349 - Epicardial Fat as a Radiomic Predictor of Cardiac Outcomes following Thoracic Radiotherapy for Lung Cancer
Presenter(s)
Q. Jiao1, M. Bakhtiar2, C. E. Kehayias3, C. V. Guthier4, J. He3, A. Nikolova5, R. H. Mak6, and K. M. Atkins7; 1Harvard Medical School, Boston, MA, 2Harvard Radiation Oncology Program, Boston, MA, 3Department of Radiation Oncology, Dana-Farber Cancer Institute/Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, 4Dana-Farber Brigham Cancer Center, Boston, MA, 5Department of Cardiology, Cedars-Sinai Medical Center, Los Angeles, CA, 6Department of Radiation Oncology, Brigham and Women's Hospital/Dana-Farber Cancer Institute, Boston, MA, 7Department of Radiation Oncology, Cedars-Sinai Medical Center, Los Angeles, CA
Purpose/Objective(s): Epicardial fat is a cardiometabolic risk factor associated with pro-inflammatory changes that contribute to atherosclerosis and the risk of coronary events and heart disease. Given the heightened cardiovascular (CV) risk in patients with lung cancer, we explored the relationship between epicardial fat, baseline CV risk, and major adverse cardiovascular events (MACE) following thoracic radiotherapy (RT) for lung cancer.
Materials/Methods: This is a retrospective analysis of 696 patients with non-small cell lung cancer (NSCLC) treated with RT between 2001 and 2014. Epicardial fat volumes were generated using an in-house trained deep learning auto-segmentation algorithm, applied to pre-treatment diagnostic or RT computed tomography (CT) simulation scans and indexed to body surface area. MACE included myocardial infarction, unstable angina, heart failure hospitalization/urgent visit, coronary revascularization, or CV death. Wilcoxon rank-sum testing and area under the receiver operating characteristic curve (AUC) analyses were used to compare epicardial fat index by baseline CV risk factors and evaluate epicardial fat index as a predictor of CV risk factors and MACE, respectively. Fine-Gray regressions predicting MACE were performed (non-cardiac death as a competing risk).
Results: The median follow-up was 23.2 months (interquartile range [IQR]: 10.5-48.7). Median age at diagnosis was 65 years (IQR: 57-73), 49% female. At baseline, 52% percent of patients had hypertension, 49% had hyperlipidemia, 14% had diabetes, and 36% had coronary heart disease. Median epicardial fat index was 51.5 cm3/m2 (IQR: 36.5-73.0). Patients with baseline hypertension and hyperlipidemia had a significantly higher epicardial fat index (p<0.001). In the AUC analysis, epicardial fat index was modestly predictive of post-RT MACE (AUC = 0.62) and cardiac death (AUC = 0.61), including among individuals with no known cardiac history (AUC for MACE = 0.60). Adjusting for age, sex, baseline CV risk factors (hypertension, diabetes, arrhythmia, known coronary heart disease), RT technique, and the volume (V) of the left anterior-descending coronary artery (LAD) receiving 15Gy =10%, there was an increased risk of MACE among patients with epicardial fat index in the upper quartile (=75th percentile) (subdistribution hazard ratio 1.86, 95% confidence interval 1.10-3.16; p=0.022).
Conclusion: Among high CV risk patients with NSCLC treated with RT, pre-treatment epicardial fat index was significantly associated with baseline hypertension and hyperlipidemia, as well as the risk of MACE, even after adjusting for known cardiometabolic and CV risk factors. These findings suggest epicardial fat index may provide additional explanatory power and augment MACE prediction models after thoracic RT. Future studies are warranted to validate these findings in larger cohorts and explore optimal CT-based epicardial fat features that may help identify high CV risk patients for intensified risk mitigation.