食管癌放疗患者发生中重度急性放射性食管炎风险因素分析及预测模型的建立  

Analysis of risk factors of moderate and severe acute radiation esophagitis in esophageal cancer patients underwent radiotherapy and construction of prediction model

作　　者：叶海主 彭小保 袁华琴 胡宏霞 陈文强 YE Haizhu;PENG Xiaobao;YUAN Huaqin;HU Hongxia;CHEN Wenqiang(Department of Oncology,Nanjing Gaochun People's Hospital,Jiangsu Nanjing 211300,China)

机构地区：[1]南京市高淳人民医院肿瘤科,江苏南京211300

出　　处：《现代肿瘤医学》2023年第16期3032-3037,共6页Journal of Modern Oncology

基　　金：江苏大学医学临床科技发展基金项目(编号:JLY20160187)。

摘　　要：目的:分析食管癌患者发生中重度急性放射性食管炎(MSARE)的风险因素,并构建风险预测列线图模型。方法:收集2019年01月至2022年09月在我院行放射治疗的82例食管癌患者,根据有无MSARE的发生分为发生组(32例)和未发生组(50例)。采用单因素及多因素Logistic回归分析筛选发生MSARE的独立危险因素,纳入这些因素构建列线图预测模型,并对模型的区分度、校准度及临床实用性进行验证。结果:多因素Logistic分析结果表明肿瘤位置(下段:OR=0.1,95%CI:0.01~0.9,P=0.04)、N分期(N_(2)+N_(3):OR=3.71,95%CI:0.99~13.88,P=0.049)、糖尿病(OR=6.63,95%CI:1.48~29.67,P=0.013)、高血压(OR=5.19,95%CI:1.22~22.16,P=0.026)、同步化疗(OR=5.86,95%CI:1.43~23.99,P=0.014)、食管照射长度(OR=1.33,95%CI:1.07~1.65,P=0.011)均为MSARE发生的独立危险因素,利用这些因素成功构建列线图模型,ROC曲线下面积(AUC)(0.888,95%CI:0.815~0.961)表明模型预测准确度高;校准曲线与理想曲线接近,模型校准能力较好;决策曲线分析(DCA)提示阈值大约为0.06~1具有模型的最大收益范围,模型具有临床实用性。结论:成功构建食管癌患者发生MSARE风险预测模型,模型效能良好,有助于临床医师对高风险MSARE患者的早期识别,从而早期预防及改进治疗计划。Objective:To analyze the risk factors of moderate and severe acute radiation esophagitis(MSARE)in patients with esophageal cancer and construct a risk prediction nomogram model.Methods:From January 2019 to September 2022,82 patients with esophageal cancer who underwent radiotherapy in our hospital were divided into occurrence group(32 cases)and non-occurrence group(50 cases)according to the presence or absence of MSARE.Univariate and multivariate Logistic regression analysis was used to screen independent risk factors for the occurrence of MSARE.These factors were used to construct a nomogram prediction model and validate the discrimination,calibration,and clinical utility of the model.Results:Multivariate Logistic analysis showed tumor location(lower:OR=0.1,95%CI:0.01~0.9,P=0.04),N stage(N_(2)+N_(3):OR=3.71,95%CI:0.99~13.88,P=0.049),diabetes(OR=6.63,95%CI:1.48~29.67,P=0.013),hypertension(OR=5.19,95%CI:1.22~22.16,P=0.026),concurrent chemotherapy(OR=5.86,95%CI:1.43~23.99,P=0.014)and length of esophageal irradiation(OR=1.33,95%CI:1.07~1.65,P=0.011)were independent risk factors for the development of MSARE,and nomogram models were successfully constructed using these factors.The area under the ROC curve(AUC)(0.888,95%CI:0.815~0.961)indicated that the model had high prediction accuracy.The calibration curve was close to the ideal curve.The model had good calibration ability.The decision curve analysis(DCA)suggested that a threshold of approximately 0.06 to 1 had the largest margin of benefit for the model and that the model was clinically useful.Conclusion:We successfully constructed a MSARE risk prediction model for esophageal cancer patients,with good model efficacy,which is helpful for clinicians to identify high-risk MSARE patients early,so as to prevent and improve treatment plans early.

关 键 词：食管癌 中重度急性放射性食管炎 列线图 

分 类 号：R735.1[医药卫生—肿瘤]