Stage-specific impact of pelvic lymph node dissection on survival in patients with non-metastatic bladder cancer treated with radical cystectomy

OBJECTIVE To examine the effect of stage-specific pelvic lymph node dissection (PLND) on cancer-specific (CSM) and overall mortality (OM) rates at radical cystectomy (RC) for bladder cancer. METHODS Overall, 11 183 patients were treated with RC within the Surveillance, Epidemiology, and End Results database. Univariable and multivariable Cox regression analyses tested the effect of PLND on CSM and OM rates, after stratifying according to pathological tumour stage. RESULTS Overall, PLND was omitted in 25% of patients, and in 50, 35, 27, 16 and 23% of patients with respectively pTa/is, pT1, pT2, pT3 and pT4 disease (P < 0.001). For the same stages, the 10-year CSM-free rates for patients undergoing PLND compared with those with no PLND were, respectively, 80 vs 71.9% (P = 0.02), 81.7 vs 70.0% (P < 0.001), 71.5 vs 56.1% (P = 0.001), 43.7 vs 38.8% (P = 0.006), and 35.1 vs 32.0% (P = 0.1). In multivariable analyses, PLND omission was associated with a higher CSM in patients with pTa/is, pT1 and pT2 disease (all P <= 0.01), but failed to achieve independent predictor status in patients with pT3 and pT4 disease (both P >= 0.05). Omitting PLND predisposed to a higher OM across all tumour stages (all P <= 0.03). CONCLUSIONS Our results indicate that PLND was more frequently omitted in patients with organ-confined disease. The beneficial effect of PLND on cancer control outcomes was more evident in these patients than in those with pT3 or pT4 disease. PLND at RC should always be considered, regardless of tumour stage.