Precursor trispecific antibody constructs and methods of use thereof

Abstract

The name of the invention is precursor trispecific antibody constructs and methods of use thereof. Described herein are precursor trispecific antibody constructs comprising at least one tumor-associated antigen binding domain, a T cell binding domain, and a regulatory domain having an enhanced half-life. Further, methods of producing the precursor trispecific antibody constructs are disclosed. Pharmaceutical compositions comprising the precursor constructs and their use for the treatment of tumors are also disclosed.

Description

[0001] This application is a divisional application. The filing date of the original application is March 26, 2020, the application number is 2020800340187, and the invention title is "Precursor Trispecific Antibody Constructs and Methods of Use Thereof".

[0002] Cross - reference to related applications

[0003] This application claims the benefit of U.S. Serial No. 62 / 844,303, filed May 7, 2019. The entire content and disclosure of the prior application are incorporated herein by reference in their entirety.

[0004] Sequence listing statement

[0005] This application contains a sequence listing, which has been electronically submitted in ASCII format and is incorporated herein by reference in its entirety. The ASCII copy was created on March 25, 2020, named P - 586968 - PC - 25MAR2020.txt, and is 279 Kb in size. Technical field

[0006] Disclosed herein are precursor trispecific antibody constructs and methods of use for these precursor constructs. The methods of use include use for treating cancer, wherein the precursor constructs include prodrugs having tumor - restricted activation and multiple antigen - binding sites. Background art

[0007] The functions of monoclonal antibodies (unconjugated or naked antibodies) currently approved by global drug regulatory agencies for clinical use in oncology settings are known to use one or a combination of the following mechanisms: 1) blocking cell growth signal transduction, 2) blocking blood supply to cancer cells, 3) directly mediating apoptosis, 4) triggering immune effector functions such as antibody - dependent cell cytotoxicity (ADCC), antibody - dependent cell phagocytosis (ADCP), and complement - dependent cytotoxicity (CDC), and 5) promoting adaptive immunity against tumors.

[0008] Monoclonal antibody therapies have demonstrated survival benefits in clinical settings. However, compared to chemotherapy, the overall response rate in cancer patients is lower, and the survival benefit is marginal (a few months). Although the exact underlying reasons for the lack of robust clinical anti - cancer activity are 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 thought to be the effective cancer - initiating cells, have lower activity in cell proliferation, and thus they tend to better maintain a lack of growth signals.

[0009] Multispecific antibodies are being developed to improve the anti-tumor activity of monoclonal antibodies. Different from classical monoclonal antibodies (the standard first-line therapy for several tumor entities), these multispecific antibodies can combine tumor cells with means to destroy tumor cells, thereby improving the treatment efficiency. These multispecific antibodies provide new treatment options for cancer patients.

[0010] Another anti-cancer therapeutic approach is to utilize T cells. T cells provide defense against cancer throughout life by patrolling the body to seek out newly emerging cancer cells and eliminating them effectively and rapidly. Therapeutic approaches using T cells have been shown to be successful at least in the cancer treatment of metastatic melanoma, metastatic renal carcinoma, asymptomatic metastatic hormone-refractory prostate cancer, and advanced melanoma.

[0011] Another consideration for tumor cytotoxicity is the tumor microenvironment (TME). The TME includes new targets that can help guide and improve the action of antibody therapies by enhancing the host anti-tumor immune response. For example, T cells play an unexpectedly crucial role in anti-tumor antigen antibody therapies, although their importance is often not observed due to studies conducted in immunodeficient mice.

[0012] 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 such binding results in cytotoxicity. Thus, the "off-target" binding of multispecific and bispecific antibodies poses potential challenges 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 disadvantage of many bispecific or multispecific antibodies is their short half-life.

[0013] There is still a need to provide multispecific trivalent antibodies that have cytotoxicity specifically targeted to tumor cells while reducing toxic side effects and maintaining the efficacy of the antibodies. Reducing the non-specific toxic side effects of multispecific antibodies and simultaneously enhancing the efficacy of these antibodies requires antibodies in the following precursor forms: (1) that conjugate to targets associated with tumor cells, tumor-associated cells, or the tumor cell environment, and (2) that activate cytotoxic cells, such as T cells, once localized to the tumor microenvironment. Further, it is necessary that such multispecific antibodies do not significantly reduce the immunogenicity against 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 conjugating to cells that provide toxicity prior to the antibody binding to the tumor or tumor-associated target. Summary of the Invention

[0014] In one aspect, the present disclosure provides 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; 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 the extracellular epitope of human CD3ε.

[0015] In one aspect, the present disclosure provides 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 or a second sub-regulatory domain, the first sub-regulatory domain comprising a first protease cleavage domain and a half-life extension (HLP) domain, the second sub-regulatory domain comprising a second protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to the 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 the extracellular epitope of CD3ε.

[0016] In one aspect, the present disclosure provides 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 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 the extracellular epitope of human CD3ε.

[0017] 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.

[0018] In a related aspect, 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 angiogenesis antigen in the TME, an antigen on blood vessels in the TME, or a cytokine antigen in the TME, or any combination thereof. In another related aspect, the TAA can be one of the following: 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, 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, glypicans, EphA2, CD138, B7-H3, B7-H4, gpA33, GPC3, SSTR2, ROR1, 5T4, and VEGF-R2 or any combination thereof. In a further related aspect, the TAAs bound by the first binding domain or the second binding domain or both are selected from 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 NO: 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.

[0019] In another related aspect, the tumor microenvironment antigen can be a KIR, LILR or TIGIT antigen. In another related aspect, the stromal antigen in the tumor microenvironment can be fibroblast activation protein (FAP), α-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 blood vessels in the tumor microenvironment comprises endothelial cell surface antigens 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.

[0020] In a related aspect, the HLP domain comprises a human serum albumin (HSA) polypeptide.

[0021] In a related aspect, the CAP component of the second sub-regulatory domain comprises the amino acid sequence of the extracellular epitope of human CD3ε. In another related aspect, the amino acid sequence of the CAP component is described in SEQ ID NO:5 or its homolog.

[0022] In a related aspect, each of the first binding domain, the second binding domain or both comprises a single-chain variable fragment (scFv). In another related aspect, the third binding domain comprises a Fab antigen-binding fragment.

[0023] 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 can be cleaved by a serine protease, a cysteine protease, an aspartic protease, or a matrix metalloproteinase (MMP), or is a combinatorial substrate that is cleaved by one or more of MMP2 / 9, uPA, cathepsin, and asparaginyl endopeptidase, 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 the combinatorial substrate that is cleaved by one or more of MMP2 / 9, uPA, cathepsin, and asparaginyl endopeptidase is recited in SEQ ID NO:35. In a further related aspect, the first and / or second protease cleavage domain includes a 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 still another aspect, the first and / or second protease cleavage domain comprises the sequence recited in SEQ ID NO:35.

[0024] 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 assembly 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 assembly located at the N-terminus of the protease cleavage domain is located at the N-terminus of the VL3 region.

[0025] In another aspect, the precursor trispecific antibody construct comprises two polypeptides - polypeptide A and polypeptide B, wherein polypeptide A comprises components having the following N-terminal to C-terminal sequence: HLP domain, protease cleavage domain, third binding domain (VH3 region), first binding domain (VL-VH); and polypeptide B comprises components having the following N-terminal to C-terminal sequence: CAP component, protease cleavage domain, third binding domain (VL region), second binding domain (VL-VH); or polypeptide A comprises components having the following N-terminal to C-terminal sequence: HLP domain, protease cleavage domain, third binding domain (VH3 region), first binding domain (VH-VL); and polypeptide B comprises components having the following N-terminal to C-terminal sequence: CAP component, protease cleavage domain, third binding domain (VL region), second binding domain (VH-VL); or polypeptide A comprises components having the following N-terminal to C-terminal sequence: HLP domain, protease cleavage domain, third binding domain (VL3 region), first binding domain (VL-VH); and polypeptide B comprises components having the following N-terminal to C-terminal sequence: CAP component, protease cleavage domain, third binding domain (VH3 region), second binding domain (VL-VH); or polypeptide A comprises components having the following N-terminal to C-terminal sequence: HLP domain, protease cleavage domain, third binding domain (VL3 region), first binding domain (VH-VL); and polypeptide B comprises components having the following N-terminal to C-terminal sequence: CAP component, protease cleavage domain, third binding domain (VH region), second binding domain (VH-VL); or polypeptide A comprises components having the following N-terminal to C-terminal sequence: HLP domain, protease cleavage domain, third binding domain (VL3 region), first binding domain (VL-VH); and polypeptide B comprises components having the following N-terminal to C-terminal sequence: CAP component, protease cleavage domain, third binding domain (VH region), second binding domain (VH-VL); or polypeptide A comprises components having the following N-terminal to C-terminal sequence: HLP domain, protease cleavage domain, third binding domain (VL3 region), first binding domain (VH-VL); and polypeptide B comprises components having the following N-terminal to C-terminal sequence: CAP component, protease cleavage domain, third binding domain (VH3 region), second binding domain (VH-VL); or polypeptide A comprises components having the following N-terminal to C-terminal sequence: HLP domain, protease cleavage domain, third binding domain (VH3 region), first binding domain (VL-VH); and polypeptide B comprises components having the following N-terminal to C-terminal sequence: CAP component, protease cleavage domain, third binding domain (VL3 region), second binding domain (VL-VH);Alternatively, polypeptide A comprises components having the following N-terminal to C-terminal sequence: HLP domain, protease cleavage domain, third binding domain (VH3 region), first binding domain (VH-VL); and polypeptide B comprises components having the following N-terminal to C-terminal sequence: CAP component, protease cleavage domain, third binding domain (VL3 region), second binding domain (VH-VL).;

[0026] In a related aspect, the third binding domain comprises a VL3 region and a VH3 region, wherein the VL3 region comprises CDR-Ll (selected from SEQ ID NO: 107-109), CDR-L2 (SEQ ID NO: 110), and CDR-L3 (selected from SEQ ID NO: 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 recited in any one of SEQ ID NO: 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 recited in any one of SEQ ID NO: 46-72 and 114, or an amino acid sequence having at least 80% homology thereto.

[0027] In one aspect, disclosed herein is a pharmaceutical composition comprising a precursor trispecific antibody construct and a pharmaceutically acceptable carrier, 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 that comprises a protease cleavage domain and a half-life prolonging (HLP) domain; and a second sub-regulatory domain that comprises a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to the extracellular epitope of human CD3ε.

[0028] 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, 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 that comprises a protease cleavage domain and a half-life prolonging (HLP) domain; and a second sub-regulatory domain that comprises a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to the extracellular epitope of human CD3ε.

[0029] In one aspect, the present disclosure provides an expression vector comprising one or more nucleic acid constructs encoding one or more polypeptides of the precursor trispecific antibody constructs disclosed herein.

[0030] In one aspect, the present disclosure provides an isolated host cell comprising a nucleic acid sequence or 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 prolongation (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 the extracellular epitope of human CD3ε.

[0031] In one aspect, the present disclosure provides a method of treating, preventing, inhibiting growth, delaying disease progression, reducing tumor burden, or reducing the incidence of cancer or tumors, or any combination thereof, in a subject in need thereof, the method comprising 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 prolongation (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 the extracellular epitope of human CD3ε; wherein the method treats, prevents, inhibits growth, delays disease progression, reduces tumor burden, or reduces the incidence of cancer or tumors in the subject.

[0032] In a related aspect, compared to a subject to whom the pharmaceutical compositions disclosed herein have not been administered, using the methods disclosed herein for treating a subject in need thereof reduces minimal residual disease, improves remission, improves 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. In another related aspect, the cancer or tumor comprises a solid tumor or a non-solid tumor, or wherein the cancer or tumor comprises metastases of the cancer or tumor. In a further related aspect, the non-solid cancer or tumor can be a hematological malignancy, blood cell cancer, leukemia, myelodysplastic syndrome, 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, fibrosarcoma, myxosarcoma, liposarcoma, chondrosarcoma, osteosarcoma, chordoma, angiosarcoma, endotheliosarcoma, lymphangiosarcoma, lymphangioendotheliosarcoma, synovioma, mesothelioma, Ewing sarcoma, leiomyosarcoma, rhabdomyosarcoma, colon cancer or tumor, 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, cholangiocarcinoma, 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.

[0033] In one aspect, methods are disclosed herein for treating, preventing, inhibiting growth, delaying disease progression, reducing tumor burden, or reducing the incidence of 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 constructs comprising a nucleic acid sequence or sequences encoding a precursor trispecific antibody construct, the precursor trispecific antibody construct comprising: a first binding domain that binds a first tumor-associated antigen (TAA); a second binding domain that binds a second tumor-associated antigen (TAA); a third binding domain that binds an extracellular epitope of human CD3ε; a first sub-regulatory domain that comprises a protease cleavage domain and a half-life prolongation (HLP) domain; and a second sub-regulatory domain that comprises a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind the extracellular epitope of human CD3ε; wherein the method treats, prevents, inhibits growth, delays disease progression, reduces tumor burden, or reduces the incidence of cancer or tumor in the subject.

[0034] In relevant aspects, compared to subjects to whom the pharmaceutical composition has not been administered, the methods disclosed herein for treating a subject in need reduce minimal residual disease, increase remission, increase duration of remission, decrease tumor recurrence rate, prevent metastasis of a tumor or cancer, or decrease the metastasis rate of a tumor or cancer, or any combination thereof.

[0035] In one aspect, the present disclosure provides a method for generating 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 that comprises a protease cleavage domain and a half-life prolonging (HLP) domain; and a second sub-regulatory domain that comprises a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to the extracellular epitope of human CD3ε; the method comprising the steps of: culturing a host cell comprising nucleic acid sequences 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.

[0036] In relevant aspects, in the method for generating a precursor trispecific antibody construct, the expression comprises expressing from the same host cell or comprising two host cells that each separately express different polypeptides (polypeptide A and polypeptide B). BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The subject matter regarded as the precursor trispecific (trisome) antibody construct disclosed herein is particularly pointed out and distinctly claimed in the concluding portion of the specification. However, the precursor trispecific (trisome) antibody construct may be best understood when read in conjunction with the accompanying drawings, in which:

[0038] Figure 1 A schematic embodiment of a precursor trispecific (trisome) antibody construct with modular components is shown, with exemplary but not limited to impaired modular components: (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 1Embodiments of the precursor trispecific (trimeric) antibody construct have three antibody binding domains, where the Fab portion recognizes the CD3 surface antigen, and the components and regions of the different domains are identified. In the embodiments shown herein, the precursor trispecific (trimeric) antibody construct is formed from two polypeptides, where each polypeptide includes an anti-tumor associated antigen (TAA) binding domain (first and second binding domains), which is the C-terminus (third binding domain) of the anti-CD3 Fab binding domain on each polypeptide. Further, each polypeptide includes a regulatory domain (first and second sub-regulatory domains) at the N-terminus of the anti-CD3 Fab binding domain, where 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-terminus to C-terminus order of one of the ScFvs is variable light chain region (VL2), followed by linker (L4), followed by variable heavy chain region (VH2), followed by linker (L5), and the N-terminus to C-terminus order of the other ScFv is variable heavy chain region (VH3), followed by linker (L9), followed by variable light chain region (VL3), followed by linker (L10). In embodiments not shown, the N-terminus to C-terminus order of the scFv can be reversed and can be the same or different for the two scFv binding domains. In embodiments not shown, the two sub-regulatory domains can be detached from their current linkage 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 components and between 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 of binding site three, and VH1 is the variable heavy chain region of binding site three. The oval designated as HSA is the human serum albumin component. The shape designated as CP is the cleavage peptide. The triangle is the CAP component.

[0039] Figures 2A to 2F Shows various embodiments of the precursor and active trispecific (trimeric) antibody constructs described herein. Figure 2A and 2B Shows a schematic embodiment of a precursor trispecific antibody construct containing two anti-tumor associated antigen binding domains, where the tumor associated antigen is EGFR, where the N-terminus to C-terminus 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 comprises 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 a 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 the components in the regulatory domain is CAP or HSA-linker-protease-cleavable linker from N-terminus to C-terminus. L1, L2, etc. represent possible linkers between different domains or domain components. The linker may or may not be present. Figure 2C shows a schematic diagram of the precursor trispecific antibody construct as Figure 2B shown, but lacking the regulatory domain containing the half-life extension component (HSA). Figure 2D shows a schematic diagram of the precursor trispecific antibody construct as Figure 2B shown, but lacking the sub-regulatory domain containing the CD3 CAP domain. Figure 2E shows a schematic diagram of the active trispecific (trimeric) antibody construct, wherein Figure 2B the precursor construct is in the active form and lacks the two sub-regulatory domains. Figure 2F shows a schematic diagram of the precursor trispecific antibody construct, wherein the regulatory domain comprises a single regulatory domain that comprises a CAP domain, an HSA sequence, and a protease-cleavable linker on the same polypeptide.

[0040] Figure 3A and 3B shows a flow chart of protease-specific activation within tumor tissue or within the tumor environment of the precursor trispecific antibody construct, wherein T cell engagement and activation are restricted to the tumor site. Figure 3AShows the effect of the cancer (tumor) microenvironment on a precursor trispecific antibody construct. The precursor trispecific antibody construct contains two protease-cleavable domains, one of which is the C-terminus of an HSA half-life-extended 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, known to be rich in proteases secreted by cancer cells, results in protease cleavage and removal of the HSA and CAP regulatory components, where in some embodiments, the CAP contains 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 and activate T cells. If the precursor construct binds to the TAA outside the tumor microenvironment, protease cleavage does not occur and T cell activation does 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 Shows protease-specific cleavage of HSA and CD3 CAP in the cancer microenvironment, followed by T cell activation and binding of the activated trispecific construct to T cells and tumor cells. The precursor trispecific antibody construct remains intact in circulation or when present in normal tissues, with an extended in vivo half-life. 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 has a reduced limited half-life of hours compared to days to weeks for the precursor trispecific antibody construct. (Data not shown). Additionally, in some embodiments, the cleaved and activated trispecific antibody has a smaller size than the precursor construct, which can improve tumor penetration of the activated trispecific antibody.

[0041] Figure 4A and 4B Shows an embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of an activated trispecific (trisome) antibody construct (Construct 1; VLVH) and the optimized nucleotide sequence encoding the heavy chain (HC) of the activated construct. The amino acid sequence is shown from the N-terminus to the C-terminus and the nucleic acid sequence is shown 5' to 3'. Figure 4A Shows an embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of the activated construct, with the N-terminus to C-terminus 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 constant heavy chain region 1 (SEQ ID NO: 113), followed by two labeled cysteine residues (labeled bold and underlined) that can participate in disulfide double bonds, followed by the anti-EGFR scFv VL (SEQ ID NO: 34)-linker (SEQ ID NO: 39)-VH (SEQ ID NO: 37) chain. Figure 4B One embodiment shows an optimized nucleic acid sequence (DNA) encoding the heavy chain (HC) polypeptide of an activated trispecific (trimeric) construct, having the 5' to 3' order and the following components (SEQ ID NO: 150). Encoding Figure 4A The nucleic acid sequence of the component parts of the HC polypeptide shown includes: a linker (such as Figure 2E L7 of) (SEQ ID NO: 154), an anti-CD3ε variable heavy chain and constant heavy chain region 1 (SEQ ID NO: NO: 155), followed by two labeled cysteine residues (labeled bold and underlined) that can participate in disulfide double bonds, followed by the anti-EGFR scFv VL (SEQ ID NO: 36)-linker (SEQ ID NO: 40)-VH (SEQ ID NO: 38) chain.

[0042] Figure 5A and 5B One embodiment shows the amino acid sequence of the light chain (LC) polypeptide of an activated trispecific (trimeric) antibody construct (Construct 1; VLVH) and the optimized nucleotide sequence encoding the light chain (LC) of the activated construct. The amino acid sequence is shown from the N-terminus to the C-terminus, and the nucleic acid sequence is shown 5' to 3'. Figure 5A One embodiment shows the amino acid sequence of the light chain (LC) polypeptide of the activated construct, having the N-terminus to C-terminus order as follows and components: h1F3.1-λLC anti-EGFRVL-linker-VH (SEQ ID NO: 139). Figure 5A The amino acid sequence of the component parts of the LC polypeptide shown includes: a linker (such as Figure 2E L2 of) (SEQ ID NO: 158), an anti-CD3ε variable light chain and λ light chain (SEQ ID NO: 74), followed by a labeled cysteine residue (labeled bold and underlined) that can participate in disulfide double bonds, followed by the anti-EGFR scFv VL (SEQ IDNO: 34)-linker (SEQ ID NO: 40)-VH (SEQ ID NO: 37) chain. Figure 5BShows one embodiment of an optimized nucleic acid sequence (DNA) encoding the light chain (LC) polypeptide of an activated trispecific (trimeric) construct, having a 5' to 3' sequence and the following components (SEQ ID NO:151). Encoding Figure 5A The nucleic acid sequence of the component part of the LC polypeptide shown includes: a linker (e.g., Figure 2E L2) (SEQ ID NO:154), an anti-CD3ε variable light chain and λ light chain region (SEQ ID NO:159), followed by a labeled cysteine residue (labeled bold and underlined), which can participate in a disulfide double bond, followed by an anti-EGFR scFv VL (SEQ ID NO:34)-linker (SEQ ID NO:40)-VH (SEQ ID NO:37) chain.

[0043] Figure 6A And 6B Shows an embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of an activated trispecific (trimeric) 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 the N-terminus to the C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 6A Shows an embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of the activated construct, having an N-terminus to C-terminus sequence and the following components: h1F3.5-G1Fd-(VH-linker-VL) (SEQ ID NO:140). Figure 6A The amino acid sequence of the component part of the HC polypeptide shown includes: a linker (SEQ ID NO:158), an anti-CD3ε variable heavy chain and constant heavy chain region 1 (SEQ ID NO:113), followed by two labeled cysteine residues (labeled 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 6B Shows one embodiment of an optimized nucleic acid sequence (DNA) encoding the heavy chain (HC) polypeptide of an activated trispecific (trimeric) construct, having a 5' to 3' sequence and the following components (SEQ ID NO:152). Encoding Figure 6A The nucleic acid sequence of the component part of the HC polypeptide shown includes: a linker (SEQ ID NO:154), an anti-CD3ε variable heavy chain and constant heavy chain region 1 (SEQ ID NO:155), followed by two labeled cysteine residues (labeled 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)-VH (SEQ ID NO:36) chain.

[0044] Figure 7A and 7B Disclosed is an embodiment of the amino acid sequence of the light chain (LC) polypeptide of an activated trispecific (trimeric) 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 the N-terminus to the C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 7A Disclosed is an embodiment of the amino acid sequence of the light chain (LC) polypeptide of the activated construct, having the N-terminus to C-terminus sequence and the following components: h1F3.1-λLC-anti-EGFR (VH-linker-VL) (SEQ ID NO:141). Figure 7A The amino acid sequences of the component parts of the LC polypeptide shown include: linker (SEQ ID NO:158), anti-CD3ε variable light chain and λ light chain (SEQ ID NO:74), followed by a labeled cysteine residue (marked in bold and underlined) that can participate in a disulfide double bond, followed by the anti-EGFR scFv VH (SEQ ID NO:37)-linker (SEQ ID NO:40)-VL (SEQ ID NO:34) chain. Figure 7B Disclosed is an embodiment of the optimized nucleic acid sequence (DNA) encoding the light chain (LC) polypeptide of the activated trispecific (trimeric) construct, having the 5' to 3' sequence and the following components (SEQID NO:153). Encoding Figure 7A The nucleic acid sequences of the component parts of the LC polypeptide shown include: linker (SEQ ID NO:154), anti-CD3ε variable light chain and λ light chain region (SEQ ID NO:159), followed by a labeled cysteine residue (marked in bold and underlined) that can participate in a disulfide double bond, and then the anti-EGFR scFv VH (SEQ ID NO:38)-linker (SEQ ID NO:40)-VL (SEQ ID NO:36) chain.

[0045] Figure 8A and 8B Disclosed is an embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of a precursor trispecific (trimeric) antibody construct (Construct 3; VLVH) and an optimized nucleotide sequence encoding the heavy chain (HC) of the precursor construct. The amino acid sequence is shown from the N-terminus to the C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 8A Disclosed is an embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of the precursor construct, having the N-terminus to C-terminus sequence 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 sequences of the component parts of the HC polypeptide shown include: 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 constant heavy chain region 1 (SEQ ID NO:113), followed by two labeled cysteine residues (labeled bold and underlined) that can participate in disulfide double bonds, and then an anti-EGFR scFv VL (SEQ ID NO:34)-linker (SEQ ID NO:40)-VH (SEQ ID NO:37) chain. Figure 8B An embodiment of the optimized nucleic acid sequence (DNA) encoding the heavy chain (HC) polypeptide of a precursor trispecific (trimeric) antibody construct is shown, having a 5' to 3' sequence and the following components (SEQ ID NO:142). Encoding Figure 8A The nucleic acid sequences of the component parts of the HC polypeptide shown include: human serum albumin (HSA) (SEQ ID NO:8), an MMP2 / 9 protease-cleavable linker (SEQ ID NO:161 (linker with cleavable sequence and SEQ ID NO:33 (cleavable sequence)), an anti-CD3ε variable heavy chain and constant heavy chain region 1 (SEQ ID NO:155), followed by two labeled cysteine residues (labeled 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.

[0046] Figure 9A and 9B Embodiments of the amino acid sequence of the light chain (LC) polypeptide of a precursor trispecific (trimeric) antibody construct (Construct 3; VLVH) and the optimized nucleotide sequence encoding the light chain (LC) of the precursor construct are shown. The amino acid sequence is shown from the N-terminus to the C-terminus, and the nucleic acid sequence is shown from 5’ to 3’. Figure 9A An embodiment of the amino acid sequence of the light chain (LC) polypeptide of the precursor construct is shown, having an N-terminus to C-terminus sequence and the following components: Cap-h1F3.1-λLC anti-EGFR VL-linker-VH MM2 / 9 cleavage; (SEQ ID NO:131). Figure 9AThe amino acid sequences of the component parts of the LC polypeptide shown include: CAP (SEQ ID NO:5), MMP2 / 9 protease-cleavable linker (SEQ ID NO:160 and SEQ ID NO:9 (cleavable sequence)), anti-CD3ε variable light chain and λ light chain (SEQ ID NO:74), followed by a labeled cysteine residue (labeled bold and underlined) that can participate in a disulfide double bond, followed by the anti-EGFR scFv VL (SEQ ID NO:34)-linker (SEQ ID NO:40)-VH (SEQ ID NO:37) chain. Figure 9B Displays one embodiment of an optimized nucleic acid sequence (DNA) encoding the light chain (LC) polypeptide of a precursor trispecific (trimeric) antibody construct, having a 5' to 3' sequence and the following components (SEQ ID NO:143). Encoding Figure 9A The nucleic acid sequences of the component parts of the LC polypeptide shown include: 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 λ light chain region (SEQ ID NO:159), followed by a labeled cysteine residue (labeled bold and underlined) that can participate in a disulfide double bond, followed by the anti-EGFR scFv VL (SEQ ID NO:36)-linker (SEQ ID NO:40)-VH (SEQ ID NO:38) chain.

[0047] Figure 10A and 10B Displays embodiments of the amino acid sequence of the heavy chain (HC) polypeptide of a precursor trispecific (trimeric) antibody construct (Construct 4; VHVL) and the optimized nucleotide sequence encoding the heavy chain (HC) of the precursor construct. The amino acid sequence is shown from the N-terminus to the C-terminus and the nucleic acid sequence is shown 5’ to 3’. Figure 10A Displays one embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of the precursor construct, having an N-terminus to C-terminus sequence 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 part)), the variable heavy chain and constant heavy chain region 1 of anti-CD3ε (SEQ ID NO:113), followed by two labeled cysteine residues (labeled bold and underlined) that 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 10B Shows an embodiment of an optimized nucleic acid sequence (DNA) encoding the heavy chain (HC) polypeptide of a precursor trispecific (trimeric) antibody construct, having a 5' to 3' sequence and the following components (SEQ ID NO:144). Encoding 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)), the variable heavy chain and constant heavy chain region 1 of anti-CD3ε (SEQ ID NO:155), followed by two labeled cysteine residues (labeled bold and underlined) that can participate in disulfide double bonds, followed by the anti-EGFR scFv VH (SEQ ID NO:38)-linker (SEQ ID NO:40)-VL (SEQ ID NO:36) chain.

[0048] Figure 11A and 11B Shows embodiments of the amino acid sequence of the light chain (LC) polypeptide of a precursor trispecific (trimeric) antibody construct (Construct 4; VLVH) and the optimized nucleotide sequence encoding the light chain (LC) of the precursor construct. The amino acid sequence is shown from the N-terminus to the C-terminus and the nucleic acid sequence is shown from 5’ to 3’. Figure 11A Shows an embodiment of the amino acid sequence of the light chain (LC) polypeptide of a precursor construct, having an N-terminus to C-terminus sequence 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 sequences of the component parts of the LC polypeptide shown include: CAP (SEQ ID NO:5), an MMP2 / 9 protease-cleavable linker (SEQ ID NO:160 and SEQ ID NO:9 (cleavable linker)), an anti-CD3ε variable light chain and a λ light chain (SEQ ID NO:74), followed by a labeled cysteine residue (labeled bold and underlined) that 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 11B An embodiment of an optimized nucleic acid sequence (DNA) encoding the light chain (LC) polypeptide of a precursor trispecific (trimeric) antibody construct is shown, having a 5' to 3' order and the following components (SEQ ID NO:145). Encoding Figure 11A The nucleic acid sequences of the component parts of the LC polypeptide shown include: CAP (SEQ ID NO:164), an MMP2 / 9 protease-cleavable linker (SEQ ID NO:161 and SEQ ID NO:33 (cleavable sequence)), an anti-CD3ε variable light chain and a λ light chain region (SEQ ID NO:159), followed by a labeled cysteine residue (labeled bold and underlined) that 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.

[0049] Figure 12A and 12B Embodiments showing the amino acid sequence of the heavy chain (HC) polypeptide of a non-cleavable (non-activated) precursor trispecific (trimeric) antibody construct (Construct 5; VLVH) and the optimized nucleotide sequence encoding the heavy chain (HC) of the non-cleavable precursor construct. The amino acid sequence is shown from the N-terminus to the C-terminus and the nucleic acid sequence is shown 5' to 3'. Figure 12A An embodiment showing the amino acid sequence of the heavy chain (HC) polypeptide of the non-cleavable precursor construct, having an N-terminus to C-terminus 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 sequences of the component parts of the HC polypeptide shown include: human serum albumin (HSA) (SEQ ID NO:7), an uncleavable linker (SEQ ID NO:162), the anti-CD3ε variable heavy chain and constant heavy chain region 1 (SEQ ID NO:113), followed by two labeled cysteine residues (labeled bold and underlined) that can participate in a disulfide double bond, followed by the anti-EGFR scFv VL (SEQ ID NO:34)-linker (SEQ ID NO:40)-VH (SEQ ID NO:37) chain. Figure 12B One embodiment shows an optimized nucleic acid sequence (DNA) encoding the heavy chain (HC) polypeptide of an uncleavable precursor trispecific (trimeric) antibody construct, having a 5' to 3' order and the following components (SEQ ID NO:146). Encoding Figure 12A The nucleic acid sequences of the component parts of the HC polypeptide shown include: human serum albumin (HSA) (SEQ ID NO:8), an uncleavable linker (SEQ ID NO:163), the anti-CD3ε variable heavy chain and constant heavy chain region 1 (SEQ ID NO:155), followed by two labeled cysteine residues (labeled bold and underlined) that can participate in a disulfide double bond, followed by the anti-EGFR scFv VL (SEQ ID NO:36)-linker (SEQ IDNO:40)-VH (SEQ ID NO:38) chain.

[0050] Figure 13A and 13B One embodiment shows the amino acid sequence of the light chain (LC) polypeptide of an uncleavable precursor trispecific (trimeric) antibody construct (Construct 5; VLVH) and an embodiment of the optimized nucleotide sequence encoding the light chain (LC) of the uncleavable precursor construct. The amino acid sequence is shown from the N-terminus to the C-terminus and the nucleic acid sequence is shown 5’ to 3’. Figure 13A One embodiment shows the amino acid sequence of the light chain (LC) polypeptide of the uncleavable precursor construct, having an N-terminus to C-terminus order and the following components: Cap-(h1F3.1-λLC anti-EGFR VL-linker-VH uncleavable) (SEQ ID NO:135). Figure 13A The amino acid sequences of the component parts of the LC polypeptide shown include: CAP (SEQ ID NO:5), an uncleavable linker (SEQ ID NO:162), the anti-CD3ε variable light chain and λ light chain (SEQ ID NO:74), followed by a labeled cysteine residue (labeled bold and underlined) that can participate in a disulfide double bond, 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) that encodes the light chain (LC) polypeptide of an uncleavable precursor trispecific (trimeric) antibody construct, having a 5' to 3' sequence and the following components (SEQ ID NO: 147). Encoding Figure 13A The nucleic acid sequence of the component portion of the LC polypeptide shown includes: CAP (SEQ ID NO: 164), an uncleavable linker (SEQ ID NO: 163), an anti-CD3ε variable light chain and λ light chain region (SEQ ID NO: 159), followed by a labeled cysteine residue (labeled bold and underlined) that 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.

[0051] Figure 14A and 14B An embodiment showing the amino acid sequence of the heavy chain (HC) polypeptide of an uncleavable precursor trispecific (trimeric) antibody construct (Construct 6; VHVL) and the optimized nucleotide sequence encoding the heavy chain (HC) of the uncleavable precursor construct. The amino acid sequence is shown from the N-terminus to the C-terminus and the nucleic acid sequence is shown 5' to 3'. Figure 14A One embodiment showing the amino acid sequence of the heavy chain (HC) polypeptide of the uncleavable precursor construct, having an N-terminus to C-terminus sequence 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 portion of the HC polypeptide shown includes: human serum albumin (HSA) (SEQ ID NO: 7), an uncleavable linker (SEQ ID NO: 162), an anti-CD3ε variable heavy chain and constant heavy chain region 1 (SEQ ID NO: 113), followed by two labeled cysteine residues (labeled bold and underlined) that 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 14B One embodiment showing the optimized nucleic acid sequence (DNA) that encodes the heavy chain (HC) polypeptide of an uncleavable precursor trispecific (trimeric) antibody construct, having a 5' to 3' sequence and the following components (SEQ ID NO: 148). Encoding Figure 14AThe nucleic acid sequences of the component parts of the HC polypeptide shown include: human serum albumin (HSA) (SEQ ID NO:8), an uncleavable linker (SEQ ID NO:163), the variable heavy chain and constant heavy chain region 1 of anti-CD3ε (SEQ ID NO:155), followed by two labeled cysteine residues (labeled in bold and underlined) that can participate in disulfide double bonds, followed by the anti-EGFR scFv VH (SEQ ID NO:38)-linker (SEQ ID NO:40)-VL (SEQ ID NO:36) chain.

[0052] Figure 15A and 15B Disclosed is an embodiment of the amino acid sequence of the light chain (LC) polypeptide of an uncleavable precursor trispecific (trimeric) antibody construct (Construct 6; VLVH) and the optimized nucleotide sequence encoding the light chain (LC) of the uncleavable precursor construct. The amino acid sequence is shown from the N-terminus to the C-terminus, and the nucleic acid sequence is shown from 5' to 3'. Figure 15A Disclosed is an embodiment of the amino acid sequence of the light chain (LC) polypeptide of the precursor construct, having the N-terminus to C-terminus sequence and the following components: Cap (NC)-h1F3.1-λLC-anti-EGFR (VH-linker-V) (SEQ ID NO:137). Figure 15A The amino acid sequences of the component parts of the LC polypeptide shown include: CAP (SEQ ID NO:5), an uncleavable linker (SEQ ID NO:162), the variable light chain of anti-CD3ε and the λ light chain (SEQ ID NO:74), followed by a labeled cysteine residue (labeled in bold and underlined) that can participate in disulfide double bonds, followed by the anti-EGFR scFv VH (SEQ ID NO:37)-linker (SEQ ID NO:40)-VL (SEQ IDNO:34) chain. Figure 15B Disclosed is an embodiment of the optimized nucleic acid sequence (DNA) encoding the light chain (LC) polypeptide of an uncleavable precursor trispecific (trimeric) antibody construct, having the 5' to 3' sequence and the following components (SEQ ID NO:149). Encoding Figure 15A The nucleic acid sequences of the component parts of the LC polypeptide shown include: CAP (SEQ ID NO:164), an uncleavable linker (SEQ ID NO:163), the variable light chain of anti-CD3ε and the λ light chain region (SEQ ID NO:159), followed by a labeled cysteine residue (labeled in bold and underlined) that can participate in disulfide double bonds, followed by the anti-EGFR scFv VH (SEQ ID NO:38)-linker (SEQ ID NO:40)-VL (SEQ ID NO:36) chain.

[0053] Figure 16 shows the SDS-PAGE of the activated trisomes and precursor trisome antibody constructs. In Figure 16 , all trisome and precursor trisome constructs were analyzed by SDS-PAGE, either in the reduced (R) form or the non-reduced (NR) form. Lane 1 is construct 1: the activated trisome construct (VL-VH); lane 2 is construct 2: the activated trisome construct (VH-VL); lane 3 is construct 3: the precursor trisome construct (VL-VH); lane 4 is construct 4: the precursor trisome construct (VH-VL); lane 5 is construct 5: the uncleavable precursor trisome construct (VL-VH); and lane 6 is construct 6: the uncleavable precursor trisome construct (VH-VL). The lanes are labeled with the construct number and the reduced (R) or non-reduced (NR) form. The trisome molecular weight is 100112 Da, consisting of the fusion of Fd and λC, with MWs of 51105 and 49029 respectively.

[0054] As can be seen from Figure 16 , all proteins are pure and migrate at the appropriate MW as expected, whether in their reduced form or non-reduced form. The precursor trisome molecular weight is 170390 Da, consisting of the fusion of Fd and λC, with MWs of 117649 and 52762 respectively. As can be seen from Figure 16 , all proteins are pure and migrate at the appropriate MW as expected, whether in their reduced form or non-reduced form.

[0055] Figures 17A - 17F shows analytical HPLC-size exclusion chromatography of the active trisomes and precursor trisome antibody constructs. In Figures 17A - 17F , all trisomes and precursor trisomes were analyzed in their native form by analytical HPLC-size exclusion chromatography (SEC). The trisome molecular weight is 100,112 Da and migrates as a monomer at its expected size compared to known MW markers. The precursor trisome molecular weight is 170,390 Da and migrates as a monomer at the expected size compared to known MW markers. Figure 17A shows the results of construct 1, 17B shows the results of construct 2, 17C shows the results of construct 3, 17D shows the results of construct 4, 17E shows the results of construct 5, and 17F shows the results of construct 6.

[0056] Figure 18A and 18B shows ELISA binding studies. Figure 18AShows ELISA binding studies of trispecific and pro-trispecific antibody constructs against human EGFR antigen. The binding of trispecific and pro-trispecific antibodies (both VL-VH or VH-VL forms) to their extracellular fusion antigen of EGFR (hEGFR-Fc) was tested using the ELISA method. As Figure 18A can be seen, all trispecific and pro-trispecific forms bind to the extracellular domain of hEGFR with similar affinities. In addition, there is no significant difference between the VL-VH and VH-VL forms of the anti-EGFR scFv. Figure 18B Shows that the ELISA binding of trispecific and pro-trispecific antibodies against cynomolgus monkey EGFR antigen trispecific and pro-trispecific antibodies (either VL-VH or VH-VL form) to their binding with cynomolgus monkey EGFR extracellular fusion antigen (cynomolgus monkey EGFR-Fc) was tested using the ELISA method. As Figure 18B can be seen, all trispecific and pro-trispecific forms bind to the extracellular domain of cynomolgus monkey EGFR with similar affinities. There is no significant difference between the VL-VH and VH-VL forms of the anti-EGFR scFv. In addition, for human and cynomolgus monkey EGFR, the binding affinities of all trispecific and pro-trispecific are comparable. In Figure 18A and 18B the figures: construct 1 is represented by small circles, construct 2 is represented by small squares, construct 3 is represented by upward triangles, construct 4 is represented by inverted triangles, construct 5 is represented by diamonds, and construct 6 is represented by large circles.

[0057] Figure 19A and 19B Show ELISA binding studies against human and cynomolgus (cyno) CD3ε antigens. Figure 19A Shows the ELISA binding of trispecific and pro-trispecific antibodies to human CD3ε antigen. The binding of trispecific and pro-trispecific antibodies (both VL-VH or VH-VL forms) to their extracellular fusion antigen of human CD3ε (human CD3ε-histidine) was tested using the ELISA method. As can be seen from Figure 19A , the trispecific antibody binds to hCD3ε with sub-nM affinity, while the pro-trispecific forms (both cleaved (C) and uncleaved (NC)) bind to the extracellular domain of hCD3ε with a higher EC50, indicating hindrance of CD3ε binding to its antigen. There is no significant difference between the VL-VH and VH-VL forms of the anti-CD3ε. Figure 19B Shows the ELISA binding of trispecific and pro-trispecific antibodies to cynomolgus monkey CD3ε antigen. The binding of trispecific and pro-trispecific antibodies (both VL-VH or VH-VL forms) to their extracellular fusion antigen of human CD3ε (canine CD3ε-histidine) was tested using the ELISA method. As Figure 19BAs shown, the trispecific antibody binds cynomolgus CD3ε with sub-nM affinity, while the pro-trispecific forms (both cleaved (C) and uncleaved (NC)) bind to the extracellular domain of hCD3ε with higher EC50, indicating that the binding of CD3ε to its antigen is hindered. There is no significant difference between the VLVH and VHVL forms of anti-CD3ε. In addition, for human and rhesus CD3ε, the binding affinities of all trispecific and pro-trispecific forms are comparable.

[0058] Figure 20 SDS-PAGE of trispecific and pro-trispecific (VHVL) digested by MMP9 is shown. In Figure 20 , all trispecific and pro-trispecific forms were digested by MMP9, and their cleavage products were analyzed by SDS-PAGE under non-reducing (NR) conditions. The trispecific form has a molecular weight of 100112 Da and is composed of the fusion of Fd and λC. MMP9 cleavage has no obvious activity on the trispecific form because it does not have any MMP9 cleavage sequences. The pro-trispecific form has a molecular weight of 170390 Da. Pro-trispecific-C represents the pro-trispecific form with MMP9 cleavage sequences at the C-terminus of the half-life extension part (human serum albumin) and the C-terminus of the CD3ε CAP masking part. Pro-trispecific-NC represents the pro-trispecific form lacking MMP9 cleavage sequences and thus should not be cleaved by MMP9. It can be seen that only pro-trispecific-C is cleaved by MMP9, generating two bands - the trispecific form and the half-life extension part (human serum albumin (HSA)).

[0059] Figure 21A and 21B shows the ELISA binding of pro-trispecific-C (VHVL) and pro-trispecific-NC (VHVL) to human CD3e antigen: cleaved by MMP-9. Before and after MMP9 cleavage, the binding of the VH-VL form antibody of pro-trispecific to its extracellular fusion antigen with human CD3ε (human CD3ε-histidine) was tested using the ELISA method. As Figure 21A can be seen, the pro-trispecific-C antibody does not bind to CD3ε before being cleaved by MMP9, while after being cleaved by MMP9, the pro-trispecific-C binds to hCD3ε with sub-nM affinity. The pro-trispecific-NC antibody does not bind to CD3ε in the presence or absence of MMP9 cleavage ( Figure 21B ). The data indicate that the hindrance of CAP and HSA in the pro-form leads to very little binding of the pro-form to the CD3ε antigen. Once pro-trispecific-C is cleaved by MMP9, thus releasing the HSA and CAP hindrance of CD3ε, the EC50 of the binding to human CD3ε is significantly improved towards nM binding affinity. The pro-trispecific-NC antibody is not cleaved and thus has little binding to CD3ε.

[0060] Figure 22A and 22BShown is the ELISA binding of pro - trimer - C(VHVL) and pro - trimer - NC(VHVL) to cynomolgus monkey CD3e antigen with and without MMP - 9 cleavage. Before and after MMP9 cleavage, the binding of the pro - trimer VH - VL antibody to its extracellular fusion antigen with cynomolgus monkey CD3ε (cynomolgus monkey CD3ε - histidine) was tested using the ELISA method. As Figure 22A seen, the pro - trimer - C antibody did not bind CD3ε before being cleaved by MMP9, while after being cleaved by MMP9, the pro - trimer - C bound hCD3ε with sub - nM affinity. The pro - trimer - NC antibody did not bind cynomolgus monkey CD3ε in the presence or absence of MMP9 cleavage ( Figure 22B ). The data indicate that the hindrance of CAP and HSA in the precursor results in very little binding to the CD3ε antigen. Once the pro - trimer - C is cleaved by MMP9, thus releasing the HSA and CAP hindrance of cynomolgus monkey CD3ε, the EC50 of the binding to human CD3ε is significantly improved towards nM binding affinity. The pro - trimer - NC antibody was not cleaved and thus had little binding to CD3ε.

[0061] Figure 23 Shown is the ELISA binding of trimer (VHVL), pro - trimer - C(VHVL) and pro - trimer - NC(VHVL) to human CD3e antigen: cleaved by MMP - 9. Before and after MMP9 cleavage, the binding of the trimer VHVL and pro - trimer VHVL - form antibodies to their extracellular fusion antigen with human CD3ε (human CD3ε - histidine) was tested using the ELISA method. As Figure 23 seen, the pro - trimer - C antibody did not bind CD3ε before being cleaved by MMP9, while after being cleaved by MMP9, the pro - trimer - C bound hCD3ε with sub - nM affinity, similar to the trimer. The pro - trimer - NC antibody did not bind CD3ε in the presence or absence of MMP9 cleavage.

[0062] The data indicate that the hindrance of CAP and HSA in the precursor antibody results in very little binding to the CD3ε antigen. Once the pro - trimer - C is cleaved by MMP9, thus releasing the HSA and CAP hindrance of CD3ε, the EC50 of the binding to human CD3ε is significantly improved towards nM binding affinity. The pro - trimer - NC antibody was not cleaved and thus had little binding to CD3ε.

[0063] Figure 24Shows ELISA binding of trisomes (VHVL), pro-trisome-C (VHVL), and pro-trisome-NC (VHVL) to cynomolgus monkey CD3e antigen: Cleaved by MMP-9. Before and after MMP9 cleavage, ELISA was used to test the binding of trisome VHVL and pro-trisome VHVL form antibodies to their extracellular fusion antigen with cynomolgus monkey CD3ε (cynomolgus monkey CD3ε-histidine). As Figure 24 can be seen, the pro-trisome-C antibody did not bind CD3ε before being cleaved by MMP9, while after being cleaved by MMP9, the pro-trisome-C bound hCD3ε with sub-nM affinity, similar to the trisome. The pro-trisome-NC antibody did not bind CD3ε in the presence or absence of MMP9 cleavage.

[0064] Data indicate that the hindrance of CAP and HSA in the precursor antibody results in very little binding to the CD3ε antigen. Once the pro-trisome-C is cleaved by MMP9, thus releasing the HSA and CAP hindrance of CD3ε, the binding EC50 to human CD3ε is significantly improved towards a low nM binding affinity. The pro-trisome-NC antibody was not cleaved and thus had little binding to CD3ε.

[0065] Figure 25 Shows FACS binding of trisomes (VHVL), pro-trisome-C (VHVL), and pro-trisome-NC (VHVL) to Jurkat cells (human CD3e). The FACS method was used to test the binding of trisome VHVL and pro-trisome VHVL form antibodies to their binding to Jurkat cells (human CD3e). As can be seen from Figure 25 it, the pro-trisome-C and pro-trisome-NC antibodies did not bind to Jurkat cells, while the trisome bound with low nM affinity. Data indicate that the hindrance of CAP and HAS in the precursor structure results in very little binding to Jurkat cells expressing human CD3ε. Once the HSA and CAP hindrance of CD3ε is absent (trisome (VHVL)), the binding EC50 to human CD3ε is significantly improved towards a low nM binding affinity.

[0066] Figures 26A - 26B Shows an embodiment of the amino acid and nucleic acid sequences of scFv anti-ROR1 VL-VH. Figure 26A Shows an embodiment of the amino acid sequence of scFv anti-ROR1 as follows, with 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 Shows an embodiment of the optimized nucleic acid sequence encoding scFv anti-ROR1 as follows, with N-terminal to C-terminal order: VL-VH (SEQ ID NO:157).

[0067] Figures 27A - 27B Disclosed is an embodiment showing the amino acid and nucleic acid sequences of scFv anti-ROR1 VH-VL. Figure 27A Disclosed is an embodiment of the amino acid sequence of scFv anti-ROR1 having an 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 Disclosed is an embodiment of the optimized nucleic acid sequence encoding scFv anti-ROR1 having an N-terminal to C-terminal order: VH-VL (SEQ ID NO: 167).

[0068] Figures 28A - 28B Disclosed are embodiments showing the amino acid and nucleic acid sequences of scFv anti-PSMA VL-VH. Figure 28A Disclosed is an embodiment of the amino acid sequence of scFv anti-PSMA having an 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 Disclosed is an embodiment of the optimized nucleic acid sequence encoding scFv anti-PSMA having an N-terminal to C-terminal order: VL-VH (SEQ ID NO: 169).

[0069] Figures 29A - 29B Disclosed are embodiments showing the amino acid and nucleic acid sequences of scFv anti-PSMA VH-VL. Figure 29A Disclosed is an embodiment of the amino acid sequence of scFv anti-PSMA having an 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 Disclosed is an embodiment of the optimized nucleic acid sequence encoding scFv anti-PSMA having an N-terminal to C-terminal order: VH-VL (SEQ ID NO: 171).

[0070] Figures 30A - 30B Disclosed are embodiments showing the amino acid and nucleic acid sequences of scFv anti-5T4 VL-VH. Figure 30A Disclosed is an embodiment of the amino acid sequence of scFv anti-5T4 having an 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 Disclosed is an embodiment of the optimized nucleic acid sequence encoding scFv anti-5T4 having an N-terminal to C-terminal order: VL-VH (SEQ ID NO: 173).

[0071] Figures 31A - 31B Disclosed are embodiments showing the amino acid and nucleic acid sequences of the scFv anti-5T4 VH-VL. Figure 31A Disclosed is an embodiment of the amino acid sequence of the scFv anti-5T4 having an 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 Disclosed is an embodiment of the optimized nucleic acid sequence encoding the scFv anti-5T4 having an N-terminal to C-terminal order: VH-VL (SEQ ID NO:1175).

[0072] Figure 32 Shown are the FACS binding data of the EGFR-binding trispecific and precursor trispecific antibody constructs cleaved by MMP9 with Jurkat cells. The MFI (mean fluorescence intensity) provides a relative measure of antibody binding. Constructs analyzed: trispecific (VHVL) is construct 2 of Example 1 (small circles, squares); pro-trispecific-C (VHVL) is construct 4 of Example 1 (upward and downward triangles); and pro-trispecific-NC (VHVL) is construct 6 of Example 1 (diamonds and large circles). -MMP9 and +MMP9 indicate the absence (-) or presence (+) of the MMP9 protease.

[0073] Figures 33A - 33B Shown are the in vivo pharmacokinetic values (PK values) of the EGFR-activated ( Figure 33A ) trispecific construct and the precursor ( Figure 33B ) trispecific construct administered to mice. Antibody constructs: trispecific EGFR (VL-VH) is construct 1 of Example 1 ( Figure 33A ); pro-trispecific-C EGFR (VL-VH) is construct 3 of Example 1 ( Figure 33B ); and pro-trispecific-NC EGFR (VL-VH) is construct 5 of Example 1 (Figure 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.

[0074] Figures 34A - 34F Shown are the SDS-PAGE results (34A and 34D) and the analytical HPLC-size exclusion chromatography results (Figure 34B, Table 34C, Figure 34E, and Table 34F). Figures 34A - 34C Shown are the results of the trispecific-ROR1 (VL-VH) construct in which both the first and second binding sites bind ROR1 and the third binding site is the CD3ε binding site. No regulatory arms are present (construct 7 of Example 8). Figures 34D - 34FShows the results of the first trisome-ROR1 (VL-VH) construct, where both the first and second binding sites bind ROR1 and the third binding site is the CD3ε binding site, where the construct includes a CAP regulatory domain that masks CD3ε binding and an HLP regulatory domain that contains HSA, where the two regulatory domains are linked to the CD3ε binding site at the N-terminus by an uncleavable linker (construct 11 of Example 8).

[0075] Figures 35A - 35I Shows the SDS-PAGE results (35A, 35D, and 35G) under reducing and non-reducing conditions, as well as the analytical HPLC-size exclusion chromatography results (Figure 35B, Table 35C, Figure 35E, Table 35F, Figure 35H, and Table 35I). Figures 35A - 35C Shows the results of the trisome-5T4 (VL-VH) construct, where both the first and second binding sites bind 5T4 and the third binding site is the CD3ε binding site (construct 13 of Example 8). Figures 35D - 35F Shows the results of the first trisome-5T4 (VL-VH) construct, where both the first and second binding sites bind 5T4 and the third binding site is the CD3ε binding site, where the construct includes a CAP regulatory domain that masks CD3ε binding and an HLP regulatory domain that contains HSA, where the two regulatory domains are linked to the CD3ε binding site at the N-terminus by a cleavable linker (construct 15 of Example 8). Figures 35G - 35I Shows the results of the first trisome-5T4 (VL-VH) construct, where both the first and second binding sites bind 5T4 and the third binding site is the CD3ε binding site, where the construct includes a CAP regulatory domain that masks CD3ε binding and an HLP regulatory domain that contains HSA, where the two regulatory domains are linked to the CD3ε binding site at the N-terminus by an uncleavable linker (construct 17 of Example 8).

[0076] Figure 36Shows the binding curves, which show the binding of trisomes and pro-trisome antibodies to the 5T4 antigen. 5T4-trisome is an activated trisome construct in which the first and second binding domains bind 5T4, the third domain binds CD3ε, and there is no regulatory domain. 5T4-PT trisome-C is a pro-trisome 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 sub-regulatory domain being linked to the N-terminus of the Fab of the CD3ε binding domain, wherein the linker within the regulatory domain is MMP9-cleavable. 5T4-PT trisome-NC is a pro-trisome 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 sub-regulatory domain being linked to the N-terminus of the Fab of the CD3ε binding domain, wherein the linker within the regulatory domain is non-cleavable.

[0077] Figure 37 Shows the SDS-PAGE results of MMP9 protease cleavage of 5T4 pro-trisome antibody constructs. Trisome-5T4 (VL-VH) is an activated trisome 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). Pro-trisome-5T4-C-(VL-VH) is a pro-trisome 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 sub-regulatory domain being linked to the N-terminus of the Fab of the CD3ε binding domain, wherein the linker within the regulatory domain is MMP9-cleavable (construct 15 of Example 8). Pro-trisome-5T4-NC-(VL-VH) is a pro-trisome 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 sub-regulatory domain being linked to the N-terminus of the Fab of the CD3ε binding domain, wherein the linker within the regulatory domain is non-cleavable (construct 17 of Example 8).

[0078] Figure 38Binding curves are shown that demonstrate the binding of trispecific and pro-trispecific antibody constructs to human CD3ε antigen in the presence or absence of MMP9 protease ((+MMP9) or (-MMP9)), respectively. 5T4-trispecific is an activated trispecific construct where the first and second binding domains bind 5T4, the third domain binds CD3ε, and there is no regulatory domain. 5T4-PT trispecific-C is a pro-trispecific construct where the first and second binding domains bind 5T4, the third domain binds CD3ε, and there are first and second sub-regulatory domains, each connected to the N-terminus of the Fab of the CD3ε binding domain, where the linker within the regulatory domain is MMP9-cleavable. 5T4-PT trispecific-NC is a pro-trispecific construct where the first and second binding domains bind 5T4, the third domain binds CD3ε, and there are first and second sub-regulatory domains, each connected to the N-terminus of the Fab of the CD3ε binding domain, where the linker within the regulatory domain is non-cleavable.

[0079] Figure 39 FACS binding of trispecific-5T4, pro-trispecific-5T4-C, and pro-trispecific-5T4-NC to Jurkat cells (human CD3ε) is shown in the presence and absence of MMP9 protease.

[0080] Figure 40 ELISA binding studies of trispecific and pro-trispecific antibody constructs against human 5T4 antigen are shown.

[0081] Figure 41 FACS binding data of 5T4-binding trispecific and pro-trispecific antibody constructs to CHO cells expressing human 5T4 are shown. MFI (mean fluorescence intensity) provides a relative measure of antibody binding.

[0082] Figure 42 FACS binding data of 5T4-binding trispecific and pro-trispecific antibody constructs to the MCF7 breast cancer cell line, known to highly express human 5T4, are shown. MFI (mean fluorescence intensity) provides a relative measure of antibody binding.

[0083] Figure 43 and 44 Cytotoxicity assays of 5T4-binding trispecific and pro-trispecific antibody constructs against the MCF7 breast cancer cell line Figure 43 ( Figure 44 ) known to highly express human 5T4 and the NCI-H226 lung cancer cell line

[0084] Figures 45A - 45B Two embodiments of the amino acid sequence of the heavy chain (HC) polypeptide of a pro-trispecific (trispecific) antibody construct are shown (e.g. Figure 2FEmbodiments of the HC, where both scFvs bind to EGFR). The amino acid sequence is shown from the N-terminus to the C-terminus. Figure 45A and 45B Shows embodiments of the amino acid sequence of the heavy chain (HC) polypeptide of the precursor construct, with the N-terminus to C-terminus order and the following components: CAP (bold), linker (italic), human serum albumin (underlined); protease cleavage sequence (bold and italic); VH1 / CH1 of anti-CD3e Fab (bold and underlined), VL (italic and underlined), VH (italic, bold and underlined) (EGFR scFv). Two marked cysteine residues (double underlined) can participate in disulfide 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 recited in SEQ ID NO:28. Figure 45B The amino acid sequence of is recited in SEQ ID NO:31. Embodiments of the sequences of the component parts are described throughout this application.

[0085] Figure 46 Shows embodiments of the amino acid sequence of the light chain (LC) polypeptide of the precursor trispecific (trimeric) antibody construct (e.g., Figure 2F Embodiments of the LC, where both scFvs bind to EGFR). The amino acid sequence is shown from the N-terminus to the C-terminus. Figure 46 Shows an embodiment of the amino acid sequence of the light chain (LC) polypeptide of the precursor construct, with the N-terminus to C-terminus order and the following components: linker (italic), VL1-CL of anti-CD3e Fab (double underlined and bold), VL (italic and underlined), and VH (italic and underlined) (where VH is the EGFR scFv). Two marked cysteine residues (double underlined) can participate in disulfide bonds. Figure 46 The amino acid sequence of the LC shown in is recited in SEQ ID NO:32. Embodiments of the amino acid sequences of the component parts of the LC polypeptide are described throughout this application.

[0086] Figures 47A - 47B Shows two embodiments of the amino acid sequence of the heavy chain (HC) polypeptide of the precursor trispecific (trimeric) antibody construct (e.g., Figure 2F Embodiments of the HC, where both scFvs bind to 5T4). The amino acid sequence is shown from the N-terminus to the C-terminus. Figure 47A and 47BShows an embodiment of the amino acid sequence of the heavy chain (HC) polypeptide of the precursor construct, having an N-terminal to C-terminal sequence and the following components: CAP (bold), linker (italic), human serum albumin (underlined); protease cleavage sequence (bold and italic); VH1 / CH1 of anti-CD3e Fab (bold and underlined), VL (italic and underlined), VH (italic, bold and underlined) (5T4 scFv). Two marked cysteine residues (double underlined) can participate in disulfide bonds. Figure 47A The protease cleavage sequence in Figure 47A is a multiple protease cleavage sequence. Figure 47B The protease cleavage sequence in Figure 47B is the MMP2 / 9 protease cleavage sequence. Figure 47A The amino acid sequence of Figure 47A is set forth in SEQ ID NO:118. Figure 47B The amino acid sequence of Figure 47B is set forth in SEQ ID NO:176. Embodiments of the sequences of the component parts are described throughout this application.

[0087] Figure 48 Shows an embodiment of the amino acid sequence of the light chain (LC) polypeptide of the precursor trispecific (trimeric) antibody construct (e.g., Figure 2F an embodiment of the LC of Figure 2F , wherein both scFvs bind 5T4). The amino acid sequence is shown from N-terminal to C-terminal. Figure 48 Shows an embodiment of the amino acid sequence of the light chain (LC) polypeptide of the precursor construct, having an N-terminal to C-terminal sequence and the following components: linker (italic), VL1-CL of anti-CD3e Fab (double underlined and bold), VL (italic and underlined), and VH (italic and underlined), wherein VH is 5T4 scFv). Two marked cysteine residues (double underlined) can participate in disulfide bonds. Figure 48 The amino acid sequence of the LC shown in Figure 48 is set forth in SEQ ID NO:177. Embodiments of the amino acid sequences of the component parts of the LC polypeptide are described throughout this application.

[0088] 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, for clarity, the dimensions of some elements may be exaggerated relative to other elements. Further, where considered appropriate, reference numerals may be repeated in the figures to indicate corresponding or similar elements. Detailed Description

[0089] In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the precursor trispecific antibody construct. However, those skilled in the art will understand that the precursor construct, its production and use as shown herein can be practiced without these specific details. In other instances, well-known methods, procedures, and components have not been described in detail so as not to obscure the disclosure.

[0090] This document describes precursor trispecific antibody constructs that include separate cleavable masking domains and half-life extension domains, where these cleavable regulatory domains provide reduced binding of the precursor trispecific construct to T cells outside the tumor microenvironment (TME) and provide an extended half-life. The half-life extension may be limited to the time the precursor trispecific construct is outside the cancer microenvironment, or it may be extended to the time the precursor trispecific construct resides in the cancer microenvironment. The advantages of the precursor trispecific antibody constructs described herein having protease-cleavable masking domains and protease-cleavable half-life extension (HLP) domains can be improved protease-activated controlled release of the masking CAP and HLP domains.

[0091] The reduced T cell binding may result in reduced 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.

[0092] In some embodiments, the precursor trispecific antibody constructs described herein provide regulatable T cell activation, where the precursor construct provides T cell activation limited to the tumor microenvironment. In some embodiments, compared to non-precursor multivalent antibodies that are always active, the precursor trispecific antibody constructs described herein have an enhanced half-life and provide T cell activation limited to the tumor microenvironment. In some embodiments, compared to non-precursor multivalent antibodies that are always active, the precursor trispecific antibody constructs described herein have reduced T cell activation in non-tumor microenvironments.

[0093] In some embodiments, compared to non-precursor trispecific antibodies, the precursor trispecific antibody constructs described herein have an extended half-life in non-tumor microenvironments. In some embodiments, compared to non-precursor multivalent antibodies that are always active, the precursor trispecific antibody constructs described herein have reduced T cell binding and / or activation in non-tumor microenvironments and have an extended half-life in non-tumor microenvironments.

[0094] In some embodiments, the precursor trispecific antibody constructs described herein can bind to T cells. In some embodiments, the precursor trispecific antibody constructs described herein can bind to T cells within the tumor microenvironment (TME). In some embodiments, the precursor trispecific antibody constructs described herein can bind to T cells and two different TAAs. In some embodiments, the precursor trispecific 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 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. 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 TME antigens. 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 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 antigen on the surface of the blood vessel is an endothelial cell surface antigen. In some embodiments, the antigen on the surface of the blood vessel is an endothelial cell surface antigen 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.

[0095] In some embodiments, the stromal antigens in the TME include fibroblast activation protein (FAP). In some embodiments, the stromal antigens in the TME include alpha smooth muscle actin (αSMA). In some embodiments, the stromal antigens in the TME include PDGFRα. In some embodiments, the stromal antigens in the TME include integrin α11β1 (ITGA11). In some embodiments, the stromal antigens in the TME include VEGF. In some embodiments, the stromal antigens in the TME include tenascin-C, periostin. In some embodiments, the stromal antigens in the TME include fibroblast specific protein 1 (S10A4, FSP1). In some embodiments, the stromal antigens in the TME include desmin. In some embodiments, the stromal antigens in the TME include vimentin. In some embodiments, the stromal antigens in the TME include palladin. In some embodiments, the stromal antigens in the TME include urokinase-type plasminogen activator receptor-associated protein (UPARAP). In some embodiments, the stromal antigens in the TME include galectin-3. In some embodiments, the stromal antigens in the TME include podoplanin. In some embodiments, the stromal antigens in the TME include platelets. In some embodiments, the stromal antigens in the TME include CCL2. In some embodiments, the stromal antigens in the TME include CXCL12. In some embodiments, the stromal antigens in the TME are selected from any of 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, palladin, urokinase-type plasminogen activator receptor-associated protein (UPARAP), galectin-3, podoplanin, platelets, CCL2 or CXCL12.

[0096] In some embodiments, the present disclosure describes a pharmaceutical composition comprising a precursor trispecific antibody construct that provides regulatable T cell activation in a non-tumor microenvironment. In some embodiments, the present disclosure describes a pharmaceutical composition comprising a precursor trispecific antibody construct that has an increased half-life and provides T cell activation restricted to the tumor microenvironment. In some embodiments, the present disclosure describes a pharmaceutical composition comprising a precursor trispecific antibody construct that has an extended half-life in a non-tumor microenvironment. In some embodiments, the present disclosure describes a pharmaceutical composition comprising a precursor trispecific antibody construct that has an extended half-life in a non-tumor microenvironment, wherein the half-life in the tumor microenvironment is reduced compared to the half-life in the non-tumor microenvironment. In some embodiments, the pharmaceutical composition comprising the precursor construct comprises a construct that recognizes a T cell and a TAA. In some embodiments, the pharmaceutical composition comprising the precursor construct comprises a construct that recognizes a T cell and two TAAs, wherein each TAA is a different antigen. In some embodiments, the pharmaceutical composition comprising the precursor construct comprises a construct that recognizes a T cell and two TAAs, wherein the TAAs recognize the same antigen.

[0097] In some embodiments, the present disclosure describes methods of using a precursor trispecific antibody construct as disclosed herein for treating, preventing, inhibiting growth, delaying disease progression, reducing tumor burden, or reducing the incidence of cancer or a tumor in a subject, or any combination thereof. In some embodiments, compared to a subject to whom the pharmaceutical composition has not been administered, the disclosed methods of treatment reduce minimal residual disease, improve 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 the treated subject.

[0098] Precursor trispecific antibody construct

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

[0100] In some embodiments, there are provided herein 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, the regulatory domain comprising either a first and a second sub-regulatory domain or a single regulatory domain, the first sub-regulatory domain comprising a first protease cleavage domain and a half-life prolongation (HLP) domain, and the second sub-regulatory domain comprising a second protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to the extracellular epitope of human CD3ε; the single regulatory domain comprising a protease cleavage domain, a half-life prolongation (HLP) domain, and a CAP component that reduces the ability of the third binding domain to bind to the extracellular epitope of human CD3ε.

[0101] In some embodiments, there are provided herein 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 prolongation (HLP) domain, and a CAP component that reduces the ability of the third binding domain to bind to the extracellular epitope of human CD3ε.

[0102] In some embodiments, a 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 prolongation (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 the extracellular epitope of human CD3ε.

[0103] In some embodiments, a 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 prolongation (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 the extracellular epitope of human CD3ε.

[0104] For the precursor constructs described throughout the text, one of ordinary skill in the art will understand that, based on the amino acid sequences contained in the first and second binding sites, the modular structure of the constructs allows for different binding partners.

[0105] Those skilled in the art will recognize that a precursor antibody construct is a precursor form or "Pro" form of an active antibody protein. In some embodiments, the term "Pro" is used interchangeably with the term "precursor" and has the same meaning and nature.

[0106] Those skilled in the art 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" used throughout the text can be used interchangeably and have the same nature and meaning. Additionally, in some embodiments, the term "trispecific" can be replaced with the term "trimeric", as it is recognized that the antibody constructs disclosed herein have three antibody variable regions, where each region can bind a different antigen (trimeric or trispecific) or two of the three binding regions can bind the same antigen (trimeric or trispecific, where two specifically bind the same antigen). Thus, the terms listed above, for example, in some embodiments, "precursor trispecific antibody construct" can be used interchangeably with the term "precursor trimeric construct" and have the same meaning and nature.

[0107] In some embodiments, a 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 that comprises a protease cleavage domain and a half-life prolongation (HLP) domain; and a second sub-regulatory domain that comprises a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to the 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.

[0108] In some embodiments, a 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 that comprises a protease cleavage domain and a half-life prolongation (HLP) domain; and a second sub-regulatory domain that comprises a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to the extracellular epitope of human CD3ε, wherein the first TAA or the second TAA or both the first TAA and the second TAA include 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 that comprises a protease cleavage domain and a half-life prolongation (HLP) domain; and a second sub-regulatory domain that comprises a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to the 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.

[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 that comprises a protease cleavage domain and a half-life prolongation (HLP) domain; and a second sub-regulatory domain that comprises a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to the extracellular epitope of human CD3ε, wherein the first TAA or the second TAA or both the first TAA and the second TAA comprise an angiogenesis antigen 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 that comprises a protease cleavage domain and a half-life prolongation (HLP) domain; and a second sub-regulatory domain that comprises a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to the 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 blood vessels 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 that comprises a protease cleavage domain and a half-life prolongation (HLP) domain; and a second sub-regulatory domain that comprises a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to the 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.

[0113] 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 that comprises a protease cleavage domain and a half-life prolongation (HLP) domain; and a second sub-regulatory domain that comprises a protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to the 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 a 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.

[0114] Those skilled in the art will understand that in some embodiments, the term "tumor-associated antigen" (TAA) may include a molecule or a portion thereof displayed on the cell surface or a molecule present in the tumor environment (i.e., within the tumor microenvironment). In some embodiments, the TAA includes a cell surface tumor-associated antigen (TAA). 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 vascular 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.

[0115] 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.

[0116] 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 by a tumor cell into the TME. In some embodiments, the TAA comprises an effector molecule secreted by a tumor cell into the TME. In some embodiments, the effector molecule comprises a cytokine antigen. In some embodiments, the effector molecule comprises a cytokine antigen in the TME.

[0117] In some embodiments, the cytokine antigens in the TME include TNF-α, IL-6, TGF-β, IL-10, IL-8, IL-17, IL-21, INF, or VEGF. In some embodiments, the TAAs are selected from TNF-α, IL-6, TGF-β, IL-10, IL-8, IL-17, IL-21, INF, or VEGF. In some embodiments, the cytokine antigens used as TAAs include cytokine antigens known in the art.

[0118] 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 angiogenesis antigen in the TME, an antigen on a blood vessel in the TME, or a cytokine antigen in the TME, or any combination thereof.

[0119] One skilled in the art will understand that the terms "tumor microenvironment" (TME), "cancer microenvironment", and "tumor environment" can be used interchangeably, have the same nature and meaning, and encompass the microenvironment of tumor development. While the normal cell microenvironment can inhibit the growth of malignant cells, the alterations that occur in the tumor microenvironment can synergistically support cell proliferation.

[0120] 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.

[0121] 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, an extracellular epitope of a tumor cell surface antigen, a TME antigen, a stromal antigen in the TME, an angiogenesis antigen in the TME, an antigen on a blood vessel in the TME, or a cytokine in the TME.

[0122] One skilled in the art will understand that the term "antigen" or "immunogen" includes peptides, proteins or polypeptides having immunogenicity or any fragment thereof. In some embodiments, the antigen is capable of eliciting an immune response in a mammal and thus contains at least one and may contain multiple epitopes. The "antigen" molecule or a part of the molecule can be bound by a selective binder, such as the antigen-binding portion of a Fab fragment or the antigen-binding portion of a single-chain variable fragment (scFv). In addition, the "antigen" can be used in animals to produce antibodies capable of binding to the epitopes of the antigen. In some embodiments, the CAP component contains an antigen portion bound by a third binding domain.

[0123] The term "epitope" includes any determinant, and in certain embodiments, the polypeptide determinant can specifically bind to an anti-TAA binding domain or an anti-T cell receptor binding domain. An epitope is the antigenic region bound by an antibody or its antigen-binding fragment. In some embodiments, the CAP component contains an epitope bound by a third binding domain.

[0124] In certain embodiments, the epitope determinant includes the chemical reactive surface grouping of a molecule, such as amino acids, sugar side chains, phosphoryl groups or sulfonyl groups, and in certain embodiments may have specific three-dimensional structural features and / or specific charge features. In certain embodiments, when a precursor trispecific antibody construct preferably recognizes a target antigen in a complex mixture of proteins and / or macromolecules, the precursor trispecific antibody construct is said to specifically bind the antigen. When the equilibrium dissociation constant ≤ 10 -5 、10 -6 or 10 -7 M, the precursor trispecific antibody construct is said to specifically bind 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 can be ≤ 10 -10 M or 10 -11 M. The antigens disclosed herein include, but are not limited to, TAAs, CAP components and immune effector molecules, such as human CD3ε polypeptide.

[0125] In some embodiments, the tumor-associated antigen (TAA) is a tumor antigen. In some embodiments, the tumor antigen includes 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.

[0126] In some embodiments, when the TAA is a tumor cell antigen, the tumor cells include cells from solid tumors. Solid tumors can 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 neoplasm (new growth of cells) or lesion (damage to anatomical structure or disorder of physiological function) formed by abnormal growth of body tissue cells other than blood, bone marrow, or lymphocytes. In some embodiments, a solid tumor consists of abnormal cell populations that may originate from different tissue types, such as the liver, colon, breast, or lung, and initially grow in the organ from which their cells are derived. However, these cancers may spread to other organs through metastatic tumor growth in the advanced stage of the disease.

[0127] In some embodiments, the solid tumors include sarcoma or carcinoma, fibrosarcoma, myxosarcoma, liposarcoma, chondrosarcoma, osteosarcoma, chordoma, angiosarcoma, endotheliosarcoma, lymphangiosarcoma, lymphangioendotheliosarcoma, synovioma, mesothelioma, Ewing's 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, cholangiocarcinoma, choriocarcinoma, seminoma, embryonal carcinoma, Wilms' 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. In some embodiments, the solid tumors include adrenocortical tumors (adenomas and carcinomas), carcinomas, colorectal carcinomas, desmoid tumors, desmoplastic small round cell tumors, endocrine tumors, Ewing's sarcoma, germ cell tumors, hepatoblastoma, hepatocellular carcinoma, melanoma, neuroblastoma, osteosarcoma, retinoblastoma, rhabdomyosarcoma, soft tissue sarcomas 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.

[0128] In some embodiments, the tumor cells include cells from non-solid tumors, i.e., non-solid cancers. In some embodiments, the cancer can be a diffuse cancer, where the cancer spreads widely; is not confined or restricted. In some embodiments, diffuse cancers can include non-solid tumors. Examples of diffuse cancers include leukemia. Leukemia includes cancers that start in hematopoietic tissues (such as the bone marrow) and result in the production of large numbers of abnormal blood cells that enter the bloodstream.

[0129] In some embodiments, diffuse cancers include B-cell malignancies. In some embodiments, diffuse cancers include leukemia. In some embodiments, the cancer is lymphoma. In some embodiments, the lymphoma is large B-cell lymphoma.

[0130] In some embodiments, diffuse cancers or tumors include hematological malignancies. In some embodiments, hematological malignancies are types of cancer that affect the blood, bone marrow, and lymph nodes. Hematological malignancies can originate from either of two major blood cell lineages: the myeloid and lymphoid cell lines. The myeloid cell line typically produces granulocytes, red blood cells, platelets, macrophages, and mast cells, while the lymphoid cell line produces B, T, and plasma cells. Lymphomas (such as Hodgkin lymphoma), lymphocytic leukemia, and myeloma originate from the lymphoid lineage, while acute and chronic myeloid leukemia (AML, CML), myelodysplastic syndromes, and myeloproliferative disorders are of myeloid origin.

[0131] In some embodiments, non-solid (diffuse) cancers or tumors include hematopoietic system malignancies, blood cell cancers, leukemia, myelodysplastic syndromes, lymphoma, multiple myeloma (plasma cell myeloma), acute lymphocytic leukemia, acute myeloid leukemia, chronic myeloid leukemia, Hodgkin lymphoma, non-Hodgkin lymphoma, or plasma cell leukemia.

[0132] In some embodiments, the tumor or cancer contains metastases of the tumor or cancer.

[0133] 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 in a tertiary form from the exterior of the cell. In some embodiments, the cell surface TAA located in the plasma membrane is a transmembrane protein that contains hydrophilic and hydrophobic regions in its tertiary conformation.

[0134] These antigens can be presented on the cell surface together with the extracellular portion, which is usually combined with the transmembrane and cytoplasmic portions of the molecule. These antigens are sometimes presented only by tumor cells and never by normal cells. Tumor antigens can be specifically expressed on tumor cells or can represent tumor-specific mutations compared to normal cells. In this case, they are called tumor-specific antigens. More commonly, antigens are presented by both tumor cells and normal cells. In some embodiments, TAAs include antigens specifically expressed on tumor cells. In some embodiments, TAAs include antigens expressed on both tumor cells and normal cells.

[0135] In some embodiments, compared to normal cells, TAAs can be overexpressed on tumor cells, or due to the less compact structure of tumor tissue compared to normal tissue, TAAs can be more readily antibody-bound in tumor cells.

[0136] In some embodiments, the precursor trispecific antibody construct described herein comprises (a) an scFv fragment comprising a first binding domain (TAA binding domain) that binds to a TAA; (b) an scFv fragment comprising a second binding domain (TAA binding domain) that binds to a TAA; (c) a Fab fragment comprising a third binding domain (CD3 binding domain) that binds to an extracellular epitope of human CD3ε; (d) a first sub-regulatory domain comprising a protease cleavage domain and a half-life prolongation (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 the extracellular epitope of CD3ε. In some embodiments, the precursor trispecific antibody construct described herein comprises (a) an scFv fragment comprising a first binding domain (TAA binding domain) that binds to a TAA; (b) an scFv fragment comprising a second binding domain (TAA binding domain) that binds to a TAA; (c) a Fab fragment comprising a third binding domain (CD3 binding domain) that binds to an extracellular epitope of human CD3ε; (d) a first sub-regulatory domain comprising a protease cleavage domain and a half-life prolongation (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 the extracellular epitope of CD3ε.

[0137] Those skilled in the art will understand that in some embodiments, the precursor antibody construct comprises a precursor or derivative form of a pharmaceutically active antibody. In some embodiments, the pharmaceutical formulation comprises a precursor antibody construct. In some embodiments, the formulation comprises a precursor antibody construct. In some embodiments, the precursor antibody construct has reduced adverse effects compared to the activated antibody. In some embodiments, the precursor antibody construct has reduced adverse effects compared to the activated antibody, 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.

[0138] In certain embodiments, the precursor antibody construct has an extended half-life compared to the activated antibody. In certain embodiments, the precursor antibody construct has an extended half-life compared to the activated antibody, wherein the precursor antibody can be enzymatically activated or converted into the active form of the antibody, and the active form has a reduced half-life compared to the precursor antibody construct.

[0139] 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 the 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 the activated antibody, wherein the precursor antibody can be enzymatically activated or converted into the active form of the antibody.

[0140] In certain embodiments, the precursor antibody construct has both an extended half-life and a reduced ability to activate T cells compared to the activated antibody. In certain embodiments, the precursor antibody construct has both an extended half-life and a reduced ability to bind to T cells compared to the activated antibody. In certain embodiments, the precursor antibody construct has both an extended half-life and a reduced ability to activate T cells compared to the activated antibody, wherein the precursor antibody can be enzymatically activated or converted into the active form of the antibody. In certain embodiments, the precursor antibody construct has an extended half-life and a reduced ability to bind to T cells, wherein the precursor antibody can be enzymatically activated or converted into the active form of the antibody.

[0141] In some embodiments, the ability of the precursor antibody to bind to T cells is reduced, where the regulatory domain containing the CAP component is cleaved but the regulatory domain containing the HLP is not cleaved, and where the "partially" activated antibody can bind to T cells and maintain an extended half-life. In some embodiments, the binding of the partially activated precursor antibody to T cells is reduced compared to the fully activated antibody, where both regulatory arms have been proteolytically cleaved. In some embodiments, the ability of the precursor antibody to activate T cells is reduced, where the regulatory domain containing the CAP component is cleaved but the regulatory domain containing the HLP is not cleaved, and where the "partially" activated antibody can activate T cells and maintain an extended half-life. In some embodiments, the activation of T cells is reduced after binding to the partially activated precursor construct compared to the fully activated antibody, where both regulatory arms have been proteolytically cleaved.

[0142] In some embodiments, the precursor antibody construct is synthesized in vitro. In some embodiments, the precursor antibody construct does not convert to the active form of the antibody when the precursor is present in vivo in a non-tumor microenvironment (e.g., in circulation).

[0143] 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 enzymatically cleavable regulatory domains. In some embodiments, in addition to the antigen-binding domain, the precursor antibody construct further comprises a plurality of regulatory domains, where a portion of the regulatory domains is enzymatically cleavable. In some embodiments, in addition to the antigen-binding domain, the precursor antibody construct further comprises two regulatory domains, where a portion of the regulatory domains is enzymatically cleavable. In some embodiments, in addition to three antigen-binding domains, the precursor antibody construct further comprises two regulatory domains, where the regulatory domains are enzymatically cleavable.

[0144] In some embodiments, prior to the cytotoxic activation of T cells, the precursor trispecific antibodies described herein exhibit enhanced selectivity for targeting tumor cells compared to targeting normal cells.

[0145] Immunobinding generally refers to non-covalent interactions that occur between an immunoglobulin molecule and an antigen specific for the immunoglobulin, for example, as a result of, by way of illustration and not limitation, electrostatic, ionic, hydrophilic, and / or hydrophobic attractions or repulsions, steric forces, hydrogen bonds, van der Waals forces, and other interactions. The strength or affinity of an immunobinding interaction can be expressed by the dissociation constant (K d ) of the interaction, where a smaller K drepresents greater affinity. The immunobinding properties of the selected polypeptides can be quantified using methods well known in the art. One such method involves measuring the rates of antigen-binding site / antigen complex formation and dissociation, where these rates depend on the concentration of the complex partners, the affinity of the interaction, and geometric parameters that also affect the bidirectional rates. Thus, the "association 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. Thus, the ratio of k off / k on equals the dissociation constant K D . See generally Davies et al. (1990) Annual Rev. Biochem. 59:439-473.

[0146] One of ordinary skill in the art will understand that a "binding domain" or related expressions, such as a domain that "binds" or is "reactive with / to" a particular target, includes the ability of the domain to distinguish its respective antigen and 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 having the ability to specifically recognize and bind to a biomolecule (such as a cell surface receptor or a tumor protein, or a component thereof, such as its extracellular component). Binding domains include any naturally occurring, synthetic, semi-synthetic, or recombinantly produced binding partner of the biomolecule of interest. For example, and as further described herein, the binding domain can be the variable regions of the antibody light and heavy chains, or the light and heavy chain variable regions can be joined together in a single chain and in either orientation (e.g., VL-VH or VH-VL). A variety of assays are known for identifying the binding domains of the present disclosure that specifically bind to a particular target, including Western blotting, ELISA, flow cytometry, or surface plasmon resonance analysis (e.g., using BIACORE™ analysis).

[0147] In some embodiments, a binding domain or a portion thereof "specifically binds" to a target molecule if it binds to or associates with the target molecule with an affinity or Ka (i.e., the equilibrium association constant of the specific binding interaction, in units of 1 / M) (e.g., greater than or equal to about 10 5 M -1 ). In certain embodiments, the binding domain or a portion thereof binds to or associates with the target molecule with an affinity 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 and 10 11 M -1 and 10 12 M -1 or 10 13 M -1 Ka binding targets of or 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 or higher. Optionally, affinity can be defined as the equilibrium dissociation constant (Kd) of a specific binding interaction with units of M (e.g., 10 -5 M to 10 - 13 M or less). The affinity of the binding domain polypeptides and portions thereof as described herein can be readily determined using conventional techniques (see, for example, Scatchard et al. (1949) Ann. N.Y. Acad. Sci. 51:660; and U.S. Patent Application Nos. 5,283,173; 5,468,614 or equivalents, which are incorporated herein by reference in their entirety).

[0148] Exemplary binding domains are described herein. In certain embodiments, the target molecule can be a cell surface-expressed protein, such as a receptor or 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, extracellular domains of receptors, ligands of cell surface molecules / receptors, or their receptor-binding domains, and tumor-binding proteins. In certain embodiments, the antigen-binding domain can be scFv, VH, VL, domain antibody variants (dAb), camelid antibodies (VHH), fibronectin type III 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).

[0149] Thus, in some embodiments, the binding domain comprises an antibody-derived binding domain, but can be a non-antibody-derived binding domain. An antibody-derived binding domain can be a fragment of an antibody or a genetically engineered product of one or more fragments of an antibody that participates in binding 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), Fab, single-domain camelid antibodies (camelid VHH), and single-domain antibodies (dAb).

[0150] The present disclosure provides a precursor trispecific antibody construct that comprises 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 that comprises a cleavable half-life extension domain and one regulatory domain that comprises 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 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 a single regulatory domain that 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 (such as but not limited to when the third domain binds to the extracellular epitope of human CD3ε) ( Figure 2F ).

[0151] 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 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 a regulatory domain that comprises a cleavable CAP (masking) domain ( Figure 2C)。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 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 a regulatory domain that comprises a cleavable half-life extension domain ( Figure 2D )。

[0152] In some embodiments, each of the first and second binding domains comprises a single-chain variable fragment (ScFv). Those skilled in the art will understand that an ScFv is not actually a fragment of an antibody, but rather a fusion polypeptide that comprises the variable heavy chain (VH) and variable light chain (VL) regions of an immunoglobulin, which are linked by a short linker peptide of 10 to about 25 amino acids ( Figure 1 and Figures 2A - 2D and 2F).

[0153] In some embodiments, the third binding domain comprises a Fab fragment, wherein 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. Optionally, the third binding domain comprises a Fab fragment, wherein 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.

[0154] In some embodiments, the third binding domain comprises a Fab fragment, wherein a first regulatory domain, such as a CAP masking domain, is attached to the N-terminus of the VL chain and a second regulatory domain, such as 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, wherein a first regulatory domain, such as a CAP masking domain, is attached to the N-terminus of the VL chain and a second regulatory domain, such as an HSA HLP domain, is attached to the N-terminus of the VH chain.

[0155] In some embodiments, the third binding domain comprises a Fab fragment, wherein a regulatory domain that comprises, 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, wherein a regulatory domain that comprises, for example, a CAP masking domain, an HSA HLP domain, and a protease linker is attached to the N-terminus of the VL chain.

[0156] 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, linker sequences may be present separately between the scFvs of the first and second binding domains and between the CL and CH1 of the third binding domain.

[0157] In some embodiments, a cleavable linker sequence may be present between the first sub-regulatory domain and the VH chain of the third binding domain. In some embodiments, a cleavable linker sequence may be present between the first sub-regulatory domain and the VL chain of the third binding domain. In some embodiments, a cleavable linker sequence may be present between the second sub-regulatory domain and the VH chain of the third binding domain. In some embodiments, a cleavable linker sequence may be present between the second sub-regulatory domain and the VL chain of the third binding domain. In some embodiments, cleavable linker sequences may be present separately between the first and second sub-regulatory domains and the VH and VL chains of the third binding domain.

[0158] In some embodiments, a cleavable linker sequence may be present between a single regulatory domain and the VH chain of the third binding domain. In some embodiments, a cleavable linker sequence may be present between a single regulatory domain and the VL chain of the third binding domain.

[0159] These general forms are the basic structures that can be constructed and can be used to construct the precursor trispecific (trimeric) antibody constructs described herein ( Figure 1 and Figure 2A 、 2B and 2F).

[0160] 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 2F)。In some embodiments, the regulatory domain comprises a protease-cleavable linker component, and either a human serum albumin polypeptide sequence component or 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, there is only a single regulatory domain comprising a protease-cleavable linker component and a CAP amino acid (masking) component ( 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 ). One of ordinary skill in the art will appreciate that the presence of a linker, such as any of the linkers shown in the constructs of Figure 1 and 2A -F, provides flexibility to the polypeptide, while not necessarily providing the necessary regulatory features of the regulatory domain, such as those provided by the CAP (masking activity) or HAS component (increasing half-life). In some embodiments, where the linker comprises a protease cleavage linker, the linker also provides a 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.

[0161] In some embodiments, the third binding domain binds specifically to the CAP amino acid component against an immune effector molecule, such as but not limited to an extracellular epitope of the anti-CD3ε chain (CD3ε). 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, the activation of cytotoxicity against the target is specifically masked by the CAP component. In some embodiments, where the regulatory domain comprises a cleavable CAP component, the activation of cytotoxicity is limited to the tumor environment ( Figures 3A - 3B ). In some embodiments, the sources of the first or second binding domain target (such as a TAA) include those shown in Figure 3B (right side).

[0162] In some embodiments, the CAP component comprises an amino acid sequence present within the human CD3ε polypeptide chain. In some embodiments, the CAP component comprises an amino acid sequence that is part of the extracellular portion of the human CD3ε chain. In some embodiments, the CAP component comprises an amino acid sequence selected from the amino acid sequence at the N-terminus of the human CD3ε precursor polypeptide. In some embodiments, the CAP component comprises an amino acid sequence selected from the amino acid sequence at the N-terminus of the human CD3ε mature polypeptide.

[0163] The amino acid sequence of precursor human CD3ε is set forth in SEQ ID NO:1. Human CD3ε is expressed in precursor form, wherein amino acids 1-21 form a signal peptide. The amino acid sequence of mature human CD3ε 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 CD3ε is located within the precursor sequence, as set forth in SEQ ID NO:3. In some embodiments, the extracellular epitope of mature human CD3ε 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 CD3ε is located within the amino acids QDGNEEMGGITQTPYKVSISGTTVILT (SEQ ID NO:5; AA1-27).

[0164] In some embodiments, the amino acid sequence of the CAP component is set forth in SEQ ID NO:5 or a homolog thereof. In some embodiments, the amino acid sequence of the CAP component is a contiguous sequence selected from SEQ ID NO:4 or a homolog thereof.

[0165] In some embodiments, the homolog of SEQ ID NO:5 or the 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.

[0166] 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 CD3ε homologous polypeptide or a portion thereof, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) with default parameters.

[0167] One skilled in the art 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 in a macromolecule with a similar monomer sequence. When used in relation to nucleic acid sequences, the term "homology" refers to the similarity between two or more nucleic acid sequences (e.g., genes) or fragments thereof. Generally, the similarity between two or more nucleic acid sequences refers to the similarity in the composition, order, or arrangement of two or more nucleotide bases (or other genotypic features) of the two or more nucleic acid sequences. The term "homologous nucleic acid" generally refers to a nucleic acid containing nucleotide sequences that have similarity in terms of nucleotide base composition, arrangement, or order. The two or more nucleic acids can be of the same or different species or groups. When used in relation to nucleic acid sequences, the term "percentage homology" generally refers to the percentage of similarity between the nucleotide sequences of two or more nucleic acids.

[0168] When used in relation to polypeptide (or protein) sequences, the term "homology" refers to the similarity between two or more polypeptide (or protein) sequences (e.g., genes) or fragments thereof. Generally, the similarity between two or more polypeptide (or protein) sequences refers to the similarity in the composition, order, or arrangement of two or more amino acids of the two or more polypeptides (or proteins). The two or more polypeptides (or proteins) can be of the same or different species or groups. When used in relation to polypeptide (or protein) sequences, the term "percentage homology" generally refers to the percentage of similarity between the amino acid sequences of two or more polypeptides (or proteins). The terms "homologous polypeptide" or "homologous protein" generally refer to polypeptides or proteins having similar amino acid sequences and functions, respectively. Such homologous polypeptides or proteins can be related by having similar amino acid sequences and functions, but are derived or evolved from different or the same species using the techniques described herein.

[0169] 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 the polypeptide or a portion thereof disclosed herein, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) with 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 the nucleic acid sequence or a portion thereof disclosed herein, as determined using the BlastN software of the National Center for Biotechnology Information (NCBI) with default parameters.

[0170] In some embodiments, homology also includes deletion, insertion, or substitution variants thereof and of its bioactive polypeptide fragments, including amino acid substitutions. In one embodiment, the variant comprises conservative substitutions, or deletions, insertions, or substitutions that do not significantly alter the three-dimensional structure of the polypeptide component of interest described herein. In some embodiments, the deletion, insertion, or substitution does not alter the function of interest of the polypeptide component of interest disclosed herein.

[0171] In some embodiments, homology also includes deletion, insertion, or substitution variants thereof and of its bioactive polypeptide fragments, including amino acid substitutions. In one embodiment, the variant comprises conservative substitutions, or deletions, insertions, or substitutions that do not significantly alter the three-dimensional structure of the CAP component, such as the 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, which in some embodiments binds to the third binding domain, or reduces T cell binding, or reduces T cell activation, or any combination thereof.

[0172] In some embodiments, the length of the CAP component is 6 - 110 amino acids. In some embodiments, the length of the CAP component is between about 6 - 10 amino acids. In some embodiments, the length of the CAP component is between about 10 - 20 amino acids. In some embodiments, the length of the CAP component is between about 20 - 30 amino acids. In some embodiments, the length of the CAP component is between about 20 - 40 amino acids. In some embodiments, the length of the CAP component is between about 30 - 40 amino acids. In some embodiments, the length of the CAP component is between about 40 - 60 amino acids. In some embodiments, the length of the CAP component is between about 60 - 80 amino acids. In some embodiments, the length of the CAP component is between about 80 - 100 amino acids. In some embodiments, the length of the CAP component is between about 80 - 110 amino acids.

[0173] In some embodiments, the length of the CAP component is 6 amino acids. In some embodiments, the length of the CAP component is 7 amino acids. In some embodiments, the length of the CAP component is 8 amino acids. In some embodiments, the length of the CAP component is 9 amino acids. In some embodiments, the length of the CAP component is 10 amino acids. In some embodiments, the length of the CAP component is 11 amino acids. In some embodiments, the length of the CAP component is 12 amino acids. In some embodiments, the length of the CAP component is 13 amino acids. In some embodiments, the length of the CAP component is 14 amino acids.

[0174] In some embodiments, the length of the CAP component is 15 amino acids. In some embodiments, the length of the CAP component is 16 amino acids. In some embodiments, the length of the CAP component is 17 amino acids. In some embodiments, the length of the CAP component is 18 amino acids. In some embodiments, the length of the CAP component is 19 amino acids. In some embodiments, the length of the CAP component is 20 amino acids. In some embodiments, the length of the CAP component is 21 amino acids. In some embodiments, the length of the CAP component is 22 amino acids. In some embodiments, the length of the CAP component is 23 amino acids. In some embodiments, the length of the CAP component is 24 amino acids. In some embodiments, the length of the CAP component is 25 amino acids. In some embodiments, the length of the CAP component is 26 amino acids. In some embodiments, the length of the CAP component is 27 amino acids. In some embodiments, the length of the CAP component is 28 amino acids. In some embodiments, the length of the CAP component is 29 amino acids. In some embodiments, the length of the CAP component is 30 amino acids. In some embodiments, the length of the CAP component is 31, 32, 33, 34, 35, 36, 37, 38, 39, or 40 amino acids.

[0175] In some embodiments, the CAP component specifically binds to the third binding region, thereby reducing the T cell binding of the precursor construct. In some embodiments, the CAP component specifically binds to the third binding region, thereby inhibiting the T cell binding of the precursor construct. In some embodiments, the CAP component specifically binds to the third binding region, thereby reducing the T cell activation of the precursor construct. In some embodiments, the CAP component specifically binds to the third binding region, thereby inhibiting the T cell activation of the precursor construct.

[0176] In some embodiments, the regulatory domain comprises a cleavable half-life extension domain. In some embodiments, the cleavable half-life extension domain comprises an HSA polypeptide.

[0177] In some embodiments, there is a linker between 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).

[0178] In some embodiments, the regulatory domain comprises a cleavable half-life extension domain that comprises a protease-cleavable domain and a human serum albumin polypeptide (HSA). In some embodiments, the order of components in the regulatory domain is (N-terminus to C-terminus) 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 that contains a CAP amino acid (masking) component, a protease-cleavable domain and a human serum albumin polypeptide (HSA). In some embodiments, the order of components in the regulatory domain is (N-terminus to C-terminus) CAP-L-HSA-L-protease-cleavable domain, where L is a possible linker amino acid sequence( Figure 2F ). In some embodiments, where the protease-cleavable domain is at the C-terminus of the HSA polypeptide sequence, the precursor construct has an adjustable enhanced half-life, where the precursor construct has an enhanced half-life in the in vivo circulation and in the absence of the tumor microenvironment.

[0179] In some embodiments, the regulatory domain comprises a cleavable half-life extension domain that comprises a protease-cleavable domain and a CAP masking domain. In some embodiments, the order of components in the regulatory domain is (N-terminus to C-terminus) CAP-L-protease-cleavable domain, where L is a possible linker amino acid sequence( Figure 1 , 2A, 2B, and 2C). In some embodiments, the component order in the regulatory domain is (N-terminus to C-terminus) CAP-L-HSA-L-protease-cleavable domain, where L is a possible linker amino acid sequence. Prior to the precursor antibody entering the tumor microenvironment, the trispecific precursor construct is effectively prevented from binding to immune effector target molecules.

[0180] In some embodiments, there is one regulatory domain having a cleavable half-life extension domain and a CAP masking domain. In some embodiments, there are two regulatory domains: one having a cleavable half-life extension domain and one having a cleavable CAP masking domain. In some embodiments, there are three regulatory domains: one having a cleavable half-life extension domain and two having cleavable CAP masking domains. 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 during circulation in the body and in the absence of a tumor environment. In some embodiments, the activation of cytotoxicity of the precursor trispecific antibody construct is limited to the tumor environment. In some embodiments, the precursor construct maintains an enhanced half-life during circulation in the body and effectively prevents binding to immune effector target molecules during circulation in the body in a non-tumor environment ( Figures 3A - 3B ). In some embodiments, the activation of cytotoxicity to the target is specifically masked by the CAP component of the precursor construct during circulation and in the absence of a tumor environment, and the precursor construct has an enhanced half-life during circulation in the body and in the presence or absence of a tumor environment. In some embodiments, the activation of cytotoxicity is limited to the tumor environment.

[0181] In some embodiments, the activation of cytotoxicity to the target is specifically masked by the CAP component during circulation and in a non-tumor environment. In some embodiments, the activation of cytotoxicity is limited to the tumor environment. In some embodiments, the activation of T cells is specifically masked by the CAP component.

[0182] In some embodiments, the amino acid sequence of the HSA component is set forth in SEQ ID NO:6. In some embodiments, the amino acid sequence of the HSA component is set forth in SEQ ID NO:7.

[0183] 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 comprises, for example but not limited to, any human albumin sequence disclosed in a known database (e.g., a 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 as a synthetic construct.

[0184] In some embodiments, the HSA component is encoded by the nucleotide sequence set forth in SEQ ID NO:8.

[0185] 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 a known database (e.g., a 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 as a synthetic construct.

[0186] In some embodiments, homologs of the HSA component comprise polypeptides 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 an amino acid sequence. In some embodiments, the homologs comprise polypeptides 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 an HAS polypeptide or a portion thereof, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) with default parameters. In some embodiments, the nucleic acids encoding the homologs of the HSA component comprise 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 homologs encode polypeptides 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 an HAS polypeptide or a portion thereof, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) with default parameters.

[0187] In some embodiments, homology also includes deletion, insertion, or substitution variants thereof and of its bioactive polypeptide fragments, including amino acid substitutions. In one embodiment, the variant comprises conservative substitutions, or deletions, insertions, or substitutions that do not significantly alter the three-dimensional structure of the HSA component. In some embodiments, the deletion, insertion, or substitution does not alter the function of interest of the HSA component, and in some embodiments, it provides a half-life extending domain.

[0188] Linear representations of embodiments of the regulatory domain of the precursor trispecific antibody constructs disclosed herein include, but are not limited to, (N-terminus to C-terminus)

[0189] (1) CAP-L-protease-cleavable domain-L, where L may or may not be present;

[0190] (2) HSA-L-protease-cleavable domain-L, where L may or may not be present;

[0191] (3) CAP-L-non-cleavable domain-L, where L may or may not be present;

[0192] (4) HSA-L-non-cleavable domain-L, where L may or may not be present;

[0193] (5) Protease-cleavable domain-L-CAP;

[0194] (6) Protease-cleavable domain-L-HSA;

[0195] (7) Non-cleavable domain-L-CAP; and

[0196] (8) Non-cleavable domain-L-HSA.

[0197] In some embodiments, the precursor trispecific antibody constructs disclosed herein comprise precursor constructs having an improved therapeutic window, wherein their restricted presence provides the ability to target a broad range of new targets or provides improved activity or a combination thereof, such as but not limited to, the ability to activate T cells only in the cancer microenvironment, and to target cancer-specific TAAs according to the cancer type, as well as specific TAAs uniquely expressed by proteases produced by binding to that cancer type. In some embodiments, the precursor construct has the ability to activate T cells only in the TME and targets cancer-specific TAAs and different TAAs present in the TME.

[0198] As used herein, the "C-terminus" of a polypeptide, such as carboxyl-terminus, carboxy-terminus, C-terminal tail, C-terminus or COOH-terminus, is the end of an amino acid chain (protein or polypeptide) 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 place the C-terminus on the right and write the sequence from the N- to the C-terminus. In some embodiments, the C-terminus of a polypeptide includes up to the last amino acid residue of the polypeptide, which contributes its amino group to form a peptide bond with the carboxyl group of its adjacent amino acid residue.

[0199] As used herein, the "N-terminus" of a polypeptide, such as amino-terminus, NH 2 -terminus, N-terminus or amine-terminus, is the start of a protein or polypeptide, which refers to the free amino group (-NH 2)。Generally, an amino group bonds to another carboxyl group in a protein to form a chain. However, since only one of the two ends of the protein is chained, the free amino group refers to the N-terminus. As described above, by convention, peptide sequences are written from the N-terminus to the C-terminus, from left to right in the LTR language. This associates the translation direction with the text direction (because when a protein is translated from messenger RNA, it is produced from the N-terminus to the C-terminus - amino acids are added to the carbonyl end). In some embodiments, the N-terminus of a polypeptide contains the first amino acid of the polypeptide, which contributes its carboxyl group to form a peptide bond with the amino group of an adjacent amino acid residue.

[0200] One of ordinary skill in the art will understand that the linker assembly can include amino acid peptides that are connected by one or more chemical bonds or indirectly connected through one or more linkers. Any suitable chemical bond can be used for direct connection, including but not limited to covalent bonds such as peptide bonds and disulfide bonds, and non-covalent bonds such as hydrogen bonds, hydrophobic bonds, ionic bonds, and van der Waals bonds.

[0201] "Covalent bond" as used herein 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, "peptide bond" refers to a covalent bond formed between the carboxyl group of an amino acid and the amino group of an adjacent amino acid. As used herein, "disulfide bond" refers to a covalent bond formed between two sulfur atoms. A disulfide bond can be formed by the oxidation of two thiol groups. In certain embodiments, the covalent connection is a direct connection through a covalent bond. In certain embodiments, the covalent connection is a direct connection through a peptide bond or a disulfide bond.

[0202] "Non-covalent bond" as used herein refers to an attractive interaction between two molecules or two chemical groups that does not involve electron sharing. Examples of non-covalent bonds include but are not limited to hydrogen bonds, hydrophobic bonds, ionic bonds, and van der Waals bonds. "Hydrogen bond" as used herein refers to the attraction between a hydrogen atom of a first molecule / group and an electronegative atom of a second molecule / group. "Hydrophobic bond" as used herein refers to the force that causes hydrophobic or non-polar molecules / groups to aggregate or associate together in an aqueous environment. "Ionic bond" as used herein refers to the attraction between a cation and an anion. "Van der Waals bond" as used herein refers to a non-specific attraction between two adjacent molecules / groups that have transient random fluctuations in their electron distribution. In certain embodiments, the covalent connection is a direct connection through a non-covalent bond. In certain embodiments, the covalent connection is a direct connection through a hydrogen bond, a hydrophobic bond, an ionic bond, or a van der Waals bond.

[0203] One of ordinary skill in the art will understand that the protease-cleavable domains described herein include linkers that contain protease cleavage sites. Thus, the terms "protease-cleavable domain" and "protease-cleavable linker" are used interchangeably herein and have the same meaning and properties.

[0204] Those skilled in the art will understand that the terms "tumor microenvironment", "cancer microenvironment", "TME", and "tumor environment" can be used interchangeably, have the same nature and meaning, and encompass the microenvironment of tumor development. While the normal cell microenvironment can inhibit the growth of malignant cells, the alterations occurring in the tumor microenvironment can synergistically support cell proliferation.

[0205] The tumor shapes its microenvironment and supports 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 (EPCs) for angiogenesis under hypoxic conditions is also associated with metastasis. In some embodiments, the TAA comprises a cell surface antigen associated with angiogenesis. In some embodiments, the TAA is overexpressed by cancer cells. In some embodiments, the TAA is expressed on embryonic cells. In some embodiments, the TAA is expressed on both embryonic cells and cancer cells, but not expressed or only minimally expressed on normal adult cells. In some embodiments, the TAA is expressed on solid tumor cells. In some embodiments, the TAA is expressed on non-solid cancer cells. In some embodiments, the TAA is expressed on angiogenic tissue cells.

[0206] In addition, the proteins secreted by the tumor alter the microenvironment by contributing growth factors and proteases that degrade the extracellular matrix and affect cell movement 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 related to cancer development. Tumor-associated epithelial cells express MMPs at higher levels compared to normal epithelial cells. In some embodiments, the microenvironment of the tumor contains increased protease activity compared to the non-tumor environment.

[0207] Figure 3A and 3B Non-limiting examples of precursor trispecific antibody constructs are provided that can have an enhanced half-life in vivo and in the non-tumor environment. Further, when the precursor construct is in the non-tumor environment, the anti-CD3 third binding domain of the precursor trispecific antibody construct is blocked and may not interact or bind with target T cells. In some embodiments, as shown by the scissors in Figure 3A and just before Figure 3B T cell activation in Figure 2E , the cancer microenvironment provides protease cleavage of the precursor trispecific antibody construct, which removes the half-life extension regulatory domain (HSA) and the CD3εCAP regulatory domain, resulting in the presence of the activated EGFR(2scFvs)xCD3ε antibody (

[0208] In some embodiments, the protease-cleavable domain comprises a protease-cleavable amino acid sequence (cleavable peptide / cleavable linker; CP) that comprises a peptide cleavable by a serine protease, a cysteine protease, an aspartic protease, or a matrix metalloproteinase (MMP) cleavage sequence. In some embodiments, the protease-cleavable domain comprises a protease-cleavable amino acid sequence (cleavable peptide / cleavable linker; CP) that comprises a peptide that is a substrate for cleavage by a variety of different proteases. In some embodiments, the protease-cleavable domain comprises a protease-cleavable amino acid sequence (cleavable peptide / cleavable linker; CP) that comprises a peptide that is a substrate for cleavage by MMP2 / MMP9 proteases or urokinase-type plasminogen activator (uPA) protease, or proteinase, or asparaginyl endopeptidase protease. In some embodiments, serine protease, cysteine protease, aspartic protease, uPA protease, proteinase, asparaginyl endopeptidase protease, or matrix metalloproteinase (MMP) is expressed at a higher level in the tumor microenvironment. In some embodiments, matrix metalloproteinase is expressed at a higher level in the tumor microenvironment.

[0209] In some embodiments, the protease-cleavable sequence is an 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.

[0210] In some embodiments, the protease-cleavable sequence is a uPA (urokinase-type plasminogen activator) cleavable sequence. In some embodiments, the protease-cleavable sequence is an MT-SP1 (proteinase) cleavable sequence.

[0211] In some embodiments, the protease-cleavable sequence is an MMP, uPA, proteinase, and asparaginyl endopeptidase cleavable sequence.

[0212] 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 length of the protease-cleavable domain is 1 to 20 amino acids.

[0213] 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 substrates. Substrate sequences that can be cleaved by MMPs have been widely 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 MMPs. 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 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 includes 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.

[0214] In some embodiments, the protease-cleavable sequence comprising a uPA-cleavable sequence comprises a sequence 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).

[0215] In some embodiments, the protease-cleavable sequence comprising a proteolytic enzyme-cleavable sequence comprises a sequence recited in QRRVVGG (SEQ ID NO:27), QAR,AANL (SEQ ID NO:29), PTNL (SEQ ID NO:30), PTN, or SAN.

[0216] In some embodiments, the cleavable peptide is encoded by the nucleic acid sequence recited in SEQ ID NO:33: CCACTGGGCCTGGCCGGC.

[0217] In some embodiments, the amino acid sequence of the protease-cleavable sequence that serves as a substrate for MMP2 / 9, uPA, proteolytic enzyme, and asparaginyl endopeptidase-cleavable sequences is recited in PLGLAGSGRSDNH (SEQ ID NO:35). In some embodiments, all protease-cleavable sequences included in the precursor construct comprise SEQ ID NO:35. In some embodiments, at least one of the protease-cleavable sequences included in the precursor construct comprises SEQ ID NO:35. In some embodiments, at least 2 of the protease-cleavable sequences included in the precursor construct comprise SEQ ID NO:35. In some embodiments, at least 3 of the protease-cleavable sequences included in the precursor construct comprise SEQ ID NO:35.

[0218] In some embodiments, the sequence of the protease-cleavable peptide component of regulatory domain one is the same as that of the protease-cleavable peptide component of regulatory domain two. In some embodiments, the sequence of the protease-cleavable peptide component of regulatory domain one is different from that of the protease-cleavable peptide component of regulatory domain two. In some embodiments, the protease that cleaves the protease-cleavable peptide component of regulatory domain one is the same as the protease that cleaves the protease-cleavable peptide component of regulatory domain two. In some embodiments, the protease that cleaves the protease-cleavable peptide component of regulatory domain one is different from the protease that cleaves the protease-cleavable peptide component of regulatory domain two.

[0219] In some embodiments, the protease that cleaves the first and second sub-regulatory domains is an 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 proteolytic enzyme protease. In some embodiments, one of the first or second sub-regulatory domains is cleaved by an 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 an 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 an MMP protease, while the other regulatory domain is cleaved by a proteolytic enzyme 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.

[0220] A stable linker or protease-resistant linker refers to a linker peptide sequence that does not belong to the known protease substrate sequence and thus does not result in significant cleavage product formation when incubated with proteases.

[0221] In some embodiments, the cleavage substrate (or cleavage sequence) of the linker may include an amino acid sequence that can serve as a substrate for a protease (usually an extracellular protease). In other embodiments, the cleavage sequence contains a cysteine-cysteine pair capable of forming a disulfide bond, which can be cleaved by the action of a reducing agent. In other embodiments, the cleavage sequence contains a substrate that can be cleaved upon photolysis.

[0222] The cleavage substrate is located within the protease-cleavable domain 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 disrupted via reduction upon exposure to a reducing agent) or cleaved by photoinduced photolysis, in the presence of a target, a cleavage product having various functional properties as described herein is produced. 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 ).

[0223] The cleavage substrate of the cleavage domain can be selected based on a protease that is co-localized in diseased tissue or on the cell surface of an interesting target antigen that expresses the binding domain of the fusion moiety. A variety of different conditions are known in which the interesting target is co-localized with the protease, and the substrates of the protease are known in the art. In the example of cancer, the target tissue can be cancerous tissue, specifically cancerous tissue of a solid tumor. Elevated levels of proteases with known substrates have been reported in many cancers (e.g., solid tumors) in the literature. 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 injury, ischemia, etc. In addition, anti-angiogenic targets, such as VEGF, are known.

[0224] In some embodiments, when the TAAs of the first or second binding domain are selected such that they can bind to tumor antigens, a suitable cleavage substrate sequence for the linker will be a cleavage substrate sequence that contains a peptide substrate cleavable by a protease that is present at the cancer treatment site, i.e., the tumor microenvironment, and is present at an elevated level at the cancer treatment site compared to non-cancerous tissue.

[0225] In some embodiments, the first or second, or both the first and second binding domains of the precursor constructs disclosed herein can bind to TAAs, such as EGFR, and the cleavage substrate sequence can be a matrix metalloproteinase (MMP) substrate and thus cleavable by MMP. In other embodiments, the TAA contains ROR1 and the cleavage substrate sequence can be a proteinase (MT-SP1, TADG-15, epithin, ST14) substrate and thus cleavable by the proteinase. In other embodiments, the first or second, or both the first and second binding domains of the precursor construct can bind to the interesting target, and the cleavage substrate present in the cleavable domain can be, for example, asparaginyl 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.

[0226] In some embodiments, the precursor trispecific antibody construct can bind to the 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 due to the presence of the CAP component, the third binding domain of the precursor construct or the partially cleaved precursor construct may be specifically unavailable to the target CD3ε antigen. In some embodiments, the precursor trispecific antibody construct can bind to the 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 due to the presence of a half-life extension domain (such as an HSA polypeptide sequence), the precursor construct or the partially cleaved precursor construct has an enhanced half-life, and the third binding domain is available or partially available to the target CD3ε antigen. In some embodiments, the precursor trispecific antibody construct can bind to the TAA via the first or second, or both the first and second binding domains, wherein the cleavable domains of both regulatory arms remain uncleaved, wherein due to the presence of a half-life extension domain (such as, an HSA polypeptide sequence), the precursor construct has an enhanced half-life, and due to the presence of the CAP component, the third binding domain remains specifically unavailable to the target CD3ε antigen.

[0227] In some embodiments, there is a linker (L)( Figure 1 and 2A -2D) between any component parts of the precursor trispecific antibody construct. In some embodiments, the linker of the precursor construct (e.g., the linker between the VL and VH of the Fab and the regulatory domain) comprises a cleavable domain linker. In some embodiments, for each linker, the ability to be cleaved is independently selected. In some embodiments, for each linker, the ability to be cleaved by a protease is independently selected. In some embodiments, the linker is cleavable by a protease. In some embodiments, the linker is not cleavable by a protease. In some embodiments, the linker between the CH1 or CL of the 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 the Fab and the ScFv of the second binding domain comprises a non-cleavable linker.

[0228] One of ordinary skill in the art will understand that in some embodiments, the linker comprises a spacer between two active components or two regions of an active component.

[0229] One of ordinary skill in the art will understand that the cleavable domain comprises a linear amino acid sequence containing an enzyme cleavage site and may be referred to as a "cleavable linker" or "linker" or "cleavable peptide" or "CP" in certain embodiments, wherein the linkers disclosed herein can be cleavable or non-cleavable.

[0230] 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.

[0231] In some embodiments, the linker is a single amino acid. In some embodiments, the cleavable linker comprises the amino acid sequence recited in any of SEQ ID NO:9-32. In some embodiments, the cleavable linker is encoded by the nucleic acid sequence recited in SEQ ID NO:33. In some embodiments, the cleavable linker is encoded by the nucleic acid sequence recited in SEQ ID NO:35.

[0232] In some embodiments, the non-cleavable linker comprises the amino acid sequence recited in SEQ ID NO:162. In some embodiments, the non-cleavable linker is encoded by the nucleic acid sequence recited in SEQ ID NO:163.

[0233] For specific cleavage by the enzyme protease, contact is made between the enzyme and the cleavage substrate. When there is sufficient enzyme activity in the precursor construct comprising a first and a second binding domain that binds TAA, a third binding domain that binds to the extracellular epitope of CD3ε, and two regulatory domains comprising a cleavable linker, the cleavable domain can be cleaved. Sufficient enzyme activity can refer to the ability of the enzyme to contact the protease-cleavable domain with a cleavage site and effect cleavage. In some embodiments, the enzyme may be near the precursor construct but unable to cleave due to other cellular factors or protein modification of the enzyme.

[0234] In some embodiments, the cleavable domain substrate may include, but is not limited to, substrates cleavable 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 enkephalinase (HNE), caspase, caspase 14, asparaginyl endopeptidase, NS3 / 4A, caspase 1, cathepsin, proteinase 2, plasmin, caspase 2, cathepsin A, transmembrane peptidase, 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, proteinases (MT-SP1, TADG-15, epithin, ST14), and MT1-MMP.

[0235] In another embodiment, cleavage of the substrate may involve a disulfide bond of a cysteine pair and thus may 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 amounts in the tissue of or around solid tumors.

[0236] Other suitable protease cleavage sites for the cleavable linker used herein are known in the art or can be identified using, for example, the method described by Turk et al., 2001 Nature Biotechnology 19, 661-667.

[0237] In some embodiments, the first, second, and third binding domains of the precursor trispecific antibody construct can all bind their respective human and non-chimpanzee primate target molecules. Thus, the first binding domain and / or the second binding domain bind a human cell surface tumor-associated antigen (TAA) and the corresponding homolog of the cell surface TAA in non-chimpanzee primates. The identification and determination of the homologs of human cell surface TAAs 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 an antigen comprising an extracellular epitope of human CD3ε and can bind the corresponding homolog of CD3ε in non-chimpanzee primates. In some embodiments, the first, second, or third binding domain, or any combination thereof, also binds its respective chimpanzee target molecule.

[0238] Those skilled in the art will understand that in some embodiments, the cell surface tumor-associated antigen (TAA) includes molecules presented on the cell surface. In some embodiments, the cell is a tumor cell. In some embodiments, the cell is a non-tumor cell present in the tumor microenvironment, such as but not limited to cells present in the vascular tissue associated with a tumor or cancer.

[0239] Those skilled in the art will understand that the term "antigen" or "immunogen" includes peptides, proteins, polypeptides having immunogenicity. In some embodiments, the antigen is capable of eliciting an immune response in a mammal and thus contains at least one and may contain multiple epitopes. An "antigen" molecule or a portion of a molecule can be bound by a selective binder, such as the antigen-binding portion of a Fab fragment or the antigen-binding portion of an scFv fragment. In addition, an "antigen" can be used in an animal to generate antibodies capable of binding the epitopes of the antigen. In some embodiments, the CAP component contains a portion of the antigen bound by the second binding domain.

[0240] In certain embodiments, the term "epitope" includes any determinant capable of specifically binding to a TAA or an immunoglobulin or a T cell receptor. An epitope is the antigenic region bound by an antibody or its antigen-binding fragment. In some embodiments, the CAP component contains the epitope bound by the third binding domain.

[0241] In certain embodiments, the epitope determinant includes the chemically active surface grouping of a molecule, such as an amino acid, a sugar side chain, a phosphoryl group, or a sulfonyl group, and in certain embodiments may have specific three-dimensional structural features and / or specific charge features. In certain embodiments, when the 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 the antigen. When the equilibrium dissociation constant ≤ 10 -5 、10 -6 or 10-7 At time M, the precursor trispecific antibody construct is referred to as specifically binding an 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 can 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 the human CD3ε polypeptide.

[0242] 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.

[0243] In some embodiments, when the TAA is a tumor cell antigen, the tumor cells include cells from solid tumors. Solid tumors can be benign (not cancerous) or malignant (cancerous). 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 neoplasm (new growth of cells) or lesion (damage to an anatomical structure or disorder of a physiological function) formed by the abnormal growth of cells of body tissues other than blood, bone marrow, or lymphocytes. In some embodiments, a solid tumor consists of an abnormal population of cells that may be derived from different tissue types, such as the liver, colon, breast, or lung, and initially grows in the organ from which its cells are derived. However, these cancers may spread to other organs through metastatic tumor growth in the advanced stages of the disease.

[0244] In some embodiments, solid tumors include sarcoma or carcinoma, fibrosarcoma, myxosarcoma, liposarcoma, chondrosarcoma, osteosarcoma, chordoma, angiosarcoma, endotheliosarcoma, lymphangiosarcoma, lymphangioendotheliosarcoma, synovioma, mesothelioma, Ewing's 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, cholangiocarcinoma, choriocarcinoma, seminoma, embryonal carcinoma, Wilms' 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. In some embodiments, solid tumors include adrenocortical tumors (adenomas and carcinomas), carcinoma, colorectal cancer, desmoid tumor, desmoplastic small round cell tumor, endocrine tumor, Ewing's 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.

[0245] In some embodiments, tumor cells include cells from non-solid tumors, i.e., non-solid cancers. In some embodiments, the cancer can be a diffuse cancer, where the cancer spreads widely; is not localized or restricted. In some embodiments, diffuse cancers can include non-solid tumors. Examples of diffuse cancers include leukemia. Leukemia includes cancers that start in hematopoietic tissues (such as the bone marrow) and result in the production of large numbers of abnormal blood cells that enter the bloodstream.

[0246] In some embodiments, diffuse cancers include B cell malignancies. In some embodiments, diffuse cancers include leukemia. In some embodiments, the cancer is lymphoma. In some embodiments, the lymphoma is large B cell lymphoma.

[0247] In some embodiments, diffuse cancers or tumors include hematologic malignancies. In some embodiments, hematologic malignancies are types of cancer that affect the blood, bone marrow, and lymph nodes. Hematologic malignancies can originate from either of two major blood cell lineages: the myeloid and lymphoid lineages. The myeloid lineage typically gives rise to granulocytes, red blood cells, platelets, macrophages, and mast cells, while the lymphoid lineage gives rise to B, T, and plasma cells. Lymphomas (such as Hodgkin lymphoma), lymphocytic leukemias, and myelomas originate from the lymphoid lineage, while acute and chronic myeloid leukemias (AML, CML), myelodysplastic syndromes, and myeloproliferative disorders are of myeloid origin.

[0248] In some embodiments, non-solid (diffuse) cancers or tumors include hematopoietic malignancies, blood cell cancers, leukemias, myelodysplastic syndromes, lymphomas, multiple myelomas (plasma cell myelomas), acute lymphocytic leukemia, acute myeloid leukemia, chronic myeloid leukemia, Hodgkin lymphoma, non-Hodgkin lymphoma, or plasma cell leukemia.

[0249] In some embodiments, a tumor or cancer comprises metastases of the tumor or cancer.

[0250] In some embodiments, a 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 in a tertiary form from the exterior of the cell. In some embodiments, a cell surface TAA located in the plasma membrane is a transmembrane protein that includes a hydrophilic region and a hydrophobic region in its tertiary conformation.

[0251] These antigens can be presented on the cell surface together with an extracellular portion that is typically associated with the transmembrane and cytoplasmic portions of the molecule. These antigens are sometimes presented only by tumor cells and never by normal cells. Tumor antigens can be specifically expressed 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 commonly, antigens are presented by both tumor cells and normal cells. In some embodiments, TAAs include antigens specifically expressed on tumor cells. In some embodiments, TAAs include antigens expressed on both tumor cells and normal cells.

[0252] In some embodiments, compared to normal cells, TAAs can be overexpressed on tumor cells or, due to the less compact structure of tumor tissue compared to normal tissue, TAAs can be more readily antibody-bound in tumor cells.

[0253] In some embodiments, the first binding domain or the second binding domain or both that bind to the cell surface TAA comprise amino acid sequences that bind to human TAA. In some embodiments, the anti-scFv comprises a heavy chain variable region and a light chain variable region, each of which further comprises complementarity determining regions (CDRs). In some embodiments, the first binding domain or the second binding domain or both that bind to the cell surface TAA comprise a linker between the scFv variable light chain (VL) region and the scFv variable heavy chain (VH) region. In some embodiments, the first binding domain or the second binding domain that binds to the cell surface TAA comprises a linker between the C-terminus of the CL region of the scFv fragment and the third binding domain. In some embodiments, the first binding domain or the second binding domain that binds to the cell surface TAA comprises a linker between the C-terminus of the CH1 region of the scFv fragment and the third binding domain. In some embodiments, the first binding domain that binds to the cell surface TAA comprises a linker between the C-terminus of the CH1 region of the scFv fragment and the third binding domain, and the second binding domain that binds to the cell surface TAA comprises a linker between the C-terminus of the CL region of the scFv fragment and the third binding domain. In some embodiments, the first binding domain that binds to the cell surface TAA comprises a linker between the C-terminus of the CL region of the scFv fragment and the third binding domain, and the second binding domain that binds to the cell surface TAA comprises a linker between the C-terminus of the CH1 region of the scFv fragment and the third binding domain.

[0254] Figures 2A - 2FDisplay embodiments, where the TAA is EGFR (the first and second binding domains contain anti-EGFR scFv). Those skilled in the art will understand that in other embodiments, the TAA can be a TAA known in the art, such as but not limited to TAAs including: 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;Erb2), 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, 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, 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 includes 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.;

[0255] Trispecific antibodies can be designed to bind to at least one tumor-associated antigen (TAA), which, in some embodiments, includes a tumor cell surface antigen, a T cell antigen, and a second TAA, with the aim of the antibody binding and killing tumor cells (more selectively than normal cells) and ultimately improving the efficacy and safety of monospecific reagents, 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 binding order and may thus bind to T cells before or in the absence of binding to a TAA, wherein the cytotoxicity provided by activated T cells may actually cause harmful side effects by non-specifically causing non-tumor cell death. In some embodiments, the TAA comprises a human antigen.

[0256] In some embodiments, the TME antigen comprises a KIR, LILR, or TIGIT antigen.

[0257] In some embodiments, the first binding domain or the second binding domain or both that bind to cell surface TAAs bind to polypeptide targets, which in some embodiments are associated with one or more specific cancers or disease conditions, such as, but not limited to, TAAs including: 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; various malignancies, including ovarian cancer and colorectal cancer) and TRAIL receptor 2, prostate-specific membrane antigen (PSMA; prostate cancer) protein, prostate stem cell antigen (PSCA) protein (prostatic adenocarcinoma), CA125 (various cancers, including ovarian cancer), tumor-associated protein carbonic anhydrase IX (CAIX; 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, CD20 (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) (multiple tumors, including sarcomas, ovarian cancer and pancreatic cancer), Cripto (TDGF1) (epithelial malignancies, including colon cancer, breast cancer, lung cancer, ovarian cancer and pancreatic cancer), ACVRL1 / ALK1 (multiple malignancies, including leukemia and lymphoma), MUC5AC (epithelial malignancies, including breast adenocarcinoma), PRAME, P15, RU1, RU2, SART-1, SART-3, WT1, AFP, beta-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 cancer and pancreatic cancer), mesothelin (many tumors, including mesothelioma and ovarian and pancreatic cancer), DR5 (multiple malignancies, including ovarian cancer 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 cancer, brain cancer and breast cancer), EphA2 (multiple cancers,including neuroblastoma, melanoma, breast cancer, and small cell lung cancer), CD138 (multiple myeloma), B7-H3 (CSC, stroma, NSCLC, bladder tumor, mesothelioma, melanoma), gpA33 (colorectal cancer), GPC3 (liver cancer, lung cancer, esophageal cancer, gastric cancer, head and neck cancer), SSTR2 (neuroendocrine tumors, GIST), ROR1 (hematological 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 (vascular system associated with most malignant tumors including epithelial adenocarcinoma). Examples of harmful target cells or cancer cells associated with the presented TAA are included in italics in parentheses.,

[0258] In some embodiments, the TAA is selected from EGFR, ROR1, PSMA, and 5T4. In some embodiments, the first binding domain comprises an scFv that binds to human EGFR (anti-hEGFR), or human ROR1 (anti-ROR1), or human PSMA (anti-PSMA), or human 5T4 (anti-5T4).

[0259] In some embodiments, the TAA is EGFR. In some embodiments, the first binding domain comprises an scFv that binds to human EGFR (anti-hEGFR). In some embodiments, the amino acid sequence of the light chain variable region (VL) of anti-hEGFR-scFv 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.

[0260] In some embodiments, the light chain variable region (VL) of anti-hEGFR-scFv is encoded by the nucleic acid sequence set forth in SEQ ID NO:36. In some embodiments, the light chain variable region (VL) of anti-hEGFR-scFv is encoded by a homolog of the nucleic acid sequence set forth in SEQ ID NO:36.

[0261] In some embodiments, the amino acid sequence of the heavy chain variable region (VH) of 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.

[0262] In some embodiments, the heavy chain variable region (VH) of anti-hEGFR-scFv is encoded by the nucleic acid sequence set forth in SEQ ID NO:38. In some embodiments, the heavy chain variable region (VH1) of anti-hEGFR-scFv is encoded by a homolog of the nucleic acid sequence set forth in SEQ ID NO:38.

[0263] 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 recited by 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 recited 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 recited in SEQ ID NO:40.

[0264] 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 ).

[0265] In some embodiments, the anti-EGFR scFV sequence comprising the linker sequence comprises the sequence SEQ ID NO:41. In some embodiments, the anti-EGFR scFv comprising the linker sequence comprises a homolog of SEQ ID NO:41.

[0266] In some embodiments, the anti-EGFR scFV sequence comprising the linker sequence comprises the sequence SEQ ID NO:42. In some embodiments, the anti-EGFR scFv comprising the linker sequence comprises a homolog of SEQ ID NO:42.

[0267] In some embodiments, the amino acid sequences of the light chain variable regions of anti-hROR1-scFv, or anti-PSMA-scFv, or anti-5T4-scFv are recited in Table 1 below:

[0268] Table 1: Amino acid sequences and optimized nucleotide sequences encoding anti-hROR1-scFv, or anti-PSMA-scFv, or anti-5T4-scFv

[0269] 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

[0270] 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 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 the anti-EGFR scFv, or any anti-ROR1 scFv, or anti-PSMA scFv, or anti-5T4 scFv.

[0271] In some embodiments, the nucleotide sequence encoding the anti-EGFR scFV comprising a linker sequence comprises SEQ ID NO: 43. In some embodiments, the anti-EGFR scFv comprising a linker sequence comprises a homolog of SEQ ID NO: 43.

[0272] In some embodiments, the nucleotide sequence encoding the anti-EGFR scFV comprising a linker sequence comprises SEQ ID NO: 44. In some embodiments, the anti-EGFR scFv comprising a linker sequence comprises a homolog of SEQ ID NO: 44.

[0273] 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 the anti-EGFR scFv. In some embodiments, the homolog comprises a nucleotide 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 nucleic acid sequence of the anti-EGFR scFv or any anti-ROR1 scFv, or anti-PSMA scFv, or anti-5T4 scFv.

[0274] In some embodiments, 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) are disclosed herein, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) with default parameters. In some embodiments, homologs of nucleotide sequences 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) are disclosed herein, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) with default parameters.

[0275] In some embodiments, homologs of anti-hROR1 scFv VL-VH (SEQ ID NO:156) or anti-hROR1 scFv VH-VL (SEQ ID NO:169) are disclosed herein, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) with default parameters. In some embodiments, homologs of nucleotide sequences encoding anti-hROR1 scFv VL-VH (SEQ ID NO:157) or anti-hROR1 scFv VH-VL (SEQ ID NO:167) are disclosed herein, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) with default parameters.

[0276] In some embodiments, homologs of anti-hPSMA scFv VL-VH (SEQ ID NO:168) or anti-hPSMA scFv VH-VL (SEQ ID NO:170) are disclosed herein, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) with default parameters. In some embodiments, homologs of nucleotide sequences encoding anti-hPSMA scFv VL-VH (SEQ ID NO:169) or anti-hPSMA scFv VH-VL (SEQ ID NO:171) are disclosed herein, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) with default parameters.

[0277] In some embodiments, homologs of anti-h5T4scFv VL-VH (SEQ ID NO:172) or anti-h5T4scFv VH-VL (SEQ ID NO:174) are disclosed herein, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) with default parameters. In some embodiments, homologs of the nucleotide sequences encoding anti-h5T4scFv VL-VH (SEQ ID NO:173) or anti-h5T4 scFv VH-VL (SEQ ID NO:174) are disclosed herein, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) with default parameters.

[0278] In some embodiments, homology also includes deletion, insertion, or substitution variants thereof and of its bioactive polypeptide fragments, including amino acid substitutions. In one embodiment, the variant comprises conservative substitutions, or deletions, insertions, or substitutions that do not significantly alter the three-dimensional structure of the polypeptide of interest, such as the VL or VH region of the first binding domain, specifically in the region of the CDR epitope-binding domain. 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 and second binding domains or both the first and second binding domains of the precursor construct, which in some embodiments binds 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-5T4scFv 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 ROR1, PSMA, or 5T4, respectively, on target tumor cells.

[0279] 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 recited 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 recited in SEQ ID NO:37 or a homolog thereof.

[0280] 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 recited in any of SEQ ID NO:34, 37, 156, 166, 168, 170, 172, or 174 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 comprise the sequence recited in SEQ ID NO:41 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 recited in SEQ ID NO:42 or a homolog thereof.

[0282] 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 NO:36, 38, 157, 167, 169, 171, 173, or 175 or a homolog thereof.

[0283] 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 SEQ ID NO:36 or a homolog thereof, and the sequence recited in SEQ ID NO:38 or a homolog thereof.

[0284] 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 SEQ ID NO:43 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 recited in SEQ ID NO:44 or a homolog thereof.

[0285] 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 both the first and second binding domains is optimized for mammalian transcription and translation.

[0286] In another embodiment, the TAAs provided herein are angiogenic antigens that are expressed on both activated pericytes and pericytes of the tumor angiogenic vasculature, the latter being associated with new blood vessel formation in vivo. Angiogenic antigens are known in the art, see, for example, WO2010 / 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) / scatter factor (SF); interleukin-8 (IL-8); leptin; midkine; placental growth factor; platelet-derived endothelial cell growth factor (PD-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).

[0287] As described above and throughout, in some embodiments, the first binding domain or the second binding domain, or the first and second binding domains (the TAA binding domains) comprise single-chain variable fragments (scFv).

[0288] 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 is described in more detail throughout, for example, as contained in polypeptide A and polypeptide B.

[0289] In some embodiments, the precursor trispecific antibody construct comprises a Fab fragment at its core, which in some embodiments comprises the third binding domain. Those skilled in the art will understand that a Fab fragment is an antigen-binding fragment of an antibody. A Fab consists of one constant and one 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 a disulfide bond between the heavy and light chain constant regions (see, for example, Figure 1 and 2A -the figures in 2B and 2F, and in Figure 8A 、 9A, the amino acid sequences shown in 10A, 11A, 45A - 45B, 46, 47A - 47B, and 48, where the cysteine residues that indicate that a disulfide bond (Cys - SS - Cys bond) can be formed between polypeptide A and B of the precursor construct (highlighted in bold and underlined) are shown. The codons encoding these Cys residues are shown in Figure 8B , 9B , in the nucleic acid sequences shown in 10B and 11B (highlighted in bold and underlined). Thus, one skilled in the art will understand that the term "Fab" with respect to an antibody generally includes the portion of the antibody consisting of a single light chain (both variable and constant regions) that is bound to the variable region and the first constant region of a single heavy chain via a disulfide bond.

[0290] One skilled in the art will recognize that the disulfide bond between the heavy and light chains is preferred but not essential for function (Orcutt et al. (2010), PEDS, 23:221 - 228). Thus, in certain embodiments, the Fab fragments disclosed herein may not contain a disulfide bond. In this regard, the heavy and light chains can be engineered in such a way that they stably interact without the need for a disulfide bond. For example, in certain embodiments, the heavy or light chain can be engineered to remove cysteine residues, and the heavy and light chains still stably interact and perform the function of Fab. In some embodiments, mutations are made to promote stable interaction between the heavy and light chains. For example, the "knobs into hole" engineering strategy can be used to promote dimerization between the heavy and light chains of Fab (see, e.g., 1996 Protein Engineering, 9:617 - 621). Using this strategy, a "knob" is created by replacing small amino acid side chains at the interface between interacting domains with larger side chains. A corresponding "hole" is created at the interface between the interacting molecules by replacing larger side chains with smaller side chains. Thus, the use of variant Fab fragments designed for specific purposes is also contemplated herein, such as 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.

[0291] In some embodiments, the conformation of the variable and constant regions within the Fab fragment may be different from that found in native 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 the other (Shaefer et al. (2011), PNAS, 108:111870 - 92). Such modified Fab fragments still function to bind their specific target antigen and are contemplated for use in the precursor constructs disclosed herein. Thus, in this regard, the variable and constant regions that make up Fab are considered modular.

[0292] 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 their subtypes, such as IgG1, IgG2, IgG3, and IgG4. The light chain domain can be derived from a κ or λ chain. The Fab fragments used herein can be prepared recombinantly.

[0293] As is well known in the art, an antibody is an immunoglobulin molecule that can specifically bind to a target, such as a carbohydrate, polynucleotide, lipid, polypeptide, etc., through at least one epitope recognition site located in the variable region of the immunoglobulin molecule. Those skilled in the art will understand that the term "antibody" includes not only intact polyclonal or monoclonal antibodies, but also humanized antibodies, chimeric antibodies, antibody fragments including antibody fragments lacking the Fc region, and any other modified configurations of immunoglobulin molecules that contain an antigen-binding site or the desired specificity (epitope recognition site), including scFv fragments and Fab fragments. In some embodiments, the precursor antibody constructs described herein lack the Fc region.

[0294] The Fab fragments disclosed herein comprise an antigen-binding portion (the third binding domain) consisting of the variable region of the immunoglobulin heavy chain and the variable region of the immunoglobulin light chain (VH and VL, respectively). Similarly, the aforementioned scFv fragments (the first or second binding domain) comprise an antigen-binding portion consisting of the variable region of the immunoglobulin heavy chain and the variable region of the immunoglobulin light chain (VH and VL, respectively). More specifically, as used herein, the term "antigen-binding portion" refers to a polypeptide fragment that contains at least one CDR of the immunoglobulin heavy chain and / or light chain and 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 constructs described herein can contain 1, 2, 3, 4, 5, or all 6 CDRs of the VH and VL sequences of the parental antibody that bind to the target antigen of interest. In certain embodiments, the antigen-binding portion of the scFv fragment (the first or second binding domain, or both the first and second binding domains) of the precursor trispecific antibody construct binds to a 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 to CD3.

[0295] In some 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 / VLs with desired properties, such as increased affinity for a 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.

[0296] Other methods can also be used to mix and match CDRs to identify Fabs with desired binding activities (e.g., binding 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, where CDR3 and FR4 are retained from the mouse VH. After obtaining the antibody, the VH is screened against a human VL library to obtain an antigen - binding antibody. 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 retaining the mouse CDR3. Antibodies capable of binding the antigen are obtained. Rader et al., PNAS (1998) 95:8910 - 8915 describe a process similar to that of Beiboer et al. above.

[0297] These techniques just described are known per se in the art. However, one skilled in the art will be able to use such techniques to obtain antigen - binding fragments of antibodies according to several embodiments of the disclosure herein using conventional methods in the art.

[0298] Also disclosed herein are methods for obtaining antibody antigen - binding domains specific for a target antigen (e.g., CD3 or any target antigen for the binding domains described elsewhere herein for the constructs described herein), the method comprising providing a VH domain that is an amino - acid sequence variant of the VH domain by adding, deleting, substituting, or inserting one or more amino acids in the amino - acid sequence of the VH domain listed herein, optionally combining the thus - provided VH domain with one or more VL domains, and testing the VH domain or one or more VH / VL combinations to identify a specific - binding member or antibody antigen - binding domain that is specific for a target antigen of interest (e.g., CD3) and optionally has one or more desired properties. The VL domain can have an amino - acid sequence substantially as listed herein. Similar methods can be employed in combining one or more sequence variants of the VL domains disclosed herein with one or more VH domains.

[0299] One of ordinary skill in the art will understand that "specifically binds" or "preferentially binds" (used interchangeably herein) to an epitope of an antibody or polypeptide is a term 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 as compared to an alternative cell or substance. An antibody or its Fab or scFv "specifically binds" or "preferentially binds" to a target if its binding to the target is of greater affinity, avidity, more rapid, and / or of longer duration than its binding to other substances. For example, an antibody that specifically binds or preferentially binds to a CD3 epitope is one that binds one CD3 epitope with greater affinity, avidity, more rapid, and / or of longer duration than it binds other CD3 epitopes or non-CD3 epitopes. 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 bind or preferentially bind to a second target. Thus, "specific binding" or "preferential binding" does not necessarily require (although it can include) exclusive binding. Usually (but not necessarily), reference to binding means preferential binding.

[0300] In certain embodiments, the antigen-binding portion of a Fab fragment (third binding domain) as described herein includes a heavy-chain and a light-chain CDR set, which are inserted between sets of heavy-chain and light-chain framework regions (FRs), respectively, that provide support for the CDRs and define the spatial relationship of the CDRs relative to each other. As used herein, the term "CDR set" refers to the three hypervariable regions of the heavy or light chain V region. Starting from the N-terminus of the heavy or light chain, these regions are designated "CDR1", "CDR2", and "CDR3", respectively. Thus, the antigen-binding site includes six CDRs, comprising a CDR set from each of the heavy-chain and light-chain V regions. A polypeptide containing 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 has shown that the amino acid residues of the CDRs make extensive contacts with the bound antigen, with the most extensive antigen contacts being with the heavy-chain CDR3. Thus, the molecular recognition unit is primarily responsible for the specificity of the antigen-binding site.

[0301] 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 V region of a heavy or light chain. Some FR residues may contact the bound antigen; however, the FRs are primarily responsible for folding the V region into the antigen-binding site, particularly the FR residues immediately adjacent to the CDRs. Within the FRs, certain amino acid residues and certain structural features are highly conserved. In this regard, all V region sequences contain an internal disulfide loop of approximately 90 amino acid residues. When the V region folds into the binding site, the CDRs are shown to form protruding loop motifs that make up the antigen-binding surface. It is recognized that there are conserved structural regions of the FRs that influence the folding shape of the CDR loops into certain "canonical" structures - independent of the exact CDR amino acid sequence. In addition, certain FR residues are known to participate in non-covalent interdomain contacts that stabilize the heavy and light chain interactions of antibodies.

[0302] The structure and location of immunoglobulin variable regions can be determined by reference to Kabat, E.A. et al., Sequences of Proteins of Immunological Interest. 4th Edition. US Department of Health and Human Services. 1987 and its updates (now available on the Internet at immune.bme.nwu.edu).

[0303] One of ordinary skill in the art will recognize that the term "monoclonal antibody" encompasses a homogeneous population of antibodies, where the monoclonal antibodies are composed of amino acids (naturally occurring and non-naturally occurring) that participate in the selective binding of an epitope. Monoclonal antibodies are highly specific and directly target a single epitope. The term "monoclonal antibody" includes not only intact monoclonal antibodies and full-length monoclonal antibodies, but also their fragments (e.g., Fab, Fab', F(ab')2, Fv), single chains (ScFv), their variants, fusion proteins containing an antigen-binding portion, humanized monoclonal antibodies, chimeric monoclonal antibodies, and any other modified configuration of an immunoglobulin molecule (the immunoglobulin molecule containing an antigen-binding fragment (epitope recognition site) with the desired specificity and ability to bind the epitope). This is not intended to limit the source of the antibody or the manner in which it is prepared (e.g., by hybridoma, phage selection, recombinant expression, transgenic animals, etc.). The term includes intact immunoglobulins as well as fragments as described herein.

[0304] The proteolytic enzyme papain preferentially cleaves the IgG molecule to produce several fragments, two of which (F(ab) fragments) each contain a covalently linked heterodimer containing a complete antigen-binding site. Pepsin is capable of cleaving the IgG molecule to provide several fragments, including the F(ab')2 fragment containing two antigen-binding sites. The Fv fragments used according to certain embodiments disclosed herein can be generated 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-covalently linked VH::VL heterodimers that include an antigen-binding site that retains most of the antigen recognition and binding capacity of the 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.

[0305] In some embodiments of the present disclosure, a Fab fragment comprising a third binding domain binds to CD3. In some embodiments of the present disclosure, a Fab fragment comprising a third binding domain binds to CD3ε.

[0306] The "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 disulfide-linked heterodimer of highly variable (α) and (β) chains. In other T cells, an alternative receptor consisting of variable γ and (δ) chains is expressed. Each chain of the 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 located 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 from a variety of animal species, including humans, mice, rats, or other mammals.

[0307] "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), TCRγ, or TCR(δ) chain). In certain embodiments, the anti-TCR Fab binds to TCR(α), TCR(β), or both. One of ordinary skill in the art 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 where the binding properties of the third binding domain are mentioned.

[0308] "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 the CD3(γ) chain, the CD3(δ) chain, two CD3(epsilon; ε) chains, and a homodimer of 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, which is a characteristic that allows these chains to associate with the 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 the ITAM is important for the signaling capacity 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.

[0309] As used herein, "anti-CD3 Fab" refers to a Fab that comprises a third binding domain that specifically binds to an individual CD3 chain (e.g., CD3(γ) chain, CD3(δ) chain, or CD3(epsilon; ε) chain) or a complex formed by two or more individual 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 CD3(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.

[0310] In some embodiments, the anti-CD3 Fab binds to an epitope contained within amino acids 1-27 of CD3ε. In some embodiments, the anti-CD3 Fab binds to amino acids 1-27 of CD3ε. In some embodiments, the anti-CD3 Fab binds to amino acids 1-27 of human CD3ε. Amino acids 1-27 of CD3ε are recited in SEQ ID NO:5.

[0311] One of ordinary skill in the art will understand that, in some embodiments, the term "anti-CD3 Fab" may include the third binding domain of the precursor trispecific antibody construct described herein. In some embodiments, the term "anti-CD3 Fab" may include a precursor construct, where the binding properties of the third binding domain are mentioned.

[0312] 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, where the term includes the third binding domain of the precursor construct. In some embodiments, the term "Fab" may be used interchangeably with the phrase "third binding domain", both having the same nature and meaning.

[0313] In some embodiments, the precursor trispecific antibody construct comprises a third binding domain that binds to an extracellular epitope of CD3ε. In some embodiments, the precursor trispecific antibody construct comprises a third binding domain that binds to the N-terminus of CD3ε. In some embodiments, the precursor trispecific antibody construct comprises a third binding domain that binds to an epitope having amino acids 1-27 of CD3ε. In some embodiments, the anti-CD3 Fab binds to amino acids 1-27 of CD3ε. In some embodiments, the anti-CD3 Fab binds to amino acids 1-27 of human CD3ε. Amino acids 1-27 of CD3ε are recited in SEQ ID NO:5.

[0314] As used herein, a "TCR complex" refers to a complex formed by the association of a TCR with CD3. For example, a TCR complex can be composed of a CD3(γ) chain, a CD3(δ) chain, two CD3(ε) chains, a homodimer of CD3(ζ) chains, a TCR(α) chain, and a TCR(β) chain. Optionally, a TCR complex can be composed of a CD3(γ) chain, a CD3(δ) chain, two CD3(ε) chains, a homodimer of CD3(ζ) chains, a TCRγ chain, and a TCR(δ) chain.

[0315] As used herein, "components of the TCR complex" refers to TCR chains (i.e., TCR(α), TCR(β), TCRγ or TCR(δ)), CD3 chains (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).

[0316] As background, the TCR complex is generally responsible for initiating the T cell response to an antigen bound to an MHC molecule. It is believed that the binding of the peptide:MHC ligand to the TCR and co-receptor (i.e., CD4 or CD8) brings together the TCR complex, co-receptor and CD45 tyrosine phosphatase. This allows CD45 to remove inhibitory phosphate groups, thereby activating the Lck and Fyn protein kinases. Activation of these protein kinases leads to phosphorylation of ITAM on the CD3(ζ) chain, which in turn enables these chains to bind the cytoplasmic tyrosine kinase ZAP-70. Subsequently, activation of the bound ZAP-70 by phosphorylation triggers three signaling pathways, two of which are initiated by phosphorylation and activation of PLC-(γ), which then cleaves phosphatidylinositol phosphate (PIP) into diacylglycerol (DAG) and inositol triphosphate (IP3). Activation of protein kinase C by DAG leads to activation of the transcription factor NFκB. Due to the action of IP3, a sudden increase in intracellular free Ca 2+ 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 a member of the AP-1 transcription factor family; a dimer of members of the Fos and Jun transcriptional regulatory factor families.

[0317] The third signaling pathway initiated by activated ZAP-70 is activation of Ras and subsequent activation of the MAP kinase cascade. This ultimately leads to activation of Fos and thus activation of the AP-1 transcription factor. NFκB, NFAT and AP-1 act together on the T cell chromosome to initiate new gene transcription leading to T cell differentiation, proliferation and effector function. See Pitcher et al., 2003., TRENDS in Immunol. 24, 554-560; Smith-Garvin et al., Annu Rev Immunol. 2009; 27:591-619.

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

[0319] In certain embodiments, the third binding domain specifically binds to an individual human CD3 chain (e.g., human CD3(γ) chain, human CD3(δ) chain, or human CD3(ε) chain) or a combination of two or more individual 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 the human CD3(ε) chain. In certain embodiments, the third binding domain specifically binds to the extracellular epitope of the human CD3(ε) chain. In certain embodiments, the third binding domain specifically binds to the epitope within SEQ ID NO:3.

[0320] In certain other embodiments, the Fab of the present disclosure comprising the third binding domain specifically binds to TCR(α), TCR(β), or the heterodimer formed by TCR(α) and TCR(β). In certain embodiments, the Fab specifically binds to one or more of human TCR(α), human TCR(β), or the heterodimer formed by human TCR(α) and human TCR(β).

[0321] In certain embodiments, the Fab of the present disclosure that includes 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 CD3(γ) chain, CD3(δ) chain, CD3(ε) chain, TCR(α) chain, or 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 human CD3(γ) chain, human CD3(δ) chain, human CD3(ε), human TCR(α) chain, or human TCR(β) chain, or any combination thereof. In certain embodiments, the Fab of the present disclosure binds 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.

[0322] The Fab of the present disclosure can be produced as described herein or by a variety of methods known in the art (see, e.g., U.S. Patent Nos. 6,291,161; 6,291,158). Sources of Fabs include monoclonal antibody nucleic acid sequences from different species (which can be converted to antibodies, Fvs, scFvs, or Fabs, e.g., in phage libraries), said 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.

[0323] Anti-human CD3 antibodies that cross-react with simian CD3 are particularly desirable, such as the SP34 murine monoclonal antibody, which specifically binds to denatured forms of human CD3 (western blot or dot blot) and native form (on T cells) of human CD3 (Pressano, S. The EMBO J. 4:337-344, 1985; Alarcon, B. EMBO J. 10:903-912, 1991). The SP34 murine monoclonal antibody also binds to COS cells transfected with CD3c alone and 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 crosslinking, SP34 activates T cells (Yang et al., J. Immunol. 137:1097-1100, 1986). Cross-reactivity with simian CD3 is important because it allows the direct use of clinical candidates in non-human primates for toxicity studies, rather than in chimpanzees or using surrogate molecules. Thus, toxicity studies using such cross-reactive anti-CD3 Fabs in the precursor trispecific antibody constructs of the present disclosure can provide a more relevant safety assessment.

[0324] Other exemplary anti-CD3 antibodies include the Cris-7 monoclonal antibody (Reinherz, E.L. et al. (eds.), Leukocyte typing II., Springer Verlag, New York, (1986)), the 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 the 145-2C11 monoclonal antibody (Hirsch et al. (1988) J. Immunol. 140:3766). Further CD3-binding molecules contemplated for use herein include UCHT-1 (Beverley, P C and Callard, R.E. (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 by reference.

[0325] 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, 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 incorporated herein by reference.

[0326] An exemplary anti-TCR antibody is the H57 monoclonal antibody (Lavasani et al. (2007) Scandinavian Journal of Immunology 65:39-47).

[0327] Antigen-binding fragment sequences of Fab fragments (e.g., heavy and light chain variable region sequences) are available in public databases or can be used in conventional systems (e.g., mice, HuMAb mice.RTM., TC mice.TM., KM-mice.RTM., llamas, chickens, rats, hamsters, rabbits, etc.) using components of traditional hybridoma development strategies with CD3 chains, TCR components, or other Fab-binding targets as immunogens for the development of Fabs for use herein. Those skilled in the art will understand that Fab fragments can be generated using a variety of techniques known in the art, including antibody display techniques such as phage, yeast, ribosome, and mRNA display techniques; B cell culture techniques such as SLAM technique; or high-throughput gene sequencing techniques for B cells or plasma B cells isolated from immunized animal subjects or immunized human subjects.

[0328] In some embodiments, the third binding domain (Fab) disclosed herein comprises the native sequence of a humanized FR amino acid sequence and the CDR amino acid sequence of a murine monoclonal antibody. Examples of those anti-CD3ε amino acid sequences in which the FR sequences have been humanized while the CDR amino acid sequences remain those of the SP34 murine monoclonal antibody are disclosed in International Application Publication No. WO2007 / 042261, the entire content of which is incorporated herein by reference.

[0329] 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 respectively below. 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 within the precursor trispecific antibody constructs of the present disclosure include the VH, CH1, VL, and CL amino acid sequences 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.

[0330] Table 2: Amino acid sequences of anti-CD3 VH, VL, HC, LC, and CDRs and their combinations.

[0331]

[0332]

[0333]

[0334] 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 that comprises a variable heavy chain region (VH) containing CDR-H1, CDR-H2, and CDR-H3 and a variable light chain region (VL) containing 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.

[0335] In some embodiments, the amino acid sequence of the anti-human CD3ε CDR-H1 is set forth in SEQ ID NO:104. In some embodiments, the amino acid sequence of the anti-human CD3ε CDR-H2 is set forth in SEQ ID NO:105. In some embodiments, the amino acid sequence of the anti-human CD3ε CDR-H3 is set forth in SEQ ID NO:106. In some embodiments, the amino acid sequence of the anti-human CD3ε CDR-L1 is set forth in any of SEQ ID NO:107-109. In some embodiments, the amino acid sequence of the anti-human CD3 3ε CDR-L1 is set forth in SEQ ID NO:110. In some embodiments, the amino acid sequence of the anti-human CD33ε CDR-L1 is set forth in SEQ ID NO:111-112.

[0336] In some embodiments, the third binding domain comprises a Fab fragment that comprises a variable heavy chain region (VH) and a variable light chain region (VL) that bind to the extracellular domain of the human CD3ε polypeptide. In some embodiments, the amino acid sequences of the VH and VL of anti-human CD3ε are selected from the amino acid sequences set forth in any of SEQ ID NO:46-72 and 114 (VH) and 75-103 and 116 (VL). In some embodiments, the amino acid sequences of the VH and VL of anti-human CD3ε comprise sequence homologs of the amino acid sequences set forth in any of SEQ ID NO:46-72 and 114 (VH) and 75-103 and 116 (VL).

[0337] In some embodiments, the amino acid sequence of the VH of the third binding domain (VH1) of human CD3ε is selected from the amino acid sequences recited in any one of SEQ ID NOs: 75 - 103 and 116 or homologs thereof. In some embodiments, the amino acid sequence of the VL of the third binding domain (VL1) of human CD3ε is selected from the amino acid sequences recited in any one of SEQ ID NOs: 75 - 103 and 116 or homologs thereof.

[0338] 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 - CD3ε. 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 - CD3ε. In some embodiments, the homologs include polypeptides 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 the VH of anti - human CD3ε or the VL of anti - human CD3ε, as determined using the BlastP software of the National Center for Biotechnology Information (NCBI) with default parameters.

[0339] In some embodiments, homology also includes deletion, insertion or substitution variants of it and its bioactive polypeptide fragments, including amino acid substitutions. In one embodiment, the variant comprises conservative substitutions, or deletions, insertions or substitutions that do 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, and in some embodiments, it binds to the CD3ε sequence on the target T cell.

[0340] In some embodiments, the third binding domain that binds to the CD3ε cell surface epitope VH1 comprises the sequences recited in SEQ ID NOs: 46-72 and 114 or homologs thereof. In some embodiments, the second binding domain that binds to the CD3ε cell surface epitope VL1 comprises the sequences recited in SEQ ID NOs: 75-103 and 116 or homologs thereof. In some embodiments, the third binding domain that binds to the CD3ε cell surface epitope comprises a sequence selected from the sequences recited in SEQ ID NOs: 46-72, 74 and 114 or homologs thereof, and a sequence selected from the sequences recited in SEQ ID NOs: 75-103, 113, 115 and 116 or homologs thereof. In some embodiments, the third binding domain that binds to the CD3ε cell surface epitope comprises the sequence recited in SEQ ID NO: 113 or homologs thereof, and the sequence recited in SEQ ID NO: 74 or homologs thereof.

[0341] In some embodiments, the VL region of the third binding domain comprises the amino acid sequences recited 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 VH region of the third binding domain comprises CDR-H1 (SEQ ID NO: 104), CDR-H2 (SEQ ID NO: 105), and CDR-H3 (SEQ ID NO: 106).

[0342] In some embodiments of the precursor trispecific antibody construct, the VL region of the third binding domain comprises the amino acid sequence recited in any of SEQ ID NOs: 75-103 and 116, or an amino acid sequence having at least 80% homology thereto. In some embodiments, the VL region of the third binding domain comprising an amino acid sequence having at least 80% homology thereto comprises a framework sequence having at least 80% homology, wherein the CDR regions are "as is" in the selected amino acid sequences (SEQ ID NOs: 107-112).

[0343] In some embodiments of the precursor trispecific antibody construct, the VH region of the third binding domain comprises the amino acid sequence recited in any of SEQ ID NOs: 46-72 and 114, or an amino acid sequence having at least 80% homology thereto. In some embodiments, the VH region of the third binding domain comprising an amino acid sequence having at least 80% homology thereto comprises a framework sequence having at least 80% homology, wherein the CDR regions are "as is" in the selected amino acid sequences (SEQ ID NOs: 104-106).

[0344] 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, or second or third binding domain, or any combination thereof comprises a humanized binding domain.

[0345] 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 first, second, and third constant regions, and each light chain consisting 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 containing α, δ, Ε, γ, and μ heavy chains are classified as immunoglobulin (Ig) A, IgD, IgE, IgG, and IgM. A complete antibody forms a "Y" shape. The stem of the Y consists of the second and third constant regions (for IgE and IgM, the fourth constant region) of the two heavy chains bound together and forms a disulfide bond (interchain) in the hinge. Heavy chains γ, α, and δ have constant regions consisting of three tandem (in a straight line) Ig domains and a hinge region for increased flexibility; heavy chains μ and ε have constant regions consisting of four immunoglobulin domains. The second and third constant regions are referred to as the "CH2 domain" and the "CH3 domain", respectively. Each arm of the Y includes the variable region and the first constant region of a single heavy chain bound to the variable region and the constant region of a single light chain. The variable regions of the light and heavy chains are responsible for antigen binding.

[0346] The "complementary determining region" or "CDR" of an antibody refers to the highly variable loops in the variable region of the antibody heavy or light chain. The CDRs can interact with the antigen conformation and largely determine binding to the antigen (although some framework regions are known to be involved in binding). The variable region of the heavy chain and the variable region of the light chain each contain 3 CDRs. The CDRs can be defined or identified by conventional methods, such as according to the sequences of Kabat et al. (Wu, T T and Kabat, E.A., J Exp Med. 132(2):211-50, (1970); Borden, P. and Kabat E.A., PNAS, 84:2440-2443 (1987); Kabat, E.A. et al, Sequences of proteins of immunological interest, published by DIANE Publishing, 19922), or according to the structures of Chothia et al. (Choithia, C. and Lesk, A.M., J. Mol. Biol., 196(4):901-917 (1987), Choithia, C. et al, Nature, 342:877-883 (1989)).

[0347] The "variable heavy chain" or "VH" of an antibody refers to a fragment of the heavy chain that contains three CDRs inserted between flanking segments called framework regions, which are more highly conserved than the CDRs and form a scaffold to support the CDRs.

[0348] The "variable light chain" or "VL" of an antibody refers to a fragment of the light chain that contains three CDRs inserted between framework regions.

[0349] The "Fv" of an antibody refers to the smallest fragment of an antibody that bears a complete antigen-binding site. The Fv fragment consists of the variable region of a single light chain associated with the variable region of a single heavy chain.

[0350] The "single-chain Fv antibody" or "scFv" of an antibody refers to a engineered antibody consisting of a variable light chain and a variable heavy chain that are directly linked to each other or linked through a peptide linker sequence.

[0351] 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)). A "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. Patent No. 6,005,079).

[0352] A "single-domain antibody" or "dAb" refers to an antibody fragment consisting of the variable region of an antibody heavy chain (VH domain) or the variable region of an antibody light chain (VL domain) (Holt, L., et al, Trends in Biotechnology, 21(11):484-490).

[0353] As used herein, the term "disulfide bond" refers to the binding of a heavy-chain fragment and a light-chain fragment through 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 the light-chain fragment, respectively.

[0354] "Variable region linker sequence" is an amino acid sequence that links the heavy chain variable region to the light chain variable region and provides a linker function compatible with the interaction of 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 link a binding domain to an immunoglobulin CH2 or CH3 region polypeptide can be used as the variable region linker sequence.

[0355] 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-terminus of the VL or 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-terminus of the VL region of the third binding domain, and the second sub-regulatory domain is located at the N-terminus 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-terminus of the VH region of the third binding domain, and the second sub-regulatory domain is located at the N-terminus of the VL region of the third binding domain.

[0356] 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 the 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. Those skilled in the art will understand 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 VH of the third binding domain, the first sub-regulatory domain is located at the N-terminus of VL of the third binding domain. Similarly, those skilled in the art will understand 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.

[0357] Alternative sources of binding domains can include sequences encoding random peptide libraries or sequences encoding engineered amino acid diversity in loop regions of alternative non-antibody scaffolds such as fibronectin 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., WO2006 / 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.TM. (see, e.g., PCT Patent Application Publication No. WO 2007 / 098934; WO 2006 / 072620), etc.

[0358] As Figure 1 and 2A- Depicted in -2F, the scFv is an illustrative binding domain. In some embodiments, the first or second binding domain comprising the scFv fragment, or both the first and second binding domains, 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;Erb2), 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 TAAs include 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 TAAs are selected from PSMA, ROR1, 5T4 and EGFR. These and other tumor proteins or tumor-associated proteins are known to those skilled in the art.;

[0359] In certain embodiments, the first or second binding domain, or both the first and second binding domains, specifically bind to an antigen target associated with a disease condition. Disease conditions can include physiological disorders, pathological disorders, and cosmetic disorders. Examples of illustrative disorders include, but are not limited to, cancer, inflammatory disorders, allografts, type I diabetes, type II diabetes, and multiple sclerosis.

[0360] In some embodiments, the specific structural components of the precursor trispecific antibody construct comprise a first and a second binding domain (e.g., but not limited to, scFv fragments), a third binding domain (e.g., 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.

[0361] 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 the HC polypeptide and polypeptide B comprises the LC polypeptide. In other embodiments, polypeptide A comprises the LC polypeptide and polypeptide B comprises the HC polypeptide.

[0362] 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 the 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;

[0363] wherein the first and second binding domains are positioned as described above, and

[0364] 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 or VL region of the third binding domain;

[0365] 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.

[0366] 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".

[0367] In some embodiments, the precursor trispecific antibody construct described herein comprises two polypeptides - polypeptide A and polypeptide B, wherein polypeptide A comprises components in an N-terminal to C-terminal sequence: HLP domain, protease cleavage domain, CD3 third binding domain VH-CH1 region, first binding domain (VL-VH); and polypeptide B comprises components in a sequence from the N-terminus to the 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 the order of N-terminus to C-terminus):

[0368] Polypeptide A: HLP-L-CP-L-CD3VH-L-CD3CH1-L-TAA VL-L-TAAVH

[0369] Polypeptide B: CAP-L-CP-L-CD3VL-L-CD3CL-L-TAAVL-L-TAA VH,

[0370] wherein "L" is a linker, which may or may not be present in each embodiment.

[0371] In some embodiments, the precursor trispecific antibody construct described herein comprises two polypeptides - polypeptide A and polypeptide B, wherein polypeptide A comprises components in an N-terminal to C-terminal sequence: HLP domain, protease cleavage domain, CD3 third binding domain VH-CH1 region, first binding domain (VH-VL); and polypeptide B comprises components in a sequence from the N-terminus to the 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 the order of N-terminus to C-terminus)

[0372] Polypeptide A: HLP-L-CP-L-CD3VH-L-CD3CH1-L-TAA VH-L-TAAVL

[0373] Polypeptide B: CAP-L-CP-L-CD3VL-L-CD3CL-L-TAAVL-L-TAA VH,

[0374] wherein "L" is a linker, which may or may not be present in each embodiment Figure 1 )

[0375] In some embodiments, the precursor trispecific antibody construct described herein comprises two polypeptides - Polypeptide A and Polypeptide B, wherein Polypeptide A comprises components in 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 in 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 the order of N-terminal to C-terminal)

[0376] Polypeptide A: HLP-L-CP-L-CD3VH-L-CD3CH1-L-TAAVH-L-TAAVL

[0377] Polypeptide B: AP-L-CP-L-CD3VL-L-CD3CL-L-TAAVH-L-TAAVL,

[0378] wherein "L" is a linker, which may or may not be present in each embodiment

[0379] In some embodiments, the precursor trispecific antibody construct described herein comprises two polypeptides - Polypeptide A and Polypeptide B, wherein Polypeptide A comprises components in 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 in 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 the order of N-terminal to C-terminal)

[0380] Polypeptide A: HLP-L-CP-L-CD3VH-L-CD3CH1-L-TAAVL-L-TAAVH

[0381] Polypeptide B: CAP-L-CP-L-CD3VL-L-CD3CL-L-TAAVH-L-TAA VL,

[0382] wherein "L" is a linker, which may or may not be present in each embodiment.

[0383] The above four constructs provide non-limiting examples of trispecific precursor antibody constructs, wherein the positions of VL and VH within the scFv of the first or second binding region are alternating. Those skilled in the art will understand that these alternative positions will also be included in the following embodiments, where the scFv is represented as VL-VH, in order to illustrate other alternative non-limiting combinations of components.

[0384] In some embodiments, the precursor trispecific antibody constructs described herein comprise two polypeptides - Polypeptide A and Polypeptide B, wherein Polypeptide A comprises components in 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 in 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 constructs described herein comprise the following two polypeptides (in the order of N-terminal to C-terminal)

[0385] Polypeptide A: CAP-L-CP-L-CD3VH-L-CD3CH1-L-TAA VL-L-TAA VH

[0386] Polypeptide B: HLP-L-CP-L-CD3VL-L-CD3CL-L-TAAVL-L-TAAVH,

[0387] wherein "L" is a linker, which may or may not be present in each embodiment.

[0388] The above constructs provide non-limiting examples of trispecific precursor antibody constructs, wherein the positions of the two regulatory domains are alternating. Those skilled in the art will understand that these alternative positions can also be included in other embodiments, where the regulatory domain positions are represented such that the HLP domain is a component of the HC and CAP is a component of the LC, and the scFv is represented as VL-VH, in order to illustrate other alternative non-limiting combinations of components.

[0389] In some embodiments, the precursor trispecific antibody construct described herein comprises two polypeptides - polypeptide A and polypeptide B, wherein polypeptide A comprises components in an N-terminal to C-terminal sequence: HLP domain, protease cleavage domain, CD3 third binding domain VL-CL region, first binding domain (VL-VH); and polypeptide B comprises components in an N-terminal to C-terminal sequence: 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 of N-terminal to C-terminal)

[0390] Polypeptide A: HLP-L-CP-L-CD3VL-L-CD3CL-L-TAA VH-L-TAA VL

[0391] Polypeptide B: CAP-L-CP-L-CD3VH-L-CD3CH1-L-TAA VH-L-TAAVL,

[0392] wherein "L" is a linker, which may or may not be present in each embodiment.

[0393] Those skilled in the art will understand that the names "polypeptide A" and "polypeptide B" are merely names denoting two heterologous polypeptide chains, and thus the names themselves can be interchanged or changed, for example, polypeptide 1 and polypeptide 2. Further, the term includes two structurally different polypeptide chains that together form the precursor trispecific antibody construct as described herein. Further, those skilled in the art will understand the modular nature of the precursor construct described herein, where modules can replace one another, where they provide similar or different activities. For example, but not limited to, ScFv may have to be sorted in an N-terminal to C-terminal VL-VH or VH-VL order; or the regulatory domain can include a CAP component or an HLP domain.

[0394] The order of the components can be alternated as shown in the above non-limiting examples, where in some embodiments, the first or second sub-regulatory domain is a component of either HC or LC and can be located at the N-terminal of the third binding domain, and the first or second binding domain comprises an scFv, where the components (VL, L, and VH) of the scFv are independently sorted as VL-L-VH or VH-L-VL, which are located at the C-terminal of the third binding domain and are components of either HC or LC, respectively.

[0395] Those skilled in the art will understand that depending on the desired functionality of the precursor construct, the precursor construct can include different regulatory domains. For example, the trispecific precursor construct can contain only a single regulatory domain, either the HLP domain or the CAP domain (see respectivelyFigure 2C and 2D )。

[0396] 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 are directly linked to respective ends of VH-CH1 or VL-CL of a third binding domain (e.g., Fab) (i.e., without addition of additional amino acids therebetween). In other embodiments, the linkage to the third binding domain (e.g., Fab) involves 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 the first or second sub-regulatory domain or both, depending on the surrounding space of the third binding domain target and the first and second binding domain targets on the cell surface (i.e., e.g., accessibility of the CD3ε target on the T cell cell surface).

[0397] In other 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 N-terminus of the heavy chain and / or light chain of the third binding domain. In still further 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 N-terminus of the first or second binding domain or both, and / or the C-terminus of the first or second sub-regulatory 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 linkage between the first or second binding domain and / or the first or second sub-regulatory domain and the VH-CH1 and VL-CL chains of the third binding domain may be the same or different.

[0398] The linkage 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 chains can utilize a combination of deletions and linkers as needed. As will be understood by those skilled in the art, the linkages between the third binding domain VH-CH1 and VL-CL chains and the first or second binding domain or both and / or the first or second sub-regulatory domain or both can be correspondingly adjusted and tested for the desired functionality (e.g., binding affinity, T cell activity) using methods known in the art and described herein.

[0399] As described herein, the linkages between domains or between components within a domain include linkers. In some embodiments, there are linkers between domains. In some embodiments, there are no linkers between domains. In some embodiments, there are linkers between the components that make up a domain. In some embodiments, there are no linkers between the components that make up a domain.

[0400] Figure 1 and 2A - Figures 2D and 2F illustrate the positions where linkers can be present in embodiments of the precursor trispecific antibody constructs disclosed herein.

[0401] 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.

[0402] 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.

[0403] In some embodiments, the length of the linker between components within the first or second binding domain or both is 1 to 10 amino acids. In other embodiments, the length of the linker between components within the first or second binding domain or both is 1 to 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.

[0404] In some embodiments, the length of the linker between components within the first or second sub-regulatory domain or both is 1 to 10 amino acids, it being understood that the linkers between different components need not have the same length. In other embodiments, the length of the linker between components within the first or second sub-regulatory domain or both is 1 to 20 or 20 amino acids, it being understood that the linkers between different components need not have the same length. In this regard, the length of the linker 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 further embodiments, the length of the linker between each set of components can be 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30 amino acids.

[0405] In some embodiments, linkers are present between the components within the polypeptide components VH and CH1 of the HC or LC Fab and the components within the polypeptide components VL and CL, respectively. In some embodiments, the length of the linker is 1 to 10 amino acids, it being understood that the linkers between different components need not have the same length. In other embodiments, the length of the linker between components within the third binding domain is 1 to 20 or 20 amino acids, it being understood that the linkers between different components need not have the same length. In this regard, the length of the linker 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 further embodiments, the length of the linker between each set of components can be 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30 amino acids.

[0406] In certain embodiments, the linker suitable for the precursor constructs described herein is a flexible linker. A suitable linker can be readily selected and can have any suitable different lengths, such as from 1 amino acid (e.g., 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, and can be 1, 2, 3, 4, 5, 6, or 7 amino acids.

[0407] In some embodiments, the flexible linker includes 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 may thus be able to act as neutral tethers between components. Glycine is significantly closer to the phi-psi space than alanine and is much less restricted than residues with longer side chains (see Scheraga, Rev. Computational Chem. 11173-142 (1992)). In some embodiments, the flexible linker includes, but is 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 in the art will recognize that the design of the precursor trispecific antibody construct can include all or part of a flexible linker such that the linker can include a flexible linker and one or more portions that confer a less flexible structure to provide the desired precursor trispecific antibody construct structure.

[0408] 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. Linkers known in the art are described at least 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 (https: / / www.ccdc.cam.ac.uk / solutions / partnersoftware / csdlinkerdatabase / ) provided by The Cambridge Crystallographic Data Centre.

[0409] 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, specifically MMP). In certain embodiments, the linker is a peptide linker.

[0410] In some embodiments, the linker between the third binding domain VH-CH1 or VL-CL chain and the first or second sub-regulatory 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 sub-regulatory domain or both comprises a protease substrate cleavage sequence, such as MMP2 / 9, uPA, proteinase, and asparaginyl endopeptidase substrate cleavage sequences. 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, proteinase, and asparaginyl endopeptidase.

[0411] 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 MMP. SEQ ID NO:9 is a commonly used MMP substrate cleavage sequence (see, for example, 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 still 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.

[0412] In some embodiments, all protease sites comprise the same proteolytic sequence. In other embodiments, the protease sites of the precursor construct are different. The difference in the proteolytic sequences within the precursor construct can provide additional regulation of the function of the precursor or partially activated construct.

[0413] A stable linker or protease non-cleavable linker refers to a linker peptide sequence that does not belong to a known protease substrate sequence and thus does not result in significant cleavage product formation upon incubation with protease.

[0414] 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 (usually an extracellular protease). In other embodiments, the cleavage sequence comprises a cysteine-cysteine pair capable of forming 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.

[0415] The cleavage substrate is located in the linker such that when the cleavage substrate is cleaved by a cleaving agent (e.g., the cleavage substrate of the linker is cleaved by a protease and / or the cysteine-cysteine disulfide bond is disrupted via reduction by exposure to a reducing agent) or by photoinduced photolysis, in the presence of a target, a cleavage product is produced having a variety of functional properties as described herein.

[0416] The cleavage substrate of the linker can be selected based on a protease that is co-localized in diseased tissue or expressed on the cell surface of a binding domain of the fusion moiety of interest for a target antigen. A variety of different diseases are known in which the target of interest is co-localized with a protease, and the substrates of the protease are known in the art. In the example of cancer, the target tissue can be cancerous tissue, specifically cancerous tissue of a solid tumor. An increase in the level of proteases having known substrates has been reported in many cancers (e.g., solid tumors) in the literature (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 injury, ischemia, etc. In addition, anti-angiogenic targets, such as VEGF, are known. Thus, when selecting the binding domain of the fusion moiety of the precursor trispecific antibody construct of the present disclosure to enable binding to a TAA, a suitable cleavage substrate sequence for the protease-cleavable linker will be a cleavage substrate sequence that contains a peptide substrate that can be cleaved by a protease present at an elevated level in the cancerous treatment site, particularly compared to non-cancerous tissue.

[0417] In some embodiments, the first or second binding domain, or both, of the precursor trispecific antibody construct can bind, for example, Her2, and the cleavage substrate sequence can be a matrix metalloproteinase (MMP) substrate and thus cleavable by MMP. In other embodiments, the first or second binding domain, or both, of the fusion portion in the precursor trispecific antibody construct can bind a target of interest or two targets of interest, and the cleavage substrate present in the linker can be, for example, asparaginyl 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, or both, of the fusion portion in the precursor trispecific antibody construct can bind 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, asparaginyl endopeptidase, uPA, and a proteolytic enzyme. In some embodiments, the first or second binding domain, or both, of the fusion portion in the precursor trispecific antibody construct can bind a target of interest or two targets of interest, and the cleavage substrate present in the linker comprises a combination of MMP2 / 9, asparaginyl endopeptidase, uPA, and a proteolytic enzyme, as shown in SEQ ID NO:35. In other embodiments, in diseases other than cancer, such as multiple sclerosis or rheumatoid arthritis, the cleavage substrate is cleaved by other disease-specific proteases.

[0418] The unmodified or uncleaved linker can permit tethering of the binding domain (first, second, or third, or a combination thereof) and the regulatory domain (first or second, or both).

[0419] The linker 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 sub-regulatory domain, or both) can comprise the same cleavage substrate or can comprise different cleavage substrates, e.g., the fir...

Claims

1. A precursor trispecific antibody construct, comprising: (a) a first binding domain that binds to a first tumor-associated antigen (TAA); (b) a second binding domain that binds to a second TAA; (c) a third binding domain that binds to an extracellular epitope of human CD3ε; and (d) a regulatory domain, the regulatory domain comprising (i) a first sub-regulatory domain and a second sub-regulatory domain, the first sub-regulatory domain comprising a first protease cleavage domain and a half-life prolongation (HLP) domain, and the second sub-regulatory domain comprising a second protease cleavage domain and a CAP component that reduces the ability of the third binding domain to bind to the extracellular epitope of human CD3ε; or (ii) a single regulatory domain that comprises a protease cleavage domain, a half-life prolongation (HLP) domain, and a CAP component that reduces the ability of the third binding domain to bind to the extracellular epitope of human CD3ε.

2. The precursor trispecific antibody construct according to claim 1, wherein the TAA bound by the first binding domain or the second binding domain or both is selected from EGFR, ROR1, and PSMA.

3. The precursor trispecific antibody construct according to claim 2, wherein (a) when the TAA is EGFR, the first binding domain or the second binding domain or both comprises the amino acid sequence recited in any one of SEQ ID NO: 34, 37 or a combination thereof; (b) when the TAA is ROR1, the first binding domain or the second binding domain or both comprises the amino acid sequence recited in any one of SEQ ID NO: 156 and 166 or a combination thereof; and (c) when the TAA is PSMA, the first binding domain or the second binding domain or both comprises the amino acid sequence recited in any one of SEQ ID NO: 168 and 170 or a combination thereof.

4. The precursor trispecific antibody construct according to claim 1, wherein the HLP domain comprises a human serum albumin (HSA) polypeptide.

5. The precursor trispecific antibody construct according to claim 1, wherein the CAP component comprises the amino acid sequence of the extracellular epitope of human CD3ε.

6. The precursor trispecific antibody construct according to claim 1, wherein the first binding domain, the second binding domain or both each comprises a single-chain variable fragment (scFv), and wherein the third binding domain comprises a Fab antigen-binding fragment.

7. The precursor trispecific antibody construct according to claim 1, wherein one or both of the first protease cleavage domain and the second protease cleavage domain comprises a protease-cleavable amino acid sequence that can be cleaved by a serine protease, a cysteine protease, an aspartic protease, a matrix metalloproteinase (MMP), or is a combined substrate that is cleaved by one or more of MMP2 / 9, uPA, cathepsin, and asparaginyl endopeptidase or any combination thereof.

8. A pharmaceutical composition comprising the precursor trispecific antibody construct according to claim 1 and a pharmaceutically acceptable carrier.

9. A nucleic acid construct or expression vector comprising one or more nucleic acid sequences encoding the precursor trispecific antibody construct according to claim 1.

10. Use of the precursor trispecific antibody construct according to any one of claims 1-7 or the pharmaceutical composition according to claim 8 in the preparation of a medicament for treating cancer or a tumor, preventing cancer or a tumor, inhibiting the growth of cancer or a tumor, delaying the disease progression of cancer or a tumor, reducing the tumor burden of cancer or a tumor, or reducing the incidence of cancer or a tumor.

Images