Radiomics using contrast-enhanced T1-weighted imaging and clinical features for predicting response to EGFR-TKI in EGFR-mutated non-small cell lung cancer patients with brain metastases

Radiomics using contrast-enhanced T1-weighted imaging and clinical features for predicting response to EGFR-TKI in EGFR-mutated non-small cell lung cancer patients with brain metastases | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Radiomics using contrast-enhanced T1-weighted imaging and clinical features for predicting response to EGFR-TKI in EGFR-mutated non-small cell lung cancer patients with brain metastases Lian-Yu Sui, Tian-Ye Zhang, Cheng Cheng, Li-Hong Xing, Huan Meng, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8521529/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 12 You are reading this latest preprint version Abstract Objectives To construct and validate a model based on clinical characteristics and magnetic resonance imaging (MRI) radiomics to predict 1-year efficacy of epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) in patients with EGFR-mutant non-small cell lung cancer (NSCLC) brain metastases (BMs). Methods This study retrospectively analyzed data from 338 patients with EGFR-mutant NSCLC BMs from three centers, including MRI, clinical and pathological data, and radiological features. Based on the selected significant intratumoral features, seven machine learning algorithms were applied to compare model efficacy, and the best algorithm was selected for model construction. In the model predicting 1-year therapeutic efficacy, clinical, radiomic, and combined models were constructed separately. The model performance was evaluated using receiver operating characteristic curves. Results The final development cohort comprised 285 patients from Center 1, while the external validation set included 57 patients from Centers 2 and 3. In the model predicting 1-year EGFR-TKI efficacy, the random forest algorithm, which showed the best application, was used to construct the model. Compared with the radiomic and clinical models, the combined model exhibited superior area under the curve performance in the test set (0.756 vs 0.644 vs 0.668). In the external validation set, the combined model achieved an area under the curve of 0.743. Conclusion Compared to single clinical or radiomic models, the combined model was more effective in predicting the 1-year efficacy of EGFR-TKIs in patients with NSCLC BMs with EGFR mutations. Brain metastases non-small cell lung cancer EGFR-TKIs Clinical characteristics MRI Radiomics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Brain metastases (BMs) are secondary tumors originating from solid organs outside the central nervous system and represent the most common type of brain tumor in adults [ 1 ] . During the clinical course, 10–40% of patients with solid tumors develop BMs, and the median survival for patients with synchronous BMs from almost all primary sites does not exceed 12 months [ 2 ] . Non-small cell lung cancer (NSCLC) is the primary source of BMs, accounting for ~ 50% of cases [ 3 ] . BMs are typically associated with poor prognosis and high mortality in lung cancer. Therefore, precise diagnosis and treatment are important for clinical management of BMs. Epidermal growth factor receptor (EGFR) mutations are widely recognized as the most common genetic alteration in NSCLC. Compared to NSCLC without BMs, the frequency of EGFR mutations is higher in NSCLC with BMs, and patients with NSCLC harboring EGFR mutations are more prone to developing BMs than those with wild-type EGFR [ 4 ] . Given the high sensitivity and specificity of EGFR tyrosine kinase inhibitors (TKIs) in treating EGFR-mutated lung cancer, and their increasingly vital role in BM management, these agents are associated with favorable prognosis [ 5 ] . EGFR-TKIs therapy has significantly improved survival and life expectancy of patients with EGFR-mutated NSCLC [ 6 , 7 ] . However, due to tumor heterogeneity and patient-specific factors, there is considerable variability in the sensitivity of NSCLC BMs to EGFR-TKIs. Therefore, personalized and precise diagnosis and treatment of BMs are crucial for improving prognosis and quality of life. Magnetic resonance imaging (MRI) offers the advantage of noninvasively and repeatedly visualizing comprehensive, three-dimensional monitoring for intra-, inter-, and peritumoral information. This effectively compensates for deficiencies such as difficulties in obtaining adequate tissue samples, high costs, and long turnaround times for results. Compared to tissue-based molecular analyses, imaging remains the primary means for noninvasive monitoring of therapeutic responses and surveillance of disease progression [ 8 ] . Contrast-enhanced T1-weighted imaging (CE-T1WI) is the preferred conventional method for diagnosing BMs because it enhances the contrast between BMs and surrounding tissues. It can reflect signal variations caused by differences in the marginal regions of BMs and the degree of tumor enhancement, making it more advantageous for capturing lesion heterogeneity [ 9 ] . CE-T1WI is the most frequently applied sequencing method for diagnosis and treatment of BMs [ 10 ] . CE-T1WI has significant potential for visualizing and quantifying pathophysiological changes in BMs before and after treatment, and can be used to assess the therapeutic efficacy of EGFR-TKIs. MRI is an indispensable component of BMs diagnosis and post-treatment monitoring; however, current applications remain at a macroscopic level, primarily limited to subjective observations. In contrast to the limited information provided by traditional MRI diagnostic methods, MRI radiomics can automatically quantify tumor heterogeneity, thereby improving the accuracy of diagnostic and prognostic models. The visualization of tumor heterogeneity also aids physicians in formulating more precise treatment plans. Therefore, by studying characteristic changes in tumors on MR images, radiomics methods can predict tumor sensitivity to molecular therapies. Such techniques contribute to making more optimized clinical decisions without incurring additional costs [ 11 ] . There is a high incidence of EGFR-mutated NSCLC with BMs, and BMs are an indicator of therapeutic efficacy; therefore, MRI of BMs has the potential to identify novel biomarkers for predicting response to EGFR-TKIs. Previous studies, in predicting the therapeutic efficacy of EGFR-TKIs, have primarily relied on constructing models using radiomic features alone, overlooking the potential complementary value of clinical variables. Integration of clinical characteristics with radiomic analysis can enhance the robustness of models and facilitate exploration of their clinical translational value [ 12 ] . The application of MRI radiomics, a noninvasive, quantifiable, and reproducible intelligent method, to evaluate the therapeutic efficacy of malignant tumors in clinical settings has become an inevitable trend. Predictive models constructed based on radiomic features and clinical characteristics demonstrate significant promise in forecasting therapeutic outcomes. The incorporation of radiomic features into clinical models can enhance prediction, thereby improving clinical decision-making without incurring additional costs. This approach is conducive to optimizing treatment plans tailored to the individual needs and risk profiles of patients [ 13 , 14 ] . Therefore, the present study extracted radiomic features from MRI scans of patients with EGFR-mutated NSCLC and BMs, integrated multiple data sources, and developed and externally validated a radiomic framework in conjunction with clinical characteristics to predict the therapeutic response to EGFR-TKIs. This study also showed the potential clinical value of the combined model in guiding personalized treatment strategies. Materials and methods Inclusion and exclusion criteria of patients This was a retrospective analysis of brain MRI data from 230 patients treated at Center 1 from April 2016 to January 2024, who served as the primary cohort. The training and internal test sets of the primary cohort were allocated in an 8:2 ratio for model development and validation. Additionally, 80 NSCLC patients with BMs treated at Center 2 from June 2020 to January 2024 and Center 3 from November 2021 to March 2024 were combined into a new dataset as the external validation set to balance the distribution of demographic and clinicopathological factors. The Institutional Review Board waived the requirement for written consent because of the retrospective nature of the study and the absence of identifiable patient information. Patients with EGFR-mutated NSCLC received EGFR-TKI therapy in accordance with current clinical guidelines. For each patient, the response to EGFR-TKI treatment was assessed using the response evaluation criteria in solid tumors (RECIST) version 1.1 [ 15 ] . Clinical characteristics, including age, gender, EGFR mutation type, EGFR-TKI type, extracranial metastasis, number of BMs, size of BMs, and peritumoral edema, were obtained from clinical and pathological records, as well as baseline cranial MRI images. The therapeutic efficacy of EGFR-TKIs was determined by evaluating changes in the patients' condition within 1 year after initiation of EGFR-TKI treatment. Intracranial efficacy was assessed using the Response Assessment in Neuro-Oncology (RANO) [ 16 ] criteria, categorized as intracranial complete response (iCR), intracranial partial response (iPR), intracranial stable disease (iSD), and intracranial progressive disease (iPD). The intracranial objective response rate represented the proportion of patients achieving iCR and iPR, while the intracranial disease control rate represented the proportion of patients achieving iCR, iPR, and iSD. Patients with iCR and iPR were classified into the effective treatment group, whereas those with iSD and iPD were classified into the ineffective treatment group. The inclusion criteria were: (1) definitive diagnosis of NSCLC was established through lung biopsy or open surgery; (2) enhanced brain MRI was performed at our hospital within 2 weeks prior to treatment; (3) brain MRI revealed at least one BM; (4) EGFR mutation was confirmed by genetic testing, and the patient received EGFR-TKI treatment; and (5) complete medical records were available, including treatment information and monitoring examination results. The exclusion criteria were: (1) patients did not receive EGFR-TKIs after onset of BMs; (2) poor image quality affected quantitative analysis; (3) incomplete clinical information; (4) other concurrent cancers; and (5) patients who simultaneously received local treatment, such as whole-brain radiotherapy and surgical resection. For more detailed information on the inclusion and exclusion criteria for NSCLC BMs in this study, please refer to Fig. 1 . MRI All three research centers utilized 1.5 T or 3 T MR scanners to perform cranial MRI. After administering a weight-adjusted dose of 0.1 mmol/kg gadobenate dimeglumine contrast agent via an elbow vein using a high-pressure injector at an injection rate of 2 ml/s, axial CE-T1WI sequences were acquired. The imaging protocols varied among the different scanners at each research center. Detailed scanning parameters are provided in Table 1 . Radiomics analysis MRI segmentation Figure 2 illustrates the specific workflow of radiomic analysis. All imaging data were retrieved from the Picture Archiving and Communication System and stored in DICOM format. Two radiologists used ITK-SNAP version 3.8.0 ( www.itksnap.org ) to delineate regions of interest (ROIs) on axial CE-T1WI scans. A computer-generated random number table was used to select images from 30 patients for intra- and interobserver consistency analysis. The imaging data of these 30 patients were used for tumor ROI delineation by two radiologists who were blinded to the patients' pathological results. One of the radiologists repeated the tumor ROI delineation 1 month after initial delineation. Interobserver consistency analysis was based on the data from the initial delineation by both radiologists, while intraclass correlation coefficient (ICC) was conducted using the data from the two delineations performed by the same radiologist (with 10 years of experience). Radiomic feature extraction and selection The original images underwent N4_Normalization processing, and radiomic features were extracted from the ROIs of each patient using Pyradiomics version 3.1.0. The radiomic features extracted from each ROI included: morphology (e.g., volume, surface area, and diameter); first-order statistical features (e.g., mean, standard deviation, and entropy); and second-order features. The second-order features included: gray-level co-occurrence matrix (describing gray-level spatial correlation); gray-level run-length matrix (analyzing texture roughness); gray-level size zone matrix (quantifying distribution of region sizes); gray-level dependence matrix (describing local gray-level variations); and neighborhood gray-tone difference matrix (analyzing local heterogeneity). Firstly, to eliminate interference caused by outliers during Z-score normalization, all radiomic features were normalized to ensure that feature selection was not affected by large differences in magnitude. Secondly, an ICC consistency test was conducted, and features with an ICC threshold > 0.8 were retained. From each patient, 1505 features were extracted, and 430 features with an ICC intra- and interclass correlation coefficient < 0.8 were deleted, leaving 1075 features for further selection. Subsequently, Spearman correlation analysis was used for initial dimensionality reduction of the radiomic features, eliminating those with an absolute correlation value > 0.9. A recursive feature elimination algorithm was used to progressively eliminate unimportant features. Finally, the least absolute shrinkage and selection operator (LASSO) combined with 10-fold cross-validation was utilized for final feature dimensionality reduction, resulting in selection of the optimal features. Construction and validation of predictive models Model training Different combinations of feature types and classifiers can be customized to suit specific classification tasks. The motivation for using multiple classifiers was to enhance the robustness of the results, as previous studies have indicated that selecting an appropriate algorithm plays a crucial role in classification outcomes [ 17 , 18 ] . This study recognized that no single model in machine learning is universally applicable, and explored various classifiers to develop a high-performance and robust model for predicting the therapeutic efficacy of EGFR-TKIs in NSCLC BMs. To adhere to best ML practices, a combined classifier comprising seven commonly used ML algorithms was adopted, comprising: logistic regression as a linear model; random forest (RF) as an ensemble method; k-nearest neighbors for instance-based modeling; linear support vector machine; decision tree as a tree-structured supervised learning model; Gaussian naïve Bayes as a probability classification model; and extreme gradient boosting as a gradient boosting framework. Subsequently, the seven ML models were evaluated, and their efficacies were comprehensively assessed through area under the curve (AUC) analysis, decision curve analysis to determine standardized net benefit, and calibration curves for each trained model to identify the optimal model. After performing feature extraction and selection from within the BMs, as well as from automatically expanded 3-, 5-, and 8-mm peritumoral regions, radiomic models were constructed based on optimal ML algorithms according to model training, and their performances were compared. Predictive model construction Clinical and pathological data were retrospectively collected from electronic medical records and digital pathology systems across various centers, including: age; gender; smoking status; Karnofsky Performance Status score; primary tumor site; pathological classification; tumor markers [carcinoembryonic antigen, neuron-specific enolase, cytokeratin 19 fragment (CY211)]; extracranial metastasis; TNM staging (describing the size and invasiveness of the primary tumor, lymph node metastasis, and systemic metastasis); EGFR mutation subtype; and EGFR-TKI type. Additionally, MRI morphological features commonly used in clinical assessments of central nervous system tumors were analyzed, including; number and grouping of BMs; location, maximum diameter, and grouping of the largest lesion; enhancement pattern; and peritumoral edema. Logistic regression analysis was conducted to select clinically significant features ( P < 0.05) for model construction. Finally, radiomic, clinical, and combined models were constructed based on the optimal ML algorithm. The optimal model was determined by comparing the AUCs using the Delong test, and the model's generalization capability was further evaluated in an external validation set. Statistical analysis We used SPSS 22.0 software (IBM Corporation, Chicago, IL, USA) for statistical analysis. In descriptive statistics, continuous variables were expressed as mean ± standard deviation or median (interquartile range), while categorical variables were presented as frequencies and percentages. For analysis of differences among the training, test, and external validation sets, continuous variables were compared using Student's t test or Mann–Whitney U test depending on the data type, while categorical variables were compared using χ 2 test or Fisher's exact test. The discriminatory ability of predictive models was evaluated and compared using the 95% confidence interval (CI) of the area under the receiver operating characteristic curve (ROC). Model performance was further assessed by calculating AUC, accuracy (ACC), sensitivity (SEN), specificity (SPE), and F1 score. The Delong test was used to compare the AUC values of different models and evaluate their robustness. The performance of radiomic, clinical, and combined models was evaluated separately in the test and external validation sets. Calibration curves were used to assess the deviation in the distribution of predicted values, reflecting the consistency between model predictions and actual probabilities. Based on this, decision curve analysis was conducted to calculate the net benefit at different threshold probabilities, thereby evaluating the model's value in differential diagnosis. All statistical analyses were performed with a significance level set at P < 0.05. Results Patient basic information and treatment efficacy A total of 338 patients from three centers were included. Table 2 summarizes the clinical characteristics, MRI features, and therapeutic efficacy of EGFR-TKIs. The clinical features included KPS score, extracranial metastases, M staging of the TNM classification, and clinical staging. The MRI morphological features included the number of lesions, necrosis, and EGFR-TKI type. Significant differences were observed across the training, test, and external validation sets (all P < 0.05). The assessment showed that 39 patients achieved iCR, 104 iPR, 60 iSD, and 131 iPD. The intracranial objective response rate was 42.3% and intracranial disease control rate was 60.1%. Predictive models evaluation A higher AUC indicated greater classification accuracy, while a smaller AUC gap between the training and test sets suggested stronger model generalizability and lower risk of overfitting. The model training results (Fig. 3 ) indicated that the radiomic model constructed based on the RF algorithm emerged as the top performer. Comparison of RF models derived from intratumoral features of BMs and extended 3-, 5-, and 8-mm peritumoral regions revealed that intratumoral features demonstrated better AUC performance in the test set (Fig. 4 ). Therefore, intratumoral features were selected for constructing the radiomic model. LASSO was applied to determine the optimal parameter value (λ = 0.0285) based on intratumoral features of BMs, ultimately selecting eight significant features that made important contributions to the classification of the predictive model (Fig. 5 ). Three meaningful clinical features (age, CY211, and EGFR-TKI grouping) were identified through univariate logistic regression (Table 3 ). The ROC curves of the clinical, radiomic, and combined model constructed based on RF in the training and test sets are presented in Fig. 6 a, b. Figure 6 c, d further clarifies the performance differences among the three models in predicting the therapeutic efficacy of EGFR-TKIs. The corresponding AUC, ACC, SEN, SPE, and F1 scores of the three predictive models in the training and test sets were calculated to assess their quantitative classification performance (Table 4 ). In the predictive model for EGFR-TKI therapeutic efficacy within 1 year, the combined model based on CE-T1WI demonstrated the best performance compared to the radiomic and clinical models. The AUC performance of the combined model in the test set outperformed that of the radiomic and clinical models (0.756 vs 0.644 vs 0.668). The Delong test confirmed that the combined model achieved impressive predictive performance (Z-score: −2.166 vs −4.640, P = 0.030 vs < 0.001). It also obtained optimal metrics in the test set, with an AUC of 0.756 (95% CI: 0.627–0.886), ACC 0.702, SEN 0.844, SPE 0.520, and F1 score 0.683. In the external validation set, the ROC curve of the combined model is shown in Fig. 7 a, with an AUC of 0.743 (95% CI: 0.604–0.881), ACC 0.679, SEN 0.480, SPE 0.857, and F1 score 0.662. These results confirmed the reliability and generalizability of the combined model in predicting the therapeutic efficacy of EGFR-TKIs. The calibration curves of the combined model in the training, test, and external validation sets are shown in Fig. 7 b, verifying the reliability and generalizability of the fused model. Discussion In this comprehensive study, we utilized multicohort data from patients with EGFR-mutant NSCLC BMs from three institutions to develop and externally validate an MRI radiomic framework for predicting the therapeutic response to EGFR-TKIs. The radiomic phenotypic features proposed demonstrated excellent predictive capabilities. We discovered a synergistic effect between radiomic features and clinical characteristics. The integration of clinical features and MRI radiomics ultimately facilitated the development of a combined model that used multiple data sources to optimize predictive performance. The validation framework underscored the pivotal role of radiomic features extracted from MRI. These features provide complementary insights to established MRI biomarkers, bringing us closer to achieving personalized treatment strategies for NSCLC BMs. Compared to traditional visual assessment methods based on enhanced MRI of BMs, the radiomic model exhibited significant advantages in terms of accuracy and sensitivity in predicting the therapeutic response of BMs, thereby aiding clinicians in formulating more personalized treatment plans and promoting the application and development of radiomic approaches. MRI radiomics can help predict the efficacy of targeted therapy, serving as a valuable tool for clinicians and patients. It provides clinicians with valuable, objective, and consistent information to determine optimal treatment strategies and patient follow-up plans. The motivation for using multiple classifiers was to enhance the robustness of the results, as previous studies have indicated that selecting an appropriate algorithm plays a crucial role in classification outcomes [ 19 , 20 ] . Zhao et al. [ 21 ] constructed support vector machine, RF, and k-nearest neighbors models using LASSO based on CE-T1WI and T2-fluid-attenuated inversion recovery sequences to differentiate between primary central nervous system lymphoma and BMs. The RF model exhibited the best performance (AUC = 0.73). Kanakarajan et al. [ 22 ] developed a radiomic model based on CE-T1WI to predict the efficacy of radiotherapy for BMs. By incorporating clinical features, the combined model using an RF classifier achieved the highest AUC of 0.89, with an accuracy rate of 87%. Similar studies [ 23 ] have shown that radiomic models constructed using an RF classifier also achieved favorable performance in predicting the efficacy of radiotherapy for BMs, with an AUC of up to 0.83. In summary, the RF classifier has demonstrated excellent performance in predicting the therapeutic efficacy for BMs, which agrees with the findings of our study. Our model training also indicated the significant efficacy of the RF classifier. Previous studies have shown that extracting radiomic features from the tumor region and peritumoral edema of BMs is reasonable and reliable. This approach can encompass most of the details in the heterogeneous regions of BMs and provide better prediction of therapeutic efficacy in patients with BMs, compared with features from a single region [ 24 ] . Fan et al. [ 25 ] also indicated that the tumor region and peritumoral edema in NSCLC BMs were complementary, allowing for extraction of rich radiomic phenotypic features to enhance model robustness and achieve more reliable prediction of the therapeutic efficacy of EGFR-TKIs. Based on previous studies, we constructed models using intratumoral and peritumoral features based on the optimal ML algorithm. However, the results indicated that the model constructed using features extracted from within the tumor exhibited better performance. This may have been because NSCLC patients with EGFR mutations often develop multiple small BMs, with mild or no peritumoral edema [ 26 ] . Therefore, it is impossible to extract critical information from the peritumoral edema area pertaining to the therapeutic efficacy against BMs. Among patients with NSCLC BMs, those with EGFR mutations have longer survival and better control of intracranial disease [ 27 ] . Fan et al. [ 25 ] incorporated pretreatment CE-T1WI and T2WI from NSCLC BMs patients and constructed models based on tumor area analysis, peritumoral edema, margin/background parenchyma, and multiregional fusion to predict the therapeutic response to EGFR-TKIs. The fusion model achieved the best performance in the external validation set, proving that subregional radiomics, as a novel noninvasive approach, can help guide personalized treatment strategies with EGFR-TKIs for NSCLC BMs. However, our study only utilized radiomic features to construct models for predicting the therapeutic efficacy of EGFR-TKIs using MRI radiomics, neglecting the potential complementary value of clinical variables. A predictive model combining radiomics with clinical predictive features will enable earlier detection of disease progression and benefit patients by guiding clinicians to conduct more intensive MRI follow-up of patients at high risk of intracranial progression. Similarly, Qu et al. [ 28 ] included 212 lesions from 70 patients with NSCLC BMs who received first-line EGFR-TKI treatment. Radiomic features were extracted from the brain tumor regions on pretreatment CE-T1WI, and a radiomic score was calculated based on the selected features. The combined predictive model incorporating EGFR-19del mutation, third-generation EGFR-TKI treatment, and the average rad-score outperformed models constructed using each of these three predictors alone. The combined model demonstrated good predictive value for intracranial progression within 1 year after EGFR-TKI treatment. The results confirmed that the nomogram combining clinical and radiomic features outperformed nomograms based solely on either features alone. Unlike the aforementioned validation studies, in our study, age, EGFR-TKIs, and CY211 were identified as significant clinical features. CY211 was the most sensitive tumor marker in NSCLC, which is consistent with the literature [ 29 ] . Qi et al. [ 30 ] included 117 patients with NSCLC BMs for individualized prediction of the therapeutic efficacy of first-generation EGFR-TKIs. The LASSO algorithm was used to screen radiomic features extracted from multiparametric MRI. In the short-term efficacy predictive model, clinical, radiomic, and combined nomograms were constructed separately. The results demonstrated good consistency between predicted risk and actual outcomes, confirming that the combined nomogram outperformed nomograms based solely on clinical or radiomic features, with a C-index of 0.843. Qi et al. [ 31 ] subsequently developed and validated a nomogram based on clinical characteristics and MRI radiomics for predicting the short-term efficacy of third-generation EGFR-TKIs in patients with EGFR-mutant lung adenocarcinoma BMs, with a C-index of 0.803. The studies confirmed the favorable efficacy of first- and third-generation EGFR-TKIs in treating EGFR-mutant NSCLC BMs. The predictive performance of nomograms established by combining radiomics and clinical features outperformed nomograms based solely on radiomics or clinical features, and served as a noninvasive predictive tool that aided timely personalized adjustment of treatment plans. Association of radiomic characteristics with clinical features is an emerging field that enhances the correlation between radiomic features and molecular characteristics, potentially reflecting gene expression differences within and between tumors. The aforementioned studies indicate that MRI radiomics combined with clinical features can provide noninvasive and intelligent prediction of the therapeutic efficacy of EGFR-TKIs in EGFR-mutant NSCLC BMs. However, because of the design of case follow-up, the research samples were from a single center and limited in number, which restricts the generalizability and clinical effectiveness of the results. Despite the use of cross-validation and the good performance of the models in the training and test sets, the generalizability of the models to unseen data requires further investigation. In this study, radiomic analysis was performed using pretreatment CE-T1WI from NSCLC BM patients. To maximize the sample size, patients with different EGFR mutation types receiving different EGFR-TKIs were included. Based on a multi-institutional and large sample size, combined with clinical data, the effectiveness of EGFR-TKIs in local tumor control was predicted. The results indicated that radiomic features combined with clinical features served as complementary biomarkers for predicting the therapeutic efficacy in NSCLC BM patients, achieving an AUC of 0.743 in the external validation set, demonstrating good generalizability. Our study demonstrated that combining MRI radiomics with clinical features improved the accuracy of EGFR-TKI therapeutic monitoring based on imaging, and supported preoperative diagnosis, treatment planning, and outcome determination of BMs. Our study had several limitations. Firstly, although this was a multicenter study with an external validation set, the model was developed based on retrospective data, which had the inherent biases of retrospective analyses and challenges of adequately controlling for confounding variables. Future confirmation through prospective studies is needed to refine and evaluate the clinical utility of the model. Secondly, the EGFR mutation status data were derived from primary lung cancer tissues, which may not directly reflect secondary mutations occurring in metastatic lesions. Thirdly, the MRI equipment and scanning parameters showed heterogeneity, and the treatment strategies received by patients were diverse and at different stages of the disease. These factors may all affect the generalizability of our findings. Additionally, this study was based on a single CE-T1WI sequence, as some patients did not undergo scanning with other conventional sequences. However, diffusion-weighted imaging and diffusion tensor imaging are also recommended as standards for evaluating BMs. Incorporating more sequences into future predictive models may enhance their predictive capabilities. Finally, future research should integrate multimodal imaging and data for in-depth exploration and methodological optimization to improve model calibration. Additionally, a thorough analysis of the associations between radiomic features and tumor biological characteristics will contribute to a better understanding of tumor pathogenesis and treatment response mechanisms, thereby facilitating clinical translation. Conclusion This study developed and externally validated a model that combined clinical and radiomic features to support treatment decision-making in patients with NSCLC BMs. The results provide robust evidence for predicting the therapeutic efficacy of EGFR-TKIs and add value to targeted treatment of patients. However, it is important to note that these findings require further confirmation through future prospective studies to enhance their clinical generalizability. Abbreviations ACC, Accuracy AD, adenocarcinoma AUC, area under the curve BMs, Brain metastasis CEA, Carcinoembryonic antigen CE-T1WI, Contrast-enhanced T1 weighted imaging CY211, Cytokeratin 19 fragment DT, Decision tree EGFR, epidermal growth factor receptor EGFR-TKIs, EGFR tyrosine kinase inhibitors GLCM, Gray-level co-occurrence matrix GLDM, Gray-level dependence matrix GLRLM, Gray-level run-length matrix GLSZM, Gray-level size zone matrix GNB, gaussian naive bayes ICC, Intraclass correlation coefficient iCR, Intracranial complete response iDCR, Intracranial disease control rate iORR,Intracranial objective response rate iPD, Intracranial progressive disease iPR, Intracranial partial response iSD, Intracranial stable disease KNN, K-nearest neighbors LASSO, Least absolute shrinkage and selection operator LR, Logistic regression MRI, Magnetic resonance imaging NGTDM, Neighboring gray-tone difference matrix NSE, Neuron-specific enolase NSCLC, Non-small-cell lung cancer OR, Odds ratio RECIST, Response evaluation criteria in solid tumors ROI, Region of interest SCC, Squamous cell cancer SPE, Specificity SEN, Sensitivity SVM, Support vector machine XGB, Extreme gradient boosting Declarations Ethics approval and consent to participate This retrospective study was approved by the institutional review board (“Ethics committee of the Affiliated Hospital of Hebei University”, Dongfeng Road, No. 212, Lianchi District, Baoding 071000, China, Number HDFYLL-KY-2024-037; date of approval 02/28/2024)., was conducted in accordance with the Declaration of Helsinki. Informed consent was waived for all patients due to the retrospective nature of this study. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Funding This study was supported by the Hebei University Graduate Student Innovation Funding Project (CXZZBS2025028). Author Contribution Conception and design of the research: XPY, JNW. Acquisition of data: LYS. Analysis and interpretation of the data: LYS. Statistical analysis: LYS. Writing of the manuscript: LYS. Critical revision of the manuscript for intellectual content: LYS, LHX, HM, YZ, QW, CL and TSZ. All authors read and approved the final draft. Acknowledgements None. Data Availability The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request. References ACHROL A S, RENNERT R C, ANDERS C, et al. Brain metastases [J]. Nat Reviews Disease Primers. 2019;5(1):5. SUNG K S. Clinical Practice Guidelines for Brain Metastasis From Solid Tumors [J]. Brain Tumor Res Treat. 2024;12(1):14–22. LAMBA N, WEN P Y, AIZER AA. Epidemiology of brain metastases and leptomeningeal disease [J]. Neurooncology. 2021;23(9):1447–56. GE M, ZHUANG Y, ZHOU X, et al. High probability and frequency of EGFR mutations in non-small cell lung cancer with brain metastases [J]. J Neurooncol. 2017;135(2):413–8. KHANDEKAR MJ, PIOTROWSKA Z, WILLERS H, et al. Role of Epidermal Growth Factor Receptor (EGFR) Inhibitors and Radiation in the Management of Brain Metastases from EGFR Mutant Lung Cancers [J]. Oncologist. 2018;23(9):1054–62. SIEGEL R L, KRATZER T B, GIAQUINTO A N, et al. Cancer statistics, 2025 [J]. CA Cancer J Clin. 2025;75(1):10–45. BOIRE A, BRASTIANOS P K, GARZIA L, et al. Brain metastasis [J]. Nat Rev Cancer. 2020;20(1):4–11. KHALIGHI S, REDDY K, MIDYA A, et al. Artificial intelligence in neuro-oncology: advances and challenges in brain tumor diagnosis, prognosis, and precision treatment [J]. NPJ Precis Oncol. 2024;8(1):80. DREGELY I, KELLY-MORLAND C PREZZID, et al. Imaging biomarkers in oncology: Basics and application to MRI [J]. J Magn Reson Imaging. 2018;48(1):13–26. LE RHUN E, GUCKENBERGER M, SMITS M, et al. EANO-ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up of patients with brain metastasis from solid tumours [J]. Ann Oncol. 2021;32(11):1332–47. LAMBIN, P, LEIJENAAR R T H, DEIST T M, et al. Radiomics: the bridge between medical imaging and personalized medicine [J]. Nat Rev Clin Oncol. 2017;14(12):749–62. ZHANG X, ZHANG G, QIU X, et al. Non-invasive decision support for clinical treatment of non-small cell lung cancer using a multiscale radiomics approach [J]. Radiother Oncol. 2024;191:110082. WANG T-W, CHAO H-S, CHIU H-Y, et al. Radiomics of metastatic brain tumor as a predictive image biomarker of progression-free survival in patients with non-small-cell lung cancer with brain metastasis receiving tyrosine kinase inhibitors [J]. Translational Oncol. 2024;39:101826. MOURAVIEV A, DETSKY J, SAHGAL A, et al. Use of radiomics for the prediction of local control of brain metastases after stereotactic radiosurgery [J]. Neuro Oncol. 2020;22(6):797–805. LIN N U, LEE E Q AOYAMAH, et al. Response assessment criteria for brain metastases: proposal from the RANO group [J]. Lancet Oncol. 2015;16(6):e270–8. OCAñA-TIENDA B, PéREZ-BETETA J, ROMERO-ROSALES J A, et al. Volumetric analysis: Rethinking brain metastases response assessment [J]. Neurooncol Adv. 2024;6(1):vdad161. XIA X, QIU J, TAN Q, et al. Radiomics-Based Differentiation of Primary Central Nervous System Lymphoma and Solitary Brain Metastasis Using Contrast-Enhanced T1-Weighted Imaging: A Retrospective Machine Learning Study [J]. Acad Radiol. 2025;32(9):5401–12. XIA X, WU W. Interpretable Machine Learning Models for Differentiating Glioblastoma From Solitary Brain Metastasis Using Radiomics [J]. Acad Radiol. 2025;32(9):5388–400. ARTZI M, BRESSLER I, BEN BASHAT D. Differentiation between glioblastoma, brain metastasis and subtypes using radiomics analysis [J]. J Magn Reson Imaging. 2019;50(2):519–28. ORTIZ-RAMON R, LARROZA A et al. ARANA E,. A radiomics evaluation of 2D and 3D MRI texture features to classify brain metastases from lung cancer and melanoma [J]. Annu Int Conf IEEE Eng Med Biol Soc, 2017, 2017: 493-6. ZHAO LM, XIE F F HUR, et al. Radiomic-Based MRI for Classification of Solitary Brain Metastases Subtypes From Primary Lymphoma of the Central Nervous System [J]. J Magn Reson Imaging. 2023;57(1):227–35. KANAKARAJAN H, DE BAENE W, SITSKOORN M et al. Predicting local control of brain metastases after stereotactic radiosurgery with clinical, radiomics and deep learning features [J]. medRxiv, 2024: 2024.05. 13.24307241. RESSA G, LEVI R, SAVINI G, et al. AI differentiates radionecrosis from true progression in brain metastasis upon stereotactic radiosurgery: analysis of 124 histologically assessed lesions [J]. Neuro Oncol; 2025. XU J, WANG P, LI Y, et al. Development and validation of an MRI-Based nomogram to predict the effectiveness of immunotherapy for brain metastasis in patients with non-small cell lung cancer [J]. Front Immunol. 2024;15:1373330. FAN Y, WANG X, DONG Y, et al. Multiregional radiomics of brain metastasis can predict response to EGFR-TKI in metastatic NSCLC [J]. Eur Radiol. 2023;33(11):7902–12. CHAMBERLAIN M C, BAIK C S, GADI V K, et al. Systemic therapy of brain metastases: non–small cell lung cancer, breast cancer, and melanoma [J]. Neurooncology. 2016;19(1):i1–24. EICHLER A F, KAHLE K T, WANG D L, et al. EGFR mutation status and survival after diagnosis of brain metastasis in nonsmall cell lung cancer [J]. Neuro Oncol. 2010;12(11):1193–9. QU J, ZHANG T, ZHANG X, et al. MRI radiomics for predicting intracranial progression in non-small-cell lung cancer patients with brain metastases treated with epidermal growth factor receptor tyrosine kinase inhibitors [J]. Clin Radiol. 2024;79(4):e582–91. LIU L, TENG J, ZHANG L et al. The Combination of the Tumor Markers Suggests the Histological Diagnosis of Lung Cancer [J]. Biomed Res Int, 2017, 2017: 2013989. QI H, HOU Y, ZHENG Z, et al. MRI radiomics predicts the efficacy of EGFR-TKI in EGFR-mutant non-small-cell lung cancer with brain metastasis [J]. Clin Radiol. 2024;79(7):515–25. QI H, HOU Y, ZHENG Z, et al. Clinical characteristics and MRI based radiomics nomograms can predict iPFS and short-term efficacy of third-generation EGFR-TKI in EGFR-mutated lung adenocarcinoma with brain metastases [J]. BMC Cancer. 2024;24(1):362. Tables Table 1 to 4 are available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files Tables.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 12 Mar, 2026 Reviews received at journal 11 Mar, 2026 Reviewers agreed at journal 11 Mar, 2026 Reviews received at journal 11 Mar, 2026 Reviewers agreed at journal 10 Mar, 2026 Reviews received at journal 09 Mar, 2026 Reviewers agreed at journal 09 Mar, 2026 Reviewers invited by journal 12 Feb, 2026 Editor assigned by journal 12 Feb, 2026 Editor invited by journal 03 Feb, 2026 Submission checks completed at journal 02 Feb, 2026 First submitted to journal 02 Feb, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8521529","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":590341337,"identity":"eab35b2a-1fba-4195-b483-5fb90859572b","order_by":0,"name":"Lian-Yu Sui","email":"","orcid":"","institution":"Affiliated Hospital of Hebei University/ School of Clinical Medicine of Hebei University","correspondingAuthor":false,"prefix":"","firstName":"Lian-Yu","middleName":"","lastName":"Sui","suffix":""},{"id":590341339,"identity":"d4bc8d67-b6a7-4dea-b5eb-57bfa350872f","order_by":1,"name":"Tian-Ye Zhang","email":"","orcid":"","institution":"College of Quality and Technical Supervision, Hebei University","correspondingAuthor":false,"prefix":"","firstName":"Tian-Ye","middleName":"","lastName":"Zhang","suffix":""},{"id":590341341,"identity":"814bec23-3bdf-405f-bb38-321125f07560","order_by":2,"name":"Cheng Cheng","email":"","orcid":"","institution":"College of Quality and Technical Supervision, Hebei University","correspondingAuthor":false,"prefix":"","firstName":"Cheng","middleName":"","lastName":"Cheng","suffix":""},{"id":590341346,"identity":"a42b2ab8-ee06-49ef-9b67-b955b9d7acb3","order_by":3,"name":"Li-Hong Xing","email":"","orcid":"","institution":"Hebei Key Laboratory of Precise Imaging of Inflammation Related Tumors, the Affiliated Hospital of Hebei University","correspondingAuthor":false,"prefix":"","firstName":"Li-Hong","middleName":"","lastName":"Xing","suffix":""},{"id":590341351,"identity":"17305699-a7fa-4d8c-ad57-07d2f51ce094","order_by":4,"name":"Huan Meng","email":"","orcid":"","institution":"Hebei Key Laboratory of Precise Imaging of Inflammation Related Tumors, the Affiliated Hospital of Hebei University","correspondingAuthor":false,"prefix":"","firstName":"Huan","middleName":"","lastName":"Meng","suffix":""},{"id":590341356,"identity":"96f34abd-4c68-43dd-bf70-316231088b71","order_by":5,"name":"Chong Liu","email":"","orcid":"","institution":"Baoding First Central Hospital","correspondingAuthor":false,"prefix":"","firstName":"Chong","middleName":"","lastName":"Liu","suffix":""},{"id":590341359,"identity":"459377a6-6ded-4933-9828-9623b96e2054","order_by":6,"name":"Qi Wang","email":"","orcid":"","institution":"Tumor Hospital of Hebei Medical University","correspondingAuthor":false,"prefix":"","firstName":"Qi","middleName":"","lastName":"Wang","suffix":""},{"id":590341363,"identity":"151bba5e-7f08-4c43-a3c8-cfb60c8bd7dd","order_by":7,"name":"Jia-Ning Wang","email":"","orcid":"","institution":"Hebei Key Laboratory of Precise Imaging of Inflammation Related Tumors, the Affiliated Hospital of Hebei University","correspondingAuthor":false,"prefix":"","firstName":"Jia-Ning","middleName":"","lastName":"Wang","suffix":""},{"id":590341365,"identity":"1182abd8-eead-456c-8246-650f1f74c9d0","order_by":8,"name":"Tian-Shuo Z","email":"","orcid":"","institution":"Xiangya School of Medicine, Central South University","correspondingAuthor":false,"prefix":"","firstName":"Tian-Shuo","middleName":"","lastName":"Z","suffix":""},{"id":590341367,"identity":"14eb4961-a5c7-449c-8e0b-00c6e76d1d9c","order_by":9,"name":"Kun Liu","email":"","orcid":"","institution":"College of Quality and Technical Supervision, Hebei University","correspondingAuthor":false,"prefix":"","firstName":"Kun","middleName":"","lastName":"Liu","suffix":""},{"id":590341369,"identity":"cc21101b-fa42-46ec-859b-5e9e565b5e37","order_by":10,"name":"Xiao-Ping Yin","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9UlEQVRIiWNgGAWjYDACZiBOAGI2Bgb2Hx8qGHhI0sIgOeMMMVqQgTRvGxGqDI7zHpN4UGOTxyfdfsFw5rw6GXP2A4wfPubg0XKYL9kg4VhaMZvMmYKEj9sO81j2JDBLztyGTwuP4YMEtsOJbRI5CQdnbjvAY3AggY2ZF78WoJp/YC2Jzbxz6ngMzj8gqMXwQWIbSEv6YWbeBmYegxsEbJE8zGNskNgH9ItEDhvjjGNAS288bMbrF77zZ8wkf3yzyZOfkf6M4UNNnb3B+eSDHz7i0aJwAEInMDDwGEDFGBtwqwcC+Qa4FvYHeFWOglEwCkbByAUANGVSeMHXYCcAAAAASUVORK5CYII=","orcid":"","institution":"Hebei Key Laboratory of Precise Imaging of Inflammation Related Tumors, the Affiliated Hospital of Hebei University","correspondingAuthor":true,"prefix":"","firstName":"Xiao-Ping","middleName":"","lastName":"Yin","suffix":""}],"badges":[],"createdAt":"2026-01-05 12:53:06","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8521529/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8521529/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102862413,"identity":"7863038e-5c1a-49f2-b431-36748ff8274b","added_by":"auto","created_at":"2026-02-17 16:17:20","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":254154,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart of patient inclusion and exclusion criteria. BMs, brain metastases; EGFR, epidermal growth factor receptor; EGFR-TKIs, EGFR tyrosine kinase inhibitors; NSCLC, non-small cell lung cancer.\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8521529/v1/879f4b8d2f95857a0b4f525a.jpeg"},{"id":102862412,"identity":"105f1fec-99f3-4892-8edd-99b4550a23f0","added_by":"auto","created_at":"2026-02-17 16:17:20","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":196813,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart of the proposed model analyses. DT, decision tree; GLCM, gray-level co-occurrence matrix; GLDM, gray-level dependence matrix; GLRLM, gray-level run-length matrix; GLSZM, gray-level size zone matrix; GNB, Gaussian naive Bayes; KNN, k-nearest neighbors; LASSO, least absolute shrinkage and selection operator; NGTDM, Neighboring gray-tone difference matrix; LR, logistic regression; MRI, magnetic resonance imaging; RF, random forest; SVM, support vector machine; XGB, extreme gradient boosting.\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8521529/v1/a231f731b4b7d2ef502f6c0a.jpeg"},{"id":102862414,"identity":"c74bfff0-a7b9-4240-b251-43d20067bea7","added_by":"auto","created_at":"2026-02-17 16:17:20","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":145892,"visible":true,"origin":"","legend":"\u003cp\u003eML Model training. ROC curve analysis and AUC (a, b), decision curve analysis (DCA) (c, d), and calibration curve for the validation set (e) of the classifier models in the training and validation sets. The ROC curve describes the performance of the classifier models in predicting treatment efficacy. The AUC values for the training and test sets are provided to evaluate the discriminatory ability of the models. A higher AUC value indicates greater classification accuracy, while a smaller gap in AUC between the training and validation sets suggests stronger model generalization ability and a lower risk of overfitting. DCA directly calculates the practical value of the models under different decision thresholds. The calibration curve assesses the degree of match between the predicted probabilities and the actual probabilities for each model. AUC, area under the curve; DT, decision tree; GNB, Gaussian naive Bayes; KNN, k-nearest neighbors; LR, logistic regression; ML, machine learning; RF, random forest; SVM, support vector machine; XGB, extreme gradient boosting.\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8521529/v1/a955699d91e214d5290f7141.jpeg"},{"id":102964283,"identity":"3bc80374-6243-4d05-a9f7-285c105fbde0","added_by":"auto","created_at":"2026-02-19 04:21:59","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":82449,"visible":true,"origin":"","legend":"\u003cp\u003eRF-based radiomics prediction models for intratumoral and peritumoral (3/5/8 mm) regions. ROC curve and AUC (a) and decision curve (b) in the validation set. AUC, area under the curve; RF, random forest; ROC, receiver operating characteristic.\u003c/p\u003e","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8521529/v1/cab8dfd965eb027b91c56712.jpeg"},{"id":102963644,"identity":"7c15268a-68df-413b-a9d9-84b2b202611a","added_by":"auto","created_at":"2026-02-19 04:19:43","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":102867,"visible":true,"origin":"","legend":"\u003cp\u003eFeature selection using 10-fold cross-validation LASSO. Eight optimal features and their weights were selected (c). LASSO, least absolute shrinkage and selection operator.\u003c/p\u003e","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8521529/v1/6ca227b65b50f2ed0e434a17.jpeg"},{"id":102862416,"identity":"3ba42fc9-4159-4a06-b5e8-02ebabfb3edf","added_by":"auto","created_at":"2026-02-17 16:17:20","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":87645,"visible":true,"origin":"","legend":"\u003cp\u003ePerformance comparison of the RF-based prediction models in the training and validation cohorts. ROC curve analysis and AUC (a, b) and decision curve analysis (c, d) in the training and validation sets. AUC, area under the curve; RF, random forest; ROC, receiver operating characteristic.\u003c/p\u003e","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8521529/v1/65df17f5f4e3940e23f4c59f.jpeg"},{"id":102862417,"identity":"c357e529-4c71-4ced-9942-fc3f5bb68346","added_by":"auto","created_at":"2026-02-17 16:17:20","extension":"jpeg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":67525,"visible":true,"origin":"","legend":"\u003cp\u003ePerformance of the fusion prediction model. ROC curve and AUC (a) in the external validationset, and calibration curve (b) in the training set, test set, and external validationset. AUC, area under the curve; ROC, receiver operating characteristic.\u003c/p\u003e","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8521529/v1/b40e559c11cfb07d5cc12707.jpeg"},{"id":102965334,"identity":"ba024d7e-f42d-49d1-bb50-c0d0a965a313","added_by":"auto","created_at":"2026-02-19 04:31:24","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1643723,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8521529/v1/357fecda-5d54-4067-9d56-3970d504d3e3.pdf"},{"id":102862411,"identity":"ca427c71-4552-4edb-b600-d6374bb7069f","added_by":"auto","created_at":"2026-02-17 16:17:20","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":31769,"visible":true,"origin":"","legend":"","description":"","filename":"Tables.docx","url":"https://assets-eu.researchsquare.com/files/rs-8521529/v1/7e3b4a49e9a255e5401cf3ac.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Radiomics using contrast-enhanced T1-weighted imaging and clinical features for predicting response to EGFR-TKI in EGFR-mutated non-small cell lung cancer patients with brain metastases","fulltext":[{"header":"Introduction","content":"\u003cp\u003eBrain metastases (BMs) are secondary tumors originating from solid organs outside the central nervous system and represent the most common type of brain tumor in adults\u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]\u003c/sup\u003e. During the clinical course, 10\u0026ndash;40% of patients with solid tumors develop BMs, and the median survival for patients with synchronous BMs from almost all primary sites does not exceed 12 months\u003csup\u003e[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]\u003c/sup\u003e. Non-small cell lung cancer (NSCLC) is the primary source of BMs, accounting for ~\u0026thinsp;50% of cases\u003csup\u003e[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]\u003c/sup\u003e. BMs are typically associated with poor prognosis and high mortality in lung cancer. Therefore, precise diagnosis and treatment are important for clinical management of BMs. Epidermal growth factor receptor (EGFR) mutations are widely recognized as the most common genetic alteration in NSCLC. Compared to NSCLC without BMs, the frequency of EGFR mutations is higher in NSCLC with BMs, and patients with NSCLC harboring EGFR mutations are more prone to developing BMs than those with wild-type EGFR\u003csup\u003e[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]\u003c/sup\u003e. Given the high sensitivity and specificity of EGFR tyrosine kinase inhibitors (TKIs) in treating EGFR-mutated lung cancer, and their increasingly vital role in BM management, these agents are associated with favorable prognosis\u003csup\u003e[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]\u003c/sup\u003e. EGFR-TKIs therapy has significantly improved survival and life expectancy of patients with EGFR-mutated NSCLC\u003csup\u003e[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/sup\u003e. However, due to tumor heterogeneity and patient-specific factors, there is considerable variability in the sensitivity of NSCLC BMs to EGFR-TKIs. Therefore, personalized and precise diagnosis and treatment of BMs are crucial for improving prognosis and quality of life.\u003c/p\u003e \u003cp\u003eMagnetic resonance imaging (MRI) offers the advantage of noninvasively and repeatedly visualizing comprehensive, three-dimensional monitoring for intra-, inter-, and peritumoral information. This effectively compensates for deficiencies such as difficulties in obtaining adequate tissue samples, high costs, and long turnaround times for results. Compared to tissue-based molecular analyses, imaging remains the primary means for noninvasive monitoring of therapeutic responses and surveillance of disease progression\u003csup\u003e[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/sup\u003e. Contrast-enhanced T1-weighted imaging (CE-T1WI) is the preferred conventional method for diagnosing BMs because it enhances the contrast between BMs and surrounding tissues. It can reflect signal variations caused by differences in the marginal regions of BMs and the degree of tumor enhancement, making it more advantageous for capturing lesion heterogeneity\u003csup\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/sup\u003e. CE-T1WI is the most frequently applied sequencing method for diagnosis and treatment of BMs\u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e. CE-T1WI has significant potential for visualizing and quantifying pathophysiological changes in BMs before and after treatment, and can be used to assess the therapeutic efficacy of EGFR-TKIs. MRI is an indispensable component of BMs diagnosis and post-treatment monitoring; however, current applications remain at a macroscopic level, primarily limited to subjective observations. In contrast to the limited information provided by traditional MRI diagnostic methods, MRI radiomics can automatically quantify tumor heterogeneity, thereby improving the accuracy of diagnostic and prognostic models. The visualization of tumor heterogeneity also aids physicians in formulating more precise treatment plans. Therefore, by studying characteristic changes in tumors on MR images, radiomics methods can predict tumor sensitivity to molecular therapies. Such techniques contribute to making more optimized clinical decisions without incurring additional costs \u003csup\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThere is a high incidence of EGFR-mutated NSCLC with BMs, and BMs are an indicator of therapeutic efficacy; therefore, MRI of BMs has the potential to identify novel biomarkers for predicting response to EGFR-TKIs. Previous studies, in predicting the therapeutic efficacy of EGFR-TKIs, have primarily relied on constructing models using radiomic features alone, overlooking the potential complementary value of clinical variables. Integration of clinical characteristics with radiomic analysis can enhance the robustness of models and facilitate exploration of their clinical translational value\u003csup\u003e[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e. The application of MRI radiomics, a noninvasive, quantifiable, and reproducible intelligent method, to evaluate the therapeutic efficacy of malignant tumors in clinical settings has become an inevitable trend. Predictive models constructed based on radiomic features and clinical characteristics demonstrate significant promise in forecasting therapeutic outcomes. The incorporation of radiomic features into clinical models can enhance prediction, thereby improving clinical decision-making without incurring additional costs. This approach is conducive to optimizing treatment plans tailored to the individual needs and risk profiles of patients\u003csup\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/sup\u003e. Therefore, the present study extracted radiomic features from MRI scans of patients with EGFR-mutated NSCLC and BMs, integrated multiple data sources, and developed and externally validated a radiomic framework in conjunction with clinical characteristics to predict the therapeutic response to EGFR-TKIs. This study also showed the potential clinical value of the combined model in guiding personalized treatment strategies.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eInclusion and exclusion criteria of patients\u003c/h2\u003e \u003cp\u003eThis was a retrospective analysis of brain MRI data from 230 patients treated at Center 1 from April 2016 to January 2024, who served as the primary cohort. The training and internal test sets of the primary cohort were allocated in an 8:2 ratio for model development and validation. Additionally, 80 NSCLC patients with BMs treated at Center 2 from June 2020 to January 2024 and Center 3 from November 2021 to March 2024 were combined into a new dataset as the external validation set to balance the distribution of demographic and clinicopathological factors. The Institutional Review Board waived the requirement for written consent because of the retrospective nature of the study and the absence of identifiable patient information. Patients with EGFR-mutated NSCLC received EGFR-TKI therapy in accordance with current clinical guidelines. For each patient, the response to EGFR-TKI treatment was assessed using the response evaluation criteria in solid tumors (RECIST) version 1.1\u003csup\u003e[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/sup\u003e. Clinical characteristics, including age, gender, EGFR mutation type, EGFR-TKI type, extracranial metastasis, number of BMs, size of BMs, and peritumoral edema, were obtained from clinical and pathological records, as well as baseline cranial MRI images.\u003c/p\u003e \u003cp\u003eThe therapeutic efficacy of EGFR-TKIs was determined by evaluating changes in the patients' condition within 1 year after initiation of EGFR-TKI treatment. Intracranial efficacy was assessed using the Response Assessment in Neuro-Oncology (RANO)\u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/sup\u003e criteria, categorized as intracranial complete response (iCR), intracranial partial response (iPR), intracranial stable disease (iSD), and intracranial progressive disease (iPD). The intracranial objective response rate represented the proportion of patients achieving iCR and iPR, while the intracranial disease control rate represented the proportion of patients achieving iCR, iPR, and iSD. Patients with iCR and iPR were classified into the effective treatment group, whereas those with iSD and iPD were classified into the ineffective treatment group.\u003c/p\u003e \u003cp\u003eThe inclusion criteria were: (1) definitive diagnosis of NSCLC was established through lung biopsy or open surgery; (2) enhanced brain MRI was performed at our hospital within 2 weeks prior to treatment; (3) brain MRI revealed at least one BM; (4) EGFR mutation was confirmed by genetic testing, and the patient received EGFR-TKI treatment; and (5) complete medical records were available, including treatment information and monitoring examination results. The exclusion criteria were: (1) patients did not receive EGFR-TKIs after onset of BMs; (2) poor image quality affected quantitative analysis; (3) incomplete clinical information; (4) other concurrent cancers; and (5) patients who simultaneously received local treatment, such as whole-brain radiotherapy and surgical resection. For more detailed information on the inclusion and exclusion criteria for NSCLC BMs in this study, please refer to Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eMRI\u003c/h3\u003e\n\u003cp\u003eAll three research centers utilized 1.5 T or 3 T MR scanners to perform cranial MRI. After administering a weight-adjusted dose of 0.1 mmol/kg gadobenate dimeglumine contrast agent via an elbow vein using a high-pressure injector at an injection rate of 2 ml/s, axial CE-T1WI sequences were acquired. The imaging protocols varied among the different scanners at each research center. Detailed scanning parameters are provided in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \n\u003ch3\u003eRadiomics analysis\u003c/h3\u003e\n\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eMRI segmentation\u003c/h2\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e illustrates the specific workflow of radiomic analysis. All imaging data were retrieved from the Picture Archiving and Communication System and stored in DICOM format. Two radiologists used ITK-SNAP version 3.8.0 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e\u003ca href=\"http://www.itksnap.org\" target=\"_blank\"\u003ewww.itksnap.org\u003c/a\u003e\u003c/span\u003e\u003cspan address=\"http://www.itksnap.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) to delineate regions of interest (ROIs) on axial CE-T1WI scans. A computer-generated random number table was used to select images from 30 patients for intra- and interobserver consistency analysis. The imaging data of these 30 patients were used for tumor ROI delineation by two radiologists who were blinded to the patients' pathological results. One of the radiologists repeated the tumor ROI delineation 1 month after initial delineation. Interobserver consistency analysis was based on the data from the initial delineation by both radiologists, while intraclass correlation coefficient (ICC) was conducted using the data from the two delineations performed by the same radiologist (with 10 years of experience).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eRadiomic feature extraction and selection\u003c/h3\u003e\n\u003cp\u003eThe original images underwent N4_Normalization processing, and radiomic features were extracted from the ROIs of each patient using Pyradiomics version 3.1.0. The radiomic features extracted from each ROI included: morphology (e.g., volume, surface area, and diameter); first-order statistical features (e.g., mean, standard deviation, and entropy); and second-order features. The second-order features included: gray-level co-occurrence matrix (describing gray-level spatial correlation); gray-level run-length matrix (analyzing texture roughness); gray-level size zone matrix (quantifying distribution of region sizes); gray-level dependence matrix (describing local gray-level variations); and neighborhood gray-tone difference matrix (analyzing local heterogeneity).\u003c/p\u003e \u003cp\u003eFirstly, to eliminate interference caused by outliers during Z-score normalization, all radiomic features were normalized to ensure that feature selection was not affected by large differences in magnitude. Secondly, an ICC consistency test was conducted, and features with an ICC threshold\u0026thinsp;\u0026gt;\u0026thinsp;0.8 were retained. From each patient, 1505 features were extracted, and 430 features with an ICC intra- and interclass correlation coefficient\u0026thinsp;\u0026lt;\u0026thinsp;0.8 were deleted, leaving 1075 features for further selection. Subsequently, Spearman correlation analysis was used for initial dimensionality reduction of the radiomic features, eliminating those with an absolute correlation value\u0026thinsp;\u0026gt;\u0026thinsp;0.9. A recursive feature elimination algorithm was used to progressively eliminate unimportant features. Finally, the least absolute shrinkage and selection operator (LASSO) combined with 10-fold cross-validation was utilized for final feature dimensionality reduction, resulting in selection of the optimal features.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eConstruction and validation of predictive models\u003c/h2\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003eModel training\u003c/h2\u003e \u003cp\u003eDifferent combinations of feature types and classifiers can be customized to suit specific classification tasks. The motivation for using multiple classifiers was to enhance the robustness of the results, as previous studies have indicated that selecting an appropriate algorithm plays a crucial role in classification outcomes \u003csup\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/sup\u003e. This study recognized that no single model in machine learning is universally applicable, and explored various classifiers to develop a high-performance and robust model for predicting the therapeutic efficacy of EGFR-TKIs in NSCLC BMs. To adhere to best ML practices, a combined classifier comprising seven commonly used ML algorithms was adopted, comprising: logistic regression as a linear model; random forest (RF) as an ensemble method; k-nearest neighbors for instance-based modeling; linear support vector machine; decision tree as a tree-structured supervised learning model; Gaussian na\u0026iuml;ve Bayes as a probability classification model; and extreme gradient boosting as a gradient boosting framework. Subsequently, the seven ML models were evaluated, and their efficacies were comprehensively assessed through area under the curve (AUC) analysis, decision curve analysis to determine standardized net benefit, and calibration curves for each trained model to identify the optimal model. After performing feature extraction and selection from within the BMs, as well as from automatically expanded 3-, 5-, and 8-mm peritumoral regions, radiomic models were constructed based on optimal ML algorithms according to model training, and their performances were compared.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e\n\u003ch3\u003ePredictive model construction\u003c/h3\u003e\n\u003cp\u003eClinical and pathological data were retrospectively collected from electronic medical records and digital pathology systems across various centers, including: age; gender; smoking status; Karnofsky Performance Status score; primary tumor site; pathological classification; tumor markers [carcinoembryonic antigen, neuron-specific enolase, cytokeratin 19 fragment (CY211)]; extracranial metastasis; TNM staging (describing the size and invasiveness of the primary tumor, lymph node metastasis, and systemic metastasis); EGFR mutation subtype; and EGFR-TKI type. Additionally, MRI morphological features commonly used in clinical assessments of central nervous system tumors were analyzed, including; number and grouping of BMs; location, maximum diameter, and grouping of the largest lesion; enhancement pattern; and peritumoral edema. Logistic regression analysis was conducted to select clinically significant features (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) for model construction. Finally, radiomic, clinical, and combined models were constructed based on the optimal ML algorithm. The optimal model was determined by comparing the AUCs using the Delong test, and the model's generalization capability was further evaluated in an external validation set.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eWe used SPSS 22.0 software (IBM Corporation, Chicago, IL, USA) for statistical analysis. In descriptive statistics, continuous variables were expressed as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation or median (interquartile range), while categorical variables were presented as frequencies and percentages. For analysis of differences among the training, test, and external validation sets, continuous variables were compared using Student's \u003cem\u003et\u003c/em\u003e test or Mann\u0026ndash;Whitney \u003cem\u003eU\u003c/em\u003e test depending on the data type, while categorical variables were compared using χ\u003csup\u003e2\u003c/sup\u003e test or Fisher's exact test. The discriminatory ability of predictive models was evaluated and compared using the 95% confidence interval (CI) of the area under the receiver operating characteristic curve (ROC). Model performance was further assessed by calculating AUC, accuracy (ACC), sensitivity (SEN), specificity (SPE), and F1 score. The Delong test was used to compare the AUC values of different models and evaluate their robustness. The performance of radiomic, clinical, and combined models was evaluated separately in the test and external validation sets. Calibration curves were used to assess the deviation in the distribution of predicted values, reflecting the consistency between model predictions and actual probabilities. Based on this, decision curve analysis was conducted to calculate the net benefit at different threshold probabilities, thereby evaluating the model's value in differential diagnosis. All statistical analyses were performed with a significance level set at \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003ePatient basic information and treatment efficacy\u003c/h2\u003e \u003cp\u003eA total of 338 patients from three centers were included. Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e summarizes the clinical characteristics, MRI features, and therapeutic efficacy of EGFR-TKIs. The clinical features included KPS score, extracranial metastases, M staging of the TNM classification, and clinical staging. The MRI morphological features included the number of lesions, necrosis, and EGFR-TKI type. Significant differences were observed across the training, test, and external validation sets (all \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The assessment showed that 39 patients achieved iCR, 104 iPR, 60 iSD, and 131 iPD. The intracranial objective response rate was 42.3% and intracranial disease control rate was 60.1%.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003ePredictive models evaluation\u003c/h2\u003e \u003cp\u003eA higher AUC indicated greater classification accuracy, while a smaller AUC gap between the training and test sets suggested stronger model generalizability and lower risk of overfitting. The model training results (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) indicated that the radiomic model constructed based on the RF algorithm emerged as the top performer. Comparison of RF models derived from intratumoral features of BMs and extended 3-, 5-, and 8-mm peritumoral regions revealed that intratumoral features demonstrated better AUC performance in the test set (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Therefore, intratumoral features were selected for constructing the radiomic model. LASSO was applied to determine the optimal parameter value (λ\u0026thinsp;=\u0026thinsp;0.0285) based on intratumoral features of BMs, ultimately selecting eight significant features that made important contributions to the classification of the predictive model (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThree meaningful clinical features (age, CY211, and EGFR-TKI grouping) were identified through univariate logistic regression (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The ROC curves of the clinical, radiomic, and combined model constructed based on RF in the training and test sets are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea, b. Figure\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec, d further clarifies the performance differences among the three models in predicting the therapeutic efficacy of EGFR-TKIs. The corresponding AUC, ACC, SEN, SPE, and F1 scores of the three predictive models in the training and test sets were calculated to assess their quantitative classification performance (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn the predictive model for EGFR-TKI therapeutic efficacy within 1 year, the combined model based on CE-T1WI demonstrated the best performance compared to the radiomic and clinical models. The AUC performance of the combined model in the test set outperformed that of the radiomic and clinical models (0.756 vs 0.644 vs 0.668). The Delong test confirmed that the combined model achieved impressive predictive performance (Z-score: \u0026minus;2.166 vs \u0026minus;4.640, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.030 vs\u0026thinsp;\u0026lt;\u0026thinsp;0.001). It also obtained optimal metrics in the test set, with an AUC of 0.756 (95% CI: 0.627\u0026ndash;0.886), ACC 0.702, SEN 0.844, SPE 0.520, and F1 score 0.683. In the external validation set, the ROC curve of the combined model is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea, with an AUC of 0.743 (95% CI: 0.604\u0026ndash;0.881), ACC 0.679, SEN 0.480, SPE 0.857, and F1 score 0.662. These results confirmed the reliability and generalizability of the combined model in predicting the therapeutic efficacy of EGFR-TKIs. The calibration curves of the combined model in the training, test, and external validation sets are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb, verifying the reliability and generalizability of the fused model.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this comprehensive study, we utilized multicohort data from patients with EGFR-mutant NSCLC BMs from three institutions to develop and externally validate an MRI radiomic framework for predicting the therapeutic response to EGFR-TKIs. The radiomic phenotypic features proposed demonstrated excellent predictive capabilities. We discovered a synergistic effect between radiomic features and clinical characteristics. The integration of clinical features and MRI radiomics ultimately facilitated the development of a combined model that used multiple data sources to optimize predictive performance. The validation framework underscored the pivotal role of radiomic features extracted from MRI. These features provide complementary insights to established MRI biomarkers, bringing us closer to achieving personalized treatment strategies for NSCLC BMs. Compared to traditional visual assessment methods based on enhanced MRI of BMs, the radiomic model exhibited significant advantages in terms of accuracy and sensitivity in predicting the therapeutic response of BMs, thereby aiding clinicians in formulating more personalized treatment plans and promoting the application and development of radiomic approaches. MRI radiomics can help predict the efficacy of targeted therapy, serving as a valuable tool for clinicians and patients. It provides clinicians with valuable, objective, and consistent information to determine optimal treatment strategies and patient follow-up plans.\u003c/p\u003e \u003cp\u003eThe motivation for using multiple classifiers was to enhance the robustness of the results, as previous studies have indicated that selecting an appropriate algorithm plays a crucial role in classification outcomes\u003csup\u003e[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]\u003c/sup\u003e. Zhao et al.\u003csup\u003e[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]\u003c/sup\u003e constructed support vector machine, RF, and k-nearest neighbors models using LASSO based on CE-T1WI and T2-fluid-attenuated inversion recovery sequences to differentiate between primary central nervous system lymphoma and BMs. The RF model exhibited the best performance (AUC\u0026thinsp;=\u0026thinsp;0.73). Kanakarajan et al.\u003csup\u003e[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/sup\u003e developed a radiomic model based on CE-T1WI to predict the efficacy of radiotherapy for BMs. By incorporating clinical features, the combined model using an RF classifier achieved the highest AUC of 0.89, with an accuracy rate of 87%. Similar studies\u003csup\u003e[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]\u003c/sup\u003e have shown that radiomic models constructed using an RF classifier also achieved favorable performance in predicting the efficacy of radiotherapy for BMs, with an AUC of up to 0.83. In summary, the RF classifier has demonstrated excellent performance in predicting the therapeutic efficacy for BMs, which agrees with the findings of our study. Our model training also indicated the significant efficacy of the RF classifier. Previous studies have shown that extracting radiomic features from the tumor region and peritumoral edema of BMs is reasonable and reliable. This approach can encompass most of the details in the heterogeneous regions of BMs and provide better prediction of therapeutic efficacy in patients with BMs, compared with features from a single region\u003csup\u003e[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]\u003c/sup\u003e. Fan et al.\u003csup\u003e[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/sup\u003e also indicated that the tumor region and peritumoral edema in NSCLC BMs were complementary, allowing for extraction of rich radiomic phenotypic features to enhance model robustness and achieve more reliable prediction of the therapeutic efficacy of EGFR-TKIs. Based on previous studies, we constructed models using intratumoral and peritumoral features based on the optimal ML algorithm. However, the results indicated that the model constructed using features extracted from within the tumor exhibited better performance. This may have been because NSCLC patients with EGFR mutations often develop multiple small BMs, with mild or no peritumoral edema\u003csup\u003e[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]\u003c/sup\u003e. Therefore, it is impossible to extract critical information from the peritumoral edema area pertaining to the therapeutic efficacy against BMs.\u003c/p\u003e \u003cp\u003eAmong patients with NSCLC BMs, those with EGFR mutations have longer survival and better control of intracranial disease \u003csup\u003e[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]\u003c/sup\u003e. Fan et al. \u003csup\u003e[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/sup\u003e incorporated pretreatment CE-T1WI and T2WI from NSCLC BMs patients and constructed models based on tumor area analysis, peritumoral edema, margin/background parenchyma, and multiregional fusion to predict the therapeutic response to EGFR-TKIs. The fusion model achieved the best performance in the external validation set, proving that subregional radiomics, as a novel noninvasive approach, can help guide personalized treatment strategies with EGFR-TKIs for NSCLC BMs. However, our study only utilized radiomic features to construct models for predicting the therapeutic efficacy of EGFR-TKIs using MRI radiomics, neglecting the potential complementary value of clinical variables. A predictive model combining radiomics with clinical predictive features will enable earlier detection of disease progression and benefit patients by guiding clinicians to conduct more intensive MRI follow-up of patients at high risk of intracranial progression. Similarly, Qu et al. \u003csup\u003e[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e included 212 lesions from 70 patients with NSCLC BMs who received first-line EGFR-TKI treatment. Radiomic features were extracted from the brain tumor regions on pretreatment CE-T1WI, and a radiomic score was calculated based on the selected features. The combined predictive model incorporating EGFR-19del mutation, third-generation EGFR-TKI treatment, and the average rad-score outperformed models constructed using each of these three predictors alone. The combined model demonstrated good predictive value for intracranial progression within 1 year after EGFR-TKI treatment. The results confirmed that the nomogram combining clinical and radiomic features outperformed nomograms based solely on either features alone. Unlike the aforementioned validation studies, in our study, age, EGFR-TKIs, and CY211 were identified as significant clinical features. CY211 was the most sensitive tumor marker in NSCLC, which is consistent with the literature\u003csup\u003e[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]\u003c/sup\u003e. Qi et al.\u003csup\u003e[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]\u003c/sup\u003e included 117 patients with NSCLC BMs for individualized prediction of the therapeutic efficacy of first-generation EGFR-TKIs. The LASSO algorithm was used to screen radiomic features extracted from multiparametric MRI. In the short-term efficacy predictive model, clinical, radiomic, and combined nomograms were constructed separately. The results demonstrated good consistency between predicted risk and actual outcomes, confirming that the combined nomogram outperformed nomograms based solely on clinical or radiomic features, with a C-index of 0.843. Qi et al.\u003csup\u003e[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]\u003c/sup\u003e subsequently developed and validated a nomogram based on clinical characteristics and MRI radiomics for predicting the short-term efficacy of third-generation EGFR-TKIs in patients with EGFR-mutant lung adenocarcinoma BMs, with a C-index of 0.803. The studies confirmed the favorable efficacy of first- and third-generation EGFR-TKIs in treating EGFR-mutant NSCLC BMs. The predictive performance of nomograms established by combining radiomics and clinical features outperformed nomograms based solely on radiomics or clinical features, and served as a noninvasive predictive tool that aided timely personalized adjustment of treatment plans. Association of radiomic characteristics with clinical features is an emerging field that enhances the correlation between radiomic features and molecular characteristics, potentially reflecting gene expression differences within and between tumors. The aforementioned studies indicate that MRI radiomics combined with clinical features can provide noninvasive and intelligent prediction of the therapeutic efficacy of EGFR-TKIs in EGFR-mutant NSCLC BMs. However, because of the design of case follow-up, the research samples were from a single center and limited in number, which restricts the generalizability and clinical effectiveness of the results. Despite the use of cross-validation and the good performance of the models in the training and test sets, the generalizability of the models to unseen data requires further investigation. In this study, radiomic analysis was performed using pretreatment CE-T1WI from NSCLC BM patients. To maximize the sample size, patients with different EGFR mutation types receiving different EGFR-TKIs were included. Based on a multi-institutional and large sample size, combined with clinical data, the effectiveness of EGFR-TKIs in local tumor control was predicted. The results indicated that radiomic features combined with clinical features served as complementary biomarkers for predicting the therapeutic efficacy in NSCLC BM patients, achieving an AUC of 0.743 in the external validation set, demonstrating good generalizability. Our study demonstrated that combining MRI radiomics with clinical features improved the accuracy of EGFR-TKI therapeutic monitoring based on imaging, and supported preoperative diagnosis, treatment planning, and outcome determination of BMs.\u003c/p\u003e \u003cp\u003eOur study had several limitations. Firstly, although this was a multicenter study with an external validation set, the model was developed based on retrospective data, which had the inherent biases of retrospective analyses and challenges of adequately controlling for confounding variables. Future confirmation through prospective studies is needed to refine and evaluate the clinical utility of the model. Secondly, the EGFR mutation status data were derived from primary lung cancer tissues, which may not directly reflect secondary mutations occurring in metastatic lesions. Thirdly, the MRI equipment and scanning parameters showed heterogeneity, and the treatment strategies received by patients were diverse and at different stages of the disease. These factors may all affect the generalizability of our findings. Additionally, this study was based on a single CE-T1WI sequence, as some patients did not undergo scanning with other conventional sequences. However, diffusion-weighted imaging and diffusion tensor imaging are also recommended as standards for evaluating BMs. Incorporating more sequences into future predictive models may enhance their predictive capabilities. Finally, future research should integrate multimodal imaging and data for in-depth exploration and methodological optimization to improve model calibration. Additionally, a thorough analysis of the associations between radiomic features and tumor biological characteristics will contribute to a better understanding of tumor pathogenesis and treatment response mechanisms, thereby facilitating clinical translation.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study developed and externally validated a model that combined clinical and radiomic features to support treatment decision-making in patients with NSCLC BMs. The results provide robust evidence for predicting the therapeutic efficacy of EGFR-TKIs and add value to targeted treatment of patients. However, it is important to note that these findings require further confirmation through future prospective studies to enhance their clinical generalizability.\u003c/p\u003e"},{"header":"Abbreviations","content":" \u003cp\u003eACC, Accuracy\u003c/p\u003e \u003cp\u003eAD, adenocarcinoma\u003c/p\u003e \u003cp\u003eAUC, area under the curve\u003c/p\u003e \u003cp\u003eBMs, Brain metastasis\u003c/p\u003e \u003cp\u003eCEA, Carcinoembryonic antigen\u003c/p\u003e \u003cp\u003eCE-T1WI, Contrast-enhanced T1 weighted imaging\u003c/p\u003e \u003cp\u003eCY211, Cytokeratin 19 fragment\u003c/p\u003e \u003cp\u003eDT, Decision tree\u003c/p\u003e \u003cp\u003eEGFR, epidermal growth factor receptor\u003c/p\u003e \u003cp\u003eEGFR-TKIs, EGFR tyrosine kinase inhibitors\u003c/p\u003e \u003cp\u003eGLCM, Gray-level co-occurrence matrix\u003c/p\u003e \u003cp\u003eGLDM, Gray-level dependence matrix\u003c/p\u003e \u003cp\u003eGLRLM, Gray-level run-length matrix\u003c/p\u003e \u003cp\u003eGLSZM, Gray-level size zone matrix\u003c/p\u003e \u003cp\u003eGNB, gaussian naive bayes\u003c/p\u003e \u003cp\u003eICC, Intraclass correlation coefficient\u003c/p\u003e \u003cp\u003eiCR, Intracranial complete response\u003c/p\u003e \u003cp\u003eiDCR, Intracranial disease control rate\u003c/p\u003e \u003cp\u003eiORR,Intracranial objective response rate\u003c/p\u003e \u003cp\u003eiPD, Intracranial progressive disease\u003c/p\u003e \u003cp\u003eiPR, Intracranial partial response\u003c/p\u003e \u003cp\u003eiSD, Intracranial stable disease\u003c/p\u003e \u003cp\u003eKNN, K-nearest neighbors\u003c/p\u003e \u003cp\u003eLASSO, Least absolute shrinkage and selection operator\u003c/p\u003e \u003cp\u003eLR, Logistic regression\u003c/p\u003e \u003cp\u003eMRI, Magnetic resonance imaging\u003c/p\u003e \u003cp\u003eNGTDM, Neighboring gray-tone difference matrix\u003c/p\u003e \u003cp\u003eNSE, Neuron-specific enolase\u003c/p\u003e \u003cp\u003eNSCLC, Non-small-cell lung cancer\u003c/p\u003e \u003cp\u003eOR, Odds ratio\u003c/p\u003e \u003cp\u003eRECIST, Response evaluation criteria in solid tumors\u003c/p\u003e \u003cp\u003eROI, Region of interest\u003c/p\u003e \u003cp\u003eSCC, Squamous cell cancer\u003c/p\u003e \u003cp\u003eSPE, Specificity\u003c/p\u003e \u003cp\u003eSEN, Sensitivity\u003c/p\u003e \u003cp\u003eSVM, Support vector machine\u003c/p\u003e \u003cp\u003eXGB, Extreme gradient boosting\u003c/p\u003e "},{"header":"Declarations","content":"\u003cp\u003e \u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e \u003cp\u003e This retrospective study was approved by the institutional review board (\u0026ldquo;Ethics committee of the Affiliated Hospital of Hebei University\u0026rdquo;, Dongfeng Road, No. 212, Lianchi District, Baoding 071000, China, Number HDFYLL-KY-2024-037; date of approval 02/28/2024)., was conducted in accordance with the Declaration of Helsinki. Informed consent was waived for all patients due to the retrospective nature of this study.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent for publication\u003c/strong\u003e \u003cp\u003eNot applicable.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eCompeting interests\u003c/strong\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThis study was supported by the Hebei University Graduate Student Innovation Funding Project (CXZZBS2025028).\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eConception and design of the research: XPY, JNW. Acquisition of data: LYS. Analysis and interpretation of the data: LYS. Statistical analysis: LYS. Writing of the manuscript: LYS. Critical revision of the manuscript for intellectual content: LYS, LHX, HM, YZ, QW, CL and TSZ. All authors read and approved the final draft.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eNone.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eACHROL A S, RENNERT R C, ANDERS C, et al. Brain metastases [J]. Nat Reviews Disease Primers. 2019;5(1):5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSUNG K S. Clinical Practice Guidelines for Brain Metastasis From Solid Tumors [J]. Brain Tumor Res Treat. 2024;12(1):14\u0026ndash;22.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLAMBA N, WEN P Y, AIZER AA. Epidemiology of brain metastases and leptomeningeal disease [J]. Neurooncology. 2021;23(9):1447\u0026ndash;56.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGE M, ZHUANG Y, ZHOU X, et al. High probability and frequency of EGFR mutations in non-small cell lung cancer with brain metastases [J]. J Neurooncol. 2017;135(2):413\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKHANDEKAR MJ, PIOTROWSKA Z, WILLERS H, et al. Role of Epidermal Growth Factor Receptor (EGFR) Inhibitors and Radiation in the Management of Brain Metastases from EGFR Mutant Lung Cancers [J]. Oncologist. 2018;23(9):1054\u0026ndash;62.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSIEGEL R L, KRATZER T B, GIAQUINTO A N, et al. Cancer statistics, 2025 [J]. CA Cancer J Clin. 2025;75(1):10\u0026ndash;45.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBOIRE A, BRASTIANOS P K, GARZIA L, et al. Brain metastasis [J]. Nat Rev Cancer. 2020;20(1):4\u0026ndash;11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKHALIGHI S, REDDY K, MIDYA A, et al. Artificial intelligence in neuro-oncology: advances and challenges in brain tumor diagnosis, prognosis, and precision treatment [J]. NPJ Precis Oncol. 2024;8(1):80.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDREGELY I, KELLY-MORLAND C PREZZID, et al. Imaging biomarkers in oncology: Basics and application to MRI [J]. J Magn Reson Imaging. 2018;48(1):13\u0026ndash;26.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLE RHUN E, GUCKENBERGER M, SMITS M, et al. EANO-ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up of patients with brain metastasis from solid tumours [J]. Ann Oncol. 2021;32(11):1332\u0026ndash;47.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLAMBIN, P, LEIJENAAR R T H, DEIST T M, et al. Radiomics: the bridge between medical imaging and personalized medicine [J]. Nat Rev Clin Oncol. 2017;14(12):749\u0026ndash;62.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZHANG X, ZHANG G, QIU X, et al. Non-invasive decision support for clinical treatment of non-small cell lung cancer using a multiscale radiomics approach [J]. Radiother Oncol. 2024;191:110082.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWANG T-W, CHAO H-S, CHIU H-Y, et al. Radiomics of metastatic brain tumor as a predictive image biomarker of progression-free survival in patients with non-small-cell lung cancer with brain metastasis receiving tyrosine kinase inhibitors [J]. Translational Oncol. 2024;39:101826.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMOURAVIEV A, DETSKY J, SAHGAL A, et al. Use of radiomics for the prediction of local control of brain metastases after stereotactic radiosurgery [J]. Neuro Oncol. 2020;22(6):797\u0026ndash;805.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLIN N U, LEE E Q AOYAMAH, et al. Response assessment criteria for brain metastases: proposal from the RANO group [J]. Lancet Oncol. 2015;16(6):e270\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOCA\u0026ntilde;A-TIENDA B, P\u0026eacute;REZ-BETETA J, ROMERO-ROSALES J A, et al. Volumetric analysis: Rethinking brain metastases response assessment [J]. Neurooncol Adv. 2024;6(1):vdad161.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXIA X, QIU J, TAN Q, et al. Radiomics-Based Differentiation of Primary Central Nervous System Lymphoma and Solitary Brain Metastasis Using Contrast-Enhanced T1-Weighted Imaging: A Retrospective Machine Learning Study [J]. Acad Radiol. 2025;32(9):5401\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXIA X, WU W. Interpretable Machine Learning Models for Differentiating Glioblastoma From Solitary Brain Metastasis Using Radiomics [J]. Acad Radiol. 2025;32(9):5388\u0026ndash;400.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eARTZI M, BRESSLER I, BEN BASHAT D. Differentiation between glioblastoma, brain metastasis and subtypes using radiomics analysis [J]. J Magn Reson Imaging. 2019;50(2):519\u0026ndash;28.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eORTIZ-RAMON R, LARROZA A et al. ARANA E,. A radiomics evaluation of 2D and 3D MRI texture features to classify brain metastases from lung cancer and melanoma [J]. Annu Int Conf IEEE Eng Med Biol Soc, 2017, 2017: 493-6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZHAO LM, XIE F F HUR, et al. Radiomic-Based MRI for Classification of Solitary Brain Metastases Subtypes From Primary Lymphoma of the Central Nervous System [J]. J Magn Reson Imaging. 2023;57(1):227\u0026ndash;35.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKANAKARAJAN H, DE BAENE W, SITSKOORN M et al. Predicting local control of brain metastases after stereotactic radiosurgery with clinical, radiomics and deep learning features [J]. medRxiv, 2024: 2024.05. 13.24307241.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRESSA G, LEVI R, SAVINI G, et al. AI differentiates radionecrosis from true progression in brain metastasis upon stereotactic radiosurgery: analysis of 124 histologically assessed lesions [J]. Neuro Oncol; 2025.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXU J, WANG P, LI Y, et al. Development and validation of an MRI-Based nomogram to predict the effectiveness of immunotherapy for brain metastasis in patients with non-small cell lung cancer [J]. Front Immunol. 2024;15:1373330.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFAN Y, WANG X, DONG Y, et al. Multiregional radiomics of brain metastasis can predict response to EGFR-TKI in metastatic NSCLC [J]. Eur Radiol. 2023;33(11):7902\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCHAMBERLAIN M C, BAIK C S, GADI V K, et al. Systemic therapy of brain metastases: non\u0026ndash;small cell lung cancer, breast cancer, and melanoma [J]. Neurooncology. 2016;19(1):i1\u0026ndash;24.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEICHLER A F, KAHLE K T, WANG D L, et al. EGFR mutation status and survival after diagnosis of brain metastasis in nonsmall cell lung cancer [J]. Neuro Oncol. 2010;12(11):1193\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQU J, ZHANG T, ZHANG X, et al. MRI radiomics for predicting intracranial progression in non-small-cell lung cancer patients with brain metastases treated with epidermal growth factor receptor tyrosine kinase inhibitors [J]. Clin Radiol. 2024;79(4):e582\u0026ndash;91.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLIU L, TENG J, ZHANG L et al. The Combination of the Tumor Markers Suggests the Histological Diagnosis of Lung Cancer [J]. Biomed Res Int, 2017, 2017: 2013989.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQI H, HOU Y, ZHENG Z, et al. MRI radiomics predicts the efficacy of EGFR-TKI in EGFR-mutant non-small-cell lung cancer with brain metastasis [J]. Clin Radiol. 2024;79(7):515\u0026ndash;25.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQI H, HOU Y, ZHENG Z, et al. Clinical characteristics and MRI based radiomics nomograms can predict iPFS and short-term efficacy of third-generation EGFR-TKI in EGFR-mutated lung adenocarcinoma with brain metastases [J]. BMC Cancer. 2024;24(1):362.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTable 1 to 4 are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-medical-imaging","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bmim","sideBox":"Learn more about [BMC Medical Imaging](http://bmcmedimaging.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bmim/default.aspx","title":"BMC Medical Imaging","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Brain metastases, non-small cell lung cancer, EGFR-TKIs, Clinical characteristics, MRI, Radiomics","lastPublishedDoi":"10.21203/rs.3.rs-8521529/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8521529/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eObjectives\u003c/h2\u003e \u003cp\u003eTo construct and validate a model based on clinical characteristics and magnetic resonance imaging (MRI) radiomics to predict 1-year efficacy of epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) in patients with EGFR-mutant non-small cell lung cancer (NSCLC) brain metastases (BMs).\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eThis study retrospectively analyzed data from 338 patients with EGFR-mutant NSCLC BMs from three centers, including MRI, clinical and pathological data, and radiological features. Based on the selected significant intratumoral features, seven machine learning algorithms were applied to compare model efficacy, and the best algorithm was selected for model construction. In the model predicting 1-year therapeutic efficacy, clinical, radiomic, and combined models were constructed separately. The model performance was evaluated using receiver operating characteristic curves.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe final development cohort comprised 285 patients from Center 1, while the external validation set included 57 patients from Centers 2 and 3. In the model predicting 1-year EGFR-TKI efficacy, the random forest algorithm, which showed the best application, was used to construct the model. Compared with the radiomic and clinical models, the combined model exhibited superior area under the curve performance in the test set (0.756 vs 0.644 vs 0.668). In the external validation set, the combined model achieved an area under the curve of 0.743.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eCompared to single clinical or radiomic models, the combined model was more effective in predicting the 1-year efficacy of EGFR-TKIs in patients with NSCLC BMs with EGFR mutations.\u003c/p\u003e","manuscriptTitle":"Radiomics using contrast-enhanced T1-weighted imaging and clinical features for predicting response to EGFR-TKI in EGFR-mutated non-small cell lung cancer patients with brain metastases","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-17 16:17:15","doi":"10.21203/rs.3.rs-8521529/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-12T04:45:15+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-11T14:26:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"246749382143643036969009099315947130729","date":"2026-03-11T14:21:58+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-11T13:44:58+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"279554823129645364150232342368553418703","date":"2026-03-10T06:33:48+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-10T02:06:02+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"144916382428983413091575794949510127423","date":"2026-03-10T01:50:38+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-12T06:54:07+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-12T06:53:27+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-02-03T13:00:56+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-02T15:31:18+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Medical Imaging","date":"2026-02-02T14:39:18+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-medical-imaging","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bmim","sideBox":"Learn more about [BMC Medical Imaging](http://bmcmedimaging.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bmim/default.aspx","title":"BMC Medical Imaging","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"251607e5-03f4-45e6-aba0-16ff169cdad2","owner":[],"postedDate":"February 17th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-08T14:11:56+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-17 16:17:15","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8521529","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8521529","identity":"rs-8521529","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}