CT-based prediction of <i>EGFR</i> mutation subtypes and <i>TP53</i> co-mutation status in lung adenocarcinoma.
Authors
Affiliations (10)
Affiliations (10)
- Respiratory and Critical Care Medicine Center, Weifang People's Hospital, Weifang, China.
- National Clinical Research Center for Medical Auxiliary Technology(Laboratory Medicine), Department of Laboratory Medicine, The First Hospital of China Medical University, Shenyang, China.
- Research Unit of Medical Laboratory, Chinese Academy of Medical Sciences, Shenyang, China.
- School of Mechanical and Automation, Weifang University, Weifang, China.
- Shandong Key Laboratory of Intelligent Manufacturing, Weifang, China.
- Department of Pathology, Yantai Yuhuangding Hospital, Yantai, China.
- Department of Pulmonary and Critical Care Medicine, Affiliated Hospital of Shandong Second Medical University, Weifang, China.
- Department of Pulmonary and Critical Care Medicine, Shandong Provincial Chest Hospital, Jinan, China.
- Department of Pulmonary and Critical Care Medicine, The Second Affiliated Hospital of Shandong First Medical University, Tai'an, China.
- School of Equipment Engineering, Shenyang Ligong University, Shenyang, China.
Abstract
Epidermal growth factor receptor (<i>EGFR</i>) mutation status is a key determinant of targeted treatment in lung adenocarcinoma, whereas co-occurring <i>TP53</i> alterations may further affect therapeutic response and prognosis. This study aimed to develop a noninvasive CT-based approach for predicting <i>EGFR</i> mutation subtypes and <i>TP53</i> co-mutation status in lung adenocarcinoma, thereby providing imaging support for pretreatment risk stratification and clinical decision-making. Patients with pathologically confirmed lung adenocarcinoma who underwent pretreatment chest CT and molecular testing were retrospectively enrolled from a single-center cohort. A total of 1,633 patients were included and categorized into five molecular groups: <i>EGFR</i> wild type, <i>EGFR</i> 19del, <i>EGFR</i> L858R, <i>EGFR</i> 19del with <i>TP53</i> co-mutation, and <i>EGFR</i> L858 R with <i>TP53</i> co-mutation. Based on chest CT images, key slice extraction, lesion ROI generation, automatic segmentation, and image standardization were performed. Semantic CT features, radiomics features, and deep CT representations were then derived. Under a unified data partitioning and standardized evaluation framework, four models were constructed and compared for the five-class classification task: a random forest model based on semantic CT features, a support vector machine model based on radiomics features, a Qwen-based semantic classification model, and a MedicalDINOv3-based deep learning model. To eliminate confounding bias induced by multimodal feature fusion, two ablation experiments were supplemented to split the model input features, and two MedicalDINOv3 models were built with inputs restricted to semantic features only and CT images only, respectively. Model performance was assessed using accuracy, macro-averaged area under the receiver operating characteristic curve (AUC), and macro-F1 score. External validation was further conducted using public and multicenter datasets. Among the 1,633 patients in the internal cohort, 731 were <i>EGFR</i> wild type, 295 had <i>EGFR</i> 19del, 399 had <i>EGFR</i> L858R, 118 had <i>EGFR</i> 19del with <i>TP53</i> co-mutation, and 90 had <i>EGFR</i> L858 R with <i>TP53</i> co-mutation. To quantify the independent predictive value of different input modalities, ablation experiments on three variants of MedicalDINOv3 were performed. Ablation results revealed distinct performance among three input strategies. The MedicalDINOv3 with only semantic input obtained a test accuracy of 31.55% and an AUC of 0.6305 with limited predictive ability. The branch solely fed with CT images reached an accuracy of 55.61% and a macro-average AUC of 0.6347. The complete MedicalDINOv3 integrating CT deep features, clinical and semantic features performed best, with a test accuracy of 0.6905 and a macro-average AUC of 0.8788. This comparison proved that CT deep representations outperform traditional handcrafted imaging and semantic features, and multi-modal fusion further improves subtype stratification performance. The support vector machine model achieved an accuracy of 0.527, a macro-AUC of 0.792 and a macro-F1 score of 0.524; the random forest model achieved an accuracy of 0.455, a macro-AUC of 0.819 and a macro-F1 score of 0.412; and the Qwen-based model achieved an accuracy of 0.339, a macro-AUC of 0.700 and a macro-F1 score of 0.316. Receiver operating characteristic analyses showed that the deep learning model provided relatively stable discrimination for <i>EGFR</i> wild type and <i>TP53</i> co-mutated categories, whereas overlap remained between the two major <i>EGFR</i>-sensitive subtypes, <i>EGFR</i> 19del and <i>EGFR</i> L858 R. In external validation, the public dataset used for <i>EGFR</i> binary classification yielded an overall accuracy of 67.0% and an AUC of 0.6666. The raw overall accuracy of the cohort was 75.5%, and raw overall accuracy is prone to overestimation when the cohort has severe class imbalance dominated by <i>EGFR</i> wild-type samples, which can only serve as an auxiliary reference. Multicenter five-class external validation yielded a balanced accuracy of 58.2% and a macro F1 of 0.574. The subtype-specific AUCs for <i>EGFR</i> wild type, <i>EGFR</i> 19del, <i>EGFR</i> L858R, <i>EGFR</i> 19del with <i>TP53</i> co-mutation, and <i>EGFR</i> L858 R with <i>TP53</i> co-mutation were 0.8342, 0.6504, 0.7221, 0.6586, and 0.7563, respectively. A CT-based deep learning model can provide noninvasive stratification of EGFR mutation subtypes and TP53 co-mutation status in lung adenocarcinoma. Compared with semantic CT features and radiomics features, deep CT representations showed superior overall discriminative performance. This approach may serve as an adjunctive tool to molecular testing for pretreatment assessment in lung adenocarcinoma. However, the model showed inconsistent generalization performance across different molecular subtypes, with notably limited predictive efficacy for EGFR 19del and EGFR 19del+TP53 subtypes. Therefore, predictions for these categories should be interpreted with caution, and the model requires further validation in larger, multicenter, and standardized datasets before clinical application.