Heterodimeric antibodies that bind ENPP3 and CD3

Abstract

The present invention relates to antibodies comprising novel antigen-binding domains, and heterodimeric antibodies that bind to extracellular nucleotide pyrophosphatase / phosphodiesterase family member 3 (ENPP3).

Description

[0001] Priority claim

[0002] This application claims priority to U.S. Provisional Application No. 62 / 812,922, filed on March 1, 2019, and U.S. Provisional Application No. 62 / 929,687, filed on November 1, 2019, which are hereby incorporated by reference in their entireties. Background Art

[0003] Antibody-based therapeutics have been successfully used to treat a variety of diseases, including cancer. An increasingly common approach being explored is the engineering of a single immunoglobulin molecule that co-binds two different antigens. This alternative antibody format that binds two different antigens is often referred to as a bispecific antibody. Because the considerable diversity of the antibody variable region (Fv) allows for the generation of Fvs that recognize virtually any molecule, the typical approach to generating bispecific antibodies is to introduce a novel variable region into the antibody.

[0004] A particularly useful approach for bispecific antibodies is to engineer a first binding domain that engages CD3 and a second binding domain that engages an antigen associated with or upregulated on cancer cells, such that the bispecific antibody redirects CD3. + T cells to destroy cancer cells. It has been previously reported that extracellular nucleotide pyrophosphatase / phosphodiesterase family member 3 (ENPP3) is highly expressed in renal cell carcinoma and minimally expressed in normal tissues. In view of this, it is believed that anti-ENPP3 antibodies can be used, for example, to target anti-tumor therapeutic agents (e.g., chemotherapeutic agents and T cells) to such ENPP3-expressing tumors. + Novel bispecific antibodies against CD3 and ENPP3 localize effector T cells to ENPP3-expressing tumors. Summary of the Invention

[0005] Thus, provided herein are ENPP3 antigen-binding domains and anti-ENPP3 antibodies (eg, bispecific antibodies).

[0006] In one aspect, provided herein is a composition comprising an extracellular nucleotide pyrophosphatase / phosphodiesterase family member 3 (ENPP3) binding domain comprising the variable heavy chain complementarity determining regions 1-3 (vhCDR1-3) and the variable light chain complementarity determining regions (vlCDR1-3) of any of the following ENPP3 binding domains: AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, H16-1.67, H16-1.23, H16-1.44, and Ha16-1.67. (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I ). In some embodiments, the vhCDR1-3 and the vlCDR1-3 are selected from Figure 12 、 13A to 13B ,and 14A to 14I The vhCDR1-3 sequences and vlCDR1-3 sequences of the ENPP3 binding domain are provided in .

[0007] In another aspect, provided herein is a composition comprising an extracellular nucleotide pyrophosphatase / phosphodiesterase family member 3 (ENPP3) binding domain comprising a variable heavy chain domain and a variable light chain domain of any of the following ENPP3 binding domains: AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, H16-9.44, H16-1.67, Ha16-1(2). (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha16-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I ).

[0008] In another aspect, the present invention provides a composition comprising an extracellular nucleotide pyrophosphatase / phosphodiesterase family member 3 (ENPP3) binding domain, wherein the ENPP3 binding domain is selected from the following ENPP3 binding domains: AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, H16-9.44, H16-1.67, Ha16-1(2). (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I ).

[0009] In another aspect, the present invention provides a nucleic acid composition comprising: a) a first nucleic acid encoding a variable heavy chain domain comprising variable heavy chain complementary determining regions 1-3 (vhCDR1-3) of an ENPP3 binding domain; and b) a second nucleic acid encoding a variable light chain domain comprising variable light chain complementary determining regions 1-3 (vlCDR1-3) of an ENPP3 binding domain, wherein the ENPP3 binding domain is one of the following ENPP3 binding domains: AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.8, AN1[ENPP3] H1 L1.8. L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1 (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha16-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1、H16-1.61、H16-1(3,5)5、H16-7.200、H16-1(3,5)42、H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I ). In some embodiments, the vhCDR1-3 and the vlCDR1-3 are selected from Figure 12 、 13A to 13B ,and 14A to 14I The vhCDR1-3 sequences and vlCDR1-3 sequences are provided in .

[0010] In another aspect, the present invention provides a nucleic acid composition comprising: a) a first nucleic acid encoding a variable heavy chain domain comprising a variable heavy chain domain of an ENPP3 binding domain; and b) a second nucleic acid encoding a variable light chain domain comprising a variable light chain domain of an ENPP3 binding domain, wherein the ENPP3 binding domain is any one of the following ENPP3 binding domains: AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1 (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1、H16-1.61、H16-1(3,5)5、H16-7.200、H16-1(3,5)42、H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I ).

[0011] In some embodiments, the present invention provides an expression vector composition comprising: a) a first expression vector comprising a first nucleic acid; b) a second expression vector comprising a second nucleic acid. In other embodiments, the present invention provides a host cell comprising the expression vector composition.

[0012] In some embodiments, the present invention provides a method for preparing an extracellular nucleotide pyrophosphatase / phosphodiesterase family member 3 (ENPP3) binding domain, the method comprising culturing a host cell under conditions that express the ENPP3 binding domain, and recovering the ENPP3 binding domain.

[0013] In another aspect, the present invention provides an anti-ENPP3 antibody comprising an extracellular nucleotide pyrophosphatase / phosphodiesterase family member 3 (ENPP3) binding domain, wherein the ENPP3 binding domain comprises the variable heavy chain complementary determining regions 1-3 (vhCDR1-3) and the variable light chain complementary determining regions (vlCDR1-3) of any of the following ENPP3 binding domains: AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, H16-1(1)23, H16-9.44, H16-1.67, Ha16-1(1)23. 6-1 (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I ). In some embodiments, vhCDR1-3 and vlCDR1-3 are selected from Figure 12 、 13A to 13B ,and 14A to 14I The vhCDR1-3 and vlCDR1-3 of any of the following ENPP3 binding domains in.

[0014] In another aspect, the present invention provides an anti-ENPP3 antibody comprising an extracellular nucleotide pyrophosphatase / phosphodiesterase family member 3 (ENPP3) binding domain, wherein the ENPP3 binding domain comprises a variable heavy chain domain and a variable light chain domain of any of the following ENPP3 binding domains: AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, H16-9.44, H16-1.67, Ha16-1(2). (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I ).

[0015] In another aspect, provided herein is an anti-ENPP3 antibody comprising an extracellular nucleotide pyrophosphatase / phosphodiesterase family member 3 (ENPP3) binding domain, wherein the ENPP3 binding domain is selected from any one of the following ENPP3 binding domains: AN1[ENPP3] H1L1, AN1[ENPP3] H1L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, H16-9.44, H16-1.67, Ha16-1(2). 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I ).

[0016] In some embodiments, the antibody comprises: a) a first monomer comprising a first antigen-binding domain and a first constant domain; and b) a second monomer comprising a second antigen-binding domain and a second constant domain, wherein either the first antigen-binding domain or the second antigen-binding domain is an ENPP3 binding domain. In other embodiments, the first antigen-binding domain and the second antigen-binding domain bind to different antigens. In other embodiments, the first antigen-binding domain is an ENPP3 binding domain, and the second antigen-binding domain is a CD3 binding domain. In other embodiments, the CD3 binding domain comprises the vhCDR1-3 and vlCDR1-3 of any of the following CD3 binding domains: H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10F In other embodiments, the vhCDR1-3 and the vlCDR1-3 of the CD3 binding domain are selected from 10A to 10F vhCDR1-3 and vlCDR1-3 in.

[0017] In some embodiments, the CD3 binding domain comprises the variable heavy chain domain and the variable light chain domain of any of the following CD3 binding domains: H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10F ).

[0018] In some embodiments, the CD3 binding domain is an anti-CD3 scFv.

[0019] In some embodiments, the first constant domain and the second constant domain each comprise CH2-CH3.

[0020] In some embodiments, the first constant domain and the second constant domain each comprise CH1-hinge-CH2-CH3.

[0021] In some embodiments, the first constant domain and the second constant domain are each variant constant domains.

[0022] In some embodiments, the first monomer and the second monomer comprise a set of heterodimerization variants, wherein the set of heterodimerization variants is Figures 1A to 1E In some embodiments, the set of heterodimerization variants includes one of the following set of variants: S364K / E357Q:L368D / K370S; S364K:L368D / K370S; S364K:L368E / K370S; D401K:T411E / K360E / Q362E; and T366W:T366S / L368A / Y407V.

[0023] In some embodiments, the first monomer and the second monomer each further comprise an ablation variant. In other embodiments, the ablation variant is E233P / L234V / L235A / G236del / S267K.

[0024] In some embodiments, at least one of the first monomer or the second monomer further comprises a pI variant. In some embodiments, the pI variant is N208D / Q295E / N384D / Q418E / N421D. In some embodiments, the scFv comprises a charged scFv linker.

[0025] In some embodiments, the present invention provides a nucleic acid composition comprising a nucleic acid encoding an anti-ENPP3. In some embodiments, the composition comprises a nucleic acid encoding a first monomer and a second monomer. In some embodiments, the present invention provides an expression vector comprising the nucleic acid. In some embodiments, the present invention provides a host cell transformed with the expression vector.

[0026] In some embodiments, the present invention provides a method for producing an anti-ENPP3 antibody according to any one of claims B1 to B21. The method comprises culturing the host cell according to claim B25 under conditions that express the anti-ENPP3 antibody, and recovering the anti-ENPP3 antibody. In some embodiments, the present invention provides a method for treating cancer, comprising administering the antibody to a patient in need thereof.

[0027] In another aspect, the present invention provides a heterodimeric antibody comprising: a) a first monomer comprising: i) an anti-CD3 scFv comprising a first variable light chain domain, an scFv linker, and a first variable heavy chain domain; and ii) a first Fc domain, wherein the scFv is covalently attached to the N-terminus of the first Fc domain using a domain linker; b) a second monomer comprising a VH2-CH1-hinge-CH2-CH3 monomer, wherein VH is a second variable heavy chain domain and CH2-CH3 is a second Fc domain; and c) a light chain comprising a second variable light chain domain, wherein the second variable heavy chain domain and the second variable light chain domain form an ENPP3 binding domain.

[0028] In some embodiments, the ENPP3 binding domain comprises the vhCDR1-3 and vlCDR1-3 of any of the following ENPP3 binding domains: AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1(1)23. (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I ).

[0029] In some embodiments, the vhCDR1-3 and vlCDR1-3 of the ENPP3 binding domain are selected from Figure 12 、 13A to 13B ,and 14A to 14I The vhCDR1-3 sequences and vlCDR1-3 sequences of the ENPP3 binding domain are provided in .

[0030] In some embodiments, the second heavy chain variable domain comprises a heavy chain variable domain, and the second light chain variable domain comprises a variable light chain domain of any of the following ENPP3 binding domains: AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1(1)23. (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I ).

[0031] In some embodiments, the anti-CD3 scFv comprises the vhCDR1-3 and vlCDR1-3 of any of the following CD3 binding domains: H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10F ).

[0032] In some embodiments, the vhCDR1-3 and vlCDR1-3 of the anti-CD3 scFv are selected from 10A to 10F vhCDR1-3 and vlCDR1-3 in.

[0033] In some embodiments, the anti-CD3 scFv comprises the variable heavy chain domain and the variable light chain domain of any of the following CD3 binding domains: H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10F ).

[0034] In some embodiments, the first variable light domain is covalently attached to the N-terminus of the first Fc domain using a domain linker.

[0035] In some embodiments, the first variable heavy chain domain is covalently attached to the N-terminus of the first Fc domain using a domain linker.

[0036] In some embodiments, the scFv linker is a charged scFv linker.

[0037] In some embodiments, the first Fc domain and the second Fc domain are variant Fc domains.

[0038] In some embodiments, the first monomer and the second monomer include a monomer selected from Figures 1A to 1E In some embodiments, the set of heterodimerization variants is selected from the group consisting of: S364K / E357Q:L368D / K370S; S364K:L368D / K370S; S364K:L368E / K370S; D401K:T411E / K360E / Q362E; and T366W:T366S / L368A / Y407V, wherein the numbering is according to EU numbering.

[0039] In some embodiments, the first monomer and the second monomer further comprise an ablation variant. In some embodiments, the ablation variant is E233P / L234V / L235A / G236del / S267K, wherein the numbering is according to EU numbering.

[0040] In some embodiments, one of the first monomer or the second monomer comprises a pi variant.

[0041] In some embodiments, the pi variant is N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is according to EU numbering.

[0042] In some embodiments, the first monomer comprises the amino acid variant S364K / E357Q / E233P / L234V / L235A / G236del / S267K, wherein the second monomer comprises the amino acid variant L368D / K370S / N208D / Q295E / N384D / Q418E / N421D / E233P / L234V / L235A / G236del / S267K, and wherein numbering is according to EU numbering.

[0043] In some embodiments, the scFv linker is a charged scFv linker having the amino acid sequence (GKPGS)4.

[0044] In some embodiments, the first monomer and the second monomer each further comprise amino acid variant 428 / 434S.

[0045] In some embodiments, the heterodimeric antibody comprises the following heterodimeric antibodies: XENP24804, XENP26820, XENP28287, XENP28925, XENP29516, XENP30262, XENP26821, XENP29436, XENP28390, XENP29463, and XENP30263.

[0046] In another aspect, the present invention provides a heterodimeric antibody comprising: a) a first monomer comprising, from N-terminus to C-terminus, an scFv-linker-CH2-CH3, wherein the scFv is an anti-CD3 scFV and CH2-CH3 is a first Fc domain; b) a second monomer comprising, from N-terminus to C-terminus, VH-CH1-hinge-CH2-CH3, wherein CH2-CH3 is a second Fc domain; and c) a light chain comprising VL-CL; wherein the first variant Fc domain comprises amino acid variants S364K / E357Q, wherein the second variant Fc domain comprises amino acid variants L368D / K370S, wherein the first variant Fc domain and the second The variant Fc domains each comprise the amino acid variants E233P / L234V / L235A / G236del / S267K, wherein the hinge-CH2-CH3 of the second monomer comprises the amino acid variants N208D / Q295E / N384D / Q418E / N421D, wherein the VH and VL form an ENPP3 binding domain comprising a variable heavy chain domain and a variable light chain domain, respectively, of an ENPP3 binding domain selected from the group consisting of: AN1 [ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1 (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I), wherein the anti-CD3 scFv comprises a variable heavy chain domain and a variable light chain domain selected from the group consisting of: H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10F ), and the numbering is based on EU numbering.

[0047] In some embodiments, the scFv comprises a charged scFv linker having the amino acid sequence (GKPGS)4.

[0048] In some embodiments, the first variant Fc domain and the second variant Fc domain each further comprise amino acid variant 428 / 434S, wherein numbering is according to EU numbering.

[0049] In some embodiments, the present invention provides a nucleic acid composition comprising nucleic acids encoding a first monomer and a second monomer and a light chain of an antibody.

[0050] In some embodiments, the present invention provides an expression vector comprising a nucleic acid. In some embodiments, the present invention provides a host cell transformed with the expression vector.

[0051] In some embodiments, the present invention provides a method of treating an ENPP3-related cancer, comprising administering to a patient in need thereof any one of the antibodies provided herein.

[0052] In another aspect, the present invention provides a heterodimeric antibody comprising: a) a first monomer comprising, from N-terminus to C-terminus, VH1-CH1-Linker 1-scFv-Linker 2-CH2-CH3, wherein VH1 is a first variable heavy chain domain, scFv is an anti-CD3 scFV, Linker 1 and Linker 2 are a first domain linker and a second domain linker, respectively, and CH2-CH3 is a first Fc domain; b) a second monomer comprising, from N-terminus to C-terminus, VH2-CH1-hinge-CH2-CH3, wherein VH2 is a second variable heavy chain domain and CH2-CH3 is a second Fc domain; and c) a common light chain comprising a variable light chain domain; wherein the first variable heavy chain domain and the variable light chain domain form a first ENPP3 binding domain, and the second variable heavy chain domain and the variable light chain domain form a second ENPP3 binding domain.

[0053] In some embodiments, the first ENPP3 binding domain and the second ENPP3 binding domain each comprise the vhCDR1-3 and vlCDR1-3 of any of the following ENPP3 binding domains: AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1(1)23. (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha16-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I ).

[0054] In some embodiments, the vhCDR1-3 and vlCDR1-3 of the first ENPP3 binding domain and the second ENPP3 binding domain are selected from Figures 14 and Figure 45 vhCDR1-3 and vlCDR1-3 provided in.

[0055] In some embodiments, the first variable heavy chain domain and the second variable heavy chain domain each comprise a variable heavy chain domain of an ENPP3 binding domain, and the first variable light chain domain and the second variable light chain domain each comprise a variable light chain domain of an ENPP3 binding domain, wherein the ENPP3 binding domain is any of the following ENPP3 binding domains: AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, H16-9.44, H16-1.67, Ha16-1(2). (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I ).

[0056] In some embodiments, the scFv comprises the vhCDR1-3 and vlCDR1-3 of any of the following CD3 binding domains: H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10F ).

[0057] In some embodiments, the vhCDR1-3 and vlCDR1-3 of the scFv are selected from 10A to 10F vhCDR1-3 and vlCDR1-3 in.

[0058] In some embodiments, the scFv comprises the variable heavy chain domain and the variable light chain domain of any of the following CD3 binding domains: H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10F ).

[0059] In some embodiments, the scFv comprises a scFv variable heavy chain domain, a scFv variable light chain domain, and a scFv linker connecting the scFv variable heavy chain domain and the scFv variable light chain domain.

[0060] In some embodiments, the scFv variable heavy chain domain is attached to the C-terminus of the CH1 of the first monomer using the first domain linker, and the scFv variable light chain domain is covalently attached to the N-terminus of the first Fc domain using the second domain linker.

[0061] In some embodiments, the scFv variable light chain domain is attached to the C-terminus of the CH1 of the first monomer using the first domain linker, and the scFv variable heavy chain domain is covalently attached to the N-terminus of the first Fc domain using the second domain linker.

[0062] In some embodiments, the scFv linker is a charged scFv linker.

[0063] In some embodiments, the first Fc domain and the second Fc domain are variant Fc domains.

[0064] In some embodiments, the first monomer and the second monomer comprise a monomer selected from Figures 1A to 1E A set of heterodimerization variants of the heterodimerization variants depicted in .

[0065] In some embodiments, the set of heterodimerization variants is selected from the group consisting of: S364K / E357Q:L368D / K370S; S364K:L368D / K370S; S364K:L368E / K370S; D401K:T411E / K360E / Q362E; and T366W:T366S / L368A / Y407V, wherein the numbering is according to EU numbering.

[0066] In some embodiments, the first monomer and the second monomer further comprise an ablation variant.

[0067] In some embodiments, the ablation variant is E233P / L234V / L235A / G236del / S267K, wherein the numbering is according to EU numbering.

[0068] In some embodiments, one of the first monomer or the second monomer further comprises a pi variant.

[0069] In some embodiments, the pi variant is N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is according to EU numbering.

[0070] In some embodiments, the first variant Fc domain comprises the amino acid variants S364K / E357Q / E233P / L234V / L235A / G236del / S267K, wherein the second variant Fc domain comprises the amino acid variants L368D / K370S / N208D / Q295E / N384D / Q418E / N421D / E233P / L234V / L235A / G236del / S267K, wherein numbering is according to EU numbering.

[0071] In some embodiments, the scFv linker is a charged scFv linker having the amino acid sequence (GKPGS)4.

[0072] In some embodiments, the first variant Fc domain and the second variant Fc domain each further comprise amino acid variant 428 / 434S, wherein numbering is according to EU numbering.

[0073] In some embodiments, the heterodimeric antibodies include the following heterodimeric antibodies: XENP29437, XENP29520, XENP30264, XENP26822, XENP28438, XENP29438, XENP29467, XENP30469, XENP30470, XENP30819, XENP30821, XENP31148, XENP31149, XENP31150, XENP31419, and XENP31471.

[0074] In another aspect, the heterodimeric antibody comprises: a) a first monomer comprising, from N-terminus to C-terminus, VH1-CH1-linker 1-scFv-linker 2-CH2-CH3, wherein the scFv is an anti-CD3 scFV and CH2-CH3 is a first Fc domain; b) a second monomer comprising, from N-terminus to C-terminus, VH1-CH1-hinge-CH2-CH3, wherein CH2-CH3 is a second Fc domain; and c) a common light chain comprising VL-CL; wherein the first variant Fc domain comprises amino acid variants S364K / E357Q, wherein the second variant Fc domain comprises amino acid variants L368D / K370S, wherein wherein the first variant Fc domain and the second variant Fc domain each comprise the amino acid variants E233P / L234V / L235A / G236del / S267K, wherein the hinge-CH2-CH3 of the second monomer comprises the amino acid variants N208D / Q295E / N384D / Q418E / N421D, wherein the VH and VL comprise a variable heavy chain domain and a variable light chain domain selected from the group consisting of the following ENPP3 binding domains: AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1 (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I), wherein the anti-CD3 scFv comprises a variable heavy chain domain and a variable light chain domain selected from the group consisting of: H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10F ), and the numbering is based on EU numbering.

[0075] In some embodiments, the scFv comprises a charged scFv linker having the amino acid sequence (GKPGS)4.

[0076] In some embodiments, the first variant Fc domain and the second variant Fc domain each further comprise amino acid variant 428 / 434S.

[0077] In some embodiments, the first monomer and the second monomer are a common light chain of an antibody. In some embodiments, the present invention provides an expression vector comprising a nucleic acid. In some embodiments, the present invention provides a host cell transformed with the expression vector. In some embodiments, the present invention provides a method for treating an ENPP3-related cancer, comprising administering an antibody to a patient in need thereof.

[0078] In another aspect, the present invention provides a heterodimeric antibody comprising the following heterodimeric antibodies: XENP24804, XENP26820, XENP28287, XENP28925, XENP29516, XENP30262, XENP26821, XENP29436, XENP28390, XENP29463, and XENP30263.

[0079] In another aspect, the present invention provides a heterodimeric antibody, comprising the following heterodimeric antibodies: XENP29437, XENP29520, XENP30264, XENP26822, XENP28438, XENP29438, XENP29467, XENP30469, XENP30470, XENP30819, XENP30821, XENP31148, XENP31149, XENP31150, XENP31419, and XENP31471. In some embodiments, the present invention provides a nucleic acid composition comprising a nucleic acid encoding the heterodimeric antibody. In some embodiments, the present invention provides an expression vector comprising a nucleic acid. In some embodiments, the present invention provides a host cell transformed with an expression vector.

[0080] In some embodiments, the present invention provides a method of treating an ENPP3-related cancer, comprising administering to a patient in need thereof any one of the heterodimeric antibodies provided herein. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figures 1A to 1E Useful pairs of Fc heterodimerization variants are depicted (including skew variants and pi variants). Some variants do not have a corresponding "monomer 2" variant; these are pi variants that can be used alone on either monomer.

[0082] Figure 2 Depicts a list of homozygous variant antibody constant regions and their corresponding substitutions. pi_(-) represents a lower pi variant, while pi_(+) represents a higher pi variant. These can optionally and independently be combined with other heterodimerization variants of the antibodies described herein (as well as other variant types, as outlined herein).

[0083] Figure 3 Useful ablation variants are depicted that ablate FcγR binding (sometimes referred to as “knockout” or “KO” variants). Typically, the ablation variant is found on both monomers, although in some cases it may be on only one monomer.

[0084] Figure 4 Particularly useful embodiments of the "non-Fv" components of the antibodies described herein are depicted.

[0085] Figure 5 Depicted are a plurality of charged scFv linkers as described herein for increasing or decreasing the pI of heterodimeric bsAbs of the invention utilizing one or more scFvs as components. (+H) positive linkers are particularly useful herein, especially with the anti-CD3 V described herein. L and VH sequences. According to Whitlow et al., Protein Engineering 6(8):989-995 (1993), a single prior art scFv linker with a single charge is referred to as a "Whitlow." It should be noted that such a linker is used to reduce aggregation and enhance proteolytic stability in scFv. Such a charged scFv linker can be used in any of the antibody formats disclosed herein, comprising scFv (e.g., a 1+1 Fab-scFv-Fc format and a 2+1 Fab2-scFv-Fc format).

[0086] Figure 6 Many exemplary domain linkers are depicted. In some embodiments, these linkers are used to connect single-chain Fv to Fc chains. In some embodiments, these linkers can be combined. For example, a GGGGS linker can be combined with a "half-hinge" linker.

[0087] 7A to 7DDepicted are the sequences of several useful heavy chain backbones of 1+1 Fab-scFv-Fc bispecific antibody formats based on human IgG1, without Fv sequences (e.g., scFv and VH on the Fab side). Backbone 1 is based on human IgG1 (356E / 358M allotype) and includes an S364K / E357Q:L368D / K370S skew variant, a C220S on one chain with an S364K / E357Q skew variant, an N208D / Q295E / N384D / Q418E / N421D pI variant on one chain with an L368D / K370S skew variant, and an E233P / L234V / L235A / G236del / S267K ablation variant on both chains. Backbone 2 is based on human IgG1 (356E / 358M allotype) and includes an S364K:L368D / K370S skew variant, a C220S on one chain with an S364K skew variant, an N208D / Q295E / N384D / Q418E / N421D pI variant on one chain with an L368D / K370S skew variant, and an E233P / L234V / L235A / G236del / S267K ablation variant on both chains. Backbone 3 is based on human IgG1 (356E / 358M allotype) and includes an S364K:L368E / K370S skew variant, a C220S on one chain with an S364K skew variant, an N208D / Q295E / N384D / Q418E / N421D pI variant on one chain with an L368D / K370S skew variant, and an E233P / L234V / L235A / G236del / S267K ablation variant on both chains. Backbone 4 is based on human IgG1 (356E / 358M allotype) and includes a D401K:K360E / Q362E / T411E skew variant, C220S on one chain with the D401K skew variant, an N208D / Q295E / N384D / Q418E / N421D pI variant on one chain with the K360E / Q362E / T411E skew variant, and an E233P / L234V / L235A / G236del / S267K ablation variant on both chains. Backbone 5 is based on human IgG1 (356E / 358M allotype) and includes a S364K / E357Q:L368D / K370S skew variant, a C220S on one chain with the S364K / E357Q skew variant, a N208D / Q295E / N384D / Q418E / N421D pI variant on one chain with the L368D / K370S skew variant, and an E233P / L234V / L235A / G236del / S267K ablation variant on both chains.Backbone 6 is based on human IgG1 (356E / 358M allotype) and includes an S364K / E357Q:L368D / K370S skew variant, a C220S on one chain with the S364K / E357Q skew variant, an N208D / Q295E / N384D / Q418E / N421D pI variant on one chain with the L368D / K370S skew variant, and an E233P / L234V / L235A / G236del / S267K ablation variant on both chains, as well as an N297A variant on both chains. Backbone 7 is identical to Backbone 6 except that the mutation is N297S. Backbone 8 is based on human IgG4 and includes a S364K / E357Q:L368D / K370S skew variant, an N208D / Q295E / N384D / Q418E / N421D pI variant on the chain with the L368D / K370S skew variant, and an S228P (EU numbering, this is S241P in Kabat) variant on both chains that ablate Fab arm exchange as known in the art. Backbone 9 is based on human IgG2 and includes a S364K / E357Q:L368D / K370S skew variant, an N208D / Q295E / N384D / Q418E / N421D pI variant on the chain with the L368D / K370S skew variant. Backbone 10 is based on human IgG2 and includes an S364K / E357Q:L368D / K370S skew variant, an N208D / Q295E / N384D / Q418E / N421D pI variant on one chain with the L368D / K370S skew variant, and an S267K variant on both chains. Backbone 11 is identical to Backbone 1 except that it includes the M428L / N434S Xtend mutation. Backbone 12 is based on human IgG1 (356E / 358M allotype) and comprises an S364K / E357Q:L368D / K370S skew variant, a C220S and P217R / P229R / N276K pI variant on one chain with the S364K / E357Q skew variant, and an E233P / L234V / L235A / G236del / S267K ablation variant on both chains. Included within each of these backbones are sequences that are 90%, 95%, 98%, and 99% identical (as defined herein) to the recited sequences and / or contain 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 additional amino acid substitutions (compared to the "parent" of the figure, which, as one skilled in the art will appreciate, already contain multiple amino acid modifications compared to the parent human IgG1 (or IgG2 or IgG4, depending on the backbone)). That is, in addition to the skew variants, pi variants, and ablation variants contained in the backbone of this figure, the enumerated backbone may contain additional amino acid modifications (typically amino acid substitutions).

[0088] Figures 8A to 8CDepicted are the sequences of several useful heavy chain backbones of 2+1 Fab2-scFv-Fc bispecific antibody formats based on human IgG1, without Fv sequences (e.g., scFv and VH on the Fab side). Backbone 1 is based on human IgG1 (356E / 358M allotype) and contains a S364K / E357Q:L368D / K370S skew variant, a N208D / Q295E / N384D / Q418E / N421D pI variant on one chain with the L368D / K370S skew variant, and an E233P / L234V / L235A / G236del / S267K ablation variant on both chains. Backbone 2 is based on human IgG1 (356E / 358M allotype) and contains an S364K:L368D / K370S skew variant, an N208D / Q295E / N384D / Q418E / N421D pI variant on one chain with the L368D / K370S skew variant, and an E233P / L234V / L235A / G236del / S267K ablation variant on both chains. Backbone 3 is based on human IgG1 (356E / 358M allotype) and contains an S364K:L368E / K370S skew variant, an N208D / Q295E / N384D / Q418E / N421D pI variant on one chain with the L368E / K370S skew variant, and an E233P / L234V / L235A / G236del / S267K ablation variant on both chains. Backbone 4 is based on human IgG1 (356E / 358M allotype) and contains a D401K:K360E / Q362E / T411E skew variant, an N208D / Q295E / N384D / Q418E / N421D pI variant on one chain with the K360E / Q362E / T411E skew variant, and an E233P / L234V / L235A / G236del / S267K ablation variant on both chains. Backbone 5 is based on human IgG1 (356D / 358L allotype) and contains a S364K / E357Q:L368D / K370S skew variant, a N208D / Q295E / N384D / Q418E / N421D pI variant on the chain with the L368D / K370S skew variant, and an E233P / L234V / L235A / G236del / S267K ablation variant on both chains.Backbone 6 is based on human IgG1 (356E / 358M allotype) and includes an S364K / E357Q:L368D / K370S skew variant, an N208D / Q295E / N384D / Q418E / N421D pI variant on one chain with the L368D / K370S skew variant, and an E233P / L234V / L235A / G236del / S267K ablation variant on both chains, as well as an N297A variant on both chains. Backbone 7 is identical to Backbone 6 except that the mutation is N297S. Backbone 8 is identical to Backbone 1 except that it includes the M428L / N434S Xtend mutation. Backbone 9 is based on human IgG1 (356E / 358M allotype) and comprises a S364K / E357Q:L368D / K370S skew variant, a P217R / P229R / N276K pI variant on one chain with the S364K / E357Q skew variant, and an E233P / L234V / L235A / G236del / S267K ablation variant on both chains. Included within each of these backbones are sequences that are 90%, 95%, 98%, and 99% identical (as defined herein) to the recited sequences and / or contain 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 additional amino acid substitutions (compared to the "parent" of the figure, which, as one skilled in the art will appreciate, already contain multiple amino acid modifications compared to the parent human IgG1 (or IgG2 or IgG4, depending on the backbone)). That is, in addition to the skew variants, pi variants, and ablation variants contained in the backbone of this figure, the enumerated backbone may contain additional amino acid modifications (typically amino acid substitutions).

[0089] Figure 9 Depicted are sequences of several useful constant light chain domain backbones based on human IgG1, without Fv sequences (e.g., scFv or Fab). The constant light backbone sequences encompassed herein are 90%, 95%, 98%, and 99% identical (as defined herein) to the recited sequences, and / or contain 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 additional amino acid modifications.

[0090] 10A to 10F The sequences of exemplary anti-CD3 scFvs suitable for use in the bispecific antibodies described herein are depicted. The CDRs are underlined and the scFv linker is double-underlined (in the sequence, the scFv linker is the positively charged scFv(GKPGS)4 linker (SEQ ID NO: XXX), but as will be appreciated by those skilled in the art, this linker can be replaced by other linkers, including uncharged linkers or negatively charged linkers, some of which are depicted in FIG. Figure 5), and slashes indicate one or more boundaries of the variable domain. In addition, the naming convention shows the orientation of the scFv from the N-terminus to the C-terminus. As indicated herein and for each sequence containing CDRs herein, as shown in Table 2, the precise designation of the CDR positions may vary slightly depending on the numbering used, and therefore not only the underlined CDRs but also CDRs included within the VH and VL domains using other numbering systems are included herein. In addition, for all sequences in the figures, these V H and V L The sequences can be used in scFv format or Fab format.

[0091] Figures 11A to 11B Antigen sequences for a variety of antigens used in the antibodies described herein, both human and cynomolgus monkey, are depicted to aid in the development of antigen binding domains that bind to both for ease of clinical development.

[0092] Figure 12 Depicted are the variable heavy and light chain sequences of an exemplary humanized ENPP3 binding domain, referred to herein as AN1, and the sequence of XENP28278, an anti-ENPP3 mAb based on AN1 with the E233P / L234V / L235A / G236del / S267K ablation variant and an IgG1 backbone. The CDRs are underlined, and slashes indicate the boundaries between the variable and constant domains. As noted herein, and indeed for each sequence containing CDRs herein, as shown in Table 2, the precise designation of the CDR positions may vary slightly depending on the numbering used, and therefore not only the underlined CDRs but also CDRs included within VH and VL domains using other numbering systems are included herein. In addition, for all sequences in the figures, these V H and V L The sequences can be used in scFv format or Fab format.

[0093] 13A to 13B Depicted are the variable heavy and light chain sequences of AN1 variants engineered to achieve improved purification and / or modulation of ENPP3 binding affinity and / or potency. The CDRs are underlined, and slashes indicate the boundary between the variable region and the constant domain. As mentioned herein and true for each sequence containing the CDRs, depending on the numbering scheme used (e.g., Figure 12 ), the exact designation of the CDR positions may vary slightly, and therefore not only the underlined CDRs are included herein but also CDRs included within VH and VL domains using other numbering systems. Further, as for all sequences in the accompanying figures, these V H and V LThe sequences can be used in scFv or Fab formats. In addition, each of the variable heavy chain domains depicted herein can be paired with any other αENPP3 variable light chain domain; and each of the variable light chain domains depicted herein can be paired with any other αENPP3 variable heavy chain domain.

[0094] 14A to 14I The variable regions of additional ENPP3 antigen binding domains that can be used for αENPP3xαCD3 antibodies are depicted. The CDRs are underlined. As mentioned herein and true for each sequence containing CDRs, depending on the numbering scheme used (e.g. Figure 12 ), the exact designation of the CDR positions may vary slightly, and therefore not only the underlined CDRs but also CDRs included within VH and VL domains using other numbering systems are included herein. In addition, for all sequences in the figures, these V H and V L The sequences can be used in scFv format or Fab format.

[0095] FIG. 15A to FIG. 15B Several formats of the antibodies described herein are depicted. Figure 15A A "1+1 Fab-scFv-Fc" format is depicted, wherein the first arm comprises an ENPP3-binding Fab and the second arm comprises a CD3-binding scFv. Figure 30 B depicts a "2+1 Fab2-scFv-Fc" format, wherein the first arm comprises an ENPP3-binding Fab, and the second arm comprises a Fab and a scFv, wherein the Fab binds ENPP3 and the scFv binds CD3.

[0096] Figure 16 The amino acid sequence of a control anti-RSV x anti-CD3 bispecific antibody (Fab-scFv-Fc) is depicted in a corkscrew format. The antibody is named using the Fab variable region first and the scFv variable region second, separated by a dotted line. The CDRs are underlined and slashes indicate the boundaries of the variable region. The scFv domain has a V H -scFv linker-V L The orientation of the sequences is (N-terminal to C-terminal), but this can be reversed. In addition, each of the sequences outlined herein may or may not include the M428L / N434S variant in one or preferably both Fc domains, which results in a longer serum half-life.

[0097] 17A to 17CDepicted is the sequence of an illustrative αENPP3 x αCD3 bsAb in a 1+1 Fab-scFv-Fc format comprising the H1.30_L1.47 anti-CD3 scFv (also known as CD3 high [VHVL]). The CDRs are underlined, and slashes indicate the boundaries between the variable region and other chain components (e.g., constant region and domain linker). It should be noted that the αENPP3 x αCD3 bsAb can utilize variable region, Fc region, and constant domain sequences that are 90%, 95%, 98%, and 99% identical (as defined herein) and / or comprise 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid substitutions. In addition, each of the sequences outlined herein may or may not comprise the M428L / N434S variant in one or, preferably, both Fc domains, which results in a longer serum half-life.

[0098] 18A to 18C Depicted is the sequence of an illustrative αENPP3 x αCD3 bsAb in a 1+1 Fab-scFv-Fc format comprising the H1.32_L1.47 anti-CD3 scFv (also known as CD3 high-Int number 1 [VHVL]). The CDRs are underlined, and slashes indicate the boundaries between the variable region and other chain components (e.g., constant region and domain linker). It should be noted that the αENPP3 x αCD3 bsAb can utilize variable region, Fc region, and constant domain sequences that are 90%, 95%, 98%, and 99% identical (as defined herein) and / or comprise 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid substitutions. In addition, each of the sequences outlined herein may or may not comprise the M428L / N434S variant in one or, preferably, both Fc domains, which results in a longer serum half-life.

[0099] 19A to 19C Depicted is the sequence of an illustrative αENPP3 x αCD3 bsAb in a 2+1 Fab2-scFv-Fc format comprising the H1.30_L1.47 anti-CD3 scFv (also known as CD3 high [VHVL]). The CDRs are underlined, and slashes indicate the boundaries between the variable region and other chain components (e.g., constant region and domain linker). It should be noted that the αENPP3 x αCD3 bsAb can utilize variable region, Fc region, and constant domain sequences that are 90%, 95%, 98%, and 99% identical (as defined herein) and / or comprise 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid substitutions. In addition, each of the sequences outlined herein may or may not comprise the M428L / N434S variant in one or, preferably, both Fc domains, which results in a longer serum half-life.

[0100] 20A to 20D Depicted is the sequence of an illustrative αENPP3 x αCD3 bsAb in a 2+1 Fab2-scFv-Fc format comprising the H1.32_L1.47 anti-CD3 scFv (also known as CD3 high-Int number 1 [VHVL]). The CDRs are underlined, and slashes indicate the boundaries between the variable region and other chain components (e.g., constant region and domain linker). It should be noted that the αENPP3 x αCD3 bsAb can utilize variable region, Fc region, and constant domain sequences that are 90%, 95%, 98%, and 99% identical (as defined herein) and / or comprise 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid substitutions. In addition, each of the sequences outlined herein may or may not comprise the M428L / N434S variant in one or, preferably, both Fc domains, which results in a longer serum half-life.

[0101] Figure 21 Depicted is the sequence of an illustrative αENPP3 x αCD3 bsAb in a 2+1 Fab2-scFv-Fc format comprising the L1.47_H1.30 anti-CD3 scFv (also known as CD3 high [VLVH]). The CDRs are underlined, and slashes indicate the boundaries between the variable region and other chain components (e.g., constant region and domain linker). It should be noted that the αENPP3 x αCD3 bsAb can utilize variable region, Fc region, and constant domain sequences that are 90%, 95%, 98%, and 99% identical (as defined herein) and / or comprise 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid substitutions. In addition, each of the sequences outlined herein may or may not comprise the M428L / N434S variant in one or, preferably, both Fc domains, which results in a longer serum half-life.

[0102] Figures 22A to 22CDepicted is the sequence of an illustrative αENPP3 x αCD3 bsAb in a 2+1 Fab2-scFv-Fc format comprising the L1.47_H1.32 anti-CD3 scFv (also known as CD3 high-Int number 1 [VHVL]). The CDRs are underlined, and slashes indicate the boundaries between the variable region and other chain components (e.g., constant region and domain linker). It should be noted that the αENPP3 x αCD3 bsAb can utilize variable region, Fc region, and constant domain sequences that are 90%, 95%, 98%, and 99% identical (as defined herein) and / or comprise 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid substitutions. In addition, each of the sequences outlined herein may or may not comprise the M428L / N434S variant in one or, preferably, both Fc domains, which results in a longer serum half-life.

[0103] Figures 23A to 23E Depicted is the sequence of an illustrative αENPP3 x αCD3 bsAb in a 2+1 Fab2-scFv-Fc format comprising the L1.47_H1.89 anti-CD3 scFv (also known as CD3 high-Int number 2 [VLVH]). The CDRs are underlined, and slashes indicate the boundaries between the variable region and other chain components (e.g., constant region and domain linker). It should be noted that the αENPP3 x αCD3 bsAb can utilize variable region, Fc region, and constant domain sequences that are 90%, 95%, 98%, and 99% identical (as defined herein) and / or comprise 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid substitutions. In addition, each of the sequences outlined herein may or may not comprise the M428L / N434S variant in one or, preferably, both Fc domains, which results in a longer serum half-life.

[0104] FIG. 24A to FIG. 24B Depicts the induction of RTCC in CFSE-labeled KU812 cells. A) As indicated by CFSE + A) As indicated by the reduction in KU812 cell numbers and B) CFSE staining with Zombie Aqua after 24 h of incubation of CFSE-labeled KU812 with human PBMCs (effector to target ratio of 10:1) and αENPP3 x αCD3 bispecific antibodies (XENP26820, XENP26821, XENP28287, and XENP28390). +The percentage of KU812 cells was indicated. Controls used were an αRSV x αCD3 bispecific antibody (XENP13245), effector and target cells alone, and target cells alone. Overall, the data showed that the prototype αENPP3 x αCD3 bsAb dose-dependently induced redirected T cell cytotoxicity (RTCC) on KU812 cells; CD3 binding affinity correlated with RTCC potency (i.e., bsAbs with CD3 high induced RTCC more potently than bsAbs with CD3 high-Int No. 1); and bsAbs with the AN1 binding domain induced RTCC more potently than bsAbs with the H16-7.8 binding domain.

[0105] Figures 25A to 25C Depicts activated CD4 + T cells, as indicated by: A) CD4 + MFI of CD107a on T cells, B) CD4 + CD25 MFI on T cells, and C) CD4 after 24 h of incubation of CFSE-labeled KU812 with human PBMCs (effector to target ratio of 10:1) and αENPP3 x αCD3 bispecific antibodies (XENP26820, XENP26821, XENP28287, and XENP28390). + CD69 MFI on T cells. Controls used were αRSV x αCD3 bispecific antibody (XENP13245), effector and target cells only, and target cells only. Consistent with RTCC data, αENPP3 x αCD3 bsAb dose-dependently induced CD4 + T cell activation; CD3 binding affinity correlates with activation potency (i.e., bsAbs with CD3 high induce CD4 more efficiently than bsAbs with CD3 high-Int No. 1). + T cell activation); and bsAbs with AN1 binding domains induced CD4 T cell activation more effectively than bsAbs with H16-7.8 binding domains. + T cell activation.

[0106] Figures 26A to 26C Depicts activated CD8 + T cells, as indicated by: A) CD8 + MFI of CD107a on T cells, B) CD8 +CD25 MFI on T cells, and C) CD8 after 24 h of incubation of CFSE-labeled KU812 with human PBMCs (effector to target ratio of 10:1) and αENPP3 x αCD3 bispecific antibodies (XENP26820, XENP26821, XENP28287, and XENP28390). + CD69 MFI on T cells. Controls used were αRSV x αCD3 bispecific antibody (XENP13245), effector and target cells only, and target cells only. Consistent with RTCC data, αENPP3 x αCD3 bsAb dose-dependently induced CD8 + T cell activation; CD3 binding affinity correlates with activation potency (i.e., bsAbs with CD3 high induce CD8 more efficiently than bsAbs with CD3 high-Int No. 1). + T cell activation); and bsAbs with AN1 binding domains induced CD8 T cell activation more effectively than bsAbs with H16-7.8 binding domains. + T cell activation.

[0107] FIG. 27A to FIG. 27B Depicts RTCC induction in CFSE-labeled RXF393 cells. A) As indicated by CFSE + As indicated by the decrease in the number of RXF393 cells and B) CFSE staining with Zombie Aqua after 24 h of incubation of CFSE-labeled RXF393 with human PBMCs (effector to target ratio of 20:1) and αENPP3 x αCD3 bispecific antibodies (XENP26820, XENP26821, XENP28287, and XENP28390). + The percentage of RXF393 cells was indicated. Controls used were an αRSV x αCD3 bispecific antibody (XENP13245), effector and target cells alone, and target cells alone. Consistent with the data on KU812 cells, the data showed that the prototype αENPP3 x αCD3 bsAb dose-dependently induced redirected T cell cytotoxicity (RTCC) on RXF393 cells; CD3 binding affinity correlated with RTCC potency (i.e., bsAbs with CD3 high induced RTCC more potently than bsAbs with CD3 high-Int No. 1); and bsAbs with the AN1 binding domain induced RTCC more potently than bsAbs with the H16-7.8 binding domain.

[0108] Figures 28A to 28C Depicts activated CD4 + T cells, as indicated by: A) CD4+ CD107a MFI on T cells, B) CD4 + CD25 MFI on T cells, and C) CD4 after 24 h of incubation of CFSE-labeled RXF393 with human PBMCs (effector to target cell ratio of 20:1) and αENPP3 x αCD3 bispecific antibodies (XENP26820, XENP26821, XENP28287, and XENP28390). + CD69 MFI on T cells. Controls used were αRSV x αCD3 bispecific antibody (XENP13245), effector and target cells only, and target cells only. Consistent with RTCC data, αENPP3 x αCD3 bsAb dose-dependently induced CD4 + T cell activation; CD3 binding affinity correlates with activation potency (i.e., bsAbs with CD3 high induce CD4 more efficiently than bsAbs with CD3 high-Int No. 1). + T cell activation); and bsAbs with AN1 binding domains induced CD4 T cell activation more effectively than bsAbs with H16-7.8 binding domains. + T cell activation.

[0109] Figures 29A to 29C Depicts activated CD8 + T cells, as indicated by: A) CD8 + MFI of CD107a on T cells, B) CD8 + CD25 MFI on T cells, and C) CD8 after 24 h of incubation of CFSE-labeled RXF393 with human PBMCs (effector to target ratio 20:1) and αENPP3 x αCD3 bispecific antibodies (XENP26820, XENP26821, XENP28287, and XENP28390). + CD69 MFI on T cells. Controls used were αRSV x αCD3 bispecific antibody (XENP13245), effector and target cells only, and target cells only. Consistent with RTCC data, αENPP3 x αCD3 bsAb dose-dependently induced CD8 + T cell activation; CD3 binding affinity correlates with activation potency (i.e., bsAbs with CD3 high induce CD8 more efficiently than bsAbs with CD3 high-Int No. 1). + T cell activation); and bsAbs with AN1 binding domains induced CD8 T cell activation more effectively than bsAbs with H16-7.8 binding domains. + T cell activation.

[0110] Figure 30 Depicted are A) chromatograms showing purified fraction 2 of XENP28287 (cation exchange chromatography followed by protein A chromatography), and Figure 30 Purity and homogeneity of peaks B and BC (as well as pre-purified material) isolated from the cation exchange separation depicted in A, as determined by B) analytical size exclusion chromatography with multi-angle light scattering (aSEC-MALS) and C) analytical cation exchange chromatography (aCIEX). Figure 30 B also depicts the molecular weight of the protein species in the peak as determined by multi-angle light scattering.

[0111] Figure 31 Depicted are A) chromatograms showing purified fraction 2 of XENP28925 (cation exchange chromatography followed by protein A chromatography), and Figure 31 Purity and homogeneity of Peak B (as well as pre-purified material) isolated from the cation exchange separation depicted in A, as determined by B) analytical size exclusion chromatography with multi-angle light scattering (aSEC-MALS) and C) analytical cation exchange chromatography (aCIEX). Figure 31 B also depicts the molecular weight of the protein species in the peak as determined by multi-angle light scattering.

[0112] Figure 32 Depicted are A) a chromatogram demonstrating purified fraction 2 of XENP31149 (cation exchange chromatography following Protein A chromatography), and B) the identity of peaks A and B (and pre-purified material) isolated from the cation exchange separation as depicted in Figure XA, as determined by analytical size exclusion chromatography with multi-angle light scattering (aSEC-MALS).

[0113] Figure 33 Depicted are A) a chromatogram demonstrating purified fraction 2 of XENP31419 (cation exchange chromatography following Protein A chromatography), and B) the identity of peaks A and B (and pre-purified material) isolated from the cation exchange separation as depicted in Figure XA, as determined by analytical size exclusion chromatography with multi-angle light scattering (aSEC-MALS).

[0114] Figure 34 Depicts the expression of CFSE-labeled KU812 (solid line, ENPP3 高 ) or CFSE-labeled RCC4 (dashed line, ENPP3 低RTCC induction on CFSE-labeled target cells was performed by incubating CFSE-labeled target cells with human PBMCs (at a 10:1 ratio of effector cells to target cells) and the αENPP3 x αCD3 bispecific antibody XENP28925 for 18 hours and staining the CFSE with Zombie Aqua. + The percentages of cells are indicated.

[0115] Figure 35 Depicted affinity-engineered αENPP3 x αCD3 1+1 bsAb with ENPP3 高 Binding of KU812 cells.

[0116] Figure 36 Depicts the expression of CFSE-labeled KU812 (solid line, ENPP3 高 ) or CFSE-labeled RCC4 (dashed line, ENPP3 低 ) cells, as determined by Zombie Aqua staining of CFSE after 42 h of incubation of CFSE-labeled target cells with human PBMCs (effector to target ratio of 10:1) and the αENPP3 x αCD3 bispecific antibodies XENP28925 (WT high ENPP3 binding), XENP29516 (moderate ENPP3 binding), or XENP30262 (low ENPP3 binding). + The data showed that both XENP29516 and XENP30262 showed inhibition of ENPP3 compared to XENP28925. 低 Substantially less potent induction of RTCC on RCC4 cells, where RTCC potency correlated with binding potency, as shown above. XENP29516 and XENP30262 also showed inhibition of ENPP3 高 Less efficient induction of RTCC on cells.

[0117] Figures 37A to 37C Depicted are the induction of A) IFNγ, B) IL-6, and C) TNFα release in human PBMCs incubated for 18 hours with KU812 cells (10:1 effector to target ratio) and the αENPP3 x αCD3 bispecific antibodies XENP28925 (CD3 high) or XENP29436 (CD3 high - Int No. 1). Data demonstrate that XENP29436 exhibits substantially less potent induction of cytokine release than XENP28925.

[0118] Figure 38Depicted is the induction of IFNγ release from human PBMCs incubated for 18 hours with RCC4 cells (effector to target ratio 10:1) and the αENPP3 x αCD3 bispecific antibodies XENP28925 (CD3 high) or XENP29436 (CD3 high - Int No. 1). 低 In the case of RCC4 cells, XENP29436 showed negligible induction of cytokine release compared to XENP28925.

[0119] Figure 39 Depicts the expression of CFSE-labeled KU812 (solid line, ENPP3 高 ) or CFSE-labeled RCC4 (dashed line, RCC4 低 ) cells, as determined by CFSE staining with Zombie Aqua after 42 h of incubation of CFSE-labeled target cells with human PBMCs (effector to target ratio 10:1) and the αENPP3 x αCD3 bispecific antibodies XENP28925 (CD3 high) or XENP29436 (CD3 high - Int No. 1). + The data indicate that XENP29436 exhibits substantially less efficacy against ENPP3 compared to XENP28925. 低 However, XENP29436 also showed reduced induction of ENPP3 高 Efficacy of RTCC on cells.

[0120] Figure 40 Depicted is the induction of IFNγ release in human PBMCs incubated with KU812 cells (10:1 effector to target ratio) and the αENPP3 x αCD3 bispecific antibodies XENP28925 (ENPP3 High; CD3 High), XENP29436 (ENPP3 High; CD3 High - Int No. 1), XENP29518 (ENPP3 Intermediate; CD3 High), XENP29463 (ENPP3 Intermediate; CD3 High - Int No. 1), XENP30262 (ENPP3 Low; CD3 High), or XENP30263 (ENPP3 Low; CD3 High - Int No. 1). The data show that reducing CD3 or ENPP3 binding potency reduces the induction of cytokine release. Notably, reducing both CD3 and ENPP3 binding potency further reduced the induction of cytokine release.

[0121] Figure 41 Depicts the expression of CFSE-labeled KU812 (solid line, ENPP3 高) or CFSE-labeled RCC4 (dashed line, ENPP3 低 ) cells, as determined by CFSE staining with Zombie Aqua after 42 h of incubation of CFSE-labeled target cells with human PBMCs (effector to target ratio 10:1) and the αENPP3 x αCD3 bispecific antibodies XENP28925 (WT high ENPP3 binding; CD3 high; monovalent ENPP3 binding), XENP29516 (moderate ENPP3 binding; CD3 high; monovalent ENPP3 binding), or XENP29520 (moderate ENPP3 binding; CD3 high; bivalent ENPP3 binding). + The data show that bivalent binding (with moderate ENPP3 binding) is the most common in ENPP3 低 Reduced RTCC efficacy was maintained on cells but not on ENPP3 高 The cells recovered RTCC efficacy close to that exhibited by XENP28925.

[0122] Figure 42 Depicts the expression of CFSE-labeled KU812 (solid line, ENPP3 高 ) or CFSE-labeled RCC4 (dashed line, ENPP3 低 ) cells, as determined by CFSE staining with Zombie Aqua after 42 h of incubation of CFSE-labeled target cells with human PBMCs (effector to target ratio 10:1) and the αENPP3 x αCD3 bispecific antibodies XENP28925 (WT high ENPP3 binding; CD3 high; monovalent ENPP3 binding), XENP30262 (low ENPP3 binding; CD3 high; monovalent ENPP3 binding), or XENP30264 (low ENPP3 binding; CD3 high; bivalent ENPP3 binding). + The data show that bivalent binding (with low ENPP3 binding) further reduced ENPP3 低 RTCC efficacy on cells and restored ENPP3 高 Certain RTCC efficacy on cells.

[0123] Figure 43Depicts RTCC induction on CFSE-labeled KU812, as shown by Zombie Aqua staining of CFSE after 44 hours of incubation of CFSE-labeled target cells with human PBMCs (effector to target ratio 10:1) and the αENPP3 x αCD3 bispecific antibodies XENP28925 (CD3 high; monovalent ENPP3 binding), XENP29437 (CD3 high; bivalent ENPP3 binding), XENP29436 (CD3 high-Int No. 1; monovalent ENPP3 binding), or XENP29438 (CD3 high-Int No. 1; bivalent ENPP3 binding). + Unexpectedly, XENP29438 was unable to induce RTCC in KU812 cells.

[0124] Figure 44 Depicts the expression of CFSE-labeled KU812 (solid line, ENPP3 高 ) or CFSE-labeled RCC4 (dashed line, ENPP3 低 ) as induced by RTCC on CFSE-labeled target cells, as determined by Zombie Aqua staining of CFSE after 24 h of incubation with human PBMCs (effector to target ratio 10:1) and the αENPP3 x αCD3 bispecific antibodies XENP29437 (CD3 high VH / VL; bivalent ENPP3 binding), XENP30469 (CD3 high VL / VH; bivalent ENPP3 binding), XENP29428 (CD3 high-Int No. 1 VH / VL; bivalent ENPP3 binding), or XENP30470 (CD3 high-Int No. 2 VL / VH; bivalent ENPP3 binding). + The data show that swapping the orientation of the variable heavy and light domains in the CD3high-Int No. 1 scFv restored its activity in the context of a 2+1 Fab2-scFv-Fc bsAb format (XENP29438 vs. XENP30470). Swapping the orientation of the variable heavy and light domains in the CD3high scFv in the context of a 2+1 Fab2-scFv-Fc bsAb format (XENP29437 vs. XENP30469) resulted in a more modest improvement in RTCC potency.

[0125] Figure 45 Depicts the expression of CFSE-labeled KU812 (solid line, ENPP3 高 ) or CFSE-labeled RCC4 (dashed line, ENPP3 低) on RTCC, as determined by incubation of CFSE-labeled target cells with human PBMCs (at a 10:1 ratio of effector cells to target cells) and αENPP3 x CFSE-stained with Zombie Aqua after incubation for 40 hours with the αCD3 bispecific antibodies XENP29520 (CD3 High [VH / VL]; bivalent ENPP3 moderate binding), XENP30819 (CD3 High-Int No. 1 [VL / VHL]; bivalent ENPP3 moderate binding), XENP31149 (CD3 High-Int No. 2 [VL / VHL]; bivalent ENPP3 moderate binding), XENP30264 (CD3 High [VH / VL]; bivalent ENPP3 low binding), XENP30821 (CD3 High-Int No. 1 [VL / VHL]; bivalent ENPP3 low binding), or XENP31150 (CD3 High-Int No. 2 [VL / VHL]; bivalent ENPP3 low binding). + The percentages of cells are indicated.

[0126] Figure 46 Depicted are the sequences of XENP16432, an anti-PD-1 mAb based on nivolumab and an Ig backbone with the E233P / L234V / L235A / G236del / S267K ablation variant; and XENP21461 (pembrolizumab).

[0127] Figure 47 Depicted are changes in tumor volume over time (as determined by caliper measurement) in KU812- and huPBMC-engrafted NSG mice dosed with PBS, XENP16432 (a bivalent anti-PD-1 mAb), or illustrative αENPP3 x αCD3 2+1 bsAbs (XENP30819, XENP30821, or XENP31419) alone or in combination with XENP16432. Each of the αENPP3 x αCD3 bsAbs, at low and / or higher doses, enhanced the allogeneic anti-tumor effect of T cells against KU812 cells and combined well with PD-1 blockade.

[0128] Figures 48A to 48C Depicted are A) CD45 in the blood of NSG mice transplanted with KU812 and huPBMCs dosed with PBS, XENP16432 (a bivalent anti-PD-1 mAb), or an illustrative αENPP3 x αCD3 2+1 bsAb (XENP30819, XENP30821, or XENP31419) alone or in combination with XENP16432 by day 14. + Lymphocytes, B) CD8 +T cells and C) CD4 + T cell expansion. In all cases, combination with PD-1 blockade enhanced lymphocyte expansion.

[0129] Figure 49 Depicted are changes in tumor volume over time (as determined by caliper measurement) in RXF-393 and huPBMC-engrafted NSG mice dosed with PBS, XENP16432 (a bivalent anti-PD-1 mAb), or illustrative αENPP3 x αCD3 2+1 bsAbs (XENP30819 or XENP31419) alone or in combination with XENP16432. Each of the αENPP3 x αCD3 bsAbs, at low, medium, and / or high doses, enhanced the allogeneic anti-tumor effect of T cells against KU812 cells and combined well with PD-1 blockade.

[0130] Figures 50A to 50N Depicted are changes in tumor volume over time (as determined by caliper measurement) in KU812- and huPBMC-engrafted NSG mice dosed with A) PBS, B) XENP16432 (a bivalent anti-PD-1 mAb), or illustrative αENPP3 x αCD3 2+1 bsAbs (XENP30819, XENP30821, or XENP31419) alone or in combination with XENP16432. Each of the αENPP3 x αCD3 bsAbs, at low and / or higher doses, enhanced the allogeneic anti-tumor effect of T cells against KU812 cells and combined well with PD-1 blockade.

[0131] Figures 51A to 51L Depicted are changes in tumor volume over time (as determined by caliper measurement) in RXF-393- and huPBMC-engrafted NSG mice dosed with A) PBS, B) XENP16432 (a bivalent anti-PD-1 mAb), or illustrative αENPP3 x αCD3 2+1 bsAbs (XENP30819 or XENP31419) alone or in combination with XENP16432. Each of the αENPP3 x αCD3 bsAbs, at low, medium, and / or high doses, enhanced the allogeneic anti-tumor effect of T cells against KU812 cells and combined well with PD-1 blockade.

[0132] Figures 52A to 52KSeveral formats for the anti-ENPP3 x anti-CD3 bispecific antibodies disclosed herein are depicted. The first is a "1+1 Fab-scFv-Fc" format (also known as a "bottle opener" or "triple F" format), in which the first antigen-binding domain is a Fab domain and the second anti-antigen-binding domain is a scFv domain ( Figure 1A In addition, “mAb-Fv,” “mAb-scFv,” “2+1 Fab2-scFv-Fc” (also called “central scFv” or “central scFv” format), “central Fv,” “single-armed central scFv,” “single scFv-mAb,” “scFv-mAb,” “dual scFv,” “trident,” and non-heterodimer bispecific formats are shown. Figure 49 The scFv domains depicted in FIG52 can be variable heavy chain-(optional linker)-variable light chain, or variable light chain-(optional linker)-variable heavy chain, from N-terminus to C-terminus. In addition, for the single-arm scFv-mAb, the scFv can be attached to the N-terminus of the heavy chain monomer or to the N-terminus of the light chain. In certain embodiments, "anti-antigen 1" in FIG52 refers to an ENPP3 binding domain. In certain embodiments, "anti-antigen 1" in FIG52 refers to a CD3 binding domain. In certain embodiments, "anti-antigen 2" in FIG52 refers to an ENPP3 binding domain. In certain embodiments, "anti-antigen 2" in FIG52 refers to a CD3 binding domain. In certain embodiments, "anti-antigen 1" in FIG52 refers to an ENPP3 binding domain and "anti-antigen 2" in FIG52 refers to a CD3 binding domain. In certain embodiments, "anti-antigen 1" in FIG52 refers to an ENPP3 binding domain and "anti-antigen 2" in FIG52 refers to a CD3 binding domain. In certain embodiments, "anti-antigen 1" in FIG52 refers to a CD3 binding domain and "anti-antigen 2" in FIG52 refers to an ENPP3 binding domain. Any of the disclosed ENPP3 binding domains and CD3 binding domains can be included in the bispecific format of FIG. 52 .

[0133] Figure 53 Schematic diagrams of the heterodimeric Fc proteins described herein are provided, including 2:1 Fab2-scFv-Fc, 1:1 Fab=scFv-Fc, Y / Z-Fc (e.g., non-targeted interleukin-Fc), anti-X x Y / ZF (e.g., targeted interleukin-Fc), and one-armed Fc proteins.

[0134] Figure 54 A structural model of the CH3-CH3 interface constructed using MOE based on Protein Data Bank entry 3AVE is provided. A set of novel Fc substitutions enables a heterodimer yield exceeding 95% with little change in thermal stability.

[0135] Figure 55Isoallergic substitutions are depicted to minimize effects on tertiary structure. Engineered isoelectric point differences in the Fc region allow or facilitate direct purification of Fc heterodimers.

[0136] Figure 56 Hinge and CH2 substitutions are depicted to abrogate FcγR binding.

[0137] Figures 57A to 57C The 2:1 Fab2-scFv-Fc format was shown to enable targeting of tumor antigens at low density on normal cells. Tuning TAA potency and TAA / CD3 affinity enabled selective cytotoxicity of cell lines that mimic cancer and normal tissues (high / low antigen density). The tuned 2:1 bispecific antibody also reduced interference from soluble antigens and reduced cytokine release.

[0138] Figure 57A Tuning FAP potency and FAP / CD3 affinity enables selective cytotoxicity of cell lines mimicking cancer and normal tissue (high / low antigen density). XENP23535 represents a tuned 1:1 format targeting FAP. XENP25967 represents a tuned 2:1 format targeting FAP.

[0139] Figure 57B Tuning SSTR2 potency and SSTR2 / CD3 affinity enables selective cytotoxicity against cell lines that mimic cancer and normal tissues (high / low antigen density). XENP18087 represents a tuned 1:1 format targeting SSTR2. XENP30458 represents a tuned 2:1 format targeting SSTR2.

[0140] Figure 57C Tuning ENPP3 potency and ENPP3 / CD3 affinity enables selective cytotoxicity of cell lines mimicking cancer and normal tissues (high / low antigen density). XENP28925 represents a tuned 1:1 format targeting ENPP3. XENP31149 represents a tuned 2:1 format targeting ENPP3.

[0141] Figure 58 The advantages of using the method described herein for research-scale production of heterodimeric Fc proteins are described. This method can be used to directly produce heterodimeric Fc proteins.

[0142] Figure 59 Results for stable cell line development are shown for clones with high titers and high heterodimer prevalence. Shake flask yields for top clones ranged from 1-2 g / L with approximately 90% heterodimer content. Data were obtained after only a standard Protein A purification step.

[0143] Figure 60Depicted are the induction of RTCC by A) XENP18087 or B) XENP30458 on A549 cells transfected with SSTR2 (high, medium, and low density).

[0144] Figure 61 Depicted are the reductions in A) target cell numbers and B) IL-6, C) TNFα, D) IFNγ, and E) IL-1β release by effector cells after 48 hours of incubation of CFSE-labeled SSTR2+COR-L279 target cells with human PBMCs (effector:target ratio 20:1) in the presence of XENP18087 or XENP30458.

[0145] Figures 62A to 62D Sequences of illustrative 1:1 and 2:1 tuned TAA x CD3 bispecifics described herein. Anti-TTA (e.g., anti-FAP, anti-SSTR2, and anti-ENPP3) components such as variable regions, anti-CD3 components such as variable regions, constant / Fc regions, and linkers are shown. The linker is double-underlined (although one skilled in the art will appreciate that the linker can be replaced with another linker). Slashes ( / ) indicate the boundaries between the variable region, constant / Fc region, and linker. CDRs are underlined. In some embodiments, the 1:1 TAA x CD3 bispecific is XENP23535, XENP18087, or XENP28925. In some embodiments, the 2:1 TAA x CD3 bispecific is XENP25967, XENP30458, or XENP31149.

[0146] Figure 63 The sequence of the SSTR2 binding domain [αSSTR2]_H1.24_L1.30 is depicted. DETAILED DESCRIPTION

[0147] I. Overview

[0148] Bispecific antibodies that co-bind CD3 and tumor antigen targets are used to redirect T cells to attack and lyse targeted tumor cells. Examples include BiTE® and DART formats, which monovalently bind CD3 and tumor antigens. While CD3-targeted approaches have shown considerable promise, a common side effect of such therapies is associated cytokine production, often leading to toxic cytokine release syndrome. Because the anti-CD3 binding domain of the bispecific antibody binds to all T cells, it recruits a high cytokine-producing CD4 T cell subset. Furthermore, this CD4 T cell subset includes regulatory T cells, the recruitment and expansion of which may lead to immunosuppression and negatively impact long-term tumor suppression. Furthermore, these formats do not contain an Fc domain and exhibit very short serum half-lives in patients.

[0149] Provided herein are novel anti-CD3 x anti-ENPP3 (also referred to as anti-ENPP3 x anti-CD3, αCD3 x αENPP3, or αENPP3 x αCD3) heterodimeric bispecific antibodies and methods of using such antibodies to treat cancer. Specifically, provided herein are various forms of anti-CD3, anti-ENPP3 bispecific antibodies, such as Figure 15A and Figure 15B These bispecific antibodies can be used to treat cancers, particularly those with increased ENPP3 expression, such as renal cell carcinoma. Such antibodies are used to direct CD3+ effector T cells to ENPP3+ tumors, thereby allowing the CD3+ effector T cells to attack and lyse the ENPP3+ tumors.

[0150] Furthermore, in some embodiments, the present disclosure provides bispecific antibodies with different binding affinities for human CD3, which can modify or reduce potential side effects of anti-CD3 therapies. That is, in some embodiments, the antibodies described herein provide antibody constructs comprising an anti-CD3 antigen-binding domain that is a "strong" or "high-affinity" binder to CD3 (e.g., one example is the heavy and light chain variable domains described as H1.30_L1.47 (optionally including an appropriate charged linker)) and also binds to ENPP3. In other embodiments, the antibodies described herein provide antibody constructs comprising an anti-CD3 antigen-binding domain that is a "mild" or "low-affinity" binder to CD3. Further embodiments provide antibody constructs comprising an anti-CD3 antigen-binding domain that has a moderate or "intermediate" affinity for CD3 that also binds to ENPP3. While a very large number of anti-CD3 antigen binding domains (ABDs) can be used, a particularly useful embodiment uses six different anti-CD3 ABDs, although these can be used in two scFv orientations as discussed herein. Affinity is typically measured using a Biacore assay.

[0151] It should be understood that the "high, medium, and low" anti-CD3 sequences provided herein can be used in a variety of heterodimeric forms, such as Figure 15A 、 Figure 15B Typically, due to potential side effects of T cell recruitment, exemplary embodiments utilize formats that only monovalently bind CD3, e.g. Figure 15A and Figure 15B As depicted in, and in the form depicted herein, the CD3 ABD is a scFv as more fully described herein. In contrast, the subject bispecific antibodies can be monovalent (e.g. Figure 15A ) or bivalent (e.g. Figure 15B ) binds ENPP3.

[0152] Provided herein are compositions comprising ENPP3 binding domains, comprising antibodies having such ENPP3 binding domains (e.g., ENPP3 x CD3 bispecific antibodies). The subject antibodies comprising such ENPP3 binding domains advantageously elicit a range of different immune responses, depending on the specific ENPP3 binding domain used. For example, the subject antibodies exhibit differences in selectivity for cells with varying ENPP3 expression, potency against ENPP3-expressing cells, ability to elicit cytokine release, and sensitivity to soluble ENPP3. Such ENPP3 binding domains and related antibodies can be used, for example, to treat ENPP3-associated cancers.

[0153] Thus, in one aspect, provided herein are heterodimeric antibodies that bind to two different antigens, e.g., antibodies that are "bispecific" because they bind to two different target antigens, typically ENPP3 and CD3 as described herein. These heterodimeric antibodies can bind to these target antigens monovalently (e.g., a single antigen-binding domain is present, such as a variable heavy chain and variable light chain domain pair) or bivalently (two antigen-binding domains are present that each independently bind to an antigen). In some embodiments, the heterodimeric antibodies provided herein comprise one CD3-binding domain and one ENPP3-binding domain (e.g., a "1+1 Fab-scFv-Fc" format heterodimeric antibody described herein). In other embodiments, the heterodimeric antibodies provided herein comprise one CD3-binding domain and two ENPP3-binding domains (e.g., a "2+1 Fab2-scFv-Fc" format heterodimeric antibody described herein). The heterodimeric antibodies provided herein are based on the use of different monomers containing amino acid substitutions that "skew" the formation of heterodimers compared to homodimers, as more fully outlined below, in combination with "pI variants" that allow for easy purification of heterodimers away from homodimers, as similarly outlined below. The heterodimeric bispecific antibodies provided generally rely on the use of engineered or variant Fc domains that can self-assemble in production cells to produce heterodimeric proteins, as well as methods for producing and purifying such heterodimeric proteins.

[0154] II. Nomenclature

[0155] The antibodies provided herein are listed in several different formats. In some instances, each monomer in a particular antibody is given a unique "XENP" number, but as will be understood in the art, longer sequences may contain shorter XENP numbers. For example, the "scFv-Fc" monomer of an antibody in the format of 1+1 Fab-scFv-Fc may have a first XENP number, while the scFv domain itself will have a different XENP number. Some molecules have three polypeptides, so the XENP numbers of the components are used as the names. Thus, the molecule XENP29520 in the format of 2+1 Fab2-scFv-Fc includes three sequences (see Figure 19A ) 1) "Fab-Fc heavy chain" monomer; 2) "Fab-scFv-Fc heavy chain" monomer; and 3) "light chain" monomer or equivalent, but those skilled in the art will be able to readily identify these sequences through sequence alignment. These XENP numbers are provided in the sequence listings and identifiers, and are used in the accompanying figures. Furthermore, a single molecule containing three components generates multiple sequence identifiers. For example, the listing for Fab includes the full heavy chain sequence, the variable heavy chain domain sequence, and the three CDRs of the variable heavy chain domain sequence; the full light chain sequence, the variable light chain domain sequence, and the three CDRs of the variable light chain domain sequence. The Fab-scFv-Fc monomer includes the full-length sequence, the variable heavy chain domain sequence, the three heavy chain CDRs, and an scFv sequence (including the scFv variable heavy chain domain sequence, the scFv variable light chain domain sequence, and the scFv linker). Note that some molecules with scFv domains described herein use a single charged scFv linker (+H), but others may also be used. Additionally, the nomenclature for specific antigen binding domains (e.g., ENPP3 and CD3 binding domains) uses the format "Hx.xx_Ly.yy," where the numbering serves as a unique identifier for a specific variable chain sequence. Thus, the variable domain on the Fab side of the CD3 binding domain AN1[ENPP3] H1L1 (e.g., Figure 12 ) is "H1 L1," which indicates that the variable heavy chain domain H1 is combined with the light chain domain L1. When these sequences are used as scFvs, the designation "H1 L1" indicates that the variable heavy chain domain, H1, is combined with the light chain domain L1 and is in a VH-linker-VL orientation from N-terminus to C-terminus. Such a molecule having identical sequences for the heavy and light chain variable domains, but in reverse order (VL-linker-VH orientation from N-terminus to C-terminus), would be designated "L1_H1.1." Similarly, different constructs may "mix and match" heavy and light chains, as will be apparent from the sequence listing and accompanying figures.

[0156] III. Definitions

[0157] In order that this application may be more fully understood, several definitions are set forth below. Such definitions are intended to encompass grammatical equivalents.

[0158] "ENPP3" or "ectonucleotide pyrophosphatase / phosphodiesterase family member 3" (e.g., GenBank Accession No. NP005012.2) refers herein to a protein that belongs to a family of extracellular enzymes that hydrolyze residual extracellular nucleotides. The ENPP3 sequence is depicted in e.g. Figure 11A and Figure 11B ENPP3 is expressed in certain cancers, including renal cell carcinoma.

[0159] By "ablate" herein is meant a reduction or removal of an activity. Thus, for example, "ablates FcγR binding" means that the Fc region amino acid variant has less than 50% of the initial binding, preferably greater than 70% to 80% to 90% to 95% to 98% loss of activity compared to an Fc region without the specific variant, and typically, the activity is below the level of detectable binding in a Biacore, SPR or BLI assay. Particularly useful for ablation of FcγR binding are Figure 5 Those variations shown, which typically add to two monomers.

[0160] As used herein, "ADCC" or "antibody-dependent cell-mediated cytotoxicity" refers to a cell-mediated reaction in which nonspecific cytotoxic cells expressing FcγRs recognize bound antibody on a target cell and subsequently cause lysis of the target cell. ADCC is associated with binding to FcγRIIIa; increased binding to FcγRIIIa results in increased ADCC activity.

[0161] As used herein, "ADCP" or antibody-dependent cell-mediated phagocytosis refers to a cell-mediated reaction in which nonspecific phagocytes expressing FcγRs recognize bound antibody on a target cell and subsequently cause phagocytosis of the target cell.

[0162] As used herein, the term “antibody” is generally used. The antibodies described herein can take a variety of forms as described herein, including traditional antibodies as well as antibody derivatives, fragments and mimetics as described herein.

[0163] Traditional immunoglobulin (Ig) antibodies are "Y"-shaped tetramers. Each tetramer is typically composed of two identical pairs of polypeptide chains, each pair having one "light chain" monomer (usually with a molecular weight of about 25 kDa) and one "heavy chain" monomer (usually with a molecular weight of about 50-70 kDa).

[0164] Other useful antibody formats include, but are not limited to, the 1+1 Fab-scFv-Fc and 2+1 Fab-scFv-Fc antibody formats described herein, as well as "mAb-Fv," "mAb-scFv," "central-Fv," "single-armed scFv-mAb," "scFv-mAb," "double scFv," and "trident" antibody formats, such as Figure 49 shown.

[0165] An antibody heavy chain typically comprises a variable heavy chain (VH) domain, comprising vh CDR1-3, and an Fc domain, comprising a CH2-CH3 monomer. In some embodiments, an antibody heavy chain comprises a hinge and CH1 domain. Traditional antibody heavy chains are monomers organized from N-terminus to C-terminus as follows: VH-CH1-hinge-CH2-CH3. CH1-hinge-CH2-CH3 are collectively referred to as the antibody heavy chain "constant domain" or "constant region," of which there are five distinct classes or "isotypes": IgA, IgD, IgG, IgE, and IgM. Thus, as used herein, "isotype" refers to any subclass of immunoglobulins defined by the chemical and antigenic characteristics of their constant regions. It should be understood that therapeutic antibodies may also comprise hybrids of isotypes and / or subclasses. For example, as described in U.S. Publication No. 2009 / 0163699, incorporated by reference, the antibodies described herein comprise a human IgG1 / G2 hybrid.

[0166] In some embodiments, the antibodies provided herein include IgG isotype constant domains, which have several subclasses, including but not limited to IgG1, IgG2, IgG3, and IgG4. Within the IgG subclass of immunoglobulins, several immunoglobulin domains are present in the heavy chain. By "immunoglobulin (Ig) domain" herein is meant a region of an immunoglobulin with different tertiary structures. Of interest in the antibodies described herein are the heavy chain domains, which include the constant heavy chain (CH) domain and the hinge domain. In the context of IgG antibodies, each IgG isotype has three CH regions. Thus, in the context of IgG, the "CH" domains are as follows: "CH1" refers to positions 118-220 according to EU numbering as in Kabat. "CH2" refers to positions 237-340 according to EU numbering as in Kabat, and "CH3" refers to positions 341-447 according to EU numbering as in Kabat. As shown herein and described below, pI variants can be in one or more CH regions as well as the hinge region, as discussed below.

[0167] It should be noted that IgG1 has different allotypes with polymorphisms at 356 (D or E) and 358 (L or M). The sequences described herein use the 356D / 358M allotype, however other allotypes are contemplated herein. That is, any sequence comprising an IgG1 Fc domain encompassed herein may have 356E / 358L in place of the 356D / 358M allotypes. It should be understood that therapeutic antibodies may also comprise hybrids of isotypes and / or subclasses. For example, as described in U.S. Publication No. 2009 / 0163699, incorporated by reference, the antibodies of the present invention, in some embodiments, comprise an IgG1 / IgG2 hybrid.

[0168] As used herein, "Fc" or "Fc region" or "Fc domain" refers to a polypeptide comprising the constant region of an antibody, in some cases excluding all of the first constant region immunoglobulin domain (e.g., C1) or a portion thereof, and in some cases, optionally comprising all or a portion of the hinge. For IgG, the Fc domain includes the immunoglobulin domains CH2 and CH3 (Cγ2 and Cγ3), and optionally all or a portion of the hinge region between CH1 (Cγ1) and CH2 (Cγ2). Thus, in some cases, the Fc domain comprises, from N-terminus to C-terminus, CH2-CH3 and hinge-CH2-CH3. In some embodiments, the Fc domain is an Fc domain from IgG1, IgG2, IgG3, or IgG4, with IgG1 hinge-CH2-CH3 and IgG4 hinge-CH2-CH3 being particularly useful in many embodiments. Additionally, in the case of a human IgG1 Fc domain, the hinge often comprises a C220S amino acid substitution. In addition, in the case of a human IgG4 Fc domain, the hinge often comprises an S228P amino acid substitution. Although the boundaries of the Fc region can vary, the human IgG heavy chain Fc region is generally defined as comprising residues E216, C226, or A231 at its carboxyl terminus, where numbering is performed according to the EU index as in Kabat. In some embodiments, as described more fully below, amino acid modifications are made to the Fc region, for example, to alter binding to one or more FcγRs or to FcRn.

[0169] By "heavy chain constant region" herein is meant the CH1-hinge-CH2-CH3 portion of an antibody (or fragment thereof), excluding the variable heavy chain domain; in EU numbering of human IgG1, this is amino acids 118 to 447. By "heavy chain constant region fragment" herein is meant a heavy chain constant region that contains fewer amino acids from either or both of the N-terminus and the C-terminus but still retains the ability to form a dimer with another heavy chain constant region.

[0170] Another type of Ig domain of the heavy chain is the hinge region. "Hinge" or "hinge region" or "antibody hinge region" or "hinge domain" herein refers to the flexible polypeptide comprising the amino acids between the first constant domain and the second constant domain of an antibody. Structurally, the IgG CH1 domain ends at EU position 215, and the IgG CH2 domain begins at residue EU position 231. Therefore, herein, for IgG, the antibody hinge is defined to comprise positions 216 (E216 in IgG1) to 230 (p230 in IgG1), where numbering is according to the EU index as in Kabat. In some cases, a "hinge fragment" is used, which contains fewer amino acids at either or both of the N-terminus and C-terminus of the hinge domain. As mentioned herein, pI variants can also be made in the hinge region. Many antibodies herein have at least one cysteine ​​at position 220 according to the EU code (hinge region) replaced by a serine. Typically, this modification is on the "scFv monomer" side of most of the sequences depicted herein, but it can also be on the "Fab monomer" side or on both to reduce disulfide formation. Specifically included in the sequences herein is a substitution of one or both of these cysteines (C220S).

[0171] As will be appreciated by those skilled in the art, the exact numbering and position of the heavy chain constant region domains may differ between different numbering systems. A useful comparison of the heavy chain constant regions according to EU and Kabat numbering is as follows, see Edelman et al., 1969, Proc Natl Acad Sci USA 63:78-85 and Kabat et al., 1991, Sequences of Proteins of Immunological Interest, 5th ed., United States Public Health Service, National Institutes of Health, Bethesda, which are incorporated herein by reference in their entirety.

[0172] Table 1

[0173]

[0174] An antibody light chain typically comprises two domains: a variable light chain domain (VL), which comprises the light chain CDRs vl and CDR1-3, and a constant light chain region (often referred to as CL or CK). Antibody light chains are typically organized from N-terminus to C-terminus as follows: VL-CL.

[0175] As used herein, "antigen binding domain" or "ABD" refers to a set of six complementarity determining regions (CDRs) that, when present as part of a polypeptide sequence, specifically bind to a target antigen as discussed herein (e.g., ENPP3 or CD3). As is known in the art, these CDRs typically exist as a first set of variable heavy chain CDRs (vhCDRs or VHCDRs) and a second set of variable light chain CDRs (vlCDRs or VLCDRs), each comprising three CDRs: vhCDR1, vhCDR2, and vhCDR3, variable heavy chain CDRs, and vlCDR1, vlCDR2, and vlCDR3, and vhCDR3, variable light chain CDRs. The CDRs are present in the variable heavy chain domain (vhCDR1-3) and the variable light chain domain (vlCDR1-3). The variable heavy chain domain and variable light chain domain are derived from the Fv region.

[0176] The antibodies described herein provide a large number of different CDR sets. In this case, a "complete CDR set" comprises three variable light chains and three variable heavy chain CDRs, such as vlCDR1, vlCDR2, vlCDR3, vhCDR1, vhCDR2, and vhCDR3. These can be part of a larger variable light chain domain or a variable heavy chain domain, respectively. In addition, as more fully outlined herein, when heavy and light chains are used (e.g., when Fab is used), the variable heavy chain domain and the variable light chain domain can be on separate polypeptide chains, or in the case of scFv sequences, on a single polypeptide chain.

[0177] As will be appreciated by those skilled in the art, the exact numbering and placement of the CDRs in different numbering systems may be different. However, it should be understood that disclosure of the variable heavy chain sequence and / or variable light chain sequence encompasses disclosure of the associated (intrinsic) CDRs. Thus, the disclosure of each variable heavy region is the disclosure of the vhCDRs (e.g., vhCDR1, vhCDR2, and vhCDR3), and the disclosure of each variable light region is the disclosure of the vlCDRs (e.g., vlCDR1, vlCDR2, and vlCDR3). A useful comparison of the CDR numbering is as follows, see Lafranc et al., Dev. Comp. Immunol. 27(1):55-77 (2003):

[0178] Table 2

[0179]

[0180] Throughout this specification, the Kabat numbering system (e.g., Kabat et al., supra (1991)) is generally used when referring to residues in the variable domain (approximately residues 1-107 in the light chain variable region and residues 1-113 in the heavy chain variable region), as well as the EU numbering system for the Fc region.

[0181] CDRs contribute to the formation of the antigen-binding domain and the antigen-binding site of an antibody, or more specifically, the epitope binding site. An "epitope" refers to a determinant that interacts with a specific antigen-binding site in the variable region of an antibody molecule, called a paratope. An epitope is a group of molecules, such as amino acids or sugar side chains, and typically has specific structural characteristics as well as specific charge characteristics. A single antigen may have more than one epitope.

[0182] An epitope can include amino acid residues that are directly involved in binding (also called the immunodominant component of the epitope) and other amino acid residues that are not directly involved in binding, such as amino acid residues that are effectively blocked by the specific antigen-binding peptide; in other words, the amino acid residues are within the coverage area of ​​the specific antigen-binding peptide.

[0183] Epitopes can be conformational or linear. Conformational epitopes are generated by spatially juxtaposed amino acids from different segments of a linear polypeptide chain. Linear epitopes are epitopes generated by adjacent amino acid residues in a polypeptide chain. Conformational and non-conformational epitopes can be distinguished by the loss of binding to the former, but not the latter, in the presence of denaturing solvents.

[0184] An epitope typically comprises at least 3, and more typically at least 5 or 8-10, amino acids in a unique spatial conformation. Antibodies that recognize the same epitope can be verified in a simple immunoassay by demonstrating the ability of one antibody to block the binding of another antibody to the target antigen, such as "box-in." As outlined below, the present disclosure includes not only the antigen-binding domains and antibodies listed herein, but also antigen-binding domains and antibodies that compete for binding to the epitope to which the listed antigen-binding domains bind.

[0185] In some embodiments, the six CDRs of the antigen-binding domain are derived from a variable heavy chain domain and a variable light domain. In the "Fab" format, a set of six CDRs is derived from two different polypeptide sequences, a variable heavy chain domain (vh or VH; containing vhCDR1, vhCDR2, and vhCDR3) and a variable light domain (vl or VL; containing vlCDR1, vlCDR2, and vlCDR3), with the C-terminus of the vh domain linked to the N-terminus of the CH1 domain of the heavy chain and the C-terminus of the vl domain linked to the N-terminus of the constant light domain (and thus forming a light chain). In the scFv format, the vh and vl domains are typically bivalently linked into a single polypeptide sequence using a linker as outlined herein ("scFv linker"), which can be (starting from the N-terminus) vh-linker-vl or vl-linker-vh, with the former generally preferred (including optional domain linkers on each side, depending on the format used (e.g., according to Figure 1)). Typically, the C-terminus of the scFv domain is attached to the N-terminus of the hinge in the second monomer.

[0186] As used herein, "variable region" or "variable domain" means a region of an immunoglobulin that includes one or more Ig domains that are substantially encoded by any of the Vκ, Vλ and / or VH genes that constitute the κ, λ and heavy chain immunoglobulin gene loci, respectively, and contains the CDRs that confer antigen specificity. Thus, a "variable heavy chain domain" is paired with a "variable light domain" to form an antigen binding domain ("ABD"). In addition, each variable domain includes three hypervariable regions ("complementarity determining regions," "CDRs") arranged in the following order from amino-terminus to carboxyl-terminus (VHCDR1, VHCDR2 and VHCDR3 of the variable heavy chain domain and VLCDR1, VLCDR2 and VLCDR3 of the variable light domain) and four framework (FR) regions: FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4. The hypervariable region generally encompasses amino acid residues 24-34 (LCDR1; "L" for light chain), 50-56 (LCDR2), and 89-97 (LCDR3) in the light chain variable region and approximately 31-35B (HCDR1; "H" for heavy chain), 50-65 (HCDR2), and 95-102 (HCDR3) in the heavy chain variable region; Kabat et al., SEQUENCES OF PROTEINS OF IMMUNOLOGICAL INTEREST, 5th ed., Public Health Service, National Institutes of Health, Bethesda, MD. Bethesda, Md.) (1991) and / or those residues forming the hypervariable loops (e.g., residues 26-32 (LCDR1), 50-52 (LCDR2), and 91-96 (LCDR3) in the light chain variable region and 26-32 (HCDR1), 53-55 (HCDR2), and 96-101 (HCDR3) in the heavy chain variable region; Chothia and Lesk (1987) J. Mol. Biol. 196:901-917. Specific CDRs of the invention are described in Table 2.

[0187] As used herein, "Fab" or "Fab region" refers to a polypeptide comprising the VH, CH1, VL, and CL immunoglobulin domains, typically located on two different polypeptide chains (e.g., VH-CH1 on one chain and VL-CL on another chain). Fab can refer to this region in isolation or in the context of the bispecific antibodies described herein. In the context of Fab, in addition to the CH1 and CL domains, Fab also includes the Fv region.

[0188] As used herein, "Fv" or "Fv fragment" means a polypeptide comprising the VL and VH domains of an ABD. The Fv region can be formatted as both a Fab (as discussed above, typically two different polypeptides that also contain a constant region as outlined above) and a scFv, in which the VL and VH domains are combined (typically with a linker as discussed herein) to form the scFv.

[0189] By "single-chain Fv" or "scFv" herein is meant a variable heavy chain domain covalently attached to a variable light chain domain, typically using an scFv linker as discussed herein, to form an scFv or scFv domain. The scFv domains can be oriented in either direction from N-terminus to C-terminus (VH-linker-VL or VL-linker-VH). In the sequences depicted in the sequence listing and figures, the order of the VH and VL domains is indicated in the name, e.g., H.X_L.Y means N-terminus to C-terminus is VH-linker-VL, and L.Y_H.X is VL-linker-VH.

[0190] Some embodiments of the antibodies of the invention provided herein include at least one scFv domain that, although not naturally occurring, typically comprises a variable heavy chain domain and a variable light domain connected together by an scFv linker. As outlined herein, although scFv domains are typically oriented from N-terminus to C-terminus as VH-scFv linker-VL, this orientation can be reversed for any one scFv domain (or those constructed using the vh and vl sequences of a Fab) to become VL-scFv linker-VH, with an optional linker located at one or both ends depending on the format.

[0191] As used herein, "modification" refers to amino acid substitutions, insertions, and / or deletions in a polypeptide sequence or alterations to moieties chemically linked to a protein. For example, a modification can be an altered carbohydrate or PEG structure attached to a protein. As used herein, "amino acid modification" refers to amino acid substitutions, insertions, and / or deletions in a polypeptide sequence. For clarity, unless otherwise indicated, amino acid modifications are always directed to amino acids encoded by DNA, e.g., the 20 amino acids having codons in DNA and RNA.

[0192] As used herein, an "amino acid substitution" or "substitution" refers to replacing an amino acid at a specific position in a parent polypeptide sequence with a different amino acid. In particular, in some embodiments, the substitution is for an amino acid that does not naturally occur at a specific position, does not naturally occur in an organism, or in any organism. For example, the substitution E272Y refers to a variant polypeptide, in this case an Fc variant, in which the glutamic acid at position 272 is replaced with tyrosine. For clarity, a protein that has been engineered to alter the nucleic acid coding sequence without changing the starting amino acid (e.g., replacing CGG (encoding arginine) with CGA (still encoding arginine) to increase expression levels in a host organism) is not an "amino acid substitution"; that is, although a new gene encoding the same protein is generated, if the protein has the same amino acid at the specific position where it starts, the protein is not an amino acid substitution.

[0193] As used herein, "amino acid insertion" or "insertion" means adding an amino acid sequence at a specific position in a parent polypeptide sequence. For example, -233E or 233E indicates the insertion of glutamic acid after position 233 and before position 234. In addition, -233ADE or A233ADE indicates the insertion of AlaAspGlu after position 233 and before position 234.

[0194] As used herein, "amino acid deletion" or "deletion" refers to the removal of an amino acid sequence at a specific position in a parent polypeptide sequence. For example, E233- or E233#, E233(), or E233del represents a deletion of glutamic acid at position 233. In addition, EDA233- or EDA233# represents a deletion of the sequence GluAspAla starting at position 233.

[0195] As used herein, a "variant protein" or "protein variant" or "variant" refers to a protein that differs from a parent protein by at least one amino acid modification. A protein variant has at least one amino acid modification compared to the parent protein, but not so many that the variant protein will not align with the parent protein using an alignment program as described below. Typically, variant proteins (such as the variant Fc domains outlined herein, etc.) are typically at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to the parent protein using an alignment program such as BLAST as described below. As used herein, "variant" also refers to a specific amino acid modification that confers a specific function (e.g., "heterodimerization variant," "pi variant," "ablation variant," etc.).

[0196] As described below, in some embodiments, the parent polypeptide, e.g., the Fc parent polypeptide, is a human wild-type sequence, such as a heavy constant domain or Fc region from IgG1, IgG2, IgG3, or IgG4, but human sequences with variants can also be used as "parent polypeptides," such as the IgG1 / 2 hybrids of U.S. Publication No. 2006 / 0134105. Protein variant sequences herein will preferably have at least about 80% identity to the parent protein sequence, and most preferably at least about 90% identity, more preferably at least about 95% to 98% to 99% identity. Thus, as used herein, "antibody variant" or "variant antibody" means an antibody that differs from a parent antibody by at least one amino acid modification, as used herein, "IgG variant" or "variant IgG" means an antibody that differs from a parent IgG (again, in many cases, from a human IgG sequence) by at least one amino acid modification, and as used herein, "immunoglobulin variant" or "variant immunoglobulin" means an immunoglobulin sequence that differs from a parent immunoglobulin sequence by at least one amino acid modification. As used herein, "Fc variant" or "variant Fc" means a protein that includes amino acid modifications in the Fc domain as compared to the Fc domain of human IgG1, IgG2, or IgG4.

[0197] As used herein, "Fc variant" or "variant Fc" refers to a protein that includes amino acid modifications in the Fc domain. Modifications can be additions, deletions, or substitutions. Fc variants are defined according to the amino acid modifications that constitute them. Thus, for example, N434S or 434S is an Fc variant having a serine substitution at position 434 relative to the parent Fc polypeptide, where numbering is according to the EU index. Similarly, M428L / N434S defines an Fc variant having the substitutions M428L and N434S relative to the parent Fc polypeptide. The identity of the WT amino acids can be non-specific, in which case the variant is referred to as 428L / 434S. It should be noted that the order in which the substitutions are presented is arbitrary, that is, for example, 428L / 434S is the same Fc variant as 434S / 428L, etc. For all positions discussed herein with respect to antibodies or derivatives and fragments thereof (e.g., Fc domains), amino acid position numbering is according to the EU index unless otherwise indicated. The "EU index" or "EU index" as in Kabat or "EU numbering" scheme refers to the EU numbering of antibodies (Edelman et al., 1969, Proc. Natl. Acad. Sci. USA 63:78-85, hereby incorporated by reference in its entirety).

[0198] Typically, the variant Fc domain is at least about 80%, 85%, 90%, 95%, 97%, 98% or 99% identical to the corresponding parent human IgG Fc domain (using the identity algorithm discussed below, one embodiment of which utilizes the BLAST algorithm as known in the art, using default parameters). Alternatively, the variant Fc domain may have 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 amino acid modifications compared to the parent Fc domain. Alternatively, the variant Fc domain may have up to 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acid modifications compared to the parent Fc domain. Additionally, as discussed herein, the variant Fc domains described herein still retain the ability to form a dimer with another Fc domain as measured using known techniques such as native gel electrophoresis as described herein.

[0199] As used herein, "protein" refers to at least two covalently linked amino acids, including proteins, polypeptides, oligopeptides, and peptides. In addition, the polypeptides constituting the antibodies described herein may include one or more side chains or terminal synthetic derivatizations, glycosylation, pegylation, cyclic transformations, cyclizations, linkers to other molecules, fusions to proteins or protein domains, and the addition of peptide tags or labels.

[0200] As used herein, "residue" refers to a position in a protein and its associated amino acid entity. For example, asparagine 297 (also known as Asn297 or N297) is the residue at position 297 in the human antibody IgG1.

[0201] As used herein, "IgG subclass modification" or "isotype modification" refers to an amino acid modification that converts one amino acid of an IgG isotype to the corresponding amino acid in a differently aligned IgG isotype. For example, because IgG1 contains a tyrosine and IgG2 contains a phenylalanine at EU position 296, a F296Y substitution in IgG2 is considered an IgG subclass modification.

[0202] As used herein, "non-naturally occurring modification" means an amino acid modification that is not an isotype. For example, because human IgG does not include a serine at position 434, the substitution 434S of IgG1, IgG2, IgG3, or IgG4 (or a hybrid thereof) is considered a non-naturally occurring modification.

[0203] As used herein, "amino acid" and "amino acid identity" mean one of the 20 naturally occurring amino acids encoded by DNA and RNA.

[0204] As used herein, "effector function" refers to the biochemical events resulting from the interaction of an antibody Fc region with an Fc receptor or ligand. Effector functions include, but are not limited to, ADCC, ADCP, and CDC.

[0205] As used herein, "IgG Fc ligand" means a molecule, preferably a polypeptide, from any organism that binds to the Fc region of an IgG antibody to form an Fc / Fc ligand complex. Fc ligands include, but are not limited to, FcγRI, FcγRII, FcγRIII, FcRn, C1q, C3, mannan binding lectin, mannose receptor, staphylococcal protein A, streptococcal protein G, and viral FcγR. Fc ligands also include Fc receptor homologs (FcRH), which are Fc receptor families homologous to FcγR (Davis et al., 2002, "Immunological Reviews" 190: 123-136, which are incorporated herein by reference in their entirety). Fc ligands may include undiscovered molecules that bind to Fc. Specific IgG Fc ligands are FcRn and Fcγ receptors. As used herein, "Fc ligand" means a molecule, preferably a polypeptide, from any organism that binds to the Fc region of an antibody to form an Fc / Fc ligand complex.

[0206] As used herein, "Fcγ receptor" or "FcγR" or "FcgammaR" means any member of the family of proteins that bind to the Fc region of an IgG antibody and are encoded by an FcγR gene. In humans, this family includes, but is not limited to, FcγRI (CD64), including isoforms FcγRIa, FcγRIb, and FcγRIc; FcγRII (CD32), including isoforms FcγRIIa (including allotypes H131 and R131), FcγRIIb (including FcγRIIb-1 and FcγRIIb-2), and FcγRIIc; and FcγRIII (CD16), including isoforms FcγRIIIa (including allotypes V158 and F158) and FcγRIIIb (including allotypes FcγRIIIb-NA1 and FcγRIIIb-NA2) (Jefferis et al., 2002, Immunol Lett. 82:57-65, which is incorporated herein by reference in its entirety), as well as any undiscovered human FcγR or FcγR isoform or allotype. FcγRs can be from any organism, including but not limited to humans, mice, rats, rabbits, and monkeys. Mouse FcγRs include but are not limited to FcγRI (CD64), FcγRII (CD32), FcγRIII (CD16), and FcγRIII-2 (CD16-2), as well as any undiscovered mouse FcγR or FcγR isoform or allotype.

[0207] As used herein, "FcRn" or "neonatal Fc receptor" refers to a protein that binds to the Fc region of an IgG antibody and is at least partially encoded by the FcRn gene. FcRn can be from any organism, including but not limited to humans, mice, rats, rabbits, and monkeys. As is known in the art, a functional FcRn protein comprises two peptides, commonly referred to as a heavy chain and a light chain. The light chain is beta-2 microglobulin and the heavy chain is encoded by the FcRn gene. Unless otherwise indicated herein, FcRn or FcRn protein refers to a complex of the FcRn heavy chain and beta-2-microglobulin. Various FcRn variants are used to enhance binding to the FcRn receptor and, in some cases, increase serum half-life. "FcRn variants" enhance binding to the FcRn receptor, and suitable FcRn variants are shown below.

[0208] As used herein, "parent polypeptide" means the starting polypeptide that is subsequently modified to produce a variant. The parent polypeptide can be a naturally occurring polypeptide or a variant or engineered version of a naturally occurring polypeptide. Thus, as used herein, "parent immunoglobulin" means an unmodified immunoglobulin polypeptide that is modified to produce a variant, and as used herein, "parent antibody" means an unmodified antibody that is modified to produce a variant antibody. It should be noted that "parent antibody" includes known commercial recombinantly produced antibodies as outlined below. In this context, "parent Fc domain" will be relative to the enumerated variants; thus, a "variant human IgG1 Fc domain" will be compared to the parent Fc domain of human IgG1, a "variant human IgG4 Fc domain" will be compared to the parent Fc domain of human IgG4, etc.

[0209] As used herein, "position" refers to a position in a protein sequence. Positions can be numbered sequentially or according to a defined format (e.g., the EU index for antibody numbering).

[0210] As used herein, "target antigen" means a molecule that is specifically bound by an antigen binding domain comprising the variable region of a given antibody.

[0211] The lower context of the monomers of the heterodimeric antibodies described herein is that by "strandedness" is meant to be similar to two DNA chains that are "matched", and heterodimerization variants are incorporated into each monomer to retain the ability to "match" to form heterodimers. For example, if some pI variants are engineered into monomer A (e.g., to make the pI higher), steric variants that can also be used as "charge pairs" will not interfere with the pI variants, for example, charge variants that make the pI higher are placed on the same "strand" or "monomer" to retain both functions. Similarly, for "skewed" variants that appear in pairs in a group as more fully outlined below, those skilled in the art consider pI when deciding which chain or monomer the pair will enter, so that the pI separation is also maximized using the pI of the skewed variant.

[0212] As used herein, "target cell" refers to a cell that expresses a target antigen.

[0213] In the context of producing the bispecific antibodies described herein, a "host cell" refers to a cell that contains exogenous nucleic acid encoding the components of the bispecific antibody and is capable of expressing the bispecific antibody under suitable conditions. Suitable host cells are discussed below.

[0214] "Wild type or WT" herein means an amino acid sequence or nucleotide sequence found in nature, including allelic variants. A WT protein has an amino acid sequence or nucleotide sequence that has not been intentionally modified.

[0215] Provided herein are multiple antibody domains having sequence identity to human antibody domains. The sequence identity between two similar sequences (e.g., antibody variable domains) can be measured by algorithms such as Smith, TF and Waterman, MS (1981) "Comparison of Biosequences", Adv. Appl. Math. 2:482 [local homology algorithm]; Needleman, SB and Wunsch, CD. (1970) "A General Method Applicable To The Search For Similarities In The Amino Acid Sequence Of Two Proteins", J. Mol. Biol. 48:443 [homologous alignment algorithm], Pearson, WR and Lipman, DJ (1988) "Improved Tools For Biological Sequence Comparison", The National Academy of Sciences of the United States of America (USA) 85:2444 [Similarity search method]; or Altschul, SF et al. (1990) "Basic Local Alignment Search Tool", Journal of Molecular Biology 215:403-10, "BLAST" algorithm, see https: / / blast.ncbi.nlm.nih.gov / Blast.cgi. When using any of the aforementioned algorithms, the default parameters (for window length, gap penalty, etc.) are used. In one embodiment, sequence identity is determined using the BLAST algorithm using the default parameters.

[0216] The antibodies described herein are generally isolated or recombinant. When used to describe the various polypeptides disclosed herein, "isolated" means a polypeptide that has been identified and separated and / or recovered from a cell or cell culture expressing the polypeptide. Typically, an isolated polypeptide will be prepared by at least one purification step. An "isolated antibody" refers to an antibody that is substantially free of other antibodies having different antigenic specificities. "Recombinant" means that the antibody is produced using recombinant nucleic acid technology in an exogenous host cell, and the antibody may also be isolated.

[0217] "Specific binding" or "specifically binds to" or "specific for" a particular antigen or epitope means binding that is measurably different from non-specific interactions. Specific binding can be measured, for example, by determining the binding of a molecule compared to the binding of a control molecule, which is typically a molecule of similar structure that has no binding activity. For example, specific binding can be determined by competition with a control molecule that is similar to the target.

[0218] Specific binding for a particular antigen or epitope can be exhibited, for example, by an antibody having a KD for the antigen or epitope of at least about 10 -4 M, at least about 10 -5 M, at least about 10 -6 M, at least about 10 -7 M, at least about 10 -8 M, at least about 10 -9 M, or at least about 10 -10 M, at least about 10 -11 M, at least about 10 -12 M or greater, where KD refers to the off-rate of a specific antibody-antigen interaction. Typically, the KD of an antibody that specifically binds to an antigen is 20-fold, 50-fold, 100-fold, 500-fold, 1000-fold, 5,000-fold, 10,000-fold, or more greater than the KD of a control molecule relative to the antigen or epitope.

[0219] Furthermore, specific binding for a particular antigen or epitope can be demonstrated, for example, by an antibody having a KA or Ka against the antigen or epitope that is at least 20-fold, 50-fold, 100-fold, 500-fold, 1000-fold, 5,000-fold, 10,000-fold, or more greater than that against the epitope relative to a control, where KA or Ka refers to the on-rate of the specific antibody-antigen interaction. Binding affinity is typically measured using Biacore, SPR, or BLI assays.

[0220] IV. ENPP3 Binding Domain

[0221] In one aspect, provided herein are ENPP3 antigen-binding domains (ABDs) and compositions comprising such ENPP3 antigen-binding domains (ABDs), including anti-ENPP3 antibodies. The subject antibodies comprising such ENPP3 antigen-binding domains (e.g., anti-ENPP3 x anti-CD3 bispecific antibodies) advantageously elicit a range of different immune responses (see Examples 5 and 6). Such ENPP3 binding domains and related antibodies can be used, for example, to treat ENPP3-associated cancers.

[0222] As will be appreciated by those skilled in the art, suitable ENPP3 binding domains may include those described in the sequence listing and Figure 12 、 13A to 13B ,and 14A to 14I A set of 6 CDRs as depicted in or as if the CDRs are underlined, or where a different numbering scheme is used as described herein and as shown in Table 2, is used. Figure 12 、 13A to 13B ,and 14A to 14I Other aligned CDRs identified within the variable heavy (VH) chain domain and variable light domain (VL) sequences of those depicted in the sequence listing (see Table 2). Suitable ENPP3 ABDs may also comprise the entire VH and VL sequences for use as scFv or Fab domains as depicted in these sequences and the accompanying figures.

[0223] In one embodiment, the ENPP3 antigen binding domain comprises the six CDRs (ie, vhCDR1-3 and vlCDR1-3) of the ENPP3 ABD described herein, including the Figures and Sequence Listing. In exemplary embodiments, the ENPP3 ABD is one of the following ENPP3 ABDs: AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19 and Ha16-1.80 ( Figure 12 、 13A to 13B and 14A to 14I ).

[0224] In addition to the parent CDR sets that form the ABD for ENPP3 disclosed in the Figures and Sequence Listing, ENPP3 ABD variants having CDRs that include at least one modification of the ENPP3 ABD CDRs disclosed herein are also provided herein. In one embodiment, the ENPP3 ABD comprises a set of six CDRs having one, two, three, four, five, six, seven, eight, nine, or ten amino acid modifications compared to the six CDRs of the ENPP3 ABD depicted herein, including in the Figures and Sequence Listing. In exemplary embodiments, the ENPP3 ABD comprises a set of 6 CDRs having 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid modifications compared to the 6 CDRs of one of the following ENPP3 ABDs: AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, H16-9.44, H16-1.67, Ha16-1.23, Ha16-1.67, Ha16-1.76. (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19 and Ha16-1.80 ( Figure 12 、 13A to 13B and 14A to 14I In certain embodiments, the variant ENPP3 ABD is capable of binding to an ENPP3 antigen as determined by at least one of a Biacore, surface plasmon resonance (SPR), and / or BLI (bio-layer interferometry, such as an octet assay), the latter assay being particularly useful in many embodiments. In a specific embodiment, the ENPP3 ABD is capable of binding to a human ENPP3 antigen (see Example 5).

[0225] In one embodiment, the ENPP3 ABD comprises six CDRs that are at least 90%, 95%, 97%, 98% or 99% identical to the six CDRs of the ENPP3 ABD as described herein, including the Figures and Sequence Listing. In exemplary embodiments, the ENPP3 ABD comprises six CDRs that are at least 90%, 95%, 97%, 98%, or 99% identical to the six CDRs of one of the following ENPP3 ABDs: AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, H16-9.44, H16-1.67, Ha16-1.23, Ha16-1.69, Ha16-1.76. (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19 and Ha16-1.80 ( Figure 12 、 13A to 13B and 14A to 14I In certain embodiments, the ENPP3 ABD is capable of binding to TIM-3 as determined by at least one of a Biacore, surface plasmon resonance (SPR) and / or BLI (biolayer interferometry, e.g., Octet assay) assay, the latter assay being particularly useful in many embodiments. In a specific embodiment, the ENPP3 ABD is capable of binding to a human ENPP3 antigen (see Figure 2 ).

[0226] In another exemplary embodiment, the ENPP3 ABD comprises the variable heavy (VH) domain and the variable light (VL) domain of any one of the ENPP3 ABDs described herein, including in the Figures and Sequence Listing. In exemplary embodiments, the ENPP3 ABD is one of the following ENPP3 ABDs: AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19 and Ha16-1.80 ( Figure 12 、 13A to 13B and 14A to 14I ).

[0227] In addition to the parent ENPP3 variable heavy and variable light domains disclosed herein, provided herein are ENPP3 ABDs comprising a variable heavy domain and / or a variable light domain that are variants of the ENPP3 ABD VH and VL domains disclosed herein. In one embodiment, the variant VH and / or VL domains have 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid changes compared to the VH and / or VL domains of the ENPP3 ABDs described herein, including the figures and sequence listing. In exemplary embodiments, the variant VH domain and / or VL domain has 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid changes compared to the VH and / or VL domain of one of the following ENPP3 ABDs: AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, H16-9.44, H16-1.67, Ha16-1(2). (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19 and Ha16-1.80 ( Figure 12 、 13A to 13B and 14A to 14I In certain embodiments, the ENPP3 ABD is capable of binding to the ENPP3 as determined by at least one of a Biacore, surface plasmon resonance (SPR), and / or BLI (biolayer interferometry, such as an octet assay), the latter assay being particularly useful in many embodiments. In a specific embodiment, the ENPP3 ABD is capable of binding to a human ENPP3 antigen (see Example 5).

[0228] In one embodiment, the variant VH and / or VL domains are at least 90%, 95%, 97%, 98% or 99% identical to the VH and / or VL of an ENPP3 ABD as described herein, including the Figures and Sequence Listing. In exemplary embodiments, the variant VH and / or VL domains are at least 90%, 95%, 97%, 98% or 99% identical to the VH and / or VL of one of the following ENPP3 ABDs: AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1(1)23. (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19 and Ha16-1.80 ( Figure 12 、 13A to 13B and 14A to 14I In certain embodiments, the ENPP3 ABD is capable of binding to the ENPP3 as determined by at least one of a Biacore, surface plasmon resonance (SPR), and / or BLI (biolayer interferometry, such as an octet assay), the latter assay being particularly useful in many embodiments. In a specific embodiment, the ENPP3 ABD is capable of binding to a human ENPP3 antigen (see Example 5).

[0229] V. Antibodies

[0230] In one aspect, provided herein are antibodies that bind to ENPP3 (e.g., anti-ENPP3 antibodies). In certain embodiments, the antibodies bind to human ENPP3 ( Figure 11A ). The subject anti-ENPP3 antibodies include monospecific ENPP3 antibodies, as well as multispecific (e.g., bispecific) anti-ENPP3 antibodies. In certain embodiments, the anti-ENPP3 antibodies have Figure 15A 、 Figure 15B and Figures 52A to 52K The invention can be in the form of any of the antibody formats described in .

[0231] In some embodiments, the subject compositions comprise an ENPP3 binding domain. In some embodiments, the compositions comprise an antibody having an ENPP3 binding domain. The antibodies provided herein comprise one, two, three, four, and five or more ENPP3 binding domains. In certain embodiments, the ENPP3 binding domain comprises a Figure 12 、 13A to 13B and 14A to 14I In some embodiments, the ENPP3 binding domain comprises any of the vhCDR1, vhCDR2, vhCDR3, vlCDR1, vlCDR2, and vlCDR3 sequences of the ENPP3 binding domains depicted in Figure 12 、 13A to 13B and 14A to 14I The underlined vhCDR1, vhCDR2, vhCDR3, vlCDR1, vlCDR2, and vlCDR3 sequences of the ENPP3 binding domains are those depicted in FIG. In some embodiments, the ENPP3 binding domain comprises a Figure 12 、 13A to 13B and 14A to 14I The variable heavy chain domain and variable light domain of the ENPP3 binding domain are as described in those depicted in FIG. Figure 12 、 13A to 13B and 14A to 14I The ENPP3 binding domains described in the study include AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, and Ha16-1. 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80.

[0232] In one aspect, provided herein are bispecific antibodies that bind ENPP3 and CD3 in various formats as described below and are generally depicted in Figure 15A and Figure 15B These bispecific heterodimeric antibodies include an ENPP3 binding domain. In certain embodiments, the ENPP3 binding domain includes a moiety selected from Figure 12 、 13A to 13B and 14A to 14I In some embodiments, the ENPP3 binding domain comprises a VHCDR1, VHCDR2, VHCDR3, VLCDR1, VLCDR2 and VLCDR3 sequence selected from the group consisting of those described in Figure 12 、 13A to 13B and 14A to 14I Underlined VHCDR1, VHCDR2, VHCDR3, VLCDR1, VLCDR2 and VLCDR3 sequences of the ENPP3 binding domains are those depicted in FIG.

[0233] These bispecific heterodimeric antibodies bind to ENPP3 and CD3. Such antibodies comprise a CD3 binding domain and at least one ENPP3 binding domain. Any suitable ENPP3 binding domain can be included in the anti-ENPP3 x anti-CD3 bispecific antibody. In some embodiments, the anti-ENPP3 x anti-CD3 bispecific antibody comprises one, two, three, four, or more ENPP3 binding domains, including but not limited to Figure 12 、 13A to 13B and 14A to 14I In certain embodiments, the anti-ENPP3 x anti-CD3 antibody comprises an ENPP3 binding domain comprising a moiety selected from Figure 12 、 13A to 13B and 14A to 14I In some embodiments, the anti-ENPP3 x anti-CD3 antibody comprises an ENPP3 binding domain comprising a VHCDR1, VHCDR2, VHCDR3, VLCDR1, VLCDR2, and VLCDR3 sequence selected from the group consisting of those described in Figure 12 、 13A to 13B and 14A to 14I In some embodiments, the anti-ENPP3 x anti-CD3 antibody comprises an ENPP3 binding domain comprising a cleaved ENPP3 binding domain selected from the group consisting of: Figure 12、 13A to 13B and 14A to 14I In an exemplary embodiment, the anti-ENPP3 x anti-CD3 antibody comprises an anti-ENPP3AN1[ENPP3]_H1L1 binding domain.

[0234] The anti-ENPP3 x anti-CD3 antibodies provided herein can comprise any suitable CD3 binding domain. In certain embodiments, the anti-ENPP3 x anti-CD3 antibodies comprise a CD3 binding domain comprising a protein selected from the group consisting of 10A to 10F In some embodiments, the anti-ENPP3 x anti-CD3 antibody comprises a CD3 binding domain comprising a VHCDR1, VHCDR2, VHCDR3, VLCDR1, VLCDR2, and VLCDR3 sequence selected from the group consisting of those described in 10A to 10F In some embodiments, the anti-ENPP3 x anti-CD3 antibody comprises a CD3 binding domain comprising a CD3 binding domain selected from the group consisting of 10A to 10F In some embodiments, the CD3 binding domain is selected from the group consisting of the CD3 binding domains depicted in . In some embodiments, the CD3 binding domain is selected from anti-CD3 H1.30_L1.47, anti-CD3 H1.32_L1.47; anti-CD3 H1.89_L1.48; anti-CD3 H1.90_L1.47; anti-CD3 H1.33_L1.47; and anti-CD3 H1.31_L1.47. As described herein, these anti-CD3 antigen binding domains (CD3-ABD) can be used in scFv format in any orientation (e.g., VH-scFv linker-VL or VL-scFv linker-VH from N-terminus to C-terminus).

[0235] The antibodies provided herein comprise different antibody domains. As described herein and known in the art, the antibodies described herein comprise different domains within the heavy and light chains, which domains may also overlap. These domains include, but are not limited to, an Fc domain, a CH1 domain, a CH2 domain, a CH3 domain, a hinge domain, a heavy chain constant domain (CH1-hinge-Fc domain or CH1-hinge-CH2-CH3), a variable heavy chain domain, a variable light chain domain, a light chain constant domain, a FAb domain, and a scFv domain.

[0236] As indicated herein, there are a number of suitable linkers (for use as domain linkers or scFv linkers) that can be used to covalently link the enumerated domains (e.g., scFv, Fab, Fc domain, etc.), including traditional peptide bonds produced by recombinant techniques. Figure 6 Exemplary linkers for attaching the domains of the subject antibodies to each other are depicted in . In some embodiments, the linker peptide may primarily comprise the following amino acid residues: Gly, Ser, Ala, or Thr. The linker peptide should have a length sufficient to connect the two molecules so that they assume the correct conformation relative to each other so that they retain the desired activity. In one embodiment, the linker has a length of about 1-50 amino acids, preferably a length of about 1-30 amino acids. In one embodiment, a linker with a length of 1-20 amino acids can be used, and in some embodiments, about 5 to about 10 amino acids can be used. Useful linkers include glycine-serine polymers (including, for example, (GS)n, (GSGGS)n, (GGGGS)n, and (GGGS)n, where n is an integer of at least one (and typically 3 to 4)), glycine-alanine polymers, alanine-serine polymers, and other flexible linkers, some of which are in Figure 5 and Figure 6 Alternatively, a variety of non-protein polymers including, but not limited to, polyethylene glycol (PEG), polypropylene glycol, polyalkylene oxide, or copolymers of polyethylene glycol and polypropylene glycol may be used as linkers.

[0237] Other linker sequences can include any sequence of any length of the CL / CH1 domain, but not including all residues of the CL / CH1 domain; for example, the first 5-12 amino acid residues of the CL / CH1 domain. The linker can be derived from an immunoglobulin light chain, such as Cκ or Cλ. The linker can be derived from an immunoglobulin heavy chain of any isotype, including, for example, Cγ1, Cγ2, Cγ3, Cγ4, Cα1, Cα2, Cδ, Cε, and Cμ. The linker sequence can also be derived from other proteins, such as Ig-like proteins (e.g., TCR, FcR, KIR), hinge region-derived sequences, and other native sequences from other proteins.

[0238] In some embodiments, the linker is a "domain linker," which is used to connect any two domains as outlined herein. For example, in Figure 15BIn the embodiment of the present invention, there may be a domain linker connecting the C-terminus of the CH1 domain of the Fab to the N-terminus of the scFv, wherein another optional domain linker connects the C-terminus of the scFv to the CH2 domain (although in many embodiments, the hinge serves as this domain linker). Although any suitable linker can be used, many embodiments utilize glycine-serine polymers as domain linkers, including, for example, (GS)n, (GSGGS)n, (GGGGS)n, and (GGGS)n, where n is an integer of at least one (and typically 3 to 4 to 5), and any peptide sequence that allows for recombinant connection of the two domains and has sufficient length and flexibility so that each domain retains its biological function. In some cases, and with attention to the "chain type", as outlined below, charged domain linkers such as those used in some embodiments of scFv linkers can be used. Figure 6 Exemplary useful domain linkers are depicted in .

[0239] With particular reference to domain linkers used to attach scFv domains to Fc domains in a "2+1" format, there are several domain linkers with particular utility, including, for example, Figure 6 The "full hinge C220S variant," "flexible half hinge," "charged half hinge 1," and "charged half hinge 2" are shown.

[0240] In some embodiments, the linker is a "scFv linker," which is used to covalently attach the VH and VL domains as discussed herein. In many cases, the scFv linker is a charged scFv linker, a plurality of which are shown in FIG. Figure 5 In. Therefore, in some embodiments, the antibodies described herein also provide charged scFv joints, to promote the pI separation between the first monomer and the second monomer. That is, by incorporating positive or negative (or both, when using the main chain of scFv on different monomers) electrically charged scFv joints, the monomer comprising a charged joint is allowed to change pI without producing further changes in the Fc domain. These charged joints can be substituted into any scFv containing a standard joint. Again, as will be appreciated by those skilled in the art, according to the desired pI changes, a charged scFv joint is used on the correct " chain " or monomer. For example, as discussed herein, in order to form a heterodimeric antibody in the form of 1+1 Fab-scFv-Fc, the original pI of the Fv region of each desired antigen-binding domain is calculated, and one is selected to form scFv, and according to pI, a positive or negative joint is selected.

[0241] Charged domain linkers can also be used to increase the pi separation of monomers of the antibodies described herein, and thus Figure 5 Those contained within can be used herein in any embodiment utilizing a linker.

[0242] Specifically, Figure 15A and 15B The format depicted is an antibody, often referred to as a "heterodimeric antibody," meaning a protein having at least two related Fc sequences that self-assemble into a heterodimeric Fc domain and at least two Fv regions, either as a Fab or as a scFv.

[0243] The provided ENPP3 binding domains can be included in any useful antibody format, including, for example, canonical immunoglobulins, as well as the 1+1 Fab-scFv-Fc and 2+1 Fab2-scFv-Fv formats provided herein. Other useful antibody formats include, but are not limited to, "mAb-Fv," "mAb-scFv," "central-Fv," "single-armed scFv-mAb," "scFv-mAb," "dual scFv," and "trident" formats of antibodies, such as Figures 52A to 52K In the public.

[0244] In some embodiments, the subject antibodies comprise one or more of the ENPP3 ABDs provided herein. In some embodiments, the antibodies comprise one ENPP3 ABD. In other embodiments, the antibodies comprise two ENPP3 ABDs. In exemplary embodiments, the ENPP3 ABD comprises the variable heavy chain domain and the variable light chain domain of one of the following ENPP3 ABDs: AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19 and Ha16-1.80 ( Figure 12 、 13A to 13B and 14A to 14I). In some embodiments, the ENPO3 ABD is one of the following ENPP3 ABDs: AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1 (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha16-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19 and Ha16-1.80 ( Figure 12 、 13A to 13B and 14A to 14I ).

[0245] In exemplary embodiments, the antibody is a bispecific antibody comprising one or two ENPP3 ABDs, including any of the ENPP3 ABDs provided herein. Bispecific antibodies comprising such ENPP3 ABDs include, for example, 1+1 Fab-scFv-Fc and 2+1 Fab2-scFv-Fc bispecific antibody formats. In exemplary embodiments, the ENPP3 ABD is one of the following B7H3 ABDs: AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19 and Ha16-1.80 ( Figure 12 、 13A to 13B and 14A to 14I In exemplary embodiments, the ENPP3 binding domain is a Fab. In some embodiments, such bispecific antibodies are heterodimeric bispecific antibodies comprising any heterodimerization skew variant, pI variant, and / or ablation variant described herein.

[0246] A. Chimeric and Humanized Antibodies

[0247] In certain embodiments, the antibodies described herein comprise a heavy chain variable region from a specific germline heavy chain immunoglobulin gene and / or a light chain variable region from a specific germline light chain immunoglobulin gene. For example, such antibodies may comprise or consist of human antibodies comprising a heavy chain or light chain variable region that is "the product of" or "derived from" a specific germline sequence. Human antibodies that are "the product of" or "derived from" a human germline immunoglobulin sequence can be identified by comparing the amino acid sequence of the human antibody to the amino acid sequence of a human germline immunoglobulin and selecting the human germline immunoglobulin sequence that is closest in sequence (i.e., greatest % identity) to the sequence of the human antibody (using the methods outlined herein). Human antibodies that are "the product of" or "derived from" a specific human germline immunoglobulin sequence may contain amino acid differences compared to the germline sequence due to, for example, naturally occurring somatic mutations or intentionally introduced site-directed mutations. In some cases, humanized antibodies can be at least 95%, 96%, 97%, 98% or 99% identical to the amino acid sequence encoded by the germline immunoglobulin genes, or even at least 96%, 97%, 98% or 99% identical to the amino acid sequence encoded by the germline immunoglobulin genes. Typically, humanized antibodies derived from specific human germline sequences will show no more than 10-20 amino acid differences (before introducing any skew variants, pI variants and ablation variants herein) with the amino acid sequence encoded by the germline immunoglobulin genes; That is, before introducing variants described herein, the number of variants is typically lower. In certain cases, a humanized antibody may display no more than 5, or even no more than 4, 3, 2, or 1 amino acid difference from the amino acid sequence encoded by the germline immunoglobulin gene (again, before the introduction of any skew, pi, and ablation variants herein; that is, the number of variants is generally low before the introduction of the variants described herein).

[0248] In one embodiment, the parent antibody has been affinity matured, as is known in the art. Humanization and affinity maturation can be performed using structure-based methods, for example as described in USSN 11 / 004,590. Selection-based methods can be used to humanize and / or affinity mature antibody variable regions, including but not limited to the methods described in Wu et al., 1999, J. Mol. Biol. 294: 151-162; Baca et al., 1997, J. Biol. Chem. 272(16): 10678-10684; Rosok et al., 1996, J. Biol. Chem. 271(37): 22611-22618; Rader et al., 1998, PNAS 95: 8910-8915; Krauss et al., 2003, Protein Engineering 16(10): 753-759, which are incorporated herein by reference in their entirety. Other humanization approaches may involve grafting only portions of the CDRs, including but not limited to those described in USSN 09 / 810,510; Tan et al., 2002, J. Immunol. 169:1119-1125; De Pascalis et al., 2002, J. Immunol. 169:3076-3084, all of which are incorporated herein by reference in their entirety.

[0249] B. Heterodimeric Antibodies

[0250] In exemplary embodiments, the bispecific antibodies provided herein are heterodimeric bispecific antibodies comprising two variant Fc domain sequences. Such variant Fc domains comprise amino acid modifications to facilitate self-assembly and / or purification of the heterodimeric antibody.

[0251] One ongoing issue in antibody technology is the desire for "bispecific" antibodies to bind to two different antigens simultaneously, often allowing access to the different antigens and thereby generating new functions and therapies. Typically, these antibodies are produced by incorporating genes for each heavy and light chain into host cells. This typically results in the formation of the desired heterodimer (AB) as well as two homodimers, AA and BB (which do not involve the light chain heterodimer issue). However, the main obstacle to the formation of bispecific antibodies is the difficulty in biasing the formation of the desired heterodimeric antibody towards the formation of the homodimer and / or the difficulty in purifying the heterodimeric antibody from the homodimeric antibody.

[0252] There are multiple mechanisms that can be used to produce heterodimeric antibodies of the present invention. In addition, as will be appreciated by those skilled in the art, these different mechanisms can be combined to ensure high heterodimerization. The amino acid modifications that promote the production and purification of heterodimers are generally referred to as "heterodimerization variants." As discussed below, heterodimerization variants include "skewed" variants (e.g., "knob and hole" and "charge pair" variants described below) and "pI variants," which allow purification of heterodimers from homodimers. As generally described in U.S. Pat. No. 9,605,084 (the entire contents of which are incorporated herein by reference), and particularly as discussed below for heterodimerization variants, useful mechanisms of heterodimerization include the "knob and mortar" ("KIH") described in U.S. Pat. No. 9,605,084, the "electrostatic manipulation" or "charge pair" described in U.S. Pat. No. 9,605,084, the pi variants described in U.S. Pat. No. 9,605,084, and generally other Fc variants summarized in U.S. Pat. No. 9,605,084, and below.

[0253] Heterodimerization variants that can be used to form and purify heterodimeric antibodies (eg, bispecific antibodies) of the invention are discussed in further detail below.

[0254] 1. Skew variant

[0255] In some embodiments, the heterodimeric antibody comprises a skew variant, which is one or more amino acid modifications in the first Fc domain (A) and / or the second Fc domain (B) that favor the formation of Fc heterodimers (Fc dimers comprising the first Fc domain and the second Fc domain); (AB) over Fc homodimers (Fc dimers comprising two of the first Fc domain or two of the second Fc domain; AA or BB). Suitable skew variants are included in Figure 29 of U.S. Published Application No. 2016 / 0355608, which is hereby incorporated by reference in its entirety and specifically for its disclosure of skew variants, and in Figures 1A to 1E and Figure 4 middle.

[0256] One mechanism is often referred to in the art as "knob and hole," and so-called amino acid engineering can optionally be used to create steric influences that favor heterodimer formation over homodimer formation; this is sometimes referred to as "knob and hole," as described in: USSN 61 / 596,846; Ridgway et al., Protein Eng. 9(7):617 (1996); Atwell et al., J. Mol. Biol. 1997 270:26; U.S. Pat. No. 8,216,805, all of which are incorporated herein by reference in their entirety. The figure shows the number of "monomer A-monomer B" pairs that depend on the "knob and hole." Additionally, as described in Merchant et al., Nature Biotech. 16:677 (1998), these "knob and hole" mutations can be combined with disulfide bonds to skew heterodimerization.

[0257] Another mechanism for producing heterodimeric antibodies is sometimes referred to as "electrostatic steering," as described in Gunasekaran et al., J. Biol. Chem. 285(25): 19637 (2010), which is incorporated herein by reference in its entirety. This is sometimes referred to herein as a "charge pair." In this embodiment, electrostatics are used to skew the formation toward heterodimerization. As will be appreciated by those skilled in the art, these charge pairs may also affect pI and therefore purification, and therefore may also be considered pI variants in some cases. However, since these are generated to force heterodimerization and are not used as purification tools, they are classified as "steric variants." These include, but are not limited to, D221E / P228E / L368E paired with D221R / P228R / K409R (eg, these are the "monomer counterparts") and C220E / P228E / 368E paired with C220R / E224R / P228R / K409R.

[0258] In some embodiments, the skew variants advantageously and simultaneously favor heterodimerization based on the "knob and mortar" mechanism and the "electrostatic manipulation" mechanism. In some embodiments, the heterodimeric antibody comprises one or more sets of such heterodimerization skew variants. These variants appear as "sets" of "pairs." That is, one set of pairs is incorporated into a first monomer and another set of pairs is incorporated into a second monomer. It should be noted that these sets do not necessarily behave as "knob-into-hole" variants, where there is a one-to-one correspondence between residues on one monomer and residues on another monomer. That is, these paired sets can instead form an interface between the two monomers that encourages heterodimer formation and discourages homodimer formation, thereby allowing the percentage of heterodimers that spontaneously form under biological conditions to exceed 90%, rather than the expected 50% (25% homodimer A / A: 50% heterodimer A / B: 25% homodimer B / B). Figure 4 Exemplary heterodimerization "skewed" variants are depicted in . In exemplary embodiments, the heterodimeric antibody comprises S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L; K370S:S364K / E357Q; or T366S / L368A / Y407V:T366W (optionally including a bridging disulfide bond, T366S / L368A / Y407V / Y349C:T366W / S354C) a "skewed" variant amino acid substitution set. In an exemplary embodiment, the heterodimeric antibody comprises the "S364K / E357Q:L368D / K370S" amino acid substitution set. For nomenclature purposes, the pairing "S364K / E357Q:L368D / K370S" means that one monomer comprises an Fc domain comprising amino acid substitutions S364K and E357Q, and the other monomer comprises an Fc domain comprising amino acid substitutions L368D and K370S; as described above, the "strandedness" of these pairings depends on the starting pI.

[0259] In some embodiments, the skew variants provided herein can be optionally and independently incorporated into one or both of the first and second Fc domains of a heterodimeric antibody along with any other modifications, including but not limited to other skew variants (see, e.g., FIG. 37 of U.S. Published Application No. 2012 / 0149876, which is incorporated herein by reference, particularly for its disclosure of skew variants), pI variants, isotype variants, FcRn variants, ablation variants, etc. In addition, a single modification can also be independently and optionally included in or excluded from a subject heterodimeric antibody.

[0260] Additional monomer A and monomer B variants can optionally and independently be combined in any amount with other variants (such as the pi variants outlined herein or other steric variants shown in Figure 37 of US 2012 / 0149876, the figures and legends and SEQ ID NOs of which are hereby incorporated by reference).

[0261] In some embodiments, the steric variants outlined herein can be optionally and independently incorporated into one or two monomers with any pi variant (or other variant, such as an Fc variant, an FcRn variant, etc.), and can be independently and optionally included in or excluded from the proteins of the antibodies described herein.

[0262] A list of suitable skew variants can be found in Figures 1A to 1E Among them Figure 4 Specific uses of some pairs in many embodiments are shown. In various embodiments, pairs including, but not limited to, the following groups are particularly useful: S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, and K370S:S364K / E357Q. In terms of nomenclature, the pair "S364K / E357Q:L368D / K370S" means that one of the monomers has the double variant group S364K / E357Q and the other has the double variant group L368D / K370S.

[0263] 2. pI (isoelectric point) variants for heterodimers

[0264] In some embodiments, the heterodimeric antibody comprises a purification variant that advantageously allows separation of the heterodimeric antibody (eg, an anti-ENPP3 x anti-CD3 bispecific antibody) from the homodimeric protein.

[0265] There are several basic mechanisms that can lead to easy purification of heterodimeric antibodies. For example, modification of one or both of the antibody heavy chain monomers A and B so that each monomer has a different pI allows electropurification of heterodimeric AB antibodies from monomers AA and BB proteins. Alternatively, some scaffold formats, such as "1+1 Fab-scFv-Fc" and "2+1 Fab2-scFv-Fc" formats, also allow separation based on size. As described herein, it is also possible to use skew variants that "skew" the formation of heterodimers relative to homodimers. Therefore, the combination of heterodimerization skew variants and pI variants is particularly useful in the heterodimeric antibodies provided herein.

[0266] In addition, as more fully outlined below, depending on the format of the heterodimeric antibody, pi variants and / or domain linkers contained within the constant region and / or Fc domain of the monomer can be used. In some embodiments, the heterodimeric antibody comprises additional modifications for alternative functions that can also produce pi changes, such as Fc, FcRn and KO variants.

[0267] In some embodiments, the heterodimeric antibodies of the invention provided herein comprise at least one monomer having one or more modifications that alter the pI of the monomer (i.e., a "pI variant"). Generally, as will be appreciated by those skilled in the art, pI variants exist in two general categories: variants that increase the pI of a protein (basic changes) and variants that decrease the pI of a protein (acidic changes). As described herein, all combinations of these variants can be made: one monomer can be wild-type, or a variant that does not exhibit a pI significantly different from wild-type, and the other can be more basic or more acidic. Alternatively, each monomer can be altered so that one is more basic and one is more acidic.

[0268] Depending on the format of the heterodimeric antibody, pI variants can be included within the constant and / or Fc domains of the monomer, or charged linkers, domain linkers, or scFv linkers can be used. That is, antibody formats utilizing one or more scFvs, such as a "1 + 1 Fab-scFv-Fc," can include a charged scFv linker (positively or negatively charged) to provide an additional pI boost for purification purposes. As will be appreciated by those skilled in the art, some 1 + 1 Fab-scFv-Fc formats can be used with only a charged scFv linker without additional pI adjustment, but the antibodies described herein do also provide pI variants and / or charged domain linkers on one or both of the monomers. Additionally, additional amino acid engineering for alternative functions can also confer pI variation, such as in Fc, FcRn, and KO variants.

[0269] In heterodimeric antibodies of the present invention that utilize pI as a separation mechanism to allow purification of heterodimeric proteins, amino acid variants are introduced into one or both of the monomeric polypeptides. That is, the pI of one of the monomers (referred to herein as "monomer A" for simplicity) can be engineered to be away from monomer B, or both monomer A and monomer B can be altered as the pI of monomer A increases and the pI of monomer B decreases. As outlined more fully below, changes in the pI of either or both monomers can be accomplished by removing or adding charged residues (e.g., replacing neutral amino acids with positively or negatively charged amino acid residues, e.g., glycine to glutamic acid), changing a charged residue from positively or negatively charged to the oppositely charged residue (e.g., aspartic acid to lysine), or changing a charged residue to a neutral residue (e.g., loss of charge; lysine to serine). Figure 3 and Figure 4 A variety of these variations are shown in .

[0270] Thus, in some embodiments, the heterodimeric antibodies of the invention comprise amino acid modifications in the constant region that alter the isoelectric point (pI) of at least one (if not both) of the monomers of the dimeric protein by incorporating amino acid substitutions ("pI variants" or "pI substitutions") into one or both of the monomers to form a "pI antibody." As shown herein, separation of the heterodimer from the two homodimers can be achieved if the pIs of the two monomers differ by as little as 0.1 pH units, with values ​​of 0.2, 0.3, 0.4, and 0.5 or higher being useful in the antibodies described herein.

[0271] As will be appreciated by those skilled in the art, the number of pi variants to be included on each or both monomers to achieve good separation will depend in part on the starting pis of the components, e.g., the starting pis of the scFv and Fab of interest in the 1+1 Fab-scFv-Fc and 2+1 Fab2-scFv-Fc formats. That is, to determine which monomer to engineer or in which "direction" (e.g., more positive or more negative), the Fv sequences of the two target antigens are calculated and a decision is made accordingly. As is known in the art, different Fvs will have different starting pis for use in the antibodies described herein. Typically, as outlined herein, the pis are engineered so that the overall pi difference of the individual monomers is at least about 0.1 log, preferably 0.2 to 0.5 as outlined herein.

[0272] In cases where pi variants are used to achieve heterodimerization, a more modular approach is provided for designing and purifying bispecific proteins (including antibodies) by using one or more constant regions of one or more heavy chains. Thus, in some embodiments, heterodimerization variants (including skewed and pi heterodimerization variants) are not included in the variable region, necessitating the engineering of each individual antibody. Additionally, in some embodiments, the likelihood of pi variants causing immunogenicity is significantly reduced by introducing pi variants from different IgG isotypes to alter the pi without introducing significant immunogenicity. Therefore, another problem to be addressed is to elucidate low pi constant domains with high human sequence content, for example, minimizing or avoiding non-human residues at any particular position. Alternatively, or in addition to isotype substitutions, the likelihood of pi variants causing immunogenicity is significantly reduced by utilizing homozygous substitutions (e.g., Asn to Asp; and Gln to Glu).

[0273] As discussed below, the negative effects that may occur with this pI engineering are also increased serum half-life and increased FcRn binding. That is, as described in U.S. Published Application No. US 2012 / 0028304 (incorporated by reference in its entirety), reducing the pI of antibody constant domains (those found in antibody and Fc fusion) may cause serum to remain in the body for a longer period of time. These pI variants with increased serum half-life also promote pI changes for purification.

[0274] Additionally, it should be noted that pi variants provide additional benefits to the analysis and quality control processes of bispecific antibodies, as the ability to eliminate, minimize, and differentiate when homodimers are present is significant. Similarly, the ability to reliably test the reproducibility of heterodimeric antibody production is important.

[0275] Generally, embodiments of particular uses rely on a variant panel comprising skew variants that promote heterodimerization compared to homodimerization in combination with pi variants that increase the pi difference between the two monomers to facilitate purification of heterodimers from homodimers.

[0276] Exemplary combinations of pi variants are shown in U.S. Published Application No. 2016 / 0355608. Figure 4 and Figure 5 as well as Figure 30 All of the aforementioned documents are hereby incorporated by reference in their entirety, particularly with respect to their disclosure of pi variants. Figure 2 Preferred combinations of pI variants are shown in FIG. As outlined herein and shown in the figures, these changes are shown relative to IgG1, but all isotypes can be altered in this manner, as well as isotype hybrids. In the case where the heavy chain constant domain is from IgG2-4, R133E and R133Q can also be used.

[0277] In one embodiment, a preferred combination of pI variants comprises one monomer (negative Fab side) comprising the 208D / 295E / 384D / 418E / 421D variant (N208D / Q295E / N384D / Q418E / N421D relative to human IgG1) and a second monomer (positive scFv side) comprising a positively charged scFv linker comprising (GKPGS)4 (SEQ ID NO: XX). However, as will be appreciated by those skilled in the art, the first monomer comprises the CH1 domain, which includes position 208. Therefore, in constructs that do not comprise a CH1 domain (e.g., for antibodies that do not utilize a CH1 domain as one of the domains), a preferred negative pI variant Fc panel comprises the 295E / 384D / 418E / 421D variant (Q295E / N384D / Q418E / N421D relative to human IgG1).

[0278] Thus, in some embodiments, a monomer has a Figure 2 and the other monomer has a charged linker (in the form of a charged scFv linker, since the monomer comprises an scFv as indicated by the format or a charged domain linker, which may be selected from Figure 5 those depicted in ).

[0279] In some embodiments, modifications are made in the hinge of the Fc domain, including positions 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, and 230 based on EU numbering. Thus, pi mutations and particularly substitutions can be made in one or more of positions 216-230, with 1, 2, 3, 4, or 5 mutations being useful. Again, all possible combinations, alone or with other pi variants in other domains, are contemplated.

[0280] Specific substitutions that can be used to lower the pi of the hinge domain include, but are not limited to, a deletion at position 221, a non-natural valine or threonine at position 222, a deletion at position 223, a non-natural glutamate at position 224, a deletion at position 225, a deletion at position 235, and a deletion or non-natural alanine at position 236. In some cases, pi substitutions are made only in the hinge domain, and in other cases, these substitutions are added in any combination to other pi variants in other domains.

[0281] In some embodiments, mutations can be made in the CH2 region, including positions 233, 234, 235, 236, 274, 296, 300, 309, 320, 322, 326, 327, 334, and 339 based on EU numbering. It should be noted that changes from 233 to 236 can be made to increase effector function in the IgG2 backbone (as well as 327A). Again, all possible combinations of these 14 positions can be made; for example, variant Fc domains with 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 CH2 pI substitutions can be included.

[0282] Specific substitutions that can be used to lower the pI of the CH2 domain include, but are not limited to, a non-natural glutamine or glutamic acid at position 274, a non-natural phenylalanine at position 296, a non-natural phenylalanine at position 300, a non-natural valine at position 309, a non-natural glutamic acid at position 320, a non-natural glutamic acid at position 322, a non-natural glutamic acid at position 326, a non-natural glycine at position 327, a non-natural glutamic acid at position 334, a non-natural threonine at position 339, and all possible combinations within CH2 and with other domains.

[0283] In this embodiment, the modification can be independently and optionally selected from positions 355, 359, 362, 384, 389, 392, 397, 418, 419, 444, and 447 (EU numbering) of the CH3 region. Specific substitutions that can be used to reduce the pI of the CH3 domain include, but are not limited to, a non-natural glutamine or glutamic acid at position 355, a non-natural serine at position 384, a non-natural asparagine or glutamic acid at position 392, a non-natural methionine at position 397, a non-natural glutamic acid at position 419, a non-natural glutamic acid at position 359, a non-natural glutamic acid at position 362, a non-natural glutamic acid at position 389, a non-natural glutamic acid at position 418, a non-natural glutamic acid at position 444, and a deletion or non-natural aspartic acid at position 447.

[0284] Generally, as will be appreciated by those skilled in the art, there are two general categories of pI variants: variants that increase the pI of a protein (alkaline changes) and variants that decrease the pI of a protein (acidic changes). As described herein, all combinations of these variants can be made: one monomer can be wild-type, or a variant that does not exhibit a pI significantly different from wild-type, and the other can be more basic or more acidic. Alternatively, each monomer can be altered so that one is more basic and one is more acidic.

[0285] Preferred combinations of pI variants are shown in Figure 4 As outlined herein and shown in the figures, these changes are shown relative to IgG1, but all isotypes can be altered in this manner, as well as isotype hybrids. In cases where the heavy chain constant domain is from IgG2-4, R133E and R133Q can also be used.

[0286] In one embodiment, for example, Figure 15A and Figure 15BIn the format, a preferred combination of pI variants has one monomer (negative Fab side) comprising the 208D / 295E / 384D / 418E / 421D variant (N208D / Q295E / N384D / Q418E / N421D relative to human IgG1) and a second monomer (positive scFv side) comprising a positively charged scFv linker comprising (GKPGS)4 (SEQ ID NO: XXX). However, as will be appreciated by those skilled in the art, the first monomer comprises a CH1 domain, which comprises position 208. Thus, in constructs that do not comprise a CH1 domain (e.g., for antibodies that do not utilize a CH1 domain on one of the domains, such as in a dual scFv format or a "one-arm" format, such as Figure 52B 、 52C or 52D), a preferred negative pi variant Fc set comprises the 295E / 384D / 418E / 421D variant (Q295E / N384D / Q418E / N421D relative to human IgG1).

[0287] Thus, in some embodiments, a monomer has a Figure 4 and the other monomer has a charged linker (in the form of a charged scFv linker, as the monomer comprises an scFv as indicated by the format or a charged domain linker, which can be selected from Figure 5 those depicted in ).

[0288] 3. Isotype variants

[0289] In addition, many embodiments of the antibodies described herein rely on "importing" pi amino acids at specific positions from one IgG isotype into another IgG isotype, thereby reducing or eliminating the possibility of introducing undesirable immunogenicity into the variant. Figure 21 A variety of these variants are shown in [ 1 ], which is hereby incorporated by reference. That is, for various reasons, IgG1 is a commonly used isotype for therapeutic antibodies, containing high effector functions. However, the heavy chain constant region of IgG1 has a higher pI than that of IgG2 (8.10 versus 7.31). By introducing IgG2 residues at specific positions into the IgG1 backbone, the pI of the resulting monomer can be lowered (or increased) and additionally exhibit a longer serum half-life. For example, IgG1 has a glycine at position 137 (pI of 5.97), while IgG2 has a glutamic acid (pI of 3.22); the introduction of the glutamic acid will affect the pI of the resulting protein. As described below, multiple amino acid substitutions are generally required to significantly affect the pI of a variant antibody. However, it should be noted that, as discussed below, uniform changes within an IgG2 molecule can result in an extended serum half-life.

[0290] In other embodiments, non-isotypic amino acid changes are made to lower the overall charge state of the resulting protein (e.g., by changing a higher pi amino acid to a lower pi amino acid), or to allow for structural adaptations for stability, etc., as described further below.

[0291] In addition, by engineering the pIs of the heavy and light chain constant domains, it was found that the individual monomers in the heterodimers were significantly altered. As discussed herein, having the pIs of the two monomers differ by at least 0.5 can allow separation by ion exchange chromatography or isoelectric focusing or other methods sensitive to the isoelectric point.

[0292] 4. Calculate pI

[0293] The pI of each monomer can depend on the pI of the variant heavy chain constant domain and the pI of the total monomer, which includes the variant heavy chain constant domain and the fusion partner. Therefore, in some embodiments, using the chart of Figure 19 in U.S. Publication 2014 / 0370013, pI changes are calculated based on the variant heavy chain constant domain: pI changes are calculated based on the variant heavy chain constant domain. As discussed herein, which monomer to be engineered is generally determined by the inherent pI of the Fv and scaffold regions. Alternatively, the pI of each monomer can be compared.

[0294] 5. pi variants that also confer better FcRn binding in vivo

[0295] Where the pi variant reduces the monomeric pi, this may have the added benefit of improving serum retention time in vivo.

[0296] Although still under review, it is believed that the Fc region has a longer half-life in vivo because Fc is sequestered in the endosome at pH 6 by binding to FcRn (Ghetie and Ward, 1997, Immunol Today. 18(12):592-598, which is hereby incorporated by reference in its entirety). The endosomal compartment then recycles Fc to the cell surface. Once the compartment is open to the extracellular space, the higher pH (approximately 7.4) induces the release of Fc back into the blood. Dall'Acqua et al. showed that in mice, Fc mutants with enhanced FcRn binding at pH 6 and pH 7.4 actually had reduced serum concentrations and the same half-life as wild-type Fc (Dall'Acqua et al., 2002, J. Immunol. 169:5171-5180, which is hereby incorporated by reference in its entirety). It is believed that the increased affinity of Fc for FcRn at pH 7.4 prevents Fc release back into the blood. Therefore, Fc mutations that increase Fc's in vivo half-life ideally enhance FcRn binding at lower pH while still allowing Fc release at higher pH. The amino acid histidine changes its charge state within the pH range of 6.0-7.4. Therefore, it is not surprising that His residues are found at key positions within the Fc / FcRn complex.

[0297] Recently, it has been proposed that antibodies with variable regions with lower isoelectric points may also have longer serum half-lives (Igawa et al., 2010 Protein Engineering Design and Selection (PEDS) 23(5): 385-392, which is incorporated by reference in its entirety). However, the mechanism for this remains poorly understood. Moreover, variable regions vary from antibody to antibody. Constant region variants with reduced pI and extended half-life would provide a more modular approach to improving the pharmacokinetic properties of antibodies, as described herein.

[0298] C. Additional Fc Variants for Additional Functions

[0299] In addition to the heterodimerization variants discussed above, there are many useful Fc amino acid modifications that can be made for a variety of reasons, including but not limited to altering binding to one or more FcγR receptors, altering binding to FcRn receptors, etc., as discussed below.

[0300] Therefore, the antibodies provided herein (heterodimers and homodimers) can include such amino acid modifications with or without the heterodimerization variants (e.g., pI variants and steric variants) outlined herein. Each set of variants can independently and optionally include any specific heterodimer protein or exclude any specific heterodimer protein.

[0301] 1. FcγR variants

[0302] Thus, a variety of useful Fc substitutions can be made to alter binding to one or more FcγR receptors. In certain embodiments, the antibodies of the invention comprise modifications that alter binding to one or more FcγR receptors (i.e., "FcγR variants"). Substitutions that result in increased binding as well as decreased binding can be useful. For example, it is known that increased binding to FcγRIIIa generally results in increased ADCC (antibody-dependent cell-mediated cytotoxicity; a cell-mediated reaction in which nonspecific toxic cells expressing FcγRs recognize bound antibody on a target cell and subsequently cause lysis of the target cell). Similarly, in some cases, decreased binding to FcγRIIb (an inhibitory receptor) may also be advantageous. Amino acid substitutions useful for the antibodies described herein are included in U.S. Patent No. 8,188,321 (specifically, Figure 41 ) and No. 8,084,582 and U.S. Published Application Nos. 20060235208 and 20070148170, all of which are expressly incorporated herein by reference in their entireties, particularly with respect to the variants disclosed therein. Specific variants that may be used include, but are not limited to, 236A, 239D, 239E, 332E, 332D, 239D / 332E, 267D, 267E, 328F, 267E / 328F, 236A / 332E, 239D / 332E / 330Y, 239D / 332E / 330L, 243A, 243L, 264A, 264V, and 299T.

[0303] In addition, as specifically disclosed in USSN 12 / 341,769, which is hereby incorporated by reference in its entirety, there are additional Fc substitutions that can be used to increase binding to the FcRn receptor and extend serum half-life, including but not limited to 434S, 434A, 428L, 308F, 259I, 428L / 434S, 259I / 308F, 436I / 428L, 436I or V / 434S, 436V / 428L, and 259I / 308F / 428L. Such modifications may be included in one or both Fc domains of the antibodies of the invention.

[0304] 2. Ablation variants

[0305] Similarly, another type of functional variant is an "FcγR ablation variant" or "Fc knockout (FcKO or KO)" variant. In these embodiments, in some therapeutic applications, it is desirable to reduce or eliminate the normal binding of the Fc domain to one or more or all Fcγ receptors (e.g., FcγR1, FcγRIIa, FcγRIIb, FcγRIIIa, etc.) to avoid additional mechanisms of action. That is, for example, in many embodiments, particularly in the use of bispecific antibodies that bind monovalently to CD3, it is often desirable to ablate FcγRIIIa binding to eliminate or significantly reduce ADCC activity. One of the Fc domains comprises one or more Fcγ receptor ablation variants. These ablation variants are depicted in Figure 14 and each can be independently and optionally included or not, with preferred aspects utilizing an ablation variant selected from the group consisting of: G236R / L328R, E233P / L234V / L235A / G236del / S239K, E233P / L234V / L235A / G236del / S267K, E233P / L234V / L235A / G236del / S239K / A327G, E233P / L234V / L235A / G236del / S267K / A327G, and E233P / L234V / L235A / G236del. It should be noted that the ablation variants mentioned herein ablate FcγR binding, but generally do not ablate FcRn binding.

[0306] As is known in the art, the Fc domain of human IgG1 has the highest binding to Fcγ receptors, and therefore, when the constant domain (or Fc domain) in the backbone of the heterodimeric antibody is IgG1, ablation variants can be used. For example, alternatively or in addition to ablation variants in the IgG1 context, a mutation at glycosylation position 297 (typically to A or S) can significantly ablate binding to FcγRIIIa. Human IgG2 and IgG4 have naturally reduced binding to Fcγ receptors, and therefore those backbones can be used with or without ablation variants.

[0307] D. Combination of heterodimers and Fc variants

[0308] As will be understood by those skilled in the art, all of the heterodimerization variants (including skew and / or pI variants) can be optionally and independently combined in any manner as long as they retain their "chainness" or "monomer partitioning". In some embodiments, the heterodimeric antibodies provided herein include combinations of heterodimerization skew variants, isoprene pI substitutions, and FcKO variants, such as Figure 4 In addition, all of these variants can be combined into any of the heterodimeric forms.

[0309] In the case of pi variants, while a particularly useful embodiment is shown in the figures, other combinations can be generated based on the basic principle of varying the pi difference between two monomers to facilitate purification.

[0310] Additionally, any of the heterodimerization variants, skew and pi, are also independently and optionally combined with Fc ablation variants, Fc variants, FcRn variants, as generally outlined herein.

[0311] Exemplary combinations of variants included in some embodiments of heterodimeric 1+1 Fab-scFv-Fc and 2+1 Fab2-scFv-Fc formats of antibodies include Figure 4 In certain embodiments, the antibody is a heterodimeric 1+1 Fab-scFv-Fc or 2+1 Fab2-scFv-Fc antibody, such as Figure 15A and Figure 15B shown.

[0312] E. Anti-ENPP3 x anti-CD3 bispecific antibody

[0313] In another aspect, provided herein are anti-ENPP3 x anti-CD3 (also referred to herein as "αENPP3 x αCD3") bispecific antibodies. These antibodies comprise at least one ENPP3 binding domain and at least one CD3 binding domain. In some embodiments, the bispecific αENPP3 x αCD3 antibodies provided herein selectively induce an immune response at the site of a tumor expressing ENPP3.

[0314] It should be noted that, unless otherwise indicated herein, the order in which the antigens are listed by name does not confer structure; that is, an ENPP3 x CD3 1+1 Fab-scFv-Fc antibody may have an scFv that binds to either ENPP3 or CD3, but in some cases the order specifies the structure as indicated.

[0315] As more fully outlined herein, these combinations of ABDs can be in various formats as outlined below, typically a combination in which one ABD is in Fab format and the other is in scFv format. Exemplary formats for the bispecific antibodies provided herein include 1+1 Fab-scFv-Fc and 2+1 Fab2-scFv-Fv formats (see, e.g., Figure 15A and Figure 15B Other useful antibody formats include, but are not limited to, "mAb-Fv," "mAb-scFv," "central-Fv," "single-arm scFv-mAb," "scFv-mAb," "dual scFv," and "trident" forms of antibodies, such as Figures 52A to 52K In the public.

[0316] In addition, typically one of the ABDs comprises an scFv in an N- to C-terminal orientation as outlined herein, either VH-scFv linker-VL or VL-scFv linker-VH. Depending on the format, one or both of the other ABDs are typically Fabs comprising a VH domain on one protein chain (usually as a component of the heavy chain) and a VL on another protein chain (usually as a component of the light chain).

[0317] As will be appreciated by those skilled in the art, any set of six CDRs or VH and VL domains can be in scFv format or Fab format, which is then added to heavy and light constant domains, wherein the heavy constant domain includes variants (contained within the CH1 domain as well as the Fc domain). The scFv sequences contained in the sequence listing utilize specific charged linkers, while as outlined herein, uncharged or other charged linkers can be used, including Figure 5 and Figure 6 those described in .

[0318] Additionally, as discussed above, the numbering used in the sequence listing to identify the CDRs is Kabat, however, different numbering may be used which would alter the amino acid sequences of the CDRs as shown in Table 2.

[0319] For all variable heavy and light domains listed herein, additional variants can be prepared. As outlined herein, in some embodiments, a set of 6 CDRs can have 0, 1, 2, 3, 4, or 5 amino acid modifications (where amino acid substitutions are particularly useful) and changes in the framework regions of the variable heavy and light domains, as long as the framework (excluding the CDRs) remains at least about 80%, 85%, or 90% identical to a human germline sequence selected from those sequences listed in U.S. Patent No. 7,657,380 in Figure 1, the drawings and legends of which are incorporated herein by reference in their entirety. Thus, for example, the same CDRs as described herein can be combined with different framework sequences from human germline sequences, as long as the framework regions remain at least 80%, 85%, or 90% identical to a human germline sequence selected from those sequences listed in U.S. Patent No. 7,657,380 in Figure 1. Alternatively, the CDRs can have amino acid modifications (e.g., 1, 2, 3, 4, or 5 amino acid modifications in a set of CDRs (that is, the CDRs can be modified as long as the total number of changes in a set of 6 CDRs is less than 6 amino acid modifications, wherein any combination of CDRs is altered; for example, there can be one alteration in vlCDR1, two alterations in vhCDR2, no alterations in vhCDR3, etc.)), and have framework region alterations, so long as the framework regions maintain at least 80%, 85%, or 90% identity to a human germline sequence selected from those sequences listed in U.S. Patent No. 7,657,380 in Figure 1 .

[0320] The anti-ENPP3 x anti-CD3 bispecific antibody may comprise any suitable CD3 ABD, including those described herein (see e.g., 10A to 10F In some embodiments, the CD3 ABD of the anti-ENPP3 x anti-CD3 bispecific antibody comprises the variable heavy chain domain and the variable light chain domain of the CD3 ABD provided herein, including 10A to 10F and those described in the sequence listing. In some embodiments, the CD3 ABD comprises the variable heavy chain domain and variable light chain domain of any of the following CD3 ABDs: H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10FIn exemplary embodiments, the CD3 ABD is one of the following CD3 ABDs: H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10F ) or their variations.

[0321] The anti-ENPP3 x anti-CD3 bispecific antibody may comprise any suitable ENPP3 ABD, including those described herein (see, e.g., Figure 12 、 13A to 13B and 14A to 14I In some embodiments, the ENPP3 ABD of the anti-ENPP3 x anti-CD3 bispecific antibody comprises the variable heavy chain domain and the variable light domain of the ENPP3 ABD provided herein, included in Figure 12 、 13A to 13B and 14A to 14I and those described in the Sequence Listing. In some embodiments, the ENPP3 ABD comprises the variable heavy chain domain and the variable light chain domain of one of the following ENPP3 ABDs: AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19 and Ha16-1.80 ( Figure 12 、 13A to 13B and 14A to 14I). In exemplary embodiments, the ENPP3 ABD is one of the following ENPP3 ABDs: AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19 and Ha16-1.80 ( Figure 12 、 13A to 13B and 14A to 14I ) or their variations.

[0322] F. Anti-SSTR2 x anti-CD3 bispecific antibody

[0323] In another aspect, provided herein are anti-SSTR2 x anti-CD3 (also referred to herein as "αSSTR2 x αCD3") bispecific antibodies. These antibodies comprise at least one SSTR2-binding domain and at least one CD3-binding domain. In some embodiments, the bispecific αSSTR2 x αCD3 antibodies provided herein selectively induce an immune response at the site of a tumor expressing SSTR2.

[0324] It should be noted that, unless otherwise indicated herein, the order in which the antigens are listed by name does not confer structure; that is, an SSTR2 x CD3 1 + 1 Fab-scFv-Fc antibody can have an scFv that binds to either SSTR2 or CD3, but in some cases the order specifies the structure as indicated.

[0325] As more fully outlined herein, these combinations of ABDs can be in various formats as outlined below, typically a combination in which one ABD is in Fab format and the other is in scFv format. Exemplary formats for the bispecific antibodies provided herein include 1+1 Fab-scFv-Fc and 2+1 Fab2-scFv-Fv formats (see, e.g., Figure 15A and Figure 15B Other useful antibody formats include, but are not limited to, "mAb-Fv," "mAb-scFv," "central-Fv," "single-arm scFv-mAb," "scFv-mAb," "dual scFv," and "trident" forms of antibodies, such as Figures 52A to 52K In the public.

[0326] In addition, typically one of the ABDs comprises an scFv in an N- to C-terminal orientation as outlined herein, either VH-scFv linker-VL or VL-scFv linker-VH. Depending on the format, one or both of the other ABDs are typically Fabs comprising a VH domain on one protein chain (usually as a component of the heavy chain) and a VL on another protein chain (usually as a component of the light chain).

[0327] As will be appreciated by those skilled in the art, any set of six CDRs or VH and VL domains can be in scFv format or Fab format, which is then added to heavy and light constant domains, wherein the heavy constant domain includes variants (contained within the CH1 domain as well as the Fc domain). The scFv sequences contained in the sequence listing utilize specific charged linkers, while as outlined herein, uncharged or other charged linkers can be used, including Figure 5 and Figure 6 those described in .

[0328] Additionally, as discussed above, the numbering used in the sequence listing to identify the CDRs is Kabat, however, different numbering may be used which would alter the amino acid sequences of the CDRs as shown in Table 2.

[0329] For all variable heavy and light domains listed herein, additional variants can be prepared. As outlined herein, in some embodiments, a set of 6 CDRs can have 0, 1, 2, 3, 4, or 5 amino acid modifications (where amino acid substitutions are particularly useful) and changes in the framework regions of the variable heavy and light domains, as long as the framework (excluding the CDRs) remains at least about 80%, 85%, or 90% identical to a human germline sequence selected from those sequences listed in U.S. Patent No. 7,657,380 in Figure 1, the drawings and legends of which are incorporated herein by reference in their entirety. Thus, for example, the same CDRs as described herein can be combined with different framework sequences from human germline sequences, as long as the framework regions remain at least 80%, 85%, or 90% identical to a human germline sequence selected from those sequences listed in U.S. Patent No. 7,657,380 in Figure 1. Alternatively, the CDRs can have amino acid modifications (e.g., 1, 2, 3, 4, or 5 amino acid modifications in a set of CDRs (that is, the CDRs can be modified as long as the total number of changes in a set of 6 CDRs is less than 6 amino acid modifications, wherein any combination of CDRs is altered; for example, there can be one alteration in vlCDR1, two alterations in vhCDR2, no alterations in vhCDR3, etc.)), and have framework region alterations, so long as the framework regions maintain at least 80%, 85%, or 90% identity to a human germline sequence selected from those sequences listed in U.S. Patent No. 7,657,380 in Figure 1 .

[0330] The anti-SSTR2 x anti-CD3 bispecific antibody may comprise any suitable CD3 ABD, including those described herein (see e.g., 10A to 10F In some embodiments, the CD3 ABD of the anti-SSTR2 x anti-CD3 bispecific antibody comprises the variable heavy chain domain and the variable light chain domain of the CD3 ABD provided herein, including 10A to 10F and those described in the sequence listing. In some embodiments, the CD3 ABD comprises the variable heavy chain domain and variable light chain domain of any of the following CD3 ABDs: H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10FIn exemplary embodiments, the CD3 ABD is one of the following CD3 ABDs: H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10F ) or their variations.

[0331] The anti-SSTR2 x anti-CD3 bispecific antibody may comprise [αSSTR2] H1.24_L1.30 ( Figure 63 ), or variants thereof.

[0332] G. Useful Forms of the Invention

[0333] As will be appreciated by those skilled in the art and as discussed more fully below, the heterodimeric bispecific antibodies provided herein can be in various configurations, as generally depicted in FIG1 . Some of the figures depict a "single-ended" configuration, in which one type of specificity is present on one "arm" of the molecule and a different specificity is present on the other "arm." Other figures depict a "double-ended" configuration, in which the "top" of the molecule has at least one type of specificity and the "bottom" of the molecule has one or more different specificities. Thus, in some embodiments, the antibodies described herein are directed against novel immunoglobulin compositions that co-bind different first and second antigens.

[0334] As will be appreciated by those skilled in the art, the heterodimeric forms described herein can have different valencies and can be bispecific. That is, the heterodimeric antibodies described herein can be bivalent and bispecific, with one target tumor antigen (e.g., CD3) bound by one binding domain and another target tumor antigen (e.g., ENPP3) bound by a second binding domain. Heterodimeric antibodies can also be trivalent and bispecific, with the first antigen bound by two binding domains and the second antigen bound by the second binding domain. As outlined herein, when CD3 is one of the target antigens, it is preferred that CD3 be bound only monovalently to reduce potential side effects.

[0335] The antibodies described herein utilize a combination of an anti-CD3 antigen binding domain and an anti-ENPP3 binding domain. As will be appreciated by those skilled in the art, any set of anti-CD3 CDRs, anti-CD3 variable light and heavy chain domains, Fabs, and scFvs as depicted in any of the Figures may be used. Similarly, any anti-ENPP3 antigen binding domain may be used, regardless of whether a combination of anti-CD3 CDRs, anti-CD3 variable light and heavy chain domains, Fabs, and scFvs as depicted in any of the Figures (e.g., Figure 12 、 13A to 13B ,and 14A to 14I ) are optionally and independently combined in any combination.

[0336] 1. 1+1 Fab-scFv-Fc format

[0337] One heterodimeric scaffold that has found particular use in the antibodies described herein is the "1+1 Fab-scFv-Fc" or "bottle opener" format, e.g. Figure 15A As shown, an exemplary combination of a CD3 binding domain and a tumor target antigen (ENPP3) binding domain is provided. In this embodiment, one heavy chain monomer of the antibody contains a single-chain Fv ("scFv", as defined below) and an Fc domain. The scFv comprises a variable heavy chain domain (VH1) and a variable light chain domain (VL1), wherein VH1 is attached to VL1 using a chargeable scFv linker (see, e.g., Figure 5 ). The scFv is attached to the heavy chain using a domain linker (see, e.g., Figure 6 ). The other heavy chain monomer is a "conventional" heavy chain (VH-CH1-hinge-CH2-CH3). The 1+1 Fab-scFv-Fc also contains a light chain that interacts with the VH-CH1 to form the Fab. This structure is sometimes referred to herein as a "bottle opener" format due to its general visual resemblance to a bottle opener. The two heavy chain monomers are brought together by using amino acid variants in the constant region (e.g., Fc domain, CH1 domain, and / or hinge region) that promote heterodimeric antibody formation as described more fully below (e.g., the heterodimerization variants discussed above).

[0338] The "1+1 Fab-scFv-Fc" format of the present invention offers several distinct advantages. As is known in the art, antibody analogs relying on two scFv constructs often suffer from stability and aggregation issues, which can be alleviated in the antibodies described herein by adding "conventional" heavy and light chain pairings. Furthermore, unlike formats relying on two heavy chains and two light chains, there is no risk of improper heavy and light chain pairing (e.g., heavy chain 1 paired with light chain 2).

[0339] Many of the embodiments outlined herein generally rely on a 1+1 Fab-scFv-Fc or "bottle opener" format antibody comprising a first monomer comprising an scFv comprising a variable heavy chain domain and a variable light domain covalently attached using a scFv linker (charged in many, but not all, cases), wherein the scFv is typically covalently attached to the N-terminus of the first Fc domain via a domain linker. The domain linker can be charged or uncharged and exogenous or endogenous (e.g., all or part of a native hinge domain). Any suitable linker can be used to attach the scFv to the N-terminus of the first Fc domain. In some embodiments, the domain linker is selected from Figure 6 The second monomer in the 1+1 Fab-scFv-Fc format or "bottle opener" format is a heavy chain, and the composition also comprises a light chain.

[0340] Generally, in many preferred embodiments, the scFv is the CD3 binding domain and the Fab forms the ENPP3 binding domain. Figure 15A An exemplary anti-ENPP3 x anti-CD3 bispecific antibody in the form of 1+1 Fab-scFv-Fc is depicted in FIG. 17A to 17C and 18A to 18C Depicted in.

[0341] Additionally, the Fc domains of the antibodies described herein typically comprise skew variants (e.g., Figure 3 and Figure 9 70S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W and T366S / L368A / Y407V / Y349C:T366W / S354C), optionally ablated variants (comprising Figure 3 ), optionally charged scFv linkers (comprising Figure 5 ), and the heavy chain includes pi variants (comprising Figure 4 those shown in ).

[0342] In certain embodiments, the 1+1 Fab-scFv-Fc scaffold format comprises a first monomer comprising an scFv-domain linker-CH2-CH3 monomer; a second monomer comprising a first variable heavy chain domain-CH1-hinge-CH2-CH3 monomer; and a third monomer comprising a first variable light chain domain. In some embodiments, the CH2-CH3 of the first monomer is a first variant Fc domain, and the CH2-CH3 of the second monomer is a second variant Fc domain. In some embodiments, the scFv comprises an scFv variable heavy chain domain and an scFv variable light domain that form the CD3 binding portion. In certain embodiments, the scFv variable heavy chain domain and the scFv variable light domain are covalently linked using an scFv linker (charged, in many but not all cases). See, for example, Figure 5 In some embodiments, the first variable heavy chain domain and the first variable light domain form an ENPP3 binding domain. Particularly useful ENPP3 and CD3 combinations for 1+1 Fab-scFv-Fc ENPP3 xCD3 bispecific antibody formats are 17A to 17C and 18A to 18C 8 L1.77 x CD3 H.130 L1.47. In some embodiments, the 1+1 Fab-scFv-Fc format comprises a skew variant, a pI variant, and an ablation variant. Thus, some embodiments comprise a 1+1 Fab-scFv-Fc format comprising: a) a first monomer ("scFv monomer") comprising a charged scFv linker ( Figure 5The invention further comprises the following: a) a first monomer comprising a skew variant L368D / K370S, a pI variant N208D / Q295E / N384D / Q418E / N421D, a first monomer comprising a skew variant L368D / K370S, a pI variant N208D / Q295E / N384D / Q418E / N421D, a first monomer comprising a skew variant E233P / L234V / L235A / G236del / S267K, and a variable heavy chain domain; and c) a light chain comprising a variable light chain domain (VL) and a constant light chain domain (CL), wherein numbering is according to EU numbering. The variable heavy chain domain and the variable light chain domain constitute the ENPP3 binding portion. CD3 binding domain sequences particularly useful in these embodiments include, but are not limited to, H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31, as well as 10A to 10F ENPP3 binding domain sequences particularly useful in these embodiments include, but are not limited to, AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B and 14A to 14IParticularly useful ENPP3 and CD3 sequence combinations for 1+1 Fab2-scFv-Fc format antibodies include, for example, ENPP3 H16-1.93 x CD3 H1.30 L1.47, ENPP3 H16-7.8 x CD3 H1.30 L1.47, ENPP3 AN1 [ENPP3] H1L1 x CD3 H1.30 L1.47, ENPP3 AN1 [ENPP3] H1.8 L1 x CD3 H1.30 L1.47, ENPP3 AN1 [ENPP3] H1.8 L1 x CD3 H1.30 L1.47, ENPP3 AN1 [ENPP3] H1.8 L1.33 x CD3 H1.30L1.47, and ENPP3 H1.8 L1.77 x CD3 H.130 L1.47.

[0343] Exemplary variable heavy and light domains of scFvs that bind to CD3 are contained in 10A to 10F Exemplary variable heavy and light domains of Fvs that bind to ENPP3 are included in Figure 12 、 13A to 13B and 14A to 14I In an exemplary embodiment, the ENPP3 binding domain of the 1+1 Fab-scFv-Fc ENPP3 x CD3 bispecific antibody comprises the VH and VL of one of the following ENPP3 binding domains: AN1[ENPP3] H1L1, AN1[ENPP3] H1L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14IIn one embodiment, the CD3 binding domain of the 1+1 Fab-scFv-Fc ENPP3 x CD3 bispecific antibody comprises the VH and VL of one of the following CD3 binding domains: H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10F Particularly useful ENPP3 and CD3 combinations for 1+1 Fab-scFv-Fc ENPP3 x CD3 bispecific antibody formats are 17A to 17C and 18A to 18C 3 x CD3 H1.30 L1.47, and ENPP3 H1.8 L1.77 x CD3 H.130 L1.47.

[0344] In some embodiments, the 1+1 Fab-scFv-Fc format comprises a skew variant, a pI variant, an ablation variant, and an FcRn variant. Thus, some embodiments comprise a 1+1 Fab-scFv-Fc format comprising: a) a first monomer ("scFv monomer") comprising a charged scFv linker ( Figure 6+H sequence is preferred in some embodiments), skew variant S364K / E357Q, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S and a scFv that binds to CD3 as outlined herein; b) a second monomer ("Fab monomer") comprising the skew variant L368D / K370S, pI variant N208D / Q295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S and a variable heavy chain domain; and c) a light chain comprising a variable light chain domain (VL) and a constant light chain domain (CL), wherein numbering is according to EU numbering. The variable heavy chain domain and the variable light domain constitute the ENPP3 binding domain. The CD3 binding domain sequences particularly useful in these embodiments include, but are not limited to, H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31, as well as 10A to 10F ENPP3 binding domain sequences particularly useful in these embodiments include, but are not limited to, AN1[ENPP3] H1L1, AN1[ENPP3] H1L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80, as in Figure 12 、 13A to 13B and 14A to 14IParticularly useful ENPP3 and CD3 sequence combinations for use in the 1+1 Fab2-scFv-Fc format of antibodies include, for example, 17A to 17C and 18A to 18C and including: ENPP3 H16-1.93 x CD3 H1.30 L1.47, ENPP3 H16-7.8 x CD3 H1.30 L1.47, ENPP3 AN1 [ENPP3] H1L1 x CD3 H1.30 L1.47, ENPP3 AN1 [ENPP3] H1.8 L1 x CD3 H1.30 L1.47, ENPP3 AN1 [ENPP3] H1.8 L1 x CD3 H1.30 L1.47, ENPP3 AN1 [ENPP3] H1.8 L1.33 x CD3 H1.30 L1.47, and ENPP3 H1.8 L1.77 x CD3 H.130 L1.47.

[0345] 7A to 7D Some exemplary Fc domain sequences useful in 1+1 Fab-scFv-Fc format antibodies are shown. 7A to 7D The "monomer 1" sequence depicted in the diagram generally refers to the Fc domain of a "Fab-Fc heavy chain," while the "monomer 2" sequence refers to the Fc domain of a "scFv-Fc heavy chain." Figure 9 Useful CL sequences are provided that can be used in this form.

[0346] In some embodiments, any of the VH and VL sequences depicted herein (including all VH and VL sequences depicted in the Figures and Sequence Listing, including sequences directed against ENPP3) can be added to a scFv using any of the anti-CD3 scFv sequences shown in the Figures and Sequence Listing. 7A to 7D The corkscrew backbone forms as the "Fab side".

[0347] For those from Figure 7A The opener backbone 1 (optionally comprising a 428L / 434S variant), in particular CD binding domain sequences for use in these embodiments include but are not limited to CD3 binding domains anti-CD3 H1.30_L1.47), anti-CD3 H1.32_L1.47, anti-CD3 H1.89_L1.47, anti-CD3 H1.90_L1.47, anti-CD3 H1.33_L1.47 and anti-CD3 H1.31_L1.47, as well as in 10A to 10F Those described in 7A to 7D The scFv side of the backbone is connected as shown in FIG.

[0348] 17A to 17C and 18A to 18CParticularly useful ENPP3 and CD3 sequence combinations (optionally including the 428L / 434S variant) are disclosed in .

[0349] 2. mAb-Fv

[0350] One heterodimeric scaffold particularly useful in the antibodies described herein is the mAb-Fv format. In this embodiment, the format relies on the use of an "additional" variable heavy chain domain attached to the C-terminus of one monomer and an "additional" variable light chain domain attached to the C-terminus of another monomer, thereby forming a third antigen binding domain, wherein the Fab portions of both monomers bind ENPP3 and the "additional" scFv domain binds CD3.

[0351] In this embodiment, the first monomer comprises a first heavy chain comprising a first variable heavy chain domain and a first heavy chain constant domain comprising a first Fc domain, wherein a first variable light chain domain is covalently attached to the C-terminus of the first Fc domain using a domain linker (VH1-CH1-hinge-CH2-CH3-[optional linker]-VL2). The second monomer comprises a second variable heavy chain domain, which belongs to a second heavy chain constant domain comprising a second Fc domain, and a third variable heavy chain domain, which is covalently linked to the C-terminus of the second Fc domain using a domain linker (vh1-CH1-hinge-CH2-CH3-[optional linker]-VH2). The two C-terminally attached variable domains constitute the Fv that binds to CD3 (as having bivalent CD3 binding is less preferred). This embodiment further utilizes a light chain comprising a variable light chain domain and a light chain constant domain, which associates with the heavy chain to form two identical Fabs that bind to ENPP3. For many embodiments herein, these constructs comprise skew variants, pi variants, ablation variants, additional Fc variants, etc., as desired and described herein.

[0352] The antibodies described herein are provided in mAb-Fv format, wherein the CD3 binding domain sequence is as follows 10A to 10F The antibodies described herein are provided in mAb-Fv format, wherein the ENPP3 binding domain sequence is as shown in Figure 12 、 13A to 13B and 14A to 14I shown.

[0353] In addition, the Fc domain of the mAb-Fv format contains skew variants (e.g. Figure 3and a set of amino acid substitutions shown in Figure 8, and particularly useful skew variants are selected from the group consisting of: S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W and T366S / L368A / Y407V / Y349C:T366W / S354C; optionally ablation variants (including Figure 3 those shown in ); optionally charged scFv linkers (including Figure 5 those shown in ); and heavy chains comprising pI variants (including Figure 2 those shown in ).

[0354] In some embodiments, the mAb-Fv format comprises a skew variant, a pI variant, and an ablation variant. Thus, some embodiments comprise a mAb-Fv format comprising: a) a first monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, and a first variable heavy chain domain and a second variable heavy chain domain, wherein the first variable heavy chain domain and the first variable light domain of the light chain constitute an Fv that binds to ENPP3; b) a second monomer comprising the skew variant L368D / K370S, the pI variant N 208D / Q295E / N384D / Q418E / N421D, the ablation variant E233P / L234V / L235A / G236del / S267K and a first variable heavy chain domain and a second variable light chain, wherein the first variable heavy chain domain and the first variable light domain constitute an Fv that binds to ENPP3 as outlined herein, and the second variable light chain together with the second variable heavy chain domain forms an Fv that binds to CD3 (ABD); and c) a light chain comprising a first variable light domain and a constant light domain.

[0355] In some embodiments, the mAb-Fv format comprises a skew variant, a pI variant, an ablation variant, and an FcRn variant. Thus, some embodiments comprise a mAb-Fv format comprising: a) a first monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, the FcRn variant M428L / N434S, and a first variable heavy chain domain and a second variable heavy chain domain, wherein the first variable heavy chain domain and the first variable light chain domain form an Fv that binds to ENPP3; b) a second monomer comprising the skew variant L368D / K370S, the pI variant N208 D / Q295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S and a first variable heavy chain domain and a second variable light chain, wherein the first variable heavy chain domain and the first variable light domain constitute an Fv that binds to ENPP3 as outlined herein, and the second variable light chain together with the second variable heavy chain domain of the first monomer forms an Fv that binds to CD3 (ABD); and c) a light chain comprising a first variable light domain and a constant light domain.

[0356] 3. mAb-scFv

[0357] One heterodimeric scaffold particularly useful in the antibodies described herein is the mAb-scFv format. In this embodiment, the format relies on the use of an scFv attached to the C-terminus of one of the monomers, thereby forming a third antigen-binding domain, wherein the Fab portions of both monomers bind to ENPP3 and the "additional" scFv domain binds to CD3. Thus, the first monomer comprises a first heavy chain (comprising a variable heavy domain and a constant domain) with a C-terminal covalently linked scFv comprising an scFv variable light domain, an scFv linker, and an scFv variable heavy domain, in either orientation. This embodiment further utilizes a light chain comprising a variable light domain and a light chain constant domain, which associates with the heavy chain to form two identical Fabs that bind to ENPP3. For many embodiments herein, these constructs include skew variants, pI variants, ablated variants, additional Fc variants, and the like, as desired and described herein.

[0358] The antibodies described herein are provided in the form of mAb-scFv, wherein the CD binding domain sequence is as follows 10A to 10F The ENPP3 binding domain sequence is shown in Figure 12 、 13A to 13B and 14A to 14I shown.

[0359] In addition, the Fc domain of the mAb-scFv format comprises skew variants (e.g., the set of amino acid substitutions shown in FIG. 1 , and particularly useful skew variants are selected from the group consisting of: S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W, and T366S / L368A / Y407V / Y349C:T366W / S354C); optionally, ablation variants (including Figure 3 those shown in ); optionally charged scFv linkers (including Figure 5 those shown in ); and heavy chains comprising pI variants (including Figure 2 those shown in ).

[0360] In some embodiments, the mAb-scFv format comprises skew variants, pI variants, and ablation variants. Thus, some embodiments comprise a mAb-scFv format comprising: a) a first monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, a variable heavy chain domain, and a scFv domain, wherein the variable heavy chain domain and the variable light domain of the common light chain form an Fv that binds to ENPP3 as described herein, and the scFv domain binds to CD3; b) a second monomer The invention also provides a monomer comprising a skew variant L368D / K370S, a pI variant N208D / Q295E / N384D / Q418E / N421D, an ablation variant E233P / L234V / L235A / G236del / S267K and a variable heavy chain domain, wherein the variable heavy chain domain and the variable light domain of the common light chain constitute an Fv that binds to ENPP3 as outlined herein; and c) a common light chain comprising a variable light domain and a constant light domain.

[0361] In some embodiments, the mAb-scFv format comprises a skew variant, a pI variant, an ablation variant, and an FcRn variant. Thus, some embodiments include a mAb-scFv format comprising: a) a first monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, the FcRn variant M428L / N434S, and a variable heavy chain domain (which together with the variable light chain domain of the common light chain constitutes an Fv that binds to ENPP3 as outlined herein and a ScFv domain that binds to CD3); b) a second monomer comprising The skew variant L368D / K370S, the pI variant N208D / Q295E / N384D / Q418E / N421D, the ablation variant E233P / L234V / L235A / G236del / S267K, the FcRn variant M428L / N434S and the variable heavy chain domain (which together with the variable light chain domain of the common light chain constitute an Fv that binds to ENPP3 as outlined herein); and c) a common light chain comprising a variable light chain domain and a light chain constant domain.

[0362] 4. 2+1 Fab2-scFv-Fc format

[0363] One heterodimeric scaffold found to be particularly useful in the antibodies described herein is the "2+1 Fab2-scFv-Fc" format (also referred to as the "central-scFv format" in previous related documents), e.g. Figure 15BAn exemplary combination of a CD3-binding domain and two tumor target antigen (ENPP3)-binding domains is shown. In this embodiment, the format relies on the use of an inserted scFv domain to form a third antigen-binding domain, with the Fab portions of two monomers binding to ENPP3 and the "additional" scFv domain binding to CD3. The scFv domain is inserted between the Fc domain and the CH1-Fv region of one of the monomers, providing the third antigen-binding domain. For example, an ENPP3 x CD3 bispecific antibody with a 2+1 Fab2-scFv-Fc format is effective in inducing redirected T cell cytotoxicity in cells expressing low levels of ENPP3. Furthermore, as shown in the examples, the ENPP3 x CD3 bispecific antibody with a 2+1 Fab2-scFv-Fc format allows for "fine-tuning" of the immune response, as such antibodies exhibit a wide variety of different properties depending on the ENPP3 and / or CD3-binding domains used. For example, such antibodies exhibit differences in selectivity for cells with different ENPP3 expression, potency against ENPP3-expressing cells, ability to trigger cytokine release, and sensitivity to soluble ENPP3. These ENPP3 antibodies can be used, for example, to treat ENPP3-associated cancers.

[0364] In this embodiment, a monomer comprises a first heavy chain comprising a first variable heavy chain domain, a CH1 domain (and optionally a hinge), and an Fc domain, wherein the scFv comprises an scFv variable light domain, an scFv linker, and an scFv variable heavy chain domain. The scFv is covalently linked between the C-terminus of the CH1 domain of the heavy chain constant domain and the N-terminus of the first Fc domain using an optional domain linker (VH1-CH1-[optional linker]-VH2-scFv linker-VL2-[optional linker including hinge]-CH2-CH3, or the reverse orientation of the scFv VH1-CH1-[optional linker]-VL2-scFv linker-VH2-[optional linker including hinge]-CH2-CH3). The optional linker can be any suitable peptide linker, including, for example, Figure 6 In some embodiments, the optional linker is a hinge or a fragment thereof. The other monomer is a standard Fab side (i.e., VH1-CH1-hinge-CH2-CH3). This embodiment further utilizes a light chain comprising a variable light chain domain and a light chain constant domain, which associates with the heavy chain to form two identical Fabs that bind to ENPP3. For many embodiments herein, these constructs include skew variants, pI variants, ablation variants, additional Fc variants, etc., as desired and described herein.

[0365] In one embodiment, the 2+1 Fab2-scFv-Fc format antibody comprises 10A to 10F In one embodiment, the 2+1 Fab2-scFv-Fc format antibody comprises a VH and VL scFv having a CD3 binding domain sequence as described in the sequence listing. Figure 12 、 13A to 13B and 14A to 14I and two Fabs of the VH and VL of the ENPP3 binding domains shown in the sequence listing. In an exemplary embodiment, the ENPP3 binding domain of the 2+1 Fab2-scFv-Fc ENPP3 x CD3 bispecific antibody comprises the VH and VL of one of the following ENPP3 binding domains: AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, H16-9.44, H16-1.67, Ha16-1(2). (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I In one embodiment, the CD3 binding domain of the antibody in the 2+1 Fab2-scFv-Fc format comprises the VH and VL of one of the following CD3 binding domains: H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10F A particularly useful ENPP3 and CD3 combination for a 2+1 Fab2-scFv-Fc antibody format is 19A to 19C 、 20A to 20D 、 Figure 21 、 Figures 22A to 22C 、 Figures 23A to 23E ENPP3 H1.8 L1 x CD3 H1.30 L1.47, ENPP3 H1.8 L1.33 x CD3 H1.30 L1.47, ENPP3H1.8 L1.77 x CD3 H1.30 L1.47, ENPP3 H16-7.8 x CD3 H1.32 L1.47, ENPP3 AN[ENPP3]H1L1 x CD3 H1.32 L1.47, ENPP3 H1.8 L1 x CD3 H1.32 L1.47, ENPP3 H1.8 L1 x CD3 H1.32 L1.47, ENPP3 H1.8 L1 x CD3 H1.32 L1.47, ENPP3 H1.8 L1 x CD3 H1.32 L1.47 L1.47 H1.32、ENPP3 H1.8 L1.77 x CD3 L1.47 H1.32、ENPP3H1.8 L1 x CD3 L1.47 H1.89、ENPP3 H1.8 L1.33 x CD3 L1.47 H1.89、ENPP3 H1.8 L1.77x CD3 L1.47 H1.89、ENPP3 H1.8 L1.33 x CD3 L1.47 H1.89, and ENPP3 H1.8 L1.77 x CD3L1.47 H1.89.

[0366] In addition, the Fc domain of the 2+1 Fab2-scFv-Fc format comprises a skew variant (e.g., a set of amino acid substitutions shown in FIG1 , and particularly useful skew variants are selected from the group consisting of: S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W, and T366S / L368A / Y407V / Y349C:T366W / S354C); optionally, an ablation variant (including Figure 3 those shown in ); optionally charged scFv linkers (including Figure 5 those shown in ); and heavy chains comprising pI variants (including Figure 2those shown in ).

[0367] In some embodiments, the 2+1 Fab2-scFv-Fc format antibody comprises a skew variant, a pI variant, and an ablation variant. Thus, some embodiments comprise a 2+1 Fab2-scFv-Fc format comprising: a) a first monomer (Fab-scFv-Fc side), the first monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, and a variable heavy chain domain, the variable heavy chain domain and the variable light chain domain of the common light chain constituting an Fv binding to ENPP3 as outlined herein and an scFv domain binding to CD3; b) a second monomer (Fab-Fc side), the The second monomer includes the skew variant L368D / K370S, the pI variant N208D / Q295E / N384D / Q418E / N421D, the ablation variant E233P / L234V / L235A / G236del / S267K and a variable heavy chain domain, which together with the variable light chain domain of the common light chain constitutes an Fv that binds to ENPP3 as described herein; and c) a light chain comprising a variable light domain and a constant light domain, wherein numbering is according to EU numbering. In some embodiments, the common light chain and variable heavy chain domain on each monomer form an ENPP3 binding domain. CD3 binding domain sequences particularly useful in these embodiments include, but are not limited to, H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31, as well as 10A to 10FENPP3 binding domain sequences particularly useful in these embodiments include, but are not limited to, AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80, as in Figure 12 、 13A to 13B and 14A to 14I Described in.

[0368] In some embodiments, the 2+1 Fab2-scFv-Fc format antibody comprises a skew variant, a pI variant, an ablation variant, and an FcRn variant. Thus, some embodiments comprise a 2+1 Fab2-scFv-Fc format comprising: a) a first monomer (Fab-scFv-Fc side), the first monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, the FcRn variant M428L / N434S, and a variable heavy domain, the variable heavy domain and the variable light chain domain of the common light chain constituting an Fv binding to ENPP3 as outlined herein and an scFv domain binding to CD3; b) a second monomer (Fab-Fc side), the The second monomer includes the skew variant L368D / K370S, the pI variant N208D / Q295E / N384D / Q418E / N421D, the ablation variant E233P / L234V / L235A / G236del / S267K, the FcRn variant M428L / N434S and a variable heavy domain, which together with the variable light domain of the common light chain constitutes an Fv that binds to ENPP3 as described herein; and c) a light chain comprising a variable light domain and a constant light domain, wherein numbering is according to EU numbering. In some embodiments, the common light chain and variable heavy chain domain on each monomer form an ENPP3 binding domain. CD3 binding domain sequences particularly useful in these embodiments include, but are not limited to, H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31, as well as 10A to 10FENPP3 binding domain sequences particularly useful in these embodiments include, but are not limited to, AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80, as in Figure 12 、 13A to 13B and 14A to 14I Described in.

[0369] Figures 8A to 8C Shown are some exemplary Fc domain sequences that can be used in a 2+1 Fab2-scFv-Fc format. Figures 8A to 8C The "monomer 1" sequence depicted in the diagram generally refers to the Fc domain of a "Fab-Fc heavy chain," while the "monomer 2" sequence refers to the Fc domain of a "Fab-scFv-Fc heavy chain." Figure 9 Useful CL sequences are provided that can be used in this form.

[0370] 5. Center-Fv

[0371] One heterodimeric scaffold particularly useful in the antibodies described herein is a central-Fv format. In this embodiment, the format relies on the use of an inserted Fv domain (i.e., a central Fv domain) to form an "additional" third antigen-binding domain, wherein the Fab portion of two monomers binds to ENPP3 and the "additional" central Fv domain binds to CD3. The "additional" central Fv domain is inserted between the Fc domain and the CH1-Fv region of the monomer, thereby providing a third antigen-binding domain (i.e., the "additional" central Fv domain), wherein each monomer contains the components of the "additional" central Fv domain (i.e., one monomer includes the variable heavy chain domain and the other includes the variable light domain of the "additional" central Fv domain).

[0372] In this embodiment, one monomer includes a first heavy chain comprising a first variable heavy chain domain, a CH1 domain, an Fc domain, and an additional variable light domain. The light chain domain is covalently linked between the C-terminus of the CH1 domain of the heavy chain constant domain and the N-terminus of the first Fc domain using a domain linker (VH1-CH1-[optional linker]-VL2-hinge-CH2-CH3). The other monomer includes a first heavy chain comprising a first variable heavy chain domain, a CH1 domain, an Fc domain, and an additional variable heavy chain domain (VH1-CH1-[optional linker]-VH2-hinge-CH2-CH3). The light chain domain is covalently linked between the C-terminus of the CH1 domain of the heavy chain constant domain and the N-terminus of the first Fc domain using a domain linker.

[0373] This embodiment further utilizes a light chain comprising a variable light chain domain and a light chain constant domain that associates with the heavy chain to form two identical Fabs, each binding to ENPP3. For many embodiments herein, these constructs comprise skew variants, pI variants, ablation variants, additional Fc variants, etc., as desired and described herein.

[0374] The antibodies described herein provide a central-Fv format in which the CD3 binding domain sequence is as follows 10A to 10F The ENPP3 binding domain sequence is shown in Figure 12 、 13A to 13B and 14A to 14I shown.

[0375] 6. Single-arm central-scFv

[0376] One heterodimeric scaffold particularly useful in the antibodies described herein is the one-arm central-scFv format. In this embodiment, one monomer includes only the Fc domain, while the other monomer comprises a Fab domain (first antigen-binding domain), an scFv domain (second antigen-binding domain), and an Fc domain, with the scFv domain inserted between the Fc domains. In this format, the Fab portion binds to one receptor target, and the scFv binds to another receptor target. In this format, the Fab portion binds to ENPP3 and the scFv binds to CD3, or vice versa.

[0377] In this embodiment, one monomer comprises a first heavy chain comprising a first variable heavy chain domain, a CH1 domain, and an Fc domain, wherein the scFv comprises an scFv variable light domain, an scFv linker, and an scFv variable heavy chain domain. The scFv is covalently linked in either orientation between the C-terminus of the CH1 domain of the heavy constant domain and the N-terminus of the first Fc domain using a domain linker (VH1-CH1-[optional domain linker]-VH2-scFv linker-VL2-[optional domain linker]-CH2-CH3 or VH1-CH1-[optional domain linker]-VL2-scFv linker-VH2-[optional domain linker]-CH2-CH3). The second monomer comprises an Fc domain (CH2-CH3). This embodiment further utilizes a light chain comprising a variable light domain and a constant light domain, which associates with the heavy chain to form a Fab.

[0378] For many embodiments herein, these constructs comprise skew variants, pi variants, ablation variants, additional Fc variants, etc., as desired and described herein.

[0379] The antibodies described herein provide a central-Fv format in which the CD3 binding domain sequence is as follows 10A to 10F The ENPP3 binding domain sequence is shown in Figure 12 、 13A to 13B and 14A to 14I shown.

[0380] In addition, the Fc domain of the one-arm central-scFv format typically comprises a skew variant (e.g., a set of amino acid substitutions as shown in FIG1 , wherein particularly useful skew variants are selected from the group consisting of: S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W, and T366S / L368A / Y407V / Y349C:T366W / S354C), optionally ablated variants (e.g., Figure 3 ), optionally charged scFv linkers (comprising Figure 5 ), and the heavy chain includes pi variants (comprising Figure 2 those shown in ).

[0381] In some embodiments, the one-armed Central-scFv format comprises a skew variant, a pi variant, and an ablation variant. Thus, some embodiments of the one-arm central-scFv format include: a) a first monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, and a variable heavy chain domain and a scFv domain, wherein the variable heavy chain domain and the variable light domain of the light chain constitute an Fv that binds to ENPP3 as outlined herein, and the scFv domain binds to CD3; b) a second monomer comprising an Fc domain having the skew variant L368D / K370S, the pI variant N208D / Q295E / N384D / Q418E / N421D, the ablation variant E233P / L234V / L235A / G236del / S267K; and c) a light chain comprising a variable light domain and a constant light domain.

[0382] In some embodiments, the one-arm central-scFv format comprises a skew variant, a pI variant, an ablation variant, and an FcRn variant. Thus, some embodiments of the one-arm central-scFv format comprise: a) a first monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, the FcRn variant M428L / N434S, and a variable heavy chain domain and a scFv domain, wherein the variable heavy chain domain and the variable light domain of the light chain constitute an Fv that binds to ENPP3 as outlined herein, The scFv domain binds to CD3; b) a second monomer comprising an Fc domain with the skew variant L368D / K370S, the pI variant N208D / Q295E / N384D / Q418E / N421D, the ablation variant E233P / L234V / L235A / G236del / S267K, and the FcRn variant M428L / N434S; and c) a light chain comprising a variable light domain and a constant light domain.

[0383] 7. Single-arm scFv-mAb

[0384] One heterodimeric scaffold particularly useful in the antibodies described herein is the one-armed scFv-mAb format. In this embodiment, one monomer contains only the Fc domain, while the other utilizes the scFv domain linked to the N-terminus of the heavy chain, typically through the use of a linker: VH-scFv linker-VL-[optional domain linker]-CH1-hinge-CH2-CH3 or (in the reverse orientation) VL-scFv linker-VH-[optional domain linker]-CH1-hinge-CH2-CH3. In this format, each Fab portion encodes PP3 and the scFv binds CD3. This embodiment further utilizes a light chain comprising a variable light domain and a light chain constant domain, which associates with the heavy chain to form the Fab. For many embodiments herein, these constructs include skew variants, pI variants, ablation variants, additional Fc variants, and the like, as desired and described herein.

[0385] The antibodies described herein provide a one-armed scFv-mAb format, wherein the CD3 binding domain sequence is as follows 10A to 10F As shown, and the ENPP3 binding domain sequence is as shown Figure 12 、 13A to 13B and 14A to 14I shown.

[0386] Furthermore, the Fc domain of the one-armed scFv-mAb format often includes skewed variants (e.g., Figure 3 and 8, and particularly useful skew variants are selected from the group consisting of: S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W and T366S / L368A / Y407V / Y349C:T366W / S354C; optionally ablation variants (including Figure 3 those shown in ); optionally charged scFv linkers (including Figure 5 those shown in ); and heavy chains comprising pI variants (including Figure 2 those shown in ).

[0387] In some embodiments, the one-armed scFv-mAb format comprises a skew variant, a pi variant, and an ablation variant. Thus, some embodiments of the one-armed scFv-mAb format include: a) a first monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K and a variable heavy domain and a scFv domain, wherein the variable heavy domain and the variable light domain of the light chain constitute an Fv that binds to ENPP3 as outlined herein, and the scFv domain binds to CD3; b) a second monomer comprising an Fc domain having the skew variant L368D / K370S, the pI variant N208D / Q295E / N384D / Q418E / N421D, the ablation variant E233P / L234V / L235A / G236del / S267K; and c) a light chain comprising a variable light domain and a constant light domain.

[0388] In some embodiments, the single-arm scFv-mAb format comprises a skew variant, a pI variant, an ablation variant, and an FcRn variant. Thus, some embodiments of the single-arm scFv-mAb format comprise: a) a first monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, the FcRn variant M428L / N434S and a variable heavy domain and a scFv domain, wherein the variable heavy domain and the variable light domain of the light chain constitute an Fv that binds to ENPP3 as outlined herein, wherein the sc The Fv domain binds to CD3; b) a second monomer comprising an Fc domain with the skew variant L368D / K370S, the pI variant N208D / Q295E / N384D / Q418E / N421D, the ablation variant E233P / L234V / L235A / G236del / S267K, and the FcRn variant M428L / N434S; and c) a light chain comprising a variable light domain and a constant light domain.

[0389] 8. scFv-mAb

[0390] One heterodimeric scaffold particularly useful in the antibodies described herein is the mAb-scFv format. In this embodiment, the format relies on the use of a scFv attached to the N-terminus of one of the monomers, thereby forming a third antigen binding domain, wherein the Fab portion of both monomers binds to ENPP3 and the "additional" scFv domain binds to CD3.

[0391] In this embodiment, the first monomer comprises a first heavy chain (comprising a variable heavy domain and a constant domain), wherein the N-terminus covalently linked scFv comprises an scFv variable light domain, an scFv linker, and an scFv variable heavy domain in either orientation ((VH1-scFv linker-VL1-[optional domain linker]-VH2-CH1-hinge-CH2-CH3) or (scFv in the opposite orientation) ((VL1-scFv linker-VH1-[optional domain linker]-VH2-CH1-hinge-CH2-CH3)). This embodiment further utilizes a light chain comprising a variable light domain and a light chain constant domain, which associates with the heavy chain to form two identical Fabs that bind ENPP3. For many embodiments herein, these constructs comprise skew variants, pI variants, ablation variants, additional Fc variants, etc., as desired and described herein.

[0392] The antibodies described herein are provided in the form of scFv-mAbs, wherein the CD3 binding domain sequence is as follows 10A to 10F As shown, and the ENPP3 binding domain sequence is as shown Figure 12 、 13A to 13B and 14A to 14I shown.

[0393] In addition, the Fc domain of the scFv-mAb format typically comprises a skew variant (e.g., a set of amino acid substitutions as shown in FIG1 , wherein particularly useful skew variants are selected from the group consisting of: S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W, and T366S / L368A / Y407V / Y349C:T366W / S354C), optionally ablated variants (e.g., Figure 3 ), optionally charged scFv linkers (comprising Figure 5 ), and the heavy chain includes pi variants (comprising Figure 2 those shown in ).

[0394] In some embodiments, the scFv-mAb format comprises a skew variant, a pI variant, and an ablation variant. Thus, some embodiments comprise a scFv-mAb format comprising: a) a first monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, a variable heavy chain domain, and a scFv domain, wherein the variable heavy chain domain and the variable light domain of the common light chain form an Fv that binds to ENPP3 as outlined herein, and wherein the scFv domain binds to CD3; b) a second monomer The invention also provides a monomer comprising a skew variant L368D / K370S, a pI variant N208D / Q295E / N384D / Q418E / N421D, an ablation variant E233P / L234V / L235A / G236del / S267K and a variable heavy chain domain, wherein the variable heavy chain domain and the variable light domain of the common light chain constitute an Fv that binds to ENPP3 as outlined herein; and c) a common light chain comprising a variable light domain and a constant light domain.

[0395] In some embodiments, the scFv-mAb format comprises a skew variant, a pI variant, an ablation variant, and an FcRn variant. Thus, some embodiments include a scFv-mAb format comprising: a) a first monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, the FcRn variant M428L / N434S, and a variable heavy chain domain (which together with the variable light chain domain of the common light chain constitutes an Fv that binds to ENPP3 as outlined herein and an ScFv domain that binds to CD3); b) a second monomer comprising The skew variant L368D / K370S, the pI variant N208D / Q295E / N384D / Q418E / N421D, the ablation variant E233P / L234V / L235A / G236del / S267K, the FcRn variant M428L / N434S and the variable heavy chain domain (which together with the variable light chain domain of the common light chain constitute an Fv that binds to ENPP3 as outlined herein); and c) a common light chain comprising a variable light chain domain and a light chain constant domain.

[0396] 9. Dual scFv Format

[0397] The antibodies described herein also provide a dual scFv format known in the art. In this embodiment, the ENPP3xCD3 heterodimeric bispecific antibody is composed of two scFv-Fc monomers (both monomers are in the form of (VH-scFv linker-VL-[optional domain linker]-CH2-CH3) or (VL-scFv linker-VH-[optional domain linker]-CH2-CH3), or one monomer in one orientation and the other monomer in another orientation).

[0398] The antibodies described herein provide dual scFv formats, wherein the CD3 binding domain sequences are as follows 10A to 10F As shown, and the ENPP3 binding domain sequence is as shown Figure 12 、 13A to 13B and 14A to 14I As shown. In some embodiments, the dual scFv format comprises a skew variant, a pI variant, and an ablation variant. Thus, some embodiments comprise a dual scFv format comprising: a) a first monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, and a first scFv that binds to CD3 or ENPP3; and b) a second monomer comprising the skew variant L368D / K370S, the pI variant N208D / Q295E / N384D / Q418E / N421D, the ablation variant

[0399] E233P / L234V / L235A / G236del / S267K and a second scFv that binds to CD3 or ENPP3. In some embodiments, the dual scFv format comprises a skew variant, a pI variant, an ablation variant, and an FcRn variant. In some embodiments, the dual scFv format comprises a skew variant, a pI variant, and an ablation variant. Thus, some embodiments comprise a dual scFv format comprising: a) a first monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, the FcRn variant M428L / N434S, and a first scFv that binds to CD3 or ENPP3; and b) a second monomer comprising the skew variant L368D / K370S, the pI variant N208D / Q295E / N384D / Q418E / N421D, the ablation variant E233P / L234V / L235A / G236del / S267K, the FcRn variant M428L / N434S, and a second scFv that binds to CD3 or ENPP3.

[0400] 10. Non-heterodimerable bispecific antibodies

[0401] As will be appreciated by those skilled in the art, the ENPP3 and CD3 Fv sequences outlined herein can also be used in both monospecific antibodies (eg, "traditional monoclonal antibodies") or non-heterodimeric bispecific formats.

[0402] Particularly useful CD3 binding domain sequences include, but are not limited to, H1.30_L1.47, H1.32_L1.47, H1.89_L1.47, H1.90_L1.47, H1.33_L1.47, H1.31_L1.47, L1.47_H1.30, L1.47_H1.30, L1.47_H1.32, L1.47_H1.89, L1.47_H1.90, L1.47_H1.33, and L1.47_H1.31 ( 10A to 10F ).

[0403] Particularly useful ENPP3 binding domain sequences include, but are not limited to, AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B and 14A to 14I ).

[0404] Suitable non-heterodimeric bispecific formats are known in the art and include a number of different formats as generally described in Spiess et al., Molecular Immunology (67):95-106 (2015) and Kontermann, mAbs 4:2,182-197 (2012), both of which are expressly incorporated by reference, particularly with respect to the drawings, figure legends, and references to the formats therein.

[0405] 11. Trident form

[0406] In some embodiments, the bispecific antibodies described herein are in a "tripartite" format as generally described in WO2015 / 184203, which is hereby expressly incorporated by reference in its entirety, particularly with respect to the figures, legends, definitions, and sequences of the "Heterodimer-Promoting Domain" or "HPD" (including "K-helix" and "E-helix" sequences). The tripartite format relies on the use of two different HPDs that associate to form a heterodimer structure as components of the structure, see Figure 1K. In this embodiment, the tripartite format comprises "traditional" heavy and light chains (e.g., VH1-CH1-hinge-CH2-CH3 and VL1-CL), a third chain comprising a first "diabody-type binding domain" or "DART®," VH2-(linker)-VL3-HPD1, and a fourth chain comprising a second DART®, VH3-(linker)-(linker)-VL2-HPD2. VH1 and VL1 form a first ABD, VH2 and VL2 form a second ABD, and VH3 and VL3 form a third ABD. In some cases, as shown in FIG1K , the second and third ABDs bind to the same antigen, typically ENPP3 in this case, e.g., bivalently, wherein the first ABD binds monovalently to CD3.

[0407] 12. Monospecific monoclonal antibodies

[0408] As will be appreciated by those skilled in the art, the novel Fv sequences outlined herein can also be used in both monospecific antibodies (e.g., "traditional monoclonal antibodies") or non-heterodimer bispecific formats. Thus, in some embodiments, the antibodies described herein provide monoclonal (monospecific) antibodies comprising the six CDRs and / or the vh and vl sequences from the accompanying figures, typically with an IgG1, IgG2, IgG3, or IgG4 constant region, with IgG1, IgG2, and IgG4 (including IgG4 constant regions comprising the S228P amino acid substitution) being particularly useful in some embodiments. That is, any sequence designated "H_L" herein can be linked to the constant region of a human IgG1 antibody.

[0409] In some embodiments, the monospecific antibody is an ENPP3 monospecific antibody. In certain embodiments, the monospecific anti-ENPP3 antibody comprises six CDRs of any anti-ENPP3 antibody selected from the group consisting of: AN1[ENPP3] H1 L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1 L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha16-1(1)23. (3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha16-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19, and Ha16-1.80 ( Figure 12 、 13A to 13B ,and 14A to 14I ).

[0410] H. Antigen binding domain

[0411] As discussed herein, the subject heterodimeric antibodies comprise two antigen binding domains (ABDs), each binding to ENPP3 or CD3. As outlined herein, these heterodimeric antibodies can be bispecific and bivalent (e.g., Figure 15A in the depicted formats, with each antigen bound to a single ABD) or bispecific and trivalent (e.g. Figure 15B As depicted in , one antigen binds to a single ABD and the other antigen binds to two).

[0412] In addition, typically one of the ABDs comprises an scFv in an N- to C-terminal orientation as outlined herein, either VH-scFv linker-VL or VL-scFv linker-VH. Depending on the format, one or both of the other ABDs are typically Fabs comprising a VH domain on one protein chain (usually as a component of the heavy chain) and a VL on another protein chain (usually as a component of the light chain).

[0413] The present disclosure provides a variety of ABDs that bind to a variety of different checkpoint proteins, as outlined below. As will be understood by those skilled in the art, any set of six CDRs or VH and VL domains can be in scFv or Fab form, which is then added to the heavy constant domain and the light constant domain, wherein the heavy constant domain includes variants (contained within the CH1 domain and the Fc domain). The scFv sequences contained in the sequence listing utilize specific charged linkers, while as outlined herein, uncharged or other charged linkers can be used, including those described in Figure 7.

[0414] Additionally, as discussed above, the numbering used in the sequence listing to identify the CDRs is Kabat, however, different numbering may be used which would alter the amino acid sequences of the CDRs as shown in Table 2.

[0415] For all variable heavy and light domains listed herein, additional variants can be prepared. As outlined herein, in some embodiments, a set of 6 CDRs can have 0, 1, 2, 3, 4, or 5 amino acid modifications (where amino acid substitutions are particularly useful) and changes in the framework regions of the variable heavy and light domains, as long as the framework (excluding the CDRs) remains at least about 80%, 85%, or 90% identical to a human germline sequence selected from those sequences listed in U.S. Patent No. 7,657,380 in Figure 1, the drawings and legends of which are incorporated herein by reference in their entirety. Thus, for example, the same CDRs as described herein can be combined with different framework sequences from human germline sequences, as long as the framework regions remain at least 80%, 85%, or 90% identical to a human germline sequence selected from those sequences listed in U.S. Patent No. 7,657,380 in Figure 1. Alternatively, the CDRs can have amino acid modifications (e.g., 1, 2, 3, 4, or 5 amino acid modifications in a set of CDRs (that is, the CDRs can be modified as long as the total number of changes in a set of 6 CDRs is less than 6 amino acid modifications, wherein any combination of CDRs is altered; for example, there can be one alteration in VLCDR1, two alterations in VHCDR2, no alterations in VHCDR3, etc.)) and have framework region alterations, so long as the framework regions retain at least 80%, 85%, or 90% identity to a human germline sequence selected from those sequences listed in Figure 1 of U.S. Patent No. 7,657,380.

[0416] 1. ENPP3 antigen-binding domain

[0417] In some embodiments, one of the ABDs binds ENPP3. Figure 12 、 13A to 13B and 14A to 14ISuitable groups of six CDRs and / or VH and VL domains are depicted in . In some embodiments, the heterodimeric antibody is an antibody in the form of 1+1 Fab-scFv-Fc or 2+1 Fab2-scFv-Fv (see, e.g., Figure 15A and Figure 15B ).

[0418] In one embodiment, the ENPP3 antigen binding domain comprises the six CDRs (ie, vhCDR1-3 and vlCDR1-3) of the ENPP3 ABD described herein, including the Figures and Sequence Listing. In exemplary embodiments, the ENPP3 ABD is one of the following ENPP3 ABDs: AN1[ENPP3] H1L1, AN1[ENPP3] H1 L1.33, AN1[ENPP3] H1L1.77, AN1[ENPP3] H1.8 L1, AN1[ENPP3] H1.8 L1.33, AN1[ENPP3] H1 L1.77, H16-7.213, H16-9.69, H16-1.52, Ha16-1(1)23, H16-9.44, H16-1.67, Ha1 6-1(3,5)36, H16-1.86, H16-9.10, H16-9.33, H16-1.68, Ha16-1(1)1, Ha1 6-1(3,5)18, Ha16-1(2,4)4, Ha16-1(3,5)56, H16-7.8, H16-1.93, Ha1 6-1(3,5)27.1, H16-1.61, H16-1(3,5)5, ​​H16-7.200, H16-1(3,5)42, H1 6-9.65, Ha1-1(3,5)19 and Ha16-1.80 ( Figure 12 、 13A to 13B and 14A to 14I ).

[0419] As will be understood by those skilled in the art, a suitable ENPP3 binding domain may include a set of six CDRs as depicted in the Figures or as underlined, or, where a different numbering scheme is used as described herein and as shown in Table 2, using the CDRs in Table 2. Figure 12 、 13A to 13B and 14A to 14I Suitable ABDs may also comprise the entire VH and VL sequences used as scFv or Fab as depicted in these sequences and the accompanying figures. In many embodiments herein containing an Fv to ENPP3, it is the Fab monomer that binds to ENPP3.

[0420] In addition to the parent CDR sets disclosed in the Figures and Sequence Listing that form the ABD to ENPP3, the present disclosure provides variant CDR sets. In one embodiment, a set of six CDRs may have one, two, three, four, or five amino acid changes from the parent CDRs, as long as the ENPP3 ABD is still capable of binding to the target antigen as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biolayer interferometry, such as the Octet assay), the latter being particularly useful in many embodiments.

[0421] In addition to the parent variable heavy chain domain and variable light domain disclosed herein that form an ABD to ENPP3, the present disclosure provides variant VH and VL domains. In one embodiment, the variant VH and VL domains can each have 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10 amino acid changes from the parent VH and VL domains, as long as the ABD is still capable of binding to the target antigen as measured by at least one of Biacore, surface plasmon resonance (SPR) and / or BLI (biolayer interferometry, such as the Octet assay), the latter being particularly useful in many embodiments. In another embodiment, the variant VH and VL are at least 90%, 95%, 97%, 98% or 99% identical to the corresponding parent VH or VL, so long as the ABD is still capable of binding to the target antigen as measured by at least one of a Biacore, surface plasmon resonance (SPR) and / or BLI (biolayer interferometry, e.g., Octet assay) assay, the latter being particularly useful in many embodiments.

[0422] 2. CD3 antigen-binding domain

[0423] In some embodiments, one of the ABDs binds to CD3. 10A to 10F Suitable sets of six CDRs and / or VH and VL domains and scFv sequences are depicted in the sequence listing. Particularly useful CD3 binding domain sequences include, but are not limited to, anti-CD3 H1.30_L1.47, anti-CD3 H1.32, anti-CD3 L1.47, anti-CD3 H1.89_L1.47, anti-CD3 H1.90_L1.47, anti-CD3 H1.33_L1.47, anti-CD3 H1.31_L1.47, anti-CD3 L1.47_H1.30, anti-CD3 L1.47_H1.30, anti-CD3 L1.47_H1.32, anti-CD3 L1.47_H1.89, anti-CD3 L1.47_H1.90, anti-CD3 L1.47_H1.33, and anti-CD3 L1.47_H1.31, as described herein. 10A to 10Fshown.

[0424] As will be appreciated by those skilled in the art, suitable CD3 binding domains may include, for example, 10A to 10F A set of 6 CDRs as depicted in or as underlined in FIG, or where a different numbering scheme is used as described herein and as shown in Table 2, is used. 10A to 10F The CDRs identified by other alignments within the VH and VL sequences of those depicted in the Figures. Suitable ABDs may also comprise the entire VH and VL sequences used as scFv or Fab as depicted in these sequences and the accompanying figures. In many embodiments herein containing an Fv to CD3, it is the scFv monomer that binds to CD3.

[0425] In addition to the parent CDR sets that form ABDs to CD3 disclosed in the Figures and Sequence Listing, the present disclosure provides variant CDR sets. In one embodiment, a set of six CDRs may have one, two, three, four, or five amino acid changes from the parent CDRs, as long as the CD3 ABD is still capable of binding to the target antigen as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biolayer interferometry, such as the Octet assay), the latter being particularly useful in many embodiments.

[0426] In addition to the parent variable heavy chain domain and variable light domain disclosed herein that form an ABD to CD3, the present disclosure provides variant VH and VL domains. In one embodiment, the variant VH and VL domains can each have 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10 amino acid changes from the parent VH and VL domains, as long as the ABD is still capable of binding to the target antigen as measured by at least one of Biacore, surface plasmon resonance (SPR) and / or BLI (biolayer interferometry, such as the Octet assay), the latter being particularly useful in many embodiments. In another embodiment, the variant VH and VL are at least 90%, 95%, 97%, 98% or 99% identical to the corresponding parent VH or VL, so long as the ABD is still capable of binding to the target antigen as measured by at least one of a Biacore, surface plasmon resonance (SPR) and / or BLI (biolayer interferometry, e.g., Octet assay) assay, the latter being particularly useful in many embodiments.

[0427] VI. SSTR2 Binding Domain

[0428] In one aspect, provided herein are somatostatin receptor 2 (SSTR2) antigen binding domains (ABDs) and compositions comprising such SSTR2 antigen binding domains (ABDs), including anti-SSTR2 antibodies.

[0429] Somatostatin receptors (SSTRs) belong to the G protein-coupled receptor (GPCR) superfamily. Each GPCR consists of a single polypeptide chain consisting of extracellular / intracellular domains and seven transmembrane domains. SSTRs are highly expressed in a variety of cultured tumor cells and primary tumor tissues, including NETs (lung cancer, gastrointestinal (GI) cancer, pancreatic cancer, pituitary cancer, medullary carcinoma, prostate cancer, pancreatic lung cancer, osteosarcoma, etc.) and non-NET cancers (breast cancer, lung cancer, colorectal cancer, ovarian cancer, cervical cancer, etc.) (Reubi et al., 2003, Endocr. Rev. 24: 389-427; Volante et al., 2008, Mol. Cell. Endocrinol. 286: 219-229; and Schulz et al., 2003, Gynecol. Oncol. 89: 385-390). To date, five SSTR receptor subtypes have been identified (Patel et al., 1997, Trends Endocrinol. Metab. 8: 398-405). In particular, SSTR2 is expressed at high concentrations on many tumor cells (Volante et al., 2008, Mol. Cell. Endocrinol. 286: 219-229; and Reubi et al., 2003, Eur. J. Nucl. Med. Mol. Imaging 30: 781-793), making it a candidate target antigen for bispecific antibody cancer-targeted therapies. Given the high concentrations of SSTR2 expressed on various tumors, it is believed that anti-SSTR2 antibodies can be used, for example, to localize anti-tumor therapeutics (e.g., chemotherapeutic agents and T cells) to such SSTR2-expressing tumors.

[0430] The subject antibodies comprising the SSTR2 antigen binding domains provided herein (e.g., anti-SSTR2 x anti-CD3 bispecific antibodies) advantageously elicit a range of different immune responses. Such SSTR2 binding domains and related antibodies can be used, for example, to treat SSTR2-associated cancers.

[0431] As will be appreciated by those skilled in the art, suitable SSTR2 binding domains may include, for example, Figure 63 A set of 6 CDRs as depicted in or as underlined in FIG, or where a different numbering scheme is used as described herein and as shown in Table 2, is used. Figure 63CDRs identified by other alignments within the VH and VL sequences of those depicted in the Figures. Suitable ABDs may also comprise the entire VH and VL sequences used as scFv or Fab as depicted in these sequences and the Figures. In many embodiments herein comprising an Fv to SSTR2, it is the Fab monomer that binds to SSTR2. In one embodiment, the SSTR2 antigen binding domain comprises the six CDRs of [αSSTR2]H1.24_L1.30 (i.e., vhCDR1-3 and vlCDR1-3) ( Figure 63 ).

[0432] In addition to the parent CDR sets that form ABDs for SSTR2 disclosed in the Figures and Sequence Listing, provided herein are SSTR2 ABD variants having CDRs that include at least one modification of the SSTR2 ABD CDRs disclosed herein. In one embodiment, the SSTR2 ABD comprises a set of six CDRs that have one, two, three, four, five, six, seven, eight, nine, or ten amino acid modifications compared to the six CDRs of the SSTR2 ABD depicted herein, including in the Figures and Sequence Listing. In an exemplary embodiment, the SSTR2 ABD comprises a set of six CDRs that have one, two, three, four, five, six, seven, eight, nine, or ten amino acid modifications compared to the six CDRs of [αSSTR2] H1.24_L1.30 ( Figure 63 In certain embodiments, the variant SSTR2 ABD is capable of binding to an SSTR2 antigen as determined by at least one of a Biacore, surface plasmon resonance (SPR), and / or BLI (bio-layer interferometry, e.g., an Octet assay), the latter assay being particularly useful in many embodiments. In specific embodiments, the SSTR2 ABD is capable of binding to a human SSTR2 antigen.

[0433] In one embodiment, the SSTR2 ABD comprises six CDRs that are at least 90%, 95%, 97%, 98%, or 99% identical to the six CDRs of the SSTR2 ABD as described herein, including the Figures and Sequence Listing. In exemplary embodiments, the SSTR2 ABD comprises six CDRs that are at least 90%, 95%, 97%, 98%, or 99% identical to the six CDRs of [αSSTR2] H1.24_L1.30 ( Figure 63 In certain embodiments, the SSTR2 ABD is capable of binding to an SSTR2 antigen as determined by at least one of a Biacore, surface plasmon resonance (SPR), and / or a BLI (bio-layer interferometry, such as an Octet assay), the latter assay being particularly useful in many embodiments. In specific embodiments, the SSTR2 ABD is capable of binding to a human SSTR2 antigen.

[0434] In another exemplary embodiment, the SSTR2 ABD comprises the variable heavy (VH) domain and the variable light (VL) domain of any of the ENPP3 ABDs described herein, including the Figures and Sequence Listing. In an exemplary embodiment, the SSTR2 ABD is [αSSTR2] H1.24_L1.30 ( Figure 63 ). .

[0435] In addition to the parent SSTR2 variable heavy and variable light domains disclosed herein, provided herein are SSTR2 ABDs comprising a variable heavy domain and / or variable light domain that is a variant of the SSTR2 ABD VH and VL domains disclosed herein. In one embodiment, the variant VH domain and / or VL domain has 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid changes compared to the VH and / or VL domains of the SSTR2 ABDs described herein, including the figures and sequence listing. In one embodiment, the variant VH domain and / or VL domain has 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid changes relative to the VH and / or VL domains of [αSSTR2] H1.24_L1.30 ( Figure 63 In certain embodiments, the SSTR2 ABD is capable of binding to SSTR2 as determined by at least one of a Biacore, surface plasmon resonance (SPR), and / or BLI (bio-layer interferometry, e.g., an Octet assay), the latter assay being particularly useful in many embodiments. In specific embodiments, the SSTR2 ABD is capable of binding to a human SSTR2 antigen.

[0436] In one embodiment, the variant VH and / or VL domains are at least 90%, 95%, 97%, 98% or 99% identical to the VH and / or VL of the SSTR2 ABD as described herein, including the Figures and the Sequence Listing. In exemplary embodiments, the variant VH and / or VL domains are at least 90%, 95%, 97%, 98% or 99% identical to the VH and / or VL of [αSSTR2] H1.24_L1.30 ( Figure 63 In certain embodiments, the SSTR2 ABD is capable of binding to SSTR2 as determined by at least one of a Biacore, surface plasmon resonance (SPR), and / or BLI (bio-layer interferometry, e.g., an Octet assay), the latter assay being particularly useful in many embodiments. In specific embodiments, the SSTR2 ABD is capable of binding to a human SSTR2 antigen.

[0437] In some embodiments, the subject antibodies described herein comprise at least one SSTR2 binding domain. In certain embodiments, the antibodies are heterodimeric antibodies. In some embodiments, the heterodimeric antibodies are antibodies in the form of 1+1 Fab-scFv-Fc or 2+1 Fab2-scFv-Fv (see, e.g., Figure 15A and Figure 15B Such heterodimeric antibodies may comprise any of the Fc variant amino acid substitutions provided herein, either individually or in combination (e.g., skew, pi, and ablation variants, including those of Figures 1 to 2). Figure 4 Particularly useful skew variants are selected from the group consisting of: S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L; K370S:S364K / E357Q; T366S / L368A / Y407V:T366W and T366S / L368A / Y407V / Y349C:T366W / S354C), optionally ablative variants (comprising Figure 3 ), optionally charged scFv linkers (comprising Figure 5 ), and the heavy chains include pi variants (comprising Figure 2 those shown in ).

[0438] A. Useful implementation methods

[0439] Useful embodiments include a 1+1 Fab-scFv-Fc format comprising: a) a first monomer ("scFv monomer") comprising a charged scFv linker ( Figure 5 The invention further comprises a) a first monomer comprising a skew variant L368D / K370S, a pI variant N208D / Q295E / N384D / Q418E / N421D, a second monomer comprising a skew variant L368D / K370S, a pI variant N208D / Q295E / N384D / Q418E / N421D, a second monomer comprising a skew variant E233P / L234V / L235A / G236del / S267K, and a variable heavy chain domain; and c) a light chain comprising a variable light chain domain (VL) and a constant light chain domain (CL), wherein numbering is according to EU numbering. In some embodiments, the variable heavy chain domain and the variable light chain domain constitute the ENPP3 binding portion.

[0440] Other useful embodiments include a 1+1 Fab-scFv-Fc format comprising: a) a first monomer ("scFv monomer") comprising a charged scFv linker ( Figure 5 The invention further comprises the following: a) a first monomer comprising a skew variant L368D / K370S, a pI variant N208D / Q295E / N384D / Q418E / N421D, a second monomer comprising a variable heavy chain domain (VL) and a constant light chain domain (CL), wherein numbering is according to EU numbering. In some embodiments, the variable heavy chain domain and the variable light chain domain constitute the SSTR2 binding portion.

[0441] Other useful embodiments include a 2+1 Fab2-scFv-Fc format comprising: a) a first monomer (Fab-scFv-Fc side) comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K and a variable heavy chain domain, which together with the variable light chain domain of the common light chain constitutes an Fv that binds to ENPP3 as outlined herein and an scFv domain that binds to CD3; b) a second monomer (Fab-Fc side) comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K and a variable heavy chain domain; The second monomer includes the skew variant L368D / K370S, the pI variant N208D / Q295E / N384D / Q418E / N421D, the ablation variant E233P / L234V / L235A / G236del / S267K and a variable heavy chain domain, which together with the variable light chain domain of the common light chain constitutes an Fv that binds to ENPP3 as described herein; and c) a light chain comprising a variable light domain and a constant light domain, wherein numbering is according to EU numbering. In some embodiments, the common light chain and variable heavy chain domain on each monomer form an ENPP3 binding domain.

[0442] Other useful embodiments include a 2+1 Fab2-scFv-Fc format comprising: a) a first monomer (Fab-scFv-Fc side) comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K and a variable heavy chain domain, which together with the variable light chain domain of the common light chain constitutes an Fv binding to SSTR2 as outlined herein and an scFv domain binding to CD3; b) a second monomer (Fab-Fc side) comprising the variable heavy chain domain of the common light chain; The second monomer includes the skew variant L368D / K370S, the pI variant N208D / Q295E / N384D / Q418E / N421D, the ablation variant E233P / L234V / L235A / G236del / S267K, and a variable heavy chain domain, which together with the variable light chain domain of the common light chain constitutes an Fv that binds to SSTR2 as described herein; and c) a light chain comprising a variable light domain and a constant light domain, wherein numbering is according to EU numbering. In some embodiments, the common light chain and variable heavy chain domain on each monomer form an SSTR2 binding domain (e.g., [αSSTR2] H1.24_L1.30 ( Figure 63 )).

[0443] Some useful embodiments include: XENP24804, XENP26820, XENP28287, XENP28925, XENP29516, XENP30262, XENP26821, XENP29436, XENP28390, XENP29463, and XENP30263.

[0444] Other useful embodiments include: XENP29437, XENP29520, XENP30264, XENP26822, XENP28438, XENP29438, XENP29467, XENP30469, XENP30470, XENP30819, XENP30821, XENP31148, XENP31149, XENP31150, XENP31419, and XENP31471.

[0445] Another useful embodiment is XENP30458.

[0446] VII. Nucleic Acids of the Invention

[0447] The present disclosure also provides nucleic acid compositions encoding the anti-ENPP3 antibodies provided herein, including but not limited to anti-ENPP3 x anti-CD3 bispecific antibodies and ENPP3 monospecific antibodies.

[0448] As will be appreciated by those skilled in the art, the nucleic acid composition will depend on the format and scaffold of the heterodimeric protein. Thus, for example, when the format requires three amino acid sequences (e.g., a 1+1 Fab-scFv-Fc format (e.g., the first amino acid monomer comprises the Fc domain and scFv, and the second amino acid monomer comprises the heavy and light chains)), the three nucleic acid sequences can be incorporated into one or more expression vectors for expression. Similarly, some formats (e.g., the dual scFv format disclosed in Figure 1) require only two nucleic acid sequences; similarly, these can be placed into one or two expression vectors.

[0449] As is known in the art, depending on the host cell used to produce the heterodimeric antibodies described herein, nucleic acids encoding components of the antibodies described herein can be incorporated into expression vectors as known in the art. Typically, the nucleic acid is operably linked to any number of regulatory elements (promoters, origins of replication, selectable markers, ribosome binding sites, inducers, etc.). The expression vector can be an extrachromosomal or integrative vector.

[0450] The nucleic acids and / or expression vectors for the antibodies described herein are then transformed into any number of different types of host cells well known in the art, including mammalian, bacterial, yeast, insect, and / or fungal cells, with mammalian cells (e.g., CHO cells) being used in many embodiments.

[0451] In some embodiments, if applicable depending on the format, the nucleic acid encoding each monomer and the optional nucleic acid encoding the light chain are each contained in a single expression vector, typically under different or identical promoter control conditions. In embodiments particularly for antibodies described herein, each of the two or three nucleic acids is contained on a different expression vector. As shown herein and in 62 / 025,931 (hereby incorporated by reference), different vector ratios can be used to drive heterodimer formation. That is, surprisingly, although the protein includes a 1:1:2 ratio of first monomer:second monomer:light chain (in the case of many embodiments herein having three polypeptides comprising heterodimeric antibodies), these are not the ratios that yield the best results.

[0452] Heterodimeric antibodies as described herein are prepared by culturing host cells comprising one or more expression vectors as well known in the art. Once produced, conventional antibody purification steps are performed, including ion exchange chromatography steps. As discussed herein, making the pI of the two monomers differ by at least 0.5 allows separation to be achieved by ion exchange chromatography or isoelectric focusing or other methods sensitive to the isoelectric point. In other words, the pI substitution comprising changing the isoelectric point (pI) of each monomer allows each monomer to have a different pI, and the heterodimer also has a different pI, thereby promoting the isoelectric purification (e.g., anion exchange column, cation exchange column) of "1+1 Fab-scFv-Fc" and "2+1" heterodimers. These substitutions also help to measure and monitor any contaminated dual scFv-Fc and purified mAb homodimers (e.g., IEF gel, cIEF, and analytical IEX columns).

[0453] VIII. Biological and Biochemical Functions of Heterodimeric Bispecific Antibodies

[0454] Typically, a bispecific ENPP3 x CD3 antibody as described herein is administered to a patient with cancer, and efficacy is assessed in a variety of ways as described herein. Thus, while standard efficacy assays such as assessment of cancer burden, tumor size, presence or extent of metastasis, etc., can be performed, immuno-oncology treatments can also be evaluated based on immune status assessment. This can be accomplished in a variety of ways, including in vitro and in vivo assays.

[0455] IX. Treatment

[0456] Once prepared, the compositions of the antibodies described herein can be used in a number of applications. ENPP3 is highly expressed in renal cell carcinoma, and thus, the heterodimeric compositions of the antibodies described herein can be used to treat such ENPP3-positive cancers.

[0457] X. Antibody Compositions for In Vivo Administration

[0458] Formulations of antibodies for use according to the present invention are prepared by mixing the antibodies having the desired degree of purity with optional pharmaceutically acceptable carriers, excipients, or stabilizers (Remington's Pharmaceutical Sciences, 16th ed., Osol, A., ed.

[1980] ) for storage as lyophilized formulations or aqueous solutions. Acceptable carriers, excipients, or stabilizers are nontoxic to recipients at the dosages and concentrations employed and include buffers such as phosphates, citrates, and other organic acids; antioxidants including ascorbic acid and methionine; preservatives (such as octadecyldimethylbenzylammonium chloride; hexahydroxyquaternium chloride; benzalkonium chloride, benzethonium chloride; phenol, butyl alcohol, or benzyl alcohol; alkyl parabens such as methyl or propyl paraben; catechol; resorcinol; cyclohexanol; 3-pentanol; and m-cresol); low molecular weight (less than about 10 residues) polypeptides; proteins , such as serum albumin, gelatin, or immunoglobulins; hydrophilic polymers, such as polyvinylpyrrolidone; amino acids, such as glycine, glutamine, asparagine, histidine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates, including glucose, mannose, or dextrins; chelating agents, such as EDTA; sugars, such as sucrose, mannitol, trehalose, or sorbitol; salt-forming counterions, such as sodium; metal complexes (e.g., zinc protein complexes); and / or nonionic surfactants, such as TWEEN™, PLURONICS™, or polyethylene glycol (PEG).

[0459] XI. Mode of Administration

[0460] The antibodies and chemotherapeutic agents described herein are administered to a subject according to known methods, such as intravenous administration (either as a bolus or by continuous infusion over a period of time).

[0461] XII. Treatment

[0462] In the methods described herein, therapy is used to provide a positive therapeutic response to a disease or condition. A "positive therapeutic response" is intended to improve the disease or condition and / or improve symptoms associated with the disease or condition. For example, a positive therapeutic response would refer to one or more of the following improvements in a disease: (1) a decrease in the number of neoplastic cells; (2) an increase in neoplastic cell death; (3) inhibition of neoplastic cell survival; (5) inhibition (i.e., slowing to some extent and preferably stopping) of tumor growth; (6) an increase in patient survival; and (7) some relief of one or more symptoms associated with the disease or condition.

[0463] A positive treatment response for any given disease or condition can be determined by standardized response criteria specific for that disease or condition. Tumor responses to changes in tumor morphology (i.e., overall tumor burden, tumor size, etc.) can be assessed using screening techniques such as magnetic resonance imaging (MRI) scans, x-ray imaging, computed tomography (CT) scans, bone scan imaging, endoscopy, and tumor biopsy sampling, including bone marrow aspirate (BMA) and counting circulating tumor cells.

[0464] In addition to these positive therapeutic responses, subjects undergoing therapy may experience the beneficial effect of improvement in symptoms associated with the disease.

[0465] Treatment according to the present disclosure involves the use of a "therapeutically effective amount" of the drug. A "therapeutically effective amount" refers to an amount effective, measured at dosages and for periods of time necessary, to achieve the desired therapeutic result.

[0466] A therapeutically effective amount may vary according to factors such as the disease state, age, sex, and weight of the individual, and the ability of the drug to elicit a desired response in the individual. A therapeutically effective amount is also one in which any toxic or detrimental effects of the antibody or antibody portion outweigh the therapeutically beneficial effects.

[0467] A "therapeutically effective amount" for tumor treatment can also be measured by its ability to stabilize disease progression. The ability of a compound to inhibit cancer can be assessed in animal model systems that are predictive of efficacy in human tumors.

[0468] Alternatively, this property of the composition can be assessed by examining the compound's ability to inhibit tumor growth or induce apoptosis, using in vitro assays known to skilled practitioners. A therapeutically effective amount of a therapeutic compound can reduce tumor size or otherwise alleviate symptoms in a subject. One of ordinary skill in the art will be able to determine this amount based on factors such as the size of the subject, the severity of the subject's symptoms, and the particular composition or route of administration selected.

[0469] The dosage regimen is adjusted to provide the optimal desired response (e.g., a therapeutic response). For example, a single bolus may be administered, or several divided doses may be administered over time, or the dose may be proportionally reduced or increased as indicated by the exigencies of the therapeutic situation. For ease of administration and uniformity of dosage, parenteral compositions may be formulated in dosage units. As used herein, a dosage unit refers to a physically discrete unit suitable as a single dosage for the subject to be treated; each unit contains a predetermined amount of active compound calculated to produce the desired therapeutic effect, in association with a desired pharmaceutical carrier.

[0470] The specification for the dosage unit forms of the present disclosure are dictated by and directly dependent on (a) the unique characteristics of the active compound and the particular therapeutic effect to be achieved, and (b) the limitations inherent in the art imposed by individual sensitivities to compounding such an active compound for therapy.

[0471] Effective dosages and dosage regimens for the bispecific antibodies described herein depend on the disease or condition to be treated and can be determined by one skilled in the art.

[0472] An exemplary, non-limiting range for a therapeutically effective amount of a bispecific antibody for use in the antibodies described herein is about 0.1-100 mg / kg.

[0473] All cited references are expressly incorporated herein by reference in their entirety.

[0474] While specific embodiments of the present disclosure have been described above for purposes of illustration, those skilled in the art will appreciate that numerous changes in detail may be made without departing from the invention as described in the appended claims.

[0475] Example

[0476] Provide embodiment below to illustrate antibody as herein described.These embodiments are not intended to limit antibody as herein described to any specific application or theory of operation.For all constant region positions discussed in antibody as herein described, according to the EU index in as Kabat (Kabat etc., 1991, Sequences of Proteins of Immunological Interest, the 5th edition, Bethesda U.S. National Institutes of Health Public Health Service (United States Public Health Service, National Institutes of Health, Bethesda), it is incorporated herein by reference in its entirety) number.Technicians in the field of antibodies will understand that this convention is made up of non-sequential numbering in the specific region of immunoglobulin sequence, and described non-sequential numbering makes it possible to carry out standardized reference to the conserved position in immunoglobulin family.Therefore, the position of any given immunoglobulin as defined by EU index does not necessarily correspond to its sequential sequence.

[0477] General and specific scientific techniques are summarized in US Publication Nos. 2015 / 0307629, 2014 / 0288275, and WO 2014 / 145806, which are all expressly incorporated herein by reference in their entireties, and particularly for the techniques summarized therein.

[0478] Example 1: Binding Domain

[0479] 1A: CD3 binding domain

[0480] The sequences of CD3 binding domains with different CD3 binding affinities are shown in FIG10 .

[0481] 1B: ENPP3 binding domain

[0482] 1B(a): ENPP3 binding domain AN1

[0483] The variable region of the murine ENPP3 binding domain was humanized using string content optimization (see, e.g., U.S. Patent No. 7,657,380, issued February 2, 2010). The sequence of the humanized ENPP3 binding domain, referred to herein as AN1, is as follows: 10A to 10F shown.

[0484] AN1 variants were engineered for improved purification (in the context of an αENPP3 x αCD3 bispecific antibody) and to modulate ENPP3 binding affinity / potency. The sequences of illustrative such variants are depicted in FIG13 .

[0485] 1B(b): Additional ENPP3 binding domain

[0486] The sequences of additional ENPP3 binding domains that can be used in the αENPP3 x αCD3 bispecific antibodies described herein are depicted in FIG14 .

[0487] Example 2: Engineering and Production of αENPP3 x αCD3 Bispecific Antibodies

[0488] Various formats of αENPP3 x αCD3 bispecific antibodies (bsAbs) are envisioned, illustrative formats of which are outlined below and in FIG15 .

[0489] One such format is the 1+1 Fab-scFv-Fc format, which comprises a single-chain Fv ("scFv") covalently attached to a first heterodimeric Fc domain, a heavy chain variable region (VH) covalently attached to a complementary second heterodimeric Fc domain, and a light chain (LC) that is transfected separately so that the Fab domain is formed together with the variable heavy chain domain.

[0490] Another format is the 2+1 Fab2-scFv-Fc format, which comprises a VH domain covalently attached to a CH1 domain covalently attached to a scFv covalently attached to a first heterodimeric Fc domain (VH-CH1-scFv-Fc); a VH domain covalently attached to a complementary second heterodimeric Fc domain; and LC, which is transfected separately such that the Fab domain is formed together with the VH domain.

[0491] The DNA coding strand of the αENPP3 x αCD3 bsAb was generated by standard gene synthesis followed by isothermal cloning (Gibson assembly) or subcloning into a vector containing a fusion partner (e.g. Figure 6 The domain linkers depicted in FIG7 and / or FIG7 Figure 9 The DNA was transfected into the pTT5 expression vector (with the backbone depicted in Figure 1). The DNA was expressed in HEK293E cells. The sequences of illustrative αENPP3 x αCD3 bsAbs (based on the binding domains described in Example 1) in the 1+1 Fab-scFv-Fc format and the 2+1 Fab2-scFv-Fc format are depicted in Figures 17 to 23, respectively.

[0492] Example 3: αENPP3 x αCD3 bsAb redirects T cells to destroy ENPP3-expressing cells

[0493] A prototype αENPP3 xαCD3 bsAb in a 1+1 Fab-scFv-Fc format was engineered using the binding domains described in Example 1. Specifically, the sequences of XENP26820 (comprising the ENPP3 binding domain clone H16-7.8 and CD3 high scFv), XENP26821 (comprising the ENPP3 binding domain clone H16-7.8 and CD3 high-Int No. 1 scFv), XENP28287 (comprising the ENPP3 binding domain clone AN1 and CD3 scFv), and XENP28390 (comprising the ENPP3 binding domain clone AN1 and CD3 high Int No. 1 scFv) are depicted in Figures 17 and 18. XENP13245 (comprising the RSV binding domain based on motavizumab and anti-CD3 high; the sequence is depicted in Figures 17 and 18). Figure 16 middle) were used as controls.

[0494] Research prototype αENPP3 x αCD3 bispecific antibody (bsAb) redirects CD3 + The potential of effector T cells to destroy cell lines expressing ENPP3. In the first experiment, KU812 (ENPP3 高Basophilic leukemia cell lines) were incubated with human PBMCs (at an effector to target ratio of 10:1) and the indicated concentrations of the test articles as described above at 37°C for 24 hours. Following incubation, the cells were stained with Aqua Zombie dye for 15 minutes at room temperature. The cells were then washed and stained with antibodies against cell surface markers and analyzed by flow cytometry. Two different approaches were used to investigate the induction of redirected T cell cytotoxicity (RTCC): a) Reduction of the number of CSFE+ target cells (data depicted in Figure 2) Figure 24A ), and b) Zombie Aqua staining of CSFE+ target cells (data depicted in Figure 24B CD4 + and CD8 + T cell activation and degranulation were also determined based on CD107a, CD25, and CD69 expression (data for which are depicted in Figures 25-26).

[0495] In a second experiment, RXF393 (a clinically relevant renal cell carcinoma cell line expressing ENPP3) cells were incubated with human PBMCs (at an effector to target cell ratio of 20:1) and the indicated concentrations of the prototype test articles as described above for 24 hours at 37°C. Following incubation, the cells were stained with Aqua Zombie dye for 15 minutes at room temperature. The cells were then washed and stained with antibodies against cell surface markers and analyzed by flow cytometry. As described above, two different approaches were used to study the induction of RTCC: a) reduction in the number of CSFE+ target cells (data for which are depicted in Figure 2A); and b) reduction in the number of CSFE+ target cells (data for which are depicted in Figure 2B). Figure 27A ), and b) Zombie Aqua staining of CSFE+ target cells (data depicted in Figure 27B CD4 + and CD8 + T cell activation and degranulation were also determined based on CD107a, CD25, and CD69 expression (data for which are depicted in Figures 28-29).

[0496] Overall, the data showed that the prototype αENPP3 x αCD3 bsAb dose-dependently induced RTC on ENPP3 cells; CD3 binding affinity correlated with RTCC potency (i.e., bsAbs with CD3 high induced RTCC more effectively than bsAbs with CD3 high-Int No. 1); and bsAbs with an AN1-based binding domain induced RTCC more effectively than bsAbs with an H16-7.8-based binding domain. Consistent with the RTCC data, αENPP3 x αCD3 bsAbs dose-dependently induced T cell activation; CD3 binding affinity correlated with activation potency (i.e., bsAbs with CD3 high induced T cell activation more effectively than bsAbs with CD3 high-Int No. 1); and bsAbs with an AN1-binding domain induced T cell activation more effectively than bsAbs with an H16-7.8-binding domain.

[0497] Example 4: Improving αENPP3 x αCD3 Production

[0498] Typically, bispecific antibodies were produced by transient transfection in HEK293E cells and purified by a two-step purification process involving protein A chromatography (purification part 1) followed by ion exchange chromatography (purification part 2).

[0499] 4A: Engineering AN1 variants to improve yield

[0500] 4A(a): Production of XENP28287 results in a homogenous population containing aggregates and unpaired monomers

[0501] XENP28287 was purified from HEK293E supernatant as described above. Figure 30 A depicts a chromatogram showing purified fraction 2 of XENP28287 (cation exchange chromatography followed by Protein A chromatography). The chromatogram shows the separation of two peaks (peak B and peak BC), which were further characterized for identity, purity, and homogeneity by analytical size exclusion chromatography with multi-angle light scattering (aSEC-MALS) and analytical cation exchange chromatography (aCIEX), as generally described below.

[0502] Peaks B and BC, isolated from purified fraction 2 of XENP28287 (as well as prepurified material), were analyzed using aSEC-MALS to infer their component protein species. Analysis was performed on an Agilent 1200 high-performance liquid chromatography (HPLC) system. Samples were injected onto a Superdex™ 200 10 / 300 GL column (GE Healthcare Life Sciences) at 1.0 ml / min for 25 minutes at 4°C using a UV detection wavelength of 280 nM and a mobile phase of 1× PBS, pH 7.4. MALS was performed on a miniDAWN® TREOS® (Wyatt Technology, Santa Barbara, California) with an Optilab® T-rEX refractive index detector. Analysis was performed using an Agilent OpenLab Chromatography Data System (CDS) ChemStation version AIC C.01.07 and ASTRA version 6.1.7.15. Chromatograms depicting the SEC separation profiles of the pre-purified material, Peak B and Peak BC, and the approximate MW of the component species determined by MALS are depicted in Figure 30 Peak B contains a major species at approximately 126 kDa, consistent with the calculated molecular weight of the XENP28287 heterodimer (based on the amino acid sequence), but also contains a contaminating species at 75 kDa (likely a monomer). Peak BC contains peaks with species at 308 kDa (likely an aggregate), 121 kDa (XENP28287), and 82 kDa (a contaminating monomer). Notably, the fractionation spectrum of the prepurified material indicates that less than 85% of the material is bispecific antibody heterodimer.

[0503] Peaks from purified fraction 2 were also analyzed using analytical CIEX to further assess the purity and homogeneity of peaks B and BC. Analysis was performed on an Agilent 1200 high-performance liquid chromatography (HPLC) system. Samples were injected at 1.0 mL / min onto a Proteomix SCX-NP5 5µM non-porous column (Sepax Technologies, Inc., Newark, Del.) using a 0–40% NaCl gradient in 20 mM MES, pH 6.0 buffer, using a UV detection wavelength of 280 nM. Analyses were performed using Agilent OpenLAB CDS ChemStation Edition AIC C.01.07. Figure 30A chromatogram depicting the CIEX separation of peaks B and BC is depicted in C. Notably, the aCIEX separation shows that in the peak BC material, in addition to the main peak, there are many charge variants.

[0504] 4A(b): AN1 VH variant H1.8 enables improved separation

[0505] A number of AN1 variable heavy chain (VH) domains were designed with the goal of improving bispecific antibody production. One particular VH variant (H1.8; SEQ ID NO: XXX; also depicted in Figure 13) enabled improved separation of bispecific antibody heterodimers from contaminating species. To illustrate this, XENP28925 (containing an ENPP3-binding domain with the AN1 H1.8 VH variant; sequence depicted in Figure 17) was produced and purified from HEK293E supernatant as described above. Figure 31 A depicts a chromatogram showing purified fraction 2 of XENP28925 (cation exchange chromatography followed by Protein A chromatography). The chromatogram shows the separation of one major peak (Peak B), which was further characterized by aSEC-MALS and aCIEX) for identity, purity, and homogeneity as described above.

[0506] Figure 31 B to Figure 31 A chromatogram depicting the aSEC separation profile of the pre-purified material and Peak B (where the MW of the component species are as determined by MALS) and the aCIEX separation profile of Peak B is depicted in C. The profile shows that Peak B contains a major species of approximately 128 kDa, which is consistent with the calculated molecular weight of the XENP28925 heterodimer (based on the amino acid sequence). Notably, the separation profile of the pre-purified material shows that greater than 97% of this material is the bispecific antibody heterodimer.

[0507] Overall, this demonstrates that the AN1 H1.8 VH variant enables increased yield of αENPP3 x αCD3 heterodimers, as well as improved separation of heterodimers from contaminating species.

[0508] 4B: Backbone Engineering of a 2+1 Fab2-scFv-Fc Bispecific Format to Improve Yield

[0509] 4B(a): Production of XENP31419 results in a protein population skewed toward VH-Fc homodimers

[0510] XENP31149 (αENPP3 x αCD3 bsAb in 2+1 Fab2-scFv-Fc format; sequence depicted in FIG23 ) was purified from HEK293E supernatant as described above. Figure 32A depicts a chromatogram showing purified fraction 2 of XENP31149 (cation exchange chromatography followed by Protein A chromatography). The chromatogram shows the separation of two peaks (Peak A and Peak B), which were further characterized by analytical size exclusion chromatography with multi-angle light scattering (aSEC-MALS) to obtain identity, as generally described above.

[0511] Figure 32 B depicts the chromatogram of the aSEC separation profile of peaks A and B, along with the MWs of the component species determined by MALS. The spectra show that the major peak A contains a species with a molecular weight of 148.4 kDa, consistent with the calculated molecular weight of the VH-Fc homodimer, while the minor peak B contains a species with a molecular weight of 173.9 kDa, consistent with the calculated molecular weight of the XENP31149 heterodimer. Thus, XENP31149 was produced at a very low titer of 12.4 mg / L.

[0512] 4B(b): Engineering a full hinge in the Fab-scFv-Fc chain improves the yield of 2+1 Fab2-scFv-Fc heterodimers

[0513] Various approaches were investigated to increase the yield of 2+1 Fab2-scFv-Fc heterodimers, including altering the linker between the VH and scFv or the linker between the scFv and CH2 in the Fab-scFv-Fc chain. XENP31419 (sequence depicted in FIG23 ) was engineered into its XENP31149 counterpart, which features a full hinge (EPKSCDKTHTCPPCP; SEQ ID NO: XX) rather than a flexible half hinge (GGGGSGGGGSKTHTCPPCP; SEQ ID NO: XX) between the scFv and CH2 regions in the Fab-scFv-Fc chain. XENP31419 was produced and purified from HEK293E supernatant as described above. Figure 33 A depicts a chromatogram showing purified fraction 2 of XENP31419 (cation exchange chromatography followed by protein A chromatography). The chromatogram shows the separation of two peaks (minor peak A and major peak B), which were further characterized by analytical size exclusion chromatography with multi-angle light scattering (aSEC-MALS) to obtain identity, as generally described above.

[0514] Figure 33B depicts the chromatogram of the aSEC separation profile of peaks A and B, along with the MWs of the component species determined by MALS. The spectra show that minor peak A contains a species with a molecular weight of 152.2 kDa, consistent with the calculated molecular weight of the VH-Fc homodimer, while major peak B contains a species with a molecular weight of 180 kDa, consistent with the calculated molecular weight of the XENP31419 heterodimer. Thus, production yielded a significantly increased titer of 107.8 mg / L of XENP31419 heterodimer.

[0515] Example 5: Tuning the αENPP3 x αCD3 bsAb to improve selectivity and therapeutic index

[0516] The following experiments usually use KU812 as ENPP3 高 Target cells (as ENPP3 + tumor cells) or RCC4 as ENPP3 低 The study was performed using target cells (as a surrogate for cells outside the tumor environment). Target cells were incubated with human PBMCs and the test article at the specified effector-to-target cell ratio at 37°C. Following incubation, cells were stained with AquaZombie dye for 15 minutes at room temperature. Cells were then washed and stained with antibodies against cell surface markers and analyzed by flow cytometry. Induction of RTCC was determined using Zombie Aqua staining of CSFE+ target cells; and T cell activation and degranulation were determined by expression of CD107a, CD25, and CD69 on lymphocytes. It should also be noted that some of the datasets were from the same experiment, as several engineering approaches were explored simultaneously.

[0517] To investigate the on-target / off-tumor killing potential of a prototype 1+1 Fab-scFv-Fc bispecific antibody with both high-affinity CD3 binding and high-affinity ENPP3 binding, KU812 cells and RCC4 cells were incubated with human PBMCs (effector to target cell ratio of 10:1) and the indicated concentrations of XENP28925 at 37°C for 18 h. Figure 34 The data depicted in the present study show that XENP28925 is involved in the expression of ENPP3 高 Induced RTCC in KU812 cells; however, XENP28925 also 低 The induction of RTCC on RCC4 cells indicated that there was room for improvement in the therapeutic index of the αENPP3 x αCD3 bispecific antibody. Therefore, the prototype αENPP3 x αCD3 bsAb was further engineered with the goal of improving selectivity and therapeutic index.

[0518] 5A: Tuning ENPP3 binding affinity

[0519] The first approach explored tuning ENPP3 binding affinity. Variant ENPP3 binding arms were engineered with variable light chain domain variants with the goal of creating an ENPP3 binding affinity ladder, the illustrative sequences of which are depicted in Figure 13 (for the variable region) and Figure 16 (in the context of a 1+1 Fab-scFv-Fc bsAb).

[0520] The binding of affinity-engineered αENPP3 x αCD3 bsAb to cell surface ENPP3 was studied. KU812 cells were incubated with the indicated test articles at the indicated concentrations. The cells were then stained with an Fcγ fragment-specific secondary antibody to detect the test articles and analyzed by flow cytometry. Figure 35 The data depicted show that the affinity-engineered αENPP3 x αCD3 bsAbs exhibited affinity to ENPP3 ranging from high (XENP28925 with L1 variable light chain) to moderate (XENP29516 with L1.33 variable light chain) to low (XENP30262 with L1.77 variable light chain). 高 A range of binding efficiencies in KU812 cells.

[0521] Next, to investigate the effect of modulating ENPP3 binding affinity on the selectivity of the bispecific antibody, KU812 (ENPP3 高 ) and RCC4 (ENPP3 低 ) cells were incubated with human PBMCs (effector to target ratio of 10:1) and the indicated concentrations of the following bispecific antibodies with fixed CD3 potency (CD3 high) at 37°C for 42 h: XENP28925 (WT high ENPP3 binding), XENP29516 (moderate ENPP3 binding), or XENP30262 (low ENPP3 binding). Figure 36 The data depicted show that both XENP29516 and XENP30262 showed inhibition of ENPP3 compared to XENP28925. 低 Substantially less potent induction of RTCC on cells where RTCC potency correlated with binding potency as shown above. However, XENP29516 and XENP30262 also showed inhibition of ENPP3 高 Less efficient induction of RTCC on cells.

[0522] 5B: Tuning CD3 binding potency

[0523] Reducing affinity for CD3 was also explored to improve pharmacokinetics and attenuate cytokine release. An αENPP3 x αCD3 1+1 Fab-scFv-Fc bsAb was engineered with a CD3 high-Int no. 1 scFv, an illustrative sequence of which is depicted in FIG18 .

[0524] To investigate the induction of cytokine release, an experiment was performed in which KU812 and RCC4 cells were incubated with human PBMCs (effector to target ratio of 10:1) and the indicated concentrations of XENP28925 (CD3 high) or XENP29436 (CD3 high - Int No. 1) for 18 hours at 37°C. The release of IFNγ, IL-6, and TNFα was measured using the V-PLEX Proinflammatory Panel 1 Human Kit (according to the manufacturer's protocol; Meso Scale Discovery, Rockville, Md.), and the data are depicted in Figures 37 and 37. Figure 38 middle.

[0525] To investigate whether tuning CD3 binding potency had any effect on selectivity, another experiment was performed in which KU812 (ENPP3 高 ) and RCC4 (ENPP3 低 ) cells were incubated with human PBMCs (effector to target ratio 10:1) and the indicated concentrations of XENP28925 (CD3 high) or XENP29436 (CD3 high - Int No. 1) at 37°C for 42 h. Figure 39 The data presented indicate that XENP29436 exhibits substantially less efficacy against ENPP3 compared to XENP28925. 低 However, XENP29436 also showed reduced induction of ENPP3 高 Efficacy of RTCC on cells.

[0526] 5C: Tuning both ENPP3 binding affinity and CD3 binding potency

[0527] Next, the effects of reducing both ENPP3 and CD3 binding potency were investigated.An αENPP3 x αCD3 1+1 Fab-scFv-Fc bsAb with a reduced potency ENPP3 binding domain and a CD3 High-Int No. 1 scFv was engineered, the sequence of which is depicted in FIG18 .

[0528] KU812 cells were incubated with human PBMCs (at an effector to target ratio of 10:1) and the indicated concentrations of XENP28925 (ENPP3 high; CD3 high), XENP29436 (ENPP3 high; CD3 high-Int No. 1), XENP29518 (ENPP3 medium; CD3 high), XENP29463 (ENPP3 medium; CD3 high-Int No. 1), XENP30262 (ENPP3 low; CD3 high), or XENP30263 (ENPP3 low; CD3 high-Int No. 1) at 37°C for 18 hours. IFNγ release was measured using the V-PLEX Proinflammatory Panel 1 Human Kit. Figure 40 The data depicted in show that reducing the potency of CD3 or ENPP3 binding reduces the induction of cytokine release. Notably, reducing the potency of CD3 and ENPP3 binding further reduces the induction of cytokine release.

[0529] 5D: Tuning both ENPP3 binding potency and ENPP3 binding valence

[0530] It is hypothesized that although the reduced ENPP3 binding efficiency reduces the 低 Cells and ENPP3 高 cells, but increased binding valence may restore ENPP3 高 Therefore, an αENPP3 x αCD3 bispecific antibody with reduced ENPP3 binding potency was engineered into a 2+1 Fab2-scFv-Fc format, the sequence of which is depicted in FIG19 .

[0531] KU812 (ENPP3 高 ) and RCC4 (ENPP3 低 ) cells were incubated with human PBMCs (effector to target cell ratio of 10:1) and the indicated concentrations of XENP28925 (monovalent high ENPP3 binding), XENP29516 (monovalent moderate ENPP3 binding), XENP29520 (bivalent moderate ENPP3 binding), XENP30262 (monovalent low ENPP3 binding), or XENP30264 (bivalent low ENPP3 binding) at 37°C for 42 h. ...

Claims

1. A heterodimeric anti-CD3 x anti-ENPP3 antibody, comprising: a. a first monomer, the first monomer consisting of SEQ ID NO: 531; b. a second monomer, the second monomer consisting of SEQ ID NO: 532; and c. a light chain consisting of SEQ ID NO:

533.

2. A heterodimeric antibody, comprising: a. a first monomer encoded by a nucleic acid encoding SEQ ID NO: 531; b. a second monomer encoded by a nucleic acid encoding SEQ ID NO: 532; and c. A light chain encoded by the nucleic acid encoding SEQ ID NO:

533.

3. A nucleic acid composition, comprising: a. a first nucleic acid, wherein the first nucleic acid encodes the first monomer according to claim 1 or claim 2; b. a second nucleic acid encoding a second monomer according to claim 1 or claim 2; and c. a third nucleic acid encoding the light chain according to claim 1 or claim 2.

4. An expression vector composition, comprising: a. a first expression vector, the first expression vector comprising the first nucleic acid according to claim 3; b. a second expression vector, the second expression vector comprising the second nucleic acid according to claim 3; and c. a third expression vector, comprising the third nucleic acid according to claim 3. A host cell comprising the expression vector composition according to claim 4 .

6. A method for preparing the heterodimeric antibody according to claim 1 or claim 2, the method comprising culturing the host cell according to claim 5 under conditions where the heterodimeric antibody is expressed, and recovering the heterodimeric antibody.

Images