Journal of International Oncology ›› 2026, Vol. 53 ›› Issue (6): 339-345.doi: 10.3760/cma.j.cn371439-20251229-00055

• Original Article • Previous Articles     Next Articles

Factors influencing the occurrence of high-grade acute radiodermatitis in patients undergoing image guided radiation therapy after breast cancer surgery and the construction of a prediction model

Su Jialin, Zheng Jin, Zhao Canjun, Han Lijun, Zhang Jie, Wang Liping()   

  1. Department of Traditional Chinese MedicineTangdu Hospital,Air Force Medical UniversityXi'an 710038, China
  • Received:2025-12-29 Online:2026-06-08 Published:2026-06-05
  • Contact: Wang Liping E-mail:chenchenmama2022@163.com
  • Supported by:
    Key Research and Development Program of Shaanxi Province of China(2024SF-YBXM-459)

Abstract:

Objective To analyze and predict the influencing factors for patients with breast cancer who received image guided radiation therapy(IGRT)after surgery developing high-grade(grade Ⅲ and above)acute radiodermatitis, and to construct a prediction model. Methods A total of 455 patients who underwent IGRT for acute radiodermatitis after breast-conserving or modified radical mastectomy at Tangdu Hospital, Air Force Medical University from January 2023 to January 2025 were selected as the research subjects. The general clinical data of the patients with high-grade and low-grade(Ⅰ-Ⅱ)acute radiodermatitis were compared. Multivariate logistic regression analysis was used to identify the independent factors influencing the occurrence of high-grade acute radiodermatitis. The receiver operator characteristic(ROC)curve was drawn to evaluate the predictive performance of the each influencing factor and their combinations for the occurrence of high-grade acute radiodermatitis in patients. A prediction model was constructed based on the results of multivariate analysis. The calibration curve was used to verify the consistency between the occurrence probability and the predicted probability. The clinical decision curve analysis was conducted to assess the clinical application value of the prediction model. Results Among the 455 patients with breast cancer who underwent postoperative IGRT and developed acute radiodermatitis, 374 cases were of low-grade acute radiodermatitis, including 270 cases of grade Ⅰ and 104 cases of grade Ⅱ; 81 cases were of high-grade, including 75 cases of grade Ⅲ and 6 cases of grade Ⅳ. There were statistically significant differences in body mass index(BMI)(t=10.23, P<0.001), smoking history(χ2=9.69, P=0.002), diabetes(χ2=4.52, P=0.034), ferritin level(t=31.55, P<0.001), high-sensitivity C-reactive protein(hs-CRP)level(t=32.78, P<0.001), and CD3+ T cell level(t=46.50, P<0.001)between the high-grade and low-grade acute radiodermatitis patients. Multivariate logistic regression analysis showed that, BMI(OR=1.52, 95%CI:1.15-2.53, P=0.019), smoking history(OR=1.44, 95%CI:1.09-1.98, P=0.024), diabetes(OR=1.62, 95%CI:1.26-3.05, P=0.013), ferritin level(OR=1.78, 95%CI:1.49-3.55, P=0.009), hs-CRP level(OR=2.09, 95%CI:1.63-4.46, P<0.001), and CD3+ T cell level(OR=1.96, 95%CI:1.50-3.83, P=0.004)were the independent factors influencing the occurrence of high-grade acute radiodermatitis in the patients undergoing IGRT after breast cancer surgery. The ROC curve analysis showed that, the area under the curve(AUC)for BMI, smoking history, diabetes, ferritin level, hs-CRP level, and CD3+ T cell level alone in predicting the occurrence of high-grade acute radiodermatitis in the patients undergoing IGRT after breast cancer surgery were 0.69, 0.67, 0.71, 0.73, 0.77, and 0.74, respectively. The AUC of the combined prediction of these six factors was 0.88, indicating that the combined prediction was more valuable than individual predictions based on BMI, smoking history, diabetes, ferritin level, hs-CRP level, and CD3+ T cell level(Z=3.42, P=0.001; Z=3.35, P=0.001; Z=3.30, P=0.001; Z=2.00, P=0.025; Z=2.55, P=0.007; Z=2.10, P=0.019). The calibration curve showed that the actual occurrence probability of high-grade acute radiodermatitis in patients receiving IGRT after breast cancer surgery predicted by the nomogram prediction model was relatively consistent with the predicted probability, with a consistency index of 0.87(95%CI:0.84-0.89). The calibration curve fitted well(χ2=7.49, P=0.485), indicating high calibration accuracy of the model. The clinical decision curve showed that the prediction model exhibited good discriminative ability, indicating its potential clinical application value. Conclusions BMI, smoking history, diabetes, ferritin level, hs-CRP level, and CD3+ T cell level are the independent factors influencing the occurrence of high-grade acute radiodermatitis in the patients undergoing IGRT after breast cancer surgery. The nomogram prediction model constructed based on these influencing factors has certain clinical reference and application value.

Key words: Breast neoplasms, Radiotherapy, image-guided, Radiodermatitis, Root cause analysis