Pathological features of Prostate Imaging Reporting and Data System (PI-RADS) 3 MRI lesions in biopsy and radical prostatectomy specimens.

Objective To assess the whole pathology spectrum of Prostate Imaging Reporting and Data System (PI-RADS) 3 lesions, identified on magnetic resonance imaging, using systematic (SB), targeted biopsy (TB) and radical prostatectomy (RP) specimen analysis. Methods From a prospective database of patients undergoing RP after a combination of SB (median 12 cores) and fusion TB (median 3 cores), we included 150 PI-RADS 3 cases. Clinically significant prostate cancer (csPCa) was defined by a Grade Group 2 or more. The primary endpoints were unfavourable features in RP specimens. Results Targeted biopsy was negative in 20.7% of patients. Final Grade Group 3 or more and a pT3 stage was reported in 36.7% and 38.7% of RP specimens. The upgrading rate was 38.2% between biopsy and RP specimens. The concordance rate between Grade Group on TB and RP was only 38.0%. The two independent predictive factors for unfavourable disease (pT3-4 and/or final Grade Group 3-5) were prostate-specific antigen density (PSAD; P = 0.001) and presence of csPCa on TB (odds ratio 3.7; P = 0.001). The risk of unfavourable disease was increased 2.3-fold and 5.8-fold, respectively, for patients with a PSAD between 0.15 and 0.20, and a PSAD >0.20 ng/mL/g. The 5-year biochemical recurrence-free survival rate was 93.2%. Conclusions PI-RADS 3 lesions exhibited aggressive features in almost 40% of cases. PSAD and presence of csPCa on TB are independent predictive factors for high-grade and/or extraprostatic disease. A combination of SB and TB improve grade prediction compared to use of TB alone.