Understanding treatment disconnect and mortality trends in renal cell carcinoma using tumor registry data.

Abstract

403 Background: Previous research has demonstrated a rise in overall and cancer specific mortality over time despite increased detection and treatment of kidney cancer, termed “treatment disconnect.” Our aim was to assess the impact of missing data and rising incidence on mortality trends using national tumor registry data. Methods: Using Surveillance, Epidemiology, and End Results (SEER) data, we identified all patients with a kidney cancer diagnosis from 1973-2009. We estimated overall mortality rates by tumor size after accounting for lags from diagnosis to death and using multiple imputation for missing data. Mortality rates for all patients irrespective of tumor size were also estimated after adjustment for prior cumulative incidence using ridge regression. Results: 73,647 patients met inclusion criteria, of which 22.7% had missing tumor size data. Compared to observed rates, significant decreases in the change in mortality rate per 100,000 US population from 1983 to 2009 were noted for 2-4cm (Δobs1.58 versus Δimp1.34; p=0.003), 4-7cm (Δobs1.61 versus Δimp0.79; p<0.001), and >7cm (Δobs1.55 versus Δimp0.19; p<0.001) tumors following missing data imputation, while no differences in change of rate were noted for tumors <2cm (Δobs0.29 versus Δimp0.25). In a separate analysis, holding the incidence of kidney cancer constant over time, the temporal increase in overall mortality for all patients was attenuated (p<0.001 for test of difference in slope) when comparing observed (3.9 to 6.8) and estimated (5.9 to 7.1) rates, suggesting that rapidly rising incidence may also influence reported mortality trends. Conclusions: Missing data and rising incidence rates may contribute substantially to the “treatment disconnect” phenomenon when examining mortality rates in kidney cancer using tumor registry data.