External validation of a pre-operative prognostic nomogram for renal cell carcinoma in two patient populations: a retrospective cohort study.

The aim of this study was to establish the discriminating accuracy of Kanao's pre-operative nomogram for renal cell carcinoma in predicting cause-specific survival among representative patients who underwent nephrectomy. Patients originated from two centers: Chiba University Hospital (n= 151) and Chiba Cancer Center (n = 91). We validated the predictive accuracy, which was assessed using Harrell's concordance-index. The concordance-index values were 0.692 and 0.834 for Chiba University Hospital and Chiba Cancer Center, respectively, although it was 0.822 for the combined data sets. Results of external validation were different at each cohort. We constructed calibration plots of Kanao's nomogram and confirmed the tendency at each institution. Inconsistency of results among two centers makes it difficult to reach a valid conclusion. Therefore, the predictive accuracy of Kanao's nomogram was not settled. Clinicians need to confirm the predictive accuracy of Kanao's nomogram and construct calibration plots when applying this nomogram to different patient populations.