Diabetes and kidney cancer survival in patients undergoing nephrectomy: A Canadian multi-center, propensity score analysis

Abstract

Diabetes has been associated with worse survival outcomes in various malignancies; however, there are conflicting data in kidney cancer. Determining whether diabetes is associated with survival in kidney cancer may help guide treatment in a comorbid patient population. We used the Canadian Kidney Cancer information system database to identify patients undergoing partial or radical nephrectomy between 1989 and 2017 for localized renal cell carcinoma at 16 institutions across Canada. We derived inverse probability of treatment weights (IPTW) from a propensity score model based on various clinical, surgical, and pathological characteristics. We used Cox proportional hazard models to evaluate the association between diabetes and cancer-specific and overall survival, in the sample weighted by the IPTW. 4828 patients met inclusion criteria, of whom 948 (19.6%) were diabetic. Median follow-up in those without death was 26.6 months (interquartile range 9.7-53.8). Among the entire cohort, 901 deaths were from any cause, and 299 deaths from kidney cancer. Before propensity score methods, diabetics were older, more likely to have comorbidities and clear cell histopathology. After propensity score adjustment, all characteristics were balanced between groups (standardized difference <0.10). IPTW-adjusted Cox proportional hazard models demonstrated no significant association between diabetes and cancer-specific (hazard ratio 1.13, 95% confidence interval 0.78-1.62), or overall survival (hazard ratio 1.14, 95% confidence interval 0.94-1.38). Our multi-centre study found that diabetes and nondiabetics have similar survival following nephrectomy for kidney cancer.