Abstracts from the New England Section of the AUA 2021

© The Canadian Journal of Urology TM : International Supplement, October 2021 Figure 1: IPW-adjusted Kaplan-Meier plot of overall survival for radical cystectomy (RC) versus trimodality therapy (TMT). Figure 2: Heterogeneity of treatment effects according to cT stage, age, and Charlson score. 1 2 Radical Cystectomy versus Trimodality Therapy for Muscle-Invasive Urothelial Carcinoma of the Bladder Kenneth A. Softness, MD 1 , Sumedh Kaul, MS 1 , Aaron Fleishman, MPH 1 , JasonA. Efstathiou, MD, PhD 2 , Joaquim Bellmunt, MD, PhD 1 , Simon P. Kim, MD 3 , Ruslan Korets, MD 1 , Peter F. Chang, MD, MPH 1 , Andrew A. Wagner, MD 1 , Aria F. Olumi, MD 1 , Boris Gershman, MD 1 1 Beth Israel Deaconess Medical Center, Boston, MA, USA, 2 Massachusetts General Hospital, Boston, MA, USA, 3 University of Colorado, Aurora, CO, USA Introduction : The comparative effectiveness of radical cystectomy (RC) and trimodality therapy (TMT) for muscle-invasive bladder cancer remains uncertain, as no randomized data exist. Aphase 3 trial (SPARE) was attempted in the UK, but randomization was deemed infeasible and the trial was closed. Herein, we emulated the SPARE trial to evaluate the comparative effectiveness of RC versus TMT. Materials & Methods: We used theNCDB to emulate a target trial designed to resemble the SPARE trial.We identified patients aged 40-79with cT2-3 cN0 cM0 urothelial carcinoma of the bladder diagnosed from2006-2015whowere treated withmultiagent neoadjuvant chemotherapy +RCwith lymphadenectomy (RC arm) or multiagent chemotherapy + 3D conformal radiotherapy to the bladder (≥64 Gy; or ≥55 Gy in ≥2.5 Gy/fraction; or 39-45 Gy with salvage RC; TMT arm). We fit a flexible logistic regression model for treatment to estimate the propensity score, and thenused inverse probability of treatment weights (IPWs) to evaluate the associations of treatment group with overall survival (OS). Results : A total of 2,048 patients were included, of whom 1,812 underwent RC and 236 underwent TMT. After IPW-adjustment, baseline characteristics were balanced. Median follow-up was 29.0 months, during which time 838 deaths occurred. The 5-year IPW-adjusted OS was 53% for RC and 44% for TMT (p=0.42; Figure 1 ). Compared to RC, TMTwas not associatedwith a statistically significant difference inOS (HR0.97; 95%CI 0.64-1.19; p=0.40).When examining heterogeneity of treatment effects according to cT stage, age, andCharlson score, RCappeared tobe associatedwith improvedOSonly for patientswithcT3disease (HR 0.42, p=0.01; Figure 2 ). Similar resultswere observed in sensitivity analyses. Conclusions : In observational analyses designed to emulate the SPARE trial, there was no statistically significant difference in OS between RC and TMT. Heterogeneity of treatment effects suggested improved survival with RC only for cT3 disease. Pre-Nephroureterectomy Diagnosis of Low-Grade Urothelial Carcinoma Does Not Predict Low Grade Disease on Final Pathology Mohammad H. Hout, MdD , Borivoj Golijanin, BS, Chris Tucci, MS, RN, Frances Kazal, BS, Timothy K. O’Rourke, Jr, MD, David Sobel, MD, Gyan Pareek, MD, Dragan Golijanin, MD Brown University, Providence, RI, USA Introduction: Upper tract urothelial carcinoma (UTUC) treatment depends on stage and grade of the disease. The gold standard of treatment is radical nephroureterectomy. In recent years UTUC treatment trends for low grade (LG) disease have shifted more towards minimally invasive and endoscopic approaches. Due to the potentially aggressive nature of UTUC, there is a risk of undertreatment especially if high grade (HG) disease is not confirmed on endoscopic biopsies. We sought to explore the risk of upgrading of UTUC, pathological, and long-term outcomes. Materials & Methods: A retrospective analysis of nephroureterectomy for UTUC cases performed at our hospital system from 1/1/2006 - 12/31/2020 was completed. Clinicopathologic features of patients were collected. Pre- operative pathology and diagnostic methods were analyzed, and descriptive statistics were summarized. Paired, nominal data of pre-operative and post- operative grading were compared using McNemar’s test. All analyses were completed using SPSS Version 26 (IBM Corp, Armonk NY). Results: 97 patients were included. 68/97 (70%) of patients were diagnosed with UTUC pre-operatively via endoscopic biopsy. 11 (11%) were diagnosed visually by endoscopy and 14 (14%) were diagnosed by cross sectional imaging. Of the 68 patients with biopsies, 37 (54%) were LG and 31 (46%) were HG. Of all patients with preop LG UTUC 25/37 (68%) were upgraded to HG UTUC (p<.0001) on final pathology. A total of 56/68 (84%) patients hadHGUTUC on final pathology. Of patients with upgraded final pathology (n=25), 16 (64%) were cT1, 8 (32%) cT2, and 1 (4%) was cT3 and were changed to 10 (40%) pTa, 7 (28%) pT1, 2 (8%) pT2, 5 (20%) pT3, and 1 (4%) pT4. Of upgraded patients with LG and HG on final pathology 2 (16%) and 11 (44%) ultimately passed away, respectively. Type of endoscopic biopsy device was not associated with a difference in LG biopsy upgrading. Conclusions: Management of LG UTUC on endoscopic biopsies carries significant risk due to potential of undergrading. Caution is highly advised when selecting patients for minimally invasive or endoscopic management even if adequate pathology specimen has been harvested and results in LG disease. Apre-nephroureterectomy diagnosis of LGUTUC is a poor predictor of final pathology. Scientific Session I: Oncology I 2

RkJQdWJsaXNoZXIy OTk5Mw==