Cardiac Risk After Heart-Sparing Breast Radiotherapy

Radiotherapy for breast cancer exposes the heart to incidental radiation, and historical data have shown a dose-dependent increase in associated cardiac events. Although whole-heart dose metrics are commonly used for risk assessment, emerging evidence suggests that dose to the left anterior descending coronary artery (LAD) may better capture risk.

To compare the ability of heart and LAD radiation dose metrics to predict cardiac risk after breast radiotherapy.

This was a cross-sectional study of patients with breast cancer who were treated with either 3-dimensional conformal or intensity-modulated radiotherapy from 2008 to 2018 at a tertiary care center in Canada. The primary analysis included patients with left-sided breast cancer. Cardiac events were assessed using longitudinal follow-up data. Dosimetry was derived from computed tomography plans using automated segmentation and converted to equivalent dose in 2-Gy fractions (EQD2). Data were analyzed from September 2024 to April 2026.

Adverse cardiac events defined as myocardial infarction or hospital admission or emergency department visit for unstable angina (ie, acute coronary syndrome), arrhythmia, heart failure, pericarditis, or myocarditis. The incidence of coronary angiography and coronary revascularization was also captured as coronary artery disease (CAD). Discrimination for dose metrics was performed with receiver operator characteristic curves and competing-risks regression (Fine and Gray), adjusted for cardiovascular risk factors.

The analysis included 4908 patients with breast cancer of whom 2223 had left-sided breast cancer. During a median (IQR) follow-up period of 10.8 (8.4-13.1) years, cumulative incidence of cardiac event or CAD was 5.0 (95% CI, 4.1-6.0) at 10 years. A data-driven cut point analysis identified 12 Gy EQD2 as the maximum LAD dose that best stratified risk. Among patients with left-sided breast cancer, maximum LAD dose (concordance [C] index, 0.58; 95% CI, 0.52-0.64) discriminated better than mean heart dose (C index, 0.53; 95% CI, 0.47-0.60). In multivariable analysis, maximum LAD of 12 Gy or greater was independently associated with higher cardiac risk (subdistribution hazard ratio = 1.81; 95% CI, 1.04-3.16; P = .04), whereas mean heart dose of 2 Gy or greater was not associated (P = .99). For clinical context, the 12-Gy EQD2 for maximum LAD dose corresponds to a physical dose of approximately 10.5 Gy for 42.5 Gy in 16 fractions and 7 Gy for 26 Gy in 5 fractions.

In this cross-sectional study of heart-sparing breast radiotherapy, LAD dose was associated with cardiac events, whereas whole-heart metrics was not. These findings support LAD-based planning and respiratory motion management to reduce long-term cardiovascular risk in patients with breast cancer undergoing radiotherapy.