Predicting the EGFR mutation status of lung adenocarcinoma manifesting as ground-glass nodules based on 18F-FDG PET/CT radiomics

Yunmei Shi; Rong Niu; Xiaonan Shao; Yuetao Wang

Abstract

3161

Introduction: To explore the predictive value of combining ¹⁸F-FDG PET/CT based radiomics and machine learning methods in distinguishing the mutation status of epidermal growth factor receptor (EGFR) in lung adenocarcinoma manifesting as ground-glass nodules (GGNs).

Methods: We retrospectively analyzed 106 patients with lung adenocarcinoma who underwent ¹⁸F-FDG PET/CT examination followed by surgical resection within one month. The patients were divided into EGFR mutant group and wild-type group. By semi-automatic segmentation of the tumor area on PET/CT image, 3562 radiomic features (1781 PET features, 1781 CT features) were extracted. The data set was divided into training set (n=68) and testing set (n=38) by stratified random sampling. In the training set, Mann-Whitney U test and the Least Absolute Shrinkage Sum Selection Operator (LASSO) algorithm were used to select the best 14 prediction features (2 PET features and 12 CT features), and the four machine learning classifiers was used to build a model to predict EGFR mutation status. Then, the 5-fold cross-validation method was used for verification. The receiver operating characteristic (ROC) curve was performed on the testing set to evaluate the model performance, and the area under the curve (AUC), sensitivity, specificity, and accuracy were calculated.

Results: Among the 106 nodules, 81 had EGFR mutations (76.4%). There were no significant differences in general data, morphological characteristics and PET/CT parameters between the EGFR mutation group and the wild group (P>0.05). Among the four models in the test set, XGBoost showed the best performance (AUC=0.798, 95%CI: 0.627-0.904), and was significantly better than Random Forest (AUC=0.680, 95%CI: 0.509-0.822) (Z=2.122, P=0.034).

Conclusions: The combination of ¹⁸F-FDG PET/CT radiomics and machine learning methods is a potential non-invasive method for predicting the EGFR mutation status of GGNs lung adenocarcinoma.