Re-thinking How We Use Prostate Health Index for African American Men

Abstract

Objective: To assess how the validated Prostate Health Index (PHI) risk stratifications perform with African American (AA) men and establish a threshold PHI value to potentially rule out the need for prostate biopsy.

Materials and methods: AA men meeting FDA-specified indications for PHI testing (>50 years old, PSA 4-10 and negative DRE) who underwent subsequent biopsy were included. Rates of clinically significant prostate cancer (csPCa, as defined by Gleason score ≥7) across accepted PHI stratifications were recorded. Receiver operator curve (ROC) analysis was undertaken to assess PHI performance to predict csPCa. A phi cutoff providing 90% sensitivity was identified. Among AA men with PSA 4-10 ng/mL, the proportion of men who proceeded to biopsy upon physician recommendation was determined.

Results: Two hundred nine patients met primary criteria; 91 (43.5%) of which had csPCA. The area under the curve for PHI predicting csPCa was 0.68 (95% CI: 0.61-0.75). Using a phi threshold of <23.0 to avoid biopsy provided 98.9% sensitivity, 9.3% specificity, and would have avoided 4.7% of biopsies. The proportion of those who proceeded to biopsy upon physician recommendation was 81.8%.

Conclusions: PHI demonstrated limited performance in our cohort, with current stratifications featuring misleadingly low cancer detection rates for these men. Furthermore, PHI had limited use to avoid prostate biopsy, as the proposed threshold of 23.0 only allowed 4.7% of men to avoid biopsy. Further work is needed to assess and optimize PHI usage in AA men; nonetheless, it may still have use in increasing compliance with biopsy recommendation.