LAUSR

1551Background: The Prostate Cancer Prevention Trial (PCPT) reported decreased risk of all prostate cancers but increased risk of high-grade prostate cancer in the finasteride arm. The aim of the present analysis was to identify genotypes that modify the effect of finasteride using case-only methods for estimating gene-treatment interactions. Methods: Germline genetic data from 1157 prostate cancer cases in the Prostate Cancer Prevention Trial (PCPT, NCT00288106) were analyzed by case-only methods. Genotypes included 357 single nucleotide polymorphisms (SNPs) from 83 candidate genes in androgen metabolism, inflammation, circadian rhythm and other pathways. Univariate case-only analysis was conducted to evaluate whether individual SNPs modified the finasteride effect on the risk of high-grade and low-grade prostate cancer. Case-only classification trees and random forests, which are powerful machine learning methods with resampling-based controls for model complexity, were employed to identify a predictive...