The current tumor-node-metastasis (TNM)–staging system for urothelial carcinoma of the bladder (UCB) is based on the number and size of the largest positive lymph node (LN). The aggregate LN metastasis diameter (ALNMD) may better reflect the burden of metastatic disease and improve the ability to predict recurrence-free (RFS) and overall survival (OS).

Clinical characteristics and follow-up information of 134 patients with LN-positive UCB treated by radical cystectomy was modeled using Cox proportional hazards regression analysis to predict OS. Pathologic specimens were retrospectively reviewed by a single genitourinary pathologist unaware of treatment outcome to determine the greatest dimension of metastasis in all affected LN. The median follow-up of survivors was 23 months.

The median OS was 17 months; median LN density, 17%; and median number of LN removed, 14. ALNMD was a significant predictor of RFS and OS after adjusting for pathologic T stage, lymphovascular invasion, LN density, comorbidity, and extranodal extension (adjusted HR 1.1; P = .02), even when restricting the analysis to patients in whom 10 or more LN have been removed. The predictive accuracy of a model for OS that contained ALNMD was superior to the one without this parameter and the TNM-staging system (c-index 0.71 vs 0.67 vs 0.62).

ALNMD is a significant predictor of RFS and OS after adjusting for standard prognostic parameters among patients with LN-positive UCB and may be a useful parameter to include in future predictive nomograms and TNM-staging systems.