Primary Lung Adenocarcinoma With Or Without EGFR Mutation

ObjectiveTo identify imaging markers that reflect the epidermal growth factor receptor(EGFR)mutation status by comparing computed tomography(CT)imaging-based histogram features between bone metastases in patients of primary lung adenocarcinoma with and without EGFR mutation.Materials and MethodsThis retrospective study included 57 patients with bone metastasis of primary lung adenocarcinoma between June 2014 and December 2017.These patients were pathologically confirmed as lung adenocarcinoma;EGFR mutation status was confirmed by gene detection;and metastatic bone lesions were analyzed by CT examination.We collected CT images of bone metastases in these patients.The region of interest(ROI)of bone metastases was manually segmented for three dimensions(3D)by the radiologists.A total of 42 CT imaging-based histogram features were automatically extracted.Feature selection was conducted using statistical difference test between groups(Shapiro-Wilk test,Bartlett test,Student's t-test,Mann-Whitney U test),single-factor logistic regression analysis and Spearman correlation analysis.A receiver operating characteristic(ROC)curve was plotted to compare the effectiveness of features in distinguishing between EGFR(+)and EGFR(-)groups.DeLong's test was used to analyze the differences between the area under the curve(AUC)values.ResultsThrough the statistical difference test between groups,34 histogram features with potential predictive ability were preliminarily obtained.After using single-factor logistic regression analysis,31 features were found to be statistically significant(p<0.05).Finally,Spearman correlation analysis was conducted and three histogram features,namely Range,Skewness,and Quantile 0.975,were the most representative and were significantly correlated with the EGFR mutation status.AUC values of Range,Skewness,and Quantile 0.975 were 0.765,0.699,and 0.749,respectively.Sensitivity and specificity of these three features were 0.788 and 0.708,0.417 and 0.909,and 0.667 and 0.750,respectively.After combining these three features,AUC increased to 0.783 and sensitivity and specificity were 0.708 and 0.788,respectively.Finally,the differences between AUC values in four groups were analyzed in pairs by Delong's test,and they were statistically insignificant.ConclusionCT imaging-based histogram features of bone metastases in patients of primary lung adenocarcinoma with and without EGFR mutation were identified,and they may contribute to diagnosis and prediction of EGFR mutation status.