Critical analysis and validation of lymph node density as prognostic variable in urothelial carcinoma of bladder.

OBJECTIVE: To validate the prognostic relevance of lymph node density (LND) and identify its optimal cut-points in a large international multicenter series of patients treated with radical cystectomy (RC) for invasive bladder cancer. METHODS: From 1993 to 2005, 4,430 bladder cancer patients who underwent RC without neoadjuvant chemotherapy were reviewed; of these, 1,038 were pN+M0 disease and form the basis of this report. RESULTS: Median age of patients was 67 years with median follow-up in survivors of 33 months. Overall, 5-year DSS estimate was 36%. Median number of lymph nodes removed was 18 (IQR, 11-32), median number of positive lymph nodes was 2 (IQR, 1-5), and median LND was 14.3% (IQR, 6.67-33.3%). LND as continuous variable was a stronger prognostic factor for DSS in patients that underwent a more extensive PLND (P < 0.001). HR for inverse association of LND with DSS increased incrementally with increasing LND cut-points. Categorizing LND into quintiles revealed strong tertiary distribution of risk based on LND <6%, 6%-41%, or >41% with cumulative 5-year DSS of 47%, 36%, and 21%, respectively (P < 0.001). When patients were stratified by adjuvant chemotherapy, LND remains independently prognostic in patients who received adjuvant chemotherapy as well as those who did not. CONCLUSION: Lymph node density is prognostic in bladder cancer patients who undergo a more extensive PLND and remains prognostic even when adjuvant chemotherapy is used. Prognostic value of LND is best represented as a continuum of risk and LND <6% represents the best possible outcome in patients with nodal disease.