Multispecific polypeptide constructs containing constrained CD3 binding domains and receptor binding regions and methods of using same - Patents.com

Abstract

The present invention generally relates to multispecific polypeptides that bind to at least CD3, a second antigen, and a T cell receptor, such as a costimulatory receptor or an inhibitory receptor, where the multispecific polypeptide is capable of engaging CD3. In some embodiments, the multispecific polypeptide construct binds to a costimulatory receptor, providing costimulatory binding activity. In some embodiments, the multispecific polypeptide construct binds to an inhibitory receptor, blocking inhibitory activity. In some aspects, the multispecific polypeptide has constrained CD3 binding, binding or engaging CD3 only upon binding to a second antigen, such as a tumor-associated antigen. In some embodiments, the multispecific polypeptide contains a cleavable linker that, when cleaved, provides dual effector function. Methods of making and using these multispecific polypeptides in a variety of therapeutic, diagnostic, and prophylactic applications are also provided.

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application is related to U.S. Provisional Application No. 62 / 702,888, filed July 24, 2018, entitled "Multispecific Polypeptide Constructs Containing a Constrained CD3 Binding Domain and a Costimulatory Receptor Binding Region and Methods of Using the Same"; U.S. Provisional Application No. 62 / 744,641, filed October 11, 2018, entitled "Multispecific Polypeptide Constructs Containing a Constrained CD3 Binding Domain and a Costimulatory Receptor Binding Region and Methods of Using the Same"; and U.S. Provisional Application No. 62 / 744,641, filed October 11, 2018, entitled "Multispecific Polypeptide Constructs Containing a Constrained CD3 Binding Domain and a Costimulatory Receptor Binding Region and Methods of Using the Same." This application claims priority to U.S. Provisional Application No. 62 / 832,268, filed April 10, 2019, entitled "METHODS OF USING THE SAME," the contents of each of which are incorporated by reference in their entirety.

[0002] INCORPORATION-BY-REFERENCE TO SEQUENCE LISTING This application has been submitted with a Sequence Listing in electronic format. The Sequence Listing is provided as a file entitled 744952000840SeqList.TXT, created on July 23, 2019, and having a size of 311 kilobytes. The information in the electronic format of the Sequence Listing is incorporated by reference in its entirety.

[0003] Field The present invention generally relates to multispecific polypeptides that bind to at least CD3, a second antigen, and a T cell receptor, such as a costimulatory receptor or an inhibitory receptor, where the multispecific polypeptide is capable of engaging CD3. In some embodiments, the multispecific polypeptide construct binds to a costimulatory receptor, providing costimulatory binding activity. In some embodiments, the multispecific polypeptide construct binds to an inhibitory receptor, blocking inhibitory activity. In some aspects, the multispecific polypeptide has constrained CD3 binding, binding or engaging CD3 only upon binding to a second antigen, such as a tumor-associated antigen. In some embodiments, the multispecific polypeptide construct contains a cleavable linker that, when cleaved, provides dual effector function. Methods of making and using these multispecific polypeptides in a variety of therapeutic, diagnostic, and prophylactic applications are also provided. [Background technology]

[0004] background Therapeutic antibodies that cause target cell depletion generally rely on effector function mediated through interactions with Fcγ receptors (FcγR) and complement proteins. FcγR-expressing effector cells are primarily those of the innate immune system. T cells are not the direct effector cells involved in antibody-mediated target cell depletion.

[0005] The CD3 (cluster of differentiation 3) T-cell coreceptor is a multimeric protein composed of four distinct polypeptide chains called the epsilon, gamma, delta, and zeta chains. The CD3 complex serves as the signaling module of the T-cell receptor (TCR), noncovalently associating with the antigen-binding a / b chains of the TCR.

[0006] Direct engagement of CD3 leads to T cell activation and is therefore a desirable target for a variety of therapeutic and / or diagnostic applications. Thus, there is a need for antibodies and therapeutic agents that target the CD3 / TCR pathway. Summary of the Invention

[0007] overview Provided herein is a multispecific polypeptide construct comprising a first component comprising an immunoglobulin Fc region and a second component comprising a CD3 binding region, wherein the first and second components are coupled by a linker, and the Fc region is located N-terminal to the CD3 binding region; one or both of the first and second components comprises at least one antigen binding domain that binds to a tumor-associated antigen (TAA); and one or both of the first and second components comprises at least one costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor. In some embodiments, the CD3 binding region binds to CD3 (CD3ε). In some aspects, the multispecific polypeptide construct comprises at least or about two antigen binding domains that bind to a TAA and a CRBR. Exemplary constructs are described herein.

[0008] Also provided herein is a multispecific polypeptide construct comprising a first component comprising an immunoglobulin Fc region and a second component comprising a CD3-binding region, wherein the first and second components are coupled by a linker, and the Fc region is located N-terminal to the CD3-binding region; one or both of the first and second components comprises at least one antigen-binding domain that binds to a tumor-associated antigen (TAA); and one or both of the first and second components comprises at least one inhibitory receptor-binding region (IRBR) that binds to an inhibitory receptor. In some embodiments, the CD3-binding region binds to CD3 (CD3ε). In some aspects, the multispecific polypeptide construct contains at least or about two antigen-binding domains that bind to a TAA and an IRBR. Exemplary constructs are described herein.

[0009] Also provided herein is a multispecific polypeptide construct comprising a first component comprising an immunoglobulin Fc region and a second component comprising a CD3-binding region, wherein the first and second components are coupled by a linker, and the Fc region is located N-terminal to the CD3-binding region; one or both of the first and second components comprises at least one antigen-binding domain that binds to a tumor-associated antigen (TAA); one or both of the first and second components comprises at least one inhibitory receptor-binding region (IRBR) that binds to an inhibitory receptor; and one or both of the first and second components comprises at least one costimulatory receptor-binding region (CRBR) that binds to a costimulatory receptor. In some embodiments, the CD3-binding region binds to CD3 (CD3ε). In some aspects, the multispecific polypeptide construct contains at least or about two antigen-binding domains that bind to a TAA, a CRBR, and an IRBR. Exemplary constructs are described herein.

[0010] In some embodiments, at least one antigen binding domain is positioned amino terminal to the Fc region and / or carboxy terminal to the CD3 binding region of the multispecific polypeptide construct. In some embodiments, at least one costimulatory receptor binding region (CRBR) is positioned amino terminal to the Fc region and / or carboxy terminal to the CD3 binding region of the multispecific polypeptide construct.

[0011] In some embodiments, the first component comprises a first antigen-binding domain and the second component comprises a second antigen-binding domain, each of which binds to a tumor-associated antigen (TAA). In some embodiments, the first antigen-binding domain is positioned amino-terminal to the Fc region of the multispecific construct, and the second antigen-binding domain is positioned carboxy-terminal to the CD3-binding region of the multispecific construct. In some embodiments, the first or second component further comprises a costimulatory receptor binding region (CRBR).

[0012] Provided herein is a multispecific polypeptide construct comprising, in order from N-terminus to C-terminus, a costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor and / or an antigen-binding domain that binds to a tumor-associated antigen (TAA); an immunoglobulin Fc region; a linker; a CD3-binding region that binds to CD3 (CD3ε); and a costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor and / or an antigen-binding domain that binds to a tumor-associated antigen (TAA), wherein the multispecific polypeptide construct comprises at least one CRBR and at least one antigen-binding domain. In some embodiments, the multispecific polypeptide construct contains only one costimulatory receptor binding region (CRBR). In some embodiments, the multispecific polypeptide construct contains two antigen-binding domains that bind to a TAA. In some embodiments, the antigen-binding domains bind to the same tumor-associated antigen (TAA). In some embodiments, one antigen-binding domain is located amino-terminal to the Fc region and one antigen-binding domain is located carboxy-terminal to the CD3-binding region.

[0013] Provided herein is a multispecific polypeptide construct comprising, from N-terminus to C-terminus, an immunoglobulin Fc region; a linker; a CD3-binding region that binds to CD3 (CD3ε); and an antigen-binding domain that binds to a tumor-associated antigen (TAA) and a costimulatory receptor-binding region (CRBR) that binds to a costimulatory receptor. Provided herein is a multispecific polypeptide construct comprising, from N-terminus to C-terminus, an antigen-binding domain that binds to a tumor-associated antigen (TAA) and a costimulatory receptor-binding region (CRBR) that binds to a costimulatory receptor; an immunoglobulin Fc region; a linker; and a CD3-binding region that binds to CD3 (CD3ε).

[0014] In some such embodiments, the Fc region is a homodimeric Fc region. In some examples, the Fc region is the Fc region of human IgG1, human IgG2, human IgG3, or human IgG4, or an immunologically active fragment thereof. In some cases, the Fc region comprises a polypeptide comprising the amino acid sequence set forth in SEQ ID NO:1, or a sequence of amino acids having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to SEQ ID NO:1. In some embodiments, the Fc region comprises a polypeptide comprising the amino acid sequence set forth in SEQ ID NO:2, or a sequence of amino acids having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to SEQ ID NO:2; the Fc region comprises a polypeptide comprising the amino acid sequence set forth in SEQ ID NO:4, or a sequence of amino acids having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to SEQ ID NO:4; or the Fc region comprises a polypeptide comprising the amino acid sequence set forth in SEQ ID NO:5, or a sequence of amino acids having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to SEQ ID NO:5.

[0015] In some embodiments, the Fc region is a heterodimeric Fc region. In some embodiments, one or both Fc polypeptides of the heterodimeric Fc region comprise at least one modification to induce heterodimerization compared to the polypeptide of the homodimeric Fc region, optionally compared to the Fc polypeptide set forth in SEQ ID NO:1 or an immunologically active fragment thereof. In some embodiments, each of the Fc polypeptides of the heterodimeric Fc independently comprises at least one amino acid modification. In some embodiments, each of the Fc polypeptides of the heterodimeric Fc comprises a knob-into-hole modification or a charge mutation to increase electrostatic complementarity of the polypeptides. In some cases, the amino acid modification is a knob-into-hole modification.

[0016] In some embodiments, the first Fc polypeptide of the heterodimeric Fc comprises a modification selected from Thr366Ser, Leu368Ala, Tyr407Val, and a combination thereof, and the second Fc polypeptide of the heterodimeric Fc comprises the modification Thr366Trp. In some embodiments, the first and second Fc polypeptides further comprise a modification of a non-cysteine ​​residue to a cysteine ​​residue, wherein the modification of the first polypeptide is at one of positions Ser354 and Tyr349, and the modification of the second Fc polypeptide is at the other of positions Ser354 and Tyr349. In some embodiments, the amino acid modification is a charge mutation to increase electrostatic complementarity of the polypeptides.

[0017] In some embodiments, each of the first and / or second Fc polypeptides, or the first and second Fc polypeptides, comprises a modification at a complementary position, where the modification is a substitution of an amino acid having the opposite charge to the complementary amino acid of the other polypeptide. In some embodiments, one of the first or second Fc polypeptides of the heterodimeric Fc further comprises a modification at residue Ile253. In some cases, the modification is Ile253Arg. In some instances, one of the first or second Fc polypeptides of the heterodimeric Fc further comprises a modification at residue His435. In some instances, the modification is His435Arg. In some embodiments, the Fc region comprises a polypeptide lacking Lys447.

[0018] In some embodiments, the Fc region comprises a polypeptide containing at least one modification to enhance FcRn binding. In some instances, the modification is at a position selected from the group consisting of Met252, Ser254, Thr256, Met428, Asn434, and combinations thereof. In some embodiments, the modification is at a position selected from the group consisting of Met252Y, Ser254T, Thr256E, Met428L, Met428V, Asn434S, and combinations thereof. In some embodiments, the modification is at positions Met252 and Met428. In some embodiments, the modification is Met252Y and Met428L. In some embodiments, the modification is Met252Y and Met428V.

[0019] In some embodiments, the first polypeptide of the heterodimeric Fc comprises a sequence of amino acids set forth in any of SEQ ID NOs: 82, 86, 94, or 96, and the second polypeptide of the heterodimeric Fc comprises a sequence of amino acids set forth in any of SEQ ID NOs: 83, 87, 90, 92, 98, or 100. In some embodiments, the Fc region comprises a polypeptide comprising at least one amino acid modification that reduces effector function and / or reduces binding to an effector molecule selected from an Fcγ receptor or C1q. In some embodiments, the one or more amino acid modifications are deletions of one or more of Glu233, Leu234, or Leu235. In some embodiments, the first polypeptide of the heterodimeric Fc comprises a sequence of amino acids set forth in any of SEQ ID NOs: 84, 88, 95, or 97, and the second polypeptide of the heterodimeric Fc comprises a sequence of amino acids set forth in any of SEQ ID NOs: 85, 89, 91, 93, 99, or 101. In some embodiments, the Fc region comprises a polypeptide comprising at least one modification to enhance FcγR binding. In some embodiments, the modification is at Ser239 or Ile332. In some embodiments, the glycosylation of the Fc region is modified to enhance FcγR binding compared to an unmodified Fc region. In some embodiments, the Fc region lacks fucose or has a reduced fucose content.

[0020] In some such embodiments, the CD3 binding region is an anti-CD3 antibody or antigen-binding fragment. In some embodiments, the anti-CD3 antibody or antigen-binding fragment comprises a variable heavy chain region (VH) and a variable light chain region (VL). In some embodiments, the CD3 binding region is monovalent. In some embodiments, the CD3 binding region is a variable fragment (Fv) comprising a variable heavy chain region (VH) and a variable light chain region (VL). In some embodiments, the anti-CD3 antibody or antigen-binding fragment is not a single-chain antibody, and optionally is not a single-chain variable fragment (scFv). In some embodiments, the Fc is a heterodimeric Fc, wherein the VH and VL comprising the anti-CD3 antibody or antigen-binding fragment are linked to opposing polypeptides of the heterodimeric Fc.

[0021] In some embodiments, the CD3 binding region is incapable or substantially incapable of binding or engaging CD3 unless at least one of the antigen binding domains binds to its TAA, hi some embodiments, the CD3 binding region is incapable or substantially incapable of binding or engaging CD3 unless at least two of the antigen binding domains bind to its TAA.

[0022] In some embodiments, the linker is a polypeptide linker. In some examples, the linker is a polypeptide up to 25 amino acids in length. In some embodiments, the linker is 2 to 24 amino acids, 2 to 20 amino acids, 2 to 18 amino acids, 2 to 14 amino acids, 2 to 12 amino acids, 2 to 10 amino acids, 2 to 8 amino acids, 2 to 6 amino acids, 6 to 24 amino acids, 6 to 20 amino acids, 6 to 18 amino acids, 6 to 14 amino acids, 6 to 12 amino acids, 6 to 10 amino acids, 6 to 8 amino acids, 8 to 24 amino acids, 8 to 20 amino acids, 8 to 18 amino acids, 8 to 14 amino acids, 8 to 12 amino acids, 8 to 10 amino acids, 10 to 24 amino acids, 10 to 20 amino acids, 10 to 18 amino acids, 10 to 14 amino acids, 10 to 12 amino acids, 12 to 24 amino acids, 12 to 20 amino acids, 12 to 18 amino acids, 12 to 14 amino acids, 14 to 24 amino acids, 14 to 20 amino acids, 14 to 18 amino acids, 18 to 24 amino acids, 18 to 20 amino acids, or 20 to 24 amino acids. or a polypeptide of about 2 to 24 amino acids, 2 to 20 amino acids, 2 to 18 amino acids, 2 to 14 amino acids, 2 to 12 amino acids, 2 to 10 amino acids, 2 to 8 amino acids, 2 to 6 amino acids, 6 to 24 amino acids, 6 to 20 amino acids, 6 to 18 amino acids, 6 to 14 amino acids, 6 to 12 amino acids, 6 to 10 amino acids, 6 to 8 amino acids, 8 to 24 amino acids, 8 to 20 amino acids, 8 to 18 amino acids, 8 to 14 amino acids, 8 to 12 amino acids, 8 to 10 amino acids, 10 to 24 amino acids, 10 to 20 amino acids, 10 to 18 amino acids, 10 to 14 amino acids, 10 to 12 amino acids, 12 to 24 amino acids, 12 to 20 amino acids, 12 to 18 amino acids, 12 to 14 amino acids, 14 to 24 amino acids, 14 to 20 amino acids, 14 to 18 amino acids, 18 to 24 amino acids, 18 to 20 amino acids, or 20 to 24 amino acids. In some embodiments, the linker is a polypeptide that is 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acids in length. In some cases, the linker is a polypeptide that is 3 to 18 amino acids in length.In some embodiments, the linker is a polypeptide that is 12 to 18 amino acids in length. In some embodiments, the linker is a polypeptide that is 15 to 18 amino acids in length.

[0023] In some embodiments, the linker is a non-cleavable linker. In some embodiments, the non-cleavable linker does not contain a substrate recognition site that is specifically recognized for cleavage by a protease. In some embodiments, the non-cleavable linker comprises GS, GGS, GGGGS (SEQ ID NO: 149), GGGGGS (SEQ ID NO: 135), and combinations thereof. In some embodiments, the non-cleavable linker comprises (GGS)n (n is 1 to 10). In some embodiments, the non-cleavable linker comprises (GGGGS)n (SEQ ID NO: 173) (n is 1 to 10). In some embodiments, the non-cleavable linker comprises (GGGGGS)n (SEQ ID NO: 172) (n is 1 to 4). In some embodiments, the non-cleavable linker comprises GGS. In some embodiments, the non-cleavable linker comprises GGGGS (SEQ ID NO: 149). In some embodiments, the non-cleavable linker comprises GGGGGS (SEQ ID NO: 135). In some embodiments, the non-cleavable linker comprises (GGS) (SEQ ID NO: 10). In some embodiments, the non-cleavable linker comprises GGSGGSGGS (SEQ ID NO: 11). In some embodiments, the non-cleavable linker comprises GGSGGSGGSGGS (SEQ ID NO: 12). In some embodiments, the non-cleavable linker comprises In some embodiments, the non-cleavable linker comprises: In some embodiments, the non-cleavable linker comprises: In some embodiments, the non-cleavable linker comprises: Includes TIFF0007651450000004.tif4128.

[0024] In some embodiments, the linker is a cleavable linker.

[0025] Provided herein is a multispecific polypeptide construct comprising a first component comprising a heterodimeric Fc region and a second component comprising an anti-CD3 antibody or antigen-binding fragment comprising a variable heavy chain region (VH) and a variable light chain region (VL), wherein the VH and VL constituting the anti-CD3 antibody or antigen-binding fragment are linked to opposing polypeptides of the heterodimeric Fc; the first and second components are coupled by a cleavable linker, and the heterodimeric Fc region is positioned N-terminal to the anti-CD3 antibody; one or both of the first and second components comprises at least one antigen-binding domain that binds to a tumor-associated antigen (TAA); and one or both of the first and second components comprises at least one costimulatory receptor-binding region (CRBR) that binds to a costimulatory receptor. In some embodiments, the binding of the CD3-binding region to CD3 is substantially reduced when the multispecific polypeptide construct is in an uncleaved state compared to a cleaved state. In some embodiments, in the cleaved state, the first and second components are unlinked.

[0026] In some embodiments, the cleavable linker is a polypeptide that functions as a substrate for a protease. In some embodiments, the protease is produced by an immune effector cell, by a tumor, or by a cell present in the tumor microenvironment. In some embodiments, the protease is produced by an immune effector cell, and the immune effector cell is an activated T cell, a natural killer (NK) cell, or an NK T cell. In some embodiments, the protease is selected from the group consisting of matriptase, matrix metalloproteinase (MMP), granzyme B, and combinations thereof. In some embodiments, the protease is granzyme B.

[0027] In some embodiments, the cleavable linker has the general formula P4 P3 P2 P1↓P1′ (SEQ ID NO:150) (P4 is an amino acid I, L, Y, M, F, V, or A; P3 is an amino acid A, G, S, V, E, D, Q, N, or Y; P2 is an amino acid H, P, A, V, G, S, or T; P1 is an amino acid D or E; P1′ is an amino acid I, L, Y, M, F, V, T, S, G, or A). In some embodiments, the cleavable linker has the general formula P4 P3 P2 P1↓P1′ (SEQ ID NO:151) (P4 is an amino acid I or L; P3 is an amino acid E; P2 is an amino acid P or A; P1 is an amino acid D; P1′ is an amino acid I, V, T, S, or G). In some embodiments, the cleavable linker comprises the amino acid sequence IEPDI (SEQ ID NO:136), LEPDG (SEQ ID NO:152), LEADT (SEQ ID NO:137), IEPDG (SEQ ID NO:138), IEPDV (SEQ ID NO:139), IEPDS (SEQ ID NO:140), IEPDT (SEQ ID NO:141), or LEADG (SEQ ID NO:153). In some embodiments, the cleavable linker comprises an amino acid sequence selected from the group consisting of SEQ ID NOs:22, 105-112, 136-141, 148, and 150-153. In some examples, the cleavable linker comprises the amino acid sequence set forth in SEQ ID NO:105.

[0028] In some embodiments, the protease is matriptase. In some embodiments, the cleavable linker comprises the sequence P1QAR↓(A / V) (SEQ ID NO:154) (P1 is any amino acid); or the cleavable linker comprises the sequence RQAR(A / V) (SEQ ID NO:155). In some examples, the cleavable linker comprises the sequence RQARV (SEQ ID NO:156). In some embodiments, the cleavable linker comprises an amino acid sequence selected from the group consisting of SEQ ID NOs:23, 154-156. In some embodiments, the protease is an MMP. In some embodiments, the MMP is MMP-2. In some examples, the cleavable linker contains the general formula P3 P2 P1↓P1' (SEQ ID NO:157) (P3 is P, V, or A; P2 is Q or D; P1 is A or N; and P1' is L, I, or M). In some embodiments, the cleavable linker contains the general formula P3 P2 P1↓P1' (SEQ ID NO:158) (P3 is P; P2 is Q or D; P1 is A or N; P1' is L or I). In some cases, the cleavable linker comprises the sequence PAGL (SEQ ID NO:24). In some embodiments, the cleavable linker comprises an amino acid sequence selected from the group consisting of SEQ ID NOs:22-31, 104-114, 117-118, 136-144, 148, 150-158.

[0029] In some embodiments, the multispecific polypeptide construct comprises at least a first polypeptide comprising a first Fc polypeptide of a heterodimeric Fc region, a linker, and a VH domain or VL domain of an anti-CD3 antibody or antigen-binding fragment; and a second polypeptide comprising a second Fc polypeptide of the heterodimeric Fc region, a linker, optionally the same linker as present in the first polypeptide, and the other of the VH domain or VL domain of the anti-CD3 antibody or antigen-binding fragment, wherein one or both of the first and second components comprise at least one antigen-binding domain that binds to a tumor-associated antigen (TAA), and one or both of the first and second components comprise at least one costimulatory receptor-binding region (CRBR) that binds to a costimulatory receptor, and the multispecific polypeptide construct contains at least one CRBR and at least one antigen-binding domain.

[0030] In some embodiments, only one of the first or second polypeptides comprises at least one antigen-binding domain that binds to a TAA. In some embodiments, at least one antigen-binding domain is positioned amino-terminal to the Fc region and / or carboxy-terminal to the CD3-binding region of one of the first or second polypeptides of the multispecific polypeptide construct. In some embodiments, at least one antigen-binding domain is positioned amino-terminal to the Fc region of the multispecific construct, and the second antigen-binding domain is positioned carboxy-terminal to the CD3-binding region of the multispecific construct. In some embodiments, only one of the first or second polypeptides comprises at least one costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor. In some embodiments, the costimulatory receptor binding region (CRBR) is positioned amino-terminal to the Fc region or carboxy-terminal to the CD3-binding region of one of the first or second polypeptides of the multispecific polypeptide construct.

[0031] In some embodiments, the first polypeptide contains, from N-terminus to C-terminus, a first antigen-binding domain that binds to a tumor-associated antigen (TAA), a first Fc polypeptide of the heterodimeric Fc region, a linker, a VL or VH of an anti-CD3 antibody or antigen-binding fragment, and a second antigen-binding domain that binds to a tumor-associated antigen (TAA); and the second polypeptide contains, from N-terminus to C-terminus, a second Fc polypeptide of the heterodimeric Fc region, a linker, optionally the same linker as present in the first polypeptide, the other of the VH or VL of the anti-CD3 antibody or antigen-binding fragment, and a costimulatory receptor-binding region (CRBR) that binds to a costimulatory receptor.

[0032] In some embodiments, the antigen-binding domain, or each of the antigen-binding domains independently, contains the extracellular domain of the native cognate binding partner of the TAA or a binding fragment thereof, or a variant thereof that exhibits binding activity with the TAA. In some embodiments, the antigen-binding domain, or each of the antigen-binding domains independently, is an antibody or antigen-binding fragment thereof selected from the group consisting of a Fab fragment, a F(ab')2 fragment, an Fv fragment, an scFv, an scAb, a dAb, a single-domain heavy chain antibody, and a single-domain light chain antibody.

[0033] In some embodiments, at least one costimulatory receptor binding region (CRBR) is or comprises the extracellular domain of the native cognate binding partner of the costimulatory receptor or a binding fragment thereof, or a variant thereof that exhibits binding activity with the costimulatory receptor. In some embodiments, at least one costimulatory receptor binding region (CRBR) is an antibody or antigen-binding fragment thereof selected from the group consisting of a Fab fragment, a F(ab')2 fragment, a Fv fragment, an scFv, an scAb, a dAb, a single-domain heavy chain antibody, and a single-domain light chain antibody. In some embodiments, the antibody or antigen-binding fragment thereof is an Fv, scFv, Fab, a single-domain antibody (sdAb), a VNAR, or a VHH. In some examples, the antibody or antigen-binding fragment thereof is an sdAb. In some embodiments, the sdAb is a human sdAb or a humanized sdAb. In some embodiments, the sdAb is a VHH, a VNAR, a modified VH domain, or a modified VK domain. In some embodiments, the antibody or antigen-binding fragment thereof is an scFv. In some embodiments, the antibody or antigen-binding fragment thereof is a Fab.

[0034] In some embodiments, the multispecific polypeptide construct comprises a first polypeptide comprising a first Fc polypeptide of a heterodimeric Fc region, a linker, and a VH domain of an anti-CD3 antibody or antigen-binding fragment; a second polypeptide comprising a second Fc polypeptide of the heterodimeric Fc region, a linker, and a VL domain of an anti-CD3 antibody or antigen-binding fragment; a third polypeptide comprising a VH-CH1(Fd) or VL-CL of a Fab antibody fragment that binds to a tumor-associated antigen; and a fourth polypeptide comprising a VH-CH1(Fd) or VL-CL of a Fab antibody fragment that binds to a costimulatory receptor, wherein the first and / or second polypeptide further comprise (1) the other of the VH-CH1(Fd) or VL-CL of the Fab antibody fragment that binds to the tumor-associated antigen, and (2) the other of the VH-CH1(Fd) or VL-CL of the Fab antibody fragment that binds to a costimulatory receptor.

[0035] In some embodiments, the multispecific polypeptide construct comprises a first polypeptide comprising a first Fc polypeptide of a heterodimeric Fc region, a linker, and a VH domain of an anti-CD3 antibody or antigen-binding fragment; a second polypeptide comprising a second Fc polypeptide of the heterodimeric Fc region, a linker, and a VL domain of an anti-CD3 antibody or antigen-binding fragment; and a third polypeptide comprising a VH-CH1(Fd) or VL-CL of a Fab antibody fragment that binds to a costimulatory receptor, wherein the first and / or second polypeptide further comprises the other of the VH-CH1(Fd) or VL-CL of the Fab antibody fragment that binds to a costimulatory receptor, and the first and / or second polypeptide further comprises at least one antigen-binding domain that binds a tumor-associated antigen (TAA).

[0036] In some embodiments, the multispecific polypeptide construct comprises a first polypeptide comprising a first Fc polypeptide of a heterodimeric Fc region, a linker, and a VH domain of an anti-CD3 antibody or antigen-binding fragment; a second polypeptide comprising a second Fc polypeptide of the heterodimeric Fc region, a linker, and a VL domain of an anti-CD3 antibody or antigen-binding fragment; and a third polypeptide comprising a VH-CH1(Fd) or VL-CL of a Fab antibody fragment that binds to a tumor-associated antigen, wherein the first and / or second polypeptide further comprises the other of the VH-CH1(Fd) or VL-CL of the Fab antibody fragment that binds to the tumor-associated antigen, and the first and / or second polypeptide further comprises at least one costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor.

[0037] In some embodiments, only one of the first or second polypeptides contains the other of the VH-CH1(Fd) or VL-CL of the Fab antibody fragment. In some embodiments, both the first or second polypeptides contain the other of the VH-CH1(Fd) or VL-CL of the Fab antibody fragment. In some embodiments, the other of the VH-CH1(Fd) or VL-CL of the Fab antibody fragment is positioned amino-terminal to the Fc region and / or carboxy-terminal to the CD3-binding region of one of the first or second polypeptides of the multispecific polypeptide construct. In some embodiments, the other of the VH-CH1(Fd) or VL-CL of the Fab antibody fragment is positioned amino-terminal to the Fc region of the first or second polypeptide and carboxy-terminal to the CD3-binding region of the other of the first or second polypeptide.

[0038] In some such embodiments, the antigen binding domain, or each of the antigen binding domains independently, is selected from the group consisting of 1-92-LFA-3, 5T4, α4 integrin, αV integrin, α4β1 integrin, α4β7 integrin, AGR2, anti-Lewis Y, apelin J receptor, APRIL, B7-H3, B7-H4, BAFF, BTLA, C5 complement, C-242, CA9, CA19-9 (Lewis a), carbonic anhydrase 9, CD2, C D3, CD6, CD9, CD11a, CD19, CD20, CD22, CD24, CD25, CD27, CD28, CD30, CD33, CD38, CD40, CD40L, CD41, CD44, CD44v6, CD 47, CD51, CD52, CD56, CD64, CD70, CD71, CD74, CD80, CD81, CD86, CD95, CD117, CD123, CD125, CD132(IL-2RG), CD133, C D137, CD138, CD166, CD172A, CD248, CDH6, CEACAM5 (CEA), CEACAM6 (NCA-90), claudin 3, claudin 4, cMet, collagen, Cripto, CSFR, CSFR-1, CTLA-4, CTGF, CXCL10, CXCL13, CXCR1, CXCR2, CXCR4, CYR61, DL44, DLK1, DLL3, DLL4, DPP-4, DSG1, ED A, EDB, EGFR, EGFRviii, endothelin B receptor (ETBR), ENPP3, EpCAM, EPHA2, EPHB2, ERBB3, RSV F protein, FAP, FGF-2, FGF8, FGFR1, FGFR2, FGFR3, FGFR4, FLT-3, folate receptor alpha (FRα), GAL3ST1, G-CSF, G-CSFR, GD2, GITR, GLUT1, GLUT4, GM-CSF, GM-CSFR, GPIIb / IIIa receptor, Gp130, GPIIB / IIIA, GPNMB, GRP78, HER2 / neu, HER3, HER4, HGF, hGH, HVEM, hyaluronidase, ICOS, IFNα, IFNβ, IFNγ, IgE, IgE receptor (FceRI), IGF, IGF1R, IL1B, IL1R, IL2, IL11, IL12, IL12p40, IL-12R, IL-12Rβ1, IL13, IL13R, IL15, IL17, IL18, IL21, I L23, IL23R, IL27 / IL27R (wsx1), IL29, IL-31R, IL31 / IL31R, IL2R, IL4, IL4R, IL6, IL6R, insulin receptor, Jagged ligand, Jagged1, Jagged2, KISS1-R, LAG-3, LIF-R, Lewis X, LIGHT, LRP4, LRRC26, Ly6G6D, LyPD1, MCSP, mesothelin, MRP4, MUC1, mucin 16 (MUC16, CA-125), Na / K ATPase, NGF, Nicastrin, Notch receptor, Notch1, Notch2, Notch3, Notch4, NOV, OSM-R, OX-40, PAR2, PDGF-AA, PDGF-BB, PDGFRα, PDGFRβ, PD-1, PD-L1, PD-L2, phosphatidylserine, P1GF, PSCA, PSMA, PSGR, RAAG12, RAGE, SLC44A4, sphingosine-1-phosphate, STEAP1, STEAP2, TAG-72, TAPA1, TEM-8, TGFβ, TIGIT , TIM-3, TLR2, TLR4, TLR6, TLR7, TLR8, TLR9, TMEM31, TNFα, TNFR, TNFRS12A, TRAIL-R1, TRAIL-R2, transferrin, transferrin receptor, TRK-A, TRK-B, uPAR, VAP1, VCAM-1, VEGF, VEGF-A, VEGF-B, VEGF-C, VEGF-D, VEGFR1, VEGFR2, VEGFR3, VISTA, WISP-1, WISP-2, and WISP-3.

[0039] In some embodiments, the antigen-binding domain comprises at least a first antigen-binding domain and a second antigen-binding domain, wherein the first antigen-binding domain and the second antigen-binding domain bind to the same TAA. In some embodiments, the first antigen-binding domain and the second antigen-binding domain bind to different epitopes of the same TAA. In some embodiments, the first antigen-binding domain and the second antigen-binding domain bind to the same epitope of the same TAA. In some embodiments, the antigen-binding domain comprises at least a first antigen-binding domain and a second antigen-binding domain, wherein the first antigen-binding domain and the second antigen-binding domain bind to different TAAs.

[0040] In some embodiments, the costimulatory receptor binding region (CRBR) comprises at least a first CRBR and a second CRBR, and the first CRBR and the second CRBR bind to the same costimulatory receptor. In some embodiments, the first costimulatory receptor binding region (CRBR) and the second CRBR bind to different epitopes of the same costimulatory receptor. In some embodiments, the first costimulatory receptor binding region (CRBR) and the second CRBR bind to the same epitope of the same costimulatory receptor. In some embodiments, the costimulatory receptor binding region (CRBR) comprises at least a first CRBR and a second CRBR. In some embodiments, the first CRBR and the second CRBR bind to different costimulatory receptors.

[0041] In some embodiments, at least one costimulatory receptor binding region (CRBR) binds to a costimulatory receptor selected from 41BB (CD137), OX40 (CD134), CD27, glucocorticoid-inducible TNFR-related protein (GITR), CD28, ICOS, CD40, B-cell activating factor receptor (BAFF-R), B-cell maturation antigen (BCMA), transmembrane activator and CAML interactor (TACI), and NKG2D. In some embodiments, at least one costimulatory receptor binding region (CRBR) binds to a costimulatory receptor selected from 41BB (CD137), OX40 (CD134), and glucocorticoid-inducible TNFR-related protein (GITR).

[0042] In any of the provided embodiments, at least one inhibitory receptor binding region (IRBR) is an antibody or antigen-binding fragment thereof selected from the group consisting of a Fab fragment, a F(ab')2 fragment, an Fv fragment, an scFv, an scAb, a dAb, a single-domain heavy chain antibody, and a single-domain light chain antibody. In some embodiments, the antibody or antigen-binding fragment thereof is an Fv, scFv, Fab, a single-domain antibody (sdAb), a VNAR, or a VHH. In some embodiments, the antibody or antigen-binding fragment is an sdAb. In some embodiments, the sdAb is a human sdAb or a humanized sdAb. In some embodiments, the sdAb is a VHH, a VNAR, a modified VH domain, or a modified VK domain. In some embodiments, the antibody or antigen-binding fragment thereof is an scFv. In some embodiments, the antibody or antigen-binding fragment thereof is a Fab. In any of the embodiments provided herein that include a construct containing at least one inhibitory receptor binding domain, the inhibitory receptor is selected from PD-1, CTLA-4, TIGIT, VISTA, or TIM3 as expressed on T cells, e.g., human T cells. In some embodiments, the T cells are activated T cells. In any of the embodiments provided, at least one inhibitory receptor binding region (IRBR) binds to PD-1.

[0043] In some embodiments, the multispecific polypeptide construct contains a first connecting peptide (LP1) between the first antigen-binding domain and the Fc region. In some embodiments, the multispecific polypeptide construct contains a second connecting peptide (LP2) between the CD3-binding region and the second antigen-binding domain. In some embodiments, the multispecific polypeptide construct contains a first connecting peptide (LP1) between the first costimulatory receptor binding region (CRBR) and the Fc region. In some embodiments, the multispecific polypeptide construct contains a second connecting peptide (LP2) between the CD3-binding region and the second costimulatory receptor binding region (CRBR). In some embodiments, the multispecific polypeptide construct comprises a first connecting peptide (LP1) between the antigen binding domain or costimulatory receptor binding region and the Fc region and a second connecting peptide (LP2) between the CD3 binding region and the antigen binding domain or CRBR, and has the following structural arrangement from N-terminus to C-terminus: first antigen binding domain or CRBR - LP1 - Fc region - linker - CD3 binding region - LP2 - second antigen binding domain or CRBR.

[0044] In some embodiments, the linker is a cleavable linker. In some embodiments, the two connecting peptides are not identical to each other. In some embodiments, LP1 or LP2 are independently peptides of about 1 to 20 amino acids in length. In some embodiments, LP1 or LP2 independently contain a peptide that is or includes a Gly-Ser linker or GGS as set forth in SEQ ID NOs: 10-13, 119, 135, 147, 149.

[0045] In some embodiments, the anti-CD3 antibody or antigen-binding fragment is an Fv antibody fragment. In some embodiments, the Fv antibody fragment comprises a disulfide-stabilized anti-CD3 binding Fv (dsFv). In some embodiments, the anti-CD3 antibody or antigen-binding fragment comprises a VH CDR1 comprising the amino acid sequence TYAMN (SEQ ID NO:16); a VH CDR3 comprising the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO:18), a VL CDR1 comprising the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO:19); a VL CDR2 comprising the amino acid sequence GTNKRAP (SEQ ID NO:20); and a VL CDR3 comprising the amino acid sequence ALWYSNLWV (SEQ ID NO:21).

[0046] In some examples, the anti-CD3 Fv comprises a VH having an amino acid sequence of any of SEQ ID NOs: 14 and 32-62, or a sequence that exhibits at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to any of SEQ ID NOs: 14 and 32-62; and a VL having an amino acid sequence of any of SEQ ID NOs: 15 and 63-81, or a sequence that exhibits at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to any of SEQ ID NOs: 15 and 63-81.

[0047] In some embodiments, the anti-CD3 antibody or antigen-binding fragment is an Fv. In some embodiments, the anti-CD3 Fv comprises a VH having an amino acid sequence of any of SEQ ID NOs: 14, 32-43, 45-47, 48, and 287, or a sequence that exhibits at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to any of SEQ ID NOs: 14, 32-43, 45-47, 48, and 287; and a VL having an amino acid sequence of any of SEQ ID NOs: 15, 63, 65-71, 73, 75, 77, and 288, or a sequence that exhibits at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to any of SEQ ID NOs: 15, 63, 65-71, 73, 75, 77, and 288. In some embodiments, the anti-CD3 Fv comprises the amino acid sequence of SEQ ID NO: 14 and the amino acid sequence of SEQ ID NO: 15. In other cases, the anti-CD3 Fv comprises the amino acid sequence of SEQ ID NO: 287 and the amino acid sequence of SEQ ID NO: 288.

[0048] In some embodiments, the VH chain region and VL chain region of the CD3-binding domain each independently comprise at least one amino acid modification. In some embodiments, the at least one amino acid modification in the VH chain region and VL chain region of the CD3-binding domain increases the stability of the CD3-binding domain. In some embodiments, the at least one amino acid modification in the VH chain region and VL chain region of the CD3-binding domain increases the ability of the CD3-binding domain to bind to CD3. In some embodiments, the at least one amino acid modification in the VH chain region and VL chain region of the CD3-binding domain increases the stability of the CD3-binding domain by creating a disulfide bond between the VH chain region and the VL chain region.

[0049] In some embodiments, the CD3 binding region has a disulfide-stabilized bond between the VH and VL regions. In some embodiments, the anti-CD3 antibody or antigen-binding fragment is a disulfide-stabilized Fv (dsFv). In some embodiments, the disulfide-stabilized anti-CD3 Fv comprises an anti-CD3 VH comprising a Cys mutation at position 44 and an anti-CD3 VL comprising a Cys mutation at position 100 according to the Kabat numbering. In some embodiments, the disulfide-stabilized anti-CD3 Fv comprises an anti-CD3 VH comprising a G44C mutation and an anti-CD3 VL comprising a G100C mutation according to the Kabat numbering. In some embodiments, the disulfide-stabilized anti-CD3 Fv comprises an anti-CD3 VH comprising a Cys mutation at position 105 and an anti-CD3 VL comprising a Cys mutation at position 43 according to the Kabat numbering. In some embodiments, the anti-CD3 dsFv comprises a VH having the amino acid sequence of any of SEQ ID NOs: 44, 49-62, 290, and 311, or a sequence that exhibits at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to any of SEQ ID NOs: 44, 49-62, 290, and 311; and a VL having the amino acid sequence of any of SEQ ID NOs: 64, 72, 74, 76, 78-81, 241, and 289, or a sequence that exhibits at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to any of SEQ ID NOs: 64, 72, 74, 76, 78-81, 241, and 289. In some embodiments, the anti-CD3 dsFv comprises the amino acid sequence of SEQ ID NO:44 and the amino acid sequence of SEQ ID NO:72. In some embodiments, the anti-CD3 dsFv comprises the amino acid sequence of SEQ ID NO:44 and the amino acid sequence of SEQ ID NO:241. In some embodiments, the anti-CD3 epsilon-binding domain comprises a variable heavy chain (VH) comprising the amino acid sequence of SEQ ID NO:290 and a variable light chain (VL) comprising the amino acid sequence of SEQ ID NO:289.In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (VH) comprising the amino acid sequence of SEQ ID NO:311 and a variable light chain (VL) comprising the amino acid sequence of SEQ ID NO:289.

[0050] In some such embodiments, the multispecific polypeptide construct is conjugated to an agent. In some embodiments, the agent is a therapeutic agent, an anti-tumor agent, a toxin or fragment thereof, a detectable moiety, or a diagnostic agent. In some embodiments, the agent is conjugated to the multispecific polypeptide construct via a linker.

[0051] Provided herein are polynucleotides encoding any of the provided multispecific polypeptide constructs. Provided herein are polynucleotides encoding any of the provided polypeptide chains of any of the provided multispecific polypeptide constructs. Provided herein are polynucleotides containing a first nucleic acid sequence encoding a first polypeptide of any of the provided multispecific constructs and a second nucleic acid sequence encoding any of the provided second polypeptides of the multispecific constructs, wherein the first and second nucleic acid sequences are separated by an internal ribosome entry site (IRES) or a nucleic acid encoding a self-cleaving peptide or a peptide that causes ribosome skipping. In some embodiments, the first nucleic acid sequence and the second nucleic acid sequence are operably linked to the same promoter. In some embodiments, the multispecific polypeptide construct contains a third polypeptide chain, and the polynucleotide further contains a third nucleic acid encoding the third polypeptide of the multispecific construct. In some embodiments, the third nucleic acid is separated from the first and / or second polypeptide by an internal ribosome entry site (IRES) or a nucleic acid encoding a self-cleaving peptide or a peptide that causes ribosome skipping, and / or the third nucleic acid sequence is operably linked to the same promoter as the first and / or second nucleic acid sequence. In some embodiments, the nucleic acid encoding the self-cleaving peptide or a peptide that causes ribosome skipping is selected from T2A, P2A, E2A, or F2A.

[0052] Provided herein is a vector containing any of the provided polynucleotides. In some embodiments, the vector is an expression vector. In some embodiments, the vector is a viral vector or a eukaryotic vector, and optionally, the eukaryotic vector is a mammalian vector.

[0053] Provided herein are cells containing a polynucleotide or any of the provided polynucleotides, a vector or any of the provided vectors. In some embodiments, the cells are recombinant or isolated. In some embodiments, the cells are mammalian cells. In some embodiments, the cells are HEK293 cells or CHO cells.

[0054] Provided herein is a method of making a multispecific polypeptide construct comprising introducing a polynucleotide or any of the provided polynucleotides, or a vector or any of the provided vectors into a cell and culturing the cells under conditions wherein the multispecific polypeptide construct is produced.

[0055] Provided herein are methods for producing a multispecific polypeptide construct, comprising culturing any of the provided cells under conditions in which the multispecific polypeptide is produced by the cells. In some embodiments, the cells are mammalian cells. In some embodiments, the cells are HEK293 cells or CHO cells. In some embodiments, the method further comprises isolating or purifying the multispecific polypeptide construct from the cells. In some embodiments, the multispecific polypeptide construct is a heterodimer.

[0056] Provided herein are multispecific polypeptide constructs produced by any of the provided methods.

[0057] Provided herein are pharmaceutical compositions containing any of the provided multispecific polypeptide constructs and a pharmaceutically acceptable carrier. In some embodiments, the pharmaceutical composition is sterile.

[0058] Provided herein are methods for stimulating or inducing an immune response, comprising contacting a target cell and a T cell with any of the provided multispecific polypeptide constructs or pharmaceutical compositions, wherein the target cell expresses a tumor-associated antigen recognized by the multispecific polypeptide construct. In some embodiments, the target cell is a tumor cell expressing a tumor-associated antigen (TAA). In some embodiments, the multispecific polypeptide construct comprises a cleavable linker that functions as a substrate for a protease, and the induction or stimulation of an immune response is increased in the presence of the protease. In some embodiments, the protease is produced by an immune effector cell, by the tumor, or by a cell present in the tumor microenvironment. In some embodiments, the protease is produced by an immune effector cell, and the immune effector cell is an activated T cell, a natural killer (NK) cell, or an NK T cell. In some embodiments, the immune effector cell is in the vicinity of a cell expressing the antigen. In some embodiments, the protease is produced by a tumor in the vicinity of cells expressing the TAA in the tissue and / or is co-localized with the TAA in the tissue, and when the multispecific polypeptide construct is exposed to the protease, the protease cleaves the cleavable linker in the multispecific polypeptide construct.

[0059] In some embodiments, the protease is selected from the group consisting of matriptase, matrix metalloproteinase (MMP), granzyme B, and combinations thereof. In some embodiments, the protease is granzyme B. In some embodiments, the contacting is performed ex vivo or in vitro. In some embodiments, the contacting is performed in vivo in a subject.

[0060] Provided herein are methods for stimulating or inducing an immune response in a subject, comprising administering a therapeutically effective amount of any of the provided multispecific conjugates or pharmaceutical compositions to a subject in need thereof. In some embodiments, the method increases cellular immunity. In some embodiments, the method increases T cell activity. In some embodiments, the method increases cytolytic T cell (CTL) activity. In some embodiments, the immune response against tumors or cancer is increased. In some embodiments, the method treats a disease or condition in the subject.

[0061] Provided herein are methods for treating a disease or condition in a subject, comprising administering to a subject in need thereof a therapeutically effective amount of any of the provided multispecific conjugates or any of the provided pharmaceutical compositions. In some embodiments, the disease or condition is a tumor or cancer. In some embodiments, the subject is a human.

[0062] Those skilled in the art will recognize that the antibodies of the present disclosure have a variety of uses. For example, the proteins of the present disclosure are used as therapeutic agents for a variety of disorders. The antibodies of the present disclosure are also used as reagents in diagnostic kits or as diagnostic tools, or they may be used in competitive assays to generate therapeutic reagents. [The present invention 1001] 1. A multispecific polypeptide construct comprising a first component comprising an immunoglobulin Fc region and a second component comprising a CD3 binding region, the first and second components are coupled by a linker, and the Fc region is positioned N-terminal to the CD3 binding region; one or both of the first and second components comprises at least one antigen-binding domain that binds to a tumor-associated antigen (TAA); One or both of the first and second components comprises at least one costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor; Multispecific polypeptide constructs. [The present invention 1002] 1. A multispecific polypeptide construct comprising a first component comprising an immunoglobulin Fc region and a second component comprising a CD3 binding region, the first and second components are coupled by a linker, and the Fc region is positioned N-terminal to the CD3 binding region; one or both of the first and second components comprises at least one antigen-binding domain that binds to a tumor-associated antigen (TAA); One or both of the first and second components comprises at least one inhibitory receptor binding region (IRBR) that binds to an inhibitory receptor; Multispecific polypeptide constructs. [The present invention 1003] 1. A multispecific polypeptide construct comprising a first component comprising an immunoglobulin Fc region and a second component comprising a CD3 binding region, the first and second components are coupled by a linker, and the Fc region is positioned N-terminal to the CD3 binding region; one or both of the first and second components comprises at least one antigen-binding domain that binds to a tumor-associated antigen (TAA); one or both of the first and second components comprises at least one costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor; One or both of the first and second components comprises at least one inhibitory receptor binding region (IRBR) that binds to an inhibitory receptor; Multispecific polypeptide constructs. [The present invention 1004] 1004. The multispecific polypeptide construct of any of claims 1001 to 1003, wherein the CD3 binding region binds to CD3 (CD3ε). [The present invention 1005] 1005. The multispecific polypeptide construct of any of claims 1001 to 1004, wherein at least one antigen-binding domain is positioned amino-terminal to the Fc region and / or carboxy-terminal to the CD3-binding region of the multispecific polypeptide construct. [The present invention 1006] 1006. The multispecific polypeptide construct of any of claims 1001 to 1005, wherein the first component comprises a first antigen-binding domain and the second component comprises a second antigen-binding domain, each of the antigen-binding domains binding to a tumor-associated antigen (TAA). [The present invention 1007] 1006. The multispecific polypeptide construct of the invention, wherein the first antigen-binding domain is positioned amino-terminal to the Fc region of the multispecific construct and the second antigen-binding domain is positioned carboxy-terminal to the CD3-binding region of the multispecific construct. [The present invention 1008] The multispecific polypeptide construct of any of claims 1001 and 1003 to 1007, wherein at least one costimulatory receptor binding region (CRBR) is positioned amino-terminal to the Fc region and / or carboxy-terminal to the CD3 binding region of the multispecific polypeptide construct. [The present invention 1009] The multispecific polypeptide construct of any of claims 1001 and 1003 to 1008, wherein at least one costimulatory receptor binding region (CRBR) is positioned carboxy-terminal to the CD3 binding region of the multispecific polypeptide construct. [The present invention 1010] 8. The multispecific polypeptide construct of any of claims 1002 to 1007, wherein at least one inhibitory receptor binding region (IRBR) is positioned amino-terminal to the Fc region and / or carboxy-terminal to the CD3 binding region of the multispecific polypeptide construct. [The present invention 1011] The multispecific polypeptide construct of any of claims 1002 to 1007 and 1010, wherein at least one inhibitory receptor binding region (IRBR) is positioned carboxy-terminal to the CD3 binding region of the multispecific polypeptide construct. [The present invention 1012] 1012. The multispecific polypeptide construct of any of claims 1001 to 1011, wherein the first component comprises a first antigen-binding domain and the second component comprises a second antigen-binding domain, each of the antigen-binding domains binding to a tumor-associated antigen (TAA). [The present invention 1013] 1012. The multispecific polypeptide construct of the invention, wherein the first antigen-binding domain is positioned amino-terminal to the Fc region of the multispecific construct and the second antigen-binding domain is positioned carboxy-terminal to the CD3-binding region of the multispecific construct. [The present invention 1014] The multispecific polypeptide construct of invention 1012 or invention 1013, wherein the first or second component further comprises a costimulatory receptor binding region (CRBR). [The present invention 1015] The multispecific polypeptide construct of invention 1012 or invention 1013, wherein the first or second component further comprises an inhibitory receptor binding region (IRBR). [The present invention 1016] From the N-terminus to the C-terminus, a costimulatory receptor-binding region (CRBR) that binds to a costimulatory receptor and / or an antigen-binding domain that binds to a tumor-associated antigen (TAA); immunoglobulin Fc region; Linker; a CD3 binding region that binds to CD3 (CD3ε); and a costimulatory receptor-binding region (CRBR) that binds to a costimulatory receptor and / or an antigen-binding domain that binds to a tumor-associated antigen (TAA) 1. A multispecific polypeptide construct comprising: A multispecific polypeptide construct comprising at least one CRBR and at least one antigen-binding domain. [The present invention 1017] 1016. A multispecific polypeptide construct of the present invention comprising only one costimulatory receptor binding region (CRBR). [The present invention 1018] From the N-terminus to the C-terminus, an inhibitory receptor-binding region (IRBR) that binds to an inhibitory receptor and / or an antigen-binding domain that binds to a tumor-associated antigen (TAA); immunoglobulin Fc region; Linker; a CD3 binding region that binds to CD3 (CD3ε); and Inhibitory receptor-binding region (IRBR) that binds to inhibitory receptors and / or antigen-binding domain that binds to tumor-associated antigens (TAA) 1. A multispecific polypeptide construct comprising: A multispecific polypeptide construct comprising at least one IRBR and at least one antigen-binding domain. [The present invention 1019] 1018. A multispecific polypeptide construct of the present invention comprising only one inhibitory receptor binding region (IRBR). [The present invention 1020] From the N-terminus to the C-terminus, an inhibitory receptor-binding region (IRBR) that binds to an inhibitory receptor or a costimulatory receptor-binding region (CRBR) that binds to a costimulatory receptor, and / or an antigen-binding domain that binds to a tumor-associated antigen (TAA); immunoglobulin Fc region; Linker; a CD3 binding region that binds to CD3 (CD3ε); and An antigen-binding domain that binds to the other of IRBR or CRBR, and / or a tumor-associated antigen (TAA) 1. A multispecific polypeptide construct comprising: A multispecific polypeptide construct comprising at least one IRBR, at least one CRBR, and at least one antigen-binding domain. [The present invention 1021] 10. The multispecific polypeptide construct of any of claims 1016 to 1020, comprising two antigen-binding domains that bind to a TAA. [The present invention 1022] 1021. A multispecific polypeptide construct of the present invention, wherein the antigen-binding domains bind to the same tumor-associated antigen (TAA). [The present invention 1023] 1023. The multispecific polypeptide construct of invention 1021 or invention 1022, wherein one antigen-binding domain is positioned amino-terminal to the Fc region and one antigen-binding domain is positioned carboxy-terminal to the CD3-binding region. [The present invention 1024] From the N-terminus to the C-terminus, immunoglobulin Fc region; Linker; a CD3 binding region that binds to CD3 (CD3ε); and The antigen-binding domain binds to tumor-associated antigens (TAA) and the costimulatory receptor-binding region (CRBR) binds to costimulatory receptors. 1. A multispecific polypeptide construct comprising: [The present invention 1025] From the N-terminus to the C-terminus, an antigen-binding domain that binds to tumor-associated antigens (TAAs) and a costimulatory receptor-binding region (CRBR) that binds to costimulatory receptors; immunoglobulin Fc region; a linker; and CD3 binding domain that binds to CD3 (CD3ε) 1. A multispecific polypeptide construct comprising: [The present invention 1026] From the N-terminus to the C-terminus, immunoglobulin Fc region; Linker; a CD3 binding region that binds to CD3 (CD3ε); and An antigen-binding domain that binds to tumor-associated antigens (TAA) and an inhibitory receptor-binding region (IRBR) that binds to inhibitory receptors 1. A multispecific polypeptide construct comprising: [The present invention 1027] From the N-terminus to the C-terminus, an antigen-binding domain that binds to tumor-associated antigens (TAAs) and an inhibitory receptor-binding region (IRBR) that binds to inhibitory receptors; immunoglobulin Fc region; a linker; and CD3 binding domain that binds to CD3 (CD3ε) 1. A multispecific polypeptide construct comprising: [The present invention 1028] The multispecific polypeptide construct of any of claims 1001 to 1027, wherein the Fc region is a homodimeric Fc region. [The present invention 1029] The multispecific polypeptide construct of any of claims 1001 to 1028, wherein the Fc region is the Fc region of human IgG1, human IgG2, human IgG3, or human IgG4, or an immunologically active fragment thereof. [The present invention 1030] 10. The multispecific polypeptide construct of any of claims 1001 to 1029, wherein the Fc region comprises a polypeptide comprising the amino acid sequence set forth in SEQ ID NO:1, or a sequence of amino acids having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to SEQ ID NO:1. [The present invention 1031] the Fc region comprises a polypeptide comprising the amino acid sequence set forth in SEQ ID NO:2, or a sequence of amino acids having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to SEQ ID NO:2; the Fc region comprises a polypeptide comprising the amino acid sequence set forth in SEQ ID NO:4, or a sequence of amino acids having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to SEQ ID NO:4; or the Fc region comprises a polypeptide comprising the amino acid sequence set forth in SEQ ID NO:5, or a sequence of amino acids having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to SEQ ID NO:5; A multispecific polypeptide construct according to any one of claims 1001 to 1029 of the present invention. [The present invention 1032] The multispecific polypeptide construct of any of claims 1001 to 1023 and 1029 to 1031, wherein the Fc region is a heterodimeric Fc region. [The present invention 1033] 1032. The multispecific polypeptide construct of the present invention, wherein one or both Fc polypeptides of the heterodimeric Fc region comprise at least one modification to induce heterodimerization compared to the polypeptide of the homodimeric Fc region, and optionally compared to the Fc polypeptide set forth in SEQ ID NO:1 or an immunologically active fragment thereof. [The present invention 1034] The multispecific polypeptide construct of the invention 1033, wherein each of the Fc polypeptides of the heterodimeric Fc independently comprises at least one amino acid modification. [This invention 1035] The multispecific polypeptide construct of the present invention 1034, wherein each of the Fc polypeptides of the heterodimeric Fc comprises a knob-into-hole modification or a charge mutation to increase the electrostatic complementarity of the polypeptides. [The present invention 1036] 1035. The multispecific polypeptide construct of the invention, wherein the amino acid modifications are knob-into-hole modifications. [This invention 1037] 1037. The multispecific polypeptide construct of any of claims 1032 to 1036, wherein the CD3 binding region comprises a variable heavy chain region (VH) and a variable light chain region (VL), wherein the VL is C-terminal to a first Fc polypeptide of the heterodimeric Fc region and the VH is C-terminal to a second Fc polypeptide of the heterodimeric Fc region, wherein the first Fc polypeptide comprises a hole mutation and the second Fc polypeptide comprises a knob mutation. [The present invention 1038] 1038. The multispecific polypeptide construct of any of claims 1032 to 1037, wherein the first Fc polypeptide of the heterodimeric Fc comprises a modification selected from Thr366Ser, Leu368Ala, Tyr407Val, and combinations thereof, and the second Fc polypeptide of the heterodimeric Fc comprises the modification Thr366Trp. [This invention 1039] 1038. The multispecific polypeptide construct of the invention, wherein the first and second Fc polypeptides further comprise a modification of a non-cysteine ​​residue to a cysteine ​​residue, the modification of the first polypeptide being at one of positions Ser354 and Tyr349 and the modification of the second Fc polypeptide being at the other of positions Ser354 and Tyr349. [The present invention 1040] 1035. The multispecific polypeptide construct of the present invention, wherein the amino acid modifications are charge mutations to increase electrostatic complementarity of the polypeptide. [The present invention 1041] The multispecific polypeptide construct of any of claims 1032 to 1035 and 1040, wherein the first and / or second Fc polypeptide or the first and second Fc polypeptide each comprises a modification at a complementary position, the modification being a substitution of an amino acid having an opposite charge to the complementary amino acid of the other polypeptide. [The present invention 1042] 1042. The multispecific polypeptide construct of any of claims 1033 to 1041, wherein one of the first or second Fc polypeptides of the heterodimeric Fc further comprises a modification at residue Ile253. [This invention 1043] 1042. The multispecific polypeptide construct of the present invention, wherein said modification is Ile253Arg. [This invention 1044] 1043. The multispecific polypeptide construct of any of claims 1033 to 1043, wherein one of the first or second Fc polypeptides of the heterodimeric Fc further comprises a modification at residue His435. [This invention 1045] The multispecific polypeptide construct of the present invention 1044, wherein said modification is His435Arg. [The present invention 1046] 1046. The multispecific polypeptide construct of any of claims 1001 to 1045, wherein the Fc region comprises a polypeptide lacking Lys447. [This invention 1047] 1047. The multispecific polypeptide construct of any of claims 1001 to 1046, wherein the Fc region comprises a polypeptide comprising at least one modification to enhance FcRn binding. [This invention 1048] 1047. The multispecific polypeptide construct of the present invention, wherein said modification is at a position selected from the group consisting of Met252, Ser254, Thr256, Met428, Asn434, and combinations thereof. [This invention 1049] 1048. The multispecific polypeptide construct of the present invention, wherein said modification is at a position selected from the group consisting of Met252Y, Ser254T, Thr256E, Met428L, Met428V, Asn434S, and combinations thereof. [The present invention 1050] 1048. The multispecific polypeptide construct of the invention, wherein said modifications are at position Met252 and at position Met428. [This invention 1051] 1050. The multispecific polypeptide construct of the present invention, wherein said modifications are Met252Y and Met428L. [This invention 1052] 1050. The multispecific polypeptide construct of the present invention, wherein said modifications are Met252Y and Met428V. [This invention 1053] The multispecific polypeptide construct of any of claims 1032 to 1052, wherein the first polypeptide of the heterodimeric Fc comprises the amino acid sequence set forth in any of SEQ ID NOs: 82, 86, 94, or 96, and the second polypeptide of the heterodimeric Fc comprises the amino acid sequence set forth in any of SEQ ID NOs: 83, 87, 90, 92, 98, or 100. [This invention 1054] The multispecific polypeptide construct of any of claims 1032 to 1053, wherein the first polypeptide of the heterodimeric Fc comprises the amino acid sequence set forth in any of SEQ ID NOs: 291, 293, 297, or 298, and the second polypeptide of the heterodimeric Fc comprises the amino acid sequence set forth in any of SEQ ID NOs: 292, 294, 295, 296, 299, or 300. [This invention 1055] The Fc region Polypeptides comprising at least one amino acid modification that reduces effector function and / or reduces binding to an effector molecule selected from an Fcγ receptor or C1q The multispecific polypeptide construct of any of claims 1001 to 1054, comprising: [The present invention 1056] 1055. The multispecific polypeptide construct of the invention, wherein the one or more amino acid modifications is a deletion of one or more of Glu233, Leu234, or Leu235. [This invention 1057] 1056. The multispecific polypeptide construct of any of claims 1032 to 1056, wherein the first polypeptide of the heterodimeric Fc comprises the amino acid sequence set forth in any of SEQ ID NOs: 84, 88, 95, or 97, and the second polypeptide of the heterodimeric Fc comprises the amino acid sequence set forth in any of SEQ ID NOs: 85, 89, 91, 93, 99, or 101. [This invention 1058] The multispecific polypeptide construct of any of claims 1032 to 1056, wherein the first polypeptide of the heterodimeric Fc comprises the amino acid sequence set forth in any of SEQ ID NOs: 291, 293, 297, or 298, and the second polypeptide of the heterodimeric Fc comprises the amino acid sequence set forth in any of SEQ ID NOs: 292, 294, 295, 296, 299, or 300. [This invention 1059] 10. The multispecific polypeptide construct of any of claims 1001 to 1054, wherein the Fc region comprises a polypeptide comprising at least one modification to enhance FcγR binding. [The present invention 1060] 1059. The multispecific polypeptide construct of the present invention, wherein said modification is a modification at Ser239 or Ile332. [This invention 1061] 1059. The multispecific polypeptide construct of any of claims 1001 to 1054 and 1059, wherein the glycosylation of the Fc region is modified to enhance FcγR binding compared to an unmodified Fc region. [This invention 1062] 1061. A multispecific polypeptide construct of the invention, wherein the Fc region lacks fucose or has a reduced fucose content. [The present invention 1063] 1062. The multispecific polypeptide construct of any of claims 1001 to 1062, wherein the CD3 binding region is an anti-CD3 antibody or antigen-binding fragment. [The present invention 1064] 1063. The multispecific polypeptide construct of the invention, wherein the anti-CD3 antibody or antigen-binding fragment comprises a variable heavy chain region (VH) and a variable light chain region (VL). [This invention 1065] The multispecific construct of the present invention 1064, wherein the VL of the anti-CD3 antibody or antigen-binding fragment is linked to a first Fc polypeptide of a heterodimeric Fc comprising a modification selected from Thr366Ser, Leu368Ala, Tyr407Val, and combinations thereof, and the VH of the anti-CD3 antibody or antigen-binding fragment is linked to a second Fc polypeptide of the heterodimeric Fc comprising the modification Thr366Trp. [The present invention 1066] 1066. The multispecific polypeptide construct of any of claims 1001 to 1065, wherein the CD3 binding region is monovalent. [This invention 1067] 1067. The multispecific polypeptide construct of any of claims 1001 to 1066, wherein the CD3 binding region is a variable fragment (Fv) comprising a variable heavy chain region (VH) and a variable light chain region (VL). [The present invention 1068] 1067. The multispecific polypeptide construct of any of claims 1063 to 1066, wherein the anti-CD3 antibody or antigen-binding fragment is not a single chain antibody, and optionally is not a single chain variable fragment (scFv). [The present invention 1069] The multispecific polypeptide construct of any of claims 1064 to 1068, wherein the Fc is a heterodimeric Fc, and the VH and VL constituting the anti-CD3 antibody or antigen-binding fragment are linked to opposing polypeptides of the heterodimeric Fc. [The present invention 1070] 1069. The multispecific polypeptide construct of any of claims 1001 to 1069, wherein the CD3 binding region is incapable or substantially incapable of binding to or engaging with CD3 unless at least one of the antigen binding domains is bound to its TAA. [This invention 1071] 10. The multispecific polypeptide construct of any of claims 1001 to 1070, wherein the CD3 binding region is incapable or substantially incapable of binding to or engaging with CD3 unless at least two of the antigen binding domains are bound to its TAA. [This invention 1072] 1072. The multispecific polypeptide construct of any of claims 1001 to 1071, wherein the linker is a polypeptide linker. [This invention 1073] 1072. The multispecific polypeptide construct of the present invention, wherein the linker is a polypeptide of up to 25 amino acids in length. [This invention 1074] The linker may be selected from the group consisting of 2 to 24 amino acids, 2 to 20 amino acids, 2 to 18 amino acids, 2 to 14 amino acids, 2 to 12 amino acids, 2 to 10 amino acids, 2 to 8 amino acids, 2 to 6 amino acids, 6 to 24 amino acids, 6 to 20 amino acids, 6 to 18 amino acids, 6 to 14 amino acids, 6 to 12 amino acids, 6 to 10 amino acids, 6 to 8 amino acids, 8 to 24 amino acids, 8 to 20 amino acids, 8 to 18 amino acids, 8 to 14 amino acids, 8 to 12 amino acids, 8 to 10 amino acids, 10 to 24 amino acids, 10 to 20 amino acids, 10 to 18 amino acids, 10 to 14 amino acids, 10 to 12 amino acids, 12 to 24 amino acids, 12 to 20 amino acids, 12 to 18 amino acids, 12 to 14 amino acids, 14 to 24 amino acids, 14 to 20 amino acids, 14 to 18 amino acids, 18 to 24 amino acids, 18 to 20 amino acids, or 20 to 24 amino acids, or about 2 to 24 amino acids, 2 to 20 amino acids the multispecific polypeptide construct of invention 1072 or 1073, which is a polypeptide of 2 to 18 amino acids, 2 to 14 amino acids, 2 to 12 amino acids, 2 to 10 amino acids, 2 to 8 amino acids, 2 to 6 amino acids, 6 to 24 amino acids, 6 to 20 amino acids, 6 to 18 amino acids, 6 to 14 amino acids, 6 to 12 amino acids, 6 to 10 amino acids, 6 to 8 amino acids, 8 to 24 amino acids, 8 to 20 amino acids, 8 to 18 amino acids, 8 to 14 amino acids, 8 to 12 amino acids, 8 to 10 amino acids, 10 to 24 amino acids, 10 to 20 amino acids, 10 to 18 amino acids, 10 to 14 amino acids, 10 to 12 amino acids, 12 to 24 amino acids, 12 to 20 amino acids, 12 to 18 amino acids, 12 to 14 amino acids, 14 to 24 amino acids, 14 to 20 amino acids, 14 to 18 amino acids, 18 to 24 amino acids, 18 to 20 amino acids, or 20 to 24 amino acids. [This invention 1075] The multispecific polypeptide construct of any of claims 1072 to 1074, wherein the linker is a polypeptide that is 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acids in length. [This invention 1076] 1076. The multispecific polypeptide construct of any of claims 1072 to 1075, wherein the linker is a polypeptide having a length of 3 to 18 amino acids. [This invention 1077] 1076. The multispecific polypeptide construct of any of claims 1072 to 1075, wherein the linker is a polypeptide having a length of 12 to 18 amino acids. [This invention 1078] 1076. The multispecific polypeptide construct of any of claims 1072 to 1075, wherein the linker is a polypeptide that is 15 to 18 amino acids in length. [This invention 1079] 1078. The multispecific polypeptide construct of any of claims 1001 to 1078, wherein the linker is a non-cleavable linker. [The present invention 1080] 1079. The multispecific polypeptide construct of the present invention, wherein the non-cleavable linker does not contain a substrate recognition site that is specifically recognized for cleavage by a protease. [This invention 1081] The multispecific polypeptide construct of invention 1079 or invention 1080, wherein the non-cleavable linker comprises GS, GGS, GGGGS (SEQ ID NO: 149), GGGGGS (SEQ ID NO: 135), and combinations thereof. [This invention 1082] 1079-1081. The multispecific polypeptide construct of any of claims 1079-1081, wherein the non-cleavable linker comprises (GGS)n, where n is 1 to 10. [This invention 1083] 1079-1082. The multispecific polypeptide construct of any of claims 1079-1082, wherein the non-cleavable linker comprises (GGGGS)n (SEQ ID NO:173), wherein n is 1 to 10. [This invention 1084] 1079-1083. The multispecific polypeptide construct of any of claims 1079-1083, wherein the non-cleavable linker comprises (GGGGGS)n (SEQ ID NO:172), wherein n is 1 to 4. [This invention 1085] The multispecific polypeptide construct of any of claims 1079 to 1084, wherein the non-cleavable linker comprises GGS. [The present invention 1086] The multispecific polypeptide construct of any of claims 1079 to 1084, wherein the non-cleavable linker comprises GGGGS (SEQ ID NO: 149). [This invention 1087] The multispecific polypeptide construct of any of claims 1079 to 1084, wherein the non-cleavable linker comprises GGGGGS (SEQ ID NO: 135). [This invention 1088] The non-cleavable linker is (GGS) 2 A multispecific polypeptide construct according to any one of claims 1079 to 1084, comprising (SEQ ID NO: 10). [This invention 1089] The multispecific polypeptide construct of any of claims 1079 to 1084, wherein the non-cleavable linker comprises GGSGGSGGS (SEQ ID NO:11). [The present invention 1090] The multispecific polypeptide construct of any of claims 1079 to 1084, wherein the non-cleavable linker comprises GGSGGSGGSGGS (SEQ ID NO: 12). [This invention 1091] The non-cleavable linker is TIFF0007651450000006.tif4128 A multispecific polypeptide construct according to any one of claims 1079 to 1084, comprising: [This invention 1092] The non-cleavable linker is TIFF0007651450000007.tif4128 A multispecific polypeptide construct according to any one of claims 1079 to 1084, comprising: [This invention 1093] The non-cleavable linker is TIFF0007651450000008.tif4128 A multispecific polypeptide construct according to any one of claims 1079 to 1084, comprising: [This invention 1094] The non-cleavable linker is TIFF0007651450000009.tif4128 A multispecific polypeptide construct according to any one of claims 1079 to 1084, comprising: [This invention 1095] 1078. The multispecific polypeptide construct of any of claims 1001 to 1078, wherein the linker is a cleavable linker. [This invention 1096] 1. A multispecific polypeptide construct comprising a first component comprising a heterodimeric Fc region and a second component comprising an anti-CD3 antibody or antigen-binding fragment comprising a variable heavy chain region (VH) and a variable light chain region (VL), the VH and VL constituting the anti-CD3 antibody or antigen-binding fragment are linked to opposing polypeptides of a heterodimeric Fc; the first and second components are coupled by a cleavable linker, and the heterodimeric Fc region is positioned N-terminal to the anti-CD3 antibody; one or both of the first and second components comprises at least one antigen-binding domain that binds to a tumor-associated antigen (TAA); One or both of the first and second components comprises at least one costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor; Multispecific polypeptide constructs. [This invention 1097] 1096. The multispecific polypeptide construct of the present invention, wherein one or both of the first and second components further comprises at least one inhibitory receptor binding region (IRBR) that binds to an inhibitory receptor. [This invention 1098] 1. A multispecific polypeptide construct comprising a first component comprising a heterodimeric Fc region and a second component comprising an anti-CD3 antibody or antigen-binding fragment comprising a variable heavy chain region (VH) and a variable light chain region (VL), the VH and VL constituting the anti-CD3 antibody or antigen-binding fragment are linked to opposing polypeptides of a heterodimeric Fc; the first and second components are coupled by a cleavable linker, and the heterodimeric Fc region is positioned N-terminal to the anti-CD3 antibody; one or both of the first and second components comprises at least one antigen-binding domain that binds to a tumor-associated antigen (TAA); One or both of the first and second components comprises at least one inhibitory receptor binding region (IRBR) that binds to an inhibitory receptor; Multispecific polypeptide constructs. [This invention 1099] The multispecific polypeptide construct of the present invention 1098, wherein one or both of the first and second components further comprises at least one costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor. [The present invention 1100] A multispecific polypeptide construct according to any one of claims 1096 to 1099, wherein the binding of the CD3 binding region to CD3 is substantially reduced when the multispecific polypeptide construct is in an uncleaved state compared to when the multispecific polypeptide construct is in a cleaved state. [The present invention 1101] 1100. The multispecific polypeptide construct of any of claims 1096 to 1100, wherein in the cleaved state, the first and second components are unlinked. [The present invention 1102] 1102. The multispecific polypeptide construct of any of claims 1096 to 1101, wherein the cleavable linker is a polypeptide that functions as a substrate for a protease. [The present invention 1103] The multispecific polypeptide construct of the present invention 1102, wherein the protease is produced by immune effector cells, by the tumor, or by cells present in the tumor microenvironment. [The present invention 1104] The multispecific polypeptide construct of the present invention 1103, wherein the protease is produced by an immune effector cell, and the immune effector cell is an activated T cell, a natural killer (NK) cell, or an NK T cell. [This invention 1105] The multispecific polypeptide construct of any of claims 1102 to 1104, wherein the protease is selected from matriptase, matrix metalloproteinase (MMP), granzyme B, and combinations thereof. [The present invention 1106] 1105. The multispecific polypeptide construct of the present invention, wherein the protease is Granzyme B. [This invention 1107] 1096-1106. The multispecific polypeptide construct of any of claims 1096 to 1106, wherein the cleavable linker comprises an amino acid sequence of the general formula P4 P3 P2 P1↓P1' (SEQ ID NO:150), wherein P4 is an amino acid I, L, Y, M, F, V, or A; P3 is an amino acid A, G, S, V, E, D, Q, N, or Y; P2 is an amino acid H, P, A, V, G, S, or T; P1 is an amino acid D or E; and P1' is an amino acid I, L, Y, M, F, V, T, S, G, or A. [This invention 1108] 1096-1107. The multispecific polypeptide construct of any of claims 1096 to 1107, wherein the cleavable linker comprises an amino acid sequence of the general formula P4 P3 P2 P1↓P1' (SEQ ID NO:151), wherein P4 is amino acid I or L; P3 is amino acid E; P2 is amino acid P or A; P1 is amino acid D; and P1' is amino acid I, V, T, S, or G. [This invention 1109] 109. The multispecific polypeptide construct of any of claims 1096 to 1108, wherein the cleavable linker comprises the amino acid sequence IEPDI (SEQ ID NO: 136), LEPDG (SEQ ID NO: 152), LEADT (SEQ ID NO: 137), IEPDG (SEQ ID NO: 138), IEPDV (SEQ ID NO: 139), IEPDS (SEQ ID NO: 140), IEPDT (SEQ ID NO: 141), or LEADG (SEQ ID NO: 153). [The present invention 1110] 109. The multispecific polypeptide construct of any of claims 1096 to 1109, wherein the cleavable linker comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 22, 105-112, 136-141, 148, 150-153. [The present invention 1111] 11. The multispecific polypeptide construct of any of claims 1096 to 1110, wherein the cleavable linker comprises the amino acid sequence shown in SEQ ID NO: 105. [The present invention 1112] 1111. The multispecific polypeptide construct of the present invention, wherein the protease is matriptase. [The present invention 1113] the cleavable linker comprises the sequence P1QAR↓(A / V) (SEQ ID NO:154), where P1 is any amino acid; or the cleavable linker comprises the sequence RQAR(A / V) (SEQ ID NO: 155); A multispecific polypeptide construct according to any one of claims 1096 to 1112. [This invention 1114] 13. The multispecific polypeptide construct of any of claims 1096 to 1113, wherein the cleavable linker comprises the sequence RQARV (SEQ ID NO: 156). [This invention 1115] 15. The multispecific polypeptide construct of any of claims 1096 to 1114, wherein the cleavable linker comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 23, 154 to 156. [The present invention 1116] The multispecific polypeptide construct of the present invention 1105, wherein the protease is an MMP. [This invention 1117] 1116. The multispecific polypeptide construct of the present invention, wherein the MMP is MMP-2. [This invention 1118] 1096-1117. The multispecific polypeptide construct of any of claims 1096-1117, wherein the cleavable linker comprises the general formula P3 P2 P1↓P1' (SEQ ID NO:157), where P3 is P, V, or A; P2 is Q or D; P1 is A or N; and P1' is L, I, or M. [This invention 1119] 1096-1118. The multispecific polypeptide construct of any of claims 1096 to 1118, wherein the cleavable linker comprises the general formula P3 P2 P1↓P1' (SEQ ID NO:158), where P3 is P; P2 is Q or D; P1 is A or N; and P1' is L or I. [The present invention 1120] 1096-1119. The multispecific polypeptide construct of any of claims 1096-1119, wherein the cleavable linker comprises the sequence PAGL (SEQ ID NO: 24). [This invention 1121] 11. The multispecific polypeptide construct of any of claims 1096 to 1120, wherein the cleavable linker comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 22-31, 104-114, 117-118, 136-144, 148, 150-158. [This invention 1122] The multispecific polypeptide construct comprises: (i) a first polypeptide comprising a first Fc polypeptide of a heterodimeric Fc region, a linker, and a VH domain or VL domain of an anti-CD3 antibody or antigen-binding fragment; and (ii) a second polypeptide comprising a second Fc polypeptide of the heterodimeric Fc region, a linker, optionally the same linker as that present in the first polypeptide, and the other of the VH domain or VL domain of the anti-CD3 antibody or antigen-binding fragment. At least one or both of the first and second polypeptides comprises at least one antigen-binding domain that binds to a tumor-associated antigen (TAA), and one or both of the first and second components comprises at least one costimulatory receptor-binding region (CRBR) that binds to a costimulatory receptor; The multispecific polypeptide construct comprises at least one CRBR and at least one antigen-binding domain. A multispecific polypeptide construct according to any one of claims 1069 to 1121 of the present invention. [This invention 1123] 10. The multispecific polypeptide construct of any of claims 1001 to 1122, wherein one or more antigen-binding domains that bind to the TAA result in monovalent, divalent, trivalent, or tetravalent binding to the TAA. [This invention 1124] The multispecific polypeptide construct of invention 1122 or invention 1123, wherein only one of the first or second polypeptide comprises at least one antigen binding domain that binds to a TAA. [Invention 1125] 1125. The multispecific polypeptide construct of any of claims 1122 to 1124, wherein at least one antigen-binding domain is positioned amino-terminal to an Fc region and / or carboxy-terminal to a CD3-binding region of one of the first or second polypeptides of the multispecific polypeptide construct. [The present invention 1126] 1125. The multispecific polypeptide construct of any of claims 1122 to 1124, wherein at least one antigen-binding domain is positioned amino-terminal to the Fc region of the multispecific construct and a second antigen-binding domain is positioned carboxy-terminal to the CD3-binding region of the multispecific construct. [This invention 1127] 1127. The multispecific polypeptide construct of any of claims 1122 to 1126, wherein only one of the first or second polypeptides comprises at least one costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor. [This invention 1128] 1128. The multispecific polypeptide construct of any of claims 1122 to 1127, wherein the costimulatory receptor binding region (CRBR) is positioned amino-terminal to the Fc region of one of the first or second polypeptides of the multispecific polypeptide construct, or carboxy-terminal to the CD3 binding region. [This invention 1129] the first polypeptide comprises, in order from N-terminus to C-terminus, a first antigen-binding domain that binds to a tumor-associated antigen (TAA), a first Fc polypeptide of the heterodimeric Fc region, a linker, a VL or VH of the anti-CD3 antibody or antigen-binding fragment, and a second antigen-binding domain that binds to the tumor-associated antigen (TAA); the second polypeptide comprises, in order from N-terminus to C-terminus, a second Fc polypeptide of the heterodimeric Fc region, a linker, optionally the same linker as that present in the first polypeptide, the other of the VL or VH of the anti-CD3 antibody or antigen-binding fragment, and a costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor; A multispecific polypeptide construct according to any one of claims 1122 to 1128 of the present invention. [The present invention 1130] 1129. The multispecific polypeptide construct of any of claims 1122 to 1129, wherein the heterodimeric Fc region comprises an Fc hole polypeptide and an Fc knob polypeptide, wherein the VL of the anti-CD3 antibody or antigen-binding fragment is positioned C-terminal to the Fc hole and the VH of the anti-CD3 antibody or antigen-binding fragment is positioned C-terminal to the Fc knob. [This invention 1131] 10. The multispecific polypeptide construct of any of claims 1001 to 1130, wherein the antigen-binding domain, or each of the antigen-binding domains independently, comprises the extracellular domain of the native cognate binding partner of the TAA or a binding fragment thereof, or a variant thereof that exhibits binding activity with the TAA. [This invention 1132] The multispecific polypeptide construct of any of claims 1001 to 1131, wherein the antigen-binding domain, or each of the antigen-binding domains independently, is an antibody or antigen-binding fragment thereof selected from the group consisting of a Fab fragment, a F(ab')2 fragment, an Fv fragment, an scFv, an scAb, a dAb, a single-domain heavy chain antibody, and a single-domain light chain antibody. [This invention 1133] The multispecific polypeptide construct of any of claims 1001 to 1132, wherein at least one costimulatory receptor binding region (CRBR) is or comprises the extracellular domain of a native cognate binding partner of the costimulatory receptor or a binding fragment thereof, or a variant thereof that exhibits binding activity with the costimulatory receptor. [This invention 1134] The multispecific polypeptide construct of any of claims 1001 to 1132, wherein at least one costimulatory receptor binding region (CRBR) is an antibody or an antigen-binding fragment thereof selected from the group consisting of a Fab fragment, a F(ab')2 fragment, an Fv fragment, a scFv, a scAb, a dAb, a single-domain heavy chain antibody, and a single-domain light chain antibody. [This invention 1135] The multispecific polypeptide construct of the invention 1132 or 1134, wherein the antibody or antigen-binding fragment thereof is an Fv, scFv, Fab, or single domain antibody (sdAb). [This invention 1136] The multispecific polypeptide construct of invention 1132, invention 1134 or invention 1122, wherein the antibody or antigen-binding fragment is an sdAb. [This invention 1137] 1001 to 1136. The multispecific polypeptide construct of any of claims 1001 to 1136, wherein one or both of the first and second components comprise an antigen-binding domain that binds to a tumor-associated antigen (TAA), and the antigen-binding domain is a single-chain antibody fragment. [This invention 1138] The multispecific polypeptide construct of the present invention 1137, wherein the single-chain antibody fragment is a single-domain antibody or a single-chain variable fragment (scFv). [This invention 1139] 1139. The multispecific polypeptide construct of any of claims 1136 to 1138, wherein the sdAb is a human sdAb or a humanized sdAb. [This invention 1140] 1139. The multispecific polypeptide construct of any of claims 1136 to 1139, wherein the sdAb is a VHH, a VNAR, a modified VH domain, or a modified VK domain. [This invention 1141] The multispecific polypeptide construct of the present invention 1132, 1134, or 1135, wherein the antibody or antigen-binding fragment thereof is an scFv. [This invention 1142] The multispecific polypeptide construct of the present invention 1132, 1134 or 1135, wherein the antibody or antigen-binding fragment thereof is a Fab. [This invention 1143] The multispecific polypeptide construct comprises: (i) a first polypeptide comprising a first Fc polypeptide of a heterodimeric Fc region, a linker, and a VH domain of an anti-CD3 antibody or antigen-binding fragment; (ii) a second polypeptide comprising a second Fc polypeptide of the heterodimeric Fc region, a linker, and a VL domain of an anti-CD3 antibody or antigen-binding fragment; (iii) a third polypeptide comprising VH-CH1(Fd) or VL-CL of a Fab antibody fragment that binds to a tumor-associated antigen; (iv) a fourth polypeptide comprising the VH-CH1(Fd) or VL-CL of a Fab antibody fragment that binds to a costimulatory receptor. Including, the first and / or second polypeptide is (1) the other of VH-CH1(Fd) or VL-CL of a Fab antibody fragment that binds to a tumor-associated antigen, and (2) the other of VH-CH1(Fd) or VL-CL of the Fab antibody fragment that binds to the costimulatory receptor further comprising: A multispecific polypeptide construct according to any one of 1001 to 1132 and 1134 to 1142 of the present invention. [This invention 1144] The multispecific polypeptide construct comprises: (i) a first polypeptide comprising a first Fc polypeptide of a heterodimeric Fc region, a linker, and a VH domain of an anti-CD3 antibody or antigen-binding fragment; (ii) a second polypeptide comprising a second Fc polypeptide of the heterodimeric Fc region, a linker, and a VL domain of an anti-CD3 antibody or antigen-binding fragment; and (iii) a third polypeptide comprising the VH-CH1(Fd) or VL-CL of a Fab antibody fragment that binds to a costimulatory receptor; Including, the first and / or second polypeptide further comprises the other of VH-CH1(Fd) or VL-CL of a Fab antibody fragment that binds to a costimulatory receptor; the first and / or second polypeptide further comprises at least one antigen-binding domain that binds to a tumor-associated antigen (TAA); A multispecific polypeptide construct according to any one of 1001 to 1132 and 1134 to 1142 of the present invention. [Invention 1145] The multispecific polypeptide construct comprises: (i) a first polypeptide comprising a first Fc polypeptide of a heterodimeric Fc region, a linker, and a VH domain of an anti-CD3 antibody or antigen-binding fragment; (ii) a second polypeptide comprising a second Fc polypeptide of the heterodimeric Fc region, a linker, and a VL domain of an anti-CD3 antibody or antigen-binding fragment; and (iii) a third polypeptide comprising VH-CH1(Fd) or VL-CL of a Fab antibody fragment that binds to a tumor-associated antigen; Including, the first and / or second polypeptide further comprises the other of VH-CH1(Fd) or VL-CL of a Fab antibody fragment that binds to a tumor-associated antigen; the first and / or second polypeptide further comprises at least one costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor; A multispecific polypeptide construct according to any one of 1001 to 1132 and 1134 to 1142 of the present invention. [Invention 1146] The multispecific polypeptide construct of the invention 1144 or 1145, wherein only one of the first or second polypeptides comprises the other of the VH-CH1(Fd) or VL-CL of the Fab antibody fragment. [This invention 1147] The multispecific polypeptide construct of the present invention 1144 or 1145, wherein both the first or second polypeptide comprises the other of the VH-CH1(Fd) or VL-CL of the Fab antibody fragment. [Invention 1148] The multispecific polypeptide construct of the present invention 1146 or 1147, wherein the other of the VH-CH1(Fd) or VL-CL of the Fab antibody fragment is positioned amino-terminal to the Fc region and / or carboxy-terminal to the CD3 binding region of one of the first or second polypeptides of the multispecific polypeptide construct. [This invention 1149] 1149. The multispecific polypeptide construct of any of claims 1146 to 1148, wherein the other of the VH-CH1(Fd) or VL-CL of the Fab antibody fragment is positioned amino-terminal to the Fc region of the first polypeptide or the second polypeptide and carboxy-terminal to the CD3 binding region of the other of the first or second polypeptide. [This invention 1150] The antigen-binding domain, or each antigen-binding domain independently, may be: 1-92-LFA-3, 5T4, α4 integrin, αV integrin, α4β1 integrin, α4β7 integrin, AGR2, anti-Lewis Y, apelin J receptor, APRIL, B7-H3, B7-H4, BAFF, BTLA, C5 complement, C-242, CA9, CA19-9 (Lewis a), carbonic anhydrase 9, CD2, CD3, CD6, CD9, CD11 a, CD19, CD20, CD22, CD24, CD25, CD27, CD28, CD30, CD33, CD38, CD40, CD40L, CD41, CD44, CD44v6, CD47, CD51, CD52 , CD56, CD64, CD70, CD71, CD74, CD80, CD81, CD86, CD95, CD117, CD123, CD125, CD132(IL-2RG), CD133, CD137, CD138 , CD166, CD172A, CD248, CDH6, CEACAM5 (CEA), CEACAM6 (NCA-90), claudin 3, claudin 4, cMet, collagen, Cripto, CSFR, CSFR-1, CTLA-4, CTGF, CXCL10, CXCL13, CXCR1, CXCR2, CXCR4, CYR61, DL44, DLK1, DLL3, DLL4, DPP-4, DSG1, EDA, EDB , EGFR, EGFRviii, endothelin B receptor (ETBR), ENPP3, EpCAM, EPHA2, EPHB2, ERBB3, RSV F protein, FAP, FGF-2, FGF8, FGFR1, FGFR2, FGFR3, FGFR4, FLT-3, folate receptor alpha (FRα), GAL3ST1, G-CSF, G-CSFR, GD2, GITR, GLUT1, GLUT4, GM-CSF, GM-CSFR, GPIIb / IIIa receptor, Gp130, GPIIB / IIIA, GPNMB, GRP78, HER2 / neu, HER3, HER4, HGF, hGH, HVEM, hyaluronidase, ICOS, IFNα, IFNβ, IFNγ, IgE, IgE receptor (FceRI), IGF, IGF1R, IL1B, IL1R, IL2, IL11, IL12, IL12p40, IL-12R, IL-12Rβ1, IL13, IL13R, IL15, IL17, IL18, IL21, I L23, IL23R, IL27 / IL27R (wsx1), IL29, IL-31R, IL31 / IL31R, IL2R, IL4, IL4R, IL6, IL6R, insulin receptor, Jagged ligand, Jagged1, Jagged2, KISS1-R, LAG-3, LIF-R, Lewis X, LIGHT, LRP4, LRRC26, Ly6G6D, LyPD1, MCSP, mesothelin, MRP4, MUC1, mucin 16 (MUC16, CA-125), Na / K ATPase, NGF, Nicastrin, Notch receptor, Notch1, Notch2, Notch3, Notch4, NOV, OSM-R, OX-40, PAR2, PDGF-AA, PDGF-BB, PDGFRα, PDGFRβ, PD-1, PD-L1, PD-L2, phosphatidylserine, P1GF, PSCA, PSMA, PSGR, RAAG12, RAGE, SLC44A4, sphingosine 1 phosphate, STEAP1, STEAP2, TAG-72, TAPA1, TEM-8, TGFβ, TIGIT, TIM-3, TLR2, TLR4, 1001-1149. The multispecific polypeptide construct of any of claims 1001 to 1149, which binds to a tumor antigen selected from TLR6, TLR7, TLR8, TLR9, TMEM31, TNFα, TNFR, TNFRS12A, TRAIL-R1, TRAIL-R2, transferrin, transferrin receptor, TRK-A, TRK-B, uPAR, VAP1, VCAM-1, VEGF, VEGF-A, VEGF-B, VEGF-C, VEGF-D, VEGFR1, VEGFR2, VEGFR3, VISTA, WISP-1, WISP-2, and WISP-3. [This invention 1151] The multispecific polypeptide construct of any of claims 1001 to 1150, wherein the antigen-binding domain comprises at least a first antigen-binding domain and a second antigen-binding domain, and the first antigen-binding domain and the second antigen-binding domain bind to the same TAA. [This invention 1152] 1151. The multispecific polypeptide construct of the present invention, wherein the first antigen-binding domain and the second antigen-binding domain bind to different epitopes of the same TAA. [This invention 1153] 1151. The multispecific polypeptide construct of the present invention, wherein the first antigen-binding domain and the second antigen-binding domain bind to the same epitope of the same TAA. [This invention 1154] The multispecific polypeptide construct of any of claims 1001 to 1153, wherein the antigen-binding domain comprises at least a first antigen-binding domain and a second antigen-binding domain, and the first antigen-binding domain and the second antigen-binding domain bind to different TAAs. [This invention 1155] The multispecific polypeptide construct of any of claims 1001, 1003 to 1017, 1020 to 1025, 1028 to 1097, and 1099 to 1154, wherein the costimulatory receptor binding region (CRBR) comprises at least a first CRBR and a second CRBR, and the first CRBR and the second CRBR bind to the same costimulatory receptor. [Invention 1156] The multispecific polypeptide construct of the present invention 1155, wherein the first costimulatory receptor binding region (CRBR) and the second CRBR bind to different epitopes of the same costimulatory receptor. [This invention 1157] The multispecific polypeptide construct of the present invention 1155, wherein the first costimulatory receptor binding region (CRBR) and the second CRBR bind to the same epitope of the same costimulatory receptor. [This invention 1158] The multispecific polypeptide construct of any of claims 1001, 1003 to 1017, 1020 to 1025, 1028 to 1097, and 1099 to 1157, wherein the costimulatory receptor binding region (CRBR) comprises at least a first CRBR and a second CRBR. [This invention 1159] 1159. The multispecific polypeptide construct of any of claims 1155 to 1158, wherein the first CRBR and the second CRBR bind to different costimulatory receptors. [The present invention 1160] The multispecific polypeptide construct of any of claims 1001, 1003 to 1017, 1020 to 1025, 1028 to 1097, and 1099 to 1159, wherein at least one costimulatory receptor binding region (CRBR) binds to a costimulatory receptor selected from 41BB (CD137), OX40 (CD134), CD27, glucocorticoid-inducible TNFR-related protein (GITR), CD28, ICOS, CD40, B-cell activating factor receptor (BAFF-R), B-cell maturation antigen (BCMA), transmembrane activator and CAML interactor (TACI), and NKG2D. [This invention 1161] The multispecific polypeptide construct of any of claims 1001, 1003 to 1017, 1020 to 1025, 1028 to 1097, and 1099 to 1160, wherein at least one costimulatory receptor binding region (CRBR) binds to a costimulatory receptor selected from 41BB (CD137), OX40 (CD134), and glucocorticoid-inducible TNFR-related protein (GITR). [This invention 1162] The multispecific polypeptide construct of any of claims 1001, 1003 to 1017, 1020 to 1025, 1028 to 1097, and 1099 to 1161, wherein at least one costimulatory receptor binding region (CRBR) binds to 41BB (CD137). [This invention 1163] The multispecific polypeptide construct of any of claims 1001, 1003 to 1017, 1020 to 1025, 1028 to 1097, and 1099 to 1162, wherein at least one costimulatory receptor comprises the amino acid sequence set forth in SEQ ID NO:215 or an amino acid sequence having at least 99% sequence identity to the amino acid sequence set forth in SEQ ID NO:215. [This invention 1164] The multispecific polypeptide construct of any of claims 1001, 1003-1017, 1020-1025, 1028-1097, and 1099-1163, wherein at least one costimulatory receptor comprises the amino acid sequence set forth in SEQ ID NO:321. [Invention 1165] The multispecific polypeptide construct of any of claims 1002 to 1015, 1018 to 1023, 1026 to 1095, 1097 to 1121, and 1123 to 1164, wherein at least one inhibitory receptor binding region (IRBR) is or comprises the extracellular domain of a native cognate binding partner of a costimulatory receptor or a binding fragment thereof, or a variant thereof that exhibits binding activity with a costimulatory receptor. [Invention 1166] The multispecific polypeptide construct of any of claims 1002 to 1015, 1018 to 1023, 1026 to 1095, 1097 to 1121, and 1123 to 1164, wherein at least one inhibitory receptor binding region (IRBR) is an antibody or an antigen-binding fragment thereof selected from the group consisting of a Fab fragment, an F(ab')2 fragment, an Fv fragment, an scFv, an scAb, a dAb, a single-domain heavy chain antibody, and a single-domain light chain antibody. [This invention 1167] 1166. The multispecific polypeptide construct of the present invention, wherein the antibody or antigen-binding fragment thereof is an Fv, scFv, Fab, or single domain antibody (sdAb). [Invention 1168] The multispecific polypeptide construct of the invention 1166 or 1122, wherein the antibody or antigen-binding fragment thereof is an sdAb. [This invention 1169] 1168. The multispecific polypeptide construct of the invention, wherein the sdAb is a human sdAb or a humanized sdAb. [This invention 1170] The multispecific polypeptide construct of the present invention 1168 or 1169, wherein the sdAb is a VHH, a VNAR, a modified VH domain, or a modified VK domain. [This invention 1171] The multispecific polypeptide construct of the present invention 1166 or 1167, wherein the antibody or antigen-binding fragment thereof is an scFv. [This invention 1172] The multispecific polypeptide construct of the present invention 1166 or 1167, wherein the antibody or antigen-binding fragment thereof is a Fab. [This invention 1173] The multispecific polypeptide construct of any of claims 1002 to 1015, 1018 to 1023, 1026 to 1095, 1097 to 1121, and 1123 to 1172, wherein the inhibitory receptor binding region (IRBR) comprises at least a first IRBR and a second IRBR, and the first IRBR and the second IRBR bind to the same inhibitory receptor. [This invention 1174] The multispecific polypeptide construct of the present invention 1173, wherein the first inhibitory receptor binding region (IRBR) and the second IRBR bind to different epitopes of the same inhibitory receptor. [Invention 1175] The multispecific polypeptide construct of the present invention 1173, wherein the first inhibitory receptor binding region (IRBR) and the second IRBR bind to the same epitope of the same inhibitory receptor. [Invention 1176] The multispecific polypeptide construct of any one of 1002 to 1015, 1018 to 1023, 1026 to 1095, 1097 to 1121, and 1123 to 1175, wherein the inhibitory receptor binding region (IRBR) comprises at least a first IRBR and a second IRBR. [This invention 1177] 1177. The multispecific polypeptide construct of any of claims 1173 to 1176, wherein the first IRBR and the second IRBR bind to different inhibitory receptors. [This invention 1178] The multispecific polypeptide construct of any of claims 1002 to 1015, 1018 to 1023, 1026 to 1095, 1097 to 1121, and 1123 to 1177, wherein at least one receptor binding region (IRBR) binds to an inhibitory receptor selected from PD-1, CTLA-4, TIGIT, VISTA, or TIM3. [This invention 1179] The multispecific polypeptide construct of any of claims 1002 to 1015, 1018 to 1023, 1026 to 1095, 1097 to 1121, and 1123 to 1178, wherein at least one inhibitory receptor binding region (IRBR) binds to PD-1. [This invention 1180] 1001 to 1179. The multispecific polypeptide construct of any of claims 1001 to 1179, comprising a first connecting peptide (LP1) between the first antigen-binding domain and the Fc region. [This invention 1181] 11. The multispecific polypeptide construct of any of claims 1001 to 1180, comprising a second connecting peptide (LP2) between the CD3 binding region and the second antigen binding domain. [This invention 1182] The multispecific polypeptide construct of any of claims 1001 to 1180, comprising a first connecting peptide (LP1) between the costimulatory receptor binding region (CRBR) and the Fc region. [This invention 1183] 1001-1182. The multispecific polypeptide construct of any of claims 1001-1182, comprising a second connecting peptide (LP2) between the CD3 binding region and the second costimulatory receptor binding region (CRBR). [This invention 1184] a first connecting peptide (LP1) between the antigen-binding domain or costimulatory receptor-binding region and the Fc region, and a second connecting peptide (LP2) between the CD3-binding region and the antigen-binding domain or CRBR; having the following structural arrangement from N-terminus to C-terminus: first antigen-binding domain or CRBR - LP1 - Fc region - linker - CD3-binding region - LP2 - second antigen-binding domain or CRBR; A multispecific polypeptide construct according to any one of claims 1001 to 1183 of the present invention. [This invention 1185] The multispecific polypeptide construct of the present invention 1184, wherein the linker is a cleavable linker. [Invention 1186] The multispecific polypeptide construct of the present invention 1184, wherein the linker is a non-cleavable linker. [This invention 1187] 1187. The multispecific polypeptide construct of any of claims 1184 to 1186, wherein the two connecting peptides are not identical to each other. [This invention 1188] 1188. The multispecific polypeptide construct of any of claims 1180 to 1187, wherein LP1 or LP2 is independently a peptide of about 1 to 20 amino acids in length. [This invention 1189] 1188. The multispecific polypeptide construct of the present invention, wherein LP1 or LP2 independently comprise a peptide which is or comprises a Gly-Ser linker or GGS as shown in SEQ ID NOs: 10-13, 119, 135, 147, 149. [This invention 1190] 1001 to 1189. The multispecific polypeptide construct of any of claims 1001 to 1189, wherein the anti-CD3 antibody or antigen-binding fragment is an Fv antibody fragment. [This invention 1191] 1190. The multispecific polypeptide construct of the present invention, wherein the Fv antibody fragment is a disulfide-stabilized anti-CD3 binding Fv fragment (dsFv). [This invention 1192] The anti-CD3 antibody or antigen-binding fragment is VH CDR1 containing the amino acid sequence TYAMN (SEQ ID NO: 16); amino acid sequence TIFF0007651450000010.tif4128 a VH CDR3 comprising the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO:18); a VL CDR1 comprising the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO:19); a VL CDR2 comprising the amino acid sequence GTNKRAP (SEQ ID NO:20); and a VL CDR3 comprising the amino acid sequence ALWYSNLWV (SEQ ID NO:21). The multispecific polypeptide construct of any one of claims 1001 to 1191, comprising: [This invention 1193] Anti-CD3 dsFv, a VH having the amino acid sequence of any of SEQ ID NOs: 14, 32-62, 287, 290, and 311, or a sequence exhibiting at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to any of SEQ ID NOs: 14, 32-62, 287, 290, and 311; and VL having an amino acid sequence of any of SEQ ID NOs: 15, 63-81, 241, 288, and 289, or a sequence that exhibits at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to any of SEQ ID NOs: 15, 63-81, 241, 288, and 289. The multispecific polypeptide construct of the present invention 1160 or 1161, comprising: [This invention 1194] The multispecific polypeptide construct of any of claims 1190 to 1193, wherein the anti-CD3 dsFv comprises the amino acid sequence of SEQ ID NO:14 and the amino acid sequence of SEQ ID NO:15. [This invention 1195] The multispecific polypeptide construct of any of claims 1190 to 1194, wherein the anti-CD3 dsFv comprises the amino acid sequence of SEQ ID NO:44 and the amino acid sequence of SEQ ID NO:72; the anti-CD3 dsFv comprises the amino acid sequence of SEQ ID NO:44 and the amino acid sequence of SEQ ID NO:241; the anti-CD3 dsFv comprises the amino acid sequence of SEQ ID NO:287 and the amino acid sequence of SEQ ID NO:288; or the anti-CD3 dsFv comprises the amino acid sequence of SEQ ID NO:311 and the amino acid sequence of SEQ ID NO:289. [Invention 1196] 1001-1195. The multispecific polypeptide construct of any of claims 1001-1195, conjugated to an agent. [This invention 1197] 1196. The multispecific polypeptide construct of the present invention, wherein the agent is a therapeutic agent, an anti-tumor agent, a toxin or fragment thereof, a detectable moiety, or a diagnostic agent. [This invention 1198] 1197. The multispecific polypeptide construct of the present invention, wherein the agent is conjugated to the multispecific polypeptide construct via a linker. [This invention 1199] A polynucleotide encoding any one of the multispecific polypeptide constructs of the present inventions 1001 to 1198. [The present invention 1200] A polynucleotide encoding any one of the polypeptide chains of the multispecific polypeptide constructs of the present invention according to any one of 1001 to 1198. [The present invention 1201] A polynucleotide comprising a first nucleic acid sequence encoding a first polypeptide of a multispecific construct of any one of claims 1001 to 1198 and a second nucleic acid sequence encoding a second polypeptide of the multispecific construct, the first and second nucleic acid sequences are separated by an internal ribosome entry site (IRES) or by a nucleic acid encoding a self-cleaving peptide or a peptide that causes ribosome skipping; Polynucleotide. [This invention 1202] The polynucleotide of the present invention 1201, wherein the first nucleic acid sequence and the second nucleic acid sequence are operably linked to the same promoter. [This invention 1203] The polynucleotide of claim 1201 or claim 1202, wherein the multispecific polypeptide construct comprises a third polypeptide chain and the polynucleotide further comprises a third nucleic acid encoding the third polypeptide of the multispecific construct. [The present invention 1204] the third nucleic acid is separated from the first and / or second polypeptide by an internal ribosome entry site (IRES) or by a nucleic acid encoding a self-cleaving peptide or a peptide that causes ribosome skipping; and / or the third nucleic acid sequence is operably linked to the same promoter as the first and / or second nucleic acid sequence; The polynucleotide of the present invention 1203. [This invention 1205] The polynucleotide of any one of 1201 to 1204, wherein the nucleic acid encoding the self-cleaving peptide or the peptide that causes ribosome skipping is selected from T2A, P2A, E2A, and F2A. [This invention 1206] A vector comprising any one of the polynucleotides of the present invention 1199 to 1205. [This invention 1207] The vector of the present invention 1206, which is an expression vector. [This invention 1208] 12. The vector of claim 1206 or 1207, which is a viral vector or a eukaryotic vector, and optionally, the eukaryotic vector is a mammalian vector. [This invention 1209] A cell comprising one or more polynucleotides according to any one of 1199 to 1205 of the present invention or one or more vectors according to any one of 1206 to 1208 of the present invention. [The present invention 1210] The cell of the present invention 1209, which is recombinant or isolated. [The present invention 1211] The cell of the present invention 1210, which is a mammalian cell. [The present invention 1212] The cell of the present invention 1211, which is a HEK293 cell or a CHO cell. [This invention 1213] A step of introducing one or more polynucleotides according to any one of claims 1199 to 1205 of the present invention or one or more vectors according to any one of claims 1206 to 1208 of the present invention into a cell; Culturing the cells under conditions such that the multispecific polypeptide construct is produced. 1. A method for producing a multispecific polypeptide construct, comprising: [This invention 1214] A method for producing a multispecific polypeptide construct, comprising the step of culturing any of the cells of the present inventions 1209 to 1213 under conditions in which the multispecific polypeptide is produced by the cells. [This invention 1215] The cell of the present invention 1213 or 1214, which is a mammalian cell. [This invention 1216] The cell of the present invention 1215, which is a HEK293 cell or a CHO cell. [This invention 1217] The method of claim 1213 or claim 1214, further comprising the step of isolating or purifying the multispecific polypeptide construct from the cells. [This invention 1218] 1218. The method of any of claims 1213 to 1217, wherein the multispecific polypeptide construct is a heterodimer. [This invention 1219] A multispecific polypeptide construct produced by any of the methods of 1213 to 1218 of the present invention. [The present invention 1220] A pharmaceutical composition comprising any one of the multispecific polypeptide constructs of the present inventions 1001 to 1198 or 1219 and a pharmaceutically acceptable carrier. [This invention 1221] The pharmaceutical composition of the present invention 1220, which is sterile. [This invention 1222] A method for stimulating or inducing an immune response, comprising the step of contacting target cells and T cells with any of the multispecific polypeptide constructs of inventions 1001 to 1198 or 1219, or the pharmaceutical composition of invention 1220 or 1221, wherein the target cells express a tumor-associated antigen recognized by the multispecific polypeptide construct. [This invention 1223] The method of claim 1222, wherein the target cell is a tumor cell expressing a tumor-associated antigen (TAA). [This invention 1224] The method of claim 1222 or claim 1223, wherein the multispecific polypeptide construct comprises a cleavable linker that functions as a substrate for a protease, and wherein the induction or stimulation of an immune response is increased in the presence of the protease. [This invention 1225] The method of claim 1224, wherein the protease is produced by an immune effector cell, by the tumor, or by cells present in the tumor microenvironment. [This invention 1226] The method of claim 1224 or claim 1225, wherein the protease is produced by an immune effector cell, and the immune effector cell is an activated T cell, a natural killer (NK) cell, or an NK T cell. [This invention 1227] The method of claim 1226, wherein the immune effector cells are in proximity to cells expressing the antigen. [This invention 1228] the protease is produced by the tumor in the vicinity of cells expressing the TAA in the tissue and / or is co-localized with the TAA in the tissue; when the multispecific polypeptide construct is exposed to a protease, the protease cleaves the cleavable linker within the multispecific polypeptide construct; Any of the methods of 1222 to 1227 of the present invention. [This invention 1229] The method of any of claims 1222 to 1228, wherein the protease is selected from matriptase, matrix metalloproteinase (MMP), granzyme B, and combinations thereof. [This invention 1230] 1229. The method of claim 1229, wherein the protease is granzyme B. [This invention 1231] The method of any of claims 1222 to 1230, wherein the contacting step is carried out ex vivo or in vitro. [This invention 1232] The method of any of claims 1222 to 1231, wherein the contacting step is carried out in vivo in a subject. [This invention 1233] A method for stimulating or inducing an immune response in a subject, comprising the step of administering a therapeutically effective amount of any of the multispecific conjugates of the present inventions 1001 to 1198 or 1219, or the pharmaceutical composition of the present invention 1220 or 1221, to a subject in need thereof. [This invention 1234] The method of any one of claims 1222 to 1233, which increases cellular immunity. [This invention 1235] The method of any of claims 1222 to 1234 of the present invention, which increases T cell activity. [This invention 1236] The method of any one of claims 1222 to 1235, which increases cytolytic T cell (CTL) activity. [This invention 1237] The method of any of claims 1222 to 1236, wherein the immune response against the tumor or cancer is increased. [This invention 1238] The method of any of claims 1222 to 1236 for treating a disease or condition in a subject. [This invention 1239] A method for treating a disease or condition in a subject, comprising administering a therapeutically effective amount of any of the multispecific polypeptide constructs of the present inventions 1001 to 1198 or the pharmaceutical composition of the present invention 1220 or 1221 to a subject in need thereof. [This invention 1240] The method of claim 1238 or claim 1239, wherein the disease or condition is a tumor or cancer. [This invention 1241] The method of any of claims 1238 to 1240, wherein the disease or condition expresses a tumor-associated antigen (TAA). [This invention 1242] The method of any one of claims 1222 to 1241, wherein the subject is a human. [This invention 1243] A multispecific polypeptide construct of any of inventions 1001 to 1198 or a pharmaceutical composition of invention 1220 or invention 1221 for use in treating a disease or condition in a subject. [This invention 1244] Use of a multispecific polypeptide construct of any of inventions 1001 to 1198 or a pharmaceutical composition of invention 1220 or invention 1221 for the manufacture of a medicament for use in treating a disease or condition in a subject. [Invention 1245] A multispecific polypeptide construct or pharmaceutical composition for use according to the present invention 1243, or the use according to the present invention 1244, wherein the disease or condition is a tumor or cancer. [Invention 1246] A multispecific polypeptide construct or pharmaceutical composition for use of 1243 of the invention or 1245 of the invention, or the use of 1244 of the invention or 1245 of the invention, wherein the disease or condition expresses a tumor-associated antigen (TAA). [This invention 1247] A multispecific polypeptide construct or pharmaceutical composition for use according to the present invention 1243, the present invention 1245, or the present invention 1246, or the use of the present invention 1244, the present invention 1245, or the present invention 1246, wherein the subject is a human. [Brief explanation of the drawings]

[0063] [Figure 1]Figure 1 is a schematic diagram of the basic components of a multispecific polypeptide construct of the present disclosure with constrained CD3 binding. The antigen-binding domain is located at the amino- and / or carboxy-terminus. An Fc region, such as a heterodimeric Fc region, is located N-terminal to the CD3-binding region. This positioning of the Fc in close proximity to the CD3-binding region interferes with CD3 binding. A costimulatory receptor binding region (CRBR) is located at the amino- and / or carboxy-terminus. The illustrated exemplary multispecific construct may further contain a second CRBR if the multispecific molecule has one or two antigen-binding domains. [Figure 2] Figures 2A-2F illustrate various representative B7H3-targeted restricted CD3 engagers, with or without CRBR. Figure 2A illustrates a representative B7H3-targeted restricted CD3 engager that does not contain a 4-1BB binding domain as CRBR (designated cx3095 or cx3834). Because cx3834 has an opposite knob and hole configuration compared to cx3095, the first polypeptide chain of cx3834 depicted in Figure 2A contains an Fc hole polypeptide, and the second polypeptide depicted has an Fc knob polypeptide. Figure 2B illustrates a representative B7H3-targeted restricted CD3 engager that contains a 4-1BB binding domain as CRBR (designated cx3091). cx3091 has a 4-1BB-targeting sdAb positioned at the C-terminus of the opposite chain of the heterodimer, thereby exhibiting monovalent binding to 4-1BB. Representative constructs cx3095, cx3834, and cx3091 each have a B7H3-targeting sdAb positioned at the N- and C-termini of one chain of the heterodimer, thereby exhibiting bivalent binding to B7H3. Figures 2C-2F illustrate representative B7H3-targeted restricted CD3 engagers that include a B7H3-targeting sdAb positioned at the N- and C-termini of one chain of the heterodimer and a CRBR (e.g., OX40, GITR, ICOS, or CD28 sdAb) positioned at the C-terminus of the opposite chain of the heterodimer. [Figure 3]Figure 3 illustrates representative 5T4-targeted restricted CD3 engagers, one without (top) and one with (bottom) a 41BB-binding domain. cx3546 and cx3547 have the same 5T4-targeting sdAb (12E9) positioned at the N-terminus of one chain of the heterodimer and a different 5T4-targeting sdAb (14B5 or 16G10, respectively) positioned at the C-terminus of one chain of the heterodimer. cx3499 and cx3497 are identical to cx3546 and cx3547, respectively, but have a 41BB-targeting sdAb positioned at the C-terminus of the opposite chain of the Fc heterodimer. cx3546, cx3547, cx3499, and cx3497 each contain a cleavable linker between the Fc region and the CD3-binding region. cx5185 and cx5951 are identical except that cx5185 contains a 4-1BB-targeting sdAb positioned C-terminal to the CD3-binding region; both contain a non-cleavable linker between the Fc region and the CD3-binding region, and contain humanized versions of the 5T4-targeting sdAb, hz12E9v9 and hz16G10v11, positioned at the N- and C-termini of the Fc heterodimer. These representative 5T4-targeted restricted CD3 engagers exhibit bivalent binding to 5T4, and in constructs containing the 41BB-binding domain, monovalent binding to 41BB. [Figure 4]Figures 4A-4D illustrate cell binding by representative B7H3-targeting restricted CD3 engager constructs cx3091 (containing the 41BB binding domain) and cx3095 (not containing the 41BB binding domain) compared to the DART-Fc format B7H3xCD3. Figures 4A and 4B show binding of each of the evaluated constructs to A375 cells (a B7H3-positive melanoma cell line). Figures 4C and 4D illustrate binding to T cells, demonstrating that only the DART-Fc format B7H3xCD3 binds, while restricted CD3 engagers containing the 41BB binding domain (cx3091) or lacking the 41BB binding domain (cx3095) do not. Figures 4A and 4C show histograms of normalized cell number versus fluorescence at 200 nM for each construct. Full titration of each construct in various cell types is shown in Figures 4B and 4D. In Figures 4A and 4C, the secondary anti-human APC antibody-only control is shown as a solid black trace, binding of the positive control anti-CD3 is shown as a white trace, and binding of cx3091, cx3095, and DART-Fc B7H3xCD3 is shown as a gray shaded trace in Figures 4A and 4C. [Figure 5] Figures 5A-5D illustrate cell binding by representative constrained CD3-engaging constructs cx3497 (containing the 41BB-binding domain) and cx3547 (not containing the 41BB-binding domain) targeting 5T4. Figures 5A and 5B show binding to Ovcar-5 cells (a 5T4-positive melanoma cell line). Figures 5C and 5D illustrate binding to T cells, showing that the tested constructs do not bind to isolated T cells. Figures 5A and 5C show histograms of normalized cell number versus fluorescence at 200 nM for each construct. A complete titration of each construct in various cell types is shown in Figures 5B and 5D. In Figures 5A and 5C, the secondary anti-human APC antibody only control is shown as a solid black trace, binding of the positive control anti-CD3 is shown as a white trace, and cx3497 and cx3547 are shown as gray shaded traces in Figures 5A and 5C. [Figure 6]Figures 6A-6C illustrate the ability of representative B7H3-targeted, restricted CD3-engaging constructs to mediate target antigen-specific agonism of costimulatory receptors (41BB or OX40). In Figure 6A, a representative B7H3-targeted, restricted CD3-engaging construct, cx3091, containing the 41BB-binding domain, was compared to a similar construct, cx3095, lacking the 41BB-binding domain, in Jurkat 41BB NFκB luciferase reporter cells to assess 41BB agonism in the presence of the B7H3-positive cell line, A375, and the cognate A375 B7H3 knockout cell line (A375ΔB7H3). Notably, cx3091 was capable of potent and robust 41BB agonism. cx3095 induced trace amounts of NFκB signaling, likely mediated by TCR / CD3 signaling independent of 41BB signaling. In Figures 6B and 6C, representative B7H3-targeted, restricted CD3-engaged constructs containing the 41BB-binding domain (cx3091), the OX40-binding domain (cx3723), or no CRBD (cx3834) were evaluated against the OX40-reporter Jurkat cell line (Figure 6B) or the 41BB-reporter Jurkat cell line (Figure 6C) co-cultured with A375 target antigen-expressing cells. [Figure 7] Figures 7A-7B illustrate the ability of representative B7H3-targeted, restricted CD3-engaging constructs cx3095 and cx3091, as well as other DART-Fc formats targeting B7H3 and CD3, to mediate target antigen-specific T cell activation. Jurkat CD3 NFAT-GFP reporter cells were used to assess T cell activation in the presence of the B7H3-positive cell line A375 (Figure 7A) and the A375 B7H3 knockout cell line A375ΔB7H3 (Figure 7B). Notably, the 41BB-containing construct cx3091 exhibited enhanced NFAT signaling, likely due to increased Jurkat cell viability and / or reduced exhaustion following TCR / CD3 signaling. [Figure 8]Figures 8A-8D illustrate the efficacy of T cell-mediated cytotoxicity driven by representative B7H3-targeted constrained CD3-engaging constructs, cx3091 containing the 41BB-binding domain and cx3095 lacking the 41BB-binding domain, compared with alternative DART-Fc formats targeting B7H3 and CD3. The titration range of 50 nM to 80 pM of the CD3-engaging constructs in the B7H3-positive A375 cell line is shown in Figures 8A and 8C, and in the B7H3 knockout A549 cell line is shown in Figures 8B and 8D. Figures 8A and 8B show T cell-mediated cytotoxicity as determined by total cell apoptosis (cells positive for caspase 3 / 7 substrates) due to caspase activation in labeled target cells in each cell line. Figures 8C and 8D illustrate target cell depletion resulting from T cell-mediated cytotoxicity. [Figure 9] Figures 9A-9E are a series of graphs illustrating the efficacy of target-dependent T cell-mediated cytotoxicity induced against A375 cells by B7H3-targeted, restricted CD3-engaged constructs incorporating various costimulatory receptor binding domains. Exemplary B7H3-targeted, restricted CD3-engaged constructs were used here that incorporated either no costimulatory receptor binding domain (cx3834), an OX40 binding domain (Figure 9A cx3723), a 41BB binding domain (Figure 9B cx3091), a GITR binding domain (Figure 9C cx3724), an ICOS binding domain (Figure 9D cx3721), or a CD28 binding domain (Figure 9E cx3726). Cytotoxicity is indicated by the total overlap area, representing target cell apoptosis, measured by quantifying double-positive fluorescence of labeled target cells (A375) and caspase 3 / 7 substrates. [Figure 10]Figures 10A and 10B illustrate the kinetics of T cell cytotoxicity mediated by representative 5T4-targeting constrained CD3-engaging constructs cx3497 (containing the 41BB-binding domain) and cx3547 (not containing the 41BB-binding domain) against the 5T4-positive cell line Ovcar-5 (Figure 10A) and the 5T4-negative cell line CCRF-CEM (Figure 10B). The total overlap area represents cleaved caspase 3 / 7 substrate in double-positive fluorescently labeled target cells. The initial cytotoxicity observed in the antigen-negative cell line was likely mediated by an MHC mismatch between the target cell and the T cell. Notably, only cx3497, which contains the 41BB-binding domain, was able to induce cytotoxicity, with kinetics consistent with that of 41BB upregulation after TCR signaling. [Figure 11] Figures 11A-11D illustrate the efficacy of T cell activation driven by representative B7H3-targeted constrained CD3-engaging constructs, cx3091 containing the 41BB-binding domain and cx3095 lacking the 41BB-binding domain, compared with other DART-Fc formats targeting B7H3 and CD3. The titration range of 50 nM to 80 pM of the CD3-engaging constructs in the B7H3-positive A375 cell line is shown in Figures 11A and 11C, and in the B7H3 knockout A549 cell line is shown in Figures 11B and 11D. Figures 11A and 11B show CD4 T cell activation, and Figures 11C and 11D show CD8 T cell activation. T cell activation, evidenced by CD71 upregulation, was measured by flow cytometry. [Figure 12] Figures 12A and 12B show a comparison of IFNγ production by T cells treated with representative B7H3-targeted restricted CD3-engaging constructs, cx3091 containing the 41BB-binding domain and cx3095 not containing the 41BB-binding domain, in the presence of B7H3-positive A375 target cells (Figure 12A) and B7H3-negative CCRF-CEM target cells (Figure 12B). Cytokine production was monitored by flourospot assay. [Figure 13]Figure 13 shows a comparison of IFNγ production by T cells treated with representative 5T4-targeted restricted CD3-engaging constructs, cx3499 and cx3497, which contain the 41BB-binding domain, and cx3546 and cx3547, which do not contain the 41BB-binding domain, in the presence of 5T4-positive Ovcar-5 cells and 5T4-negative CCRF-CEM cells. Cytokine production was monitored by ELISA. cx3546 and cx3499 contain the same 5T4-targeting sdAb, and 3547 and 3497 contain the same 5T4-targeting sdAb. Thus, the only difference between these sets of constructs is the addition of the 41BB-binding domain. 5T4-targeted restricted CD3-engaging constructs incorporating the 41BB-binding domain exhibit enhanced ability to induce IFNγ production from T cells in an antigen-dependent manner. [Figure 14] Figures 14A and 14B demonstrate the ability of constrained CD3-engaging constructs incorporating costimulatory receptor binding domains to mediate specific activation of the respective costimulatory signaling pathways. Here, Jurkat OX40 (Figure 14A) or Jurkat 41BB (Figure 14B) NFκB luciferase reporter cells were used. Exemplary constrained CD3-engaging constructs targeting B7H3 were used here that either lacked a costimulatory receptor binding domain (cx3834), or incorporated an OX40-binding domain (cx3723), or an 41BB-binding domain (cx3091). cx3723 and cx3091, incorporating either an OX40-binding domain or an 41BB-binding domain, were found to induce specific activation of the targeted costimulatory receptors, respectively. Some degree of induced NFκB signaling was observed with all constructs, including those that did not contain costimulatory receptor binding domains, suggesting that target-dependent CD3 signaling mediates NFκB signaling in this system. The B7H3-positive cell line, A375, was used. [Figure 15]Figures 15A-15C illustrate the binding of 18H10 and its humanized variant hz18H10v7 to FreeStyle 293 cells expressing human PD-1 (Figure 15A), cynomolgus monkey PD-1 (Figure 15B), or mouse PD-1 (Figure 15C). Binding to untransfected (293) cells was also assessed and shown. [Figure 16] 16 is a graph illustrating the binding of 18H10 and its humanized variant hz18H10v7 to human activated T cells. Binding of test articles to activated T cells was quantified by flow cytometry. [Figure 17] FIG. 17 is a graph illustrating the ability of 18H10 and its humanized variant hz18H10v7 to block PD1 / PDL1-mediated inhibition of T cell receptor (TCR) signaling in a Jurkat reporter luciferase assay system. [Figure 18] 18A-18B illustrate a series of schematic diagrams depicting various TAA-targeted, restricted CD3-engaging constructs containing an anti-PD1-binding domain. The basic component of the TAA-targeted, restricted CD3-engaging constructs of the present disclosure has restricted CD3 binding. The antigen-binding domain is positioned at the amino-terminus and / or carboxy-terminus. An Fc region, such as a heterodimeric Fc region, is positioned N-terminal to the CD3-binding region. This positioning of the Fc in close proximity to the CD3-binding region interferes with CD3 binding. In an exemplary embodiment, the anti-PD1-binding domain is positioned N-terminal to the heterodimeric Fc, and a 41BB costimulatory binding domain may be contained C-terminal to the anti-CD3-binding domain. [Figure 19]Figure 19A is a schematic diagram of various constrained CD3 constructs targeting B7H3, composed of two polypeptides, chain 1 and chain 2. Chain 1 contains either a heterodimeric Fc "hole" linked via a non-cleavable linker to an anti-CD3 VL domain modified at G100C (top); a B7H3-targeting sdAb linked to a heterodimeric Fc "hole" linked via a non-cleavable linker to an anti-CD3 VL domain (middle); or a B7H3-targeting sdAb linked to a heterodimeric Fc "hole" linked via a non-cleavable linker to an anti-CD3 VL domain modified at G100C (bottom). Chain 2 contains either a B7H3-targeting sdAb linked to a complementary heterodimeric Fc "knob" linked via a linker to an anti-CD3 VH domain modified at G44C, which is linked to a second B7H3 sdAb (top); a B7H3-targeting sdAb linked to a complementary heterodimeric Fc "knob" linked via a linker to an anti-CD3 VH domain (middle); or a B7H3-targeting sdAb linked to a complementary heterodimeric Fc "knob" linked via a linker to an anti-CD3 VH domain modified by G44C (bottom). When coexpressed, the CD3-binding domains assemble properly via VL:VH association at the hole and knob, respectively. Where indicated, the VH:VL interaction is stabilized by an engineered disulfide bond between modified residues G44C in the VH domain and G100C in the VL domain. Figure 19B is a schematic diagram of various B7H3-targeting constrained CD3 constructs composed of two polypeptides, chain 1 and chain 2. Chain 1 contains a heterodimeric Fc "hole" linked via a non-cleavable linker to an anti-CD3 VL domain modified at G100C, which is linked to a costimulatory receptor-targeting sdAb.Chain 2 contains either a B7H3-targeting sdAb linked to a complementary heterodimeric Fc "knob" linked via a linker to an anti-CD3 VH domain modified at G44C, which is linked to a second B7H3-targeting sdAb (top); a heterodimeric Fc "knob" linked via a linker to an anti-CD3 VH domain modified at G44C, which is linked to a B7H3-targeting sdAb (middle); or a B7H3-targeting sdAb linked to a complementary heterodimeric Fc "knob" linked via a linker to an anti-CD3 VH domain modified by G44C (bottom). When coexpressed, the CD3-binding domains are properly assembled via VL:VH association at the hole and knob, respectively. The VH:VL interaction is stabilized by an engineered disulfide bond between modified residues G44C in the VH domain and G100C in the VL domain. The resulting constructs engage B7H3 either bivalently (top) or monovalently (middle and bottom). All constructs herein contain a costimulatory receptor-targeting sdAb. Figure 19C shows a schematic representation of various constrained B7H3-targeting CD3 constructs, composed of three polypeptides, chain 1, chain 2, and chain 3, in which the B7H3-targeting domain is FAB. Chain 1 contains a BH73-targeting VH, IgG constant heavy chain 1 (CH1), linked via a hinge to the first member of a heterodimeric Fc (Fc-Het-1), which is linked via a linker to an anti-CD3 VL domain lacking (top) or containing (bottom) the G100C modification. Chain 2 contains a BH73-targeting VH, IgG constant heavy chain 1 (CH1), linked via a hinge to the second member of the heterodimeric Fc (Fc-Het-2), which is linked via the linker to an anti-CD3 VH domain lacking (top) or containing (bottom) the G44C modification. Chain 3 contains a complementary B7H3-targeting VL domain linked to a human Ig constant light chain (CL) region. When coexpressed, the CD3-binding domains are properly assembled via VL:VH association in the complementary heterodimeric Fc regions.Where indicated, the VH:VL interaction is stabilized by an engineered disulfide bond between modified residues, G44C in the VH domain and G100C in the VL domain. [Figure 20] Figure 20 is a schematic diagram of three constrained CD3 constructs targeting DLL3, composed of two polypeptides, chain 1 and chain 2. Chain 1 contains a heterodimeric Fc "hole" linked via a non-cleavable linker to an anti-CD3 VL domain modified at G100C, which is linked to a costimulatory receptor-targeting sdAb. Chain 2 contains either a DLL3-targeting sdAb linked to a complementary heterodimeric Fc "knob" linked via the linker to an anti-CD3 VH domain modified at G44C linked to a second DLL3-targeting sdAb (top); a heterodimeric Fc "knob" linked via the linker to an anti-CD3 VH domain modified at G44C linked to a DLL3-targeting sdAb (middle); or a DLL3-targeting sdAb linked to a complementary heterodimeric Fc "knob" linked via the linker to an anti-CD3 VH domain modified by G44C (bottom). [Figure 21]Figures 21A and 21B illustrate cell binding of representative constrained CD3-engaging constructs targeting B7H3. cx5187 and cx5823 each contain two B7H3-binding domains, and conjugates cx5873 and cx5965 each contain one B7H3-binding domain. Figure 21A shows binding to B7H3-positive A375 cells. Figure 21B shows lack of binding to B7H3-negative CCRF-CEM cells and isolated T cells. Figures 21C and 21D illustrate the ability of representative constrained CD3-engaging constructs targeting B7H3 to agonize CD3 in a target-dependent manner. Figure 21C shows that engagement of B7H3-positive A375 cells with a molecule that is bivalent, two epitopes for B7H3 (cx5187) induced stronger CD3 signaling than constructs that are monovalent for B7H3 (cx5873 and cx5965). Figure 21D shows the lack of T cell activation in the presence of B7H3-negative CCRF-CEM cells. The Jurkat CD3 NFAT-GFP reporter cell line was used to assess CD3 agonism. [Figure 22] Figures 22A and 22B illustrate the ability of representative restricted CD3-engaged constructs targeting B7H3 to induce T cell-mediated cytotoxicity in a target-dependent manner. Figure 22A shows that targeting B7H3-positive A375 cells with a bivalent, biepitopic construct for B7H3 (cx5187) induced more potent T cell-mediated cytotoxicity than constructs monovalent for B7H3 (cx5873 and cx5965). Figure 22B illustrates the lack of T cell-mediated cytotoxicity against B7H3-negative CCRF-CEM cells. [Figure 23]Figures 23A-D illustrate that a representative restricted CD3-engaging construct targeting B7H3 can activate T cells in the presence of B7H3-positive A375 cells but not in the presence of B7H3-negative CCRF-CEM cells. Figures 23A and 23B show that targeting B7H3-positive A375 cells with a bivalent, biepitope construct (cx5187) for B7H3 induced stronger CD25 expression in CD4+ or CD8+ T cells than constructs monovalent for B7H3 (cx5873 and cx5965). Figures 23C and 23D show the lack of CD25 expression in CD4+ or CD8+ T cells, respectively, in the presence of B7H3-negative CCRF-CEM cells. [Figure 24] Figures 24A and 24B demonstrate that monovalent and bivalent restricted CD3-engaging constructs targeting DLL3 bound to the DLL3-expressing cell line SHP-77 (Figure 24A) but not to isolated T cells (Figure 24B). Binding was assessed by flow cytometry. Figure 24C illustrates the ability of representative restricted CD3-engaging constructs targeting DLL3 to agonize CD3 signaling in the presence of DLL3-positive SHP-77 cells. Engagement of DLL3-positive cells with a bivalent, biepitopic construct against DLL3 (cx5352) induced more potent T cell activation than monovalent constructs against DLL3 (cx5800 and cx5801). The Jurkat CD3 NFAT luciferase reporter cell line was used to assess CD3 signaling. [Figure 25] Figures 25A and 25B illustrate the efficacy of T cell-mediated cytotoxicity driven by exemplary 5T4-targeted constrained CD3-engaging constructs, cx5185, which contains the 41BB-binding domain, and cx5951, which does not contain the 41BB-binding domain. The titration range of 200 nM to 3.1 pM of the CD3-engaging constructs in the 5T4-positive Ovcar5 cell line is shown in Figure 25A, and in the 5T4-negative CCRF-CEM cell line is shown in Figure 25B. [Figure 26]Figures 26A-F illustrate T cell activation, as assessed by CD25 expression, in CD4 cells (Figures 26A-C) and CD8 cells (Figures 26D-F) after incubation with representative 5T4-targeted constrained CD3-engaging constructs, cx5185 containing the 41BB-binding domain and cx5951 not containing the 41BB-binding domain, in the presence of various 5T4-positive cells, A375 (Figures 26A and 26D), Ovcar-5 (Figures 26B and 26E), and SHP-77 (Figures 26C and 26F). T cell activation was assessed by flow cytometry monitoring cell surface expression of CD25. [Figure 27] Figures 27A-C show a comparison of IFNγ production by T cells treated with titrations of representative 5T4-targeted restricted CD3-engaging constructs, cx5185 containing the 41BB binding domain and cx5951 not containing the 41BB binding domain, in the presence of 5T4-positive cell lines, A375 (Figure 27A), SHP-77 (Figure 27B), and Ovcar5 (Figure 27C). [Figure 28] Figures 28A-F illustrate T cell proliferation in CD4 T cells (Figures 28A-C) and CD8 T cells (Figures 28D-F) after incubation with representative 5T4-targeted constrained CD3-engaging constructs, cx5185 containing the 41BB-binding domain and cx5951 without the 41BB-binding domain, in the presence of various 5T4-positive cell lines, A375 (Figures 28A and 28D), Ovcar-5 (Figures 28B and 28E), and SHP-77 (Figures 28C and 28F). T cell proliferation was monitored via dilution of CellTrace™ dye. Results are shown as the percentage of proliferating cells monitored by dilution of CellTrace™ Violet dye. [Figure 29]Figures 29A-F illustrate the assessment of mitochondrial function in CD4 T cells (Figures 29A-C) and CD8 T cells (Figures 29D-F) after incubation with representative 5T4-targeted constrained CD3-engaging constructs, cx5185 containing the 41BB-binding domain and cx5951 without the 41BB-binding domain, in the presence of various 5T4-positive cells, A375 (Figures 29A and 29D), Ovcar-5 (Figures 29B and 29E), and SHP-77 (Figures 29C and 29F). Mitochondrial function was assessed by flow cytometry using MitoTracker Green, a mitochondria-selective fluorescent probe that accumulates in active mitochondria. [Figure 30] Figure 30 shows that the 5T4-targeted constrained CD3-engaging construct cx5185, which contains the 41BB binding domain, can mediate 41BB signaling, whereas the identical construct cx5951, which lacks the 41BB binding domain, cannot. 41BB signaling was monitored using Jurkat 41BB NFkB luciferase reporter cells and plate-bound recombinant 5T4 as a source of antigen. [Figure 31]Figure 31A illustrates various representative B7H3-targeted restricted CD3 engagers containing the 4-1BB binding domain as a CRBR. cx5841 and cx5187 have a B7H3-targeting sdAb positioned at the N-terminus and C-terminus of the Fc knob on one chain of the heterodimer, and a 4-1BB-targeting sdAb positioned at the C-terminus of the Fc hole on the other chain of the heterodimer, but with the VH and VL of the CD3-binding Fv positioned opposite each other. Figures 31B-E illustrate the results of T cell reporter assays for the exemplary constructs described in Figure 31A. Figures 31B and 31C illustrate the mean fluorescence intensity (MFI) of the GFP reporter when the B7H3-positive cell line A375 or the B7H3-negative cell line CCRF-CEM, respectively, were cocultured with Jurkat CD3 NFAT-GFP reporter cells. Figures 31D and 31E illustrate the relative light units (RLU) of the luciferase reporter when the B7H3-positive cell line A375 or the B7H3-negative cell line CCRF-CEM, respectively, were co-cultured with Jurkat CD3 NFAT luciferase reporter cells. [Figure 32] Figure 32 illustrates representative DLL3-targeted restricted CD3 engagers, cx5499 (Figure 32A) which does not contain a 41BB-binding domain, and cx5352 (Figure 32B) which does contain a 41BB-binding domain. cx5499 and cx5352 have a DLL3-targeting sdAb positioned at the N-terminus of one chain of the heterodimer and at the C-terminus of the same chain of the heterodimer. cx5352 and cx5499 are identical to each other, except that cx5352 has a 41BB-targeting sdAb positioned at the C-terminus of the opposite chain of the heterodimer, whereas cx5499 does not. These representative DLL3-targeted restricted CD3 engagers exhibit bivalent binding to DLL3, and in constructs containing the 41BB-binding domain, exhibit monovalent binding to 41BB. [Figure 33]Figure 33 shows that the DLL3-targeted, restricted CD3-engaging construct cx5352, which contains the 41BB-binding domain, can mediate 41BB signaling, whereas the identical construct cx5499, which lacks the 41BB-binding domain, cannot. DLL3-dependent 41BB signaling was monitored using Jurkat 41BB NFkB luciferase reporter cells cocultured with either parental HEK-293 cells or HEK-293 cells transiently expressing a truncated version of DLL3. Activity is shown as relative luciferase units (RLU). [Figure 34] Figures 34A-B illustrate the efficacy of T cell-mediated cytotoxicity driven by an exemplary DLL3-targeted, restricted CD3-engaging construct, cx5352, containing the 41BB-binding domain against the DLL3-positive cell line, SHP-77 (Figure 34A) or the DLL3-negative cell line, HEK-293FS (Figure 34B). Three different T cell donors were used as sources of effector cells in this assay. [Figure 35] Figures 35A-C show the differential efficacy of T cell-mediated cytotoxicity driven by exemplary DLL3-targeted restricted CD3-engaging constructs, cx5352 containing the 41BB-binding domain and cx5499 lacking the 41BB-binding domain. Various effector (T cells) to target cell (SHP-77) ratios (effector:target cell ratios) of 10:1 (Figure 35A), 5:1 (Figure 35B), or 1.25:1 (Figure 35C) were compared. Human PBMCs were used as the source of T cells. [Figure 36] Figure 36 shows a comparison of IFNγ production by T cells treated with titrations of representative DLL3-targeted restricted CD3-engaging constructs, cx5352 containing the 41BB binding domain and cx5499 not containing the 41BB binding domain, in the presence of the DLL3-positive cell line SHP-77. DETAILED DESCRIPTION OF THE INVENTION

[0064] Detailed Description The present disclosure provides a fusion protein that engages restricted T cells in the form of a multispecific polypeptide construct that binds at least to CD3, a second antigen, and a costimulatory receptor. Also provided in the present disclosure is a fusion protein that engages T cells in the form of a multispecific polypeptide construct that binds at least to CD3, a second antigen, and an inhibitory receptor. Also provided in the present disclosure is a fusion protein that engages T cells in the form of a multispecific polypeptide construct that binds at least to CD3, a second antigen, a costimulatory receptor, and an inhibitory receptor.

[0065] In some embodiments, the multispecific polypeptide constructs provided herein comprise at least a first component comprising an immunoglobulin Fc region, a second component comprising one or more copies of at least one binding domain that binds to CD3 (referred to herein as an anti-CD3 binding domain or CD3 binding region, terms used interchangeably herein), and a linker, such as a polypeptide linker, joining the first and second components. In the provided multispecific polypeptide constructs, one or both of the first and second components contain at least one antigen-binding domain that, when engaged by binding to an antigen, enables the constrained CD3-binding region to substantially bind to CD3. In some embodiments, the antigen is a tumor-associated antigen (TAA). Furthermore, in some aspects, one or both of the first and second components of the multispecific polypeptide construct also comprise at least one costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor. In other aspects, one or both of the first and second components of the multispecific polypeptide construct also comprise at least one inhibitory receptor binding region (IRBR) that binds to an inhibitory receptor. In a specific aspect of the present disclosure provided, one of the first and second components of the multispecific polypeptide construct contains a CRBR, and the other of the first and second components contains an IRBR. In some embodiments, the linker is a non-cleavable linker. In some embodiments, the linker is a cleavable linker.

[0066] In some embodiments, the provided multispecific polypeptide constructs comprise an arrangement in which a first component containing an Fc region is N-terminal to a second component containing a CD3-binding region. In such embodiments, the first and second components are joined via a linker that is C-terminal to the end of the Fc region. In some embodiments, the antigen-binding domain and / or CRBR is located in the amino-terminal (N-terminal) region of the multispecific polypeptide construct. In some embodiments, the antigen-binding domain and / or CRBR is located in the carboxy-terminal (C-terminal) region of the multispecific polypeptide construct. In some embodiments, the antigen-binding domain and / or CRBR is located in both the N-terminal and C-terminal regions of the multispecific polypeptide construct. Various arrangements of the multispecific polypeptide constructs provided herein are shown in Figure 1.

[0067] In some embodiments, the provided multispecific polypeptide constructs comprise an arrangement in which a first component containing an Fc region is N-terminal to a second component containing a CD3-binding region. In such embodiments, the first and second components are joined via a linker that is C-terminal to the end of the Fc region. In some embodiments, the antigen-binding domain and / or IRBR is located in the amino-terminal (N-terminal) region of the multispecific polypeptide construct. In some embodiments, the antigen-binding domain and / or IRBR is located in the carboxy-terminal (C-terminal) region of the multispecific polypeptide construct. In some embodiments, the antigen-binding domain and / or IRBR is located in both the N-terminal and C-terminal regions of the multispecific polypeptide construct. Various arrangements of the multispecific polypeptide constructs provided herein are shown in Figure 18A.

[0068] In some embodiments, the provided multispecific polypeptide constructs comprise an arrangement in which a first component containing an Fc region is N-terminal to a second component containing a CD3-binding region. In such embodiments, the first and second components are joined via a linker that is C-terminal to the end of the Fc region. In some embodiments, the antigen-binding domain and / or CRBR is located in the amino-terminal (N-terminal) region of the multispecific polypeptide construct, and the antigen-binding domain and / or IRBR is located in the carboxy-terminal (C-terminal) region of the multispecific polypeptide construct. In some embodiments, the antigen-binding domain and / or IRBR is located in the amino-terminal (N-terminal) region of the multispecific polypeptide construct, and the antigen-binding domain and / or CRBR is located in the carboxy-terminal (C-terminal) region of the multispecific polypeptide construct. Various arrangements of the multispecific polypeptide constructs provided herein are shown in Figure 18B.

[0069] Because the provided multispecific polypeptide constructs substantially bind to CD3 only after antigen binding via the antigen-binding domain, such constructs exhibit restricted T cell-engaging activity. This is exemplified in the Examples and Figures provided herein, which demonstrate that restricted CD3-engaging proteins can efficiently bind to TAA-positive cells with little or no binding to T cells. This unique property allows restricted CD3-engaging proteins to distribute to sites where TAAs are present without binding to peripheral T cells. This format differs from other CD3-engaging multispecific constructs in which constitutive CD3 binding is inhibited or eliminated, and provides a significant advantage by avoiding peripheral T cell binding and allowing preferential distribution to sites where the antigen recognized by the antigen-binding domain is present. For example, as shown in the Examples, the restricted CD3-engaging format enables potency similar to that of the DART-Fc format (e.g., published PCT application WO2017 / 030926), but with significantly attenuated peripheral T cell binding. Furthermore, while other CD3-engaging constructs mediate antigen-dependent T cell activation, the multispecific polypeptide constructs provided herein mediate both antigen-dependent T cell binding and activation.

[0070] The restricted T cell-engaging activity of the provided multispecific polypeptide constructs is, in some aspects, due to the positioning of the Fc region N-terminal to the CD3-binding region. In some embodiments, such positioning reduces, attenuates, interferes with, and / or prevents CD3 binding by the CD3-binding region. The multispecific polypeptide constructs provided herein exhibit reduced or eliminated CD3 binding and T cell activation in the absence of antigen binding by the antigen-binding domain. In some embodiments, the ability of the CD3-binding region to bind CD3 is greatly enhanced in the presence of an antigen-binding event mediated by the antigen-binding domain of the multispecific polypeptide construct. In some embodiments, the ability to activate T cells is greatly enhanced in the presence of an antigen-binding event mediated by the antigen-binding domain of the multispecific polypeptide construct. Engagement of its cognate antigen by the antigen-binding domain within the multispecific polypeptide construct results in subsequent T cell engagement and mediates antigen-dependent T cell activation, such as cytotoxicity, cytokine release, degranulation, and proliferation. In some embodiments, the provided multispecific polypeptide constructs can be used to increase an immune response, for example, to enhance T cell activity, including cytolytic (or cytotoxic) T cell activity. Modulation of the immune response can, in some aspects, treat a disease or condition in a subject.

[0071] In some embodiments, the multispecific polypeptide constructs of the present disclosure exhibit specificity for CD3, one or more other antigens, and one or more costimulatory receptors. In some embodiments, the multispecific polypeptide constructs of the present disclosure exhibit specificity for CD3, one or more other antigens, and one or more inhibitory receptors. In some embodiments, the multispecific polypeptide constructs of the present disclosure exhibit specificity for CD3, one or more other antigens, one or more costimulatory receptors, and one or more inhibitory receptors. In some embodiments, the multispecific polypeptide constructs may contain one or more antigen-binding domains, e.g., one, two, or three antigen-binding domains, e.g., two or three antigen-binding domains, capable of binding to one or more TAAs. In some embodiments, the multispecific polypeptide constructs may contain one or more costimulatory receptor-binding regions (CRBRs), e.g., one, two, or three CRBRs, capable of binding to one or more costimulatory receptors, e.g., two or three costimulatory receptors. In some embodiments, the multispecific polypeptide construct may contain one or more inhibitory receptor binding regions (IRBRs), e.g., one, two, or three IRBRs, that can bind to one or more inhibitory receptors, e.g., two or three inhibitory receptors. Exemplary construct formats are shown in Figures 1-3 and 18A-B.

[0072] In some embodiments, one or more CRBRs bind to a costimulatory receptor expressed on the surface of T cells. Non-limiting examples of costimulatory receptors that can be targeted by CRBRs include, for example, 41BB, OX40, CD27, GITR, CD28, ICOS, CD40L, BAFFR, BCMA, TACI, or NKG2d. In some embodiments, one or more CRBRs are or include a binding domain (e.g., sdAb, scFv, or Fab) that binds to a costimulatory receptor. As shown herein, the presence of CRBRs in the provided multispecific polypeptide constructs increases or enhances T cell activity upon simultaneous engagement of CD3 by the CD3-binding region.

[0073] In some cases, one or more CRBRs bind to a costimulatory receptor that is not constitutively expressed in T cells. In some embodiments, one or more CRBRs bind to a costimulatory receptor that is upregulated, induced, or expressed upon T cell activation. In some embodiments, CRBRs bind to a costimulatory receptor that is a member of the tumor necrosis factor (TNF) receptor family, such as a TNFRSF member, that is upregulated, induced, or expressed upon T cell activation. In some embodiments, CRBRs bind to a costimulatory receptor that is a member of the B7 family, such as a B7 family member, that is upregulated, induced, or expressed upon T cell activation. Such costimulatory receptors whose expression is upregulated or whose surface expression is induced in activated T cells include, for example, 41BB, OX40, and GITR. In some aspects, the CD3 binding region of the provided multispecific polypeptide construct induces or enhances T cell activation when the CD3 binding region binds to CD3, which in turn leads to increased or upregulated expression of such costimulatory receptors. In such instances, binding of CRBR to a costimulatory receptor, e.g., 41BB, OX40, or GITR, is increased or enabled only after engagement of CD3 by the CD3-binding region of the multispecific polypeptide construct, thereby controlling the costimulatory activity of the provided multispecific construct on target cells. In some cases, this can solve the problem of other T cell engagers, as it avoids the antigen "sink" effect caused by binding to CD3 and / or costimulatory receptors expressed on peripheral T cells, such as resting T cells or unwanted cells, which can accelerate systemic clearance. In some embodiments, the provided multispecific polypeptide constructs exhibit increased half-life and / or improved pharmacokinetic or pharmacodynamic properties, e.g., better tumor exposure, compared to other T cell engagers that can bind to peripheral cells, such as bispecific T cell engagers, e.g., DART-Fc.

[0074] In some embodiments, one or more IRBRs bind to inhibitory receptors expressed on the surface of T cells. In some aspects, the inhibitory receptors are those whose expression is upregulated or increased in activated T cells. Non-limiting examples of inhibitory receptors that can be targeted by IRBRs include, for example, PD-1, CTLA-4, TIGIT, VISTA, TIM3, or LAG3. In some embodiments, one or more IRBRs are or include a binding domain (e.g., sdAb, scFv, or Fab) that binds to an inhibitory receptor.

[0075] In some embodiments, one or more IRBRs bind to an inhibitory receptor and block or inhibit its interaction with its ligand, thereby blocking the inhibitory activity of T cells. In some embodiments, the IRBR is a PD-1-binding polypeptide that binds to PD-1. In some cases, the provided PD-1-binding polypeptide directly blocks or inhibits the interaction between PD-L1 / L2 and PD-1. In some aspects, inclusion of an IRBR in a multispecific polypeptide construct inhibits or reduces the interaction between an inhibitory receptor and its ligand, e.g., between PD-L1 and / or PD-L2 and PD-1, thereby modulating the immune response. Transmission of an inhibitory signal results in down-modulation of the immune cell response (and consequently down-modulation of the overall immune response), while blocking the inhibitory signal in an immune cell results in up-modulation of the immune cell response (and consequently up-modulation of the immune response). In some cases, modulation by enhancing the immune response can be used to treat certain diseases or conditions in which the immune response is suppressed, such as cancer. Generally, the ability to block inhibitory interactions occurs in the immune synapse and / or tumor microenvironment when the CD3-binding region binds to CD3 on a T cell and / or the antigen-binding domain binds to a TAA. In specific embodiments, the provided multispecific polypeptide constructs reduce, suppress, or abrogate inhibitory signals mediated by inhibitory receptors in cells such as T cells.

[0076] In some embodiments, one or more antigen-binding domains bind to antigens of tumor cells or cells in the tumor microenvironment. In some aspects, the provided multispecific polypeptide constructs can be used to increase immune responses, such as T cell activity, e.g., cytotoxic activity, against tumors or cancer. In some embodiments, the provided multispecific polypeptide constructs can be used to treat tumors or cancer in a subject.

[0077] In some embodiments, one or more antigen-binding domains bind to the same antigen. In some embodiments, a multispecific polypeptide construct comprises multiple antigen-binding domains that bind to different epitopes of the same antigen. In some embodiments, a multispecific polypeptide construct comprises multiple antigen-binding domains that bind to one or more different antigens. In some embodiments, a multispecific polypeptide construct comprises multiple antigen-binding domains that bind to different epitopes of the same antigen, and also comprises additional antigen-binding domains that bind to one or more different antigens. In some examples, the provided multispecific polypeptide construct provides multivalent engagement of one or more TAAs, e.g., through at least a first antigen-binding domain and a second antigen-binding domain. For example, in some embodiments, the polypeptide construct comprises at least a first antigen-binding single domain antibody (sdAb) and a second antigen-binding sdAb. In some aspects, the first and second antigen-binding domains bind to the same antigen.

[0078] Furthermore, in some aspects, because the CD3-binding region of the multispecific polypeptide constructs of the present disclosure is constrained or otherwise blocked and / or inhibited by the presence of the Fc region, these constructs ensure that CD3-mediated T cell binding does not occur or is reduced in peripheral blood. Thus, the multispecific polypeptide constructs of the present disclosure offer numerous advantages. In some aspects, these constructs limit the sink effect caused by binding to all T cells. In some aspects, these constructs reduce systemic toxicity.

[0079] In some embodiments, the multispecific polypeptide constructs provided herein allow for controlled biodistribution to desired sites in a subject, such as, for example, sites of tumor-associated antigen (TAA) expression, including, for example, the tumor and surrounding tumor microenvironment.

[0080] In some embodiments, the multispecific polypeptide constructs provided herein exist in two states with respect to their ability to bind CD3 and subsequently activate T cells: (1) an "inactive" state occurs when the binding of any or all of the antigen binding domains is absent, such that CD3 binding is restricted and T cell interaction is eliminated or reduced; and (2) an "active" state occurs when any or all of the antigen binding domains are bound to antigen, such that the CD3 binding region can bind CD3 and T cell interaction is possible.

[0081] In some embodiments, the Fc region is linked to the CD3 binding domain via a linker. In some embodiments, the Fc region is linked to the CD3 binding region via a non-cleavable linker. In some embodiments, the Fc region is linked to the CD3 binding region via a cleavable or otherwise labile linker.

[0082] In some aspects, the multispecific polypeptide constructs of the present disclosure enable therapeutic efficacy in the absence of cleavage, e.g., in the absence of proteolysis. In some embodiments, the linker is a non-cleavable linker.

[0083] In some embodiments, the Fc region and the CD3-binding region are linked by a cleavable linker, such as a linker that can be specifically cleaved in the presence of a protease. In some aspects, enhanced CD3 binding occurs after cleavage of the cleavable linker. In some such aspects, the "active" state can be further amplified through several mechanisms, for example, through cleavage of the linker joining the CD3-binding region and the Fc region. In some embodiments, the cleavable linker is a linker containing a substrate recognition site for a protease. In some embodiments where the Fc region and the CD3-binding region are linked by a cleavable linker, enhanced CD3 binding can occur after cleavage within the linker.

[0084] In some embodiments, the Fc region is a homodimeric Fc region. In some embodiments, the Fc region is a heterodimeric Fc region. In some embodiments, the Fc region is a monomeric Fc region. In some embodiments, the Fc region of the multispecific polypeptide construct can interact with FcγR and mediate innate immune effector functions, such as antibody-dependent cellular cytotoxicity (ADCC) and antibody-dependent cellular phagocytosis (ADCP). In some embodiments, the Fc region of the multispecific polypeptide construct can interact with complement proteins, i.e., C1q, and mediate complement-dependent cytotoxicity. Thus, in some aspects, the multispecific polypeptide constructs of the present disclosure enable multiple immune effector mechanisms, including innate immune effectors and T cells.

[0085] In some embodiments, the linker is a cleavable linker. In some embodiments, where the Fc region and the CD3-binding region are operably linked by a cleavable linker, cleavage of the linker between the Fc region and the CD3-binding region can separate the multispecific polypeptide construct into a first component and a second component. Depending on the composition of the multispecific polypeptide construct, the first and second components may have different functionalities. In some embodiments, the Fc region is a region that exhibits one or more effector functions, such as ADCC, CDC, or ADCP. In such examples, the multispecific polypeptide constructs of the present disclosure can be used to create a self-amplifying system. For example, the multispecific construct can be used as follows: ADCC mediated by NK cells after TAA targeting and CD16 binding of the Fc region results in the release of granzyme B, which is capable of extracellular proteolysis, and cleavage of the linker between the first and second components of the multispecific polypeptide construct.

[0086] In some embodiments, the multispecific polypeptide construct provides a two-in-one therapeutic group with dual effector functions, in which proteolytic activation of the multispecific polypeptide construct produces two components, each with biological activity. The multispecific polypeptide constructs of the present disclosure can provide Fc-mediated effector functions, such as ADCC (e.g., release of granzyme B by NK cells), ADCP, and / or CDC.

[0087] Constrained CD3-engaging constructs can be used with any TAA-binding domain and are intended to avoid interactions with peripheral T cells and mediate potent TAA-dependent T cell cytotoxicity, thereby enabling better therapeutic exposure in tumors or tumor microenvironments. Incorporation of a protease-cleavable linker between the Fc and CD3-binding domain components allows for full exposure of the CD3-binding domain, thereby amplifying T cell activation capacity. Depending on the specific linker included, the amplification step may be mediated by tumor-associated proteases or by granzymes released after antigen-dependent T cell activation. When a tumor protease-cleavable linker is included, amplification is mediated by the tumor or tumor microenvironment. On the other hand, when a granzyme B-cleavable linker is included, amplification can be self-mediated by T cells after antigen-dependent activation. Furthermore, in cases where an effector-enabled Fc is included in the construct, amplification through ADCC mechanisms can be mediated by granzymes released from NK cells.

[0088] In some embodiments, the cleavable linker is cleaved by a protease produced in the tumor microenvironment and / or a protease produced or secreted by T cells upon T cell activation, such as may be induced by initial binding of the CD3-binding region to CD3 in the tumor microenvironment via binding of the antigen-binding domain to a TAA. In some embodiments, the protease is Granzyme B. In some aspects, the multispecific polypeptide constructs of the present disclosure exploit the ability of proteases and / or Granzyme B in the tumor microenvironment to cleave the linker within the multispecific polypeptide construct below the Fc immunoglobulin polypeptide, thereby generating two therapeutically active proteins, in some cases with different effector cell engagement. In some aspects, upon cleavage of the cleavable linker, the cleaved first portion or component retains Fc effector function and bivalent targeting of a first antigen, e.g., via the first antigen-binding domain, such as a TAA, and the second portion or component retains the ability for T cell engagement, as separation of the CD3-binding region from the Fc region allows for CD3 binding. The cleaved second portion or component also retains the ability to bind the TAA, which in some cases may be bivalently bound, via the second antigen binding domain.

[0089] The multispecific polypeptide constructs of the present disclosure are designed to ensure that the protease that cleaves the cleavable linker does not need to be tumor-specific (e.g., it does not need to be differentially expressed only at the tumor site and / or in the tumor environment). Rather, these multispecific polypeptide constructs only require that the protease be co-located with the TAA. The valency of these constructs will drive biodistribution and retention in the tumor and / or tumor microenvironment.

[0090] In some embodiments, the second portion or component contains a CD3-binding region that is monovalent for CD3, so that T cell activation does not occur unless a TAA is present. In some aspects, when the multivalent polypeptide construct contains a cleavable linker, the cleaved second portion or component enables TAA-dependent T cell-mediated cytotoxicity. In some cases, the cleaved second portion or component ensures that FcRn interaction does not occur. Furthermore, the cleaved second portion or component is sufficiently small in size, e.g., only approximately 50 kDa, which will ensure rapid clearance if, for some reason, the cleaved second portion or component is distributed outside the tumor site and / or is aberrantly cleaved outside the tumor site.

[0091] In some embodiments, the multispecific polypeptide constructs of the present disclosure allow for simultaneous T cell and NK cell-mediated cytotoxicity. In some cases, such activity can occur in a multispecific polypeptide construct containing a first antigen-binding domain, e.g., a first anti-TAA antigen-binding domain, and a second antigen-binding domain, e.g., a second anti-TAA antigen-binding domain, that can target different and / or non-competing epitopes of a given TAA.

[0092] In some aspects, the multispecific polypeptide constructs of the present disclosure offer numerous advantages over current bispecific therapeutics. The multispecific polypeptide constructs of the present disclosure are smaller than conventional therapeutic antibodies, e.g., 125 kDa versus 150 kDa, which may allow for better penetration into targets, e.g., tumors. First, the overall size of the multispecific polypeptide construct provides a long half-life for the uncleaved construct, and upon cleavage of the construct, the cleaved second portion or component will be small enough to ensure a short half-life. In some aspects, because CD3 binding by the CD3 binding region depends on TAA engagement before CD3 engagement occurs, the multispecific polypeptide constructs of the present disclosure exhibit reduced systemic toxicity or toxicity in areas outside the tumor and / or tumor microenvironment. In some cases, the inclusion of a cleavable linker specific for a protease in the tumor environment reduces CD3 binding by the multispecific construct until proteolytic activation and TAA engagement amplify or enhance CD3 engagement. Furthermore, the additional presence of CRBR in the provided multispecific polypeptide constructs provides antigen-dependent specific costimulatory signaling to further increase or enhance T cell activity at a desired site or location, such as a tumor and / or tumor microenvironment. Similarly, inclusion of IRBR in the provided multispecific polypeptide constructs can also enhance T cell activity by removing or blocking checkpoint inhibitory signals.

[0093] All publications and patent documents cited herein are incorporated by reference as if each such publication or document was specifically and individually indicated to be incorporated herein by reference. Citation of publications and patent documents is not intended as an admission that any is relevant prior art, nor does it constitute an admission as to the contents or date thereof. While the present invention has been described by way of illustration, those skilled in the art will recognize that the invention can be embodied in a variety of embodiments, and that the foregoing description and following examples are intended to illustrate, but not limit, the scope of the claims which follow.

[0094] I. Definition Unless otherwise defined, scientific and technical terms used in connection with this disclosure shall have the meanings commonly understood by those of ordinary skill in the art. The term "a" entity or "an" entity refers to one or more of that entity. For example, a compound refers to one or more compounds. Thus, the terms "(a)," "(an)," "one or more," and "at least one" can be used interchangeably. Furthermore, unless otherwise required by context, the singular shall include the plural and the plural shall include the singular. Generally, the nomenclature utilized in connection with, and techniques of, cell and tissue culture, molecular biology, protein and oligonucleotide or polynucleotide chemistry, and hybridization described herein are those well known and commonly used in the art. Standard techniques are used for recombinant DNA, oligonucleotide synthesis, and tissue culture and transformation (e.g., electroporation, lipofection). Enzymatic reactions and purification techniques are performed according to manufacturer's specifications or as commonly accomplished in the art or as described herein. The above techniques and procedures are generally carried out according to conventional methods well known in the art, as described in the various general and more specific references cited and described herein. See, for example, Sambrook et al. Molecular Cloning: A Laboratory Manual (2nd ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (1989)). The nomenclature utilized in connection with analytical chemistry, synthetic organic chemistry, and medicinal and pharmaceutical chemistry described herein, as well as the laboratory procedures and techniques thereof, are well known and commonly used in the art. Standard techniques are used for chemical synthesis, chemical analysis, pharmaceutical preparation, formulation, and delivery, and treatment of patients.

[0095] As utilized in accordance with the present disclosure, the following terms, unless otherwise indicated, shall be understood to have the following meanings:

[0096] As used herein, the term "antibody" refers to immunoglobulin molecules and antigen-binding portions or fragments of immunoglobulin (Ig) molecules, i.e., molecules that contain an antigen-binding site that specifically binds (immunoreacts with) an antigen. The term antibody encompasses not only intact polyclonal or monoclonal antibodies, but also dAbs, Fabs, Fab's, F(ab')2, Fvs, single-chain (scFvs), or single-domain antibodies (sdAbs). Typically, an "antigen-binding fragment" contains at least one CDR of an immunoglobulin heavy and / or light chain that binds to at least one epitope of an antigen of interest. In this regard, an antigen-binding fragment may contain one, two, three, four, five, or all six CDRs of the variable heavy (VH) and variable light (VL) chain sequences from an antibody that binds to the antigen, generally including six CDRs ("CDR1," "CDR2," and "CDR3" for each of the heavy and light chains) for antibodies containing a VH and a VL, or three CDRs for antibodies containing a single variable domain. Antigen-binding fragments include single-domain antibodies, e.g., those containing only a VH, or those containing only a VL, e.g., a V H H, V NAR , Modified V H Domain or modified V K Contains the domain.

[0097] The basic antibody structural unit is known to comprise a tetramer. Each tetramer is composed of two identical pairs of polypeptide chains, each pair having one "light chain" (approximately 25 kDa) and one "heavy chain" (approximately 50-70 kDa). The amino-terminal portion of each chain contains a variable region of approximately 100-110 amino acids or more primarily responsible for antigen recognition. The carboxy-terminal portion of each chain defines a constant region primarily responsible for effector function. Generally, antibody molecules obtained from humans belong to one of the classes IgG, IgM, IgA, IgE, and IgD, which differ from each other depending on the nature of the heavy chain present in the molecule. Certain classes have subclasses, such as IgG1, IgG2, IgG3, IgG4, and others. Furthermore, in humans, light chains can be kappa or lambda chains.

[0098] The term "monoclonal antibody" (mAb) or "monoclonal antibody composition," as used herein, refers to a population of antibody molecules containing only one molecular species of antibody molecule, consisting of a unique light chain gene product and a unique heavy chain gene product. Specifically, the complementarity-determining regions (CDRs) of a monoclonal antibody are identical in all molecules of the population. MAbs contain an antigen-binding site capable of immunoreacting with a particular epitope of an antigen, characterized by a unique binding affinity.

[0099] As used herein, the term "epitope" includes a specific portion of an antigen targeted by an antibody, antibody fragment, or other binding domain. The term "epitope" includes any protein region to which specific binding is directed. The term "epitope" includes any protein determinant involved in specific binding with a binding molecule such as an antibody or antigen-binding fragment. Epitope determinants generally consist of chemically active surface groups of molecules such as amino acids or sugar side chains, and generally have specific three-dimensional structural characteristics and specific charge characteristics. For example, antibodies can be raised against N-terminal, central, or C-terminal peptides of a polypeptide. Furthermore, antibodies can be raised against linear or discontinuous epitopes of a polypeptide.

[0100] As used herein, the terms "specific binding" and "specifically bind" refer to the ability of a binding molecule, such as an antibody or antigen-binding fragment, to preferentially bind to an antigen in a complex mixture of proteins and / or macromolecules. A binding molecule is said to exhibit "specific binding" or "preferential binding" if it reacts or associates with a particular cell or target antigen more frequently, rapidly, for a longer period of time, and / or with greater affinity than with another cell or target antigen. A binding molecule specifically binds or preferentially binds to a target if it binds to the target with greater affinity, avidity, rapidly, and / or for a longer period of time than other substances. In some aspects, specific binding can refer to non-covalent interactions of the type that occur between an immunoglobulin molecule and an antigen for which the immunoglobulin is specific. It is understood that specific or preferential binding does not necessarily require (although it can include) exclusive binding. Various known methods can be used to quantify or assess binding. The strength or affinity of a binding interaction can be measured by the dissociation constant (K d ) and the smaller K d represents a greater affinity. One such method involves measuring the rates of formation and dissociation of the antigen-binding site / antigen complex, which depend on the concentrations of the complex partners, the affinity of the interaction, and geometric parameters that affect the rates in both directions equally. Thus, the "association rate constant" (K on ) and "dissociation rate constant" (K off ) can be determined by calculation of the concentration and the actual rates of association and dissociation (see Nature 361:186-87 (1993)). off / K on The ratio of α to β allows for the cancellation of all parameters unrelated to affinity, and the dissociation constant K d(See generally Davies et al. (1990) Annual Rev Biochem 59:439-473.) A binding molecule such as an antibody or antigen-binding fragment has a binding constant (K) measured by an assay such as a radioligand binding assay or similar assay known to those skilled in the art. d ) is said to specifically bind when it is ≦1 μM, for example, in some embodiments ≦100 nM, in some embodiments ≦10 nM, and in some embodiments ≦100 pM to about 1 pM.

[0101] The terms "polypeptide" and "protein" are used interchangeably herein and refer to a molecular chain of two or more amino acids linked by peptide bonds. The term includes post-translational modifications of the polypeptide, such as glycosylation, acetylation, phosphorylation, and the like. The term also includes molecules containing one or more amino acid analogs or non-standard or unnatural amino acids, which can be synthesized or recombinantly expressed using known protein engineering techniques. Furthermore, proteins can be derivatized. A protein can be a single polypeptide chain or a multimer (dimer) of at least two polypeptide chains covalently linked to each other, for example, by interchain disulfide bonds. Thus, monomeric, dimeric, and higher order multimeric polypeptides are included within the defined term. Multimeric polypeptides can be homomultimers (of identical polypeptide chains) or heteromultimers (of non-identical polypeptide chains).

[0102] The term "isolated protein" as referred to herein means that the protein of interest (1) is at least partially free from other proteins with which it is naturally found in nature; (2) is essentially free from other proteins from the same source, e.g., from the same species; (3) is expressed by cells from a different species; (4) is separated from at least about 50% of the polynucleotides, lipids, carbohydrates, or other materials with which it is naturally associated; (5) is not associated (by covalent or non-covalent interactions) with portions of proteins with which it is naturally associated; (6) is functionally associated (by covalent or non-covalent interactions) with polypeptides with which it is not naturally associated; or (7) does not exist in nature. Such isolated proteins may be encoded by genomic DNA, cDNA, mRNA, or other RNA, may be of synthetic origin, or a combination thereof. In certain embodiments, an isolated protein is substantially free from proteins or polypeptides or other contaminants found in its natural environment that would interfere with its use (therapeutic, diagnostic, preventative, research, or otherwise).

[0103] The term "polynucleotide," as referred to herein, means a nucleotide of at least 10 bases in length, either ribonucleotides or deoxynucleotides, or a modified form of either type of nucleotide. The term includes single- and double-stranded forms of DNA. Polynucleotides according to the present disclosure include nucleic acid molecules that encode the polypeptides of the multispecific polypeptide constructs set forth herein.

[0104] The term "isolated polynucleotide," as used herein, is intended to mean a polynucleotide of genomic, cDNA, or synthetic origin, or a combination thereof, that, because of its origin, (1) is not associated in whole or in part with polynucleotides found in nature, (2) is operably linked to polynucleotides to which it is not linked in nature, or (3) does not occur in nature as part of a larger sequence.

[0105] The term "operably linked," as used herein, refers to the positioning of the components so described being in a relationship permitting them to function in their intended manner. A regulatory sequence "operably linked" to a coding sequence is ligated in such a way that expression of the coding sequence is achieved under conditions compatible with the regulatory sequences.

[0106] The term "substantial identity" as applied to polypeptides means that two peptide sequences, when optimally aligned using default gap weights, such as by the programs GAP or BESTFIT, share at least 80 percent sequence identity, e.g., in some embodiments, at least 90 percent sequence identity, in some embodiments, at least 95 percent sequence identity, and in some embodiments, at least 99 percent sequence identity. Specifically, conservative amino acid substitutions are contemplated. Conservative substitutions are those that take place within a family of amino acids whose side chains are related. Genetically encoded amino acids are generally classified into the following families: (1) acidic amino acids are aspartic acid and glutamic acid; (2) basic amino acids are lysine, arginine, and histidine; (3) nonpolar amino acids are alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine, and tryptophan; and (4) uncharged polar amino acids are glycine, asparagine, glutamine, cysteine, serine, threonine, and tyrosine. Hydrophilic amino acids include arginine, asparagine, aspartic acid, glutamine, glutamic acid, histidine, lysine, serine, and threonine. Hydrophobic amino acids include alanine, cysteine, isoleucine, leucine, methionine, phenylalanine, proline, tryptophan, tyrosine, and valine. Other families of amino acids include: (i) the aliphatic hydroxy family, serine and threonine; (ii) the amide-containing family, asparagine and glutamine; (iii) the aliphatic family, alanine, valine, leucine, and isoleucine; and (iv) the aromatic family, phenylalanine, tryptophan, and tyrosine.For example, it is reasonable to expect that the isolated substitution of leucine with isoleucine or valine, aspartic acid with glutamic acid, threonine with serine, or similar substitutions between structurally related amino acids will not have a significant effect on the binding or properties of the resulting molecule, especially when the substitution does not involve an amino acid in a framework position.Whether an amino acid change results in a functional peptide can be readily determined by assaying the specific activity of the polypeptide derivative, which assays are described in detail herein.

[0107] The term "agent" is used herein to refer to a chemical compound, a mixture of chemical compounds, a biological macromolecule, or an extract made from biological material.

[0108] As used herein, the term "label" or "labeled" refers to the incorporation of a detectable marker, for example, by incorporation of a radiolabeled amino acid into a polypeptide or by attachment of a biotinyl group that can be detected by marked avidin (e.g., streptavidin containing a fluorescent marker or enzymatic activity that can be detected by optical or calorimetric methods). In certain circumstances, the label or marker may also be therapeutic. A variety of methods of labeling polypeptides and glycoproteins are known in the art and can be used. Examples of labels for polypeptides include radioisotopes or radionuclides (e.g., 3 H, 14 C. 15 N, 35 S, 90 Y, 99 Tc, 111 In, 125 I, 131 I), fluorescent labels (e.g., fluorophores, rhodamine, lanthanide phosphors), enzymatic labels (e.g., horseradish peroxidase, p-galactosidase, luciferase, alkaline phosphatase), chemiluminescence, biotinyl groups, predetermined polypeptide epitopes recognized by secondary reporters (e.g., leucine zipper pair sequences, binding sites for secondary antibodies, metal binding domains, epitope tags). In some embodiments, labels are attached by spacer arms of various lengths to reduce potential steric hindrance.

[0109] The term "tumor-associated antigen" or "TAA," as used herein, refers to a counter-structure that is present primarily on tumor cells of a mammalian subject and is not generally found on normal cells of the mammalian subject. A tumor-specific antigen need not be exclusive to tumor cells; the proportion of cells bearing a particular mammalian tumor-associated antigen or the level of the tumor-associated antigen on the surface of the tumor is sufficiently high that it can be targeted by an anti-tumor therapeutic agent, such as a provided multispecific polypeptide construct, to provide prevention or treatment for the effects of the mammalian tumor. In some embodiments, in a random statistical sample of cells derived from a mammal with a tumor, at least 50% of cells exhibiting a TAA are cancer cells. In other embodiments, at least 60%, 70%, 80%, 85%, 90%, 95%, or 99% of cells exhibiting a TAA are cancer cells.

[0110] As used herein, a composition refers to any mixture of two or more products, substances, or compounds, such as cells, which may be a solution, suspension, liquid, powder, paste, aqueous, non-aqueous, or any combination thereof.

[0111] The term "pharmaceutical composition" refers to a composition suitable for pharmaceutical use in a mammalian subject, often a human. A pharmaceutical composition typically comprises an effective amount of an active agent (e.g., a multispecific polypeptide construct) and a carrier, excipient, or diluent. The carrier, excipient, or diluent is typically a pharmaceutically acceptable carrier, excipient, or diluent, respectively.

[0112] The terms "treating," "treatment," or "therapy" of a disease or disorder, as used herein, refer to the slowing, halting, or reversal of the progression of a disease or disorder, as evidenced by a reduction, arrest, or elimination of clinical or diagnostic symptoms, by administering a pharmaceutical composition of the present disclosure, alone or in combination with another compound described herein. "Treating," "treatment," or "therapy" also refers to a reduction in the severity of symptoms or a reduction in the rate of recurrence in acute or chronic diseases or disorders. As used herein, the terms "treating," or "inhibiting," "inhibiting," or "inhibition" of cancer, with respect to cancer, refer to at least one of a statistically significant reduction in the rate of tumor growth, a halt in tumor growth, or a decrease in tumor size, mass, metabolic activity, or volume, or a statistically significant increase in progression-free survival (PFS) or overall survival (OS), as measured by standard criteria, such as, but not limited to, Response Evaluation Criteria for Solid Tumors (RECIST). "Preventing," "prophylaxis," or "prevention" of a disease or disorder refers to the administration of a pharmaceutical composition, alone or in combination with another compound, to prevent the onset or development of a disease or disorder, or some or all of the symptoms of a disease or disorder, or to reduce the likelihood of developing a disease or disorder.

[0113] The term "effective amount" or "therapeutically effective amount" refers to an amount and / or concentration of a composition that, when administered to a patient alone (i.e., as monotherapy) or in combination with an additional therapeutic agent, provides a statistically significant reduction in disease progression, e.g., by ameliorating or eliminating the symptoms and / or causes of the disease. An effective amount can be an amount that relieves, reduces, or alleviates at least one symptom or biological response or effect associated with a disease or disorder, prevents the progression of a disease or disorder, or improves the patient's physical function.

[0114] As used herein, "substantially pure" means that a species is the predominant species present (i.e., more abundant than other individual species in the composition on a molar basis), and a substantially purified fraction is a composition in which a species constitutes at least about 50 percent (on a molar basis) of all macromolecular species present. Generally, a substantially pure composition will constitute greater than about 80 percent of all macromolecular species present in the composition, e.g., in some embodiments, greater than about 85%, 90%, 95%, and 99%. In some embodiments, a species is purified to essential homogeneity, such that the composition consists essentially of a single macromolecular species (contaminating species cannot be detected in the composition by conventional detection methods).

[0115] As used herein, a "subject" is a mammal, such as a human or other animal, typically a human. A subject, e.g., a patient, includes one in need of treatment to treat a disease or disorder. The term patient includes human and animal subjects. A subject may be male or female and may be of any appropriate age, including infant, juvenile, adolescent, adult, and geriatric subjects.

[0116] Other chemical terms herein are used in accordance with conventional usage in the art, as exemplified by the McGraw-Hill Dictionary of Chemical Terms (Parker, S., Ed., McGraw-Hill, San Francisco (1985)).

[0117] The term "about" as used herein refers to a general error range for each value, which is readily known to those skilled in the art. Reference to a value or parameter with "about" herein includes (describes) aspects related to the value or parameter itself. For example, a description of "about X" includes the description of "X."

[0118] II. Multispecific Polypeptide Constructs Provided herein is a multispecific polypeptide construct comprising a first component comprising an immunoglobulin Fc region and a second component comprising a CD3-binding region, wherein the first and second components are coupled by a linker, and the Fc region is positioned N-terminal to the CD3-binding region; one or both of the first and second components comprises (1) an antigen-binding domain that binds to a tumor-associated antigen (TAA) and (2) a costimulatory receptor-binding region (CRBR) that binds to a costimulatory receptor. Also provided herein is a multispecific polypeptide construct comprising a first component comprising an immunoglobulin Fc region and a second component comprising a CD3-binding region, wherein the first and second components are coupled by a linker, and the Fc region is positioned N-terminal to the CD3-binding region; one or both of the first and second components comprises (1) an antigen-binding domain that binds to a tumor-associated antigen (TAA) and (2) an inhibitory receptor-binding region (IRBR) that binds to an inhibitory receptor. In some embodiments, the multispecific polypeptide construct contains at least one antigen-binding domain that binds to a TAA. In some embodiments, the multispecific polypeptide construct contains at least a first antigen-binding domain that binds to a TAA and a second antigen-binding domain that binds to a TAA. In some embodiments, the multispecific polypeptide construct contains at least one CRBR that binds to a costimulatory receptor and / or at least one IRBR that binds to an inhibitory receptor. Exemplary formats of the multispecific constructs provided herein are shown in Figures 1-3 and 18A and 18B.

[0119] In some embodiments, the multispecific polypeptide construct contains, from N-terminus to C-terminus, an immunoglobulin Fc region; a linker; a CD3 binding region that binds to CD3 (CD3ε); and at least two binding domains, one an antigen binding domain that binds to a TAA and the other a CRBR that binds to a costimulatory receptor.

[0120] In some embodiments, the multispecific polypeptide construct comprises, in order from N-terminus to C-terminus, at least two binding domains, one an antigen binding domain that binds to a tumor-associated antigen (TAA) and the other a CRBR that binds to a costimulatory receptor; an immunoglobulin Fc region; a linker; and a CD3 binding domain that binds to CD3 (CD3ε).

[0121] In some embodiments, the multispecific polypeptide construct contains, from N-terminus to C-terminus, an immunoglobulin Fc region; a linker; a CD3 binding region that binds to CD3 (CD3ε); and at least two binding domains, one an antigen binding domain that binds to a TAA and the other an IRBR that binds to an inhibitory receptor.

[0122] In some embodiments, the multispecific polypeptide construct contains, from N-terminus to C-terminus, an immunoglobulin Fc region; a linker; a CD3 binding region that binds to CD3 (CD3ε); and at least two binding domains, one an antigen binding domain that binds to a TAA and the other an IRBR that binds to an inhibitory receptor.

[0123] In some embodiments, the multispecific polypeptide construct contains, in order from N-terminus to C-terminus, at least two binding domains, one an antigen-binding domain that binds to a tumor-associated antigen (TAA) and the other an IRBR that binds to an inhibitory receptor; an immunoglobulin Fc region; a linker; and a CD3-binding region that binds to CD3 (CD3ε).

[0124] In some embodiments, the multispecific polypeptide construct contains, in order from N-terminus to C-terminus, at least one antigen-binding domain that binds to a tumor-associated antigen (TAA) or a CRBR that binds to a costimulatory receptor; an immunoglobulin Fc region; a linker; a CD3-binding region that binds to CD3 (CD3ε); and at least one antigen-binding domain that binds to a TAA or the other of the CRBRs that bind to a costimulatory receptor. In some embodiments, the N-terminal or C-terminal portion of the multispecific polypeptide construct contains one antigen-binding domain that binds to a TAA. In some embodiments, the N-terminal or C-terminal portion of the multispecific polypeptide construct contains two antigen-binding domains that bind to a TAA. In some embodiments, each of the antigen-binding domains that bind to a TAA are identical. In some embodiments, at least the first antigen-binding domain and the second antigen-binding domain bind to different TAAs or different epitopes of the same TAA.

[0125] In some embodiments, the multispecific polypeptide construct contains, from N-terminus to C-terminus, at least one antigen-binding domain that binds to a tumor-associated antigen (TAA) or an IRBR that binds to an inhibitory receptor; an immunoglobulin Fc region; a linker; a CD3-binding region that binds to CD3 (CD3ε); and at least one antigen-binding domain that binds to a TAA or an IRBR that binds to an inhibitory receptor. In some embodiments, the N-terminal or C-terminal portion of the multispecific polypeptide construct contains one antigen-binding domain that binds to a TAA. In some embodiments, the N-terminal or C-terminal portion of the multispecific polypeptide construct contains two antigen-binding domains that bind to a TAA. In some embodiments, each of the antigen-binding domains that bind to a TAA are identical. In some embodiments, at least the first antigen-binding domain and the second antigen-binding domain bind to different TAAs or different epitopes of the same TAA.

[0126] In some embodiments, the multispecific polypeptide construct contains, in order from N-terminus to C-terminus, a CRBR that binds to a costimulatory receptor; an immunoglobulin Fc region; a linker; a CD3 binding region that binds to CD3 (CD3ε); and at least one antigen binding domain that binds to a TAA. In some embodiments, the C-terminal portion of the multispecific polypeptide construct contains one antigen binding domain that binds to a TAA. In some embodiments, the C-terminal portion of the multispecific polypeptide construct contains two antigen binding domains that bind to a TAA. In some embodiments, each of the antigen binding domains that bind to a TAA are identical. In some embodiments, at least the first antigen binding domain and the second antigen binding domain bind to different TAAs or different epitopes of the same TAA.

[0127] In some embodiments, the multispecific polypeptide construct contains, from N-terminus to C-terminus, an IRBR that binds to an inhibitory receptor; an immunoglobulin Fc region; a linker; a CD3 binding region that binds to CD3 (CD3ε); and at least one antigen-binding domain that binds to a TAA. In some embodiments, the C-terminal portion of the multispecific polypeptide construct contains one antigen-binding domain that binds to a TAA. In some embodiments, the C-terminal portion of the multispecific polypeptide construct contains two antigen-binding domains that bind to a TAA. In some embodiments, each of the antigen-binding domains that bind to a TAA are identical. In some embodiments, at least the first antigen-binding domain and the second antigen-binding domain bind to different TAAs or different epitopes of the same TAA.

[0128] In some embodiments, the multispecific polypeptide construct contains, in order from N-terminus to C-terminus, at least one antigen-binding domain that binds to a TAA; an immunoglobulin Fc region; a linker; a CD3-binding region that binds to CD3 (CD3ε); and a CRBR that binds to a costimulatory receptor. In some embodiments, the N-terminal portion of the multispecific polypeptide construct contains one antigen-binding domain that binds to a TAA. In some embodiments, the N-terminal portion of the multispecific polypeptide construct contains two antigen-binding domains that bind to a TAA. In some embodiments, each of the antigen-binding domains that bind to a TAA are identical. In some embodiments, at least the first antigen-binding domain and the second antigen-binding domain bind to different TAAs or different epitopes of the same TAA.

[0129] In some embodiments, the multispecific polypeptide construct contains, in order from N-terminus to C-terminus, at least one antigen-binding domain that binds to a TAA; an immunoglobulin Fc region; a linker; a CD3-binding region that binds to CD3 (CD3ε); and an IRBR that binds to an inhibitory receptor. In some embodiments, the N-terminal portion of the multispecific polypeptide construct contains one antigen-binding domain that binds to a TAA. In some embodiments, the N-terminal portion of the multispecific polypeptide construct contains two antigen-binding domains that bind to a TAA. In some embodiments, each of the antigen-binding domains that bind to a TAA are identical. In some embodiments, at least the first antigen-binding domain and the second antigen-binding domain bind to different TAAs or different epitopes of the same TAA.

[0130] In some embodiments, the multispecific polypeptide construct contains, in order from N-terminus to C-terminus, at least one antigen-binding domain that binds to a TAA; an immunoglobulin Fc region; a linker; a CD3-binding region that binds to CD3 (CD3ε); and at least two binding domains, one antigen-binding domain that binds to a TAA and the other CRBR that binds to a costimulatory receptor. In some embodiments, the C-terminal portion of the multispecific polypeptide construct contains two antigen-binding domains that bind to a TAA. In some embodiments, each of the antigen-binding domains that bind to a TAA is identical. In some embodiments, at least the first antigen-binding domain and the second antigen-binding domain bind to different TAAs or different epitopes of the same TAA.

[0131] In some embodiments, the multispecific polypeptide construct contains, in order from N-terminus to C-terminus, at least one antigen-binding domain that binds to a TAA; an immunoglobulin Fc region; a linker; a CD3-binding region that binds to CD3 (CD3ε); and at least two binding domains, one antigen-binding domain that binds to a TAA and the other IRBR that binds to an inhibitory receptor. In some embodiments, the C-terminal portion of the multispecific polypeptide construct contains two antigen-binding domains that bind to a TAA. In some embodiments, each of the antigen-binding domains that bind to a TAA is identical. In some embodiments, at least the first antigen-binding domain and the second antigen-binding domain bind to different TAAs or different epitopes of the same TAA.

[0132] In some embodiments, the multispecific polypeptide construct contains, in order from N-terminus to C-terminus, at least two binding domains, one antigen-binding domain that binds to a tumor-associated antigen (TAA) and the other CRBR that binds to a costimulatory receptor; an immunoglobulin Fc region; a linker; a CD3-binding domain that binds to CD3 (CD3ε); and at least one antigen-binding domain that binds to a TAA. In some embodiments, the N-terminal portion of the multispecific polypeptide construct contains two antigen-binding domains that bind to a TAA. In some embodiments, each of the antigen-binding domains that bind to a TAA is identical. In some embodiments, at least the first antigen-binding domain and the second antigen-binding domain bind to different TAAs or different epitopes of the same TAA.

[0133] In some embodiments, the multispecific polypeptide construct contains, in order from N-terminus to C-terminus, at least two binding domains, one antigen-binding domain that binds to a tumor-associated antigen (TAA) and the other IRBR that binds to an inhibitory receptor; an immunoglobulin Fc region; a linker; a CD3-binding region that binds to CD3 (CD3ε); and at least one antigen-binding domain that binds to a TAA. In some embodiments, the N-terminal portion of the multispecific polypeptide construct contains two antigen-binding domains that bind to a TAA. In some embodiments, each of the antigen-binding domains that bind to a TAA is identical. In some embodiments, at least the first antigen-binding domain and the second antigen-binding domain bind to different TAAs or different epitopes of the same TAA.

[0134] In some embodiments, the multispecific polypeptide construct contains, in order from N-terminus to C-terminus, at least two binding domains, one antigen-binding domain that binds to a tumor-associated antigen (TAA) and the other CRBR that binds to a costimulatory receptor; an immunoglobulin Fc region; a linker; a CD3 binding region that binds to CD3 (CD3ε); and at least two binding domains, one antigen-binding domain that binds to a TAA and the other IRBR that binds to an inhibitory receptor. In some embodiments, each of the antigen-binding domains that bind to a TAA are identical. In some embodiments, at least the first antigen-binding domain and the second antigen-binding domain bind to different TAAs or different epitopes of the same TAA.

[0135] In some embodiments, the multispecific polypeptide construct contains, in order from N-terminus to C-terminus, at least two binding domains, one of which is an antigen-binding domain that binds to a tumor-associated antigen (TAA) and the other of which is an IRBR that binds to an inhibitory receptor; an immunoglobulin Fc region; a linker; a CD3 binding region that binds to CD3 (CD3ε); and at least two binding domains, one of which is an antigen-binding domain that binds to a TAA and the other of which is a CRBR that binds to a costimulatory receptor. In some embodiments, each of the antigen-binding domains that bind to a TAA are identical. In some embodiments, at least the first antigen-binding domain and the second antigen-binding domain bind to different TAA or different epitopes of the same TAA.

[0136] In some embodiments, the multispecific polypeptide construct is a dimer, and dimerization is formed by covalent or non-covalent interactions between two polypeptide chains. In some embodiments, the two polypeptide chains are covalently linked to each other, for example, by an interchain disulfide bond. In some embodiments, the Fc region mediates dimerization via an interchain disulfide bond. In some embodiments, the multispecific polypeptide construct contains a homodimeric Fc region, and in some cases, both polypeptide chains of the multispecific polypeptide construct are identical (homodimers). In some embodiments, the multispecific polypeptide construct contains a heterodimeric Fc region, and in some cases, the polypeptide chains of the multispecific polypeptide construct are different (heterodimers). In a specific example of a heterodimeric multispecific polypeptide construct, the CD3 binding region is a two-chain polypeptide containing a VH chain and a VL chain, for example, an Fv antibody fragment containing a VH and a VL chain. In some embodiments, the Fv antibody fragment comprises a disulfide-stabilized anti-CD3 binding Fv fragment (dsFv).

[0137] In some embodiments, the multispecific polypeptide construct is formed from or comprises two polypeptides, including a first polypeptide comprising a first Fc polypeptide of a heterodimeric Fc region, a linker, and a VH domain of an anti-CD3 antibody or antigen-binding fragment (e.g., Fv); and a second polypeptide comprising a second Fc polypeptide of a heterodimeric Fc region, a linker, and a VL domain of an anti-CD3 antibody or antigen-binding fragment (e.g., Fv). In some embodiments, the first polypeptide contains one or two antigen-binding domains, or chains thereof, that bind to a TAA. In some embodiments, the second polypeptide contains one or two antigen-binding domains, or chains thereof, that bind to a TAA. In some embodiments, each antigen-binding domain that binds to a TAA is positioned N-terminal to the Fc polypeptide and / or C-terminal to a chain of the CD3-binding region. In some embodiments, the first polypeptide, the second polypeptide, or both the first and second polypeptides further comprise a CRBR, such that the multispecific polypeptide construct contains at least one CRBR. In some embodiments, the multispecific polypeptide construct contains two IRBRs. In some embodiments, the IRBR of the first and / or second polypeptide may be located N-terminal to the Fc polypeptide and / or C-terminal to the chain of the CD3 binding region. In some embodiments, the first polypeptide or the second polypeptide, or both the first and second polypeptides, further comprise an IRBR, such that the multispecific polypeptide construct contains at least one IRBR. In some embodiments, the multispecific polypeptide construct contains two IRBRs. In some embodiments, the IRBR of the first and / or second polypeptide may be located N-terminal to the Fc polypeptide and / or C-terminal to the chain of the CD3 binding region.In some embodiments, the first polypeptide or the second polypeptide or both the first and second polypeptides further comprise a CRBR, and the first polypeptide or the second polypeptide or both the first and second polypeptides further comprise an IRBR, such that the multispecific polypeptide construct contains at least one CRBR and at least one IRBR.

[0138] Various exemplary multispecific polypeptide constructs containing at least two polypeptide chains are described herein, for example, in Figures 1-3 and 18A-B. Any of such configurations of multispecific polypeptide constructs are contemplated.

[0139] In some embodiments, the multispecific polypeptide construct contains at least two antigen-binding domains that bind to tumor-associated antigens (TAA) and at least one costimulatory receptor-binding region (CRBR) that binds to a costimulatory receptor. In some embodiments, the multispecific polypeptide construct comprises: (1) a first polypeptide comprising, from N-terminus to C-terminus, a first antigen-binding domain that binds to a tumor-associated antigen (TAA), a first Fc polypeptide of a heterodimeric Fc region, a linker (e.g., a cleavable linker), a chain (e.g., VH or VL) of an anti-CD3 antibody or antigen-binding fragment (e.g., Fv or dsFv), and a second antigen-binding domain that binds to a tumor-associated antigen (TAA); and (2) a second polypeptide comprising, from N-terminus to C-terminus, a second Fc polypeptide of a heterodimeric Fc region, an identical linker (e.g., the same cleavable linker), the other chain (the other of VH or VL) of the anti-CD3 antibody or antigen-binding fragment, and a costimulatory receptor-binding region (CRBR) that binds to a costimulatory receptor. In some embodiments, the multispecific polypeptide construct contains: (1) a first polypeptide comprising, from N-terminus to C-terminus, a first antigen-binding domain that binds to a tumor-associated antigen (TAA), a first Fc polypeptide of a heterodimeric Fc region, a linker (e.g., a cleavable linker), a chain (e.g., VH or VL) of an anti-CD3 antibody or antigen-binding fragment (e.g., Fv or dsFv), and a second antigen-binding domain that binds to the tumor-associated antigen (TAA); and (2) a second polypeptide comprising, from N-terminus to C-terminus, a costimulatory receptor-binding region (CRBR) that binds to a costimulatory receptor, a second Fc polypeptide of a heterodimeric Fc region, an identical linker (e.g., the same cleavable linker), and the other chain (the other of VH or VL) of the anti-CD3 antibody or antigen-binding fragment.

[0140] In some embodiments, the multispecific polypeptide construct contains at least two antigen-binding domains that bind to tumor-associated antigens (TAA) and at least one inhibitory receptor-binding region (IRBR) that binds to an inhibitory receptor. In some embodiments, the multispecific polypeptide construct contains: (1) a first polypeptide comprising, from N-terminus to C-terminus, a first antigen-binding domain that binds to a tumor-associated antigen (TAA), a first Fc polypeptide of a heterodimeric Fc region, a linker (e.g., a cleavable linker), a chain (e.g., VH or VL) of an anti-CD3 antibody or antigen-binding fragment (e.g., Fv or dsFv), and a second antigen-binding domain that binds to the tumor-associated antigen (TAA); and (2) a second polypeptide comprising, from N-terminus to C-terminus, a second Fc polypeptide of a heterodimeric Fc region, an identical linker (e.g., the same cleavable linker), the other chain (the other of VH or VL) of the anti-CD3 antibody or antigen-binding fragment, and an inhibitory receptor-binding region (IRBR) that binds to an inhibitory receptor. In some embodiments, the multispecific polypeptide construct contains: (1) a first polypeptide comprising, from N-terminus to C-terminus, a first antigen-binding domain that binds to a tumor-associated antigen (TAA), a first Fc polypeptide of a heterodimeric Fc region, a linker (e.g., a cleavable linker), a chain (e.g., VH or VL) of an anti-CD3 antibody or antigen-binding fragment (e.g., Fv or dsFv), and a second antigen-binding domain that binds to the tumor-associated antigen (TAA); and (2) a second polypeptide comprising, from N-terminus to C-terminus, an inhibitory receptor-binding region (IRBR) that binds to an inhibitory receptor, a second Fc polypeptide of a heterodimeric Fc region, an identical linker (e.g., the same cleavable linker), and the other chain (the other of VH or VL) of the anti-CD3 antibody or antigen-binding fragment.

[0141] In some embodiments, the multispecific polypeptide construct contains at least two antigen binding domains that bind to tumor-associated antigens (TAA), at least one costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor, and at least one inhibitory receptor binding region (IRBR) that binds to an inhibitory receptor. In some embodiments, the multispecific polypeptide construct contains: (1) a first polypeptide comprising, from N-terminus to C-terminus, a first antigen-binding domain that binds to a tumor-associated antigen (TAA), a first Fc polypeptide of a heterodimeric Fc region, a linker (e.g., a cleavable linker), a chain (e.g., VH or VL) of an anti-CD3 antibody or antigen-binding fragment (e.g., Fv or dsFv), and a second antigen-binding domain that binds to a tumor-associated antigen (TAA); and (2) a second polypeptide comprising, from N-terminus to C-terminus, one of CRBR or IRBR, a second Fc polypeptide of a heterodimeric Fc region, an identical linker (e.g., the same cleavable linker), the other chain (the other of VH or VL) of the anti-CD3 antibody or antigen-binding fragment, and the other of CRBR or IRBR.

[0142] Each of the components of the multispecific polypeptide constructs of the present disclosure is described in more detail below.

[0143] 1. Anti-CD3 binding domain The multispecific polypeptide constructs of the present disclosure comprise one or more copies of an anti-CD3 binding domain. The anti-CD3 binding domain of the present disclosure activates T cells through engagement of CD3ε on T cells. The anti-CD3 binding domain of the present disclosure agonizes, stimulates, activates, and / or otherwise augments CD3-mediated T cell activation. The biological activity of CD3 includes, for example, T cell activation and other signaling through the interaction between CD3 and the antigen-binding subunit of the T cell receptor (TCR). For example, the anti-CD3 binding domain of the present disclosure activates T cells completely or partially through engagement of CD3ε on T cells by partially or completely modulating, e.g., agonizing, stimulating, activating, or otherwise augmenting, CD3-mediated T cell activation.

[0144] In preferred embodiments, the anti-CD3 binding domain of the present disclosure specifically binds to the ε chain of CD3, also known as CD3ε. The anti-CD3ε binding domain of the present disclosure activates T cells through engagement of CD3ε in T cells. The anti-CD3ε binding domain of the present disclosure includes monoclonal antibodies, such as mammalian monoclonal antibodies, primate monoclonal antibodies, fully human monoclonal antibodies, including humanized monoclonal antibodies and chimeric antibodies, as well as antigen-binding fragments thereof. In some embodiments, the anti-CD3ε binding domain comprises one or more copies of an antibody or its antigen-binding fragment.

[0145] In some embodiments, the anti-CD3ε binding domain comprises one or more copies of an antibody or antigen-binding fragment thereof selected from the group consisting of a Fab fragment, a F(ab')2 fragment, an Fv fragment, an scFv, a scAb, a dAb, a single-domain heavy chain antibody, and a single-domain light chain antibody. In some embodiments, the anti-CD3ε binding domain comprises an Fv antibody fragment that binds to CD3ε (referred to herein as an anti-CD3ε Fv fragment). In some embodiments, the anti-CD3ε Fv antibody fragment is a disulfide-stabilized anti-CD3 binding Fv fragment (dsFv). In some embodiments, the anti-CD3 binding domain is monovalent for CD3 binding.

[0146] In some embodiments, the CD3 binding region is not a single chain antibody, e.g., in some aspects, the CD3 binding region is not a single chain variable fragment (scFv).

[0147] In some embodiments, the CD3 binding region is an Fv antibody fragment containing a variable heavy chain (Hv, also referred to as VH) and a variable light chain (Lv, also referred to as VL), such as any of those described. In aspects of such embodiments, the immunoglobulin Fc region is a heterodimeric Fc region containing two different Fc polypeptides, such as any of those described in Section II.2, that are capable of heterodimeric association between both polypeptides of the Fc heterodimer. In such embodiments, the variable heavy chain (Hv) and variable light chain (Lv) of the CD3 binding region are linked to opposite chains of the heterodimeric Fc.

[0148] In some embodiments, the anti-CD3ε binding domain comprises a combination of a VH CDR1 sequence, a VH CDR2 sequence, and a VH CDR3 sequence, wherein at least one of the VH CDR1 sequence, the VH CDR2 sequence, and the VH CDR3 sequence comprises at least the amino acid sequence TYAMN (SEQ ID NO: 16); TIFF0007651450000011.tif4128; and a VH CDR3 sequence comprising at least the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO: 18).

[0149] In some embodiments, the anti-CD3ε binding domain comprises a combination of a VL CDR1 sequence, a VL CDR2 sequence, and a VL CDR3 sequence, wherein at least one of the VL CDR1 sequence, the VL CDR2 sequence, and the VL CDR3 sequence is selected from a VL CDR1 sequence comprising at least the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO: 19); a VL CDR2 sequence comprising at least the amino acid sequence GTNKRAP (SEQ ID NO: 20); and a VL CDR3 sequence comprising at least the amino acid sequence ALWYSNLWV (SEQ ID NO: 21).

[0150] In some embodiments, the anti-CD3ε binding domain comprises a VH CDR1 sequence comprising at least the amino acid sequence TYAMN (SEQ ID NO:16); TIFF0007651450000012.tif4128; a VH CDR3 sequence comprising at least the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO:18), a VL CDR1 sequence comprising at least the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO:19); a VL CDR2 sequence comprising at least the amino acid sequence GTNKRAP (SEQ ID NO:20); and a VL CDR3 sequence comprising at least the amino acid sequence ALWYSNLWV (SEQ ID NO:21).

[0151] In some embodiments, the anti-CD3ε binding domain comprises a combination of a VH CDR1 sequence, a VH CDR2 sequence, and a VH CDR3 sequence, wherein at least one of the VH CDR1 sequence, the VH CDR2 sequence, and the VH CDR3 sequence comprises a VH CDR1 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence TYAMN (SEQ ID NO:16); a VH CD2 sequence comprising a sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to TIFF0007651450000013.tif4128; and a VH CDR3 sequence comprising a sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO:18).

[0152] In some embodiments, the anti-CD3 epsilon binding domain comprises a combination of a VL CDR1 sequence, a VL CDR2 sequence, and a VL CDR3 sequence, wherein at least one of the VL CDR1 sequence, the VL CDR2 sequence, and the VL CDR3 sequence comprises a VL CDR1 sequence comprising a sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO:19); a VL CDR2 sequence comprising a sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence GTNKRAP (SEQ ID NO:20); and an amino acid sequence ALWYSNLWV (SEQ ID NO:21). The VL CDR3 sequence is selected from VL CDR3 sequences containing sequences that are at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the VL CDR3 sequence of ...

[0153] In some embodiments, the anti-CD3 epsilon binding domain comprises a VH CDR1 sequence comprising at least the amino acid sequence GFTFNTYAMN (SEQ ID NO:312); a VH CDR2 sequence comprising at least the amino acid sequence RIRSKYNNYATY (SEQ ID NO:318); a VH CDR3 sequence comprising at least the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO:18), a VL CDR1 sequence comprising at least the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO:19); a VL CDR2 sequence comprising at least the amino acid sequence GTNKRAP (SEQ ID NO:20); and a VL CDR3 sequence comprising at least the amino acid sequence ALWYSNLWV (SEQ ID NO:21).

[0154] In some embodiments, the anti-CD3ε binding domain comprises a VH CDR1 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence TYAMN (SEQ ID NO:16); a VH CD2 sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to TIFF0007651450000014.tif4128; a VH CDR3 sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO:18); a VL CDR1 sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO:19); and a VL CDR3 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence ALWYSNLWV (SEQ ID NO:21).

[0155] In some embodiments, the anti-CD3 epsilon binding domain comprises a VH CDR1 sequence comprising at least the amino acid sequence GFTFNTYAMN (SEQ ID NO:312); a VH CDR2 sequence comprising at least the amino acid sequence RIRSKYNNYATY (SEQ ID NO:318); a VH CDR3 sequence comprising at least the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO:18), a VL CDR1 sequence comprising at least the amino acid sequence GSSTGAVTTSNYAN (SEQ ID NO:319); a VL CDR2 sequence comprising at least the amino acid sequence GTNKRAP (SEQ ID NO:320); and a VL CDR3 sequence comprising at least the amino acid sequence ALWYSNHWV (SEQ ID NO:315).

[0156] In some embodiments, the anti-CD3 epsilon binding domain comprises a VH CDR1 sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence GFTFNTYAMN (SEQ ID NO:312); a VH CDR2 sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence RIRSKYNNYATY (SEQ ID NO:313); a VH CDR3 sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO:18); a VL CDR1 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence GTNKRAP (SEQ ID NO:320); and a VL CDR3 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence ALWYSNHWV (SEQ ID NO:315).

[0157] In some embodiments, the anti-CD3ε binding domain is an Fv fragment comprising a combination of a heavy chain variable amino acid sequence and a light chain variable amino acid sequence. In some embodiments, the anti-CD3ε binding domain is an Fv fragment comprising a combination of a heavy chain variable amino acid sequence and a light chain variable amino acid sequence comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to an amino acid sequence selected from the group consisting of SEQ ID NOs: 32-81 and 241. In some embodiments, the anti-CD3ε binding domain is an Fv fragment comprising a combination of a heavy chain variable amino acid sequence and a light chain variable amino acid sequence comprising an amino acid sequence selected from the group consisting of SEQ ID NOs: 32-81 and 241. In some embodiments, the anti-CD3ε binding domain is an Fv fragment comprising a combination of a heavy chain variable amino acid sequence selected from the group consisting of SEQ ID NOs: 32-62 and a light chain variable amino acid sequence selected from the group consisting of SEQ ID NOs: 63-81 and 241. In some embodiments, the anti-CD3ε binding domain is an Fv fragment comprising a combination of a heavy chain variable amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to an amino acid sequence selected from the group consisting of SEQ ID NOs: 32-62 and a light chain variable amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to an amino acid sequence selected from the group consisting of SEQ ID NOs: 63-81 and 241.

[0158] In some embodiments, the anti-CD3ε binding domain comprises a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence comprising an amino acid sequence selected from the group of SEQ ID NOs: 32-81 and 241. In some embodiments, the anti-CD3ε binding domain comprises a combination of a heavy chain variable region amino acid sequence selected from the group of SEQ ID NOs: 32-62 and a light chain variable region amino acid sequence comprising an amino acid sequence selected from the group of SEQ ID NOs: 63-81 and 241.

[0159] In some embodiments, the anti-CD3ε Fv antibody fragment comprises a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence selected from the group consisting of SEQ ID NOs: 14, 15, 32-81, 241, 287-291, and 311. In some embodiments, the anti-CD3ε Fv antibody fragment comprises a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to an amino acid sequence selected from the group consisting of SEQ ID NOs: 14, 15, 32-81, 241, 287-291, and 311. In some embodiments, the anti-CD3ε Fv antibody fragment comprises a combination of a variable heavy chain amino acid sequence selected from the group of SEQ ID NOs: 14, 32-62, and 287, 290, and 311, and a variable light chain amino acid sequence selected from the group consisting of SEQ ID NOs: 15, 63-81, 241, 288, and 289. In some embodiments, the anti-CD3ε Fv antibody fragment comprises a combination of a variable heavy chain amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to an amino acid sequence selected from the group consisting of SEQ ID NOs: 14, 32-62, 287, 290, and 311, and a variable light chain amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to an amino acid sequence selected from the group consisting of SEQ ID NOs: 15, 63-81, 241, 288, and 289.

[0160] In some embodiments, the anti-CD3ε binding domain is an Fv fragment comprising a combination of a heavy chain variable amino acid sequence and a light chain variable amino acid sequence. In some embodiments, the anti-CD3ε Fv antibody fragment comprises a variable heavy chain amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to an amino acid sequence selected from the group consisting of SEQ ID NOs: 14, 32-43, 45-47, 48, and 287, in combination with a variable light chain amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to an amino acid sequence selected from the group consisting of SEQ ID NOs: 15, 63, 65-71, 73, 75, 77, and 288. In some embodiments, the anti-CD3ε Fv antibody fragment comprises a combination of a variable heavy chain amino acid sequence selected from the group of SEQ ID NOs: 14, 32-43, 45-47, 48, and 287 and a variable light chain amino acid sequence selected from the group consisting of SEQ ID NOs: 15, 63, 65-71, 73, 75, 77, and 288.

[0161] In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (Hv) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO: 14. In some embodiments, the anti-CD3ε binding domain comprises a variable light chain (Lv) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO: 15. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (Hv) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO: 14 and a variable light chain (Lv) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO: 15. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (Hv) comprising the amino acid sequence of SEQ ID NO: 14. In some embodiments, the anti-CD3ε binding domain comprises a variable light chain (Lv) comprising the amino acid sequence of SEQ ID NO: 15. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (Hv) comprising the amino acid sequence of SEQ ID NO:14 and a variable light chain (Lv) comprising the amino acid sequence of SEQ ID NO:15.

[0162] In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (VH) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO: 287. In some embodiments, the anti-CD3ε binding domain comprises a variable light chain (VL) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO: 288. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (VH) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO:287 and a variable light chain (VL) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO:288. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (VH) comprising the amino acid sequence of SEQ ID NO:287. In some embodiments, the anti-CD3ε binding domain comprises a variable light chain (VL) comprising the amino acid sequence of SEQ ID NO:288. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (VH) comprising the amino acid sequence of SEQ ID NO:287 and a variable light chain (VL) comprising the amino acid sequence of SEQ ID NO:288.

[0163] In a specific embodiment, Fv is V H -V LThe heterodimer is a disulfide-stabilized Fv (dsFv) stabilized by an interchain disulfide bond. In some embodiments, the interchain disulfide bond is altered by mutation of a position in the framework position of the VH chain and / or the VL chain. In some embodiments, the VH chain contains a Cys mutation at position 44 and the VL chain contains a Cys mutation at position 100, each according to the Kabat numbering. For example, in some embodiments, the VH chain contains a G44C mutation and the VL chain contains a G100C mutation, each according to the Kabat numbering. In some embodiments, the disulfide-stabilized anti-CD3 Fv comprises an anti-CD3 VH containing a Cys mutation at position 105 and an anti-CD3 VL containing a Cys mutation at position 43, each according to the Kabat numbering.

[0164] In some embodiments, the anti-CD3ε Fv comprises a combination of a variable heavy chain amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to an amino acid sequence selected from the group consisting of SEQ ID NOs: 44, and 49-62, 290, and 311, and a variable light chain amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to an amino acid sequence selected from the group consisting of SEQ ID NOs: 64, 72, 74, 76, 78-81, 241, and 289. In some such embodiments, the anti-CD3 Fv is a dsFv having a VH chain containing a G44C mutation and a VL chain containing a G100C mutation, each according to the Kabat numbering. In some embodiments, the anti-CD3ε Fv antibody fragment comprises a combination of a variable heavy chain amino acid sequence selected from the group of SEQ ID NOs: 44, and 49-62, 290, and 311, and a variable light chain amino acid sequence selected from the group consisting of SEQ ID NOs: 64, 72, 74, 76, 78-81, 241, and 289.

[0165] In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (Hv) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO: 44. In some embodiments, the anti-CD3ε binding domain comprises a variable light chain (Lv) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO: 72. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (Hv) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO:44, and a variable light chain (Lv) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO:72. In some such embodiments, the anti-CD3 Fv is a dsFv having a VH chain containing a G44C mutation and a VL chain containing a G100C mutation, each according to the Kabat numbering. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (Hv) comprising the amino acid sequence of SEQ ID NO:44. In some embodiments, the anti-CD3ε binding domain comprises a variable light chain (Lv) comprising the amino acid sequence of SEQ ID NO: 72. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (Hv) comprising the amino acid sequence of SEQ ID NO: 44 and a variable light chain (Lv) comprising the amino acid sequence of SEQ ID NO: 72.

[0166] In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (Hv) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO: 44. In some embodiments, the anti-CD3ε binding domain comprises a variable light chain (Lv) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO: 241. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (Hv) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO:44, and a variable light chain (Lv) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO:241. In some such embodiments, the anti-CD3 Fv is a dsFv having a VH chain containing a G44C mutation and a VL chain containing a G100C mutation, each according to the Kabat numbering. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (Hv) comprising the amino acid sequence of SEQ ID NO:44. In some embodiments, the anti-CD3ε binding domain comprises a variable light chain (Lv) comprising the amino acid sequence of SEQ ID NO: 241. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (Hv) comprising the amino acid sequence of SEQ ID NO: 44 and a variable light chain (Lv) comprising the amino acid sequence of SEQ ID NO: 241.

[0167] In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (VH) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO: 290. In some embodiments, the anti-CD3ε binding domain comprises a variable light chain (VL) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO: 289. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (VH) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO:290, and a variable light chain (VL) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO:289. In some such embodiments, the anti-CD3 Fv is a dsFv having a VH chain containing a G44C mutation and a VL chain containing a G100C mutation, each according to the Kabat numbering. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (VH) comprising the amino acid sequence of SEQ ID NO:290. In some embodiments, the anti-CD3ε binding domain comprises a variable light chain (VL) comprising the amino acid sequence of SEQ ID NO: 289. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (VH) comprising the amino acid sequence of SEQ ID NO: 290 and a variable light chain (VL) comprising the amino acid sequence of SEQ ID NO: 289.

[0168] In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (VH) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO: 311. In some embodiments, the anti-CD3ε binding domain comprises a variable light chain (VL) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO: 289. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (VH) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO:311, and a variable light chain (VL) comprising an amino acid sequence at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence of SEQ ID NO:289. In some such embodiments, the anti-CD3 Fv is a dsFv having a VH chain containing a G44C mutation and a VL chain containing a G100C mutation, each according to the Kabat numbering. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (VH) comprising the amino acid sequence of SEQ ID NO:311. In some embodiments, the anti-CD3ε binding domain comprises a variable light chain (VL) comprising the amino acid sequence of SEQ ID NO: 289. In some embodiments, the anti-CD3ε binding domain comprises a variable heavy chain (VH) comprising the amino acid sequence of SEQ ID NO: 311 and a variable light chain (VL) comprising the amino acid sequence of SEQ ID NO: 289.

[0169] 2. Immunoglobulin Fc Polypeptides The multispecific polypeptide constructs of the present disclosure comprise an immunoglobulin Fc region. In some embodiments, the immunoglobulin Fc region is an IgG isotype selected from the group consisting of IgG1 isotype, IgG2 isotype, IgG3 isotype, and IgG4 subclass. In some embodiments, the Fc region is human Fc. In some embodiments, the immunoglobulin Fc region is a polypeptide comprising an amino acid sequence selected from the group consisting of SEQ ID NOs: 1-6. In some embodiments, the immunoglobulin Fc region contains an Fc chain that is an immunoactive fragment of any of SEQ ID NOs: 1-6. In some embodiments, the immunoglobulin Fc region contains an Fc polypeptide chain that is at least 50%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of any of SEQ ID NOs: 1-6, or an immunoactive fragment thereof.

[0170] In some embodiments, the multispecific polypeptide construct is a dimer formed by polypeptides each containing an Fc. In some specific embodiments, identical or substantially identical polypeptides will dimerize to create a homodimer. In some embodiments, the dimer is a homodimer in which the two polypeptides of the multispecific polypeptide construct are identical. In other cases, the Fc region is formed by an Fc domain that has been mutated or modified to promote heterodimerization, in which different polypeptides dimerize to give a heterodimer. Thus, in some embodiments, the dimer is a heterodimer in which the two polypeptide chains of the multispecific polypeptide construct are different. Exemplary modifications to promote heterodimerization are known and include the following:

[0171] Generally, the Fc region is responsible for effector functions such as complement-dependent cytotoxicity (CDC) and antibody-dependent cellular cytotoxicity (ADCC) in addition to antigen-binding ability, which is the primary function of immunoglobulins. Furthermore, the FcRn sequence present in the Fc region plays a role in regulating serum IgG levels by increasing in vivo half-life through conjugation with the in vivo FcRn receptor. In some embodiments, such functions may be altered, for example, reduced or enhanced, in the Fc for use with the provided multispecific polypeptide constructs.

[0172] In some embodiments, the Fc region of the provided multispecific polypeptide constructs exhibits one or more effector functions. In some cases, the Fc region can provide Fc-mediated effector functions, such as, for example, ADCC (e.g., release of granzyme B by NK cells), ADCP, and / or CDC. In some embodiments in which the multispecific polypeptide construct contains a cleavable linker, cleavage of the linker can produce two components, each with biological activity: a CD3-binding region capable of binding and engaging CD3 on T cells (which in some aspects can also include CRBR for inducing costimulatory signals on T cells), and an Fc region linked to a TAA antigen-binding domain capable of exhibiting target-specific effector function. In certain embodiments provided herein, the multispecific polypeptide constructs contain a non-cleavable linker and, in some aspects, may not exhibit independent Fc-mediated effector function.

[0173] In some embodiments, the Fc region comprises an Fc polypeptide that has been mutated or modified to alter one or more effector functions. Various examples of mutations in Fc polypeptides to alter, e.g., reduce, effector function are known and include those described below. In some embodiments, unless described with reference to a particular SEQ ID NO, references to amino acid substitutions in the Fc region are in accordance with EU numbering according to Kabat (also referred to as Kabat numbering). EU numbering is known and is according to the EU index reported in the latest IMGT Scientific Chart (IMGT®, the international ImMunoGeneTics information system®, http: / / www.imgt.org / IMGTScientificChart / Numbering / Hu_IGHGnber.html (created: 17 May 2001, last updated: 10 Jan 2013) and Kabat, EA et al. Sequences of Proteins of Immunological interest. 5th ed. US Department of Health and Human Services, NIH publication No. 91-3242 (1991).

[0174] In some embodiments, provided multispecific polypeptide constructs containing an Fc region exhibiting reduced effector function may be desirable candidates for applications in which constrained CD3 binding is desired, but certain effector functions (such as CDC and ADCC) are unnecessary or deleterious. In vitro and / or in vivo cytotoxicity assays can be performed to confirm reduced / depleted CDC and / or ADCC activity. For example, Fc receptor (FcR) binding assays can be performed to ensure that the multispecific polypeptide construct and / or truncated components thereof lack FcγR binding (and thus likely lack ADCC activity) but retain FcRn binding ability. The primary cells for mediating ADCC, NK cells, express only FcγRIII, while monocytes express FcγRI, FcγRII, and FcγRIII. Non-limiting examples of in vitro assays to assess ADCC activity of a molecule of interest are described in U.S. Pat. No. 5,500,362 (see, e.g., Hellstrom, I. et al. Proc. Nat'l Acad. Sci. USA 83:7059-7063(1986)) and Hellstrom, I. et al., Proc. Nat'l Acad. Sci. USA 82:1499-1502(1985); U.S. Pat. No. 5,821,337 (see, Bruggemann, M. et al., J. Exp. Med. 166: 1351-1361(1987)). Alternatively, non-radioactive assay methods may be utilized (see, e.g., ACTI™ Non-Radioactive Cytotoxicity Assay for Flow Cytometry (CellTechnology, Inc. Mountain View, Calif.); and CytoTox 96™ Non-Radioactive Cytotoxicity Assay (Promega, Madison, Wis.)). Useful effector cells for such assays include peripheral blood mononuclear cells (PBMCs) and natural killer (NK) cells.Alternatively, or in addition, the ADCC activity of the molecule of interest can be assessed in vivo, for example, in an animal model such as that disclosed in Clynes et al. Proc. Nat'l Acad. Sci. USA 95:652-656 (1998). C1q binding assays can also be performed to confirm that the multispecific polypeptide construct or a truncated component thereof is unable to bind C1q and therefore lacks CDC activity. See, for example, the C1q binding ELISA and C3c binding ELISA of WO 2006 / 029879 and WO 2005 / 100402. To assess complement activation, a CDC assay can be performed (see, e.g., Gazzano-Santoro et al., J. Immunol. Methods 202:163(1996); Cragg, MS et al., Blood 101:1045-1052(2003); and Cragg, MS and MJ Glennie, Blood 103:2738-2743(2004)). Determination of FcRn binding and in vivo clearance / half-life can also be performed using methods known in the art (see, e.g., Petkova, SB et al., Int'l. Immunol. 18(12):1759-1769(2006)).

[0175] In some embodiments, the immunoglobulin Fc region or immunologically active fragment thereof is of the IgG isotype. For example, the immunoglobulin Fc region of the fusion protein is of the human IgG1 isotype, having the following amino acid sequence: TIFF0007651450000015.tif31128

[0176] In some embodiments, the immunoglobulin Fc region or immunologically active fragment thereof comprises a human IgG1 polypeptide sequence that is at least 50%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of SEQ ID NO:1.

[0177] In some embodiments, an IgG1 Fc polypeptide or variant thereof, such as any of those described below, can be made with the G1 m1 or G1 m3 allotype. In some embodiments, the Fc region can contain amino acids of the human G1 m1 allotype, such as residues containing Asp (D) and Leu (L) at positions 356 and 358, as shown in SEQ ID NO:1. In some cases, the Fc polypeptide can contain amino acid substitutions E356D and M358L to reconstitute residues of allotype G1 m1. In other embodiments, the Fc region can contain amino acids of the human G1 m3 allotype, such as residues Glu (E) and Met (M) at positions 356 and 358, according to EU numbering, as shown in SEQ ID NOs:194 and 195. In some cases, the Fc polypeptide may contain amino acid substitutions D356E and L358M to reconstitute residues of allotype G1 m3.

[0178] In some embodiments, the human IgG1 Fc region is modified to alter antibody-dependent cellular cytotoxicity (ADCC) and / or complement-dependent cytotoxicity (CDC). For example, Natsume et al., 2008 Cancer Res,68(10): 3863-72;Idusogie et al., 2001 J Immunol, 166 (4):2571-5;Moore et al., 2010 mAbs, 2 (2): 181-189;Lazar et al., 2006 PNAS, 103 (11): 4005-4010, Shields et al., 2001 JBC, 276 (9): 6591-6604;Stavenhagen et al., 2007 Cancer Res,67(18): 8882-8890;Stavenhagen et al.,2008 Advan. Enzyme Regul., 48: 152-164;Alegre et al, 1992 J Immunol, 148: 3461-3468; and amino acid modifications reviewed in Kaneko and Niwa, 2011 Biodrugs, 25(1):1-11, the contents of each of which are incorporated herein by reference in their entirety.

[0179] In some embodiments, an Fc region, such as a human IgG1 Fc region, is modified to enhance ADCC or CDC activity. Examples of mutations that enhance ADCC include modifications at Ser239 and Ile332, such as Ser239Asp and Ile332Glu (S239D, I332E). Examples of mutations that enhance CDC include modifications at Lys326 and Glu333. In some embodiments, the Fc region is modified at one or both of these positions using the Kabat numbering system (e.g., Lys326Ala and / or Glu333Ala (K326A and E333A)).

[0180] In some embodiments, the Fc region is modified to provide reduced Fc-mediated effector function, e.g., through reduced Fc receptor binding, e.g., FcγR binding rather than the usual FcRn binding. In some embodiments, the human IgG1 Fc region fusion protein of the present disclosure lacks or has reduced fucose attached to the N-linked glycan chain at N297. There are numerous ways to prevent fucosylation, including, but not limited to, production in a FUT8-deficient cell line; addition of inhibitors, e.g., castanospermine, to mammalian cell culture medium; and metabolic engineering of the production cell line. In some embodiments, the human IgG1 Fc region is modified at amino acid Asn297 (boxed, Kabat numbering) (e.g., Asn297Ala (N297A) or Asn297Asp (N297D)) to prevent glycosylation of the fusion protein.

[0181] In some embodiments, the Fc region of the fusion protein is altered at one or more of the following positions to reduce Fc receptor binding: Leu234 (L234), Leu235 (L235), Asp265 (D265), Asp270 (D270), Ser298 (S298), Asn297 (N297), Asn325 (N325), or Ala327 (A327). For example, Leu234Ala (L234A), Leu235Ala (L235A), Asp265Asn (D265N), Asp270Asn (D270N), Ser298Asn (S298N), Asn297Ala (N297A), Asn325Glu (N325E), or Ala327Ser (A327S). In some embodiments, the Fc region of the fusion protein is modified at amino acid Leu235 (within the box of SEQ ID NO:1, Kabat numbering) to alter Fc-receptor interactions (e.g., Leu235Glu (L235E) or Leu235Ala (L235A)). In some embodiments, the Fc region of the fusion protein is modified at amino acid Leu234 (in the box of SEQ ID NO:1, Kabat numbering) to alter Fc receptor interactions (e.g., Leu234Ala (L234A)). In some embodiments, the Fc region of the fusion protein is altered at both amino acids 234 and 235 (e.g., Leu234Ala and Leu235Ala (L234A / L235A) or Leu234Val and Leu235Ala (L234V / L235A)). In preferred embodiments, modifications within the Fc region reduce binding to the Fc receptor gamma receptor but have minimal effect on binding to the neonatal Fc receptor (FcRn).

[0182] In some embodiments, the human IgG Fc region is modified to enhance FcRn binding. Examples of Fc mutations that enhance FcRn binding are Met252Tyr, Ser254Thr, Thr256Glu (M252Y, S254T, T256E, respectively) (Kabat numbering, Dall'Acqua et al. 2006, J. Biol Chem Vol. 281(33)23514-23524), Met428Leu and Asn434Ser (M428L, N434S) (Zalevsky et al. 2010 Nature Biotech, Vol. 28(2)157-159) (Kabat et al. 1991 EU Index of Sequences of Proteins of Immunological Interest). In some embodiments, the variant or modified Fc polypeptide comprises the following mutations, using the Kabat numbering system: Met252Tyr and Met428Leu or Met252Tyr and Met428Val (M252Y, M428L or M252Y, M428V).

[0183] In some embodiments, the Fc region of the fusion protein lacks amino acids at one or more of the following positions to reduce Fc receptor binding: Glu233 (E233), Leu234 (L234), or Leu235 (L235). In these embodiments, the Fc deletion of these three amino acids reduces complement protein C1q binding. TIFF0007651450000016.tif30128

[0184] In some embodiments, the Fc region is mutated at one or more of the following positions to reduce Fc receptor binding: Glu233 (E233), Leu234 (L234), or Leu235 (L235). The one or more mutations may include E233P, L234V, and / or L235A.

[0185] In some embodiments, the Fc region of the fusion protein is altered at Gly236 (within the box of SEQ ID NO:1) to reduce Fc receptor binding. For example, Gly236 is deleted from the fusion protein. In some embodiments, a human IgG1 Fc region is modified at amino acid Gly236 (e.g., Gly236Ala (G236A)) to enhance interaction with CD32A.

[0186] In specific embodiments, mutations in the Fc region to reduce Fc effector function, e.g., via reduced binding of the Fc receptor to FcγR, include any of the following mutations: G236R / L328R, E233P / L234V / L235A / G236del / S239K, E233P / L234V / L235A / G236del / S267K, E233P / L234V / L235A / G236del / S239K / A327G, E233P / L234V / L235A / G236del / S267K / A327G, or E233P / L234V / L235A / G236del.

[0187] In some embodiments, the human IgG1 Fc region lacks Lys447 (Kabat et al 1991 EU Index of Sequences of Proteins of Immunological Interest).

[0188] In some embodiments, the fusion or immunologically active fragment thereof comprises a human IgG2 polypeptide sequence that is at least 50%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of SEQ ID NO:2.

[0189] In some embodiments, the immunoglobulin Fc region of the fusion protein, or an immunologically active fragment thereof, is of the human IgG2 isotype, having the following amino acid sequence: TIFF0007651450000017.tif30128

[0190] In some embodiments, the fusion or immunologically active fragment thereof comprises a human IgG2 polypeptide sequence that is at least 50%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of SEQ ID NO:3.

[0191] In some embodiments, the human IgG2 Fc region is modified at amino acid Asn297 (intrabox, to prevent glycosylation of the antibody, e.g., Asn297Ala (N297A) or Asn297Asp (N297D)). In some embodiments, the human IgG2 Fc region lacks Lys447 (Kabat et al 1991 Sequences of Proteins of Immunological Interest, EU Index).

[0192] In some embodiments, the immunoglobulin Fc region or immunologically active fragment of the fusion protein is of the human IgG3 isotype, having the following amino acid sequence: TIFF0007651450000018.tif31128

[0193] In some embodiments, the antibody or immunologically active fragment thereof comprises a human IgG3 polypeptide sequence that is at least 50%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of SEQ ID NO:4.

[0194] In some embodiments, the human IgG3 Fc region is modified at amino acid Asn297 (boxed, Kabat numbering) to prevent glycosylation of the antibody (e.g., Asn297Ala (N297A) or Asn297Asp (N297D)). In some embodiments, the human IgG3 Fc region is modified at amino acid 435 (e.g., Arg435His (R435H)) to extend half-life. In some embodiments, the human IgG3 Fc region lacks Lys447 (Kabat et al 1991 Sequences of Proteins of Immunological Interest, EU Index).

[0195] In some embodiments, the immunoglobulin Fc region or immunologically active fragment of the fusion protein is of the human IgG4 isotype having the following amino acid sequence: TIFF0007651450000019.tif31128

[0196] In some embodiments, the antibody or immunologically active fragment thereof comprises a human IgG4 polypeptide sequence that is at least 50%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of SEQ ID NO:5.

[0197] In some embodiments, the immunoglobulin Fc region or immunologically active fragment of the fusion protein is of the human IgG4 isotype, having the following amino acid sequence: TIFF0007651450000020.tif30128

[0198] In some embodiments, the antibody or immunologically active fragment thereof comprises a human IgG4 polypeptide sequence that is at least 50%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of SEQ ID NO:6.

[0199] In other embodiments, the human IgG4 Fc region is modified at amino acid 235 (e.g., Leu235Glu(L235E)) to alter Fc-receptor interactions. In some embodiments, the human IgG4 Fc region is modified at amino acid Asn297 (boxed, Kabat numbering) to prevent glycosylation of the antibody (e.g., Asn297Ala (N297A) or Asn297Asp (N297D)). In some embodiments, the human IgG4 Fc region lacks Lys447 (Kabat et al 1991 Sequences of Proteins of Immunological Interest EU Index).

[0200] In some embodiments, the human IgG Fc region is modified to stabilize homodimerization at the CH3:CH3 interface by introducing two disulfide bonds by changing Ser354 to Cys (S354C) and Tyr349 to Cys (Y349C) (S354C / Y349C).

[0201] In certain embodiments of the multispecific polypeptide constructs provided herein, the human IgG Fc region is modified to induce heterodimerization. Various methods are known for promoting heterodimerization of complementary Fc polypeptides. For example, Ridgway et al, Protein Eng. 9:617-621(1996);Merchant et al, Nat. Biotechnol. 16(7):677-81(1998);Moore et al. (2011)MAbs,3:546-57;Von Kreudenstein et al. MAbs, (2013)5: 646-54;Gunasekaran et al. (2010) J. Biol. Chem., 285: 19637-46;Leaver-Fay et al. (2016) Structure, 24:641-51;Ha et al. (2016)Frontiers in Immunology,7:1;Davis et al. (2010) Protein Eng Des Sel, 23: 195-202; Published International PCT Application No. WO 1998 / 050431, WO2009 / 089004, WO2011143545, WO2014 / 067011, WO2012 / 058768, WO2018027025; published U.S. patent application numbers US20140363426, US20150307628, US20180016354, US20150239991; and U.S. Patent Nos. 5,731,168, 7,183,076, 9,701,759, 9,605,084, and 9,650,446. Methods for promoting heterodimerization of the Fc chain include mutagenesis of the Fc region, for example, by including a set of "knob-into-hole" mutations or mutations to achieve electrostatic steering of the Fc to favor attractive interactions between different polypeptide chains.For example, in some embodiments, heterodimeric Fc polypeptides contain mutations to alter the charge polarity at the Fc dimer interface so that coexpression of electrostatically matched Fc chains favors favorable attractive interactions, thereby promoting the formation of desired Fc heterodimers, and unfavorable repulsive charge interactions suppress the formation of undesired Fc homodimers (Guneskaran et al., (2010) JBC, 285:19637-19646). When coexpressed in cells, interchain association is possible, but due to charge repulsion, the chains do not substantially self-associate. Another strategy for generating heterodimeric Fc involves mixing CH3 domain segments from human IgG and IgA to create heterodimers of complementary CH3s, referred to as SEED Fc.

[0202] Heterodimerization methods and variants also include those described in published International PCT application WO2014 / 145806, including "knobs and holes" mutations (also called "skew" variants), mutations involving "electrostatic steering" or "charge pairing," and pI variants. Heterodimer variants also include any described in published U.S. application numbers US2012 / 0149876 or US2018 / 011883.

[0203] In some embodiments, both polypeptides of the Fc heterodimer contain paired or complementary amino acid modifications to promote heterodimerization. Exemplary pairs of amino acid modifications for polypeptides of Fc fusions are shown in Table 1.

[0204] Table 1. Amino acid pairs of heterodimeric Fc TIFF0007651450000021.tif47133

[0205] In some embodiments, the modification comprises introducing a protuberance (knob) into a first Fc polypeptide and a cavity (hole) into a second Fc polypeptide, such that the protuberance is positioned in the cavity to facilitate complexation of the first and second Fc-containing polypeptides. The amino acids targeted for substitution and / or modification to create the protuberance or cavity in the polypeptide are typically interface amino acids that interact with or contact one or more amino acids at the interface of the second polypeptide.

[0206] In some embodiments, the first Fc polypeptide modified to contain a protuberance (hole) amino acid comprises a substitution of a native or initial amino acid with an amino acid having at least one side chain that protrudes from the interface of the first Fc polypeptide and thus is positioned in a compensatory cavity (hole) in the adjacent interface of the second polypeptide. The replacement amino acid most often has a side chain volume larger than that of the initial amino acid residue. Methods for determining and / or evaluating the properties of amino acid residues to identify ideal replacement amino acids for creating a protuberance are known to those skilled in the art. In some embodiments, the replacement residue for forming the protuberance is a naturally occurring amino acid residue, including, for example, arginine (R), phenylalanine (F), tyrosine (Y), or tryptophan (W). In some examples, the initial residue identified for substitution is an amino acid residue with a small side chain, such as, for example, alanine, asparagine, aspartic acid, glycine, serine, threonine, or valine.

[0207] In some embodiments, the second Fc polypeptide to be modified to contain a cavity (hole) is recessed from the interface of the second polypeptide and thus includes a substitution of the native or initial amino acid with an amino acid having at least one side chain that can accommodate a corresponding protrusion from the interface of the first polypeptide. The substituted amino acid most often has a smaller side chain volume than the initial amino acid residue. Methods for determining and / or evaluating the characteristics of amino acid residues to identify ideal replacement residues for cavity formation are known to those skilled in the art. Generally, replacement residues for cavity formation are naturally occurring amino acids, including, for example, alanine (A), serine (S), threonine (T), and valine (V). In some examples, the initial amino acid identified for substitution is an amino acid with a large side chain, such as, for example, tyrosine, arginine, phenylalanine, or tryptophan.

[0208] The CH3 interface of human IgG1, for example, comprises 16 residues from each domain located on four antiparallel β-strands buried 1090 Å from each surface (see, for example, Deisenhofer et al. (1981) Biochemistry, 20:2361-2370; Miller et al., (1990) J Mol. Biol., 216, 965-973; Ridgway et al., (1996) Prot. Engin., 9:617-621; U.S. Patent No. 5,731,168). Modifications of the CH3 domain to create protuberances or cavities are described, for example, in U.S. Patent No. 5,731,168; International Patent Applications WO98 / 50431 and WO2005 / 063816; and Ridgway et al., (1996) Prot. Engin. 617-621. In some instances, modifications of the CH3 domain to create a bulge or cavity are typically targeted to residues located in the two central antiparallel β-strands, with the goal being to minimize the risk that the resulting bulge will be accommodated by protruding into the surrounding solvent rather than being accommodated in a compensatory cavity in the partner CH3 domain.

[0209] For example, in some embodiments, the heterodimeric Fc comprises a polypeptide having an amino acid modification at Thr366 in a CH3 domain, which, when replaced with a bulkier amino acid, e.g., Try (T366W), can preferentially pair with a second CH3 domain having an amino acid modification at positions Thr366, Leu368, and Tyr407 to a less bulky amino acid, e.g., Ser, Ala, or Val (T366S / L368A / Y407V). Heterodimerization via CH3 modifications can be further stabilized by the introduction of disulfide bonds, e.g., by changing Ser354 to Cys (S354C) and Tyr349 to Cys (Y349C) in the opposing CH3 domain (reviewed in Carter, 2001 Journal of Immunological Methods, 248:7-15).

[0210] In specific embodiments, the multispecific polypeptide construct contains a first and a second Fc capable of mediating Fc heterodimerization, wherein the first Fc polypeptide contains the mutations T366W and S354C, and the second Fc polypeptide contains the mutations T366S, L368A, Y407V, and Y349C. In some embodiments, the first Fc polypeptide is selected from an Fc polypeptide comprising the sequence set forth in SEQ ID NO:291 or 297, and the second Fc polypeptide is selected from an Fc polypeptide comprising the sequence set forth in SEQ ID NO:292, 295, or 299. In some embodiments, the first Fc polypeptide is or comprises the sequence of amino acids set forth in any of SEQ ID NOs:82, 86, 94, or 96, and the second Fc polypeptide is or comprises the sequence of amino acids set forth in any of SEQ ID NOs:83, 87, 90, 92, 98, or 100.

[0211] In some embodiments, the Fc polypeptide exhibits features that provide Fc-mediated effector functions. In a specific example, the first Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:291, and the second Fc polypeptide is or comprises SEQ ID NO:292 or 295. In some embodiments, the first Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:82, and the second Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:83 or 90. In some embodiments, the first Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:86, and the second Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:87 or 92. The first and second Fc polypeptides may be formatted in either polypeptide chain of the construct.

[0212] In some embodiments, one or both of the first and second Fc polypeptides may further comprise one or more amino acid mutations to further reduce one or more Fc effector functions, such as reduced Fc receptor binding. Exemplary mutations for reducing Fc effector function include any of those described. In some embodiments, the modification may be a deletion of one or more positions Glu233 (E233), Leu234 (L234), or Leu235 (L235), such as a deletion of Glu233 (E233), Leu234 (L234), and Leu235 (L235). In some embodiments, the first Fc polypeptide is selected from an Fc polypeptide comprising a sequence set forth in SEQ ID NO: 293 or 298, and the second Fc polypeptide is selected from an Fc polypeptide comprising a sequence set forth in SEQ ID NO: 294, 296, or 300. In some embodiments, the first Fc polypeptide is or comprises the sequence of amino acids set forth in any of SEQ ID NOs:84, 88, 95, or 97, and the second Fc polypeptide is or comprises the sequence of amino acids set forth in any of SEQ ID NOs:85, 89, 91, 93, 99, or 101.

[0213] In a specific example, the first Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:293, and the second Fc polypeptide is or comprises SEQ ID NO:294 or 296. In some embodiments, the first Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:84, and the second Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:85 or 91. In some embodiments, the first Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:88, and the second Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:89 or 93. The first and second Fc polypeptides may be formatted in either polypeptide chain of the construct.

[0214] In some embodiments, the first Fc polypeptide or the second Fc polypeptide further comprises the mutation M252Y and / or M428V. In a specific example, the first Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:297, and the second Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:299. In some embodiments, the first Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:94, and the second Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:98. In some embodiments, the first Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:96, and the second Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:100. In other examples, the first Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:298, and the second Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:300. In some embodiments, the first Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:95, and the second Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:99. In some embodiments, the first Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:97, and the second Fc polypeptide is or comprises the sequence set forth in SEQ ID NO:101. The first and second Fc polypeptides may be formatted in either polypeptide chain of the construct.

[0215] Additional examples of variants that can facilitate the promotion of heterodimers include S364K / E357Q and L368D / K370S; L368D / K370S and S364K; L368E / K370S and S364K; T411T / E360E / Q362E and D401K; L368D / K370S and S364K / E357L, K370S and S364K / E357Q and and T366S / L368A / Y407V and T366W, or 366S / L368A / Y407V / Y349C and T366W / S354C (each pair representing mutations in the first and second Fc polypeptides). In a specific embodiment, the provided constructs contain first and second Fc polypeptides containing the mutation pairs L368D / K370S and S364K and E357Q.

[0216] An additional mechanism that can be used in generating heterodimers is sometimes referred to as "electrostatic steering," as described in Gunasekaran et al., J. Biol. Chem. 285 (25): 19637 (2010). This is sometimes referred to herein as "charge pairing." In this embodiment, electrostatics are used to bias formation toward heterodimerization. As one skilled in the art will appreciate, these may also have an effect on pI, and therefore purification, and therefore, in some cases, may also be considered pI variants. However, because these were generated to force heterodimerization and not used as purification tools, they are classified as "steric variants." In one embodiment, the first Fc polypeptide may contain the mutations D221E / P228E / L368E, and the second Fc polypeptide may contain the mutations D221R / P228R / K409R. In another embodiment, the first Fc polypeptide may contain the mutations C220E / P228E / 368E and the second Fc polypeptide may contain the mutations C220R / E224R / P228R / K409R.

[0217] In some embodiments, heterodimerization can be facilitated by pI variants. In some aspects, pI variants can include those that increase the pI of the protein (basic changes). In other aspects, pI variants can include those that decrease the pI of the protein (acidic changes). In some cases, all combinations of these variants can be made, including combinations in which one Fc polypeptide can be wild-type or a variant that does not exhibit a pI significantly different from wild-type, and the other Fc polypeptide can be more basic or more acidic. Alternatively, one Fc polypeptide can be made more basic and the other more acidic. In some embodiments, at least one Fc polypeptide is a negative pI variant Fc containing the mutations Q295E / N384D / Q418E / N421D.

[0218] In some embodiments, a combination of steric heterodimerization variants (eg, knobs and holes) and pI variants or charge pair variants may be used.

[0219] In specific embodiments, the provided constructs contain (a) a first Fc polypeptide comprising the scubariant S364K / E357Q; and (b) a second Fc polypeptide containing the scubariant L368D / K370S and pI variants N208D / Q295E / N384D / Q418E / N421D. In some embodiments, one or both of the first and second polypeptides may contain additional mutations to reduce Fc effector activity, such as the exemplary mutations E233P / L234V / L235A / G236del / S267K. Examples of such first and second Fc polypeptides capable of mediating Fc heterodimerization include the sequences set forth in SEQ ID NOs: 285 and 286. The first and second Fc polypeptides may be formatted in either polypeptide chain of the construct.

[0220] The resulting multispecific polypeptide construct can be purified by an appropriate method, such as affinity chromatography on a Protein A or Protein G column. When two nucleic acid molecules encoding different polypeptides are transformed into a cell, homodimers and heterodimers will form. Expression conditions can be adjusted so that heterodimer formation is favored over homodimer formation.

[0221] Techniques for recovering heterodimers from homodimers based on the differential affinity of the heterodimers for affinity reagents are known. In some aspects, such techniques involve designing heterodimers such that one of the Fc polypeptide chains does not bind to the affinity reagent Protein A. In some cases, one of the polypeptide chains may contain one or more amino acid substitutions to eliminate or reduce the affinity of one of the polypeptides of the Fc heterodimer for the Protein A reagent. See, e.g., WO2017134440, WO2010151792, and Jendeberg et al. (Jendeberg et al., (1997) J. Immunol. Methods, 201(1): 25-34). In some of these embodiments, the Fc region may be modified in the Protein A binding site of one member of the heterodimer to prevent Protein A binding, thereby allowing for more efficient purification of the heterodimeric fusion protein. An exemplary modification at this binding site is Ile253, e.g., Ile253Arg (I253R). In some embodiments, the modification can be H435R or H435R / Y436F. In some embodiments, the Fc polypeptide of the Fc heterodimer can contain a modification (pA+ / pG-) that allows it to bind to Protein A but not Protein G. Exemplary pA+ / pG- amino acid modifications include, for human IgG1, an Fc containing serine at position 428, serine at position 434, and optionally, histidine at position 436, or those residues at the corresponding positions in human IgG2, 3, or 4. In some aspects, such amino acid modifications at positions 428, 434, and optionally, at position 436 of an IgG Fc polypeptide reduce or prevent Protein G binding and enhance protein purification.

[0222] In some embodiments, any of these modifications that confer differential affinity to affinity reagents can be combined with one or more of the other amino acid modifications described above. For example, the I253R modification can be combined with either the T366S / L368A / Y407V or T366W modification. The T366S / L368A / Y407V modified Fc can form homodimers because there is no steric hindrance at the dimerization interface that exists in the case of the T366W modified Fc. Thus, in some embodiments, the I253R modification is combined with the T366S / L368A / Y407V modified Fc to disallow purification of any homodimeric Fc that may have formed. A similar modification can be utilized by combining T366S / L368A / Y407V with H453R.

[0223] In some embodiments, the Fc region of the heterodimeric molecule may further comprise one or more other Fc mutations, such as any of those described above, hi some embodiments, the heterodimeric molecule comprises an Fc region comprising a mutation that reduces effector function.

[0224] In some embodiments, one Fc polypeptide of the heterodimeric Fc comprises the amino acid sequence set forth in either SEQ ID NO:291 (e.g., SEQ ID NO:82 or 86) or 297 (e.g., SEQ ID NO:94 or 96), and the other Fc polypeptide of the heterodimeric Fc contains the amino acid sequence set forth in either SEQ ID NO:201 (e.g., SEQ ID NO:83 or 87), 295 (e.g., SEQ ID NO:90 or 92), or 299 (e.g., SEQ ID NO:98 or 100). In some embodiments, one Fc polypeptide of the heterodimeric Fc comprises the amino acid sequence set forth in either SEQ ID NO:293 (e.g., SEQ ID NO:84 or 88) or 298 (e.g., SEQ ID NO:95 or 97), and the other Fc polypeptide of the heterodimeric Fc comprises the amino acid sequence set forth in either SEQ ID NO:294 (e.g., SEQ ID NO:85 or 89), 296 (e.g., SEQ ID NO:91 or 93), or 300 (e.g., SEQ ID NO:99 or 101).

[0225] In some embodiments, the human IgG Fc region is modified to prevent dimerization. In these embodiments, the fusion protein of the present disclosure is a monomer. For example, modification of residue Thr366 to a charged residue, such as Thr366Lys, Thr366Arg, Thr366Asp, or Thr366Glu (T366K, T366R, T366D, or T366E, respectively), prevents CH3-CH3 dimerization.

[0226] In some embodiments, the Fc region of the fusion protein is altered at one or more of the following positions to reduce Fc receptor binding: Leu234 (L234), Leu235 (L235), Asp265 (D265), Asp270 (D270), Ser298 (S298), Asn297 (N297), Asn325 (N325), or Ala327 (A327). For example, Leu234Ala (L234A), Leu235Ala (L235A), Asp265Asn (D265N), Asp270Asn (D270N), Ser298Asn (S298N), Asn297Ala (N297A), Asn325Glu (N325E), or Ala327Ser (A327S). In preferred embodiments, modifications within the Fc region reduce binding to the Fc receptor gamma receptor but have minimal effect on binding to the neonatal Fc receptor (FcRn).

[0227] In some embodiments, the fusion protein contains a polypeptide derived from an immunoglobulin hinge region. The hinge region can be selected from any of the human IgG subclasses. For example, the fusion protein can contain a modified IgG1 hinge having the sequence EPKSSDKTHTCPPC (SEQ ID NO:7), in which Cys220, which forms a disulfide with the C-terminal cysteine ​​of the light chain, is mutated to serine (e.g., Cys220Ser (C220S)). In other embodiments, the fusion protein contains a shortened hinge having the sequence DKTHTCPPC (SEQ ID NO:8).

[0228] In some embodiments, the fusion protein has a modified hinge derived from IgG4 modified to prevent or reduce strand exchange (e.g., Ser228Pro (S228P)), having the sequence ESKYGPPCPPC (SEQ ID NO:9). In some embodiments, the fusion protein contains a linker polypeptide. In other embodiments, the fusion protein contains a linker and hinge polypeptide.

[0229] 3. Linker The provided multispecific polypeptide constructs contain a linker joining or coupling a first component containing an immunoglobulin Fc region to a second component containing a CD3-binding region. In some embodiments, the linker is positioned at the end of the C-terminal region of the Fc region such that the Fc region is N-terminal to the CD3-binding region. Because the provided multispecific polypeptide constructs are multimers, such as dimers, containing a first and second polypeptide that together form the first and second components, the provided constructs contain a linker joining the Fc portion and CD3-binding region of the first polypeptide and a linker joining the Fc portion and CD3-binding region of the second polypeptide. In some embodiments, the first polypeptide comprises a first Fc polypeptide, a linker, and a first domain (e.g., VH) of the CD3-binding region of the heterodimeric Fc region, and the second polypeptide comprises a second Fc polypeptide, a linker, and a second domain (e.g., VL) of the CD3-binding region of the heterodimeric Fc region. Typically, the linkers present in the first and second polypeptides of the multispecific polypeptide construct are the same. Thus, in some embodiments, each CD3 binding domain is linked to the opposing polypeptide of an Fc, such as a heterodimeric Fc, via a linker, such as the same linker.

[0230] A variety of polypeptide linkers for use in fusion proteins are known (see, e.g., Chen et al. (2013) Adv. Drug. Deliv. 65:1357-1369; and International PCT Publication Nos. WO2014 / 099997, WO2000 / 24884; U.S. Patent No. 5,258,498; U.S. Patent No. 5,525,491; U.S. Patent No. 5,525,491, U.S. Patent No. 6,132,992).

[0231] In some embodiments, the linker is selected so that, when the CD3-binding region is conjugated to the Fc region of the multispecific polypeptide conjugate, the CD3-binding region is constrained and is unable or substantially unable to bind or engage with CD3 on the surface of cells, e.g., T cells, when the multispecific polypeptide construct contacts the cells. Various assays can be used to evaluate CD3 binding or engagement by the multispecific polypeptide construct, including assays to assess T cell binding, NFAT activation using reporter systems, cytolytic T cell activity, cytokine production, and / or expression of T cell activation markers. Exemplary assays are shown in the provided examples. Typically, the linker ensures correct folding of the polypeptide construct, does not exhibit charges inconsistent with the activity or function of the linked polypeptide, and does not form bonds or other interactions with amino acid residues in one or more of the domains that would weaken or alter the activity of the linked polypeptide. In some embodiments, the linker is a polypeptide linker. The polypeptide linker can be a flexible linker or a non-flexible linker, or a combination of both.

[0232] In some aspects, the linker is a short, medium, or long linker. In some embodiments, the linker is up to 40 amino acids in length. In some embodiments, it is up to 25 amino acids in length. In some embodiments, the linker is at least about 2 amino acids in length. In some aspects, a suitable length is, for example, at least 1 amino acid residue, typically less than about 40 amino acid residues, e.g., 2 to 25 amino acid residues, 5 to 20 amino acid residues, 5 to 15 amino acid residues, or 8 to 12 amino acids in length. In some embodiments, the linker is about 2 to 24 amino acids, 2 to 20 amino acids, 2 to 18 amino acids, 2 to 14 amino acids, 2 to 12 amino acids, 2 to 10 amino acids, 2 to 8 amino acids, 2 to 6 amino acids, 6 to 24 amino acids, 6 to 20 amino acids, 6 to 18 amino acids, 6 to 14 amino acids, 6 to 12 amino acids, 6 to 10 amino acids, 6 to 8 amino acids, 8 to 24 amino acids, 8 to 20 amino acids, 8 to 18 amino acids, 8 to 14 amino acids, 8 to 12 amino acids, 8 to 10 amino acids, 10 to 24 amino acids, 10 to 20 amino acids, 10 to 18 amino acids, 10 to 14 amino acids, 10 to 12 amino acids, 12 to 24 amino acids, 12 to 20 amino acids, 12 to 18 amino acids, 12 to 14 amino acids, 14 to 24 amino acids, 14 to 20 amino acids, 14 to 18 amino acids, 18 to 24 amino acids, 18 to 20 amino acids, or 20 to 24 amino acids. In some embodiments, the linker is 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acids in length.

[0233] In certain aspects, the longer the linker length, the greater the CD3 binding when the multispecific polypeptide conjugate is bound to its antigen, e.g., a TAA. Thus, in some aspects, the linker is greater than 12 amino acids in length, e.g., greater than 13, 14, 15, 16, 17, or 18 amino acids in length. In some embodiments, the linker is 12-40 amino acids in length, 12-30 amino acids, 12-24 amino acids, 12-18 amino acids, 12-15 amino acids, 15-40 amino acids, 15-30 amino acids, 15-24 amino acids, 15-18 amino acids, 18-40 amino acids, 18-30 amino acids, 18-24 amino acids, 24-40 amino acids, 24-30 amino acids, or 30-40 amino acids.

[0234] The linker may be naturally occurring, synthetic, or a combination of both. Particularly suitable linker polypeptides primarily comprise amino acid residues selected from glycine (Gly), serine (Ser), alanine (Ala), and threonine (Thr). For example, the linker may contain at least 75%, e.g., at least 80%, at least 85%, or at least 90% (calculated based on the total number of residues present in the peptide linker) of amino acid residues selected from Gly, Ser, Ala, and Thr. The linker may consist exclusively of Gly, Ser, Ala, and / or Thr residues. In some embodiments, the linker contains 1-25 glycine residues, 5-20 glycine residues, 5-15 glycine residues, or 8-12 glycine residues. In some aspects, suitable peptide linkers typically contain at least 50% glycine residues, e.g., at least 75% glycine residues. In some embodiments, the peptide linker comprises only glycine residues. In some embodiments, the peptide linker comprises only glycine and serine residues.

[0235] In some embodiments, these linkers are composed primarily of the amino acids glycine and serine, and are referred to herein as GS linkers. In some embodiments, the linker contains (GGS)n (n is 1-10, e.g., 1-5, e.g., 1-3), e.g., GGS(GGS)n (SEQ ID NO:171) (n is 0-10). In a specific embodiment, the linker contains the sequence (GGGGS)n (SEQ ID NO:173) (n is 1-10 or n is 1-5, e.g., 1-3). In a further embodiment, the linker contains (GGGGGS)n (SEQ ID NO:172) (n is 1-4, e.g., 1-3). The linker may include any combination of the above, e.g., 2, 3, 4, or 5 repeats of the GS linker, GGS linker, GGGGS linker, and / or GGGGGS linker may be combined. In some embodiments, such linkers are 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, or 19 amino acids in length.

[0236] In some embodiments, the linker is as follows (single letter amino acid code): GGS, GGGGS (SEQ ID NO: 149), or GGGGGS (SEQ ID NO: 135). In some embodiments, the GS linker is TIFF0007651450000022.tif37144. In some embodiments, the linker is GGGG (SEQ ID NO:103). In some embodiments, the linker is GGGGG (SEQ ID NO:250). In some of the foregoing examples, serine may be substituted with alanine (e.g., (Gly4Ala) or (Gly3Ala)).

[0237] In some embodiments, the linker comprises the amino acid sequence Gly x -Xaa-Gly y -Xaa-Gly z(SEQ ID NO:174) wherein each Xaa is independently selected from alanine (Ala), valine (Val), leucine (Leu), isoleucine (Ile), methionine (Met), phenylalanine (Phe), tryptophan (Trp), proline (Pro), glycine (Gly), serine (Ser), threonine (Thr), cysteine ​​(Cys), tyrosine (Tyr), asparagine (Asn), glutamine (Gln), lysine (Lys), arginine (Arg), histidine (His), aspartic acid (Asp), and glutamic acid (Glu); and wherein x, y, and z are each integers ranging from 1 to 5. In some embodiments, each Xaa is independently selected from the group consisting of Ser, Ala, and Thr. In a particular variation, x, y, and z each equal 3, thus providing a peptide linker having the amino acid sequence Gly-Gly-Gly-Xaa-Gly-Gly-Gly-Xaa-Gly-Gly-Gly (SEQ ID NO:175), where each Xaa is selected as described above.

[0238] In some embodiments, the linker is a serine-rich linker based on repeating (SSSSG)n (SEQ ID NO:185) motifs, where n is at least 1, but can also be 2, 3, 4, 5, 6, 7, 8, and 9.

[0239] In some cases, it may be desirable to provide a peptide linker with a degree of immobility. This can be achieved by including proline residues in the amino acid sequence of the peptide linker. Thus, in some embodiments, the linker includes at least one proline residue in the amino acid sequence of the peptide linker. For example, the peptide linker may have an amino acid sequence in which at least 25% (e.g., at least 50% or at least 75%) of the amino acid residues are proline residues. In one specific embodiment, the peptide linker includes only proline residues.

[0240] In some aspects, the peptide linker comprises at least one cysteine ​​residue, e.g., one cysteine ​​residue. For example, in some embodiments, the linker comprises at least one cysteine ​​residue and an amino acid residue selected from the group consisting of Gly, Ser, Ala, and Thr. In some such embodiments, the linker comprises only glycine and cysteine ​​residues, e.g., only glycine and cysteine ​​residues. Typically, only one cysteine ​​residue is included in each peptide linker. An example of a specific linker containing a cysteine ​​residue is the amino acid sequence Gly m -Cys-Gly n (n and m are each an integer from 1 to 12, e.g., 3 to 9, 4 to 8, or 4 to 7.) In a particular variation, such a peptide linker has the amino acid sequence GGGGG-C-GGGGG (SEQ ID NO:177).

[0241] In some embodiments, the linker of the fusion protein is a structured or constrained linker. In specific embodiments, the structured linker contains the sequence (AP)n or (EAAAK)n (SEQ ID NO:178) (n is 2 to 20, preferably 4 to 10), such as, but not limited to, AS-(AP)n-GT (SEQ ID NO:179) or AS-(EAAAK)n-GT (SEQ ID NO:180) (n is 2 to 20, e.g., 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15). In other embodiments, the linker contains the sequence TIFF0007651450000023.tif17144 (n is 2 to 20). In some embodiments, the linker TIFF0007651450000024.tif4166. In some embodiments, such linkers, due to their structure, may be more resistant to proteolytic cleavage and therefore may offer advantages when injected in vivo.

[0242] In some embodiments, the linker may be referred to as a non-cleavable linker rather than a cleavable linker. In some embodiments, the linker is not cleavable by a protease. In some embodiments, a linker that is not cleavable or not cleavable by a protease is generally stable for in vivo delivery or recombinant production. In some aspects, a linker that is not cleavable by a protease includes one that does not contain at least one peptide bond that is preferably located within a cleavable peptide sequence or protease recognition site. In specific embodiments, a non-cleavable linker is not a target substrate of a protease, and therefore is not preferentially or specifically cleaved by a protease compared to a linker that contains a substrate recognition site for the same protease.

[0243] In some embodiments, the linker does not contain a substrate recognition site or cleavage site for a specific protease, which is a sequence recognized by the active site of the protease and cleaved by the protease. Typically, for example, the cleavage sequence for a serine protease is composed of the P1-P4 and P1'-P4' amino acids in the substrate, with cleavage occurring after the P1 position. Typically, the cleavage sequence for a serine protease is six residues long to match the extended substrate specificity of many proteases, but may be longer or shorter depending on the protease. Typically, the linker does not contain a releasable P1-P1' bond sequence recognized by the protease. In some aspects, a non-cleavable linker, or a linker that does not contain a substrate recognition site specifically recognized for cleavage by the protease, is cleaved by the protease substantially less than the cleavage of the target substrate of the protease.

[0244] In some embodiments, the linker is a cleavable linker. In some aspects, the cleavable linker is a linker such as any of those described above, further comprising a sequence that is a substrate for a protease due to the presence of at least one bond that can be degraded under physiological conditions. In some cases, the cleavable linker is susceptible to cleavage or susceptible to cleavage under certain conditions present in vivo, for example, after exposure to extracellular proteases, including those present in the cellular environment in vivo. In some cases, the proteases may be present in certain physiological microenvironments, such as tumor microenvironments, thereby limiting the sites at which cleavage can occur.

[0245] Proteases typically exhibit specificity or preference for cleaving a particular target substrate compared to other non-target substrates. The degree of such specificity can be determined based on the rate constant of cleavage of a sequence, e.g., a linker, which is a measure of the preference of the protease for the substrate and the efficiency of the enzyme. A method of determining the rate of increase in cleavage over time in the presence of ...

Claims

1. 1. A multispecific polypeptide construct comprising a first component comprising an immunoglobulin Fc region and a second component comprising a CD3 binding region, the CD3 binding region is a disulfide-stabilized anti-CD3 binding Fv fragment (dsFv) comprising a variable heavy chain region (VH) and a variable light chain region (VL); the Fc region is a heterodimeric Fc region, and the VH and VL of the anti-CD3 dsFv are linked to opposing polypeptides of the heterodimeric Fc; the first and second components are coupled by a polypeptide linker that is 3-18 amino acids in length and is composed of at least 80% glycine or serine residues, and the Fc region is positioned N-terminal to the CD3 binding region; one or both of the first and second components comprises at least one antigen binding domain that binds to a tumor associated antigen (TAA), each of the at least one antigen binding domain being an sdAb; The second component comprises at least one costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor, and each of the at least one CRBR is an sdAb; Multispecific polypeptide constructs.

2. 2. The multispecific polypeptide construct of claim 1, wherein at least one antigen binding domain is positioned amino-terminal to the Fc region and / or carboxy-terminal to the CD3 binding region of the multispecific polypeptide construct.

3. 3. The multispecific polypeptide construct of claim 1 or claim 2, wherein at least one costimulatory receptor binding region (CRBR) is positioned amino-terminal to the Fc region and carboxy-terminal to the CD3 binding region of the multispecific polypeptide construct.

4. 4. The multispecific polypeptide construct of any one of claims 1 to 3, wherein the first component comprises a first antigen-binding domain and the second component comprises a second antigen-binding domain, each of the antigen-binding domains binding to a tumor-associated antigen (TAA).

5. From the N-terminus to the C-terminus, An antigen-binding domain that binds to a tumor-associated antigen (TAA); a heterodimeric immunoglobulin Fc region comprising a first Fc polypeptide and a second Fc polypeptide; a polypeptide linker that is 3 to 18 amino acids in length and is composed of at least 80% glycine residues; an anti-CD3 dsFv that binds to CD3 (CD3ε) comprising a variable heavy chain region (VH) and a variable light chain region (VL), wherein the VH and VL of the anti-CD3 dsFv are linked to opposing polypeptides of a heterodimeric Fc; and Costimulatory receptor binding domain (CRBR) that binds to a costimulatory receptor and / or an antigen-binding domain that binds to a tumor-associated antigen (TAA) A multispecific polypeptide construct comprising: A multispecific polypeptide construct comprising at least one CRBR which is an sdAb and at least one antigen binding domain which is an sdAb.

6. The Fc region an amino acid sequence set forth in SEQ ID NO:1, or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to SEQ ID NO:1; an amino acid sequence as set forth in SEQ ID NO:2, or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to SEQ ID NO:2; an amino acid sequence as set forth in SEQ ID NO:4, or a sequence of amino acids having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to SEQ ID NO:4; or An amino acid sequence set forth in SEQ ID NO:5, or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to SEQ ID NO:

5.

6. The multispecific polypeptide construct of any one of claims 1 to 5, comprising a polypeptide comprising:

7. and / or the Fc region comprises a polypeptide that includes at least one modification to enhance FcRn binding. the Fc region comprises a polypeptide comprising at least one amino acid modification that reduces effector function and / or reduces binding to an effector molecule selected from an Fcγ receptor or C1q; A multispecific polypeptide construct according to any one of claims 1 to 6.

8. 8. The multispecific polypeptide construct of any one of claims 1 to 7, wherein the CD3 binding region is unable to bind or engage CD3 unless at least one of the antigen binding domains is bound to its TAA.

9. The polypeptide linker is 6 to 18 amino acids, 6 to 14 amino acids, 6 to 12 amino acids, 6 to 10 amino acids, 6 to 8 amino acids, 8 to 18 amino acids, 8 to 14 amino acids, 8 to 12 amino acids, 8 to 10 amino acids, 10 to 18 amino acids, 10 to 14 amino acids, 10 to 12 amino acids, 12 to 18 amino acids, 12 to 14 amino acids, or 14 to 18 amino acids in length. A multispecific polypeptide construct according to any one of claims 1 to 8.

10. The polypeptide linker is 12 to 18 amino acids in length, or the polypeptide linker is 15-18 amino acids in length; A multispecific polypeptide construct according to any one of claims 1 to 8.

11. A multispecific polypeptide construct described in any one of claims 1 to 10, wherein the polypeptide linker is a non-cleavable linker.

12. The non-cleavable linker is GGS, GGGGS (SEQ ID NO:149), GGGGGS (SEQ ID NO:135), and combinations thereof; (GGS)n, n is 1 to 6; (GGGGS)n (SEQ ID NO:173), where n is 1 to 3; and / or (GGGGGS)n (SEQ ID NO:172), wherein n is 1 to 3; 12. The multispecific polypeptide construct of claim 11.

13. The non-cleavable linker is G.G.S., GGGGS (SEQ ID NO:149), GGGGGS (SEQ ID NO:135), (GGS) 2 (SEQ ID NO:10)、 GGSGGSGGS (SEQ ID NO:11), GGSGGSGGSGGS (SEQ ID NO:12), GGSGGSGGSGGSGGS (SEQ ID NO: 13), GGGGGSGGGGGSGGGGGS (SEQ ID NO: 119), GGSGGGGSGGGGSGGGGS (SEQ ID NO: 147), or GGGGSGGGGSGGGGS (SEQ ID NO:170) 13. The multispecific polypeptide construct of claim 11 or 12, comprising:

14. The multispecific polypeptide construct according to any one of claims 1 to 10, wherein the linker is a cleavable linker.

15. 1. A multispecific polypeptide construct comprising a first component comprising a heterodimeric Fc region and a second component comprising an anti-CD3 dsFv comprising a variable heavy region (VH) and a variable light region (VL), the VH and VL constituting the anti-CD3 dsFv are linked to opposing polypeptides of a heterodimeric Fc; the first and second components are coupled by a cleavable linker that is a substrate for a protease selected from matriptase, matrix metalloprotease (MMP), granzyme B, and combinations thereof, the cleavable linker being 6 to 24 amino acids in length and comprising the amino acid sequence set forth in SEQ ID NO: 150, 151, 154, 155, 157, or 158; a heterodimeric Fc region positioned N-terminal to the anti-CD3 antibody; one or both of the first and second components comprises at least one antigen binding domain that binds to a tumor associated antigen (TAA), each of the at least one antigen binding domain being an sdAb; The second component comprises at least one costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor, and each of the at least one CRBR is an sdAb; Multispecific polypeptide constructs.

16. 16. The multispecific polypeptide construct of any one of claims 1 to 15, wherein the sdAb is a human sdAb or a humanized sdAb.

17. The antigen-binding domain, or each of the antigen-binding domains independently, is: 1-92-LFA-3, 5T4, α4 integrin, αV integrin, α4β1 integrin, α4β7 integrin, AGR2, anti-Lewis Y, apelin J receptor, APRIL, B7-H3, B7-H4, BAFF, BTLA, C5 complement, C-242, CA9, CA19-9 (Lewis a), carbonic anhydrase 9, CD2, CD3, CD6, CD9, CD11 a, CD19, CD20, CD22, CD24, CD25, CD27, CD28, CD30, CD33, CD38, CD40, CD40L, CD41, CD44, CD44v6, CD47, CD51, CD52 , CD56, CD64, CD70, CD71, CD74, CD80, CD81, CD86, CD95, CD117, CD123, CD125, CD132 (IL-2RG), CD133, CD137, CD138 , CD166, CD172A, CD248, CDH6, CEACAM5 (CEA), CEACAM6 (NCA-90), claudin 3, claudin 4, cMet, collagen, Cripto, CSFR, CSFR-1, CTLA-4, CTGF, CXCL10, CXCL13, CXCR1, CXCR2, CXCR4, CYR61, DL44, DLK1, DLL3, DLL4, DPP-4, DSG1, EDA, EDB , EGFR, EGFRviii, endothelin B receptor (ETBR), ENPP3, EpCAM, EPHA2, EPHB2, ERBB3, RSV F protein, FAP, FGF-2, FGF8, FGFR1, FGFR2, FGFR3, FGFR4, FLT-3, folate receptor alpha (FRα), GAL3ST1, G-CSF, G-CSFR, GD2, GITR, GLUT1, GLUT4, GM-CSF, GM-CSFR, GPIIb / IIIa receptor, Gp130, GPIIB / IIIA, GPNMB, GRP78, HER2 / neu, HER3, HER4, HGF, hGH, HVEM, hyaluronidase, ICOS, IFNα, IFNβ, IFNγ, IgE, IgE receptor (FceRI), IGF, IGF1R, IL1B, IL1R, IL2, IL11, IL12, IL12p40, IL-12R, IL-12Rβ1, IL13, IL13R, IL15, IL17, IL18, IL21, I L23, IL23R, IL27 / IL27R (wsx1), IL29, IL-31R, IL31 / IL31R, IL2R, IL4, IL4R, IL6, IL6R, insulin receptor, Jagged ligand, Jagged1, Jagged2, KISS1-R, LAG-3, LIF-R, Lewis X, LIGHT, LRP4, LRRC26, Ly6G6D, LyPD1, MCSP, mesothelin, MRP4, MUC1, mucin 16 (MUC16, CA-125), Na / K ATPase, NGF, Nicastrin, Notch receptor, Notch1, Notch2, Notch3, Notch4, NOV, OSM-R, OX-40, PAR2, PDGF-AA, PDGF-BB, PDGFRα, PDGFRβ, PD-1, PD-L1, PD-L2, phosphatidylserine, P1GF, PSCA, PSMA, PSGR, RAAG12, RAGE, SLC44A4, sphingosine-1-phosphate, STEAP1, STEAP2, TAG-72, TAPA1, TEM-8, TGFβ, TIGIT, TIM-3, TLR2, TLR4 17. The multispecific polypeptide construct of any one of claims 1 to 16, which binds to a tumor antigen selected from: TLR6, TLR7, TLR8, TLR9, TMEM31, TNFα, TNFR, TNFRS12A, TRAIL-R1, TRAIL-R2, transferrin, transferrin receptor, TRK-A, TRK-B, uPAR, VAP1, VCAM-1, VEGF, VEGF-A, VEGF-B, VEGF-C, VEGF-D, VEGFR1, VEGFR2, VEGFR3, VISTA, WISP-1, WISP-2, and WISP-3.

18. (a) the multispecific polypeptide construct comprises at least a first antigen-binding domain and a second antigen-binding domain, wherein the first antigen-binding domain and the second antigen-binding domain bind to the same TAA, or (b) the multispecific polypeptide construct comprises at least a first antigen-binding domain and a second antigen-binding domain, wherein the first antigen-binding domain and the second antigen-binding domain bind to different TAAs; A multispecific polypeptide construct according to any one of claims 1 to 17.

19. 19. The multispecific polypeptide construct of any one of claims 1 to 18, wherein at least one costimulatory receptor binding region (CRBR) binds to a costimulatory receptor selected from 41BB (CD137), OX40 (CD134), CD27, Glucocorticoid-Inducible TNFR-Related Protein (GITR), CD28, ICOS, CD40, B-cell activating factor receptor (BAFF-R), B-cell maturation antigen (BCMA), Transmembrane activator and CAML interactor (TACI), and NKG2D.

20. Anti-CD3 dsFv, (a) a VH CDR1 comprising the amino acid sequence GFTFNTYAMN (SEQ ID NO: 312); a VH CDR2 comprising the amino acid sequence RIRSKYNNYATY (SEQ ID NO: 313); a VH CDR3 comprising the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO: 18); a VL CDR1 comprising the amino acid sequence GSSTGAVTTSNYAN (SEQ ID NO: 319); a VL CDR2 comprising the amino acid sequence GTNKRAP (SEQ ID NO: 320); and a VL CDR3 comprising the amino acid sequence ALWYSNHWV (SEQ ID NO: 315); and / or (b) a VH having an amino acid sequence of any of SEQ ID NOs: 14, 32-62, 287, 290, and 311, or a sequence that exhibits at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to any of SEQ ID NOs: 14, 32-62, 287, 290, and 311; and VL having an amino acid sequence of any of SEQ ID NOs: 15, 63-81, 241, 288, and 289, or a sequence that exhibits at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to any of SEQ ID NOs: 15, 63-81, 241, 288, and 289.

20. The multispecific polypeptide construct according to any one of claims 1 to 19, comprising:

21. A multispecific polypeptide construct described in any one of claims 1 to 20, wherein the first polypeptide of the heterodimeric Fc comprises an amino acid sequence set forth in any of SEQ ID NOs: 82, 84, 86, 88, 94-97, 291, 293, 297, 298, or 285, and the second polypeptide of the heterodimeric Fc comprises an amino acid sequence set forth in any of SEQ ID NOs: 83, 85, 87, 89-93, 98-101, 292, 294-296, 295, 299, 300, or 286.

22. 22. The multispecific polypeptide construct of any one of claims 1 to 21, conjugated to an agent.

23. A polynucleotide encoding the multispecific polypeptide construct of any one of claims 1 to 22.

24. A polynucleotide comprising a first nucleic acid sequence encoding a first polypeptide of the multispecific polypeptide construct according to any one of claims 1 to 22 and a second nucleic acid sequence encoding a second polypeptide of the multispecific polypeptide construct, the first and second nucleic acid sequences are separated by an internal ribosome entry site (IRES) or by a nucleic acid encoding a self-cleaving peptide or a peptide that causes ribosome skipping; Polynucleotide.

25. A vector comprising the polynucleotide of claim 23 or 24.

26. 26. A cell comprising one or more polynucleotides of claim 23 or 24 or one or more vectors of claim 25.

27. introducing one or more polynucleotides according to claim 23 or 24 or one or more vectors according to claim 25 into a cell; and Culturing the cells under conditions whereby the multispecific polypeptide construct is produced. Contains, or Culturing the cell of claim 26 under conditions such that the multispecific polypeptide is produced by the cell. Methods for producing multispecific polypeptide constructs.

28. A pharmaceutical composition comprising the multispecific polypeptide construct of any one of claims 1 to 22 and a pharma- ceutically acceptable carrier.

29. A composition for use in stimulating or inducing an immune response in a subject having a disease or condition, comprising a multispecific polypeptide construct according to any one of claims 1 to 22 or a pharmaceutical composition according to claim 28.

30. A composition comprising a multispecific polypeptide construct according to any one of claims 1 to 22 or a pharmaceutical composition according to claim 28 for use in the treatment of a disease or condition in a subject.

31. 29. Use of a multispecific polypeptide construct according to any one of claims 1 to 22 or a pharmaceutical composition according to claim 28 for the manufacture of a medicament for use in the treatment of a disease or condition in a subject.

Images