Risk of Symptomatic Stroke After Radiation Therapy for Childhood Cancer: A Long-Term Follow-Up Cohort Analysis

Purpose Long-term childhood cancer survivors are at high risk of late adverse effects, including stroke. We aimed to determine the cumulative incidence of clinically validated symptomatic stroke (transient ischemic attack [TIA], cerebral infarction, and intracerebral hemorrhage [ICH]) and to quantify dose-effect relationships for cranial radiation therapy (CRT) and supradiaphragmatic radiation therapy (SDRT). Methods and Materials Our single-center study cohort included 1362 survivors of childhood cancer that were diagnosed between 1966 and 1996. Prescribed CRT and SDRT doses were converted into the equivalent dose in 2-Gy fractions (EQD2 ). Multivariate Cox regression models were used to analyze the relationship between the EQD2 and stroke. Results After a median latency time of 24.9 years and at a median age of 31.2 years, 28 survivors had experienced a first stroke: TIA (n=5), infarction (n=13), and ICH (n=10). At an attained age of 45 years, the estimated cumulative incidences, with death as competing risk, among survivors treated with CRT only, SDRT only, both CRT and SDRT, and neither CRT nor SDRT were, respectively, 10.0% (95% confidence interval [CI], 2.5%-17.0%), 5.4% (95% CI, 0%-17.0%), 12.5% (95% CI, 5.5%-18.9%), and 0.1% (95% CI, 0%-0.4%). Radiation at both locations significantly increased the risk of stroke in a dose-dependent manner (hazard ratios: HRCRT 1.02 Gy−1 ; 95% CI, 1.01-1.03, and HRSDRT 1.04 Gy−1 ; 95% CI, 1.02-1.05). Conclusions Childhood cancer survivors treated with CRT, SDRT, or both have a high stroke risk. One in 8 survivors treated at both locations will have experienced a symptomatic stroke at an attained age of 45 years. Further research on the pathophysiologic processes involved in stroke in this specific group of patients is needed to enable the development of tailored secondary prevention strategies.