Development and Validation of a Nomogram Prediction Model for Moderate-to-Severe Acute Radiation Dermatitis in Patients with Breast Cancer: A Retrospective Study.

Abstract

Aims/Background Acute radiation dermatitis is the most common complication of radiotherapy in patients with breast cancer, with mild severity relieved by symptomatic treatment and moderate-to-severe severity leading to compromised skin integrity and affecting the patient's quality of life. Therefore, this study aims to develop a prediction model for moderate-to-severe acute radiation dermatitis in patients with breast cancer to reduce its severity. Methods A retrospective analysis of 713 patients receiving radiotherapy for breast cancer at the Affiliated Cancer Hospital of Xinjiang Medical University from January 2019 to December 2023 was conducted, with January 2019 to December 2021 serving as the training group (497 patients) and January 2022 to December 2023 serving as the validation group (216 patients). Patients in the training group were classified as having mild (383 patients) or moderately severe (114 patients) acute radiation dermatitis. Binary logistic regression was used to analyze the independent effects on moderately severe acute radiation dermatitis in patients with breast cancer, and a predictive model of the bar-folding plot was constructed and validated. Results Univariable analysis revealed that age, body mass index, targeted therapy, oral tamoxifen use, hyperlipidemia, diabetes, positive regional lymph node metastasis, value-added index, and triple-negative breast cancer were factors influencing moderate-to-severe acute radiation dermatitis in patients with breast cancer. Multivariate analysis showed that body mass index, hyperlipidemia, diabetes, positive regional lymph node metastasis, and value-added index were independent influencing factors for moderate-to-severe acute radiation dermatitis in patients with breast cancer. A nomogram prediction model was constructed, and the area under the receiver operating characteristic curve of the model was 0.814 and 0.743 for internal and external validation, respectively. The calibration curve showed that the model predicted moderate-to-severe acute radiation dermatitis better, and the decision curve analysis curve showed that the model had a high clinical benefit. Conclusion This risk prediction model can predict moderate-to-severe acute radiation dermatitis in patients with breast cancer, and help clinical providers screen high-risk patients and reduce acute radiation dermatitis severity.