Contemporary use trends and effect on survival of pelvic lymph node dissection for non-muscle-invasive bladder cancer

BACKGROUND: Patients diagnosed with non-muscle-invasive bladder cancer (NMIBC) who are at a very high risk of disease progression and failure of Bacillus Calmette-Guerin treatment are recommended to undergo immediate radical cystectomy (RC). The role and optimal degree of pelvic lymph node dissection (PLND) during RC for NMIBC patients, however, have not been well investigated. PATIENTS AND METHODS: The Surveillance, Epidemiology, and End Results (SEER) database was used to identify patients. Overall survival (OS) was assessed with the Kaplan–Meier technique. Multivariable Cox regression analysis was conducted to determine independent factors of OS. RESULTS: A total of 1,701 patients were identified in the SEER database from 2004 to 2015. Any level of PLND (>0 lymph nodes examined) was performed in 1,092 patients (64.2%). The median number of lymph nodes examined was 8 (interquartile range, 0–20) in T1, 0 (interquartile range, 0–11) in Ta, and 0 (interquartile range, 0–14) in Tia patients. Compared with non-PLND, any level of PLND improved OS in T1 but not in Ta or Tis patients. Compared to limited (1–9 lymph nodes examined) and non-PLND, extensive PLND (lymph nodes examined ≥10) resulted in better OS only in T1 patients (all p < 0.001, adjusted significance level = 0.017). PLND was identified as a independent protective factor for OS. CONCLUSION: Based on the SEER database, we found that PLND during RC led to better OS and extensive PLND was associated with better OS in T1 but not in Ta or Tis patients. The implementation of PLND was insufficient both in population proportions and scope.