Fitting the Relative Seriality Model on Cardiac Toxicity and Dose to Different Parts of Heart after Radiation Therapy for Stage III Non-Small Cell Lung Cancer
Abstract
Purpose
To determine which cardiac substructures are associated with radiation-induced cardiac symptoms following radiotherapy for stage III non-small cell lung cancer (NSCLC), and to estimate the corresponding dose–response relationships using a widely applied normal tissue complication probability (NTCP) model.
Methods
A total of 112 patients treated with 70–90 Gy across six prospective clinical trials were analyzed. Dose–volume histograms were generated for the left and right ventricles (LV, RV), left and right atria (LA, RA), left anterior descending artery (LAD), whole heart, and pericardium. Cardiac toxicity was categorized as ischemic, arrhythmic, pericardial, or a combination of these events. Baseline cardiac risk was evaluated using the WHO/ISH risk score. The Relative Seriality (RS) NTCP model was applied to fit the clinical data. Model performance was assessed using the area under the receiver operating characteristic curve (AUC) and odds ratios (ORs).
Results
After a median follow-up of 26 months post-radiotherapy, ischemic, arrhythmic, pericardial events and at least one cardiac event occurred in 6.3%, 10.7%, 8.0% and 22.3% of the patients, respectively. Combined cardiac events were correlated with dose to heart (AUC = 0.71, OR = 3.3 (1.1-9.3), p = 0.03) and LV (AUC = 0.68, OR = 3.6 (1.4-9.6), p = 0.01). The corresponding biological dose constraints were 31Gy for heart and 21Gy for LV, respectively. The RS model parameters for heart were D50 = 46.7 (32.7-74.7), gamma = 0.34 (0.21-0.49) and s = 0.92 (0.09-6.43), whereas for LV they were 59.9 (44.2-93.6), 0.35 (0.19-0.49) and 9.04 (4.97-63.26), respectively.
Conclusion
Specific cardiac substructures were identified as being significantly associated with distinct radiation-induced cardiac symptoms. The RS NTCP model demonstrated a statistically significant fit to the clinical data in most cases. The resulting parameter sets may be applied as NTCP-based objectives in radiotherapy treatment plan optimization.