Abstract
PURPOSE: To compare the difference in normal tissue complication probability (ΔNTCP) between proton therapy (PT) and photon therapy plans using radiation-induced lung injury (RILI) as an endpoint and to analyze its correlation with thoracic anatomic features in breast cancer patients. MATERIALS AND METHODS: A total of 409 breast cancer patients receiving photon intensity-modulated radiation therapy were randomly split into training and testing sets at an 8:2 ratio. A dose-modifying-factors (DMFs)-incorporated Lyman-Kutcher-Burman NTCP model was developed by maximum likelihood estimation with the training set to predict the risk of grade ≥ 1 RILI (CTCAE 5.0) within 1 year after radiotherapy. The DMFs stood for baseline risk factors were identified by least absolute shrinkage and selection operator regression and uni-multivariable logistic regression. After model validation, PT plans were generated for 80 patients from the dataset. The Pearson/Spearman rank correlation coefficient followed by linear regression was used to assess the correlation between anatomic features and lung ΔNTCP between photon and proton plans. RESULTS: BMI ≥ 23.52 kg/m² (P = .049) and interval between last cycle of chemotherapy and radiotherapy (ICR) ≤ 20 days (P = .014) were found to be independent risk factors for RILI. The optimal NTCP parameters were: n = 0.40, m = 0.22, TD50 = 24.66Gy, DMF-BMI = 0.88, and DMF-ICR = 0.92. The model performed well in area under the receiver operating curve (training set 0.754, testing set 0.733) and other validation tests. Among the 80 patients with photon and proton plans, the mean ΔNTCP was 57.45% ± 10.51%. Linear regression showed a significant positive correlation between Arc Height to Base Ratio at the transverse plane of the sternal angle and ΔNTCP (regression coefficient 56.56, P = .049). CONCLUSION: BMI ≥ 23.52 kg/m² and ICR ≤ 20 days are risk factors for RILI. In patients with larger Arc Height to Base Ratio at the transverse plane of the sternal angle, PT plans suggest greater lung sparing in comparison to intensity-modulated radiation therapy. Further studies are needed to validate this association.