A nomogram predicting disease-specific survival after nephrectomy for papillary renal cell carcinoma

Abstract

5091 Background: Whereas multiple nomograms have been developed to assess outcomes of patients with clear cell renal cell carcinoma, a model to assess prognosis of papillary renal cell carcinoma (PRCC) has not yet been developed. After data collection and slide review of a large cohort of patients, the aim of this study was to develop and to internally validate a nomogram for prediction of disease-specific survival for PRCC. Methods: Out of 2,687 patients who underwent surgery for a renal tumor between 1989 and 2008 at two institutions, 258 (10%) were found to have PRCC. H&E slides were reviewed by one uro-pathologist at each institution for papillary sub-type, tumor grade, microvascular invasion, sarcomatoid features, collecting system invasion and presence and extent of tumor necrosis. A nomogram was constructed as a graphical representation of significant variables of disease-specific survival in multivariate Cox proportional hazards regression analysis. The discrimination and calibration of the nomogram were assessed, both utilizing bootstrapping to obtain relatively unbiased estimates. Results: After a median follow-up of 35 months, 49 PRCC-related deaths (19%) had occurred. In univariate analysis, incidental detection, T, N, M stage, grade, microvascular invasion, collecting system invasion, papillary sub-type, sarcomatoid features, and necrosis were all associated with prognosis. Multivariate Cox proportional hazards analysis, however, identified incidental detection, T stage, M stage, microvascular invasion, and necrosis, but not papillary sub-type as independent prognostic factors of disease-specific survival. These variables formed the basis of the nomogram that predicted 5-year disease-specific survival probability. The nomogram predicted well, with a bootstrapped corrected concordance index of 0.93, and showed good calibration. Conclusions: A highly accurate tool utilizing basic clinical and pathological information for predicting disease-specific survival was developed specifically for PRCC. This tool should be helpful for identification of the subset of PRCC patients with aggressive clinical behavior, and may contribute to the ability to individualize postoperative surveillance and therapy. No significant financial relationships to disclose.