Validation of an integrated staging system toward improved prognostication of patients with localized renal cell carcinoma in an international population.

PURPOSE Outcome prediction for patients with renal cell carcinoma is based on a combination of factors. In this study a previously published clinical outcome algorithm based on 1997 T stage, Fuhrman grade and performance score is validated using an international database. MATERIALS AND METHODS A total of 1,060 patients from Nijmegen, the Netherlands (NN), MD Anderson (MDA) and University of California, Los Angeles (UCLA) who had localized renal cell carcinoma were evaluated for outcome prediction using a clinical outcome algorithm previously shown to stratify patients into low, intermediate and high risk groups. Validation was performed by comparing the 3 risk groups separately within the 3 centers as well as by comparing hazard ratios and concordance indices among the 3 centers. RESULTS Estimated disease specific survival rates at 5 years for the low risk groups were 94% (NN), 92% (MDA) and 93% (UCLA). The 5-year disease specific survival rates for the intermediate risk groups were 65% (NN), 73% (MDA) and 78% (UCLA), while the rates for the high risk groups were 40% (NN), 30% (MDA) and 48% (UCLA). The concordance indices for each of the databases were 79% (NN), 86% (MDA) and 84% (UCLA). CONCLUSIONS A clinical algorithm that uses only 3 prognostic variables (1997 T stage, Fuhrman grade and performance status) to stratify patients with localized renal cell carcinoma into 3 risk groups has been shown to be applicable to external databases. This algorithm may be useful for patient counseling, surveillance and identification of high risk patients for enrollment in clinical trials.