Ⅰ~ⅢB期可切除周围型非小细胞肺癌患者临床及高分辨率CT影像特征与EGFR基因突变的相关性研究

Correlation between clinical and HRCT imaging features with epidermal growth factor receptor gene mutation in patients with stage Ⅰ−ⅢB resectable peripheral non-small cell lung cancer

目的探讨Ⅰ~ⅢB期可切除周围型非小细胞肺癌(NSCLC)患者的临床及高分辨率CT(HRCT)影像特征与表皮生长因子受体(EGFR)基因突变的相关性。方法回顾性分析2019年1月至2021年4月于惠州市中心人民医院行手术切除、术后行EGFR基因检测且临床分期为Ⅰ~ⅢB期的周围型NSCLC患者164例,其中男性86例、女性78例,年龄(60.3±9.3)岁。根据EGFR基因检测结果将患者分为EGFR基因突变型组和野生型组,分析患者的临床及HRCT影像特征并进行统计学分析。计量资料的比较采用两独立样本t检验,计数资料的比较采用χ^(2)检验或Fisher确切概率检验。筛选差异有统计学意义的指标纳入二元Logistic回归分析。采用受试者工作特征(ROC)曲线的曲线下面积对构建的回归模型的诊断效能进行评估。结果164例NSCLC患者中,EGFR基因突变型组114例(69.51%)、野生型组50例(30.49%)。与EGFR野生型组比较,EGFR基因突变型组多见于女性、年龄<60岁、无吸烟史、病理类型为肺腺癌的患者,且2组间的差异均有统计学意义(χ^(2)=21.91、−4.71、34.64,Fisher确切概率法,均P<0.05)。HRCT影像特征中,EGFR基因突变型组肿瘤最大径小于野生型组[(2.52±1.51)cm对(4.12±6.07)cm],且肿瘤多位于右肺,密度类型多为磨玻璃影和(或)混合磨玻璃影,多伴有分叶征、支气管充气征和胸膜牵拉征,与EGFR野生型组相比,差异均有统计学意义(t=2.14,χ^(2)=10.13~19.05,均P<0.05)。2组患者在术后临床分期、癌胚抗原水平、肿瘤伴有毛刺征、坏死、血管集束征、空泡征、空洞征、胸腔淋巴结肿大中的差异均无统计学意义(χ^(2)=0.40~2.33,Fisher确切概率法,均P>0.05)。Logistic回归分析结果显示,患者无吸烟史(OR=0.225,95%CI:0.066~0.764)和肿瘤伴有分叶征(OR=3.344,95%CI:1.079~10.360)是EGFR基因突变的独立预测因子(均P<0.05),回归模型的ROC曲线的曲线下面积为0.858。结论Ⅰ~ⅢB期可切除周围型NSCLC的临床及HRCT影像特征与EGFR基因突变具有一定的相关性,对指导临床评估患者情况有重要意义。

Objective To investigate the correlation between epidermal growth factor receptor(EGFR)gene mutation with clinical and high-resolution CT(HRCT)imaging features in patients with stagesⅠ−ⅢB resectable peripheral non-small cell lung cancer(NSCLC).Methods A total of 164 patients with peripheral NSCLC who underwent lung cancer resection,EGFR genetic testing,and clinical staging ofⅠ−ⅢB in Huizhou Central People's Hospital from January 2019 to April 2021 were retrospectively analyzed.The cohort included 86 males and 78 females,aged 60.3±9.3 years.According to the results of EGFR genetic testing,the patients were divided into EGFR mutation and wild-type groups.The clinical and HRCT imaging features of the patients were analyzed and statistically analyzed.The measurement data were compared by two-independent-sample t-test,and the counting data were compared by chi-square and Fisher's exact probability tests.Indicators with statistical differences were selected for binary Logistic regression analysis.The area under curve of the receiver operating characteristic(ROC)curve was used in evaluating the diagnostic efficiency of the regression model.Results Among the 164 patients with NSCLC,114(69.51%)were confirmed to have EGFR mutations,and 50(30.49%)had wild-type EGFR.Compared with the wild-type EGFR,EGFR mutations were more frequent in non smoking,female patients aged<60 years with a pathological type of adenocarcinoma.Differences between the groups were statistically significant(χ^(2)=21.91,−4.71,34.64;Fisher's exact probability method,all P<0.05).As observed in HRCT imaging features,the maximum diameter of EGFR mutation group was smaller than that of the wild-type group((2.52±1.51)cm vs.(4.12±6.07)cm),and the tumors were mostly located in the right lung,with ground-glass opacity or mixed ground-glass opacity,lobulation sign,bronchial inflation sign,and pleural traction sign,compared with the wild-type group,and the differences were statistically significant(t=2.14,χ^(2)=10.13–19.05,all P<0.05).No significant differences in postoperative clinical stage,carcinoembryonic antigen level,lesions with burr signs,necrosis,vascular convergence sign,vacuole sign,cavity sign and thoracic lymphadenopathy were found between the groups(χ^(2)=0.40–2.33,Fisher's exact probability method,all P>0.05).The results of Logistic regression analysis showed that patients with no smoking history(OR=0.225,95%CI:0.066–0.764)and tumor accompanied by lobulation sign(OR=3.344,95%CI:1.079–10.360)are the independent predictors of EGFR gene mutation(all P<0.05),and the area under the curve of the regression model was 0.858.Conclusion The clinical and imaging features of resectable peripheral NSCLC in stagesⅠ−ⅢB are correlated with EGFR gene mutation,which are of great significance to assess the patient's condition.