5062 Background: Guideline recommended genetic testing in advanced prostate cancer (PC) is underutilized and not always accepted by patients (pts). We assessed attitudes and decisional conflict (DC) surrounding genetic testing associated with subsequent test completion, and differences between white and nonwhite pts. Methods: Eligibility for this prospective single institution study included pts with N1 or M1 PC who had not yet completed genetic testing. Upon informed consent, pts were given a 24-question survey using a Likert scale of 0 (strongly agree) to 4 (strongly disagree) to assess attitudes toward genetic testing including a validated assessment of DC. DC and DC subscores (range 0-100, with 100 being highest DC) were calculated from subsets of survey responses. Self-identified race was obtained from the EMR. Two-group comparisons between white and nonwhite pts, and between those who completed genetic testing and who did not, were conducted with SAS v9.4 software using Fisher’s exact test for categorical variables and Wilcoxon rank sum test for Likert scale survey questions and DC score variables. Results: Of 42 enrolled pts (21 white, 17 black, 1 Asian, 3 declined), 22 (52.4%) completed genetic testing. Compared to white pts, nonwhite pts expressed more concern about test result privacy (mean = 1.72 v 2.95, p = 0.002), test results being used for non-healthcare purposes (1.78 v 3.00, p = 0.003), and trying unproven treatments (1.72 v 2.67, p = 0.01). Nonwhite pts felt more external pressure in decision-making compared to white pts (0.67 v 0.29, p = 0.04). No significant differences were appreciated in completion of testing, DC, or any subscore between racial groups. Compared to pts who did not complete testing, those who completed testing were more likely to report they knew which options were available (0.73. v 1.25, p = 0.05), knew the benefits of each option (0.77 v 1.30, p = 0.04), knew the risk and side effects of each option (0.95 v 1.50, p = 0.05), were clear about which benefits matter most to themselves (0.73 v 1.37, p = 0.02), and were clear about the best choice for themselves (0.73 v 1.35, p = 0.02). DC (28.59 v 18.11, p = 0.03) was higher in pts who did not complete testing, along with uncertainty (31.25 v 19.32, p = 0.02) and informed (31.25 v 20.45, p = 0.03) subscores. No differences were seen in values clarity, support, or effective decision subscores. Conclusions: In our study, nonwhite pts expressed greater concern about privacy, data misuse, and trying unproven treatments. Those who did not complete testing had more DC with greater uncertainty about knowledge and decision making. These findings will help direct targeted interventions to increase knowledge, trust, and decisional certainty about genetic testing in pts with advanced PC. Ongoing studies will assess the impact of these interventions on rates of testing completion at our institution.