Prostate cancer (PCa) remains an incurable disease once progression to the metastatic castration-resistant (mCRPC) state occurs; killing >30,000 U.S. men/yr. Unfortunately, each of the FDA-approved agents for mCRPC produces only modest increases in overall survival followed by the emergence of a resistant and more aggressive phenotype. Thus, there urgent unmet need for innovative therapies with novel mechanisms of action that allow discrimination between normal and PCa cells. Herein, we put forward an innovative strategy to selectively exploit selective vulnerabilities in neuroendocrine prostate cancer (NEPC), a highly aggressive variant of mCRPC with poor prognosis, using an orally-dosed and highly specific lysine-specific demethylase 1 (LSD1) inhibitor, bomedemstat, in clinical development for unrelated indications to maximize near-term patient benefit. The prevalence of NEPC is increasing in mCRPC patients following prolonged treatment with potent androgen receptor (AR) axis-targeted therapies (e.g., abiraterone, enzalutamide) as an adaptive mechanism of resistance. Comprehensive genome-scale analyses integrated with pharmacogenomics pipelines will provide a rich mechanistic understanding of the genetic and epigenetic features underlying the sensitivity differential between NEPC and AR+ prostate cancer (ARPC) subtypes to LSD1 inhibition (LSD1i). It should be noted that although NEPC is more sensitive, LSD1i also has significant activity against ARPC tumors, which has important clinical implications for the growing number of advanced mCRPC patients with heterogeneous lesions encompassing mixed molecular and pathological subtypes. LSD1-depedent vulnerabilities are identified that will guide patient selection for a precision medicine approach. Additionally, combination therapy targeting LSD1-induced synthetic lethal susceptibilities are rationally designed to provide more potent and durable responses to extend the clinical utility of these well-tolerated clinically-tested agents with favorable safety and pharmacokinetic profiles. Our proposed work has both conceptual and technical innovations. We will 1) validate LSD1 as a target in NE+ and AR+ tumors via comprehensive mechanistic interrogation using unique patient-derived xenograft and organoid (PDX/O) models; 2) compare responders vs. non-responders across subtypes to inform patient selection in future biomarker-driven clinical trials using state-of-the-art genome-wide profiling technologies; 3) dissect the functional effects of LSD1i in advanced preclinical model systems using these same advanced profiling strategies; 4) identify rational combination therapies based on LSD1i-induced synthetic lethal vulnerabilities to produce durable responses using a high-throughput PDO screening platform; and 5) use clinical-grade drugs to maximize near-term therapeutic potential against this highly aggressive lethal mCRPC variant emerging as a resistance mechanism to standard-of-care therapies with increa...