DESCRIPTION (provided by applicant): There are several critical unmet needs in the management of localized prostate cancer. Central among them is the development of minimally invasive tools to distinguish localized cancers that are truly indolent from cancers that are progressive and potentially lethal. To address this key need, we first propose to perform an integrated, multi-dimensional genomic, epigenomic and expression analysis to uncover novel molecular pathways that characterize indolent vs. aggressive prostate cancers. In this approach we define indolent tumors as those screen detected (e.g. PSA screening) lesions that are Gleason score 6 (or less) that are organ confined at radical prostatectomy. We consider these tumors indolent as they do not appear capable of metastasis. In contrast, we equate Gleason score 8-10 tumors as "interval" or symptomatic since, even with primary treatment, these tumors often recur and metastasize at high frequencies. Additionally, we will validate our key markers/pathways discovered in this project using additional populations with long term outcomes. We hypothesize that our multi-modality genomic-based integrated approach, contrasting these two divergent tumor types, will reveal signatures that distinguish cancers with dichotomous phenotypes. We also hypothesize that these signatures will vary based on race and thus in parallel we will comprehensively characterize African American prostate cancers to reveal molecular features driving racial disparities in outcomes. We will validate the signatures obtained using large cohorts of cases with established outcomes including: (1) the Johns Hopkins Active surveillance cohort and (2) Prostate cancer cases from the BLSA (Baltimore Longitudinal Study of Aging), an observational cohort of men followed since 1954 with autopsy documented indolent or aggressive/lethal disease. We also propose that these signatures will be able to predict outcomes of cancers with indeterminate kinetics and propose to test this through analysis of cases of intermediate risk prostate cancer with long-term follow-up and known outcomes from Johns Hopkins and in collaboration with colleagues from Harvard, from the Physician's Health and Health Professionals follow-up studies. Together this work will yield highly relevant information that can be directly applied to the clinical management of localized prostate cancer. Specifically, it will yield an integrated signature that distinguishes localized - indolent tumors from localized tumors with lethal potential. Additionally we believe these signatures will be critical in determining treatment strategies for individuals with prostate cancers of indeterminate kinetics.