LAUSR

OBJECTIVE To use econometric methods to assess comparative overall survival of patients with metastatic renal cell carcinoma (mRCC) managed with initial cytoreductive nephrectomy (CN) vs initial systemic therapy. Randomized data demonstrate improved survival for CN preceding cytokine-based therapy in mRCC. This benefit may be attenuated in the contemporary mRCC era given more effective systemic therapies. METHODS Patients over age 65 with mRCC from the Surveillance, Epidemiology, and End Results registries linked with Medicare claims from 2006 to 2011 were categorized by initial treatment. We applied sequential survival analysis methods to assess the association between initial CN and overall survival (OS) including Cox proportional hazards models, propensity scoring, and instrumental variable analysis to account for measured and unmeasured selection bias. RESULTS Of 537 patients analyzed, 190 had initial CN followed by targeted therapy and 347 had initial targeted therapy. Median OS in the initial CN group was 17.4 months (interquartile range 9.8-32.0), compared with 9.2 months (interquartile range 4.3-18.0) for initial targeted therapy. Cox proportional hazards analysis revealed initial CN was associated with improved OS (hazard ratio 0.50, 95% confidence interval [CI] 0.38-0.65). Propensity matching demonstrated a survival advantage for initial CN of 5.8 months (95% CI 1.9-9.7). Accounting for unmeasured confounding with instrumental variable analysis demonstrated a trend toward improved survival with initial CN (hazard ratio 0.29 [95% CI 0.08-1.00]). CONCLUSION Initial CN is associated with improved survival compared with initial systemic therapy in a contemporary population-based mRCC cohort.