临床、病理特征联合能谱CT评估非小细胞肺癌表皮生长因子受体基因突变  

Clinical,pathological and spectral CT characteristics in evaluation on epidermal growth factor receptor gene mutation of non-small cell lung cancer

作　　者：何小群[1] 李琦[1] 罗天友[1] 吕发金[1] 刘筱霜 霍继文 HE Xiaoqun;LI Qi;LUO Tianyou;LYU Fajin;LIU Xiaoshuang;HUO Jiwen(Department of Radiology,the First Affiliated Hospital of Chongqing Medical University,Chongqing 400016,China)

机构地区：[1]重庆医科大学附属第一医院放射科,重庆400016

出　　处：《中国医学影像技术》2021年第2期230-234,共5页Chinese Journal of Medical Imaging Technology

基　　金：重庆市卫生和计划生育委员会医学科研计划(2017MSXM010);重庆市科学技术委员会科技计划(cstc2017jcyjAX0281)。

摘　　要：目的观察临床、病理特征及常规CT征象联合能谱CT定量参数预测非小细胞肺癌(NSCLC)表皮生长因子受体(EGFR)基因突变的价值。方法回顾性分析91例NSCLC患者,根据EGFR基因检测结果分为突变阳性组(n=47)和阴性组(n=44)。比较组间临床、病理表现、常规CT征象及能谱CT定量参数差异。针对差异有统计学意义的参数构建预测NSCLC患者EGFR基因突变的Logistic回归模型1和回归模型2,以ROC曲线评价模型预测效能。结果相比阴性组,阳性组多为女性、不吸烟及腺癌患者(P均<0.05)。阳性组病灶多见含气腔隙(P<0.05),阴性组病灶多见钙化及坏死(P均<0.05)。阳性组动、静脉期碘浓度值及水浓度值均高于阴性组(P均<0.05)。回归模型2预测NSCLC患者EGFR基因突变效能(AUC=0.788)优于回归模型1(AUC=0.686,Z=2.606,P=0.019)。结论临床、病理特征及常规CT征象联合能谱CT定量参数可提高预测NSCLC患者EGFR基因突变的效能。Objective To explore the value of clinical,pathological and conventional CT features combined with spectral CT quantitative parameters in predicting epidermal growth factor receptor(EGFR)mutation of non-small cell lung cancer(NSCLC).Methods Data of 91 NSCLC patients were retrospectively analyzed.The patients were divided into EGFR mutation-positive group(n=47,positive group)and mutation-negative group(n=44,negative group)according to EGFR gene test results.Based on parameters being statistical different between groups,Logistic regression model 1 and 2 were constructed to predict EGFR mutation status of NSCLC,respectively,and ROC curve was used to evaluate the corresponding diagnostic efficacy.Results Compared with negative group,female,non-smoking and adenocarcinoma patients were more common in positive group(all P<0.05),while more lesions in positive group showed air space(P<0.05),more lesions in negative group showed calcifications and necroses(both P<0.05).Iodine concentration and water concentration in positive group were higher than those in negative group at the arterial and venous phases(both P<0.05).The efficacy of predicting EGFR gene mutation in NSCLC of model 2(AUC=0.788)was better than that of model 1(AUC=0.686,Z=2.606,P=0.019).Conclusion Combining clinical,pathological characteristics and conventional CT features with spectral CT quantitative parameters could effectively improve the prediction efficacy of EGFR gene mutations of NSCLC.

关 键 词：肺肿瘤 表皮生长因子受体 体层摄影术 X线计算机 基因突变 

分 类 号：R734.2[医药卫生—肿瘤] R814.42[医药卫生—临床医学]