Session Item

There is emerging evidence that cardiac dose has a negative impact on cardiac events after thoracic radiotherapy (RT). We studied the association between cardiac substructure dose and cardiac hospitalization within six months post-RT.

2,037 lung cancer patients treated with curative-intent RT (55Gy in 20 fractions) were analyzed. Cardiac hospitalization within six months post-RT as well as medical cardiac history prior to RT including ischemic heart disease, arrhythmia, heart failure, pericardial, myocardial, valve and artery disease were obtained from relevant WHO-ICD10 codes from Hospital Episode Statistics. Frequency of cardiac hospitalization was categorized as no, low (1-2), and high (3+) number of admissions. For each patient, planning CT scan was deformably registered to 5 template patients with 14 cardiac substructures segmented. Then, mean, and maximum dose to each substructure were calculated and averaged over 5 templates. To assess associations between cardiac hospitalization and 14 cardiac substructure doses, elastic-net regressions were performed with 1000 bootstraps. Structures with highest selection frequency (>60%) were combined into one structure and mean dose for this combined structure calculated. Associations between priori selected variables (including sex, age, tumor volume, cancer type, N stage, performance status, baseline cardiac admission, and mean lung dose), and dose to combined structure were tested against 3 related outcomes: cardiac hospitalization, number of cardiac hospitalizations, and frequency of cardiac hospitalization using logistic, linear, and ordinal logistic regression, respectively.

From 2,037 patients analyzed, 472 (23%) were hospitalized for cardiac causes at least once within six months post-RT. Of these, 155 (33%) did not have any history of cardiac admission prior to RT. The numbers of no, low, and high cardiac hospitalization were 1565 (77%), 303 (15%), and 169 (8%), respectively. Maximum number of hospitalizations within six months was 12. Combined structure was defined as: right atrium, right coronary artery, left anterior descending artery, aortic valve, and left main coronary artery (all selected >60%) with average dose of 8.8Gy (range:0.51-51.7Gy). From 3 multivariable analyses, higher risk of cardiac hospitalization, number of hospitalizations, and frequency of hospitalization was associated with being male, older, history of cardiac hospitalization prior to RT, worse performance status, larger tumor, and higher dose to the combined structure (Table 1).

Our results suggest the possibility that early cardiac hospitalization post-RT can be minimized by prioritizing dose reduction to the combined structure and provides a further rationale to avoid dose to the base of the heart, where these structures are located. This is so far the biggest single cohort study linking substructure dose to cardiac related hospitalization post-RT. Patients with history of cardiac admission should be monitored closely.