Main Session
Sep
29
PQA 07 - Head and Neck Cancer, Lung Cancer/Thoracic Malignancies, and Nursing and Supportive Care
3573 - Statewide Analysis of Clinical and Dosimetric Predictors of Cardiac Hospitalization after Thoracic RT for NSCLC
Presenter(s)
Batoul Nasser, MD, MS - University of Michigan Hospital, Ann Arbor, MI
B. Nasser1, W. Wang2, M. Schipper1,3, N. Libbey2, T. Pflederer1,4, H. Yin1, J. Johnson-Olokesusi5, P. A. Paximadis6, G. W. Tormoen7, M. M. Matuszak1, and S. Jolly1,8; 1Department of Radiation Oncology, University of Michigan, Ann Arbor, MI, 2University of Michigan, Ann Arbor, MI, 3Department of Biostatistics, University of Michigan, Ann Arbor, MI, 4Department of Radiation Oncology, Veterans Affairs Ann Arbor Health System, Ann Arbor, MI, 5Department of Radiation Oncology, Ann Arbor, MI, 6Department of Radiation Oncology, Corewell Health South, St. Joseph, MI, 7West Michigan Cancer Cancer, Kalamazoo, MI, 8Michigan Radiation Oncology Quality Consortium Coordinating Center, Ann Arbor, MI
Purpose/Objective(s):
Cardiac morbidity remains a concern in patients with lung cancer receiving thoracic RT. We identified clinical and dosimetric factors associated with cardiac-related hospitalizations (CRH), hypothesizing that baseline cardiac comorbidities and heart dose metrics (mean heart dose [MHD], heart V5–V60) predict post-treatment cardiac events.Materials/Methods:
Data were obtained from a prospective, statewide quality consortium. 2084 eligible patients had stage I–III NSCLC and received definitive RT from 2018 to 2024 (1105 SBRT; 1949 non-SBRT). The primary endpoint was CRH. CRH included arrhythmia, congestive heart failure (CHF), myocardial infarction (MI), or other cardiac events. Cumulative incidence was estimated under a competing-risks framework, treating death and hospice entry as competing events. Multivariable models included age, sex, ECOG, stage, treatment modality, and baseline cardiac comorbidities. LASSO Fine-Gray models were used to select patient factors associated with risk of any CRH. Cardiac dose metrics were added to build a MV model. Sensitivity analyses excluded SBRT-treated patients.Results:
During a median follow-up of 14 months, 168 (8.1%) patients experienced > 1 CRH. The 1-, 2- and 3-year cumulative incidence of CRH was 6.6%, 9.3% and 13.3%. The most common diagnoses were arrhythmia (4.6%), CHF (2.2%), and MI (2.0%); categories were not mutually exclusive. Overall, the median MHD was 3.07Gy (0 to 32.72Gy), with median MHD was 0.26Gy (0.10 to 1.60Gy) in SBRT and 6.46Gy (2.60 to 11.40Gy) in non-SBRT treated patients. In univariable competing-risks analyses, older age, ECOG =2, and baseline cardiac comorbidity were associated with higher CRH risk. Pre-existing cardiac disease was the strongest individual predictor, including CHF (p<0.001), prior MI (p=0.002), and hypertension (p<0.001). Higher MHD was associated with increased risk (HR 1.03 per Gy, 95% CI 1.00-1.05, p=0.047). 1833 patients had complete covariate data. The final MV model of any CRH included age, COPD, supplemental oxygen use, ECOG, any baseline cardiac comorbidities and MHD (Table 1). The most significant predictors of any CRH included age, baseline cardiac comorbidities, with MHD demonstrating a borderline association (Table 1). Results were consistent after excluding SBRT patients.Conclusion:
Older age and baseline cardiac comorbidities are major predictors for CRH risk following thoracic RT for lung cancer. Higher MHD was associated with increased cardiac events independent of treatment modality. These findings support cardiac risk stratification and heart-sparing radiation techniques. Table 1: MV model for any CRH * Variables selected for inclusion via LASSO and not by p-value criteria.| Any CRH | ||
| HR (95% CI) | P-Value* | |
| Age (Years) | 1.02(1.00, 1.04) | 0.039 |
| COPD (Y vs N) | 1.37(0.93, 2.03) | 0.110 |
| O2 (Y vs N) | 1.18(0.75, 1.86) | 0.480 |
| ECOG (2+ vs 0/1) | 1.43(0.91, 2.23) | 0.120 |
| Any baseline cardiac comorbidity (Y vs N) | 2.34(1.43, 3.82) | <.001 |
| MHD (Gy) | 1.04(0.99, 1.07) | 0.057 |