Purpose: Similar to prostate-specific antigen (PSA) density, PCA3 density (PCA3D: ratio of urinary PCA3 score/prostate volume) can be calculated, but whether it can be an aid to decide biopsy in patients at risk of prostate cancer (PCa) is uncertain. The objective was to demonstrate that PCA3D provides better specificity than PCA3 in predicting initial prostate biopsy outcome.
Methods: Serum and urine samples were obtained from 595 consecutive patients scheduled for initial prostate biopsy. The urinary PCA3 test was performed before biopsy. Additional measures were prostate volume, PSA density (PSAD) and PCA3D. Multivariate logistic regression models including baseline characteristics and the markers were evaluated. The presumed net benefit was assessed through decision curve analyses.
Results: PSAD and PCA3D performed better than PSA and PCA3 score, respectively. PCA3D provided the best specificity (76 %). The best calculated cutoff for PCA3D was 1. The risk of positive biopsy significantly increased to 70 % if PCA3D ≥ 1 versus 29 % if PCA3D was <1. Using a cutoff at 0.5 for PCA3D, biopsies could have been avoided in up to 52 % of the patients without PCa while missing 15 % of any PCa and 10 % of PCa with Gleason score ≥7. Decision curve analyses showed that PSAD was the best predictor of Gleason score at biopsy while PCA3D best predicted the proportion of invaded cores.
Conclusions: PCA3D showed a significant increase in specificity when compared with PSA, PSAD and PCA3. PCA3D can be considered an easy-to-use mini-nomogram with a 70 % risk of positive initial biopsy when PCA3D > 1, i.e., PCA3 score > prostate volume.