Introduction

African American men demonstrate significant disparities in prostate cancer outcomes compared to men of other ethnicities. This study aims to assess the difference in biochemical recurrence between African American men and non-African American men with high risk prostate cancer treated by radical prostatectomy.

Methods

This is a multi-institutional observational study comprised of 1869 men with NCCN high risk prostate cancer managed primarily by radical prostatectomy at the Cleveland Clinic, Johns Hopkins University, and MD Anderson Cancer Center. In total, 233 (12.5%) men in this cohort self-identified as AA. A Cox regression model was constructed to predict the risk of biochemical recurrence (BCR) while adjusting for differences in clinical covariates between African American and non-African American men. All data is presented as median[IQR].

Results

On univariate analysis, African American men demonstrated differences in median age (60.0 years [54.0-65.0] vs. 57.0 years [52.0-67.0]), initial prostate specific antigen (PSA) (8.8 ng/ml [5.2-21.5] vs. 7.1 ng/ml [4.9-13.4]), number of cores with Gleason score 8 or greater disease (1[0-3] vs. 2[1-4]), T stage (1[1-2] vs. 2[1-2]), and grade group (4[4-5] vs. 4[4-5]), compared to non-African American men (all p < 0.01). African American men in the cohort had median follow-up of 36.5 months [14.3-59.5] compared to 36.2 months [14.1-62.0] among non-African American men (p = 0.81). Overall, the BCR-free probability was 54.7% and 51.2% for African American men and 61.3% and 51.4% for non-African American men at 3- and 5- years, respectively. On Cox regression, African American ethnicity was associated with hazard ratio 1.31 (95% CI 1.05, 1.63, p = 0.02) in a model controlling for age, initial PSA, clinical stage, grade group, and number of biopsy cores with Gleason 8 or higher disease.

Conclusions

African American ethnicity was associated with a 31% increased risk of biochemical recurrence when adjusted comparisons were performed. Further research is needed to determine if these observed differences in the rate of BCR are modifiable among AA men with high risk disease.

Funding

none