Multispecific and costimulatory antibody constructs and methods of use thereof

Abstract

The present disclosure describes costimulatory multivalent and multispecific antibody constructs capable of binding at least two different antigens on a cytotoxic effector cell and at least one disease-associated antigen (e.g., a TAA) on a target cell. Pharmaceutical compositions comprising such antibody constructs and methods of preparing and using such constructs and compositions, e.g., for the treatment of a disease such as cancer, are also disclosed.

Description

MULTISPECIFIC AND COSTIMULATORY ANTIBODY CONSTRUCTS ANDMETHODS OF USE THEREOFCROSS-REFERENCE

[0001] This application claims priority to and the benefit of U.S. Provisional Application No. 63 / 730,303, filed December 10, 2024, U.S. Provisional Application No. 63 / 685,886, filed August 22, 2024, U.S. Provisional Application No. 63 / 638,215, filed April 24, 2024, U.S. Provisional Application No. 63 / 575,343, filed April 5, 2024, and U.S. Provisional Application No. 63 / 560,940, filed March 4, 2024, the entire content of each of which is incorporated by reference herein for all purposes.TECHNICAL FIELD

[0002] The present disclosure generally relates to multispecific T cell engaging antibody constructs that can comprise a first binding domain capable of binding a first antigen on a cytotoxic effector cell, a second binding domain capable of binding a second antigen on a cytotoxic effector cell, and a third binding domain capable of binding a disease-associated antigen, such as a tumor- associated antigen (TAA) on a tumor cell.BACKGROUND

[0003] Cancer continues to pose a major unmet medical need, despite the considerable progress that has been made in its treatment over the past decades. While the current standard of care, as well as more recently developed anti -cancer therapies, have shown some clinical progress, various indications such as those with low T cell tumor infiltration still present a major clinical challenge. Bispecific T cell engagers (TCEs) have exhibited clinical successes in the treatment of hematological cancers, while treatment of solid tumors remains a challenge. Treatment of solid tumors with conventional CD3-engaging TCEs can result in limited T cell proliferation and recruitment to the tumor site, and treatment related T cell anergy, thus restricting the ability of bispecific TCEs to inhibit growth of these poorly infiltrated and rapidly growing tumors.SUMMARY

[0004] In various embodiments, the present disclosure describes multispecific antibody constructs that are capable of engaging two different antigens on one or more immune cell(s) (e.g., T cell(s)), as well as a disease-associated antigen, e.g., an antigen on a tumor cell. In certain embodiments ofthe present disclosure, described herein are costimulatory, multivalent and multi specific antibody constructs that comprise at least three binding domains, wherein a first binding domain is capable of binding a first antigen on a cytotoxic effector cell, a second binding domain capable of binding a second antigen on a cytotoxic effector cell, and a third binding domain capable of binding a disease-associated antigen, such as a TAA on a tumor cell. A costimulatory antibody construct of the present disclosure can be multivalent and multi specific and comprises the two effector cell (e.g., T cell) antigen binding domains in a specific configuration, that is the two effector cell antigen binding domains are in a certain proximity and relative orientation to each other, as further described herein.

[0005] In some embodiments, the present disclosure describes an antibody construct, comprising: (i) a first immunoglobulin (Ig) heavy chain, Hl, comprising, from N- to C-terminus, a first Fab heavy chain coupled to a first Fc polypeptide, (ii) a first Ig light chain, LI, comprising, from N- to C-terminus, a first Fab light chain coupled to a first scFv domain, and (iii) a second Ig heavy chain, H2, comprising a second Fc polypeptide, wherein: (a) the first Fab heavy chain of Hl and the first Fab light chain of LI form a first Fab domain that is capable of binding CD3 (e.g., CD3e or CD3 epsilon) on a cytotoxic effector cell; (b) the first scFv domain is capable of binding CD28 on a cytotoxic effector cell; and (c) the first Fc polypeptide and the second Fc polypeptide form a dimeric Fc domain. In certain embodiments, another (e.g., third) antigen binding domain is coupled to the N- or C-terminus of the second Fc polypeptide, wherein such an additional antigen binding domain is capable of binding a disease-associated antigen, such as a TAA on a tumor cell. The first light chain, LI, of an antibody construct described herein thus comprises the domain structure: (VL-CL)Fab-(VL-VH)SCFv or (VL-CL)Fab-(VH-VL)SCFv, optionally including one or more linkers as further described herein. Antibody constructs comprising such light chain LI can have a format also referred to herein as “Format A.”

[0006] In various embodiments, the disease-associated antigen is a TAA.

[0007] In some embodiments, the present disclosure describes an antibody construct, comprising: (i) a first Fab domain capable of binding CD3 on a cytotoxic effector cell; (ii) a first scFv domain capable of binding CD28 on a cytotoxic effector cell; (iii) a third binding domain capable of binding a TAA on a tumor cell; and (iv) a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the third binding domain is coupled to the N-terminus of the second Fcpolypeptide, and (c) the first scFv domain is coupled to the C-terminus of the light chain of the first Fab domain.

[0008] In some embodiments, the third (disease-associated) antigen binding domain is a second Fab domain. In other embodiments, the third antigen binding domain is a second scFv domain.

[0009] In some embodiments, the present disclosure describes an antibody construct comprising two immunoglobulin heavy chains, Hl and H2, and two immunoglobulin light chains, LI and L2, wherein: (i) Hl comprises, from N- to C-terminus, a first Fab heavy chain, comprising a first VH sequence and a first CHI sequence, coupled to a first Fc polypeptide; (ii) H2 comprises, from N- to C-terminus, a second Fab heavy chain, comprising a second VH sequence and a second CHI sequence, coupled to a second Fc polypeptide; (iii) LI comprises, from N- to C-terminus, a first Fab light chain, comprising a first VL sequence and a first CL sequence, coupled to a first scFv domain comprising a third VL sequence coupled to a third VH sequence; and (iv) L2 comprises, from N- to C-terminus, a second Fab light chain comprising a second VL sequence and a second CL sequence, wherein: (a) the first Fab heavy chain of Hl and the first Fab light chain of LI form a first Fab domain capable of binding CD3 on a cytotoxic effector cell; (b) the second Fab heavy chain of H2 and the second Fab light chain of L2 form a second Fab domain capable of binding a TAA on a tumor cell; (c) the first scFv domain is capable of binding CD28 on a cytotoxic effector cell; and (d) the first Fc polypeptide and the second Fc polypeptide form a dimeric Fc domain.

[0010] In some embodiments, the present disclosure describes an antibody construct comprising two immunoglobulin heavy chains, Hl and H2, and one immunoglobulin light chain, LI, wherein: (i) Hl comprises, from N- to C-terminus, a first Fab heavy chain, comprising a first VH sequence and a first CHI sequence, coupled to a first Fc polypeptide; (ii) LI comprises, from N- to C- terminus, a first Fab light chain, comprising a first VL sequence and a first CL sequence, coupled to a first scFv domain comprising a second VL sequence coupled to a second VH sequence; and (iii) H2 comprises, from N- to C-terminus, a second scFv domain comprising a third VL sequence and a third VH sequence, coupled to a second Fc polypeptide; wherein: (a) the first Fab heavy chain of Hl and the first Fab light chain of LI form a first Fab domain capable of binding CD3 on a cytotoxic effector cell; (b) the first scFv domain is capable of binding CD28 on a cytotoxic effector cell; (c) the second scFv domain is capable of binding a TAA on a tumor cell; and (d) the first Fc polypeptide and the second Fc polypeptide form a dimeric Fc domain.

[0011] Further described herein are pharmaceutical compositions comprising one or more of the antibody constructs of the present disclosure, and a pharmaceutically acceptable carrier, excipient, diluent, or combination thereof.

[0012] Further described herein is a method of producing an antibody construct of the present disclosure, the method comprising: (a) obtaining a host cell culture comprising at least one host cell comprising one or more nucleic acid molecules encoding one or more, two or more, three or more, or four or more of the polypeptide chains that form the antibody construct; and (b) recovering the antibody construct from the host cell culture.

[0013] Further described herein is a method of treating a tumor in a subject in need thereof, the method comprising administering to the subject an antibody construct of the present disclosure. In some embodiments, such tumor can be a solid tumor, such as a carcinoma, a sarcoma, or a lymphoma (e.g., Non-Hodgkin Lymphoma). In other embodiments, the tumor can be a liquid tumor (e.g., a hematologic malignancy) such as a leukemia.BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The implementations disclosed herein are illustrated by way of example, and not by way of limitation, in the accompanying drawings. The description and drawings are only for the purpose of illustration and as an aid to understanding and are not intended as a definition of the limits of the antibody constructs, pharmaceutical compositions, and methods of the present disclosure.

[0015] FIGS. 1A-1D and 1G-1H show schematic representations of the format and geometry of trivalent and trispecific antibody constructs that comprise binding domains capable of engaging CD3, CD28, and a TAA, according to certain embodiments of the present disclosure. FIGS. 1A and IB are representative depictions of antibody constructs having a format referred to herein “Format A.” FIGS. 1E-1F show schematic representations of bispecific control constructs used as described and used in this disclosure.

[0016] FIGS. 2A-2B show binding of certain anti-CD3 (Fab) / anti-CD28 (scFv) / anti-MSLN (scFv) trivalent and trispecific constructs (v34914, v34915, v34916 and v34917, FIG. 2A), and anti-CD3 (scFv) / anti-CD28 (Fab) / anti-MSLN (scFv) trivalent and trispecific constructs (v34913 and v34918, FIG. 2B) to CD4+ (left) and CD8+ (right) T cells, as measured by flow cytometry.

[0017] FIGS. 3A-3B show binding of certain trivalent and trispecific anti-CD3 / anti-CD28 / anti- Cldnl8.2 antibody constructs to CLDN18.2+ SNU 601 cells as measured by flow cytometry. Both,the constructs that comprised an anti-CD3 Fab domain and an anti-CD28 scFv domain (v37633, v37634 and v37635) as well as those that comprised an anti-CD28 Fab domain and an anti-CD3 scFv domain (v37638, v37640 and v37642), were compared against an anti-(Cldnl8.2xCD3) bispecific construct (v37663), a one-armed anti-Cldnl8.2 antibody (v37675), a benchmark control (v35923, also known as AMG910), an anti-Cldnl8.2 monoclonal antibody (mAb), and the anti- RSV protein F mAb palivizumab.

[0018] FIGS. 4A-4B show binding of certain anti-CD3 (Fab) / anti-CD28 (scFv) / anti- Cldnl8.2 (scFv) trivalent and trispecific constructs (v37633-v37637, FIG. 4A), and anti-CD3 (scFv) / anti- CD28 (Fab) / anti- Cldnl8.2 (scFv) trivalent and trispecific constructs (v37638-v37642, FIG. 4B) to CD4+ (left) and CD8+ (right) T cells, as measured by flow cytometry.

[0019] FIG. 5 shows binding of selected trivalent and trispecific anti-(CLDN18.2xCD3xCD28) antibody constructs, and the format-matched single arm binding controls, to CD3+CD28+ human T cells measured by flow cytometry.

[0020] FIG. 6 shows binding curves to (i) wildtype Jurkat cells (e.g., T cells expressing both CD3 and CD28), (ii) CD28 knock-out (KO) Jurkat cells, and (iii) CD3 knock-out (KO) Jurkat cells for certain trivalent and trispecific antibody constructs as well as for bispecific control constructs.

[0021] FIGS. 7A-7B show that certain anti-CD3 (Fab) / anti-CD28 (scFv) / anti-MSLN (scFv) trivalent and trispecific constructs (v34914, v34915, v34916 and v34917, FIG. 7A), and anti-CD3 (scFv) / anti-CD28 (Fab) / anti-MSLN (scFv) trivalent and trispecific constructs (v34913 and v34918, FIG. 7B) directed T cells from healthy donors to toward MSLN+ H292 target cells, inducing T cell mediated killing of the tumor cells.

[0022] FIGS. 8A-8D show in vitro T cell-dependent cytotoxicity induced by the trivalent and trispecific antibody constructs v34914 (FIG. 8A), v34916 (FIG. 8B), v34913 (FIG. 8C) and v34918 (FIG. 8D) and compared to the bispecific anti-(MSLNxCD3) benchmark control construct v34919 against MSLNhighH292 cells (152,986 MSLN / cell) and MSLNlowOVTOKO cells (9,752 MSLN / cell).

[0023] FIGS. 9A-9B show that the trivalent and trispecific antibody constructs v34913, v34916 and v34918 directed T cells from healthy donors to kill MSLN+ H292 target cells expressing moderate levels of MSLN. This long-term T cell-dependent cellular cytotoxicity (TDCC) study utilized a low E:T ratio of 1 :5 and an incubation period of either 3 days (FIG. 9A) or 7 days (FIG.9B), e.g., long-term co-culture at low E:T ratios to reflect conditions more accurately in certain (e.g., solid) tumor types.

[0024] FIGS. 10A-10B show that certain anti-CD3 (Fab) / anti-CD28 (scFv) / anti-Cldnl8.2 (scFv) trivalent and trispecific constructs (v37633-v37637, FIG. 10A), and anti-CD3 (scFv) / anti-CD28 (Fab) / anti- Cldnl8.2 (scFv) trivalent and trispecific constructs (v37638-v37642, FIG. 10B) directed T cells from healthy donors to kill CLDN18.2+ SNU 601 target cells. These trispecific constructs were compared against a bispecific anti-(Cldnl8.2xCD3) control (v37663), a bispecific anti-(Cldnl8.2xCD28) control (v37665) and a negative control (Het-Fc version of palivizumab, v22277).

[0025] FIGS. 11A-11D show in vitro T cell-dependent cytotoxicity induced by the trivalent and trispecific antibody constructs v37633 (FIG. 11A), v37634 (FIG. 11B), v37638 (FIG. 11C) and v37642 (FIG. 11D) and compared to the bispecific anti-(MSLNxCD3) benchmark control construct v35923 against CLDN18.2hi§hSNU 601 and CLDN18.2lowSKOV-3 target cells.

[0026] FIGS. 12A-12B show concentration response curves for a long-term (7 days) TDCC study for the tested anti-CLDN18.2 trivalent and trispecific constructs that either comprised anti-CD3 (Fab) / anti-CD28 (scFv) / anti-Cldnl8.2 (scFv) (v37633-v36735, FIG. 12A) or anti-CD3 (scFv) / anti-CD28 (Fab) / anti- Cldnl8.2 (scFv) (v37638, v37640, v37642, FIG. 12B) and showthat the constructs directed T cells from healthy donors to kill CLDN18.2+ target cells expressing varying levels of CLDN18.2 using an E:T ratio of 1 : 1. Top left: SNU-601 (276,125 CLDN18.2 / cell); top right: KATO-III (63,566 CLDN18.2 / cell); bottom: DAN-G (33,164 CLDN18.2 / cell).

[0027] FIG. 13 shows that several tested anti -CLDN18.2 trivalent and trispecific constructs with varying affinity for CD28, v37638, v37683, v37689, v37692 and v37694, directed T cells from healthy donors to kill CLDN18.2+ SNU 601 cells and compared to bispecific (v37663) and negative (v22277) control constructs.

[0028] FIG. 14A shows concentration response curves and in vitro cytotoxicities for the tested constructs in various DLL3 -positive cell lines in co-culture with human T cells. For this study, T cells were thawed at 37 °C and mixed with DLL3+ NCI-H82-RFP, SHP-77-RFP or COR-L279- RFP cells such that the ratio of T cells to tumor cells was adjusted to an effector to target cell (E:T) ratio of 1 :5, 1 :2 and 1 :2 for the H82, SHP-77 and COR-L279 cell lines, respectively. FIG. 14B shows concentration response curves and in vitro cytotoxicities for certain tested constructs invarious DLL3-positive cell lines in co-culture with human T cells in a follow-on study in which human T cells were thawed at 37 °C and mixed with DLL3+ NCI-H82-RFP, SHP-77-RFP, COR- L279-RFP, NCI-H69-RFP, NCI-H524-RFP and NCI-H526-RFP cells such that the ratio of T cells to tumor cells was adjusted to an effector to target cell (E:T) ratio of 1 :5 for NCI-H82 and 1 :2 for the remaining cell lines.

[0029] FIGS. 14C-14E show dose-response curves of primary human T cells in co-culture with DLL3+ NCI-H82-RFP cells as well as primary human Pan-T cells in co-culture with DLL3+ NCI- H82-RFP cells following treatment with the respective anti-(DLL3xCD3xCD28) trivalent and trispecific antibody constructs and bispecific controls, respectively.

[0030] FIGS. 15A-15B show that certain anti-CD3 (Fab) / anti-CD28 (scFv) / anti-MSLN (scFv) trivalent and trispecific constructs (v34914, v34915, v34916 and v34917, FIG. 15A), and anti- CD3 (scFv) / anti-CD28 (Fab) / anti-MSLN (scFv) trivalent and trispecific constructs (v34913 and v34918, FIG. 15B) induced Interleukin-2 (IL-2, left) and / or tumor necrosis factor alpha (TNFa, right) production from T cells when co-incubated with MSLN+ H292 cells (E:T of 2: 1) for 72 hours (hrs).

[0031] FIGS. 16A-16B show that certain anti-CD3 (Fab) / anti-CD28 (scFv) / anti-Cldnl8.2 (scFv) trivalent and trispecific constructs (v37633-v37637, FIG. 16A), and anti-CD3 (scFv) / anti-CD28 (Fab) / anti- Cldnl8.2 (scFv) trivalent and trispecific constructs (v37638-v37642, FIG. 16B) induced IL-2 (top) and / or TNFa (bottom) production from T cells co-incubated with CLDN18.2+ SNU 601 cells (E:T of 2: 1) for 72 hrs.

[0032] FIGS. 17A-17B show that certain anti -CLDN18.2 trivalent and trispecific constructs that either comprised anti-CD3 (Fab) / anti-CD28 (scFv) / anti-Cldnl8.2 (scFv) (v37633-v36735, FIG. 17A) or anti-CD3 (scFv) / anti-CD28 (Fab) / anti- Cldnl8.2 (scFv) (v37638, v37640, v37642, FIG. 17B) were capable of inducing the production of several cytokines (i.e., fFNy (top), IL-2 (middle), TNFa (bottom)) in T cells in the presence of CLDN18.2+ target SNU 601 using a 72 hours incubation period and an E:T ratio of 2: 1, and in comparison to certain bispecific and benchmark control constructs.

[0033] FIGS. 18A-18B show that several tested anti-CLDN18.2 trivalent and trispecific constructs with varying affinity for CD28, v37638, v37683, v37689, v37692 and v37694, induced IL-2 (FIG. 18A) or TNFa (FIG. 18B) production from T cells co-incubated with CLDN18.2+SNU 601 cells (E:T of 2: 1) for 72 hrs, and in comparison to bispecific (v37663) and negative (v22277) control constructs.

[0034] FIGS. 18C-18D show curves of the percent of cells proliferated with the respective titrated antibody constructs. FIG. 18C shows T cell proliferation in the presence of DLL3+NCI-H82 cells, and FIG. 18D shows T cell proliferation in the absence of DLL3+ cells.

[0035] FIG. 18E shows upregulation of Bcl-XL expression in activated T-cells using certain trispecific antibody constructs of the present disclosure as well as bispecific and benchmark controls. The data show that the trivalent and trispecific antibody construct, v39917, which comprised a light chain comprising a C-terminally attached anti-CD28 scFv domain, induced Bcl- XL upregulation similar to that of the combination treatment including the two bispecific anti- (CD28xDLL3) and anti-(CD3xDLL3) control constructs.

[0036] FIG. 19 shows in vivo anti-tumor activities in donor X-engrafted mice treated with either the bispecific control constructs, v35923 or v38417, or the trispecific and trivalent antibody construct, v37634. FIG. 20 shows the body weights of mice treated over the course of the in vivo efficacy study referenced in FIG. 19.

[0037] FIG. 21 shows flow cytometric analysis of T cell populations in blood and tumor following administration of constructs including the trivalent and trispecific antibody construct v37634.

[0038] FIG. 22 shows percent body weight loss compared to baseline (e.g., body weight measured prior to PBMC engraftment) following treatment with vehicle control, 1 mg / kg of v37634, v37663, or ANC28.1 / 5D10, as well as 0.25 mg / kg of OKT3 (also known as Orthoclone OKT3 or Muromonab-CD3).

[0039] FIG. 23A shows the change in tumor volume and FIG. 23B shows the change in body weight following IV administration of 0.05 mg / kg (qlwx4) of either the trivalent and trispecific antibody construct v37634, a clinical benchmark, or a negative control to mice bearing SNU620 gastric tumors.

[0040] FIG. 24 shows tumor growth delay observed in an in vivo study using donor A-engrafted mice treated with 2.5 mg / kg of the trivalent and trispecific antibody construct v38967 when compared to the clinical benchmark. FIG. 25 shows that percent survival in this in vivo study was significantly higher in mice receiving 2.5 mg / kg of v38967 compared to mice receiving equivalent doses of clinical benchmark (Long-rank p = 0.0047). FIG. 26 shows that body weights of mice treated with v38967 over the course of this in vivo study remained stable.

[0041] FIG. 27 shows tumor regression observed in donor A-engrafted mice treated with the trivalent and trispecific antibody construct v39917 compared to the clinical benchmark (v38589) and irrelevant mAb (v22277). FIG. 28 and FIG. 29 show the percent survival and change in body weight, respectively, over the course of this in vivo study.

[0042] FIG. 30 shows the qualitative food consumption of male cynomolgus monkeys (n=3) that were given a repeat intravenous (i.v.) dose of 3 mg / kg of v40063 once weekly for two weeks. FIGS. 31A-31C show hematological changes, more specifically red cell mass parameters (red blood cells (FIG. 31A), hematocrit (FIG. 31B) and hemoglobin (FIG. 31C)), measured in male cynomolgus monkeys (n=3) that were given a repeat i.v. dose of 3 mg / kg of v40063 once weekly for two weeks. FIGS. 32A-32D show additional hematological changes observed in the study referred to in FIG. 30 including lymphocyte count (FIG. 32A), reticulocyte count (FIG. 32B), neutrophil count (FIG. 32C), and monocyte count (FIG. 32D). FIGS. 33A-33D show changes in serum concentrations of the following cytokines: IL-6 (FIG. 33A), IFN-y (FIG. 33B), TNF-a (FIG. 33C) and MCP-1 (FIG. 33D).

[0043] FIG. 34 depicts curves showing the percent of cells proliferated with the respective titrated antibody constructs tested.

[0044] FIG. 35 shows tumor regression observed in donor B-engrafted mice treated with v41086, CD28-KO (v41753), and the clinical benchmark (v38589), compared to the irrelevant mAb (v40552). Treatment with 6.6 nmol / kg and 3.3 nmol / kg of v41086 resulted in full tumor regression in 7 out of 7 mice. Body weights remained stable for all mice as shown in FIG. 36.

[0045] FIG. 37 shows tumor growth inhibition as observed in donor-engrafted mice treated with 2.85 nmol / kg of the trivalent and trispecific antibody construct v41086 when compared to treatment with the clinical benchmark (v38985).

[0046] FIG. 38 shows serum cytokine (IFNy, IL-2, TNFa and IL-6; in pg / mL) concentrations measured 72 hours post-treatment of humanized mice with the respective construct.

[0047] FIG. 39 shows serum concentrations of the measured cytokines IL-6, IL- 10, MCP-1 and IL-2 in cynomolgus monkeys that were given two i.v. doses of 10 mg / kg of the respective tested constructs.

[0048] FIG. 40 shows the impact of certain antibody constructs having a format according to either Format A, B or C, on the viability of T cells in the absence of tumor cells.

[0049] FIG. 41 shows target-independent cytokine induction by certain antibody constructs having a format according to either Format A, B or C.

[0050] FIG. 42 shows the percentage bridging of CD3(KO) T cells to CD28(KO) T cells by certain antibody constructs having a format according to either Format A, B or C.DETAILED DESCRIPTION

[0051] In various embodiments, the present disclosure describes costimulatory antibody constructs that can be multivalent and multispecific and capable of binding two different antigens on a cytotoxic effector cell (e.g., a T cell), and either (i) a TAA on a tumor cell or (ii) an antigen associated with another disease such as an inflammatory or an autoimmune disease. Such antibody construct can comprise at least a first binding domain capable of binding a first antigen on a cytotoxic effector cell, a second binding domain capable of binding a second antigen on a cytotoxic effector cell, and a third binding domain capable of binding a TAA on a tumor cell or another disease-associated antigen (such as an inflammatory or autoimmune disease-associated antigen).

[0052] In certain embodiments, described herein are multivalent and multispecific T cell engaging antibody constructs capable of costimulating an effector cell (e.g., a T cell), e.g., by engaging two antigens on such effector cell. In some embodiments, these effector cell antigens are CD3 (which includes the epsilon chain of CD3, e.g., CD3s or CD3-epsilon, and CD28. As further described herein, the costimulatory T cell engaging antibody constructs of the present disclosure can comprise a certain format and geometry that enables obligate cis binding of CD3 and CD28 on the same cytotoxic effector cell (as further defined and described herein), rather than binding CD3 on one cell (e.g., a first T cell) and CD28 on another cell (e.g., a second T cell) and cross-linking such effector cells (e.g., T cells).

[0053] In various embodiments, described herein are multivalent and multispecific antibody constructs that comprise an immunoglobulin (Ig) light chain, LI, comprising a Fab light chain portion (e.g., a first VL sequence and a CL sequence) and an scFv portion (e.g., a second VL sequence and a VH sequence), also referred to herein as “Format A”. In various embodiments, the Fab light chain portion of LI can form an anti-CD3 Fab domain with a Fab heavy chain portion of a corresponding Ig heavy chain, Hl, while the scFv domain of LI is capable of engaging CD28. The format (e.g., Fab, scFv, etc.), spatial orientation and relative spatial distance between the anti- CD3 and anti-CD28 binding domains was optimized to generate costimulatory antibody constructs with increased target antigen (e.g., TAA)-dependent activity (e.g., T cell activation, cytokinerelease, etc.), reduced T cell toxicity, obligate cis binding of CD3 and CD28 on the same T cell (e.g., reducing the risk of T cell mediated fratricide), and conditional engagement of CD28 requiring CD3 binding for induction CD28 signaling.I. DEFINITIONS

[0054] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art.

[0055] The term “about,” as used herein in the context of a numerical value or range, generally refers to ±10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or ±1% of the numerical value or range recited or claimed, unless otherwise specified. In various embodiments, the term “about” refers to an approximately ±10% variation from a given value or range. In other embodiments, the term “about” refers to an approximately ±5% variation from a given value or range. In yet other embodiments, the term “about” refers to an approximately ±1% variation from a given value or range. It is to be understood that such a variation is always included in any given value provided herein, whether it is specifically referred to or not.

[0056] The use of the word “a” or “an,” when used herein in conjunction with the term “comprising,” can mean “one,” but it is also consistent with the meaning of “one or more,” “at least one” and “one or more than one.”

[0057] As used herein, the terms “comprising,” “having,” “including” and “containing,” and grammatical variations thereof, are inclusive or open-ended and do not exclude additional, unrecited elements and / or method steps. The term “consisting essentially of’ when used herein in connection with a construct, composition, use or method, denotes that additional features, elements and / or method steps can be present, but that these additions do not materially affect the manner in which the recited construct, composition, method or use functions. The term “consisting of,” when used herein in connection with a construct, composition, use or method, excludes the presence of additional elements and / or method steps. An antibody construct, composition, use, or method described herein as comprising certain elements and / or steps can also, in certain embodiments, consist essentially of those elements and / or steps, and in other embodiments, consist of those elements and / or steps, whether or not these embodiments are specifically referred to.

[0058] The terms “subject” and “patient” can be used interchangeably herein and generally refer to an animal in need of treatment. An animal in need of treatment can be a human or a non-human animal, such as a mammal, bird, or fish. In certain embodiments, the subject or patient is amammal. In some embodiments, the subject is a human. In other embodiments, the subject is a rodent or a non-human primate.

[0059] An “effective amount” of a costimulatory antibody construct described herein, or a pharmaceutical composition comprising such antibody construct, in respect of a particular result to be achieved is an amount sufficient to achieve the desired result. For example, an “effective amount” of an antibody construct or pharmaceutical composition when referred to in respect of the killing of cancer cells, refers to an amount of antibody construct or composition comprising the antibody construct sufficient to produce a killing effect.

[0060] Unless specified otherwise, the terms “Fc region,” “Fc” and “Fc domain” are used interchangeably herein and refer to a C-terminal region of an immunoglobulin (Ig) heavy chain that contains at least a portion of a constant region. In various embodiments, an Fc domain herein can be dimeric. Such dimeric Fc domain can comprise a first Fc polypeptide and a second Fc polypeptide, wherein each Fc polypeptide can comprise a CH2 domain and a CH3 domain. Such dimeric Fc can either be homodimeric, i.e., comprising first and second Fc polypeptides that have identical amino acid sequences, or heterodimeric, i.e., comprising first and second Fc polypeptides that have different amino acid sequences, e.g., sequences that share about 95%, 96%, 97%, 98%, or about 99% sequence identity. In some embodiments, an antibody construct of the present disclosure comprises a homodimeric Fc domain. In yet other embodiments, and as further described herein, an antibody construct comprises a heterodimeric Fc domain in which at least one of the CH2 and / or CH3 domains of the first and second Fc polypeptides have amino acid sequences that share about 99% or less, 98% or less, or about 97% or less sequence identity.

[0061] The term “antibody construct,” as used herein, generally refers to an antibody-like molecule comprising at least one immunoglobulin (Ig) domain, such as an Fc domain, a Fab domain, etc., and comprising two or more, three or more, or four or more polypeptide chains that are covalently and / or non-covalently associated with each other. The polypeptide chains of an antibody construct can include one or more Ig heavy chain(s) and one or more Ig light chain(s). A costimulatory antibody construct, according to various embodiments of the present disclosure, comprises at least three or at least four polypeptide chains including two Ig heavy chains (e.g., Hl, H2, etc.) and at least one light chain, wherein the light chain is part of a costimulatory portion of the antibody construct and comprises a Fab portion capable of binding a first antigen (e.g., CD3) on a cytotoxic effector cell and an scFv portion capable of binding a second antigen (e.g., CD28)on a cytotoxic effector cell. FIGs. 1A and IB illustrate certain embodiments of costimulatory antibody constructs of the present disclosure comprising at least three (e.g., two Ig heavy and one Ig light chain, FIG. 1A) or at least four (e.g., two Ig heavy and two Ig light chains, FIG. IB).

[0062] The term “multispecific,” as used herein in the context of an antibody construct, refers to a biologically functional protein (e.g., an antibody construct as described herein) which is “at least bispecific,” i.e., it comprises at least a first binding domain and a second binding domain, wherein such first and second binding domain can bind specifically two distinct epitopes, e.g., a first epitope and a second epitope. Such first and second epitopes can be located on the same antigen or on different antigens, e.g., a first epitope on cluster of differentiation 3 (CD3, e.g., CD3s) and a second epitope on cluster of differentiation 28 (CD28), or a first epitope on CD3 or CD28 and a second epitope on a TAA. Accordingly, in some embodiments, antibody constructs according to the present disclosure can comprise specificities for at least two different epitopes located on two different antigens (e.g., CD3 and CD28, or CD3 and a TAA). In some of these embodiments, a multispecific antibody construct of this disclosure can comprise specificities for at least three different epitopes located on three different antigens (e.g., CD3, CD28, and a TAA). Hence, the term “multispecific” in the context of an antibody construct herein encompasses antibody constructs that are at least bispecific (i.e., comprising two binding domains with specificities for two different antigens or targets), at least trispecific (i.e., comprising three binding domains with specificities for three different antigens or targets, e.g., CD3, CD28, and a TAA), or even tetraspecific and so forth.

[0063] The term “trispecific,” as used herein in the context of an antibody construct, refers to a biologically functional protein (e.g., an antibody construct as described herein) which is “at least trispecific,” i.e., it comprises at least a first binding domain, a second binding domain and a third binding domain, wherein such first, second and third binding domains can bind specifically three distinct epitopes, e.g., a first epitope, a second epitope and a third epitope. Such first, second and third epitopes can be located on the same antigen or on different antigens, e.g., a first epitope on CD3, a second epitope on CD28, and a third epitope on a TAA (e.g., a Claudin (Cldn) such as Cldn6 or Cldnl8.2, mesothelin (MSLN), or another antigen expressed by a tumor cell). Accordingly, in some embodiments, trispecific antibody constructs according to the present disclosure can comprise specificities for at least three different antigens or targets. Hence, the specificity in the context of an antibody construct herein describes the total number of differentepitopes and / or antigens an antibody construct can bind to, e.g., a monospecific antibody construct comprises one binding domain with a specificity for one epitope or antigen, a bispecific antibody construct comprises two binding domains with specificities for two different epitopes and / or antigens, and a trispecific antibody construct comprises three binding domains with specificities for three different epitopes and / or antigens, and so forth.

[0064] The term “multivalent,” as used herein in the context of an antibody construct, refers to a biologically functional protein (e.g., an antibody construct as described herein) which is “at least bivalent,” i.e., it comprises at least two binding domains, e.g., at least a first binding domain and a second binding domain, wherein each of the first and second binding domains is capable of specifically binding an epitope and / or antigen, e.g., CD3, CD28, or a TAA. The at least two binding domains can bind to the same epitope or antigen or to different epitopes or antigens. The term “multivalent” encompasses antibody constructs that are bivalent, trivalent, tetravalent, and so forth.

[0065] The term “trivalent,” as used herein in the context of an antibody construct, refers to a biologically functional protein (e.g., an antibody construct as described herein) which is “at least trivalent,” i.e., it comprises three binding domains, e.g., at least a first binding domain, a second binding domain and a third binding domain, wherein each of the first, second, and third binding domains is capable of specifically binding an epitope and / or antigen, e.g., CD3, CD28, or a TAA. The three binding domains can either have specificities for three different epitopes or antigens, or two or more of the three binding domains have a specificity for the same epitope or antigen. Hence, the valency in the context of an antibody construct herein, e.g., being mono-, bi-, or trivalent, describes the total number of antigen binding domains of an antibody construct. Accordingly, the valency of an antibody construct generally is at least equal to its specificity, i.e., a trispecific antibody construct has to be at least trivalent. In embodiments in which the antibody construct is multivalent and multispecific, each of the three binding domains of a construct is capable of binding a different epitope or antigen, and thus the construct engages each of the three epitopes and / or antigens monovalently.

[0066] As used herein, the term “format” in the context of an antibody construct described herein generally describes attributes of the antibody construct such as its antigen valency (e.g., a construct being mono- or bivalent for a given antigen), the type(s) of binding domain(s) (e.g., possessing one or more scFv domain(s), Fab domain(s), etc.) present in an antibody construct, as well as thepresence, absence, and / or type of an Fc domain (e.g., homodimeric, heterodimeric, containing one or more constant heavy domains, CH2, CH3, etc.). As an example, in some embodiments, an antibody construct of the present disclosure can be multivalent and multispecific in a 1+1+1 format illustrating that the construct contains three binding domains, wherein each binding domain has an affinity for a different antigen (e.g., CD3, CD28 and a TAA), i.e., the trivalent construct is monovalent (as indicated by “1”) for each of the three antigens.

[0067] The term “dissociation constant (KD or Ka)” as used herein, is intended to refer to the equilibrium dissociation constant of a particular ligand-protein interaction. As used herein, ligandprotein interactions refer to, but are not limited to protein-protein interactions or antibody-antigen interactions. The KD measures the propensity of two proteins complexed together (e.g. AB) to dissociate reversibly into constituent components (A+B), and is defined as the ratio of the rate of dissociation, also called the “off-rate (koff)”, to the association rate, or “on-rate (kon)”. Thus, KD equals koff / konand is expressed as a molar concentration (M). It follows that the smaller the KD, the stronger the affinity of binding, and thus a decrease in KD indicates an increase in affinity. Therefore, a KD of 1 mM indicates weak binding affinity compared to a KD of 1 nM. Affinity is sometimes measured in terms of a KA or Ka, which is the reciprocal of the KD or Kd. KD values for antigen-binding constructs can be determined using methods well established in the art. One method for determining the KD of an antigen-binding construct is by using surface plasmon resonance (SPR), typically using a biosensor system such as a Biacore® system. Isothermal titration calorimetry (ITC) is another method that can be used to measure KD.

[0068] As used herein, the term “geometry” in the context of an antibody construct described herein generally describes the overall (e.g., 3 -dimensional) structure of an antibody construct, such as the relative spatial localization, arrangement and / or connectivity of the various domains of an antibody construct, e.g., the relative arrangement and connectivity of binding domains and Fc domain, as further described herein and as illustrated in FIGS. 1A-1D according to certain embodiments of the present disclosure.

[0069] As used herein, the term “costimulatory” or “costimulation” in the context of an antibody construct described herein generally refers to the ability of an antibody construct to bind CD3 and CD28 on a cytotoxic effector cell, e.g., a T cell, in vitro and / or in vivo. In various embodiments, the term “costimulatory” in the context of an antibody construct described herein refers to the ability of an antibody construct to bind CD3 and CD28 on the same cytotoxic effector cell, e.g., aT cell, in vitro and / or in vivo, as compared to binding CD3 and CD28 on the same cytotoxic effector cell on two different cytotoxic effector cells.

[0070] In various embodiments, an antibody construct of the present disclosure that is capable of binding CD3 refers to a construct that is able to bind an epitope on the epsilon chain of CD3, also referred to herein as “CD3s,” “CD3e” or “CD3 epsilon.”

[0071] Generally, and unless specified otherwise, an amino acid sequence of a polypeptide described herein is described and defined in the direction from its N- to its C-terminus. As an example, a polypeptide described as comprising an scFv domain that is coupled to an Fc polypeptide is defined herein as a polypeptide in which the C-terminus of the scFv domain is coupled, either with or without a linker, to the N-terminus of the Fc polypeptide and the domain structure can be described as: scFv-Fc, or with the inclusion of a linker as: scFv-LinkerscFv'Fc-Fc.

[0072] As used herein, abbreviations such as “Hl” and “H2,” or “A” and “B,” are generally used as generic heavy chain identifiers and broadly refer to a first heavy chain and a second heavy chain of an antibody construct, respectively, and thus are not intended to be limited to any specific heavy chain amino acid (or polynucleotide) sequences.

[0073] The term “amino acid modification,” as used herein in the context of an amino acid sequence of a polypeptide that has been modified relative to a reference (e.g., wildtype (WT)) sequence, generally refers to an amino acid sequence of a polypeptide in which one or more amino acid substitution(s), one or more amino acid insertion(s), and / or one or more amino acid deletion(s) have been introduced relative to a corresponding unmodified (e.g., WT) amino acid sequence of the polypeptide.

[0074] Descriptions of antibody constructs such as “anti-(TAAxCD28xCD3)” and “anti- TAA / anti-CD28 / anti-CD3” can be used interchangeably herein and generally refer to an antibody construct that is at least trispecific and thus contains at least three binding domains that are capable of binding an epitope on TAA, CD28 and CD3, respectively. In various embodiments, such description refers to a costimulatory, multivalent and multispecific antibody construct having at least three binding domains, one capable of binding an epitope on TAA, one an epitope on CD28 and one an epitope on CD3, respectively, wherein the TAA can be DLL3, MSLN, Cldnl8.2 or, in various embodiments, any other tumor antigen described herein and known in the art.

[0075] Generally, it is to be understood that the positive recitation of a feature in one embodiment serves as a basis for excluding the feature in an alternative embodiment. In particular, where a listof options is presented for a given embodiment or claim, it is to be understood that one or more option can be deleted from the list and the shortened list can form an alternative embodiment, whether or not such an alternative embodiment is specifically referred to.

[0076] It is further contemplated that any embodiment discussed herein can be implemented with respect to any antibody construct, method, use, or composition disclosed herein, and vice versa. Furthermore, modifications of the specific embodiments described herein that would be apparent to those skilled in the art are intended to be included within the scope of the claims recited herein.II. COSTIMULATORY ANTIBODY CONSTRUCTS

[0077] As further described herein, in various embodiments, the present disclosure relates to costimulatory, multivalent and multispecific antibody constructs that comprise at least three binding domains, with two of such binding domains directed against CD3 and CD28 expressed by an immune cell, e.g., T cell, and a disease-associated antigen, such as a TAA expressed by a tumor cell, or an antigen associated with an inflammatory or autoimmune disease.A. Format and Geometry of Antibody Constructs

[0078] In various embodiments, the antibody constructs described in the present disclosure are costimulatory, multivalent and multispecific and comprise at least three antigen binding domains (which can also be referred to herein as “paratopes”), with each of the three binding domains being capable of binding a different antigen.

[0079] In certain embodiments, a multivalent and multispecific antibody construct of the present disclosure can consist of or comprise a format of 1+1+1, indicating that each of the at least three binding domains binds a different antigen, and hence such construct binds each antigen in a monovalent manner. Such 1+1+1-format antibody constructs can also be described as trivalent and trispecific.

[0080] In other embodiments, a multivalent and multispecific antibody construct of the present disclosure can comprise at least four antigen binding domains and can consist of or comprise a format of 2+1+1, indicating that two of the at least four binding domains each bind a different antigen in a monovalent (“1”) manner, and the other two of the at least four binding domains bind a third antigen in a bivalent (“2”) manner. Such 2+1+1-format antibody constructs can also be described as tetravalent and trispecific.

[0081] In yet other embodiments, a multivalent and multispecific antibody construct of the present disclosure can comprise at least four antigen binding domains and can consist of or comprise aformat of l+l+l+l, indicating that each of the at least four binding domains binds a different antigen in a monovalent (“1”) manner. Such l+l+l+l -format antibody constructs can be described as tetraval ent and tetraspecific.

[0082] As further described herein, any one of the 2, 3, 4 or more antigen binding domains of an antibody construct can have a different structure or format. In some embodiments, any one of the 2, 3, 4 or more antigen binding domains of an antibody construct can comprise or consist of a single-domain antibody (e.g., a VH domain), a VHH antibody, an scFv domain, a Fab domain, or any other binding domain structure known in the art, such as a linear polypeptide sequence.

[0083] In certain embodiments, a costimulatory antibody construct of the present disclosure comprises one or more scFv domain(s) and / or one or more Fab domain(s) as antigen binding domains.

[0084] As further described herein, in various embodiments, a multivalent and multispecific antibody construct of this disclosure can comprise two structural portions: (i) a costimulation (or costimulatory) portion that comprises or consists of a first heavy (Hl) and first light (LI) chain pair, e.g., a Hl-Ll pair, that comprises at least an anti-CD3 binding domain and an anti-CD28 binding domain, and (ii) a target antigen (e.g., TAA)-binding portion, comprising a target antigen binding domain that can either comprise or consist of a second heavy chain, H2 (e.g., in instances in which the target antigen binding domain is an scFv domain or a single-chain Fab), or a second heavy and light chain pair, e.g., H2-L2 (e.g., in instance in which the target antigen binding domain is a Fab domain). These multivalent and multispecific antibody constructs can have a format also referred to herein as “Format A,” as further described elsewhere herein, and which generally refers to an antibody construct in which the costimulatory portion comprises an anti-CD3 Fab domain coupled to the N-terminus of a first Fc polypeptide, and an anti-CD28 scFv domain, and wherein the anti-CD28 scFv domain is coupled to the C-terminus of the light chain of the anti-CD3 Fab domain; the target antigen binding portion may comprise a target antigen binding domain coupled to the N-terminus of a second Fc polypeptide, wherein the format of the target antigen binding domain may vary (e.g., may comprise an scFv domain or a Fab domain, as shown in, e.g., FIG. 1A and IB).

[0085] In various embodiments, described herein are multivalent and multispecific antibody constructs comprising three antigen binding domains, two of which are capable of binding two different antigens on a cytotoxic effector cell (e.g., an immune cell such as a T cell), and onebinding domain is capable of binding a disease-associated antigen (e.g., a TAA) on a target cell (e.g., a tumor cell). In certain embodiments, such costimulatory multivalent and multispecific antibody constructs comprise a (i) costimulation portion comprising a Hl-Ll pair that comprises an anti-CD3 Fab domain and an anti-CD28 scFv domain as well as a first Fc polypeptide, and in which the anti-CD28 scFv domain is coupled to the C-terminus of the Fab light chain of the anti- CD3 Fab domain, and (ii) a TAA-binding portion comprising an anti-TAA binding domain and a second Fc polypeptide.

[0086] In some embodiments, described herein is a multivalent and multispecific antibody construct, comprising: (i) a first immunoglobulin (Ig) heavy chain, Hl, comprising, from N- to C- terminus, a first Fab heavy chain coupled to a first Fc polypeptide, (ii) a first Ig light chain, LI, comprising, from N- to C-terminus, a first Fab light chain coupled to a first scFv domain, and (iii) a second Ig heavy chain, H2, comprising a second Fc polypeptide, wherein: (a) the first Fab heavy chain of Hl and the first Fab light chain of LI form a first Fab domain that is capable of binding CD3 on a cytotoxic effector cell; (b) the first scFv domain is capable of binding CD28 on a cytotoxic effector cell; and (c) the first Fc polypeptide and the second Fc polypeptide form a dimeric Fc domain. In some embodiments, such antibody construct further comprises a binding domain coupled to the N-terminus of the second Fc polypeptide, wherein such a binding domain is capable of binding a TAA on a tumor cell.

[0087] In some embodiments, described herein is a multivalent and multispecific antibody construct, comprising: (i) a first immunoglobulin (Ig) heavy chain, Hl, comprising, from N- to C- terminus, a first Fab heavy chain coupled to a first Fc polypeptide, (ii) a first Ig light chain, LI, comprising, from N- to C-terminus, a first Fab light chain coupled to a first scFv domain, and (iii) a second Ig heavy chain, H2, comprising a second Fc polypeptide, wherein: (a) the first Fab heavy chain of Hl and the first Fab light chain of LI form a first Fab domain that binds CD3 on a cytotoxic effector cell; (b) the first scFv domain that is coupled to the C-terminus of the first Fab light chain binds CD28 on a cytotoxic effector cell; and (c) the first Fc polypeptide and the second Fc polypeptide form a dimeric Fc domain.

[0088] In some embodiments, such multivalent and multispecific antibody construct further comprises a binding domain coupled to the N-terminus of the second Fc polypeptide, wherein such binding domain binds a TAA on a tumor cell. In some of these embodiments, H2 further comprises a second Fab heavy chain that is coupled to the N-terminus of the second Fc polypeptide and asecond Ig light chain, L2, comprising a second Fab light chain, and wherein the second Fab heavy chain of H2 and the second Fab light chain of L2 form the second Fab domain.

[0089] In some embodiments, Hl further comprises a first linkerFab'Fcthat couples the first Fab heavy chain to the first Fc polypeptide. Such first linkerFab'Fccan comprise or consist of an IgGl, IgG2 or IgG4 hinge region, or are a derivative thereof, wherein the derivative has at least about 80%, 85%, 90%, 93%, 95%, 97%, or 99% sequence identity to a wildtype IgGl, IgG2 or IgG4 hinge region. In some embodiments, the first linkerFab'Fcand the second linkerFab'Fcindependently comprise or consist of an IgGl, IgG2 or IgG4 hinge region, or are a derivative thereof, wherein each of the derivatives has at least about 80%, 85%, 90%, 93%, 95%, 97%, or 99% sequence identity to a wildtype IgGl, IgG2 or IgG4 hinge region.

[0090] In some embodiments, the first linkerFab'Fcand the second linkerFab'Fcindependently comprise or consist of an amino acid sequence having at least about 80%, 85%, 90%, or at least about 95% sequence identity to the sequence set forth in SEQ ID NO: 56 or 54.

[0091] In some embodiments, LI further comprises a linkerFab'scFvthat couples the first Fab light chain to the first scFv domain. Such linkerFab'scFvcan comprise or consist of the amino acid sequence set forth in SEQ ID NO: 58, or a sequence having 1, 2, or 3 amino acid modifications relative thereto.

[0092] In some embodiments, the first scFv domain that is capable of binding CD28 and comprises an anti-CD28 VH sequence and an anti-CD28 VL sequence can further comprise a first LinkerscFv(e.g., LinkerscFvl). In some embodiments, the first scFv domain has the domain structure, from N- to C-terminus, of VL-LinkerscFvl-VH. In other embodiments, the first scFv domain has the domain structure, from N- to C-terminus, of VH-LinkerscFvl-VL.

[0093] In some embodiments, described herein are costimulatory, multivalent and multispecific antibody constructs that comprise at least two Fab domains and at least one scFv domain as binding domains that each target a different antigen. Such antibody constructs can also be referred to as heterodimeric Fab or “Het-Fab” antibody constructs because each Fab domain binds a different antigen and can thus comprise VH and VL sequences with different sequences and, optionally, can contain one or more amino acid modifications in the CHI and / or CL sequences of each Fab domain to facilitate correct heavy and light chain pairing, as further described herein.

[0094] Thus, in various embodiments, the binding domain that is coupled to the N-terminus of the second Fc polypeptide of an antibody construct herein can be a second Fab domain. FIG. IB showsa schematic representation of such an antibody construct according to certain embodiments of this disclosure.

[0095] In some embodiments, described herein is a multivalent and multispecific antibody construct (e.g., a Fab2construct such as a Het-Fab antibody construct), comprising: (i) a first Fab domain capable of binding CD3 on a cytotoxic effector cell; (ii) a first scFv domain capable of binding CD28 on a cytotoxic effector cell; (iii) a second Fab domain capable of binding a TAA on a tumor cell; and (iv) a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the second Fab domain is coupled to the N-terminus of the second Fc polypeptide, and (c) the first scFv domain is coupled to the C-terminus of the light chain of the first Fab domain.

[0096] In certain embodiments of a multivalent and multispecific Het-Fab antibody construct, the first Fab domain, the second Fab domain, or both Fab domains comprise one or more amino acid substitutions relative to a reference Fab heavy and / or light chain sequence that promotes correct heavy and light chain pairing. Amino acid modifications that can promote heavy and light chain pairing are known in the art and are described in, e.g., International Patent Publication Nos. WO 2014 / 082179, WO 2015 / 181805, and WO 2017 / 059551, which are incorporated herein by reference in their entireties.

[0097] In some embodiments of a multivalent and multispecific Het-Fab antibody construct, the first Fab domain targeting CD3 and the second Fab domain targeting a TAA each comprise a kappa light chain.

[0098] In other embodiments of a multivalent and multispecific Het-Fab antibody construct, the first Fab domain and the second Fab domain each comprise a lambda light chain.

[0099] In yet other embodiments of a costimulatory Het-Fab antibody construct, one of the first Fab domain and the second Fab domain comprises a kappa light chain and the other Fab domain comprises a lambda light chain.

[0100] In certain embodiments of a multivalent and multispecific Het-Fab antibody construct in which one Fab domain comprises a kappa light chain and the other Fab domain comprises a lambda light chain, the Fab domain comprising the kappa light chain comprises one or more of the amino acid substitutions 143E, 145T, 179E and 228D in the CHI sequence of the heavy chain and one or more of the amino acid substitutions 121K, 124R and 178R in the CL sequence of the kappa light chain, and the other Fab domain comprising the lambda light chain comprises one or more of theamino acid substitutions 125R and 188K in the CHI sequence of the heavy chain and one or more of the amino acid substitutions 122D, 129T, 176E and 178E in the CL sequence of the lambda light chain, wherein the amino acid residues are identified according to the Kabat numbering system.

[0101] In certain embodiments of a multivalent and multispecific Het-Fab antibody construct in which one Fab domain comprises a kappa light chain and the other Fab domain comprises a lambda light chain, the Fab domain comprising the kappa light chain comprises one or more of the amino acid substitutions 124R and 186K in the CHI sequence of the heavy chain and one or more of the amino acid substitutions 133G, 135W, 176D and 180E in the CL sequence of the kappa light chain, and the other Fab domain comprising the lambda light chain comprises one or more of the amino acid substitutions 139W, 143D and 145T in the CHI sequence of the heavy chain and one or more of the amino acid substitutions 124Q and 131R in the CL sequence of the lambda light chain, wherein the amino acid residues are identified according to the Kabat numbering system.

[0102] In various other embodiments, described herein are multivalent and multispecific antibody constructs that comprise at least one Fab domain and at least two scFv domains as binding domains. Such antibody constructs can also be referred to as costimulatory multivalent and multispecific scFv2antibody constructs indicating that such constructs comprise at least two scFv binding domains.

[0103] In such embodiments, the binding domain that is coupled to the N-terminus of the second Fc polypeptide of an antibody construct herein is a second scFv domain comprising a second VH sequence and a second VL sequence. FIG. 1A shows a schematic of such an scFv2antibody construct according to certain embodiments of this disclosure.

[0104] In some embodiments, described herein is a multivalent and multispecific antibody construct (e.g., a scFv2antibody construct), comprising: (i) a first Fab domain capable of binding CD3 on a cytotoxic effector cell; (ii) a first scFv domain capable of binding CD28 on a cytotoxic effector cell; (iii) a second scFv domain capable of binding a TAA on a tumor cell; and (iv) a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide, and (c) the first scFv domain is coupled to the C-terminus of the light chain of the first Fab domain.

[0105] In some of these embodiments, H2 can further comprise a first linkerscFv'Fcthat couples the second scFv domain to the second Fc polypeptide. Such first linkerscFv'Fccan comprise or consist of an IgGl, IgG2 or IgG4 hinge region, or are a derivative thereof, wherein the derivative has at least about 80%, 85%, 90% or 95% sequence identity to a wildtype IgGl, IgG2 or IgG4 hinge region. In some embodiments, the first linkerscFv'Fccan comprise or consist of an amino acid sequence having at least about 805, 85%, 90%, or at least about 95% sequence identity to the sequence set forth in SEQ ID NO: 56 or 57.

[0106] In some embodiments, the second scFv domain that is capable of binding a TAA on a tumor cell comprises a second LinkerscFv(LinkerscFv2) and has the domain structure, from N- to C- terminus, of VH-LinkerscFv2-VL.

[0107] In various embodiments, the LinkerscFvland the LinkerscFv2of an scFv2antibody construct herein can each independently comprise or consist of the amino acid sequence of (G4S)X, wherein x is 1, 2, 3, 4 or 5 (SEQ ID NO: 63). In some embodiments, x is 3 or 4.

[0108] In certain embodiments, the present disclosure relates to a multivalent and multispecific antibody construct, comprising: (i) a first Fab domain capable of binding CD3 on a cytotoxic effector cell; (ii) a first scFv domain capable of binding CD28 on a cytotoxic effector cell; (iii) a third binding domain capable of binding a TAA on a tumor cell; and (iv) a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the third binding domain is coupled to the N-terminus of the second Fc polypeptide, and (c) the first scFv domain is coupled to the C- terminus of the light chain of the first Fab domain.

[0109] Such multivalent and multispecific antibody construct can comprise a first Ig heavy chain (Hl) comprising a first Fab heavy chain comprising a first VH sequence and a first CHI sequence and the first Fc polypeptide, a second Ig heavy chain (H2) comprising the second Fc polypeptide, and a first Ig light chain (LI) comprising a first Fab light chain coupled to the first scFv domain, wherein the first Fab heavy chain and the first Fab light chain form the first Fab domain. In some embodiments, Hl further comprises a first linkerFab'Fcthat couples the C-terminus of the first Fab heavy chain to the N-terminus of the first Fc polypeptide.

[0110] In some embodiments, LI of such multivalent and multispecific antibody construct comprises, from N- to C-terminus, the first Fab light chain, comprising a first VL sequence and a first CL sequence, coupled to the first scFv domain, wherein the first scFv domain comprises, fromN- to C-terminus, either (i) a third VH sequence coupled to a third VL sequence or (ii) a third VL sequence coupled to a third VH sequence. In some embodiments, LI further comprises a linkerFab'scFvthat couples the first Fab light chain to the first scFv domain. Such linkerFab'scFvcan comprise or consist of the amino acid sequence set forth in SEQ ID NO: 58, or a sequence having 1, 2, or 3 amino acid modifications relative thereto.[OHl] In some embodiments, the first scFv domain that is capable of binding CD28 further comprises a first LinkerscFv(LinkerscFvl) and has the domain structure, from N- to C-terminus, of VL- LinkerscFvl-VH.

[0112] In various embodiments, such a multivalent and multispecific antibody construct is a Het- Fab construct in which the third binding domain is a second Fab domain.

[0113] In such embodiments, the second Ig heavy chain (H2) further comprises a second Fab heavy chain comprising a second VH sequence and a second CHI sequence, and the antibody construct further comprises a second Ig light chain (L2) comprising a second Fab light chain comprising a second VL sequence and a second CL sequence, and wherein the second Fab heavy chain and the second Fab light chain form the second Fab domain.

[0114] In some embodiments, H2 further comprises a second linkerFab'Fcthat couples the C- terminus of the second Fab heavy chain to the N-terminus of the second Fc polypeptide. The first linkerFab'Fc, the second linkerFab'Fc, or both such linkers of an antibody construct, can be polypeptide linkers independently comprising or consisting of about 5 to about 50 consecutive amino acid residues. In some embodiments, the first linkerFab'Fcand the second linkerFab'Fcindependently comprise or consists of an IgGl, IgG2 or IgG4 hinge region, or a derivative thereof, wherein the derivative has at least about 80%, 85%, 90%, 93%, 95%, 97%, or 99% sequence identity to a wildtype IgGl, IgG2 or IgG4 hinge region.

[0115] In certain embodiments of a multivalent and multispecific Het-Fab antibody construct, the first Fab domain, the second Fab domain, or both Fab domains comprise one or more amino acid substitutions relative to a reference Fab sequence that promotes correct heavy and light chain pairing. Amino acid modifications that can promote heavy and light chain pairing are known in the art and are described in, e.g., International Patent Publication Nos. WO 2014 / 082179, WO 2015 / 181805, and WO 2017 / 059551, which are incorporated herein by reference in their entireties.

[0116] As further described herein, the first Fab domain, the second Fab domain, or both Fab domains, can each independently comprise a kappa or a lambda light chain. In some embodiments,the first Fab domain and the second Fab domain (i) each comprise a kappa light chain, (ii) each comprise a lambda light chain, or (iii) one Fab domain comprises a kappa light chain and the other Fab domain comprises a lambda light chain.

[0117] In various other embodiments, a multivalent and multispecific antibody construct herein is a scFv2construct in which the third binding domain is a second scFv domain comprising a second VH sequence and a second VL sequence.

[0118] In such embodiments, H2 further comprises a linkerscFv'Fcthat couples the second scFv domain to the second Fc polypeptide. Such linkerscFv'Fccan comprise or consist of an IgGl, IgG2 or IgG4 hinge region, or a derivative thereof, wherein the derivative has at least about 80%, 85%, 90% or 95% sequence identity to a wildtype IgGl, IgG2 or IgG4 hinge region.

[0119] In some embodiments, the second scFv domain comprises a second LinkerscFv(LinkerscFv2) and has the domain structure, from N- to C-terminus, of VH-LinkerscFv2-VL.

[0120] In various embodiments, the LinkerscFvland the LinkerscFv2each independently comprise or consist of the amino acid sequence of (G4S)X, wherein x is 1, 2, 3, 4 or 5 (SEQ ID NO: 63).

[0121] In various embodiments of the present disclosure, a multivalent and multi specific antibody construct described herein can comprise or consist of at least four polypeptide chains. Such four chains can include two Ig heavy chains, Hl and H2, and two Ig light chains, LI and L2. In various embodiments, such costimulatory multivalent and multispecific antibody constructs can include Het-Fab constructs comprising at least two Fab domains.

[0122] Hence, in some embodiments, described herein is a multivalent and multispecific antibody construct comprising two immunoglobulin heavy chains, Hl and H2, and two immunoglobulin light chains, LI and L2, wherein: (i) Hl comprises, from N- to C-terminus, a first Fab heavy chain comprising a first VH sequence and a first CHI sequence, coupled to a first Fc polypeptide; (ii) H2 comprises, from N- to C-terminus, a second Fab heavy chain comprising a second VH sequence and a second CHI sequence, coupled to a second Fc polypeptide; (iii) LI comprises, from N- to C-terminus, a first Fab light chain comprising a first VL sequence and a first CL sequence, coupled to a first scFv domain comprising a third VL sequence coupled to a third VH sequence; and (iv) L2 comprises, from N- to C-terminus, a second Fab light chain comprising a second VL sequence and a second CL sequence, wherein: (a) the first Fab heavy chain of Hl and the first Fab light chain of LI form a first Fab domain capable of binding CD3 on a cytotoxic effector cell; (b) the second Fab heavy chain of H2 and the second Fab light chain of L2 form a second Fab domaincapable of binding a TAA on a tumor cell; (c) the first scFv domain is capable of binding CD28 on a cytotoxic effector cell; and (d) the first Fc polypeptide and the second Fc polypeptide form a dimeric Fc domain.

[0123] In some embodiments, Hl further comprises a first linkerFab'Fcthat couples the first Fab heavy chain to the first Fc polypeptide, and H2 further comprises a second linkerFab'Fcthat couples the second Fab heavy chain to the second Fc polypeptide.

[0124] In further embodiments, the first Fab domain, the second Fab domain, or both Fab domains comprise one or more amino acid substitutions relative to a reference Fab sequence that promotes correct heavy and light chain pairing, as further described elsewhere herein. In some embodiments, the first Fab domain and the second Fab domain (i) each comprise a kappa light chain, (ii) each comprise a lambda light chain, or (iii) one Fab domain comprises a kappa light chain and the other Fab domain comprises a lambda light chain.

[0125] In various embodiments of the present disclosure, a multivalent and multi specific antibody construct can comprise or consist of at least three polypeptide chains. Such three chains can include two Ig heavy chains, Hl and H2, and one Ig light chain, LI. In various embodiments, such multivalent and multispecific antibody constructs can include scFv2constructs comprising at least two scFv domains.

[0126] In certain embodiments, described herein is a multivalent and multispecific antibody construct comprising two immunoglobulin heavy chains, Hl and H2, and one immunoglobulin light chain, LI, wherein: (i) Hl comprises, from N- to C-terminus, a first Fab heavy chain comprising a first VH sequence and a first CHI sequence, coupled to a first Fc polypeptide; (ii) LI comprises, from N- to C-terminus, a first Fab light chain comprising a first VL sequence and a first CL sequence, coupled to a first scFv domain comprising a second VL sequence coupled to a second VH sequence; and (iii) H2 comprises, from N- to C-terminus, a second scFv domain comprising a third VL sequence and a third VH sequence, coupled to a second Fc polypeptide; wherein: (a) the first Fab heavy chain of Hl and the first Fab light chain of LI form a first Fab domain capable of binding CD3 on a cytotoxic effector cell; (b) the first scFv domain is capable of binding CD28 on a cytotoxic effector cell; (c) the second scFv domain is capable of binding a TAA on a tumor cell; and (d) the first Fc polypeptide and the second Fc polypeptide form a dimeric Fc domain.

[0127] In some of these embodiments, Hl further comprises a first linkerFab'Fcthat couples the first Fab heavy chain to the first Fc polypeptide, and H2 further comprises a linkerscFv'Fcthat couples the second scFv domain to the second Fc polypeptide.

[0128] In some embodiments, one or more of the first linkerFab'Fc, the second linkerFab'Fc, and the linkerscFv'Fcof the antibody constructs described herein (e.g., as applicable to Het-Fab or scFv2antibody constructs) can independently comprise or consists of an IgGl, IgG2 or IgG4 hinge region, or a derivative thereof, wherein the derivative has at least about 80%, 85%, 90% or 95% sequence identity to a wildtype IgGl, IgG2 or IgG4 hinge region.

[0129] In various embodiments, a multivalent and multispecific antibody construct of the present disclosure can comprise a first light chain, LI, which further comprises a linkerFab'scFvthat couples the first Fab light chain to the first scFv domain.

[0130] As further described herein, a multivalent and multispecific antibody construct of this disclosure (e.g., a Het-Fab or an scFv2antibody construct) can comprise a dimeric Fc domain. In various embodiments, such dimeric Fc domain is a heterodimeric Fc domain in which the amino acid sequences of the first Fc polypeptide and the second Fc polypeptide differ in at least one residue, e.g., have about 99%, 98%, 97%, 95% or about 90% sequence identity to one another.

[0131] In various embodiments of a costimulatory antibody construct herein that comprises an Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, the first Fc polypeptide comprises a first CH2 sequence and a first CH3 sequence and the second Fc polypeptide comprises a second CH2 sequence and a second CH3 sequence.

[0132] In some embodiments, one or more of the constant domain sequences (e.g., one or more of CHI, CH2 and CH3 sequences) of an Ig heavy chain of an antibody construct described herein are IgG, IgA or IgE constant domains. In some embodiments, one or more of the constant domain sequences of an Ig heavy chain of an antibody construct described herein are IgG constant domains. Such one or more IgG constant domains can independently be of the IgGl, IgG2, IgG3 or IgG4 subclass. In some embodiments, one or more constant domains of an antibody construct are IgGl constant domains. In other embodiments, one or more constant domains of an antibody construct are IgG2 constant domains. In yet other embodiments, one or more constant domains of an antibody construct are IgG4 constant domains.

[0133] In some of these embodiments, at least one of the first and second IgG CH2 sequences comprises one or more amino acid substitutions relative to a corresponding wildtype IgG CH2 sequence that reduce or ablate binding of the antibody construct to one or more Fey receptors.

[0134] In further embodiments of herein described antibody constructs, at least one of the first and second IgG CH3 sequences comprises one or more amino acid substitutions relative to a corresponding wildtype IgG CH3 sequence that promotes formation of a heterodimeric Fc domain compared to a formation of a homodimeric Fc domain. As further described herein, several approaches that promote heterodimeric Fc formation are known in the art. In certain embodiments, the CH3 sequence of one of the Fc polypeptides comprises a set of amino acid substitutions selected from the group consisting of: L351Y_F405A_Y407V, T350V_L351Y_F405A_Y407V and T350V_L351Y_S400E_F405A_Y407V, and the CH3 sequence of the other Fc polypeptide comprises a set of amino acid substitutions selected from the group consisting of: T366L K392M T394W, T366L K392L T394W, T350V_T366L_K392L_T394W,T350V_T366L_K392M_T394W and T350V_T366L_N390R_K392M_T394W, and wherein the numbering of amino acid residues in the Fc polypeptides is according to the EU numbering system.

[0135] As described further herein, the multivalent and multispecific antibody constructs of the present disclosure can comprise several different immunoglobulin domains (e.g., one or more of an Fc domain, a Fab domain, an scFv domain, etc.) and further exhibit certain physicochemical and biological properties that may be a result of their specific format, geometry and / or structure in combination with certain binding affinities for their biological targets.B. Domains of an Antibody Construct

[0136] A costimulatory, multivalent and multispecific antibody construct of the present disclosure can comprise one or more antibody domains. In various embodiments, a multivalent and multispecific antibody construct of the present disclosure comprises a plurality of (i.e., two or more) antibody domains. Such plurality of antibody domains can include Ig-derived domains and comprise (i) one or more Fc domain(s), wherein an Fc domain can comprise a first Fc polypeptide and a second Fc polypeptide, and can be either homodimeric or heterodimeric, (ii) one or more Fab domain(s), wherein a Fab domain can comprise a heavy chain polypeptide sequence (also referred to as a Fab heavy chain) comprising a heavy chain variable domain (VH) sequence and a heavy constant domain (CHI) sequence and a light chain polypeptide sequence (also referred to as a Fab light chain) comprising a light variable domain (VL) sequence and light constant domain(CL) sequence, and / or (iii) one or more scFv domains, wherein an scFv domain can comprise an scFv VH sequence coupled to an scFv VL sequence, optionally via a linkerscFvas further described herein. The various domains an antibody construct can comprise are further described herein.

[0137] In some embodiments, a multivalent and multispecific antibody construct of the present disclosure can comprise additional types of domains, which may or may not be an Ig domain or Ig-derived, such as VHH domains, single-chain Fabs (scFabs), single-domain antibody (sdAb) domains, or other amino acid or non-amino acid-based moieties such as fluorophores, other detectable labels, low molecular weight (MW) organic compounds with, e.g., MW <700 Da, etc.

[0138] An Ig structural unit is typically composed of two pairs of polypeptide chains, each pair having one “light” chain (about 25 kilodalton (kD)) and one “heavy” chain (about 50-70 kD). Light chains can be classified as either kappa or lambda. The “class” of an Ig refers to the type of constant domain possessed by its heavy chain. There are five major classes of antibodies: IgA, IgD, IgE, IgG and IgM, and several of these can be further divided into subclasses (isotypes), for example, IgGl, IgG2, IgG3, IgG4, IgAl and IgA2. The heavy chain constant domains that correspond to the different classes of immunoglobulins (Igs) are called alpha (a), delta (8), epsilon (a), gamma (y) and mu (p), respectively.

[0139] In various embodiments, a multivalent and multispecific antibody construct described herein is based on an IgG class immunoglobulin, for example, an IgGl, IgG2, IgG3 or IgG4 immunoglobulin. In some embodiments, an antibody construct described herein is based on an IgGl, IgG2 or IgG4 immunoglobulin. In certain embodiments, an antibody construct described herein is based on an IgGl immunoglobulin. In other embodiments, an antibody construct described herein is based on an IgG2 immunoglobulin. In yet other embodiments, an antibody construct described herein is based on an IgG4 immunoglobulin. In the context of the present disclosure, when an antibody construct is based on a specified Ig isotype, it refers to an antibody construct that comprises either all or a portion of the constant region (i.e., Fc domain and / or CHI domain) of the specified Ig isotype. It is to be understood that an antibody construct can also comprise hybrids of isotypes and / or subclasses, according to certain embodiments of this disclosure.

[0140] Generally, in antibodies, the N-terminal domain of each polypeptide chain (e.g., heavy and light chain) usually defines a variable region (e.g., VH or VL) of about 100 to 110 or more amino acids in length that is primarily responsible for antigen recognition. The terms variable light chain(VL) and variable heavy chain (VH) refer to these domains in the light and heavy chain, respectively. As described herein, in various embodiments, a multivalent and multispecific antibody construct of the present disclosure can comprise two or more variable domain sequences as part of its antigen binding domains. In various embodiments, a multivalent and multi specific antibody construct of the present disclosure comprises two variable domain sequence per binding domain (e.g., per Fab and / or scFv), and thus by being trivalent (i.e., containing three binding domains) and trispecific (i.e., each of the three binding domains targets a different antigen) can comprise a total of at least six variable domain sequences, e.g., three VH domain sequences and three VL domain sequences.

[0141] In some embodiments, two or more of such variable domains are coupled to one another in tandem and in a single polypeptide chain format, e.g., as described for scFv-type binding domains which contain (from either N- or C-terminus) a VH domain coupled to a VL domain, or as described for constructs in which an scFv domain (which contains two variable domain sequences coupled to one another in tandem) is coupled N-terminally to a Fab domain via a Fab variable domain sequence, e.g., either the Fab VH or the Fab VL sequence, depending on whether the scFv domain is coupled to the Fab heavy or light chain, respectively. In some embodiments, such construct can comprise the heavy chain domain structure, from N- to C-terminus: [(VH-VL) or (VL-VH)]scFv-(VH-CHi)Fab-Fc. The one or more variable domains can be coupled to each other in tandem either with or without one or more linkers, as further described herein, e.g., a linkerscFv, linkerscFv'Fab, etc.

[0142] Accordingly, an antibody construct of the present disclosure that is derived from an Ig molecule (i.e., an antibody construct that comprises at least one Ig domain, wherein the one Ig domain can comprise a wildtype amino acid sequence and / or an amino acid sequence that contains one or more amino acid modifications relative to a wildtype sequence) can comprise different Ig domains within its heavy and light chain(s). Heavy chain domains can include the Fc domain (or Fc region), e.g., comprising a CH2 domain and CH3 domain, a hinge domain (or hinge region), a heavy chain Fab domain comprising a variable heavy domain (VH) and a constant heavy domain (CHI), and light chain domains can include the variable light domain (VL) and the light constant domain (CL). In some embodiments, and as further described herein, a light chain of an antibody construct can further comprise an scFv domain coupled to, e.g., C-terminus of the CL constant domain of the light chain. In some embodiments, and according to certain nomenclatures, the “Fcdomain” or “Fc region” can include the CH2 and CH3 domains as well as a hinge domain (or hinge region).

[0143] In some embodiments, an antibody construct described herein can further comprise another domain or moiety that may not be derived from an Ig molecule. Such a non-Ig domain can be referred to as a moiety. Such moiety can be a detectable label (e.g., a radiolabel or fluorescent label), a low molecular weight (e.g., <750 Da) drug molecule, another peptide (e.g., signal peptide(s)) or polypeptide molecule, or combinations thereof.B.l Complementarity Determining Regions (CDRs) and Binding Domains

[0144] In each of the VH and VL domains of an antibody construct herein are three loops which are hypervariable in sequence and form an antigen-binding site. Each of these loops is referred to as a “hypervariable region” or “HVR,” or “complementarity determining region” or “CDR.” The terms hypervariable region (HVR) and complementarity determining region (CDR) are used herein interchangeably in reference to the portions of the variable domain (e.g., VH or VL) that form the antigen-binding site. With the exception of CDR1 in VH, CDRS generally comprise the amino acid residues that form the hypervariable loops. The VH and VL domains consist of relatively invariant stretches called framework regions (FRs) of between about 15 to 30 amino acids in length separated by the shorter CDRs, which are each typically between about 5 and 15 amino acids in length, although can occasionally be longer or shorter. The three CDRs and four FRs that make up each VH and VL domain are generally arranged from N- to C-terminus as follows: FR1-CDR1- FR2-CDR2-FR3-CDR3-FR4.

[0145] Several different definitions and numbering conventions of the CDR regions in Ig molecules are in common use, including those described by Kabat et al. (1983, Sequences of Proteins of Immunological Interest, NIH Publication No. 369-847, Bethesda, MD), by Chothia et al. (1987, J Mol Biol, 196:901-917), as well as the IMGT, AbM and Contact definitions. These different definitions include overlapping or subsets of amino acid residues when compared against each other. By way of example, CDR definitions according to Kabat, Chothia, IMGT, AbM and Contact are provided in TABLE 1 below.

[0146] Accordingly, as can be readily apparent to one skilled in the art, the exact numbering and placement of CDRs can differ based on the numbering system employed. However, it is to be understood that the disclosure herein of a variable heavy domain (VH) includes the disclosure of the associated (inherent) heavy chain CDRs (HCDRs or HCDR1 -3) as defined by any of the knownnumbering systems. Similarly, disclosure herein of a variable light domain (VL) includes the disclosure of the associated (inherent) light chain CDRs (LCDRs or LCDR1-3) as defined by any of the known numbering systems. In case a set of CDR sequences (e.g., the 3 HCDRs and the 3 LCDRs) for a binding domain against a certain target is given or claimed using a specific numbering system (e.g., Kabat or IMGT), it is to be understood that such claim also encompasses the corresponding sets of CDR sequences as determined using any other of the known numbering systems.

[0147] One skilled in the art can appreciate that a limited number of amino acid substitutions can be introduced into the CDR sequences or to the VH or VL sequences of known antibodies without the antibody losing its ability to bind its target, e.g., a reduction in binding affinity of at least about 10-fold, 100-fold, or 1000-fold or more. Candidate amino acid substitutions can be identified by computer modeling or by techniques such as alanine scanning, with the resulting variants being tested for binding activity (e.g., expressed as binding affinity, e.g., given as the measured KD value) by standard techniques. As an example, in certain embodiments, the CD3 binding domain(s) of antibody constructs described herein can comprise a set of CDRs (i.e., heavy chain CDR1, CDR2 and CDR3, and light chain CDR1, CDR2 and CDR3) that have 90% or greater, 95% or greater, 98% or greater, 99% or greater, or 100% sequence identity to the amino acid sequences set forth in, e.g., SEQ ID NOs: 134-139, respectively, wherein the binding domain retains or substantially retains the ability to bind CD3. In this context, the term “substantially” refers to a change in binding affinity of less than about 5%, less than about 4%, less than about 3%, less than about 2%, or less than about 1%. However, in certain other embodiments, one or skill in the art can also appreciate that introduction of one or more amino acid modifications into a set of six CDRs for a particular binding domain can result in substantial changes in one or more of the following property parameters: antigen binding affinity, stability of the binding domain, and expression or production yields.TABLE 1: Common CDR Definitions11Either the Kabat or Chothia numbering system can be used for HCDR2, HCDR3 and the light chain CDRs for all definitions except Contact, which uses Chothia numbering.2Using Kabat numbering. The position in the Kabat numbering scheme that demarcates the end of the Chothia and IMGT CDR-H1 loop varies depending on the length of the loop due to the placement of insertions outside of those CDR definitions at positions 35A and 35B in Kabat. The IMGT and Chothia CDR-H1 loop can be unambiguously defined using Chothia numbering. CDR-H1 definitions using Chothia numbering are: Kabat H31-H35, Chothia H26- H32, AbM H26-H35, IMGT H26-H33, Contact H30-H35.

[0148] In some embodiments, the antibody constructs described herein comprise at least one Ig domain from a mammalian Ig, such as a bovine Ig, a human Ig, a camelid Ig, a rat Ig, or a mouse Ig. In some embodiments, an antibody construct herein can be a chimeric construct comprising two or more Ig domains, in which at least one domain is from a first mammalian Ig, for example a human Ig, and at least a second domain is from a second mammalian Ig, for example, a mouse or rat Ig. In other embodiments, an antibody construct can be derived from Igs that are from different species, for example, an antibody construct can be chimeric or humanized. A “chimeric antibody construct” refers to an antibody that typically comprises at least one variable domain from a rodent antibody (usually a murine antibody) and at least one constant domain from a human antibody. A “humanized antibody construct” is a type of chimeric antibody that contains minimal sequence derived from a non-human antibody. In some embodiments, an antibody construct herein can comprise at least one Ig constant domain from a human Ig. In various embodiments, all domains of an antibody construct described herein can be (or be derived from) from a human Ig.

[0149] In some embodiments, and as further described herein, modifications (e.g., to the amino acid sequence) to one or more domains of an antibody construct can be made to further refine the properties and performance of an antibody construct (e.g., antigen affinity, stability, and / or pharmacokinetics, etc.). For example, framework region (FR) residues of a human Ig can be replaced by corresponding non-human residues, or the humanized antibodies can comprise residues that are not found in either the recipient antibody or the donor antibody. In general, a variable domain in a humanized antibody or a humanized antibody domain comprises all or substantially all of the hypervariable regions from a non-human Ig and all or substantially all of the FRs from a human Ig sequence. As further described herein, modifications in the Fc domain can enable preferential pairing of the Fc polypeptides to form a heterodimeric Fc domain ratherthan a homodimeric Fc domain, and / or alter interaction of the Fc domain with one or more Fey receptors.

[0150] In some embodiments, the present disclosure relates to antibody constructs that can have certain valencies, e.g., can be multivalent, such as bivalent, trivalent, tetraval ent, and so on. Hence, in various embodiments, an antibody construct herein comprises two, three, or four antigen binding domains, i.e., is at least bivalent, at least trivalent, or at least tetravalent. In certain embodiments of this disclosure, an antibody construct can be at least trispecific and trivalent, and thus such antibody construct can comprise at least three binding domains. Each of these three binding domains has a unique binding specificity for a certain epitope (located on either the same antigen and / or on different antigens). In some of these embodiments, a trispecific and trivalent antibody construct of the present disclosure comprises three binding domains, e.g., one or more Fab domain(s) and / or one or more scFv domain(s), each capable of binding an epitope on one of three different antigens (e.g., CD3, CD28, and a TAA).

[0151] In some embodiments, an antibody construct of the present disclosure can comprise (i) one or more Fab domain(s), (ii) one or more scFv domain(s), and (iii) an Fc domain comprising a first Fc polypeptide and a second Fc polypeptide.

[0152] In various embodiments, a multivalent and multispecific antibody construct as described herein comprises (i) a first Fab domain capable of binding to a first antigen, (ii) a first scFv domain capable of binding to a second antigen, and (iii) a second Fab domain capable of binding to a third antigen, and (iv) a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide.

[0153] In other embodiments, a multivalent and multispecific antibody construct as described herein comprises (i) a Fab domain capable of binding to a first antigen, (ii) a first scFv domain capable of binding to a second antigen, and (iii) a second scFv domain capable of binding to a third antigen, and (iv) a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide.

[0154] In certain embodiments, a multivalent and multispecific antibody construct described herein comprises a fourth binding domain. Such fourth binding domain can bind to an epitope that is either identical or different to the epitopes the other three binding domains bind to.

[0155] Generally, a “Fab domain,” as used herein (also commonly known as an abbreviation of “fragment, antigen binding”), comprises a constant region comprising the constant domain (CL) ofthe light chain and the first constant domain (CHI) of the heavy chain, and a variable region comprising the variable domains VL and VH on the light and heavy chains, respectively, which comprise the CDRs as described herein. In some embodiments, a Fab domain can be a single chain Fab. A single chain Fab can be a Fab molecule in which the Fab light chain and the Fab heavy constant chain are connected by a peptide linker to form a single polypeptide chain. In such embodiments, typically, the C-terminus of the Fab light chain is connected to the N-terminus of the Fab heavy chain in the single-chain Fab molecule, however, other formats are also encompassed herein. In various embodiments herein, however, a Fab domain of an antibody construct is formed by two separately expressed polypeptide chains, i.e., a Fab light chain and a Fab heavy chain. The heavy chain and light chain parts of the Fab domain can, however, be interconnected by covalent bonds, such as disulfide bonds.

[0156] An “scFv domain,” as used herein(also commonly known as an abbreviation of “singlechain fragment, variable”), generally comprises a heavy chain variable domain (VH) and a light chain variable domain (VL) in a single polypeptide chain format. The scFv can optionally comprise a peptide linker between the VH and VL domains which can assist the scFv in forming a functional structure for antigen binding. Hence, in various embodiments, an scFv domain herein can include a VL domain that is coupled via its C-terminus to the N-terminus of a VH domain by a linkerscFv, i.e., an scFv domain can have the domain structure: VL-linkerscFv-VH, or alternatively, an scFv can comprise a VH connected by its C-terminus to the N-terminus of a VL by a linkerscFv, i.e., having the domain structure: Vu-linkerscFv-VL.B.2 Binding Domains against Antigens on a Cytotoxic Effector Cell

[0157] As further described herein, an antibody construct of the present disclosure can comprise at least two binding domains capable of binding to one or more molecule(s), e.g., a polypeptide(s), on the surface of a cytotoxic effector cell. Such cytotoxic effector cell can be an immune cell. Such immune cell can comprise a T cell, a macrophage, a dendritic cell, a neutrophil, a B-cell, an NK cell, or a combination thereof.

[0158] In various embodiments, an antibody construct of the present disclosure comprises at least one binding domain capable of binding a first antigen on a cytotoxic effector cell, and at least one binding domain capable of binding a second antigen on a cytotoxic effector cell. In various embodiments, such first and second antigens are different from each other and are located on the same cytotoxic effector cell, e.g., T cell.

[0159] In some embodiments of a multivalent and multispecific T cell-engaging antibody construct herein, the first antigen on the cytotoxic effector cell is CD3, and the second antigen on the cytotoxic effector cell is CD28. In some of these embodiments, both CD3 and CD28 can be engaged by a multivalent and multispecific antibody construct when located on the same cytotoxic effector cell surface. In other embodiments, both antigens CD3 and CD28 can be engaged by a multivalent and multispecific antibody construct when located on the surfaces of different cytotoxic effector cells.

[0160] The binding of an antibody construct of the present disclosure to two different antigens located on one cytotoxic effector cell can also be referred to as “cis” -binding. In other embodiments, the cytotoxic effector cell is more than one cell, i.e., the first and the second antigens that an antibody constructs of this disclosure can bind to are located on the surfaces of different cells. The binding of an antibody construct to two different antigens located on two different cytotoxic effector cells can also be referred to as “trans” -binding or cross-linking of effector cells (e.g., T cells).

[0161] In various embodiments, a multivalent and multispecific antibody construct of the present disclosure that binds CD3 and CD28 and that comprises a light chain, L, comprising an anti-CD3 Fab light chain C-terminally coupled to an anti-CD28 scFv domain, binds CD3 and CD28 in “cis,” i.e., on the same effector cell (e.g., T cell).

[0162] In some embodiments, the present disclosure relates to multivalent and multispecific antibody constructs comprising at least a first binding domain capable of binding CD3 on a cytotoxic effector cell, a second binding domain capable of binding CD28 on a cytotoxic effector cell, and a third binding domain capable of binding a target antigen, such as a TAA on a tumor cell, wherein the antibody construct binds CD3 and CD28 on the same cytotoxic effector cell. Hence, the antibody constructs described herein can also be referred to as “T cell engagers,” “TCEs” or “T cell engager molecules,” describing the ability of a construct to bind both, antigens on one or more effector cell(s) (e.g., T cell(s)) as well as a TAA on a tumor cell. In some embodiments, the engagement of a multivalent and multispecific antibody construct with two different effector cell antigens on an effector cell and a TAA on a tumor cell can be - at least temporarily - simultaneous, and thereby establish a TCR-independent immune synapse, and direct T cell-mediated cytotoxic activity to a tumor environment which contains tumor cells expressing the TAA. In various embodiments, and as further described herein, a multivalent and multispecificantibody construct may cause a significantly reduced immune cell (e.g., T cell) activation in the absence of the TAA, e.g., when the immune synapse cannot be fully formed due to an absence of the TAA, and relative to immune cell activation in the presence of the TAA. Such TAA-dependent immune cell activation can provide for a broader therapeutic window compared to conventional T cell engager modalities.

[0163] In various embodiments, the first antigen binding domain capable of binding CD3 on a cytotoxic effector cell can be a Fab domain or an scFv domain. In these embodiments, the second antigen binding domain capable of binding CD28 on a cytotoxic effector cell can also be a Fab domain or an scFv domain.

[0164] In various embodiments of a multivalent and multispecific T cell-engaging antibody construct herein, the first binding domain capable of binding CD3 on a cytotoxic effector cell is a Fab domain, and the second binding domain capable of binding CD28 on a cytotoxic effector cell is an scFv domain, wherein CD3 and CD28 is engaged by the construct on the same effector cell.

[0165] In some embodiments, the Fab domain that is capable of binding CD3 can comprise a heavy chain constant domain (CHI) comprising or consisting of an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 3 or SEQ ID NO: 4. In some embodiments, such CHI domain sequence comprises or consists of an amino acid sequence having at least about 90% sequence identity to the amino acid sequence set forth in SEQ ID NO: 3 or SEQ ID NO: 4. In some embodiments, such CHI domain sequence comprises or consists of an amino acid sequence having at least about 95% sequence identity to the amino acid sequence set forth in SEQ ID NO: 3 or SEQ ID NO: 4. In some embodiments, such CHI domain sequence comprises or consists of an amino acid sequence having at least about 97% sequence identity to the amino acid sequence set forth in SEQ ID NO: 3 or SEQ ID NO: 4. In some embodiments, such CHI domain sequence comprises or consists of an amino acid sequence having at least about 99% sequence identity to the amino acid sequence set forth in SEQ ID NO: 3 or SEQ ID NO: 4. In some embodiments, such CHI domain sequence comprises or consists of the amino acid sequence set forth in SEQ ID NO: 3 or SEQ ID NO: 4.

[0166] In some embodiments, the Fab domain that is capable of binding CD3 can comprise a light chain constant domain (CL) comprising or consisting of an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 6 or SEQ ID NO: 7. In some embodiments, such CL domain sequence comprises orconsists of an amino acid sequence having at least about 90% sequence identity to the amino acid sequence set forth in SEQ ID NO: 6 or SEQ ID NO: 7. In some embodiments, such CL domain sequence comprises or consists of an amino acid sequence having at least about 95% sequence identity to the amino acid sequence set forth in SEQ ID NO: 6 or SEQ ID NO: 7. In some embodiments, such CL domain sequence comprises or consists of an amino acid sequence having at least about 97% sequence identity to the amino acid sequence set forth in SEQ ID NO: 6 or SEQ ID NO: 7. In some embodiments, such CL domain sequence comprises or consists of an amino acid sequence having at least about 99% sequence identity to the amino acid sequence set forth in SEQ ID NO: 6 or SEQ ID NO: 7. In some embodiments, such CL domain sequence comprises or consists of the amino acid sequence set forth in SEQ ID NO: 6 or SEQ ID NO: 7.

[0167] In various embodiments, the cytotoxic effector cell that expresses CD3 and CD28 which the antibody construct engages is a T cell.B.2.1 Binding Domains Against CD 3

[0168] As described herein, in some embodiments, a multivalent and multispecific antibody construct of the present disclosure comprises a binding domain capable of binding cluster of differentiation 3 (CD3, e.g., CD3s) on a cytotoxic effector cell (e.g., T cell). Such multivalent and multispecific antibody construct can bind one or more of human, cynomolgus (cyno) and murine CD3.

[0169] In various embodiments, an anti-CD3 binding domain (e.g., an scFv domain or a Fab domain) of an antibody construct herein can have an affinity for CD3 (e.g., given as a KD value for binding CD3) that is not more than about 1 nM, 5 nM, 10 nM, 20 nM, or not more than about 30 nM. Thus, in various embodiments, an anti-CD3 binding domain of an antibody construct herein has a KD value for binding CD3 that is from about 20 nM to about 200 nM, from about 30 nM to about 150 nM, from about 40 nM to about 100 nM, or from 50 nM to about 80 nM, from about 20 nM to about 80 nM, from about 30 nM to about 60 nM, or from about 40 nM to about 50 nM. In some embodiments, an anti-CD3 binding domain of an antibody construct herein has a KD value for binding CD3 that is not more than about 30 nM, 40 nM, 50 nM, or about 60 nM. In various embodiments, an antibody construct herein comprises an anti-CD3 binding domain that has a KD value for binding CD3 from about 20 nM to about 40 nM, e.g., of about 30 nM.

[0170] In various embodiments, the anti-CD3 binding domain of an antibody construct herein is a Fab domain.

[0171] In certain embodiments, the anti-CD3 binding domain of an antibody construct has an affinity for CD3 that is about 2-fold, 3-fold, 5-fold, 7-fold, 10-fold, or about 15-fold higher than the binding affinity of the anti-CD28 domain for CD28. In some of these embodiments, the difference in affinities of the anti-CD3 and anti-CD28 binding domains for their respective targets, and, in some embodiments, in combination with their structure and relative spatial orientation, can provide for conditional binding of CD28 by the construct that is dependent on engaging CD3, i.e., there is no or significantly reduced CD28 binding of the construct in the absence of CD3 engagement.

[0172] In some embodiments, an antibody construct herein comprising an anti-CD3 binding domain, an anti-CD28 binding domain, and an anti-TAA binding domain, binds CD3 with a KD from about 0.1 nM to about 100 nM, from about 1 nM to about 100 nM, or from about 10 nM to about 75 nM.

[0173] In various embodiments, an antibody construct herein comprises a binding domain capable of binding CD3 on a T cell, wherein such binding domain has a KD value for binding CD3 from about 20 nM to about 40 nM, e.g., of about 30 nM, and comprises the CDR sequences of the VH sequence as set forth in SEQ ID NOS: 49, 50 and 52 or SEQ ID NOS: 49, 51 and 52, and the CDR sequences of the VL sequence as set forth in SEQ ID NO: 53-55.

[0174] In various embodiments, an antibody construct herein comprises a binding domain capable of binding CD3 on a T cell, wherein such binding domain has a KD value for binding CD3 from about 20 nM to about 40 nM, e.g., of about 30 nM, and comprises a VH domain comprising a HCDR1 sequence comprising the sequence GVTFNYYG (SEQ ID NO: 49), a HCDR2 sequence comprising the sequence ITSSGGRI (SEQ ID NO: 50) or ITRSGGRI (SEQ ID NO: 51), and a HCDR3 sequence comprising the sequence TLDGRDGWVAY (SEQ ID NO: 52), and a VL domain comprising a LCDR1 sequence comprising the sequence TGNIGSNY (SEQ ID NO: 53), a LCDR2 sequence comprising the sequence RND, and a LCDR3 sequence comprising the sequence QSYSSGFI (SEQ ID NO: 55).

[0175] In some embodiments, the anti-CD3 binding domain comprises a VH domain comprising or consisting of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 1 or SEQ ID NO: 2, and a VL domain comprising or consisting of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 5. In certainembodiments, the anti-CD3 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 1 or SEQ ID NO: 2, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 5.

[0176] In some embodiments, the CDRs of an anti-CD3 paratope used in an antibody construct of the present disclosure comprise one or more amino acid modifications in one or more of the CDR sequences set forth in SEQ ID NOs: 49, 50 and 52-55 or SEQ ID NOS: 49 and 51-55, wherein at least about 80%, 90%, or 95% binding affinity to CD3 is retained compared to the paratope without such amino acid modifications.

[0177] In various embodiments of a multivalent and multispecific antibody construct, the anti- CD3 binding domain is a Fab domain as described herein.

[0178] The VH domain of an anti-CD3 Fab domain of an antibody construct herein can comprise or consist of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 1 or SEQ ID NO: 2. In some embodiments, the VH domain of such Fab domain comprises or consists of an amino acid sequence having at least about 90% sequence identity to the amino acid sequence set forth in SEQ ID NO: 1 or SEQ ID NO: 2. In some embodiments, the VH domain of such Fab domain comprises or consists of an amino acid sequence having at least about 95% sequence identity to the amino acid sequence set forth in SEQ ID NO: 1 or SEQ ID NO: 2. In some embodiments, the VH domain of such Fab domain comprises or consists of an amino acid sequence having at least about 97% sequence identity to the amino acid sequence set forth in SEQ ID NO: 1 or SEQ ID NO: 2. In some embodiments, the VH domain of such Fab domain comprises or consists of an amino acid sequence having at least about 99% sequence identity to the amino acid sequence set forth in SEQ ID NO: 1 or SEQ ID NO: 2. In yet other embodiments, the VH domain of such Fab domain comprises or consists of the amino acid sequence set forth in SEQ ID NO: 1 or SEQ ID NO: 2.

[0179] The VL domain of an anti-CD3 Fab domain of an antibody construct herein, which can be part of a light chain that pairs with an anti-CD3 Fab domain sequence of a heavy chain to form the Fab domain, can comprise or consist of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 5. In some embodiments, the VL domain of such scFv or Fab domain comprises or consists of an amino acid sequence having at least about 90% sequence identity to the amino acid sequence setforth in SEQ ID NO: 5. In some embodiments, the VL domain of such scFv or Fab domain comprises or consists of an amino acid sequence having at least about 95% sequence identity to the amino acid sequence set forth in SEQ ID NO: 5. In some embodiments, the VL domain of such scFv or Fab domain comprises or consists of an amino acid sequence having at least about 97% sequence identity to the amino acid sequence set forth in SEQ ID NO: 5. In some embodiments, the VL domain of such scFv or Fab domain comprises or consists of an amino acid sequence having at least about 99% sequence identity to the amino acid sequence set forth in SEQ ID NO: 5. In various embodiments, the VL domain of such scFv or Fab domain comprises or consists of the amino acid sequence set forth in SEQ ID NO: 5.

[0180] In embodiments in which the anti-CD3 domain is a Fab domain, which further comprises a CHI domain and a CL domain in its heavy and light chains, respectively, the CHI domain of an anti-CD3 Fab domain of an antibody construct herein can comprise or consist of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 3 or SEQ ID NO: 4. In some embodiments, the CHI domain of such Fab domain comprises or consists of an amino acid sequence having at least about 90% sequence identity to the amino acid sequence set forth in SEQ ID NO: 3 or SEQ ID NO: 4. In some embodiments, the CHI domain of such Fab domain comprises or consists of an amino acid sequence having at least about 95% sequence identity to the amino acid sequence set forth in SEQ ID NO: 3 or SEQ ID NO: 4. In some embodiments, the CHI domain of such Fab domain comprises or consists of an amino acid sequence having at least about 97% sequence identity to the amino acid sequence set forth in SEQ ID NO: 3 or SEQ ID NO: 4. In some embodiments, the CHI domain of such Fab domain comprises or consists of an amino acid sequence having at least about 99% sequence identity to the amino acid sequence set forth in SEQ ID NO: 3 or SEQ ID NO: 4. In some embodiments, the CHI domain of such Fab domain comprises or consists of the amino acid sequence set forth in SEQ ID NO: 3 or SEQ ID NO: 4.

[0181] Moreover, the CL domain of an anti-CD3 Fab domain of an antibody construct herein can comprise or consist of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 6 or SEQ ID NO: 7. In some embodiments, the CL domain of such Fab domain comprises or consists of an amino acid sequence having at least about 90% sequence identity to the amino acid sequence set forth in SEQ ID NO: 6 or SEQ ID NO: 7. In some embodiments, the CL domain of such Fab domaincomprises or consists of an amino acid sequence having at least about 95% sequence identity to the amino acid sequence set forth in SEQ ID NO: 6 or SEQ ID NO: 7. In some embodiments, the CL domain of such Fab domain comprises or consists of an amino acid sequence having at least about 97% sequence identity to the amino acid sequence set forth in SEQ ID NO: 6 or SEQ ID NO: 7. In some embodiments, the CL domain of such Fab domain comprises or consists of an amino acid sequence having at least about 99% sequence identity to the amino acid sequence set forth in SEQ ID NO: 6 or SEQ ID NO: 7. In yet other embodiments, the CL domain of such Fab domain comprises or consists of the amino acid sequence set forth in SEQ ID NO: 6 or SEQ ID NO: 7.

[0182] In some embodiments, an anti-CD3 binding domain of an antibody construct is a Fab domain. In some embodiments, such an anti-CD3 Fab domain comprises a heavy chain, or a portion thereof (e.g., in cases in which the heavy chain further comprises an Fc portion, etc.), and a light chain. The heavy chain of the anti-CD3 Fab domain can comprise or consist of a VH domain coupled to a CHI domain, from N- to C-terminus. In various embodiments, the anti-CD3 Fab domain comprises (i) a VH domain comprising or consisting of an amino acid sequence having at least about 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 1 or SEQ ID NO: 2, (ii) a VL domain comprising or consisting of an amino acid sequence having at least about 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 5, (iii) a CHI domain comprising or consisting of an amino acid sequence having at least about 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 3 or SEQ ID NO: 4, and (iv) CL domain comprising or consisting of an amino acid sequence having at least about 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 6 or SEQ ID NO: 7.

[0183] In certain embodiments, the anti-CD3 Fab domain comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 1 or SEQ ID NO: 2, a CHI domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 3, a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 5, and a CL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 6.

[0184] In certain embodiments, the anti-CD3 Fab domain comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 1 or SEQ ID NO: 2, a CHI domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 4, a VL domaincomprising or consisting of the amino acid sequence set forth in SEQ ID NO: 5, and a CL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 7.

[0185] In certain embodiments, the anti-CD3 Fab domain comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 1, a CHI domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 4, a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 5, and a CL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 7.

[0186] In certain embodiments, the anti-CD3 Fab domain comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 1, a CHI domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 3, a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 5, and a CL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 7.

[0187] In certain embodiments, the anti-CD3 Fab domain comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 1, a CHI domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 3, a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 5, and a CL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 6.

[0188] In certain embodiments, the anti-CD3 Fab domain comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 2, a CHI domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 3, a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 5, and a CL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 7.

[0189] In certain embodiments, the anti-CD3 Fab domain comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 2, a CHI domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 3, a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 5, and a CL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 7.

[0190] In certain embodiments, the anti-CD3 Fab domain comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 2, a CHI domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 3, a VL domain comprising orconsisting of the amino acid sequence set forth in SEQ ID NO: 5, and a CL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 6.

[0191] As further described herein, in some embodiments, the binding affinity of a multivalent and multispecific antibody construct of the present disclosure for one or more specific targets (e.g., CD3) can - at least in part - be dependent on its format and / or geometry. As an example, the relative orientation and vicinity of an anti-CD3 binding domain in an antibody construct can affect the binding domain’s ability to interact with the target epitope it is binding, e.g., through steric hindrance, conformational changes occurring when the construct interacts with one or more of its targets (e.g., conditional degrees of freedom for a binding domain such as its steric flexibility), etc.

[0192] In further embodiments, the CD3 binding affinity of an anti-CD3 binding domain of an antibody construct herein can be engineered and altered (e.g., increased / decreased relative to unmodified domains), e.g., by using one or more amino acid modifications relative to one or more reference sequence(s). In some embodiments, a tri specific and trivalent antibody construct of the present disclosure can comprise a variant anti-CD3 binding domain that comprises one or more amino acid modifications in its VH and / or VL domain(s) compared to the anti-CD3 binding domains described herein which comprise a VH sequence set forth in, e.g., SEQ ID NOs: 1 or 2 and a VL sequence set forth in, e.g., SEQ ID NO: 5. Such one or more amino acid modifications can reduce or increase the binding affinity of the variant anti-CD3 binding domain to CD3 when compared to the binding affinity of a corresponding anti-CD3 binding domain that does not comprise such one or more amino acid modifications.

[0193] In some embodiments, the one or more amino acid modifications used to alter the binding affinity of an anti-CD3 binding domain can include one or more amino acid substitution(s), one or more amino acid addition(s), and / or one or more amino acid deletion(s). In certain embodiments, the one or more amino acid modifications used to alter the binding affinity of an anti-CD3 binding domain comprise or consist of one or more amino acid substitution(s) relative to an unmodified binding domain sequence (e.g., an anti-CD3 VH or VL sequence).

[0194] In some embodiments, the anti-CD3 affinity of an affinity-altered CD3 binding domain can be about ±2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, or 10-fold higher or lower than a corresponding parental anti-CD3 binding domain.

[0195] In some embodiments, other anti-CD3 binding domains, such as those known in the art (e.g., SP34, OKT3, etc.), as well as variants thereof, can also be used in the multivalent and multispecific antibody constructs described herein.

[0196] In some embodiments, an antibody construct herein comprises an anti-CD3 Fab domain comprising a VH sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 82 and a VL sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 83. In certain embodiments, an antibody construct herein comprises an anti-CD3 Fab domain comprising a VH sequence having at least about 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 82 and a VL sequence having at least about 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 83. In certain embodiments, an antibody construct herein comprises an anti-CD3 Fab domain comprising a VH sequence having at least about 99% or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 82 and a VL sequence having at least about 99% or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 83.

[0197] In some embodiments, an antibody construct herein comprises an anti-CD3 Fab domain comprising a VH sequence comprising the HCDR1-3 sequence set forth in SEQ ID NOs: 134-136, respectively, and a VL sequence comprising the LCDR1-3 sequence set forth in SEQ ID NOs: 137-139, respectively.

[0198] In some embodiments, an antibody construct herein comprises an anti-CD3 Fab domain comprising a VH sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 84 and a VL sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 85. In certain embodiments, an antibody construct herein comprises an anti-CD3 Fab domain comprising a VH sequence having at least about 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 84 and a VL sequence having at least about 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 85. In certain embodiments, an antibody construct herein comprises an anti-CD3 Fab domain comprising a VH sequence having at least about 99% or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 84 and a VL sequence having at least about 99% or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 85.

[0199] In some embodiments, an antibody construct herein comprises an anti-CD3 Fab domain comprising a VH sequence comprising the HCDR1-3 sequence set forth in SEQ ID NOs: 134, 140 and 141, respectively, and a VL sequence comprising the LCDR1-3 sequence set forth in SEQ ID NOs: 142, 138 and 139, respectively.

[0200] In some embodiments, an antibody construct herein comprises an anti-CD3 Fab domain comprising a VH sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 86 and a VL sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 87. In certain embodiments, an antibody construct herein comprises an anti-CD3 Fab domain comprising a VH sequence having at least about 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 86 and a VL sequence having at least about 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 87. In certain embodiments, an antibody construct herein comprises an anti-CD3 Fab domain comprising a VH sequence having at least about 99% or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 86 and a VL sequence having at least about 99% or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 87.

[0201] In some embodiments, an antibody construct herein comprises an anti-CD3 Fab domain comprising a VH sequence comprising the HCDR1-3 sequence set forth in SEQ ID NOs: 143, 135 and 141, respectively, and a VL sequence comprising the LCDR1-3 sequence set forth in SEQ ID NOs: 137-139, respectively.B.2.2 Binding Domains Against CD28

[0202] As further described herein, in various embodiments, a multivalent and multispecific antibody construct of the present disclosure (e.g., a Het-Fab or scFv2antibody construct) comprises at least one binding domain capable of binding cluster of differentiation 28 (CD28). In some embodiments, such construct binds one or more of human, cyno and murine CD28.

[0203] In various embodiments, such anti-CD28 binding domain (e.g., an scFv domain of a Fab domain) of an antibody construct herein can have an affinity for CD28 (given as a KD value for binding CD28) that is from about 10 nM to about 500 nM, from about 20 nM to about 250 nM, from about 15 nM to about 35 nM or from about 5 nM to about 50 nM.

[0204] In various embodiments of a multivalent and multi specific antibody construct, the anti- CD28 binding domain comprises or consists of an scFv domain.

[0205] In various embodiments, the anti-CD28 binding domain of an antibody construct herein comprises or consists of an scFv domain, and is coupled to the C-terminus of a light chain of an anti-CD3 Fab domain of multivalent and multispecific antibody construct. Such multivalent and multispecific antibody construct can engage CD3 and CD28 in “cis” on the same effector cell, such as a T cell, as further described herein.

[0206] In some embodiments, an antibody construct herein comprises an scFv domain capable of binding CD28 on a T cell, wherein such binding domain has an affinity for CD28 that is from about 15 nM to about 35 nM or from about 5 nM to about 50 nM and comprises an anti-CD28 VH sequence comprising a HCDR1 having the sequence SYGVH (SEQ ID NO: 28), a HCDR2 having the sequence VIWPGGGTNFNSALMS (SEQ ID NO: 31), and a HCDR3 having the sequence DRAYGNYLYAMDY (SEQ ID NO: 35), and an anti-CD28 VL sequence comprising a LCDR1 having the sequence RASES VEYYVTSLMQ (SEQ ID NO: 41), a LCDR2 having the sequence AASNVDS (SEQ ID NO: 44), and a LCDR3 having the sequence QQSRKVPFT (SEQ ID NO: 48).

[0207] In some of these embodiments, an antibody construct herein comprises an scFv domain capable of binding CD28 on a T cell, wherein such binding domain comprises a VH domain comprising or consisting of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 15, and a VL domain comprising or consisting of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 16. In some embodiments, the anti-CD28 binding domain comprises a VH domain comprising or consisting of the sequence set forth in SEQ ID NO: 15, and a VL domain comprising or consisting of the sequence set forth in SEQ ID NO: 16.

[0208] In other embodiments, a multivalent and multispecific antibody construct of the present disclosure comprises an anti-CD28 binding domain (e.g., an scFv domain) comprising (i) a VH domain that comprises a sequence having one or more amino acid substitution(s) compared to the sequence set forth in SEQ ID NO: 15, and / or (ii) a VL domain that comprises a sequence having one or more amino acid substitution(s) compared to the sequence set forth in SEQ ID NO: 16, wherein the position of such amino acid substitution can be provided herein according to the IMGT, Kabat, or another numbering system known to the skilled artisan.

[0209] In some embodiments, such one or more amino acid substitutions in either the anti-CD28 VH domain and / or the anti-CD28 VL domain can reduce the binding affinity of the corresponding anti-CD28 binding domain to CD28 by about 1.5-fold, 1.6-fold, 1.7-fold, 1.8-fold, 2.0-fold, 3.0- fold, 3.3-fold, 3.5-fold, 3.7-fold, 3.9-fold, 5.0-fold, 5.2-fold, 5.5-fold, 6.0-fold, 7.0-fold, 8.0-fold, 8.5-fold, 9.0-fold, 10-fold, 20-fold, or 25-fold, or from about 1.5-fold to about 25-fold, from about 2.0-fold to about 20-fold, from about 3.0-fold to about 20-fold, or from about 5.0-fold to about 10- fold, when compared to the binding affinity of an anti-CD28 binding domain that does not contain a VH and / or VL sequence with such amino acid substitutions (e.g., a binding domain comprising the VH and VL sequence set forth in SEQ ID NOs: 15 and 16, respectively).

[0210] Hence, in some embodiments, an anti-CD28 binding domain (e.g., an scFv domain) of an antibody construct herein can have an affinity for CD28 (given as a KD value for binding CD28) that is from about 10 nM to about 500 nM, from about 20 nM to about 600 nM, from about 20 nM to about 250 nM, from about 20 nM to about 150 nM, from about 20 nM to about 100 nM, or from about 20 nM to about 50 nM.

[0211] In some embodiments, an antibody construct herein binds CD28 with a KD from about 1 nM to about 100 nM or from about 10 nM to about 75 nM.

[0212] In some embodiments, an anti-CD28 binding domain herein comprises a VH domain comprising the amino acid substitution P1058A relative to the amino acid sequence set forth in SEQ ID NO: 15. In some of these embodiments, the anti-CD28 VH domain comprises the amino acid sequence set forth in SEQ ID NO: 21.

[0213] In some embodiments, an anti-CD28 binding domain herein comprises a VH domain comprising the amino acid substitution G1064S relative to the amino acid sequence set forth in SEQ ID NO: 15, and as determined using the IMGT numbering scheme. In some of these embodiments, the anti-CD28 VH domain comprises the amino acid sequence set forth in SEQ ID NO: 24.

[0214] In some embodiments, an anti-CD28 binding domain herein comprises a VL domain comprising the amino acid substitution V1035G relative to the amino acid sequence set forth in SEQ ID NO: 16. In some of these embodiments, the anti-CD28 VL domain comprises the amino acid sequence set forth in SEQ ID NO: 17.

[0215] In some embodiments, an anti-CD28 binding domain herein comprises a VL domain comprising the amino acid substitution D1068E relative to the amino acid sequence set forth inSEQ ID NO: 16. In some of these embodiments, the anti-CD28 VL domain comprises the amino acid sequence set forth in SEQ ID NO: 26.

[0216] In some embodiments, an anti-CD28 binding domain herein comprises a VH domain comprising the amino acid substitution E1080K relative to the amino acid sequence set forth in SEQ ID NO: 15. In some of these embodiments, the anti-CD28 VH domain comprises the amino acid sequence set forth in SEQ ID NO: 20.

[0217] In some embodiments, an anti-CD28 binding domain herein comprises a VH domain comprising the amino acid substitution Y1110S relative to the amino acid sequence set forth in SEQ ID NO: 15. In some of these embodiments, the anti-CD28 VH domain comprises the amino acid sequence set forth in SEQ ID NO: 18.

[0218] In some embodiments, an anti-CD28 binding domain herein comprises a VH domain comprising the amino acid substitution N111 laA relative to the amino acid sequence set forth in SEQ ID NO: 15. In some of these embodiments, the anti-CD28 VH domain comprises the amino acid sequence set forth in SEQ ID NO: 25.

[0219] In some embodiments, an anti-CD28 binding domain herein comprises a VH domain comprising the amino acid substitution Y1112S relative to the amino acid sequence set forth in SEQ ID NO: 15. In some of these embodiments, the anti-CD28 VH domain comprises the amino acid sequence set forth in SEQ ID NO: 76.

[0220] In some embodiments, an anti-CD28 binding domain herein comprises a VH domain comprising the amino acid substitution LI 112aN relative to the amino acid sequence set forth in SEQ ID NO: 15. In some of these embodiments, the anti-CD28 VH domain comprises the amino acid sequence set forth in SEQ ID NO: 23.

[0221] In some embodiments, an anti-CD28 binding domain herein comprises a VH domain comprising the amino acid substitution Y1113S relative to the amino acid sequence set forth in SEQ ID NO: 15. In some of these embodiments, the anti-CD28 VH domain comprises the amino acid sequence set forth in SEQ ID NO: 27.

[0222] In some embodiments, an anti-CD28 binding domain herein comprises a VH domain comprising the amino acid substitution Y1037A relative to the amino acid sequence set forth in SEQ ID NO: 15. In some of these embodiments, the anti-CD28 VH domain comprises the amino acid sequence set forth in SEQ ID NO: 22.

[0223] In some embodiments, an anti-CD28 binding domain herein comprises a VL domain comprising the amino acid substitution Y1031A relative to the amino acid sequence set forth in SEQ ID NO: 16. In some of these embodiments, the anti-CD28 VL domain comprises the amino acid sequence set forth in SEQ ID NO: 77.

[0224] In some embodiments, an anti-CD28 binding domain herein comprises a VL domain comprising the amino acid substitution N1066A relative to the amino acid sequence set forth in SEQ ID NO: 16. In some of these embodiments, the anti-CD28 VL domain comprises the amino acid sequence set forth in SEQ ID NO: 19.

[0225] In some embodiments, an antibody construct comprising an anti-CD28 binding domain comprising one or more of the VH and / or VL domain substitutions described herein can exhibit a reduced non-specific anti-tumor activity in vitro and / or in vivo. In one such embodiment, an antibody construct comprising a mutated (e.g., relative to the huTN228 wildtype sequence) anti- CD28 binding domain can induce less non-specific (e.g., in the absence of the TAA) immune cell activity, e.g., having reduced non-specific cytokine production by the immune cells. In certain embodiments, an antibody construct comprising an anti-CD28 binding domain that carries the N1066A substitution in the VL domain relative to huTN228 wildtype can induce a reduced nonspecific T cell activity, e.g., cytokine production, e.g., in the absence of the TAA. In another embodiment, an antibody construct comprising an anti-CD28 binding domain that carries the Y1031A substitution in the VL domain relative to huTN228 wildtype can induce a reduced nonspecific T cell activity, e.g., cytokine production, e.g., in the absence of the TAA. In some embodiments, non-specific (e.g., in the absence of the TAA) T cell activity, e.g., cytokine production such as production of TNFa, IL-2, etc., can be reduced by about 10-fold, 20-fold, 30- fold, 50-fold, 60-fold, 70-fold, or about 100-fold.

[0226] In various embodiments, an antibody construct herein comprises a binding domain capable of binding CD28 on a T cell, wherein such binding domain has a KD value for binding CD28 from about 20 nM to about 600 nM and comprises a VH domain comprising a HCDR1 having the sequence SXiGVH (SEQ ID NO: 30), a HCDR2 having the sequence VIWX2GGX3TNFNSALMS (SEQ ID NO: 34), and a HCDR3 having the sequence DRAX4GX5YX6X7AMDY (SEQ ID NO: 40) and a VL sequence comprising a LCDR1 having the sequence RASESVEYYXsTSLMQ (SEQ ID NO: 43), a LCDR2 having the sequence AASX9VX10S (SEQ ID NO: 47), and a LCDR3 havingthe sequence QQSRKVPFT (SEQ ID NO: 48), and wherein Xi = Y, A; X2= P, A; X3= G, S; X4= S, Y; X5= N, A; X6= L, N; X7= S, Y; X8= G, V; X9= N, A; and Xio = E, D.

[0227] In some embodiments, an anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO:15, and a VL domain comprising or consisting of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO:16. In some of these embodiments, the anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 15, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 16.

[0228] In some embodiments, the anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 21, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 16.

[0229] In some embodiments, the anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 24, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 16.

[0230] In some embodiments, the anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 15, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 17.

[0231] In some embodiments, the anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 15, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 26.

[0232] In some embodiments, the anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 20, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 16.

[0233] In some embodiments, the anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 18, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 16.

[0234] In some embodiments, the anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 25, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 16.

[0235] In some embodiments, the anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 76, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 16.

[0236] In some embodiments, the anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 23, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 16.

[0237] In some embodiments, the anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 27, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 16.

[0238] In some embodiments, the anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 22, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 16.

[0239] In some embodiments, the anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 15, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 77.

[0240] In some embodiments, the anti-CD28 binding domain of an antibody construct herein comprises a VH domain comprising or consisting of the amino acid sequence set forth in SEQ IDNO: 15, and a VL domain comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 16.

[0241] In various embodiments of a multivalent and multispecific antibody construct described herein, the anti-CD28 binding domain comprises or consists of an scFv domain. As further described herein, such anti-CD28 binding domain can have the domain structure, from N- to C- terminus, of: VL-LinkerscFv-VH or VH-LinkerscFv-VL.

[0242] In embodiments in which the anti-CD28 binding domain of an antibody construct is an scFv domain, such anti-CD28 scFv domain can comprise, from either N- to C-terminus of C- to N-terminus, a VH domain comprising the amino acid sequence set forth in any one of SEQ ID NOs: 15, 18, 20, 21, 22, 23, 24, 25, 27, or 76, coupled to a VL domain comprising the amino acid sequence set forth in any one of SEQ ID NOs: 16, 17, 19, 26, or 77, via a linkerscFvwhich sequence is (GnS)m, wherein n, m can independently be 1, 2, 3, 4 or 5, and as set forth in SEQ ID NO: 69. In some of these instances, the linkerscFvcomprises or consists of the amino acid sequence set forth in SEQ ID NO: 59. And in some embodiments, the anti-CD28 scFv domain has a domain structure, from N- to C-terminus, of: Vu-linkerscFv-VL. In other embodiments, the anti-CD28 scFv domain has a domain structure, from N- to C-terminus, of: VL-linkerscFv-Vu.

[0243] In certain embodiments, described herein is an antibody construct comprising a binding domain capable of binding CD28, wherein the binding domain comprises a VH sequence comprising a HCDR1 having the sequence SXiGVH (SEQ ID NO: 30), a HCDR2 having the sequence VIWX2GGX3TNFNSALMS (SEQ ID NO: 34), and a HCDR3 having the sequence DRAX4GX5YX6X7AMDY (SEQ ID NO: 40), and a VL sequence comprising a LCDR1 having the sequence RASESVEYYXsTSLMQ (SEQ ID NO: 43), a LCDR2 having the sequence AASX9VX10S (SEQ ID NO: 47), and a LCDR3 having the sequence QQSRKVPFT (SEQ ID NO: 48), and comprises one or more of the following amino acid substitutions at the positions as identified in the CDR sequences: Xi: Y to A, X2: P to A, X3: G to S, X4: S to Y, X5: N to A, Xe: L to N, X7: S to Y, X8: G to V, X9: N to A, or Xi0: E to D.

[0244] In some embodiments, the antibody construct comprises an anti-CD28 binding domain comprising the substitution Xi: Y to A. In some embodiments, the antibody construct comprises an anti-CD28 binding domain comprising the substitution X2: P to A. In some embodiments, the antibody construct comprises an anti-CD28 binding domain comprising the substitution X3: G to S. In some embodiments, the antibody construct comprises an anti-CD28 binding domaincomprising the substitution X4: S to Y. In some embodiments, the antibody construct comprises an anti-CD28 binding domain comprising the substitution X5: N to A. In some embodiments, the antibody construct comprises an anti-CD28 binding domain comprising the substitution Xe: L to N. In some embodiments, the antibody construct comprises an anti-CD28 binding domain comprising the substitution X7: S to Y. In some embodiments, the antibody construct comprises an anti-CD28 binding domain comprising the substitution Xs: G to V. In some embodiments, the antibody construct comprises an anti-CD28 binding domain comprising the substitution X9: N to A. In some embodiments, the antibody construct comprises an anti-CD28 binding domain comprising the substitution X10: E to D.

[0245] In some embodiments, such antibody construct can have a binding affinity for CD28 that is reduced by about 1.5-fold to about 25-fold, by about 2.0-fold to about 20-fold, by about 3.0-fold to about 20-fold, or by about 5.0-fold to about 10-fold, when compared to the binding affinity of an antibody construct comprising an anti-CD28 binding domain that does not contain the one or more amino acid substitutions in one or more of the CDR sequences.

[0246] In some embodiments, such antibody construct can comprise an anti-CD28 binding domain that comprises the CDRs of a VH domain as set forth in SEQ ID NOS: 28, 31, and 35, and the CDRs of the VL sequence set forth in SEQ ID NOS: 41, 44, and 48.

[0247] In some embodiments, such antibody construct can comprise an anti-CD28 binding domain that comprises an anti-CD28 VH sequence comprising a HCDR1 having the sequence SYGVH (SEQ ID NO: 28), a HCDR2 having the sequence VIWPGGGTNFNSALMS (SEQ ID NO: 31), and a HCDR3 having the sequence DRAYGNYLYAMDY (SEQ ID NO: 35), and an anti-CD28 VL sequence comprising a LCDR1 having the sequence RASESVEYYVTSLMQ (SEQ ID NO: 41), a LCDR2 having the sequence AASNVDS (SEQ ID NO: 44), and a LCDR3 having the sequence QQSRKVPFT (SEQ ID NO: 48).

[0248] As further described herein, in various embodiments, an anti-CD28 binding domain of a multivalent and multi specific antibody construct is an scFv domain. Such anti-CD28 scFv domain can be coupled to a Fab light chain sequence, e.g., to the C-terminus of a Fab light chain sequence.

[0249] In some embodiments, other anti-CD28 binding domains, such as those known in the art (e.g., CD28.3, TGN1412, etc.), as well as variants thereof, can also be used in the multivalent and multispecific antibody constructs described herein.

[0250] In certain embodiments, a multivalent and multispecific antibody construct comprises anti- CD28 scFv domain comprising a VH sequence having an amino acid sequence of at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 102, and a VL sequence having an amino acid sequence of at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 103. In some of these embodiments, the anti-CD28 scFv domain comprises a VH sequence having the sequence set forth in SEQ ID NO: 102, and a VL sequence having the sequence set forth in SEQ ID NO: 103.

[0251] In some embodiments, an anti-CD28 scFv domain comprises a VH sequence comprising the HCDRsl-3 set forth in SEQ ID NO: 106-108, and a VL sequence comprising the LCDRsl-3 set forth in SEQ ID NO: 109-111.

[0252] In certain embodiments, a multivalent and multispecific antibody construct comprises an anti-CD28 scFv domain comprising a VH sequence having an amino acid sequence of at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 104, and a VL sequence having an amino acid sequence of at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 105. In some of these embodiments, the anti-CD28 scFv domain comprises a VH sequence having the sequence set forth in SEQ ID NO: 104, and a VL sequence having the sequence set forth in SEQ ID NO: 105.

[0253] In some embodiments, an anti-CD28 scFv domain comprises a VH sequence comprising the HCDRsl-3 set forth in SEQ ID NO: 112-114, and a VL sequence comprising the LCDRsl-3 set forth in SEQ ID NO: 115-117.C. Binding Domains Against a Disease-Associated Antigen

[0254] As described herein, in various embodiments, a multivalent and multispecific antibody construct of the present disclosure comprises at least one binding domain capable of binding a disease-associated antigen.

[0255] The disease-associated antigen can be a TAA on tumor cell, or an antigen associated with an inflammatory or autoimmune disease.

[0256] In various embodiments, the disease-associated antigen targeted by an antibody construct of the present disclosure is a TAA on tumor cell.

[0257] In various embodiments, and as further described herein, an antibody construct herein can be at least trivalent and trispecific and thus comprises a first binding domain capable of binding a first antigen on a cytotoxic effector cell, a second binding domain capable of binding a secondantigen on a cytotoxic effector cell, and a third binding domain, wherein such third binding domain is capable of binding the TAA.

[0258] In some embodiments, an antibody construct herein binds a TAA monovalently, i.e., via a single binding domain, such as a Fab domain or an scFv domain.

[0259] In other embodiments, an antibody construct herein binds to a TAA bivalently, i.e., via two binding domains. In such embodiments, an antibody construct can comprise two anti-TAA binding domains that target the same epitope on the TAA. In other embodiments, an antibody construct can comprise two anti-TAA binding domains that target different epitopes on the TAA, e.g., binds the TAA bi-paratopically.

[0260] In various embodiments, and as further described herein, the binding domain of a multivalent and multispecific antibody construct that binds a disease-associated antigen such as a TAA comprises or consists of a Fab domain (e.g., in a Het-Fab antibody construct). In other embodiments, and as further described herein, the binding domain of a multivalent and multispecific antibody construct that binds a disease-associated antigen such as a TAA comprises or consists of as scFv domain (e.g., in a scFv2antibody construct).

[0261] In the context of the present disclosure, the term “tumor-associated antigen” generally relates to proteins that are under normal (e.g., healthy, or non-disease) conditions specifically expressed in a limited number of tissues and / or organs, and / or in specific developmental stages, and are expressed or aberrantly expressed in one or more tumor tissues. In the context of the present disclosure, the TAA can be associated with the cell surface of a tumor cell and may not be or only rarely expressed in normal tissues. A TAA herein can be associated with a certain type of tumor based on the type of cell and / or tissue it originated from.

[0262] In some embodiments, a TAA that can be targeted by an antibody construct of this disclosure can be associated with a carcinoma, a sarcoma, a myeloma, a leukemia, a lymphoma, or a mixed type tumor. In certain embodiments, a multivalent and multispecific antibody construct of the present disclosure can bind a TAA of a solid tumor. Solid tumors can include carcinomas, sarcomas, or lymphomas. In other embodiments, a TAA that can be targeted by an antibody construct of the present disclosure can be that of a non-solid tumor (e.g., a liquid tumor), such as a leukemia.

[0263] In some embodiments, the TAA can be Fibroblast activation protein alpha (FAPa), Trophoblast glycoprotein (5T4), Tumor-associated calcium signal transducer 2 (Trop2),Fibronectin EDB (EDB-FN), fibronectin F.IIIB domain, CGS-2, EpCAM, EGFR, HER-2, HER- 3, cMet, GPC3, FOLR1, EpCAM, PSMA, CD38, DLL3, MSLN or Cldnl8.2.

[0264] In some embodiments, the TAA can be associated with a solid tumor. Examples of such TAAs can include GPC3, MSLN, DLL3, HER2, FAP, EGFRvIII, PSMA, Cldnl8.2, Mucin-1 and EpCAM.

[0265] In various embodiments, any solid tumor-associated antigen may be targeted by a multivalent and multispecific antibody construct of the present disclosure as long as the antibody construct comprises a binding domain that is capable of engaging the solid tumor-associated antigen with a high enough affinity to allow formation of a TCR-independent immune synapse between an immune cell (e.g., T cell, engaged via the constructs anti-CD3 and anti-CD28 domains) and the tumor cell expressing the antigen, followed by induction of an anti-tumor immune response resulting in tumor cell lysis or other events that are tumor cell-static or -toxic.

[0266] In some embodiments, a multivalent and multispecific antibody construct of the present disclosure comprises a binding domain capable of binding Delta-like ligand 3 (DLL3). In various embodiments, an antibody construct herein can be at least trivalent and trispecific and comprises a first binding domain capable of binding a first antigen on a cytotoxic effector cell, a second binding domain capable of binding a second antigen on a cytotoxic effector cell, and a third binding domain, wherein such third binding domain is capable of binding DLL3. Generally, the anti-DLL3 binding domain of an antibody construct herein can comprise or consist of an scFv domain or a Fab domain. In various embodiments of a multivalent and multispecific antibody construct, the anti-DLL3 binding domain comprises a Fab domain. In other embodiments of a multivalent and multispecific antibody construct, the anti-DLL3 binding domain comprises an scFv domain. The anti-DLL3 binding domains that can be part of a multivalent and multispecific antibody construct described herein can encompass binding domains capable of recognizing and binding any one of the epitopes, portions, domains or motifs located on, e.g., mouse, cynomolgus and / or human DLL3. In various embodiments, an anti-DLL3 binding domain of a multivalent and multispecific antibody construct described herein can bind to an epitope of human DLL3, which can include an N-terminal ECD domain, EGF1-6, and / or the membrane proximal peptide. Anti-DLL3 binding domains against one or more of these human DLL3 epitopes are known in the art, and any such binding domain which is capable of binding DLL3 can generally be used in the multivalent and multispecific antibody constructs described herein. In various embodiments, a multivalent andmultispecific antibody construct of the present disclosure, e.g., a multivalent and multispecific Het-Fab or scFv2antibody construct, comprises an anti-DLL3 binding domain (e.g., an scFv domain or a Fab domain) that has a KD value for binding DLL3 from about 0.01 nM to about 100 nM or from about 0.1 nM to about 50 nM. In some embodiments, a multivalent and multispecific antibody construct has a binding affinity (e.g., provided as a KD value) for DLL3 of at least about 40 nM, 30 nM, 20 nM, 10 nM, 5 nM, 3 nM, 2 nM, 1 nM, 0.5 nM, or at least about 0.2 nM. In some embodiments, a multivalent and multispecific antibody construct herein has a binding affinity (e.g., provided as a KD value) for DLL3 of from about 40 nM to about 0.1 nM, from about 20 nM to about 0.2 nM, from about 10 nM to about 0.2 nM, or from about 5 nM to about 0.2 nM, e.g., using SPR or other methods known in the art.

[0267] In some embodiments, an anti-DLL3 binding domain, e.g., an anti-DLL3 VH and / or an anti-DLL3 VL sequence, of a multivalent and multispecific antibody construct of this disclosure can be one that is known in the art, or a variant (e.g., having about thereof, e.g., an anti-DLL3 binding domain that is described in any one of International Patent Publication Nos. WO 2013 / 126746, WO 2016 / 138038, WO 2017 / 021349, WO 2021 / 200898, WO 2019 / 234220, WO 2021 / 173307, WO 2021 / 226204 and WO 2021 / 155380, which are incorporated herein by reference in their entireties.

[0268] In certain embodiments, a multivalent and multispecific antibody construct of the present disclosure, e.g., a multivalent and multispecific (e.g., a Het-Fab or scFv2) antibody construct, comprises an anti-DLL3 binding domain (e.g., a Fab domain or an scFv domain, respectively) that comprises a VH sequence comprising the HCDR1 sequence as set forth in SEQ ID NO: 158, the HCDR2 sequence as set forth in SEQ ID NO: 159 and the HCDR3 sequence as set forth in SEQ ID NO: 160, and a VL sequence comprising the LCDR1 sequence as set forth in SEQ ID NO: 161, the LCDR2 sequence as set forth in SEQ ID NO: 162 and the LCDR3 sequence as set forth in SEQ ID NO: 163.

[0269] In some embodiments, a multivalent and multispecific antibody construct of the present disclosure, e.g., a trivalent and trispecific Het-Fab or scFv2antibody construct, comprises an anti- DLL3 binding domain (e.g., a Fab domain or an scFv domain, respectively) that comprises a VH sequence comprising the amino acid sequence set forth in SEQ ID NO: 152, and a VL sequence comprising the amino acid sequence set forth in SEQ ID NO: 153.

[0270] In certain embodiments, a multivalent and multispecific antibody construct of the present disclosure, e.g., a trivalent and trispecific Het-Fab or scFv2antibody construct, comprises an anti- DLL3 binding domain (e.g., a Fab domain or an scFv domain, respectively) that comprises a VH sequence comprising the HCDR1 sequence as set forth in SEQ ID NO: 158, the HCDR2 sequence as set forth in SEQ ID NO: 159 and the HCDR3 sequence as set forth in SEQ ID NO: 160, and a VL sequence comprising the LCDR1 sequence as set forth in SEQ ID NO: 164, the LCDR2 sequence as set forth in SEQ ID NO: 165 and the LCDR3 sequence as set forth in SEQ ID NO: 163.

[0271] In some embodiments, a multivalent and multispecific antibody construct of the present disclosure, e.g., a trivalent and trispecific Het-Fab or scFv2antibody construct, comprises an anti- DLL3 binding domain (e.g., a Fab domain or an scFv domain, respectively) that comprises a VH sequence comprising the amino acid sequence set forth in SEQ ID NO: 154, and a VL sequence comprising the amino acid sequence set forth in SEQ ID NO: 155.

[0272] In some embodiments, a multivalent and multispecific antibody construct of the present disclosure, e.g., a trivalent and trispecific Het-Fab or scFv2antibody construct, comprises an anti- DLL3 binding domain (e.g., a Fab domain or an scFv domain, respectively) that comprises a VH sequence comprising the amino acid sequence set forth in SEQ ID NO: 156, and a VL sequence comprising the amino acid sequence set forth in SEQ ID NO: 157.

[0273] As described herein, in some embodiments, the anti-TAA binding domain of an antibody construct herein can be capable of binding mesothelin (MSLN). In various embodiments, the anti- TAA binding domain (e.g., scFv domain or Fab domain) of an antibody construct herein is capable of binding MSLN. In some embodiments, such anti -MSLN binding domain comprises a VH sequence comprising a HCDR1 having the sequence GYTMN (SEQ ID NO: 118), a HCDR2 having the sequence LITPYSGASSYAQKFQG (SEQ ID NO: 119), and a HCDR3 having the sequence GGYDGRGFDY (SEQ ID NO: 120) and a VL sequence comprising a LCDR1 having the sequence SASSSVSYMH (SEQ ID NO: 121), a LCDR2 having the sequence DTSKLAS (SEQ ID NO: 122), and a LCDR3 having the sequence QQWSGHPLT (SEQ ID NO: 123). In some embodiments, the anti-MSLN binding domain of an antibody construct herein comprises a VH sequence that comprises or consists of an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 132, and a VL sequence that comprises or consists of an amino acid sequence having at least about90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 133. Other anti-MSLN binding domains that are known in the art can also be used in the context of the multivalent and multispecific antibody constructs described herein.

[0274] In yet other embodiments, the anti-TAA binding domain of an antibody construct herein can be capable of binding Claudinl8.2 (Cldnl8.2). Generally, such anti-TAA binding domain herein can comprise or consist of an scFv domain or a Fab domain. In various embodiments, the anti-TAA binding domain of a multivalent and multispecific antibody construct herein is an scFv domain. In other embodiments, the anti-TAA binding domain of a multivalent and multispecific antibody construct herein is a Fab domain. In various embodiments, the anti-TAA binding domain (e.g., scFv domain or Fab domain) of an antibody construct herein is capable of binding Cldnl8.2. In some embodiments, such anti-Cldnl8.2 binding domain comprises a HCDR1 having the sequence SNPMI (SEQ ID NO: 124), a HCDR2 having the sequence IIDTDGSTYYADWAKG (SEQ ID NO: 125), and a HCDR3 having the sequence RLHGSSNGYYDDL (SEQ ID NO: 126) and a VL sequence comprising a LCDR1 having the sequence QASQSIYSYLS (SEQ ID NO: 127), a LCDR2 having the sequence KASTLAS (SEQ ID NO: 128), and a LCDR3 having the sequence QQGYTVTNVDKNT (SEQ ID NO: 129). In some embodiments, an anti-Cldnl8.2 binding domain (e.g., scFv domain or Fab domain) of an antibody construct herein comprises a VH sequence comprising or consisting of an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 130, and a VL sequence comprising or consisting of an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 131. In some embodiments, the anti-Cldnl8.2 binding domain of an antibody construct herein is an scFv domain and comprises, either from N- to C-terminus or from C- to N-terminus, a VH sequence comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 130, coupled to a VL sequence comprising or consisting of the amino acid sequence set forth in SEQ ID NO: 131, via a linkerscFvhaving the amino acid sequence set forth in SEQ ID NO: 59. Other anti-Cldnl8.2 binding domains that are known in the art can also be used in the context of the multivalent and multi specific antibody constructs described herein.

[0275] Generally, in various embodiments, an antibody construct of this disclosure can bind a TAA with a KD from about 0.01 nM to about 10 nM, from about 0.05 nM to about 5 nM, or from about 0.1 nM to about 5 nM.D. Fc Domains

[0276] As described herein, a multivalent and multispecific antibody construct of this disclosure can comprise an Fc domain (or Fc region, or Fc, also commonly known as an abbreviation of “fragment, crystallizable”) comprising a first Fc polypeptide and a second Fc polypeptide. In various embodiments, the Fc domain is a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein the first Fc polypeptide and the second Fc polypeptide share at least about 90%, 95%, 97%, or about 99% amino acid sequence identity, e.g., each comprising one or more asymmetric amino acid substitutions that can promote preferential pairing for the Fc polypeptides to form the heterodimeric Fc domain (e.g., H1-H2) compared to formation of a respective homodimeric Fc domain (e.g., Hl-Hl or H2-H2).

[0277] The term “Fc domain,” as used herein, includes native (or wildtype) sequence Fc domains as well as variant Fc domains comprising one or more amino acid modifications relative to a corresponding native or wildtype Fc domain. Unless otherwise specified herein, numbering of amino acid residues in the Fc domain or constant region is according to the EU numbering system, also called the EU index, as described, e.g., in Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD (1991). An “Fc polypeptide” of a dimeric (e.g., heterodimeric) Fc domain refers to one of the two polypeptide chains (e.g., a first and second Fc polypeptide) forming the dimeric (e.g., heterodimeric) Fc domain. In some embodiments, Fc polypeptides can comprise a C-terminal constant region of an Ig heavy chain that is capable of stable self-association. In various embodiments, and as further described herein, an Fc polypeptide (e.g., a first or a second Fc polypeptide) comprises at least one of a CH2 domain and / or a CH3 domain. In certain embodiments, an Fc polypeptide of an antibody construct described herein comprises a CH2 domain and a CH3 domain.

[0278] As disclosed herein, a multivalent and multispecific antibody construct can comprise an Fc domain, wherein such Fc domain can be a heterodimeric Fc domain. The Fc domain, e.g., heterodimeric Fc domain, of an antibody construct, unless otherwise specified, comprises a first Fc polypeptide and a second Fc polypeptide. Generally, each Fc polypeptide of a (e.g., heterodimeric) Fc domain can comprise a CH2 domain, a CH3 domain, or, as described in various embodiments herein, both a CH2 domain and a CH3 domain.

[0279] In certain embodiments, an antibody construct herein comprises an Fc domain based on a human IgG Fc domain, such as an IgGl, IgG2, IgG3 or IgG4 domain. In some embodiments, an antibody construct comprises an Fc domain based on a human IgGl Fc domain. In other embodiments, an antibody construct comprises an Fc domain based on a human IgG4 Fc domain. In various embodiments, an antibody construct comprises a heterodimeric IgG Fc domain comprising two different Fc polypeptides, e.g., a first Fc polypeptide and a second polypeptide, wherein the first and second Fc polypeptides have different amino acid sequences, e.g., amino acid sequences that have about 90%, 95%, 97%, or 99% sequence identity when compared and aligned to one another, e.g., as further described herein. In some embodiments, the differences in the amino acid sequences of a first and second Fc polypeptide can be due to asymmetric amino acid substitutions that can be introduced into each Fc polypeptide chain to promote preferential paring of the heavy chains to form the heterodimeric Fc domain, compared to a corresponding homodimeric Fc domain.

[0280] In various embodiments, a multivalent and multispecific antibody construct herein comprises an Fc domain that is a modified IgG Fc domain, and in which at least the CH3 domain of at least one Fc polypeptide comprises one or more amino acid modifications compared to a respective wildtype CH3 domain. In some embodiments, an antibody construct herein comprises an Fc domain that is a modified IgG Fc domain in which at least the CH2 domain of at least one Fc polypeptide comprises one or more amino acid modifications compared to a respective wildtype CH2 domain. In some embodiments, an antibody construct comprises an Fc domain that is a modified IgG Fc domain in which both the CH3 domain and the CH2 domain of at least one Fc polypeptide comprises one or more amino acid modifications compared to respective wildtype CH3 and CH2 domains. In various embodiments, both Fc polypeptides of a heterodimeric Fc domain can comprise one or more amino acid modifications in their CH3 domains. In some embodiments, both Fc polypeptides of a heterodimeric Fc domain can comprise one or more amino acid modifications in their CH2 domains. In yet other embodiments, both Fc polypeptides of a heterodimeric Fc domain can comprise one or more amino acid modifications in their CH2 domains and CH3 domains.Modified Fc Domains

[0281] In some embodiments, the present disclosure relates to multivalent and multispecific antibody constructs that can comprise a heterodimeric Ig (e.g., IgG) Fc domain comprising amodified heterodimeric CH3 domain, wherein the modified heterodimeric CH3 domain comprises one or more asymmetric amino acid modifications, i.e., one or both of the first and the second Fc polypeptide(s) each comprise one or more amino acid modifications in their CH3 domain sequences compared to respective wildtype sequences. As used herein, the term “asymmetric amino acid modification” generally refers to a modification in which an amino acid at a specific position on the first Fc polypeptide is different to the amino acid at the corresponding position on the second Fc polypeptide. These asymmetric amino acid modifications can comprise modifications of only one of the two amino acids at the corresponding position on each Fc polypeptide, or they can comprise modifications of both amino acids at the corresponding positions on each of the first and second Fc polypeptides. In various embodiments, an “asymmetric amino acid modification” is an asymmetric amino acid substitution.

[0282] In some embodiments, an antibody construct herein comprises a heterodimeric Fc domain comprising a modified CH3 domain (i.e., a heterodimeric CH3 domain consisting of the two CH3 domains of the first and second Fc polypeptides), wherein the modified CH3 domain comprises one or more asymmetric amino acid modifications that promote formation of the heterodimeric Fc domain (e.g., pairing of a first Fc polypeptide with a second Fc polypeptide) over formation of a corresponding homodimeric Fc domain (e.g., pairing of a first Fc polypeptide with another first Fc polypeptide). Amino acid modifications that can be made to the CH3 domain of an Fc domain in order to promote formation of a heterodimeric Fc domain are known in the art and include, for example, those described in International Publication No. WO 96 / 027011 (“knobs into holes”), Gunasekaran etal., 2010, J Biol Chem, 285, 19637-46 (“electrostatic steering”), Davis etal., 2010, Prot Eng Des Sei, 23(4): 195-202 (strand exchange engineered domain (SEED) technology) and Labrijn et al., 2013, Proc Natl Acad Sci USA, 110(13): 5145-50 (Fab-arm exchange). Other examples include approaches combining positive and negative design strategies to produce stable asymmetrically modified Fc regions as described in International Patent Publication Nos. WO 2012 / 058768 and WO 2013 / 063702.

[0283] In certain embodiments, an antibody construct described herein comprises a heterodimeric Fc domain comprising a modified heterodimeric CH3 domain in which at least one, or both of the Fc polypeptide chains comprise one or more amino acid modifications, as described in International Publication No. WO 2012 / 058768 or International Patent Publication No. WO 2013 / 063702.

[0284] In some embodiments, an antibody construct described herein comprises a heterodimeric human IgGl Fc domain having a modified CH3 domain. TABLE 2 herein provides the amino acid sequence of a human IgGl Fc domain sequence (e.g., a sequence that a first and / or a second Fc polypeptide can be derived from), corresponding to amino acids 231 to 447 of a full-length human IgGl heavy chain (e.g., one that comprises VH, CHI, Hinge, CH2 and CH3 domains), and identified by SEQ ID NO: 78. The CH2 domain is typically defined as comprising amino acids 231-340 of the full-length human IgGl heavy chain and the CH3 domain is typically defined as comprising amino acids 341-447 of the full-length human IgGl heavy chain.

[0285] As described herein, an antibody construct can comprise a heterodimeric Fc domain having a modified CH3 domain comprising one or more asymmetric amino acid modifications that promote formation of the heterodimeric Fc domain over formation of a homodimeric Fc domain, and in which the modified CH3 domain comprises a first Fc polypeptide including amino acid modifications at positions F405 and Y407, relative to SEQ ID NO: 78, and a second Fc polypeptide including amino acid modifications at positions T366 and T394, relative to SEQ ID NO: 78. In various embodiments, the one or more amino acid modifications comprise one or more amino acid substitutions. Hence, in some embodiments, the amino acid modification at position F405 of the first Fc polypeptide of the modified CH3 domain is F405A, F405I, F405M, F405S, F405T or F405V. In some embodiments, the amino acid modification at position Y407 of the first Fc polypeptide of the modified CH3 domain is Y407I or Y407V. In some embodiments, the amino acid modification at position T366 of the second Fc polypeptide of the modified CH3 domain is T366I, T366L or T366M. In some embodiments, the amino acid modification at position T394 of the second Fc polypeptide of the modified CH3 domain is T394W. In some embodiments, the modified CH3 domain of a first Fc polypeptide further includes an amino acid modification at position L351, relative to SEQ ID NO: 78. In some embodiments, the amino acid modification at position L351 in the first Fc polypeptide of the modified CH3 domain is L351Y. In some embodiments, the second Fc polypeptide of the modified CH3 domain further includes an amino acid modification at position K392, relative to SEQ ID NO: 78. In some embodiments, the amino acid modification at position K392 in the second Fc polypeptide of the modified CH3 domain is K392F, K392L or K392M. In some embodiments, one or both of the first and second Fc polypeptides of the modified CH3 domain further comprises the amino acid modification T350V.

[0286] In certain embodiments, an antibody construct herein comprises a heterodimeric Fc domain having a modified CH3 domain comprising one or more asymmetric amino acid modifications that promote formation of the heterodimeric Fc domain over formation of a homodimeric Fc domain, and in which the modified CH3 domain comprises a first Fc polypeptide including the amino acid modification F405A, F405I, F405M, F405S, F405T or F405V together with the amino acid modification Y407I or Y407V, and relative to SEQ ID NO: 78, and a second Fc polypeptide including the amino acid modification T366I, T366L or T366M, together with the amino acid modification T394W, and relative to SEQ ID NO: 78. In some embodiments, the first Fc polypeptide of the modified CH3 domain further includes the amino acid modification L351Y. In some embodiments, the second Fc polypeptide of the modified CH3 domain further includes the amino acid modification K392F, K392L or K392M. In some embodiments, one or both of the first and second Fc polypeptides having a modified CH3 domain further comprises the amino acid modification T350V.TABLE 2: Exemplary Human IgGl Fc Domain Sequences and Variants Thereof* “A” corresponds to a first Fc polypeptide chain and “B” to corresponds to a second Fc polypeptide chain.

[0287] In certain embodiments, an antibody construct herein comprises a heterodimeric Fc domain comprising a modified CH3 domain having a first Fc polypeptide that comprises amino acid modifications at positions F405 and Y407, and optionally further comprises an amino acidmodification at position L351, and a second Fc polypeptide that comprises amino acid modifications at positions T366 and T394, and optionally further comprises an amino acid modification at position K392, as described above, and the first Fc polypeptide further comprises an amino acid modification at one or both of positions S400 or Q347 and / or the second Fc polypeptide further comprises an amino acid modification at one or both of positions K360 or N390, wherein the amino acid modification at position S400 is S400E, S400D, S400R or S400K; the amino acid modification at position Q347 is Q347R, Q347E or Q347K; the amino acid modification at position K360 is K360D or K360E, and the amino acid modification at position N390 is N390R, N390K or N390D, relative to SEQ ID NO: 78.

[0288] In some embodiments, an antibody construct comprises a heterodimeric Fc domain comprising a modified CH3 domain comprising the modifications of any one of Variant 1, Variant 2, Variant 3, Variant 4 or Variant 5, as shown in TABLE 2.

[0289] In various embodiments, an antibody construct of the present disclosure can comprise a heterodimeric Fc domain comprising a first Fc polypeptide (A) and a second Fc polypeptide (B), wherein the first Fc polypeptide (A) and the second Fc polypeptide (B) comprise the amino acid substitutions in their CH3 domains according to variant #1 as shown in TABLE 2. In other embodiments, an antibody construct of the present disclosure can comprise a heterodimeric Fc domain comprising a first Fc polypeptide (A) and a second Fc polypeptide (B), wherein the first Fc polypeptide (A) and the second Fc polypeptide (B) comprise the amino acid substitutions in their CH3 domains according to variant #2 as shown in TABLE 2. In some embodiments, an antibody construct of the present disclosure can comprise a heterodimeric Fc domain comprising a first Fc polypeptide (A) and a second Fc polypeptide (B), wherein the first Fc polypeptide (A) and the second Fc polypeptide (B) comprise the amino acid substitutions in their CH3 domains according to variant #3 as shown in TABLE 2. In some embodiments, an antibody construct of the present disclosure can comprise a heterodimeric Fc domain comprising a first Fc polypeptide (A) and a second Fc polypeptide (B), wherein the first Fc polypeptide (A) and the second Fc polypeptide (B) comprise the amino acid substitutions in their CH3 domains according to variant #4 as shown in TABLE 2. In yet other embodiments, an antibody construct of the present disclosure can comprise a heterodimeric Fc domain comprising a first Fc polypeptide (A) and a second Fc polypeptide (B), wherein the first Fc polypeptide (A) and the second Fc polypeptide (B)comprise the amino acid substitutions in their CH3 domains according to variant #5 as shown in TABLE 2

[0290] In certain embodiments, the CH3 domain of a first Fc polypeptide of an antibody construct herein has an amino acid sequence that is at least about 80%, about 85%, about 90%, about 95%, about 97%, or at least about 99% identical to the amino acid sequence set forth in SEQ ID NO: 9. In certain embodiments, the CH3 domain of a second Fc polypeptide of an antibody construct herein has an amino acid sequence that is at least about 80%, about 85%, about 90%, about 95%, about 97%, or at least about 99% identical to the amino acid sequence set forth SEQ ID NO: 10. In some embodiments, the CH3 domain of a first Fc polypeptide of an antibody construct herein has the amino acid sequence set forth in SEQ ID NO: 9 and a second Fc polypeptide of the antibody construct herein has the amino acid sequence set forth SEQ ID NO: 10.

[0291] In certain embodiments, an antibody construct herein comprises a heterodimeric Fc domain based on an IgG Fc domain having a modified CH2 domain (i.e., a heterodimeric CH2 domain consisting of the two CH2 domain sequences of the respective first and second Fc polypeptides). In some embodiments, an antibody construct comprises a heterodimeric Fc domain based on an IgG Fc domain having a modified CH2 domain, wherein the modification(s) of the CH2 domain result(s) in altered (e.g., reduced or abated) binding to one or more Fc receptors (FcRs) such as receptors of the FcyRI, FcyRII and FcyRIII subclasses.

[0292] Several amino acid modifications to the CH2 domain of the first and / or second Fc polypeptide(s) of an Fc domain that selectively alter the affinity of such Fc domain for different Fey receptors are known in the art. Amino acid modifications that result in increased binding and amino acid modifications that result in decreased binding can both be useful in certain indications. For example, increasing binding affinity of an Fc for FcyRIIIa (an activating receptor) can result in increased antibody dependent cell-mediated cytotoxicity (ADCC), which in turn can result in increased lysis of the target cell. Decreased binding to FcyRIIb (an inhibitory receptor) likewise can be beneficial in some circumstances. In certain indications, a decrease in, or elimination of, ADCC and complement-mediated cytotoxicity (CDC) can be desirable. In such embodiments, modified CH2 domains comprising amino acid modifications that result in increased binding to FcyRIIb or amino acid modifications that can decrease or eliminate binding of the Fc region to all of the Fey receptors (“knock-out” variants) can be useful.

[0293] Non-limiting examples of amino acid modifications to the CH2 domain that alter binding of the Fc domain by Fey receptors include, but are not limited to, the following: S298A / E333A / K334A and S298A / E333A / K334A / K326A (increased affinity for FcyRIIIa) (Lu, et al., 2011, J Immunol Methods, 365(1-2): 132-41); F243L / R292P / Y300L / V305I / P396L (increased affinity for FcyRIIIa) (Stavenhagen, et al., 2007, Cancer Res, 67(18):8882-90); F243L / R292P / Y300L / L235V / P396L (increased affinity for FcyRIIIa) (Nordstrom JL, et a!., 2011, Breast Cancer Res, 13(6):R123); F243L (increased affinity for FcyRIIIa) (Stewart, et al, 2011, Protein Eng Des Se , 24(9):671-8); S298A / E333A / K334A (increased affinity for FcyRIIIa) (Shields, et al., 2001, J Biol Chem, 276(9):6591-604); S239D / I332E / A330L and S239D / I332E (increased affinity for FcyRIIIa) (Lazar, et al., 2006, Proc Natl Acad Sci USA, 103(11):4005-10), and S239D / S267E and S267E / L328F (increased affinity for FcyRIIb) (Chu, et al., 2008, Mol Immunol, 45(15):3926-33). Additional modifications that affect Fc domain binding to Fey receptors are described in Therapeutic Antibody Engineering (Strohl & Strohl, Woodhead Publishing series in Biomedicine No 11, ISBN 1 907568 37 9, Oct 2012, page 283).

[0294] In various embodiments, an antibody construct of the present disclosure comprises a heterodimeric Fc domain based on an IgG Fc domain having a modified CH2 domain, in which one or both of the CH2 sequences (i.e., of the first / second Fc polypeptide) of the modified dimeric CH2 domain comprise one or more amino acid modifications that can result in decreased or eliminated binding of the Fc domain to one or more, or all of the Fey receptors (i.e., a “knock-out” or “KO” variant).

[0295] Various publications describe strategies that have been used to engineer antibodies to produce “knock-out” Fc variants (see, for example, Strohl, 2009, Curr Opin Biotech 20:685-691, and Strohl & Strohl, “ Antibody Fc engineering for optimal antibody performance" In Therapeutic Antibody Engineering, Cambridge: Woodhead Publishing, 2012, pp 225-249). These strategies include reduction of effector function through modification of glycosylation, use of IgG2 / IgG4 scaffolds, or the introduction of mutations in the hinge or CH2 domain of the Fc (see also, U.S. Patent Publication No. 2011 / 0212087, International Patent Publication No. WO 2006 / 105338, U.S. Patent Publication No. 2012 / 0225058, U.S. Patent Publication No. 2012 / 0251531, and Strop et a!., 2012, J. Mol. Biol., 420: 204-219).

[0296] In some embodiments, an Fc domain of an antibody construct can comprise one or more of known amino acid modifications to reduce FcyR and / or complement binding of the Fc domain. In some embodiments, such modifications can include those identified in TABLE 3.TABLE 3: Modifications to Reduce Fey Receptor or Complement Binding

[0297] Additional examples herein include Fc domains engineered to include the amino acid modifications L235A / L236A / D265S, e.g., based on the sequence set forth in SEQ ID NO: 78. In addition, asymmetric amino acid modifications in the CH2 domain that decrease binding of the Fc to all Fey receptors are described in International Patent Publication No. WO 2014 / 190441.

[0298] In certain embodiments, the CH2 domain of a first and a second Fc polypeptide herein comprises or consists of an amino acid sequence having at least about 80%, about 85%, about 90%, about 95%, about 97%, or at least about 99% sequence identity to the sequence set forth in SEQ ID NO: 8. In some embodiments, the CH2 domain of a first and / or a second Fc polypeptide herein comprises or consists of the sequence set forth in SEQ ID NO: 8.

[0299] In certain embodiments, an antibody construct herein comprises a heterodimeric Fc domain in which native glycosylation has been modified. As is known in the art, glycosylation of an Fc can be modified to increase or decrease effector function. For example, mutation of the conserved asparagine residue at position 297 to alanine, glutamine, lysine, or histidine (i.e., N297A, Q, K or H) results in an aglycoslated Fc that lacks all effector function (Bolt et a!.. 1993, Eur. J. Immunol., 23:403-411; Tao & Morrison, 1989, J. Immunol., 143:2595-2601). Conversely, removal of fucose from heavy chain N297-linked oligosaccharides has been shown to enhance ADCC, based on improved binding to FcyRIIIa (see, for example, Shields et al., 2002, J Biol Chem., 277:26733- 26740, and Niwa et al., 2005, J. Immunol. Methods, 306: 151-160). Such low fucose antibody constructs can be produced, for example in knockout Chinese hamster ovary (CHO) cells lackingfucosyltransferase (FUT8) (Yamane-Ohnuki et al., 2004, Biotechnol. Bioeng., 87:614-622), in the variant CHO cell line, Lee 13, that has a reduced ability to attach fucose to N297-linked carbohydrates (International Publication No. WO 03 / 035835), or in other cells that generate afucosylated antibodies (see, for example, Li et al., 2006, Nat Biotechnol, 24:210-215; Shields et al., 2002, ibid, and Shinkawa et al., 2003, J. Biol. Chem., 278:3466-3473). In addition, International Patent Publication No. WO 2009 / 135181 describes the addition of fucose analogs to culture medium during antibody production to inhibit incorporation of fucose into the carbohydrate on the antibody.E. Linkers

[0300] In various embodiments, a multivalent and multispecific antibody construct described herein can comprise one or more linkers. In some embodiments, such one or more linkers are peptide (also referred to herein as “peptitic”) linkers comprising or consisting of an amino acid sequence of about 1, 2, 3, 5, 10, 15, 20, 25, 30, 40, or about 50 consecutive amino acid residues in length. The one or more peptide linker of an antibody construct can comprise or consist of an amino acid sequence from 1 to about 50, from 2 to about 40, from 3 to about 30, or from 5 to about 25 consecutive amino acid residues in length.

[0301] Such peptide linkers can couple, or link, two or more polypeptide sequences and / or domains of an antibody construct to each other. In various embodiments, a linker herein can couple a first polypeptide chain, e.g., a heavy chain constant domain (CHI) of a Fab domain, to an Fc polypeptide. Thus, a linker can be used to couple one domain of an antibody construct to another domain. Examples include (from N- to C-terminus) (i) a linkerFab'Fcthat couples a Fab domain to an Fc domain, (ii) a linkerscFvthat couples VH domain to a VL domain, or vice versa depending on the scFv domain structure, (iii) a linkerscFv'Fcthat couples an scFv domain to an Fc domain, and (iv) a linkerFab'scFvthat couples a Fab domain (e.g., a Fab light chain) to an scFv domain. In embodiments in which an antibody construct comprises two scFv domains (e.g., an scFv2antibody construct), both such scFv domains of the antibody construct can contain a linkerscFvwith either identical or different amino acid sequence. In embodiments in which both scFv domains contain an identical linkerscFv, such antibody construct can be described as comprising a linkerscFv(instead of specifying that it contains a linkerscFvland a linkerscFv2as both linkers contain the same amino acid sequence). In other embodiments in which the two linkerscFvof both scFv domains have different amino acid sequences, such antibody construct can be described as comprising alinkerscFvland a linkerscFv2, indicating that the two linker sequences differ in at least one amino acid residue.

[0302] In embodiments in which a linker couples, e.g., a heavy chain variable domain (VH) to, e.g., a light chain variable domain (VL), the linker can be of sufficient length to allow both domains to elicit their biological function. In addition to providing a spacing function, a linker herein (e.g., a peptide linker) can provide flexibility or rigidity suitable for properly orienting the one or more domains of an antibody construct, both within the antibody construct itself and between the antibody construct and its biological target(s).

[0303] Furthermore, a linker herein (e.g., a peptide linker) can support (i) expression of a full- length fusion protein, e.g., a full-length polypeptide chain Hl, LI, H2, and / or L2 etc. of an antibody construct, and (ii) provide increased stability of the purified protein both in vitro and in vivo, e.g., following administration to a subject in need thereof, such as a human. The one or more linkers used in antibody constructs herein are generally non-immunogenic or poorly immunogenic in mammalian subjects that the construct may be administered to. In certain embodiments, one or more of the linker(s) used in an antibody construct herein can comprise part or all of (i) a human Ig hinge region, (ii) a stalk region of C-type lectins, (iii) a family of type II membrane proteins, or combinations thereof. In certain embodiments, one or more of the linker(s) used in an antibody construct herein can comprise part or all of a human Ig hinge region, such as an IgGl hinge region, or is a derivative thereof comprising one or more amino acid modification relative to a wildtype hinge region. Examples of linkers that may comprise part or all of a human Ig hinge region, or a derivative thereof, are linkerFab'Fcor linkerscFv'Fc.

[0304] In certain embodiments, each linker used in an antibody construct herein can comprise or consist of an amino acid sequence having a length of 2 to about 50 amino acids. In some embodiments, each linker used in an antibody construct herein can comprise or consist of an amino acid sequence having a length from about 3 to about 40 amino acids, from about 10 to about 50 amino acids, from about 2 to about 40 amino acids, from about 5 to about 30 amino acids, from about 5 to about 25 amino acids, from about 4 to about 30 amino acids, from about 10 to about 30 amino acids, or from about 15 to about 25 amino acids. In some embodiments, the one or more linkers of an antibody construct can each comprise an amino acid sequence comprising or consisting of 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26,27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, or 50 consecutive amino acids.

[0305] In certain embodiments, a linker (e.g., a linkerscFv, linkerFab'scFv, linkerFab'Fc, etc.) of an antibody construct herein comprises of consists of the amino acid sequence (EAAAK)nwherein n is an integer from 1 to 5 (SEQ ID NO: 60). In some embodiments, a linker comprises or consists of the sequence EAAAK (SEQ ID NO: 61). In some embodiments, a linker comprises or consists of the sequence EAAAKEAAAK (SEQ ID NO: 62). In some embodiments, a linker comprises a polyproline linker, e.g., having an amino acid sequence of PPP or PPPP (SEQ ID NO: 64). In certain embodiments, a linker is a glycine (G)-proline (P) polypeptide linker, e.g., comprising or consisting of one or more of GPPPG (SEQ ID NO: 65), GGPPPGG (SEQ ID NO: 66), GPPPPG (SEQ ID NO: 67), or GGPPPPGG (SEQ ID NO: 68). In some embodiments, a linker herein is a (GnS)m linker, wherein n and m are independently integers from 1 to 5 (SEQ ID NO: 69). In certain embodiments, a linker comprises or consists of an amino acid sequence of (G3S)n(G4S)i (SEQ ID NO: 70), (G3S)i(G4S)n(SEQ ID NO: 71), (G3S)n(G4S)n(SEQ ID NO: 72), wherein each n is an integer from 1 to 5, or (G4S)X(SEQ ID NO: 63) wherein each x is an integer from 1 to 5. In certain embodiments, a linker herein is suitable for connecting two different domains of an antibody construct and comprises a sequence comprising glycine-serine linkers, for example, but not limited to, (GmS)n-GG (SEQ ID NO: 73), (SGn)m(SEQ ID NO: 74), or (SEGn)m(SEQ ID NO: 75), wherein m and n are independently integers from 1 to 20.

[0306] In some embodiments, an antibody construct described herein comprises any one or more of the linkers described herein. In some embodiments, an antibody construct comprises a plurality of linkers, e.g., 1, 2, 3, 4, 5 or more linkers, which can include one or more linkerFab'Fc, one or more linkerscFv'Fc, one or more linkerFab'scFv, and / or one or more linkerscFv.

[0307] In certain embodiments, a linkerscFvof an antibody construct herein comprises or consists of the amino acid sequence (GnS)mlinker, wherein n and m are independently integers from 1 to 5 (SEQ ID NO: 69). In such embodiments, n and m can both be 4, and thus the one or more linkerscFvof an antibody construct can comprise or consist of the sequence (G4S)4(SEQ ID NO: 59). In some embodiments, the one or more linkerscFvof an antibody construct (e.g., a linkerscFvland a linkerscFv2) can comprise or consist of an amino acid sequence having about 80%, 90%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 59.

[0308] In some embodiments, an antibody construct of the present disclosure can comprise a linkerFab'scFvthat couples the C-terminus of a Fab light chain to the N-terminus of an scFv domain. In some embodiments, such linkerFab'scFvcan comprise or consist of the amino acid sequence set forth in SEQ ID NO: 58 (GGGGSG), or a sequence having about 40%, 60%, or 80% sequence identity thereto.

[0309] In certain embodiments, one or more of the linkers that an antibody construct comprises can be an amino acid sequence obtained, derived, or designed from an antibody hinge region sequence. In some embodiments, such linker can have at least one cysteine capable of participating in at least one disulfide bond under physiological conditions or other standard peptide conditions (e.g., peptide purification conditions, conditions for peptide storage, etc.). In certain embodiments, a linker corresponding to, or similar to, an Ig hinge region peptide retains a cysteine that corresponds to the hinge cysteine disposed toward the amino (or N-) terminus of that hinge region. In further embodiments, a linker is derived from an IgGl hinge region and can be modified to remove any one or more of the cysteine residues. In some embodiments, a linker comprises an IgGl hinge region that has one cysteine or two cysteines that correspond to wildtype hinge cysteine residues.

[0310] In certain embodiments, a linker of an antibody construct described herein can comprise an “altered wildtype Ig hinge region” or an “altered Ig hinge region”. Such altered hinge regions can refer to (a) a wild type Ig hinge region with up to 30 percent amino acid changes (or up to 25 percent, 20 percent, 15 percent, 10 percent, or 5 percent amino acid substitutions, insertions or deletions), (b) a portion of a wild type Ig hinge region that is at least 10 amino acids (e.g., at least 12, 13, 14 or 15 amino acids) in length with up to 30 percent amino acid changes (or up to 25 percent, 20 percent, 15 percent, 10 percent, or 5 percent amino acid substitutions or deletions), (c) a portion of a wild type Ig hinge region that comprises the core hinge region, which portion can be 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15, or at least 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15 amino acids in length, or (d) a combination of any of (a)-(c). In certain embodiments, one or more cysteine residues in a wildtype Ig hinge region, such as an IgGl hinge region comprising the upper and core regions, can be substituted by one or more other amino acid residues (e.g., one or more serine residues). An altered Ig hinge region can alternatively or additionally have a proline residue of a wildtype Ig hinge region, such as an IgGl hinge region, substituted by another amino acid residue (e.g., a serine residue).

[0311] In some embodiments, an antibody construct of this disclosure comprises a linkerFab'Fccomprising or consisting of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 56 or 54. In some embodiments, an antibody construct comprises a linkerscFv'Fccomprising or consisting of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 57. In some embodiments, an antibody construct comprises a linkerFab'scFvcomprising or consisting of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence set forth in SEQ ID NO: 58.

[0312] In certain embodiments, a multivalent and multispecific (e.g., trivalent and trispecific) Het- Fab antibody construct of this disclosure can comprise a first linkerFab'Fcthat couples one Fab heavy chain to one Fc polypeptide and a second linkerFab'Fcthat couples the other Fab heavy chain to the other Fc polypeptide. In some embodiments, the first linkerFab'Fccan comprise or consist of the amino acid sequence set forth in SEQ ID NO: 56, and the second linkerFab'Fccan comprise or consist of the amino acid sequence set forth in SEQ ID NO: 54.

[0313] In various embodiments, an antibody construct of this disclosure comprises a linker that couples a Fab domain to a first Fc polypeptide, linkerFab'Fc, wherein the linkerFab'Fccomprises or consists of the amino acid sequence set forth in SEQ ID NO: 56 or 54, and another linker coupling a first or second scFv domain to a first or second Fc polypeptide, linkerscFv'Fc, wherein the linkerscFv'Fccomprises or consists of the amino acid sequence set forth in SEQ ID NO: 57.

[0314] Any of these antibody constructs can further comprise one or more linkerscFvthat couple the two variable domains VH and VL of an scFv domain to one another, wherein such linkerscFvcan comprise or consist of the amino acid sequence set forth in SEQ ID NO: 57, as well as a linkerFab'scFvthat couples the C-terminus of a Fab light chain to the N-terminus of an scFv domain, wherein such linkerFab'scFvcomprises or consists of the amino acid sequence set forth in SEQ ID NO: 58.F. Certain Embodiments of Costimulatory Antibody Constructs

[0315] The present disclosure describes multivalent and multispecific antibody constructs that can have different formats and / or geometries, e.g., spatial molecular structures. The multivalent and multispecific antibody constructs described herein have the ability to engage at least three different antigens, two different antigens on one or more cytotoxic effector cell(s) (e.g., T cell(s)) and adisease-associated antigen, such as TAA on a tumor cell, wherein each of these antigens is engaged at least in a monovalent manner.

[0316] Certain embodiments of the present disclosure relate to multivalent and multispecific antibody constructs that are at least trivalent and trispecific Het-Fab antibody constructs. Such Het- Fab antibody constructs can comprise at least three antigen binding domains, two of which are Fab domains. And since both Fab domains have binding specificities for different antigens, the two Fab domains comprise different heavy and light chain amino acid sequences (e.g., different sequences in the corresponding VH / VL pairs) and thus can be described as hetero-Fabs or in short “Het-Fabs” (e.g., compared to homodimeric Fabs that share heavy and light chains with identical amino acid sequences and target the same antigen, e.g., “Hom-Fabs”).

[0317] In various embodiments, described herein is a multivalent and multispecific Het-Fab antibody construct, comprising: (i) a first Fab domain capable of binding a first antigen on a cytotoxic effector cell; (ii) a first scFv domain capable of binding a second antigen on a cytotoxic effector cell; (iii) athird binding domain capable of binding a disease-associated antigen on a target cell; and (iv) a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the third binding domain is coupled to the N-terminus of the second Fc polypeptide, and (c) the first scFv domain is coupled to the C-terminus of the light chain of the first Fab domain. In some embodiments, the first Fab domain binds CD3 and the first scFv domain binds CD28. In some embodiments, the third binding domain comprises or consists of a second Fab domain or a second scFv domain. In some embodiments, the third binding domain binds a TAA on a tumor cell.

[0318] In various embodiments, described herein is a multivalent and multispecific Het-Fab antibody construct, comprising: (i) a first Fab domain capable of binding a first antigen on a cytotoxic effector cell; (ii) a second Fab domain capable of binding a TAA on a tumor cell; (iii) an scFv domain capable of binding a second antigen on a cytotoxic effector cell; and (iv) a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the second Fab domain is coupled to the N-terminus of the second Fc polypeptide, and (c) the scFv domain is coupled to the C-terminus of a Fab light chain. In some embodiments, the scFv domain is coupled to the C- terminus of the light chain of the first Fab domain.

[0319] In some embodiments, a multivalent and multispecific Het-Fab antibody construct of the present disclosure is capable of monovalently engaging CD3, CD28 and a TAA.

[0320] In some embodiments, a multivalent and multispecific Het-Fab antibody construct comprises: (i) a first Fab domain, wherein the first Fab domain is capable of binding CD3 on a cytotoxic effector cell; (ii) a second Fab domain, wherein the second Fab domain is capable of binding a TAA on the tumor cell; (iii) an scFv domain, wherein the scFv domain is capable of binding CD28 on a cytotoxic effector cell; and (iv) a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N- terminus of the first Fc polypeptide, (b) the second Fab domain is coupled to the N-terminus of the second Fc polypeptide, and (c) the scFv domain is coupled to the C-terminus of the light chain of the first Fab domain.

[0321] In some embodiments, a multivalent and multispecific Het-Fab antibody construct of the present disclosure that is capable of binding CD3, CD28 and a TAA comprises or consists of four polypeptide chains, at least two immunoglobulin heavy chains (e.g., Hl and H2) and at least two immunoglobulin light chains (e.g., LI and L2). In some embodiments, the four polypeptide chains associate, either covalently and / or non-covalently, to form the multivalent and multispecific Het- Fab antibody construct.

[0322] In some embodiments, a multivalent and multispecific Het-Fab antibody construct of the present disclosure comprises a heterodimeric Fc domain. In such embodiments, Hl and / or H2 can comprise one or more amino acid modifications that promote formation of the heavy chain heterodimer H1-H2 compared to formation of the respective homodimers Hl -Hl or H2-H2. In some of these embodiments, and as further described herein, Hl and H2 each comprise a CH3 sequence, wherein each CH3 sequence comprises one or more amino acid substitutions, relative to a respective wildtype CH3 sequence, that promote heterodimer H1-H2 formation compared to homodimer formation.

[0323] In some embodiments of a multivalent and multispecific Het-Fab antibody construct, the CH3 sequence of one of the Fc polypeptides comprises a set of amino acid substitutions selected from the group consisting of: L351Y_F405A_Y407V, T350V_L351Y_F405A_Y407V and T350V_L351Y_S400E_F405A_Y407V, and the CH3 sequence of the other Fc polypeptide comprises a set of amino acid substitutions selected from the group consisting of: T366L K392M T394W, T366L K392L T394W, T350V T366L K392L T394W,T350V_T366L_K392M_T394W and T350V_T366L_N390R_K392M_T394W, and wherein the numbering of amino acid residues in the Fc polypeptides is according to the EU numbering system.

[0324] In some embodiments of a multivalent and multispecific Het-Fab antibody construct, Hl, H2, LI and / or L2 each comprise one or more amino acid modifications that promote correct heavy and light chain pairing to form the desired Hl -LI and H2-L2 heterodimer pairs, when compared to the formation of incorrectly paired heterodimers, e.g., H1L2 and H2L1. In some embodiments, Hl, H2, LI and / or L2 each comprise one or more amino acid substitutions in their CHI and CL sequences, respectively, that promote correct heavy and light chain pairing, and relative to unmodified parental and / or wildtype CHI and CL sequences.

[0325] In some embodiments, a multivalent and multispecific Het-Fab antibody construct of the present disclosure comprises one Fab domain comprising one or more of the amino acid substitutions 143E, 145T, 179E and 228D in the CHI sequence of the heavy chain and one or more of the amino acid substitutions 121K, 124R and 178R in the CL sequence of the light chain, and another Fab domain comprising one or more of the amino acid substitutions 125R and 188K in the CHI sequence of the heavy chain and one or more of the amino acid substitutions 122D, 129T, 176E and 178E in the CL sequence of the light chain, wherein these amino acid substitutions are according to the Kabat numbering scheme and relative to unmodified parental and / or wildtype CHI and CL sequences and promote correct heavy and light chain pairing for the formation of the desired Hl -LI and H2-L2 heterodimers, compared to formation of the incorrectly paired heterodimers H1L2 and H2L1.

[0326] Hence, in various embodiments, a multivalent and multispecific Het-Fab antibody construct of the present disclosure comprises (i) amino acid modification(s) in one or more of its CH3 sequences of the Fc domain to promote correct heavy chain heterodimer pairing (i.e., H1-H2 as compared to H1H1 or H2H2), and in addition to that (ii) amino acid modification(s) in the CHI / CL sequences of its respective Fab domains to promote correct heavy and light chain pairing (i.e., Hl-Ll and H2-L2 as compared to HILI and H2L2). Together, these amino acid modifications can enable correct chain pairing to yield the antibody construct, which chain configuration can be described as: L1HLH2L2.

[0327] In various embodiments, the present disclosure relates to multivalent and multispecific scFv2antibody constructs. Such scFv2antibody constructs can comprise at least three antigen binding domains, at least two of which are scFv domains, thus the descriptor “scFv2”.

[0328] In some embodiments, a multivalent and multispecific scFv2antibody construct of the present disclosure comprises: (i) a Fab domain capable of binding a first antigen on a cytotoxic effector cell; (ii) a first scFv domain capable of binding a TAA on a tumor cell; (iii) a second scFv domain capable of binding a second antigen on a cytotoxic effector cell; and (iv) a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the first scFv domain is coupled to the N-terminus of the second Fc polypeptide, and (c) the second scFv domain is coupled to the C-terminus of the Fab light chain.

[0329] In some embodiments, a multivalent and multispecific scFv2antibody construct of the present disclosure comprises: (i) the Fab domain, wherein the Fab domain is capable of binding CD3 on a cytotoxic effector cell; (ii) the first scFv domain capable of binding a TAA on a tumor cell; (iii) the second scFv domain, wherein the second scFv domain is capable of binding CD28 on a cytotoxic effector cell; and (iv) the dimeric Fc domain comprising the first Fc polypeptide and the second Fc polypeptide, wherein: (a) the Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the first scFv domain is coupled to the N-terminus of the second Fc polypeptide, and (c) the second scFv domain is coupled to the C-terminus of the Fab light chain.

[0330] In some embodiments, a multivalent and multispecific scFv2antibody construct of the present disclosure that is capable of binding CD3, CD28 and a TAA comprises or consists of at least three polypeptide chains, at least two immunoglobulin heavy chains (e.g., Hl and H2) and at least one immunoglobulin light chain (e.g., L). In some embodiments, the three polypeptide chains associate, either covalently and / or non-covalently, to form the multivalent and multispecific scFv2antibody construct.

[0331] In some embodiments, a multivalent and multispecific scFv2antibody construct of the present disclosure comprises a heterodimeric Fc domain. In such embodiments, Hl and / or H2 can comprise one or more amino acid modifications that promote formation of the heterodimer Hl -H2 compared to formation of respective homodimers Hl -Hl or H2-H2. In some of these embodiments, and as further described herein, Hl and H2 each comprise a CH3 sequence, wherein each CH3 sequence comprises one or more amino acid substitutions, relative to a respective wildtype CH3 sequence, that promote heterodimer H1-H2 formation compared to homodimer formation.

[0332] In some embodiments of a multivalent and multispecific scFv2antibody construct, the CH3 sequence of one of the Fc polypeptides comprises a set of amino acid substitutions selected from the group consisting of: L351Y_F405A_Y407V, T350V_L351Y_F405A_Y407V and T350V_L351Y_S400E_F405A_Y407V, and the CH3 sequence of the other Fc polypeptide comprises a set of amino acid substitutions selected from the group consisting of: T366L K392M T394W, T366L K392L T394W, T350V_T366L_K392L_T394W,T350V_T366L_K392M_T394W and T350V_T366L_N390R_K392M_T394W, and wherein the numbering of amino acid residues in the Fc polypeptides is according to the EU numbering system.

[0333] In certain embodiments, described herein is an antibody construct, comprising: (i) a first Fab domain capable of binding CD3 on a cytotoxic effector cell; (ii) a first scFv domain capable of binding CD28 on a cytotoxic effector cell; (iii) a third binding domain capable of binding a TAA on a tumor cell, wherein the third binding domain comprises or consists of a second Fab domain or a second scFv domain; and (iv) a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the third binding domain is coupled to the N-terminus of the second Fc polypeptide, and (c) the first scFv domain is coupled to the C-terminus of the light chain of the first Fab domain. In some embodiments, the third binding domain comprises or consists of a second Fab domain. In other embodiments, the third binding domain comprises or consists of a second scFv domain. The dimeric Fc domain can be a heterodimeric Fc domain, as further described herein. FIGS. 1A (e.g., the third binding domain is a second scFv domain) and IB (e.g., the third binding domain is a second Fab domain) show the format and geometry of such trivalent and trispecific antibody constructs, according to certain embodiments of this disclosure.

[0334] In various embodiments, a multivalent and multispecific antibody construct (e.g., a Het- Fab or scFv2construct) comprises a light chain having the following domain structure, from N- to C-terminus, of VL-CL-scFv, wherein the scFv domain can comprise, from N- to C-terminus, a VH coupled to a VL sequence, or a VL sequence coupled to a VH sequence. Thus, in certain embodiments, such Fab light chain has the domain structure, from N- to C-terminus, of (VL- CL)Fab-(VL-VH)scFv. In other embodiments, the light chain has the domain structure, from N- to C-terminus, of (VL-CL)Fab-(VH-VL)ScFv.

[0335] In certain embodiments, described herein is a multivalent and multispecific antibody construct (e.g., a Het-Fab or scFv2construct) comprising a light chain having the following domainstructure, from N- to C-terminus, of (VL-CL)Fab-(VL-VH)SCFv. In some embodiments, such light chain can further comprise one or more linkers, as further described herein. In some embodiments, the light chain comprises a peptide linkerFab'scFvbetween the Fab portion and the scFv portion which couples the Fab portion to the scFv portion to yield the light chain with the domain structure (VL-CL)Fab-LinkerFab'scFv-(VL-VH)scFv. The linkerFab'scFvcan comprise or consist of the amino acid sequence set forth in SEQ ID NO: 58. As further described herein, the scFv domain of the light chain can also comprise a linkerscFvthat coupled the VL sequence to the VH sequence. Thus, in certain embodiments, the multivalent and multispecific antibody construct comprises a light chain with the domain structure (VL-CL)Fab-LinkerFab'scFv-(VL-LinkerscFv-VH)SCFv. The linkerscFvcan comprise or consist of the amino acid sequence set forth in SEQ ID NO: 59.

[0336] In certain embodiments, described herein is a multivalent and multispecific antibody construct comprising a light chain comprising an amino acid sequence having at least about 80%, 85%, 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence set forth in SEQ ID NO: 13. In some embodiments, the light chain of a multivalent and multispecific antibody construct can comprise an amino acid sequence having the amino acid sequence set forth in SEQ ID NO: 13.

[0337] In various other embodiments, and as further described herein, a Fab light chain that comprises a Fab portion (e.g., a VL and a CL domain) binding CD3 and an scFv portion binding CD28 can have a different amino acid sequence when compared to SEQ ID NO: 13, e.g., can have less than 98%, 95%, 90%, or less than 85% sequence identity to SEQ ID NO: 13, because different combinations of anti-CD3 and / or anti-CD28 binding domain sequences (i.e., paratopes) may be used in the costimulatory constructs described herein. As further described herein, as long as such anti-CD3 and / or anti-CD28 binding domains fulfill certain criteria such as relative CD3 and CD28 binding affinities, induction of CD3 and CD28 engagement of the construct on an immune cell (e.g., T cell), as determined by, e.g., SPR or a comparable method known in the art, various different anti-CD3 and / or anti-CD28 binding domains known in the art may be used in the antibody constructs described herein.G. Certain Properties of Costimulatory Antibody Constructs

[0338] The costimulatory multivalent and multispecific antibody constructs of the present disclosure (e.g., Het-Fab as well as scFv2antibody constructs as described herein) can have several specific properties due to the specific combination of antibody format and antigen bindingaffinities. As disclosed herein, a costimulatory multivalent and multispecific antibody construct comprises a light chain comprising a Fab portion and an scFv portion, as further described herein, and thus can possess certain properties that can be unique to a construct with that specified format (i.e., Format A).

[0339] In some embodiments, binding of a multivalent and multi specific antibody construct to two different antigens on a T cell and to a TAA on a tumor cell, e.g., in a tumor (micro)environment, can be - at least temporarily - simultaneous, thereby establishing a TCR-independent immune synapse, which can direct T cell-mediated cytotoxic activity to a tumor environment that contains tumor cells expressing the TAA. In various embodiments, and as further described herein, a multivalent and multispecific antibody construct may cause a significantly reduced immune cell (e.g., T cell) activation in the absence of the TAA, e.g., when the immune synapse cannot be fully formed due to an absence of the TAA. Such property can be advantageous over those of conventional constructs as it allows, for example, activation of a subject’s immune system in a more TAA-dependent manner and thus may cause less off-target effects in the subject compared to conventional constructs or antibodies that act in a less TAA-dependent manner.

[0340] In various embodiments, and in order to address certain treatment challenges with known and conventional therapeutic T cell engaging antibodies, e.g., durability of response, off-target toxicities and / or sustained T cell activation, the herein disclosed multivalent and multispecific costimulatory T cell engager antibody constructs have been rationally designed to, e.g., optimally engage CD3 and CD28 on a T cell in obligate cis (i.e., on the same T cell), and redirect and enhance cytotoxic T cell responses to TAA-expressing tumor cells while maintaining an improved safety profile when compared to conventional therapeutic modalities.

[0341] In some embodiments, the antibody constructs described herein can posses an enhanced anti-tumor activity in tumors that have relatively low T cell infiltration when compared to conventional constructs that target only one immune cell antigen (e.g., CD3 or CD28, but not both), due to their costimulatory activity by being capable of engaging both CD3 and CD28 on the same immune cell (e.g., T cell).

[0342] In various embodiments, binding of a multivalent and multispecific antibody construct of the present disclosure to CD3 comprises or consists of binding of such construct to CD3e, i.e., the epsilon subunit of the T cell receptor complex.

[0343] Generally, and as it may be appreciated by a person of ordinary skill in the art, natural T cell activation can require both TCR (e.g., involving the CD3 complex) and CD28 stimulation. The antibody constructs of the present disclosure have been specifically designed, e.g., through their format and geometry, to provide both CD3 and CD28 costimulation. Furthermore, and according to various embodiments of the present disclosure, the anti-CD3 and anti-CD28 binding affinities of the anti-CD3 and anti-CD28 binding domains (e.g., scFv’s, Fab’s, etc.) of antibody constructs described herein, in combination with their relative positioning within the construct, can be specifically selected and engineered to generate signals for immune cell (e.g., T cell) activation with appropriate strength to reduce both T cell anergy on the one side and T cell overreaction and dysfunction on the other side of the spectrum, and to provide an improved ratio of anti-tumor / on-target to healthy tissue / off-target activity. In various embodiments, such improved on-target-to-off-target activities can be achieved by optimizing the format and geometry of the antibody constructs in a way such that engagement of all three antigens, namely CD3, CD28 and TAA, at the same time allows for the most potent anti-tumor activity, compared to instances in which, e.g., only CD3 and / or CD28 are engaged and bound by the construct.

[0344] In some embodiments, costimulatory antibody construct of the present disclosure can mediate a sustained T cell-mediated cytotoxicity toward tumor cells over multiple cycles of T cell stimulation. In some embodiments, such antibody construct can be used to stimulate a T cell population about 4, 5 or 6 times over a 14-day period (e.g., first day of stimulation is on day 0) and achieve anti-tumor cell cytotoxicity of at least about 70%, 75%, 80%, 85%, or 90% and T cell viability of at least about 60%, 65%, 70%, 75%, or 80% as measured at certain time points over the 14-day period.

[0345] In some embodiments, a costimulatory antibody construct of the present disclosure does bind CD3 and CD28 on the same cytotoxic effector cells (e.g., T cell) and does not bind CD3 and CD28 on two different effector cells, as measured, e.g., in a cell-cell bridging assay with recorded double positive events of less than about 5, less than about 3, or less than about 2. Hence, in various embodiments, a costimulatory antibody construct of the present disclosure co-engages CD3 and CD28 on the surface of the same cytotoxic effector cell (also referred to herein as “cis” binding of CD3 and CD28). In various embodiments, a costimulatory antibody construct binds CD28 with a KD that is about 2-fold, 4-fold, 6-fold, 8-fold or 10-fold higher compared to that of CD3, when measured using a cytotoxic effector cell that expresses both CD3 and CD28 (and when comparedto a cytotoxic effector cell that expresses CD28 but not CD3), and when measured using SPR or an alternative method known in the art for measuring cell binding of antibody constructs.

[0346] In various embodiments, and without being bound by any theory, a costimulatory antibody construct of the present disclosure can bind CD3 on the surface of an effector cell (e.g., T cell) first via its anti-CD3 Fab domain before CD28 on the same effector cell is bound by the anti-CD28 scFv domain. The binding of a costimulatory antibody construct to CD28 hence can be conditional upon CD3 binding, another property specific to costimulatory antibody constructs described herein that comprise an anti-CD3 Fab domain, and wherein the anti-CD3 Fab light chain is C-terminally coupled to an anti-CD28 scFv domain. A multivalent and multispecific antibody construct of the present disclosure can exhibit conditional binding to CD28, i.e., binds CD28 with a significantly lower dissociation constant (i.e., higher affinity) when co-engaged with CD3, and compared to binding of CD28 alone, e.g., in the absence of CD3 (e.g., when using a CD3 knock-out (KO) cell line). In some embodiments, a multivalent and multi specific antibody construct of the present disclosure binds CD28 with an at least about 2-fold, 3-fold, 5-fold, 7-fold, 8-fold, or at least about 10-fold increased affinity (e.g., expressed in change in KD) when co-engaged with CD3, and when compared to the binding of CD28 in the absence of CD3.

[0347] In some embodiments, a multivalent and multispecific antibody construct of the present disclosure binds a cytotoxic effector cell (e.g., a T cell) that expresses CD3 and CD28 with an affinity (e.g., KD) from about 5 nM to about 100 pM, from about 1 nM to about 100 pM, from about 1 nM to about 250 pM, from about 1 nM to about 500 pM, or from about 1 nM to about 750 pM. In certain embodiments, such antibody construct binds the cytotoxic effector cell that expresses CD3 and CD28 with an affinity that is about 2-fold, 5-fold, 10-fold, 20-fold, 50-fold, 100-fold, or about 200-fold higher, and / or from about 2-fold to about 200-fold, from about 2-fold to about 150-fold, from about 2-fold to about 100-fold, or from about 20-fold to about 200-fold higher than that of a corresponding bispecific anti-CD3xTAA and / or anti-CD28xTAA antibody construct.

[0348] Furthermore, in various embodiments, the multivalent and multispecific antibody constructs of the present disclosure may provide a target cell (e.g., tumor cell) dependent cytotoxicity profile, e.g., as shown herein when cytokine release is significantly reduced in the presence of isolated T cells only compared to conditions in which the T cells are in co-culture with TAA-expressing tumor cells.

[0349] In some embodiments, a multivalent and multispecific antibody construct has a melting temperature at Tml, Tm2, and / or Tm3 that is within 10 °C, 5 °C, 2 °C, or within 1 °C degree of that of a bivalent and monospecific (e.g., monospecific for either CD3, CD28 or TAA) IgGl monoclonal antibody, as determined using, e.g., differential scanning calorimetry (DSC) or differential scanning fluorometry (DSF). In some embodiments, such bivalent and monospecific IgGl monoclonal antibody can be any conventional IgGl antibody capable of binding a specific antigen. In certain embodiments, such monoclonal, monospecific and bivalent IgGl reference antibody is one that comprises two Fab domains as described herein in the context of trispecific antibody constructs. In some embodiments, the bivalent and monospecific IgGl monoclonal antibody comprises two anti-CD3 Fab domains as described herein, two anti-CD28 Fab domains as described herein, or two anti-TAA Fab domains as described herein.

[0350] In some embodiments, the multivalent and multispecific antibody constructs disclosed herein can have a thermal stability when measured at 40 °C and over a time period of about 2, 3, 5, 7, 10, or 14 days of at least about 90%, 95%, 97%, 98% or 99%, i.e., 90%, 95%, 97%, 98% or 99% of intact construct is measured using, e.g., size-exclusion chromatography, or other comparable methods known in the art. In certain embodiments, such constructs can comprise at least one Fab domain capable of binding CD3, and two scFv domains, wherein one such scFv domain is capable of binding CD28, and the other scFv domain is capable of binding a TAA. In other embodiments, such constructs can comprise at least two Fab domains, one of which is capable of binding either CD3 and the other Fab domain is capable of binding a TAA, and one scFv domain capable of binding CD28. In some embodiments, the stability of a trispecific and trivalent antibody construct over a period of 14 days at 40 °C is at least about 97% or 98% percent (i.e., at least about 97% or 98% of construct is intact as measured, e.g., using size-exclusion chromatography). In some embodiments, the concentration of the construct in such stability experiment is about 1 mg / mL.

[0351] In various embodiments, a costimulatory antibody construct comprising a costimulatory portion and a TAA-engaging portion, wherein the costimulatory portion comprises an anti-CD3 Fab domain and an anti-CD28 scFv domain and wherein the anti-CD28 scFv domain is coupled to the C-terminus of the Fab light chain of the anti-CD3 Fab domain, can comprise various different combinations of different anti-CD3, anti-CD28 and anti-TAA binding domain sequences. In some embodiments, the anti-CD3, anti-CD28 and anti-TAA binding domains can have certain ratios ofaffinities for their respective targets, and wherein these affinities can be determined either in a one- armed antibody (OAA) context (e.g., a monovalent, monospecific construct comprising an Fc and a single binding domain (e.g., a Fab or an scFv) against a given antigen), or in a multispecific antibody context (e.g., when using a multivalent and multivalent antibody construct as described herein.

[0352] In some embodiments, the KD ratio of the anti-CD3, anti-CD28 and anti-TAA binding domains for their respective targets can be about (1-1 l)[CD3]:(5-160)[CD28]:(0.01-0.5[TAA], l[CD3]:(l-10)[CD28]:(0.01-l)[TAA], l[CD3]:(l-5)[CD28]:(0.01-l)[TAA], 1[CD3]:(1-10)[CD28]:(0.01-0.5)[TAA] or about l[CD3]:(l-5)[CD28]:(0.01-0.5)[TAA], Such a ratio of antigen affinities of binding domains, in combination with the structure of the costimulatory portion of an antibody construct as described herein, can provide for, e.g., CD3 dependent engagement of CD28 (e.g., resulting in less CD28-dependent toxicities), TAA-driven cytotoxic activity of the construct.III. SEQUENCE IDENTITY OF AMINO ACID AND NUCLEIC ACID SEQUENCES

[0353] As described in other parts of this disclosure, certain embodiments herein relate to an isolated polypeptide or a set of isolated polypeptides (e.g., polypeptide chains Hl, H2, LI, L2 or Hl, H2 and L etc., or portions, e.g., domains, thereof) of a multivalent and multispecific antibody construct, as well as to a polynucleotide or a set of polynucleotides encoding the one or more polypeptide chains of an antibody construct described herein. A polynucleotide in this context can encode all or part of an antibody construct, such as one or more polypeptide chains (e.g., Hl, H2, LI, L2 etc.) of an antibody construct.

[0354] In some embodiments, described herein is a nucleic acid molecule or a set of nucleic acid molecules encoding one or more, two or more, three or more, or four or more polypeptide chains that form any of the multivalent and multispecific antibody constructs disclosed herein.

[0355] The terms “nucleic acid,” “nucleic acid molecule” and “polynucleotide” are used interchangeably herein and refer to a polymeric form of nucleotides of any length, either deoxyribonucleotides or ribonucleotides, or analogues thereof. Non-limiting examples of polynucleotides include a gene, a gene fragment, messenger RNA (mRNA), cDNA, recombinant polynucleotides, plasmids, vectors, isolated DNA of any sequence, isolated RNA of any sequence, nucleic acid probes, and primers.

[0356] In some embodiments, described herein is a vector or a set of vectors comprising the nucleic acid molecule or the set of nucleic acid molecules that encode the one or more polypeptide chains (e.g., one or more of Hl, H2, LI, L2 etc.) of an antibody construct disclosed herein.

[0357] A polynucleotide that “encodes” a given polypeptide is a polynucleotide that is transcribed (in the case of DNA) and translated (in the case of mRNA) into a polypeptide in vivo when placed under the control of appropriate regulatory sequences. The boundaries of the coding sequence are determined by a start codon at the 5' (amino) terminus and a translation stop codon at the 3' (carboxy) terminus. A transcription termination sequence can be located 3' to the coding sequence.

[0358] In certain embodiments, the present disclosure relates to polynucleotide and / or polypeptide sequences that are identical or substantially identical to another polynucleotide and / or polypeptide sequence. The term “identical,” in the context of two or more polynucleotide or polypeptide sequences, refers to two or more sequences or subsequences that are the same, i.e., have the identical sequence of nucleotide or amino acid monomers (i.e., 100% sequence identity), respectively. Polypeptide or polynucleotide sequences herein share “sequence identity” if they have a percentage or a certain number of amino acid residues or nucleotides, respectively, that are at least about 80%, about 85%, about 90%, about 95%, about 97%, or at least about 99% identity over a specified region when compared and aligned for maximum correspondence over a comparison window or over a designated region as measured using one of the commonly used sequence comparison algorithms as known to persons of ordinary skill in the art or by manual alignment and visual inspection. This definition also refers to the complement of a test polynucleotide sequence. The identity can exist over a region that is at least about 50 amino acids or nucleotides in length, or over a region that is from about 75 to about 100 amino acids or nucleotides in length, or, where not specified, across the entire sequence of a polypeptide or polynucleotide. For sequence comparison, typically test sequences are compared to a designated reference sequence. When using a sequence comparison algorithm, test and reference sequences are entered into a computer, subsequence coordinates are designated, if necessary, and sequence algorithm program parameters are designated. Default program parameters can be used, or alternative parameters can be designated. The sequence comparison algorithm then calculates the percent (%) sequence identities for the test sequences relative to the reference sequence, based on the program parameters.

[0359] The term “comparison window,” as used herein, refers to a segment of a sequence comprising contiguous amino acid or nucleotide positions which can be from about 20 to about 1000 contiguous amino acid or nucleotide positions, for example from about 50 to about 600 or from about 100 to about 300 or from about 150 to about 200 contiguous amino acid or nucleotide positions over which a test sequence can be compared to a reference sequence of the same number of contiguous positions after the two sequences are optimally aligned. Longer segments up to and including the full-length sequence may also be used as a comparison window in certain embodiments. Methods of alignment of sequences for comparison are known to those of ordinary skill in the art. Optimal alignment of sequences for comparison can be conducted, for example, by the local homology algorithm of Smith & Waterman, 1970, Adv. Appl. Math., 2:482c; by the homology alignment algorithm of Needleman & Wunsch, 1970, J. Mol. Biol., 48:443; by the search for similarity method of Pearson & Lipman, 1988, Proc. Natl. Acad. Sci. USA, 85:2444, or by computerized implementations of these algorithms (for example, GAP, BESTFIT, FASTA or TFASTA in the Wisconsin Genetics Software Package, Genetics Computer Group, Madison, WI), or by manual alignment and visual inspection (see, for example, Ausubel et al., Current Protocols in Molecular Biology, (1995 supplement), Cold Spring Harbor Laboratory Press). Examples of available algorithms suitable for determining percent sequence identity are the BLAST and BLAST 2.0 algorithms, which are described in Altschul et al., 1997, Nuc. Acids Res., 25:3389- 3402, and Altschul et al., 1990, J. Mol. BioL, 215:403-410, respectively. Software for performing BLAST analyses is publicly available through the website for the National Center for Biotechnology Information (NCBI).

[0360] Certain embodiments described herein relate to variant sequences (e.g., variant VH domains, variant Fc polypeptides, etc.) that comprise one or more amino acid modification, e.g., one or more amino acid insertions, one or more amino acid deletions, and / or one or more amino acid substitutions, when compared to, e.g., a reference such as a wildtype sequence (e.g., a sequence that is disclosed in a known reference). In certain embodiments, the one or more amino acid modification of a variant sequence comprises one or more amino acid substitutions when compared to a reference such as a wildtype sequence. In some embodiments, the one or more amino acid substitutions are one or more non-conservative substitutions. In other embodiments, the one or more amino acid substitutions are one or more conservative substitutions.

[0361] In general, a “conservative substitution,” as used herein, is considered to be a substitution of one amino acid with another amino acid having similar physical, chemical and / or structural properties. Common conservative substitutions are listed under Column 1 of TABLE 4.TABLE 4: Conservative Amino Acid Substitutions

[0362] One skilled in the art will appreciate that the main factors in determining what constitutes a conservative substitution are usually the size of the amino acid side chain and its physical / chemi cal properties, but that certain environments allow for substitution of a given amino acid with a broader range of amino acids than those listed in Column 1 of TABLE 4. These additional amino acids tend to either have similar properties to the amino acid being substituted but to vary more widely in size or be of similar size but vary more widely in phy si cal / chemi calproperties. This broader range of conservative substitutions is listed under Column 2 of TABLE 4. The skilled person can readily ascertain the most appropriate group of substituents to select from in view of the particular protein environment in which the amino acid substitution is being made.IV. PHARMACEUTICAL COMPOSITIONS

[0363] In certain embodiments, the present disclosure relates to pharmaceutical compositions that can comprise one or more of the multivalent and multispecific antibody constructs described herein. In various embodiments, a pharmaceutical composition herein can further comprise a pharmaceutically acceptable excipient, carrier, buffer, stabiliser, or other materials well known to those skilled in the art. Such materials are generally non-toxic and do not interfere with the efficacy of the active ingredient (i.e., antibody construct). The precise nature of a carrier or other material can depend on the route of administration. Hence, a pharmaceutical composition herein can be formulated for various used and administration routes, e.g., for oral, intravenous, cutaneous, subcutaneous, nasal, intramuscular, or intraperitoneal administration routes.

[0364] A pharmaceutical composition for oral administration can be in tablet, capsule, powder, or liquid form. A tablet can include a solid carrier such as gelatin or an adjuvant. Liquid pharmaceutical compositions generally include a liquid carrier such as water, petroleum, animal or vegetable oils, mineral oil, or synthetic oil. Physiological saline solution, dextrose or other saccharide solution or glycols such as ethylene glycol, propylene glycol or polyethylene glycol can be included.

[0365] For intravenous, cutaneous, or subcutaneous injection, or injection at the site of affliction (e.g., at a tumor site), the active ingredient (i.e., antibody construct) can be in the form of a parenterally acceptable aqueous solution which is pyrogen-free and has suitable pH, isotonicity and stability. Those of relevant skill in the art are well able to prepare suitable solutions using, for example, isotonic vehicles such as Sodium Chloride Injection, Ringer's Injection, Lactated Ringer's Injection. Preservatives, stabilisers, buffers, antioxidants and / or other additives can be included, as required.

[0366] For antibody constructs according to the present disclosure that are administered to a subject, administration is preferably in a “therapeutically effective amount” that is sufficient to show benefit to the individual, as further described herein. The actual amount administered, and rate and time-course of administration, can depend on the nature and severity of the disease (e.g., cancer) being treated. Prescription of treatment, e.g., decisions on dosage etc., is within theresponsibility of general practitioners and other medical doctors, and typically takes account of the disorder to be treated, the condition of the individual patient, the site of delivery, the method of administration and other factors known to practitioners. Examples of the techniques and protocols mentioned above can be found in Remington's Pharmaceutical Sciences, 16th edition, Osol, A. (ed), 1980.

[0367] In some embodiments, a pharmaceutical composition can comprise a second active ingredient (e.g., another protein or small molecule) in addition to an antibody construct described herein.

[0368] Hence, also described herein is a pharmaceutical composition comprising any one or more of the multivalent and multispecific antibody construct(s) disclosed herein, and a pharmaceutically acceptable carrier, excipient, diluent, or combination thereof.V. KITS

[0369] The present disclosure also describes kits comprising one or more of the multivalent and multispecific antibody constructs described herein, or a pharmaceutical composition as described herein and that comprises such antibody construct(s), as well as instructions for use. Thus, in certain embodiments, described herein are kits comprising vectors for expressing an antibody construct described herein and instructions for use. In certain embodiments, described herein are kits comprising host cells comprising a vector for expressing an antibody construct and instructions for use. In some embodiments, the present disclosure relates to kits comprising a purified antibody construct and instructions for use. The purified antibody construct can be lyophilized or provided in a dry form, such as a powder or granules, and the kit can additionally contain a suitable solvent for reconstitution of the lyophilized or dried component(s).

[0370] A kit can further comprise a container and a label and / or package insert on or associated with the container. The label or package insert contains instructions customarily included in commercial packages of therapeutic products, providing information or instructions about the indications, usage, dosage, administration, contraindications and / or warnings concerning the use of such therapeutic products (e.g., an antibody construct described herein). The label or package insert can further include a notice in the form prescribed by a governmental agency regulating the manufacture, use or sale of pharmaceuticals or biological products, which notice reflects approval by the agency of manufacture, for use or sale for human or animal administration. The container can hold a composition comprising an antibody construct of this disclosure. In some embodiments,the container can have a sterile access port. For example, the container can be an intravenous solution bag or a vial having a stopper that can be pierced by a hypodermic injection needle.

[0371] In addition to the container containing a composition comprising an antibody construct, the kit can further comprise one or more additional containers comprising other components of the kit. For example, a pharmaceutically acceptable buffer (such as bacteriostatic water for injection) (BWFI), phosphate-buffered saline, Ringer's solution, or dextrose solution), or other buffers or diluents can be included in such kit.

[0372] Suitable containers can include, for example, bottles, vials, syringes, intravenous solution bags, and the like. The containers can be formed from a variety of materials such as glass or plastic. If appropriate, one or more components (e.g., an antibody construct) of the kit can be lyophilized or provided in a dry form, such as a powder or granules, and the kit can additionally contain a suitable solvent for reconstitution of the lyophilized or dried component(s).

[0373] A kit herein can further include other materials desirable from a commercial or user standpoint, such as filters, needles, and syringes.VI. METHODS

[0374] Further described herein are methods of producing and using the multivalent and multispecific antibody constructs of the present disclosure.A. Methods of Producing an Antibody Construct

[0375] In some embodiments, the present disclosure relates to methods for preparing the multivalent and multispecific antibody constructs described herein. In various embodiments, an antibody construct of the present disclosure can be produced using standard recombinant methods known in the art (see, for example, U.S. Patent No. 4,816,567 and “ Antibodies: A Laboratory Manual ” 2ndEdition, Ed. Greenfield, Cold Spring Harbor Laboratory Press, New York, 2014).

[0376] For recombinant production of an antibody construct described herein, a polynucleotide or set of polynucleotides encoding the antibody construct can be generated and inserted into one or more vectors for further cloning and / or expression in a host cell. Polynucleotide(s) encoding the antibody construct can be produced by standard methods known in the art (see, for example, Ausubel et al., Current Protocols in Molecular Biology, John Wiley & Sons, New York, 1994 & update, and “ Antibodies: A Laboratory Manual,” 2ndEdition, Ed. Greenfield, Cold Spring Harbor Laboratory Press, New York, 2014). As would be appreciated by one of skill in the art, the number of polynucleotides required for expression of the antibody construct may be dependent on theformat and / or geometry of the antibody construct, including, for example, the number of polypeptide chains that the antibody construct is comprised of. For example, when an antibody construct comprises three polypeptide chains (e.g., Hl, H2 and L), three polynucleotides each encoding one polypeptide chain can be used, and when an antibody construct comprises four polypeptide chains (e.g., Hl, H2, LI and L2), four polynucleotides each encoding one polypeptide chain can be used. In embodiments in which two or more polynucleotides are used, such two or more polynucleotides can be incorporated into one vector or into more than one vector (e.g., two or three separate vectors).

[0377] Generally, for expression, the polynucleotide or set of polynucleotides encoding an antibody construct herein can be incorporated into an expression vector together with one or more regulatory elements, such as transcriptional elements, which can be used for efficient transcription of the polynucleotide(s). Examples of such regulatory elements include, but are not limited to, promoters, enhancers, terminators, and polyadenylation signals. One skilled in the art will appreciate that the choice of regulatory elements can be dependent on the host cell selected for expression of the polypeptides of the antibody construct and that such regulatory elements can be derived from a variety of sources, including bacterial, fungal, viral, mammalian or insect genes. The expression vector can optionally further contain heterologous nucleic acid sequences that facilitate expression or purification of the expressed protein. Examples include, but are not limited to, signal peptides and affinity tags such as metal-affinity tags, histidine tags, avidin / streptavidin encoding sequences, glutathione-S-transferase (GST) encoding sequences and biotin encoding sequences. The expression vector can be an extrachromosomal vector or an integrating vector. Hence, in some embodiments, the amino acid sequences of the polypeptide chains of an expressed antibody construct described herein, e.g., chains Hl, H2, LI and / or L2 etc., can comprise a signal peptide sequence. Such signal peptide sequences may vary depending on the expression system and conditions used for producing an antibody construct. Exemplary signal peptide sequences can comprise the amino acid sequence METDTLLLWVLLLWVPGSTG (SEQ ID NO: 79) or MRPTWAWWLFLVLLLALWAPARG (SEQ ID NO: 80), e g., for Hl, H2, etc., or MRPTWAWWLFLVLLLALWAPARG (SEQ ID NO: 80) or MGWSCIILFLVATATGVHS (SEQ ID NO: 81), e.g., for LI, L2, etc. In certain embodiments, one or more heavy chains (e.g., Hl, H2, etc.) of an antibody construct described herein can comprise a C-terminal lysine residue following expression of the polypeptide chains inside the cell. In various embodiments, such C-terminal lysine residue may be enzymatically cleaved from the polypeptide chains prior to further processing (e.g., purification, formulation, etc.) and prior to use of the corresponding antibody construct, e.g., prior to administration of the construct to a subject in need thereof.

[0378] Certain embodiments for producing an antibody construct of the present disclosure relate to vectors (such as expression vectors) comprising one or more polynucleotides encoding at least a portion of an antibody construct described herein. The polynucleotide(s) can be comprised by a single vector or by more than one vector. In some embodiments, the polynucleotides are comprised by a multi-cistronic vector. Expression vectors that can be used to express polynucleotides include but are not limited to pTT5 and pUC15 cells comprising vectors encoding an antibody construct.

[0379] Suitable host cells for cloning or expression of the antibody construct polypeptides include various prokaryotic or eukaryotic cells as known in the art. Eukaryotic host cells include, for example, mammalian cells, plant cells, insect cells and yeast cells (such as Saccharomyces or Pichia cells). Prokaryotic host cells include, for example, E. coli, A. salmonicida or B. subtilis cells. In certain embodiments, an antibody construct can be produced in bacteria, in particular when glycosylation and Fc effector function are not needed or desired for the indented purpose of the antibody construct, as described for example in U.S. Patent Nos. 5,648,237, 5,789,199, and 5,840,523, and in Charlton, Methods in Molecular Biology, Vol. 248, pp. 245-254, B.K.C. Lo, ed., Humana Press, Totowa, N.J., 2003. Eukaryotic microbes such as filamentous fungi or yeast are suitable expression host cells in certain embodiments, in particular fungi and yeast strains whose glycosylation pathways have been “humanized” resulting in the production of an antibody with a partially or fully human glycosylation pattern (see, for example, Gemgross, 2004, Nat. Biotech. 22: 1409-1414, and Li et al., 2006, Nat. Biotech. 24:210-215).

[0380] Suitable host cells for the expression of glycosylated antibody constructs are, in various embodiments, eukaryotic cells. For example, U.S. Patent Nos. 5,959,177, 6,040,498, 6,420,548, 7,125,978 and 6,417,429 describe PLANTIBODIES™ technology for producing antibodies and portions thereof (e.g., scFv(s), Fab(s), etc.) in transgenic plants. Mammalian cell lines adapted to grow in suspension are particularly useful for the expression of antibody constructs described herein. Examples include, but are not limited to, monkey kidney CV1 line transformed by SV40 (COS-7), human embryonic kidney (HEK) line 293 or 293 cells (see, for example, Graham et al., 1977, J. Gen Virol., 36:59), baby hamster kidney cells (BHK), mouse sertoli TM4 cells (see, for example, Mather, 1980, Biol Reprod, 23:243-251); monkey kidney cells (CV1), African greenmonkey kidney cells (VERO-76), human cervical carcinoma (HeLa) cells, canine kidney cells (MDCK), buffalo rat liver cells (BRL 3 A), human lung cells (W138), human liver cells (Hep G2), mouse mammary tumour (MMT 060562), TRI cells (see, for example, Mather etal., 1982, Annals N.Y. AcadSci, 383:44-68), MRC 5 cells, FS4 cells, Chinese hamster ovary (CHO) cells (including DHFR CHO cells, see Urlaub et al., 1980, Proc Natl Acad Sci USA, 77:4216), and myeloma cell lines (such as Y0, NSO and Sp2 / 0). Exemplary mammalian host cell lines suitable for production of antibodies are reviewed in Yazaki & Wu, Methods in Molecular Biology, Vol. 248, pp. 255-268 (B.K.C. Lo, ed., Humana Press, Totowa, N.J., 2003).

[0381] In certain embodiments, the host cell used to produce a multivalent and multispecific antibody construct herein is a transient or stable higher eukaryotic cell line, such as a mammalian cell line. In some embodiments, the host cell is a mammalian HEK293T, CHO, HeLa, NSO or COS cell. In some embodiments, the host cell is a stable cell line that allows for mature glycosylation of the antibody construct.

[0382] The host cells comprising the expression vector(s) encoding the antibody construct can be cultured using routine methods to produce the antibody construct. Alternatively, in some embodiments, host cells comprising the expression vector(s) encoding the antibody construct can be used therapeutically or prophylactically to deliver the antibody construct to a subject, or polynucleotides or expression vectors can be administered to a cell from a subject ex vivo and the cell then returned to the body of the subject.

[0383] In some embodiments, a host cell comprises (for example, has been transformed with) a vector comprising a polynucleotide that encodes a VL and a VH of a binding domain of an antibody construct described herein. In some embodiments, a host cell comprises (for example, has been transformed with) a vector comprising a polynucleotide that encodes a full-length polypeptide chain of an antibody construct described herein, e.g., Hl, H2, LI and / or L2 as described herein. In another example, a host cell comprises a first vector comprising a polynucleotide that encodes the VL of a binding domain and a second vector comprising a polynucleotide that encodes the corresponding VH of the binding domain. In various embodiments, the host cell is eukaryotic, for example, a Chinese Hamster Ovary (CHO) cell, a human embryonic kidney (HEK) cell or a lymphoid cell (e.g., Y0, NSO, Sp20 cell). In certain embodiments, the host cell is Expi293™ (Thermo Fisher, Waltham, MA). In certain embodiments, the host cell used herein is CHO-S cells (National Research Council Canada) or HEK293 cells.

[0384] Certain embodiments of the present disclosure relate to a method of making an antibody construct comprising culturing a host cell into which one or more polynucleotides encoding the antibody construct, or one or more expression vectors encoding the antibody construct, have been introduced, under conditions suitable for expression of the antibody construct. Such method can further comprise recovering the antibody construct from the host cell (or from host cell culture medium). In some embodiments, such method can further comprise purifying the antibody construct.

[0385] Cell culture media that can be used include, but are not limited to, DMEM (Thermo Fisher, Waltham, MA), Opti-MEM™ (Thermo Fisher, Waltham, MA), Opti-MEM™ I Reduced Serum Medium (Thermo Fisher, Waltham, MA), RPMI-1640 medium, Expi293™ Expression Medium (Thermo Fisher, Waltham, MA), and FreeStyle CHO expression medium (Thermo Fisher Scientific, Waltham, MA). The cell culture medium can be supplemented with serum, e.g., fetal bovine serum (FBS), amino acids, e.g., L-glutamine, antibiotics, e.g., penicillin, and streptomycin, and / or antimycotics, e.g., amphotericin, or any other supplements routinely used in the to support cell culture.

[0386] In various embodiments, an antibody construct of the present disclosure is purified after expression. Proteins, such as an antibody construct of the present disclosure, can be isolated or purified in a variety of ways known to those skilled in the art (see, for example, Protein Purification: Principles and Practice, 3rdEd., Scopes, Springer-Verlag, NY, 1994). Standard purification methods that can be used for the antibody constructs disclosed herein include chromatographic techniques, including ion exchange, hydrophobic interaction, affinity, sizing or gel filtration, and reverse-phase, carried out at atmospheric pressure or at high pressure using systems such as FPLC and HPLC. Additional purification methods include electrophoretic, immunological, precipitation, dialysis and chromatofocusing techniques. Ultrafiltration and diafiltration techniques, in conjunction with protein concentration, can also be used. As is well known in the art, a variety of natural proteins bind Fc domains and other structural elements of an antibody construct, and, in some embodiments, these proteins can be used for purification of an antibody construct. For example, the bacterial proteins A and G can bind to the Fc domain of some antibody constructs. Likewise, the bacterial protein L can bind to the Fab domain of some antibody constructs. Purification can also be enabled by a particular fusion partner. For example, antibody constructs can be purified using glutathione resin if a GST fusion is employed, Ni+2affinitychromatography if a His-tag is employed or immobilized anti-flag antibody if a flag-tag is used. The degree of purification necessary may vary depending on the use of the antibody constructs. Hence, in some embodiments, no purification may be necessary.

[0387] In certain embodiments, an antibody construct of this disclosure is substantially pure. The term “substantially pure” (or “substantially purified”) when used in reference to an antibody construct described herein, refers to an antibody construct as substantially or essentially free of components that normally accompany or interact with the protein as found in its naturally occurring environment, such as a native cell, or a host cell in the case of a recombinantly produced antibody construct. In certain embodiments, an antibody construct that is substantially pure is an antibody construct purified to have less than about 30%, less than about 25%, less than about 20%, less than about 15%, less than about 10%, less than about 5%, or less than about 2% (by dry weight) of other contaminating protein species.

[0388] Assessment of antibody construct purity and / or homogeneity can be performed by any method known in the art, including, but not limited to, non-reducing / reducing CE-SDS, non- reducing / reducing SDS-PAGE, Ultra-high performance liquid chromatography-size exclusion chromatography (UPLC-SEC), High Performance Liquid Chromatography (HPLC), mass spectrometry, multi angle light scattering (MALS), and dynamic light scattering (DLS).

[0389] In certain embodiments, an antibody construct described herein can comprise one or more post-translational modifications. Such post-translational modifications can occur in vivo, or they be conducted in vitro after isolation of the antibody construct from the host cell.

[0390] Post-translational modifications can include various modifications as are known in the art (see, for example, Proteins - Structure and Molecular Properties, 2nd Ed., T. E. Creighton, W. H. Freeman and Company, New York, 1993; Post-Translational Covalent Modification of Proteins, B. C. Johnson, Ed., Academic Press, New York, pgs. 1-12, 1983; Seifter et al., 1990, Meth. Enzymol., 182:626-646, and Rattan et al., 1992, Ann. N.Y. Acad. Sci., 663:48-62). In those embodiments in which an antibody construct comprises one or more post-translational modifications, the antibody construct can comprise the same type of modification at one or several sites (e.g., amino acid residues), or it can comprise different modifications at different sites.

[0391] Examples of post-translational modifications can include glycosylation, acetylation, phosphorylation, amidation, derivatization by known protecting / blocking groups, formylation,oxidation, reduction, proteolytic cleavage or specific chemical cleavage by cyanogen bromide, trypsin, chymotrypsin, papain, V8 protease or NaBEU.

[0392] Other examples of post-translational modifications can include, for example, addition or removal of N-linked or O-linked carbohydrate chains, chemical modifications of N-linked or O- linked carbohydrate chains, processing of N-terminal or C-terminal ends, attachment of chemical moieties to the amino acid backbone, and addition or deletion of an N-terminal methionine residue resulting from prokaryotic host cell expression. Post-translational modifications can also include modification with a detectable label, such as an enzymatic, fluorescent, isotopic or affinity label to allow for detection and isolation of the protein. Examples of suitable enzyme labels include, but are not limited to, horseradish peroxidase, alkaline phosphatase, beta-galactosidase and acetylcholinesterase. Examples of suitable prosthetic group complexes include, but are not limited to, streptavidin / biotin and avidin / biotin. Examples of suitable fluorescent materials include, but are not limited to, umbelliferone, fluorescein, fluorescein isothiocyanate, rhodamine, dichlorotriazinylamine fluorescein, dansyl chloride and phycoerythrin. An example of a luminescent material is luminol, examples of bioluminescent materials include luciferase, luciferin and aequorin, and examples of suitable radioactive materials include iodine, carbon, sulfur, tritium, indium, technetium, thallium, gallium, palladium, molybdenum, xenon, and fluorine.

[0393] Additional examples of post-translational modifications can include acylation, ADP- ribosylation, amidation, covalent attachment of flavin, covalent attachment of a heme moiety, covalent attachment of a nucleotide or nucleotide derivative, covalent attachment of a lipid or lipid derivative, covalent attachment of phosphotidylinositol, cross-linking, cyclization, disulfide bond formation, demethylation, formation of covalent cross-links, formation of cysteine, formation of pyroglutamate, gamma-carboxylation, GPI anchor formation, hydroxylation, iodination, methylation, myristylation, pegylation, prenylation, racemization, selenoylation, sulfation, transfer-RNA mediated addition of amino acids to proteins such as arginylation, and ubiquitination.

[0394] In some embodiments, described herein is a method of producing a multivalent and multispecific construct of the present disclosure, the method comprising: (a) obtaining a host cell culture comprising at least one host cell comprising one or more nucleic acid molecules encoding one or more, two or more, three or more, or four or more polypeptide chains that form the antibody construct; and (b) recovering the antibody construct from the host cell culture. In someembodiments, such method can further comprise, subsequent to step (b), purifying the antibody construct.B. Methods of Using an Antibody Construct of the Present Disclosure

[0395] In certain embodiments, the present disclosure relates to methods of using multivalent and multispecific antibody construct of the present disclosure. In some embodiments, described herein are methods of using an antibody construct described herein for the treatment of a disease or condition in a subject in need thereof.

[0396] Such method can comprise administering a multivalent and multispecific antibody construct, or a pharmaceutical composition comprising such antibody construct, to a subject in need thereof. In certain embodiments, the subject is a mammal. In some embodiments, the subject is a human.

[0397] In some embodiments, the present disclosure relates to a method of treating a cancer, an autoimmune disease, or an inflammatory condition in a subject in need thereof, the method comprising administering to the subject a multivalent and multispecific antibody construct of the present disclosure, or a pharmaceutical composition comprising such antibody construct, and wherein the antibody construct comprises a binding domain capable of binding an antigen associated with a cancer, an autoimmune disease, or an inflammatory condition.

[0398] In various embodiments, the method is directed to treating a cancer in a subject in need thereof. Cancers that can be treated using the methods and antibody constructs disclosed herein can include, but are not limited to, hematologic neoplasms (including leukemias, myelomas and lymphomas), carcinomas (including adenocarcinomas and squamous cell carcinomas), melanomas and sarcomas. Carcinomas and sarcomas are also frequently referred to as “solid tumors”. In certain embodiments, the cancer is a solid tumor.

[0399] In certain embodiments, the present disclosure relates to a method of treating a solid tumor in a subject in need thereof, the method comprising administering to the subject a multivalent and multispecific antibody construct of the present disclosure, or a pharmaceutical composition comprising such antibody construct. In some embodiments, solid tumors include carcinomas, sarcomas, and lymphomas. A solid tumor target that can be engaged by an antibody construct of the present disclosure include any known solid tumor target, including MSLN, Cldnl8.2, DLL3, HER2, GPC3, amongst others.

[0400] In certain embodiments, the present disclosure relates to a method of treating a liquid tumor in a subject in need thereof, the method comprising administering to the subject a multivalent and multispecific antibody construct of the present disclosure, or a pharmaceutical composition comprising such antibody construct. In some embodiments, the liquid tumor can be a leukemia.

[0401] When used in a method described herein, an antibody construct of this disclosure can exert either a cytotoxic or cytostatic effect that can result in one or more of a reduction in the size of a tumor, the slowing or prevention of an increase in the size of a tumor, an increase in the disease- free survival time between the disappearance or removal of a tumor and its reappearance, prevention of an initial or subsequent occurrence of a tumor (e.g., metastasis), an increase in the time to progression, reduction of one or more adverse symptom(s) associated with a tumor, an increase in the overall survival time of a subject having a tumor, or a combination of the above.

[0402] The methods described herein can comprise administering a multivalent and multispecific antibody construct to a subject in need thereof. An antibody construct can be administered to a subject by any appropriate route of administration. As will be appreciated by the person of skill in the art, the route and / or mode of administration can vary depending upon the desired therapeutic results. In various embodiments, antibody constructs of this disclosure can be administered by systemic administration or local administration. Local administration can be at the site of a tumor or into a tumor draining lymph node. Generally, the antibody constructs can be administered by parenteral administration, for example, by intravenous, intramuscular, intradermal, intraperitoneal, subcutaneous, or spinal administration, such as by injection or infusion.

[0403] A treatment (e.g., of a cancer, an autoimmune and / or inflammatory condition in a subject) can be achieved by administration of a therapeutically effective amount of a multivalent and multispecific antibody construct to a subject in need thereof. A “therapeutically effective amount,” as used herein, generally refers to an amount of an antibody construct described herein that is effective, at dosages and for periods of time necessary, to achieve a desired therapeutic result. A therapeutically effective amount can vary according to factors such as the disease state, age, sex, and weight of the subject. A therapeutically effective amount is also one in which any potential toxic or detrimental effects of the antibody constructs are outweighed by the therapeutically beneficial effects. “Sufficient amount” generally refers to an amount sufficient to produce a desired effect, e.g., an amount sufficient to generate an anti-tumor immune response to a target (e.g.,tumor) cell or tissue, e.g., by engaging an immune cell (e.g., T cell) using a multivalent and multispecific antibody construct described herein.

[0404] A suitable dosage of a multivalent and multispecific antibody construct described herein can be determined by a skilled medical practitioner. The selected dosage level may depend upon a variety of pharmacokinetic factors including the activity (e.g., antigen affinity(ies)) of the particular antibody construct employed, the route of administration, the time of administration, the rate of excretion of the construct, the duration of the treatment, other drugs, compounds and / or materials used in combination with the antibody construct, e.g., anti-cancer agents, the age, sex, weight, condition, general health and prior medical history of the subject being treated, and like factors well known in the medical arts.

[0405] In some embodiments, a method of treating a disease (e.g., a cancer, an autoimmune and / or inflammatory condition) in a subject comprises administering a second active ingredient (e.g., another protein or small molecule) in addition to an antibody construct described herein. Such second active ingredient can be administered simultaneously or sequentially with an antibody construct dependent upon the condition to be treated.

[0406] In some embodiments, the present disclosure relates to a method of eliciting an anti-tumor immune response in a cell population comprising immune cells expressing CD3 and CD28 and tumor cells expressing a TAA, the method comprising contacting the cell population with an effective amount of a multivalent and multispecific antibody construct of the present disclosure. In some embodiments, such multivalent and multispecific antibody construct binds CD3 and CD28 on one or more cytotoxic effector cell, such as an immune cell, and the TAA on a tumor cell, thereby forming a TCR-independent immune synapse, and the antibody construct is either an scFv2antibody construct as described herein, comprising at least: (i) a first Fab domain capable of binding CD3 on an immune cell; (ii) a first scFv domain capable of binding CD28 on an immune cell, (iii) a second scFv domain capable of binding the TAA on a tumor cell; and (iv) a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the first scFv domain is coupled to the C-terminus of the Fab light chain, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide, or the antibody construct is a Het-Fab antibody construct as described herein, comprising: (i) a first Fab domain capable of binding CD3 on an immune cell; (ii) a second Fab domain capable of binding the TAA on a tumor cell; (iii) afirst scFv domain capable of binding CD28 on an immune cell; and (iv) a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the second Fab domain is coupled to the N-terminus of the second Fc polypeptide, and (c) the scFv domain is coupled to the C-terminus of the light chain of the first Fab domain.

[0407] In some embodiments, the present disclosure relates to a method of inhibiting the proliferation of tumor cells expressing the TAA in a cell population comprising the tumor cells and immune cells expressing CD3 and CD28, the method comprising contacting the cell population with an effective amount of a multivalent and multispecific antibody construct of the present disclosure. In some embodiments, such multivalent and multispecific antibody construct binds CD3 and CD28 on one or more immune cell(s) and the TAA on a tumor cell, thereby forming a TCR-independent immune synapse, and the antibody construct is either an scFv2antibody construct as described herein, comprising: (i) a first Fab domain capable of binding CD3 on an immune cell; (ii) a first scFv domain capable of binding CD28 on an immune cell, (iii) a second scFv domain capable of binding the TAA on a tumor cell; and (iv) a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the first scFv domain is coupled to the C-terminus of the Fab light chain, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide, or the antibody construct is a Het-Fab antibody construct as described herein, comprising: (i) a first Fab domain capable of binding CD3 on an immune cell; (ii) a second Fab domain capable of binding the TAA on a tumor cell; (iii) a first scFv domain capable of binding CD28 on an immune cell; and (iv) a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N- terminus of the first Fc polypeptide, (b) the second Fab domain is coupled to the N-terminus of the second Fc polypeptide, and (c) the first scFv domain is coupled to the C-terminus of the light chain of the first Fab domain.

[0408] In some embodiments, the present disclosure relates to a method of killing tumor cells expressing a TAA, such method comprising contacting a cell population comprising the tumor cells and immune cells expressing CD3 and CD28 with an effective amount of a multivalent and multispecific antibody construct of the present disclosure. In various embodiments, such multivalent and multispecific antibody construct binds CD3 and CD28 on one or more immunecell(s) and the TAA on a tumor cell, thereby forming a TCR-independent immune synapse, and the antibody construct is either an scFv2antibody construct as described herein, comprising: (i) a first Fab domain capable of binding CD3 on an immune cell; (ii) a first scFv domain capable of binding CD28 on an immune cell, (iii) a second scFv domain capable of binding the TAA on a tumor cell; and (iv) a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the first scFv domain is coupled to the C-terminus of the Fab light chain, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide, or the antibody construct is a Het-Fab antibody construct as described herein, comprising: (i) a first Fab domain capable of binding CD3 on an immune cell; (ii) a second Fab domain capable of binding the TAA on a tumor cell; (iii) a first scFv domain capable of binding CD28 on an immune cell; and (iv) a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the second Fab domain is coupled to the N-terminus of the second Fc polypeptide, and (c) the scFv domain is coupled to the C-terminus of the light chain of the first Fab domain.

[0409] In various embodiments, the TAA is expressed by a solid tumor. The solid tumor can be a carcinoma, a sarcoma, or a lymphoma.

[0410] In other embodiments, the TAA is expressed by a liquid tumor such as a leukemia.

[0411] In any of the methods described herein, the immune cell is a T cell, and an immune cell population can comprise or consist of one or more type(s) of T cells.

[0412] As described herein, in various embodiments, an antibody construct herein can be administered to a subject in need thereof, for example, a subject suffering from a cancer, an autoimmune disease and / or an inflammatory condition, in order to modulate an immune response in the subject. The immune response that can be modulated using an antibody construct of this disclosure can be an anti-tumor immune response in the subject, e.g., in various embodiments, such modulated immune response can occur locally at a tumor site. Thus, in certain embodiments, an antibody construct described herein can initiate and / or upregulate a local immune response, e.g., an anti -tumor response of a subject’s immune system in order to elicit a localized cytotoxic effect against the tumor at the tumor site.

[0413] In various embodiments, an antibody construct described herein, e.g., a multivalent and multispecific antibody construct capable of monovalent binding of CD3 (e.g., via one Fab domain),monovalent binding of CD28 (e.g., via one scFv domain), and monovalent binding of a TAA (e.g., via an scFv or Fab domain), can have a broader therapeutic window, compared to comparable conventional molecules, and can allow administration of higher doses of the herein described constructs, leading potentially to increased anti-tumor effects without inducing, or inducing lower grades of, side effects and / or off-target effects. Such broader therapeutic window can be due to certain properties of the antibody constructs described herein, including higher ratios of anti -tumor activity compared to cytokine induction, i.e., higher tumor cell killing activities can be achieved at lower cytokine induction levels.

[0414] In some embodiments, the present disclosure relates to a method of inhibiting the growth of a TAA-expressing tumor and / or reducing the volume of the tumor in a subject in need thereof, the method comprising administering to the subject an effective amount of a multivalent and multispecific antibody construct of the present disclosure. In various embodiments, such in vivo anti-tumor effect is elicited by the multivalent and multispecific antibody construct that binds (e.g., simultaneously) CD3 and CD28 on one or more immune cell(s) and the TAA on a tumor cell, thereby forming a TCR-independent immune synapse, and the antibody construct is either an scFv2antibody construct as described herein, comprising: (i) a first Fab domain capable of binding CD3 on an immune cell; (ii) a first scFv domain capable of binding CD28 on an immune cell, (iii) a second scFv domain capable of binding the TAA on a tumor cell; and (iv) a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the first scFv domain is coupled to the C-terminus of the Fab light chain, and (c) the second scFv domain is coupled to the N- terminus of the second Fc polypeptide, or the antibody construct is a Het-Fab antibody construct as described herein, comprising: (i) a first Fab domain capable of binding CD3 on an immune cell; (ii) a second Fab domain capable of binding the TAA on a tumor cell; (iii) a first scFv domain capable of binding CD28 on an immune cell; and (iv) a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the second Fab domain is coupled to the N-terminus of the second Fc polypeptide, and (c) the scFv domain is coupled to the C-terminus of the light chain of the first Fab domain.

[0415] In various embodiments, and as further described herein, inhibition of tumor growth and / or a reduction in tumor volume in a subject can be elicited by simultaneous binding of the antibodyconstruct to CD3 and CD28 on one or more immune cell(s) and to a TAA on a tumor cell and formation of a TCR-independent artificial immune synapse within a tumor environment in the subject, thereby eliciting an anti -turn or cytotoxic effect mediated by the activated immune cell and directed against the tumor cell. The immune cell can be a T cell.

[0416] In various embodiments of the methods described herein, the multivalent and multispecific antibody constructs binds CD3 and CD28 on the same immune cell (e.g., T cell). In other embodiments, the multivalent and multispecific antibody constructs binds CD3 on a first immune cell (e.g., T cell) and CD28 on a second immune cell (e.g., T cell), wherein the first and the second immune cells are different cells.

[0417] In certain other embodiments, a method herein is directed to treating an autoimmune disease in a subject in need thereof.

[0418] In yet other embodiments, a method herein is directed to treating an inflammatory condition in a subject in need thereof.

[0419] In various embodiments of the methods described herein, the subject is a rodent, a nonhuman primate, or a human.

[0420] In further embodiments, and in relation to a method described herein, administration of a sufficient amount of a multivalent and multispecific antibody construct to a subject in need thereof can provide one or more of the following to activate or upregulate an immune response in the subject: (i) modulation of T-cell receptor signaling, (ii) modulation of T-cell activation, (iii) modulation of pro-inflammatory cytokines, (iv) modulation of interferon-y (IFNy) production by T cells, (v) modulation of T-cell suppression, (vi) modulation of M2-type tumor associated macrophages (TAM) or myeloid-derived suppressor cell (MDSC) survival and / or differentiation, and / or (vii) modulation of cytotoxic or cytostatic effects on cells, e.g., cancer cells.

[0421] In some embodiments, the present disclosure relates to methods of modulating an immune response in a cell population or in a subject using one or more of the multivalent and multispecific antibody con struct / s) of the present disclosure, wherein such modulation can comprise one or more of (i) immune cell activation, (ii) stimulation of T-cell receptor signaling, (iii) stimulation of antibody-dependent cellular cytotoxicity (ADCC), (iv) T cell-dependent cytotoxicity (TDCC), (v) cell-dependent cytotoxicity (CDC), (vi) antibody-dependent cellular phagocytosis (ADCP), and combinations of the above. As described herein, in certain embodiments, an antibody construct of the present disclosure activates T effector cells. In some embodiments, and as demonstrated herein,an antibody construct increases production of one or more cytokines and / or signalling molecules, such as GM-CSF, TNFa, a MIP-1, IFN-y, IL-2, IL-12, IL-17, IL-21 and / or C-X-C motif ligand 13 (CXCL13) by T effector cells.

[0422] As described herein, in various embodiments, a multivalent and multispecific antibody construct of the present disclosure comprises an Fc domain comprising a first and a second Fc polypeptide, wherein one or more of the Fc polypeptides can comprise a modified CH2 domain (e.g., compared to a WT domain) that comprises one or more amino acid modifications that can result in a decrease or elimination of binding of the Fc domain to one or more, or to all of the Fey receptors (also referred to herein as an Fc “knock-out” or “KO” variant).C. Experimental Methods

[0423] In some embodiments, the present disclosure relates to experimental methods for analyzing and / or detecting a multivalent and multispecific antibody construct of the present disclosure. Such methods can be used, for example, to assess in vitro and / or in vivo properties of such antibody construct such as its pharmacokinetic (PK) and pharmacodynamic (PD) properties. Other properties and characteristics of an antibody construct can be evaluated such as its stability under certain conditions (e.g., temperature, pH, etc.), its solubility, or its behaviour in the presence of certain other chemical components such as other proteins or cells.

[0424] Specific binding of a multivalent and multispecific antibody construct described herein to an antigen (e.g., CD3, CD28, TAA) can be measured, for example, through an enzyme-linked immunosorbent assay (ELISA), a surface plasmon resonance (SPR) technique (employing, for example, a BIAcore instrument) (Liljeblad et al., 2000, Glyco J, 17:323-329), or a traditional binding assay (Heeley, 2002, Endocr Res, 28:217-229). In certain embodiments, specific binding is defined as the extent of binding to an unrelated protein being less than about 10% of the binding to the target antigen (e.g., CD3, CD28, TAA, etc.) as measured by SPR, for example.

[0425] Additional experimental methods are described in, e.g., EXAMPLES 1-58 herein.VII. CERTAIN EMBODIMENTS OF THE DISCLOSURE

[0426] Certain embodiments of the present disclosure relate to embodiments 1-143 below, as well as to any combination of one or more of embodiments 1-143.

[0427] Embodiment 1. An antibody construct, comprising: (i) a first immunoglobulin (Ig) heavy chain, Hl, comprising, from N- to C-terminus, a first Fab heavy chain coupled to a first Fc polypeptide, (ii) a first Ig light chain, LI, comprising, from N- to C-terminus, a first Fab light chaincoupled to a first scFv domain, and (iii) a second Ig heavy chain, H2, comprising a second Fc polypeptide, wherein: (a) the first Fab heavy chain of Hl and the first Fab light chain of LI form a first Fab domain that is capable of binding CD3 on a cytotoxic effector cell; (b) the first scFv domain is capable of binding CD28 on a cytotoxic effector cell; and (c) the first Fc polypeptide and the second Fc polypeptide form a dimeric Fc domain.

[0428] Embodiment 2. The antibody construct of embodiment 1, wherein the antibody construct further comprises a binding domain coupled to the N-terminus of the second Fc polypeptide, wherein the binding domain is capable of binding a disease-associated antigen, optionally wherein the disease-associated antigen is a TAA on a tumor cell.

[0429] Embodiment 3. The antibody construct of embodiment 1 or embodiment 2, wherein Hl further comprises a first linkerFab'Fcthat couples the first Fab heavy chain to the first Fc polypeptide.

[0430] Embodiment 4. The antibody construct of embodiments 3, wherein the first linkerFab'Fccomprises or consists of an IgGl, IgG2 or IgG4 hinge region, or are a derivative thereof, wherein the derivative has at least about 80%, 85%, 90%, 93%, 95%, 97%, or 99% sequence identity to a wildtype IgGl, IgG2 or IgG4 hinge region.

[0431] Embodiment 5. The antibody construct of any one of embodiments 1-4, wherein LI further comprises a linkerFab'scFvthat couples the first Fab light chain to the first scFv domain.

[0432] Embodiment 6. The antibody construct of any one of embodiments 1-5, wherein the first scFv domain comprises a first LinkerscFv(LinkerscFvl) and has the domain structure, from N- to C- terminus, of VL-LinkerscFvl-VH.

[0433] Embodiment 7. The antibody construct of any one of embodiments 2-6, wherein the binding domain is a second Fab domain.

[0434] Embodiment 8. The antibody construct of embodiment 7, wherein H2 further comprises a second Fab heavy chain that is coupled to the N-terminus of the second Fc polypeptide and a second Ig light chain, L2, comprising a second Fab light chain, and wherein the second Fab heavy chain of H2 and the second Fab light chain of L2 form the second Fab domain.

[0435] Embodiment 9. The antibody construct of embodiment 8, wherein H2 further comprises a second linkerFab'Fccomprising or consisting of an IgGl, IgG2 or IgG4 hinge region, or are a derivative thereof, wherein the derivative has at least about 80%, 85%, 90%, 93%, 95%, 97%, or 99% sequence identity to a wildtype IgGl, IgG2 or IgG4 hinge region.

[0436] Embodiment 10. The antibody construct of any one of embodiments 7-9, wherein the first Fab domain, the second Fab domain, or both Fab domains comprise one or more amino acid substitutions relative to a reference Fab sequence that promote correct heavy and light chain pairing.

[0437] Embodiment 11. The antibody construct of any one of embodiments 7-10, wherein the first Fab domain and the second Fab domain (i) each comprise a kappa light chain, (ii) each comprise a lambda light chain, or (iii) one Fab domain comprises a kappa light chain and the other Fab domain comprises a lambda light chain.

[0438] Embodiment 12. The antibody construct of any one of embodiments 2-6, wherein the binding domain is a second scFv domain comprising a second VH sequence and a second VL sequence.

[0439] Embodiment 13. The antibody construct of embodiment 12, wherein H2 further comprises the second scFv domain coupled to the N-terminus of the second Fc polypeptide.

[0440] Embodiment 14. The antibody construct of embodiment 12 or embodiment 13, wherein H2 further comprises a first linkerscFv'Fcthat couples the second scFv domain to the second Fc polypeptide.

[0441] Embodiment 15. The antibody construct of embodiment 14, wherein the first linkerscFv'Fccomprises or consists of an IgGl, IgG2 or IgG4 hinge region, or are a derivative thereof, wherein the derivative has at least about 80%, 85%, 90% or 95% sequence identity to a wildtype IgGl, IgG2 or IgG4 hinge region.

[0442] Embodiment 16. The antibody construct of any one of embodiments 12-15, wherein the second scFv domain comprises a second LinkerscFv(LinkerscFv2) and has the domain structure, from N- to C-terminus, of VH-LinkerscFv2-VL.

[0443] Embodiment 17. The antibody construct of embodiment 6 or embodiment 16, wherein the LinkerscFvland the LinkerscFv2each independently comprise or consist of the amino acid sequence of (G4S)X, wherein x is 1, 2, 3, 4 or 5 (SEQ ID NO: 63).

[0444] Embodiment 18. An antibody construct, comprising: (i) a first Fab domain capable of binding CD3 on a cytotoxic effector cell; (ii) a first scFv domain capable of binding CD28 on a cytotoxic effector cell; (iii) a third binding domain capable of binding a TAA on a tumor cell; and (iv) a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide, (b) the thirdbinding domain is coupled to the N-terminus of the second Fc polypeptide, and (c) the first scFv domain is coupled to the C-terminus of the light chain of the first Fab domain.

[0445] Embodiment 19. The antibody construct of embodiment 18, wherein the antibody construct comprises a first Ig heavy chain (Hl) comprising a first Fab heavy chain comprising a first VH sequence and a first CHI sequence and the first Fc polypeptide, a second Ig heavy chain (H2) comprising the second Fc polypeptide, and a first Ig light chain (LI) comprising a first Fab light chain and the first scFv domain, wherein the first Fab heavy chain and the first Fab light chain form the first Fab domain.

[0446] Embodiment 20. The antibody construct of embodiment 19, wherein LI comprises, from N- to C-terminus, the first Fab light chain, comprising a first VL sequence and a first CL sequence, coupled to the first scFv domain, wherein the first scFv domain comprises, from N- to C-terminus, either (i) a third VH sequence coupled to a third VL sequence or (ii) a third VL sequence coupled to a third VH sequence.

[0447] Embodiment 21. The antibody construct of embodiment 19 or embodiment 20, wherein LI further comprises a linkerFab'scFvthat couples the first Fab light chain to the first scFv domain.

[0448] Embodiment 22. The antibody construct of any one of embodiments 18-21, wherein the first scFv domain further comprises a first LinkerscFv(LinkerscFvl) and has the domain structure, from N- to C-terminus, of VL- LinkerscFvl-VH.

[0449] Embodiment 23. The antibody construct of any one of embodiments 19-22, wherein the third binding domain is a second Fab domain.

[0450] Embodiment 24. The antibody construct of embodiment 23, wherein the second Ig heavy chain (H2) further comprises a second Fab heavy chain comprising a second VH sequence and a second CHI sequence, and the antibody construct further comprises a second Ig light chain (L2) comprising a second Fab light chain comprising a second VL sequence and a second CL sequence, and wherein the second Fab heavy chain and the second Fab light chain form the second Fab domain.

[0451] Embodiment 25. The antibody construct of embodiment 23 or embodiment 24, wherein H2 further comprises a second linkerFab'Fcthat couples the C-terminus of the second Fab heavy chain to the N-terminus of the second Fc polypeptide.

[0452] Embodiment 26. The antibody construct of any one of embodiments 20-25, wherein the first linkerFab'Fc, the second linkerFab'Fc, or both, are polypeptide linkers each independently comprising or consisting of about 5 to about 50 consecutive amino acid residues.

[0453] Embodiment 27. The antibody construct of embodiment 26, wherein the first linkerFab'Fcand the second linkerFab'Fcindependently comprise or consists of an IgGl, IgG2 or IgG4 hinge region, or a derivative thereof, wherein the derivative has at least about 80%, 85%, 90%, 93%, 95%, 97%, or 99% sequence identity to a wildtype IgGl, IgG2 or IgG4 hinge region.

[0454] Embodiment 28. The antibody construct of any one of embodiments 23-27, wherein the first Fab domain, the second Fab domain, or both Fab domains comprise one or more amino acid substitutions relative to a reference Fab sequence that promotes correct heavy and light chain pairing.

[0455] Embodiment 29. The antibody construct of any one of embodiments 23-28, wherein the first Fab domain and the second Fab domain (i) each comprise a kappa light chain, (ii) each comprise a lambda light chain, or (iii) one Fab domain comprises a kappa light chain and the other Fb domain comprises a lambda light chain.

[0456] Embodiment 30. The antibody construct of any one of embodiments 19-22, wherein the third binding domain is a second scFv domain comprising a second VH sequence and a second VL sequence.

[0457] Embodiment 31. The antibody construct of embodiment 30, wherein H2 further comprises a linkerscFv'Fcthat couples the second scFv domain to the second Fc polypeptide.

[0458] Embodiment 32. The antibody construct of embodiment 31, wherein the linkerscFv'Fccomprises or consists of an IgGl, IgG2 or IgG4 hinge region, or a derivative thereof, wherein the derivative has at least about 80%, 85%, 90% or 95% sequence identity to a wildtype IgGl, IgG2 or IgG4 hinge region.

[0459] Embodiment 33. The antibody construct of any one of embodiments 30-32, wherein the second scFv domain comprises a second LinkerscFv(LinkerscFv2) and has the domain structure, from N- to C-terminus, of VH- LinkerscFv2-VL.

[0460] Embodiment 34. The antibody construct of embodiment 22 or embodiment 33, wherein the LinkerscFvland the LinkerscFv2each independently comprise or consist of the amino acid sequence of (G4S)X, wherein x is 1, 2, 3, 4 or 5 (SEQ ID NO: 63).

[0461] Embodiment 35. An antibody construct comprising two immunoglobulin heavy chains, Hl and H2, and two immunoglobulin light chains, LI and L2, wherein: (i) Hl comprises, from N- to C-terminus, a first Fab heavy chain comprising a first VH sequence and a first CHI sequence, coupled to a first Fc polypeptide; (ii) H2 comprises, from N- to C-terminus, a second Fab heavy chain comprising a second VH sequence and a second CHI sequence, coupled to a second Fc polypeptide; (iii) LI comprises, from N- to C-terminus, a first Fab light chain comprising a first VL sequence and a first CL sequence, coupled to a first scFv domain comprising a third VL sequence coupled to a third VH sequence; and (iv) L2 comprises, from N- to C-terminus, a second Fab light chain comprising a second VL sequence and a second CL sequence, wherein: (a) the first Fab heavy chain of Hl and the first Fab light chain of LI form a first Fab domain capable of binding CD3 on a cytotoxic effector cell; (b) the second Fab heavy chain of H2 and the second Fab light chain of L2 form a second Fab domain capable of binding a TAA on a tumor cell; (c) the first scFv domain is capable of binding CD28 on a cytotoxic effector cell; and (d) the first Fc polypeptide and the second Fc polypeptide form a dimeric Fc domain.

[0462] Embodiment 36. The antibody construct of embodiment 35, wherein Hl further comprises a first linkerFab'Fcthat couples the first Fab heavy chain to the first Fc polypeptide, and H2 further comprises a second linkerFab'Fcthat couples the second Fab heavy chain to the second Fc polypeptide.

[0463] Embodiment 37. The antibody construct of embodiment 35 or embodiment 36, wherein the first Fab domain, the second Fab domain, or both Fab domains comprise one or more amino acid substitutions relative to a reference Fab sequence that promotes correct heavy and light chain pairing.

[0464] Embodiment 38. The antibody construct of any one of embodiments 35-37, wherein the first Fab domain and the second Fab domain (i) each comprise a kappa light chain, (ii) each comprise a lambda light chain, or (iii) one Fab domain comprises a kappa light chain and the other Fb domain comprises a lambda light chain.

[0465] Embodiment 39. An antibody construct comprising two immunoglobulin heavy chains, Hl and H2, and one immunoglobulin light chain, LI, wherein: (i) Hl comprises, from N- to C- terminus, a first Fab heavy chain comprising a first VH sequence and a first CHI sequence, coupled to a first Fc polypeptide; (ii) LI comprises, from N- to C-terminus, a first Fab light chain comprising a first VL sequence and a first CL sequence, coupled to a first scFv domain comprisinga second VL sequence coupled to a second VH sequence; and (iii) H2 comprises, from N- to C- terminus, a second scFv domain comprising a third VL sequence and a third VH sequence, coupled to a second Fc polypeptide; wherein: (a) the first Fab heavy chain of Hl and the first Fab light chain of LI form a first Fab domain capable of binding CD3 on a cytotoxic effector cell; (b) the first scFv domain is capable of binding CD28 on a cytotoxic effector cell; (c) the second scFv domain is capable of binding a TAA on a tumor cell; and (d) the first Fc polypeptide and the second Fc polypeptide form a dimeric Fc domain.

[0466] Embodiment 40. The antibody construct of embodiment 39, wherein Hl further comprises a first linkerFab'Fcthat couples the first Fab heavy chain to the first Fc polypeptide, and H2 further comprises a linkerscFv'Fcthat couples the second scFv domain to the second Fc polypeptide.

[0467] Embodiment 41. The antibody construct of embodiment 36 or embodiment 40, wherein one or more of the first linkerFab'Fc, the second linkerFab'Fc, and the linkerscFv'Fcindependently comprise or consists of an IgGl, IgG2 or IgG4 hinge region, or a derivative thereof, wherein the derivative has at least about 80%, 85%, 90% or 95% sequence identity to a wildtype IgGl, IgG2 or IgG4 hinge region.

[0468] Embodiment 42. The antibody construct of any one of embodiments 35-41, wherein LI further comprises a linkerFab'scFvthat couples the first Fab light chain to the first scFv domain.

[0469] Embodiment 43. The antibody construct of any one of embodiments 1-42, wherein the first Fc polypeptide comprises a first CH2 sequence and a first CH3 sequence and the second Fc polypeptide comprises a second CH2 sequence and a second CH3 sequence.

[0470] Embodiment 44. The antibody construct of any one of embodiments 1-43, wherein the dimeric Fc domain is a heterodimeric Fc domain.

[0471] Embodiment 45. The antibody construct of embodiment 43 or embodiment 44, wherein at least one of the first and second CH2 sequences comprises one or more amino acid substitutions relative to a corresponding wildtype CH2 sequence that reduce or ablate binding of the antibody construct to one or more Fey receptors.

[0472] Embodiment 46. The antibody construct of any one of embodiments 43-45, wherein at least one of the first and second CH3 sequences comprises one or more amino acid substitutions relative to a corresponding wildtype CH3 sequence that promotes formation of a heterodimeric Fc domain compared to a formation of a homodimeric Fc domain.

[0473] Embodiment 47. The antibody construct of embodiment 46, wherein the CH3 sequence of one of the Fc polypeptides comprises a set of amino acid substitutions selected from the group consisting of: L351Y_F405A_Y407V, T350V_L351Y_F405A_Y407V andT350V_L351Y_S400E_F405A_Y407V, and the CH3 sequence of the other Fc polypeptide comprises a set of amino acid substitutions selected from the group consisting of: T366L K392M T394W, T366L K392L T394W, T350V_T366L_K392L_T394W,T350V_T366L_K392M_T394W and T350V_T366L_N390R_K392M_T394W, and wherein the numbering of amino acid residues in the Fc polypeptides is according to the EU numbering system.

[0474] Embodiment 48. The antibody construct of any one of embodiments 1-47, wherein the antibody construct binds CD3 with a KD from about 0.1 nM to about 100 nM, from about 1 nM to about 100 nM, or from about 10 nM to about 75 nM, optionally, wherein CD3 comprises the CD3 epsilon chain.

[0475] Embodiment 49. The antibody construct of any one of embodiments 1-48, wherein the first Fab domain that is capable of binding CD3 comprises HCDR1-3 as set forth in SEQ ID NOs: 49-52 and LCDR1-3 as set forth in SEQ ID NOs: 53-55.

[0476] Embodiment 50. The antibody construct of any one of embodiments 1-49, wherein the first Fab domain that is capable of binding CD3 comprises a VH sequence having at least about 90%, 95%, 97%, or 99% sequence identity to any one of the sequences set forth in SEQ ID NOs: 1, 2, 82, 84 or 86, and a VL sequence having at least about 90%, 95%, 97%, or 99% sequence identity to any one of the sequences set forth in SEQ ID NOs: 5, 83, 85 or 87.

[0477] Embodiment 51. The antibody construct of any one of embodiments 1-50, wherein the antibody construct binds CD28 with a KD from about 1 nM to about 100 nM or from about 10 nM to about 75 nM.

[0478] Embodiment 52. The antibody construct of any one of embodiments 1-51, wherein the first scFv domain that is capable of binding CD28 comprises HCDR1-3 as set forth in SEQ ID NOs: 30, 34 and 40, and LCDR1-3 as set forth in SEQ ID NOs: 43, 47 and 48.

[0479] Embodiment 53. The antibody construct of any one of embodiments 1-52, wherein the first scFv domain that is capable of binding CD28 comprises a VH sequence having at least about 90%, 95%, 97%, or 99% sequence identity to any one of the sequences set forth in SEQ ID NOs: 15, 18, 20, 21, 22, 23, 24, 25, 27 and 76, and a VL sequence having at least about 90%, 95%, 97%,or 99% sequence identity to any one of the sequences set forth in SEQ ID NOs: 16, 17, 19, 26 and 77.

[0480] Embodiment 54. The antibody construct of any one of embodiments 1-53, wherein the antibody construct does not bind CD3 and CD28 on two different cytotoxic effector cells as measured in a cell-cell bridging assay with recorded double positive events of less than about 5, less than about 3, or less than about 2.

[0481] Embodiment 55. The antibody construct of embodiment 54, wherein the antibody construct co-engages CD3 and CD28 on the surface of the same cytotoxic effector cell.

[0482] Embodiment 56. The antibody construct of any one of embodiments 1-55, wherein the antibody construct binds CD28 with a KD that is about 2-fold, 4-fold, 6-fold, 8-fold or 10-fold higher on a cytotoxic effector cell that expresses both CD3 and CD28, compared to a cytotoxic effector cell that expresses CD28 but not CD3, and when measured using SPR.

[0483] Embodiment 57. The antibody construct of any one of embodiments 1-56, wherein the antibody construct binds the TAA with a KD from about 0.01 nM to about 10 nM, from about 0.05 nM to about 5 nM, or from about 0.1 nM to about 5 nM.

[0484] Embodiment 58. The antibody construct of any one of embodiments 1-57, wherein the antibody construct is capable of forming a TCR-independent immune synapse by binding CD3 and CD28 on the surface of the cytotoxic effector cell and the TAA on the tumor cell.

[0485] Embodiment 59. The antibody construct of any one of embodiments 1-58, wherein the antibody construct comprises a fourth binding domain.

[0486] Embodiment 60. The antibody construct of any one of embodiments 1-59, wherein the cytotoxic effector cell is an immune cell, optionally wherein the immune cell is a T cell.

[0487] Embodiment 61. The antibody construct of any one of embodiments 1-60, wherein the TAA is expressed by a solid tumor.

[0488] Embodiment 62. The antibody construct of any one of embodiments 1-61, wherein the solid tumor is a carcinoma, a sarcoma, or a lymphoma.

[0489] Embodiment 63. A pharmaceutical composition comprising the antibody construct of any one of embodiments 1-62, and a pharmaceutically acceptable carrier, excipient, diluent, or combination thereof.

[0490] Embodiment 64. A nucleic acid molecule or a set of nucleic acid molecules encoding one or more, two or more, three or more, or four or more polypeptide chains that form the antibody construct of any one of embodiments 1-62.

[0491] Embodiment 65. A vector or a set of vectors comprising the nucleic acid molecule or the set of nucleic acid molecules of embodiment 64.

[0492] Embodiment 66. A cell comprising the nucleic acid molecule or the set of nucleic acid molecules of embodiment 64, or the vector or set of vectors of embodiment 65.

[0493] Embodiment 67. A method of producing an antibody construct of any one of embodiments 1-62, the method comprising: (a) obtaining a host cell culture comprising at least one host cell comprising one or more nucleic acid molecules encoding one or more, two or more, three or more, or four or more of the polypeptide chains that form the antibody construct; and (b) recovering the antibody construct from the host cell culture.

[0494] Embodiment 68. The method of embodiment 67, further comprising, subsequent to step (b), purifying the antibody construct.

[0495] Embodiment 69. A method of eliciting an anti-tumor immune response in a cell population comprising immune cells and tumor cells, the method comprising contacting the cell population with an effective amount of the antibody construct of any one of embodiments 1-62, wherein the immune cells express CD3 and CD28 and the tumor cells express a TAA.

[0496] Embodiment 70. A method of inhibiting the proliferation of tumor cells, the method comprising contacting a cell population comprising the tumor cells and immune cells with an effective amount of the antibody construct of any one of embodiments 1-62, wherein the immune cells express CD3 and CD28 and the tumor cells express a TAA.

[0497] Embodiment 71. A method of killing tumor cells, the method comprising contacting a cell population comprising the tumor cells and immune cells with an effective amount of the antibody construct of any one of embodiments 1-62, wherein the immune cells express CD3 and CD28 and the tumor cells express a TAA.

[0498] Embodiment 72. The method of any one of embodiments 69-71, wherein the immune cells comprise T cells.

[0499] Embodiment 73. The method of any one of embodiments 69-72, wherein the antibody construct binds (i) CD3 and CD28 on the same immune cell and (ii) the TAA on a tumor cell.

[0500] Embodiment 74. The method of embodiment 73, wherein the binding of the antibody construct to CD3 and CD28 on the immune cell and the TAA on the tumor cell forms a TCR- independent artificial immune synapse between the immune cell and the tumor cell, thereby eliciting a cytotoxic immune response of the immune cell against the tumor cell.

[0501] Embodiment 75. The method of any one of embodiments 69-74, wherein the cell population is within a subject.

[0502] Embodiment 76. A method of treating a tumor in a subject in need thereof, the method comprising administering to the subject an antibody construct of any one of embodiments 1-62.

[0503] Embodiment 77. The method of embodiment 76, wherein a cytotoxic immune response against the tumor is elicited in the subject, thereby treating the tumor in the subject.

[0504] Embodiment 78. The method of embodiment 76 or embodiment 77, wherein the tumor is a solid tumor.

[0505] Embodiment 79. An antibody construct of any one of embodiments 1-62 for use in the treatment of a tumor, optionally wherein the tumor is a sol...

Claims

1. CLAIMSWHAT IS CLAIMED IS:

1. An antibody construct, comprising:(i) a first Fab domain capable of binding CD3 on a cytotoxic effector cell,(ii) a first scFv domain capable of binding CD28 on a cytotoxic effector cell,(iii) a third binding domain capable of binding a disease-associated antigen, and(iv) a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein:(a) the first Fc polypeptide and / or the second Fc polypeptide comprise one or more amino acid modifications that promote formation of the heterodimeric Fc domain;(b) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide;(c) the first scFv domain is coupled to the C-terminus of the light chain of the first Fab domain; and(d) the third binding domain is coupled to the N-terminus of the second Fc polypeptide.

2. The antibody construct of claim 1, wherein the first Fc polypeptide and / or the second Fc polypeptide comprise one or more amino acid modifications that reduce binding to at least one FcyR.

3. An antibody construct, comprising:(i) a first Fab domain capable of binding CD3 on a cytotoxic effector cell,(ii) a first scFv domain capable of binding CD28 on a cytotoxic effector cell,(iii) a third binding domain capable of binding a disease-associated antigen, and(iv) a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein:(a) the first Fc polypeptide and / or the second Fc polypeptide comprise one or more amino acid modifications that reduce binding to at least one FcyR;(b) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide;(c) the first scFv domain is coupled to the C-terminus of the light chain of the first Fab domain; and(d) the third binding domain is coupled to the N-terminus of the second Fc polypeptide.

4. The antibody construct of claim 3, wherein the dimeric Fc domain is a heterodimeric Fc domain and the first Fc polypeptide and / or the second Fc polypeptide comprise one or more amino acid modifications that promote formation of the heterodimeric Fc domain.

5. The antibody construct of any one of claims 1-4, wherein the third binding domain comprises or consists of a second Fab domain or a second scFv domain.

6. The antibody construct of any one of claims 1-5, wherein the disease-associated antigen is a tumor-associated antigen (TAA).

7. The antibody construct of claim 6, wherein the TAA is an antigen associated with a solid tumor.

8. The antibody construct of claim 6, wherein the TAA is an antigen associated with a liquid tumor.

9. The antibody construct of any one of claims 6-8, wherein the KD ratio of the three binding domains for their respective targets is about (1-1 l)[CD3]:(5-160)[CD28]:(0.01- 0.5[TAA], l[CD3]:(l-10)[CD28]:(0.01-l)[TAA] or l[CD3]:(l-5)[CD28]:(0.01- 0.5)[TAA],10. The antibody construct of any one of claims 1-9, wherein the first scFv domain has the domain structure, from N- to C-terminus, of VH-LinkerscFvl-VL or VL-LinkerscFvl-VH, and wherein the LinkerscFvlcomprises or consists of the amino acid sequence set forth in SEQ ID NO: 63.

11. The antibody construct of any one of claims 1-10, wherein the first scFv domain has the domain structure, from N- to C-terminus, of VL-LinkerscFvl-VH.

12. The antibody construct of any one of claims 1-11, wherein the antibody construct comprises at least, from N- to C-terminus, (i) a first immunoglobulin (Ig) heavy chain (Hl) comprising a first Fab heavy chain comprising a first VH sequence and a first CHI sequence, coupled to the first Fc polypeptide, (ii) a second Ig heavy chain (H2)comprising the second Fc polypeptide, and (iii) a first Ig light chain (LI) comprising a first Fab light chain comprising a first VL sequence and a first CL sequence, coupled to the first scFv domain, wherein the first Fab heavy chain of Hl and the first Fab light chain of LI form the first Fab domain.

13. The antibody construct of claim 12, wherein LI further comprises a linkerFab'scFvthat couples the first Fab light chain to the first scFv domain.

14. The antibody construct of claim 13, wherein the linkerFab'scFvcomprises or consists of the amino acid sequence set forth in SEQ ID NO: 58, or an amino acid sequence having one or more amino acid substitutions thereto.

15. The antibody construct of any one of claims 12-14, wherein Hl further comprises a first linkerFab'Fcthat couples the first Fab heavy chain to the first Fc polypeptide.

16. The antibody construct of claims 15, wherein the first linkerFab'Fccomprises or consists of an IgGl hinge region, or are a derivative thereof, wherein the linkerFab'Fchas at least about 80%, 85%, 90%, 93%, 95%, 97%, 99% or 100% sequence identity to the sequence set forth in SEQ ID NO: 54 or 56.

17. The antibody construct of any one of claims 1-16, wherein the first Fc polypeptide comprises a first CH2 sequence and a first CH3 sequence and the second Fc polypeptide comprises a second CH2 sequence and a second CH3 sequence.

18. The antibody construct of claim 17, wherein the first CH2 sequence and / or the second CH2 sequence comprise one or more amino acid modifications that reduce binding to at least one FcyR.

19. The antibody construct of claim 18, wherein the first CH2 sequence and / or the second CH2 sequence comprise one or more of the amino acid substitutions L234A, L235A and D265S, and wherein the numbering of amino acid residues in the Fc polypeptides is according to the EU numbering system.

20. The antibody construct of any one of claims 17-19, wherein the first CH3 sequence and / or the second CH3 sequence comprise one or more amino acid modifications that promote formation of a heterodimeric Fc domain when compared to the formation of a homodimeric Fc domain.

21. The antibody construct of claim 20, wherein the CH3 sequence of one of the Fc polypeptides comprises a set of amino acid substitutions selected from the groupconsisting of: L351Y_F405A_Y407V, T350V_L351Y_F405A_Y407V and T350V_L351Y_S400E_F405A_Y407V, and the CH3 sequence of the other Fc polypeptide comprises a set of amino acid substitutions selected from the group consisting of: T366L K392M T394W, T366L K392L T394W, T350V_T366L_K392L_T394W, T350V_T366L_K392M_T394W and T350V_T366L_N390R_K392M_T394W, and wherein the numbering of amino acid residues in the Fc polypeptides is according to the EU numbering system.

22. The antibody construct of any one of claims 1-21, wherein the antibody construct binds CD3 with a KD from about 0.1 nM to about 100 nM, from about 1 nM to about 100 nM, or from about 10 nM to about 75 nM.

23. The antibody construct of any one of claims 1-22, wherein the antibody construct binds CD28 with a KD from about 1 nM to about 100 nM or from about 10 nM to about 75 nM.

24. The antibody construct of any one of claims 1-23, wherein the antibody construct coengages CD3 and CD28 on the surface of the same cytotoxic effector cell, as measured in a cell-cell bridging assay with recorded double positive events of less than about 5, less than about 3, or less than about 2.

25. The antibody construct of any one of claims 1-24, wherein the antibody construct binds CD28 on a cytotoxic effector cell that expresses both CD3 and CD28 with a KD that is about 2-fold, 4-fold, 6-fold, 8-fold or 10-fold higher when compared to a cytotoxic effector cell that expresses CD28 but not CD3, when measured using SPR.

26. The antibody construct of any one of claims 1-25, wherein the cytotoxic effector cell is an immune cell, optionally wherein the immune cell is a T cell.

27. The antibody construct of claim 26, wherein the antibody construct does not reduce the viability of a population of T cells by more than 5%, 3%, 1%, or by 0% after incubation for 48 hours when compared to T cells treated with a negative control construct that does not contain a binding domain against the disease-associated antigen.

28. The antibody construct of claim 12-27, wherein H2 further comprises a second Fab heavy chain comprising a second VH sequence and a second CHI sequence, wherein the second Fab heavy chain is coupled to the N-terminus of the second Fc polypeptide.

29. The antibody construct of claim 28, wherein H2 further comprises a second linkerFab'Fccomprising an IgGl hinge region, or are a derivative thereof, wherein the secondlinkerFab'Fchas at least about 80%, 85%, 90%, 93%, 95%, 97%, 99% or 100% sequence identity to the sequence set forth in SEQ ID NO: 54 or 56.

30. The antibody construct of claim 28, further comprising a second Ig light chain (L2) comprising a second Fab light chain comprising a second VL sequence and a second CL sequence, wherein the second Fab heavy chain of H2 and the second Fab light chain of L2 form a second Fab domain as the third binding domain.

31. An antibody construct, comprising:(i) a first Fab domain capable of binding CD3 on a cytotoxic effector cell,(ii) a first scFv domain capable of binding CD28 on a cytotoxic effector cell,(iii) a third binding domain comprising a second Fab domain and capable of binding a disease-associated antigen, and(iv) a dimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein:(a) the first Fab domain is coupled to the N-terminus of the first Fc polypeptide;(b) the second Fab domain is coupled to the N-terminus of the second Fc polypeptide; and(c) the first scFv domain is coupled to the C-terminus of the light chain of the first Fab domain.

32. The antibody construct of any one of claims 28-31, wherein one of the Fab domains comprises a kappa light chain and the other Fab domain comprises a lambda light chain.

33. The antibody construct of any one of claims 28-32, wherein one or both Fab domains comprise one or more amino acid substitutions relative to a reference Fab sequence that promote correct heavy and light chain pairing of Hl with LI as compared to L2, and H2 with L2 as compared to LI .

34. The antibody construct of claim 12-27, wherein H2 further comprises a second scFv domain comprising a second VH sequence and a second VL sequence, wherein the second scFv domain is coupled to the N-terminus of the second Fc polypeptide.

35. The antibody construct of claim 34, wherein H2 further comprises a first linkerscFv'Fcthat couples the second scFv domain to the second Fc polypeptide.

36. The antibody construct of claim 35, wherein the first linkerscFv'Fccomprises or consists of an IgGl hinge region, or are a derivative thereof, wherein the first linkerscFv'Fchas at least about 80%, 85%, 90%, 95% or 100% sequence identity to the sequence set forth in SEQ ID NO: 57.

37. The antibody construct of claim 35 or claim 36, wherein the second scFv domain comprises a second LinkerscFv(LinkerscFv2) and has the domain structure, from N- to C- terminus, of VH-LinkerscFv2- VL, wherein the LinkerscFv2comprises or consists of the amino acid sequence set forth in SEQ ID NO: 63.

38. The antibody construct of any one of claims 1-37, wherein the first Fab domain that is capable of binding CD3 (i) comprises HCDR1-3 as set forth in SEQ ID NOs: 49, 50 and 52, respectively, or 49, 51 and 52, respectively, and LCDR1 and 3 as set forth in SEQ ID NOs: 53 and 55, respectively, with LCDR2 having the sequence of RND, according to the IMGT numbering system; (ii) comprises HCDR1-3 as set forth in SEQ ID NOs: 134, 140 and 141, respectively, and LCDRl-3 set forth in SEQ ID NOs: 142, 138 and 139, respectively; or (iii) comprises HCDR1-3 as set forth in SEQ ID NOs: 143, 135 and 141, respectively, and LCDR1-3 set forth in SEQ ID NOs: 137-139, respectively.

39. The antibody construct of any one of claims 1-38, wherein the first Fab domain that is capable of binding CD3 comprises a VH sequence having at least about 90%, 95%, 97%, or 99% sequence identity to any one of the sequences set forth in SEQ ID NOs: 1, 2, 82, 84 or 86, and a VL sequence having at least about 90%, 95%, 97%, or 99% sequence identity to any one of the sequences set forth in SEQ ID NOs: 5, 83, 85 or 87.

40. The antibody construct of any one of claims 1-39, wherein the first scFv domain that is capable of binding CD28 (i) comprises HCDR1-3 as set forth in SEQ ID NOs: 30, 34 and 40, and LCDR1-3 as set forth in SEQ ID NOs: 43, 47 and 48, (ii) comprises HCDR1-3 as set forth in SEQ ID NOs: 106-108, and LCDR1-3 as set forth in SEQ ID NOs: 109-111, or (iii) comprises HCDR1-3 as set forth in SEQ ID NOs: 112-114, and LCDR1-3 as set forth in SEQ ID NOs: 115-117.

41. The antibody construct of any one of claims 1-40, wherein the first scFv domain that is capable of binding CD28 comprises a VH sequence having at least about 90%, 95%, 97%, or 99% sequence identity to any one of the sequences set forth in SEQ ID NOs: 15, 18, 20, 21, 22, 23, 24, 25, 27 and 76, and a VL sequence having at least about 90%, 95%,97%, or 99% sequence identity to any one of the sequences set forth in SEQ ID NOs: 16, 17, 19, 26 and 77.

42. The antibody construct of any one of claims 1-41, wherein the antibody construct binds the disease-associated antigen with a KD from about 0.01 nM to about 10 nM, from about 0.05 nM to about 5 nM, or from about 0.1 nM to about 5 nM.

43. The antibody construct of any one of claims 1-42, wherein the antibody construct comprises a fourth binding domain.

44. A pharmaceutical composition comprising the antibody construct of any one of claims 1- 43, and a pharmaceutically acceptable carrier, excipient, diluent, or combination thereof.

45. A nucleic acid molecule or a set of nucleic acid molecules encoding one or more, two or more, three or more, or four or more polypeptide chains that form the antibody construct of any one of claims 1-43.

46. A vector or a set of vectors comprising the nucleic acid molecule or the set of nucleic acid molecules of claim 45.

47. A cell comprising the nucleic acid molecule or the set of nucleic acid molecules of claim 45, or the vector or set of vectors of claim 46.

48. A method of producing an antibody construct of any one of claims 1-43, the method comprising:(a) obtaining a host cell culture comprising at least one host cell comprising one or more nucleic acid molecules encoding one or more, two or more, three or more, or four or more of the polypeptide chains that form the antibody construct; and(b) recovering the antibody construct from the host cell culture.

49. A method of eliciting an immune response in a cell population comprising immune cells and target cells, the method comprising contacting the cell population with an effective amount of the antibody construct of any one of claims 1-43, wherein the immune cells express CD3 and CD28 and the target cells express a disease-associated antigen.

50. A method of inhibiting the proliferation of target cells, the method comprising contacting a cell population comprising the target cells and immune cells with an effective amount of the antibody construct of any one of claims 1-43, wherein the immune cells express CD3 and CD28 and the target cells express a disease-associated antigen.

51. The method of any one of claims 49-50, wherein the target cells comprise tumor cells and the disease-associated antigen is a TAA.

52. The method of any one of claims 49-51, wherein the cell population is within a subject.

53. A method of treating a tumor in a subject in need thereof, the method comprising administering to the subject an antibody construct of any one of claims 1-43.

54. The method of claim 53, further comprising eliciting a cytotoxic immune response against the tumor in the subject, thereby treating the tumor in the subject.

55. The method of claim 53 or claim 54, wherein the tumor is a solid tumor or a liquid tumor.

56. An antibody construct of any one of claims 1-43 for use in the treatment of a tumor.

57. Use of an antibody construct of any one of claims 1-43 in the manufacture of a medicament for the treatment of a tumor.

Images