Inhibitory Checkpoint Receptors Research Articles

Inhibitory checkpoint receptors control CD8+ resident memory T cells to prevent skin allergy

Tissue-resident memory T (Trm) cells are detrimental in patients with numerous chronic inflammatory diseases, including allergic contact dermatitis (ACD). We sought to analyze the contribution of Trm cells to the chronicity and severity of ACD and to define the local parameters regulating their development and functions. We used an experimental model of ACD (ie, contact hypersensitivity to 2,4-dinitrofluorobenzene) that is mediated by CD8+ T cells. Our data show that early effector T cells accumulated in the skin during the acute contact hypersensitivity reaction and gave rise to epidermal CD8+ Trm cells expressing a specific set of inhibitory checkpoint receptors (ICRs), such as programmed cell death protein 1 (PD-1) and T cell immunoglobulin and mucin domain 3 (TIM-3). Those Trm cells remained in the epidermis for several weeks and mediated the eczema exacerbations, which developed on allergen re-exposure without the contribution of circulating specific T cells. Furthermore, allergen-induced Trm cell reactivation was constrained because treatment with ICR antagonists dramatically enhanced the magnitude and severity of eczema exacerbations. Finally, we show that the persistence of the allergen in the epidermis for long periods of time was responsible for both the development and maintenance of epidermal Trm cells, as well as the sustained expression of ICRs. Although CD8+ Trm cells are key for the pathophysiology of ACD, intrinsic mechanisms control their reactivation to prevent damaging immunopathology. Developing strategies targeting the reactivation of skin Trm cells in situ through their ICRs should open new perspectives for the treatment of ACD.

IPSC-Derived NK Cells and Anti-PD-1 Antibody Synergize to Enhance T Cell Cytokine and Cytolytic Responses Against Multiple Tumors

The development of immunotherapeutic monoclonal antibodies targeting checkpoint inhibitory receptors (CIR), such as programmed death 1 (PD-1), has transformed the oncology landscape. However, many tumor subtypes are resistant to CIR-targeted therapy, and relapse remains a significant concern. Therefore, combination of novel immunotherapies with CIR targeting remains a promising and widely investigated approach to bolster anti-tumor responses and to overcome tumor resistance to CIR therapy.Natural killer (NK) cells mediate direct tumor cell lysis and are key regulators of T cell responses through the production of inflammatory cytokines and chemokines. In many cancers, NK cell numbers are low and their functional responses are sub-optimal. The use of allogeneic NK cell immunotherapy has shown significant clinical promise for the treatment of acute myelogenous leukemia (AML). However, this approach has inherent limitations with respect to the number of NK cells that can be isolated and variability in the quantity and quality of NK cells between donors. To overcome these barriers, we have developed a system for large scale expansion of NK cells derived from induced pluripotent stem cells (iPSCs) to be combined with CIR antibodies for multiple tumor types (Figure 1A). iPSC derive NK (iNK) cells (defined as CD45+CD3-CD56+) differentiated with high efficiency in this culture system (Figure 1B), and overall expansion from the hematopoietic progenitor stage to end of the protocol was approximately 1 million-fold (Figure 1C).iNK cells displayed a cell surface phenotype typical of primary peripheral blood NK cells, with high expression of the cytotoxicity molecules granzyme B and perforin. Functionally, iNK cells degranulated and produced the pro-inflammatory cytokines tumor necrosis factor (TNF) and interferon (IFN)-γ at high frequencies in response to K562 cells (a myeloid leukemia cell line) (Figure 1D). These cells also effectively killed a broad range of solid tumor targets, including ovarian cancer (SKOV-3), lung cancer (A549) and pancreatic cancer (PANC-1) cell lines, in 2D tissue culture assays (Figure 1E). To interrogate the ability of iPSC-derived NK cells to synergize with CIR therapy, we used an in vitro 3D tumor spheroid system to model the combinatorial effects of T cells, iNK cells, and checkpoint blockade in the context of anti-tumor function. Using SKOV-3 spheroids as targets in a 160-hour killing assay, we found that iNK cells could mediate significant, but not complete destruction of tumor spheroids (46% tumor reduction). Addition of twice as many activated T cells by themselves induced tumor spheroid destruction (58% tumor reduction). The combination of iNK and activated T cells led to robust target cell destruction (71% tumor reduction). Importantly, the combination of activated T cells, iNK cells and anti-PD-1 antibody led to a near complete elimination of tumor spheroid targets, with greater than 99% tumor reduction (Figure 1F). In addition to synergistic lysis of tumor spheroids, analyses of cytokine secretion in iNK, activated T cell and CIR spheroid cultures revealed synergistic production of TNF and IFN-g (Figure 1G).The results from our 3D spheroid assays led us to hypothesize that iNK cells could recruit T cells to promote tumor lysis. To test this hypothesis, we performed transwell migration assays to test the relative abilities of iNK cells and K562 cells to promote T cell migration. We found that, relative to media controls, both iNK cells and K562 cells promoted T cell migration across transwells, and the highest amount of degree of migration was observed in the presence of both iNK and K562 cells (Figure 1H). Finally, we tested iNK-mediated T cell recruitment in an in vivo setting where iNK cells and IL-2 were injected with or without K562 cells into the peritoneal cavities of immune-deficient NSG mice, and activated T cells were injected retro-orbitally into the blood. Mice were sacrificed 4 days later, and total T cells in the peritoneal cavities were determined. Compared to the T cell only control group, iNK cells promoted significant T cell recruitment to the peritoneum, which was amplified upon iNK cell activation with the addition of K562 cells (Figure 1I). Together, our data demonstrate that iNK cells can serve as an off-the-shelf source of NK cells and have the potential to synergize with anti-PD-1 CIR therapy to enhance anti-tumor T cell responses. DisclosuresCichocki:Fate Therapeutics Inc.: Consultancy, Research Funding. Bjordahl:Fate Therapeutics Inc.: Employment. Gaidarova:Fate Therapeutics Inc: Employment. Mahmood:Fate Therapeutics Inc.: Employment. Rogers:Fate Therapeutics Inc: Employment. Ge:Fate Therapeutics Inc.: Employment. Kaufman:Fate Therapeutics: Consultancy, Research Funding. Valamehr:Fate Therapeutics Inc.: Employment.

Adaptive NK Cells Are Expanded By NKG2C-Agonist Antibodies and Retain Their Cytotoxic and Secretory Properties

Natural killer (NK) cells can eliminate myeloid leukemia cells after stem cell transplant, yet many patients relapse. Several reports have described significant decreases in relapse rates post-transplant in recipients that experienced cytomegalovirus (CMV) reactivation. Our group has shown that reduced relapse rates correlated with expansion of adaptive NK cells, which can be defined by surface expression of the maturation marker CD57 and the activating receptor NKG2C. Adaptive NK cells live longer, are more resistant to immune suppression than canonical NK cells, exhibit properties of immune memory, and their presence in peripheral blood is only observed in CMV seropositive individuals. Given the in vitro and in vivo evidence for enhanced function of adaptive NK cells against malignant cells, we sought to develop a method to preferentially expand this subset in vitro. To this end, we stimulated NK cells isolated from the peripheral blood of healthy CMV seropositive donors for 7 days with IL-15 and either a plate-bound control IgG isotype antibody or an antibody that was determined to be a strong agonist of NKG2C signaling in redirected antibody-dependent cellular cytotoxicity (ADCC) assays. We observed robust proliferation of NKG2C+ NK cells when stimulated with the anti-NKG2C agonist antibody (Figure 1A). High proliferation rates were accompanied by significant induction of the checkpoint inhibitory receptors programmed death-1 (PD-1) and lymphocyte activation gene-3 (LAG-3) (Figure 1B), with LAG-3 expression nearly doubling on NKG2C+ NK cells. LAG-3 is a unique checkpoint inhibitor that ligates MHC class II molecules and is highly expressed on lymphocytes in patients with various types of malignancy.The stimulated NK cells were also highly active, as evidenced by a complete switch from a CD45RA+CD45RO- to CD45RA-CD45RO+ phenotype (Figure 1C). CD45ROexpression on lymphocytes has been associated with better prognosis in solid tumors including melanoma, lung, and endometrial cancers suggesting a potent ability to target malignant cells.Despite the increased expression of checkpoint inhibitory receptors, expanded NKG2C+ NK cells remained highly functional. In assays where NK cells were used as effectors against K562 cells, (a myeloid leukemia cell line), we found that NKG2C agonist antibody-expanded NK cells exhibited higher frequencies of tumor necrosis factor (TNF) and interferon (IFN)-γ production compared to NK cells from control IgG isotype cultures. Surprisingly, we found that NK cells expressing LAG-3, instead of exhibiting functional exhaustion, were highly responsive to tumor targets (Figure 1D). To directly test the cytotoxic function of adaptive NK cells expanded with the anti-NKG2C antibody, we analyzed NK cell-mediated killing of dye-labeled K562 cells over the course of 48 hours by live imaging using an IncuCyte Zoom instrument. We observed more rapid killing kinetics and more robust overall target cell killing by adaptive NK cells expanded with the anti-NKG2C agonist antibody relative to the IgG isotype control condition (Figure 1E).Our data show that CD57+NKG2C+, adaptive NK cells can be efficiently expanded using NKG2C-agonist antibodies. These cells exhibit potent killing abilities and enhanced cytokine secretion, making them ideal candidates for adoptive cell transfer in hematologic malignancies. We also show that LAG-3, a checkpoint inhibitory receptor, is actually associated with enhanced NK cell function in these in vitro assays. This is likely because of the relatively short period of activation (7 days) to which the NK cells are exposed during expansion. There are at least four LAG-3 targeted therapies currently in clinical trials with more in development. A better understanding of the mechanisms of expression of this important receptor and its impact on NK cell function will aid in the rational use of checkpoint inhibitory receptor therapies for cancer patients. [Display omitted] DisclosuresHouchins:Bio-Techne Inc: Employment. Cichocki:Fate Therapeutics Inc.: Consultancy, Research Funding.

The Two SH2 Domains of SHP-2 Bridge Two PD-1 Molecules Resulting in SHP-2 Activation and PD-1-Mediated Inhibition

Programmed cell death-1 (PD-1) is a checkpoint inhibitory receptor member of the B7-CD28 family, which promotes peripheral tolerance and restrains anti-viral and anti-tumor immunity. Although PD-1 blockade leads to durable clinical responses in a fraction of patients, the majority of patients display only transient responses, emphasizing the need for better understanding of the mechanism of PD-1-mediated T cell inhibition. In contrast to other CD28 family family members, which form disulfide-linked homodimers, PD-1 lacks the juxtamembrane cysteine residue responsible for homodimer formation and has been shown to exist as a monomer. The cytoplasmic tail of PD-1 has one immunoreceptor tyrosine-based inhibitory motif (ITIM), centered on Y223 residue, and one immunoreceptor tyrosine-based switch motif (ITSM), centered on Y248 residue. Mutational studies have shown that PD-1-mediated inhibition relies on the interaction of the ITSM with SHP-2 but the mechanism by which PD-1 induces SHP-2 activation is unknown. In this study we sought to determine how PD-1: SHP-2 interaction leads to inhibition of T-cell responses. SHP-2 contains a phosphatase (PTP) domain and two SH2 domains, N-SH2 and C-SH2. To determine whether PD-1 selectively interacts with a specific SH2 domain of SHP-2, we generated five different GST-fusion proteins in which GST was fused with either SHP-2 full length (FL), SHP-2-N-SH2, SHP-2-C-SH2, SHP-2-ΔN-SH2 (lacking the N-terminus SH2 domain) or SHP-2-PTP. Surprisingly, pull-down assays with GST-fusion proteins using lysates from primary human T cells or Jurkat T cell line revealed that PD-1 interacted with SHP-2 via both SH2 domains of SHP-2. Biacore assays confirmed that phosphorylated ITSM Y248 interacted with SHP-2 N-SH2 and C-SH2. To determine whether one of the SHP-2 SH2 domains might preferentially interact with PD-1 in live cells, we mutagenized the functional sites of the N-SH2 or C-SH2 domain at arginine R32 and R138, respectively, and transfected COS cells with cDNA of SHP-2 wild type or each SHP-2 SH2 mutant, together with PD-1 and the TCR proximal kinase Fyn, required for PD-1 phosphorylation and interaction with SHP-2. Immunoprecipitation and immunoblot assays showed that mutagenesis of either SH2 domain abrogated interaction of SHP-2 with PD-1 ITSM Y248, providing evidence that both SH2 domains of SHP-2 are involved in the interaction with PD-1. Because each PD-1 molecule has only one ITSM, these results indicate that SHP-2 interacts via its two SH2 domains with two PD-1 molecules. Assessment of PD-1: PD-1 dimer formation in live cells by split luciferase complementation, using NanoBiT proximity assays, showed that upon PD-1 phosphorylation, PD-1: PD1 interaction occurs only in the presence of SHP-2 with intact N-SH2 and C-SH2 domains. The SH2 domains of SHP-2 have a crucial and distinct role in regulating SHP-2 PTPase activity. In the absence of a tyrosine-phosphorylated binding ligand, N-SH2 is bound to the PTP domain leading to an auto-inhibitory closed conformation that blocks the PTP active site. Phosphorylation of Y542 in the SHP-2 C-terminus tail leads to intramolecular interaction of Y542 with the N-SH2 domain to relieve binding to PTP domain. Similarly, intermolecular interaction of the N-SH2 domain with specific phosphorylated ligands disrupts its PTP recognition surface reversing the auto-inhibitory conformation and activates the phosphatase. We determined that monomeric PD-1 ITIM-pY223 or ITSM-pY248 peptides did not activate the PTP. A dimeric phosphopeptide generated by covalent joining of two PD-1 ITSM-pY248 phosphopeptides with a linker that matches the distance between the binding sites of the two SHP-2 SH2 domains induced rapid activation of SHP-2. In contrast, similarly designed dimeric peptides joining either two PD-1 ITIM-pY223 or ITIM-pY223-and-ITSM-pY248 phosphopeptides did not activate PTP activity of SHP-2. The ability of PD1 ITSM-pY248 to induce SHP-2 activation was correlated with inhibition of antigen-mediated IL-2 production, which was abrogated when Y248 was mutagenized to phenylalanine. Our results reveal the geometry of PD-1: SHP-2 interaction that leads to SHP-2 activation and have implications for the development of PD-1-binding compounds to selectively suppress T cell responses by dimerizing PD-1 or enhance T cell responses by disrupting PD-1 dimerization and SHP-2 activation. DisclosuresFreeman:EMD-Serono: Patents & Royalties; Xios: Membership on an entity's Board of Directors or advisory committees; Origimed: Membership on an entity's Board of Directors or advisory committees; Roche: Membership on an entity's Board of Directors or advisory committees; AstraZeneca: Patents & Royalties; Bristol-Myers-Squibb: Patents & Royalties; Bristol-Myers-Squibb: Membership on an entity's Board of Directors or advisory committees; Roche: Patents & Royalties; Dako: Patents & Royalties; Boehringer-Ingelheim: Patents & Royalties; Merck: Patents & Royalties; Novartis: Patents & Royalties.

Ibrutinib-Based Therapy Improves Anti-Tumor T Cell Killing Function Allowing Effective Pairing with Anti-PD-L1 Immunotherapy Compared to Traditional FCR Chemoimmunotherapy; Implications for Therapy and Correlative Immune Functional Data from the Phase III E1912 Trial

The targeted therapy ibrutinib inhibits B cell receptor signaling (BTK inhibitor) and has yielded high response rates and durable remissions in patients with chronic lymphocytic leukemia (CLL). However, it is widely believed that the addition of immune therapies to targeted drugs will be required to activate anti-tumor immunity and work towards curative therapy. Identifying effective combinations of targeted drugs and/or standard chemotherapy with immunotherapy is a priority research area and particularly relevant for CLL, as patients' T cells have been shown to exhibit profound tolerance/exhaustion and notably, no activity was reported in a recent trial of anti-PD-1 immunotherapy for relapsed disease.Ibrutinib has shown beneficial immunomodulatory activity in CLL by inhibiting IL-2-inducible T cell kinase (ITK) as well BTK that is associated with increased effector CD4+and CD8+ T cell numbers and decreased expression of inhibitory checkpoint receptors such as PD-1 on patient T cells. Here we have performed comparative immune bioassays from a randomized phase III trial comparing ibrutinib-based therapy to traditional FCR chemoimmunotherapy to assess the effects of treatments on anti-tumor T cell function.Viable peripheral blood mononuclear cell samples were collected serially (baseline, 6 months and 12 months) from CLL patients on the randomized phase III E1912 trial of ibrutinib and rituximab versus FCR for previously untreated disease to allow longitudinal batched immune analysis.Cytotoxicity assays revealed that highly purified CD3+ T cells from the FCR treated patients at 6 and 12-month time-points did not change their activated killing function against autologous baseline CD19+ CLL tumor B cells (acting as target antigen-presenting cells pulsed with superantigen, sAg) compared to pre-treatment/baseline exhausted T cells (n=22). In contrast, patients treated with ibrutinib-based therapy had a significant increase in activated anti-tumor T cell killing function (P<.01, n=22) at both 6-month (66% increase) and 12-month (89% increase) time-points.Flow cytometric analysis of circulating immune subsets revealed that the percentage of PD-1 and PD-L1 positive cells among CD8+ and CD4+ T cells (particularly effector compartments) were reduced with ibrutinib-based therapy, whereas only a partial reduction was detected following FCR treatment. However, patients' T cells from both treatment arms responded normally to T cell receptor engagement by upregulating these checkpoint molecules. This led us to explore ex vivo treatment of highly purified CD3+ T cells and CD19+ CLL B cells from both treatment arms with anti-PD-L1 or anti-PD-1 immunotherapy prior to cytotoxicity assays. Our functional data revealed that the T cells from both FCR time-points (6 and 12-months) were not sensitive to either anti-PD-L1 (n=14) or anti-PD-1 (n=14) treatment. In contrast, ibrutinib-based treatment sensitized anti-tumor T function (23% increase in killing) following anti-PD-L1 treatment (n=14) at the 6-month time-point only (P<.01) but not with anti-PD-1.To investigate the mechanism underlying these effector function differences, we compared the ability of highly purified CD3+ T cells from each treatment arm (n=45) time-point to form F-actin immunological synapses with baseline autologous CLL tumor B cells. Quantitative confocal image analysis revealed that ibrutinib-based therapy significantly (P<.01) enhanced polarization of F-actin, tyrosine-phosphorylated proteins and granzyme B at immune synapses with tumor cells at both 6 and 12-month time-points, whereas FCR treated patient T cells failed to mobilize these lytic synapse molecules. Importantly, our assays have revealed that T cells from both FCR treatment time-points formed “non-polarized” immune synapses with tumor cells, in keeping with cytotoxic dysfunction and insensitivity to additional checkpoint immunotherapy. In contrast, our functional correlative bioassays have revealed that ibrutinib-based therapy can reactivate exhausted cytolytic T cell function and suggest to us, a potential therapeutic window for anti-PD-L1 immunotherapy at the earlier 6-month time-point.We believe this data supports the concept of incorporating functional bioassays to immune-monitoring assays associated to clinical trials that should aid knowledge-led design of future combination immunotherapy. DisclosuresKay:Janssen: Membership on an entity's Board of Directors or advisory committees; Agios Pharm: Membership on an entity's Board of Directors or advisory committees; Acerta: Research Funding; Cytomx Therapeutics: Membership on an entity's Board of Directors or advisory committees; Infinity Pharm: Membership on an entity's Board of Directors or advisory committees; Pharmacyclics: Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees; Morpho-sys: Membership on an entity's Board of Directors or advisory committees; Gilead: Membership on an entity's Board of Directors or advisory committees; Tolero Pharmaceuticals: Membership on an entity's Board of Directors or advisory committees, Research Funding. Shanafelt:Pharmacyclics: Research Funding; Genentech: Research Funding; GlaxoSmithKline: Research Funding; Jansen: Research Funding. Ramsay:Celgene Corporation: Research Funding; Roche Glycart AG: Research Funding; MedImmune: Research Funding.

Nivolumab for Relapsed or Residual Haematological Malignancies after Allogeneic Haematopoietic Stem Cell Transplantation (NIVALLO)

Aim: To evaluate the safety and efficacy of nivolumab for the treatment of relapsed or residual haematological malignancies after allogeneic stem cell transplantation (alloSCT).Background: Relapse of haematological malignancies following alloSCT is a major cause of post-transplant mortality. Interaction between programmed cell death protein-1 (PD-1) and its ligand (PD-L1) inhibits T-cell alloreactivity and contributes to immune escape. Nivolumab inhibits PD-1 signalling and augments T-cell cytotoxicity. The safety and efficacy of nivolumab post-alloSCT has not been evaluated in a clinical trial.Method: In this investigator-initiated phase IIa clinical trial, patients with relapsed or persistent haematological malignancies following alloSCT receive nivolumab 3mg/kg for up to 48 weeks. Patients with current graft-versus-host disease (GVHD) or prior grade ≥2 acute GVHD or chronic GVHD are excluded.Results: Six participants have received at least one dose of nivolumab at this interim assessment. Primary haematological malignancies relapsing post-alloSCT included Hodgkin lymphoma (HL, 2 patients), acute myeloid leukaemia (AML, 2), transformed chronic lymphocytic leukaemia (tCLL, 1) and mantle cell lymphoma (MCL, 1). The median time from alloSCT to first dose of nivolumab was 25.5 months. Two participants developed grade 3 acute GVHD at 6 days and 13 days following the first dose of nivolumab. Complete or partial responses were observed in 3 participants (50%). Two participants with HL achieved complete responses. One participant with MCL had a complete nodal response with small volume persistent bone marrow disease. One participant with monosomal karyotype AML achieved initial blast reduction (23% to 13%) however subsequently developed progressive AML. T-cell phenotyping at first AML relapse (prior to nivolumab) demonstrated a high proportion of CD8+ T cells that expressed PD-1 and T-cell immunoglobulin and mucin domain 3 (TIM-3) consistent with T-cell exhaustion. Following treatment with nivolumab there was an increase in TNFα production by CD8+ T-cells at day 7 post nivolumab, demonstrating augmentation of T-cell activity. Despite continued nivolumab treatment TNFα production subsequently declined and correlated with loss of clinical response. TIM-3 expression was further upregulated at post-nivolumab progression suggesting this inhibitory checkpoint receptor may have contributed to nivolumab resistance.Conclusion: Nivolumab treatment after alloSCT results in potent immune stimulation with a high rate of clinical responses, albeit with a risk of GVHD. Acquired resistance to nivolumab may develop via upregulation of alternative inhibitory checkpoints. DisclosuresSzer:Alexion Pharmaceuticals, Inc.: Consultancy, Honoraria, Other: Travel Support , Research Funding. Grigg:BMS: Membership on an entity's Board of Directors or advisory committees; Roche: Membership on an entity's Board of Directors or advisory committees; Takeda: Membership on an entity's Board of Directors or advisory committees; Gilead: Membership on an entity's Board of Directors or advisory committees.

Abstract 3824: Antitumor activity of REGN4659, a fully human anti-CTLA-4 monoclonal antibody, against MC38.Ova tumors grown in immunocompetent human CTLA-4 knock-in mice

Abstract Cytotoxic T-lymphocyte associated antigen 4 (CTLA-4) is an inhibitory checkpoint receptor that is expressed by activated conventional T cells and regulatory T cells and negatively regulates T cell activation. CTLA-4 plays a key role in restraining the adaptive immune response of T cells against a variety of antigens, including self-antigens. However, in the tumor microenvironment, the co-inhibitory CTLA-4 receptor contributes to tumor immune escape. CTLA-4 shares its two ligands, CD80 and CD86, which are largely expressed on antigen-presenting cells, with CD28, a co-activator of T cells; however, CTLA-4 binds CD80 and CD86 with higher affinity than CD28. Blockade of CTLA-4 elicits an effective antitumor immunity in preclinical animal models and has demonstrated favorable clinical results in several solid malignancies. REGN4659 is a human IgG1 antibody that binds to the extracellular domain of human and monkey CTLA-4 with high affinity and specificity. REGN4659 blocks CTLA-4 interaction with CD80 and CD86 ligands and enhances the cytokine release of activated primary human T cells in vitro. The effects of treatment with REGN4659 on the subcutaneous growth of a mouse adenocarcinoma tumors expressing chicken ovalbumin (MC38.Ova) were evaluated in immunocompetent mice genetically engineered to express a human CTLA-4 chimeric protein from the endogenous mouse locus (CTLA-4 hum/hum knock-in). Prophylactic treatment with REGN4659 significantly suppressed MC38.Ova tumor growth in CTLA-4 knock-in mice in a dose dependent manner, and improved mouse survival. Because patients treated systemically with CTLA-4 blockade have suffered from immune-related adverse events, we investigated the efficacy of a peritumoral low-dose administration of a mouse-reactive surrogate anti-CTLA-4 antibody in an attempt to minimize systemic exposure. Peritumoral injection of anti-CTLA-4 antibody was as efficacious as intraperitoneal delivery at inducing anti-tumor immunity. Peritumoral anti-CTLA-4 antibody delivery was associated with lower circulating antibody levels but nonetheless caused systemic antitumor immunity as revealed by growth retardation of concomitantly implanted distant tumors. The mechanism of action and robust anti-tumor efficacy of REGN4659 supports its clinical development for the treatment of human cancers. Citation Format: Elena Burova, Omaira Allbritton, Susannah Brydges, William Poueymirou, Robert Durso, Douglas MacDonald, Ella Ioffe, Markus Mohr, William Olson, Gavin Thurston. Antitumor activity of REGN4659, a fully human anti-CTLA-4 monoclonal antibody, against MC38.Ova tumors grown in immunocompetent human CTLA-4 knock-in mice [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3824.

Checkpoint inhibition: NK cells enter the scene.

Tumor-associated natural killer cells have elevated expression of the checkpoint inhibitory receptor TIGIT. Monoclonal antibody–mediated blockade targeting TIGIT unleashes the anti-tumor activity of both natural killer cells and T cells in preclinical mouse models and efficiently delays tumor growth.

Expression of the DNA-Binding Factor TOX Promotes the Encephalitogenic Potential of Microbe-Induced Autoreactive CD8+ T Cells

Infections are thought to trigger CD8+ cytotoxic Tlymphocyte (CTL) responses during autoimmunity. However, the transcriptional programs governing the tissue-destructive potential of CTLs remain poorly defined. In a model of central nervous system (CNS) inflammation, we found that infection with lymphocytic choriomeningitis virus (LCMV), but not Listeria monocytogenes (Lm), drove autoimmunity. The DNA-binding factor TOX was induced in CTLs during LCMV infection and was essential for their encephalitogenic properties, and its expression was inhibited by interleukin-12 during Lm infection. TOX repressed the activity of several transcription factors (including Id2, TCF-1, and Notch) that are known to drive CTL differentiation. TOX also reduced immune checkpoint sensitivity by restraining the expression of the inhibitory checkpoint receptor CD244 on the surface of CTLs, leading to increased CTL-mediated damage in the CNS. Our results identify TOX as a transcriptional regulator of tissue-destructive CTLs in autoimmunity, offering a potential mechanistic link to microbial triggers.

The multi-receptor inhibitor axitinib reverses tumor-induced immunosuppression and potentiates treatment with immune-modulatory antibodies in preclinical murine models.

Cancer immunotherapies have significantly improved the prognosis of cancer patients. Despite the clinical success of targeting inhibitory checkpoint receptors, including PD-1 and/or CTLA-4 on T cells, only a minority of patients derive benefit from these therapies. New strategies to improve cancer immunotherapy are therefore needed. Combination therapy of checkpoint inhibitors with targeted agents has promisingly shown to increase the efficacy of immunotherapy. Here, we analyzed the immunomodulatory effects of the multi-receptor tyrosine kinase inhibitor axitinib and its efficacy in combination with immunotherapies. In different syngeneic murine tumor models, axitinib showed therapeutic efficacy that was not only mediated by VEGF-VEGFR inhibition, but also through the induction of anti-cancer immunity. Mechanistically, a significant reduction of immune-suppressive cells, including a decrease of tumor-promoting mast cells and tumor-associated macrophages was observed upon axitinib treatment. Inhibition of mast cells by axitinib as well as their experimental depletion led to reduced tumor growth. Of note, treatment with axitinib led to an improved T cell response, while the latter was pivotal for the therapeutic efficacy. Combination with immunecheckpoint inhibitors anti-PD-1 and anti-TIM-3 and/or agonistic engagement of the activating receptor CD137 resulted in a synergistic therapeutic efficacy. This demonstrates non-redundant immune activation induced by axitinib via modulation of myeloid and mast cells. These findings provide important mechanistic insights into axitinib-mediated anti-cancer immunity and provide rationale for clinical combinations of axitinib with different immunotherapeutic modalities.

Modeling tumor immunity of mouse glioblastoma by exhausted CD8+ T cells

T cell exhaustion occurs during chronic infection and cancers. Programmed cell death protein-1 (PD-1) is a major inhibitory checkpoint receptor involved in T cell exhaustion. Blocking antibodies (Abs) against PD-1 or its ligand, PD-L1, have been shown to reverse T cell exhaustion during chronic infection and cancers, leading to improved control of persistent antigen. However, modeling tumor-specific T cell responses in mouse has been difficult due to the lack of reagents to detect and phenotype tumor-specific immune responses. We developed a novel mouse glioma model expressing a viral epitope derived from lymphocytic choriomeningitis virus (LCMV), which allowed monitoring of tumor-specific CD8 T-cell responses. These CD8 T cells express high levels of PD-1 and are unable to reject tumors, but this can be reversed by anti-PD-1 treatment. These results suggest the efficacy of PD-1 blockade as a treatment for glioblastoma, an aggressive tumor that results in a uniformly lethal outcome. Importantly, this new syngeneic tumor model may also provide further opportunities to characterize anti-tumor T cell exhaustion and develop novel cancer immunotherapies.

Abstract 5666: Safety of the first-in-class anti-NKG2A monoclonal antibody monalizumab in combination with cetuximab: a phase Ib/II study in recurrent or metastatic squamous cell carcinoma of the head and neck (R/M SCCHN)

Abstract Background: Monalizumab (IPH2201) is a first-in-class immune checkpoint inhibitor targeting CD94-NKG2A receptors expressed on tumor infiltrating cytotoxic CD8 T lymphocytes and NK cells. HLA-E, the ligand of this inhibitory checkpoint receptor is up-regulated in SCCHN, protecting cancer from killing by CD94-NKG2A+ cells. Monalizumab blocks binding of CD94-NKG2A to HLA-E, reducing inhibitory signaling and thereby enhancing NK and T cell anti-tumor responses. Cetuximab is an anti-EGFR monoclonal antibody blocking oncogenic signaling and inducing Fcγ receptor-mediated antibody dependent cellular cytotoxicity (ADCC). In vitro cetuximab-mediated ADCC is inhibited by HLA-E expression on target cells and this inhibition can be circumvented with CD94-NKG2A blockade. Combination of monalizumab and cetuximab might provide greater antitumor activity than either drug alone. Methods: A multicenter, non-randomized dose-escalation and expansion study is evaluating monalizumab plus cetuximab in patients with R/M SCCHN (NCT02643550). Patients ≥ 18 years who progressed after platinum-based chemotherapy (regardless of the number of previous lines of treatment) were enrolled, without regard to HLA-E or human papilloma virus status. Using a 3+3 design, 5 dose levels of monalizumab (0.4, 1, 2, 4 or 10 mg/kg every 2 weeks) were explored with fixed doses of cetuximab (400 mg/m² load followed by 250 mg/m² weekly). Patients were treated until disease progression or unacceptable toxicity. The primary objective was to evaluate safety and Dose Limiting Toxicity (DLT). The secondary objectives were to estimate the Maximum Tolerated Dose (MTD) and the Recommended Phase II Dose (RP2D), and to determine the pharmacokinetics, pharmacodynamics and immunogenicity of the combination. Results: Enrolment began in December, 2015. As of October 11, 2016, 13 patients with R/M SCCHN were enrolled at dose levels 0.4, 1, 2 and 4 mg/kg. Median age was 60 years (range: 40-74); 92% were male; PS was 0 or 1; all patients had received prior systemic therapy (1 prior line in 1 patient and &amp;gt; 2 lines in 12 patients) for R/M SCCHN including platinum based chemotherapy (100% of the patients) and cetuximab (85%). There were no DLTs, infusion related reactions, immune related disorders or deaths related to treatment. No discontinuation attributable to treatment-related adverse events and no treatment-related grade 3 and 4 adverse events were reported, except fatigue (grade 3) in one patient. Updated data including pharmacodynamics and pharmacokinetics on the full dose escalation part will be presented. Conclusion: Monalizumab + cetuximab were well tolerated with no additional safety concerns compared to monalizumab or cetuximab alone. The dose-expansion phase of the study will be initiated in the near future. Citation Format: Roger B. Cohen, Sébastien Salas, Caroline Even, Nuria Kotecki, Antonio Jimeno, Anne-Marie Soulié, Anne Tirouvanziam-Martin, Robert Zerbib, Pascale André, Agnès Boyer-Chammard, Jérôme Fayette. Safety of the first-in-class anti-NKG2A monoclonal antibody monalizumab in combination with cetuximab: a phase Ib/II study in recurrent or metastatic squamous cell carcinoma of the head and neck (R/M SCCHN) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5666. doi:10.1158/1538-7445.AM2017-5666

Abstract 568: Regulatory T cell recruitment limits the effectiveness of checkpoint inhibition for claudin-low breast cancer

Abstract Introduction: Breast cancer can be separated into five intrinsic subtypes based on differences in the transcriptome of the tumor. We propose that the intrinsic differences of specific tumor subytpes lead to extrinsic differences in the tumor microenvironment. Methods: We utilized human clinical and genetically engineered mouse model (GEMM) samples of the intrinsic subtypes luminal A, basal-like, and claudin-low breast cancers to evaluate the immune landscape in the tumor microenvironment by histology and microarray analysis. Our claudin-low GEMM was derived from BRCA-/-/p53-/- mice. The HER-2 overexpressing, basal-like, and luminal A models have been previously descried. We utilized the FoxP3-DTR transgenic mouse model as a method of regulatory T cell (Treg) depletion to evaluate their function in these GEMMs. Results: The claudin-low human tumors were heavily infiltrated with immune cells, with CD4+ T cells being the most prominent, when compared to the luminal A subtype (P = 0.01). There were also increased focal areas of Tregs in human claudin-low tumors. To evaluate the mechanism for these findings, we utilized a GEMM of claudin-low tumors in addition HER-2 overexpressing, basal-like, and luminal A models. Mice with claudin-low tumors recruited elevated numbers of immune cells to the tumor microenvironment when compared to other breast cancer subtypes (P=0.01). Additionally, there was increased expression of multiple chemokine ligands in the tumor microenvironment among claudin-low tumors, with CXCL12 being the most highly overexpressed. Because the claudin-low tumors were heavily immune infiltrated, we hypothesized that blockade of the inhibitory checkpoint receptors programmed death-1 (PD-1) and cytotoxic T-lymphocyte associated antigen-4 (CTLA-4) could delay tumor growth and improve anti-tumor immune response. Surprisingly, we saw no delay in tumor growth in the claudin-low model using checkpoint inhibition. To investigate if presence of Tregs limited the function of checkpoint inhibitors, mice with claudin-low tumors were treated with AMD3100, a CXCR4 inhibitor. This decreased Treg infiltration into the tumor but did not alter tumor growth. We then utilized the FoxP3-DTR transgenic mouse model, where depletion of Tregs alone resulted in a very modest decrease in tumor growth, while depletion of Tregs plus checkpoint inhibition significantly improved survival (P = 0.03) and increased cytokine production by CD8+ T cells. Conclusion: We found that an effective anti-tumor immune response in claudin-low tumors is inhibited by the recruitment of Tregs to the tumor microenvironment. These data highlight early Treg recruitment as a possible mechanism for the lack of response to immune checkpoint inhibition therapy for claudin-low breast cancer. Note: This abstract was not presented at the meeting. Citation Format: Sarah C. Vick, Nicholas A. Taylor, Michael D. Iglesia, W June Brickey, Lisa A. Carey, Bentley R. Midkiff, Karen P. McKinnon, Shannon Reisdorf, Joel S. Parker, Charles M. Perou, Benjamin G. Vincent, Jonathan S. Serody. Regulatory T cell recruitment limits the effectiveness of checkpoint inhibition for claudin-low breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 568. doi:10.1158/1538-7445.AM2017-568

Immunotherapy in lymphoma.

Immunotherapy in lymphoma.

High-fidelity genome editing using NextGEN CRISPR (Clo51-dCas9) system for the production of allogeneic CAR-T cells.

3048 Background: Autologous chimeric antigen receptor (CAR) T therapies are highly efficient at targeting hematological malignancies, but the clinical applications have been limited by individualized manufacturing. Furthermore, there has been little success in treating solid tumors due to immunosuppressive microenvironments. Currently, genome editing technologies are being used to address both issues. However, the CRISPR/Cas9 system has significant safety concerns due to high incidence of off-target mutations and TALEN only works sufficiently in activated cells. A hybrid gene editing system, NextGEN (NG) Clo51-dCas9, can be targeted using gRNA, like CRISPR/Cas9, but exhibits little-to-no off-target cutting like TALEN, thereby overcoming limitations in the genome editing of resting T cells. Methods: We successfully developed a platform for production of allogeneic CAR-T cells with reduced receptivity to inhibitory signaling. Here, T cells were modified by piggyBac-mediated BCMA CAR gene delivery, along with NG reagents to knock out critical genes mediating rejection responses. Gene edited CAR-T cells were assessed by mixed lymphocyte reaction (MLR) and tumor killing. In addition, NG was used to knockout multiple checkpoint inhibitory receptors known to mediate key suppressive signals in T cells. Results: NG demonstrated high gene disruption efficiencies for all targets (84% for TCRα, 91% for TCRβ, 64% for β-2 microglobulin, and 40-60% for the surface inhibitory receptors PD-1, CTLA-4, Tim3, Lag-3, and TGFBRII). In contrast to CRISPR/Cas9, no off-target mutations were detected for multiple targets by deep-sequencing. In MLRs, disruption of TCR eliminated the GvHD response, while disruption of MHCI completely abrogated graft-rejection. Lastly, TCR/MHCI double knockout did not affect the ability to kill BCMA+ multiple myeloma cells in vitro. Conclusions: NG overcomes significant limitations of the CRISPR/Cas9 and TALEN systems with highly efficient genome editing in resting T cells. NG has great potential and flexibility for the manufacture of allogeneic CAR-T cells and for enhancing efficacy against solid tumors.

Adaptive resistance to anti-PD1 therapy by Tim-3 upregulation is mediated by the PI3K-Akt pathway in head and neck cancer

ABSTRACTProgrammed Death 1 (PD-1) and T cell Ig and mucin domain-3 protein (Tim-3) are immune checkpoint receptors that are expressed on tumor-infiltrating lymphocytes (TIL) in tumor-bearing mice and humans. As anti-PD-1 single agent response rates are only <20% in head and neck squamous cell carcinoma (HNSCC) patients, it is important to understand how multiple inhibitory checkpoint receptors maintain suppressed cellular immunity. One such receptor, Tim-3, activates downstream proliferative pathways through Akt/S6, and is highly expressed in dysfunctional TIL. We observed that PD-1 and Tim-3 co-expression was associated with a more exhausted phenotype, with the highest PD-1 levels on TIL co-expressing Tim-3. Dampened Akt/S6 phosphorylation in these PD-1+Tim-3+ TIL, when the PD-1 pathway was ligated, suggested that signaling cross-talk could lead to escape through Tim-3 expression. Indeed, PD-1 blockade of human HNSCC TIL led to further Tim-3 upregulation, supporting a circuit of compensatory signaling and potentially permitting escape from anti-PD-1 blockade in the tumor microenvironment. Also, in a murine HNC tumor model that is partially responsive to anti-PD-1 therapy, Tim-3 was upregulated in TIL from persistently growing tumors. Significant antitumor activity was observed after sequential addition of anti-Tim-3 mAb to overcome adaptive resistance to anti-PD-1 mAb. This increased Tim-3-mediated escape of exhausted TIL from PD-1 inhibition that was mediated by phospho-inositol-3 kinase (PI3K)/Akt complex downstream of TCR signaling but not cytokine-mediated pathways. Taken together, we conclude that during PD-1 blockade, TIL upregulate Tim-3 in a PI3K/Akt-dependent manner, providing further support for dual targeting of these molecules for more effective cancer immunotherapy.

Abstract B129: Multiple immune checkpoint receptors are co-expressed on tumor antigen-specific T cells and contribute to tumor immune evasion

Abstract Immune checkpoint blockade has revolutionized cancer immunotherapy, and combination treatment with antibodies against multiple inhibitory receptors promises to improve efficacy substantially. However, the spatiotemporal and combinatorial expression profile of various checkpoint receptors on effector T cells and the effect of combined blockade on immune function remain poorly defined. In this study we set out to determine the co-expression kinetics and functional impact of multiple checkpoint receptors (PD-1, LAG-3, TIM-3, and BTLA) on tumor antigen-specific effector T cells in controlled in vitro culture systems and in the tumor environment. To elicit tumor antigen-specific responses we immunized mice with irradiated MC38 tumor cells engineered to express chicken ovalbumin (OVA; MC38.OVA) and used pentamer reagents to identify OVA257-264-specific CD8 T cells. Repeated stimulation with irradiated MC38.OVA cells or irradiated OVA-pulsed splenocytes ex vivo resulted in robust proliferation and activation of antigen-specific T cells as determined by BrdU incorporation and production of IFN-γ. Most OVA-specific cells expressed PD-1 ex vivo and maintained PD-1 expression during repeated cycles of in vitro stimulation. In contrast, antigen-specific T cells did not express TIM-3 or LAG-3 ex vivo, yet both receptors were induced on the majority of PD1 positive, but not PD-1 negative, cells upon repeated stimulation ex vivo, resulting in double and triple positive cells. The frequency of BTLA expressing cells remained low under all conditions. The addition of PD-1 and TIM-3 blocking antibodies during ex vivo restimulation enhanced T cell activation, demonstrating that these receptors mediate inhibitory function. Consistent with our in vitro data demonstrating that repeated antigen stimulation promotes the coordinated expression of checkpoint receptors, we found that the vast majority of MC38.OVA or Colon26 tumor infiltrating T cells (TILs) express PD-1, and most PD-1 positive cells also co-express TIM-3 and LAG-3, whereas these receptors are largely absent from T cells in secondary lymphoid organs. Importantly, combination treatment of tumor-bearing mice with anti-PD-1 and anti-TIM-3, or anti-PD-1 and anti-LAG-3 antibodies enhanced the immune control of tumor growth in both models as compared to the respective monotherapies. Finally, in vitro activation of human PBMC-derived T cells resulted in a similar sequential upregulation of checkpoint inhibitors ultimately resulting in their co-expression. Here, we have established a culture system to delineate the hierarchical expression of inhibitor checkpoint receptors on tumor antigen-specific murine T cells during repeated antigen stimulation. Defining the inhibitory checkpoint receptor landscape will guide the process of identifying the most promising combinations of checkpoint blockers in cancer immunotherapy. Citation Format: Robert J. Durso, Jerry Pei, Michelle Russell, Pratha Budhani, Elena Burova, Chandrika Taduriyasas, Ella Ioffe, Markus Mohrs, Gavin Thurston, Jie Dai. Multiple immune checkpoint receptors are co-expressed on tumor antigen-specific T cells and contribute to tumor immune evasion [abstract]. In: Proceedings of the Second CRI-CIMT-EATI-AACR International Cancer Immunotherapy Conference: Translating Science into Survival; 2016 Sept 25-28; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2016;4(11 Suppl):Abstract nr B129.

Abstract 4364: Adenosine A2a receptor blockade as a means of enhancing immune checkpoint inhibition and adoptive T-cell therapy

Abstract Adenosine signaling through the A2a receptor (A2aR) has pleiotropic immunosuppressive effects on a broad range of immune cells. Extracellular adenosine levels, which are typically very low in normal tissues, are elevated in the tumor microenvironment, allowing increased signaling through the A2aR. Previously, our lab and others have shown increased rejection of tumors in A2aR-null mice, demonstrating that adenosine signaling through A2aR limits immunologic response to tumors. Present work in our lab demonstrates the ability of a potent and specific, orally administered A2aR antagonist, CPI-444, to enhance immunologic response in mice. First, A2aR blockade with CPI-444 (1 mg/kg) in the peri-vaccination period led to significant expansion of antigen-specific T cells over untreated controls. Interestingly, antigen-specific CTLs from CPI-444 treated mice showed significantly lower expression of inhibitory checkpoint receptors, including PD-1 and TIM-3. Second, in an MC38 tumor model, daily treatment with CPI-444 (100 mg/kg) led to tumor growth suppression and survival benefit compared with vehicle-treated controls. Notably, tumor infiltrating regulatory T cells had significantly lower CTLA-4 and FoxP3 expression in CPI-444 treated mice. These initial immunologic findings of enhanced early CTL expansion and suppression of inhibitory co-signaling have important implications for applying A2aR blockade to immunotherapy in cancer. First, given the ability of CPI-444 to augment CTL expansion, we investigated its application to a therapeutic adoptive T cell transfer system using OVA-expressing B16 melanoma cell line. In mice with established tumors, daily CPI-444 (100 mg/kg) treatment initiated on the day of adoptive transfer of activated OVA-specific OT-I T cells led to suppressed tumor growth and increased survival versus vehicle treated controls. Second, in light of our data demonstrating its ability to suppress PD-1 in CTLs, we theorized that CPI-444 may lower the threshold for effective anti-PD-1 therapy and achieve synergistic effects as combination therapy. Indeed, daily CPI-444 (100 mg/kg) treatment combined with anti-PD-1 treatment showed robust antitumor effect in the CT26 tumor model, with complete tumor rejection in 7/9 mice (vs. 2/10 mice in the anti-PD-1 alone group). Survival was markedly improved in mice receiving combination treatment versus either CPI-444 or anti-PD-1 alone. In as much as increased adenosine in the tumor microenvironment is partly a consequence of dysregulated cancer cell metabolism, we hypothesized that targeting tumor metabolism would enhance the efficacy of A2aR antagonism. To this end, inhibition of tumor glutamine metabolism potently augmented the efficacy of CPI-444 in a CT26 model. Overall, our data suggest that A2aR antagonism can enhance the efficacy of immunotherapy in the form of checkpoint blockade and adoptive cellular therapy. Citation Format: Robert D. Leone, Judson M. Englert, Chih-Hsien Cheng, Jiayu Wen, Min-Hee Oh, Im-Hong Sun, Chirag Patel, Ian A. Bettencourt, Jonathan D. Powell. Adenosine A2a receptor blockade as a means of enhancing immune checkpoint inhibition and adoptive T-cell therapy. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 4364.

Effects of anti-NKG2A antibody administration on leukemia and normal hematopoietic cells.

Natural killer cells are key cells of the innate immune system. Natural killer cell receptor repertoires are diversified by a stochastic expression of killer-cell-immunoglobulin-like receptors and lectin-like receptors such as NKG2 receptors. All individuals harbor a subset of natural killer cells expressing NKG2A, the inhibitory checkpoint receptor for HLA-E. Most neoplastic and normal hematopoietic cells express HLA-E, the inhibitory ligand of NKG2A. A novel anti-human NKG2A antibody induced tumor cell death, suggesting that the antibody could be useful in the treatment of cancers expressing HLA-E. We found that immunodeficient mice, co-infused with human primary leukemia or Epstein-Barr virus cell lines and NKG2A(+) natural killer cells, pre-treated with anti-human NKG2A, were rescued from disease progression. Human NKG2A(+) natural killer cells reconstituted in immunodeficient mice after transplantation of human CD34(+) cells. These natural killer cells are able to kill engrafted human primary leukemia or Epstein-Barr virus cell lines by lysis after intraperitoneal administration of anti-human NKG2A. Thus, this anti-NKG2A may exploit the anti-leukemic action of the wave of NKG2A(+) natural killer cells recovering after hematopoietic stem cell transplants or adoptive therapy with natural killer cell infusions from matched or mismatched family donors after chemotherapy for acute leukemia, without the need to search for a natural killer cell alloreactive donor.

PD-1 blockade upregulate TIM-3 expression as a compensatory regulation of immune check point receptors in HNSCC TIL

Programmed Death 1 (PD-1) and T cell Ig and mucin domain-3 protein (Tim-3) are immune check point receptors that are upregulated on tumor infiltrating lymphocytes (TIL) in tumor-bearing mice and humans. As anti-PD-1 single agent response rates are still relatively low (20%) in HNC patients, it is important to learn how different inhibitory check point receptors work together to maintain the suppressive status of immune system. We observed that PD-1 and Tim-3 co-expression is associated with an exhausted phenotype of HNSCC TIL of patients, demonstrating the highest PD-1 levels in Tim-3 double positive TIL. We also observed that PD-1+/Tim-3+ TIL manifest dampened in Akt/p-S6 activation upon TCR stimulation, leading us to infer the potential for signaling cross-talk between PD-1 and Tim-3 downstream signaling pathways. Indeed, in freshly isolated HNSCC TIL, PD-1-/TIM-3+ T cells showed higher baseline expression of p-SHP-2 (p < 0.01) which is triggered upon PD-1 ligation. In addition, PD-1 blockade using nivolumab of human HNSCC TIL led to upregulation of Tim-3 expression, suggesting a circuit of compensatory, cross-talk signaling and permitting escape from anti-PD-1 blockade during TCR stimulation. In a murine HNC tumor model, anti-PD-1 treatment modestly suppressed tumor growth, but in TIL from persistent tumors in these mice, Tim-3 expression was dramatically upregulated after PD-1 blockade. Taken together, in response to PD-1 blockade, the most exhausted TIL appear to upregulate Tim-3 as a compensatory mechanism, supporting dual targeting, which may provide new therapeutic strategy for cancer immunotherapy. Figure 1 Figure 2 Figure 3 Figure 4