Impact of the extent of lymph node dissection on survival outcomes in clinically lymph node-positive bladder cancer.

OBJECTIVE: To determine the oncological impact of extended pelvic lymph node dissection (ePLND) vs standard PLND (sPLND) during radical cystectomy (RC) in clinically lymph node-positive (cN+) bladder cancer (BCa). PATIENTS AND METHODS: In this retrospective, multicentre study we included 969 patients who underwent RC with sPLND (internal/external iliac and obturator lymph nodes) or ePLND (sPLND plus common iliac and presacral nodes) with or without platin-based peri-operative chemotherapy for cTany N1-3 M0 BCa between 1991 and 2022. We assessed the impact of ePLND on recurrence-free survival (RFS) and the distribution of recurrences (locoregional and distant recurrences). The secondary endpoint was overall survival (OS). We performed propensity-score matching using covariates associated with the extent of PLND in univariable logistic regression analysis. The association of the extent of PLND with RFS and OS was investigated using Cox regression models. RESULTS: Of 969 cN+ patients, 510 were 1:1 matched on propensity scores. The median (interquartile range [IQR]) time to recurrence was 8 (4-16) months, and median (IQR) follow-up of alive patients was 30 (13-51) months. Disease recurrence was observed in 104 patients in the ePLND and 107 in the sPLND group. Of these, 136 (27%), 47 (9.2%) and 19 patients (3.7%) experienced distant, locoregional, or both distant and locoregional disease recurrence, respectively. When stratified by the extent of PLND, we did not find a difference in recurrence patterns (P > 0.05). ePLND improved neither RFS (hazard ratio [HR] 0.91, 95% confidence interval [CI] 0.70-1.19; P = 0.5) nor OS (HR 0.78, 95% CI 0.60-1.01; P = 0.06) compared to sPLND. Stratification by induction chemotherapy did not change outcomes. CONCLUSION: Performing an ePLND at the time of RC in cN+ patients improved neither RFS nor OS compared to sPLND, regardless of induction chemotherapy status. Pretreatment risk stratification is paramount to identify ideal candidates for RC with ePLND as part of a multimodal treatment approach.