王延虎, 肖韡, 方申存, 王慧驰, 洪梅, 柴林, 常远, 郑玮, 朱诺. 预测食管癌调强适形放疗致急性放射性皮炎的列线图模型的建立与验证[J]. 实用临床医药杂志, 2023, 27(1): 26-30, 59. DOI: 10.7619/jcmp.20222370
引用本文: 王延虎, 肖韡, 方申存, 王慧驰, 洪梅, 柴林, 常远, 郑玮, 朱诺. 预测食管癌调强适形放疗致急性放射性皮炎的列线图模型的建立与验证[J]. 实用临床医药杂志, 2023, 27(1): 26-30, 59. DOI: 10.7619/jcmp.20222370
WANG Yanhu, XIAO Wei, FANG Shencun, WANG Huichi, HONG Mei, CHAI Lin, CHANG Yuan, ZHENG Wei, ZHU Nuo. Establishment and validation of a Nomogram model for predicting acute radiation dermatitis caused by intensity modulated radiotherapy for esophageal cancer[J]. Journal of Clinical Medicine in Practice, 2023, 27(1): 26-30, 59. DOI: 10.7619/jcmp.20222370
Citation: WANG Yanhu, XIAO Wei, FANG Shencun, WANG Huichi, HONG Mei, CHAI Lin, CHANG Yuan, ZHENG Wei, ZHU Nuo. Establishment and validation of a Nomogram model for predicting acute radiation dermatitis caused by intensity modulated radiotherapy for esophageal cancer[J]. Journal of Clinical Medicine in Practice, 2023, 27(1): 26-30, 59. DOI: 10.7619/jcmp.20222370

预测食管癌调强适形放疗致急性放射性皮炎的列线图模型的建立与验证

Establishment and validation of a Nomogram model for predicting acute radiation dermatitis caused by intensity modulated radiotherapy for esophageal cancer

  • 摘要:
    目的 建立预测食管癌调强适形放疗致急性放射性皮炎(ARD)的列线图模型,并验证其预测能力。
    方法 选取行调强适形放疗的179例胸段食管癌患者为研究对象,根据患者发生3级及以上ARD情况分为≥3级ARD组和对照组。应用单因素分析、LASSO回归分析和多因素Logistic回归分析筛选食管癌调强适形放疗致≥3级ARD的预测因素,并据此建立列线图模型。
    结果 糖尿病、白蛋白水平 < 30 g/L、卡氏评分 < 80分、放疗剂量≥65 Gy和同期化疗是食管癌调强适形放疗致≥3级ARD的独立危险因素(P < 0.05)。模型验证结果显示,一致性指数(C-index)为0.771,校准曲线趋近于理想曲线;受试者工作特征(ROC)曲线的曲线下面积(AUC)为0.778(95%CI:0.738~0.818),12%~64%预测范围内模型净获益,表明该列线图模型具有良好的预测能力。
    结论 食管癌调强适形放疗致≥3级ARD的危险因素包括糖尿病、白蛋白水平 < 30 g/L、卡氏评分 < 80分、放疗剂量≥65 Gy和同期化疗,基于危险因素建立的列线图模型具有良好的预测能力。

     

    Abstract:
    Objective To establish a Nomogram model for predicting acute radiation dermatitis (ARD) caused by intensity modulated radiotherapy for esophageal cancer and verify its predictive ability.
    Methods A total of 179 thoracic esophageal cancer patients with intensity modulated radiation therapy were selected as the research objects, and they were divided into ≥ grade 3 ARD group and control group according to the incidence of ARD grade 3 or above. The predictors of ≥ grade 3 ARD caused by intensity modulated radiation therapy for esophageal cancer were screened by univariate, LASSO regression and multivariate Logistic regression analyses, and the Nomogram model was established.
    Results Diabetes, albumin level < 30 g/L, Karnofsky score < 80, radiation dose ≥ 65 Gy and concurrent chemotherapy were the independent risk factors of ≥ grade 3 ARD caused by intensity modulated radiation therapy for esophageal cancer (P < 0.05). The model verification results showed that the consistency index (C-index) was 0.771, and the calibration curve was close to the ideal curve; the area under the curve (AUC) of receiver operating characteristic (ROC) curve was 0.778 (95%CI, 0.738 to 0.818), and the net benefit of the model was in the prediction range of 12% to 64%, indicating that the Nomogram model had good prediction ability.
    Conclusion The risk factors of ≥ grade 3 ARD caused by intensity modulated radiation therapy for esophageal cancer include diabetes, albumin level < 30 g/L, Karnofsky score < 80, radiotherapy dose ≥ 65 Gy and concurrent chemotherapy, and the Nomogram model based on risk factors has good predictive ability.

     

/

返回文章
返回