Author: Yufeng Cao, Arun Gopal, Kai Huang, Kai Wang 👨🔬
Affiliation: Department of Radiation Oncology, University of Maryland School of Medicine, Department of Radiation Oncology, University of Maryland Medical Center, University of Maryland, Baltimore 🌍
Purpose: The treatment of left-sided breast tumors poses significant concerns regarding the risk of radiation-induced damage to nearby organs, particularly the heart. In clinical practice, breath-hold and free-breathing techniques are commonly compared to evaluate cardiac sparing, requiring both plans to be optimized for each patient. We propose a novel distance metric to predict breath-hold and free-breathing heart dose differences based on the spatial relationship between the breast and the heart.
Methods: A cohort study was conducted involving 24 early-stage left breast cancer patients treated with tangent fields (40-43Gy in 15-16 fractions). Among these, 16 patients underwent treatment using the free-breathing (FB) technique, while 8 patients were treated using the breath-hold (BH) technique. To evaluate and compare the dose distribution, a novel distance metric was defined as follows: D = α (MinFB - MinBH)(Heart, Breast) + β (MaxFB - MaxBH)(Heart, Breast) α, β represent adjustable parameters that control the relative weighting of the minimum and maximum distances between the heart and the breastThis metric is designed to reflect the effective geometric separation between the heart and breath volumes that can be tuned to correlate with and predict cardiac dose as a function of patient anatomy and treatment technique.
Results: Our model demonstrated a high level of accuracy, correctly predicting the treatment technique for 14 out of 16 patients (87.5%) treated with FB and 7 out of 8 patients (87.5%) treated with BH. These results highlight the potential of the model in effectively distinguishing between techniques and offering valuable support in treatment planning and evaluation.
Conclusion: This novel metric serves as a valuable tool for analyzing and predicting dosimetric differences between techniques, contributing to the enhanced personalization of radiotherapy for left-sided breast cancer patients. Additionally,it streamlines the planning process as an early automated predictive tool to identify appropriate candidates for breath-hold radiotherapy.