Predictors of Positive Biopsy Following a PIRADS 3 Finding on MRI
Andrew Tam, MD, Ilene Staff, PhD, Kevin Pinto, BS, Joseph Tortora, MS, Tara McLaughlin, PhD, Joseph Wagner, MD.
Hartford Hospital, Hartford, CT, USA.
BACKGROUND: MRI is increasingly utilized as an adjunct procedure in the evaluation of patients with elevated prostate-specific antigen (PSA). While the current consensus is to biopsy PI-RADS 4 and 5 lesions, recommendations for PI-RADS 3 lesions are less clear. Historically, the rates of PI-RADS 3 lesions harboring clinically significant (Gleason Grade Group ≥2) prostate cancer (csPCa) vary widely (2-30%). The objective of our study is to evaluate predictors of PCa and csPCa in patients with PI-RADS 3 lesions on MRI. METHODS: We queried our electronic records to identify patients with PI-RADS 3 lesions on MRI from 2014-2022 who subsequently underwent a software targeted (UroNav®) biopsy with concurrent systematic prostate biopsies. In a multifaceted design, patients positive vs negative for PCa on biopsy were compared on demographics, biopsy history and other clinical characteristics. A second analysis compared men positive for csPCa with all others (negative and Gleason Grade Group (GGG) 1). Chi square tests and Wilcoxon rank sum tests were used for analyses of categorical and continuous variables respectively. Logistic regressions were used to explore the independence of predictors, limited in scope by the volume of positive biopsy findings. A separate set of analyses focused only on the MRI targeted cores exploring potential predictors, such as lesion size and location that are specific to the lesion. Given that each set of analyses was repeated for PCa and csPCa, a Bonferroni correction was applied; p<0.025 was used as the criterion value to achieve an overall significance level of p<0.05. RESULTS: 184 patients met inclusion criteria; 16/184 (8.7%) patients had csPCa; 51/184 (27.7%) patients had any PCa. For the main analyses, age, lack of prior biopsy, prostate size, and PSA density were associated with PCa diagnosis; age, prostate size, and PSA density were associated with csPCa (Table 1). PSA density remained strongly associated with both PCa (OR (95% CI) = 4.1 (1.9- 8.9); p<0.001) and csPCa (OR =12.6 (3.7-43.1); p<0.001) in multivariate analyses independent of age and prior biopsy which were significant for all PCa but not csPCa (Table 2). At the individual lesion level, 30/200 (15.0%) of the targeted biopsies demonstrated PCa; 14/200 (7.0%) harbored csPCa. There was a higher proportion of anterior lesions among those positive for PCa and csPCa but the associations were not significant in this small sample. CONCLUSIONS: With low rates of csPCa, patients with PI-RADS 3 lesions are faced with significant uncertainty. Prostate size, age, and especially PSA density are other factors to consider and may help clinicians identify those men for whom biopsy would be most beneficial. 
Back to 2023 Abstracts