不可切除局部晚期食管癌患者放免联合治疗后发生≥2级放射性肺炎的列线图预测模型  

Construction of a nomogram prediction model of grade≥2 radiation pneumonitis in patients with unresectable locally advanced esophageal carcinoma after combined radiation/immunotherapy

作　　者：渠琬溪 王晨阳 刘兹昉 丁昕 QU Wanxi;WANG Chenyang;LIU Zifang;DING Xin(The First School of Clinical Medicine,Xuzhou Medical University,Xuzhou,Jiangsu 221004,China;Department of Radiation Oncology,the Affiliated Hospital of Xuzhou Medical University,Xuzhou,Jiangsu 221002)

机构地区：[1]徐州医科大学第一临床医学院,江苏徐州221004 [2]徐州医科大学附属医院放疗科,江苏徐州221002

出　　处：《徐州医科大学学报》2024年第4期271-278,共8页Journal of Xuzhou Medical University

基　　金：国家自然科学基金(81972845)。

摘　　要：目的 探讨不可切除局部晚期食管癌患者放免联合治疗后发生≥2级放射性肺炎的危险因素,并构建列线图预测模型。方法 选取2020年1月—2023年1月于徐州医科大学附属医院接受放免联合治疗的不可切除局部晚期食管癌患者110例,收集临床资料进行回顾性分析。观察指标为胸部放疗完成后6个月内放射性肺炎的发生情况。采用单因素和多因素logistic回归分析确定不可切除局部晚期食管癌患者治疗后发生≥2级放射性肺炎的独立危险因素,并构建预测模型。结果 本研究纳入的110例食管癌患者中,33例(30.0%)出现≥2级放射性肺炎。多因素logistic回归分析显示,年龄、吸烟史、计划靶体积(PTV)和V25(肺部接受25 Gy剂量的体积百分比)是不可切除局部晚期食管癌患者治疗后出现≥2级放射性肺炎的独立预测因子。受试者工作特征(ROC)曲线显示,本研究所构建的列线图敏感度和特异度分别为0.758和0.818,曲线下面积(AUC)值达到0.851(95%CI:0.776~0.926),说明列线图具有良好的区分能力。校准曲线显示,≥2级放射性肺炎的实际发生率与列线图预测的≥2级放射性肺炎发生率之间具有较高的一致性。决策曲线分析进一步证实,该模型显示出良好的临床效益。结论 本研究构建了包含年龄、吸烟史、PTV和V25的列线图预测模型,可用于食管癌患者放免联合治疗后发生≥2级放射性肺炎的个体化预测。Objective To explore the risk factors for grade≥2 radiation pneumonitis in patients with unresectable locally advanced esophageal carcinoma after combined radiation/immunotherapy,and to construct a nomogram prediction model.Methods A total of 110 patients with unresectable locally advanced esophageal carcinoma who received the combined radiotherapy/immunotherapy in the Affiliated Hospital of Xuzhou Medical University from January 2020 to January 2023 were selected,and their clinical data were collected for retrospective analysis.The observation index was the occurrence of radiation pneumonitis within six months after the completion of chest radiotherapy.Univariate and multivariate logistic regression analyses were used to determine the independent risk factors for the occurrence of grade≥2 radiation pneumonitis in patients with unresectable locally advanced esophageal carcinoma after treatment and a prediction model was constructed.Results Among 110 esophageal carcinoma patients included in this study,33(30.0%)patients developed grade≥2 radiation pneumonitis.Multivariate logistic regression analysis showed that age,smoking history,planned target volume(PTV),and V25(percentage of volume of the lungs receiving a dose of 25 Gy)were the independent predictors of the occurrence of grade≥2 radiation pneumonitis after treatment in patients with unresectable locally advanced esophageal carcinoma. The receiver operating characteristic (ROC) curve showed that thesensitivity and specificity of the established nomogram in the current study were 0. 758 and 0. 818, with an area undercurve (AUC) value of 0. 851 (95%CI: 0. 776-0. 926), indicating that the nomogram had good discriminatory ability.The calibration curve showed high consistency between the actual incidence of grade≥2 radiation pneumonitis and thepredicted incidence of grade≥2 radiation pneumonitis from the nomogram. Decision curve analysis further confirmed thatthe model showed good clinical benefit. Conclusions The study provides a nomogram model contai

关 键 词：放射性肺炎 食管癌 免疫治疗 剂量体积直方图 预测模型 毒性 

分 类 号：R735.1[医药卫生—肿瘤]