Tailored Use of PSA Density According to Multiparametric MRI Index Lesion Location: Results of a Large, Multi-Institutional Series

Multiparametric magnetic resonance imaging (mpMRI) of the prostate has become the standard of care for the diagnosis of prostate cancer (PCa) in patients with elevated prostate-specific antigen (PSA) levels, primarily due to its high negative predictive value (NPV). However, the positive predictive value (PPV) of mpMRI exhibits substantial variation across both prostate imaging reporting and data system (PI-RADS) scores and individual patients’ clinical risk. Moreover, it remains suboptimal for PI-RADS 4 and 5 (40% and 69%), and very low (13%) for PI-RADS 3 lesions. Improving the PPV of mpMRI as a decision-making tool to reduce the number of unnecessary prostate biopsies becomes pivotal.

PSA density (PSAd) has been proposed as a biomarker to identify patients with negative mpMRI who may still require biopsy. While combining PSAd with mpMRI improves the detection of clinically significant prostate cancer (csPCa), the influence of the index lesion location (ILL) on the performance of PSAd remains unclear. In fact, the historical cutoff of 0.15 ng/ml/ml proposed before mpMRI era is not generalizable to contemporary cohorts, since patients with positive mpMRI have a higher probability of harboring csPCa in comparison to their counterparts considered for prostate biopsy in the pre-mpMRI era.

In this context, we evaluated the diagnostic performance of PSAd for csPCa according to ILL and after stratifying patients according to PI-RADS score in a large multi-institutional series of men with positive mpMRI (PI-RADS ≥3) who underwent mpMRI-targeted plus systematic biopsy. Our findings support considering patients with PI-RADS ≥4 at higher risk of csPCa, being the risk ≥20% regardless of PSAd and ILL. Conversely, patients with PI-RADS 3 lesions located in the peripheral zone and transition zone have a csPCa risk <10% in case of PSAd values <0.05 ng/ml/ml and <0.13 ng/ml/ml, respectively.

From a clinical standpoint, this study may gain significant clinical relevance in the classic diagnostic algorithm of PCa. Specifically, ILL may be incorporated into existing models aimed at predicting the need for prostate biopsy in patients with PI-RADS 3 lesions. In fact, our findings show that PSAd may not categorize patients with PI-RADS 4–5 lesions, as the risk of harboring csPCa remains substantial even among those with low PSAd levels. However, we found that PSAd may be a valuable tool for stratifying PI-RADS 3 lesions, also taking into account ILL, which appears to be crucial for the accuracy of this biomarker.

Written by: Leonardo Quarta, MD, Unit of Urology/Division of Oncology, Gianfranco Soldera Prostate Cancer Lab, Urological Research Institute (URI), IRCCS San Raffaele Scientific Institute, Milan, Italy

Read the Abstract

Login