Anti-CD28 X anti-ENPP3 antibody

Abstract

Provided herein are novel anti-CD28x anti-ENPP3 antibodies and methods of using such antibodies to treat ENPP3 related cancers. The subject anti-CD28x anti-ENPP3 antibodies are capable of agonically binding to a CD28 costimulatory molecule on a T cell and to ENPP3 on a tumor cell. Thus, such antibodies selectively enhance anti-tumor activity at the tumor site while minimizing peripheral toxicity. The subject antibodies provided herein may particularly be used in combination with other anti-cancer therapies, including, for example, bispecific antibodies for the treatment of ENPP3-associated cancers.

Description

[0001] Priority Statement This application claims priority and benefit to U.S. Provisional Application No. 63 / 496,367, filed April 14, 2023, the contents of which are hereby incorporated by reference in their entirety. Background Technology

[0002] Antibody-based therapeutics have been successfully used to treat a variety of diseases, including cancer. A growing and exploring approach is to engineer single immunoglobulin molecules that co-link two different antigens. Such alternative antibody forms that link two different antigens are often referred to as bispecific antibodies. A specific approach to bispecific antibodies involves engineering a first binding domain to CD3 and a second binding domain to an antigen associated with or upregulated on cancer cells (e.g., ENPP3), thus redirecting the bispecific antibody to CD3. + T cells destroy cancer cells.

[0003] However, over time, TILs lose their cytotoxic capacity due to the upregulation of inhibitory immune checkpoints. While checkpoint blockade has been shown to increase clinical response rates compared to other treatment options, many patients still fail to respond to it. Binding to co-stimulatory receptors on TILs has the potential to provide a positive signal capable of overcoming the negative signaling of immune checkpoints. Preclinical and clinical studies of antibodies agonistic to co-stimulatory receptors have indeed demonstrated that agonism of co-stimulatory receptors can produce an impressive anti-tumor response, thereby activating T cells to attack tumor cells.

[0004] Enhancing antitumor activity by specifically destroying tumor cells while minimizing peripheral toxicity is also important for cancer therapy. In this context, it is crucial that T cells receive co-stimulatory signals only in the presence of target tumor cells. However, the agonistic effect of monospecific full-length antibodies on co-stimulatory receptors may not distinguish between TILs and peripheral T cells reacting to autoantigens that cause autoimmune toxicity.

[0005] Exonucleotide pyrophosphatase / phosphodiesterase family member 3 (ENPP3) is a type II transmembrane glycoprotein belonging to the nucleotide pyrophosphatase / phosphodiesterase family. These proteins are involved in the hydrolysis of extracellular nucleotides and also possess ATPase and ATP pyrophosphatase activities. ENPP3 is overexpressed in a variety of cancers, including most renal cell carcinomas and some liver cancers. Due to its overexpression in multiple tumors, ENPP3 is a candidate for targeted therapeutics. Although immunotherapies targeting ENPP3 have been explored, novel immune-enhancing compositions are still needed to treat ENPP3-related cancers. Summary of the Invention

[0006] This article provides a novel anti-CD28 x anti-ENPP3 antibody and a method for using such antibodies to treat ENPP3-related cancers. The subject anti-CD28 x anti-ENPP3 antibody is capable of binding agonistically to CD28 co-stimulatory molecules on T cells and ENPP3 on tumor cells. Therefore, such antibodies selectively enhance antitumor activity at the tumor site while minimizing peripheral toxicity. The subject antibody provided herein is particularly suitable for use in combination with other anticancer therapies, including, for example, bispecific antibodies for the treatment of ENPP3-related cancers.

[0007] In a first aspect, this document provides an anti-CD28 x anti-ENPP3 heterodimer antibody comprising a) a first monomer; b) a second monomer; and c) a light chain. The first monomer comprises: i) a single-chain variable fragment (scFv); and ii) a first Fc domain, wherein the scFv is covalently linked to the N-terminus of the first Fc domain using a domain linker. The second monomer comprises VH1-CH1-hinge-CH2-CH3 from the N-terminus to the C-terminus, wherein VH1 is a first variable heavy chain domain and CH2-CH3 is a second Fc domain. The light chain comprises VL1-CL from the N-terminus to the C-terminus, wherein VL1 is a first variable light chain domain and CL is a constant light chain domain. The scFv comprises a second VH domain (VH2), an scFv linker, and a second variable light chain domain (VL2). VH1 and VL1 together form a first antigen-binding domain (ABD), and VH2 and VL2 together form a second ABD. Furthermore, one of the first ABD and the second ABD is a CD28 binding domain, and the other of the first ABD and the second ABD is an exonucleotide pyrophosphatase / phosphodiesterase 3 (ENPP3) binding domain.

[0008] In some embodiments, the scFv includes a VH2-scFv connector-VL2 from the N-end to the C-end. In some embodiments, the scFv includes a VL2-scFv connector-VH2 from the N-end to the C-end.

[0009] In an exemplary embodiment, the first ABD is an ENPP3 binding domain, and the second ABD is a CD28 binding domain. In some embodiments, VH1 and VL1 are VH and VL of any ENPP3 binding domain in FIG22 or variations thereof. In some embodiments, VH2 and VL2 are selected from one of the following: 1) Figure 15 , VH and VL or variants thereof of any CD28 binding domain in Figures 18 and 21; or 2) (i) Figure 15 Or VH or its variants as shown in Figure 16; and (ii) Figure 15 Or VL or its variants as shown in Figure 17.

[0010] In some implementations, the first Fc domain and the second Fc domain are each variant Fc domains.

[0011] In some embodiments, the first and second Fc domains comprise a set of heterodimerization skew variants selected from the following heterodimerization variants: 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 designations. In some embodiments, the first and second Fc domains comprise the heterodimerization skew variant S364K / E357Q: L368D / K370S, wherein the numbering is according to EU designations.

[0012] In some embodiments, the first and second Fc domains each contain one or more ablation variants. In an exemplary embodiment, the one or more ablation variants include E233P / L234V / L235A / G236del / S267K, where the numbering is based on EU designations.

[0013] In some embodiments, one of the first or second monomers further comprises one or more pI variants. In some embodiments, the CH1-hinge-CH2-CH3 of the second monomer comprises pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is in accordance with EU designations.

[0014] In some embodiments, the CH1-hinge-CH2-CH3 of the second monomer comprises amino acid variants E233P / L234V / L235A / G236del / S267K / L368D / K370S / N208D / Q295E / N384D / Q418E / N421D, and the first Fc domain comprises amino acid variants E233P / L234V / L235A / G236del / S267K / S364K / E357Q, wherein the numbering is according to EU designations. In an exemplary embodiment, the Fc domains of both the first and second variants further comprise amino acid variants 428L / 434S.

[0015] In some implementations, the scFv connector is GKPGSGKPGSGKPGSGKPGS.

[0016] On the other hand, this paper provides an anti-CD28 x anti-ENPP3 heterodimer antibody, the antibody comprising a) a first monomer; b) a second monomer; c) a first light chain; and d) a second light chain. The first monomer comprises, from N-terminus to C-terminus, VH1-CH1-first domain linker-scFv-second domain linker-CH2-CH3, wherein VH1 is a first variable heavy chain domain and CH2-CH3 is a first Fc domain. The second monomer comprises, from N-terminus to C-terminus, VH1-CH1-hinge-CH2-CH3, wherein CH2-CH3 is a second Fc domain. The first and second light chains each comprise, from N-terminus to C-terminus, VL1-CL, wherein VL1 is a first variable light chain domain and CL is a constant light chain domain. The scFv comprises a second VH domain (VH2), an scFv linker, and a second variable light chain domain (VL2). The VH1 of the first monomer and the VL1 of the first light chain, and the VH1 of the second monomer and the VL1 of the second light chain, each form a first antigen-binding domain (ABD), and the VH2 and the VL2 form a second ABD. Furthermore, the first ABD is an exonucleotide pyrophosphatase / phosphodiesterase 3 (ENPP3) binding domain, and the second ABD is a CD28 binding domain; alternatively, the first ABD is a CD28 binding domain, and the second ABD is an ENPP3 binding domain.

[0017] In some embodiments, the scFv includes a VH2-scFv connector-VL2 from the N-end to the C-end. In some embodiments, the scFv includes a VL2-scFv connector-VH2 from the N-end to the C-end.

[0018] In an exemplary embodiment, the first ABD is an ENPP3 binding domain, and the second ABD is a CD28 binding domain. In some embodiments, VH1 and VL1 are selected from: 1) any VH and VL of the ENPP3 binding domain in FIG22 or a variant thereof. In some embodiments, VH2 and VL2 are selected from: 1) Figure 15 , VH and VL or variants thereof of any CD28 binding domain in Figures 18 and 21; or 2) (i) Figure 15 Or VH or its variants as shown in Figure 16; and (ii) Figure 15 Or VL or its variants as shown in Figure 17.

[0019] In some implementations, the first Fc domain and the second Fc domain are each variant Fc domains.

[0020] In some embodiments, the first and second Fc domains comprise a set of heterodimerization skew variants selected from the following heterodimerization variants: 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 designations. In some embodiments, the first and second Fc domains comprise the heterodimerization skew variant S364K / E357Q: L368D / K370S, wherein the numbering is according to EU designations. In some embodiments, the first and second Fc domains each contain one or more ablation variants. In an exemplary embodiment, the one or more ablation variants include E233P / L234V / L235A / G236del / S267K, where the numbering is based on EU designations.

[0021] In some embodiments, one of the first or second monomers further comprises one or more pI variants. In some embodiments, the CH1-hinge-CH2-CH3 of the second monomer comprises pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is in accordance with EU designations.

[0022] In some embodiments, the CH1-hinge-CH2-CH3 of the second monomer comprises amino acid variants E233P / L234V / L235A / G236del / S267K / L368D / K370S / N208D / Q295E / N384D / Q418E / N421D, and the first Fc domain comprises amino acid variants E233P / L234V / L235A / G236del / S267K / S364K / E357Q, wherein the numbering is according to EU designations. In an exemplary embodiment, the Fc domains of both the first and second variants further comprise amino acid variants 428L / 434S.

[0023] In some implementations, the scFv connector is GKPGSGKPGSGKPGSGKPGS.

[0024] On the other hand, this paper provides an anti-CD28 x anti-ENPP3 heterodimer antibody, the antibody comprising a) a first monomer; b) a second monomer; c) a first light chain; and d) a second light chain. The first monomer comprises, from N-terminus to C-terminus, VH1-CH1-hinge-CH2-CH3-domain linker-scFv, wherein VH1 is a first variable heavy chain domain and CH2-CH3 is a first Fc domain. The second monomer comprises, from N-terminus to C-terminus, VH1-CH1-hinge-CH2-CH3, wherein VH1 is a first variable heavy chain domain and CH2-CH3 is a second Fc domain. The first and second light chains each comprise, from N-terminus to C-terminus, VL1-CL, wherein VL1 is a first variable light chain domain and CL is a constant light chain domain. The scFv comprises a second VH domain (VH2), an scFv linker, and a second variable light chain domain (VL2). The VH1 of the first monomer and the VL1 of the first light chain, and the VH1 of the second monomer and the VL1 of the second light chain, each form a first antigen-binding domain (ABD), and the VH2 and the VL2 form a second ABD. Furthermore, the first ABD is an exonucleotide pyrophosphatase / phosphodiesterase 3 (ENPP3) binding domain, and the second ABD is a CD28 binding domain; alternatively, the first ABD is a CD28 binding domain, and the second ABD is an ENPP3 binding domain.

[0025] In some embodiments, the scFv includes a VH2-scFv connector-VL2 from the N-end to the C-end. In some embodiments, the scFv includes a VL2-scFv connector-VH2 from the N-end to the C-end.

[0026] In an exemplary embodiment, the first ABD is an ENPP3 binding domain, and the second ABD is a CD28 binding domain. In some embodiments, VH1 and VL1 are selected from: 1) any VH and VL of the ENPP3 binding domain in FIG22 or a variant thereof. In some embodiments, VH2 and VL2 are selected from: 1) Figure 15 , VH and VL or variants thereof of any CD28 binding domain in Figures 18 and 21; or 2) (i) Figure 15 Or VH or its variants as shown in Figure 16; and (ii) Figure 15 Or VL or its variants as shown in Figure 17.

[0027] In some implementations, the first Fc domain and the second Fc domain are each variant Fc domains.

[0028] In some embodiments, the first and second Fc domains comprise a set of heterodimerization skew variants selected from the following heterodimerization variants: 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 designations. In some embodiments, the first and second Fc domains comprise the heterodimerization skew variant S364K / E357Q: L368D / K370S, wherein the numbering is according to EU designations.

[0029] In some embodiments, the first and second Fc domains each contain one or more ablation variants. In an exemplary embodiment, the one or more ablation variants include E233P / L234V / L235A / G236del / S267K, where the numbering is based on EU designations.

[0030] In some embodiments, one of the first or second monomers further comprises one or more pI variants. In some embodiments, the CH1-hinge-CH2-CH3 of the second monomer comprises pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is in accordance with EU designations.

[0031] In some embodiments, the CH1-hinge-CH2-CH3 of the second monomer comprises amino acid variants E233P / L234V / L235A / G236del / S267K / L368D / K370S / N208D / Q295E / N384D / Q418E / N421D, and the first Fc domain comprises amino acid variants E233P / L234V / L235A / G236del / S267K / S364K / E357Q, wherein the numbering is according to EU designations. In an exemplary embodiment, the Fc domains of both the first and second variants further comprise amino acid variants 428L / 434S.

[0032] In some implementations, the scFv connector is GKPGSGKPGSGKPGSGKPGS.

[0033] On the other hand, this paper provides a heterodimer antibody comprising a) a first monomer; b) a second monomer; c) a first light chain; and d) a second light chain. The first monomer comprises, from N-terminus to C-terminus, a VH1-CH1-domain linker-VH1-CH1-hinge-CH2-CH3, wherein each VH1 is a first variable heavy chain domain, and CH2-CH3 is a first Fc domain. The second monomer comprises, from N-terminus to C-terminus, an scFv-domain linker-CH2-CH3, wherein VH1 is a first variable heavy chain domain and CH2-CH3 is a second Fc domain. The first and second light chains each comprise, from N-terminus to C-terminus, VL1-CL, wherein VL1 is a first variable light chain domain and CL is a constant light chain domain. The scFv comprises a second VH domain (VH2), an scFv linker, and a second variable light chain domain (VL2). The VH1 of the first monomer and the VL1 of the first light chain, and the VH1 of the second monomer and the VL1 of the second light chain, each form a first antigen-binding domain (ABD), and the VH2 and the VL2 form a second ABD. Furthermore, the first ABD is an exonucleotide pyrophosphatase / phosphodiesterase 3 (ENPP3) binding domain, and the second ABD is a CD28 binding domain; alternatively, the first ABD is a CD28 binding domain, and the second ABD is an ENPP3 binding domain.

[0034] In some embodiments, the first and second Fc domains comprise a set of heterodimerization skew variants selected from the following heterodimerization variants: 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 designations. In some embodiments, the first and second Fc domains comprise the heterodimerization skew variant S364K / E357Q: L368D / K370S, wherein the numbering is according to EU designations.

[0035] In some embodiments, the first and second Fc domains each contain one or more ablation variants. In an exemplary embodiment, the one or more ablation variants include E233P / L234V / L235A / G236del / S267K, where the numbering is based on EU designations.

[0036] In some embodiments, one of the first or second monomers further comprises one or more pI variants. In some embodiments, the CH1-hinge-CH2-CH3 of the second monomer comprises pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is in accordance with EU designations.

[0037] In some embodiments, the CH1-hinge-CH2-CH3 of the second monomer comprises amino acid variants E233P / L234V / L235A / G236del / S267K / L368D / K370S / N208D / Q295E / N384D / Q418E / N421D, and the first Fc domain comprises amino acid variants E233P / L234V / L235A / G236del / S267K / S364K / E357Q, wherein the numbering is according to EU designations. In an exemplary embodiment, the Fc domains of both the first and second variants further comprise amino acid variants 428L / 434S.

[0038] In some implementations, the scFv connector is GKPGSGKPGSGKPGSGKPGS.

[0039] On the other hand, this document provides a bispecific antibody comprising: a) an ENPP3 binding domain comprising i) a first variable heavy chain domain (VH1) and ii) a first variable light chain domain (VL1); and b) an anti-CD28 binding domain comprising i) a second variable heavy chain domain (VH2) and ii) a second variable light chain domain (VL2). In some embodiments, VH1 and VL1 are selected from: 1) any VH and VL of the ENPP3 binding domain in Figure 22 or a variant thereof. In some embodiments, VH2 and VL2 are selected from: 1) Figure 15 , VH and VL or variants thereof of any CD28 binding domain in Figures 18 and 21; or 2) (i) Figure 15 Or VH or its variants as shown in Figure 16; and (ii) Figure 15 Or VL or its variants as shown in Figure 17.

[0040] In some embodiments of the bispecific antibody, the first Fc domain and the second Fc domain are each variant Fc domains.

[0041] In some embodiments of the bispecific antibody, the first and second Fc domains comprise a set of heterodimerization skewed variants selected from the following heterodimerization variants: 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 designations. In some embodiments, the first and second Fc domains comprise the heterodimerization skewed variant S364K / E357Q: L368D / K370S, wherein the numbering is according to EU designations.

[0042] In some embodiments, the first and second Fc domains each contain one or more ablation variants. In an exemplary embodiment, the one or more ablation variants include E233P / L234V / L235A / G236del / S267K, where the numbering is based on EU designations.

[0043] In some embodiments, one of the first or second monomers further comprises one or more pI variants. In some embodiments, the CH1-hinge-CH2-CH3 of the second monomer comprises pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is in accordance with EU designations.

[0044] This document also provides nucleic acid compositions comprising nucleic acids encoding the antibodies described herein, expression vector compositions comprising such nucleic acids, host cells comprising the expression vector compositions for preparing the antibodies, and methods for preparing the antibodies.

[0045] On the other hand, this article provides a method for treating ENPP3-related cancers in patients in need, the method comprising administering to the patient an anti-CD28 x anti-ENPP3 bispecific antibody as described herein.

[0046] On the other hand, this article provides a method for treating ENPP3-related cancers in patients in need, the method comprising administering to the patient an anti-CD28 x anti-ENPP3 bispecific antibody as described herein; and an anti-CD3 x anti-ENPP3 bispecific antibody. Attached Figure Description

[0047] Figure 1A and Figure 1B The sequences of CD28 in humans, mice, and cynomolgus monkeys were depicted. These CD28 sequences could be used to develop cross-reactive CD28 antigen-binding domains for clinical development.

[0048] Figure 2A and Figure 2B The antigen sequences of various antigens used in this invention are described, including human and cyno antigens, to facilitate the development of antigen-binding domains that bind to both, thereby facilitating clinical development.

[0049] Figures 3A-3F Several pairs of useful heterodimer variants (including skewed variants and pI variants) were depicted. Figure 3F Among them, there are variants that do not have a corresponding "monomer 2" variant. Such variants are pI variants that can be used alone on either monomer of a bispecific antibody (e.g., ENPP3 x CD28 bsAb), or, for example, included on a non-scFv side in the form of utilizing scFv as a component, and can use an appropriate charged scFv linker on a second monomer utilizing scFv as the CD28 binding domain. Figure 6 Suitable charged connectors are shown. The heterodimer yield (%) and CH3T of preferred Fc heterodimer variants have been previously described. m (°C) (See, for example, U.S. Patent Application No. 2019 / 0248898) Figure 8 ).

[0050] Figure 4 A list depicting the constant regions of isosteric variant antibodies and their respective substitutions is provided. pI_(-) indicates a lower pI variant, while pI_(+) indicates a higher pI variant. These variants may optionally and independently be combined with other variants, including heterodimerized variants, outlined herein.

[0051] Figure 5 Useful ablation variants (also known as “knockout” or “KO” variants) that ablate FcγR binding are described. In some embodiments, such ablation variants include Fc domains of both monomers of the subject antibody described herein. In other embodiments, the ablation variants include only one variant Fc domain.

[0052] Figure 6Various charged scFv linkers are described, which can be used to increase or decrease the pI of subject heterodimeric bispecific antibodies (e.g., ENPP3 x CD28 bsAb) that utilize one or more scFvs as components, as described herein. (+H) positively charged linkers are particularly used herein, especially in conjunction with the anti-CD28 VL and VH sequences shown herein. 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 “Whitlow”. It should be noted that such linkers are used to reduce aggregation in scFvs and enhance proteolytic stability in scFvs. Such charged scFv linkers can be used in any subject antibody form comprising scFvs disclosed herein (e.g., 1+1 Fab-scFv-Fc and 2+1 Fab2-scFv-Fc forms).

[0053] Figure 7 Several exemplary structural domain connectors are depicted. In some embodiments, these connectors can be used to connect a single chain Fv to an Fc chain. In some embodiments, these connectors can be combined in any orientation. For example, a GGGGS connector can be combined with a "lower half hinge" connector at either the N-end or the C-end.

[0054] Figure 8 A bispecific antibody platform particularly useful for the ENPP3 x CD28 bsAb of the present invention is shown. Although the platform is described in the context of a 1+1 Fab-scFv-Fc form, it can be applied to other bispecific antibody forms.

[0055] Figures 9A-9DSequences of several useful heterodimeric ENPP3 x CD28 bsAb backbones based on human IgG1 are shown, excluding cytokine sequences. Heterodimeric Fc backbone 1 is based on human IgG1 (356E / 358M allotypes) and includes the L368D / K370S skewed variant and the Q295E / N384D / Q418E / N421D pI variant on the first heterodimeric Fc chain, the S364K / E357Q skewed variant on the second heterodimeric Fc chain, and the E233P / L234V / L235A / G236del / S267K ablation variant on both chains. The heterodimeric Fc backbone 2 is based on human IgG1 (356E / 358M allotype) and includes the L368D / K370S skewed variant and the Q295E / N384D / Q418E / N421D pI variant on the first heterodimeric Fc chain, the S364K skewed variant on the second heterodimeric Fc chain, and the E233P / L234V / L235A / G236del / S267K ablation variant on both chains. The heterodimeric Fc backbone 3 is based on human IgG1 (356E / 358M allotype) and includes the L368E / K370S skewed variant and the Q295E / N384D / Q418E / N421D pI variant on the first heterodimeric Fc chain, the S364K skewed variant on the second heterodimeric Fc chain, and the E233P / L234V / L235A / G236del / S267K ablation variant on both chains. The heterodimeric Fc backbone 4 is based on human IgG1 (356E / 358M allotype) and includes the K360E / Q362E / T411E skewed variant and the Q295E / N384D / Q418E / N421D pI variant on the first heterodimeric Fc chain, the D401K skewed variant on the second heterodimeric Fc chain, and the E233P / L234V / L235A / G236del / S267K ablation variant on both chains. The heterodimeric Fc backbone 5 is based on human IgG1 (356D / 358L allotypes) and includes the L368D / K370S skewed variant and the Q295E / N384D / Q418E / N421D pI variant on the first heterodimeric Fc chain, the S364K / E357Q skewed variant on the second heterodimeric Fc chain, and the E233P / L234V / L235A / G236del / S267K ablation variant on both chains.The heterodimeric Fc backbone 6 is based on human IgG1 (356E / 358M allotype) and includes the L368D / K370S skewed variant and the Q295E / N384D / Q418E / N421D pI variant on the first heterodimeric Fc chain, the S364K / E357Q skewed variant on the second heterodimeric Fc chain, and the E233P / L234V / L235A / G236del / S267K ablation variant and the deglycosylated N297A variant on both chains. The heterodimeric Fc backbone 7 is based on human IgG1 (356E / 358M allotype) and includes the L368D / K370S skewed variant and the Q295E / N384D / Q418E / N421D pI variant on the first heterodimeric Fc chain, the S364K / E357Q skewed variant on the second heterodimeric Fc chain, and the E233P / L234V / L235A / G236del / S267K ablation variant and the deglycosylated N297S variant on both chains. The heterodimeric Fc backbone 8 is based on human IgG4 and includes the L368D / K370S skewed variant and the Q295E / N384D / Q418E / N421D pI variant on the first heterodimeric Fc chain, the S364K / E357Q skewed variant on the second heterodimeric Fc chain, and the S228P (S241P in Kabat, according to the EU designation) variant with ablation Fab arm exchanges (as known in the art) on both chains. The heterodimeric Fc backbone 9 is based on human IgG2 and includes the L368D / K370S skewed variant and the Q295E / N384D / Q418E / N421D pI variant on the first heterodimeric Fc chain, and the S364K / E357Q skewed variant on the second heterodimeric Fc chain. The heterodimeric Fc backbone 10 is based on human IgG2 and includes the L368D / K370S skewed variant and the Q295E / N384D / Q418E / N421D pI variant on the first heterodimeric Fc chain, the S364K / E357Q skewed variant on the second heterodimeric Fc chain, and the S267K ablation variant on both chains. The heterodimeric Fc backbone 11 is based on human IgG1 (356E / 358M allotype) and includes the L368D / K370S skewed variant and the Q295E / N384D / Q418E / N421D pI variant on the first heterodimeric Fc chain, the S364K / E357Q skewed variant on the second heterodimeric Fc chain, and the E233P / L234V / L235A / G236del / S267K ablation variant and the M428L / N434S Xtend variant on both chains.The heterodimeric Fc backbone 12 is based on human IgG1 (356E / 358M allotype) and includes the L368D / K370S skewed variant on the first heterodimeric Fc chain, the S364K / E357Q skewed variant and the P217R / P229R / N276K pI variant on the second heterodimeric Fc chain, and the E233P / L234V / L235A / G236del / S267K ablation variant on both chains. The heterodimeric Fc backbone 13 is based on human IgG1 (356D / 358L allotypes) and includes the L368D / K370S skewed variant and the Q295E / N384D / Q418E / N421D pI variant on the first heterodimeric Fc chain, the S364K / E357Q skewed variant on the second heterodimeric Fc chain, and the E233P / L234V / L235A / G236del / S267K ablation variant and the M428L / N434S Xtend variant on both chains. The heterodimeric Fc backbone 14 is based on human IgG1 (356E / 358M allotype) and includes the L368D / K370S skewed variant and the Q295E / N384D / Q418E / N421D pI variant on the first heterodimeric Fc chain, the S364K / E357Q skewed variant on the second heterodimeric Fc chain, and the E233P / L234V / L235A / G236del / S267K ablation variant and the M428L / N434AXtend variant on both chains. The heterodimeric Fc backbone 15 is based on human IgG1 (356D / 358L allotypes) and includes the L368D / K370S skewed variant and the Q295E / N384D / Q418E / N421D pI variant on the first heterodimeric Fc chain, the S364K / E357Q skewed variant on the second heterodimeric Fc chain, and the E233P / L234V / L235A / G236del / S267K ablation variant and the M428L / N434A Xtend variant on both chains.

[0056] Each of these backbones includes sequences that are 90%, 95%, 98%, and 99% identical to the listed sequences (as defined herein) and / or contain 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 additional amino acid substitutions (compared to the “parent” in the figure, as those skilled in the art will understand, the “parent” already contains multiple amino acid modifications compared to parent IgG1 (or IgG2 or IgG4, depending on the backbone). That is, in addition to or as alternatives to the skewed variants, pI variants, and ablation variants contained in the backbones in this figure, the listed backbones may contain additional amino acid modifications (typically amino acid substitutions). Furthermore, the backbones depicted herein may include the deletion of C-terminal glycine (K446_) and / or lysine (K447_). The deletion of C-terminal glycine and / or lysine may be intentionally engineered to reduce heterogeneity, or in the context of certain bispecific forms (such as mAb-scFv forms). In addition, the loss of C-terminal glycine and / or lysine can occur naturally, for example, during production and storage.

[0057] Figure 10 Illustrative sequences for the heterodimeric ENPP3 x CD28 bsAb backbone in the 2 + 1 mAb-scFv form are depicted. The form depicted herein is based on the heterodimeric Fc backbone 1 depicted in Figure 9, but also includes G446_ on monomer 1 (-) and G446_ / K447_ on monomer 2 (+). It should be noted that any additional backbone depicted in Figure 9 can be applied to the 2 + 1 mAb-scFv form, whether or not K447_ is included on one or both chains. It should be noted that these sequences may also include the M428L / N434S variant.

[0058] Figure 11 The sequence of “CH1” in an implementation that can be used in ENPP3 x CD28 bsAb is described.

[0059] Figure 12 A sequence of "hinges" is depicted in an implementation that can be used in ENPP3 x CD28 bsAb.

[0060] Figure 13 A constant domain for homologous light chains in the subject ENPP3 x CD28 bsAb, which can be used to utilize Fab-binding domains, is depicted.

[0061] Figures 14A-14N depict the bispecific forms of the present invention. Figure 14A depicts a “1 + 1 Fab-scFv-Fc” form having a first Fab arm that binds a first antigen and a second scFv arm that binds a second antigen. The 1 + 1 Fab-scFv-Fc form comprises a first monomer containing (optionally via a linker) a first heavy chain variable region (VH1) covalently linked to the N-terminus of a first heterodimeric Fc backbone; a second monomer containing (optionally via a linker) a single-chain Fv covalently linked to the N-terminus of a second corresponding heterodimeric Fc backbone; and a third monomer containing a light chain variable region covalently linked to a light chain constant domain, wherein the light chain variable region is complementary to VH1. Figure 14B depicts a “2 + 1 Fab2-scFv-Fc” form having a first Fab arm and a second Fab-scFv arm, wherein the Fab binds the first antigen and the scFv binds the second antigen. The 2+1 Fab2-scFv-Fc form comprises a first monomer containing a first heavy chain variable region (VH1) covalently linked (optionally via a linker) to the N-terminus of a first heterodimeric Fc backbone; a second monomer containing a VH1 covalently linked (optionally via a linker) to a single-chain Fv, the single-chain Fv being covalently linked (optionally via a linker) to the N-terminus of a second corresponding heterodimeric Fc backbone; and a third monomer containing a light chain variable region covalently linked to a light chain constant structural domain, wherein the light chain variable region is complementary to VH1. Figure 14C depicts a “1+1 common light chain” or “1+1 CLC” form having a first Fc containing a first Fab arm binding a first antigen and a second Fc containing a second Fab arm binding a second antigen. The 1+1 CLC form comprises a first monomer containing VH1-CH1-hinge-CH2-CH3; a second monomer containing VH2-CH1-hinge-CH2-CH3; and a third monomer containing VL-CL. VL pairs with VH1 to form a binding domain with first antigen binding specificity; and VL pairs with VH2 to form a binding domain with second antigen binding specificity. Figure 14D depicts a “2+1 common light chain” or “2+1 CLC” form having a first Fc containing two Fab arms, each binding a first antigen, and a second Fc containing one Fab arm binding a second antigen. The 2+1 CLC form comprises a first monomer containing VH1-CH1-hinge-VH1-CH1-hinge-CH2-CH3; a second monomer containing VH2-CH1-hinge-CH2-CH3; and a third monomer containing VL-CL.VL pairs with first and second VH1 to form a binding domain with binding specificity to a first antigen; and VL pairs with VH2 to form a binding domain with binding specificity to a second antigen. Figure 14E depicts a “2 + 1 mAb-scFv” form having a first Fc containing an N-terminal Fab arm binding to the first antigen and a second Fc containing an N-terminal Fab arm binding to the first antigen and a C-terminal scFv binding to the second antigen. The 2 + 1 mAb-scFv form comprises a first monomer containing VH1-CH1-hinge-CH2-CH3; a second monomer containing VH1-CH1-hinge-CH2-CH3-scFv; and a third monomer containing VL-CL. VL pairs with first and second VH1 to form a binding domain with binding specificity to the first antigen. Figure 14F depicts a “2 + 1 Fab2-Fc x scFv-Fc” form, having a first Fc containing two N-terminal Fab arms that bind to a first antigen and a second Fc domain containing an N-terminal scFv that binds to a second antigen. The 2 + 1 Fab2-Fc x scFv-Fc form comprises a first monomer containing VH1-CH1-VH2-CH1-hinge-CH2-CH3; a second monomer containing scFv-domain structure-CH2-CH3; and a third and fourth monomer containing VL-CL. VL pairs with VH1 and VH2 to form a binding domain specific to the first antigen. Other bispecific forms include G) dual scFv, H) single-arm scFv-mAb, I) scFv-mAb, J) bispecific mAb, K) single-arm central scFv, L) mAb-Fv, M) central Fv, and N) trident.

[0062] Figure 15 The variable heavy chain and variable light chain sequences of 1A7 (an exemplary phage-derived CD28 binding domain) and the sequence of XENP28428 (an anti-CD28 mAb based on 1A7 and an IgG1 backbone with E233P / L234V / L235A / G236del / S267K ablation variants) are depicted. CDRs are underlined, and slashes indicate the boundary between the variable region and the constant domain. As described herein and for every sequence containing a CDR, the precise identification of CDR locations may vary slightly depending on the numbering used (as shown in Table 2), and therefore, this document includes not only underlined CDRs but also CDRs included within the VH and VL domains using other numbering systems. Furthermore, for all sequences in the figure, these VH and VL sequences may be used in scFv or Fab form.

[0063] Figure 16A-16F The sequence of an affinity-optimized variable heavy chain domain from the anti-CD28 clone 1A7 was depicted. It should be noted that the variable heavy chain domain can be associated with… Figure 15 Pairing with any other variable light chain domains depicted in Figures 17 and 18 (including SEQ ID NO: XXX-YYY).

[0064] Figure 17A-17I The sequence of an affinity-optimized variable light chain domain from the anti-CD28 clone 1A7 was depicted. It should be noted that the variable heavy chain domain can be associated with… Figure 15 Pairing with any other variable light chain domains depicted in Figures 16 and 18 (including SEQ ID NO: XXX-YYY).

[0065] Figures 18A-18C Illustrative sequences of affinity-optimized 1A7 VH / VL pairs are depicted. It should be noted that these pairs can be constructed in either Fab or scFv form. Furthermore, in the scFv form, these pairs can be constructed in either VHVL or VLVH orientations.

[0066] Figure 19A and Figure 19B The common frame region (FR) and complementarity determination region (CDR) of the anti-CD28 clone 1A7 variable heavy chain domain variant and variable light chain domain variant are depicted (as in Kabat).

[0067] Figure 20 An illustrative affinity engineering of the 1A7 VH / VL pair and its binding affinity is depicted in the context of scFv (in the form of 1+1 Fab-scFv-Fc bsAb).

[0068] Figures 21A-21H Variable heavy chain sequences and variable light chain sequences of additional CD28 binding domains that can be used in the ENPP3 x CD28 bsAb of this invention are depicted. As described herein and for every sequence containing a CDR herein, the precise identification of the CDR position may vary slightly depending on the numbering used (as shown in Table 2), and therefore this document includes not only underlined CDRs but also CDRs included within the VH and VL domains using other numbering systems. Furthermore, for all sequences in the figures, these VH and VL sequences may be used in scFv form or Fab form.

[0069] Figure 22A-22KExemplary variable heavy chain and variable light chain sequences of the ENPP3 binding domain that can be used in the ENPP3 x CD28 bsAb of this invention are depicted. As described herein and for each sequence containing a CDR, the identification of the CDR position can be determined according to the numbering scheme shown in Table 2. Furthermore, for all sequences in the figure, these VH and VL sequences can be used in scFv form or in Fab form.

[0070] Figure 23 A) the classical T cell / APC interaction is depicted, and B) the classical T cell / APC interaction is replicated by combining a CD3 bispecific antibody with a CD28 bispecific antibody. In the classical T cell / APC interaction, there is a first signal (signal 1) provided by the reactivity of the TCR with peptide-MHC and a second signal (signal 2) provided by the crosslinking of CD80 / CD86 expressed on the APC with CD28, which together fully activate the T cell. In contrast, in treatment using a CD3 bispecific antibody, only the first signal is provided. The CD28 signal can be provided by a CD28 bispecific antibody, designed to promote activation and proliferation through CD28 co-stimulation. In some embodiments, TAA1 and TAA2 can be different antigens. In some embodiments, TAA1 and TAA2 can be the same antigen but with different epitopes. In some embodiments, TAA1 and TAA2 can be the same antigen and the same epitope.

[0071] Figures 24A-24H Illustrative sequences of ENPP3 x CD28 bsAbs in the form of 1+1 Fab-scFv-Fc are depicted. The slashes indicate the boundaries between the variable domain and other domains (e.g., domain linkers and constant domains). Additionally, the nomenclature conventions specify the orientation of the scFv from the N-terminus to the C-terminus. As described herein and for each sequence containing a CDR, the identification of the CDR position can be determined according to the numbering scheme shown in Table 2. It should be noted that ENPP3 x CD28 bsAbs can utilize 90, 95, 98, and 99% identical (as defined herein) sequences of the variable region, Fc region, and constant domain, and / or contain 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid substitutions. Furthermore, each sequence outlined herein may include or exclude the M428L / N434S variant in one or preferably two Fc domains; inclusion of M428L / N434S results in a longer serum half-life.

[0072] Figure 25A-25FSequences of exemplary anti-CD3 binding domains suitable for use with CD28 bispecific antibodies of the present invention are depicted. CDR is underlined, scFv adapter is double-underlined (in the sequence, the scFv adapter is a positively charged scFv(GKPGS)4 adapter (SEQ ID NO: XXX), however, those skilled in the art will understand that this adapter can be replaced by other adapters, including uncharged or negatively charged adapters, some of which are depicted in...). Figure 6 (in the middle), and the slash indicates the boundary of the variable structural domain. Additionally, the naming convention specifies the orientation of the scFv from the N-end to the C-end. As described herein and for every sequence containing a CDR, the precise identification of the CDR location may vary slightly depending on the numbering used (as shown in Table 2), and therefore this document includes not only underlined CDRs but also CDRs included within the VH and VL structural domains using other numbering systems. Furthermore, for all sequences in the figure, these VH and VL sequences can be used in scFv form or Fab form.

[0073] Figure 26 The purified T cells were depicted with RXF-393 (ENPP3). + Induction of A) IL2 and B) IFNγ secretion when tumor cells are incubated with 1 µg / ml of illustrative ENPP3x CD3 bsAb and dose-escalating illustrative ENPP3x CD28 bsAb.

[0074] Figure 27 The release of IL-2 in purified T cells from 14 unique T cell donors was depicted. These cells were incubated with TUHR10TKB ENPP3+ target cells at a 1:1 E:T ratio and treated with a combination of 10 µg / ml ENPP3 x CD28 bsAb and 1 µg / ml illustrative ENPP3 x CD3.

[0075] Figure 28 The study depicted the target cell killing of VMRC-RCW target cells mediated by ENPP3 x CD28 enhanced by (A) ENPP3 x CD3 and (B) B7H3 x CD3.

[0076] Figure 29 The induction of BCIXL by ENPP3 x CD28 amplification in (A) ENPP3 x CD3 and (B) B7H3 x CD3-mediated RTCC was depicted.

[0077] Figure 30The study depicted how ENPP3 x CD28 enhances T cell cytotoxicity in CD3 x ENPP3-mediated RTCC, as measured by granzyme B and CD107a.

[0078] Figure 31A and Figure 31B The study depicted how ENPP3 increases T cell activation and upregulates PD1 in RTCC mediated by A) ENPP3 x CD3 and B) B7H3 x CD3.

[0079] Figure 32 The study depicted the upregulation of intracellular IFNg by ENPP3 x CD28 in ENPP3 x CD3-mediated RTCC.

[0080] Figure 33 The induction of IL2 secretion was described when purified T cells were incubated with VMRC-RCW tumor cells (approximately 13K ENPP3 antigens) at a 1:1 E:T ratio and with 1 µg / ml of illustrative ENPP3 x CD3 and dose-escalating ENPP3 x CD28 bsAb in various forms.

[0081] Figure 34 The study depicted the induction of IL2 secretion when purified T cells were incubated with VMRC-RCW tumor cells (approximately 13K ENPP3 antigens) at a 1:1 E:T ratio and with 1 µg / ml of illustrative ENPP3 x CD3 and dose-escalating ENPP3 x CD28 bsAbs (both in 1+1 Fab-scFv-Fc form) containing multiple ENPP3 and CD28 binding domains.

[0082] Figure 35 The study demonstrated that XENP46666 enhanced IL2 release at 24 hours in antigen-specific RTCC assays, with a 2-3 fold increase compared to T cells alone + ENPP3 + pp65-MDA-MB-231 target cells.

[0083] Figure 36 The antitumor activity of the combination of ENPP3 x CD28 bsAb XENP46666 and XENP46667 with ENPP3 x CD3 bsAb was described.

[0084] Figure 37 Pharmacokinetic data for XENP46666 and its alternative XENP46674 in cynomolgus monkeys were presented.

[0085] Figure 38 The study described how ENPP3 x CD28 enhances EpCAM x CD3-mediated cell killing.

[0086] Figure 39 The ENPP3 x CD28-enhanced EpCAM x CD3-mediated IFNg release was described.

[0087] Figures 40A-40B The ENPP3 x CD28 bsAb in the form of a single-arm central scFv is depicted.

[0088] Figure 41A-41F The ENPP3 x CD28 bsAb in the form of 2 + 1 mAb-scFv was depicted.

[0089] Figures 42A-42F The ENPP3 x CD28 bsAb in the form of 2 + 1 Fab2-scFv-Fc is depicted. Figure 43 The increased thermal stability of disulfide bond-stabilized 1A7 scFv was described.

[0090] Figure 44 The CD28 affinity of additional 1A7scFv variants obtained from different experiments in the context of different CD28 multispecific antibodies was depicted, with or without disulfide bond stabilization (as indicated by [SS]). Although all scFvs depicted below use the scFv adapter (GKPGS)4, any scFv adapter (including...) can be used. Figure 5 (Those in the example). In addition, scFv can be in either orientation, i.e., VH-scFv connector-VL or VL-scFv connector-VH.

[0091] Figures 45A-45E An illustrative 1A7-based scFv is described, featuring an engineered cysteine ​​residue for disulfide bond stabilization. While all scFvs described below use the scFv adapter (GKPGS)4, any scFv adapter (including...) can be used. Figure 5 (Those in the example). In addition, the scFv can be in either orientation, namely VH-scFv connector-VL or VL-scFv connector-VH, but only the former is shown.

[0092] Figure 46 The KD values ​​of steady-state (SS) and dynamic (1:1 Langmuir) fits of the ENPP3 binding domain to human and cynomolgus monkey ENPP3 were depicted.

[0093] Figures 47A-47C Novel 1A7 VHs (although they can also pair with any 1A7 VL) that have been engineered for affinity pairing with IGKV1-39 germline sequences are depicted, along with shared frame region (FR) and complementarity-determining region (CDR).

[0094] Figure 48 The affinity of the novel 1A7 variant paired with the IGKV1-39 hominin VL for human CD28 was described. Detailed Implementation

[0095] I. Overview Exonucleotide pyrophosphatase / phosphodiesterase family member 3 (ENPP3) is a type II transmembrane glycoprotein belonging to the nucleotide pyrophosphatase / phosphodiesterase family. These proteins are involved in the hydrolysis of extracellular nucleotides and also possess ATPase and ATP pyrophosphatase activities. ENPP3 is overexpressed in a variety of cancers, including most renal cell carcinomas and some liver cancers. Due to its overexpression in multiple tumors, ENPP3 is a candidate for targeted therapeutics.

[0096] In cancer treatment, the activation of T cells is being extensively studied. Complete activation and differentiation of T cells require multiple signals. For example... Figure 23 As shown in Figure A, Signal 1 (promoted by the T cell receptor (TCR) recognition peptide-MHC (pMHC) complex) is absolutely necessary for T cell activation. Signal 2 synergizes with and amplifies Signal 1, typically provided by the interaction of CD28 ligands CD80 and CD86 with CD28 itself. Although CD28 co-stimulation alone is generally inert, when combined with Signal 1 activation, it promotes additional activation, survival, and proliferation signals, including IL-2 secretion. Since CD80 and CD86 are expressed only naturally by specialized antigen-presenting cells (APCs), the degree of CD28 co-stimulation in the tumor setting can be highly variable. Therefore, this invention targets a novel class of tumor-targeting anti-CD28 x anti-ENPP3 bispecific antibodies whose CD28 CD80 / CD86 co-stimulation mimics the CD28 CD80 / CD86 co-stimulation, thereby providing an artificial source of Signal 2 (…). Figure 23 B). It is worth noting that signal 1 can be provided by the recognition of the tumor cell’s natural TCR:pMHC, or it can be provided by a combination of CD28 bispecific antibodies and CD3 bispecific antibodies (e.g., anti-CD3 x anti-ENPP3), thus mimicking signal 1.

[0097] Therefore, this article presents novel anti-CD28 x anti-ENPP3 (also known as "αCD28 x αENPP3", and sometimes referred to as "CD28 x ENPP3") bispecific antibodies and methods for using such antibodies to treat ENPP3-related cancers. In many cases, these bispecific antibodies are heterodimers. The subject αCD28 x αENPP3 antibody is able to bind agonistically to the CD28 co-stimulatory molecule on T cells and target ENPP3 on tumor cells expressing ENPP3. Thus, such antibodies selectively enhance antitumor activity at tumor sites expressing ENPP3 while minimizing peripheral toxicity. The subject antibody presented in this article is particularly useful for enhancing antitumor activity, either alone as a monotherapy or when used in combination with other anticancer therapies, as described more broadly herein.

[0098] Therefore, on the one hand, this document provides heterodimeric antibodies that bind to two different antigens, for example, said antibodies are “bispecific” because they bind to two different target antigens, typically CD28 and ENPP3 as described below. These heterodimeric antibodies can bind to each target antigen monovalently (e.g., having a single antigen-binding domain, such as a variable heavy chain domain and a variable light chain domain pair) or bivalently (having two antigen-binding domains, each binding the antigen independently). In some embodiments, the heterodimeric antibodies provided herein comprise a CD28-binding domain and an ENPP3-binding domain (e.g., heterodimeric antibodies in the form of a “1+1 Fab-scFv-Fc” as described herein, and are therefore bispecific and bivalent). In other embodiments, the heterodimer antibodies provided herein comprise a CD28-binding domain and two ENPP3-binding domains (e.g., heterodimer antibodies described herein in the form of “2+1 Fab2-scFv-Fc”, “2+1 mAb-scFv”, and “2+1 Fab2-Fc x scFv-Fc”, and are therefore bispecific but trivalent because they contain three antigen-binding domains (ABDs)). The heterodimer antibodies provided herein are based on the use of different monomers containing amino acid substitutions (i.e., “skewed variants”) that cause the formation of heterodimers to be “skewed” relative to homodimers, as outlined more fully below. In some embodiments, the heterodimer antibodies are also conjugated to purification variants (e.g., “pI variants”), thereby allowing for the simple purification of heterodimers from homodimers, as similarly outlined below. The heterodimeric bispecific antibodies provided typically rely on the use of engineered or variant Fc domains that can self-assemble in production cells to generate heterodimeric proteins, as well as methods for generating and purifying such heterodimeric proteins.

[0099] II. Nomenclature The nomenclature for specific antigen-binding domains (e.g., ENPP3 and CD28 binding domains) uses the form "Hx.xx_Ly.yy", where the number is a unique identifier for a specific variable strand sequence. Thus, for example, the CD28 binding domain "1A7[CD28]_H1_L1" ( Figure 15 The scFv includes a variable heavy chain domain H1 and a variable light chain domain L1. When these sequences are used as scFv, the name "H1_L1" indicates that the binding domain comprises a combination of the variable heavy chain domain "H1" and the variable light chain domain "L1," and is VH-joint-VL oriented from the N-terminus to the C-terminus. A molecule having the same heavy chain and light chain variable domain sequences but in reverse order (VL-joint-VH oriented from the N-terminus to the C-terminus) will be designated "L1_H1". Similarly, different constructs may "mix and match" heavy and light chains, as will be apparent from the sequence listing and figures.

[0100] III. Definition To provide a more complete understanding of this application, several definitions are set forth below. These definitions are intended to cover grammatically equivalent terms.

[0101] In this article, “CD28,” “differentiation cluster 28,” and “Tp44” (e.g., Genebank accessions NP_001230006 (human), NP_001230007 (human), NP_006130 (human), and NP_031668 (mouse)) refer to the B7 receptor expressed on T cells, which provides the co-stimulatory signal required for T cell activation and survival. In addition to the T cell receptor (TCR), T cell stimulation via CD28 also provides a potent signal for the production of various interleukins. CD28 is the receptor for CD80 (B7.1) and CD86 (B7.2) proteins. CD28 includes an intercellular domain with a YMNM motif, which is essential for recruiting SH2-containing proteins (particularly PI3K). CD28 also includes two proline-rich motifs capable of binding SH3-containing proteins. Figure 1 depicts an exemplary CD28 sequence. Unless otherwise stated, references to CD28 refer to the human CD28 sequence.

[0102] In this article, "ENPP3" or "exonucleotide pyrophosphatase / phosphodiesterase family member 3" or "B10" or "CD203c" or "NPP3" or "PD-IBETA" or "PDNP3" refer to a type II transmembrane glycoprotein belonging to the nucleotide pyrophosphatase / phosphodiesterase family. These proteins are involved in the hydrolysis of extracellular nucleotides and also possess ATPase and ATP pyrophosphatase activities. For example, Figure 2 depicts the ENPP3 sequence. ENPP3 is overexpressed in a variety of cancers, including renal cell carcinoma.

[0103] In this text, "ablation" means a reduction or removal of activity. Therefore, for example, "ablation of FcγR binding" means that an Fc region amino acid variant has less than 50% initial binding compared to an Fc region without the specific variant, with a loss of activity exceeding 70-80-90-95-98% being preferred, and generally, the activity is below the binding level detectable in Biacore, SPR, or BLI assays. Particularly useful in the ablation of FcγR binding is… Figure 5 Those shown are typically added to two monomers.

[0104] As used herein, “ADCC” or “antibody-dependent cell-mediated cytotoxicity” refers to a cell-mediated reaction in which nonspecific cytotoxic cells expressing FcγR recognize antibodies bound to target cells and subsequently cause lysis of the target cells. ADCC is associated with FcγRIIIa binding; increased binding to FcγRIIIa leads to increased ADCC activity.

[0105] As used in this article, “ADCP” refers to a cell-mediated response in which nonspecific phagocytes expressing FcγR recognize antibodies bound to target cells and subsequently induce phagocytosis of the target cells.

[0106] As used herein, the term "antibody" is used extensively. The antibodies described herein may take many forms as described herein, including conventional antibodies as well as antibody derivatives, fragments, and mimics.

[0107] Traditional immunoglobulin (Ig) antibodies are Y-shaped tetramers. Each tetramer typically consists of two pairs of identical polypeptide chains, each pair having a "light chain" monomer (typically with a molecular weight of about 25 kDa) and a "heavy chain" monomer (typically with a molecular weight of about 50-70 kDa).

[0108] Other useful antibody forms include, but are not limited to, the “1+1 Fab-scFv-Fc” and “2+1 Fab2-scFv-Fc”, “2+1 mAb-scFv”, and “2+1 Fab2-Fc x scFv-Fc” forms provided herein (see, for example, Figure 14). Other useful antibody forms include, but are not limited to, “1+1 common light chain” and “2+1 common light chain”, “mAb-Fv”, “mAb-scFv”, “central Fv”, “single-arm scFv-mAb”, “scFv-mAb”, “double scFv”, and “trident” antibody forms (Figure 14). See also US20180127501A1, which is incorporated herein by reference, particularly the relevant portions relating to antibody forms (see, for example, Figure 2 of US20180127501A1).

[0109] Antibody heavy chains typically include variable heavy chain (VH) domains, comprising vhCDR1-3, and Fc domains, comprising CH2-CH3 monomers. In some embodiments, the antibody heavy chain includes a hinge and a CH1 domain. A conventional antibody heavy chain is a monomer organized from the N-terminus to the C-terminus as follows: VH-CH1-hinge-CH2-CH3. The CH1-hinge-CH2-CH3 domains are collectively referred to as the antibody heavy chain's "constant domains" or "constant regions," which contain five distinct classes or "isotypes": IgA, IgD, IgG, IgE, and IgM.

[0110] 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 immunoglobulin IgG subclasses, several immunoglobulin domains are present in the heavy chain. Herein, "immunoglobulin (Ig) domain" refers to a region of an immunoglobulin with a distinctive tertiary structure. This invention focuses on the heavy chain domains, including constant heavy chain (CH) domains and hinge domains. In the context of IgG antibodies, each IgG isotype has three CH regions. Therefore, in the context of IgG, the "CH" domains are as follows: "CH1" refers to positions 118-215 according to the EU index as in Kabat. "Hinge" refers to positions 216-230 according to the EU index as in Kabat. "CH2" refers to positions 231-340 according to the EU index as in Kabat, and "CH3" refers to positions 341-447 according to the EU index as in Kabat. As shown in Table 1, the exact numbering and arrangement of the heavy chain domains may vary in different numbering systems. As shown in this document and described below, pI variants may reside in one or more CH regions and hinge regions discussed below.

[0111] It should be noted that IgG1 has different allotypes, exhibiting polymorphism at 356 (D or E) and 358 (L or M). The sequences described herein use the 356E / 358M allotype; however, other allotypes are included herein. That is, any sequence including the IgG1 Fc domain included herein may have 356D / 358L, replacing the 356E / 358M allotype. It should be understood that therapeutic antibodies may also comprise hybrids of allotypes and / or subtypes. For example, as shown in U.S. Publication 2009 / 0163699, which is incorporated herein by reference, in some embodiments, the antibodies of the present invention comprise human IgG1 / G2 hybrids.

[0112] As used herein, “Fc” or “Fc region” or “Fc domain” refers to a polypeptide containing a constant region of an antibody, in some cases excluding all or a portion of the first constant region immunoglobulin domain (e.g., CH1), and in some cases optionally including all or a portion of the hinge. For IgG, the Fc domain comprises 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 CH2-CH3 and hinge-CH2-CH3 from the N-terminus to the C-terminus. In some embodiments, the Fc domain is an Fc domain derived from IgG1, IgG2, IgG3, or IgG4, wherein IgG1 hinge-CH2-CH3 and IgG4 hinge-CH2-CH3 are particularly useful in many embodiments. Additionally, in the case of the human IgG1 Fc domain, the hinge may include a C220S amino acid substitution. Additionally, in the case of the human IgG4 Fc domain, the hinge may include an S228P amino acid substitution. Although the boundaries of the Fc region can vary, the human IgG heavy chain Fc region is generally defined to include residues E216, C226, or A231 at its carboxyl terminus, where the numbering is based on the EU index, as in Kabat. In some embodiments, as described more fully below, the Fc region is modified with amino acids, for example, to alter its binding with one or more FcγRs or with FcRn.

[0113] In this article, "heavy chain constant region" refers to the CH1-hinge-CH2-CH3 portion of an antibody (or a fragment thereof), excluding the variable heavy chain domain; in the EU numbering of human IgG1, this is amino acid 118-447. In this article, "heavy chain constant region fragment" refers to a heavy chain constant region containing fewer amino acids from either the N-terminus or C-terminus, but still retaining the ability to form a dimer with another heavy chain constant region.

[0114] Another type of domain in heavy chains is the hinge region. In this document, “hinge,” “hinge region,” “antibody hinge region,” or “hinge domain” refers to a flexible polypeptide containing amino acids between the first and second constant domains of an antibody. Structurally, the IgG CH1 domain terminates at EU position 215, and the IgG CH2 domain begins at residue EU position 231. Therefore, for IgG, the antibody hinge is defined herein as encompassing positions 216 (E216 in IgG1) through 230 (P230 in IgG1), where the numbering is based on EU indexing as in Kabat. In some cases, “hinge fragments” are used, which contain fewer amino acids at either end or both ends of the hinge domain, either N-terminus or C-terminus. As described herein, pI variants can also be prepared in the hinge region. In this document, many antibodies have at least one cysteine ​​replaced by a serine at position 220 according to the EU number (hinge region). Generally, this modification occurs on the "scFv monomer" side of most sequences described herein (when using the 1+1 or 2+1 form), although it can also occur on the "Fab monomer" side or on both sides to reduce disulfide bond formation. In particular, the sequences described herein include the substitution of one or both of these cysteine ​​residues (C220S).

[0115] Those skilled in the art will understand that the exact numbering and arrangement of the heavy chain constant region domains (i.e., CH1, hinge, CH2, and CH3 domains) may vary in different numbering systems. A useful comparison of the heavy chain constant region numbering according to the EU and Kabat is provided below, 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, all of which are incorporated herein by reference in their entirety.

[0116] Table 1

[0117] Antibody light chains typically contain two domains: variable light chain domains (VL), which include light chain CDRs vlCDR1-3; and constant light chain regions (often referred to as CL or Cκ). Antibody light chains are typically organized from the N-terminus to the C-terminus as follows: VL-CL.

[0118] In this document, "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 target antigens such as those discussed herein (e.g., ENPP3 or CD28). 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 containing three CDRs: vhCDR1, vhCDR2, and vhCDR3 variable heavy chain CDRs, and vlCDR1, vlCDR2, and vlCDR3 variable light chain CDRs. The CDRs are located within the variable heavy chain domains (vhCDR1-3) and the variable light chain domains (vlCDR1-3). The variable heavy chain domains and variable light chain domains originate from the Fv region.

[0119] This invention provides a large number of different CDR groups. In this case, a “full CDR group” comprises three variable light chain CDRs 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. Additionally, as summarized more fully herein, when using heavy and light chains (e.g., when using Fab), the variable heavy chain domain and the variable light chain domain can be located on separate polypeptide chains, or, in the case of an scFv sequence, on a single polypeptide chain.

[0120] Those skilled in the art will understand that the exact numbering and arrangement of CDRs may vary in different numbering systems. However, it should be understood that the disclosure of a variable heavy chain sequence and / or a variable light chain sequence includes the disclosure of the associated (inherent) CDR. Therefore, the disclosure of each variable heavy chain region is the disclosure of vhCDRs (e.g., vhCDR1, vhCDR2, and vhCDR3), and the disclosure of each variable light chain region is the disclosure of vlCDRs (e.g., vlCDR1, vlCDR2, and vlCDR3). A useful comparison of CDR numbering is provided below; see [link to relevant documentation]. Lafranc et al. , Dev. Comp. Immunol. 27(1):55-77(2003): Table 2

[0121] Throughout this specification, when referring to residues in the variable domains (roughly residues 1-107 of the light chain variable region and residues 1-113 of the heavy chain variable region), the Kabat numbering system is generally used, and for the Fc region, the EU numbering system is generally used (e.g., Kabat et al., ibid. (1991)).

[0122] CDRs (Corrective Derivatives) facilitate the formation of antigen-binding domains and antigen-binding sites on antibodies, or more specifically, epitope-binding sites. An epitope is a determinant cluster that interacts with a specific antigen-binding site (called a complementary site) in the variable region of an antibody molecule. Epitopes are molecular groups such as amino acid or sugar side chains and typically possess specific structural and charge characteristics. A single antigen may have more than one epitope.

[0123] Epitopes can contain amino acid residues that directly participate in binding (also known as the immunodominant component of the epitope) and other amino acid residues that do not directly participate in binding, such as amino acid residues that are effectively blocked by specific antigen-binding peptides; in other words, the amino acid residues are within the footprint of specific antigen-binding peptides.

[0124] 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 generated by adjacent amino acid residues in the polypeptide chain. The difference between conformational and non-conformational epitopes lies in the loss of binding to the former rather than the latter in the presence of denaturing solvents.

[0125] In a unique spatial conformation, an epitope typically comprises at least three, and more commonly at least five, or eight to ten amino acids. Antibodies recognizing the same epitope can be verified using a simple immunoassay that demonstrates the ability of one antibody to block the binding of another antibody to a target antigen, such as "binning." As outlined below, the present invention includes not only the antigen-binding domains and antibodies listed herein, but also those that compete for binding to epitopes bound to the listed antigen-binding domains.

[0126] In some implementations, the six CDRs of the antigen-binding domain are provided by a variable heavy chain domain and a variable light chain domain. In the “Fab” form, a set of six CDRs is provided by two distinct polypeptide sequences, a variable heavy chain domain (vh or VH; containing vhCDR1, vhCDR2, and vhCDR3) and a variable light chain domain (vl or VL; containing vlCDR1, vlCDR2, and vlCDR3), wherein the C-terminus of the vh domain is attached to the N-terminus of the CH1 domain of the heavy chain and the C-terminus of the vl domain is attached to the N-terminus of the constant light chain domain (and thus forms a light chain). In the scFv form, the vh and vl domains are typically covalently linked into a single polypeptide sequence using a linker (“scFv linker”) as outlined herein, which can be (starting from the N-terminus) vh-linker-vl or vl-linker-vh, the former being generally preferred (including optional linkers on each side, depending on the form used). Generally, the C-end of the scFv structural domain is connected to the N-end of all or part of the hinge in the second monomer.

[0127] As used herein, a “variable region” or “variable domain” refers to a region in an immunoglobulin that contains one or more Ig domains generally encoded by any of the Vκ, Vλ, and / or VH genes that constitute the κ, λ, and heavy chain immunoglobulin loci, respectively, and contains a CDR that confers antigen specificity. Therefore, a “variable heavy chain domain” pairs with a “variable light chain domain” to form an antigen-binding domain (“ABD”). Additionally, each variable domain contains three hypervariable regions (“complementarity-determining regions”, “CDRs”) arranged from the amino terminus to the carboxyl terminus (vhCDR1, vhCDR2, and vhCDR3 of the variable heavy chain domain, and vlCDR1, vlCDR2, and vlCDR3 of the variable light chain domain) and four frame (FR) regions: FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4.

[0128] As used herein, "Fab" or "Fab region" refers to an antibody region containing the VH, CH1, VL, and CL immunoglobulin domains, which are typically located on two different polypeptide chains (e.g., VH-CH1 on one chain and VL-CL on another). Fab can refer to this region alone, or to this region in the context of the bispecific antibody of this invention. In the context of Fab, in addition to the CH1 and CL domains, Fab also includes the Fv region.

[0129] As used herein, “Fv” or “Fv fragment” or “Fv region” refers to an antibody region containing the VL and VH domains. Fv regions can be constructed as Fab (as discussed above, typically two distinct peptides that also include constant regions as outlined above) and as single-chain Fv (scFv) forms, where the vl and vh domains are contained within a single peptide, typically linked by linkers as discussed herein.

[0130] In this document, "single-chain Fv" or "scFv" refers to a variable heavy-chain domain typically covalently connected to a variable light-chain domain using scFv joints as discussed herein to form an scFv or scFv domain. The scFv domain can be oriented in either direction from the N-terminus to the C-terminus (vh-joint-vl or vl-joint-vh). In the sequences depicted in the sequence listing and figures, the order of the vh and vl domains is indicated by name; for example, H.X_L.Y indicates vh-joint-vl from the N-terminus to the C-terminus, and L.Y_H.X indicates vl-joint-vh.

[0131] Some embodiments of the subject antibody presented herein include at least one scFv domain, which, while not naturally occurring, typically comprises a variable heavy chain domain and a variable light chain domain linked together by an scFv linker. As outlined herein, while the scFv domain is typically oriented VH-scFv linker-VL from the N-terminus to the C-terminus, this can be reversed for any scFv domain (or those constructed using Fab's vh and vl sequences) to VL-scFv linker-VH, with optional linkers at one or both ends, depending on the form.

[0132] In this article, "modification" or "variation" means the substitution, insertion, and / or deletion of amino acids in a polypeptide sequence, or an alteration of the portion of the protein chemically linked to it. For example, a modification could be a change in the structure of a carbohydrate or PEG linked to a protein. In this article, "amino acid modification" means the substitution, insertion, and / or deletion of amino acids in a polypeptide sequence. For clarity, unless otherwise stated, amino acid modifications always refer to amino acids encoded by DNA, such as the 20 amino acids that have codons in DNA and RNA.

[0133] In this document, "amino acid substitution" or "replacement" means replacing an amino acid at a specific position in the parental polypeptide sequence with a different amino acid. Specifically, in some embodiments, substitution is for an amino acid that is not naturally present at that specific position, i.e., not naturally present in any organism or any living organism. For example, substitution of E272Y refers to a variant polypeptide, in this case, the Fc variant, where the glutamic acid at position 272 is replaced by a tyrosine. For clarity, a protein engineered to alter its nucleic acid coding sequence but not its starting amino acid (e.g., exchanging CGG (encoding arginine) for CGA (still encoding arginine) to increase expression levels in a host organism) is not "amino acid substitution"; that is, although a new gene encoding the same protein is generated, if the protein has the same amino acid at its starting specific position, then the protein is not an amino acid substitution.

[0134] As used herein, “amino acid insertion” or “insertion” means the addition of an amino acid sequence at a specific position in the parent polypeptide sequence. For example, -233E or 233E indicates the insertion of glutamic acid after position 233 and before position 234. Similarly, -233ADE or A233ADE indicates the insertion of AlaAspGlu after position 233 and before position 234.

[0135] As used herein, “amino acid deletion” or “deletion” means the removal of an amino acid sequence at a specific position in the parent polypeptide sequence. For example, E233- or E233#, E233(), or E233del indicates the deletion of glutamic acid at position 233. Additionally, EDA233- or EDA233# indicates the deletion of the sequence GluAspAla starting at position 233.

[0136] As used herein, “variant protein” or “protein variant” or “variant” means a protein that differs from the parent protein due to 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 cannot be aligned with the parent protein using alignment procedures (such as those described below). Generally, using alignment procedures (such as BLAST) described below, variant proteins (such as the variant Fc domain outlined herein) are typically found to be at least 75, 80, 85, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to the parent protein.

[0137] As used in this article, “variant” also refers to a specific amino acid modification that imparts a particular function (e.g., “heterodimerization variant”, “pI variant”, “ablation variant”, etc.).

[0138] As described below, in some embodiments, the parental polypeptide (e.g., the Fc parental polypeptide) is a human wild-type sequence, such as the heavy chain constant domain or Fc region derived from IgG1, IgG2, IgG3, or IgG4; however, human sequences with variants may also be used as “parental polypeptides,” for example, including the IgG1 / 2 hybrid of U.S. Publication 2006 / 0134105. The protein variant sequences used herein preferably have at least about 80% identity with the parental protein sequence, and most preferably at least about 90% identity, more preferably at least about 95-98-99% identity. Therefore, as used herein, “antibody variant” or “variant antibody” means an antibody that differs from the parental antibody due to at least one amino acid modification; as used herein, “IgG variant” or “variant IgG” means an antibody that differs from the parental IgG (again, in many cases, derived from a human IgG sequence) due to at least one amino acid modification; and as used herein, “immunoglobulin variant” or “variant immunoglobulin” means an immunoglobulin sequence that differs from the parental immunoglobulin sequence due to at least one amino acid modification. As used in this article, "Fc variant" or "variant Fc" refers to a protein that contains amino acid modifications in its Fc domain compared to the Fc domain of human IgG1, IgG2, or IgG4.

[0139] As used herein, “Fc variant” or “variant Fc” means a protein that contains an amino acid modification in its 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 with a serine substitution at position 434 relative to a parental Fc polypeptide, where the numbering is based on the EU index. Similarly, M428L / N434S defines an Fc variant with substitutions of M428L and N434S relative to a parental Fc polypeptide. The identity of the WT amino acids may not be indicated, in which case the aforementioned variant is referred to as 428L / 434S. It should be noted that the order of substitutions provided is arbitrary, that is, for example, 428L / 434S is the same Fc variant as 434S / 428L, and so on. For all positions involving antibodies or their derivatives and fragments (e.g., Fc domains) discussed herein, unless otherwise stated, the amino acid position numbering is based on the EU index. The “EU index” or “EU index as in Kabat” or “EU number” scheme refers to the numbering of EU antibodies (Edelman et al., 1969, Proc Natl Acad Sci USA 63:78-85, hereby incorporated in its entirety by reference). Modifications can be additions, deletions, or substitutions.

[0140] Generally, the variant Fc domain has at least about 80, 85, 90, 95, 97, 98, or 99% identity with the corresponding parental IgG Fc domain (using the identity algorithm discussed below, one implementation utilizing the BLAST algorithm 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 parental 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. Furthermore, as discussed herein, the variant Fc domains described herein retain the ability to form dimers with another Fc domain, as measured using known techniques such as non-denaturing gel electrophoresis.

[0141] As used herein, "protein" means at least two covalently linked amino acids, including proteins, polypeptides, oligopeptides, and peptides. Additionally, the polypeptide constituting the antibody of this invention may include synthetic derivatization, glycosylation, PEGylation, cyclic arrangement, cyclization, linker to other molecules, fusion with a protein or protein domain, and addition of peptide tags or labels.

[0142] As used in this article, "residue" refers to a position in a protein and its associated amino acid identity. For example, asparagine 297 (also known as Asn297 or N297) is the residue at position 297 in human antibody IgG1.

[0143] As used herein, “IgG subclass modification” or “isotype modification” refers to an amino acid modification that converts one amino acid of an IgG isotype into the corresponding amino acid in a different aligned IgG isotype. For example, because IgG1 contains tyrosine at EU position 296 and IgG2 contains phenylalanine, the F296Y substitution in IgG2 is considered an IgG subclass modification.

[0144] As used in this article, “non-naturally occurring modification” means an amino acid modification that is not of the same type. For example, because human IgG does not contain serine at position 434, the substitution 434S in IgG1, IgG2, IgG3, or IgG4 (or their hybrids) is considered a non-naturally occurring modification.

[0145] As used in this article, "amino acid" and "amino acid identity" refer to one of the 20 naturally occurring amino acids encoded by DNA and RNA.

[0146] As used herein, "effective function" refers to a biochemical event resulting from the interaction between the antibody's Fc region and an Fc receptor or ligand. Effector functions include, but are not limited to, ADCC, ADCP, and CDC.

[0147] 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 the Fc receptor family homologous to FcγR (Davis et al., 2002, Immunological Reviews 190:123-136, incorporated herein by reference in its entirety). Fc ligands may include undiscovered Fc-binding molecules. 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.

[0148] As used in this article, “Fcγ receptor” and “FcγR (FcgammaR)” refer to any member of the protein family that binds to the Fc region of IgG antibodies and is encoded by the 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 alloforms H131 and R131), FcγRIIb (including FcγRIIb-1 and FcγRIIb-2), and FcγRIIc; and FcγRIII (CD16), including isoforms FcγRIIIa (including alloforms V158 and F158) and FcγRIIIb (including alloforms FcγRIIb-NA1 and FcγRIIb-NA2) (Jefferis et al., 2002, Immunol Lett 82:57-65, incorporated herein by reference in its entirety) and any undiscovered human FcγR or FcγR isoforms or alloforms. FcγR can be derived from any organism, including but not limited to humans, mice, rats, rabbits, and monkeys. Mouse FcγR includes but is not limited to FcγRI (CD64), FcγRII (CD32), FcγRIII (CD16), and FcγRIII-2 (CD16-2), as well as any undiscovered mouse FcγR or FcγR isoforms or allotypes.

[0149] 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 originate from any organism, including but not limited to humans, mice, rats, rabbits, and monkeys. As is known in the art, functional FcRn proteins comprise two polypeptides, commonly referred to as the heavy chain and the light chain. The light chain is β-2-microglobulin, and the heavy chain is encoded by the FcRn gene. Unless otherwise indicated herein, FcRn or FcRn protein refers to the complex of the FcRn heavy chain and β-2-microglobulin. Various FcRn variants are used to increase binding to the FcRn receptor and, in some cases, to increase serum half-life. An “FcRn variant” is an amino acid modification that contributes to increased binding to the FcRn receptor, and suitable FcRn variants are shown below.

[0150] As used herein, “parental polypeptide” means the starting polypeptide that is subsequently modified to generate a variant. A parental polypeptide can be a naturally occurring polypeptide, or a variant or engineered form of a naturally occurring polypeptide. Therefore, as used herein, “parental immunoglobulin” means an unmodified immunoglobulin polypeptide modified to generate a variant, and as used herein, “parental antibody” means an unmodified antibody modified to generate a variant antibody. It should be noted that “parental antibody” includes known commercially recombinant antibodies as outlined below. In this context, “parental Fc domain” will be relative to the listed variants; thus, “variant human IgG1 Fc domain” is compared to the parental Fc domain of human IgG1, “variant human IgG4 Fc domain” is compared to the parental Fc domain of human IgG4, and so on.

[0151] As used herein, “position” refers to a location within a protein sequence. Positions may be numbered sequentially or according to established formats, such as the EU index used for numbering antibody domains (e.g., CH1, CH2, CH3, or hinge domains).

[0152] As used in this article, “target antigen” means a molecule that is specifically bound by an antigen-binding domain containing a variable region of a given antibody.

[0153] In this document, in the context of the monomers of the heterodimer antibodies of the present invention, "strandedness" refers to two "matching" strands similar to DNA, incorporating heterodimerizing variants into each monomer to maintain the ability to "match" to form heterodimers. For example, if some pI variants are engineered to monomer A (e.g., to make the pI higher), spatial 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 maintain both functionalities. Similarly, for "skewed" variants that appear in pairs as outlined more fully below, those skilled in the art will consider the pI to determine which strand or monomer one of the pairs will enter, such that the pI of the skewed variant is also used to maximize pI separation.

[0154] As used in this article, "target cell" refers to a cell that expresses the target antigen.

[0155] In this document, in the context of generating bispecific antibodies according to the present invention, "host cell" means a cell containing exogenous nucleic acid encoding components of the bispecific antibody and capable of expressing the bispecific antibody under suitable conditions. Suitable host cells are discussed below.

[0156] In this article, "wild-type" or "WT" refers to an amino acid or nucleotide sequence found in nature, including allelic variations. WT proteins have an unmodified amino acid or nucleotide sequence.

[0157] This article provides several antibody domains (e.g., Fc domains) that have sequence identity with human antibody domains. Sequence identity between two similar sequences (e.g., antibody variable domains) can be measured by algorithms, such as those in the following literature: 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 [Homology alignment algorithm]; Pearson, WR and Lipman, DJ (1988) “Improved Tools For Biological Sequence Comparison,” Proc. Natl. Acad. Sci. (USA) 85:2444 [Similarity search method]; or Altschul, SF et al., (1990) “Basic Local Alignment Search Tool,” J. Mol. Biol. 215:403-10, namely the "BLAST" algorithm, see https: / / blast.ncbi.nlm.nih.gov / Blast.cgi. When using any of the aforementioned algorithms, use the default parameters (regarding window length, space penalty, etc.). In one implementation, the BLAST algorithm is used, with the default parameters used to perform sequence identity verification. The antibodies of this invention are typically isolated or recombinant. In describing the various polypeptides disclosed herein, “isolated” means a polypeptide that has been identified and isolated and / or recovered from cells or cell cultures expressing the polypeptide. Typically, isolated polypeptides are prepared by at least one purification step. “Isolated antibody” refers to an antibody that substantially does not contain other antibodies with different antigen specificities. “Recombinant” means antibodies generated in exogenous host cells using recombinant nucleic acid technology, and that they may also be isolated.

[0158] "Specific binding" refers to the binding of a specific antigen or epitope to which it "specifically binds" or is "specifically" associated with, meaning a binding that is significantly 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 with a similar structure that does not have binding activity. For example, specific binding can be determined by competition with a control molecule that is similar to the target.

[0159] Specific binding to a particular antigen or epitope can be demonstrated, for example, by antibodies whose KD to the antigen or epitope is 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 dissociation rate of a specific antibody-antigen interaction. Typically, the KD of an antibody that specifically binds to an antigen against a control molecule will be 20, 50, 100, 500, 1000, 5,000, 10,000 or more times greater than that of the antigen or epitope.

[0160] Additionally, specific binding to a particular antigen or epitope can be demonstrated, for example, by antibodies whose KA or Ka value for the antigen or epitope is at least 20, 50, 100, 500, 1000, 5000, 10000, or more times greater than that for the epitope relative to a control, where KA or Ka refers to the association rate of the specific antibody-antigen interaction. Binding affinity is typically measured using Biacore, SPR, or BLI assays.

[0161] IV. Anti-CD28 x Anti-ENPP3 antibody On the one hand, this article provides novel anti-CD28 x anti-ENPP3 antibodies. In some embodiments, the anti-CD28 x anti-ENPP3 antibodies described herein are capable of binding agonistically to CD28 co-stimulatory molecules on T cells and ENPP3 on tumor cells. Such antibodies selectively enhance antitumor activity at ENPP3-associated tumor sites while minimizing peripheral toxicity. The subject antibody provided herein is particularly suitable for use in combination with other anticancer therapies, including, for example, bispecific antibodies for the treatment of ENPP3-associated cancers.

[0162] The anti-CD28 x anti-ENPP3 antibody is multivalent and includes at least two antigen-binding domains (ABDs), wherein at least one antigen-binding domain is a CD28-binding domain and at least one antigen-binding domain is an ENPP3-binding domain. The subject anti-CD28 x anti-ENPP3 antibody may include any suitable CD28-binding domain and ENPP3-binding domain, including, for example, the CD28-binding domain and ENPP3-binding domain provided herein.

[0163] The antigen-binding domains described herein typically include a variable heavy chain domain (VH) having VH-CDR1, VH-CDR-2, and VH-CDR-3; and a variable light chain domain (VL) having VL-CDR1, VL-CDR-2, and VL-CDR-3.

[0164] Additionally, as discussed above, the number used to identify CDRs in the sequence listing and figures is Kabat; however, different numbers may be used, which would alter the amino acid sequence of the CDRs as shown in Table 2.

[0165] Other variants can be prepared for all the variable heavy and light chain domains listed herein. As outlined herein, in some embodiments, a set of six CDRs may have 0, 1, 2, 3, 4, or 5 amino acid modifications (ammonia substitution being particularly useful), and variations in the frame regions of the variable heavy and light chain domains, provided that the frame (other than the CDR) maintains at least about 80, 85, 90, 95, or 99% identity with the phylogenetic sequences selected from those listed in Figure 1 of U.S. Patent No. 7,657,380, the figures and illustrations of which are incorporated herein by reference in their entirety. Thus, for example, the same CDR as described herein may be combined with different frame sequences from phylogenetic sequences, provided that the frame regions maintain at least 80, 85, or 90% identity with the phylogenetic sequences selected from those listed in Figure 1 of U.S. Patent No. 7,657,380. Alternatively, the CDR may have amino acid modifications (e.g., 1, 2, 3, 4, or 5 amino acid modifications in a set of CDRs; that is, the CDR may be modified as long as the total number of variations in a set of 6 CDRs is less than 6 amino acid modifications, wherein any combination of CDRs is a variation; for example, there may be one variation in vlCDR1, two variations in vhCDR2, no variation in vhCDR3, etc.)), and frame region variations, provided that the frame region maintains at least 80, 85, 95 to 99% identity with the phylogenetic sequences selected from those listed in Figure 1 of U.S. Patent No. 7,657,380.

[0166] Those skilled in the art will understand that any set of six CDR or VH and VL domains can be in the form of scFv or Fab, which are then added to the heavy chain constant domain and the light chain constant domain, wherein the heavy chain constant domain contains variants (including those within the CH1 domain and the Fc domain).

[0167] Additionally, in embodiments where the subject antibody includes scFv, the scFv may be oriented from the N-terminus to the C-terminus as a VH-scFv linker-VL or a VL-scFv linker-VH. In some forms, one or more ABDs are typically Fabs, which include a VH domain (typically as a component of the heavy chain) on one protein chain and a VL domain (typically as a component of the light chain) on another protein chain. Figure 6 An exemplary scFv connector for the subject antibody is depicted.

[0168] In some embodiments, the anti-CD28 x anti-ENPP3 antibody is a bispecific antibody. In some embodiments, the anti-CD28 x anti-ENPP3 antibody is a bivalent antibody. In some embodiments, the anti-CD28 x anti-ENPP3 antibody is a trivalent antibody. In some embodiments, the anti-CD28 x anti-ENPP3 antibody is a bispecific bivalent antibody. In some embodiments, the anti-CD28 x anti-ENPP3 antibody includes one CD28-binding domain and one ENPP3-binding domain. In an exemplary embodiment, the anti-CD28 x anti-ENPP3 antibody is a bispecific trivalent antibody. In some embodiments, the anti-CD28 x anti-ENPP3 antibody includes one CD28-binding domain and two ENPP3-binding domains.

[0169] The anti-CD28 x anti-ENPP3 antibody described herein may be in any useful form, including, for example, a canonical immunoglobulin, and the “1 + 1 Fab-scFv-Fc,” “2 + 1 “mAb-scFv,” “2 + 1 Fab2-scFv-Fc,” and “2 + 1 Fab2-Fc x scFv-Fc” forms described herein (Figure 14). Other useful forms include, but are not limited to, the “mAb-Fv,” “central Fv,” “single-arm scFv-mAb,” “scFv-mAb,” “double scFv,” and “trident” forms described herein (see, for example, Figure 14). See also US20180127501A1, which is incorporated herein by reference, particularly the relevant portion relating to antibody forms (see, for example, Figure 2). In some embodiments, the anti-CD28 x anti-ENPP3 antibody is a heterodimeric bispecific antibody comprising a variant Fc domain having any of the heterodimeric skewed variants, pI variants, and / or ablation variants described herein. See, for example… Figure 8 .

[0170] It should be noted that, unless otherwise stated herein, the order of antigens listed by name does not confer a structure; that is, an anti-ENPP3 x anti-CD28 1+1 Fab-scFv-Fc antibody may have an scFv that binds to either ENPP3 or CD28, although in some cases the order specifies the indicated structure.

[0171] The anti-CD28 x anti-ENPP3 antibody described herein also includes different antibody domains. As described herein and as is known in the art, the antibody described herein includes different domains within the heavy and light chains, and these domains may overlap. These domains include, but are not limited to, Fc domains, CH1 domains, CH2 domains, CH3 domains, hinge domains, heavy chain constant domains (CH1-hinge-Fc domains or CH1-hinge-CH2-CH3), variable heavy chain domains, variable light chain domains, light chain constant domains, Fab domains, and scFv domains.

[0172] As shown in this paper, there are several suitable linkers (used as domain linkers or scFv linkers) that can be used to covalently link the listed domains (e.g., scFv, Fab, Fc domains, VH domains, VL domains, etc.), including conventional peptide bonds generated by recombination techniques. Figure 7Exemplary linkers are depicted that connect the domains of a subject antibody to each other. In some embodiments, the linker peptide may primarily comprise the following amino acid residues: Gly, Ser, Ala, or Thr. The length of the linker peptide should be sufficient to link two molecules in such a way that they present the correct conformation relative to each other, thereby retaining their desired activity. In one embodiment, the linker is about 1 to 50 amino acid long, preferably about 1 to 30 amino acid long. In one embodiment, a linker of 1 to 20 amino acid long can be used, with about 5 to about 10 amino acids being useful in some embodiments. 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 1 (and typically 3 to 4), glycine-alanine polymers, alanine-serine polymers, and other flexible linkers. Alternatively, a variety of non-protein polymers can be used as linkers, including but not limited to polyethylene glycol (PEG), polypropylene glycol, polyoxyalkylene oxide, or copolymers of polyethylene glycol and polypropylene glycol.

[0173] Other linker sequences can include any sequence of any length from the CL / CH1 domain, but not all residues of the CL / CH1 domain; for example, the first 5-12 amino acid residues of the CL / CH1 domain. Linkers can be derived from immunoglobulin light chains, such as Cκ or Cλ. Linkers can be derived from any isotype of immunoglobulin heavy chain, including, for example, Cγ1, Cγ2, Cγ3, Cγ4, Cα1, Cα2, Cδ, Cε, and Cµ. Linker sequences can also be derived from other proteins, such as Ig-like proteins (e.g., TCR, FcR, KIR), hinge region-derived sequences, and other natural sequences from other proteins.

[0174] In some embodiments, the connector is a “domain connector” for joining any two domains as outlined herein. For example, in the 2+1 Fab2-scFv-Fc form, a domain connector may be present to link the C-terminus of the CH1 domain of the Fab to the N-terminus of the scFv, wherein another optional domain connector links the C-terminus of the scFv to the CH2 domain (though in many embodiments, a hinge is used as this domain connector). While any suitable connector may be used, many embodiments utilize glycine-serine polymers (including, for example, (GS)n, (GSGGS)n, (GGGGS)n, and (GGGS)n, where n is an integer of at least 1 (and typically 3 to 4 to 5)) as domain connectors, and any peptide sequence that allows the two domains to recombine and has sufficient length and flexibility for each domain to retain its biological function. In some cases, and where “chain-like properties” are of interest, charged domain connectors, as outlined below, may be used as in some embodiments of the scFv connector. Exemplary useful domain connectors are depicted in Figure 7 middle.

[0175] In some embodiments, the connector is an scFv connector for covalently connecting VH and VL structural domains as discussed herein. In many cases, the scFv connector is a live scFv connector, wherein multiple are in... Figure 6 As shown in the diagram. Therefore, this document provides charged scFv connectors to facilitate pI separation between the first and second monomers. That is, by incorporating charged scFv connectors with positive or negative charges (or both in the case of scFv scaffolds using different monomers), this allows the monomer containing the charged connector to change the pI without further altering the Fc domain. These charged connectors can be substituted into any scFv containing a standard connector. Again, those skilled in the art will understand that charged scFv connectors are used on the correct “chain” or monomer depending on the desired pI change. For example, as discussed herein, to prepare heterodimeric antibodies in the form of 1+1 Fab-scFv-Fc, the original pI of the Fv region for each desired antigen-binding domain is calculated and selected to prepare the scFv, and a positively or negatively charged connector is selected based on the pI. Charged domain connectors can also be used to increase pI separation of the monomers of the present invention, and therefore, Figure 7 Those included herein can be used in any implementation of the connectors described herein.

[0176] For example, Figure 24 depicts an exemplary subject-specific anti-CD28 x anti-ENPP3 antibody. In the cell culture production of the anti-CD28 x anti-ENPP3 antibody provided herein, C-terminal lysine residues or C-terminal lysine and glycine residues may be cleaved from the heavy chain monomer, resulting in a variant with a C-terminal “prune.” See, for example, Jiang et al., Journal of Pharmaceutical Sciences 105:2066-2072 (2016). Therefore, in some embodiments provided herein, the anti-CD28 x anti-ENPP3 antibody is a variant of the anti-CD28 x anti-ENPP3 antibody comprising the deletion of a C-terminal lysine (-K) terminus or a C-terminal lysine and glycine (-GK) residue in one or both Fc domains of the anti-CD28 x anti-ENPP3 antibody described herein. In some embodiments, the deletion is G446del and / or K447del (EU number).

[0177] The anti-CD28 x anti-ENPP3 antibody is described in more detail below.

[0178] A. CD28 binding domain The anti-CD28 x anti-ENPP3 antibody provided herein includes at least one CD28 binding domain. Any suitable CD28 binding domain may be included in the anti-CD28 x anti-ENPP3 antibody provided herein. In an exemplary embodiment, the CD28 binding domain is an agonistic CD28 ABD, which advantageously provides T cell co-stimulatory activity.

[0179] Those skilled in the art will understand that a suitable CD28 binding domain may comprise a set of six CDRs as depicted in the figures (they are underlined or, in the case of using different numbering schemes, as described herein and shown in Table 2), i.e., in Figure 15 The CD28 binding domain sequences depicted in Figures 18 and 21 are used with additional alignments to identify the CDRs within the variable heavy chain (VH) and variable light chain (VL) domain sequences. Figures 16 and 17 depict additional VH and VL sequences of exemplary CD28 binding domains that can be used in the subject antibody. Suitable CD28 ABDs may also include the complete VH and VL sequences as depicted in these sequences and figures, used as scFv or Fab.

[0180] In one embodiment, the CD28 antigen-binding domain includes any CD28-binding domain described herein (including, but not limited to, those described herein). Figure 15The six CDRs (i.e., vhCDR1-3 and vlCDR1-3) depicted in Figures 18 and 21. In some embodiments, the CD28 ABD that binds to human CD28 is one of the following CD28 ABDs: 1A7[CD28]_H1L1, 1A7[CD28]_H1.1_L1, 1A7[CD28]_H1_L1.71, 1A7[CD28]_H1.1_L1.71, 1A7[CD28]_H1.14_L1, 1A7[CD28]_H1.14_L1.71, CD28.3[CD28]_H0L0, hCD28.3[CD28]_H1L1, 5.11A1[CD28]_H0L0, TGN1412_H1L1, 3 41VL34[CD28]_H1L1, 341VL36[CD28]_H1L1, 281VL4[CD28]_H1L1, HuTN228[CD28]_H1L1, PV1[CD28]_H0L0, m9.3[CD28]_H0L0, hu9.3[CD28]_H1L1, 9G2[CD28]_H0L0, 9G2[CD28]_H1L1, 2F10A3.140[CD28]_H1L1 and TN228[CD28]_H4L2 ( Figure 15 (Figures 18 and 21). In some embodiments, the CD28 ABD includes components selected from... Figure 15 The VH / VL pairs of VH and VL are depicted in Figures 16 and 17.

[0181] In addition to the parental CDR groups forming CD28 ABD disclosed in the accompanying drawings and sequence listings, this document also provides variants of CD28 ABD having CDRs with at least one modification including the CD28 ABD CDRs disclosed herein (e.g., ( Figure 15(See Figures 18 and 21, and the sequence listing). In one embodiment, the CD28 ABD of the subject anti-CD28 x anti-ENPP3 antibody comprises a set of 6 CDRs having amino acid modifications of 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 compared to the 6 CDRs of the CD28 ABD as described herein (including the figures and sequence listing). In an exemplary embodiment, the CD28 ABD of the subject anti-CD28 x anti-ENPP3 antibody comprises a set of 6 CDRs having amino acid modifications of 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 compared to the 6 CDRs of the CD28 ABD as described herein (including the figures and sequence listing). One of the six CDRs of ABD has modifications of 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 amino acids: 1A7[CD28]_H1L1, 1A7[CD28]_H1.1_L1, 1A7[CD28]_H1_L1.71, 1A7[CD28]_H1.1_L1.71, 1A7[CD28]_H1.14_L1, 1A7[CD28]_H1.14_L1.71, CD28.3[CD28]_H0L0, hCD28.3[CD28]_H1L1, 5.11A1[CD28]_H 0L0, TGN1412_H1L1, 341VL34[CD28]_H1L1, 341VL36[CD28]_H1L1, 281VL4[CD28]_H1L1, HuTN228[CD28]_H1L1, PV1[CD28]_H0L0, m9.3[CD28]_H0L0, hu9.3[CD28]_H1L1, 9G2[CD28]_H0L0, 9G2[CD28]_H1L1, 2F10A3.140[CD28]_H1L1 and TN228[CD28]_H4L2 ( Figure 15 (Figures 18 and 21). In some embodiments, the CD28 ABD of the subject anti-CD28 x anti-ENPP3 antibody is capable of binding the CD28 antigen, as measured by at least one of the following: Biacore, surface plasmon resonance (SPR), flow cytometry, and / or BLI (biolayer interferometry, such as Octet assay), the latter being particularly useful in many embodiments. In a particular embodiment, the CD28 ABD is capable of binding the human CD28 antigen (see Figure 1).

[0182] In some embodiments, the CD28 ABD of the subject anti-CD28 x anti-ENPP3 antibody comprises six CDRs that are at least 90, 95, 97, 98, or 99% identical to the six CDRs of the CD28 ABD as described herein (including the figures and sequence listing). In an exemplary embodiment, the CD28 ABD of the subject anti-CD28 x anti-ENPP3 antibody comprises six CDRs that are at least 90, 95, 97, 98, or 99% identical to the six CDRs of the CD28 ABD as described herein (including the figures and sequence listing). The ABD comprises six CDRs, which are at least 90%, 95%, 97%, 98%, or 99% identical to the six CDRs of one of the following CD28 ABDs: 1A7[CD28]_H1L1, 1A7[CD28]_H1.1_L1, 1A7[CD28]_H1_L1.71, 1A7[CD28]_H1.1_L1.71, 1A7[CD28]_H1.14_L1, 1A7[CD28]_H1.14_L1.71, CD28.3[CD28]_H0L0, hCD28.3[CD28]_H1L1, 5.11A1[C D28]_H0L0, TGN1412_H1L1, 341VL34[CD28]_H1L1, 341VL36[CD28]_H1L1, 281VL4[CD28]_H1L1, HuTN228[CD28]_H1L1, PV1[CD28]_H0L0, m9.3[CD28]_H0L0, hu9.3[CD28]_H1L1, 9G2[CD28]_H0L0, 9G2[CD28]_H1L1, 2F10A3.140[CD28]_H1L1 and TN228[CD28]_H4L2 ( Figure 15 (Figures 18 and 21). In some embodiments, the CD28 ABD is capable of binding to CD28, as measured by at least one of the following: Biacore, surface plasmon resonance (SPR), flow cytometry, and / or BLI (biolayer interferometry, such as Octet assay), the latter being particularly useful in many embodiments. In a particular embodiment, the CD28 ABD is capable of binding to the human CD28 antigen (see Figure 1).

[0183] In another exemplary embodiment, the CD28 ABD of the subject anti-CD28 x anti-ENPP3 antibody includes the variable heavy chain (VH) domain and variable light chain (VL) domain of any of the CD28 ABDs described herein (including the figures and sequence listing). In an exemplary embodiment, the CD28 ABD is the following CD28 One of ABD: 1A7[CD28]_H1L1, 1A7[CD28]_H1.1_L1, 1A7[CD28]_H1_L1.71, 1A7[CD28]_H1.1_L1.71, 1A7[CD28]_H1.14 _L1, 1A7[CD28]_H1.14_L1.71, CD28.3[CD28]_H0L0, hCD28.3[CD28]_H1L1, 5.11A1[CD28]_H0L0, TGN1412_H1L1, 3 41VL34[CD28]_H1L1, 341VL36[CD28]_H1L1, 281VL4[CD28]_H1L1, HuTN228[CD28]_H1L1, PV1[CD28]_H0L0, m9.3[CD28]_H0L0, hu9.3[CD28]_H1L1, 9G2[CD28]_H0L0, 9G2[CD28]_H1L1, 2F10A3.140[CD28]_H1L1 and TN228[CD28]_H4L2 ( Figure 15 (Figures 18 and 21). In some embodiments, the CD28 ABD includes components selected from... Figure 15 The VH and VL pairs depicted in Figures 16 and 17.

[0184] In some embodiments, the anti-CD28 x anti-ENPP3 antibody comprises CD28 ABD, which includes a variable heavy chain domain and / or a variable light chain domain, said domain being a variant of the CD28 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 variations compared to the VH and / or VL domains of the CD28 ABD described herein (including the figures and sequence listing). In an exemplary embodiment, the variant VH and / or VL domains are associated with the following CD28 ABD... Compared to the VH and / or VL domains of one of the ABDs, there are 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid variations: 1A7[CD28]_H1L1, 1A7[CD28]_H1.1_L1, 1A7[CD28]_H1_L1.71, 1A7[CD28]_H1.11_L1.71, 1A7[CD28]_H1.14_L1, 1A7[CD28]_H1.14_L1.71, CD28.3[CD28]_H0L0, hCD28.3[CD28]_H1L1, 5.11A1[CD28] _H0L0, TGN1412_H1L1, 341VL34[CD28]_H1L1, 341VL36[CD28]_H1L1, 281VL4[CD28]_H1L1, HuTN228[CD28]_H1L1, PV1[CD28]_H0L0, m9.3[CD28]_H0L0, hu9.3[CD28]_H1L1, 9G2[CD28]_H0L0, 9G2[CD28]_H1L1, 2F10A3.140[CD28]_H1L1 and TN228[CD28]_H4L2 ( Figure 15 (Figures 18 and 21). In some embodiments, the changes are in Figure 15 -18 and the VH structural domain depicted in Figure 21. In some embodiments, the change is located within... Figure 15 -18 and the VL structural domain depicted in Figure 21. In some embodiments, the variation is located within... Figure 15-18 and the VH and VL domains depicted in Figure 21. In some embodiments, the one or more amino acid changes are located in the VH and / or VL framework regions (FR1, FR2, FR3 and / or FR4). In some embodiments, the one or more amino acid changes are located in one or more CDRs. In some embodiments, the CD28 ABD is capable of binding to CD28, as measured by at least one of the following: Biacore, surface plasmon resonance (SPR), flow cytometry and / or BLI (biolayer interferometry, such as Octet assay), the latter being particularly useful in many embodiments. In a particular embodiment, the CD28 ABD is capable of binding to the human CD28 antigen (see Figure 1).

[0185] 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 CD28 ABD as described herein (including the figures and sequence listing). In an exemplary embodiment, the variant VH and / or VL domains are identical to the following CD28... One of the ABDs has VH and / or VL that are at least 90%, 95%, 97%, 98%, or 99% identical: 1A7[CD28]_H1L1, 1A7[CD28]_H1.1_L1, 1A7[CD28]_H1_L1.71, 1A7[CD28]_H1.1_L1.71, 1A7[CD28]_H1.14_L1, 1A7[CD28]_H1.14_L1.71, CD28.3[CD28]_H0L0, hCD28.3[CD28]_H1L1, 5.11A1[CD28]_H0L0, T GN1412_H1L1, 341VL34[CD28]_H1L1, 341VL36[CD28]_H1L1, 281VL4[CD28]_H1L1, HuTN228[CD28]_H1L1, PV1[CD28]_H0L0, m9.3[CD28]_H0L0, hu9.3[CD28]_H1L1, 9G2[CD28]_H0L0, 9G2[CD28]_H1L1, 2F10A3.140[CD28]_H1L1 and TN228[CD28]_H4L2 ( Figure 15 (Figures 18 and 21). In some embodiments, the CD28 ABD includes... Figure 15 -18 and the VH structural domains depicted in Figure 21 are at least 90%, 95%, 97%, 98%, or 99% identical to each other. In some embodiments, the CD28 ABD includes VHs that are identical to those depicted in Figure 21. Figure 15 -18 and the VL structural domains depicted in Figure 21 are at least 90%, 95%, 97%, 98%, or 99% identical to each other. In some embodiments, the CD28 ABD includes VLs that are identical to those depicted in Figure 21. Figure 15 The VH and VL domains depicted in Figures -18 and 21 are at least 90%, 95%, 97%, 98%, or 99% identical. In some embodiments, the CD28 ABD is capable of binding to CD28, as measured by at least one of the following: Biacore, surface plasmon resonance (SPR), flow cytometry, and / or BLI (biolayer interferometry, such as Octet assay), the latter being particularly useful in many embodiments. In a particular embodiment, the CD28 ABD is capable of binding to the human CD28 antigen (see Figure 1).

[0186] In some embodiments, the CD28-binding domain of the subject anti-CD28 x anti-ENPP3 antibody includes a VH, said VH comprising Figure 19A Either of the depicted VHCDR1-3 or HFR1-4 sequences. In some embodiments, the CD28 binding domain includes a VL, which includes... Figure 19B Either of the described VLCDR1-3 or LFR1-4 sequences.

[0187] In some embodiments, the anti-CD28 x anti-ENPP3 antibody includes a CD28-binding domain, the domain comprising VH and VL selected from: (i) VH, which includes vhCDR1, vhCDR2 and vhCDR3, each having Figure 15 (ii) the amino acid sequences of vhCDR1, vhCDR2, and vhCDR3 of any CD28-binding domain VH or variant thereof depicted in Figures 18 and 21; and (ii) VL, wherein the VL comprises vlCDR1, vlCDR2, and vlCDR3, which respectively have Figure 15 The amino acid sequences of vlCDR1, vlCDR2, and vlCDR3 of the CD28 binding domain VL or its variants depicted in Figures 18 and 21; or (i) VH, which includes vhCDR1, vhCDR2 and vhCDR3, each having Figure 15 Or the amino acid sequences of vhCDR1, vhCDR2, and vhCDR3 of VH or its variants depicted in Figure 16; and (ii) VL, which comprises vlCDR1, vlCDR2, and vlCDR3, each having Figure 15 Or the amino acid sequences of vlCDR1, vlCDR2 and vlCDR3 of VL or its variants depicted in Figure 17.

[0188] In some embodiments, the anti-CD28 x anti-ENPP3 antibody includes a CD28-binding domain, the domain comprising VH and VL selected from: (i) VH, wherein the VH has Figure 15 (ii) the amino acid sequence of VH or a variant thereof of any CD28 binding domain depicted in Figures 18 and 21; and (ii) VL, wherein VL has Figure 15 The amino acid sequences of VL or variants of the CD28 binding domain depicted in Figures 18 and 21; or (i) VH, wherein the VH has Figure 15 Or the amino acid sequence of VH or its variants depicted in Figure 16; and (ii) VL, which has Figure 15 Or the amino acid sequence of VL or its variants as depicted in Figure 17.

[0189] B. ENPP3 binding domain The anti-CD28 x anti-ENPP3 antibody presented herein includes at least one ENPP3 binding domain. Subject antibodies including such ENPP3 antigen-binding domains (e.g., anti-ENPP3 x anti-CD3 bispecific antibodies) advantageously target cells expressing high levels of ENPP3, rather than those expressing multiple levels of ENPP3 (e.g., normal cells).

[0190] In some embodiments, the ENPP3 ABD of the anti-CD28 x anti-ENPP3 antibody includes a set of six CDRs as depicted in the sequence listing and Figure 22 (the CDRs are underlined or, in the case of using different numbering schemes, as described herein and shown in Table 2), i.e., other CDRs identified by alignment within the variable heavy chain (VH) domain and variable light chain (VL) domain sequences depicted in Figure 22 and the sequence listing (see Table 2).

[0191] In an exemplary embodiment, the ENPP3 ABD of the anti-CD28 x anti-ENPP3 antibody includes the variable heavy chain (VH) domain and the variable light chain (VL) domain of any of the ENPP3 ABDs described herein (including the figures and sequence listing). In an exemplary embodiment, the ENPP3 ABD of the anti-CD28 x anti-ENPP3 antibody is the ENPP3 ABD depicted in Figure 22.

[0192] In addition to the parental ENPP3 variable heavy chain domain and variable light chain domain disclosed herein, this document also provides ENPP3 ABD, which includes a variable heavy chain domain and / or a variable light chain domain, said domains being 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 variations compared to the VH and / or VL domains of the ENPP3 ABD described herein (including the figures and sequence listing). In an exemplary embodiment, the variant VH and / or VL domains have 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid variations compared to the VH and / or VL domains of the ENPP3 ABD depicted in Figure 22. In some embodiments, said variations are located in the VH domain depicted in Figure 22. In some embodiments, said variations are located in the VL domain depicted in Figure 22. In some embodiments, the changes are located in the VH and VL domains depicted in Figure 22. In some embodiments, the one or more amino acid changes are located in the VH and / or VL framework regions (FR1, FR2, FR3, and / or FR4). In some embodiments, the one or more amino acid changes are located in one or more of vhCDR1-3 and / or vlCDR1-3. In some embodiments, the ENPP3 variant includes vhCDR1-3 and / or vlCDR1-3 of any ENPP3ABD in Figure 22. In some embodiments, the ENPP3ABD of the anti-CD28 x anti-ENPP3 antibody is capable of binding to ENPP3, as measured by at least one of the following: Biacore, surface plasmon resonance (SPR), flow cytometry, and / or BLI (biolayer interferometry, such as Octet assay), the latter being particularly useful in many embodiments. In a particular embodiment, the ENPP3 ABD is capable of binding to the human ENPP3 antigen (Figure 2).

[0193] 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 ENPP3 ABD as described herein (including the figures and sequence listing). In an exemplary 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 ENPP3 ABD depicted in FIG. 22. In some embodiments, the ENPP3 ABD includes a VH that is at least 90%, 95%, 97%, 98%, or 99% identical to the VH domain depicted in FIG. 22. In some embodiments, the ENPP3 ABD includes a VL that is at least 90%, 95%, 97%, 98%, or 99% identical to the VL domain depicted in FIG. 22. In some embodiments, the ENPP3 ABD includes a VH and a VL that are at least 90%, 95%, 97%, 98%, or 99% identical to the VH and VL domains depicted in FIG. 22. In some embodiments, the ENPP3 ABD comprises the six CDRs (vhCDR1-3 and vlCDR1-3) of any ENPP3 ABD in Figure 22. In some embodiments, the ENPP3 ABD of the anti-CD28 x anti-ENPP3 antibody is capable of binding to ENPP3, as measured by at least one of the following: Biacore, surface plasmon resonance (SPR), flow cytometry, and / or BLI (biolayer interferometry, such as Octet assay), the latter being particularly useful in many embodiments. In a particular embodiment, the ENPP3 ABD is capable of binding to the human ENPP3 antigen (Figure 2).

[0194] In some embodiments, the anti-CD28 x anti-ENPP3 antibody is a bivalent antibody comprising one ENPP3-binding domain (e.g., a 1+1 Fab-scFv-Fc form antibody). In other embodiments, the anti-CD28 x anti-ENPP3 antibody is a trivalent antibody comprising two ENPP3-binding domains (e.g., 2+1 mAb-scFv, 2+1 Fab2). - Antibodies in the form of scFv-Fc and 2+1Fab2-Fc x scFv-Fc.

[0195] C. Chimeric antibodies and humanized antibodies In some embodiments, the subject antibodies provided herein comprise heavy chain variable regions from a specific germline heavy chain immunoglobulin gene and / or light chain variable regions from a specific germline light chain immunoglobulin gene. For example, such antibodies may comprise or be composed of human antibodies containing heavy or light chain variable regions that are either “products” of or “derived from” a specific germline sequence. Thus, human antibodies that are “products” of or “derived from” a human germline immunoglobulin sequence can be identified by comparing the amino acid sequence of the human antibody with the amino acid sequence of a human germline immunoglobulin and selecting the human germline immunoglobulin sequence that is sequence-closest (i.e., has the greatest identity%) to the sequence of the human antibody (using the methods outlined herein). Human antibodies that are “products” 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. However, the amino acid sequence of a humanized antibody is generally at least 90% identical to the amino acid sequence encoded by a human germline immunoglobulin gene, and the amino acid residues contained therein identify the antibody as derived from a human sequence when compared with germline immunoglobulin amino acid sequences of other species (e.g., murine germline sequences). In some cases, the amino acid sequence of a humanized antibody may be at least 95%, 96%, 97%, 98%, or 99%, or even at least 96%, 97%, 98%, or 99%, identical to the amino acid sequence encoded by a germline immunoglobulin gene. Typically, a humanized antibody derived from a specific human germline sequence will exhibit no more than 10-20 amino acid differences compared to the amino acid sequence encoded by a human germline immunoglobulin gene (prior to any skewed variants, pI variants, and ablation variants introduced herein; that is, prior to the variants of the present invention, the number of variants is generally low). In some cases, the humanized antibody may exhibit no more than 5 amino acid differences or even no more than 4, 3, 2 or 1 amino acid difference compared to the amino acid sequence encoded by germline immunoglobulin genes (again, prior to any skewed variants, pI variants and ablation variants introduced herein; that is, prior to the introduction of the variants of the present invention, the number of variants was generally low).

[0196] 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 mature the variable region of antibodies. These methods include, but are not limited to, those described in the following literature: 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, Proc. Natl. Acad. Sci. USA 95: 8910-8915; Krauss et al., 2003, Protein Engineering 16(10):753-759, all of which are incorporated herein by reference in their entirety. Other humanization methods may involve porting only a portion of the CDR, including but not limited to the methods described in the following literature: 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.

[0197] D. Heterodimer antibody In an exemplary embodiment, the anti-CD28 x anti-ENPP3 antibody provided herein is a heterodimer bispecific antibody comprising two variant Fc domain sequences. Such variant Fc domains include amino acid modifications to facilitate the self-assembly and / or purification of the heterodimer antibody.

[0198] A persistent problem in antibody technology is the desire for “bispecific” antibodies to bind to two different antigens simultaneously, typically allowing these different antigens to approach each other, thereby generating novel functions and therapies. These antibodies are usually produced by incorporating the genes for each heavy and light chain into the host cell. This typically results in the formation of the desired heterodimer (AB) and two homodimers (AA and BB (excluding the issue of light chain heterodimerization)). However, a major obstacle to forming bispecific antibodies is the difficulty in favoring the formation of the desired heterodimer rather than the homodimer, and / or the difficulty in purifying the heterodimer from the homodimer.

[0199] Multiple mechanisms are available for generating heterodimer antibodies. Furthermore, those skilled in the art will understand that these different mechanisms can be combined to ensure high heterodimerization. Amino acid modifications that promote the generation and purification of heterodimers are generally referred to collectively as “heterodimerization variants.” As discussed below, heterodimerization variants include “skew” variants (e.g., the “mortar and pestle” and “charge pair” variants described below) and “pI variants,” which allow the purification of heterodimers from homodimers. As generally described in U.S. Patent No. 9,605,084 (which is hereby incorporated in its entirety by reference and specifically for the discussion of heterodimerization variants below), useful heterodimerization mechanisms include the “mortar and pestle” (“KIH”) as described in U.S. Patent No. 9,605,084, the “electrostatic orientation” or “charge pair” as described in U.S. Patent No. 9,605,084, the pI variant as described in U.S. Patent No. 9,605,084, and additional Fc variants as generally outlined in U.S. Patent No. 9,605,084 and below.

[0200] Heterodimerization variants that can be used to form and purify thematic heterodimer antibodies (e.g., bispecific antibodies) are discussed in more detail below.

[0201] 1. Skew variant In some embodiments, the heterodimer antibody includes a skewed variant that is a modification of one or more amino acids in a first Fc domain (A) and / or a second Fc domain (B), the modification being conducive to the formation of an Fc heterodimer (including an Fc dimer comprising the first Fc domain and the second Fc domain; (AB)) rather than an Fc homodimer (including an Fc dimer comprising both first Fc domains or both second Fc domains; AA or BB). Suitable skewed variants include those disclosed in U.S. Publication No. 2016 / 0355608 (which is hereby incorporated herein by reference in its entirety and specifically for its disclosure regarding skewed variants). Figure 29 , and Figure 3 and Figure 8 middle.

[0202] A particular type of skewed variant, commonly referred to in the art as a "mortar and pestle," refers to an amino acid engineering modification that produces spatial effects that favor heterodimer formation while discouraging homodimer formation, as described in the following publications: USSN 61 / 596,846; Ridgway et al., Protein Engineering 9(7):617 (1996); Atwell et al., J. Mol. Biol. 1997 270:26; US Patent No. 8,216,805, all of which are hereby incorporated by reference in their entirety and specifically for the disclosure of the "mortar and pestle" mutation. This is sometimes referred to herein as a "spatial variant." The figures identify several "monomer A - monomer B" pairs that depend on the "mortar and pestle." In addition, as described in Merchant et al., Nature Biotech. 16:677 (1998), these "mortar and pestle" mutations can combine with disulfide bonds to further favor the formation of Fc heterodimers.

[0203] Another method that can be used to generate heterodimers is sometimes referred to as “electrostatic redirection,” as described in Gunasekaran et al., J. Biol. Chem. 285(25):19637 (2010), which is hereby incorporated in its entirety by reference. This is sometimes referred to herein as “charge pair.” In this embodiment, electrostatics are used to deflect the formation toward heterodimerization. Those skilled in the art will understand that these variants may also affect the pI and thus the purification, and therefore may be considered pI variants in some cases. However, since these variants are generated to force heterodimerization and are not used as purification tools, they are classified as “deflection variants.” These variants include, but are not limited to, D221E / P228E / L368E paired with D221R / P228R / K409R (e.g., these variants are “monomer counterparts”), and C220E / P228E / 368E paired with C220R / E224R / P228R / K409R.

[0204] In some embodiments, the skewed variants advantageously and simultaneously facilitate heterodimerization based on both "mortar and pestle" and "electrostatic reversal" mechanisms. In some embodiments, the heterodimer antibody comprises one or more such heterodimerizing skewed variants. These variants appear as "paired" "sets." That is, one set of said pairs is incorporated into a first monomer and another set of said pairs is incorporated into a second monomer. It should be noted that these sets do not necessarily exhibit "mortar and pestle" variants, where there is a one-to-one correspondence between residues on one monomer and residues on the other. That is, these paired sets can instead form an interface between the two monomers that promotes heterodimer formation but not homodimer formation, thereby allowing the percentage of spontaneously formed heterodimers under biological conditions to exceed 90%, rather than the expected 50% (25% homodimer A / A: 50% heterodimer A / B: 25% homodimer B / B). Figure 3 and Figure 8 Exemplary heterodimerization “tilted” variants are depicted. Such “tilted” variants include, but are not limited to: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364K; T411T / E360E / Q362E: D401K; L368D / K370S: S364K / E357L; K370S: S364K / E357Q (EU designations).

[0205] In exemplary embodiments, the heterodimer antibody includes 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 bridging disulfide bonds, T366S / L368A / Y407V / Y349C: T366W / S354C) “skewed” variant amino acid substitution groups (EU numbers). In one exemplary embodiment, the heterodimeric antibody comprises an amino acid substitution set of “S364K / E357Q : L368D / K370S”. For nomenclature purposes, “S364K / E357Q : L368D / K370S” means that one monomer comprises an Fc domain containing amino acid substitutions for S364K and E357Q, and another monomer comprises an Fc domain containing amino acid substitutions for L368D and K370S; as above, the “chain-like properties” of these pairs depend on the starting pI.

[0206] In some embodiments, the skew variants provided herein may optionally and independently be modified with any other modifications (including, but not limited to, other skew variants, see, for example, U.S. Publication Application No. 2012 / 0149876). Figure 37 (The following, by reference, are incorporated herein by reference, particularly their disclosures regarding skewed variants, pI variants, isotype variants, FcRn variants, ablation variants, etc.) are incorporated together with one or both of the first and second Fc domains of the heterodimer antibody. Furthermore, individual modifications may be independently and optionally included in or excluded from the subject heterodimer antibody.

[0207] In some implementations, the skewed variants outlined herein may optionally and independently be incorporated into one or two heavy chain monomers along with any pI variant (or other variants, such as Fc variants, FcRn variants, etc.), and may independently and optionally be included in or excluded from the subject heterodimer antibody.

[0208] 2. Purified variants In some embodiments, the heterodimeric antibody includes a purified variant that advantageously allows the separation of the heterodimeric protein (e.g., an anti-CD28 x anti-ENPP3 bispecific antibody) from the homodimeric protein.

[0209] Several basic mechanisms facilitate the easy purification of heterodimer antibodies. For example, modification of one or both of the anti-weight chain monomers A and B, resulting in each monomer having a different pI, allows for isoelectric purification of heterodimer AB antibodies from monomeric AA and BB proteins. Alternatively, some scaffold forms allow for size-based separation, such as the "1 + 1 Fab-scFv-Fc" and "2 + 1 Fab2-scFv-Fc" forms. As mentioned above, it is also possible to use skewed variants to "skew" the formation of heterodimers relative to homodimers. Therefore, combinations of heterodimerizing skewed variants and purification variants are particularly useful in the heterodimer antibodies presented herein.

[0210] Additionally, as outlined more fully below, depending on the form of the heterodimeric antibody, purified variants contained within the constant region and / or Fc domain of the monomer, and / or domain linkers, may be used. In some embodiments, the heterodimeric antibody includes additional modifications for alternative functions, which may also produce pI changes, such as Fc, FcRn, and KO variants.

[0211] In some embodiments, the heterodimeric antibodies of the subject matter provided herein comprise at least one monomer having one or more modifications that alter the pI of the monomer (i.e., "pI variants"). Generally, those skilled in the art will understand that pI variants exist in two general categories: those that increase the pI of the protein (basic changes) and those that decrease the pI of the protein (acidic changes). As described herein, all combinations of these variants are possible: one monomer may be wild-type, or a variant that does not exhibit a pI significantly different from the wild-type, and the other monomer may be more basic or more acidic. Alternatively, each monomer may be modified, with one monomer becoming more basic and the other more acidic.

[0212] Depending on the form of the heterodimeric antibody, the pI variant may be contained within a constant domain and / or Fc domain of the monomer, or a charged linker (domain linker or scFv linker) may be used. That is, antibody forms utilizing scFv (such as the "1+1 Fab-scFv-Fc" form) may include a charged scFv linker (positively or negatively charged), which provides a further pI enhancement for purification purposes. Those skilled in the art will understand that some 1+1 Fab-scFv-Fc and 2+1 Fab2-scFv-Fc forms are useful only when used with a charged scFv linker without additional pI modulation; however, the present invention also provides pI variants and / or charged domain linkers on one or both monomers. Additionally, further amino acid engineering for alternative functions may also confer pI variations, such as Fc, FcRn, and KO variants.

[0213] In heterodimeric antibodies that use pI as a separation mechanism to allow purification of heterodimeric proteins, amino acid variants are introduced into one or both monomeric peptides. That is, the pI of one monomer (referred to herein as "monomer A" for simplicity) can be engineered to be far removed from monomer B, or both monomers A and B can be altered, with the pI of monomer A increasing and the pI of monomer B decreasing. As outlined more fully below, changes in the pI of either monomer can be made by: removing or adding charged residues (e.g., replacing neutral amino acids with positively or negatively charged amino acid residues, such as glycine becoming glutamic acid), changing charged residues from positively or negatively charged to oppositely charged (aspartic acid becoming lysine), or changing charged residues to neutral residues (e.g., losing charge; lysine becoming serine). Figure 3 and Figure 4 Several of these variations are shown.

[0214] Therefore, in some embodiments, the heterodimeric antibody includes an amino acid modification in the constant region, which alters the isoelectric point (pI) of at least one (if not two) monomers of the dimeric protein to form a "pI antibody" by incorporating an amino acid substitution ("pI variant" or "pI substitution") into one or two monomers. As shown herein, the separation of the heterodimer from two homodimers can be achieved if the pI difference between the two monomers is as small as 0.1 pH units, wherein 0.2, 0.3, 0.4, and 0.5 or higher are all possible values ​​for this invention.

[0215] Those skilled in the art will understand that, to achieve good separation, the number of pI variants intended to be included on each or both monomers will depend in part on the initial pI of the component, for example, in the 1+1 Fab-scFv-Fc, 2+1 Fab2-scFv-Fc, 1+1 CLC, and 2+1 CLC forms, depending on the scFv of interest (1+1 Fab-scFv-Fc, 2+1 Fab2-scFv-Fc) and the initial pI of the Fab. 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 the decision is made accordingly. As is known in the art, different Fvs will have different initial pIs, which are explored in this invention. Generally, as outlined herein, pIs are engineered such that the total pI difference for each monomer reaches at least about 0.1 log, preferably 0.2 to 0.5, as outlined herein.

[0216] In the use of pI variants to achieve heterodimerization, a more modular approach to designing and purifying bispecific proteins (including antibodies) is provided by utilizing the constant region of the heavy chain. Therefore, in some embodiments, heterodimerizing variants (including skewed and pI heterodimerizing variants) are not included in the variable region, necessitating 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 issue to be addressed is elucidating low pI constant domains with high human sequence content, e.g., minimizing or avoiding non-human residues at any specific position. Alternatively, in addition to isotype substitution, the likelihood of pI variants causing immunogenicity is significantly reduced by utilizing isosteric substitutions (e.g., Asn to Asp; and Gln to Glu).

[0217] As discussed below, the potential collateral benefits of this pI engineering include prolonged serum half-life and increased FcRn binding. In other words, as described in US Publication No. US 2012 / 0028304 (incorporated in its entirety by reference), reducing the pI of antibody constant domains (including those found in antibody and Fc fusions) can result in longer serum retention in vivo. These pI variants, which increase serum half-life, also facilitate pI changes for purification.

[0218] Additionally, it should be noted that the pI variants offer further benefits for the analysis and quality control processes of bispecific antibodies, as the ability to eliminate, minimize, and distinguish homodimers in their presence is important. Similarly, the ability to reliably test the reproducibility of heterodimeric antibody production is also crucial.

[0219] Generally, implementations for special purposes rely on a group of variants, including skewed variants that promote heterodimerization rather than homodimerization, plus pI variants that increase the pI difference between the two monomers to facilitate the purification of heterodimers from homodimers.

[0220] Exemplary combinations of pI variants are shown in Figure 3 and Figure 4 And the US publication of application number 2016 / 0355608 Figure 30 As shown in Figure 3, all references are incorporated herein by reference in their entirety, specifically relating to public information concerning pI variants. Figure 4 Preferred combinations of pI variants are shown. As outlined herein and illustrated in the figures, these variations are shown relative to IgG1, but all isotypes and isotype heterozygotes can be modified in this manner. R133E and R133Q may also be used in cases where the heavy chain constant domain is derived from IgG2-4.

[0221] In one embodiment, a preferred combination of pI variants comprises a monomer (negatively charged Fab side) containing the 208D / 295E / 384D / 418E / 421D variant (N208D / Q295E / N384D / Q418E / N421D relative to human IgG1) and a second monomer (positively charged scFv side) containing a positively charged scFv linker (including (GKPGS)4 (SEQ ID NO:XXX)). However, those skilled in the art will understand that the first monomer includes a CH1 domain, including position 208. Therefore, in constructs that do not include a CH1 domain (e.g., for antibodies that do not utilize a CH1 domain on one of their domains), the preferred negatively charged pI variant Fc group comprises the 295E / 384D / 418E / 421D variant (Q295E / N384D / Q418E / N421D relative to human IgG1).

[0222] Therefore, in some implementations, a monomer has Figure 8 One set of replacements and another monomer having an electrical connector (in the form of an electrical scFv connector, since the monomer contains scFv), or in the form of an electrical structural domain connector as indicated, the electrical structural domain connector being selectable from Figure 6 Those that are depicted.

[0223] In some implementations, modifications are made to the hinges 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. Therefore, pI mutations, and particularly substitutions, can be made in one or more of positions 216-230, where 1, 2, 3, 4, or 5 mutations are useful. Similarly, all possible combinations are considered (alone, or together with other pI variants in other domains).

[0224] Specific substitutions that can be used to reduce pI in the hinge domain include, but are not limited to, deletion at position 221, non-natural valine or threonine at position 222, deletion at position 223, non-natural glutamic acid at position 224, deletion at position 225, deletion at position 235, and deletion or non-natural alanine at position 236. In some cases, pI substitution is made only in the hinge domain, while in others, these substitutions are added in any combination to other pI variants in other domains.

[0225] In some implementations, 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 can be made to positions 233-236 to increase effector function in the IgG2 backbone (along with 327A). Similarly, all possible combinations of these 14 positions are possible; for example, the antibodies provided herein may include variant Fc domains with 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 CH2 pI substitutions.

[0226] Specific substitutions that can be used to reduce the pI of the CH2 domain include, but are not limited to, non-natural glutamine or glutamic acid at position 274, non-natural phenylalanine at position 296, non-natural phenylalanine at position 300, non-natural valine at position 309, non-natural glutamic acid at position 320, non-natural glutamic acid at position 322, non-natural glutamic acid at position 326, non-natural glycine at position 327, non-natural glutamic acid at position 334, non-natural threonine at position 339, and all possible combinations within CH2 and with other domains.

[0227] In this embodiment, the modification may be independently and optionally selected from positions 355, 359, 362, 384, 389, 392, 397, 418, 419, 444, and 447 (EU numbers) of the CH3 region. Specific substitutions that can be used to reduce the pI of the CH3 domain include, but are not limited to, non-natural glutamine or glutamic acid at position 355, non-natural serine at position 384, non-natural asparagine or glutamic acid at position 392, non-natural methionine at position 397, non-natural glutamic acid at position 419, non-natural glutamic acid at position 359, non-natural glutamic acid at position 362, non-natural glutamic acid at position 389, non-natural glutamic acid at position 418, non-natural glutamic acid at position 444, and deletion or non-natural aspartic acid at position 447.

[0228] In some embodiments, the anti-CD28 x anti-ENPP3 antibody includes an amino acid substitution in one of its Fc domains that reduces binding to protein A. Such purification variants produce heterodimers with asymmetric binding to protein A, which can then be used to separate heterodimer populations from homodimer populations via a pH gradient. Exemplary purification amino acid substitutions that reduce binding to protein A include, but are not limited to, H435R and Y436F (IgG1 CH3 domain, EU number). See, for example, US2010331527, which is incorporated herein by reference in its entirety and specifically relates to the relevant disclosure concerning Fc domain modifications to reduce protein A binding.

[0229] 3. Allotype variants Furthermore, many embodiments of thematic heterodimeric antibodies rely on “introducing” a pI amino acid from a specific position of one IgG isoform into another IgG isoform, thereby reducing or eliminating the possibility of introducing unwanted immunogenicity into the variant. Figure 21 of U.S. Publication 2014 / 0370013 illustrates several of these variants, which is hereby incorporated by reference. That is, IgG1 is a common isoform for therapeutic antibodies for a variety of reasons, including high effector function. However, the heavy chain constant region of IgG1 has a higher pI (8.10 vs. 7.31) than the heavy chain constant region of IgG2. By introducing IgG2 residues at specific positions into the IgG1 backbone, the resulting monomer exhibits a lower (or higher) pI and additionally a longer serum half-life. For example, IgG1 has glycine at position 137 (pI 5.97), while IgG2 has glutamate (pI 3.22); introducing glutamate will affect the pI of the resulting protein. As described below, multiple amino acid substitutions are typically required to significantly affect the pI of the variant antibody. However, it should be noted that, as discussed below, even changes in the IgG2 molecule can allow for an increase in serum half-life.

[0230] In other embodiments, different types of amino acids are modified to reduce the total charge state of the resulting protein (e.g., by changing amino acids with higher pI to amino acids with lower pI), or to allow structural modifications to achieve stability, as further described below.

[0231] Furthermore, significant changes in the pI of each monomer in the heterodimer can be observed through pI engineering of the constant structural domains of the heavy and light chains. As discussed in this paper, a pI difference of at least 0.5 between the two monomers allows for separation by ion exchange chromatography, isoelectric focusing, or other methods sensitive to the isoelectric point.

[0232] 4. Calculate pI The pI of each monomer of the antibody presented herein may depend on the pI of the variable heavy chain constant domain and the pI of the total monomer, which includes the variable heavy chain constant domain and the fusion cohesive. Therefore, in some embodiments, the pI variation is calculated based on the variable heavy chain constant domain using the graph in Figure 19 of U.S. Publication 2014 / 0370013. As discussed herein, which monomer is engineered is typically determined by the inherent pI of the Fv and the scaffold region. Alternatively, the pI of each monomer can be compared.

[0233] 5. It also endows the pI variant with better in vivo binding of FcRn. In cases where the pI variant reduces the pI of the monomer, the pI variant may have the additional benefit of improved serum retention in vivo.

[0234] Although still under investigation, the Fc region is thought to have a long half-life in vivo because the binding of FcRn in the endosome at pH 6 isolates Fc (Ghetie and Ward, 1997 Immunol Today. 18(12): 592-598, incorporated herein by reference). The endosome compartment then allows Fc to recirculate to the cell surface. Once the compartment opens to the extracellular space, a higher pH (around 7.4) induces the release of Fc back into the bloodstream. In mice, Dall' Acqua et al. showed that Fc mutants with increased 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, incorporated herein by reference). The increased affinity of Fc for FcRn at pH 7.4 is thought to prevent the release of Fc back into the bloodstream. Therefore, Fc mutations that increase the in vivo half-life of Fc would ideally increase FcRn binding at lower pH levels while still allowing Fc release at higher pH levels. Within a pH range of 6.0 to 7.4, the amino acid histidine alters its charge state. Therefore, it is not surprising to find His residues at key positions in the Fc / FcRn complex.

[0235] Recently, it has been proposed that antibodies containing variable regions with lower isoelectric points can also have longer serum half-lives (Igawa et al., 2010 PEDS. 23(5): 385-392, incorporated herein by reference in its entirety). However, the mechanisms underlying this remain poorly understood. Furthermore, the variable regions vary from antibody to antibody. As discussed in this paper, constant region variants with reduced pI and extended half-lives would offer a more modular approach to improving the pharmacokinetic properties of antibodies.

[0236] E. Additional Fc variants for other functions In addition to the heterodimerization variants discussed above, there are a variety of useful Fc amino acid modifications, which can be made for various reasons, including but not limited to altering the binding to one or more FcγR receptors, altering the binding to FcRn receptors, etc., as discussed below.

[0237] Therefore, the antibodies (heterodimeric antibodies and homodimeric antibodies) presented herein may include such amino acid modifications, with or without the heterodimerization variants (e.g., pI variants and spatial variants) outlined herein. Each group of variants may be independently and optionally included in or excluded from any particular heterodimeric protein.

[0238] 1. FcγR and FcRn variants Therefore, various useful Fc substitutions can be performed to alter binding to one or more FcγR receptors. In some embodiments, the subject antibody includes modifications that alter binding to one or more FcγR receptors (i.e., "FcγR variants"). Substitutions that result in both increased and decreased binding can be useful. For example, increased binding to FcγRIIIa is known to generally lead to increased ADCC (antibody-dependent cell-mediated cytotoxicity; a cell-mediated reaction in which nonspecific toxic cells expressing FcγR recognize binding antibodies on target cells and subsequently cause lysis of the target cells). Similarly, in some cases, decreased binding to FcγRIIb (an inhibitory receptor) may also be beneficial. Amino acid substitutions that can be used in the subject antibody include those listed in U.S. Patent Nos. 8,188,321 (particularly Figure 41) and 8,084,582, and U.S. Publication Applications Nos. 20060235208 and 20070148170, all of which are expressly incorporated herein by reference in their entirety and specifically with respect to the variants disclosed therein that affect Fcγ receptor binding. Specific variants available 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. Such modifications can be included in one or both Fc domains of the subject antibody.

[0239] In some embodiments, the subject antibody includes one or more Fc modifications that increase serum half-life. Fc substitutions that can increase binding to the FcRn receptor and increase serum half-life, as specifically disclosed in USSN 12 / 341,769 (hereinafter incorporated herein by reference in its entirety), include, but are not limited to, 434S, 434A, 428L, 308F, 259I, 428L / 434S, 259I / 308F, 436I / 428L, 436I or V / 434S, 436V / 428L, 259I / 308F / 428L, and M252Y / S254T / T256E. Such modifications may be included in one or both Fc domains of the subject antibody.

[0240] 2. Ablation variants In some embodiments, the heterodimeric antibody includes one or more modifications that 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. Such modifications are referred to as “FcγR ablation variants” or “Fc knockout (FcKO or KO)” variants. In these embodiments, for 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 when using bispecific antibodies that monovalently bind to CD28, it is generally desirable to ablate FcγRIIIa binding to eliminate or significantly reduce ADCC activity. In some embodiments, in the subject antibody described herein, at least one Fc domain comprises one or more Fcγ receptor ablation variants. In some implementations, in the subject antibody described herein, both Fc domains contain one or more Fcγ receptor ablation variants. Figure 5 These ablation variants are described, and each ablation variant may be independently and optionally included or excluded, wherein preferred aspects utilize ablation variants selected from the group consisting of: L234A / L235A / D265S, 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 in this article ablate FcγR binding, but typically do not ablate FcRn binding.

[0241] As is known in the art, the Fc domain of human IgG1 has the highest binding affinity to the Fcγ receptor, and therefore, ablative variants can be used when the constant domain (or Fc domain) in the backbone of a heterodimer antibody is IgG1. Alternatively, in addition to ablative variants in the IgG1 background, mutations at glycosylation position 297 (typically A or S) can significantly ablate binding to, for example, FcγRIIIa. Human IgG2 and IgG4 have naturally reduced binding affinity to the Fcγ receptor, and therefore those backbones can be used with or without ablative variants.

[0242] F. Combination of heterodimeric variant and Fc variant Those skilled in the art will understand that all the listed heterodimer variants (including skewed and / or pI variants) can be optionally and independently combined in any way, as long as they retain their “chain-like characteristics” or “monomer distribution”. Furthermore, all these variants can be combined into any heterodimer form.

[0243] Regarding pI variants, while the accompanying figures illustrate particularly useful implementations, other combinations can be generated by following the basic rule of altering the pI difference between the two monomers to facilitate purification.

[0244] In addition, any heterodimerization variant (skew variant and pI variant) can also be combined independently and optionally with Fc ablation variant, Fc variant, FcRn variant, as outlined in this paper.

[0245] Figure 8 Exemplary combinations of variants included in some embodiments of antibodies in the form of heterodimeric 1+1 Fab-scFv-Fc, 2+1 mAb-Fc, 2+1 Fab2-scFv-Fc, and 2+1 Fab2-Fc x scFv-Fc. In some embodiments, the heterodimeric antibody comprises, for example, Figure 8 The combination of the variants described.

[0246] G. Useful antibody forms Those skilled in the art will understand, and as will be discussed more fully below, that the heterodimeric bispecific antibody presented herein can be configured in several different ways, as generally depicted in Figure 14.

[0247] Those skilled in the art will understand that the heterodimeric form of the present invention can have different valence states and can be bispecific. That is, the heterodimeric antibody of the present invention can be bivalent and bispecific, or trivalent and bispecific, wherein the first antigen is bound by both binding domains, and the second antigen is bound by the second binding domain. As outlined herein, when CD28 is one of the target antigens, CD28 preferably binds monovalently only.

[0248] This invention utilizes a combination of CD28 bonding domains and ENPP3 bonding domains. Those skilled in the art will understand that the methods described in any of the accompanying drawings (in particular, see...) can be used... Figure 15 The anti-CD28 CDR, anti-CD28 variable light chain domain and variable heavy chain domain, Fab and scFv or variants thereof depicted in Figures 18 and 21) may also be used, optionally and independently in any combination, as depicted in any figure (e.g., Figure 22) of the CDR, variable light chain domain and variable heavy chain domain, Fab and scFv or variants thereof.

[0249] 1. 1 + 1 Fab-scFv-Fc format One heterodimeric antibody form that can be specifically used in the anti-CD28 x anti-ENPP3 antibody presented herein is a “1+1 Fab-scFv-Fc” or “opener” form, as shown in Figure 14A. The 1+1 Fab-scFv-Fc antibody comprises a first monomer that is a “conventional” heavy chain (VH1-CH1-hinge-CH2-CH3), wherein VH1 is a first variable heavy chain domain and CH2-CH3 is a first Fc domain. The 1+1 Fab-scFv-Fc also includes a light chain comprising a first variable light chain domain VL1 and a constant light chain domain CL. The light chain interacts with the VH1-CH1 of the first monomer to form a first antigen-binding domain, i.e., Fab. The second monomer of the antibody comprises a second binding domain and a second Fc domain, the second binding domain being a single-chain Fv (“scFv”, as defined below). The scFv includes a second variable heavy chain structural domain (VH2) and a second variable light chain structural domain (VL2), wherein VH2 is connected to VL2 using a potentially charged scFv connector (see example). Figure 6 The scFv is connected to the heavy chain using a domain connector (see, for example...). Figure 7 The two monomers are bound together by using amino acid variants (e.g., the heterodimerization variants discussed above) that promote the formation of heterodimerized antibodies in constant regions (e.g., Fc domains, CH1 domains, and / or hinge regions), as described more fully below. Due to its rough visual resemblance to a bottle opener, this structure is sometimes referred to herein as the "bottle opener" form. In some embodiments, the 1+1 Fab-scFv-Fc form antibody is a bivalent antibody.

[0250] The “1+1 Fab-scFv-Fc” form of this invention has several unique advantages. As is known in the art, antibody analogs that rely on two scFv constructs often suffer from stability and aggregation problems, which can be mitigated in this invention by adding “conventional” heavy and light chain pairings. Furthermore, unlike forms that rely on two heavy chains and two light chains, there is no problem of mispairing of heavy and light chains (e.g., heavy chain 1 paired with light chain 2, etc.).

[0251] In some embodiments of the 1+1 Fab-scFv-Fc antibody, one of the first or second antigen-binding domains is a CD28-binding domain and the other is an ENPP3-binding domain. In some embodiments of the 1+1 Fab-scFv-Fc, scFv binds to CD28 and Fab binds to ENPP3. Figure 24 depicts an exemplary anti-CD28 x anti-ENPP3 bispecific antibody in the form of a 1+1 Fab-scFv-Fc.

[0252] In some embodiments, the first and second Fc domains of the 1+1 Fab-scFv-Fc form antibody are heterodimerized skewed variants (e.g., as shown in Figure 3 and...). Figure 8 The Fc domain of the variants (shown as a set of amino acid substitutions) is particularly useful. Particularly useful heterodimerized skewed variants include 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 (EU designations)). In an exemplary embodiment, one of the first or second variant Fc domains includes the heterodimerization skew variant L368D / K370S, and the other of the first or second variant Fc domains includes the heterodimerization skew variant S364K / E357Q, wherein the designations are based on EU designations. In another exemplary embodiment, the first variant Fc domain includes the heterodimerization skew variant L368D / K370S, and the second variant Fc domain includes the heterodimerization skew variant S364K / E357Q, wherein the designations are based on EU designations.

[0253] In some implementations, the variant Fc domain includes an ablation variant (including...) Figure 5 (Those shown). In some embodiments, each of the first and second variant Fc domains includes the ablation variant E233P / L234V / L235A / G236_ / S267K, where the numbering is according to the EU numbering.

[0254] In some embodiments, the constant structural domain (CH1-hinge-CH2-CH3) of the first monomer includes pI variants (including those shown in Figure 3 and...). Figure 4(Those shown). In an exemplary embodiment, the constant structural domains (CH1-hinge-CH2-CH3) of the first monomer include pI variants N208D / Q295E / N384D / Q418E / N421D, where the numbering is in accordance with EU numbering.

[0255] In an exemplary embodiment, the 1+1 Fab-scFv-Fc form antibody comprises as follows Figure 8 The described combination of amino acid modifications. In such embodiments, the CH1-hinge-CH2-CH3 of the first monomer contains the amino acid variants L368D / K370S / N208D / Q295E / N384D / Q418E / N421D / E233P / L234V / L235A / G236del / S267K, and the second Fc domain contains the amino acid variants S364K / E357Q / E233P / L234V / L235A / G236del / S267K, wherein the numbering is in accordance with EU designations.

[0256] In some implementations, the scFv of the 1+1 Fab-scFv-Fc form antibody provided herein includes a charged scFv adapter (including... Figure 6 (Those shown). In some embodiments, the 1+1 Fab-scFv-Fc form antibody provided herein includes the FcRn variant M428L / N434S, where the numbering is based on the EU numbering.

[0257] In an exemplary embodiment of the 1+1 Fab-scFv-Fc form antibody, the first Fc domain includes a heterodimerized skewed variant L368D / K370S, and the second Fc domain includes a heterodimerized skewed variant S364K / E357Q; each of the first and second Fc domains includes an ablation variant E233P / L234V / L235A / G236_ / S267K; and the constant domain (CH1-hinge-CH2-CH3) of the first monomer includes pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is according to EU numbering. In some embodiments, the scFv of the 1+1 Fab-scFv-Fc form antibody provided herein includes a (GKPGS)4 charged scFv connector. In some embodiments, the 1+1 Fab-scFv-Fc form antibody provided herein comprises the FcRn variant M428L / N434S, wherein the designation is based on the EU designation. In some embodiments, the scFv of the 1+1 Fab-scFv-Fc form antibody provided herein comprises a charged scFv adapter (including... Figure 6 Those shown).

[0258] Subject 1 + 1 Fab-scFv-Fc form antibodies may include any suitable CD28 binding domain, including any CD28 binding domain provided herein. In some embodiments, the CD28 binding domain is one of the following CD28 binding domains or variants thereof: 1A7[CD28]_H1L1, 1A7[CD28]_H1.1_L1, 1A7[CD28]_H1_L1.71, 1A7[CD28]_H1.1_L1.71, 1A7[CD28]_H1.14_L1, 1A7[CD28]_H1.14_L1.71, CD28.3[CD28]_H0L0, hCD28.3[CD28]_H1L1, 5.11A1[CD28]_H0L 0, TGN1412_H1L1, 341VL34[CD28]_H1L1, 341VL36[CD28]_H1L1, 281VL4[CD28]_H1L1, HuTN228[CD28]_H1L1, PV1[CD28]_H0L0, m9.3[CD28]_H0L0, hu9.3[CD28]_H1L1, 9G2[CD28]_H0L0, 9G2[CD28]_H1L1, 2F10A3.140[CD28]_H1L1 and TN228[CD28]_H4L2 ( Figure 15 (Figures 18 and 21). Figure 15 Figures 16 and 17 depict additional VH and VL sequences of the exemplary CD28 binding domain that can be used for antibodies in the form of Topic 1+1 Fab-scFv-Fc.

[0259] In some embodiments of the 1+1 Fab-scFv-Fc form, the anti-CD28 ABD has VH and VL domains selected from the following: (i) VH, which includes vhCDR1, vhCDR2 and vhCDR3, each having Figure 15 (ii) the amino acid sequences of vhCDR1, vhCDR2, and vhCDR3 of any CD28-binding domain VH or variant thereof depicted in Figures 18 and 21; and (ii) VL, wherein the VL comprises vlCDR1, vlCDR2, and vlCDR3, which respectively have Figure 15 The amino acid sequences of vlCDR1, vlCDR2, and vlCDR3 of the CD28 binding domain VL or its variants depicted in Figures 18 and 21; or (i) VH, which includes vhCDR1, vhCDR2 and vhCDR3, each having Figure 15Or the amino acid sequences of vhCDR1, vhCDR2, and vhCDR3 of VH or its variants depicted in Figure 16; and (ii) VL, which comprises vlCDR1, vlCDR2, and vlCDR3, each having Figure 15 Or the amino acid sequences of vlCDR1, vlCDR2 and vlCDR3 of VL or its variants depicted in Figure 17.

[0260] In some embodiments of the 1+1 Fab-scFv-Fc form, the anti-CD28 ABD has VH and VL domains selected from the following: (i) VH, wherein the VH has Figure 15 (ii) the amino acid sequence of VH or a variant thereof of any CD28 binding domain depicted in Figures 18 and 21; and (ii) VL, wherein VL has Figure 15 The amino acid sequences of VL or variants of the CD28 binding domain depicted in Figures 18 and 21; or (i) VH, wherein the VH has Figure 15 Or the amino acid sequence of VH or its variants depicted in Figure 16; and (ii) VL, which has Figure 15 Or the amino acid sequence of VL or its variants as depicted in Figure 17.

[0261] The Topic 1+1 Fab-scFv-Fc form antibody may include any suitable ENPP3 binding domain, including any ENPP3 binding domain provided herein. Figure 22 depicts exemplary ENPP3 binding domains that may be used for the Topic 1+1 Fab-scFv-Fc form antibody.

[0262] In some embodiments of the 1+1 Fab-scFv-Fc form, the ENPP3 ABD has (i) VH, which has an amino acid sequence of VH or a variant thereof of any of the ENPP3 binding domains depicted in FIG22; and (ii) VL, which has an amino acid sequence of VL or a variant thereof of the ENPP3 binding domain depicted in FIG22.

[0263] Figure 9 illustrates some exemplary Fc domain sequences that can be used in 1+1 Fab-scFv-Fc antibody forms. The “Single 1” sequence depicted in Figure 9 generally refers to the Fc domain of the “Fab-Fc heavy chain,” and the “Single 2” sequence refers to the Fc domain of the “scFv-Fc heavy chain.” Additionally, Figure 11 and Figure 12 Exemplary CH1-hinge structural domains, CH1 structural domains, and hinge structural domains, which may be included in a first or second monomer of the form 1 + 1 Fab-scFv-Fc, are provided. Furthermore, Figure 13Useful CL sequences are provided for this form.

[0264] 2. 2+1 mAb-scFv form One heterodimeric antibody form that can be specifically used in the subject matter of bispecific anti-CD28 x anti-ENPP3 antibodies is the 2+1 mAb-scFv form shown in Figure 14E. This antibody form comprises three antigen-binding domains: two Fab moieties and an scFv domain linked to the C-terminus of a heavy chain. In some embodiments of this form, each of the Fab moieties binds to ENPP3 (in this case, human ENPP3), and the “additional” scFv domain binds to CD28. That is, this mAb-scFv form is a trivalent antibody.

[0265] In these embodiments, the first chain or monomer comprises VH1-CH1-hinge-CH2-CH3 from the N-end to the C-end, and the second monomer comprises VH1-CH1-hinge-CH2-CH3-domain connector-scFv domain from the N-end to the C-end, wherein the scFv domain comprises a second VH (VH2), a second VL (VL2), and an scFv connector. Regarding all scFv domains herein, the scFv domains may be oriented in any direction from the N-end to the C-end, i.e., VH2-scFv connector-VL2 or VL2-scFv connector-VH2. Therefore, the second monomer may comprise VH1-CH1-hinge-CH2-CH3-domain connector-VH2-scFv connector-VL2 or VH1-CH1-hinge-CH2-CH3-domain connector-VL2-scFv connector-VH2 from the N-end to the C-end. The composition also comprises a light chain VL1-CL. In some embodiments, this form comprises two identical light chains (VL1-CL). In this form, VH1 is each a first variable heavy chain (VH) domain, VL1 is each a first variable light chain (VL) domain, VH2 is a second variable heavy chain domain, and VL2 is a second variable light chain domain. In some embodiments, the form comprises two identical common light chains (each a VL1-CL), wherein each of the common light chains associates with a VH1-CH1 of the first and second monomers to form two identical Fabs (i.e., the first antigen-binding domain). In some embodiments, the scFv is the second antigen-binding domain. In some embodiments, the first ABD binds to human ENPP3, and the second ABD binds to human CD28.

[0266] In some embodiments, the first and second Fc domains of the 2+1 mAb-scFv form antibody are heterodimerized skewed variants (e.g., as shown in Figure 3 and...). Figure 8The Fc domain of variants (shown as a set of amino acid substitutions) is also shown. Particularly useful heterodimerized skewed variants include 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 (EU number)). In an exemplary embodiment, one of the first or second variant Fc domains includes the heterodimerization skew variant L368D / K370S, and the other of the first or second variant Fc domains includes the heterodimerization skew variant S364K / E357Q, wherein the designations are based on EU designations. In another exemplary embodiment, the first variant Fc domain includes the heterodimerization skew variant L368D / K370S, and the second variant Fc domain includes the heterodimerization skew variant S364K / E357Q, wherein the designations are based on EU designations.

[0267] In some implementations, the variant Fc domain includes an ablation variant (including...) Figure 5 (Those shown). In some embodiments, each of the first and second variant Fc domains includes the ablation variant E233P / L234V / L235A / G236_ / S267K, where the numbering is according to the EU numbering.

[0268] In some embodiments, the constant structural domain (CH1-hinge-CH2-CH3) of the first monomer includes pI variants (including those shown in Figure 3 and...). Figure 4 (Those shown). In an exemplary embodiment, the constant structural domains (CH1-hinge-CH2-CH3) of the first monomer include pI variants N208D / Q295E / N384D / Q418E / N421D, where the numbering is in accordance with EU numbering.

[0269] In some embodiments, the scFv of the 2+1 mAb-scFv antibody provided herein includes a charged scFv adapter (including... Figure 6 (Those shown). In some embodiments, the 2+1 mAb-scFv form antibody provided herein includes the FcRn variant M428L / N434S, where the numbering is based on the EU numbering.

[0270] In an exemplary embodiment, the 2+1 mAb-scFv antibody comprises as follows Figure 8The described combination of amino acid modifications. In such embodiments, the first variant Fc domain comprises a heterodimerized skewed variant L368D / K370S, and the second variant Fc domain comprises a heterodimerized skewed variant S364K / E357Q; each of the first and second variant Fc domains comprises an ablation variant E233P / L234V / L235A / G236_ / S267K; and the constant domain (CH1-hinge-CH2-CH3) of the first monomer comprises pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is according to EU numbering. In some embodiments, the scFv of the 2+1 mAb-scFv form antibody provided herein comprises a (GKPGS)4 charged scFv linker. In some implementations, the 2+1 mAb-scFv form antibody provided herein includes the FcRn variant M428L / N434S, where the numbering is based on the EU numbering.

[0271] In some embodiments, the scFv of the second monomer of the 2+1 Fab2-scFv-Fc form antibody is CD28-binding, and the VH1 of the first and second monomers and the VL1 of the common light chain each form an ENPP3-binding domain. The 2+1 mAb-scFv form antibody may include any suitable ENPP3-binding domain and CD28-binding domain, including any ENPP3-binding domain and CD28-binding domain provided herein, as well as the associated VH and VL or variants thereof (see, for example...). Figure 15 -18 and Figure 21).

[0272] In some embodiments of the 2 + 1 mAb-scFv form, the anti-CD28 ABD has VH and VL domains selected from the following: (i) VH, which includes vhCDR1, vhCDR2 and vhCDR3, each having Figure 15 (ii) the amino acid sequences of vhCDR1, vhCDR2, and vhCDR3 of any CD28-binding domain VH or variant thereof depicted in Figures 18 and 21; and (ii) VL, wherein the VL comprises vlCDR1, vlCDR2, and vlCDR3, which respectively have Figure 15 The amino acid sequences of vlCDR1, vlCDR2, and vlCDR3 of the CD28 binding domain VL or its variants depicted in Figures 18 and 21; or (i) VH, which includes vhCDR1, vhCDR2 and vhCDR3, each having Figure 15Or the amino acid sequences of vhCDR1, vhCDR2, and vhCDR3 of VH or its variants depicted in Figure 16; and (ii) VL, which comprises vlCDR1, vlCDR2, and vlCDR3, each having Figure 15 Or the amino acid sequences of vlCDR1, vlCDR2 and vlCDR3 of VL or its variants depicted in Figure 17.

[0273] In some embodiments of the 2 + 1 mAb-scFv form, the anti-CD28 ABD has VH and VL domains selected from the following: (i) VH, wherein the VH has Figure 15 (ii) the amino acid sequence of VH or a variant thereof of any CD28 binding domain depicted in Figures 18 and 21; and (ii) VL, wherein VL has Figure 15 The amino acid sequences of VL or variants of the CD28 binding domain depicted in Figures 18 and 21; or (i) VH, wherein the VH has Figure 15 Or the amino acid sequence of VH or its variants depicted in Figure 16; and (ii) VL, which has Figure 15 Or the amino acid sequence of VL or its variants as depicted in Figure 17.

[0274] In some embodiments of the 2 + 1 mAb-scFv form, each of the ENPP3 ABDs has (i) a VH having an amino acid sequence of the VH or a variant thereof of any of the ENPP3 binding domains depicted in FIG22; and (ii) a VL having an amino acid sequence of the VL or a variant thereof of the ENPP3 binding domain depicted in FIG22.

[0275] Figure 10 Some exemplary Fc domain sequences that can be used in the form of 2+1 mAb-scFv antibodies are shown. Additionally, Figure 11 and Figure 12 Exemplary CH1-hinge structural domains, CH1 structural domains, and hinge structural domains, which may be included in a first or second monomer of the form 2+1 mAb-scFv, are provided. Furthermore, Figure 13 Useful CL sequences are provided for this form.

[0276] Figure 41 depicts an illustrative sequence of ENPP3 x CD28 bsAb in the form of 2 + 1 mAb-scFv.

[0277] 3. 2 + 1 Fab2-scFv-Fc form One heterodimeric antibody form particularly useful in the anti-CD28 x anti-ENPP3 antibody presented herein is the 2+1 Fab2-scFv-Fc form (also referred to as the “central scFv form”) shown in Figure 14B. This antibody form comprises three antigen-binding domains: two Fab moieties and an scFv inserted between the VH-CH1 and CH2-CH3 regions of a monomer. In some embodiments of this form, each of the Fab moieties binds ENPP3, and the “additional” scFv domain binds CD28. In some embodiments, the 2+1 Fab2-scFv-Fc antibody is a trivalent antibody.

[0278] In some embodiments of the 2+1 Fab2-scFv-Fc form, the first monomer comprises a standard heavy chain (i.e., VH1-CH1-hinge-CH2-CH3), where VH1 is a first variable heavy chain domain and CH2-CH3 are first Fc domains. The second monomer comprises another first variable heavy chain domain (VH1), a CH1 domain (and optionally a hinge), a second Fc domain, and scFv, which comprises a scFv variable light chain domain (VL2), a scFv connector, and a scFv variable heavy chain domain (VH2). The scFv is covalently linked to the C-terminus of the CH1 domain of the second monomer and the N-terminus of the second Fc domain using an optional domain linker (VH1-CH1-[optional linker]-VH2-scFv linker-VH2-[optional linker]-CH2-CH3, or in the opposite orientation for scFv, i.e., VH1-CH1-[optional linker]-VL2-scFv linker-VH2-[optional linker]-CH2-CH3). The optional linker can be any suitable peptide linker, including, for example... Figure 7 The domain connectors included herein. This embodiment also utilizes a common light chain comprising a variable light chain domain (VL1) and a constant light chain domain (CL). In some embodiments, the form comprises two identical common light chains (each a VL1-CL), wherein each of the common light chains associates with a VH1-CH1 of the first and second monomers to form two identical Fabs. In some embodiments, the identical Fabs each bind ENPP3, and the scFv binds CD28. With respect to many embodiments herein, these constructs may include, as desired and as described herein, skew variants, pI variants, ablation variants, additional Fc variants, etc.

[0279] In some embodiments, the first and second Fc domains of the 2+1 Fab2-scFv-Fc form antibody are heterodimerized skewed variants (e.g., as shown in Figure 3 and...). Figure 8The Fc domain of the variants (shown as a set of amino acid substitutions) is particularly useful. Particularly useful heterodimerized skewed variants include 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 (EU designations)). In an exemplary embodiment, one of the first or second variant Fc domains includes the heterodimerization skew variant L368D / K370S, and the other of the first or second variant Fc domains includes the heterodimerization skew variant S364K / E357Q, wherein the designations are based on EU designations. In another exemplary embodiment, the first variant Fc domain includes the heterodimerization skew variant L368D / K370S, and the second variant Fc domain includes the heterodimerization skew variant S364K / E357Q, wherein the designations are based on EU designations.

[0280] In some implementations, the variant Fc domain includes an ablation variant (including...) Figure 5 (Those shown). In some embodiments, each of the first and second variant Fc domains includes the ablation variant E233P / L234V / L235A / G236_ / S267K, where the numbering is according to the EU numbering.

[0281] In some embodiments, the constant structural domain (CH1-hinge-CH2-CH3) of the first monomer includes pI variants (including those shown in Figure 3 and...). Figure 4 (Those shown). In an exemplary embodiment, the constant structural domains (CH1-hinge-CH2-CH3) of the first monomer include pI variants N208D / Q295E / N384D / Q418E / N421D, where the numbering is in accordance with EU numbering.

[0282] In some implementations, the scFv of the 2+1 Fab2-scFv-Fc form antibody provided herein includes a charged scFv adapter (including... Figure 6 (Those shown). In some embodiments, the 2+1 Fab2-scFv-Fc form antibody provided herein includes the FcRn variant M428L / N434S, where the numbering is based on the EU numbering.

[0283] In an exemplary embodiment, the 2+1 Fab2-scFv-Fc form antibody comprises as follows Figure 8 The described combination of amino acid modifications. In such embodiments, the first variant Fc domain comprises a heterodimerized skewed variant L368D / K370S, and the second variant Fc domain comprises a heterodimerized skewed variant S364K / E357Q; each of the first and second variant Fc domains comprises an ablation variant E233P / L234V / L235A / G236_ / S267K; and the constant domain (CH1-hinge-CH2-CH3) of the first monomer comprises pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is according to EU numbering. In some embodiments, the scFv of the 2+1 Fab2-scFv-Fc form antibody provided herein comprises a (GKPGS)4 charged scFv linker. In some implementations, the 2+1 Fab2-scFv-Fc form antibody provided herein includes the FcRn variant M428L / N434S, where the numbering is based on the EU numbering.

[0284] In some embodiments, the CH1-hinge-CH2-CH3 of the first monomer comprises amino acid variants L368D / K370S / N208D / Q295E / N384D / Q418E / N421D / E233P / L234V / L235A / G236del / S267K, and the second Fc domain comprises amino acid variants S364K / E357Q / E233P / L234V / L235A / G236del / S267K, wherein the numbering is in accordance with EU numbering.

[0285] In some embodiments, the scFv of the second monomer of the 2+1 Fab2-scFv-Fc form antibody is CD28-binding, and the VH1 of the first and second monomers and the VL1 of the common light chain each form an ENPP3-binding domain. The 2+1 Fab2-scFv-Fc form antibody may include any suitable CD28-binding domain, including any CD28-binding domain provided herein. In some embodiments, the CD28 binding domain is one of the following CD28 binding domains or variations thereof: 1A7[CD28]_H1L1, 1A7[CD28]_H1.1_L1, 1A7[CD28]_H1_L1.71, 1A7[CD28]_H1.1_L1.71, 1A7[CD28]_H1.14_L1, 1A7[CD28]_H1.14_L1.71, CD28.3[CD28]_H0L0, hCD28.3[CD28]_H1L1, 5.11A1[CD28]_H0L 0, TGN1412_H1L1, 341VL34[CD28]_H1L1, 341VL36[CD28]_H1L1, 281VL4[CD28]_H1L1, HuTN228[CD28]_H1L1, PV1[CD28]_H0L0, m9.3[CD28]_H0L0, hu9.3[CD28]_H1L1, 9G2[CD28]_H0L0, 9G2[CD28]_H1L1, 2F10A3.140[CD28]_H1L1 and TN228[CD28]_H4L2 ( Figure 15 (Figures 18 and 21).

[0286] In some embodiments of the 2+1 Fab2-scFv-Fc form, the anti-CD28 ABD has VH and VL domains selected from the following: (i) VH, which includes vhCDR1, vhCDR2 and vhCDR3, each having Figure 15 (ii) the amino acid sequences of vhCDR1, vhCDR2, and vhCDR3 of any CD28-binding domain VH or variant thereof depicted in Figures 18 and 21; and (ii) VL, wherein the VL comprises vlCDR1, vlCDR2, and vlCDR3, which respectively have Figure 15 The amino acid sequences of vlCDR1, vlCDR2, and vlCDR3 of the CD28 binding domain VL or its variants depicted in Figures 18 and 21; or (i) VH, which includes vhCDR1, vhCDR2 and vhCDR3, each having Figure 15Or the amino acid sequences of vhCDR1, vhCDR2, and vhCDR3 of VH or its variants depicted in Figure 16; and (ii) VL, which comprises vlCDR1, vlCDR2, and vlCDR3, each having Figure 15 Or the amino acid sequences of vlCDR1, vlCDR2 and vlCDR3 of VL or its variants depicted in Figure 17.

[0287] In some embodiments of the 2+1 Fab2-scFv-Fc form, the anti-CD28 ABD has VH and VL domains selected from the following: (i) VH, wherein the VH has Figure 15 (ii) the amino acid sequence of VH or a variant thereof of any CD28 binding domain depicted in Figures 18 and 21; and (ii) VL, wherein VL has Figure 15 The amino acid sequences of VL or variants of the CD28 binding domain depicted in Figures 18 and 21; or (i) VH, wherein the VH has Figure 15 Or the amino acid sequence of VH or its variants depicted in Figure 16; and (ii) VL, which has Figure 15 Or the amino acid sequence of VL or its variants as depicted in Figure 17.

[0288] In some embodiments, the VH1 of the first and second monomers and the VL1 of the common light chain of the 2+1 Fab2-scFv-Fc form antibody each form an ENPP3-binding domain. The 2+1 Fab2-scFv-Fc form antibody may include any suitable ENPP3-binding domain, including any ENPP3-binding domain provided herein. Figure 22 depicts exemplary ENPP3-binding domains that may be used in the 2+1 Fab2-scFv-Fc form antibody.

[0289] In some embodiments of the 2+1 Fab2-scFv-Fc form, each of the ENPP3 ABDs comprises (i) a VH having an amino acid sequence of the VH or a variant thereof of any of the ENPP3 binding domains depicted in FIG22; and (ii) a VL having an amino acid sequence of the VL or a variant thereof of the ENPP3 binding domain depicted in FIG22.

[0290] Figure 9 shows some exemplary Fc domain sequences that can be used in the 2+1 Fab2-scFv-Fc form of antibodies. Additionally, Figure 11 and Figure 12Exemplary CH1-hinge structural domains, CH1 structural domains, and hinge structural domains, which may be included in a first or second monomer of the form 2 + 1 Fab2-scFv-Fc, are provided. Furthermore, Figure 13 Useful CL sequences are provided for this form.

[0291] Figure 42 depicts an illustrative sequence of ENPP3 x CD28 bsAb in the form of 2 + 1 Fab2-scFv-Fc.

[0292] 4. 2 + 1 Fab2-Fc x scFv-Fc form One heterodimeric antibody form particularly useful in the anti-CD28 x anti-ENPP3 antibody presented herein is the 2+1 Fab2-Fc x scFv-Fc form (also referred to as the “stacked opener”) shown in Figure 14F. This form comprises a first monomer, a second monomer, and a common light chain. The first monomer comprises, from the N-terminus to the C-terminus: VH1-CH1-domain linker-VH1-CH1-hinge-CH2-CH3, wherein each of VH1 is a first variable heavy chain domain, and CH2-CH3 is a first variant Fc domain. The domain linker can be any useful domain linker (see, for example...). Figure 7 The second monomer comprises a single-chain Fv (“scFv”) covalently linked to a second variant Fc domain via a domain connector (scFv-domain connector-CH2-CH3). The common light chain comprises VL1-CL, where VL1 is a first variable light chain domain. The scFv of the second monomer comprises a second variable heavy chain domain (VH2) linked to a second variable light chain domain (VL2) via an scFv connector. In some embodiments, this form comprises two identical common light chains (VL1-CL). In this embodiment, the two VH1-CH1s of the first monomer each interact with the common light chain to form two identical Fabs, which are first antigen-binding domains, and VH2 and VL2 form a second antigen-binding domain. In some embodiments, each first antigen-binding domain is an ENPP3-binding domain, and the second antigen-binding domain is a CD28-binding domain. With respect to many embodiments herein, these constructs may include, as desired and as described herein, skewed variants, pI variants, ablation variants, additional Fc variants, etc.

[0293] In some embodiments, the first and second Fc domains of the 2+1 Fab2-Fc x scFv-Fc form antibody are heterodimerized skewed variants (e.g., as shown in Figure 3 and...). Figure 8The Fc domain of variants (shown as a set of amino acid substitutions) is also shown. Particularly useful heterodimerized skewed variants include 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 (EU designations)). In an exemplary embodiment, one of the first or second variant Fc domains includes the heterodimerization skew variant L368D / K370S, and the other of the first or second variant Fc domains includes the heterodimerization skew variant S364K / E357Q, wherein the designations are based on EU designations. In another exemplary embodiment, the first variant Fc domain includes the heterodimerization skew variant L368D / K370S, and the second variant Fc domain includes the heterodimerization skew variant S364K / E357Q, wherein the designations are based on EU designations.

[0294] In some implementations, the variant Fc domain includes an ablation variant (including...) Figure 5 (Those shown). In some embodiments, each of the first and second variant Fc domains includes the ablation variant E233P / L234V / L235A / G236_ / S267K, where the numbering is according to the EU numbering.

[0295] In some embodiments, the constant structural domain (CH1-hinge-CH2-CH3) of the first monomer includes pI variants (including those shown in Figure 3 and...). Figure 4 (Those shown). In an exemplary embodiment, the constant structural domains (CH1-hinge-CH2-CH3) of the first monomer include pI variants N208D / Q295E / N384D / Q418E / N421D, where the numbering is in accordance with EU numbering.

[0296] In some implementations, the 2+1 Fab2-Fc x scFv-Fc form antibody provided herein includes a charged scFv adapter (including... Figure 6 (Those shown). In some embodiments, the 2+1 Fab2-scFv-Fc form antibody provided herein includes the FcRn variant M428L / N434S, where the numbering is based on the EU numbering.

[0297] In an exemplary embodiment, the 2+1 Fab2-Fc x scFv-Fc form antibody comprises as follows Figure 8 The described combination of amino acid modifications. In such embodiments, the first variant Fc domain comprises a heterodimerized skewed variant L368D / K370S, and the second variant Fc domain comprises a heterodimerized skewed variant S364K / E357Q; each of the first and second variant Fc domains comprises an ablation variant E233P / L234V / L235A / G236_ / S267K; and the constant domain (CH1-hinge-CH2-CH3) of the first monomer comprises pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is according to EU designations. In some embodiments, the scFv of the 2+1 Fab2-scFv-Fc form antibody provided herein comprises a (GKPGS)4 charged scFv linker (SEQ ID NO:24). In some implementations, the 2+1Fab2-scFv-Fc form antibody provided herein includes the FcRn variant M428L / N434S, where the numbering is based on the EU numbering.

[0298] In some embodiments, the CH1-hinge-CH2-CH3 of the first monomer comprises amino acid variants L368D / K370S / N208D / Q295E / N384D / Q418E / N421D / E233P / L234V / L235A / G236del / S267K, and the second Fc domain comprises amino acid variants S364K / E357Q / E233P / L234V / L235A / G236del / S267K, wherein the numbering is in accordance with EU numbering.

[0299] In some embodiments, the scFv of the second monomer of the 2+1 Fab2-Fc x scFv-Fc form antibody is CD28-binding, and the VH1 of the first and second monomers and the VL1 of the common light chain each form an ENPP3-binding domain. The 2+1 Fab2-scFv-Fc form antibody may include any suitable CD28-binding domain, including any CD28-binding domain provided herein. In some embodiments, the CD28 binding domain is one of the following CD28 binding domains or variations thereof: 1A7[CD28]_H1L1, 1A7[CD28]_H1.1_L1, 1A7[CD28]_H1_L1.71, 1A7[CD28]_H1.1_L1.71, 1A7[CD28]_H1.14_L1, 1A7[CD28]_H1.14_L1.71, CD28.3[CD28]_H0L0, hCD28.3[CD28]_H1L1, 5.11A1[CD28]_H0L 0, TGN1412_H1L1, 341VL34[CD28]_H1L1, 341VL36[CD28]_H1L1, 281VL4[CD28]_H1L1, HuTN228[CD28]_H1L1, PV1[CD28]_H0L0, m9.3[CD28]_H0L0, hu9.3[CD28]_H1L1, 9G2[CD28]_H0L0, 9G2[CD28]_H1L1, 2F10A3.140[CD28]_H1L1 and TN228[CD28]_H4L2 ( Figure 15 (Figures 18 and 21).

[0300] In some embodiments of the 2 + 1 Fab2-Fc x scFv-Fc form, the anti-CD28 ABD has VH and VL domains selected from the following: (i) VH, which includes vhCDR1, vhCDR2 and vhCDR3, each having Figure 15 (ii) the amino acid sequences of vhCDR1, vhCDR2, and vhCDR3 of any CD28-binding domain VH or variant thereof depicted in Figures 18 and 21; and (ii) VL, wherein the VL comprises vlCDR1, vlCDR2, and vlCDR3, which respectively have Figure 15 The amino acid sequences of vlCDR1, vlCDR2, and vlCDR3 of the CD28 binding domain VL or its variants depicted in Figures 18 and 21; or (i) VH, which includes vhCDR1, vhCDR2 and vhCDR3, each having Figure 15Or the amino acid sequences of vhCDR1, vhCDR2, and vhCDR3 of VH or its variants depicted in Figure 16; and (ii) VL, which comprises vlCDR1, vlCDR2, and vlCDR3, each having Figure 15 Or the amino acid sequences of vlCDR1, vlCDR2 and vlCDR3 of VL or its variants depicted in Figure 17.

[0301] In some embodiments of the 2 + 1 Fab2-Fc x scFv-Fc, the anti-CD28 ABD has VH and VL domains selected from the following: (i) VH, wherein the VH has Figure 15 (ii) the amino acid sequence of VH or a variant thereof of any CD28 binding domain depicted in Figures 18 and 21; and (ii) VL, wherein VL has Figure 15 The amino acid sequences of VL or variants of the CD28 binding domain depicted in Figures 18 and 21; or (i) VH, wherein the VH has Figure 15 Or the amino acid sequence of VH or its variants depicted in Figure 16; and (ii) VL, which has Figure 15 Or the amino acid sequence of VL or its variants as depicted in Figure 17.

[0302] In some embodiments, the VH1 of the first and second monomers and the VL1 of the common light chain of the 2+1 Fab2-Fc x scFv-Fc form antibody each form an ENPP3-binding domain. The 2+1 Fab2-Fc x scFv-Fc form antibody may include any suitable ENPP3-binding domain, including any ENPP3-binding domain provided herein. Figure 22 depicts exemplary ENPP3-binding domains that may be used in the 2+1 Fab2-Fc x scFv-Fc form antibody.

[0303] In some embodiments of the 2+1 Fab2-Fc x scFv-Fc form, each of the ENPP3 ABDs includes (i) a VH having an amino acid sequence of the VH or a variant thereof of any of the ENPP3 binding domains depicted in Figure 22.

[0304] Figure 9 shows some exemplary Fc domain sequences that can be used in the form of 2+1 Fab2-Fc x scFv-Fc antibodies. Additionally, Figure 11 and Figure 12 Exemplary CH1-hinge structural domains, CH1 structural domains, and hinge structural domains are provided, which may be included in a first or second monomer of the form 2 + 1 Fab2-Fc x scFv-Fc. Furthermore, Figure 13 Useful CL sequences are provided for this form.

[0305] 5. 1 + 1 CLC form One heterodimeric antibody form particularly useful in the anti-CD28 x anti-ENPP3 antibody presented herein is a “1+1 common light chain” or “1+1 CLC” form, depicted in Figure 14C. The 1+1 CLC antibody comprises a first monomer comprising VH1-CH1-hinge-CH2-CH3, wherein VH1 is a first variable heavy chain domain and CH2-CH3 is a first Fc domain; a second monomer comprising VH2-CH1-hinge-CH2-CH3, wherein VH2 is a second variable heavy chain domain and CH2-C3 is a second Fc domain; and a third monomer, a “common light chain” comprising VL-CL, wherein VL is a common variable light chain domain and CL is a constant light chain domain. In such embodiments, VL pairs with VH1 to form a first binding domain having a first antigen binding specificity; and VL pairs with VH2 to form a second binding domain having a second antigen binding specificity. In some implementations, the 1+1 CLC antibody is a bivalent antibody.

[0306] In some embodiments, the first and second Fc domains of the 1+1 CLC form are heterodimerized skewed variants (e.g., as shown in Figure 3 and...). Figure 8 The Fc domain of variants (shown as a set of amino acid substitutions) is also shown. Particularly useful heterodimerized skewed variants include 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 (EU numbers). In an exemplary embodiment, one of the first or second variant Fc domains includes the heterodimerization skew variant L368D / K370S, and the other of the first or second variant Fc domains includes the heterodimerization skew variant S364K / E357Q, wherein the designations are based on EU designations. In another exemplary embodiment, the first variant Fc domain includes the heterodimerization skew variant L368D / K370S, and the second variant Fc domain includes the heterodimerization skew variant S364K / E357Q, wherein the designations are based on EU designations.

[0307] In some implementations, the variant Fc domain includes an ablation variant (including...) Figure 5 (Those shown). In some embodiments, each of the first and second variant Fc domains includes the ablation variant E233P / L234V / L235A / G236_ / S267K, where the numbering is according to the EU numbering.

[0308] In some embodiments, the constant structural domain (CH1-hinge-CH2-CH3) of the first or second monomer includes pI variants (including...) Figure 4 (as shown). In an exemplary embodiment, the constant structural domain (CH1-hinge-CH2-CH3) of the first or second monomer includes pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is in accordance with EU numbering.

[0309] In some implementations, the 1+1 CLC antibody form provided herein includes the FcRn variant M428L / N434S, where the numbering is based on the EU numbering.

[0310] In an exemplary embodiment, the first variant Fc domain includes heterodimerized skewed variants L368D / K370S, and the second variant Fc domain includes heterodimerized skewed variants S364K / E357Q; each of the first and second variant Fc domains includes ablation variants E233P / L234V / L235A / G236_ / S267K; and the constant domain (CH1-hinge-CH2-CH3) of the first monomer includes pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is in accordance with EU numbering.

[0311] In some embodiments, the CH1-hinge-CH2-CH3 of the first monomer comprises amino acid variants L368D / K370S / N208D / Q295E / N384D / Q418E / N421D / E233P / L234V / L235A / G236del / S267K, and the second Fc domain comprises amino acid variants S364K / E357Q / E233P / L234V / L235A / G236del / S267K, wherein the numbering is in accordance with EU numbering.

[0312] In some implementations, the 1+1 CLC antibody provided herein also includes the FcRn variant M428L / N434S, where the designation is based on the EU designation.

[0313] In some implementations, one of the first or second binding domains binds CD28, and the other binding domain binds ENPP3. Topic 1+1 CLC-form antibodies may include any suitable CD28-binding domain and ENPP3 domain, including any CD28-binding domain and ENPP3-binding domain or variants thereof provided herein (see, for example...). Figure 15 -18, Figures 21 and 22).

[0314] 6. 2 + 1 CLC form Another heterodimeric antibody form that can be specifically used in the CD28 x anti-ENPP3 antibody presented herein is the “2+1 common light chain” or “2+1 CLC” form, which is depicted in Figure 14D. The 2+1 CLC form comprises a first monomer comprising VH1-CH1-linker-VH1-CH1-hinge-CH2-CH3, wherein each of VH1 is a first variable heavy chain domain and CH2-CH3 is a first Fc domain; a second monomer comprising VH2-CH1-hinge-CH2-CH3, wherein VH2 is a second variable heavy chain domain and CH2-CH3 is a second Fc domain; and a third monomer comprising the “common light chain” VL-CL, wherein VL is a common variable light chain domain and CL is a constant light chain domain. VL pairs with each VH1 of the first monomer to form two first binding domains, each domain having first antigen binding specificity; and VL pairs with VH2 to form a second binding domain having second antigen binding specificity. The linker for the first monomer can be any suitable linker, including... Figure 7 Any of the described domain linkers or combinations thereof. In some embodiments, the 2+1 CLC form antibody is a trivalent antibody.

[0315] In some embodiments, the first and second Fc domains of the 2+1 CLC form are heterodimerized skewed variants (e.g., as shown in Figure 3 and...). Figure 8The Fc domain of variants (shown as a set of amino acid substitutions) is also shown. Particularly useful heterodimerized skewed variants include 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 (EU designations)). In an exemplary embodiment, one of the first or second variant Fc domains includes the heterodimerization skew variant L368D / K370S, and the other of the first or second variant Fc domains includes the heterodimerization skew variant S364K / E357Q, wherein the designations are based on EU designations. In another exemplary embodiment, the first variant Fc domain includes the heterodimerization skew variant L368D / K370S, and the second variant Fc domain includes the heterodimerization skew variant S364K / E357Q, wherein the designations are based on EU designations.

[0316] In some implementations, the variant Fc domain includes an ablation variant (including...) Figure 5 (Those shown). In some embodiments, each of the first and second variant Fc domains includes the ablation variant E233P / L234V / L235A / G236_ / S267K, where the numbering is according to the EU numbering.

[0317] In some embodiments, the constant structural domain (CH1-hinge-CH2-CH3) of the first or second monomer includes pI variants (including...) Figure 4 (as shown). In an exemplary embodiment, the constant structural domain (CH1-hinge-CH2-CH3) of the first or second monomer includes pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is in accordance with EU numbering.

[0318] In some implementations, the 2+1 CLC form antibody provided herein also includes the FcRn variant M428L / N434S, where the designation is based on the EU designation.

[0319] In exemplary embodiments, the first variant Fc domain includes the heterodimerized skewed variant L368D / K370S, and the second variant Fc domain includes the heterodimerized skewed variant S364K / E357Q; each of the first and second variant Fc domains includes the ablation variants E233P / L234V / L235A / G236_ / S267K; and the constant domain (CH1-hinge-CH2-CH3) of the first monomer includes the pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is in accordance with EU designations. In some embodiments, the 2+1 CLC antibody provided herein also includes the FcRn variant M428L / N434S, wherein the numbering is in accordance with EU designations.

[0320] In some embodiments, the CH1-hinge-CH2-CH3 of the second monomer comprises amino acid variants L368D / K370S / N208D / Q295E / N384D / Q418E / N421D / E233P / L234V / L235A / G236del / S267K, and the first Fc domain comprises amino acid variants S364K / E357Q / E233P / L234V / L235A / G236del / S267K, wherein the numbering is in accordance with EU numbering.

[0321] In some embodiments, each of the two first binding domains binds to tumor ENPP3, and the second binding domain binds to CD28. Topic 2+1 CLC-form antibodies may include any suitable CD28-binding domain and ENPP3 domain, including any CD28-binding domain and ENPP3-binding domain or variants thereof provided herein (see, for example...). Figure 15 -18, Figures 21 and 22).

[0322] 7. Dual SCFV format One heterodimeric antibody form that can be particularly used in the subject matter of bispecific anti-CD28 x anti-ENPP3 antibodies is a dual scFv form, as known in the art and illustrated in Figure 14G. In this embodiment, the heterodimeric bispecific antibody is composed of two scFv-Fc monomers (both 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.

[0323] In this case, all ABDs are in scFv form. Anti-CD28 x anti-ENPP3 antibodies in dual scFv form may include any suitable ENPP3 binding domain and CD28 binding domain, including any ENPP3 binding domain and CD28 binding domain provided herein.

[0324] Additionally, the Fc structural domain of the double scFv form includes skewed variants (e.g., as shown in Figure 3 and...). Figure 8 The group of amino acid substitutions shown includes particularly useful skewed variants 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), and optional ablation variants (including...). Figure 5 Those shown), optional live SCFV connectors (including) Figure 6 Those shown), and the heavy chain contains pI variants (including Figure 4 Those shown).

[0325] In some implementations, the dual scFv form includes a skew variant, a pI variant, and an ablation variant. Therefore, some embodiments include the following: a) a first monomer comprising the skewed variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, and a scFv (VH1-scFv linker-VL1-[optional domain linker]-CH2-CH3 or VL1-scFv linker-VH1-[optional domain linker]-CH2-CH3) binding to the first antigen; and b) a first monomer comprising the skewed variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and a scFv (VH1-scFv linker-VL1-[optional domain linker]-CH2-CH3 or VL1-scFv linker-VH1-[optional domain linker]-CH2-CH3) binding to the second antigen. pI variants may be as outlined herein, but the most common will be charged scFv connectors with opposite charges for each monomer. FcRn variants may optionally be included, particularly 428L / 434S.

[0326] The dual scFv form may include any suitable ENPP3 binding domain and CD28 binding domain, including any ENPP3 binding domain and CD28 binding domain provided herein, as well as the associated VH and VL or variants thereof (see, for example...). Figure 15 -18, Figures 21 and 22).

[0327] 8. Single-arm center scFv One heterodimeric antibody form that can be specifically used in the anti-CD28 x anti-ENPP3 antibody presented herein is the single-arm central scFv form shown in Figure 14K. In this embodiment, one monomer contains only the Fc domain, while the other monomer includes a Fab domain (first antigen-binding domain), an scFv domain (second antigen-binding domain), and an Fc domain, wherein the scFv domain is inserted between the Fc domains.

[0328] In this embodiment, a monomer comprises a first heavy chain containing a first variable heavy chain domain, a CH1 domain, and an Fc domain, wherein the scFv comprises a scFv variable light chain domain, an scFv connector, and a scFv variable heavy chain domain. The scFv is covalently connected 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 connector, in any orientation, i.e., VH1-CH1-[optional domain connector]-VH2-scFv connector-VL2-[optional domain connector]-CH2-CH3 or VH1-CH1-[optional domain connector]-VL2-scFv connector-VH2-[optional domain connector]-CH2-CH3. A second monomer comprises an Fc domain (CH2-CH3). This embodiment also utilizes a light chain containing a variable light chain domain and a constant light chain domain, which associates with the heavy chain to form a Fab. Regarding the many implementations described herein, these constructs include, as desired and as described herein, skew variants, pI variants, ablation variants, additional Fc variants, etc.

[0329] The single-arm center scFv form may include any suitable ENPP3-binding domain and CD28-binding domain, including any ENPP3-binding domain and CD28-binding domain provided herein, as well as the associated VH and VL or variants thereof (see, for example...). Figure 15 -18, Figures 21 and 22).

[0330] Figure 40 depicts an illustrative ENPP3 x CD28 bsAb sequence in the form of a single-arm central scFv.

[0331] 9. Single-arm scFv-mAb form One heterodimeric antibody form particularly suitable for use in the anti-CD28 x anti-ENPP3 antibody presented herein is the single-arm mAb-scFv form shown in Figure 14H. This form comprises: 1) a first monomer containing an “empty” Fc domain; 2) a second monomer comprising a first variable heavy chain domain (VH), an scFv domain (a second antigen-binding domain), and an Fc domain, wherein the scFv domain is attached to the N-terminus of the first variable heavy chain domain; and 3) a light chain comprising a first variable light chain domain and a constant light chain domain. The first variable heavy chain domain and the first variable light chain domain form a first antigen-binding domain, and the scFv is the second antigen-binding domain. In this form, one of the first antigen-binding domain and the second antigen-binding domain binds CD28, and the other antigen-binding domain binds ENPP3. With respect to many embodiments described herein, these constructs include, as desired and as described herein, skewed variants, pI variants, ablation variants, additional Fc variants, etc.

[0332] The single-arm scFv-mAb antibody may include any suitable ENPP3 binding domain and CD28 binding domain, including any ENPP3 binding domain and CD28 binding domain provided herein, as well as associated VH and VL or variants thereof (see, for example...). Figure 15 -18, Figures 21 and 22).

[0333] 10. scFv-mAb format One heterodimeric antibody form particularly useful in the anti-CD28 x anti-ENPP3 antibody presented herein is the mAb-scFv form shown in Figure 14I. In this embodiment, the form relies on the use of scFv linked to the N-terminus of a monomer to form a third antigen-binding domain, wherein the Fab portions of the two monomers each bind to a target, and the “additional” scFv domain binds to a different target.

[0334] In this embodiment, the first monomer comprises a first heavy chain (containing a variable heavy chain domain and a constant domain), wherein the N-terminally covalently connected scFv comprises a scFv variable light chain domain, a scFv connector, and a scFv variable heavy chain domain, in either orientation ((vh1-scFv connector-vl1-[optional domain connector]-vh2-CH1-hinge-CH2-CH3) or (when the scFv is in the opposite orientation) ((vl1-scFv connector-vh1-[optional domain connector]-vh2-CH1-hinge-CH2-CH3)). The second monomer comprises the heavy chain VH2-CH1-hinge-CH2-CH3. This embodiment also utilizes a common light chain containing a variable light chain domain and a constant light chain domain, which associates with the heavy chain to form two identical Fabs. Regarding many embodiments herein, these constructs include, as desired and as described herein, skewed variants, pI variants, ablation variants, additional Fc variants, etc.

[0335] The scFv-mAb antibody form may include any suitable ENPP3 binding domain and CD28 binding domain, including any ENPP3 binding domain and CD28 binding domain provided herein, as well as associated VH and VL or variants thereof (see, for example...). Figure 15 -18, Figures 21 and 22).

[0336] 11. mAb-Fv form One heterodimeric antibody form particularly suitable for use in the anti-CD28 x anti-ENPP3 antibody presented herein is the mAb-Fv form (Figure 14L). In this embodiment, the form relies on the use of an “additional” variable heavy chain domain linked to the C-terminus of one monomer and an “additional” variable light chain domain linked to the C-terminus of another monomer to form a third antigen-binding domain (i.e., the “additional” Fv domain), wherein the Fab portions of both monomers bind CD28 and the “additional” Fv domain binds ENPP3.

[0337] In this embodiment, the first monomer comprises a first heavy chain, the first heavy chain comprising a first variable heavy chain structural domain and a first constant heavy chain structural domain comprising a first Fc structural domain, wherein the first variable light chain structural domain is covalently connected to the C end of the first Fc structural domain using a structural domain joint (vh1-CH1-hinge-CH2-CH3-[optional joint]-vl2). The second monomer comprises a second variable heavy chain domain, a second constant heavy chain domain comprising a second Fc domain, and a third variable heavy chain domain (vh1-CH1-hinge-CH2-CH3-[optional connector]-vh2) covalently connected to the C-terminus of the second Fc domain using a domain connector. This embodiment also utilizes a common light chain comprising a variable light chain domain and a constant light chain domain, which associates with the heavy chain to form two identical Fabs, each comprising two identical Fvs. The two C-terminally connected variable domains constitute an “additional” third Fv. Regarding the many embodiments described herein, these constructs include, as desired and as described herein, skew variants, pI variants, ablation variants, additional Fc variants, etc.

[0338] The mAb-Fv form may include any suitable ENPP3 binding domain and CD28 binding domain, including any ENPP3 binding domain and CD28 binding domain provided herein, as well as the associated VH and VL or variants thereof (see, for example...). Figure 15 -18, Figures 21 and 22).

[0339] 12. Central Fv form One heterodimeric antibody form particularly useful in the anti-CD28 x anti-ENPP3 antibody presented herein is the central Fv form shown in Figure 14M. In this embodiment, the form relies on the use of an inserted Fv domain to form an “additional” third antigen-binding domain, wherein the Fab portions of both monomers bind ENPP3, and the “additional” central Fv domain binds CD28. The Fv domain is inserted between the Fc domain and the CH1-Fv region of the monomers to provide the third antigen-binding domain, wherein each monomer contains an Fv component (e.g., one monomer contains a variable heavy chain domain, and the other monomer contains a variable light chain domain of the “additional” central Fv domain).

[0340] In this embodiment, one monomer comprises a first heavy chain, which includes a first variable heavy chain structural domain, a CH1 structural domain, an Fc structural domain, and an additional variable light chain structural domain. The additional variable light chain structural domain is covalently connected between the C-end of the CH1 structural domain of the heavy chain constant structural domain and the N-end of the first Fc structural domain using a structural domain joint (vh1-CH1-[optional joint]-vh2-hinge-CH2-CH3). Another monomer comprises a first heavy chain, which includes a first variable heavy chain structural domain, a CH1 structural domain, an Fc structural domain, and an additional variable heavy chain structural domain (vh1-CH1-[optional joint]-vh2-hinge-CH2-CH3). The additional variable heavy chain structural domain is covalently connected between the C-end of the CH1 structural domain of the heavy chain constant structural domain and the N-end of the first Fc structural domain using a structural domain joint. This implementation utilizes a common light chain comprising a variable light chain domain and a constant light chain domain, which associates with a heavy chain to form two identical Fabs, each of which binds ENPP3. The additional variable heavy chain domain and the additional variable light chain domain form an “extra” central Fv that binds CD28. Regarding the many implementations described herein, these constructs include, as desired and as described herein, skew variants, pI variants, ablation variants, additional Fc variants, etc.

[0341] The central Fv form may include any suitable ENPP3 binding domain and CD28 binding domain, including any ENPP3 binding domain and CD28 binding domain provided herein, as well as the associated VH and VL or variants thereof (see, for example...). Figure 15 -18, Figures 21 and 22).

[0342] 13. Non-heterodimer bispecific antibody Those skilled in the art will understand that the anti-CD28 x anti-ENPP3 antibody provided herein may also be included in a non-heterodimeric bispecific form (see Figure 14J). In this form, the anti-CD28 x anti-ENPP3 comprises: 1) a first monomer containing VH1-CH1-hinge-CH2-CH3; 2) a second monomer containing VH2-CH1-hinge-CH2-CH3; 3) a first light chain containing VL1-CL; and 4) a second light chain containing VL2-CL. In such embodiments, VH1 and VL1 form a first antigen-binding domain, and VH2 and VL2 form a second antigen-binding domain. One of the first or second antigen-binding domains binds CD28, and the other antigen-binding domain binds ENPP3.

[0343] Anti-CD28 x anti-ENPP3 antibodies in the form of non-heterodimeric bispecific antibodies may include any suitable ENPP3-binding domain and CD28-binding domain, including any ENPP3-binding domain and CD28-binding domain provided herein, as well as associated VH and VL or variants thereof (see, for example...). Figure 15 -18, Figures 21 and 22).

[0344] 14. Trident Form In some embodiments, the anti-CD28 x anti-ENPP3 antibody provided herein is in a “trident” form, as generally described in WO2015 / 184203, which is hereby expressly incorporated herein by reference in its entirety, with particular attention to the figures, illustrations, definitions, and sequences relating to the “heterodimer promoting domain” or “HPD” (including “K-helix” and “E-helix” sequences). The trident relies on the use of two distinct HPDs that associate to form a heterodimeric structure as a component of the structure, see Figure 14N. In this embodiment, the trident form comprises a “conventional” heavy chain and a light chain (e.g., VH1-CH1-hinge-CH2-CH3 and VL1-CL), a third chain (VH2-(linker)-VL3-HPD1) containing a first “bifunctional antibody type binding domain” or “DART®”, and a fourth chain (VH3-(linker)-(linker)-VL2-HPD2) containing a second DART®. VH1 and VL1 form the first ABD, VH2 and VL2 form the second ABD, and VH3 and VL3 form the third ABD. In some cases, as shown in Figure 14P, the second and third ABDs bind to the same antigen.

[0345] The trident form may include any suitable ENPP3 binding domain and CD28 binding domain, including any ENPP3 binding domain and CD28 binding domain provided herein, as well as the associated VH and VL or variants thereof (see, for example...). Figure 15 -18, Figures 21 and 22).

[0346] V. Nucleic Acids On the other hand, this article provides a nucleic acid composition encoding the anti-CD28 x anti-ENPP3 antibody provided herein. The nucleic acid composition may refer to one or more polynucleotides.

[0347] Those skilled in the art will understand that the nucleic acid composition will depend on the form and scaffold of the heterodimeric protein. Therefore, for example, when the form requires a three-amino acid sequence, such as for the 1+1 Fab-scFv-Fc, 2+1 mAb-scFv, 2+1 Fab2-scFv-Fc, and 2+1 Fab2-Fc x scFv-Fc forms, the three polynucleotides can be incorporated into one or more expression vectors for expression. In an exemplary embodiment, each polynucleotide is incorporated into a different expression vector.

[0348] As is known in the art, nucleic acids encoding the binding domains and antibody components disclosed herein may be incorporated into expression vectors, as is known in the art, and depend on the host cell used to generate the heterodimeric antibodies of the present invention. Typically, the nucleic acid is operatively linked to multiple regulatory elements (promoters, origins of replication, selectivity markers, ribosome binding sites, inducers, etc.). The expression vector may be an extrachromosomal or integrated vector.

[0349] Next, the polynucleotides and / or expression vectors of the present invention are transformed into a variety of different types of host cells (including mammalian, bacterial, yeast, insect and / or fungal cells) well known in the art, wherein mammalian cells (e.g., CHO cells) can be used in many embodiments.

[0350] In some embodiments, the polynucleotide encoding each monomer is contained in a single expression vector, typically under the control of different or the same promoter. In embodiments particularly used in this invention, each of these polynucleotides is contained on a different expression vector. As shown herein and in US 62 / 025,931 (hereinafter incorporated by reference), different vector ratios can be used to drive heterodimer formation. That is, surprisingly, although the protein contains a 1:1:2 ratio of first monomer: second monomer: light chain (in many embodiments herein having three polypeptides constituting a heterodimeric antibody), these are not ratios that produce optimal results.

[0351] The antibodies presented herein are prepared by culturing host cells containing expression vectors well known in the art. Once generated, conventional antibody purification steps, including ion-exchange chromatography, are performed. As discussed herein, ensuring that the pIs of the two monomers differ by at least 0.5 allows for separation by ion-exchange chromatography, isoelectric focusing, or other isoelectric point-sensitive methods. That is, pI substitutions, including altering the isoelectric point (pI) of each monomer, result in each monomer having a different pI and the heterodimer also having a unique pI, thereby facilitating isoelectric purification of the “1+1 Fab-scFv-Fc” heterodimer (e.g., anion-exchange columns, cation-exchange columns). These substitutions also facilitate the identification and monitoring of any contaminating bis-scFv-Fc and mAb homodimers after purification (e.g., IEF gels, cIEF, and analytical IEX columns).

[0352] VI. Biological and biochemical functions of anti-CD28 x anti-ENPP3 antibodies The anti-CD28 x anti-ENPP3 antibodies described herein are typically administered to patients with ENPP3-related cancers, and their efficacy is assessed in a variety of ways as described herein. Therefore, immuno-oncology therapy can also be evaluated based on immune status assessments, where standard efficacy assays (such as cancer burden, tumor size, assessment of the presence or extent of metastasis) are available. This can be performed in various ways, including in vitro and in vivo assays.

[0353] A. Antibody compositions for in vivo administration Embodiments of the present invention relate to a pharmaceutical composition comprising any of the anti-CD28 x anti-ENPP3 antibodies described herein and a pharmaceutically acceptable carrier. Formulations of the anti-CD28 x anti-ENPP3 antibodies described herein are prepared by mixing an antibody having the desired level of purity with an optional pharmaceutically acceptable carrier, excipient, or stabilizer (Remington's Pharmaceutical Sciences, 16th edition, Osol, A. ed.

[1980] ) for storage as a lyophilized formulation or an aqueous solution. Acceptable carriers, excipients, or stabilizers are non-toxic to the recipient at the doses and concentrations used and include buffers such as phosphates, citrates, and other organic acids; antioxidants, including ascorbic acid and methionine; preservatives (e.g., octadecyl dimethyl benzyl ammonium chloride; hexamethyl diammonium chloride; benzalkonium chloride; benzyl chloride; phenol, butyl, or benzyl alcohol; alkyl esters of p-hydroxybenzoate, such as methylparaben or propylparaben; catechol; resorcinol; cyclohexanol; 3-pentanol; and m-cresol); low molecular weight peptides (less than about 10 residues); 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 dextrin; chelating agents, such as EDTA; sugars, such as sucrose, mannitol, trehalose, or sorbitol; salt-forming counterions, such as sodium; metal complexes (e.g., Zn-protein complexes); and / or nonionic surfactants, such as TWEEN™, PLURONICS™, or polyethylene glycol (PEG).

[0354] VII. Treatment Once prepared, the compositions of this invention can be used for a variety of oncology applications to treat cancer, typically by enhancing immune responses (e.g., T cell activation and proliferation), particularly when used in conjunction with anticancer therapies (such as antitumor bispecific antibodies). In some embodiments, the antibodies provided herein enhance immune responses (e.g., T cell activation and proliferation) by providing agonistic co-stimulation to T cells in a tumor microenvironment expressing ENPP3.

[0355] In some implementations, the anti-CD28 x anti-ENPP3 bispecific antibody provided herein is administered in conjunction with an antitumor therapy, which includes, for example, an antitumor-associated antigen (TAA) bispecific antibody.

[0356] A. Anti-CD28 x anti-ENPP3 / anti-ENPP3 bispecific antibody In some embodiments, the antiCD28 x antiENPP3 antibody provided herein is administered together with an antiENPP3 bispecific antibody, said bispecific antibody being a T-cell binding bispecific antibody, such as those that bind to human CD3.

[0357] In classical T cell / APC interactions, there is a first signal (signal 1) provided by the responsiveness of the TCR to peptide-MHC and a second signal (signal 2) provided by the crosslinking of CD80 / CD86 expressed on APCs with CD28, which together fully activate the T cell (see [link to T cell interaction]). Figure 23 A). In contrast, treatment with CD3 bispecific antibodies targeting tumor-associated antigens (TAAs), i.e., anti-CD3 x anti-ENPP3 bispecific antibodies, only provided a first signal.

[0358] Unbound by any particular operational theory, the anti-CD28 x anti-ENPP3 bispecific antibody presented herein is believed to enhance the anti-tumor response of the anti-CD3 x anti-ENPP3 bispecific antibody through CD28 co-stimulation (see [link to article]). Figure 23 B). Therefore, on the one hand, this article provides a method for treating ENPP3-related cancers in patients by administering the anti-CD3 x anti-ENPP3 bispecific antibodies and anti-CD28 x anti-ENPP3 bispecific antibodies provided herein.

[0359] Anti-CD3 x anti-ENPP3 antibodies that can be used in combination with the subject anti-CD28 x anti-ENPP3 antibody to provide "Signal 1" include those having any of the CD3-binding domains and ENPP3-binding domains described herein (see, for example, Figures 22 and 25). Suitable antibody forms of such anti-CD3 x anti-ENPP3 antibodies include, but are not limited to, the antibody forms described herein (see, for example, Figure 14). In some embodiments, the combined anti-CD3 x anti-ENPP3 antibody and the anti-CD28 x anti-ENPP3 antibody bind to the same ENPP3 epitope. In some embodiments, the combined anti-CD3 x anti-ENPP3 antibody and the anti-CD28 x anti-ENPP3 antibody bind to different ENPP3 epitopes.

[0360] B. Application mode The antibodies presented in this article are administered to subjects according to known methods, such as intravenous administration by bolus injection or by continuous infusion over a period of time.

[0361] C. Treatment Model In the method of the present invention, the therapy is used to provide a positive therapeutic response to a disease or ailment.

[0362] “Positive treatment response” means improvement of the disease or condition, and / or improvement of symptoms associated with the disease or condition. For example, a positive treatment response would refer to one or more of the following improvements: (1) a reduction in the number of neoplasms; (2) an increase in neoplasm death; (3) inhibition of neoplasm survival; (5) inhibition (i.e., to some extent slowing, 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.

[0363] A positive treatment response to any given disease or condition can be determined using standardized response criteria specific to that disease or condition. Screening techniques such as magnetic resonance imaging (MRI), X-ray imaging, computed tomography (CT), bone scan imaging, endoscopy, and tumor biopsy sampling (including bone marrow aspiration (BMA) and counting of circulating tumor cells) can be used to assess tumor response based on changes in tumor morphology (i.e., overall tumor burden, tumor size, etc.).

[0364] In addition to these positive treatment responses, subjects receiving therapy may also experience beneficial effects such as improvement in disease-related symptoms.

[0365] The treatment according to the present invention includes a "therapeuticly effective amount" of the drug used. "Therapeuticly effective amount" refers to the amount that effectively achieves the desired therapeutic outcome within the necessary dosage and time period.

[0366] Therapeutic effective doses can vary depending on factors such as an individual's disease state, age, sex, weight, and the drug's ability to elicit the desired response in the individual. Therapeutic effective doses are also the amount by which any toxic or harmful effects of the antibody or antibody component are exceeded by the beneficial therapeutic effect.

[0367] The "therapeutic effective dose" of cancer therapies can also be measured by their ability to stabilize disease progression. The ability of compounds to inhibit cancer can be evaluated in animal model systems that predict human tumor efficacy.

[0368] Alternatively, this property of the composition can be assessed by using in vitro assays known to skilled practitioners to examine the compound's ability to inhibit cell growth or induce apoptosis. Therapeutic amounts of a therapeutic compound can reduce tumor size or otherwise improve symptoms in a subject. Those skilled in the art will be able to determine such amounts based on factors such as the subject's body size, the severity of the subject's symptoms, and the specific composition or route of administration chosen.

[0369] Dosing regimens are adjusted to provide the optimal desired response (e.g., therapeutic response). For example, a single large pill may be administered, several separate doses may be administered over time, or the dose may be proportionally reduced or increased as indicated by the urgency of the treatment situation. Parenteral compositions may be formulated in unit dosage form to facilitate administration and uniformity of dosage. As used herein, unit dosage form refers to a physically discrete unit suitable as a unit dose to a subject to be treated; each unit contains a predetermined amount of the active compound, said amount calculated to combine with the desired drug carrier to produce the desired therapeutic effect.

[0370] The specifications of the dosage unit form of the present invention are subject to and directly depend on (a) the unique characteristics of the active compound and the specific therapeutic effect to be achieved, and (b) the inherent limitations of the field of mixing such active compounds for treating individual sensitivities.

[0371] The effective dose and dosage regimen of the bispecific antibody used in this invention depend on the disease or ailment to be treated and can be determined by those skilled in the art.

[0372] Example Example 1: CD28 binding domain The sequences of CD28 in humans, mice, and cynomolgus monkeys are depicted in Figure 1 and can be used to develop cross-reactive CD28 antigen-binding domains for clinical development.

[0373] 1A: Novel CD28 Binding Domain One approach considered to avoid hyperexcitatory effects associated with TGN1412 is to generate novel CD28-binding domains that have lower binding affinity for CD28 compared to TGN1412 and / or bind to different CD28 epitopes. In one activity generating such novel CD28-binding domains, an internal de novo phage library was screened for CD28.

[0374] 1A(a): Phage clone 1A7 It should be noted that this phage library utilizes human VL and introduces diversity into LCDR3. Figure 15 The amino acid sequence of an exemplary phage-derived clone 1A7 is depicted.

[0375] To optimize CD28 bsAb, a variety of 1A7 affinity variants were developed by engineering the VH variant (illustrated sequence as depicted in Figure 16 and SEQ ID NO: XXX-YYY), the VL variant (illustrated sequence as depicted in Figure 17 and SEQ ID NO: XXX-YYY), the common sequence (depicted in Figure 19), and combinations thereof (illustrated sequence as depicted in Figure 18). Figure 20The monovalent affinity of the illustrative variant in the scFv context is depicted.

[0376] Unexpectedly, it was found that the variable light chain domain of clone 1A7 differed from the parental line IGKV1-39 VL only in a single amino acid in LCDR3; however, despite this single amino acid difference, the VH of 1A7 did not pair well with the parental line VL, resulting in weakened binding. For example, 1A7_H1.14_L1 showed an affinity of approximately 200 nM, while 1A7_H1.14_IGKV1-39 showed an affinity >1 μM. Importantly, however, the 1A7 VH pairs with IGKV1-39, allowing them to be used in a common light chain bispecific form and to pair with other binding domains utilizing the variable light chain domain of IGKV1-39. Therefore, an affinity-optimized phage library was generated, focusing only on substitutions in the variable heavy chain domain of 1A7. Figure 47 and SEQ ID No: XXX-YYY depict exemplary affinity-optimized amino acid sequences of the 1A7 variable domain paired with IGKV1-39. The unit price KD of the affinity-optimized 1A7 Fab variant was determined using Octet, and its data are plotted on... Figure 48 middle.

[0377] After identifying these additional 1A7 VH variants, preferred variants were paired with the original 1A7_L1 and L1.71 VL to generate more affinity variants (in the scFv background). The KD of these additional variants is depicted in... Figure 44 Finally, to stabilize the form containing scFv, a disulfide-stabilized scFv can be used. This can potentially improve behavior in high-concentration formulations, enhance thermal stability, and prevent Fab-LC intercalation. A disulfide-stabilized scFv can be achieved by including cysteine ​​substitutions in VH and VL to form a disulfide bond between the two cysteine ​​residues. An illustrative example of such a disulfide-stabilized scFv is depicted in Figure 45, and KD is depicted in... Figure 44 In addition, such as Figure 43 The description indicates that disulfide bond stabilization improved the melting temperature by 5.7°C.

[0378] 1B: Another CD28 binding domain Figure 21 depicts a sequence of additional CD28 binding domains that can be used in the ENPP3 x CD28 bsAb of this invention.

[0379] Example 2: ENPP3 binding domain The sequences of human, mouse, and cynomolgus monkey ENPP3 are depicted in Figure 2 and can be used to develop cross-reactive ENPP3 antigen-binding domains for clinical development. Figure 22 depicts the sequence of the ENPP3 binding domain in the ENPP3 x CD28 bsAb that can be used in this invention. The KD values ​​of certain ENPP3 binding domains and human and cynomolgus monkey ENPP3 were measured using Biacore. The bivalent anti-ENPP3 IgG1 mAb was diluted in HBS-EP+ containing 0.5% BSA. The anti-ENPP3 mAb was captured at 10 nM on an anti-human Fc chip prepared by amine coupling of 50 μg / ml Affinipure goat anti-human Fc mAb for 10 s. The human ENPP3 (Acro Biosystem EN3-H52H4) or cynomolgus monkey ENPP3 analytes were flowed at 30 μl / min at 37°C for 5 min for association time and 5 min for dissociation time, and the results are depicted in Figure 22. Figure 46 middle.

[0380] Example 3: Engineering Modification of ENPP3 x CD28 bsAb Complete activation and differentiation of T cells require multiple signals. Signal 1 (promoted by the T cell receptor (TCR) recognition peptide-MHC (pMHC) complex) is absolutely necessary for T cell activation. Signal 2 works synergistically with and amplifies Signal 1, typically provided by the interaction of CD28 ligands CD80 and CD86 with CD28 itself. Although CD28 linkage alone is usually inert, when combined with Signal 1 activation, it promotes additional activation, survival, and proliferation signals, including IL2 secretion (see [link to relevant documentation]). Figure 23 Since CD80 and CD86 are naturally expressed only by specialized antigen-presenting cells (APCs), the degree of CD28 co-stimulation in the tumor environment can be highly variable. By creating novel tumor-targeting CD28 bispecific antibodies, the CD80 / CD86 linkage of CD28 can be mimicked, thereby providing an artificial source of signal 2. Notably, the signal can be provided through natural TCR:pMHC recognition in tumor cells, or it can be provided by a combination of a CD28 bispecific antibody and a CD3 bispecific antibody (which mimics signal 1). With these concepts in mind, the ENPP3 x CD28 bsAb was conceived and considered for use in various forms, a schematic outline of which is shown in Figure 14.

[0381] 3A: 1 + 1 Fab-scFv-Fc form An exemplary form utilizing the Fab domain and scFv is the 1+1 Fab-scFv-Fc form (schematically depicted in Figure 14A), comprising a first monomer containing a single-chain Fv (“scFv”) covalently linked to a first heterodimeric Fc domain and having first antigen-binding specificity, i.e., scFv-domain linker-CH2-CH3; a second monomer containing a heavy chain, i.e., VH-CH1-hinge-CH2-CH3, wherein CH2-CH3 is a second heterodimeric Fc domain complementary to the first heterodimeric Fc domain; and a light chain (LC) that is transfected separately to form a Fab domain with a variable heavy chain domain and having second antigen-binding specificity. Various heterodimerization methods known in the art can be used in this (and other) bispecific forms, in conjunction with various methods for purifying heterodimers from contaminating homodimers (including those depicted in Figure 3). Several linkers known in the art can be used to connect the VH and VL domains of scFv. Finally, maximizing the serum half-life of bsAbs may be useful, and several half-life-extending variants known in the art can be used in these bsAbs.

[0382] Specifically, the 1+1 Fab-scFv-Fc bsAb can utilize backbone 1 or 11 in Figure 9. This backbone utilizes the L368D / K370S (on HC):S364K / E357Q (on scFv-Fc) heterodimeric Fc variant. The HC side also includes the pI variants N208D / Q295E / N384D / Q418E / N421D to increase the negative charge of the heavy chain. The scFv utilizes a positively charged (GKPGS)4 linker between the VH and VL domains to increase the positive charge of the scFv-Fc chain. In general, both methods facilitate the easy purification of heterodimers from contaminating homodimers. The FcγR ablation variant utilized in this platform is an E233P / L234V / L235A / G236_ / S267K substitution on both the HC and scFv-Fc monomers. In some cases, bsAbs include M428L / N434S half-life extended variants. Figure 24 depicts an illustrative sequence of ENPP3 x CD28bsAbs in the form of 1 + 1 Fab-scFv-Fc.

[0383] 3B: Single-arm center scFv form Another exemplary form of utilizing the Fab domain and scFv is the single-arm central scFv (or "semi-stacked") form depicted in Figure 14K. In this embodiment, one monomer contains only the Fc domain, while the other monomer includes a Fab domain (first antigen-binding domain), an scFv domain (second antigen-binding domain), and an Fc domain, wherein the scFv domain is inserted between the Fc domains.

[0384] In this embodiment, a monomer comprises a first heavy chain containing a first variable heavy chain domain, a CH1 domain, and an Fc domain, wherein the scFv comprises a scFv variable light chain domain, an scFv connector, and a scFv variable heavy chain domain. The scFv is covalently connected 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 connector, in any orientation, i.e., VH1-CH1-[optional domain connector]-VH2-scFv connector-VL2-[optional domain connector]-CH2-CH3 or VH1-CH1-[optional domain connector]-VL2-scFv connector-VH2-[optional domain connector]-CH2-CH3. A second monomer comprises an Fc domain (CH2-CH3). This embodiment also utilizes a light chain containing a variable light chain domain and a constant light chain domain, which associates with the heavy chain to form a Fab. Regarding the many implementations described herein, these constructs include, as desired and as described herein, skew variants, pI variants, ablation variants, additional Fc variants, etc.

[0385] The single-arm center scFv form may include any suitable ENPP3-binding domain and CD28-binding domain, including any ENPP3-binding domain and CD28-binding domain provided herein, as well as the associated VH and VL or variants thereof (see, for example...). Figure 15 -18, Figures 21 and 22). Figure 40 depicts an exemplary ENPP3 x CD28bsAb in the form of a single-arm central scFv.

[0386] 3C : 2 + 1 mAb-scFv form Another exemplary form utilizing the Fab domain and scFv is the 2+1 mAb-scFv form (schematically depicted in Figure 14E), which comprises a first monomer containing a first heavy chain covalently linked to a single-chain Fv (“scFv”) having first antigen-binding specificity, namely VH-CH1-hinge-CH2-CH3-domain linker-scFv, where CH2-CH3 is a first heterodimeric Fc domain; a second monomer containing a heavy chain, namely VH-CH1-hinge-CH2-CH3, where CH2-CH3 is a second heterodimeric Fc domain complementary to the first heterodimeric Fc domain; and a light chain (LC) that is transfected separately to form a Fab domain with the two VH domains having second antigen-binding specificity. Specifically, the 2+1 Fab-scFv-Fc bsAb can utilize... Figure 10 The backbone utilizes a heterodimeric Fc variant of L368D / K370S (on HC):S364K / E357Q (on the HC-scFv side). HC also includes pI variants N208D / Q295E / N384D / Q418E / N421D to increase the negative charge of the heavy chain. The scFv utilizes a positively charged (GKPGS)4 linker between the VH and VL domains to increase the positive charge of the HC-scFv chain. In general, both methods facilitate the easy purification of heterodimers from contaminating homodimers. The FcγR ablation variant utilized in this platform is an E233P / L234V / L235A / G236del / S267K substitution on the HC and scFv-Fc monomers. In some cases, bsAb includes a half-life-extended variant of M428L / N434S. Figure 41 depicts an illustrative sequence of ENPP3 x CD28 bsAb in the form of 2 + 1 mAb-scFv.

[0387] 3D: 2 + 1 Fab2-scFv-Fc format Another exemplary form utilizing Fab domains and scFv is the 2+1 Fab2-scFv-Fc form (schematically depicted in Figure 14B), which can also be referred to as the 2+1 central scFv-Fc form. The 2+1 Fab2-scFv-Fc comprises a first monomer containing a heavy chain, namely VH-CH1-hinge-CH2-CH3, where CH2-CH3 is a first heterodimeric Fc domain; a second monomer containing VH-CH1-linker-scFv-linker-CH2-CH3, where CH2-CH3 is a second heterodimeric Fc domain complementary to the first heterodimeric Fc domain and where the scFv has a first antigen-binding specificity; and a light chain (LC) that is transfected separately to form a Fab domain with the two VH domains, exhibiting a second antigen-binding specificity. Figure 42 depicts an illustrative sequence of ENPP3 x CD28 bsAb in the form of 2 + 1 Fab2-scFv-Fc.

[0388] While a particularly useful framework has been provided above, a variety of heterodimerization methods known in the art can be used in the bispecific form, combined with various methods for purifying heterodimers from contaminating homodimers (including those depicted in Figure 3). A variety of linkers known in the art can be used to connect the VH and VL domains of scFv. Finally, maximizing the serum half-life of the bsAb can be useful, and a variety of half-life-extending variants known in the art can be used in these bsAbs. Regardless of the bsAb form, the CD28 bispecific antibody is monovalent to CD28 and incorporates Fc variants engineered to ablate FcγR binding to avoid potential hyperactivation. Such Fc variants include... Figure 5 Those that are described.

[0389] Example 4: Effective combination of ENPP3 x CD28 bsAb and CD3 bsAb As described in Example 3, the ENPP3 x CD28 bsAb is intended to be combined with a CD3 bsAb. Such CD3 bsAbs can utilize the CD3-binding domain depicted in Figure 24. For example, the ENPP3 x CD28 bsAb can be combined with another ENPP3 x CD3 bsAb (see, for example, U.S. Patent Nos. 11,472,890 and WO 2020 / 180726). Alternatively, the ENPP3 x CD28 bsAb can be combined with a CD3 bsAb targeting other antigens.

[0390] For research purposes, RXF-393 (ENPP3) +Tumor cells were mixed with purified T cells and incubated with escalating doses of the illustrative ENPP3 x CD28 bsAb XENP44587 (sequence depicted in Figure 24) and 1 µg / ml of the illustrative ENPP3 x CD3 bsAb. IL-2 and IFNγ levels in the cell supernatant were measured using MSD (Meso Scale Discovery, Rockville, Md.). Figure 26 The data depicted show that each ENPP3 x CD28 bsAb dose-dependently enhances IL-2 secretion driven by ENPP3 x CD3 bsAb.

[0391] Example 5: Further investigation into ENPP3 x CD28-enhanced ENPP3 x CD3 and B7H3 x CD3-mediated RTCC With or without 1 μg / mL XENP44587 ENPP3 (ENPP3-A) x CD28 (1A7 H1.14_L1.71, 37 nM) in 1+1 Fab-scFv-Fc form, overnight recovered donor PBMCs (80K, 2:1 E:T) were incubated with or without CFSE-labeled VMRC-RCW (40K) in a semi-logarithmic serial dilution of illustrative B7H3 x CD3 or ENPP3 x CD3. VMRC-RCW target cells, exhibiting ENPP3 antigen-binding capacity in the approximately 10K–20K range, represented papillary renal cell carcinoma. After 48 hours, cells were stained with surface antibodies to identify cell subpopulations. Cell viability was measured using Zombie Aqua. Figure 28-32 The results are described. PBMC+VMRC-RCW only refers to PBMC and CFSE-labeled VMRC-RCW incubated together in R10 medium at a 2:1 E:T ratio, without the addition of CD3 bsAb or 1 μg / mL CD28 bsAb. 1 μg / mL ENPP3 x CD28 refers to PBMC and CFSE-labeled VMRC-RCW incubated together in R10 medium at a 2:1 E:T ratio, without the addition of CD3 bsAb, but with the addition of 1 μg / mL CD28 bsAb. Figure 28 The described ENPP3 x CD28 bsAb enhances ENPP3 x CD3 ( Figure 28 A) and B7H3 x CD3 ( Figure 28 B) Mediated target cell killing. For example... Figure 29 As depicted, ENPP3 x CD28bsAb amplifies BclXL-induced RTCC in CD3 bsAb-mediated RTCC. (As shown...) Figure 30As depicted, ENPP3 x CD28 bsAb increases T cell cytotoxicity in CD3 bsAb-mediated RTCC, as measured by granzyme B and CD107a. Figure 31 illustrates that ENPP3 x CD28 bsAb increases T cell activation and PD1 upregulation in CD3 bsAb-mediated RTCC. Figure 32 As described, ENPP3 x CD28 upregulates intracellular IFNg in ENPP3 x CD3-mediated RTCC.

[0392] Example 6: Adjusting the efficacy of ENPP3 x CD28 bsAb As described in Example 3, various forms of ENPP3 x CD28 bsAb were conceived. Furthermore, as described in Examples 1 and 2, an experiment was conducted to investigate the effects of different forms and binding domains on the IL2 release efficacy of ENPP3 x CD28 bsAb. In this experiment, 8K adherent VMRC-RCW cells and 8K T cells were incubated overnight with a 1:1 E:T ratio of ENPP3 x CD28 diluted 3-fold and 1 μg / mL of ENPP3 x CD3. After incubation at 37°C for 24 h, the supernatant was collected for MSD analysis. T cells were isolated using the StemCell EasySep™ Human T Cell Enrichment Kit (catalog number 19051). The experimental results are depicted in... Figure 33 The study showed that the 1+1 Fab-scFv-Fc form exhibited the most efficient IL2 induction compared to 2+1 mAb-scFv, 2+1 Fab2-scFv-Fc (which can also be referred to as 2+1 central scFv), or single-arm central scFv (which can also be referred to as the "semi-stacked" form). Specifically, XENP46666, possessing the 1+1 Fab-scFv-Fc form along with the ENPP3-A ENPP3 binding domain and the H1.129_L1 SS (disulfide-stable) CD28 binding domain, showed the most efficient IL2 induction compared to other binding domains in the same 1+1 Fab-scFv-Fc form. Figure 34 The description.

[0393] In another experiment, T cells from 14 unique donors were isolated using the StemCell EasySep™ Human T Cell Enrichment Kit (catalog number 19051) and seeded with TUHR10TKB cells at a 1:1 E:T ratio (8K T cells: 8K target cells) with 1 μg / mL of illustrative ENPP3 x CD3, with or without 10 μg / mL of ENPP3 x CD28. TUHR10TKB cells exhibited approximately 52K ENPP3 antigen-binding capacity, similar to target cells of clear cell renal cell carcinoma. After incubation at 37°C for 24 h, the supernatant was collected for MSD analysis. Figure 27 As described, XENP4666 again exhibits significantly higher levels of IL-2 release compared to the “semi-stacked” form or the same form with the AN1ENPP3 binding domain.

[0394] Furthermore, another experiment demonstrated that ENPP3 x CD28 bsAb enhances EpCAM x CD3-mediated RTCC and IFNg production. Overnight recovered donor PBMCs (80K, 2:1E:T) were incubated with, or without, CFSE-labeled VMRC-RCW (40K) in a semi-logarithmic serial dilution of illustrative EpCAM x CD3, with or without 1 μg / mL of ENPP3 x CD28 test material. After 48 hours, cells were stained with the surface antibody to identify cell subpopulations. Cell viability was measured using Zombie Aqua. Figure 38 and Figure 39 As shown, compared with the AN1 ENPP3 binding domain of XENP46658, the ENPP3-A ENPP3 binding domain of XENP46666 is superior in enhancing target cell killing and INFg production.

[0395] Example 7: XENP46666 enhances IL2 and IFNylate release in antigen-specific RTCC assay For further investigation, pp65-MDA-MB-231 cells were transfected to express ENPP3, which has an antigen-binding capacity exceeding 500K. Next, 20K ENPP3+ pp65-MDA-MB-231 cells were adhered overnight and then incubated with 400K purified T cells from a CMV+HLA-A*02:01 donor at a 20:1 E:T ratio using a 3-fold serial dilution of XENP46666. CMV-responsive T cells were purified using the StemCell EasySep™ Human T Cell Enrichment Kit (catalog number 19051). After incubation at 37°C for 24 hours, the supernatant was collected for MSD analysis of cytokines. Figure 35The study described that XENP46666 enhanced IL2 release by 2-3 times compared to T cells alone plus target cells.

[0396] Example 8: In a mouse model, the combination of XENP46666 and XENP46667 with ENPP3 x CD3 showed antitumor efficacy. To investigate the in vivo antitumor efficacy of ENPP3 x CD28 bsAb, 150 female NSG-DKO mice were intradermally inoculated with 1x10⁻⁶ bsAb on day -7. 6 100 RXF-393 cells / mouse, volume 0.1 mL. On day -1, caliper measurements were performed to divide mice into 13 groups. On day 0, mice were implanted intraperitoneally with 5 x 10 RXF-393 cells / mouse. 6 One huPBMC / mouse, with a volume of 0.5 ml, followed by the first dose administered intraperitoneally in a volume of 0.1 ml. Figure 36 The test items indicated. As depicted, both XENP46666 and XENP46667, when administered in combination with ENPP3 x CD3, exhibited strong antitumor activity compared to ENPP3 x CD3 alone.

[0397] Example 9: XENP46666 and its alternative XENP46674 were well tolerated in cynomolgus monkeys. When using XENP46666 or XENP46674 (due to their stronger binding to cyno ENPP3, they are used as cyno alternatives, such as...) Figure 46 When the drug was administered to cynomolgus monkeys (as described), no related deaths or clinical outcomes, qualitative changes in food consumption, body weight, or hematological parameters were observed. Figure 37 The described half-life ranges from 8.59 to 10.6 days, and C... max and AUC 0-最后 The measured values ​​are dose-proportional across different dose levels.

Claims

1. A heterodimer antibody, said heterodimer antibody comprising: a) A first monomer, the first monomer comprising: i) Single-chain variable fragments (scFv); and ii) A first Fc structural domain, wherein the scFv is covalently connected to the N end of the first Fc structural domain using a structural domain connector; b) A second monomer, the second monomer comprising VH1-CH1-hinge-CH2-CH3 from the N-terminus to the C-terminus, wherein VH1 is a first variable heavy chain domain and CH2-CH3 is a second Fc domain; and c) A light chain comprising VL1-CL from the N-end to the C-end, wherein VL1 is a first variable light chain structural domain and CL is a constant light chain structural domain. The scFv includes a second VH structural domain (VH2), an scFv connector, and a second variable light chain structural domain (VL2). Wherein VH1 and VL1 together form a first antigen-binding domain (ABD), and VH2 and VL2 together form a second ABD, and One of the first ABD and the second ABD is a CD28 binding domain, and the other of the first ABD and the second ABD is an exonucleotide pyrophosphatase / phosphodiesterase 3 (ENPP3) binding domain.

2. The heterodimer antibody according to claim 1, wherein the scFv comprises a VH2-scFv linker-VL2 from the N-terminus to the C-terminus.

3. The heterodimer antibody according to claim 1, wherein the scFv includes a VL2-scFv linker-VH2 from the N-terminus to the C-terminus.

4. The heterodimeric antibody according to any one of claims 1 to 3, wherein the first ABD is the ENPP3 binding domain and the second ABD is the CD28 binding domain.

5. The heterodimer antibody according to claim 4, wherein VH1 and VL1 are VH and VL of any ENPP3 binding domain in Figure 22 or a variant thereof.

6. The heterodimer antibody according to claim 4 or 5, wherein VH2 and VL2 are selected from one of the following: (1) VH and VL or variants thereof of any CD28 binding domain in Figures 15, 18 and 21; or (2) (i) VH or a variant thereof in Figure 15 or Figure 16; and (ii) VL or a variant thereof in Figure 15 or Figure 17.

7. The heterodimeric antibody according to any one of claims 1 to 6, wherein the first Fc domain and the second Fc domain are each variant Fc domains.

8. The heterodimer antibody of claim 7, wherein the first and second Fc domains comprise a set of heterodimerization skewed variants selected from the following heterodimerization variants: S364K / E357Q: L368D / K370S; S364K: L368D / K370S; S364K: L368E / K370S; D401K: T411E / K360E / Q362E; and T366W: T366S / L368A / Y407V, wherein the numbering is based on EU numbering.

9. The heterodimer antibody of claim 8, wherein the first and second Fc domains comprise heterodimerized skewed variants S364K / E357Q: L368D / K370S, wherein the numbering is based on EU designations.

10. The heterodimeric antibody according to any one of claims 7 to 9, wherein each of the first and second Fc domains comprises one or more ablation variants.

11. The heterodimer antibody of claim 10, wherein the one or more ablation variants comprise E233P / L234V / L235A / G236del / S267K, wherein the numbering is in accordance with EU numbering.

12. The heterodimeric antibody according to any one of claims 7 to 11, wherein one of the first or second monomers further comprises one or more pI variants.

13. The heterodimer antibody of claim 12, wherein the CH1-hinge-CH2-CH3 of the second monomer comprises pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is in accordance with EU numbering.

14. The heterodimer antibody according to any one of claims 7 to 13, wherein the CH1-hinge-CH2-CH3 of the second monomer comprises amino acid variants E233P / L234V / L235A / G236del / S267K / L368D / K370S / N208D / Q295E / N384D / Q418E / N421D. The first Fc domain contains amino acid variants E233P / L234V / L235A / G236del / S267K / S364K / E357Q. Furthermore, the numbering is based on the EU numbering.

15. The heterodimeric antibody according to any one of claims 8-14, wherein the first and second variant Fc domains each further comprise amino acid variants 428L / 434S.

16. The heterodimeric antibody according to any one of claims 1 to 15, wherein the scFv linker is GKPGSGKPGSGKPGSGKPGS.

17. A heterodimer antibody, said heterodimer antibody comprising: a) A first monomer, which contains VH1-CH1-first domain connector-scFv-second domain connector-CH2-CH3 from the N end to the C end; VH1 is the first variable heavy chain domain, and CH2-CH3 is the first Fc domain; b) A second monomer, which contains VH1-CH1-hinge-CH2-CH3 from the N-terminus to the C-terminus, wherein CH2-CH3 is a second Fc domain; c) A first light chain, the first light chain comprising VL1-CL from the N-end to the C-end, wherein VL1 is a first variable light chain structural domain and CL is a constant light chain structural domain; and d) A second light chain, which comprises VL1-CL from the N-end to the C-end, wherein VL1 is a first variable light chain structural domain and CL is a constant light chain structural domain. The scFv includes a second VH structural domain (VH2), an scFv connector, and a second variable light chain structural domain (VL2). The first monomer's VH1 and the first light chain's VL1, and the second monomer's VH1 and the second light chain's VL1, each form a first antigen-binding domain (ABD), and the VH2 and the VL2 form a second ABD. The first ABD is an exonucleotide pyrophosphatase / phosphodiesterase 3 (ENPP3) binding domain, and the second ABD is a CD28 binding domain, or the first ABD is a CD28 binding domain and the second ABD is an ENPP3 binding domain.

18. The heterodimer antibody of claim 17, wherein the scFv comprises a VH2-scFv linker-VL2 from the N-terminus to the C-terminus.

19. The heterodimer antibody of claim 17, wherein the scFv comprises a VL2-scFv linker-VH2 from the N-terminus to the C-terminus.

20. The heterodimeric antibody according to any one of claims 17 to 19, wherein the first ABD is the ENPP3 binding domain and the second ABD is the CD28 binding domain.

21. The heterodimer antibody according to claim 20, wherein VH1 and VL1 are VH and VL of any ENPP3 binding domain in Figure 22.

22. The heterodimer antibody according to claim 20 or 21, wherein VH2 and VL2 are selected from one of the following: (1) VH and VL or variants thereof of any CD28 binding domain in Figures 15, 18 and 21; or (2) (i) VH or a variant thereof in Figure 15 or Figure 16; and (ii) VL or a variant thereof in Figure 15 or Figure 17.

23. The heterodimeric antibody according to any one of claims 17 to 22, wherein the first Fc domain and the second Fc domain are each variant Fc domains.

24. The heterodimerizing antibody of claim 23, wherein the first and second Fc domains comprise a set of heterodimerizing skewed variants selected from the following heterodimerizing variants: S364K / E357Q: L368D / K370S; S364K: L368D / K370S; S364K: L368E / K370S; D401K: T411E / K360E / Q362E; and T366W: T366S / L368A / Y407V, wherein the numbering is based on EU numbering.

25. The heterodimer antibody of claim 24, wherein the first and second Fc domains comprise heterodimerized skew variants S364K / E357Q: L368D / K370S, wherein the designation is based on EU designation.

26. The heterodimeric antibody according to any one of claims 23 to 25, wherein each of the first and second Fc domains comprises one or more ablation variants.

27. The heterodimer antibody of claim 26, wherein the one or more ablation variants comprise E233P / L234V / L235A / G236del / S267K, wherein the numbering is in accordance with EU numbering.

28. The heterodimeric antibody according to any one of claims 23 to 27, wherein one of the first or second monomers further comprises one or more pI variants.

29. The heterodimer antibody of claim 28, wherein the CH1-hinge-CH2-CH3 of the second monomer comprises pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is in accordance with EU numbering.

30. The heterodimer antibody according to any one of claims 23 to 29, wherein the CH1-hinge-CH2-CH3 of the second monomer comprises the amino acid variants E233P / L234V / L235A / G236del / S267K / L368D / K370S / N208D / Q295E / N384D / Q418E / N421D. The first Fc domain contains amino acid variants E233P / L234V / L235A / G236del / S267K / S364K / E357Q. Furthermore, the numbering is based on the EU numbering.

31. The heterodimeric antibody according to any one of claims 24 to 30, wherein the first and second variant Fc domains each further comprise amino acid variants 428L / 434S.

32. The heterodimeric antibody according to any one of claims 17 to 31, wherein the scFv linker is GKPGSGKPGSGKPGSGKPGS.

33. A heterodimer antibody, said heterodimer antibody comprising: a) A first monomer, the first monomer comprising VH1-CH1-hinge-CH2-CH3-structural domain connector-scFv from the N end to the C end; VH1 is the first variable heavy chain domain, and CH2-CH3 is the first Fc domain; b) A second monomer, which comprises VH1-CH1-hinge-CH2-CH3 from the N-end to the C-end, wherein VH1 is a first variable heavy chain domain and CH2-CH3 is a second Fc domain. c) A first light chain, the first light chain comprising VL1-CL from the N-end to the C-end, wherein VL1 is a first variable light chain structural domain and CL is a constant light chain structural domain; and d) A second light chain, which comprises VL1-CL from the N-end to the C-end, wherein VL1 is a first variable light chain structural domain and CL is a constant light chain structural domain. The scFv includes a second VH structural domain (VH2), an scFv connector, and a second variable light chain structural domain (VL2). The first monomer's VH1 and the first light chain's VL1, and the second monomer's VH1 and the second light chain's VL1, each form a first antigen-binding domain (ABD), and the VH2 and the VL2 form a second ABD. The first ABD is an exonucleotide pyrophosphatase / phosphodiesterase 3 (ENPP3) binding domain, and the second ABD is a CD28 binding domain, or the first ABD is a CD28 binding domain and the second ABD is an ENPP3 binding domain.

34. The heterodimer antibody of claim 33, wherein the scFv comprises a VH2-scFv linker-VL2 from the N-terminus to the C-terminus.

35. The heterodimer antibody of claim 33, wherein the scFv comprises a VL2-scFv linker-VH2 from the N-terminus to the C-terminus.

36. The heterodimeric antibody according to any one of claims 33 to 35, wherein the first ABD is the ENPP3 binding domain and the second ABD is the CD28 binding domain.

37. The heterodimer antibody according to claim 36, wherein VH1 and VL1 are VH and VL of any ENPP3 binding domain in Figure 22 or a variant thereof.

38. The heterodimer antibody according to claim 36 or 37, wherein VH2 and VL2 are selected from one of the following: (1) VH and VL or variants thereof of any CD28 binding domain in Figures 15, 18 and 21; or (2) (i) VH or a variant thereof in Figure 15 or Figure 16; and (ii) VL or a variant thereof in Figure 15 or Figure 17.

39. The heterodimeric antibody according to any one of claims 33 to 38, wherein the first Fc domain and the second Fc domain are each variant Fc domains.

40. The heterodimer antibody of claim 39, wherein the first and second Fc domains comprise a set of heterodimerization skewed variants selected from the following heterodimerization variants: S364K / E357Q: L368D / K370S; S364K: L368D / K370S; S364K: L368E / K370S; D401K: T411E / K360E / Q362E; and T366W: T366S / L368A / Y407V, wherein the numbering is based on EU numbering.

41. The heterodimer antibody of claim 40, wherein the first and second Fc domains comprise heterodimerized skew variants S364K / E357Q: L368D / K370S, wherein the designation is based on EU designation.

42. The heterodimeric antibody according to any one of claims 39 to 41, wherein each of the first and second Fc domains comprises one or more ablation variants.

43. The heterodimer antibody of claim 42, wherein the one or more ablation variants comprise E233P / L234V / L235A / G236del / S267K, wherein the numbering is in accordance with EU numbering.

44. The heterodimeric antibody according to any one of claims 39 to 43, wherein one of the first or second monomers further comprises one or more pI variants.

45. The heterodimer antibody of claim 44, wherein the CH1-hinge-CH2-CH3 of the second monomer comprises pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is in accordance with EU numbering.

46. ​​The heterodimer antibody according to any one of claims 39 to 45, wherein the CH1-hinge-CH2-CH3 of the second monomer comprises the amino acid variants E233P / L234V / L235A / G236del / S267K / L368D / K370S / N208D / Q295E / N384D / Q418E / N421D. The first Fc domain contains amino acid variants E233P / L234V / L235A / G236del / S267K / S364K / E357Q. Furthermore, the numbering is based on the EU numbering.

47. The heterodimeric antibody according to any one of claims 40 to 46, wherein the first and second variant Fc domains each further comprise amino acid variants 428L / 434S.

48. The heterodimeric antibody according to any one of claims 33 to 47, wherein the scFv linker is GKPGSGKPGSGKPGSGKPGS.

49. A heterodimer antibody, said heterodimer antibody comprising: a) A first monomer, the first monomer comprising VH1-CH1-structural domain connector-VH1-CH1-hinge-CH2-CH3 from the N end to the C end; Each VH1 is a first variable heavy chain domain, and CH2-CH3 is a first Fc domain; b) A second monomer, which contains scFv-domain connector-CH2-CH3 from the N-end to the C-end, wherein VH1 is a first variable heavy chain domain and CH2-CH3 is a second Fc domain; c) A first light chain, the first light chain comprising VL1-CL from the N-end to the C-end, wherein VL1 is a first variable light chain structural domain and CL is a constant light chain structural domain; and d) A second light chain, which comprises VL1-CL from the N-end to the C-end, wherein VL1 is a first variable light chain structural domain and CL is a constant light chain structural domain. The scFv includes a second VH structural domain (VH2), an scFv connector, and a second variable light chain structural domain (VL2). The first monomer's VH1 and the first light chain's VL1, and the second monomer's VH1 and the second light chain's VL1, each form a first antigen-binding domain (ABD), and the VH2 and the VL2 form a second ABD. The first ABD is an exonucleotide pyrophosphatase / phosphodiesterase 3 (ENPP3) binding domain, and the second ABD is a CD28 binding domain, or the first ABD is a CD28 binding domain and the second ABD is an ENPP3 binding domain.

50. The heterodimer antibody of claim 49, wherein the scFv comprises a VH2-scFv linker-VL2 from the N-terminus to the C-terminus.

51. The heterodimer antibody of claim 49, wherein the scFv comprises a VL2-scFv linker-VH2 from the N-terminus to the C-terminus.

52. The heterodimeric antibody according to any one of claims 49 to 51, wherein the first ABD is the ENPP3 binding domain and the second ABD is the CD28 binding domain.

53. The heterodimer antibody according to claim 52, wherein VH1 and VL1 are VH and VL of any ENPP3 binding domain in Figure 22 or a variant thereof.

54. The heterodimer antibody according to claim 52 or 53, wherein VH2 and VL2 are selected from one of the following: (1) VH and VL or variants thereof of any CD28 binding domain in Figures 15, 18 and 21; or (2) (i) VH or a variant thereof in Figure 15 or Figure 16; and (ii) VL or a variant thereof in Figure 15 or Figure 17.

55. The heterodimeric antibody according to any one of claims 49 to 54, wherein the first Fc domain and the second Fc domain are each variant Fc domains.

56. The heterodimer antibody of claim 55, wherein the first and second Fc domains comprise a set of heterodimerization skewed variants selected from the following heterodimerization variants: S364K / E357Q: L368D / K370S; S364K: L368D / K370S; S364K: L368E / K370S; D401K: T411E / K360E / Q362E; and T366W: T366S / L368A / Y407V, wherein the numbering is based on EU numbering.

57. The heterodimer antibody of claim 56, wherein the first and second Fc domains comprise heterodimerized skew variants S364K / E357Q: L368D / K370S, wherein the designation is based on EU designation.

58. The heterodimeric antibody according to any one of claims 55 to 57, wherein each of the first and second Fc domains comprises one or more ablation variants.

59. The heterodimer antibody of claim 58, wherein the one or more ablation variants comprise E233P / L234V / L235A / G236del / S267K, wherein the numbering is in accordance with EU numbering.

60. The heterodimeric antibody according to any one of claims 55 to 59, wherein one of the first or second monomers further comprises one or more pI variants.

61. The heterodimer antibody of claim 60, wherein the CH1-hinge-CH2-CH3 of the second monomer comprises pI variants N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is in accordance with EU numbering.

62. The heterodimer antibody according to any one of claims 55 to 61, wherein the CH1-hinge-CH2-CH3 of the second monomer comprises amino acid variants E233P / L234V / L235A / G236del / S267K / L368D / K370S / N208D / Q295E / N384D / Q418E / N421D. The first Fc domain contains amino acid variants E233P / L234V / L235A / G236del / S267K / S364K / E357Q. Furthermore, the numbering is based on the EU numbering.

63. The heterodimeric antibody according to any one of claims 56 to 62, wherein the first and second variant Fc domains each further comprise amino acid variants 428L / 434S.

64. The heterodimeric antibody according to any one of claims 49 to 63, wherein the scFv linker is GKPGSGKPGSGKPGSGKPGS.

65. A bispecific antibody, said bispecific antibody comprising: a) A carcinoembryonic antigen-associated cell adhesion molecule 5 (ENPP3) binding domain, said domain comprising i) a first variable heavy chain domain (VH1) and ii) a first variable light chain domain (VL1); and b) An anti-CD28 binding domain comprising i) a second variable heavy chain domain (VH2) and ii) a second variable light chain domain (VL2).

66. The bispecific antibody of claim 65, wherein VH1 and VL1 are VH and VL of any ENPP3 binding domain or a variant thereof in Figure 22.

67. The bispecific antibody according to claim 65 or 66, wherein VH2 and VL2 are selected from one of the following: (1) VH and VL or variants thereof of any CD28 binding domain in Figures 15, 18 and 21; or (2) (i) VH or a variant thereof in Figure 15 or Figure 16; and (ii) VL or a variant thereof in Figure 15 or Figure 17.

68. The bispecific antibody according to any one of claims 65 to 67, wherein the bispecific antibody further comprises a first Fc domain and a second Fc domain.

69. The bispecific antibody of claim 68, wherein the first and second Fc domains comprise a set of heterodimerization skewed variants selected from the following heterodimerization variants: S364K / E357Q: L368D / K370S; S364K: L368D / K370S; S364K: L368E / K370S; D401K: T411E / K360E / Q362E; and T366W: T366S / L368A / Y407V, wherein the numbering is based on EU designations.

70. The bispecific antibody of claim 69, wherein the first and second Fc domains comprise heterodimerized skewed variants S364K / E357Q: L368D / K370S, wherein the numbering is based on EU designations.

71. The bispecific antibody according to any one of claims 68 to 70, wherein each of the first and second Fc domains comprises one or more ablation variants.

72. The bispecific antibody of claim 71, wherein the one or more ablation variants comprise E233P / L234V / L235A / G236del / S267K, wherein the numbering is in accordance with EU numbering.

73. The bispecific antibody according to any one of claims 68 to 72, wherein one of the first or second monomers further comprises one or more pI variants.

74. The bispecific antibody according to claim 73, wherein the pI variant is N208D / Q295E / N384D / Q418E / N421D, wherein the numbering is according to EU numbering.

75. A nucleic acid composition comprising: a) A first nucleic acid, wherein the first nucleic acid encodes a first monomer as described in any one of claims 1 to 64; b) a second nucleic acid, the second nucleic acid encoding a second monomer as described in any one of claims 1 to 64; and c) A third nucleic acid, said third nucleic acid encoding a light chain as described in any one of claims 1 to 64.

76. An expression vector composition, said expression vector composition comprising: respectively, a) A first expression vector, the first expression vector comprising the first nucleic acid as described in claim 75; b) A second expression vector, the second expression vector comprising the second nucleic acid as described in claim 75; and c) A third expression vector comprising the third nucleic acid as described in claim 75.

77. A host cell comprising the expression vector composition of claim 76.

78. A method for preparing a heterodimeric antibody according to any one of claims 1 to 64, the method comprising culturing a host cell as described in claim 77 under conditions in which the heterodimeric antibody is expressed, and recovering the heterodimeric antibody.

79. A method of treating ENPP3-associated cancer in a patient in need, the method comprising administering to the patient a heterodimeric antibody according to any one of claims 1 to 64.

80. A method for treating ENPP3-associated cancer in a patient in need, the method comprising administering to the patient a heterodimeric antibody according to any one of claims 1 to 64; and an anti-CD3 x anti-ENPP3 bispecific antibody.

81. A method of treating ENPP3-associated cancer in a patient in need, the method comprising administering to the patient a bispecific antibody according to any one of claims 65 to 74.

82. A method for treating ENPP3-related cancer in a patient in need, the method comprising administering to the patient a bispecific antibody according to any one of claims 65 to 74; and an anti-CD3 x anti-ENPP3 bispecific antibody.

Images