Expanding Active Surveillance Inclusion Criteria: A Novel Nomogram Including Preoperative Clinical Parameters and Magnetic Resonance Imaging Findings.

BACKGROUND: Current European Association of Urology, American Urological Association, and National Comprehensive Cancer Network guidelines recommend active surveillance (AS) for selected intermediate-risk prostate cancer (PCa) patients. However, limited evidence exists regarding which men can be selected safely. OBJECTIVE: To externally validate the Gandaglia risk calculator (Gandaglia-RC), designed to predict adverse pathology (AP) at radical prostatectomy (RP) and thus able to improve selection of intermediate-risk PCa patients suitable for AS, and to assess whether addition of magnetic resonance imaging (MRI) findings (MAP model) improves the predictive ability of Gandaglia-RC. DESIGN, SETTING, AND PARTICIPANTS: This is a retrospective analysis of a single-center cohort of 1284 consecutive men with low- and intermediate-risk PCa treated with RP between 2013 and 2019. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: AP was defined as non-organ-confined disease and/or lymph node invasion and/or pathological grade group≥3 at RP. Logistic regression was used to calculate the predictors of AP; calculated coefficients were used to develop a risk score. Receiver operating characteristic curve analysis and decision curve analysis were performed to evaluate the net benefit within models. RESULTS AND LIMITATIONS: At multivariable analysis, age at surgery, prostate-specific antigen, systematic and targeted biopsy Gleason grade group, MRI prostate volume, Prostate Imaging Reporting and Data System score, and MRI extraprostatic extension were significantly associated with AP. The model significantly improved the ability of Gandaglia-RC to predict AP (area under the curve 0.71 vs 0.63 [p<0.0001]). Using a 30% threshold, the proportions of men eligible for AS were 45% and 77% and the risks of AP were 16% and 17%, for Gandaglia-RC and MAP model, respectively. CONCLUSIONS: Compared with Gandaglia-RC, the MAP model significantly increased the number of patients eligible for AS without significantly increasing the risk of AP at RP. PATIENT SUMMARY: In this report, we have developed a risk prediction tool to select men for conservative treatment of prostate cancer. Using the novel tool, more men could safely be allocated to conservative treatment rather than surgery or radiation.

Using biomarkers in patients with positive multiparametric magnetic resonance imaging: 4Kscore predicts the presence of cancer outside the index lesion.

OBJECTIVES: To evaluate if the blood biomarker, 4Kscore, in addition to multiparametric magnetic resonance imaging information could identify patients who would benefit from undergoing only a targeted biopsy. METHODS: We retrospectively analyzed a population of 256 men with positive multiparametric magnetic resonance imaging who underwent standard + targeted biopsy at Mount Sinai Hospital, New York, NY, USA. 4Kscore (OPKO Health, Miami, FL, USA) was sampled from all patients before biopsy. Uni- and multivariable binary logistic regression analyses were carried out to predict clinically significant prostate cancer, defined as International Society of Urological Pathology grade group ≥2, in standard biopsy cores. The model with the best area under the curve was selected and internal validation was carried out using the leave-one-out cross-validation. RESULTS: The developed model showed an area under the curve of 0.86. Carrying out only targeted biopsy in patients with a model-derived probability <12.5% resulted in 39.5% (n = 101) fewer standard biopsies and a 33.9% (n = 20) reduction of detecting grade group 1 disease, while missing grade group ≥2 in 5.2% (n = 4) using standard biopsy only and 1.1% (n = 1) using standard biopsy + targeted biopsy. CONCLUSIONS: 4Kscore in combination with multiparametric magnetic resonance imaging can help to reduce unnecessary standard biopsy and decrease detection of clinically insignificant prostate cancer.

Performance of prostate multiparametric MRI for prediction of prostate cancer extra-prostatic extension according to NCCN risk categories: implication for surgical planning.

BACKGROUND: Prediction of extra-prostatic extension (EPE) in men undergoing radical prostatectomy (RP) is of utmost importance. Great variability in the performance of multiparametric magnetic resonance imaging (mpMRI) has been reported for prediction of EPE. The present study aimed to determine the diagnostic performance of mpMRI for predicting EPE in different National Comprehensive Cancer Network (NCCN) risk categories. METHODS: Overall 664 patients who underwent radical prostatectomy with a staging mpMRI were enrolled in this single-center, retrospective study. Patients with mpMRI report non-compliant with PI-RADSv2.0, were excluded. Patients were stratified according to NCCN criteria: very low/low (VLR-LR) to High Risk (HR) in order to assess final pathology EPE rates (focal and established). Sensitivity, specificity, positive and negative predictive values of staging mpMRI were computed in each group. Univariable and multivariable analysis were used to evaluate predictors of positive surgical margins. RESULTS: Pathological evaluation demonstrated established and focal EPE in 60 (9%) and 106 (16%) patients, respectively, while mpMRI suspicion for EPE was present in 180 (27%) patients. Age, preoperative PSA, PSA density, number of positive cores, NCCN groups, prostate volume, mpMRI suspicion for EPE, PIRADSv2.0 and lesion size differed significantly between the patients with any EPE and without EPE (all P≤0.05). The sensitivity of mpMRI in detecting any EPE varied from 12% (95% CI: 0.6-53%) in VLR-LR to 83% (66-93%) in HR while the corresponding values for the specificity were 92% (85-96%) and 63% (45-78%), respectively. Patients with false-negative mpMRI EPE prediction were more likely to have positive surgical margins in univariable (OR: 2.14; CI: 1.18, 3.87) as well as multivariable analysis adjusting for NCCN risk categories (OR: 1.97; CI: 1.08, 3.60). CONCLUSIONS: The performance of mpMRI for prediction of EPE varies greatly between different NCCN risk categories with a low positive predicting value in patients at low to favorable intermediate risk and a low negative predictive value in patients at Unfavorable intermediate to high risk PCa. Given that mpMRI EPE misdiagnosis could have a negative impact on oncological and functional outcomes, NCCN risk categories should be considered when interpreting mpMRI findings in PCa patients.