Background: Few prospective studies have focused on the performance of the Prostate Health Index (PHI) in Asian populations. Therefore, we aimed to evaluate the performance of the PHI in predicting prostate cancer (PCa) compared with standard prostate-specific antigen (PSA) tests.
Methods: We prospectively enrolled patients with suspected PCa with a total PSA (tPSA) level 4 to 10 ng/mL or tPSA <4 ng/mL and a suspicious digital rectal examination between February 2017 and September 2018. All of the patients underwent a 12-core transrectal ultrasound-guided prostate biopsy. Prebiopsy blood samples were analyzed for tPSA, free PSA (fPSA), percentage of fPSA (%fPSA), [-2]proPSA (p2PSA), and percentage of p2PSA (%p2PSA). The PHI was calculated as (p2PSA/fPSA) × √tPSA. The areas under the receiver operating characteristic curve (AUCs) were estimated for the PSA derivatives in addition to their specificities at a prespecified sensitivity of 90%.
Results: Of the 307 enrolled patients, 95 (30.9%) had PCa on biopsy. Excluding fPSA, all of the PSA derivatives were significantly different between the positive and negative biopsy groups. Of the various derivatives, the PHI (AUC: 0.783) showed the best performance in predicting the results of the initial biopsy compared with tPSA (AUC: 0.611). At a sensitivity of 90%, the PHI had the best specificity of 46.7% compared with 23.2% for tPSA. Using a PHI cutoff value of 35.15 for biopsy, 108 (35.2%) patients could have avoided undergoing a biopsy. To detect Gleason score ≥ 7 disease at 90% sensitivity, the threshold for PHI was 36.96 with a specificity of 52.1%.
Conclusion: PHI was the best biomarker among the PSA derivatives in predicting PCa at biopsy in men with tPSA < 10 ng/mL. The risk of a Gleason score ≥ 7 increased with increasing PHI.