Coronary heart disease (CHD) and stroke risk associated with breast cancer (CA) irradiation (RT) based on patient baseline cardiac risk profile

Abstract

606 Background: Although population-based studies showed increased fatal and non-fatal CHD risk associated with left-sided breast RT, presumably from “bystander” cardiac dose, how RT interacts with underlying cardiac risk factors to influence these outcomes is unclear. We used data from the Nurses’ Health Study (NHS) to examine the effect of RT on fatal and non-fatal CHD, stratified by baseline CHD risk factors. Methods: 121,700 female registered nurses enrolled in 1976. Ongoing biennial medical information collected includes CHD risk factors, cardiac events, CA diagnosis, treatment, and cause of death. Medical record review confirms CA diagnoses and cardiac events. Numerous publications establishing the causes of CHD and cardiac mortality have resulted. 3863 women have developed stage 0-III breast CA and received RT. Proportional hazards models were used to assess the association between left-sided RT and CHD development, after adjusting for age and CHD risk factors. Results: During 38,538 person-years follow-up after breast CA diagnosis, 126 CHD events developed. In age- adjusted univariate models, left-sided RT (vs. right-sided), diabetes, parental CHD before age 60, and smoking were significantly associated with fatal and non-fatal CHD; aspirin and postmenopausal hormone use were protective. Crude incidence of CHD after left-sided RT was 396 per 100,000 person-years, compared to 259 after right-sided RT. A multivariable model that accounted for the above plus other risk factors (hypertension, hyperlipidemia, physical activity, alcohol use) and stage showed that left-sided RT was independently associated with CHD development (rate ratio, RR 1.5, 95% CI 1.1–2.2). Stratified analyses by the presence of hypertension, hyperlipidemia, and smoking showed similar RR from left-sided RT in all subgroups. In contrast, left-sided RT did not affect risk of stroke in multivariable analysis (RR 0.7, 95% CI 0.4–1.2). Conclusions: Left-sided RT is an independent CHD risk factor, and indicates a similar RR to patients in all subgroups of baseline CHD risk, though absolute risk increase was larger in women who had additional risk factors. Breast CA survivors after left-sided RT may need long-term cardiac follow-up and management. No significant financial relationships to disclose.