Mechanisms of Clearance of MHC I-deficient Tumors Induced by STING Agonists

Immunotherapies, such as “checkpoint blockade” have revolutionized modern cancer treatment, greatly increasing patient survival and leading to cures in a significant number of people. These groundbreaking drugs target inhibitory receptors on CD8 T cells, increasing T cell activation and promoting tumor destruction. However, tumors may evade such therapies via loss of MHC molecules or because they contain few/no neoantigens. Therefore, approaches to mobilize immune cells to kill CD8 T cell-resistant tumors are needed to combat these potential escape mechanisms. Natural killer (NK) cells recognize stress-induced ligands on tumor cells and do not require that tumors display neoantigens. CD4 T cells recognize epitopes presented by MHC II molecules that can serve as alternative targets for anti-tumor responses and are capable of activating other immune cells within the tumor microenvironment to kill tumor cells. Therefore, immunotherapies that mobilize NK cells or CD4 T cells have potential for targeting tumors resistant to CD8 T cells.Cyclic-dinucleotides (CDNs) are a class of immunostimulatory molecules that bind to the STING protein and induce type I interferons (IFN) and other pro-inflammatory mediators. Recently, in mouse transplantable tumor models, it was shown that intratumoral CDN injections stimulate powerful CD8 T cell-mediated tumor rejection, often leading to long-lasting tumor regression and protection from rechallenge. Compounds that activate the STING pathway are currently being tested in clinical trials for treating cancer.My thesis work examines the potential of CDNs to mobilize anti-tumor immune responses against MHC I-deficient tumors. The data show that CDNs trigger potent tumor rejection in several different tumor models, independent of CD8 T cells. The antitumor effects are dependent on NK and in some cases CD4 T cells, which can mediate rejection independently of each other. CDNs enhanced NK cell activation, cytotoxicity, and antitumor effects in part by inducing type I IFN (IFN). IFN acted in part directly on NK cells in vivo, and in part indirectly via dendritic cells. Upon applying CDNs in vivo, dendritic cells (DCs) upregulated IL-15R in an IFN-dependent manner, and IL-15 action was important for CDN-induced NK activation and tumor control. Mice lacking IFNAR specifically on DCs had reduced NK cell activation and tumor control.CD4 T cells also mediate potent antitumor responses in some tumor models, independently of CD8 T cells and, in the primary response, independently of NK cells, B cells, and T cells. CDN treatments led to increased tumor-specific priming of CD4 T cells and enhanced effector functions, such as production of IFN-, IL-2, and TNF. Tumor-specific CD4 T cells in tumors treated with CDNs had a less exhausted, Th1-like phenotype, with increased production of IFN-, which was necessary for the antitumor response. Mice that cleared their primary tumors exhibited a long-lasting antitumor memory response, which was dependent on CD4 T cells, IFN- and partially dependent on B cells, myeloid cells and, albeit to a minor extent, NK cells. Interestingly, the antitumor response did not rely on MHC II expression by the tumor cells, suggesting that CD4 T cells either initiate antitumor effects indirectly, without direct recognition of the tumor cell by CD4 T cells, or engage ligands other than conventional MHC II (and MHC I) on tumor cells. In the final chapter of the thesis I described several genetic screens in cell lines aimed at identifying novel cellular ligands for NK cell activating receptors. The screens employed chimeric antigen receptor (CAR) T cells that incorporate NK cell activating receptors (NK CAR T cells) as selecting agents against human tumor cell lines mutagenized with a retroviral gene-trap or stably expressing CRISPR-Cas9 and a genome-wide gRNA library. I successfully generated multiple NK CAR T cells and performed several screens using NKp44 and NKp46 CAR T cells. These screens resulted in lists of genes, that when mutated, were enriched in the surviving cellular population. Future work will be needed to validate the hits from these screens and perform additional screens if necessary.This thesis aims to identify mechanisms of immune cell activation induced by STING agonists as well as to unearth novel mechanisms of NK cell tumor recognition. This work is significant because it addresses CD8 T cell-independent immunological rejection of MHC-deficient tumors induced by CDNs and mediated by NK and CD4 T cells. A greater understanding of mechanisms leading to immunological clearance of MHC-deficient tumors will be necessary to develop approaches to overcome mechanisms tumors employ to escape CD8 T cell control and to design rational combination therapies with existing anti-cancer drugs. Therapeutics that activate NK cells and CD4 T cells may represent next-generation approaches to cancer immunotherapy