Black Race Predicts Poor Compliance and Higher Prostate Cancer Risk in a Multinational Repeat Biopsy Cohort: Secondary Analysis From the Reduction by Dutasteride of Prostate Cancer Events (REDUCE) Clinical Trial.

BACKGROUND: On a population level, black patients have more prostate cancer (PC), particularly aggressive PC. Similarly, on initial biopsy, black race has been linked with higher PC and high-grade PC risk. Whether this extends to first repeat biopsy is unknown. We tested if black race predicts PC risk, grade, and biopsy compliance within a phase 3 PC prevention trial with study-mandated biopsies among patients with an initial negative prestudy biopsy. METHODS: Analysis of 7445 patients (2.4% black, 97.6% white) from REDUCE, a 4-year, double-blind, placebo-controlled trial testing dutasteride versus placebo on PC risk. Patients had a single, negative, prestudy biopsy, PSA 2.5-10 ng/mL, and study-mandated biopsies at 2 and 4-years. Multivariable logistic and multinomial regressions were used to test if self-reported race predicted PC, grade (low, Grade-Group 1 [GG1] vs. high, Grade-Group > 2 [GG2+]), and biopsy compliance. Given concerns about noncompliance affecting analyses, primary analyses for PC examined 2-year biopsy outcomes among patients who underwent biopsy. RESULTS: On multivariable analysis, black patients were less likely to receive the 2-year biopsy (OR: 0.53, 95%CI: 0.39-0.74), but had higher PC risk, which approached, but did not reach significance (OR 1.56, 95%CI: 0.98-2.46). When stratified by grade, on multivariable analysis, black race was unrelated to GG1 (RRR: 1.17, 95%CI: 0.65-2.12) but associated with GG2 + PC (RRR: 2.44, 95%CI: 1.29-4.64) at the 2-year biopsy. CONCLUSION: Among patients with a negative prestudy biopsy on a phase 3 trial, black patients had lower compliance with study-mandated biopsy but were 2.44 times more likely to have GG2 + PC on the 2-year biopsy versus white patients. These data extend population and initial biopsy data to first repeat biopsy, strongly supporting that black race is linked with aggressive PC. Given differential biopsy compliance may be greater in the real-world, these data suggest population data may underestimate true racial disparities in PC risk and aggressiveness.