Incorporating mpMRI Biopsy data into established pre-RP nomograms: Potential impact of an increasingly common clinical scenario.

Abstract

32 Background: Current pre-radical prostatectomy (RP) nomograms predicting lymph node involvement (LNI) are based on systematic 12-core prostate biopsies (PBx). With the introduction of mpMRI, cognitive or fusion biopsies have become prevalent, often in the absence of systematic cores. We examine the practical application of MR biopsy data using established pre-RP nomograms and the potential implications on RP intra-operative decision making. Methods: Utilizing a prospectively maintained single institution database, all patients who underwent MRI-based PBx prior to RP were identified. Each patient was assessed using the MSKCC Kattan nomogram and the Briganti nomogram using the following iterations: 1) Targeted [T] (targeted cores alone), 2) Targeted & Systematic [TS] and 3) Targeted Augmented [TA]. The TA iteration utilized targeted core data alone and assumed negative remaining systematic cores for a total 12 core. Nomogram outcomes, specifically risk of LNI, was compared across iterations. Clinically significant impact was defined as a change in risk above or below 2% (Δ2) or 5% (Δ5), based on current guidelines recommendations for lymph node dissection. Results: 69 men met inclusion criteria (6 targeted, 63 systematic + targeted PBx). In the 6 men with targeted only biopsies, using the Kattan and Briganti nomograms, Δ2 occurred in 1 patient (16.7%) and Δ5 in 1-2 patients (16.7-33.3%); in all, TA iteration LNI was lower than the T iteration. In the 58 patients with positive targeted biopsy cores, Δ2 and Δ5 were 8.62-32.76% and 25.86-37.93%, respectively. In the subset of 52 patients with both targeted and systematic biopsies, using their TS nomogram as an internal validation, the TA iteration was a better approximation of their TS outcomes than their T iteration in 48% (Kattan) and 67% (Briganti) of patients. Conclusions: mpMRI-based PBx results, and in particular those from targeted biopsy cores alone, yield significantly different results using established pre-RP nomograms. Therefore, future nomograms must better incorporate MRI biopsy data and provide guidelines on how to account for targeted cores. In the interim, augmenting targeted biopsy data may bridge the gap.