Prostate cancer patients acquire resistance to standard-of-care strategies progressing to advanced disease and resulting in 350,000 yearly deaths. As acquired resistance is mediated by increased androgen receptor (AR) expression, we are developing a “Bipolar Androgen” therapy (BAT) cycling serum testosterone from supraphysiological to near-castrate levels, maximizing toxicity to high and low AR-expressing cells respectively. BAT is a clinically effective, safe and unique approach to treat castration-resistant prostate cancer (CRPC) patients that improves quality of life, produces biochemical and objective responses, and re-sensitizes tumors to AR inhibitors. Our data from a recent clinical trial (NCT03554317) shows that prostate tumor cells produce inflammatory cytokines following BAT, and patients who benefited most from this therapy have an enriched inflammatory transcriptional signature in tumors. Thus, despite its conception as a “targeted” therapy, our understanding of BAT has broadened to consider its effects on the immune system, which appears critical for success. To capitalize on this unappreciated potential and bridge the gap between patients who benefitted or not from this novel strategy, we characterized patient peripheral blood mononuclear cells (PBMCs) before and after treatment with BAT to define the changes it induces in immune cells. Our observations indicate that BAT elicits the expansion of monocytic-myeloid derived suppressor cells (M-MDSCs) in patients with poor therapy responses. The association between the abundance of M-MDSCs and reduced therapeutic efficacy, the precedent in the literature that testosterone dampens the pro-inflammatory phenotype of macrophages, plus the critical role of the inflammatory response in controlling tumor growth following BAT, lead us to the hypothesis that increased numbers of M-MDSCs induced by BAT restrict antitumor immunity leading to reduced therapeutic efficacy. To test this hypothesis, we propose three aims: (1) we will characterize PBMCs from metastatic CRPC patients treated with BAT utilizing next generation sequencing, to identify transcriptional signatures in M-MDSCs that are associated with therapeutic response and model M-MDSC tumor engraftment as these cells contribute to the immunosuppressive tumor microenvironment; (2) we will use high dimensional flow cytometry to demonstrate that immunosuppressive M-MDSCs are induced by BAT and not by standard-of-care therapy in an independent cohort of patients, and describe the mechanism behind this induction by exposing isolated human PBMCs in vitro to supraphysiological levels of testosterone; (3) we will utilize mouse models of prostate cancer to assess the therapeutic benefit of M-MDSC depletion in combination with BAT. This collaborative effort across the disciplines of computational biology, oncology, and myeloid cell biology, will build a detailed understanding of how BAT reprograms tumor immunity and determine if M-MDSCs expansion un...