Standard Versus Extended Lymph Node Dissection at the Time of Radical Cystectomy for Bladder Cancer: Emulation of a Clinical Trial
Alejandro Abello, MD MPH1, Sumedh Kaul, MS1, Aaron Fleishman, MPH1, Joaquim Bellmunt, MD MPH1, Irving Kaplan, MD1, Simon Kim, MD2, Peter Chang, MD MPH1, Andrew Wagner, MD1, Ruslan Korets, MD1, Aria Olumi, MD1, Boris Gershman, MD1.
1Beth Israel Deaconess Medical Center, Boston, MA, USA, 2University of Colorado, Aurora, CO, USA.
BACKGROUND: It is uncertain whether lymphadenectomy (LND) provides a survival benefit in patients undergoing radical cystectomy (RC). In the only completed randomized trial on this topic - LEA AUO AB 25/02 - extended LND (eLND) was associated with improved survival compared to limited/standard (sLND), although these associations did not reach statistical significance. Herein, we emulated a pragmatic clinical trial designed to resemble the LEA trial.
METHODS: We identified patients in the National Cancer Database who met the following eligibility criteria based on the LEA trial: adult 40-79 years old, Charlson 0-1, underwent RC with LND for high-grade cT1 / cT2-T4a cNany cM0 urothelial carcinoma of the bladder from 2006-2015, without neoadjuvant chemotherapy, at a hospital performing ≥16 RC/year. sLND and eLND were defined as removal of 4-11 and ≥12 lymph nodes based on the LEA trial. A propensity score (PS) was estimated for receipt of eLND, and the associations of LND type with overall survival (OS) were evaluated adjusting with inverse probability of treatment weights (IPW).
RESULTS: A total of 2248 patients formed the study cohort, including 436 with sLND and 1812 with eLND. Baseline characteristics were well-balanced after PS adjustment. During a median follow-up of 37.5 months, eLND was associated with significantly improved 5-year OS compared to sLND (60% vs 48%; HR 0.72, 95%CI: 0.61-0.85, p<0.01; Figure 1). Effect estimates were consistent across all potential treatment effect modifiers, including cT stage, cN stage, and age (Figure 2). Results were robust in sensitivity analyses that modified the definitions for LND and relaxed the annual RC hospital volume requirement.
CONCLUSIONS: In observational analyses designed to emulate a completed clinical trial, eLND was associated with improved OS compared to sLND among patients undergoing RC. Survival and effect estimates were similar to those in the LEA trial but statistically significant due to larger sample size. 
Back to 2021 Abstracts