Using combined CT-clinical radiomics models to identify epidermal growth factor receptor mutation subtypes in lung adenocarcinoma.

This study aimed to explore the predictive ability of deep learning (DL) for the common epidermal growth factor receptor (EGFR) mutation subtypes in patients with lung adenocarcinoma. A total of 665 patients with lung adenocarcinoma (528/137) were recruited from two different institutions. In the training set, an 18-layer convolutional neural network (CNN) and fivefold cross-validation strategy were used to establish a CNN model. Subsequently, an independent external validation cohort from the other institution was used to evaluate the predictive efficacy of the CNN model. Grad-weighted class activation mapping (Grad-CAM) technology was used for the visual interpretation of the CNN model. In addition, this study also compared the prediction abilities of the radiomics and CNN models. Receiver operating characteristic (ROC) curves, accuracy and precision values, and recall and F1-score were used to evaluate the effectiveness of the CNN model and compare its performance with that of the radiomics model. In the validation set, the micro- and macroaverage values of the area under the ROC curve of the CNN model to identify the three EGFR subtypes were 0.78 and 0.79, respectively. All evaluation indicators of the CNN model were better than those of the radiomics model. Our study confirmed the potential of DL for predicting the EGFR mutation status in lung adenocarcinoma. The imaging phenotypes of the three mutation subtypes were found to be different, which can provide a basis for choosing more accurate and personalized treatment in patients with lung adenocarcinoma.

Lung adenocarcinoma (LADC) with epidermal growth factor receptor (EGFR) mutation is considered a subgroup of lung cancer sensitive to EGFR-targeted tyrosine kinase inhibitors. We aimed to develop and validate a computed tomography (CT)-based radiomics signature for prediction of EGFR mutation status in LADC appearing as a subsolid nodule. A total of 467 eligible patients were divided into training and validation cohorts (n = 306 and 161, respectively). Radiomics features were extracted from unenhanced CT images by using Pyradiomics. A CT-based radiomics signature for distinguishing EGFR mutation status was constructed using the random forest (RF) method in the training cohort and then tested in the validation cohort. A combination of the radiomics signature with a clinical factors model was also constructed using the RF method. The performance of the model was evaluated using the area under the curve (AUC) of a receiver operating characteristic curve. In this study, 64.2% (300/467) of the patients showed EGFR mutations. L858R mutation of exon 21 was the most common mutation type (185/301). We identified a CT-based radiomics signature that successfully discriminated between EGFR positive and EGFR negative in the training cohort (AUC = 0.831) and the validation cohort (AUC = 0.789). The radiomics signature combined with the clinical factors model was not superior to the simple radiomics signature in the two cohorts (p > .05). As a noninvasive method, the CT-based radiomics signature can be used to predict the EGFR mutation status of LADC appearing as a subsolid nodule. Lung adenocarcinoma (LADC) with epidermal growth factor receptor (EGFR) mutation is considered a subgroup of lung cancer that is sensitive to EGFR-targeted tyrosine kinase inhibitors. However, some patients with inoperable subsolid LADC are unable to undergo tissue sampling by biopsy for molecular analysis in clinical practice. A computed tomography-based radiomics signature may serve as a noninvasive biomarker to predict the EGFR mutation status of subsolid LADCs when mutational profiling is not available or possible.

ObjectiveIn this study, we aim to establish a non-invasive tool to predict epidermal growth factor receptor (EGFR) mutation status and subtypes based on radiomic features of computed tomography (CT).MethodsA total of 233 lung adenocarcinoma patients were investigated and randomly divided into the training and test cohorts. In this study, 2300 radiomic features were extracted from original and filtered (Exponential, Laplacian of Gaussian, Logarithm, Gabor, Wavelet) CT images. The radiomic features were divided into four categories, including histogram, volumetric, morphologic, and texture features. An RF-BFE algorithm was developed to select the features for building the prediction models. Clinicopathological features (including age, gender, smoking status, TNM staging, maximum diameter, location, and growth pattern) were combined to establish an integrated model with radiomic features. ROC curve and AUC quantified the effectiveness of the predictor of EGFR mutation status and subtypes.ResultsA set of 10 features were selected to predict EGFR mutation status between EGFR mutant and wild type, while 9 selected features were used to predict mutation subtypes between exon 19 deletion and exon 21 L858R mutation. To predict the EGFR mutation status, the AUC of the training cohort was 0.778 and the AUC of the test cohort was 0.765. To predict the EGFR mutation subtypes, the AUC of training cohort was 0.725 and the AUC of test cohort was 0.657. The integrated model showed the most optimal predictive performance with EGFR mutation status (AUC = 0.870 and 0.759) and subtypes (AUC = 0.797 and 0.554) in the training and test cohorts.ConclusionCT-based radiomic features can extract information on tumor heterogeneity in lung adenocarcinoma. In addition, we have established a radiomic model and an integrated model to non-invasively predict the EGFR mutation status and subtypes of lung adenocarcinoma, which is conducive to saving clinical costs and guiding targeted therapy.

Objective To explore the association between epidermal growth factor receptor (EGFR) mutation subtypes and the prognosis of brain metastasis in patients with lung adenocarcinoma. Methods A retrospective analysis was performed on the clinical data of 256 patients who were admitted to our hospital and confirmed with brain metastases of lung adenocarcinoma by EGFR mutation detection from 2010 to 2015. The prognostic factors for brain metastases were analyzed. The survival rate was calculated by the Kaplan-Meier method and analyzed by the log-rank test. The univariate and multivariate prognostic analyses were performed by the log-rank test and the Cox proportional hazards model. Results The median survival time was 10.13 months in all patients. The univariate analysis showed that sex, EGFR mutation status, exon 19 deletion, the Karnofsky Performance Status (KPS) score of brain metastases, and targeted therapy were prognostic predictors (P=0.006, 0.001, 0.010, 0.000, 0.003). The multivariate analysis showed that the KPS score and exon 19 deletion were prognostic factors for brain metastases (P=0.000, 0.045). When grouped into the recursive partitioning analysis classes, all the patients were split into three subgroups with significantly different prognosis (P=0.000). Conclusions Exon 19 deletion is a prognostic predictor of brain metastases in patients with lung adenocarcinoma, which can be integrated into the prognosis scoring system for brain metastases of lung adenocarcinoma. EGFR tyrosine kinase inhibitors improve the survival in patients with brain metastases of lung adenocarcinoma and EGFR mutation, particularly, in those with exon 19 deletion. Key words: Neoplasm metastasis, brain; Lung neoplasms; Epidermal growth factor receptor; Prognosis

Despite the increasing interest in radiogenomic prediction, few studies have directly compared the performance of logistic regression and decision tree models in distinguishing epidermal growth factor receptor (EGFR) mutation subtypes. This study provides the first systematic comparison of the predictive performance of these two models in identifying exon 19 deletions (19Del) and exon 21 L858R point mutations (21L858R) in patients with lung adenocarcinoma. By leveraging imaging and clinical parameters, we aimed to address a critical gap in the literature by establishing an optimal prediction model and providing a noninvasive tool to support personalized treatment strategies for patients with unknown EGFR mutation status. We retrospectively collected clinical and radiological data from 193 patients with histologically confirmed lung adenocarcinoma who were admitted to the Second Affiliated Hospital of Anhui Medical University between May 2018 and June 2024. Based on EGFR genotyping results, patients were stratified into two groups: the EGFR 19Del mutation group and the EGFR 21L858R mutation group. Comparative statistical analyses-including Student's t-test, Mann-Whitney U test, chi-square test, or Fisher's exact test-were performed to evaluate differences in clinical and CT imaging characteristics between groups. Variables with P < 0.05 in the univariate analysis were subsequently included in both logistic regression and decision tree models to identify independent predictors of EGFR mutation subtype. Model performance was assessed using ROC curve analysis. The area under the curve (AUC) was calculated for each model, and their predictive accuracy was further compared using DeLong's test, net reclassification improvement (NRI), and integrated discrimination improvement (IDI). In the decision tree model, age and brain metastasis emerged as key decision nodes for differentiating 19Del and 21L858R mutations, with an AUC of 0.712 (95% CI: 0.639-0.785). In contrast, the logistic regression model identified age, pleural thickening, lymphadenopathy, and brain metastasis as independent predictors, achieving a higher AUC of 0.740 (95% CI: 0.671-0.810). The NRI and IDI values were 0.498 (P < 0.001, 95% CI: 0.238-0.758) and 0.043 (P = 0.004, 95% CI: 0.013-0.072), respectively, suggesting improved reclassification and discrimination by the logistic model. However, DeLong's test revealed no statistically significant difference between the AUCs of the two models (Z = 1.314, P = 0.189). Both logistic regression and decision tree models demonstrated value in predicting EGFR 19Del and 21L858R mutations in lung adenocarcinoma, each offering distinct methodological advantages. The logistic regression model exhibited higher interpretability and statistical robustness, making it well-suited for clinical decision-making. Meanwhile, the decision tree model offered superior visual clarity and intuitive structure, which may enhance practical utility. A combined modeling approach that harnesses the strengths of both methods may provide a more accurate and comprehensive tool for early mutation identification and individualized treatment planning in patients with lung adenocarcinoma.

MRI-Based Radiomics Nomogram as a Potential Biomarker to Predict the EGFR Mutations in Exon 19 and 21 Based on Thoracic Spinal Metastases in Lung Adenocarcinoma

To predict epidermal growth factor receptor (EGFR) mutation status in lung adenocarcinoma using MR-based radiomics signature of brain metastasis and explore the optimal MR sequence for prediction. Data from 52 patients with brain metastasis from lung adenocarcinoma (28 with mutant EGFR, 24 with wild-type EGFR) were retrospectively reviewed. Contrast-enhanced T1-weighted imaging (T1-CE), T2 fluid-attenuated inversion recovery (T2-FLAIR), T2WI, and DWI sequences were selected for radiomics features extraction. A total of 438 radiomics features were extracted from each MR sequence. All sequences were randomly divided into training and validation cohorts. The least absolute shrinkage selection operator was used to select informative features, a radiomics signature was built with the logistic regression model of the training cohort, and the radiomics signature performance was evaluated using the validation cohort and an independent testing data set. The radiomics signature built on 9 selected features showed good discrimination in both the training and validation cohorts for T2-FLAIR. The radiomics signature of T2-FLAIR yielded an AUC of 0.987, a classification accuracy of 0.991, sensitivity of 1.000, and specificity of 0.980 in the validation cohort. The AUC was 0.871 in the independent testing data set. The AUCs of our radiomics signature to differentiate exon 19 and exon 21 mutations were 0.529, 0.580, 0.645, and 0.406 for T1-CE, T2-FLAIR, T2WI, and DWI, respectively. We developed a T2-FLAIR radiomics signature that can be used as a noninvasive auxiliary tool for predicting EGFR mutation status in lung adenocarcinoma, which is helpful to guide therapeutic strategies. • MR-based radiomics signature of brain metastasis may help predict EGFR mutation status in lung adenocarcinoma, especially using T2-FLAIR. • Nine radiomics features extracted from T2-FLAIR sequence strongly correlate with EGFR mutation status. • Radiomics features reflect tumor heterogeneity through potential changes in tissue morphology caused by EGFR mutation.

Radiomics signature: A potential and incremental predictor for EGFR mutation status in NSCLC patients, comparison with CT morphology

Neoadjuvant chemotherapy and radical surgery versus exclusive radiotherapy in locally advanced squamous cell cervical cancer: results from the Italian multicenter randomized study

The histologic heterogeneity of lung adenocarcinoma is well known. Many histologic subtypes have been described, and recently their prognostic and predictive value has emerged. Laser capture microdissection may aid in the isolation of cancer cells from distinct subtypes of lung adenocarcinoma, thus enabling the description of their specific molecular features. Characterization of epidermal growth factor receptor (EGFR) mutations in histologic subtypes of lung adenocarcinoma has become an important issue. The purpose of this study was to analyze EGFR mutations in exons 18-21 in single histologic subtypes of lung adenocarcinoma after laser capture microdissection. A revision and reclassification of a series of 208 non-small cell lung cancers was conducted, and 62 adenocarcinomas with a total of 119 histologic component subtypes were identified. Laser capture microdissection of each subtype was performed. EGFR mutations in exons 18-21 were detected using polymerase chain reaction single-strand conformation polymorphism and direct DNA sequencing. EGFR mutations were detected only in 3 out of the 62 adenocarcinomas analyzed. Two adenocarcinomas harbored EGFR mutations in exon 19 (the E746-T751 deletion VA insertion and the LREAT deletion) and one adenocarcinoma the EGFR exon 21 L858R missense point mutation. EGFR mutations were observed in all component subtypes. This suggests that, in a patient with lung adenocarcinoma, EGFR mutations are not associated with particular component histologic subtypes and probably occur at an early stage of tumorigenesis. Notably, 2 out of the 3 mutated adenocarcinomas had a bronchioloalveolar component, whereas the third mutated adenocarcinoma had a papillary subtype. Although we detected EGFR mutations only in 3 out of 62 adenocarcinomas and EGFR mutations were present in every subtype of each mutated adenocarcinoma, our research might represent a basis for further studies in characterizing molecular profiles of different component subtypes of lung adenocarcinoma.

Objective To retrospectively investigate computed tomographic (CT) quantitative analysis of ground-glass opacity (GGO) volume percentage and morphologic features of resected lung adenocarcinomas according to epidermal growth factor receptor (EGFR) mutation status and subtypes. Methods Amplification refractory mutation system was used to detect mutations in the EGFR gene. Distribution of demographics and GGO volume percentage were performed according to EGFR mutation status and subtypes. Results EGFR mutations were significantly more frequent in women (55.2% vs. 37.0%, p=0.001) and in never-smokers (59.5% vs. 38.4%, p < 0.001) than those without EGFR mutation. GGO volume percentage was significantly higher in tumors with EGFR mutation than in tumors without EGFR mutation (52.8±25.7% vs. 29.0±20.7%, p < 0.001). The GGO volume percentages in tumors with exon 21 mutation and EGFR mutation showed a significant difference compared with those without EGFR mutation (p < 0.001, area under the curve=0.871, sensitivity=94.6%, specificity=73.8%, and p < 0.001, area under the curve=0.783, sensitivity=69.9%, specificity=75.4%, resp.), with cut-off values of 37.7% and 34.3% in receiver operating characteristic curve analysis. Conclusion GGO volume percentage in adenocarcinomas with EGFR mutation was significantly higher than that in tumors without EGFR mutation, and adenocarcinomas with exon 21 mutation showed significantly higher GGO volume percentage than in tumors with exon 19 mutation and those without EGFR mutation. Our results indicate that GGO volume percentage cut-off values of more than 37.7% and 34.3% were predictors of positive exon 21 mutation and EGFR mutation, respectively.

The exon 19 and 21 in Epidermal Growth Factor Receptor (EGFR) mutation are the most common subtype of lung adenocarcinoma, and the strongest predictive biomarker for progression-free survival and tumor response. Although some studies have shown differences in radiological features between cases with and without EFGR mutations, they lacked necessary stratification. This article is to evaluate the association of CT features between the wild type and the subtype (exon 19 and 21) of EGFR mutations in patients with lung adenocarcinoma. Of the 721 finally included patients, 132 were positive for EGFR mutation in exon 19, 140 were positive for EGFR mutation in exon 21, and 449 were EGFR wild type. EGFR mutation in exon 19 was associated with a small-maximum diameter (28.51 ± 14.07) (p < 0.0001); sex (p < 0.0001); pleural retraction (p = 0.0034); and the absence of fibrosis (p < 0.0001), while spiculated margins (p = 0.0095), subsolid density (p < 0.0001) and no smoking (p < 0.0001) were associated with EGFR mutation in exon 21. Receiver Operating Characteristic (ROC) curves suggested that the maximum Area Under the Curve (AUC) was related to the female gender (AUC = 0.636) and the absence of smoking (AUC = 0.681). This study demonstrated the radiological and clinical features could be used to prognosticate EGFR mutation subtypes in exon 19 and 21.

To investigate the use of radiomics in the in-depth identification of epidermal growth factor receptor (EGFR) mutation status in patients with lung adenocarcinoma. Computed tomography images of 438 patients with lung adenocarcinoma were collected in two different institutions, and 496 radiomic features were extracted. In the training set, lasso logistic regression was used to establish radiomic signatures. Combining radiomic index and clinical features, five machine learning methods, and a tenfold cross-validation strategy were used to establish combined models for EGFR+ vs EGFR- , and 19Del vs L858R, groups. The predictive power of the models was then evaluated using an independent external validation cohort. In the EGFR+ vs EGFR- and 19Del vs L858R groups, radiomic signatures consisting of 12 and 7 radiomic features were established, respectively; the area under the curves (AUCs) of the lasso logistic regression model on the validation set was 0.76 and 0.71, respectively. After inclusion of the clinical features, the maximum AUC of combined models on the validation set was 0.79 and 0.74, respectively. Logistic regression analysis showed good performance in the two groups, with AUCs of 0.79 and 0.71 on the validation set. Additionally, the AUC of combined models in the EGFR+ vs EGFR- group was higher than that of the 19Del vs L858R group. Our study shows the potential of radiomics to predict EGFR mutation status. There are imaging phenotypic differences between EGFR+ and EGFR- , and between 19Del and L858R; these can be used to allow patients with lung adenocarcinoma to choose more appropriate and personalized treatment options.

The purpose of this study is to develop the predictive models for epidermal growth factor receptor (EGFR) mutation status and subtypes [exon 21-point mutation (L858R) and exon 19 deletion mutation (19Del)] and evaluate their clinical usefulness. Total 172 patients with lung adenocarcinoma were retrospectively analyzed. The analysis of variance and the least absolute shrinkage were used for feature selection from plain computed tomography images. Then, radiomic score (rad-score) was calculated for the training and test cohorts. Two machine learning (ML) models with 5-fold were applied to construct the predictive models with rad-score, clinical features, and the combination of rad-score and clinical features. The nomogram was developed using rad-score and clinical features. The prediction performance was evaluated by the area under the receiver operating characteristic curve (AUC). Finally, decision curve analysis (DCA) was performed using the best ML and nomogram models. In the test cohorts, the AUC of the best ML and the nomogram model were 0.73 (95% confidence interval, 0.59-0.87) and 0.79 (0.65-0.92) in the EGFR mutation groups, 0.83 (0.67-0.99) and 0.85 (0.72-0.97) in the L858R mutation groups, as well as 0.77 (0.58-0.97) and 0.77 (0.60-0.95) in the 19Del groups. The DCA showed that the nomogram models have comparable results with ML models. We constructed two predictive models for EGFR mutation status and subtypes. The nomogram models had comparable results to the ML models. Because the superiority of the performance of ML and nomogram models varied depending on the prediction groups, appropriate model selection is necessary.

Identifying an epidermal growth factor receptor (EGFR) mutation is important because EGFR tyrosine kinase inhibitors are the first-line treatment of choice for patients with EGFR mutation-positive lung adenocarcinomas (LUAC). This study is aimed at developing and validating a radiomics-based machine learning (ML) approach to identify EGFR mutations in patients with LUAC. We retrospectively collected data from 201 patients with positive EGFR mutation LUAC (140 in the training cohort and 61 in the validation cohort). We extracted 1316 radiomics features from preprocessed CT images and selected 14 radiomics features and 1 clinical feature which were most relevant to mutations through filter method. Subsequently, we built models using 7 ML approaches and established the receiver operating characteristic (ROC) curve to assess the discriminating performance of these models. In terms of predicting EGFR mutation, the model derived from radiomics features and combined models (radiomics features and relevant clinical factors) had an AUC of 0.79 (95% confidence interval (CI): 0.77-0.82), 0.86 (0.87-0.88), respectively. Our study offers a radiomics-based ML model using filter methods to detect the EGFR mutation in patients with LUAC. This convenient and low-cost method may be of help to noninvasively identify patients before obtaining tumor sample for molecule testing.