Precursor trispecific antibody constructs and methods of use thereof

Abstract

The present invention describes a precursor trispecific antibody construct comprising at least one tumor-associated antigen binding domain, a T cell binding domain, and a regulatory domain with enhanced half-life. Further, a method for producing the precursor trispecific antibody construct is disclosed. A pharmaceutical composition comprising the precursor construct and its use for treating tumors is also disclosed.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of U.S. Serial No. 62 / 844,303, filed May 7, 2019. The entire contents and disclosure of the prior application are incorporated herein by reference in their entirety.

[0003] Sequence Listing Statement

[0004] This application contains a sequence listing, which has been submitted electronically in ASCII format and is incorporated herein by reference in its entirety. The ASCII copy was created on March 25, 2020, is named P-586968-PC-25MAR2020.txt, and is 279Kb in size. Technical Field

[0005] Disclosed herein are precursor trispecific antibody constructs and methods of use thereof. Methods of use include for treating cancer, wherein the precursor constructs comprise a prodrug having tumor-restricted activation and multiple antigen binding sites. Background Art

[0006] Monoclonal antibodies (unconjugated or naked) that are currently approved by global drug regulatory agencies for clinical use in the oncology setting are known to function using one or a combination of the following mechanisms: 1) blocking cell growth signaling, 2) blocking blood supply to cancer cells, 3) directly mediating apoptosis, 4) triggering immune effector functions such as antibody-dependent cellular cytotoxicity (ADCC), antibody-dependent cellular phagocytosis (ADCP), and complement-dependent cytotoxicity (CDC), and 5) promoting adaptive immunity against tumors.

[0007] Monoclonal antibody therapy has demonstrated survival benefits in the clinic. However, compared with chemotherapy, the overall response rate for cancer patients is low and the survival benefit is minimal (a few months). Although the root cause of the lack of robust clinical anti-cancer activity is not fully understood, studies have shown that cancer cells often rapidly develop compensatory signaling pathways to evade cell death. In addition, cancer stem cells (CSCs), which are considered to be efficient cancer-initiating cells, are less active in cell proliferation, so they tend to better maintain the lack of growth signals.

[0008] To improve the anti-tumor activity of monoclonal antibodies, multispecific antibodies are being developed. Unlike classic monoclonal antibodies (standard first-line therapy for several tumor entities), these multispecific antibodies can combine tumor cells with the means to destroy them, thereby improving the efficiency of treatment. These multispecific antibodies provide new treatment options for cancer patients.

[0009] Another therapeutic approach to fighting cancer is to utilize T cells. T cells provide a defense against cancer throughout life by patrolling the body to find newly emerged cancer cells and eliminating them efficiently and rapidly. Therapeutic approaches utilizing T cells have proven successful in at least the treatment of metastatic melanoma, metastatic renal cancer, asymptomatic metastatic hormone-refractory prostate cancer, and advanced melanoma.

[0010] Another consideration for tumor cytotoxicity is the tumor microenvironment (TME). The TME includes novel targets that can help guide and improve the effects of antibody therapies by enhancing host anti-tumor immune responses. For example, T cells play an unexpectedly critical role in antibody therapies against tumor antigens, although their importance is often unobserved because studies are conducted in immunodeficient mice.

[0011] A drawback of antibody therapies for cancer treatment is the "off-target" binding of antibodies to non-cancer tumor-associated antigen-expressing cells, especially if this binding leads to cytotoxicity. Therefore, the "off-target" binding of multispecific and bispecific antibodies poses a potential challenge to controlling their "off-target" activity against normal tissues that also express the antigen (even at very low levels). These "off-target" effects severely limit multispecific and bispecific antibody therapies. Another shortcoming of many bispecific or multispecific antibodies is their short half-life.

[0012] There remains a need to provide multispecific trivalent antibodies that have cytotoxicity specific for tumor cells while reducing toxic side effects and maintaining the quality of the antibody's effectiveness. Reducing the nonspecific toxic side effects of multispecific antibodies while enhancing the effectiveness of these antibodies requires antibodies in precursor form that (1) engage targets associated with tumor cells, tumor-associated cells, or the tumor cell environment, and (2) activate cytotoxic cells, such as T cells, once localized in the tumor microenvironment. Furthermore, it is essential that such multispecific antibodies do not significantly reduce immunogenicity to tumors or tumor-associated targets. The precursor trispecific antibody constructs described herein meet this need by attaching an adjustable half-life enhancing component and a blocking component that inhibits the antibody from engaging toxic cells before binding to the tumor or tumor-associated target. Summary of the Invention

[0013] In one aspect, disclosed herein are precursor trispecific antibody constructs comprising: a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; a first regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε.

[0014] In one aspect, disclosed herein is a precursor trispecific antibody construct comprising: a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; and a regulatory domain comprising either a first sub-regulatory domain and a second sub-regulatory domain, wherein the first sub-regulatory domain comprises a first protease cleavage domain and a half-life extension (HLP) domain, and the second sub-regulatory domain comprises a second protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε; or a single regulatory domain comprising a protease cleavage domain, a half-life extension (HLP) domain, and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of CD3ε.

[0015] In one aspect, disclosed herein are precursor trispecific antibody constructs comprising: a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; and a regulatory domain comprising a protease cleavage domain, a half-life extension (HLP) domain, and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε.

[0016] In a related aspect, the first binding domain and the second binding domain bind to the same TAA. In another related aspect, the first binding domain and the second binding domain bind to different TAAs.

[0017] In related aspects, the first TAA or the second TAA or both the first TAA and the second TAA can be an extracellular epitope of a tumor cell surface antigen, a tumor microenvironment antigen, a stromal antigen in the tumor microenvironment (TME), an angiogenic antigen in the TME, an antigen on a blood vessel in the TME, or a cytokine antigen in the TME, or any combination thereof. In another related aspect, the TAA can be one of: EGFR, FcγRI, FcγRIIa FcγRIIb, FcγRIIIb, CD28, CD137, CTLA-4, FAS, fibroblast growth factor receptor 1 (FGFR1), FGFR2, FGFR3, FGFR4, glucocorticoid-induced TNFR-related (GITR) protein, lymphotoxin-β receptor (LTβR), toll-like receptor (TLR), tumor necrosis factor-related apoptosis-inducing ligand-receptor 1 (TRAIL receptor 1), TRAIL receptor 2, prostate-specific membrane antigen (PSMA) protein, prostate stem cell antigen (PSCA) protein, tumor-associated protein Carbonic anhydrase IX (CAIX), epidermal growth factor receptor 1 (EGFR1), EGFRvIII, human epidermal growth factor receptor 2 (Her2 / neu; Erb2), ErbB3 (HER3), folate receptor, ephrin receptor, PDGFRa, ErbB-2, CD20, CD22, CD30, CD33, CD40, CD37, CD38, CD70, CD74, CD40), CD80, CD86, CD2, p53, cMet (tyrosine protein kinase Met, hepatocyte growth factor receptor (HGFR)), MAGE -A1, MAGE-A2, MAGE-A3, MAGE-A4, MAGE-A6, MAGE-A10, MAGE-A12, BAGE, DAM-6, DAM-10, GAGE-1, GAGE-2, GAGE-8, GAGE-3, GAGE-4, GAGE-5, GAGE-6, GAGE-7B, NA88-A, NY-ESO-1, BRCA1, BRCA2, MART-1, MC1R, Gp100, PSA, PSM, tyrosinase, Wilms' tumor antigen (WT1), TRP-1, T RP-2, ART-4, CAMEL, Cyp-B, hTERT, hTRT, iCE, MUC1, MUC2, P-cadherin, myostatin (GDF8), Cripto (TDGF1), MUC5AC, PRAME, P15, RU1, RU2, SART-1, SART-3, WT1, AFP, β-catenin / m, caspase-8 / m, CDK-4 / m, ELF2M, GnT-V, G250, HSP70-2M, HST-2, KIAA0205, MUM-1, MUM-2,In a further related aspect, the TAA bound by the first binding domain or the second binding domain or both is selected from the group consisting of EGFR, ROR1, PSMA, and 5T4. In another related aspect, when the TAA antigen is EGFR, the amino acid sequence of the first binding domain or the second binding domain, or both, comprises the amino acid sequence recited in any one of SEQ ID NOs: 34, 37, or a combination thereof. In another related aspect, when the TAA antigen is ROR1, the amino acid sequence of the first binding domain or the second binding domain, or both, comprises the amino acid sequence recited in any one of SEQ ID NOs: 156 and 166, or a combination thereof. In another related aspect, when the TAA antigen is PSMA, the amino acid sequence of the first binding domain or the second binding domain, or both, comprises the amino acid sequence recited in any one of SEQ ID NOs: 168 and 170, or a combination thereof. In another related aspect, when the TAA antigen is 5T4, the amino acid sequence of the first binding domain or the second binding domain, or both, comprises the amino acid sequence recited in any one of SEQ ID NOs: 172 and 174, or a combination thereof.

[0018] In another related aspect, the tumor microenvironment antigen can be a KIR, LILR or TIGIT antigen. In another related aspect, the matrix antigen in the tumor microenvironment can be fibroblast activation protein (FAP), alpha smooth muscle actin (αSMA), PDGFRα, integrin α11β1 (ITGA11) VEGF, tenascin-C, periostin, fibroblast-specific protein 1 (S10A4, FSP1), desmin, vimentin, paladin, urokinase-type plasminogen activator receptor-associated protein (UPARAP), galectin-3, podoplanin, platelets, CCL2 or CXCL12. In another related aspect, the angiogenesis antigen in the tumor microenvironment can be bFGF, INF or VEGF. In a further related aspect, the antigen on the surface of the blood vessels in the tumor microenvironment comprises an endothelial cell surface antigen selected from CD31, CD105, CD146 and CD144. In yet another related aspect, the cytokine antigen can be TNF-α, IL-6, TGF-β, IL-10, IL-8, IL-17, IL-21, INF, or VEG.

[0019] In a related aspect, the HLP domain comprises a human serum albumin (HSA) polypeptide.

[0020] In a related aspect, the CAP component of the second sub-regulatory domain comprises the amino acid sequence of an extracellular epitope of human CD3 epsilon. In another related aspect, the amino acid sequence of the CAP component is set forth in SEQ ID NO: 5 or a homolog thereof.

[0021] In a related aspect, the first binding domain, the second binding domain, or both each comprise a single chain variable fragment (scFv).In another related aspect, the third binding domain comprises a Fab antigen binding fragment.

[0022] In a related aspect, the protease cleavage domains in the first and second sub-regulatory domains are cleaved by the same protease. In another related aspect, the protease cleavage domains in the first and second sub-regulatory domains are cleaved by different proteases. In another related aspect, the first and / or second protease cleavage domains comprise a protease cleavable amino acid sequence that is cleavable by a serine protease, a cysteine ​​protease, an aspartic protease, or a matrix metalloproteinase (MMP), or is a composite substrate cleaved by one or more of MMP2 / 9, uPA, matrices, and asparagine endopeptidases, or any combination thereof. In another related aspect, the MMP can be matrix metalloproteinase 1 (MMP-1), matrix metalloproteinase 2 (MMP-2), matrix metalloproteinase 9 (MMP-9), or matrix metalloproteinase 14 (MMP-14). In another related aspect, the serine protease can be urokinase-type plasminogen activator (uPA) protease or membrane-type serine protease (MT-SP1). In another related aspect, the amino acid sequence of a composite substrate cleaved by one or more of MMP2 / 9, uPA, matrices, and asparagine endopeptidases is set forth in SEQ ID NO:35. In further related aspects, the first and / or second protease cleavage domain comprises the sequence recited in SEQ ID NO: 9 (PLGLAG), SEQ ID NO: 10 (GPLGMLSQ), SEQ ID NO: 11 (GPLGLWAQ), SEQ ID NO: 12 (GPLGLAG), SEQ ID NO: 13 (KKNPAELIGPVD), or SEQ ID NO: 14 (KKQPAANLVAPED), or SEQ ID NO: 35. In yet another related aspect, the first and / or second protease cleavage domain comprises the sequence recited in SEQ ID NO: 9. In yet another aspect, the first and / or second protease cleavage domain comprises the sequence recited in SEQ ID NO: 35.

[0023] In a related aspect, the third binding domain comprises a variable heavy chain (VH3-CH1) region and a variable light chain (VL3-CL) region; wherein the first binding domain is located at the C-terminus of the VL3-CL or the VH3-CH1 region of the third binding domain; wherein when the first binding domain is located at the C-terminus of the VL3-CL region, the second sub-regulatory domain is located at the C-terminus of the VH3-CH1 region, and when the first binding domain is located at the C-terminus of the VH3-CH1 region, the second sub-regulatory domain is located at the C-terminus of the VL3-CL region. In another related aspect, the third binding domain comprises a variable heavy chain (VH3) region and a variable light chain (VL3) region; wherein the first sub-regulatory domain comprising the HLP domain located at the N-terminus of the protease cleavage domain is located at the N-terminus of the VH3 region or the VL3 region of the third binding domain; wherein when the first sub-regulatory domain is located at the N-terminus of the VL3 region, the second sub-regulatory domain comprising the CAP component located at the N-terminus of the protease cleavage domain is located at the N-terminus of the VH3 region, and when the first sub-regulatory domain is located at the N-terminus of the VH3 region, the second sub-regulatory domain comprising the CAP component located at the N-terminus of the protease cleavage domain is located at the N-terminus of the VL3 region.

[0024] In another aspect, the precursor trispecific antibody construct comprises two polypeptides, polypeptide A and polypeptide B, wherein polypeptide A comprises a component having the following N-terminal to C-terminal order: HLP domain, protease cleavage domain, third binding domain (VH3 region), first binding domain (VL-VH); and polypeptide B comprises a component having the following N-terminal to C-terminal order: CAP component, protease cleavage domain, third binding domain (VL region), second binding domain (VL-VH); or polypeptide A comprises a component having the following N-terminal to C-terminal order: HLP domain, protease cleavage domain, third binding domain (VH3 region), first binding domain (VH-VL); and polypeptide B comprises a component having The components with the following N-terminal to C-terminal order: CAP component, protease cleavage domain, third binding domain (VL region), second binding domain (VH-VL); or polypeptide A comprises components with the following N-terminal to C-terminal order: HLP domain, protease cleavage domain, third binding domain (VL3 region), first binding domain (VL-VH); and polypeptide B comprises components with the following N-terminal to C-terminal order: CAP component, protease cleavage domain, third binding domain (VH3 region), second binding domain (VL-VH); or polypeptide A comprises components with the following N-terminal to C-terminal order: HLP domain, protease cleavage domain, third binding domain (VL3 region), first binding domain (VL-VH). Region), a first binding domain (VH-VL); and polypeptide B comprises a component having the following N- to C-terminal order: CAP component, protease cleavage domain, a third binding domain (VH region), a second binding domain (VH-VL); or polypeptide A comprises a component having the following N- to C-terminal order: HLP domain, protease cleavage domain, a third binding domain (VL3 region), a first binding domain (VL-VH); and polypeptide B comprises a component having the following N- to C-terminal order: CAP component, protease cleavage domain, a third binding domain (VH region), a second binding domain (VL-VH); or polypeptide A comprises a component having the following N- to C-terminal order: HLP domain, a protease cleavage domain, a third binding domain (VL3 region), a first binding domain (VH-VL); and polypeptide B comprises a module having the following N-terminal to C-terminal order: a CAP module, a protease cleavage domain, a third binding domain (VH3 region), a second binding domain (VH-VL); or polypeptide A comprises a module having the following N-terminal to C-terminal order: an HLP domain, a protease cleavage domain, a third binding domain (VH3 region), a first binding domain (VL-VH); and polypeptide B comprises a module having the following N-terminal to C-terminal order: a CAP module, a protease cleavage domain, a third binding domain (VL3 region), a second binding domain (VL-VH);Or polypeptide A comprises a module having the following N-terminal to C-terminal order: HLP domain, protease cleavage domain, third binding domain (VH3 region), first binding domain (VH-VL); and polypeptide B comprises a module having the following N-terminal to C-terminal order: CAP module, protease cleavage domain, third binding domain (VL3 region), second binding domain (VH-VL).

[0025] In a related aspect, the third binding domain comprises a VL3 region and a VH3 region, wherein the VL3 region comprises CDR-L1 (selected from SEQ ID NOs: 107-109), CDR-L2 (SEQ ID NO: 110), and CDR-L3 (selected from SEQ ID NOs: 111-112), and the VH3 region comprises CDR-H1 (SEQ ID NO: 104), CDR-H2 (SEQ ID NO: 105), and CDR-H3 (SEQ ID NO: 106). In another related aspect, the VL3 region comprises an amino acid sequence selected from any one of SEQ ID NOs: 75-103 and 116, or an amino acid sequence having at least 80% homology thereto. In another related aspect, the VH3 region comprises an amino acid sequence selected from any one of SEQ ID NOs: 46-72 and 114, or an amino acid sequence having at least 80% homology thereto.

[0026] In one aspect, disclosed herein is a pharmaceutical composition comprising a precursor trispecific antibody construct and a pharmaceutically acceptable carrier, wherein the precursor trispecific antibody construct comprises: a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second tumor-associated antigen (TAA); a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε.

[0027] In one aspect, disclosed herein is a nucleic acid construct comprising a nucleic acid sequence, or multiple nucleic acid sequences, encoding a precursor trispecific antibody construct comprising: a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second tumor-associated antigen (TAA); a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε.

[0028] In one aspect, disclosed herein are expression vectors comprising one or more nucleic acid constructs encoding one or more polypeptides of the precursor trispecific antibody constructs disclosed herein.

[0029] In one aspect, disclosed herein is an isolated host cell comprising a nucleic acid sequence or multiple nucleic acid sequences encoding a precursor trispecific antibody construct, the precursor trispecific antibody construct comprising: a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second tumor-associated antigen (TAA); a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε.

[0030] In one aspect, disclosed herein is a method of treating, preventing, inhibiting growth, delaying disease progression, reducing tumor burden, or reducing the incidence of a cancer or tumor, or any combination thereof, in a subject in need of such treatment, comprising the step of administering to the subject a pharmaceutical composition comprising a precursor trispecific antibody construct comprising: a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second tumor-associated antigen (TAA); a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε; wherein the method treats, prevents, inhibits growth, delays disease progression, reduces tumor burden, or reduces the incidence of a cancer or tumor in the subject.

[0031] In related aspects, the methods disclosed herein for treating a subject in need thereof reduce minimal residual disease, increase remission, increase duration of remission, reduce tumor recurrence rate, prevent metastasis of a tumor or cancer, or reduce the rate of metastasis of a tumor or cancer, or any combination thereof, compared to a subject not administered a pharmaceutical composition disclosed herein. In another related aspect, the cancer or tumor comprises a solid tumor or a non-solid tumor, or wherein the cancer or tumor comprises metastasis of a cancer or tumor. In further related aspects, the non-solid cancer or tumor can be a hematopoietic malignancy, a blood cell cancer, a leukemia, a myelodysplastic syndrome, a lymphoma, multiple myeloma (plasma cell myeloma), acute lymphoblastic leukemia, acute myeloid leukemia, chronic myeloid leukemia, Hodgkin lymphoma, non-Hodgkin lymphoma, or plasma cell leukemia; or wherein the solid tumor can be a sarcoma or carcinoma, a fibrosarcoma, a myxosarcoma, a liposarcoma, a chondrosarcoma, an osteosarcoma, a chordoma, an angiosarcoma, an endosarcoma, a sarcoma, a fibrosarcoma, a myxosarcoma, a chondrosarcoma, an osteosarcoma, a chordoma, an angiosarcoma, an endosarcoma, a fibrosarcoma ... cutaneous sarcoma, lymphangiosarcoma, lymphangioendotheliosarcoma, synovioma, mesothelioma, Ewing sarcoma, leiomyosarcoma, rhabdomyosarcoma, colon cancer, pancreatic cancer or tumor, breast cancer or tumor, ovarian cancer or tumor, prostate cancer or tumor, squamous cell carcinoma, basal cell carcinoma, adenocarcinoma, sweat gland carcinoma, sebaceous gland carcinoma, papillary carcinoma, papillary adenocarcinoma, cystadenocarcinoma, medullary carcinoma, bronchogenic carcinoma, renal cell carcinoma, liver cancer, bile duct cancer, choriocarcinoma, seminoma, embryonal carcinoma, Wilm's tumor, cervical cancer or tumor, uterine cancer or tumor, testicular cancer or tumor, lung cancer, small cell lung cancer, bladder cancer, epithelial carcinoma, glioma, astrocytoma, medulloblastoma, craniopharyngioma, ependymoma, pinealoma, hemangioblastoma, acoustic neuroma, oligodendroglioma, schwannoma, meningioma, melanoma, neuroblastoma, or retinoblastoma.

[0032] In one aspect, disclosed herein is a method for treating, preventing, inhibiting growth, delaying disease progression, reducing tumor burden, or reducing the incidence of a cancer or tumor, or any combination thereof, in a subject in need of such treatment, comprising the step of administering to the subject a pharmaceutical composition comprising a nucleic acid construct or multiple nucleic acid constructs comprising a nucleic acid sequence or multiple nucleic acid sequences encoding a precursor trispecific antibody construct, the precursor trispecific antibody construct comprising: a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second tumor-associated antigen (TAA); a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε; wherein the method treats, prevents, inhibits growth, delays disease progression, reduces tumor burden, or reduces the incidence of a cancer or tumor in the subject.

[0033] In related aspects, use of the methods disclosed herein for treating a subject in need thereof reduces minimal residual disease, increases remission, increases duration of remission, reduces the rate of tumor recurrence, prevents metastasis of a tumor or cancer, or reduces the rate of metastasis of a tumor or cancer, or any combination thereof, compared to a subject not administered the pharmaceutical composition.

[0034] In one aspect, disclosed herein is a method for producing a precursor trispecific antibody construct, which comprises: a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second tumor-associated antigen (TAA); a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε; the method comprises the following steps: culturing a host cell comprising a nucleic acid sequence encoding precursor trispecific antibody construct polypeptides A and B, expressing the polypeptides A and B, isolating the expressed precursor trispecific antibody construct polypeptides A and B, and dimerizing the polypeptides A and B.

[0035] In a related aspect, in the method of producing a precursor trispecific antibody construct, expression comprises expression from the same host cell or comprises two host cells each expressing a different polypeptide (polypeptide A and polypeptide B). BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The subject matter regarded as the precursor trispecific (tribody) antibody constructs disclosed herein is particularly pointed out and distinctly claimed in the concluding portion of the specification. However, the precursor trispecific (tribody) antibody constructs, both as to organization and method of use, together with objects, features, and advantages thereof, may be best understood by reference to the following detailed description when read in conjunction with the accompanying drawings, in which:

[0037] Figure 1 Schematic embodiments of precursor trispecific (trimeric) antibody constructs having modular components are shown, including but not limited to: (1) modular regulatory domains, which in certain embodiments include modular functional components and modular protease cleavage peptides, and (2) modular binding domains, which in certain embodiments include modular anti-tumor associated antigen binding domains. Figure 1The embodiment of the precursor trispecific (tribody) antibody construct has three antibody binding domains, wherein the Fab portion recognizes the CD3 surface antigen, wherein the components and regions of the different domains are identified. In the embodiment shown herein, the precursor trispecific (tribody) antibody construct is formed by two polypeptides, wherein each polypeptide includes an anti-tumor associated antigen (TAA) binding domain (the first and second binding domains), which is the C-terminus (the third binding domain) of the anti-CD3 Fab binding domain on each polypeptide. Further, each polypeptide includes a regulatory domain (the first and second sub-regulatory domains) at the N-terminus of the anti-CD3 Fab binding domain, wherein the first sub-regulatory domain includes a protease cleavage domain and a half-life extension domain (in this embodiment, human serum albumin), and the second sub-regulatory domain includes a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to the human CD3 surface antigen. As shown herein, each anti-TAA binding domain is a single-chain variable fragment (ScFv). As shown herein, the N-terminal to C-terminal order of one of the ScFvs is the variable light chain region (VL2), followed by a linker (L4), followed by a variable heavy chain region (VH2), followed by a linker (L5), while the N-terminal to C-terminal order of the other ScFv is the variable heavy chain region (VH3), followed by a linker (L9), followed by a variable light chain region (VL3), followed by a linker (L10). In embodiments not shown, the N-terminal to C-terminal order of the scFvs can be reversed, and the binding domains can be the same or different for the two scFvs. In embodiments not shown, the two sub-regulatory domains can be linked from their current bonds to another polypeptide chain, for example, the HSA regulatory domain can be linked to the N-terminus of the variable light chain (VL1) of the Fab fragment and the CAP regulatory domain can be linked to the N-terminus of the variable heavy chain (VH1) of the Fab fragment. Linkers between modules and domains are identified by "L" followed by a number, e.g., L1, L2, L3, L4, L5, L6, L7, L8, L9, L10. Linkers may or may not be present. VL1 is the variable light chain region that binds to site three, and VH1 is the variable heavy chain region that binds to site three. The oval, designated HSA, is the human serum albumin module. The shape designated CP is the cleavage peptide. The triangle is the CAP module.

[0038] Figures 2A to 2F Various embodiments of precursor and active trispecific (trimeric) antibody constructs described herein are shown. Figure 2A and 2B A schematic embodiment of a precursor trispecific antibody construct comprising two anti-tumor associated antigen binding domains is shown, wherein the tumor associated antigen is EGFR, wherein the N-terminal to C-terminal order of one of the variable regions is VL2-L4-VH2 and the other is VH3-L9-VL3 ( Figure 2A) or one scFv is VH2-L4-VL2 and the other is VL3-L9-VH3 ( Figure 2B ). The precursor trispecific antibody construct further includes an anti-CD3ε Fab domain at the N-terminus of the scFv binding domain and two sub-regulatory domains comprising a protease-cleavable linker, and a human serum albumin (HSA) polypeptide sequence or CD3 CAP, wherein in some embodiments, the amino acids of the CAP are amino acids 1-27 of the mature CD3ε polypeptide (SEQ ID NO: 4) at the N-terminus of the Fab domain. The order of components in the regulatory domain is N-terminus to C-terminus: CAP or HSA-linker-protease-cleavable linker. L1, L2, etc. represent possible linkers between different domains or domain components. Linkers may or may not be present. Figure 2C Shows Figure 2B Schematic representation of the precursor trispecific antibody construct shown, but lacking the regulatory domain containing the half-life extending assembly (HSA). Figure 2D Shows Figure 2B Schematic representation of the precursor trispecific antibody construct shown, but lacking the subregulatory domain comprising the CD3 CAP domain. Figure 2E Schematic diagram showing active trispecific (trimeric) antibody constructs, wherein Figure 2B The precursor construct of is in active form and lacks the two sub-regulatory domains. Figure 2F A schematic diagram of a precursor trispecific antibody construct is shown, wherein the regulatory domain comprises a single regulatory domain comprising a CAP domain, an HSA sequence, and a protease-cleavable linker on the same polypeptide.

[0039] Figure 3A and 3B Flowchart showing protease-specific activation within tumor tissue or within the tumor environment of a precursor trispecific antibody construct, where T cell engagement and activation is restricted to the tumor site. Figure 3AThe effect of the cancer (tumor) microenvironment on a precursor trispecific antibody construct is shown. The precursor trispecific antibody construct comprises two protease-cleavable domains, one of which is the C-terminus of an HSA half-life-extending polypeptide and the other is a CAP component that can be specifically bound by a third (anti-CD3 Fab) binding domain. Entry into the cancer microenvironment, which is known to be rich in proteases secreted by cancer cells, results in protease cleavage and removal of HSA and the CAP regulatory component, wherein in some embodiments, the CAP comprises an extracellular CD3ε epitope. In some embodiments, the protease-cleavable domains can be cleaved by the same or different proteases. The resulting activated antibody (activated trispecific antibody construct) can now bind to and activate T cells. If the precursor construct is bound to a TAA and is not in the tumor microenvironment, protease cleavage will not occur and T cell activation will not occur. The design of the precursor trispecific antibody construct provides improved protease-activated controlled release of the CAP and HSA regulatory domains with trispecific binding epitopes. Figure 3B The protease-specific cleavage of HSA and CD3 CAP in the cancer microenvironment is shown, followed by T cell activation and binding of the activated tribody construct to T cells and tumor cells. The precursor trispecific antibody construct remains intact in the circulation or when present in normal tissues, with an extended half-life in vivo. After binding to the TAA target antigen, in this case EGFR, where the target antigen is present on the tumor surface (in the tumor microenvironment), protease-specific activation may occur, resulting in cleavage of the two sub-regulatory domains and exposure of the anti-CD3 binding site. The activated trispecific antibody construct antibody has a reduced limited half-life of several hours compared to several days to several weeks for the precursor trispecific antibody construct. (Data not shown). In addition, in some embodiments, the cleaved and activated trispecific antibody comprises a smaller size than the precursor construct, which can improve the activated trispecific antibody tumor penetration.

[0040] Figure 4A and 4B Shown are the amino acid sequences of the heavy chain (HC) polypeptides of an activated trispecific (tribody) antibody construct (Construct 1; VLVH) and embodiments of optimized nucleotide sequences encoding the activated construct heavy chain (HC). The amino acid sequence is shown from N-terminus to C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 4A One embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of the activated construct is shown, having an N-terminal to C-terminal order and the following components: h1F3.5-G1Fd anti-EGFR VL-linker-VH (SEQ ID NO: 138). Figure 4AThe amino acid sequence of the component parts of the HC polypeptide shown includes: a linker (SEQ ID NO: 158), an anti-CD3ε variable heavy chain and a constant heavy chain region 1 (SEQ ID NO: 113), followed by two labeled cysteine ​​residues (labeled in bold and underlined), which can participate in disulfide double bonds, followed by an anti-EGFR scFv VL (SEQ ID NO: 34)-linker (SEQ ID NO: 39)-VH (SEQ ID NO: 37) chain. Figure 4B One embodiment of an optimized nucleic acid sequence (DNA) encoding the heavy chain (HC) polypeptide of an activated trispecific (tribody) construct is shown, having a 5' to 3' order and the following components (SEQ ID NO: 150). Figure 4A The nucleic acid sequences of the component parts of the HC polypeptide shown include: linkers (e.g. Figure 2E L7 of (SEQ ID NO: 154), anti-CD3ε variable heavy chain and constant heavy chain region 1 (SEQ ID NO: NO: 155), followed by two labeled cysteine ​​residues (marked in bold and underlined), which can participate in disulfide double bonds, followed by anti-EGFR scFv VL (SEQ ID NO: 36)-linker (SEQ ID NO: 40)-VH (SEQ ID NO: 38) chain.

[0041] Figure 5A and 5B Shown are embodiments of the amino acid sequence of the light chain (LC) polypeptide of an activated trispecific (tribody) antibody construct (Construct 1; VLVH) and an optimized nucleotide sequence encoding the light chain (LC) of the activated construct. The amino acid sequence is shown from N-terminus to C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 5A One embodiment of the amino acid sequence of the light chain (LC) polypeptide of the activated construct is shown, having the following N-terminal to C-terminal order and components: h1F3.1-λLCanti-EGFRVL-Linker-VH (SEQ ID NO: 139). Figure 5A The amino acid sequences of the component parts of the LC polypeptide shown include: linkers (e.g. Figure 2E L2) (SEQ ID NO: 158), anti-CD3ε variable light chain and lambda light chain (SEQ ID NO: 74), followed by labeled cysteine ​​residues (marked in bold and underlined), which can participate in disulfide bonds, followed by anti-EGFR scFv VL (SEQ ID NO: 34)-linker (SEQ ID NO: 40)-VH (SEQ ID NO: 37) chains. Figure 5BOne embodiment of an optimized nucleic acid sequence (DNA) encoding the light chain (LC) polypeptide of an activated trispecific (tribody) construct is shown, having a 5' to 3' sequence and the following components (SEQ ID NO: 151). Figure 5A The nucleic acid sequences of the component parts of the LC polypeptide include: linkers (e.g. Figure 2E L2) (SEQ ID NO: 154), anti-CD3ε variable light chain and lambda light chain region (SEQ ID NO: 159), followed by labeled cysteine ​​residues (marked in bold and underlined), which can participate in disulfide double bonds, followed by anti-EGFR scFv VL (SEQ ID NO: 34)-linker (SEQ ID NO: 40)-VH (SEQ ID NO: 37) chains.

[0042] Figure 6A and 6B Shown are embodiments of the amino acid sequence of the heavy chain (HC) polypeptide of an activated trispecific (tribody) antibody construct (Construct 2; VHVL) and the optimized nucleotide sequence encoding the heavy chain (HC) of the activated construct. The amino acid sequence is shown from N-terminus to C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 6A One embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of the activated construct is shown, having an N-terminal to C-terminal order and the following components: h1F3.5-G1Fd-(VH-Linker-VL) (SEQ ID NO: 140). Figure 6A The amino acid sequence of the component parts of the HC polypeptide shown includes: a linker (SEQ ID NO: 158), an anti-CD3ε variable heavy chain and a constant heavy chain region 1 (SEQ ID NO: 113), followed by two labeled cysteine ​​residues (labeled in bold and underlined), which can participate in disulfide double bonds, followed by an anti-EGFR scFv VH (SEQ ID NO: 37)-linker (SEQ ID NO: 40)-VL (SEQ ID NO: 34) chain. Figure 6B One embodiment of an optimized nucleic acid sequence (DNA) encoding the heavy chain (HC) polypeptide of an activated trispecific (tribody) construct is shown, having a 5' to 3' sequence and the following components (SEQ ID NO: 152). Figure 6A The nucleic acid sequence of the component parts of the HC polypeptide shown includes: a linker (SEQ ID NO: 154), an anti-CD3ε variable heavy chain and a constant heavy chain region 1 (SEQ ID NO: 155), followed by two labeled cysteine ​​residues (marked in bold and underlined), which can participate in disulfide double bonds, followed by an anti-EGFR scFv VH (SEQ ID NO: 38)-linker (SEQ ID NO: 40)-VH (SEQ ID NO: 36) chain.

[0043] Figure 7A and 7B Shown are embodiments of the amino acid sequence of the light chain (LC) polypeptide of an activated trispecific (tribody) antibody construct (Construct 2; VHVL) and an optimized nucleotide sequence encoding the light chain (LC) of the activated construct. The amino acid sequence is shown from N-terminus to C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 7A One embodiment of the amino acid sequence of the light chain (LC) polypeptide of the activated construct is shown, having the following N-terminal to C-terminal order and components: h1F3.1-λLC-anti-EGFR (VH-Linker-VL) (SEQ ID NO: 141). Figure 7A The amino acid sequence of the component parts of the LC polypeptide shown includes: a linker (SEQ ID NO: 158), an anti-CD3ε variable light chain and a lambda light chain (SEQ ID NO: 74), followed by a labeled cysteine ​​residue (marked in bold and underlined), which can participate in a disulfide double bond, followed by an anti-EGFR scFv VH (SEQ ID NO: 37)-linker (SEQ ID NO: 40)-VL (SEQ ID NO: 34) chain. Figure 7B One embodiment of an optimized nucleic acid sequence (DNA) encoding the light chain (LC) polypeptide of an activated trispecific (tribody) construct is shown, having a 5' to 3' sequence and the following components (SEQ ID NO: 153). Figure 7A The nucleic acid sequence of the component parts of the LC polypeptide shown includes: a linker (SEQ ID NO: 154), an anti-CD3ε variable light chain and a lambda light chain region (SEQ ID NO: 159), followed by a labeled cysteine ​​residue (marked in bold and underlined), which can participate in a disulfide double bond, followed by an anti-EGFR scFv VH (SEQ ID NO: 38)-linker (SEQ ID NO: 40)-VL (SEQ ID NO: 36) chain.

[0044] Figure 8A and 8B An embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of a precursor trispecific (tribody) antibody construct (Construct 3; VLVH) and an optimized nucleotide sequence encoding the heavy chain (HC) of the precursor construct is shown. The amino acid sequence is shown from N-terminus to C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 8A One embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of the precursor construct is shown, having an N-terminal to C-terminal order and the following components: hALB-G-PLGLAG(MMP2 / 9)-(clone)-h1F3.5-G1Fd anti-EGFR VL-linker-VH (SEQ ID NO: 130). Figure 8A The amino acid sequence of the component parts of the HC polypeptide shown includes: human serum albumin (HSA) (SEQ ID NO: 7), an MMP2 / 9 protease cleavable linker (SEQ ID NO: 160 (linker with cleavable sequence) and SEQ ID NO: 9 (cleavable sequence)), an anti-CD3ε variable heavy chain and a constant heavy chain region 1 (SEQ ID NO: 113), followed by two labeled cysteine ​​residues (labeled in bold and underlined) that can participate in disulfide double bonds, followed by an anti-EGFR scFv VL (SEQ ID NO: 34)-linker (SEQ ID NO: 40)-VH (SEQ ID NO: 37) chain. Figure 8B One embodiment of an optimized nucleic acid sequence (DNA) encoding the heavy chain (HC) polypeptide of a precursor trispecific (tribody) antibody construct is shown, having a 5' to 3' sequence and the following components (SEQ ID NO: 142). Figure 8A The nucleic acid sequence of the component parts of the HC polypeptide shown includes: human serum albumin (HSA) (SEQ ID NO: 8), an MMP2 / 9 protease cleavable linker (SEQ ID NO: 161 (linker with a cleavable sequence and SEQ ID NO: 33 (cleavable sequence)), an anti-CD3ε variable heavy chain and a constant heavy chain region 1 (SEQ ID NO: 155), followed by two labeled cysteine ​​residues (marked in bold and underlined) that can participate in disulfide double bonds, followed by an anti-EGFR scFv VL (SEQ ID NO: 36)-linker (SEQ ID NO: 40)-VH (SEQ ID NO: 38) chain.

[0045] Figure 9A and 9B Shown are embodiments of the amino acid sequence of the light chain (LC) polypeptide of a precursor trispecific (tribody) antibody construct (Construct 3; VLVH) and the optimized nucleotide sequence encoding the light chain (LC) of the precursor construct. The amino acid sequence is shown from N-terminus to C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 9A One embodiment of the amino acid sequence of the light chain (LC) polypeptide of the precursor construct is shown, having an N-terminal to C-terminal order and the following components: Cap-hlF3.1-λLC anti-EGFR VL-linker-VH MM2 / 9 cleavage; (SEQ ID NO: 131). Figure 9AThe amino acid sequence of the component parts of the LC polypeptide shown includes: CAP (SEQ ID NO: 5), MMP2 / 9 protease cleavable linker (SEQ ID NO: 160 and SEQ ID NO: 9 (cleavable sequence)), anti-CD3 epsilon variable light chain and lambda light chain (SEQ ID NO: 74), followed by the labeled cysteine ​​residues (marked in bold and underlined), which can participate in disulfide double bonds, followed by the anti-EGFR scFv VL (SEQ ID NO: 34)-linker (SEQ ID NO: 40)-VH (SEQ ID NO: 37) chain. Figure 9B One embodiment of an optimized nucleic acid sequence (DNA) encoding the light chain (LC) polypeptide of a precursor trispecific (tribody) antibody construct is shown, having a 5' to 3' sequence and the following components (SEQ ID NO: 143). Figure 9A The nucleic acid sequence of the component parts of the LC polypeptide shown includes: CAP (SEQ ID NO: 164), MMP2 / 9 protease cleavable linker (SEQ ID NO: 161 and SEQ ID NO: 33 (cleavable sequence)), anti-CD3 epsilon variable light chain and lambda light chain region (SEQ ID NO: 159), followed by labeled cysteine ​​residues (marked in bold and underlined), which can participate in disulfide double bonds, followed by anti-EGFR scFv VL (SEQ ID NO: 36)-linker (SEQ ID NO: 40)-VH (SEQ ID NO: 38) chains.

[0046] Figure 10A and 10B An embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of a precursor trispecific (tribody) antibody construct (Construct 4; VHVL) and the optimized nucleotide sequence encoding the heavy chain (HC) of the precursor construct is shown. The amino acid sequence is shown from N-terminus to C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 10A One embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of the precursor construct is shown, having an N-terminal to C-terminal order and the following components: hALB-G-PLGLAG(MMP2 / 9)-(clone)-h1F3.5-G1Fd anti-EGFR (VH-linker-VL) (SEQ ID NO: 132). Figure 10AThe amino acid sequence of the component parts of the HC polypeptide shown includes: human serum albumin (HSA) (SEQ ID NO:7), an MMP2 / 9 protease cleavable linker (SEQ ID NO:160 and SEQ ID NO:9 (cleavable portion)), an anti-CD3ε variable heavy chain and a constant heavy chain region 1 (SEQ ID NO:113), followed by two labeled cysteine ​​residues (labeled in bold and underlined) that can participate in disulfide double bonds, followed by an anti-EGFR scFv VH (SEQ ID NO:37)-linker (SEQ ID NO:40)-VL (SEQ ID NO:34) chain. Figure 10B One embodiment of an optimized nucleic acid sequence (DNA) encoding the heavy chain (HC) polypeptide of a precursor trispecific (tribody) antibody construct is shown, having a 5' to 3' sequence and the following components (SEQ ID NO: 144). Figure 10A The nucleic acid sequence of the component parts of the HC polypeptide shown includes: human serum albumin (HSA) (SEQ ID NO: 8), an MMP2 / 9 protease cleavable linker (SEQ ID NO: 161 and SEQ ID NO: 33 (cleavable linker)), an anti-CD3ε variable heavy chain and a constant heavy chain region 1 (SEQ ID NO: 155), followed by two labeled cysteine ​​residues (marked in bold and underlined) that can participate in disulfide double bonds, followed by an anti-EGFR scFv VH (SEQ ID NO: 38)-linker (SEQ ID NO: 40)-VL (SEQ ID NO: 36) chain.

[0047] Figure 11A and 11B An embodiment of the amino acid sequence of the light chain (LC) polypeptide of a precursor trispecific (tribody) antibody construct (Construct 4; VLVH) and an optimized nucleotide sequence encoding the light chain (LC) of the precursor construct is shown. The amino acid sequence is shown from N-terminus to C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 11A One embodiment of the amino acid sequence of the light chain (LC) polypeptide of the precursor construct is shown, having an N-terminal to C-terminal order and the following components: Cap-MMP2 / 9 cleavage-h1F3.1-λLC-anti-EGFR (VH-linker-VL) (SEQ ID NO: 133; Plasmid 7). Figure 11AThe amino acid sequence of the component parts of the LC polypeptide shown includes: CAP (SEQ ID NO: 5), MMP2 / 9 protease cleavable linker (SEQ ID NO: 160 and SEQ ID NO: 9 (cleavable linker)), anti-CD3 epsilon variable light chain and lambda light chain (SEQ ID NO: 74), followed by the labeled cysteine ​​residues (marked in bold and underlined), which can participate in disulfide double bonds, followed by the anti-EGFR scFv VH (SEQ ID NO: 37)-linker (SEQ ID NO: 40)-VL (SEQ ID NO: 34) chain. Figure 11B One embodiment of an optimized nucleic acid sequence (DNA) encoding the light chain (LC) polypeptide of a precursor trispecific (tribody) antibody construct is shown, having a 5' to 3' sequence and the following components (SEQ ID NO: 145). Figure 11A The nucleic acid sequence of the component parts of the LC polypeptide shown includes: CAP (SEQ ID NO: 164), MMP2 / 9 protease cleavable linker (SEQ ID NO: 161 and SEQ ID NO: 33 (cleavable sequence)), anti-CD3ε variable light chain and lambda light chain region (SEQ ID NO: 159), followed by labeled cysteine ​​residues (marked in bold and underlined), which can participate in disulfide double bonds, followed by anti-EGFR scFv VH (SEQ ID NO: 38)-linker (SEQ ID NO: 40)-VL (SEQ ID NO: 36) chains.

[0048] Figure 12A and 12B Shown are embodiments of the amino acid sequence of the heavy chain (HC) polypeptide of a non-cleavable (non-activatable) precursor trispecific (tribody) antibody construct (Construct 5; VLVH) and an optimized nucleotide sequence encoding the heavy chain (HC) of the non-cleavable precursor construct. The amino acid sequence is shown from N-terminus to C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 12A One embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of the non-cleavable precursor construct is shown, having an N-terminal to C-terminal order and the following components: hALB-G-PLGLAG NC-h1F3.5-G1Fd anti-EGFR VL-linker-VH (SEQ ID NO: 134). Figure 12AThe amino acid sequence of the component parts of the HC polypeptide shown includes: human serum albumin (HSA) (SEQ ID NO: 7), a non-cleavable linker (SEQ ID NO: 162), an anti-CD3ε variable heavy chain and a constant heavy chain region 1 (SEQ ID NO: 113), followed by two labeled cysteine ​​residues (marked in bold and underlined), which can participate in disulfide double bonds, followed by an anti-EGFR scFv VL (SEQ ID NO: 34)-linker (SEQ ID NO: 40)-VH (SEQ ID NO: 37) chain. Figure 12B One embodiment of an optimized nucleic acid sequence (DNA) encoding the heavy chain (HC) polypeptide of a non-cleavable precursor trispecific (tribody) antibody construct is shown, having a 5' to 3' sequence and the following components (SEQ ID NO: 146). Figure 12A The nucleic acid sequence of the component parts of the HC polypeptide shown includes: human serum albumin (HSA) (SEQ ID NO: 8), a non-cleavable linker (SEQ ID NO: 163), an anti-CD3ε variable heavy chain and a constant heavy chain region 1 (SEQ ID NO: 155), followed by two labeled cysteine ​​residues (marked in bold and underlined), which can participate in disulfide double bonds, followed by an anti-EGFR scFv VL (SEQ ID NO: 36)-linker (SEQ ID NO: 40)-VH (SEQ ID NO: 38) chain.

[0049] Figure 13A and 13B Shown are embodiments of the amino acid sequence of the light chain (LC) polypeptide of a non-cleavable precursor trispecific (tribody) antibody construct (Construct 5; VLVH) and an optimized nucleotide sequence encoding the light chain (LC) of the non-cleavable precursor construct. The amino acid sequence is shown from N-terminus to C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 13A One embodiment of the amino acid sequence of the light chain (LC) polypeptide of the non-cleavable precursor construct is shown, having an N-terminal to C-terminal order and the following components: Cap-(hlF3.1-λLC anti-EGFR VL-linker-VH non-cleavable) (SEQ ID NO: 135). Figure 13A The amino acid sequence of the component parts of the LC polypeptide shown includes: CAP (SEQ ID NO: 5), a non-cleavable linker (SEQ ID NO: 162), anti-CD3 epsilon variable light chain and lambda light chain (SEQ ID NO: 74), followed by the labeled cysteine ​​residues (marked in bold and underlined), which can participate in disulfide double bonds, followed by the anti-EGFR scFv VL (SEQ ID NO: 34)-linker (SEQ ID NO: 40)-VH (SEQ ID NO: 37) chain. Figure 13B One embodiment of an optimized nucleic acid sequence (DNA) encoding the light chain (LC) polypeptide of a non-cleavable precursor trispecific (tribody) antibody construct is shown, having a 5' to 3' sequence and the following components (SEQ ID NO: 147). Figure 13A The nucleic acid sequence of the component parts of the LC polypeptide shown includes: CAP (SEQ ID NO: 164), a non-cleavable linker (SEQ ID NO: 163), an anti-CD3ε variable light chain and a lambda light chain region (SEQ ID NO: 159), followed by a labeled cysteine ​​residue (marked in bold and underlined), which can participate in a disulfide double bond, followed by an anti-EGFR scFv VL (SEQ ID NO: 36)-linker (SEQ ID NO: 40)-VH (SEQ ID NO: 38) chain.

[0050] Figure 14A and 14B Shown are embodiments of the amino acid sequence of the heavy chain (HC) polypeptide of a non-cleavable precursor trispecific (tribody) antibody construct (Construct 6; VHVL) and an optimized nucleotide sequence encoding the heavy chain (HC) of the non-cleavable precursor construct. The amino acid sequence is shown from N-terminus to C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 14A One embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of the non-cleavable precursor construct is shown, having an N-terminal to C-terminal order and the following components: hALB-G-PLGLAG(NC)-(Clone)-h1F3.5-G1Fd-(VH-Linker-VL) (SEQ ID NO: 136). Figure 14A The amino acid sequence of the component parts of the HC polypeptide shown includes: human serum albumin (HSA) (SEQ ID NO: 7), a non-cleavable linker (SEQ ID NO: 162), an anti-CD3ε variable heavy chain and a constant heavy chain region 1 (SEQ ID NO: 113), followed by two labeled cysteine ​​residues (marked in bold and underlined), which can participate in disulfide double bonds, followed by an anti-EGFR scFv VH (SEQ ID NO: 37)-linker (SEQ ID NO: 40)-VL (SEQ ID NO: 34) chain. Figure 14B One embodiment of an optimized nucleic acid sequence (DNA) encoding the heavy chain (HC) polypeptide of a non-cleavable precursor trispecific (tribody) antibody construct is shown, having a 5' to 3' sequence and the following components (SEQ ID NO: 148). Figure 14AThe nucleic acid sequence of the component parts of the HC polypeptide shown includes: human serum albumin (HSA) (SEQ ID NO: 8), a non-cleavable linker (SEQ ID NO: 163), an anti-CD3ε variable heavy chain and a constant heavy chain region 1 (SEQ ID NO: 155), followed by two labeled cysteine ​​residues (marked in bold and underlined), which can participate in disulfide double bonds, followed by an anti-EGFR scFv VH (SEQ ID NO: 38)-linker (SEQ ID NO: 40)-VL (SEQ ID NO: 36) chain.

[0051] Figure 15A and 15B Shown are embodiments of the amino acid sequence of the light chain (LC) polypeptide of a non-cleavable precursor trispecific (tribody) antibody construct (Construct 6; VLVH) and an optimized nucleotide sequence encoding the light chain (LC) of the non-cleavable precursor construct. The amino acid sequence is shown from N-terminus to C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 15A One embodiment of the amino acid sequence of the light chain (LC) polypeptide of the precursor construct is shown, having an N-terminal to C-terminal order and the following components: Cap(NC)-h1F3.1-λLC-anti-EGFR(VH-Linker-V) (SEQ ID NO: 137). Figure 15A The amino acid sequence of the component parts of the LC polypeptide shown includes: CAP (SEQ ID NO: 5), a non-cleavable linker (SEQ ID NO: 162), anti-CD3ε variable light chain and lambda light chain (SEQ ID NO: 74), followed by labeled cysteine ​​residues (marked in bold and underlined), which can participate in disulfide double bonds, followed by anti-EGFR scFv VH (SEQ ID NO: 37)-linker (SEQ ID NO: 40)-VL (SEQ ID NO: 34) chains. Figure 15B One embodiment of an optimized nucleic acid sequence (DNA) encoding the light chain (LC) polypeptide of a non-cleavable precursor trispecific (tribody) antibody construct is shown, having a 5' to 3' sequence and the following components (SEQ ID NO: 149). Figure 15A The nucleic acid sequence of the component parts of the LC polypeptide shown includes: CAP (SEQ ID NO: 164), a non-cleavable linker (SEQ ID NO: 163), an anti-CD3ε variable light chain and a lambda light chain region (SEQ ID NO: 159), followed by a labeled cysteine ​​residue (marked in bold and underlined), which can participate in a disulfide double bond, followed by an anti-EGFR scFv VH (SEQ ID NO: 38)-linker (SEQ ID NO: 40)-VL (SEQ ID NO: 36) chain.

[0052] Figure 16 Shown is an SDS-PAGE of activated triabody and pro-triabody antibody constructs. Figure 16 In the , all trimeric and pro-trimeric constructs were analyzed by SDS-PAGE, whether in reduced (R) or non-reduced (NR) form. Number 1 is construct 1: activated trimeric construct (VL-VH); number 2 is construct 2: activated trimeric construct (VH-VL); number 3 is construct 3: pro-trimeric construct (VL-VH); number 4 is construct 4: pro-trimeric construct (VH-VL); number 5 is construct 5: non-cleavable pro-trimeric construct (VL-VH); and number 6 is construct 6: non-cleavable pro-trimeric construct (VH-VL). Lanes are labeled with the construct number and reduced (R) or non-reduced (NR) form. The molecular weight of the trimeric construct is 100112 Da and is composed of a fusion of Fd and λC, with MWs of 51105 and 49029, respectively.

[0053] from Figure 16 As can be seen in Figure 3, all proteins are pure and migrate at the appropriate MW as expected, regardless of their reduced or non-reduced forms. The precursor trimer has a molecular weight of 170,390 Da and consists of a fusion of Fd and λC with MWs of 117,649 and 52,762, respectively. Figure 16 As can be seen in the Figure 3, all proteins were pure and migrated at the appropriate MW as expected, both in their reduced and non-reduced forms.

[0054] Figures 17A-17F Analytical HPLC-size exclusion chromatography showing active triameric and pro-triameric antibody constructs. Figures 17A-17F All trimers and precursor trimers were analyzed in their native form by analytical HPLC-size exclusion chromatography (SEC). The trimer had a molecular weight of 100,112 Da and migrated as a monomer at its expected size compared to known MW markers. The precursor trimer had a molecular weight of 170,390 Da and migrated as a monomer at its expected size compared to known MW markers. Figure 17A 17B shows the results for construct 1, 17B shows the results for construct 2, 17C shows the results for construct 3, 17D shows the results for construct 4, 17E shows the results for construct 5, and 17F shows the results for construct 6.

[0055] Figure 18A and 18B ELISA binding studies are shown. Figure 18AELISA binding studies of triabody and precursor triabody antibody constructs for human EGFR antigen are shown. Triabody and precursor triabody antibodies (both VL-VH or VH-VL formats) were tested for their binding to the EGFR extracellular fusion antigen (hEGFR-Fc) using the ELISA method. Figure 18A As can be seen, all trimeric and protrimeric forms bind to the extracellular domain of hEGFR with similar affinity. In addition, there are no major differences between the VL-VH and VH-VL formats of the anti-EGFR scFv. Figure 18B The ELISA method was used to test the ELISA binding of the trio and precursor trio antibodies to the rhesus EGFR antigen trio and precursor trio antibodies (whether in VL-VH or VH-VL format) for their binding to the rhesus EGFR extracellular fusion antigen (rhesus EGFR-Fc). Figure 18B As can be seen, all triads and precursor triads bind to the extracellular domain of rhesus EGFR with similar affinity. There are no major differences between the VL-VH and VH-VL formats of anti-EGFR scFv. In addition, all triads and precursor triads have comparable binding affinities to human and rhesus EGFR. Figure 18A and 18B In the figure: construct 1 is represented by a small circle, construct 2 is represented by a small square, construct 3 is represented by an upward triangle, construct 4 is represented by an inverted triangle, construct 5 is represented by a diamond, and construct 6 is represented by a large circle.

[0056] Figure 19A and 19B ELISA binding studies to human and cynomolgus monkey (cyno) CD3ε antigen are shown. Figure 19A ELISA binding of triabody and pre-triabody antibodies to human CD3ε antigen is shown. Triabody and pre-triabody antibodies (both VL-VH or VH-VL formats) were tested for their binding to human CD3ε extracellular fusion antigen (human CD3ε-histidine) using the ELISA method. Figure 19A As can be seen in the Figure 5, the triabody binds hCD3ε with sub-nM affinity, while the pre-triabody forms (both cleaved (C) and uncleaved (NC)) bind to the extracellular domain of hCD3ε with a higher EC50, suggesting that CD3ε is blocked from binding to its antigen. There were no major differences between the VL-VH and VH-VL formats of anti-CD3ε. Figure 19B ELISA binding of triabody and pre-triabody antibodies to cynomolgus monkey CD3ε antigen is shown. Triabody and pre-triabody antibodies (both VL-VH or VH-VL formats) were tested for their binding to human CD3ε extracellular fusion antigen (canine CD3ε-histidine) using the ELISA method. Figure 19BAs shown, the triabody binds to cynomolgus monkey CD3ε with sub-nM affinity, while the pre-tribody forms (both cleaved (C) and uncleaved (NC)) bind to the extracellular domain of hCD3ε with a higher EC50, suggesting that CD3ε is blocked from binding to its antigen. There are no major differences between the VLVH and VHVL forms of anti-CD3ε. In addition, all triabody and pre-tribody bind with comparable affinity to human and rhesus monkey CD3ε.

[0057] Figure 20 Shown is an SDS-PAGE of the trisome and pretrisome (VHVL) digested with MMP9. Figure 20 In the present study, all trisomes and pre-trisomes were digested by MMP9, and their cleavage products were analyzed by SDS-PAGE under non-reducing (NR) conditions. The trisome has a molecular weight of 100112 Da and consists of a fusion of Fd and λC. MMP9 cleavage has no significant activity against the trisome because it does not have any MMP9 cleavage sequence. The pre-trisome has a molecular weight of 170390 Da. Pre-trisome-C represents the pre-trisome with the MMP9 cleavage sequence at the C-terminus of the half-life extending moiety (human serum albumin) and the C-terminus of the CD3ε CAP masking moiety. Pre-trisome-NC represents the pre-trisome that lacks the MMP9 cleavage sequence and therefore should not be cleaved by MMP9. It can be seen that only pre-trisome-C is cleaved by MMP9, resulting in two bands - the trisome and the half-life extending moiety (human serum albumin (HSA)).

[0058] Figure 21A and 21B ELISA binding of pre-tribody-C (VHVL) and pre-tribody-NC (VHVL) to human CD3 e antigen is shown: cleaved by MMP-9. The pre-tribody VH-VL form of the antibody was tested for its binding to the human CD3 e extracellular fusion antigen (human CD3 e-histidine) using an ELISA method before and after MMP9 cleavage. Figure 21A As can be seen in Figure 3, the pre-tribody-C antibody does not bind to CD3ε before being cleaved by MMP9, but after being cleaved by MMP9, the pre-tribody-C binds to hCD3ε with sub-nM affinity. The pre-tribody-NC antibody does not bind to CD3ε in the presence or absence of MMP9 cleavage ( Figure 21B The data demonstrate that the CAP and HSA blockages in the pro-form result in minimal binding of the pro-form to the CD3ε antigen. Once the pro-trimer-C is cleaved by MMP9, releasing the HSA and CAP blockages for CD3ε, the EC50 for binding to human CD3ε significantly improves toward nanomolar affinity. The pro-trimer-NC antibody is not cleaved and therefore exhibits minimal binding to CD3ε.

[0059] Figure 22A and 22BELISA binding of pre-tribody-C (VHVL) and pre-tribody-NC (VHVL) to cynomolgus monkey CD3 epsilon antigen is shown with and without MMP-9 cleavage. The pre-tribody VH-VL antibodies were tested for their binding to cynomolgus monkey CD3 epsilon extracellular fusion antigen (cynomolgus monkey CD3 epsilon-histidine) using an ELISA method before and after MMP9 cleavage. Figure 22A As can be seen in Figure 3, the pre-tribody-C antibody does not bind to CD3ε before being cleaved by MMP9, but after being cleaved by MMP9, the pre-tribody-C binds to hCD3ε with sub-nM affinity. The pre-tribody-NC antibody does not bind to cynomolgus monkey CD3ε in the presence or absence of MMP9 cleavage ( Figure 22B The data demonstrate that the CAP and HSA blockages in the precursor result in minimal binding to the CD3ε antigen. Once the protrimer-C is cleaved by MMP9, releasing the HSA and CAP blockages of cynomolgus monkey CD3ε, the EC50 for binding to human CD3ε significantly improves toward nanomolar affinity. The protrimer-NC antibody is not cleaved and therefore exhibits minimal binding to CD3ε.

[0060] Figure 23 ELISA binding of trisomy (VHVL), pre-trisomy-C (VHVL) and pre-trisomy-NC (VHVL) to human CD3 epsilon antigen is shown: cleaved by MMP-9. Trisomy VHVL and pre-trisomy VHVL forms of the antibody were tested for their binding to the human CD3 epsilon extracellular fusion antigen (human CD3 epsilon-histidine) using an ELISA method before and after MMP9 cleavage. Figure 23 As can be seen in Figure 3, the pre-tribody-C antibody does not bind CD3ε before cleavage by MMP9, whereas after cleavage by MMP9, the pre-tribody-C antibody binds hCD3ε with sub-nM affinity, similar to the tribody. The pre-tribody-NC antibody does not bind CD3ε in the presence or absence of MMP9 cleavage.

[0061] The data demonstrate that the CAP and HSA blockages in the precursor antibody result in minimal binding to the CD3ε antigen. Once the protrimer-C is cleaved by MMP9, releasing the HSA and CAP blockages on CD3ε, the EC50 for binding to human CD3ε significantly improves toward nanomolar affinity. The protrimer-NC antibody is not cleaved and therefore exhibits minimal binding to CD3ε.

[0062] Figure 24ELISA binding of triabody (VHVL), pre-triabody-C (VHVL) and pre-triabody-NC (VHVL) to cynomolgus monkey CD3 epsilon antigen is shown: cleaved by MMP-9. Triabody VHVL and pre-triabody VHVL forms of the antibody were tested for their binding to cynomolgus monkey CD3 epsilon extracellular fusion antigen (cynomolgus monkey CD3 epsilon-histidine) using an ELISA method before and after MMP9 cleavage. Figure 24 As can be seen in Figure 3, the pre-tribody-C antibody does not bind CD3ε before cleavage by MMP9, whereas after cleavage by MMP9, the pre-tribody-C antibody binds hCD3ε with sub-nM affinity, similar to the tribody. The pre-tribody-NC antibody does not bind CD3ε in the presence or absence of MMP9 cleavage.

[0063] The data demonstrate that the CAP and HSA blockages in the precursor antibody result in minimal binding to the CD3ε antigen. Once the protrimer-C is cleaved by MMP9, releasing the HSA and CAP blockages on CD3ε, the EC50 for binding to human CD3ε improves significantly toward low nanomolar affinity. The protrimer-NC antibody is not cleaved and therefore exhibits minimal binding to CD3ε.

[0064] Figure 25 FACS binding of trisomy (VHVL), pre-trisomy-C (VHVL) and pre-trisomy-NC (VHVL) to Jurkat cells (human CD3e) is shown. The trisomy VHVL and pre-trisomy VHVL forms of the antibody were tested for their binding to Jurkat cells (human CD3e) using the FACS method. Figure 25 As can be seen in the Figure 3 , the pre-tribody-C and pre-tribody-NC antibodies do not bind to Jurkat cells, while the tribody binds with low nM affinity. The data indicate that the CAP and HSA blockade of the pre-structure results in very little binding to Jurkat cells expressing human CD3ε. Once the HSA and CAP blockade of CD3ε is absent (tribody (VHVL)), the EC50 for binding to human CD3ε improves significantly toward low nM binding affinity.

[0065] Figures 26A-26B Embodiments of the amino acid and nucleic acid sequences of the scFv anti-ROR1 VL-VH are shown. Figure 26A One embodiment of the amino acid sequence of the scFv anti-ROR1 is shown as follows, with the N-terminal to C-terminal order: VL-VH, and the following components: anti-ROR1 VL, linker, and anti-ROR1 VH (SEQ ID NO: 156). Figure 26B One embodiment of an optimized nucleic acid sequence encoding a scFv anti-ROR1 is shown below, with an N-terminal to C-terminal order: VL-VH (SEQ ID NO: 157).

[0066] Figures 27A-27B Embodiments of the amino acid and nucleic acid sequences of the scFv anti-ROR1 VH-VL are shown. Figure 27A One embodiment of the amino acid sequence of the scFv anti-ROR1 is shown as follows, with the N-terminal to C-terminal order: VH-VL, and the following components: anti-ROR1 VH, linker, and anti-ROR1 VL (SEQ ID NO: 166). Figure 27B One embodiment of an optimized nucleic acid sequence encoding a scFv anti-ROR1 is shown below, with an N-terminal to C-terminal order: VH-VL (SEQ ID NO: 167).

[0067] Figures 28A-28B Embodiments of the amino acid and nucleic acid sequences of the scFv anti-PSMA VL-VH are shown. Figure 28A One embodiment of the amino acid sequence of the scFv anti-PSMA is shown as follows, with the N-terminal to C-terminal order: VL-VH, and the following components: anti-PSMA VL, linker, and anti-PSMA VH (SEQ ID NO: 168). Figure 28B One embodiment of an optimized nucleic acid sequence encoding a scFv anti-PSMA is shown below, with an N-terminal to C-terminal order: VL-VH (SEQ ID NO: 169).

[0068] Figures 29A-29B Embodiments of the amino acid and nucleic acid sequences of the scFv anti-PSMA VH-VL are shown. Figure 29A One embodiment of the amino acid sequence of the scFv anti-PSMA is shown as follows, with the N-terminal to C-terminal order: VH-VL, and the following components: anti-PSMA VH, linker, and anti-PSMA VL (SEQ ID NO: 170). Figure 29B One embodiment of an optimized nucleic acid sequence encoding a scFv anti-PSMA is shown below, with an N-terminal to C-terminal order: VH-VL (SEQ ID NO: 171).

[0069] Figures 30A-30B Embodiments of the amino acid and nucleic acid sequences of the scFv anti-5T4 VL-VH are shown. Figure 30A One embodiment of the amino acid sequence of the scFv anti-5T4 is shown as follows, with the N-terminal to C-terminal order: VL-VH, and the following components: anti-5T4 VL, linker, and anti-5T4 VH (SEQ ID NO: 172). Figure 30B One embodiment of an optimized nucleic acid sequence encoding a scFv anti-5T4 is shown below, with an N-terminal to C-terminal order: VL-VH (SEQ ID NO: 173).

[0070] Figures 31A-31B Embodiments of the amino acid and nucleic acid sequences of the scFv anti-5T4 VH-VL are shown. Figure 31A One embodiment of the amino acid sequence of the scFv anti-5T4 is shown as follows, with the N-terminal to C-terminal order: VH-VL, and the following components: anti-5T4 VH, linker, and anti-5T4 VL (SEQ ID NO: 1174). Figure 31B One embodiment of an optimized nucleic acid sequence encoding a scFv anti-5T4 is shown below, with an N-terminal to C-terminal order: VH-VL (SEQ ID NO: 1175).

[0071] Figure 32 FACS binding data for EGFR-binding triameric and pre-triameric antibody constructs cleaved by MMP9 to Jurkat cells are shown. The MFI (mean fluorescence intensity) provides the relative proportion of antibody binding. Constructs analyzed: Triameric(VHVL) is construct 2 from Example 1 (small circles, squares); pre-triameric-C(VHVL) is construct 4 from Example 1 (upward and downward triangles); and pre-triameric-NC(VHVL) is construct 6 from Example 1 (diamonds and large circles). -MMP9 and +MMP9 indicate the absence (-) or presence (+) of MMP9 protease.

[0072] Figures 33A-33B EGFR activated ( Figure 33A ) trimeric constructs and precursors ( Figure 33B ) In vivo pharmacokinetic values ​​(PK values) of the trimeric construct. Antibody construct: Trimeric EGFR (VL-VH) is the construct 1 ( Figure 33A ); Protribody-C EGFR (VL-VH) is construct 3 of Example 1 ( Figure 33B ); and protrimer-NC EGFR (VL-VH) is construct 5 of Example 1 ( FIG. 33 ). The constructs were administered intravenously at two concentrations: 0.5 mg / kg and 2 mg / kg. The results provide half-life information for each construct.

[0073] Figures 34A-34F SDS-PAGE results (34A and 34D) and analytical HPLC-size exclusion chromatography results (Figure 34B, Table 34C, Figure 34E, and Table 34F) are shown. Figures 34A-34C Results are shown for a trio-ROR1 (VL-VH) construct in which both the first and second binding sites bind ROR1 and the third binding site is a CD3ε binding site. No regulatory arms are present (Construct 7 of Example 8). Figures 34D-34FResults are shown for a protrimer-ROR1 (VL-VH) construct, in which both the first and second binding sites bind ROR1 and the third binding site is a CD3ε binding site, wherein the construct comprises a CAP regulatory domain that masks CD3ε binding and an HLP regulatory domain comprising HSA, wherein both regulatory domains are linked at the N-terminus to the CD3ε binding site by a non-cleavable linker (construct 11 of Example 8).

[0074] Figures 35A-35I Shown are the results of SDS-PAGE under reducing and non-reducing conditions (35A and 35D and 35G), and the results of analytical HPLC-size exclusion chromatography (Figure 35B, Table 35C, Figure 35E, Table 35F, Figure 35H and Table 35I). Figures 35A-35C Results are shown for a tria-5T4 (VL-VH) construct, in which both the first and second binding sites bind 5T4 and the third binding site is a CD3ε binding site (construct 13 of Example 8). Figures 35D-35F Results are shown for a protrimer-5T4 (VL-VH) construct, in which both the first and second binding sites bind 5T4 and the third binding site is a CD3ε binding site, wherein the construct comprises a CAP regulatory domain that masks CD3ε binding and an HLP regulatory domain comprising HSA, wherein both regulatory domains are linked at the N-terminus to the CD3ε binding site via a cleavable linker (construct 15 of Example 8). Figures 35G-35I Results are shown for a protrimer-5T4 (VL-VH) construct, in which both the first and second binding sites bind 5T4 and the third binding site is a CD3ε binding site, wherein the construct comprises a CAP regulatory domain that masks CD3ε binding and an HLP regulatory domain comprising HSA, wherein both regulatory domains are linked at the N-terminus to the CD3ε binding site by a non-cleavable linker (construct 17 of Example 8).

[0075] Figure 36Binding curves are shown showing binding of triabody and pre-triabody antibodies to the 5T4 antigen. 5T4-Triabody is an activated triabody construct in which the first and second binding domains bind 5T4, the third domain binds CD3ε, and there is no regulatory domain. 5T4-PT Triabody-C is a pre-triabody construct in which the first and second binding domains bind 5T4, the third domain binds CD3ε, and has first and second sub-regulatory domains, each linked to the N-terminus of the Fab of the CD3ε binding domain, wherein the linker within the regulatory domain is cleavable by MMP9. 5T4-PT Triabody-NC is a pre-triabody construct in which the first and second binding domains bind 5T4, the third domain binds CD3ε, and has first and second sub-regulatory domains, each linked to the N-terminus of the Fab of the CD3ε binding domain, wherein the linker within the regulatory domain is non-cleavable.

[0076] Figure 37 SDS-PAGE results of MMP9 protease cleavage of 5T4 pre-trimeric antibody constructs are shown. Tribody-5T4 (VL-VH) is an activated tribody construct in which the first and second binding domains bind 5T4, the third domain binds CD3ε, and there is no regulatory domain (Construct 13 of Example 8). Pre-tribody-5T4-C-(VL-VH) is a pre-tribody construct in which the first and second binding domains bind 5T4, the third domain binds CD3ε, and has a first and a second sub-regulatory domain, each of which is linked to the N-terminus of the Fab of the CD3ε binding domain, wherein the linker within the regulatory domain is cleavable by MMP9 (Construct 15 of Example 8). Protrin-5T4-NC-(VL-VH) is a protrin construct in which the first and second binding domains bind to 5T4, the third domain binds to CD3ε, and has a first and a second sub-regulatory domain, each of which is connected to the N-terminus of the Fab of the CD3ε binding domain, wherein the linker within the regulatory domain is not cleavable (Construct 17 of Example 8).

[0077] Figure 38Binding curves are shown showing binding of triabody and pre-triabody antibody constructs to human CD3ε antigen in the presence or absence of MMP9 protease ((+MMP9) or (-MMP9)), respectively. 5T4-Triabody is an activated triabody construct in which the first and second binding domains bind 5T4, the third domain binds CD3ε, and there is no regulatory domain. 5T4-PT Triabody-C is a pre-triabody construct in which the first and second binding domains bind 5T4, the third domain binds CD3ε, and has first and second sub-regulatory domains, each linked to the N-terminus of the Fab of the CD3ε binding domain, wherein the linker within the regulatory domain is cleavable by MMP9. 5T4-PT Triabody-NC is a pre-triabody construct in which the first and second binding domains bind 5T4, the third domain binds CD3ε, and has first and second sub-regulatory domains, each linked to the N-terminus of the Fab of the CD3ε binding domain, wherein the linker within the regulatory domain is non-cleavable.

[0078] Figure 39 Shown are FACS binding of trisomy-5T4, pretrisomy-5T4-C, and pretrisomy-5T4-NC to Jurkat cells (human CD3e) in the presence and absence of MMP9 protease.

[0079] Figure 40 ELISA binding studies of triameric and pre-triameric antibody constructs against human 5T4 antigen are shown.

[0080] Figure 41 Shown are FACS binding data for 5T4 binding triameric and pro-triameric antibody constructs to CHO cells expressing human 5T4. The MFI (mean fluorescence intensity) provides the relative proportion of antibody binding.

[0081] Figure 42 Shown are FACS binding data for 5T4 binding triabody and pro-triabody antibody constructs to the MCF7 breast cancer cell line known to highly express human 5T4. The MFI (mean fluorescence intensity) provides the relative proportion of antibody binding.

[0082] Figure 43 and 44 The 5T4 binding triabody and pro-triabody antibody constructs were shown to be effective against the MCF7 breast cancer cell line ( Figure 43 ) and NCI-H226 lung cancer cell lines ( Figure 44 ) cytotoxicity assay.

[0083] Figures 45A-45B Two embodiments of the amino acid sequences of the heavy chain (HC) polypeptides of the precursor trispecific (trimeric) antibody constructs are shown (e.g. Figure 2F(a) an embodiment of a HC wherein both scFvs bind EGFR. The amino acid sequences are shown from N-terminus to C-terminus. Figure 45A and 45B An embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of the precursor construct is shown, with N-terminal to C-terminal order and the following components: CAP (bold), linker (italic), human serum albumin (underlined); protease cleavage sequence (bold and italic); VH1 / CH1 (bold and underlined), VL (italic and underlined), VH (italic, bold and underlined) of anti-CD3 e Fab (EGFR scFv). The two labeled cysteine ​​residues (double underlined) can participate in disulfide double bonds. Figure 45A The protease cleavage sequence in is a multiple protease cleavage sequence. Figure 45B The protease cleavage sequence in is the MMP2 / 9 protease cleavage sequence. Figure 45A The amino acid sequence of is set forth in SEQ ID NO:28. Figure 45B The amino acid sequence of is set forth in SEQ ID NO: 31. Embodiments of sequences of component parts are described throughout this application.

[0084] Figure 46 Embodiments of the amino acid sequences of the light chain (LC) polypeptides of the precursor trispecific (trimeric) antibody constructs are shown (e.g. Figure 2F (a) an embodiment of an LC wherein both scFvs bind EGFR. The amino acid sequences are shown from N-terminus to C-terminus. Figure 46 One embodiment of the amino acid sequence of the light chain (LC) polypeptide of the precursor construct is shown, with N-terminal to C-terminal order and the following components: linker (italic), VL1-CL (double underlined and bold), VL (italic and underlined) and VH (italic and underlined) of the anti-CD3 e Fab (wherein VH is an EGFR scFv). The two labeled cysteine ​​residues (double underlined) can participate in disulfide double bonds. Figure 46 The amino acid sequence of LC shown in is set forth in SEQ ID NO: 32. Embodiments of the amino acid sequences of the component parts of the LC polypeptide are described throughout this application.

[0085] Figures 47A-47B Two embodiments of the amino acid sequences of the heavy chain (HC) polypeptides of the precursor trispecific (trimeric) antibody constructs are shown (e.g. Figure 2F (a) an embodiment of a HC wherein both scFvs bind 5T4. Amino acid sequences are shown from N-terminus to C-terminus. Figure 47A and 47BAn embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of the precursor construct is shown, with N-terminal to C-terminal order and the following components: CAP (bold), linker (italic), human serum albumin (underlined); protease cleavage sequence (bold and italic); VH1 / CH1 (bold and underlined), VL (italic and underlined), VH (italic, bold and underlined) of anti-CD3 e Fab (5T4scFv). The two labeled cysteine ​​residues (double underlined) can participate in disulfide double bonds. Figure 47A The protease cleavage sequence in is a multiple protease cleavage sequence. Figure 47B The protease cleavage sequence in is the MMP2 / 9 protease cleavage sequence. Figure 47A The amino acid sequence of is set forth in SEQ ID NO:118. Figure 47B The amino acid sequence of is set forth in SEQ ID NO: 176. Embodiments of sequences of component parts are described throughout this application.

[0086] Figure 48 Embodiments showing the amino acid sequences of the light chain (LC) polypeptides of the precursor trispecific (trimeric) antibody constructs (e.g. Figure 2F (a) an embodiment of an LC wherein both scFvs bind 5T4. Amino acid sequences are shown from N-terminus to C-terminus. Figure 48 An embodiment of the amino acid sequence of the light chain (LC) polypeptide of the precursor construct is shown, with the following components in N-terminal to C-terminal order: linker (italic), VL1-CL (double underlined and bold), VL (italic and underlined), and VH (italic and underlined) of an anti-CD3 epsilon Fab, wherein VH is a 5T4 scFv. The two labeled cysteine ​​residues (double underlined) can participate in disulfide double bonds. Figure 48 The amino acid sequence of LC shown in is set forth in SEQ ID NO: 177. Embodiments of the amino acid sequences of the modular portions of the LC polypeptide are described throughout this application.

[0087] It should be understood that for simplicity and clarity of illustration, the elements shown in the figures are not necessarily drawn to scale. For example, the dimensions of some elements may be exaggerated relative to other elements for clarity. Further, where deemed appropriate, reference numerals may be repeated in the figures to indicate corresponding or similar elements. DETAILED DESCRIPTION

[0088] In the following detailed description, many specific details are described to provide a thorough understanding of the precursor trispecific antibody constructs. However, those skilled in the art will appreciate that the precursor constructs, their production and use described herein can be practiced without these specific details. In other cases, well-known methods, procedures and components have not been described in detail in order to avoid obscuring the present disclosure.

[0089] Described herein are precursor trispecific antibody constructs comprising a separate cleavable masking domain and a half-life extension domain, wherein these cleavable regulatory domains provide reduced precursor trispecific constructs for binding to T cells outside the tumor microenvironment (TME), and provide an extended half-life. Half-life extension may be limited to the time of the precursor trispecific construct outside the cancer microenvironment, or it may be extended to the time when the precursor trispecific construct resides in the cancer microenvironment. The advantages of the precursor trispecific antibody constructs described herein having a protease-cleavable masking domain and a protease-cleavable half-life extension (HLP) domain may be the controlled release of improved protease activation of the masking CAP and HLP domains.

[0090] The reduction of T cell binding may lead to a reduction of T cell activation. In some embodiments, the precursor trispecific antibody constructs described herein are regulatable precursor constructs. The regulatable precursor trispecific antibody constructs described herein may have an extended half-life, or reduced T cell binding, or reduced T cell activation, or any combination thereof.

[0091] In some embodiments, the precursor trispecific antibody constructs described herein provide regulatable T cell activation, wherein the precursor construct provides T cell activation limited to the tumor microenvironment. In some embodiments, the precursor trispecific antibody constructs described herein have an enhanced half-life and provide T cell activation limited to the tumor microenvironment compared to a non-precursor always active multivalent antibody. In some embodiments, the precursor trispecific antibody constructs described herein have reduced T cell activation in a non-tumor microenvironment compared to a non-precursor always active multivalent antibody.

[0092] In some embodiments, the precursor trispecific antibody constructs described herein have an extended half-life in a non-tumor microenvironment compared to a non-precursor trispecific antibody. In some embodiments, the precursor trispecific antibody constructs described herein have reduced T cell binding and / or activation in a non-tumor microenvironment and an extended half-life in a non-tumor microenvironment compared to a non-precursor always active multivalent antibody.

[0093] In some embodiments, the precursor tri-specific antibody constructs described herein can bind to T cells. In some embodiments, the precursor tri-specific antibody constructs described herein can bind to T cells within the tumor microenvironment (TME). In some embodiments, the precursor tri-specific antibody constructs described herein can bind to T cells and two different TAAs. In some embodiments, the precursor tri-specific antibody constructs described herein can bind to T cells and two different TAAs, wherein the TAAs comprise different extracellular epitopes of tumor cell surface antigens. In some embodiments, the precursor tri-specific antibody constructs described herein can bind to T cells and two different TAAs, wherein the TAAs comprise extracellular epitopes of tumor cell surface antigens. In some embodiments, the precursor tri-specific antibody constructs described herein can bind to T cells and two different TAAs, wherein the TAAs comprise extracellular epitopes of tumor cell surface antigens and TME antigens. In some embodiments, the precursor tri-specific antibody constructs described herein can bind to T cells and two different TAAs, wherein the TAAs comprise extracellular epitopes of tumor cell surface antigens and stromal antigens in the TME. In some embodiments, the precursor trispecific antibody constructs described herein can bind to T cells and two different TAAs, wherein the TAAs comprise extracellular epitopes of tumor cell surface antigens and angiogenic antigens in the TME. In some embodiments, the precursor trispecific antibody constructs described herein can bind to T cells and two different TAAs, wherein the TAAs comprise extracellular epitopes of tumor cell surface antigens and antigens on the surface of blood vessels in the TME. In some embodiments, the antigens on the surface of blood vessels are endothelial cell surface antigens. In some embodiments, the antigens on the surface of blood vessels are endothelial cell surface antigens selected from CD31, CD105, CD146, and CD144. In some embodiments, the precursor trispecific antibody constructs described herein can bind to T cells and two different TAAs, wherein the TAAs comprise extracellular epitopes of tumor cell surface antigens and cytokines in the TME.

[0094] In some embodiments, the matrix antigens in the TME comprise fibroblast activation protein (FAP). In some embodiments, the matrix antigens in the TME comprise alpha smooth muscle actin (αSMA). In some embodiments, the matrix antigens in the TME comprise PDGFRα. In some embodiments, the matrix antigens in the TME comprise integrin α11β1 (ITGA11). In some embodiments, the matrix antigens in the TME comprise VEGF. In some embodiments, the matrix antigens in the TME comprise tenascin-C, periostin. In some embodiments, the matrix antigens in the TME comprise fibroblast-specific protein 1 (S10A4, FSP1). In some embodiments, the matrix antigens in the TME comprise desmin. In some embodiments, the matrix antigens in the TME comprise vimentin. In some embodiments, the matrix antigens in the TME comprise paladin. In some embodiments, the matrix antigens in the TME comprise urokinase-type plasminogen activator receptor-associated protein (UPARAP). In some embodiments, the matrix antigens in the TME comprise galectin-3. In some embodiments, the matrix antigens in the TME comprise podoplanin. In some embodiments, the matrix antigen in the TME comprises platelets. In some embodiments, the matrix antigen in the TME comprises CCL2. In some embodiments, the matrix antigen in the TME comprises CXCL12. In some embodiments, the matrix antigen in the TME is selected from fibroblast activation protein (FAP), alpha smooth muscle actin (αSMA), PDGFRα, integrin α11β1 (ITGA11) VEGF, tenascin-C, periostin, fibroblast-specific protein 1 (S10A4, FSP1), desmin, vimentin, paladin, urokinase-type plasminogen activator receptor-associated protein (UPARAP), galectin-3, podoplanin, platelets, CCL2 or CXCL12.

[0095] In some embodiments, described herein are pharmaceutical compositions comprising precursor trispecific antibody constructs that provide regulatable T cell activation in a non-tumor microenvironment. In some embodiments, described herein are pharmaceutical compositions comprising precursor trispecific antibody constructs that have an increased half-life and provide T cell activation limited to a tumor microenvironment. In some embodiments, described herein are pharmaceutical compositions comprising precursor trispecific antibody constructs that include an extended half-life in a non-tumor microenvironment. In some embodiments, described herein are pharmaceutical compositions comprising precursor trispecific antibody constructs that include an extended half-life in a non-tumor microenvironment, wherein the half-life in a tumor microenvironment is reduced compared to the half-life in a non-tumor microenvironment. In some embodiments, the pharmaceutical composition comprising a precursor construct includes a construct that recognizes T cells and TAAs. In some embodiments, the pharmaceutical composition comprising a precursor construct includes a construct that recognizes T cells and two TAAs, wherein each TAA is a different antigen. In some embodiments, a pharmaceutical composition comprising a precursor construct comprises a construct that recognizes a T cell and two TAAs, wherein the TAAs recognize the same antigen.

[0096] In some embodiments, described herein are methods of using the precursor trispecific antibody constructs as disclosed herein for treating, preventing, inhibiting growth, slowing disease progression, reducing tumor burden, or reducing the incidence of a cancer or tumor in a subject, or any combination thereof. In some embodiments, the treatment methods disclosed herein reduce minimal residual disease, increase remission, increase the duration of remission, reduce the rate of tumor recurrence, prevent metastasis of a tumor or cancer, or reduce the rate of metastasis of a tumor or cancer, or any combination thereof, in a treated subject compared to a subject not administered the pharmaceutical composition.

[0097] Precursor trispecific antibody constructs

[0098] In some embodiments, the precursor trispecific antibody construct comprises a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε.

[0099] In some embodiments, disclosed herein is a precursor trispecific antibody construct comprising: a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; and a regulatory domain, wherein the regulatory domain includes either a first and a second sub-regulatory domain or a single regulatory domain, wherein the first sub-regulatory domain comprises a first protease cleavage domain and a half-life extension (HLP) domain, and the second sub-regulatory domain comprises a second protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε; the single regulatory domain comprises a protease cleavage domain, a half-life extension (HLP) domain, and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε.

[0100] In some embodiments, disclosed herein are precursor trispecific antibody constructs comprising: a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; and a regulatory domain comprising a protease cleavage domain, a half-life extension (HLP) domain, and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε.

[0101] In some embodiments, the precursor trispecific antibody construct comprises a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε.

[0102] In some embodiments, the precursor trispecific antibody construct comprises a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε.

[0103] With respect to the precursor constructs described throughout, the skilled artisan will appreciate that the modular structure of the constructs allows for different binding partners based on the amino acid sequences contained in the first and second binding sites.

[0104] The skilled artisan will recognize that a precursor antibody construct is a precursor or "pre-(Pro)" form of an active antibody protein. In some embodiments, the term "pre" is used interchangeably with the term "precursor" and has the same meaning and properties.

[0105] The skilled artisan will understand that, in some embodiments, the terms "precursor trispecific antibody construct," "precursor trispecific antibody construct," "precursor antibody," "precursor construct," "precursor antibody construct," "precursor trispecific antibody," "trispecific antibody," "antibody," "trispecific antibody construct," and "trispecific construct" as used herein can be used interchangeably and all have the same properties and meanings. Additionally, in some embodiments, the term "trispecific" can be replaced with the term "tribody" as it is recognized that the antibody constructs disclosed herein have three antibody body regions, each of which can bind to a different antigen (tribody or trispecific) or two of the three binding regions can bind to the same antigen (tribody or trispecific, wherein the two specific binding antigens are the same). Thus, the terms listed above, for example, in some embodiments, "precursor trispecific antibody construct" can be used interchangeably with the term "precursor tribody construct" and all have the same meanings and properties.

[0106] In some embodiments, the precursor trispecific antibody construct comprises a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε, wherein the first TAA or the second TAA or both the first TAA and the second TAA comprise an extracellular epitope of a tumor cell surface antigen.

[0107] In some embodiments, the precursor trispecific antibody construct comprises a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε, wherein the first TAA or the second TAA or both the first TAA and the second TAA comprise a TME.

[0108] In some embodiments, the precursor trispecific antibody construct comprises a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε, wherein the first TAA or the second TAA or both the first TAA and the second TAA comprise a stromal antigen in the TME.

[0109] In some embodiments, the precursor trispecific antibody construct comprises a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε, wherein the first TAA or the second TAA or both the first TAA and the second TAA comprise an angiogenic antigen in the TME.

[0110] In some embodiments, the precursor trispecific antibody construct comprises a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε, wherein the first TAA or the second TAA or both the first TAA and the second TAA comprise an antigen on the surface of a blood vessel in the TME.

[0111] In some embodiments, the precursor trispecific antibody construct comprises a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε, wherein the first TAA or the second TAA or both the first TAA and the second TAA comprise a cytokine antigen in the TME.

[0112] In some embodiments, the precursor trispecific antibody construct comprises a first binding domain that binds to a first tumor-associated antigen (TAA); a second binding domain that binds to a second TAA; a third binding domain that binds to an extracellular epitope of human CD3ε; a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of human CD3ε, wherein the first TAA or the second TAA or both the first TAA and the second TAA comprise an extracellular epitope of a tumor cell surface antigen or TME, or a stromal antigen in the TME, or an angiogenic antigen in the TME, or an antigen on the surface of a blood vessel in the TME, or a cytokine antigen in the TME, or any combination thereof.

[0113] The skilled artisan will understand that in some embodiments, the term "tumor-associated antigen" (TAA) may include molecules or portions thereof displayed on the surface of a cell or molecules present in the tumor environment (i.e., within the tumor microenvironment). In some embodiments, TAAs include cell surface tumor-associated antigens (TAAs). In some embodiments, the cell is a tumor cell. In some embodiments, the cell is a non-tumor cell present in the tumor environment, such as, but not limited to, a cell present in the vasculature tissue associated with a tumor or cancer. In some embodiments, the TAA is an angiogenic antigen in the tumor microenvironment. In some embodiments, the TAA is an antigen on a blood vessel in the tumor microenvironment. In some embodiments, the cell is a stromal cell present in the tumor environment. In some embodiments, the TAA is a stromal cell antigen within the tumor microenvironment. In some embodiments, the TAA includes an extracellular epitope of a tumor cell surface antigen. In some embodiments, the TAA includes an extracellular matrix antigen.

[0114] In some embodiments, the angiogenic antigen comprises bFGF. In some embodiments, the angiogenic antigen comprises INF. In some embodiments, the angiogenic antigen comprises VEGF. In some embodiments, the angiogenic antigen comprises bFGF, INF, or VEGF.

[0115] In some embodiments, the TAA comprises an antigen present in the TME. In some embodiments, the TAA comprises a cytokine antigen in the TME. In some embodiments, the TAA comprises a molecule secreted into the TME by a tumor cell. In some embodiments, the TAA comprises an effector molecule secreted into the TME by a tumor cell. In some embodiments, the effector molecule comprises a cytokine antigen. In some embodiments, the effector molecule comprises a cytokine antigen in the TME.

[0116] In some embodiments, the cytokine antigen in the TME comprises TNF-α, IL-6, TGF-β, IL-10, IL-8, IL-17, IL-21, INF, or VEGF. In some embodiments, the TAA is selected from TNF-α, IL-6, TGF-β, IL-10, IL-8, IL-17, IL-21, INF, or VEGF. In some embodiments, the cytokine antigen used as a TAA includes cytokine antigens known in the art.

[0117] In some embodiments, the first TAA or the second TAA or both the first TAA and the second TAA comprise an extracellular epitope of a tumor cell surface antigen, a tumor microenvironment antigen, a stromal antigen in the tumor microenvironment (TME), an angiogenic antigen in the TME, an antigen on a blood vessel in the TME, or a cytokine antigen in the TME, or any combination thereof.

[0118] The skilled artisan will understand that the terms "tumor microenvironment" (TME), "cancer microenvironment," and "tumor environment" are used interchangeably, have the same nature and meaning, and encompass the microenvironment in which tumors develop. While the normal cellular microenvironment can inhibit the growth of malignant cells, changes in the tumor microenvironment can synergistically support cell proliferation.

[0119] In some embodiments, the first binding domain and the second binding domain of the precursor construct disclosed herein bind to the same TAA. In some embodiments, the first binding domain and the second binding domain of the precursor construct disclosed herein bind to different TAAs. In some embodiments, the first binding domain and the second binding domain of the precursor construct disclosed herein bind to different TAAs on the same cell. In some embodiments, the first binding domain and the second binding domain of the precursor construct disclosed herein bind to different TAAs on different cells. In some embodiments, the TAA comprises a cell surface antigen on a tumor cell. In some embodiments, the TAA comprises a cell surface antigen on a cell in the TME.

[0120] In some embodiments, the first binding domain and the second binding domain bind to different TAAs. The different TAAs can be, for example, but not limited to, extracellular epitopes of tumor cell surface antigens, TME antigens, stromal antigens in the TME, angiogenic antigens in the TME, antigens on blood vessels in the TME, or cytokines in the TME.

[0121] The skilled artisan will understand that the term "antigen" or "immunogen" includes an immunogenic peptide, protein or polypeptide or any fragment thereof. In some embodiments, an antigen is capable of eliciting an immune response in a mammal and therefore comprises at least one and may comprise multiple epitopes. An "antigen" molecule or a portion of a molecule is capable of being bound by a selective binding agent, such as the antigen binding portion of a Fab fragment or the antigen binding portion of a single-chain variable fragment (scFv). In addition, an "antigen" can be used in an animal to produce an antibody capable of binding to an epitope of the antigen. In some embodiments, the CAP assembly comprises a portion of an antigen that is bound by a third binding domain.

[0122] The term "epitope" includes any determinant, and in certain embodiments, a polypeptide determinant is capable of specific binding to an anti-TAA binding domain or an anti-T cell receptor binding domain. An epitope is a region of an antigen that is bound by an antibody or antigen-binding fragment thereof. In some embodiments, the CAP assembly comprises an epitope that is bound by a third binding domain.

[0123] In certain embodiments, epitope determinants include chemically active surface groupings of molecules, such as amino acids, sugar side chains, phosphoryl or sulfonyl groups, and in certain embodiments may have specific three-dimensional structural characteristics and / or specific charge characteristics. In certain embodiments, when a precursor trispecific antibody construct preferentially recognizes a target antigen in a complex mixture of proteins and / or macromolecules, the precursor trispecific antibody construct is said to specifically bind to the antigen. When the equilibrium dissociation constant is ≤10 -5 , 10 -6 or 10 -7 When M, the precursor trispecific antibody construct is said to specifically bind to the antigen. In some embodiments, the equilibrium dissociation constant can be ≤10 -8 M or 10 -9 M. In some further embodiments, the equilibrium dissociation constant may be ≤10 -10 M or 10 -11 M. Antigens disclosed herein include, but are not limited to, TAAs, CAP components, and immune effector molecules, such as human CD3ε polypeptide.

[0124] In some embodiments, the tumor associated antigen (TAA) is a tumor antigen. In some embodiments, the tumor antigen comprises those antigens presented on tumor cells. In some embodiments, the tumor antigen is present on solid tumor cells. In some embodiments, the tumor antigen is a cancer antigen present on non-solid tumor cells.

[0125] In some embodiments, when the TAA is a tumor cell antigen, the tumor cell includes a cell from a solid tumor. Solid tumors may be benign (not cancer) or malignant (cancer). Different types of solid tumors are named after the cell types that form them. Examples of solid tumors are sarcomas, carcinomas, and lymphomas. In some embodiments, a solid tumor is a vegetation (new growth of cells) or a lesion (damage to anatomy or a disorder of physiological function) formed by the abnormal growth of body tissue cells other than blood, bone marrow, or lymphocytes. In some embodiments, a solid tumor is composed of abnormal cell populations that may be derived from different tissue types, such as the liver, colon, breast, or lung, and initially grow in the organ from which the cells originate. However, these cancers may spread to other organs through metastatic tumor growth in the late stages of the disease.

[0126] In some embodiments, the solid tumor comprises a sarcoma or carcinoma, a fibrosarcoma, a myxosarcoma, a liposarcoma, a chondrosarcoma, an osteosarcoma, a chordoma, an angiosarcoma, an endotheliosarcoma, a lymphangiosarcoma, a lymphangioendotheliosarcoma, a synovioma, a mesothelioma, an Ewing sarcoma, a leiomyosarcoma, a rhabdomyosarcoma, a colon cancer, a pancreatic cancer or tumor, a breast cancer or tumor, an ovarian cancer or tumor, a prostate cancer or tumor, a squamous cell carcinoma, a basal cell carcinoma, an adenocarcinoma, a sweat gland carcinoma, a sebaceous gland carcinoma, a papillary carcinoma, a papillary adenocarcinoma, a cystic gland, or a cystic gland. In some embodiments, the solid tumor comprises adrenal cortex tumor (adenoma and carcinoma), carcinoma, colorectal cancer, desmoid tumor, desmoid small round cell tumor, endocrine tumor, Ewing sarcoma, germ cell tumor, hepatoblastoma, hepatocellular carcinoma, melanoma, neuroblastoma, osteosarcoma, retinoblastoma, rhabdomyosarcoma, soft tissue sarcoma other than rhabdomyosarcoma and Wilms' tumor. In some embodiments, the solid tumor is a breast tumor. In another embodiment, the solid tumor is prostate cancer. In another embodiment, the solid tumor is colon cancer. In some embodiments, the tumor is a brain tumor. In another embodiment, the tumor is a pancreatic tumor. In another embodiment, the tumor is a colorectal tumor.

[0127] In some embodiments, the tumor cells include cells from a non-solid tumor, i.e., a non-solid cancer. In some embodiments, the cancer may be a diffuse cancer, in which the cancer spreads widely; it is not localized or restricted. In some embodiments, diffuse cancers may include non-solid tumors. Examples of diffuse cancers include leukemias. Leukemias include cancers that begin in hematopoietic tissue (e.g., bone marrow) and cause large numbers of abnormal blood cells to be produced and enter the blood.

[0128] In some embodiments, the diffuse cancer comprises a B cell malignancy. In some embodiments, the diffuse cancer comprises a leukemia. In some embodiments, the cancer is a lymphoma. In some embodiments, the lymphoma is a large B cell lymphoma.

[0129] In some embodiments, diffuse cancer or tumor comprises hematologic malignancy.In some embodiments, hematologic malignancy is the type of cancer that affects blood, bone marrow and lymph nodes.Hematologic malignancy may be derived from any of two major blood cell lineages: myeloid and lymphoid cell lineages.Myeloid cell lineages produce granulocytes, erythrocytes, platelets, macrophages and mast cells conventionally, while lymphoid cell lineages produce B, T and plasma cells.Lymphoma (such as Hodgkin's lymphoma), lymphocytic leukemia and myeloma are derived from lymphoid lineages, while acute and chronic myeloid leukemia (AML, CML), myelodysplastic syndrome and myeloproliferative diseases are bone marrow derived.

[0130] In some embodiments, the non-solid (diffuse) cancer or tumor comprises a hematopoietic malignancy, a blood cell cancer, a leukemia, a myelodysplastic syndrome, a lymphoma, multiple myeloma (plasma cell myeloma), acute lymphoblastic leukemia, acute myeloid leukemia, chronic myeloid leukemia, Hodgkin lymphoma, non-Hodgkin lymphoma, or plasma cell leukemia.

[0131] In some embodiments, the tumor or cancer comprises a metastasis of the tumor or cancer.

[0132] In some embodiments, the cell surface TAA is located in or on the plasma membrane of the cell such that at least a portion of the molecule remains accessible from outside the cell in a tertiary form. In some embodiments, the cell surface TAA located in the plasma membrane is a transmembrane protein comprising a hydrophilic region and a hydrophobic region in its tertiary conformation.

[0133] These antigens can be presented on the cell surface along with an extracellular portion, which is often combined with the transmembrane and cytoplasmic portions of the molecule. These antigens are sometimes only presented by tumor cells and never by normal cells. Tumor antigens can be expressed exclusively on tumor cells or can represent tumor-specific mutations compared to normal cells. In this case, they are referred to as tumor-specific antigens. More common are antigens presented by both tumor cells and normal cells. In some embodiments, TAAs include antigens that are expressed exclusively on tumor cells. In some embodiments, TAAs include antigens that are expressed on both tumor cells and normal cells.

[0134] In some embodiments, TAAs may be overexpressed on tumor cells compared to normal cells, or TAAs may be readily accessible to antibody binding in tumor cells due to the less compact structure of tumor tissue compared to normal tissue.

[0135] In some embodiments, the precursor trispecific antibody construct described herein comprises (a) an scFv fragment comprising a first binding domain that binds to TAA (TAA binding domain); (b) an scFv fragment comprising a second binding domain that binds to TAA (TAA binding domain); (c) a Fab fragment comprising a third binding domain that binds to an extracellular epitope of human CD3ε (CD3 binding domain); (d) a first sub-regulatory domain comprising a protease cleavage domain and a half-life extension (HLP) domain; and (e) a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of CD3ε. In some embodiments, the precursor trispecific antibody constructs described herein comprise (a) an scFv fragment comprising a first binding domain that binds to TAA (TAA binding domain); (b) an scFv fragment comprising a second binding domain that binds to TAA (TAA binding domain); (c) a Fab fragment comprising a third binding domain that binds to an extracellular epitope of human CD3ε (CD3 binding domain); (d) a first sub-regulatory domain comprising a protease cleavage domain and a half-life extending (HLP) domain comprising a human serum albumin (HSA) polypeptide; and (e) a second sub-regulatory domain comprising a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to an extracellular epitope of CD3ε.

[0136] Technicians will understand that in some embodiments, precursor antibody constructs include precursors or derivative forms of pharmaceutically active antibodies. In some embodiments, pharmaceutical preparations include precursor antibody constructs. In some embodiments, preparations include precursor antibody constructs. In some embodiments, compared with activated antibodies, precursor antibody constructs have reduced adverse effects. In some embodiments, compared with activated antibodies, precursor antibody constructs have reduced adverse effects, wherein the precursor antibody can be enzymatically activated or converted into the active form of the antibody. In some embodiments, the precursor trispecific antibody construct antibody described herein is a precursor trispecific antibody construct.

[0137] In some embodiments, the precursor antibody construct has an extended half-life compared to an activated antibody. In some embodiments, the precursor antibody construct has an extended half-life compared to an activated antibody, wherein the precursor antibody can be enzymatically activated or converted into an active form of the antibody, and the active form has a reduced half-life compared to the precursor antibody construct.

[0138] In some embodiments, the ability of the precursor antibody construct to bind to T cells is reduced. In certain embodiments, the ability of the precursor antibody construct to activate T cells is reduced compared to an activated antibody. In some embodiments, the ability of the precursor antibody construct to bind to T cells is reduced. In certain embodiments, the ability of the precursor antibody construct to activate T cells is reduced compared to an activated antibody, wherein the precursor antibody can be enzymatically activated or converted into an active form of an antibody.

[0139] In some embodiments, compared with activated antibody, the precursor antibody construct has the ability of the activated T cell of the half-life of prolongation and reduction. In some embodiments, compared with activated antibody, the precursor antibody construct has the ability of the activated T cell of the half-life of prolongation and reduction. In some embodiments, compared with activated antibody, the precursor antibody construct has the ability of the activated T cell of the half-life of prolongation and reduction, and wherein the precursor antibody can be enzymatically activated or converted into the active form of antibody. In some embodiments, compared with activated antibody, the precursor antibody construct has the ability of the activated T cell of the half-life of prolongation and reduction, and wherein the precursor antibody can be enzymatically activated or converted into the active form of antibody.

[0140] In some embodiments, the ability of a precursor antibody to bind to T cells is reduced, wherein the regulatory domain comprising the CAP component is cleaved but the regulatory domain comprising the HLP is not cleaved, wherein the "partially" activated antibody can bind to T cells and maintain an extended half-life. In some embodiments, the binding of a partially activated precursor antibody to T cells is reduced compared to a fully activated antibody, wherein both regulatory arms have been proteolytically cleaved. In some embodiments, the ability of a precursor antibody to activate T cells is reduced, wherein the regulatory domain comprising the CAP component is cleaved but the regulatory domain comprising the HLP is not cleaved, wherein the "partially" activated antibody can activate T cells and maintain an extended half-life. In some embodiments, activation of T cells is reduced upon binding to a partially activated precursor construct, wherein both regulatory arms have been proteolytically cleaved, compared to a fully activated antibody.

[0141] In some embodiments, the precursor antibody construct is synthesized in vitro. In some embodiments, the precursor antibody construct is not converted into the active form of the antibody when the precursor exists in vivo in a non-tumor microenvironment (e.g., in the circulation).

[0142] In some embodiments, in addition to the antigen binding domain, the precursor antibody construct further comprises a plurality of regulatory domains. In some embodiments, in addition to the antigen binding domain, the precursor antibody construct further comprises two regulatory domains. In some embodiments, in addition to the antigen binding domain, the precursor antibody construct further comprises an enzymatically cleavable regulatory domain. In some embodiments, in addition to the antigen binding domain, the precursor antibody construct further comprises a plurality of regulatory domains, wherein a portion of the regulatory domain is enzymatically cleavable. In some embodiments, in addition to the antigen binding domain, the precursor antibody construct further comprises two regulatory domains, wherein a portion of the regulatory domain is enzymatically cleavable. In some embodiments, in addition to the antigen binding domain, the precursor antibody construct further comprises two regulatory domains, wherein a portion of the regulatory domain is enzymatically cleavable. In some embodiments, in addition to the three antigen binding domains, the precursor antibody construct further comprises two regulatory domains, wherein the regulatory domain is enzymatically cleavable.

[0143] In some embodiments, the precursor trispecific antibodies described herein comprise enhanced selectivity for targeting tumor cells over normal cells prior to cytotoxic activation of T cells.

[0144] Immunological binding generally refers to the non-covalent interaction that occurs between an immunoglobulin molecule and an antigen for which the immunoglobulin is specific, for example, as a result of electrostatic, ionic, hydrophilic and / or hydrophobic attractions or repulsions, steric forces, hydrogen bonding, van der Waals forces, and other interactions, by way of illustration and not limitation. The strength or affinity of an immunological binding interaction can be measured using the dissociation constant (K) of the interaction. d ) indicates that the smaller K d=Affinity represents a greater affinity. The immunological binding properties of a selected polypeptide can be quantified using methods known in the art. One such method requires measuring the rates of formation and dissociation of the antigen binding site / antigen complex, where these rates depend on the concentration of the complex partner, the affinity of the interaction, and geometric parameters that also affect the two-way rate. Thus, the "binding rate constant" (k on ) can be calculated from the concentration and the actual association rate and the "dissociation rate constant" (k off ) can be determined from the actual dissociation rate. Therefore, k off / k on The ratio is equal to the dissociation constant K D See generally Davies et al. (1990) Annual Rev. Biochem. 59:439-473.

[0145] The skilled artisan will understand that a "binding domain" or related expressions, such as a domain that "binds" or is "reactive with / toward" a specific target, includes the ability of the domain to distinguish between the respective antigens and to specifically associate with the target antigen. A "binding domain" or "binding region" according to the present disclosure can be, for example, any protein, polypeptide, oligopeptide, or peptide that has the ability to specifically recognize and bind to a biomolecule (e.g., a cell surface receptor or tumor protein, or a component thereof, e.g., an extracellular component thereof). Binding domains include any naturally occurring, synthetic, semisynthetic, or recombinantly produced binding partner for the biomolecule of interest. For example, and as further described herein, the binding domain can be an antibody light chain and heavy chain variable region, or the light chain and heavy chain variable regions can be linked together as a single chain and in either orientation (e.g., VL-VH or VH-VL). A variety of assays are known for identifying binding domains of the present disclosure that specifically bind to a specific target, including Western blot, ELISA, flow cytometry, or surface plasmon resonance analysis (e.g., using BIACORE™ analysis).

[0146] In some embodiments, if a binding domain or portion thereof binds with an affinity or Ka (i.e., the equilibrium association constant for a particular binding interaction, in units of 1 / M) (e.g., greater than or equal to about 10 5 M -1 ) binds to a target molecule or associates with a target molecule, then it "specifically binds" to a target molecule. In certain embodiments, the binding domain or portion thereof is greater than or equal to about 10 6 M -1 , 10 7 M -1 , 10 8 M -1 , 10 9 M -1 , 10 10 M-1 , 10 11 M -1 , 10 12 M -1 or 10 13 M -1 A "high affinity" binding domain may comprise a domain with a Ka of at least 10 7 M -1 , at least 10 8 M -1 , at least 10 9 M -1 , at least 10 10 M -1 , at least 10 11 M -1 , at least 10 12 M -1 , at least 10 13 M -1 Alternatively, affinity can be defined as having units M (e.g., 10 -5 M to 10 - 13 The equilibrium dissociation constant (Kd) of a specific binding interaction of 5-M or less. The affinities of binding domain polypeptides and portions thereof as described herein can be readily determined using conventional techniques (see, e.g., Scatchard et al. (1949) Ann. NY Acad. Sci. 51:660; and U.S. Patent Application Nos. 5,283,173; 5,468,614, or equivalents, which are incorporated herein in their entireties).

[0147] Illustrative binding domains are described herein. In certain embodiments, the target molecule can be a protein expressed on the cell surface, such as a receptor or a tumor antigen. In some embodiments, the target molecule is a tumor-associated antigen (TAA). Exemplary binding domains include immunoglobulin antigen binding domains, such as scFv, scTCR, the extracellular domain of a receptor, a ligand of a cell surface molecule / receptor, or its receptor binding domain and tumor binding protein. In certain embodiments, the antigen binding domain can be scFv, VH, VL, domain antibody variants (dAb), camelid antibodies (VHH), fibronectin 3 domain variants, ankyrin repeat variants and other antigen-specific binding domains derived from other protein scaffolds (Owen, B. (2017) Nat Biotechnol Jul 12: 35 (7): 602-603).

[0148] Thus, in certain embodiments, the binding domain comprises an antibody-derived binding domain, but may be a non-antibody-derived binding domain. An antibody-derived binding domain may be a fragment of an antibody or a genetically modified product of one or more fragments of an antibody that participate in binding to an antigen. Examples include, but are not limited to, complementarity determining regions (CDRs), variable regions (Fv), heavy chain variable regions (VH), light chain variable regions (VL), heavy chains, light chains, single chain variable regions (scFv), Fabs, single domain camelid antibodies (Camelidae VHH), and single domain antibodies (dAbs).

[0149] The present disclosure provides precursor trispecific antibody constructs comprising a first binding domain that binds to a cell surface tumor associated antigen (TAA), such as, but not limited to, a TAA that is an epidermal growth factor receptor (EGFR) antigen; and a second binding domain that binds to a cell surface tumor associated antigen (TAA), such as, but not limited to, a second TAA that is an epidermal growth factor receptor (EGFR) antigen; a third binding domain that binds to an immune effector molecule, such as, but not limited to, an extracellular epitope of the CD3ε chain (CD3ε); and two regulatory domains, such as, but not limited to, one regulatory domain comprising a cleavable half-life extension domain and one regulatory domain comprising a cleavable CAP (masking) domain ( Figure 1 and 2A -2B). In some embodiments, the precursor trispecific antibody construct comprises a first binding domain that binds to a cell surface tumor-associated antigen (TAA), such as, but not limited to, a TAA for an epidermal growth factor receptor (EGFR) antigen; and a second binding domain that binds to a cell surface tumor-associated antigen (TAA), such as, but not limited to, a second TAA for an epidermal growth factor receptor (EGFR) antigen; a third binding domain that binds to an immune effector molecule, such as, but not limited to, an extracellular epitope of the CD3ε chain (CD3ε); and a single regulatory domain comprising a protease cleavage domain, a half-life extension (HLP) domain, and a CAP component that reduces the ability of the third binding domain to bind (such as, but not limited to, when the third domain binds to an extracellular epitope of human CD3ε). Figure 2F ).

[0150] In some embodiments, the precursor trispecific antibody construct comprises a first binding domain that binds to a cell surface tumor associated antigen (TAA), such as, but not limited to, a TAA for an epidermal growth factor receptor (EGFR) antigen; and a second binding domain that binds to a cell surface tumor associated antigen (TAA), such as, but not limited to, a second TAA for an epidermal growth factor receptor (EGFR) antigen; a third binding domain that binds to an immune effector molecule, such as, but not limited to, an extracellular epitope of the CD3ε chain (CD3ε); and a regulatory domain ( Figure 2CIn some embodiments, the precursor trispecific antibody construct comprises a first binding domain that binds to a cell surface tumor-associated antigen (TAA), such as but not limited to a TAA for an epidermal growth factor receptor (EGFR) antigen; and a second binding domain that binds to a cell surface tumor-associated antigen (TAA), such as but not limited to a second TAA for an epidermal growth factor receptor (EGFR) antigen; a third binding domain that binds to an immune effector molecule such as but not limited to an extracellular epitope of the CD3ε chain (CD3ε); and a regulatory domain comprising a cleavable half-life extension domain ( Figure 2D ).

[0151] In some embodiments, the first and second binding domains each comprise a single chain variable fragment (ScFv). The skilled artisan will appreciate that ScFv is not actually an antibody fragment, but rather a fusion polypeptide comprising the variable heavy chain (VH) and variable light chain (VL) regions of an immunoglobulin, connected by a short linker peptide of 10 to about 25 amino acids ( Figure 1 and Figures 2A-2D and 2F).

[0152] In some embodiments, the third binding domain comprises a Fab fragment in which the first binding domain is attached to the C-terminus of the CL chain and the second domain is attached to the C-terminus of the CH1 chain. Alternatively, the third binding domain comprises a Fab fragment in which the second binding domain is attached to the C-terminus of the CH1 chain and the first domain is attached to the C-terminus of the CL chain.

[0153] In some embodiments, the third binding domain comprises a Fab fragment in which a first regulatory domain, e.g., a CAP masking domain, is attached to the N-terminus of the VL chain and a second regulatory domain, e.g., an HSA HLP domain, is attached to the N-terminus of the VH chain. Figure 1 、 2A and 2B). In some embodiments, the third binding domain comprises a Fab fragment in which a first regulatory domain, e.g., a CAP masking domain, is attached to the N-terminus of the VL chain and a second regulatory domain, e.g., an HSA HLP domain, is attached to the N-terminus of the VH chain.

[0154] In some embodiments, the third binding domain comprises a Fab fragment in which a regulatory domain comprising, for example, a CAP masking domain, an HSA HLP domain, and a protease linker is attached to the N-terminus of the VH chain ( Figure 2F In some embodiments, the third binding domain comprises a Fab fragment in which a regulatory domain comprising, for example, a CAP masking domain, an HSA HLP domain, and a protease linker is attached to the N-terminus of the VL chain.

[0155] In some embodiments, a linker sequence may be present between the scFv of the first binding domain and the CL of the third binding domain. In some embodiments, a linker sequence may be present between the scFv of the first binding domain and the CH1 of the third binding domain. In some embodiments, a linker sequence may be present between the scFv of the second binding domain and the CL of the third binding domain. In some embodiments, a linker sequence may be present between the scFv of the second binding domain and the CH1 of the third binding domain. In some embodiments, a linker sequence may be present between the scFv of the first binding domain and the second binding domain and between the CL and CH1 of the third binding domain, respectively.

[0156] In some embodiments, there may be a cleavable linker sequence between the first sub-regulatory domain and the VH chain of the third binding domain. In some embodiments, there may be a cleavable linker sequence between the first sub-regulatory domain and the VL chain of the third binding domain. In some embodiments, there may be a cleavable linker sequence between the second sub-regulatory domain and the VH chain of the third binding domain. In some embodiments, there may be a cleavable linker sequence between the second sub-regulatory domain and the VL chain of the third binding domain. In some embodiments, there may be a cleavable linker sequence between the first sub-regulatory domain and the VH and VL chains of the second sub-regulatory domain and the third binding domain, respectively.

[0157] In some embodiments, there may be a cleavable linker sequence between the single regulatory domain and the VH chain of the third binding domain. In some embodiments, there may be a cleavable linker sequence between the single regulatory domain and the VL chain of the third binding domain.

[0158] These general forms are basic structures that can be constructed to construct the precursor trispecific (trimeric) antibody constructs described herein ( Figure 1 and Figure 2A 、 2B and 2F).

[0159] In some embodiments, the regulatory domain comprises a protease-cleavable linker component and a human serum albumin polypeptide (HSA) sequence component ( Figure 1 、 2A , 2B and 2D). In some embodiments, only a single regulatory domain comprising a protease-cleavable linker component and a human serum albumin polypeptide (HSA) sequence component is present ( Figure 2D In some embodiments, only a single regulatory domain comprising a protease-cleavable linker component, a human serum albumin polypeptide (HSA) sequence component, and a CAP component is present ( Figure 2FIn some embodiments, the regulatory domain comprises a protease-cleavable linker component and one of a human serum albumin polypeptide sequence component and at least one CAP amino acid component ( Figure 1 and 2A -2B). In some embodiments, there are two regulatory domains, one consisting essentially of a protease-cleavable linker component and a human serum albumin polypeptide sequence component, and the other consisting essentially of a protease-cleavable linker and a CAP amino acid (masking) component ( Figure 1 and 2A -2B). In some embodiments, only a single regulatory domain comprising a protease-cleavable linker component and a CAP amino acid (masking) component is present ( Figure 2C In some embodiments, there is a single regulatory domain comprising a CAP amino acid (masking) component, a protease-cleavable linker component, and a human serum albumin polypeptide (HSA) sequence component ( Figure 2F ). The skilled person will appreciate the presence of linkers, e.g. Figure 1 and 2A -F, any linker displayed in the construct shown in the embodiment, will provide flexibility to the polypeptide, while not necessarily providing the necessary regulatory features for the regulatory domain, such as provided by the CAP (masking activity) or HAS component (increasing half-life). In some embodiments, wherein the linker comprises a protease cleavable linker, the linker also provides regulatory function, wherein cleavage of the protease cleavable linker can remove the CAP (masking) component, remove the half-life extension component, or can remove both the CAP (masking) component and the half-life extension component.

[0160] In some embodiments, the third binding domain is an anti-immune effector molecule, such as, but not limited to, an anti-CD3ε chain (CD3ε) extracellular epitope, that specifically binds to the CAP amino acid component. In some embodiments, the CAP component effectively blocks the binding of the precursor trispecific antibody construct to an immune effector target molecule, such as a T cell. In some embodiments, activation of cytotoxicity to the target is specifically masked by the CAP component. In some embodiments, wherein the regulatory domain comprises a cleavable CAP component, activation of cytotoxicity is limited to the tumor environment ( Figures 3A-3B In some embodiments, the source of the first or second binding domain target (e.g., TAA) includes Figure 3B Those shown in (right).

[0161] In some embodiments, the CAP component comprises an amino acid sequence present within a human CD3 epsilon polypeptide chain. In some embodiments, the CAP component comprises an amino acid sequence present as part of the extracellular portion of a human CD3 epsilon chain. In some embodiments, the CAP component comprises an amino acid sequence selected from the amino acid sequence at the N-terminus of a human CD3 epsilon precursor polypeptide. In some embodiments, the CAP component comprises an amino acid sequence selected from the amino acid sequence at the N-terminus of a human CD3 epsilon mature polypeptide.

[0162] The amino acid sequence of precursor human CD3 epsilon is set forth in SEQ ID NO: 1. Human CD3 epsilon is expressed in a precursor form, wherein amino acids 1-21 form a signal peptide. The amino acid sequence of mature human CD3 epsilon is set forth in amino acids 22-207 of SEQ ID NO: 1, as set forth herein in SEQ ID NO: 2. In some embodiments, the extracellular epitope of human CD3 epsilon is located within the precursor sequence, as set forth in SEQ ID NO: 3. In some embodiments, the extracellular epitope of mature human CD3 epsilon is located within amino acids 1-27 of the precursor sequence, as set forth in SEQ ID NO: 4. In some embodiments, the extracellular epitope of human CD3 epsilon is located within amino acids QDGNEEMGGITQTPYKVSISGTTVILT (SEQ ID NO: 5; AA1-27).

[0163] In some embodiments, the amino acid sequence of the CAP module is set forth in SEQ ID NO: 5 or a homolog thereof. In some embodiments, the amino acid sequence of the CAP module is a contiguous sequence selected from SEQ ID NO: 4 or a homolog thereof.

[0164] In some embodiments, SEQ ID NO:5 or a homolog of a CAP sequence selected from SEQ ID NO:4 comprises a polypeptide that is at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99% homologous to the amino acid sequence.

[0165] In some embodiments, the homolog comprises a polypeptide that is at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% homologous to a human CD3 epsilon homologous polypeptide or portion thereof as determined using BlastP software from the National Center for Biotechnology Information (NCBI) using default parameters.

[0166] Technicians will understand that the term "homology" and its grammatical forms include the similarity between two or more structures. The term "homologous sequence" refers to a region with a similar monomer order in a macromolecule. When used in relation to a nucleic acid sequence, the term "homology" refers to the similarity between two or more nucleic acid sequences (e.g., a gene) or a fragment thereof. Typically, the similarity between two or more nucleic acid sequences refers to the similarity of the composition, order, or arrangement of two or more nucleotide bases (or other genotypic characteristics) of two or more nucleic acid sequences. The term "homologous nucleic acids" typically refers to a nucleic acid comprising a nucleotide sequence with similarity in terms of nucleotide base composition, arrangement, or order. Two or more nucleic acids can be identical or different species or groups. When used in relation to a nucleic acid sequence, the term "homology percentage" typically refers to the similarity percentage between the nucleotide sequences of two or more nucleic acids.

[0167] When used in connection with a polypeptide (or protein) sequence, the term "homology" refers to the similarity between two or more polypeptide (or protein) sequences (e.g., genes) or fragments thereof. Typically, the similarity between two or more polypeptide (or protein) sequences refers to the similarity in composition, order, or arrangement of two or more amino acids of the two or more polypeptides (or proteins). Two or more polypeptides (or proteins) may be of the same or different species or groups. When used in connection with a polypeptide (or protein) sequence, the term "homology percentage" typically refers to the similarity percentage between the amino acid sequences of two or more polypeptide (or protein) sequences. The term "homologous polypeptides" or "homologous proteins" typically refers to polypeptides or proteins having similar amino acid sequences and functions, respectively. Such homologous polypeptides or proteins may be related by having similar amino acid sequences and functions, but derived or evolved from different or the same species using the techniques described herein.

[0168] In some embodiments, the homolog comprises a polypeptide that is at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% homologous to a polypeptide disclosed herein, or a portion thereof, as determined using BlastP software from the National Center for Biotechnology Information (NCBI) using default parameters. In some embodiments, a homolog comprises a nucleotide sequence that is at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% homologous to a nucleic acid sequence disclosed herein, or a portion thereof, as determined using BlastN software from the National Center for Biotechnology Information (NCBI) using default parameters.

[0169] In some embodiments, homology also includes deletion, insertion, or substitution variants thereof and their biologically active polypeptide fragments, including amino acid substitutions. In one embodiment, the variant comprises a conservative substitution, or a deletion, insertion, or substitution that does not significantly alter the three-dimensional structure of the polypeptide module of interest described herein. In some embodiments, the deletion, insertion, or substitution does not alter the function of interest of the polypeptide module of interest disclosed herein.

[0170] In some embodiments, homology also includes deletion, insertion or substitution variants thereof and their biologically active polypeptide fragments, including amino acid substitutions. In one embodiment, the variant comprises a conservative substitution, or a deletion, insertion or substitution that does not significantly change the three-dimensional structure of the CAP component, such as a portion of the human CD3ε polypeptide present in the CAP component, particularly in the region of the epitope recognized and bound by the third binding domain. In some embodiments, the deletion, insertion or substitution does not alter the function of interest of the CAP component, in some embodiments, it binds to the third binding domain, or reduces T cell binding, or reduces T cell activation, or any combination thereof.

[0171] In some embodiments, the CAP assembly is 6-110 amino acids in length. In some embodiments, the CAP assembly is between about 6-10 amino acids in length. In some embodiments, the CAP assembly is between about 10-20 amino acids in length. In some embodiments, the CAP assembly is between about 20-30 amino acids in length. In some embodiments, the CAP assembly is between about 20-40 amino acids in length. In some embodiments, the CAP assembly is between about 30-40 amino acids in length. In some embodiments, the CAP assembly is between about 40-60 amino acids in length. In some embodiments, the CAP assembly is between about 60-80 amino acids in length. In some embodiments, the CAP assembly is between about 80-100 amino acids in length. In some embodiments, the CAP assembly is between about 80-110 amino acids in length.

[0172] In some embodiments, the CAP component is 6 amino acids in length. In some embodiments, the CAP component is 7 amino acids in length. In some embodiments, the CAP component is 8 amino acids in length. In some embodiments, the CAP component is 9 amino acids in length. In some embodiments, the CAP component is 10 amino acids in length. In some embodiments, the CAP component is 11 amino acids in length. In some embodiments, the CAP component is 12 amino acids in length. In some embodiments, the CAP component is 13 amino acids in length. In some embodiments, the CAP component is 14 amino acids in length.

[0173] In some embodiments, the CAP component is 15 amino acids in length. In some embodiments, the CAP component is 16 amino acids in length. In some embodiments, the CAP component is 17 amino acids in length. In some embodiments, the CAP component is 18 amino acids in length. In some embodiments, the CAP component is 19 amino acids in length. In some embodiments, the CAP component is 20 amino acids in length. In some embodiments, the CAP component is 21 amino acids in length. In some embodiments, the CAP component is 22 amino acids in length. In some embodiments, the CAP component is 23 amino acids in length. In some embodiments, the CAP component is 24 amino acids in length. In some embodiments, the CAP component is 25 amino acids in length. In some embodiments, the CAP component is 26 amino acids in length. In some embodiments, the CAP component is 27 amino acids in length. In some embodiments, the CAP component is 28 amino acids in length. In some embodiments, the CAP component is 29 amino acids in length. In some embodiments, the CAP component is 30 amino acids in length. In some embodiments, the CAP module is 31, 32, 33, 34, 35, 36, 37, 38, 39, or 40 amino acids in length.

[0174] In some embodiments, the CAP component specifically binds to the third binding region, thereby reducing T cell binding of the precursor construct. In some embodiments, the CAP component specifically binds to the third binding region, thereby inhibiting T cell binding of the precursor construct. In some embodiments, the CAP component specifically binds to the third binding region, thereby reducing T cell activation of the precursor construct. In some embodiments, the CAP component specifically binds to the third binding region, thereby inhibiting T cell activation of the precursor construct.

[0175] In some embodiments, the regulatory domain comprises a cleavable half-life extending domain. In some embodiments, the cleavable half-life extending domain comprises an HSA polypeptide.

[0176] In some embodiments, there are linkers between the components of the regulatory domain. In some embodiments, there is a linker between the regulatory domain and the N-terminus of the VH chain of the Fab fragment. In some embodiments, there is a linker between the regulatory domain and the N-terminus of the VL chain of the Fab fragment. In some embodiments, there is a linker between the regulatory domain and the N-terminus of the VH chain of the Fab fragment, and there is a linker between the regulatory domain and the N-terminus of the VL chain of the Fab fragment. In some embodiments, the linker between the components of the regulatory domain and the N-terminus of the Fab fragment polypeptide is a cleavable linker. In some embodiments, any linker between the components of the regulatory domain and the Fab polypeptide is a cleavable linker. In some embodiments, the linker between the components of the regulatory domain and the Fab polypeptide is non-cleavable ( Figure 1 、 2A -2D and 3A-3B).

[0177] In some embodiments, the regulatory domain comprises a cleavable half-life extension domain comprising a protease cleavable domain and a human serum albumin polypeptide (HSA). In some embodiments, the order of components in the regulatory domain is (N-terminal to C-terminal) HSA-L-protease cleavable domain, where L is a possible linker amino acid sequence ( Figure 1 、 2A , 2B and 2D). In some embodiments, the regulatory domain comprises a cleavage regulatory domain comprising a CAP amino acid (masking) module, a protease cleavable domain, and a human serum albumin polypeptide (HSA). In some embodiments, the order of modules in the regulatory domain is (N-terminal to C-terminal) CAP-L-HSA-L-protease cleavable domain, where L is a possible linker amino acid sequence ( Figure 2F In some embodiments, wherein the protease cleavable domain is at the C-terminus of the HSA polypeptide sequence, the precursor construct has a tunable enhanced half-life, wherein the precursor construct has an enhanced half-life while circulating in vivo and in the absence of a tumor microenvironment.

[0178] In some embodiments, the regulatory domain comprises a cleavable half-life extension domain comprising a protease cleavable domain and a CAP masking domain. In some embodiments, the order of components in the regulatory domain is (N-terminal to C-terminal) CAP-L-protease cleavable domain, where L is a possible linker amino acid sequence ( Figure 1 、 2A, 2B and 2C). In some embodiments, the order of components in the regulatory domain is (N-terminal to C-terminal) CAP-L-HSA-L-protease cleavable domain, where L is a possible linker amino acid sequence. Before the precursor antibody enters the tumor microenvironment, the trispecific precursor construct is effectively prevented from binding to the immune effector target molecule.

[0179] In some embodiments, there is one regulatory domain comprising a cleavable half-life extension domain and a CAP masking domain. In some embodiments, there are two regulatory domains: one comprising a cleavable half-life extension domain and one comprising a cleavable CAP masking domain. In some embodiments, there are three regulatory domains: one comprising a cleavable half-life extension domain, and two comprising a cleavable CAP masking domain. The precursor trispecific construct having an HSA regulatory domain and at least one CAP regulatory domain has an adjustable enhanced half-life, wherein the precursor trispecific antibody construct has an enhanced half-life and is effectively prevented from binding to at least one immune effector target molecule. The half-life may be enhanced in the body circulation and in the absence of a tumor environment. In some embodiments, the activation of cytotoxicity by the precursor trispecific antibody construct is limited to the tumor environment. In some embodiments, the precursor construct maintains an enhanced half-life in the body circulation and is effectively prevented from binding to the immune effector target molecule in the body circulation in a non-tumor environment ( Figures 3A-3B In some embodiments, the activation of cytotoxicity against the target is specifically masked by the CAP component of the precursor construct in the circulation and in the absence of the tumor environment, and the precursor construct comprises an enhanced half-life in the circulation in vivo and in the absence or presence of the tumor environment. In some embodiments, the activation of cytotoxicity is limited to the tumor environment.

[0180] In some embodiments, activation of cytotoxicity against a target is specifically masked by the CAP component in the circulation and in the non-tumor environment. In some embodiments, activation of cytotoxicity is limited to the tumor environment. In some embodiments, activation of T cells is specifically masked by the CAP component.

[0181] In some embodiments, the amino acid sequence of the HSA module is set forth in SEQ ID NO: 6. In some embodiments, the amino acid sequence of the HSA module is set forth in SEQ ID NO: 7.

[0182] In some embodiments, the amino acid sequence of the HSA component is any HSA polypeptide sequence known in the art, or a portion thereof, or a homolog thereof. In some embodiments, the HSA component of the precursor trispecific antibody construct includes, for example, but not limited to, any human albumin sequence disclosed in a known database (e.g., the protein database that is part of the National Center for Biotechnology Information (NCBI) or Swiss-Prot), where the sequence can be specifically identified as human or can be identified as a synthetic construct.

[0183] In some embodiments, the HSA component is encoded by the nucleotide sequence set forth in SEQ ID NO:8.

[0184] In some embodiments, the nucleic acid sequence of the HSA component is any HSA nucleotide sequence known in the art, or a portion thereof, or a homolog thereof. In some embodiments, the HSA component of the precursor trispecific antibody construct comprises a nucleic acid sequence encoding, for example, but not limited to, any human albumin sequence disclosed in known databases (e.g., the protein database that is part of the National Center for Biotechnology Information (NCBI) or Swiss-Prot), where the sequence can be specifically identified as human or can be identified as a synthetic construct.

[0185] In some embodiments, a homolog of an HSA component comprises a polypeptide that is at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% homologous to an amino acid sequence. In some embodiments, a homolog comprises a polypeptide that is at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% homologous to an HSA polypeptide or portion thereof as determined using BlastP software from the National Center for Biotechnology Information (NCBI) using default parameters. In some embodiments, the homolog encoding an HSA component comprises nucleotides that are at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% homologous to a nucleic acid sequence. In some embodiments, the homolog encodes a polypeptide that is at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% homologous to a HSA polypeptide or a portion thereof as determined using BlastP software from the National Center for Biotechnology Information (NCBI) using default parameters.

[0186] In some embodiments, homology also includes deletion, insertion, or substitution variants thereof, including amino acid substitutions, and their biologically active polypeptide fragments. In one embodiment, the variant comprises a conservative substitution, or a deletion, insertion, or substitution that does not significantly alter the three-dimensional structure of the HSA module. In some embodiments, the deletion, insertion, or substitution does not alter the function of interest of the HSA module, and in some embodiments, provides a half-life-extending domain.

[0187] Linear representations of embodiments of the regulatory domains of the precursor trispecific antibody constructs disclosed herein include, but are not limited to (N-terminus to C-terminus)

[0188] (1) CAP-L-protease cleavable domain-L, where L may or may not be present;

[0189] (2) HSA-L-protease cleavable domain-L, where L may or may not be present;

[0190] (3) CAP-L-non-cleavable domain-L, where L may or may not be present;

[0191] (4) HSA-L-non-cleavable domain-L, where L may or may not be present;

[0192] (5) protease-cleavable domain-L-CAP;

[0193] (6) protease cleavable domain-L-HSA;

[0194] (7) non-cleavable domain - L-CAP; and

[0195] (8) Non-cleavable domain-L-HSA.

[0196] In some embodiments, the precursor trispecific antibody constructs disclosed herein comprise precursor constructs with an improved therapeutic window, wherein their limited presence provides the ability to target a wide range of new targets or provide improved activity or a combination thereof, such as, but not limited to, the ability to activate T cells only in the cancer microenvironment and, depending on the cancer type, target cancer-specific TAAs, as well as specific TAAs uniquely expressed by that cancer type in conjunction with proteases produced by that cancer type. In some embodiments, the precursor constructs have the ability to activate T cells only in the TME and target cancer-specific TAAs and different TAAs present in the TME.

[0197] As used herein, the "C-terminus" of a polypeptide, etc., e.g., carboxyl-terminus, carboxyl-terminus, C-terminal tail, C-terminus, or COOH-terminus) is the end of an amino acid chain (protein or polypeptide) that is terminated by a free carboxyl group (-COOH). When a protein is translated from messenger RNA, it is produced from the N-terminus to the C-terminus. The convention for writing peptide sequences is to put the C-terminus on the right and write the sequence from N- to C-terminus. In some embodiments, the C-terminus of a polypeptide includes up to the last amino acid residue of the polypeptide that donates its amine group to form a peptide bond with the carboxyl group of its adjacent amino acid residue.

[0198] As used herein, the "N-terminus" of a polypeptide (e.g., amino-terminus, NH2-terminus, N-terminus, or amine-terminus) is the beginning of a protein or polypeptide and refers to the free amine group (-NH2) at the end of the polypeptide. Typically, the amine group bonds to another carboxyl group in the protein to form a chain, but since only one of the two regions of the protein end is chained, the free amine group refers to the N-terminus. As described above, by convention, peptide sequences are written from N-terminus to C-terminus, from left to right, in LTR languages. This relates the direction of translation to the direction of the text (because when a protein is translated from messenger RNA, it is produced from N-terminus to C-terminus - amino acids are added to the carbonyl end). In some embodiments, the N-terminus of a polypeptide comprises the first amino acid of the polypeptide, which contributes its carboxyl group to form a peptide bond with the amine group of its adjacent amino acid residue.

[0199] The skilled artisan will appreciate that the linker component can include amino acid peptides connected by one or more chemical bonds or indirectly connected by one or more linkers. Any suitable chemical bond can be used to make a direct connection, including but not limited to covalent bonds such as peptide bonds and disulfide bonds, non-covalent bonds such as hydrogen bonds, hydrophobic bonds, ionic bonds and van der Waals bonds.

[0200] As used herein, a "covalent bond" refers to a stable association between two atoms that share one or more electrons. Examples of covalent bonds include, but are not limited to, peptide bonds and disulfide bonds. As used herein, a "peptide bond" refers to a covalent bond formed between the carboxyl group of an amino acid and the amine group of an adjacent amino acid. As used herein, a "disulfide bond" refers to a covalent bond formed between two sulfur atoms. A disulfide bond can be formed by oxidation of two thiol groups. In certain embodiments, the covalent linkage is a direct linkage via a covalent bond. In certain embodiments, the covalent linkage is a direct linkage via a peptide bond or a disulfide bond.

[0201] "Non-covalent bond" refers to an attractive interaction between two molecules or two chemical groups that does not involve electron sharing in this article. Examples of non-covalent bonds include, but are not limited to, hydrogen bonds, hydrophobic bonds, ionic bonds, and van der Waals bonds. "Hydrogen bond" refers to the attraction between the hydrogen atoms of a first molecule / group and the electronegative atoms of a second molecule / group in this article. "Hydrophobic bond" refers to the force that causes hydrophobic or non-polar molecules / groups to aggregate or associate together in an aqueous environment in this article. "Ionic bond" refers to the attraction between cations and anions in this article. "Van der Waals bond" refers to the non-specific attraction between two adjacent molecules / groups with instantaneous random fluctuations in electron distribution in this article. In certain embodiments, covalent attachment is a direct connection through a non-covalent bond. In certain embodiments, covalent attachment is a direct connection through a hydrogen bond, hydrophobic bond, ionic bond, or van der Waals bond.

[0202] The skilled artisan will appreciate that the protease-cleavable domains described herein include joints comprising a protease cleavage site. Therefore, the terms "protease-cleavable domain" and "protease-cleavable joint" are used interchangeably herein and have the same meaning and properties.

[0203] The skilled artisan will understand that the terms "tumor microenvironment," "cancer microenvironment," "TME," and "tumor environment" are used interchangeably, have the same nature and meaning, and encompass the microenvironment in which tumors develop. While the normal cellular microenvironment can inhibit the growth of malignant cells, changes in the tumor microenvironment can synergistically support cell proliferation.

[0204] Tumors shape their microenvironment and support the development of tumor cells and non-malignant cells. The tumor microenvironment affects angiogenesis by interfering with the signaling pathways required for cell recruitment and blood vessel construction. The recruitment of endothelial progenitor cells (EPC) for angiogenesis under hypoxic conditions is also associated with metastasis. In some embodiments, TAAs comprise cell surface antigens associated with angiogenesis. In some embodiments, TAAs are overexpressed by cancer cells. In some embodiments, TAAs are expressed on embryonic cells. In some embodiments, TAAs are expressed on embryonic cells and cancer cells, but not expressed or only minimally expressed on normal adult cells. In some embodiments, TAAs are expressed on solid tumor cells. In some embodiments, TAAs are expressed on non-solid cancer cells. In some embodiments, TAAs are expressed on angiogenic tissue cells.

[0205] In addition, tumor-secreted proteins alter the microenvironment by contributing growth factors and proteases that degrade the extracellular matrix and influence cell motility and adhesion. Stromal cells secrete ECM proteins, cytokines, growth factors, proteases, protease inhibitors, and endoglycosidases, such as heparanase. Matrix metalloproteinases (MMPs) are important secreted proteins closely associated with cancer development. Compared to normal epithelial cells, tumor-associated epithelial cells express MMPs at higher levels. In some embodiments, the microenvironment of a tumor comprises increased protease activity compared to the non-tumor environment.

[0206] Figure 3A and 3B Non-limiting examples of precursor trispecific antibody constructs are provided that may have enhanced half-lives while circulating in vivo and in a non-tumor environment. Further, when the precursor construct is in a non-tumor environment, the anti-CD3 third binding domain of the precursor trispecific antibody construct is blocked and may not interact or bind to target T cells. In some embodiments, as Figure 3A The scissors are shown in Figure 3BPrior to T cell activation in the cancer microenvironment, protease cleavage of the precursor trispecific antibody construct removes the half-life extension regulatory domain (HSA) and the CD3ε CAP regulatory domain, resulting in the presence of an activating EGFR(2scFvs)xCD3ε antibody ( Figure 2E ) and T cell activation.

[0207] In some embodiments, the protease cleavable domain comprises a protease cleavable amino acid sequence (cleavable peptide / cleavable linker; CP) comprising a peptide cleavable by a serine protease, cysteine ​​protease, aspartic protease, or matrix metalloproteinase (MMP) cleavable sequence. In some embodiments, the protease cleavable domain comprises a protease cleavable amino acid sequence (cleavable peptide / cleavable linker; CP) comprising a peptide that is a substrate for cleavage by multiple different proteases. In some embodiments, the protease cleavable domain comprises a protease cleavable amino acid sequence (cleavable peptide / cleavable linker; CP) comprising a peptide that is a substrate for cleavage by MMP2 / MMP9 protease or urokinase-type plasminogen activator (uPA) protease, or a matrices, or an asparagine endopeptidase protease. In some embodiments, serine proteases, cysteine ​​proteases, aspartic proteases, uPA proteases, matrices, asparagine endopeptidase proteases, or matrix metalloproteinases (MMPs) are expressed at higher levels in the tumor microenvironment. In some embodiments, matrix metalloproteinases are expressed at higher levels in the tumor microenvironment.

[0208] In some embodiments, the protease cleavable sequence is a MMP cleavable sequence. In some embodiments, the matrix metalloproteinase cleavable sequence can be a matrix metalloproteinase 1 (MMP-1), matrix metalloproteinase 2 (MMP-2), matrix metalloproteinase 9 (MMP-9) or matrix metalloproteinase 14 (MMP-14) cleavable sequence.

[0209] In some embodiments, the protease cleavable sequence is a uPA (urokinase-type plasminogen activator) cleavable sequence. In some embodiments, the protease cleavable sequence is a MT-SP1 (matrix) cleavable sequence.

[0210] In some embodiments, the protease cleavable sequence is a MMP, uPA, matriptase, and asparagine endopeptidase cleavable sequence.

[0211] In some embodiments, the protease cleavable domain comprises an amino acid sequence that is 1 to 10 amino acids in length. In some embodiments, the protease cleavable domain is 1 to 20 amino acids in length.

[0212] In some embodiments, the protease cleavable domain comprises a protease substrate cleavage sequence, such as, but not limited to, an MMP substrate cleavage sequence. Most MMPs can cleave the well-known PLGLAG peptide sequence (SEQ ID NO: 9) in the substrate. Substrate sequences that can be cleaved by MMPs have been extensively studied. A protease substrate cleavage sequence refers to a peptide sequence that can be cleaved by protease treatment. An MMP substrate sequence refers to a peptide sequence that can be cleaved by incubation with an MMP. SEQ ID NO: 9 is a commonly used MMP substrate cleavage sequence (see, e.g., Jiang, PNAS (2004) 101: 17867-72; Olson, PNAS (2010) 107: 4311-6). In another embodiment, the protease cleavage site is recognized by MMP-2, MMP-9, or a combination thereof. In another embodiment, the protease site comprises a sequence recited in GPLGMLSQ (SEQ ID NO: 10), GPLGLWAQ (SEQ ID NO: 11), GPLGLAG (SEQ ID NO: 12), KKNPAELIGPVD (SEQ ID NO: 13), KKQPAANLVAPED (SEQ ID NO: 14), GPLGIAGQ (SEQ ID NO: 15), or PVGLIG (SEQ ID NO: 16). In some embodiments, the protease cleavage site comprises any protease cleavage site (protease cleavable peptide; CP) known in the art that is sensitive to proteases present in the tumor environment, such as, but not limited to, the protease cleavage sites disclosed in Eckhard, U et al. Matrix Biol. Jan; 49: 37-60.

[0213] In some embodiments, the protease cleavable sequence comprising a uPA cleavable sequence comprises a sequence as recited in NSGRAV (SEQ ID NO: 17), SGRSA (SEQ ID NO: 18), LGGSGRSANAILE (SEQ ID NO: 19), SGRS (SEQ ID NO: 20), GGSGRSANK (SEQ ID NO: 21), LGGSGRSANAILEC (SEQ ID NO: 22), GGGRR (SEQ ID NO: 23), TGRGPS (SEQ ID NO: 24), LSGRSDNH (SEQ ID NO: 25), or PLTGRSGG (SEQ ID NO: 26).

[0214] In some embodiments, the protease cleavable sequence comprising a matrices cleavable sequence comprises a sequence as recited in QRRVVGG (SEQ ID NO: 27), QAR,AANL (SEQ ID NO: 29), PTNL (SEQ ID NO: 30), PTN, or SAN.

[0215] In some embodiments, the cleavable peptide is encoded by the nucleic acid sequence set forth in SEQ ID NO: 33: CCACTGGGCCTGGCCGGC.

[0216] In some embodiments, the amino acid sequence of a protease cleavable sequence that serves as a substrate for a cleavable sequence for MMP2 / 9, uPA, matriptase, and asparagine endopeptidase is set forth in PLGLAGSGRSDNH (SEQ ID NO: 35). In some embodiments, all protease cleavable sequences contained in the precursor construct comprise SEQ ID NO: 35. In some embodiments, at least one of the protease cleavable sequences contained in the precursor construct comprises SEQ ID NO: 35. In some embodiments, at least two of the protease cleavable sequences contained in the precursor construct comprise SEQ ID NO: 35. In some embodiments, at least three of the protease cleavable sequences contained in the precursor construct comprise SEQ ID NO: 35.

[0217] In some embodiments, the sequence of the protease-cleavable peptide module of regulatory domain one is the same as the sequence of the protease-cleavable peptide module of regulatory domain two. In some embodiments, the sequence of the protease-cleavable peptide module of regulatory domain one is different from the sequence of the protease-cleavable peptide module of regulatory domain two. In some embodiments, the protease that cleaves the cleavable peptide module of regulatory domain one is the same as the protease that cleaves the protease-cleavable peptide module of regulatory domain two. In some embodiments, the protease that cleaves the cleavable peptide module of regulatory domain one is different from the protease that cleaves the protease-cleavable peptide module of regulatory domain two.

[0218] In some embodiments, the protease that cleaves the first and second sub-regulatory domains is a MMP protease. In some embodiments, the protease that cleaves the first and second sub-regulatory domains is a uPA protease. In some embodiments, the protease that cleaves the first and second sub-regulatory domains is a matriptase protease. In some embodiments, one of the first or second sub-regulatory domains is cleaved by a MMP protease, while the other regulatory domain is cleaved by a non-MMP protease. In some embodiments, one of the first or second sub-regulatory domains is cleaved by a MMP protease, while the other regulatory domain is cleaved by a uPA protease. In some embodiments, one of the first or second sub-regulatory domains is cleaved by a MMP protease, while the other regulatory domain is cleaved by a matriptase protease. In some embodiments, one of the first or second sub-regulatory domains is cleaved by one MMP protease, while the other regulatory domain is cleaved by another MMP protease.

[0219] A stable linker or protease non-cleavable linker refers to a linker peptide sequence that is not among the known protease substrate sequences and therefore does not result in the formation of significant cleavage products upon incubation with the protease.

[0220] In some embodiments, the cleavage substrate (or cleavage sequence) of the linker can include an amino acid sequence that can serve as a substrate for a protease (typically an extracellular protease). In other embodiments, the cleavage sequence comprises a cysteine-cysteine ​​pair that can form a disulfide bond, which can be cleaved by the action of a reducing agent. In other embodiments, the cleavage sequence comprises a substrate that can be cleaved upon photolysis.

[0221] The cleavage substrate is located within the protease cleavable domain such that when the cleavage substrate is cleaved by a cleavage agent (e.g., a cleavage substrate of a linker is cleaved by a protease and / or a cysteine-cysteine ​​disulfide bond is disrupted via reduction by exposure to a reducing agent) or by light-induced photolysis, in the presence of the target, a cleavage product is generated having various functional properties as described herein. In some embodiments, the cleavage product has a reduced half-life. In some embodiments, the cleavage product has the ability to activate T cells ( Figures 3A-3B ).

[0222] The cleavage substrate of the cleavage domain can be selected based on a protease on the cell surface of the target antigen of interest that is co-localized in the diseased tissue or expresses the binding domain of the fusion portion. A variety of different conditions are known, wherein the target of interest is co-localized with a protease, wherein the substrate of the protease is known in the art. In the example of cancer, the target tissue can be a cancerous tissue, specifically a cancerous tissue of a solid tumor. The literature reports the increased levels of proteases with known substrates in many cancers (e.g., solid tumors). See, for example, [La Rocca et al, (2004) British J. of Cancer 90 (7): 1414-1421. Radisky ES, Front Biosci (Landmark Ed). 2015 Jun 1; 20: 1144-63; Miao C, et al., Oncotarget. 2017 May 9; 8 (19): 32309-32321]. Non-limiting examples of diseases include: all types of cancer (breast cancer, lung cancer, colorectal cancer, prostate cancer, head and neck cancer, pancreatic cancer, etc.), rheumatoid arthritis, Crohn's disease, melanoma, SLE, cardiovascular damage, ischemia, etc. In addition, anti-angiogenic targets, such as VEGF, are known.

[0223] In some embodiments, when the TAA of the first or second binding domain is selected such that it is capable of binding to a tumor antigen, a suitable cleavage substrate sequence for the linker will be a cleavage substrate sequence comprising a peptide substrate cleavable by a protease that is present at the cancer treatment site, i.e., the tumor microenvironment, which is particularly present at elevated levels at the cancer treatment site compared to non-cancerous tissue.

[0224] In some embodiments, the first or second, or both the first and second binding domains of the precursor construct disclosed herein can bind to a TAA, such as EGFR, and the cleavage substrate sequence can be a matrix metalloproteinase (MMP) substrate, and thus can be cleaved by an MMP. In other embodiments, the TAA comprises ROR1 and the cleavage substrate sequence can be a protein cleavage enzyme (MT-SP1, TADG-15, epithein, ST14) substrate, and thus can be cleaved by a protein cleavage enzyme. In other embodiments, the first or second, or first and second binding domains of the precursor construct can bind to a target of interest, and the cleavage substrate present in the cleavage domain can be, for example, asparagine endopeptidase, plasmin, TMPRSS-3 / 4, MMP-9, MT1-MMP, cathepsin, caspase, human neutrophil elastase, β-secretase, uPA, or PSA. In other embodiments, in diseases other than cancer, such as multiple sclerosis or rheumatoid arthritis, the cleavage domain is cleaved by other disease-specific proteases.

[0225] In some embodiments, the precursor trispecific antibody construct can bind to a TAA via the first or second, or both the first and second binding domains, wherein the cleavable domain of the regulatory arm remains uncleaved, and thus the third binding domain of the precursor construct or partially cleaved precursor construct may be specifically unavailable for the target CD3ε antigen due to the presence of the CAP component. In some embodiments, the precursor trispecific antibody construct can bind to a TAA via the first or second, or both the first and second binding domains, wherein the cleavable domain of the regulatory arm remains uncleaved, wherein the precursor construct or partially cleaved precursor construct has an enhanced half-life due to the half-life extension domain (e.g., an HSA polypeptide sequence), and the third binding domain is available or partially available for the target CD3ε antigen. In some embodiments, a precursor trispecific antibody construct can bind to a TAA via either the first or the second, or both the first and second binding domains, wherein the cleavable domains of the two regulatory arms remain uncleaved, wherein the precursor construct has an enhanced half-life due to the half-life extension domain (e.g., an HSA polypeptide sequence), and wherein the third binding domain remains unavailable for specificity for the target CD3ε antigen due to the presence of the CAP component.

[0226] In some embodiments, a linker (L) ( Figure 1 and 2A -2D). In some embodiments, the linker of the precursor construct (e.g., the linker between the VL and VH of a Fab and the regulatory domain) comprises a cleavable domain linker. In some embodiments, the ability to be cleaved is independently selected for each linker. In some embodiments, the ability to be cleaved by a protease is independently selected for each linker. In some embodiments, the linker can be cleaved by a protease. In some embodiments, the linker cannot be cleaved by a protease. In some embodiments, the linker between the CH1 or CL of a Fab and the ScFv of the first binding domain comprises a non-cleavable linker. In some embodiments, the linker between the CH1 or CL of a Fab and the ScFv of the second binding domain comprises a non-cleavable linker.

[0227] The skilled artisan will appreciate that in some embodiments, a linker comprises a spacer region between two active modules or between two regions of an active module.

[0228] A skilled artisan will appreciate that a cleavable domain comprises a linear amino acid sequence that contains an enzymatic cleavage site and may in certain embodiments be referred to as a "cleavable linker" or "linker" or "cleavable peptide" or "CP," wherein the linkers disclosed herein may be cleavable or non-cleavable.

[0229] In some embodiments, the linker is present at the C-terminus of the constant heavy chain (CH1) of the Fab fragment. In some embodiments, the linker is present at the C-terminus of the constant light chain (CL) of the Fab fragment. In some embodiments, the C-terminal linker of CH1 is cleavable. In some embodiments, the C-terminal linker of CH1 is non-cleavable. In some embodiments, the C-terminal linker of CL is cleavable. In some embodiments, the C-terminal linker of CL is non-cleavable.

[0230] In some embodiments, the linker is a single amino acid. In some embodiments, the cleavable linker comprises an amino acid sequence recited in any one of SEQ ID NOs: 9-32. In some embodiments, the cleavable linker is encoded by a nucleic acid sequence recited in SEQ ID NO: 33. In some embodiments, the cleavable linker is encoded by a nucleic acid sequence recited in SEQ ID NO: 35.

[0231] In some embodiments, the non-cleavable linker comprises the amino acid sequence set forth in SEQ ID NO: 162. In some embodiments, the non-cleavable linker is encoded by the nucleic acid sequence set forth in SEQ ID NO: 163.

[0232] For the specific cleavage of enzyme proteases, contact is made between the enzyme and the cleavage substrate. When there is enough enzymatic activity in a precursor construct comprising the first and second binding domains in conjunction with the extracellular epitope of CD3ε and two regulatory domains comprising a cleavable linker, the cleavable domain can be cleaved. Sufficient enzymatic activity can refer to the ability of the enzyme to contact the protease cleavable domain with a cleavage site and achieve cleavage. In some embodiments, the enzyme may be near the precursor construct but cannot be cleaved due to protein modifications of other cellular factors or the enzyme.

[0233] In some embodiments, the cleavable domain substrate can include, but is not limited to, a substrate that can be cleaved by one or more of the following enzymes or proteases: ADAM10; Caspase 8, Cathepsin, MMP 8, ADAM12, Caspase 9, FAP, MMP 9, ADAM17, Caspase 10, Granzyme B, MMP 13, ADAMTS, Caspase 11, Guanidinobenzotase (GB), MMP 14, ADAMTS5, Caspase 12, Hepsin, MT-SP1, BACE, Caspase 13, Human Neutrophil Elastase Neprilysin (HNE), Caspase, Caspase 14, Asparagine Endopeptidase, NS3 / 4A, Caspase 1, Cathepsin, Matriptase 2, Plasmin, Caspase 2, Cathepsin A, Penetrans, PSA, Caspase 3, Cathepsin B, MMP 1, PSMA, Caspase 4, Cathepsin D, MMP 2, TACE, caspase 5, cathepsin E, MMP 3, TMPRSS 3 / 4, caspase 6, cathepsin K, MMP 7, uPA, caspase 7, matriptase (MT-SP1, TADG-15, epithelin, ST14) and MT1-MMP.

[0234] In another embodiment, the cleavage substrate may involve a disulfide bond of a cysteine ​​pair, which can therefore be cleaved by a reducing agent, such as, but not limited to, cellular reducing agents, such as glutathione (GSH), thioredoxin, NADPH, flavin, ascorbic acid, etc., which may be present in large quantities in the tissues of or around solid tumors.

[0235] Other suitable protease cleavage sites for use in the cleavable linkers herein are known in the art or can be identified using the methods described, for example, by Turk et al., 2001 Nature Biotechnology 19, 661-667.

[0236] In some embodiments, the first binding domain, the second binding domain, and the third binding domain of the precursor trispecific antibody construct can all bind to their respective human and non-chimpanzee primate target molecules. Thus, the first binding domain and / or the second binding domain binds to a human cell surface tumor-associated antigen (TAA) and a corresponding homolog of a cell surface TAA in a non-chimpanzee primate. The identification and determination of human cell surface TAA homologs in non-chimpanzee primates are well known to those skilled in the art and can be performed, for example, by sequence alignment. The third binding domain can bind to an antigen comprising an extracellular epitope of human CD3ε and can bind to a corresponding homolog of CD3ε in a non-chimpanzee primate. In some embodiments, the first or second or third binding domain, or any combination thereof, also binds to their respective chimpanzee target molecules.

[0237] As will be appreciated by those skilled in the art, in some embodiments, a cell surface tumor associated antigen (TAA) comprises a molecule displayed on the surface of a cell. In some embodiments, the cell is a tumor cell. In some embodiments, the cell is a non-tumor cell present in a tumor environment, such as, but not limited to, a cell present in vasculature tissue associated with a tumor or cancer.

[0238] As will be understood by those skilled in the art, the terms "antigen" or "immunogen" include peptides, proteins, polypeptides that are immunogenic. In some embodiments, an antigen is capable of eliciting an immune response in a mammal and therefore contains at least one and may contain multiple epitopes. An "antigen" molecule or portion of a molecule is capable of being bound by a selective binding agent, such as the antigen binding portion of a Fab fragment or the antigen binding portion of a scFv fragment. In addition, an "antigen" can be administered to an animal to produce antibodies that are capable of binding to an epitope of the antigen. In some embodiments, the CAP assembly comprises a portion of the antigen to which the second binding domain binds.

[0239] In certain embodiments, the term "epitope" includes any determinant capable of specific binding to a TAA, an immunoglobulin, or a T cell receptor. An epitope is a region of an antigen that is bound by an antibody or antigen-binding fragment thereof. In some embodiments, the CAP component comprises an epitope that is bound by a third binding domain.

[0240] In certain embodiments, epitope determinants include chemically active surface groupings of molecules, such as amino acids, sugar side chains, phosphoryl or sulfonyl groups, and in certain embodiments may have specific three-dimensional structural characteristics and / or specific charge characteristics. In certain embodiments, when a precursor trispecific antibody construct preferentially recognizes a target antigen in a complex mixture of proteins and / or macromolecules, the precursor trispecific antibody construct is said to specifically bind to the antigen. When the equilibrium dissociation constant is ≤10 -5 , 10 -6 or 10-7 When M, the precursor trispecific antibody construct is said to specifically bind to the antigen. In some embodiments, the equilibrium dissociation constant can be ≤10 -8 M or 10 -9 M. In some further embodiments, the equilibrium dissociation constant may be ≤10 -10 M or 10 -11 M. Antigens disclosed herein include, but are not limited to, TAAs, CAP components, and immune effector molecules, such as human CD3ε polypeptide.

[0241] In some embodiments, the tumor associated antigen (TAA) is a tumor antigen. In some embodiments, the tumor antigen comprises those antigens presented on tumor cells. In some embodiments, the tumor antigen is present on solid tumor cells. In some embodiments, the tumor antigen is a cancer antigen present on non-solid tumor cells.

[0242] In some embodiments, when the TAA is a tumor cell antigen, the tumor cells include cells from solid tumors. Solid tumors may be benign (not cancer) or malignant (cancer). Different types of solid tumors are named after the cell types that form them. Examples of solid tumors are sarcomas, carcinomas, and lymphomas. In some embodiments, solid tumors are neoplasms (new growths of cells) or lesions (damage to anatomical structures or disorders of physiological functions) formed by abnormal growth of body tissue cells other than blood, bone marrow, or lymphocytes. In some embodiments, solid tumors are composed of abnormal cell populations that may be derived from different tissue types, such as the liver, colon, breast, or lungs, and initially grow in the organs from which their cells originate. However, these cancers may spread to other organs through metastatic tumor growth in the late stages of the disease.

[0243] In some embodiments, the solid tumor comprises a sarcoma or carcinoma, a fibrosarcoma, a myxosarcoma, a liposarcoma, a chondrosarcoma, an osteosarcoma, a chordoma, an angiosarcoma, an endotheliosarcoma, a lymphangiosarcoma, a lymphangioendotheliosarcoma, a synovioma, a mesothelioma, an Ewing sarcoma, a leiomyosarcoma, a rhabdomyosarcoma, a colon cancer, a pancreatic cancer or tumor, a breast cancer or tumor, an ovarian cancer or tumor, a prostate cancer or tumor, a squamous cell carcinoma, a basal cell carcinoma, an adenocarcinoma, a sweat gland carcinoma, a sebaceous gland carcinoma, a papillary carcinoma, a papillary adenocarcinoma, a cystic gland, or a cystic gland. In some embodiments, the solid tumor comprises adrenal cortex tumor (adenoma and carcinoma), carcinoma, colorectal cancer, desmoid tumor, desmoid small round cell tumor, endocrine tumor, Ewing sarcoma, germ cell tumor, hepatoblastoma, hepatocellular carcinoma, melanoma, neuroblastoma, osteosarcoma, retinoblastoma, rhabdomyosarcoma, soft tissue sarcoma other than rhabdomyosarcoma and Wilms' tumor. In some embodiments, the solid tumor is a breast tumor. In another embodiment, the solid tumor is prostate cancer. In another embodiment, the solid tumor is colon cancer. In some embodiments, the tumor is a brain tumor. In another embodiment, the tumor is a pancreatic tumor. In another embodiment, the tumor is a colorectal tumor.

[0244] In some embodiments, the tumor cells include cells from a non-solid tumor, i.e., a non-solid cancer. In some embodiments, the cancer may be a diffuse cancer, in which the cancer spreads widely; it is not localized or restricted. In some embodiments, diffuse cancers may include non-solid tumors. Examples of diffuse cancers include leukemias. Leukemias include cancers that begin in hematopoietic tissue (e.g., bone marrow) and cause large numbers of abnormal blood cells to be produced and enter the blood.

[0245] In some embodiments, the diffuse cancer comprises a B cell malignancy. In some embodiments, the diffuse cancer comprises a leukemia. In some embodiments, the cancer is a lymphoma. In some embodiments, the lymphoma is a large B cell lymphoma.

[0246] In some embodiments, diffuse cancer or tumor comprises hematologic malignancy.In some embodiments, hematologic malignancy is the type of cancer that affects blood, bone marrow and lymph nodes.Hematologic malignancy may be derived from any of two major blood cell lineages: myeloid and lymphoid cell lineages.Myeloid cell lineages produce granulocytes, erythrocytes, platelets, macrophages and mast cells conventionally, while lymphoid cell lineages produce B, T and plasma cells.Lymphoma (such as Hodgkin's lymphoma), lymphocytic leukemia and myeloma are derived from lymphoid lineages, while acute and chronic myeloid leukemia (AML, CML), myelodysplastic syndrome and myeloproliferative diseases are bone marrow derived.

[0247] In some embodiments, the non-solid (diffuse) cancer or tumor comprises a hematopoietic malignancy, a blood cell cancer, a leukemia, a myelodysplastic syndrome, a lymphoma, multiple myeloma (plasma cell myeloma), acute lymphoblastic leukemia, acute myeloid leukemia, chronic myeloid leukemia, Hodgkin lymphoma, non-Hodgkin lymphoma, or plasma cell leukemia.

[0248] In some embodiments, the tumor or cancer comprises a metastasis of the tumor or cancer.

[0249] In some embodiments, the cell surface TAA is located in or on the plasma membrane of the cell such that at least a portion of the molecule remains accessible from outside the cell in a tertiary form. In some embodiments, the cell surface TAA located in the plasma membrane is a transmembrane protein comprising a hydrophilic region and a hydrophobic region in its tertiary conformation.

[0250] These antigens can be presented on the cell surface along with an extracellular portion, which is often combined with the transmembrane and cytoplasmic portions of the molecule. These antigens are sometimes only presented by tumor cells and never by normal cells. Tumor antigens can be expressed exclusively on tumor cells or can represent tumor-specific mutations compared to normal cells. In this case, they are referred to as tumor-specific antigens. More common are antigens presented by both tumor cells and normal cells. In some embodiments, TAAs include antigens that are expressed exclusively on tumor cells. In some embodiments, TAAs include antigens that are expressed on both tumor cells and normal cells.

[0251] In some embodiments, TAAs may be overexpressed on tumor cells compared to normal cells, or TAAs may be readily accessible to antibody binding in tumor cells due to the less compact structure of tumor tissue compared to normal tissue.

[0252] In some embodiments, the first binding domain or the second binding domain or both of the binding cell surface TAA comprises an amino acid sequence that binds to human TAA. In some embodiments, the anti-scFv comprises a heavy chain variable region and a light chain variable region, wherein each region further comprises a complementary determining region (CDR). In some embodiments, the first binding domain or the second binding domain or both of the binding cell surface TAA are included in a joint between the scFv variable light chain (VH) region and the scFv variable heavy chain (VH) region. In some embodiments, the first binding domain or the second binding domain of the binding cell surface TAA comprises a joint between the C-terminus of the scFv fragment and the CL region of the third binding domain. In some embodiments, the first binding domain or the second binding domain of the binding cell surface TAA comprises a joint between the C-terminus of the scFv fragment and the CH1 region of the third binding domain. In some embodiments, the first binding domain of the binding cell surface TAA comprises a joint between the C-terminus of the scFv fragment and the CH1 region of the third binding domain, and the second binding domain of the binding cell surface TAA comprises a joint between the C-terminus of the scFv fragment and the CL region of the third binding domain. In some embodiments, the first binding domain that binds to a cell surface TAA comprises a linker between the scFv fragment and the C-terminus of the CL region of the third binding domain, and the second binding domain that binds to a cell surface TAA comprises a linker between the scFv fragment and the C-terminus of the CH1 region of the third binding domain.

[0253] Figures 2A-2FEmbodiments are shown in which the TAA is EGFR (the first and second binding domains comprise anti-EGFR scFv). The skilled artisan will appreciate that in other embodiments, the TAA may be a TAA known in the art, such as, but not limited to, a TAA comprising the following: FcγRI, FcγRIIa, FcγRIIb, FcγRIIIb, CD28, CD137, CTLA-4, FAS, fibroblast growth factor receptor 1 (FGFR1), FGFR2, FGFR3, FGFR4, glucocorticoid-induced TNFR-related (GITR) protein, lymphotoxin-β receptor (LTβR), toll-like receptor (TLR), tumor necrosis factor-related apoptosis-inducing ligand-receptor 1 (TRAIL receptor 1) and TRAIL receptor 2, prostate-specific membrane antigen (PSMA) protein, prostate stem cell antigen (PSCA) protein, tumor-associated protein carbonic anhydrase IX (CAIX), epidermal growth factor receptor 1 (EGFR1), EGFRvIII, human epidermal growth factor receptor 2 (Her2 / neu;ErbB2), ErbB3 (also known as HER3), folate receptor, ephrin receptor, PDGFRa, ErbB-2, CD20, CD22, CD30, CD33, CD40, CD37, CD38, CD70, CD74, CD40), CD80, CD86, CD2, p53, cMet (also known as tyrosine protein kinase Met or hepatocyte growth factor receptor (HGFR)), MAGE-A1, MAGE-A2, MAGE-A3, MAGE-A4, MAGE-A6, MAGE-A10, M AGE-A12, BAGE, DAM-6, DAM-10, GAGE-1, GAGE-2, GAGE-8, GAGE-3, GAGE-4, GAGE-5, GAGE-6, GAGE-7B, NA88-A, NY-ESO-1, BRCA1, BRCA2, MART-1, MC1R, Gp100, PSA, PSM, PSMA, tyrosinase, Wilms' tumor antigen (WT1), TRP-1, TRP-2, ART-4, CAMEL, Cyp-B, hTERT, hTRT, iCE, MUC1, MUC2, P-cadherin, myostatin (GDF8), Cripto (TDGF1), MUC5AC, PRAME, P15, RU1, RU2, SART-1, SART-3, WT1, AFP, β-catenin / m, caspase-8 / m, CDK-4 / m, ELF2M, GnT-V, G250, HSP70-2M, HST-2, KIAA0205, MUM-1, MUM-2, MUM-3, myosin / m, RAGE, SART-2, TR P-2 / INT2, 707-AP, Annexin II, CDC27 / m, TPI / mbcr-abl, ETV6 / AML, LDLR / FUT, Pml / RARα, TEL / AML1, CD28, CD137, CanAg, Mesothelin, DR5, PD-1, PD1L, IGF-1R, CXCR4, Neuropilin 1, Glypican, EphA2, CD138, B7-H3, B7-H4, gpA33, GPC3, SSTR2, ROR1, 5T4, or VEGF-R2. In some embodiments, the TAA comprises PSMA, CD30, B7-H3, B7-H4, gpA33, HER2, P-cadherin, gp100, DR5, GPC3, SSTR2, Mesothelin, ROR1, 5T4, folate receptor, or EGFR. In some embodiments, the TAA comprises EGFR. In some embodiments, the TAA comprises ROR1. In some embodiments, the TAA comprises PSMA. In some embodiments, the TAA comprises 5T4.

[0254] Trispecific antibodies can be designed to bind to at least one tumor-associated antigen (TAA), which in some embodiments comprises a tumor cell surface antigen, a T cell antigen, and a second TAA, with the goal that the antibody binds to and kills tumor cells (more selectively than normal cells), and ultimately improves the efficacy and safety of monospecific agents, wherein the trispecific antibody comprises a binding domain that binds to at least one cell surface tumor-associated antigen (TAA), a binding domain that binds to a second TAA, and a binding domain that binds to an extracellular epitope of a T cell. However, such trispecific antibodies cannot regulate the order of binding and may therefore bind to T cells before or in the absence of binding to the TAA, wherein the cytotoxicity provided by the activated T cells may actually lead to harmful side effects by nonspecifically causing the death of non-tumor cells. In some embodiments, the TAA comprises a human antigen.

[0255] In some embodiments, the TME antigen comprises a KIR, LILR, or TIGIT antigen.

[0256] In some embodiments, the first binding domain or the second binding domain, or both, that binds to a cell surface TAA binds to a polypeptide target that, in some embodiments, is associated with one or more specific cancers or disease conditions, such as, but not limited to, TAAs comprising the following: FcγRI, FcγRIIa, FcγRIIb, FcγRIIIb, CD28, CD137, CTLA-4, FAS, fibroblast growth factor receptor 1 (FGFR1), FGFR2, FGFR3, FGFR4, glucocorticoid-induced TNFR-related (GITR) protein, lymphotoxin-β receptor (LTβR), toll-like receptors (TLRs), tumor necrosis factor-related apoptosis-inducing ligand-receptor 1 (TRAIL receptor 1; various malignancies, including ovarian and colorectal cancers) and TRAIL receptor 2, prostate-specific membrane antigen (PSMA; prostate cancer) protein, prostate stem cell antigen (PSCA) protein (prostate adenocarcinoma), CA125 (various cancers, including ovarian cancer), tumor-associated protein carbonic anhydrase IX (C AIX; various cancers, including renal cell carcinoma), epidermal growth factor receptor 1 (EGFR1; epithelial malignancies), EGFR (non-small cell lung cancer, epithelial ovarian cancer, colorectal cancer, head and neck cancer, breast cancer, lung cancer, esophageal cancer), EGFRvIII, human epidermal growth factor receptor 2 (Her2 / neu; Erb2; epithelial malignancies), ErbB3 (also known as HER3 (epithelial malignancies)), folate receptor, ephrin receptor, PDGFRa (epithelial malignancies), ErbB-2, CD 20 (B cells, autoimmune, allergic, or malignant), CD22 (B cells, autoimmune or malignant), CD30 (B cell malignancies), CD33 (myeloid malignancies), CD40, CD37, CD38, CD70 (B cells, autoimmune, allergic, or malignant), CD74 (B cells, autoimmune, allergic, or malignant), CD40 (B cells, autoimmune, allergic, or malignant); CD80 (B cells, autoimmune, allergic, or malignant), CD86 (B cells,autoimmune, allergic, or malignant), CD2 (T cells), p53, cMet (also known as tyrosine protein kinase Met or hepatocyte growth factor receptor (HGFR; gastrointestinal and liver malignancies)), MAGE-A1, MAGE-A2, MAGE-A3, MAGE-A4, MAGE-A6, MAGE-A10, MAGE-A12, BAGE, DAM-6, DAM-10, GAGE-1, GAGE-2, GAGE-8, GAGE-3, GAGE-4, GAGE-5, GAGE-6, GAGE-7B, NA88-A, NY-ESO-1, BRCA1, BRCA2, MART-1, MC1R, Gp100, PSA, PSM, tyrosinase, Wilms tumor antigen (WT1), TRP-1, TRP-2, ART-4, CAMEL, Cyp-B, hTERT, hTRT, iCE, MUC1 (epithelial malignancies), MUC2, P-cadherin (epithelial malignancies, including breast cancer), myostatin (GDF8) (various tumors, including sarcoma, ovarian cancer and pancreatic cancer), Cripto (TDGF1) (epithelial malignancies, including colon cancer, breast cancer, lung cancer, ovarian cancer and pancreatic cancer), ACVRL1 / ALK1 (various malignancies, including leukemia and lymphoma), MUC5AC (epithelial malignancies, including breast adenocarcinoma), PRAME, P15, RU1, RU2, SART-1, SART-3, WT1, AFP, β-catenin / m, caspase-8 / m, CDK-4 / m, ELF2M, GnT-V, G250, HSP70-2M, HST-2, KIAA0205, MUM-1, MUM-2, MUM-3, myosin / m, RAGE, SART-2, TRP-2 / INT2, 707-AP, Annexin II, CDC27 / m, TPI / mbcr-abl, ETV6 / AML, LDLR / FUT, Pml / RARα, TEL / AML1, CD28, CD137 (B cells or T cells, autoimmune, allergic, or malignant), CanAg (tumors, such as colon and pancreatic cancer), Mesothelin (many tumors, including mesothelioma and ovarian and pancreatic cancer), DR5 (multiple malignancies, including ovarian and colorectal cancer), PD-1 (B cells, autoimmune, allergic, or malignant), PD1L (multiple malignancies, including epithelial adenocarcinoma), IGF-1R (most malignancies, including epithelial adenocarcinoma), CXCR4 (B cells or T cells, autoimmune, allergic, or malignant), Neuropilin 1 (epithelial malignancies, including lung cancer), Glypican (multiple cancers, including liver, brain, and breast cancer), EphA2 (multiple cancers,Examples of TAAs that are expressed include neuroblastoma, melanoma, breast cancer, and small cell lung cancer), CD138 (myeloma), B7-H3 (CSCs, stroma, NSCLC, bladder tumors, mesothelioma, melanoma), gpA33 (colorectal cancer), GPC3 (liver cancer, lung cancer, esophageal cancer, gastric cancer, head and neck cancer), SSTR2 (neuroendocrine tumors, GIST), ROR1 (hematologic cancers, pancreatic cancer, ovarian cancer, renal cell carcinoma, NSCLC, and triple-negative breast cancer), 5T4 (mesothelioma, gastric cancer, ovarian cancer, renal cancer, cancer stem cells in NSCLC, head and neck cancer), or VEGF-R2 (vasculature associated with most malignancies, including epithelial adenocarcinomas). Examples of deleterious target cells or cancer cells associated with the presented TAAs are included in parentheses in italics.

[0257] In some embodiments, the TAA is selected from EGFR, ROR1, PSMA, and 5T4. In some embodiments, the first binding domain comprises a scFv that binds to human EGFR (anti-hEGFR), or human ROR1 (anti-ROR1), or human PSMA (anti-PSMA), or human 5T4 (anti-5T4).

[0258] In some embodiments, the TAA is EGFR. In some embodiments, the first binding domain comprises a scFv that binds to human EGFR (anti-hEGFR). In some embodiments, the amino acid sequence of the anti-hEGFR-scFv light chain variable region (VL) is set forth in SEQ ID NO: 34. In some embodiments, the anti-hEGFR scFv VL sequence comprises a homolog of SEQ ID NO: 34.

[0259] In some embodiments, the anti-hEGFR-scFv light chain variable region (VL) is encoded by the nucleic acid sequence set forth in SEQ ID NO: 36. In some embodiments, the anti-hEGFR-scFv light chain variable region (VL) is encoded by a homolog of the nucleic acid sequence set forth in SEQ ID NO: 36.

[0260] In some embodiments, the amino acid sequence of the heavy chain variable region (VH) of the anti-hEGFR-scFv is set forth in SEQ ID NO: 37. In some embodiments, the anti-hEGFR scFv VH sequence comprises a homolog of SEQ ID NO: 37.

[0261] In some embodiments, the anti-hEGFR-scFv heavy chain variable region (VH) is encoded by the nucleic acid sequence set forth in SEQ ID NO: 38. In some embodiments, the anti-hEGFR-scFv heavy chain variable region (VH1) is encoded by a homolog of the nucleic acid sequence set forth in SEQ ID NO: 38.

[0262] In some embodiments, the anti-EGFR scFV comprises a linker between the VL and VH regions. In some embodiments, the linker between the VL and VH regions comprises any linker disclosed herein. In some embodiments, the amino acid sequence of the linker between the VL and VH regions of the anti-EGFR scFV is set forth in SEQ ID NO: 39. In some embodiments, the linker between the VL and VH regions of the anti-EGFR scFV comprises a homolog of SEQ ID NO: 39. In some embodiments, the linker between the VL and VH regions of the anti-EGFR scFV is encoded by the nucleic acid sequence set forth in SEQ ID NO: 40. In some embodiments, the linker between the VL and VH regions of the anti-EGFR scFV is encoded by a homolog of the nucleic acid sequence set forth in SEQ ID NO: 40.

[0263] In some embodiments, the components of the anti-EGFR scFv comprise a VL-linker-VH sequence (N-terminus to C-terminus) ( Figure 8A and 9A In some embodiments, the components of the anti-EGFR scFv comprise a VH-linker-VL sequence (N-terminus to C-terminus) ( Figure 8B and 9B ).

[0264] In some embodiments, the anti-EGFR scFv sequence comprising a linker sequence comprises the sequence of SEQ ID NO: 41. In some embodiments, the anti-EGFR scFv comprising a linker sequence comprises a homolog of SEQ ID NO: 41.

[0265] In some embodiments, the anti-EGFR scFv sequence comprising a linker sequence comprises the sequence of SEQ ID NO: 42. In some embodiments, the anti-EGFR scFv sequence comprising a linker sequence comprises a homolog of SEQ ID NO: 42.

[0266] In some embodiments, the amino acid sequence of the light chain variable region of anti-hROR1-scFv, or anti-PSMA-scFv, or anti-5T4-scFv is described in Table 1 below:

[0267] Table 1: Amino acid sequences and optimized nucleotide sequences encoding anti-hROR1-scFv, anti-PSMA-scFv, or anti-5T4-scFv

[0268] Antigen binding (anti-antigen) SEQ ID NO: ROR1(VL-VH) 156 ROR1(VL-VH) 157 ROR1(VH-VL) 166 ROR1(VH-VL) 167 PSMA(VL-VH) 168 PSMA(VL-VH) 169 PSMA(VH-VL) 170 PSMA(VH-VL) 171 5T4(VL-VH) 172 5T4(VL-VH) 173 5T4(VH-VL) 174 5T4(VH-VL) 175

[0269] In some embodiments, the homolog comprises a polypeptide that is at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% homologous to the amino acid sequence of an anti-EGFR scFv. In some embodiments, the homolog comprises a polypeptide that is at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequence of an anti-EGFR scFv, or any anti-ROR1 scFv, or anti-PSMA scFv, or anti-5T4 scFv.

[0270] In some embodiments, the nucleotide sequence encoding the anti-EGFR scFV comprising a linker sequence comprises the sequence of SEQ ID NO: 43. In some embodiments, the anti-EGFR scFv comprising a linker sequence comprises a homolog of SEQ ID NO: 43.

[0271] In some embodiments, the nucleotide sequence encoding the anti-EGFR scFV comprising a linker sequence comprises the sequence of SEQ ID NO: 44. In some embodiments, the anti-EGFR scFv comprising a linker sequence comprises a homolog of SEQ ID NO: 44.

[0272] In some embodiments, the homolog comprises a polypeptide that is at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% homologous to the nucleic acid sequence of an anti-EGFR scFv. In some embodiments, the homolog comprises nucleotides that are at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the nucleic acid sequence of an anti-EGFR scFv, or any anti-ROR1 scFv, or an anti-PSMA scFv, or an anti-5T4 scFv.

[0273] In some embodiments, disclosed herein are homologs of anti-hEGFR scFv VL (SEQ ID NO: 34 or SEQ ID NO: 35), or anti-hEGFR scFv VH (SEQ ID NO: 37), or anti-hEGFR scFv (SEQ ID NO: 41), or anti-hEGFR scFv (SEQ ID NO: 42), respectively, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) using default parameters. In some embodiments, disclosed herein are homologs of the nucleotide sequence encoding anti-hEGFR scFv VL (SEQ ID NO: 36), or anti-hEGFR scFv VH (SEQ ID NO: 38), or anti-hEGFR scFv (SEQ ID NO: 43), or anti-hEGFR scFv (SEQ ID NO: 44), respectively, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) using default parameters.

[0274] In some embodiments, disclosed herein are homologs of anti-hROR1 scFv VL-VH (SEQ ID NO: 156) or anti-hROR1 scFv VH-VL (SEQ ID NO: 169), as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) using default parameters. In some embodiments, disclosed herein are homologs of the nucleotide sequence encoding anti-hROR1 scFv VL-VH (SEQ ID NO: 157) or anti-hROR1 scFv VH-VL (SEQ ID NO: 167), as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) using default parameters.

[0275] In some embodiments, disclosed herein are homologs of anti-hPSMA scFv VL-VH (SEQ ID NO: 168) or anti-hPSMA scFv VH-VL (SEQ ID NO: 170), as determined using the BlastP software from the National Center for Biotechnology Information (NCBI) using default parameters. In some embodiments, disclosed herein are homologs of the nucleotide sequence encoding anti-hPSMA scFv VL-VH (SEQ ID NO: 169) or anti-hPSMA scFv VH-VL (SEQ ID NO: 171), as determined using the BlastP software from the National Center for Biotechnology Information (NCBI) using default parameters.

[0276] In some embodiments, disclosed herein are homologs of anti-h5T4scFv VL-VH (SEQ ID NO: 172) or anti-h5T4scFv VH-VL (SEQ ID NO: 174), as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) using default parameters. In some embodiments, disclosed herein are homologs of the nucleotide sequence encoding anti-h5T4scFv VL-VH (SEQ ID NO: 173) or anti-h5T4 scFv VH-VL (SEQ ID NO: 174), as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) using default parameters.

[0277] In some embodiments, homology also includes deletion, insertion, or substitution variants thereof and their biologically active polypeptide fragments, including amino acid substitutions. In one embodiment, the variant comprises a conservative substitution, or a deletion, insertion, or substitution that does not significantly alter the three-dimensional structure of the polypeptide of interest, such as the VL or VH region of the first binding domain, particularly in the region of the CDR epitope binding region. In some embodiments, the deletion, insertion, or substitution does not alter the function of interest of the anti-hEGFR VL or anti-hEGFR VH or anti-EGFR ScFv present in the first or second binding domain, or both the first and second binding domains, of the precursor construct, which in some embodiments binds to EGFR on target tumor cells. In some embodiments, the deletion, insertion, or substitution does not alter the function of interest of the anti-ROR1 ScFv, or anti-PSMA, or anti-5T4 scFv present in the first or second binding domain, or both the first and second binding domains of the precursor construct, which in some embodiments binds to ROR1, PSMA, or 5T4, respectively, on target tumor cells.

[0278] In some embodiments, the first or second binding domain, or both the first and second binding domains, that bind to a cell surface tumor associated antigen comprise the sequence set forth in SEQ ID NO: 34 or a homolog thereof. In some embodiments, the first or second binding domain, or both the first and second binding domains, that bind to a cell surface tumor associated antigen comprise the sequence set forth in SEQ ID NO: 37 or a homolog thereof.

[0279] In some embodiments, the first or second binding domain, or the first and second binding domains, that bind to a cell surface tumor-associated antigen comprise a sequence recited in any one of SEQ ID NOs: 34, 37, 156, 166, 168, 170, 172, or 174, or a homolog thereof.

[0280] In some embodiments, the first or second binding domain, or the first and second binding domain, that binds to a cell surface tumor associated antigen comprises the sequence set forth in SEQ ID NO: 41 or a homolog thereof. In some embodiments, the first or second binding domain, or the first and second binding domain, that binds to a cell surface tumor associated antigen comprises the sequence set forth in SEQ ID NO: 42 or a homolog thereof.

[0281] In some embodiments, the first or second binding domain, or both the first and second binding domains, that bind to a cell surface tumor-associated antigen are encoded by a nucleotide sequence comprising the sequence recited in any one of SEQ ID NOs: 36, 38, 157, 167, 169, 171, 173, or 175, or a homolog thereof.

[0282] In some embodiments, the first or second binding domain, or the first and second binding domains, that bind to a cell surface tumor-associated antigen are encoded by a nucleotide sequence comprising the sequence set forth in SEQ ID NO: 36, or a homolog thereof, and the sequence set forth in SEQ ID NO: 38, or a homolog thereof.

[0283] In some embodiments, the first or second binding domain, or the first and second binding domains, that bind to a cell surface tumor associated antigen are encoded by a nucleotide sequence comprising the sequence set forth in SEQ ID NO: 43 or a homolog thereof. In some embodiments, the first or second binding domain, or the first and second binding domains, that bind to a cell surface tumor associated antigen comprise the sequence set forth in SEQ ID NO: 44 or a homolog thereof.

[0284] In some embodiments, the nucleotide sequence encoding the precursor trispecific antibody construct polypeptide is optimized for mammalian transcription and translation. In some embodiments, the nucleotide sequence encoding the first binding domain or the second binding domain or both the first and second binding domains of the precursor trispecific antibody construct polypeptide is optimized for mammalian transcription and translation. In some embodiments, the nucleotide sequence of the VL or VH or both the VL and VH regions of the first binding domain or the second binding domain or the first and second binding domains is optimized for mammalian transcription and translation.

[0285] In another embodiment, the TAAs provided herein are angiogenic antigens that are expressed on both activated pericytes and pericytes of tumor angiogenic vasculature, the latter of which is associated with neovascularization in vivo. Angiogenic antigens are known in the art, see, for example, WO 2010 / 102140, which is incorporated herein by reference. For example, angiogenic antigens can be selected from: angiopoietin-1 (Ang1), angiopoietin 3, angiopoietin 4, angiopoietin 6; Del-1; fibroblast growth factors: acidic (aFGF) and basic (bFGF); follistatin; granulocyte colony stimulating factor (G-CSF); hepatocyte growth factor (HGF) / spreading factor (SF); interleukin-8 (IL-8); leptin; midkine; placental growth factor; platelet-derived endothelial growth factor (PD-1); platelet-derived endothelial growth factor (PD-2); platelet-derived endothelial growth factor (PD-3); platelet-derived endothelial growth factor (PD-4); platelet-derived endothelial growth factor (PD-6); platelet-derived endothelial growth factor (PD-7); platelet-derived endothelial growth factor (PD-8); platelet-derived endothelial growth factor (PD-9); platelet-derived endothelial growth factor (PD-11); platelet-derived endothelial growth factor (PD-12); platelet-derived endothelial growth factor (PD-13); platelet-derived endothelial growth factor (PD-14); platelet-derived endothelial growth factor (PD-15); platelet-derived endothelial growth factor (PD-16); platelet-derived endothelial growth factor (PD-17); platelet-derived endo -ECGF); platelet-derived growth factor-BB (PDGF-BB); pleiotrophin (PTN); progranulin; proliferin; survivin; transforming growth factor-α (TGF-α); transforming growth factor-β (TGF-β); tumor necrosis factor-α (TNF-α); vascular endothelial growth factor (VEGF) / vascular permeability factor (VPF).

[0286] As described above and throughout, in some embodiments, the first binding domain or the second binding domain, or both the first and second binding domains (TAA binding domains) comprise a single chain variable fragment (scFv).

[0287] In some embodiments, the third binding domain (the extracellular epitope binding domain of CD3ε) comprises a Fab fragment. The specific structural order of the components of the precursor trispecific antibody construct, for example contained in polypeptide A and polypeptide B, is described in more detail throughout.

[0288] In some embodiments, the precursor trispecific antibody construct comprises a Fab fragment at its core, which in some embodiments comprises a third binding domain. It will be understood by those skilled in the art that a Fab fragment is an antigen-binding fragment of an antibody. Fab consists of a constant region and a variable region of an immunoglobulin heavy chain and an immunoglobulin light chain. The heavy chain constant (CH1) and variable (VH) regions heterodimerize with the light chain variable (VL) and constant (CL) regions and are typically covalently linked by disulfide bonds between the heavy and light chain constant regions (see, e.g., Figure 1 and 2A -2B and 2F, and in Figure 8A 、 9A, 10A, 11A, 45A-45B, 46, 47A-47B and 48, which indicate the cysteine ​​residues that can form disulfide bonds (Cys-SS-Cys bonds) between polypeptides A and B of the precursor construct (highlighted in bold and underlined). The codons encoding these Cys residues are shown in Figure 8B 、 9B , 10B and 11B (highlighted in bold and underlined). Thus, one of skill will understand that the term "Fab" with respect to an antibody generally includes the portion of an antibody consisting of a single light chain (both variable and constant regions) bound by disulfide bonds to the variable region and first constant region of a single heavy chain.

[0289] Those skilled in the art will recognize that disulfide bonds between heavy and light chains are preferred, but are not essential for function (Orcutt et al. (2010), PEDS, 23: 221-228). Therefore, in certain embodiments, the Fab fragment disclosed herein may not contain disulfide bonds. In this regard, the heavy and light chains can be modified in such a way that they stably interact without the need for disulfide bonds. For example, in certain embodiments, the heavy or light chain can be modified to remove cysteine ​​residues, and wherein the heavy and light chains still stably interact and perform the function of Fab. In some embodiments, mutations are performed to promote stable interactions between the heavy and light chains. For example, a "knobs into hole" modification strategy can be used to promote dimerization between the heavy and light chains of Fab (see, for example, 1996 Protein Engineering, 9: 617-621). Using this strategy, a "knob" is created by replacing the small amino acid side chains at the interface between the interaction domains with larger side chains. By replacing larger side chains with smaller ones, a corresponding "hole" is generated at the interface between the interacting molecules. Therefore, variant Fab fragments designed for specific purposes are also considered for use herein, for example, amino acid changes in the constant domains of CH1 and / or CL, as well as removal of disulfide bonds or addition of tags for purification.

[0290] In some embodiments, the configuration of the variable and constant regions within the Fab fragment may be different from that found in natural Fab. In other words, in one embodiment, the orientation of the variable and constant regions can be VH-CL in one chain and VL-CH1 in another (Shaefer et al. (2011), PNAS, 108: 111870-92). Such modified Fab fragments still function to bind to their specific target antigens and are intended to be used in the precursor constructs disclosed herein. Therefore, in this regard, the variable and constant regions that make up Fab are considered to be modular.

[0291] In certain embodiments, the Fab fragments of the present disclosure are derived from monoclonal antibodies and can be derived from any type of antibody, including IgA, IgM, IgD, IgG, IgE, and subtypes thereof, such as IgG1, IgG2, IgG3, and IgG4. The light chain domain can be derived from a kappa or lambda chain. The Fab fragments used herein can be recombinantly prepared.

[0292] As is well known in the art, antibodies are immunoglobulin molecules that can specifically bind to targets, such as carbohydrates, polynucleotides, lipids, polypeptides, etc., through at least one epitope recognition site located in the variable region of the immunoglobulin molecule. Technicians will understand that the term "antibody" includes not only complete polyclonal or monoclonal antibodies, but also humanized antibodies, chimeric antibodies, antibody fragments including antibody fragments lacking Fc regions, and any other modified configurations of immunoglobulin molecules containing antigen binding sites or desired specific fragments (epitope recognition sites), including scFv fragments and Fab fragments. In some embodiments, the precursor antibody constructs described herein lack Fc regions.

[0293] As disclosed herein, the Fab fragment comprises an antigen binding portion (third binding domain) consisting of an immunoglobulin heavy chain variable region and an immunoglobulin light chain variable region (VH and VL, respectively). Similarly, the above-mentioned scFv fragment (first or second binding domain) comprises an antigen binding portion consisting of an immunoglobulin heavy chain variable region and an immunoglobulin light chain variable region (VH and VL, respectively). More specifically, as used herein, the term "antigen binding portion" refers to a polypeptide fragment comprising at least one CDR of an immunoglobulin heavy chain and / or light chain, which binds to a target antigen of interest, such as the TAA of the first or second binding region, or the CD3 molecule of the third binding region. In this regard, the antigen binding portion of the precursor construct described herein may comprise 1, 2, 3, 4, 5 or all 6 CDRs of the VH and VL sequences of the parent antibody that binds to the target antigen of interest. In certain embodiments, the antigen binding portion of the scFv fragment (first or second binding domain, or both the first and second binding domains) of the precursor trispecific antibody construct binds to TAA, such as, but not limited to, human EGFR. In certain embodiments, the antigen binding portion of the Fab fragment of the precursor trispecific antibody construct binds CD3.

[0294] In certain embodiments, the specific VH and / or VL of the precursor trispecific antibody constructs described herein can be used to screen libraries of complementary variable regions to identify VH / VL with desired properties, such as improved affinity for the target antigen of interest. Such methods are described, for example, in Portolano et al., J. Immunol. (1993) 150: 880-887; Clarkson et al., Nature (1991) 352: 624-628.

[0295] Other methods can also be used to mix and match CDRs to identify Fabs with desired binding activity (e.g., binding to CD3 or other target antigens of interest as described herein for other binding domains present in the precursor trispecific antibody constructs). For example: Klimka et al., British Journal of Cancer (2000) 83:252-260, describe a screening process using mouse VL and human VH libraries, wherein the CDR3 and FR4 are retained from the mouse VH. After obtaining the antibody, the VH is screened against the human VL library to obtain antibodies that bind to the antigen. Beiboer et al., J. Mol. Biol. (2000) 296:833-849 describe a screening process using a complete mouse heavy chain and human light chain library. After obtaining the antibody, one VL is combined with a human VH library that retains the mouse CDR3. Antibodies capable of binding to the antigen are obtained. Rader et al., PNAS (1998) 95:8910-8915 describe a process similar to that described above by Beiboer et al.

[0296] These just described techniques are known in the art per se. However, the skilled person will be able to use such techniques to obtain antigen-binding fragments of antibodies according to several embodiments of the disclosure described herein using routine methods in the art.

[0297] Also disclosed herein are methods for obtaining antibody antigen-binding domains specific for a target antigen (e.g., CD3 or any target antigen for which a binding domain is described herein elsewhere), the methods comprising providing a VH domain by adding, deleting, substituting, or inserting one or more amino acids in the amino acid sequence of a VH domain listed herein, which is an amino acid sequence variant of the VH domain, optionally combining the VH domain thus provided with one or more VL domains, and testing the VH domains or one or more VH / VL combinations to identify specific binding members or antibody antigen-binding domains that are specific for the target antigen of interest (e.g., CD3) and optionally have one or more desired properties. The VL domain may have an amino acid sequence substantially as set forth herein. Similar methods may be employed in combining one or more sequence variants of the VL domains disclosed herein with one or more VH domains.

[0298] Those skilled in the art will understand that "specifically binds" or "preferentially binds" (used interchangeably herein) to an epitope of an antibody or polypeptide are terms well known in the art, and methods for determining such specific or preferential binding are also well known in the art. A molecule is said to exhibit "specific binding" or "preferential binding" if it reacts or associates with a particular cell or substance more frequently, more rapidly, for a longer duration, and / or with greater affinity than with an alternative cell or substance. An antibody or its Fab or scFv "specifically binds" or "preferentially binds" to a target if it binds with greater affinity, avidity, more rapidly, and / or for a longer duration than it binds to another substance. For example, an antibody that specifically or preferentially binds to a CD3 epitope is an antibody that binds to one CD3 epitope with greater affinity, avidity, more rapidly, and / or for a longer duration than it binds to another CD3 epitope or a non-CD3 epitope. It will also be understood from this definition that, for example, an antibody (or portion or epitope) that specifically or preferentially binds to a first target may or may not specifically or preferentially bind to a second target. Thus, "specific binding" or "preferential binding" does not necessarily require (although it can include) exclusive binding. Typically (but not necessarily), reference to binding means preferential binding.

[0299] In certain embodiments, the antigen-binding portion of the Fab fragment (the third binding domain) as described herein includes heavy chain and light chain CDR groups, which are inserted between heavy chain and light chain framework region (FR) groups, respectively, and the framework region (FR) groups provide support for CDR and define the spatial relationship of CDR relative to each other. As used herein, the term "CDR group" refers to three hypervariable regions in the heavy chain or light chain V region. Starting from the N- end of the heavy chain or light chain, these regions are respectively expressed as "CDR1", "CDR2" and "CDR3". Therefore, the antigen binding site includes six CDRs, comprising a CDR group from each of the heavy chain and light chain V regions. The polypeptide comprising a single CDR (e.g., CDR1, CDR2 or CDR3) is referred to herein as a "molecular recognition unit". Crystallographic analysis of many antigen-antibody complexes shows that the amino acid residues of CDR form extensive contacts with the bound antigen, wherein the most extensive antigen contact is with the heavy chain CDR3. Therefore, the molecular recognition unit is primarily responsible for the specificity of the antigen binding site.

[0300] As used herein, the term "FR group" refers to the four flanking amino acid sequences of the CDRs that make up the CDR group of the heavy or light chain V region. Some FR residues may contact the bound antigen; however, FR is primarily responsible for folding the V region into an antigen binding site, particularly the FR residues directly adjacent to the CDR. In FR, certain amino acid residues and certain structural features are very highly conserved. In this regard, all V region sequences contain an internal disulfide loop of approximately 90 amino acid residues. When the V region is folded into a binding site, the CDR appears as a protruding loop motif that forms the antigen binding surface. It is recognized that there are conserved structural regions of FR that affect the folding shape of the CDR loop into certain "canonical" structures - regardless of the precise CDR amino acid sequence. In addition, certain FR residues are known to participate in non-covalent interdomain contacts that stabilize the interaction between the heavy and light chains of antibodies.

[0301] The structure and location of immunoglobulin variable regions can be determined by reference to Kabat, EA et al., Sequences of Proteins of Immunological Interest. 4th Edition. US Department of Health and Human Services. 1987 and updates thereof, now available on the internet (mmune.bme.nwu.edu).

[0302] The skilled person will recognize that the term "monoclonal antibody" includes a homogeneous antibody population, wherein the monoclonal antibody is composed of amino acids (naturally occurring and non-naturally occurring) that participate in epitope selective binding. Monoclonal antibodies are highly specific and are directed against a single epitope. The term "monoclonal antibody" includes not only complete monoclonal antibodies and full-length monoclonal antibodies, but also fragments thereof (e.g., Fab, Fab', F(ab')2, Fv), single chains (ScFv), variants thereof, fusion proteins comprising antigen-binding portions, humanized monoclonal antibodies, chimeric monoclonal antibodies, and any other modified configurations of immunoglobulin molecules (the immunoglobulin molecules comprising antigen-binding fragments (epitope recognition sites) with the desired specificity and ability to bind epitopes). This is not intended to limit the source of the antibody or its preparation method (e.g., by hybridoma, phage selection, recombinant expression, transgenic animals, etc.). The term includes complete immunoglobulins as described herein as well as fragments, etc.

[0303] The proteolytic enzyme papain preferentially cleaves IgG molecules to produce several fragments, two of which (F(ab) fragments) each comprise a covalent heterodimer containing an intact antigen-binding site. Pepsin is capable of cleaving IgG molecules to provide several fragments, including a F(ab')2 fragment comprising two antigen-binding sites. The Fv fragments used in accordance with certain embodiments disclosed herein can be produced by preferential proteolytic cleavage of IgM, and in a few cases by preferential proteolytic cleavage of IgG or IgA immunoglobulin molecules. However, Fv fragments are more commonly derived using recombinant techniques known in the art. Fv fragments include non-covalent VH::VL heterodimers that include an antigen-binding site that retains most of the antigen recognition and binding capabilities of a native antibody molecule. Inbar et al. (1972) Proc. Nat. Acad. Sci. USA 69:2659-2662; Hochman et al. (1976) Biochem 15:2706-2710; and Ehrlich et al. (1980) Biochem 19:4091-4096.

[0304] In some embodiments of the present disclosure, the Fab fragment comprising the third binding domain binds to CD3. In some embodiments of the present disclosure, the Fab fragment comprising the third binding domain binds to CD3ε.

[0305] "T cell receptor" (TCR) is a molecule found on the surface of T cells that, together with CD3, is generally responsible for recognizing antigens that bind to major histocompatibility complex (MHC) molecules. In most T cells, it consists of a heterodimer of highly variable (α) and (β) chains connected by disulfide bonds. In other T cells, an optional receptor consisting of variable γ and (δ) chains is expressed. Each chain of TCR is a member of the immunoglobulin superfamily and has an N-terminal immunoglobulin variable region, an immunoglobulin constant region, a transmembrane region, and a short cytoplasmic tail at the C-terminus (see Abbas and Lichtman, Cellular and Molecular Immunology (5th Ed.), Editor: Saunders, Philadelphia, 2003; Janeway et al, Immunobiology: The Immune System in Health and Disease, 4th Ed., Current Biology Publications, p 148, 149, and 172, 1999). The TCRs used in the present disclosure can be derived from a variety of animal species, including humans, mice, rats, or other mammals.

[0306] "Anti-TCR Fab" or "anti-TCR precursor trispecific antibody construct" refers to a Fab or a precursor trispecific antibody construct comprising a Fab that specifically binds to a TCR molecule or one of its individual chains (e.g., TCR (α), TCR (β), TCR Y or TCR (δ) chain). In certain embodiments, the anti-TCR Fab binds to TCR (α), TCR (β) or both. The skilled artisan will understand that in some embodiments, the term "anti-TCR Fab" may include the third binding domain of the precursor trispecific antibody construct described herein. In some embodiments, the term "anti-TCR Fab" may include a precursor construct in which the binding properties of the third binding domain are mentioned.

[0307] "CD3" is known in the art as a six-chain multiprotein complex (see Smith-Garvin et al., Annu Rev Immunol. 2009; 27: 591-619). In mammals, the complex comprises a homodimer of the CD3 (γ) chain, the CD3 (δ) chain, two CD3 (epsilon; ε) chains, and the CD3 (ζ) chain. The CD3 (γ), CD3 (δ), and CD3 (ε) chains are highly related cell surface proteins of the immunoglobulin superfamily that contain a single immunoglobulin domain. The transmembrane regions of the CD3 (γ), CD3 (δ), and CD3 (ε) chains are negatively charged, a feature that allows these chains to associate with positively charged T cell receptor chains. The intracellular tails of the CD3 (γ), CD3 (δ), and CD3 (ε) chains each contain a single conserved motif, known as an immunoreceptor tyrosine-based activation motif or ITAM, while each CD3 (ζ) chain has three. Without wishing to be bound by theory, it is believed that ITAMs are important for the signaling ability of the TCR complex.The CD3 used in the present disclosure can be from a variety of animal species, including human, mouse, rat or other mammals.

[0308] As used herein, "anti-CD3 Fab" refers to a Fab comprising a third binding domain that specifically binds to a single CD3 chain (e.g., CD3 (γ) chain, CD3 (δ) chain, or CD3 (epsilon; ε) chain) or a complex formed by two or more single CD3 chains (e.g., a complex of more than one CD3 (ε) chain, a complex of CD3 (γ) and CD3 (ε) chains, a complex of CD3 (δ) and CD3 (ε) chains). In certain embodiments, the anti-CD3 Fab specifically binds to CD3 (γ), CD3 (δ), or CD 3 (ε), or any combination thereof, and in certain embodiments, specifically binds to CD3 (ε). In some embodiments, the anti-CD3 Fab binds to the N-terminus of CD3ε. In some embodiments, the anti-CD3 Fab binds to an extracellular epitope of CD3ε.

[0309] In some embodiments, the anti-CD3 Fab binds to an epitope contained within amino acids 1-27 of CD3 epsilon. In some embodiments, the anti-CD3 Fab binds to amino acids 1-27 of CD3 epsilon. In some embodiments, the anti-CD3 Fab binds to amino acids 1-27 of human CD3 epsilon. Amino acids 1-27 of CD3 epsilon are set forth in SEQ ID NO:5.

[0310] The skilled artisan will understand that in some embodiments, the term "anti-CD3 Fab" may include the third binding domain of a precursor trispecific antibody construct described herein. In some embodiments, the term "anti-CD3 Fab" may include a precursor construct in which the binding properties of the third binding domain are mentioned.

[0311] In some embodiments, the third binding domain of the precursor construct comprises a Fab. In some embodiments, when referring to the third binding domain of the precursor construct, the term "Fab" will be used, wherein the term includes the third binding domain of the precursor construct. In some embodiments, the term "Fab" can be used interchangeably with the phrase "third binding domain" and both have the same properties and meaning.

[0312] In some embodiments, the precursor trispecific antibody construct comprises a third binding domain that binds to an extracellular epitope of CD3 epsilon. In some embodiments, the precursor trispecific antibody construct comprises a third binding domain that binds to the N-terminus of CD3 epsilon. In some embodiments, the precursor trispecific antibody construct comprises a third binding domain that binds to an epitope comprising amino acids 1-27 of CD3 epsilon. In some embodiments, the anti-CD3 Fab binds to amino acids 1-27 of CD3 epsilon. In some embodiments, the anti-CD3 Fab binds to amino acids 1-27 of human CD3 epsilon. Amino acids 1-27 of CD3 epsilon are set forth in SEQ ID NO:5.

[0313] As used herein, "TCR complex" refers to a complex formed by the association of CD3 with TCR. For example, the TCR complex can be composed of a CD3 (γ) chain, a CD3 (δ) chain, two CD3 (ε) chains, a homodimer of a CD3 (ζ) chain, a TCR (α) chain, and a TCR (β) chain. Alternatively, the TCR complex can be composed of a CD3 (γ) chain, a CD3 (δ) chain, two CD3 (ε) chains, a homodimer of a CD 3 (ζ) chain, a TCRγ chain, and a TCR (δ) chain.

[0314] As used herein, "component of a TCR complex" refers to a TCR chain (i.e., TCR(α), TCR(β), TCRγ, or TCR(δ)), a CD3 chain (i.e., CD3(γ), CD3(δ), CD3(ε), or CD3(ζ)), or a complex formed by two or more TCR chains or CD3 chains (e.g., a complex of TCR(α) and TCR(β), a complex of TCRγ and TCR(δ), a complex of CD3(ε) and CD3(δ), a complex of CD3(γ) and CD3(ε), or a sub-TCR complex of TCR(α), TCR(β), CD3(γ), CD3(δ), and two CD3(ε) chains).

[0315] As background, the TCR complex is generally responsible for initiating T cell responses to antigens bound to MHC molecules. It is believed that the binding of peptide:MHC ligands to the TCR and co-receptors (i.e., CD4 or CD8) brings together the TCR complex, co-receptors, and CD45 tyrosine phosphatase. This allows CD45 to remove inhibitory phosphate groups, thereby activating Lck and Fyn protein kinases. Activation of these protein kinases leads to phosphorylation of ITAMs on the CD3 (ζ) chains, which in turn enables these chains to bind to the intracytoplasmic tyrosine kinase ZAP-70. The phosphorylation of the bound ZAP-70 is then triggered by phosphorylation, which triggers three signaling pathways, two of which are initiated by the phosphorylation and activation of PLC-(γ), which then cleaves phosphatidylinositol phosphate (PIP) into diacylglycerol (DAG) and inositol triphosphate (IP3). DAG activates protein kinase C, leading to the activation of the transcription factor NFKB. Due to the action of IP3, intracellular free Ca 2+ The sudden increase in ATP activates the cytoplasmic phosphatase calcineurin, which causes the transcription factor NFAT (nuclear factor of activated T cells) to translocate from the cytoplasm to the nucleus. Full transcriptional activity of NFAT also requires members of the AP-1 family of transcription factors; dimers of members of the Fos and Jun families of transcriptional regulators.

[0316] The third signaling pathway initiated by activated ZAP-70 is the activation of Ras and the subsequent activation of the MAP kinase cascade. This ultimately leads to the activation of Fos and, consequently, the activation of the AP-1 transcription factor. NFKB, NFAT, and AP-1 act together on T cell chromosomes to initiate the transcription of new genes that lead to T cell differentiation, proliferation, and effector functions. See Pitcher et al., 2003., TRENDS in Immunol. 24, 554-560; Smith-Garvin et al., Annu Rev Immunol. 2009; 27: 591-619.

[0317] In certain embodiments, the Fab specifically binds to a single human CD3 chain (e.g., a human CD3 (γ) chain, a human CD3 (δ) chain, or a human CD3 (ε) chain) or a combination of two or more single human CD3 chains (e.g., a complex of human CD3 (γ) and human CD3 (ε) or a complex of human CD3 (δ) and human CD3 (ε)). In certain embodiments, the Fab specifically binds to a human CD3 (ε) chain. In certain embodiments, the Fab specifically binds to an extracellular epitope of the human CD3 (ε) chain. In certain embodiments, the Fab specifically binds to an epitope within SEQ ID NO: 3.

[0318] In certain embodiments, the third binding domain specifically binds to a single human CD3 chain (e.g., a human CD3 (γ) chain, a human CD3 (δ) chain, or a human CD3 (ε) chain) or a combination of two or more single human CD3 chains (e.g., a complex of human CD3 (γ) and human CD3 (ε) or a complex of human CD3 (δ) and human CD3 (ε)). In certain embodiments, the third binding domain specifically binds to a human CD3 (ε) chain. In certain embodiments, the third binding domain specifically binds to an extracellular epitope of the human CD3 (ε) chain. In certain embodiments, the third binding domain specifically binds to an epitope within SEQ ID NO: 3.

[0319] In certain other embodiments, a Fab of the present disclosure comprising a third binding domain specifically binds to TCR(α), TCR(β), or a heterodimer formed by TCR(α) and TCR(β). In certain embodiments, the Fab specifically binds to one or more of a human TCR(α), a human TCR(β), or a heterodimer formed by a human TCR(α) and a human TCR(β).

[0320] In certain embodiments, the Fab of the present disclosure comprising a third binding domain binds to a complex formed by one or more CD3 chains and one or more TCR chains, such as a complex formed by a CD3 (γ) chain, a CD3 (δ) chain, a CD3 (ε) chain, a TCR (α) chain, or a TCR (β) chain, or any combination thereof. In other embodiments, the Fab of the present disclosure binds to a complex formed by one CD3 (γ) chain, one CD3 (δ) chain, two CD3 (ε) chains, one TCR (α) chain, and one TCR (β) chain. In further embodiments, the Fab of the present disclosure binds to a complex formed by one or more human CD3 chains and one or more human TCR chains, such as a complex formed by a human CD3 (γ) chain, a human CD3 (δ) chain, a human CD3 (ε), a human TCR (α) chain, or a human TCR (β) chain, or any combination thereof. In certain embodiments, the Fabs of the present disclosure bind to a complex formed by one human CD3(γ) chain, one human CD3(δ) chain, two human CD3(ε) chains, one human TCR(α) chain, and one human TCR(β) chain.

[0321] Fab of the present disclosure can be produced as described herein or by various methods known in the art (see, e.g., U.S. Patent Nos. 6,291,161; 6,291,158). Sources of Fab include monoclonal antibody nucleic acid sequences (which can be converted into antibodies, Fv, scFv or Fab, e.g., in phage libraries) from various species, including humans, camelids (from camels, dromedaries or llamas; Hamers-Casterman et al. (1993) Nature, 363:446 and Nguyen et al. (1998) J. Mol. Biol., 275:413), sharks (Roux et al. (1998) Proc. Nat'l. Acad. Sci. (USA) 95:11804), fish (Nguyen et al. (2002) Immunogenetics, 54:39), rodents, birds or sheep.

[0322] Anti-human CD3 antibodies that cross-react with monkey CD3 are particularly desirable, such as the SP34 mouse monoclonal antibody, which specifically binds to both denatured human CD3 (by Western blot or dot blot) and native human CD3 (on T cells) (Pressano, S. The EMBO J. 4:337-344, 1985; Alarcon, B. EMBO J. 10:903-912, 1991). The SP34 mouse monoclonal antibody also binds to COS cells transfected with CD3c alone, as well as to CD3ε / γ or CD3ε / δ double transfectants (Salmeron A. et al., J. Immunol. 147:3047-52, 1991). The SP34 antibody also cross-reacts with non-human primates (Yoshino N. et al., Exp. Anim 49:97-110, 2000; Conrad M L. et al., Cytometry 71A:925-33, 2007). In addition, upon cross-linking, SP34 activates T cells (Yang et al., J. Immunol. 137:1097-1100, 1986). Cross-reactivity with monkey CD3 is important because it allows toxicity studies to be conducted directly with clinical candidates in non-human primates, rather than in chimpanzees or using surrogate molecules. Therefore, toxicity studies using such cross-reactive anti-CD3 Fabs in the precursor trispecific antibody constructs of the present disclosure can provide more relevant safety assessments.

[0323] Other exemplary anti-CD3 antibodies include Cris-7 monoclonal antibody (Reinherz, EL et al. (eds.), Leukocyte typing II., Springer Verlag, New York, (1986)), BC3 monoclonal antibody (Anasetti et al. (1990) J. Exp. Med. 172: 1691), OKT3 (Ortho multicenter Transplant Study Group (1985) N. Engl. J. Med. 313: 337) and its derivatives such as OKT3 ala-ala (Herold et al. (2003) J. Clin. Invest. 11: 409), visilizumab (Carpenter et al. (2002) Blood 99: 2712), and 145-2C11 monoclonal antibody (Hirsch et al. (1988) J. Immunol. 140: 3766). Further CD3 binding molecules contemplated for use herein include UCHT-1 (Beverley, PC and Callard, RE (1981) Eur. J. Immunol. 11:329-334) and the CD3 binding molecules described in WO2004 / 106380; WO2010 / 037838; WO2008 / 119567; WO2007 / 042261; WO2010 / 0150918, the entire contents of which are incorporated herein.

[0324] In some embodiments, the amino acid sequence of the third binding region comprising anti-CD3ε binding activity comprises any anti-CD3ε sequence known in the art. In some embodiments, the amino acid sequence of the third binding region comprising binding activity to anti-CD3ε or its derivatives or antibody fragments thereof comprises any anti-CD3ε sequence known in the art. Examples of known anti-CD3ε amino acid sequences can be found in, for example, but not limited to, U.S. Patent Nos.: 9,822,180; 9,493,563; 9,587,021; 9,562,073; U.S. Published Application Nos.: 2013 / 0129729; 2017 / 0247476; 2016 / 0194399; 2010 / 0150918; 2018 / 0112011; and WO2017 / 162587, the entire contents of which are contained herein.

[0325] An exemplary anti-TCR antibody is the H57 monoclonal antibody (Lavasani et al. (2007) Scandinavian Journal of Immunology 65:39-47).

[0326] Antigen-binding fragment sequences (e.g., heavy and light chain variable region sequences) of Fab fragments are available in public databases or can be used to develop Fab for use herein using conventional hybridoma development strategy components in conventional systems (e.g., mouse, HuMAb mouse.RTM., TC mouse.TM., KM-mouse.RTM., llama, chicken, rat, hamster, rabbit, etc.) using CD3 chains, TCR components, or other Fab binding targets as immunogens. Those skilled in the art will appreciate that Fab fragments can be produced using a variety of techniques known in the art, including antibody display technologies such as phage, yeast, ribosome, and mRNA display technologies; B cell culture technologies such as SLAM technology; or high-throughput gene sequencing technologies for B cells or plasma B cells isolated from immunized animal subjects or immunized human subjects.

[0327] In some embodiments, the third binding domain (Fab) disclosed herein comprises a humanized FR amino acid sequence and a native sequence of a CDR amino acid sequence of a mouse monoclonal antibody. Examples of anti-CD3ε amino acid sequences in which the FR sequences are humanized while the CDR amino acid sequences remain those of the SP34 mouse monoclonal antibody are disclosed in International Application Publication No. WO2007 / 042261, the entire contents of which are incorporated herein.

[0328] Exemplary third binding domain (e.g., but not limited to, anti-CD3ε Fab) sequences included in the precursor trispecific antibody constructs of the present disclosure include VH, CH1, VL, and CL amino acid sequences, and polynucleotides encoding them, as shown in Tables 1 and 2 below, respectively. Amino acid sequences comprising the third binding domain include those recited below: SEQ ID NOs: 46-72 and 114 (VH) and 75-103 and 116 (VL), including their CDRs, such as those recited in SEQ ID NOs: 104-112. In some embodiments, the third binding domain (e.g., Fab) sequences included in the precursor trispecific antibody constructs of the present disclosure include VH, CH1, VL, and CL amino acid sequences as recited in Table 2, or homologs thereof. In some embodiments, homologs of SEQ ID NOs: 46-72 and 114 and 75-103 and 116 retain their CDR regions, such as those recited in SEQ ID NOs: 104-112.

[0329] Table 2: Amino acid sequences of anti-CD3 VH, VL, HC, LC and CDRs and their combinations.

[0330]

[0331]

[0332]

[0333] In some embodiments, the third binding domain binds to the CD3ε polypeptide. In some embodiments, the third binding domain binds to the extracellular domain of the human CD3ε polypeptide. In some embodiments, the third binding domain comprises a Fab fragment comprising a variable heavy chain region (VH) comprising CDR-H1, CDR-H2, and CDR-H3 and a variable light chain region (VL) comprising CDR-L1, CDR-L2, and CDR-L3, wherein the third binding domain binds to the extracellular domain of the human CD3ε polypeptide. In some embodiments, the third binding domain binds to an epitope within SEQ ID NO: 3. In some embodiments, the third binding domain binds to SEQ ID NO: 5.

[0334] In some embodiments, the amino acid sequence of anti-human CD3ε CDR-H1 is set forth in SEQ ID NO: 104. In some embodiments, the amino acid sequence of anti-human CD3ε CDR-H2 is set forth in SEQ ID NO: 105. In some embodiments, the amino acid sequence of anti-human CD3ε CDR-H3 is set forth in SEQ ID NO: 106. In some embodiments, the amino acid sequence of anti-human CD3ε CDR-L1 is set forth in any one of SEQ ID NOs: 107-109. In some embodiments, the amino acid sequence of anti-human CD3 3ε CDR-L1 is set forth in SEQ ID NO: 110. In some embodiments, the amino acid sequence of anti-human CD3 3ε CDR-L1 is set forth in SEQ ID NO: 111-112.

[0335] In some embodiments, the third binding domain comprises a Fab fragment comprising a variable heavy chain region (VH) and a variable light chain region (VL) that bind to the extracellular domain of a human CD3ε polypeptide. In some embodiments, the amino acid sequences of the anti-human CD3ε VH and VL are selected from the group consisting of SEQ ID NOs: 46-72 and 114 (VH) and 75-103 and 116 (VL). In some embodiments, the amino acid sequences of the anti-human CD3ε VH and VL comprise a sequence homolog selected from the group consisting of SEQ ID NOs: 46-72 and 114 (VH) and 75-103 and 116 (VL).

[0336] In some embodiments, the amino acid sequence of VH of the third binding domain (VH1) of human CD3ε is selected from the amino acid sequence recited in any one of SEQ ID NOs: 75-103 and 116, or a homolog thereof. In some embodiments, the amino acid sequence of VL of the third binding domain (VL1) of human CD3ε is selected from the amino acid sequence recited in any one of SEQ ID NOs: 75-103 and 116, or a homolog thereof.

[0337] In some embodiments, the homolog comprises a polypeptide that is at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% homologous to the amino acid sequence of the variable light chain or variable heavy chain of anti-CD3 epsilon. In some embodiments, the homolog comprises a polypeptide that is at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% identical to the amino acid sequence of the variable light chain or variable heavy chain of anti-CD3 epsilon. In some embodiments, the homolog comprises a polypeptide that is at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 87%, at least 89%, at least 91%, at least 93%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% homologous to anti-human CD3ε VH or anti-human CD3ε VL, respectively, as determined using BlastP software from the National Center for Biotechnology Information (NCBI) using default parameters.

[0338] In some embodiments, homology also includes deletion, insertion, or substitution variants thereof and their biologically active polypeptide fragments, including amino acid substitutions. In one embodiment, the variant comprises a conservative substitution, or a deletion, insertion, or substitution that does not significantly alter the three-dimensional structure of the polypeptide of interest, such as the VL or VH region, specifically in the region of the CDR epitope binding region. In some embodiments, the deletion, insertion, or substitution does not alter the function of interest of the anti-human CD3ε Fab, which in some embodiments binds to the CD3ε sequence on the target T cell.

[0339] In some embodiments, the third binding domain that binds to the CD3 epsilon extracellular epitope VH1 comprises a sequence recited in SEQ ID NOs: 46-72 and 114, or a homolog thereof. In some embodiments, the second binding domain that binds to the CD3 epsilon extracellular epitope VL1 comprises a sequence recited in SEQ ID NOs: 75-103 and 116, or a homolog thereof. In some embodiments, the third binding domain that binds to the CD3 epsilon extracellular epitope comprises a sequence selected from the group consisting of SEQ ID NOs: 46-72, 74, and 114, or a homolog thereof, and a sequence selected from the group consisting of SEQ ID NOs: 75-103, 113, 115, and 116, or a homolog thereof. In some embodiments, the third binding domain that binds to the CD3 epsilon extracellular epitope comprises a sequence recited in SEQ ID NOs: 113, or a homolog thereof, and a sequence recited in SEQ ID NO: 74, or a homolog thereof.

[0340] In some embodiments, the third binding domain VL region comprises the amino acid sequence as described for CDR-L1 (selected from SEQ ID NOs: 107-109), CDR-L2 (SEQ ID NO: 110), and CDR-L3 (selected from SEQ ID NOs: 111 and 112), and the third binding domain VH region comprises CDR-H1 (SEQ ID NO: 104), CDR-H2 (SEQ ID NO: 105), and CDR-H3 (SEQ ID NO: 106).

[0341] In some embodiments of the precursor trispecific antibody construct, the VL region of the third binding domain comprises an amino acid sequence as recited in any one of SEQ ID NOs: 75-103 and 116, or an amino acid sequence having at least 80% homology thereto. In some embodiments, the third binding VL region comprising an amino acid sequence having at least 80% homology thereto comprises a framework sequence having at least 80% homology, wherein the CDR region is "as is" in the selected amino acid sequence (SEQ ID NOs: 107-112).

[0342] In some embodiments of the precursor trispecific antibody construct, the third combined VH region comprises an amino acid sequence as recited in any one of SEQ ID NOs: 46-72 and 114, or an amino acid sequence having at least 80% homology thereto. In some embodiments, the third combined VH region comprising an amino acid sequence having at least 80% homology thereto comprises a framework sequence having at least 80% homology, wherein the CDR region is "as is" in the selected amino acid sequence (SEQ ID NOs: 104-106).

[0343] In some embodiments, the first binding domain comprises a humanized binding domain. In some embodiments, the second binding domain comprises a humanized binding domain. In some embodiments, the third binding domain comprises a humanized binding domain. In some embodiments, the first, second, or third binding domain, or any combination thereof, comprises a humanized binding domain.

[0344] As will be understood by those skilled in the art and as described herein, in some embodiments, a complete antibody comprises two heavy chains and two light chains, each heavy chain consisting of a variable region and a first, second, and third constant region, while each light chain consists of a variable region and a constant region. Mammalian heavy chains are classified as α, δ, Ε, γ, and μ, and mammalian light chains are classified as λ or κ. Immunoglobulins comprising α, δ, Ε, γ, and μ heavy chains are classified as immunoglobulins (Ig) A, IgD, IgE, IgG, and IgM. Complete antibodies form a "Y" shape. The stem of the Y is composed of the second and third constant regions (for IgE and IgM, the fourth constant region) of the two heavy chains that are bound together, and disulfide bonds (interchain) are formed in the hinge. Heavy chains γ, α, and δ have a constant region consisting of three Ig domains in series (in a straight line) and a hinge region for increasing flexibility; heavy chains μ and ε have a constant region consisting of four immunoglobulin domains. The second and third constant regions are referred to as the "CH2 domain" and "CH3 domain," respectively. Each arm of the Y includes the variable region of a single heavy chain and the first constant region, combined with the variable region and constant region of a single light chain. The variable regions of the light and heavy chains are responsible for antigen binding.

[0345] "Complementarity determining regions" or "CDRs" with respect to antibodies refer to the highly variable loops in the variable regions of either the heavy or light chains of an antibody. CDRs can interact with the conformation of an antigen and largely determine binding to the antigen (although some framework regions are known to participate in binding). The heavy and light chain variable regions each contain three CDRs. CDRs can be defined or identified by conventional methods, for example, according to the sequences of Kabat et al. (Wu, TT and Kabat, EA, J Exp Med. 132(2):211-50, (1970); Borden, P. and Kabat EA, PNAS, 84:2440-2443 (1987); Kabat, EA et al, Sequences of proteins of immunological interest, DIANE Press, 19922), or according to the structures of Chothia et al. (Choithia, C. and Lesk, AM, J. Mol. Biol., 196(4):901-917 (1987), Choithia, C. et al, Nature, 342:877-883 (1989)).

[0346] "Heavy chain variable region" or "VH" with respect to antibodies refers to the fragment of the heavy chain containing the three CDRs interposed between flanking segments called framework regions, which are more highly conserved than the CDRs and form a scaffold to support the CDRs.

[0347] "Light chain variable region" or "VL" with reference to an antibody refers to the fragment of a light chain that contains three CDRs interposed between framework regions.

[0348] "Fv" with respect to antibodies refers to the smallest antibody fragment with a complete antigen-binding site. An Fv fragment consists of the variable region of a single light chain bound to the variable region of a single heavy chain.

[0349] "Single-chain Fv antibody" or "scFv" with respect to antibodies refers to an engineered antibody consisting of a light chain variable region and a heavy chain variable region linked to each other directly or through a peptide linker sequence.

[0350] As used herein, "single-domain camelid antibody" or "camelid VHH" refers to the smallest known antigen-binding unit of a heavy chain antibody (Koch-Nolte et al., FASEB J., 21: 3490-3498 (2007)). "Heavy chain antibody" or "camelid antibody" refers to an antibody that contains two VH domains and no light chain (Riechmann L. et al., J. Immunol. Methods 231: 25-38 (1999); WO94 / 04678; WO94 / 25591; U.S. Pat. No. 6,005,079).

[0351] "Single-domain antibody" or "dAb" refers to an antibody fragment consisting of the variable region of the antibody heavy chain (VH domain) or the variable region of the antibody light chain (VL domain) (Holt, L., et al, Trends in Biotechnology, 21(11):484-490).

[0352] As used herein, the term "disulfide bond" refers to the binding of a heavy chain fragment to a light chain fragment via one or more disulfide bonds. One or more disulfide bonds can be formed between the two fragments by linking thiol groups in the two fragments. In certain embodiments, one or more disulfide bonds can be formed between one or more cysteine ​​residues in the heavy chain fragment and one or more cysteine ​​residues in the light chain fragment, respectively.

[0353] A "variable region linking sequence" is an amino acid sequence that connects the heavy chain variable region to the light chain variable region and provides a linker function compatible with the interaction of the two sub-binding domains, such that the resulting polypeptide retains specific binding affinity for the same target molecule as an antibody comprising the same light and heavy chain variable regions. In certain embodiments, a hinge that can be used to connect a binding domain to an immunoglobulin CH2 or CH3 region polypeptide can be used as the variable region linking sequence.

[0354] In some embodiments, the third binding domain comprises a variable heavy chain (VH) region and a variable light chain (VL) region, wherein the first or second sub-regulatory domain is located at the N- end of the VL region or the VH region of the third binding domain. In some embodiments, the third binding domain comprises a variable heavy chain (VH) region and a variable light chain (VL) region, wherein the first sub-regulatory domain is located at the N- end of the VL region of the third binding domain, and the second sub-regulatory domain is located at the N- end of the VH region of the third binding domain. In some embodiments, the third binding domain comprises a variable heavy chain (VH) region and a variable light chain (VL) region, wherein the first sub-regulatory domain is located at the N- end of the VH region of the third binding domain, and the second sub-regulatory domain is located at the N- end of the VL region of the third binding domain.

[0355] In some embodiments, the third binding domain comprises a constant heavy chain (CH1) region and a constant light chain (CL) region, wherein the first or second binding domain is located at the C-terminus of the CH1 region or CL region of the third binding domain. In some embodiments, the third binding domain comprises a constant heavy chain (CH1) region and a constant light chain (CL) region, wherein the first binding domain is located at the C-terminus of the CL region of the third binding domain, and the second binding domain is located at the C-terminus of the CH1 region of the third binding domain. In some embodiments, the third binding domain comprises a constant heavy chain (CH1) region and a constant light chain (CL) region, wherein the first binding domain is located at the C-terminus of the CH1 region of the third binding domain, and the second binding domain is located at the C-terminus of the CL region of the third binding domain. The skilled artisan will appreciate that the first and second sub-regulatory domains are located at the N-terminus of VH and VL, wherein when the first sub-regulatory domain is located at the N-terminus of VH, the second sub-regulatory domain is located at the N-terminus of VL, and vice versa, when the second sub-regulatory domain is located at the N-terminus of the VH of the third binding domain, the first sub-regulatory domain is located at the N-terminus of the VL of the third binding domain. Similarly, the skilled artisan will appreciate that the first and second binding domains are located at the C-terminus of the CH1 and CL domains of the third binding domain, wherein when the first binding domain is located at the C-terminus of CH1, the second binding domain is located at the C-terminus of CL, and vice versa, when the second binding domain is located at the C-terminus of CH1, the first binding domain is located at the C-terminus of CL of the third binding domain.

[0356] Alternative sources of binding domains can include sequences encoding random peptide libraries or sequences encoding engineered amino acid diversity in the loop regions of alternative non-antibody scaffolds, such as fibrinogen domains (see, e.g., Weisel et al. (1985) Science 230: 1388), Kunitz domains (see, e.g., U.S. Patent No. 6,423,498), lipocalin domains (see, e.g., WO 2006 / 095164), V-like domains (see, e.g., U.S. Patent Application Publication No. 2007 / 0065431), C-type lectin domains (Zelensky and Gredy (2005) FEBS J. 272: 6179), or Fcab.™ (see, e.g., PCT Patent Application Publication Nos. WO 2007 / 098934; WO 2006 / 072620), among others.

[0357] like Figure 1 and 2A-2F, scFv is a specific illustrative binding domain. In some embodiments, the first or second binding domain, or both the first and second binding domains, comprising the scFv fragment can bind to any of a variety of target molecules, including but not limited to: FcγRI, FcγRIIa, FcγRIIb, FcγRIIIb, CD28, CD137, CTLA-4, FAS, fibroblast growth factor receptor 1 (FGFR1), FGFR2, FGFR3, FGFR4, glucocorticoid-induced TNFR-related (GITR) protein, lymphotoxin-β receptor (LTβR), toll-like receptor (TLR), tumor necrosis factor-related apoptosis-inducing ligand-receptor 1 (TRAIL receptor 1) and TRAIL receptor 2, prostate-specific membrane antigen (PSMA) protein, prostate stem cell antigen (PSCA) protein, tumor-associated protein carbonic anhydrase IX (CAIX), epidermal growth factor receptor 1 (EGFR1), EGFRvIII, human epidermal growth factor receptor 2 (Her2 / neu;, ErbB2), ErbB3 (also known as HER3), folate receptor, ephrin receptor, PDGFRa, ErbB-2, CD20, CD22, CD30, CD33, CD40, CD37, CD38, CD70, CD74, CD40), CD80, CD86, CD2, p53, cMet (also known as tyrosine protein kinase Met or hepatocyte growth factor receptor (HGFR)), MAGE-A1, MAGE-A2, MAGE-A3, MAGE-A4, MAGE-A6, MAGE-A10, MAGE-A12, BAGE, DAM-6, DAM-10, GAGE-1, GAGE-2, GAGE-8, GAGE-3, GAGE-4, GAGE-5, GAGE-6, GAGE-7B, NA88-A, NY-ESO-1, BRCA1, BRCA2, MART-1, MC1R, Gp100, PSA, PSM, tyrosinase, Wilms' tumor antigen (WT1), TRP-1, TRP-2, ART-4, CAMEL, Cyp-B, hTERT, hTRT, iCE , MUC1, MUC2, P-cadherin, myostatin (GDF8), Cripto (TDGF1), MUC5AC, PRAME, P15, RU1, RU2, SART-1, SART-3, WT1, AFP, β-catenin / m, caspase-8 / m, CDK-4 / m, ELF2M, GnT-V, G250, HSP70-2M, HST-2, KIAA0205, MUM-1, MUM-2, MUM-3, myosin / m, RAGE, SART-2, TRP- 2 / INT2, 707-AP, Annexin II, CDC27 / m, TPI / mbcr-abl, ETV6 / AML, LDLR / FUT, Pml / RARα, TEL / AML1, CD28, CD137, CanAg, Mesothelin, DR5, PD-1, PD1L, IGF-1R, CXCR4, Neuropilin 1, Glypican, EphA2, CD138, B7-H3, B7-H4, gpA33, GPC3, SSTR2, ROR1, 5T4, or VEGF-R2. In some embodiments, the TAA comprises PSMA, CD30, B7-H3, B7-H4, gpA33, HER2, P-cadherin, gp100, DR5, GPC3, SSTR2, Mesothelin, ROR1, 5T4, folate receptor, or EGFR. In some embodiments, the TAA is selected from PSMA, ROR1, 5T4, and EGFR. These and other tumor proteins or tumor-associated proteins are known to those skilled in the art.

[0358] In certain embodiments, the first or second binding domain, or both the first and second binding domains, specifically binds to an antigenic target associated with a disease condition. Disease conditions can include physiological conditions, pathological conditions, and cosmetic conditions. Examples of illustrative conditions include, but are not limited to, cancer, inflammatory disorders, allografts, type I diabetes, type II diabetes, and multiple sclerosis.

[0359] In some embodiments, the specific structural components of the precursor trispecific antibody construct comprise first and second binding domains (such as, but not limited to, scFv fragments), a third binding domain (such as, but not limited to, Fab fragments), a linker region, and first and second sub-regulatory domains, as described in detail herein, wherein the regulatory domains can each comprise a protease cleavable domain and an HSA polypeptide sequence or a protease cleavable domain and a CAP component, and a linker, or any combination thereof.

[0360] In some embodiments, the precursor trispecific antibody construct comprises two polypeptides. In some embodiments, these polypeptides can be identified based on the heavy chain (HC) or light chain (LC) component of the third binding domain. In some embodiments, these polypeptides can be identified as polypeptide A and polypeptide B. In some embodiments, polypeptide A comprises an HC polypeptide and polypeptide B comprises an LC polypeptide. In other embodiments, polypeptide A comprises an LC polypeptide and polypeptide B comprises an HC polypeptide.

[0361] In some embodiments, the precursor trispecific antibody construct described herein comprises a third binding domain comprising a variable heavy chain (VH) region and a variable light chain (VL) region; wherein the first binding domain is located at the C-terminus of the CL or CH1 region of the third binding domain; wherein when the first binding domain is located at the C-terminus of the CL region, the second binding domain is located at the C-terminus of the CH1 region, and when the first binding domain is located at the C-terminus of the CH1 region, the second binding domain is located at the C-terminus of the CL region. In some embodiments, the precursor trispecific antibody construct described herein comprises a third binding domain comprising a variable heavy chain (VH) region and a variable light chain (VL) region;

[0362] wherein the first and second binding domains are positioned as described above, and

[0363] wherein the first sub-regulatory domain comprising the HLP domain located at the N-terminus of the protease cleavage domain is located at the N-terminus of the VH region or the VL region of the third binding domain;

[0364] wherein (a) when the first sub-regulatory domain is located at the N-terminus of the VL region, the second sub-regulatory domain comprising the CAP component located at the N-terminus of the protease cleavage domain is located at the N-terminus of the VH region, and (b) when the first sub-regulatory domain is located at the N-terminus of the VH region, the second sub-regulatory domain comprising the CAP component located at the N-terminus of the protease cleavage domain is located at the N-terminus of the VL region.

[0365] For clarity in the following schematic description, the first and second anti-TAA binding sites will be represented by the term "TAA" and the anti-CD3 third binding site will be represented by the term "CD3."

[0366] In some embodiments, the precursor trispecific antibody construct described herein comprises two polypeptides, polypeptide A and polypeptide B, wherein polypeptide A comprises the following components in the order from N-terminus to C-terminus: HLP domain, protease cleavage domain, CD3 third binding domain VH-CH1 region, first binding domain (VL-VH); and polypeptide B comprises the following components in the order from N-terminus to C-terminus: CAP component, protease cleavage domain, CD3 third binding domain VL-CL region, second binding domain (VL-VH). In some embodiments, the precursor trispecific antibody construct described herein comprises the following two polypeptides (in order from N-terminus to C-terminus):

[0367] Peptide A: HLP-L-CP-L-CD3VH-L-CD3CH1-L-TAA VL-L-TAA VH

[0368] Peptide B: CAP-L-CP-L-CD3VL-L-CD3CL-L-TAA VL-L-TAA VH,

[0369] wherein "L" is a linker, which may or may not be present in each embodiment.

[0370] In some embodiments, the precursor trispecific antibody construct described herein comprises two polypeptides - polypeptide A and polypeptide B, wherein polypeptide A comprises components having an N-terminal to C-terminal order: HLP domain, protease cleavage domain, CD3 third binding domain VH-CH1 region, first binding domain (VH-VL); and polypeptide B comprises components having an N-terminal to C-terminal order: CAP component, protease cleavage domain, CD3 third binding domain VL-CL region, second binding domain (VL-VH). In some embodiments, the precursor trispecific antibody construct described herein comprises the following two polypeptides (in order from N-terminal to C-terminal):

[0371] Peptide A: HLP-L-CP-L-CD3VH-L-CD3CH1-L-TAA VH-L-TAA VL

[0372] Peptide B: CAP-L-CP-L-CD3VL-L-CD3CL-L-TAA VL-L-TAA VH,

[0373] Wherein "L" is a linker, which may or may not be present in each embodiment ( Figure 1 ).

[0374] In some embodiments, the precursor trispecific antibody construct described herein comprises two polypeptides - polypeptide A and polypeptide B, wherein polypeptide A comprises components having an N-terminal to C-terminal order: HLP domain, protease cleavage domain, CD3 third binding domain VH-CH1 region, first binding domain (VH-VL); and polypeptide B comprises components having an N-terminal to C-terminal order: CAP component, protease cleavage domain, CD3 third binding domain VL-CL region, second binding domain (VH-VL). In some embodiments, the precursor trispecific antibody construct described herein comprises the following two polypeptides (in order from N-terminal to C-terminal):

[0375] Peptide A: HLP-L-CP-L-CD3VH-L-CD3CH1-L-TAA VH-L-TAA VL

[0376] Peptide B: AP-L-CP-L-CD3VL-L-CD3CL-L-TAA VH-L-TAA VL,

[0377] wherein "L" is a linker, which may or may not be present in each embodiment.

[0378] In some embodiments, the precursor trispecific antibody construct described herein comprises two polypeptides - polypeptide A and polypeptide B, wherein polypeptide A comprises components having an N-terminal to C-terminal order: HLP domain, protease cleavage domain, CD3 third binding domain VH-CH1 region, first binding domain (VL-VH); and polypeptide B comprises components having an N-terminal to C-terminal order: CAP component, protease cleavage domain, CD3 third binding domain VL-CL region, second binding domain (VH-VL). In some embodiments, the precursor trispecific antibody construct described herein comprises the following two polypeptides (in order from N-terminal to C-terminal):

[0379] Peptide A: HLP-L-CP-L-CD3VH-L-CD3CH1-L-TAA VL-L-TAA VH

[0380] Peptide B: CAP-L-CP-L-CD3VL-L-CD3CL-L-TAA VH-L-TAA VL,

[0381] wherein "L" is a linker, which may or may not be present in each embodiment.

[0382] The above four constructs provide non-limiting examples of trispecific precursor antibody constructs in which the positions of VL and VH within the scFv of the first or second binding region are alternated. The skilled artisan will appreciate that these alternative positions will also be included in the following embodiments, in which scFv is represented as VL-VH, in order to illustrate other alternative non-limiting combinations of components.

[0383] In some embodiments, the precursor trispecific antibody construct described herein comprises two polypeptides, polypeptide A and polypeptide B, wherein polypeptide A comprises components having an N-terminal to C-terminal order: CAP domain, protease cleavage domain, CD3 third binding domain VH-CH1 region, first binding domain (VL-VH); and polypeptide B comprises components having an N-terminal to C-terminal order: HLP component, protease cleavage domain, CD3 third binding domain VL-CL region, second binding domain (VL-VH). In some embodiments, the precursor trispecific antibody construct described herein comprises the following two polypeptides (in order from N-terminal to C-terminal):

[0384] Peptide A: CAP-L-CP-L-CD3VH-L-CD3CH1-L-TAA VL-L-TAA VH

[0385] Polypeptide B: HLP-L-CP-L-CD3VL-L-CD3CL-L-TAA VL-L-TAA VH,

[0386] wherein "L" is a linker, which may or may not be present in each embodiment.

[0387] The above constructs provide non-limiting examples of trispecific precursor antibody constructs in which the positions of the two regulatory domains are alternated. The skilled artisan will appreciate that these alternative positions can also be included in other embodiments in which the regulatory domain positions are indicated as HLP domains being components of the HC and CAP being components of the LC, and the scFv being indicated as VL-VH, in order to illustrate other alternative, non-limiting combinations of components.

[0388] In some embodiments, the precursor trispecific antibody construct described herein comprises two polypeptides, polypeptide A and polypeptide B, wherein polypeptide A comprises components in the order from N-terminus to C-terminus: HLP domain, protease cleavage domain, CD3 third binding domain VL-CL region, first binding domain (VL-VH); and polypeptide B comprises components in the order from N-terminus to C-terminus: CAP component, protease cleavage domain, CD3 third binding domain VH-CH1 region, second binding domain (VL-VH). In some embodiments, the precursor trispecific antibody construct described herein comprises the following two polypeptides (in the order from N-terminus to C-terminus):

[0389] Peptide A: HLP-L-CP-L-CD3VL-L-CD3CL-L-TAA VH-L-TAA VL

[0390] Peptide B: CAP-L-CP-L-CD3VH-L-CD3CH1-L-TAA VH-L-TAA VL,

[0391] wherein "L" is a linker, which may or may not be present in each embodiment.

[0392] Those skilled in the art will understand that the names "Polypeptide A" and "Polypeptide B" are merely names for two heterologous polypeptide chains, and thus the names themselves can be interchanged or changed, for example, Polypeptide 1 and Polypeptide 2. Further, the term encompasses two structurally different polypeptide chains that together form a precursor trispecific antibody construct, as described herein. Further, the skilled artisan will understand the modular nature of the precursor constructs described herein, where modules can be substituted for one another, where they provide similar or different activities. For example, but not limited to, the ScFv may necessarily be ordered N-terminus to C-terminus VL-VH or VH-VL; or the regulatory domain may include a CAP module or an HLP domain.

[0393] The order of the components can be alternated as shown in the non-limiting examples above, where in some embodiments, the first or second sub-regulatory domain is a component of either the HC or LC and can be located N-terminally to the third binding domain, and the first or second binding domain comprises an scFv, wherein the components of the scFv (VL, L, and VH) are independently ordered VL-L-VH or VH-L-VL, which are located C-terminally to the third binding domain and are components of either the HC or LC, respectively.

[0394] The skilled person will appreciate that the precursor construct may include different regulatory domains depending on the desired functionality of the precursor construct. For example, a trispecific precursor construct may comprise only a single regulatory domain, either the HLP domain or the CAP domain (see, respectively). Figure 2C and 2D ).

[0395] In certain embodiments, the first or second binding domain, or both the first and second binding domains, or the first or second sub-regulatory domain, or both the first and second sub-regulatory domains, or a combination thereof, is directly linked to the respective ends of the VH-CH1 or VL-CL of the third binding domain (e.g., Fab) (i.e., without the addition of additional amino acids therebetween). In other embodiments, the connection to the third binding domain (e.g., Fab) comprises the use of a linker as described above (with additional amino acids as described below). In some embodiments, it may be necessary to delete several amino acids (e.g., 1-3 amino acids or 1-10 amino acids; e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acids) from the C-terminus of a given first or second binding domain or both, and / or first or second sub-regulatory domain or both, depending on the space surrounding the third binding domain target and the first and second binding domain targets on the cell surface (i.e., for example, the accessibility of the CD3ε target on the surface of T cells).

[0396] In other embodiments, it may be desirable to delete several amino acids (e.g., 1-3 amino acids or 1-10 amino acids; e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acids) from the N-terminus of the heavy and / or light chain of the third binding domain. In still further embodiments, it may be desirable to delete several amino acids (e.g., 1-3 amino acids or 1-10 amino acids; e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acids) from the N-terminus of the first or second binding domain, or both, and / or from the C-terminus of the first or second subregulatory domain, or both, and simultaneously delete several amino acids (e.g., 1-3 amino acids or 1-10 amino acids; e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acids) from the N-terminus and / or C-terminus of the third binding domain chain (VH-CH1 or VL-CL). The length and sequence of the connection between the first or second binding domain and / or the first or second subregulatory domain and the third binding domain VH-CH1 and VL-CL chain can be the same or different.

[0397] The connection between the first or second binding domain or both, and / or the first or second sub-regulatory domain or both and the third binding domain VH-CH1 and VL-CL chain can be made using a combination of deletions and linkers as needed. As will be appreciated by those skilled in the art, the connection between the third binding domain VH-CH1 and VL-CL chain and the first or second binding domain or both and / or the first or second sub-regulatory domain or both can be adjusted accordingly and tested for desired functionality (e.g., binding affinity, T cell activity) using methods known in the art and described herein.

[0398] As described herein, the connections between domains or between components within a domain include joints. In some embodiments, there are joints between domains. In some embodiments, there are no joints between domains. In some embodiments, there are joints between components that comprise a domain. In some embodiments, there are no joints between components that comprise a domain.

[0399] Figure 1 and 2A - 2D and 2F illustrate where linkers may be present in embodiments of the precursor trispecific antibody constructs disclosed herein.

[0400] In some embodiments, the length of the linker between the first or second binding domain or both and the third binding domain VH-CH1 or VL-CL is 1-10 amino acids. In other embodiments, the length of the linker between the first or second binding domain or both and the third binding domain VH-CH1 or VL-CL is 1-20 or 20 amino acids. In this regard, the length of the linker can be 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 amino acids. In further embodiments, the length of the linker can be 21, 22, 23, 24, 25, 26, 27, 28, 29 or 30 amino acids.

[0401] In some embodiments, the length of the linker between the first or second sub-regulatory domain or both and the third binding domain VH-CH1 or VL-CL is 1-10 amino acids. In other embodiments, the length of the linker between the first or second sub-regulatory domain or both and the third binding domain VH-CH1 or VL-CL is 1-20 or 20 amino acids. In this regard, the length of the linker can be 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 amino acids. In further embodiments, the length of the linker can be 21, 22, 23, 24, 25, 26, 27, 28, 29 or 30 amino acids.

[0402] In some embodiments, the length of the linker between the first or second binding domain or components within both is 1-10 amino acids. In other embodiments, the length of the linker between the first or second binding domain or components within both is 1-20 or 20 amino acids. In this regard, the length of the linker between components can be 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acids. In further embodiments, the length of the linker between components can be 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30 amino acids.

[0403] In some embodiments, the length of the joint between the first or second sub-regulatory structural domain or the components in the two is 1-10 amino acid, and it is wherein understood that the joint between different components does not need to have the same length. In other embodiments, the length of the joint between the first or second sub-regulatory structural domain or the components in the two is 1-20 or 20 amino acid, and it is wherein understood that the joint between different components does not need to have the same length. In this regard, the length of the joint between each group of components can be 1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19 or 20 amino acid. In a further embodiment, the length of the joint between each group of components can be 21,22,23,24,25,26,27,28,29 or 30 amino acid.

[0404] In some embodiments, linkers are present between the components within the VH and CH1 polypeptide modules of the HC or LC Fab and between the components within the VL and CL polypeptide modules, respectively. In some embodiments, the linkers are 1-10 amino acids in length, with the understanding that linkers between different components need not be of the same length. In other embodiments, the linkers between components within the third binding domain are 1-20 or 20 amino acids in length, with the understanding that linkers between different components need not be of the same length. In this regard, the linkers between each set of components can be 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acids in length. In further embodiments, the linkers between each set of components can be 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30 amino acids in length.

[0405] In some embodiments, the linker that is applicable to precursor construct as herein described is a flexible linker. Suitable linkers can be easily selected and can have any suitable different lengths, for example from 1 amino acid (for example Gly) to 20 amino acids, from 2 amino acids to 15 amino acids, from 3 amino acids to 12 amino acids, including 4 amino acids to 10 amino acids, 5 amino acids to 9 amino acids, 6 amino acids to 8 amino acids or 7 amino acids to 8 amino acids, can be 1, 2, 3, 4, 5, 6 or 7 amino acids.

[0406] In some embodiments, flexible linkers include glycine polymers (G)n, glycine-serine polymers (including, for example, (GS)n, (GSGGS)n (SEQ ID NO: 119), and (GGGS)n (SEQ ID NO: 120), where n is an integer of at least 1), glycine-alanine polymers, alanine-serine polymers, and other flexible linkers known in the art. Glycine and glycine-serine polymers are relatively unstructured and therefore may be able to act as neutral tethers between components. Glycine is significantly closer to the phi-psi space than alanine and is much less constrained than residues with longer side chains (see Scheraga, Rev. Computational Chem. 11173-142 (1992)). In some embodiments, flexible linkers include, but are not limited to, Gly-Gly-Ser-Gly (GGSG; SEQ ID NO: 121), Gly-Gly-Ser-Gly-Gly (GGSGG; SEQ ID NO: 122), Gly-Ser-Gly-Ser-Gly (GSGSG; SEQ ID NO: 123), Gly-Ser-Gly-Gly-Gly (GSGGG; SEQ ID NO: 124), Gly-Gly-Gly-Ser-Gly (GGGSG; SEQ ID NO: 125), Gly-Ser-Ser-Ser-Gly (GSSSG; SEQ ID NO: 126), etc. One of ordinary skill will recognize that the design of the precursor trispecific antibody construct may include a fully or partially flexible linker, such that the linker may include a flexible linker and one or more portions that confer a less flexible structure to provide the desired precursor trispecific antibody construct structure.

[0407] In some embodiments, the flexible linker used in the precursor construct includes any flexible linker known in the art. In some embodiments, the flexible linker includes a flexible unstructured linker. At least one linker known in the art is described in: Chengcheng Liu, Ju Xin Chin, Dong-Yup Lee; SynLinker: an integrated system for designing linkers and synthetic fusion proteins, Bioinformatics, Volume 31, Issue 22, 15 November 2015, Pages 3700–3702; Fusion protein linkers: property, design and functionality Chen X et al, Adv Drug Deliv Rev. 2013 Oct; 65(10): 1357-69; The Linker Data base provided by The Centre for Integrative Bioinformatics vrije Universiteit Amsterdam ( http: / / www.ibi.vu.nl / programs / linkerdbwww ); and the CSD Linker Database provided by The Cambridge Crystallographic Data Centre (https: / / www.ccdc.cam.ac.uk / solutions / partnersoftware / csdlinkerdatabase / ).

[0408] In certain embodiments, the linker between the third binding domain and the first or second binding domain or both, or the first or second sub-regulatory domain or both (binding and regulatory domains) is a stable linker (not cleavable by proteases, particularly MMPs). In certain embodiments, the linker is a peptide linker.

[0409] In some embodiments, the linker between the third binding domain VH-CH1 or VL-CL chain and the first or second subregulatory domain, or both, comprises a protease substrate cleavage sequence, such as an MMP substrate cleavage sequence. In some embodiments, the linker between the third binding domain VH-CH1 or VL-CL chain and the first or second subregulatory domain, or both, comprises a protease substrate cleavage sequence, such as an MMP2 / 9, uPA, matriptase, and asparagine endopeptidase substrate cleavage sequence. The peptide sequence of SEQ ID NO: 9 in the substrate can be cleaved by most MMPs. The peptide sequence of SEQ ID NO: 35 in the substrate can be cleaved by MMP2 / 9, uPA, matriptase, and asparagine endopeptidase.

[0410] A protease substrate cleavage sequence refers to a peptide sequence that can be cleaved by protease treatment. An MMP substrate sequence refers to a peptide sequence that can be cleaved by incubation with an MMP. SEQ ID NO:9 is a commonly used MMP substrate cleavage sequence (see, e.g., Jiang, PNAS (2004) 101:17867-72; Olson, PNAS (2010) 107:4311-6). In another embodiment, the protease cleavage site is recognized by MMP-2, MMP-9, or a combination thereof. In yet another embodiment, the protease site comprises a sequence selected from (SEQ ID NO:9, SEQ ID NO:10, SEQ ID NO:11, and SEQ ID NO:35). In a further embodiment, the protease site comprises the sequence recited in SEQ ID NO:35.

[0411] In some embodiments, all protease sites comprise identical proteolytic sequences. In other embodiments, the protease sites of the precursor construct are different. The difference of the proteolytic sequences in the precursor construct can provide additional regulation of the function of the precursor or partially activated construct.

[0412] A stable linker or protease non-cleavable linker refers to a linker peptide sequence that is not among the known protease substrate sequences and therefore does not result in the formation of significant cleavage products upon incubation with the protease.

[0413] In some embodiments, the cleavage substrate (or cleavage sequence or protease cleavage domain) of the linker can include an amino acid sequence that can serve as a substrate for a protease (typically an extracellular protease). In other embodiments, the cleavage sequence comprises a cysteine-cysteine ​​pair that can form a disulfide bond, which can be cleaved by the action of a reducing agent. In other embodiments, the cleavage sequence comprises a substrate that can be cleaved upon photolysis.

[0414] The cleavage substrate is positioned in the linker such that when the cleavage substrate is cleaved by a cleavage agent (e.g., the cleavage substrate of the linker is cleaved by a protease and / or the cysteine-cysteine ​​disulfide bond is broken via reduction by exposure to a reducing agent) or by light-induced photolysis, in the presence of the target, a cleavage product is generated having various functional properties as described herein.

[0415] The cleavage substrate of the linker can be selected based on a protease that is co-localized in the diseased tissue or on the surface of a cell expressing the target antigen of interest of the binding domain of the fusion moiety. A variety of different conditions are known in which the target of interest is co-localized with a protease, wherein substrates of the protease are known in the art. In the example of cancer, the target tissue can be cancerous tissue, in particular cancerous tissue of a solid tumor. Increased levels of proteases with known substrates have been reported in the literature in many cancers (e.g., solid tumors) (see, e.g., La Rocca et al., (2004) British J. of Cancer 90(7): 1414-1421). Non-limiting examples of diseases include: all types of cancer (breast cancer, lung cancer, colorectal cancer, prostate cancer, head and neck cancer, pancreatic cancer, etc.), rheumatoid arthritis, Crohn's disease, melanoma, SLE, cardiovascular damage, local ischemia, etc. In addition, anti-angiogenic targets, such as VEGF, are known. Thus, when the binding domain of the fusion portion of the precursor trispecific antibody construct of the present disclosure is selected so that it is capable of binding to a TAA, a suitable cleavage substrate sequence for the protease-cleavable linker will be a cleavage substrate sequence comprising a peptide substrate that can be cleaved by a protease that is present at a cancerous treatment site, particularly at elevated levels compared to non-cancerous tissue.

[0416] In some embodiments, the first or second binding domain of the precursor trispecific antibody construct or both can bind to, for example, Her2, and the cleavage substrate sequence can be a matrix metalloproteinase (MMP) substrate and can therefore be cleaved by MMP. In other embodiments, the first or second binding domain of the fusion portion in the precursor trispecific antibody construct or both can bind to a target of interest or two targets of interest, and the cleavage substrate present in the linker can be, for example, asparagine endopeptidase, plasmin, TMPRSS-3 / 4, MMP-9, MT1-MMP, cathepsin, caspase, human neutrophil elastase, β-secretase, uPA or PSA. In other embodiments, the first or second binding domain of the fusion portion in the precursor trispecific antibody construct or both can bind to a target of interest or two targets of interest, and the cleavage substrate present in the linker can be, for example, a combination of MMP2 / 9, asparagine endopeptidase, uPA and protein cleavage enzymes. In some embodiments, the first or second binding domain of the fusion moiety, or both, in the precursor trispecific antibody construct binds to a target of interest or two targets of interest, and the cleavage substrate present in the linker comprises a combination of MMP2 / 9, asparagine endopeptidase, uPA, and a protease, as set forth in SEQ ID NO: 35. In other embodiments, the cleavage substrate is cleaved by other disease-specific proteases in diseases other than cancer, such as multiple sclerosis or rheumatoid arthritis.

[0417] An unmodified or uncleaved linker can allow for tethering of the binding domain (first, second, or third, or a combination thereof) and the regulatory domain (first or second, or both).

[0418] The linkers of the precursor trispecific antibody construct (e.g., the linker between the CH1 or CL of the third binding domain and the first or second binding domain or both, and the linker between the VH or VL of the third binding domain and the first or second subregulatory domain or both) may comprise the same cleavage substrate or may comprise different cleavage substrates, for example, the first linker may comprise a first cleavage substrate and the second linker may comprise a second cleavage substrate, etc. The cleavage substrates may be different substrates of the same enzyme (e.g., exhibiting different binding affinities for the enzyme), or different substrates of different enzymes, or one of the cleavage substrates may be an enzyme substrate and the other of the cleavage substrates may be a photolysis substrate, or the other of the cleavage substrates may be a substrate for reduction, or any combination thereof.

[0419] In some embodiments, some linkers may be non-cleavable, while other linkers are cleavable linkers. For example, but not limited to, the linkers between Fab CH1 and CL and the first and second binding domains are non-cle...

Claims

1. A precursor trispecific antibody construct consisting of two polypeptides consisting of: (a) a first binding domain comprising a single chain variable fragment (scFv) that binds to the tumor associated antigen (TAA) 5T4, wherein the scFv sequence consists of SEQ ID NO: 172 or 174; (b) a second binding domain comprising a single chain variable fragment (scFv) that binds to the TAA 5T4, wherein the scFv sequence consists of SEQ ID NO: 172 or 174; (c) a third binding domain comprising a Fab antigen-binding fragment that binds to an extracellular epitope of human CD3ε, wherein each of the first binding domain and the second binding domain is located at the C-terminus of the third binding domain; and (d) a regulatory domain located at the N-terminus of the third binding domain, the regulatory domain consisting of a protease cleavage domain, a half-life extension (HLP) domain and a CAP component, the protease cleavage domain being a combined substrate cleavable by one or more of MMP2 / 9, uPA, matriptase and asparagine endopeptidase or any combination thereof, the half-life extension (HLP) domain consisting of a human serum albumin (HSA) polypeptide having an amino acid sequence as recited in SEQ ID NO:6, the CAP component consisting of an amino acid sequence as recited in SEQ ID NO:5, wherein the precursor trispecific antibody construct has a structure selected from the group consisting of: (i) heavy chain: N'-human serum albumin (HSA)-linker-protease cleavable peptide MMP2 / 9-linker-(anti-CD3εVH1-linker-CH1)-linker-(anti-5T4VL3-linker-VH3)-linker-C', and light chain: N'-CAP-linker-protease cleavable peptide MMP2 / 9-linker-(anti-CD3εVL1-linker-CL1)-linker-(anti-5T4VL2-linker-VH2)-linker-C'; and (ii) Heavy chain: N'-human serum albumin (HSA)-linker-protease cleavable peptide MMP2 / 9-linker-(anti-CD3εVH1-linker-CH1)-linker-(anti-5T4 VH3-linker-VL3)-linker-C', and light chain: N'-CAP-linker-protease cleavable peptide MMP2 / 9-linker-(anti-CD3εVL1-linker-CL1)-linker-(anti-5T4 VH2-linker-VL2)-linker-C'.

2. The precursor trispecific antibody construct according to claim 1, wherein the protease cleavage domain comprises the amino acid sequence recited in SEQ ID NO:

35.

3. A pharmaceutical composition comprising the precursor trispecific antibody construct according to any one of claims 1-2 and a pharmaceutically acceptable carrier.

4. A nucleic acid construct comprising one or more nucleic acid sequences encoding the precursor trispecific antibody construct according to any one of claims 1-2. An expression vector comprising the nucleic acid construct according to claim 4 .

6. An isolated host cell comprising the expression vector of claim 5, wherein the host cell is selected from the group consisting of a prokaryotic cell, a fungal cell, a yeast cell, and a mammalian cell.

7. Use of the precursor trispecific antibody construct according to any one of claims 1-2 or the pharmaceutical composition according to claim 3 in the preparation of a medicament for treating, preventing, inhibiting growth, delaying disease progression, reducing tumor burden or reducing the incidence of a cancer or tumor known to express TAA 5T4 in a subject in need thereof, or any combination thereof, wherein the cancer or tumor is a solid tumor and metastatic tumor selected from breast tumors, prostate cancer, brain tumors, pancreatic tumors, colorectal tumors, gastric cancer, ovarian cancer, NSCLC and head and neck cancer.

8. The use according to claim 7, wherein the cancer or tumor is colon cancer.

Images