Abstract Genetic testing, as a key component of precision medicine, may reduce prostate cancer treatment-related disparities, but only if widely disseminated outside of tertiary cancer care settings. This study sought to characterize predictors of germline genetic testing referrals and use among patients diagnosed with prostate cancer at a safety-net hospital using electronic medical record data. Men who had a confirmed diagnosis of prostate cancer between January 1, 2011 and March 10, 2020 were identified via the tumor registry. Using a centralized clinical data warehouse, we collected data on age, race, ethnicity, primary language, marital status, clinical stage, and insurance. The primary outcome was receipt of a referral to genetic counseling. We hypothesized that men who were foreign-born, non-English speaking, identified as Black race or Hispanic ethnicity, and were older would be less likely to be referred for genetic testing. A secondary outcome was the completion of genetic testing. Descriptive statistics (means, standard deviations, frequencies) described the cohort. In multivariable analyses, logistic regression estimated odds ratios (OR) and 95% confidence intervals (95% CI) for factors hypothesized to influence referral to genetic testing: age, race (Black, White, Asian, Other), ethnicity (Hispanic vs. non-Hispanic), language (English vs. non-English), country of origin (US vs. Other), insurance (Medicare, Medicaid, Private, Other), and clinical cancer stage (Local, Regional, Metastatic). Overall, 1,877 patients were diagnosed with prostate cancer in the study period. The mean age was 65 years (SD=9). 44% identified as Black race, 32% White. Ethnic composition was 17% Hispanic, 80% Non-Hispanic. Almost half (49%) were married, 46% were foreign born, and 34% had Medicaid insurance at diagnosis. Two-thirds (67%) spoke English, and 33% were non-English speakers. The majority (65%) had local-only disease at diagnosis, 3% had regional disease, 9% metastatic, and 22% had missing clinical stage data. For the primary outcome, 163 (9%) of all patients received at least one genetics referral. In multivariate models, we found that those who were older (OR=0.95, 95% CI: 0.93, 0.98) and identified as White race (OR=0.60, 95% CI: 0.38, 0.96) had lower odds of receiving a referral. Those with regional or metastatic disease at diagnosis were significantly more likely to receive referral, as expected (OR= 4.45 and 4.78, respectively). No other demographics significantly predicted referral. Of the 163 men referred to genetics, 136 (83%) had at least one genetics encounter and 19 (14%) had genetic testing. In sum, few patients received referrals for genetic counseling and testing from 2010-2021, with 80% occurring post-guideline changes in 2018. When referrals were made, our sample had high rates of genetics encounters, although lower rates of testing completion. Low rates of referral and testing indicate opportunities to improve both identification of eligible patients and resolve barriers to completing genetic testing post-encounter. Citation Format: Christine M. Gunn, Gretchen Gignac, Magdalena Pankowska, Kimberly Zayhowski, Catharine Wang. Characterizing access to genetics referrals for prostate cancer in a safety net hospital [abstract]. In: Proceedings of the AACR Virtual Conference: 14th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2021 Oct 6-8. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2022;31(1 Suppl):Abstract nr PO-050.