African-American (AA) men exhibit one of the highest risks for prostate cancer (PC). They present with later-stage disease and have worse outcomes than Caucasian (CA) men. The genomics of AA PCs has been dramatically understudied compared to CA PCs. It is unclear whether oncogenic pathways drive PC in a race- sensitive manner, and how such pathways might be targeted. We recently developed and reported on a novel algorithm for classification of PC that we believe represents a significant advance over previous approaches. We found that by analyzing transcriptome data alone, most PCs can be assigned to one of only three novel subtypes, named PCS1 (luminal), PCS2 (luminal), and PCS3 (basal). As validated in over 10 independent cohorts, the PCS1 subtype demonstrates the worst clinical outcome, including in tumors with low Gleason score. The PCS1 and PCS3 subtypes are over-represented in castration-resistant PCs (CRPC). Genomics, transcriptomics, and immunohistochemical studies of AA PCs are consistent with the hypothesis that the PCS1 and PCS3 subtypes are over-represented in PCs arising in men of African ancestry. We have used this novel system, in combination with other computational and laboratory approaches, to identify a developmental transcription factor, ONECUT2 (OC2), which appears to operate as a ?master regulator? in a subset of CRPCs. This protein appears to be most active in the PCS1 and PCS3 subtypes, and therefore may be a critical mediator of progression in tumors seen in AA men. OC2 appears targetable with a small molecule and therefore may be a viable drug target in the context of aggressive PCs in AA men. We will test two hypotheses in this study: Hypothesis 1) The PCS system can identify distinct molecular features of PCs from AA men that can be used clinically to aid risk stratification and to identify actionable targets relevant to AA men; and Hypothesis 2) OC2 is a viable drug target in PCs that arise in AA men.
In Aim 1 we will 1.1) determine the distribution of PC subtypes in AA men, and assess whether they similarly predict progression to advanced disease in AA and CA men; 1.2) identify master regulator proteins operating in AA PCs; and 1.3) determine whether application of the PCS system can be used to improve the predictive power of cancer-relevant prognostic signatures.
In Aim 2 we will develop methods using radical prostatectomy and biopsy specimens from AA and CA men to determine which tumors will be most sensitive to OC2-targeted therapy.
In Aim 3 we will determine whether OC2 is a viable drug target. This study will be the largest integrated analysis of RNA expression data ever performed in AA PCs. It will provide a rigorous test of the hypothesis that molecular mechanisms specifying gene regulation and signal transduction differ in AA vs. CA men. These studies will also evaluate a novel drug target that appears to be a central driver of many PCs affecting men of African ancestry.
This project will test the hypothesis that a newly developed computational method of prostate cancer (PC) subtyping can identify cancers that are more likely to progress in African-American (AA) men. The study will also test whether the developmental transcription factor, ONECUT2, is a viable drug target in PCs from AA men.
