Immunomodulatory triple-specific T cell engager fusion protein

Abstract

The present disclosure provides fusion proteins with multifunctional biological designs for programmed target engagement. In certain embodiments, the fusion proteins described herein provide simultaneous T cell engagement with a target antigen and immune checkpoint targeting.

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Provisional Patent Application No. 63 / 303,481, filed January 26, 2022, and U.S. Provisional Patent Application No. 63 / 417,655, filed October 19, 2022, which are incorporated herein by reference for all purposes. [Background technology]

[0002] background Checkpoint inhibitors, such as PD-1 and PD-L1 inhibitors, used to treat cancer work by blocking the interaction of immune checkpoint proteins expressed on T cells and cancer cells, thereby reducing the immunosuppression that would otherwise occur. Several anti-PD-1 antibodies (e.g., pembrolizumab and nivolumab) and anti-PD-L1 antibodies (e.g., atezolizumab) have been approved for the treatment of certain types of cancer. These checkpoint inhibitors may be administered alone or in combination with another antibody against an antigen overexpressed on cancer cells, for example, in combination with an anti-HER2 antibody. Summary of the Invention

[0003] overview In certain embodiments, described herein is a trispecific fusion protein comprising: (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold. Such trispecific fusion proteins can be trivalent or tetravalent, i.e., can include another binding domain, e.g., a second anti-TAA binding domain.

[0004] In some embodiments, described herein are tetravalent and trispecific fusion proteins comprising: (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain and a third binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a tumor cell; (iii) a fourth binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, the third binding domain, and the fourth binding domain are operably linked to the scaffold.

[0005] As further described herein, the trispecific fusion proteins of the present disclosure may have a wide variety of formats, and such fusion proteins may include one or more peptide linker moieties capable of interconnecting the various polypeptides and domains that may comprise the fusion protein.

[0006] In some embodiments, described herein is a trispecific fusion protein comprising an anti-CD3 binding domain comprising a VH region and a VL region that specifically binds to a CD3 (cluster of differentiation 3) antigen on the surface of a T cell; one or two anti-TAA binding domains comprising a VH region and a VL region that specifically binds to a TAA (tumor-associated antigen) on the surface of a tumor cell; and a PD-1 polypeptide capable of binding to a PD-L1 receptor on the surface of a tumor cell, wherein the anti-CD3 domain, the anti-TAA domain(s), and the PD-1 polypeptide are operably linked to a scaffold. The scaffold may be an antibody or a portion thereof, and may have, comprise, or consist of a heterodimeric immunoglobulin (Ig) Fc region.

[0007] In certain embodiments, the anti-CD3 binding domain has a Fab format, the anti-TAA binding domain(s) have an scFv format, and the PD-1 polypeptide is fused to the N-terminus of the VH of the anti-CD3 binding domain Fab via a peptide linker. In certain embodiments, the anti-CD3 binding domain has a Fab format, the anti-TAA binding domain has an scFv format, and the PD-1 polypeptide is fused to the N-terminus of the VL of the anti-CD3 binding domain Fab via a peptide linker.

[0008] In certain embodiments, the PD-1 polypeptide used in the fusion proteins described herein is wild-type PD-1. In certain embodiments, the PD-1 polypeptide contains one or more mutations that increase or decrease its binding affinity for PD-L1, where such one or more mutations (e.g., amino acid substitution(s)) are relative to a particular wild-type or non-mutated PD-1 polypeptide amino acid sequence. In certain embodiments, the PD-1 polypeptide of the fusion proteins herein has an affinity for PD-L1 of about 100 μM to about 10 pM, or about 10 μM to about 150 pM, or about 100 nM to about 150 pM.

[0009] In certain embodiments, the trispecific fusion proteins described herein comprise an anti-CD3 binding domain comprising a VH region and a VL region that specifically binds to a CD3 (cluster of differentiation 3) antigen on the surface of a T cell; one or two anti-TAA binding domains comprising a VH region and a VL region that specifically binds to a TAA (tumor-associated antigen) on the surface of a tumor cell; and a PD-1 polypeptide capable of binding to a PD-L1 receptor on the surface of a tumor cell, wherein the fusion protein has one of the formats exemplified in Figures 3A-3LL herein.

[0010] In some embodiments, the TAA can be any of mesothelin (MSLN), claudin 18.2 (Cldn18.2), GPC3, DLL3, PSMA, MUC17, LIV1, ROR1, and EGFRvIII.

[0011] In some embodiments, the peptide linker is selected from SEQ ID NOs: 15, 16, 17, 18, 19, and 20.

[0012] In some embodiments, the trispecific fusion protein reduces or inhibits binding of a PD-1 polypeptide located on a T cell to a PD-L1 polypeptide located on a cancer cell.

[0013] Also described herein are pharmaceutical formulations / compositions of the trispecific fusion proteins described herein.

[0014] Also described herein is a method of treating cancer in a subject in need thereof, comprising administering to the subject a trispecific fusion protein.

[0015] Also described herein are methods of overcoming or preventing T cell exhaustion, comprising exposing the T cells to the trispecific fusion proteins described herein, wherein the T cells can be located in a subject, such as a human or a rodent.

[0016] These and other features, aspects, and advantages of the present invention will become better understood with reference to the following description and accompanying drawings (also referred to herein as "figures"). [Brief explanation of the drawings]

[0017] [Figure 1] Figure 1(A) shows a diagram of a possible mechanism by which the trispecific fusion protein (PD-1 / anti-CD3 / anti-TAA) described herein can exert its unique activity as a T cell engager therapeutic. The molecule is thought to co-engage tumor-associated antigens (TAA) and PD-L1 on cancer cells, simultaneously binding to CD3 on the T cell surface, thereby bringing T cells into close proximity with the cancer cells. This blocks the PD-L1 / PD-1 interaction, resulting in checkpoint inhibition and T cell activation. If such an interaction is not blocked, T cell activation can be prevented. Figure 1(B) shows a spatial model of the antibody (PDB 1HZH), with the CH / CL of one arm representing the size of the scFv domain that can interact with the HER2 protein as a TAA (PDB 1N8Z). The Fab interacts with CD3 (PDB 6JXR) and has a model helical linker connecting it to the PD-1 polypeptide (PDB:3BIK), which interacts with PD-L1. From a spatial perspective, this model shows that the interactions of the various binding domains of the fusion protein with their respective targets are spatially feasible. [Figure 2] FIG. 1 is a schematic diagram of an exemplary trispecific PD-1 / anti-CD3 / anti-TAA fusion protein where the TAA is a HER, according to certain embodiments of the present disclosure. [Figure 3-1]Figure 3 shows a series of diagrams illustrating various non-limiting examples of possible formats for the PD-1 / anti-CD3 / anti-TAA trispecific fusion proteins described herein (e.g., Figures 3(A)-3(LL)). The PD-1 domain may be fused via a linker to the N- or C-terminus of the anti-CD3 heavy or light chain, or may be linked to the C-terminus of the Fc domain. The PD-1 domain may be fused via a linker to the N- or C-terminus of the anti-TAA heavy or light chain. The VH and VL domains of the antigen-binding domain may be, for example, in a Fab format, a hybrid format comprising one Fab and one scFv, or part of a dual scFv. Multiple anti-TAA antigen-binding domains may be present, each comprising a VH and a VL with affinity for a TAA. [Figure 3-2] See description of Figure 3-1. [Figure 3-3] See description of Figure 3-1. [Figure 4A]

[0023] Figure 1 shows a graph of in vivo tumor growth in animals treated with bispecific or trispecific antibody fusion proteins. Tumor growth over time is shown for animals treated once daily for 4 weeks with either 1 mg / kg or 0.5 mg / kg of either bispecific antibody fusion protein (v32497, black line, n=8 at each dose) or trispecific antibody fusion protein (v31929, gray line, n=8 at each dose). The days of test article injection are indicated by vertical dotted lines with arrows above them. [Figure 4B] Figure 1 shows in vivo tumor growth in animals treated with bispecific or trispecific antibody fusion proteins. Tumor growth is shown for animals treated with a combination of antibodies designed to independently target the same epitope as the trispecific antibody fusion protein (v31929). Bispecific v32497 was administered at 1 mg / kg once daily, and anti-PD-L1 antibody (atezolizumab, v33449) was administered at 5 mg / kg twice daily for 4 weeks (n=8). The days of injection of the trispecific or bispecific test article are indicated by vertical dotted lines with arrows above. [Figure 4C]1 is a graph showing in vivo tumor growth in animals treated with bispecific or trispecific antibody fusion proteins. The frequency of durable responses for each treatment regimen is shown, based on animals whose tumors regressed to 0 mm3 and remained that way for the duration of the study (n / a = not applicable). [Figure 4D] Figure 4 shows graphs depicting in vivo tumor growth in animals treated with bispecific or trispecific antibody fusion proteins. Results are shown for the same study shown in Figures 4(A)-4(C), but using different donors for animal reconstitution. The trispecific fusion protein v31929 was shown to be able to induce a significantly stronger T cell-mediated antitumor response when compared to either the bispecific construct alone or in combination with a checkpoint inhibitor, resulting in a complete response in 6 out of 8 animals (compared to the control cohort and no response with the control / benchmark construct). [Figure 5] The expression levels of exhaustion markers PD-1, LAG-3, and TIM-3, which increased with sequential CD3 / CD28 stimulation, are shown. T cells were stimulated with CD3 / CD28 Dynabeads every two days for a total of 8 days. T cells were sampled on days 0 and 2, 4, 6, and 8. Naive T cells served as controls. Surface expression of PD-1, LAG-3, and TIM-3 was determined by flow cytometry. Bar graphs show the percentage of marker-expressing cells, determined by gating on marker-positive cells based on FMO controls within the live / CD3+ gate (A). [Figure 6] The levels of the inflammatory cytokines IFN gamma (also referred to herein as "IFNγ" or "IFNg", Figure 6(A)), IL-2 (Figure 6(B)), and TNF alpha (TNFα, Figure 6(C)) in supernatants obtained from cell cultures of CD3 / CD28 stimulated T cells over 8 days of culture are shown. [Figure 7] 1 is a bar graph showing the proliferation of exhausted T cells 5 days after exposure to JIMT-1 tumor cells in the presence of various concentrations of the trispecific PD-1 / anti-CD3 / anti-HER2 fusion protein (v31929) and control constructs. [Figure 8]The levels of the inflammatory cytokines IFN-gamma (Figure 8(A)) and IL-2 (Figure 8(B)) produced by exhausted T cells 3 days after exposure to JIMT-1 tumor cells are shown in the presence of various concentrations of the trispecific PD-1 / anti-CD3 / anti-HER2 fusion protein (v31929) and control construct. [Figure 9] Figures 9 and 10 are graphs showing the percent survival of JIMT-1 target cells after 2 days of culture with exhausted T cells in the presence of trispecific PD-1 / anti-CD3 / anti-HER2 fusion proteins and controls. [Figure 10] Figures 9 and 10 are graphs showing the percent survival of JIMT-1 target cells after 2 days of culture with exhausted T cells in the presence of trispecific PD-1 / anti-CD3 / anti-HER2 fusion proteins and controls. [Figure 11] Oncocytic binding of the trispecific PD-1 / anti-CD3 / anti-Her2 fusion protein (v31929) on naive (top) and exhausted (bottom) T cells is shown. Binding of atezolizumab (anti-PD-L1) and the CD3 bispecific benchmark v35923 is also shown for comparison. [Figure 12] Graphs (top) depict percent survival of H292 target tumor cells after 72 hours of co-culture with T cells in the presence of different trispecific and trivalent or tetravalent PD-1 / anti-MSLN / anti-CD3 fusion protein constructs in various formats. Exemplary graphs comparing the efficacy of two trispecific and tetravalent formats, v38449 (middle) and v38450 (bottom), against the anti-MSLN MH6T TriTAC benchmark and format-matched bispecific controls are also shown. [Figure 13] Figure 1 shows a graph of percent survival of SNU-216, H292, HCT-116, SKOV-3, and OVCAR-3 target tumor cells after 72 hours of co-culture with T cells in the presence of the trispecific trivalent PD-1 / anti-MSLN / anti-CD3 fusion protein v38344 and the control construct v38345 (comprising a PD-1 moiety with reduced affinity for PD-L1 compared to the format-matched anti-CD3 x MSLN bispecific + PD-L1 polypeptide used in v38344). [Figure 14]Graphs and tables showing quantification of MSLN and PD-L1 surface-expressed receptors for tumor cell lines SNU-216, H292, HCT-116, SKOV-3, and OVCAR-3. Quantification of PD-L1 surface-expressed receptors after 24 hours of incubation with 20 ng / mL IFNg is also shown. [Figure 15] Figure 1 shows the results of a hybrid PD-1 / PD-L1 reporter gene assay investigating crosslinking of T cells with HCT-116 tumor cells and blockade of PD-1 / PD-L1 checkpoint engagement. Results are shown for a trispecific tetravalent PD-1 / anti-CD3 / anti-MSLN fusion protein (v38449), a format-matched bispecific control containing an attenuated (KO) PD-1 moiety (v38454), and the combination of the bispecific control (v38454) with atezolizumab (v33449). [Figure 16] Percent survival of SNU-620 target tumor cells after 96 hours of co-culture with T cells in the presence of different trivalent or tetravalent trispecific PD-1 / anti-CLDN18.2 / anti-CD3 fusion proteins and control constructs is shown (top). Also shown is an exemplary graph comparing the efficacy of two trivalent trispecific formats (v38408 (bottom left) and v38410 (bottom right)) against AMG-910 and a format-matched bispecific control. [Figure 17] Percent survival of SNU-620 target tumor cells after 72 hours of co-culture with T cells in the presence of different tetravalent trispecific PD-1 / anti-CLDN18.2 / anti-CD3 fusion proteins and control constructs is shown (top). Also shown is an exemplary graph comparing the efficacy of two tetravalent trispecific formats (v38999 (bottom left) and v39003 (bottom right)) against AMG-910 and a format-matched bispecific control. [Figure 18] Percent survival of SNU-620 (top), KATO-III (middle), and SNU-601 (bottom) target tumor cells after 96 or 120 hours of co-culture with T cells using various T cell to tumor cell ratios in the presence of the trivalent trispecific PD-1 / anti-CLDN18.2 / anti-CD3 fusion protein v38410 or a control construct is shown. [Figure 19]Graphs and tables showing quantification of CLDN and PD-L1 surface-expressed receptors for tumor cell lines SNU-620, SNU-601, and KATO-III. Quantification of PD-L1 surface-expressed receptors after 24 hours of incubation with 20 ng / mL IFNg is also shown. [Figure 20] Figure 1 shows the results of a hybrid PD-1 / PD-L1 reporter gene assay investigating crosslinking of T cells with SNU-620 tumor cells and blockade of PD-1 / PD-L1 checkpoint engagement. Results are shown for three trispecific PD-1 / anti-CD3 / anti-Cldn18.2 variants and their respective format-matched bispecific controls containing attenuated (KO) PD-1 moieties, as well as the combination of the bispecific controls with atezolizumab (v38408, top; v38410, bottom left; v39007, bottom right). [Figure 21] Figure 1 shows the percent survival of HCT-116 and H292 target tumor cells after 72 hours of co-culture with T cells in the presence of trispecific PD-1 / anti-MSLN / anti-CD3 fusion proteins containing affinity-regulated or wild-type (WT) PD-1 moieties, as further described herein. Results for the respective format-matched bispecific controls are also shown. [Figure 22] Figure 1 shows the percent survival of SNU-620 target tumor cells after 72 hours of co-culture with T cells in the presence of trispecific PD-1 / anti-Cldn18.2 / anti-CD3 fusion proteins containing affinity-regulated or wild-type (WT) PD-1 moieties, as further described herein. Results for the respective format-matched bispecific controls are also shown. [Figure 23]Figure 1 shows the results of a hybrid PD-1 / PD-L1 reporter gene assay investigating crosslinking of T cells with HCT-116 tumor cells and blockade of PD-1 / PD-L1 checkpoint engagement. Results are shown for three trispecific PD-1 / anti-CD3 / anti-MSLN variants (v38910, top left; v38449, top right; v38450, bottom) containing affinity-regulated or wild-type (WT) PD-1. Results are also shown for the respective format-matched bispecific controls (v38454, v38455) containing attenuated (KO) PD-1, and the combination of the bispecific controls with atezolizumab. The control variant v38354 is a "true" anti-MSLN / anti-CD3 bispecific construct that does not contain the PD-1 moiety. [Figure 24] Figure 1 shows the results of a hybrid PD-1 / PD-L1 reporter gene assay examining crosslinking of T cells with SNU-620 tumor cells and blockade of PD-1 / PD-L1 checkpoint engagement. Results are shown for three trispecific PD-1 / anti-CD3 / anti-Cldn18.2 formats using variants containing affinity-tuned or wild-type (WT) PD-1. Also shown are format-matched bispecific controls for each containing attenuated (KO) PD-1, and the combination of the bispecific controls with atezolizumab. [Figure 25] Cytokine production 72 hours after TDCC using H292 tumor cells is shown. Tumor cells and T cells were treated with three trispecific PD-1 / anti-CD3 / anti-MSLN variants containing affinity-tuned or wild-type (WT) PD-1. Results for a format-matched bispecific control containing attenuated (KO) PD-1 are also shown. Production of TNFα, IL-2, and IFNγ was measured using the MSD assay. [Figure 26] Flow cytometry results of the DC differentiation protocol show upregulation of DC surface markers CD11c, CD80, CD86, and PD-L1 and downregulation of the monocyte marker CD14 by day 7 after exposure to the trispecific fusion protein v31929. [Figure 27]Results of an autologous DC-T cell co-culture assay demonstrate that the trispecific fusion protein (v31929) co-engages PD-L1-expressing APCs and T cells, resulting in high levels of T cell activation as measured by upregulation of CD25 and proliferation of both CD4 and CD8 T cells. [Figure 28] Figure 1 shows evaluation of dendritic cell (DC)-T cell co-engagement by trispecific PD-1 / anti-CD3 / anti-Her2 fusion proteins containing affinity-tuned or wild-type PD-1 moieties. Trispecific constructs containing high-affinity PD-1 (v31929) exhibit superior T cell activation capacity compared to format-matched variants with lower PD-1 affinity. DETAILED DESCRIPTION OF THE INVENTION

[0018] Detailed Description In various embodiments, disclosed herein are trispecific fusion proteins comprising a PD-1 domain derived from the extracellular IgV domain of a PD-1 polypeptide, an anti-CD3-binding domain, and an anti-TAA (tumor-associated antigen)-binding domain linked to a scaffold. Also disclosed herein are pharmaceutical compositions comprising the trispecific fusion proteins disclosed herein and methods of using the same. Schematic diagrams of exemplary PD-1 / anti-CD3 / anti-TAA fusion proteins in various formats and geometries are shown in Figure 3 (e.g., Figures 3(A)-3(LL)). A schematic diagram of a specific embodiment of a trispecific fusion protein in which the TAA is HER2 is shown in Figure 2. Figure 1A shows a schematic diagram of co-engagement of T cells and cancer cells by a trispecific fusion protein. The anti-CD3 domain, while bound to the cancer cell via the TAA, prompts the T cell to form a tight synapse (e.g., a TCR-independent immune synapse), triggering T cell activation for cancer cell killing. At the same time, the PD-1 domain can bind to PD-L1 on cancer cells (which may, in some cases, be different from those to which the fusion protein binds via the anti-TAA binding domain), thereby acting as a checkpoint inhibitor. Trispecific fusion proteins are trifunctional molecules that combine anti-TAA, anti-CD3, and anti-PD-L1 activity. In certain embodiments, the trifunctional format and geometry of the fusion protein may facilitate binding that brings all three targets into close proximity at the immunological synapse, potentially allowing antigen binding to benefit from avidity and local concentration effects. Figure 1(B) provides a hypothetical three-dimensional model of a trispecific fusion protein interacting with CD3, HER2 (i.e., a TAA), and PD-L1.

[0019] I. Definition Terms used in the claims and specification are defined briefly here and in more detail below.

[0020] The term "fusion protein," as used herein, refers to a protein comprising multiple (e.g., two, three, four, or more) polypeptide regions, chains, or domains linked together, for example, by peptide bonds or other covalent bonds (e.g., disulfide bonds). Thus, as used herein in the context of a single fusion polypeptide chain, "fused" refers to polypeptide sequences linked together via peptide bonds. Examples include immunomodulatory ligands, antigen-binding domains, immunomodulatory receptors, antibodies fused to antibodies or antibody fragments, or scaffolds. Fusion proteins comprising two or more polypeptide chains described herein may also be referred to as "variants" or "constructs."

[0021] The term "biologically functional protein," as used herein, broadly refers to a polypeptide or protein that has a biological function, such as an antibody or portion thereof, e.g., a dimeric Fc or Fab domain.

[0022] The term "antibody," as used herein, generally refers to immunoglobulins (Ig) and Ig-derived polypeptide constructs, which can include naturally occurring Ig from various species (e.g., human IgG, IgA, IgE, rodent, camel, shark, etc.), and non-naturally occurring Ig-like molecules such as the fusion proteins described in this disclosure.

[0023] The term "ligand-receptor pair," as used herein, refers to a receptor polypeptide and a ligand polypeptide that specifically bind to each other. An example is the PD-1-PD-L1 pair.

[0024] An "immunomodulatory" molecule, as used herein, generally refers to a molecule that has the ability to modulate, either directly or indirectly, an immune response in a subject, e.g., upregulating or downregulating an immune response and / or modulating immune cell activity.

[0025] The term "peptide linker," as used herein, refers to a peptide that bonds, couples, or links other peptides or polypeptides. Such terms may be used interchangeably herein with the terms "peptide linker" or "peptide-based linker," and generally refer to a linker moiety that includes at least one amino acid residue. In many embodiments, the linkers herein comprise or consist of a contiguous sequence of two or more amino acid residues. In some embodiments, the linkers of the trispecific fusion proteins herein comprise or consist of a contiguous sequence of about 2, 5, 10, 15, 20, 25, 30, 40, or about 50 amino acid residues.

[0026] The terms "Fc region," "Fc," and "Fc domain" are used interchangeably herein and refer to the C-terminal region of an immunoglobulin heavy chain that contains at least a portion of the constant region.

[0027] The term "bispecific," as used herein in the context of a bispecific fusion protein or antibody, generally refers to a biologically functional protein (e.g., a fusion protein described herein) that is capable of specifically binding to two distinct epitopes, which may be on the same or different antigens. In some embodiments herein, the term "bispecific" may be used for a fusion protein that contains three binding domains for three different antigens (i.e., may be considered trispecific), but in which one of such binding domains is a knockout (KO) binding domain that has reduced, or in some cases no significant, binding to its respective epitope or antigen. As an example, a fusion protein herein containing a KO PD-1 domain (e.g., containing one or more knockout mutations) may be referred to as "bispecific" and can be used as a control construct having the same format but with reduced PD-L1 binding.

[0028] The term "trispecific," as used herein in the context of a trispecific fusion protein or antibody, generally refers to a biologically functional protein (e.g., a fusion protein described herein) that is capable of specifically binding to three distinct epitopes, which may be on the same or different antigens.

[0029] The term "multispecific," as used herein in the context of a multispecific fusion protein or antibody, generally refers to a biologically functional protein (e.g., a fusion protein described herein) that is capable of specifically binding to two or more distinct target molecules or epitopes, which may be on the same or different antigens.

[0030] The "valency" of a fusion protein described herein refers to the total number of antigen-binding domains (e.g., scFv, Fab, or other polypeptide moieties) contained in the fusion protein. As an example, as described in various embodiments herein, a fusion protein herein can be trivalent, thus comprising three antigen-binding domains. In other embodiments, a fusion protein herein can be tetravalent, i.e., comprising four antigen-binding domains. Such three or four antigen-binding domains of a trivalent or tetravalent fusion protein can each be specific for the same or different antigens and / or epitopes. As described in various embodiments herein, a fusion protein of the present disclosure can be trispecific, where such a trispecific fusion protein is at least trivalent because it comprises three antigen-binding domains, each binding domain having binding specificity for a different epitope and / or antigen. In some embodiments, as further described herein, trispecific fusion proteins of the present disclosure may be tetravalent and comprise four binding domains, wherein three of such four binding domains have binding specificities for three different epitopes and / or antigens and two of such four binding domains have binding specificity for the same epitope and antigen.

[0031] The term "immune checkpoint," as used herein, generally refers to a regulatory pathway of the immune system that controls activation of the immune system.

[0032] The term "specifically binds" (and grammatical variations thereof), when referring to the binding of a fusion protein herein to a specific antigen, epitope, ligand, or receptor, means binding that is measurably different from non-specific interactions.

[0033] As detailed below, "mammal" includes both humans and non-humans, including, but not limited to, humans, non-human primates, canines, felines, murines, bovines, equines, and porcines.

[0034] It should be noted that as used in this specification and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise.

[0035] Abbreviations used in this application include: PD-1 (programmed cell death protein 1); PD-L1 (programmed death ligand 1); CD3 (cluster of differentiation 3); and CD28 (cluster of differentiation 28).

[0036] As used herein, the term "about" refers to approximately a ±10% variation from a given value. It is to be understood that such a variation is always included in any given value provided herein, whether or not it is specifically referred to.

[0037] As used herein, the terms "comprising," "having," "including," and "containing," and grammatical variations thereof, are inclusive or open-ended and do not exclude additional, unrecited elements and / or method steps. When used herein in connection with a fusion protein, composition, use, or method, the term "consisting essentially of" indicates that additional elements and / or method steps may be present, but that these additions do not substantially affect the manner in which the recited fusion protein, composition, method, or use functions. When used herein in connection with a fusion protein, composition, use, or method, the term "consisting of" excludes the presence of additional elements and / or method steps. A fusion protein, composition, use, or method described herein as including certain elements and / or steps may also, in certain embodiments, consist essentially of those elements and / or steps, and in other embodiments, consist of those elements and / or steps, whether or not those embodiments are specifically referred to.

[0038] It is contemplated that any embodiment discussed in this specification can be implemented with respect to any method, use, fusion protein, or composition disclosed herein, and vice versa.

[0039] It should also be understood that the positive enumeration of a feature in one embodiment serves as a basis for excluding the feature in another embodiment. In particular, where a list of alternatives is presented for a given embodiment or claim, it should be understood that one or more alternatives may be deleted from the list, and that the shortened list may form an alternative embodiment, whether or not such alternative embodiment is specifically referenced.

[0040] The sequences of the various amino acid sequences and clones and variants referred to herein are set forth, for example, in Table AA parts 1, 2 and 3, and Table BB.

[0041] II. Trispecific Fusion Proteins In certain embodiments, the present disclosure describes trispecific fusion proteins, such trispecific fusion proteins comprising: (i) a first binding domain capable of binding to an antigen on a cytotoxic effector cell; (ii) a second binding domain capable of binding to a TAA on a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on a tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, and the third binding domain are each operably linked to the scaffold, either directly or via one or more different linkers, e.g., peptide linkers. In certain embodiments, the antigen on the cytotoxic effector cell is CD3, and the third binding domain is a wild-type or variant PD-1 polypeptide.

[0042] In various embodiments, the present disclosure describes a trispecific fusion protein comprising: (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell (e.g., a T cell); (ii) a second binding domain capable of binding to a TAA on the surface of a tumor cell; (iii) a PD-1 polypeptide capable of binding to PD-L1 on the surface of a tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, and the PD-1 polypeptide are operably linked to the scaffold.

[0043] In some embodiments, such trispecific fusion proteins are trivalent, i.e., they comprise three binding domains capable of binding to three different epitopes and / or antigens.

[0044] In some embodiments, the cytotoxic effector cell is an immune cell. The immune cell can be a T cell. In some embodiments, the TAA and PD-L1 that the fusion protein engages and binds are located on the surface of the same tumor cell. In other embodiments, the TAA and PD-L1 are located on the surface of two different tumor cells.

[0045] In various embodiments, the first binding domain of the trispecific fusion protein is a Fab domain, wherein the Fab domain comprises a heavy chain comprising a VH sequence and a CH1 sequence, and a light chain comprising a VL sequence and a CL sequence, as further described and defined herein. Thus, in some embodiments, described herein is a trispecific fusion protein comprising: (i) a Fab domain (first binding domain) capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a TAA on the surface of a tumor cell; (iii) a PD-1 polypeptide capable of binding to PD-L1 on the surface of a tumor cell; and (iv) a scaffold, wherein the Fab domain, second binding domain, and PD-1 polypeptide are operably linked to the scaffold.

[0046] In various embodiments, the second binding domain of the trispecific fusion proteins herein is an scFv domain comprising a VL sequence linked to a VH sequence. The order or connectivity of these two scFv domain sequences can be, from N- to C-terminal, that the VH sequence is linked to the VL sequence, or that the VL sequence is linked to the VH sequence. The VH and VL sequences can be linked, for example, via a peptide linker, referred to herein in the context of scFv domains as a linker-scFv. Such a linker-scFv can be a peptide or peptide linker comprising or consisting of a contiguous sequence of amino acid residues, for example, about 2, 5, 10, 15, 20, 25, 30 or more contiguous amino acid residues.

[0047] In some embodiments, described herein is a trispecific fusion protein comprising: (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) an scFv domain (second binding domain) capable of binding to a TAA on the surface of a tumor cell; (iii) a PD-1 polypeptide capable of binding to PD-L1 on the surface of a tumor cell; and (iv) a scaffold, wherein the first binding domain, the scFv domain, and the PD-1 polypeptide are operably linked to the scaffold.

[0048] In certain embodiments, described herein is a trispecific fusion protein comprising: (i) a Fab domain (first binding domain) capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) an scFv domain (second binding domain) capable of binding to a TAA on the surface of a tumor cell; (iii) a PD-1 polypeptide capable of binding to PD-L1 on the surface of a tumor cell; and (iv) a scaffold, wherein the first binding domain, the scFv domain, and the PD-1 polypeptide are operably linked to the scaffold.

[0049] In any of these embodiments, the scaffold is the moiety to which the three binding domains are operably linked, either directly or via one or more distinct peptide linkers, as further described herein. The term "operably linked," as used herein in the context of a trispecific fusion protein, generally refers to the direct or indirect connection of a first domain of the trispecific fusion protein, e.g., a first binding domain, to a second domain of the trispecific fusion protein, e.g., the scaffold, wherein each operably linked domain of the fusion protein is capable of performing its function(s) (e.g., biological, physiological, and / or chemical function(s)) similarly or identically to the respective domain(s) isolated and not operably linked as part of the trispecific fusion protein. Furthermore, in cases where a first domain of a trispecific fusion protein is directly and operably linked to a second domain of the trispecific fusion protein, such first domain is linked to the second domain via a direct covalent bond, e.g., a peptide bond, without a linker. Alternatively, the first domain of the trispecific fusion protein may be indirectly operably linked to the second domain of the trispecific fusion protein in several ways, such as (i) by a linker moiety and / or (ii) through another domain that may be located between the first and second domains, for example, in a fusion polypeptide chain in which an scFv domain is linked to an Fc polypeptide via a Fab heavy chain.

[0050] In various embodiments, the backbone of the trispecific fusion proteins herein is an Fc domain, which comprises a first Fc polypeptide and a second Fc polypeptide. At least one of the first or second Fc polypeptides may comprise a CH2 domain and / or a CH3 domain sequence. In certain embodiments, both the first and second Fc polypeptides comprise a CH2 domain sequence and a CH3 domain sequence, respectively. In some embodiments, both the first and second Fc polypeptides comprise or consist of the same amino acid sequence. Such an Fc domain comprising Fc polypeptides with the same amino acid sequence may be referred to as a homodimeric Fc domain. In yet other embodiments, the trispecific fusion proteins of the present disclosure comprise a heterodimeric Fc domain, which comprises a first Fc polypeptide and a second Fc polypeptide, wherein the amino acid sequences of the first and second Fc polypeptides comprise or consist of amino acid sequences that differ by at least one amino acid residue. Thus, the first and second Fc polypeptides of a heterodimeric Fc domain can comprise amino acid sequences that share about 99%, 98%, 97%, 96%, or about 95% sequence identity. In some embodiments, the first and second Fc polypeptides of the heterodimeric Fc domain of the trispecific constructs described herein comprise one or more asymmetric amino acid modifications (e.g., substitutions) that promote preferential pairing of the first and second Fc polypeptides to form the heterodimeric Fc domain relative to the formation of a homodimeric Fc domain and relative to an Fc domain that does not comprise such one or more asymmetric modifications.

[0051] In some embodiments, described herein is a trispecific fusion protein comprising: (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold. In some embodiments, the scaffold comprises a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide. The dimeric Fc domain may be a heterodimeric Fc domain, wherein the amino acid sequence of the first Fc polypeptide differs from the amino acid sequence of the second Fc polypeptide by at least one amino acid residue.

[0052] In some embodiments, the first binding domain is linked to the N-terminus of the first Fc polypeptide and the second binding domain is linked to the N-terminus of the second binding domain.

[0053] In some embodiments, the first binding domain and the second binding domain are each independently a Fab or an scFv. Thus, in some embodiments, a) the first binding domain is a Fab and the second binding domain is an scFv; or b) the first binding domain is an scFv and the second binding domain is a Fab; or c) the first binding domain is a Fab and the second binding domain is a Fab; or d) the first binding domain is an scFv and the second binding domain is an scFv.

[0054] In some embodiments, the first binding domain comprises a first Fc polypeptide followed by a first linker. Fc In some embodiments, the second binding domain is linked to the second Fc polypeptide via a second linker. Fc In some of these embodiments, the first linker Fc, a second linker Fc , or both comprise or consist of an IgG hinge region or a portion or variant thereof. In some embodiments, the first linker Fc , a second linker Fc , or both comprise or consist of an amino acid sequence having at least 80%, 90%, or 95% sequence identity to the amino acid sequence set forth in SEQ ID NO: 50 or a fragment thereof.

[0055] In some embodiments, the first binding domain is a Fab and is linked via its C-terminus to the N-terminus of the first Fc polypeptide. In certain embodiments, the second binding domain is an scFv comprising, from N- to C-terminal, a VH domain linked to a VL domain, or a VL domain linked to a VH domain, and is linked via its C-terminus to the N-terminus of the second Fc polypeptide.

[0056] In various embodiments herein, the trispecific fusion proteins herein comprise a first immunoglobulin G heavy chain, a second immunoglobulin G heavy chain, and an immunoglobulin light chain, wherein the first heavy chain comprises, from N-terminal to C-terminal, a Fab VH and CH1 Domains linked to a first Fc polypeptide; the second heavy chain comprises, from N-terminal to C-terminal, scFv VH and VL or VL and VH Domains linked to a second Fc polypeptide; and the light chain comprises, from N-terminal to C-terminal, a Fab VL and CL Domains.

[0057] In some embodiments, the third binding domain is linked to (i) the first binding domain, (ii) the second binding domain, or (iii) the scaffold. In some embodiments, the third binding domain is linked to (i) the N-terminus of the Fab VH domain, (ii) the N-terminus of the Fab VL domain, (iii) the C-terminus of the CL domain, (iv) the N-terminus of the scFv VH or VL domain, (v) the C-terminus of the first Fc polypeptide, or (vi) the C-terminus of the second Fc polypeptide.

[0058] In some embodiments, the third binding domain comprises a PD-1 polypeptide. In certain embodiments, the third binding domain consists of a PD-1 polypeptide. In some embodiments, the PD-1 polypeptide is a wild-type PD-1 polypeptide comprising or consisting of a portion of the amino acid sequence set forth in SEQ ID NO:7, or a fragment thereof. In other embodiments, the PD-1 polypeptide comprises one or more amino acid modifications relative to a corresponding wild-type PD-1 polypeptide that increase or decrease the binding affinity of the PD-1 polypeptide to PD-L1 relative to the binding affinity of the corresponding wild-type PD-1 polypeptide to PD-L1. In some of these embodiments, the one or more amino acid modifications comprise one or more amino acid substitutions. In some embodiments, the PD-1 polypeptide used in the trispecific fusion protein has a binding affinity for PD-L1 of about 100 μM to about 10 pM, about 10 μM to about 150 pM, or about 100 nM to 150 pM. In some embodiments, the PD-1 polypeptide comprises or consists of an amino acid sequence having at least about 80%, 90%, 95%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO:9, as further described herein.

[0059] In some embodiments, the fusion proteins described herein are trispecific and trivalent, i.e., comprise three antigen-binding domains: one anti-CD3 domain, one anti-TAA domain, and one anti-PD-L1 domain. In other embodiments herein, the fusion proteins are trispecific and tetravalent, i.e., comprise four antigen-binding domains: one anti-CD3 domain, two anti-TAA domains, and two anti-PD-L1 domains.

[0060] In certain embodiments, the trivalent and trispecific fusion protein is not v31929.

[0061] In certain embodiments, the trivalent and trispecific fusion proteins or tetravalent and trispecific fusion proteins herein are capable of binding to a TAA other than HER2, i.e., the TAA is not HER2.

[0062] In some embodiments, the trispecific fusion protein is tetravalent. Thus, in such embodiments, the trispecific fusion protein comprises a fourth binding domain, wherein the fourth binding domain is linked, either directly or via a linker, to the first binding domain, the second binding domain, the third binding domain, or the scaffold. Such a fourth binding domain is capable of binding to the same TAA, thereby allowing the tetravalent fusion protein to bind to the TAA via two binding domains. In some embodiments, the second binding domain and the fourth binding domain capable of binding to the TAA are both scFv domains. In such embodiments, the second binding domain and the fourth binding domain capable of binding to the TAA each comprise or consist of the same anti-TAA VH and VL sequences.

[0063] In some embodiments, described herein are tetravalent and trispecific fusion proteins comprising: (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain and a third binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a first tumor cell; (iii) a fourth binding domain capable of binding to PD-L1 on the surface of a second tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, the third binding domain, and the fourth binding domain are operably linked to the scaffold.

[0064] In some embodiments, described herein is a trispecific fusion protein comprising: (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a TAA on the surface of a first tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a second tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold, and the trispecific fusion protein is not v31929.

[0065] In some embodiments, described herein is a trispecific fusion protein comprising: (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a TAA on the surface of a first tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a second tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold, and the TAA is not HER2.

[0066] In some embodiments, described herein are trispecific fusion proteins comprising: (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a first tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a second tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold, and the third binding domain is linked to the second binding domain.

[0067] In some embodiments, described herein is a trispecific fusion protein comprising: (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a first tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a second tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold, and wherein the third binding domain is linked to the scaffold.

[0068] In some embodiments, described herein are trispecific fusion proteins comprising: (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a first tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a second tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold, and the third binding domain is linked to either (i) the second binding domain or (ii) the scaffold.

[0069] In some embodiments, described herein is a trispecific fusion protein comprising: (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell, where the first binding domain is a Fab domain comprising a Fab heavy chain and a Fab light chain; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a first tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a second tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold, and the third binding domain is linked to (i) the N-terminus or C-terminus of the Fab light chain, (ii) the second binding domain, or (iii) the scaffold.

[0070] In various embodiments, as further described herein, the trispecific fusion protein, when bound to CD3 on a cytotoxic effector cell, a TAA on a tumor cell, and PD-L1 on a tumor cell, forms a TCR-independent immune synapse that is capable of inducing effector cell-mediated cytotoxicity against tumor cells.

[0071] In certain embodiments, the trispecific fusion protein binds to a TAA and PD-L1 on the same tumor cell. In other embodiments, the trispecific fusion protein binds to a TAA and PD-L1 on different tumor cells.

[0072] III. PD-1 Domain In some embodiments, described herein are trispecific fusion proteins, each of which comprises a PD-1 domain (i.e., as a third binding domain) capable of binding to its cognate or naturally occurring ligand, PD-L1, on tumor cells. PD-1 and PD-L1 belong to the immunoglobulin superfamily (IgSF) (Natarajan, Kannan; Mage, Michael G; and Margulies, David H (April 2015) Immunoglobulin Superfamily. In: eLS. John Wiley & Sons, Ltd: Chichester., AF Williams 1, AN Barclay (1988) The Immunoglobulin Superfamily—Domains for Cell Surface Recognition Annu Rev Immunol 6:381-405).

[0073] The immunoglobulin superfamily (IgSF) is divided into commonly found domains based on the core fold of immunoglobulins (Igs). This Ig fold consists of a beta sandwich consisting of seven antiparallel beta strands arranged in two three-stranded and two four-stranded beta sheets. The two beta sandwiches are interconnected via a disulfide bridge between strands B and F. A commonly recognized structural motif in Ig folds is the "Greek Key" motif. Common subgroups of IgSF are the IgV, IgC1, and IgC2 domains. Each member is distinguished based on shared structural features and beta-strand arrangement. The IgC domain contains seven beta strands arranged in two three-stranded and two four-stranded sheets, while the IgV domain contains nine beta strands arranged in two four-stranded and two five-stranded sheets. IgC1 and IgC2 differ in the structural arrangement of their strands. IgSF domains can be found in a wide variety of biologically important proteins, including antigen receptors, immunoglobulins, and immunoregulatory receptors. Surface-exposed residues of the core beta sandwich and loops connecting the beta strands can function as interaction interfaces for antigen recognition, tertiary / quaternary assembly of other structural domains, or receptor / ligand pairs. Because the antigen recognition site of an immunoglobulin (the VH-VL pair in antibodies such as IgG1) comprises a dimer of two IgV domains, a dimer of either the IgSF or IgV domain is structurally compatible to form a steric mask at the antigen recognition site when covalently linked to the N-terminus of the antibody.

[0074] The ligand-receptor pair may be immunomodulatory, e.g., an immune checkpoint, modulating immune cell effector function, modulating T cell receptor signaling, modulating interactions between antigen-presenting cells and effector cells, or a combination thereof. In certain embodiments, the ligand-receptor pair comprises the extracellular portion of an IgSF receptor and its cognate ligand or a receptor-binding fragment thereof. A receptor-binding fragment refers to any polypeptide that specifically binds to the receptor of a ligand-receptor pair, and may be naturally occurring or non-naturally occurring. "Naturally occurring," as used herein and when applied to the subject matter herein, such as a polypeptide, refers to the fact that the subject matter, e.g., a polypeptide, is found in nature. For example, a polypeptide or polynucleotide sequence that is present in an organism (e.g., a mammal or plant), can be isolated from a natural source, and has not been artificially modified in a laboratory, is naturally occurring. In certain embodiments, the ligand-receptor pair may be two interacting protein domains belonging to the immunoglobulin domain superfamily, including a wild-type PD-1 polypeptide. "Non-naturally occurring," as used herein, refers to engineered polypeptide sequences that are structurally similar to IgSF, such as, for example, variants of naturally occurring proteins. Examples of immunoregulatory pairs of ligand-receptor domains belonging to the immunoglobulin superfamily include, but are not limited to, pairs of the B7 / CD28 family (e.g., PD1-PDL1, PD1-PDL2, CTLA4-CD80, CD28-CD80, CD28-CD86, CTLA4-CD86, PDL1-CD80, as well as ICOS-ICOSL, NCR3LG1-NKp30, HHLA2-CD28H, and CD47-SIRPα. CD80 (also known as B7-1), CD86 (B7-2), PDL1 (B7-H1), ICOSL (B7-H2), PDL2 (B7-DC), CD276 (B7-H3), VTCN1 (B7-H4), VISTA (B7-H5), NCR3LG1 (B7-H6), and HHLA2 (B7-H7) belong to the B7 family.The B7 family of proteins is typically considered a ligand and pairs with members of the CD28 family, including CD28, CTLA4, CD28H, NKp30, PD1, and ICOS. (SMWest and XADeng. Considering B7-CD28 as a family through sequence and structure. Exp Biol Med (Maywood) 2019;244(17):1577-1583; doi:10.1177 / 1535370219855970)

[0075] In some embodiments, the PD-1 domain may be the IgV domain of PD-1 (Uniprot ID Q15116, 18-132). The terms "PD-1 domain" and "PD-1 polypeptide" are used interchangeably herein. In certain embodiments, the PD-1 domain may be a fragment of PD-1 that retains the ability to bind to PD-L1. In certain embodiments described herein, the affinity of the PD-1 domain of the trispecific fusion protein for PD-L1 is altered (e.g., increased or decreased) compared to wild-type PD-1. In certain embodiments, the PD-1 domain is engineered to contain a sequence that differs from wild-type PD-1.

[0076] In certain embodiments, the PD-1 domain of the fusion proteins herein comprises one or more mutations that increase the binding affinity of PD-1 for its cognate binding partner, PD-L1. In certain embodiments, the relative binding affinity of the modified PD-1 polypeptides herein is greater than about 1, 1.5, 2, 2.53, 5, 10, 20, 30, 40, 50, 100, 500, 1000, 5,000, 10,000, 50,000, or about 100,000 times greater than the binding affinity of the corresponding wild-type PD-1 for its naturally occurring cognate binding partner, PD-L1.

[0077] In certain embodiments, the PD-1 domain of the fusion proteins herein comprises one or more mutations that reduce its binding affinity to PD-L1. In certain embodiments, the relative binding affinity of PD-1 compared to wild-type PD-L1 is more than 1, 1.5, 2, 2.53, 5, 10, 20, 30, 40, 50, 100, 500, 1000, 5,000, 10,000, 50,000, or 100,000 times less than the binding affinity of the wild-type PD-1 ligand to its naturally occurring cognate binding partner, PD-L1.

[0078] In some embodiments, the PD-1 of the fusion proteins herein has an amino acid sequence corresponding to SEQ ID NO: 7 or 11. In certain embodiments, the PD-1 has an amino acid sequence that is substantially identical to SEQ ID NO: 7 or 11. In certain embodiments, the PD-1 has an amino acid sequence that is about 80%, about 85%, about 90%, or about 95% identical to SEQ ID NO: 7 or 11. In certain embodiments, the PD-1 has an amino acid sequence that is about 96%, 97%, 98%, or 99% identical to SEQ ID NO: 7 or 11. Any PD-1 variant can be used, for example, a high-affinity variant known in the art, such as those provided in R.L. Maute et al., Engineering high-affinity PD-1 variants for optimized immunotherapy and immuno-PET imaging. Proc Natl Acad Sci USA 112, E6506-6514 (2015), WO2016 / 022994A2 or E. Lazar-Molnar et al., Structure-guided development of a high affinity human Programmed Cell Death-1: Implications for tumor immunotherapy EBIOMedicine 17, 30-44 (2017) and WO2019 / 241758A1.

[0079] In certain embodiments, the PD-1 polypeptide of the fusion proteins herein has an affinity for PD-L1 of about 100 μM to about 10 pM. In certain embodiments, the PD-1 polypeptide has an affinity for PD-L1 of about 10 μM to about 150 pM. In certain embodiments, the PD-1 polypeptide has an affinity for PD-L1 of about 100 nM to about 150 pM. In certain embodiments, the PD-1 polypeptide has an affinity for PD-L1 of about 10 pM, 50 pM, 100 pM, 150 pM, 200 pM, 400 pM, 500 pM, 600 pM, 700 pM, 800 pM, 900 pM, or about 1 nM. In certain embodiments, the PD-1 polypeptide has an affinity for PD-L1 of about 1 nM, 5 nM, 10 nM, 20 nM, 30 nM, 40 nM, 50 nM, 60 nM, 70 nM, 80 nM, 90 nM or about 100 nM.

[0080] In some embodiments, the PD-1 polypeptide of the trispecific fusion proteins herein has a binding affinity for PD-L1 of between about 1 nM and about 200 nM, between about 10 nM and about 100 nM, between about 20 nM and about 75 nM, or between about 30 nM and about 50 nM.

[0081] IV. Fusion Protein Formats The trispecific fusion proteins described herein can be in many different formats, as shown, for example, in Figure 3. The fusion proteins described herein can be considered to have a modular structure consisting of several IgG-derived domains, e.g., scFv, Fab, Fc domains, etc., operably linked to each other via a scaffold, and further include at least a PD-1 domain fused, either directly or via a peptide linker, to a biologically functional protein (e.g., an antibody or portion thereof), to form the trispecific fusion protein described herein. The biologically functional protein, e.g., an Ig-like portion, then includes at least a first and a second polypeptide chain. For example, either the N-terminus or C-terminus of the PD-1 domain can be fused to the first or second polypeptide of the biologically functional protein, e.g., via a peptide linker. Various non-limiting formats of the trispecific fusion proteins described herein are illustrated in Figure 3. For example, in some embodiments, the PD-1 domain can be fused to the N-terminus of the anti-CD3 light chain of the fusion protein via a peptide linker. In some embodiments, the PD-1 domain may be fused to the N-terminus of the anti-CD3 heavy chain of the fusion protein via a peptide linker. In some embodiments, the PD-1 domain may be fused to the C-terminus of the anti-CD3 light chain of the fusion protein via a peptide linker. In some embodiments, the PD-1 domain may be fused to the C-terminus of the anti-TAA light chain of the fusion protein via a peptide linker.

[0082] In certain embodiments, the VH and VL of the anti-CD3 binding domain are in Fab format. In certain embodiments, the VH and VL of the anti-CD3 binding domain are in scFv format. In certain embodiments, the VH and VL of the anti-TAA binding domain are in Fab format. In certain embodiments, the VH and VL of the anti-TAA binding domain are in scFv format. In certain embodiments, the VH and VL of the anti-CD3 binding domain are in Fab format, and the VH and VL of the anti-TAA binding domain are in scFv format. In some embodiments, the VH and VL of the CD3 antigen binding domain are in Fab format, and the VH and VL of the anti-TAA binding domain are in Fab format. In certain embodiments, a fusion protein comprises multiple anti-TAA antigen binding domains comprising VH and VL, which can be in Fab format or scFv format. In certain embodiments, two anti-TAA binding domains are in scFv format, as further described herein.

[0083] In the embodiments described below and illustrated in Figure 3, the scFv domain comprises a VH and a VL and can be in one of two orientations: (1) VH, VL, from N- to C-terminus; or (2) VL, VH, from N- to C-terminus. The VH and VL regions can be linked by a peptide linker, e.g., GGGGSGGGGSGGGGSGGGGS.

[0084] In some embodiments, for example, as illustrated in Figure 3(A), the fusion protein comprises three polypeptides (1, light chain (LC)), (2, heavy chain (HC)), and (3, heavy chain (HC)) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 VL, anti-CD3 CL; (2) PD1 polypeptide, anti-CD3 VH, anti-CD3 CH1, CH2, CH3; and (3) anti-TAA scFv, CH2, CH3.

[0085] In some embodiments, for example, as illustrated in Figure 3(B), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus, (1) a PD1 polypeptide, an anti-CD3 VL, an anti-CD3 CL; (2) an anti-CD3 VH, an anti-CD3 CH1, CH2, CH3; and (3) an anti-TAA scFv, CH2, CH3.

[0086] In some embodiments, for example, as illustrated in Figure 3(C), the fusion protein comprises four polypeptides (1, LC), (2, HC), (3, HC), and (4, LC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 VL, anti-CD3 CL; (2) PD1 polypeptide, anti-CD3 VH, anti-CD3 CH1, CH2, CH3; (3) anti-TAA VH, CH1, CH2, CH3; and (4) anti-TAA VL, CL.

[0087] In some embodiments, for example, as illustrated in Figure 3(D), the fusion protein comprises four polypeptides (1, LC), (2, HC), (3, HC), and (4, LC) (from left to right), including, from N-terminus to C-terminus, (1) a PD1 polypeptide, an anti-CD3 VL, and an anti-CD3 CL; (2) an anti-CD3 VH, and an anti-CD3 CH1, CH2, and CH3; (3) an anti-TAA VH, CH1, CH2, and CH3; and (4) an anti-TAA VL and CL.

[0088] In some embodiments, for example, as illustrated in Figure 3(E), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 VL, anti-CD3 CL; (2) anti-CD3 VH, anti-CD3 CH1, CH2, CH3, PD-1 polypeptide; and (3) anti-TAA scFv, CH2, CH3.

[0089] In some embodiments, for example, as illustrated in Figure 3(F), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 VL, anti-CD3 CL; (2) anti-CD3 VH, anti-CD3 CH1, CH2, CH3; and (3) anti-TAA scFv, CH2, CH3, PD-1 polypeptides.

[0090] In some embodiments, for example, as illustrated in Figure 3(G), the fusion protein comprises four polypeptides (1, LC), (2, HC), (3, HC), and (4, LC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 VL, anti-CD3 CL; (2) anti-CD3 VH, anti-CD3 CH1, CH2, CH3, PD-1 polypeptide; (3) anti-TAA VH, CH1, CH2, CH3; and (4) anti-TAA VL, CL.

[0091] In some embodiments, for example, as illustrated in Figure 3(H), the fusion protein comprises four polypeptides (1, LC), (2, HC), (3, HC), and (4, LC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 VL, anti-CD3 CL; (2) anti-CD3 VH, anti-CD3 CH1, CH2, CH3; (3) anti-TAA VH, CH1, CH2, CH3 PD-1 polypeptide; and (4) anti-TAA VL, CL.

[0092] In some embodiments, for example, as illustrated in Figure 3(I), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 VL, anti-CD3 CL, PD-1 polypeptide; (2) anti-CD3 VH, anti-CD3 CH1, CH2, CH3; and (3) anti-TAA scFv, CH2, CH3.

[0093] In some embodiments, for example, as illustrated in Figure 3(J), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 VL, anti-CD3 CL; (2) anti-CD3 VH, anti-CD3 CH1, CH2, CH3; and (3) PD1 polypeptide, anti-TAA scFv, CH2, CH3.

[0094] In some embodiments, for example, as illustrated in Figure 3(K), the fusion protein comprises four polypeptides (1, LC), (2, HC), (3, HC), and (4, LC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 VL, anti-CD3 CL, PD-1 polypeptide; (2) anti-CD3 VH, anti-CD3 CH1, CH2, CH3; (3) anti-TAA VH, CH1, CH2, CH3; and (4) anti-TAA VL, CL.

[0095] In some embodiments, for example, as illustrated in Figure 3(L), ​​the fusion protein comprises four polypeptides (1, LC), (2, HC), (3, HC), and (4, LC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 VL, anti-CD3 CL; (2) anti-CD3 VH, anti-CD3 CH1, CH2, CH3; (3) anti-TAA VH, CH1, CH2, CH3; and (4) anti-TAA VL, CL, PD-1 polypeptides.

[0096] In some embodiments, for example, as illustrated in Figure 3(M), the fusion protein comprises four polypeptides (1, LC), (2, HC), (3, HC), and (4) (from left to right), including, from N-terminus to C-terminus, (1) anti-CD3 VL, anti-CD3 CL; (2) anti-CD3 VH, anti-CD3 CH1, CH2, CH3; (3) PD-1 polypeptide, anti-TAA VH, CH1, CH2, CH3; and (4) anti-TAA VL, CL.

[0097] In some embodiments, for example, as illustrated in Figure 3(N), the fusion protein comprises four polypeptides (1, LC), (2, HC), (3, HC), and (4, LC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 VL, anti-CD3 CL; (2) anti-CD3 VH, anti-CD3 CH1, CH2, CH3; (3) anti-TAA VH, CH1, CH2, CH3; and (4) PD-1 polypeptide, anti-TAA VL, CL.

[0098] In some embodiments, for example, as illustrated in Figure 3(O), the fusion protein comprises two polypeptides (1, HC) and (2, HC) (from left to right), comprising, from N-terminus to C-terminus: (1) a PD-1 polypeptide, an anti-CD3 scFv, CH2, CH3; and (2) an anti-TAA scFv, CH2, CH3.

[0099] In some embodiments, for example, as illustrated in Figure 3(P), the fusion protein comprises two polypeptides (1, HC) and (2, HC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 scFv, CH2, CH3; and (2) PD1-polypeptide, anti-TAA scFv, CH2, CH3.

[0100] In some embodiments, for example, as illustrated in Figure 3(Q), the fusion protein comprises four polypeptides (1, LC), (2, HC), (3, HC), and (4, LC) (from left to right), comprising, from N-terminus to C-terminus, (1)) anti-TAA VL, CL, anti-CD3 VL, anti-CD3 CL; (2) PD-1 polypeptide, anti-TAA VH, CH1, anti-CD3 VH, CH1, CH2, CH3; (3) anti-TAA VH, CH1, CH2, CH3; and (4) anti-TAA VL, CL.

[0101] In some embodiments, for example, as illustrated in Figure 3(R), the fusion protein comprises four polypeptides (1), (2), (3), and (4) (from left to right), including, from N-terminus to C-terminus, (1) a PD-1 polypeptide, an anti-TAA VL, CL, an anti-CD3 VL, and an anti-CD3 CL; (2) an anti-TAA VH, CH1, an anti-CD3 VH, CH1, CH2, and CH3; (3) an anti-TAA VH, CH1, CH2, and CH3; and (4) an anti-TAA VL and CL.

[0102] In some embodiments, for example, as illustrated in Figure 3(S), the fusion protein comprises five polypeptides (1, LC), (2, LC), (3, HC), (4, HC), and (5, LC) (from left to right), including, from N-terminus to C-terminus, (1) anti-TAA VL, CL1, PD-1 polypeptide; (2) anti-CD3 VL, anti-CD3 CL; (3) anti-TAA VH, CH1, anti-CD3 VH, CH1, CH2, CH3; (4) anti-TAA VH, CH1, CH2, CH3; and (5) anti-TAA VL, CL.

[0103] In some embodiments, for example, as illustrated in Figure 3(T), the fusion protein comprises two polypeptides (1, HC) and (2, HC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 scFv, CH2, CH3, PD-1 polypeptide; and (2) anti-TAA scFv, CH2, CH3.

[0104] In some embodiments, for example, as illustrated in Figure 3(U), the fusion protein comprises two polypeptides (1, HC) and (2, HC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 scFv, CH2, CH3; and (2) anti-TAA scFv, CH2, CH3, PD-1 polypeptide.

[0105] In some embodiments, for example, as illustrated in Figure 3(V), the fusion protein comprises five polypeptides (1, LC), (2, LC), (3, HC), (4, HC), and (5, LC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-TAA VL, CL; (2) PD-1 polypeptide, anti-CD3 VL, CL; (3) anti-TAA VH, CH1, anti-CD3 VH, CH1, CH2, CH3; (4) anti-TAA VH, CH1, CH2, CH3, and (5) anti-TAA VL, CL.

[0106] In some embodiments, for example, as illustrated in Figure 3(W), the fusion protein comprises four polypeptides (1, LC), (2, HC), (3, HC), and (4, LC) (from left to right), comprising, from N-terminus to C-terminus: (1) anti-TAA VL, CL, anti-CD3 VL, CL, PD-1 polypeptide; (2) anti-TAA VH, CH1, anti-CD3 VH, CH1, CH2, CH3; (3) anti-TAA VH, CH1, CH2, CH3, and (4) anti-TAA VL, CL.

[0107] In some embodiments, for example, as illustrated in Figure 3(X), the fusion protein comprises four polypeptides (1, LC), (2, HC), (3, HC), and (4, LC) (from left to right), including, from N-terminus to C-terminus, (1) anti-TAA VL, CL, anti-CD3 VL, CL; (2) anti-TAA VH, CH1, anti-CD3 VH, CH1, CH2, CH3; (3) PD-1 polypeptide, anti-TAA VH, CH1, CH2, CH3, and (4) anti-TAA VL, CL.

[0108] In some embodiments, for example, as illustrated in Figure 3(Y), the fusion protein comprises four polypeptides (1, LC), (2, HC), (3, HC), and (4, LC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-TAA VL, CL, anti-CD3 VL, CL; (2) anti-TAA VH, CH1, anti-CD3 VH, CH1, CH2, CH3; (3) anti-TAA VH, CH1, CH2, CH3, and (4) PD-1 polypeptide, anti-TAA VL, CL.

[0109] In some embodiments, for example, as illustrated in Figure 3(Z), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus: (1) anti-CD3 VL, CL; (2) anti-TAA scFv, anti-CD3 VH, CH1, CH2, CH3; and (3) PD-1 polypeptide, anti-TAA CH2, CH3.

[0110] In some embodiments, for example, as illustrated in Figure 3(AA), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus: (1) anti-CD3 VL, CL; (2) anti-TAA scFv, anti-CD3 VH, CH1, CH2, CH3; and (3) PD-1 polypeptide, anti-TAA scFv, anti-TAA CH2, CH3.

[0111] In some embodiments, for example, as illustrated in Figure 3(BB), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus: (1) a PD-1 polypeptide, anti-CD3 VL, CL; (2) an anti-TAA scFv, anti-CD3 VH, CH1, CH2, CH3; and (3) an anti-TAA scFv, CH2, CH3.

[0112] In some embodiments, for example, as illustrated in Figure 3(CC), the fusion protein comprises three polypeptides (1), (2), and (3) (from left to right), including, from N-terminus to C-terminus: (1) anti-CD3 VL, CL; (2) PD-1 polypeptide, anti-CD3 VH, CH1, CH2, CH3, anti-TAA scFv; and (3) anti-TAA scFv, CH2, CH3.

[0113] In some embodiments, for example, as illustrated in Figure 3(DD), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus: (1) anti-CD3 VL, CL; (2) anti-CD3 VH, CH1, CH2, CH3, anti-TAA scFv; and (3) PD-1 polypeptide, anti-TAA scFv, CH2, CH3.

[0114] In some embodiments, for example, as illustrated in Figure 3(EE), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 VL, CL; PD-1 polypeptide; (2) anti-CD3 VH, CH1, CH2, CH3, anti-TAA scFv; (3) anti-TAA scFv, CH2, CH3.

[0115] In some embodiments, for example, as illustrated in Figure 3 (FF), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 VL, CL; (2) anti-TAA scFv, anti-CD3 VH, CH1, CH2, CH3; and (3) anti-TAA scFv, CH2, CH3, PD-1 polypeptide.

[0116] In some embodiments, for example, as illustrated in Figure 3 (GG), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus: (1) anti-CD3 VL, CL; (2) anti-TAA scFv, anti-CD3 VH, CH1, CH2, CH3, PD-1 polypeptide; and (3) anti-TAA scFv, CH2, CH3.

[0117] In some embodiments, for example, as illustrated in Figure 3(HH), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), including, from N-terminus to C-terminus: (1) anti-CD3 VL, CL, PD-1 polypeptide; (2) anti-TAA scFv, anti-CD3 VH, CH1, CH2, CH3; and (3) anti-TAA scFv, CH2, CH3.

[0118] In some embodiments, for example, as illustrated in Figure 3(II), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus: (1) anti-CD3 VL, CL; (2) PD-1 polypeptide, anti-CD3 VH, CH1, CH2, CH3; and (3) anti-TAA scFv, CH2, CH3, anti-CD3 scFv.

[0119] In some embodiments, for example, as illustrated in Figure 3(JJ), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus, (1) anti-CD3 VL, CL; (2) anti-CD3 VH, CH1, CH2, CH3, anti-TAA scFv; and (3) anti-TAA scFv, CH2, CH3, PD-1 polypeptide.

[0120] In some embodiments, for example, as illustrated in Figure 3(KK), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), including, from N-terminus to C-terminus: (1) a PD-1 polypeptide, anti-CD3 VL, CL; (2) an anti-CD3 VH, CH1, CH2, CH3, anti-TAA scFv; and (3) an anti-TAA scFv, CH2, CH3.

[0121] In some embodiments, for example, as illustrated in Figure 3(LL), the fusion protein comprises three polypeptides (1, LC), (2, HC), and (3, HC) (from left to right), comprising, from N-terminus to C-terminus: (1) anti-CD3 scFv, CH2, CH3; (2) PD-1 polypeptide, anti-TAA VH, CH1, CH2, CH3; (3) anti-TAA VL, CL.

[0122] In some embodiments, the fusion protein is trivalent and trispecific and the TAA is HER2.

[0123] In various embodiments herein, the amino acid sequences of the trispecific fusion proteins may be identified by clone numbers, which are further specified and defined herein.

[0124] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 22080 and 23734, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 21490, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 12985.

[0125] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 22080, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 21490, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 12985.

[0126] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 23734, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 21490, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 12985.

[0127] In some embodiments, the fusion protein is trivalent and trispecific and the TAA is MSLN.

[0128] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29207, 29208, 29276, 29238, 29282, 22080 or 23734; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 23867, 23270, 29275 or 25095; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412, 23570 or 12985.

[0129] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29207, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29275, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0130] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29208, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in v29275, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0131] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29276, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 25095, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 23570.

[0132] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29238, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29275, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0133] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 22080, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 23867, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 12985.

[0134] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 23734, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 23867, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 12985.

[0135] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29282, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 23270, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0136] In some embodiments, the fusion protein is trivalent and trispecific and the TAA is Cldn18.2.

[0137] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29241, 29238, 29208 or 29211; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29264, 29261, 29267 or 28373; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0138] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29241, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 28373, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0139] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29238, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 28373, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0140] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29208, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 28373, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0141] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29211, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 28373, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0142] In some embodiments, the trispecific fusion protein is tetravalent and the TAA is MSLN.

[0143] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29257 or 29283, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 23867, 29258, 29264, 23867, 29263, 29267 or 29261, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412, 29226 or 29220.

[0144] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29257, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 23867, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29220.

[0145] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29283, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29258, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0146] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29283, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29264, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0147] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29283, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 23867, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29220.

[0148] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29283, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 23867, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29226.

[0149] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29283, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29263, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0150] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29283, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29264, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0151] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29283, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29261, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0152] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29283, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29267, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0153] In some embodiments, the trispecific fusion protein is tetravalent and the TAA is Cldn18.2.

[0154] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29244, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29245, 29248, 29251 or 29254, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0155] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29244, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29245, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0156] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29244, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29248, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0157] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29244, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29251, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in v16412.

[0158] In some embodiments, the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29244, (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29254, and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412.

[0159] In some embodiments, a trispecific fusion protein of the present disclosure does not comprise a first binding domain that is a Fab capable of binding CD3, a second binding domain that is an scFv capable of binding HER2, and a PD-1 polypeptide, wherein (i) the PD-1 polypeptide is linked to the N-terminus of the anti-CD3 Fab domain, (ii) the anti-CD3 Fab domain is linked to the N-terminus of the first Fc polypeptide of the heterodimeric Fc domain, and (iii) the anti-HER2 scFv domain is linked to the second Fc polypeptide of the heterodimeric Fc domain.

[0160] In some embodiments, the trispecific fusion protein consists of three polypeptide chain clones 22080 (heavy chain (H)1), 12985 (light chain (L)1), and 21490 (heavy chain, H2), which have the amino acid sequences set forth in SEQ ID NO:141, SEQ ID NO:102, and SEQ ID NO:119, respectively, and are not v31929.

[0161] In some embodiments, the trispecific fusion protein is v38449, i.e., tetravalent, and comprises or consists of (ii) a first heavy chain comprising the clone sequence described in 29283, (iii) a second heavy chain comprising the clone sequence described in 29258, and (iv) a light chain comprising the clone sequence described in 16412.

[0162] In some embodiments, the trispecific fusion protein is v38917, i.e., tetravalent, and comprises or consists of (ii) a first heavy chain comprising the clone sequence described in 29283, (iii) a second heavy chain comprising the clone sequence described in 29261, and (iv) a light chain comprising the clone sequence described in 16412.

[0163] In some embodiments, the trispecific fusion protein is v38410, i.e., trivalent, and comprises or consists of (ii) a first heavy chain comprising the clone sequence described in 29283, (iii) a second heavy chain comprising the clone sequence described in 28373, and (iv) a light chain comprising the clone sequence described in 16412.

[0164] In some embodiments, the trispecific fusion protein is v38729, i.e., trivalent, and comprises or consists of (ii) a first heavy chain comprising the clone sequence described in 29209, (iii) a second heavy chain comprising the clone sequence described in 28373, and (iv) a light chain comprising the clone sequence described in 16412.

[0165] In some embodiments, the trispecific fusion protein is v38999, i.e., tetravalent, and comprises or consists of (ii) a first heavy chain comprising the clone sequence described in 29244, (iii) a second heavy chain comprising the clone sequence described in 29245, and (iv) a light chain comprising the clone sequence described in 16412.

[0166] In some embodiments, the trispecific fusion protein is v39000, i.e., tetravalent, and comprises or consists of: (i) a first heavy chain comprising the clone sequence described in 29244, (ii) a second heavy chain comprising the clone sequence described in 29248, and (iii) a light chain comprising the clone sequence described in 16412.

[0167] In some embodiments, the trispecific fusion protein is v39003, i.e., tetravalent, and comprises or consists of (ii) a first heavy chain comprising the clone sequence described in 29244, (iii) a second heavy chain comprising the clone sequence described in 29251, and (iv) a light chain comprising the clone sequence described in 16412.

[0168] In some embodiments, the trispecific fusion protein is v39004, i.e., tetravalent, and comprises or consists of (ii) a first heavy chain comprising the clone sequence described in 29244, (iii) a second heavy chain comprising the clone sequence described in 29254, and (iv) a light chain comprising the clone sequence described in 16412.

[0169] In some embodiments, the trispecific fusion proteins herein are trivalent and capable of binding to HER2 and are selected from the group consisting of v38400, v38403, v38404, v38405, v31927, v31928, and v38407.

[0170] In some embodiments, the trispecific fusion proteins herein are trivalent, capable of binding to MSLN, and selected from the group consisting of v38520, v38440, v38441, v38442, v38443, v38344, v38345, v38910, v38911, v38913, and v38914. In other embodiments, the trispecific fusion proteins herein are tetravalent, capable of binding to MSLN, and selected from the group consisting of v38448, v38449, v38450, v38451, v38452, v38915, v38919, v38921, and v38922.

[0171] In some embodiments, the trispecific fusion proteins herein are trivalent, capable of binding to Cldn18.2, and selected from the group consisting of v38408, v38409, v38410, v38411, v38522, v38729, v38731, v38732, v38733, v38735, v38736, v38737, v38739, and v38740. In other embodiments, the trispecific fusion proteins herein are tetravalent, capable of binding to Cldn18.2, and selected from the group consisting of v38412, v38413, v38414, v38415, v38416, v38741, v38743, v38744, v38917, v38918, v38999, v39000, v39003, v39004, and v39007.

[0172] In some embodiments, the trispecific fusion proteins of the present disclosure have an EC of about 0.01 pM to about 3 pM, about 0.02 pM to about 2.5 pM, about 0.0.5 pM to about 1 pM, or about 0.02 pM to about 1 pM for killing tumor cells in a TDCC assay. 50 In some embodiments, the trispecific fusion proteins of the present disclosure have an EC value that is about 1,500-fold to about 180,000-fold higher than the corresponding bispecific construct having the same format but not including an anti-PD-L1 moiety (e.g., a PD-1 polypeptide described herein). 50 In some embodiments, the trispecific fusion protein is trivalent or tetravalent and comprises a PD-1 polypeptide having an amino acid sequence having at least about 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO:9 or 10.

[0173] The sequences of each polypeptide chain (ie, clone) that may comprise or consist of some of the fusion proteins herein are shown in Table X1 and in the sequence listing herein.

[0174] In certain embodiments, the fusion protein is conjugated to another therapeutic and / or diagnostic moiety, eg, a chemotherapeutic agent, or a radioisotope.

[0175] V. Biologically Functional Proteins In certain embodiments, the biologically functional proteins of the trispecific fusion proteins described herein comprise at least one antigen-binding domain. The binding domain can be, for example, an immunoglobulin-based binding domain or a non-immunoglobulin-based antibody mimetic, or other polypeptide or small molecule, such as a natural or artificial ligand, capable of specifically binding to their target. Non-immunoglobulin-based antibody mimetic formats include, for example, anticalins, finomers, affimers, alphabodies, DARPins, and avimers.

[0176] The fusion proteins described herein comprise or consist of biologically functional proteins. Examples of biologically functional proteins include, but are not limited to, antibodies and antibody-derived molecules, such as polypeptides comprising antigen-binding domains and polypeptide scaffolds, such as dimeric Fc domains. Thus, in certain embodiments, one or more polypeptide chains, e.g., the first and second polypeptides of the biologically functional proteins described herein, are polypeptides comprising antibody variable and / or constant domains, or other domains that confer antigen-binding or scaffolding functions to the fusion protein.

[0177] antibody In certain embodiments, the biologically functional protein is an antibody, i.e., an immunoglobulin. Antibodies according to the present disclosure can take a variety of formats, as described herein, including antibody fragments, multivalent and / or multispecific antibodies, etc. Thus, in certain embodiments, the biologically functional protein is a multivalent, multispecific antibody. The terms "antibody" and "immunoglobulin" are used interchangeably herein to refer to a polypeptide encoded by one or more immunoglobulin genes, or modified forms of immunoglobulin genes, which polypeptide, or a portion thereof, specifically binds to an antigen.

[0178] Specific binding of the fusion proteins described herein can be measured, for example, by enzyme-linked immunosorbent assay (ELISA), surface plasmon resonance (SPR) technology (e.g., employing a BIAcore instrument) (Liljeblad et al., 2000, Glyco J, 17:323-329), or traditional binding assays (Heeley, 2002, Endocr Res, 28:217-229). In certain embodiments, specific binding is defined as the extent of binding to unrelated proteins being less than about 10% of the binding to the target antigen, as measured, for example, by SPR. In certain embodiments, specific binding of an antibody or antibody fragment to a particular antigen or epitope is characterized by a dissociation constant (K) of <1 μM, e.g., <100 nM, <10 nM, <1 nM, <0.1 nM, <0.01 nM, or <0.001 nM. D In certain embodiments, the specific binding of an antibody or antibody fragment to a particular antigen or epitope is defined as about 10 -6 M or less, e.g., about 10 -7 M or less, or about 10 -8 The dissociation constant (K D In some embodiments, the specific binding of an antibody or antibody fragment to a particular antigen or epitope is defined as about 10 -6 M ~ about 10 -13 M, for example, about 10 -7 M ~ about 10 -13 M, about 10 -8M ~ about 10 -13 M, or about 10 -9 M ~ about 10 -13 Dissociation constant of M (K D ) is defined by

[0179] A conventional immunoglobulin structural unit typically consists of two pairs of polypeptide chains, each pair having one "light" chain (approximately 25 kilodaltons (kD)) and one "heavy" chain (approximately 50-70 kD). Light chains are classified as either kappa or lambda. The "class" of an immunoglobulin refers to the type of constant domain possessed by its heavy chain. There are five major classes of antibodies: IgA, IgD, IgE, IgG, and IgM, several of which can be further divided into subclasses (isotypes), e.g., IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2. The heavy-chain constant domains corresponding to the different classes of immunoglobulins are called alpha (α), delta (δ), epsilon (ε), gamma (γ), and mu (μ), respectively.

[0180] In certain embodiments, the antibodies described herein are based on an IgG class immunoglobulin, e.g., an IgG1, IgG2, IgG3, or IgG4 immunoglobulin. In some embodiments, the antibodies described herein are based on an IgG1, IgG2, or IgG4 immunoglobulin. In some embodiments, the antibodies described herein are based on an IgG1 immunoglobulin. In the context of the present disclosure, when an antibody is based on a particular immunoglobulin isotype, it means that the antibody comprises all or a portion of the constant region of the particular immunoglobulin isotype. It is understood that antibodies may also, in some embodiments, comprise isotype and / or subclass hybrids.

[0181] The N-terminal domain of each polypeptide chain of an immunoglobulin defines a variable region of about 100-110 amino acids or more in length that is primarily responsible for antigen recognition. The terms variable light chain (VL) and variable heavy chain (VH) refer to these domains in the light and heavy chains, respectively.

[0182] It will therefore be understood that immunoglobulins may comprise different domains within their heavy and light chains. Such domains may overlap and may include an Fc domain (or Fc region), a CH1 domain, a CH2 domain, a CH3 domain, a hinge domain, a heavy chain constant domain (CH1-hinge-Fc or CH1-hinge-CH2-CH3), a variable heavy domain (VH), a variable light domain (VL), and a light chain constant domain (CL). An "Fc domain" comprises the CH2 and CH3 domains, and optionally the hinge domain (or hinge region).

[0183] Each of the VH and VL domains of immunoglobulins contains three loops whose sequences are hypervariable and form the antigen-binding site. Each of these loops is referred to as a "hypervariable region" or "HVR." The terms hypervariable region (HVR) and complementarity-determining region (CDR) are used interchangeably herein to refer to the portions of the variable region that form the antigen-binding domain. Except for CDR1 of VH, CDRs generally contain amino acid residues that form the hypervariable loops. The VH and VL domains consist of multiple relatively invariant stretches called framework regions (FRs), each about 15 to 30 amino acids long, separated by shorter CDRs, each of which is typically about 5 to 15 amino acids long but can sometimes be longer or shorter. The three CDRs and four FRs that make up each of the VH and VL domains are arranged N-terminally to C-terminally as FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4.

[0184] Many different definitions of CDR regions are commonly used, including those described by Kabat et al. (1983, Sequences of Proteins of Immunological Interest, NIH Publication No. 369-847, Bethesda, MD), Chothia et al. (1987, J Mol Biol, 196:901-917), as well as the IMGT, AbM, and Contact definitions. These different definitions include overlapping or subsets of amino acid residues when compared with each other. By way of example, CDR definitions according to Kabat, Chothia, IMGT, AbM, and Contact are set forth in Table 1 below. Thus, as one skilled in the art will readily appreciate, the exact numbering and arrangement of CDRs may vary depending on the numbering system employed. However, it is understood that the disclosure herein of a variable heavy chain domain (VH) includes the disclosure of the associated (unique) heavy chain CDRs (HCDRs) defined by any of the known numbering systems. Similarly, disclosure herein of a variable light chain domain (VL) includes disclosure of the associated (unique) heavy chain CDRs (HCDRs), as defined by any known numbering system.

[0185] (Table 1) TIFF2025504909000002.tif160165

[0186] Those skilled in the art will understand that a limited number of amino acid substitutions can be introduced into the CDR sequences or VH or VL sequences of a known antibody or antibody fragment without destroying the antibody or fragment's ability to bind to its target. Candidate amino acid substitutions can be identified by techniques such as computer modeling or the alanine scanning described above, and the resulting variants are tested for binding activity using standard techniques. For example, in certain embodiments, the EGFR-binding domain contained in the fusion protein comprises a set of CDRs (i.e., heavy chain CDR1, CDR2, and CDR3, and light chain CDR1, CDR2, and CDR3) derived from cetuximab or panitumumab that have 90% or more, 95% or more, 98% or more, 99% or more, or 100% sequence identity, wherein the binding domain retains the ability to bind to EGFR. In certain embodiments, the EGFR-binding domain included in the fusion protein includes variants of these CDR sequences containing 1 to 10 amino acid substitutions across the three CDRs, e.g., 1 to 7 amino acid substitutions, 1 to 5 amino acid substitutions, 1 to 4 amino acid substitutions, 1 to 3 amino acid substitutions, 1 to 2 amino acid substitutions, or 1 amino acid substitution across the CDRs (i.e., the CDRs can be modified by including up to 10 amino acid substitutions in any combination of the modified CDRs), where the variant retains the ability to bind to EGFR. Typically, in some embodiments, such amino acid substitutions can be conservative amino acid substitutions as outlined in column 1 or column 2 of Table 4 below.

[0187] In certain embodiments, antibodies, e.g., trispecific fusion proteins described herein, comprise at least one immunoglobulin domain derived from a mammalian immunoglobulin, e.g., bovine immunoglobulin, human immunoglobulin, camelid immunoglobulin, rat immunoglobulin, or mouse immunoglobulin. In some embodiments, biologically functional proteins, e.g., trispecific fusion proteins described herein, may be chimeric antibodies and comprise two or more immunoglobulin domains, with at least one domain derived from a first mammalian immunoglobulin, e.g., a human immunoglobulin, and at least a second domain derived from a second mammalian immunoglobulin, e.g., a mouse or rat immunoglobulin. In some embodiments, biologically functional proteins, e.g., trispecific fusion proteins described herein, comprise at least one immunoglobulin constant domain derived from a human immunoglobulin.

[0188] Those skilled in the art will appreciate that these antibody domains can be combined in various ways to provide antibodies with different formats, e.g., multispecific antibodies in different formats, such as those described in Figure 3. In some embodiments, these formats are generally based on antibody formats known in the art (see, e.g., the review by Brinkmann & Kontermann, 2017, MABS, 9(2):182-212, and Muller & Kontermann, "Bispecific Antibodies" in Handbook of Therapeutic Antibodies, Wiley-VCH Verlag GmbH & Co. (2014)).

[0189] Antibodies of the biologically functional proteins described herein, such as the trispecific fusion proteins described herein, can have different valencies. In certain embodiments, the biologically functional proteins described herein comprise a single antigen-binding domain. In certain embodiments, the biologically functional proteins described herein comprise two or more antigen-binding domains, e.g., three antigen-binding domains. In certain embodiments, the biologically functional proteins include antibodies with different valencies and specificities. As used herein, a "bispecific antibody" comprises two binding domains. In certain embodiments, each of the two binding domains has a unique binding specificity for an epitope on the antigen. As used herein, a "multispecific antibody" comprises two or more binding domains, e.g., three or four binding domains. In certain embodiments, each of the two or more binding domains has a unique binding specificity for an epitope on the antigen. In some embodiments, at least two of the two or more binding domains have unique binding specificities for two epitopes, which may be on the same antigen or different antigens. For example, antibodies as used herein may be bivalent and bispecific, or bivalent and monospecific, such as, for example, naturally occurring IgG antibodies without modification. Alternatively, antibodies as used herein may be trivalent and bispecific, i.e., the antibody comprises three binding domains, two of which have specificity for different epitopes, e.g., on the same or different antigens. Antibodies may also be bispecific and tetravalent, i.e., the antibody comprises two pairs of binding domains (i.e., four binding domains), each pair specifically binding to a particular epitope. Other valencies are possible, as further described herein.

[0190] When an antibody contains two binding domains that bind to the same target molecule, the binding domains may bind to the same epitope on the target molecule or different epitopes on the target molecule. In some embodiments, an antibody contains two binding domains that bind to the same epitope on the same target molecule (e.g., an antigen such as HER2 or MSLN). In some embodiments, an antibody contains two binding domains that bind to different epitopes on the target molecule. The term "biparatopic" can be used to refer to an antibody that contains two binding domains that bind to different epitopes on the same target molecule (e.g., an antigen such as HER2 or MSLN). A biparatopic antibody may bind to a single antigen molecule via two different epitopes, or it may bind to two separate antigen molecules via different epitopes.

[0191] In certain embodiments, the antibodies described herein are biparatopic for the first antigen and monoparatopic for the second antigen, e.g., trivalent and trispecific, in that they comprise a first binding domain and a second binding domain that each bind to a different epitope on the first target molecule (biparatopic binding), and a third binding domain that binds to the second target molecule (monoparatopic binding). Alternatively, a tetraspecific biparatopic antibody may comprise a first binding domain and a second binding domain that each bind to a different epitope on the first target molecule (biparatopic binding), and a third binding domain and a fourth binding domain that each bind to a different epitope on the second target molecule (biparatopic binding).

[0192] In some embodiments, the antibodies described herein further comprise a scaffold, and the binding domains are operably linked to the scaffold. "Operably linked," as used herein, means that the described components are in a relationship that allows each to function in its intended manner. The binding domains can be directly or indirectly linked to the scaffold. "Indirectly linked" means that a given binding domain is linked to the scaffold via another component, such as a linker or one of the other binding domains. Various formats of fusion proteins comprising scaffolds are described in more detail below.

[0193] VI. Antigen-binding domain formats In some embodiments, the fusion proteins described herein comprise an antibody having at least one antigen-binding domain that is an antibody fragment, such as a Fab, Fab', single-chain Fab (scFab), single-chain Fv (scFv), single-domain antibody (sdAb), or a combination thereof. In various embodiments, the trispecific fusion proteins of the present disclosure comprise three binding domains, wherein at least one of such domains is a Fab domain and at least one other domain is an scFv domain, as described herein.

[0194] As used herein, "Fab" or "Fab fragment" contains the light chain and heavy chain variable domains VL and VH, respectively, containing the CDRs, as well as the light chain constant domain (CL) and the first heavy chain constant domain (CH1). Fab' or Fab' fragment differs from a Fab fragment in that it has several additional amino acid residues, including one or more cysteine ​​residues from the hinge region, at the C-terminus of the heavy chain CH1 domain.

[0195] A Fab fragment can comprise or consist of two individual polypeptide chains (light and heavy chains), or can be a single-chain Fab. A single-chain Fab is a Fab molecule in which the Fab light chain and the Fab heavy chain are connected by a peptide linker to form a single peptide chain. Typically, in a single-chain Fab molecule, the C-terminus of the Fab light chain is connected to the N-terminus of the Fab heavy chain, although other formats are possible. In various embodiments of the present disclosure, the Fab of the trispecific fusion protein is not a single-chain Fab, but comprises a light chain (CL + VL sequence) and a heavy chain (CH1 + VH sequence) as described herein.

[0196] As used herein, an "scFv" or "scFv domain" refers to a single polypeptide chain comprising an antibody heavy chain variable domain (VH) and light chain variable domain (VL). The scFv of a trispecific fusion protein can optionally comprise a polypeptide linker between the VH and VL domains, which can assist the scFv in forming the desired structure for antigen binding. An scFv can comprise a VL connected from its C-terminus to the N-terminus of the VH by a linker, i.e., VL-linker-VH (N-terminus to C-terminus), or a scFv can comprise a VH connected via its C-terminus to the N-terminus of the VL by a linker, i.e., VH-linker-VL (N-terminus to C-terminus). For a review of scFvs, see Pluckthun in The Pharmacology of Monoclonal Antibodies, vol. 113, Rosenburg and Moore eds., Springer-Verlag, New York, pp. 269-315 (1994). In some embodiments, the linker connecting the VL and VH sequences of the scFv used in the fusion proteins herein can be a peptide linker. Such a peptide linker can comprise or consist of a sequence of about 5, 10, 15, 20, 25, or more consecutive amino acid residues. In some embodiments, such a linker can be a peptide linker having the sequence (G a S b)x, where a, b, and x are independently integers from 1 to 5. In some embodiments, such a linker comprises or consists of the amino acid sequence (GS)x, where x is an integer from 1 to 5. In some embodiments, such a linker comprises or consists of the amino acid sequence set forth in SEQ ID NO:284.

[0197] The term "sdAb" refers to a single immunoglobulin domain. sdAbs can be derived, for example, from camelids. Camelid antibodies lack light chains and their antigen-binding site consists of a single domain called "VHH". sdAbs form three CDR / hypervariable loops that form the antigen-binding site: CDR1, CDR2 and CDR3. sdAbs are fairly stable and are easily expressed, for example, as fusions with the Fc chain of an antibody (see, e.g., Harmsen & De Haard, 2007, Appl. Microbiol Biotechnol. 77(1):13-22).

[0198] In some embodiments, one or more of the binding domains included in the antibodies and trispecific fusion proteins described herein can be a natural (e.g., wild-type amino acid sequence) or artificial (e.g., a sequence having one or more amino acid modifications relative to the wild-type amino acid sequence) ligand or binding domain for a target receptor (e.g., an antigen), or a functional fragment of such a ligand, i.e., a fragment capable of specifically binding to a target receptor (e.g., an antigen).

[0199] In some embodiments, the antigen-binding domains of the trispecific fusion proteins described herein may be in the form of individual combinations of scFvs, Fabs, sdAbs, etc. For example, when the binding domains are in the form of scFvs, formats such as tandem scFvs ((scFv)2 or taFv) or triplebody (three scFvs) can be constructed, in which the scFvs are connected together by a flexible linker. ScFvs can also be used to construct diabody, triabody, and tetrabody (tandem diabody or TandAb) formats, which contain two, three, and four scFvs, respectively, connected by short linkers. Due to the limited length of the linker (usually about 5 amino acids long), the scFvs dimerize head-to-tail. In any of the aforementioned formats, the scFvs can be further stabilized by including interdomain disulfide bonds. For example, by introducing additional cysteine ​​residues into each chain (e.g., at position 44 in the VH and position 100 in the VL), a disulfide bond can be introduced between the VL and VH (see, e.g., Fitzgerald et al., 1997, Protein Engineering, 10:1221-1225), or a disulfide bond can be introduced between two VHs to provide an antigen-binding domain with a DART format (see, e.g., Johnson et al., 2010, J Mol. Biol., 399:436-449).

[0200] Similarly, fusion protein formats comprising two or more sdAbs, such as VH or VHH, connected to each other by a suitable linker can also be used for biologically functional proteins. Other examples of scaffold-less antibody formats include those based on Fab fragments, such as the Fab2, F(ab')2 and F(ab')3 formats, where the Fab fragments are connected by a linker or IgG hinge region.

[0201] Combinations of different forms of antigen-binding domains can also be employed to generate alternative formats. For example, an scFv or sdAb can be fused to the C-terminus of either or both of the light and heavy chains of a Fab fragment, resulting in a bivalent (Fab-scFv) or (Fab-sdAb) or trivalent (Fab-(scFv)2 or Fab-(sdAb)2). Similarly, one or two scFvs or sdAbs can be fused to the hinge region of an F(ab')2 fragment to produce a trivalent or tetravalent F(ab')2-scFv / sdAb. The binding domain can be one or a combination of the above forms (e.g., scFv, Fab and / or sdAb, or ligand-based binding domains).

[0202] In certain specific embodiments, the biologically functional protein of the trispecific fusion proteins herein comprises a bispecific antibody that binds to an immune cell antigen, e.g., CD3, and a tumor-associated antigen (TAA), e.g., HER2. In certain more specific embodiments, the biologically functional protein comprises a bispecific antibody in Fab-scFv format, where the Fab binds to the immune cell antigen and the scFv binds to the TAA. In certain more specific embodiments, the biologically functional protein comprises a bispecific antibody in Fab-scFv format, where the Fab binds to CD3 and the scFv binds to HER2. In some embodiments, the biologically functional protein comprises a bispecific antibody in Fab-Fab format, where one Fab binds to CD3 and the other Fab binds to HER2.

[0203] In certain embodiments, a biologically functional protein comprises two or more antigen-binding domains operably linked to a heterodimeric Fc. In this context, a biologically functional protein can be bivalent, trivalent, or tetravalent. Non-limiting examples of formats are described below and further described herein, e.g., in Figure 3. Other structures are known in the art (see, e.g., Spiess et al., 2015, Mol Immunol., 67:95-106).

[0204] Exemplary structures of the biologically functional proteins herein, i.e., bivalent antibodies, comprising two binding domains operably linked to a heterodimeric Fc include, but are not limited to: a) a mAb format, in which the first binding domain is a Fab operably linked to the N-terminus of a first Fc polypeptide of the heterodimeric Fc, and the second binding domain is a Fab operably linked to the N-terminus of a second Fc polypeptide; b) a hybrid format, in which the first binding domain is an scFv operably linked to the N-terminus of one Fc polypeptide of the heterodimeric Fc, and the second binding domain is a Fab operably linked to the N-terminus of the other Fc polypeptide; and c) a dual scFv format, in which the first binding domain is an scFv operably linked to the N-terminus of a first Fc polypeptide of the heterodimeric Fc, and the second binding domain is an scFv operably linked to the N-terminus of a second Fc polypeptide.

[0205] Other examples include antibodies comprising one binding domain (either first or second) that is a Fab or scFv operably linked to the N-terminus of a first Fc polypeptide and another binding domain that is a Fab or scFv operably linked to the C-terminus of a second Fc polypeptide.

[0206] Exemplary structures of a multispecific antibody (e.g., a fusion protein) comprising three binding domains operably linked to a heterodimeric Fc (i.e., a trivalent antibody that may be mono-, bi-, or trispecific) include, but are not limited to, the following: A) a first binding domain is a Fab operably linked to the N-terminus of a first Fc polypeptide of the heterodimeric Fc, a second binding domain is a Fab operably linked to the N-terminus of a second Fc polypeptide, and a third binding domain is attached to the C-terminus of one of the Fc polypeptides. A) a mAb-Fv format, in which the first binding domain is a Fab operably linked to the N-terminus of a first Fc polypeptide of the heterodimeric Fc, the second binding domain is a Fab operably linked to the N-terminus of a second Fc polypeptide, and the third binding domain is an scFv operably linked to the C-terminus of either the first or second Fc polypeptide; B) a mAb-scFv format, in which the first binding domain is a Fab operably linked to the N-terminus of a first Fc polypeptide of the heterodimeric Fc, the second binding domain is a Fab operably linked to the N-terminus of a second Fc polypeptide, and the third binding domain is an scFv operably linked to the C-terminus of either the first or second Fc polypeptide; C) a mAb-scFv format, in which the first binding domain is a Fab operably linked to the N-terminus of a first Fc polypeptide of the heterodimeric Fc. A) an scFv-mAb format, in which the first binding domain is a Fab operably linked to the N-terminus of a first Fc polypeptide of the heterodimeric Fc, the second binding domain is a Fab operably linked to the N-terminus of a second Fc polypeptide, and the third binding domain is an scFv operably linked to the N-terminus of either the first or second binding domain; B) an scFv-mAb format, in which the first binding domain is a scFv operably linked to the N-terminus of one Fc polypeptide of the heterodimeric Fc, and the second binding domain is an scFv operably linked to the N-terminus of the other Fc polypeptide. a central scFv format in which the first binding domain is an scFv operably linked to the N-terminus of one Fc polypeptide of a heterodimeric Fc, the second binding domain is a Fab operably linked to the N-terminus of the other Fc polypeptide, and the third binding domain is a Fab operably linked to the N-terminus of the first or second binding domain; E) a Fab hybrid format in which the first binding domain is an scFv operably linked to the N-terminus of one Fc polypeptide of a heterodimeric Fc, the second binding domain is a Fab operably linked to the N-terminus of the other Fc polypeptide, and the third binding domain is a Fab operably linked to the N-terminus of the first or second binding domain;F) an scFv hybrid format in which the first binding domain is an scFv operably linked to the N-terminus of one Fc polypeptide of the heterodimeric Fc, the second binding domain is a Fab operably linked to the N-terminus of the other Fc polypeptide, and the third binding domain is an scFv operably linked to the N-terminus of the first or second binding domain; G) a hybrid scFv format in which the first binding domain is an scFv operably linked to the N-terminus of one Fc polypeptide of the heterodimeric Fc, the second binding domain is a Fab operably linked to the N-terminus of the other Fc polypeptide, and the third binding domain is an scFv operably linked to the C-terminus of the first or second Fc polypeptide; H a) a hybrid Fab format in which the first binding domain is an scFv operably linked to the N-terminus of one Fc polypeptide of the heterodimeric Fc, the second binding domain is a Fab operably linked to the N-terminus of the other Fc polypeptide, and the third binding domain is a Fab operably linked to the C-terminus of the first or second Fc polypeptide; and I) a Fab-mAb format in which the first binding domain is a Fab operably linked to the N-terminus of the first Fc polypeptide of the heterodimeric Fc, the second binding domain is a Fab operably linked to the N-terminus of the second Fc polypeptide, and the third binding domain is a Fab operably linked to the N-terminus of either the first or second binding domain;

[0207] In various embodiments, the trivalent fusion proteins of the present disclosure are trispecific. Such trivalent trispecific fusion proteins herein may comprise a first binding domain (e.g., a Fab or scFv domain) capable of binding to an antigen (e.g., CD3) on a cytotoxic effector cell (e.g., an immune cell such as a T cell), a second binding domain (e.g., a Fab or scFv domain) capable of binding to a TAA on a tumor cell, and a third binding domain (e.g., a PD-1 polypeptide) capable of binding to PD-L1 on a tumor cell.

[0208] Exemplary structures of multispecific antibodies (e.g., fusion proteins) comprising four binding domains operably linked to a heterodimeric Fc, i.e., tetravalent antibodies, which may be mono-, bi-, or trispecific, include, but are not limited to: i) a first binding domain is an scFv operably linked to the N-terminus of one Fc polypeptide of the heterodimeric Fc, a second binding domain is an scFv operably linked to the N-terminus of the other Fc polypeptide, and a third binding domain is an Fab operably linked to one of the scFvs; and ii) a central scFv2 format, in which the first binding domain is a Fab operably linked to the N-terminus of one Fc polypeptide of a heterodimeric Fc, the second binding domain is a Fab operably linked to the N-terminus of the other Fc polypeptide, the third binding domain is an scFv operably linked to one of the Fabs, and the fourth binding domain is an scFv operably linked to the other Fab. In various other embodiments herein, a tetravalent fusion protein comprises a first binding domain that is a Fab domain, a second binding domain that is a first scFv domain, a third binding domain that is a second scFv domain, and a fourth binding domain. In some embodiments, the fourth binding domain is a Fab or scFv domain. In other embodiments, the fourth binding domain may be a polypeptide chain having a non-Fab or non-scFv structure and specificity for a particular antigen. In various embodiments, the fourth binding domain is a PD-1 polypeptide, either a wild-type PD-1 protein or an engineered (e.g., mutated and / or truncated) variant thereof.

[0209] The biologically functional proteins described herein, such as antibodies of fusion proteins, can contain a label, a drug, or a combination thereof. Any label known in the art that is suitable for detecting the fusion proteins described herein can be used. Antibody-drug conjugates are described in more detail below.

[0210] In certain embodiments, the antigen-binding domains of antibodies of the biologically functional proteins described herein bind to the same antigen on the same cell. In certain embodiments, the antigen-binding domains bind to multiple antigens on the same cell. In certain embodiments, the antigen-binding domains bind to multiple antigens, where at least one antigen is on a different cell than the other antigens. In certain embodiments, the antigen-binding domain(s) of the antibody bind to a tumor cell or an immune cell. In certain embodiments, the antigen-binding domain of the antibody binds to a tumor cell and an immune cell.

[0211] Chimeric, humanized and variant antibodies In some embodiments, the antibodies of the fusion proteins described herein can be derived from immunoglobulins from different species; for example, the antibodies can be chimeric or humanized antibodies, as described herein. A "chimeric antibody" typically refers to an antibody that contains at least one variable domain derived from a rodent antibody (usually a mouse antibody) and at least one constant domain derived from a human antibody. A "humanized antibody" is a type of chimeric antibody that contains minimal sequence derived from a non-human antibody.

[0212] The human constant domains of a chimeric antibody need not be of the same isotype as the non-human constant domains they replace. Chimeric antibodies are discussed, for example, in Morrison et al., 1984, Proc. Natl. Acad. Sci. USA, 81:6851-55, and U.S. Patent No. 4,816,567. Generally, a humanized antibody is a human immunoglobulin (recipient antibody) in which residues from a hypervariable region of the recipient are replaced by residues from a hypervariable region of a non-human species (donor antibody), such as mouse, rat, rabbit, or non-human primate, that has the desired specificity and affinity for the target antigen. This technique for producing humanized antibodies is often referred to as "CDR grafting." Both "chimeric antibodies" and "humanized antibodies" generally refer to antibodies that combine immunoglobulin regions or domains from multiple species.

[0213] In some cases, additional modifications are made to further refine antibody performance. For example, framework region (FR) residues of the human immunoglobulin may be replaced by corresponding non-human residues, or the humanized antibody may contain residues that are not found in either the recipient antibody or the donor antibody. Generally, the variable domains of humanized antibodies will contain all or substantially all of the hypervariable regions from the non-human immunoglobulin and all or substantially all of the FRs from the human immunoglobulin sequence. Humanized antibodies are described in detail in, for example, Jones, et al., 1986, Nature, 321:522-525; Riechmann, et al., 1988, Nature, 332:323-329, and Presta, 1992, Curr. Op. Struct. Biol., 2:593-596.

[0214] Numerous approaches for selecting optimal human frameworks for grafting nonhuman CDRs are known in the art. Early approaches used a limited subset of well-characterized human antibodies, regardless of their sequence identity with the nonhuman antibody providing the CDRs (the "fixed framework" approach). More recent approaches employ variable regions with high amino acid sequence identity with the variable regions of the nonhuman antibody providing the CDRs (the "homology matching" or "best-fit" approach). Another approach is to select fragments of framework sequences within the light or heavy chain variable regions from several different human antibodies. CDR grafting can sometimes result in partial or complete loss of affinity of the grafted molecule for its target antigen. In such cases, affinity can be restored by backmutating some of the human-derived residues to the corresponding nonhuman-derived residues. Methods for preparing humanized antibodies by these approaches are well known in the art (see, e.g., Tsurushita & Vasquez, 2004, Humanization of Monoclonal Antibodies, Molecular Biology of B Cells, 533-545, Elsevier Science (USA); Jones et al., 1986, Nature, 321:522-525; Riechmann et al., 1988, Nature, 332:323-329; Presta et al., 1997, Cancer Res, 57(20):4593-4599).

[0215] Alternatively or in addition to such traditional approaches, more recent techniques can be employed to further reduce the immunogenicity of CDR-grafted humanized antibodies. For example, instead of a human framework with somatic mutation(s), a framework based on a human germline sequence or consensus sequence can be employed as the acceptor human framework. Another technique aimed at reducing the potential immunogenicity of non-human CDRs is to graft only the specificity-determining residues (SDRs). In this approach, only the minimum CDR residues ("SDRs") required for antigen-binding activity are grafted onto a human germline framework. This method improves the "humanness" of humanized antibodies (i.e., similarity to human germline sequences), thereby helping to reduce the risk of immunogenicity of the variable regions. These techniques are described in various publications (see, e.g., Almagro & Fransson, 2008, Front Biosci, 13:1619-1633; Tan, et al., 2002, J Immunol, 169:1119-1125; Hwang, et al., 2005, Methods, 36:35-42; Pelat, et al., 2008, J Mol Biol, 384:1400-1407; Tamura, et al., 2000, J Immunol, 164:1432-1441; Gonzales, et al., 2004, Mol Immunol, 1:863-872, and Kashmiri, et al., 2005, Methods, 36:25-34).

[0216] In certain embodiments, the antibody of the fusion protein herein comprises a humanized antibody sequence, e.g., one or more humanized variable domains. In some embodiments, the antibody of the fusion protein herein is a humanized antibody.

[0217] In certain embodiments, the antigen-binding domain contained in the fusion protein is a substitution variant of a known antibody that contains one or more amino acid substitutions in the CDRs of the parent antibody. In certain embodiments, the substitution variant has altered (e.g., improved) certain biological properties compared to the parent antibody. For example, the substitution variant may have increased or decreased affinity for the target protein or may have reduced immunogenicity. In some embodiments, the substitution variant substantially retains certain biological properties of the parent antibody.

[0218] CDR hotspots are residues encoded by codons that undergo frequent mutation during the somatic maturation process (see, e.g., Chowdhury, 2008, Methods Mol. Biol., 207:179-196). Affinity maturation by secondary library construction and reselection has been described (see, e.g., Hoogenboom et al. in Methods in Molecular Biology, 178:1-37, O'Brien et al., ed., Human Press, Totowa, NJ (2001)).

[0219] Methods for affinity maturation are well known in the art. For example, diversity can be introduced into the variable genes selected for maturation by various techniques, including, for example, error-prone PCR, chain shuffling, or oligonucleotide-directed mutagenesis. A secondary library is then generated and screened to identify any antibody variants with the desired affinity. Another method for introducing diversity involves a CDR-directed approach, in which several CDR residues (e.g., 2, 3, 4, or more residues at a time) are randomized. In many cases, the CDR3 of either the heavy chain or the light chain or both is targeted in a CDR-directed approach. CDR residues involved in antigen binding can be identified, for example, by alanine scanning mutagenesis (see, for example, Cunningham and Wells, 1989, Science, 244:1081-1085) or by computer modeling, which uses a crystal structure of an antigen-antibody complex to identify contact points between the antibody and the antigen.

[0220] In certain embodiments, substitution variants comprise one or more substitutions in one or more CDRs, provided that the substitutions do not substantially reduce the binding ability of the binding domain to its target antigen. For example, substitution variants may comprise one or more conservative substitutions described herein in one or more CDRs that do not substantially reduce binding affinity. In some embodiments, substitution variants comprise one or more amino acid substitutions in a CDR that are not involved in antigen-contacting amino acids. In some embodiments, substitution variants comprise variant VH or VL sequences in which each CDR is unchanged or contains no more than one, two, or three amino acid substitutions.

[0221] Glycosylation variants In certain embodiments, the fusion proteins described herein comprise biologically functional proteins based on IgG Fc with altered native glycosylation. As is known in the art, glycosylation of Fc can be altered to increase or decrease effector function.

[0222] For example, mutation of the conserved asparagine residue at position 297 to alanine, glutamine, lysine, or histidine (i.e., N297A, Q, K, or H) results in a deglycosylated Fc that lacks all effector function (Bolt et al., 1993, Eur. J. Immunol., 23:403-411; Tao & Morrison, 1989, J. Immunol., 143:2595-2601).

[0223] Conversely, removal of fucose from the oligosaccharide linked to heavy chain N297 has been shown to enhance ADCC due to improved binding to FcγRIIIa (see, e.g., Shields et al., 2002, J. Biol. Chem., 277:26733-26740, and Niwa et al., 2005, J. Immunol. Methods, 306:151-160). Such low-fucose antibodies can be produced, for example, in knockout Chinese hamster ovary (CHO) cells lacking fucosyltransferase (FUT8) (Yamane-Ohnuki et al., 2004, Biotechnol. Bioeng., 87:614-622), in the variant CHO cell line Lec13 with reduced ability to attach fucose to N297-linked glycans (International Publication No. WO 03 / 035835), or in other cells that produce defucosylated antibodies (e.g., Li et al., 2006, Nat Biotechnol., 24:210-215; Shields et al., 2002, ibid., and Shinkawa et al., 2003, J. Biol. Chem., 278:3466-3473). In addition, International Publication No. WO2009 / 135181 describes adding a fucose analog to the culture medium during antibody production to inhibit incorporation of fucose into the sugar chains on the antibody.

[0224] Other methods for producing antibodies with little or no fucose on the Fc glycosylation site (N297) are well known in the art, e.g., GlymaX® technology (ProBioGen AG) (see von Horsten et al., 2010, Glycobiology, 20(12):1607-1618 and U.S. Patent No. 8,409,572).

[0225] Other glycosylation variants include those containing bisected oligosaccharides, e.g., variants in which the biantennary oligosaccharides attached to the Fc region of an antibody are bisected by N-acetylglucosamine (GlcNAc). Such glycosylation variants have reduced fucosylation and / or improved ADCC function. See, e.g., International Publication No. WO 2003 / 011878, U.S. Patent No. 6,602,684, and U.S. Patent Application Publication No. US 2005 / 0123546. Useful glycosylation variants also include those containing at least one galactose residue in the oligosaccharide attached to the Fc region, which may have improved ADCC function (see, e.g., International Publication Nos. WO 1997 / 030087, WO 1998 / 58964, and WO 1999 / 22764).

[0226] Polypeptide backbone In certain embodiments, the biologically functional protein of the fusion proteins described herein comprises a polypeptide backbone, which can function, for example, to stabilize or extend the in vivo half-life of a ligand receptor pair.

[0227] In certain embodiments, the biologically functional protein comprises or consists of a dimeric Fc region. In certain embodiments, the first and second polypeptides of the biologically functional protein comprise or consist of a dimeric Fc, where the first polypeptide consists of a first Fc polypeptide and the second polypeptide consists of a second Fc polypeptide, and the first and second Fc polypeptides form a dimeric Fc region. In certain embodiments, the dimeric Fc region is a heterodimeric Fc. Heterodimeric Fc regions are described in more detail herein.

[0228] In certain embodiments, the polypeptide backbone comprises a first and a second polypeptide. In certain embodiments, the PD-1 domain is fused to the first polypeptide (e.g., a first heavy chain) or the second polypeptide (e.g., a second heavy chain) via a peptide linker. Thus, in certain embodiments, the PD-1 domain is fused to the N-terminus of the first polypeptide or the N-terminus of the second polypeptide via a peptide linker. Conversely, in certain embodiments, the PD-1 domain is fused to the C-terminus of the first polypeptide or the second polypeptide via a peptide linker.

[0229] In certain embodiments, the biologically functional protein comprises a polypeptide backbone comprised of a dimeric Fc. In some embodiments, the polypeptide backbone comprised of a heterodimeric Fc comprises a modified CH3 and / or CH2 domain of Table 2 and / or Table 3, respectively.

[0230] Fc domain In certain embodiments, the trispecific fusion proteins described herein comprise a biologically functional protein, e.g., an antibody and / or polypeptide scaffold comprising a dimeric immunoglobulin Fc region. The term "Fc region" includes native sequence Fc regions and variant Fc regions. Unless otherwise specified herein, the numbering of amino acid residues in an Fc region or constant region is according to the EU numbering system, also known as the EU index, as described in Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD (1991). The "Fc polypeptide" of a dimeric Fc refers to one of the two polypeptides that form the dimeric Fc region, i.e., the polypeptide comprising the C-terminal constant region of an immunoglobulin heavy chain capable of stable self-association.

[0231] The Fc region can comprise either a CH3 domain or a CH3 and a CH2 domain. The CH3 domain comprises two CH3 sequences, each of which comprises one of the two Fc polypeptides of a dimeric Fc. Similarly, the CH2 domain comprises two CH2 sequences, each of which comprises one of the two Fc polypeptides of a dimeric Fc.

[0232] In certain embodiments, the fusion proteins described herein comprise an Fc based on human IgG Fc. In some embodiments, the fusion proteins comprise an Fc based on human IgG1 Fc. In some embodiments, the fusion proteins comprise an Fc based on a heterodimeric Fc comprising two different Fc polypeptides.

[0233] In certain embodiments, the fusion proteins described herein comprise an Fc based on a modified IgG Fc, wherein the CH3 domain comprises one or more amino acid modifications. In some embodiments, the fusion proteins comprise an Fc based on a modified IgG Fc, wherein the CH2 domain comprises one or more amino acid modifications. In some embodiments, the fusion proteins comprise an Fc based on a modified IgG Fc, wherein the CH3 domain comprises one or more amino acid modifications, and the CH2 domain comprises one or more amino acid modifications.

[0234] Modified Fc CH3 domain In certain embodiments, the fusion proteins herein comprise a heterodimeric immunoglobulin Fc comprising a modified CH3 domain, wherein the modified CH3 domain comprises one or more asymmetric amino acid modifications. As used herein, "asymmetric amino acid modification" refers to a modification in which an amino acid at a particular position on a first Fc polypeptide is different from the amino acid at the corresponding position on a second Fc polypeptide. These asymmetric amino acid modifications may include modification of only one of the two amino acids at the corresponding positions on each Fc polypeptide, or may include modification of both amino acids at the corresponding positions on each of the first and second Fc polypeptides.

[0235] In certain embodiments, the fusion protein comprises a heterodimeric Fc comprising a modified CH3 domain, the modified CH3 domain comprising one or more asymmetric amino acid modifications that promote heterodimeric Fc formation over homodimeric Fc formation. Amino acid modifications that can be made to the CH3 domain of an Fc to promote heterodimeric Fc formation are known in the art, including, for example, those described in International Publication No. WO 96 / 027011 ("knobs-into-holes"), Gunasekaran et al., 2010, J Biol Chem, 285, 19637-46 ("electrostatic steering"), Davis et al., 2010, Prot Eng Des Sel, 23(4):195-202 (strand exchange engineered domain (SEED) technology), and Labrijn et al., 2013, Proc Natl Acad Sci USA, 110(13):5145-50 (Fab arm exchange). Other examples include approaches that combine positive and negative design strategies to result in asymmetrically modified stable Fc regions, as described in International Publication Nos. WO2012 / 058768 and WO2013 / 063702.

[0236] In certain embodiments, the fusion protein comprises a heterodimeric Fc with a modified CH3 domain as described in International Publication No. WO2012 / 058768 or International Patent Publication No. WO2013 / 063702.

[0237] In some embodiments, the fusion protein comprises a heterodimeric human IgG1 Fc with a modified CH3 domain. Table 2 below provides the amino acid sequence of the human IgG1 Fc sequence corresponding to amino acids 231-447 of the full-length human IgG1 heavy chain. The CH2 domain is typically defined as comprising amino acids 231-340 of the full-length human IgG1 heavy chain, and the CH3 domain is typically defined as comprising amino acids 341-447 of the full-length human IgG1 heavy chain.

[0238] In certain embodiments, the fusion proteins herein comprise a heterodimeric Fc having a modified CH3 domain comprising one or more asymmetric amino acid modifications that promote heterodimeric Fc formation over homodimeric Fc formation, wherein the modified CH3 domain comprises a first Fc polypeptide comprising amino acid modifications at positions F405 and Y407 and a second Fc polypeptide comprising amino acid modifications at positions T366 and T394. In some embodiments, the amino acid modification at position F405 of the first Fc polypeptide of the modified CH3 domain is F405A, F405I, F405M, F405S, F405T, or F405V. In some embodiments, the amino acid modification at position Y407 of the first Fc polypeptide of the modified CH3 domain is Y407I or Y407V. In some embodiments, the amino acid modification at position T366 of the second Fc polypeptide of the modified CH3 domain is T366I, T366L, or T366M. In some embodiments, the amino acid modification at position T394 of the second Fc polypeptide of the modified CH3 domain is T394W. In some embodiments, the first Fc polypeptide of the modified CH3 domain further comprises an amino acid modification at position L351. In some embodiments, the amino acid modification at position L351 in the first Fc polypeptide of the modified CH3 domain is L351Y. In some embodiments, the second Fc polypeptide of the modified CH3 domain further comprises an amino acid modification at position K392. In some embodiments, the amino acid modification at position K392 in the second Fc polypeptide of the modified CH3 domain is K392F, K392L, or K392M. In some embodiments, one or both of the first and second Fc polypeptides of the modified CH3 domain further comprise the amino acid modification T350V.

[0239] In certain embodiments, the fusion proteins herein comprise a heterodimeric Fc having a modified CH3 domain comprising one or more asymmetric amino acid modifications that promote heterodimeric Fc formation over homodimeric Fc formation, wherein the modified CH3 domain comprises a first Fc polypeptide comprising the amino acid modification F405A, F405I, F405M, F405S, F405T, or F405V together with the amino acid modification Y407I or Y407V, and a second Fc polypeptide comprising the amino acid modification T366I, T366L, or T366M together with the amino acid modification T394W. In some embodiments, the first Fc polypeptide of the modified CH3 domain further comprises the amino acid modification L351Y. In some embodiments, the second Fc polypeptide of the modified CH3 domain further comprises the amino acid modification K392F, K392L, or K392M. In some embodiments, one or both of the first and second Fc polypeptides of the modified CH3 domain further comprises the amino acid modification T350V.

[0240] In certain embodiments, the fusion proteins herein comprise a heterodimeric Fc comprising a modified CH3 domain, as described above, having a first Fc polypeptide comprising amino acid modifications at positions F405 and Y407, and optionally further comprising an amino acid modification at position L351, and a second Fc polypeptide comprising amino acid modifications at positions T366 and T394, and optionally further comprising an amino acid modification at position K392, wherein the first Fc polypeptide has an amino acid modification at one or both of positions S400 or Q347. and / or the second Fc polypeptide further comprises an amino acid modification at one or both of positions K360 or N390, wherein the amino acid modification at position S400 is S400E, S400D, S400R or S400K, the amino acid modification at position Q347 is Q347R, Q347E or Q347K, the amino acid modification at position K360 is K360D or K360E, and the amino acid modification at position N390 is N390R, N390K or N390D.

[0241] In certain embodiments, the fusion proteins herein comprise a heterodimeric Fc comprising a modified CH3 domain comprising any one of variant 1, variant 2, variant 3, variant 4, or variant 5 modifications shown in Table 2. In certain embodiments, the CH3 domain has an amino acid sequence corresponding to SEQ ID NO: 4 or SEQ ID NO: 5. In certain embodiments, the CH3 has an amino acid sequence that is substantially identical to SEQ ID NO: 4 or SEQ ID NO: 5. In certain embodiments, the CH3 domain has an amino acid sequence that is about 80%, about 85%, about 90%, or about 95% identical to SEQ ID NO: 4 or SEQ ID NO: 5.

[0242] (Table 2) TIFF2025504909000003.tif119165

[0243] Modified Fc CH2 domain In certain embodiments, the fusion proteins herein comprise an Fc based on an IgG Fc with a modified CH2 domain, hi some embodiments, the fusion proteins comprise an Fc based on an IgG Fc with a modified CH2 domain, wherein the modification of the CH2 domain results in altered binding to one or more Fc receptors (FcR), such as receptors of the FcγRI, FcγRII, and FcγRIII subclasses.

[0244] Numerous amino acid modifications to the CH2 domain that selectively alter the affinity of Fc for different Fcγ receptors are known in the art. Both amino acid modifications that result in increased binding and amino acid modifications that result in decreased binding may be useful in certain indications. For example, increasing the binding affinity of Fc to FcγRIIIa (an activating receptor) may result in increased antibody-dependent cell-mediated cytotoxicity (ADCC), thereby increasing lysis of target cells. Decreased binding to FcγRIIb (an inhibitory receptor) may also be beneficial in some situations. In certain indications, reduction or elimination of ADCC and complement-mediated cytotoxicity (CDC) may be desirable. In such cases, modified CH2 domains containing amino acid modifications that result in increased binding to FcγRIIb or that reduce or eliminate binding of the Fc region to all Fcγ receptors ("knockout" variants) may be useful.

[0245] Examples of amino acid modifications to the CH2 domain that alter Fc binding by Fcγ receptors include, but are not limited to, S298A / E333A / K334A and S298A / E333A / K334A / K326A (increased affinity for FcγRIIIa) (Lu, et al., 2011, J Immunol Methods, 365(1-2):132-41); F243L / R292P / Y300L / V305I / P396L (increased affinity for FcγRIIIa) (Stavenhagen, et al., 2007, Cancer Res, 67(18):8882-90); F243L / R292P / Y300L / L235V / P396L (increased affinity for FcγRIIIa) (Nordstrom JL, et al., 2011, Breast Cancer Res, 13(6):R123); F243L (increased affinity for FcγRIIIa) (Stewart, et al., 2011, Protein Eng Des Sel., 24(9):671-8); S298A / E333A / K334A (increased affinity for FcγRIIIa) (Shields, et al., 2001, J Biol Chem, 276(9):6591-604); S239D / I332E / A330L and S239D / I332E (increased affinity for FcγRIIIa) (Lazar, et al., 2006, Proc Natl Acad Sci USA, 103(11):4005-10), and S239D / S267E and S267E / L328F (increased affinity for FcγRIIb) (Chu, et al., 2008, Mol Immunol, 45(15):3926-33).

[0246] Additional modifications that affect Fc binding to Fcγ receptors are described in Therapeutic Antibody Engineering (Strohl & Strohl, Woodhead Publishing series in Biomedicine No 11, ISBN 1 907568 37 9, Oct 2012, page 283).

[0247] In certain embodiments, the fusion protein comprises an Fc based on an IgG Fc with a modified CH2 domain, wherein the modified CH2 domain comprises one or more amino acid modifications (i.e., a "knockout" or "KO" variant) that result in reduced or eliminated binding of the Fc region to all of the Fcγ receptors.

[0248] Various publications describe strategies used to engineer antibodies to generate "knockout" variants (see, e.g., Strohl, 2009, Curr Opin Biotech 20:685-691, and Strohl & Strohl, "Antibody Fc engineering for optimal antibody performance" In Therapeutic Antibody Engineering, Cambridge: Woodhead Publishing, 2012, pp 225-249). These strategies include reducing effector function through modified glycosylation (described in more detail below), using an IgG2 / IgG4 scaffold, or introducing mutations in the hinge or CH2 domain of the Fc (see also U.S. Patent Publication No. 2011 / 0212087, International Publication No. WO2006 / 105338, U.S. Patent Publication No. 2012 / 0225058, U.S. Patent Publication No. 2012 / 0251531, and Strop et al., 2012, J. Mol. Biol., 420:204-219).

[0249] Specific non-limiting examples of known amino acid modifications to reduce FcγR and / or complement binding to Fc include those identified in Table 3.

[0250] (Table 3) TIFF2025504909000004.tif113165

[0251] Additional examples include Fc regions engineered to contain the amino acid modifications L235A / L236A / D265S, and asymmetric amino acid modifications in the CH2 domain that reduce Fc binding to all Fcγ receptors are described in International Publication No. WO2014 / 190441.

[0252] In certain embodiments, the CH2 domain has an amino acid sequence corresponding to SEQ ID NO: 6. In certain embodiments, the CH2 has an amino acid sequence that is substantially identical to SEQ ID NO: 6. In certain embodiments, the CH2 domain has an amino acid sequence that is about 80%, about 85%, about 90%, or about 95% identical to SEQ ID NO: 6.

[0253] Antibody-drug conjugates Certain embodiments of the fusion proteins described herein include biologically functional proteins that are antibodies conjugated to drugs, i.e., antibody-drug conjugates (ADCs). The drugs in ADCs can be any therapeutic molecule, e.g., toxins, chemotherapeutic agents, or small molecule inhibitors. The ADCs can be conjugated to drugs via linkers that can be cleavable or non-cleavable. Cleavable linkers can be easily cleaved under intracellular conditions, e.g., by lysosomal processes. Examples of cleavable linkers include linkers that are protease-sensitive, acid-sensitive, reduction-sensitive, or photocleavable. Drug conjugation can be carried out by any method known in the art, including, but not limited to, lysine or cysteine ​​conjugation, bisthiol linkers, conjugation using glycosylation sites on antibodies, ultraviolet light conjugation, and the use of unnatural amino acids.

[0254] VII. Peptide Linkers The trispecific fusion proteins described herein may contain at least one peptide linker. A peptide linker is a peptide that connects or joins other peptides or polypeptides of the fusion protein. In certain embodiments, the peptide linker fuses a biologically functional polypeptide of a protein, such as an anti-CD3 / anti-TAA antibody portion, to the PD-1 domain (i.e., the third binding domain).

[0255] In some embodiments, the peptide linker connecting the PD-1 domain to the remainder of the fusion protein is long enough to allow the antigen-binding domain of the fusion protein to bind to its epitope. In addition to providing spacing functionality, the peptide linker may provide suitable flexibility or rigidity for properly orienting one or more domains of the fusion proteins herein within the fusion protein and between or among the fusion protein and its target(s). Furthermore, the peptide linker may support expression of the full-length fusion protein and stability of the purified protein both in vitro and in vivo after administration to a subject in need thereof, e.g., a human, and preferably is non-immunogenic or has reduced immunogenicity in those same subjects. In certain embodiments, the peptide linker may comprise part or all of a human immunoglobulin hinge, the stalk region of C-type lectins, a family of type II membrane proteins, or a combination thereof.

[0256] In certain embodiments, the one or more peptide linker(s) of the fusion protein are long enough to allow the other components of the fusion protein (e.g., the anti-CD3 and anti-TAA binding domains) to bind to their respective epitopes, and are about 2 to about 150 amino acids in length. In certain embodiments, the peptide linker is about 3 to about 50 amino acids in length, or about 5 to about 20 amino acids in length, or about 10 to about 50 amino acids in length, or about 2 to about 40 amino acids in length, or about 8 to about 20 amino acids in length, or about 10 to about 60 amino acids in length, or about 10 to about 30 amino acids in length, or about 15 to about 25 amino acids in length. In some embodiments, the peptide linker is 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, or 60 amino acids.

[0257] In certain embodiments, a fusion protein comprising a peptide linker(s) described herein comprises at least two heterologous polypeptides, i.e., a first polypeptide located at the amino (N) terminus of the peptide linker and a second polypeptide located at the carboxyl (C) terminus of the peptide linker, such that the two heterologous polypeptides are separated by the peptide linker.

[0258] In certain embodiments, the peptide linker of the fusion protein comprises the amino acid sequence (EAAAK)n, where n is an integer from 1 to 5. In some embodiments, the peptide linker is EAAAK. In some embodiments, the peptide linker is EAAAKEAAAK. In some embodiments, the peptide linker comprises a polyproline linker, optionally having the amino acid sequence of PPP (SEQ ID NO: 41) or PPPP (SEQ ID NO: 40). In certain embodiments, the linker is a glycine (G)-proline (P) polypeptide linker, optionally GPPPG, GGPPPGG, GPPPPG, or GGPPPGG. In certain embodiments, the peptide linker comprises a Gly n In certain embodiments, the peptide linker is (Gly3Ser) n (Gly4Ser)1, (Gly3Ser)1(Gly4Ser) n , (Gly3Ser) n (Gly4Ser) n , or (Gly4Ser) n where n is an integer from 1 to 5. In certain embodiments, peptide linkers suitable for connecting different domains include sequences containing glycine-serine linkers, such as, but not limited to, EAAAKEAAAKEAAAK, (G3S) n , PPPPP, P n , G.P. n G, GGP n G.G., G. n (G m S) n -GG, (SG n ) m , (SEG n ) m where m and n are integers from 0 to 20, or from 0 to about 20. Many peptide linkers are known in the art, for example, Chen et al. Adv Drug Deliv Rev. 2013 Oct 15;65(10): 1357-1369.

[0259] In certain embodiments, the peptide linker is an amino acid sequence obtained, derived, or designed from an antibody hinge region sequence, a sequence that links a binding domain to a receptor, or a sequence that links a binding domain to a cell surface transmembrane region or membrane anchor. In some embodiments, the peptide linker has at least one cysteine ​​that can participate in at least one disulfide bond under physiological conditions or other standard peptide conditions (e.g., peptide purification conditions, conditions for peptide storage). In certain embodiments, a peptide linker corresponding to or similar to an immunoglobulin hinge peptide retains a cysteine ​​corresponding to the hinge cysteine ​​located toward the amino terminus of the hinge. In further embodiments, the peptide linker is from an IgG1 hinge, modified to remove any cysteine ​​residues, or an IgG1 hinge with one or two cysteines corresponding to the hinge cysteines.

[0260] In certain embodiments, peptide linkers for use in the fusion proteins herein may comprise an "altered wild-type immunoglobulin hinge region" or an "altered immunoglobulin hinge region." Such an altered hinge region refers to (a) a wild-type immunoglobulin hinge region having up to 30 percent amino acid changes (e.g., up to 25, 20, 15, 10, or 5 percent amino acid substitutions or deletions), (b) a portion of a wild-type immunoglobulin hinge region that is at least 10 amino acids in length (e.g., at least 12, 13, 14, or 15 amino acids) having up to 30 percent amino acid changes (e.g., up to 25, 20, 15, 10, or 5 percent amino acid substitutions or deletions), or (c) a portion of a wild-type immunoglobulin hinge region that includes the core hinge region (the portion can be 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15, or at least 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15 amino acids in length). In certain embodiments, one or more cysteine ​​residues in a wild-type immunoglobulin hinge region, such as an IgG1 hinge, including the upper and core regions, may be substituted with one or more other amino acid residues (e.g., one or more serine residues). The altered immunoglobulin hinge region may alternatively or additionally have a proline residue in the wild-type immunoglobulin hinge region, such as an IgG1 hinge, including the upper and core regions, substituted with another amino acid residue (e.g., a serine residue).

[0261] Alternative hinge and linker sequences that can be used as connecting regions can be derived from portions of cell surface receptors that connect IgV-like or IgC-like domains. Regions between IgV-like domains in cell surface receptors containing multiple tandem IgV-like domains and regions between IgC-like domains in cell surface receptors containing multiple tandem IgC-like domains can also be used as connecting regions or linker peptides. In certain embodiments, hinge and linker sequences are 5-60 amino acids in length and can be primarily flexible, although they can also provide more rigid properties and contain a primarily helical structure with minimal beta-sheet structure.

[0262] VII. Target and Tumor-Associated Antigens In some embodiments, the antigen-binding domain of the fusion protein described herein specifically binds to a cell surface molecule. In certain embodiments, the antigen-binding domain of the fusion protein specifically binds to a tumor-associated antigen (TAA). A TAA is any antigenic substance expressed on the surface of tumor cells. In some embodiments, the antigen binding domain is selected from the group consisting of fibroblast activation protein alpha (FAPa), trophoblast glycoprotein (5T4), tumor-associated calcium signaling factor 2 (Trop2), fibronectin EDB (EDB-FN), fibronectin F.IIIB domain, CGS-2, EpCAM, EGFR, HER-2, HER-3, cMet, CEA, and FOLR1, EpCAM, EGFR, HER-2, HER-3, cMet, CEA, and FOLR1, EpCAM, EGFR, HER-2, HER-3, c-Met, FOLR1, PSMA, CD38, BCMA, and CEA, 5T4, AFP, B7-H3, cadherin-6, CAIX, CD117, CD123, CD138, CD166, C D19, CD20, CD205, CD22, CD30, CD33, CD40, CD352, CD37, CD44, CD52, CD56, CD70, CD71, CD74, CD79b , DLL3, DR5, EphA2, FAP, FGFR2, FGFR3, GPC3, gpA33, FLT-3, gpNMB, HPV-16E6, HPV-16E7, ITGA2, IT The antigen-binding domain specifically binds to a TAA selected from GA3, SLC39A6, MAGE, mesothelin (MSLN), Mucl, Mucl6, NaPi2b, nectin-4, P-cadherin, NY-ESO-1, PRLR, PSCA, PTK7, ROR1, SLC44A4, SLTRK5, SLTRK6, STEAP1, TIM1, tissue factor (TF), Trop2, and WT1. In certain embodiments, the antigen-binding domain specifically binds to a TAA selected from mesothelin, claudin 18.2 (Cldn18.2), GPC3, DLL3, PSMA, MUC17, LIV1, ROR1, and EGFRvIII.

[0263] In some embodiments, the antigen binding domain specifically binds to an antigen expressed on a virus-infected cell, a bacteria-infected cell, an injured red blood cell, an arterial plaque cell, an inflammatory or fibrotic tissue cell.

[0264] In certain embodiments, the antigen-binding domain specifically binds to a cytokine receptor. Examples of cytokine receptors include type I cytokine receptors, such as GM-CSF receptor, G-CSF receptor, type I IL receptor, Epo receptor, LIF receptor, CNTF receptor, and TPO receptor; type II cytokine receptors, such as IFN-alpha receptors (IFNAR1, IFNAR2), IFB-beta receptors, IFN-gamma receptors (IFNGR1, IFNGR2), and type II IF receptors; chemokine receptors, such as CC chemokine receptors, CXC chemokine receptors, CX3C chemokine receptors, and XC chemokine receptors. These include, but are not limited to, kine receptors; tumor necrosis receptor superfamily receptors, such as TNFRSF5 / CD40, TNFRSF8 / CD30, TNFRSF7 / CD27, TNFRSFlA / TNFRl / CD120a, TNFRSF1B / TNFR2 / CD120b; TGF-beta receptors, such as TGF-beta receptor 1 and TGF-beta receptor 2; and Ig superfamily receptors, such as IF-1 receptor, CSF-1R, PDGFR (PDGFRA, PDGFRB), and SCFR.

[0265] In certain embodiments, the antigen-binding domain of the fusion proteins described herein specifically binds to at least one molecule or target of interest in vivo. In certain embodiments, the target of interest is cluster of differentiation 3 (CD3), human epidermal growth factor receptor 2 (HER2), epidermal growth factor receptor (EGFR), mesothelin (MSLN), tissue factor (TF), cluster of differentiation 19 (CD19), tyrosine protein kinase Met (c-Met), cluster of differentiation 40 (CD40), cadherin 3 (CDH3), or a combination thereof. In certain embodiments, the fusion protein comprises an antibody, wherein at least one antigen-binding domain of the antibody binds to an epitope on CD3, HER2, EGFR, MSLN, TF, CD19, c-Met, CD40, CDH3, or a combination thereof.

[0266] In some embodiments, the target of interest is HER2. In such embodiments, the anti-HER2 binding domain of the fusion protein has a VH having an amino acid sequence corresponding to SEQ ID NO: 120 and a VL having an amino acid sequence corresponding to SEQ ID NO: 124. In certain embodiments, the anti-HER2 paratope has a VH amino acid sequence substantially identical to SEQ ID NO: 120 and a VL amino acid sequence substantially identical to SEQ ID NO: 124. In certain embodiments, the anti-HER2 paratope has a VH amino acid sequence that is about 80%, about 85%, about 90%, or about 95% identical to SEQ ID NO: 120 and a VL amino acid sequence that is about 80%, about 85%, about 90%, or about 95% identical to SEQ ID NO: 124. In certain embodiments, the anti-HER2 paratope has a VH amino acid sequence that is about 96%, about 97%, about 98%, or about 99% identical to SEQ ID NO: 120 and a VL amino acid sequence that is about 96%, about 97%, about 98%, or about 99% identical to SEQ ID NO: 124. In some embodiments, the anti-HER2 paratope comprises an scFv having an amino acid sequence corresponding to SEQ ID NO: 3. In some embodiments, the anti-HER2 has a VH having three CDRs, HCDR1, HDR2, and HCDR3, having amino acid sequences corresponding to SEQ ID NOs: 121, 122, and 123, respectively, and a VL having three CDRs, LCDR1, LCDR2, and LCDR3, having amino acid sequences corresponding to SEQ ID NOs: 125, 126, and 127, respectively.

[0267] In some embodiments, the TAA is HER2, and the anti-HER2 VH sequence of the second binding comprises an HCDR1 sequence comprising DTYIH (SEQ ID NO: 121), an HCDR2 sequence comprising RIYPTNGYTRYADSVKG (SEQ ID NO: 122), and an HCDR3 sequence comprising WGGDGFYAMDY (SEQ ID NO: 123), and the anti-HER2 VL sequence of the second binding comprises an LCDR1 sequence comprising RASQDVNTAVA (SEQ ID NO: 125), an LCDR2 sequence comprising SASFLYS (SEQ ID NO: 126), and an LCDR3 sequence comprising QQHYTTPPT (SEQ ID NO: 127). In some embodiments, the anti-HER2 binding domain comprises a VH domain comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NOs: 120 and 231, and a VL domain comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 124.

[0268] In certain embodiments, the antigen-binding domain of the fusion protein specifically binds to a molecule, e.g., a polypeptide, on an immune cell. In certain embodiments, the fusion protein comprises both an antigen-binding domain that specifically binds to a TAA and an antigen-binding domain that specifically binds to a molecule on an immune cell, e.g., a polypeptide such as CD3. Thus, in certain embodiments, the fusion protein binds to both a tumor cell and an immune cell. In certain embodiments, the immune cell is a T cell. In certain embodiments, the immune cell is a macrophage, dendritic cell, neutrophil, B cell, or NK cell. In certain embodiments, the fusion protein binds to the CD3 antigen on a T cell and one or more TAAs on a tumor cell.

[0269] In certain embodiments, the trispecific fusion proteins of the present disclosure do not bind to HER2, i.e., they comprise a binding domain capable of binding to a TAA other than HER2.

[0270] In some embodiments, the fusion proteins herein are capable of binding to MSLN. In certain embodiments, such fusion proteins comprise a binding domain capable of binding to MSLN. Thus, in some embodiments, the TAA is MSLN, and the anti-MSLN VH sequence of the second binding domain comprises an HCDR1 sequence comprising GYTMN (SEQ ID NO: 286), an HCDR2 sequence comprising LITPYNGASSYNQKFRG (SEQ ID NO: 288), and an HCDR3 sequence comprising GGYDGRGFDY (SEQ ID NO: 285), and the anti-MSLN VL sequence of the second binding domain comprises an LCDR1 sequence comprising SASSSVSYMH (SEQ ID NO: 300), an LCDR2 sequence comprising DTSKLAS (SEQ ID NO: 279), and an LCDR3 sequence comprising QQWSGYPLT (SEQ ID NO: 294).

[0271] In some embodiments, the anti-MSLN VH sequence of the second binding domain comprises an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 297, and the anti-MSLN VL sequence of the second binding domain comprises an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 276 or 277. In embodiments describing trispecific tetravalent fusion proteins, the second and third binding domains may both be capable of binding to MSLN, as further described herein. In some of these embodiments, the second and third binding domains each comprise or consist of the same anti-MSLN VH and VL sequences and therefore bind to the same epitope on the MSLN target protein.

[0272] In some embodiments, the fusion proteins described herein comprise a binding domain capable of binding to Cldn18.2.

[0273] In some embodiments, the TAA is Cldn18.2, and the anti-Cldn18.2 VH sequence of the second binding domain comprises an HCDR1 sequence comprising SNPMI (SEQ ID NO: 310), an HCDR2 sequence comprising IIDTDGSTYYADWAKG (SEQ ID NO: 311), and an HCDR3 sequence comprising RLHGSSNGYYDDL (SEQ ID NO: 312), and the anti-MSLN VL sequence of the second binding domain comprises an LCDR1 sequence comprising QASQSIYSYLS (SEQ ID NO: 313), an LCDR2 sequence comprising KASTLAS (SEQ ID NO: 314), and an LCDR3 sequence comprising QQGYTVTNVDKNT (SEQ ID NO: 315).

[0274] In some embodiments, the anti-Cldn18.2 VH sequence of the second binding domain comprises an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 316, and the anti-Cldn18.2 VL sequence of the second binding domain comprises an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 317.

[0275] In embodiments describing trispecific tetravalent fusion proteins, the second and third binding domains may both be capable of binding to Cldn18.2, as further described herein. In some of these embodiments, the second and third binding domains each comprise or consist of the same anti-Cldn18.2 VH and VL sequences and therefore bind to the same epitope on the Cldn18.2 target protein.

[0276] T Cell Engager T cell engagers (TCEs) are polypeptide constructs, often bispecific antibodies, that simultaneously bind to a TAA on tumor cells and a CD3 epitope on T cells, forming an artificial TCR-independent immune synapse. This activates T cells and allows them to exert cytotoxic effects against tumor cells. Bispecific antibodies capable of targeting T cells to tumor cells have been identified and validated for cancer therapy. Blinatumomab is an example of a bispecific anti-CD3-CD19 antibody in a format called BiTE™ (bispecific T cell engager), which has been identified for the treatment of B cell disorders such as relapsed B cell non-Hodgkin's lymphoma and chronic lymphocytic leukemia (Baeuerle et al. (2009) Cancer Research 12:4941-4944) and is FDA-approved. T cell enhancers directed against other tumor-associated target antigens have also been generated, and several have entered clinical trials: AMG110 / MT110 EpCAM for lung, gastric, and colorectal cancer; AMG211 / MEDI565 CEA for gastrointestinal adenocarcinoma; and AMG 212 / BAY2010112 PSMA for prostate cancer (see Suruadevara, CM et al., Oncoimmunology. 2015 Jun;4(6):e1008339). While these studies demonstrated promising clinical efficacy, they were again hampered by severe dose-limiting toxicities, primarily due to cytokine release syndrome (CRS). This narrowed the therapeutic window. The use of T cell-binding paratopes and fusion proteins that are more specifically activated in the tumor microenvironment (e.g., exerting more TAA-dependent T cell cytotoxicity) may reduce the toxicity of TCEs.

[0277] In certain embodiments, the fusion proteins described herein bind to a CD3 antigen on a T cell at one of their binding domains and to both a TAA and an IgSF ligand on a tumor cell at the second and third binding domains, respectively. In certain embodiments, as shown, for example, in Figures 1(A) and 1(B), binding of an IgSF ligand (e.g., PD-L1) on a tumor cell inhibits binding of its IgSF receptor (e.g., PD-1) on a T cell, thereby blocking checkpoint inhibition.

[0278] In certain embodiments, the fusion protein comprises an anti-CD3 paratope VH and VL that are substantially identical to those of the paratopes shown in Table BB. In certain embodiments, the CD3 paratope comprises the following VH and VL amino acid sequences: (a) a VH comprising an amino acid sequence corresponding to SEQ ID NO: 2 and a VL comprising an amino acid sequence according to SEQ ID NO: 1; (b) a VH comprising an amino acid sequence corresponding to SEQ ID NO: 206 and a VL comprising an amino acid sequence corresponding to SEQ ID NO: 210; (c) a VH comprising an amino acid sequence corresponding to SEQ ID NO: 215 and a VL comprising an amino acid sequence corresponding to SEQ ID NO: 219; (d) a VH comprising an amino acid sequence corresponding to SEQ ID NO: 223 and a VL comprising an amino acid sequence corresponding to SEQ ID NO: 227; (d) a VH comprising an amino acid sequence corresponding to SEQ ID NO: 231 and a VL comprising an amino acid sequence corresponding to SEQ ID NO: 235; or (e) VH comprising an amino acid sequence corresponding to SEQ ID NO: 239 and VL comprising an amino acid sequence corresponding to SEQ ID NO: 243.

[0279] In certain embodiments, the CD3 paratope comprises a VH and VL that are about 90%, about 91%, about 92%, about 93%, about 94%, about 95%, about 96%, about 97%, about 98%, or about 99% identical to the following: (a) a VH comprising an amino acid sequence corresponding to SEQ ID NO: 2 and a VL comprising an amino acid sequence according to SEQ ID NO: 1; (b) a VH comprising an amino acid sequence corresponding to SEQ ID NO: 206 and a VL comprising an amino acid sequence corresponding to SEQ ID NO: 210; (c) a VH comprising an amino acid sequence corresponding to SEQ ID NO: 215 and a VL comprising an amino acid sequence corresponding to SEQ ID NO: 219; (d) a VH comprising an amino acid sequence corresponding to SEQ ID NO: 223 and a VL comprising an amino acid sequence corresponding to SEQ ID NO: 227; (e) a VH comprising an amino acid sequence corresponding to SEQ ID NO: 231 and a VL comprising an amino acid sequence corresponding to SEQ ID NO: 235; or (f) VH comprising an amino acid sequence corresponding to SEQ ID NO: 239 and VL comprising an amino acid sequence corresponding to SEQ ID NO: 243.

[0280] In certain embodiments, the anti-CD3 paratope comprises a VH comprising three heavy chain CDRs, HCDR1, HCDR2, and HCDR3, comprising amino acid sequences corresponding to SEQ ID NOs: 207, 208, and 209, and a VL comprising three light chain CDRs, LCDR1, LCDR2, and LCDR3, comprising amino acid sequences corresponding to SEQ ID NOs: 211, 212, and 214. In certain embodiments, the anti-CD3 paratope comprises a VH comprising three heavy chain CDRs, HCDR1, HCDR2, and HCDR3, comprising amino acid sequences corresponding to SEQ ID NOs: 224, 225, and 226, and a VL comprising three light chain CDRs, LCDR1, LCDR2, and LCDR3, comprising amino acid sequences corresponding to SEQ ID NOs: 228, 229, and 230. In certain embodiments, the anti-CD3 paratope comprises a VH comprising three heavy chain CDRs, HCDR1, HCDR2, and HCDR3, comprising amino acid sequences corresponding to SEQ ID NOs: 232, 233, and 234, and a VL comprising three light chain CDRs, LCDR1, LCDR2, and LCDR3, comprising amino acid sequences corresponding to SEQ ID NOs: 236, 237, and 238. In certain embodiments, the anti-CD3 paratope comprises a VH comprising three heavy chain CDRs, HCDR1, HCDR2, and HCDR3, comprising amino acid sequences corresponding to SEQ ID NOs: 240, 241, and 242, and a VL comprising three light chain CDRs, LCDR1, LCDR2, and LCDR3, comprising amino acid sequences corresponding to SEQ ID NOs: 244, 245, and 246.

[0281] In some embodiments, the anti-CD3 binding domain comprises a VH domain comprising an HCDR1 sequence selected from the group consisting of RSTMH (SEQ ID NO: 207), YYGMS (SEQ ID NO: 303), KYAMN (SEQ ID NO: 224), and TYAMN (SEQ ID NO: 232), an HCDR2 sequence selected from the group consisting of YINPSSAYTNYNQKFKD (SEQ ID NO: 208), SITSSGGRIYYPDSVKG (SEQ ID NO: 301), SITRSGGRIYYPDSVKG (SEQ ID NO: 217), RIRSKYNNYATYYADSVKD (SEQ ID NO: 225), and RIRSKYNNYATYYADSVKG (SEQ ID NO: 233), and an HCDR3 sequence selected from the group consisting of PQVHYDYNGFPY (SEQ ID NO: 209), DGRDGWVAY (SEQ ID NO: 275), HGNFGNSYISYWAY (SEQ ID NO: 226), and HGNFGNSYVSWFAY (SEQ ID NO: 234). and a VL domain comprising an LCDR1 sequence selected from the group consisting of ASSSVSYMN (SEQ ID NO:211), KRNTGNIGSNYVN (SEQ ID NO:287), TGNTGNIGSNYVN (SEQ ID NO:220), GSSTGAVTSGNYPN (SEQ ID NO:228), and GSSTGAVTTSNYAN (SEQ ID NO:236), an LCDR2 sequence selected from the group consisting of DSSKLAS (SEQ ID NO:212), RNDKRPD (SEQ ID NO:298), RDDKRPS (SEQ ID NO:221), GTKFLAP (SEQ ID NO:229), RSYQRPS (SEQ ID NO:199), and GTNKRAP (SEQ ID NO:237), and an LCDR3 sequence selected from the group consisting of QQWSRNPPT (SEQ ID NO:214), QSYSSGFI (SEQ ID NO:295), VLWYSNRWV (SEQ ID NO:230), ATWDDSLDGWV (SEQ ID NO:200), and ALWYSNLWV (SEQ ID NO:238).

[0282] VIII. Sequence homology Certain embodiments of the present disclosure relate to an isolated polynucleotide or set of polynucleotides that encodes the fusion proteins described herein. A polynucleotide in this context can encode all or part of a fusion protein.

[0283] The terms "nucleic acid," "nucleic acid molecule," and "polynucleotide" are used interchangeably herein and refer to a polymer of nucleotides of any length, either deoxyribonucleotides or ribonucleotides, or analogs thereof. Non-limiting examples of polynucleotides include genes, gene fragments, messenger RNA (mRNA), cDNA, recombinant polynucleotides, plasmids, vectors, isolated DNA of any sequence, isolated RNA of any sequence, nucleic acid probes, and primers.

[0284] A polynucleotide that "encodes" a given polypeptide is one that is transcribed (in the case of DNA) and translated into a polypeptide (in the case of mRNA) in vivo when placed under the control of appropriate regulatory sequences. The boundaries of the coding sequence are determined by a start codon at the 5' (amino) terminus and a translation stop codon at the 3' (carboxy) terminus. A transcription termination sequence may be located 3' to the coding sequence.

[0285] In certain embodiments, the present disclosure relates to polynucleotides and polypeptide sequences that are identical or substantially identical to polynucleotides encoding at least a portion of a fusion protein described herein, e.g., a first or second polypeptide of a biologically functional protein. The terms "identical" or "percent identical" in the context of two or more polynucleotide or polypeptide sequences refer to two or more sequences or subsequences that are the same. Sequences are "substantially identical" when compared and aligned for best correspondence over a comparison window or over a designated region, as determined using one of the commonly used sequence comparison algorithms known to those skilled in the art, or by manual alignment and visual inspection, if the sequences have the same percentage of amino acid residues or nucleotides (e.g., about 80%, about 85%, about 90%, or about 95% identity over a designated region). This definition also refers to the complement of a test polynucleotide sequence. Identity can exist over a region of at least about 50 amino acids or nucleotides in length, or over a region of 75-100 amino acids or nucleotides in length, or, if not specified, over the entire polypeptide or polynucleotide sequence. For sequence comparison, a test sequence is typically compared to a designated reference sequence. When using a sequence comparison algorithm, test and reference sequences are entered into a computer, subsequence coordinates are designated, if necessary, and sequence algorithm program parameters are designated. Default program parameters can be used, or alternative parameters can be designated. The sequence comparison algorithm then calculates the percent sequence identity of the test sequence relative to the reference sequence based on the program parameters.

[0286] As used herein, a "comparison window" refers to a segment of a sequence comprising 20 to 1000 contiguous amino acid or nucleotide positions, e.g., about 50 to about 600, about 100 to about 300, or about 150 to about 200 contiguous amino acid or nucleotide positions, over which a test sequence can be compared to a reference sequence over the same number of contiguous positions after the two sequences are optimally aligned. Longer segments, up to the full-length sequence, can also be used as the comparison window in certain embodiments. Methods for aligning sequences for comparison purposes are known to those of skill in the art. Optimal sequence alignment for comparison can be performed, for example, by the local homology algorithm of Smith & Waterman, 1970, Adv. Appl. Math., 2:482c, by the homology alignment algorithm of Needleman & Wunsch, 1970, J. Mol. Biol., 48:443, by the search for similarity method of Pearson & Lipman, 1988, Proc. Natl. Acad. Sci. USA, 85:2444, or by computer implementations of these algorithms (e.g., GAP, BESTFIT, FASTA, or TFASTA (Wisconsin Genetics Software Package, Genetics Computer Group, Madison, WI)), or by manual alignment and visual inspection (see, for example, Ausubel et al., Current Protocols in Molecular Biology, (1995 supplement), Cold Spring Harbor Laboratory Press). Examples of available algorithms suitable for determining percent sequence identity are the BLAST and BLAST 2.0 algorithms, which are described in Altschul et al., 1997, Nuc. Acids Res., 25:3389-3402, and Altschul et al., 1990, J. Mol. Biol., 215:403-410, respectively.Software for performing BLAST analyses is publicly available from the National Center for Biotechnology Information (NCBI) website. Certain embodiments described herein relate to variant sequences containing one or more amino acid substitutions. In some embodiments, the amino acid substitutions are conservative substitutions. Generally, a "conservative substitution" is considered to be the replacement of one amino acid with another amino acid that has similar physical, chemical, and / or structural properties. Common conservative substitutions are listed in column 1 of Table 4.

[0287] (Table 4) TIFF2025504909000005.tif203165

[0288] Those of skill in the art will understand that, although the primary factors determining what constitutes a conservative substitution are usually the size of the amino acid side chain and its physical / chemical properties, in certain circumstances a wider range of amino acids than those listed in column 1 can be substituted for a given amino acid. These additional amino acids tend to have similar properties to the amino acid being substituted, but significantly different sizes, or similar sizes, but significantly different physical / chemical properties. This broad range of conservative substitutions is listed in column 2 of Table 4. Those of skill in the art will readily be able to determine the most appropriate set of substituents to select, given the particular protein environment in which the amino acid substitution is to be made.

[0289] IX. Preparation of Fusion Proteins The fusion proteins described herein can be produced using standard recombinant methods known in the art (see, e.g., U.S. Patent No. 4,816,567 and "Antibodies: A Laboratory Manual," 2 nd Edition, Ed. Greenfield, Cold Spring Harbor Laboratory Press, New York, 2014).

[0290] Vector encoding the fusion protein For recombinant production of the fusion proteins described herein, a polynucleotide or set of polynucleotides encoding the fusion protein is generated and inserted into one or more vectors for further cloning and / or expression in a host cell. The polynucleotide(s) encoding the fusion protein can be generated by standard methods known in the art (e.g., Ausubel et al., Current Protocols in Molecular Biology, John Wiley & Sons, New York, 1994 & update, and "Antibodies: A Laboratory Manual," 2002). nd (See, for example, "The Genetic Algorithm of the Invention," Ed. Greenfield, Cold Spring Harbor Laboratory Press, New York, 2014). As will be understood by those skilled in the art, the number of polynucleotides required for expression of a fusion protein will depend on the format of the fusion protein, for example, whether the fusion protein includes an antibody, and the number of polypeptides in the fusion protein. For example, if the fusion protein includes two polypeptide chains, two polynucleotides encoding each of the polypeptide chains will be required. Similarly, in certain embodiments, if the fusion protein includes a biologically functional protein in mAb format, two polynucleotides encoding each of the polypeptide chains may be required. If multiple polynucleotides are required, they can be incorporated into a single vector or multiple vectors.

[0291] Generally, a polynucleotide or set of polynucleotides is incorporated into an expression vector for expression along with one or more regulatory elements, such as transcriptional elements necessary for efficient transcription of the polynucleotide. Examples of such regulatory elements include, but are not limited to, promoters, enhancers, terminators, and polyadenylation signals. Those skilled in the art will understand that the choice of regulatory element will depend on the host cell selected for expression of the polypeptide of the fusion protein, and that such regulatory elements can be derived from a variety of sources, including bacterial, fungal, viral, mammalian, or insect genes. The expression vector can optionally further contain heterologous nucleic acid sequences that facilitate expression or purification of the expressed protein. Examples include, but are not limited to, signal peptides and affinity tags, such as metal affinity tags, histidine tags, avidin / streptavidin coding sequences, glutathione-S-transferase (GST) coding sequences, and biotin coding sequences. The expression vector can be an extrachromosomal vector or an integrating vector.

[0292] Certain embodiments of the present disclosure relate to vectors (such as expression vectors) that contain one or more polynucleotides encoding at least a portion of the fusion proteins described herein. The polynucleotide(s) can be contained in one vector or multiple vectors. In some embodiments, the polynucleotides are contained in a multicistronic vector.

[0293] Expression vectors used to express the polynucleotides include, but are not limited to, pTT5 and pUC15.

[0294] Cells containing a vector encoding the fusion protein Suitable host cells for cloning or expressing fusion protein polypeptides include various prokaryotic or eukaryotic cells known in the art. Eukaryotic host cells include, for example, mammalian cells, plant cells, insect cells, and yeast cells (such as Saccharomyces cells or Pichia cells). Prokaryotic host cells include, for example, E. coli cells, A. salmonicida cells, or B. subtilis cells.

[0295] In certain embodiments, fusion proteins, particularly where glycosylation and Fc effector function are not required, are produced in bacteria as described, for example, in U.S. Patent Nos. 5,648,237, 5,789,199, and 5,840,523, and Charlton, Methods in Molecular Biology, Vol. 248, pp. 245-254, BKC Lo, ed., Humana Press, Totowa, NJ, 2003.

[0296] Eukaryotic microbes, such as filamentous fungi or yeast, are suitable expression host cells in certain embodiments, particularly fungal and yeast strains in which the glycosylation pathway has been "humanized," resulting in the production of antibodies with partial or fully human glycosylation patterns (see, e.g., Gerngross, 2004, Nat. Biotech. 22:1409-1414, and Li et al., 2006, Nat. Biotech. 24:210-215).

[0297] Suitable host cells for the expression of glycosylated fusion proteins are usually eukaryotic cells. For example, U.S. Patent Nos. 5,959,177, 6,040,498, 6,420,548, 7,125,978 and 6,417,429 describe the PLANTIBODIES™ technology for producing antibodies in transgenic plants. For the expression of fusion proteins, mammalian cell lines adapted to grow in suspension are particularly useful. Examples include SV40-transformed monkey kidney line CV1 (COS-7), human embryonic kidney (HEK) line 293 or 293 cells (see, e.g., Graham et al., 1977, J. Gen Virol., 36:59), baby hamster kidney cells (BHK), mouse Sertoli TM4 cells (see, e.g., Mather, 1980, Biol Reprod., 23:243-251); monkey kidney cells (CV1), African green monkey kidney cells (VERO-76), human cervical carcinoma (HeLa) cells, canine kidney cells (MDCK), buffalo rat liver cells (BRL3A), human lung cells (W138), human liver cells (HepG2), mouse mammary tumor (MMT060562), TRI cells (see, e.g., Mather et al., 1982, Annals of NY Acad. Sci, 383: 44-68), MRC5 cells, FS4 cells, Chinese hamster ovary (CHO) cells (DHFR - Examples of suitable mammalian host cell lines for antibody production include, but are not limited to, CHO cells; see Urlaub et al., 1980, Proc Natl Acad Sci USA, 77:4216), and myeloma cell lines (such as Y0, NS0, and Sp2 / 0). Exemplary mammalian host cell lines suitable for antibody production are reviewed in Yazaki & Wu, Methods in Molecular Biology, Vol. 248, pp. 255-268 (BKC Lo, ed., Humana Press, Totowa, NJ, 2003).

[0298] In certain embodiments, the host cell is a transient or stable higher eukaryotic cell line, such as a mammalian cell line. In some embodiments, the host cell is a mammalian HEK293T cell, CHO cell, HeLa cell, NS0 cell, or COS cell. In some embodiments, the host cell is a stable cell line that allows mature glycosylation of the fusion protein.

[0299] Host cells containing an expression vector(s) encoding a fusion protein can be cultured using conventional methods to produce the fusion protein. Alternatively, in some embodiments, host cells containing an expression vector(s) encoding a fusion protein can be used to deliver the fusion protein therapeutically or prophylactically to a subject, or the polynucleotide or expression vector can be administered ex vivo to cells from a subject, which can then be returned to the subject.

[0300] In some embodiments, the host cell comprises (e.g., is transformed with) a vector comprising a polynucleotide encoding the VL of a binding domain and the VH of a binding domain described herein. In some embodiments, the host cell comprises a first vector comprising a polynucleotide encoding the VL of a binding domain described herein and a second vector comprising a polynucleotide encoding the VH of the corresponding binding domain. In some embodiments, the host cell is a eukaryotic organism, such as a Chinese hamster ovary (CHO) cell, a human embryonic kidney (HEK) cell, or a lymphoid cell (e.g., a YO, NS0, or Sp20 cell).

[0301] In certain embodiments, the host cells are Expi293™ (Thermo Fisher, Waltham, Mass.). In certain embodiments, the host cells are CHO-S cells (National Research Council Canada) or HEK293 cells.

[0302] Certain embodiments of the present disclosure relate to methods of making a fusion protein, comprising culturing a host cell into which one or more polynucleotides encoding the fusion protein or one or more expression vectors encoding the fusion protein have been introduced under conditions suitable for expression of the fusion protein, and optionally recovering the fusion protein from the host cell (or from the host cell culture medium).

[0303] Cell culture media that may be used include, but are not limited to, DMEM (Thermo Fisher, Waltham, MA), Opti-MEM™ (Thermo Fisher, Waltham, MA), Opti-MEM™ I Reduced Serum Medium (Thermo Fisher, Waltham, MA), RPMI-1640 medium, Expi293™ Expression Medium (Thermo Fisher, Waltham, MA), and FreeStyle CHO Expression Medium (Thermo Fisher Scientific, Waltham, MA).

[0304] The cell culture medium may be supplemented with serum, e.g., fetal bovine serum (FBS), amino acids, e.g., L-glutamine, antibiotics, e.g., penicillin and streptomycin, and / or antimycotics, e.g., amphotericin, or any other additives routinely used to support cell culture.

[0305] Purification of fusion proteins Typically, fusion proteins are purified after expression. Proteins can be isolated or purified by various methods known to those skilled in the art (see, for example, Protein Purification: Principles and Practice, 3). rd(See, Ed., Scopes, Springer-Verlag, NY, 1994). Standard purification methods include chromatographic techniques, including ion exchange, hydrophobic interaction, affinity, size or gel filtration, and reversed phase, performed at atmospheric or elevated pressure using systems such as FPLC and HPLC. Additional purification methods include electrophoretic, immunological, precipitation, dialysis, and chromatofocusing techniques. Ultrafiltration and diafiltration techniques, coupled with protein concentration, are also useful. As is well known in the art, various natural proteins bind to Fc and antibodies, and these proteins are used to purify specific antibodies. For example, bacterial proteins A and G bind to the Fc region. Similarly, bacterial protein L binds to the Fab region of some antibodies. Purification can also be enabled by specific fusion partners. For example, antibodies can be purified using glutathione resins when GST fusions are used, or Ni-tags when His-tags are used. +2 Purification can be achieved using affinity chromatography or, if a flag tag is used, immobilized anti-flag antibodies. The degree of purification required will vary depending on the use of the antibody. In some cases, no purification may be necessary.

[0306] In certain embodiments, the fusion protein is substantially pure. The term "substantially pure" (or "substantially purified"), when used with respect to a fusion protein described herein, means that the fusion protein is substantially or essentially free from components that normally accompany or interact with the protein as found in its naturally occurring environment, e.g., native cells, or, in the case of a recombinantly produced fusion protein, host cells. In certain embodiments, a substantially pure fusion protein is a protein preparation having less than about 30%, less than about 25%, less than about 20%, less than about 15%, less than about 10%, or less than about 5% (by dry weight) of contaminating proteins.

[0307] Protein purification and / or assessment of homogeneity can be performed by any method known in the art, including, but not limited to, non-reducing / reducing CE-SDS, non-reducing / reducing SDS-PAGE, ultra-high performance liquid chromatography-size exclusion chromatography (UPLC-SEC), high performance liquid chromatography (HPLC), mass spectrometry, multi-angle light scattering (MALS), dynamic light scattering (DLS).

[0308] Post-translational modifications In certain embodiments, the fusion proteins described herein comprise one or more post-translational modifications, which may occur in vivo or may be performed in vitro after the fusion protein is isolated from a host cell.

[0309] Post-translational modifications include various modifications known in the art (see, e.g., Proteins—Structure and Molecular Properties, 2nd Ed., TECreighton, W.H. Freeman and Company, New York, 1993; Post-Translational Covalent Modification of Proteins, B.C. Johnson, Ed., Academic Press, New York, pp. 1-12, 1983; Seifter et al., 1990, Meth. Enzymol., 182:626-646, and Rattan et al., 1992, Ann. N.Y. Acad. Sci., 663:48-62). In those embodiments in which the fusion protein contains one or more post-translational modifications, the fusion protein may contain the same type of modification at one or several sites, or may contain different modifications at different sites.

[0310] Examples of post-translational modifications include glycosylation, acetylation, phosphorylation, amidation, derivatization with known protecting / blocking groups, formylation, oxidation, reduction, proteolytic cleavage or specific chemical cleavage (by cyanogen bromide, trypsin, chymotrypsin, papain, V8 protease or NaBH4).

[0311] Other examples of post-translational modifications include, for example, the addition or removal of N- or O-linked glycans, chemical modification of N- or O-linked glycans, N- or C-terminal processing, conjugation of chemical moieties to the amino acid backbone, and the addition or deletion of N-terminal methionine residues resulting from expression in prokaryotic host cells. Post-translational modifications can also include modification with detectable labels, such as enzymatic, fluorescent, isotopic, or affinity labels, to enable protein detection and isolation. Examples of suitable enzymatic labels include, but are not limited to, horseradish peroxidase, alkaline phosphatase, beta-galactosidase, and acetylcholinesterase. Examples of suitable prosthetic group complexes include, but are not limited to, streptavidin / biotin and avidin / biotin. Examples of suitable fluorescent substances include, but are not limited to, umbelliferone, fluorescein, fluorescein isothiocyanate, rhodamine, dichlorotriazinylamine fluorescein, dansyl chloride, and phycoerythrin. An example of a luminescent material is luminol, examples of bioluminescent materials include luciferase, luciferin and aequorin, and examples of suitable radioactive materials include iodine, carbon, sulfur, tritium, indium, technetium, thallium, gallium, palladium, molybdenum, xenon and fluorine.

[0312] Additional examples of post-translational modifications include acylation, ADP-ribosylation, amidation, covalent attachment of flavin, covalent attachment of a heme moiety, covalent attachment of a nucleotide or nucleotide derivative, covalent attachment of a lipid or lipid derivative, covalent attachment of phosphatidylinositol, cross-linking, cyclization, disulfide bond formation, demethylation, formation of covalent cross-links, formation of cysteine, formation of pyroglutamate, gamma-carboxylation, GPI anchor formation, hydroxylation, iodination, methylation, myristylation, pegylation, prenylation, racemization, selenoylation, sulfation, transfer-RNA-mediated addition of amino acids to proteins such as arginylation, and ubiquitination.

[0313] X. Treatment method In certain aspects, the present disclosure includes a method for treating a disease or condition comprising administering to a subject in need thereof a fusion protein described herein. In certain embodiments, the subject is a mammal. In certain embodiments, the subject is a human.

[0314] In certain embodiments, the methods disclosed herein are for the treatment of cancer. Cancers may include, but are not limited to, hematopoietic malignancies (including leukemia, myeloma, and lymphoma), carcinomas (including adenocarcinoma and squamous cell carcinoma), melanoma, and sarcomas. Carcinomas and sarcomas are often referred to as "solid tumors." In certain embodiments, the cancer is a solid tumor. In certain embodiments, the cancer is a leukemia. In certain embodiments, the cancer is a lymphoma.

[0315] The fusion protein can exert either a cytotoxic or cytostatic effect, and can result in one or more of: a reduction in tumor size, a slowing or prevention of tumor growth, an increase in disease-free survival from tumor disappearance or removal until its recurrence, prevention of tumor onset or subsequent development (e.g., metastasis), an increase in progression-free time, a reduction in one or more adverse symptoms associated with the tumor, or an increase in overall survival in a tumor-bearing subject.

[0316] In certain embodiments, the methods disclosed herein are for the treatment of an immune deficiency disorder or disease.

[0317] In certain embodiments, the methods disclosed herein are for the treatment of an autoimmune disease or condition.

[0318] The methods described herein include administering a pharmaceutical composition comprising a fusion protein described herein to a subject in need thereof. The fusion protein can be administered to the subject by an appropriate route of administration. As will be understood by one of skill in the art, the route and / or mode of administration can vary depending on the desired results. Typically, immunotherapeutic antibodies are administered by systemic or local administration. Local administration can be to the tumor site or tumor-draining lymph nodes. Generally, the fusion protein is administered parenterally, for example, intravenously, intramuscularly, intradermally, intraperitoneally, subcutaneously, or spinally, for example, by injection or infusion.

[0319] Treatment is achieved by administering a "therapeutically effective amount" of the fusion protein. A "therapeutically effective amount" refers to an amount effective, at the dosage and for the period of time necessary, to achieve the desired therapeutic result. A therapeutically effective amount may vary depending on factors such as the disease state, age, sex, and weight of the subject. A therapeutically effective amount is also one in which any toxic or detrimental effects of the fusion protein are outweighed by the therapeutically beneficial effects. A "sufficient amount" means an amount sufficient to produce the desired effect, e.g., an amount sufficient to modulate the immune response against the target cell or tissue by binding of the immunomodulatory ligand-receptor to the immune cell.

[0320] Suitable dosages of pharmaceutical compositions containing fusion proteins can be determined by skilled medical practitioners. The selected dosage level will depend on various pharmacokinetic factors, including the activity of the particular fusion protein employed, the route of administration, the time of administration, the rate of excretion of the polypeptide, the duration of treatment, other drugs, compounds and / or substances used in combination with the fusion protein, e.g., anti-cancer drugs, the age, sex, weight, condition, general health and past medical history of the subject being treated, and similar factors well known in the medical arts.

[0321] Methods for modulating immune cells or immune responses In certain embodiments, the fusion proteins described herein are administered to a subject in need thereof, e.g., a subject with cancer, to modulate the subject's immune system. Thus, in certain embodiments, the fusion proteins described herein downregulate an immune response or upregulate an immune response.

[0322] According to this embodiment, administration of a sufficient amount of the fusion protein to a subject can affect one of the following to activate or upregulate an immune response: modulation of an immune checkpoint, modulation of T cell receptor signaling, modulation of T cell activation, modulation of pro-inflammatory cytokines, modulation of interferon-γ production by T cells, modulation of T cell suppression, modulation of survival and / or differentiation of M2-type tumor-associated macrophages (TAMs) or myeloid-derived suppressor cells (MDSCs), and / or modulation of a cytotoxic or cytostatic effect on cells.

[0323] In certain embodiments, provided herein are methods of modulating an immune response, including inhibiting an immune checkpoint, stimulating an immune checkpoint, activating an immune cell, stimulating T cell receptor signaling, and stimulating antibody-dependent cellular cytotoxicity (ADCC), T cell-dependent cytotoxicity (TDCC), T cell-dependent cytotoxicity (CDC), or antibody-dependent cellular phagocytosis (ADCP).

[0324] In certain embodiments, the fusion protein is capable of agonizing a costimulatory receptor on a target leukocyte. The functional effects of leukocyte costimulatory receptor agonism include activation of T effector cells. Activation of T effector cells can result in increased production by the T cell of one or more cytokines, such as interferon gamma (IFN-γ), interleukin-2 (IL-2), interleukin-12 (IL-12), interleukin-17 (IL-17), interleukin-21 (IL-21), granulocyte-macrophage colony-stimulating factor (GM-CSF), tumor necrosis factor-α (TNF-α), macrophage inflammatory protein 1 (MIP-1β), and / or C-X-C motif ligand 13 (CXCL13). Increased production of IL-21 and CXCL13 by T effector cells may, for example, support the differentiation and activation of inflammatory myeloid cells in the TME, recruit anti-tumor lymphoid cells such as B cells and NKT cells, and / or support the formation of tertiary lymphoid structures.

[0325] In certain embodiments, the fusion protein activates T effector cells, hi some embodiments, the fusion protein increases production of GM-CSF, TNF-α, MIP-1β, IL-17, IL-12, IL-21, and / or C-X-C motif ligand 13 (CXCL13) by T effector cells.

[0326] Certain embodiments of the present disclosure relate to methods of modulating costimulatory receptor agonism in leukocytes in vivo using fusion proteins, for example, to treat cancer.

[0327] In certain embodiments, the methods relate to suppressing or downregulating immune cells or immune responses, for example, to treat autoimmune diseases or disorders. Thus, in certain embodiments, the fusion protein is administered in an amount sufficient to modulate immune cells. In certain embodiments, the downregulation of the immune response is by modulating immune checkpoints, T cell receptor signaling, T cell activation, pro-inflammatory cytokines, interferon-γ production by T cells, T cell suppression, modulating survival and / or differentiation of M2-type tumor-associated macrophages (TAMs) or myeloid-derived suppressor cells (MDSCs), and / or modulating cytotoxic or cytostatic effects on cells.

[0328] Methods for Altering ADCC of Target Cells In certain embodiments, the fusion proteins described herein induce antibody-dependent cell-mediated cytotoxicity (ADCC), thereby resulting in increased lysis of target cells. In certain embodiments, the fusion proteins comprise an Fc region with increased binding affinity of the Fc to FcRγIIIa (an activating receptor), resulting in increased antibody-dependent cell-mediated cytotoxicity (ADCC) and increased lysis of target cells. In certain embodiments, the Fc region comprises a modified CH2 domain comprising an amino acid modification that results in increased binding affinity of the Fc to FcRγIIIa (an activating receptor), resulting in increased antibody-dependent cell-mediated cytotoxicity (ADCC).

[0329] In certain embodiments, the fusion proteins described herein reduce antibody-dependent cell-mediated cytotoxicity (ADCC). In certain indications, reduced or eliminated ADCC and complement-mediated cytotoxicity (CDC) are desirable. In certain embodiments, the fusion proteins comprise Fc regions that contain modified CH2 domains containing amino acid modifications that result in increased binding to FcγRIIb or that reduce or eliminate binding of the Fc region to all Fcγ receptors ("knockout" variants) may be useful. In certain embodiments, the fusion proteins comprise Fc regions with reduced binding to FcγRIIb (an inhibitory receptor).

[0330] In some embodiments, described herein are methods of treating a disease in a subject in need thereof, the method comprising administering to the subject a trispecific fusion protein comprising: (i) a first binding domain capable of binding to an antigen on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a first tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a second tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold.

[0331] In yet another embodiment, described herein are methods of killing cancer cells in a subject in need thereof, the method comprising administering to the subject a trispecific fusion protein comprising: (i) a first binding domain capable of binding to an antigen on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a first tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a second tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold.

[0332] In various embodiments, such trispecific fusion proteins used in the methods described herein may be either trivalent or tetravalent and may have formats and properties as further described herein.

[0333] In various embodiments, the anti-PD-L1 / anti-CD3 / anti-TAA trispecific fusion proteins of the present disclosure are capable of greater PD-1:PD-L1 checkpoint blockade in a cell population comprising CD3-expressing T cells and TAA- and PD-L1-expressing tumor cells, when compared to (i) a format-matched anti-CD3 / anti-TAA bispecific antibody, and / or (ii) an anti-PD-L1 agent, e.g., a format-matched anti-CD3 / anti-TAA bispecific antibody in combination with an anti-PD-L1 antibody. In some embodiments, such an increase in checkpoint blockade is about 1.2-fold, 1.5-fold, 2-fold, 3-fold, or more relative to the combination treatment, as measured, for example, using a reporter gene assay (RGA). In some embodiments, such an increase in checkpoint blockade is about 1.1-fold to about 1.5-fold greater, about 1.2-fold to about 1.7-fold greater, or about 1.2-fold to about 2-fold greater relative to the combination treatment.

[0334] XI. Pharmaceutical Compositions Fusion proteins according to the present disclosure can be formulated into pharmaceutical compositions. These compositions can contain, in addition to one or more fusion proteins, pharmaceutically acceptable excipients, carriers, buffers, stabilizers, or other materials well known to those skilled in the art. Such materials must be non-toxic and must not interfere with the efficacy of the active ingredient. The precise nature of the carrier or other material may depend on the route of administration, e.g., oral, intravenous, cutaneous or subcutaneous, nasal, intramuscular, or intraperitoneal.

[0335] Oral pharmaceutical compositions can be in the form of tablets, capsules, powders, or liquids. Tablets can contain solid carriers such as gelatin or adjuvants. Liquid pharmaceutical compositions generally contain liquid carriers such as water, petroleum, animal or vegetable oils, mineral oil, or synthetic oil. Physiological saline, dextrose, or other sugar solutions, or glycols such as ethylene glycol, propylene glycol, or polyethylene glycol can be included.

[0336] For intravenous, cutaneous or subcutaneous injection, or injection into an affected area, the active ingredient may be in the form of a parenterally acceptable aqueous solution that is pyrogen-free and has suitable pH, isotonicity, and stability. Those skilled in the art can easily prepare suitable solutions using isotonic vehicles such as sodium chloride, Ringer's injection, lactated Ringer's injection, etc. Preservatives, stabilizers, buffers, antioxidants, and / or other additives may be included as necessary.

[0337] In the case of fusion proteins according to the present disclosure administered to an individual, the administration is preferably a "therapeutically effective amount" sufficient to show benefit to the individual. A "prophylactically effective amount" can also be administered, provided that it is sufficient to show benefit to the individual. The actual amount administered, as well as the rate and duration of administration, will depend on the nature and severity of the protein aggregation disorder being treated. Prescribing treatment, e.g., determining dosage, is the responsibility of a general practitioner or other physician, and typically takes into account the disorder being treated, the condition of the individual patient, the site of delivery, the method of administration, and other factors known to the practitioner. Examples of the above-mentioned techniques and protocols can be found in Remington's Pharmaceutical Sciences, 16th edition, Osol, A. (ed), 1980.

[0338] The compositions may be administered alone or in combination with other treatments, either simultaneously or sequentially, depending on the condition being treated.

[0339] In some embodiments, described herein are pharmaceutical compositions comprising a trispecific fusion protein, the trispecific fusion protein comprising: (i) a first binding domain capable of binding to an antigen on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a first tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a second tumor cell; and (iv) a scaffold, wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold.

[0340] In various embodiments, such trispecific fusion proteins used in the pharmaceutical compositions described herein may be either trivalent or tetravalent and may have formats and properties as further described herein.

[0341] XII. Kit The present disclosure also provides kits comprising one or more trispecific fusion proteins and / or pharmaceutical compositions described herein and instructions for use. Thus, in certain embodiments, described herein are kits comprising a vector for expressing a fusion protein described herein and instructions for use. In certain embodiments, described herein are kits comprising a host cell comprising a vector for expressing the fusion protein and instructions for use. In certain embodiments, a kit comprises a purified fusion protein and instructions for use. The purified fusion protein can be lyophilized or provided in a dried form such as a powder or granules, and the kit may further contain a solvent suitable for reconstituting the lyophilized or dried component(s).

[0342] The kit typically includes a container and a label and / or package insert on or associated with the container. The label or package insert includes instructions customarily included in commercial packaging for therapeutic products, such as those providing information or instructions regarding the indications, methods of use, dosage, administration, contraindications, and / or warnings for use of the therapeutic product. The label or package insert may further include a notice in a form prescribed by a government agency regulating the manufacture, use, or sale of pharmaceutical or biological products, indicating approval by that agency for the manufacture, use, or sale for administration to humans or animals. The container holds a composition including the fusion protein. In some embodiments, the container may have a sterile access port. For example, the container may be an intravenous infusion bag or a vial having a stopper pierceable by a hypodermic injection needle.

[0343] In addition to the container holding the composition comprising the fusion protein, the kit may include one or more additional containers containing other components of the kit, such as a pharmaceutically acceptable buffer (such as bacteriostatic water for injection (BWFI), phosphate-buffered saline, Ringer's solution, or dextrose solution), other buffers, or diluents.

[0344] Suitable containers include, for example, bottles, vials, syringes, intravenous infusion bags, etc. The containers can be made of various materials, such as glass or plastic. If necessary, one or more components of the kit can be lyophilized or provided in a dry form, such as a powder or granules, and the kit can further contain a suitable solvent for reconstituting the lyophilized or dried component(s).

[0345] Kits may further include other materials desirable from a commercial or user standpoint, including filters, needles, and syringes.

[0346] The following examples are offered for illustrative purposes only and are not intended to limit the scope of the invention in any way. Efforts have been made to ensure accuracy with respect to numbers used (e.g., amounts, temperatures, etc.), but some experimental error and deviation should, of course, be allowed for.

[0347] The practice of the present disclosure will employ, unless otherwise indicated, conventional methods of protein chemistry, biochemistry, recombinant DNA technology, and pharmacology, within the skill of the art. Such techniques are fully explained in the literature, see, e.g., T.E. Creighton, Proteins: Structures and Molecular Properties (W.H. Freeman and Company, 1993); A.L. Lehninger, Biochemistry (Worth Publishers, Inc., current addition); Sambrook, et al., Molecular Cloning: A Laboratory Manual (2nd Edition, 1989); Methods In Enzymology (S. Colowick and N. Kaplan eds., Academic Press, Inc.); Remington's Pharmaceutical Sciences, 18th Edition (Easton, Pennsylvania: Mack Publishing Company, 1990); Carey and Sundberg Advanced Organic Chemistry 3 rd Ed. (Plenum Press) Vols A and B (1992).

[0348] XIII. Specific Embodiments of the Present Disclosure In various embodiments, the present disclosure relates to any one of embodiments 1-218 and combinations thereof.

[0349] Embodiment 1. (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) Skeleton 1. A trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold. The trispecific fusion protein.

[0350] Embodiment 2. The trispecific fusion protein of embodiment 1, wherein the scaffold comprises a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide.

[0351] Embodiment 3. The trispecific fusion protein of embodiment 2, wherein the dimeric Fc domain is a heterodimeric Fc domain and the amino acid sequence of the first Fc polypeptide differs from the amino acid sequence of the second Fc polypeptide by at least one amino acid residue.

[0352] Embodiment 4. The trispecific fusion protein of any one of embodiments 2-3, wherein the first binding domain is linked to the N-terminus of the first Fc polypeptide and the second binding domain is linked to the N-terminus of the second binding domain.

[0353] Embodiment 5. The trispecific fusion protein of any one of embodiments 1 to 4, wherein the first binding domain and the second binding domain are each independently a Fab or an scFv.

[0354] Embodiment 6.a) the first binding domain is a Fab and the second binding domain is an scFv; or b) the first binding domain is an scFv and the second binding domain is a Fab; or c) the first binding domain is a Fab and the second binding domain is a Fab; or d) the first binding domain is an scFv and the second binding domain is an scFv; 6. The trispecific fusion protein of embodiment 5.

[0355] Embodiment 7. The first binding domain is linked to the first Fc polypeptide via a first linker. Fc 7. The trispecific fusion protein according to any one of embodiments 2 to 6, wherein the trispecific fusion protein is linked via

[0356] Embodiment 8. The second binding domain is linked to the second Fc polypeptide via a second linker. Fc 8. The trispecific fusion protein according to any one of embodiments 2 to 7, wherein the trispecific fusion protein is linked via

[0357] Embodiment 9. The first linker Fc , the second linker Fc 9. The trispecific fusion protein of any one of embodiments 7 to 8, wherein said fusion protein comprises or consists of an IgG hinge region or a part or variant thereof.

[0358] Embodiment 10. The first linker Fc , the second linker Fc 10. The trispecific fusion protein of any one of embodiments 7 to 9, wherein either or both of said fusion proteins comprise or consist of an amino acid sequence having at least 80%, 90%, or 95% sequence identity to the amino acid sequence set forth in SEQ ID NO: 50 or a fragment thereof.

[0359] Embodiment 11. The trispecific fusion protein of any one of embodiments 2 to 10, wherein the first binding domain is a Fab and is linked to the N-terminus of the first Fc polypeptide via the C-terminus of the Fab.

[0360] Embodiment 12. The trispecific fusion protein of any one of embodiments 2 to 11, wherein the second binding domain is an scFv comprising, in an N-terminal to C-terminal direction, a VH domain linked to a VL domain, or a VL domain linked to a VH domain, and is linked via its C-terminus to the N-terminus of the second Fc polypeptide.

[0361] Embodiment 13. The trispecific fusion protein of any one of embodiments 2 to 5, wherein the fusion protein comprises a first immunoglobulin G heavy chain, a second immunoglobulin G heavy chain, and an immunoglobulin light chain, wherein the first heavy chain comprises, from N-terminal to C-terminal, a Fab VH and CH1 Domain linked to the first Fc polypeptide, the second heavy chain comprises, from N-terminal to C-terminal, scFv VH and VL or VL and VH Domains linked to the second Fc polypeptide, and the light chain comprises, from N-terminal to C-terminal, a Fab VL and CL Domain.

[0362] Embodiment 14. The trispecific fusion protein of any one of embodiments 1 to 13, wherein the third binding domain is linked to (i) the first binding domain, (ii) the second binding domain, or (iii) the scaffold.

[0363] Embodiment 15. The trispecific fusion protein of embodiment 14, wherein the third binding domain is linked to (i) the N-terminus of the Fab VH domain, (ii) the N-terminus of the Fab VL domain, (iii) the C-terminus of the CL domain, (iv) the N-terminus of the scFv VH or VL domain, (v) the C-terminus of the first Fc polypeptide, or (vi) the C-terminus of the second Fc polypeptide.

[0364] Embodiment 16. The trispecific fusion protein of any one of Embodiments 1 to 15, wherein the third binding domain comprises a PD-1 polypeptide.

[0365] Embodiment 17. The trispecific fusion protein of any one of embodiments 1 to 16, wherein the third binding domain consists of a PD-1 polypeptide.

[0366] Embodiment 18. The PD-1 polypeptide is: A wild-type PD-1 polypeptide comprising or consisting of a portion of the amino acid sequence set forth in SEQ ID NO:7 or a fragment thereof. 18. The trispecific fusion protein of any one of embodiments 16 to 17, wherein

[0367] Embodiment 19. The PD-1 polypeptide has, compared to a corresponding wild-type PD-1 polypeptide: one or more amino acid modifications that increase or decrease the binding affinity of the PD-1 polypeptide to PD-L1 compared to the binding affinity of the corresponding wild-type PD-1 polypeptide to PD-L1; 18. The trispecific fusion protein of any one of embodiments 16 to 17, comprising:

[0368] Embodiment 20. The trispecific fusion protein of embodiment 19, wherein said one or more amino acid modifications comprise one or more amino acid substitutions.

[0369] Embodiment 21. The trispecific fusion protein of Embodiment 20, wherein the PD-1 polypeptide has a binding affinity for PD-L1 of about 100 μM to about 10 pM, about 10 μM to about 150 pM, about 100 nM to 150 pM, or about 5 nM to about 90 nM.

[0370] Embodiment 22. The trispecific fusion protein of any one of Embodiments 16 to 21, wherein the PD-1 polypeptide comprises, or consists of, an amino acid sequence having at least about 80%, 90%, 95%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO:9.

[0371] Embodiment 23. The anti-CD3 binding domain comprises: a VH domain comprising an HCDR1 sequence selected from the group consisting of RSTMH (SEQ ID NO: 207), YYGMS (SEQ ID NO: 303), KYAMN (SEQ ID NO: 224), and TYAMN (SEQ ID NO: 232), an HCDR2 sequence selected from the group consisting of YINPSSAYTNYNQKFKD (SEQ ID NO: 208), SITSSGGRIYYPDSVKG (SEQ ID NO: 301), SITRSGGRIYYPDSVKG (SEQ ID NO: 217), RIRSKYNNYATYYADSVKD (SEQ ID NO: 225), and RIRSKYNNYATYYADSVKG (SEQ ID NO: 233), and an HCDR3 sequence selected from the group consisting of PQVHYDYNGFPY (SEQ ID NO: 209), DGRDGWVAY (SEQ ID NO: 275), HGNFGNSYISYWAY (SEQ ID NO: 226), and HGNFGNSYVSWFAY (SEQ ID NO: 234); a VL domain comprising an LCDR1 sequence selected from the group consisting of SASSSVSYMN (SEQ ID NO:211), KRNTGNIGSNYVN (SEQ ID NO:287), TGNTGNIGSNYVN (SEQ ID NO:220), GSSTGAVTSGNYPN (SEQ ID NO:228), and GSSTGAVTTSNYAN (SEQ ID NO:236); an LCDR2 sequence selected from the group consisting of DSSKLAS (SEQ ID NO:212), RNDKRPD (SEQ ID NO:298), RDDKRPS (SEQ ID NO:221), GTKFLAP (SEQ ID NO:229), RSYQRPS (SEQ ID NO:199), and GTNKRAP (SEQ ID NO:237); and an LCDR3 sequence selected from the group consisting of QQWSRNPPT (SEQ ID NO:214), QSYSSGFI (SEQ ID NO:295), VLWYSNRWV (SEQ ID NO:230), ATWDDSLDGWV (SEQ ID NO:200), and ALWYSNLWV (SEQ ID NO:238); 23. The trispecific fusion protein of any one of embodiments 1 to 22, comprising:

[0372] Embodiment 24. The trispecific fusion protein of any one of embodiments 1 to 23, wherein the anti-CD3 binding domain comprises a VH domain comprising the HCDR sequences YYGMS (SEQ ID NO: 303), SITSSGGRIYYPDSVKG (SEQ ID NO: 301), and DGRDGWVAY (SEQ ID NO: 275), and a VL domain comprising the LCDR sequences KRNTGNIGSNYVN (SEQ ID NO: 287), RNDKRPD (SEQ ID NO: 298), and QSYSSGFI (SEQ ID NO: 295).

[0373] Embodiment 25. The trispecific fusion protein of any one of embodiments 1 to 23, wherein the anti-CD3 binding domain comprises a VH domain comprising the HCDR sequences YYGMS (SEQ ID NO: 302), SITRSGGRIYYPDSVKG (SEQ ID NO: 217), and DGRDGWVAY (SEQ ID NO: 275), and a VL domain comprising the LCDR sequences TGNTGNIGSNYVN (SEQ ID NO: 220), RDDKRPS (SEQ ID NO: 221), and QSYSSGFI (SEQ ID NO: 295).

[0374] Embodiment 26. The trispecific fusion protein of any one of embodiments 1 to 23, wherein the anti-CD3 binding domain comprises a VH domain comprising the HCDR sequences KYAMN (SEQ ID NO: 224), RIRSKYNNYATYYADSVKD (SEQ ID NO: 225), and HGNFGNSYISYWAY (SEQ ID NO: 226), and a VL domain comprising the LCDR sequences GSSTGAVTSGNYPN (SEQ ID NO: 228), GTKFLAP (SEQ ID NO: 229), and VLWYSNRWV (SEQ ID NO: 230).

[0375] Embodiment 27. The trispecific fusion protein of any one of embodiments 1 to 23, wherein the anti-CD3 binding domain comprises a VH domain comprising the HCDR sequences TYAMN (SEQ ID NO: 232), RIRSKYNNYATYYADSVKG (SEQ ID NO: 233), and HGNFGNSYVSWFAY (SEQ ID NO: 234), and a VL domain comprising the LCDR sequences GSSTGAVTTSNYAN (SEQ ID NO: 236), GTNKRAP (SEQ ID NO: 237), and ALWYSNLWV (SEQ ID NO: 238).

[0376] Embodiment 28. The trispecific fusion protein of any one of embodiments 1 to 23, wherein the anti-CD3 binding domain comprises a VH domain comprising the HCDR sequences RSTMH (SEQ ID NO: 207), YINPSSAYTNYNQKFKD (SEQ ID NO: 208), and PQVHYDYNGFPY (SEQ ID NO: 209), and a VL domain comprising the LCDR sequences SASSSVSYMN (SEQ ID NO: 211), DSSKLAS (SEQ ID NO: 212), and QQWSRNPPT (SEQ ID NO: 214).

[0377] Embodiment 29. The trispecific fusion protein of any one of embodiments 1 to 23, wherein the anti-CD3 binding domain comprises a VH domain comprising the HCDR sequences RSTMH (SEQ ID NO: 207), YINPSSAYTNYNQKFKD (SEQ ID NO: 208), and PQVHYDYNGFPY (SEQ ID NO: 209), and a VL domain comprising the LCDR sequences RSYQRPS (SEQ ID NO: 199) and ATWDDSLDGWV (SEQ ID NO: 200).

[0378] Embodiment 30. The anti-CD3 binding domain comprises: a VH domain comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to any one of SEQ ID NOs: 2, 215, 223, and 231; a VL domain comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to any one of SEQ ID NOs: 1, 219, 227, and 235; 24. The trispecific fusion protein of any one of embodiments 1 to 23, comprising:

[0379] Embodiment 31. The trispecific fusion protein of any one of embodiments 1 to 30, wherein the TAA is not HER2.

[0380] Embodiment 32. The trispecific fusion protein of any one of embodiments 1 to 31, wherein the TAA is Cldn18.2.

[0381] Embodiment 33. The trispecific fusion protein of embodiment 32, wherein the anti-Cldn18.2 VH sequence of the second binding comprises an HCDR1 sequence comprising SNPMI (SEQ ID NO: 310), an HCDR2 sequence comprising IIDTDGSTYYADWAKG (SEQ ID NO: 311), and an HCDR3 sequence comprising RLHGSSNGYYDDL (SEQ ID NO: 312), and the anti-Cldn18.2 VL sequence of the second binding comprises an LCDR1 sequence comprising QASQSIYSYLS (SEQ ID NO: 313), an LCDR2 sequence comprising KASTLAS (SEQ ID NO: 314), and an LCDR3 sequence comprising QQGYTVTNVDKNT (SEQ ID NO: 315).

[0382] Embodiment 34. The trispecific fusion protein of any one of embodiments 32 to 33, wherein the anti-Cldn18.2 VH sequence of the second binding domain comprises an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 316, and the anti-Cldn18.2 VL sequence of the second binding domain comprises an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 317.

[0383] Embodiment 35. The trispecific fusion protein of any one of embodiments 2 to 34, wherein the first Fc polypeptide and the second Fc polypeptide each comprise a CH2 domain comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO:6.

[0384] Embodiment 36. The trispecific fusion protein of any one of embodiments 2 to 35, wherein one of the first Fc polypeptide or the second Fc polypeptide comprises a CH3 domain comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO:4, and the other Fc polypeptide comprises a CH3 domain comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO:5.

[0385] Embodiment 37. The trispecific fusion protein of embodiment 1, wherein the trispecific fusion protein is not v31929.

[0386] Embodiment 38. The trispecific fusion protein of any one of embodiments 1 to 30, wherein the TAA is HER2 and the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 22080 or 23734; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 21490; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 12985.

[0387] Embodiment 39. The trispecific fusion protein of embodiment 38, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 22080; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 21490; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 12985.

[0388] Embodiment 40. The trispecific fusion protein of embodiment 38, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23734; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 21490; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 12985.

[0389] Embodiment 41. The trispecific fusion protein of any one of embodiments 1 to 31, wherein the TAA is MSLN and the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29207, 29208, 29276, 29238, 29282, 22080 or 23734; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 23867, 23270, 29275 or 25095; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412, 23570 or 12985.

[0390] Embodiment 42. The trispecific fusion protein of embodiment 41, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29207; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29275; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0391] Embodiment 43. The trispecific fusion protein of embodiment 41, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29208; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in v29275; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0392] Embodiment 44. The trispecific fusion protein of embodiment 41, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29276; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 25095; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23570.

[0393] Embodiment 45. The trispecific fusion protein of embodiment 41, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29238; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29275; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0394] Embodiment 46. The trispecific fusion protein of any one of embodiments 1 to 3, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29282; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23270; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0395] Embodiment 47. The trispecific fusion protein of embodiment 41, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 22080; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23867; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 12985.

[0396] Embodiment 48. The trispecific fusion protein of embodiment 41, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23734; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23867; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 12985.

[0397] Embodiment 49. The trispecific fusion protein of any one of embodiments 1 to 36, wherein the TAA is Cldn18.2 and the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29241, 29238, 29208 or 29211; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29264, 29261, 29267 or 28373; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0398] Embodiment 50. The trispecific fusion protein of embodiment 49, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29241; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 28373; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0399] Embodiment 51. The trispecific fusion protein of embodiment 49, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29238; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 28373; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0400] Embodiment 52. The trispecific fusion protein of embodiment 49, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29208; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 28373; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0401] Embodiment 53. The trispecific fusion protein of embodiment 49, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29211; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 28373; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0402] Embodiment 54. The trispecific fusion protein of any one of embodiments 1 to 36, wherein the trispecific fusion protein comprises a fourth binding domain, wherein the fourth binding domain is linked, either directly or via a linker, to the first binding domain, the second binding domain, the third binding domain, or the scaffold.

[0403] Embodiment 55 The trispecific fusion protein of embodiment 54, wherein the fourth binding domain is capable of binding to the TAA.

[0404] Embodiment 56. The trispecific fusion protein of embodiment 55, wherein the second binding domain and the fourth binding domain capable of binding to the TAA are both scFv domains.

[0405] Embodiment 57. The trispecific fusion protein of embodiment 56, wherein the second binding domain and the fourth binding domain capable of binding to the TAA comprise or consist of the same anti-TAA VH and VL sequences, respectively.

[0406] Embodiment 58. The trispecific fusion protein of any one of embodiments 55 to 57, wherein the TAA is MSLN and the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 29257 or 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 23867, 29258, 29264, 23867, 29263, 29267 or 29261; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence set forth in 16412, 29226 or 29220.

[0407] Embodiment 59. The trispecific fusion protein of embodiment 58, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29257; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23867; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29220.

[0408] Embodiment 60. The trispecific fusion protein of embodiment 58, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29258; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0409] Embodiment 61. The trispecific fusion protein of embodiment 58, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29264; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0410] Embodiment 62. The trispecific fusion protein of embodiment 58, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23867; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29220.

[0411] Embodiment 63. The trispecific fusion protein of embodiment 58, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23867; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29226.

[0412] Embodiment 64. The trispecific fusion protein of embodiment 58, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29263; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0413] Embodiment 65. The trispecific fusion protein of embodiment 58, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29264; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0414] Embodiment 66. The trispecific fusion protein of embodiment 58, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29261; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0415] Embodiment 67. The trispecific fusion protein of embodiment 58, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29267; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0416] Embodiment 68. The trispecific fusion protein of any one of embodiments 55 to 57, wherein the TAA is Cldn18.2 and the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29244; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29245, 29248, 29251 or 29254; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0417] Embodiment 69. The trispecific fusion protein of embodiment 68, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29244; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29245; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0418] Embodiment 70. The trispecific fusion protein of embodiment 68, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29244; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29248; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0419] Embodiment 71. The trispecific fusion protein of embodiment 68, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29244; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29251; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in v16412.

[0420] Embodiment 72. The trispecific fusion protein of embodiment 68, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29244; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29254; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0421] Embodiment 73. The trispecific fusion protein of any one of embodiments 1 to 72, wherein the trispecific fusion protein, when bound to CD3 on the cytotoxic effector cell, a TAA on the tumor cell, and PD-L1 on the tumor cell, forms a TCR-independent immune synapse capable of inducing effector cell-mediated cytotoxicity against the tumor cell.

[0422] Embodiment 74. The trispecific fusion protein of any one of embodiments 1 to 73, wherein the trispecific fusion protein binds to the TAA and PD-L1 on the same tumor cell.

[0423] Embodiment 75. The trispecific fusion protein of any one of embodiments 1 to 73, wherein the trispecific fusion protein binds to the TAA and PD-L1 on different tumor cells.

[0424] Embodiment 76. (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain and a third binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a tumor cell; (iii) a fourth binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) Skeleton 1. A tetravalent trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, the third binding domain, and the fourth binding domain are operably linked to the scaffold. The tetravalent trispecific fusion protein.

[0425] Embodiment 77. A pharmaceutical composition comprising the trispecific fusion protein of any one of embodiments 1 to 76 and a pharmaceutically acceptable carrier, excipient, diluent, or combination thereof.

[0426] Embodiment 78. A nucleic acid molecule or set of nucleic acid molecules encoding a trispecific fusion protein according to any one of embodiments 1 to 76.

[0427] Embodiment 79. A vector or a set of vectors comprising the nucleic acid molecule or set of nucleic acid molecules described in embodiment 78.

[0428] Embodiment 80. A cell comprising a nucleic acid molecule or set of nucleic acid molecules according to embodiment 78, or a vector or set of vectors according to embodiment 79.

[0429] Embodiment 81. A method for producing a trispecific fusion protein according to any one of embodiments 1 to 76, comprising: (a) obtaining a host cell culture comprising at least one host cell comprising one or more nucleic acid molecules encoding said trispecific fusion protein; (b) recovering the trispecific fusion protein from the host cell culture; The method comprising:

[0430] Embodiment 82. The method of embodiment 81, further comprising, after step (b), purifying the trispecific fusion protein.

[0431] Embodiment 83. A method of eliciting an anti-tumor immune response in a cell population comprising cytotoxic effector cells and tumor cells, said method comprising contacting said cell population with an effective amount of the trispecific fusion protein of any one of embodiments 1 to 76, wherein said cytotoxic effector cells express CD3, and said tumor cells express TAA and PD-L1.

[0432] Embodiment 84. A method of inhibiting the proliferation of tumor cells that express a TAA and PD-L1, the method comprising contacting a cell population comprising the tumor cells and cytotoxic effector cells with an effective amount of the trispecific fusion protein of any one of embodiments 1 to 76, wherein the cytotoxic effector cells express CD3.

[0433] Embodiment 85. A method of killing tumor cells that express a TAA and PD-L1, the method comprising contacting a cell population comprising the tumor cells and cytotoxic effector cells with an effective amount of the trispecific fusion protein of any one of embodiments 1 to 76, wherein the cytotoxic effector cells express CD3.

[0434] Embodiment 86. The method of any one of embodiments 83 to 85, wherein the cytotoxic effector cells comprise T cells.

[0435] Embodiment 87. The method of any one of embodiments 83 to 86, wherein the TAA and PD-L1 are located on the same tumor cell.

[0436] Embodiment 88. The method of any one of embodiments 83 to 86, wherein the TAA and PD-L1 are located on different tumor cells.

[0437] Embodiment 89. The method of any one of embodiments 83-88, wherein binding of the CD3, TAA, and PD-L1 by the trispecific fusion protein forms an artificial TCR-independent immune synapse between the immune cell and the tumor cell(s), thereby eliciting a cytotoxic immune response of the immune cell against the tumor cell(s).

[0438] Embodiment 90. The method of any one of embodiments 83 to 89, wherein the cell population is located within a living subject.

[0439] Embodiment 91. A method for treating cancer in a subject in need thereof, comprising administering to the subject a trispecific fusion protein of any one of embodiments 1 to 76.

[0440] Embodiment 92 The method of embodiment 91, wherein the trispecific fusion protein elicits a cytotoxic immune response against the cancer in the subject, thereby treating the cancer in the subject.

[0441] Embodiment 93. The method of any one of embodiments 90 to 92, wherein the subject is a rodent, a non-human primate, or a human.

[0442] Embodiment 94. The trispecific fusion protein of any one of embodiments 1 to 76 for use in the treatment of cancer.

[0443] Embodiment 95. Use of a trispecific fusion protein according to any one of embodiments 1 to 76 in the manufacture of a medicament for the treatment of cancer.

[0444] Embodiment 96. (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a TAA on the surface of a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) Skeleton 1. A trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold; and the trispecific fusion protein is not v31929; The trispecific fusion protein.

[0445] Embodiment 97. (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a TAA on the surface of a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) Skeleton 1. A trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold; and the TAA is not HER2; The trispecific fusion protein.

[0446] Embodiment 97. The trispecific fusion protein of any one of embodiments 96 to 97, wherein the third binding domain is linked to (i) the first binding domain, (ii) the second binding domain, or (iii) the scaffold.

[0447] Embodiment 98. (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) Skeleton 1. A trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold; and the third binding domain is linked to the second binding domain; The trispecific fusion protein.

[0448] Embodiment 99. (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) Skeleton 1. A trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold; and the third binding domain is linked to the scaffold; The trispecific fusion protein.

[0449] Embodiment 100. (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of the first tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a second tumor cell; and (iv) Skeleton 1. A trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold; and the third binding domain is linked to (i) the second binding domain or (ii) the scaffold; The trispecific fusion protein.

[0450] Embodiment 101. a) the first binding domain is a Fab and the second binding domain is an scFv; or b) the first binding domain is an scFv and the second binding domain is a Fab; or c) the first binding domain is a Fab and the second binding domain is a Fab; or d) the first binding domain is an scFv and the second binding domain is an scFv; 101. The trispecific fusion protein of any one of embodiments 96 to 100.

[0451] Embodiment 102. (i) a first binding domain that is a Fab domain, capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) Skeleton 1. A trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold; and the third binding domain is linked to (i) the N-terminus or C-terminus of the Fab light chain, (ii) the second binding domain, or (iii) the scaffold; The trispecific fusion protein.

[0452] Embodiment 103. The trispecific fusion protein of any one of embodiments 96 to 102, wherein the scaffold comprises or consists of a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide.

[0453] Embodiment 104. The trispecific fusion protein of embodiment 103, wherein the dimeric Fc domain is a heterodimeric Fc domain and the amino acid sequence of the first Fc polypeptide differs from the amino acid sequence of the second Fc polypeptide by at least one amino acid residue.

[0454] Embodiment 105. The first binding domain is linked to the first Fc polypeptide via a first linker. Fc 105. The trispecific fusion protein of any one of embodiments 103 to 104, wherein the trispecific fusion protein is linked via

[0455] Embodiment 106. The second binding domain is linked to the second Fc polypeptide via a second linker. Fc 106. The trispecific fusion protein of any one of embodiments 103 to 105, wherein the trispecific fusion protein is linked via

[0456] Embodiment 107. The first linker Fc , the second linker Fc 107. The trispecific fusion protein of any one of embodiments 105-106, wherein said fusion protein comprises or consists of an IgG hinge region or a part or variant thereof.

[0457] Embodiment 108. The first linker Fc , the second linker Fc 108. The trispecific fusion protein of any one of embodiments 105 to 107, wherein either or both of said fusion proteins comprise or consist of an amino acid sequence having at least 80%, 90%, or 95% sequence identity to the amino acid sequence set forth in SEQ ID NO: 50 or a fragment thereof.

[0458] Embodiment 109. The trispecific fusion protein of any one of embodiments 96 to 108, wherein the first binding domain is a Fab domain and is linked to the N-terminus of the first Fc polypeptide via the C-terminus of the Fab heavy chain.

[0459] Embodiment 110. The trispecific fusion protein of any one of embodiments 96 to 109, wherein the second binding domain is an scFv domain and is linked via its C-terminus to the N-terminus of the second Fc polypeptide.

[0460] Embodiment 111. The trispecific fusion protein of any one of embodiments 96 to 110, wherein the trispecific fusion protein comprises or consists of three polypeptide chains comprising two immunoglobulin G heavy chains and one immunoglobulin light chain, wherein the first heavy chain comprises, from N-terminal to C-terminal, a Fab VH and CH1 Domain linked to the CH2 and CH3 Domain of the first Fc polypeptide; the second heavy chain comprises, from N-terminal to C-terminal, an scFv VH and VL or VL and VH Domain linked to the CH2 and CH3 Domain of the second Fc polypeptide; and the light chain comprises, from N-terminal to C-terminal, a Fab VL and CL Domain, wherein the light chain is capable of forming a Fab domain with the Fab VH and CH1 Domain of the first heavy chain.

[0461] Embodiment 112. The trispecific fusion protein of any one of embodiments 96 to 111, wherein the third binding domain comprises a PD-1 polypeptide.

[0462] Embodiment 113. The trispecific fusion protein of any one of embodiments 96 to 112, wherein the third binding domain consists of a PD-1 polypeptide.

[0463] Embodiment 114. The PD-1 polypeptide is: A wild-type PD-1 polypeptide comprising or consisting of a portion of the amino acid sequence set forth in SEQ ID NO:7 or a fragment thereof. 114. The trispecific fusion protein of any one of embodiments 112 to 113, wherein

[0464] Embodiment 115. The PD-1 polypeptide has a nucleotide sequence similar to that of a corresponding wild-type PD-1 polypeptide (e.g., SEQ ID NO: 7): one or more amino acid modifications that increase or decrease the binding affinity of the PD-1 polypeptide to PD-L1 compared to the binding affinity of the corresponding wild-type PD-1 polypeptide to PD-L1; 114. The trispecific fusion protein of any one of embodiments 112-113, comprising:

[0465] Embodiment 116. The trispecific fusion protein of embodiment 115, wherein the one or more amino acid modifications comprise one or more amino acid substitutions.

[0466] Embodiment 117. The trispecific fusion protein of Embodiment 116, wherein the PD-1 polypeptide has a binding affinity for PD-L1 of about 100 μM to about 10 pM, about 10 μM to about 150 pM, about 100 nM to 150 pM, or about 5 nM to about 90 nM.

[0467] Embodiment 118. The trispecific fusion protein of any one of Embodiments 112 to 117, wherein the PD-1 polypeptide comprises or consists of an amino acid sequence having at least about 80%, 90%, 95%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO:9.

[0468] Embodiment 119. The anti-CD3 binding domain comprises: a VH domain comprising an HCDR1 sequence selected from the group consisting of RSTMH (SEQ ID NO: 207), YYGMS (SEQ ID NO: 303), KYAMN (SEQ ID NO: 224), and TYAMN (SEQ ID NO: 232), an HCDR2 sequence selected from the group consisting of YINPSSAYTNYNQKFKD (SEQ ID NO: 208), SITSSGGRIYYPDSVKG (SEQ ID NO: 301), SITRSGGRIYYPDSVKG (SEQ ID NO: 217), RIRSKYNNYATYYADSVKD (SEQ ID NO: 225), and RIRSKYNNYATYYADSVKG (SEQ ID NO: 233), and an HCDR3 sequence selected from the group consisting of PQVHYDYNGFPY (SEQ ID NO: 209), DGRDGWVAY (SEQ ID NO: 275), HGNFGNSYISYWAY (SEQ ID NO: 226), and HGNFGNSYVSWFAY (SEQ ID NO: 234); a VL domain comprising an LCDR1 sequence selected from the group consisting of SASSSVSYMN (SEQ ID NO:211), KRNTGNIGSNYVN (SEQ ID NO:287), TGNTGNIGSNYVN (SEQ ID NO:220), GSSTGAVTSGNYPN (SEQ ID NO:228), and GSSTGAVTTSNYAN (SEQ ID NO:236); an LCDR2 sequence selected from the group consisting of DSSKLAS (SEQ ID NO:212), RNDKRPD (SEQ ID NO:298), RDDKRPS (SEQ ID NO:221), GTKFLAP (SEQ ID NO:229), RSYQRPS (SEQ ID NO:199), and GTNKRAP (SEQ ID NO:237); and an LCDR3 sequence selected from the group consisting of QQWSRNPPT (SEQ ID NO:214), QSYSSGFI (SEQ ID NO:295), VLWYSNRWV (SEQ ID NO:230), ATWDDSLDGWV (SEQ ID NO:200), and ALWYSNLWV (SEQ ID NO:238); 119. The trispecific fusion protein of any one of embodiments 96 to 118, comprising:

[0469] Embodiment 120. The trispecific fusion protein of any one of embodiments 96 to 119, wherein the anti-CD3 binding domain comprises a VH domain comprising the HCDR sequences YYGMS (SEQ ID NO: 303), SITSSGGRIYYPDSVKG (SEQ ID NO: 301), and DGRDGWVAY (SEQ ID NO: 275), and a VL domain comprising the LCDR sequences KRNTGNIGSNYVN (SEQ ID NO: 287), RNDKRPD (SEQ ID NO: 298), and QSYSSGFI (SEQ ID NO: 295).

[0470] Embodiment 121. The trispecific fusion protein of any one of embodiments 96 to 119, wherein the anti-CD3 binding domain comprises a VH domain comprising the HCDR sequences YYGMS (SEQ ID NO: 302), SITRSGGRIYYPDSVKG (SEQ ID NO: 217), and DGRDGWVAY (SEQ ID NO: 275), and a VL domain comprising the LCDR sequences TGNTGNIGSNYVN (SEQ ID NO: 220), RDDKRPS (SEQ ID NO: 221), and QSYSSGFI (SEQ ID NO: 295).

[0471] Embodiment 122. The trispecific fusion protein of any one of embodiments 96 to 119, wherein the anti-CD3 binding domain comprises a VH domain comprising the HCDR sequences KYAMN (SEQ ID NO: 224), RIRSKYNNYATYYADSVKD (SEQ ID NO: 225), and HGNFGNSYISYWAY (SEQ ID NO: 226), and a VL domain comprising the LCDR sequences GSSTGAVTSGNYPN (SEQ ID NO: 228), GTKFLAP (SEQ ID NO: 229), and VLWYSNRWV (SEQ ID NO: 230).

[0472] Embodiment 123. The trispecific fusion protein of any one of embodiments 96 to 119, wherein the anti-CD3 binding domain comprises a VH domain comprising the HCDR sequences TYAMN (SEQ ID NO: 232), RIRSKYNNYATYYADSVKG (SEQ ID NO: 233), and HGNFGNSYVSWFAY (SEQ ID NO: 234), and a VL domain comprising the LCDR sequences GSSTGAVTTSNYAN (SEQ ID NO: 236), GTNKRAP (SEQ ID NO: 237), and ALWYSNLWV (SEQ ID NO: 238).

[0473] Embodiment 124. The trispecific fusion protein of any one of embodiments 96 to 119, wherein the anti-CD3 binding domain comprises a VH domain comprising the HCDR sequences RSTMH (SEQ ID NO: 207), YINPSSAYTNYNQKFKD (SEQ ID NO: 208), and PQVHYDYNGFPY (SEQ ID NO: 209), and a VL domain comprising the LCDR sequences SASSSVSYMN (SEQ ID NO: 211), DSSKLAS (SEQ ID NO: 212), and QQWSRNPPT (SEQ ID NO: 214).

[0474] Embodiment 125. The trispecific fusion protein of any one of embodiments 96 to 119, wherein the anti-CD3 binding domain comprises a VH domain comprising the HCDR sequences RSTMH (SEQ ID NO: 207), YINPSSAYTNYNQKFKD (SEQ ID NO: 208), and PQVHYDYNGFPY (SEQ ID NO: 209), and a VL domain comprising the LCDR sequences RSYQRPS (SEQ ID NO: 199) and ATWDDSLDGWV (SEQ ID NO: 200).

[0475] Embodiment 126. The anti-CD3 binding domain comprises: a VH domain comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to any one of SEQ ID NOs: 2, 215, 223, and 231; a VL domain comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to any one of SEQ ID NOs: 1, 219, 227, and 235; 120. The trispecific fusion protein of any one of embodiments 96 to 119, comprising:

[0476] Embodiment 127. The trispecific fusion protein of any one of embodiments 96 to 126, wherein said TAA is not HER2.

[0477] Embodiment 128. The trispecific fusion protein of any one of embodiments 96 to 127, wherein the TAA is Cldn18.2.

[0478] Embodiment 129. The trispecific fusion protein of any one of embodiments 96 or 98 to 127, wherein the TAA is HER2, and the anti-HER2 VH sequence of the second binding domain comprises an HCDR1 sequence comprising DTYIH (SEQ ID NO: 121), an HCDR2 sequence comprising RIYPTNGYTRYADSVKG (SEQ ID NO: 122), and an HCDR3 sequence comprising WGGDGFYAMDY (SEQ ID NO: 123), and the anti-HER2 VL sequence of the second binding domain comprises an LCDR1 sequence comprising RASQDVNTAVA (SEQ ID NO: 125), an LCDR2 sequence comprising SASFLYS (SEQ ID NO: 126), and an LCDR3 sequence comprising QQHYTTPPT (SEQ ID NO: 127).

[0479] Embodiment 130. The anti-HER2 binding domain comprises: a VH domain comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NOs: 120 and 231; a VL domain comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 124; 130. The trispecific fusion protein of embodiment 129, comprising:

[0480] Embodiment 131. The trispecific fusion protein of any one of embodiments 96 to 127, wherein the TAA is MSLN, and the anti-MSLN VH sequence of the second binding domain comprises an HCDR1 sequence comprising GYTMN (SEQ ID NO: 286), an HCDR2 sequence comprising LITPYNGASSYNQKFRG (SEQ ID NO: 288), and an HCDR3 sequence comprising GGYDGRGFDY (SEQ ID NO: 285), and the anti-MSLN VL sequence of the second binding domain comprises an LCDR1 sequence comprising SASSSVSYMH (SEQ ID NO: 300), an LCDR2 sequence comprising DTSKLAS (SEQ ID NO: 279), and an LCDR3 sequence comprising QQWSGYPLT (SEQ ID NO: 294).

[0481] Embodiment 132. The trispecific fusion protein of embodiment 131, wherein the anti-MSLN VH sequence of the second binding comprises an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 297, and the anti-MSLN VL sequence of the second binding comprises an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 276 or 277.

[0482] Embodiment 133. The trispecific fusion protein of embodiment 128, wherein the TAA is Cldn18.2, and the anti-Cldn18.2 VH sequence of the second binding domain comprises an HCDR1 sequence comprising SNPMI (SEQ ID NO: 310), an HCDR2 sequence comprising IIDTDGSTYYADWAKG (SEQ ID NO: 311), and an HCDR3 sequence comprising RLHGSSNGYYDDL (SEQ ID NO: 312), and the anti-Cldn18.2 VL sequence of the second binding domain comprises an LCDR1 sequence comprising QASQSIYSYLS (SEQ ID NO: 313), an LCDR2 sequence comprising KASTLAS (SEQ ID NO: 314), and an LCDR3 sequence comprising QQGYTVTNVDKNT (SEQ ID NO: 315).

[0483] Embodiment 134. The trispecific fusion protein of embodiment 133, wherein the anti-Cldn18.2 VH sequence of the second binding domain comprises an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 316, and the anti-Cldn18.2 VL sequence of the second binding domain comprises an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 317.

[0484] Embodiment 135. The trispecific fusion protein of any one of embodiments 103 to 134, wherein the first Fc polypeptide and the second Fc polypeptide each comprise a CH2 domain comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO:6.

[0485] Embodiment 136. The trispecific fusion protein of any one of embodiments 103 to 135, wherein one of the first Fc polypeptide or the second Fc polypeptide comprises a CH3 domain comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO:4, and the other Fc polypeptide comprises a CH3 domain comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO:5.

[0486] Embodiment 137. The trispecific fusion protein of any one of embodiments 96 or 100-103, wherein the TAA is HER2 and the trispecific fusion protein comprises: (i) a 96th heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23734; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 21490; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 12985.

[0487] Embodiment 138. The TAA is MSLN, and the trispecific fusion protein comprises (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to a clone sequence described in 29207, 29208, 29276, 29238, 29282, 22080, or 23734, (ii) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to a clone sequence described in 23867, 23270, 29275, or 25095. 104. The trispecific fusion protein of any one of embodiments 96-97 or 100-103, comprising (iii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the cloned sequence, and (iv) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the cloned sequence set forth in 16412, 23570 or 12985.

[0488] Embodiment 139. The trispecific fusion protein of any one of embodiments 96 to 100, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29207; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29275; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0489] Embodiment 140. The trispecific fusion protein of any one of embodiments 96 to 100, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29208; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29275; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0490] Embodiment 141. The trispecific fusion protein of any one of embodiments 96 to 100, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29276; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 25095; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23570.

[0491] Embodiment 142. The trispecific fusion protein of any one of embodiments 96 to 100 or 102 to 103, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29238; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29275; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0492] Embodiment 143. The trispecific fusion protein of any one of embodiments 96 to 97, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29282; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23270; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0493] Embodiment 144. The trispecific fusion protein of any one of embodiments 96 to 100, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 22080; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23867; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 12985.

[0494] Embodiment 145. The trispecific fusion protein of any one of embodiments 96 to 100, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23734; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23867; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 12985.

[0495] Embodiment 146. The trispecific fusion protein of any one of embodiments 96 to 103, wherein the TAA is Cldn18.2 and the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29241, 29238, 29208 or 29211; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29264, 29261, 29267 or 28373; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0496] Embodiment 147. The trispecific fusion protein of any one of embodiments 96 to 100 or 103, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29241; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 28373; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0497] Embodiment 148. The trispecific fusion protein of any one of embodiments 96 to 100 or 103, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29238; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 28373; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0498] Embodiment 149. The trispecific fusion protein of any one of embodiments 96 to 100, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29208; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 28373; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0499] Embodiment 150. The trispecific fusion protein of any one of embodiments 96 to 100, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29211; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 28373; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0500] Embodiment 151. (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain and a third binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a tumor cell; (iii) a fourth binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) Skeleton 1. A tetravalent trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, the third binding domain, and the fourth binding domain are operably linked to the scaffold. The tetravalent trispecific fusion protein.

[0501] Embodiment 152. The tetravalent trispecific fusion protein of embodiment 151, wherein the scaffold comprises or consists of a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide.

[0502] Embodiment 153. The tetravalent trispecific fusion protein of embodiment 152, wherein the dimeric Fc domain is a heterodimeric Fc domain and the amino acid sequence of the first Fc polypeptide differs from the amino acid sequence of the second Fc polypeptide by at least one amino acid residue.

[0503] Embodiment 154. a) said first binding domain is a Fab and said second and said third binding domains are both scFv domains; or b) the first binding domain is an scFv and the second and third binding domains are both Fab domains; or c) the first binding domain, the second binding domain, and the third binding domain are Fab domains; or d) The tetravalent trispecific fusion protein according to any one of embodiments 151 to 153, wherein the first binding domain, the second binding domain, and the third binding domain are scFv domains.

[0504] Embodiment 155. A tetravalent trispecific fusion protein according to any one of embodiments 151 to 154, wherein (i) the first binding domain is a Fab domain, (ii) the second binding domain is a first scFv domain, and (iii) the third binding domain is a second scFv domain.

[0505] Embodiment 156. The tetravalent trispecific fusion protein of any one of embodiments 151 to 155, wherein the first binding domain is linked to the N-terminus of the first Fc polypeptide and the second binding domain is linked to the N-terminus of the second Fc polypeptide.

[0506] Embodiment 157. The tetravalent trispecific fusion protein of any one of embodiments 151 to 156, wherein the third binding domain is linked to (i) the first binding domain, (ii) the second binding domain, or (iii) the scaffold.

[0507] Embodiment 158. (a) the first binding domain is a Fab domain comprising a Fab heavy chain and a Fab light chain, wherein the C-terminus of the Fab heavy chain is linked to the N-terminus of the first Fc polypeptide; (b) the second binding domain is a first scFv domain comprising, from N- to C-terminal, a VH-linker-VL or VL-linker-VH structure, wherein the C-terminus of the first scFv domain is linked to the N-terminus of the second Fc polypeptide; (c) the third binding domain is a second scFv domain comprising, in the N-terminal to C-terminal direction, the structure VH-linker-VL or VL-linker-VH, wherein the C-terminus of the second scFv domain is linked to (i) the N-terminus of the Fab heavy or light chain, or (ii) the C-terminus of the first or second Fc polypeptide; 158. A tetravalent trispecific fusion protein according to any one of embodiments 151 to 157.

[0508] Embodiment 159. The tetravalent trispecific fusion protein of any one of embodiments 151 to 158, wherein the second binding domain and the third binding domain are capable of binding to the same TAA.

[0509] Embodiment 160. The tetravalent trispecific fusion protein of embodiment 159, wherein the second binding domain and the third binding domain are capable of binding to the same epitope on the TAA.

[0510] Embodiment 161. The tetravalent trispecific fusion protein of any one of embodiments 151 to 160, wherein the second binding domain and the third binding domain comprise the same anti-TAA VH and VL sequences.

[0511] Embodiment 162. The tetravalent trispecific fusion protein of any one of embodiments 151 to 161, wherein the fourth binding domain is linked to (i) the first binding domain, (ii) the second binding domain, (iii) the third binding domain, or (iv) the scaffold.

[0512] Embodiment 163. The tetravalent trispecific fusion protein of any one of embodiments 151 to 155, wherein the TAA is MSLN and the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29257 or 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23867, 29258, 29264, 23867, 29263, 29267 or 29261; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412, 29226 or 29220.

[0513] Embodiment 164. The tetravalent trispecific fusion protein of embodiment 163, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29257; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23867; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29220.

[0514] Embodiment 165. The tetravalent trispecific fusion protein of embodiment 163, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29258; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0515] Embodiment 166. The tetravalent trispecific fusion protein of embodiment 163, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29264; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0516] Embodiment 167. The tetravalent trispecific fusion protein of embodiment 163, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23867; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29220.

[0517] Embodiment 168. The tetravalent trispecific fusion protein of embodiment 163, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 23867; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29226.

[0518] Embodiment 169. The tetravalent trispecific fusion protein of embodiment 163, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29263; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0519] Embodiment 170. The tetravalent trispecific fusion protein of embodiment 163, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29264; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0520] Embodiment 171. The tetravalent trispecific fusion protein of embodiment 163, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29261; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0521] Embodiment 172. The tetravalent trispecific fusion protein of embodiment 163, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29283; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29267; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0522] Embodiment 173. The tetravalent trispecific fusion protein of any one of embodiments 151 to 155, wherein the TAA is Cldn18.2 and the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29244; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29245, 29248, 29251 or 29254; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0523] Embodiment 174. The tetravalent trispecific fusion protein of embodiment 173, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29244; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29245; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0524] Embodiment 175. The tetravalent trispecific fusion protein of embodiment 173, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29244; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29248; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0525] Embodiment 176. The tetravalent trispecific fusion protein of embodiment 173, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29244; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29251; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in v16412.

[0526] Embodiment 177. The tetravalent trispecific fusion protein of embodiment 173, wherein the trispecific fusion protein comprises: (i) a first heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29244; (ii) a second heavy chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 29254; and (iii) a light chain having at least 90%, 95%, 97%, 99% or 100% sequence identity to the clone sequence described in 16412.

[0527] Embodiment 178. The trispecific fusion protein of any one of embodiments 96 to 150 or the tetravalent trispecific fusion protein of any one of embodiments 151 to 177, wherein the trispecific fusion protein, when bound to CD3 on the cytotoxic effector cell, a TAA on the tumor cell, and PD-L1 on the tumor cell, forms a TCR-independent immune synapse capable of inducing effector cell-mediated cytotoxicity against the tumor cell.

[0528] Embodiment 179. The trispecific fusion protein of any one of embodiments 96 to 150 or the tetravalent trispecific fusion protein of any one of embodiments 151 to 177, wherein the trispecific fusion protein binds to a TAA and PD-L1 on the same tumor cell.

[0529] Embodiment 180. The trispecific fusion protein of any one of embodiments 96 to 150 or the tetravalent trispecific fusion protein of any one of embodiments 151 to 177, wherein the trispecific fusion protein binds to a TAA and PD-L1 on different tumor cells.

[0530] Embodiment 181. A pharmaceutical composition comprising a trispecific fusion protein according to any one of embodiments 96 to 150 or a tetravalent trispecific fusion protein according to any one of embodiments 151 to 177, and a pharmaceutically acceptable carrier, excipient, diluent, or combination thereof.

[0531] Embodiment 182. A nucleic acid molecule or a set of nucleic acid molecules encoding a trispecific fusion protein according to any one of embodiments 96 to 150 or a tetravalent trispecific fusion protein according to any one of embodiments 151 to 177.

[0532] Embodiment 183. A vector or a set of vectors comprising the nucleic acid molecule or set of nucleic acid molecules according to embodiment 182.

[0533] Embodiment 184. A cell comprising a nucleic acid molecule or a set of nucleic acid molecules according to embodiment 182, or a vector or a set of vectors according to embodiment 183.

[0534] Embodiment 185. A method for producing a trispecific fusion protein according to any one of embodiments 96 to 150 or a tetravalent trispecific fusion protein according to any one of embodiments 151 to 177, comprising: (a) obtaining a host cell culture comprising at least one host cell comprising one or more nucleic acid molecules encoding said trispecific fusion protein; (b) recovering the trispecific fusion protein from the host cell culture; The method comprising:

[0535] Embodiment 186. The method of embodiment 185, further comprising, after step (b), purifying the trispecific fusion protein.

[0536] Embodiment 187. A method for inducing an anti-tumor immune response in a cell population comprising cytotoxic effector cells and tumor cells, said method comprising contacting said cell population with an effective amount of the trispecific fusion protein of any one of embodiments 96 to 150 or the tetravalent trispecific fusion protein of any one of embodiments 151 to 177, wherein said cytotoxic effector cells express CD3 and said tumor cells express TAA and PD-L1.

[0537] Embodiment 188. A method for inhibiting the proliferation of tumor cells that express a TAA and PD-L1, the method comprising contacting a cell population comprising the tumor cells and cytotoxic effector cells with an effective amount of the trispecific fusion protein of any one of embodiments 96 to 150 or the tetravalent trispecific fusion protein of any one of embodiments 151 to 177, wherein the cytotoxic effector cells express CD3.

[0538] Embodiment 189. A method for killing tumor cells expressing a TAA and PD-L1, the method comprising contacting a cell population comprising the tumor cells and cytotoxic effector cells with an effective amount of the trispecific fusion protein of any one of embodiments 96 to 150 or the tetravalent trispecific fusion protein of any one of embodiments 151 to 177, wherein the cytotoxic effector cells express CD3.

[0539] Embodiment 190. The method of any one of embodiments 187-189, wherein the cytotoxic effector cells comprise T cells.

[0540] Embodiment 191. The method of any one of embodiments 187 to 190, wherein the TAA and PD-L1 are located on the same tumor cell.

[0541] Embodiment 192. The method of any one of embodiments 187 to 191, wherein the TAA and PD-L1 are located on different tumor cells.

[0542] Embodiment 193. The method of any one of embodiments 187 to 192, wherein the binding of CD3, TAA, and PD-L1 forms an artificial TCR-independent immune synapse between the immune cell and the tumor cell, thereby eliciting a cytotoxic immune response of the immune cell against the tumor cell.

[0543] Embodiment 194. The method of any one of embodiments 187 to 193, wherein the cell population is located in a living subject.

[0544] Embodiment 195. A method for treating cancer in a subject in need thereof, comprising administering to the subject a trispecific fusion protein of any one of embodiments 96 to 150 or a tetravalent trispecific fusion protein of any one of embodiments 151 to 177.

[0545] Embodiment 196. The method of embodiment 195, wherein the trispecific fusion protein induces a cytotoxic immune response against the cancer in the subject, thereby treating the cancer in the subject.

[0546] Embodiment 197. The method of any one of embodiments 194 to 196, wherein the subject is a rodent, a non-human primate, or a human.

[0547] Embodiment 198. A trispecific fusion protein according to any one of embodiments 96 to 150 or a tetravalent trispecific fusion protein according to any one of embodiments 151 to 177 for use in the treatment of cancer.

[0548] Embodiment 199. Use of a trispecific fusion protein according to any one of embodiments 96 to 150 or a tetravalent trispecific fusion protein according to any one of embodiments 151 to 177 in the manufacture of a medicament for the treatment of cancer.

[0549] Embodiment 200. A pharmaceutical composition comprising a trispecific fusion protein, wherein the trispecific fusion protein: (i) a first binding domain capable of binding to an antigen on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of the first tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a second tumor cell; and (iv) Skeleton Including, wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold. The pharmaceutical composition.

[0550] Embodiment 201. A method of treating a disease in a subject in need thereof, said method comprising administering to said subject a trispecific fusion protein, wherein said trispecific fusion protein: (i) a first binding domain capable of binding to an antigen on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of the first tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a second tumor cell; and (iv) Skeleton Including, wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold. The method.

[0551] Embodiment 202. A method of killing cancer cells in a subject in need thereof, said method comprising administering to said subject a trispecific fusion protein, said trispecific fusion protein comprising: (i) a first binding domain capable of binding to an antigen on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of the first tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a second tumor cell; and (iv) Skeleton Including, wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold. The method.

[0552] Embodiment 203.a) an anti-CD3 binding domain comprising a VH region and a VL region that specifically binds to the CD3 (cluster of differentiation 3) antigen on the surface of a T cell; b) one or two anti-TAA binding domains, comprising a VH region and a VL region that specifically bind to a TAA (tumor-associated antigen) on the surface of a tumor cell; c) a PD-1 polypeptide capable of binding to a PD-L1 receptor on the surface of a tumor cell. 1. A trispecific fusion protein...

Claims

1. (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a TAA on the surface of a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) skeleton 1. A trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold; the third binding domain is linked to (i) the first binding domain, (ii) the second binding domain, or (iii) the scaffold; and the trispecific fusion protein is not v31929; The trispecific fusion protein.

2. (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a TAA on the surface of a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) skeleton 1. A trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold; the third binding domain is linked to (i) the first binding domain, (ii) the second binding domain, or (iii) the scaffold; and the TAA is not HER2; The trispecific fusion protein.

3. (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) skeleton 1. A trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold; and the third binding domain is linked to the second binding domain; The trispecific fusion protein.

4. (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) skeleton 1. A trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold; and the third binding domain is linked to the scaffold; The trispecific fusion protein.

5. (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) skeleton 1. A trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold; and the third binding domain is linked to (i) the second binding domain or (ii) the scaffold; The trispecific fusion protein.

6. a) the first binding domain is a Fab and the second binding domain is an scFv, or b) the first binding domain is an scFv and the second binding domain is a Fab, or c) the first binding domain is a Fab and the second binding domain is a Fab, or d) the first binding domain is an scFv and the second binding domain is an scFv; The trispecific fusion protein according to any one of claims 1 to 5.

7. (i) a first binding domain that is a Fab domain, capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a tumor cell; (iii) a third binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) a scaffold comprising a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide.

1. A trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, and the third binding domain are operably linked to the scaffold; and the third binding domain is linked to (i) the N-terminus or C-terminus of the Fab light chain, (ii) the second binding domain, or (iii) the scaffold; The trispecific fusion protein.

8. 7. The trispecific fusion protein of claim 6, wherein the scaffold comprises or consists of a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide.

9. the first binding domain is a Fab domain and is linked to the N-terminus of the first Fc polypeptide via the C-terminus of the Fab heavy chain, and the second binding domain is an scFv domain and is linked to the N-terminus of the second Fc polypeptide via its C-terminus. The trispecific fusion protein of claim 6.

10. The method of claim 1, wherein the first binding domain is a Fab domain and is linked to the N-terminus of the first Fc polypeptide via the C-terminus of a Fab heavy chain; and the second binding domain is an scFv domain and is linked via its C-terminus to the N-terminus of the second Fc polypeptide; The trispecific fusion protein of claim 7.

11. 8. The trispecific fusion protein of any one of claims 1 to 5 and 7, wherein the third binding domain comprises a PD-1 polypeptide.

12. 12. The trispecific fusion protein of claim 11, wherein the PD-1 polypeptide comprises or consists of an amino acid sequence having at least about 80%, 90%, 95%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO:9 or SEQ ID NO:

10.

13. the anti-CD3 binding domain is HCDR1 sequences selected from the group consisting of RSTMH (SEQ ID NO: 207), YYGMS (SEQ ID NO: 303), KYAMN (SEQ ID NO: 224) and TYAMN (SEQ ID NO: 232), YINPSSAYTNYNQKFKD (SEQ ID NO: 208), SITSSGGRIYYPDSVKG (SEQ ID NO: 301), SITRSGGRIYYPDSVKG (SEQ ID NO: 217), RIRSKYNNYATYYADSVKD ( a VH domain comprising an HCDR2 sequence selected from the group consisting of PQVHYDYNGPY (SEQ ID NO:209), DGRDGWVAY (SEQ ID NO:275), HGNFGNSYISYWAY (SEQ ID NO:226), and HGNFGNSYVSWFAY (SEQ ID NO:234); and LCDR1 sequences selected from the group consisting of SASSSVSYMN (SEQ ID NO: 211), KRNTGNIGSNYVN (SEQ ID NO: 287), TGNTGNIGSNYVN (SEQ ID NO: 220), GSSTGAVTSGNYPN (SEQ ID NO: 228) and GSSTGAVTTSNYAN (SEQ ID NO: 236), DSSKLAS (SEQ ID NO: 212), RNDKRPD (SEQ ID NO: 298), RDDKRPS (SEQ ID NO: 221), GTKFLAP ( a VL domain comprising an LCDR2 sequence selected from the group consisting of QQWSRNPPT (SEQ ID NO:214), QSYSSGFI (SEQ ID NO:295), VLWYSNRWV (SEQ ID NO:230), ATWDDSLDGWV (SEQ ID NO:200), and ALWYSNLWV (SEQ ID NO:238); 8. The trispecific fusion protein of any one of claims 1 to 5 and 7, comprising:

14. 8. The trispecific fusion protein of any one of claims 1 to 5 and 7, wherein the TAA is Cldn18.2, and the anti-Cldn18.2 VH sequence of the second binding domain comprises a HCDR1 sequence comprising SNPMI (SEQ ID NO: 310), a HCDR2 sequence comprising IIDTDGSTYYADWAKG (SEQ ID NO: 311), and a HCDR3 sequence comprising RLHGSSNGYYDDL (SEQ ID NO: 312), and the anti-Cldn18.2 VL sequence of the second binding domain comprises a LCDR1 sequence comprising QASQSIYSYLS (SEQ ID NO: 313), a LCDR2 sequence comprising KASTLAS (SEQ ID NO: 314), and a LCDR3 sequence comprising QQGYTVTNVDKNT (SEQ ID NO: 315).

15. (i) a first binding domain capable of binding to CD3 on the surface of a cytotoxic effector cell; (ii) a second binding domain and a third binding domain capable of binding to a tumor-associated antigen (TAA) on the surface of a tumor cell; (iii) a fourth binding domain capable of binding to PD-L1 on the surface of a tumor cell; and (iv) skeleton 1. A tetravalent trispecific fusion protein comprising: wherein the first binding domain, the second binding domain, the third binding domain, and the fourth binding domain are operably linked to the scaffold. The tetravalent trispecific fusion protein.

16. 16. The tetravalent trispecific fusion protein of claim 15, wherein the scaffold comprises or consists of a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide.

17. a) the first binding domain is a Fab and the second and third binding domains are both scFv domains, or b) the first binding domain is an scFv and the second and third binding domains are both Fab domains, or c) the first binding domain, the second binding domain, and the third binding domain are Fab domains, or d) the first binding domain, the second binding domain, and the third binding domain are scFv domains; 16. The tetravalent trispecific fusion protein of claim 15.

18. 18. The tetravalent trispecific fusion protein of any one of claims 15 to 17, wherein (i) the first binding domain is a Fab domain, (ii) the second binding domain is a first scFv domain, and (iii) the third binding domain is a second scFv domain.

19. 18. The tetravalent trispecific fusion protein of any one of claims 15 to 17, wherein the first binding domain is linked to the N-terminus of the first Fc polypeptide and the second binding domain is linked to the N-terminus of the second Fc polypeptide.

20. 18. The tetravalent trispecific fusion protein of any one of claims 15 to 17, wherein the third binding domain is linked to (i) the first binding domain, (ii) the second binding domain, or (iii) the scaffold.

21. The tetravalent trispecific fusion protein of any one of claims 15 to 17, wherein the second binding domain and the third binding domain are capable of binding to the same TAA.

22. 16. The trispecific fusion protein of any one of claims 1 to 5, 7, and 15, wherein the trispecific fusion protein binds to the TAA and PD-L1 on the same tumor cell or on different tumor cells.

23. A pharmaceutical composition comprising the trispecific fusion protein of any one of claims 1 to 5, 7, and 15, and a pharmaceutically acceptable carrier, excipient, diluent, or combination thereof.

24. A nucleic acid molecule or set of nucleic acid molecules encoding the trispecific fusion protein of any one of claims 1 to 5, 7 and 15.

25. 25. A vector or set of vectors comprising the nucleic acid molecule or set of nucleic acid molecules of claim 24.

26. A method for producing the trispecific fusion protein of any one of claims 1 to 5, 7 and 15, comprising: (a) obtaining a host cell culture comprising at least one host cell comprising one or more nucleic acid molecules encoding said trispecific fusion protein; (b) recovering the trispecific fusion protein from the host cell culture; The method comprising:

27. A pharmaceutical for treating cancer, comprising the trispecific fusion protein of any one of claims 1 to 5, 7, and 15.

28. 20. Use of the trispecific fusion protein of any one of claims 1 to 5, 7 and 15 in the manufacture of a medicament for the treatment of cancer.

Images