Purpose: To evaluate the prognostic significance of different detection modalities of renal cell carcinoma (RCC) in a large cohort of patients who had been previously submitted to surgery in two teaching hospitals in Italy.
Materials and methods: We reviewed the clinical records of 1446 patients who had been submitted to surgical treatment for RCC at the Departments of Urology of Padua (n=747) and Verona (n=699) from 1976 to 2000. Patients were classified into two groups according to the detection mode: symptomatic and incidental. The cancer-specific survival probability was estimated according to the Kaplan-Meier method. In order to compare the survival curves the log rank test was used. The predictive independent value of the variables was examined using the Cox proportional hazards model.
Results: Six hundred and thirty patients (43.6%) were treated for incidental RCC and 816 (56.4%) for symptomatic RCC. In the incidental group, the size (p<0.001), the pathological stage (p<0.001) and the nuclear grading (p<0.001) of tumors were lower than those causing symptoms. The 5-year and 10-year cancer-specific survival probability were 84% and 75% in the incidental group, and 66% and 54.5% in the symptomatic group (p<0.0001), respectively. At a multivariate analysis, the mode of detection was an independent predictive variable (H.R. 1.559), as well as pathological stage (H.R. 1.809), nuclear grading (H.R. 1.411), size <or=4 cm (H.R. 1.667), and venous involvement (H.R. 1.526).
Conclusion: In patients with RCC, the detection modality can predict the cancer-specific survival rate independently of tumor pathological stage and grading.