Multi-specific antigen binding molecules for cell targeting and uses thereof

Abstract

The present invention provides multispecific antigen binding molecules that bind both T cell antigens (e.g., CD3) and target antigens (e.g., tumor associated antigens, viral or bacterial antigens) and that comprise a single polypeptide chain that is multivalent (e.g., divalent) in terms of T cell antigen binding, as well as uses thereof.

Description

[0001] This application is a divisional application of PCT application PCT / US2020 / 046352, filed on August 14, 2020, with the invention name “Multispecific antigen-binding molecules for cell targeting and their uses”. The date on which the PCT application entered the Chinese national phase is March 1, 2022, and the application number is 202080061412.X.

[0002] Reference to a sequence listing

[0003] This application incorporates by reference a sequence listing submitted in computer readable form as file 10606WO01-Sequence.txt, which was created on August 7, 2020 and contains 64,570 bytes. Technical Field

[0004] The present invention relates to alternative forms of multivalent antigen binding proteins and methods of using the same. Multivalent antigen binding proteins, including bispecific and multispecific molecules, include a first polypeptide chain having both N-terminal and C-terminal antigen binding domains that specifically bind to a T cell antigen (e.g., CD3) and a second polypeptide chain including at least one antigen binding domain that binds to a target antigen (e.g., a tumor cell antigen). Background Art

[0005] Bispecific and multispecific antibodies and antigen-binding molecules are known in the art (see, e.g., Brinkmann and Kontermann, MABS, 9(2):182-212, 2017). In such known forms, there is an FcFc* ( Figure 1A structure), a traditional bispecific antibody having a Fab antigen binding domain on either arm of the antibody, and an Fc region having a modified CH3 domain that alters protein A binding affinity to allow separation of heterodimers from homodimer impurities (ibid., p. 184, Figure 2 , Figure 7, last structure). This traditional bispecific antibody format has been used to make bispecific antibodies in which one arm of the antibody targets a tumor cell antigen and the second arm targets a T cell antigen, such as CD3. Another conventional format is IgG-HC-scFv ( Figure 1B structure), a bispecific antibody in which two N-terminal Fab domains bind to a first antigen and two scFv domains connected to the C-terminus of the Fc region bind to a second antigen (ibid., p. 184, Figure 2, Figure 10, first structure). The art needs new and useful forms of bispecific or multispecific antigen-binding molecules that improve desired functionality. Although Brinkmann et al. generally refer to "building blocks" for generating any homodimeric or heterodimeric antigen-binding molecules (page 183, Figure 1), the possibilities are virtually endless, and only a few have been reported to be prepared. Figure 2 (page 184). Furthermore, Brinkmann does not consider specific antigen binding domains, specifically, molecules that include T cell antigen binding domains at both the N-terminus and the C-terminus of a single polypeptide chain that forms part of a multispecific molecule. Summary of the invention

[0006] In general, the invention provides multispecific antigen-binding molecules that bind both a T cell antigen (TCA) (e.g., CD3) and a target antigen (TA) (e.g., a tumor-associated antigen, a viral or bacterial antigen), and which comprise a single polypeptide chain that is multivalent (e.g., bivalent) with respect to T cell antigen binding.

[0007] In one aspect, the present invention provides a multispecific antigen-binding molecule comprising: (a) a first polypeptide, which comprises, from N-terminus to C-terminus, (i) a first antigen-binding domain that specifically binds to a T cell antigen, (ii) a first multimerization domain, and (iii) a second antigen-binding domain that specifically binds to a T cell antigen; and (b) a second polypeptide, which comprises, from N-terminus to C-terminus, (i) a third antigen-binding domain that specifically binds to a target antigen and (ii) a second multimerization domain, wherein the first multimerization domain and the second multimerization domain associate with each other to form a molecule.

[0008] In some embodiments, the second polypeptide further includes a fourth antigen-binding domain at the C-terminus of the second multimerization domain. In some cases, the fourth antigen-binding domain specifically binds to a target antigen. In some cases, the third antigen-binding domain and the fourth antigen-binding domain specifically bind to different target antigens. In some cases, the different target antigens are expressed (or present) on the surface of the same cell. In some cases, the different target antigens are expressed (or present) on the surface of different cells. In this article, the reference to the target antigen expressed (or present) on the cell surface includes proteins embedded or spanning the cell membrane expressed by the cell, and peptides presented by the cell in the groove background of the major histocompatibility complex (MHC) protein. In some cases, the third antigen-binding domain and the fourth antigen-binding domain specifically bind to the same target antigen. In some embodiments, the fourth antigen-binding domain specifically binds to a T cell antigen. In some cases, the first antigen-binding domain and the second antigen-binding domain specifically bind to the same T cell antigen. In some cases, the first antigen-binding domain and the second antigen-binding domain specifically bind to different T cell antigens. In some embodiments, the first antigen binding domain specifically binds to a first T cell antigen as a costimulatory molecule, and the second antigen binding domain specifically binds to a second T cell antigen as a checkpoint inhibitor. In some cases, the costimulatory molecule is CD28, and the checkpoint inhibitor is PD-1. In some cases, the first, second, and fourth antigen binding domains specifically bind to the same T cell antigen. In some cases, the first, second, and fourth antigen binding domains bind to different T cell antigens. In some cases, the first and fourth antigen binding domains specifically bind to the same T cell antigen. In some cases, the second and fourth antigen binding domains specifically bind to the same T cell antigen.

[0009] In various embodiments, one or more of the antigen binding domains are Fab. In various embodiments, one or more of the antigen binding domains are scFv. In some embodiments, the multispecific molecule contains both Fab and scFv antigen binding domains. In some cases, the first antigen binding domain and the third antigen binding domain are Fab. In some cases, the second antigen binding domain is scFv. In some cases, the fourth antigen binding domain is scFv. In some embodiments, the first, second and third antigen binding domains are Fab. In some cases, the first and third antigen binding domains are Fab domains, and the second antigen binding domain is a scFv domain. In some embodiments, the first, second, third and fourth antigen binding domains are Fab. In some cases, the first, second, third and fourth antigen binding domains are Fab domains. In some cases, the first and third antigen binding domains are Fab domains, and the second and fourth antigen binding domains are scFv domains. In some cases, the first, second, third and fourth antigen binding domains are Fab domains. In some cases, the first and third antigen binding domains are Fab domains, and the second and fourth antigen binding domains are scFv domains. In some cases, the first, second, third and fourth antigen binding domains are Fab domains.

[0010] In any embodiment wherein the antigen binding domain is a scFv domain, the scFv domain may include a heavy chain variable region (HCVR) including a cysteine ​​mutation at residue 44 and a light chain variable region (Kabat numbering) including a cysteine ​​mutation at residue 100. In some cases, the scFv includes a HCVR and a LCVR linked together by a 10 to 30 amino acid polypeptide linker, optionally a (G4S)4 linker. In some embodiments, the scFv is linked to the C-terminus of the first and / or second multimerization domains by a 5 to 25 amino acid linker, optionally a (G4S)3 linker.

[0011] In some embodiments, the T cell antigen is a T cell receptor complex antigen (i.e., any protein subunit constituting the T cell receptor complex). In some cases, the T cell antigen is CD3. In some cases, the T cell antigen is a co-stimulatory molecule or a checkpoint inhibitor on a T cell. In some embodiments, the T cell antigen is selected from the group consisting of CD27, CD28, 4-1BB and PD-1. In some embodiments, the T cell antigen is selected from the group consisting of CD3, CD27, CD28, 4-1BB and PD-1.

[0012] In some embodiments, the target antigen is a tumor-associated antigen. In some embodiments, the target antigen is a viral or bacterial antigen. In some embodiments, the target antigen is a fungal antigen or a parasitic antigen.

[0013] In some embodiments, the first and second multimerization domains are immunoglobulin Fc domains. In some cases, the first multimerization domain and the second multimerization domain are human IgG1 or human IgG4 Fc domains. In some cases, the first and second multimerization domains include the immunoglobulin hinge domain, CH2 domain and CH3 domain of human IgG polypeptides (e.g., IgG1, IgG2, IgG3 or IgG4). In some cases, the first and second multimerization domains include the hinge domain, CH2 domain and CH3 domain of human IgG1 polypeptides. In some cases, the first and second multimerization domains include the hinge domain, CH2 domain and CH3 domain of human IgG4 polypeptides. In some embodiments, the first and second multimerization domains are associated with each other by disulfide bonds.

[0014] In some embodiments, the first multimerization domain or the second multimerization domain includes an amino acid substitution that reduces the affinity for protein A binding compared to the wild-type Fc domain of the same isotype. In some cases, the amino acid substitution includes H435R modification, or H435R and Y436F modification (EU numbering). In some cases, the first multimerization domain includes the H435R and Y436F modifications. In some cases, the second multimerization domain includes the H435R and Y436F modifications. In some embodiments, the first polypeptide, the second polypeptide, or the first polypeptide and the second polypeptide both include a modified hinge domain, and the modified hinge domain reduces the binding affinity to Fcγ receptors compared to the wild-type hinge domain of the same isotype.

[0015] On the other hand, the present invention provides a multispecific antigen-binding molecule, comprising: (a) a first polypeptide, which comprises, from N-terminus to C-terminus, (i) a first Fab that specifically binds to a T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a first scFv that specifically binds to a T cell antigen; and (b) a second polypeptide, which comprises, from N-terminus to C-terminus, (i) a second Fab that specifically binds to a target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a second scFv that specifically binds to a target antigen, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule.

[0016] In some embodiments, the second Fab and the second scFv specifically bind to different target antigens. In some cases, the different target antigens are expressed on the surface of the same cell. In some embodiments, the second Fab and the second scFv specifically bind to the same target antigen.

[0017] On the other hand, the present invention provides a multispecific antigen-binding molecule, comprising: (a) a first polypeptide, which comprises, from N-terminus to C-terminus, (i) a first Fab that specifically binds to a T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a second Fab that specifically binds to a T cell antigen; and (b) a second polypeptide, which comprises, from N-terminus to C-terminus, (i) a third Fab that specifically binds to a target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a fourth Fab that specifically binds to a target antigen, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule.

[0018] In some embodiments, the third Fab and the fourth Fab specifically bind to different target antigens. In some cases, the different target antigens are expressed on the surface of the same cell. In some embodiments, the third Fab and the fourth Fab specifically bind to the same target antigen.

[0019] On the other hand, the present invention provides a multispecific antigen-binding molecule, comprising: (a) a first polypeptide, which comprises, from N-terminus to C-terminus, (i) a first Fab that specifically binds to a T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a first scFv that specifically binds to a T cell antigen; and (b) a second polypeptide, which comprises, from N-terminus to C-terminus, (i) a second Fab that specifically binds to a target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a second scFv that specifically binds to a T cell antigen, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule.

[0020] On the other hand, the present invention provides a multispecific antigen-binding molecule, comprising: (a) a first polypeptide, which comprises, from N-terminus to C-terminus, (i) a first Fab that specifically binds to a T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a second Fab that specifically binds to a T cell antigen; and (b) a second polypeptide, which comprises, from N-terminus to C-terminus, (i) a second Fab that specifically binds to a target antigen and (ii) a second immunoglobulin Fc domain, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule.

[0021] In various embodiments, as any one of the embodiments mentioned above or herein, the T cell antigen is a T cell receptor complex antigen (that is, any protein subunit constituting the T cell receptor complex). In some cases, the T cell antigen is CD3. In some cases, the T cell antigen is a co-stimulatory molecule or a checkpoint inhibitor on a T cell. In some embodiments, the T cell antigen is selected from the group consisting of CD27, CD28, 4-1BB and PD-1. In some embodiments, the T cell antigen is selected from the group consisting of CD3, CD27, CD28, 4-1BB and PD-1.

[0022] In various embodiments, such as any of those embodiments mentioned above or herein, the target antigen is a tumor-associated antigen. In some embodiments, the target antigen is a viral or bacterial antigen. In some embodiments, the target antigen is a fungal antigen or a parasitic antigen.

[0023] In some embodiments, as any one of the embodiments mentioned above or herein, the first and second multimerization domains are immunoglobulin Fc domains. In some cases, the first multimerization domain and the second multimerization domain are human IgG1 or human IgG4 Fc domains. In some cases, the first and second multimerization domains include the immunoglobulin hinge domain, CH2 domain and CH3 domain of human IgG polypeptides (e.g., IgG1, IgG2, IgG3 or IgG4). In some cases, the first and second multimerization domains include the hinge domain, CH2 domain and CH3 domain of human IgG1 polypeptides. In some cases, the first and second multimerization domains include the hinge domain, CH2 domain and CH3 domain of human IgG4 polypeptides. In some embodiments, the first and second multimerization domains are associated with each other by disulfide bonds.

[0024] In some embodiments, as any one of the embodiments mentioned above or herein, the first multimerization domain or the second multimerization domain includes an amino acid substitution that reduces the affinity for protein A binding compared to the wild-type Fc domain of the same isotype. In some cases, the amino acid substitution includes H435R modification, or H435R and Y436F modification (EU numbering). In some cases, the first multimerization domain includes the H435R and Y436F modifications. In some cases, the second multimerization domain includes the H435R and Y436F modifications. In some embodiments, the first polypeptide, the second polypeptide, or the first polypeptide and the second polypeptide both include a modified hinge domain, and the modified hinge domain reduces the binding affinity to Fcγ receptors compared to the wild-type hinge domain of the same isotype.

[0025] In another aspect, the present invention provides a pharmaceutical composition comprising any of the multispecific molecules discussed above or herein, and a pharmaceutically acceptable carrier or diluent.

[0026] In another aspect, the invention provides a method of treating cancer comprising administering any one of the multispecific molecules discussed above or herein to a subject in need thereof.

[0027] In another aspect, the invention provides a method of treating an infection, the method comprising administering any of the multispecific molecules discussed above or herein to a subject in need thereof. In some cases, the infection is a bacterial infection. In some cases, the infection is a viral infection. In some cases, the infection is a fungal infection. In some cases, the infection is a parasitic infection.

[0028] In various embodiments, the target antigen is present at a density of 10 to 10,000,000 copies per target cell. In various embodiments, the target antigen is present at a density of 100 to 10,000,000 copies per target cell. In various embodiments, the target antigen is present at a density of 100 to 1,000,000 copies per target cell. In some embodiments, the target antigen is present at a density of 50 to 10,000. In some embodiments, the target antigen is present at a density of 100 to 5000. In some embodiments, the target antigen is present at a density of 100 to 20,000. In some embodiments, the target antigen is present at a density of 500 to 1,000,000 copies per target cell. In some embodiments, the target antigen is present at a density of 1000 to 20,000 copies per target cell. In some embodiments, the target antigen is present at a density of greater than 20,000 copies per target cell. In some embodiments, the target antigen is present at about 10, about 50, about 100, about 200, about 300, about 400, about 500, about 1000, about 2000, about 3000, about 4000, about 5000, about 6000, about 7000, about 8000, about 9000, about 10,000, about 15,000, about 20,000, about 25,000, about 50,000 per target cell. , about 75,000, about 100,000, about 200,000, about 300,000, about 400,000, about 500,000, about 600,000, about 700,000, about 800,000, about 900,000, about 1,000,000, about 2,000,000, about 3,000,000, about 4,000,000 or about 5,000,000 copies. As used herein, "low density antigen" is an antigen found to be no more than 5000 copies on a target cell. Reference to a low density of antigen includes situations in which the cells have no more than 4000, no more than 3000, no more than 2000, no more than 1000, no more than 900, no more than 800, no more than 700, no more than 600, no more than 500, no more than 400, no more than 300, no more than 200, no more than 100, or no more than 50 copies of the target antigen.

[0029] In various embodiments, the multispecific molecule is administered in combination with a second therapeutic agent to treat a disease or condition. In some cases, the second therapeutic agent includes a bispecific antigen binding molecule comprising a first antigen binding domain that binds a target antigen (TA) and a second antigen binding domain that binds a T cell antigen. In some cases, the target antigen is a tumor cell antigen. In some embodiments, the second therapeutic agent includes a bispecific anti-TA x anti-CD28 antibody. In some embodiments, the second therapeutic agent includes a bispecific anti-EGFR x anti-CD28 antibody. In some embodiments, the second therapeutic agent includes an antibody that binds a checkpoint inhibitor on a T cell. In some embodiments, the second therapeutic agent includes an anti-PD-1 antibody. In some cases, the multispecific molecule is administered in combination with two or more second therapeutic agents.

[0030] In another aspect, the invention provides a use of any one of the multispecific molecules discussed above or herein in the preparation of a medicament for treating a disease or disorder (eg, cancer or infection) in a subject in need thereof.

[0031] In another aspect, the invention provides for use of any of the multispecific molecules discussed above or herein in medicine or in treating a disease or disorder (eg, cancer or infection).

[0032] In another aspect, the invention provides a multispecific molecule as discussed above or herein for use in medicine or in the treatment of a disease or disorder (eg, cancer or infection).

[0033] In any of the embodiments discussed above or herein, the target antigen can be a peptide in the context of a groove of a major histocompatibility complex (MHC) protein.

[0034] In various embodiments, any of the features or components of the embodiments discussed above or herein can be combined, and such combinations are encompassed within the scope of the present disclosure. Any specific value discussed above or herein can be combined with another related value discussed above or herein to enumerate a range of values ​​having an upper and lower end representing a range, and such ranges are encompassed within the scope of the present disclosure.

[0035] Other embodiments will become apparent upon reading the ensuing detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1A and 1B Known bispecific antibody and antigen-binding molecule formats are shown.

[0037] Figure 1C , 1E, 1F, 1G, 1H, 1I, 1J, 1K, 1L, 1M, 1N, 1O, 1P, 1Q, 1R and 1S show bispecific or multispecific antigen binding molecule formats according to embodiments of the present invention. In each of these formats, the first polypeptide chain includes an N-terminal and C-terminal antigen binding domain (e.g., Fab or scFv) that specifically binds to a T cell antigen (TCA) (e.g., CD3), and the second polypeptide chain includes at least one antigen binding domain (e.g., Fab or scFv) that binds to a target antigen (TA) (e.g., a tumor cell antigen). Figure 1D A format is presented in which two antigen binding domains that specifically bind a T cell antigen (eg, CD3) are located on different polypeptide chains (at the N-terminus on one polypeptide chain, and at the C-terminus on a second polypeptide chain).

[0038] Figure 2 Compared with T cell only control (zero) and positive control, the Figure 1A , 1B T cell activation induced by each form of the molecule shown in Figure 1C. In the absence of target cells, none of these molecules activated T cells.

[0039] Figure 3 Shown is a positive control that induced maximal cell killing in the presence of human PBMCs and target cells (A375). Figure 1A , 1B The cytotoxic activity of each form of the molecule shown in 1C. The CD3 binding domain of the molecule includes the variable region of the 7221G anti-CD3 antibody. Figure 1C The molecules with Figure 1A and 1B The molecules of the structure are more effective.

[0040] Figure 4A , 4B and 4C show the maximal cell killing induced by the positive control in the presence of human PBMCs and target cells (A375) in combination with an anti-PD-1 antibody ( Figure 4A ), co-stimulatory bispecific EGFR x CD28 antibody ( Figure 4B ) or anti-PD-1 antibody and co-stimulatory bispecific EGFR x CD28 antibody ( Figure 4C ) combination, with Figure 1A , 1B The cytotoxic activity of each form of the molecule shown in 1C. The CD3 binding domain of the molecule includes the variable region of the 7221G anti-CD3 antibody. Figure 1C The structure of the molecule combined with these additional antibodies is significantly higher than that with Figure 1A and 1BThe molecules of the structure are more effective.

[0041] Figure 5 Shown with Figure 1A (Left) Compared with the structure of the molecule, Figure 4A , 4B The maximum antibody concentrations shown in 4C were measured Figure 1C (Right) Cytokine levels of the structured molecule. The CD3 binding domain of the molecule includes the variable region of the 7221G anti-CD3 antibody. Although it has significantly greater cytotoxic activity, it has Figure 1C The structured molecules did not show higher levels of cytokine release.

[0042] Fig. 6A , 6B , 6C and 6D show the Figure 1C The structured molecule and a modified version of the molecule (with an inactive domain – marked with an X in the figure legend) were compared with Raji cells overexpressing the MAGEA4 peptide ( Fig. 6A ) or A375 cells ( Figure 6C ) or CD3+Jurkat cells ( Figure 6B and 6D ) combination. Fig. 6A and 6B The CD3 binding domain of the molecule shown in FIG. 5 comprises the variable region of the 7195P anti-CD3 antibody. Figure 6C and 6D The CD3 binding domain of the molecule shown in FIG. 1 includes the variable region of the 7221G anti-CD3 antibody. Fig. 6A , 6B , 6C and 6D, the presence of two active antigen binding domains improves binding to the target antigen, and similar binding is observed regardless of the source of the anti-CD3 binding domain. As shown in these figures, binding is most affected when the N-terminal Fab domain is removed.

[0043] Fig. 7A and 7B Shows Fig. 6A and 6B ( Fig. 7A )as well as Figure 6C and 6D ( Figure 7B ) has the same cytotoxic activity as shown in Figure 1C The molecule with the structure of α-CD3 showed the greatest cytotoxic potency, followed by the molecule with two active T cell antigen (e.g., CD3) binding domains. A similar cytotoxic pattern was observed regardless of the source of the anti-CD3 binding domain.

[0044] Fig. 8A and 8BShows that there is Figure 1C The structured molecule and modified versions of the molecule (with a C-terminal Fab domain or an inactive domain – indicated by an X in the figure legend) were compared with Raji cells overexpressing the MAGEA4 peptide ( Fig. 8A ) or CD3+Jurkat cells ( Figure 8B The CD3 binding domain of the molecule includes the variable region of the 7195P anti-CD3 antibody. Fig. 8A and 8B As shown, the C-terminal scFv domain provides better binding to the target antigen compared to the C-terminal Fab domain.

[0045] Fig. 9 Shows Fig. 8A and 8B The cytotoxic activity of the same molecule shown in . The CD3 binding domain of the molecule includes the variable region of the 7195P anti-CD3 antibody. Figure 1C The molecules with Figure 1E Structure of molecules.

[0046] Fig. 10A and 10B Shows that there is Figure 1C and 1D The structure of the molecules and A375 cells overexpressing MAGEA4 peptide ( Fig. 10A ) or CD3+Jurkat cells ( Fig. 10B The CD3 binding domain of the molecule included the variable region of the 7221G anti-CD3 antibody. The two molecules showed similar binding to both cell types relative to each other.

[0047] Fig.11A and 11B Shows Fig. 10A and 10B The same molecule shown in Figure 2 shows cytotoxic activity against A375 cells from two different donors. The CD3 binding domain of the molecule includes the variable region of the 7221G anti-CD3 antibody. Figure 1C The molecular structure has Figure 1D The molecules of the structure are more effective.

[0048] Fig. 12A and 12B They are shown with Figure 1A Compared with the molecules with Figure 1C and Figure 1FThe relative cytotoxic activity and potency of the molecules with the structures described above were determined. As discussed in Example 7, these molecules were tested alone and in combination with a co-stimulatory bispecific EGFR x CD28 antibody and an anti-PD-1 antibody. The CD3 binding domain of the molecule included the variable region of the 7195P anti-CD3 antibody. Figure 1F The molecules of the structure target two different epitopes of the same target antigen with two TA antigen binding domains, while Figure 1C The structure of the molecule targets the same epitope of the target antigen with two TA antigen binding domains. Figure 1F The molecular structure has Figure 1C The molecules with the structure are more effective, and both molecules are better than those with Figure 1A In each case, the combination of these molecules with the co-stimulatory bispecific antibody and the anti-PD-1 antibody resulted in even greater cytotoxic potency, similar to Figures 4A-4C The results shown in .

[0049] Fig.13 Shows the Figure 1F Relative binding affinity of molecules of structure, wherein the CD3 binding domain is derived from an anti-CD3 antibody with strong, medium or weak binding affinity to CD3. The "strong" binding domain is derived from the 7195P anti-CD3 antibody. The "medium" binding domain is derived from the 7221G anti-CD3 antibody. The "weak" binding domain is derived from the 7221G20 anti-CD3 antibody. References to, for example, "strong / strong" refer to Fab anti-CD3 binding domains and scFc anti-CD3 binding domains, respectively. As expected, binding to CD3-positive Jurkat cells is related to the affinity strength of the anti-CD3 binding domain in the molecule.

[0050] Fig.14A and 14B Shows Fig.13 Relative cytotoxic activity and potency of the indicated molecules in MAGEA4 positive A375 cells. As discussed in Example 8, these molecules were tested individually ( Fig.14A ), and tested in combination with co-stimulatory bispecific EGFRxCD28 and anti-PD-1 antibodies ( Fig. 14B ). There is a clear correlation between the strength of the anti-CD3 binding domain and the potency of the molecule. "Control" is a positive control that targets a scaffold of all HLA molecules to provide maximum cytotoxicity compared to the other molecules.

[0051] Fig.15A and 15B Shows Fig.13 Relative cytotoxic activity and potency of the indicated molecules in MAGEA4-positive ScaBER cells. As discussed in Example 8, these molecules were tested individually ( Fig.15A ), and tested in combination with co-stimulatory bispecific EGFRxCD28 and anti-PD-1 antibodies ( Fig. 15B ). There is a clear correlation between the strength of the anti-CD3 binding domain and the potency of the molecule. "Control" is a positive control that targets a scaffold of all HLA molecules to provide maximum cytotoxicity compared to the other molecules.

[0052] Fig.16A , 16B and 16C shows a Figure 1A The molecules with structures of 1C (molecule C), 1C (molecule B) and IF (molecules A and D) have an effect on NYESO-1 positive cells ( Fig.16A )、MAGEA4(peptide 1) positive cells( Fig. 16B ) and MAGEA4 (peptide 2) positive cells ( Fig. 16C ). As expected, molecule D without the NYESO-1 binding domain did not bind to NYESO-1 expressing cells ( Fig.16A ), while molecules lacking the relevant MAGEA4 binding domain do not bind to MAGEA4-expressing cells, e.g. Fig. 16B and 16C The CD3 binding domain of the molecule includes the variable region of the 7195P anti-CD3 antibody. The "HLA-targeted bispecific" positive control binds to both HLA molecules and CD3. The "isotype control multispecific" is a Figure 1C A structured molecule having a binding domain to an unrelated target antigen.

[0053] Fig.17A and 17B The binding sites of HLA molecules and CD3 are shown respectively. Figure 1A Compared with the positive control of the structure Figure 1C and Figure 1F The relative cytotoxic activity and potency of molecules with Figure 1C A molecule having a binding domain to an unrelated target antigen and having Figure 1A Structured molecules having binding domains to CD3 and an unrelated target antigen. As discussed in Example 9, these molecules were tested alone and in combination with a co-stimulatory bispecific EGFR x CD28 antibody and an anti-PD-1 antibody. The CD3 binding domain of the molecule includes the variable region of the 7195P anti-CD3 antibody. Figure 1F The molecule with the structure targets two different antigens (NYESO-1 and MAGEA4) with two TA antigen-binding domains, while Figure 1C The structure of the molecule targets a single antigen with two TA antigen-binding domains. Figure 1FMolecules that target two different antigens are more likely to have Figure 1C In each case, the combination of these molecules with the co-stimulatory bispecific antibody and the anti-PD-1 antibody produced even greater cytotoxic potency than the individual molecules alone, similar to Figures 4A-4C The results shown in .

[0054] Fig. 17C and 17D Shows the combination Fig.17A and 17B Relative T cell activation of the molecules discussed.

[0055] Fig.18A and 18B They are shown with Figure 1A Compared with the molecules with Figure 1C and Figure 1F The relative cytotoxic activity and potency of molecules with different structures. Figure 1A The isotype control contains a molecule with a structure that binds to human leukocyte antigen (HLA) molecules and CD3. Figure 1C A molecule having a binding domain to an unrelated target antigen and having Figure 1A Structured molecules having binding domains to CD3 and an unrelated target antigen. As discussed in Example 9, these molecules were tested alone and in combination with a co-stimulatory bispecific EGFR x CD28 antibody and an anti-PD-1 antibody. The CD3 binding domain of the molecule included the variable region of the 7195P anti-CD3 antibody. Fig.18A As shown, Figure 1F Molecules with Figure 1C The molecule with the structure of TA-binding domain (targeting a single epitope with two TA-binding domains) was more effective, and both molecules were more effective than those with Figure 1A The molecules of the structure are more effective. Similarly, Fig.18B As shown, Figure 1F Molecules with Figure 1C The molecule with the structure of TA-binding domain (targeting a single antigen with two TA-binding domains) was more effective, and both molecules were more effective than those with Figure 1A In each case, the combination of these molecules with the co-stimulatory bispecific antibody and the anti-PD-1 antibody produced even greater cytotoxic potency than the individual molecules alone, similar to Figures 4A-4C The results shown in .

[0056] Fig.18C , 18D , 18E and 18F show the combination Fig.18A and 18BRelative T cell activation of the molecules discussed.

[0057] Fig.19A and 19B They are shown respectively with Figure 1F The structure of the molecule is relative to that of Figure 1A Cytotoxic activity and potency and T cell activation of a combination of two molecules having a structure Figure 1F The molecules of the same structure bind to the same target antigen pair. Fig.19A and 19B As shown, Figure 1F The molecular structure has Figure 1A The combination of the two molecules in the structure more effectively kills tumor cells and increases T cell activation. DETAILED DESCRIPTION

[0058] Before describing the present invention in more detail, it should be understood that the present invention is not limited to the specific methods and experimental conditions described, because such methods and conditions can vary. It should also be understood that the terms used herein are only for the purpose of describing specific embodiments, and are not intended to be restrictive, because the scope of the present invention is limited only by the appended claims.

[0059] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those of ordinary skill in the art to which the invention belongs. As used herein, when used to refer to a specific listed value, the term "about" means that the numerical value may differ from the listed value by no more than 1%. For example, as used herein, the expression "about 100" includes 99 and 101 and all values ​​therebetween (e.g., 99.1, 99.2, 99.3, 99.4, etc.).

[0060] Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, the preferred methods and materials are now described. All patents, applications and non-patent publications mentioned in this specification are incorporated herein by reference in their entirety.

[0061] definition

[0062] The term "T cell" refers to immune cells expressing CD3, including CD4+ cells (helper T cells), CD8+ cells (cytotoxic T cells), regulatory T cells (Tregs), and tumor-infiltrating lymphocytes.

[0063] The expression "T cell antigen" refers to a protein expressed on the cell surface present on a T cell, and includes a "co-stimulatory molecule". A "co-stimulatory molecule" refers to a protein expressed by a T cell that binds to a cognate ligand or receptor (e.g., on an antigen presenting cell) to provide a stimulation signal that, in combination with the primary signal provided by the engagement of the TCR of the T cell with the peptide / MHC, can stimulate the activity of the T cell. The stimulation of T cells may include activation, proliferation and / or survival of the T cell.

[0064] As used herein, the expression "cell surface expressed" or "cell surface molecule" means one or more proteins that are expressed on the surface of a cell in vitro or in vivo such that at least a portion of the protein is exposed on the extracellular side of the cell membrane and is accessible to the antigen-binding portion of an antibody or the antigen-binding domain of a multispecific antigen-binding molecule discussed herein.

[0065] The expression "CD3" used herein refers to an antigen expressed on T cells as part of a multimolecular T cell receptor (TCR), and is composed of homodimers or heterodimers formed by the association of two of the following four receptor chains: CD3-ε, CD3-δ, CD3-ζ and CD3-γ. All references to proteins, polypeptides and protein fragments herein are intended to refer to the human version of the corresponding protein, polypeptide or protein fragment, unless it is explicitly stated that it is from a non-human species. Therefore, the expression "CD3" means human CD3, unless specified as being from a non-human species, such as "mouse CD3", "monkey CD3" and the like.

[0066] As used herein, "antibodies that bind to CD3" or "anti-CD3 antibodies" include antibodies and antigen-binding fragments thereof that specifically recognize a single CD3 subunit (e.g., ε, δ, γ or ζ), as well as antibodies and antigen-binding fragments thereof that specifically recognize a dimeric complex of two CD3 subunits (e.g., γ / ε, δ / ε and ζ / ζ CD3 dimers). The antigen-binding domains of the present invention may bind to soluble CD3 and / or CD3 expressed on the cell surface. Soluble CD3 includes native CD3 protein and recombinant CD3 protein variants such as monomeric and dimeric CD3 constructs that lack a transmembrane domain or are otherwise unrelated to the cell membrane.

[0067] As used herein, the expression "CD3 expressed on the cell surface" refers to one or more CD3 proteins that are expressed on the cell surface in vitro or in vivo so that at least a portion of the CD3 protein is exposed to the extracellular side of the cell membrane and can be approached by the antigen-binding portion of the antibody. "CD3 expressed on the cell surface" includes CD3 proteins contained in the context of a functional T cell receptor in the membrane of the cell. The expression "CD3 expressed on the cell surface" includes CD3 proteins (e.g., γ / ε, δ / ε and ζ / ζ CD3 dimers) expressed as a part of a homodimer or heterodimer on the cell surface. The expression "CD3 expressed on the cell surface" also includes CD3 chains (e.g., CD3-ε, CD3-δ or CD3-γ) that are expressed on the cell surface by themselves without other CD3 chain types. "CD3 expressed on the cell surface" may include or consist of CD3 proteins expressed on the surface of cells that normally express CD3 proteins. Alternatively, "cell surface expressed CD3" may comprise or consist of CD3 protein expressed on the surface of cells that normally do not express human CD3 on their surface but have been artificially engineered to express CD3 on their surface.

[0068] The term "antigen binding domain" refers to that portion of a multispecific molecule or corresponding antibody that specifically binds to a predetermined antigen (e.g., CD3 or a tumor-associated antigen). Reference to a "corresponding antibody" refers to the antibody from which the CDRs or variable regions (HCVR and LCVR) used in the multispecific molecule are derived. For example, the Figure 1C The molecules of the structure comprise Fab and scFv with variable regions derived from specific anti-CD3 antibodies and anti-MAGEA4 antibodies. These antibodies are the "counterpart antibodies" of the corresponding multispecific molecules.

[0069] The term "multispecific antigen-binding molecule" includes molecules that bind two or more (e.g., three or four) different epitopes or antigens. In some cases, the multispecific antigen-binding molecule is bispecific. In some cases, the multispecific antigen-binding molecule is trispecific. In some cases, the multispecific antigen-binding molecule is tetraspecific.

[0070] The term "antibody" means any antigen binding molecule or molecular complex comprising at least one complementary determining region (CDR) that specifically binds or interacts with a specific antigen (e.g., CD3 or target antigen (TA)). The term "antibody" includes immunoglobulin molecules comprising four polypeptide chains, two heavy (H) chains and two light (L) chains interconnected by disulfide bonds, and multimers thereof (e.g., IgM). The term "antibody" also includes immunoglobulin molecules consisting of four polypeptide chains, two heavy (H) chains and two light (L) chains interconnected by disulfide bonds. Each heavy chain includes a heavy chain variable region (abbreviated herein as HCVR or VH ) and the heavy chain constant region. The heavy chain constant region consists of three domains C H 1. C H 2 and C H 3. Each light chain includes a light chain variable region (abbreviated herein as LCVR or V L ) and the light chain constant region. The light chain constant region consists of a domain (C L 1) You can set V H Area and V L The V region is further subdivided into regions of hypervariability, called complementarity determining regions (CDRs), interspersed with regions that are more conserved, called framework regions (FRs). H and V L It is composed of three CDRs and four FRs, arranged from the amino terminus to the carboxyl terminus in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. In different embodiments of the present invention, the FRs of the anti-TA antibody or anti-CD3 antibody (or its antigen-binding portion) may be identical to human germline sequences, or may be natural or artificially modified. An amino acid consensus sequence may be defined based on a side-by-side analysis of two or more CDRs.

[0071] As used herein, the term "antibody" also includes the antigen binding fragment of a complete antibody molecule. As used herein, the term "antigen binding portion thereof", "antigen binding fragment" of an antibody, etc., include any naturally occurring, enzymatically obtainable, synthetic or genetically engineered polypeptide or glycoprotein, which specifically binds to an antigen to form a complex. The antigen binding fragment of an antibody can be derived from a complete antibody molecule, for example, using any suitable standard technique, such as proteolytic digestion or involving manipulation and expression of DNA encoding variable and optionally constant domains of an antibody. Such DNA is known and / or is easy to obtain from, for example, commercial sources, DNA libraries (comprising, for example, phage-antibody libraries), or can be synthesized. DNA can be sequenced and manipulated by chemical methods or by molecular biological techniques, for example, one or more variable domains and / or constant domains are arranged into a suitable configuration, or codons are introduced, cysteine ​​residues are produced, modified, added or deleted amino acids, etc.

[0072] Non-limiting examples of antigen-binding fragments include: (i) Fab fragments; (ii) F(ab')2 fragments; (iii) Fd fragments; (iv) Fv fragments; (v) single-chain Fv (scFv) molecules; (vi) dAb fragments; and (vii) minimal recognition units consisting of amino acid residues that mimic the hypervariable region of an antibody (e.g., isolated complementary determining regions (CDRs), such as CDR3 peptides) or constrained FR3-CDR3-FR4 peptides. Other engineered molecules, such as domain-specific antibodies, single domain antibodies, domain-deleted antibodies, chimeric antibodies, CDR-grafted antibodies, diabodies, triabodies, tetrabodies, minibodies, nanobodies (e.g., monovalent nanobodies, bivalent nanobodies, etc.), small modular immunopharmaceuticals (SMIPs) and shark variant IgNAR domains are also encompassed within the expression "antigen-binding fragment" as used herein.

[0073] The antigen-binding fragment of an antibody generally includes at least one variable domain. The variable domain can be of any size or amino acid composition and will generally include at least one CDR adjacent to or within one or more framework sequences. L Domain-associated V H In the antigen-binding fragment of the structural domain, V H Domain and V L The domains may be positioned relative to each other in any suitable arrangement. For example, the variable region may be a dimer and contain V H -V H 、V H -V L or V L Alternatively, the antigen-binding fragment of the antibody may contain a monomer V H or V L Structural domain.

[0074] In certain embodiments, the antigen-binding fragment of an antibody may contain at least one variable domain covalently linked to at least one constant domain. Non-limiting exemplary configurations of variable and constant domains that may be found within the antigen-binding fragment of an antibody of the invention include: (i) V H -C H 1; (ii) V H -C H 2; (iii) V H -C H 3; (iv) V H -C H 1-C H 2; (v) V H -C H 1-C H 2-C H 3; (vi) VH -C H 2-C H 3; (vii) V H -C L ; (viii) V L -C H 1; (ix) V L -C H 2; (x) V L -C H 3; (xi) V L -C H 1-C H 2; (xii) V L -C H 1-C H 2-C H 3; (xiii) V L -C H 2-C H 3; and (xiv) V L -C L . In any configuration of variable domains and constant domains (including any of the exemplary configurations listed above), the variable domains and constant domains may be directly connected to each other or may be connected by a complete or partial hinge or linker region. The hinge region may be composed of at least 2 (e.g., 5, 10, 15, 20, 40, 60 or more) amino acids that result in a flexible or semi-flexible connection between adjacent variable domains and / or constant domains in a single polypeptide molecule. In addition, the antigen-binding fragment of the antibody of the present invention may include homodimers or heterodimers (or other multimers) of any of the variable domain and constant domain configurations listed above, which are non-covalently associated with each other and / or with one or more monomer V H or V L The domains are covalently associated (eg, through one or more disulfide bonds).

[0075] In certain embodiments of the present invention, the antibody is a human antibody. The term "human antibody" is intended to include antibodies with variable regions and constant regions derived from human germline immunoglobulin sequences. The human antibody may include amino acid residues that are not encoded by human germline immunoglobulin sequences (e.g., mutations introduced by random or site-specific mutagenesis in vitro or by somatic mutations in vivo), such as in CDR regions, and specifically, in CDR3 regions. However, as used herein, the term "human antibody" is not intended to include following antibodies, in which the CDR sequences derived from the germline of another mammalian species (e.g., mice) have been transplanted onto human framework sequences.

[0076] In some embodiments, the antibodies discussed herein can be recombinant human antibodies. The term "recombinant human antibody" is intended to include all human antibodies prepared, expressed, produced or separated by a recombinant manner, such as antibodies expressed by recombinant expression vectors transfected into host cells, antibodies separated from recombinant combined human antibody libraries, antibodies separated from human immunoglobulin gene transgenic animals (e.g., mice) (see, for example, Taylor et al., (1992), "Nucl. Acids Res.", 20: 6287-6295) or antibodies prepared, expressed, produced or separated by any other means, the other means involving splicing human immunoglobulin gene sequences to other DNA sequences. Such recombinant human antibodies have variable regions and constant regions derived from human germline immunoglobulin sequences. However, in certain embodiments, such recombinant human antibodies undergo in vitro mutagenesis (or, when using animals of transgenic human Ig sequences, undergo in vivo somatic mutagenesis), and therefore the V H Area and V L The amino acid sequence of the region is as follows: H Sequence and V L Sequences are related to but may not naturally occur in vivo in the human antibody germline repertoire.

[0077] The antibodies mentioned herein can be isolated antibodies. As used herein, "isolated antibodies" means antibodies that have been identified and separated and / or recovered from at least one component of its natural environment. For example, antibodies that have been separated or removed from at least one component of an organism, or from tissues or cells that naturally exist or naturally produce antibodies are "isolated antibodies". Isolated antibodies also include in situ antibodies in recombinant cells. Isolated antibodies are antibodies that have undergone at least one purification or separation step. Isolated antibodies can be substantially free of other cellular materials and / or chemicals.

[0078] The antibodies mentioned herein include one or more amino acid substitutions, insertions and / or deletions in the framework and / or CDR regions of the heavy and light chain variable domains compared to the corresponding germline sequences from which the antibodies were derived. Such mutations can be readily determined by comparing the amino acid sequences disclosed herein with germline sequences available from, for example, public antibody sequence databases.

[0079] The term "epitope" refers to an antigenic determinant that interacts with a specific antigen binding site in the variable region of an antibody molecule called a paratope. A single antigen can have more than one epitope. Therefore, different antibodies can bind to different regions on the antigen and can have different biological effects. Epitopes can be conformational or linear. Conformational epitopes are produced by spatially juxtaposed amino acids from different segments of a linear polypeptide chain. A linear epitope is an epitope produced by adjacent amino acid residues in a polypeptide chain. In some cases, an epitope may comprise a portion of a sugar, phosphoryl, or sulfonyl group on the antigen.

[0080] A "multimerization domain" or "multimerizing domain" is any macromolecule that has the ability to associate (covalently or non-covalently) with a second macromolecule of the same or similar structure or composition. For example, a multimerization domain may be a domain that includes an immunoglobulin C H The polypeptide of 3 domains. A non-limiting example of a multimerization domain is the Fc portion of an immunoglobulin, for example, an Fc domain of an IgG selected from isotypes IgG1, IgG2, IgG3 and IgG4 and any allotype within each isotype group. In certain embodiments, the multimerization domain is an Fc fragment or an amino acid sequence having a length of 1 to about 200 amino acids containing at least one cysteine ​​residue. In other embodiments, the multimerization domain is a cysteine ​​residue or a short peptide containing cysteine. Other multimerization domains include peptides or polypeptides including or consisting of a leucine zipper, a spiral loop motif or a coiled coil motif. In some embodiments, the multimerization domain is an immunoglobulin Fc domain, and two such Fc domains associate through an interchain disulfide bond to form a multispecific antigen-binding molecule of the present invention, such as in a conventional antibody.

[0081] The term "nucleic acid" or "polynucleotide" refers to nucleotides and / or polynucleotides, such as deoxyribonucleic acid (DNA) or ribonucleic acid (RNA), oligonucleotides, fragments generated by polymerase chain reaction (PCR), and fragments generated by any of ligation, cleavage, endonuclease action and exonuclease action. Nucleic acid molecules can be composed of monomers that are naturally occurring nucleotides (such as DNA and RNA) or analogs of naturally occurring nucleotides (e.g., enantiomeric forms of naturally occurring nucleotides) or a combination of the two. Modified nucleotides can have changes in the sugar moiety and / or in the pyrimidine or purine base moiety. Sugar modifications include, for example, replacement of one or more hydroxyls with halogens, alkyls, amines and azidos, or sugars can be functionalized as ethers or esters. In addition, the entire sugar moiety can be replaced by spatial and electronically similar structures (such as azasugars and carbocyclic sugar analogs). Examples of modifications in the base moiety include alkylated purines and pyrimidines, acylated purines or pyrimidines or other well-known heterocyclic substituents. Nucleic acid monomers can be connected by phosphodiester bonds or analogs of such connections. Nucleic acids can be single-stranded or double-stranded.

[0082] As used herein, the term "recombinant" is intended to include all molecules prepared, expressed, generated or isolated by recombinant means, such as multispecific molecules (e.g., bispecific molecules) expressed using recombinant expression vectors transfected into host cells, multispecific molecules (e.g., bispecific molecules) isolated from animals (e.g., mice) transgenic for human immunoglobulin genes (see, e.g., Taylor et al., (1992), Nucleic Acids Res. 20:6287-6295), or multispecific molecules prepared, expressed, generated or isolated by any other means involving splicing of human immunoglobulin and / or MHC gene sequences to other DNA sequences. Such recombinant multispecific molecules may comprise antigen binding domains having variable and constant regions derived from human germline immunoglobulin sequences.

[0083] As used herein, the term "subject" or "patient" includes all members of the animal kingdom, including non-human primates and humans. In one embodiment, the patient is a person suffering from a disease or disorder, such as an infection or cancer.

[0084] The term "substantial identity" or "substantially identical" when referring to a nucleic acid or a fragment thereof means a nucleotide sequence identity of at least about 95%, more preferably at least about 96%, 97%, 98% or 99% of the nucleotide bases when optimally aligned with another nucleic acid (or its complementary strand) by appropriate nucleotide insertions or deletions, as measured by any well-known sequence identity algorithm such as FASTA, BLAST or Gap, as discussed below. In some cases, a nucleic acid molecule having substantial identity to a reference nucleic acid molecule can encode a polypeptide having an amino acid sequence that is identical or substantially similar to a polypeptide encoded by the reference nucleic acid molecule.

[0085] When applied to these polypeptides, the term "substantially similar" or "substantially similar" means that two peptide sequences share at least 95% sequence identity, even more preferably at least 98% or 99% sequence identity when optimally aligned using the default gap weights by the programs GAP or BESTFIT. Preferably, the residue positions that are not identical differ by conservative amino acid substitutions. "Conservative amino acid substitutions" are amino acid substitutions in which an amino acid residue is substituted by another amino acid residue having a similar side chain (R group) with chemical properties (e.g., charge or hydrophobicity). In general, conservative amino acid substitutions do not substantially change the functional properties of proteins. In the case where two or more amino acid sequences differ from each other due to conservative substitutions, the sequence identity percentage or degree of similarity can be adjusted upward to correct the conservative nature of the substitution. Methods for making this adjustment are well known to those skilled in the art. See, for example, Pearson (1994) Methods Mol. Biol. 24: 307-331, which is incorporated herein by reference. Examples of amino acid groups having chemically similar side chains include (1) aliphatic side chains: glycine, alanine, valine, leucine, and isoleucine; (2) aliphatic hydroxyl side chains: serine and threonine; (3) amide-containing side chains: asparagine and glutamine; (4) aromatic side chains: phenylalanine, tyrosine, and tryptophan; (5) basic side chains: lysine, arginine, and histidine; (6) acidic side chains: aspartic acid and glutamic acid; and (7) sulfur-containing side chains are cysteine ​​and methionine. Preferred conservative amino acid substitutions are: valine-leucine-isoleucine, phenylalanine-tyrosine, lysine-arginine, alanine-valine, glutamic acid-aspartic acid, and asparagine-glutamine. Alternatively, a conservative substitution is any change that has a positive value in the PAM250 log-likelihood matrix disclosed in Gonnet et al., (1992), Science, 256: 1443-1445, incorporated herein by reference. A "moderately conservative" substitution is any change that has a non-negative value in the PAM250 log-likelihood matrix.

[0086] Sequence similarity of polypeptides also referred to as sequence identity is usually measured using sequence analysis software. Protein analysis software uses a measure of the similarity assigned to various substitutions, deletions and other modifications (comprising conservative amino acid substitutions) to match similar sequences. For example, GCG software contains programs such as Gap and Bestfit, which can be used with default parameters to determine closely related polypeptides, such as sequence homology or sequence identity between homologous polypeptides from different organism species or between wild-type proteins and their mutant proteins. See, for example, 6.1 version GCG. FASTA using default or recommended parameters can also be used to compare polypeptide sequences, which is a program in 6.1 version GCG. FASTA (e.g., FASTA2 and FASTA3) provides alignment and sequence identity percentage of the best overlapping region between the query sequence and the search sequence (Pearson (2000), supra). When the sequence of the present invention is compared with a database containing a large number of sequences from different organisms, another preferred algorithm is a computer program BLAST using default parameters, especially BLASTP or TBLASTN. See, e.g., Altschul et al. (1990) J. Mol. Biol. 215:403-410 and Altschul et al. (1997) Nucleic Acids Res. 25:3389-402, each of which is incorporated herein by reference.

[0087] The terms "vector" and "expression vector" include but are not limited to viral vectors, plasmids, RNA vectors or linear or circular DNA or RNA molecules, which may be composed of chromosomes, non-chromosomes, semisynthetic or synthetic nucleic acids. In some cases, a vector is capable of autonomous replication (episomal vectors) and / or expression of nucleic acids (expression vectors) connected thereto. A large number of suitable vectors are known to those skilled in the art and are commercially available. Viral vectors include retroviruses, adenoviruses, parvoviruses (e.g., adeno-associated viruses), coronaviruses, negative-strand RNA viruses such as orthomyxoviruses (e.g., influenza viruses), rhabdoviruses (e.g., rabies and vesicular stomatitis viruses), paramyxoviruses (e.g., measles and Sendai), positive-strand RNA viruses such as picornaviruses and alphaviruses, and double-stranded DNA viruses, including adenoviruses, herpes viruses (e.g., herpes simplex virus type 1 and type 2, Epstein-Barr virus (Epstein-Barr virus), cytomegalovirus) and poxviruses (e.g., vaccinia, fowlpox and canarypox). Other viruses include, for example, Norwalk virus, togavirus, flavivirus, reovirus, papovavirus, hepadnavirus and hepatitis virus. Examples of retroviruses include: avian leukosis sarcoma, mammalian C-type, B-type virus, D-type virus, HTLV-BLV group and lentivirus.

[0088] Multispecific antigen binding molecules

[0089] The multispecific antigen-binding molecules (e.g., bispecific or trispecific or tetraspecific antigen-binding molecules) of the present invention include (a) a first polypeptide, which includes, from N-terminus to C-terminus, (i) a first antigen-binding domain that specifically binds to a T cell antigen, (ii) a first multimerization domain, and (iii) a second antigen-binding domain that specifically binds to a T cell antigen; and (b) a second polypeptide, which includes, from N-terminus to C-terminus, (i) a third antigen-binding domain that specifically binds to a target antigen and (ii) a second multimerization domain, wherein the first multimerization domain and the second multimerization domain associate with each other (e.g., through an interchain disulfide bond) to form a molecule.

[0090] In some embodiments, a multispecific antigen-binding molecule (e.g., a bispecific or trispecific or tetraspecific antigen-binding molecule) of the invention comprises (a) a first polypeptide comprising, from N-terminus to C-terminus, (i) a first antigen-binding domain that specifically binds to a T cell antigen, (ii) a first multimerization domain, and (iii) a second antigen-binding domain that specifically binds to a T cell antigen; and (b) a second polypeptide comprising, from N-terminus to C-terminus, (i) a third antigen-binding domain that specifically binds to a target antigen, (ii) a second multimerization domain, and (iii) a fourth antigen-binding domain that specifically binds to a target antigen, wherein the first multimerization domain and the second multimerization domain associate with each other (e.g., through an interchain disulfide bond) to form a molecule.

[0091] The antigen binding domains mentioned above and herein may be Fab domains, which include a heavy chain variable region (HCVR) and a heavy chain CH1 domain paired with a light chain variable region (LCVR) and a CL domain. The antigen binding domains mentioned above and herein may also be single chain variable fragment (scFv) domains, which include HCVR and LCVR connected together by a short peptide linker of, for example, about 10 to about 25 amino acids. Specific linkers include (G4S) nLinker, wherein n=1-10, or n is 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10. In some cases, the linker between the HCVR and LCVR of each scFv is (G4S)4. Unless otherwise defined, the antigen binding domains of the multispecific molecules of the present invention can be all Fab domains, all scFv domains, or a combination of Fab domains and scFv domains. In some cases, one or more of the antigen binding domains are Fab domains. In some cases, one or more of the antigen binding domains are scFv domains. In some cases, the first antigen binding domain and the third antigen binding domain are Fab domains. In some cases, the second antigen binding domain is a scFv domain. In some cases, the fourth antigen binding domain is a scFv domain. In some cases, the first and third antigen binding domains are Fab domains, and the second and fourth antigen binding domains are scFv domains. In some cases, the first, second and third antigen binding domains are Fab domains. In some cases, the first, second, third, and fourth antigen binding domains are Fab domains.

[0092] In various embodiments, the scFv domain is connected to the C-terminal of the corresponding multimerization domain by a joint peptide. In some cases, the length of the joint is between 1-10 amino acid. In certain embodiments, the length of the joint is between 1-20 amino acid. In this regard, the length of the joint can be 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 amino acid. In certain embodiments, the length of the joint can be 21, 22, 23, 24, 25, 26, 27, 28, 29 or 30 amino acid. The scope comprising the numerals discussed herein is also encompassed in the present disclosure, for example, a length of 10 to 30 amino acid joints. In certain embodiments, the joint is a flexible joint. Suitable linkers can be readily selected and can have any suitable different lengths, such as 1 amino acid (e.g., Gly) to 20 amino acids, 2 amino acids to 15 amino acids, 3 amino acids to 12 amino acids, including 4 amino acids to 10 amino acids, 5 amino acids to 9 amino acids, 6 amino acids to 8 amino acids or 7 amino acids to 8 amino acids, and can be 1, 2, 3, 4, 5, 6 or 7 amino acids. Exemplary flexible linkers include glycine polymers (G)n, glycine-serine polymers (GS) n(wherein n is an integer of at least 1 (e.g., 1-20)), glycine-alanine polymers, alanine-serine polymers, and other flexible linkers known in the art. Specific linkers include (G4S) n A linker, wherein n=1-10, or n is 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10. In some cases, the linker between each scFv domain and the C-terminus of the corresponding multimerization domain is (G4S)3.

[0093] In those embodiments where one or more antigen binding domains are scFvs, the scFv can be a stabilized scFv in which one or more modifications are made to the HCVR and / or LCVR sequences to generate and maintain the correct conformation of the scFv. In some embodiments, the scFv comprises cysteine ​​mutations at residue 44 of the HCVR and residue 100 of the LCVR (Kabat numbering) to generate disulfide bonds between the variable regions (see Zhao et al., Int. J. Mol. Sci, 12: 1-11, 2011; and Weatherill et al., Protein Engineering, Design and Selection, 25(7): 321-329, 2012). In some embodiments, the scFv comprises mutations at residue 39 of the HCVR and residue 38 of the LCVR (Kabat numbering) to modify glutamine residues to glutamic acid or lysine residues to inhibit conformational isomerization (see Igawa et al., Protein Engineering, Design and Selection, 23(8):667-677, 2010).

[0094] In various embodiments, the LCVR (and optionally, the CL) of any antigen binding domain may be a homologous LCVR corresponding to the HCVR, or the LCVR (and optionally, the CL) may be a common LCVR shared by multiple antigen binding domains. In some embodiments, the light chain of the Fab domain is a common light chain. In some embodiments, the light chain of the Fab domain is a homologous light chain corresponding to the target antigen binding domain, and the light chain is common to both Fab domains. In some embodiments, the LCVR of the scFv domain is a homologous LCVR. In some embodiments, the light chain of the Fab domain is a common light chain and the LCVR of the scFv domain is a homologous LCVR.

[0095] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a first scFv that specifically binds to a T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a second Fab that specifically binds to a target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a second scFv that specifically binds to a target antigen, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule. Exemplary structures of such molecules are as follows. Figure 1C shown.

[0096] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a second Fab that specifically binds to a T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a third Fab that specifically binds to a target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a fourth Fab that specifically binds to a target antigen, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule. Exemplary structures of such molecules are as follows Figure 1E shown.

[0097] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a first scFv that specifically binds to a T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a second Fab that specifically binds to a first target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a second scFv that specifically binds to a second target antigen different from the first target antigen, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule. Exemplary structures of such molecules are as follows. Figure 1F shown.

[0098] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a second Fab that specifically binds to a T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a third Fab that specifically binds to a first target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a fourth Fab that specifically binds to a second target antigen different from the first target antigen, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule. Exemplary structures of such molecules are as follows: Figure 1G shown.

[0099] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a first scFv that specifically binds to a T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a second Fab that specifically binds to a target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a second scFv that specifically binds to a T cell antigen, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule. Exemplary structures of such molecules are as follows. Figure 1H shown.

[0100] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a second Fab that specifically binds to a T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a third Fab that specifically binds to a target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a fourth Fab that specifically binds to a T cell antigen, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule. Exemplary structures of such molecules are as follows Fig. 1I shown.

[0101] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a second Fab that specifically binds to a T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a second Fab that specifically binds to a target antigen and (ii) a second immunoglobulin Fc domain, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule. Exemplary structures of such molecules are as follows Figure 1J shown.

[0102] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a second Fab that specifically binds to a T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a third Fab that specifically binds to a target antigen and (ii) a second immunoglobulin Fc domain, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule. Exemplary structures of such molecules are as follows Figure 1K shown.

[0103] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a first T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a first scFv that specifically binds to a second T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a second Fab that specifically binds to a target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a second scFv that specifically binds to a target antigen, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule. Exemplary structures of such molecules are as follows. Figure 1L shown.

[0104] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a first T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a second Fab that specifically binds to a second T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a third Fab that specifically binds to a target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a fourth Fab that specifically binds to a target antigen, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule. Exemplary structures of such molecules are as follows Figure 1M shown.

[0105] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a first T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a first scFv that specifically binds to a second T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a second Fab that specifically binds to a first target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a second scFv that specifically binds to a second target antigen different from the first target antigen, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule. Exemplary structures of such molecules are as follows. Figure 1N shown.

[0106] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a first T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a second Fab that specifically binds to a second T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a third Fab that specifically binds to a first target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a fourth Fab that specifically binds to a second target antigen different from the first target antigen, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule. Exemplary structures of such molecules are as follows. Fig.1O shown.

[0107] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a first T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a first scFv that specifically binds to a second T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a second Fab that specifically binds to a target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a second scFv that specifically binds to a T cell antigen (optionally can bind to a first T cell antigen, a second T cell antigen, or a third T cell antigen), wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through disulfide bonds to form a molecule. Exemplary structures of such molecules are as follows Figure 1P shown.

[0108] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a first T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a second Fab that specifically binds to a second T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a third Fab that specifically binds to a target antigen, (ii) a second immunoglobulin Fc domain, and (iii) a fourth Fab that specifically binds to a T cell antigen (optionally can bind to a first T cell antigen, a second T cell antigen, or a third T cell antigen), wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through disulfide bonds to form a molecule. Exemplary structures of such molecules are as follows Figure 1Q shown.

[0109] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a first T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a second Fab that specifically binds to a second T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a second Fab that specifically binds to a target antigen and (ii) a second immunoglobulin Fc domain, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule. Exemplary structures of such molecules are as follows. Figure 1R shown.

[0110] In some embodiments, the multispecific antigen-binding molecules of the present invention include: (a) a first polypeptide, which includes from N-terminus to C-terminus (i) a first Fab that specifically binds to a first T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a second Fab that specifically binds to a second T cell antigen; and (b) a second polypeptide, which includes from N-terminus to C-terminus (i) a third Fab that specifically binds to a target antigen and (ii) a second immunoglobulin Fc domain, wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule. Exemplary structures of such molecules are as follows Figure 1S shown.

[0111] Unless otherwise defined, and when present, the fourth antigen binding domain can specifically bind to target antigen or T cell antigen.In some cases, the third antigen binding domain and the fourth antigen binding domain specifically bind to different target antigens (different epitopes on the same protein, or different proteins).In some cases, the different target antigens are expressed on the surface of the same target cell (e.g., tumor cell).In some cases, the third antigen binding domain and the fourth antigen binding domain specifically bind to the same target antigen (the same epitope on the same protein).In various embodiments, the first and second antigen binding domains and the fourth antigen binding domain (when present, and for T cell antigens) can be combined with the same or different T cell antigens, as shown in the figure.In some cases, the first, second and fourth antigen binding domains specifically bind to different T cell antigens (different epitopes on the same protein, or different proteins).In some cases, the first, second and fourth antigen binding domains specifically bind to the same T cell antigen (the same epitope on the same protein).In some cases, different T cell antigens are co-stimulatory molecules (e.g., CD28) and checkpoint inhibitors (e.g., PD-1) on the surface of T cells. In such embodiments, the multispecific molecules of the present invention can provide co-stimulatory signals to T cells and prevent checkpoint inhibition. As used herein, reference to "same" target antigen or T cell antigen does not necessarily mean that the antigen binding domain is bound to the same surface molecule, but that the antigen binding domain has the same specificity (e.g., each of which binds CD3 or TA). Similarly, reference to "different" target antigen or T cell antigen means that it is different from another target antigen (e.g., MAGEA4 vs. EGFR) or another T cell antigen (e.g., CD28 vs. PD-1), or another epitope on the same protein.

[0112] In any of the embodiments discussed above or herein, the target antigen can be a tumor-associated antigen or an infection-associated antigen (e.g., a viral antigen, a bacterial antigen, a fungal antigen, or an antigen expressed by a parasite). In some cases, the target antigen is a tumor-associated antigen. In some cases, the target antigen is an infection-associated antigen. In some cases, the target antigen is a viral antigen. In some cases, the target antigen is a bacterial antigen. In some cases, the target antigen is a fungal antigen. In some embodiments, the target antigen is an antigen expressed by a parasite.

[0113] In some cases, the target antigen is a peptide (PiG) in the groove background of the major histocompatibility complex (MHC) protein. In some embodiments, PiG is a peptide consisting of about 5 to about 40 amino acid residues, about 6 to about 30 amino acid residues, about 8 to about 20 amino acid residues, or about 9, 10 or 11 amino acid residues. In some cases, PiG is a fragment of a tumor-associated antigen, a viral antigen, a bacterial antigen, a fungal antigen, or a parasitic antigen. In various embodiments, the target antigen is a peptide in the groove background of any class, subtype, or allele of a human leukocyte antigen, including any one of HLA-A, HLA-B, HLA-C, HLA-DR, HLA-DQ, or HLA-DP. In some embodiments, the target antigen is a peptide / MHC complex. In some cases, the peptide in the peptide / MHC complex is a fragment of a tumor-associated antigen, a fragment of a bacterial antigen, a fragment of a viral antigen, a fragment of a fungal antigen, or a fragment of a parasitic antigen.

[0114] In some cases, the antigen is a tumor-associated antigen or an antigen expressed by a tumor cell. In some embodiments, the tumor-associated antigen is selected from the group consisting of AFP, ALK, BAGE protein, BIRC5 (survivin), BIRC7, β-catenin, brc-abl, BRCA1, BORIS, CA9, carbonic anhydrase IX, caspase-8, CALR, CCR5, CD19, CD20 (MS4A1), CD22, CD40, CD70, CDK4, CEA, cyclin-B1, CYP1B1, EGFR, EGFRvIII, ErbB2 / Her2, ErbB3, ErbB4, ETV6-AML, EpCAM, EphA2, Fra-1, FOLR1, GAGE ​​protein (e.g., GAGE-1, 2), GD2, GD3, GloboH, phosphatidylinositol glycan-3, GM3, gp100, Her2, HLA / B-raf, HL A / k-ras, HLA / MAGE-A3, hTERT, IL-10, LMP2, MAGE proteins (e.g., MAGE-1, 2, 3, 4, 6, and 12), MART-1, mesothelin, ML-IAP, Muc1, Muc2, Muc3, Muc4, Muc5, Muc16 (CA-125), MUM1, NA17, NY-BR1, NY-BR62, NY-BR85, NY-ESO1, p15, p53, PAP, PAX3, PAX5, PCTA-1, PLAC1, PRLR, PRAME, PSMA (FOLH1), RAGE proteins, Ras, RGS5, Rho, SART-1, SART-3, STEAP1, STEAP2, TAG-72, TGF-β, TMPRSS2, Thompson-nouvelle antigen (Thompson-nouvelle antigen antigen, Tn), TRP-1, TRP-2, tyrosinase and uroplakin-3.

[0115] In some cases, the antigen is a viral antigen or a bacterial antigen. In some embodiments, the viral antigen is associated with or expressed by a virus selected from the group consisting of adenovirus, astrovirus, chikungunya virus, cytomegalovirus, dengue, Ebola, EBV, Hantavirus, HBsAg, hepatitis A, hepatitis B, hepatitis C, hepatitis D, hepatitis E, herpes, HIV, HPIV, HTLV, influenza, Japanese encephalitis virus, lassa, measles, metapneumovirus, mumps, norovirus, oropauche, HPV, parvovirus, rotavirus, RSV, rubella, SARS, TBEV, usutu, , vaccinia, varicella, West Nile, yellow fever and Zika, or the bacterial antigen is derived from a bacterium selected from the group consisting of methicillin-resistant Staphylococcus aureus (MRSA), Clostridium difficile, carbapenem-resistant Enterobacter, drug-resistant gonococci, multidrug-resistant Acinetobacter, drug-resistant Campylobacter, fluconazole-resistant Candida, extended-spectrum β-lactamase-producing bacteria, vancomycin-resistant Enterococci, multidrug-resistant Pseudomonas aeruginosa, drug-resistant non-typhoidal Salmonella, drug-resistant Salmonella serotypes, drug-resistant Shigella, drug-resistant Streptococcus pneumoniae, drug-resistant Mycobacterium tuberculosis, vancomycin-resistant Staphylococcus aureus, erythromycin-resistant group A Streptococcus, and clindamycin-resistant group B Streptococcus.

[0116] In any embodiment discussed above or herein, T cell antigens can be antigens expressed at the surface of T cells, T cell receptor complex antigens, costimulatory molecules or checkpoint inhibitors on T cells, CD3, CD27, CD28, 4-1BB or PD-1. In some cases, T cell antigens are T cell receptor complex antigens. In some cases, the T cell antigen is CD3. In some cases, the T cell antigen is a costimulatory molecule or a checkpoint inhibitor on T cells. In some cases, the T cell antigen is selected from the group consisting of CD27, CD28, 4-1BB and PD-1. In some cases, the T cell antigen is selected from the group consisting of CD3, CD27, CD28, 4-1BB and PD-1. In some cases, T cell antigens are selected from the group consisting of: CD28, ICOS, HVEM, CD27, 4-1BB, OX40, DR3, GITR, CD30, SLAM, CD2, 2B4, CD226, TIM1 and TIM2.

[0117] In certain embodiments in which the T cell antigen is CD3, the CD3 binding domain binds to human CD3 and induces human T cell activation. In certain embodiments, the CD3 binding domain binds weakly to human CD3 and induces human T cell activation. In some embodiments, the CD3 binding domain binds weakly to human CD3 and induces killing of tumor-associated antigen-expressing cells. In some embodiments, the CD3 binding domain weakly binds or associates with human and cynomolgus macaque (monkey) CD3, but the binding interaction cannot be detected by in vitro assays known in the art. In some embodiments, the CD3 binding domain binds to human CD3 with weak affinity. In some embodiments, the CD3 binding domain binds to human CD3 with moderate affinity. In some embodiments, the CD3 binding domain binds to human CD3 with high affinity. In some embodiments, the CD3 binding domain binds to human CD3 with a K of less than 15nM. D Binds to human CD3 (e.g., at 25° C.) as measured by surface plasmon resonance (e.g., mAb-capture or antigen-capture format) or a substantially similar assay. In some embodiments, the CD3 binding domain binds to human CD3 with a K of greater than about 15 nM, greater than about 20 nM, greater than about 30 nM, greater than about 40 nM, greater than about 50 nM, greater than about 60 nM, greater than about 100 nM, greater than about 200 nM, or greater than about 300 nM. D Binds to human CD3 as measured in a surface plasmon resonance binding assay (e.g., a mAb-capture or antigen-capture format) or a substantially similar assay. In some embodiments, the antibodies or antigen-binding fragments of the invention have a K of less than about 5 nM, less than about 2 nM, less than about 1 nM, less than about 800 pM, less than about 600 pM, less than about 500 pM, less than about 400 pM, less than about 300 pM, less than about 200 pM, less than about 180 pM, less than about 160 pM, less than about 140 pM, less than about 120 pM, less than about 100 pM, less than about 80 pM, less than about 60 pM, less than about 40 pM, less than about 20 pM, or less than about 10 pM. D Binding to human CD3 as measured by surface plasmon resonance, e.g., using an assay format as defined in Example 3 herein (e.g., a mAb-capture or antigen-capture format), or a substantially similar assay.

[0118] In some embodiments, the CD3 binding domain exhibits an EC of less than about 50 nM, less than about 40 nM, less than about 30 nM, less than about 20 nM, less than about 10 nM, less than about 5 nM, less than about 4 nM, less than about 3 nM, less than about 2 nM, less than about 1 nM, less than 900 pM, less than 800 pM, less than 700 pM, less than 600 pM, or less than 500 pM. 50Values ​​as measured in an in vitro flow cytometry binding assay. In some embodiments, the CD3 binding domain exhibits an EC of about or greater than about 1 nM, 2 nM, 3 nM, 4 nM, 5 nM, 10 nM, 25 nM, 50 nM, 100 nM, 500 nM, or 1 μM. 50 Values, as measured in an in vitro flow cytometric binding assay.

[0119] In any embodiment, the CD3 binding domain may include any of the HCVR / LCVR or CDR (e.g., six CDRs contained in a pair of HCVR / LCVR sequences) amino acid sequences of anti-CD3 antibodies disclosed in WO 2014 / 047231 (9250-WO) or WO2017 / 053856 (10151WO01), the anti-CD3 antibodies comprising antibodies identified as 7195P, 7221G, 7221G5, and 7221G20. In various embodiments, the anti-CD3 antibodies identified as "strong binders" have an affinity for human CD3 in the single-digit nanomolar range (e.g., 1-9 nM), as measured in a surface plasmon resonance assay (e.g., measured on a T200 BIACORE instrument at 25°C in antigen-capture format). In various embodiments, the anti-CD3 antibodies identified as "moderate binders" have an affinity for human CD3 in the double-digit nanomolar range (e.g., 10-99 nM, optionally, 10-50 nM or 10-25 nM), as measured in a surface plasmon resonance assay. In various embodiments, the anti-CD3 antibodies identified as "weak binders" have an affinity for human CD3 in the triple-digit nanomolar range (e.g., 100-999 nM, optionally, 100-500 nM or 500 nM to 1 μM), as measured in a surface plasmon resonance assay. In various embodiments, the anti-CD3 antibodies identified as "very weak binders" have an affinity for human CD3 greater than 10 μM or undetectable, as measured in a surface plasmon resonance assay.

[0120] In any embodiment, the CD3 binding domain can include any of the HCVR / LCVR or CDR (e.g., the six CDRs contained in a pair of HCVR / LCVR sequences) amino acid sequences listed in the table below (the "G" version is taken from WO 2017 / 053856). In some embodiments, the CD3 binding domain (e.g., in a Figure 1C In other words, the cognate light chain of the target antigen binding domain is common to both the target antigen binding domain and the CD3 binding domain (e.g., in Figure 1C or in the N-terminal Fab domain of the structure of 1F).

[0121] Table 1: Heavy chain amino acid sequence identifiers

[0122]

[0123] Table 2: Heavy chain nucleic acid sequence identifiers

[0124]

[0125] Table 3: Light chain amino acid sequence identifiers

[0126]

[0127] Table 4: Light chain nucleic acid sequence identifiers

[0128]

[0129] Each antibody listed in Table 1 includes a common light chain variable region including the amino acid sequence listed in Table 3. Each "G" designated antibody may also be referred to herein by a "7221" prefix, e.g., 7221G, 7221G5, 7221G20, etc. In the scFv version of the antigen binding domain, amino acid residue 44 of the heavy chain variable region may be substituted with a cysteine ​​residue, e.g., as shown in SEQ ID NO: 169 (modified heavy chain corresponding to 7195P) or SEQ ID NO: 170 (modified heavy chain corresponding to 7221G).

[0130] The multispecific antigen-binding molecules of the present invention (e.g., bispecific or trispecific or tetraspecific antigen-binding molecules) include two polypeptide chains, each of which includes a multimerization domain that promotes the association of the two polypeptide chains (e.g., through an interchain disulfide bond) to form a single multispecific antigen-binding molecule. In any embodiment discussed above or herein, the first and second multimerization domains can be immunoglobulin Fc domains (e.g., of human IgG isotype). In some cases, the first and second multimerization domains are associated with each other through disulfide bonds. In some embodiments, the first multimerization domain and the second multimerization domain are human IgG1 or human IgG4 Fc domains. In some cases, the first and second multimerization domains include hinge domains, CH2 domains, and CH3 domains of human IgG1 or human IgG4.

[0131] In some embodiments, the first multimerization domain or the second multimerization domain include an amino acid replacement that reduces the affinity for protein A binding compared to the wild-type Fc domain of the same isotype (e.g., human IgG1 or human IgG4). In some cases, the amino acid replacement includes H435R modification, or H435R and Y436F modification (EU numbering). In some cases, the first multimerization domain includes H435R and Y436F modification. In some cases, the second multimerization domain includes H435R and Y436F modification.

[0132] In some embodiments, the first polypeptide, the second polypeptide, or both the first polypeptide and the second polypeptide comprise a modified hinge domain that has reduced binding affinity to an Fcγ receptor compared to a wild-type hinge domain of the same isotype (e.g., human IgG1 or human IgG4).

[0133] In various embodiments in which the multimerization domain comprises a heavy chain constant region comprising a hinge domain, the constant region can be chimeric, combining sequences derived from more than one immunoglobulin isotype. For example, a chimeric Fc domain can comprise sequences derived from human IgG1, human IgG2, or human IgG4. H C in Zone 2 H 2 sequence and a portion or all of a C sequence derived from human IgG1, human IgG2 or human IgG4 H 3 sequences. The chimeric Fc domain may also contain a chimeric hinge region. For example, a chimeric hinge may include an "upper hinge" sequence derived from a human IgG1, human IgG2, or human IgG4 hinge region, which is combined with a "lower hinge" sequence derived from a human IgG1, human IgG2, or human IgG4 hinge region. A specific example of a chimeric Fc domain that may be included in any antigen-binding molecule shown herein includes, from N-terminus to C-terminus: [IgG4 CH1]-[IgG4 upper hinge]-[IgG2 lower hinge]-[IgG4 CH2]-[IgG4 CH3]. Another example of a chimeric Fc domain that may be included in any antigen-binding molecule listed herein includes, from N-terminus to C-terminus: [IgG1 CH1]-[IgG1 upper hinge]-[IgG2 lower hinge]-[IgG4 CH2]-[IgG1 CH3]. These and other examples of chimeric Fc domains that may be included in any antigen binding molecule of the invention are described in WO 2014 / 121087 (8550-WO). Chimeric Fc domains and variants thereof having these general structural arrangements may have altered Fc receptor binding, which in turn affects Fc effector function.

[0134] In various embodiments in which the multimerization domain includes a heavy chain constant region comprising a hinge domain, positions 233-236 within the hinge domain can be G, G, G and vacant; G, G, vacant and vacant; G, vacant, vacant and vacant; or all vacant, wherein the positions are numbered according to EU numbering. Optionally, the heavy chain constant region includes a hinge domain, a CH2 domain and a CH3 domain from N-terminus to C-terminus. Optionally, the heavy chain constant region includes a CH1 domain, a hinge domain, a CH2 domain and a CH3 domain from N-terminus to C-terminus. Optionally, the CH1 region (if present), the rest of the hinge region (if present), the CH2 region and the CH3 region are the same human isotype. Optionally, the CH1 region (if present), the rest of the hinge region (if present), the CH2 region and the CH3 region are human IgG1. Optionally, the CH1 region (if present), the rest of the hinge region (if present), the CH2 region and the CH3 region are human IgG2. Optionally, the CH1 region (if present), the remainder of the hinge region (if any), the CH2 region and the CH3 region are human IgG4. Optionally, the constant region has a CH3 domain modified to reduce binding to protein A. These and other examples of multimeric heavy chain constant regions that may be included in any antigen binding molecule of the invention are described in WO 2016 / 161010 (10140WO01).

[0135] In embodiments of the present invention, the association of one multimerization domain with another multimerization domain promotes the association between two antigen binding domains, thereby forming a multispecific antigen binding molecule. The multimerization domain can be any macromolecule, protein, polypeptide, peptide or amino acid that has the ability to associate with a second multimerization domain having the same or similar structure or composition. For example, the multimerization domain can be a protein that includes immunoglobulin C. H A non-limiting example of a multimeric component is the Fc portion of an immunoglobulin (including the C H 2-C H 3 domains), e.g., an IgG selected from the group consisting of isotypes IgG1, IgG2, IgG3, and IgG4, and an Fc domain of any isotype within each isotype group.

[0136] In some embodiments, the first multimerization domain and the second multimerization domain can be the same IgG isotype, such as IgG1 / IgG1, IgG2 / IgG2, IgG4 / IgG4, etc. Alternatively, the first multimerization domain and the second multimerization domain can be different IgG isotypes, such as IgG1 / IgG2, IgG1 / IgG4, IgG2 / IgG4, etc.

[0137] In certain embodiments, the multimerization domain is an Fc fragment or an amino acid sequence of 1 to about 200 amino acids in length containing at least one cysteine ​​residue. In other embodiments, the multimerization domain is a cysteine ​​residue or a peptide containing short cysteine. Other multimerization domains include peptides or polypeptides including or consisting of a leucine zipper, a spiral loop motif, or a coiled coil motif.

[0138] A multimerization domain, such as an Fc domain (with or without a hinge), may include one or more amino acid changes (e.g., insertions, deletions, or substitutions) compared to a wild-type, naturally occurring version of the Fc domain. For example, the present invention comprises a bispecific antigen-binding molecule comprising one or more modifications in the Fc domain that result in a modified Fc domain with a modified binding interaction (e.g., enhanced or reduced) between Fc and FcRn. In one embodiment, the bispecific antigen-binding molecule comprises a C H Zone 2 or C H 3 region, wherein the modification increases the affinity of the Fc domain for FcRn in an acidic environment (e.g., in an endosome with a pH range of about 5.5 to about 6.0). Non-limiting examples of such Fc modifications include, for example, modifications at position 250 (e.g., E or Q); modifications at positions 250 and 428 (e.g., L or F); modifications at positions 252 (e.g., L / Y / F / W or T), 254 (e.g., S or T), and 256 (e.g., S / R / Q / E / D or T); or modifications at positions 428 and / or 433 (e.g., L / R / S / P / Q or K) and / or 434 (e.g., H / F or Y); or modifications at positions 250 and / or 428; or modifications at positions 307 or 308 (e.g., 308F, V308F) and 434. In one embodiment, the modifications include 428L (e.g., M428L) and 434S (e.g., N434S) modifications; 428L, 259I (e.g., V259I) and 308F (e.g., V308F) modifications; 433K (e.g., H433K) and 434 (e.g., 434Y) modifications; 252, 254 and 256 (e.g., 252Y, 254T and 256E) modifications; 250Q and 428L modifications (e.g., T250Q and M428L); and 307 and / or 308 modifications (e.g., 308F or 308P).

[0139] The present invention also includes a multispecific antigen-binding molecule comprising a first Ig C H 3 domain and the second Ig C H 3 domains, wherein the first Ig C H 3 domain and the second Ig C HThe three domains differ from each other by at least one amino acid, and wherein the at least one amino acid difference reduces binding of the bispecific antibody to protein A compared to a bispecific antibody lacking the amino acid difference. H 3 domains bind protein A, and the second Ig C H The 3 domain contains a mutation that reduces or eliminates protein A binding, such as the H95R modification (by IMGT exon numbering; H435R by EU numbering). H 3 may further include a Y96F modification (by IMGT; Y436F by EU). See, e.g., U.S. Pat. No. 8,586,713. H Further modifications found within 3 include: in the case of IgG1 antibodies, D16E, L18M, N44S, K52N, V57M and V82I (by IMGT; D356E, L358M, N384S, K392N, V397M and V422I by EU); in the case of IgG2 antibodies, N44S, K52N and V82I (IMGT; N384S, K392N and V422I by EU); and in the case of IgG4 antibodies, Q15R, N44S, K52N, V57M, R69K, E79Q and V82I (by IMGT; Q355R, N384S, K392N, V397M, R409K, E419Q and V422I by EU).

[0140] Preparation of antigen binding domains and construction of bispecific molecules

[0141] Antigen binding domains specific for specific antigens can be prepared by any antibody generation technology known in the art. Once obtained, different antigen binding domains specific for two or more different antigens (e.g., CD3 and target antigen) can be appropriately arranged relative to each other to produce the structure of the multispecific antigen binding molecule of the present invention using conventional methods. In certain embodiments, one or more individual components (e.g., heavy and light chains or portions thereof) of the multispecific antigen binding molecules of the present invention are derived from chimeric antibodies, humanized antibodies or fully human antibodies. Methods for preparing such antibodies are well known in the art. For example, VELOCIMMUNE can be used. TM One or more heavy chains and / or light chains of the multispecific antigen-binding molecules of the invention are prepared using VELOCIMMUNE TMUsing the PCR technology (or any other human antibody generation technology), high affinity chimeric antibodies are initially isolated for a specific antigen (e.g., CD3 or target antigen) having human variable regions and mouse constant regions. The antibodies are characterized and selected for desired properties including affinity, selectivity, epitope, etc. The mouse constant regions are replaced with the desired human constant regions to generate fully human heavy and / or light chains that can be incorporated into the multispecific antigen-binding molecules of the invention.

[0142] Genetically engineered animals can be used to prepare human multispecific antigen-binding molecules. For example, genetically modified mice that cannot rearrange and express endogenous mouse immunoglobulin light chain variable sequences can be used, wherein the mouse only expresses one or two human light chain variable domains encoded by human immunoglobulin sequences, and the human immunoglobulin sequences are operably connected to the mouse κ constant gene at the endogenous mouse κ locus. Such genetically modified mice can be used to produce fully human multispecific antigen-binding molecules including two different heavy chains, and the heavy chain is associated with the same light chain including the variable domain derived from one of two different human light chain variable region gene segments. (See, for example, US2011 / 0195454). Fully human refers to an antibody or its antigen-binding fragment or immunoglobulin domain comprising an amino acid sequence encoded by DNA, and the DNA is derived from a human sequence within the entire length of each polypeptide of an antibody or its antigen-binding fragment or immunoglobulin domain. In some cases, fully human sequences are derived from human endogenous proteins. In other cases, fully human proteins or protein sequences include chimeric sequences, wherein each component sequence is derived from a human sequence. While not being bound by any one theory, chimeric proteins or chimeric sequences are generally designed to minimize the generation of immunogenic epitopes at the junction of the component sequences, for example, compared to any wild-type human immunoglobulin region or domain.

[0143] In various embodiments, the methods and techniques discussed above are used to generate antibodies against T cell antigens and target antigens, and the antigen binding domains (e.g., HCVRs, LCVRs, or CDRs) of these antibodies are used to generate multispecific antigen binding molecules discussed herein or having, for example, Figure 1C and 1E -Multispecific antigen-binding molecule with the structure shown in 1S.

[0144] Binding properties of antigen binding domains

[0145] As used herein, the term "binding" in the context of binding of an antibody (e.g., a corresponding antibody), an immunoglobulin, an antigen binding domain or a multispecific antigen binding molecule to, for example, a predetermined antigen (e.g., a cell surface protein or fragment thereof) generally refers to an interaction or association between a minimum of two entities or molecular structures, such as an antigen binding domain / antigen interaction.

[0146] For example, when measured by, e.g., surface plasmon resonance (SPR) technology in a BIAcore 3000 instrument using an antigen as a ligand and an antibody, Ig, antibody binding domain or multispecific antigen-binding molecule as an analyte (or anti-ligand), binding affinity is typically about 10 -7 M or lower (e.g. about 10 -8 M or lower, such as about 10 -9 or lower) D Flow cytometry is also often used.

[0147] Thus, an antibody (e.g., a corresponding antibody), antigen-binding domain, or multispecific antigen-binding molecule of the invention binds to a predetermined antigen or cell surface molecule that has a K D The affinity corresponding to the K value is at least ten times lower than the affinity of the antigen or cell surface molecule binding to a non-specific antigen (e.g., BSA, casein). According to the present invention, an affinity corresponding to the K value equal to or less than ten times that of the non-specific antigen is at least ten times lower than the affinity of the antigen or cell surface molecule binding to the non-specific antigen (e.g., BSA, casein). D The affinity of an antibody (e.g., a corresponding antibody), antigen-binding domain, or multispecific antigen-binding molecule corresponding to the value can be considered to have no detectable binding, but such an antibody can be paired with a second antigen-binding arm to generate a bispecific antibody of the invention.

[0148] The term "K D "(M)" refers to the dissociation equilibrium constant of a specific antibody (or antigen binding domain)-antigen interaction, or the dissociation equilibrium constant of the binding of an antibody (or antigen binding domain) or antibody binding fragment to an antigen. K D There is an inverse relationship between K and binding affinity, so K D The smaller the value, the higher the affinity, ie the stronger it is. Therefore, the term "higher affinity" or "stronger affinity" relates to a higher ability to form an interaction and therefore to a smaller K D value, and conversely, the term "lower affinity" or "weaker affinity" relates to a lower ability to form an interaction, and thus K D In some cases, a particular molecule (e.g., an antibody or antigen binding domain) has a higher binding affinity (or K) for a partner molecule (e.g., antigen X) with which it interacts, as compared to the binding affinity of the molecule (e.g., an antibody or antigen binding domain) for another interacting partner molecule (e.g., antigen Y). D ) can be expressed as D value (lower or weaker affinity) divided by the smaller K D A binding ratio determined by comparing a binding affinity of a given protein to a given protein (higher or stronger affinity) can be expressed, for example, as a 5-fold or 10-fold greater binding affinity, as appropriate.

[0149] The term “k d "(seconds-1 or 1 / s) refers to the dissociation rate constant for a particular antibody (or antigen binding domain)-antigen interaction, or the dissociation rate constant for an antibody or antibody binding domain. This value is also referred to as k off value.

[0150] The term “k a "(M-1×sec-1 or 1 / M) refers to the association rate constant for a specific antibody (or antigen binding domain)-antigen interaction, or the association rate constant of an antibody or antibody binding domain.

[0151] The term "K A "(M-1 or 1 / M) refers to the association equilibrium constant of a specific antibody (or antigen binding domain)-antigen interaction, or the association equilibrium constant of the antibody or antibody binding domain. k a Divide by k d The association equilibrium constant was obtained.

[0152] The term "EC50" or "EC 50 ” refers to the half maximal effective concentration, which comprises the concentration of the antibody (or antigen-binding domain or multispecific molecule) that induces a response intermediate between the baseline and maximum after a specified exposure time. EC 50 Substantially represents the concentration of an antibody (or antigen binding domain or multispecific molecule) at which 50% of its maximal effect is observed. 50 The value is equal to the concentration of the multispecific molecule of the invention that binds to cells expressing CD3 or a target antigen (e.g., a tumor-associated antigen) at half-maximal binding as determined by, for example, a flow cytometric binding assay. Thus, decreased or weaker binding is observed, said binding having an increased EC 50 Or half of the maximum effective concentration.

[0153] In one embodiment, decreased binding can be defined as an increased EC 50 The concentration of the molecule is such that it is able to bind to half of the target cells at maximum.

[0154] In another embodiment, EC 50 The values ​​represent the concentration of the molecule of the invention which triggers half-maximal depletion of target cells by T cell cytotoxic activity. Thus, increased cytotoxic activity (e.g., T cell-mediated tumor cell killing) is observed with a decreased EC 50 Or the effective concentration value of half the maximum value.

[0155] pH-dependent binding

[0156] The present invention includes antigen binding domains and multispecific antigen binding molecules with pH-dependent binding characteristics. For example, compared with neutral pH, the molecules of the present invention can show reduced binding to T cell antigens or target antigens at acidic pH. Alternatively, compared with neutral pH, the molecules of the present invention can show enhanced binding to T cell antigens or target antigens at acidic pH. The expression "acidic pH" includes a pH value less than about 6.2, for example, about 6.0, 5.95, 5.9, 5.85, 5.8, 5.75, 5.7, 5.65, 5.6, 5.55, 5.5, 5.45, 5.4, 5.35, 5.3, 5.25, 5.2, 5.15, 5.1, 5.05, 5.0 or less. As used herein, the expression "neutral pH" means that the pH is about 7.0 to about 7.4. The expression "neutral pH" encompasses pH values ​​of about 7.0, 7.05, 7.1, 7.15, 7.2, 7.25, 7.3, 7.35 and 7.4.

[0157] In certain instances, "reduced binding at acidic pH compared to neutral pH" refers to the K of a molecule (or antigen binding domain) binding to its antigen at acidic pH. D The value is related to the K of the molecule (or antigen binding domain) binding to its antigen at neutral pH. D For example, if a molecule or antigen binding domain exhibits an acidic / neutral K of about 3.0 or greater, D For purposes of the present invention, a molecule or antigen binding domain may be considered to exhibit "reduced binding to a T cell antigen or target antigen at acidic pH compared to neutral pH" if the acidic / neutral K of a molecule or antigen binding domain of the invention is greater than or equal to 1. D The ratio may be about 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, 7.0, 7.5, 8.0, 8.5, 9.0, 9.5, 10.0, 10.5, 11.0, 11.5, 12.0, 12.5, 13.0, 13.5, 14.0, 14.5, 15.0, 20.0, 25.0, 30.0, 40.0, 50.0, 60.0, 70.0, 100.0 or more.

[0158] Multispecific molecules with pH-dependent binding properties can be obtained by, for example, screening a panel of corresponding antibodies for reduced (or enhanced) binding to a specific antigen at acidic pH compared to neutral pH. Additionally, modification of the antigen binding domain at the amino acid level can generate molecules with pH-dependent properties. For example, by replacing one or more amino acids of an antigen binding domain (e.g., within a CDR) with histidine residues, a molecule can be obtained that has reduced binding to an antigen at acidic pH relative to neutral pH.

[0159] Biological properties of multispecific antigen-binding molecules

[0160] The present invention may include multispecific antigen-binding molecules and antigen-binding domains thereof that are capable of simultaneously binding to a human T cell antigen (eg, CD3) and a human target antigen (eg, a tumor-associated antigen).

[0161] The present invention may comprise a multispecific antigen-binding molecule that binds to a human T cell antigen (e.g., CD3) and induces T cell activation in the presence of a target cell. For example, in some embodiments, the present invention comprises a multispecific antigen-binding molecule that binds to a human T cell antigen (e.g., CD3) and induces T cell cytotoxic activity in the presence of cells expressing one or more target antigens (e.g., tumor-associated antigens).

[0162] The present invention may comprise a multispecific antigen-binding molecule that binds to a human T cell antigen (e.g., CD3) and induces T cell activation without causing any abnormalities relative to conventional bispecific anti-CD3 x anti-TA antibodies (e.g., Figure 1A ) Increase cytokine production.

[0163] The present invention may comprise multispecific antigen-binding molecules that are capable of depleting or reducing a population of cells in which the cells express a target antigen or target antigens. The multispecific antigen-binding molecules of the present invention are capable of depleting or reducing a population of cells in which the cells express a target antigen or target antigens. Figure 1A and 1B ) more effectively induces T cell-mediated cytotoxicity.

[0164] The invention may comprise multispecific antigen-binding molecules that bind to a human T cell antigen (e.g., CD3) and two different target antigens (e.g., Figure 1F and induce cytotoxic activity and / or T cell activation in the presence of cells expressing both target antigens.

[0165] Many cancers express multiple intracellular antigens that are processed by proteosomes within the cell and the associated peptides are presented on the cell surface in the context of HLA molecules. Targeting peptides from different proteins can be used to increase the specificity of the multispecific molecules of the present invention. In some cases, cancers characterized by PiG antigens or low-density cancer antigens escape conventional cancer treatments because they are typically present in tumors with low target copy numbers. In addition, solid tumors characterized by PiG or low-density cancer antigens may be more resistant to treatment and more difficult to treat because they are not cell surface antigens, but are present in grooves within cancer-associated peptides. Therefore, the use of multispecific molecules of the present invention that target two different antigens (e.g., low-density antigens) can effectively target PiG and / or low-density cancer antigens to increase / enhance the efficacy of cancer treatment, especially those cancers characterized by solid tumors.

[0166] In various embodiments, the multispecific antigen-binding molecules of the invention are capable of inducing T cell-mediated cytotoxicity in a cell population when the target antigen is present at a density ranging from about 100 copies per cell to about 1 million copies per cell or more. In some cases, the target antigen is present at about 100, about 200, about 300, about 400, about 500, about 1000, about 2000, about 3000, about 4000, about 5000, about 6000, about 7000, about 8000, about 9000, about 10000, about 15000, about 20000, about 25000, about 30000, about 4000, about 5000, about 6000, about 7000, about 8000, about 9000, about 10000, about 15000, about 20000, about 25000, about 30000, about 4000 Densities of 35,000, about 40,000, about 45,000, about 50,000, about 75,000, about 100,000 (i.e., 100K), about 200K, about 300K, about 400K, about 500K, about 600K, about 700K, about 800K, about 900K, about 1 million, about 2 million, about 3 million, about 4 million, about 5 million or about 10 million copies exist.

[0167] Without being bound by theory, the inventors hypothesize that the improved cytotoxic potency of the molecular format of the invention is a function of the presence of two T cell antigen (e.g., CD3) binding domains on a single chain of the molecule. Specifically, it is hypothesized that the geometry of the molecular structure of the invention selectively induces lytic synapse formation at low concentrations and not stimulatory synapse formation, the latter being responsible for cytokine production by cytotoxic T lymphocytes.

[0168] Epitope mapping and related technologies

[0169] The epitope on the T cell antigen (e.g., CD3) and / or target antigen (e.g., tumor-associated antigen) bound by the antigen binding molecules of the present invention can be composed of a single continuous sequence of 3 or more (e.g., 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 or more) amino acids of the protein. Alternatively, the epitope can be composed of multiple non-continuous amino acids (or amino acid sequences) of the protein. The molecules of the present invention can interact with, for example, amino acids contained in a single CD3 chain (e.g., CD3-ε, CD3-δ or CD3-γ), or can interact with amino acids on two or more different CD3 chains. As used herein, the term "epitope" refers to an antigenic determinant that interacts with a specific antigen binding site in the variable region of an antigen binding domain known as a paratope. A single antigen can have more than one epitope. Therefore, different antigen binding domains can bind to different regions on the antigen and can have different biological effects. Epitopes can be conformational or linear. Conformational epitopes are produced by spatially juxtaposed amino acids from different segments of a linear polypeptide chain. Linear epitopes are epitopes produced by adjacent amino acid residues in a polypeptide chain. In some cases, an epitope can comprise a sugar, phosphoryl or sulfonyl moiety on the antigen.

[0170] Various techniques known to those of ordinary skill in the art can be used to determine whether an antigen binding domain of a molecule "interacts with one or more amino acids" within a polypeptide or protein. Exemplary techniques include, for example, conventional cross-blocking assays (such as described in Antibody ( Antibodies, cross-blocking assays described in Harlow and Lane (Cold Spring Harbor Press, Cold Spring Harbor Laboratory, New York), alanine scanning mutation analysis, peptide blot analysis (Reineke, 2004, Methods in Molecular Biology, 248:443-463) and peptide cleavage analysis. In addition, methods such as epitope excision, epitope extraction and chemical modification of antigens can be used (Tomer, 2000, Protein Science, 9:487-496). Another method that can be used to identify amino acids within a polypeptide that interact with the antigen binding domain of a molecule is hydrogen / deuterium exchange detected by mass spectrometry. In general, hydrogen / deuterium exchange methods involve deuterium labeling of the protein of interest and then combining the molecule with the deuterium-labeled protein. Next, the protein / molecule complex is transferred to water to allow hydrogen-deuterium exchange to occur at all residues except the molecule-protected residues (which remain deuterium-labeled). After dissociation of the molecule, the target protein is subjected to protease cleavage and mass spectrometry analysis, thereby revealing deuterium-labeled residues corresponding to specific amino acids that interact with the molecule. See, for example, Ehring, (1999) Analytical Biochemistry, 267(2):252-259; Engen and Smith (2001) Anal. Chem., 73:256A-265A. X-ray crystallography of antigen / molecule complexes can also be used for epitope mapping.

[0171] Bioequivalent

[0172] The present invention includes multispecific antigen-binding molecules that are bioequivalent to any of the exemplary multispecific antigen-binding molecules listed herein. If, for example, two antigen-binding proteins are drug equivalents or drug substitutes that do not show significant differences in absorption rate and absorption extent when applied with the same molar dose (single dose or multiple doses) under similar experimental conditions, the two antigen-binding proteins or antibodies are considered to be bioequivalent. If some antigen-binding proteins are equivalent in absorption extent but their absorption rates are not equivalent, they are considered to be equivalents or drug substitutes, and can be considered to be bioequivalent, because this intentional and reflected difference in absorption rate on the label is not necessary for obtaining effective body drug concentrations when, for example, long-term use, and is considered to be medically insignificant for the specific drug product under study.

[0173] In one embodiment, two antigen binding proteins are bioequivalent if there are no clinically meaningful differences in safety, purity, and potency.

[0174] In one embodiment, two antigen binding proteins are bioequivalent if patients can switch between the reference product and the biological product one or more times without an expected increased risk of adverse effects, including clinically significant changes in immunogenicity or decreased effectiveness, compared to continued treatment without switching.

[0175] In one embodiment, two antigen binding proteins are bioequivalent if they both act by one or more common mechanisms of action for one or more conditions of use, provided such mechanisms are known.

[0176] Bioequivalence can be demonstrated by in vivo and in vitro methods. Bioequivalence measurements include, for example: (a) in vivo tests performed in humans or other mammals, where the concentration of the antigen-binding protein or its metabolites is measured over time in blood, plasma, serum or other biological fluids; (b) in vitro tests that are correlated with and reasonably predictive of bioavailability data in humans; (c) in vivo tests in humans or other mammals, where appropriate acute pharmacological effects of the antigen-binding protein (or its target) are measured over time; and (d) in well-controlled clinical trials to establish the safety, efficacy or bioavailability or bioequivalence of the antigen-binding protein.

[0177] Bioequivalent variants of the exemplary multispecific antigen-binding molecules listed herein can be constructed, for example, by making various substitutions of residues or sequences or deleting terminal or internal residues or sequences that are not required for biological activity. For example, cysteine ​​residues that are not required for biological activity can be deleted or replaced with other amino acids to prevent the formation of unnecessary or incorrect intramolecular disulfide bonds upon renaturation. In other contexts, bioequivalent antigen-binding proteins can include variants of the exemplary multispecific antigen-binding molecules listed herein that include amino acid changes that alter the glycosylation characteristics of the molecule, for example, mutations that eliminate or remove glycosylation.

[0178] Species selectivity and cross-species reactivity

[0179] According to certain embodiments of the present invention, there is provided an antigen binding molecule that is combined with a human T cell antigen (e.g., CD3) but not combined with the same antigen from other species. There is also provided an antigen binding molecule that is combined with a human target antigen (e.g., tumor antigen) but not combined with the same target antigen from other species. The present invention also includes an antigen binding molecule that is combined with a human antigen and a corresponding antigen from one or more non-human species.

[0180] According to certain exemplary embodiments of the present invention, antigen binding molecules are provided that bind to human CD3 and / or human tumor antigens, and may or may not bind to one or more of mouse, rat, guinea pig, hamster, gerbil, pig, cat, dog, rabbit, goat, sheep, cattle, horse, camel, cynomolgus, marmoset, rhesus monkey or chimpanzee CD3 and / or tumor antigens, as appropriate. For example, in a specific exemplary embodiment of the present invention, a multispecific antigen binding molecule is provided that includes a first antigen binding domain that binds to human CD3 and cynomolgus CD3 and a second antigen binding domain that specifically binds to a human tumor antigen.

[0181] Immunoconjugates

[0182] The present invention encompasses antigen binding molecules conjugated to therapeutic moieties ("immunoconjugates"), such as cytotoxins, chemotherapeutic drugs, immunosuppressants, or radioisotopes. Cytotoxic agents include any agent that is harmful to cells. Examples of suitable cytotoxic and chemotherapeutic agents for forming immunoconjugates are known in the art (see, e.g., WO 05 / 103081).

[0183] Therapeutic Formulations and Administration

[0184] The invention provides pharmaceutical compositions comprising the multispecific antigen-binding molecules of the invention. The pharmaceutical compositions of the invention are formulated with suitable carriers, excipients, and other agents that provide improved transfer, delivery, tolerability, and the like. Many suitable formulations can be found in the formulary known to all pharmaceutical chemists: Remington's Pharmaceutical Sciences, Mack Publishing Company, Easton, PA. These formulations include, for example, powders, pastes, ointments, gels, waxes, oils, lipids, lipids (cationic or anionic) containing vesicles (such as LIPOFECTIN TM , Life Technologies, Carlsbad, CA), DNA conjugates, anhydrous absorption creams, oil-in-water and water-in-oil emulsions, polyethylene glycol emulsions (polyethylene glycol of various molecular weights), semisolid gels, and semisolid mixtures containing polyethylene glycol. See also Powell et al. "Compendium of excipients for parenteral formulations" PDA (1998) "Journal of Pharmaceutical Science and Technology ( ...

[0185] The dosage of the antigen binding molecule applied to the patient can vary according to the patient's age and body size, target disease, condition, route of administration, etc. The preferred dosage is usually calculated based on body weight or body surface area. When the multispecific antigen binding molecules of the present invention are used in adult patients for therapeutic purposes, it may be advantageous to administer the multispecific antigen binding molecules of the present invention intravenously, usually at a single dose of about 0.01 to about 20 mg / kg body weight, more preferably about 0.02 to about 7 mg / kg body weight, about 0.03 to about 5 mg / kg body weight, or about 0.05 to about 3 mg / kg body weight. Depending on the severity of the condition, the frequency and duration of treatment can be adjusted. The effective dose and schedule for administering the multispecific antigen binding molecule can be determined empirically; for example, patient progress can be monitored by regular assessment, and the dosage can be adjusted accordingly. In addition, interspecies scaling of dosage can be performed using methods well known in the art (e.g., Mordenti et al., 1991, Pharmaceut. Res. 8: 1351).

[0186] Various delivery systems are known and can be used to administer the pharmaceutical compositions of the invention, e.g., encapsulation in liposomes, microparticles, microcapsules, recombinant cells capable of expressing mutant viruses, receptor-mediated endocytosis (see, e.g., Wu et al. 1987 J. Biol. Chem. 262:4429-4432). Methods of introduction include, but are not limited to, intradermal, intramuscular, intraperitoneal, intravenous, subcutaneous, intranasal, epidural, and oral routes. The composition can be administered by any convenient route, e.g., by infusion or bolus injection, absorption through epithelial or mucocutaneous linings (e.g., oral mucosa, rectal and intestinal mucosa, etc.), and can be administered with other bioactive agents. Administration can be systemic or local.

[0187] The pharmaceutical compositions of the present invention can be delivered subcutaneously or intravenously with a standard needle and syringe. In addition, for subcutaneous delivery, pen delivery devices are readily applicable to deliver the pharmaceutical compositions of the present invention. Such pen delivery devices can be reusable or disposable. Reusable pen delivery devices typically utilize a replaceable cartridge containing the pharmaceutical composition. Once the entire pharmaceutical composition within the cartridge has been administered and the cartridge is empty, the empty cartridge can be easily discarded and replaced with a new cartridge containing the pharmaceutical composition. The pen delivery device can then be reused. In a disposable pen delivery device, there is no replaceable cartridge. Instead, the disposable pen delivery device is preloaded with the pharmaceutical composition held in a reservoir within the device. Once the pharmaceutical composition in the reservoir is emptied, the entire device is discarded.

[0188] Many reusable pen and auto-injector delivery devices are useful for subcutaneous delivery of the pharmaceutical compositions of the present invention. Examples include, but are not limited to, AUTOPENTM (Owen Mumford, Inc., Woodstock, UK), DISETRONIC TM pen (Disetronic Medical Systems, Bergdorf, Switzerland), HUMALOG MIX 75 / 25 TM Pen, HUMALOG TM Pen, HUMALIN 70 / 30 TM PEN (Eli Lilly and Co., Indianapolis, Indiana), NOVOPEN TM I, II and III (Novo Nordisk, Copenhagen, Denmark), NOVOPENJUNIOR TM (Novo Nordisk, Copenhagen, Denmark), BD TM PEN (Becton Dickinson, Franklin Lakes, NJ), OPTIPEN TM 、OPTIPEN PRO TM 、OPTIPEN STARLET TM OPTICLIK TM (sanofi-aventis, Frankfurt, Germany), to name a few. Examples of disposable pen delivery devices for subcutaneous delivery of the pharmaceutical compositions of the present invention include, but are not limited to, SOLOSTAR TM Pen (Sanofi), FLEXPEN TM (Novo Nordisk) and KWIKPEN TM (Eli Lilly), SURECLICK TM Autoinjector (Amgen, Thousand Oaks, CA), PENLET TM (Haselmeier, Stuttgart, Germany), EPIPEN (Dey, LP), and HUMIRA TM PEN (Abbott Labs, Abbott Technology Park, IL), to name a few.

[0189] In some cases, the pharmaceutical composition can be delivered in a controlled release system. In one embodiment, a pump can be used (see Langer, supra; Sefton, 1987, CRC: Biomedical Engineering Review (CRC Crit. Ref. Biomed. Eng.), 14: 201). In another embodiment, a polymer material can be used; see, Medical Applications of Controlled Release, Langer and Wise (ed.), 1974, CRC Press, Boca Raton, Florida (CRC Pres., Boca Raton, Florida). In yet another embodiment, a controlled release system can be placed near the target of the composition, so only a small portion of the systemic dose is required (see, for example, Goodson, 1984, Medical Applications of Controlled Release, supra, Vol. 2, pp. 115-138). Other controlled release systems are discussed in the review of Langer, 1990, Science, 249: 1527-1533.

[0190] Injectable preparations can include dosage forms for intravenous, subcutaneous, intradermal and intramuscular injections, instillations, etc. These injectable preparations can be prepared by known methods. For example, injectable preparations can be prepared by, for example, dissolving, suspending or emulsifying the above-mentioned antigen binding molecules or their salts in a sterile aqueous medium or a conventional oily medium for injection. As an aqueous medium for injection, there are, for example, physiological saline, isotonic solutions containing glucose and other adjuvants, etc., which can be used in combination with suitable solubilizing agents (such as alcohol (for example, ethanol), polyols (for example, propylene glycol, polyethylene glycol)), nonionic surfactants [for example, polysorbate 80, HCO-50 (polyoxyethylene (50mol) adducts of hydrogenated castor oil)], etc. As an oily medium, for example, sesame oil, soybean oil, etc. can be used, which can be used in combination with solubilizing agents such as benzyl benzoate, benzyl alcohol, etc. The injection prepared in this way is preferably filled in a suitable ampoule.

[0191] Advantageously, the pharmaceutical composition for oral or parenteral use is prepared into a dosage form suitable for a unit dose of a certain dose of active ingredient. Such dosage forms of unit doses include, for example, tablets, pills, capsules, injections (ampoules), suppositories, etc. The content of the above-mentioned antigen binding molecules is generally about 5 mg to about 500 mg per unit dose dosage form; particularly in the form of injection, it is preferred that the content of the above-mentioned antigen binding molecules is about 5 mg to about 100 mg and about 10 mg to about 250 mg for other dosage forms.

[0192] Therapeutic uses of antigen binding molecules

[0193] The present invention includes methods comprising administering a therapeutic composition to a subject in need thereof, the therapeutic composition comprising a multispecific antigen-binding molecule that specifically binds to a T cell antigen (e.g., CD3) and a target antigen (e.g., a tumor-associated antigen). The therapeutic composition may include any of the multispecific antigen-binding molecules disclosed herein, and a pharmaceutically acceptable carrier or diluent. As used herein, the expression "subject in need thereof" means a human or non-human animal that exhibits one or more symptoms or signs of cancer, or a human or non-human animal that would otherwise benefit from inhibition or reduction of target antigen activity or depletion of target antigen-positive cells (e.g., tumor cells).

[0194] The multispecific antigen-binding molecules of the invention (and therapeutic compositions comprising the same) are particularly useful for treating any disease or condition in which stimulation, activation and / or targeting of an immune response would be beneficial. In particular, the multispecific antigen-binding molecules of the invention can be used to treat, prevent and / or ameliorate any disease or condition associated with or mediated by targeted antigen expression or activity or proliferation of target antigen-positive cells. The mechanism of action for achieving the therapeutic methods of the invention comprises killing cells expressing the target antigen in the presence of T cells.

[0195] The multispecific antigen-binding molecules of the present invention can be used to treat diseases or conditions associated with target antigen expression, including, for example, cancer. Analysis / diagnostic methods such as tumor scanning known in the art can be used to determine whether a patient carries tumor cells that are positive for the target antigen. In some cases, the cancer is selected from solid tumors, cervical cancer, squamous cell carcinoma of the head and neck, melanoma, prostate cancer, acute myeloid leukemia, pancreatic cancer, colon cancer, acute lymphocytic leukemia, non-Hodgkin's lymphoma, gastric cancer, post-transplantation lymphoproliferative disease, ovarian cancer, lung cancer, squamous cell carcinoma, non-small cell lung cancer, esophageal cancer, bladder cancer, nasopharyngeal cancer, uterine cancer, liver cancer, testicular cancer, or breast cancer.

[0196] The present invention also encompasses methods for treating residual cancer in a subject.As used herein, the term "residual cancer" means the presence or persistence of one or more cancer cells in a subject after treatment with an anti-cancer therapy.

[0197] According to certain aspects, the invention provides methods for treating a disease or condition associated with target antigen expression (e.g., cancer), the method comprising administering to the subject one or more multispecific antigen-binding molecules described elsewhere herein after it has been determined that the subject suffers from a target antigen-positive cancer. For example, the invention comprises methods for treating cancer comprising administering to the patient a multispecific antigen-binding molecule 1 day, 2 days, 3 days, 4 days, 5 days, 6 days, 1 week, 2 weeks, 3 weeks or 4 weeks, 2 months, 4 months, 6 months, 8 months, 1 year or more after the subject has received other immunotherapy or chemotherapy.

[0198] Combination therapies and formulations

[0199] The present invention provides a method including administering a pharmaceutical composition, the pharmaceutical composition including any exemplary multispecific antigen binding molecules described herein and a combination of one or more other therapeutic agents. The exemplary additional therapeutic agents that can be combined or co-administered with the antigen binding molecules of the present invention include, for example, anti-tumor agents (e.g., chemotherapeutics). In certain embodiments, the second therapeutic agent can be a monoclonal antibody, an antibody drug conjugate, a bispecific antibody conjugated with an anti-tumor agent, a checkpoint inhibitor, or a combination thereof. Other agents that can be beneficially combined with the antigen binding molecules of the present invention include cytokine inhibitors, the cytokine inhibitors include small molecule cytokine inhibitors and antibodies that bind to cytokines such as IL-1, IL-2, IL-3, IL-4, IL-5, IL-6, IL-8, IL-9, IL-10, IL-11, IL-12, IL-13, IL-17, IL-18, or their corresponding receptors. The pharmaceutical compositions of the present invention (e.g., pharmaceutical compositions comprising multispecific antigen-binding molecules as disclosed herein) can also be administered as part of a treatment regimen comprising one or more therapeutic combinations selected from the following: monoclonal antibodies that can interact with different antigens on the surface of cells; bispecific antibodies, one arm of which binds to an antigen on the surface of a tumor cell and the other arm binds to an antigen on a T cell; antibody drug conjugates; bispecific antibodies conjugated to an anti-tumor agent; checkpoint inhibitors, such as checkpoint inhibitors targeting PD-1 or CTLA-4; or a combination thereof. In certain embodiments, the checkpoint inhibitor can be selected from a PD-1 inhibitor, such as pembrolizumab (Keytruda), nivolumab (Opdivo), or cemiplimab (REGN2810). In certain embodiments, the checkpoint inhibitor may be selected from a PD-L1 inhibitor, such as atezolizumab (Tecentriq), avelumab (Bavencio), or durvalumab (Imfinzi). In certain embodiments, the checkpoint inhibitor may be selected from a CTLA-4 inhibitor, such as ipilimumab (Yervoy). Other combinations that may be used in combination with the antibodies of the present invention are described above.

[0200] The present invention also comprises a therapeutic combination comprising any of the antigen binding molecules mentioned herein and an inhibitor of one or more of VEGF, Ang2, DLL4, EGFR, ErbB2, ErbB3, ErbB4, EGFRvIII, cMet, IGF1R, IL-10, B-raf, PDGFR-α, PDGFR-β, FOLH1 (PSMA), PRLR, STEAP1, STEAP2, TMPRSS2, MSLN, CA9, uroplakin, or any of the above cytokines, wherein the inhibitor is an aptamer, an antisense molecule, a ribozyme, siRNA, a peptibody, a nanobody, an antibody, a bispecific antibody, or an antibody fragment (e.g., a Fab fragment; a F(ab')2 fragment; a Fd fragment; a Fv fragment; a scFv; a dAb fragment; or other engineered molecules, such as diabodies, triabodies, tetrabodies, minibodies, and minimal recognition units). The antigen binding molecules of the present invention can also be used in combination with antiviral agents, antibiotics, analgesics, corticosteroids and / or NSAIDs and / or co-formulated. The antigen binding molecules of the present invention can also be used as part of a treatment regimen that also includes radiotherapy and / or conventional chemotherapy.

[0201] The additional therapeutically active component may be administered prior to, simultaneously with, or shortly after administration of the antigen binding molecules of the invention; (for purposes of this disclosure, such administration regimens are considered administration of the antigen binding molecules "in combination" with the additional therapeutically active component).

[0202] The invention encompasses pharmaceutical compositions in which the antigen binding molecules of the invention are co-formulated with one or more additional therapeutically active ingredients as described elsewhere herein.

[0203] Administration regimen

[0204] According to certain embodiments of the present invention, multispecific antigen binding molecules can be applied to a subject within a limited time course. The method according to this aspect of the present invention includes sequentially applying multiple doses of the antigen binding molecules of the present invention to a subject. As used herein, "sequentially applying" means that the antigen binding molecules of each dose are applied to a subject at different time points, for example, at different dates separated by a predetermined interval (e.g., hours, days, weeks or months). The present invention includes methods, the methods including sequentially applying a single initial dose of antigen binding molecules to a patient, then applying one or more second doses of antigen binding molecules, and optionally subsequently applying one or more third doses of antigen binding molecules.

[0205] The terms "initial dose", "secondary dose" and "tertiary dose" refer to the time series of the antigen binding molecules of the present invention. Therefore, "initial dose" is the dosage (also referred to as "baseline dose") applied at the beginning of the treatment regimen; "secondary dose" is the dosage applied after the initial dose; "tertiary dose" is the dosage applied after the second dose. Initial dose, second dose and third dose can all contain the same amount of antigen binding molecules, but may generally be different from each other in terms of frequency of application. However, in certain embodiments, the amount of antigen binding molecules contained in initial dose, second dose and / or third dose during treatment is different from each other (for example, appropriately adjusted or reduced). In certain embodiments, two or more (for example, 2, 3, 4 or 5) dosages are applied as "loading doses" at the beginning of the treatment regimen, followed by subsequent doses (for example, "maintenance doses") applied on a less frequent basis.

[0206] In an exemplary embodiment of the present invention, each second dose and / or third dose is 1 to 26 (e.g., 1, 1 1 / 2,2,2 1 / 2,3,3 1 / 2,4,4 1 / 2,5,5 1 / 2,6,6 1 / 2,7,7 1 / 2,8,8 1 / 2,9,9 1 / 2, 10, 10 1 / 2, 11, 11 1 / 2, 12, 12 1 / 2, 13, 13 1 / 2, 14, 14 1 / 2, 15, 15 1 / 2, 16, 16 1 / 2, 17, 17 1 / 2, 18, 18 1 / 2, 19, 19 1 / 2, 20, 20 1 / 2, 21, 21 1 / 2, 22, 22 1 / 2, 23, 23 1 / 2, 24, 24 1 / 2, 25, 25 1 / 2, 26, 26 1 / 2 or more) weeks of administration. As used herein, the phrase "previous dose" means the dosage of the antigen binding molecules administered to the patient before the next dose in the sequence in the absence of an intermediate dose in a sequence of multiple administrations.

[0207] The method according to this aspect of the present invention can include administering to the patient any number of second and / or third doses of antigen binding molecules. For example, in certain embodiments, only a single second dose is administered to the patient. In other embodiments, two or more (e.g., 2, 3, 4, 5, 6, 7, 8 or more) second doses are administered to the patient. Similarly, in certain embodiments, only a single third dose is administered to the patient. In other embodiments, two or more (e.g., 2, 3, 4, 5, 6, 7, 8 or more) third doses are administered to the patient.

[0208] In embodiments involving multiple second doses, each second dose can be administered at the same frequency as other second doses. For example, each second dose can be administered to the patient 1 to 2 weeks after the previous dose. Similarly, in embodiments involving multiple third doses, each third dose can be administered at the same frequency as other third doses. For example, each third dose can be administered to the patient 2 to 4 weeks after the previous dose. Alternatively, the frequency of administering the second dose and / or the third dose to the patient can vary during the course of the treatment regimen. The frequency of administration can also be adjusted during the physician's treatment process according to the needs of individual patients after clinical examination.

[0209] Preferred implementation scheme:

[0210] 1. A multispecific antigen-binding molecule comprising:

[0211] (a) a first polypeptide comprising, from N-terminus to C-terminus, (i) a first antigen-binding domain that specifically binds to a T-cell antigen, (ii) a first multimerization domain, and (iii) a second antigen-binding domain that specifically binds to a T-cell antigen; and

[0212] (b) a second polypeptide, which comprises, from N-terminus to C-terminus, (i) a third antigen-binding domain that specifically binds to a target antigen and (ii) a second multimerization domain,

[0213] wherein said first multimerization domain and said second multimerization domain associate with each other to form a molecule.

[0214] 2. A molecule according to embodiment 1, wherein the second polypeptide further comprises a fourth antigen binding domain located at the C-terminus of the second multimerization domain.

[0215] 3. A molecule according to embodiment 2, wherein the fourth antigen binding domain specifically binds to a target antigen.

[0216] 4. A molecule according to embodiment 2, wherein the fourth antigen binding domain specifically binds to a T cell antigen.

[0217] 5. A molecule according to embodiment 3, wherein the third antigen binding domain and the fourth antigen binding domain specifically bind to different target antigens.

[0218] 6. The molecule of embodiment 5, wherein the different target antigens are expressed on the surface of the same cell.

[0219] 7. A molecule according to embodiment 3, wherein the third antigen binding domain and the fourth antigen binding domain specifically bind to the same target antigen.

[0220] 8. A molecule according to any one of embodiments 1 to 7, wherein the first antigen binding domain and the second antigen binding domain specifically bind to the same T cell antigen.

[0221] 9. A molecule according to any one of embodiments 1 to 7, wherein the first antigen binding domain and the second antigen binding domain specifically bind to different T cell antigens.

[0222] 10. A molecule according to embodiment 9, wherein the first antigen binding domain specifically binds to a first T cell antigen that is a co-stimulatory molecule, and the second antigen binding domain specifically binds to a second T cell antigen that is a checkpoint inhibitor.

[0223] 11. A molecule according to embodiment 10, wherein the co-stimulatory molecule is CD28 and the checkpoint inhibitor is PD-1.

[0224] 12. A molecule according to embodiment 4, wherein the first, second and fourth antigen binding domains specifically bind to the same T cell antigen.

[0225] 13. A molecule according to embodiment 4, wherein the first, second and fourth antigen binding domains bind different T cell antigens.

[0226] 14. A molecule according to embodiment 4, wherein the first and fourth antigen binding domains specifically bind to the same T cell antigen.

[0227] 15. A molecule according to embodiment 4, wherein the second and fourth antigen binding domains specifically bind to the same T cell antigen.

[0228] 16. The molecule of any one of embodiments 1 to 15, wherein one or more of the antigen binding domains is a Fab domain.

[0229] 17. The molecule of any one of embodiments 1 to 16, wherein one or more of said antigen binding domains is a scFv domain.

[0230] 18. A molecule according to embodiment 17, wherein the scFv domain comprises a heavy chain variable region (HCVR) comprising a cysteine ​​mutation at residue 44 and a light chain variable region comprising a cysteine ​​mutation at residue 100 (Kabat numbering).

[0231] 19. A molecule according to embodiment 17 or 18, wherein the scFv comprises a HCVR and a LCVR connected together by a polypeptide linker of 10 to 30 amino acids, optionally a (G4S)4 linker.

[0232] 20. A molecule according to embodiment 17 or 18, wherein the scFv is linked to the C-terminus of the first and / or second multimerization domain via a linker of 5 to 25 amino acids, optionally a (G4S)3 linker.

[0233] 21. The molecule of any one of embodiments 1 to 20, wherein the first antigen binding domain and the third antigen binding domain are Fab domains.

[0234] 22. The molecule of any one of embodiments 1 to 21, wherein the second antigen binding domain is a scFv domain.

[0235] 23. The molecule of any one of embodiments 2 to 22, wherein the fourth antigen binding domain is a scFv domain.

[0236] 24. The molecule of embodiment 1, wherein the first, second and third antigen binding domains are Fab domains.

[0237] 25. A molecule according to embodiment 1, wherein the first and third antigen binding domains are Fab domains and the second antigen binding domain is a scFv domain.

[0238] 26. A molecule according to embodiment 2, wherein the first, second, third and fourth antigen binding domains are Fab domains.

[0239] 27. A molecule according to embodiment 3, wherein the first, second, third and fourth antigen binding domains are Fab domains.

[0240] 28. A molecule according to embodiment 5, wherein the first and third antigen binding domains are Fab domains, and the second and fourth antigen binding domains are scFv domains.

[0241] 29. The molecule of embodiment 5, wherein the first, second, third and fourth antigen binding domains are Fab domains.

[0242] 30. A molecule according to embodiment 4, wherein the first and third antigen binding domains are Fab domains and the second and fourth antigen binding domains are scFv domains.

[0243] 31. A molecule according to embodiment 4, wherein the first, second, third and fourth antigen binding domains are Fab domains.

[0244] 32. A molecule according to any one of embodiments 9 to 11, wherein the first and third antigen binding domains are Fab domains and the second and fourth antigen binding domains are scFv domains.

[0245] 33. The molecule of any one of embodiments 9 to 11, wherein the first, second, third and fourth antigen binding domains are Fab domains.

[0246] 34. A molecule according to any one of embodiments 1 to 33, wherein the T cell antigen is a T cell receptor complex antigen.

[0247] 35. A molecule according to embodiment 34, wherein the T cell antigen is CD3.

[0248] 36. A molecule according to any one of embodiments 1 to 33, wherein the T cell antigen is a co-stimulatory molecule or a checkpoint inhibitor on a T cell.

[0249] 37. A molecule according to any one of embodiments 1 to 33, wherein the T cell antigen is selected from the group consisting of CD27, CD28, 4-1BB and PD-1.

[0250] 38. A molecule according to any one of embodiments 1 to 37, wherein the target antigen is a tumor-associated antigen.

[0251] 39. A molecule according to any one of embodiments 1 to 38, wherein the first and second multimerization domains are immunoglobulin Fc domains.

[0252] 40. A molecule according to embodiment 39, wherein the first and second multimerization domains are associated with each other via a disulfide bond.

[0253] 41. The molecule of any one of embodiments 1 to 40, wherein the first multimerization domain and the second multimerization domain are human IgG1 or human IgG4 Fc domains.

[0254] 42. A molecule according to any one of embodiments 39 to 41, wherein the first multimerization domain or the second multimerization domain comprises an amino acid substitution that reduces affinity for protein A binding compared to a wild-type Fc domain of the same isotype.

[0255] 43. A molecule according to embodiment 42, wherein the amino acid substitutions include H435R or H435R and Y436F modifications (EU numbering).

[0256] 44. A molecule according to embodiment 43, wherein the first multimerization domain includes the H435R and Y436F modifications.

[0257] 45. A molecule according to any one of embodiments 1 to 44, wherein the first polypeptide, the second polypeptide, or both the first polypeptide and the second polypeptide comprise a modified hinge domain, wherein the modified hinge domain has reduced binding affinity to an Fcγ receptor compared to a wild-type hinge domain of the same isotype.

[0258] 46. ​​A multispecific antigen-binding molecule comprising:

[0259] (a) a first polypeptide, which comprises, from N-terminus to C-terminus, (i) a first Fab that specifically binds to a T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a first scFv that specifically binds to a T cell antigen; and

[0260] (b) a second polypeptide, which comprises, from N-terminus to C-terminus, (i) a second Fab that specifically binds to the target antigen,

[0261] (ii) a second immunoglobulin Fc domain and (iii) a second scFv that specifically binds to a target antigen,

[0262] wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule.

[0263] 47. A molecule according to embodiment 46, wherein the second Fab and the second scFv specifically bind to different target antigens.

[0264] 48. A molecule according to embodiment 47, wherein the different target antigens are expressed on the surface of the same cell.

[0265] 49. A molecule according to embodiment 46, wherein the second Fab and the second scFv specifically bind to the same target antigen.

[0266] 50. A multispecific antigen-binding molecule comprising:

[0267] (a) a first polypeptide, which comprises, from N-terminus to C-terminus, (i) a first Fab that specifically binds to a T-cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a second Fab that specifically binds to a T-cell antigen; and

[0268] (b) a second polypeptide, wherein the second polypeptide comprises, from N-terminus to C-terminus, (i) a third Fab that specifically binds to the target antigen,

[0269] (ii) a second immunoglobulin Fc domain and (iii) a fourth Fab that specifically binds to a target antigen,

[0270] wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule.

[0271] 51. A molecule according to embodiment 50, wherein the third Fab and the fourth Fab specifically bind to different target antigens.

[0272] 52. The molecule of embodiment 51, wherein the different target antigens are expressed on the surface of the same cell.

[0273] 53. A molecule according to embodiment 50, wherein the third Fab and the fourth Fab specifically bind to the same target antigen.

[0274] 54. A multispecific antigen-binding molecule comprising:

[0275] (a) a first polypeptide, which comprises, from N-terminus to C-terminus, (i) a first Fab that specifically binds to a T cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a first scFv that specifically binds to a T cell antigen; and

[0276] (b) a second polypeptide, which comprises, from N-terminus to C-terminus, (i) a second Fab that specifically binds to the target antigen,

[0277] (ii) a second immunoglobulin Fc domain and (iii) a second scFv that specifically binds to a T cell antigen,

[0278] wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule.

[0279] 55. A multispecific antigen-binding molecule comprising:

[0280] (a) a first polypeptide, which comprises, from N-terminus to C-terminus, (i) a first Fab that specifically binds to a T-cell antigen, (ii) a first immunoglobulin Fc domain, and (iii) a second Fab that specifically binds to a T-cell antigen; and

[0281] (b) a second polypeptide, which comprises, from N-terminus to C-terminus, (i) a second Fab that specifically binds to a target antigen and (ii) a second immunoglobulin Fc domain,

[0282] wherein the first immunoglobulin domain and the second immunoglobulin domain are associated with each other through a disulfide bond to form a molecule.

[0283] 56. A molecule according to any one of embodiments 46 to 55, wherein the T cell antigen is a T cell receptor complex antigen.

[0284] 57. A molecule according to any one of embodiments 46 to 55, wherein the T cell antigen is CD3.

[0285] 58. A molecule according to any one of embodiments 46 to 55, wherein the T cell antigen is a co-stimulatory molecule or a checkpoint inhibitor on a T cell.

[0286] 59. A molecule according to any one of embodiments 46 to 55, wherein the T cell antigen is selected from the group consisting of CD27, CD28, 4-1BB and PD-1.

[0287] 60. The molecule of any one of embodiments 46 to 59, wherein the target antigen is a tumor-associated antigen.

[0288] 61. The molecule of any one of embodiments 46 to 60, wherein the first immunoglobulin Fc domain and the second immunoglobulin Fc domain are human IgG1 or human IgG4 Fc domains.

[0289] 62. A molecule according to embodiment 61, wherein the first immunoglobulin Fc domain or the second immunoglobulin Fc domain comprises amino acid substitutions that reduce affinity for protein A binding compared to a wild-type Fc domain of the same isotype.

[0290] 63. A molecule according to embodiment 62, wherein the amino acid substitutions include H435R or H435R and Y436F modifications (EU numbering).

[0291] 64. A molecule according to embodiment 63, wherein the first immunoglobulin Fc domain comprises the H435R and Y436F modifications.

[0292] 65. A molecule according to any one of embodiments 46 to 64, wherein the first polypeptide, the second polypeptide, or both the first polypeptide and the second polypeptide comprise a modified hinge domain, wherein the modified hinge domain has reduced binding affinity to an Fcγ receptor compared to a wild-type hinge domain of the same isotype.

[0293] 66. A molecule according to any one of embodiments 1 to 65, wherein the target antigen is a peptide in the context of the groove of a major histocompatibility complex (MHC) protein.

[0294] 67. A pharmaceutical composition comprising a molecule according to any one of embodiments 1 to 66 and a pharmaceutically acceptable carrier or diluent.

[0295] 68. A method of treating cancer comprising administering a molecule according to any one of embodiments 1 to 66 to a subject in need thereof.

[0296] 69. A method of treating an infection comprising administering to a subject in need thereof a molecule according to any one of embodiments 1 to 66.

[0297] 70. The method of embodiment 69, wherein the infection is a bacterial infection.

[0298] 71. The method of embodiment 69, wherein the infection is a viral infection.

[0299] 72. The method of embodiment 69, wherein the infection is a fungal infection.

[0300] 73. The method of embodiment 69, wherein the infection is a parasitic infection.

[0301] 74. The method of any one of embodiments 68 to 73, wherein the target antigen is present at a density of 10 to 10,000,000 copies per target cell, 100 to 1,000,000 copies per target cell, 100 to 20,000 copies per target cell, or 100 to 5000 copies per target cell.

[0302] 75. A method according to any one of embodiments 68 to 73, wherein the target antigen is present at a density of 1000 to 20,000 copies per target cell or greater than 20,000 copies per target cell.

[0303] 76. The method of any one of embodiments 68 to 75, wherein the molecule is administered in combination with a second therapeutic agent.

[0304] 77. A method according to embodiment 76, wherein the second therapeutic agent comprises a bispecific antigen binding molecule, which comprises a first antigen binding domain that binds to a target antigen (TA) and a second antigen binding domain that binds to a T cell antigen.

[0305] 78. A method according to embodiment 77, wherein the target antigen is a tumor cell antigen.

[0306] 79. The method of any one of embodiments 76 to 78, wherein the second therapeutic agent comprises a bispecific anti-TAx anti-CD28 antibody.

[0307] 80. The method of embodiment 79, wherein the second therapeutic agent comprises a bispecific anti-EGFR x anti-CD28 antibody.

[0308] 81. The method of embodiment 76, wherein the second therapeutic agent comprises an antibody that binds a checkpoint inhibitor on T cells.

[0309] 82. The method of embodiment 81, wherein the second therapeutic agent comprises an anti-PD-1 antibody.

[0310] 83. The method of any one of embodiments 76 to 82, wherein the molecule is administered in combination with two or more second therapeutic agents.

[0311] 84. Use of a molecule according to any one of embodiments 1 to 66 in the preparation of a medicament for treating cancer in a subject in need thereof.

[0312] 85. Use of a molecule according to any one of embodiments 1 to 66 for the preparation of a medicament for treating an infection in a subject in need thereof.

[0313] 86. The use according to embodiment 85, wherein the infection is a bacterial infection, a viral infection, a fungal infection or a parasitic infection.

[0314] 87. Use of a molecule according to any one of embodiments 1 to 66 in the treatment of cancer.

[0315] 88. Use of a molecule according to any one of embodiments 1 to 66 in treating an infection.

[0316] 89. The use according to embodiment 88, wherein the infection is a bacterial infection, a viral infection, a fungal infection or a parasitic infection.

[0317] 90. A molecule according to any one of embodiments 1 to 66 for use in medicine.

[0318] 91. A molecule according to any one of embodiments 1 to 66 for use in treating cancer.

[0319] 92. A molecule according to any one of embodiments 1 to 66 for use in treating an infection.

[0320] 93. The molecule of embodiment 92, wherein the infection is a bacterial infection, a viral infection, a fungal infection, or a parasitic infection.

[0321] Examples

[0322] The following examples are presented to provide a complete disclosure and description of how to prepare and use the methods and compositions of the present invention to those of ordinary skill in the art, and are not intended to limit the scope of what the inventors consider to be their invention. Efforts have been made to ensure accuracy regarding the numerals used (e.g., amounts, temperatures, etc.), but some experimental errors and deviations should be considered. Unless otherwise indicated, parts are parts by weight, molecular weight is average molecular weight, temperature is degrees Celsius, and pressure is or is close to atmospheric pressure.

[0323] Methods for binding by flow cytometry: In the following examples, the following flow cytometry method was used to determine the binding of various molecules. Flow cytometric analysis was used to determine the binding of the MAGEA4xCD3 multispecific molecule to RAJI / HLA-A2 / B2M / MAGEA4 (peptide a), A375 / hHLA-A2 / B2M / MAGEA4 (peptide b), RAJI / HLA-A2 / B2M / NY-ESO-1, and JURKAT, followed by detection with an APC-labeled anti-human IgG antibody. Briefly, 1x10 5 Cells / well were incubated at 4°C for 30 minutes with serial dilutions of MAGEA4xCD3 multispecific molecules or isotype controls (a human IgG4 stealth antibody that binds to human antigens and does not cross-react with human MAGEA4 or CD3). After incubation, the cells were washed twice with cold PBS containing 1% filtered FBS, and PE-conjugated anti-human secondary antibodies were added to the cells and incubated for another 30 minutes. Wells without antibodies or only with secondary antibodies were used as controls. After incubation, the cells were washed, resuspended in 200 μL of cold PBS containing 1% filtered FBS, and analyzed by flow cytometry on a BD FACS Canto II.

[0324] Methods used for cytotoxicity assays:In the following examples, the following cytotoxicity assays were used to determine the cytotoxicity of various molecules. To monitor the killing of MAGEA4+ cells in the presence of MAGEA4xCD3 as a single agent or in combination with EGFRxCD28 bispecific antibodies and / or PD-1 antibodies, A375 cells, ScaBER cells, NCI-H1755 metastatic (liver-derived) cells, and NCI-H1755 cells were labeled with 1 uM of the fluorescent tracking dye Violet Cell Tracker. After labeling, cells were plated overnight at 37°C. Separately, human PBMCs were plated at 1x10 6 Cells / mL were plated in supplemented RPMI medium and incubated overnight at 37°C to enrich lymphocytes by depleting adherent macrophages, dendritic cells and some monocytes. The next day, the target cells were co-incubated with the original PBMCs (effector / target cell 10:1 ratio) depleted of adherent cells, serial dilutions of MAGEA4xCD3 multispecific molecules and fixed concentrations of EGFRxCD28 and / or anti-PD1 antibodies at 37°C for 96 hours. Cells were removed from the cell culture plate using trypsin-EDTA dissociation buffer and analyzed by FACS on FACS BD LSRFortessa-X20. For FACS analysis, cells were stained with dead / alive near IR reaction (Invitrogen) dyes. 5E05 counting beads were added to each well immediately before FACS analysis. 1E05 beads were collected for each sample. In order to evaluate the specificity of killing, cells were gated on the population of live Violet markers. The percentage of the viable population was recorded and used to calculate survival.

[0325] Example 1: T cell activation depends on the presence of target cells

[0326] right Figure 1A , 1B T cell activation was assessed for each molecular format as shown in Figure 1C. T cell activation and upregulation of PD-1 markers were assessed by incubating cells with antibodies directly conjugated to CD2, CD4, CD8, CD25, and PD-1 and reporting the percentage of late activated (CD25+ / CD8+) T cells and PD-1+ / CD4+ T cells to total T cells (CD2+).

[0327] like Figure 2 As shown, exemplary multispecific molecules of the invention ( Figure 1C The "zero" represents a T cell only control.

[0328] Example 2: Cytotoxicity of multispecific molecules relative to conventional formats

[0329] Exemplary multispecific molecules of the invention ( Figure 1C structure) and compared it with a conventional form of a molecule with the same antigen binding domain ( Figure 1A and 1B ) cytotoxicity. The CD3 binding domain used in this example has a moderate binding affinity to human CD3. The target antigen binding domain used in this example binds to the MAGEA4 (melanoma associated antigen A4) peptide. The "control" is a positive control that targets a scaffold of all HLA molecules to provide maximum cytotoxicity compared to other formats.

[0330] like Figure 3 As shown, exemplary multispecific molecules of the invention ( Figure 1C structure) than molecules with conventional bispecific formats ( Figure 1A Structure and Figure 1B structure) to kill target cells more effectively.

[0331] Example 3: Cytotoxicity of multispecific molecules relative to conventional formats in combination with anti-PD-1 antibodies, co-stimulatory bispecific antibodies, or both

[0332] Exemplary multispecific molecules of the invention ( Figure 1C structure) and compared it with a conventional form of a molecule with the same antigen binding domain ( Figure 1A and 1B ) were compared with the cytotoxicity of anti-PD-1 antibody, co-stimulatory bispecific EGFRxCD28 antibody, or a combination of anti-PD-1 antibody and co-stimulatory bispecific EGFR x CD28 antibody. Positive control, CD3, and target antigen binding domains are as described in Example 2 above.

[0333] like Figure 4A , 4B As shown in Figures 4 and 4C, the addition of anti-PD-1 antibodies, co-stimulatory bispecific EGFR x CD28 antibodies, or both further enhances the exemplary multispecific molecules of the invention ( Figure 1C The solid line represents the cytotoxicity of a single agent (e.g. Figure 3 ), and the dotted lines represent the cytotoxicity of the corresponding combinations.

[0334] In addition to cytotoxicity, supernatants from assay wells of human PBMC assays were assessed for Th1 / Th2 cytokine release using the BD Cytometric Bead Array Human Kit and following the manufacturer's protocol. Figure 5 As shown, compared with conventional bispecific antibody formats ( Figure 1A Structure), compared to the exemplary multispecific molecule of the invention ( Figure 1CThe greater cytotoxicity of the construct) did not result in any greater cytokine release.

[0335] This set of experiments demonstrated that: (a) in the cytotoxicity assay, at the maximum concentration, Figure 1C The structure of the molecule showed a higher release rate than that of the Figure 1A Molecules with a stronger structure; (b) Figure 1C The cytotoxicity EC50 of the molecule (single agent) with Figure 1A (c) In a cytotoxicity assay, at the maximum concentration, the observed EC50 of a molecule with Figure 1A The structured molecules showed a higher release of cytokines than those with comparable levels when combined with anti-PD-1 antibodies. Figure 1A The molecules with the structure (anti-PD-1 combination) are more effective; (d) Figure 1C The cytotoxicity EC50 of the molecule with the structure combined with anti-PD-1 antibody was lower than that of the Figure 1A The observed EC50 of the molecule with the structure (anti-PD-1 combination); (e) in the cytotoxicity assay, at the maximum concentration, Figure 1A The structured molecules showed comparable levels of cytokine release when combined with the anti-EGFR xCD28 bispecific antibody Figure 1A The structure of the molecule (anti-EGFR x CD28 combination) is more potent; (f) Figure 1C The cytotoxicity EC50 of the molecule with the structure combined with the anti-EGFR x CD28 bispecific antibody was lower than that of the Figure 1A The observed EC50 for the molecule with the structure (anti-EGFR x CD28 combination); (g) in the cytotoxicity assay, at the maximum concentration, Figure 1A The structured molecules showed comparable levels of cytokine release when combined with anti-PD-1 antibody and anti-EGFR xCD28 bispecific antibody. Figure 1A The structure of the molecule (triplet combination) is more effective; (h) has Figure 1C The cytotoxicity EC50 of the molecule with the structure combined with anti-PD-1 antibody and anti-EGFR x CD28 bispecific antibody was lower than that of the Figure 1A EC50 observed for the structured molecules (triplet combinations).

[0336] Example 4: Potency of multispecific molecules enhanced by two effector binding domains

[0337] Exemplary multispecific molecules ( Figure 1C The binding of the 5′-HRP construct to target cells overexpressing the MAGEA4 peptide and to CD3+ Jurkat cells was also evaluated. Figure 1C Modification of the structure in which one or more antigen binding domains are inactivated. Inactivated domains are indicated by "X" in the legend.

[0338] like Figures 6A-6D As shown, the binding data show that the combination of two antigen binding domains (e.g., a single Fab and a single scFv) binds to target cells with greater affinity (lower EC50) than the binding of a molecule with a single Fab domain or a molecule with a single scFv domain. As expected, the isotype control molecule showed no binding. No difference in binding pattern was observed regardless of the source of the anti-CD3 binding domain.

[0339] In addition to binding, the cytotoxicity of these molecules was also determined using the above methods. Fig. 7A and 7B As shown, exemplary multispecific molecules of the invention ( Figure 1C The structure) showed the greatest cytotoxic potency, followed by two modified molecules that included two T cell antigen (CD3) binding domains but only a single target antigen (MAGEA4) binding domain (scFv or Fab). Likewise, the same cytotoxic pattern was observed regardless of the source of the anti-CD3 binding domain. The negative control ( Figure 1A format) includes an unrelated target antigen binding domain.

[0340] Example 5: C-terminal scFv domains enhance the potency of multispecific molecules relative to C-terminal Fab domains

[0341] Exemplary multispecific molecules ( Figure 1C The binding of the 5′-HRP construct to target cells overexpressing the MAGEA4 peptide and to CD3+ Jurkat cells was also evaluated. Figure 1C Modification of the structure to use the Fab domain ( Figure 1E The C-terminal scFv domain is replaced by a C-terminal scFv domain (structure), or the N-terminal Fab domain is inactivated in the modification. The inactivated domain is indicated by "X" in the legend.

[0342] Similar to the combination discussed in Example 4, and as Fig. 8A and 8B As shown, the binding data show that the combination of two antigen binding domains (e.g., a single Fab and a single scFv, or two Fabs) binds to target cells with greater affinity (lower EC50) compared to the binding of a molecule with a single Fab domain or a molecule with a single scFv domain. Fig. 8A and 8B As shown in the table, Figure 1C and Figure 1EThe molecules of the structures bound with comparable binding titers.

[0343] In addition to binding, the cytotoxicity of these molecules was also determined using the above methods. Fig. 9 As shown, exemplary multispecific molecules of the invention ( Figure 1C structure) showed the greatest cytotoxic potency, followed by a modified molecule including a C-terminal Fab domain in place of the two scFv domains.

[0344] Example 6: Single-chain bivalent T cell antigens enhance the potency of multispecific molecules relative to multi-chain bivalents

[0345] Exemplary multispecific molecules ( Figure 1C The binding of the 5′-HRP-positive glioma cell line to target cells overexpressing the MAGEA4 peptide and to CD3+ Jurkat cells was also evaluated. Figure 1D Binding to these cells of a molecule of the structure shown, in which the MAGEA4 binding domain and the CD3 binding domain are swapped so that the two sets of antigen binding domains are located on two separate polypeptide chains.

[0346] like Fig. 10A and 10B As shown, the binding data showed similar binding of the two molecular structures to each of the two cell types.

[0347] In addition to binding, the cytotoxicity of these molecules was also determined using the above methods. Fig.11A and 11B As shown, exemplary multispecific molecules of the invention ( Figure 1C Structure) relative to having Figure 1D The structured molecules showed greater cytotoxic potency, thus confirming that the presence of two T cell antigen binding domains on a single polypeptide chain provides enhanced cytotoxic potency.

[0348] Example 7: Relative cytotoxicity of multispecific molecules targeting one or two antigens relative to conventional formats alone or in combination with anti-PD-1 antibodies and co-stimulatory bispecific antibodies

[0349] Two exemplary multispecific molecules of the invention ( Figure 1C and Figure 1F structure) and combined it with conventional formats of molecules ( Figure 1A ) were compared to the cytotoxicity of the cells. This example uses a positive control with higher specificity than the control used in the previous example to show that Figure 1C and 1FThe structure of the molecules, and the greater difference between these molecules and the combination of co-stimulatory bispecific antibodies and anti-PD-1 antibodies. The CD3 antigen binding domain used in this example has a strong binding affinity to human CD3, and the target antigen binding domain (MAGEA4a) is as described in Example 2 above. Negative control ( Figure 1A The target antigen binding domain is not included in the form of Figure 1F The second target antigen binding domain (MAGEA4b) of the structured molecule binds to a completely different epitope of MAGEA4 than the epitope bound by the first target antigen binding domain.

[0350] like Fig. 12A and 12B As shown, multispecific molecules targeting two different low-density antigens on tumor cells showed increased efficacy compared to multispecific molecules targeting only a single tumor antigen, and both molecules showed Figure 1A The addition of anti-PD-1 antibody and co-stimulatory bispecific EGFR x CD28 antibody further enhances the exemplary multispecific molecule of the present invention ( Figure 1C and 1F structure).

[0351] Example 8: Relative cytotoxicity of multispecific molecules is related to the affinity of the T cell antigen binding domain

[0352] Preparation of anti-CD3 binding domains with different affinities Figure 1F Exemplary multispecific molecules of structure (eg Fig.13 Five molecules were prepared according to the following parameters:

[0353] Molecule A with CD3 arms 7195P (strong) fab and 7195P (strong) scfv;

[0354] Molecule B with CD3 arms 7221G (medium) fab and 7221G (medium) scfv;

[0355] Molecule C with CD3 arms 7221G20(weak)fab and 7221G20(weak)scfv;

[0356] Molecule D with CD3 arms 7221G20 (weak) fab and 7221G (medium) scFv; and

[0357] Molecule E with CD3 arms 7221G (medium) fab and 7195P (strong) scFv.

[0358] The binding titration range of these five molecules to T cells was tested by flow cytometry and correlated with the strength of the CD3 binding domain, as Fig.13 Indicated relative to isotype control.

[0359] In cytotoxicity assays targeting two different MAGEA4+ cell lines (A375 and ScaBER), the potency of the molecule was shown to decrease when the strength of the effector arm (e.g., anti-CD3 binding domain) was reduced, either as a single agent, or in combination with an EGFRxCD28 bispecific antibody and an anti-PD1 antibody, as Fig.14A , 14B , 15A and 15B. Each molecule contains the same target antigen binding domain (with non-overlapping MAGEA4 peptide 1 and MAGEA4 peptide 2).

[0360] Example 9: Relative cytotoxicity of multispecific molecules targeting two antigens relative to conventional formats alone or in combination with anti-PD-1 antibodies and co-stimulatory bispecific antibodies

[0361] Three exemplary multispecific molecules of the invention ( Figure 1C and Figure 1F structure) and combined it with conventional formats of molecules ( Figure 1A ) were used for comparison. Figure 1A Positive controls for structures that bind CD3 and HLA. The CD3 antigen binding domain used in this example has a strong binding affinity for human CD3 (derived from 7195P), and the target antigen binding domain is directed against one or two non-overlapping MAGEA4 (melanoma-associated antigen A4) peptides (MAGEA4Aa and MAGEA4b) or a peptide against NY-ESO-1 (New York esophageal squamous cell carcinoma 1). Two isotype negative controls ( Figure 1A Form and Figure 1C format) that includes an unrelated target antigen binding domain.

[0362] like Fig.16A , 16B As shown in Figures 16C and 16C, the molecules bound to cells expressing NY-ESO-1, MAGEA4a, or MAGEA4b by flow cytometry, as expected.

[0363] like Fig.17A and 17B As shown, multispecific molecules targeting two different antigens (molecule A) or two different epitopes of a single antigen (molecule B) effectively induced metastatic non-small cell lung cancer (NSCLC) cells ( Fig.17A ) and NSCLC cells ( Fig. 17B ), wherein the multispecific molecule targeting two different antigens (molecule A) showed increased potency relative to the multispecific molecule targeting two different epitopes of the same antigen (molecule B). The addition of anti-PD-1 antibody and co-stimulatory bispecific EGFR x CD28 antibody further enhanced the cytotoxicity of the exemplary multispecific molecule of the invention ( Figure 1F The relative induction of T cell activation by these molecules was also assessed, as Fig. 17C (in metastatic NSCLC cells) and 17D (in NSCLC cells).

[0364] Will target one or two antigens (different epitopes or different antigens) and have Figure 1C and 1F The relative cytotoxic activity and potency of the multispecific molecules of the structure were compared with the conventional format of the molecule ( Figure 1A ) were compared with the cytotoxicity of the positive control and isotype control as described in this example. Fig.18A and 18B As shown, multispecific molecules are more effective than conventional formats and target two different epitopes ( Fig.18A ) or two different antigens ( Fig.18B ) are more effective than multispecific molecules that target the same antigen at both target antigen binding domains. The relative induction of T cell activation by these molecules is shown in Fig.18C , 18D , 18E and 18F.

[0365] Example 10: Relative cytotoxicity of a multispecific molecule targeting two antigens relative to a combination of molecules targeting the same antigen in conventional form

[0366] An exemplary multispecific molecule of the invention targeting two different antigens was measured as described above. Figure 1F structure) and compared it with a conventional form of the molecule ( Figure 1A The cytotoxicity of the conventional format of the molecule was compared with that of the combination of an anti-PD-1 antibody and a co-stimulatory bispecific EGFR x CD28 antibody.

[0367] Cytotoxicity assays targeted MAGEA4-expressing-SCaBER cells (bladder) and demonstrated that multispecific molecules targeting MAGEA4a and MAGEA4b (which are non-overlapping peptides of MAGEA4) were more effective than combinations of conventional bispecific antibodies targeting the same two MAGEA4 peptides, such as Fig.19AThe addition of anti-PD-1 antibody and co-stimulatory bispecific EGFR xCD28 antibody further enhances the exemplary multispecific molecule of the present invention ( Figure 1F The relative induction of T cell activation by these same molecules is shown in Fig.19B middle.

[0368] The present invention is not limited to a certain scope by the specific embodiments described herein. In fact, various modifications of the present invention, in addition to those described herein, will become apparent to those skilled in the art from the foregoing description. Such modifications are intended to fall within the scope of the appended claims.

Claims

1. A multispecific antigen-binding molecule comprising: (a) a first polypeptide comprising, from N-terminus to C-terminus, (i) a first antigen-binding domain that specifically binds to a T-cell antigen, (ii) a first multimerization domain, and (iii) a second antigen-binding domain that specifically binds to a T-cell antigen; and (b) a second polypeptide, which comprises, from N-terminus to C-terminus, (i) a third antigen-binding domain that specifically binds to a target antigen and (ii) a second multimerization domain, wherein said first multimerization domain and said second multimerization domain associate with each other to form a molecule.

2. The molecule of claim 1, wherein the first antigen binding domain and the second antigen binding domain specifically bind to the same T cell antigen.

3. The molecule of claim 1, wherein the first antigen binding domain and the second antigen binding domain specifically bind to different T cell antigens.

4. The molecule of claim 3, wherein the first antigen binding domain specifically binds a first T cell antigen that is a co-stimulatory molecule, and the second antigen binding domain specifically binds a second T cell antigen that is a checkpoint inhibitor.

5. The molecule of claim 4, wherein the co-stimulatory molecule is CD28 and the checkpoint inhibitor is PD-1.

6. The molecule of any one of claims 1 to 5, wherein one or more of the antigen binding domains is a Fab domain.

7. The molecule according to any one of claims 1 to 6, wherein the T cell antigen is a T cell receptor complex antigen.

8. The molecule of claim 7, wherein the T cell antigen is CD3.

9. The molecule according to any one of claims 1 to 6, wherein the T cell antigen is a co-stimulatory molecule or a checkpoint inhibitor on T cells.

10. The molecule according to any one of claims 1 to 6, wherein the T cell antigen is selected from the group consisting of CD27, CD28, 4-1BB and PD-1.

11. The molecule according to any one of claims 1 to 10, wherein the target antigen is a tumor-associated antigen.

12. The molecule of any one of claims 1 to 11, wherein the first and second multimerization domains are immunoglobulin Fc domains.

13. The molecule of claim 12, wherein the first and second multimerization domains are associated with each other via a disulfide bond.

14. The molecule according to any one of claims 1 to 13, wherein the first multimerization domain and the second multimerization domain are human IgGl or human IgG4 Fc domains.

15. The molecule of any one of claims 12 to 14, wherein the first multimerization domain or the second multimerization domain comprises an amino acid substitution that reduces affinity for protein A binding compared to a wild-type Fc domain of the same isotype.

16. The molecule of claim 15, wherein the amino acid substitutions comprise H435R or H435R and Y436F modifications (EU numbering).

17. The molecule of any one of claims 1 to 16, wherein the first polypeptide, the second polypeptide, or both the first polypeptide and the second polypeptide comprise a modified hinge domain that has reduced binding affinity to an Fcγ receptor compared to a wild-type hinge domain of the same isotype.

18. A pharmaceutical composition comprising a molecule according to any one of claims 1 to 17 and a pharmaceutically acceptable carrier or diluent.

19. Use of a multispecific antigen-binding molecule in the preparation of a medicament for treating cancer, the multispecific antigen-binding molecule comprising: (a) a first polypeptide comprising, from N-terminus to C-terminus, (i) a first antigen-binding domain that specifically binds to a T-cell antigen, (ii) a first multimerization domain, and (iii) a second antigen-binding domain that specifically binds to a T-cell antigen; and (b) a second polypeptide, which comprises, from N-terminus to C-terminus, (i) a third antigen-binding domain that specifically binds to a target antigen and (ii) a second multimerization domain, wherein said first multimerization domain and said second multimerization domain associate with each other to form a molecule.

20. Use of a multispecific antigen-binding molecule in the preparation of a medicament for treating infection, the multispecific antigen-binding molecule comprising: (a) a first polypeptide comprising, from N-terminus to C-terminus, (i) a first antigen-binding domain that specifically binds to a T-cell antigen, (ii) a first multimerization domain, and (iii) a second antigen-binding domain that specifically binds to a T-cell antigen; and (b) a second polypeptide, which comprises, from N-terminus to C-terminus, (i) a third antigen-binding domain that specifically binds to a target antigen and (ii) a second multimerization domain, wherein said first multimerization domain and said second multimerization domain associate with each other to form a molecule.

Images