Diabetes risk in childhood cancer survivors: a population-based study

Background — With continued improvements in survival following childhood cancer, the long-term health consequences of cancer treatments are receiving greater focus. Diabetes is emerging as a potential late-effect of cancer therapy; however, the magnitude of risk has not been well established. The goal of our study was to evaluate the risk of diabetes in childhood cancer survivors in Ontario using validated population-based databases.

Methods — We used provincial administrative health databases to compare the rate of diabetes in survivors of childhood cancer and controls from the general population. Diabetes was measured using a validated algorithm. We used a cause-specific hazard regression model in which death and development of another cancer were treated as competing risks.

Results — We identified 10,438 1-year survivors of childhood cancer that were diagnosed before age 21 between January 1, 1990, and December 31, 2010. The mean follow up was 11.2 years (standard deviation, 6.9). Cancer survivors had a 55% increased rate of developing diabetes compared with matched controls (HR 1.51, 95% CI 1.28 to 1.78) in models adjusted for rurality and income. Individuals treated for cancer between ages 6 and 10 had the highest increased rates of diabetes (HR 3.89, 95% CI 2.26 to 6.68). Among cancer types, leukemia and lymphoma were associated with the greatest increased risk for diabetes.

Conclusions — Our study supports evidence of an increased risk for diabetes in survivors of childhood cancer. Future research is warranted for better identification of treatment-related risk factors for diabetes in this population.