Objective: To determine the value of clinical and pathological parameters defining the Störkel score in order to predict outcomes of patients with surgically treated renal cell carcinoma (RCC).
Material and methods: A total of 834 consecutive patients having radical or partial nephrectomy were retrospectively reviewed. For each patient with RCC, the prognostic Störkel score was calculated according to the following variables: Robson stage, Thoenes nuclear grading, histological type, pattern of growth, and age. Based on the Störkel score, patients were divided into groups: those with good prognosis (GP), intermediate prognosis (IP), and poor prognosis (PP). Cancer-specific survival (CSS) and overall survival (OS) were estimated using the Kaplan-Meier method. The accuracy of prediction of CSS and OS with the Störkel score was analyzed using Kaplan-Meier analysis, proportional hazards regression, and graphic representation [(Kaplan-Meier curves, area under the curve (AUC)]. In 564 patients who were still alive, the median follow-up was 79 months (mean 84.8 months).
Results: In the GP, IP, and PP groups, CSS after 8 years was 86.7%, 75.6%, and 13.7%, respectively (p<0.001). In the multiple analysis, only the Robson stage and Thoenes nuclear grading independently predicted CSS. Accordingly, the prognostic accuracy of the Störkel score (CSS prediction: AUC=0.744, 95% CI=0.70-0.79) was not better than with a reduced model that included the Robson stage and grading only (CSS prediction: AUC=0.765, 95%CI=0.72-0.81).
Conclusions: Of all parameters included in the Störkel score, only the Robson stage and nuclear grading are significant prognostic factors. Hence, we recommend an accordant modification of the score with additional variables.