Scholarlink

Objective <p>To develop a risk prediction model for radiation dermatitis in cancer patients during radiotherapy, validate its predictive performance, and provide a theoretical basis for reducing the incidence of radiation dermatitis in clinical practice.</p> Methods <p>A prospective study design was adopted. A convenience sampling method was used to select 296 cancer patients who received radiotherapy at Fudan Tumor Minhang Branch from January 2023 to January 2025 as the modeling group. Based on the occurrence of radiation dermatitis, patients were divided into a radiation dermatitis group (<i>n</i> = 150) and a non-radiation dermatitis group (<i>n</i> = 146). Data from the two groups were compared, and a risk prediction model was constructed based on the results of logistic regression analysis. The predictive performance of the model was evaluated using the area under the receiver operating characteristic (ROC) curve. From February to June 2025, 126 cancer patients receiving radiotherapy were selected as the validation group for external validation of the model.</p> Results <p>Among the 296 patients in the modeling group, the mean age of the radiation dermatitis group was 61.45 ± 9.67&#xa0;years, while that of the non-radiation dermatitis group was 57.06 ± 10.78&#xa0;years. The incidence of radiation dermatitis was 50.67% (150/296) in the modeling group and 42.9% (54/126) in the validation group. The final predictive variables included in the model were age (OR = 1.030, 95% CI 1.004–1.057), total radiation dose (OR = 1.035, 95% CI 1.002–1.069), body mass index (BMI) (OR = 1.255, 95% CI 1.141–1.380), serum retinol (OR = 1.035, 95% CI 1.016–1.054), and serum albumin (OR = 3.078, 95% CI 1.759–5.385) (all <i>P</i> &lt; 0.05). The area under the ROC curve for the prediction model was 0.792 (95% CI 0.741–0.843), with a sensitivity of 0.80, specificity of 0.78, and Youden index of 0.58. Validation results showed an area under the ROC curve of 0.768 (95% CI: 0.710–0.826), a maximum Youden index of 0.56, sensitivity of 0.78, specificity of 0.78, and an accuracy of 78.6%, indicating good predictive performance.</p> Conclusion <p>The risk prediction model for radiation dermatitis in cancer patients during radiotherapy developed in this study demonstrates good predictive performance. It can effectively predict the incidence of radiation dermatitis during radiotherapy and provides a reference for clinical staff in designing and implementing intervention measures.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Development and validation of a predictive model for radiation dermatitis risk in cancer patients during radiotherapy

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月16日

查看全文
  • Fengrong Tang,
  • Aihong Wan,
  • Lei Zhang,
  • Xiaochen Mo

摘要

Objective

To develop a risk prediction model for radiation dermatitis in cancer patients during radiotherapy, validate its predictive performance, and provide a theoretical basis for reducing the incidence of radiation dermatitis in clinical practice.

Methods

A prospective study design was adopted. A convenience sampling method was used to select 296 cancer patients who received radiotherapy at Fudan Tumor Minhang Branch from January 2023 to January 2025 as the modeling group. Based on the occurrence of radiation dermatitis, patients were divided into a radiation dermatitis group (n = 150) and a non-radiation dermatitis group (n = 146). Data from the two groups were compared, and a risk prediction model was constructed based on the results of logistic regression analysis. The predictive performance of the model was evaluated using the area under the receiver operating characteristic (ROC) curve. From February to June 2025, 126 cancer patients receiving radiotherapy were selected as the validation group for external validation of the model.

Results

Among the 296 patients in the modeling group, the mean age of the radiation dermatitis group was 61.45 ± 9.67 years, while that of the non-radiation dermatitis group was 57.06 ± 10.78 years. The incidence of radiation dermatitis was 50.67% (150/296) in the modeling group and 42.9% (54/126) in the validation group. The final predictive variables included in the model were age (OR = 1.030, 95% CI 1.004–1.057), total radiation dose (OR = 1.035, 95% CI 1.002–1.069), body mass index (BMI) (OR = 1.255, 95% CI 1.141–1.380), serum retinol (OR = 1.035, 95% CI 1.016–1.054), and serum albumin (OR = 3.078, 95% CI 1.759–5.385) (all P < 0.05). The area under the ROC curve for the prediction model was 0.792 (95% CI 0.741–0.843), with a sensitivity of 0.80, specificity of 0.78, and Youden index of 0.58. Validation results showed an area under the ROC curve of 0.768 (95% CI: 0.710–0.826), a maximum Youden index of 0.56, sensitivity of 0.78, specificity of 0.78, and an accuracy of 78.6%, indicating good predictive performance.

Conclusion

The risk prediction model for radiation dermatitis in cancer patients during radiotherapy developed in this study demonstrates good predictive performance. It can effectively predict the incidence of radiation dermatitis during radiotherapy and provides a reference for clinical staff in designing and implementing intervention measures.