ECOG performance status 0 or 1 and symptom classification do not improve the ability to predict renal cell carcinoma-specific survival

Objectives: We tested and compared the improvement in prognostic ability related to the consideration of either ECOG performance status (ECOGPS) and/or symptom classification (S-CLASS) in renal cell carcinoma specific mortality (RCC-SM) predictions. Methods: Univariate and multivariate Cox regression analyses targeted RCC-SM in 2570 RCC patients treated with either partial or radical nephrectomy. The increment in predictive accuracy related to the addition of either ECOGPS, S-CLASS or both was quantified using Harrell's concordance index. Results: Follow-up ranged from 0.1 to 23 years (median 3.2) and 610 patients (23.7%) died of RCC. In multivariable analyses, ECOGPS and S-CLASS represented independent predictors of RCC-SM. The addition of ECOGPS to established RCC-SM predictors increased the predictive accuracy by 0.3% (p = 0.8) versus 0.6% (p = 0.5) for S-CLASS versus 0.6% (p = 0.5) for both. Conclusions: Neither ECOGPS nor S-CLASS improves the ability to predict RCC-SM. Therefore, these variables may be safely omitted when RCC-SM risk is quantified. (C) 2007 Elsevier Ltd. All rights reserved.