Abstract
BACKGROUND: Radiodermatitis is the predominant acute toxicity in locally advanced breast cancer (BC) radiotherapy. Early-onset radiodermatitis substantially increases the risk of high-grade injury and potential treatment interruption. This study aimed to identify risk factors for early-onset radiodermatitis by analyzing patient characteristics and dose-distribution profiles from step-and-shoot intensity-modulated radiotherapy (IMRT) and volumetric-modulated arc therapy (VMAT). METHODS: This retrospective analysis included 128 women with locally advanced unilateral BC who received postoperative IMRT or VMAT. Patient demographics, treatment parameters, and three-dimensional dose-volume metrics were extracted. Early-onset radiodermatitis was defined as CTCAE v5.0 grade ≥ 1 developing before the 20th treatment fraction. Univariable and multivariable logistic regression identified independent risk factors, and dosimetric variables were compared between IMRT and VMAT. RESULTS: Early-onset radiodermatitis was observed in 35 of 128 patients (27.3%). Four independent predictors were identified: treatment technique (IMRT versus VMAT: odds ratio [OR] 4.25; 95% confidence interval [CI] 1.30-13.94; p = 0.017), left-sided irradiation (OR 20.98; 95% CI 2.39-184.27; p = 0.006), heart V(5) (OR 0.85 per 1% increase; 95% CI 0.74-0.97; p = 0.012), and PTV(Breast) volume (OR 1.002 per cc; 95% CI 1.000-1.004; p = 0.023). VMAT provided superior target coverage and reduced high-dose exposure to the heart and ipsilateral lung (V(40)), whereas IMRT better limited low-dose spill to the contralateral breast and lung and to the ipsilateral lung (V(5)). CONCLUSIONS: IMRT, left-sided irradiation, low heart V(5), and large PTV(Breast) volume were independent risk factors for early-onset radiodermatitis. Awareness of these factors can guide prophylactic skin care and adaptive planning. While VMAT provides advantages in target coverage and cardiac/high-dose lung sparing against IMRT's superior control of low-dose exposure to normal tissues.