For a study, researchers sought to determine the practice-level variance in the management of magnetic resonance imaging (MRI) Prostate Imaging-Reporting and Data System (PI-RADS) 3 lesions in men with favorable-risk prostate cancer (FRPC) who were contemplating or were on active surveillance (AS).

They looked through the Michigan Urological Surgery Improvement Collaborative registry for FRPC males (GG1 & low-volume GG2) who had MRIs between January 2013 and March 2020. The primary goal was to examine practice-level variance in the interval between MRI and biopsy and MRI and therapy for PI-RADS 3 lesions. The MRIs were taken both after the diagnostic biopsy and while on AS. The Kaplan-Meier technique was used to quantify biopsy-free survival from MRI to surveillance biopsy, and multivariable Cox proportional hazards models were used to identify clinical and demographic characteristics related to the time it took to receive a biopsy after discovering PI-RADS 3 lesions. 

They found 3,172 FRPC males having MRIs, 473 of whom had a PI-RADS 3 score. There was considerable practice-level heterogeneity in biopsy rates among patients with PI-RADS 3 MRI data (log-rank test, P<.001), with biopsy-free likelihood ranging from 28% to 69% at 6 months (median: 59%). They were unable to find any parameters that had a significant relationship with biopsy time. Time from PI-RADS 3 to treatment, on the other hand, showed less variation (log-rank test, P=.2), while several clinical factors had statistically significant associations: age (P=.018), Prostate Specific Antigen-Density 0.1-0.2 (P=.035), ISUP-GG 2 (P=.002), and a number of positive cores (P<.001), as expected. 

The most important factor influencing the choice to biopsy PI-RADS 3 lesions in FRPC men contemplating or on AS is the urology practice, not GG or the magnitude of biopsy positive. Future work is required to aid decision-making and minimise unpredictability.