Background: Depressive disorder is a common cause of morbidity among men with prostate cancer (PC) and may contribute to known racial disparities in PC outcomes. We estimated the incidence, management, and impact of depressive disorder (depression) on overall mortality among African American (AA) and non-Hispanic White (W) Veterans with localized PC. Methods: In this retrospective cohort study, we used linked administrative, survey, and electronic health record data from the Veterans Health Administration (VA) Corporate Data Warehouse to identify AA and W Veterans with no preexisting depression who were diagnosed with localized PC between January 1, 2004 and December 31, 2013. Patients were followed through December 31, 2019. The primary outcomes were incident depression (defined from diagnosis codes and PHQ-2 and -9 screenings between six months to five years after PC diagnosis), receipt of anti-depressant therapy, and all-cause mortality. We used logistic and Cox regression models, adjusted for sociodemographic factors, PSA, Gleason score, and prostate cancer treatment, to estimate associations with all outcomes, using race-by-depression and race-by-treatment interaction terms to investigate racial disparities. Results: Among 32,194 Veterans diagnosed with localized prostate cancer (median age 67 years [interquartile range [IQR] 62 to 73 years], median follow-up 9.9 years [IQR 8.0 to 12.1 years]), 8,177 (25.4%) were AA. Overall, 8,285 (25.7%) Veterans were diagnosed with depression after PC diagnosis, and 2,525 (30.5%) of depressed Veterans received an antidepressant. Compared to Veterans without depression, Veterans with incident depression had higher all-cause mortality (adjusted hazard ratio [aHR] 1.30 [95% CI 1.25-1.35]). Race moderated all outcomes: AAs were more likely than Ws to be diagnosed with depression. However, among those with depression, AAs were less likely than Ws to receive an antidepressant. Interaction analyses showed that the HR of all-cause mortality associated with depression among AAs was significantly greater than that of Ws. Antidepressant receipt was not associated with improved mortality (aHR 1.05 [95% CI 0.97-1.13]); this finding was not moderated by race. Conclusions: Depression was common among men with prostate cancer within a large equal-access health care system, and African American men had more adverse depression-related outcomes than White men. Identifying and managing incident depression should be a key target of efforts to improve prostate cancer outcomes and disparities.