Abstract
INTRODUCTION: We aimed to determine the minimum cross-sectional ellipsoid area on magnetic resonance (MR) of intraprostatic nodules that best predicts for subsequent targeted biopsies revealing ≥ grade group (GG) 2 disease. METHODS: Forty-six patients previously diagnosed with GG 1 prostate adenocarcinoma who received cognitively fused, MR-guided, transperineal targeted biopsies in addition to six random biopsies were included in this analysis. A Youden cutpoint analysis was used to determine the ellipsoid area in the axial plane best predicting for ≥GG 2 disease within the targeted biopsy cores and logistic regression used to assess the result. RESULTS: Median time from MR imaging to targeted biopsy was 2.4 (1.4-5.5) months. Forty of 46 (87%) patients had one nodule and 6/46 (13%) had two separate nodules on MR that received targeted biopsy. Of the 52 nodules, five (10%), 33 (63%), and 14 (27%) were Prostate Imaging-Reporting and Data System (PI-RADS) 3, 4, and 5, respectively. Thirteen (25%), six (12%), and 33 (64%) were in the anterior, medial, and posterior regions of the prostate, respectively. Median area was 0.72 (0.49-1.29) cm(2) (average diameter 9.5 mm). Fifteen of 46 (33%) patients had ≥1 random biopsy and 20/52 (38%) nodules had ≥1 targeted biopsy revealing ≥GG 2 disease. The optimal area cutpoint was ≥0.7 cm(2), with an area under the curve of 0.671 (0.510-0.832). On logistic regression, area ≥0.7 cm(2) was solely predictive of targeted biopsy revealing ≥GG 2 disease (odds ratio 6.5, 1.3-32.4, p=0.022). CONCLUSIONS: Nodule area ≥0.7 cm(2) may predict for transperineal-based targeted biopsies being positive for ≥GG 2 disease when 1-2 cores are taken.