Presentation Authors: Bashir Al Hussein Al Awamlh*, New York, NY, Leonard S Marks, Los Angelos , CA, Geoffrey Sonn, Palo Alto, CA, Michael Gross, Jim Hu, New York, NY
Introduction: PI-RADS 3 lesions by definition, are equivocal lesions on multi-parametric MRI with a lower likelihood for detection of clinically significant prostate cancer. We aim to assess for clinical characteristics that might aid in discriminating clinically significant prostate cancer (csPCa) in men with PI-RADS (version 2) 3 lesions.
Methods: Multicenter (Weill Cornell Medicine, UCLA and Stanford) study of biopsy naive men (n = 211) who underwent MRI-targeted and systematic biopsies for PI-RADS 3 using Artemis (Eigen, Grass Valley, CA). Clinical parameters analyzed included age, PSA level prior to biopsy, prostate volume and PSA density (PSAD). Men with Grade Group â‰¥ 2 on either target or template cores were considered harboring csPCa. Chi-squared tests and independent two-sample t-tests compared groups and receiver operating characteristic (ROC) curves explored the prognostic ability of continuous parameters. Multivariable logistic regression assessed for parameters independently associated with csPCa.
Results: The mean age of the cohort was 63.4 years. Of the 211 men that underwent targeted biopsy, 52 (24.6%) had csPCa. Patients with csPCa had similar PSA levels (mean 6.99 vs. 6.11 ng/ml, p = 0.42) to those without csPCa. However, men with csPCa had lower prostate volume (mean 40 vs 56 ml, p < 0.001) and higher PSA density (0.2 vs. 0.12 , p = 0.03). From ROC analyses, prostate volume demonstrated the largest AUC compared with PSA and prostate density (0.73 [95% CI 0.65-0.81] vs. 0.51 and 0.64, respectively). Youden's J statistic statistic determined 37.35 ml as an optimal cutoff for prostate volume (sensitivity 63%, specificity 82%). Multivariable modeling revealed older age and smaller prostate volume were independently associated with Grade Group â‰¥ 2 disease (Table).
Conclusions: We found biopsy naive older men with lower prostate volumes more likely to have csPCA with PI-RADS 3 lesions. However, the clinical application of prostate volume as a decision aid is limited by its sensitivity. Our findings demonstrate an opportunity for bio-marker discovery or MRI refinement to improve csPCa stratification for PIRADS 3 lesions.