Bispecific antibodies and uses thereof

Abstract

The present disclosure relates to bispecific antibodies or antigen-binding fragments thereof, which specifically bind to two different antigens with different binding affinities.

Description

[Technical Field]

[0001] Priority claims This application claims the benefit of U.S. Provisional Patent Application No. 62 / 539,970, filed August 1, 2017, and U.S. Provisional Patent Application No. 62 / 654,112, filed April 6, 2018, the entire contents of which are incorporated herein by reference.

[0002] Technical Field The present disclosure relates to bispecific antibodies or antigen-binding fragments thereof. [Background technology]

[0003] background Bispecific antibodies are engineered proteins that can simultaneously bind to two different antigens or epitopes. This dual specificity opens up a wide range of applications, including redirecting T cells to tumor cells, simultaneously blocking two different signaling pathways, dual targeting of different disease mediators, and delivering payloads to target sites. The approval of catumasomab (anti-EpCAM and anti-CD3) and blinatumomab (anti-CD19 and anti-CD3) marked major milestones in the development of bispecific antibodies.

[0004] The wide range of uses for bispecific antibodies necessitates the continued development of various therapeutic approaches based on bispecific antibodies. Summary of the Invention

[0005] overview The present disclosure relates to unbalanced bispecific antibodies or antigen-binding fragments, in which the bispecific antibodies or antigen-binding fragments specifically bind to two different antigens with different binding affinities.

[0006] In some embodiments, the present disclosure relates to a bispecific antibody or antigen-binding fragment comprising a first heavy chain variable region, a second heavy chain variable region, a first light chain variable region, and a second light chain variable region, wherein the first heavy chain variable region and the first light chain variable region bind to each other to form a 10 7 M -1 , 10 8 M -1 , 10 9 M -1 , 10 10 M -1 , 10 11 M -1 , or 10 12 M -1 and the second heavy chain variable region and the second light chain variable region bind to each other to form a first antigen-binding region that specifically binds to a first antigen with a binding affinity higher than 10 9 M -1 , 10 8 M -1 , 10 7 M -1 , 10 6 M -1 , 10 5 M -1 , or 10 4 M -1 a second antigen-binding region that specifically binds to a second antigen with a lower binding affinity than the first antigen-binding region;

[0007] In some embodiments, the second antigen-binding region comprises 10 7 M -1 , 10 6 M -1 , 10 5 M -1 , or 10 4 M -1 specifically binds to a second antigen with a higher binding affinity than the first antigen.

[0008] In some embodiments, the binding affinity of the first antigen-binding region when binding to the first antigen is at least 100-fold, 1000-fold, or 10000-fold higher than the binding affinity of the second antigen-binding region when binding to the second antigen.

[0009] In some embodiments, the first light chain variable region and the second light chain variable region are at least 90%, 95%, 99%, or 100% identical.

[0010] In some embodiments, the present disclosure relates to a bispecific antibody or antigen-binding fragment comprising a first arm comprising a first heavy chain variable region and a first light chain variable region, and a second arm comprising a second heavy chain variable region and a second light chain variable region, wherein the first arm is 10 7 M -1 , 10 8 M -1 , 10 9 M -1 , 10 10 M -1 , 10 11 M -1 , 10 12 M -1 and the second arm specifically binds to the first antigen with a binding affinity greater than 10 9 M -1 , 10 8 M -1 , 10 7 M -1 , 10 6 M -1 , 10 5 M -1 , or 10 4 M -1 specifically binds to the second antigen with a lower binding affinity than the first antigen.

[0011] In some embodiments, the second arm is 7 M -1 , 10 6 M -1 , 10 5 M -1 , or 10 4 M -1 It specifically binds to a second antigen with higher binding affinity.

[0012] In some embodiments, the binding affinity of the first arm when binding to the first antigen is at least 100-fold, 1000-fold, or 10000-fold higher than the binding affinity of the second arm when binding to the second antigen.

[0013] In some embodiments, the first light chain variable region and the second light chain variable region are at least 90%, 95%, 99%, or 100% identical.

[0014] In some embodiments, the present disclosure relates to a bispecific antibody or antigen-binding fragment comprising a first heavy chain comprising a first heavy chain variable region, a second heavy chain comprising a second heavy chain variable region, a first light chain comprising a first light chain variable region, and a second light chain comprising a second light chain variable region, wherein the first heavy chain variable region and the first light chain variable region bind to each other to form a bispecific antibody or antigen-binding fragment. 7 M -1 , 10 8 M -1 , 10 9 M -1 , 10 10 M -1 , 10 11 M -1 , 10 12 M -1 and a first antigen-binding region that specifically binds to a first antigen with a binding affinity higher than 10. 9 M -1 , 10 8 M -1 , 10 7 M -1 , 10 6 M -1 , 10 5 M -1 , or 10 4 M -1 a second antigen-binding region that specifically binds to a second antigen with a lower binding affinity than the first antigen-binding region;

[0015] In some embodiments, the second antigen-binding region comprises 10 7 M -1 , 10 6 M -1 , 10 5 M -1 , or 10 4 M -1 It specifically binds to a second antigen with higher binding affinity.

[0016] In some embodiments, the binding affinity of the first antigen-binding region when binding to the first antigen is at least 100-fold, 1000-fold, or 10000-fold higher than the binding affinity of the second antigen-binding region when binding to the second antigen.

[0017] In some embodiments, the first light chain and the second light chain are at least 90%, 95%, 99%, or 100% identical.

[0018] In some embodiments, the first heavy chain and the second chain are linked to each other by the "knobs into holes" method.

[0019] In some embodiments, the first antigen is a cancer-specific antigen and the second antigen is CD3.

[0020] In some embodiments, the first antigen is CD20 and the second antigen is CD3.

[0021] In some embodiments, the first heavy chain variable region comprises a sequence at least 80%, 85%, 90%, or 95% identical to SEQ ID NO:1, the second heavy chain variable region comprises a sequence at least 80%, 85%, 90%, or 95% identical to SEQ ID NO:2, and the first and second light chain variable regions comprise a sequence at least 80%, 85%, 90%, or 95% identical to SEQ ID NO:3.

[0022] In some embodiments, the first antigen is a cancer-specific antigen and the second antigen is a cancer-associated antigen.

[0023] In some embodiments, the first antigen is PD-L1 and the second antigen is CD55.

[0024] In some embodiments, the first heavy chain variable region comprises a sequence at least 80%, 85%, 90%, or 95% identical to SEQ ID NO:4, the second heavy chain variable region comprises a sequence at least 80%, 85%, 90%, or 95% identical to SEQ ID NO:5, and the first and second light chain variable regions comprise a sequence at least 80%, 85%, 90%, or 95% identical to SEQ ID NO:6 or SEQ ID NO:7.

[0025] In some aspects, the present disclosure relates to a method of making a bispecific antibody or antigen-binding fragment, the method comprising: selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment that binds to the first antigen and a second antibody or antigen-binding fragment that binds to the second antigen, wherein the first antibody or antigen-binding fragment comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); determining the amino acid sequences of VHa, VLa, VHb, and VLb; aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is greater than 80%; designing a common light chain variable region (VLc), wherein the VLc maintains affinity for the first antigen when combined with the VHa; redesigning the VHa and VHb sequences to obtain VHa' and VHb' that increase the difference in biochemical or biophysical properties between a first protein comprising two polypeptides each comprising a VHa' and two polypeptides each comprising a VLc, and a second protein comprising two polypeptides each comprising a VHb' and two polypeptides each comprising a VLc; and Producing a bispecific antibody or antigen-binding fragment having two light chain variable regions and two heavy chain variable regions, wherein each of the two light chain variable regions comprises a VLc, and the two heavy chain variable regions comprise a VHa' and a VHb', respectively. Includes.

[0026] In some embodiments, in step (d), the binding affinity of the VLc-VHb for the second antigen may be reduced.

[0027] In some embodiments, the method further comprises creating a buffer system for purifying the bispecific antibody or antigen-binding fragment.

[0028] In some aspects, the present disclosure relates to a method of making a bispecific antibody or antigen-binding fragment, the method comprising: selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment that binds to the first antigen and a second antibody or antigen-binding fragment that binds to the second antigen, wherein the first antibody or antigen-binding fragment comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); determining the amino acid sequences of VHa, VLa, and VLb; aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is less than 80%; replacing all light chain variable regions in the phage display antibody library with VLa and panning against a second antigen to obtain a third heavy chain variable region (VHc); redesigning the VHa and VHc sequences to obtain VHa' and VHc' that increase the difference in biochemical or biophysical properties between a first protein comprising two polypeptides each comprising a VHa' and two polypeptides each comprising a VLa, and a second protein comprising two polypeptides each comprising a VHc' and two polypeptides each comprising a VLa; and Producing a bispecific antibody or antigen-binding fragment having two light chain variable regions and two heavy chain variable regions, wherein each of the two light chain variable regions comprises a VLa, and the two heavy chain variable regions comprise a VHa' and a VHc', respectively. Includes.

[0029] In some embodiments, the method further comprises creating a buffer system for purifying the bispecific antibody or antigen-binding fragment.

[0030] In some aspects, the present disclosure relates to a method of making a bispecific antibody or antigen-binding fragment, the method comprising: selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment that binds to the first antigen and a second antibody or antigen-binding fragment that binds to the second antigen, wherein the first antibody or antigen-binding fragment comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); determining the amino acid sequences of VHa, VLa, VHb, and VLb; aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is less than 80%; replacing all light chain variable regions in a phage display antibody library with a plurality of light chain variable regions, wherein the light chain variable regions are at least 80%, 85%, 90%, 95%, or 99% identical to VLa or VLb; panning against a second antigen; selecting a common light chain variable region (VLc), and a third heavy chain variable region (VHc), wherein the VHa-VLc binds to a first antigen with a desired affinity and the VHc-VLc binds to a second antigen with a desired affinity; redesigning the VHa and VHc sequences to obtain VHa' and VHc' that increase the difference in biochemical or biophysical properties between a first protein comprising two polypeptides each comprising a VHa' and two polypeptides each comprising a VLc, and a second protein comprising two polypeptides each comprising a VHc' and two polypeptides each comprising a VLc; and Producing a bispecific antibody or antigen-binding fragment having two light chain variable regions and two heavy chain variable regions, wherein each of the two light chain variable regions comprises a VLc, and the two heavy chain variable regions comprise a VHa' and a VHc', respectively. Includes.

[0031] In some embodiments, in step (d), the plurality of light chain variable regions is generated by error-prone PCR.

[0032] In some embodiments, the method further comprises creating a buffer system for purifying the bispecific antibody or antigen-binding fragment.

[0033] In one aspect, the present disclosure provides a method of making a bispecific antibody or antigen-binding fragment thereof, the method comprising one or more of the following steps: (a) selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment thereof that binds to the first antigen and a second antibody or antigen-binding fragment thereof that binds to the second antigen. In some embodiments, the first antibody or antigen-binding fragment thereof comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment thereof comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); (b) determining the amino acid sequences of VHa, VLa, VHb, and VLb; (c) aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is less than 80%; (d) replacing all light chain variable regions in the phage display antibody library with a plurality of light chain variable regions, in some embodiments, that are at least 80%, 85%, 90%, 95%, or 99% identical to VLa or VLb; (e) panning against a second antigen; (f) selecting a common light chain variable region (VLc) and a third heavy chain variable region (VHc), in some embodiments, the VHc-VLc binds to the second antigen with the desired affinity; (g) determining that the homology between VLa and VLc is greater than 80%; (h) designing a common light chain variable region (VLd), in some embodiments, that maintains affinity for the first antigen when bound to the VHa and has a desired affinity for the second antigen when bound to the VHc; (i) optionally, redesigning the VHa and VHc sequences to obtain VHa' and VHc' that increase the difference in biochemical or biophysical properties between a first protein comprising two polypeptides each comprising a VHa' and two polypeptides each comprising a VLd, and a second protein comprising two polypeptides each comprising a VHc' and two polypeptides each comprising a VLd; and (j) optionally, generating a bispecific antibody or antigen-binding fragment thereof having two light chain variable regions and two heavy chain variable regions, in some embodiments, each of the two light variable regions comprises a VLd and the two heavy chain variable regions comprise a VHa' and a VHc', respectively.

[0034] In one aspect, the present disclosure provides a method of making a bispecific antibody or antigen-binding fragment thereof, the method comprising one or more of the following steps: (a) selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment thereof that binds to the first antigen and a second antibody or antigen-binding fragment thereof that binds to the second antigen. In some embodiments, the first antibody or antigen-binding fragment thereof comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment thereof comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); (b) determining the amino acid sequences of VHa, VLa, VHb, and VLb; (c) aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is greater than 80%; (d) designing a common light chain variable region (VLc), in some embodiments, the VLc maintains affinity for the first antigen when combined with the VHa; and (e) optionally, generating a bispecific antibody or antigen-binding fragment thereof having two light chain variable regions and two heavy chain variable regions, in some embodiments, each of the two light variable regions comprises a VLc, and the two heavy chain variable regions comprise a VHa and a VHb, respectively.

[0035] In one aspect, the present disclosure also provides a method of making a bispecific antibody or antigen-binding fragment thereof, the method comprising one or more of the following steps: (a) selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment thereof that binds to the first antigen and a second antibody or antigen-binding fragment thereof that binds to the second antigen. In some embodiments, the first antibody or antigen-binding fragment thereof comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment thereof comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); (b) determining the amino acid sequences of VHa, VLa, and VLb; (c) aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is less than 80%; (d) replacing all light chain variable regions in the phage display antibody library with VLa and panning against a second antigen to obtain a third heavy chain variable region (VHc); and (e) optionally, generating a bispecific antibody or antigen-binding fragment thereof having two light chain variable regions and two heavy chain variable regions, in some embodiments, each of the two light variable regions comprises a VLa, and the two heavy chain variable regions comprise a VHa and a VHc, respectively.

[0036] In one aspect, the present disclosure further provides a method of making a bispecific antibody or antigen-binding fragment thereof, the method comprising one or more of the following steps: (a) selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment thereof that binds to the first antigen and a second antibody or antigen-binding fragment thereof that binds to the second antigen. In some embodiments, the first antibody or antigen-binding fragment thereof comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment thereof comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); (b) determining the amino acid sequences of VHa, VLa, VHb, and VLb; (c) aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is less than 80%; (d) replacing all light chain variable regions in the phage display antibody library with a plurality of light chain variable regions, in some embodiments, that are at least 80%, 85%, 90%, 95%, or 99% identical to VLa or VLb; (e) panning against the first and / or second antigen; (f) selecting a common light chain variable region (VLc), and a third heavy chain variable region (VHc), in some embodiments, where the VHa-VLc binds the first antigen with the desired affinity and the VHc-VLc binds the second antigen with the desired affinity; and (g) optionally, generating a bispecific antibody or antigen-binding fragment thereof having two light chain variable regions and two heavy chain variable regions, in some embodiments, each of the two light variable regions comprises a VLc and the two heavy chain variable regions comprise a VHa and a VHc, respectively.

[0037] In another aspect, the present disclosure provides an antibody or antigen-binding fragment thereof that binds to CD3, which a heavy chain variable region (VH) comprising complementarity determining regions (CDRs) 1, 2, and 3, wherein the VH CDR1 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR1 amino acid sequence, the VH CDR2 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR2 amino acid sequence, and the VH CDR3 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR3 amino acid sequence; and A light chain variable region (VL) comprising CDR1, 2, and 3, wherein the VL CDR1 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR1 amino acid sequence, the VL CDR2 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR2 amino acid sequence, and the VL CDR3 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR3 amino acid sequence. wherein the selected VH CDR1, 2, 3 amino acid sequences and the selected VL CDR1, 2, 3 amino acid sequences are one of the following: The amino acid sequences of the selected VH CDR1, 2, and 3 are set forth in SEQ ID NOs: 22 to 24, respectively, and the amino acid sequences of the selected VL CDR1, 2, and 3 are set forth in SEQ ID NOs: 28 to 30, respectively.

[0038] In some embodiments, the VH comprises CDR1, 2, and 3 having the amino acid sequences set forth in SEQ ID NOs: 22, 23, and 24, respectively, and the VL comprises CDR1, 2, and 3 having the amino acid sequences set forth in SEQ ID NOs: 28, 29, and 30, respectively.

[0039] In some embodiments, the antibody or antigen-binding fragment specifically binds to human CD3.

[0040] In some embodiments, the antibody or antigen-binding fragment is a bispecific antibody.

[0041] In another embodiment, the disclosure also provides an antibody or antigen-binding fragment thereof that binds to PD-L1, which a heavy chain variable region (VH) comprising complementarity determining regions (CDRs) 1, 2, and 3, wherein the VH CDR1 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR1 amino acid sequence, the VH CDR2 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR2 amino acid sequence, and the VH CDR3 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR3 amino acid sequence; and A light chain variable region (VL) comprising CDR1, 2, and 3, wherein the VL CDR1 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR1 amino acid sequence, the VL CDR2 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR2 amino acid sequence, and the VL CDR3 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR3 amino acid sequence. wherein the amino acid sequences of the selected VH CDRs 1, 2, 3 and the amino acid sequences of the selected VL CDRs 1, 2, 3 are one of the following: (1) The amino acid sequences of the selected VH CDR1, 2, and 3 are set forth in SEQ ID NOs: 41 to 43, respectively, and the amino acid sequences of the selected VL CDR1, 2, and 3 are set forth in SEQ ID NOs: 53 to 55, respectively; (2) The amino acid sequences of the selected VH CDR1, 2, and 3 are set forth in SEQ ID NOs: 41 to 43, respectively, and the amino acid sequences of the selected VL CDR1, 2, and 3 are set forth in SEQ ID NOs: 59 to 61, respectively.

[0042] In some embodiments, the VH comprises CDR1, 2, and 3 having the amino acid sequences set forth in SEQ ID NOs: 41-43, respectively, and the VL comprises CDR1, 2, and 3 having the amino acid sequences set forth in SEQ ID NOs: 59-61, respectively.

[0043] In some embodiments, the antibody or antigen-binding fragment specifically binds to human CD3. In some embodiments, the antibody or antigen-binding fragment is a bispecific antibody.

[0044] In another aspect, the present disclosure provides an antibody or antigen-binding fragment thereof that binds to CD55, which a heavy chain variable region (VH) comprising complementarity determining regions (CDRs) 1, 2, and 3, wherein the VH CDR1 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR1 amino acid sequence, the VH CDR2 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR2 amino acid sequence, and the VH CDR3 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR3 amino acid sequence; and A light chain variable region (VL) comprising CDR1, 2, and 3, wherein the VL CDR1 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR1 amino acid sequence, the VL CDR2 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR2 amino acid sequence, and the VL CDR3 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR3 amino acid sequence. wherein the selected VH CDR1, 2, 3 amino acid sequence and the selected VL CDR1, 2, 3 amino acid sequence are one of the following: (1) The amino acid sequences of the selected VH CDR1, 2, and 3 are set forth in SEQ ID NOs: 47 to 49, respectively, and the amino acid sequences of the selected VL CDR1, 2, and 3 are set forth in SEQ ID NOs: 53 to 55, respectively; (2) The amino acid sequences of the selected VH CDR1, 2, and 3 are set forth in SEQ ID NOs: 47 to 49, respectively, and the amino acid sequences of the selected VL CDR1, 2, and 3 are set forth in SEQ ID NOs: 59 to 61, respectively.

[0045] In some embodiments, the VH comprises CDR1, 2, and 3 having the amino acid sequences set forth in SEQ ID NOs: 47-49, respectively, and the VL comprises CDR1, 2, and 3 having the amino acid sequences set forth in SEQ ID NOs: 59-61, respectively.

[0046] In some embodiments, the antibody or antigen-binding fragment specifically binds to human CD3.

[0047] In some embodiments, the antibody or antigen-binding fragment is a bispecific antibody.

[0048] In one aspect, the disclosure provides a nucleic acid comprising a polynucleotide encoding a polypeptide comprising an immunoglobulin light chain or fragment thereof comprising a VL comprising CDR1, 2, 3 comprising the amino acid sequences set forth in SEQ ID NOs: 53-55, respectively. In some embodiments, the VL binds to PD-L1 when paired with a VH comprising the amino acid sequence set forth in SEQ ID NO:4, and / or binds to CD55 when paired with a VH comprising the amino acid sequence set forth in SEQ ID NO:5.

[0049] In one aspect, the disclosure provides a nucleic acid comprising a polynucleotide encoding a polypeptide comprising an immunoglobulin light chain or fragment thereof comprising a VL comprising CDR1, 2, 3 comprising the amino acid sequences set forth in SEQ ID NOs:59-61, respectively. In some embodiments, the VL binds to PD-L1 when paired with a VH comprising the amino acid sequence set forth in SEQ ID NO:4, and / or binds to CD55 when paired with a VH comprising the amino acid sequence set forth in SEQ ID NO:5.

[0050] In some embodiments, the nucleic acid encodes a bispecific antibody. In some embodiments, the nucleic acid is a cDNA.

[0051] In one aspect, the disclosure provides a vector comprising one or more of the nucleic acids described herein.

[0052] In one aspect, the disclosure provides a cell comprising a vector described herein. In some embodiments, the cell is a CHO cell.

[0053] In one aspect, the disclosure provides a cell comprising one or more of the nucleic acids described herein.

[0054] In one aspect, the present disclosure provides a bispecific antibody or antigen-binding fragment thereof that binds to CD20 and CD3, comprising: a first polypeptide comprising a first heavy chain variable region (VH) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:1; a second polypeptide comprising a second heavy chain variable region (VH) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:2; a third polypeptide comprising a first light chain variable region (VL) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:3; and a fourth polypeptide comprising a second light chain variable region (VL) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:3.

[0055] In some embodiments, the first heavy chain variable region (VH) comprises SEQ ID NO:1, the second heavy chain variable region (VH) comprises SEQ ID NO:2, the first light chain variable region (VL) comprises SEQ ID NO:3, and the second light chain variable region (VL) comprises SEQ ID NO:3.

[0056] In some embodiments, the first polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:34, 35, or 36; the second polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:37, 38, or 39; the third polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:40; and the fourth polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:40.

[0057] In some embodiments, the first polypeptide comprises the amino acid sequence set forth in SEQ ID NO:35 and the second polypeptide comprises the amino acid sequence shown in SEQ ID NO:38.

[0058] In one aspect, the disclosure provides a bispecific antibody or antigen-binding fragment thereof that binds PD-L1 and CD55, comprising: a first polypeptide comprising a first heavy chain variable region (VH) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:4; a second polypeptide comprising a second heavy chain variable region (VH) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:5; a third polypeptide comprising a first light chain variable region (VL) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:6 or 7; and a fourth polypeptide comprising a second light chain variable region (VL) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:6 or 7.

[0059] In some embodiments, the first heavy chain variable region (VH) comprises SEQ ID NO:4, the second heavy chain variable region (VH) comprises SEQ ID NO:5, the first light chain variable region (VL) comprises SEQ ID NO:7, and the second light chain variable region (VL) comprises SEQ ID NO:7.

[0060] In some embodiments, the first polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:65, the second polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:66, the third polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:67 or 68, and the fourth polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:67 or 68.

[0061] In some embodiments, the first polypeptide comprises the amino acid sequence set forth in SEQ ID NO:65, the second polypeptide comprises the amino acid sequence set forth in SEQ ID NO:66, the third polypeptide comprises the amino acid sequence set forth in SEQ ID NO:68, and the fourth polypeptide comprises the amino acid sequence set forth in SEQ ID NO:68.

[0062] In one aspect, the present disclosure provides an antibody-drug conjugate comprising an antibody or antigen-binding fragment thereof described herein covalently attached to a therapeutic agent. In some embodiments, the therapeutic agent is a cytotoxic or cytostatic agent.

[0063] In one aspect, the present disclosure provides a method of treating a subject with cancer. The method comprises administering to the subject a therapeutically effective amount of a composition comprising an antibody or antigen-binding fragment thereof described herein, or an antibody-drug conjugate described herein. In some embodiments, the subject has a solid tumor. In some embodiments, the cancer is melanoma, pancreatic cancer, or a hematological malignancy. In some embodiments, the cancer is non-Hodgkin's lymphoma, lymphoma, or chronic lymphocytic leukemia.

[0064] In one aspect, the disclosure provides a method for reducing the rate of tumor growth, comprising contacting tumor cells in a subject with an effective amount of a composition comprising an antibody or antigen-binding fragment thereof described herein, or an antibody-drug conjugate described herein.

[0065] In one aspect, the disclosure provides a method of killing tumor cells, the method comprising contacting tumor cells in a subject with an effective amount of a composition comprising an antibody or antigen-binding fragment thereof described herein, or an antibody-drug conjugate described herein.

[0066] In one aspect, the present disclosure provides a pharmaceutical composition comprising an antibody or antigen-binding fragment thereof described herein and a pharmaceutically acceptable carrier.

[0067] In one aspect, the present disclosure provides a pharmaceutical composition comprising an antibody-drug conjugate described herein and a pharmaceutically acceptable carrier.

[0068] Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention belongs. Although methods and materials for use in the present invention are described herein, other suitable methods and materials known in the art can also be used. Materials, methods, and examples are for illustrative purposes only and are not intended to be limiting. All publications, patent applications, patents, sequences, database entries, and other references mentioned herein are incorporated by reference in their entirety. In the event of any conflict, the present specification, including definitions, will control.

[0069] [The present invention 1001] a first heavy chain variable region, a second heavy chain variable region, a first light chain variable region, and Second light chain variable region A bispecific antibody or antigen-binding fragment thereof comprising: The first heavy chain variable region and the first light chain variable region bind to each other to form 107 M -1 、10 8 M -1 、10 9 M -1 、10 10 M -1 、10 11 M -1 , or 10 12 M -1 forming a first antigen-binding region that specifically binds to the first antigen with a binding affinity greater than The second heavy chain variable region and the second light chain variable region bind to each other to form 10 9 M -1 、10 8 M -1 、10 7 M -1 、10 6 M -1 、10 5 M -1 , or 10 4 M -1 forming a second antigen-binding region that specifically binds to a second antigen with a lower binding affinity than The bispecific antibody or antigen-binding fragment thereof. [The present invention 1002] The second antigen-binding region is 7 M -1 、10 6 M -1 、10 5 M -1 , or 10 4 M -1 1001. A bispecific antibody or antigen-binding fragment thereof of the present invention, which specifically binds to a second antigen with a higher binding affinity than the first antibody or antigen-binding fragment thereof. [The present invention 1003] 1001. A bispecific antibody or antigen-binding fragment thereof of the present invention, wherein the binding affinity of the first antigen-binding region when binding to the first antigen is at least 100-fold, 1000-fold, or 10000-fold higher than the binding affinity of the second antigen-binding region when binding to the second antigen. [The present invention 1004] 1001. The bispecific antibody or antigen-binding fragment thereof of the present invention, wherein the first light chain variable region and the second light chain variable region are at least 90%, 95%, 99%, or 100% identical. [The present invention 1005] a first arm comprising a first heavy chain variable region and a first light chain variable region; and a second arm comprising a second heavy chain variable region and a second light chain variable region; A bispecific antibody or antigen-binding fragment thereof comprising: The first arm is 10 7 M -1 、10 8 M -1 、10 9 M -1 、10 10 M -1 、10 11 M -1 、10 12 M -1 and the second arm specifically binds to the first antigen with a binding affinity greater than 10 9 M -1 、10 8 M -1 、10 7 M -1 、10 6 M -1 、10 5 M -1 , or 10 4 M -1 specifically binds to a second antigen with a lower binding affinity than The bispecific antibody or antigen-binding fragment thereof. [The present invention 1006] The second arm is 10 7 M -1 、10 6 M -1 、10 5 M-1 , or 10 4 M -1 1005. The bispecific antibody or antigen-binding fragment thereof of the present invention, which specifically binds to a second antigen with a higher binding affinity than the first antibody or antigen-binding fragment thereof. [The present invention 1007] 1005. The bispecific antibody or antigen-binding fragment thereof of the present invention, wherein the binding affinity of said first arm when binding to a first antigen is at least 100-fold, 1000-fold, or 10000-fold higher than the binding affinity of said second arm when binding to a second antigen. [The present invention 1008] 1005. The bispecific antibody or antigen-binding fragment thereof of the present invention, wherein the first light chain variable region and the second light chain variable region are at least 90%, 95%, 99%, or 100% identical. [The present invention 1009] a first heavy chain comprising a first heavy chain variable region; a second heavy chain comprising a second heavy chain variable region; a first light chain comprising a first light chain variable region, and a second light chain comprising a second light chain variable region A bispecific antibody or antigen-binding fragment thereof comprising: The first heavy chain variable region and the first light chain variable region bind to each other to form 10 7 M -1 、10 8 M -1 、10 9 M -1 、10 10 M -1 、10 11 M -1 、10 12 M -1 forming a first antigen-binding region that specifically binds to the first antigen with a binding affinity greater than The second heavy chain variable region and the second light chain variable region bind to each other to form 10 9 M -1 、10 8 M -1 、10 7 M -1 、10 6 M -1 、10 5 M -1 , or 10 4 M -1 forming a second antigen-binding region that specifically binds to a second antigen with a lower binding affinity than The bispecific antibody or antigen-binding fragment thereof. [The present invention 1010] The second antigen-binding region is 7 M -1 、10 6 M -1 、10 5 M -1 , or 10 4 M -1 1009. The bispecific antibody or antigen-binding fragment thereof of the present invention, which specifically binds to a second antigen with a higher binding affinity than the first antibody or antigen-binding fragment thereof. [The present invention 1011] 1009. The bispecific antibody or antigen-binding fragment thereof of the present invention, wherein the binding affinity of the first antigen-binding region when binding to the first antigen is at least 100-fold, 1000-fold, or 100,000-fold higher than the binding affinity of the second antigen-binding region when binding to the second antigen. [The present invention 1012] 1009. The bispecific antibody or antigen-binding fragment thereof of the present invention, wherein the first light chain and the second light chain are at least 90%, 95%, 99%, or 100% identical. [The present invention 1013] 1009. The bispecific antibody or antigen-binding fragment thereof of the present invention, wherein the first heavy chain and the second heavy chain are bound to each other by the knobs-into-holes method. [The present invention 1014] The bispecific antibody or antigen-binding fragment thereof of any of claims 1001 to 1013, wherein the first antigen is a cancer-specific antigen and the second antigen is CD3. [The present invention 1015] The bispecific antibody or antigen-binding fragment thereof of any of claims 1001 to 1013, wherein the first antigen is CD20 and the second antigen is CD3. [The present invention 1016] 1015. A bispecific antibody or antigen-binding fragment thereof of the invention, wherein the first heavy chain variable region comprises a sequence that is at least 80%, 85%, 90%, or 95% identical to SEQ ID NO:1, the second heavy chain variable region comprises a sequence that is at least 80%, 85%, 90%, or 95% identical to SEQ ID NO:2, and the first and second light chain variable regions comprise a sequence that is at least 80%, 85%, 90%, or 95% identical to SEQ ID NO:3. [The present invention 1017] The bispecific antibody or antigen-binding fragment thereof of any of claims 1001 to 1013, wherein the first antigen is a cancer-specific antigen and the second antigen is a cancer-associated antigen. [The present invention 1018] The bispecific antibody or antigen-binding fragment thereof of any of claims 1001 to 1013, wherein the first antigen is PD-L1 and the second antigen is CD55. [The present invention 1019] 1018. A bispecific antibody or antigen-binding fragment thereof of the invention, wherein the first heavy chain variable region comprises a sequence that is at least 80%, 85%, 90%, or 95% identical to SEQ ID NO:4, the second heavy chain variable region comprises a sequence that is at least 80%, 85%, 90%, or 95% identical to SEQ ID NO:5, and the first and second light chain variable regions comprise a sequence that is at least 80%, 85%, 90%, or 95% identical to SEQ ID NO:6 or SEQ ID NO:7. [The present invention 1020] 1. A method of making a bispecific antibody or antigen-binding fragment thereof, comprising: (a) selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment thereof that binds to the first antigen and a second antibody or antigen-binding fragment thereof that binds to the second antigen, wherein the first antibody or antigen-binding fragment thereof comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment thereof comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); (b) determining the amino acid sequences of VHa, VLa, VHb, and VLb; (c) aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is greater than 80%; (d) designing a common light chain variable region (VLc), wherein the VLc maintains affinity for the first antigen when combined with the VHa; (e) redesigning the sequences of VHa and VHb to obtain VHa' and VHb' that increase the difference in biochemical or biophysical properties between a first protein comprising two polypeptides each comprising a VHa' and two polypeptides each comprising a VLc, and a second protein comprising two polypeptides each comprising a VHb' and two polypeptides each comprising a VLc; and (f) producing a bispecific antibody or antigen-binding fragment thereof having two light chain variable regions and two heavy chain variable regions, wherein each of the two light chain variable regions comprises a VLc, and the two heavy chain variable regions comprise a VHa' and a VHb', respectively. [The present invention 1021] The method of claim 1020, wherein in step (d), the binding affinity of the VLc-VHb for the second antigen can be reduced. [The present invention 1022] The method of the present invention 1020 further comprises: (g) creating a buffer system for purifying said bispecific antibody or antigen-binding fragment thereof. [The present invention 1023] 1. A method of making a bispecific antibody or antigen-binding fragment thereof, comprising: (a) selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment thereof that binds to the first antigen and a second antibody or antigen-binding fragment thereof that binds to the second antigen, wherein the first antibody or antigen-binding fragment thereof comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment thereof comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); (b) determining the amino acid sequences of VHa, VLa, and VLb; (c) aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is less than 80%; (d) replacing all light chain variable regions in the phage display antibody library with VLa and panning against a second antigen to obtain a third heavy chain variable region (VHc); (e) redesigning the sequences of VHa and VHc to obtain VHa' and VHc' that increase the difference in biochemical or biophysical properties between a first protein comprising two polypeptides each comprising a VHa' and two polypeptides each comprising a VLa, and a second protein comprising two polypeptides each comprising a VHc' and two polypeptides each comprising a VLa; and (f) producing a bispecific antibody or antigen-binding fragment thereof having two light chain variable regions and two heavy chain variable regions, wherein each of the two light chain variable regions comprises a VLa, and the two heavy chain variable regions comprise a VHa' and a VHc', respectively. [The present invention 1024] The method of the present invention 1023 further includes: (g) creating a buffer system for purifying said bispecific antibody or antigen-binding fragment thereof. [The present invention 1025] 1. A method of making a bispecific antibody or antigen-binding fragment thereof, comprising: (a) selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment thereof that binds to the first antigen and a second antibody or antigen-binding fragment thereof that binds to the second antigen, wherein the first antibody or antigen-binding fragment thereof comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment thereof comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); (b) determining the amino acid sequences of VHa, VLa, VHb, and VLb; (c) aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is less than 80%; (d) replacing all light chain variable regions in the phage display antibody library with a plurality of light chain variable regions, wherein the light chain variable regions are at least 80%, 85%, 90%, 95%, or 99% identical to VLa or VLb; (e) panning against said second antigen; (f) selecting a common light chain variable region (VLc) and a third heavy chain variable region (VHc), wherein the VHa-VLc binds to the first antigen with the desired affinity and the VHc-VLc binds to the second antigen with the desired affinity; (g) redesigning the sequences of VHa and VHc to obtain VHa' and VHc' that increase the difference in biochemical or biophysical properties between a first protein comprising two polypeptides each comprising a VHa' and two polypeptides each comprising a VLc, and a second protein comprising two polypeptides each comprising a VHc' and two polypeptides each comprising a VLc; and (h) producing a bispecific antibody or antigen-binding fragment thereof having two light chain variable regions and two heavy chain variable regions, wherein each of the two light chain variable regions comprises a VLc, and the two heavy chain variable regions comprise a VHa' and a VHc', respectively. [The present invention 1026] The method of claim 1025, wherein in step (d), the plurality of light chain variable regions is generated by error-prone PCR. [The present invention 1027] The method of the present invention 1025 further includes: (i) creating a buffer system for purifying said bispecific antibody or antigen-binding fragment thereof. [The present invention 1028] 1. A method of making a bispecific antibody or antigen-binding fragment thereof, comprising: (a) selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment thereof that binds to the first antigen and a second antibody or antigen-binding fragment thereof that binds to the second antigen, wherein the first antibody or antigen-binding fragment thereof comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment thereof comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); (b) determining the amino acid sequences of VHa, VLa, VHb, and VLb; (c) aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is less than 80%; (d) replacing all light chain variable regions in the phage display antibody library with a plurality of light chain variable regions, wherein the light chain variable regions are at least 80%, 85%, 90%, 95%, or 99% identical to VLa or VLb; (e) panning against said second antigen; (f) selecting a common light chain variable region (VLc) and a third heavy chain variable region (VHc), wherein the VHc-VLc binds to the second antigen with a desired affinity; (g) determining that the homology between VLa and VLc is greater than 80%; (h) designing a common light chain variable region (VLd), wherein the VLd maintains affinity for the first antigen when bound to the VHa and has a desired affinity for the second antigen when bound to the VHc; (i) optionally, redesigning the sequences of VHa and VHc to obtain VHa' and VHc' that increase the difference in biochemical or biophysical properties between a first protein comprising two polypeptides each comprising a VHa' and two polypeptides each comprising a VLd, and a second protein comprising two polypeptides each comprising a VHc' and two polypeptides each comprising a VLd; and (j) optionally, generating a bispecific antibody or antigen-binding fragment thereof having two light chain variable regions and two heavy chain variable regions, wherein each of the two light chain variable regions comprises a VLd and the two heavy chain variable regions comprise a VHa' and a VHc', respectively. [The present invention 1029] 1. A method of making a bispecific antibody or antigen-binding fragment thereof, comprising: (a) selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment thereof that binds to the first antigen and a second antibody or antigen-binding fragment thereof that binds to the second antigen, wherein the first antibody or antigen-binding fragment thereof comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment thereof comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); (b) determining the amino acid sequences of VHa, VLa, VHb, and VLb; (c) aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is greater than 80%; (d) designing a common light chain variable region (VLc), wherein the VLc maintains affinity for the first antigen when combined with the VHa; and (e) optionally, generating a bispecific antibody or antigen-binding fragment thereof having two light chain variable regions and two heavy chain variable regions, wherein each of the two light chain variable regions comprises a VLc, and the two heavy chain variable regions comprise a VHa and a VHb, respectively. [The present invention 1030] 1. A method of making a bispecific antibody or antigen-binding fragment thereof, comprising: (a) selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment thereof that binds to the first antigen and a second antibody or antigen-binding fragment thereof that binds to the second antigen, wherein the first antibody or antigen-binding fragment thereof comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment thereof comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); (b) determining the amino acid sequences of VHa, VLa, and VLb; (c) aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is less than 80%; (d) replacing all light chain variable regions in the phage display antibody library with VLa and panning against a second antigen to obtain a third heavy chain variable region (VHc); and (e) optionally, generating a bispecific antibody or antigen-binding fragment thereof having two light chain variable regions and two heavy chain variable regions, wherein each of the two light chain variable regions comprises a VLa, and the two heavy chain variable regions comprise a VHa and a VHc, respectively. [The present invention 1031] 1. A method of making a bispecific antibody or antigen-binding fragment thereof, comprising: (a) selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment thereof that binds to the first antigen and a second antibody or antigen-binding fragment thereof that binds to the second antigen, wherein the first antibody or antigen-binding fragment thereof comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment thereof comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); (b) determining the amino acid sequence of VHa, VLa, VHb, and / or VLb; (c) aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is less than 80%; (d) replacing all light chain variable regions in the phage display antibody library with a plurality of light chain variable regions, wherein the light chain variable regions are at least 80%, 85%, 90%, 95%, or 99% identical to VLa or VLb; (e) panning against the first and / or second antigen; (f) selecting a common light chain variable region (VLc) and a third heavy chain variable region (VHc), wherein the VHa-VLc binds to the first antigen with the desired affinity and the VHc-VLc binds to the second antigen with the desired affinity; and (g) optionally, generating a bispecific antibody or antigen-binding fragment thereof having two light chain variable regions and two heavy chain variable regions, wherein each of the two light chain variable regions comprises a VLc, and the two heavy chain variable regions comprise a VHa and a VHc, respectively. [The present invention 1032] a heavy chain variable region (VH) comprising complementarity determining regions (CDRs) 1, 2, and 3, wherein the VH CDR1 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR1 amino acid sequence, the VH CDR2 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR2 amino acid sequence, and the VH CDR3 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR3 amino acid sequence; and a light chain variable region (VL) comprising CDR1, 2, and 3, wherein the VL CDR1 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR1 amino acid sequence, the VL CDR2 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR2 amino acid sequence, and the VL CDR3 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR3 amino acid sequence. An antibody or antigen-binding fragment thereof that binds to CD3, comprising: the amino acid sequences of the selected VH CDR1, 2, and 3 are set forth in SEQ ID NOs: 22 to 24, respectively, and the amino acid sequences of the selected VL CDR1, 2, and 3 are set forth in SEQ ID NOs: 28 to 30, respectively; The antibody or antigen-binding fragment thereof. [The present invention 1033] The antibody or antigen-binding fragment thereof of the present invention 1032, wherein VH comprises CDR1, 2, and 3 having the amino acid sequences set forth in SEQ ID NOs: 22, 23, and 24, respectively, and VL comprises CDR1, 2, and 3 having the amino acid sequences set forth in SEQ ID NOs: 28, 29, and 30, respectively. [The present invention 1034] Any one of the antibodies 1032 to 1033 of the present invention or an antigen-binding fragment thereof, which specifically binds to human CD3. [This invention 1035] Any one of the antibodies 1032 to 1034 of the present invention or an antigen-binding fragment thereof, which is a bispecific antibody. [The present invention 1036] a heavy chain variable region (VH) comprising complementarity determining regions (CDRs) 1, 2, and 3, wherein the VH CDR1 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR1 amino acid sequence, the VH CDR2 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR2 amino acid sequence, and the VH CDR3 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR3 amino acid sequence; and a light chain variable region (VL) comprising CDR1, 2, and 3, wherein the VL CDR1 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR1 amino acid sequence, the VL CDR2 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR2 amino acid sequence, and the VL CDR3 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR3 amino acid sequence. 1. An antibody or antigen-binding fragment thereof that binds to PD-L1, comprising: the selected VH CDR1, 2, and 3 amino acid sequences and the selected VL CDR1, 2, and 3 amino acid sequences are (1) The amino acid sequences of the selected VH CDR1, 2, and 3 are set forth in SEQ ID NOs: 41 to 43, respectively, and the amino acid sequences of the selected VL CDR1, 2, and 3 are set forth in SEQ ID NOs: 53 to 55, respectively; (2) The amino acid sequences of the selected VH CDR1, 2, and 3 are set forth in SEQ ID NOs: 41 to 43, respectively, and the amino acid sequences of the selected VL CDR1, 2, and 3 are set forth in SEQ ID NOs: 59 to 61, respectively. One of the The antibody or antigen-binding fragment thereof. [This invention 1037] The antibody or antigen-binding fragment thereof of the present invention 1036, wherein VH comprises CDR1, 2, and 3 having the amino acid sequences set forth in SEQ ID NOs: 41 to 43, respectively, and VL comprises CDR1, 2, and 3 having the amino acid sequences set forth in SEQ ID NOs: 59 to 61, respectively. [The present invention 1038] Any one of the antibodies 1036 to 1037 of the present invention or an antigen-binding fragment thereof, which specifically binds to human CD3. [This invention 1039] Any one of the antibodies 1036 to 1038 of the present invention or an antigen-binding fragment thereof, which is a bispecific antibody. [The present invention 1040] a heavy chain variable region (VH) comprising complementarity determining regions (CDRs) 1, 2, and 3, wherein the VH CDR1 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR1 amino acid sequence, the VH CDR2 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR2 amino acid sequence, and the VH CDR3 region comprises an amino acid sequence that is at least 80% identical to a selected VH CDR3 amino acid sequence; and a light chain variable region (VL) comprising CDR1, 2, and 3, wherein the VL CDR1 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR1 amino acid sequence, the VL CDR2 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR2 amino acid sequence, and the VL CDR3 region comprises an amino acid sequence that is at least 80% identical to a selected VL CDR3 amino acid sequence. An antibody or antigen-binding fragment thereof that binds to CD55, comprising: the selected VH CDR1, 2, and 3 amino acid sequences and the selected VL CDR1, 2, and 3 amino acid sequences are (1) The amino acid sequences of the selected VH CDR1, 2, and 3 are set forth in SEQ ID NOs: 47 to 49, respectively, and the amino acid sequences of the selected VL CDR1, 2, and 3 are set forth in SEQ ID NOs: 53 to 55, respectively; (2) The amino acid sequences of the selected VH CDR1, 2, and 3 are set forth in SEQ ID NOs: 47 to 49, respectively, and the amino acid sequences of the selected VL CDR1, 2, and 3 are set forth in SEQ ID NOs: 59 to 61, respectively. One of the The antibody or antigen-binding fragment thereof. [This invention 1041] The antibody or antigen-binding fragment thereof of the present invention 1040, wherein VH comprises CDR1, 2, and 3 having the amino acid sequences set forth in SEQ ID NOs: 47 to 49, respectively, and VL comprises CDR1, 2, and 3 having the amino acid sequences set forth in SEQ ID NOs: 59 to 61, respectively. [The present invention 1042] Any one of the antibodies 1040 to 1041 of the present invention or an antigen-binding fragment thereof, which specifically binds to human CD3. [This invention 1043] Any one of the antibodies 1040 to 1042 of the present invention or an antigen-binding fragment thereof, which is a bispecific antibody. [This invention 1044] An immunoglobulin light chain or fragment thereof comprising a VL comprising CDR1, 2, and 3 comprising the amino acid sequences set forth in SEQ ID NOs: 53 to 55, respectively. A nucleic acid comprising a polynucleotide encoding a polypeptide comprising: the VL binds to PD-L1 when paired with a VH comprising the amino acid sequence set forth in SEQ ID NO:4, and / or binds to CD55 when paired with a VH comprising the amino acid sequence set forth in SEQ ID NO:5. The nucleic acid. [This invention 1045] An immunoglobulin light chain or fragment thereof comprising a VL comprising CDR1, 2, and 3 comprising the amino acid sequences set forth in SEQ ID NOs: 59 to 61, respectively. A nucleic acid comprising a polynucleotide encoding a polypeptide comprising: the VL binds to PD-L1 when paired with a VH comprising the amino acid sequence set forth in SEQ ID NO:4, and / or binds to CD55 when paired with a VH comprising the amino acid sequence set forth in SEQ ID NO:5. The nucleic acid. [The present invention 1046] A nucleic acid according to any one of 1044 to 1045 of the present invention, which encodes a bispecific antibody. [This invention 1047] The nucleic acid of any one of 1044 to 1045 of the present invention, which is a cDNA. [This invention 1048] A vector comprising one or more of the nucleic acids of any one of 1044 to 1047 of the present invention. [This invention 1049] A cell containing the vector of the present invention. [The present invention 1050] The cell of the present invention 1049, which is a CHO cell. [This invention 1051] A cell containing one or more of the nucleic acids of any one of 1044 to 1047 of the present invention. [This invention 1052] a first polypeptide comprising a first heavy chain variable region (VH) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:1; a second polypeptide comprising a second heavy chain variable region (VH) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:2; a third polypeptide comprising a first light chain variable region (VL) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:3; a fourth polypeptide comprising a second light chain variable region (VL) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:3. A bispecific antibody or antigen-binding fragment thereof that binds to CD20 and CD3, comprising: [This invention 1053] the first heavy chain variable region (VH) comprises SEQ ID NO:1; the second heavy chain variable region (VH) comprises SEQ ID NO:2; the first light chain variable region (VL) comprises SEQ ID NO:3; and the second light chain variable region (VL) comprises SEQ ID NO:3; The bispecific antibody or antigen-binding fragment thereof of the present invention. [This invention 1054] the first polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:34, 35, or 36; the second polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:37, 38, or 39; the third polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:40; and the fourth polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:40; The bispecific antibody or antigen-binding fragment thereof of the present invention. [This invention 1055] the first polypeptide comprises the amino acid sequence set forth in SEQ ID NO:35; and the second polypeptide comprises the amino acid sequence set forth in SEQ ID NO:38; The bispecific antibody or antigen-binding fragment thereof of the present invention. [This invention 1056] a first polypeptide comprising a first heavy chain variable region (VH) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:4; a second polypeptide comprising a second heavy chain variable region (VH) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:5; a third polypeptide comprising a first light chain variable region (VL) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 6 or 7; a fourth polypeptide comprising a second light chain variable region (VL) comprising an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 6 or 7. A bispecific antibody or antigen-binding fragment thereof that binds to PD-L1 and CD55, comprising: [This invention 1057] the first heavy chain variable region (VH) comprises SEQ ID NO:4; the second heavy chain variable region (VH) comprises SEQ ID NO:5; the first light chain variable region (VL) comprises SEQ ID NO:7; and the second light chain variable region (VL) comprises SEQ ID NO:7; The bispecific antibody or antigen-binding fragment thereof of the present invention. [This invention 1058] the first polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:65; the second polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO:66; the third polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 67 or 68; and the fourth polypeptide comprises an amino acid sequence at least 90%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 67 or 68; The bispecific antibody or antigen-binding fragment thereof of the present invention. [This invention 1059] the first polypeptide comprises the amino acid sequence set forth in SEQ ID NO:65; the second polypeptide comprises the amino acid sequence set forth in SEQ ID NO:66; the third polypeptide comprises the amino acid sequence set forth in SEQ ID NO:68; and the fourth polypeptide comprises the amino acid sequence set forth in SEQ ID NO:68; The bispecific antibody or antigen-binding fragment thereof of the present invention. [The present invention 1060] An antibody-drug conjugate comprising an antibody or antigen-binding fragment thereof of any of inventions 1001-1019, 1032-1043, and 1052-1059, covalently linked to a therapeutic agent. [The present invention 1061] The antibody-drug conjugate of the present invention 1060, wherein said therapeutic agent is a cytotoxic agent or a cytostatic agent. [The present invention 1062] A method for treating a subject suffering from cancer, comprising administering to the subject a therapeutically effective amount of a composition comprising any one of the antibodies or antigen-binding fragments thereof of 1001 to 1019, 1032 to 1043, and 1052 to 1059 of the present invention, or the antibody-drug conjugate of 1060 or 1061 of the present invention. [The present invention 1063] The method of claim 1062, wherein the subject has a solid tumor. [This invention 1064] 1063. The method of claim 1062, wherein said cancer is melanoma, pancreatic cancer, or a hematological malignancy. [This invention 1065] 1063. The method of claim 1062, wherein said cancer is non-Hodgkin's lymphoma, lymphoma, or chronic lymphocytic leukemia. [The present invention 1066] A method for reducing the rate of tumor growth, comprising contacting tumor cells of a subject with an effective amount of a composition comprising any one of antibodies or antigen-binding fragments thereof of 1001 to 1019, 1032 to 1043, and 1052 to 1059 of the present invention, or an antibody-drug conjugate of 1060 or 1061 of the present invention. [This invention 1067] A method for killing tumor cells, comprising contacting tumor cells of a subject with an effective amount of a composition comprising any one of antibodies or antigen-binding fragments thereof of 1001 to 1019, 1032 to 1043, and 1052 to 1059 of the present invention, or an antibody-drug conjugate of 1060 or 1061 of the present invention. [The present invention 1068] A pharmaceutical composition comprising any one of the antibodies 1001 to 1019, 1032 to 1043, and 1052 to 1059 of the present invention or an antigen-binding fragment thereof, and a pharmaceutically acceptable carrier. [The present invention 1069] A pharmaceutical composition comprising the antibody-drug conjugate of the present invention 1060 or 1061 and a pharmaceutically acceptable carrier. Other features and advantages of the invention will be apparent from the following detailed description and drawings, and from the claims. [Brief explanation of the drawings]

[0070] [Figure 1] Figure 1A is a graph showing that redesigned antibody A from Example 1 binds to Raji cells expressing CD20. Figure 1B is a graph showing that redesigned antibody B from Example 1 binds to Jukart cells expressing CD3. [Figure 2A] CD20+ Raji cell binding assay results. [Figure 2B] Results of CD3+ Jurkat cell binding assay. [Figure 3] T cell activation assay (CD20 / 3 in the figure is CD20 / CD3 BsMab. A10 and A11 indicate different elution fractions. The isotype is IgG1 antibody, which was used as a control). [Figure 4] Titration curves of T cell activation for different antibodies. [Figure 5] Antibody-mediated CD20+ Raji cell killing in the presence of peripheral blood mononuclear cells (PBMCs). [Figure 6] Antibody-mediated CD20+ Raji cell killing in the presence of PBMCs, where T cells were activated with IL-2 and CD3 / CD28 beads for 4 days. [Figure 7] Antibody-mediated CD20+ Raji cell killing in the presence of PBMCs, where T cells were activated with IL-2 and CD3 / CD28 beads for 7 days. [Figure 8] Antibody-mediated CD3+ Jurkat cell killing in the presence of PBMCs. [Figure 9] Antibody-mediated CD3+ Jurkat cell killing in the presence of PBMCs, where T cells were activated with IL-2 and CD3 / CD28 beads for 4 days. [Figure 10] Antibody-mediated CD3+ Jurkat cell killing in the presence of PBMCs, where T cells were activated with IL-2 and CD3 / CD28 beads for 7 days. [Figure 11] Antibody-induced depletion of activated T cells in PBMCs. [Figure 12] Antibody-induced depletion of inactivated T cells within PBMCs. [Figure 13] Complement-dependent cytotoxicity (CDC)-mediated Raji cell lysis measured by FACS. [Figure 14] CDC-mediated Raji cell lysis determined by calcein release. [Figure 15] CDC-mediated Jurkat cell lysis measured by FACS (A11 and B3 are different elution fractions). [Figure 16] Depletion of T cells within PBMCs in the presence of human complement-enriched serum. [Figure 17] Lysis of rituximab-resistant cells mediated by T cell activation based on PBMCs from three different donors. [Figure 18] Lysis of rituximab-resistant cells mediated by T cell activation based on PBMCs from three different donors. [Figure 19] Lysis of rituximab-resistant cells mediated by T cell activation based on PBMCs from three different donors. [Figure 20] T cell activation to assess purified antibodies. [Figure 21A] Average body weight of mice in each group after injection with phosphate buffered saline PBS (G1), CD20 / CD3 BsMab (G2, "BIS") or rituximab (G3, "RTX"). [Figure 21B] Average imaging intensity of luciferase-labeled Raji cells in each group after injection with phosphate buffered saline PBS (G1), CD20 / CD3 BsMab (G2, "BIS"), or rituximab (G3, "RTX"). [Figure 22]Figure 22A: Reduced capillary electrophoresis sodium dodecyl sulfate (Re-CE-SDS) results for purified CD20 / CD3 bispecific antibody samples. Figure 22B: Non-reduced CE (Non-Re-CE-SDS) results for purified CD20 / CD3 bispecific antibody samples. [Figure 23] Figure 23A: Binding affinity for avelumab (PD-L1 wt) and a designed PD-L1 single dimeric IgG antibody (PD-L1 V1) comprising a VHa (SEQ ID NO:4) and a common VL (SEQ ID NO:6) for PD-L1. Figure 23B: Binding affinity for the parent anti-CD55 antibody (CD55 wt) and a designed CD55 single dimeric IgG antibody (CD55 V1) comprising a VHb (SEQ ID NO:5) and a common VL (SEQ ID NO:6) for CD55. [Figure 24] Alignment of the common light chain for BsMab v1 (SEQ ID NO:67) and the common light chain for BsMab v2 (SEQ ID NO:68). [Figure 25] Figure 25A: Binding affinity for avelumab (PD-L1 wt) and a redesigned PD-L1 single dimeric IgG antibody (PD-L1 V2) comprising a VHa for PD-L1 (SEQ ID NO:4) and a common VL v2 (SEQ ID NO:7). Figure 25B: Binding affinity for the parent anti-CD55 antibody (CD55 wt) and a redesigned CD55 single dimeric IgG antibody (CD55 V2) comprising a VHb for CD55 (SEQ ID NO:5) and a common VL v2 (SEQ ID NO:7). [Figure 26A] Antibody-mediated CDC in MDA231 cells. [Figure 26B] Antibody-mediated CDC in MDA231 cells. [Figure 27A] Results of antibody internalization assay using MDA231 cells. [Figure 27B] Results of antibody internalization assay using SIHA cells. [Figure 28]FIG. 28 is a schematic diagram showing how a bispecific antibody that binds to CD3 and a cancer antigen (eg, a cancer-specific antigen) recognizes and kills tumor cells. [Figure 29] FIG. 29 is a schematic diagram showing how bispecific antibodies that bind to cancer-specific and cancer-associated antigens recognize and kill tumor cells. DETAILED DESCRIPTION OF THE INVENTION

[0071] Detailed Description A bispecific antibody or antigen-binding fragment thereof is an artificial protein that can simultaneously bind to two different antigens. In some embodiments, a bispecific antibody or antigen-binding fragment thereof can have two arms (arms A and B). Each arm contains one heavy chain variable region and one light chain variable region.

[0072] Bispecific antibodies or antigen-binding fragments thereof can be IgG-like or non-IgG-like. An IgG-like bispecific antibody can have two Fab arms and one Fc region, with the two Fab arms binding to different antigens. A non-IgG-like bispecific antibody or antigen-binding fragment can be, for example, a chemically linked Fab (e.g., the two Fab regions are chemically linked) or a single-chain variable fragment (scFV). For example, an scFV can have two heavy chain variable regions and two light chain variable regions.

[0073] In an unbalanced bispecific antibody or antigen-binding fragment thereof, the two arms (arms: A and B) or two antigen-binding regions (antigen-binding regions: A and B) can bind to their respective target antigens with different affinities. The binding affinity can be expressed by the association constant (Ka) as follows: Ka = [antibody-antigen] / [antibody] [antigen]

[0074] High affinity antibodies are usually 10 7 M -1 The Ka for one arm or one antigen-binding region is 105 M -1 , 10 6 M -1 , 10 7 M -1 , 10 8 M -1 , 10 9 M -1 , 10 10 M -1 , 10 11 M -1 or 10 12 M -1 In some embodiments, Ka can be greater than 10 5 M -1 , 10 6 M -1 , 10 7 M -1 , 10 8 M -1 , 10 9 M -1 , 10 10 M -1 , 10 11 M -1 or 10 12 M -1 It may be lower than

[0075] The binding affinity (A) of the first arm or first antigen-binding region can be higher than the binding affinity (B) of the second arm or second antigen-binding region. Bispecific antibodies with unequal affinities can have various advantages. For example, bispecific antibodies with unequal affinities can be used to target a cancer-specific antigen on cancer cells and CD3 on T cells. In this case, the high affinity for the cancer-specific antigen can result in better capture of cancer cells by T cells, while the low affinity for CD3 can avoid triggering T cell signals by CD3 (Figure 28). Only when the bispecific antibody is presented to T cells in a more multivalent manner can the T cells be activated and kill the target cancer cells. Furthermore, bispecific antibodies with unequal affinities can be used to target a cancer-specific antigen and a cancer-associated antigen (Figure 29). In this case, the bispecific antibody binds only weakly to non-cancerous cells that express low levels of the cancer-associated antigen, but binds strongly to cancer cells that express both the cancer-specific antigen and high levels of the cancer-associated antigen.

[0076] For bispecific antibodies with unequal affinities, the Ka (A) of the first arm or first antigen-binding region is 10 7 M -1 , 10 8 M -1 , 10 9 M -1 , 10 10 M -1 , 10 11 M -1 or 10 12 M -1 In some embodiments, the Ka(A) of the first arm or first antigen-binding region may be 10-fold, 100-fold, 1000-fold, 10,000-fold, or 100,000-fold higher than the Ka(B) of the second arm or second antigen-binding region. Thus, in some embodiments, the Ka(B) of the second arm or second antigen-binding region may be 10 5 M -1 , 10 6 M -1 , 10 7 M-1 , 10 8 M -1 or 10 9 M -1 In some embodiments, the K a (B) of the second arm or second antigen-binding region may be lower than or equal to 10 while still maintaining reasonable affinity (e.g., 10 4 M -1 , 10 5 M -1 or 10 6 M -1 specifically binds to the target antigen at a specific binding site (higher than

[0077] Binding affinity can also be expressed by the dissociation constant (Kd). Kd = [antibody] [antigen] / [antibody-antigen]

[0078] Kd is 10 -5 M, 10 -6 M, 10 -7 M, 10 -8 M, 10 -9 M, 10 -10 M, 10 -11 M or 10 -12 In some embodiments, the Kd can be less than 10 -5 M, 10 -6 M, 10 -7 M, 10 -8 M, 10 -9 M, 10 -10 M, 10 -11 M or 10 -12 It can be higher than M.

[0079] In some embodiments, the binding affinity (A) of the first arm or first antigen-binding region can be higher than the binding affinity (B) of the second arm or second antigen-binding region. For example, the Kd(B) of the second arm or second antigen-binding region can be 10-fold, 100-fold, 1000-fold, 10,000-fold, or 100,000-fold higher (and therefore have lower affinity) than the Kd(A) of the first arm or first antigen-binding region. Thus, in some embodiments, the Kd(A) of the first arm or first antigen-binding region can be 10-fold, 100-fold, 10000-fold, 100,000-fold, or 100,000-fold higher (and therefore have lower affinity).-7 M, 10 -8 M, 10 -9 M, 10 -10 M, 10 -11 M or 10 -12 The Kd(B) of the second arm or second antigen-binding region may be lower than 10 -5 M, 10 -6 M, 10 -7 M, 10 -8 M or 10 -9 It can be higher than M.

[0080] In some embodiments, a bispecific antibody or antigen-binding fragment thereof comprises two light chains and two heavy chains. Each of the two light chains has one light chain variable region (VL) and one light chain constant region (CL). Each of the two heavy chains has one heavy chain variable region (VH) and three heavy chain constant regions (CH1, CH2, CH3). In some embodiments, the two light chains in arm A and arm B are the same. Therefore, the CDRs in the VLs of the two light chains can be the same. In some embodiments, the two heavy chains in a bispecific antibody or antigen-binding fragment thereof are different. Therefore, the CDRs in the VHs of the two heavy chains are different.

[0081] Various methods can be used to ensure that identical heavy chains do not bind to each other when creating bispecific antibodies. For example, the "knobs-into-holes" method involves introducing mutations of amino acids with large side chains into one heavy chain and mutations of amino acids with small side chains into the other heavy chain. Thus, identical heavy chains are less likely to bind to each other, and two different heavy chains are more likely to bind to each other. The "knobs-into-holes" method is described, for example, in Ridgway, John B.B., Leonard G. Presta, and Paul Carter, "'Knobs-into-holes' engineering of antibody CH3 domains for heavy chain heterodimerization." Protein Engineering, Design and Selection, 9.7 (1996), which is incorporated herein by reference in its entirety.

[0082] Unbalanced bispecific antibodies that bind to T cell-specific antigens and cancer antigens Bispecific antibodies (BsAbs or BsMabs) with T cell-specific antigen (e.g., CD3, CD4, or CD8)-binding arms capable of recruiting and activating T cells have been widely investigated for cancer therapy. However, many effector functions of such bispecific antibodies have been eliminated due to safety concerns. Because antibody effector functions, such as ADCC and CDC, have been shown to play important roles in cancer cell killing, "safely" maintaining antibody effector functions could improve the antibody's cancer-killing capabilities and expand the mechanism of action of therapeutic antibodies. To "safely" maintain effector functions and expand the application of such bispecific antibodies, an imbalanced bispecific antibody technology platform based on computational antibody design has been developed.

[0083] In this design, the first antigen-binding region targets a cancer-specific antigen, and the second antigen-binding region targets a T cell-specific antigen (e.g., CD3, CD4, or CD8) to generate T cells to attack cancer with the cancer-specific antigen (Figure 28).

[0084] As used herein, the term "cancer-specific antigen" refers to an antigen that is specifically expressed on the surface of cancer cells. These antigens can be used to identify tumor cells. Normal cells rarely express cancer-specific antigens. Some typical cancer-specific antigens include, for example, CD20, PSA, PSCA, PD-L1, Her2, Her3, Her1, β-catenin, CD19, CEACAM3, EGFR, c-Met, EPCAM, PSMA, CD40, MUC1, and IGF1R. PSA is primarily expressed on prostate cancer cells, and Her2 is primarily expressed on breast cancer cells.

[0085] Bispecific antibodies that bind to CD20 and CD3 are described herein. These bispecific antibodies can be used to target multiple CD20-positive cancers, such as CD20-positive non-Hodgkin's lymphoma (NHL), and thus can be used to treat a subject's non-Hodgkin's lymphoma. Because bispecific antibodies accommodate a different mechanism of action for treating cancer compared to therapeutic antibodies targeting CD20 alone, they can be used as complementary treatments for CD20-positive cancers, particularly those that do not respond well to current CD20 therapies (e.g., rituximab-resistant NHL).

[0086] Antibodies with high affinity for CD3 may induce T cell signaling and cause unwanted immune responses. Therefore, to "safely" maintain the effector function of the antibody while reducing the risk of T cell signaling by CD3, antibodies with low affinity for CD3 (e.g., Ka of 10) are recommended. 5 M -1 , 10 6 M -1 or 10 7 M -1A multivalent binding domain (which may be less than 1000 uM) is required. As used herein, the term "safely maintaining antibody effector function" means that the antibody does not induce ADCC or CDC on normal cells (e.g., non-cancer cells). When multiple bispecific antibodies are presented on target cancer cells (e.g., in clusters) and bridge the interaction between the cancer cells and T cells, these bispecific antibodies can induce T cell signals via CD3 in a multivalent form, and the activated T cells kill the target cancer cells.

[0087] Thus, disclosed herein is a bispecific antibody or antigen-binding fragment thereof comprising two heavy chain variable regions and two light chain variable regions, wherein a first heavy chain variable region comprises a sequence at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:1, a second heavy chain variable region comprises a sequence at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:2, and the first and second light chain variable regions comprise a sequence at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:3.

[0088] In some embodiments, the CDR sequences for binding to CD20 comprise the heavy chain variable domain CDRs of SEQ ID NOs:16-18 and the light chain variable domain CDRs of SEQ ID NOs:28-30 (as defined by Kabat numbering). In Chothia numbering, the heavy chain variable domain CDR sequences are set forth in SEQ ID NOs:19-21 and the light chain variable domain CDRs are set forth in SEQ ID NOs:31-33.

[0089] In some embodiments, the CDR sequences for binding to CD3 comprise the heavy chain variable domain CDRs of SEQ ID NOs:22-24 and the light chain variable domain CDRs of SEQ ID NOs:28-30 (as defined by Kabat numbering). In Chothia numbering, the heavy chain variable domain CDR sequences are set forth in SEQ ID NOs:25-27 and the light chain variable domain CDRs are set forth in SEQ ID NOs:31-33.

[0090] In some embodiments, the bispecific antibody or antigen-binding fragment thereof comprises: a first heavy chain amino acid sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:34, 35 or 36; a second heavy chain amino acid sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:37, 38 or 39; a first light chain amino acid sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:40; and a second light chain amino acid sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:41. and a second light chain amino acid sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to NO: 40. In some embodiments, the first light chain amino acid sequence and the second light chain amino acid sequence are identical.

[0091] Unbalanced bispecific antibodies that bind to cancer-specific and cancer-associated antigens Also disclosed herein are unbalanced bispecific antibodies having a first antigen-binding region that targets a cancer-specific antigen and a second antigen-binding region that targets a cancer-associated antigen.

[0092] As used herein, the term "cancer-associated antigen" refers to an antigen that is expressed at relatively high levels on cancer cells but may also be expressed at relatively low levels on normal cells. CD55, CD59, CD46, and various adhesion molecules, such as N-cadherin, VE-cadherin, NCAM, Mel-CAM, ICAM, NrCAM, VCAM1, ALCAM, and MCAM, are cancer-associated antigens. While both cancer-specific and cancer-associated antigens are expressed on the surface of cancer cells, the difference between cancer-specific and cancer-associated antigens is that cancer-associated antigens are also expressed on normal cells, but at relatively lower levels than on cancer cells. In contrast, cancer-specific antigens are rarely expressed on normal cells, and even when expressed on normal cells, the amount is extremely low. Antibodies targeting cancer-specific antigens typically do not induce antibody-dependent cellular cytotoxicity (ADCC) or complement-dependent cytotoxicity (CDC) on normal cells. In contrast, antibodies targeting cancer-associated cells with high affinity can cause cytotoxic effects on normal cells. It is therefore important that the bispecific antibody binds to the cancer-associated antigen with relatively low affinity (Figure 29).

[0093] A bispecific antibody that binds to PD-L1 and CD55 is described in the Examples. This antibody can be used to treat subjects with PD-L1 and CD55-positive cancers through ADCC or CDC, as well as by blocking PD-L1 / PD1 interaction to activate T cell-dependent immune responses and reduce CD55 suppression of CDC. Furthermore, because cancer cells can become resistant to PD-L1 antibodies, binding between the second arm and CD55 on cancer cells can provide an additional therapeutic effect.

[0094] Accordingly, the present disclosure relates to a bispecific antibody or antigen-binding fragment thereof comprising two heavy chain variable regions and two light chain variable regions, wherein a first heavy chain variable region comprises a sequence at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:4, the second heavy chain variable region comprises a sequence at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:5, and the first and second light chain variable regions comprise a sequence at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:6 or 7.

[0095] In some embodiments, the CDR sequences for binding to PD-L1 comprise the CDRs of the heavy chain variable domain (SEQ ID NOs:41-43) and the CDRs of the light chain variable domain (SEQ ID NOs:53-55 or 59-61) (as defined by Kabat numbering). In Chothia numbering, the CDR sequences of the heavy chain variable domain are set forth in SEQ ID NOs:44-46, and the CDRs of the light chain variable domain are set forth in SEQ ID NOs:56-58 or 62-64.

[0096] In some embodiments, the CDR sequences for binding to CD55 comprise the heavy chain variable domain CDRs of SEQ ID NOs:47-49 and the light chain variable domain CDRs of SEQ ID NOs:53-55 or 59-61 (as defined by Kabat numbering). In Chothia numbering, the heavy chain variable domain CDR sequences are set forth in SEQ ID NOs:50-52 and the light chain variable domain CDRs are set forth in SEQ ID NOs:56-58 or 62-64.

[0097] In some embodiments, the bispecific antibody or antigen-binding fragment thereof comprises a first heavy chain amino acid sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:65; a second heavy chain amino acid sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:66; a first light chain amino acid sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:67 or 68; and a second light chain amino acid sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:69 or 70. and a second light chain amino acid sequence that is at least 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to NO:67 or 68. In some embodiments, the first light chain amino acid sequence and the second light chain amino acid sequence are identical.

[0098] Creating unbalanced bispecific antibodies or antigens, or antigen-binding fragments thereof Bispecific antibodies or antigens, or antigen-binding fragments thereof, can be produced in the following manner: (1) Two target antigens are selected, and the sequences of the heavy chain variable region (VHa) and light chain variable region (VLa) of an antibody that binds to the first antigen (antibody A) are determined, and the sequences of the heavy chain variable region (VHb) and light chain variable region (VLb) of an antibody that binds to the second antigen (antibody B) are determined. (2) VLa and VLb are aligned, and if the sequence homology exceeds 80%, a consensus VL is designed using computer modeling tools (e.g., BioLuminate, Schordingerm, Cambridge, Massachusetts). During the design process, efforts are made to maintain the affinity of VLa, but this may result in some sacrifice of the affinity of VLb. The consensus VL can be VLa or VLb itself, or it can be a new VLc whose sequence shares high homology with VLa and VLb. The three-dimensional structures of VLa and VLb can be determined, for example, from structural modeling or crystal structures. This process can begin with the sequence of VLa. Based on the three-dimensional structure, amino acids in the light chain that are important for binding to the second antigen (e.g., when combined with VHb) but not involved in binding to the first antigen (e.g., when combined with VHa) can be identified, and amino acids in VLa can be changed to the corresponding amino acids in VLb. After repeating this process several times, a consensus VLc can be obtained. (3) If the homology between VLa and VLb is less than 80%, a human ScFV or Fab phage library is created by replacing the VL of an existing human naive ScFV library with the VL of antibody A, followed by error-prone PCR to induce less than 20% nucleotide mutations in the VL, and panning against the antigen of antibody B to obtain a new antibody B' having VLa or its homolog (more than 80% homology) as its VL. If the VL is not VLa but a VLa homolog (e.g., more than 80% homology), step (2) is repeated to design a consensus VL. (4) Using computer modeling tools, the sequences of VHa and VHb are redesigned to increase the difference in biochemical and biophysical properties (e.g., three-dimensional isoelectric point (PI)) between A and B. During this process, the affinity of A cannot be reduced, but the affinity of B can be reduced to a certain extent. (5) Create a buffer system and purify the imbalanced bispecific antibody.

[0099] The isoelectric point (PI) of a peptide is the pH at which, on statistical average, a particular molecule has no net charge. The amino acids that make up a peptide can be positive, negative, neutral, or polar in nature, and when combined, give the protein its overall charge. However, certain amino acids within a protein are buried within the protein and do not interact with the surrounding solution. 3D PI takes into account the three-dimensional structure of the protein and provides a better estimate of the pH value at which, on statistical average, the protein has no net charge when properly folded. (The inventors used gradient pH buffers from published literature; however, the buffers are not our invention; the purification process still needs to be optimized.)

[0100] In some embodiments, bispecific antibodies or antigens, or antigen-binding fragments thereof, can also be generated by the following methods: (a) selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment thereof that binds to the first antigen and a second antibody or antigen-binding fragment thereof that binds to the second antigen, wherein the first antibody or antigen-binding fragment thereof comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment thereof comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); (b) determining the amino acid sequences of VHa, VLa, and VLb; (c) aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is less than 80%; (d) replacing all light chain variable regions in the phage display antibody library with VLa and panning against a second antigen to obtain a third heavy chain variable region (VHc); (e) redesigning the sequences of VHa and VHc to obtain VHa' and VHc' that increase the difference in biochemical or biophysical properties between a first protein comprising two polypeptides each comprising a VHa' and two polypeptides each comprising a VLa, and a second protein comprising two polypeptides each comprising a VHc' and two polypeptides each comprising a VLa; and (f) producing a bispecific antibody or antigen-binding fragment thereof having two light chain variable regions and two heavy chain variable regions, wherein each of the two light chain variable regions comprises a VLa, and the two heavy chain variable regions comprise a VHa' and a VHc', respectively.

[0101] In some embodiments, bispecific antibodies or antigens, or antigen-binding fragments thereof, can also be generated by the following methods: (a) selecting a first antigen and a second antigen, and identifying a first antibody or antigen-binding fragment thereof that binds to the first antigen and a second antibody or antigen-binding fragment thereof that binds to the second antigen, wherein the first antibody or antigen-binding fragment thereof comprises a first heavy chain variable region (VHa) and a first light chain variable region (VLa), and the second antibody or antigen-binding fragment thereof comprises a second heavy chain variable region (VHb) and a second light chain variable region (VLb); (b) determining the amino acid sequences of VHa, VLa, VHb, and VLb; (c) aligning the amino acid sequences of VLa and VLb and determining that the sequence homology between VLa and VLb is less than 80%; (d) replacing all light chain variable regions in the phage display antibody library with a plurality of light chain variable regions, wherein the light chain variable regions are at least 80%, 85%, 90%, 95%, or 99% identical to VLa or VLb; (e) panning against the first and / or second antigen (e.g., the second antigen); (f) selecting a common light chain variable region (VLc) and a third heavy chain variable region (VHc), wherein the VHa-VLc binds to the first antigen with the desired affinity and the VHc-VLc binds to the second antigen with the desired affinity; (g) redesigning the sequences of VHa and VHc to obtain VHa' and VHc' that increase the difference in biochemical or biophysical properties between a first protein comprising two polypeptides each comprising a VHa' and two polypeptides each comprising a VLc, and a second protein comprising two polypeptides each comprising a VHc' and two polypeptides each comprising a VLc; and (h) producing a bispecific antibody or antigen-binding fragment thereof having two light chain variable regions and two heavy chain variable regions, wherein each of the two light chain variable regions comprises a VLc, and the two heavy chain variable regions comprise a VHa' and a VHc', respectively.

[0102] In some embodiments, if the VHa-VLc cannot bind to the first antigen with the desired affinity, additional steps can be taken. For example, if the VLc is at least 80% identical to the VLa, a new common light chain can be designed. In some embodiments, the process begins with the VLa, and amino acids can be mutated to amino acids within the VLc based on the methods described herein (e.g., based on the three-dimensional structures of the VLa and VLc).

[0103] In some embodiments, the design of a common light chain variable region involves aligning VLa and VLb and studying the different residues between VLa and VLb at the same Kabat position. If the different residue on VLb does not contact a CDR, an interface residue, a canonical residue, or a Vernier zone residue on the B Fv structure, the residue on VLb is mutated to the residue at the same Kabat position on VLa. Otherwise, the residue on VLb is retained.

[0104] In some embodiments, redesigning the heavy chain variable region involves calculating the 3D PI of Fv A and Fv B using BioLuminate and mutating non-CDR, non-canonical, non-interface, and non-Vernier zone residues to make Fvs with high 3D PI higher and Fvs with low 3D PI lower.

[0105] Instructions for using BioLuminate are described, for example, in the BioLuminate User Guide, which is incorporated herein by reference in its entirety.

[0106] Antibodies and antigen-binding fragments The present disclosure provides antibodies and antigen-binding fragments thereof comprising the complementarity-determining regions (CDRs), heavy chain variable regions, light chain variable regions, heavy chains, or light chains described in this disclosure. In some embodiments, the antibodies and antigen-binding fragments thereof are unbalanced bispecific antibodies and antigen-binding fragments thereof.

[0107] Generally, antibodies (also called immunoglobulins) are composed of two classes of polypeptide chains: light chains and heavy chains. Non-limiting antibodies in this disclosure can be intact, four-immunoglobulin chain antibodies consisting of two heavy chains and two light chains. The heavy chains of the antibodies can be of an isotype, including IgM, IgG, IgE, IgA, or IgD, or a subisotype, including IgG1, IgG2, IgG2a, IgG2b, IgG3, IgG4, IgE1, IgE2, etc. The light chains can be kappa or lambda light chains. An antibody can be composed of two identical copies of a light chain and / or two identical copies of a heavy chain, each containing a variable domain (or variable region, VH) and multiple constant domains (or constant regions), linked to each other via disulfide bonds within the constant domains to form the antibody "stem." Each light chain contains one variable domain (or variable region, VL) and one constant domain (or constant region), which are linked to one heavy chain via disulfide bonds. The variable region of each light chain is aligned with the variable region of the heavy chain to which it is linked. Both the light and heavy chain variable regions contain three hypervariable regions flanked by more conserved framework regions (FR).

[0108] These hypervariable regions, known as complementarity-determining regions (CDRs), form the loops that comprise the primary antigen-binding surface of an antibody. The four framework regions adopt a mostly β-sheet conformation, and the CDRs form loops that connect, or in some cases form part of, this β-sheet structure. The CDRs of each chain are held in close proximity by the framework regions and the CDRs of the other chain, contributing to the formation of the antigen-binding region.

[0109] Methods for identifying the CDR regions of an antibody by analyzing its amino acid sequence are well known, and a series of CDR definitions are commonly used: the Kabat definition is based on sequence variability, and the Chothia definition is based on the location of structural loop regions. These methods and definitions are described, for example, in Martin, "Protein sequence and structure analysis of antibody variable domains," Antibody engineering, Springer Berlin Heidelberg, 2001, pp. 422-439; Abhinandan, et al., "Analysis and improvements to Kabat and structurally correct numbering of antibody variable domains," Molecular immunology 45.14(2008):3832-3839; Wu, TT and Kabat, EA(1970) J. Exp. Med. 132:211-250; Martin et al., Methods Enzymol. 203:121-53(1991); Morea et al., Biophys Chem. 68(1-3):9-16(Oct. 1997); Morea et al., J. Mol. Biol. 275(2):269-94(Jan. 1998); Chothia et al., Nature 342(6252):877-83 (Dec. 1989); Ponomarenko and Bourne, BMC Structural Biology 7:64 (2007); the entirety of each cited reference is incorporated herein by reference. In this disclosure, Kabat numbering is used as the default unless otherwise indicated.

[0110] CDR is important for recognizing the epitope of an antigen. As used herein, "epitope" refers to the smallest part of a target molecule that has the ability to be specifically bound by the antigen-binding domain of an antibody. The minimum size of an epitope is generally 3, 4, 5, 6, or 7 amino acids, but these amino acids do not need to be in a continuous linear sequence in the primary structure of the antigen, because the epitope can be based on the three-dimensional configuration based on the secondary and tertiary structure of the antigen.

[0111] In some embodiments, these antibodies can be intact immunoglobulin molecules (e.g., IgG1, IgG2a, IgG2b, IgG3, IgM, IgD, IgE, IgA). The IgG subclasses (IgG1, IgG2, IgG3, and IgG4) are highly conserved and differ in their constant regions, particularly their hinge and upper CH2 domains. The sequences and differences of IgG subclasses are well known in the art and are described, for example, in Vidarsson, et al., "IgG subclasses and allotypes: from structure to effector functions," Frontiers in immunology 5 (2014); Irani, et al. "Molecular properties of human IgG subclasses and their implications for designing therapeutic monoclonal antibodies against infectious diseases," Molecular immunology 67.2 (2015):171-182; Shakib, Farouk, ed. The human IgG subclasses: molecular analysis of structure, function and regulation. Elsevier, 2016; the entire contents of each cited reference are incorporated herein by reference.

[0112] The antibody may be an immunoglobulin molecule derived from any species (e.g., human, rodent, mouse, rat, camelid). Antibodies disclosed herein include, but are not limited to, polyclonal, monoclonal, monospecific, multispecific antibodies, and chimeric antibodies, including those in which an immunoglobulin binding domain is fused to another polypeptide. The term "antigen-binding domain" or "antigen-binding fragment" refers to a portion of an antibody that retains the specific binding activity of the intact antibody, i.e., any portion of an antibody that has the ability to specifically bind to the epitope on the intact antibody's target molecule. These include, for example, Fab, Fab', F(ab')2, and variants of these fragments. Thus, in some embodiments, an antibody or antigen-binding fragment thereof can be, for example, an scFv, aFv, aFd, adAb, diabody, bispecific scFv, diabody, linear antibody, single-chain antibody molecule, multispecific antibody formed from antibody fragments, and any polypeptide that is or comprises a binding domain cognate with an antibody binding domain. Non-limiting examples of antigen-binding domains include, for example, intact antibody heavy and / or light chain CDRs, intact antibody heavy and / or light chain variable regions, intact antibody full-length heavy and / or light chains, or CDRs from either intact antibody heavy and / or light chains.

[0113] In some embodiments, the scFV has two heavy chain variable domains and two light chain variable domains, and in some embodiments, the scFV has two antigen-binding regions (antigen-binding regions: A and B), and these two antigen-binding regions can bind to their respective target antigens with different affinities.

[0114] In some embodiments, these antigen-binding fragments can form a portion of a chimeric antigen receptor (CAR). In some embodiments, the chimeric antigen receptor is a fusion of a single-chain variable fragment (scFv) disclosed herein fused to the CD3-zeta transmembrane and intracellular domains. In some embodiments, the chimeric antigen receptor also comprises intracellular signaling domains from various costimulatory receptors (e.g., CD28, 41BB, ICOS). In some embodiments, the chimeric antigen receptor comprises multiple signaling domains, e.g., CD3z-CD28-41BB or CD3z-CD28-OX40, in order of increasing strength. Thus, in one aspect, the present disclosure further provides a cell (e.g., a T cell) expressing a chimeric antigen receptor described herein.

[0115] In some embodiments, the antibody or antigen-binding fragment thereof is capable of binding to two different antigens or two different epitopes.

[0116] In some embodiments, the antibody or antigen-binding fragment thereof can comprise one, two, or three heavy chain variable region CDRs selected from Table 1, Table 2, Table 11, and Table 12. In some embodiments, the antibody or antigen-binding fragment thereof can comprise one, two, or three light chain variable region CDRs selected from Table 3, Table 13, and Table 14.

[0117] In some embodiments, the antibody A heavy chain variable region (VH) comprising complementarity determining regions (CDRs) 1, 2 and 3, wherein the CDR1 region comprises or consists of an amino acid sequence that is at least 80%, 85%, 90%, or 95% identical to the amino acid sequence of a selected VHCDR1, the CDR2 region comprises or consists of an amino acid sequence that is at least 80%, 85%, 90%, or 95% identical to the amino acid sequence of a selected VHCDR2, and the CDR3 region comprises or consists of an amino acid sequence that is at least 80%, 85%, 90%, or 95% identical to the amino acid sequence of a selected VHCDR3; and A light chain variable region (VL) comprising CDR1, 2 and 3, wherein the CDR1 region comprises or consists of an amino acid sequence that is at least 80%, 85%, 90%, or 95% identical to a selected VLCDR1 amino acid sequence, the CDR2 region comprises or consists of an amino acid sequence that is at least 80%, 85%, 90%, or 95% identical to a selected VLCDR2 amino acid sequence, and the CDR3 region comprises or consists of an amino acid sequence that is at least 80%, 85%, 90%, or 95% identical to the amino acid sequence of a selected VLCDR3. The amino acid sequences of the selected VH CDRs 1, 2, 3 are shown in Tables 1, 2, 11, and 12, and the amino acid sequences of the selected VL CDRs 1, 2, 3 are shown in Tables 3, 13, and 14.

[0118] In some embodiments, the antibodies or antigen-binding fragments described herein can comprise a heavy chain variable domain comprising one, two, or three CDRs selected from Table 1, Table 2, Table 11, and Table 12, and zero, one, or two amino acid insertions, deletions, or substitutions.

[0119] In some embodiments, the antibodies or antigen-binding fragments described herein can comprise a light chain variable domain comprising one, two, or three CDRs selected from Table 3, Table 13, and Table 14, and zero, one, or two amino acid insertions, deletions, or substitutions.

[0120] The insertions, deletions and substitutions can be within the CDR sequences or at one or both termini of the CDR sequences.

[0121] The Fc region of the antibodies and antibody fragments of the disclosure can be modified to provide desired effector functions or serum half-lives.

[0122] Antibody multimerization can be achieved through spontaneous antibody aggregation or through chemical or recombinant linkage techniques. For example, a small percentage of purified antibody preparations (e.g., purified IgG1 molecules) spontaneously form protein aggregates containing antibody homodimers and other higher order antibody multimers.

[0123] Any antibody or antigen-binding fragment described herein can be linked to a stabilizing molecule (e.g., a molecule that increases the half-life of the antibody or antigen-binding fragment thereof in a subject or solution). Non-limiting examples of stabilizing molecules include polymers (e.g., polyethylene glycol) or proteins (e.g., serum albumin, such as human serum albumin). Linking to a stabilizing molecule increases the half-life or extends the biological activity of the antibody or antigen-binding fragment in vitro (e.g., in tissue culture or when stored as a pharmaceutical composition) or in vivo (e.g., in humans).

[0124] In some embodiments, the antibodies or antigen-binding fragments (e.g., bispecific antibodies) described herein can be conjugated to a therapeutic agent. The antibody-drug conjugates comprising the antibodies or antigen-binding fragments thereof can be covalently or non-covalently bound to a therapeutic agent. In some embodiments, the therapeutic agent is a cytotoxic or cytostatic agent (e.g., cytochalasin B, gramicidin D, ethidium bromide, emytin, mitomycin, etoposide, teniposide, vincristine, vinblastine, colchicine, doxorubicin, daunorubicin, dihydroxyanthracin, maytansinoids such as DM-1 and DM-4, ​​dione, mitoxantrone, mithramycin, actinomycin D, 1-dehydrotestosterone, glucocorticoids, procaine, tetracaine, lidocaine, propranolol, puromycin, epirubicin, and cyclophosphamide and congeners).

[0125] Antibody properties The antibodies or antigen-binding fragments thereof (e.g., bispecific antibodies) described herein can increase an immune response. In some embodiments, the antibodies or antigen-binding fragments thereof described herein can increase the immune response, activity, or number of T cells (e.g., CD3+ cells, CD8+ and / or CD4+ cells) by at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, 2-fold, 3-fold, 5-fold, 10-fold, or 20-fold.

[0126] In some embodiments, the antibodies or antigen-binding fragments thereof described herein can reduce the activity or number of T cells by at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, 2-fold, 3-fold, 5-fold, 10-fold, or 20-fold.

[0127] In some embodiments, the antibodies or antigen-binding fragments thereof described herein do not induce an immune response in normal cells (e.g., non-tumor cells) or in the absence of tumor cells.

[0128] In some embodiments, the antibodies or antigen-binding fragments thereof (e.g., bispecific antibodies) can bind to PD-L1 or PD-L2. Thus, the antibodies or antigen-binding fragments thereof described herein can block the binding between PD-1 and PD-L1 and / or the binding between PD-1 and PD-L2. In some embodiments, upon binding to PD-L1 or PD-L2, the antibodies can inhibit the PD-1 signaling pathway and upregulate an immune response. Thus, in some embodiments, the antibodies or antigen-binding fragments described herein are PD-1 antagonists. In some embodiments, the antibodies or antigen-binding fragments thereof are PD-1 agonists.

[0129] In some embodiments, the antibodies or antigen-binding fragments thereof (e.g., bispecific antibodies) can bind to CD3. Thus, the antibodies or antigen-binding fragments thereof described herein can recruit T cells to target cells.

[0130] In some embodiments, the antibody (or antigen-binding fragment thereof) has a dissociation rate (koff) of 0.1 s -1 Less than 0.01s -1 Less than 0.001s -1 Less than 0.0001s -1 Less than, or 0.00001s -1 In some embodiments, the dissociation rate (koff) is less than 0.01 s -1 Exceeds 0.001s -1 Exceeds 0.0001s -1 Exceeds 0.00001s -1 or 0.000001s -1 In some embodiments, the kinetic binding rate (k) is greater than 1x10 2 / Ms over 1x10 3 / Ms over 1x10 4 / Ms over 1x10 5 / Ms or 1x106 In some embodiments, the kinetic binding rate (k) is greater than 1x10 5 / Ms is less than 1x10 6 / Ms or less than 1x10 7 / Ms is less than.

[0131] Affinity can be estimated from the quotient of the kinetic rate constants (Kd=koff / kon). In some embodiments, Kd is greater than or equal to 1x10 -4 Less than M, 1x10 -5 Less than M, 1x10 -6 Less than M, 1x10 -7 Less than M, 1x10 -8 Less than M, 1x10 -9 Less than m or 1x10 -10 In some embodiments, the Kd is less than 50 nM, 30 nM, 20 nM, 15 nM, 10 nM, 9 nM, 8 nM, 7 nM, 6 nM, 5 nM, 4 nM, 3 nM, 2 nM, or 1 nM. In some embodiments, the Kd is less than 1 x 10 -4 Over M, 1x10 -5 Over M, 1x10 -6 Over M, 1x10 -7 Over M, 1x10 -8 Over M, 1x10 -9 Over M, 1x10 -10 Over M, 1x10 -11 Over M or 1x10 -12 exceeds M. Furthermore, Ka can be calculated from Kd by the formula Ka=1 / Kd.

[0132] Common techniques for measuring the affinity of an antibody for an antigen include, for example, ELISA, RIA, and surface plasmon resonance (SPR).

[0133] In some embodiments, thermal stability is determined. The antibodies or antigen-binding fragments described herein can have a Tm of greater than 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, or 95°C.

[0134] Because IgG can be described as a multi-domain protein, the melting curve sometimes shows two or three transitions, with a first denaturation temperature, Tm D1, a second denaturation temperature, Tm D2, and optionally a third denaturation temperature, Tm D3.

[0135] In some embodiments, the antibodies or antigen-binding fragments described herein have a Tm D1 of greater than 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, or 95°C. In some embodiments, the antibodies or antigen-binding fragments described herein have a Tm D2 of greater than 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, or 95°C. In some embodiments, the antibodies or antigen-binding fragments described herein have a Tm D3 of greater than 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, or 95°C.

[0136] In some embodiments, the Tm, Tm D1, Tm D2, Tm D3 is less than 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, or 95°C.

[0137] In some embodiments, the antibodies or antigen-binding fragments described herein do not begin to form aggregates at temperatures below 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, or 95°C. In some embodiments, Tagg266 or Tagg473 is less than 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, or 95°C.

[0138] In some embodiments, the antibodies or antigen-binding fragments described herein have a pI of greater than 7.0, 7.1, 7.2, 7.3, 7.4, 7.5, 7.6, 7.7, 7.8, 7.9, 8.0, 8.1, 8.2, 8.3, 8.4, 8.5, 8.6, 8.7, 8.8, 8.9, 9.0, 9.1, 9.2, 9.3, 9.4, 9.5, 9.6, 9.7, 9.8, or 9.9. In some embodiments, the antibodies or antigen-binding fragments described herein have a pI of less than 7.0, 7.1, 7.2, 7.3, 7.4, 7.5, 7.6, 7.7, 7.8, 7.9, 8.0, 8.1, 8.2, 8.3, 8.4, 8.5, 8.6, 8.7, 8.8, 8.9, 9.0, 9.1, 9.2, 9.3, 9.4, 9.5, 9.6, 9.7, 9.8, or 9.9.

[0139] In some embodiments, the antibody has a tumor growth inhibition percentage (TGI%) of greater than 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, 110%, 120%, 130%, 140%, 150%, 160%, 170%, 180%, 190%, or 200%. In some embodiments, the antibody has a tumor growth inhibition percentage of less than 60%, 70%, 80%, 90%, 100%, 110%, 120%, 130%, 140%, 150%, 160%, 170%, 180%, 190%, or 200%. The TGI% can be determined, for example, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30 days after initiation of treatment, or can be determined 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 months after initiation of treatment. As used herein, tumor growth inhibition percentage (TGI%) is calculated using the following formula: TGI(%)=[1-(Ti-T0) / (Vi-V0)]×100

[0140] Ti is the mean tumor volume of the treatment group on day i. T0 is the mean tumor volume of the treatment group on day 0. Vi is the mean tumor volume of the control group on day i. V0 is the mean tumor volume of the control group on day 0.

[0141] In some embodiments, the antibody or antigen-binding fragment can increase complement dependent cytotoxicity (CDC) by at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, 2-fold, 3-fold, 5-fold, 10-fold, or 20-fold.

[0142] In some embodiments, the antibody or antigen-binding fragment can increase antibody-dependent cell-mediated cytotoxicity (ADCC) by at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, 2-fold, 3-fold, 5-fold, 10-fold, or 20-fold.

[0143] In some embodiments, the antibody or antigen-binding fragment can increase the internalization rate by at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, 2-fold, 3-fold, 5-fold, 10-fold, or 20-fold.

[0144] In some embodiments, the antibody or antigen-binding fragment can increase the phagocytosis rate by at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, 2-fold, 3-fold, 5-fold, 10-fold, or 20-fold.

[0145] In some embodiments, the antibody or antigen-binding fragment can enhance T cell function, for example, by increasing effector T cell proliferation and / or increasing gamma interferon production by effector T cells (e.g., by comparing proliferation and / or cytokine production prior to treatment with the antibody or antigen-binding fragment).

[0146] In some embodiments, the antibody or antigen-binding fragment can enhance CD4+ effector T cell function, e.g., by increasing CD4+ effector T cell proliferation and / or increasing gamma interferon production by CD4+ effector T cells (e.g., by comparing proliferation and / or cytokine production prior to treatment with the antibody or antigen-binding fragment). In some embodiments, the cytokine is gamma interferon. In some embodiments, the antibody or antigen-binding fragment increases the number of intratumoral (infiltrating) CD4+ effector T cells (e.g., the total number of CD4+ effector T cells, or, e.g., the percentage of CD4+ cells among CD45+ cells), e.g., compared to the number of intratumoral (infiltrating) CD4+ T cells prior to treatment with the antibody or antigen-binding fragment. In some embodiments, the antibody or antigen-binding fragment increases the number of intratumoral (infiltrating) CD4+ effector T cells that express gamma interferon (e.g., total gamma interferon-expressing CD4+ cells, or, e.g., percentage of gamma interferon-expressing CD4+ cells among total CD4+ cells), e.g., compared to the number of intratumoral (infiltrating) CD4+ T cells that express gamma interferon before treatment.

[0147] In some embodiments, the antibody or antigen-binding fragment increases the number of intratumor (infiltrating) CD8+ effector T cells (e.g., the total number of CD8+ effector T cells, or, e.g., the percentage of CD8+ cells among CD45+ cells), e.g., compared to the number of intratumor (infiltrating) CD8+ T cells prior to treatment with the antibody or antigen-binding fragment. In some embodiments, the antibody or antigen-binding fragment increases the number of intratumor (infiltrating) CD8+ effector T cells that express gamma interferon (e.g., the percentage of gamma interferon-expressing CD8+ cells among total CD8+ cells), e.g., compared to the number of intratumor (infiltrating) CD8+ T cells that express gamma interferon prior to treatment with the antibody.

[0148] In some embodiments, the antibody or antigen-binding fragment enhances memory T cell function, for example, by increasing memory T cell proliferation and / or increasing cytokine (eg, gamma interferon) production by memory cells.

[0149] In some embodiments, the antibody or antigen-binding fragment has a functional Fc region. In some embodiments, the effector function of the functional Fc region is antibody-dependent cell-mediated cytotoxicity (ADCC). In some embodiments, the effector function of the functional Fc region is phagocytosis. In some embodiments, the effector function of the functional Fc region is ADCC and phagocytosis. In some embodiments, the Fc region is human IgG1, human IgG2, human IgG3, or human IgG4.

[0150] In some embodiments, the antibody or antigen-binding fragment is capable of inducing apoptosis.

[0151] In some embodiments, the antibody or antigen-binding fragment does not have a functional Fc region, e.g., the antibody or antigen-binding fragment is a Fab, Fab', F(ab')2, and Fv fragment.

[0152] In some embodiments, the antibody or antigen-binding fragment is a humanized antibody. The percentage of humanization refers to the percentage identity of the heavy or light chain variable region sequence compared to human antibody sequences in the International Immunogenetics Information System (IMGT) database. In some embodiments, the percentage of humanization is greater than 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, or 95%. Detailed descriptions of how to determine the percentage of humanization are well known in the art and are described, for example, in Jones, Tim D., et al. "The INNs and outs of antibody nonproprietary names," MAbs. Vol. 8, No. 1, Taylor & Francis, 2016, the entire contents of which are incorporated herein by reference. A high percentage of humanization often has various advantages, such as being safer and more effective in humans, potentially being better tolerated in human subjects, and / or causing fewer side effects, hi some embodiments, the antibody or antigen-binding fragment is a human antibody.

[0153] Recombinant vector The present disclosure also provides recombinant vectors (e.g., expression vectors) comprising the isolated polynucleotides disclosed herein (e.g., polynucleotides encoding the polypeptides disclosed herein), host cells into which the recombinant vectors have been introduced (i.e., such that the host cell contains the polynucleotides, and / or vectors comprising the polynucleotides), and the production of recombinant antibody polypeptides or fragments thereof by recombinant techniques.

[0154] As used herein, a "vector" refers to any construct capable of delivering one or more polynucleotides of interest to a host cell when introduced into the host cell. An "expression vector" is capable of delivering one or more polynucleotides of interest to be expressed as encoded polypeptides when introduced into a host cell. Thus, in an expression vector, a polynucleotide of interest is placed in the vector for expression by operably linking it to regulatory elements such as a promoter, an enhancer, and / or a polyA tail at, near, or flanking the insertion site of the polynucleotide of interest in the vector or in the genome of the host cell, so that the polynucleotide of interest is translated in the host cell into which the expression vector is introduced.

[0155] Vectors can be introduced into host cells by methods well known in the art, such as, for example, electrophoresis, chemical transfer (e.g., DEAE-dextran), transformation, transfection, and infection and / or transduction (e.g., by recombinant viruses). Thus, non-limiting examples of vectors include viral vectors (which can be used to generate recombinant viruses), naked DNA or RNA, plasmids, cosmids, phage vectors, and DNA or RNA expression vectors coupled with cationic condensing agents.

[0156] In some implementations, a polynucleotide disclosed herein (e.g., a polynucleotide encoding a polypeptide disclosed herein) is introduced using a viral expression system (e.g., cowpox or other poxvirus, retrovirus, or adenovirus), which can include the use of non-pathogenic (defective), replication-competent, or replication-defective viruses, in which case viral propagation generally occurs only in complementing viral packaging cells. Suitable systems are described, for example, in Fisher-Hoch et al., 1989, Proc. Natl. Acad. Sci. USA 86:317-321; Flexner et al., 1989, Ann. NY Acad Sci. 569:86-103; Flexner et al., 1990, Vaccine, 8:17-21; U.S. Patent Nos. 4,603,112, 4,769,330, and 5,017,487; WO 89 / 01973; U.S. Patent Application No. 4,777,127; British Patent Application No. 2,200,651; European Patent Application No. 0,345,242; WO 91 / 02805; Berkner-Biotechniques, 6:616-627, 1988; These expression systems are described in Rosenfeld et al., 1991, Science, 252:431-434; Kolls et al., 1994, Proc. Natl. Acad. Sci. USA, 91:215-219; Kass-Eisler et al., 1993, Proc. Natl. Acad. Sci. USA, 90:11498-11502; Guzman et al., 1993, Circulation, 88:2838-2848; and Guzman et al., 1993, Cir. Res., 73:1202-1207. Techniques for incorporating DNA into such expression systems are well known to those skilled in the art. The DNA may be "naked," for example, as described in Ulmer et al., 1993, Science, 259:1745-1749 and Cohen, 1993, Science, 259:1691-1692.The uptake of naked DNA can be increased by coating the DNA with biodegradable beads that are efficiently transported into cells.

[0157] For expression, the DNA insert, i.e., the antibody-encoding or polypeptide-encoding polynucleotide disclosed herein, is operably linked to a suitable promoter (e.g., a heterologous promoter), such as the phage lambda PL promoter, the E. coli lac, trp, and tac promoters, the SV40 early and late promoters, and promoters of retroviral long terminal repeats, to name a few. Suitable promoters are well known to those of skill in the art. Expression constructs will further include transcription initiation and termination sites, and, in the transcribed region, a ribosome binding site for translation. The coding portion of the mature transcript produced by the construct will include a translation initiation at the beginning and may include an appropriately positioned termination codon (UAA, UGA, or UAG) at the end of the polypeptide to be translated.

[0158] As indicated, the expression vector may contain at least one selectable marker. Such markers include dihydrofolate reductase or neomycin resistance genes for eukaryotic cell culture, and tetracycline or ampicillin resistance genes for culture in E. coli and other bacteria. Representative examples of suitable hosts include, but are not limited to, bacterial cells such as E. coli, Streptomyces, and Salmonella typhimurium cells; fungal cells such as yeast cells; insect cells such as Drosophila S2 and Spodoptera Sf9 cells; animal cells such as CHO, COS, Bowes melanoma, and HK293 cells; and plant cells. Appropriate culture media and conditions for the host cells described herein are well known in the art.

[0159] Non-limiting vectors for use in bacteria include pQE70, pQE60, and pQE-9 available from Qiagen; pBS vectors, Phagescript vectors, Bluescript vectors, pNH8A, pNH16a, pNH18A, and pNH46A available from Stratagene; and ptrc99a, pKK223-3, pKK233-3, pDR540, and pRIT5 available from Pharmacia. Non-limiting eukaryotic vectors include pWLNEO, pSV2CAT, pOG44, pXT1, and pSG available from Stratagene; and pSVK3, pBPV, pMSG, and pSVL available from Pharmacia. Other suitable vectors will be readily apparent to those skilled in the art.

[0160] Non-limiting bacterial promoters suitable for use include the E. coli lacI and lacZ promoters, T3 and T7 promoters, gpt promoter, lambda PR and PL promoters, and trp promoter. Suitable eukaryotic promoters include the CMV immediate early promoter, HSV thymidine kinase promoter, early and late SV40 promoters, promoters of retroviral LTRs such as Rous sarcoma virus (RSV), and metallothionein promoters such as the mouse metallothionein-I promoter.

[0161] In the yeast Saccharomyces cerevisiae, a series of vectors containing constitutive or inducible promoters, such as α-factor, alcohol oxidase, and PGH, are available. For reviews, see Ausubel et al. (1989) Current Protocols in Molecular Biology, John Wiley & Sons, New York, NY, and Grant et al., Method Enzymol., 153:516-544 (1997).

[0162] Introduction of the construct into the host cell can be accomplished by calcium phosphate transfection, DEAE-dextran mediated transfection, cationic lipid mediated transfection, electrophoresis, transduction, infection, or other methods described in standard laboratory manuals such as Davis et al., Basic Methods In Molecular Biology (1986), which is incorporated herein by reference in its entirety.

[0163] Transcription of DNA encoding the antibodies of this disclosure by higher eukaryotes can be increased by inserting an enhancer sequence into the vector. Enhancers are cis-acting elements of DNA, usually about 10 to 300 bp, that act to increase transcriptional activity of a promoter in a given host cell type. Examples of enhancers include the SV40 enhancer, which is located on the late side of the replication origin at base pairs 100 to 270, the cytomegalovirus early promoter enhancer, the polyoma enhancer on the late side of the replication origin, and adenovirus enhancers.

[0164] For secretion of the translated protein into the lumen of the endoplasmic reticulum, the periplasmic space or the extracellular environment, appropriate secretion signals may be incorporated into the expressed polypeptide, which signals may be endogenous to the polypeptide or they may be heterologous signals.

[0165] The polypeptides (e.g., antibodies) can be expressed in modified forms, such as fusion proteins (e.g., GST-fusions) or with histidine tags, and can contain not only secretion signals but also other heterologous functional regions. For example, a region of additional amino acids, particularly charged amino acids, can be added to the N-terminus of the polypeptide to improve stability and persistence in host cells during purification or during subsequent processing or storage. Furthermore, peptide moieties can be added to polypeptides to facilitate purification. Such regions can be removed prior to final purification of the polypeptide. The addition of peptide moieties to polypeptides to improve stability and facilitate purification, to effect secretion or excretion, and to facilitate secretion are, inter alia, familiar and routine techniques in the art.

[0166] The present disclosure provides nucleic acid sequences that are at least 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% identical to any of the nucleotide sequences described herein. , and amino acid sequences that are at least 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% identical to any amino acid sequence described herein.

[0167] The present disclosure also provides nucleic acid sequences having at least 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% homology to any of the nucleotide sequences described herein. and amino acid sequences having at least 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% homology to any amino acid sequence described herein.

[0168] In some embodiments, the disclosure relates to a nucleotide sequence encoding any of the peptides described herein or any amino acid sequence encoded by any of the nucleotide sequences described herein. In some embodiments, the nucleic acid sequence is less than 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 150, 200, 250, 300, 350, 400, 500, or 600 nucleotides. In some embodiments, the amino acid sequence is less than 5, 6, 7, 8, 9, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200, 250, 300, 350, or 400 amino acid residues.

[0169] In some embodiments, the amino acid sequence (i) comprises an amino acid sequence; or (ii) consists of an amino acid sequence, which is any one of the sequences described herein.

[0170] In some embodiments, the nucleic acid sequence (i) comprises a nucleic acid sequence; or (ii) consists of a nucleic acid sequence, which is any one of the sequences described herein.

[0171] To determine the percent identity of two amino acid sequences or two nucleic acid sequences, the sequences are aligned for optimal comparison purposes (e.g., gaps can be introduced into one or both of the first and second amino acid or nucleic acid sequences to ensure optimal alignment and to allow non-cognate sequences to be ignored for comparison purposes). The length of a reference sequence aligned for comparison purposes is at least 80% of the length of the reference sequence, and in some embodiments, at least 90%, 95%, or 100%. The amino acid residues or nucleotides at corresponding amino acid or nucleotide positions are then compared. If a position in the first sequence is occupied by the same amino acid residue or nucleotide as the corresponding position in the second sequence, then the molecules are identical at that position (as used herein, amino acid or nucleic acid "identity" is equivalent to amino acid or nucleic acid "homology"). The percent identity between two sequences is a function of the number of identical positions shared by the sequences, taking into account the number of gaps and gap length that need to be introduced for optimal alignment of the two sequences. For purposes of the present invention, sequence comparison and determination of percent identity between two sequences may be accomplished using a Blossum62 scoring matrix with a gap penalty of 12, a gap extend penalty of 4, and a frameshift gap penalty of 5.

[0172] Percentage sequence homology (e.g., amino acid sequence homology or nucleic acid homology) can also be determined. Methods for determining percentage sequence homology are well known in the art. In some embodiments, conserved amino acid residues with similar physicochemical properties (percent homology), e.g., leucine and isoleucine, can be used to calculate sequence similarity. Families of amino acid residues with similar physicochemical properties have been defined in the art. These families include, for example, amino acids with basic side chains (e.g., lysine, arginine, histidine), acidic side chains (e.g., aspartic acid, glutamic acid), uncharged polar side chains (e.g., glycine, asparagine, glutamine, serine, threonine, tyrosine, cysteine), nonpolar side chains (e.g., alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine, tryptophan), beta-branched side chains (e.g., threonine, valine, isoleucine), and aromatic side chains (e.g., tyrosine, phenylalanine, tryptophan, histidine). In many cases, the percentage homology is higher than the percentage identity.

[0173] Antibody production method Isolated fragments of human proteins (e.g., CD55, CD3, cancer-specific antigens, or cancer-associated antigens) can be used as immunogens to generate antibodies using standard techniques for polyclonal and monoclonal antibody preparation. Polyclonal antibodies can be raised in animals by multiple injections (e.g., subcutaneous or intraperitoneal) of antigenic peptides or proteins. In some embodiments, the antigenic peptides or proteins are injected with at least one adjuvant. In some embodiments, the antigenic peptides or proteins are conjugated to an agent that is immunogenic in the species being immunized. Animals can be injected with the antigenic peptides or proteins two or more times (e.g., two, three, or four times).

[0174] The full-length polypeptide or protein can be used, or alternatively, an antigenic peptide fragment thereof can be used as the immunogen. An antigenic peptide of a protein includes at least 8 (e.g., at least 10, 15, 20, or 30) amino acid residues of the amino acid sequence of the protein and encompasses an epitope of the protein such that antibodies raised against the peptide can form a specific immune complex with the protein.

[0175] An immunogen is typically used to prepare antibodies by immunizing a suitable subject (e.g., a human or transgenic animal expressing at least one human immunoglobulin locus). An appropriate immunogenic preparation can contain, for example, a recombinantly expressed or chemically synthesized polypeptide. The preparation can further include an adjuvant, such as Freund's complete or incomplete adjuvant, or a similar immunostimulant.

[0176] Polyclonal antibodies can be prepared, as described above, by immunizing a suitable subject using a polypeptide, or an antigenic peptide thereof (e.g., a portion of the protein) as an immunogen. The antibody titers in the immunized subject can be monitored over time by standard techniques, such as enzyme-linked immunosorbent assay (ELISA) using immobilized polypeptide or peptide. If desired, the antibody molecules can be isolated from the mammal (e.g., from its blood) and further purified by well-known techniques, such as protein A followed by protein G chromatography to obtain the IgG fraction. After immunization, after an appropriate time has elapsed, e.g., when antibody titers are at their peak, antibody-producing cells can be obtained from the subject and used to prepare monoclonal antibodies using standard techniques, such as the hybridoma technique first described by Kohler et al. (Nature 256:495-497, 1975), the human B cell hybridoma technique (Kozbor et al., Immunol. Today 4:72, 1983), the EBV-hybridoma technique (Cole et al., Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, Inc., pp. 77-96, 1985), or the trioma technique. Techniques for producing hybridomas are well known (see generally Current Protocols in Immunology, 1994, Coligan et al. (Eds.), John Wiley & Sons, Inc., New York, NY). Hybridoma cells producing a monoclonal antibody can be detected by screening the hybridoma culture supernatants for antibodies that bind the polypeptide or epitope of interest, eg, using an ELISA assay.

[0177] Mutations of the antibodies or antigen-binding fragments described herein can be prepared by introducing appropriate nucleotide changes into DNA encoding the human, humanized, or chimeric antibodies or antigen-binding fragments thereof described herein, or by peptide synthesis. Such mutations include, for example, deletion, insertion, or substitution of residues within the amino acid sequence forming the antigen-binding site or domain of the antibody. In a population of such mutations, some antibodies or antigen-binding fragments have increased affinity for the target protein. Any combination of deletions, insertions, and / or combinations can be made to arrive at an antibody or antigen-binding fragment thereof with increased binding affinity for the target. The amino acid changes introduced into the antibody or antigen-binding fragment can alter or introduce new post-translational modifications of the antibody or antigen-binding fragment, such as changing the number (e.g., increasing or decreasing) of glycosylation sites, altering glycosylation sites (e.g., altering the amino acid sequence so that different sugars are attached by enzymes present in cells), or introducing new glycosylation sites.

[0178] The antibodies disclosed herein can be derived from any species of animal, including mammals. Non-limiting examples of natural antibodies include antibodies derived from humans, primates, such as monkeys and apes, cows, pigs, horses, sheep, camelids (e.g., camels and llamas), chickens, goats, and rodents (e.g., rats, mice, hamsters, and rabbits), including transgenic rodents that have been genetically engineered to produce human antibodies.

[0179] Phage display (panning) can be used to optimize antibody sequences with desired binding affinities. In this technique, a gene encoding a single-chain Fv (including VH or VL) can be inserted into a phage coat protein gene, thereby displaying an scFv on the outside of the phage, while the gene for the protein is contained inside the phage, resulting in a link between genotype and phenotype. These display phages can then be screened against a target antigen to detect the interaction between the displayed antigen-binding site and the target antigen. Thus, a large library of proteins can be screened and amplified in a process called in vitro selection, allowing antibody sequences with desired binding affinities to be obtained.

[0180] Human and humanized antibodies include antibodies with variable and constant regions derived from (or having amino acid sequences identical to) human germline immunoglobulin sequences. Human antibodies may include amino acid residues (e.g., by in vitro random or site-specific mutagenesis or in vivo somatic mutation) not encoded by human germline immunoglobulin sequences, for example, in the CDRs.

[0181] Humanized antibodies typically have a human framework (FR) into which nonhuman CDRs have been grafted. Thus, a humanized antibody has one or more amino acid sequences introduced into it from a nonhuman source. These nonhuman amino acid residues are often referred to as "import" residues, and they are typically taken from an "import" variable domain. Humanization can essentially be performed, for example, by substituting rodent CDRs or CDR sequences for the corresponding sequences of a human antibody. These methods are described, for example, in Jones et al., Nature, 321:522-525 (1986); Riechmann et al., Nature, 332:323-327 (1988); Verhoeye et al., Science, 239:1534-1536 (1988); each of which is incorporated herein by reference. Thus, a "humanized" antibody is a chimeric antibody in which significantly less than an intact human V domain has been substituted by the corresponding sequence from a nonhuman species. In practice, humanized antibodies are typically murine antibodies in which some CDR residues and some FR residues are substituted by residues from analogous sites in human antibodies.

[0182] It is further important that antibodies be humanized with retention of high specificity and affinity for the antigen and other favorable biological properties. To achieve this goal, humanized antibodies can be prepared by a process of analysis of the parental sequences and various conceptual humanized products using three-dimensional models of the parental and humanized sequences. Three-dimensional immunoglobulin models are commonly available and are familiar to those skilled in the art. Computer programs are available which illustrate and display probable three-dimensional conformational structures of selected candidate immunoglobulin sequences. Inspection of these displays permits analysis of the likely role of the residues in the functioning of the candidate immunoglobulin sequence, i.e., the analysis of residues that influence the ability of the candidate immunoglobulin to bind to its antigen. In this way, FR residues from the recipient can be selected and combined with the import sequence to achieve the desired antibody characteristic, such as increased affinity for the antigen.

[0183] Identity or homology with respect to the original sequence is the percentage of amino acid residues present in the candidate sequence that are identical to sequences present in a human, humanized, or chimeric antibody or fragment, aligning the sequences and introducing gaps, if necessary, to achieve the maximum percent sequence identity, and not considering any conventional substitutions as part of the sequence identity.

[0184] In some embodiments, covalent modifications can be made to antibodies or antigen-binding fragments thereof. These covalent modifications can be produced by chemical or enzymatic synthesis, or by enzymatic or chemical cleavage. Other types of covalent modifications of antibodies or antibody fragments introduced into the molecule can be induced by reacting targeted amino acid residues of the antibody or fragment with organic derivatizing agents capable of reacting with selected side chains, or with N- or C-terminal residues.

[0185] In some embodiments, antibody variants are provided having carbohydrate structures that lack fucose attached (directly or indirectly) to the Fc region. For example, the amount of fucose in such antibodies can be 1%-80%, 1%-65%, 5%-65%, or 20%-40%. The amount of fucose can be determined by calculating the average amount of fucose in the glycan at Asn297 relative to the sum of all glycans attached to Asn297 (e.g., complex, hybrid, and high mannose structures) as measured by MALDI-TOF mass spectrometry as described in WO 2008 / 077546. For example, Asn297 refers to the asparagine residue located at approximately position 297 of the Fc region (Eu numbering of Fc region residues; or position 314 in Kabat numbering); however, Asn297 can be located approximately ±3 amino acids upstream or downstream of position 297, i.e., between positions 294 and 300, due to minor sequence variations in the antibody. Such fucosylation variations improve ADCC function. In some embodiments, the Fc region of the antibody can be further engineered to replace the asparagine at position 297 with alanine (N297A) to reduce glycan heterogeneity.

[0186] In some embodiments, the Fc region of the antibody can be further engineered to replace the serine at position 228 (EU numbering) of IgG4 with proline (S228P) to promote production efficiency by avoiding Fab-arm exchange. A detailed description of the S228 mutation is found, for example, in Silva et al., "The S228P mutation prevents in vivo and in vitro IgG4 Fab-arm exchange as demonstrated using a combination of novel quantitative immunoassays and physiological matrix preparation." Journal of Biological Chemistry 290.9 (2015):5462-5469, which is incorporated by reference in its entirety.

[0187] In some embodiments, the methods described herein are designed to generate bispecific antibodies. Bispecific antibodies can be generated by engineering the interface between a pair of antibody molecules to maximize the percentage of heterodimers recovered from recombinant cell culture. For example, the interface can comprise at least a portion of the CH3 domain of an antibody constant domain. In this method, one or more small amino acid side chains from the interface of a first antibody molecule are replaced with larger side chains (e.g., tyrosine or tryptophan). Compensatory "cavities" of identical or similar size for the larger side chains are generated on the interface of a second antibody molecule by replacing the large amino acid side chains with smaller ones (e.g., alanine or threonine). This provides a mechanism for increasing the yield of heterodimers over other unwanted end-products, such as homodimers. This method is described, for example, in WO 96 / 27011, incorporated by reference in its entirety.

[0188] In some embodiments, one or more amino acid residues of the CH3 portion of IgG are substituted. In some embodiments, one heavy chain has one or more of the following substitutions: Y349C and T366W. The other heavy chain has one or more of the following substitutions: E356C, T366S, L368A, and Y407V. Further, the substitution (-ppcpScp-->-ppcpPcp-) can also be introduced into both substituted IgG hinge regions. In some embodiments, one heavy chain has a T366Y (knob) substitution and the other heavy chain has a Y407T (hole) substitution.

[0189] Furthermore, anion exchange chromatography can be used in the production of bispecific antibodies. Anion exchange chromatography is a process of separating substances based on their charges using an ion exchange resin containing a positively charged group such as a diethylaminoethyl (DEAE) group. In solution, this resin is covered by positively charged counterions (cations). The anion exchange resin binds to negatively charged molecules and replaces the counterions. Anion exchange chromatography can be used to purify proteins based on their isoelectric point (pI). The isoelectric point is defined as the pH at which a protein has no net charge overall. When pH > pI, the protein has an overall negative charge, and when pH < pI, the protein has an overall positive charge. Thus, in some embodiments, different amino acid substitutions can be introduced into the two heavy chains, and the pI for the homodimer containing two arm A's and the pI for the homodimer containing two arm B's are different. The pI for the bispecific antibody having arm A and arm B is somewhere between the two pI's of the homodimers. Thus, the two homodimers and the bispecific antibody are eluted under different pH conditions. The present disclosure shows that substitutions of some amino acid residues can introduce pI adjustment into the heavy chain.

[0190] Thus, in some embodiments, the amino acid residue at position 83 (Kabat numbering) is lysine, arginine, or histidine. In some embodiments, the amino acid residue at one or more of positions 1, 6, 43, 81, and 105 (Kabat numbering) is aspartic acid or glutamic acid.

[0191] In some embodiments, one or more of the amino acid residues at positions 13 and 105 (Kabat numbering) are aspartic acid or glutamic acid. In some embodiments, one or more of the amino acid residues at positions 13 and 42 (Kabat numbering) are lysine, arginine, histidine, or glycine.

[0192] Bispecific antibodies can include, for example, cross-linked or "heteroconjugate" antibodies. For example, one of the antibodies in the heteroconjugate can be coupled to avidin, and the other to biotin. Heteroconjugate antibodies can be made using any convenient cross-linking method. Suitable cross-linking agents and techniques are well known in the art and are described in U.S. Pat. No. 4,676,980, incorporated herein by reference in its entirety.

[0193] Methods for generating bispecific antibodies from antibody fragments are also well known in the art. For example, bispecific antibodies can be prepared using chemical linkage. Brenna et al. (Science 229:81, 1985) describe a procedure in which intact antibodies are proteolytically cleaved to generate F(ab')2 fragments. These fragments are reduced in the presence of the dithiol complexing agent, sodium arsenite, to stabilize vicinal dithiols and prevent intermolecular disulfide bond formation. The generated Fab' fragments are then converted to thionitrobenzoic acid (TNB) derivatives. One of the Fab'TNB derivatives is then reconverted to the Fab' thiol by reduction with mercaptoethylamine and mixed with an equimolar amount of another Fab'TNB derivative to form the bispecific antibody.

[0194] treatment The methods described herein include methods for the treatment of cancer-related diseases. Generally, the methods involve administering a therapeutically effective amount of an engineered bispecific antibody (e.g., an imbalanced bispecific antibody) or antigen-binding fragment thereof described herein to a subject in need thereof or determined to be in need of such treatment.

[0195] As used in this context, "treat" means improving at least one symptom of a cancer-related disease. Cancer often results in death; therefore, treatment results in increased life expectancy (e.g., at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 months, or at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 years). Administration of a therapeutically effective amount of a described agent (e.g., an unbalanced bispecific antibody) for the treatment of a cancer-related condition results in a reduced number of cancer cells and / or alleviated symptoms.

[0196] As used herein, the term "cancer" refers to an abnormal state or condition characterized by cells capable of autonomous growth, i.e., rapidly proliferating cell growth. The term is meant to include all types of cancerous growth or oncogenic processes, metastatic tissues, or malignantly transformed cells, tissues, or organs, regardless of histopathological type or invasive stage. The term "tumor" as used herein refers to cancerous cells, e.g., a mass of cancerous cells. Cancers that can be treated or diagnosed using the methods described herein include malignancies of various organ systems affecting the lung, breast, thyroid, lymphatic, gastrointestinal, and genitourinary tracts, as well as most adenocarcinomas, such as colon cancer, renal cell carcinoma, prostate cancer, and / or testicular cancer, non-small cell lung cancer, small intestine cancer, and esophageal cancer. In some embodiments, the agents described herein are designed to treat or diagnose cancer in a subject. "Cancer" is art-recognized and refers to malignant tumors of epithelial or endocrine tissue, including respiratory cancer, digestive cancer, genitourinary system cancer, testicular cancer, breast cancer, prostate cancer, endocrine system cancer, and melanoma. In some embodiments, the cancer is renal carcinoma or melanoma. Exemplary cancers include those forming from tissue of the cervix, lung, prostate, breast, head and neck, colon, and ovary. The term also includes carcinosarcomas, which include malignant tumors composed of carcinomatous and sarcomatous tissue. "Adenocarcinoma" refers to cancer derived from glandular tissue or in which the tumor cells form recognizable glandular structures. The term "sarcoma" is art-recognized and refers to malignant tumors derived from mesenchyme.

[0197] In some embodiments, the cancer is a Rituximab (Rituxan®) resistant cancer.

[0198] In one aspect, the present disclosure also provides a method for treating cancer in a subject, a method for reducing the rate of growth of a tumor over time in a subject, a method for reducing the risk of developing metastasis, or a method for reducing the risk of developing additional metastases in a subject. In some embodiments, the treatment can stop, slow, delay, or inhibit the progression of cancer. In some embodiments, the treatment can result in a reduction in one or more of the number, severity, and / or duration of symptoms of cancer in a subject.

[0199] In one aspect, the disclosure features a method comprising administering a therapeutically effective amount of an antibody or antigen-binding fragment thereof, or an antibody drug conjugate disclosed herein, to a subject in need thereof, e.g., a subject having, identified, or diagnosed with cancer, such as breast cancer (e.g., triple-negative breast cancer), carcinoid cancer, cervical cancer, endometrial cancer, glioma, head and neck cancer, liver cancer, lung cancer, small cell lung cancer, lymphoma, melanoma, ovarian cancer, pancreatic cancer, prostate cancer, renal cancer, colorectal cancer, gastrointestinal cancer, testicular cancer, thyroid cancer, bladder cancer, urethral cancer, or a hematological malignancy.

[0200] As used herein, the terms "subject" and "patient" are used interchangeably throughout the specification and refer to an animal, human or non-human, to which treatment according to the methods of the present invention is provided. Veterinary and non-veterinary applications are contemplated by the present invention. A human patient can be an adult human or a juvenile human (e.g., a human under the age of 18). In addition to humans, patients include, but are not limited to, mice, rats, hamsters, guinea pigs, rabbits, ferrets, cats, dogs, and primates. Included are, for example, non-human primates (e.g., monkeys, chimpanzees, gorillas, etc.), rodents (e.g., rats, mice, gerbils, hamsters, ferrets, rabbits), lagomorphs, swine (e.g., pigs, minipigs), equines, canines, felines, bovines, and other domestic animals, and zoo animals.

[0201] In some embodiments, the cancer is unresectable or metastatic melanoma, non-small cell lung cancer (NSCLC), small cell lung cancer (SCLC), bladder cancer, or metastatic hormone-refractory prostate cancer. In some embodiments, the subject has a solid tumor. In some embodiments, the cancer is squamous cell carcinoma of the head and neck (SCCHN), renal cell carcinoma (RCC), triple-negative breast cancer (TNBC), or colorectal cancer. In some embodiments, the subject has Hodgkin's lymphoma. In some embodiments, the subject has triple-negative breast cancer (TNBC), gastrointestinal cancer, urothelial carcinoma, Merkel cell carcinoma, or head and neck cancer. In some embodiments, the cancer is melanoma, pancreatic cancer, mesothelioma, a hematological malignancy, particularly non-Hodgkin's lymphoma, lymphoma, chronic lymphocytic leukemia, or an advanced solid tumor.

[0202] In some embodiments, the compositions and methods disclosed herein may be used to treat patients at risk for cancer, as well as patients with cancer, which can be identified by a variety of methods known in the art.

[0203] As used herein, an "effective amount" means an amount or dosage sufficient to achieve a beneficial or desired result, including, for example, stopping, slowing, retarding, or inhibiting the progression of a disease, e.g., cancer. The effective amount depends, for example, on the age and weight of the subject to whom the antibody, antigen-binding fragment, antibody-drug conjugate, antibody-encoding polynucleotide, vector comprising the polynucleotide, and / or composition thereof is administered, the severity of the symptoms, and the route of administration, and therefore dosing is determined on an individual basis.

[0204] An effective amount can be administered in one or more doses. For example, an effective amount of an antibody, antigen-binding fragment, or antibody-drug conjugate is an amount sufficient to ameliorate, stop, stabilize, reverse, inhibit, slow down, and / or delay the progression of a patient's autoimmune disease or cancer, or an amount sufficient to ameliorate, stop, stabilize, reverse, slow down, and / or delay the growth of cells (e.g., biopsied cells, any cancer cells described herein, or cell lines (e.g., cancer cell lines)) in vitro. As understood in the art, the effective amount of an antibody, antigen-binding fragment, or antibody-drug conjugate can vary depending on other factors, such as the patient's medical history and the type (and / or dosage) of antibody used, among others.

[0205] Effective amounts and schedules for administering the antibodies, antibody-encoding polynucleotides, antibody-drug conjugates, and / or compositions disclosed herein can be determined empirically, and such determinations are within the skill of the art. One of skill in the art will understand that the dosage to be administered will vary depending, for example, on the mammal receiving the antibodies, antibody-encoding polynucleotides, antibody-drug conjugates, and / or compositions disclosed herein, the route of administration, the particular type of antibodies, antibody-encoding polynucleotides, antigen-binding fragments, antibody-drug conjugates, and / or compositions disclosed herein used, and other agents administered to the mammal. Guidance for selecting appropriate dosages for antibodies or antigen-binding fragments can be found in literature on the therapeutic use of antibodies and antigen-binding fragments, e.g., Handbook of Monoclonal Antibodies, Ferrone et al., eds., Noges Publications, Park Ridge, NJ, 1985, ch. 22 and pp. 303-357; Smith et al., Antibodies in Human Diagnosis and Therapy, Haber et al., eds., Raven Press, New York, 1977, pp. 365-389.

[0206] An effective daily dose of an antibody is between 0.01 mg / kg and 100 mg / kg. In some embodiments, the dose is less than 100 mg / kg, 10 mg / kg, 9 mg / kg, 8 mg / kg, 7 mg / kg, 6 mg / kg, 5 mg / kg, 4 mg / kg, 3 mg / kg, 2 mg / kg, 1 mg / kg, 0.5 mg / kg, or 0.1 mg / kg. In some embodiments, the dose is greater than 10 mg / kg, 9 mg / kg, 8 mg / kg, 7 mg / kg, 6 mg / kg, 5 mg / kg, 4 mg / kg, 3 mg / kg, 2 mg / kg, 1 mg / kg, 0.5 mg / kg, 0.1 mg / kg, 0.05 mg / kg, or 0.01 mg / kg. In some embodiments, the dose is about 10 mg / kg, 9 mg / kg, 8 mg / kg, 7 mg / kg, 6 mg / kg, 5 mg / kg, 4 mg / kg, 3 mg / kg, 2 mg / kg, 1 mg / kg, 0.9 mg / kg, 0.8 mg / kg, 0.7 mg / kg, 0.6 mg / kg, 0.5 mg / kg, 0.4 mg / kg, 0.3 mg / kg, 0.2 mg / kg, or 0.1 mg / kg.

[0207] In any of the methods described herein, at least one antibody, antigen-binding fragment thereof, antibody-drug conjugate, or pharmaceutical composition (e.g., any of the antibodies, antigen-binding fragments, antibody-drug conjugates, or pharmaceutical compositions described herein), and optionally at least one additional therapeutic agent, can be administered to the subject at least once a week (e.g., once a week, twice a week, three times a week, four times a week, once a day, twice a day, or three times a day). In some embodiments, at least two different antibodies and / or antigen-binding fragments can be administered in the same composition (e.g., a liquid composition). In some embodiments, at least one antibody, antigen-binding fragment, antibody-drug conjugate, and at least one additional therapeutic agent can be administered in the same composition (e.g., a liquid composition). In some embodiments, at least one antibody or antigen-binding fragment and at least one additional therapeutic agent are administered in two different compositions (e.g., a liquid composition comprising at least one antibody or antigen-binding fragment, and a solid oral composition comprising at least one additional therapeutic agent). In some embodiments, the at least one additional therapeutic agent can be administered as a pill, tablet, or capsule. In some embodiments, the at least one additional therapeutic agent is administered as a long-acting oral formulation.

[0208] In some embodiments, the one or more additional therapeutic agents can be administered to the subject prior to or after administration of at least one antibody, antigen-binding antibody fragment, antibody-drug conjugate, or pharmaceutical composition (e.g., any of the antibodies, antigen-binding antibody fragments, or pharmaceutical compositions described herein). In some embodiments, the one or more additional therapeutic agents and at least one antibody, antigen-binding antibody fragment, antibody-drug conjugate, or pharmaceutical composition (e.g., any of the antibodies, antigen-binding antibody fragments, or pharmaceutical compositions described herein) can be administered to the subject such that there is an overlap in the period of biological activity in the subject.

[0209] In some embodiments, the subject can be administered at least one antibody, antigen-binding antibody fragment, antibody-drug conjugate, or pharmaceutical composition (e.g., any of the antibodies, antigen-binding antibody fragments, or pharmaceutical compositions described herein) for an extended period of time (e.g., at least 1 week, 2 weeks, 3 weeks, 1 month, 2 months, 3 months, 4 months, 5 months, 6 months, 7 months, 8 months, 9 months, 10 months, 11 months, 12 months, 1 year, 2 years, 3 years, 4 years, or 5 years). A skilled medical professional can determine the duration of treatment using any of the diagnostic methods described herein or by following the effectiveness of the treatment (e.g., observation of at least one symptom of cancer). As described herein, a skilled medical professional can also vary the identity and number (e.g., increase or decrease) of antibodies or antigen-binding antibody fragments, antibody-drug conjugates (and / or one or more additional therapeutic agents) administered to a subject, and can adjust the dose (e.g., increase or decrease) or frequency of administration of at least one antibody or antigen-binding antibody fragment (and / or one or more additional therapeutic agents) to said subject based on an evaluation of the effectiveness of the treatment (e.g., using methods described herein and methods well known in the art).

[0210] In some embodiments, one or more additional therapeutic agents can be administered to the subject. The additional therapeutic agents can include one or more inhibitors selected from the following: B-Raf inhibitors, EGFR inhibitors, MEK inhibitors, ERK inhibitors, K-Ras inhibitors, c-Met inhibitors, anaplastic lymphoma kinase (ALK) inhibitors, phosphatidylinositol-3-kinase (PI3K) inhibitors, Akt inhibitors, mTOR inhibitors, dual PI3K / mTOR inhibitors, Bruton's tyrosine kinase (BTK) inhibitors, and isocitrate dehydrogenase 1 (IDH1) inhibitors and / or isocitrate dehydrogenase 2 (IDH2) inhibitors. In some embodiments, the additional therapeutic agent is an indoleamine 2,3-dioxygenase-1 (IDO1) inhibitor (e.g., epacadostat).

[0211] In some embodiments, the additional therapeutic agent may comprise one or more inhibitors selected from the group consisting of a HER3 inhibitor, an LSD1 inhibitor, an MDM2 inhibitor, a BCL2 inhibitor, a CHK1 inhibitor, an activated hedgehog signaling pathway inhibitor, and an agent that selectively degrades the estrogen receptor.

[0212] In some embodiments, the additional therapeutic agent is selected from the group consisting of trabectedin, nanoparticle albumin-bound paclitaxel (nab-paclitaxel), trebananib, pazopanib, cediranib, palbociclib, everolimus, fluoropyrimidine, IFL, regorafenib, reolysin, Alimta, Zykadia, Sutent, temsirolimus, axitinib, everolimus, sorafenib, Votrient, pazopanib ib, IMA-901, AGS-003, cabozantinib, vinflunine, Hsp90 inhibitors, Ad-GM-CSF, temozolomide, IL-2, IFNa, vinblastine, thalomid, dacarbazine, cyclophosphamide, lenalidomide, azacitidine, lenalidomide, bortezomid, amrubicin, carfilzomib, pralatrexate, and enzastaurin.

[0213] In some embodiments, the additional therapeutic agents may comprise one or more therapeutic agents selected from the group consisting of an adjuvant, a TLR agonist, tumor necrosis factor (TNF) alpha, IL-1, HMGB1, an IL-10 antagonist, an IL-4 antagonist, an IL-13 antagonist, an IL-17 antagonist, an HVEM antagonist, an ICOS agonist, a therapeutic target CX3CL1, a therapeutic target CXCL9, a therapeutic target CXCL10, a therapeutic target CCL5, an LFA-1 agonist, an ICAM1 agonist, and a selectin agonist.

[0214] In some embodiments, carboplatin, nanoparticle albumin-bound paclitaxel, paclitaxel, cisplatin, pemetrexed, gemcitabine, FOLFOX, or FOLFIRI is administered to the subject.

[0215] In some embodiments, the additional therapeutic agent is an anti-OX40 antibody, an anti-PD-1 antibody, an anti-PD-L1 antibody, an anti-PD-L2 antibody, an anti-LAG-3 antibody, an anti-TIGIT antibody, an anti-BTLA antibody, an anti-CTLA-4 antibody, or an anti-GITR antibody.

[0216] Pharmaceutical Compositions and Routes of Administration Also provided herein is a pharmaceutical composition comprising at least one (e.g., one, two, three, or four) of the antibodies, antigen-binding fragments, or antibody-drug conjugates described herein. Two or more (e.g., two, three, or four) of any of the antibodies, antigen-binding fragments, or antibody-drug conjugates described herein can be present in any combination in the pharmaceutical composition. The pharmaceutical composition can be formulated by any method known in the art.

[0217] Pharmaceutical compositions are formulated to be compatible with the intended route of administration (e.g., intravenous, intraarterial, intramuscular, intradermal, subcutaneous, or intraperitoneal). The compositions may contain a sterile diluent (e.g., sterile water or saline), fixed oils, polyethylene glycol, glycerin, propylene glycol, or other synthetic solvents; antibacterial or antifungal agents such as benzyl alcohol or methylparabens, chlorobutanol, phenol, ascorbic acid, or thimerosal; antioxidants such as ascorbic acid or sodium bisulfite; chelating agents such as ethylenediaminetetraacetic acid; buffers such as acetates, citrates, or phosphates; and isotonic agents such as sugars (e.g., dextrose), polyalcohols (e.g., mannitol or sorbitol), or salts (e.g., sodium chloride), or any combination thereof. Liposomal suspensions may also be used as pharmaceutically acceptable carriers (see, e.g., U.S. Pat. No. 4,522,811). The compositions may be formulated and enclosed in ampoules, disposable syringes, or multiple-dose vials. Where necessary (e.g., injectable formulations), proper fluidity can be maintained by the use of a coating, such as lecithin or a surfactant. Absorption of the antibody or antigen-binding fragment thereof can be prolonged by the inclusion of agents that delay absorption (e.g., aluminum monostearate and gelatin). Alternatively, controlled release can be achieved by implants and microencapsulated delivery systems that may include biodegradable, biocompatible polymers (e.g., ethylene vinyl acetate, polyanhydrides, polyglycolic acid, collagen, polyorthoesters, and polylactic acid; Alza Corporation and Nova Pharmaceutical, Inc.).

[0218] Compositions comprising one or more of any of the antibodies, antigen-binding fragments, and antibody-drug conjugates described herein can be formulated for parenteral (e.g., intravenous, intraarterial, intramuscular, intradermal, subcutaneous, or intraperitoneal) administration in dosage unit form (i.e., physically discrete units containing a predetermined amount of active compound for ease of administration and uniformity of dosage).

[0219] The toxicity and therapeutic efficacy of a composition can be determined by standard pharmaceutical procedures in cell cultures or experimental animals (e.g., monkeys). The LD50 (the dose lethal to 50% of the population) and ED50 (the dose therapeutically effective in 50% of the population) can be determined; the therapeutic index is the ratio of LD50:ED50. Drugs that exhibit a high therapeutic index are preferred. When a drug exhibits unwanted side effects, care must be taken to minimize potential damage (i.e., reduce the unwanted side effects). Toxicity and therapeutic efficacy can be determined by other standard pharmaceutical procedures.

[0220] Data obtained from cell culture assays and animal studies are used to formulate appropriate dosages of any given agent for use in subjects (e.g., humans). A therapeutically effective amount of one or more (e.g., one, two, three, or four) antibodies or antigen-binding fragments thereof (e.g., any of the antibodies or antibody fragments described herein) is an amount that treats a disease (e.g., kills cancer cells) in a subject (e.g., a human subject identified as having cancer) or reduces the severity, frequency, and / or duration of one or more symptoms of the disease in a subject (e.g., a human) identified as being at risk for developing the disease (e.g., a subject who previously had cancer but has now been cured). The efficacy or dosage of any of the antibodies or antigen-binding fragments described herein can be determined by a medical or veterinary professional using methods well known in the art, as well as by observing one or more symptoms of the disease in a subject (e.g., a human). Certain factors can influence the dosage and timing required to effectively treat a subject (e.g., the severity of the disease or condition, previous treatments, the general health and / or age of the subject, and the presence of other diseases).

[0221] Exemplary doses include milligrams or micrograms of an antibody or antigen-binding fragment, or antibody-drug conjugate described herein per kg of subject body weight (e.g., about 1 μg / kg to about 500 mg / kg; about 100 μg / kg to about 500 mg / kg; about 100 μg / kg to about 50 mg / kg; about 10 μg / kg to about 5 mg / kg; about 10 μg / kg to about 0.5 mg / kg; or about 1 μg / kg to about 50 μg / kg). These doses cover a wide range, but those of skill in the art will understand that effective amounts of therapeutic agents, including antibodies and antigen-binding fragments thereof, of varying potencies can be determined by methods well known in the art. Typically, a relatively low dose is administered initially, and the attending physician or veterinary professional (for therapeutic uses) or researcher (if still working in the development stages) will subsequently gradually increase the dose until an appropriate response is obtained. Moreover, it is understood that the specific dosage level for any particular subject will depend on a variety of factors, including the activity of the particular compound used, the age, body weight, general health, sex, and diet of the subject, the time of administration, the route of administration, the excretion rate, and the half-life of the antibody or antibody fragment in vivo.

[0222] The pharmaceutical compositions can be included in a container, pack, or dispenser together with instructions for administration. The present disclosure also provides methods for producing antibodies or antigen-binding fragments thereof, or antibody-drug conjugates for the various uses described herein. [Example]

[0223] The invention is further described in the following examples, which do not limit the scope of the invention described in the claims.

[0224] Example 1: Methods and Materials The following assays are used in the examples.

[0225] Binding assay (a) 5x10 in 50µL of medium 5 The cells are dispensed into each well of a 96-well plate. (b) 100 μL of antibody is added into the wells at different dilutions. (c) Incubate the plate at room temperature (RT) for 60 minutes. (d) The cells are pelleted by centrifugation and washed three times with phosphate buffered saline (PBS) containing 0.1% bovine serum albumin (BSA). (e) Resuspend the cell pellet in 100 μL of PBS and 0.1% BSA containing 1:500 Cy3-conjugated goat anti-human IgG antibody. (f) Incubate in the dark at room temperature for 30 minutes. (g) Cells are washed three times and resuspended in fluorescence activated cell sorting (FACS) buffer. (h) Analyze the cells using a flow cytometer.

[0226] Antibody-dependent cytotoxicity (ADCC) assay (a) Target cells were washed once with PBS before calcein AM labeling. (b) 1:333 of 2.5 mM calcein AM stock solution is used for labeling of target cells. (c) Incubate the cells at 37°C for 30 minutes in the dark. (d) Wash the cells three times with PBS. (e) 1x10 4 Dispense 50 μL of calcein AM labeled target cells into each well. (f) Add 100 μL of diluted antibody to the wells. (g) Incubate the plate at room temperature for 60 minutes. (h) 5x10 4 of PBMS (which affect the cells) is added to 50 μL of medium in each well (E / T ratio of 5). (i) Incubate the plate at 37°C for 4 hours. (j) The cells are spun down and 180 μL of the supernatant is transferred to another 96-well plate with a translucent bottom and black walls. (k) Read the plate at 485 excitation and 520 emission wavelengths.

[0227] Complement-dependent cytotoxicity (CDC) assay (a) Target cells are harvested and stained with calcein AM for ADCC (assay of calcein release only). (b) 50 μL of target cells (1 x 10 5 Cells) are seeded into each well of a 96-well plate. (c) Add 100 μL of antibody at different concentrations into the wells. (d) Incubate the plate at room temperature for 15 minutes. (e) Add 50 μL of 10% complement-enriched human serum to each well (final concentration 5%). (f) Incubate the plate at room temperature for 45 minutes. (g) Transfer 180 μL of the supernatant to another 96-well plate with a translucent bottom and black walls (for calcein release assay only). (h) Wash the cells three times with PBS containing 0.1% BSA. (i) Cells are stained with 2 μL per well of 7-aminoactinomycin D (7AAD) for 15 minutes at room temperature in the dark. (j) Wash the cells three times and analyze the cells on a flow cytometer.

[0228] T cell activation Pre-activated peripheral blood mononuclear cells (PBMCs) were used in some ADCC experiments. (a) Dynabeads (human T cell activator CD3 / CD28) are used to activate PBMCs. (b) After washing with buffer, Dynabeads are added to PBMCs at a 1:1 ratio along with 30 U / mL interleukin-2 (IL2). (c) Incubating the cell mixture for a sufficient period of time. (d) At the end of the incubation, the beads are removed with a magnet and the stimulated PBMCs are then used in the ADCC assay.

[0229] Cell binding assay with MDA231 cells (a) MDA231 cells were cultured in medium at 1x10 6Prepare at a concentration of cells / mL. (b) Dilute the antibody sample to the appropriate concentration. (c) Transfer 50 μL of cells to each well of a 96-well V-bottom plate. (d) Transfer 50 μL of antibody to each well of a 96-well V-bottom plate. (e) The cell mixture is incubated for 60 minutes at room temperature. (f) Cells are spun down and washed twice with FACS buffer. (g) Resuspend the cells in 100 μL of FACS buffer containing Cy3-conjugated goat anti-human (GAH) IgG antibody in each well (1:500). (h) Incubate at room temperature for 30 minutes and wash twice with FACS buffer. (i) FACS analysis.

[0230] Internalization assay involving MDA231 and SIHA cells (a) 50 μL of cell suspension (MDA231 or SIHA cells) at 1 x 10 6 Add 100 ml of each solution to each well of a 96-well plate. (b) Add 50 μL of Ab to the corresponding wells. (c) Incubate at 37°C for 30 minutes. (d) 100 μL of pHrodo Red-labeled GAH IgG is added to each well and incubated at 37° C. for 24 hours. (d) Trypsinize and collect cells, wash twice, and perform FACS.

[0231] Complement-dependent cytotoxicity (CDC) assay involving MDA231 cells (a) Target cells: MDA231 cells were washed twice with PBS and then diluted to 0.5x10 in PBS. 6 Adjust to a concentration of 1 / mL. (b) Cells are seeded into flat-bottom 24-well plates at 300 μL / well. (c) Add 300 μL of 20 μg / mL antibody to the corresponding wells to a final concentration of 10 μg / mL. (d) Incubate the plate at 37°C for 48 hours. (e) At the end of the incubation, the cells are trypsinized and washed twice with plain medium. (f) Resuspend the cell pellet in 100 μL simple medium and transfer the cells to a 96-well plate. (g) Add 100 μL of 10% complement-enriched serum to each well. (h) Cells are incubated for 4 hours at 37°C. (i) Wash the cells twice with FACS buffer. (j) Add 100 μL trypsin to detach the cells for 3 minutes. (k) Resuspend the cell pellet in FACS buffer containing 7AAD (1:50 dilution). (l) After 15 minutes of incubation at room temperature, the cells are washed twice. (m) Perform FACS analysis.

[0232] Example 2: Bispecific antibodies that bind to CD20 and CD3 A bispecific antibody was designed to bind to CD20 and CD3. This bispecific antibody has two common light chains (with identical sequences) and two different heavy chains. The variable regions of the two heavy chains and the sequence of the common light chain are shown below.

[0233] VHa against CD20 (designed from rituximab VH): TIFF0007801095000001.tif19159 VHb against CD3 (designed from MAb12F6VH): TIFF0007801095000002.tif18160 Common VL (Ritumaxib VL) TIFF0007801095000003.tif11160

[0234] The 12F6 antibody is described, for example, in "Construction and characterization of a humanized anti-human CD3 monoclonal antibody 12F6 with effective immunoregulatory functions," Immunology, 116(4), 487-498 (2005), which is incorporated herein by reference in its entirety. The sequence for the parent antibody is also shown below for comparison purposes.

[0235] Parent CD20VH (ritumaxib VH): TIFF0007801095000004.tif18160 Parent CD20VL (ritumaxib): TIFF0007801095000005.tif11160 Parental CD3VH (MAb12F6VH): TIFF0007801095000006.tif18161 Parental CD3VL (MAb12F6VL): TIFF0007801095000007.tif11158

[0236] The CDR sequences of the redesigned VH and VL are also summarized in the table below.

[0237] Table 1. VHa for CD20 heavy chain TIFF0007801095000008.tif37161

[0238] Table 2: VHb for CD3 heavy chain TIFF0007801095000009.tif37161

[0239] Table 3: VL for common light chains TIFF0007801095000010.tif37161

[0240] The 3D (three-dimensional) isoelectric point (PI) of the ritumaxib Fv (VH+VL) is 9.9, and the 3D PI of the 12F6 Fv is 9.8. After sequence redesign, the 3D PI of VHa+common VL was 10.0, and the 3D PI of VHb+common VL was 9.1. The PI change did not affect the binding affinity of CD20, and the second antigen-binding region still maintained reasonable binding affinity for CD3. The mutations in the two VH chains are shown in the table below.

[0241] Table 4: Altered amino acids in VH (CD20) TIFF0007801095000011.tif15161

[0242] Table 5: Altered amino acids in VH (CD3) TIFF0007801095000012.tif35161

[0243] In Figures 1A and 1B, the antigen-binding capabilities of redesigned ritumaxib (antibody A) and redesigned 12F6 (antibody B), respectively, were tested. Figure 1A shows that redesigned ritumaxib (antibody A) binds to CD20-positive Raji cells. Antibody A is a homodimer with two VHb (SEQ ID NO:1) and two common VL (SEQ ID NO:3). Figure 1B shows that redesigned 12F6 (antibody B) binds to CD3-positive Jurkat cells. Antibody B is also a homodimer with two VHb (SEQ ID NO:2) and two common VL (SEQ ID NO:3). These data suggest that the redesigned ritumaxib heavy chain, redesigned 12F6 heavy chain, and common light chain can be combined into a functional bispecific antibody, for example, via "knob-into-hole" technology.

[0244] Therefore, a CD20 / CD3 "unbalanced bispecific antibody" was designed. Knobandhole mutations were also introduced into the constant region of the heavy chain to facilitate bispecific antibody formation.

[0245] The full-length sequences of the heavy and light chains are shown below.

[0246] Full-length CD20 heavy chain version 1 (wild-type IgG1Fc) TIFF0007801095000013.tif54161 CD20 heavy chain version 2 full length (IgG1 Fc with Y407T (hole) mutation): TIFF0007801095000014.tif55161 CD20 heavy chain version 3 full length (IgG1 Fc with T366Y (knob) mutation): TIFF0007801095000015.tif54161 Full length CD3 heavy chain version 1 (wild type IgG1Fc): TIFF0007801095000016.tif54161 CD3 heavy chain version 2 full length (IgG1 Fc with T366Y (knob) mutation): TIFF0007801095000017.tif54161 CD3 heavy chain version 3 full length (IgG1 Fc with Y407T (hole) mutation): TIFF0007801095000018.tif54161 Total length of common light chain: TIFF0007801095000019.tif26160

[0247] An IgG1 heavy chain against CD20 (version 2; SEQ ID NO: 35) with a Y407T (EU numbering) mutation and an IgG1 heavy chain against CD3 with a T366Y (EU numbering) mutation (version 2; SEQ ID NO: 38) were chosen to generate a bispecific antibody for further experiments, which also has two common light chains (SEQ ID NO: 40).

[0248] This unbalanced bispecific antibody also contains the following features: (1) significantly reduced CD3 binding affinity to increase safety; (2) maintained ADCC / CDC effector function to broaden clinical implementation; and (3) differentiated biochemical and biophysical properties of the CD20-binding arm and the CD3-binding arm to enable better separation of the bispecific antibody during downstream purification processes.

[0249] As shown in the following examples, this antibody had better CD20+ Raji cell-killing efficacy than CD20 homodimers and ritumaxib in the presence of human PBMCs. Meanwhile, under the same conditions, this antibody failed to kill CD3+ Jurkat cells or exhausted normal T cells. Therefore, this antibody demonstrated promise for broader clinical applications than current anti-CD20 cancer therapies: (1) compared to ritumaxib, this antibody possesses T cell-recruiting function; (2) compared to CAR-T / other T cell-recruiting therapies, this antibody maintains functional effector function; and (3) this antibody does not exhibit any safety concerns in vitro. Taken together, the CD20 / CD3 bispecific antibody and platform described in this disclosure can address unmet needs in the field of targeted cancer therapy.

[0250] The bispecific antibodies disclosed herein were purified through two stages: affinity purification using Protein A (Round 1) and anion exchange purification using monoQ5 / 50 (Round 2). In the second round, a gradient pH buffer (e.g., PBS) was used to elute the antibodies. T cell activation assays were performed on different fractions after elution. In Figure 20, the numbers indicate different fractions. Only CD20 / CD3 bispecific antibodies can activate T cells, so the T cell activation assay can assess the purity and content of CD20 / CD3 bispecific antibodies in each fraction. The results showed that fractions 4 to 7 contained relatively pure CD20 / CD3 bispecific antibodies, demonstrating that CD20 / CD3 bispecific antibodies can be purified by the methods described herein.

[0251] Additionally, the pI of the antibodies described herein was also determined, and this information can be useful for choosing an appropriate pH for elution.

[0252] (Table 6) TIFF0007801095000020.tif78128

[0253] Example 3: Binding affinity of bispecific antibodies After computational design, the CD20 homodimeric IgG containing the designed VH sequence (SEQ ID NO:1) and the consensus VL sequence (SEQ ID NO:3) showed similar binding ability to CD20 compared with the parental anti-CD20 IgG (parental CD20). Cell-binding affinity assays were performed using Raji cells (expressing CD20). The binding results are shown in Figure 2A.

[0254] The CD3 homodimeric IgG containing the designed VH sequence (SEQ ID NO:2) and the consensus VL sequence (SEQ ID NO:3) had reduced binding ability to CD3 compared to the parental anti-CD3 IgG (parental CD3). Cell-binding affinity assays were performed using Jurkat cells (expressing CD3). The binding results are shown in Figure 2B.

[0255] Example 4: T cell activation by an imbalanced CD20 / CD3 bispecific antibody An imbalanced CD20 / CD3 bispecific monoclonal antibody (BsMab) activated only T cells in the presence of target tumor cells. Using Raji cells as CD20+ target tumor cells, 293 cells as CD20 control cells, and Jurkat cells as a T cell model, the following experiment was performed to test whether CD20 / CD3 BsMab could activate T cells in the presence of target tumor cells due to clusters formed from multiple BsMabs that bound to both T cells and target tumor cells. In contrast, in the presence of CD20 control cells, CD20 / CD3 BsMab did not activate T cells due to weak binding by one arm to CD3 on T cells.

[0256] The following experimental procedures were used in this example: (1) Raji & Jurkat, Jurkat & 293 1x10 5 are separately seeded into U-bottom 96-well plates. (2) Add test antibody and incubate overnight (19 hours). (3) Wash the cells once with PBS + 0.1% BSA. (4) Add anti-human CD69 antibody (labeled with PE) (1.5 μL / well) and incubate at room temperature for 30 minutes. (5) Wash the cells once. (6) Read.

[0257] The results are shown in Figure 3. The T cell activation efficacy of different concentrations of the test antibody in the presence of Raji cells is shown in Figure 4. An isotype antibody (non-specific IgG1 antibody) was used as a control. T cell activation was measured by the expression of CD69 on the surface of Jurkat cells.

[0258] Example 5: Imbalanced CD20 / CD3 BsMabs induce PBMC-mediated cell killing The imbalanced CD20 / CD3 BsMab showed better PBMC-mediated cell killing than ritumaxib and CD20 homodimeric antibody before and after T cell activation.

[0259] Before T cell activation: Fresh peripheral blood mononuclear cells (PBMCs) from healthy donors were incubated overnight at 37°C and then with calcein-labeled CD20+ Raji cells for 4 hours in the presence of different antibodies as indicated. Cell death was measured by calcein release. The results are shown in Figure 5.

[0260] After T cell activation (4 days): Fresh PBMCs from healthy donors were incubated with recombinant IL-2 and CD3 / CD28 beads for 4 days to activate T cells, followed by incubation with calcein-labeled CD20+ Raji cells for 2 hours in the presence of different antibodies as indicated. Cell death was measured by calcein release. The results are shown in Figure 6.

[0261] After T cell activation (7 days): Fresh PBMCs from healthy donors were incubated with recombinant IL-2 and CD3 / CD28 for 7 days to activate T cells, followed by incubation with calcein-labeled CD20+ Raji cells for 2 hours in the presence of different antibodies as indicated. Cell death was measured by calcein release. The results are shown in Figure 7.

[0262] To address safety concerns about whether the imbalanced CD20 / CD3 BsMAb also kills CD3+ T cells under the same conditions as those described in the PBMC killing assay, CD3+ Jurkat cells were used as a control in each experiment. Only the highest antibody concentration (10 μg / mL) was tested. The results of T cell activation are shown in Figure 8. The results after T cell activation (4 days) are shown in Figure 9. The results after T cell activation (7 days) are shown in Figure 10. The number of Jurkat cells in the PBS-treated group was set as the baseline. Therefore, if the number of Jurkat cells is equal to that of the PBS-treated group, the percentage of killing will be 0. If the number of cells is greater than that of the PBS-treated group, the percentage of killing will be negative.

[0263] As a result, no killing of Jurkat cells was observed before or 4 days after T cell activation. However, killing of Jurkat cells was observed 7 days after T cell activation, and under the same conditions, ritumaxib and CD20 homodimer IgG did not cause killing of Raji cells. This suggests that the killing of Jurkat cells after 7 days was caused by T cell superactivation. To further test whether natural T cells activated for 7 days were also killed in the presence of the imbalanced CD20 / CD3 BsMAb, PBMCs activated for the same 7 days were incubated overnight with the different antibodies shown in the figure, and T cell depletion was checked. The results are shown in Figure 11. LALA in the figure is a CD20 / CD3 BsMAb with L234A and L235A mutations (EU numbering). The antibody with the L234A and L235A mutations has no Fc effector function and was used as a negative control.

[0264] Example 6: T cell depletion with an asymmetric bispecific CD20 / CD3 antibody Experiments were also performed to test whether pre-activated T cells could be depleted by an imbalanced CD20 / CD3 BsMab.

[0265] FIG. 12 shows that resting T cells in PBMCs were not depleted by overnight incubation with imbalanced CD20 / CD3 BsMab.

[0266] Example 7: Induction of complement-dependent cytotoxicity Because the CD20 / CD3 BsMab has an arm that binds to CD20 with high affinity, we performed an experiment to test whether CD20-arm binding is sufficient to induce complement-dependent cytotoxicity. The antibody was incubated with human complement-enriched serum and CD20+ Raji cells. The imbalanced CD20 / CD3 BsMab reduced CDC efficacy compared to ritumaxib and a CD20 homodimeric antibody. The results of FACS (7AAD) detection are shown in Figure 13. The results of calcein release detection are shown in Figure 14.

[0267] Example 8: Safety evaluation We also tested whether CD3+ Jurkat cells and normal T cells were killed by the imbalanced CD20 / CD3 BsMab. Figure 15 shows that high doses of the imbalanced CD20 / CD3 BsMab did not induce CDC on Jurkat cells.

[0268] FIG. 16 shows that the imbalanced CD20 / CD3 BsMab did not induce T cell death after co-incubation with PBMCs and human serum, including human complement-enriched serum.

[0269] Example 9: Imbalanced CD20 / CD3 BsMab can kill ritumaxib-resistant Raji cells To test whether imbalanced CD20 / CD3 BsMabs could kill ritumaxib-resistant Raji cells (RRCLs), RRCLs were incubated with 7-day-activated PBMCs from three different donors in the presence of the indicated antibodies. Significant killing of RRCLs in the presence of imbalanced CD20 / CD3 BsMabs was observed (Figures 17-19).

[0270] Example 10: Animal studies on imbalanced CD20 / CD3 BsMabs Experiments were performed to evaluate the efficacy of CD20 / CD3 BsMab in animals.

[0271] Raji cells, human PBMCs, and unbalanced CD20 / CD3 BsMA were mixed and injected intravenously into mice. These Raji cells were luciferase-labeled. Each mouse in the treatment group (B-NDG, Biocytogen, Beijing, Cat# 201811808) received 5x10 5 Raji cells, 2.5x10 6 Mice were administered human PBMC cells and 60 μg of antibody. Mice were imaged on days 0, 2, 3, and every 3 days thereafter to track Raji cell depletion.

[0272] On day 0, luciferase-labeled Raji cells and human PBMC cells were mixed with either phosphate-buffered saline (PBS) (group G1; reference; n=4), CD20 / CD3 BsMab (group G2; n=4), or ritumaxib (anti-CD20 antibody; group G3; n=4). The mice were first imaged 15 minutes after intravenous (iv) injection, then imaged on days 2, 3, and every 3 days thereafter.

[0273] Figure 21A shows that CD20 / CD3 BsMab and ritumaxib had no obvious toxic effects. Figure 21B shows that both CD20 / CD3 BsMab and ritumaxib had tumor-inhibitory effects, with ritumaxib being less effective than CD20 / CD3 BsMab. The difference in tumor-inhibitory effect was observed 16 days after injection.

[0274] Example 11: Characterization of unbalanced bispecific antibodies Experiments were performed to characterize the purified CD20 / CD3 bispecific antibody samples.

[0275] First, reduced capillary electrophoresis (Re-CE-SDS) was performed on the purified CD20 / CD3 bispecific antibody sample. The results showed three main peaks. Based on molecular weight, peak #1 was the common light chain (LC), and peaks #2 and #3 were the two distinct heavy chains (HC) (Figure 22A).

[0276] Non-reduced CE (non-Re-CE-SDS) was also performed. The results showed that there was one main peak of CD20 / CD3 bispecific IgG (Figure 22B). The results in Figures 22A and 22B suggest that the CD20 / CD3 bispecific antibody sample had good purity.

[0277] Second, differential scanning fluorimetry (DSF) was performed to measure the protein melting temperature (Tm), and static light scattering (SLS) was performed to measure the aggregation temperature at 266 nm (Tagg266) and 473 nm (Tagg473). These samples were submitted to the UNcle system for analysis. A 1°C / min temperature ramp was performed for DSF and SLS, with monitoring from 20°C to 95°C. UNcle measured SLS at 266 nm and 473 nm. Tm and Tag were calculated and analyzed using UNcle analysis software.

[0278] Some tested antibodies had two Tm values, and some had three. This is due to the multidomain structure of IgG. The CH2 domain typically has a Tm of approximately 70°C in PBS, while the CH3 domain is more stable, with a Tm of approximately 80°C. Fabs have a wide range of Tm values, from approximately 50 to 85°C, due to their large sequence variability. Therefore, the Tm values ​​measured by various analytical techniques are usually "apparent" transition temperatures rather than formal melting temperatures. For whole IgG antibodies, there are often two to three Tm values ​​in DSF measurements. It is not easy to determine which Tm represents which domain.

[0279] In the case of this bispecific antibody, the Tm of 86.7°C likely represents the CH3 domain alone. The other one or two lower Tms represent Fab, CH2, or Fab+CH2.

[0280] Tagg is the temperature at which SLS begins to detect agglomeration. Tagg266 measures SLS at 266nm, which is more sensitive and better suited to detecting smaller particles. Tagg473 measures at 473nm, better suited to measuring larger particles.

[0281] Both the DSF and SLS data indicate that the CD20 / CD3 bispecific antibody has good thermal stability.

[0282] (Table 7) TIFF0007801095000021.tif31161

[0283] Third, dynamic light scattering (DLS) detected only molecular particles of one size (10.15 nm), indicating the absence of aggregation in the sample.

[0284] (Table 8) TIFF0007801095000022.tif25161

[0285] These characterization data suggest that CD20 / CD3 bispecific antibodies have good potential as therapeutic antibodies.

[0286] Example 12: Bispecific antibodies that bind to PD-L1 and CD55 Two versions of the bispecific antibody were designed to bind to PD-L1 and CD55 (PD-L1 / CD55 BsMab v1 and PD-L1 / CD55 BsMab v2). These bispecific antibodies have two common light chains and two different heavy chains.

[0287] The sequences of the variable regions of the two heavy chains and the common light chain of the first version of the bispecific antibody (PD-L1 / CD55 BsMab v1) are shown below.

[0288] VHa against PD-L1 (designed from avelumab): TIFF0007801095000023.tif19161 VHb against CD55 (designed from CD55 ScFV): TIFF0007801095000024.tif19160 Common VL (designed from CD55 ScFV): TIFF0007801095000025.tif18159

[0289] CD55 ScFV is described, for example, in "Identification of a human anti-CD55 single-chain Fv by subtractive panning of a phage library using tumor and nontumor cell lines," Cancer Res. 59(11), 2718-2723 (1999), which is incorporated herein by reference in its entirety. The sequence of this parent antibody is also shown below for comparison purposes.

[0290] Parent PD-L1VH: TIFF0007801095000026.tif18161 Parental PD-L1VL: TIFF0007801095000027.tif19159 Parental CD55VH: TIFF0007801095000028.tif18160 Parental CD55VL: TIFF0007801095000029.tif18159

[0291] The 3DPI of the avelumab Fv (VH+VL) was 9.4, and the 3DPI of the anti-CD55 Fv was 9.8. After sequence redesign, the 3DPI of the VHa+common VL was 9.9, and the 3DPI of the VHb+common VL was 9.3. The mutations for the two VH chains are shown in the table below.

[0292] Table 9. Altered amino acids in VH (PD-L1) TIFF0007801095000030.tif20161

[0293] Table 10. Altered amino acids in VH (CD55) TIFF0007801095000031.tif20161

[0294] Example 13: Binding Affinities of Newly Designed PD-L1 and CD55 Antibodies Experiments were performed to determine the binding affinities of newly designed PD-L1 and CD55 antibodies.

[0295] Anti-PD-L1 homodimeric IgG (PD-L1v1), containing a designed VH sequence (SEQ ID NO:4) and a consensus VL sequence (SEQ ID NO:6), had weaker binding affinity than the parental anti-PD-L1 antibody (PD-L1wt) (Figure 23A). Anti-CD55 homodimeric IgG (CD55v1), containing a designed VH sequence (SEQ ID NO:5) and a consensus VL sequence (SEQ ID NO:6), had similar binding affinity compared to the parental anti-CD55 antibody (CD55wt) (Figure 23B).

[0296] The antibodies (CD55v1 and PD-L1v1) did not meet this requirement because the bispecific antibody must bind to the cancer-specific antigen (PD-L1) with high affinity and the other arm of the bispecific antibody must bind to the cancer-associated antigen (CD55) with low affinity.

[0297] Therefore, a second version of the bispecific antibody was designed to bind to PD-L1 and CD55 (PD-L1 / CD55 BsMab v2). The VHa and VHb for the second version of the bispecific antibody were identical to the VHa and VHb of the same first version of the bispecific antibody. However, the common light chain was redesigned based on the methods described herein. The sequence of the redesigned common light chain is shown below.

[0298] Common VL2 (redesigned from SEQ ID NO:6): TIFF0007801095000032.tif19159

[0299] An alignment of consensus VL (SEQ ID NO:6) and consensus VL2 (SEQ ID NO:7) is shown in Figure 24. The underlined sequence is the sequence of the light chain constant region.

[0300] The CDR sequences of the redesigned VH and VL are shown below.

[0301] Table 11. VHa for PD-L1 heavy chain TIFF0007801095000033.tif37161

[0302] Table 12. VHbs for CD55 heavy chain TIFF0007801095000034.tif37161

[0303] Table 13: VL version 1 for common VL TIFF0007801095000035.tif37161

[0304] (Table 14) VL version 2 for common VL TIFF0007801095000036.tif37161

[0305] Furthermore, since lambda light chains are less common than kappa light chains in human serum, in the examples the constant region of the lambda light chain was replaced with the constant region of the kappa light chain.

[0306] Experiments were conducted to determine the binding affinity of the second version of the antibody. Anti-PD-L1 homodimeric IgG (PD-L1v2), containing a designed VH sequence (SEQ ID NO:4) and a consensus VL2 sequence (SEQ ID NO:7), had similar binding affinity compared to the parent anti-PD-L1 antibody (PD-L1wt) (Figure 25A). Anti-CD55 homodimeric IgG (CD55v2), containing a designed VH sequence (SEQ ID NO:5) and a consensus VL2 sequence (SEQ ID NO:7), had weaker binding affinity compared to the parent anti-CD55 antibody (CD55wt) (Figure 25B). Therefore, the binding affinity of the antibody with the redesigned sequence met the requirements, and PD-L1 / CD55 BsMab v2 was selected for further experiments. PD-L1 / CD55 BsMab v2 has two common light chains (κ chains) comprising SEQ ID NO: 7, one IgG1 heavy chain comprising SEQ ID NO: 4, and one IgG1 heavy chain comprising SEQ ID NO: 5. Furthermore, the heavy chain for PD-L1 has a Y407T mutation (EU numbering), and the IgG1 heavy chain for CD55 has a T366Y mutation (EU numbering).

[0307] The full-length sequences of these heavy and light chains are shown below.

[0308] PD-L1 heavy chain length: TIFF0007801095000037.tif54161 CD55 heavy chain full length: TIFF0007801095000038.tif54161 CD55 common light chain version 1 full length: TIFF0007801095000039.tif26160 CD55 common light chain version 2 full length: TIFF0007801095000040.tif26160

[0309] Additionally, the pI of the antibodies described herein was also determined, and this information is important for selecting the appropriate pH for elution.

[0310] (Table 15) TIFF0007801095000041.tif60132

[0311] Example 14: Complement-dependent cytotoxicity (CDC) for PD-L1 / CD55 bispecific antibodies Experiments were conducted to test complement-dependent cytotoxicity for PD-L1 / CD55 bispecific antibodies. The parent antibodies were included for comparison purposes. The assay was performed on MDA231 cells according to the protocol described herein, and the concentration of each antibody was 10 μg / mL. The results are shown in Figures 26A-26B.

[0312] As shown in the figure, both the anti-PD-L1 (PD-L1wt) and anti-CD55 (CD55wt) parental antibodies can induce CDC. PD-L1 / CD55 bispecific antibody v1 had much weaker CDC compared to the parental anti-PD-L1 antibody (PD-L1wt) and parental anti-CD55 antibody (CD55wt). In contrast, PD-L1 / CD55 bispecific antibody v2 had much stronger CDC compared to the first version, parental anti-PD-L1 antibody (PD-L1wt), and parental anti-CD55 antibody (CD55wt). The CDC efficacy of PD-L1 / CD55 bispecific antibody v2 was approximately 4.5-fold higher than the CDC efficacy of the first version of the bispecific antibody.

[0313] Example 15: Internalization induced by PD-L1 / CD55 bispecific antibodies Experiments were performed to assess internalization induced by PD-L1 / CD55 bispecific antibodies.

[0314] Internalization assays were performed on two versions of the PD-L1 / CD55 bispecific antibody and its parent antibody. MDA231 cells were used in the first internalization experiment (Figure 27A), and SIHA cells were used in the second internalization experiment (Figure 27B). Cells were mixed with 20 μg / mL antibody and incubated at 37°C for 30 minutes. pHrodo-labeled secondary antibody was then added, and the cells were incubated at 37°C for 24 hours. Cells were then harvested and analyzed by FACS.

[0315] CD55 is a receptor for echovirus and coxsackie B virus infection and is known to be an internalization receptor. Accordingly, the anti-CD55 parental monoclonal antibody (CD55wt) induced rapid internalization of CD55. As shown in Figure 27A, in MDA231 cells with similar PD-L1 and CD55 expression levels, the internalization induced by the anti-PD-L1 antibody was much slower than that of CD55. However, both PD-L1 / CD55 bispecific antibodies v1 and v2 were able to induce internalization, and this internalization rate was comparable to that of the anti-CD55 parental monoclonal antibody.

[0316] In Figure 27B, CD55 expression is higher than that of PD-L1 in SIHA cells. PD-L1 / CD55 bispecific antibodies v1 and v2 can induce better internalization than the parent anti-PD-L1 and anti-CD55 antibodies. Nevertheless, considering the reduced CD55 binding of PD-L1 / CD55 BsMab v2 compared to v1, PD-L1 / CD55 BsMab v2 should have a better efficacy / safety balance in vivo and be safer than PD-L1 / CD55 BsMab v1. Therefore, PD-L1 / CD55 BsMabs are expected to induce the death of target cancer cells at three different levels: (1) blocking PD1 / PD-L1 interaction; (2) inducing PD-L1 internalization; and (3) when conjugated with a drug, the antibody-drug conjugate can kill cancer cells.

[0317] Other Examples While the present invention has been described in conjunction with the detailed description thereof, it should be understood that the foregoing description is illustrative, but not limiting, of the scope of the invention as defined by the appended claims. Other aspects, advantages, and modifications are within the scope of the appended claims.

Claims

1. a first heavy chain variable region (VH1) comprising complementarity determining regions (CDRs) 1, 2, and 3; a second heavy chain variable region (VH2) comprising CDR1, 2, and 3; a first light chain variable region (VL1) comprising CDR1, 2, and 3; and A second light chain variable region (VL2) comprising CDRs 1, 2, and 3 A bispecific antibody or antigen-binding fragment thereof comprising: the VH1 CDR1 region comprises the amino acid sequence of SEQ ID NO: 16, the VH1 CDR2 region comprises the amino acid sequence of SEQ ID NO: 17, and the VH1 CDR3 region comprises the amino acid sequence of SEQ ID NO: 18; the VH2 CDR1 region comprises the amino acid sequence of SEQ ID NO: 22, the VH2 CDR2 region comprises the amino acid sequence of SEQ ID NO: 23, and the VH2 CDR3 region comprises the amino acid sequence of SEQ ID NO: 24; the VL1 CDR1 region comprises the amino acid sequence of SEQ ID NO: 28, the VL1 CDR2 region comprises the amino acid sequence of SEQ ID NO: 29, and the VL1 CDR3 region comprises the amino acid sequence of SEQ ID NO: 30; the VL2 CDR1 region comprises the amino acid sequence of SEQ ID NO: 28, the VL2 CDR2 region comprises the amino acid sequence of SEQ ID NO: 29, and the VL2 CDR3 region comprises the amino acid sequence of SEQ ID NO: 30; the first heavy chain variable region and the first light chain variable region bind to each other to form a first antigen-binding region that specifically binds to a first antigen; and the second heavy chain variable region and the second light chain variable region bind to each other to form a second antigen-binding region that specifically binds to a second antigen; the binding affinity of the first antigen-binding region when binding to the first antigen is at least 100-fold higher than the binding affinity of the second antigen-binding region when binding to the second antigen, and the binding affinity of the second antigen-binding region when binding to the second antigen is greater than 10 4 M −1 , 10 5 M −1 or 10 6 M −1 ; The binding affinity is measured by the binding constant Ka, the first light chain variable region and the second light chain variable region are identical; The first antigen is CD20 and the second antigen is CD3; The bispecific antibody or antigen-binding fragment thereof.

2. The first antigen-binding region is 8 M -1 2. The bispecific antibody or antigen-binding fragment thereof of claim 1, which specifically binds to the first antigen with a binding affinity greater than that of the first antigen.

3. 2. The bispecific antibody or antigen-binding fragment thereof according to claim 1, comprising a first heavy chain comprising the first heavy chain variable region and a second heavy chain comprising the second heavy chain variable region, wherein the first heavy chain and the second heavy chain are bound to each other by a knobs-into-holes method.

4. A cell comprising one or more nucleic acids encoding the bispecific antibody or antigen-binding fragment thereof of claim 1.

5. 2. The bispecific antibody or antigen-binding fragment thereof of claim 1, comprising a first polypeptide, a second polypeptide, a third polypeptide, and a fourth polypeptide, the first polypeptide comprises an amino acid sequence at least 90% identical to SEQ ID NO: 34; the second polypeptide comprises an amino acid sequence at least 90% identical to SEQ ID NO: 37; the third polypeptide comprises an amino acid sequence at least 90% identical to SEQ ID NO: 40; and the fourth polypeptide comprises an amino acid sequence at least 90% identical to SEQ ID NO: 40; The bispecific antibody or antigen-binding fragment thereof.

6. 10. A pharmaceutical composition for treating a subject suffering from cancer, comprising a therapeutically effective amount of a composition comprising the bispecific antibody or antigen-binding fragment thereof of claim 1.

7. 7. The pharmaceutical composition of claim 6, wherein the subject has a solid tumor, melanoma, pancreatic cancer, a hematological malignancy, non-Hodgkin's lymphoma, lymphoma, or chronic lymphocytic leukemia.

8. 2. The bispecific antibody or antigen-binding fragment thereof of claim 1, comprising a first polypeptide, a second polypeptide, a third polypeptide, and a fourth polypeptide, the first polypeptide comprises a VH1 comprising complementarity-determining regions (CDRs) 1, 2, and 3, wherein the VH1 CDR1 region comprises the amino acid sequence of SEQ ID NO: 16, the VH1 CDR2 region comprises the amino acid sequence of SEQ ID NO: 17, and the VH1 CDR3 region comprises the amino acid sequence of SEQ ID NO: 18; the second polypeptide comprises a VH2 comprising CDR1, 2, and 3, wherein the VH2 CDR1 region comprises the amino acid sequence of SEQ ID NO: 22, the VH2 CDR2 region comprises the amino acid sequence of SEQ ID NO: 23, and the VH2 CDR3 region comprises the amino acid sequence of SEQ ID NO: 24; the third polypeptide comprises a VL1 comprising CDR1, 2, and 3, wherein the VL1 CDR1 region comprises the amino acid sequence of SEQ ID NO: 28, the VL1 CDR2 region comprises the amino acid sequence of SEQ ID NO: 29, and the VL1 CDR3 region comprises the amino acid sequence of SEQ ID NO: 30; and the fourth polypeptide comprises a VL2 comprising CDR1, 2, and 3, wherein the VL2 CDR1 region comprises the amino acid sequence of SEQ ID NO: 28, the VL2 CDR2 region comprises the amino acid sequence of SEQ ID NO: 29, and the VL2 CDR3 region comprises the amino acid sequence of SEQ ID NO: 30; The bispecific antibody or antigen-binding fragment thereof.

9. VH1 comprises the amino acid sequence of SEQ ID NO: 1; VH2 comprises the amino acid sequence of SEQ ID NO: 2; VL1 comprises the amino acid sequence of SEQ ID NO: 3; and VL2 comprises the amino acid sequence of SEQ ID NO: 3; The bispecific antibody or antigen-binding fragment thereof according to claim 8.

10. the first polypeptide comprises an amino acid sequence at least 90% identical to SEQ ID NO: 34, 35, or 36; the second polypeptide comprises an amino acid sequence at least 90% identical to SEQ ID NO: 37, 38, or 39; the third polypeptide comprises an amino acid sequence at least 90% identical to SEQ ID NO: 40; and the fourth polypeptide comprises an amino acid sequence at least 90% identical to SEQ ID NO: 40; The bispecific antibody or antigen-binding fragment thereof according to claim 8.

11. the first polypeptide comprises the amino acid sequence of SEQ ID NO: 35; and the second polypeptide comprises the amino acid sequence of SEQ ID NO:38; The bispecific antibody or antigen-binding fragment thereof according to claim 8.

12. VH1, comprising a VH1 CDR1, a VH1 CDR2, and a VH1 CDR3 that are identical to the CDR1, CDR2, and CDR3 of SEQ ID NO:1; a VH2 comprising a VH2 CDR1, a VH2 CDR2, and a VH2 CDR3 that are identical to the CDR1, CDR2, and CDR3 of SEQ ID NO:2; a VL1 comprising a VL1 CDR1, a VL1 CDR2, and a VL1 CDR3 that are identical to the CDR1, CDR2, and CDR3 of SEQ ID NO:3; and VL2, comprising VL2 CDR1, VL2 CDR2, and VL2 CDR3 that are identical to CDR1, CDR2, and CDR3 of SEQ ID NO:3 2. The bispecific antibody or antigen-binding fragment thereof of claim 1, comprising: VH1 comprises an amino acid sequence at least 90% identical to SEQ ID NO:1; VH2 comprises an amino acid sequence at least 90% identical to SEQ ID NO:2; VL1 comprises an amino acid sequence at least 90% identical to SEQ ID NO:3; and VL2 comprises an amino acid sequence at least 90% identical to SEQ ID NO:3, The bispecific antibody or antigen-binding fragment thereof.

13. VH1 comprises the amino acid sequence of SEQ ID NO: 1; VH2 comprises the amino acid sequence of SEQ ID NO: 2; VL1 comprises the amino acid sequence of SEQ ID NO: 3; and VL2 comprises the amino acid sequence of SEQ ID NO: 3; The bispecific antibody or antigen-binding fragment thereof according to claim 1.

14. VH1 comprises an amino acid sequence at least 90% identical to SEQ ID NO:1; VH2 comprises an amino acid sequence at least 90% identical to SEQ ID NO:2; VL1 comprises an amino acid sequence at least 90% identical to SEQ ID NO:3; and VL2 comprises an amino acid sequence at least 90% identical to SEQ ID NO:3, The bispecific antibody or antigen-binding fragment thereof according to claim 8.

15. VH1 comprises an amino acid sequence at least 95% identical to SEQ ID NO:1; VH2 comprises an amino acid sequence at least 95% identical to SEQ ID NO:2; VL1 comprises an amino acid sequence at least 95% identical to SEQ ID NO:3; and VL2 comprises an amino acid sequence at least 95% identical to SEQ ID NO:3, The bispecific antibody or antigen-binding fragment thereof according to claim 8.

16. VH1 comprises an amino acid sequence at least 97% identical to SEQ ID NO:1; VH2 comprises an amino acid sequence at least 97% identical to SEQ ID NO:2; VL1 comprises an amino acid sequence at least 97% identical to SEQ ID NO:3; and VL2 comprises an amino acid sequence at least 97% identical to SEQ ID NO:3, The bispecific antibody or antigen-binding fragment thereof according to claim 8.

17. VH1 comprises an amino acid sequence at least 98% identical to SEQ ID NO:1; VH2 comprises an amino acid sequence at least 98% identical to SEQ ID NO:2; VL1 comprises an amino acid sequence at least 98% identical to SEQ ID NO:3; and VL2 comprises an amino acid sequence at least 98% identical to SEQ ID NO:3, The bispecific antibody or antigen-binding fragment thereof according to claim 8.

18. VH1 comprises an amino acid sequence at least 99% identical to SEQ ID NO:1; VH2 comprises an amino acid sequence at least 99% identical to SEQ ID NO:2; VL1 comprises an amino acid sequence at least 99% identical to SEQ ID NO:3; and VL2 comprises an amino acid sequence at least 99% identical to SEQ ID NO:3, The bispecific antibody or antigen-binding fragment thereof according to claim 8.

19. the first polypeptide comprises an amino acid sequence at least 95% identical to SEQ ID NO: 34, 35, or 36; the second polypeptide comprises an amino acid sequence at least 95% identical to SEQ ID NO: 37, 38, or 39; the third polypeptide comprises an amino acid sequence at least 95% identical to SEQ ID NO: 40; and The fourth polypeptide comprises an amino acid sequence that is at least 95% identical to SEQ ID NO:

40. The bispecific antibody or antigen-binding fragment thereof according to claim 10.

20. the first polypeptide comprises an amino acid sequence at least 96% identical to SEQ ID NO: 34, 35, or 36; the second polypeptide comprises an amino acid sequence at least 96% identical to SEQ ID NO: 37, 38, or 39; the third polypeptide comprises an amino acid sequence at least 96% identical to SEQ ID NO: 40; and The fourth polypeptide comprises an amino acid sequence that is at least 96% identical to SEQ ID NO:

40. The bispecific antibody or antigen-binding fragment thereof according to claim 10.

21. the first polypeptide comprises an amino acid sequence at least 97% identical to SEQ ID NO: 34, 35, or 36; the second polypeptide comprises an amino acid sequence at least 97% identical to SEQ ID NO: 37, 38, or 39; the third polypeptide comprises an amino acid sequence at least 97% identical to SEQ ID NO: 40; and The fourth polypeptide comprises an amino acid sequence that is at least 97% identical to SEQ ID NO:

40. The bispecific antibody or antigen-binding fragment thereof according to claim 10.

22. the first polypeptide comprises an amino acid sequence at least 98% identical to SEQ ID NO: 34, 35, or 36; the second polypeptide comprises an amino acid sequence at least 98% identical to SEQ ID NO: 37, 38, or 39; the third polypeptide comprises an amino acid sequence at least 98% identical to SEQ ID NO: 40; and The fourth polypeptide comprises an amino acid sequence that is at least 98% identical to SEQ ID NO:

40. The bispecific antibody or antigen-binding fragment thereof according to claim 10.

23. the first polypeptide comprises an amino acid sequence at least 99% identical to SEQ ID NO: 34, 35, or 36; the second polypeptide comprises an amino acid sequence at least 99% identical to SEQ ID NO: 37, 38, or 39; the third polypeptide comprises an amino acid sequence at least 99% identical to SEQ ID NO: 40; and The fourth polypeptide comprises an amino acid sequence that is at least 99% identical to SEQ ID NO:

40. The bispecific antibody or antigen-binding fragment thereof according to claim 10.

24. The second antigen-binding region is 6 M -1 2. The bispecific antibody or antigen-binding fragment thereof of claim 1, which specifically binds to CD3 with a binding affinity, as measured by a binding constant Ka, that is lower than

25. 2. The bispecific antibody or antigen-binding fragment thereof according to claim 1, having a functional Fc region capable of inducing antibody-dependent cell-mediated cytotoxicity (ADCC).

Images