Comparison of the UCLA Integrated Staging System and the Leibovich score in survival prediction for patients with nonmetastatic clear cell renal cell carcinoma.

OBJECTIVES

To directly compare the models-the UCLA-Integrated Scoring System (UISS) and the Leibovich models-using various survival endpoints. Several Phase III trials of adjuvant therapy in renal cell carcinoma (RCC) have been initiated after advances in targeted therapy. To select patients at high risk of relapse and mortality, 2 aforementioned prognostic models have been incorporated into these trials. These models have not been compared previously.

METHODS

A retrospective study of 355 patients with unilateral nonmetastatic clear cell RCC undergoing nephrectomy between 1990 and 2006 at the Singapore General Hospital was undertaken. Performance of the UISS and the Leibovich models, as well as corresponding trial inclusion criteria, was directly compared using log-likelihood statistics. Adequacy and concordance indices were also calculated. Study endpoints tested were overall survival (OS), cancer-specific survival (CSS), and disease-free survival (DFS).

RESULTS

Likelihood ratio testing demonstrated a significant benefit in prediction when adding the Leibovich model to the UISS model in all outcomes tested, with no benefit using the converse approach (OS: P=.002 vs P=.27; CSS: P=.0001 vs P=.57; DFS: P=<0.0001 vs P=.30). Benefit was seen primarily in disease-free survival when adding the Leibovich trial criteria to UISS trial criteria, with no benefit using the converse approach (OS: P=.16 vs P=.27; CSS: P=.17 vs P=.11; DFS: P=.01 vs P=.26).

CONCLUSIONS

Both the Leibovich model and trial criteria are superior to the UISS model and trial criteria, respectively, in estimating survival outcomes in patients with nonmetastatic clear cell RCC after nephrectomy.