Trivalent trispecific antibody constructs and methods of use thereof

Abstract

The present disclosure provides trivalent, trispecific antibody constructs capable of binding two different antigens on one or more cytotoxic effector cells and a tumor associated antigen (TAA) on a tumor cell. Also disclosed are pharmaceutical compositions comprising such antibody constructs and methods of making and using such constructs and compositions, for example, for the treatment of cancer.

Description

[0001] Cross-reference

[0002] This application claims priority and the benefit of U.S. Provisional Application No. 63 / 591,311, filed Oct. 18, 2023; U.S. Provisional Application No. 63 / 465,137, filed May 9, 2023; U.S. Provisional Application No. 63 / 458,852, filed Apr. 12, 2023; and U.S. Provisional Application No. 63 / 417,542, filed Oct. 19, 2022, the entire contents of each of these applications being incorporated herein by reference for all purposes. Technical Field

[0003] The present disclosure generally relates to trivalent trispecific T cell engaging antibody constructs that can include a first binding domain capable of binding a first antigen on a first cytotoxic effector cell, a second binding domain capable of binding a second antigen on a second cytotoxic effector cell, and a third binding domain capable of binding a tumor associated antigen (TAA) on a tumor cell. Background Art

[0004] Cancer still represents a major unmet medical need, despite considerable progress in cancer treatment over the past few decades. Although current standards of care and recently developed anti-cancer therapies have shown some clinical progress, significant clinical challenges remain in multiple indications, such as those with low T cell tumor infiltration. Summary of the Invention

[0005] In various embodiments, the present disclosure describes trispecific trivalent antibody constructs that are capable of engaging two different antigens on one or more immune cells (e.g., T cells) and an antigen (e.g., TAA) on a tumor cell. In certain embodiments of the present disclosure, trispecific trivalent antibody constructs are described that include three binding domains, wherein a first binding domain is capable of binding a first antigen on a first cytotoxic effector cell, a second binding domain is capable of binding a second antigen on a second cytotoxic effector cell, and a third binding domain is capable of binding a tumor associated antigen (TAA) on a tumor cell.

[0006] In one embodiment, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding a first antigen on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein one of the scFv domains is capable of binding a second antigen on a second cytotoxic effector cell and the other scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first antigen and the second antigen on one or more cytotoxic effector cells are different, (b) the Fab domain is coupled to the N-terminus of the first Fc polypeptide, and (c) each of the first scFv domain and the second scFv domain is independently coupled to the N-terminus of the Fab domain, the C-terminus of the Fab domain, or the N-terminus of the second Fc polypeptide.

[0007] In one embodiment, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding CD3 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein one of the scFv domains is capable of binding CD28 on a second cytotoxic effector cell and the other scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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, and (b) each of the first scFv domain and the second scFv domain is independently coupled to the N-terminus of the Fab domain, the C-terminus of the Fab domain, the C-terminus of one of the Fc polypeptides, or the N-terminus of the second Fc polypeptide.

[0008] In one embodiment, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding CD28 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein one of the scFv domains is capable of binding CD3 on a second cytotoxic effector cell and the other scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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, and (b) each of the first scFv domain and the second scFv domain is independently coupled to the N-terminus of the Fab domain, the C-terminus of the Fab domain, the C-terminus of one of the Fc polypeptides, or the N-terminus of the second Fc polypeptide.

[0009] In one embodiment, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding CD3 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding CD28 on a second cytotoxic effector cell and the second scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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 V H domain of the Fab domain, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide. An antibody construct according to such an embodiment is shown, for example, as Figure 1A shown.

[0010] In one embodiment, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding CD28 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding CD3 on a second cytotoxic effector cell and the second scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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 V H domain of the Fab domain, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide. An antibody construct according to such an embodiment is shown, for example, as Figure 1A shown.

[0011] In one embodiment, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding CD3 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding CD28 on a second cytotoxic effector cell and the second scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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 C L terminus of the light chain of the Fab domain, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide. An antibody construct according to such an embodiment is shown, for example, as Figure 1B shown.

[0012] In one embodiment, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding CD28 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding CD3 on a second cytotoxic effector cell and the second scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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 C L -terminus of the light chain of the Fab domain, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide. An antibody construct according to such an embodiment is shown, for example, as Figure 1B illustrated.

[0013] In one embodiment, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding CD3 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding CD28 on a second cytotoxic effector cell and the second scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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 C-terminus of the first Fc polypeptide, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide. An antibody construct according to such an embodiment is shown, for example, as Figure 1F illustrated.

[0014] In one embodiment, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding CD28 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding CD3 on a second cytotoxic effector cell and the second scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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 C-terminus of the first Fc polypeptide, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide. An antibody construct according to such an embodiment is shown, for example, as Figure 1F illustrated.

[0015] In one embodiment, an antibody construct is described herein, the antibody construct comprising: (i) a Fab domain capable of binding CD3 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding CD28 on a second cytotoxic effector cell and the second scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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 C-terminus of the first Fc polypeptide, and (c) the second scFv domain is coupled to the N-terminus of the V H domain of the Fab domain. An antibody construct according to such an embodiment is shown, for example, as Figure 1C shown.

[0016] In one embodiment, an antibody construct is described herein, the antibody construct comprising: (i) a Fab domain capable of binding CD28 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding CD3 on a second cytotoxic effector cell and the second scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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 C-terminus of the first Fc polypeptide, and (c) the second scFv domain is coupled to the N-terminus of the V H domain of the Fab domain. An antibody construct according to such an embodiment is shown, for example, as Figure 1C shown.

[0017] In some embodiments, a pharmaceutical composition is described herein, which comprises the trivalent trispecific antibody construct of the present disclosure and a pharmaceutically acceptable carrier, excipient, diluent, or a combination thereof.

[0018] In some embodiments, a nucleic acid molecule or collection of nucleic acid molecules is described herein, which encodes one or more, two or more, or three or more polypeptide chains that form the trivalent trispecific antibody construct of the present disclosure.

[0019] In some embodiments, a vector or collection of vectors is described herein, which comprises a nucleic acid molecule or collection of nucleic acid molecules that encodes one or more, two or more, or three or more polypeptide chains that form the trivalent trispecific antibody construct of the present disclosure.

[0020] In certain embodiments, the present disclosure relates to a method of generating a trivalent trispecific antibody construct, the method comprising: (a) obtaining a host cell culture comprising at least one host cell that comprises one or more nucleic acid molecules encoding one or more, two or more, or three or more polypeptide chains that form the antibody construct; and (b) recovering the antibody construct from the host cell culture.

[0021] In certain embodiments, the present disclosure relates to a method of eliciting an anti-tumor immune response in a cell population comprising immune cells and tumor cells expressing a TAA, the method comprising contacting the cell population with an effective amount of the trivalent trispecific antibody construct of the present disclosure, wherein the immune cells express a first antigen and a second antigen, and the tumor cells express the TAA.

[0022] In certain embodiments, the present disclosure relates to a method of inhibiting the proliferation of tumor cells expressing a TAA, the method comprising contacting a cell population comprising tumor cells and immune cells with an effective amount of the trivalent trispecific antibody construct of the present disclosure, wherein the immune cells express a first antigen and a second antigen, and the tumor cells express the TAA.

[0023] In certain embodiments, the present disclosure relates to a method of killing tumor cells expressing a TAA, the method comprising contacting a cell population comprising tumor cells and immune cells with an effective amount of the trivalent trispecific antibody construct of the present disclosure, wherein the immune cells express a first antigen and a second antigen, and the tumor cells express the TAA.

[0024] In some embodiments, the first antigen can be CD3 or CD28, the second antigen can be CD3 or CD28, wherein the first antigen and the second antigen are different antigens. Additionally, in some embodiments, the TAA can be MSLN or Cldn18.2.

[0025] In certain embodiments, the present disclosure relates to a method of treating cancer in a subject in need thereof, the method comprising administering to the subject the trivalent trispecific antibody construct of the present disclosure or a pharmaceutical composition comprising the trivalent trispecific antibody construct of the present disclosure. In some embodiments, such method can further comprise eliciting a cytotoxic immune response against the cancer in the subject by the trivalent trispecific antibody construct, thereby treating the cancer in the subject.

[0026] In some embodiments, the present disclosure relates to a trivalent trispecific antibody construct for use in cancer treatment.

[0027] In some embodiments, the present disclosure relates to a trivalent trispecific antibody construct for use in the preparation of a medicament for treating cancer.

[0028] In some embodiments, an antibody construct is described herein that comprises a binding domain capable of binding CD28, wherein the binding domain comprises: V H a sequence that comprises an HCDR1 having the sequence SX1GVH (SEQ ID NO:302), an HCDR2 having the sequence VIWX2GGX3TNFNSALMS (SEQ ID NO:306), and an HCDR3 having the sequence DRAX4GX5YX6X7AMDY (SEQ ID NO:312); and V L a sequence that comprises an LCDR1 having the sequence RASESVEYYX8TSLMQ (SEQ ID NO:315), an LCDR2 having the sequence AASX9VX 10 S (SEQ ID NO:319), and an LCDR3 having the sequence QQSRKVPFT (SEQ ID NO:320), the binding domain having one or more of the following amino acid substitutions at the positions identified in the CDR sequences: X1: Y to A (i.e., the residue Y is replaced by the residue A); X2: P to A; X3: G to S; X4: S to Y; X5: N to A; X6: L to N; X7: S to Y; X8: G to V; X9: N to A; or X 10 : E to D.

[0029] In some embodiments, an antibody construct is described herein that comprises a binding domain capable of binding Cldn18.2, wherein the binding domain comprises: V H a sequence that comprises an HCDR1 having the sequence SNPMI (SEQ ID NO:333), an HCDR2 having the sequence IIDTDGSTYYADWAKG (SEQ ID NO:334), and an HCDR3 having the sequence RLHGSSNGYYDDL (SEQ ID NO:335); and V L a sequence that comprises an LCDR1 having the sequence QASQSIYSYLS (SEQ IDNO:336), an LCDR2 having the sequence KASTLAS (SEQ ID NO:337), and an LCDR3 having the sequence QQGYTVTNVDKNT (SEQ ID NO:338). BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The embodiments disclosed herein are illustrated by way of example and not limitation in the accompanying drawings. The description and the drawings are for illustrative purposes only and to assist understanding, and are not intended as a definition of the limitations of the antibody constructs, pharmaceutical compositions, and methods of the present disclosure.

[0031] Figure 1A - 1GSchematic showing the format and geometry of a trivalent trispecific antibody construct comprising binding domains capable of engaging CD3, CD28, and TAA, according to embodiments of the present disclosure. Figure 1H - 1L Schematic showing a bispecific control construct used and described in the present disclosure.

[0032] Figure 2A - 2B Shows the binding of certain anti-CD3 (Fab) / anti-CD28 (scFv) / anti-MSLN (scFv) trivalent trispecific constructs (v34914, v34915, v34916, and v34917, Figure 2A ) and anti-CD3 (scFv) / anti-CD28 (Fab) / anti-MSLN (scFv) trivalent trispecific constructs (v34913 and v34918, Figure 2B ) to CD4+ (left) and CD8+ (right) T cells, as measured by flow cytometry.

[0033] Figure 3A - 3B Shows the binding of certain anti-CD3 (Fab) / anti-CD28 (scFv) / anti-MSLN (scFv) trivalent trispecific constructs (v34914, v34915, v34916, and v34917, Figure 3A ) and anti-CD3 (scFv) / anti-CD28 (Fab) / anti-MSLN (scFv) trivalent trispecific constructs (v34913 and v34918, Figure 3B ) directing T cells from healthy donors to MSLN+H292 target cells, inducing T cell-mediated tumor cell killing.

[0034] Figure 4A - 4B Shows the production of interleukin-2 (IL-2, left) and / or tumor necrosis factor α (TNFα, right) from T cells induced by certain anti-CD3 (Fab) / anti-CD28 (scFv) / anti-MSLN (scFv) trivalent trispecific constructs (v34914, v34915, v34916, and v34917, Figure 4A ) and anti-CD3 (scFv) / anti-CD28 (Fab) / anti-MSLN (scFv) trivalent trispecific constructs (v34913 and v34918, Figure 4B ) when co-incubated with MSLN+H292 cells (E:T of 2:1) for 72 hours (h).

[0035] Figure 5A - 5D Shows that compared to the bispecific anti-(MSLNxCD3) benchmark control construct v34919, by the trivalent trispecific antibody constructs v34914 ( Figure 5A ), v34916 ( Figure 5B ), v34913 (Figure 5C ) and v34918( Figure 5D ) induced against MSLN 高 In vitro T cell-dependent cytotoxicity against H292 cells (152,986 MSLN / cell) and MSLN 低 OVTOKO cells (9,752 MSLN / cell).

[0036] Figure 6A - 6B Shows that the trivalent trispecific antibody constructs v34913, v34916, and v34918 direct T cells from healthy donors to kill MSLN+H292 target cells expressing moderate levels of MSLN. This long-term T cell-dependent cytotoxicity (TDCC) study utilized a low E:T ratio of 1:5 and an incubation period of 3 days ( Figure 6A ) or 7 days ( Figure 6B ) to more accurately reflect conditions in certain (e.g., solid) tumor types, such as long-term co-culture at a low E:T ratio.

[0037] Figure 7A - 7B Shows T cell proliferation after 5 days of co-incubation with MSLN+OVCAR3 cells (E:T of 10:1) induced by the trivalent trispecific antibody constructs v34913 and v34918 ( Figure 7B ) containing an anti-CD28 Fab domain, and the trivalent trispecific antibody constructs v34914, v34916, and v34917 ( Figure 7A ) containing an anti-CD3 Fab domain, compared to control constructs v34919, v34927, and v31926. Figure 7C Shows T cell proliferation data measured after 3, 5, and 7 days of co-culture of the effective trivalent trispecific construct v34913, the corresponding bispecific anti-(MSLNxCD3) control construct v34919, and the negative control v22277 with H292 cells (E:T ratio = 2:1).

[0038] Figure 8A - 8BBinding of certain trivalent trispecific anti-CD3 / anti-CD28 / anti-Cldn18.2 antibody constructs to CLDN18.2+ SNU 601 cells, as measured by flow cytometry, is shown. Constructs containing an anti-CD3 Fab domain and an anti-CD28 scFv domain (v37633, v37634, and v37635) and those containing an anti-CD28 Fab domain and an anti-CD3 scFv domain (v37638, v37640, and v37642) were all compared to an anti-(Cldn18.2xCD3) bispecific construct (v37663), a single-arm anti-Cldn18.2 antibody (v37675), a benchmark control (v35923, also known as AMG910), an anti-Cldn18.2 monoclonal antibody (mAb), and an anti-RSV protein F mAb palivizumab.

[0039] Figure 9A - 9B Binding of certain anti-CD3(Fab) / anti-CD28(scFv) / anti-Cldn18.2(scFv) trivalent trispecific constructs (v37633 - v37637, Figure 9A ) and anti-CD3(scFv) / anti-CD28(Fab) / anti-Cldn18.2(scFv) trivalent trispecific constructs (v37638 - v37642, Figure 9B ) to CD4+(left) and CD8+(right) T cells, as measured by flow cytometry, is shown.

[0040] Figure 10A - 10B Binding of certain anti-CD3(Fab) / anti-CD28(scFv) / anti-Cldn18.2(scFv) trivalent trispecific constructs (v37633 - v37637, Figure 10A ) and anti-CD3(scFv) / anti-CD28(Fab) / anti-Cldn18.2(scFv) trivalent trispecific constructs (v37638 - v37642, Figure 10B ) to direct T cells from healthy donors to kill CLDN18.2+ SNU601 target cells is shown. These trispecific constructs were compared to a bispecific anti-(Cldn18.2xCD3) control (v37663), a bispecific anti-(Cldn18.2xCD28) control (v37665), and a negative control (Het-Fc form of palivizumab, v22277).

[0041] Figure 11A - 11B Binding of certain anti-CD3(Fab) / anti-CD28(scFv) / anti-Cldn18.2(scFv) trivalent trispecific constructs (v37633 - v37637, Figure 11A) and the anti-CD3 (scFv) / anti-CD28 (Fab) / anti-Cldn18.2 (scFv) trivalent trispecific construct (v37638 - v37642, Figure 11B ) induces the production of IL-2 (upper) and / or TNFα (lower) from T cells co-incubated with CLDN18.2+ SNU 601 cells (E:T ratio of 2:1) for 72 hours.

[0042] Figure 12A - 12D Shows in vitro T cell-dependent cytotoxicity against CLDN18.2 Figure 12A ) v37633, Figure 12B ) v37634, Figure 12C ) v37638, Figure 12D ) and v37642 高 SNU 601 and CLDN18.2 低 SKOV-3 target cells induced by the trivalent trispecific antibody construct compared to the bispecific anti-(MSLNxCD3) benchmark control construct v35923.

[0043] Figure 13A - 13B Shows the concentration-response curves of the long-term (7-day) TDCC study of anti-CLDN18.2 trivalent trispecific constructs tested containing anti-CD3 (Fab) / anti-CD28 (scFv) / anti-Cldn18.2 (scFv) (v37633 - v36735, Figure 13A ) or anti-CD3 (scFv) / anti-CD28 (Fab) / anti-Cldn18.2 (scFv) (v37638, v37640, v37642, Figure 13B ) and shows that the constructs direct T cells from healthy donors to kill CLDN18.2+ target cells expressing different levels of CLDN18.2, using an E:T ratio of 1:1. Upper left: SNU-601 (276,125 CLDN18.2 / cell); upper right: KATO-III (63,566 CLDN18.2 / cell); lower: DAN-G (33,164 CLDN18.2 / cell).

[0044] Figure 14A - 14B Shows, compared to certain bispecific and benchmark control constructs, in the presence of the CLDN18.2+ target cell SNU 601 using a 72-hour incubation period and an E:T ratio of 2:1, the anti-CD3 (Fab) / anti-CD28 (scFv) / anti-Cldn18.2 (scFv) (v37633 - v36735, Figure 14A) or anti-CD3 (scFv) / anti-CD28 (Fab) / anti-Cldn18.2 (scFv) (v37638, v37640, v37642, Figure 14B ) certain anti-CLDN18.2 trivalent trispecific constructs are able to induce the production of several cytokines in T cells (i.e., IFNγ (upper left), IL-2 (upper right), TNFα (lower)).

[0045] Figure 15 Shows the binding curves of several anti-CLDN18.2 trivalent trispecific constructs v37638, v37683, v37689, v37692, and v37694 containing an anti-CD28 binding domain with mutations in their V H or V L domain (and thus having different affinities for CD28) to CD3+CD28+ Jurkat cells, as measured by flow cytometry.

[0046] Figure 16 Shows that several tested anti-CLDN18.2 trivalent trispecific constructs v37638, v37683, v37689, v37692, and v37694 having different affinities for CD28 direct T cells from healthy donors to kill CLDN18.2+ SNU 601 cells compared to bispecific (v37663) and negative (v22277) control constructs.

[0047] Figure 17A - 17B Shows that several tested anti-CLDN18.2 trivalent trispecific constructs v37638, v37683, v37689, v37692, and v37694 having different affinities for CD28 induce the production of IL-2 ( Figure 17A ) or TNFα ( Figure 17B ) from T cells co-incubated with CLDN18.2+ SNU 601 cells (E:T of 2:1) for 72 hours compared to bispecific (v37663) and negative (v22277) control constructs.

[0048] Figure 18 Shows the upregulation of Bcl-xL expression in activated T cells using the trivalent trispecific antibody construct v37634 of the present disclosure or certain bispecific control constructs.

[0049] Figure 19 Shows the upregulation of T cell proliferation in activated T cells using the trivalent trispecific antibody construct v37634 of the present disclosure or certain bispecific control constructs.

[0050] Figure 20Shows in vivo anti-tumor activity in mice transplanted with donor X treated with bispecific control construct v35923 or v38417 or trispecific trivalent antibody construct v37634.

[0051] Figure 21 Shows Figure 20 The body weights of mice treated during the in vivo efficacy study involved.

[0052] Figure 22A - 22B Shows (i) a library of conventional agonist complementarity determining region (CDR) variants ( Figure 22A ) with a range of CD28 binding affinities and (ii) agonist CDR variants ( Figure 22B ) with a range of CD3 binding affinities, as determined by SPR. Figure 22C Shows a schematic diagram (left) of the effect of the complementarity determining region (scFv versus Fab) and geometry of antibody constructs on the binding affinities for CD3 and CD28 for a subset of formats with the same CD3 and CD28 CDRs, and a schematic diagram of the affinity after engineering of CD3 and CD28 CDRs for one antibody construct format, which can be transferred in the format to create a large set of antibody constructs (right).

[0053] Figure 23A Shows that after 5 cycles of freezing to -80 °C and then thawing to 4 °C, the representative trivalent trispecific antibody construct v37634 shows no significant change in purity, Figure 23B Shows that after incubation at 40 °C for 14 days in an accelerated stress test, v37634 shows only a small change in purity. The construct concentration is 1 mg / mL. Figure 23C - 23F Shows that after storage at -80 °C for 10 weeks ( Figure 23C ), after storage at 4 °C for 14 days ( Figure 23D ), and after 3 hours at low (pH 3.5, Figure 23E ) and high (pH 9, Figure 23F ) pH conditions, the trivalent trispecific antibody construct v37634 shows no significant change in purity. Detailed Description

[0054] In various embodiments, the present disclosure describes trivalent trispecific antibody constructs that are capable of binding two different antigens located on a cytotoxic effector cell (e.g., a T cell) or on two different cytotoxic effector cells (i.e., each antigen is located on a different cell) and a tumor-associated antigen (TAA) on a tumor cell. Such antibody constructs can include a first binding domain capable of binding a first antigen on a first cytotoxic effector cell, a second binding domain capable of binding a second antigen on a second cytotoxic effector cell, and a third binding domain capable of binding a TAA on a tumor cell. Thus, in certain embodiments, trivalent trispecific T cell-engaging antibody constructs are described herein that are capable of co-stimulating one or more effector cells (e.g., T cells) by engaging two effector cell antigens (e.g., CD3 and CD28).

[0055] In some embodiments, as further described herein, the trivalent trispecific antibody constructs of the present disclosure can include (i) an antigen-binding fragment (Fab) domain (i.e., the first binding domain) capable of binding a first antigen on a first cytotoxic effector cell, (ii) a first single-chain variable fragment (scFv) domain (i.e., the second binding domain) capable of binding a second antigen on a second cytotoxic effector cell, and (iii) a second scFv domain (i.e., the third binding domain) capable of binding a TAA on a tumor cell. Such trivalent trispecific antibody constructs can also include an Fc domain (e.g., a heterodimeric Fc domain) that includes a first Fc polypeptide and a second Fc polypeptide and is coupled directly (e.g., without a linker) or indirectly (e.g., via a linker and / or via another binding domain (e.g., in the case where the first binding domain is indirectly coupled to the Fc polypeptide via the second binding domain, e.g., the C-terminus of the first binding domain is coupled to the N-terminus of the second binding domain, and this second binding domain is in turn coupled to the N-terminus of the Fc polypeptide, thereby indirectly coupling the first binding domain to the Fc polypeptide)) to each binding domain.

[0056] As further described herein, Figure 1A - 1G certain antibody construct geometries with three binding domains having different relative orientations are depicted in accordance with certain embodiments of the present disclosure.

[0057] Drug compositions are further described herein that include one or more of the trivalent trispecific antibody constructs disclosed herein.

[0058] Other embodiments of the disclosure relate to a nucleic acid molecule or collection of nucleic acid molecules that encode one or more (e.g., 2, 3, or more) polypeptide chains of the antibody constructs described herein. Some embodiments relate to a vector or collection of vectors that comprise a nucleic acid molecule or collection of nucleic acid molecules encoding the antibody constructs of the disclosure.

[0059] Other embodiments of the disclosure describe methods of generating and using the trivalent trispecific antibody constructs described herein, e.g., for treating cancer in a subject such as a rodent or a human.

[0060] I. Definitions

[0061] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art.

[0062] Unless otherwise indicated, as used herein, in the context of a numerical value or range, the term “about” generally refers to ±10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2% or ±1% of the recited or claimed numerical value or range. In various embodiments, the term “about” means varying approximately ±10% from a given value or range. In other embodiments, the term “about” means varying approximately ±5% from a given value or range. In other embodiments, the term “about” means varying approximately ±1% from a given value or range. It is understood that such variations, whether or not specifically recited, are always included in any given value provided herein.

[0063] When used in conjunction with the term “comprising” herein, the use of the word “a / an” can mean “one,” but it is also consistent with the meaning of “one or more,” “at least one,” and “one or more than one.”

[0064] As used herein, the terms “comprising,” “having,” “including,” and “containing” and their grammatical variants are inclusive or open-ended and do not exclude additional, unrecited elements and / or method steps. When used in conjunction with a construct, composition, use, or method herein, the term “consisting essentially of” means that additional features, elements, and / or method steps may be present, but these additional do not materially affect the manner in which the recited construct, composition, method, or use functions. The term “consisting of” when used in conjunction with a construct, composition, use, or method herein excludes the presence of additional elements and / or method steps. Antibody constructs, compositions, uses, or methods described herein as comprising certain elements and / or steps may also, in certain embodiments, consist essentially of those elements and / or steps, and in other embodiments, consist of those elements and / or steps, whether or not those embodiments are specifically recited.

[0065] The terms "subject" and "patient" are used interchangeably herein and generally refer to an animal in need of treatment. The animal in need of treatment can be a human or non-human animal, such as a mammal, bird, or fish. In certain embodiments, the subject or patient is a mammal. In some embodiments, the subject is a human. In other embodiments, the subject is a rodent or non-human primate.

[0066] For a particular result to be achieved, an "effective amount" of a trivalent trispecific antibody construct or a pharmaceutical composition comprising such an antibody construct described herein is an amount sufficient to achieve the desired result. For example, when it comes to killing cancer cells, an "effective amount" of the antibody construct or pharmaceutical composition refers to the amount of the antibody construct or the composition comprising the antibody construct sufficient to produce a killing effect.

[0067] Unless otherwise specified, the terms "Fc region", "Fc", and "Fc domain" are used interchangeably herein and refer to the C-terminal region of an immunoglobulin (Ig) heavy chain containing at least a portion of the constant region. In various embodiments, the Fc domain herein can be dimeric. Such a dimeric Fc domain can comprise a first Fc polypeptide and a second Fc polypeptide, where each Fc polypeptide can comprise a CH2 domain and a CH3 domain. Such a dimeric Fc can be homodimeric, i.e., comprising a first Fc polypeptide and a second Fc polypeptide having the same amino acid sequence, or heterodimeric, i.e., comprising a first Fc polypeptide and a second Fc polypeptide having different amino acid sequences (e.g., sequences having about 95%, 96%, 97%, 98%, or about 99% sequence identity). In some embodiments, the antibody constructs of the present disclosure comprise a homodimeric Fc domain. In other embodiments, and as further described herein, the antibody constructs comprise a heterodimeric Fc domain, where at least one of the CH2 and / or CH3 domains of the first Fc polypeptide and the second Fc polypeptide has an amino acid sequence having about 99% or less, 98% or less, or about 97% or less sequence identity.

[0068] As used herein, in the context of an antibody construct, the term "multispecific" refers to a biologically functional protein (e.g., an antibody construct as described herein) that is "at least bispecific", i.e., it comprises at least a first binding domain and a second binding domain, wherein such first binding domain and second binding domain can specifically bind two different epitopes, e.g., a first epitope and a second epitope. Such first epitope and second epitope can be located on the same antigen or different antigens, e.g., a first epitope on cluster of differentiation 3 (CD3) or cluster of differentiation 28 (CD28) and a second epitope on a tumor-associated antigen (TAA). Thus, in some embodiments, an antibody construct according to the present disclosure can comprise specificity for at least two different antigens or at least three different antigens or targets. Thus, as used herein in the context of an antibody construct, the term "multispecific" encompasses antibody constructs that are at least bispecific (i.e., comprise two binding domains that are specific for two different antigens or targets) or at least trispecific (i.e., comprise three binding domains that are specific for three different antigens or targets (e.g., CD3, CD28, and TAA)).

[0069] As used herein, in the context of an antibody construct, the term "trispecific" refers to a biologically functional protein (e.g., an antibody construct as described herein) that 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 binding domain, second binding domain, and third binding domain can specifically bind three different epitopes, e.g., a first epitope, a second epitope, and a third epitope. Such first epitope, second epitope, and third epitope can be located on the same antigen or different antigens, e.g., a first epitope on CD3, a second epitope on CD28, and a third epitope on a TAA (e.g., Claudin (Cldn) such as Cldn6 or Cldn18.2, or mesothelin (MSLN), etc.). Thus, in some embodiments, a trispecific antibody construct according to the present disclosure can comprise specificity for at least three different antigens or targets. Thus, as used herein in the context of an antibody construct, "specificity" describes the total number of different epitopes and / or antigens to which an antibody construct can specifically bind, e.g., a monospecific antibody construct comprises one or more binding domains that are specific for one epitope or antigen, a trispecific antibody construct comprises three or more binding domains that are specific for three different epitopes and / or antigens, and so on.

[0070] As used herein, the term "trivalent" in the context of an antibody construct refers to a biologically functional protein (e.g., an antibody construct as described herein) that is "at least trivalent", i.e., it contains 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 binding domain, the second binding domain, and the third binding domain is capable of specifically binding an epitope and / or an antigen, e.g., CD3, CD28, or a TAA. The three binding domains may be specific for three different epitopes or antigens, or two or more of the three binding domains may be specific for the same epitope or antigen. Thus, the valence, e.g., monovalent, divalent, or trivalent, in the context of an antibody construct herein describes the total number of antigen-binding domains of the antibody construct. Thus, the valence of an antibody construct must be at least equal to its specificity, i.e., a trispecific antibody construct must be at least trivalent. In an embodiment where the antibody construct is trivalent and trispecific, each of the three binding domains of the construct is capable of binding a different epitope or antigen, and thus the construct binds each of the three epitopes and / or antigens monovalently.

[0071] As used herein, the term "format" in the context of an antibody construct as described herein generally describes the properties of the antibody construct, including: its antigen valence (e.g., the construct is monovalent or divalent for a given antigen), the type of binding domains present in the antibody construct (e.g., having one or more scFv domains, Fab domains, etc.), and the presence, absence, and / or type of Fc domain (e.g., homodimer, heterodimer, containing one or more constant heavy chain domains CH2, CH3, etc.). As an example, in some embodiments, the antibody constructs of the present disclosure may be trivalent and trispecific, in a "1+1+1" format, meaning the construct contains three binding domains, where each binding domain has an affinity for a different antigen (e.g., CD3, CD28, and a TAA), i.e., the trivalent construct is monovalent for each of the three antigens (as represented by the "1").

[0072] As used herein, the term "geometry" in the context of an antibody construct as described herein generally describes the overall structure of the antibody construct, including the relative spatial orientation and arrangement and / or connection of the individual domains of the antibody construct, e.g., according to certain embodiments of the present disclosure, the relative arrangement and connection of the binding domain and the Fc domain, as further described herein and as Figure 1A - 1G shown in.

[0073] Unless otherwise indicated, the amino acid sequences of the polypeptides described herein are described and defined in the direction from the N-terminus to the C-terminus. As an example, a polypeptide described as comprising an scFv domain conjugated to an Fc polypeptide is defined herein as a polypeptide in which the C-terminus of the scFv domain (with or without a linker) is conjugated to the N-terminus of the Fc polypeptide, and the domain structure can be described as scFv-Fc, or scFv-linker-Fc in the case of the inclusion of a linker. scFv-Fc -Fc.

[0074] As used herein, abbreviations such as "H1" and "H2" or "A" and "B" are generally used as general heavy chain identifiers and broadly refer, respectively, to the first and second heavy chains of an antibody construct and are not intended to be limited to any particular heavy chain amino acid (or polynucleotide) sequence.

[0075] As used herein, the term "amino acid modification" in the context of the amino acid sequence of a polypeptide generally refers to an amino acid sequence of a polypeptide in which one or more amino acid substitutions, one or more amino acid insertions, and / or one or more amino acid deletions have been introduced relative to the corresponding unmodified (e.g., WT or reference) amino acid sequence of the polypeptide.

[0076] Descriptions of antibody constructs such as "anti-(Cldn18.2xCD28xCD3)" and "anti-CD3 / anti-CD28 / anti-Cldn18.2" 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 to epitopes on Cldn18.2, CD28, and CD3, respectively. In various embodiments, such a description refers to a trivalent trispecific antibody construct having three binding domains that are capable of binding to epitopes on Cldn18.2, CD28, and CD3, respectively.

[0077] Typically, an affirmative statement of a feature in one embodiment is used as a basis for excluding that feature in an alternative embodiment. In particular, in cases where a list of options is presented for a given embodiment or claim, it should be understood that one or more options can be removed from the list and the shortened list can form an alternative embodiment, whether or not such alternative embodiments are specifically recited.

[0078] It is also contemplated that any embodiment discussed herein can be implemented with respect to any antibody construct, method, use, or composition disclosed herein, and vice versa. In addition, modifications of the specific embodiments described herein that are obvious to those of ordinary skill in the art are intended to be included within the scope of the claims described herein.

[0079] II. Antibody Constructs

[0080] In various embodiments, the present disclosure relates to trivalent trispecific T cell engager antibody constructs that comprise three binding domains capable of engaging two different antigens on one or more cytotoxic effector cells and a TAA on tumor cells.

[0081] In some embodiments, the trivalent trispecific antibody constructs of the present disclosure comprise: (i) a first binding domain capable of binding a first antigen on a first cytotoxic effector cell, (ii) a second binding domain capable of binding a second antigen on a second cytotoxic effector cell, (iii) a third binding domain capable of binding a TAA on a tumor cell, and (iv) an Fc domain comprising a first Fc polypeptide and a second Fc polypeptide. The antibody constructs described herein can have various formats and geometries. For example, as further described herein, the binding domains and the Fc domain can have various relative spatial orientations and arrangements.

[0082] A. Formats and Geometries of Antibody Constructs

[0083] In various embodiments, the antibody constructs described in the present disclosure are trivalent and trispecific and comprise three antigen-binding domains, each of the three antigen-binding domains capable of binding a different antigen. Thus, in various embodiments, the trivalent trispecific antibody constructs of the present disclosure can have a 1+1+1 format, indicating that each of the three binding domains binds a different antigen, and thus such constructs bind each antigen in a monovalent manner.

[0084] In some embodiments, such trivalent trispecific antibody constructs can comprise one or more different types of binding domains. Types of binding domains that can be used in the antibody constructs described herein include scFv domains, Fab domains, single domain antibodies (sdAbs), etc. Thus, in certain embodiments, the antibody constructs described herein can comprise one or more scFv domains and / or one or more Fab domains. In some embodiments, the antibody construct can comprise one or more scFv domains and one or more Fab domains.

[0085] In some embodiments, the trivalent trispecific antibody constructs of the present disclosure comprise two scFv domains (e.g., a first scFv domain and a second scFv domain) and one Fab domain. Such an antibody construct can also comprise an Fc domain. Thus, the two scFv domains, the Fab domain, and the Fc domain can be coupled to each other in various relative spatial arrangements to yield various construct geometries, such as as Figure 1A - 1G shown. In some embodiments, the Fc domain is a heterodimeric Fc domain.

[0086] In certain embodiments of the present disclosure, the trivalent trispecific antibody constructs do not contain a polypeptide chain comprising two or more scFv domains coupled to each other in tandem. For example, according to the domain structure, from the N-terminus to the C-terminus or from the C-terminus to the N-terminus, it is (V H / L -V L / H ) scFv1 -(V H / L -V L / H ) scFv2 , wherein such a chain may optionally contain one or more linkers that couple the V H and V L domains within the scFv domain or between two scFv domains. Expression of an antibody construct containing a polypeptide chain with two or more scFv domains coupled in tandem may reduce the producibility and stability (e.g., thermal stability) of the antibody construct. Thus, in various embodiments, compared to conventional constructs containing a polypeptide chain with two or more scFv domains coupled in tandem, the antibody constructs described herein may have higher producibility and / or thermal stability.

[0087] In some embodiments, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding a first antigen on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein one of the scFv domains is capable of binding a second antigen on a second cytotoxic effector cell and the other scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first antigen and the second antigen are different, (b) the Fab domain is coupled to the N-terminus of the first Fc polypeptide, and (c) the first scFv domain and the second scFv domain are independently coupled to the N-terminus of the Fab domain, the C-terminus of the Fab domain, the C-terminus of one of the Fc polypeptides, or the N-terminus of the second Fc polypeptide.

[0088] In some embodiments, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding to a first antigen on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein one of the scFv domains is capable of binding to a second antigen on a second cytotoxic effector cell and the other scFv domain is capable of binding to a tumor-associated antigen (TAA) on a tumor cell; and (iii) an Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein: (a) the first antigen and the second antigen are different, (b) the Fab domain is coupled to the N-terminus of the first Fc polypeptide, and (c) the first scFv domain and the second scFv domain are independently coupled to the N-terminus of the Fab domain, the C-terminus of the Fab domain, or the N-terminus of the second Fc polypeptide.

[0089] In one embodiment, the present disclosure describes a trivalent trispecific antibody construct that comprises: (i) a Fab domain capable of binding to a first antigen on a first cytotoxic effector cell; (ii) a first scFv domain capable of binding to a second antigen on a second cytotoxic effector cell; (iii) a second scFv domain capable of binding to a tumor-associated antigen (TAA); and (iv) a heterodimeric 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 or the N-terminus of the second Fc polypeptide, (b) the first scFv domain is coupled to the N-terminus of the Fab domain, the C-terminus of the Fab domain, the C-terminus of the first Fc polypeptide, or the N-terminus of the second Fc polypeptide, and (c) the second scFv domain is coupled to the N-terminus of the first Fc polypeptide, the N-terminus of the second Fc polypeptide, or the N-terminus of the Fab domain.

[0090] In one embodiment, the present disclosure describes a trivalent trispecific antibody construct that comprises: (i) a Fab domain capable of binding to a first antigen on a first cytotoxic effector cell; (ii) a first scFv domain capable of binding to a second antigen on a second cytotoxic effector cell; (iii) a second scFv domain capable of binding to a tumor-associated antigen (TAA); and (iv) a heterodimeric 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 Fab domain, the C-terminus of the Fab domain, or the C-terminus of the first Fc polypeptide, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide.

[0091] In some embodiments, the first antigen is CD3 or CD28, wherein the first antigen and the second antigen are different antigens.

[0092] In some embodiments, the second antigen is CD3 or CD28, wherein the first antigen and the second antigen are different antigens.

[0093] In certain embodiments of the present disclosure, the first antigen on the first cytotoxic effector cell is CD28, and the second antigen on the second cytotoxic effector cell is CD3. In other embodiments, the first antigen is CD3, and the second antigen is CD28.

[0094] In one embodiment, the present disclosure describes a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding CD28 on a first cytotoxic effector cell; (ii) a first scFv domain capable of binding CD3 on a second cytotoxic effector cell; (iii) a second scFv domain capable of binding a TAA on a tumor cell; and (iv) a heterodimeric 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 Fab domain, the C-terminus of the Fab domain, or the C-terminus of the first Fc polypeptide, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide.

[0095] In one embodiment, the present disclosure describes a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding CD3 on a first cytotoxic effector cell; (ii) a first scFv domain capable of binding CD28 on a second cytotoxic effector cell; (iii) a second scFv domain capable of binding a TAA on a tumor cell; and (iv) a heterodimeric 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 Fab domain, the C-terminus of the Fab domain, or the C-terminus of the first Fc polypeptide, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide.

[0096] In some embodiments, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding to a first antigen on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein one of the scFv domains is capable of binding to a second antigen on a second cytotoxic effector cell and the other scFv domain is capable of binding to a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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, and (b) the first scFv domain and the second scFv domain are independently coupled to (i) the N-terminus of the Fab domain, (ii) the C-terminus of the Fab domain, (iii) the N-terminus of one of the Fc polypeptides, or (iv) the N-terminus of the second Fc polypeptide, provided that when one of the scFv domains is coupled to the C-terminus of one of the Fc polypeptides, the first antigen is CD3 and the second antigen is CD28, or the first antigen is CD28 and the second antigen is CD3, and provided that the first scFv domain and the second scFv domain are not coupled to each other in a tandem manner.

[0097] In some embodiments of the trivalent trispecific antibody constructs described herein, the first scFv domain is coupled to the N-terminus of the Fab domain. In such embodiments, the first scFv domain may be coupled to the N-terminus of the V H sequence of the heavy chain of the Fab domain. In other embodiments, the first scFv domain may be coupled to the N-terminus of the V L sequence of the light chain of the Fab domain.

[0098] In certain embodiments, the trivalent trispecific antibody constructs of the present disclosure comprise: a) a first heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a first scFv domain that comprises a first scFv V L sequence coupled to a first scFv V H sequence (V H -V L ), or a first scFv V H sequence coupled to a first scFv V L sequence (V L -V H ), (ii) a heavy chain Fab sequence that comprises a Fab V H1 sequence coupled to a Fab C H sequence, and (iii) a first Fc polypeptide; b) a second heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a second scFv domain that comprises a second scFv V L sequence coupled to a second scFv VH Sequence (V H -V L ), or with a second scFv V H Sequence coupled second scFv V L Sequence (V L -V H ), and (ii) a second Fc polypeptide; and c) a light chain polypeptide comprising, from N-terminus to C-terminus: L Sequence coupled Fab V L Sequence, wherein: the heavy chain Fab sequence and the light chain polypeptide form a Fab domain, and the first Fc polypeptide and the second Fc polypeptide form a heterodimeric Fc domain.

[0099] In other embodiments of the trivalent trispecific antibody construct, the first scFv domain is coupled to the C-terminus of the first Fc polypeptide. In such embodiments, the trivalent trispecific antibody construct may comprise: a) a first heavy chain polypeptide comprising, from N-terminus to C-terminus: (i) a heavy chain Fab sequence comprising a Fab C-terminus; H1 Sequence coupled Fab V H sequence, (ii) a first Fc polypeptide, and (iii) a first scFv domain comprising a first scFv V L Sequence coupling of the first scFv V H Sequence (V H -V L ), or with the first scFv V H Sequence coupling of the first scFvV L Sequence (V L -V H ); b) a second heavy chain polypeptide, which comprises from N-terminus to C-terminus: (i) a second scFv domain, which comprises a second scFv V L Sequence coupled second scFv V H Sequence (V H -V L ), or with a second scFv V H Sequence coupled second scFv V L Sequence (V L -V H ), and (ii) a second Fc polypeptide; and c) a light chain polypeptide comprising a Fab C L Sequence coupled Fab V L Sequence, wherein: the heavy chain Fab sequence and the light chain polypeptide form a Fab domain, and the first Fc polypeptide and the second Fc polypeptide form a heterodimeric Fc domain. One embodiment of this trivalent trispecific antibody construct is Figure 1F Describe in.

[0100] In still further embodiments of the trivalent trispecific antibody constructs described herein, the first scFv domain is coupled to the C-terminus of the light chain of the Fab domain. In such embodiments, the antibody construct can comprise: a) a first heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a heavy chain Fab sequence that comprises a Fab V sequence coupled to a Fab C sequence, and (ii) a first Fc polypeptide; b) a second heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a second scFv domain that comprises a second scFv V sequence coupled to a second scFv V sequence (V-V), or a second scFv V sequence coupled to a second scFv V sequence (V-V), and (ii) a second Fc polypeptide; and c) a light chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a light chain Fab sequence that comprises a Fab V sequence coupled to a Fab C sequence, and (ii) a first scFv domain that comprises a first scFv V sequence coupled to a first scFv V sequence (V-V), or a first scFv V sequence coupled to a first scFv V sequence (V-V), wherein the heavy chain Fab sequence and the Fab light chain polypeptide sequence form a Fab domain, and the first Fc polypeptide and the second Fc polypeptide form a heterodimeric Fc domain. One embodiment of such a trivalent trispecific antibody construct is depicted in L sequence. In such embodiments, the antibody construct can comprise: a) a first heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a heavy chain Fab sequence that comprises a Fab V sequence coupled to a Fab C H1 sequence, and (ii) a first Fc polypeptide; b) a second heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a second scFv domain that comprises a second scFv V sequence coupled to a second scFv V H sequence, and (ii) a second Fc polypeptide; and c) a light chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a light chain Fab sequence that comprises a Fab V sequence coupled to a Fab C L sequence, and (ii) a first scFv domain that comprises a first scFv V sequence coupled to a first scFv V H sequence (V H -V L ), or a first scFv V sequence coupled to a first scFv V H sequence, and (ii) a first Fc polypeptide; b) a second heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a second scFv domain that comprises a second scFv V sequence coupled to a second scFv V L sequence (V L -V H ), and (ii) a second Fc polypeptide; and c) a light chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a light chain Fab sequence that comprises a Fab V sequence coupled to a Fab C L sequence, and (ii) a first scFv domain that comprises a first scFv V sequence coupled to a first scFv V L sequence, and (ii) a first scFv domain that comprises a first scFv V sequence coupled to a first scFv V L sequence, and (ii) a first scFv domain that comprises a first scFv V sequence coupled to a first scFv V H sequence (V H -V L ), or a first scFv V sequence coupled to a first scFv V H sequence, and (ii) a first scFv domain that comprises a first scFv V sequence coupled to a first scFv V L sequence (V L -V H ), wherein the heavy chain Fab sequence and the Fab light chain polypeptide sequence form a Fab domain, and the first Fc polypeptide and the second Fc polypeptide form a heterodimeric Fc domain. One embodiment of such a trivalent trispecific antibody construct is depicted in Figure 1B is depicted.

[0101] In one embodiment, the present disclosure describes a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding CD28 on a first cytotoxic effector cell; (ii) a first scFv domain capable of binding CD3 on a second cytotoxic effector cell; (iii) a second scFv domain capable of binding a TAA on a tumor cell; and (iv) a heterodimeric 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 or the second Fc polypeptide, (b) the first scFv domain is coupled to the N-terminus of the first Fc polypeptide, the C-terminus of the first Fc polypeptide, or the N-terminus of the second Fc polypeptide, and (c) the second scFv domain is coupled to the N-terminus of the Fab domain. Certain embodiments of such trivalent trispecific antibody constructs are depicted in Figure 1D and Figure 1F are depicted.

[0102] In another embodiment, the present disclosure describes a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding CD3 on a first cytotoxic effector cell; (ii) a first scFv domain capable of binding CD28 on a second cytotoxic effector cell; (iii) a second scFv domain capable of binding a TAA on a tumor cell; and (iv) a heterodimeric 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 or the second Fc polypeptide, (b) the first scFv domain is coupled to the N-terminus of the first Fc polypeptide, the C-terminus of the first Fc polypeptide, or the N-terminus of the second Fc polypeptide, and (c) the second scFv domain is coupled to the N-terminus of the Fab domain. Certain embodiments of such trivalent trispecific antibody constructs are depicted in Figure 1D and Figure 1F are depicted.

[0103] In certain embodiments, the Fab domain is coupled to the N-terminus of the first Fc polypeptide and the first scFv domain is coupled to the C-terminus of the first Fc polypeptide. Exemplary embodiments are described in Figure 1C and Figure 1F are described. In one such embodiment, for example as Figure 1C shown, the antibody construct may comprise: a) a first heavy chain polypeptide comprising, from the N-terminus to the C-terminus: (i) a second scFv domain comprising a second scFv V L sequence coupled to a second scFv V H sequence (V H -V L ), or a second scFv V HSequence coupled second scFv V L Sequence (V L -V H ), (ii) a heavy chain Fab sequence comprising a Fab C H1 Sequence coupled Fab V H sequence, (iii) a first Fc polypeptide, and (iv) a first scFv domain, which comprises a first scFvV L Sequence coupling of the first scFv V H Sequence (V H -V L ), or with the first scFv V H Sequence coupling of the first scFv V L Sequence (V L -V H ); b) a second heavy chain polypeptide comprising or consisting of a second Fc polypeptide; and c) a light chain polypeptide comprising, from N-terminus to C-terminus: a light chain Fab sequence comprising a Fab C L Sequence coupled Fab V L Sequence, wherein: the heavy chain Fab sequence and the light chain polypeptide form a Fab domain, and the first Fc polypeptide and the second Fc polypeptide form a heterodimeric Fc domain.

[0104] In another embodiment of the trivalent trispecific antibody constructs described herein, the Fab domain is coupled to the N-terminus of the first Fc polypeptide, and the first scFv domain is coupled to the N-terminus of the second Fc polypeptide, e.g. Figure 1E In this embodiment, the antibody construct may comprise: a) a first heavy chain polypeptide comprising, from N-terminus to C-terminus: (i) a second scFv domain comprising a second scFv V L Sequence coupled second scFv V H Sequence (V H -V L ), or with a second scFv V H Sequence coupled second scFv V L Sequence (V L -V H ), (ii) a heavy chain Fab sequence comprising a Fab C H1 Sequence coupled FabV H sequence, and (iii) a first Fc polypeptide; b) a second heavy chain polypeptide comprising, from N-terminus to C-terminus: (i) a first scFv domain comprising a first scFv V L Sequence coupling of the first scFv V H Sequence (V H -V L), or with the first scFv V H Sequence coupling of the first scFv V L Sequence (V L -V H ), and (ii) a second Fc polypeptide; and c) a light chain polypeptide comprising, from N-terminus to C-terminus: a light chain Fab sequence comprising a Fab C L Sequence coupled Fab V L Sequence, wherein: the heavy chain Fab sequence and the light chain polypeptide form a Fab domain, and the first Fc polypeptide and the second Fc polypeptide form a heterodimeric Fc domain.

[0105] In another embodiment, the Fab domain is coupled to the N-terminus of the second Fc polypeptide, and the first scFv domain is coupled to the N-terminus of the first Fc polypeptide, e.g. Figure 1D In this embodiment, the antibody construct may comprise: a) a first heavy chain polypeptide comprising, from N-terminus to C-terminus: (i) a first scFv domain comprising a first scFv V domain; L Sequence coupling of the first scFv V H Sequence (V H -V L ), or with the first scFv V H Sequence coupling of the first scFv V L Sequence (V L -V H ), and (iii) a first Fc polypeptide; b) a second heavy chain polypeptide comprising, from N-terminus to C-terminus: (i) a second scFv domain comprising a second scFvV L Sequence coupled second scFv V H Sequence (V H -V L ), or with a second scFv V H Sequence coupled second scFv V L Sequence (V L -V H ), (ii) a heavy chain Fab sequence comprising a Fab C H1 Sequence coupled Fab V H sequence, and (iii) a second Fc polypeptide; and c) a light chain polypeptide comprising, from N-terminus to C-terminus: a light chain Fab sequence comprising L Sequence coupled Fab V L Sequence, wherein: the heavy chain Fab sequence and the light chain polypeptide form a Fab domain, and the first Fc polypeptide and the second Fc polypeptide form a heterodimeric Fc domain.

[0106] In some embodiments, an antibody construct is described herein, wherein each of the first scFv construct and the second scFv domain can be independently coupled to the N-terminus of a Fab domain, the C-terminus of a Fab domain, the C-terminus of one of the Fc polypeptides, or the N-terminus of a second Fc polypeptide.

[0107] In such embodiments, the antibody construct can comprise: (i) a Fab domain capable of binding CD3 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein one of the scFv domains is capable of binding CD28 on a second cytotoxic effector cell and the other scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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, and (b) each of the first scFv domain and the second scFv domain is independently coupled to the N-terminus of the Fab domain, the C-terminus of the Fab domain, the C-terminus of one of the Fc polypeptides, or the N-terminus of the second Fc polypeptide.

[0108] In other embodiments, the antibody construct can comprise: (i) a Fab domain capable of binding CD28 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein one of the scFv domains is capable of binding CD3 on a second cytotoxic effector cell and the other scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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, and (b) each of the first scFv domain and the second scFv domain is independently coupled to the N-terminus of the Fab domain, the C-terminus of the Fab domain, the C-terminus of one of the Fc polypeptides, or the N-terminus of the second Fc polypeptide.

[0109] In some embodiments, the first scFv domain is coupled to the N-terminus of the Fab domain and the second scFv is coupled to the N-terminus of the second Fc polypeptide. In certain embodiments, the first scFv domain is coupled to the N- H terminus of the V sequence of the heavy chain of the Fab domain. In other embodiments, the first scFv domain is coupled to the N- L terminus of the V sequence of the light chain of the Fab domain.

[0110] In some embodiments, a trivalent trispecific antibody construct can comprise or consist of three polypeptide chains that can associate to form the antibody construct. In some embodiments, such a trivalent trispecific antibody construct comprises: a) a first heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a first scFv domain that comprises a first scFv V L sequence coupled to a first scFv V H sequence (V H -V L ), or a first scFv V H sequence coupled to a first scFv V L sequence (V L -V H ), (ii) a heavy chain Fab sequence that comprises a Fab V H1 sequence coupled to a Fab C H sequence, and (iii) a first Fc polypeptide; b) a second heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a second scFv domain that comprises a second scFv V L sequence coupled to a second scFv V H sequence (V H -V L ), or a second scFv V H sequence coupled to a second scFv V L sequence (V L -V H ), and (ii) a second Fc polypeptide; and c) a light chain polypeptide that, from the N-terminus to the C-terminus, comprises a Fab V L sequence coupled to a Fab C L sequence, wherein: the heavy chain Fab sequence and the light chain polypeptide form a Fab domain, and the first Fc polypeptide and the second Fc polypeptide form an Fc domain.

[0111] In some embodiments, the first scFv domain is coupled to the C-terminus of the Fab domain, and the second scFv domain is coupled to the N-terminus of the second Fc polypeptide. In such embodiments, the trivalent trispecific antibody construct can comprise the following polypeptide chains: a) a first heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a heavy chain Fab sequence that comprises a Fab V H1 sequence coupled to a Fab C H sequence, and (ii) a first Fc polypeptide; b) a second heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a second scFv domain that comprises a second scFv V L sequence coupled to a second scFv V H sequence (V H -V L), or with a second scFv V H Sequence coupled second scFv V L Sequence (V L -V H ), and (ii) a second Fc polypeptide; and c) a light chain polypeptide comprising, from N-terminus to C-terminus: (i) L Sequence coupled Fab V L sequence, and (ii) a first scFv domain comprising a first scFv V L Sequence coupling of the first scFvV H Sequence (V H -V L ), or with the first scFv V H Sequence coupling of the first scFv V L Sequence (V L -V H ), wherein: the heavy chain Fab sequence and the light chain Fab sequence form a Fab domain, and the first Fc polypeptide and the second Fc polypeptide form an Fc domain.

[0112] In some embodiments of the trivalent trispecific antibody constructs described herein, the first scFv domain is coupled to the N-terminus of the Fab domain and the second scFv domain is coupled to the C-terminus of one of the Fc polypeptides. In such embodiments, the first scFv domain can be coupled to the V-terminus of the Fab domain. H In other embodiments, the first scFv domain is coupled to the N-terminus of the Fab domain. L In some embodiments, the second scFv domain is coupled to the N-terminus of the first Fc polypeptide. In some embodiments, the second scFv domain is coupled to the C-terminus of the first Fc polypeptide. In further such embodiments, the second scFv domain is coupled to the C-terminus of the first Fc polypeptide.

[0113] In some embodiments, the trivalent trispecific antibody construct may comprise the following polypeptide chains: a) a first heavy chain polypeptide comprising, from N-terminus to C-terminus: (i) a first scFv domain comprising a first scFv V domain; L Sequence coupling of the first scFv V H Sequence (V H -V L ), or with the first scFv V H Sequence coupling of the first scFv V L Sequence (V L -V H ), (ii) a heavy chain Fab sequence comprising a Fab C H1 Sequence coupled Fab V Ha sequence, (iii) a first Fc polypeptide, and (iv) a second scFv domain comprising a second scFv V L sequence-coupled second scFv V H sequence (V H -V L ), or a second scFv V H sequence-coupled second scFv V L sequence (V L -V H ); b) a second heavy chain polypeptide comprising a second Fc polypeptide; and c) a light chain polypeptide comprising a Fab V L sequence-coupled to a Fab C L sequence, wherein: the heavy chain Fab sequence and the light chain polypeptide form a Fab domain, and the first Fc polypeptide and the second Fc polypeptide form an Fc domain.

[0114] In other embodiments, the trivalent trispecific antibody construct can comprise the following polypeptide chains: a) a first heavy chain polypeptide comprising, from the N-terminus to the C-terminus: (i) a first scFv domain comprising a first scFv V L sequence-coupled first scFv V H sequence (V H -V L ), or a first scFv V H sequence-coupled first scFv V L sequence (V L -V H ), (ii) a heavy chain Fab sequence comprising a Fab V H1 sequence-coupled to a Fab C H sequence, and (iii) a first Fc polypeptide; b) a second heavy chain polypeptide comprising, from the N-terminus to the C-terminus: (i) a second scFv domain comprising a second scFv V L sequence-coupled second scFv V H sequence (V H -V L ), or a second scFv V H sequence-coupled second scFv V L sequence (V L -V H ), and (ii) a second Fc polypeptide; and c) a light chain polypeptide comprising a Fab V L sequence-coupled to a Fab C L sequence, wherein: the heavy chain Fab sequence and the light chain polypeptide form a Fab domain, and the first Fc polypeptide and the second Fc polypeptide form an Fc domain.

[0115] In some embodiments of the trivalent trispecific antibody constructs described herein that include a first scFv domain and a second scFv domain, the first scFv domain is capable of binding to a TAA, and the second scFv domain is capable of binding to CD28 or CD3. In still other embodiments, the second scFv domain is capable of binding to a TAA, and the first scFv domain is capable of binding to CD28 or CD3.

[0116] In one embodiment, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding to CD3 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding to CD28 on a second cytotoxic effector cell, and the second scFv domain is capable of binding to a tumor-associated antigen (TAA) on a tumor cell; and (iii) an Fc domain that comprises 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 V H domain of the Fab domain, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide.

[0117] In one embodiment, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding to CD28 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding to CD3 on a second cytotoxic effector cell, and the second scFv domain is capable of binding to a tumor-associated antigen (TAA) on a tumor cell; and (iii) an Fc domain that comprises 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 V H domain of the Fab domain, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide.

[0118] In one embodiment, an antibody construct is described herein that comprises: (i) a Fab domain capable of binding to CD3 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding to CD28 on a second cytotoxic effector cell, and the second scFv domain is capable of binding to a tumor-associated antigen (TAA) on a tumor cell; and (iii) an Fc domain that comprises 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 CL coupling of the C-terminus of the domain, and (c) coupling of the second scFv domain to the N-terminus of the second Fc polypeptide.

[0119] In one embodiment, described herein is an antibody construct comprising: (i) a Fab domain capable of binding CD28 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding CD3 on a second cytotoxic effector cell and the second scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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 C- L coupling of the C-terminus of the domain, and (c) coupling of the second scFv domain to the N-terminus of the second Fc polypeptide.

[0120] In one embodiment, described herein is an antibody construct comprising: (i) a Fab domain capable of binding CD3 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding CD28 on a second cytotoxic effector cell and the second scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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 C-terminus of the first Fc polypeptide, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide.

[0121] In one embodiment, described herein is an antibody construct comprising: (i) a Fab domain capable of binding CD28 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding CD3 on a second cytotoxic effector cell and the second scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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 C-terminus of the first Fc polypeptide, and (c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide.

[0122] In one embodiment, an antibody construct is described herein, the antibody construct comprising: (i) a Fab domain capable of binding CD3 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding CD28 on a second cytotoxic effector cell and the second scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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 C-terminus of the first Fc polypeptide, and (c) the second scFv domain is coupled to the N-terminus of the V H domain of the Fab domain.

[0123] In one embodiment, an antibody construct is described herein, the antibody construct comprising: (i) a Fab domain capable of binding CD28 on a first cytotoxic effector cell; (ii) a first scFv domain and a second scFv domain, wherein the first scFv domain is capable of binding CD3 on a second cytotoxic effector cell and the second scFv domain is capable of binding a tumor-associated antigen (TAA) on a tumor cell; and (iii) an 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 C-terminus of the first Fc polypeptide, and (c) the second scFv domain is coupled to the N-terminus of the V H domain of the Fab domain.

[0124] In various embodiments, the dimeric Fc domain of the trivalent trispecific antibody construct is a heterodimeric Fc domain.

[0125] In some embodiments, the first scFv domain has a domain structure from the N-terminus to the C-terminus of V H -V L As further described herein, the first scFv domain may also comprise a linker that couples the first scFv V H sequence to the first scFv V L sequence. scFv1

[0126] In certain other embodiments, the first scFv domain has a domain structure from the N-terminus to the C-terminus of V L -V H As further described herein, the first scFv domain may also comprise a linker that couples the first scFv V L sequence to the first scFv V H sequence.scFv1 。

[0127] In some embodiments, the second scFv domain has a domain structure of V from the N-terminus to the C-terminus H -V L As further described herein, the second scFv domain may also include a linker that couples the second scFv V H sequence to the second scFv V L sequence scFv2 。

[0128] In certain embodiments, the second scFv domain has a domain structure of V from the N-terminus to the C-terminus L -V H As further described herein, the second scFv domain may also include a linker that couples the second scFv V L sequence to the second scFv V H sequence scFv2 。

[0129] In some embodiments, the first scFv domain has a domain structure of V from the N-terminus to the C-terminus L -V H and the second scFv domain has a domain structure of V from the N-terminus to the C-terminus L -V H 。

[0130] In some embodiments, the first scFv domain has a domain structure of V from the N-terminus to the C-terminus H -V L and the second scFv domain has a domain structure of V from the N-terminus to the C-terminus H -V L 。

[0131] In some embodiments, the first scFv domain has a domain structure of V from the N-terminus to the C-terminus L -V H and the second scFv domain has a domain structure of V from the N-terminus to the C-terminus H -V L 。

[0132] In some embodiments, the first scFv domain has a domain structure of V from the N-terminus to the C-terminus H -V L and the second scFv domain has a domain structure of V from the N-terminus to the C-terminus L -V H 。

[0133] In various embodiments of the trivalent trispecific antibody constructs of the present disclosure, the TAA is MSLN or Cldn18.2, as further described herein.

[0134] As further described herein, the antibody constructs of the present disclosure may also comprise one or more linkers. Such one or more linkers may be one or more peptide linkers. One or more linkers may couple one, two, or more domains and / or sequences of the antibody construct to each other. As an example, the linker herein scFv may couple the scFv V H sequence to the scFvV L sequence.

[0135] B. Domains of the Antibody Construct

[0136] The trivalent trispecific antibody constructs of the present disclosure may comprise one or more antibody domains. In various embodiments, the trivalent trispecific antibody constructs of the present disclosure comprise multiple (i.e., two or more) antibody domains. Such multiple antibody domains may comprise (i) one or more Fc domains, wherein the Fc domain may comprise a first Fc polypeptide and a second Fc polypeptide and may be homodimeric or heterodimeric; (ii) one or more Fab domains, wherein the Fab domain may comprise a heavy chain polypeptide, which comprises a heavy chain variable domain (V H ) sequence and a heavy chain constant domain (C H1 ) sequence, and a light chain polypeptide, which comprises a light chain variable domain (V L ) sequence and a light chain constant domain (C L ) sequence; and (iii) one or more scFv domains, wherein the scFv domain may comprise an scFv V L sequence coupled to an scFv V H sequence. The various domains that the antibody construct may comprise are further described herein.

[0137] An Ig structural unit typically consists of two pairs of polypeptide chains, each pair having one "light" chain (about 25 kilodaltons (kD)) and one "heavy" chain (about 50 - 70 kD). The light chain can be classified as κ or λ. The "class" of an Ig refers to the type of constant domain possessed by its heavy chain. There are five main classes of antibodies: IgA, IgD, IgE, IgG, and IgM, and several of these classes can be further divided into subclasses (isotypes), e.g., IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2. The heavy chain constant domains corresponding to the different classes of immunoglobulins (Ig) are called alpha (α), delta (δ), epsilon (ε), gamma (γ), and mu (μ), respectively.

[0138] In various embodiments, the trivalent trispecific antibody constructs described herein are based on IgG-class immunoglobulins, such as IgG1, IgG2, IgG3, or IgG4 immunoglobulins. In some embodiments, the antibody constructs described herein are based on IgG1, IgG2, or IgG4 immunoglobulins. In certain embodiments, the antibody constructs described herein are based on IgG1 immunoglobulins. In the context of the present disclosure, when an antibody construct is based on a particular Ig isotype, it refers to an antibody construct that contains all or part of the constant region (i.e., the Fc domain) of the particular Ig isotype. It should be understood that, in certain embodiments according to the present disclosure, the antibody constructs may also contain hybrids of isotypes and / or subclasses.

[0139] Generally, the N-terminal domain of each polypeptide chain in an antibody typically defines a variable region of about 100 to 110 or more amino acids that is primarily responsible for antigen recognition (e.g., V H or V L ). The terms "variable light chain (V L )" and "variable heavy chain (V H )" refer to these domains in the light and heavy chains, respectively. As described herein, in various embodiments, the trivalent trispecific antibody constructs of the present disclosure may contain two or more variable domain sequences. In various embodiments, each binding domain of the trivalent trispecific antibody constructs of the present disclosure contains two variable domains, and thus can include a total of six variable domain sequences by being trivalent (i.e., containing three binding domains) and having trispecificity (i.e., each of the three binding domains targets a different antigen), such as three V H domain sequences and three V L domain sequences.

[0140] In some embodiments, two or more of such variable domains are coupled to each other in tandem and in a single polypeptide chain format, for example, as described for an scFv-type binding domain containing (from the N-terminus to the C-terminus) a V L domain coupled to a V H domain, or as described for a construct in which an scFv domain (which contains two variable domain sequences coupled to each other in tandem) is coupled to a Fab domain via a Fab variable domain sequence (e.g., Fab V H or Fab V L sequence). In some embodiments, such a construct may contain, from the N-terminus to the C-terminus, [(V H -V L ) or (V L -V H )] scFv -V H -CH1 -Fc heavy chain domain structure.

[0141] Thus, antibody constructs derived from Ig molecules of the present disclosure may contain different Ig domains within their heavy and light chains. The heavy chain domains may include an Fc domain (or Fc region) (e.g., containing CH2 and CH3 domains), a hinge domain (or hinge region), a heavy chain Fab domain containing a variable heavy chain domain (V H ) and a constant heavy chain domain (C H1 ), and the light chain domains may include a variable light chain domain (V L ) and a light chain constant domain (C L ). In some embodiments, and according to certain nomenclatures, the "Fc domain" may include the CH2 and CH3 domains as well as the hinge domain (or hinge region).

[0142] B.1 Complementary determining regions (CDRs) and binding domains

[0143] Each V H and V L domain in the antibody constructs herein has three loops that are highly variable in sequence and form the antigen-binding site. Each of these loops is referred to as a "hypervariable region" or "HVR" or "complementary determining region" or "CDR". The terms hypervariable region (HVR) and complementary determining region (CDR) are used interchangeably herein to refer to the portion of a variable domain (e.g., V H or V L ) that forms the antigen-binding site. Except for CDR1 in V H , CDRs generally contain the amino acid residues that form the hypervariable loops. The V H and V L domains are composed of relatively invariant stretches called framework regions (FRs), which are between about 15 and 30 amino acids in length, separated by shorter CDRs, each CDR generally being between about 5 and 15 amino acids in length, but occasionally can be longer or shorter. The three CDRs and four FRs that make up each V H and V L domain are arranged from the N-terminus to the C-terminus as follows: FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4.

[0144] Several different definitions and numbering conventions for CDR regions in Ig molecules are commonly used, including those described by Kabat et al. (1983, Sequences of Proteins of Immunological Interest, NIH Publication No. 369 - 847, Bethesda, MD), Chothia et al. (1987, J Mol Biol, 196:901 - 917), as well as those described by the IMGT, AbM, and Contact definitions. These different definitions include the superposition or subsets of amino acid residues when compared to each other. For example, the CDR definitions according to Kabat, Chothia, IMGT, AbM, and Contact are provided in Table 1 below.

[0145] Thus, as will be apparent to those skilled in the art, the exact numbering and placement of CDRs can vary based on the numbering system employed. However, it should be understood that the disclosure herein of the heavy chain variable domain (V H ) includes the disclosure of the relevant (native) heavy chain CDRs (HCDRs) as defined by any known numbering system. Similarly, the disclosure herein of the variable light domain (V L ) includes the disclosure of the relevant (native) light chain CDRs (LCDRs) as defined by any known numbering system. Those skilled in the art will recognize that a limited number of amino acid substitutions can be introduced into the CDR sequences or V H or V L sequences of a known antibody without the antibody losing its ability to bind to its target, such as a decrease in binding affinity by at least about 1000 - fold or more. Candidate amino acid substitutions can be identified by computer modeling or by techniques such as alanine scanning, and the binding activity of the resulting variants (e.g., expressed as binding affinity, e.g., given as measured EC 50 values) is tested by standard techniques. As an example, in certain embodiments, the CD3 - binding domain of the antibody constructs described herein comprises a set of CDRs (i.e., heavy chain CDR1, CDR2, and CDR3 and light chain CDR1, CDR2, and CDR3) that have 90% or higher, 95% or higher, 98% or higher, 99% or higher, or 100% sequence identity to the amino acid sequences shown in SEQ ID NOs: 321 - 326, respectively, wherein the binding domain retains or substantially retains the ability to bind CD3. In this context, the term "substantially" means 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%.

[0146] Table 1: Common CDR Definitions 1

[0147]

[0148] 1 Except for Contacts using the Chothia numbering, for all definitions, the Kabat or Chothia numbering system can be used for HCDR2, HCDR3, and light chain CDRs.

[0149] 2 Use the Kabat numbering. The positions that distinguish the ends of the Chothia and IMGT CDR-H1 loops in the Kabat numbering scheme change according to the length of the loop because Kabat made insertions at positions 35A and 35B outside those CDR definitions. The IMGT and Chothia CDR-H1 loops can be clearly defined using the Chothia numbering. The CDR-H1 definitions using the Chothia numbering are: Kabat H31-H35, Chothia H26-H32, AbM H26-H35, IMGT H26-H33, Contact H30-H35.

[0150] In some embodiments, the antibody constructs described herein comprise at least one Ig domain from a mammalian Ig such as bovine Ig, human Ig, camel Ig, rat Ig, or mouse Ig. In some embodiments, the antibody constructs herein can be chimeric constructs comprising two or more Ig domains, wherein at least one domain is from a first mammalian Ig, such as human Ig, and at least a second domain is from a second mammalian Ig, such as mouse or rat Ig. In other embodiments, the antibody constructs can be derived from Igs from different species, e.g., the antibody constructs 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 class of chimeric antibodies that contain a minimal sequence derived from a non-human antibody. In some embodiments, the antibody constructs herein can comprise at least one Ig constant domain from a human Ig. In various embodiments, all domains of the antibody constructs described herein can be from (or derived from) human Ig.

[0151] In some embodiments, and as further described herein, one or more domains of the antibody construct can be modified (e.g., the amino acid sequence) to further improve the properties and performance of the antibody construct (e.g., antigen affinity, stability, and / or pharmacokinetics, etc.). For example, framework region (FR) residues of human Ig can be replaced with corresponding non-human residues, or a humanized antibody can contain residues not found in either the recipient antibody or the donor antibody. Generally, the variable domain or humanized antibody domain in a humanized antibody contains 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 Fc polypeptides to form a heterodimeric Fc domain rather than a homodimeric Fc domain.

[0152] In some embodiments, the present disclosure relates to antibody constructs that can have different valences, e.g., can be bivalent or trivalent. Thus, in various embodiments, the antibody constructs herein contain two or three antigen-binding domains, i.e., are at least bivalent or at least trivalent. In various embodiments of the present disclosure, the antibody construct can be trispecific and trivalent, and thus such an antibody construct can contain three binding domains. Each of the three binding domains can have a unique binding specificity for an antigen (the same epitope / antigen or different epitopes / antigens). In some embodiments, the trispecific trivalent antibody construct of the present disclosure contains three binding domains capable of binding three different antigens (e.g., CD3, CD28, and TAA), e.g., one or more Fab domains and / or one or more scFv domains.

[0153] In some embodiments, the antibody constructs of the present disclosure can contain (i) one or more Fab domains, (ii) one or more scFv domains, and (iii) an Fc domain containing a first Fc polypeptide and a second Fc polypeptide. In various embodiments, the trivalent trispecific antibody construct as described herein contains (i) a Fab domain capable of binding to a first antigen, (ii) a first scFv domain capable of binding to a second antigen, (iii) a second scFv domain capable of binding to a third antigen, and (iv) a heterodimeric Fc domain containing a first Fc polypeptide and a second Fc polypeptide.

[0154] Generally, as used herein, a "Fab domain" contains a constant region that contains the constant domain of the light chain (C L ) and the first constant domain of the heavy chain (C H1 ); and a variable region that contains the variable domains V L and V H. In some embodiments, the Fab domain can be a single-chain Fab. The single-chain Fab can be a Fab molecule in which the Fab light chain and the Fab heavy constant chain are linked by a peptide linker to form a single polypeptide chain. In such embodiments, typically, the C-terminus of the Fab light chain is linked to the N-terminus of the Fab heavy chain in the single-chain Fab molecule; however, other formats are also encompassed herein. However, in various embodiments herein, the Fab domain of the antibody construct is formed by two separately expressed polypeptide chains, namely the light chain and the heavy chain (or portions thereof). However, the heavy chain and light chain portions of the Fab domain can be linked to each other by covalent bonds such as disulfide bonds.

[0155] As used herein, a "scFv domain" generally comprises the heavy chain variable domain (V H ) and the light chain variable domain (V L ) in a single polypeptide chain format. The scFv can optionally include a peptide linker between the V H and V L domains, which can assist the scFv in forming a functional structure for antigen binding. Thus, in various embodiments, the scFv domain herein can include a V scFv domain coupled to the N-terminus of the V H domain via its C-terminus by a linker L , i.e., the scFv domain can have the domain structure: V L -linker scFv -V H , or alternatively, the scFv can comprise a V scFv domain linked to the N-terminus of V L via its C-terminus by a linker H , i.e., having the domain structure: V H -linker scFv -V L .

[0156] In some embodiments, the antibody constructs described herein can further comprise another domain or portion that may not be derived from an Ig molecule. Such a non-Ig domain can be referred to as a moiety. Such a moiety can be a detectable label (e.g., a radioactive label or a fluorescent label), a low molecular weight (e.g., <750 Da) drug molecule, another peptide (e.g., a signal peptide) or polypeptide molecule, or a combination thereof.

[0157] B.2 Binding Domains Against Antigens on Cytotoxic Effector Cells

[0158] As further described herein, the antibody constructs of the disclosure can include at least two binding domains capable of binding to one or more molecules (e.g., polypeptides) on the surface of one or more cytotoxic effector cells. Such one or more cytotoxic effector cells can be one or more immune cells. Such one or more immune cells can include T cells, macrophages, dendritic cells, neutrophils, B cells, NK cells, or combinations thereof.

[0159] In various embodiments, the trivalent trispecific antibody constructs of the disclosure include at least one binding domain capable of binding a first antigen on a first cytotoxic effector cell and at least one binding domain capable of binding a second antigen on a second cytotoxic effector cell.

[0160] In various embodiments, the first cytotoxic effector cell and the second cytotoxic effector cell are different cells, i.e., the first antigen and the second antigen are on the surfaces of different cells. In other embodiments, the first cytotoxic effector cell and the second cytotoxic effector cell are the same cell, e.g., the first antigen and the second antigen to which the antibody construct binds are expressed by the same cell, i.e., are on the surface of the same cell. In some embodiments, the first cytotoxic effector cell and the second cytotoxic effector cell include or consist of T cells.

[0161] In some embodiments, the first antigen is cluster of differentiation 3 (CD3) and the second antigen is CD28. In such embodiments, both CD3 and CD28 can be bound by the trivalent trispecific antibody construct when on the surface of the same cytotoxic effector cell. In other embodiments, the antigens CD3 and CD28 can both be bound by the trivalent trispecific antibody construct when on the surfaces of different cytotoxic effector cells.

[0162] Thus, in some embodiments, the present disclosure relates to trivalent trispecific antibody constructs that comprise a first binding domain capable of binding CD3, a second binding domain capable of binding CD28, and a third binding domain capable of binding a TAA. Thus, the antibody constructs described herein may also be referred to as "T cell engagers", "TCEs", or "T cell engager molecules", describing the ability of the constructs to bind antigens on one or more T cells and TAAs on tumor cells. In some embodiments, such as in the tumor (micro)environment, the engagement of a trivalent trispecific antibody construct with two different antigens on one or more T cells and an antigen on tumor cells can be—at least transiently—simultaneous, thereby establishing a TCR-independent immune synapse and directing T cell-mediated cytotoxic activity to the tumor environment containing tumor cells expressing the TAA. In various embodiments, as further described herein, the trivalent trispecific antibody constructs can cause a significant reduction in immune cell (e.g., T cell) activation in the absence of a TAA, such as when an immune synapse cannot be fully formed due to the absence of a TAA.

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

[0164] Thus, in some embodiments, the two binding domains capable of binding CD3 and CD28 on a first cytotoxic effector cell and a second cytotoxic effector cell are Fab domains, respectively.

[0165] In other embodiments, the two binding domains capable of binding CD3 and CD28 on a first cytotoxic effector cell and a second cytotoxic effector cell are scFv domains, respectively.

[0166] In yet other embodiments, the first binding domain capable of binding CD28 on a first cytotoxic effector cell is a Fab domain, and the second binding domain capable of binding CD3 on a second cytotoxic effector cell is an scFv domain.

[0167] In another embodiment, the first binding domain capable of binding CD28 on a first cytotoxic effector cell is an scFv domain, and the second binding domain capable of binding CD3 on a second cytotoxic effector cell is a Fab domain.

[0168] In various embodiments, the Fab domain capable of binding CD3 or CD28 can comprise a heavy chain constant domain (CH1 ), the heavy chain constant domain comprises or consists of an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 107. In some embodiments, such a C H1 domain sequence comprises or consists of an amino acid sequence having at least about 90% sequence identity with the amino acid sequence shown in SEQ ID NO: 107. In some embodiments, such a C H1 domain sequence comprises or consists of an amino acid sequence having at least about 95% sequence identity with the amino acid sequence shown in SEQ ID NO: 107. In some embodiments, such a C H1 domain sequence comprises or consists of an amino acid sequence having at least about 97% sequence identity with the amino acid sequence shown in SEQ ID NO: 107. In some embodiments, such a C H1 domain sequence comprises or consists of an amino acid sequence having at least about 99% sequence identity with the amino acid sequence shown in SEQ ID NO: 107. In some embodiments, such a C H1 domain sequence comprises or consists of the amino acid sequence shown in SEQ ID NO: 107.

[0169] B.2.1 Binding domain against CD3

[0170] As described herein, in some embodiments, the trivalent trispecific antibody construct of the present disclosure comprises a binding domain capable of binding CD3.

[0171] In various embodiments, the anti-CD3 binding domain (e.g., scFv domain or Fab domain) of the antibody construct herein has an affinity for CD3 (given by the EC 50 value for binding CD3) that may not exceed about 1 nM, 5 nM, 10 nM, 20 nM or not exceed about 30 nM. Thus, in various embodiments, the anti-CD3 binding domain herein has an EC 50 value for binding CD3 that does not exceed about 20 nM to about 80 nM, about 30 nM to about 60 nM, or about 40 nM to about 50 nM. In some embodiments, the anti-CD3 binding domain herein has an EC 50 value for binding CD3 that does not exceed about 30 nM, 40 nM, 50 nM or about 60 nM. In various embodiments, the antibody construct herein comprises an anti-CD3 binding domain having an EC 50 value for binding CD3 of about 20 nM to about 40 nM, such as about 30 nM.

[0172] In some embodiments, the anti-CD3 binding domain is capable of binding CD3 with a dissociation constant of 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.

[0173] In various embodiments, the antibody constructs herein comprise a binding domain capable of binding CD3 on a T cell, wherein such binding domain binds CD3 with an EC 50 value of from about 20 nM to about 40 nM, such as about 30 nM, and comprises CDR sequences of the V H sequences as shown in SEQ ID NO: 321 - 323 and CDR sequences of the V L sequences as shown in SEQ ID NO: 324 - 326.

[0174] In various embodiments, the antibody constructs herein comprise a binding domain capable of binding CD3 on a T cell, wherein such binding domain binds CD3 with an EC 50 value of from about 20 nM to about 40 nM, such as about 30 nM, and comprises a V H domain comprising: an HCDR1 comprising the sequence GVTFNYYG (SEQ ID NO: 321), an HCDR2 comprising the sequence ITSSGGRI (SEQ ID NO: 322), and an HCDR3 comprising the sequence TLDGRDGWVAY (SEQ ID NO: 323); and a V L domain comprising: an LCDR1 comprising the sequence TGNIGSNY (SEQ ID NO: 324), an LCDR2 comprising the sequence RND (SEQ ID NO: 325), and an LCDR3 comprising the sequence QSYSSGFI (SEQ ID NO: 326).

[0175] In some embodiments, the anti-CD3 binding domain comprises a V H domain comprising an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99% or 100% sequence identity to the sequence shown in SEQ ID NO: 102 or consisting of said amino acid sequence; and a V L domain comprising an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99% or 100% sequence identity to the sequence shown in SEQ ID NO: 103 or consisting of said amino acid sequence. In certain embodiments, the anti-CD3 binding domain of the antibody constructs herein comprises a V H domain comprising the amino acid sequence shown in SEQ ID NO: 102 or consisting of said amino acid sequence; and a VL A domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 103.

[0176] In some embodiments, the CDRs of the anti-CD3 complementarity determining regions used in the antibody constructs of the present disclosure comprise one or more amino acid modifications in one or more of the CDR sequences shown in SEQ ID NOs: 321 - 326, wherein at least about 80%, 90%, or 95% of the binding affinity for CD3 is retained as compared to a complementarity determining region without such amino acid modifications.

[0177] In some embodiments, the anti-CD3 binding domain comprised by the antibody constructs described herein can comprise or consist of an scFv domain or a Fab domain.

[0178] The V of the anti-CD3 scFv or Fab domain of the antibody constructs herein H 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 shown in SEQ ID NO: 102. In some embodiments, the V of such an scFv or Fab domain H domain comprises or consists of an amino acid sequence having at least about 90% sequence identity to the amino acid sequence shown in SEQ ID NO: 102. In some embodiments, the V of such an scFv or Fab domain H domain comprises or consists of an amino acid sequence having at least about 95% sequence identity to the amino acid sequence shown in SEQ ID NO: 102. In some embodiments, the V of such an scFv or Fab domain H domain comprises or consists of an amino acid sequence having at least about 97% sequence identity to the amino acid sequence shown in SEQ ID NO: 102. In some embodiments, the V of such an scFv or Fab domain H domain comprises or consists of an amino acid sequence having at least about 99% sequence identity to the amino acid sequence shown in SEQ ID NO: 102. In other embodiments, the V of such an scFv or Fab domain H domain comprises or consists of the amino acid sequence shown in SEQ ID NO: 102.

[0179] The V of the anti-CD3 scFv or Fab domain of the antibody constructs herein LThe domain can be part of a light chain (e.g., L1) that pairs with the anti-CD3 Fab domain sequence of the heavy chain in the case of a Fab domain, and 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 shown in SEQ ID NO: 103. In some embodiments, the V L domain of such an scFv or Fab domain comprises or consists of an amino acid sequence having at least about 90% sequence identity to the amino acid sequence shown in SEQ ID NO: 103. In some embodiments, the V L domain of such an scFv or Fab domain comprises or consists of an amino acid sequence having at least about 95% sequence identity to the amino acid sequence shown in SEQ ID NO: 103. In some embodiments, the V L domain of such an scFv or Fab domain comprises or consists of an amino acid sequence having at least about 97% sequence identity to the amino acid sequence shown in SEQ ID NO: 103. In some embodiments, the V L domain of such an scFv or Fab domain comprises or consists of an amino acid sequence having at least about 99% sequence identity to the amino acid sequence shown in SEQ ID NO: 103. In various embodiments, the V L domain of such an scFv or Fab domain comprises or consists of the amino acid sequence shown in SEQ ID NO: 103 or consists of said amino acid sequence.

[0180] In embodiments where the anti-CD3 domain is a Fab domain that further comprises a C H1 domain and a C L domain in its heavy and light chains, respectively, the C H1 domain of the anti-CD3 Fab domain of the 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 shown in SEQ ID NO: 107. In some embodiments, the C H1 domain of such a Fab domain comprises or consists of an amino acid sequence having at least about 90% sequence identity to the amino acid sequence shown in SEQ ID NO: 107. In some embodiments, the C H1The domain comprises or consists of an amino acid sequence having at least about 95% sequence identity with the amino acid sequence shown in SEQ ID NO: 107. In some embodiments, the C H1 The domain comprises or consists of an amino acid sequence having at least about 97% sequence identity with the amino acid sequence shown in SEQ ID NO: 107. In some embodiments, the C H1 The domain comprises or consists of an amino acid sequence having at least about 99% sequence identity with the amino acid sequence shown in SEQ ID NO: 107. In some embodiments, the C H1 The domain comprises or consists of the amino acid sequence shown in SEQ ID NO: 107.

[0181] In addition, the C L domain of the anti-CD3 Fab domain of the antibody constructs herein can comprise or consist of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 121. In some embodiments, the C L domain comprises or consists of an amino acid sequence having at least about 90% sequence identity with the amino acid sequence shown in SEQ ID NO: 121. In some embodiments, the C L domain comprises or consists of an amino acid sequence having at least about 95% sequence identity with the amino acid sequence shown in SEQ ID NO: 121. In some embodiments, the C L domain comprises or consists of an amino acid sequence having at least about 97% sequence identity with the amino acid sequence shown in SEQ ID NO: 121. In some embodiments, the C L domain comprises or consists of an amino acid sequence having at least about 99% sequence identity with the amino acid sequence shown in SEQ ID NO: 121. In other embodiments, the C L domain comprises or consists of the amino acid sequence shown in SEQ ID NO: 121.

[0182] In some embodiments, the anti-CD3 binding domain of the antibody construct is a scFv domain. In some embodiments, such anti-CD3 scFv domain has V from the N-terminus to the C-terminus H- linker scFv -V L domain structure. In other embodiments, the anti - CD3 scFv domain has V from the N - terminus to the C - terminus L - linker scFv -V H domain structure. In any of these embodiments, as further described herein, the V H domain may comprise or consist of an amino acid sequence having at least about 95%, 97%, 99%, or 100% sequence identity with the amino acid sequence shown in SEQ ID NO:102, the V L domain may comprise or consist of an amino acid sequence having at least about 95%, 97%, 99%, or 100% sequence identity with the amino acid sequence shown in SEQ ID NO:102, and the linker scFv may comprise or consist of the amino acid sequence shown in SEQ ID NO:104.

[0183] Thus, in some embodiments, the anti - CD3 scFv domain may have V from the N - terminus to the C - terminus H - linker scFv -V L or V L - linker scFv -V H domain structure, wherein the V H domain comprises or consists of the amino acid sequence shown in SEQ ID NO:102, the V L domain comprises or consists of the amino acid sequence shown in SEQ ID NO:103, and the linker scFv comprises or consists of the amino acid sequence shown in SEQ ID NO:104. In some embodiments, the anti - CD3 scFv domain has V from the N - terminus to the C - terminus H - linker scFv -V L domain structure. In other embodiments, the anti - CD3 scFv domain has V from the N - terminus to the C - terminus L - linker scFv -V H domain structure.

[0184] In some embodiments, the anti - CD3 binding domain of the 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 where the heavy chain also comprises an Fc portion, etc.) and a light chain. The heavy chain of the anti - CD3 Fab domain may comprise CH1 Domain-coupled V H domain (from the N-terminus to the C-terminus) or consists of said domain. In various embodiments, the anti-CD3 Fab domain comprises (i) a V H domain that comprises or consists of an amino acid sequence having at least about 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 102; (ii) a V L domain that comprises or consists of an amino acid sequence having at least about 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 103; (iii) a C H1 domain that comprises or consists of an amino acid sequence having at least about 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 107; and (iv) a C L domain that comprises or consists of an amino acid sequence having at least about 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 121.

[0185] In certain embodiments, the anti-CD3 Fab domain comprises a V H domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 102; a C H1 domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 107; a V L domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 103; and a C L domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 121.

[0186] In various embodiments, and as further described herein, when a portion of the heavy chain (e.g., H1) comprising the V H domain and the C H1 domain pairs with the light chain (L1) comprising the V L domain and the CL domain, the anti-CD3 Fab domain of the antibody construct can be formed. In some embodiments, comprising V H and C H1The domain and the heavy chain of the anti-CD3 Fab domain or a part thereof may have the amino acid sequence shown in SEQ ID NO: 154, and the corresponding light chain paired with the anti-CD3 part of the heavy chain may have the amino acid sequence shown in SEQ ID NO: 120.

[0187] In some embodiments, the anti-CD3 binding domain of the antibody constructs herein comprises or consists of an scFv domain. Such an scFv domain may comprise a V H domain and a V L domain or consists of said domains. In some embodiments, the V H domain is coupled to the V scFv domain via a linker L In some embodiments, the V H domain comprises or consists of the amino acid sequence shown in SEQ ID NO: 102, and the V L domain comprises or consists of the amino acid sequence shown in SEQ ID NO: 103. Thus, in certain embodiments, the anti-CD3 binding domain of the antibody constructs herein is an scFv domain that comprises a V scFv domain coupled to a V L domain via a linker H or consists of said domains, the V H domain comprises or consists of the amino acid sequence shown in SEQ ID NO: 102, the V L domain comprises or consists of the amino acid sequence shown in SEQ ID NO: 103, and the linker scFv comprises or consists of the amino acid sequence shown in SEQ ID NO: 104.

[0188] As further described herein, in some embodiments, the binding affinity of the trivalent trispecific antibody constructs of the present disclosure for one or more specific targets (e.g., CD3) may, at least in part, depend on its format and / or geometry. As an example, the relative orientation and proximity of the anti-CD3 binding domains in the antibody construct can affect the ability of the binding domains to interact with the target epitopes to which they are to bind, such as through steric hindrance, conformational changes (e.g., the conditional freedom of the binding domains, such as spatial flexibility) that occur when the construct interacts with one or more targets.

[0189] In additional embodiments, the CD3 binding affinity of the anti-CD3 binding domain of the antibody constructs herein can be engineered and altered (e.g., increased / decreased relative to the unmodified domain), e.g., by using one or more amino acid modifications. In some embodiments, the trispecific trivalent antibody constructs of the present disclosure can comprise variant anti-CD3 binding domains that, compared to the anti-CD3 binding domains comprising, e.g., the VH sequence shown in SEQ ID NO: 102 and the VL sequence shown in SEQ ID NO: 103 as described herein, comprise one or more amino acid modifications in their VH and / or VL domains. Such one or more amino acid modifications can decrease or increase the binding affinity of the variant anti-CD3 binding domain for CD3 compared to the binding affinity of the corresponding anti-CD3 binding domain that does not contain such one or more amino acid modifications.

[0190] In some embodiments, the one or more amino acid modifications for altering the binding affinity of the anti-CD3 binding domain can include one or more amino acid substitutions, one or more amino acid additions, and / or one or more amino acid deletions. In certain embodiments, the one or more amino acid modifications for altering the binding affinity of the anti-CD3 binding domain comprise one or more amino acid substitutions relative to the unmodified binding domain sequence (e.g., anti-CD3 VH or VL sequence) or consist of said one or more amino acid substitutions.

[0191] In some embodiments, the anti-CD3 affinity of the 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 that of the corresponding parental anti-CD3 binding domain.

[0192] In some embodiments, the trivalent trispecific antibody constructs of the present disclosure can have a variety of different anti-CD3 and anti-CD28 (described further below) binding affinities, providing a range of constructs with different target engagement characteristics, particularly when considering that construct format and geometry can further influence antigen binding (see, e.g., Figure 22A - 22C ).

[0193] B.2.2 Binding Domain Against CD28

[0194] As described herein, in various embodiments, the trivalent trispecific antibody constructs of the present disclosure comprise a binding domain capable of binding CD28.

[0195] In various embodiments, the affinity of such anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein for CD28 (as the EC for binding CD28 50value (i.e., K D value) given) can be from about 10 nM to about 500 nM or from about 20 nM to about 250 nM.

[0196] In some embodiments, the antibody constructs herein comprise the V H sequences of the CDRs shown in SEQ ID NOs: 300, 303, and 307 and the V L sequences of the CDRs shown in SEQ ID NOs: 313, 316, and 320.

[0197] In some embodiments, the antibody constructs herein comprise a binding domain (e.g., an scFv domain or a Fab domain) capable of binding to CD28 on T cells, wherein the binding domain has an affinity for CD28 of from about 15 nM to about 35 nM and comprises an anti-CD28 V H sequence comprising an HCDR1 having the sequence SYGVH (SEQ ID NO: 300), an HCDR2 having the sequence VIWPGGGTNFNSALMS (SEQ ID NO: 303), and an HCDR3 having the sequence DRAYGNYLYAMDY (SEQ ID NO: 307); and an anti-CD28 V L sequence comprising an LCDR1 having the sequence RASESVEYYVTSLMQ (SEQ ID NO: 313), an LCDR2 having the sequence AASNVDS (SEQ ID NO: 316), and an LCDR3 having the sequence QQSRKVPFT (SEQ ID NO: 320).

[0198] In such embodiments, the antibody constructs herein comprise a binding domain (e.g., an scFv domain or a Fab domain) capable of binding to CD28 on T cells, wherein the binding domain comprises a V H domain comprising an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence shown in SEQ ID NO: 106 or consisting of said amino acid sequence; and a V L domain comprising an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence shown in SEQ ID NO: 116 or consisting of said amino acid sequence. In some embodiments, the anti-CD28 binding domain comprises a V H domain comprising the sequence shown in SEQ ID NO: 106 or consisting of said sequence; and a V L domain comprising the sequence shown in SEQ ID NO: 116 or consisting of said sequence.

[0199] In other embodiments, the trivalent trispecific antibody construct of the present disclosure comprises an anti-CD28 binding domain that comprises (i) a V H domain that comprises a sequence having one or more amino acid substitutions as compared to the sequence shown in SEQ ID NO: 106; and / or (ii) a V L domain that comprises a sequence having one or more amino acid substitutions as compared to the sequence shown in SEQ ID NO: 116, wherein the positions of such amino acid substitutions are provided according to the IMGT numbering system.

[0200] In some embodiments, compared to the binding affinity of an anti-CD28 binding domain that does not contain such V H and / or V L sequences (e.g., binding domains comprising the V H and V L sequences shown in SEQ ID NO: 106 and 116, respectively), one or more such amino acid substitutions in the anti-CD28 V H domain and / or the anti-CD28 V L domain can reduce the binding affinity of the corresponding anti-CD28 binding domain 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 about 1.5-fold to about 25-fold, about 2.0-fold to about 20-fold, about 3.0-fold to about 20-fold, or about 5.0-fold to about 10-fold.

[0201] Thus, in some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody construct herein has an affinity for CD28 (given as the EC 50 value for binding to CD28, i.e., the K D value) that can be about 10 nM to about 500 nM, about 20 nM to about 600 nM, about 20 nM to about 250 nM, about 20 nM to about 150 nM, about 20 nM to about 100 nM, or about 20 nM to about 50 nM.

[0202] In some embodiments, the anti-CD28 binding domain herein comprises a V H domain that contains the amino acid substitution P1058A relative to the amino acid sequence shown in SEQ ID NO: 106. In some embodiments, the anti-CD28 V H domain, the anti-CD28 V HThe domain contains the amino acid sequence shown in SEQ ID NO: 204.

[0203] In some embodiments, the anti-CD28 binding domain herein contains a V H domain, and this V H domain contains the amino acid substitution G1064S relative to the amino acid sequence shown in SEQ ID NO: 106. In some embodiments, the anti-CD28 V H domain contains the amino acid sequence shown in SEQ ID NO: 207.

[0204] In some embodiments, the anti-CD28 binding domain herein contains a V L domain, and this V L domain contains the amino acid substitution V1035G relative to the amino acid sequence shown in SEQ ID NO: 116. In some embodiments, the anti-CD28 V L domain contains the amino acid sequence shown in SEQ ID NO: 200.

[0205] In some embodiments, the anti-CD28 binding domain herein contains a V L domain, and this V L domain contains the amino acid substitution D1068E relative to the amino acid sequence shown in SEQ ID NO: 116. In some embodiments, the anti-CD28 V L domain contains the amino acid sequence shown in SEQ ID NO: 209.

[0206] In some embodiments, the anti-CD28 binding domain herein contains a V H domain, and this V H domain contains the amino acid substitution E1080K relative to the amino acid sequence shown in SEQ ID NO: 106. In some embodiments, the anti-CD28 V H domain contains the amino acid sequence shown in SEQ ID NO: 203.

[0207] In some embodiments, the anti-CD28 binding domain herein contains a V H domain, and this V H domain contains the amino acid substitution Y1110S relative to the amino acid sequence shown in SEQ ID NO: 106. In some embodiments, the anti-CD28 V H domain contains the amino acid sequence shown in SEQ ID NO: 201.

[0208] In some embodiments, the anti-CD28 binding domain herein contains a V HDomain, the V H Domain contains the amino acid substitution N1111aA relative to the amino acid sequence shown in SEQ ID NO: 106. In some embodiments, the anti-CD28 V H Domain contains the amino acid sequence shown in SEQID NO: 208.

[0209] In some embodiments, the anti-CD28 binding domain herein contains V H Domain, the V H Domain contains the amino acid substitution Y1112S relative to the amino acid sequence shown in SEQ ID NO: 106. In some embodiments, the anti-CD28 V H Domain contains the amino acid sequence shown in SEQID NO: 113.

[0210] In some embodiments, the anti-CD28 binding domain herein contains V H Domain, the V H Domain contains the amino acid substitution L1112aN relative to the amino acid sequence shown in SEQ ID NO: 106. In some embodiments, the anti-CD28 V H Domain contains the amino acid sequence shown in SEQID NO: 206.

[0211] In some embodiments, the anti-CD28 binding domain herein contains V H Domain, the V H Domain contains the amino acid substitution Y1113S relative to the amino acid sequence shown in SEQ ID NO: 106. In some embodiments, the anti-CD28 V H Domain contains the amino acid sequence shown in SEQID NO: 210.

[0212] In some embodiments, the anti-CD28 binding domain herein contains V H Domain, the V H Domain contains the amino acid substitution Y1037A relative to the amino acid sequence shown in SEQ ID NO: 106. In some embodiments, the anti-CD28 V H Domain contains the amino acid sequence shown in SEQID NO: 205.

[0213] In some embodiments, the anti-CD28 binding domain herein contains V L Domain, the V L Domain contains the amino acid substitution Y1031A relative to the amino acid sequence shown in SEQ ID NO: 116. In some embodiments, the anti-CD28 V LThe domain contains the amino acid sequence shown in SEQ ID NO: 151.

[0214] In some embodiments, the anti-CD28 binding domain herein contains a V L domain, and this V L domain contains the amino acid substitution N1066A relative to the amino acid sequence shown in SEQ ID NO: 116. In some embodiments, the anti-CD28 V L domain contains the amino acid sequence shown in SEQ ID NO: 202.

[0215] In some embodiments, an antibody construct comprising an anti-CD28 binding domain substituted with one or more V H and / or V L domains may exhibit reduced in vitro and / or in vivo non-specific activity. In one such embodiment, an antibody construct comprising an anti-CD28 binding domain containing such a mutation (relative to the huTN228 wild-type sequence) may induce less non-specific immune cell activity, such as reduced non-specific cytokine production by immune cells. In certain embodiments, an antibody construct comprising an anti-CD28 binding domain carrying an N1066A substitution in the V L domain relative to the huTN228 wild-type may induce reduced non-specific T cell activity, such as cytokine production. In another embodiment, an antibody construct comprising an anti-CD28 binding domain carrying a Y1031A substitution in the V L domain relative to the huTN228 wild-type may induce reduced non-specific T cell activity, such as cytokine production. In some embodiments, non-specific T cell activity, such as cytokine production such as TNFα, IL-2, etc., may be reduced by about 10-fold, 20-fold, 30-fold, 50-fold, 60-fold, 70-fold or about 100-fold.

[0216] In various embodiments, the antibody constructs herein comprise a binding domain capable of binding CD28 on T cells, wherein such a binding domain binds CD28 with an EC 50 value of about 20 nM to about 600 nM, and comprises: a V H domain that contains an HCDR1 having the sequence SX1GVH (SEQ ID NO: 302), an HCDR2 having the sequence VIWX2GGX3TNFNSALMS (SEQ ID NO: 306), and an HCDR3 having the sequence DRAX4GX5YX6X7AMDY (SEQ ID NO: 312); and a V LA sequence that includes an LCDR1 having the sequence RASESVEYYX8TSLMQ (SEQ ID NO:315), an LCDR2 having the sequence AASX9VX 10 S (SEQ ID NO:319), and an LCDR3 having the sequence QQSRKVPFT (SEQ ID NO:320), and wherein X1 = 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 X 10 = E, D.

[0217] In some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein includes a V H domain that includes or consists of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence shown in SEQ ID NO:106; and a V L domain that includes or consists of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99%, or 100% sequence identity to the sequence shown in SEQ ID NO:116. In some embodiments, the anti-CD28 binding domain of the antibody constructs herein includes a V H domain that includes or consists of the amino acid sequence shown in SEQ ID NO:106; and a V L domain that includes or consists of the amino acid sequence shown in SEQ ID NO:116.

[0218] In some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein includes a V H domain that includes or consists of the amino acid sequence shown in SEQ ID NO:204; and a V L domain that includes or consists of the amino acid sequence shown in SEQ ID NO:116.

[0219] In some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein includes a V H domain that includes or consists of the amino acid sequence shown in SEQ ID NO:207; and a V L domain that includes or consists of the amino acid sequence shown in SEQ ID NO:116.

[0220] In some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein comprises a V H domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 106; and a V L domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 200.

[0221] In some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein comprises a V H domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 106; and a V L domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 209.

[0222] In some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein comprises a V H domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 203; and a V L domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 116.

[0223] In some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein comprises a V H domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 201; and a V L domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 116.

[0224] In some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein comprises a V H domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 208; and a V L domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 116.

[0225] In some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein comprises a V H domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 113; and a V L domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 116.

[0226] In some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein comprises a V H domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 206; and a V L domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 116.

[0227] In some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein comprises a V H domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 210; and a V L domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 116.

[0228] In some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein comprises a V H domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 205; and a V L domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 116.

[0229] In some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein comprises a V H domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 106; and a V L domain that comprises or consists of the amino acid sequence shown in SEQ ID NO: 151.

[0230] In some embodiments, the anti-CD28 binding domain (e.g., scFv domain or Fab domain) of the antibody constructs herein comprises a V HA domain comprising or consisting of the amino acid sequence shown in SEQ ID NO: 106; and V L A domain comprising or consisting of the amino acid sequence shown in SEQ ID NO: 202.

[0231] In embodiments where the anti-CD28 binding domain of the antibody construct is a scFv domain, such an anti-CD28 scFv domain can comprise, from the N-terminus to the C-terminus or from the C-terminus to the N-terminus, via a linker scFv coupled to V L domain, the V H domain, the V H domain comprising the amino acid sequence shown in any one of SEQ ID NO: 106, 113, 201, 203, 204, 205, 206, 207, 208 or 210, the V L domain comprising the amino acid sequence shown in any one of SEQ ID NO: 116, 151, 200, 202 or 209, the linker scF v sequence is (G n S) m , where n, m can independently be 1, 2, 3, 4 or 5, and as shown in SEQ ID NO: 348. In some embodiments, the linker scFv comprises or consists of the amino acid sequence shown in SEQ ID NO: 104. And in some embodiments, the anti-CD28 scFv domain has a domain structure of V H -linker scFv -V L from the N-terminus to the C-terminus. In other embodiments, the anti-CD28 scFv domain has a domain structure of V L -linker scFv -V H from the N-terminus to the C-terminus.

[0232] In some embodiments, the anti-CD28 scFv domain of the antibody construct herein comprises or consists of an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 119. In some embodiments, the anti-CD28 scFv domain of the antibody construct herein comprises or consists of the amino acid sequence shown in SEQ ID NO: 119.

[0233] In embodiments in which the anti-CD28 binding domain of the antibody construct is a Fab domain, such anti-CD28 Fab domain can comprise a heavy chain paired with a light chain, the heavy chain comprising a V H1 sequence coupled to a C H sequence, the V H sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO:106, the CH1 sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO:107, the light chain comprising a V L sequence coupled to a C L sequence, the V L sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO:116, the CL sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO:117.

[0234] In some embodiments, the anti-CD28 Fab domain of the antibody construct herein can comprise a heavy chain paired with a light chain, the heavy chain comprising a V H1 sequence coupled to a C H sequence, the V H sequence comprising or consisting of the amino acid sequence shown in any one of SEQ ID NO:106, 113, 201, 203, 204, 205, 206, 207, 208 or 210, the CH1 sequence comprising or consisting of the amino acid sequence shown in SEQ ID NO:107, the light chain comprising a V L sequence coupled to a C L sequence, the V L sequence comprising or consisting of the amino acid sequence shown in any one of SEQ ID NO:116, 151, 200, 202 or 209, the CL sequence comprising or consisting of the amino acid sequence shown in SEQ ID NO:117.

[0235] In some embodiments, the anti-CD28 Fab domain of the antibody construct herein can comprise a heavy chain paired with a light chain, the heavy chain comprising a V H1 sequence coupled to a C H sequence, the V HThe sequence comprises or consists of the amino acid sequence shown in SEQ ID NO: 106, the CH1 sequence comprises or consists of the amino acid sequence shown in SEQ ID NO: 107, and the light chain comprises a V L sequence coupled to a C L sequence, and the V L sequence comprises or consists of the amino acid sequence shown in SEQ ID NO: 116, and the CL sequence comprises or consists of the amino acid sequence shown in SEQ ID NO: 117.

[0236] In certain embodiments, an antibody construct is described herein that comprises a binding domain capable of binding CD28, wherein the binding domain comprises: a V H sequence that comprises an HCDR1 having the sequence SX1GVH (SEQ ID NO: 302), an HCDR2 having the sequence VIWX2GGX3TNFNSALMS (SEQ ID NO: 306), and an HCDR3 having the sequence DRAX4GX5YX6X7AMDY (SEQ ID NO: 312); and a V L sequence that comprises an LCDR1 having the sequence RASESVEYYX8TSLMQ (SEQ ID NO: 315), an LCDR2 having the sequence AASX9VX 10 S (SEQ ID NO: 319), and an LCDR3 having the sequence QQSRKVPFT (SEQ ID NO: 320), and the binding domain comprises one or more of the following amino acid substitutions at the positions identified in the CDR sequences: X1: Y to A; X2: P to A; X3: G to S; X4: S to Y; X5: N to A; X6: L to N; X7: S to Y; X8: G to V; X9: N to A; or X 10 : E to D.

[0237] In some embodiments, the antibody construct comprises an anti-CD28 binding domain that comprises the substitution X1: Y to A. In some embodiments, the antibody construct comprises an anti-CD28 binding domain that comprises the substitution X2: P to A. In some embodiments, the antibody construct comprises an anti-CD28 binding domain that comprises the substitution X3: G to S. In some embodiments, the antibody construct comprises an anti-CD28 binding domain that comprises the substitution X4: S to Y. In some embodiments, the antibody construct comprises an anti-CD28 binding domain that comprises the substitution X5: N to A. In some embodiments, the antibody construct comprises an anti-CD28 binding domain that comprises the substitution X6: L to N. In some embodiments, the antibody construct comprises an anti-CD28 binding domain that comprises the substitution X7: S to Y. In some embodiments, the antibody construct comprises an anti-CD28 binding domain that comprises the substitution X8: G to V. In some embodiments, the antibody construct comprises an anti-CD28 binding domain that comprises the substitution X9: N to A. In some embodiments, the antibody construct comprises an anti-CD28 binding domain that comprises the substitution X 10 : E to D.

[0238] In some embodiments, compared to the binding affinity of an antibody construct comprising an anti-CD28 binding domain that does not contain one or more amino acid substitutions in one or more CDR sequences, the binding affinity of such an antibody construct for CD28 can be reduced by about 1.5-fold to about 25-fold, about 2.0-fold to about 20-fold, about 3.0-fold to about 20-fold, or about 5.0-fold to about 10-fold.

[0239] In some embodiments, such an antibody construct can comprise an anti-CD28 binding domain that comprises the CDRs of the V H domain shown in SEQ ID NO: 300, 303, and 307 and the CDRs of the V L sequence shown in SEQ ID NO: 313, 316, and 320.

[0240] In some embodiments, such an antibody construct can comprise an anti-CD28 binding domain that comprises: anti-CD28 V HA sequence comprising an HCDR1 having the sequence SYGVH (SEQ ID NO:300), an HCDR2 having the sequence VIWPGGGTNFNSALMS (SEQ ID NO:303), and an HCDR3 having the sequence DRAYGNYLYAMDY (SEQ ID NO:307); and an anti-CD28 V L A sequence comprising an LCDR1 having the sequence RASESVEYYVTSLMQ (SEQ ID NO:313), an LCDR2 having the sequence AASNVDS (SEQ ID NO:316), and an LCDR3 having the sequence QQSRKVPFT (SEQ ID NO:320).

[0241] C. A binding domain against a tumor-associated antigen (TAA)

[0242] As described herein, in various embodiments, the trivalent trispecific antibody constructs of the present disclosure comprise a binding domain capable of binding to a TAA. In various embodiments, as further described herein, the antibody constructs herein can be trivalent and trispecific and comprise a first binding domain capable of binding to a first antigen on a first cytotoxic effector cell, a second binding domain capable of binding to a second antigen on a second cytotoxic effector cell, and a third binding domain, wherein such third binding domain is capable of binding to a TAA. The TAA can be any antigenic substance expressed on the surface of tumor cells.

[0243] As described herein, in various embodiments, the anti-TAA binding domain of the antibody construct can be capable of binding to mesothelin (MSLN).

[0244] In other embodiments, the anti-TAA binding domain of the antibody construct can be capable of binding to Claudin18.2 (Cldn18.2).

[0245] Generally, the anti-TAA binding domain of the antibody constructs herein can be an scFv domain or a Fab domain.

[0246] In various embodiments, the anti-TAA binding domain of the trivalent trispecific antibody constructs herein is an scFv domain.

[0247] In various embodiments, the anti-TAA binding domain (e.g., scFv domain) of the antibody constructs herein is capable of binding to Cldn18.2. In some embodiments, such anti-Cldn18.2 binding domain comprises: HCDR1 having the sequence SNPMI (SEQ ID NO: 333), HCDR2 having the sequence IIDTDGSTYYADWAKG (SEQ ID NO: 334), and HCDR3 having the sequence RLHGSSNGYYDDL (SEQ ID NO: 335); and V L sequence, which comprises LCDR1 having the sequence QASQSIYSYLS (SEQ ID NO: 336), LCDR2 having the sequence KASTLAS (SEQ ID NO: 337), and LCDR3 having the sequence QQGYTVTNVDKNT (SEQ ID NO: 338).

[0248] In some embodiments, the anti-Cldn18.2 binding domain (e.g., scFv domain) of the antibody constructs herein comprises V H sequence, which comprises an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 127 or consists of said amino acid sequence; and V L sequence, which comprises an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 128 or consists of said amino acid sequence.

[0249] In various embodiments, the anti-Cldn18.2 binding domain of the antibody constructs herein is an scFv domain and comprises, from the N-terminus to the C-terminus or from the C-terminus to the N-terminus, V scFv coupled via a linker L to V H sequence, which comprises the amino acid sequence shown in SEQ ID NO: 127 or consists of said amino acid sequence, which comprises the amino acid sequence shown in SEQ ID NO: 128 or consists of said amino acid sequence, and the linker has the amino acid sequence shown in SEQ ID NO: 104.

[0250] In various embodiments, the anti-TAA binding domain (e.g., scFv domain) of the antibody constructs herein is capable of binding to MSLN. In some embodiments, such anti-MSLN binding domain comprises: V HA sequence comprising an HCDR1 having the sequence GYTMN (SEQ ID NO: 327), an HCDR2 having the sequence LITPYSGASSYAQKFQG (SEQ ID NO: 328), and an HCDR3 having the sequence GGYDGRGFDY (SEQ ID NO: 329); and V L A sequence comprising an LCDR1 having the sequence SASSSVSYMH (SEQ ID NO: 330), an LCDR2 having the sequence DTSKLAS (SEQ ID NO: 331), and an LCDR3 having the sequence QQWSGHPLT (SEQ ID NO: 332).

[0251] In some embodiments, the anti-MSLN binding domain of the antibody constructs herein is a scFv domain and comprises an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 2 or consists of said amino acid sequence. In some embodiments, the anti-MSLN scFv domain of the antibody constructs herein comprises the amino acid sequence shown in SEQ ID NO: 2 or consists of said amino acid sequence.

[0252] In certain embodiments, described herein is an antibody construct comprising an anti-Cldn18.2 binding domain, the anti-Cldn18.2 binding domain comprising the V H CDRs of the sequences and the V L CDRs of the sequences.

[0253] In some embodiments, such an antibody construct may comprise an anti-Cldn18.2 binding domain, the anti-Cldn18.2 binding domain comprising: V H A sequence comprising an HCDR1 having the sequence SNPMI (SEQ ID NO: 333), an HCDR2 having the sequence IIDTDGSTYYADWAKG (SEQ ID NO: 334), and an HCDR3 having the sequence RLHGSSNGYYDDL (SEQ ID NO: 335); and V L A sequence comprising an LCDR1 having the sequence QASQSIYSYLS (SEQ ID NO: 336), an LCDR2 having the sequence KASTLAS (SEQ ID NO: 337), and an LCDR3 having the sequence QQGYTVTNVDKNT (SEQ ID NO: 338).

[0254] In some embodiments, such an antibody construct can comprise an anti-Cldn18.2 binding domain, the anti-Cldn18.2 binding domain: comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 127 or consisting of said amino acid sequence; and V L a sequence comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQID NO: 128 or consisting of said amino acid sequence. In certain embodiments, such an anti-Cldn18.2 binding domain comprises V H a sequence comprising the amino acid sequence shown in SEQID NO: 127 or consisting of said amino acid sequence; and V L a sequence comprising the amino acid sequence shown in SEQ ID NO: 128 or consisting of said amino acid sequence.

[0255] D.Fc domain

[0256] As described herein, the trivalent trispecific antibody constructs of the present disclosure can comprise an Fc domain (or Fc region), the Fc domain 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 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 of the Fc polypeptides to form a heterodimeric Fc domain as compared to a corresponding homodimeric Fc domain.

[0257] As used herein, the term "Fc domain" includes a native (or wild-type) sequence Fc domain and a variant Fc domain that contains one or more amino acid modifications relative to the corresponding native or wild-type Fc domain. Unless otherwise specified herein, the numbering of amino acid residues in the Fc domain or constant region is according to the EU numbering system, also known as the EU index, as described, for example, 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 Fc polypeptide and a second Fc polypeptide) that form the dimeric (e.g., heterodimeric) Fc domain. In some embodiments, the Fc polypeptide can comprise the C-terminal constant region of an Ig heavy chain that is capable of stabilizing self-association. In various embodiments, and as further described herein, the Fc polypeptide (e.g., the first Fc polypeptide or the second Fc polypeptide) comprises at least one of a CH2 domain and / or a CH3 domain. In certain embodiments, the Fc polypeptide of the antibody construct described herein comprises a CH2 domain and a CH3 domain.

[0258] As disclosed herein, a trivalent trispecific antibody construct can comprise an Fc domain, wherein such Fc domain can be a heterodimeric Fc domain. Unless otherwise indicated, the Fc domain of the antibody construct, e.g., the heterodimeric Fc domain, 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 both a CH2 domain and a CH3 domain as described in various embodiments herein.

[0259] In certain embodiments, the antibody construct comprises an Fc domain based on a human IgG Fc domain. In some embodiments, the antibody construct comprises an Fc domain based on a human IgG1 Fc domain. In various embodiments, the antibody construct comprises a heterodimeric IgG Fc domain containing two different Fc polypeptides (e.g., a first Fc polypeptide and a second polypeptide), wherein the first Fc polypeptide and the second Fc polypeptide have different amino acid sequences, e.g., amino acid sequences having about 90%, 95%, 97%, or 99% sequence identity when compared and aligned with each other, e.g., as further described herein. In some embodiments, the difference in the amino acid sequences of the first Fc polypeptide and the second Fc polypeptide can be attributed to asymmetric amino acid substitutions that can be introduced into each Fc polypeptide chain to facilitate preferential pairing of the heavy chains to form a heterodimeric Fc domain as compared to a corresponding homodimeric Fc domain.

[0260] In various embodiments, the trivalent trispecific antibody constructs herein comprise an Fc domain that is a modified IgG Fc domain, and wherein at least one CH3 domain of at least one Fc polypeptide contains one or more amino acid modifications as compared to the corresponding wild-type CH3 domain. In some embodiments, the antibody constructs herein comprise an Fc domain that is a modified IgG Fc domain, wherein at least one CH2 domain of at least one Fc polypeptide contains one or more amino acid modifications as compared to the corresponding wild-type CH2 domain. In some embodiments, the antibody construct comprises an Fc domain that is a modified IgG Fc domain, wherein both the CH3 domain and the CH2 domain of at least one Fc polypeptide contain one or more amino acid modifications as compared to the corresponding wild-type CH3 domain and CH2 domain. In various embodiments, the two Fc polypeptides of the heterodimeric Fc domain can contain one or more amino acid modifications in their CH3 domains. In some embodiments, the two Fc polypeptides of the heterodimeric Fc domain can contain one or more amino acid modifications in their CH2 domains. In other embodiments, the two Fc polypeptides of the heterodimeric Fc domain can contain one or more amino acid modifications in their CH2 and CH3 domains.

[0261] Modified Fc domain

[0262] In some embodiments, the present disclosure relates to a trivalent trispecific antibody construct that can comprise a heterodimeric Ig Fc domain, the heterodimeric Ig Fc domain comprising a modified 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 Fc polypeptide and the second Fc polypeptide each comprise one or more amino acid modifications in their CH3 domain sequences as compared to the respective wild-type sequences. As used herein, the term "asymmetric amino acid modification" generally refers to a modification in which the amino acid at a particular position on the first Fc polypeptide is different from the amino acid at the corresponding position on the second Fc polypeptide. These asymmetric amino acid modifications can include modification of only one of the two amino acids at the corresponding positions on each Fc polypeptide, or they can include modification of both amino acids at the corresponding positions on each of the first Fc polypeptide and the second Fc polypeptide. In various embodiments, the "asymmetric amino acid modification" is an asymmetric amino acid substitution.

[0263] In some embodiments, the antibody constructs herein comprise a heterodimeric Fc domain, the heterodimeric Fc domain comprising a modified CH3 domain (i.e., a heterodimeric CH3 domain composed of two CH3 domains of a first Fc polypeptide and a second Fc polypeptide), wherein the modified CH3 domain comprises one or more asymmetric amino acid modifications that promote the formation of the heterodimeric Fc domain (e.g., pairing of the first Fc polypeptide with the second Fc polypeptide) rather than the corresponding homodimeric Fc domain (e.g., pairing of the first Fc polypeptide with another first Fc polypeptide). Amino acid modifications that can be made to the CH3 domain of the Fc domain to promote the formation of the heterodimeric Fc domain are known in the art and include, for example, those described in International Publication No. WO96 / 027011 ("knobs-into-holes"); Gunasekaran et al., 2010, J Biol Chem, 285, 19637-46 ("electrostatic steering"); Davis et al., 2010, Prot Eng Des Sel, 23(4):195-202 (strand-exchange engineered domain (SEED) technology) and Labrijn et al., 2013, Proc Natl Acad Sci USA, 110(13):5145-50 (Fab-arm exchange). Other examples include methods that combine positive and negative design strategies to generate a stable asymmetrically modified Fc region, as described in International Publication Nos. WO 2012 / 058768 and WO2013 / 063702.

[0264] In certain embodiments, the antibody constructs described herein comprise a heterodimeric Fc domain that comprises a modified heterodimeric CH3 domain, wherein 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.

[0265] In some embodiments, the antibody constructs described herein comprise a heterodimeric human IgG1 Fc domain having a modified CH3 domain. Table 2 herein provides the amino acid sequence of the human IgG1 Fc domain sequence (e.g., the sequence from which the first Fc polypeptide and / or the second Fc polypeptide may be derived), which corresponds to amino acids 231 to 447 of the full-length human IgG1 heavy chain (e.g., the heavy chain comprising the V H , C H1 , hinge, CH2, and CH3 domains) and is identified by SEQ ID NO:1. The CH2 domain is generally defined as comprising amino acids 231-340 of the full-length human IgG1 heavy chain, and the CH3 domain is generally defined as comprising amino acids 341-447 of the full-length human IgG1 heavy chain.

[0266] As described herein, an antibody construct can comprise a heterodimeric Fc domain having a modified CH3 domain, the modified CH3 domain comprising one or more asymmetric amino acid modifications that promote the formation of the heterodimeric Fc domain rather than the homodimeric Fc domain, wherein the modified CH3 domain comprises a first Fc polypeptide having amino acid modifications at positions F405 and Y407 relative to SEQ ID NO:1 and a second Fc polypeptide having amino acid modifications at positions T366 and T394 relative to SEQ ID NO:1. In various embodiments, the one or more amino acid modifications comprise one or more amino acid substitutions. Thus, 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 the first Fc polypeptide further comprises an amino acid modification at position L351 relative to SEQ ID NO:1. In some embodiments, the amino acid modification at position L351 of the first Fc polypeptide of the modified CH3 domain is L351Y. In some embodiments, the modified CH3 domain of the second Fc polypeptide further comprises an amino acid modification at position K392 relative to SEQ ID NO:1. In some embodiments, the amino acid modification at position K392 of the second Fc polypeptide of the modified CH3 domain is K392F, K392L, or K392M. In some embodiments, one or both of the first Fc polypeptide and the second Fc polypeptide of the modified CH3 domain further comprise the amino acid modification T350V.

[0267] In certain embodiments, the antibody constructs herein comprise a heterodimeric Fc domain having a modified CH3 domain, the modified CH3 domain comprising one or more asymmetric amino acid modifications that promote the formation of the heterodimeric Fc domain rather than the homodimeric Fc domain, wherein the modified CH3 domain comprises a first Fc polypeptide comprising the amino acid modifications F405A, F405I, F405M, F405S, F405T, or F405V and the amino acid modification Y407I or Y407V (relative to SEQ ID NO: 1) and a second Fc polypeptide comprising the amino acid modifications T366I, T366L, or T366M and the amino acid modification T394W (relative to SEQ ID NO: 1). In some embodiments, the first Fc polypeptide of the modified CH3 domain further comprises the amino acid modification L351Y. In some embodiments, the second Fc polypeptide of the modified CH3 domain further comprises the amino acid modification K392F, K392L, or K392M. In some embodiments, one or both of the first and second Fc polypeptides having the modified CH3 domain further comprise the amino acid modification T350V.

[0268] Table 2: Exemplary human IgG1 Fc domain sequences and their variants

[0269]

[0270] * "A" corresponds to the first Fc polypeptide chain, and "B" corresponds to the second Fc polypeptide chain

[0271] In certain embodiments, the antibody constructs herein comprise a heterodimeric Fc domain that comprises a modified CH3 domain having a first Fc polypeptide that comprises amino acid modifications at positions F405 and Y407 and optionally also at position L351 and a second Fc polypeptide that comprises amino acid modifications at positions T366 and T394 and optionally also 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: 1.

[0272] In some embodiments, the antibody construct comprises a heterodimeric Fc domain that includes a modified CH3 domain, and the modified CH3 domain includes a modification of any one of Variant 1, Variant 2, Variant 3, Variant 4, or Variant 5 as shown in Table 2.

[0273] In various embodiments, the antibody constructs of the present disclosure may comprise a heterodimeric Fc domain that includes a first Fc polypeptide (A) and a second Fc polypeptide (B), wherein the first Fc polypeptide (A) and the second Fc polypeptide (B) include an amino acid substitution according to Variant #1 as shown in Table 2 in their CH3 domains. In other embodiments, the antibody constructs of the present disclosure may comprise a heterodimeric Fc domain that includes a first Fc polypeptide (A) and a second Fc polypeptide (B), wherein the first Fc polypeptide (A) and the second Fc polypeptide (B) include an amino acid substitution according to Variant #2 as shown in Table 2 in their CH3 domains. In some embodiments, the antibody constructs of the present disclosure may comprise a heterodimeric Fc domain that includes a first Fc polypeptide (A) and a second Fc polypeptide (B), wherein the first Fc polypeptide (A) and the second Fc polypeptide (B) include an amino acid substitution according to Variant #3 as shown in Table 2 in their CH3 domains. In some embodiments, the antibody constructs of the present disclosure may comprise a heterodimeric Fc domain that includes a first Fc polypeptide (A) and a second Fc polypeptide (B), wherein the first Fc polypeptide (A) and the second Fc polypeptide (B) include an amino acid substitution according to Variant #4 as shown in Table 2 in their CH3 domains. In other embodiments, the antibody constructs of the present disclosure may comprise a heterodimeric Fc domain that includes a first Fc polypeptide (A) and a second Fc polypeptide (B), wherein the first Fc polypeptide (A) and the second Fc polypeptide (B) include an amino acid substitution according to Variant #5 as shown in Table 2 in their CH3 domains.

[0274] In certain embodiments, the CH3 domain of the first Fc polypeptide of the antibody constructs 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 shown in SEQ ID NO: 110. In certain embodiments, the CH3 domain of the second Fc polypeptide of the antibody constructs 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 shown in SEQ ID NO: 114. In some embodiments, the CH3 domain of the first Fc polypeptide of the antibody constructs herein has the amino acid sequence shown in SEQ ID NO: 110, and the second Fc polypeptide of the antibody constructs herein has the amino acid sequence shown in SEQ ID NO: 114.

[0275] In certain embodiments, the antibody constructs herein comprise 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 corresponding first and second Fc polypeptides). In some embodiments, the antibody construct comprises a heterodimeric Fc domain based on an IgG Fc domain having a modified CH2 domain, wherein the modification of the CH2 domain results in an altered (e.g., reduced or attenuated) binding to one or more Fc receptors (FcRs) such as receptors of the FcγRI, FcγRII, and FcγRIII subclasses.

[0276] Several amino acid modifications of the CH2 domain of the first Fc polypeptide and / or the second Fc polypeptide of an Fc domain that selectively alter the affinity of such Fc domain for different Fcγ receptors are known in the art. Amino acid modifications that cause increased binding and those that cause decreased binding can both be used for certain indications. For example, increasing the binding affinity of Fc for FcγRIIIa (the activated receptor) results in increased antibody-dependent cell-mediated cytotoxicity (ADCC), which in turn leads to increased target cell lysis. In some cases, a decrease in binding to FcγRIIb (the inhibitory receptor) may also be beneficial in some cases. In certain indications, a reduction or elimination of ADCC and complement-mediated cytotoxicity (CDC) may be desired. In such embodiments, a modified CH2 domain containing an amino acid modification that results in increased binding to FcγRIIb or an amino acid modification that can reduce or eliminate the binding of the Fc region to all Fcγ receptors ("knockout" variant) may be useful.

[0277] Non-limiting examples of amino acid modifications that alter the binding of Fcγ receptors to the Fc domain of the CH2 domain include, but are not limited to, the following: S298A / E333A / K334A and S298A / E333A / K334A / K326A (increase affinity for FcγRIIIa) (Lu, et al., 2011, J Immunol Methods, 365(1-2):132-41); F243L / R292P / Y300L / V305I / P396L (increase affinity for FcγRIIIa) (Stavenhagen, et al., 2007, Cancer Res, 67(18):8882-90); F243L / R292P / Y300L / L235V / P396L (increase affinity for FcγRIIIa) (Nordstrom JL, et al., 2011, Breast Cancer Res, 13(6):R123); F243L (increase affinity for FcγRIIIa) (Stewart, et al., 2011, Protein Eng Des Sel., 24(9):671-8); S298A / E333A / K334A (increase affinity for FcγRIIIa) (Shields, et al., 2001, J Biol Chem, 276(9):6591-604); S239D / I332E / A330L and S239D / I332E (increase affinity for FcγRIIIa) (Lazar, et al., 2006, Proc Natl Acad Sci USA, 103(11):4005-10), and S239D / S267E and S267E / L328F (increase affinity for FcγRIIb) (Chu, et al., 2008, Mol Immunol, 45(15):3926-33). Additional modifications that affect the binding of the Fc domain to Fcγ receptors are described in Therapeutic Antibody Engineering (Strohl & Strohl, Woodhead Publishing series in Biomedicine No 11, ISBN 1907568 37 9, October 2012, page 283).

[0278] In various embodiments, the antibody constructs of the present disclosure include a heterodimeric Fc domain based on an IgG Fc domain having a modified CH2 domain, wherein one or both of the CH2 sequences of the modified dimeric CH2 domain (i.e., the first / second Fc polypeptide) contain one or more amino acid modifications that can result in reduced or eliminated binding of the Fc domain to one or more, or all, Fcγ receptors (i.e., "knockout" or "KO" variants).

[0279] Various publications have described strategies that have been used to engineer antibodies to produce "knockout" Fc variants (see, e.g., Strohl, 2009, Curr Opin Biotech 20:685-691, and Strohl & Strohl, "Antibody Fc engineering for optimal antibody performance" In Therapeutic Antibody Engineering, Cambridge: Woodhead Publishing, 2012, pp. 225-249). These strategies include reducing effector function by modification of glycosylation, use of IgG2 / IgG4 scaffolds, or introduction of mutations in the hinge or CH2 domain of the Fc (see also, U.S. Patent Publication No. 2011 / 0212087, International Publication No. WO2006 / 105338, U.S. Patent Publication No. 2012 / 0225058, U.S. Patent Publication No. 2012 / 0251531, and Strop et al., 2012, J. Mol. Biol., 420:204-219).

[0280] In some embodiments, the Fc domain of the antibody construct can include one or more known amino acid modifications to reduce FcγR and / or complement binding of the Fc domain. In some embodiments, such modifications can include those identified in Table 3.

[0281] Table 3: Modifications that reduce Fcγ receptor or complement binding

[0282]

[0283]

[0284] Additional examples herein include an Fc domain engineered to include the amino acid modifications L235A / L236A / D265S, such as based on the sequence shown in SEQ ID NO:1. Additionally, asymmetric amino acid modifications in the CH2 domain that reduce Fc binding to all Fcγ receptors are described in International Publication No. WO 2014 / 190441.

[0285] In certain embodiments, the CH2 domains of the first and second Fc polypeptides herein comprise or consist 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 shown in SEQ ID NO: 109. In some embodiments, the CH2 domains of the first and / or second Fc polypeptides herein comprise or consist of the sequence shown in SEQ ID NO: 109.

[0286] In certain embodiments, the antibody constructs herein comprise a heterodimeric Fc domain in which the native glycosylation has been modified. As is known in the art, the glycosylation of 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) produces an aglycosylated Fc lacking all effector functions (Bolt et al., 1993, Eur. J. Immunol., 23:403 - 411; Tao & Morrison, 1989, J. Immunol., 143:2595 - 2601). Conversely, removal of fucose from the N297 - linked oligosaccharide of the heavy chain has been shown to enhance ADCC based on improved binding to FcγRIIIa (see, e.g., Shields et al., 2002, J Biol Chem., 277:26733 - 26740, and Niwa et al., 2005, J. Immunol. Methods, 306:151 - 160). Such afucosylated antibody constructs can be produced, for example, in knockout Chinese hamster ovary (CHO) cells lacking fucosyltransferase (FUT8) (Yamane - Ohnuki et al., 2004, Biotechnol. Bioeng., 87:614 - 622), in the variant CHO cell line Lec13 having a reduced ability to attach fucose to N297 - linked carbohydrates (International Publication No. WO 03 / 035835), or in other cells that produce non - fucosylated antibodies (see, e.g., Li et al., 2006, Nat Biotechnol, 24:210 - 215; Shields et al., 2002, ibid, and Shinkawa et al., 2003, J. Biol. Chem., 278:3466 - 3473). In addition, International Publication No. WO 2009 / 135181 describes adding fucose analogs to the culture medium during antibody production to inhibit the incorporation of fucose into the carbohydrates on the antibody.

[0287] E. Linker

[0288] In various embodiments of the present disclosure, the trivalent trispecific antibody constructs described herein may comprise one or more linkers. In some embodiments, such one or more linkers are peptides (also referred to herein as "peptides") that comprise or consist of an amino acid sequence of about 1, 2, 3, 5, 10, 15, 20, 25, 30, 40, or about 50 consecutive amino acid residues. One or more peptide linkers of the antibody construct may comprise or consist of an amino acid sequence of 1 to about 50, 2 to about 40, 3 to about 30, or 5 to about 25 consecutive amino acid residues.

[0289] Such peptide linkers may couple or link two or more polypeptide sequences and / or domains of the antibody construct to each other. In various embodiments, a linker herein may couple a first polypeptide chain (e.g., a heavy chain constant domain (C H1 )) to an Fc polypeptide. Thus, a linker may be used to couple one domain of the antibody construct to another domain (from the N-terminus to the C-terminus), e.g., e.g., a linker Fab-Fc couples a Fab domain to an Fc domain, e.g., a linker scFv-Fab couples an scFv domain to a Fab domain, e.g., a linker scFv couples a V H domain to a V L domain, and so on. In embodiments in which two scFv domains of the antibody construct contain linkers scFv having the same amino acid sequence, such an antibody construct may be described as comprising linker scFv , rather than specifying that it contains linker scFv1 and linker scFv2 . However, in embodiments in which the linkers scFv of the two scFv domains have different amino acid sequences, such an antibody construct may be described as comprising linker scFv1 and linker scFv2 .

[0290] In embodiments in which a linker couples, e.g., a heavy chain variable domain (V H ) to, e.g., a light chain variable domain (V L ), the linker may have a length sufficient to allow the two domains to elicit their biological functions. In addition to providing a spacer function, linkers herein (e.g., peptide linkers) may provide flexibility or rigidity suitable for the correct orientation of one or more domains of the antibody construct within the antibody construct itself and between the antibody construct and its target.

[0291] In addition, the linkers (e.g., peptide linkers) herein can support (i) the expression of full-length fusion proteins, such as the full-length polypeptide chains H1, L1, H2, etc. of antibody constructs, and (ii) provide increased stability of the purified protein in vitro and in vivo, e.g., after administration to a subject in need (such as a human). One or more linkers used in the antibody constructs herein are generally non-immunogenic or poorly immunogenic in mammalian subjects to which the constructs can be administered. In certain embodiments, one or more linkers used in the antibody constructs herein can comprise a portion or all of a human Ig hinge region, the stalk region of a C-type lectin, a type II membrane protein family, or a combination thereof. In certain embodiments, one or more linkers used in the antibody constructs herein can comprise a portion or all of a human Ig hinge region (such as an IgG1 hinge region), such as linker Fab-Fc or linker scFv-Fc .

[0292] In certain embodiments, each linker used in the antibody constructs herein can comprise or consist of an amino acid sequence having a length of from 2 to about 50 amino acids. In some embodiments, each linker used in the antibody constructs herein can comprise or consist of an amino acid sequence having a length of 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, one or more linkers of the antibody construct can each comprise an amino acid sequence that comprises 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 contiguous amino acids or consists of said contiguous amino acids.

[0293] In certain embodiments, the linkers of the antibody constructs herein (e.g., linker scFv , linker scFv-Fab , linker Fab-Fc , etc.) comprise the amino acid sequence (EAAAK) n, where n is an integer from 1 to 5 (SEQ ID NO: 339), or consists of said amino acid sequence. In some embodiments, the linker comprises the sequence EAAAK (SEQ ID NO: 340) or consists of said sequence. In some embodiments, the linker comprises the sequence EAAAKEAAAK (SEQ ID NO: 341) or consists of said sequence. In some embodiments, the linker comprises a polyproline linker, such as having the amino acid sequence of PPP (SEQ ID NO: 342) or PPPPP (SEQ ID NO: 343). In certain embodiments, the linker is a glycine (G)-proline (P) polypeptide linker, such as comprising one or more of GPPPG (SEQ ID NO: 344), GGPPPGG (SEQ ID NO: 345), GPPPPG (SEQ ID NO: 346) or GGPPPPGG (SEQ ID NO: 347), or consisting of them. In some embodiments, the linker herein is (G n S) m linker, where n and m are independently integers from 1 to 5 (SEQ ID NO: 348). In certain embodiments, the linker comprises (G3S) n (G4S)1 (SEQ ID NO: 349), (G3S)1(G4S) n (SEQ ID NO: 350), (G3S) n (G4S) n (SEQ ID NO: 351) or (G4S) n (SEQ ID NO: 355) of amino acid sequence or consists of said amino acid sequence, where each n is an integer from 1 to 5. In certain embodiments, the linker herein is suitable for connecting two different domains of an antibody construct and comprises a sequence containing a glycine-serine linker, such as but not limited to, (G m S) n -GG (SEQ ID NO: 352), where m and n are independently integers from 0 to 20, (SG n ) m (SEQ ID NO: 353) or (SEG n ) m (SEQ ID NO: 354), where m and n are independently integers from 0 to 20, but not both 0 at the same time.

[0294] In some embodiments, the antibody constructs described herein comprise any one or more of the linkers described herein. In some embodiments, the antibody construct comprises, for example, 1, 2, 3, 4 or 5 linkers, which may include one or more linkers Fab-Fc, one or more linkers scFv-Fab and / or one or more linkers scFv .

[0295] In certain embodiments, the linker of the antibody construct herein scFv comprises the amino acid sequence (G n S) m linker, where n and m are independently integers from 1 to 5 (SEQ ID NO: 348), or consists of said linker. In such embodiments, n and m can both be 4, so one or more linkers of the antibody construct scFv can comprise the sequence (G4S)4 (SEQ ID NO: 104) or consist of said sequence. In some embodiments, one or more linkers of the antibody construct scFv (e.g., linkers having the same amino acid sequence scFv1 and linker scFv2 ) can comprise an amino acid sequence having about 80%, 90% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 104 or consist of said amino acid sequence.

[0296] In some embodiments, the antibody construct of the present disclosure can comprise a linker that couples the C-terminus of the scFv domain to the N-terminus of the Fab domain scFv-Fab . Such a linker scFv-Fa b can comprise the amino acid sequence shown in SEQ ID NO: 105 or a sequence having about 80%, 90% or 100% sequence identity thereto or consist of said sequence.

[0297] In some embodiments, the antibody construct of the present disclosure can comprise a linker that couples the C-terminus of the Fc polypeptide to the N-terminus of the scFv domain Fc-scFv . In some embodiments, such a linker Fc-scFv can also comprise the amino acid sequence shown in SEQ ID NO: 105 or a sequence having about 80%, 90% or 100% sequence identity thereto or consist of said sequence.

[0298] In certain embodiments of the present disclosure, one or more linkers included in an antibody construct can be amino acid sequences obtained from, derived from, or designed from an antibody hinge region sequence. In some embodiments, such a 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 peptide retains a cysteine corresponding to the hinge cysteine disposed towards the amino (or N) terminus of the hinge. In further embodiments, the linker is derived from an IgG1 hinge and can be modified to remove any cysteine residues, or the linker is an IgG1 hinge having one or two cysteines corresponding to the hinge cysteines.

[0299] In certain embodiments, the linker of the antibody constructs described herein can comprise a “modified wild-type Ig hinge region” or a “modified Ig hinge region”. Such modified hinge regions can refer to (a) a wild-type Ig hinge region having up to 30% amino acid variation (e.g., up to 25%, 20%, 15%, 10%, or 5% amino acid substitutions, insertions, or deletions), (b) a portion of a wild-type Ig hinge region having up to 30% amino acid variation (e.g., up to 25%, 20%, 15%, 10%, or 5% amino acid substitutions or deletions) and having a length of at least 10 amino acids (e.g., at least 12, 13, 14, or 15 amino acids), (c) a portion of a wild-type Ig hinge region that includes 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) any combination of (a)-(c). In certain embodiments, one or more cysteine residues in a wild-type Ig hinge region (such as an IgG1 hinge that includes an upper region and a core region) can be replaced by one or more other amino acid residues (e.g., one or more serine residues). The modified Ig hinge region can alternatively or additionally have a proline residue in the wild-type Ig hinge region (such as an IgG1 hinge that includes an upper and core region) replaced by another amino acid residue (e.g., a serine residue).

[0300] Thus, in some embodiments, the antibody constructs of the present disclosure comprise a linker comprising or consisting of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99% or 100% sequence identity to the sequence shown in SEQ ID NO: 108. In some embodiments, the antibody constructs comprise a linker comprising or consisting of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99% or 100% sequence identity to the sequence shown in SEQ ID NO: 111. In some embodiments, the antibody constructs comprise a linker comprising or consisting of an amino acid sequence having at least about 80%, 90%, 95%, 97%, 99% or 100% sequence identity to the sequence shown in SEQ ID NO: 112.

[0301] In various embodiments, the antibody constructs of the present disclosure comprise a linker that couples a Fab domain to a first Fc polypeptide, the linker Fab-Fc , wherein the linker Fab-Fc comprises or consists of the amino acid sequence shown in SEQ ID NO: 108, and another linker that couples a first or second scFv domain to a first or second Fc polypeptide, the linker scFv-Fc , wherein the linker scFv-Fc comprises or consists of the amino acid sequence shown in SEQ ID NO: 111 or SEQ ID NO: 112.

[0302] F. Certain Embodiments of Antibody Constructs

[0303] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding CD3 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising from the N-terminus to the C-terminus a V H1 sequence coupled to a C H sequence, and the light chain comprising from the N-terminus to the C-terminus a V L sequence coupled to a C L sequence; (ii) a first scFv domain capable of binding CD28 on a second immune cell, wherein the first scFv domain comprises from the N-terminus to the C-terminus a first scFv V scFv sequence coupled via a first linker to a first scFv V L sequence; (iii) a second scFv domain capable of binding MSLN on a tumor cell, wherein the second scFv domain comprises from the N-terminus to the C-terminus a second scFv V H sequence coupled via a second linker to a second scFv V scFv and the second linkerL The second scFv V of the sequence coupling H sequence; and (iv) a heterodimeric 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 via the C-terminus of the FabC H1 sequence and via a linker Fab-Fc and b) the C-terminus of the first scFv domain is coupled to the N-terminus of the Fab V scFv-Fab sequence via a linker H and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker scFv-Fc .

[0304] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding CD3 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising, from the N-terminus to the C-terminus, a V H1 sequence coupled to a C H sequence, the light chain comprising, from the N-terminus to the C-terminus, a V L sequence coupled to a C L sequence; (ii) a first scFv domain capable of binding CD28 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFvV scFv sequence coupled to a first scFv V H sequence via a first linker L ; (iii) a second scFv domain capable of binding MSLN on a tumor cell, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, a second scFv V scFv sequence coupled to a second scFv V L sequence via a second linker H ; and (iv) a heterodimeric 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 via the C-terminus of the FabC H1 sequence and via a linker Fab-Fc ; b) the N-terminus of the first scFv domain is coupled to the C-terminus of the Fab C Fab-scFv sequence via a linker L ; and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker scFv-Fc .

[0305] In one embodiment, described herein is a trivalent trispecific antibody construct that comprises: (i) a Fab domain capable of binding CD3 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C H1 sequence, and the light chain comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C H sequence; (ii) a first scFv domain capable of binding CD28 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V sequence coupled via a first linker L to a first scFv V L sequence; (iii) a second scFv domain capable of binding MSLN on a tumor cell, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, a second scFv V sequence coupled via a second linker scFv to a second scFv V H sequence; and (iv) a heterodimeric 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 via the C-terminus of a FabC L sequence and via a linker scFv to the N-terminus of the first Fc polypeptide, b) the N-terminus of the first scFv domain is coupled to the C-terminus of the first Fc polypeptide via a linker L to the C-terminus of the first Fc polypeptide, and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker H to the N-terminus of the second Fc polypeptide. H1 sequence and via a linker Fab-Fc to the N-terminus of the first Fc polypeptide, b) the N-terminus of the first scFv domain is coupled to the C-terminus of the first Fc polypeptide via a linker Fc-scFv to the C-terminus of the first Fc polypeptide, and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker scFv-Fc to the N-terminus of the second Fc polypeptide.

[0306] In one embodiment, described herein is a trivalent trispecific antibody construct that comprises: (i) a Fab domain capable of binding CD3 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C H1 sequence, and the light chain comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C H sequence; (ii) a first scFv domain capable of binding CD28 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V sequence coupled via a first linker L sequence, and the light chain comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C L sequence; (ii) a first scFv domain capable of binding CD28 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V sequence coupled via a first linker scFv to a first scFv V H sequence; (iii) a second scFv domain capable of binding MSLN on a tumor cell, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, a second scFv V sequence coupled via a second linker La sequence; (iii) a second scFv domain capable of binding to MSLN on tumor cells, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, via a second linker scFv and the second scFv V L sequence coupled to the second scFv V H sequence; and (iv) a heterodimeric 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 via the C-terminus of the FabC H1 sequence and via a linker Fab-Fc ; b) the N-terminus of the first scFv domain is coupled to the C-terminus of the first Fc polypeptide via a linker Fc-scFv ; and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the Fab V scFv-Fab sequence via a linker H .

[0307] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding to CD3 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising, from the N-terminus to the C-terminus, a V H1 sequence coupled to a C H sequence, and the light chain comprising, from the N-terminus to the C-terminus, a V L sequence coupled to a C L sequence; (ii) a first scFv domain capable of binding to CD28 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V scFv sequence coupled to a first scFv V H sequence via a first linker L ; (iii) a second scFv domain capable of binding to MSLN on tumor cells, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, a second scFv V scFv sequence coupled to a second scFv V L sequence via a second linker H ; and (iv) a heterodimeric 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 second Fc polypeptide via the C-terminus of the FabC H1 sequence and via a linker Fab-Fc ; b) the C-terminus of the first scFv domain is coupled to the N-terminus of the first Fc polypeptide via a linker scFv-Fc ; and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the Fab V scFv-Fab sequence via a linker H .

[0308] In one embodiment, described herein is a trivalent trispecific antibody construct that comprises: (i) a Fab domain capable of binding CD28 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C sequence, and the light chain comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C sequence; (ii) a first scFv domain capable of binding CD3 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V sequence coupled to a first scFv V sequence via a first linker; (iii) a second scFv domain capable of binding MSLN on a tumor cell, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, a second scFv V sequence coupled to a second scFv V sequence via a second linker; and (iv) a heterodimeric 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 via the C-terminus of the Fab C sequence and via a linker; b) the C-terminus of the first scFv domain is coupled to the N-terminus of the Fab V sequence via a linker; and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker. H1 sequence coupled to a V H sequence, and the light chain comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C L sequence coupled to a V L sequence; (ii) a first scFv domain capable of binding CD3 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V sequence coupled to a first scFv V sequence via a first linker scFv coupled to a first scFv V L sequence; (iii) a second scFv domain capable of binding MSLN on a tumor cell, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, a second scFv V sequence coupled to a second scFv V sequence via a second linker H coupled to a second scFv V scFv sequence; and (iv) a heterodimeric 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 via the C-terminus of the Fab C L sequence and via a linker; b) the C-terminus of the first scFv domain is coupled to the N-terminus of the Fab V H sequence via a linker; and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker. H1 sequence and via a linker; b) the C-terminus of the first scFv domain is coupled to the N-terminus of the Fab V Fab-Fc sequence via a linker; and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker. scFv-Fab coupled to the N-terminus of the Fab V H sequence via a linker; and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker. scFv-Fc coupled to the N-terminus of the second Fc polypeptide via a linker.

[0309] In one embodiment, described herein is a trivalent trispecific antibody construct that comprises: (i) a Fab domain capable of binding CD28 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C H1 sequence coupled to a V H sequence, and the light chain comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C L sequence coupled to a V L sequence; (ii) a first scFv domain capable of binding CD3 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V sequence coupled to a first scFv V sequence via a first linker scFv coupled to a first scFv V H sequence;L a sequence; (iii) a second scFv domain capable of binding to MSLN on tumor cells, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, via a second linker scFv coupled to the second scFv V L sequence; and (iv) a heterodimeric 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 via the C-terminus of the Fab C H sequence and via a linker H1 sequence, b) the N-terminus of the first scFv domain is coupled to the C-terminus of the Fab C Fab-Fc sequence via a linker Fab-scFv and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker L sequence. scFv-Fc

[0310] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding to CD28 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising, from the N-terminus to the C-terminus, a V H1 sequence coupled to a C H sequence, and the light chain comprising, from the N-terminus to the C-terminus, a V L sequence coupled to a C L sequence; (ii) a first scFv domain capable of binding to CD3 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V scFv sequence coupled to a first scFv V H sequence via a first linker L ; (iii) a second scFv domain capable of binding to MSLN on tumor cells, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, a second scFv V scFv sequence coupled to a second scFv V L sequence via a second linker H ; and (iv) a heterodimeric 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 via the C-terminus of the Fab C H1 sequence and via a linker Fab-Fc sequence, b) the N-terminus of the first scFv domain is coupled to the C-terminus of the first Fc polypeptide via a linker Fc-scFv sequence, and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker scFv-Fc sequence.​

[0311] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding CD28 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C sequence, and the light chain comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C sequence; (ii) a first scFv domain capable of binding CD3 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V sequence coupled to a first scFv V sequence via a first linker; (iii) a second scFv domain capable of binding MSLN on a tumor cell, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, a second scFv V sequence coupled to a second scFv V sequence via a second linker; and (iv) a heterodimeric 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 via the C-terminus of the Fab C sequence and via a linker, b) the N-terminus of the first scFv domain is coupled to the C-terminus of the first Fc polypeptide via a linker, and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the Fab V sequence via a linker. H1 sequence coupled to a V H sequence, and the light chain comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C L sequence coupled to a V L sequence; (ii) a first scFv domain capable of binding CD3 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V sequence coupled to a first scFv V sequence via a first linker scFv coupled to a first scFv V H sequence; (iii) a second scFv domain capable of binding MSLN on a tumor cell, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, a second scFv V sequence coupled to a second scFv V sequence via a second linker L sequence; and (iv) a heterodimeric 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 via the C-terminus of the Fab C scFv sequence and via a linker, b) the N-terminus of the first scFv domain is coupled to the C-terminus of the first Fc polypeptide via a linker, and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the Fab V L sequence via a linker; and (iv) a heterodimeric 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 via the C-terminus of the Fab C H sequence and via a linker, b) the N-terminus of the first scFv domain is coupled to the C-terminus of the first Fc polypeptide via a linker, and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the Fab V H1 sequence and via a linker, b) the N-terminus of the first scFv domain is coupled to the C-terminus of the first Fc polypeptide via a linker, and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the Fab V Fab-Fc sequence; and (iv) a heterodimeric 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 via the C-terminus of the Fab C Fc-scFv sequence and via a linker, b) the N-terminus of the first scFv domain is coupled to the C-terminus of the first Fc polypeptide via a linker, and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the Fab V scFv-Fab sequence and via a linker, b) the N-terminus of the first scFv domain is coupled to the C-terminus of the first Fc polypeptide via a linker, and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the Fab V H sequence.

[0312] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding CD28 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C H1 sequence, and the light chain comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C H sequence; (ii) a first scFv domain capable of binding CD3 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V sequence coupled to a first scFv V sequence via a first linker L sequence coupled to a V L sequence; (ii) a first scFv domain capable of binding CD3 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V sequence coupled to a first scFv V sequence via a first linker scFv coupled to a first scFv V L sequence; (ii) a first scFv domain capable of binding CD3 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V sequence coupled to a first scFv V sequence via a first linkerH a sequence; (iii) a second scFv domain capable of binding to MSLN on tumor cells, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, via a second linker scFv coupled to the second scFv V L sequence; and (iv) a heterodimeric 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 via the C-terminus of the Fab C H sequence and via a linker H1 sequence; b) the C-terminus of the first scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker Fab-Fc sequence; and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the Fab V scFv-Fc sequence via a linker scFv-Fab sequence. H

[0313] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding to CD3 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising, from the N-terminus to the C-terminus, a V H1 sequence coupled to a C H sequence, and the light chain comprising, from the N-terminus to the C-terminus, a V L sequence coupled to a C L sequence; (ii) a first scFv domain capable of binding to CD28 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V scFv sequence coupled to a first scFv V L sequence via a first linker H sequence; (iii) a second scFv domain capable of binding to Cldn18.2 on tumor cells, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, a second scFv V scFv sequence coupled to a second scFv V L sequence via a second linker H sequence; and (iv) a heterodimeric 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 via the C-terminus of the Fab C H1 sequence and via a linker Fab-Fc sequence; b) the C-terminus of the first scFv domain is coupled to the N-terminus of the Fab V scFv-Fab sequence via a linker H sequence; and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the Fab V scFv-Fc ​Coupled to the N-terminus of the second Fc polypeptide.

[0314] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding CD3 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising from the N-terminus to the C-terminus a V H1 sequence coupled to a C H sequence, and the light chain comprising from the N-terminus to the C-terminus a V L sequence coupled to a C L sequence; (ii) a first scFv domain capable of binding CD28 on a second immune cell, wherein the first scFv domain comprises from the N-terminus to the C-terminus a first scFv V scFv sequence coupled via a first linker H to a first scFv V L sequence; (iii) a second scFv domain capable of binding Cldn18.2 on a tumor cell, wherein the second scFv domain comprises from the N-terminus to the C-terminus a second scFv V scFv sequence coupled via a second linker L to a second scFv V H sequence; and (iv) a heterodimeric 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 via the C-terminus of a FabC H1 sequence and via a linker Fab-Fc , b) the N-terminus of the first scFv domain is coupled to the C-terminus of a Fab C Fab-scFv sequence via a linker L , and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker scFv-Fc .

[0315] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding CD3 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising from the N-terminus to the C-terminus a V H1 sequence coupled to a C H sequence, and the light chain comprising from the N-terminus to the C-terminus a V L sequence coupled to a C L sequence; (ii) a first scFv domain capable of binding CD28 on a second immune cell, wherein the first scFv domain comprises from the N-terminus to the C-terminus a first scFv V scFv sequence coupled via a first linker HFirst scFvV of sequential coupling L sequence; (iii) a second scFv domain capable of binding to Cldn18.2 on tumor cells, wherein the second scFv domain contains, from the N-terminus to the C-terminus, via a second linker scFv coupled to the second scFv V L sequence of the second scFv V H sequence; and (iv) a heterodimeric 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 via the C-terminus of the FabC H1 sequence and via a linker Fab-Fc coupled to the N-terminus of the first Fc polypeptide, b) the N-terminus of the first scFv domain is coupled to the C-terminus of the first Fc polypeptide via a linker Fc-scFv coupled to the C-terminus of the first Fc polypeptide, and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker scFv-Fc coupled to the N-terminus of the second Fc polypeptide.

[0316] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding to CD3 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, and the heavy chain portion contains, from the N-terminus to the C-terminus, a V H1 sequence coupled to the C H sequence, and the light chain contains, from the N-terminus to the C-terminus, a V L sequence coupled to the C L sequence; (ii) a first scFv domain capable of binding to CD28 on a second immune cell, wherein the first scFv domain contains, from the N-terminus to the C-terminus, a first scFvV scFv sequence coupled to the first scFv V H sequence via a first linker L sequence; (iii) a second scFv domain capable of binding to Cldn18.2 on tumor cells, wherein the second scFv domain contains, from the N-terminus to the C-terminus, a second scFv V scFv sequence coupled to the second scFv V L sequence via a second linker H sequence; and (iv) a heterodimeric 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 via the C-terminus of the FabC H1 sequence and via a linker Fab-Fc coupled to the N-terminus of the first Fc polypeptide, b) the N-terminus of the first scFv domain is coupled to the C-terminus of the first Fc polypeptide via a linker Fc-scFv coupled to the C-terminus of the first Fc polypeptide, and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the Fab V scFv-Fab sequence via a linkerH N - terminal coupling of the sequence.

[0317] In one embodiment, a trivalent trispecific antibody construct is described herein, the trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding CD3 on a first immune cell, wherein the Fab domain comprises a heavy - chain portion paired with a light chain, the heavy - chain portion comprising from the N - terminal to the C - terminal a V sequence coupled to a C H1 sequence, and the light chain comprising from the N - terminal to the C - terminal a V sequence coupled to a C H sequence; (ii) a first scFv domain capable of binding CD28 on a second immune cell, wherein the first scFv domain comprises from the N - terminal to the C - terminal a first scFv V sequence coupled via a first linker L sequence; (iii) a second scFv domain capable of binding Cldn18.2 on a tumor cell, wherein the second scFv domain comprises from the N - terminal to the C - terminal a second scFv V sequence coupled via a second linker L sequence; (iv) a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein a) the Fab domain is coupled to the N - terminal of the second Fc polypeptide via the C - terminal of the FabC scFv sequence and via a linker H sequence; b) the C - terminal of the first scFv domain is coupled to the N - terminal of the first Fc polypeptide via a linker L sequence; and c) the C - terminal of the second scFv domain is coupled to the N - terminal of the Fab V scFv sequence via a linker L sequence; and (iv) a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein a) the Fab domain is coupled to the N - terminal of the second Fc polypeptide via the C - terminal of the FabC H sequence and via a linker H1 sequence; b) the C - terminal of the first scFv domain is coupled to the N - terminal of the first Fc polypeptide via a linker Fab-Fc sequence; and c) the C - terminal of the second scFv domain is coupled to the N - terminal of the Fab V scFv-Fc sequence via a linker scFv-Fab sequence and to the N - terminal of the Fab V H sequence.

[0318] In one embodiment, a trivalent trispecific antibody construct is described herein, the trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding CD28 on a first immune cell, wherein the Fab domain comprises a heavy - chain portion paired with a light chain, the heavy - chain portion comprising from the N - terminal to the C - terminal a V sequence coupled to a C H1 sequence, and the light chain comprising from the N - terminal to the C - terminal a V sequence coupled to a C H sequence; (ii) a first scFv domain capable of binding CD3 on a second immune cell, wherein the first scFv domain comprises from the N - terminal to the C - terminal a first scFv V sequence coupled via a first linker L sequence; (iii) a second scFv domain capable of binding Cldn18.2 on a tumor cell, wherein the second scFv domain comprises from the N - terminal to the C - terminal a second scFv V sequence coupled via a second linker L sequence; (iii) a second scFv domain capable of binding Cldn18.2 on a tumor cell, wherein the second scFv domain comprises from the N - terminal to the C - terminal a second scFv V sequence coupled via a second linker scFv sequence to a first scFv V LThe first scFv V of the sequence conjugate H sequence; (iii) a second scFv domain capable of binding to Cldn18.2 on tumor cells, wherein the second scFv domain contains, from the N-terminus to the C-terminus, via a second linker scFv coupled to the second scFv V L sequence of the second scFvV H sequence; and (iv) a heterodimeric 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 via the C-terminus of the Fab C H1 sequence and via a linker Fab-Fc b) the C-terminus of the first scFv domain is coupled to the N-terminus of the Fab V scFv-Fab via a linker H sequence, and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker scFv-Fc sequence.

[0319] In one embodiment, a trivalent trispecific antibody construct is described herein, the trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding to CD28 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion containing, from the N-terminus to the C-terminus, a V H1 sequence coupled to a C H sequence, and the light chain containing, from the N-terminus to the C-terminus, a V L sequence coupled to a C L sequence; (ii) a first scFv domain capable of binding to CD3 on a second immune cell, wherein the first scFv domain contains, from the N-terminus to the C-terminus, a first scFv V scFv sequence coupled to a first scFv V H sequence via a first linker L sequence; (iii) a second scFv domain capable of binding to Cldn18.2 on tumor cells, wherein the second scFv domain contains, from the N-terminus to the C-terminus, a second scFvV scFv sequence coupled to a second scFv V L sequence via a second linker H sequence; and (iv) a heterodimeric 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 via the C-terminus of the Fab C H1 sequence and via a linker Fab-Fc b) the N-terminus of the first scFv domain is coupled to the Fab C Fab-scFv via a linker LC-terminal coupling of the sequence, and c) the C-terminal of the second scFv domain via a linker scFv-Fc coupled to the N-terminal of the second Fc polypeptide.

[0320] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding CD28 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising from the N-terminus to the C-terminus a V H1 sequence coupled to a C H sequence, and the light chain comprising from the N-terminus to the C-terminus a V L sequence coupled to a C L sequence; (ii) a first scFv domain capable of binding CD3 on a second immune cell, wherein the first scFv domain comprises from the N-terminus to the C-terminus a first scFv V scFv sequence coupled to a first scFv V H sequence via a first linker L ; (iii) a second scFv domain capable of binding Cldn18.2 on a tumor cell, wherein the second scFv domain comprises from the N-terminus to the C-terminus a second scFvV scFv sequence coupled to a second scFv V L sequence via a second linker H ; and (iv) a heterodimeric 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 via the C-terminus of the Fab C H1 sequence and via a linker Fab-Fc ; b) the N-terminus of the first scFv domain is coupled to the C-terminus of the first Fc polypeptide via a linker Fc-scFv ; and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the second Fc polypeptide via a linker scFv-Fc .

[0321] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding CD28 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising from the N-terminus to the C-terminus a V H1 sequence coupled to a C H sequence, and the light chain comprising from the N-terminus to the C-terminus a V L sequence coupled to a C L sequence; (ii) a first scFv domain capable of binding CD3 on a second immune cell, wherein the first scFv domain comprises from the N-terminus to the C-terminus a first scFv VscFv The first scFv V coupled with H sequence; (iii) a second scFv domain capable of binding to Cldn18.2 on tumor cells, wherein the second scFv domain contains, from the N-terminus to the C-terminus, via a second linker L sequence; (iii) a second scFv domain capable of binding to Cldn18.2 on tumor cells, wherein the second scFv domain contains, from the N-terminus to the C-terminus, via a second linker scFv coupled with the second scFv V L sequence; and (iv) a heterodimeric 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 via the C-terminus of the Fab C H sequence and via a linker H1 sequence; b) the N-terminus of the first scFv domain is coupled to the C-terminus of the first Fc polypeptide via a linker Fab-Fc sequence; and c) the C-terminus of the second scFv domain is coupled to the N-terminus of the Fab V Fc-scFv sequence via a linker scFv-Fab sequence. H sequence.

[0322] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a Fab domain capable of binding to CD28 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, and the heavy chain portion contains, from the N-terminus to the C-terminus, a V H1 sequence coupled with a C H sequence, and the light chain contains, from the N-terminus to the C-terminus, a V L sequence coupled with a C L sequence; (ii) a first scFv domain capable of binding to CD3 on a second immune cell, wherein the first scFv domain contains, from the N-terminus to the C-terminus, a first scFv V scFv sequence coupled with a first scFv V L sequence via a first linker H sequence; (iii) a second scFv domain capable of binding to Cldn18.2 on tumor cells, wherein the second scFv domain contains, from the N-terminus to the C-terminus, a second scFvV scFv sequence coupled with a second scFv V L sequence via a second linker H sequence; and (iv) a heterodimeric 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 via the C-terminus of the Fab C H1 sequence and via a linker Fab-Fc sequence; b) the C-terminus of the first scFv domain is coupled to the first Fc polypeptide via a linker scFv-Fccoupled to the N-terminus of a second Fc polypeptide, and c) the C-terminus of the second scFv domain via a linker scFv-Fab coupled to the N-terminus of Fab V H sequence.

[0323] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 100 or 118 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 111 or 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 115 or 120 or consisting of said amino acid sequence.

[0324] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 100 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 111 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0325] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 118 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 120 or consisting of said amino acid sequence.

[0326] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 122 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 111 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 123 or consisting of said amino acid sequence.

[0327] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 124 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 111 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0328] In one embodiment, a trivalent trispecific antibody construct is described herein, the trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 124 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0329] In one embodiment, a trivalent trispecific antibody construct is described herein, the trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 129 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 111 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 120 or consisting of said amino acid sequence.

[0330] In one embodiment, a trivalent trispecific antibody construct is described herein, the trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 122 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 123 or consisting of said amino acid sequence.

[0331] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 129 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 120 or consisting of said amino acid sequence.

[0332] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 100 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0333] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 134 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 135 or consisting of said amino acid sequence.

[0334] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 130 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 131 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 120 or consisting of said amino acid sequence.

[0335] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 132 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 133 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 120 or consisting of said amino acid sequence.

[0336] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 136 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 131 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0337] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 137 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 138 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0338] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 139 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0339] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 140 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0340] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 100 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 141 or consisting of said amino acid sequence.

[0341] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 100 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 142 or consisting of said amino acid sequence.

[0342] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in any of SEQ ID NOs: 143-149 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0343] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 143 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0344] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 144 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0345] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 145 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0346] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 146 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99%, or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0347] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 147 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0348] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 148 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0349] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 149 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 115 or consisting of said amino acid sequence.

[0350] In one embodiment, described herein is a trivalent trispecific antibody construct comprising: (i) a first heavy chain polypeptide (H1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 100 or consisting of said amino acid sequence; (ii) a second heavy chain polypeptide (H2) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 125 or consisting of said amino acid sequence; and (iii) a light chain polypeptide (L1) comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 150 or SEQ ID NO: 152 or consisting of said amino acid sequence.

[0351] In certain embodiments, the trivalent trispecific antibody constructs of the present disclosure are antibody constructs in which the first scFv domain and the second scFv domain are not coupled to each other in tandem (i.e., the structure from the N-terminus to the C-terminus is scFv1-scFv2 or scFv2-scFv with or without a linker sequence between scFv1 and scFv2).

[0352] G. Certain properties of the trivalent trispecific antibody constructs

[0353] The trivalent trispecific antibody constructs of the present disclosure can have several specific properties due to their format, geometry, and antigen affinities.

[0354] Thus, in some embodiments, e.g., in the tumor (micro)environment, the engagement of a trivalent trispecific antibody construct with two different antigens on one or more T cells and an antigen on tumor cells can be - at least temporarily - simultaneous, thereby establishing a TCR-independent immunological synapse and directing T cell-mediated cytotoxic activity to the tumor environment containing tumor cells expressing TAAs. In various embodiments, as further described herein, the trivalent trispecific antibody constructs can cause a significant reduction in immune cell (e.g., T cell) activation in the absence of TAAs, e.g., when an immunological synapse cannot be fully formed due to the absence of TAAs. This property can be superior to those of conventional constructs because it allows, for example, a more TAA-dependent activation of the subject's immune system and thus can cause fewer off-target effects in the subject compared to conventional antibody constructs that act in a less TAA-dependent manner.

[0355] In some embodiments, compared to conventional constructs that target only one immune cell antigen (e.g., CD3 or CD28), the antibody constructs described herein can have enhanced anti-tumor activity in tumors with relatively low T cell infiltration, due to co-stimulatory activity that can engage both CD3 and CD28 on the same immune cell (e.g., T cell) or on two different (e.g., adjacent) immune cells (e.g., T cell).

[0356] Generally, as would be understood by one of ordinary skill in the art, natural T cell activation may require TCR (e.g., involving CD3) and CD28 stimulation. The antibody constructs of the present disclosure have been specifically designed, for example, by their format and geometry, to provide CD3 and CD28 co-stimulation. Additionally, 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, Fab, etc.) of the antibody constructs described herein, as well as their relative arrangement within the construct, have been specifically selected and engineered to generate an immune cell (e.g., T cell) activation signal of appropriate strength to, on the one hand, reduce T cell exhaustion and, on the other hand, reduce T cell overreaction and dysfunction, and to provide an improved ratio of anti-tumor on-target activity to healthy tissue off-target activity. In various embodiments, such improved on-target to off-target activity is achieved by optimizing the format and geometry of the antibody construct in such a way that simultaneous engagement of all three antigens (i.e., CD3, CD28, and TAA) allows for the most effective anti-tumor activity compared to a situation where, for example, only CD3 and CD28 are engaged and bound by the construct.

[0357] As further described herein, the antibody constructs of the present disclosure can be trivalent and trispecific, and bind each antigen, e.g., CD3, CD28, TAA, monovalently via one of its three antigen-binding domains.

[0358] In some embodiments, the binding affinity of the trivalent trispecific antibody construct for the TAA is at least about 40 nM, 30 nM, 20 nM, 10 nM, or 5 nM, about 40 nM to about 5 nM, or about 30 nM to about 10 nM, for example, using SPR or other methods known in the art. In one embodiment, the TAA is MSLN. In another embodiment, the TAA is Cldn18.2.

[0359] In some embodiments, the melting temperatures of Tm1, Tm2, and / or Tm3 of the trivalent trispecific antibody construct are within 10°C, 5°C, 2°C, or 1°C of the melting temperature of a bivalent monospecific IgG1 monoclonal antibody, as determined using, for example, differential scanning calorimetry (DSC) or differential scanning fluorimetry (DSF). In some embodiments, such a bivalent monospecific IgG1 monoclonal antibody can be any conventional IgG1 antibody capable of binding a specific antigen. In certain embodiments, such a monoclonal monospecific bivalent IgG1 antibody is an antibody comprising two Fab domains as described herein in the context of the trispecific antibody construct. In some embodiments, the bivalent monospecific IgG1 monoclonal antibody comprises two anti-CD3 Fab domains as described herein, two anti-CD28 Fab domains as described herein, or two anti-MSLN Fab domains comprising the anti-MSLN VH and VL sequences as described herein, or two anti-Cldn18.2 Fab domains comprising the anti-Cldn18.2 VH and VL sequences as described herein.

[0360] In some embodiments, the trivalent trispecific antibody construct of the present disclosure binds cytotoxic effector cells (e.g., T cells) expressing CD3 and CD28 with an affinity of about 5 nM to about 100 pM, about 1 nM to about 100 pM, about 1 nM to about 250 pM, about 1 nM to about 500 pM, or about 1 nM to about 750 pM. In certain embodiments, such an antibody construct binds cytotoxic effector cells expressing 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 about 2-fold to about 200-fold, about 2-fold to about 150-fold, about 2-fold to about 100-fold, about 20-fold to about 200-fold higher than the affinity of the corresponding bispecific anti-CD3xTAA and / or anti-CD28xTAA antibody constructs.

[0361] In some embodiments, an engineered anti-CD28 binding domain with reduced CD28 affinity (e.g., compared to the parental huTN228 complementarity determining region) can be used to generate a trivalent trispecific antibody construct that induces less CD28-mediated toxicity.

[0362] In some embodiments, for the killing of tumor cells expressing a TAA by TDCC using a 2:1 E:T ratio and a 72-hour incubation period in the presence of cytotoxic effector cells, the tumor cells expressing the TAA express at least about 100,000 TAAs / cell, and the trivalent trispecific antibody construct of the present disclosure exhibits an IC 50Values are from about 50 pM to about 0.01 pM, about 25 pM to about 0.01 pM, about 10 pM to about 0.05 pM, about 10 pM to about 0.1 pM, about 10 pM to about 1 pM, about 5 pM to about 1 pM. In some such embodiments, the maximum killing of tumor cells expressing the TAA by the antibody construct reaches at least about 60%, 65%, 70%, 75%, or 80%, 85%, or 90%, 100%, or from about 60% to about 100%, from about 70% to about 90%, or from about 75% to about 85%.

[0363] In some embodiments, when there are cells expressing the TAA at up to about 100,000 TAA / cell and a 2:1 E:T ratio and a 72-hour incubation period are used, the trivalent trispecific antibody constructs of the present disclosure are capable of inducing one or more cytokines at about 300 pg / mL to about 9000 pg / mL by cytotoxic effector cells.

[0364] Furthermore, in various embodiments, the trivalent trispecific antibody constructs of the present disclosure can provide a strict target cell (e.g., tumor cell)-dependent cytotoxicity profile, e.g., as shown herein, cytokine release is significantly reduced in the presence of only T cells alone compared to conditions where T cells are co-cultured with tumor cells expressing the TAA, see, e.g., Example 25.

[0365] In some embodiments, when measured using, for example, size exclusion chromatography or other methods known in the art, the thermal stability of the trivalent trispecific antibody constructs disclosed herein when measured over a period of about 2, 3, 5, 7, 10, or 14 days at 40 °C can be at least about 90%, 95%, 97%, 98%, or 99% of the intact construct. In certain embodiments, such constructs can comprise a Fab domain capable of binding CD3 or CD28 and two scFv domains, where one such scFv domain is capable of binding CD3 or CD28 (e.g., the antigen not bound by the Fab domain), and one scFv domain is capable of binding the TAA, such as in construct v37634. In some embodiments, the stability of the trispecific trivalent antibody construct over a 14-day period at 40 °C is at least about 97% or 98% (i.e., at least about 97% or 98% of the construct is intact as measured, for example, using size exclusion chromatography). In some embodiments, the concentration of the construct in such stability experiments is about 1 mg / mL.

[0366] H. Trivalent trispecific antibody constructs comprising a light chain containing a Fab portion and an scFv portion

[0367] Certain embodiments of the present disclosure relate to trivalent trispecific antibody constructs that comprise a light chain that (from the N-terminus to the C-terminus or from the C-terminus to the N-terminus) comprises a Fab portion containing a first VL sequence and a CL sequence coupled to (ii) a scFv domain containing a second VL sequence and a VH sequence.

[0368] In some embodiments, such a trivalent trispecific antibody construct comprises a light chain having the following domain structure: from the N-terminus to the C-terminus, VL-CL-scFv, wherein the scFv domain can contain, from the N-terminus to the C-terminus, a VH coupled to a VL sequence, or a VL sequence coupled to a VH sequence. Thus, in certain embodiments, the light chain has the domain structure (VL-CL) Fab -(VL-VH) scFv from the N-terminus to the C-terminus. In other embodiments, the light chain has the domain structure (VL-CL) Fab -(VH-VL) scFv from the N-terminus to the C-terminus.

[0369] In certain embodiments, described herein is a trivalent trispecific antibody construct that comprises a light chain having the following domain structure: (VL-CL) Fab -(VL-VH) scFv from the N-terminus to the C-terminus. In some embodiments, such a light chain can further comprise one or more linkers, as further described herein. In some embodiments, the light chain comprises a peptide linker between the Fab portion and the scFv portion Fab-scFv that couples the Fab portion to the scFv portion to yield a light chain having the domain structure (VL-CL) Fab -linker Fab-scFv -(VL-VH) scFv from the N-terminus to the C-terminus. The linker Fab-scFv can comprise or consist of the amino acid sequence shown in SEQ ID NO: 105. As further described herein, the scFv domain of the light chain can also comprise a linker that couples the VL sequence to the VH sequence scFv . Thus, in certain embodiments, the trivalent trispecific antibody construct comprises a light chain having the domain structure (VL-CL) Fab -linker Fab-scFv -(VL-linker scFv -VH) scFv from the N-terminus to the C-terminus. The linker scFv can comprise or consist of the amino acid sequence shown in SEQ ID NO: 104.

[0370] In certain embodiments, described herein is a trivalent trispecific antibody construct that comprises 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 shown in SEQ ID NO: 123.

[0371] In some embodiments, described herein is a trivalent trispecific antibody construct that comprises a light chain comprising a Fab portion and a scFv portion, wherein the antibody construct comprises: (i) a Fab domain that comprises a heavy chain and a light chain, the heavy chain comprising a VH sequence and a CH1 sequence, the light chain comprising a VL sequence and a CL sequence, wherein the Fab domain is capable of binding CD3; (ii) a first scFv domain that comprises a first VH sequence and a first VL sequence, wherein the first scFv domain is capable of binding CD28; (iii) a second scFv domain that comprises a second VH sequence and a second VL sequence, wherein the second scFv domain is capable of binding Cldn18.2; and (iv) a dimeric Fc domain that comprises 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 via its CH1 sequence, b) the first scFv domain is coupled to the C-terminus of the CL sequence of the Fab light chain, and c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide.

[0372] In some embodiments, described herein is a trivalent trispecific antibody construct that comprises a light chain comprising a Fab portion and a scFv portion, wherein the antibody construct comprises: (i) a Fab domain capable of binding CD3 on a first immune cell, wherein the Fab domain comprises a heavy chain portion paired with a light chain, the heavy chain portion comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C sequence, and the light chain comprising, from the N-terminus to the C-terminus, a V sequence coupled to a C sequence; (ii) a first scFv domain capable of binding CD28 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V sequence coupled via a first linker to a first scFv V sequence; (iii) a second scFv domain capable of binding Cldn18.2 on a tumor cell, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, a second scFv V sequence coupled via a second linker to a second scFv V sequence. H1 sequence coupled to a V H sequence, and the light chain comprising, from the N-terminus to the C-terminus, a V L sequence coupled to a C L sequence; (ii) a first scFv domain capable of binding CD28 on a second immune cell, wherein the first scFv domain comprises, from the N-terminus to the C-terminus, a first scFv V scFv sequence coupled to a first scFv V H sequence via a first linker L ; (iii) a second scFv domain capable of binding Cldn18.2 on a tumor cell, wherein the second scFv domain comprises, from the N-terminus to the C-terminus, a second scFv V scFv sequence coupled to a second scFv V L sequence via a second linker Ha sequence; and (iv) a heterodimeric Fc domain comprising a first Fc polypeptide and a second Fc polypeptide, wherein a) the Fab domain is attached via the C-terminus of the Fab C H1 sequence and via a linker Fab-Fc to the N-terminus of the first Fc polypeptide, b) the N-terminus of the first scFv domain is attached via a linker Fab-scFv to the C-terminus of the Fab C L sequence, and c) the C-terminus of the second scFv domain is attached via a linker scFv-Fc to the N-terminus of the second Fc polypeptide.

[0373] In various embodiments, such a trivalent trispecific antibody construct comprises a light chain having the domain structure (VL-CL) Fab -linker Fab-scFv -(VL-linker scFv -VH) scFv In certain embodiments, such a light chain comprises an amino acid sequence having at least about 80%, 85%, 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 123. In some embodiments, such a construct comprises a light chain comprising or consisting of the amino acid sequence shown in SEQ ID NO: 123.

[0374] In various embodiments, a trivalent trispecific antibody construct comprising a light chain containing a Fab portion and an scFv portion can have certain properties that can be unique to constructs having a particular format and geometry.

[0375] In some embodiments, the trivalent trispecific antibody constructs herein that comprise a light chain containing a Fab portion and an scFv portion do not reduce T cell viability (measured after incubation of the T cells with the respective construct for 48 hours) by more than 5%, 4%, 3%, 2%, 1%, or 0% compared to T cells treated with a negative control construct that does not contain a binding domain against Cldn18.2. In some embodiments, the trivalent trispecific antibody constructs do not reduce T cell viability by more than 5%, 3%, 1%, or 0%. In some embodiments, compared to the control construct, the trivalent trispecific antibody constructs do not reduce T cell viability by more than 5%. In some embodiments, compared to the control construct, the trivalent trispecific antibody constructs do not reduce T cell viability by more than 4%. In some embodiments, compared to the control construct, the trivalent trispecific antibody constructs do not reduce T cell viability by more than 3%. In some embodiments, compared to the control construct, the trivalent trispecific antibody constructs do not reduce T cell viability by more than 2%. In some embodiments, compared to the control construct, the trivalent trispecific antibody constructs do not reduce T cell viability by more than 1%. In some embodiments, compared to the control construct, the trivalent trispecific antibody constructs do not reduce T cell viability.

[0376] In some embodiments, the trivalent trispecific antibody constructs herein that comprise a light chain containing a Fab portion and an scFv portion reduce T cell viability by about 1.5-fold to about 2-fold, about 1.5-fold to about 3-fold, or about 2-fold to about 3-fold less than antibody constructs in which the first scFv domain and the second scFv domain are conjugated to the N-terminus of the Fab heavy chain or to the N-terminus of a second Fc polypeptide, and as measured when the respective antibody constructs are incubated with T cells for 48 hours. In some embodiments, the trivalent trispecific antibody constructs that comprise a light chain containing a Fab portion and an scFv portion reduce T cell viability by about 1.5-fold to about 2-fold less than antibody constructs in which the first scFv domain and the second scFv domain are conjugated to the N-terminus of the Fab heavy chain or to the N-terminus of a second Fc polypeptide. In some embodiments, the trivalent trispecific antibody constructs that comprise a light chain containing a Fab portion and an scFv portion reduce T cell viability by about 1.5-fold to about 3-fold less than antibody constructs in which the first scFv domain and the second scFv domain are conjugated to the N-terminus of the Fab heavy chain or to the N-terminus of a second Fc polypeptide. In some embodiments, the trivalent trispecific antibody constructs that comprise a light chain containing a Fab portion and an scFv portion reduce T cell viability by about 2-fold to about 3-fold less than antibody constructs in which the first scFv domain and the second scFv domain are conjugated to the N-terminus of the Fab heavy chain or to the N-terminus of a second Fc polypeptide.

[0377] In some embodiments, compared to antibody constructs in which the first scFv domain and the second scFv domain are independently coupled to the N-terminus of the Fab heavy chain or to the N-terminus of the second Fc polypeptide, the trivalent trispecific antibody construct containing a light chain with a Fab portion and an scFv portion herein induces cytokines that are about 80-fold to about 2000-fold, about 100-fold to about 1000-fold, or about 100-fold to about 500-fold less in an assay containing only human CD3 + T cells, and wherein the corresponding antibody construct is incubated with CD3 + T cells for 48 hours. In some embodiments, compared to antibody constructs in which the first scFv domain and the second scFv domain are independently coupled to the N-terminus of the Fab heavy chain or to the N-terminus of the second Fc polypeptide, the trivalent trispecific antibody construct containing a light chain with a Fab portion and an scFv portion herein induces cytokines that are about 80-fold to about 2000-fold less in an assay containing only human CD3 + T cells. In some embodiments, compared to antibody constructs in which the first scFv domain and the second scFv domain are independently coupled to the N-terminus of the Fab heavy chain or to the N-terminus of the second Fc polypeptide, the trivalent trispecific antibody construct containing a light chain with a Fab portion and an scFv portion herein induces cytokines that are about 100-fold to about 1000-fold less in an assay containing only human CD3 + T cells. In some embodiments, compared to antibody constructs in which the first scFv domain and the second scFv domain are independently coupled to the N-terminus of the Fab heavy chain or to the N-terminus of the second Fc polypeptide, the trivalent trispecific antibody construct containing a light chain with a Fab portion and an scFv portion herein induces cytokines that are about 100-fold to about 500-fold less in an assay containing only human CD3 + T cells. In some embodiments, the trivalent trispecific antibody construct containing a light chain with a Fab portion and an scFv portion induces less than about 10 pg / mL of cytokines, such as cytokines from about 10 pg / mL to 0.5 pg / mL.

[0378] In some embodiments, compared to antibody constructs in which the first scFv domain and the second scFv domain are independently coupled to the N-terminus of the Fab heavy chain or to the N-terminus of the second Fc polypeptide, the trivalent trispecific antibody construct containing a light chain with a Fab portion and an scFv portion herein induces cytokines that are about 5-fold to about 900-fold, about 5-fold to about 500-fold, or about 5-fold to about 300-fold less in an assay containing only human PBMCs, and wherein the corresponding antibody construct is incubated with T cells for 48 hours. In some embodiments, compared to antibody constructs in which the first scFv domain and the second scFv domain are independently coupled to the N-terminus of the Fab heavy chain or to the N-terminus of the second Fc polypeptide, the trivalent trispecific antibody construct containing a light chain with a Fab portion and an scFv portion herein induces cytokines that are about 5-fold to about 900-fold less in an assay containing only human PBMCs. In some embodiments, compared to antibody constructs in which the first scFv domain and the second scFv domain are independently coupled to the N-terminus of the Fab heavy chain or to the N-terminus of the second Fc polypeptide, the trivalent trispecific antibody construct containing a light chain with a Fab portion and an scFv portion herein induces cytokines that are about 5-fold to about 500-fold less in an assay containing only human PBMCs. In some embodiments, compared to antibody constructs in which the first scFv domain and the second scFv domain are independently coupled to the N-terminus of the Fab heavy chain or to the N-terminus of the second Fc polypeptide, the trivalent trispecific antibody construct containing a light chain with a Fab portion and an scFv portion herein induces cytokines that are about 5-fold to about 300-fold less in an assay containing only human PBMCs. In some embodiments, the trivalent trispecific antibody construct containing a light chain with a Fab portion and an scFv portion induces less than about 20 pg / mL of cytokines, such as cytokines from about 20 pg / mL to 0.5 pg / mL.

[0379] In certain embodiments, the cytokines include one or more of IL-2, interleukin-6 (IL-6), IFNγ, and TNFα.

[0380] In some embodiments, compared to trivalent trispecific antibody constructs with different formats and / or geometries, the trivalent trispecific antibody construct containing a light chain with a Fab portion and an scFv portion herein induces different amounts of memory T cell subsets upon stimulation of certain immune cells. In some embodiments, compared to combination therapy with two bivalent bispecific antibody constructs that together target the same antigens as the trivalent trispecific antibody construct containing a light chain with a Fab portion and an scFv portion herein, the trivalent trispecific antibody construct containing a light chain with a Fab portion and an scFv portion herein induces similar amounts of memory T cell subsets upon stimulation of certain immune cells.

[0381] In some embodiments, a trivalent trispecific antibody construct comprising a light chain containing a Fab portion and a scFv portion comprises: (i) a Fab domain capable of binding CD3; (ii) a first scFv domain capable of binding CD28; (iii) a second scFv domain capable of binding Claudin 18.2 (Cldn18.2); and (iv) an 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 via its C H1 sequence, b) the first scFv domain is coupled to the C L terminus of the sequence of the Fab light chain, and c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide.

[0382] In certain embodiments, the trivalent trispecific antibody construct comprising a light chain containing a Fab portion and a scFv portion is v37634.

[0383] III. Sequence Identity of Amino Acid and Nucleic Acid Sequences

[0384] As described in other parts of the present disclosure, certain embodiments herein relate to an isolated polypeptide or an isolated collection of polypeptides (e.g., polypeptide chains H1, H2, L1, etc., or portions thereof, such as domains) of a trivalent trispecific antibody construct, and a polynucleotide or a collection of polynucleotides encoding one or more polypeptide chains of the antibody construct described herein. In this context, the polynucleotide may encode all or part of the antibody construct, such as one or more polypeptide chains (e.g., H1, H2, L1, etc.) of the antibody construct.

[0385] In some embodiments, described herein is a nucleic acid molecule or a collection of nucleic acid molecules encoding one or more, two or more, or three or more polypeptide chains that form any of the trivalent trispecific antibody constructs disclosed herein.

[0386] The terms "nucleic acid", "nucleic acid molecule", and "polynucleotide" are used interchangeably herein and refer to a polymeric form of nucleotides (deoxyribonucleotides or ribonucleotides or their analogs) of any length. Non-limiting examples of polynucleotides include genes, gene fragments, messenger RNA (mRNA), cDNA, recombinant polynucleotides, plasmids, vectors, isolated DNA of any sequence, isolated RNA of any sequence, nucleic acid probes, and primers.

[0387] In some embodiments, described herein is a vector or a collection of vectors comprising a nucleic acid molecule or a collection of nucleic acid molecules encoding one or more polypeptide chains (e.g., one or more of H1, H2, L1, etc.) of the antibody construct disclosed herein.

[0388] A polynucleotide that “encodes” a given polypeptide is a polynucleotide that, when placed under the control of appropriate regulatory sequences, is transcribed (in the case of DNA) or translated (in the case of mRNA) in vivo into the polypeptide. 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’ (carboxyl) terminus. Transcription termination sequences can be located 3’ of the coding sequence.

[0389] 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. In the context of two or more polynucleotide or polypeptide sequences, the term “identical” refers to the same two or more sequences or subsequences, i.e., having the same nucleotide or amino acid monomer sequences, respectively (i.e., 100% sequence identity). When sequences are compared and aligned to obtain maximum correspondence as measured over a comparison window or over a specified region using one of the commonly used sequence comparison algorithms known to those of ordinary skill in the art or by manual alignment and visual inspection, the polypeptide or polynucleotide sequences herein have “sequence identity” if they have a certain percentage or a certain number of amino acid residues or nucleotides, respectively, i.e., at least about 80%, about 85%, about 90%, about 95%, about 97%, or at least about 99% identity within the specified region. This definition also refers to the components of the test polynucleotide sequence. Identity can exist over a region of at least about 50 amino acids or nucleotides in length, or over a region of about 75 to about 100 amino acids or nucleotides in length, or, in the absence of specification, over the entire sequence of the polynucleotide or polypeptide. For sequence comparison, the test sequence is usually compared to a designated reference sequence. When using a sequence comparison algorithm, the test sequence and the reference sequence are input into a computer, and if necessary, subsequence coordinates are designated and sequence algorithm program parameters are specified. Default program parameters can be used, or alternative parameters can be specified. The sequence comparison algorithm then calculates the percentage (%) of sequence identity of the test sequence relative to the reference sequence based on the program parameters.

[0390] As used herein, the term "comparison window" refers to a sequence segment that encompasses contiguous amino acid or nucleotide positions, which can be from about 20 to about 1000 contiguous amino acid or nucleotide positions, such as 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, and the two sequences can be compared over this segment after the test sequence is optimally aligned with a reference sequence having the same number of contiguous positions. In certain embodiments, longer segments up to and including the full-length sequence may also be used as the comparison window. Methods for sequence alignment for comparison are known to those of ordinary skill in the art. Optimal sequence alignment for comparison can be performed, for example, by the local homology algorithm of Smith & Waterman, 1970, Adv. Appl. Math., 2:482c; the homology alignment algorithm of Needleman & Wunsch, 1970, J. Mol. Biol., 48:443; the similarity search method of Pearson & Lipman, 1988, Proc. Natl. Acad. Sci. USA, 85:2444; or computerized implementations of these algorithms (e.g., GAP, BESTFIT, FASTA, or TFASTA in the Wisconsin Genetics Software Package, Genetics Computer Group, Madison, WI), or by manual alignment and visual inspection (see, e.g., Ausubel et al., Current Protocols in Molecular Biology, (Supplement 1995), Cold Spring Harbor Laboratory Press). Examples of available algorithms suitable for determining the 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 analysis is publicly available through the website of the National Center for Biotechnology Information (NCBI).

[0391] Certain embodiments described herein relate to variant sequences (e.g., variant V HDomains, variant Fc polypeptides, etc.), which contain one or more amino acid modifications, such as one or more amino acid insertions, one or more amino acid deletions, and / or one or more amino acid substitutions, when compared to, for example, a reference sequence (such as a wild-type sequence). In certain embodiments, when compared to a reference sequence (such as a wild-type sequence), one or more amino acid modifications of the variant sequence comprise one or more amino acid substitutions. In such embodiments, one or more of the 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. Generally, as used herein, "conservative substitution" is considered to be the replacement of one amino acid with another amino acid having similar physical, chemical, and / or structural properties. Common conservative substitutions are listed in column 1 of Table 4.

[0392] Table 4: Conservative Amino Acid Substitutions

[0393]

[0394] Those skilled in the art will understand that the main factors determining a conservative substitution are generally the size of the amino acid side chain and its physical / chemical properties, but certain circumstances allow for the replacement of a given amino acid with a wider range of amino acids than those listed in column 1 of Table 4. These additional amino acids tend to have properties similar to the amino acid being replaced, but vary more in size, or have a similar size but vary more in physical / chemical properties. This wider range of conservative substitutions is listed in column 2 of Table 4. Given the specific protein environment in which the amino acid substitution is made, the skilled person can readily determine the most appropriate set of substituents to choose.

[0395] IV. Pharmaceutical Compositions

[0396] In certain embodiments, the present disclosure relates to pharmaceutical compositions that may comprise one or more of the trivalent trispecific antibody constructs described herein. In various embodiments, the pharmaceutical compositions herein may further comprise pharmaceutically acceptable excipients, carriers, buffers, stabilizers, 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., the antibody construct). The exact nature of the carrier or other material may depend on the route of administration. Thus, the pharmaceutical compositions herein may be formulated for various uses and routes of administration, such as for oral, intravenous, dermal, subcutaneous, nasal, intramuscular, or intraperitoneal routes of administration.

[0397] A pharmaceutical composition for oral administration can be in the form of tablets, capsules, powders or liquids. Tablets may contain solid carriers such as gelatin or adjuvants. Liquid pharmaceutical compositions generally contain liquid carriers such as water, petroleum, animal or vegetable oils, mineral oils or synthetic oils. Physiological saline solutions, dextrose or other sugar solutions or diols such as ethylene glycol, propylene glycol or polyethylene glycol may be included.

[0398] For intravenous, cutaneous or subcutaneous injection, or injection at the site of affliction (i.e., at the tumor site), the active ingredient (i.e., the antibody construct) can be in the form of a parenterally acceptable aqueous solution that is pyrogen-free and has a suitable pH, isotonicity and stability. Persons skilled in the relevant art can well prepare a suitable solution using, for example, isotonic vehicles such as sodium chloride injection, Ringer's injection, lactated Ringer's injection. Preservatives, stabilizers, buffers, antioxidants and / or other additives can be included as needed.

[0399] For the administration of an antibody construct according to the present disclosure to a subject, it is preferably administered in a "therapeutically effective amount" sufficient to show benefit to the individual, as further described herein. The actual dosage administered, as well as the rate and time course of administration, can depend on the nature and severity of the disease being treated (e.g., cancer). The determination of the treatment prescription (e.g., dosage, etc.) is the responsibility of general practitioners and other physicians and generally takes into account the disorder to be treated, the condition of the individual patient, the site of delivery, the method of administration and other factors known to the practitioner. Examples of the techniques and regimens mentioned above can be found in Remington's Pharmaceutical Sciences, 16th Edition, Osol, A. (ed.), 1980.

[0400] In some embodiments, the pharmaceutical composition can contain a second active ingredient (e.g., another protein or small molecule) in addition to the antibody construct described herein.

[0401] Accordingly, there is also described herein a pharmaceutical composition comprising any one or more of the trivalent trispecific antibody constructs disclosed herein and a pharmaceutically acceptable carrier, excipient, diluent or combination thereof.

[0402] V. Kit

[0403] The present disclosure also describes a kit, the kit comprising one or more of the trivalent trispecific antibody constructs described herein or a pharmaceutical composition comprising such an antibody construct as described herein and instructions for use. Thus, in certain embodiments, a kit is described herein that comprises a vector for expressing the antibody construct described herein and instructions for use. In certain embodiments, a kit is described herein that comprises a host cell comprising a vector for expressing the antibody construct and instructions for use. In some embodiments, the present disclosure relates to a kit that comprises a purified antibody construct and instructions for use. The purified antibody construct may be lyophilized or provided in a dry form, such as a powder or granule, and the kit may additionally contain a suitable solvent for reconstituting the lyophilized or dry components.

[0404] The kit may further comprise a container and a label / or package insert on or associated with the container. The label or package insert contains instructions typically included in the commercial packaging of a therapeutic product that provide information or instructions regarding indications, usage, dosage, administration, contraindications, and / or warnings for the use of such a therapeutic product (e.g., the antibody constructs described herein). The label or package insert may also include a notice in a form required by a government agency regulating the manufacture, use, or sale of drugs or biological products that reflects the agency's approval of the manufacture, use, or sale for human or animal administration. The container may contain a composition comprising an antibody construct of the present disclosure. In some embodiments, the container may have a sterile access port. For example, the container may be an intravenous solution bag or a vial having a stopper that can be pierced by a hypodermic needle.

[0405] In addition to the container containing the composition comprising the antibody construct, the kit may 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 buffer or diluent may be included in such a kit.

[0406] Suitable containers may include, for example, bottles, vials, syringes, and intravenous solution bags, etc. The container may be formed of various materials such as glass or plastic. As appropriate, one or more components of the kit (e.g., the antibody construct) may be lyophilized or provided in a dry form (such as a powder or granule), and the kit may additionally contain a suitable solvent for reconstituting the lyophilized or dry components.

[0407] The kit herein may further comprise other materials desirable in the commercial and user's view, such as filters, needles, and syringes.

[0408] VI. Methods

[0409] The present disclosure also describes methods of generating and using the trivalent trispecific antibody constructs of the present disclosure.

[0410] A. Methods of generating antibody constructs

[0411] In some embodiments, the present disclosure relates to methods of preparing the trivalent trispecific antibody constructs described herein. In various embodiments, the antibody constructs of the present disclosure can be produced using standard recombinant methods known in the art (see, e.g., U.S. Patent No. 4,816,567 and “Antibodies: A Laboratory Manual,” 2nd ed., Greenfield, ed., Cold Spring Harbor Laboratory Press, New York, 2014).

[0412] To recombinantly produce the antibody constructs described herein, polynucleotides or sets of polynucleotides encoding the antibody constructs can be generated and inserted into one or more vectors for further cloning and / or expression in host cells. The polynucleotides encoding the antibody constructs can be produced by standard methods known in the art (see, e.g., Ausubel et al., Current Protocols in Molecular Biology, John Wiley & Sons, New York, 1994 & updates, and “Antibodies: A Laboratory Manual,” 2nd ed., Greenfield, ed., Cold Spring Harbor Laboratory Press, New York, 2014). As will be understood by those skilled in the art, the number of polynucleotides required for antibody construct expression can depend on the format and / or geometry of the antibody construct, including, for example, the number of polypeptide chains the antibody construct comprises. For example, when the antibody construct comprises three polypeptide chains (e.g., H1, H2, and L1), three polynucleotides each encoding one polypeptide chain can be used. In embodiments where two or more polynucleotides are used, such two or more polynucleotides can be incorporated into one vector or more than one vector (e.g., two or three separate vectors).

[0413] Typically, for expression, a polynucleotide or collection 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 that can be used for efficient transcription of the polynucleotide. Examples of such regulatory elements include, but are not limited to, promoters, enhancers, terminators, and polyadenylation signals. Those skilled in the art will understand that the choice of regulatory elements can depend on the host cell selected for expressing the polypeptide of the antibody construct, and 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 the expression or purification of the expressed protein. Examples include, but are not limited to, signal peptides and affinity tags, such as metal affinity tags, histidine tags, avidin / streptavidin coding sequences, glutathione-S-transferase (GST) coding sequences, and biotin coding sequences. The expression vector can be an extrachromosomal vector or an integrating vector. Thus, in some embodiments, the amino acid sequence of the polypeptide chain (e.g., chain H1, H2, L1, etc.) of the expressed antibody construct described herein can include a signal peptide sequence. Such signal peptide sequences can vary depending on the expression system and conditions used to produce the antibody construct. Exemplary signal peptide sequences can include the amino acid sequence METDTLLLWVLLLWVPGSTG (SEQ ID NO:155) or MRPTWAWWLFLVLLLALWAPARG (SEQ ID NO:156) (e.g., for H1, H2, etc.), or MRPTWAWWLFLVLLLALWAPARG (SEQ ID NO:156) or MGWSCIILFLVATATGVHS (SEQ ID NO:157) (e.g., for L1, L2, etc.). In certain embodiments, after intracellular polypeptide chain expression, one or more heavy chains (e.g., H1, H2, etc.) of the antibody construct described herein can include a C-terminal lysine residue. In various embodiments, such C-terminal lysine residues can be enzymatically cleaved from the polypeptide chain before further processing (e.g., purification, formulation, etc.) and before using the corresponding antibody construct, e.g., before administering the construct to a subject in need.

[0414] Certain embodiments of the present disclosure for generating antibody constructs relate to a vector (such as an expression vector) comprising one or more polynucleotides encoding at least a portion of the antibody constructs described herein. The one or more polynucleotides can be contained in a single vector or more than one vector. In some embodiments, the polynucleotides are contained in a polycistronic vector. Expression vectors that can be used to express the polynucleotides include, but are not limited to, pTT5 and pUC15 cells containing vectors encoding the antibody construct.

[0415] Suitable host cells for cloning or expressing antibody construct polypeptides include a variety of prokaryotic or eukaryotic cells known in the art. Eukaryotic host cells include, for example, mammalian cells, plant cells, insect cells, and yeast cells (such as cells of the genus Saccharomyces or Pichia). Prokaryotic host cells include, for example, Escherichia coli, Aeromonas salmonicida, or Bacillus subtilis cells. In certain embodiments, antibody constructs can be produced in bacteria, particularly when glycosylation and Fc effector functions are not required or desired for the intended purpose of the antibody construct, as described, for example, in U.S. Pat. Nos. 5,648,237, 5,789,199, and 5,840,523, and Charlton, Methods in Molecular Biology, Vol. 248, pp. 245-254, edited by B.K.C. Lo, Humana Press, Totowa, N.J., 2003. In certain embodiments, eukaryotic microorganisms such as filamentous fungi or yeast are suitable expression host cells, particularly fungal and yeast strains in which the glycosylation pathway has been "humanized" such that antibodies with a partially or fully human glycosylation pattern are produced (see, for example, Gerngross, 2004, Nat. Biotech. 22:1409-1414, and Li et al., 2006, Nat. Biotech. 24:210-215).

[0416] In various embodiments, suitable host cells for expressing glycosylated antibody constructs are eukaryotic cells. For example, U.S. Pat. Nos. 5,959,177, 6,040,498, 6,420,548, 7,125,978, and 6,417,429 describe PLANTIBODIES for producing antibodies and their fragments (e.g., scFv, Fab, etc.) in transgenic plants TMTechniques. Mammalian cell lines suitable for suspension growth are particularly useful for the expression of the antibody constructs described herein. Examples include, but are not limited to, the simian kidney CV1 line transformed by SV40 (COS-7), the human embryonic kidney (HEK) line 293 or 293 cells (see, e.g., Graham et al., 1977, J. Gen Virol., 36:59), baby hamster kidney cells (BHK), mouse Sertoli TM4 cells (see, e.g., Mather, 1980, Biol Reprod, 23:243-251); simian kidney cells (CV1), African green monkey kidney cells (VERO-76), human cervical carcinoma (HeLa) cells, canine kidney cells (MDCK), buffalo rat hepatocytes (BRL3A), human lung cells (W138), human hepatocytes (HepG2), mouse mammary tumor (MMT 060562), TRI cells (see, e.g., Mather et al., 1982, Annals N.Y. Acad Sci, 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, NS0, and Sp2 / 0). An exemplary review of mammalian host cell lines suitable for antibody production is provided in Yazaki & Wu, Methods in Molecular Biology, Vol. 248, pp. 255-268 (B.K.C. Lo, ed., Humana Press, Totowa, N.J., 2003).

[0417] In certain embodiments, the host cell for generating the trivalent trispecific antibody constructs 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, NS0, or COS cell. In some embodiments, the host cell is a stable cell line that permits mature glycosylation of the antibody construct.

[0418] Host cells containing one or more expression vectors encoding the antibody construct can be cultured using conventional methods to produce the antibody construct. Alternatively, in some embodiments, host cells containing an expression vector encoding the antibody construct can be used therapeutically or prophylactically to deliver the antibody construct to a subject, or a polynucleotide or expression vector can be administered ex vivo to cells from a subject and the cells then returned to the subject's body.

[0419] In some embodiments, the host cell comprises a vector (e.g., has been transformed with a vector) that comprises a polynucleotide encoding the V of the binding domain of the antibody construct described herein L and V H polynucleotides. In some embodiments, the host cell comprises a vector (e.g., has been transformed with a vector) that comprises a polynucleotide encoding the full-length polypeptide chain of the antibody construct described herein (e.g., H1, H2, or L1 as described herein). In another example, the host cell comprises a first vector that comprises a polynucleotide encoding the V of the binding domain L and a second vector that comprises a polynucleotide encoding the corresponding V of the binding domain H . In various embodiments, the host cell is eukaryotic, such as Chinese hamster ovary (CHO) cells, human embryonic kidney (HEK) cells, or lymphoid cells (e.g., Y0, NS0, Sp20 cells). In certain embodiments, the host cell is Expi293 TM (Thermo Fisher, Waltham, MA). In certain embodiments, the host cells used herein are CHO-S cells (National Research Council Canada) or HEK293 cells.

[0420] Certain embodiments of the present disclosure relate to methods of preparing antibody constructs, the methods comprising culturing a host cell that has been introduced with one or more polynucleotides encoding the antibody construct or one or more expression vectors encoding the antibody construct under conditions suitable for expression of the antibody construct. Such methods may also include recovering the antibody construct from the host cell (or from the host cell culture medium). In some embodiments, such methods may also include purifying the antibody construct.

[0421] Cell culture media that can be used include, but are not limited to, DMEM (Thermo Fisher, Waltham, MA), Opti-MEM TM (Thermo Fisher, Waltham, MA), Opti-MEM TM I Reduced Serum Medium (Thermo Fisher, Waltham, MA), RPMI-1640 medium, Expi293 TMExpression media (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 antifungals (e.g., amphotericin) or any other supplements routinely used to support cell culture.

[0422] In various embodiments, the antibody constructs of the present disclosure are purified after expression. Proteins such as the antibody constructs of the present disclosure can be isolated or purified in a variety of ways known to those skilled in the art (see, e.g., Protein Purification: Principles and Practice, 3rd Edition, Scopes, Springer-Verlag, NY, 1994). Standard purification methods that can be used for the antibody constructs disclosed herein include chromatographic techniques using systems such as FPLC and HPLC at atmospheric or high pressure, including ion exchange, hydrophobic interaction, affinity, size exclusion or gel filtration, and reverse phase chromatography. Additional purification methods include electrophoresis, immunoprecipitation, dialysis, and chromatofocusing techniques. Ultrafiltration and diafiltration techniques in combination with protein concentration can also be used. As is well known in the art, a variety of native proteins bind to the Fc domain and other domain elements of antibody constructs, and in some embodiments, these proteins can be used for the purification of antibody constructs. For example, bacterial protein A and G can bind to the Fc domain of some antibody constructs. Similarly, bacterial protein L can bind to the Fab domain of some antibody constructs. Purification can also be achieved by specific fusion partners. For example, if a GST fusion is employed, the antibody construct can be purified using glutathione resin, if a His tag is used, it can be purified using Ni +2 affinity chromatography, or if a Flag tag is used, it can be purified using immobilized anti-Flag antibody. The degree of purification required can vary depending on the use of the antibody construct. Thus, in some embodiments, purification may not be necessary.

[0423] In certain embodiments, the antibody constructs of the present disclosure are substantially pure. The term "substantially pure" (or "substantially purified"), when used in reference to the antibody constructs described herein, means that the antibody constructs are substantially or essentially free of components (such as the starting cells, or in the case of recombinantly produced antibody constructs, the host cells) that are normally associated with or interact with the protein as found in its native environment. In certain embodiments, a substantially pure antibody construct is an antibody construct that has been 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 proteinaceous material.

[0424] Assessment of antibody construct purity and / or homogeneity can be carried out by any method known in the art, including but not limited to, non-reducing / reducing CE-SDS, non-reducing / reducing SDS-PAGE, ultra-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).

[0425] In certain embodiments, the antibody constructs described herein can comprise one or more post-translational modifications. Such post-translational modifications can occur in vivo, or they can be carried out in vitro after the antibody construct has been isolated from the host cell.

[0426] Post-translational modifications can include a variety of modifications known in the art (see, e.g., Proteins - Structure and Molecular Properties, 2nd Edition, 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, pp. 1-12, 1983; Seifter et al., 1990, Meth. Enzymol., 182:626-646, and Rattan et al., 1992, Ann. N.Y. Acad. Sci., 663:48-62). In those embodiments in which the antibody construct comprises one or more post-translational modifications, the antibody construct can comprise the same type of modification at one or more sites (e.g., amino acid residues), or it can comprise different modifications at different sites.

[0427] 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 NaBH4.

[0428] Other examples of post-translational modifications can include, for example, addition or removal of N-linked or O-linked carbohydrate chains, chemical modification of N-linked or O-linked carbohydrate chains, processing of the N-terminus or C-terminus, attachment of chemical moieties to the amino acid backbone, and addition or deletion of N-terminal methionine residues resulting from expression in a prokaryotic host cell. Post-translational modifications can also include modification with detectable labels such as enzymatic, fluorescent, isotopic, or affinity labels to permit detection and isolation of the protein. Examples of suitable enzyme labels include, but are not limited to, horseradish peroxidase, alkaline phosphatase, β-galactosidase, and acetylcholinesterase. Examples of suitable cofactor 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.

[0429] Additional examples of post-translational modifications can include acetylation, ADP-ribosylation, amidation, covalent attachment of flavin, covalent attachment of a heme moiety, covalent attachment of a nucleotide or nucleotide derivative, covalent attachmen...

Claims

1. An antibody construct comprising: (i) A Fab domain capable of binding to a first antigen on a first cytotoxic effector cell; (ii) A first scFv domain and a second scFv domain, wherein one of the scFv domains is capable of binding to a second antigen on a second cytotoxic effector cell and the other scFv domain is capable of binding to a tumor-associated antigen (TAA) on a tumor cell; and (iii) An 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, and (b) The first scFv structure and the second scFv domain are independently coupled to (i) the N-terminus of the Fab domain, (ii) the C-terminus of the Fab domain, (iii) the C-terminus of one of the Fc polypeptides, or (iv) the N-terminus of the second Fc polypeptide, provided that when one of the scFv domains is coupled to the C-terminus of one of the Fc polypeptides, the first antigen is CD3 and the second antigen is CD28, or the first antigen is CD28 and the second antigen is CD3.

2. The antibody construct according to any one of claims 1, wherein the first scFv domain and the second scFv domain are not coupled to each other in tandem.

3. The antibody construct according to any one of claims 1 or 2, wherein the first antigen is CD28 and the second antigen is CD3.

4. The antibody construct according to any one of claims 1 or 2, wherein the first antigen is CD3 and the second antigen is CD28.

5. The antibody construct according to any one of claims 1-4, wherein the first scFv domain is coupled to the N-terminus of the Fab domain, and the second scFv is coupled to the N-terminus of the second Fc polypeptide.

6. The antibody construct according to claim 5, wherein the first scFv domain is coupled to the N-terminus of the V H sequence of the heavy chain of the Fab domain.

7. The antibody construct according to claim 5, wherein the first scFv domain is coupled to the N-terminus of the V L sequence of the light chain of the Fab domain.

8. The antibody construct according to claim 5 or 6, wherein the antibody construct comprises: a) A first heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) said first scFv domain that comprises a first scFv V L sequence coupled to a first scFv V H sequence (V H -V L ), or a first scFv V H sequence coupled to a first scFv V L sequence (V L -V H ), (ii) a heavy chain Fab sequence that comprises a Fab V H1 sequence coupled to a Fab C H sequence, and (iii) said first Fc polypeptide; b) A second heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) the second scFv domain that comprises a second scFv V L sequence coupled to a second scFv V H sequence (V H -V L ), or a second scFv V H sequence coupled to a second scFv V L sequence (V L -V H ), and (ii) the second Fc polypeptide; and c) A light chain polypeptide that, from the N-terminus to the C-terminus, comprises: a Fab V L sequence coupled to a Fab C L sequence, wherein: The heavy chain Fab sequence and the light chain polypeptide associate to form the Fab domain, and The first Fc polypeptide and the second Fc polypeptide associate to form the Fc domain.

9. The antibody construct according to any one of claims 1-4, wherein the first scFv domain is coupled to the C-terminus of the light chain C of the Fab domain L sequence, and the second scFv domain is coupled to the N-terminus of the second Fc polypeptide.

10. The antibody construct according to claim 9, wherein the antibody construct comprises: a) A first heavy chain polypeptide, which from the N-terminus to the C-terminus comprises: (i) a heavy chain Fab sequence, which comprises a Fab V sequence coupled to a Fab C sequence, and (ii) the first Fc polypeptide; H1 sequence, and (ii) said first Fc polypeptide; H sequence, and (ii) the first Fc polypeptide; b) A second heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) said second scFv domain, which comprises a second scFv V L sequence coupled to a second scFv V H sequence (V H -V L ), or a second scFv V H sequence coupled to a second scFv V L sequence (V L -V H ), and (ii) said second Fc polypeptide; and c) A light chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) a light chain Fab sequence that comprises a FabV sequence coupled to a Fab C L sequence, and (ii) the first scFv domain that comprises a first scFv V sequence coupled to a first scFv V L sequence (V L -V H ), or a first scFv V sequence coupled to a first scFv V H -V L ), or a first scFv V sequence coupled to a first scFv V H sequence (V L -V L -V H ). wherein: The heavy chain Fab sequence and the light chain Fab sequence associate to form the Fab domain, and The first Fc polypeptide and the second Fc polypeptide associate to form the Fc domain.

11. The antibody construct according to any one of claims 1-4, wherein the first scFv domain is coupled to the N-terminus of the Fab domain, and the second scFv domain is coupled to the C-terminus of one of the Fc polypeptides.

12. The antibody construct according to claim 11, wherein the first scFv domain is coupled to the N-terminus of the V H domain of the Fab domain.

13. The antibody construct according to any one of claims 11-12, wherein the second scFv domain is coupled to the C-terminus of the first Fc polypeptide.

14. The antibody construct according to any one of claims 11-12, wherein the second scFv domain is coupled to the C-terminus of the second Fc polypeptide.

15. The antibody construct according to any one of claims 11-13, wherein the antibody construct comprises: a) A first heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) said first scFv domain, which comprises a first scFv V L sequence coupled to a first scFv V H sequence (V H -V L ), or a first scFv V H sequence coupled to a first scFv V L sequence (V L -V H ), (ii) a heavy chain Fab sequence that comprises a Fab V H1 sequence coupled to a Fab C H sequence, (iii) said first Fc polypeptide, and (iv) said second scFv domain, which comprises a second scFv V L sequence coupled to a second scFv V H sequence (V H -V L ), or a second scFv V H sequence coupled to a second scFv V L sequence (V L -V H ); b) a second heavy chain polypeptide comprising the second Fc polypeptide; and c) A light chain polypeptide comprising a Fab V L sequence coupled to a Fab C L sequence, wherein: the heavy chain Fab sequence and the light chain polypeptide associate to form the Fab domain, and the first Fc polypeptide and the second Fc polypeptide associate to form the Fc domain.

16. The antibody construct according to any one of claims 11-12 or 14, wherein the antibody construct comprises: a) A first heavy chain polypeptide, which from the N-terminus to the C-terminus comprises: (i) said first scFv domain, which comprises a first scFv V L sequence coupled to a first scFv V H sequence (V H -V L ), or a first scFv V H sequence coupled to a first scFv V L sequence (V L -V H ), (ii) a heavy chain Fab sequence, which comprises a Fab V H1 sequence coupled to a Fab C H sequence, and (iii) said first Fc polypeptide; b) A second heavy chain polypeptide that, from the N-terminus to the C-terminus, comprises: (i) the second Fc polypeptide, and (ii) the second scFv domain that comprises a second scFv V L sequence conjugated to a second scFv V H sequence (V H -V L ), or a second scFv V H sequence conjugated to a second scFv V L sequence (V L -V H ); and c) A light chain polypeptide comprising a Fab V L sequence coupled to a Fab C L sequence, wherein: the heavy chain Fab sequence and the light chain polypeptide associate to form the Fab domain, and the first Fc polypeptide and the second Fc polypeptide associate to form the Fc domain.

17. The antibody construct according to any one of claims 1-16, wherein the first scFv domain is capable of binding to the TAA, and the second scFv domain is capable of binding to the second antigen on the second cytotoxic effector cell.

18. The antibody construct according to any one of claims 1-16, wherein the second scFv domain is capable of binding to the TAA, and the first scFv domain is capable of binding to the second antigen on the second cytotoxic effector cell.

19. The antibody construct according to any one of claims 1, 9 or 10, which comprises: (i) the Fab domain capable of binding to CD3; (ii) the first scFv domain capable of binding to CD28; (iii) the second scFv domain capable of binding to Claudin18.2 (Cldn18.2); and (iv) the 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 via its C H1 sequence b) The first scFv domain is coupled to the C-terminus of the Fab light chain, and L the C-terminus of the sequence, and c) the second scFv domain is coupled to the N-terminus of the second Fc polypeptide.

20. The antibody construct according to claim 19, wherein the antibody construct does not reduce T cell viability by more than 5%, 3%, 1% or 0% compared to T cells treated with a negative control construct lacking a binding domain for Cldn18.2, and wherein the antibody construct is incubated with the T cells for 48 hours.

21. The antibody construct according to claim 19 or claim 20, wherein the antibody construct reduces T cell viability by about 1.5-fold to about 2-fold, about 1.5-fold to about 3-fold, or about 2-fold to about 3-fold less than an antibody construct in which the first scFv domain and the second scFv domain are independently coupled to the N-terminus of the Fab heavy chain or to the N-terminus of the second Fc polypeptide, and wherein the corresponding antibody construct is incubated with the T cells for 48 hours.

22. The antibody construct according to any one of claims 19-21, wherein, compared to an antibody construct in which the first scFv domain and the second scFv domain are independently coupled to the N-terminus of the Fab heavy chain or to the N-terminus of the second Fc polypeptide, in an assay comprising human CD3 + T cells, the antibody construct induces cytokines that are about 80-fold to about 2000-fold, about 100-fold to about 1000-fold, or about 100-fold to about 500-fold less, and wherein the corresponding antibody construct is incubated with the T cells for 48 hours.

23. The antibody construct according to any one of claims 19 - 22, wherein compared to an antibody construct in which the first scFv domain and the second scFv domain are independently coupled to the N-terminus of the Fab heavy chain or to the N-terminus of the second Fc polypeptide, the antibody construct induces cytokines that are about 5 - fold to about 900 - fold, about 5 - fold to about 500 - fold, or about 5 - fold to about 300 - fold less in an assay comprising human PBMCs, and wherein the corresponding antibody construct is incubated with the T cells for 48 hours.

24. The antibody construct according to any one of claims 22 - 23, wherein the cytokine comprises one or more of IL-2, TNFα, IFNγ, and IL-6.

25. The antibody construct according to any one of claims 1-24, wherein the first scFv domain has a domain structure of V H -V L from the N-terminus to the C-terminus.

26. The antibody construct according to any one of claims 1-24, wherein the first scFv domain has a domain structure of V L -V H from the N-terminus to the C-terminus.

27. The antibody construct according to any one of claims 1-26, wherein the second scFv domain has a domain structure of V H -V L from the N-terminus to the C-terminus.

28. The antibody construct according to any one of claims 1-26, wherein the second scFv domain has a domain structure of V L -V H from the N-terminus to the C-terminus.

29. The antibody construct according to any one of claims 1 - 28, wherein the antibody construct comprises one or more linkers.

30. The antibody construct according to claim 29, wherein the one or more linkers are peptide linkers, each linker comprising or consisting of an amino acid sequence of 1 to about 50, 2 to about 40, 3 to about 30, or 5 to about 25 consecutive amino acid residues.

31. The antibody construct according to any one of claims 29-30, wherein the first scFv domain comprises a linker scFv1 .

32. The antibody construct according to claim 31, wherein the linker scFv1 couples the N-terminus or C-terminus of the V H domain to the C-terminus or N-terminus of the V L domain, respectively, and comprises or consists of an amino acid sequence having about 80%, 90% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO:

104.

33. The antibody construct according to any one of claims 29 - 32, wherein the second scFv domain comprises a linker scFv2 .

34. The antibody construct according to claim 33, wherein the linker scFv2 couples the N-terminus or C-terminus of the V H domain to the C-terminus or N-terminus of the V L domain, respectively, and comprises or consists of an amino acid sequence having about 80%, 90% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO:

104.

35. The antibody construct according to any one of claims 1-34, wherein the Fab domain capable of binding to the first antigen on the first cytotoxic effector cell comprises a heavy chain constant domain (C H1 ), and the heavy chain constant domain 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 shown in SEQ ID NO:

107.

36. The antibody construct according to any one of claims 1-35, wherein the antibody construct is capable of binding to human CD28, and its dissociation constant (K D ) is about 10 nM to about 500 nM, about 20 nM to about 600 nM, about 20 nM to about 250 nM, about 20 nM to about 150 nM, about 20 nM to about 100 nM, or about 20 nM to about 50 nM, as determined using SPR.

37. The antibody construct according to any one of claims 1-36, wherein the antibody construct comprises: an anti-CD28 V H sequence comprising an HCDR1 having the sequence SX1GVH (SEQ ID NO: 302), an HCDR2 having the sequence VIWX2GGX3TNFNSALMS (SEQ ID NO: 306), and an HCDR3 having the sequence DRAX4GX5YX6X7AMDY (SEQ ID NO: 312); and an anti-CD28 V L sequence comprising an LCDR1 having the sequence RAS ESVEYYX8TSLMQ (SEQ ID NO: 315), an LCDR2 having the sequence AASX9VX 10 S (SEQ ID NO: 319), and an LCDR3 having the sequence QQSRKVPFT (SEQ ID NO: 320), and wherein X1 = Y or A; X2 = P or A; X3 = G or S; X4 = S or Y; X5 = N or A; X6 = L or N; X7 = S or Y; X8 = G or V; X9 = N or A; and X 10 = E or D.

38. The antibody construct according to any one of claims 1-37, wherein the antibody construct comprises an anti-CD28 V H sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 106 and an anti-CD28 V L sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO:

116.

39. The antibody construct according to claim 38, wherein the anti-CD28 V H sequence comprises an amino acid sequence selected from the group consisting of SEQ ID NO: 201, 203-208, and 210 or consists of said amino acid sequence, and the anti-CD28 V L sequence comprises an amino acid sequence selected from the group consisting of SEQ ID NO: 200, 202, and 209 or consists of said amino acid sequence.

40. The antibody construct according to any one of claims 1-39, wherein the antibody construct comprises: an anti-CD28 V H sequence comprising an HCDR1 having the sequence SYGVH (SEQ ID NO: 300), an HCDR2 having the sequence VIWPGGGTNFNSALMS (SEQ ID NO: 303), and an HCDR3 having the sequence DRAYGNYLYAMDY (SEQ ID NO: 307); and an anti-CD28 V L sequence comprising an LCDR1 having the sequence RASESVEYYVTSLMQ (SEQ ID NO: 313), an LCDR2 having the sequence AASNVDS (SEQ ID NO: 316), and an LCDR3 having the sequence QQSRKVPFT (SEQ ID NO: 320).

41. The antibody construct according to any one of claims 1-40, wherein the antibody construct is capable of binding to human CD3, and its dissociation constant (K D ) for CD3 is about 20 nM to about 200 nM, about 30 nM to about 150 nM, about 40 nM to about 100 nM, or 50 nM to about 80 nM, as determined by SPR.

42. The antibody construct according to any one of claims 1-41, wherein the antibody construct comprises: an anti-CD3 V H sequence, which respectively comprises the HCDR1-3 sequences shown in SEQ ID NO: 321-323; and an anti-CD3 V L sequence, which respectively comprises the LCDR1-3 sequences shown in SEQ ID NO: 324-326.

43. The antibody construct according to any one of claims 1-42, wherein the antibody construct comprises: an anti-CD3 V H sequence comprising an HCDR1 having the sequence GVTFNYYG (SEQ ID NO: 321), an HCDR2 having the sequence ITSSGGRI (SEQ ID NO: 322), and an HCDR3 having the sequence TLDGRDGWVAY (SEQ ID NO: 323); and an anti-CD3V L sequence comprising an LCDR1 having the sequence TGNIGSNY (SEQ ID NO: 324), an LCDR2 having the sequence RND (SEQ ID NO: 325), and an LCDR3 having the sequence QSYSSGFI (SEQ ID NO: 326).

44. The antibody construct according to any one of claims 1-43, wherein the antibody construct comprises an anti-CD3V having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 102 H sequence and an anti-CD3V having at least about 90%, 95%, 97%, 99% or 100% sequence identity to the amino acid sequence shown in SEQ ID NO: 103 L sequence.

45. The antibody construct according to any one of claims 1 - 44, wherein the first cytotoxic effector cell and the second cytotoxic effector cell are different cells.

46. The antibody construct according to any one of claims 1 - 44, wherein the first cytotoxic effector cell and the second cytotoxic effector cell are the same cell.

47. The antibody construct according to any one of claims 1 - 46, wherein the first antigen and the second antigen are located on the T cell.

48. The antibody construct according to any one of claims 1 - 47, wherein the TAA is Cldn18.

2.

49. The antibody construct according to claim 48, wherein the antibody construct comprises: an anti-Cldn18.2 V H sequence, which respectively comprises the HCDR1-3 sequences shown in SEQ ID NO: 333-335; and an anti-Cldn18.2 V L sequence, which respectively comprises the LCDR1-3 sequences shown in SEQ ID NO: 336-338.

50. The antibody construct according to claim 49, wherein the antibody construct comprises: an anti-Cldn18.2 V H sequence comprising an HCDR1 having the sequence SNPMI (SEQ ID NO: 333), an HCDR2 having the sequence IIDTDGSTYYADWAKG (SEQ ID NO: 334), and an HCDR3 having the sequence RLHGSSNGYYDDL (SEQ ID NO: 335); and an anti-Cldn18.2 V L sequence comprising an LCDR1 having the sequence QASQSIYSYLS (SEQ ID NO: 336), an LCDR2 having the sequence KASTLAS (SEQ ID NO: 337), and an LCDR3 having the sequence QQGYTVTNVDKNT (SEQ ID NO: 338).

51. The antibody construct according to any one of claims 48 - 50, wherein the anti-Cldn18.2 V H sequence comprises or consists of an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 127, and the anti-Cldn18.2 V L sequence comprises or consists of an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 128, and optionally wherein the antibody construct binds to human Cldn18.2 with an affinity of about 1 nM to about 80 nM, about 10 nM to about 60 nM, about 10 nM to about 50 nM, or about 20 nM to about 50 nM, as determined by flow cytometry.

52. The antibody construct according to any one of claims 1 - 51, wherein the first Fc polypeptide and the second Fc polypeptide of the Fc domain each comprise or consist of a CH2 sequence and a CH3 sequence.

53. The antibody construct according to claim 52, wherein at least one of the CH2 sequences of the first Fc polypeptide and the second Fc polypeptide is an IgG1 or IgG4 CH2 sequence and comprises one or more amino acid modifications compared to an unmodified wild-type IgG1 or IgG4 CH2 sequence.

54. The antibody construct according to claim 53, wherein both CH2 sequences of the first Fc polypeptide and the second Fc polypeptide are IgG1 or IgG4 CH2 sequences and comprise one or more amino acid modifications compared to an unmodified wild-type IgG1 or IgG4 CH2 sequence.

55. The antibody construct according to claims 53 - 54, wherein the one or more amino acid modifications to the CH2 sequence reduce or eliminate the interaction of the Fc domain with one or more Fc receptors, optionally with one or more Fcγ receptors.

56. The antibody construct according to any one of claims 52-55, wherein the CH2 sequence of both the first Fc polypeptide and the second Fc polypeptide 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 shown in SEQ ID NO:

109.

57. The antibody construct according to any one of claims 52-56, wherein at least one of the CH3 sequences of the first Fc polypeptide and the second Fc polypeptide is an IgG1 or IgG4 CH3 sequence and comprises one or more amino acid modifications as compared to an unmodified wild-type IgG1 or IgG4 CH3 sequence, and optionally wherein the first Fc polypeptide and the second Fc polypeptide have different amino acid sequences and form a heterodimeric Fc domain.

58. The antibody construct according to claim 57, wherein the two CH3 sequences of the first Fc polypeptide and the second Fc polypeptide are IgG1 or IgG4 CH3 sequences and comprise one or more amino acid modifications that promote preferential pairing of the first Fc polypeptide and the second Fc polypeptide to form the heterodimeric Fc domain as compared to formation of the corresponding homodimeric Fc domain.

59. The antibody construct according to any one of claims 57-58, 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 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 the amino acid residues in the Fc polypeptide is according to the EU numbering system.

60. The antibody construct according to any one of claims 52-59, wherein the CH3 sequence of one Fc polypeptide 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 shown in SEQ ID NO:

110.

61. The antibody construct according to any one of claims 52-60, wherein the CH3 sequence of the other Fc polypeptide 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 shown in SEQ ID NO:

114.

62. The antibody construct according to any one of claims 1 - 61, wherein the antibody construct is trivalent and trispecific, and binds each antigen monovalently.

63. The antibody construct according to any one of claims 1 - 62, wherein the melting temperature of Tm1, Tm2, and / or Tm3 of the antibody construct is within 10°C, 5°C, 2°C, or within 1°C of the melting temperature of the corresponding bivalent monospecific IgG1 monoclonal antibody.

64. The antibody construct according to any one of claims 1 - 63, wherein the antibody construct binds the cytotoxic effector cells expressing CD3 and / or CD28 with an affinity that is about 2 - fold, 5 - fold, 10 - fold, 20 - fold, 50 - fold, 100 - fold, or about 200 - fold higher than the affinity of the corresponding bivalent bispecific anti - CD3xTAA and / or anti - CD28xTAA antibody constructs.

65. The antibody construct according to any one of claims 1-64, wherein for killing tumor cells expressing TAA by TDCC using an E:T ratio of 2:1 and an incubation period of 72 hours in the presence of the cytotoxic effector cells, the tumor cells expressing TAA express at least about 100,000 TAAs / cell, and the antibody construct exhibits an IC 50 value of from about 50 pM to about 0.01 pM, from about 25 pM to about 0.01 pM, from about 10 pM to about 0.05 pM, from about 10 pM to about 0.1 pM, from about 10 pM to about 1 pM, from about 5 pM to about 1 pM.

66. The antibody construct according to claim 65, wherein the maximum killing of tumor cells expressing TAA by the antibody construct reaches at least about 60%, 65%, 70%, 75%, or 80%, 85%, or 90%, 100%, or about 60% to about 100%, about 70% to about 90%, or about 75% to about 85%.

67. The antibody construct according to any one of claims 1 - 66, wherein when there are cells expressing TAA at about 100,000 TAA / cell and a 2:1 E:T ratio and a 72 - hour incubation period are used, the antibody construct is capable of inducing one or more cytokines produced by the cytotoxic effector cells at about 300 pg / mL to about 9000 pg / mL.

68. A pharmaceutical composition comprising the antibody construct according to any one of claims 1 to 67, and a pharmaceutically acceptable carrier, excipient, diluent, or a combination thereof.

69. A nucleic acid molecule or collection of nucleic acid molecules encoding one or more, two or more, or three or more polypeptide chains that form the antibody construct according to any one of claims 1 to 67.

70. A vector or collection of vectors comprising the nucleic acid molecule or collection of nucleic acid molecules of claim 69.

71. A cell comprising the nucleic acid molecule or collection of nucleic acid molecules of claim 69, or the vector or collection of vectors of claim 70.

72. A method for producing the antibody construct according to any one of claims 1 to 67, the method comprising: (a) obtaining a host cell culture comprising at least one host cell, the at least one host cell comprising one or more nucleic acid molecules encoding one or more, two or more, or three or more polypeptide chains that form the antibody construct; and (b) recovering the antibody construct from the host cell culture.

73. The method according to claim 72, further comprising purifying the antibody construct after step (b).

74. 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 an antibody construct as recited in any one of claims 1-67, wherein the immune cells express the first antigen and the second antigen, and the tumor cells express the TAA.

75. 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 an antibody construct as recited in any one of claims 1-67, wherein the immune cells express the first antigen and the second antigen, and the tumor cells express the TAA.

76. 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 an antibody construct as recited in any one of claims 1-67, wherein the immune cells express the first antigen and the second antigen, and the tumor cells express the TAA.

77. The method according to any one of claims 74 to 76, wherein the immune cells comprise T cells.

78. The method according to any one of claims 74-77, wherein the TAA is Cldn18.

2.

79. The method according to any one of claims 74-78, wherein the antibody construct binds CD3 and CD28 on one T cell or two different T cells and the TAA on the tumor cells.

80. The method according to claim 79, wherein the binding of the antibody construct to the first antigen, the second antigen, and the TAA forms a TCR-independent artificial immune synapse between the one or more immune cells and the tumor cells, thereby eliciting a cytotoxic immune response of the immune cells against the tumor cells.

81. The method according to any one of claims 74 to 80, wherein the cell population is located in a subject.

82. A method for treating cancer in a subject in need thereof, the method comprising administering to the subject an antibody construct as recited in any one of claims 1 to 67.

83. The method according to claim 82, wherein a cytotoxic immune response against the cancer is elicited in the subject, thereby treating the cancer in the subject.

84. An antibody construct as recited in any one of claims 1 to 67 for use in cancer treatment.

85. Use of an antibody construct as recited in any one of claims 1 to 67 in the manufacture of a medicament for cancer treatment.

86. An antibody construct comprising a binding domain capable of binding to CD28, wherein the binding domain comprises: V H sequence comprising an HCDR1 having the sequence SX1GVH (SEQ ID NO: 302), an HCDR2 having the sequence VIWX2GGX3TNFNSALMS (SEQ ID NO: 306), and an HCDR3 having the sequence DRAX4GX5YX6X7AMDY (SEQ ID NO: 312); and V L sequence comprising an LCDR1 having the sequence RASESVEYYX8TSLMQ (SEQ ID NO: 315), an LCDR2 having the sequence AASX9VX 10 S (SEQ ID NO: 319), and an LCDR3 having the sequence QQSRKVPFT (SEQ ID NO: 320), the binding domain having one or more of the following amino acid substitutions at the positions identified in the CDR sequences as: X1: Y to A; X2: P to A; X3: G to S; X4: S to Y; X5: N to A; X6: L to N; X7: S to Y; X8: G to V; X9: N to A; and / or X 10 : E to D.

87. The antibody construct according to claim 86, wherein the binding domain comprises: V H sequence, which comprises an amino acid sequence selected from the group consisting of SEQ ID NO: 201, 203-208, and 210 or consists of the amino acid sequence; and V L sequence, which comprises an amino acid sequence selected from the group consisting of SEQ ID NO: 200, 202, and 209 or consists of the amino acid sequence.

88. An antibody construct comprising a binding domain capable of binding to Cldn18.2, wherein the binding domain comprises: V H sequence comprising an HCDR1 having the sequence SNPMI (SEQ ID NO: 333), an HCDR2 having the sequence IIDTDGSTYYADWAKG (SEQ ID NO: 334), and an HCDR3 having the sequence RLHGSSNGYYDDL (SEQ ID NO: 335); and V L sequence comprising an LCDR1 having the sequence QASQSIYSYLS (SEQ ID NO: 336), an LCDR2 having the sequence KASTLAS (SEQ ID NO: 337), and an LCDR3 having the sequence QQGYTVTNVDKNT (SEQ ID NO: 338).

89. The antibody construct according to claim 88, wherein the binding domain comprises: V H sequence comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 127 or consisting of said amino acid sequence; and V L sequence comprising an amino acid sequence having at least about 90%, 95%, 97%, 99% or 100% sequence identity with the amino acid sequence shown in SEQ ID NO: 128 or consisting of said amino acid sequence.

90. The antibody construct according to any one of claims 86-89, which further comprises one or more additional binding domains that are capable of binding one or more additional antigens.

Images