基于列线图CT影像及临床特征预测肺腺癌EGFR突变状态  

Prediction of EGFR Mutation Statusin Lung Adenocarcinoma Based on Nomograms,CT Images and Clinical Features

作　　者：韩鹏 陈芬芬 钱运红 吕梦宇 赵大海[2] 赵红[1] HAN Peng;CHEN Fen-fen;QIAN Yun-hong;LV Meng-yu;ZHAO Da-hai;ZHAO Hong(Department of Radiology,The Second Hospital of Anhui Medical University,Hefei 230601,Anhui Province,China;Department of Respiratory and Critical Care Medicine,The Second Hospital of Anhui Medical University,Hefei 230601,Anhui Province,China)

机构地区：[1]安徽医科大学第二附属医院放射科,安徽合肥230601 [2]安徽医科大学第二附属医院呼吸与危重症医学科,安徽合肥230601

出　　处：《中国CT和MRI杂志》2025年第4期57-60,共4页Chinese Journal of CT and MRI

基　　金：安徽省转化医学研究院科研基金项目(2021zhyx-c67);合肥市自然科学基金项目(2021037)。

摘　　要：目的表皮生长因子受体(EGFR)突变的识别对于肺腺癌(LUAD)患者的治疗决策至关重要。本研究旨在构建一种方便、快捷、非侵入性的列线图预测模型,预测肺腺癌患者EGFR的突变状态,实现个体化分子靶向治疗。方法收集安徽医科大学第二附属医院从2018年5月到2022年11月期间的223例肺腺癌患者的临床资料,根据是否发生EGFR突变将病人分为EGFR突变组和EGFR野生组,对肺腺癌患者的CT影像和临床特征进行Logistic回归分析,以构建和验证列线图的预测模型。结果223例病人中,EGFR野生型62例,突变型161例,发生率为72.2%。边缘清晰、支气管充气征、没有肿瘤坏死、无肺气肿和淋巴结无肿大是肺腺癌患者E G F R突变的独立预测因子,其构建列线图模型对肺腺癌具有极高的诊断效能(曲线下面积为0.792,95%CI:0.726–0.859)和校准能力(C指数为0.792)。结论包含边缘清晰、支气管充气征、没有肿瘤坏死、无肺气肿和淋巴结无肿大的CT影像特征构建的列线图模型具有较好的预测效能以及临床有效性,以便于早期、准确地和无创地预测肺腺癌患者的EGFR突变状态。Objective The identification of EGFR EGFR mutations is critical for treatment decision-making in lung adenocarcinoma(LUAD)patients.The aim of this study was to construct a convenient,fast and noninvasive nomogram prediction model to predict EGFR mutation status in lung adenocarcinoma patients,and to realize individualized molecular targeted therapy.Methods The clinical data of 223 patients with lung adenocarcinoma from the Second Affiliated Hospital of Anhui Medical University from May 2018 to November 2021,the patients were divided into EGFR-mutant group and EGFR-wild group according to whether they had EGFR mutation or not.Logistic regression analysis was perform[]ed on CT images and clinical characteristics of lung adenocarcinoma patients to construct and validate the predictive model of nomogram.Results Among the 223 patients,62 were wild type and 161 were mutant of EGFR,the incidence was 72.2%.Clear margin,air bronchogram,absence of tumor necrosis,emphysema,and lymph node enlargement were independent predictors of EGFR mutation in lung adenocarcinoma patients,the nomogram model had extremely high diagnostic efficacy for lung adenocarcinoma(area under the curve:0.792,95%CI:0.726-0.859)and calibration ability(C Index:0.792).Conclusion The Nomogram model including clear edge,air bronchogram,no tumor necrosis,no emphysema and no lymphadenopathy on CT images has good predictive and clinical validity,in order to early,accurately and non-invasive prediction of lung adenocarcinoma patients the EGFR mutation status.

关 键 词：表皮生长因子受体 肺腺癌 列线图 预测模型 

分 类 号：R445.3[医药卫生—影像医学与核医学]