PROJECT SUMMARY Androgens are well known to be immunosuppressive, yet cancer immunotherapy is used without consideration of local androgen concentrations or the sex of the patient. We present evidence that targeting the androgen receptor is necessary for effective T cell-specific immunotherapy in mouse models of prostate cancer and sarcoma and in a human clinical trial of metastatic prostate cancer. In patient biopsies, we observed a striking correlation between low androgen receptor activity in CD8 T cells and response to PD-1 targeted immunotherapy. We validated our observations in a novel mouse model of castration resistant prostate cancer and demonstrated a requirement for androgen receptor inhibition to permit effective PD-L1 immunotherapy. Moreover, tumor regression required CD8 T cells suggesting T cell intrinsic AR activity could regulate response to immunotherapy. Extending these studies, we isolated activated CD8 T cells and observed a strong association of androgen receptor with chromatin modifiers. Performing in silico experiments revealed canonical androgen response elements in open chromatin regions of the Ifng and Gzmb loci. We observed binding of the androgen receptor within these regions and importantly, the ability to inhibit this interaction with a small molecule inhibitor. Inhibiting this interaction increased the functional capacity of dysfunctional CD8 T cells; a functional state further amplified by PD-L1 blockade. Taken together, these observations suggest that androgen receptor signaling could enforce the dysfunctional chromatin state of T cells limiting reinvigoration by checkpoint blockade. To the best of our knowledge this is the first experimental model to establish a direct role for androgen receptor activity on regulating responsiveness to immunotherapy. The goal of this proposal is to understand the mechanism(s) by which androgen receptor activity in CD8 T cells limits the effectiveness of cancer immunotherapy. We hypothesize that androgen receptor activity in CD8 T cells limits functional re-invigoration to immune checkpoint blockade by transcriptional and epigenetic repression of effector genes. We will test this hypothesis in the following Aims. 1) Evaluate whether AR activity represses T cell effector genes by direct DNA binding; 2) Determine whether AR signaling enforces a repressive chromatin state in dysfunctional CD8 T cells; and 3) Determine whether loss of T cell intrinsic AR is sufficient to restore responsiveness to checkpoint blockade. Together, these studies provide a framework for understanding hormone mediated resistance to cancer immunotherapy. Importantly, we will gain a critical understanding of how androgen could limit effective immunotherapy in prostate cancer patients and anticipate that our observations will be applicable beyond this disease to improve immunotherapy outcomes in cancers where androgens are present.