PD42-12: Modeling the Contribution of Preoperative VI-RADS Score to Predict Final Pathologic Lymph-Nodes Involvement in Patients with MIBC Undergone Radical Cystectomy

Francesco Del Giudice*, Martina Pecoraro, Maurizio Del Monte, Marco Bicchetti, Emanuele Messina, Ailin Dehghanpour, Rome, Italy, Benjamin I. Chung, Stanford, CA, Ettore De Berardinis, Alessandro Sciarra, Carlo Catalano, Valeria Panebianco, Rome, Italy

Introduction: Muscle invasive bladder cancers (MIBCs) eligible for radical cystectomy (RC) are frequently associated with subsequent pathologic regional LN involvement (pN1-3). Involvement increases with the depth of invasion, thus preoperative tools aiming to predict final histology would assume particular relevance given that a significant subset of patients may be rendered long term disease free with surgery alone. We explored the ability of the novel Vesical-Imaging Reporting and Data System (VI-RADS) score to correlate with pN+ status and lymph node density (LNd) at RC pathology report.

Methods: Patients’ ineligible neoadjuvant chemotherapy who underwent multiparametric magnetic resonance imaging (mpMRI) of the bladder prior to staging trans-urethral resection of bladder tumor (TURBT) followed by RC and extended pelvic lymph-node dissection (ePLND) for cT2-4 N0 M0 MIBC. Multivariable logistic regression model and receiver operating characteristics curve (ROC) were used to asses VI-RADS features to predict final pN+ status and higher LNd.

Results: n=128 T2–T4a, cN0–M0 patients (VI-RADS Score 5, n= 39, Score 4, n=52, Score 3, n=37) were examined. pN+ rate was 27% with n=10 (7.8%), n=26 (20.3%) presenting with pN1 and pN2 respectively. Mean lymph node density (SD) was 9% (1.5%). Increasing VI-RADS score was independently associated with higher odd of LN+ status (OR, 1.79, 95%CI: 1.24 – 2.47). Moreover, VI-RADS score 5 presented an area under the curve (AUC) of 0.89 (95%CI: 0.74 – 0.96) for LNd threshold of 5.7% when compared to preoperative VI-RADS 3-4.

Conclusions: VI-RADS is a novel tool potentially able to preoperatively predict MIBCs at higher risk for presenting pN+ status at final RC histological report. In future VI-RADS could serve as a non-invasive method to aid urologists planning the extent of LN template thus balancing the benefit and morbidity of extended vs. limited LN dissection at the moment of RC.

Source of Funding: None