Accuracy of prostate imaging reporting and data system (PI-RADS v2) in the detection of prostate cancer (PCa).

Abstract

e17604 Background: MRI-targeted biopsy is increasingly utilized over standard 12-core transrectal ultrasound (TRUS) biopsy for men with MRI-visible prostate lesions. Some clinicians defer biopsy for PI-RADS v2 category 1 and 2 lesions per the PRECISION Study (Kasivisvanathan et. Al, NEJM, 2018). The aim of this study was to independently validate the accuracy of PI-RADS v2 in detecting prostate cancer (PCa) when applied to MRI/US fusion-guided biopsies in an independent cohort of 156 patients from a large integrated community health system. Methods: Men undergoing MRI/US fusion-guided biopsy from 2016-2020 in the Intermountain Healthcare system were consecutively analyzed in this retrospective study. MRI were interpreted from four abdominal fellowship trained radiologists all with at least 5 years of experience. Fusion biopsies were performed by two urologists. Men were stratified into groups based on their PI-RADS v2 category 1-5. Biopsies were considered positive when Gleason ≥3+3. Results: A total of 156 men had 258 lesions for which they underwent MRI/US fusion-guided biopsies in the Intermountain Healthcare system from 2016 to 2020. The PCa detection rate for PIRADSv2 category 1-2 was 29.8%, category 3 32.6%, and category 4-5 37.6%. PIRADS v2 category 1, 2, 3, 4, and 5 yielded any PCa in 25, 15.9, 23.8, 53.1, and 66.7%, respectively (Table). PIRADS v2 category 1-2, 3, and 4-5 yielded any PCa in 16.8%, 23.8%, and 57.7%, respectively. Conclusions: PI-RADS v2 categories generally correlate with PCa detection rates, however, to avoid biopsy, the test must be both sensitive and specific, with low false negative rates. In our institution, we show that PI-RADS 1, 2, and 3 do not rule out the presence of PCa, and therefore should not be used as the sole factor in determining the need for prostate lesion biopsy. [Table: see text]