ABSTRACT The disparity in Prostate Cancer (PCa) mortality between African American (AA) and European American (EA) men is large, with age-adjusted mortality of 62.3 per 100,000 in AA and 25.6 per 100,000 in EA. This represents a 2.4-fold greater chance of death from prostate cancer in AA than EA, which is the largest disparity in cancer mortality of any tumor site in US men or women. Although PCa is the most common non-cutaneous tumor diagnosed in US men, our ability to predict which PCa cases will have an unfavorable outcome is limited. Despite the use of PCa screening for early detection of PCa, mass screening of higher and lower risk men results in unnecessary treatment for some and insufficient treatment in others. Therefore, it is important to identify individuals who are most likely to have an adverse PCa outcome. If this information is specifically addressed to AA populations, PCa disparities may also be ameliorated. Most clinical trials resulting in FDA approved PCa therapies exlude men with comorbidities or polypharmacy, and general do not include large samples of AA men. Becasuse AA men tend to have higher rates of comorbidities than EA men, it is not clear that therapeutic interventions are optimized to treat AA men with PCa. We hypothesize that PCa can be better managed if knowledge about clinical characteristics, comorbid conditions, and novel biomarkers are considered when applying PCa treatment. We propose to address the following specific aims using a large sample of AA PCa cases available through the Men of African Descent and Carcinoma of the Prostate (MADCaP) Networl: Aim 1) Biomarkers: Identify biomarkers that improve prediction of prostate cancer outcomes in AA men; Aim 2) Comorbidities: Evaluate the effects of comorbidities on clinical outcomes in AA men; Aim 3) Treatment Adaptation Models: Incorporate biomarkers, contextual variables, comorbidities, and other factors to evaluate how existing treatment strategies can be improved by predicting unfavorable clinical outcomes (overall mortality, biochemical recurrence, short PSA doubling time, prostate cancer specific mortality) in AA men. We will use PCa cases with detailed clinical annotation and biosamples that have been collected prospectively for more than 20 years to test the performance of existing prediction tools and newly developed algorithms among AAs and EAs. These aims will address the critical need to understand which AA men will suffer disproportionately after a PCa diagnosis, an important step to improve the management of PCa and quality of life for AA men.