Genomic Scores are Independent of Disease Volume in Men with Favorable Risk Prostate Cancer: Implications for Choosing Men for Active Surveillance.

ABSTRACT

PURPOSE: We sought to determine whether disease volume at prostate biopsy would correlate with genomic scores among men with favorable risk prostate cancer. MATERIALS AND METHODS: We identified all men with NCCN® (National Comprehensive Cancer Network®) very low and low risk disease who underwent Oncotype DX® prostate testing at our institution from 2013 to 2016. Disease volume was characterized as the percent of positive cores, the number of cores with greater than 50% involvement, the largest involvement of any single core and prostate specific antigen density. Nonparametric testing was performed to compare the median Genomic Prostate Score™ and the likelihood of favorable pathology findings between quartiles of disease volume. RESULTS: We identified 112 (37.8%) and 184 men (62.2%) at NCCN very low and low risk, respectively. Median scores did not differ significantly between disease volume quartiles (all p >0.05). However, the median likelihood of favorable pathology findings statistically differed between volume quartiles (all <0.05). Seven of the 105 men (6.3%) with very low risk disease were reclassified at low risk and 13 of 181 (7.2%) with low risk disease were reclassified at intermediate risk. Genomic disease reclassification did not depend on biopsy tumor volume. CONCLUSIONS: In patients with NCCN very low and low risk prostate cancer genomic scores did not demonstrate meaningfully significant differences by volume based on clinically established cutoff points. Moreover, genomic scores identified and reclassified men with higher risk disease despite generally acceptable surveillance characteristics in this group according to grade and volume. This suggests that in patients at low risk the tumor biological potential measured by genomics rather than by volume should inform decisions on active surveillance candidacy.