WO 2015/024060 PCT/AU2014/000830 TITLE IMMUNORECEPTOR MODULATION FOR TREATING CANCER AND VIRAL INFECTIONS TECHNICAL FIELD 5 THIS INVENTION relates to the immunoreceptor CD96. More particularly, this invention relates to inhibition of CD96 to thereby enhance the ability of the immune system to target tumours and other diseases or conditions that can evade the immune system. BACKGROUND 10 The progression of a productive immune response requires that a number of immunological checkpoints be passed. Passage may require the presence of excitatory co-stimulatory signals or the avoidance of negative or co-inhibitory signals, which act to dampen or terminate immune activity. The immunoglobulin superfamily occupies a central importance in this coordination of immune responses, 15 and the CD28/cytotoxic T-Iyiphocyte antigen-4 (CTLA-4)-B7.1/B7.2 receptorligand grouping represents the archetypal example of these immune regulators. In part the role of these checkpoints is to guard against the possibility of unwanted and harmful self-directed activities. While this is a necessary function, aiding in the prevention of autoimmunity, it may act as a barrier to successful 20 immunotherapies aimed at targeting malignant self-cells that largely display the same array of surface molecules as the cells from which they derive. Therapies aimed at overcoming these mechanisms of peripheral tolerance, in particular by blocking the inhibitory checkpoints on T cells, offer the potential to generate antitumor activity, either as monotherapies or in synergism with other therapies that directly or 25 indirectly enhance presentation of tumor epitopes to the immune system. Such anti-T cell checkpoint antibodies are showing promise in early clinical trials of advanced human cancers. Furthermore, natural killer (NK) cells are innate lymphocytes critical to limit early tumor growth and metastasis 1. NK cell functions are also regulated by the 30 integration of signals transmitted by a wide range of activating and inhibitory receptors 2. For example, the recognition of pathogen-derived or stress-induced ligands by activating receptors such as NCRs, NKG2D, or DNAM-I stimulate NK cells cytotoxicity and the secretion of pro-inflammatory mediators such as interferon gamma (IFN-;) 3.In contrast, inhibitory receptors protect target cells from NK cell 1 WO 2015/024060 PCT/AU2014/000830 mediated killing 4 These receptors mostly recognize MHC class I and MHC class I related molecules and include the KIR (killer cell immunoglobulin-like receptors) and LIR (leukocyte immunoglobulin-like receptors) families, the Ly49 family in mice and the CD94/NKG2 heterodirners in both species. 5 An emerging group of immunoglobulin superfamily members that interact with ligands of the nectin and nectin-like (nel) family has recently been described to influence NK cell and T cell functions These include C.D226 (DNAM-1) 6, CD96 (TACTILE) TIGIT (T cell imnunoglobulin and ITIM domain) and CRTAM (class I restricted T cell-associated molecule) 1f DNAM-1 and TIGIT are the most 10 extensively studied members of this family and they share a common ligand, CD 155 (necl-5; PVR) and CDl 12 (nectin-2; PVRL2) 3 TIGIT also bind an additional ligand CDl 13 (PVRL3) . The functions of DNAM-1 and TIGIT on NK cells are reportedly counter-balancing 12 I vitro, DNAM-1 potentiates the cytotoxicity of NK cells against a wide range of tumor cells b_;14 and is critical for tumor 15 immunosurveillance in v141o In contrast TIGIT bear an ITIM motif and has been proposed prevent self-tissue damage similar to inhibitory Ly49 or KIR interactions with MHC class I1. Indeed, engagement of TIGIT by CD 155 has been shown to limit IFNy production and cytotoxicity by NK cells in vitro tHowever; the role of TIGIT in NK cell biology relative to the other nectin receptors DNAM-1 20 and CD96 remains to be assessed in vivo. Despite being cloned 20 years ago . little is known about CD96, the other Ig family member that shares CD155 ligand with DNAM-1 and TIGIT2 In humans, 7 CD96 expression is largely confined to NK cells, CD8 T cells, and CD4 T cells The major ligand of CD96 is CD155, but CD96 has also been reported to associate 25 with CDI I I (nectin- 1) and play a role in promoting NK and T cell adhesion 2 SUMMARY Surprisingly, the present inventors have discovered that CD96 acts as a negative regulator of T cell and NK cell anti-tumor functions. Accordingly, the 30 invention is broadly directed to use of agents that at least partly block or inhibit CD96 to thereby reduce or relieve CD96-mediated immune inhibition to enhance or restore immune surveillance in the mammal. In certain embodiments, this may 2 WO 2015/024060 PCT/AU2014/000830 facilitate treatment of diseases or conditions responsive at least partial blocking or inhibition of CD96, such as cancers and/or viral infections. In a first aspect, the invention provides a method of reducing or relieving immune inhibition in a mammal, said method including the step of at least partly 5 inhibiting or reducing CD96 activity in one or more cells of the mammal to thereby relieve immune inhibition and/or enhance or restore immune surveillance in the mammal. Suitably, the step of inhibiting or reducing CD96 activity in the mammal does not include, or at least depend upon, killing of CD96-expressing cells in the 10 mammal. In some embodiments, the step of inhibiting or reducing CD96 activity in the mammal includes inhibiting or reducing CD96 binding to CDI55 and/or intracellular signaling in one or more cells of the mammal that express CD96. In some embodiments, the step of inhibiting or reducing CD96 activity in the mammal includes removing and/or down-regulating cell surface expression of C.D96 15 In one particular embodiment, the step of inhibiting or reducing CD96 activity in the mammal includes increasing or enhancing expression, production and/or secretion of one or more cytokines or chemokines. Preferably, the cytokine is interferon y (IFN-y). Typically, the one or more cells of the mammal are T cells, inclusive of CD4 and CD8 T cells, ydT cells, NKT cells, and natural killer (NK) 20 cells In a preferred embodiment, the method relieves immune inhibition and/or enhances or restores immune surveillance in the mammal to thereby treat or prevent cancer or cancer metastasis in the mammal. In other embodiments, the method relieves immune inhibition and/or 25 enhances or restores immune surveillance in the mammal to thereby treat or prevent a viral infection in the mammal. In a second aspect, the invention provides a method of screening, designing, engineering or otherwise producing a CD96-inhibitory agent, said method including the step of determining whether a candidate molecule is capable of at least partly 30 inhibiting or reducing CD96 activity to thereby relieve immune inhibition and/or enhance or restore immune surveillance in a mammal, 3 WO 2015/024060 PCT/AU2014/000830 In a third aspect, the invention provides a CD96-inhibitory agent screened, designed, engineered or otherwise produced according to the method of the second aspect. In one embodiment, the CD96-inhibitory agent is an antibody or antibody 5 fragment. In one particular embodiment, the CD96-inhibitory agent is an anti-cancer agent. In another particular embodiment, the CD96-inhibitory agent is an anti-viral agent 10 hi a fourth aspect, the invention provides a CD96-inhibitory agent according to the third aspect for use according to the method of the first aspect, Suitably, according to the aforementioned aspects the mammal is a human. Unless the context requires otherwise, the terms "compse", "comprises" and "comprising", or similar terms are intended to mean a non-exclusive inclusion, such 15 that a recited list of elements or features does not include those stated or listed elements solely, but may include other elements or features that are not listed or stated. The indefinite articles 't' and 'an' are used here to refer to or encompass singular or plural elements or features and should not be taken as meaning or 20 defining "one" or a "single" element or feature. BRIEF DESCRIPTION OF THE FIGURES Figure 1: CD96 competes with DNAM-1 for CD155 binding. a, b The expression of CD96 was analyzed by flow cytometry on the indicated spleen 25 lymphocyte populations from C57BL/6 WT (light grey) and CD96-mice (dark grey). The representative FACS Histograms (a) and the mean + SD (b) of 3 mice from one representative experiment out of 3 are shown. c, d The expression of CD96 DNAM I and TIG IT was determined on WT spleen NK cells freshly isolated or activated for 48 hrs with IL-2 (1000 U/mIl). e. The binding of mouse CD155-Fc coupled with AF 30 647 to purified NK cells freshly isolated from WT, CD96o DAM- ~r DAbMJM CD96 mice was assessed at the indicated concentrations by flow cytometry. fL The binding of CD155-Fe coupled with AF-647 (10 pgiml) was analyzed on purified WT NK cells in the presence of anti-C.D96 and or anti-DNAM- I mAbs. g. The binding of 4 WO 2015/024060 PCT/AU2014/000830 DNAM-l-Fc labeled with AF-647 (0.5-10 pg/mil) at the cell surface of BMDC was analyzed in the presence of 50 pig/ml of control Ig recombinant CD96 or TIGIT-FE, c-g. The representative FACS Histograms and the mean + SD of triplicate wells from one representative experiment out of at least 3 experiments are shown *** p<0.001 5 Student T test, Figure 2: CD96 engagement by CD155 regulate NK cell production of IFNy. CD96 binding to CD155-Fc limits the production of IFN-y by NK cells induced by exogenous cytokines (a, b, d) and NK cell receptors (c)- a, b, d. We analyzed the intracellular production of IFN-y by freshly purified CD96- 1IH and WT NK 10 cells in the presence or absence of anti-CD96 (50 pg/ml) in response to IL-12 (25 100 pg/n!) and IL-IS (50 ng/nl) using plates coated with or without CD 155-Fe (0.5 pg/ well). c. We analyzed the intracellular production of IFN-y by 1I-2-activated NK cells from C96- and WY' mice using plates coated with anti-NK.1 (2.5 pg/ well) and CDl155-Fc (0.5 jig/ well). The representative FACS Histograms (a) and the mean 15 + SD of triplicate wells (b. c, d) from one representative experiment out of 3 are shown. *p <0.05, ** p <0.01, *** p<0.001, Student T test. Figure 3: CD96 limits NK cell-dependent tumor immunosurveillance. a, b. CD96 and DNAM- I have an opposite role in the control of B 1 6F 10 metastasis. a. 2 x 105 B16FI cells were intravenously injected into WT, CD96. DNM4 -U and 20 DNAM-ICD96C mice and metastatic burden was quantified in the lungs after 14 days. Representative experiment out of 3, b. Pictures showing the lung of WT and CD96 mice two weeks after the injection of 2 x 10 and 5 x 10 B161710 cells. Representative experiment out of two. c. CD96 and TIGIT compete with DNAM-1 for the binding of CD155 at the cell surface of Bl6F10. The binding of DNAM-i-Fc 25 labeled with AF-647 (0.5-20 ig/ml) at the cell surface of B16F10 cells was analyzed in the presence of 50 pg/nl of control Ig, recombinant CD96 or TIGIT-Fc. The FACS histograms and the mean 1 SD of triplicate wells from one representative experiment out of 3 are shown. d. A 4 hr 51 Cr release assay was performed between B6F 10 cells and IL-2-activated N-K cells from WT, DNAMA.-U and (D96C mice at 30 the indicated effector target ratios. Solid circles represent WT NK cells, open circles represent CD96- NK cells and solid squares represents DNAM-V- NK cells. e-h. CD96 and DNAM-I have an opposite role in the immunosurveillance of MCA induced fibrosarcoma mediated by NK cells. e-h Groups of 15-30 male, WT, 5 WO 2015/024060 PCT/AU2014/000830 DNAM-I and (D96 and D M-I C96 mice were injected with MICA (100 pig/mouse). The survival (e-g) and the growth curves of individual mice with sarcoma (h) are shown. f. WT mice were treated with an anti-CD96, anti-DNAM- 1 or anti-CD155 niAbs as defined in the Materials and Methods. g. WT and ('96- 5 mi ce were injected with 100 pg MCA and treated with either a control antibody, anti IFN-yantibody, or anti-asialoGM *I. p<0;05 Mantel-Cox test. Figure 4: Anti-CD96 mAb has single agent activity and enhances the anti-tumor responses of anti-PD 1 C57BL6 wild type (WT) mice were injected subcutaneously with AT3-OVA' tumor cells (1W' cells) and treated on day 16, 20 and 24 with 10 intraperitoneal injections of anti-CD96 mAb (3-3, 250 pg ip) or anti-PD-1 (RMPI 14, 250 pg i.pj. Means ± SEM. of 5 mice per group (mm 2 ) are shown (*- p<0.05 compared to clg alone by Mann-Whitney test). Figure 5: Anti-CD96 mAb enhances anti-tumor responses generated by Doxorubicin (DOX) chemotherapy. C57BL/6 wild type (WT), DNAM-1> and 15 CD96< mice were injected subcutaneously with AT3-OVA tumor cells (106 cells) and treated on day 14 with control PBS or DOX (50 microliters, 2 mM, intratumor). Some groups of WT mice also received on day 12, 14, 18, 21, 24 and 28 intraperitoneal injections of anti-CD96 mAb (3.3, 250 pg i.p) or anti-DNAMI (480.], 250 pg i.p.). Means + SEM of 5 mice per group (mm 2 ) are shown. 20 Figure 6: Enhanced anti-tumor responses of Doxorubicin (DOX) chemotherapy with host CD96 deficiency. C57BL/6 wild type (WT), DNAM- 1. and CD96- mice were injected subcutaneously with AT3-OVA.".. tumor cells (106 cells) and treated on day 1,6 with control PBS or DOX (50 microliters, 2 mM, intratumor). Means i standard errors of 5 mice per group (mm 2 ) are shown. 25 Figure 7: Anti-CD96 niAb enhances anti-tumor responses generated by Doxorubicin (DOX) chemotherapy. C57BL/6 wild type (WT) mice were injected subcutaneously with AT3-OVAd", tumor cells (106 cells) and treated on day 16 with control PBS or DOX (50 microliters, 2 mM, intratumor). Some groups of WT mice also received on day 16, 20, and 23 intraperitoneal injections of anti-CD96 mAb 30 (3.3, 250 pg i p). Means ± SEM of 5 mice per group (minm) are shown (*- p<0.05 compared to cIg alone by Mann-Whitney test). Figure 8: Early anti-CD96 mAb enhances anti-tumor responses generated by anti-PD-I and anti-CTLA-4 mAbs. C57BL/6 wild-type (WT) mice were injected 6 WO 2015/024060 PCT/AU2014/000830 subcutaneously with B16-OVA melanoma cells (105 cells) and treated on day 1, 5, and 9 with intraperitoneal injections of anti-CD96 mAb (3.3, 250 pg i.p), anti-PD-I mAb (RMPI-14, 250 ptg i.p) anti-CTLA-4 (UCIO-4F10, 250 pg i p.), anti CD96!anti-PD-1 mAbs (250 jig i.p each), anti-CD96/anti-CTLA-4 mAbs (250 pg i.p 5 each) or control Ig (cig) (2A3, 250 jig i.p) Means ± SEN of 5 mice per group (nm) are shown (*: p<0.05 compared with anti-CD96 alone, by Mann-Vhitney test), Figure 9: Late anti-CD96 mAb enhances anti-tumor responses generated by anti-PD-1 mAb. C57BL/6 wild-type (WT) mice were injected subeutaneously with B16-OVA melanoma cells (105 cells) and treated on day 16, 20, and 24 with 10 intraperitoneal injections of anti-CD96 mAb (33, 250 pg ip), anti-PD-1 mAb (RMP1-14, 250 pg i.p.), anti-CTLA-4 (UC1O-4F10, 250 pg i.p.), anti-CD96/anti PD-I mAbs (250 pg i-p each), anti-CD96/anti-CTLA-4 mAbs (250 pg i-p each) or control Ig (cig) (2A3, 250 pg i p). Means + SEM of 5 mice per group (nul) are shown (*: p<0.05 compared with anti-CD96 alone by Manr-Whitney test). 15 Figure 10: Host CD96 promotes B16FI0 lung metastasis- C57BL/6 wild type (WT), DNAM-l<, CD96 -, and DNAM- 1 CD96- mice were injected intravenously with B16FI10 melanoma cells (1-0 cells) and metastatic burden was quantified in the lungs after 14 days by counting colonies on the lung surface. Means SEM' of 9-17 mice per group are shown (* p<0-05 compared with WT by Mann-Whitney test). 20 Figure 11: Host CD96 promotes RM-1 lung metastasis. C57BL/6 wild type (WT), DNAM- CD96 and DNAM-1 -'-CD96 mice were injected intravenous with RM I prostate carcinoma cells (104 Cells) and metastatic burden was quantified in the lungs after 14 days by counting colonies on the lung surface. Means ± SEM of 10-15 mice per group are shown (*: p<0.05 compared with WT by Mann-Whitney 25 test), Figure 12: Host CD96 promotes 3LL lung metastasis. C57BL/6 wild type (WT), DNAM-U CD96", and DNAM-IVCD96 mice were injected intravenously with 3LL lung carcinoma cells (10 5 cells) and metastatic burden was quantified in the lungs after 14 days by counting colonies on the lung surface. Means i SEM of 5 30 mice per group are shown (*- p<0.05 compared with WT by Mann-Whitney test). Figure 13: Anti-CD96 suppresses B16F10 lung metastasis, alone and in combination with T cell checkpoint blockade. C57BL/6 wild type (WT) mice were injected intravenous with B16F10 melanoma cells (105 cells). On day 0 and 3 after 7 WO 2015/024060 PCT/AU2014/000830 tumor inoculation, mice were treated with intraperitoneal injections of anti-CD96 nAb (3.3, 250 pg ip), anti-PD-i mAb (RMPI14, 250 pg i-p.), anti-CTLA-4 (UC10 4F10, 250 pg i.p.)t anti-CD96/anti-PD-1 mAbs (250 pg i.p each). anrti-CD96!anti CTLA-4 mAbs (250 tg ip each) or control Ig (cig) (2A3, 250 pg ip). Metastatic 5 burden was quantified in the lungs after 14 days by counting colonies on the lmg surface. Means ± SEM of 5 mice per group are shown (*: p<0.05 compared with anti-CD96 alone by Mann-Whitney test). Figure 14: Anti-CD96 suppresses RM-1 lung metastasis alone and in combination with T cell checkpoint blockade. C57BL/6 wild type (WT) mice were 10 injected intravenous with RM-l prostate carcinoma cells (104 cells). On day 0 and 3 after tumor inoculation, mice were treated with intraperitoneal injections of anti CD96 mAb (3.3. 250 pg i.py anti-PD-i mAb (RMPI-14, 250 pg i-p.), anti-CTLA-4 (UC10-4F10, 250 pg ip.), anti-CD96/anti-PD-1 mAbs (250 pg i.p each), anti CD96/anti-CTLA-4 mAbs (250 pg i.p each) or control Ig (cig) (2A3, 250 pg ip), 15 Metastatic burden was quantified in the lungs after 14 days by counting colonies on the hng surface. Means + SEM of 5 mice per group are shown (*: p<0.05 compared with anti-CD96 alone by Mann-Whitney test). Figure 15: Late anti-CD96 mAb enhances anti-tumor responses generated by anti-P.D-1 mAb. C57B.L/6 wild-type (WT) mice were injected subcutaneously with 20 MC38-OVAa" colon adenocarcinoma cells (106 cells) and treated on day 14, 18. 22, and 26 with intraperitoneal injections of anti-CD96 mAb (3.3, 250 pg i,p), anti-PD- I mAb (RMPI-14. 250 pg i.pj, anti-CTLA-4 (UC10-4F10, 250 pg i.p.j, anti CD96/anti-PD-l mAbs (250 pg i.p each), anti-CD96/anti-CTLA-4 mAbs (250 pg i.p each) or control Ig (clg) (2A3, 250 pg i.p). Means ± SEM of 5 mice per group (mm 2 25 are shown (*: p<0.05 compared with anti-CD96 alone by Mann-Whitney test). Fig. 16. Mechanism of the anti-tumor effect of anti-CD96 alone against AT3-OVAt" mammary carcinoma. (A-B) C57BL/6 wild-type (WT) and pfp mice were injected s.c. with AT3-OVA dil mammary carcinoma (1 x 106 cells). On (A) day 16, 20, and 24 or (B) day 12, 16 and 18 after tumor inoculation mice were treated with i.p. 30 injections of cIg (250 pg i.p.) or anti-CD96 mAb (250 pg i.p.). Some groups of mice were treated on (A) days 14, 16, and 23 after tumor inoculation with clg, anti CD4/anti-CD8P (100 pg i p.) or anti-asGM l (100 pg i.p.) or (B) days 10, 12 and 18 after tumor inoculation with clg or anti-IFN-y (250 pg i.p.). Mice were then 8 WO 2015/024060 PCT/AU2014/000830 monitored for tumor growth. Means i SEM of 5 mice per group (mm 2 ) are shown (*; p<0.05 of clg-treated mice compared with anti-C.D96 mAb-treated mice by Mann Whitney test). Fig 17. in Vitro NK cell activation. To analyse the production of IFNy from human 5 NK cells, 96 well U bottom plates were coated with recombinant human CD.1 55-Fc chimera overnight at 4"C (0.25pg/well). After three washes, 2.5 x 104 human NK cells, freshly isolated from buffy coats and FACS sorted were plated in complete RPMI supplemented with human IL-12 (10 ng/ml), I-15 (100 ng/ml) and IL-i 8 (100 ng/ml) in the presence or absence of human anti-C.D96 antibody (clone 10 NK92.39, 50 pg/ml). Cultures were also set up in wells not containing CD155-Fc coating, After 24 hours incubation, cells were harvested and analysed for IFNy production by Flow Cytomety. All conditions were run in triplicate and error bars represent the ±SEM. (A) Flow cytometry results using IL-12, 18 and 15; and (B) result from a different donor (not donor I shown in 17A) showing the proportion of 15 IFNy positive N K cells. Fig 18. Binding of human CD96 mAb (NK92.39) to human NK cells reduced the levels of CD96 present on the NK cell surface. Total NK cells were purified from peripheral blood mononuclear cells (PBMCs) by negative selection using human NK cell isolation kit (Miltenyi Biotec). Isolated NK cells were then labeled with 20 carboxyfluorescein diacetate succinmidyl ester (CFSE; Biolegend) to measurecellular proliferation. CFSE-labeled NK cells were plated in 96 well U bottom plate at 5 x 104 cells/well and stimulated with recombinant IL-2 at indicated concentrations (10 units/mil and 25 uniitS/ ), in the presence of control IgG or anti CD96 nAb (clone NK92-39) at 30 pg/mL NK cells were assessed for changes in 25 proliferation (A) or the presence/absence of surface CD96; and (B) at day 3 and 6 using BD FACS Canto i (BD Biosciences) and analysis was carried out using FlowJo (Tree Star). DETAILED DESCRIPTION 30 The present invention is at least partly predicated on the unexpected discovery that CD96 is highly expressed by resting NK cells and T cell subsets and competes with DNAM-l for the binding of CD155 on resting NK cells. Using CD96~ mice, it is demonstrated that CD96 dampens or suppresses NK cell production of 9 WO 2015/024060 PCT/AU2014/000830 IFN-v in vitro and in vivo, through competition with DNAM-1I for CD1I55 binding and also through a direct inhibition. Furthermore, CD96'- mice were shown to be more resistant to 3 %methylcholanthrene (MCA)-induced tumor formation as an indicator of carcinogenesis, or B16F 10 (melanoma), RN-1I (prostate cancer), 3LL 5 (lung cancer) experimental metastasis. In human NK cells, administration of anti CD96 antibody appears to remove or result in the loss of cell surface CD96 and/or cause a down-regulation of CD96 cell surface expression. Based on these observations, it is proposed that CD96 normally acts as a negative regulator of T and NK cell anti-tumor functions, particularly although not exclusively through 10 suppression of IFN-y production and/or secretion. Accordingly, the invention provides methods of relieving or reducing the negative imunoregulatory function of CD96 to thereby promote or restore immune surveillance, particularly by T cells and NK cells, to thereby treat or prevent cancer, cancer cell metastasis and/or viral infections. 15 An aspect of the invention therefore provides a method of reducing or relieving immune inhibition in a mammal, said method including the step of at least partly inhibiting or reducing CD96 activity in one or more cells of the mammal to thereby relieve inunune inhibition and/or enhance or restore immune surveillance in the mammal. 20 By "relieving hmnine inhibition" in the context of CD96 is meant at least partly eliminating, removing or overcoming a normal activity or function of CD96 in suppressing or inhibiting one or more immune functions of cells that normally express CD96. Typically, the one or more cells that normally express CD96 are T cells, inclusive of CD4 and CD8 T cells, y6T cells, NKT cells, and natural killer 25 (NK) cells. In sonIC embodiments, relieving immune inhibition may include or relate to abrogating peripheral tolerance to foreign pathogens, host cells displaying foreign pathogens (e~g displaying foreign pathogen-derived peptides in the context of self MiC) and/or cancerous cells or tissues of the host. By "enhance or restore immune mzvelmce" is meant at least partly 30 improving or promoting the ability of one or more elements of the immune system to monitor, detect and/or respond to foreign pathogens, host cells displaying foreign pathogens (e.g displaying foreign pathogen-derived peptides in the context of self MHC) and/or cancerous cells or tissues of the host. Suitably, the elements of the 10 WO 2015/024060 PCT/AU2014/000830 immune system are one or more cells that normally express CD96, such as T cells, inclusive of CD4 and CD8 T cells 76T cells, NKT cells and natural killer (NK) cells. At least partly inhibiting or reducing CD96 activity in one or more cells of 5 the mammal may be performed, facilitated or achieved by administration of a "()96-inhihoiy agent " to the mammal. A CD96-inhihitory agent may be any molecule that possesses or displays an ability to at least partly inhibit or reduce a biological activity of CD96. Biological activities of CD96 include one or more of binding CD155, cell surface expression, eliciting intracellular signaling and/or 10 stimulating or inducing expression and/or secretion of cytokines or chemokines. Preferably, the cytokines or chemokines include any pro-inflammatory cytokine or chemokine inclusive of MIP-la, MIP-Ii, RANTES, TNF-a and IFN-y, although without limitation thereto. Preferably, the cytokine is IFN-y. As disclosed herein, CD96 is a transmembrane protein which in the human, 15 exists in two isoforms. Isoform I has been detected in acute myeloid leukaemia and includes additional amino acids compared with isoform 2. Isoform 2 is the more common form in humans and the predicted domain architecture of isoform 2 has three (3) external immnunoglobulin-like domains (domains 1, 2 and 3) as listed in Table L The amino acid sequence of human CD96 isoform 1 is set forth in SEQ ID 20 NO:2. The nucleotide sequence encoding isoform 2 is set forth in SEQ ID NO: I Marine CD96 exists as a single isoform, the amino acid sequence of which is set forth in SEQ ID NO:4. The nucleotide sequence encoding murine CD96 is set forth in SEQ ID NOJ The external immunoglobulin-like domains (domains 1, 2 and 3) of urine CD96 are also listed in Table 1. 25 In a preferred form, the CD96-inhibitory agent binds or interacts with an amino acid sequence of one or a plurality of external immunoglobulin-like domains of CD96. For example, the CD96-inhibitory agent may bind or interact with an amino acid sequence of domain 1; domain 2; domain 3; domain I and domain 2; domain I and domain 3: domain 2 and domain 3; or domain 1, domain 2 and domain 30 3. In one embodiment, the CD96-inhibitory agent binds or interacts with one or a plurality of external imnmunoglobulin-like domains of human CD96 isoform 2. 11 WO 2015/024060 PCT/AU2014/000830 It will also be appreciated that the CD96-inhibitory agent may bind or interact with other CD96 domains or amino acid sequences in addition to one or a plurality of the external or extracellular immunoglobulin-like domains. In one embodiment, the CD96-inhibitory agent inhibits, blocks or 5 antagonizes a binding interaction between CD96 and CD155. By way of example only, the CD96-inhibitory agent may bind to an extracellular domain of CD96, or a portion thereof, that is capable of interacting with CD155 (e.g binding CD155 or being bound by CD155) to thereby at least partly inhibit or block CD96 binding to CD155. 10 In another embodiment, the CD96-inhibitory agent is a molecule that possesses or displays an ability to inhibit or reduce CD96 signaling activity. Inhibition or reduction of CD96 signaling activity may be through inhibiting, blocking or antagonizing a binding interaction with CD155 or may be through blocking CD96-initiated signaling that would normally occur in response to CD155 15 binding. By way of example, CD96 comprises an immunoreceptor tyrosine-based inhibitory motif (ITIM). ITiMs are structurally defined as 6-amino acid sequences comprising a tyrosine (Y) residue with partly conserved N-terminal (Y-2) and C terminal (Y+) residues. A general but non-limiting motif is (S/I/V/LXYXXI/V/L), wherein X is any amino acid. For example, isoforn 1 of CD96 comprises the ITIM 20 sequence IKYTCI wherein Y is residue 566 It has been proposed that when co-aggregated with activating receptors, ITIMs are phosphorylated by Src-family tyrosine kinases, which enables them to recruit Src homology 2 domain-containing phosphatases (PT.Pases) that antagonize activation signals. Accordingly, in one embodiment the CD96-inhibitory agent 25 possesses or displays an ability to inhibit or reduce CD96 signaling activity mediated by the CD96 ITIM. Preferably, inhibition or reduction of CD96 signaling activity mediated by the CD96 ITIM enables increased or enhanced cytokine (e~g [FN-y) expression, production and/or secretion by a cell expressing CD96. In another embodiment, the CD96-inhibitory agent is a molecule that 30 removes cell surface CD96 and/or reduces or down-regulates cell surface expression of CD96. The CD96-inhibitory agent may be a protein (inclusive of peptides, antibodies and antibody fragments), a nucleic acid (inclusive of inhibitory RNA 12 WO 2015/024060 PCT/AU2014/000830 molecules such as ribozymes, RNAi, miRNA and siRNA, although without limitation thereto), a lipid, a carbohydrate, a small organic molecule or any combination of these (e.g a glycoprotein, a lipoprotein, a peptide-nucleic acid etc). In one particular embodiment, the CD96-inhibitory agent is an antibody or 5 antibody fragment that binds CD96. In a preferred form, the antibody or antibody fragment binds or interacts with an amino acid sequence of one or a plurality of external or extracellular immunoglobulin-like domains of CD96. For example, the antibody or antibody fragment may bind or interact with an amino acid sequence of: domain 1; domain 2, 10 domain 3; domain I and domain 2; domain I and domain 3: domain 2 and domain 3; or domain 17 domain 2 and domain 3. In one embodiment, the antibody binds or interacts with one or a plurality of external immunoglobulin-like domains of human CD96 isoform 2. In one form the antibody binds CD96 and at least partly blocks or inhibits 15 CD96 binding to CD1 55. Antibodies may be polyclonal or monoclonal, native or recombinant. Antibody fragments include Fab and Fab'2 fragments, diabodies and single chain antibody fragments (e.g. scVs), although without limitation thereto. Antibodies and antibody fragments may be modified so as to be administrable to one species having 20 being produced in, or originating from, another species without eliciting a deleterious immune response to the "foreign" antibody In the context of humans, this is "humanization" of the antibody produced in, or originating from, another species. Such methods are well known in the art and generally involve recombinant "grafting" of non-human antibody complementarity determining regions (CDRs) 25 onto a human antibody scaffold or backbone. Suitably, the step of inhibiting or reducing CD96 activity in the mammal does not include killing CD96-expressing ceus in the mammal. In this context, "killing" may refer to any antibody-mediated cytotoxic mechanism such as complement mediated cytolysis and antibody-mediated cell-mediated cytotoxicity (ADCC), the 30 latter typically mediated by natural killer (NK) cells, macrophages, neutrophils and eosinophils. In this regard, it may be advantageous to use antibody fragments lacking an Fc portion or having a mutated Fc portion. The step of inhibiting or reducing CD96 activity in the mammal may be achieved or facilitated by administering a CD96-inhibitory agent to the mammal. 13 WO 2015/024060 PCT/AU2014/000830 By "administering" is meant the introduction of the CD96-inhibitory agent into the mammal by a particular route. Suitably, a therapeutically effective amount of the CD96-inhibitory agent is administered to the mammal. The tern "thermpeutietdly effective amount" describes a quantity of a 5 specified agent sufficient to achieve a desired effect in a mammal being treated with that agent. Generally, the method of the invention may be useful in reducing or relieving CD96-mediated immune inhibition, suppression or peripheral tolerance. Suitably, the method facilitates treatment or prevention of one or more diseases or conditions that 10 are responsive to at least partly blocking CD96-mediated immune inhibition., suppression or peripheral tolerance. As used herein, treating " or "treat" or "treatment" refers to a therapeutic intervention that at least party eliminates or ameliorates one or more existing or previously identified symptoms of a disease or condition that is responsive to at least 15 partly blocking CD96-mediated immune inhibition, suppression or peripheral tolerance. As used herein, "prevewing" or "prevent" or "prevenion" refers to a prophylactic treatment initiated prior to the onset of a symptom of a disease or condition that is responsive to at least partly blocking CD96-mediated immune 20 inhibition, suppression or peripheral tolerance, so as to at least partly or temporarily prevent the occurrence of the symptom. Typically, the disease or condition that is responsive to at least partly blocking CD96-inediated immune inhibition, suppression or peripheral tolerance is any disease or condition where enhanced or restored immune surveillance can 25 benefit a subject suffering from the disease or condition. Such diseases and conditions may include those where persistence of the disease or condition can be controlled or suppressed by cell-mediated immunity. Non-limiting examples include cancers and viral infections. Particular viral infections contemplated by the invention include persistent viral infections such as caused by human immunodeficiency virus 30 (HIV), Epstein Barr Virus (EBV), Herpes Simplex Virus (HSV inclusive of HSVI and FSV2), Human Papillomavirus (HPV), Varicella zoster virus (VSV) and Cytomegalovirus (CMV), although without lImitation thereto. In a preferred embodiment, the method reduces or relieves immune inhibition in a mammal sufficient to treat or prevent cancer or cancer metastasis in the 14 WO 2015/024060 PCT/AU2014/000830 mammal. Suitably, the cancer may be any that is responsive to at least partly blocking CD96-iediated immune inhibition, suppression or peripheral tolerance. Cancers may be in the form of solid tumors, sarcomas, lymphomas, myelomas, carcinomas, melanomas, cytomas and meningiomas, although without limitation 5 thereto, Non-limiting examples of cancers include cancers of the adrenal gland, bladder, bone, bone marrow, brain, breast, cervix, gall bladder, ganglia, gastrointestinal tract, heart, kidney, liver, hmg, muscle, ovary, pancreas, pituitary parathyroid, prostate, salivary glands, skin, spleen, testis, thyroid, and uterus. Particular non-limiting examples of cancers include colon cancer, lung cancer and 10 prostate cancer. In some embodiments, the cancer is a metastatic cancer, which is capable of migrating to another site, tissue or organ in the body and forming a tumor at that site, tissue or organ. This may occur repeatedly over time. A particularly aggressive metastatic cancer contemplated by the invention is metastatic melanoma. It will also be appreciated that the method of treatment or prevention of 15 cancer may further include co-administration of one or more other therapeutic agents that facilitate cancer treatment or prevention. By way of example only, these include: chemotherapeutic agents such as paclitaxel, doxorubicin, methotrexate and cisplatin, although without limitation thereto; and/or biotherapeutic agents such as anti-PD-1 antibodies (e.g. Nivolumab) and anti-CTLA4 antibodies (e.g Ipilimumab), although 20 without limitation thereto. Also contemplated are bi-specific antibodies that bind both CD96 and one or more other molecules including but not limited to PD-I and CTLA4. The one or more other agents that facilitate cancer treatment or prevention may be administered in combination with the CD96-inhibitory agent or be 25 administered separately, as is well understood in the art. In some embodiments, the CD96-inhibitory agent may be formulated alone or together with the one or more other agents is in the form of a pharmaceutical composition. Suitable dosages of CD96-inhibitory agents, alone or together with other 30 therapeutic agents, for mammalian administration, including human administration, may be readily determined by persons skilled in the art. Suitably, the pharmaceutical composition comprises an appropriate pharmaceutically-acceptable carrier, diluent or excipient. 15 WO 2015/024060 PCT/AU2014/000830 Preferably, the pharmaceutically-acceptabIle carrier, diluent or excipient is suitable for administration to mammals, and more preferably, to humans. By "pharmaceuticaly-acceptable carrier, dluent or excipient"' is meant a solid or liquid filler, diluent or encapsulating substance that may be safely used in 5 systemic administration. Depending upon the particular route of administration, a variety of carriers, diluents and excipients well known in the art may be used These carriers may be selected from a group including sugars, starches, cellulose and its derivatives, malt, gelatine, tale, calcium sulfate, vegetable oils, synthetic oils, polyols, alginic acid, phosphate buffered solutions, eniulsifiers, isotonic saline and 10 salts such as mineral acid salts including hydrochlorides, bromides and sulfates., organic acids such as acetates, propionates and malonates, and pyrogen-free water, A useful reference describing pharmaceutically acceptable carriers, diluents and excipients is Remington's Pharmaceutical Sciences (Mack Publishing Co. NJ USA, 1991). 15 Any safe route of administration may be employed for providing a subject with compositions comprising the CD96-inhibitory agent. For example, oral, rectal. parenteral, sublingual, buccal, intravenous, intra-articular, intra-muscular, intra dermal, subcutaneous, inhalationat intraocular, intraperitoneal, intracerebroventricular, transdermal, and the like may be employed. 20 A further aspect of the invention provides a method of screening, designing, engineering or otherwise producing a CD96-inhibitory agent, said method including the step of determining whether a candidate molecule is capable of at least partly inhibiting or reducing CD96 activity to thereby relieve immune inhibition and/or enhance or restore immune surveillance in a mammal, 25 The invention also provides a CD96-inhibitory agent screened, designed, engineered or otherwise produced according to the aforementioned aspect. The candidate molecule may be a protein (inclusive of peptides, antibodies and antibody fragments), a nucleic acid (inclusive of inhibitory RNA molecules such as ribozymes, RNA.i, miRNA and siRNA, although. without limitation thereto), a 30 lipid, a carbohydrate, a small organic molecule or any combination of these (e.g a glycoprotein, a lipoprotein. a peptide-nucleic acid etc). In some embodiments, the candidate modulator may be rationally designed or engineered de novo based on desired or predicted structural characteristics or features that indicate the candidate modulator could block or inhibit one or more 16 WO 2015/024060 PCT/AU2014/000830 biological activities of CD96, such as CD155 binding, intracellular signaling and/or IFN-y production and/or secretion. In other embodiments, the candidate modulator may be identified by screening a library of molecules without initial selection based on desired or predicted structural characteristics or features that indicate the 5 candidate modulator could block or inhibit one or more biological activities of CD96. Such libraries may comprise randomly generated or directed libraries of proteins, peptides, nucleic acids, recombinant antibodies or antibody fragments (e.g. phage display libraries), carbohydrates and/or lipids, libraries of naturally-occurring molecules and/or combinatorial libraries of synthetic organic molecules. 10 Non-limiting examples of techniques applicable to the design and/or screening of candidate modulators may employ X-ray crystallography, NMR spectroscopy, computer assisted screening of structural databases, computer-assisted modelling or biochemical or biophysical techniques which detect molecular binding interactions, as are well known in the art, 15 Biophysical and biochemical techniques which identify molecular interactions include competitive radioligand binding assays, co-immunoprecipitation, fluorescence-based assays including fluorescence resonance energy transfer (FRET) binding assays, electrophysiology, analytical ultracentrifugation, label transfer, chemical cross-linking, mass spectroscopy, microcalorimetry, surface plasmon 20 resonance and optical biosensor-based methods, such as provided in Chapter 20 of CURRENT PROTOCOLS IN PROTEIN SCIENCE Eds. Coligan ei at, (John Wiley & Sons, 1997) Biochemical techniques such as two-hybrid and phage display screening methods are provided in Chapter 1.9 of CURRENT PROTOCOLS TN PROTEIN SCIENCE Eds. Coligan e al, (John Wiley & Sons, 1997). 25 Accordingly, an initial step of the method may include identifying a plurality of candidate molecules that are selected according to broad structural and/or functional attributes, such as an ability to bind CD96. The method may include a further step that measures or detects a change in one or more biological activities of CD96 in response to the candidate molecule(s). 30 These may include CD155 binding, the presence or absence of cell surface CD96. intracellular signaling, cytokine and/or chemokine production or secretion and/or protection from tumor challenge in an in vivo model. 17 WO 2015/024060 PCT/AU2014/000830 Inhibition of CDI55 binding to CD96 by a candidate molecule may be determined by any of several techniques known in the art including competitive radioligand binding assays, surface plasmon resonance (e.g. BlACore Th ' analysis), co-immunoprecipitation and fluorescence-based analysis of the ability of a candidate 5 inhibitor to block CD155 binding to CD96 (such as by flow cytometry where CD155 is labeled with a fluorophore), Non-limiting examples of fluorophores include fluorescein isothiocyanate (FITC), allophycocyanin (APC), fluoroscein derivatives such as FAM and ROX, Texas Red, tetramethylrhodamine isothiocyanate (TRITL), R-Phycoerythrin (RPE), Alexa and Bodipy fluorophores, although without limitation 10 thereto. Alternatively, this fluorescence-based analysis could include FRET analysis (eg. one protein coupled to a donor fluorophore, the other coupled to an acceptor fluorophore), although without limitation thereto. In some embodiments, intracellular signaling may be measured directly at the 15 level of C:D96, such as by measuring recruitment of SH2 domain-containing PTPases by CD96 expressed by NK cells, or T cell subsets. A candidate molecule of the invention suitably prevents or reduces recruitment of SHF2 domain-containing PTPases by CD96 in the presence of CD155. According to this embodiment, the candidate molecule may at least partly inhibit or prevent binding between CD96 and 20 CD155, thereby at least partly inhibiting or preventing intracellular signaling by CD96, and/or at least partly inhibit or prevent intracellular signaling by CD96 despite CD 155 binding. In other embodiments, an effect of a candidate molecule on CD96 may be determined by measuring expression, production and/or secretion of one or more 25 cytokines or chemokines by cells expressing CD96. Generally, changes in cytokine or chemokine expression production and/or secretion may be measured by RT-PCR of cytokine mRNA, measurement of cytokine or chemokine protein located intracellularly (e.g by immunocytochemistry using cytokine- or chemokine-specific antibodies) and/or measurement of secreted cytokines or chemokines such as by flow 30 cytometric Cytokine Bead Array (such as commercially available from B) Biosciences), by ELISA using cytokine- or chemokine-specific antibodies and by bioassays that use cytokine- or chemokine-responsive cell lines to measure cytokines and/or chemokines secreted into cell supermatants. The cytokine may be any pro 18 WO 2015/024060 PCT/AU2014/000830 inflammatory cytokine or chemokine inclusive of MIP-Ia, MIP-IP, RANTES, TNF a and IFN-y. although without limitation thereto. Preferably, the cytokine is .FN-y. The effect of a candidate CD96 inhibitory agent may be upon CD96 expression, inclusive of cell surface expression and CD96 gene expression. It will be 5 appreciated that in certain cells, such as NK cells although without limitation thereto, CD96 inhibitory agents may cause or facilitate a removal, loss andlor down regulation of CD96 expression. In some embodiments, thus may include removal, loss and/or down-regulation of CD96 expression at the cell-surface. This may occur as a result of enhanced internalization or endocytosis of cell surface CD96 and/or by 10 a down-regulation or suppression of CD96 gene expression. In particular embodiments, CD96 cell surface expression may be detected or measured by flow cytometry, immunoprecipitation, in1nmunocvtochemistry or ininiunohistochemistry, typically by way of an antibody or antibody fragment which binds CD96, as hereinbefore described. In particular embodiments, CD96 gene expression may be 15 measured by nucleic acid sequence application (e.g. PCR inclusive of quantitative and semi-quantitaive PCR) or nucleic acid hybridization techniques such as Northern blotting. Preferably, the CD96-inhibitory effect of a candidate molecule may be determined using an in vivo tumor challenge model. For example, a mouse model 20 using CD96-expressing mice may be used to determine the ability of a candidate molecule to inhibit or prevent tumor formation and/or growth in response to an administered carcinogenic agent such as nethycholanthrene (MCA). In another example, a mouse model using CD96-expressing mice may be used to determine the ability of a candidate molecule to inhibit or prevent tumor formation and/or growth 25 in response to administration of tumor cells such as melanomas, colon adenocarcinomas, prostate carcinomas and mammary carcinomas, although without limitation thereto. Other mouse models may utilize mice that are predisposed to spontaneously forming tumors including but not limited to MMTV-polyoma, MT mammary cancer, DMBAITPA induced skin cancer, p53 loss lymphoma/sarcoma 30 and TRAMP Tg prostate cancer. It will be understood that the method of this aspect may be performed iteratively, whereby multiple rounds of screening, design, and biological testing are performed. This may include where a candidate molecule is structurally modified before each round, thereby enabling "fine-tuning" of the candidate molecule 19 WO 2015/024060 PCT/AU2014/000830 It will also be appreciated that the method may be performed in a "high throughput> "automated" or "semi-automated" manner, particularly during early stages of candidate molecule identification and selection. In a preferred embodiment, the candidate molecule is an antibody or antibody 5 fragment. As hereinbefore described, antibodies may be polyclonal or monoclonal, native or recombinant. Antibody fragments include Fab and Fab'2 fragments, diabodies and single chain antibody fragments (e.g. scVs), although without limitation thereto. Well-known protocols applicable to antibody production, selection, purification and use may be found, for example, in Chapter 2 of Coligan et 10 al, CURRENT PROTOCOLS IN IMMUNOLOGY (John Wiley & Sons NY, 1991 1994) and Harlow, E. & Lane, D, Antibodies; A Laboratorv Manual, Cold Spring Harbor, Cold Spring Harbor Laboratory, 1988, which are both herein incorporated by reference. Polyclonal antibodies may be prepared for example by injecting CD96 or a 15 fragment (e.g a peptide) thereof into a production species, which may include mice or rabbits, to obtain polyclonal antisera. Methods of producing polyclonal antibodies are well known to those skilled in the art. Exemplary protocols that may be used are described for example in Coligan et at, CURRENT PROTOCOLS IN IMMUNOLOGY, supra, and in Harlow & Lane, 1988, supra. 20 Monoclonal antibodies may be produced using the standard method as for example, described in an article by Kahier & Milstein, 1975, Nature 256, 495, which is herein incorporated by reference, or by more recent modifications thereof as for example, described in Coligan el al, CURRENT PROTOCOLS IN IMMUNOLOGY, supra by immortalizing spleen or other antibody producing cells 25 derived from a production species which has been inoculated with one or more of the isolated proteins, fragments, variants or derivatives of the invention. Suitably, the antibody or antibody fragment is suitable for administration to a human. In this context, as hereinbefore described the antibody or antibody fragment may be a "humanized" form of an antibody or antibody fragment produced in, or originating 30 front another species. Such methods are well known in the art and generally involve recombinant "grafting" of non-human antibody complementarity determining regions (CDRs) onto a human antibody scaffold or backbone. In a preferred embodiment, the antibody or antibody fragment does not kill CD96-expressing cells upon administration to a human. In this context, "kiling" 20 WO 2015/024060 PCT/AU2014/000830 may refer to any antibody-mediated cytotoxic mechanism such as complement mediated cytolysis and antibody-mediated cell-mediated cytotoxicity (ADCC), the latter typically mediated by natural killer (NK) cells, macrophages, neutrophils and eosinophils. In this regard, it may be advantageous to use antibody fragments lacking 5 an Fe portion or having a human Fe portion (e.g a humanized antibody). A CD96-inhibitory agent screened, designed, engineered or otherwise produced according to the aforementioned aspect may be used according to the method of the first aspect (e.g as an anti-cancer agent and/or an anti-viral agent), preferably in the form of a pharmaceutical composition as hereinbefore described. 10 So that the invention may be readily understood and put into practical effect, reference is made to the following non-limiting examples. 21 WO 2015/024060 PCT/AU2014/000830 EXAM PLES Example I CD96 binding to CD15$ and the ef tas of CD96 inhibition and knockout in mouse 5 tumor models Materials and Methods Mice 10 Wild Type C57BL/6 mice were purchased from the Walter and Eliza Hall Institute for Medical Research or ARC Animal Resource Centre. C57BL/6 CD96 mice were generated by Dr. Marco Colonna and Dr. Susan Gilfillan at the Washington University School of Medicine (St Louis, MO, USA) as follows,. A targeting construct designed to replace exons I and 2 of CD96, including the start 15 site, with a MCI-neor gene flanked by loxP sites was electroporated into E14.1 (129P2/OlaHsd) embryonic stem cells (Fig, S I). Chimeras transmitting the targeted allele were obtained from two clones following injection into C.57BL/6 blastocysts. Mice carrying the targeted allele were bred to C57BL/6 mice expressing a Cre transgene under the CMV promoter to delete the MCI -neor gene (Schwenk et al, 20 1995). The CD96 deletion was backcrossed onto a C57BL6 background, facilitated by a genome-wide screening of polymorphic microsatellite markers at 10 centiMorgan intervals at each generation. CD96* >99% C57BL/6 mice were intercrossed to generate the CD96 mice. DNAMI-1 mice have already been described. DNAM-W CD96- were generated by intercrossing CD96" with DNAM 25 1- mice. Mice were bred and used between the ages of 6-14 weeks. All experiments were approved by animal ethics conunittee. Ce Culture B16F10, RM-1, 3LL, AT3, MC38 and YAC-1 cell lines were grown in 30 complete RPMI Medium (Gibco, luvitrogen) i.e supplemented with 10% FCS (Thermo Scientific)., L-Glutamine (Gibco), Non-Essential Amino Acids, Sodium Pyruvate, HEPES (Gibco) and Penicillin/Streptomycin (Gibco), at 37 )C in 5% C02 For cytotoxicity assays and IL-12/L-18 titration experiments, primary NK cells were harvested from the spleen, sorted using a mouse NK cell isolation kit (Mitenyi 22 WO 2015/024060 PCT/AU2014/000830 Biotec) and AutoMACS (Miltenyi Biotec), and subsequently cultured for 5 days in RPMI Mediun. supplemented with 10% FCS, L-Glutamine, Penicillin/Streptomycin, Non-Essential Amino Acids (Gibco), Sodium Pyruvate (Gibco), HEPES (Gibco), $3 2-mercaptoethanol (Calbiochem), and 1000 IU/ml recombinant human IL-2 (Chiron 5 Corporation). All cells were incubated at 37"C in 5% COs 2 in vivio LPS challenges LPS (from 16. Goli 0127138, Sigma) suspended in PBS was injected intraperitoneally into mice at the described doses- For survival curves, mice were 10 checked hourly for symptoms of sepsis. Serum from these mice was taken at various time points by retro-orbital or cardiac bleeding for cytokine analysis. Spleens were also taken from mice at various time points to analyse receptor and ligand expression, and intracellular IFN-y expression from NK cells. 15 In vivo funor challenges Mouse B16F10 or B16-OVA melanomas, RM-1 prostate carcinoma, 3LL lung carcinoma, MC38-OVAdi, colon adenocarcinoma or AT3-OVA ® mammary carcinoma, were injected into WT, DNAM1- CD967 or DNAM-4CD96" mice subcutaneously or intravenously at the indicated doses and monitored for solid tumor 20 growth or metastasis, respectively. Treatments were administered as indicated in the Figure legends. To monitor solid tumor growth, the area of the ensuing tumor was calculated by taking the length and width of palpable tumors by calipers and plotted against time. To monitor metastasis formation, 14 days after cells were injected, lungs were harvested, placed in Bouin's fixative, and metastases were counted using 25 a dissection microscope. MCA-inducedfibrosarcoma WT, DNAM- t CD96 and DNAM-lCD96 mice were injected subcutaneously on the right flank with various doses of MCA (5-400 pg, e.g. 100 pg 30 MCA) and were monitored over time for fibrosarcoma formation. In addition, some mice were treated with control antibody, depleted of NK cells by treatment with anti asialoGM I (Wako Chemicals; 100 gg injected i.p. at day -1, 0 and then weekly until 23 WO 2015/024060 PCT/AU2014/000830 week 8), neutralized for IFN-y (H22, 250 pg injected ip. at day -1, 0 and then weekly until week 8), for CD155, for DNAM-I or CD96. Dendritie ces (MDQ; NK ee coculture essays 5 BMDC were generated as described previously. Briefly, we harvested bone marrow cells from the femur and tibia of mice and cultured them in DMEM supplemented with 10% FCS, L-Glutanine, Penicillin/Streptomycin, Non-Essential Amino Acids, Sodium Pyruvate. P-2-mercaptoethanol and 250 ng/ml ilGM-CSF (eBioscience) for 6 days, WT or CD96 - NK cells were harvested from the spleens 10 and FACS sorted to purity by staining with NKl (PK136) and TCRp (H57-597) and CD3 (17A2) antibodies, NK cells were harvested on the day of the assay. For assay set up, 5 x 10 BMDM were plated in 96 well U bottom plates. NK cells were then added to the BMDM at varying titrations (2:1, 1:1, 0.5:1, and 0.25:1). BMDM only and NK only were always included in the assay as controls, Once all cells were 15 plated, each well was filled with the appropriate amount of media to yield equivalent volumes between wells. 100 ng/ml of LPS was then added to the wells for 2 h, followed by 5 mM purified ATP (Sigma) for 30 mins. This was performed at 37C in 5% CO 2 . LPS only and ATP only controls were also included in the assay as controls. After 30 mins with ATP, supernatants were harvested and stored at -20 0 C 20 until analysed. 0, Cvtooxicity .Assays Standard 5Cr cytotoxicity assays were used to analyse the ability of WT and CD96- NK cells to kill targets. Briefly 20,000 targets labeled with 100 pCi of "Cr 25 were added to V bottom plates and NK cells were then added to the targets at defined effector to target ratios. After 4 h at 37"C in 5% COI, supernatants were harvested, and the level of 5Cr was quantified by a gamma counter (Wallac Wizard). Percentage specific killing was determined using the formula (Sample Cr release Spontaneous Cr release)/(Total Cr release-Spontaneous Cr release) x 100. 30 Cy10kine Detectian All cytokine detection in serum or supernatants except IL-I 8 was achieved by utilising Cytometric Bead Array (CBA) technology (BD Biosciences). Acquisition 24 WO 2015/024060 PCT/AU2014/000830 was completed using a Canto 11 or LSRR1 Flow Cytometric Analyser (BD Biosciences). Analysis was performed using the FCAP array software. IL-18 was detected by an ELISA according to manufacturer's instructions (MRL). For intracellular cytokine detection, isolated lymphocytes were obtained from the liver, 5 stained for surface markers, fixed and permeabilised (BD Biosciences), and stained with an anti-IFN-y antibody (XMG 12). Flow CY/omnetry Analysis and Sorting Analysis of Immune Cell Homeostasis and CD96/CD155 expression Various 10 organs (lymph node, lung, spleen, bone marrow, and liver) were processed into single lymphocyte suspensions that included red blood cell lysis Between 1 x 104 and 5 x 1 0" cells were initially subject to incubation with 2.402 to block non-specific Fc antibody binding before specific antibodies were utilised. To analyse NK cell homeostasis and IFN'- production, the following antibodies were used: anti mouse 15 NKLI, -TCRp, -CD27 (LG.7F9), -CDIlb (MI/70), and -lFN-v. For T cells: anti mouse- TCRp, -CDS (53-6.7), and -CD4 (RM4-5). For B cells: anti mouse -B220 (RA3-6B2), -CD19 (ID3). For NKT cells: mouse CDld tetramer loaded with a galactosylceramide (kindly provided by Professor Dale Godfrey University of Melbourne), anti mouse- TCRP or -CD3, -CD4, and NKl 1. For macrophages: anti 20 mouse- F41180 (BN,8) and -CDI lb. For neutrophils: anti mouse- Ly6G (IA8) and CDIlb. For conventional DC: anti mouse- MiHC 11 (.M5/114.15.2) and -CDIic (N418). For v§ T cells: anti mouse -y6 TCR (GL3) and -CD3. To analyse CD96 and CD155 expression, the specific cell type of interest was gated upon using the above antibody cocktails along with anti mouse- CD96 (3.3.3) or anti mouse- CD155 25 (4.24.3)s Acquisition was performed using an LSR 1, or Canto I flow cytometric analyser (BD Biosciences). Analysis was achieved using FloNjo (Treestar). (iC/ Sorting Naive NK cells and macrophages from the spleen were prepared and stained 30 for as described above. These cells were then sorted to purity using an Aria I FACS sorter (BD Biosciences). 25 WO 2015/024060 PCT/AU2014/000830 Statistical Anuysis Statistical analysis was achieved using Graphpad Prism Software. Data was considered to be statistically significant where the p value was equal to or less than 0.05. Statistical tests used were the unpaired Student's t test, Mann Whitney t test, 5 and the Mantel-Cox test for survival. The appropriate test used is defined in the Figure legends. Results CD96 competes with DNAM-I for CD155 binding (Figure 1) and CD96 10 engagement by CD155 down-regulates N K cell production of IFNy (Figure 2) CD96 limits NK cell-dependent tumor immunosurveillance in MCA-treated mice and promotes experimental B 16F10 lung metastasis (Figure 3). The data in Figure 4 show that anti-CD96 mAb has single agent activity (ie without anti-PDI treatment) while also enhancing the anti-tumor responses of anti 15 PD1. Anti-CD96 mAb treatment also enhances and-tumor responses generated by Doxorubicin (DOX) chemotherapy (Figures 5 & 7) which is consistent with Figure 6 where enhanced anti-tumor responses to Doxorubicin (DOX) chemotherapy were observed in host with CD96 deficiency. Referring to Figures 8 & 9, given early or late, anti-CD96 mAb enhances anti-tumor responses generated by anti-PD-1 and 20 anti-CTLA-4 mAbs and shows a particularly strong synergy with anti-PD-I. The effect of CD96 in promotion of tumour metastasis was also investigated. In Figure 10, regulation of B16FlO lung metastasis was investigated in C57BL/6 wild type (WT), DNAM- I<, CD96t and DNAM- I CD96" mice In Figure 11, host CD96 promoted RM-1 lung metastasis and in Figure 12, host CD96 promoted 3LL 25 lung metastasis Figure 13 shows that anti-CD96 mAb suppresses B16F10 lung metastasis, alone and in combination with a T cell checkpoint blockade. In Figure 14, anti-CD96 mAb suppresses RM- 1 lung metastasis, alone and in combination with T cell checkpoint blockade. In Figure 15, anti-CD96 mAb enhances anti-tumor responses generated by anti-PD-1 and anti-CTLA-4 niAbs against MC38 colon 30 tumors and shows a particularly strong synergy with anti-PD- 1 26 WO 2015/024060 PCT/AU2014/000830 Example 2 Screening assays for identifying anti-CD96 antibodies Introduction 5 The following assays may be used to identify antibodies useful in the invention The first assay would be used to identify human antibodies capable of blocking or inhibiting binding between human CD96 and human CD155, The second assay may be used to test whether or not the identified antibodies cause antibody-dependent cell-mediated cytotoxicity (ADCC). The third assay can then be 10 applied to lead candidates and involves determining whether or not a human CD96 antibody can modulate human lymphocyte effector function, Materials and Methods 15 Assay 1: (D96 binding to CD155 The ability of candidate anti-CD96 antibodies to prevent the binding of CD155 to the cell surface of CD96 expressing cells (such as NK cells) will be tested as follows. Recombinant Human CD155 fused to the C terminal Fc region of human LgGI (such as CD155-Fe available from Sino Biological) will be labeled with a 20 fluorophore such as Alexa Fluor 647 (AF647) using Zenon Human IgG Labeling kit (Molecular Probe) accordingly to the manufacturer's instructions. NK cells or other CD96-expressing cells freshly isolated from the peripheral blood of healthy donors will be incubated with AF647 labeled CD155-Fc in the presence of anti-CD96 or control Ig at different concentrations ( The cells will be harvested and the cell surface 25 binding of AF647-CD155-Fc will be tested by flow cytometry) Antibodies that prevent binding of CDI 55 cells to CD96-expressing cells will be identified by their ability to block binding of CDI55-Fc to CD96-expressing cells. Assay 2: ADCC assay 30 The survival of immune cells (such as NK cells and/or T cells) in the presence of anti-CD96 antibodies will be analyzed as follows. The peripheral blood immune cells from healthy donors will be isolated by Ficoll gradient separation. Inmune cells will be plated in 96 well plates in the presence of human [L-2 at an appropriate dosage and increasing concentrations of anti-CD96 mAbs. The survival 35 as well as the percentages of CD96 expressing cells (such as NK cells and/or T cells) 27 WO 2015/024060 PCT/AU2014/000830 will be analyzed over time by flow cytonetry. A non-limiting example of a suitable commercially available kit for this assay is the Annexin V Apoptosis Detection Kit. Assay 3: Assay fir modniulion of/human ikukocye effecior imetion by hunan CD96 5 antibodies Fresh blood samples will be collected from healthy donors. Peripheral blood mononuclear cells (PBMC) will be prepared on a Ficoil-Paque density gradient by centrifugation Highly pure CD3-CD56-+ NK cells will be obtained from PBMC by magnetically activated cell sorting. To analyze the ability of CD96 to impact human 10 NK cell production of IFN-y. 96 well U bottom plates will be coated overnight at 4 0 C with recombinant Human CD 155-Fe chimera (Sino Biological Inc,; 0_25 ge/ well) or with non-relevant human IgGI antibodies. Freshly purified human NK cells will then be plated in complete RMPI media supplemented with Human IL-12, IL-18 and optionally IL-15 for 24 h and the intracellular content and the level of IFN-y in the 15 supernatant will be analysed in the different cultures. Alternatively, human NK cells will be stimulated for 24 h in wells coated with anti-NKG2D, anti-NKp46. anti NKp30 or anti-CD16 antibodies to analyze the ability of CD96 signalling to interact with other NK cells receptors. The anti-human CD96 antibodies to be tested or control antibodies will be added to the cultures prior to the cytokines or antibodies 20 above to confirm the ability of these test anti-human CD96 antibodies to enhance the IFNy production of the human NK cells. Statistical increases in IFNy production above the control would be considered significant. Example 3 25 Assay frr modulation of'mouseNK cellfunction by anti-CD96 antibodies Additional anti-mouse CD96 antibodies would also be screened for the ability to modulate CD96 signalling activity. To analyse mouse NK. cell production of IFN y, 96 well U bottom plates would be coated ovemight at 44C with recombinant 30 mouse CD155-Fc chimera (Sino Biological Inc.; 0.25 tgi well) or with non-relevant human IgG I antibodies. After three washes with PBS, freshly purified NK cells front the indicated mouse strains would be plated in complete RMPI media supplemented with mouse IL-12 (ebiosciences; 25-100 pg/ml) and mouse IL-18 (R&D; 50 ng/mi) 28 WO 2015/024060 PCT/AU2014/000830 for 24 h. Alternatively, IL-2-activated NK cells would be stimulated for 6 h in wells coated with anti-NK 1. . (PK 136; 0. 125 tg/ well). At different time points, anti mouse CD96 antibodies (50-200 ptgfml) or control antibodies will be added to determine whether these enhance NK cell IFN-g production as measured by CBA 5 analysis. Example 4 Production ofumouse and human anti-CD96 antibodies Human CD96 is a transmembrane protein that exists in two isoforms. 10 Isoform i has been detected in acute myeloid leukaemnia and includes additional amino acids compared with isoform 2. Isofonn 2 is the common form in humans and the predicted domain architecture of isoform 2 has three (3) extemal immunoglobulin-like domains (domains 1., 2 and 3), as listed in Table 1. Antibodies against isofonn 2 are preferred for use in the present invention. The nucleic and 15 amino acid sequences of isoform 2 are given in the NCBI consensus sequence number CCDS2958 I (SEQ ID NOS: I and 2 respectively). SEQ ID NO: I. Human CD96 CDNA isoforn 2. AIGGAGAAAAAATGGAAATACTGT C C TATTACAT CAT C&AGATACATTTTGTCAAGGGAGTTTG 20 GGAZAAACAGTCACACAGAAGAAATGITTATGCTACAICTTGGCICTATGIAOAACCGACCGC AAACACAGACACTAGGCTTCTTCGTGCAGATGCAATGGTCCAAGGTCACCAATAAOATAACCTGATT GCTGTCTATCATCCCCAATACGGCTTCTACTOGTGCCTATGGGAGACCCTGTGAGTCACTTGT57ACTTT CACAGAAACTCOGAGAATGGGTCAAAAGGACTCTGCACTTAOAATATGT T TGT _CAGTAGTG GAAGGTACGAOTGTATGCTTGTTCTOTACCAGAGGGCATT'AGACTAAAATCTACAAC0TTCTCATT 25 CGAC5C5CT1ACAGCAGATGAAT AAC CA ACCATACCATA0AAATA0AGATAAATCAG 0ACTT GGAAA0A5GC T STCAAAATAGCT00T AAAAATTTCATC TGGST 2T TGCATGnGTAGGTG AGGATAATGGAA C'PThAAA TTAAT0TCCAAAAT CACTAT0AGCAA5T TCACATTACTT AAA GATAGAGTCAACGTTGGTACAGA0TAOAGACTACCTCTCT0ATCCA'AATCTTCGATGATGGGCG G ATTCTCTTGCCACATTAGAGTCGGTCCTAACAAAA0T00GAAGCTCC0CACOCACTCGTT 30 TGCTAAACAGAAAATCOCOTGTATTTGGAAA:TACACTCAGGATCTTGTAGAGGAGCAT AUCTGTTACCTAAAQAATATrTCACCAAGCAAATATCAATGGTTATAGAGG0TT057 :TTC CA TGATGAAAAAOAAGGAATGGAT ATTAOTAA TG5AAAGAGAAGAAAGATGGATTT TTGACTOA AGTC~rTGTACAAGGTACATTAjCATAATACCAGCACAGCAGACAT TTGTCGATG 0CCTTOTCCACCOAGGAAACTAGTGAACAT CCATTCAGAAA TCTTTCTTTAG 35 TETGAiTCACGCCGTC7acTGAGTTTACGATCTAICCTGATaCAcTTCTC HAAGTOTATC10TOOT5CAAGTA Tm "CC"C "AATCT 0T AGCTTGTACA TOTGAGT0CC TTGAGGCAAAOAGCCA ECAACCAGCOOGAATATGAOTAOCCCAOGCTTCAACTTCCTG GA050 GGGAAGT0CAAATP TT AGTACC TAGTGAAACATACAGTT 2AT000 0CTCAGTw>aCAGGCTCAA5A05TCATGAAATGT0OTTTA0O50AAAAACTT TTCA0AAGT0 40 GCCA0CTGCCAAGGAPTOO0AGAAAACTAATCA5 GT0CATTCA0TGTATTGTGGTC0AATAAGCC CAAAGATGGAATG 0CCTGGCCAG GAT TGTAGCAGCTTACTCTTA TTGCTGCAWGATAT TGTTTGGTC T70GAGTG5AGAAAATGG07G7CAGTA20AAAAAGAAATAA5TGGAAAGACCT CCACCT TCAAG55CCAC2A CCACCTCCCATCAOTACATTGCATTCA5AGAGCCCAACGAAAGTGATCTGCCTTATCATGACATGGA GAC CTAG 45 SEQ ID NO:2 Human CD96 protein sequence isoform 2 29 WO 20 15/024060 PCT/AU2014/000830 NE.KKWtv A§IYY Ii I IFtVKGV EIVNrEEHVYATLGS DVNLTCQIQTVGFFVOtQ S VK'>oacLI AVYHPQYGFYCAYGPPOTS LVTFTETPBNCSKWTLHLRUMSCSVSGR. LYPBGI T KTYNLLM Qi'PVTADEWNSNHT!IE NQTLEI£CFfQNSSSI TSSE.FTYAWSVDNGTQBTISNFISUSTLLK DRVKLGT:YRLHLSPVQKFDDGRFSHRVGPNK LPS T\/KVFAKFE:?V:VENNSTDVLVBR 5 TLLKNVFPYAUI ThEEDC,$FLFDEKFGIYITNEDY&KDGFLELYSVLTRVES NKPXQSLDhLTIWCM.
ALSPFGNN~I SSEKTTFLLGSETDPPLS7TESTLDTQ PS PA522AYPTSTLD& LRNTTPF$NSMTTRGFNY FWSSGTDTKK$SRI £SE:YsSPSGACSTLHDNVFTSTARAFSEV wTTANGSTHTNHVH ITGIVT 7 NKPKDCMSWVIVAALLFOCMILFGLGVRKWCQYQKfIMERFPFKPP BIQIKYTC2QEPM'QPLPYFEEtL 10 Thie Mouse CD96 protein is also a transrnembrane protein but oniy a single transcript/isoform-. is known in the m-ouse as shown in SEQ ID NTs. 3 and 4. SEUQ ID NO: 3. Mouse cDNA. 15 GG~ACGAGCATTCAATcGCGAATgA CGTGTAT7CTCTACCTGOTTCTG;CATCUWFVGACCTGCC GOTOT TTACATOCCAATGOCTCTACTGCGGCAGGGCATGCCTGTGGTACAA-TGGCTGC CACAGAAAOTIGAGGTGTAACAAATTGGAOCTG iA 2:AAGGAATATC2T±'CTGGCC1GGGTG GPA GTATG ASTCATCTTAfl TCOGTT YCCAAOTCQTC AAGCAkCA GTCACAACCT<ATT T 20 GAAXCCO TATAcACAAGA:GAACAoAACTATAcAZTAGAAA:AGAGAoAAATcGAcTTGGAAATACC ATGC:TToCAAAATAccToCTCAGAAATTouAcoTAGGTTcLcTTTTATGGT TGGTGGA0AAAGATG GAGTGGAAGAAGTWCTCTTCACCQAOCAI'CACCAGT CAAiTTCCACATCA TTTkAGGCAGAATC CTGCCATGT GTCCAAOOTCTAAA S OCTGGTG:AOACAGATCCTTTTGAU ZAAC 25 CAG,:ASATCTC.7ATCGGAWA AAGAATGGTGCTAGG=AGZAGAA7WIAGCTA TAACAAOTGCACAG>-GACACOAATCAC.AACAGA2OCTGG'2TATG$ICCTCCtA GSGCCCCAGA0-?kAMJ' A 7(2RACTTCATCACAACCCAZ kJGGrYT UCTTIWGG GATAGC 30 COCAAC CAAACAT2TACCC ACTGTGACA4C.TTCTACCCTCGCITACACAACCThTTTVCACATCTCA TATXX CCTACAGG2TACTcAGCT;ACCCTnvYkGrcAATACATAAAPNr&cGTACkcCcIcAT C-CAACTCCTCAACCAGOAATTCCAGOCGACTACTAAAGATGG. .C.ACTWTGAA-OTC'ATG GACAGATGCCAAGAAC XC§OAAGAGQTCtTGCT I(TICTAAAAGTGGAICT2IGGCCITUCTIICA QVFTCTO CTCflACAATGGCACThACT§CCY 7A-ZCGTAGCTC VQ4C7ACT~CCCCACTT 35 TCAGAACLTAGAAATCGC:TOnGCGTnCCTAGICn CATCACTCAACAOTAGATTTOCAAATGTCCTCAOAAO TGOAAATGGAAOTACTAAAATTGACCATG GACCIATQACCAGTATCATAGTTAAOPAAC2CAC4IGATGGAATGTC2CGCC GTGCTIGTCGCGGCT TGo? QTTTTTCTGQCA O TAT TGTTTGGC CTTCGAGTAAtA kAATGGfl TCGQTA'TC AkkCAA AT 40 TTGCAIPGGkCT GTGTAUA&LAAG T2CTC3A SEQ ID NOA4 Mouse, CD96 amnino acid sequence MCRKWTYCVVYI T FGNUT7GLY 3UNTQTUULQQSVDKNDMI AL YHP QYGLYCGQEHACE SQVAOTE TEKGVTNNT LYLPNIS SALCGKYEC I ILYPEC I T TVYNL IV, 45 EDYTQDEENY TI~E TNRTLE I C EQNTTS SL 22 RFFSWLVEKDGVEEVLrc dhhHvNNT~SFKGR: RLQGDYRLALSPVQ IQDDGPTF$C HLI VNUtKAWKMSTTiVVASPE ILMI VE STMDVLGERVFTCL TKNVFPKANT TWF TDGRFLQCNEFC ITTTNEEKNCS 32 F7E'F SVLTRMB2GP SQSNNMTAWCMkT$P CPK FN?$SQF: TVSFD$VIAPTKHLFTVTG$TLC-TQPFSDAGVSPTGYZATPSVTTVDENGLTPD A:PQ:SNSMTTK:GNYZEAS SGTDAKKS SRAAAS.SKSGS WFFPFTS PPEWHS' 27TSTGFQE PD$PV., 50 $WPEHSPOSAHTITTFMLTAGIIMPTIVUSGSPL; LL ?FCTLLFCLGVRKWYYRYQNEiMERPPPFKPE PPPIKYTY TQEPIGCDLCCHEMEVL.
Table In External immunoglobulin like domains of CD96 from Interpro predictions 30 WO 2015/024060 PCT/AU2014/000830 Human CD96 isoform 2 Mouse CD96 residue residue numbers numbers Domain 1 30-137 30-137 Domain 2 148-250 145-247 Domain 3 253-359 250-355 The external domains of the mouse and human CD96 proteins will be cloned into appropriate expression constructs for expression in mammalian cells such as human embryonic kidney cells. Suitable expression constructs typically include a S CMV promoter to drive expression of the CD96 gene fragnent. Following transfection into mammalian cells the protein will be expressed under suitable culture conditions before being purified prior to antibody production. Four CD96 knockout mice will be immunized with the mouse CD96 external domain protein and likewise four CD96 knockout mice will be immunized with the 10 purified human CD96 external domain protein. Immunizations will be made at approximately three times at four week intervals. The mice will be bled 10-1,2 days following the third immunization and the sera titrated on the screening antigens by ELISA. The mice with the highest antibody titres would be used for fusion, otherwise if mice have not responded adequately further immunisations would be 15 undertaken. Selected hybridomas will be cloned and mAbs purified from each clone before the individual human or mouse mAbs are screened using the screening assays of Examples 2 or 3 respectively. Isotyping of clones would optionally be undertaken in order to identify antibodies that are less likely or unable to induce ADCC, such as [gG2 and IgG4 antibodies. Approxniately 20 antibodies against human CD96 and 20 20 antibodies against mouse CD96 would be obtained as listed in Table I of Example 4. Example 5 Screening of mouse and human anti-CD96 antibodies 25 Approximately 20 each of anti-mouse CD96 and anti-human CD96 monoclonal antibodies would be obtained as described in Example 4. The anti human CD96 antibodies would be screened for the ability to modulate CD96 signallig activity using the Human NK cell assays described in Example I An additional four commercially available anti-human CD96 antibodies (I CS, NK92.39 31 WO 2015/024060 PCT/AU2014/000830 3H8, MAA6359) will also be screened for the ability to modulate CD96 signalling. Anti-mouse CD96 mAbs would be also screened for their ability to modulate NK cell function as described in Example 3. As can occur with antibodies, not all antibodies are expected to have a useful 5 effect on a given target. Accordingly, CD96 signalling for each antibody will be assessed using the human or mouse NK cell assays to determine which antibodies have an effect upon CD96 signalling. Preliminary restdts using the human NK cell assay 10 Using the human NK cell assay described in Example 2, the NK92.39 human CD96 mAb was found to increase the levels of IFN-y in human NK cells as shown in Figure 17A and 17B. This result indicates that antibodies against the human CD96 receptor can be effective in increasing IFN-yproduction in human NK cells. 15 Example 6 Anti-mouse CD96 antibody testing in cancer models Anti-mouse CD96 antibodies that are found to be active in modulating CD96 signalling will be tested for their ability to relieve immune inhibition, reduce CD96 20 activity and/or enhance or restore immune surveillance in mouse cancer models. Approximately 10 active anti-mouse CD96 antibodies will be individually tested in approximately 5-7 cancer models using In vivo tumor challenges, the same or similar to those used in Example L The efficacy of each CD96 antibody against a respective tumour and/or in a 25 tumor model may vary and some may be more responsive to treatment than others. Non-limiting examples of tumours and tumor models to be tested may include breast, prostate, thing, melanoma, colorectal, pancreatic, endometrial, kidney, bowel, gastric, oesophageal, leukaemia, lymphoma, ovarian, bladder and brain cancers including primary tumours and/or metastases of the aforementioned cancers.. 30 Example.7 Human CD96-antibody binding studies A subset of anti-human CD96 antibodies will be examined further to determine which domains of CD96 are bound by effective CD96 antibodies. As 32 WO 2015/024060 PCT/AU2014/000830 discussed in Example 3 above, isoform 2 of CD96 includes three external domains (Domains 1, 2 and 3). Antibodies will be bound to the CD96 protein and the specific CD96 protein residues bound will be detennined using hydrogen/deuterium exchange and mass 5 spectrometry. Alternatively, other methods may be employed to probe antibody CD96 binding sites such as X-ray crystallography, site directed mutagenesis or other methods known in the art. It is anticipated that effective antibodies may bind one or multiple external domains of the CD96 protein. For example anfi-CD96 antibodies that modulate 10 human NK cell function may bind any of the following combinations of external CD96 domains, with each possible single or combination binding domain shown in brackets: (1), (2). (3)5 (1,2), (1,3), (2,3), (1,2,3). Example.8 15 CD96 roles in both N K and T cel function Natural killer (NK) cells are innate lymphocytes which may be critical to limit early tumour growth and metastasis while T cells may be more important in the control of established and primary tumiours. CD96 is a checkpoint immunomodulator that can affect both NK and T cell function. 20 Primary tumors were used to examine the role of CD96 in T cells. The AT3 OVAim model was used where CD8T effector cells are known to naturally control tunor growth. AT3-OVA" mammary carcinoma (lx 106 cells) were injected subcutaneously- Mice were then monitored for tumor growth and measurements made with a caliper square as the product of two perpendicular diameters (mm 2 ), 25 Anti-CD96 treatment greatly reduced the rate of tumour growth and this beneficial effect could be removed by depletion of CD4 and CD8 T cells using anti CD4/CD8 antibody or by treatment with anti-LIFN-y (Figure 16A and 16B)- This demonstrates that anti-CD96 mAb critically required CD8 T cells and WN-y for full anti-tumour activity in this particular tumour model. 30 33 WO 2015/024060 PCT/AU2014/000830 Example 9 Loss of (D96 from the NK cell surface after aCD96 mAb biniing The mechanistic/signaling effects that may occur upon antibody binding to CD96 may also reveal functional information about the effects of CD96-CDI155 5 binding. To further investigate this, total NK cells were purified from human peripheral blood mononuclear cells (PBMCs) by negative selection using human NK cell isolation kit (Miltenyi Biotec.). Isolated NK cells were then labeled with carboxyfluorescein diacetate succinmidyl ester (CFSE; Biolegend) to measurecellular proliferation, CFSE-labeled NK cells were plated in 96 well U 10 bottom plate at 5 x 10 4 cells/well and stimulated with recombinant IL-2 at indicated concentrations (10 units/ml and 25 units/ml), in the presence of control IgG or anti human CD96 mAb (clone NK92-39) at 30 pg/mL. NK cells were assessed for changes in proliferation or the presence/absence of surface CD96 at day 3 and 6 using BD FACS Canto II (BD Biosciences) and analysis was carried out using 15 FlowJo (Tree Star) (Figure 18 (A) and (B). Anti-CD96 binding to CD96 had no effect on NK cell proliferation but appeared to greatly reduce the level of CD96 on the cell surface by day 6, either by internalization of CD96 following mAb binding or possibly via a reduction in (1D96 expression. Throughout the specification the aim has been to describe the preferred 20 embodiments of the invention without limiting the invention to any one embodiment or specific collection of features. It will therefore be appreciated by those of skill in the art that, in light of the instant disclosure, various modifications and changes can be made in the particular embodiments exemplified without departing from the scope of the present invention. 25 All computer programs, algorithms, patent and scientific literature referred to herein is incorporated herein by reference. 34 WO 2015/024060 PCT/AU2014/000830 REFERENCES 11 Vivier,E, Tomasello, E., Baratin, H, ffiilzer, T & Qgoini S. Functions of natural killer cells. Nature immumology 9, 503-510 (2008), 5 2. Lanier, L Up on the fighirope: natural kiler cell activation and inhibition. Nature immunology 9, 495-502 (2008)% 3. Chan, C.J, Smyth, AIJ& Martinet, L. Molecular mechanisms of natural killer cell activation in response t cellular stress. Cell death and difkrentiation (2013% 10 4. Raidet, D.H. & Vance. R. Self-tolerance of natural killer ce/k Ndure revcvis 6, 520-531 (2006). 5. Fuchs, A. & Colonna, M lhe role of NK cell recognition of nec/in and nectin-like proteins in tumor immunosurveillance. Seminars in cancer biology 16 359-366 (2006k 15 6, Shibua, A, et at iNAM-1, a novel adhesion molecule involved in the cytolyticpmcion of?' imphocyes.Immunity 4, 573-581 (1996% 7 Wang, P.L, 07a rrell, , Clavberaer C. & Krensky, A.A1 identification and molecular cloning oftctilt A novel hinan T cell activation antigen that is a member o/the k gzene superftmil . Immunol 148 2600-2608 (1992). 20 8. Yu, X, el al The surface protein TIG T suppresses 7 cell activation by promoting ihe generation of mature immunoregulatory dendritic cel. Nature imunologv 10, 48-57 (2009). 9. Boles, K Sel ali A novel molecular interaction Pr he adhesion off licular CD4 1 cells to flilicu/ar DC European journal of immunology 39, 695-703 25 (2009). 10. Kennedy, J, et al. A molecular analysis of NKT cells: identification of a class-i restricted T cell-associated molecule (CR114M). Journal of leukocvte biology 6 7 725-734 (2000) I. Boltino, C., e al. identiication of PVi? (C155) and Nectin-2 (CD112) as 30 cell surface ligands for the human DNAM-J ((1)226) activating molecule. The Journal of experimental medicine 198, 557-567 (2003), L2. Lozano. E, Dominguez-iag Mr", Kuchroo, T & Hafiae; DA The II(5/17R )226 axis regulates human T cell fitntion. Journal of itmnlmology 188, 3869-3875 (2012 35 WO 2015/024060 PCT/AU2014/000830 13. Lakshmikanth, T, et al, NCRs and DANAM-l mediate NK cell recogznifion and lysis of human and mouse melanoma cell lines in vitro and in vivo. ie Journal of clinical investigation 119, 1251-1263 (200 9 L 14. Chan., Cd., et al. DNAM-1/155 interactions prontoec epokine and NK 5 cell-mediated suppression of poor immunogenic melanoma me/astases. j Innunol 184, 902-911 (2010). -15 Gilfillan, S., el at DNAM-1 promotes activation of cytotoxic lymphocytes by nonprofessional antigen-presenting ce/ and tumors. The Journal of experimental medicine 205, 2965-2973 (2008). 10 16. Iguchi-Manaka; A, et at. Accelerated tumor growth in mice deficient in DAM-1I receptor. The Journal of experimental medicine 205, 2959-2964 (2008). 17. Staniesk, A, ei al. The interaction of IGIT with PVR and PVTRL2 inhibits human NK cell cytotoxicityj Proceedings of the National Academy of 15 Sciences a/the United States ofAmerica 106, 17858-17863 (2009) 18. Stanieisky, A, el at Mouse TIGIT inhibits NK-cell cytotoxicity upon interaction with PYR Efuropeantournal qf immunologv (2013). 19. Di, S., et al Recruitment of (rb2 and SHIP1 by the ITT-like notif of .TIGIT suppresses granule polarization and cytotoxicit of NK cells. Cell death and 20 di/ krentfiation 20, 456-464 (2013). 20. Fuchs, A., Celia, If, Kondo, 1. & (olonna, A4. Paradoxic inhibition of human natural interferon-producing ce/s by the activating receptor NKp44 Blood 106, 2076-2082 (2005). 21 Seth, S., et al. The murine pan Tcell marker C)96 is an adhesion receptor 25 ftr CD155 and nec/in-I. Biochemical and biophysical research communications 364, 959-965 (2007). 22. Fuchs, A, Celia, A1, Giurisato, E, Shaw, AS. Colonna A. Cutting edge: ()96 activele) promotes NK ce/i-target cell adhesion by interacting with the poliovirus receptor (CI55). J Imnmunol 172, 3994-3998 (2004) 30 23. Ahn, J, Sklton, J & Yu, K Idrogen . Deuterium Exchange Mass Spectrometry: An Emerging Biophysical Tool for Probing Protein Behavior and Higher-Order Structure. LCGC NOR H AMERICA W1V'ME 31 NU MBp R 6 (2013). 36 WO 2015/024060 PCT/AU2014/000830 24 Percy A. J, Re,, M, Burns, K. If & Schrie-ner, D. C Probing protein interactions with hydrogen/deuteriu exchange and mass spectrometry A review. Analvtica Chihnica Acta 721 (20I2) 7 21 (2012). 37