Molecular Predictors of Prostate Cancer Progression and Mortality
Co-Leaders: Janet Stanford, Ph.D., Daniel Lin, M.D.
Prostate cancer (PCa) is a major cause of mortality, but it is not currently possible to accurately predict which tumors will remain indolent vs become life-threatening. To identify prognostic biomarkers, the proposed study will utilize samples and data from a population-based cohort of 1,458 PCa patients to assess genome-wide DNA methylation in tumor tissue and germline single nucleotide variants in candidate genes in relation to recurrence and PCa-specific mortality.
PCa is the most common solid tumor in men and is a major cause of cancer-related morbidity and mortality. Prostate-specific antigen (PSA) testing has increased the number of men diagnosed with PCa, but ~30–42% of these patients have indolent tumors that carry a low probability for progression to clinically significant PCa. Clinicopathological criteria alone are not always adequate for predicting which tumors will remain indolent vs become aggressive, so many patients are over-treated and some are under-treated. Thus, biomarkers to distinguish men with less vs more aggressive disease are urgently needed.
Both inherited genetic variation (eg, SNPs) and tumor epigenomic aberrations (eg, DNA hypermethylation) likely contribute to PCa aggressiveness. Preliminary evidence supports both mechanisms, which may alter host-tumor immunity, tumor growth rate, or metastatic propensity. The overall intent of this population sciences research is to validate genetic-epigenetic biomarkers for aggressive PCa that can be translated into clinical use. Toward this goal, the project has the following aims:
- To complete validation of a panel of 30 PCSM-associated SNPs in two independent PCa patient cohorts;
- To characterize genome-wide DNA methylation (450K CpG sites) profiles in prostate tumor tissue in association with PCa-specific outcomes (eg, recurrence, metastasis, PCSM); and,
- To test validated PCSM-associated SNPs (Aim 1) and top-ranked differentially methylated genes (Aim 2) as an integrated panel of prognostic biomarkers for distinguishing clinically localized aggressive PCa.
The proposed plan builds on our prior SPORE work, taking advantage of a population-based PCa cohort with germline DNA, tumor tissue, clinical and PCa-specific outcomes data, as well as other PCa cohorts with available DNA and outcomes data. Univariate, stratified, and multivariate analyses will be completed to evaluate PCSM-associated SNPs and top-ranked differentially methylated genes associated with aggressive PCa for potential clinical utility. The Cox proportional hazards model will be used to calculate hazard ratios, 95% CIs, and p-values to examine the association of individual and combinations of germline genetic and somatic (DNA methylation) biomarkers with PCa outcomes. The overall goal is to identify and validate prognostic genetic-epigenetic biomarkers that will translate into better patient management and outcomes as well as to identify new molecular targets that may lead to novel therapies or prevention approaches for PCa.