Model-based approach for quantitative estimates of skin, heart, and lung toxicity risk for left-side photon and proton irradiation after breast-conserving surgery
DOI:
https://doi.org/10.1080/0284186X.2017.1299218Abstract
Background: Proton beam therapy represents a promising modality for left-side breast cancer (BC) treatment, but concerns have been raised about skin toxicity and poor cosmesis. The aim of this study is to apply skin normal tissue complication probability (NTCP) model for intensity modulated proton therapy (IMPT) optimization in left-side BC.
Material and methods: Ten left-side BC patients undergoing photon irradiation after breast-conserving surgery were randomly selected from our clinical database. Intensity modulated photon (IMRT) and IMPT plans were calculated with iso-tumor-coverage criteria and according to RTOG 1005 guidelines. Proton plans were computed with and without skin optimization. Published NTCP models were employed to estimate the risk of different toxicity endpoints for skin, lung, heart and its substructures.
Results: Acute skin NTCP evaluation suggests a lower toxicity level with IMPT compared to IMRT when the skin is included in proton optimization strategy (0.1% versus 1.7%, p < 0.001). Dosimetric results show that, with the same level of tumor coverage, IMPT attains significant heart and lung dose sparing compared with IMRT. By NTCP model-based analysis, an overall reduction in the cardiopulmonary toxicity risk prediction can be observed for all IMPT compared to IMRT plans: the relative risk reduction from protons varies between 0.1 and 0.7 depending on the considered toxicity endpoint.
Conclusions: Our analysis suggests that IMPT might be safely applied without increasing the risk of severe acute radiation induced skin toxicity. The quantitative risk estimates also support the potential clinical benefits of IMPT for left-side BC irradiation due to lower risk of cardiac and pulmonary morbidity. The applied approach might be relevant on the long term for the setup of cost-effectiveness evaluation strategies based on NTCP predictions.
