, FR1, FR2, FR3 and FR4, respectively, and wherein each combined FR1, FR2, FR3 and FR4 framework sequence of VH has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with each combined FR1, FR2, FR3 and FR4 framework sequence of the VH sequence, and the VH CDR sequences are not mutated, and wherein each combined FR1, FR2, FR3 and FR4 framework sequence of VL has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with each combined FR1, FR2, FR3 and FR4 framework sequence of the VL sequence, and the VL CDR sequences are not mutated.

46. the first antigen-binding region capable of binding to human PD-L1 comprises: (i) the VH sequence set forth in SEQ ID NO:1 and the VL sequence set forth in SEQ ID NO:5 [338]; or (ii) the VH sequence set forth in SEQ ID NO:8 and the VL sequence set forth in SEQ ID NO:15 [511]; or (iii) the VH sequence shown in SEQ ID NO: 18 and the VL sequence shown in SEQ ID NO: 22 [547] 46. ​​The multispecific antibody of any one of claims 40 to 45, comprising:

47. A multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or to a different human PD-L1 epitope, (i) competes for binding to human PD-L1 with an antibody comprising the VH sequence set forth in SEQ ID NO:8 and the VL sequence set forth in SEQ ID NO:15, but does not compete for binding to human PD-L1 with an antibody comprising the VH sequence set forth in SEQ ID NO:18 and the VL sequence set forth in SEQ ID NO:22; or (ii) competes for binding to human PD-L1 with an antibody comprising the VH sequence set forth in SEQ ID NO:18 and the VL sequence set forth in SEQ ID NO:22, but does not compete for binding to human PD-L1 with an antibody comprising the VH sequence set forth in SEQ ID NO:8 and the VL sequence set forth in SEQ ID NO:15; The multispecific antibody.

48. The multispecific antibody of Claim 47, which competes for binding to human PD-L1 with an antibody comprising the VH sequence set forth in SEQ ID NO:1 and the VL sequence set forth in SEQ ID NO:

5.

49. A multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or to a different human PD-L1 epitope, The multispecific antibody, wherein the binding of the antibody to human PD-L1 is not displaced by an antibody comprising the VH sequence set forth in SEQ ID NO:53 and the VL sequence set forth in SEQ ID NO:

57.

50. 50. The multispecific antibody of claim 49, which inhibits the binding of human PD-L1 to human PD-1.

51. 51. The multispecific antibody of claim 49 or 50, which competes for binding to human PD-L1 with an antibody comprising the VH sequence set forth in SEQ ID NO:18 and the VL sequence set forth in SEQ ID NO:

22.

52. 52. The multispecific antibody of any one of claims 49 to 51, wherein the binding of said multispecific antibody to human PD-L1 is blocked by an antibody comprising the VH sequence as shown in SEQ ID NO:18 and the VL sequence as shown in SEQ ID NO:

22.

53. A multispecific antibody comprising a first antigen-binding region capable of binding to human PD-L1 and a second antigen-binding region capable of binding to a second antigen or to a different human PD-L1 epitope, The first antigen-binding region comprises: (i) is capable of binding to the same epitope on human PD-L1 as antibody [338] comprising the VH sequence set forth in SEQ ID NO:1 and the VL sequence set forth in SEQ ID NO:5; or (ii) is capable of binding to the same epitope on human PD-L1 as antibody [511] comprising the VH sequence set forth in SEQ ID NO:8 and the VL sequence set forth in SEQ ID NO:15; or (iii) is capable of binding to the same human PD-L1 epitope as antibody [547] comprising the VH sequence set forth in SEQ ID NO:18 and the VL sequence set forth in SEQ ID NO:22; The multispecific antibody.

54. 54. The multispecific antibody of any one of claims 40 to 53, which is bispecific.

55. 55. The multispecific antibody of claim 54, which is bivalent.

56. 56. The multispecific antibody of any one of claims 40 to 55, which is capable of binding to a second antigen, and which second antigen is not human CD3ε.

57. The antibody of any one of the preceding claims, which is a full-length antibody.

58. 58. The antibody of claim 57, which is a full-length IgG1 antibody.

59. 2. The antibody of any one of the preceding claims, which is an antibody fragment.

60. The antibody according to any one of claims 32 to 59, comprising two half molecules each comprising an antigen-binding region, (i) the half molecule is chimeric and contains an antigen-binding region capable of binding to human PD-L1; and / or (ii) is chimeric when a half molecule containing an antigen-binding region capable of binding to human CD3ε (epsilon) is present; 60. The antibody of any one of claims 32 to 59.

61. (i) the antigen-binding region capable of binding to human PD-L1 is humanized; and / or (ii) the antigen-binding region capable of binding to human CD3ε (epsilon), if present, is humanized; An antibody according to any one of the preceding claims.

62. (i) the antigen-binding region capable of binding to human PD-L1 is human; and / or (ii) the antigen-binding region capable of binding to human CD3ε (epsilon), when present, is human; An antibody according to any one of the preceding claims.

63. The antibody of any one of the preceding claims, wherein each of the antigen-binding regions comprises a heavy chain variable region (VH) and a light chain variable region (VL), and each of the variable regions comprises three CDR sequences, CDR1, CDR2, and CDR3, respectively, and four framework sequences, FR1, FR2, FR3, and FR4, respectively.

64. 64. The antibody of claim 63, comprising two heavy chain constant regions (CH) and two chain constant regions (CL).

65. 10. The antibody of claim 9, comprising a first heavy chain and a second heavy chain, each of which comprises at least a hinge region, a CH2 region, and a CH3 region, wherein the first heavy chain has at least one amino acid substitution at a position corresponding to a position selected from the group consisting of T366, L368, K370, D399, F405, Y407, and K409 (according to EU numbering), and the second heavy chain has at least one amino acid substitution at a position corresponding to a position selected from the group consisting of T366, L368, K370, D399, F405, Y407, and K409 (according to EU numbering), and wherein the first heavy chain and the second heavy chain are not substituted at the same positions.

66. 66. The antibody of claim 65, wherein (i) in the first heavy chain, the amino acid at the position corresponding to F405 (according to EU numbering) is L and in the second heavy chain, the amino acid at the position corresponding to K409 (according to EU numbering) is R, or (ii) in the first heavy chain, the amino acid at the position corresponding to K409 (according to EU numbering) is R and in the second heavy chain, the amino acid at the position corresponding to F405 (according to EU numbering) is L.

67. The antibody according to any one of the preceding claims, comprising a first heavy chain and a second heavy chain, and one or both heavy chains are modified such that the antibody induces Fc-mediated effector functions to a lesser extent than an otherwise identical antibody comprising an unmodified first and second heavy chains, An antibody according to any one of the preceding claims.

68. 68. The antibody of claim 67, wherein the Fc-mediated effector function is measured by determining Fc-mediated CD69 expression, by binding to Fcgamma receptors, by binding to C1q, or by induction of Fc-mediated FcR cross-linking.

69. 69. The antibody of claim 67 or 68, wherein the heavy and light chain constant sequences are modified to reduce Fc-mediated CD69 expression by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99%, or 100% when compared to a wild type antibody, The Fc-mediated CD69 expression is measured in a PBMC-based functional assay.

69. The antibody of claim 67 or 68.

70. The antibody of any one of the preceding claims, comprising a first heavy chain and a second heavy chain, wherein in at least one of the first heavy chain and the second heavy chain, one or more amino acids at positions corresponding to positions L234, L235, D265, N297, and P331 in a human IgG1 heavy chain according to EU numbering are not L, L, D, N, and P, respectively.

71. 71. The antibody of claim 70, wherein in the first heavy chain and the second heavy chain, the positions corresponding to positions L234 and L235 in a human IgG1 heavy chain according to EU numbering are F and E, respectively.

72. 72. The antibody of claim 71, wherein the antibody is a bispecific antibody comprising a first heavy chain and a second heavy chain, and wherein in both the first heavy chain and the second heavy chain, positions corresponding to positions L234 and L235 in a human IgG1 heavy chain according to EU numbering are F and E, respectively, and wherein (i) the position corresponding to F405 in the human IgG1 heavy chain according to EU numbering of the first heavy chain is L and the position corresponding to K409 in the human IgG1 heavy chain according to EU numbering of the second heavy chain is R, or (ii) the position corresponding to K409 in the human IgG1 heavy chain according to EU numbering of the first heavy chain is R and the position corresponding to F405 in the human IgG1 heavy chain according to EU numbering of the second heavy chain is L.

73. 71. The antibody of claim 70, wherein in the first heavy chain and the second heavy chain, the positions corresponding to positions L234, L235, and D265 in a human IgG1 heavy chain according to EU numbering are F, E, and A, respectively.

74. 74. The antibody of claim 73, wherein the antibody is a bispecific antibody comprising a first heavy chain and a second heavy chain, and wherein in both the first heavy chain and the second heavy chain, positions corresponding to positions L234, L235, and D265 in a human IgG1 heavy chain according to EU numbering are F, E, and A, respectively, and wherein (i) the position corresponding to F405 in the human IgG1 heavy chain according to EU numbering of the first heavy chain is L and the position corresponding to K409 in the human IgG1 heavy chain according to EU numbering of the second heavy chain is R, or (ii) the position corresponding to K409 in the human IgG1 heavy chain according to EU numbering of the first heavy chain is R and the position corresponding to F405 in the human IgG1 heavy chain according to EU numbering of the second heavy chain is L.

75. 2. The antibody of any one of the preceding claims, which does not bind to human PD-L2.

76. Approximately 10 -8 M or less, e.g., about 10 -9 M or less, e.g., about 10 -10 M or less K D 5. The antibody of any preceding claim, which binds to human PD-L1 at

77. The antibody of any one of the preceding claims, which mediates concentration-dependent cytotoxicity of MDA-MB-231 cells, PC-3 cells, and / or HELA cells when purified T cells are used as effector cells, when assayed as described in Example 11 herein.

78. (i) a nucleic acid sequence encoding an antibody heavy chain sequence comprising an antigen-binding region capable of binding to human PD-L1 as defined in any one of claims 1 to 31, and / or (ii) a nucleic acid sequence encoding a light chain sequence of an antibody comprising an antigen-binding region capable of binding to human PD-L1 as defined in any one of claims 1 to 31. A nucleic acid construct comprising:

79. (i) a nucleic acid sequence encoding a heavy chain sequence of an antibody comprising an antigen-binding region capable of binding to human CD3ε as defined in any one of claims 33 to 38, and (ii) a nucleic acid sequence encoding a light chain sequence of an antibody comprising an antigen-binding region capable of binding to human CD3ε as defined in any one of claims 33 to 38.

74. The nucleic acid construct of claim 73, further comprising:

80. 80. An expression vector comprising a nucleic acid construct as defined in claim 78 or 79.

81. 81. A host cell comprising a nucleic acid construct as defined in claim 78 or 79 or an expression vector as defined in claim 80.

82. 82. The host cell of claim 81, which is a mammalian cell, such as a Chinese hamster ovary cell.

83. A pharmaceutical composition comprising an antibody according to any one of claims 1 to 77 and a pharma- ceutically acceptable carrier.

84. An antibody according to any one of claims 1 to 77 or a pharmaceutical composition according to claim 83 for use as a medicament.

85. 81. An antibody according to any one of claims 1 to 80 or a pharmaceutical composition according to claim 70 for use in the treatment of cancer.

86. An antibody according to any one of claims 1 to 77 or a pharmaceutical composition according to claim 70 for use in the treatment of a cancer disease characterised by the presence of a solid tumour.

87. The antibody of any one of claims 1 to 78 or the pharmaceutical composition of claim 70 for use in the treatment of a cancer disease selected from the group consisting of melanoma, ovarian cancer, lung cancer, colon cancer and head and neck cancer.

88. A method for treating a disease comprising administering to a subject in need thereof an antibody according to any one of claims 1 to 75 or a pharmaceutical composition according to claim 83.

89. 80. Use of an antibody according to any one of claims 1 to 77 for the manufacture of a medicament, such as a medicament for the treatment of cancer, e.g. a cancer disease characterised by the presence of a solid tumour or a cancer disease selected from the group consisting of melanoma, ovarian cancer, lung cancer, colon cancer and head and neck cancer.

90. 90. The method or use of any one of claims 83 to 89, comprising combination with one or more further therapeutic substances, such as a combination with a chemotherapeutic agent.

91. (a) culturing a host cell producing a first antibody comprising an antigen-binding region capable of binding to human PD-L1 as defined in any one of claims 1 to 13, and purifying the first antibody from the culture; (b) culturing a host cell producing a second antibody that comprises an antigen-binding region capable of binding to a different PD-L1 epitope or to a different antigen, such as the human CD3ε binding region as defined in any one of claims 14 to 19, and purifying the second antibody from the culture; (c) incubating the first antibody with the second antibody under reducing conditions sufficient to allow cysteines in the hinge region to undergo disulfide bond isomerization; and (d) Obtaining a bispecific antibody A method for producing an antibody according to any one of claims 1 to 77, comprising:

92. An anti-idiotypic antibody which binds to an antigen-binding region capable of binding to human PD-L1 as defined in any one of claims 1 to 77.

Images