A new staging system for locally advanced (pT3-4) renal cell carcinoma: a multicenter European study including 2,000 patients

PURPOSE: We provide an adequate prognostic stratification for locally advanced renal cell carcinoma and propose a new TNM classification. MATERIALS AND METHODS: We analyzed clinical and pathological data on a large series of patients undergoing radical nephrectomy for pT3-4 renal cell carcinoma at 12 European centers. Cancer specific survivals were estimated using the Kaplan-Meier method. The log rank test was used for comparing survival curves and for univariate analysis. The Cox proportional hazards regression model was used for multivariate analysis. RESULTS: The analysis included 1,969 patients. Median survivor followup was 49 months. Five-year cancer specific survival was 60% for pT3a, 46.2% for pT3b, 10% for pT3c and 12% for pT4 tumors (p <0.0001). According to median survival we identified 3 prognostic groups, including 1--patients with renal vein thrombosis (117 months), fat invasion (98 months) or infradiaphragmatic vena caval thrombosis (67 months), 2--patients with adrenal invasion alone (24 months), renal vein thrombosis plus fat invasion (24 months) or infradiaphragmatic vena cava plus fat invasion (24 months) and 3--patients with renal or infradiaphragmatic caval thrombosis plus adrenal involvement (11 months), supradiaphragmatic vena caval thrombosis (12 months) or Gerota's fascia invasion (12 months). Five-year cancer specific survival rates in groups 1 to 3 were 61%, 35% and 12.9%, respectively (p <0.0001). On multivariate analysis the proposed classification had an independent prognostic value.

PURPOSE: We provide an adequate prognostic stratification for locally advanced renal cell carcinoma and propose a new TNM classification. MATERIALS AND METHODS: We analyzed clinical and pathological data on a large series of patients undergoing radical nephrectomy for pT3-4 renal cell carcinoma at 12 European centers. Cancer specific survivals were estimated using the Kaplan-Meier method. The log rank test was used for comparing survival curves and for univariate analysis. The Cox proportional hazards regression model was used for multivariate analysis. RESULTS: The analysis included 1,969 patients. Median survivor followup was 49 months. Five-year cancer specific survival was 60% for pT3a, 46.2% for pT3b, 10% for pT3c and 12% for pT4 tumors (p <0.0001). According to median survival we identified 3 prognostic groups, including 1--patients with renal vein thrombosis (117 months), fat invasion (98 months) or infradiaphragmatic vena caval thrombosis (67 months), 2--patients with adrenal invasion alone (24 months), renal vein thrombosis plus fat invasion (24 months) or infradiaphragmatic vena cava plus fat invasion (24 months) and 3--patients with renal or infradiaphragmatic caval thrombosis plus adrenal involvement (11 months), supradiaphragmatic vena caval thrombosis (12 months) or Gerota's fascia invasion (12 months). Five-year cancer specific survival rates in groups 1 to 3 were 61%, 35% and 12.9%, respectively (p <0.0001). On multivariate analysis the proposed classification had an independent prognostic value. CONCLUSIONS: Our results suggest the necessity of reclassifying locally advanced renal cell carcinoma according to the 3 described prognostic categories.