Effect of Dose to the Heart and Cardiac Substructures on Cardiac Toxicity after Breast Cancer Radiation

Abstract

Pre-existing cardiovascular disease, chemotherapy, and higher mean heart dose are known risk factors for cardiac toxicity after breast radiation therapy. However, the relationship between cardiac substructure radiation exposure and toxicity is not well understood. We hypothesized that mean heart dose is a surrogate for global cardiac radiation exposure but that more specific dosimetric thresholds for the heart and its substructures could be identified, which could be used to guide radiation planning for breast cancer patients in the future. In this cohort study, all breast cancer patients who received curative intent breast or chest wall radiotherapy at a single high-volume institution in 2014 and 2017 were included (n = 841). Baseline characteristics included hormone therapy, chemotherapy, menopausal status, diabetes, dyslipidemia, pre-existing cardiac toxicity, and age at diagnosis. Outcomes included any cardiac toxicity, arrhythmia, cardiomyopathy, ischemia, valvular, pericardial disease, and death. The heart and substructures, including left ventricle, right ventricle, left atrium, right atrium, aortic valve, pulmonic valve, mitral valve, tricuspid valve, and left anterior descending artery, were delineated on the simulation CT for each patient. Dosimetric variables, including mean dose, max dose, and V1, 2, 3, 4, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, and 70 Gy for the heart and substructures (in cc) were extracted. For each dosimetric variable, multivariable logistic regression was performed using baseline covariates in addition to the single dosimetric variable. Patients with missing data values were excluded. Results were combined and False Discovery Rate p-value correction was performed. Multiple cardiac substructure dosimetric variables were associated with increased risk of mortality on multivariable analysis (p < 0.05). For left atrium and right atrium, V2, 3, 4, and 5 Gy were all significant. For right ventricle, mean dose, V1, 2, 3, 4, 5, 10, 15, 20, 25, 30, and 35 Gy were significant. For mitral valve, mean dose, max dose, V3, 4, and 5 Gy were significant. For tricuspid valve, mean dose was significant. For aortic valve, max dose, mean dose, V4, and V5 Gy were significant. For the whole heart, V1, 2, 3, 4, 5, 10, 15, 20, 25, 30, and 35 Gy were significant. We have identified multiple dosimetric variables for the heart and its substructures which were associated with increased risk of mortality after breast cancer radiation. In fact, for certain structures, there were multiple exposure thresholds which showed increased risk of toxicity, highlighting the complex relationship between substructure dose and outcomes. Further study into these relationships will identify the most critical cardiac substructure constraints that could be used in radiation treatment planning.