2698 - Development and Validation of a Prediction Model for Major Adverse Cardiac Events in Patients with Hepatocellular Carcinoma Undergoing Stereotactic Body Radiation Therapy

H. I. Lee¹, J. Jung¹, J. H. Park¹, B. Cho¹, Y. Goh¹, E. K. Chie², H. C. Kang², K. S. Kim², J. Son², and S. M. Yoon¹; ¹Department of Radiation Oncology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea, Republic of (South), ²Department of Radiation Oncology, Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Korea, Republic of (South)

Purpose/Objective(s): Stereotactic body radiation therapy (SBRT) offers a promising curative treatment for hepatocellular carcinoma (HCC); however, the potential cardiac toxicity due to its proximity to the heart warrants investigation. This study aimed to develop and validate a prediction model for major adverse cardiac events (MACE) in HCC patients undergoing SBRT, assessing the impact of SBRT on cardiac toxicity. Materials/

Methods: We retrospectively identified 1893 HCC patients who received SBRT at two institutions between January 2011 and December 2022. One institution served as the development cohort (N=1473), and the other as the validation cohort (N=420). Thirteen clinical and 88 dosimetric parameters were evaluated using bootstrapped forward selection and area under the curve (AUC) analysis to identify significant predictors for a multivariable MACE prediction model. Based on these factors, we constructed the cardiac event index (CEI) model, categorizing patients into distinct risk groups. Model performance was assessed for discrimination, efficiency, and calibration.
Results: In the development cohort, 5.8% of patients experienced MACE, with 3.6% having coronary artery disease. In the validation cohort, the occurrence rates were 6.7% for MACE and 4.3% for coronary artery disease. Five parameters were selected as the optimal subset for predicting MACE and were incorporated into the CEI model using the equation: CEI = age score + hypertension + current smoking + (2 × history of cardiac disease) + (0.05 × heart-V5 [%]), which yielded an AUC of 0.770 for MACE and 0.750 for coronary artery disease. The CEI model effectively stratified patients into low-, intermediate-, and high-risk groups, with MACE incidences of 0.4%, 4.9%, and 22.8% respectively. The impact of heart-V5 on MACE was minimal in the low- and intermediate-risk groups but more pronounced in the high-risk group. The validation cohort confirmed the model performance with an AUC of 0.809 and 0.793 for MACE and coronary artery disease, respectively.
Conclusion: The CEI model demonstrated robust performance in both cohorts, revealing a significant, dose-dependent relationship between SBRT and cardiac events. Notably, the effect of SBRT intensified with increasing clinical risk, necessitating careful management for high-risk patients. This innovative model can inform evidence-based decisions on cardiac dose optimization in SBRT planning. The CEI model predicting MACE

Parameters	ß		OR (95% CI)		p-value
Age (per 10-year)	0.367		1.44 (1.08–1.97)		0.011
Hypertension	0.744		2.10 (1.25–3.28)		0.002
Current smoking	0.764		2.15 (1.29–3.25)		0.001
History of cardiac disease	1.521		4.58 (2.45–8.11)		<0.001
Heart-V5	0.037		1.04 (1.03–1.06)		<0.001
Groups	MACE incidence (%)		Performance
	Development	Validation	Development	Validation
Low-risk (CEI <1.5)	0.4%	0.8%	AUC: 0.770 Sensitivity: 97.7% Specificity: 89.3%	AUC: 0.809 Sensitivity: 96.4% Specificity: 85.0%
Intermediate-risk (CEI 1.5–3.0)	4.9%	4.1%
High-risk (CEI >3.0)	22.8%	23.4%