Associations of cell cycle genetic variants with aggressive prostate cancer in the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial.

Abstract

175 Background: The need to differentiate patients at risk for developing aggressive prostate cancer (CaP) from those at risk for less aggressive disease has led to efforts to identify genetic markers to predict disease course and personalize treatment. A study with participants from Johns Hopkins Medical Institutions (JHMI) and Washington University (WU) found single nucleotide polymorphisms (SNPs) in cell cycle genes were associated with risk of aggressive CaP. We sought to replicate those results in the European-American population of the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial. Methods: We focused on variants associated with aggressive CaP in the JHMI and WU cohorts, which included 13 SNPs in 12 genes ( CCNC, CCND3, CCNG1, CCNT2, CDK2, CDK6, MDM2, SKP2, TERF2, WEE1, YWHAB, YWHAH). Variants were genotyped using the Pyrosequencing assay. Patients were classified into high risk (Gleason≥8, pT3b, N+, M+), low risk (Gleason≤7, ≤pT3a, N0, M0) or non-cancer control groups based on clinicopathologic characteristics. Logistic regression analysis was used to compare genotype frequencies of each variant between groups using the dominant model. Results: There were 108 aggressive and 1080 non-aggressive CaP patients, and 1155 controls . CDK6 ( rs8) was associated with increased risk of any CaP (OR 1.2, 95% CI 1.02-1.42; p = 0.032) and high risk disease (OR 1.63, 95% CI 1.09-2.42; p = 0.017) vs. controls. In contrast, the JHMI and WU cohort found CDK6 variants to be protective against aggressive CaP. CCNG1 ( rs11541970) approached significance (p = 0.052) between high risk and control groups, and CCNC conferred a protective effect consistent with the prior study, but did not reach significance (p = 0.101). No associations with any cell cycle gene variants were detected when comparing high and low risk patients. Conclusions: Our study did not replicate the results from the JHMI and WU cohorts. CDK6 predicted an increased risk of developing any CaP and high risk CaP. However, directionality was opposite to the prior study, indicating that this variant is unlikely to be a true predictor of increased risk of or protection from aggressive CaP.