^(18)F-FDG PET/CT多代谢参数对非小细胞肺癌EGFR基因突变的预测价值

The predictive value of multiple metabolic parameters of^(18)F-FDG PET/CT for EGFR gene mutations in non-small cell lung cancer

目的:探究氟代脱氧葡萄糖正电子发射型计算机断层显像(^(18)F-FDG PET/CT)代谢参数对非小细胞肺癌(NSCLC表皮生长因子受体(EGFR)基因突变的预测价值。方法:回顾性分析2016年3月—2021年3月治疗前于我院接受^(18)F-FDG PET/CT检查并行EGFR检测的98例NSCLC患者。单因素及多因素Logistic回归分析筛选独立预测因素并构建模型,以受试者工作特征(ROC)曲线判断该模型诊断效能。结果:EGFR突变更易发生于女性、非吸烟者、周围型、最大径较短、腺癌、T1期、低最大标准化摄取值(SUVmax)、平均标准化摄取值(SUVmean)、肿瘤代谢体积(MTV)及病变糖酵解总量(TLG)患者中,4个代谢参数的ROC曲线下面积(AUC)分别为0.622、0.638、0.679及0.687。多因素分析表明非吸烟者(OR=0.152,P=0.019)、周围型病变(OR=0.5.239,P=0.033)、低SUVmax(≤7.85,OR=4.170,P=0.002)及低SUVmean(≤5.34,OR=0.082,P=0.002)为独立预测因素,4项参数拟合建立预测模型的AUC为0.907,敏感度98.2%,特异性79.1%,阳性预测值85.5%,阴性预测值81.4%,正确百分比83.7%,诊断效能较单一参数显著提高。结论:^(18)F-FDG PET/CT多代谢参数中低SUVmax、SUVmean与EGFR突变明显相关,且联合临床特征可显著提高对NSCLC患者EGFR基因突变的预测。

Objective:To investigate the predictive value of metabolic parameters of^(18)F-FDG PET/CT for epidermal growth factor receptor(EGFR)gene mutations in non-small cell lung cancer(NSCLC).Methods:A retrospective analysis was performed by reviewing 98 NSCLC who conducted EGFR gene mutation test and underwent^(18)F-FDG PET/CT examination in our hospital from March 2016 to March 2021.Univariate and multivariate Logistic regression analysis were performed to screen independent predictors and construct model.The diagnostic performance of the model was judged by receiver operating characteristic(ROC)curve.Results:The study found that EGFR mutations were more likely to occur in patients with women,non-smokers,peripheral lesions,shorter maximal tumor size,adenocarcinoma,T1 stage,low SUVmax,SUVmean,MTV and TLG.ROC curve showed that the AUCs of the four metabolic parameters were 0.622,0.638,0.679 and 0.687,respectively.Multivariate analysis showed that non-smokers(OR=0.152,P=0.019),peripheral lesions(OR=0.5.239,P=0.033),low SUVmax(≤7.85,OR=4.170,P=0.002)and SUVmean(≤5.34,OR=0.082,P=0.002)were independent predictors.After establishing the prediction model combined with the above factors,the AUC was 0.907,the sensitivity,specificity,positive and negative predictive value and correct percentage were 98.2%,79.1%,85.5%,81.4%,and 83.7%.The diagnostic efficiency was significantly higher than that of single metabolic parameter.Conclusions:Among the metabolic parameters of^(18)F-FDG PET/CT,low SUVmax and SUVmean were significantly associated with EGFR mutations,and the predictive value of EGFR mutations could be significant-ly improved when combined with clinical features.