Is race a positive predictor of cancer on repeat prostate biopsy?

PURPOSE: High grade intraepithelial neoplasia and atypical small acinar proliferation increase the probability of cancer on a subsequent prostate biopsy. We investigated whether race is prognostic for detecting cancer in patients undergoing repeat prostate biopsies.
MATERIALS AND METHODS: At a single institution 416 men underwent 2 or more prostate biopsies from January 1993 through June 2003 for a total of 1,023 biopsies. We retrospectively examined multiple factors, including patient age, race, total number of biopsy cores total number of previously negative biopsy cores, prostate specific antigen, prostate specific antigen slope, digital rectal examination and family history of prostate cancer. Previous high grade intraepithelial neoplasia, atypical small acinar proliferation and Gleason score in positive biopsies were recorded from the histopathology review. Clinical variables were compared between black and white men using the Wilcoxon rank sum and Fisher exact tests. The Cox proportional hazards model was used for multivariate analysis.
RESULTS: Of the 416 men 216 (51.9%) were black, 174 (41.8%) were white and 26 (6.3%) were another race. The average number of biopsy sessions in black and white men was 2.41 and 2.51, respectively. The cancer detection rate was 35.1% at the second biopsy, 34.6% at the third biopsy and 32.0% at the fourth biopsy. Cancer was diagnosed in 43.5% of black men compared to 25.9% of white men (p = 0.0004). When clinical and pathological variables were compared between the racial groups, black men had significantly higher prostate specific antigen (p = 0.02). There was no statistically significant difference in patient age, total number of cores, number of previous negative cores, prostate specific antigen slope, abnormal digital rectal examination, family history, or previous high grade intraepithelial neoplasia or atypical small acinar proliferation. Multivariate analysis showed that race approached but did not achieve statistical significance as a predictor of prostate cancer on repeat biopsy (p = 0.09). Previous high grade intraepithelial neoplasia (p = 0.0025), previous atypical small acinar proliferation (p = 0.0049), digital rectal examination (p = 0.0076) and prostate specific antigen slope (p = 0.0005) were independent predictors of prostate cancer on repeat biopsy. Of patients with previous atypical small acinar proliferation black men had a significantly higher rate of cancer detection on repeat biopsy.
CONCLUSIONS: Previous high grade intraepithelial neoplasia, atypical small acinar proliferation, digital rectal examination and prostate specific antigen slope were independent predictors of prostate cancer on repeat biopsy. Race approached but did not attain significance after adjusting for disease features.