Expanding inclusion criteria for active surveillance in intermediate-risk prostate cancer: a machine learning approach.

PURPOSE: To develop new selection criteria for active surveillance (AS) in intermediate-risk (IR) prostate cancer (PCa) patients. METHODS: Retrospective study including patients from 14 referral centers who underwent pre-biopsy mpMRI, image-guided biopsies and radical prostatectomy. The cohort included biopsy-naive IR PCa patients who met the following inclusion criteria: Gleason Grade Group (GGG) 1-2, PSA < 20 ng/mL, and cT1-cT2 tumors. We relied on a recursive machine learning partitioning algorithm developed to predict adverse pathological features (i.e., ≥ pT3a and/or pN + and/or GGG ≥ 3). RESULTS: A total of 594 patients with IR PCa were included, of whom 220 (37%) had adverse features. PI-RADS score (weight:0.726), PSA density (weight:0.158), and clinical T stage (weight:0.116) were selected as the most informative risk factors to classify patients according to their risk of adverse features, leading to the creation of five risk clusters. The adverse feature rates for cluster #1 (PI-RADS ≤ 3 and PSA density < 0.15), cluster #2 (PI-RADS 4 and PSA density < 0.15), cluster #3 (PI-RADS 1-4 and PSA density ≥ 0.15), cluster #4 (normal DRE and PI-RADS 5), and cluster #5 (abnormal DRE and PI-RADS 5) were 11.8, 27.9, 37.3, 42.7, and 65.1%, respectively. Compared with the current inclusion criteria, extending the AS criteria to clusters #1 + #2 or #1 + #2 + #3 would increase the number of eligible patients (+ 60 and + 253%, respectively) without increasing the risk of adverse pathological features. CONCLUSIONS: The newly developed model has the potential to expand the number of patients eligible for AS without compromising oncologic outcomes. Prospective validation is warranted.

Magnetic Resonance Imaging-based T-staging to Predict Biochemical Recurrence after Radical Prostatectomy: A Step Towards the iTNM Classification.

BACKGROUND: Local staging of prostate cancer (PCa) still relies on digital rectal examination (DRE), which therefore remains the standard for risk stratification in guideline recommendations, clinical trials, and patient counseling. This issue is increasingly controversial as multiparametric magnetic resonance imaging (mpMRI) has become the most influential diagnostic tool for local staging of PCa over the past two decades. OBJECTIVE: To compare various models of T category based on DRE or mpMRI to predict early biochemical recurrence (BCR) after radical prostatectomy (RP). DESIGN, SETTING, AND PARTICIPANTS: A retrospective multicenter cohort study was conducted between 2014 and 2021. A total of 1436 patients were recruited across eight referral centers in France, Italy, Switzerland, and Belgium. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: BCR was defined as two prostate-specific antigen values of ≥0.2 ng/ml during follow-up. Harrell's concordance index (C index) was used to compare the discrimination of four models of T staging based on DRE (model 1: cT1 vs cT2 vs cT3) or mpMRI (model 2: organ-confined disease vs extracapsular extension [iECE] vs seminal vesicle invasion [iSVI]; model 3: Prostate Imaging-Reporting and Data System [PI-RADS] ≤3 vs PI-RADS 4 vs PI-RADS 5; and model 4: iT2a [PI-RADS ≤3] vs iT2b [PI-RADS 4] vs iT2c [PI-RADS 5 excluding ECE or SVI] vs iT3a [ECE] vs iT3b [SVI]) to predict BCR. RESULTS AND LIMITATIONS: Overall, 74 (5%), 845 (59%), 482 (34%), and 35 (2%) patients had low-, intermediate-, high-, and very high-risk PCa, respectively, according to the Mazzone risk classification. After median follow-up of 16 mo, 113 patients experienced BCR. Although the new five-group mpMRI-based T classification system (model 4) had the highest prognostic discrimination (C index 0.694) for predicting early BCR on multivariable analysis, there was overlap between the 95% confidence intervals of the models. On sensitivity analysis, the new mpMRI-based T staging still had a higher C index than DRE for predicting BCR when excluding cN1 patients and comparing it with a five-group DRE-based T classification (cT1c vs cT2a vs cT2b vs cT2c vs cT3), but the overlap between the 95% confidence intervals of the models remained. The main limitation is the short follow-up. CONCLUSIONS: We described an alternative mpMRI-based T staging for prediction of early BCR after RP for PCa. Our results need to be validated externally before they can be applied in clinical practice. PATIENT SUMMARY: At present, digital rectal examination of the prostate is used to stage prostate cancer. We developed an alternative model for staging that uses information from magnetic resonance imaging (MRI) scans to predict cancer outcomes for men undergoing surgical removal of the prostate.