Are NCCN and EAU active surveillance criteria reliable in patients with ISUP Grade-2 intermediate-risk prostate cancer? a novel model integrating MRI to predict adverse pathology

Abstract

Introduction To assess adverse pathology (AP) rates in patients with grade group (GG) 2 prostate cancer (PCa) based on biopsy characteristics and treated with radical prostatectomy (RP). Performance of active surveillance (AS) guidelines in distinguishing patients with AP has also been investigated. Methods Records of 345 patients who underwent RP for GG 2 disease detected in prostate biopsy were retrospectively reviewed. Patients with suspicion of extracapsular disease on imaging, PSA >= 20 ng/dL, unavailable biopsy data, and in-bore biopsy were excluded from the study. AP was defined as the presence of ISUP GG >= 3 or extracapsular disease. AP rates in patients meeting the AS criteria of NCCN and EAU guidelines were recorded. A novel model was developed to determine AP predictors by using a multivariable logistic regression analysis and a backward stepwise method. Results Among 231 patients, median age was 64 (45-79), median PSA was 6.1 (1.2-19) ng/dL. According to biopsy and clinical characteristics, 124 patients (53.7%) met the NCCN, 31 patients (13.4%) met the EAU AS criteria. Pathological examination after RP revealed AP in 105 patients (45.5%); GG >= 3 disease in 31 (13.4%), pT3a disease in 78 (33.7%), pT3b disease in 18 (7.8%), and pN1 disease in four patients (1.7%). AP rates in patients meeting NCCN and EAU criteria were 37.9% and 22.6%, respectively. Age ( > 63.5), PSA level ( > 5.04 ng/dL), GG2 PCa-bearing index lesion size on mpMRI ( > 11.5 mm), maximum tumor length/core length ( > 51.5%) and Gleason Pattern 4 percentage (>%17.5) were independent predictors of AP in our new model. Conclusions NCCN AS criteria were associated with nearly a twofold higher rate of AP compared with patients meeting EAU criteria. Our new model, including parameters derived from age, PSA, mpMRI and biopsy characteristics, demonstrated superior performance relative to both NCCN and EAU criteria regarding AP prediction among patients with GG 2 PCa.