Predict Epidermal Growth Factor Receptor Mutation Research Articles

A CT-based Deep Learning Radiomics Nomogram for the Prediction of EGFR Mutation Status in Head and Neck Squamous Cell Carcinoma

Accurately assessing epidermal growth factor receptor (EGFR) mutation status in head and neck squamous cell carcinoma (HNSCC) patients is crucial for prognosis and treatment selection. This study aimed to construct and validate a contrast-enhanced computed tomography (CECT)-based deep learning radiomics nomogram (DLRN) to predict EGFR mutation status of HNSCC. A total of 300 HNSCC patients who underwent CECT scans were enrolled in this study. Participants from two hospitals were separated into a training set (n=200, 56 EGFR-negative and 144 EGFR-positive) from one hospital and an external test set from the other hospital (n=100, 37 EGFR-negative and 63 EGFR-positive). The least absolute shrinkage and selection operator method was used to select the key features from CECT-based manually extracted radiomics (MER) features and features automatically extracted using a deep learning model (DL, extracted using a GoogLeNet model). The selected independent clinical factors, MER features, and DL features were then combined to construct a DLRN. The DLRN's performance was evaluated using receiver operating characteristics curves. Five MER and six DL features were finally chosen. The DLRN, which includes "gender" and "necrotic areas," along with the selected features, predicted EGFR mutation status of HNSCC (EGFR-negative vs. positive) well in both the training (area under the curve [AUC], 0.901) and test (AUC, 0.875) sets. A DLRN using CECT was built to predict EGFR mutation in HNSCC. The model showed high predictive ability and may aid in treatment selection and patient prognosis.

High accuracy epidermal growth factor receptor mutation prediction via histopathological deep learning

BackgroundThe detection of epidermal growth factor receptor (EGFR) mutations in patients with non-small cell lung cancer is critical for tyrosine kinase inhibitor therapy. EGFR detection requires tissue samples, which are difficult to obtain in some patients, costing them the opportunity for further treatment. To realize EGFR mutation prediction without molecular detection, we aimed to build a high-accuracy deep learning model with only haematoxylin and eosin (H&E)-stained slides.MethodsWe collected 326 H&E-stained non-small cell lung cancer slides from Beijing Chest Hospital, China, and used 226 slides (88 with EGFR mutations) for model training. The remaining 100 images (50 with EGFR mutations) were used for testing. We trained a convolutional neural network based on ResNet-50 to classify EGFR mutation status on the slide level.ResultsThe sensitivity and specificity of the model were 76% and 74%, respectively, with an area under the curve of 0.82. When applying the double-threshold approach, 33% of the patients could be predicted by the deep learning model as EGFR positive or negative with a sensitivity and specificity of 100.0% and 87.5%. The remaining 67% of the patients got an uncertain result and will be recommenced to perform further examination. By incorporating adenocarcinoma subtype information, we achieved 100% sensitivity in predicting EGFR mutations in 37.3% of adenocarcinoma patients.ConclusionsOur study demonstrates the potential of a deep learning-based EGFR mutation prediction model for rapid and cost-effective pre-screening. It could serve as a high-accuracy complement to current molecular detection methods and provide treatment opportunities for non-small cell lung cancer patients from whom limited samples are available.

Multicenter evaluation of AI-based CT radiomics for EGFR mutation prediction in NSCLC.

e20515 Background: Tyrosine kinase inhibitors targeting the epidermal growth factor receptor (EGFR) were the first targeted therapy to enter clinical use for non-small cell lung cancer (NSCLC), thanks to its significant improvement on treatment response and patient survival. Today, tissue sampling is the standard to collect material for testing EGFR mutation status in advanced NSCLC, but has known sampling limitations due to intra-tumoral heterogeneity leading to false negative results. Computed tomography (CT) image based radiomics models offer a non-invasive solution, allowing comprehensive analysis of the entire tumor and surrounding tissue. Radiomics could provide insights during patient longitudinal follow up, or when tissue sample is not accessible is suboptimal. During the past decade, predicting EGFR mutation status using CT radiomics was widely investigated, with promising results. Nevertheless, most studies were single center retrospective studies, without external validation. Therefore, our study aims to investigate the performance of a CT radiomics EGFR mutation prediction model in a multicentral setting with an external cohort. Methods: CT images were collected from multicenter institutions and open source databases (https://www.cancerimagingarchive.net/), including a total of 1047 patients with advanced stage NSCLC with at least one measurable lung lesion (≥10mm). Each lung lesions was segmented semi-automatically followed by the validation of an experienced radiologist with > 20 years of experience. The dataset was divided to 2:1 for training and testing, with the test set originating from centers unique from the training cohort. A total of 1246 radiomics features were extracted using the Pyradiomics package. The 10 most robust features were selected first, by removing all features with variances near zero, then by applying Maximum Relevance — Minimum Redundancy (MRMR) to the remaining features. All selected features were standardized before modelling. A support vector machines (SVM) classifier was trained using a five-fold cross validation to classify EGFR mutations status (Wild type vs Mutant). Results: The SVM model trained with the radiomics features achieved mean area under curve (AUC) of 0.83 (95% CI, ±0.03) on the cross validation and an AUC of 0.76 (95% CI, ±0.03) on the test set. Conclusions: Our multicentric study demonstrated the potential of CT radiomics models in predicting EGFR mutation status in advanced NSCLC, with results similar to those found in current literature. We are optimizing the radiomics model by combining deep learning and clinical features, to improve the accurate, reliable and reproducible performance. AI powered medical image analysis has a bright future as an non-invasive and easy-to-use alternative to identify the expression status of common oncologic genes, serving as predictive biomarker for guiding the targeted therapies of the future.

Optimal 18F-FDG PET/CT radiomics model development for predicting EGFR mutation status and prognosis in lung adenocarcinoma: a multicentric study.

To develop and interpret optimal predictive models to identify epidermal growth factor receptor (EGFR) mutation status and subtypes in patients with lung adenocarcinoma based on multicentric 18F-FDG PET/CT data, and further construct a prognostic model to predict their clinical outcome. The 18F-FDG PET/CT imaging and clinical characters of 767 patients with lung adenocarcinoma from 4 cohorts were collected. Seventy-six radiomics candidates using cross-combination method to identity EGFR mutation status and subtypes were built. Further, Shapley additive explanations and local interpretable model-agnostic explanations were used for optimal models' interpretation. Moreover, in order to predict the overall survival, a multivariate Cox proportional hazard model based on handcrafted radiomics features and clinical characteristics was constructed. The predictive performance and clinical net benefit of the models were evaluated via area under receiver operating characteristic (AUC), C-index and decision curve analysis. Among the 76 radiomics candidates, light gradient boosting machine classifier (LGBM) combined with recursive feature elimination wrapped LGBM feature selection method achieved best performance in predicting EGFR mutation status (AUC reached 0.80, 0.61, 0.71 in the internal test cohort and two external test cohorts, respectively). And extreme gradient boosting classifier combined with support vector machine feature selection method achieved best performance in predicting EGFR subtypes (AUC reached 0.76, 0.63, 0.61 in the internal test cohort and two external test cohorts, respectively). The C-index of the Cox proportional hazard model achieved 0.863. The integration of cross-combination method and the external validation from multi-center data achieved a good prediction and generalization performance in predicting EGFR mutation status and its subtypes. The combination of handcrafted radiomics features and clinical factors achieved good performance in predicting prognosis. With the urgent needs of multicentric 18F-FDG PET/CT trails, robust and explainable radiomics models have great potential in decision making and prognosis prediction of lung adenocarcinoma.

Multiregional radiomics of brain metastasis can predict response to EGFR-TKI in metastatic NSCLC.

To develop radiomics signatures from multiparametric magnetic resonance imaging (MRI) scans to detect epidermal growth factor receptor (EGFR) mutations and predict the response to EGFR-tyrosine kinase inhibitors (EGFR-TKIs) in non-small cell lung cancer (NSCLC) patients with brain metastasis (BM). We included 230 NSCLC patients with BM treated at our hospital between January 2017 and December 2021 and 80 patients treated at another hospital between July 2014 and October 2021 to form the primary and external validation cohorts, respectively. All patients underwent contrast-enhanced T1-weighted (T1C) and T2-weighted (T2W) MRI, and radiomics features were extracted from both the tumor active area (TAA) and peritumoral edema area (POA) for each patient. The least absolute shrinkage and selection operator (LASSO) was used to identify the most predictive features. Radiomics signatures (RSs) were constructed using logistic regression analysis. For predicting the EGFR mutation status, the created RS-EGFR-TAA and RS-EGFR- POA showed similar performance. By combination of TAA and POA, the multi-region combined RS (RS-EGFR-Com) achieved the highest prediction performance, with AUCs of 0.896, 0.856, and 0.889 in the primary training, internal validation, and external validation cohort, respectively. For predicting response to EGFR-TKI, the multi-region combined RS (RS-TKI-Com) generated the highest AUCs in the primary training (AUC = 0.817), internal validation (AUC = 0.788), and external validation (AUC = 0.808) cohort, respectively. Our findings suggested values of multiregional radiomics of BM for predicting EGFR mutations and response to EGFR-TKI. TheapplicationofradiomicanalysisofmultiparametricbrainMRIhasproventobeapromisingtooltostratifywhichpatientscanbenefitfromEGFR-TKItherapyandtofacilitatetheprecisetherapeuticsofNSCLCpatientswithbrainmetastases. • Multiregional radiomics can improve efficacy in predicting therapeutic response to EGFR-TKI therapy in NSCLC patients with brain metastasis. • The tumor active area (TAA) and peritumoral edemaarea (POA) may hold complementary information related to the therapeutic response to EGFR-TKI. • The developed multi-region combined radiomics signature achieved the best predictive performance and may be considered as a potential tool for predicting response to EGFR-TKI.

Predicting EGFR Mutation Status in Non–Small Cell Lung Cancer Using Artificial Intelligence: A Systematic Review and Meta-Analysis

Recent advancements in artificial intelligence (AI) render a substantial promise for epidermal growth factor receptor (EGFR) mutation status prediction in non-small cell lung cancer (NSCLC). We aimed to evaluate the performance and quality of AI algorithms that use radiomics features in predicting EGFR mutation status in patient with NSCLC. We searched PubMed (Medline), EMBASE, Web of Science, and IEEExplore for studies published up to February 28, 2022. Studies utilizing an AI algorithm (either conventional machine learning [cML] and deep learning [DL]) for predicting EGFR mutations in patients with NSLCL were included. We extracted binary diagnostic accuracy data and constructed a bivariate random-effects model to obtain pooled sensitivity, specificity, and 95% confidence interval. This study is registered with PROSPERO, CRD42021278738. Our search identified 460 studies, of which 42 were included. Thirty-five studies were included in the meta-analysis. The AI algorithms exhibited an overall area under the curve (AUC) value of 0.789 and pooled sensitivity and specificity levels of 72.2% and 73.3%, respectively. The DL algorithms outperformed cML in terms of AUC (0.822 vs. 0.775) and sensitivity (80.1% vs. 71.1%), but had lower specificity (70.0% vs. 73.8%, p-value<0.001) compared to cML. Subgroup analysis revealed that the use of positron-emission tomography/computed tomography, additional clinical information, deep feature extraction, and manual segmentation can improve diagnostic performance. DL algorithms can serve as a novel method for increasing predictive accuracy and thus have considerable potential for use in predicting EGFR mutation status in patient with NSCLC. We also suggest that guidelines on using AI algorithms in medical image analysis should be developed with a focus on oncologic radiomics.

Investigation of the combination of intratumoral and peritumoral radiomic signatures for predicting epidermal growth factor receptor mutation in lung adenocarcinoma.

We investigated optimal peritumoral size and constructed predictive models for epidermal growth factor receptor (EGFR) mutation. A total of 164 patients with lung adenocarcinoma were retrospectively analyzed. Radiomic signatures for the intratumoral region and combinations of intratumoral and peritumoral regions (3, 5, and 7mm) from computed tomography images were extracted using analysis of variance and least absolute shrinkage. The optimal peritumoral region was determined by radiomics score (rad-score). Intratumoral radiomic signatures with clinical features (IRS) were used to construct predictive models for EGFR mutation. Combinations of intratumoral and 3, 5, or 7mm-peritumoral signatures with clinical features (IPRS3, IPRS5, and IPRS7, respectively) were also used to construct predictive models. Support vector machine (SVM), logistic regression (LR), and LightGBM models with five-fold cross-validation were constructed, and the receiver operating characteristics were evaluated. Area under the curve (AUC) of the training and test cohorts values were calculated. Brier scores (BS) and decision curve analysis (DCA) were used to evaluate the predictive models. The AUC values of the SVM, LR, and LightGBM models derived from IRS were 0.783 (95% confidence interval: 0.602-0.956), 0.789 (0.654-0.927), and 0.735 (0.613-0.958) for training, and 0.791 (0.641-0.920), 0.781 (0.538-0.930), and 0.734 (0.538-0.930) for test cohort, respectively. Rad-score confirmed that the 3mm-peritumoral size was optimal (IPRS3), and AUCs values of SVM, LR, and lightGBM models derived from IPRS3 were 0.831 (0.666-0.984), 0.804 (0.622-0.908), and 0.769 (0.628-0.921) for training and 0.765 (0.644-0.921), 0.783 (0.583-0.921), and 0.796 (0.583-0.949) for test cohort, respectively. The BS and DCA of the LR and LightGBM models derived from IPRS3 were better than those from IRS. Accordingly, the combination of intratumoral and 3mm-peritumoral radiomic signatures may be helpful for predicting EGFR mutations.

CT-based decision tree model for predicting EGFR mutation status in synchronous multiple primary lung cancers.

The current study aimed to construct a computed tomography (CT)-based decision tree algorithm (DTA) model to predict the epidermal growth factor receptor (EGFR) mutation status in synchronous multiple primary lung cancers (SMPLCs). The demographic and CT findings of 85 patients with molecular profiling for surgically resected SMPLCs were reviewed retrospectively. Least absolute shrinkage and selection operator (LASSO) regression was used to select the potential predictors of EGFR mutation, and a CT-DTA model was developed. Multivariate logistic regression analysis and receiver operating characteristic (ROC) curve analysis were performed to assess the performance of this CT-DTA model. The CT-DTA model was applied to predict the EGFR mutant that had ten binary split, of which eight parameters to accurately categorize the lesions as follows: the presence of bubble-like vacuole sign (19.4% importance in the development of the model), presence of air bronchogram sign (17.4% importance), smoking status (15.7% importance), types of the lesions (14.8% importance), histology (12.6% importance), presence of pleural indentation sign (7.6% importance), gender (6.9% importance), and presence of lobulation sign (5.6% importance). The ROC analysis achieved an area under the curve (AUC) of 0.854. Multivariate logistic regression analysis demonstrated that this CT-DTA model was an independent predictor of EGFR mutation (P<0.001). CT-DTA model is a simple tool to predict the status of EGFR mutation in SMPLC patients and could be considered for treatment decision-making.

Brain metastasis magnetic resonance imaging-based deep learning for predicting epidermal growth factor receptor (EGFR) mutation and subtypes in metastatic non-small cell lung cancer.

The preoperative identification of epidermal growth factor receptor (EGFR) mutations and subtypes based on magnetic resonance imaging (MRI) of brain metastases (BM) is necessary to facilitate individualized therapy. This study aimed to develop a deep learning model to preoperatively detect EGFR mutations and identify the location of EGFR mutations in patients with non-small cell lung cancer (NSCLC) and BM. We included 160 and 72 patients who underwent contrast-enhanced T1-weighted (T1w-CE) and T2-weighted (T2W) MRI at Liaoning Cancer Hospital and Institute (center 1) and Shengjing Hospital of China Medical University (center 2) to form a training cohort and an external validation cohort, respectively. A multiscale feature fusion network (MSF-Net) was developed by adaptively integrating features based on different stages of residual network (ResNet) 50 and by introducing channel and spatial attention modules. The external validation set from center 2 was used to assess the performance of MSF-Net and to compare it with that of handcrafted radiomics features. Receiver operating characteristic (ROC) curves, accuracy, precision, recall, and F1-score were used to evaluate the effectiveness of the models. Gradient-weighted class activation mapping (Grad-CAM) was used to demonstrate the attention of the MSF-Net model. The developed MSF-Net generated a better diagnostic performance than did the handcrafted radiomics in terms of the microaveraged area under the curve (AUC) (MSF-Net: 0.91; radiomics: 0.80) and macroaveraged AUC (MSF-Net: 0.90; radiomics: 0.81) for predicting EGFR mutations and subtypes. This study provides an end-to-end and noninvasive imaging tool for the preoperative prediction of EGFR mutation status and subtypes based on BM, which may be helpful for facilitating individualized clinical treatment plans.

A nomogram based on 18F-fluorodeoxyglucose PET/CT and clinical features to predict epidermal growth factor receptor mutation status in patients with lung adenocarcinoma.

Identifying epidermal growth factor receptor (EGFR) mutations in lung adenocarcinoma (LADC) is vital for treatment decision-making. This study aimed to establish a convenient and noninvasive nomogram prediction model based on 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) imaging and clinical features to predict EGFR mutation status in patients with LADC. A total of 274 patients (male 130, female 144, median age 65 years) were enrolled in this retrospective study. Imaging data from 18F-FDG PET/CT and clinical information were analyzed, with the Mann-Whitney U test, Student's t-test, and chi-square test used to compare categorical or continuous covariates as appropriate. Logistic regression analyses were performed to identify independent variables associated with EGFR mutation status, from which the nomogram prediction model was constructed. Leave-one-out cross-validation was performed, and the discrimination ability and calibration of the nomogram were assessed by calculating the area under the curve of the receiver operating characteristic curve and the calibration curve. The clinical net benefit of the nomogram was evaluated. Of the 274 patients, 143 (52.2%) had EGFR mutations. Female sex [odds ratios (OR): 2.64, 95% confidence interval (CI): 1.29-5.45, P=0.008], non-smoking status (OR: 2.78, 95% CI: 1.30-5.88, P=0.008), mean standardized uptake value ≤9.23 (OR: 2.44, 95% CI: 1.35-4.55, P=0.004), metabolic tumor volume ≤17.72 cm3 (OR: 5.00, 95% CI: 2.38-12.50, P<0.001) and the presence of pleural retraction (OR: 1.88, 95% CI: 1.05-3.40, P=0.034) were independent predictors for EGFR mutations in LADCs. The nomogram based on these risk factors showed good predictive efficacy, with an area under the curve of 0.805 (95% CI: 0.753-0.857), a sensitivity of 90.2%, a specificity of 59.5% and an accuracy of 73.0%. The nomogram prediction model incorporating sex, smoking status, mean standardized uptake value, metabolic tumor volume, and the presence of pleural retraction could effectively discriminate EGFR-mutant from wild-type LADCs.

Radiomics for prediction of response to EGFR-TKI based on metastasis/brain parenchyma (M/BP)-interface.

To evaluate the potential of subregional radiomics as a novel tumor marker in predicting epidermal growth factor receptor (EGFR) mutation status and response to EGFR-tyrosine kinase inhibitor (TKI) therapy in NSCLC patients with brain metastasis (BM). We included 230 patients from center 1, and 80 patients were included from center 2 to form a primary and external validation cohort, respectively. Patients underwent contrast-enhanced T1-weighted and T2-weighted MRI scans before treatment. The individual- and population-level clustering was used to partition the peritumoral edema area (POA) into phenotypically consistent subregions. Radiomics features were calculated and selected from the tumor active area (TAA), POA and subregions, and used to develop models. Prediction values of each region were investigated and compared with receiver operating characteristic curves and Delong test. For predicting EGFR mutations, a multi-region combined model (EGFR-Fusion) was developed based on joint of the partitioned metastasis/brain parenchyma (M/BP)-interface and TAA, and generated the highest prediction performance in the training (AUC = 0.945, SEN = 0.878, SPE = 0.937), internal validation (AUC = 0.880, SEN = 0.733, SPE = 0.969), and external validation (AUC = 0.895, SEN = 0.875, SPE = 0.800) cohorts. For predicting response to EGFR-TKI, the developed multi-region combined model (TKI-Fusion) yielded predictive AUCs of 0.869 (SEN = 0.717, SPE = 0.884), 0.786 (SEN = 0.708, SPE = 0.818), and 0.802 (SEN = 0.750, SPE = 0.800) in the training, internal validation and external validation cohort, respectively. Our study revealed that complementary information regarding the EGFR status and response to EGFR-TKI can be provided by subregional radiomics. The proposed radiomics models may be new markers to guide treatment plans for NSCLC patients with BM.

Predictive value of intratumor metabolic and heterogeneity parameters on [18F]FDG PET/CT for EGFR mutations in patients with lung adenocarcinoma.

This study aimed to investigate the value of metabolic and heterogeneity parameters of 2-deoxy-2[18F]fluoro-D-glucose ([18F]FDG) positron emission tomography/computed tomography (PET/CT) in predicting epidermal growth factor receptor (EGFR) mutations in patients with lung adenocarcinoma (ADC). A retrospective analysis was performed on 157 patients with lung ADC between September 2015 and June 2021, who had undergone both EGFR mutation testing and [18F]FDG PET/CT examination. Metabolic and heterogeneity parameters were measured and calculated, including maximum diameter (Dmax), maximum standardized uptake value (SUVmax), mean standardized uptake value (SUVmean), metabolic tumor volume (MTV), total lesion glycolysis (TLG), and heterogeneity factor (HF). Relationships between PET/CT parameters and EGFR mutation status were evaluated and a multivariate logistic regression analysis was analyzed to establish a combined prediction model. 108 (68.8%) patients exhibited EGFR mutations. EGFR mutations were more likely to occur in females (51.9% vs. 48.1%, P = 0.007), non-smokers (83.3% vs. 16.7%, P < 0.001) and right lobes (55.6% vs. 44.4%, P = 0.017). High Dmax, MTV and HF and low SUVmean were significantly correlated with EGFR mutations, and the areas under the ROC curve (AUCs) measuring 0.647, 0.701, 0.757, and 0.661, respectively. Multivariate logistic regression analysis suggested that non-smokers (OR = 0.30, P = 0.034), low SUVmean (≤ 7.75, OR = 0.63, P < 0.001) and high HF (≥ 4.21, OR = 1.80, P = 0.027) were independent predictors of EGFR mutations. The AUC of the combined prediction model measured up to 0.863, significantly higher than that of a single parameter. EGFR mutant in lung ADC patients showed more intratumor heterogeneity (HF) than EGFR wild type, which was combined clinical feature (non-smokers), and metabolic parameter (SUVmean) may be helpful in predicting EGFR mutation status, thus playing a guiding role in EGFR-tyrosine kinase inhibitors (EGFR-TKIs) targeted therapies.

Using stacked deep learning models based on PET/CT images and clinical data to predict EGFR mutations in lung cancer.

PurposeTo determine whether stacked deep learning models based on PET/CT images and clinical data can help to predict epidermal growth factor receptor (EGFR) mutations in lung cancer.MethodsWe analyzed data from two public datasets of patients who underwent 18F-FDG PET/CT. Three PET deep learning ResNet models and one CT deep learning ResNet model were trained as low-level predictors based on PET and CT images, respectively. A high-level Support Vector Machine model (Stack PET/CT and Clinical model) was trained using the prediction results of the low-level predictors and clinical data. The clinical data included sex, age, smoking history, SUVmax and SUVmean of the lesion. Fivefold cross-validation was used in this study to validate the prediction performance of the models. The predictive performance of the models was evaluated by receiver operator characteristic (ROC) curves. The area under the curve (AUC) was calculated.ResultsOne hundred forty-seven patients were included in this study. Among them, 37/147 cases were EGFR mutations, and 110/147 cases were EGFR wild-type. The ROC analysis showed that the Stack PET/CT & Clinical model had the best performance (AUC = 0.85 ± 0.09), with 0.76, 0.85 and 0.83 in sensitivity, specificity and accuracy, respectively. Three ResNet PET models had relatively higher AUCs (0.82 ± 0.07, 0.80 ± 0.08 and 0.79 ± 0.07) and outperformed the CT model (AUC = 0.58 ± 0.12).ConclusionUsing stack generalization, the deep learning model was able to efficiently combine the anatomic and biological imaging information gathered from PET/CT images with clinical data. This stacked deep learning model showed a strong ability to predict EGFR mutations with high accuracy.

Recent and current advances in PET/CT imaging in the field of predicting epidermal growth factor receptor mutations in non-small cell lung cancer.

Tyrosine kinase inhibitors (TKIs) are a significant treatment strategy for the management of non-small cell lung cancer (NSCLC) with epidermal growth factor receptor (EGFR) mutation status. Currently, EGFR mutation status is established based on tumor tissue acquired by biopsy or resection, so there is a compelling need to develop non-invasive, rapid, and accurate gene mutation detection methods. Non-invasive molecular imaging, such as positron emission tomography/computed tomography (PET/CT), has been widely applied to obtain the tumor molecular and genomic features for NSCLC treatment. Recent studies have shown that PET/CT can precisely quantify EGFR mutation status in NSCLC patients for precision therapy. This review article discusses PET/CT advances in predicting EGFR mutation status in NSCLC and their clinical usefulness.

Role of intratumoral and peritumoral CT radiomics for the prediction of EGFR gene mutation in primary lung cancer.

To determine the added value of combining intratumoral and peritumoral CT radiomics for the prediction of epidermal growth factor receptor (EGFR) gene mutations in primary lung cancer (PLC). This study included 478 patients with PLC (348 adenocarcinomas and 130 other histological types) who underwent surgical resection and EGFR gene testing. Two radiologists performed segmentation of tumors and peritumoral regions using precontrast high-resolution CT images, and 398 radiomic features (212 intra- and 186 peritumoral features) were extracted. The peritumoral region was defined as the lung parenchyma within a distance of 3 mm from the tumor border. Model performance was estimated using Random Forest, a machine-learning algorithm. EGFR mutations were found in 162 tumors; 161 adenocarcinomas, and one pleomorphic carcinoma. After exclusion of poorly reproducible and redundant features, 32 radiomic features remained (14 intra- and 18 peritumoral features) and were included in the model building. For predicting EGFR mutations, combining intra- and peritumoral radiomics significantly improved the performance compared to intratumoral radiomics alone (AUC [area under the receiver operating characteristic curve], 0.774 vs 0.730; p < 0.001). Even in adenocarcinomas only, adding peritumoral radiomics significantly increased performance (AUC, 0.687 vs 0.630; p < 0.001). The predictive performance using radiomics and clinical features was significantly higher than that of clinical features alone (AUC, 0.826 vs 0.777; p = 0.005). Combining intra- and peritumoral radiomics improves the predictive accuracy of EGFR mutations and could be used to aid in decision-making of whether to perform biopsy for gene tests. Adding peritumoral to intratumoral radiomics yields greater accuracy than intratumoral radiomics alone in predicting EGFR mutations and may serve as a non-invasive method of predicting of the gene status in PLC.

EP13.01-002 Radiomic Signature on CT Images: A Noninvasive Biomarker for Pretreatment Discrimination of EGFR Mutations in NSCLC Patients

Identification of the molecular pathways that drive malignancy led to the development of targeted therapy in cancer. Epidermal Growth Factor Receptor (EGFR), the leading one of these molecular targets, plays a key role in cell proliferation, angiogenesis, metastasis. This study aimed to assess the performance of pretreatment computed tomography (CT) radiomics features for predicting EGFR mutation status for predicting EGFR mutation status in patients with non-small cell lung cancer (NSCLC).

Computed tomography-based radiomics quantification predicts epidermal growth factor receptor mutation status and efficacy of first-line targeted therapy in lung adenocarcinoma.

BackgroundBiomarkers that predict the efficacy of first-line tyrosine kinase inhibitors (TKIs) are pivotal in epidermal growth factor receptor (EGFR) mutant advanced lung adenocarcinoma. Imaging-based biomarkers have attracted much attention in anticancer therapy. This study aims to use the machine learning method to distinguish EGFR mutation status and further explores the predictive role of EGFR mutation-related radiomics features in response to first-line TKIs.MethodsWe retrospectively analyzed pretreatment CT images and clinical information from a cohort of lung adenocarcinomas. We entered the top-ranked features into a support vector machine (SVM) classifier to establish a radiomics signature that predicted EGFR mutation status. Furthermore, we identified the best response-related features based on EGFR mutant-related features in first-line TKI therapy patients. Then we test and validate the predictive effect of the best response-related features for progression-free survival (PFS).ResultsSix hundred ninety-two patients were enrolled in building radiomics signatures. The 13 top-ranked features were input into an SVM classifier to establish the radiomics signature of the training cohort (n = 514), and the predictive score of the radiomics signature was assessed on an independent validation group with 178 patients and obtained an area under the curve (AUC) of 74.13%, an F1 score of 68.29%, a specificity of 79.55%, an accuracy of 70.79%, and a sensitivity of 62.22%. More importantly, the skewness-Low (≤0.882) or 10th percentile-Low group (≤21.132) had a superior partial response (PR) rate than the skewness-High or 10th percentile-High group (p < 0.01). Higher skewness (hazard ratio (HR) = 1.722, p = 0.001) was also found to be significantly associated with worse PFS.ConclusionsThe radiomics signature can be used to predict EGFR mutation status. Skewness may contribute to the stratification of disease progression in lung cancer patients treated with first-line TKIs.

Development and validation of MRI-based radiomics signatures as new markers for preoperative assessment of EGFR mutation and subtypes from bone metastases

BackgroundThis study aimed to develop and externally validate contrast-enhanced (CE) T1-weighted MRI-based radiomics for the identification of epidermal growth factor receptor (EGFR) mutation, exon-19 deletion and exon-21 L858R mutation from MR imaging of spinal bone metastasis from primary lung adenocarcinoma.MethodsA total of 159 patients from our hospital between January 2017 and September 2021 formed a primary set, and 24 patients from another center between January 2017 and October 2021 formed an independent validation set. Radiomics features were extracted from the CET1 MRI using the Pyradiomics method. The least absolute shrinkage and selection operator (LASSO) regression was applied for selecting the most predictive features. Radiomics signatures (RSs) were developed based on the primary training set to predict EGFR mutations and differentiate between exon-19 deletion and exon-21 L858R. The RSs were validated on the internal and external validation sets using the Receiver Operating Characteristic (ROC) curve analysis.ResultsEight, three, and five most predictive features were selected to build RS-EGFR, RS-19, and RS-21 for predicting EGFR mutation, exon-19 deletion and exon-21 L858R, respectively. The RSs generated favorable prediction efficacies for the primary (AUCs, RS-EGFR vs. RS-19 vs. RS-21, 0.851 vs. 0.816 vs. 0.814) and external validation (AUCs, RS-EGFR vs. RS-19 vs. RS-21, 0.807 vs. 0.742 vs. 0.792) sets.ConclusionsRadiomics features from the CE MRI could be used to detect the EGFR mutation, increasing the certainty of identifying exon-19 deletion and exon-21 L858R mutations based on spinal metastasis MR imaging.

Explainable Machine Learning Model to Prediction EGFR Mutation in Lung Cancer.

ObjectivesThe aim of this study is to determine whether the clinical features including blood markers can establish an explainable machine learning model to predict epidermal growth factor receptor (EGFR) mutation in lung cancer.MethodsWe retrospectively analyzed 7,413 patients with lung adenocarcinoma (LA) diagnosed by gene sequencing in West China Hospital of the Sichuan University from April 2015 to June 2019. The machine learning algorithms (MLAs) included logistic regression (LR), random forest (RF), LightGBM, support vector machine (SVM), multi-layer perceptron (MLP), extreme gradient boosting (XGBoost), and decision tree (DT). Demographic characteristics, personal history, and blood markers were taken into. The area under the receiver operating characteristic curve (AUC) and SHapley Additive exPlanation (SHAP) value were used to explain the prediction models.ResultsOf the 7,413 patients with LA (47.6%), 3,527 were identified with EGFR mutation; RF achieved greatest performance in predicting EGFR mutation AUC [0.771, 95% confidence interval (CI): 0.770, 0.772], which was like XGBoost with AUC (0.740, 95% CI: 0.739, 0.741). The five most influential features were smoking consumption, sex, cholesterol, age, and albumin globulin ratio. The SHAP summary and dependence plot have been used to explain the affection of the 12 features to this model and how a single feature influences the output, respectively.ConclusionWe established EGFR mutation prediction models by MLAs and revealed that the RF was preferred, AUC (0.771, 95% CI: 0.770, 0.772), which was better than the traditional models. Therefore, the artificial intelligence–based MLA predicting model may become a practical tool to guide in diagnosis and therapy of LA.

Predicting EGFR mutation, ALK rearrangement, and uncommon EGFR mutation in NSCLC patients by driverless artificial intelligence: a cohort study

BackgroundTimely identification of epidermal growth factor receptor (EGFR) mutation and anaplastic lymphoma kinase (ALK) rearrangement status in patients with non-small cell lung cancer (NSCLC) is essential for tyrosine kinase inhibitors (TKIs) administration. We aimed to use artificial intelligence (AI) models to predict EGFR mutations and ALK rearrangement status using common demographic features, pathology and serum tumor markers (STMs).MethodsIn this single-center study, demographic features, pathology, EGFR mutation status, ALK rearrangement, and levels of STMs were collected from Wuhan Union Hospital. One retrospective set (N = 1089) was used to train diagnostic performance using one deep learning model and five machine learning models, as well as the stacked ensemble model for predicting EGFR mutations, uncommon EGFR mutations, and ALK rearrangement status. A consecutive testing cohort (n = 1464) was used to validate the predictive models.ResultsThe final AI model using the stacked ensemble yielded optimal diagnostic performance with areas under the curve (AUC) of 0.897 and 0.883 for predicting EGFR mutation status and 0.995 and 0.921 for predicting ALK rearrangement in the training and testing cohorts, respectively. Furthermore, an overall accuracy of 0.93 and 0.83 in the training and testing cohorts, respectively, were achieved in distinguishing common and uncommon EGFR mutations, which were key evidence in guiding TKI selection.ConclusionsIn this study, driverless AI based on robust variables could help clinicians identify EGFR mutations and ALK rearrangement status and provide vital guidance in TKI selection for targeted therapy in NSCLC patients.