Purpose: To compare the utility of lymph node density (LND) with TNM nodal status in predicting disease-specific survival (DSS) after radical cystectomy.
Patients and methods: We identified 248 patients with nodal metastasis after radical cystectomy (without neoadjuvant chemotherapy): 162 patients from Memorial Sloan-Kettering Cancer Center (MSKCC) and 86 patients from M.D. Anderson Cancer Center (MDACC). We assessed the effect of several variables on DSS.
Results: After a median follow-up duration of 24 months, 134 patients died of their disease. The median DSS was 36 months, and the 1-year, 2-year, and 5-year DSS rates were 83.7%, 57.4%, and 36.6%, respectively. The median LND was 20%. The 5-year DSS rate was 54.6% for patients with LND < or = 20% v 15.3% for patients with LND higher than 20% (P < .01). Pathologic nodal (pN) status in patients was 78 for pN1 (32%), 127 for pN2 (51%), and 43 for pN3 (17%). On univariate analysis, pN status and LND were significant predictors of DSS (P < .01). However, when pN status and LND were considered jointly in a multivariate model, only LND higher than 20% predicted decreased DSS (hazard ratio [HR], 2.75; P < .01). In addition, while nonorgan-confined (ie, > pT2) primary tumor (HR, 2.40; P < .01) and adjuvant chemotherapy (HR, 0.47; P < .01) were predictors of DSS, LND remained a predictor of DSS even after accounting for adjuvant chemotherapy.
Conclusion: LND is superior to TNM nodal status in predicting DSS for patients with lymph node-positive disease after radical cystectomy, even in the context of adjuvant chemotherapy.