EA045372B1 - ANTI-TIGIT ANTIBODIES - Google Patents

Description

Cancer immunotherapy relies on modulating the immune system to enhance recognition and response against tumor cells. Such modulation can be achieved by a variety of mechanisms, including activation of co-stimulatory molecules present in immune cells or through inhibition of coinhibitory receptors. Activation of the immune response is a complex mechanism involving multiple cell populations, such as antigen-presenting cells, important for initiating an antigen-specific response, and effector cells, responsible for the destruction of tumor cells. The mechanisms that modulate the activity of effector cells such as cytotoxic T cells are numerous and represent a good target choice in the context of cancer immunotherapy.

TIGIT (T cell immunoreceptor with Ig and ITIM domains), also called WUCAM, VSIG9 or Vstm3, is a coinhibitory receptor predominantly expressed on NK, CD8+ and CD4+ T cells, as well as regulatory T cells (Treg cells or simply Treg). TIGIT is a transmembrane protein containing the known ITIM domain in the intracellular part, a transmembrane domain and an immunoglobulin variable domain in the extracellular part of the receptor. Several ligands have been described to bind to the TIGIT receptor with CD155/PVR showing the best affinity, followed by CD113/PVRL3 and CD112/PVRL2 (Yu et al. (2009) Nat. Immunol. 10:48.). DNAM/CD226, a known costimulatory receptor also expressed on NK and T cells, competes with TIGIT for binding CD155 and CD112, but with lower affinity, suggesting tight control of the activation of these effector cells to avoid uncontrolled cytotoxicity to normal cells expressing the ligand CD155.

TIGIT expression is increased on tumor-infiltrating lymphocytes (TILs) and in disease conditions such as HIV infection. TIGIT expression marks exhausted T cells that have lower effector function compared to TIGIT negative counterparts (Kurtulus et al. (2015) J. Clin. Invest. 276:112; Chew et al. (2016) Plos Pathogens. 12) . Conversely, Treg cells that express TIGIT exhibit increased immunosuppressive activity compared to the TIGIT-negative Treg population (Joller et al. (2014) Immunity. 40:569.).

Similar to other coinhibitory receptors (PD1 or CTLA4) expressed on T cells, which have been shown to be relevant targets for immunotherapy and for which antagonistic antibodies have been approved for the treatment of human cancer, the development of an antagonistic anti-TIGIT antibody may help turn on the immune system and better fight cancer cells. It has been suggested that antagonistic anti-TIGIT antibodies, alone or in combination with an a-PD1 antibody, may achieve potent antitumor efficacy in preclinical models (Johnston et al. (2014) Cancer Cell 26:1; WO 2016/028656; US 2016/ 0176963; US 2016/0376365, all of which are incorporated herein by reference).

Thus, antagonistic antibodies specific for TIGIT, which can inhibit the activity of the TIGIT receptor, provide the opportunity to reduce the immunosuppressive effect associated with the tumor microenvironment and thereby enhance the antitumor immune response against tumor cells.

The essence of the invention

This invention relates to anti-TIGIT antibodies that can reduce the immunosuppressive effect of TIGIT-mediated signaling. In particular, the antibodies or antigen-binding fragments of the present invention can inhibit TIGIT-mediated immunosuppression by preventing ligand binding to T cells (conventional αβ T cells and uncommon γδ T cells) and NK cells and/or depletion of TIGIT-positive Treg cells and /or by inducing internalization of the TIGIT receptor.

In one aspect, the present invention provides an isolated antibody or antigen binding fragment thereof that binds to human TIGIT, and which contains a heavy chain variable domain comprising heavy chain CDR1 (HCDR1), heavy chain CDR2 (HCDR2) and heavy chain CDR3 (HCDR3), selected from the HCDR1, HCDR2 and HCDR3 sequences shown in FIG. 1 and which further comprises a light chain variable domain comprising light chain CDR1 (LCDR1), light chain CDR2 (LCDR2) and light chain CDR3 (LCDR3) selected from the sequences LCDR1, LCDR2 and LCDR3 shown in FIG. 2.

In some embodiments of the present invention, the antibody or antigen binding fragment comprises a combination of HCDR1, HCDR2, HCDR3, LCDR1, LCDR2, and LCDR3, the combination selected from the group of combinations formed by HCDR from each antibody in FIG. 1 taken from the LCDR from the corresponding antibody in FIG. 2.

In some embodiments of the present invention, an antibody or antigen binding fragment of the present invention may comprise a heavy chain variable domain having an amino acid sequence selected from the group consisting of: SEQ ID NO: 211, 213, 215, 217, 219, 221, 223, 225, 227, 229, 231, 233, 235, 237, 239, 327, 329 and 331 and amino acid sequences demonstrating at least 90, 95, 97, 98 or 99% sequence identity to

- 1 045372 him; and optionally comprising a light chain variable domain having an amino acid sequence selected from the group consisting of the amino acid sequences of SEQ ID NO:

212, 214, 216, 218, 220, 222, 224, 226, 228, 230, 232, 234, 236, 238, 240, 328, 330 and 332 and amino acid sequences having at least 90%, 95%, 97 % 98% or 99% sequence identity.

In some embodiments of the present invention, the antibody or antigen binding fragment comprises a combination of a heavy chain variable domain and a light chain variable domain, the combination selected from the group of combinations formed by the VH of each antibody in FIG. 5, or an amino acid sequence exhibiting at least 90, 95, 97, 98, or 99% sequence identity to VL from the same antibody in FIG. 5, or an amino acid sequence demonstrating at least 90, 95, 97, 98, or 99% sequence identity with it.

The most preferred antibodies and antigen binding fragments provided herein are those based on the CDRs or complete variable domains of antibody 31282 provided herein.

As demonstrated herein, these preferred anti-TIGIT antibodies and antigen binding fragments based on antibody 31282 have particularly surprising and useful properties. These properties include: higher affinity for TIGIT expressed on CD8 T cells (from healthy donors or from cancer patients) compared with each previously described anti-TIGIT antibody tested; better IC ₅₀ for competition with CD155/PVR compared to each previously described anti-TIGIT antibody tested; better EC ₅₀ in T cell activation assays compared with each previously described anti-TIGIT antibody tested; and greatly increased activity in the peripheral blood T cells of the cancer patient and, most importantly, in tumor-infiltrating lymphocytes. In addition, this document unexpectedly demonstrates that the antibodies and antigen binding fragments of this invention, especially those based on antibody 31282, preferentially deplete Treg cells. That is, TIGIT-expressing Treg cells exposed to anti-TIGIT antibodies are killed in a greater proportion compared to conventional CD4 and CD8 T cells. This is surprising because conventional CD4 and CD8 T cells also express TIGIT but do not undergo cell lysis to the same extent when exposed to antibodies. In addition, it has surprisingly been demonstrated that the antibodies and antigen binding fragments of the present invention, especially those based on antibody 31282, not only stimulate normal pro-inflammatory T cell activity, but also increase the activity of unusual γδ T cells.

Thus, in certain preferred embodiments of the present invention, provided herein is an antibody or antigen binding fragment comprising HCDR1,

HCDR2, HCDR3, LCDR1, LCDR2 and LCDR3, where: HCDR1 contains or consists of SEQ ID NO HCDR2 contains or consists of SEQ ID NO HCDR3 contains or consists of SEQ ID NO LCDR1 contains or consists of SEQ ID NO: LCDR2 contains or consists of SEQ ID NO:

LCDR3 contains or consists of SEQ ID NO:

(YTFTSYYMH), (VIGPSGASTSYAQKFQG), (ARDHSDYWSGIMEV), (RASQSVRSSYLA), (GASSRAT) and (QQYFSPPWT).

In certain such embodiments of the present invention, the heavy chain variable domain comprises or consists of an amino acid sequence according to SEQ ID NO: 221 or an amino acid sequence exhibiting at least 90%, 95%, 97%, 98%, or 99% sequence identity with it, and the light chain variable domain contains or consists of an amino acid sequence in accordance with SEQ ID NO: 222 or an amino acid sequence exhibiting at least 90, 95, 97, 98 or 99% sequence identity with it.

In certain preferred embodiments of this invention, the anti-TIGIT antibody is antibody 31282, described herein.

In a further aspect, the present invention provides an isolated antibody or antigen binding fragment thereof that cross-competes for binding to human TIGIT with an antibody of the first aspect of the invention, for example, an antibody exemplified herein.

In a further aspect, the present invention provides an isolated antibody or antigen binding fragment thereof that binds to the same epitope as an antibody of the first aspect of the invention, for example, an antibody exemplified herein.

In a further aspect, the present invention provides an antibody or antigen binding fragment thereof that binds to a human TIGIT epitope comprising TIGIT residues Q56 and I109, optionally containing residues Q56, N58 and I109. In preferred embodiments

- 2 045372 of this invention provides an antibody or an antigen-binding fragment thereof that binds to a human TIGIT epitope containing TIGIT residues Q56, N58, E60, I68, L73, H76 and I109.

In some embodiments of the present invention, the antibody or antigen-binding fragment thereof binds to a human TIGIT epitope consisting of TIGIT residues Q56, N58, E60, I68,

L73, H76 and I109.

In a further aspect, the present invention provides an isolated antibody or antigen binding fragment thereof that binds to human TIGIT and that does not compete with CD155 for TIGIT binding.

In some embodiments of the present invention, an antibody or antigen binding fragment that binds to human TIGIT and that does not compete with CD155 for TIGIT binding comprises HCDR1, HCDR2, HCDR3, LCDR1, LCDR2, and LCDR3, where HCDR1 contains or consists of SEQ ID NO: 280, HCDR2 contains or consists of SEQ ID NO: 281, HCDR3 contains or consists of SEQ ID NO: 282, and LCDR1 contains or consists of SEQ ID NO: 292, LCDR2 contains or consists of SEQ ID NO: 293, and LCDR3 contains or consists of SEQ ID NO: 294.

In certain such embodiments of the present invention, the heavy chain variable domain comprises or consists of an amino acid sequence shown as SEQ ID NO: 333 or an amino acid sequence exhibiting at least 90, 95, 97, 98, or 99% sequence identity thereto, and the light chain variable domain contains or consists of the amino acid sequence shown as SEQ ID NO: 334 or an amino acid sequence exhibiting at least 90, 95, 97, 98 or 99% sequence identity thereto.

In some preferred embodiments of the present invention, an antibody that binds to human TIGIT and that does not compete with CD155 for TIGIT binding comprises a heavy chain variable domain and a light chain variable domain, wherein HCDR1 contains SEQ ID NO: 353, HCDR2 contains SEQ ID NO: 353 NO: 354, HCDR3 contains SEQ ID NO: 355, LCDR1 contains SEQ ID NO: 356, LCDR2 contains SEQ ID NO: 357, LCDR3 contains SEQ ID NO: 358.

In some such embodiments of the present invention, the heavy chain variable domain may comprise the amino acid sequence shown as SEQ ID NO: 367, or an amino acid sequence exhibiting at least 90, 95, 97, 98, or 99% sequence identity, and the variable domain the light chain may comprise the amino acid sequence shown as SEQ ID NO: 368 or an amino acid sequence exhibiting at least 90, 95, 97, 98 or 99% sequence identity thereto.

In a further aspect, the present invention provides an isolated anti-TIGIT antibody or antigen binding fragment thereof that preferentially depletes TIGIT-expressing Treg cells, optionally wherein the antibody or antigen binding fragment is an antibody or antigen binding fragment according to the first aspect of the invention, for example, the antibody described in as an example in this document.

In a further aspect, the present invention provides an affinity variant of an antibody in accordance with other aspects of the invention, for example, the antibody exemplified herein.

In a further aspect, the present invention provides an isolated polynucleotide or combination of isolated polynucleotides encoding an antibody or antigen binding fragment, in accordance with any other aspect of the invention, for example, an antibody exemplified herein.

In a further aspect, the invention provides an isolated polynucleotide encoding the VH and/or VL domain of an anti-TIGIT antibody, wherein the polynucleotide comprises one or more sequences selected from the group consisting of SEQ ID NOs: 241-270, 335-342 and 369 -370.

In a further aspect, the present invention provides an expression vector comprising a polynucleotide or combination of polynucleotides of the invention operably linked to regulatory sequences that allow expression of an antigen binding polypeptide in a host cell or cell-free expression system.

In a further aspect, the present invention relates to a host cell or cell-free expression system containing an expression vector in accordance with the present invention.

In a further aspect, the present invention provides a method for producing a recombinant antibody or antigen binding fragment thereof, which comprises culturing a host cell or cell-free expression system of the present invention under conditions that allow expression of the antibody or antigen binding fragment and recovery of the expressed antibody or antigen binding fragment.

In a further aspect, the present invention provides a pharmaceutical composition comprising an antibody or antigen-binding fragment of the invention, such as an antibody exemplified herein, and at least one pharmaceutically acceptable carrier or excipient.

In a further aspect, the present invention relates to an antibody or antigen-binding

- 3 045372 a fragment according to this invention or a pharmaceutical composition according to this invention for use in therapy.

In a further aspect, the present invention provides an antibody or antigen binding fragment of the invention (eg, an antibody exemplified herein) or a pharmaceutical composition of the invention for use in a method of treating cancer.

In a further aspect, the invention provides an antibody or antigen binding fragment of the invention (eg, an antibody exemplified herein) or a pharmaceutical composition of the invention for use in a method of treating a viral infection, optionally a CMV infection.

In a further aspect, the present invention provides a method of treating cancer in a subject, comprising administering to the subject an effective amount of an antibody or antigen binding fragment of the present invention (e.g., an antibody exemplified herein) or a pharmaceutical composition of the present invention to the subject, thereby treating the cancer.

In a further aspect, there is provided a method of treating a viral infection in a subject, comprising administering to the subject an effective amount of an antibody or antigen binding fragment in accordance with the present invention or a pharmaceutical composition in accordance with the present invention for the purpose of treating the viral infection. In preferred embodiments of the present invention, the viral infection is a CMV infection.

In a further aspect, a method of promoting T cell activity is provided, comprising contacting a population of T cells with an antibody or antigen binding fragment according to the present invention. In certain embodiments of the present invention, the method stimulates the activity of αβ T cells. In certain embodiments of the present invention, the method stimulates the activity of γδ T cells. In certain embodiments of this invention, the method is performed in vitro. In certain embodiments of the present invention, the method is performed in vivo, for example, in a human subject.

In certain embodiments, the present invention provides a method of the present invention, or an antibody, or an antigen binding fragment thereof, or a pharmaceutical composition for use in a method of the present invention, wherein the method further comprises administering one or more additional therapeutic agents. In certain preferred embodiments of the present invention, one or more additional agents are selected from: a chemotherapy agent, an anti-PD1 antibody, an anti-PD-L1 antibody, an anti-41 BB antibody, an anti-OX40 antibody, an anti-GITR antibody, and an anti-ICOS antibodies.

In an additional aspect, a combination is provided comprising an anti-TIGIT antibody or an antigen-binding fragment thereof and one or more of a chemotherapeutic agent, an anti-PD1 antibody, an anti-PD-L1 antibody, an anti-41BB antibody, an anti-OX40 antibody, an anti-GITR antibody, and anti-ICOS antibodies. In an additional aspect, a combination according to this invention is provided for use in therapy. In an additional aspect, the combination of this invention is provided for use in a method of treating cancer or for use in a method of treating a viral infection. In a further aspect, a combination of the present invention is provided for use in a method of the present invention. In a preferred embodiment, the anti-TIGIT antibody or antigen binding fragment thereof is an antibody of the present invention or an antigen binding fragment thereof.

In all relevant aspects, it is preferred that any subject to be treated is human. In all relevant aspects, it is preferred that the cells (eg, T cells) contacted with the antibodies of this invention are human cells (eg, human T cells).

Unless technically incompatible or otherwise indicated, any described preferred embodiment of the invention may optionally be used in combination with one or more of all other preferred embodiments of the invention.

Brief description of graphic materials

Fig. 1. Table showing the complementarity determining region (CDR) sequences of the heavy chain variable domain (VH) of antibodies of the present invention.

Fig. 2. Table showing the CDR sequences of the light chain variable domain (VL) of the antibodies of the present invention.

Fig. 3. Table presenting the framework (FR) sequences of the heavy chain variable domain (VH) of the antibodies of the present invention.

Fig. 4. Table presenting the framework (FR) sequences of the light chain variable domain (VL) of the antibodies of the present invention.

Fig. 5. Table showing the amino acid sequences of the heavy chain variable domain (VH) and light chain variable domain (VL) of the antibodies of the present invention.

- 4 045372

Fig. 6. Table presenting the sequences of polynucleotides encoding the VH and VL domains of antibodies according to this invention.

Fig. 7. Graph showing the results of a competition assay between hCD155 and anti-TIGIT antibody for binding to Jurkat-hTIGIT.

Fig. 8(A). Graph showing the proportion of TIGIT-positive cells in defined T cell populations of PBMCs from 7 healthy human donors. 8(B). Graph showing the proportion of TIGIT-positive cells in different immune populations of PBMCs from 7 healthy human donors.

Fig. 9 Graph showing the results of anti-TIGIT antibody clone binding assay on Jurkat-hTIGIT.

Fig. 10(A and B). Graphs showing the results of anti-TIGIT antibody binding assays on primary CD8+ T cells from PBMCs from healthy individuals. 10(C). Graph showing results of anti-TIGIT antibody binding assay to primary memory CD8+ T cells and Tregs from PBMCs from healthy individuals.

Fig. 11. Graphs showing the results of anti-TIGIT antibody binding assay on primary CD8+ T cells PBMCs from healthy Javanese subjects.

Fig. 12. Graphs demonstrating the effect of anti-TIGIT antibodies in the CHO-TCR-CD155 and Jurkat-hTIGIT bioassays.

Fig. 13. Graphs demonstrating the effect of anti-TIGIT antibodies on increasing IFNg secretion in a functional assay on primary human CD8 T cells from healthy donors activated with CHO-TCR-CD155 cells.

Fig. 14. Histograms demonstrating the effect of anti-TIGIT antibodies on increasing IFNg secretion in a functional assay on primary human CD8 ⁺ TILs from ovarian ascites activated by CHO-TCR-CD155 cells.

Fig. 15(A). Graph showing the results of a competition assay between mouse CD155 and anti-TIGIT antibodies for binding to Jurkat-mTIGIT. 15(B). Graph showing the effect of anti-TIGIT antibody on increasing IFNg secretion in a functional assay on murine OT-1 T cells. 15(C). Graph showing the effect of anti-TIGIT antibody on increasing cytotoxicity in a functional assay on murine OT-1 T cells.

Fig. 16(A). Graph demonstrating the antitumor efficacy of anti-TIGIT antibody as monotherapy in the CT26 tumor model. 16(B and C). Graphs demonstrating the antitumor efficacy of anti-TIGIT antibody in combination with anti-PD1 in the CT26 tumor model.

Fig. 17(A). Graph demonstrating the isotype-dependent antitumor efficacy of anti-TIGIT antibody as monotherapy in the CT26 tumor model. 17(B). Graph demonstrating the isotype-dependent antitumor efficacy of anti-TIGIT antibody in combination with anti-PD1 in the CT26 tumor model.

Fig. 18(A and G). Graphs demonstrating modulation of the proportion of Treg cells in the total CD4+ T cell population in a CT26 tumor treated with anti-TIGIT antibody alone or in combination with anti-PD1 antibody. 18(B and H). Graphs showing modulation of the proportion of CD8+ T cells in the total CD45+ population in a CT26 tumor treated with anti-TIGIT antibody alone or in combination with antiPD1. 18(C and I). Graphs demonstrating modulation of CD8 ⁺ T cell/Treg ratio in CT26 tumor treated with anti-TIGIT antibody alone or in combination with anti-PD1 antibody. 18(D and J). Graph showing modulation of IFNg-secreting CD4+ T cells in a CT26 tumor treated with anti-TIGIT antibody alone or in combination with anti-PD1. 18(E). Graph showing modulation of IFNg-secreting CD8+ T cells in a CT26 tumor treated with anti-TIGIT antibody. 18(L and F). Graphs showing the ratio of IFNg/IL-10 secreting CD4+ T cells in a CT26 tumor treated with anti-TIGIT antibody alone or in combination with anti-PD1. 18(K). Graph showing modulation of IL-10-secreting CD4+ T cells in a CT26 tumor treated with anti-TIGIT antibody in combination with anti-PD1 antibody.

Fig. 19(A). Volcano plot showing the effect of anti-TIGIT antibody treatment to modulate gene expression in the CT26 tumor, obtained by NanoString analysis. 19(B). Boxplot showing modulation of cytotoxic score in CT26 tumor treated with anti-TIGIT antibody alone or in combination with αntu-PD1. 19(C). Boxplot showing the modulation of CD8+ T cell counts in a CT26 tumor treated with anti-TIGIT antibody alone or in combination with anti-PD1.

Fig. 20(A). Histograms showing the proportion of TIGIT+ CD4 ⁺ , CD8+ T cell and Treg populations in PBMCs from healthy human volunteers. 20(B). Graph demonstrating the in vitro cytotoxic effect of anti-TIGIT antibody on conventional CD4 ⁺ , CD8+ T cell and Treg populations in PBMCs from healthy human volunteers.

Fig. 21. Graph demonstrating the effect of ex-vivo cytotoxicity of anti-TIGIT antibody on conventional CD4 ⁺ , CD8+ T cell and Treg populations in a CT26 tumor.

Fig. 22(A). Graph showing results of anti-TIGIT antibody clone binding assay

- 5 045372 on Jurkat-hTIGIT cells. 22(B). Graph showing the results of anti-TIGIT antibody clone binding assays on primary CD8+ T cells from PBMCs from healthy individuals. 22(C). Graph showing the results of anti-TIGIT antibody clone binding assays on primary CD8+ T cells from PBMCs from cancer patients.

Fig. 23. Graph showing the results of a competition assay between human CD155 and anti-TIGIT antibody clones for binding to Jurkat-hTIGIT.

Fig. 24. Graph showing the functional characterization of the antagonist of a-TIGIT clones. 24(A). Graphs demonstrating the effect of anti-TIGIT antibodies in a functional assay using Jurkat-hTIGIT effector cells (luciferase reporter gene assay). 24(B). Graphs demonstrating the effect of anti-TIGIT antibodies in a functional assay measuring IFNg secretion by primary human CD8+ T cells from healthy volunteers. 24(C). Graph showing the effect of anti-TIGIT antibody clone 31282 in a functional assay measuring IFNg secretion by cancer patient CD3+ T cells from PBMCs. 24(D). Graph showing the effect of anti-TIGIT antibody clone 31282 in a functional assay measuring intracellular cytokine staining in cancer patient TILs or PBMCs.

Fig. 25. Cytotoxic activity of clone 31282a-TIGIT against CD4+ or CD8+ total memory T cells and Treg populations in PBMCs from a cancer patient.

Fig. 26. Graph showing the expression pattern of TIGIT on immune populations from cancer patients. 26(A). Frequency of TIGIT expression in immune populations of PBMCs and TILs of cancer patients. (B) Absolute quantification of TIGIT expression in immune populations from PBMCs and TILs of cancer patients.

Fig. 27(A). The structure of the Fab :TIGIT complex is demonstrated as a strip diagram. 27(B). Full link interface between clone 31282 and TIGIT. 27(C). Binding interface between clone 31282 and TIGIT showing bound residues.

Fig. 28. Analysis of competition between anti-TIGIT clones 31282 and 32959.

Fig. 29. Measurement of plasma concentrations of anti-TIGIT clone 31282 after intravenous administration of a single dose of 0.1 mg/kg (top row), 1 mg/kg (middle row), or 10 mg/kg (bottom row) in the Javanese monkey ( Cynomolgus). Left column: 31282 IgG1; right column 31282 IgG4.

Fig. 30. Graph showing the expression pattern of TIGIT on malignant and normal CD4+ T cell populations from a patient with Sézary syndrome. 30(A). Gating strategy for separating malignant and normal CD4+ T cells. 30(B). MFI for TIGIT staining on 2 different populations.

Fig. 31. Graph showing the expression pattern of TIGIT on malignant and normal B cell populations from a patient with CLL. 31(A). Gating strategy for separating malignant and normal B cells. 31(B). MFI for TIGIT staining on 2 different populations.

Fig. 32(A-C). Graph showing tumor growth curves in mice inoculated with EL4-mTIGIT tumors. 32(A). Median tumor growth curves. 32(B). Growth curves of individual tumors in mice treated with the hIgG1 isotype control antibody. 32(C). Growth curves of individual tumors in mice treated with a surrogate antagonist mouse a-TIGIT antibody (hIgG1). 32(D-F). Graph showing tumor growth curves in mice inoculated with EL4-GFP tumors. 32(D). Median tumor growth curves. 32(E). Growth curves of individual tumors in mice treated with the hIgG1 isotype control antibody. 32(F). Growth curves of individual tumors in mice treated with the surrogate antagonist a-TIGIT (hIgG1).

Fig. 33(A-D). Graphs showing tumor growth curves in mice inoculated with CT26 tumors. 33(A). Growth curves of median and individual tumors and for mice treated with anti-TIGIT antibodies and anti-4-1BB antibodies. 33(B). Growth curves of median and individual tumors and for mice treated with anti-TIGIT antibodies and anti-OX-40 antibodies. 33(C). Growth curves of median and individual tumors and for mice treated with anti-TIGIT antibodies and anti-GITR antibodies. 33(D). Growth curves of median and individual tumors and for mice treated with anti-TIGIT antibodies and anti-ICOS antibodies.

Fig. 34. Graphs showing the effect of anti-TIGIT antibodies on γδ T cells. 34(A). Median proportion of TIGIT-positive cells and MFT TIGIT signal in Vδ2' γδ T-cell populations of PBMCs from CMV-positive and -negative human donors. 34(B). Graph demonstrating the activity of anti-TIGIT Ab in increasing IFNg secretion in a functional assay on isolated primary human Vδ1 ⁺ γδ T cells. 34(C). Graph demonstrating the activity of anti-TIGIT Ab in increasing IFNg secretion in a functional assay of total PBMCs.

Detailed Description of the Invention

As used herein, the term immunoglobulin includes a polypeptide having a combination of two heavy and two light chains, regardless of whether it has any corresponding specific immunoreactivity. Antibodies refer to those assemblies that have significant known specific immunoreactive activity against the antigen of interest (eg, TIGIT). The term anti-TIGIT antibodies or anti-TIGIT antibodies are used herein to refer to antibodies that exhibit immunological specificity for the TIGIT protein. Antibodies and immunoglobulins contain light and heavy chains, with or without covalent bonds between them. The basic structures of immunoglobulin in vertebrate systems are relatively well understood.

The general term immunoglobulin includes five different classes of antibodies that can be distinguished biochemically. Although all five classes of antibodies are included within the scope of this invention, the following discussion will generally be directed to the IgG class of immunoglobulin molecules. As for IgG, immunoglobulins contain two identical polypeptide light chains with a molecular weight of approximately 23,000 daltons and two identical heavy chains with a molecular weight of 53,000-70,000. The four chains are linked by disulfide bonds in a Y configuration in which the light chains enclose the heavy chains starting at the Y mouth and continuing through the variable region.

Antibody light chains are classified as kappa or lambda (κ, λ). Each heavy chain class can be linked to a kappa or lambda light chain. In general, the light and heavy chains are covalently linked to each other, and the tail portions of the two heavy chains are linked to each other by covalent disulfide bonds or non-covalent bonds when immunoglobulins are produced by B cells or genetically engineered host cells. In the heavy chain, the amino acid sequences run from the N-terminus at the branched ends of the Y-configuration to the C-terminus at the bottom of each chain. Those skilled in the art will appreciate that heavy chains are classified as gamma, mu, alpha, delta, or epsilon (γ, μ, α, δ, ε), with some subclasses among them (eg, γ1-γ4). It is the nature of this chain that determines the class of the antibody as IgG, IgM, IgA, IgD or IgE, respectively. Immunoglobulin subclasses (isotypes), such as IgG1, IgG2, IgG3, IgG4, IgA1, etc., are well characterized and are known to confer functional specialization. Modified versions of each of these classes and isotypes will be readily apparent to those skilled in the art given the disclosure herein and are accordingly within the scope of the present invention.

As stated above, the variable region of an antibody allows the antibody to selectively recognize and specifically bind epitopes on antigens. That is, the VL domain and the VH domain of an antibody combine to form a variable region that defines a three-dimensional antigen binding site. This quaternary antibody structure forms an antigen binding site present at the end of each Y arm. More specifically, the antigen binding site is defined by three complementarity determining regions (CDRs) on each of the VH and VL chains.

As used herein, the terms TIGIT protein or TIGIT antigen or TIGIT are used interchangeably and refer to the human T cell immunoreceptor (GenBank accession number: NM173799) that binds the poliovirus receptor (PVR - also known as CD155). TIGIT is also known as VSIG9, VSTM3 or WUCAM. Reference to TIGIT includes native human TIGIT protein naturally expressed in the human host and/or on the surface of cultured human cell lines, as well as recombinant forms and fragments thereof, as well as naturally occurring mutant forms.

As used herein, the term binding site includes a region of a polypeptide that is responsible for selective binding to a target antigen of interest (eg, TIGIT). Binding domains contain at least one binding site. Exemplary binding domains include an antibody variable domain. Antibody molecules of the present invention may contain one binding site or multiple (eg, two, three or four) binding sites.

As used herein, the term derived from a specified protein (eg, a TIGIT antibody or an antigen binding fragment thereof) refers to the origin of the polypeptide. In one embodiment of the invention, the polypeptide or amino acid sequence that is derived from a particular parent polypeptide is a CDR sequence or a sequence related thereto. In one embodiment of the invention, the amino acid sequence that is derived from a particular parent polypeptide is not contiguous. For example, in one embodiment of the invention, one, two, three, four, five or six CDRs are derived from the parent antibody. In one embodiment of the invention, a polypeptide or amino acid sequence that is derived from a particular parent polypeptide or amino acid sequence has an amino acid sequence that is substantially identical to the sequence of the parent sequence or a portion thereof, wherein the portion consists of at least 3-5 amino acids, at least at least 5-10 amino acids, at least 10-20 amino acids, at least 20-30 amino acids, or at least 30-50 amino acids, or which can be identified by others skilled in the art as being derived from the starting sequence. In one embodiment of the invention, one or more CDR sequences derived from the parent antibody are modified to produce variant CDR sequences, such as affinity

- 7,045,372 variants, with variant CDR sequences retaining TIGIT binding activity.

As used herein, the term conservative amino acid substitution means the replacement of an amino acid residue with an amino acid residue having a similar side chain. Families of amino acid residues having similar side chains have been defined in the art, including basic side chains (e.g., lysine, arginine, histidine), acidic side chains (e.g., aspartic acid, glutamic acid), uncharged polar side chains (e.g., glycine, asparagine, glutamine, serine, threonine, tyrosine, cysteine), non-polar side chains (e.g. alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine, tryptophan), beta-branched side chains (e.g. threonine, valine, isoleucine) and aromatic side chains (eg, tyrosine, phenylalanine, tryptophan, histidine). Thus, an optional amino acid residue in an immunoglobulin polypeptide can be replaced by another amino acid residue from the same side chain family. In another embodiment of the invention, the amino acid chain may be replaced by a structurally similar chain that differs in the order and/or composition of the side chain family members.

As used herein, the term heavy chain portion includes amino acid sequences derived from immunoglobulin heavy chain constant domains. The heavy chain portion-containing polypeptide contains at least one of a CH1 domain, a hinge (eg, upper, middle, and/or lower hinge domain) domain, a CH2 domain, a CH3 domain, or a variant or fragment thereof. In one embodiment, an antibody or antigen binding fragment of the invention may comprise an Fc portion of an immunoglobulin heavy chain (eg, a hinge portion, a CH2 domain, and a CH3 domain). In another embodiment, an antibody or antigen binding fragment of the invention may lack at least a portion of a constant domain (eg, all or part of the CH2 domain). In some embodiments of the present invention, at least one and preferably all of the constant domains are derived from a human immunoglobulin heavy chain. For example, in one preferred embodiment of the invention, the heavy chain portion comprises a fully human hinge domain. In other preferred embodiments of the present invention, the heavy chain portion comprises a fully human Fc portion (eg, hinge, CH2 and CH3 domain sequences from human immunoglobulin).

In certain embodiments of the present invention, the constituent heavy chain constant domains are derived from different immunoglobulin molecules. For example, the heavy chain portion of a polypeptide may contain a CH2 domain derived from an IgG1 molecule and a hinge region derived from an IgG3 or IgG4 molecule. In other embodiments of the present invention, constant domains are chimeric domains containing parts of different immunoglobulin molecules. For example, the hinge may comprise a first portion of an IgG1 molecule and a second portion of an IgG3 or IgG4 molecule. As stated above, one skilled in the art will appreciate that the constant domains of the heavy chain portion can be modified such that they differ in amino acid sequence from the naturally occurring (wild type) immunoglobulin molecule. That is, the polypeptides of the present invention described herein may contain changes or modifications to one or more heavy chain constant domains (CH1, hinge, CH2 or CH3) and/or a light chain constant region (CL) domain. Typical modifications include additions, deletions or substitutions of one or more amino acids in one or more domains.

As used herein, the terms variable region and variable domain are used interchangeably and have the same meaning. The term variable refers to the fact that certain parts of the VH and VL variable domains vary widely in sequence among antibodies and are used in the binding and specificity of each particular antibody for its target antigen. However, variability is not evenly distributed across antibody variable domains. It is concentrated in three segments called hypervariable loops in each of the VL domain and the VH domain, which form part of the antigen binding site. The first, second and third hypervariable loops of the light chain domain V _a m _b d _a are referred to herein as L1(λ), L2(λ) and L3(λ) and can be defined as containing residues 24-33 (L1(λ) , consisting of 9, 10 or 11 amino acid residues), 49-53 (L2(λ), consisting of 3 residues) and 90-96 (L3(λ), consisting of 5 residues) in the VL domain (Morea et al., Methods 20, 267-279, 2000). The first, second and third hypervariable loops of the V _kanna light chain domain are referred to herein as L1(k), L2(k) and L3(k) and can be defined as containing residues 25-33 (L1(k), consisting of 6 , 7, 8, 11, 12 or 13 residues), 49-53 (L2(k), consisting of 3 residues) and 90-97 (L3(k), consisting of 6 residues) in the VL domain (Morea et al. , Methods 20, 267-279, 2000). The first, second and third hypervariable loops of the VH domain are referred to herein as H1, H2 and H3 and can be defined as containing residues 25-33 (H1, consisting of 7, 8 or 9 residues), 52-56 (H2, consisting of 3 or 4 residues) and 91-105 (H3, highly variable in length) in the VH domain (Morea et al., Methods 20, 267-279, 2000).

Unless otherwise specified, the terms L1, L2 and L3 respectively refer to the first, second and third hypervariable loops of the VL domain and cover hypervariable loops derived from both the V _canna and V _lambda isotypes. The terms H1, H2 and H3 respectively refer to the first, second and third hypervariable loops of the VH domain and encompass hypervariable loops derived from any of the known heavy chain isotypes, including γ, ε, δ, α or μ.

The hypervariable loops L1, L2, L3, H1, H2 and H3 may each form part of a complementarity determining region or CDR, as defined below. The terms hypervariable loop and complementarity-determining region are not strictly synonymous, since hypervariable loops (HV) are defined on the basis of structure, whereas complementarity-determining regions (CDRs) are defined on the basis of sequence variability (Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD, 1991) and HV and CDR limits may differ in some VH and VL domains.

The CDRs of the VL and VH domains can generally be defined as containing the following amino acids: residues 24-34 (LCDR1), 50-56 (LCDR2) and 89-97 (LCDR3) in the light chain variable domain, and residues 3135 or 31-35b (HCDR1) , 50-65 (HCDR2) and 95-102 (HCDR3) in the heavy chain variable domain; (Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD, 1991). Thus, HVs may be contained within the corresponding CDRs, and references herein to the hypervariable loops of the VH and VL domains should be interpreted to also include the corresponding CDRs, and vice versa, unless otherwise indicated.

The more highly conserved portions of the variable domains are referred to as the framework region (FR), as defined below. Each variable domain of the native heavy and light chains contains four FRs (FR1, FR2, FR3 and FR4, respectively), largely adopting a β-sheet configuration, connected by three hypervariable loops. The hypervariable loops in each chain are held together in close proximity to the FR and, together with the hypervariable loops from the other chain, contribute to the formation of the antigen-binding site of antibodies. Structural analysis of antibodies has revealed a relationship between the sequence and the shape of the binding site formed by the complementarity determining regions (Chothia et al., J. Mol. Biol. 227, 799-817, 1992; Tramontano et al., J. Mol. Biol. 215, 175 -182 (1990) Despite their high sequence variability, five of the six loops adopt only a small repertoire of main chain conformations, called canonical structures. These conformations are determined, firstly, by the length of the loops and, secondly, by the presence of key residues at certain loop positions and framework regions that determine conformation through their packing, hydrogen bonding, or ability to adopt unusual backbone conformations.

As used herein, the term CDR or complementarity determining region refers to non-contiguous antigen combining sites found in the variable region of both the heavy and light chain polypeptides. These specific regions have been described by Kabat et al., J. Biol. Chem. 252, 6609-6616, 1977, by Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD, 1991, by Chothia et al., J. Mol. Biol. 196, 901-917, 1987, and by MacCallum et al., J. Mol. Biol. 262, 732-745, 1996, where the definitions include overlapping or subsets of amino acid residues when compared with each other. Amino acid residues that span the CDRs, as defined in each of the above references, are provided for comparison. Preferably, the term CDR is a CDR as determined by Kabat based on sequence comparisons.

Table 1

CDR Definitions

CDR Definitions. Kabat ¹ XoTna(Chotia) ² MacCallum ³ V _H CDR1 31-35 26-32 30-35 V _H CDR2 50-65 53-55 47-58 V _H CDR3 95-102 96-101 93-101 V _L CDR1 24-34 26-32 30-36 V _L CDR2 50-56 50-52 46-55 V _L CDR3 89-97 91-96 89-96

Residue numbering follows the nomenclature of Kabat et al., supra.

^Residue numbering follows the nomenclature of Chothia et al., above.

³ Residue numbering follows the nomenclature of MacCallum et al., supra.

As used herein, the term framework region or FR region includes amino acid residues that are part of the variable region but are not part of the CDR (eg, using Kabat's definition of CDR). Therefore, the variable region framework is about 100-120 amino acids long, but includes only those amino acids that are located beyond

- 9 045372 outside the CDR. For a specific example of a heavy chain variable domain and for a CDR as defined by Kabat et al., framework region 1 corresponds to a variable region domain spanning amino acids 1-30; framework region 2 corresponds to the variable region domain spanning amino acids 36-49; framework region 3 corresponds to the variable region domain spanning amino acids 66-94, and framework region 4 corresponds to the variable region domain from amino acid 103 to the end of the variable region. The light chain framework regions are similarly separated by each of the light chain variable region CDRs. Similarly, using the CDR definition of Chothia et al. or McCallum et al. the boundaries of the frame region are separated by the corresponding ends of the CDRs, as described above. In preferred embodiments of the present invention, the CDRs are as defined by Kabat.

In naturally occurring antibodies, the six CDRs present on each monomeric antibody are short, non-contiguous sequences of amino acids that are specifically positioned to form an antigen-binding site as the antibody takes on a three-dimensional configuration in an aqueous environment. The remainder of the variable domains of the heavy and light chains exhibit less intermolecular variability in the amino acid sequence and are called framework regions. The framework regions largely adopt a βsheet conformation, and the CDRs form loops that connect, and in some cases are part of, the βsheet structure. Thus, these framework regions act to form a framework that positions the six CDRs in the correct orientation through interchain non-covalent interactions. The antigen binding site formed by the arranged CDRs defines a surface complementary to the epitope on the immunoreactive antigen. This complementary surface facilitates non-covalent binding of the antibody to the epitope of the immunoreactive antigen. The position of the CDR can be easily identified by one skilled in the art.

As used herein, the term fragment refers to a portion or portion of an antibody or antibody chain containing fewer amino acid residues than the whole or complete antibody or antibody chain. The term antigen-binding fragment refers to a polypeptide fragment of an immunoglobulin or antibody that binds an antigen or competes with an intact antibody (ie, the intact antibody from which it was derived) for antigen binding (ie, specific binding to TIGIT). As used herein, the term antibody fragment includes antigen binding antibody fragments, e.g., antibody light chain variable domain (VL), antibody heavy chain variable domain (VH), single chain antibody (scFv), F(ab')2 fragment, Fab fragment, Fd fragment, Fv fragment and single domain antibody (DAb) fragment. The fragments can be produced, for example, by chemical or enzymatic treatment of an intact or complete antibody or antibody chain, or by recombinant methods.

As used herein, the term valency refers to the number of potential binding sites for a target in a polypeptide. Each target binding site specifically binds one target molecule or a specific site on a target molecule. When a polypeptide contains more than one target binding site, each target binding site may specifically bind the same or different molecules (eg, may bind to different ligands or different antigens or different epitopes on the same antigen). The binding molecules of this invention have at least one TIGIT-specific binding site.

As used herein, the term specificity refers to the ability to bind (eg, immunoreact to) a given target, eg, TIGIT. The polypeptide may be monospecific and contain one or more binding sites that specifically bind a target, or the polypeptide may be multispecific and contain two or more binding sites that specifically bind the same or different targets. In one embodiment of the invention, the antibody of this invention is specific for more than one target. For example, in one embodiment of the invention, the multispecific binding molecule of the invention binds TIGIT and a second target molecule. In this context, the second target molecule is a molecule other than TIGIT.

As used herein, the term synthetic in relation to polypeptides includes polypeptides that contain an amino acid sequence that does not occur in nature. For example, non-naturally occurring polypeptides that are modified forms of naturally occurring polypeptides (e.g., containing a mutation such as an addition, substitution, or deletion) or that contain a first amino acid sequence (which may or may not occur in nature) that is linked in a linear sequence of amino acids with a second amino acid sequence (which may or may not occur in nature) to which it is not naturally related.

As used herein, the term engineered includes the manipulation of nucleic acid or polypeptide molecules by synthetic methods (e.g., recombinant methods, in vitro peptide synthesis, enzymatic or chemical combination of peptides, or

- 10 045372 some combination of these methods). Preferably, the antibodies of the present invention have been engineered to improve one or more properties such as antigen binding, stability/half-life, or effector function.

As used herein, the term modified antibody includes synthetic forms of antibodies that are modified such that they do not occur naturally, for example, antibodies that contain at least two portions of a heavy chain, but not two complete heavy chains (such as deletion antibodies domain or mini-body); multispecific forms of antibodies (eg, bispecific, trispecific, etc.) modified to bind to two or more different antigens or to different epitopes on a single antigen; heavy chain molecules linked to scFv molecules and the like. ScFv molecules are known in the art and are described, for example, in US Pat. No. 5,892,019. Additionally, the term modified antibody includes multivalent forms of antibodies (e.g., trivalent, quadrivalent, etc. antibodies that bind to three or more copies of the same antigen). In another embodiment, the modified antibody of the present invention is a fusion protein comprising at least one heavy chain portion lacking a CH2 domain and comprising a polypeptide binding domain comprising the binding portion of one member of a pair of receptor ligands.

The term modified antibody may also be used herein to refer to amino acid sequence variants of the TIGIT antibody of the present invention. One skilled in the art will appreciate that the TIGIT antibody of this invention can be modified to produce a variant TIGIT antibody that varies in amino acid sequence compared to the TIGIT antibody from which it was derived. For example, nucleotide or amino acid substitutions may be made resulting in conservative substitutions or changes in non-essential amino acid residues (eg, CDRs and/or framework residues). Amino acid substitutions may include replacing one or more amino acids with a natural or unnatural amino acid.

Antibody fragments include part of a full-length antibody, typically its antigen-binding or variable domain. Examples of antigen binding antibody fragments include Fab, Fab', F(ab')2, bispecific Fab and Fv fragments, diabodies, linear antibodies, single chain antibody molecules, single chain variable fragment (scFv) and multispecific antibodies formed from antibody fragments (see Holliger and Hudson, Nature Biotechnol. 23: 1126-1136, 2005, the contents of which are incorporated herein by reference).

As used herein, the term affinity variant refers to an antibody variant that exhibits one or more changes in amino acid sequence compared to the reference TIGIT antibody of the present invention, wherein the affinity variant exhibits an altered affinity for TIGIT compared to the reference antibody. Preferably, the affinity variant will exhibit improved affinity for TIGIT compared to the reference TIGIT antibody. The improvement may be manifested in a lower KD for TIGIT or a slower dissociation rate for TIGIT. Affinity variants typically exhibit one or more changes in amino acid sequence in the CDR compared to the reference TIGIT antibody. Such substitutions may result in the replacement of the original amino acid present at a given position in the CDR with another amino acid residue, which may be a naturally occurring amino acid residue or a non-naturally occurring amino acid residue. Amino acid substitutions may be conservative or non-conservative.

As used herein, the term affinity or binding affinity is to be understood in its ordinary meaning in the art in the context of antibody binding, and reflects the strength and/or stability of binding between an antigen and a binding site on the antibody or antigen-binding fragment thereof.

The anti-TIGIT antibodies provided herein have high binding affinity to human TIGIT. Binding affinity for TIGIT can be assessed using standard techniques known to those skilled in the art.

Binding affinity can also be expressed as the dissociation constant for a particular antibody or KD. The lower the KD value, the stronger the binding interaction between the antibody and its target antigen. In one embodiment of the invention, the binding affinity of a Fab clone containing a particular VH/VL pairing can be assessed using methods known in the art, for example, using the ForteBio™ system, using MSD solution equilibrium titration (SET), or with using surface plasmon resonance, for example using the Biacore™ system, as described in the accompanying examples. Fab fragments of antibodies of this invention typically exhibit a KD for TIGIT, measured using ForteBio™, in the range of 1x10'1° to ^5x10'8 M, optionally from ^7x10'10 to ^4x10'8 M. The KD within this range can be taken as an indication that Fab and the corresponding divalent mAb exhibit high binding affinity to hTIGIT. Bivalent mAbs containing two Fabs that (individually) exhibit a KD for hTIGIT within these ranges are also taken to demonstrate high binding affinity to hTIGIT. An MSD K _D value in the range from 1x10 ^-11 to 5x10 ^-9 , optionally from 2x10 ^-11 to 1x10 ⁹ , can be considered an indicator of high binding affinity to hTIGIT. Fab fragments of antibodies of the present invention typically exhibit a KD for TIGIT, measured using Biacore™, in the range of 1x10 ^-10 M to 1x10 ^-9 M, optionally 1x10 ^-10 to 7x10 ¹⁰ M, optionally 2x10 ^-10 to 7x10 ^{- 10} M. KD within this range can be taken as an indication that the Fab and the corresponding divalent mAb exhibit high binding affinity to hTIGIT.

Binding affinity to human TIGIT can also be assessed using a cell-based system, as described in the accompanying examples, in which the mAb is tested for binding to mammalian cells (cell lines or ex vivo cells that express TIGIT), for example, using ELISA or flow cytometry. High affinity for TIGIT may be indicated, for example, by an EC ₅₀ of no more than 0.5 nM using a flow cytometry technique (e.g., FACS) such as described in Example 10. In certain embodiments of the present invention, antibodies of the present invention exhibit an EC ₅₀ cellular binding no more than 0.5 nM, optionally no more than 0.2 nM. The cell-based affinity determination, expressed as EC ₅₀ , is preferably determined using hTIGIT-expressing Jurkat cells or primary CD8 T cells from human peripheral blood mononuclear cells (PBMCs).

As used herein, Tregs, or simply Tregs, refer to regulatory CD4+ T cells, that is, T cells that reduce the effector function(s) of conventional T cells (CD8 or CD4 T cells). Tregs can be identified according to methods known in the art, for example, using flow cytometry to identify CD4 cells expressing high levels of CD25 and low levels or absence of CD127.

As summarized above, the invention relates, at least in part, to antibodies and antigen-binding fragments thereof that bind to TIGIT. The properties and characteristics of TIGIT antibodies and antibody fragments in accordance with the present invention will now be described in more detail.

Anti-TIGIT antibodies

In one aspect, the present invention provides an isolated antibody or antigen binding fragment thereof that binds to human TIGIT, and which contains a heavy chain variable domain comprising heavy chain CDR1 (HCDR1), heavy chain CDR2 (HCDR2) and heavy chain CDR3 (HCDR3), selected from the HCDR1, HCDR2 and HCDR3 sequences shown in FIG. 1, and which further comprises a light chain variable domain comprising light chain CDR1 (LCDR1), light chain CDR2 (LCDR2) and light chain CDR3 (LCDR3) selected from the sequences LCDR1, LCDR2 and LCDR3 shown in FIG. 2. Thus, the invention relates to an isolated antibody or antigen binding fragment thereof that binds to human TIGIT and which contains a heavy chain variable domain comprising heavy chain CDR1 (HCDR1), heavy chain CDR2 (HCDR2) and heavy chain CDR3 (HCDR3), Where:

(i) HCDR1 is selected from the group consisting of SEQ ID NOs: 1, 4, 7, 10, 13, 16, 19, 22, 25, 28, 31, 34, 37, 40, 43, 271, 274 and 277;

(ii) HCDR2 is selected from the group consisting of SEQ ID NOs: 2, 5, 8, 11, 14, 17, 20, 23, 26, 29, 32, 35, 38, 41, 44, 272, 275 and 278;

(iii) HCDR3 is selected from the group consisting of SEQ ID NOs: 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 33, 36, 39, 42, 45, 273, 276 and 279;

and which further comprises a light chain variable domain comprising light chain CDR1 (LCDR1), light chain CDR2 (LCDR2) and light chain CDR3 (LCDR3), wherein (iv) LCDR1 is selected from the group consisting of SEQ ID NO: 46, 49, 52, 55, 58, 61, 64, 67, 70, 73, 76, 79, 82, 85, 88, 283, 286 and 289;

(v) LCDR2 is selected from the group consisting of SEQ ID NOs: 47, 50, 53, 56, 59, 62, 65, 68, 71, 74, 77, 80, 83, 86, 89, 284, 287 and 290; and (vi) LCDR3 is selected from the group consisting of SEQ ID NOs: 48, 51, 54, 57, 60, 63, 66, 69, 72, 75, 78, 81, 84, 87, 90, 285, 288 and 291 .

Any given anti-TIGIT antibody or antigen-binding fragment thereof containing a VH domain paired with a VL domain to form an antigen binding site (human TIGIT) will contain a combination of 6 CDRs: variable heavy chain CDR3 (HCDR3), variable heavy chain CDR2 ( HCDR2), variable heavy chain CDR1 (HCDR1), variable light chain CDR3 (LCDR3), variable light chain CDR2 (LCDR2) and variable light chain CDR1 (LCDR1). Although many different combinations of the 6 CDRs are conceivable, selected from the groups of CDR sequences listed above, and within the scope of the invention, certain combinations of the 6 CDRs are particularly preferred; they are native combinations within a single mAb that exhibit high binding affinity to human TIGIT. In some embodiments of the present invention, the antibody or antigen binding fragment comprises a combination of HCDR1, HCDR2, HCDR3, LCDR1, LCDR2, and LCDR3, wherein the combination is selected from the group of combinations formed by HCDR from

- 12 045372 of each antibody in Fig. 1 taken from the LCDR from the corresponding antibody in FIG. 2.

That is, in certain embodiments of the present invention, the antibody or antigen binding fragment comprises a combination of HCDR1, HCDR2, HCDR3, LCDR1, LCDR2, and LCDR3, wherein the combination is selected from the group consisting of:

(i) HCDR1 containing SEQ ID NO: 1, HCDR2 containing SEQ ID NO: 2, HCDR3 containing SEQ ID NO: 3, LCDR1 containing SEQ ID NO: 46, LCDR2 containing SEQ ID NO: 47, and LCDR3 containing SEQ ID NO: 48;

(ii) HCDR1 containing SEQ ID NO: 4, HCDR2 containing SEQ ID NO: 5, HCDR3 containing SEQ ID NO: 6, LCDR1 containing SEQ ID NO: 49, LCDR2 containing SEQ ID NO: 50, and LCDR3 containing SEQ ID NO: 51;

(iii) HCDR1 containing SEQ ID NO: 7, HCDR2 containing SEQ ID NO: 8, HCDR3 containing SEQ ID NO: 9, LCDR1 containing SEQ ID NO: 52, LCDR2 containing SEQ ID NO: 53, and LCDR3 containing SEQ ID NO: 54;

(iv) HCDR1 containing SEQ ID NO: 10, HCDR2 containing SEQ ID NO: 11, HCDR3 containing SEQ ID NO: 12, LCDR1 containing SEQ ID NO: 55, LCDR2 containing SEQ ID NO: 56, and LCDR3 containing SEQ ID NO: 57;

(v) HCDR1 containing SEQ ID NO: 13, HCDR2 containing SEQ ID NO: 14, HCDR3 containing SEQ ID NO: 15, LCDR1 containing SEQ ID NO: 58, LCDR2 containing SEQ ID NO: 59, and LCDR3 containing SEQ ID NO: 60;

(vi) HCDR1 containing SEQ ID NO: 16, HCDR2 containing SEQ ID NO: 17, HCDR3 containing SEQ ID NO: 18, LCDR1 containing SEQ ID NO: 61, LCDR2 containing SEQ ID NO: 62, and LCDR3 containing SEQ ID NO: 63;

(vii) HCDR1 containing SEQ ID NO: 19, HCDR2 containing SEQ ID NO: 20, HCDR3 containing SEQ ID NO: 21, LCDR1 containing SEQ ID NO: 64, LCDR2 containing SEQ ID NO: 65, and LCDR3 containing SEQ ID NO: 66;

(viii) HCDR1 containing SEQ ID NO: 22, HCDR2 containing SEQ ID NO: 23, HCDR3 containing SEQ ID NO: 24, LCDR1 containing SEQ ID NO: 67, LCDR2 containing SEQ ID NO: 68, and LCDR3 containing SEQ ID NO: 69;

(ix) HCDR1 containing SEQ ID NO: 25, HCDR2 containing SEQ ID NO: 26, HCDR3 containing SEQ ID NO: 27, LCDR1 containing SEQ ID NO: 70, LCDR2 containing SEQ ID NO: 71, and LCDR3 containing SEQ ID NO: 72;

(x) HCDR1 containing SEQ ID NO: 28, HCDR2 containing SEQ ID NO: 29, HCDR3 containing SEQ ID NO: 30, LCDR1 containing SEQ ID NO: 73, LCDR2 containing SEQ ID NO: 74, and LCDR3 containing SEQ ID NO: 75;

(xi) HCDR1 containing SEQ ID NO: 31, HCDR2 containing SEQ ID NO: 32, HCDR3 containing SEQ ID NO: 33, LCDR1 containing SEQ ID NO: 76, LCDR2 containing SEQ ID NO: 77, and LCDR3 containing SEQ ID NO: 78;

(xii) HCDR1 containing SEQ ID NO: 34, HCDR2 containing SEQ ID NO: 35, HCDR3 containing SEQ ID NO: 36, LCDR1 containing SEQ ID NO: 79, LCDR2 containing SEQ ID NO: 80, and LCDR3 containing SEQ ID NO: 81;

(xiii) HCDR1 containing SEQ ID NO: 37, HCDR2 containing SEQ ID NO: 38, HCDR3 containing SEQ ID NO: 39, LCDR1 containing SEQ ID NO: 82, LCDR2 containing SEQ ID NO: 83, and LCDR3 containing SEQ ID NO: 84;

(xiv) HCDR1 containing SEQ ID NO: 40, HCDR2 containing SEQ ID NO: 41, HCDR3 containing SEQ ID NO: 42, LCDR1 containing SEQ ID NO: 85, LCDR2 containing SEQ ID NO: 86, and LCDR3 containing SEQ ID NO: 87;

(xv) HCDR1 containing SEQ ID NO: 43, HCDR2 containing SEQ ID NO: 44, HCDR3 containing SEQ ID NO: 45, LCDR1 containing SEQ ID NO: 88, LCDR2 containing SEQ ID NO: 89, and LCDR3 containing SEQ ID NO: 90;

(xvi) HCDR1 containing SEQ ID NO: 271, HCDR2 containing SEQ ID NO: 272, HCDR3 containing SEQ ID NO: 273, LCDR1 containing SEQ ID NO: 283, LCDR2 containing SEQ ID NO: 284, and LCDR3 containing SEQ ID NO: 285;

(xvii) HCDR1 containing SEQ ID NO: 274, HCDR2 containing SEQ ID NO: 275, HCDR3 containing SEQ ID NO: 276, LCDR1 containing SEQ ID NO: 286, LCDR2 containing SEQ ID NO: 287, and LCDR3 containing SEQ ID NO: 288;

(xviii) HCDR1 containing SEQ ID NO: 277, HCDR2 containing SEQ ID NO: 278, HCDR3 containing SEQ ID NO: 279, LCDR1 containing SEQ ID NO: 289, LCDR2 containing SEQ ID NO: 290, and LCDR3 containing SEQ ID NO: 291.

In some embodiments of the present invention, the antibody or antigen binding fragment comprises a heavy chain variable domain having an amino acid sequence selected from the group consisting of: SEQ ID NO: 211, 213, 215, 217, 219, 221, 223, 225, 227, 229, 231, 233, 235, 237, 239, 327, 329 and 331 and amino acid sequences demonstrating at least 90, 95, 97, 98 or 99% sequence identity thereto; and optionally contain a light chain variable domain having an amino acid sequence selected from the group consisting of the amino acid sequences SEQ ID NO: 212, 214, 216, 218, 220, 222, 224, 226,

228, 230, 232, 234, 236, 238, 240, 328, 330 and 332 and amino acid sequences having at least 90, 95, 97, 98 or 99% sequence identity with it.

While all possible pairs of VH domains and VL domains selected from the groups of VH and VL domain sequences listed above are valid, and within the scope of the invention, certain combinations of VH and VL are particularly preferred; they are native combinations within a single mAb that exhibit high binding affinity to human TIGIT.

In some embodiments of the present invention, the antibody or antigen binding fragment comprises a combination of a heavy chain variable domain and a light chain variable domain, the combination selected from the group of combinations formed by the VH of each antibody in FIG. 5, or an amino acid sequence exhibiting at least 90, 95, 97, 98, or 99% sequence identity to VL from the same antibody in FIG. 5, or an amino acid sequence demonstrating at least 90, 95, 97, 98, or 99% sequence identity. In some embodiments of the present invention, the antibody or antigen binding fragment comprises a combination of a heavy chain variable domain and a light chain variable domain, wherein the combination is selected from the group consisting of:

(i) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 211 and a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 212;

(ii) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 213 and a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 214;

(iii) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 215 and a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 216;

(iv) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 217 and a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 218;

(v) a heavy chain variable domain containing the amino acid sequence of SEQ ID NO: 219 and a light chain variable domain containing the amino acid sequence of SEQ ID NO: 220;

(vi) a heavy chain variable domain containing the amino acid sequence of SEQ ID NO: 221 and a light chain variable domain containing the amino acid sequence of SEQ ID NO: 222;

(vii) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 223 and a light chain variable domain containing the amino acid sequence of SEQ ID NO: 224;

(viii) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 225 and a light chain variable domain containing the amino acid sequence of SEQ ID NO: 226;

(ix) a heavy chain variable domain containing the amino acid sequence of SEQ ID NO: 227 and a light chain variable domain containing the amino acid sequence of SEQ ID NO: 228;

(x) a heavy chain variable domain containing the amino acid sequence of SEQ ID NO: 229 and a light chain variable domain containing the amino acid sequence of SEQ ID NO: 230;

(xi) a heavy chain variable domain containing the amino acid sequence of SEQ ID NO: 231 and a light chain variable domain containing the amino acid sequence of SEQ ID NO: 232;

(xii) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 233 and a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 234;

(xiii) a heavy chain variable domain containing the amino acid sequence of SEQ ID NO: 235 and a light chain variable domain containing the amino acid sequence of SEQ ID NO: 236;

(xiv) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 237 and a light chain variable domain containing the amino acid sequence of SEQ ID NO: 238;

(xv) a heavy chain variable domain containing the amino acid sequence of SEQ ID NO: 239, and a light chain variable domain containing the amino acid sequence

- 14 045372

SEQ ID NO: 240;

(xvi) a heavy chain variable domain containing the amino acid sequence of SEQ ID

NO: 327, or an amino acid sequence at least 90% identical thereto, and a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 328, or an amino acid sequence at least 90% identical thereto;

(xvii) a heavy chain variable domain containing the amino acid sequence of SEQ ID NO: 329 or an amino acid sequence at least 90% identical thereto, and a light chain variable domain containing the amino acid sequence of SEQ ID NO: 330 or an amino acid sequence at least 90% identical to it; and (xviii) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 331, or an amino acid sequence at least 90% identical thereto, and a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 332, or an amino acid sequence of at least 90% identical to it.

For each of the specific VH/VL combinations listed above, it is also conceivable within the scope of the invention to combine a VH domain having an amino acid sequence that is at least 90, 92, 95, 97 or 99% identical to said VH domain sequence with a VL domain having an amino acid sequence that is at least 90, 92, 95, 97 or 99% identical to the sequence of the listed VL domain. Embodiments of the invention in which the amino acid sequence of the VH domain exhibits less than 100% sequence identity to a given VH reference sequence may nonetheless contain heavy chain CDRs that are identical to HCDR1, HCDR2, and HCDR3 of the reference sequence although exhibiting amino acid sequence variation within frame sections. Likewise, embodiments of the invention in which the amino acid sequence of the VL domain exhibits less than 100% sequence identity to a given reference sequence may nonetheless contain light chain CDRs that are identical to LCDR1, LCDR2, and LCDR3 of the reference sequence although exhibiting amino acid sequence variation in within the frame areas.

In the previous paragraph and elsewhere in this specification, the structure of antibodies/antigen-binding fragments is determined based on % sequence identity to the specified reference sequence (given SEQ ID NO). In this context, % sequence identity between two amino acid sequences can be determined by comparing the two optimally aligned sequences, and in which the compared amino acid sequence may contain additions or deletions relative to a reference sequence for optimal alignment between the two sequences. Percent identity is calculated by determining the number of identical positions for which the amino acid residue is identical between two sequences, dividing this number of identical positions by the total number of positions in the comparison window and multiplying the result by 100 to obtain the percent identity between the two sequences. Typically, comparison windows correspond to the full length of the sequence being compared. For example, you can use the BLAST program BLAST 2 sequences (Tatusova et al, Blast 2 sequences - a new tool for comparing protein and nucleotide sequences, FEMS Microbiol Lett. 174: 247-250), available at http://www.ncbi. nlm.nih.gov/gorf/bl2.html, the parameters are used by default (specifically, for the parameters open gap penalty: 5, and expansion gap penalty: 2; the selected matrix, for example, the BLOSUM 62 matrix suggested by the program), the percentage of identity between two sequences to be compared is calculated directly by the program. Determining the sequence identity of sequence queries to a reference sequence is within the skill of the person skilled in the art and can be performed using commercially available analysis software such as BLAST™.

In some preferred embodiments of the present invention, the antibody or antigen binding fragment may comprise a heavy chain variable domain and a light chain variable domain, wherein HCDR1 contains SEQ ID NO: 16, HCDR2 contains SEQ ID NO: 17, HCDR3 contains SEQ ID NO: 18, and LCDR1 contains SEQ ID NO: 61, LCDR2 contains SEQ ID NO: 62, and LCDR3 contains SEQ ID NO: 63.

In some such embodiments of the present invention, the heavy chain variable domain may comprise the amino acid sequence shown as SEQ ID NO: 221, or an amino acid sequence exhibiting at least 90, 95, 97, 98, or 99% sequence identity, and the variable domain the light chain may comprise an amino acid sequence shown as SEQ ID NO: 222 or an amino acid sequence exhibiting at least 90, 95, 97, 98 or 99% sequence identity thereto. In some such embodiments of the present invention, the heavy chain variable domain and the light chain variable domain are the VH and VL domains of antibody 31282 provided herein.

- 15 045372

Antibody 31282 presented herein is derived from antibody 29489. Antibody

31282 was derived from 29489 by replacing M-T with amino acid 116 in the FR4 VH region. It is understood that this substitution removes a potential oxidation site of the antibody and thus improves stability without affecting function. Thus, antibodies 31282 and 29489 have identical sequences

HCDR1, HCDR2, HCDR3, LCDR1, LCDR2 and LCDR3, differing only in the framework.

Accordingly, in certain embodiments of the antibodies or antigen binding fragments of the present invention, the heavy chain variable domain may comprise the amino acid sequence shown as SEQ ID NO: 219, or an amino acid sequence exhibiting at least 90, 95, 97, 98 or 99%, and the light chain variable domain may comprise the amino acid sequence shown as SEQ ID NO: 220, or an amino acid sequence exhibiting at least 90, 95, 97, 98 or 99% sequence identity thereto. In some such embodiments of the present invention, the heavy chain variable domain and the light chain variable domain are the VH and VL domains of antibody 29489 provided herein.

Embodiments of the invention in which the amino acid sequence of the VH domain exhibits less than 100% sequence identity to the sequence shown as SEQ ID NO: 221 or 219 may nevertheless contain heavy chain CDRs that are identical to HCDR1, HCDR2 and HCDR3 SEQ ID NO : 221 and 219 (SEQ ID NOs: 16, 17 and 18, respectively), although showing amino acid sequence variation in the framework regions. Likewise, embodiments of the invention in which the amino acid sequence of the VL domain exhibits less than 100% sequence identity to the sequence shown as SEQ ID NO: 222 or 220 may nevertheless contain light chain CDRs that are identical to LCDR1, LCDR2 and LCDR3 of the SEQ ID NOs: 222 and 220 (SEQ ID NOs: 61, 62 and 63, respectively), while demonstrating amino acid sequence variation in the framework regions.

Exemplary anti-TIGIT antibodies described herein and having the sequences shown in FIG. 1-5 were obtained from 5 parental antibody clones. Table 2 summarizes the pedigree of the antibodies described herein. Naïve parental human anti-TIGIT antibodies were expressed in yeast, and those exhibiting high functional activity against TIGIT (gray rows named 26...) were selected and subjected to affinity maturation. The selected affinity-mature antibodies were then expressed in mammalian cells (white rows under each parent, named 29... or 3...). Additionally, antibody 31282 was generated from 29489 by replacing M-T at amino acid 116 in the FR4 VH region. It is understood that this substitution removes a potential oxidation site of the antibody and thus improves stability without affecting function. In addition, antibody 31288 was generated from 29494 by replacing V-L with amino acid 2 in the FR1 VH region and by replacing M-T with amino acid 120 in the FR4 VH region. The V-L substitution is predicted to restore the germline sequence of VH4-39 and the M-T substitution to remove a potential antibody oxidation site and thereby improve stability without affecting function.

- 16 045372

table 2

Antibody clone CDR3 VH cell line Optimization method VH germinal lines 26518 26518 Parental VH3-07 29478 26518 H1/H2/NC VH3-30 26452 26452 Parental VH1-46 29487 26452 H1/H2/NC VH1-46 29489 26452 H1/H2/NC VH1-46 31282 29489 Amino acid mutation Ml 16T VH1-46 26486 26486 Parental VH4-0B 29499 26486 H1/H2/NC VH4-39 29494 26486 H1/H2/NC VH4-39 31288 29494 Germline reversion + amino acid mutation Ml 16T VH4-39 32919 31288 L1/L2/L3 VH4-39 32931 31288 L1/L2/L3 VH4-39 26521 26521 Parental VH1-69 29513 26521 H1/H2/NC VH1-69 26493 26493 Parental VH3-09 29520 26493 H1/H2/NC VH3-09 29523 26493 H1/H2/NC VH3-33 29527 26493 H1/H2/NC VH3-30 26432 26432 Parental VH1-69 32959 26432 H1/H2/NC VH1-69

Second generation antibodies exhibit higher affinity than the corresponding parent antibodies.

In some embodiments, the invention provides anti-TIGIT antibodies or antigen binding fragments thereof, wherein the VH domain is derived from a human germline V region sequence selected from: VH3-07, VH3-30, VH1-46, VH4-0B, VH4- 39, VH1-69, VH3-09, VH333, VH3-30. In some preferred embodiments of the present invention, the antibody or antigen binding fragment thereof comprises a VH domain derived from the human germline V region VH1-46.

A VH domain is derived from a particular V region of a germline sequence if the heavy chain variable region sequence is more likely to be derived from that germline than from any other.

Epitope TIGIT

The invention also provides an antibody or antigen binding fragment thereof that binds human TIGIT at an epitope comprising residues Q56 and 1109. In some embodiments of the present invention, the antibody or antigen binding fragment thereof binds human TIGIT at at least residues Q56, N58 and 1109. In some embodiments of the present invention, an antibody or antigen binding fragment thereof binds human TIGIT to an epitope comprising residues Q56, N58 and 1109 and optionally one or more residues E60, 168, L73 and H76. In some embodiments of the present invention, the antibody or antigen-binding fragment thereof binds human TIGIT to an epitope containing residues Q56, N58, E60,168, L73, H76, and 1109.

In some embodiments of the present invention, the antibody or antigen-binding fragment thereof binds to human TIGIT at an epitope consisting of TIGIT residues Q56, N58, E60,168, L73, H76, and 1109. In some embodiments of the present invention, the antibody or antigen-binding fragment thereof binds same epitope as antibody 31282.

When an antibody or antigen binding fragment binds a human TIGIT epitope containing said TIGIT residues, the antibody binds each of these residues and optionally other TIGIT residues. When an antibody or antigen binding fragment binds a human TIGIT epitope consisting of TIGIT residues Q56, N58, E60,168, L73, H76, and 1109, the antibody binds each of these residues and no other TIGIT residues.

An antibody or antigen binding fragment binds to human TIGIT at a given epitope if it binds to the specified amino acid residue(s) of TIGIT upon binding to TIGIT. As used herein, an antibody binds to a TIGIT residue if, when present in the protein complex formed by the antibody-TIGIT binding, the residue meets each of the following criteria: (i) it has a calculated binding free energy contribution greater than 0.3

- 17045372 kcal/mol, (ii) it has an experimental average B-factor lower than the average B-factor of all residues in the X-ray structure, (iii) it creates at least 3 pairs of interatomic contacts of heavy atoms with antibody atoms at a distance of less than or equal to 4.0 Angstroms, (iv) it does not form only hydrogen bonds or solvent-exposed ionic interactions, (v) if it is a non-aromatic polar residue (Asn, Gln, Ser, Thr, Asp, Glu, Lys, or Arg) , it makes at least one hydrogen bond or ionic interaction with the antibody. Calculation of free energy of binding will be within the capabilities of a qualified person. Preferably, the binding free energy is calculated using an empirical force field, preferably FoldX. FoldX would be familiar to a specialist and publicly available at http://foldxsuite.crg.eu/. Binding free energy calculations using FoldX are also described in Guerois et al. J. Mol. Biol. 2002; 320 (2): 369-87, which is incorporated herein by reference. One skilled in the art will know that all heavy atoms are non-hydrogen atoms (including C, N, O, S).

Accordingly, the invention also relates to an antibody or antigen binding fragment thereof that binds to human TIGIT at at least residues Q56 and I109. In some embodiments of the present invention, the antibody or antigen binding fragment thereof contacts human TIGIT at at least residues Q56, N58 and I109. In some embodiments of the present invention, an antibody or antigen binding fragment thereof binds to human TIGIT at at least residues Q56, N58, and I109, and optionally one or more of residues E60, I68, L73, and H76. In some embodiments of the present invention, the antibody or antigen binding fragment thereof contacts human TIGIT at at least residues Q56, N58, E60, 168 L73, H76, and I109.

In some such embodiments of the present invention, the antibody or antigen binding fragment thereof binds to human TIGIT only at residues Q56, N58, E60, I68, L73, H76, and I109.

Means for determining which TIGIT residues bind to an antibody or antigen binding fragment are familiar to one of ordinary skill in the art, including x-ray crystallography, such as that described in Example 23.

Also provided is an isolated antibody or antigen binding fragment thereof that binds to the same epitope as the antibody or antigen binding fragment described herein.

Antibody subtypes

TIGIT antibodies can have various embodiments in which both a VH domain and a VL domain are present. The term antibody is used herein in its broadest sense to include, but is not limited to, monoclonal antibodies (including full-length monoclonal antibodies), polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies), provided that they demonstrate the corresponding immunological specificity of human TIGIT protein. As used herein, the term monoclonal antibody refers to an antibody derived from a population of substantially homogeneous antibodies, i.e. the individual antibodies that make up the population are identical, except for possible naturally occurring mutations that may be present in minute quantities. Monoclonal antibodies are highly specific and directed against a single antigenic site. Additionally, unlike conventional (polyclonal) antibody preparations, which typically include different antibodies directed against different determinants (epitopes) on an antigen, each monoclonal antibody is directed against a single determinant or epitope on an antigen.

In non-limiting embodiments of the present invention, the TIGIT antibodies provided herein may contain CH1 domains and/or CL domains whose amino acid sequence is entirely or substantially human. When a TIGIT antibody is intended for therapeutic use in humans, it is typical that all or at least a portion of the antibody constant region has an entirely or substantially human amino acid sequence. Therefore, one or more or any combination of a CH1 domain, a hinge region, a CH2 domain, a CH3 domain, and a CL domain (and a CH4 domain, if present) may be completely or substantially human with respect to its amino acid sequence. Such antibodies may be of any human isotype, with human IgG4 and IgG1 being particularly preferred.

Advantageously, the CH1 domain, the hinge region, the CH2 domain, the CH3 domain, and the CL domain (and the CH4 domain, if present) may all have an entirely or substantially human amino acid sequence. In the context of a constant region of a humanized or chimeric antibody or antibody fragment, the term substantially human refers to amino acid sequence identity of at least 90%, or at least 92%, or at least 95%, or at least 97%, or at least 99% human constant region. The term human amino acid sequence as used herein refers to the amino acid sequence that is encoded by the human immunoglobulin gene, which includes germline,

- 18 045372 rearranged and somatically mutated genes. Such antibodies may be of any human isotype, with human IgG4 and IgG1 being particularly preferred.

Also provided are TIGIT antibodies containing human sequence constant domains that have been altered by one or more amino acid additions, deletions or substitutions relative to the human sequence.

The TIGIT antibodies provided herein can be of any isotype. Antibodies intended for therapeutic use in humans are usually of the IgA, IgD, IgE IgG, IgM type, often of the IgG type, in which case they may belong to any of the four subclasses IgG1, IgG2a and b, IgG3 or IgG4. Within each of these subclasses, it is permitted to make one or more amino acid substitutions, insertions or deletions within the Fc portion, or to make other structural modifications, for example, to increase or decrease Fc-dependent functionality.

In certain preferred embodiments of the present invention, the TIGIT antibodies provided herein are IgG antibodies. In certain embodiments of the present invention, the antibodies of the present invention are IgG1 antibodies. In some alternative embodiments of the present invention, the antibodies of the present invention are IgG4 antibodies.

IgG4 antibodies are known to undergo Fab arm exchange (FAE), which can result in unpredictable pharmacodynamic properties of the IgG4 antibody. FAE has been demonstrated to be prevented by the S228P mutation in the hinge region (Silva et al. J Biol Chem. 2015 Feb 27;290(9):54625469). Therefore, in some such embodiments of the present invention, in which the antibody of the present invention is an IgG4 antibody, the antibody contains the S228P mutation, that is, a serine to proline mutation at position 228 (according to EC numbering).

In non-limiting embodiments of the present invention, it is contemplated that one or more amino acid substitutions, insertions or deletions may be made in the heavy and/or light chain constant region, especially the Fc region. Amino acid substitutions can result in the replacement of the substituted amino acid with another naturally occurring amino acid or with a non-naturally occurring or modified amino acid. Other structural modifications are also allowed, such as, for example, changes in the glycosylation pattern (eg, by adding or removing N- or O-linked glycosylation sites). Depending on the intended use of the TIGIT antibody, it may be desirable to modify the antibody of the present invention with respect to its Fc receptor binding properties, for example, to modulate effector function.

In some embodiments of the present invention, TIGIT antibodies may comprise an Fc region of a given antibody isotype, such as human IgG1, that is modified to reduce or substantially eliminate one or more antibody effector functions naturally associated with that antibody isotype.

As demonstrated herein, antibodies with lytic effector cell functions can be effective in depleting Treg cell populations, but surprisingly without adversely affecting conventional effector T cell populations. This selectivity may allow greater inhibition of the regulatory effect of Tregs while sparing antitumor effector T cells.

Therefore, in some alternative embodiments of the present invention, TIGIT antibodies retain one or more antibody effector functions naturally associated with that antibody isotype. For example, the TIGIT antibodies of the present invention may be IgG1 antibodies that retain ADCC functionality. In other embodiments of the present invention, TIGIT antibodies may comprise an Fc region of a given antibody isotype, such as human IgG1, that is modified to enhance one or more antibody effector functions naturally associated with that antibody isotype. In this context, the effector functions of antibodies include one or more or all of antibody-dependent cellular cytotoxicity (ADCC), complement-dependent cytotoxicity (CDC) and antibody-dependent cellular phagocytosis (ADCP).

In certain embodiments of the present invention, the anti-TIGIT antibody is a modified antibody.

In certain embodiments of the present invention, a bispecific antibody is provided comprising a first arm specific for TIGIT and a second arm specific for a second target. In preferred embodiments of the present invention, the second target is an immune checkpoint molecule. In some embodiments of the present invention, the second target is OX40. In some embodiments of the present invention, the second target is an ICOS. In some embodiments of the present invention, the second target is GITR. In some embodiments of the present invention, the second target is a 4-1BB. In some embodiments of the present invention, the second target is PD-1. In some embodiments of the present invention, the second target is PD-L1. In some embodiments of the present invention, the first TIGIT-specific arm contains a combination of HCDR1 sequences,

- 19 045372

HCDR2, HCDR3, LCDR1, LCDR2 and LCDR3 antibodies according to this invention. In some embodiments of the present invention, the first arm comprises a heavy chain variable domain and a light chain variable domain, wherein HCDR1 contains SEQ ID NO: 16, HCDR2 contains SEQ ID NO: 16

NO: 17, HCDR3 contains SEQ ID NO: 18, and LCDR1 contains SEQ ID NO: 61, LCDR2 contains SEQ ID

NO: 62, and LCDR3 contains SEQ ID NO: 63.

Monoclonal antibodies or antigen-binding fragments thereof that cross-compete with the TIGIT antibodies described herein are antibodies that bind human TIGIT at a site(s) that are identical to or overlapping with the site(s) at which the present TIGIT antibodies bind. Competing monoclonal antibodies or antigen binding fragments thereof can be identified, for example, using an antibody competition assay. For example, a sample of purified or partially purified human TIGIT may be bound to a solid support. An antibody compound or antigen binding fragment thereof of the present invention and a monoclonal antibody or antigen binding fragment thereof believed to be capable of competing with such antibody compound of the present invention are then added. One of the two molecules is labeled. If a tagged compound and an unlabeled compound bind to separate and discrete sites on TIGIT, then the tagged compound will bind to the same level regardless of whether the putative competing compound is present or not. However, if the interaction sites are identical or overlap, then the unlabeled compound will compete and the amount of labeled compound bound to the antigen will be reduced. If the unlabeled compound is present in excess, then very little, if any, of the labeled compound will bind.

For purposes of this invention, competitive monoclonal antibodies or antigen binding fragments thereof are those that reduce the binding of the antibody compounds to TIGIT by about 50, about 60, about 70, about 80, about 85, about 90, about 95, or about 99%. Details of procedures for performing such competition assays are well known in the art and can be found, for example, in Harlow and Lane, Antibodies, A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 1988, 567 -569, 1988 , ISBN 0-87969-314-2. Such assays can be done quantitatively using purified antibodies. A standard curve is established by titrating one antibody against itself, i.e. the same antibody is used for both tag and competitor. The ability of an unlabeled competing monoclonal antibody or antigen-binding fragment to inhibit the binding of a labeled molecule to the plate is titrated. The results are plotted and the concentrations required to achieve the desired degree of binding inhibition are compared.

Antibodies of the invention demonstrate high affinity for TIGIT and competition with CD155

In certain embodiments of the present invention, the antibodies or antigen-binding fragments of the present invention exhibit high affinity for human TIGIT. In some embodiments of the present invention, Fab fragments of antibodies according to the present invention exhibit a K _D for TIGIT measured using ForteBio™ in the range of ^1x10'10 to 5x10'8 M, optionally from 7x10'10 to 4x10'8 M. In some In embodiments, antibodies of this invention exhibit MSD K _D in the range of 1x10'11 to 5x10'9 M, optionally from 2x10'11 to 1x10'9. In some embodiments of the present invention, Fab fragments of antibodies according to the present invention exhibit a KD for TIGIT, measured by Biacore™, in the range of ^1x10'10 M to 1x10'9 M, optionally from ^1x10'10 to ^7x10'10 M, optionally from 2x10 ' ¹⁰ to 7x10 ' ¹⁰ M.

- 20 045372

Table 3

Clone ForteBio Fab KD biotinylated human TIGIT HIS (Μ) monovalent ForteBio Fab KD mouse TIGIT- Fc (M) monovalent ForteBio Fab KD Cyno TIGIT-Fc (M) monovalent ForteBio IgG KD human TIGIT- Fc(M)Avid MSD-KD single-shaft NTNOGO (M), human TIGITHis Biacore - KD monovalent (Μ), human whom TIGITHis Binding of Jurkat cells human IMM TIGIT FON (fold compared to negative control) Binding of Jurkat cells by mouse TIGIT FON (Multiplicity compared to negative control) 2651 8 1.24E-09 Ν.Β. 4.47E-09 6.30E-10 154 233 2947 8 7.03E-10 9.18E-08 1.26E-09 5.27E-10 182 500 2645 2 5.08E-09 Ν.Β. N.B. 4.74E-10 164 47 2948 7 2.08E-09 Ν.Β. 1.55E-07 3.96E-10 161 95 2948 9 8.81E-10 Ν.Β. 3.52E-08 3.53E-10 1.1E-10 2.48E- 10 162 187 3128 2 1.34E-09 Ν.Β. 3.77E-08 2.94E- 10 2648 6 2.19E-08 Ν.Β. N.B. 5.89E-10 143 199 2949 9 1.66E-09 2.55E-08 1.45E-08 3.19E-10 1.9E-11 164 541 2949 4 1.66E-09 5.36E-08 1.86E-08 3.76E-10 7.0E-11 2.70E- 10 164 511 3128 8 2.09E-09 2.51E-08 1.92E- 10

- 21 045372

3291 9 1.42E-09 6.57E-09 680 3293 1 1.18E-09 1.97E-09 741 2949 9 1.66E-09 2.55E-08 1.45E-08 3.19E-10 1.9E-11 164 541 2652 1 9.87E-09 N.B. 1.49E-07 5.41E-10 146 218 2951 3 7.74E-10 8.55E-08 9.56E-09 3.92E-10 2.5E-11 156 406 2649 3 4.06E-08 2.67E-08 N.B. 1.49E-09 80 463 2952 0 1.31E-09 1.95E-09 2.68E-09 3.84E-10 2.1E-10 7.16E- 10 166 535 2952 3 3.84E-09 1.89E-08 2.79E-08 5.31E-10 1.7E-09 150 502 2952 7 1.33E-O9 2.02E-08 1.76E-08 3.50E-10 6.4E-10 142 414 2643 2 1.31E-08 N.B. N.B. 4.62E-09

As demonstrated in the Examples, antibody 31282 exhibits remarkably high affinity for TIGIT expressed on transgenic cells. Accordingly, in certain embodiments of the present invention, an anti-TIGIT antibody or antigen binding fragment thereof provided herein exhibits a binding EC50 for human TIGIT of less than 0.5 nM. In such preferred embodiments of the present invention, the antibody or antigen binding fragment exhibits a binding EC50 of about 0.05 to about 0.4 nM, preferably from about 0.05 to about 0.3 nM, preferably from about 0.05 to about 0. 2 nM, preferably from about 0.05 to about 0.15 nM. In certain preferred embodiments of the present invention, the antibody or antigen binding fragment exhibits a binding EC ₅₀ for human TIGIT of about 0.1 nM. In preferred embodiments of the present invention, the antibody comprises antibody CDR 31282. Preferably, the EC50 is determined using Jurkat cells expressing human TIGIT as described in Example 18. In certain embodiments of the present invention, antibodies or antigen binding fragments of the present invention cross-react with mouse TIGIT and/or TIGIT Javanese.

Because 29 ... second-generation antibodies are progeny of the affinity-mature, highly functional parent antibodies, they are expected to exhibit at least the same or equivalent functional properties as the parent antibodies, and vice versa.

As described herein, in some embodiments of the present invention, an antibody or antigen binding fragment of the present invention has equivalent affinity for TIGIT expressed by CD8 T cells and expressed by Treg cells. As used herein, an antibody or antigen binding fragment has equivalent affinity for CD8 T cells and Treg cells if the affinity for CD8 T cells is in the range of 0.5 to 1.5 times the affinity for Treg cells. For example, an antibody having equivalent affinity for CD8 T cells and Treg cells that exhibits an affinity for Treg cells of 0.03 nM will exhibit an affinity for CD8 T cells in the range of 0.015-0.045 nM.

In table Figure 3 provides a summary of the affinity properties of the anti-TIGIT antibodies of this invention, where gray cells indicate parental antibody clones and second and third generation antibodies of each cell line are shown directly below the corresponding parent antibody (see also Table 2).

As demonstrated in the Examples, antibody 31282 exhibits remarkably high affinity for TIGIT expressed on primary human CD8+ T cells. Accordingly, in certain embodiments of the present invention, an anti-TIGIT antibody or antigen binding fragment thereof provided herein exhibits a binding EC ₅₀ for human TIGIT of less than 0.5 nM. In such preferred embodiments of the present invention, the antibody or antigen binding fragment exhibits a binding EC ₅₀ of about 0.05 to about 0.4 nM, preferably from about 0.1 to about 0.3 nM. In certain preferred embodiments of the present invention, the antibody or antigen binding fragment exhibits a binding EC ₅₀ for human TIGIT of about 0.2 nM. In preferred embodiments of the present invention, the antibody or antigen binding fragment comprises the antibody CDR 31282. Preferably, _EC50 is determined using CD8+ T cells from human PBMC, preferably from a healthy individual, as described in Example 18.

As demonstrated in the accompanying Examples, in some embodiments of the present invention, the antibodies or antigen binding fragment of the present invention exhibit high affinity for TIGIT-expressing CD8 T cells and high affinity for TIGIT-expressing Treg cells. In some embodiments of the present invention, antibodies or antigen binding fragment of the present invention exhibit affinity for TIGIT-expressing CD8 T cells and TIGIT-expressing Treg cells having an EC ₅₀ of less than 0.5 nM, preferably less than 0.3 nM, preferably less than 0.2 nM. In some embodiments of the present invention, the antibodies or antigen binding fragment of the present invention exhibit equivalent affinity for TIGIT-expressing CD8 T cells and for TIGIT-expressing Treg cells.

The antibodies of this invention (eg, antibody 31282) exhibit surprisingly high affinity for CD8+ T cells from cancer patients. This is particularly beneficial as enhancing T cell effector activity in cancer patients by inhibiting TIGIT signaling may lead to more effective tumor control. Accordingly, in certain embodiments of the present invention, an anti-TIGIT antibody or antigen binding fragment thereof provided herein exhibits EC ₅₀ binding of less than 0.5 nM for human TIGIT on human CD8+ T cells from cancer patients. In such preferred embodiments of the present invention, the antibody or antigen binding fragment exhibits a binding EC ₅₀ of about 0.05 to about 0.4 nM, preferably from about 0.1 to about 0.3 nM. In some preferred embodiments of the present invention, the antibody or antigen binding fragment exhibits an EC50 for human TIGIT of about 0.1 nM to about 0.2 nM. In preferred embodiments of the present invention, the antibody or antigen binding fragment comprises the antibody CDR 31282. Preferably, EC ₅₀ is determined using CD8+ T cells from PBMCs collected from a cancer patient as described in Example 18.

As demonstrated in the accompanying Examples, in certain embodiments of the present invention, antibodies or antigen binding fragments of the present invention compete with CD155/PVR for TIGIT binding. In certain embodiments of the present invention, an antibody or antigen binding fragment of the present invention competes with CD155 having an IC50 of 0.2 nM or less, preferably 0.1 nM or less. In some embodiments of the present invention, the antibody or antigen binding fragment competes with CD155 having an IC50 of about 0.05 nM or less. In some preferred embodiments of this invention, the IC ₅₀ is about 0.05 nM. Without wishing to be bound by any theory, competition of antibodies with CD155 for TIGIT binding is expected to reduce levels of CD155-induced TIGIT signaling, thereby increasing levels of effector T cell activation.

The invention further provides variations in the affinities of the antibodies described herein.

The invention also provides an isolated antibody or antigen binding fragment thereof that cross-competes for binding to human TIGIT with the antibody or antigen binding fragment described herein.

Antibodies of the invention promote pro-inflammatory activity of T cells

Antibodies of this invention (eg, antibody 31282) are surprisingly effective in stimulating pro-inflammatory activity of CD8+ T cells. As demonstrated in the Examples, the antibodies or antigen-binding fragments of this invention (especially 31282) are more effective at stimulating pro-inflammatory CD8+ T cell activity (as indicated by IFNg release) than the anti-TIGIT comparator antibodies (see FIG. 24). This improved efficacy compared to comparator antibodies was demonstrated in TIGIT-expressing transgenic Jurkat reporter cells and primary CD8 T cells. Accordingly, in certain embodiments of the present invention, the anti-TIGIT antibody or antigen binding fragment thereof provided herein exhibits an activation EC ₅₀ of less than 5 nM for human TIGIT expressed by Jurkat reporter cells as described in Example 19. In preferred such embodiments, of the present invention, the antibody or antigen binding fragment exhibits an EC ₅₀ of about 1 to about 4 nM, preferably from about 2 nM to about 4 nM.

In some embodiments of this invention, the anti-TIGIT antibody or antigen binding fragment thereof provided herein exhibits an activation EC50 of less than 0.4 nM for CD8 T cells from healthy individuals as described in Example 19. CD8 T Cell Activity (ie, proinflammatory activity) can be measured by the production of inflammatory cytokines (eg, IFNg). In preferred such embodiments of the present invention, the antibody

- 23 045372 or antigen binding fragment exhibits an EC ₅₀ of about 0.05 to about 0.4 nM, preferably from about 0.1 to about 0.2 nM. Preferably, EC ₅₀ is determined using CD8+ T cells from

PBMCs taken from a healthy person, as described in example 19.

The accompanying examples further and unexpectedly demonstrate that the provided anti-TIGIT antibodies are effective in increasing the activity of gamma delta (γδ or g/d) T cells (i.e., T cells expressing γδ TCR subunits as opposed to conventional αβ TCR subunits ). Such γδ T cells form a distinct and important component of the immune system, and the ability of the antibodies presented herein to stimulate the activity of these cells highlights the utility of antibodies.

Accordingly, this document also provides a method for promoting γδ T cell activity, comprising contacting a population of γδ T cells with an anti-TIGIT antibody. In certain embodiments of this invention, the method is performed in vitro. In certain embodiments of the present invention, the method is performed in vivo in a human subject. In some such embodiments of the present invention, the human subject has cancer. In some embodiments of the present invention, the anti-TIGIT antibody or antigen binding fragment thereof comprises a combination of the HCDR1, HCDR2, HCDR3, LCDR1, LCDR2 and LCDR3 sequences of the antibodies of the present invention. In some embodiments of the present invention, the anti-TIGIT antibody comprises a heavy chain variable domain and a light chain variable domain, wherein HCDR1 contains SEQ ID NO: 16, HCDR2 contains SEQ ID NO: 17, HCDR3 contains SEQ ID NO: 18, and LCDR1 contains SEQ ID NO: 61, LCDR2 contains SEQ ID NO: 62, and LCDR3 contains SEQ ID NO: 63.

Selective depletion of Treg cells

As demonstrated herein, anti-TIGIT antibodies are capable of selectively depleting TIGIT-expressing Treg cells. That is, anti-TIGIT antibodies reduce the proportion of TIGIT-expressing Treg cells relative to the total T cell population to a greater extent than they reduce the proportion of effector or memory CD4 or CD8 T cells.

In some embodiments of the present invention, the antibody or antigen-binding fragment thereof selectively depletes TIGIT-expressing Treg cells.

This selective depletion of TIGIT-expressing Tregs can be mediated through selective lysis of TIGIT-expressing Tregs (eg, by ADCC or CDC (see Figs. 20, 21 and 25). TIGIT-expressing Tregs are believed to be more potent regulatory cells than non-TIGIT-expressing Treg cells. Without wishing to be bound by theory, selective depletion by lysis of TIGIT-expressing Treg cells is expected to increase T cell effector function (eg, T cell-mediated cytotoxicity, release of proinflammatory cytokines) by depletion of the total number of Treg cells, but also depletion of these Treg cells exhibiting a more powerful regulatory function.This increased effector function of T cells is demonstrated in Fig. 24.

Therefore, in certain embodiments of the present invention, the antibodies or antigen-binding fragments of the present invention selectively lyse TIGIT-expressing Treg cells.

Selective depletion of Treg cells expressing TIGIT may also be mediated by inducing internalization of the TIGIT receptor such that it is no longer expressed on the cell membrane. Without wishing to be bound by theory, by inducing internalization of TIGIT such that TIGIT+ Treg cells become TIGIT-Treg cells, the regulatory function of these cells is expected to become less active (since TIGIT+ Tregs are more active regulatory cells). As a result of receptor internalization and subsequent reduction in the regulatory activity of these Tregs, an increase in T cell effector function is expected. Therefore, in certain embodiments of the present invention, antibodies or antigen-binding fragments of the present invention inhibit the suppressive activity of TIGIT-expressing Treg cells, preferably by inducing the internalization of TIGIT by TIGIT-expressing Treg cells.

It is particularly advantageous for the anti-TIGIT antibodies of the present invention to exhibit high affinity for CD8 T cells and Treg cells, as well as to selectively deplete Treg cells, thereby promoting T cell effector function through two mechanisms. Preservation of antibody effector function (eg, ADCC, CDC) results in efficient depletion of Tregs, and selectivity means that antibody effector function does not lead to undesirable depletion of effector T cells. The selectivity is particularly surprising since previous attempts to produce an anti-TIGIT antibody were aimed at eliminating the effector function of the antibody to avoid lysis of effector T cells expressing TIGIT. In addition, because the TIGIT antibodies of the present invention exhibit affinity for effector T cells (eg, CD8 T cells), TIGIT-mediated signaling in these cells can be inhibited by competing for CD155 binding and/or inducing internalization of TIGIT on effector T cells. 24 045372 cells. When combined, these effects of the antibodies of this invention can result in significant enhancement of T cell effector function.

Other unexpected beneficial properties exhibited by the antibodies and antigen binding fragments of the present invention include enhancing T cell effector function (eg, release of pro-inflammatory cytokines) in tumor infiltrating lymphocytes (TILs). Exposure to the tumor microenvironment can cause TILs to exhibit energetic or so-called exhausted phenotypes, possibly due to overexposure to antigen and/or an immunosuppressive tumor microenvironment. It is desirable to enhance the effector function of TILs, since these are the cells that infiltrate the tumor itself and are thus located at the locus most suitable for reducing tumor size or growth; however, due to the anergic or exhausted phenotype of many TILs, it is expected that it will be difficult to enhance their effector function. Therefore, the increase in pro-inflammatory response from TILs following exposure to antibodies of this invention is unexpected and indicates that antibodies may be particularly effective therapeutic agents.

Even more unexpected beneficial properties exhibited by antibodies and antigen-binding moieties include the ability to enhance the proinflammatory activity of gamma delta (γδ) T cells. The ability to stimulate the activity of unusual T cells, such as γδ T cells, has not previously been reported for an anti-TIGIT antibody, and this provides an opportunity to treat diseases other than cancer in which γδ T cells are known to be important. For example, γδ T cells have been reported to be involved in the response to pathogenic infection (bacterial, viral (eg, CMV), fungal) and also play a role in protection against autoimmune diseases. In addition, the surprising ability to stimulate the activity of unusual T cells provides additional antitumor activity to the antibodies.

In an additional aspect, a method of selectively depleting Treg cells from a T cell population is provided, comprising contacting the T cell population with an anti-TIGIT antibody or an antigen binding fragment thereof, whereby the anti-TIGIT antibody selectively depletes the Treg cell population. In certain embodiments of this invention, the method is performed in vitro. In certain embodiments of the present invention, the method is performed in vivo in a human subject. In some such embodiments of the present invention, the human subject has cancer. In some embodiments of the present invention, the anti-TIGIT antibody or antigen binding fragment thereof comprises a combination of the HCDR1, HCDR2, HCDR3, LCDR1, LCDR2 and LCDR3 sequences of the antibodies of the present invention. In some embodiments of the present invention, the anti-TIGIT antibody comprises a heavy chain variable domain and a light chain variable domain, wherein HCDR1 contains SEQ ID NO: 16, HCDR2 contains SEQ ID NO: 17, HCDR3 contains SEQ ID NO: 18, and LCDR1 contains SEQ ID NO: 61, LCDR2 contains SEQ ID NO: 62, and LCDR3 contains SEQ ID NO: 63.

As demonstrated in the accompanying Examples, the present invention also provides anti-TIGIT antibodies that do not compete with CD155/PVR for TIGIT binding. Therefore, in a further aspect, the present invention provides a human TIGIT antibody or antigen binding fragment thereof that does not compete with CD155/PVR for binding human TIGIT. In some such embodiments of the present invention, Fab fragments of the non-competitive anti-TIGIT antibodies CD155 of the present invention exhibit a K _D for TIGIT, measured using ForteBio™, in the range of 5x10'9 to ^5x10'8 M, optionally from ^1x10'8 to 3x10' ⁸ M.

In some preferred embodiments of the present invention, the antibody may comprise a heavy chain variable domain and a light chain variable domain, wherein HCDR1 contains SEQ ID NO: 280, HCDR2 contains SEQ ID NO: 281, HCDR3 contains SEQ ID NO: 282, and LCDR1 contains SEQ ID NO: 292, LCDR2 contains SEQ ID NO: 293, and LCDR3 contains SEQ ID NO: 294. In some such embodiments of the present invention, the heavy chain variable domain may comprise the amino acid sequence shown as SEQ ID NO: 333, or the amino acid a sequence exhibiting at least 90, 95, 97, 98, or 99% sequence identity, and the light chain variable domain may comprise the amino acid sequence shown as SEQ ID NO: 334 or an amino acid sequence exhibiting at least 90, 95, 97 , 98 or 99% sequence identity with it.

Embodiments of the invention in which the amino acid sequence of the VH domain exhibits less than 100% sequence identity to the sequence shown as SEQ ID NO: 333 may nevertheless contain heavy chain CDRs that are identical to HCDR1, HCDR2 and HCDR3 of SEQ ID NO: 333 (SEQ ID NOs: 280, 281 and 282, respectively), although showing amino acid sequence variation in the framework regions. Likewise, embodiments of the invention in which the amino acid sequence of the VL domain exhibits less than 100% sequence identity to the sequence shown as SEQ ID NO: 334 may nevertheless contain light chain CDRs that are identical to LCDR1, LCDR2 and LCDR3 of SEQ ID NO : 334 (SEQ ID NO: 292, 293 and 294, respectively), while demonstrating amino acid variation

- 25 045372 sequences in frame regions.

In some preferred embodiments of the present invention, the antibody may comprise a heavy chain variable domain and a light chain variable domain, wherein HCDR1 contains SEQ ID NO: 353, HCDR2 contains SEQ ID NO: 354, HCDR3 contains SEQ ID NO: 355, LCDR1 contains SEQ ID NO: 355 ID NO: 356, LCDR2 contains SEQ ID NO: 357, LCDR3 contains SEQ ID NO: 358. In some such embodiments of the present invention, the heavy chain variable domain may contain the amino acid sequence shown as SEQ ID NO: 367, or the amino acid sequence, exhibiting at least 90, 95, 97, 98, or 99% sequence identity, and the light chain variable domain may comprise the amino acid sequence shown as SEQ ID NO: 368 or an amino acid sequence exhibiting at least 90, 95, 97, 98 or 99% sequence identity with it.

Embodiments of the invention in which the amino acid sequence of the VH domain exhibits less than 100% sequence identity to the sequence shown as SEQ ID NO: 367 may nevertheless contain heavy chain CDRs that are identical to HCDR1, HCDR2 and HCDR3 of SEQ ID NO: 367 (SEQ ID nO: 353, 354 and 355, respectively), although showing amino acid sequence variation in the framework regions. Likewise, embodiments of the invention in which the amino acid sequence of the VL domain exhibits less than 100% sequence identity to the sequence shown as SEQ ID NO: 368 may nevertheless contain light chain CDRs that are identical to LCDR1, LCDR2 and LCDR3 of SEQ ID NO : 368 (SEQ ID NO: 356, 357 and 358, respectively), while demonstrating amino acid sequence variation in the framework regions.

Polynucleotides. Vectors and expression systems

The invention also provides polynucleotide molecules encoding the TIGIT antibodies of the present invention, as well as expression vectors containing nucleotide sequences that encode the TIGIT antibodies of the present invention, operably linked to regulatory sequences that allow expression of the antigen binding polypeptide in a host cell or cell-free expression system, and the cell -host or cell-free expression system containing the expression vector.

Polynucleotide molecules encoding the TIGIT antibodies of the present invention include, for example, recombinant DNA molecules. The terms nucleic acid, polynucleotide or polynucleotide molecule are used interchangeably herein and refer to any DNA or RNA molecule, single or double stranded and, if single stranded, its complementary sequence molecule. When discussing nucleic acid molecules, the sequence or structure of a particular nucleic acid molecule may be described herein in accordance with the general convention of presenting the sequence in the 5' to 3' direction. In some embodiments of the present invention, the nucleic acids or polynucleotides are isolated. This term, when applied to a nucleic acid molecule, refers to a nucleic acid molecule that is separated from the sequences with which it is immediately adjacent in the natural genome of the organism in which it originated. For example, the isolated nucleic acid may include a DNA molecule inserted into a vector, such as a plasmid or viral vector, or integrated into the genomic DNA of a prokaryotic or eukaryotic cell or non-human host organism. When applied to RNA, the term isolated polynucleotide refers primarily to the RNA molecule encoded by the isolated DNA molecule as defined above. Alternatively, the term may refer to an RNA molecule that has been purified/separated from other nucleic acids with which it would be associated in its natural state (ie, in cells or tissues). The isolated polynucleotide (either DNA or RNA) may also be a molecule produced directly by biological or synthetic means and separated from other components present during its production.

To recombinantly produce a TIGIT antibody in accordance with the present invention, a recombinant polynucleotide encoding it can be prepared (using standard molecular biology techniques) and inserted into a replicable vector for expression in a selected host cell or cell-free expression system. Suitable host cells may be prokaryotic, yeast or higher eukaryotic cells, in particular mammalian cells. Examples of suitable mammalian host cell lines include monkey kidney line CV1 transformed with SV40 (COS-7, ATCC CRL 1651); human embryonic kidney line (293 or 293 cells subcloned for growth in suspension culture, Graham et al., J. Gen. Virol. 36: 59-74, 1977); baby hamster kidney cells (BHK, ATCC CCL 10); Chinese hamster ovary/-DHER cells (CHO, Urlaub et al., Proc. Natl. Acad. Sci. USA 77: 4216, 1980; or CHO derived clones like CHO-K1, ATCC CCL-61, Kao and Puck, Genetics of somatic mammalian cells, VII. Induction and isolation of nutritional mutants in Chinese hamster cells, Proc. Natl. Acad. Sci. 60: 1275-1281, 1968); mouse Sertoli cells (TM4; Mather, Biol. Reprod. 23: 243-252, 1980); mouse myeloma cells SP2/0-AG14 (ATCC CrL 1581; ATCC CRL 8287) or nSo (HPA culture collection no. 85110503); monkey kidney cells (CV1 ATCC CCL 70); African kidney cells

- 26 045372 green monkey (VERO-76, ATCC CRL-1587); human cervical carcinoma cells (HELA, ATCC CCL 2); canine kidney cells (MDCK, ATCC CCL 34); buffalo rat liver cells (BRL 3A, ATCC CRL 1442); human lung cells (W138, ATCC CCL 75); human liver cells (Hep G2, HB 8065); mouse mammary tumor (MMT 060562, ATCC CCL51); TRI cells (Mather et al., Annals NY Acad. Sci. 383: 44-68, 1982); MRC 5 cells; FS4 cells; and the human hepatoma line (Hep G2), as well as the PERC-6 DSM cell line. Expression vectors suitable for use in each of these host cells are also generally known in the art.

It should be noted that the term host cell generally refers to the cultured cell line. In general, humans into whom an expression vector encoding an antigen binding polypeptide of the present invention has been introduced are expressly excluded from the definition of a host cell.

In an important aspect, the invention also provides a method for producing the TIGIT antibody of the present invention, which includes culturing a host cell (or cell-free expression system) containing a polynucleotide (e.g., expression vector) encoding the TIGIT antibody under conditions that allow expression of the TIGIT antibody and recovery expressed TIGIT antibody. This recombinant expression process can be used for large-scale production of TIGIT antibodies of this invention, including monoclonal antibodies intended for therapeutic use in humans. Suitable vectors, cell lines and manufacturing processes for large-scale production of recombinant antibodies suitable for therapeutic use in vivo are generally available in the art and will be well known to one skilled in the art.

Therefore, in accordance with the present invention, there is provided an isolated polynucleotide or a combination of isolated polynucleotides encoding an antibody or antigen binding fragment comprising a combination of HCDR1, HCDR2, HCDR3, LCDR1, LCDR2 and LCDR3, which combination is selected from the group consisting of:

(i) HCDR1 containing SEQ ID NO: 16, HCDR2 containing SEQ ID NO: 17, HCDR3 containing SEQ ID NO: 18, LCDR1 containing SEQ ID NO: 61, LCDR2 containing SEQ ID NO: 62, and LCDR3 containing SEQ ID NO: 63;

(ii) HCDR1 containing SEQ ID NO: 4, HCDR2 containing SEQ ID NO: 5, HCDR3 containing SEQ ID NO: 6, LCDR1 containing SEQ ID NO: 49, LCDR2 containing SEQ ID NO: 50, and LCDR3 containing SEQ ID NO: 51;

(iii) HCDR1 containing SEQ ID NO: 7, HCDR2 containing SEQ ID NO: 8, HCDR3 containing SEQ ID NO: 9, LCDR1 containing SEQ ID NO: 52, LCDR2 containing SEQ ID NO: 53, and LCDR3 containing SEQ ID NO: 54;

(iv) HCDR1 containing SEQ ID NO: 10, HCDR2 containing SEQ ID NO: 11, HCDR3 containing SEQ ID NO: 12, LCDR1 containing SEQ ID NO: 55, LCDR2 containing SEQ ID NO: 56, and LCDR3 containing SEQ ID NO: 57;

(v) HCDR1 containing SEQ ID NO: 13, HCDR2 containing SEQ ID NO: 14, HCDR3 containing SEQ ID NO: 15, LCDR1 containing SEQ ID NO: 58, LCDR2 containing SEQ ID NO: 59, and LCDR3 containing SEQ ID NO: 60;

(vi) HCDR1 containing SEQ ID NO: 1, HCDR2 containing SEQ ID NO: 2, HCDR3 containing SEQ ID NO: 3, LCDR1 containing SEQ ID NO: 46, LCDR2 containing SEQ ID NO: 47, and LCDR3 containing SEQ ID NO: 48;

(vii) HCDR1 containing SEQ ID NO: 19, HCDR2 containing SEQ ID NO: 20, HCDR3 containing SEQ ID NO: 21, LCDR1 containing SEQ ID NO: 64, LCDR2 containing SEQ ID NO: 65, and LCDR3 containing SEQ ID NO: 66;

(viii) HCDR1 containing SEQ ID NO: 22, HCDR2 containing SEQ ID NO: 23, HCDR3 containing SEQ ID NO: 24, LCDR1 containing SEQ ID NO: 67, LCDR2 containing SEQ ID NO: 68, and LCDR3 containing SEQ ID NO: 69;

(ix) HCDR1 containing SEQ ID NO: 25, HCDR2 containing SEQ ID NO: 26, HCDR3 containing SEQ ID NO: 27, LCDR1 containing SEQ ID NO: 70, LCDR2 containing SEQ ID NO: 71, and LCDR3 containing SEQ ID NO: 72;

(x) HCDR1 containing SEQ ID NO: 28, HCDR2 containing SEQ ID NO: 29, HCDR3 containing SEQ ID NO: 30, LCDR1 containing SEQ ID NO: 73, LCDR2 containing SEQ ID NO: 74, and LCDR3 containing SEQ ID NO: 75;

(xi) HCDR1 containing SEQ ID NO: 31, HCDR2 containing SEQ ID NO: 32, HCDR3 containing SEQ ID NO: 33, LCDR1 containing SEQ ID NO: 76, LCDR2 containing SEQ ID NO: 77, and LCDR3 containing SEQ ID NO: 78;

(xii) HCDR1 containing SEQ ID NO: 34, HCDR2 containing SEQ ID NO: 35, HCDR3 containing SEQ ID NO: 36, LCDR1 containing SEQ ID NO: 79, LCDR2 containing SEQ ID NO: 80, and LCDR3 containing SEQ ID NO: 81;

(xiii) HCDR1 containing SEQ ID NO: 37, HCDR2 containing SEQ ID NO: 38, HCDR3 containing SEQ ID NO: 39, LCDR1 containing SEQ ID NO: 82, LCDR2 containing SEQ ID NO: 83, and

- 27 045372

LCDR3 containing SEQ ID NO: 84;

(xiv) HCDR1 containing SEQ ID NO: 40, HCDR2 containing SEQ ID NO: 41, HCDR3 containing SEQ ID NO: 42, LCDR1 containing SEQ ID NO: 85, LCDR2 containing SEQ ID NO: 86, and

LCDR3 containing SEQ ID NO: 87;

(xv) HCDR1 containing SEQ ID NO: 43, HCDR2 containing SEQ ID NO: 44, HCDR3 containing SEQ ID NO: 45, LCDR1 containing SEQ ID NO: 88, LCDR2 containing SEQ ID NO: 89, and LCDR3 containing SEQ ID NO: 90;

(xvi) HCDR1 containing SEQ ID NO: 271, HCDR2 containing SEQ ID NO: 272, HCDR3 containing SEQ ID NO: 273, LCDR1 containing SEQ ID NO: 283, LCDR2 containing SEQ ID NO: 284, and LCDR3 containing SEQ ID NO: 285;

(xvii) HCDR1 containing SEQ ID NO: 274, HCDR2 containing SEQ ID NO: 275, HCDR3 containing SEQ ID NO: 276, LCDR1 containing SEQ ID NO: 286, LCDR2 containing SEQ ID NO: 287, and LCDR3 containing SEQ ID NO: 288;

(xviii) HCDR1 containing SEQ ID NO: 277, HCDR2 containing SEQ ID NO: 278, HCDR3 containing SEQ ID NO: 279, LCDR1 containing SEQ ID NO: 289, LCDR2 containing SEQ ID NO: 290, and LCDR3 containing SEQ ID NO: 291.

In some embodiments, the present invention provides an isolated polynucleotide or combination of isolated polynucleotides encoding an antibody or antigen binding fragment comprising a combination of HCDR1, HCDR2, HCDR3, LCDR1, LCDR2 and LCDR3, wherein:

(i) HCDR1 contains or consists of SEQ ID NO: 16, HCDR2 contains or consists of SEQ ID NO: 17, HCDR3 contains or consists of SEQ ID NO: 18, LCDR1 contains or consists of SEQ ID NO: 61, LCDR2 contains or consists of consists of SEQ ID NO: 62, and LCDR3 contains or consists of SEQ ID NO: 63.

Also provided in accordance with this invention is an isolated polynucleotide or combination of isolated polynucleotides encoding an antibody or antigen binding fragment described herein. In certain embodiments of the present invention, an isolated polynucleotide encoding antibody 31282 provided herein, or an antigen binding fragment thereof, is provided.

Also provided in accordance with this invention is an isolated polynucleotide encoding the VH and/or VL domain of an anti-TIGIT antibody, wherein the polynucleotide contains one or more sequences selected from the group consisting of SEQ ID NOs: 241-270, 335-342 and 369 -370. In certain embodiments of the present invention, the isolated polynucleotide comprises the sequence of SEQ ID NO: 251 and/or the sequence of SEQ ID NO: 252. In certain embodiments of the present invention, the polynucleotide includes the sequence of SEQ ID NO: 251 and the sequence according to SEQ ID NO: 252 sequences are contiguous. In certain embodiments of the present invention, wherein the polynucleotide includes a sequence according to SEQ ID NO: 251 and a sequence according to SEQ ID NO: 252, the sequences are not contiguous.

The present invention further provides an expression vector comprising a polynucleotide of the invention operably linked to regulatory sequences that allow expression of an antigen binding polypeptide in a host cell or cell-free expression system.

Additionally, the present invention provides a host cell or cell-free expression system containing an expression vector in accordance with the present invention.

The present invention further provides a method for producing a recombinant antibody or antigen binding fragment thereof, which comprises culturing a host cell or cell-free expression system of the present invention under conditions that allow expression of the antibody or antigen binding fragment, and reconstituting the expressed antibody or antigen binding fragment.

Pharmaceutical compositions

Also provided herein are pharmaceutical compositions containing an antibody or antigen binding fragment of the present invention formulated with one or more pharmaceutically acceptable carriers or excipients. Such compositions may include one or a combination (eg, two or more different) TIGIT antibodies. Methods for preparing antibodies for therapeutic use in humans are well known in the art and are discussed, for example, in Wang et al., Journal of Pharmaceutical Sciences, Vol. 96, pp1-26, 2007.

The TIGIT antibodies and pharmaceutical compositions provided herein have therapeutic use, particularly in the therapeutic treatment of diseases, in particular conditions that benefit from inhibition of TIGIT function.

Combination Products

As demonstrated herein, the antibodies of this invention or antigen binding fragments thereof are particularly effective when administered in combination with immune checkpoint inhibitors, particularly anti-ICOS antagonist antibodies or anti-PD-1 antibodies (then

- 28 045372 are antagonist antibodies specific for the human immunoregulatory molecule PD-1).

Administration of anti-TIGIT antibody in combination with anti-ICOS or anti-PD-1 antibody results in a synergistic reduction in tumor growth compared with antibody alone. It is expected that similar effects will be observed when using a combination of the anti-TIGIT antibody of this invention and the anti-PD-L1 antibody.

This document also demonstrates that the antibodies of this invention or antigen binding fragments thereof are particularly effective when administered in combination with an antibody agonist specific for a co-stimulatory immune checkpoint molecule, in particular an anti-4-1BB, anti-OX40 or anti-GITR antibody. agonist Administration of anti-TIGIT antibodies in combination with anti-4-1BB, anti-OX40 or anti-GITR agonist antibodies results in a synergistic reduction in tumor growth compared with antibody alone.

In a further aspect, a combination product is provided comprising an anti-TIGIT antibody or an antigen-binding fragment thereof and one or more of a chemotherapeutic agent, an anti-PD1 antibody, an anti-PD-L1 antibody, an anti-41BB antibody, an anti-OX40 antibody, an anti-GITR antibody, and an -ICOS antibodies. In some preferred embodiments of the present invention, the anti-TIGIT antibody or antigen binding fragment thereof is an antibody or antigen binding fragment provided in accordance with the present invention. In the most preferred embodiment of the invention, the anti-TIGIT antibody or antigen binding fragment thereof comprises a combination of HCDR1, HCDR2, HCDR3, LCDR1, LCDR2 and LCDR3, wherein:

HCDR1 contains SEQ ID NO: 16 (YTFTSYYMH),

HCDR2 contains SEQ ID NO: 17 (VIGPSGASTSYAQKFQG),

HCDR3 contains SEQ ID NO: 18 (aRDHSDYWSGIMEV),

LCDR1 contains SEQ ID NO: 61 (RASQSVRSSYLA),

LCDR2 contains SEQ ID NO: 62 (GASSRAT) and

LCDR3 contains SEQ ID NO: 63 (QQYFSPPWT).

Also provided is the combination provided herein for use in a method of treating cancer or a viral infection, optionally wherein the viral infection is a CMV infection. The following is provided the combination presented herein for use in the method presented herein.

As used herein, when two or more active agents are presented as a combination, therapeutic combination, or combination therapy (the terms are used interchangeably), this does not require or exclude that the active agents are present in the same formulation. Combination therapy has its traditional interpretation of two or more active agents that must be administered so that the patient can benefit from each agent. Combination therapy does not require co-formulation, co-administration, simultaneous administration, or fixed-dose administration.

Therapeutic methods

TIGIT antibodies or antigen binding fragments thereof and pharmaceutical compositions provided herein can be used to inhibit the growth of cancerous tumor cells in vivo and are therefore useful in the treatment of tumors.

Accordingly, additional aspects of the invention relate to methods of inhibiting the growth of tumor cells in a human patient, as well as methods of treating or preventing cancer, which include administering to a patient in need thereof an effective amount of a TIGIT antibody or antigen binding fragment as a pharmaceutical composition described herein , as described herein, or combinations as described herein.

Another aspect of the invention relates to a TIGIT antibody or an antigen binding fragment thereof, as described herein, for use in inhibiting the growth of tumor cells in a human patient. Another aspect of the invention relates to a TIGIT antibody or an antigen binding fragment thereof, as described herein, for use in the treatment or prevention of cancer in a human patient.

In another aspect, the present invention provides a method for selectively depleting Treg cells in a cancer patient, comprising administering to the patient an anti-TIGIT antibody or an antigen binding fragment thereof. In some embodiments of the present invention, the anti-TIGIT antibody binds to a human TIGIT epitope containing residues Q56, N58, E60, I68, L73, H76, and I109, preferably consisting of residues Q56, N58, E60, I68, L73, H76, and I109 . In certain embodiments of the present invention, the anti-TIGIT antibody is an anti-TIGIT antibody provided herein.

In certain embodiments of the present invention, the anti-TIGIT antibody comprises a combination of HCDR1, HCDR2, HCDR3, LCDR1, LCDR2, and LCDR3, where: HCDR1 contains or consists of SEQ ID NO: 16 (YTFTSYYMH), HCDR2 contains or consists of SEQ ID NO: 17 (VIGPSGASTSYAQKFQG), HCDR3 contains or consists of SEQ ID NO: 18 (ARDHSDYWSGIMEV), LCDR1 contains or consists of SEQ ID NO: 61 (RASQSVRSSYLA), LCDR2 contains or consists of SEQ ID NO: 62

- 29 045372 (GASSRAT), and LCDR3 contains or consists of SEQ ID NO: 63 (QQYFSPPWT).

In some preferred embodiments of the present invention, the patient to be treated has a cancer selected from: kidney cancer (eg, renal cell carcinoma), breast cancer, brain tumors, chronic or acute leukemia, including acute myeloid leukemia, chronic myeloid leukemia, acute lymphoblastic leukemia, chronic lymphocytic leukemia, lymphomas (eg, Hodgkin and non-Hodgkin lymphoma, lymphocytic lymphoma, primary CNS lymphoma, B-cell lymphoma (eg, CLL), T-cell lymphoma (eg, Sezary syndrome)), cancer nasopharynx, melanoma (eg, metastatic malignant melanoma), prostate cancer, colon cancer, lung cancer, bone cancer, pancreatic cancer, skin cancer, head or neck cancer (eg, head and neck squamous cell carcinoma (HNSCC)), cancer skin, malignant melanoma of the skin or intraocular canal, uterine cancer, ovarian cancer, rectal cancer, anal cancer, stomach cancer, testicular cancer, uterine cancer, fallopian tube cancer, endometrial cancer, cervical cancer, vaginal cancer, vulvar cancer, cancer esophagus, small intestine cancer, endocrine cancer, thyroid cancer, parathyroid cancer, adrenal cancer, soft tissue sarcoma, urethral cancer, penile cancer, childhood solid tumors, bladder cancer, kidney or ureter cancer, renal pelvis cancer , neoplasms of the central nervous system (CNS), tumor angiogenesis, spinal axis tumors, brain stem glioma, pituitary adenoma, Kaposi's sarcoma, epidermoid cancer, squamous cell carcinoma, mesothelioma. In some embodiments of the present invention, the inhibited cancer is lung cancer, bladder cancer, breast cancer, kidney cancer (eg, kidney cancer), head and neck cancer (eg, HNSCC), or colon cancer (eg, colon adenocarcinoma ). In certain embodiments of the present invention, the cancer is colon cancer (eg, colon adenocarcinoma) or lung cancer. In certain embodiments of the present invention, the cancer is a blood cancer. In some such embodiments of the present invention, the cancer is lymphoma. In some embodiments of the present invention, the cancer is a T-cell lymphoma or a B-cell lymphoma.

In some embodiments of the present invention, the method of treating cancer further includes administering an additional therapeutic agent, such as a chemotherapeutic agent.

As demonstrated herein, the antibodies of the present invention or antigen-binding fragments thereof are particularly effective when administered in combination with immune checkpoint inhibitors, particularly anti-ICOS antagonist antibodies or anti-PD-1 antibodies (i.e., antagonist antibodies specific for human immunoregulatory molecule PD-1). Administration of anti-TIGIT antibody in combination with anti-ICOS or anti-PD-1 antibody results in a synergistic reduction in tumor growth compared with antibody alone. It is expected that similar effects will be observed when using a combination of the anti-TIGIT antibody of this invention and the anti-PD-L1 antibody.

This document also demonstrates that the antibodies of this invention or antigen binding fragments thereof are particularly effective when administered in combination with an antibody agonist specific for a co-stimulatory immune checkpoint molecule, in particular an anti-4-1BB, anti-OX40 or anti-GITR antibody. agonist Administration of anti-TIGIT antibodies in combination with anti-4-1BB, anti-OX40 or anti-GITR agonist antibodies results in a synergistic reduction in tumor growth compared with antibody alone.

Accordingly, this document also provides a method of treating cancer in a subject, comprising administering to the subject an effective amount of an anti-TIGIT antibody or an antigen binding fragment thereof according to the present invention, as well as administering an effective amount of an anti-PD-1 antibody, an anti-PD-L1 antibody, an -41BB antibodies, anti-OX40 antibodies and anti-GITR antibodies or anti-ICOS antibodies.

In addition, the data presented herein demonstrating that anti-TIGIT antibodies can enhance the activity of γδ cells as well as conventional T cells indicate that anti-TIGIT antibodies may be used to treat conditions other than cancer. In particular, γδ T cells are known to be important in the response to infection, such as bacterial, fungal or viral infection. As demonstrated in Example 29, when exposed to anti-TIGIT antibody, γδ T cells from CMV seropositive subjects exhibit markedly increased activation, characterized by an increase in IFNg shear. The ability to thus stimulate γδ T cell activation in CMV patients suggests that administration of an anti-TIGIT antibody will stimulate γδ T cell antiviral activity.

Accordingly, provided herein is a method of treating a viral infection in a subject, comprising administering an effective amount of an anti-TIGIT antibody or antigen binding fragment thereof. Also provided is a method of treating a viral infection in a subject, comprising administering an effective amount of an anti-TIGIT antibody or antigen binding fragment or a pharmaceutical composition provided herein to the subject, thereby treating the viral infection. In preferred embodiments of the present invention, the viral infection is a CMV infection.

In certain embodiments of the present invention, the method further includes administering one or more additional therapeutic agents. In certain embodiments of the present invention, one or more therapeutic agents is selected from: anti-PD1 antibody, anti-PD-L1 antibody, anti-41BB antibody, anti-OX40 antibody, anti-GITR antibody, and anti-ICOS antibody.

As demonstrated in the examples, the anti-TIGIT antibodies described herein are effective in stimulating T cell activity, especially pro-inflammatory T cell activity. Total T cell activity can be measured by methods known to those skilled in the art, for example, by measuring IFNg production as described in the Examples.

Accordingly, also provided herein is a method of promoting T cell activity, comprising contacting a population of T cells with an antibody or antigen binding fragment as described herein.

In certain embodiments of the present invention, the method of stimulating T cell activity is carried out in vitro. In certain embodiments of the present invention, the method of stimulating T cell activity is carried out in vivo in a human. In some such embodiments of the present invention, the human subject has cancer. In certain embodiments of the present invention, the human subject has a viral infection, such as a CMV infection.

In certain embodiments of the present invention, the method stimulates normal αβ T cell activity. In certain embodiments of the present invention, the method stimulates the activity of CD4 T cells. In certain embodiments of the present invention, the method stimulates the activity of CD8 T cells. In certain embodiments of the present invention, the method stimulates the activity of γδ (gamma delta) T cells.

In addition, the examples demonstrate that the anti-TIGIT antibodies described herein will be particularly effective in stimulating T cell activity when used in combination with an anti-PD1 antibody, an anti-PD-L1 antibody, an anti-41BB antibody, an -OX40 antibody, anti-GITR antibody or anti-ICOS antibody. Importantly, the combination provides a synergistic (i.e., greater than additive) increase in T cell activity.

Accordingly, in some embodiments of the present invention, the method of promoting T cell activity further comprises contacting a population of T cells with one or more of: anti-PD1 antibody, anti-PD-L1 antibody, anti-41BB antibody, anti-OX40 antibodies, anti-GITR antibodies and anti-ICOS antibodies.

Variants and equivalents of the embodiments of the invention described herein, but not departing from the spirit and scope of the invention, will be known to one skilled in the art. The invention will be better understood with reference to the following non-limiting examples.

Examples

Example 1. Selection of TIGIT antigen-binding proteins.

TIGIT ABPs were selected from a synthetic library of human antibodies expressed and presented on the surface of yeast cells in IgG format, typically as described, for example, in WO 2009036379; WO 2010105256; WO 2012009568; and Xu et al., Protein Eng Des Sel., Vol. 26 (10), pp. 663-670 (2013)), and more specifically as indicated below. The sequences and characteristics of ABPs isolated from the recombinant library are presented in FIG. 1-6.

Eight naive human synthetic yeast libraries, each of ~10 ⁹ diversity, were propagated as described previously (see, for example: Xu et al, 2013; WO 2009036379; WO 2010105256; and WO 2012009568). For the first two rounds of selection, a magnetic bead sorting technique was performed using the Miltenyi MACS system as described (see, e.g., Siegel et al., 2004). Briefly, yeast cells (~10 ¹⁰ cells/library) were incubated with biotinylated TIGIT-Fc antigen (Creative Biomart) in FACS wash buffer (phosphate-buffered saline (PBS)/0.1% bovine serum albumin (BSA)). After washing once with 50 ml of ice-cold wash buffer, the cell pellet was resuspended in 40 ml of wash buffer and streptavidin microbeads (500 μl) were added to the yeast and incubated for 15 min at 4°C. The yeast was then pelleted, resuspended in 5 ml wash buffer and loaded onto a Miltenyi LS column. After loading 5 ml, the column was washed 3 times with 3 ml FACS wash buffer. The column was then removed from the magnetic field and the yeast was eluted with 5 ml of growth medium and then grown overnight. The following rounds of sorting were performed using flow cytometry. Approximately 1 x 10 ⁸ yeast were pelleted, washed three times with wash buffer and incubated with biotinylated TIGIT-Fc fusion antigen (10 nM) under equilibrium conditions at room temperature. The yeast was then washed twice and stained with LC-FITC (diluted 1:100) and either SA-633 (diluted 1:500) or EA-31 045372

PE (diluted 1:50) with secondary reagents for 15 minutes at 4°C. After washing twice with ice-cold wash buffer, cell pellets were resuspended in 0.4 ml wash buffer and transferred to filter-filled cap tubes. Sorting was performed using a FACS ARIA sorter (BD Biosciences) and sorting gates were assigned to select specific binders against the background control. Subsequent rounds of selection were used to reduce the number of non-specific binders using soluble membrane proteins from CHO cells (see, for example, WO 2014179363 and Xu et al., Protein Eng Des Sel, Vol. 26 (10), pp. 663 -670 (2013)) and identify binders with improved affinity for TIGIT using the TIGIT-Fc antigen. After the final round of sorting, yeasts were plated and individual colonies were selected for characterization and assignment of clones for affinity maturation. 63 clones were tested for functional activity. From the screen, clones 26518, 26452, 26486, 26521 and 26493 had the best functional activity and were selected for further optimization.

Example 2 Antibody Optimization.

Optimization of naive clones was carried out using three maturation strategies: light chain diversification; diversification of HCDR1 and HCDR2; and diversification of HCDR3 into selected diverse pools of HCDR1 and HCDR2.

Light chain diversification: Heavy chain variable regions were isolated from naive outputs (described above) and transformed into a light chain library with 1x10 ^b diversity. Screenings were performed as described above, with one round of MACS sorting and three rounds of FACS sorting using 10 nM or 1 nM biotinylated antigen TIGIT-HIS (Creative Biomart) for the corresponding rounds.

Selection of HCDR1 and HCDR2: HCDR3 from the clones selected by the light chain diversification procedure were recombined into a pre-prepared library with HCDR1 and HCDR2 variants with a diversity of 1x10 ⁸ and selection was performed using the monomeric HIS-TIGIT antigen. Affinity pressure was applied using decreasing concentrations of biotinylated HIS-TIGIT antigen (100 to 1 nM) under equilibrium conditions at room temperature.

HCDR3/HCDR1/HCDR2 selections: Oligos were ordered from IDTs that contained HCDR3 as well as a homologous flanking region on either side of HCDR3. Amino acid positions in HCDR3 varied through NNK diversity at two positions on the oligo throughout HCHR3. HCDR3 oligos were double-stranded using primers that annealed to the flanking region of HCDR3. The remaining heavy chain variable region FWR1-FWR3 were amplified from affinity-enhanced antibody pools that were isolated from the HCDR1 and HCDR2 varieties selected above. The library was then generated by transforming the double-stranded HCDR3 oligo, the combined FWR1 to FWR3 fragments, and the heavy chain expression vector into yeast already containing the light chain of the original naïve parent. Screenings were carried out as in previous rounds using FACS sorting for four rounds. For each FACS round, libraries were assessed for PSR binding, species cross-reactivity, and affinity pressure, and sorted to obtain populations with the desired characteristics. Affinity pressures for these selections were performed as described above in the HCDR1 and HCDR2 selections.

Example 3 Antibody production and purification.

A. Production in yeast.

To obtain sufficient quantities of optimized and non-optimized selected antibodies for further characterization, yeast clones were grown to saturation and then induced for 48 h at 30°C with shaking. After induction, yeast cells were pelleted and supernatants were collected for purification. IgG was purified using a protein A column and eluted with acetic acid, pH 2.0. Fab fragments were obtained by papain digestion and purified in two steps using Protein A (GE LifeSciences) and KappaSelect (GE Healthcare LifeSciences).

B. Production in mammalian cells.

To obtain sufficient quantities of optimized and non-optimized selected antibodies for further characterization, a DNA vector encoding specific antibody clones was generated and transduced into HEK cells. Codon-optimized synthetic human DNA fragments for antibody variable domains were ordered from Geneart. The variable domain sequences were smoothly ligated into pUPE expression vectors containing the mouse Igkappa signal sequence and constant regions of the corresponding class of antibodies. Expression vectors were confirmed by restriction analysis and DNA sequencing. For transient transfection, endotoxin-free DNA maxipreps were prepared (Sigma), and heavy and light chain vectors were cotransfected into HEK293EBNA1 cells in Freestyle medium (ThermoFisherScientific) according to established protocols. Primatone (0.55% final volume) was added 24 hours after transfection. Conditioned medium was collected 6 days after transfection. Antibodies were purified using batch processes using Mabselect sureLX affinity chromatography (GE Healthcare). Bound antibodies were washed in 2 steps with PBS containing 1 M NaCl and PBS.

- 32 045372

Antibodies were eluted with 20 mM citrate, 150 mM NaCl, pH 3, and neutralized to approximately pH 7 s

1/6 volume 1 MK ₂ HPO ₄ /KH ₂ PO ₄ , pH 8.

The antibodies were then further purified by gel filtration using a Superdex200 column equilibrated in PBS. Fractions were analyzed by NuPAGE and fractions containing antibodies were pooled. The final products were sterilized through a 0.22 μM syringe filter. The product was analyzed using NuPAGE and endotoxin levels were measured using the LAL assay.

Example 4. Determination of the binding affinity of anti-TIGIT antibodies to recombinant human TIGIT protein.

A. ForteBio KD measurements.

ForteBio affinity measurements for selected antibodies were performed generally as described previously (see, for example, Estep et al., Mabs, Vol. 5 (2), pp. 270-278 (2013)). Briefly, ForteBio affinity measurements were performed by loading IgG online onto AHQ sensors. The sensors were equilibrated offline in assay buffer for 30 min and then monitored online for 60 sec to establish a baseline. IgG-loaded sensors were exposed to 100 nM antigen (human TIGIT-Fc, human TIGIT-His, or human TIGIT-Fc) for 5 min and then transferred to assay buffer for 5 min to measure mismatch. Kinetics were analyzed using a 1:1 binding model. More than 90 antibodies were tested for affinity using ForteBio, and Tab. Figure 3 shows data for 15 selected anti-TIGIT antibodies demonstrating strong binding to recombinant TIGIT protein.

B. MSD-SET KD measurements.

Steady-state affinity measurements for selected antibodies were performed generally as described previously (Estep et al., Mabs, Vol. 5 (2), pp. 270-278 (2013)). Briefly, solution equilibrium titration (SET) was performed in PBS + 0.1% IgG-free BSA (PBSF) with antigen (TIGIT-His monomer) held constant at 10-100 pM and incubated with 3-5- multiple serial dilutions of Fab or mAb starting from 10 pM to 10 nM. Antibodies (20 nM in PBS) were applied to standard MSD-ECL binding plates overnight at 4°C or at room temperature for 30 min. The plates were then blocked with BSA for 30 min with shaking at 700 rpm, followed by three washes with wash buffer (PBSF + 0.05% Tween 20). SET samples were applied and incubated on the plates for 150 s with shaking at 700 rpm, followed by one wash. The antigen captured on the plate was detected with sulfotag-labeled streptavidin at a concentration of 250 ng/ml in PBSF by incubating on the plate for 3 min. The plates were washed three times with wash buffer and then read on an MSD Sector Imager 2400 using 1x Reading Buffer T with surfactant. The percentage of free antigen was plotted as a function of titrated antibody in Prism and fitted to a quadratic equation to extract the KD. Liquid handling robots were used to improve throughput during MSD-SET experiments, including SET sample preparation. The selected antibodies were tested for affinity using MSD, and Table. Figure 4 shows data for 7 anti-TIGIT clones demonstrating strong binding to recombinant TIGIT protein.

Table 4

MSD affinity analysis for selected anti-TIGIT antibodies

Clone Affinity MSD Monovalant KD (M)TIGIT- His man 29489 1DE-10 29494 7.0E-11 29499 1.9E-11 29513 2.5E-11 29520 2DE-10 29523 1.7E-09 29527 6.4E-10

C. Biacore measurement.

Biosensor analysis was performed at 25°C in the HBS-EP buffer system (10 mM HEPES, pH 7.3, 150 mM NaCl, 3 mM EDTA, 0.05% P20 surfactant) using a Biacore 8K optical biosensor coupled to a sensor chip CM5 (GE Healthcare, Marlborough, MA). The hotel sample was kept at 8°C. Goat anti-human IgG antibody (Fcy fragment specific, Jackson ImmunoResearch Laboratories, Inc., West Grove, PA; 109-005098) was immobilized (11,700 +/- 200 RU) in both flow cells of the sensor chip using

- 33 045372 standard chemistry combination of amines. This type of surface provided a format for reproducible capture of fresh assay antibody after each regeneration step. Flow cell 2 was used to assay captured antibody (60-90 RU), while flow cell 1 was used as a control flow cell. Antigen concentrations ranging from 30 to 0.123 nM (3-fold dilutions) were prepared in electrode buffer. Each antigen sample concentration was run as a single replicate. Two empty (buffer) injections were also run and used to evaluate and subtract system artifacts. Association (300 s) and dissociation (600 s) phases for all antigen concentrations were performed at a flow rate of 30 μl/min. The surface was regenerated with three sequential injections (15 s, 15 s, and 60 s) of 10 mM glycine, pH 1.5, at a flow rate of 30 μL/min. Data were aligned, double-referenced, and fit a 1:1 anchoring model using Biacore 8K evaluation software, version 1.0. The selected antibodies were tested for affinity using Biacore, and Table. Figure 5 shows data for 5 anti-TIGIT clones demonstrating strong binding to recombinant TIGIT protein.

Table 5

Biacore affinity analysis for selected anti-TIGIT antibodies

Clone Biacore: Kd monovalent (M) (IgG on CM5 chip, human TIGIT-HIS in solution (initial concentration 25 nM, 3-fold dilution) 29489 2.48E-10 31282 2.94E-10 29494 2.70E-10 29520 7.16E-10 29527 1.20E-09 31288 1.92E-10

Example 5: Analysis of competition between antagonistic anti-TIGIT antibodies and natural TIGIT ligands.

A. Octet Red384 epitope-specific ligand sorting/blocking.

Epitope-specific sorting/ligand blocking of selected antibodies was performed using a standard sandwich crossover assay. Anti-target control IgG was loaded onto AHQ sensors, and unoccupied Fc-binding sites on the sensor were blocked with an irrelevant human IgG1 antibody. The sensors were then exposed to 100 nM target antigen (hTIGIT, Creative Biomart) followed by the addition of a second anti-target antibody or ligand (anti-TIGIT antibody and CD155 or CD113 or CD112). Data were processed using ForteBio data analysis software 7.0. Additional binding by a second antibody or ligand after antigen association indicates an unoccupied epitope (non-competitor), whereas lack of binding indicates blocking of the epitope (competitor or ligand blocking). Parental antibodies (before optimization) were tested for competition with natural ligands, and in table. Figure 6 shows data obtained for competition with CD155, CD112 and CD113. Parental clone 26432 was found to not compete with CD155 for TIGIT binding. All other selected anti-TIGIT antibodies compete with the natural ligand for binding to the recombinant human TIGIT protein.

Table 6

Sorting assay against natural TIGIT ligands for non-optimized anti-TIGIT antibodies

26518 Yes Yes Yes 26452 Yes Yes Yes 26486 Yes Yes Yes 26521 Yes Yes Yes 26493 Yes Yes Yes 26432 No

B. Competition of antagonistic anti-TIGIT antibodies with CD155 on Jurkat-hTIGIT.

Jurkat cells overexpressing human TIGIT (Jurkat-hTIGIT) were collected and distributed at 10 ⁵ cells/well and incubated with anti-human TIGIT antibodies at the following concentrations: 166.6; 53.24; 17.01; 5.43; 1.73; 0.55; 0.17; 0.05; 0.01; 5.78x10 ^-3 ; 1.85x10 ^-3 ; 5.9x10 ^-3 nM in complete medium for 45 min at 37°C. Excess antibodies were washed and then cells were incubated with CD155-His at 5 μg/ml (Creative Biomart, PVR-3141H) for 45 minutes at 37°C. Bound CD155-His was then detected using anti-His tag-PE (Biolegend, 362603, 2 µl per test), incubated for 30 min

- 34 045372 at 4°C. Cells were analyzed by FACS using a BD LSRFortessa, and the half concentration (IC ₅₀ ) that prevents CD155 binding was calculated based on the geometric mean fluorescence.

The results were as follows: 0.101 nM for clone 29489; 0.07 nM for clone 29494; 0.102 nM for clone 29520 and 0.078 nM for clone 29527, for the results illustrated in FIG. 7. The values of other antibodies tested are summarized in table. 7 below. Overall, the results demonstrate strong competition between the tested antagonistic anti-TIGIT antibodies and CD155 for binding to the TIGIT-expressed membrane.

Table 7

IC ₅₀ data for CD155 competition on human TIGIT

Example 6: Characterization of hydrophobic interaction chromatography (MAbs.2015 May-Jun; 7(3):553-561).

Anti-TIGIT IgG1 antibody samples were buffer exchanged to 1 M ammonium sulfate and 0.1 M sodium phosphate at pH 6.5 using a Zeba 40 kDa 0.5 mL spin column (Thermo Pierce, cat #87766). A salt gradient was installed on a Dionex ProPac HIC-10 column of 1.8 M ammonium sulfate, 0.1 M sodium phosphate at pH 6.5 to the same conditions as without ammonium sulfate. The gradient was run for 17 min at a flow rate of 0.75 mL/min. An acetonitrile wash step was added at the end of the run to remove remaining protein, and the column was re-equilibrated with 7 column volumes before the next injection cycle. Peak retention time was monitored at A280 absorbance, and ammonium sulfate elution concentrations were calculated based on the gradient and flow rate. In table Figure 8 shows the results obtained for 15 selected anti-TIGIT antibodies.

Table 8

Hydrophobic interaction chromatography analysis of selected anti-TIGIT antibodies

Example 7. Characteristics of the PSR preparation of a polyspecific reagent.

- 35 045372

A. Preparation of a polyspecific reagent.

Polyspecific reagent (PSR) was prepared according to Xu et.al, mAbs 2013. Briefly, 2.5 L of CHO-S cells were used as starting material. Cells were pelleted at 2400xg for 5 min in 500 ml centrifuge bottles filled to 400 ml. Cell pellets were pooled and then resuspended in 25 ml buffer B and pelleted at 2400xg for 3 min. The buffer was decanted and washing was repeated once. Cell pellets were resuspended in 3x pellet volume with buffer B containing 1x protease inhibitor (Roche, cOmplete, no EDHA) using a polytron homogenizer with cells maintained on ice. The homogenate was then centrifuged at 2400xg for 5 min, and the supernatant was retained and pelleted one more time (2400xg/5 min) to ensure removal of unbroken cells, cell debris and nuclei; the resulting supernatant is a total protein preparation. The supernatant was then transferred to two 45 ml Nalgene Oak Ridge centrifuge tubes and pelleted at 40,000xg for 40 min at 4°C. Supernatants containing individual cytosolic proteins (SCPs) were then transferred to clean Oak Ridge tubes and centrifuged at 40,000xg again. In parallel, pellets containing the membrane fraction (EMF) were retained and centrifuged at 40,000 for 20 min to remove residual supernatant. The EMF beads were then washed with Buffer B. 8 mL of Buffer B was then added to the membrane pellets to move the pellets and transfer them to the Dounce homogenizer. Once the pellets were homogenized, they were transferred to a 50 mL conical tube to represent the final EMF preparation.

One billion mammalian cells (eg CHO, HEK293, Sf9) at ~10 ⁶ -10 ⁷ cells/ml were transferred from tissue culture medium into 4x250 ml conical tubes and pelleted at 550xg for 3 minutes. All subsequent steps were performed at 4°C or on ice with ice-cold buffers. Cells were washed with 100 ml PBSF (1x PBS + 1 mg/ml BSA) and pooled into one conical tube. After removal of the supernatant, the cell pellet was then resuspended in 30 ml of buffer B (50 mM HEPES, 0.15 M NaCl, 2 mM _CaCl2 , 5 mM KCl, 5 mM _MgCl2 , 10% glycerol, pH 7.2) and pelleted at 550xg within 3 min. Buffer B supernatant was decanted and cells were resuspended in 3x volume of Buffer B pellets plus 2.5χ protease inhibitor (Roche, cOmplete, no EDTA). Protease inhibitors in buffer B were included next. Cells were homogenized four times in 30-second pulses (Polyton homogenizer, PT1200E) and the membrane fraction was pelleted at 40,000xg for 1 hour at 4°C. The precipitate was washed with 1 ml of buffer B; the supernatant is retained and is s. The pellet is transferred to a Dounce homogenizer with 3 ml of buffer B and resuspended by slowly moving the pestle up and down with 30-35 strokes. The enriched membrane fraction (EMF) is transferred to a new collection tube, rinsing the pestle to collect all potential protein. Determine the protein concentration of purified EMF using the Dc Protein Assay Kit (BioRad). To solubilize EMF, transfer to solubilization buffer (50 mM HEPES, 0.15 M NaCl, 2 mM CaCl2, 5 mM KCl, 5 mM _MgCl2 , 1% n-dodecyl-b-Dmaltopyranoside (DDM), 1x protease inhibitor, pH 7.2) to a final concentration of 1 mg/ml. Spin the mixture overnight at 4°C followed by centrifugation in a 50 mL Oak Ridge tube (Fisher Scientific, 050529-ID) at 40,000xg for 1 hour. The supernatant, which represents soluble membrane proteins (SMPs), was collected and Protein yield was quantified as described above.

For biotinylation, an NHS-LC-biotin stock solution was prepared according to the manufacturer's protocol (Pierce, Thermo Fisher). Briefly, 20 μl of biotin reagent was added to each 1 mg of EMF sample and incubated at 4°C for 3 h with gentle mixing. Adjust the volume to 25 ml with buffer B and transfer to an Oak Ridge centrifuge tube. Precipitate biotinylated EMF (b-EMF) at 40,000xg for 1 h and wash twice with 3 ml of buffer C (buffer B minus glycerol) without disturbing the pellet. Remove the remaining solution. The pellet was resuspended using a Dounce homogenizer in 3 ml of buffer C as previously described. The re-suspended pellet is now biotinylated EMF (b-EMF) and is dissolved as described above to prepare b-SMP.

B. PSR Binding Assay.

PSR assays were performed generally as described in WO 2014/179363. Briefly, to characterize the PSR profile of monoclonal antibodies presented on yeast, two million IgG-presenting yeasts were transferred to a 96-well assay plate and pelleted at 3000xg for 3 minutes to remove the supernatant. The sediment was resuspended in 50 μl of a freshly prepared 1:10 dilution of the original b-PSR and incubated on ice for 20 min. Cells were washed twice with 200 μl of cold PBSF and the pellet was resuspended in 50 μl of secondary labeling mixture (Extravidin-R-PE, anti-human LC-FITC and propidium iodide). The mixture was kept on ice for 20 minutes and then washed twice with 200 μl of ice-cold PBSF. Resuspend cells in 100 μl of ice-cold PBSF and run the plate on a FACS Canto (BD Biosciences) using an HTS sample injector. Flow cytometry data were analyzed for mean fluorescence intensity in the R-PE channel and normalized to provide adequate controls to assess nonspecific binding. In table 9 summarizes the results of polyspecific reagent binding obtained for 15 selected

- 36 045372 anti-TIGIT antibodies, which confirm the low score for most clones.

Table 9

Polyspecific reagent analysis

Clone Score (0-1) Polyspecific Reagent (PSR) 26518 29478 26452 29487 29489 26486 29494 29499 26521 29513 26493 29520 29523 0.00 0.01 0.00 0.01 0.01 0.00 0.00 0.10 0.00 0.01 0.00 0.32 0.12 29527 0.12 26432 0.00 31288 0.00 32919 0.00 32931 0.00 32959 0.1

Example 8. Characterization of TIGIT expression on immune populations from PBMCs of healthy people.

A. TIGIT expression profile on T cell subsets.

Flow cytometry analyzes were performed to evaluate TIGIT expression on immune cell subsets in PBMCs freshly isolated from healthy individuals. Conjugated antibodies were purchased from Ebioscience/Thermo Fisher Scientific, BioLegend, or BD Biosciences. Cells were stained according to the manufacturer's instructions using filtered FACS buffer (PBS + 2 mM EDHA + 0.1% BSA) and Brilliant Stain buffer (BD #563794). Cells were blocked with the appropriate human FcBlock (BD #564220) before staining and fixed using IC Fixation Buffer (eBioscience #00-8222-49) until prepared. Acquisition was performed on a FACS Fortessa (BD Biosciences) and analyzed using FlowJo software (FlowJo, LLC). Viable cells were gated on forward and side scatter. Different immune cell subsets were gated as follows: CD19+ (B cells), CD3 CD19 CD14+ (Monocytes), CD3+ TCRab (TCRgd T cells), CD3+ TCRab⁺ (TCRab T cells), CD3 CD19' CD14'HLA-DR CD56^low/βhigh (NK cells), CD3^- CD19’ CD14- HLA-DR+ (dendritic cells), CD3+ TCRab⁺ CD4+ CD127^short CD25+ (regulatory T cells), CD3+ TCRab⁺ CD4+ or CD8+ CD45RO CCR7+ (CD4 or CD8 naïve T cells), CD3+ TCRab⁺ CD4+ or CD8+ CD45RO+ (memory T cells) and CD45RO'CD62L'(effector T cells).

As shown in FIG. 8A and 8B, TIGIT is predominantly expressed on NK cells, regulatory T cells, and memory CD8 T cells. It is present to a lesser extent in other T cell subsets, with a low proportion of naïve T cells exhibiting TIGIT expression. In addition, TIGIT is not expressed on monocytes, dendritic cells, and B cells (Fig. 8B). This data set is consistent with published data (Yu et al. NI 2008 and Wang et al. EJI 2015).

Example 9: Cellular binding of antagonistic anti-TIGIT antibodies.

A. Binding of anti-TIGIT antibodies to Jurkat-hTIGIT and Jurkat-mTIGIT.

The affinity of human anti-TIGIT antibodies was measured using Jurkat E6.1 cells transduced with human TIGIT (Jurkat hTIGIT) or mouse TIGIT (JurkatmTIGIT). To analyze the affinity of the selected antibodies to hTIGIT or mTIGIT, 10 ⁵ cells were distributed per well and incubated with anti-TIGIT antibody at a single dose of 100 nM (Table 3) or with decreasing concentrations (166.6; 53.24; 17.01; 5.43; 1.73; 0.55; 0.17; 0.05; 0.01; 5.78x10 ^-3 ; 1.85x10 ^-3 ; 5.9x10' ³ nM) of selected antibodies (Fig. 9) . Antibodies were incubated with cells for 20 min at 4°C in FACS buffer. After washing, the cells were incubated with anti-human Ig (specific for Fcgamma)-PE (eBioscience, 12-4998-82, at 2.5 μg/ml) for 20 min on ice and washed twice.

- 37 045372

The geometric mean fluorescence intensity was analyzed using the LSR BD Fortessa. Cell binding was recorded as the average PE fluorescence intensity of the transfected line compared to the nontransfected line for each antibody (Table 3). To calculate the binding _EC50, the half maximum binding concentration ( _EC50 ) to hTIGITJurkat was calculated using a four-variable curve fitting equation in Prism and the resulting values were: 0.082 nM for clone 29489; 0.07 nM for clone 29494; 0.119 nM for clone 29520 and 0.05 nM for clone 29527 for the data illustrated in FIG. 9. Results demonstrate strong binding to membrane-expressed human TIGIT for the anti-TIGIT antibodies tested.

B. Binding of antagonistic anti-TIGIT antibodies to primary human T cells.

Isolated human PBMCs from healthy volunteers were assayed for binding using antagonistic anti-TIGIT antibodies. Cells were distributed at ^5x105 cells per well. Cells were incubated with anti-CD16 (clone 3G8, BioLegend 302002), CD32 (clone FLI8.26, BD Bioscience 557333) and CD64 (BD Bioscience 555525) at room temperature for 10 minutes, and the indicated human anti-TIGIT antibodies were directly added in final concentration: 12.65; 4; 1.26; 0.40; 0.126; 0.040; 0.12 and 4x10'3 nM in FACS buffer and incubated for 20 min at 4°C. After washing, cells were incubated with anti-human Ig (Fc-gamma-specific)-PE (eBioscience, 12-4998-82, at 2.5 μg/ml) for 20 minutes at 4°C. The cells were then washed and incubated with the following antibodies and LVD mixture for the results of Fig. 10A and 10B: anti-CD4-PercP-Cy5.5 (clone A161A1, BioLegend 357414); anti-CD8-BV510 (clone SK1, BD Bioscience 563919) and LVD efluor 520 (eBioscience 65-0867-14). For fig. 10C cells were washed and incubated with the following antibodies and LVD mixture: LVD efluor 520 (eBioscience 65-0867-14), anti-TCRab-PercP-Cy5.5 (clone IP26, Biolegend 306723), anti-CD4-BV510 (clone SK3, BD Horizon 562970), anti-CD8-APC-Cy7 (clone SK1, Biolegend 344714), anti-CD25-BV605 (clone 2A3, Biolegend 562660), anti-CD127-APC (A019D5, Biolegend 351316), anti-CCR7-BV421 (clone G043H7, Biolegend 353207) and anti-CD45RO' PE-Cy7 (clone UCHL1, Biolegend 304229).

The EC ₅₀ value for binding to human CD8+ primary T cells was calculated using the % positive TIGIT-stained cells on gated LVD'CD8 ⁺ T cells (FIGS. 10A and 10B).

The EC ₅₀ value for binding to memory CD8+ T cells or primary human Treg cells was calculated using % positive TIGIT stained cells on gated LVD'TCRab ⁺ CD45RO+CD8+ T cells (for memory CD8+ T cells) or on gated LVD' TCRab ⁺ CD127 ^lo CD25 ^hl CD4 ⁺ T cells (for Tregs) and are illustrated in Fig. 10C.

As shown in FIG. 10A, the EC ₅₀ value for binding to total human CD8+ T cells is 0.123 nM for clone 29489; 0.181 nM for clone 29520 and 0.253 nM for clone 29527. A direct comparison was made between 29489 and 31282 (29489 mutant with M to T mutation at residue 116) and the EC ₅₀ value was 0.057 nM and 0.086 nM, respectively, demonstrating strong and similar potency binding to primary human CD8+ T cells for 2 clones (Fig. 10B). The EC ₅₀ values obtained for binding to memory CD8+ T cells and Tregs were 0.039 nM and 0.03 nM, respectively, demonstrating strong and similar affinity for both populations (Figure 10C).

C. Binding of antagonistic anti-TIGIT antibodies to primary Javanese T cells.

Isolated PBMCs from Macaca flavicularis were obtained from BioPRIM. Cells were thawed and stimulated using a non-human primate T cell activation/expansion kit (Miltenyi Biotec) at a 1:2 ratio (bead:cell ratio) according to the manufacturer's instructions. The next day, the cells were collected, counted and distributed at 5x104 cells per well. Cells were incubated with anti-CD16 (clone 3G8, BioLegend 302002), CD32 (clone FLI8.26, BD Bioscience 557333) and CD64 (BD Bioscience 555525) at room temperature for 10 minutes, and selected human anti-TIGIT antibodies were directly added in final concentration: 12.65; 4; 1.26; 0.40; 0.126; 0.040; 0.12 and 4x10'3 nM in FACS buffer and incubated for 20 min at 4°C. After washing, the cells were incubated with anti-human Ig (Fc-gamma-specific)-PE (eBioscience, 12-4998-82, at 2.5 μg/ml) for 20 min at 4°C. The cells were then washed and incubated with the following antibodies and LVD mixture for the data illustrated in FIG. 11A and 11B: antiCD4-PercP-Cy5.5 (clone A161A1, BioLegend 357414); anti-CD8-BV510 (clone SK1, BD Bioscience 563919), CD69-APC-Cy7 (clone FN50, BioLegend, 310914), and LVD efluor 520 (eBioscience 65-0867-14). Stained cells were analyzed by FACS using a BD LSR Fortessa. The EC50 binding value was calculated using the % positive TIGIT-stained cells that were localized to LVD-CD69+CD8+ T cells. As shown in FIG. 11, EC ₅₀ values for binding to Javanese CD8+ T cells were 0.487 nM for clone 29489, 1.73 nM for clone 29520, and 0.378 nM for clone 29527. Clones 29489 and 31282 (29489 mutant with M to T mutation at residue 116) were also compared and the EC ₅₀ values were 0.25 nM and 0.26 nM, respectively, for the example shown in FIG. 11B, showing similar and strong affinity for primary Javanese CD8+ T cells for the 2 clones.

Example 10. In vitro functional characterization of antagonistic anti-TIGIT activity.

A. TIGIT bioassay for co-culture of CHO-TCR-CD155 and Jurkat-hTIGIT.

- 38 045372

To characterize the functional consequences of blocking the human TIGIT receptor, we cocultured Jurkat cells, which express hTIGIT and a luciferase reporter activated upon TCR uptake (Thaw-and-Use Format TIGIT Effector Cells from Promega), with the CHO-cell line. K1, designed to express CD155 and human TCR activator (thaw and use CD155 aAPC/CHO-K1 from Promega). Activation of TIGIT-overexpressing Jurkat cells can be induced by contact with CD155-expressing CHO-K1 cells upon TCR engagement in Jurkat cells and can be increased in the presence of an antagonistic anti-TIGIT antibody. To compare the effectiveness of different antibodies in increasing the activation of Jurkat cells, the experiment was performed in the presence of increasing concentrations of antibodies and EC ₅₀ values were calculated.

CD155 aAPC/CHO-K1 cells (Promega, CS198811) thaw-and-use format were seeded according to the manufacturer's recommendations and incubated at 37°C in a 5% CO2 O/N incubator. The next day, Thaw-and-Use TIGIT effector cells (Promega, CS198811) were added to CD155 aAPC/CHO-K1 cell plates containing fresh complete medium with anti-TIGIT antibody at 133 nM according to the manufacturer's recommendations (Figure 12A). or increasing concentrations (0.22, 0.54, 1.36, 3.41, 8.53, 21.3, 53.3, 133.33, and 333 nM) of anti-TIGIT antibody (Figure 12B) and incubated at 37°C, 5% CO2 for 6 hours.

After 6 h of incubation, TIGIT effector cell activation was assessed by measuring luciferase activity using the Bio-GloTM Luciferase Assay System (Promega, G7941).

In fig. 12A demonstrates the effect of addition of selected clones on luciferase signal compared to isotype control. The data demonstrate antagonistic activity of those antibodies that resulted in stronger activation of Jurkat-hTIGIT cells. Table 10 summarizes the induction of fold change in luciferase expression obtained for various anti-TIGIT antibodies compared to the isotype control clone (03847).

Table 10

Induction of fold change by isotype control

As shown in FIG. 12B, activation of Jurkat-hTIGIT cells was assessed with anti-TIGIT antibody ranging from 0.22 nM to 333 nM and gave an EC50 value of 3.0 nM for clone 29489; 4.4 nM for clone 29494; 2.3 nM for clone 29520 and 32 nM for clone 29527; 2.7 nM for clone 32919 and 3.2 nM for clone 32931, demonstrating strong functional activity due to blocking inhibitory TIGIT signaling. Clones 29489 and 31282 (mutant 29489 with the M to T mutation at residue 116) were also compared, and the EC ₅₀ values were 4.3 nM and 8.1 nM, respectively, for the example shown in FIG. 12C, showing similar functional activity for the 2 clones.

B. Functional analysis of primary human CD8+ T cells.

To characterize the functional consequences of blocking the human TIGIT receptor, we cocultured primary human CD8+ T cells from PBMCs from healthy human donors with the CHO-K1 cell line engineered to express human CD155 and activate human T cells. We observed that IFNg release by CD8+ T cells in the presence of engineered CD155-expressing CHO-K1 cells could be increased by blocking hTIGIT

- 39 045372 antagonistic anti-TIGIT antibodies. To compare the effectiveness of these antibodies in increasing IFNg release, the experiment was performed in the presence of increasing concentrations of antibodies and EC ₅₀ values were calculated.

CD155 aAPC/CHO-K1 cells (Promega, CS198811) thaw-and-use format were seeded in 96-well U-bottom plates according to the manufacturer's recommendations and incubated at 37°C in a 5% CO2 O incubator. /N. The next day, CD8+ T cells were purified according to the manufacturer's recommendations using a negative selection kit (Stemcell Technologies, 17953) from frozen human peripheral blood mononuclear cells isolated from the blood of healthy donors (Immunehealth). Purified CD8 T cells were then incubated with increasing concentrations (0.11 nM, 0.33 nM, 1.06 nM, 3.3 nM, 10.6 nM, 33.3 nM, 105.5 nM, and 333 nM) of antibodies (100,000 CD8 T cells/100 µl complete medium containing antibodies) for 1 hour. The antibody-CD8 mixture was then added to CD155 aAPC/CHO-K1 cell plates containing 50 μl of fresh complete medium and incubated at 37°C, 5% CO2 for 5 days. Finally, IFNg concentrations were assessed in the cell supernatant using an ELISA assay (Affymetrix eBioscience, 88-7316-86), which was performed according to the manufacturer's recommendations.

As shown in FIG. 13A, all anti-TIGIT antibodies increased IFNg secretion compared to isotype control. The greatest increase was observed in clone 29489 (6.4 times), followed by 29494 (5.8 times), 29520 (5.4 times), 29499 (5.2 times), 29527 (4.5 times) and 29513 (3. 2 times).

A dose range study (between 0.22 nM and 333 nM anti-TIGIT antibody) was also conducted to evaluate the EC ₅₀ value for increasing IFNg secretion by primary human CD8 T cells. As shown in FIG. 13B, anti-TIGIT antibody 29489 showed the best activity with an EC50 of 3.5 nM, followed by clone 29527 ( _EC50 = 5.1 nM), clone 29494 ( _EC50 = 6.1 nM) and clone 29520 ( _EC50 = 11.1 nM). Finally, clone 29489 and its mutant 31282 were tested in parallel and showed similar activity with corresponding EC ₅₀ values of 0.49 nM and 0.50 nM (Figure 13C). Overall, these data demonstrate the potent functional activity of antagonistic anti-TIGIT antibodies in blocking the TIGIT inhibitory signal in human CD8+ T cells and enhancing effector functions, which is characterized by a strong increase in IFNg production.

C. Functional analysis of human TILs.

To characterize the functional consequences of blocking the human TIGIT receptor on tumor infiltrating lymphocytes (TILS) from cancer patients, we cocultured primary human CD8+ T cells from TIL from a patient with ovarian ascites with the CHO-K1 cell line engineered to express human CD155 and activate T - human cells. We observed that IFNg release by CD8+ T cells in the presence of engineered CD155-expressing CHO-K1 cells could be increased by blocking hTIGIT with antagonistic anti-TIGIT antibodies.

CD155 aAPC/CHO-K1 cells (Promega, CS198811) thaw-and-use format were seeded in 96-well U-bottom plates according to the manufacturer's recommendations and incubated at 37°C in a 5% CO2 O incubator. /N. The next day, CD8 T cells were purified according to the manufacturer's recommendations using a negative selection kit (Stemcell Technologies, 17953) from frozen human TIL isolated from ovarian ascites (Immunehealth). Purified CD8+ T cells were then incubated with anti-TIGIT antibody clone 26452, unoptimized parental clone 29489, and 31282 (100,000 CD8+ T cells/100 μl complete medium containing antibody) for 1 hour. The antibody-CD8 mixture was then added to CD155 aAPC/CHO-K1 cell plates containing 50 μl of fresh complete medium and incubated at 37°C, 5% CO2 for 5 days.

Finally, IFNg concentrations were assessed in the cell supernatant using an ELISA assay (Affymetrix eBioscience, 88-7316-86), which was performed according to the manufacturer's recommendations. As can be seen in FIG. 14, IFNg secretion increased almost 2-fold when anti-TIGIT antibody was added to the co-culture. These data demonstrate the potent functional activity of antagonistic anti-TIGIT antibodies to block the TIGIT inhibitory signal in human CD8 ⁺ TILs and to enhance T cell effector functions in tumor settings.

Example 11. Characterization of an antagonistic anti-TIGIT antibody with functional activity in the mouse.

A. Mouse CD155 competition assay for surrogate antagonistic anti-TIGIT antibody.

Jurkat cells (clone E6-1, ATCC TIB-152) engineered to overexpress mouse TIGIT (Jurkat-mTIGIT) were used for this analysis. Anti-TIGIT antibody 26493 was used as a surrogate because this antibody showed cross-reactivity to mouse TIGIT as well as binding to human TIGIT. Cells were preincubated for 45 min at 37°C with various concentrations of anti-TIGIT antibody clone 26493 (0.03 to 10 μg/ml) in 25 μl of complete medium (RPMI + 10% FBS). Cells were washed once and incubated with 4 μg/ml mouse CD155-His-Fc tag protein (Thermo Fisher, 50259M03H50) in 50 μl complete medium for 45 minutes in an incubator. Cells were washed once and stained with PE anti-His antibody (Biolegend, 362603) for 30 min at 4°C. Median fluorescence intensity (MFI) measured by FACS was used as a measure of CD155 binding to Jurkat-mTIGIT. In fig. 15A shows the dose-response curve of anti-TIGIT clone 26493 for CD155 competition, identifying 2.3 nM as the IC ₅₀ (upper dotted line represents signal from isotype, lower dotted line represents signal from cells without CD 155). These results demonstrate the functional effectiveness of the anti-TIGIT antibody to compete with the CD155 ligand for mouse TIGIT.

B. Mouse in vitro functional assay: antigen-specific cytotoxicity (OT-1).

To evaluate the antigen-specific cytotoxic activity of OT-I CD8+ T cells against OVA-pulsed target cells and the effect of anti-TIGIT antibodies in this assay, OT1 cells were isolated from the spleens of C57BL/6-Tg ^(TcraTcrb) 1100Mjb/Crl mice (Charles River ) by mechanical dissociation followed by negative selection of mouse T cells using the EasySep™ Mouse T Cell Isolation Kit (Stemcell, Catalog #19851). As antigen presenting cells, PanO2 cancer cells, which naturally express CD155, were treated with mitomycin C (25 μg/ml) and then pulsed with OVA peptide (S7951-1MG, Sigma Aldrich, 1 μg/ml, 1 h at 37°C) . CD8+ T cells and PanO2 were cocultured for 3 days in the presence of anti-TIGIT clone 26493 or isotype control at 133 nM. On day 3, the supernatant was collected for IFNg detection by ELISA (Fig. 15B) and T cell cytotoxicity assay (Fig. 15C). OVA-pulsed PanO2 was used as target cells. Target cells and non-pulsed PanO2 cells (non-target internal control), 1 x 10 ^b each, were labeled with CFSE (C1157, ThermoFisher) and CellTrace™ Far Red Cell Proliferation Detection Kit (C34564, ThermoFisher), respectively, according to the manufacturer's instructions. These cells were mixed (1:1 ratio) and seeded in an amount of 2x10 ⁴ cells per well. OT-1 stimulated CD8+ T cells were added at 1x105 cells/well (effector cells), resulting in an effector/target ratio of 10:1 in the presence of anti-TIGIT clone 26493 or isotype control at 133 nM. After 24 h, the cells were washed with PBS and recovered by trypsinization. The cells were then stained with the Violet Fixed Dead Cell Live/Dead Stain Kit (Molecular Probes, L34955). Cytotoxic killing of target cells was then measured by monitoring changes in the ratio of live target cells to non-target cells using flow cytometry.

In fig. 15B demonstrates that the anti-TIGIT antibody increases IFNg production by nearly 2-fold, while FIG. 15C demonstrated increased cytotoxic activity of mouse OT-I CD8+ T cells by approximately 60%. Overall, these results confirm the functional activity of the anti-TIGIT antibody in enhancing murine CD8+ T cell effector function.

Example 12. Antitumor activity of antagonistic anti-TIGIT antibodies in monotherapy and in combination with anti-PD1 antibodies in a mouse model.

A. In vivo antitumor activity of antagonistic anti-TIGIT antibodies in monotherapy.

For this experiment, anti-TIGIT clone 26493 was produced in mammalian cells of the mouse IgG2a isotype. Female Balb/c mice were inoculated subcutaneously with 500,000 CT26 colon cancer cells (ATCC® CRL-2638™) for 8 weeks. On day 9 postinoculation, when tumor volumes averaged approximately 45 mm ³ , mice were randomized into treatment groups of equal tumor volume (n=8 per group). Mice were treated with 200 μg anti-TIGIT or isotype control (mIgG2a, BioXcell) or 200 μg anti-PD-1 (RMP1-14, BioXcell) and 200 μg isotype control (mIgG2a, BioXcell) or 200 μg anti-PD-1 (RMP1 -14, BioXcell) and various concentrations of anti-TIGIT (200 μg, 60 μg, 20 μg) by intraperitoneal injection on day 9, day 12, and day 15. Tumor growth was monitored and tumor volumes were measured using electronic calipers three times a week from the 9th to the 36th day. Mice were sacrificed when tumor volume exceeded 2000 mm ³ . Tumor growth curves were statistically analyzed using a linear mixed model. Differences between treatment groups were assessed by testing the time*treatment group interaction. To test for a synergistic effect between anti-TIGIT and anti-PD-1, treatment groups were recoded by the combination of the two variables; anti-TIGIT (yes/no) and anti-PD-1 (yes/no). The synergistic effect, in addition to the additive effect of each treatment (anti-TIGIT*time and anti-PD-1*time), was assessed by testing the interaction term anti-TIGIT*anti-PD-1*time.

In fig. 16A shows median tumor growth curves per group as well as individual growth curves for mice treated with anti-TIGIT antibody monotherapy. While no mice in the control group experienced tumor regression, 2/8 mice treated with anti-TIGIT had a complete response. The remaining mice showed a clear delay in tumor growth. In the control group, no mice lived beyond 30 days, while in the treated group, 7/8 mice lived beyond 30 days.

In fig. 16B shows median tumor growth curves per group, as well as individual growth curves for mice that received anti-PD1 alone or in combination with anti-TIGIT. There was significant suppression of tumor growth in mice treated with anti-TIGIT+ anti-PD-1 compared with anti-PD-1 monotherapy (p<0.0001). The combination of anti-TIGIT+anti-PD-1 achieved

- 41 045372 synergistic antitumor efficacy, which was greater than the additive effect of both monotherapies (p=0.02). The combination of anti-TIGIT antibodies (at 200 μg) and anti-PD1 antibodies resulted in 7/8 mice experiencing a complete response. Antitumor efficacy was maintained with a combination of anti-PD1 and lower doses of anti-TIGIT, which resulted in a complete response in 8/8 mice when anti-TIGIT antibody was reduced to 60 μg and in 5/8 mice when anti-TIGIT antibody decreased further to 20 μg (Fig. 16C). These data demonstrate significant antitumor efficacy of anti-TIGIT therapy when used alone (p<0.0001) or in combination with an anti-PD1 antibody (p<0.0001) for the treatment of established tumors.

Example 13. Isotype-dependent antitumor activity of an antagonistic anti-TIGIT antibody in monotherapy and combination with anti-PD1 antibodies in a mouse model.

For this experiment, anti-TIGIT clone 26493 was produced in mammalian cells of the mouse IgG2a and mouse IgG1 isotypes. Female Balb/c mice were inoculated subcutaneously with 500,000 CT26 colon cancer cells (ATCC® CRL-2638™) for 8 weeks. On day 10 postinoculation, when tumor volumes averaged approximately 100 mm ³ , mice were randomized into treatment groups of equal tumor volume (n=8 per group). To evaluate monotherapy, mice were treated with 200 μg of anti-TIGIT or isotype control (mIgG2a, BioXcell) by intraperitoneal injection on days 10, 13, and 16. To evaluate the anti-PD-1 combination, mice were treated with 200 μg anti-PD-1 (RMP1-14, BioXcell) and 200 μg isotype control (mIgG2a, BioXcell) or the 200 μg αntu-PD-1 combination. (RMP1-14, BioXcell) and 200 μg anti-TIGIT by intraperitoneal injection on days 10, 13, and 16. Tumor growth was monitored and tumor volumes were measured using electronic calipers three times a week from days 10 to 33. Mice were sacrificed when tumor volume exceeded 2000 mm ³ .

In fig. 17A shows median tumor growth curves per group as well as individual growth curves for anti-TIGIT antibody monotherapy and FIG. 17B for combination therapy with anti-TIGIT and anti-PD1 antibodies. Both as monotherapy and in combination with anti-PD-1, anti-TIGIT antibody treatment resulted in significant antitumor efficacy when administered as the mouse IgG2a isotype (p=0.0001 and p=0.009, respectively). However, antitumor efficacy could not be detected using anti-TIGIT as a mouse IgG1 isotype, suggesting that Fc receptor interaction with mIgG2a is important for the antitumor activity of antagonistic anti-TIGIT antibodies in the CT26 mouse model. These data demonstrate the isotype-dependent antitumor efficacy of anti-TIGIT therapy in monotherapy or combination for the treatment of established tumors.

Example 14. Characterization of the mechanism of action of the in vivo antitumor activity of an antagonistic anti-TIGIT antibody.

A. Flow cytometry analysis of spleen and tumor.

To examine the in vivo mode of action of the antagonistic anti-TIGIT antibody, tumors were analyzed by flow cytometry for the presence of an immune cell infiltrate following treatment with anti-TIGIT antibody 26493 (IgG2a), alone and in combination with anti-PD-1. Mice were inoculated and treated as described in Example 12. Two days after the second treatment, mice (8 mice per group) were sacrificed and tumors were collected. Tumors were dissociated using a tumor dissociation kit (Miltenyi Biotec). For direct ex-vivo staining, cells were stained with anti-CD45, anti-CD4, anti-CD8, and anti-FoxP3 (all from eBioscience) after staining with intravital dye (Molecular Probes, L34955) and Fc block. For ex vivo stimulation, cells were incubated with cell stimulation cocktail (eBioscience) and protein transport inhibitor (eBioscience) for 3 h. This was followed by staining with anti-CD4 and anti-CD8 antibodies and Fc block. After fixation and permeabilization with commercial buffers (IC fixation buffer and permeabilization buffer), cells were stained with anti-IL-10 and anti-IFNg (all from eBioscience). All figures show percentage change compared to the corresponding control group (isotype control for monotherapy, anti-PD-1 for combination), with a negative value representing a decrease and a positive value representing an increase compared to the control group.

In fig. 18A demonstrates that in vivo treatment of a tumor with anti-TIGIT mIgG2a antibody results in a 28% reduction in the proportion of regulatory T cells in the CD4 ⁺ TIL population compared to controls, which is not observed after treatment with anti-TIGIT mIgG1. This shows that TIGIT+ Treg cells are depleted, which may explain the differential efficacy of the two isotypes as discussed in Example 14. FIG. 18B demonstrates that there is no depletion of CD8 ⁺ TILs, but instead a slight increase for the two isotypes (17% increase over control for mIgG1 and 16% for mIgG2a). These results collectively lead to a greater than 50% increase in the CD8/Treg ratio in the tumor treated with anti-TIGIT mIgG2a (Fig. 18C). Intratumoral T cell functionality also improved in the anti-TIGIT antibody mIgG2a group, with a strong increase in IFNg production from both CD4+ (FIG. 18D) and CD8 ⁺ TILs (FIG. 18E). This resulted in a strong increase in the ratio of IFNγ-producing cells to IL-10-producing cells after ex vivo stimulation in the CD4 ⁺ TIL/CD8 + population (Fig. 18F).

In fig. 18G demonstrated that combining anti-TIGIT mIgG2a with anti-PD-1 leads to

- 42 045372 decrease in regulatory T cells by 33% compared to anti-PD-l monotherapy. Again, the opposite was true for CD8+ T cells, with 22% and 28% increases in CD8+ T cell infiltration, respectively, for the mIgG1 and mIgG2a isotypes, compared with anti-PD- monotherapy! (Fig. 18H). Taken together, this results in a more than two-fold increase in the ratio of CD8 ⁺ TILs to Tregs in the tumor for combination with anti-TIGIT mIgG2a (Figure 18I). Additionally, treatment with anti-TIGIT antibody mIgG2a in combination with antiPD-1 demonstrates a shift to a Th1 versus Th2 phenotype for intratumoral CD4+ T cells with a marked increase in IFNg-producing CD4 cells (Fig. 18J) and a decrease in IL-10-producing CD4 cells (Fig. 18K). This resulted in a strong increase in IFNg/IL-10 producing cells following ex vivo stimulation in the CD4 ⁺ TIL population compared to mice treated with anti-PD-1 monotherapy (Figure 18L).

Table 11

Differentially expressed genes between anti-TIGIT mIgG2a and vehicle-treated mice

Gene symbol Log2 multiple Adjusted p- changes meaning Ssg2 -1.29 0.0000668 Prfl 1.79 0.0000668 Ctsg 2.13 0.0000668 Ctla4 1.72 0.00309 Gzmb 1.51 0.00309 Ccl2 0.56 0.0174 I12ra 1.61 0.0174 CD55 1.64 0.0213 I12rb 0.872 0.0379 CD274 0.982 0.0385 Klrgl 1.3 0.0402 Icos 1.26 0.0402 Him 0.87 0.0402 Cx3crl -0.82 0.0428 Clra 0.896 0.0428 CD33 -0.906 0.0479 Ccl4 0.886 0.0518

Table 12

Differentially expressed genes between mice,

who received anti-TIGIT mIgG2a + anti-PD-! and anti-PD1 Gene symbol Log2 multiple changes Adjusted value Ctsg 2.34 0.0000375 Prfl 1.69 0.000255 Gzmb 1.71 0.000766

- 43 045372

CD55 2.08 0.00131 Entpdl 0.839 0.00131 Klrgl 1.76 0.00132 Itgal 0.874 0.0017 Ctla4 1.72 0.00173 I12ra 1.82 0.00237 Itgb3 0.863 0.00237 Slclllal 0.849 0.00329 CD36 1.44 0.0049 Cdl80 0.899 0.00602 Icaml 0.893 0.00802 CD274 1.06 0.00993 CD40 0.926 0.0113 Eomes 1.28 0.0113 Abcgl 0.869 0.0113 Ccr2 -0.781 0.0122 Thyl 0.868 0.0165 Ccl2 0.501 0.0203 Gbp5 1.12 0.0216 Icos 1.24 0.0263 Tgfbr2 0.458 0.0278 H2K1 0.292 0.0307 Sh2dla 0.999 0.0307 I12rb 0.808 0.0307 Selplg 0.64 0.031 Bstl 0.702 0.0317 CD247 1 0.032 Irf8 0.699 0.0365 112 Ir 0.899 0.0392 Gbp2b 1,H 0.0392 Stall 0.865 0.0427 C4b 0.922 0.0428 Abcal 0.537 0.044 Tre m2 0.482 0.0454

B. Tumor transcriptomic analysis using NanoString.

To investigate the in vivo mode of action of the anti-TIGIT antibody, the immune cell infiltrate of tumors treated with anti-TIGIT, alone and in combination with anti-PD-1, was analyzed by transcriptomic analysis (Nanostring). Mice were inoculated and treated as described in Example 12. Two days after the third treatment with anti-TIGIT and/or anti-PD1 antibodies, mice were sacrificed and tumors were collected. RNA was isolated and the expression of a selection of 770 genes involved in cancer immunology was directly quantified using nCounter technology (PanCancer Immune Profiling Panel, Nanostring; performed by VIB Nucleomics Core). Data were analyzed using nSolver software (Nanostring).

In fig. 19A shows a volcano plot of genes that are differentially regulated between vehicle-treated and anti-TIGIT mIgG2a-treated mice. Highly statistically significant genes fall at the top of the plot, and highly differentially expressed genes appear on both sides (left: down-regulated in anti-TIGIT-treated mice, right: up-regulated in anti-TIGIT-treated mice). Examples of highly upregulated genes include perforin, granzyme B, and CTLA-4. The solid line represents the unadjusted p-value of 0.01, the dotted line the adjusted p-value of 0.05 (Benjamini-Hochberg correction). In table 11, in table. 12 demonstrates genes that were significantly differentially expressed for anti-TIGIT mIgG2a versus vehicle and aPD-1+anti-TIGIT mIgG2a versus anti-PD1, respectively. When several genes were summarized into scores for functional immune cell subsets, the most striking finding was an increase in cytotoxic cells and CD8+ T cells (Figure 19B). The same changes were present in mice treated with anti-PD-1 + anti-TIGIT mIgG2a compared with anti-PD1 alone. No changes were observed in mice treated with anti-TIGIT mIgG1, alone or in combination with anti-PD-1.

Overall, these results demonstrate that the antitumor efficacy observed after in vivo treatment with an anti-TIGIT antibody is mediated by a decrease in Treg infiltrate in the tumor, while the effector CD8+ T cell population increases. In addition, the effector function of CD4+ and CD8 ⁺ TILs is enhanced, as evidenced by a higher proportion of IFNg-producing cells, a shift toward a Th1 response, and increased expression of genes important for cytotoxic T cell functions.

Example 15 Antibody-dependent cellular toxicity (ADCC) activity induced by antagonistic anti-TIGIT antibodies.

A. In vitro ADCC on human PBMC from healthy donors.

Isolated PBMCs from healthy human donors were resuspended in complete RPMI medium (supplemented with 10% heat-inactivated FBS + 50 U penicillin + 50 U streptomycin and supplemented with 200 U IL-2/ml). 2.5x105 human PBMCs were distributed per well in a 96-well plate. Human mammalian anti-TIGIT antibody clone 26452 or IgG1 isotype control (Biolegend, 403102) was added at a final concentration of 66.6; 0.66 and 0.006 nM for each corresponding well. Cells were incubated for 20 h at 37°C with 5% CO ₂ . Cells were then collected and stained with the following panel of antibodies: LVD efluor 520 (eBioscience 65-0867-14), anti-TCRab-PercP-Cy5.5 (clone IP26, Biolegend 306723), anti-CD4-BV510 (clone SK3, BD Horizon 562970 ), anti-CD8-APC-Cy7 (clone SK1, Biolegend 344714), anti-CD25-BV605 (clone 2A3, Biolegend 562660), anti-CD127-APC (A019D5, Biolegend 351316), anti-CCR7-BV421 (clone G043H7, Biolegend 353207) and anti-CD45RO ^- PE-Cy7 (clone UCHL1, Biolegend 304229). Results are presented on gated live cells. CD45+CD4+ or CD45+CD8+ represent the total number of CD4+ or CD8+ T cells. CD45 ⁺ RO ⁺ CD4 ⁺ or CD45 ⁺ RO ⁺ CD8 ⁺ cells are CD4+ or CD8+ memory T cells, whereas CD25 ^high CD127 ^low CD4 ⁺ are Treg cells. The proportion of TIGIT+ cells in gated Tregs is higher than in gated memory CD8+ T cells and CD4+ T cells, as demonstrated in Fig. 20A.

Absolute quantitation is performed using AccuCheck Counting beads (Life technologies) according to the manufacturer's instructions. After calculating the absolute number of cells per μl, % specific lysis is calculated using the following formula = (1(absolute number of cells per μl on the sample treated with TIGIT 26452 antibody/average value of triplicates not treated with antibody)) x 100. As shown in FIG. 20B, anti-TIGIT antibody 26452 hIgG1 triggers higher specific lysis on Tregs (62.22%) than on total CD8+ T cells (12.2%) or total CD4+ T cells (16.36%).

B. Ex-vivo ADCC on mouse tumors.

To confirm that anti-TIGIT mouse IgG2a antibody can deplete TIGIT+ regulatory T cells, an ex-vivo ADCC assay was performed. Female Balb/c mice were inoculated subcutaneously with 500,000 CT26 colon cancer cells (ATCC® CRL-2638™) for 8 weeks. Three weeks after inoculation, tumors were collected and dissociated using a tumor dissociation kit (Miltenyi Biotec). The single cell suspension was incubated with 133 nM anti-TIGIT antibody 26493 (mIgG1 or mIgG2a isotype) for 20 h (1 million cells/200 μl in RPMI + 10% FBS). After 20 hours, cells were stained with anti-CD4, anti-TIGIT, anti-CD8, and anti-FoxP3 antibodies (all from eBioscience) after staining with intravital dye (Molecular Probes, L34955) and Fc block.

In fig. Figure 21 shows the % reduction in absolute TIGIT+ cell counts compared to isotype control treatment for different TIGIT+ subsets. The greatest reduction following treatment with the anti-TIGIT antibody mIgG2a was observed in regulatory T cells (approximately 40% reduction), suggesting that these cells are more susceptible to ADCC than conventional CD4+ or CD8 ⁺ T cells.

Overall, these results demonstrate the effectiveness of anti-TIGIT hIgG1 or mIgG2a in depleting TIGIT+ immune cells, with higher activity demonstrated in the Treg population.

Example 16: Prediction of immunogenicity using in silico analysis.

The immunogenic potential of clones 29494 and 29489, as well as its variant 31282, was assessed by in silico prediction using the EpiMatrix Protein Score (De Groot et al. (2009) Clinical Immunol. 131:189.). To complete the analysis, input sequences were analyzed in overlapping 9-mer frames, and each frame was scored against a panel of eight common HLA class II alleles. These alleles are supertypes. Each is functionally equivalent or nearly equivalent to many additional alleles of a family member. Taken together, these eight supertype alleles, along with their family members, cover more than 95% of the human population (Southwood et al. (1998) J. Immunol. 160:3363). Each allele-score is a statement of predicted HLA binding affinity. EpiMatrix scores range from -3 to +3 and are normally distributed. EpiMatrix scores above 1.64 are defined as hits; that is, potentially immunogenic and worthy of further consideration.

All other factors being equal, the more HLA ligands (i.e., EpiMatrix hits) a given protein contains, the more likely that protein is to elicit an immune response. The EpiMatrix protein score is the difference between the number of predicted T cell epitopes expected to be found in a protein of a given size and the number of predicted epitopes predicted by the EpiMatrix system. EpiMatrix protein scores correlate with observed immunogenicity. EpiMatrix protein scores are normalized and can be plotted on a standardized scale. The EpiMatrix protein score for the average protein is zero. EpiMatrix protein scores above zero indicate the presence of excess MHC ligands and indicate higher immunogenicity potential, while scores below zero indicate the presence of fewer potential MHC ligands than expected and lower immunogenicity potential. Proteins with a score above +20 are considered to have significant immunogenic potential.

Correction for the presence of regulatory T cell epitopes.

Antibodies are unique proteins in the sense that the amino acid sequences of their variable domains, especially their complementarity determining regions (CDRs), can vary to an extraordinary extent. It is this variability that allows antibodies to recognize a wide range of antigens. However, recombination events and mutations that control antibody maturation can also give rise to new or neo-epitopes of T cells. These neoepitopes may appear foreign to circulating T cells. The presence of neoepitopes in antibody sequences can lead to the formation of a human-anti-human antibody response; also known as a HAHA or ADA (anti-drug antibody) response.

Regulatory T cells play an important role in suppressing immune responses to fully human proteins in the periphery, including those containing mutated and/or highly variable sequences such as antibody CDRs. Regulatory T cells are recruited and activated by regulatory T cell epitopes. The inherent risk associated with the presence of neoepitopes in antibody sequences appears to be offset by the presence of naturally occurring regulatory T cell epitopes.

Through sequence screening of many human antibody isolates, EpiVax has identified several highly conserved HLA ligands that are believed to have regulatory potential. Experimental evidence suggests that many of these peptides are, in fact, actively tolerogenic in most subjects. These highly conserved, regulatory and heterogeneous T cell epitopes are now known as Tregitopes (De Groot et al. (2008) Blood 112: 3303)

In many cases, the immunogenic potential of neoepitopes contained in humanized antibodies can be effectively controlled in the presence of significant amounts of Tregitope. For the purpose of antibody immunogenicity analysis, EpiVax has developed a Tregitope-corrected EpiMatrix score and corresponding prediction of anti-therapeutic antibody response. To calculate the Tregitope-adjusted EpiMatrix score, Tregitope scores are subtracted from the EpiMatrix protein score. Tregitope-corrected scores have been shown to correlate well with the observed clinical immune response for a panel of 23 commercial antibodies (De Groot et al. (2009) Clinical Immunol. 131:189).

The antibody sequences of clones 29489, 29494 and 31282 are scored at the lower end of the EpiMatrix scale, indicating limited immunogenicity potential. Regression analysis of licensed monoclonal antibodies predicts ADA response in ~0% of patients exposed to antibodies from clones 29489 and 31282. For clone 29494, the analysis predicts ADA response in 2.78% of exposed patients for the base VH sequence and 2.88% for the variant VH sequences. The data is given in table. 13 below.

- 46 045372

Table 13

EpiMatrix and Tregitope adjusted EpiMatrix scores______

Incoming subsequence Length Ratings EpiMatrix hits Grade EpiMatrix Grade EpiMatrix adjusted for Tregitope 29489 VH 121 904 40 -19.41 -47.26 29489VL 108 800 39 -17.58 -51.75 31282 VH 121 904 40 -19.41 -47.26 31282VL 108 800 39 -17.58 -51.75 29494_VH 125 936 54 2.68 -7.18 29494_VL 107 792 40 -12.2 -38.83

Example 17 Determination of binding affinity of anti-TIGIT clones to recombinant human TIGIT protein.

Antibody 31282 was compared to anti-TIGIT antibody clones described in other patent applications. Specifically, 31282 was compared to: 4.1D3.Q1E (also referred to as 4.1D3 from WO 2017/053748); 22G2 (from WO 2016106302); 31С6 (from WO 2016/028656); 313M2 (from WO 2016/191643); and TIG1 (from WO 2017/152088). Links and sequences of the compared antibody clones are shown in Table. 14 below:

Table 14

Sequences of the VH and VL domains of comparative anti-TIGIT antibodies

Clone a- TIGIT Links Subsequence 4.1D3.Q1E VH: SEQ ID NO: 34 of WO2017/05 3748 VL: SEQ ID NO: 36 of VH sequence: EVQLQQSGPGLVKPSQTLSLTCAISGDSVSSNSAAWNWIRQSPSRGLE WLGKTYYRFKWYSDYAVSVKGRITINPDTSKNQFSLQLNSVTPEDTA VFYCTRESTTYDLLAGPFDYWGQGTLVTVSS (SEQ ID NO: 343 in this document)

- 47 045372

WO2017/05 3748 VL sequence: DIVMTQSPDSLAVSLGERATINCKSSQTVLYSSNNKKYLAWYQQKPG QPPNLLIYWASTRESGVPDRFSGSGSGTDFTLTISSLQAEDVAVYYCQ QYYSTPFTFGPGTKVEIK (SEQ ID NO: 344 in this document) 22G2 VH: SEQ ID NO:7 out WO2016/10 6302 VL: SEQ ID NO:9 out WO2016/10 6302 VH sequence: QVHLQESGPGLVKPSETLSLTCTVSGGSVSSGIYYWSWIRQPPGKGLE WIGYIYYSGSTNYNPSLKSRVTISVDTSKNQFSLKLSSVTAADTAVYY CARDYYVSGNYYNVDYYFFGVDVWGQGTTVTVSS (SEQ ID NO: 345 in this document) VL sequence: EIVLTQSPATLSLSPGERATLSCRASQSVSSYLAWYQQKPGQAPRLLIY DASNRATGIPARFSGSGSGTDFTISSLEPEDFAVYYCQQRSNWPPLF TFGPGTKVDIK (SEQ ID NO: 346 in this document) 31S6 (MEV125.3 1S6.A1.205 VH4/VL1) VH: SEQ ID NO: 127 of WO2016/02 8656 VL: SEQ ID NO: 130 of WO2016/02 8656 VH sequence: EVQLVQSGAEVKKPGASVKVSCKASGYTFSSYVMHWVRQAPGQGLE WIGYIDPYNDGAKYAQKFQGRVTLTSDKSTSTVYMELSSLRSEDTAV YYCARGGPYGWYFDVWGQGTTVTVSS (SEQ ID NO: 347 in this document) VL sequence: DIQMTQSPSSLSASVGDRVTITCRASEHIYSYLSWYQQKPGKAPKLLIY NAKTLAEGVPSRFSGSGSGTDFTLTISSLQPEDFATYYCQHHFGSPLTF GQGTRLEIK (SEQ ID NO: 348 in this document) 313M32 VH: SEQ ID NO:67 from WO2016/19 1643 VL: SEQ ID NO:68 from WO2016/19 1643 VH sequence: QVQLQESGPGLVKPSETLSLTCAVSGYSITSDYAWNWIRQPPGKGLE WIGYISYSGSTSYNPSLRSRVTISRDTSKNQFFLKLSSVTAADTAVYYC ARRQVGLGFAYWGQGTLVTVSS (SEQ ID NO: 349 in this document) VL sequence: DIQMTQSPSSLSASVGDRVTITCKASQDVSTAVAWYQQKPGKAPKLLI YSASYRYTGVPSRFSGSGSGTDFTFTISSLQPEDIATYYCQQHYSTPWT FG (SEQ ID NO: 350 in this document) TIG1 VH: SEQ ID NO: 10 from WO2017/15 2088 VL: SEQ ID NO: 14 out VH sequence: DVQLVESGGGLVQPGGSRKLSCAASGTFFSNFGMHWVRQAPEKGLE WVAFISSGSSSSIYYADTVKGRFTISRDNPKNTLFLQMTSLRSEDTAMY YCARMRLDYYAMDYWGQGTSVTVSS (SEQ ID NO: 351 in this document) VL sequence: DVQITQSPSYLAASPGETITINCRASKSISKYLAWYQEKPGKTNKLLIYS WO2017/15 2088 GSTLQSGIPSRFSGSGSGTDFTLTISSLEPEDFAMYYCQQHNEYPWTFG GGTKLEIK (SEQ ID NO: 352 in this document)

A. Production in mammalian cells.

To obtain a sufficient number of selected a-TIGIT clones for further characterization, DNA vectors encoding specific antibody clones (clones 31282_up, 4.1D3, 22G2, 31C6, 313M32, and TIG1) were generated and transduced into HEK cells to produce the human IgG1 isotype. Codon-optimized synthetic human DNA fragments for antibody variable domains were ordered from Geneart. The variable domain sequences were smoothly ligated into pUPE expression vectors containing the mouse Igkappa signal sequence and constant regions of the corresponding class of antibodies. Expression vectors were confirmed by restriction analysis and DNA sequencing. For transient transfection, maxipreps DNA was obtained that did not contain

- 48 045372 endotoxin suppressors (Sigma), and heavy and light chain vectors were co-transfected into HEK293EBNA1 cells in Freestyle medium (ThermoFisherScientific) according to established protocols. Primatone (0.55% final volume) was added 24 hours after transfection. Conditioned medium was collected 6 days after transfection. Antibodies were purified using batch processes using Mabselect sureLX affinity chromatography (GE Healthcare). Bound antibodies were washed in 2 steps with PBS containing 1 M NaCl and PBS. Antibodies were eluted with 20 mM citrate, 150 mM NaCl, pH 3, and neutralized to approximately pH 7 with 1/6 volume of 1 MK ₂ HPO ₄ /KH ₂ PO ₄ , pH 8.

The antibodies were then further purified by gel filtration using a Superdex200 column equilibrated in PBS. Fractions were analyzed by NuPAGE and fractions containing antibodies were pooled. The final products were sterilized through a 0.22 μM syringe filter. The product was analyzed using NuPAGE and endotoxin levels were measured using the LAL assay.

In addition, clone 31282 was also produced in a CHO-K1 cell as described below (clone 31282_wu) on the IgG1 or IgG4 isotype. DNA vectors encoding antibodies were constructed and transfected into CHO-K1 cells. CHO codon optimized DNA fragments for antibody variable domains were synthesized and ligated into expression vectors containing the signal sequence and constant regions of the corresponding antibody class. Expression vectors were confirmed by restriction analysis and DNA sequencing. The heavy and light chain vectors were cotransfected into CHO-K1 cells by electroporation (Bio-rad) according to established protocols. Transfected cultures were scaled up and inoculated into fed-batch cultures. Conditioned media was collected after 14 days of fed-batch cultures.

The harvested cell culture was first clarified by two stages of deep filtration with a DOHCu A1HC series connection (Millipore). The clarified collection was then first purified by affinity chromatography with MabSelect SuRe (GE Healthcare). Bound antibodies were washed in 2 steps with 50 mM NaAc-HAc (pH 5.5) containing 1 M NaCl and 50 mM NaAc-HAc (pH 5.5). Antibodies were then eluted with 50 mM NaAc-HAc (pH 3.5) and neutralized to approximately pH 5.5 with 1 M Tris-HCl (pH 9.0).

The neutralized intermediate was then further polished by anion exchange chromatography (AEX) using a POROS HQ50 (Life Tech) in flow mode. The column was equilibrated with 50 mM NaAc-HAc (pH 5.5) before loading. The AEX stream collected during the loading and recovery step was further polished by cation exchange chromatography (CEX) in coupling-eluting mode using POROS XS (Life Tech.). The CEX column was equilibrated in 50 mM NaAc-HAc (pH 5.5) and antibodies were eluted by linear gradient elution (LGE) to reach 50 mM NaAc-HAc (pH 5.5) containing 0.5 M NaCl in 10 CV. Final ultrafiltration and diafiltration (UF/DF) using Pellicon 3, ultracel 30 kDa, type A (Millipore) was performed to concentrate the CEX eluate and exchange buffer in 20 mM His-HCl (pH 5.5). Polysorbate 80 (PS80) and sucrose were then added to the diafiltered sample to obtain a final product concentration of approximately 20 g/L in 20 mM His-HCl buffer, 0.01% (w/w) PS 80 and 9 % (w/v) sucrose (pH 5.5). The product passed all PQA tests. SEC purity, endotoxin level and other criteria met the requirement.

B. Biacore measurement.

Biosensor analysis was carried out at 25°C in the HBS-EP buffer system (10 mM HEPES, pH 7.3, 150 mM NaCl, 3 mM EDTA, 0.05% Tween 20) using Biacore T200 technology, CM5 sensor chip (launched in Novalix, France). The hotel sample was kept at 8°C. Goat anti-human IgG antibody (Fcy fragment specific, Jackson ImmunoResearch Laboratories) was immobilized (10,000 RU) in both flow cells of the sensor chip using standard amine coupling chemistry. This type of surface provided a format for reproducible capture of fresh assay antibody after each regeneration step. Flow cell 2 was used to assay captured antibody, while flow cell 1 was used as a control flow cell. Six different antigen concentrations ranging from 30 to 0.123 nM were prepared in the working buffer. Each of the antigen sample concentrations was run as a single replicate, except for the 3.33 nM run in duplicate. Two empty (buffer) injections were also run and used to evaluate and subtract system artifacts. Association (300 s) and dissociation (600 s) phases for all antigen concentrations were performed at a flow rate of 30 μl/min. The surface was regenerated using three successive injections (15 s, 15 s and 60 s) of 10 mM glycine-HCl, pH 1.5. The resulting sensorgrams were fitted globally to a 1:1 model (assuming the kinetic values were the same for all concentrations applied). Affinity was also determined from the steady state for clone 313M32, as the 1:1 kinetic model fit was not robust, showing equilibrium with human TIGIT at the end of the association time. The results obtained for different aTIGIT clones are shown in table. 15.

- 49 045372

Table 15

Kinetics and affinity assessment

Kinetic model of 1:1 binding Steady State Model Clone Ka (1/s) Kd (1/s) K<1(nM) Rmax (RU) KDnM) Rmax (RU) 31282_wu 3.86^ ⁶ 4,b2' ⁰⁴ 0.120 16.7 31282_ir 3.70^ ⁶ 4.75' ⁰⁴ 0.128 15.3 4.1D3 1.07^ ⁶ 4.72' ⁰⁵ 0.044 14.4 22G2 2.51 ⁴¹⁶ 1.78' ⁰⁴ 0.071 11.1 31С6 3.10^ ⁶ 2,О9' ⁰⁴ 0.067 16.7 313M32 pa pa pa pa 10.1 17.2 TIG1 ₅₂ 4^ ⁶ 1.3 G ⁰² 2.49 N,1

Example 18 Cellular binding of antagonistic anti-TIGIT antibodies.

A. Binding of anti-TIGIT clones to Jurkat-hTIGIT.

The affinity of human anti-TIGIT antibodies was measured using Jurkat E6.1 cells transduced with human TIGIT (Jurkat hTIGIT). To analyze the affinity of selected antibodies to hTIGIT, 10 ⁵ cells were distributed per well and incubated with decreasing concentrations (8; 4; 2; 1; 0.5; 0.25; 0.125; 0.062; 0.031; 0.016; ^8x10'3 and 4x10' ³ nM) of various clones of anti-TIGIT antagonist antibodies (Fig. 2). Antibodies were incubated with cells for 20 min at 4°C in FACS buffer. After washing, cells were incubated with anti-human 1g (Fcgamma-specific)-PE (eBioscience, 12-4998-82, at 2.5 μg/ml) for 20 minutes on ice and washed twice. Fluorescence intensity was analyzed using a BD Fortessa LSR, and cell binding was recorded as the average PE fluorescence intensity in cells expressing TIGIT on their surface.

The half-maximal binding concentration (EC50) of Jurkat-hTIGIT was calculated using a four-variable curve-form equation in Prism. The results are shown in FIG. 22A, and the values are given in table. 16 below. The EC50 values for Jurkat-hTIGIT binding are very similar for clone 31282 with no noticeable difference between the antibody produced in HEK cells (31282, 0.13 nM) or in CHO-K1 cells (31282 μV, 0.10 nM). Clones 31C6 and TIG1 also show EC ₅₀ values below 0.2 nM, while the affinity for the other clones (4.1D3, 22G2 and 313M32) is lower, and the results show EC ₅ o values ranging from 0.267 to 0.445 nM. The results demonstrate strong binding to membrane-expressed human TIGIT in the engineered system for anti-TIGIT clones 31282, 31C6 and TIG1, while other clones have lower affinity.

Table 16

EC ₅ q data and comparison of different a-TIGIT clones for binding to Jurkat-hTIGIT

Clone ECso binding to JurkathTIGIT (in nM) Multiple change of ECso to ECso of the best clone (31282_wu) 31282_wu 0.10 1 31282_ir 0.13 1.3 313M32 0.44 4.2 4.1D3 0.27 2.5 22G2 0.32 3.0 31С6 0.13 1.2 TIG1 0.17 1.6

B. Binding of anti-TIGIT clones to primary CD8 T cells from PBMCs from healthy individuals.

Isolated human PBMCs from healthy volunteers were assayed for binding using anti-TIGIT antagonist antibodies. Cells were distributed at ^1x105 cells per well. Cells were incubated with anti-CD 16 (clone 3G8, BioLegend 302002), CD32 (clone FLI8.26, BD Bioscience 557333) and CD64 (BD Bioscience 555525) at room temperature for 10 minutes, and the indicated clones anti

-50045372 human TIGIT antibodies were directly added at a final concentration of 8; 4; 2; 1; 0.5; 0.25; 0125; 0062; 0031; 0016; 8x10' ³ and 4x10' ³ nM in FACS buffer and incubated for 20 min at 4°C. After washing, the cells were incubated with anti-human 1g (Pc-gamma-specific)-PE (eBioscience, 12-4998-82, at 2.5 μg/ml) for 20 min at 4°C. The cells were then washed and incubated with the following antibodies and LVD mixture: anti-CO4-PercP-Cy5.5 (clone A161A1, BioLegend 357414); antiCD8-BV510 (clone SKI, BD Bioscience 563919) and LVD efluor 660 (eBioscience 65-0864-18).

EC ₅ o values for binding to primary human CD8 ⁺ T cells were calculated using the MFI signal on live TIGIT CD8 T cells. The results are illustrated in Fig. 22B, and EC50 concentrations are given in table. 17 below. The EC50 values for binding of primary human CD8 ⁺ T cells are very similar for clone 31282 with no noticeable difference between the antibody produced in HEK cells (31282_ir, 0.21 nM) or in CHO-ΚΙ cells (31282-wu, 0.19 nM) . Comparison between different clones of a-TIGIT antagonist antibodies shows the best EC5o value for binding to primary human CD8 ⁺ T cells for clone 31282_wu (0.19 nM) and clone 31282_ir (0.21 nM). Clones 31C6 and TIG1 show a 2-fold difference in EC ₅₀ , while clones 22G2, 313M32 and 4.1D3 differ by 6.1-9.7 times. Overall, 31282_wu and 31282_ir show better binding to membrane-expressed TIGIT on primary human CD8 ⁺ T cells.

Table 17

EU ₅ data and comparison of different a-TIGIT clones for binding to primary human CD8 ⁺ T cells

Clone ECso concentrations for binding to CD8 ⁺ T cells (in nM) Multiple change of ECso to ECso of the best clone (31282_wu) 31282_wu 0.19 1 31282_up 0.21 1.1 313M32 1.45 7.5 4.1D3 1.88 9.7 22G2 1.17 6.1 31С6 0.39 2.0 TIG1 0.38 2.0

C. Binding of anti-TIGIT clones to primary CD8 T cells from PBMCs of cancer patients.

Isolated human PBMCs from cancer patients were assayed for binding to various clones of anti-TIGIT antagonist antibodies. Cells were distributed at 1 x ¹⁰⁵ cells per well. Cells were incubated with anti-CD 16 (clone 3G8, BioLegend 302002), CD32 (clone FLI8.26, BD Bioscience 557333) and CD64 (BD Bioscience 555525) at room temperature for 10 min, and the indicated human anti-TIGIT antibodies were directly added at final concentrations of 8, 4, 2, 1, 0.5, 0.25, 0.125, 0.062 and 0.031 nM in FACS buffer and incubated for 20 min at 4°C. After washing, the cells were incubated with anti-human 1g (Pc-gamma-specific)-PE (eBioscience, 12-4998-82, at 2.5 μg/ml) for 20 min at 4°C. The cells were then washed and incubated with the following antibodies and a mixture of intravital dyes (LVD): anti-SP4-RegsR-Cy5.5 (clone A161A1, BioLegend 357414); antiCD8-BV510 (clone SKI, BD Bioscience 563919) and LVD efluor 520 (eBioscience 65-0867-14). Cells were washed and fixed, and surface staining was quantified using a BD LSR Fortessa. Flow cytometry data were analyzed using FlowJo V10.1. TIGIT MFI on gated LVD'TIGIT CD8 cells was used to calculate _EC5o values. Nonlinear regression curves are shown in FIG. 22C, and the values are given in table. 18 below.

Clones 31282_wu and 31282_ir show very similar EC ₅ o values for binding to CD8 ⁺ T cells from cancer patients with a concentration of 0.14 and 0.12 nM, respectively. The remaining clones exhibited lower affinity with clones 31C6, TIG1, and 22G2, showing 1.5-, 2.7-, and 3.1-fold lower affinities, respectively. The measured value of EC ₅ o for clone 313M32 is 8.3 times lower compared to clone 31282_ir. Clone 4.1D3 exhibits the lowest affinity, binding by a 9.5-fold difference to the best clone tested.

-51 045372

Table 18

EC50 data and comparison of different a-TIGIT clones for binding to primary CD8+ T cells from cancer patients

Example 19. Competition assay between anti-TIGIT antagonist antibody clones and natural TIGIT ligand (CD155).

Jurkat cells overexpressing human TIGIT (Jurkat-hTIGIT) were collected and distributed at 5x10 ⁴ cells/well and incubated with anti-human TIGIT antibodies at the following concentrations: 133.33; 42.20; 13.33; 4.22; 1.33; 0.422; 0.133; 0.042; 0.0133; 4.2χ10 ^-3 ; 1.3χ 10 ^-3 ; 4.2x10 ^-4 ; 1.3x10 ^-4 ; 4.2x10 ^-5 nM in complete medium for 45 min at 37°C. Excess antibodies were washed, and then cells were incubated with CD155-His at 15 μg/ml (Creative Biomart, PVR-3141H) for 45 minutes at 37°C. Bound CD155-His was then detected using anti-His tag-PE (Biolegend, 362603, 2 μl per test), incubated for 30 min at 4°C. Cells were analyzed by FACS using a BD LSRFortessa, and the half concentration (IC ₅₀ ) that prevents CD155 binding was calculated based on the average PE fluorescence intensity in total cells.

The results are shown in FIG. 23, and the values are given in table. 19 below. Anti-TIGIT clones 31282_wu and 31282_up show the best IC ₅₀ values for CD155 competition on Jurkat cells engineered to express hTIGIT at concentrations of 0.05 and 0.04 nM, respectively. Other clones (4.1D3, 22G2, 31C6, TIG1) have _IC50 values of 0.07 to 0.09 nM, while clone 313M32 competes with CD155 for binding to TIGIT with a much lower efficiency (0.65 nM) .

Table 19

Data . IC ₅₀ and comparison of different a-TIGIT clones for CD155 competition on human TIGIT

Clone ICso of CD155 competition for TIGIT binding (in nM) Multiple change of ICso to ICso of the best clone (31282_tsr) 31282_wu 0.05 1.3 31282_up 0.04 1 313M32 0.65 16.7 4.1D3 0.07 1.9 22G2 0.09 2.2 31С6 0.07 1.7 TIG1 0.06 1.6

Example 20. Functional characterization of antagonistic anti-TIGIT clones.

A. Functional analysis of TIGIT with Jurkat-hTIGIT cells.

To characterize the functional consequences of blocking the human TIGIT receptor, we cocultured Jurkat cells, which express hTIGIT and a luciferase reporter activated upon TCR uptake (Thaw-and-Use Format TIGIT Effector Cells from Promega), with the CHO-cell line. K1, designed to express human PVR/CD155 and TCR activator (thaw and use CD155 aAPC/CHO-K1 from Promega). Activation of TIGIT-overexpressing Jurkat cells can be induced by contact with CD155-expressing CHO-K1 cells upon TCR engagement in Jurkat cells and can be increased in the presence of an anti-TIGIT antagonist antibody. To compare the effectiveness of different a-TIGIT clones in increasing Jurkat cell activation, the experiment was performed in the presence of increasing concentrations of antibodies and EC ₅₀ values were calculated.

CD155 aAPC/CHO-K1 cells (Promega, CS198811) were seeded according to the manufacturer's recommendations and incubated at 37°C in a 5% CO2 O/N incubator. The next day, TIGIT effector cells (Promega, CS198811) were added according to the manufacturer's recommendations to CD155 aAPC/CHO-K1 cell plates containing fresh complete medium with anti-TIGIT antibody.

- 52 045372 in increasing concentrations (0.03; 0.11; 0 33; 1.06; 3.34; 10.56; 33.38; 105.49 and 333 nM) and incubated at 37°C, 5% CO2 for 6 h. After 6 h of incubation, activation of TIGIT effector cells was assessed by measuring luciferase activity using the BioGloTM Luciferase Assay System (Promega, G7941).

As shown in FIG. 24A and summarized in table. 20, anti-TIGIT antibody 31282 has the best performance in terms of EC ₅₀ value and maximum luciferase signal induction in the assay. The activity observed for the clone produced in HEK (31282_up) or CHO-K1 (31282_wu) cells is comparable to the maximum luciferase signal, which is 8 times higher than that of the isotype control (Bioexcell, BE0297), and to the EC ₅₀ concentration measured at 3.3 nM and 3.5 nM, respectively. In comparison, clones 4.1D3, 22G2 and 31C6 have maximum activity between 5.3 and 6.7 times the isotype control, associated with an EC ₅₀ between 5 and 10 nM. EC ₅₀ values for clone 313M32 and TIG1 could not be determined due to low activity and poor fit of the curves at the concentrations tested (Fig. 24A).

Table 20

EC ₅₀ data and comparison of different a-TIGIT clones for functional activity on Jurkat-hTIGIT cells

Name clone Induction on isotype control (multiple change) ECso (nM) Multiple change of ECso to ECso of the best clone (31282_pr) 31282_wu 8.4 3.5 1.1 31282_pr 8.0 3.3 1 313M32 / P.F. / 4.1D3 5.8 10.3 3.1 22G2 5.3 5.2 1.6 31С6 6.7 5.3 1.6 TIG1 / P.F. /

P.F.: bad fit.

B. Functional analysis of TIGIT on primary human CD8+ T cells from healthy volunteers

To characterize the functional consequences of blocking the human TIGIT receptor, we cocultured primary human CD8+ T cells from PBMCs from healthy human donors with the CHO-K1 cell line engineered to express human PVR/CD155 and activate human T cells. We observed that IFNg release by CD8+ T cells in the presence of engineered CD155-expressing CHO-K1 cells could be increased by blocking hTIGIT with antagonistic anti-TIGIT antibodies. To compare the effectiveness of these antibodies in increasing IFNg release, the experiment was performed in the presence of increasing concentrations of antibodies and EC ₅₀ values were calculated.

CD155 aAPC/CHO-K1 cells (Promega, CS198811) were seeded in 96-well U-bottom plates according to the manufacturer's recommendations and incubated at 37°C in a 5% CO2 O/N incubator. The next day, CD8+ T cells were purified according to the manufacturer's recommendations using a negative selection kit (Stemcell Technologies, 17953) from frozen human peripheral blood mononuclear cells isolated from the blood of healthy donors (Immunehealth). Purified CD8 T cells and increasing concentrations (0.011 nM, 0.033 nM, 0.11 nM, 0.33 nM, 1.06 nM, 3.3 nM, 10.6 nM, 33.3 nM, and 105.5 nM) antibodies were then added to CD155 aAPC/CHO-K1 (100,000 CD8 T cells/100 μl complete medium containing antibodies) and incubated at 37°C, 5% CO ₂ for 5 days. Finally, IFNg concentrations were assessed in the cell supernatant using an ELISA assay (Affymetrix eBioscience, 88-7316-86), which was performed according to the manufacturer's recommendations.

As shown in FIG. 24B and summarized in table. 21, a-TIGIT clone 31282 and 4.1D3 demonstrate better induction of IFNg secretion with corresponding increases of 2.7- and 2.9-fold compared to the isotype control antibody. Clone 31282 has the best efficiency for inducing IFNg production in terms of EC ₅ 0 concentration, which was measured at 0.13 nM. Clone 31C6 exhibits a 2.3-fold different _EC50 value, while clone 22G2 and 4.1D3 are 3.1- and 10.8-fold less potent than clone 31282. A value cannot be determined for clone 313M32 due to low activity and poor fit of the curves at the tested concentrations (Fig. 24B).

- 53 045372

Table 21

EC50 data and comparison of different a-TIGIT clones for functional activity on primary human CD8+ T cells

Clone Induction on isotype control (fold change) ECso (nM) Multiple change of ECso to ECso of the best clone (31282_wu) 31282_wu 2.7 0.13 1 313M32 / P.F. / 4.1D3 2.9 1.43 10.8 22G2 1.5 0.41 3.1 31С6 1.6 0.30 2.3

C. Functional analysis of TIGIT on primary human CD3+ T cells from cancer patients.

To characterize the functional consequences of blocking the human TIGIT receptor on T cells from cancer patients, primary human CD3+ T cells from cancer patient PBMCs were cocultured with the CHO-K1 cell line engineered to express human PVR/CD155 and activate human T cells (CHO -TCR-CD155). We observed that IFNg release by CD3+ T cells in the presence of engineered CD155-expressing CHO-K1 cells could be increased by blocking hTIGIT with the anti-TIGIT antagonist antibody 31282.

CD155 aAPC/CHO-K1 cells (Promega, CS198811) were seeded in 96-well U-bottom plates according to the manufacturer's recommendations and incubated at 37°C in a 5% CO2 O/N incubator. The next day, CD3+ T cells were purified according to the manufacturer's recommendations using a negative selection kit (Stemcell Technologies, 17951) from fresh human peripheral blood mononuclear cells isolated from whole blood from a cancer patient (HNSCC) collected over 24 h (Biopartners ). Purified CD3+ T cells and 66.7 nM antibodies were then added to CD155 aAPC/CHO-K1 (100,000 CD3 T cells/100 μl complete medium containing antibodies) and incubated at 37°C, 5% CO2 for 5 days. Finally, IFNg concentrations were assessed in the cell supernatant using an ELISA assay (Affymetrix eBioscience, 88-7316-86), which was performed according to the manufacturer's recommendations.

As shown in FIG. 24C, antibody 31282 induced potent functional activity to increase IFNg secretion, demonstrating the ability of this a-TIGIT antibody to reactivate PBMC T cells from cancer patients.

D. Clone a-TIGIT 31282 increases intracellular cytokine production in T cells from cancer patient PBMCs and dissociated tumor cells (DTCs).

In this example, intracellular flow cytometry staining was performed to evaluate T cell cytokine production from freshly isolated matched PBMCs and tumor infiltrated lymphocytes in dissociated tumor cells (DTC) from patients with renal cell carcinoma cancer. For DTC, tumors were minced mechanically and then incubated with a tumor dissociation kit (Miltenyi Biotech #130-095-929) while rotating in a gentleMACS dissociator, following the manufacturer's instructions for specific tumor types. Cells were stimulated for 16 h with a cocktail of T cell stimulation beads (Dynabeads, Thermo Fisher) before intracellular staining was performed. During the last 3 h of stimulation, a protein transport inhibitor cocktail (eBioscience) and a cell stimulation cocktail (eBioscience) were added to the cells. Conjugated antibodies were purchased from Ebioscience/Thermo Fisher Scientific, BioLegend, or BD Biosciences. Surface staining was performed according to the manufacturer's instructions using filtered FACS buffer (PBS + 2 mM EDTA + 0.1% BSA) and Brilliant Stain buffer (BD #563794). Cells were blocked with the appropriate human FcBlock (BD #564220) prior to surface staining. For intracellular staining, cells were fixed and permeabilized using BD Cytofix/cytoperm solution (BD Biosciences). Cells were stained with the following panel of antibodies: anti-CD45-BB515 (clone HI30, BD Horizon 564585), anti-CD73-BV421 (clone AD2, BD Horizon 562430), anti-CD8a-BV510 (clone SK1, BD Horizon 563919), anti- CD3-BV650 (clone SK7, BD Horizon 563999), anti-IFNγ-BV711 (clone 4S.B3, BD Horizon 564793), anti-IL-2-APC (MQ1-17H12, eBioscience 17-7029-82), anti- CD4-APCR700 (clone RPA-T4, BD Horizon 564975), LVD efluor 780 (eBioscience 65-0865-14), anti-TIGIT-PE (clone MBSA43, eBioscience E13456-108), anti-CD39-PE-Dazzle594 (clone A1, Biolegend 328224) and TNFα-PEcy7 (clone Mab11, eBioscience 25-7349-82). Acquisition was performed on a FACS Fortessa (BD Biosciences) and analyzed using FlowJo software (FlowJo, LLC). Viable cells were gated on forward and side scatter. T cell subsets were gated as follows: CD45 ⁺ CD3 ⁺ for PBMCs and CD45 ⁺ CD3 + CD4 + and CD45 ⁺ CD3 ⁺ CD8 ⁺ for DTC. Cytokine-secreting T cells were gated using unstained and unstimulated

- 54 045372 controlled controls.

In fig. 24D demonstrates that intracellular IL2, IFNg and TNFα all increased upon activation in the presence of a-TIGIT clone 31282. This increase was observed in CD3+ T cells from PBMCs consistent with the data illustrated in FIG. 24C, as well as in CD4+ and CD8+ TILs from dissociated tumor cells. This demonstrates the ability of a-TIGIT clone 31282 to increase the activation of PBMC and TIL populations from T cells from cancer patients.

Example 21: a-TIGIT clone 31282 induces preferential Treg cytotoxicity in PBMCs from cancer patients.

In this example, isolated PBMCs from a lung cancer patient were resuspended in complete RPMI medium (supplemented with 10% heat-inactivated FBS + 50 U penicillin + 50 U streptomycin). 2.5x105 human PBMCs were distributed per well in a 96-well plate. Human anti-TIGIT antibody clone 31282, human IgG1 isotype control (BioXcell BE0297), or rituximab (InvivoGen hcd20-mab1) was added at a final concentration of 6.6 nM to each respective well. Cells were incubated for 20 h at 37°C with 5% CO ₂ . Cells were then collected and stained with the following panel of antibodies: LVD efluor 520 (eBioscience 65-0867-14), anti-TCRab-PercP-Cy5.5 (clone IP26, Biolegend 306723), anti-CD4-BV510 (clone SK3, BD Horizon 562970 ), anti-CD8-APC-Cy7 (clone SK1, Biolegend 344714), anti-CD25-BV605 (clone 2A3, Biolegend 562660), anti-CD127-APC (A019D5, Biolegend 351316), anti-CCR7-BV421 (clone G043H7, Biolegend 353207) and anti-CD45RO-PE-Cy7 (clone UCHL1, Biolegend 304229). Results are presented on gated live cells. Absolute quantitation is performed using AccuCheck Counting beads (Life technologies) according to the manufacturer's instructions. After calculating the absolute number of cells per µl, % specific lysis is calculated using the following formula = (1-(absolute number of cells per µl on the sample treated with TIGIT 31282 antibody / average of samples treated in triplicate with isotype control)) x 100. Results are presented as the mean % specific lysis of triplicates +/- SD. The cytotoxic activity of ADCC/ADCP effector cells was assessed by measuring % specific cell lysis on gated CD19+ cells when incubated with rituximab.

As shown in FIG. 25, anti-TIGIT clone 31282 triggers higher specific lysis on Treg cells (30.1 +/- 3%) than on CD45RO ⁺ CCR7 ^+/- CD8 ⁺ T cells (general memory CD8+ T cells) (-1 .48 +/-6%) or CD45RO ⁺ CCR7 ^+/- CD4 ⁺ T (common memory CD4+ T cells) (0.64 +/-3%). Positive control rituximab triggered 77.9% (+/-6.8%) specific lysis on gated CD19 ⁺ cells. Overall data demonstrate preferential depletion of Treg cells from PBMCs of cancer patients compared with general CD4+ and CD8+ memory T cell populations. A similar preferential depletion of Treg cells was observed using cells from a patient with colon adenocarcinoma.

Example 22. Characterization of TIGIT expression on immune populations of PBMCs from cancer patients and dissociated tumor cells.

Flow cytometry analyzes were performed to evaluate TIGIT expression on immune cell subsets from freshly isolated matched PBMCs and tumor infiltrated lymphocytes in dissociated tumor cells (DTCs) from cancer patients. Samples were obtained from different indications: ovarian cancer, kidney cancer, HNSSC, skin cancer, melanoma and lung cancer. For DTC, tumors were minced mechanically and then incubated with a tumor dissociation kit (Miltenyi Biotech #130-095-929) while rotating in a gentleMACS dissociator, following the manufacturer's instructions for specific tumor types. PBMCs were isolated from whole blood using density gradient media (Lymphoprep Axis-Shield # 1115758). Phenotyping data were compared with frozen PBMCs isolated from healthy individuals (n=10).

Cells were stained according to the manufacturer's instructions using filtered FACS buffer (PBS + 2 mM EDTA + 0.1% BSA) and Brilliant Stain buffer (BD #563794). Cells were blocked with the appropriate human FcBlock (BD #564220) before staining and fixed using IC Fixation Buffer (eBioscience #00-8222-49) until prepared. DTC was stained with the following panel of antibodies: anti-CD45-BB515 (clone HI30, BD Horizon 564585), anti-CD73-BV421 (clone AD2, BD Horizon 562430), anti-CD8a-BV510 (clone SK1, BD Horizon 563919), anti- CD3-BV650 (clone SK7, BD Horizon 563999), anti-CD56-BV711 (clone 5.1H11, Biolegend 362542), anti-CD279-BV785 (clone EH12.2H7, Biolegend 329930), anti-CD127-APC (clone A019D5, Biolegend 351316), anti-CD4-APC-R700 (clone RPAT4, BD Horizon 564975), LVD efluor780 (eBioscience 65-0865-14), anti-TIGIT-PE (clone MBSA43, eBioscience E13456-108), anti-CD39- PE-Dazzle594 (clone A1, Biolegend 328224) and CD25-PE-cy7 (clone BC96, Biolegend 302612). PBMCs were stained with the following panel of antibodies: anti-CD45RO-BB515 (clone UCHL1, BD Horizon 564529), anti-CD73-BV421 (clone AD2, BD Horizon 562430), anti-CD8a-BV510 (clone SK1, BD Horizon 563919), anti- CD3-BV650 (clone SK7, BD Horizon 563999), anti-CD56-BV711 (clone 5.1H11, Biolegend 362542), anti-CD197-BV786 (clone 3D12, BD Horizon 563710), anti-CD127-APC (clone A019D5, Biolegend 351316), anti-CD4-APC-R700 (clone RPA-T4, BD Horizon 564975), LVD efluor 780 (eBioscience

- 55 045372

65-0865-14), anti-TIGIT-PE (clone MBSA43, eBioscience E13456-108), anti-CD39-PE-Dazzle594 (clone A1, Biolegend 328224) and CD25-PE-cy7 (C1 bone BC96, Biolegend 302612) . Acquisitions were performed on a FACS Fortessa (BD Biosciences) and analyzed using FlowJo software (FlowJo, LLC). Viable cells were gated on forward and side scatter. Different immune cell subsets were gated as follows: CD3⁺ CD4+ CD127+ CD25 (CD3⁺ CD4+ non-Treg cells), CD3⁺ CD4⁺ CD^™™ CD25+ (regulatory T cells), CD3⁺ CD8⁺ (CD3+ CD8+ T cells), CD3^- CD56+ (NK cells), CD3⁺ CD56+ (NKT cells), CD3^- CD56'(non-T/NK cells). Quantibrite PE beads (BD #340495) were used at the same instrument settings and were used to convert fluorescence data to the amount of bound antibody per cell.

The frequency of TIGIT expression in different immune populations is presented in Fig. 26A, and the TIGIT density for each subset is presented in FIG. 26B using Box and Wixer representation using Tukey's method to calculate percentiles.

Data show that the frequency of TIGIT on a subset of T cells is higher in PBMCs from cancer patients compared to PBMCs from healthy donors. This frequency is further increased by DTC TILS (FIG. 26A). While the same observation was made when examining TIGIT density on the surface of CD3 ⁺ CD4+ non-Treg cells and CD4 ⁺ Treg cells, for CD3 ⁺ CD8+ T cells the number of TIGIT molecules per cell decreases at DTC TILS (Figure 26B) .

Example 23. Structural and functional mapping of the TIGIT epitope and clone 31282.

To further characterize and understand the interaction between anti-TIGIT mAb clone 31282 and recombinant TIGIT protein, the crystal structure of 31282 complexed with TIGIT was determined by X-ray diffraction.

A. Expression, purification and crystallization of TIGIT and Fab.

Human TIGIT residues 23–128 were from Proteros Biostractures GmbH. TIGIT (23-128) with an N-terminal HIS tag (thrombin-cleavable) was cloned into pET15b and expressed in LB medium in B121 (DE3) at 37°C in the inclusion body. Inclusion bodies (IB) were washed with buffer containing Tris/HCl pH 7.4 and Tris/HCl pH 7.4, 0.05% Brij-35. IBs were denatured with 6 M Gdm/HCl, 50 mM Tris, pH 8.5, and 10 mM DTT. Refolding was performed in 50 mM Tris/HCl, pH 8, 1 mM GSH, 0.5 mM GSSG, 150 mM NaCl. The refolded protein was purified using a HIS trap. The N-terminal HIS tag was removed by thrombin digestion and further purification on Superdex-75 equilibrated in 50 mM Tris/HCl pH 7.5, 200 mM NaCl.

For Fab fragment expression, HEK293F cells were grown in Freestyle F17 with 1% penicillin/streptomycin, 2 mM L-glutamine, and 0.1% Pluronic. Extended transfection cultures were grown in 3 L Erlenmeyer flasks (Corning, 2 L, cell culture working volume, 37°C, 8% v/v CO ₂ , 80-120 rpm, amplitude 50 mm). The culture was diluted one day before transfection, and the cell number was adjusted to 1 × 10 ^b cells/ml. The expression culture volume was 6 liters. Transient transfection was performed with plasmids for light and heavy chain Fab. MasterMix DNA/FectoPro (FectoPro, PolyPlus) was prepared in pure F17 medium and incubated for 10 minutes (according to the PolyPlus protocol). This transfection mixture was added dropwise to the cell suspension, and the booster was added immediately. Eighteen hours after transfection, the culture was fed with 3 g/L glucose.

For Fab fragment purification, 6 L of HEK293 cell culture supernatant was collected by centrifugation 6 days after transfection and applied to a 30 mL KappaSelect column. KappaSelect was washed with PBS pH 7.4, eluted with sodium citrate pH 3, and fractions containing Fab were neutralized with Trisbuffer. The Fab was then purified on a Superdex S-200 column equilibrated in 20 mM Tris, pH 8, 100 mM NaCl, and stored at -80°C until further use.

To form the Fab-TIGIT complex, purified TIGIT was mixed with purified Fab in a 1.5:1 ratio, and the complex was purified on Superdex-200 equilibrated in 20 mM Tris, pH 8, 100 mM NaCl. The Fab-TIGIT complex was concentrated to 35 mg/mL for crystallization. The Fab-TIGIT complex was crystallized at 277 K using the vapor diffusion method by mixing 0.1 μL of protein solution (35.3 mg/mL in 20 mM Tris pH 8.0; 100 mM NaCl) in a 1:1 ratio with the reservoir solution (0.10 M sodium cacodylate, pH 6.00; 15% (w/v) PEG4000). The crystals were cryo-protected by immersing them in a reservoir solution supplemented with 25% glycerol.

B. Data collection and processing.

The cryo protocol was established using standard protocols from Proteros Biostractures GmbH. The crystals were flash frozen and measured at 100 K. X-ray diffraction data were obtained from Fab:TIGIT crystals at SWISS LIGHT SOURCE (SLS, Villigen, Switzerland) under cryogenic conditions. The crystals belong to space group P1. Data were processed using XDS and XSCALE programs. Statistics for data collection and processing can be found in table. 22.

- 56 045372

Table 22

Statistics of data collection and processing_____________________

X-ray source PXII/X10SA (SLS ¹ ) Wavelength [A] 1.0000 Detector PILATUS 6M Temperature [K] 100 Space group Pl Cell: a; b; With; [A] α; β; γ; Π 41.73; 71.46; 110.26 96.7; 95.8; 106.5 Resolution [A] 2.31 (2.56-2.31) Unique Reflections 50537 (13271) Plurality 2.0 (1.9) Completeness[%] 96.1 (95.3) Yasin [%] 8D (43.5) ^average [%] 11.0 (59.1) CpenHee(I)/sd ³ 8.11(1.94)

¹ SWISS LIGHT SOURCE (SLS, Villigen, Switzerland).

The ² values in parentheses refer to the cell with the highest base resolution.

3_____ _________ ____ „ ____ „ is calculated from independent reflections.

C. Structural modeling and refinement.

The phase information needed for structure determination and analysis was obtained by molecular substitution. The previously solved Fab structure was used as a search model. Subsequent construction and refinement of the model was carried out in accordance with standard protocols with the CCP4 and COOT software packages. To calculate the free R-factor, a measure to cross-check the validity of the final model, about 2.5% of the measured reflections were excluded from the refinement procedure (see Table 23).

TLS was refined (using REFMAC5, CCP4), which led to a decrease in R factors and an increase in the quality of electronic map density. Automatically generated local NCS restrictions were applied (keyword ncsr local in newer versions of REFMAC5). Parameterization of ligands and generation of corresponding library files were performed using CHEMSKETCH and LIBCHECK (CCP4), respectively.

The water model was built using the COOT Water Finder algorithm, placing water molecules in the peaks of the Fo-Fc map with contour 3.0, followed by refinement using REFMAC5 and validating all waters using the COOT validation tool. The criteria for the list of suspected waters were: B-factor greater than 80 A2, 2Fo-Fc map less than 1.2 σ, distance to nearest contact less than 2.3 A or more than 3.5 A. Suspected water molecules and molecules in the site Ligand binding (ligand distance less than 10 Å) was checked manually. The final complex structure was refined using PHENIX. We selected a refinement parameter including XYZ coordinates, real space, individual B-factors, and group B-factors. X-ray/stereochemical weight optimization and NCS constraints were also selected for refinement. The Ramachandran plot of the final model shows 95.39% of all residues in the preferred region, 3.95% in the permitted region. Statistics of the final structure and refinement process are given in Table. 23.

- 57 045372

Table 23

Refinement statistics ¹

Resolution [A] 108.40-2.31 Number of reflections (working/test) 49289/1247 Rpa6onee 0.2025 ^free [%] 0.2466 Total number of atoms: Protein Water 8282 676 Deviation from ideal geometry: ³ Bond length [A] Bond angles [deg] 0.003 0.771 Ramachandra Chart: ² Preferred Areas [%] Permitted Areas [%] Prohibited Areas [%] 95.39 3.95 0.66

¹ values as defined in PHENIX.

² calculation with COOT.

D. General structure.

The heavy and light chains of the human antibody Fab fragment show typical human antibody folding (FIG. 27A). The asymmetric unit contains two heterotrimers with the same overall conformation. The model contains residues 23 to 128 of TIGIT, residues 1 to 224 of the heavy chain of clone 31282, and residues 1 to 214 of the light chain of clone 31282. One region of the short loop of the heavy chain is not fully defined by electron density and is therefore not included in the model.

Diffraction images were analyzed using FoldX software to evaluate the energetic contribution of the residues and identify interaction hotspots. The amino acid residues forming the binding interface are well defined in the electron density map. The interpreted X-ray diffraction data clearly shows the interaction between Fab and TIGIT (FIGS. 27B and 27C). The light chain CDRs of clone 31282 interact with 2 TIGIT regions with CDRs L1 Arg30 and Tyr33, binding to TIGIT residues Asn58 and Glu60; with CDR L1 Arg30 and CDR L3 Phe93 binding to the TIGIT residue Ile109. CDR L2 has no contact with TIGIT (Table 24). Heavy chain clone 31282 interacts with various regions of TIGIT with CDR H1 Tyr33, binding to TIGIT at the Leu73 residue; with CDR H2 Val50, Ser54 and Ser57 binding to TIGIT at the Leu73 residue; with CDR H3 Asp102, Tyr103 and Trp104 binding to TIGIT at residue Gln56, Ile68, Leu73 and His76.

Based on this crystal structure of the clone 31282/TIGIT a-TIGIT complex, the TIGIT residues that contact clone 31282 (the epitope residues for TIGIT associated with clone 31282), and the residues of clone 31282 that contact TIGIT (the paratope residues for clone 31282, associated TIGIT) were identified. In table 24 and 25 and in FIG. 27C demonstrates TIGIT residues in contact with light chain (Table 24) or heavy chain (Table 25) residues of clone 31282. Contact residues were defined as each amino acid meeting each of the following criteria: (i) it has a calculated binding free energy contribution greater than than 0.3 kcal/mol, (ii) it has an experimental average B-factor less than the average value of the B-factor of all residues in the X-ray structure, (iii) it creates at least 3 pairs of interatomic contacts of heavy atoms with antibody atoms on distance less than or equal to 4.0 Angstroms, (iv) it does not form only solvent-exposed hydrogen bonds or ionic interactions, (v) if it is a non-aromatic polar moiety (Asn, Gln, Ser, Thr, Asp, Glu, Lys or Arg) creates at least one hydrogen bond or ionic interaction with the antibody.

- 58 045372

Table 24

Summary of TIGIT epitope residues and corresponding paratope residues on the light chain of clone 31282

Amino acid TIGIT Light chain amino acid clone 31282 Asn 58 Tug 33 Gin 60 Arg 30 Tug 33 Not 109 Arg 30 Phe 93

Table 25

Summary of TIGIT epitope residues and corresponding paratope residues on the heavy chain of clone 31282

Amino acid TIGIT Heavy chain amino acid clone 31282 Gin 56 Tgr 104 Not 68 Thug 103 Tgr 104 Leu 73 Tug 33 Vai 50 Ser 54 Vai 50 His 76 Asp 102 Thug 103 Tgr 104

Example 24. Analysis of competition between a-TIGIT clones 31282 and 32959.

Human IgG1 anti-TIGIT antibody clone 32959 was produced in HEK cells and purified as described in Example 17 above.

Jurkat cells overexpressing human TIGIT (Jurkat-hTIGIT) were collected and distributed at 5x10 ⁴ cells/well and incubated with the a-TIGIT antagonist clone 31282 at the following concentrations: 0 nM (no Ab), 0.08 nM, 0.16 nM, 0.8 nM and 8 nM, which represent the concentration range from 0 to 100 greater than the Kd of this clone. Excess antibodies were washed off and cells were incubated with decreasing concentrations (8; 4; 2; 1; 0.5; 0.25; 0.125; 0.062; 0.031; 0.016; 0.008 and 0.004 nM) directly linked (AF647) anti-TIGIT clone 32959 for 30 min at 4°C. The geometric mean fluorescence intensity was analyzed using the LSR BD Fortessa. Cell binding was recorded as the average fluorescence intensity of AF647. To calculate the EC50 of binding of clone 32959, the half maximum binding concentration (EC50) of hTIGIT-Jurkat was calculated using a four-variable curve fitting equation in Prism, and the resulting values are shown in Table 1. 26 and illustrated in FIG. 28. The results show strong binding of a-TIGIT clone 32959 regardless of the concentration of clone 31282, demonstrating no competition with the a-TIGIT antagonist antibody.

- 59 045372

Table 26

EC50 concentrations for binding of a-TIGIT clone 32959 to Jurkat-hTIGIT in the presence of increasing concentrations of the antagonist a-TIGIT clone 31282

a-TIGIT 31282 at 0 nM a-TIGIT 31282 at 0.08 nm a-TIGIT 31282 at 0.16 nM a-TIGIT 31282 at 0.8 nM a-TIGIT 31282 at 8 nm ECso binding (nM) for a-TIGIT clone 32959 0.22 0.33 0.37 0.49 0.39 Human Jurkat cell binding TIGIT FON (Multiplicity compared to negative control) for a- TIGIT clone 32959 588

Example 25 Determination of the pharmacokinetic profile of clone 31282 after a single intravenous injection in a Javanese monkey.

Cynomolgus monkeys received a-TIGIT clone 31282 IgG1 or IgG4 by intravenous bolus injection. The antibody was administered at 3 different concentrations (0.1 mg/kg; 1 mg/kg; 10 mg/kg) to 2 animals (1 male and 1 female). Blood was collected 504 hours post-dose on day 1. Blood samples were processed to obtain plasma and analyzed for α-TIGIT clone 31282 IgG1 or IgG4 concentrations using an ELISA method.

Plasma concentration-time data for individual animals were used to calculate toxicokinetic parameter values for a-TIGIT IgG1 and IgG4 31282 clones after i.v. using the intravascular model in Phoenix WinNonlin (version 6.3, Pharsighl, a Cerlara company, Princeton, New York). Jersey).

Following IV bolus administration of a-TIGIT 31282 IgG1 and IgG4 clones at doses of 0.1, 1, and 10 mg/kg, IgG1 concentrations were quantified in the plasma of male and female monkeys at 240 h, 336 h, and 504 h postdose, respectively. , and IgG4 were quantified after 168 hours, 240 hours and 504 hours, respectively (Fig. 29 and Table 27). There were no apparent sex differences in systemic exposure (Cmax and AUClast) to IgG1 and IgG4 following IV bolus administration of a-TIGIT clone 31282 IgG1 or IgG4, with ratios (male/female) ranging from 0.855 to 1.16.

Following IV bolus administration of a-TIGIT clone 31282 IgG1 to male and female monkeys, plasma IgG1 concentrations decreased biphasically at all dose levels with a mean terminal half-life (t1/2) ranging from 84.7 to 174 hours (Fig. 29 ). Systemic clearance (CL) was similar for all doses tested, ranging from 0.280 to 0.392 ml/h/kg. Apparent volume of distribution at steady state (Vss) was similar among dose levels tested, with values ranging from 53.7 to 66.5 mL/kg. Increasing the dose of a-TIGIT clone 31282 IgG1 10-fold in the range of 0.1 to 1 mg/kg and 1 to 10 mg/kg resulted in an approximately proportional increase in exposure (9.57- to 14.5-fold increase).

Following IV administration of a-TIGIT clone 31282 IgG4 to male and female monkeys, plasma IgG4 concentrations decreased biphasically at all dose levels tested with t1/2 of 148-334 hours (Fig. 29 and Table 28). CL was consistent among dose levels tested, ranging from 0.160 to 0.219 mL/h/kg. The mean VSS ranged from 41.2 to 70.7 ml/kg. Increasing the dose of a-TIGIT clone 31282 IgG4 by a factor of 10 from 0.1 to 1 mg/kg and from 1 to 10 mg/kg resulted in an approximately proportional increase in IgG4 exposure (9.32 to 12.5-fold increase).

- 60 045372

Table 27

Summary of the average parameters of toxicocypegy for a-TIGIT bug 31282 human IgG1 after intravenous bolus administration in the Javanese monkey a-TIGIT clone 31282 human IgGl

.Ιο :а (μι κι) Whitefish (MKI 'm. 1) llll.IV (H) 0.1 2.34 IIV11|1I1 22.4 |II|D^ 1(1 2(18 AU (/last 224 2330 ZZ'on (h*mcg/ml) Ιι '(41 111 (Ί (ml hll) 0.202 11.302 0.280 V.. (ml/κι) 66.5 %** 53.’

Table 28

Summary of mean toxicokinetic parameters for human IgG4 a-TIGIT clone 31282 following IV bolus administration in the Javan monkey

a-TIGIT clone 31282 human IgG4 Do)a (μι member) o.1 2.81 DIM 1 IIYYY ΙΟ 283 1·^^ Speech (Ml 1......Cl) 1 ml) ATTOlatest (h*mcg/ml) 238 2690 i 39100 Ιι Pch) 334 ( 1 1m । · KI) 0.100 0.160 0.21(1 \„(ml Oilllllii D||M[ID 0. 5'.5

Example 26. Characterization of TIGIT expression in human tumor cell populations.

Flow cytometry analyzes were performed to evaluate TIGIT expression on normal and tumor T or B cells in blood samples from cancer patients with various blood cancer features.

Samples from Sézary syndrome patients were tested to compare TIGIT expression on malignant and normal CD4 T cell populations. To separate these populations, a preliminary determination of the TCR-Vb malignant clone rearrangement was performed using the Bcckman Coulter TCR-Vb kit (#IM3497). Once the malignant clone was identified, TIGIT expression was profiled on immune cells from Sézary syndrome patients using the following commercial reagents: anti-CD3 Krome Orange (#B00068), anti-CD4-PE (#A07751), anti-CD8- PC7 (#737661), anti-CD56-PC5 (#A07789), anti-CD45-Pacific Blue (#A74763), anti-CD19-AF750 (#A94681), and anti-Vb8-FITC (#IM1233) (all from Beckman-Coulter ) and anti-TIGIT-APC (clone MBSA43, ebiosciences #17-9500-42). Flow cytometric analyzes of samples from patients with Sézary syndrome were performed on a CytoFlex instrument (Beckman-Coulter). Data were analyzed using FloJo software (FlowJo, LLC).

A typical example is shown in FIG. 30A. The gating strategy for this TCR-Vb8 malignant clone donor is demonstrated in FIG. 30A, where the malignant cells are CD45-CD3 ⁺ CD4 ⁺ Vb8 ⁺ and the normal CD4+ T cells are CD45 ⁺ CD3 ⁺ CD4 ⁺ Vb8. Strong expression of TIGIT is observed on malignant CD4+ T cells compared to normal CD4+ T cells (corresponding to MFI 4987 and 999) (Fig. 30B).

Similarly, flow cytometry analyzes were performed to evaluate TIGIT expression on normal and malignant B cells in bone marrow samples from patients with CLL. Samples were stained with the following panel of antibodies: LVD efluor 780 (eBioscience 65-0865-14), anti-CD45BB515 (clone HI30, BD Horizon 564585), uhtu-CD5-BV510 (clone UCHT2, Biolegend 363381), anti-CD19BV711 (clone SJ25C1, BD Horizon 563036) and anti-TIGIT-PE (clone MBSA43, eBioscience E13456-108). Acquisition was performed on a FACS Fortessa (BD Biosciences) and analyzed using FlowJo software (FlowJo, LLC). Viable cells were gated on forward and side scatter. Different cell subsets were gated as follows: CD45 ⁺ CD19CD5− (normal B cells) and CD45− CD19− CD5+ (malignant B-CLL).

An illustrative example is shown in FIG. 31 with the gating strategy shown in FIG. 3A. A high proportion of malignant B-CLL cells are positive for TIGIT (75%), in contrast to normal B cells (1%) (MFI 1440 and 810, respectively) (Fig. 31B).

Overall, the findings demonstrate that tumor cells express TIGIT in certain blood cancer indications.

Example 27. Antitumor activity of an antagonistic anti-TIGIT antibody as monotherapy in a mouse T-cell lymphoma model.

For this experiment, EL4 T lymphoma cells (ATCC® TIB-39™) were engineered to stably express mouse TIGIT (EL4-mTIGIT). EL4 cells transduced with a similar vector encoding GFP were used as a control (EL4-GFP). Cell pools were subcloned to generate EL4-mTIGIT and EL4-GFP clones. The anti-TIGIT antibody used was a modified version of antibody 29527 (modified so that residue 27 of VH FR3 is mutated from L to V and where residue 6 of VH FR4 is mutated from M to T) and was produced on the human IgG1 isotype. Female Balb/c mice were inoculated subcutaneously with 1,000,000 EL4mTIGIT cells or 200,000 EL4-GFP cells for 8 weeks. On day 7 postinoculation, when tumor volumes averaged approximately 110 mm ³ , mice were randomized into treatment groups of equal tumor volume (n=15 per group for EL4-mTIGIT and n=10 per group for EL4-GFP). Mice were treated with 200 μg of anti-TIGIT or isotype control antibody (hIgG1, BioXcell) by intraperitoneal injection on days 7, 10, 13, and 16 after tumor inoculation. Tumor growth was monitored and tumor volumes were measured using electronic calipers three times a week from days 7 to 26. Mice were sacrificed when tumor volume exceeded 2000 mm ³ . Tumor growth curves were statistically analyzed using a linear mixed model. Differences between treatment groups were assessed by testing the time*treatment group interaction.

Fig. 32 illustrates tumor growth curves in mice inoculated with EL4-mTIGIT (AC) or EL4-GFP (DF). Shown are median tumor growth curves (A and D) as well as individual tumor growth curves for mice treated with isotype control hIgG1 (B and E) or antagonistic aTIGIT Ab (C and F). Mice inoculated with EL4-mTIGIT cells showed significant suppression of tumor growth when treated with anti-TIGIT Ab compared to the isotype control group (p<0.001). Whereas in the group receiving the isotype control antibody, 3 of 15 mice demonstrated tumor growth control with a volume below 700 mm ³ at the end of the model, this number was increased to 8 of 15 mice in the group receiving the anti-TIGIT antibody antagonist. Antitumor efficacy or complete response could not be detected in mice bearing the EL4-GFP tumor when treatment was compared with the isotype control antagonist a-TIGIT antibody. Together, these data demonstrate that the a-TIGIT antagonist antibody (hIgG1) has significant antitumor efficacy in a TIGIT-expressing tumor cell model.

Example 28. Antitumor activity of antagonistic anti-TIGIT antibodies in combination with immune checkpoint antibodies in the CT26 mouse model of colon carcinoma.

In addition to the combination of anti-TIGIT Ab with anti-PD1 antibody (Examples 12, 13 and 14), the antitumor efficacy of anti-TIGIT Ab was also evaluated in combination with agonist antibodies specific for the co-stimulatory molecules 4-1BB, OX40 and GITR, and also with an antagonist antibody specific for the ICOS checkpoint inhibitory molecule.

The CT26 tumor cell line was purchased from ATCC® (CRL-2638™). Female balb/c mice were inoculated subcutaneously in the right flank with 500,000 cells for 8 weeks. On day 9 postinoculation, when tumor volumes averaged approximately 75 mm ³ , mice were randomized into treatment groups of equal tumor volume (n=10 mice per group). All antibodies were administered intraperitoneally every 3 days, starting from the day of randomization, for a total of 3 injections. The anti-TIGIT antibody used was a modified version of antibody 29527 (modified such that residue 27 of VH FR3 is mutated from L to V and where residue 6 of VH FR4 is mutated from M to T) and was produced on the mouse IgG2a isotype, which was given at a dose of 20 µg/mouse. Anti-4-1BB (clone 3H3, BioXCell, BE0239) was administered at a dose of 5 μg/mouse, a-OX-40 (clone OX-86, BioXCell, BE0031) was administered at a dose of 20 μg/mouse, a-GITR (clones DTA -1, BioXCell, BE0063) was administered at a dose of 10 μg/mouse; and a-ICOS (clone 7E.17G9, BioXCell, BE0059) were administered at a dose of 200 μg/mouse. Tumor growth was monitored and tumor volumes were measured using electronic calipers three times a week from days 7 to 35. Mice were sacrificed when tumor volume exceeded 2000 mm ³ . Tumor growth curves were statistically analyzed using a linear mixed model on log-transformed tumor volumes. Differences between treatment groups were assessed by testing the time*treatment group interaction. This resulted in a good model fit for the vast majority of data, except for very small tumor volumes (below 10 mm ³ ). Therefore, these small tumor volumes were treated as missing values. To test for a synergistic effect resulting from combining an anti-TIGIT antibody with a corresponding immune checkpoint antibody (IC—i.e., anti-41BB, anti-OX40, anti-GITR, and anti-ICOS), treatment groups were recoded using a combination of the two variables; anti-TIGIT (yes/no) and IC (yes/no). The synergistic effect, in addition to the additive effect of each treatment (anti-TIGIT*time and IC*time), was assessed by testing the interaction term anti-TIGIT*IC*time.

In fig. 33A shows median tumor growth curves per group, as well as individual growth curves for mice that received anti-TIGIT alone or in combination with anti-TIGIT.

- 62 045372

4-1BB. There was significant suppression of tumor growth in mice treated with anti-TIGIT + anti4-1BB compared with anti-TIGIT or anti-4-1BB monotherapy (p=0.0005 and p<0.0001, respectively). The combination of anti-TIGIT antibodies and anti-4-1BB antibodies gave 6/10 mice a complete response (tumor <30 ^mm3 and considered not measurable) compared to a complete response of 1/10 or 0/10 in the a-treated groups, respectively. TIGIT or a-4-1BB as sole agent. These data demonstrate significant antitumor efficacy of anti-TIGIT therapy in combination with anti-4-1BB for the treatment of established tumors.

In fig. 33B shows median tumor growth curves per group, as well as individual growth curves for mice that received anti-TIGIT alone or in combination with antiOX-40. There was significant suppression of tumor growth in mice treated with anti-TIGIT + antiOX-40 compared with anti-TIGIT or anti-OX-40 alone (p=0.0002 and p<0.0001, respectively). The combination of anti-TIGIT+anti-OX-40 achieved synergistic antitumor efficacy that was greater than the additive effect of both monotherapies (p=0.02). The combination of anti-TIGIT and anti-OX-40 antibodies resulted in 7/10 mice having a complete response, compared with 1/10 or 0/10 complete responses in the a-TIGIT or a-OX-40 groups, respectively. as sole agent. These data demonstrate the significant and synergistic antitumor efficacy of anti-TIGIT therapy in combination with anti-OX-40 for the treatment of established tumors.

In fig. 33C shows median tumor growth curves per group, as well as individual growth curves for mice that received anti-TIGIT alone or in combination with antiGITR. There was significant suppression of tumor growth in mice treated with anti-TIGIT + antiGITR compared with anti-TIGIT or anti-GITR monotherapy (p<0.0001). The combination of antiTIGIT+anti-GITR achieved synergistic antitumor efficacy that was greater than the additive effect of both monotherapies (p=0.01). The combination of anti-TIGIT and anti-GITR antibodies resulted in 6/10 mice experiencing a complete response, compared with 1/10 or 0/10 in the groups receiving anti-TIGIT or anti-GITR as a single agent, respectively. These data demonstrate the significant and synergistic antitumor efficacy of anti-TIGIT therapy in combination with anti-GITR for the treatment of established tumors.

In fig. 33D shows median tumor growth curves per group, as well as individual growth curves for mice that received anti-TIGIT alone or in combination with anti-ICOS. There was significant suppression of tumor growth in mice treated with anti-TIGIT + anti-ICOS compared with anti-TIGIT or anti-ICOS monotherapy (p=0.003 and p=0.0001, respectively). The combination of anti-TIGIT and anti-ICOS antibodies resulted in 1/10 mice having a complete response (where tumor is <30 ^mm3 and considered not measurable) compared to 1/10 or 0/10 in the groups receiving respectively anti-TIGIT or anti-ICOS antibodies as the sole agent. These data demonstrate the significant and synergistic antitumor efficacy of anti-TIGIT therapy in combination with anti-ICOS for the treatment of established tumors.

Example 29: Activity of antagonistic anti-TIGIT antibody on γδ T cells.

γδ (gamma delta or g/d) T cells are a population of unusual T cells with described antitumor activity (Zhao et al. 2018. J Transl Med. 16:122) and antiviral activity (eg, CMV infection), and are also implicated in autoimmune diseases (Malik S et al. 2016. Front Immunol. 7:14).

Flow cytometry analysis was performed to evaluate TIGIT expression on γδ T cells on PBMCs freshly isolated from pitomegalovirus (CMV) seronegative or seropositive healthy individuals (CMV status assessed by EFS Nouvelle Aquitaine, Bordeaux, France). Cells were stained according to the manufacturer's instructions using filtered FACS buffer (PBS + 2 mM EDTA + 0.1% BSA). Acquisition was performed on a FACS Fortessa (BD Biosciences) and analyzed using BD FACS DIVA software (BD Biosciences). Cells were gated for forward and side scatter and viability. γδ T cells were gated as follows: CD3+ TCRγδ ⁺ Vδ2'(Vδ2' γδ T cells) using the following antibodies: anti-TCR γδ APC, clone REA591 #130-109-280 from Miltenyi; anti-TCR Vδ2-PE-Vio 770, clone REA771, #130-111012 from Miltenyi; BV421 mouse CD3 anti-human, clone UCHT1, #560365 from BD Biosciences; Zombie Aqua Fixable Vitality Kit, #423101 from Biolegend.

Like conventional αβ T cells, unusual Vδ2' γδ T cells express TIGIT in both CMV-negative and positive human populations (anti-TIGIT, clone MBSA43, #12-98500-42 from eBioscience) (Fig. 34A). To characterize the functional consequences of blocking the TIGIT receptor in this cell population, magnetically isolated Vδ1 ⁺ γδ T cells (anti-TCR Vd1-FITC, clone REA173 #130-100-532 and anti-FITC Microbeads #130-048-701 both from Miltenyi ) or total PBMCs from CMV-positive donors were activated with anti-Vδ1 (10 μg/ml) (clone R9.1, #IM1761 from Beckman Coulter) and IL15 (20 ng/ml), #200-15-50UG from Peprotech), IL- 2 (100 U/ml, #200-02-1MG Peprotech) was further added to isolated Vδ1 ⁺ γδT cells in the presence or absence of TIGIT ligand CD155

- 63 045372 (#9174-CD-050 from R&D Systems). In fig. 34B demonstrates a dose-dependent reduction in IFNu secretion (ELISA kit, #3420-1h-20 from Mabtech) mediated by the addition of TIGIT ligand CD155 (0, 0.1, 1 and 10 μg/ml) with maximum inhibition achieved at a concentration of 1 μg/ml CD155. Addition of anti-TIGIT Ab clone 31282 (10 μg/ml) fully restored IFNu production to levels equal to or greater than the CD155 ligand-free condition, whereas the human IgG1 isotype control had a very limited effect. In fig. 34C demonstrated a similar inhibitory effect mediated by CD155 (10 μg/ml) after anti-Vδ1 activation of total PBMCs and complete restoration of IFN secretion when clone 31282 was added to the a-TIGIT mixture. These data demonstrate that, like αβ T cells, γδ activity T cells can be impaired by ligation of CD155 to TIGIT, and that anti-TIGIT antibodies completely prevent this inhibition.

Claims (17)

1. An isolated antibody or antigen-binding fragment thereof that binds to human TIGIT, wherein said antibody or antigen-binding fragment comprises a combination of HCDR1, HCDR2, HCDR3, LCDR1, LCDR2 and LCDR3, where: HCDR1 contains SEQ ID NO: 16 (YTFTSYYMH), HCDR2 contains SEQ ID NO: 17 (VIGPSGASTSYAQKFQG), HCDR3 contains SEQ ID NO: 18 (aRDHSDYWSGIMEV), LCDR1 contains SEQ ID NO: 61 (RASQSVRSSYLA), LCDR2 contains SEQ ID NO: 62 (GASSRAT), and LCDR3 contains SEQ ID NO: 63 (QQYFSPPWT). 2. The isolated antibody or antigen binding fragment thereof according to claim 1, wherein the heavy chain variable domain comprises an amino acid sequence shown as SEQ ID NO: 221, or an amino acid sequence exhibiting at least 90, 95, 97, 98 or 99% sequence identity with it, and the light chain variable domain contains the amino acid sequence shown as SEQ ID NO: 222 or an amino acid sequence exhibiting at least 90, 95, 97, 98 or 99% sequence identity with it. 3. An antibody or antigen-binding fragment according to any of the previous paragraphs, which is a human IgG antibody. 4. The antibody or antigen binding fragment according to claim 3, which is a human IgG1 antibody. 5. An isolated polynucleotide encoding an antibody or antigen-binding fragment according to any of the previous paragraphs. 6. An expression vector comprising a polynucleotide according to claim 5, operably linked to regulatory sequences that enable expression of the antibody or antigen binding fragment in a host cell or cell-free expression system. 7. A host cell containing the expression vector of claim 6. 8. A method for producing a recombinant antibody or antigen-binding fragment thereof, which includes culturing the host cell according to claim 7 under conditions that ensure expression of the antibody or antigen-binding fragment and isolation of the expressed antibody or antigen-binding fragment. 9. Use of an antibody or antigen binding fragment according to any one of claims 1 to 4 for the treatment of cancer. 10. A pharmaceutical composition containing an antibody or antigen-binding fragment according to any one of claims 1 to 4 and at least one pharmaceutically acceptable carrier or excipient. 11. Use of the pharmaceutical composition according to claim 10 for the treatment of cancer. 12. A method for stimulating T cell activity, comprising contacting a population of T cells with an antibody or antigen binding fragment according to any one of claims 1 to 4. 13. The method of stimulating T cell activity according to claim 12, wherein said method is carried out in vivo in a human subject, wherein said human subject has cancer. 14. A combination comprising an anti-TIGIT antibody or an antigen-binding fragment thereof and one or more of a chemotherapeutic agent, an anti-PD1 antibody, an anti-PD-L1 antibody, an anti-41BB antibody, an anti-OX40 antibody, an anti-GITR antibody, and an anti- ICOS antibodies, wherein the anti-TIGIT antibody or antigen binding fragment thereof is an antibody or antigen binding fragment according to any one of claims 1 to 4, for treating cancer. 15. Use according to claim 9 or 11, wherein the cancer is selected from the group consisting of: acute lymphoblastic leukemia, acute myeloid leukemia, B-cell lymphoma, bladder cancer, blood cancer, bone cancer, brain tumors, brain stem glioma , breast cancer, adrenal cancer, anal cancer, endocrine cancer, head or neck cancer, esophageal cancer, parathyroid cancer, penile cancer, small intestinal cancer, thyroid cancer, urethral cancer, cervical cancer, endometrial cancer , fallopian tube cancer, renal pelvis cancer, vaginal cancer, vulvar cancer, chronic lymphocytic leukemia, chronic myeloid leukemia, colon cancer, skin cancer, malignant melanoma of the skin or intraocular canal, epidermoid cancer, head and neck squamous cell carcinoma (HNSCC), lymphoma Hodgkin's, Kaposi's sarcoma, kidney cancer, lung cancer, lymphocytic lymphoma, mesothelioma, nasopharyngeal cancer, central nervous system (CNS) neoplasm, non-Hodgkin's lymphoma, ovarian cancer, pancreatic cancer, pituitary adenoma, primary CNS lymphoma, prostate cancer, rectal cancer colon, kidney cancer, sarcoma, skin cancer, spinal axis tumor, squamous cell carcinoma, gastric cancer, T-cell lymphoma, testicular cancer and uterine cancer. - 171 045372 16. The method of claim 13, wherein the cancer is selected from the group consisting of: acute lymphoblastic leukemia, acute myeloid leukemia, B-cell lymphoma, bladder cancer, blood cancer, bone cancer, brain tumors, brain stem glioma, cancer breast, adrenal cancer, anal cancer, endocrine cancer, head or neck cancer, esophageal cancer, parathyroid cancer, penile cancer, small intestine cancer, thyroid cancer, urethral cancer, cervical cancer, endometrial cancer, cancer fallopian tubes, renal pelvis cancer, vaginal cancer, vulvar cancer, chronic lymphocytic leukemia, chronic myeloid leukemia, colon cancer, skin cancer, malignant melanoma of the skin or intraocular duct, epidermoid cancer, head and neck squamous cell carcinoma (HNSCC), Hodgkin's lymphoma, Kaposi's sarcoma, kidney cancer, lung cancer, lymphocytic lymphoma, mesothelioma, nasopharyngeal cancer, central nervous system (CNS) neoplasms, non-Hodgkin's lymphoma, ovarian cancer, pancreatic cancer, pituitary adenoma, primary CNS lymphoma, prostate cancer, rectal cancer, kidney cancer, sarcoma, skin cancer, spinal axis tumor, squamous cell carcinoma, gastric cancer, T-cell lymphoma, testicular cancer and uterine cancer. 17. The combination of claim 14, wherein the cancer is selected from the group consisting of: acute lymphoblastic leukemia, acute myeloid leukemia, B-cell lymphoma, bladder cancer, blood cancer, bone cancer, brain tumors, brain stem glioma, cancer breast, adrenal cancer, anal cancer, endocrine cancer, head or neck cancer, esophageal cancer, parathyroid cancer, penile cancer, small intestine cancer, thyroid cancer, urethral cancer, cervical cancer, endometrial cancer, cancer fallopian tubes, renal pelvis cancer, vaginal cancer, vulvar cancer, chronic lymphocytic leukemia, chronic myeloid leukemia, colon cancer, skin cancer, malignant melanoma of the skin or intraocular duct, epidermoid cancer, head and neck squamous cell carcinoma (HNSCC), Hodgkin's lymphoma, Kaposi's sarcoma, kidney cancer, lung cancer, lymphocytic lymphoma, mesothelioma, nasopharyngeal cancer, central nervous system (CNS) neoplasms, non-Hodgkin's lymphoma, ovarian cancer, pancreatic cancer, pituitary adenoma, primary CNS lymphoma, prostate cancer, rectal cancer, kidney cancer, sarcoma, skin cancer, spinal axis tumor, squamous cell carcinoma, gastric cancer, T-cell lymphoma, testicular cancer and uterine cancer.