Radiogenomic modeling of EGFR mutation status in brain metastases from lung adenocarcinoma: a multicenter study with biological interpretability
7 affiliations
- 1The department of oncology, Xiangya Hospital, Central South University, Changsha, China
- 2National Clinical Research Center for Geriatric Disorders, Xiangya Hospital, Central South University, Changsha, China
- 3Xiangya Lung Cancer Center, Xiangya Hospital, Central South University, Changsha, China
- 4Bioengineering Department and Imperial-X, Imperial College London, London, UK
- 5National Heart and Lung Institute, Imperial College London, London, UK
- 6Cardiovascular Research Centre, Royal Brompton Hospital, London, UK
- 7School of Biomedical Engineering & Imaging Sciences, King’s College London, London, UK
Abstract
Accurate prediction of epidermal growth factor receptor (EGFR) mutation status in lung adenocarcinoma (LUAD) with brain metastases (BMs) is crucial for guiding targeted therapy. However, noninvasive and biologically interpretable tools remain limited. In this multicenter radiogenomic study, we analyzed a total of 1303 BMs from 421 LUAD patients across three institutions. 3435 radiomic features were extracted from T1, T2, and contrast-enhanced T1 sequences. A four-task classification framework was developed to predict EGFR mutation status (EGFR+, 19Del, L858R, or sensitizing mutation) using an adaptive LightGBM-based modeling pipeline. The models achieved excellent performance in the internal cohort (AUCs up to 0.95) and were further validated in 94 lesions with pathologically confirmed EGFR status, reaching an accuracy of 83.0%, sensitivity of 84.7%, and specificity of 80.0%. SHAP and LIME analyses revealed that shape-based radiomic features, particularly original_shape_sphericity, were the most important predictors of EGFR mutational subtypes. Then, we conducted transcriptomic analysis on 38 matched surgical specimens. Radiogenomic correlation revealed that sphericity negatively correlated with RNF125 and SLC37A2. Downstream enrichment analysis identified EGFR-associated features linked to DNA replication, sister chromatid segregation, and ERBB signaling. The study demonstrates that radiogenomic modeling, grounded in interpretable biology, holds promise as a non-invasive, clinical strategy for precision stratification of LUAD BMs.
Related
- Colonic Metastases from Lung Adenocarcinoma Detected During an Evaluation of Recurrent Diarrhea Associated with Immune-Related Colitis
- 971 The impact of liver metastases on immune-related adverse events in stage IV patients treated with immune-checkpoint inhibitors: a real-world, multicenter, retrospective study
- Lung Metastases of Pancreatic Adenocarcinoma Masquerading as Temporary Improvement of Bilateral Diffuse Lung Opacities Following Treatment for Interstitial Lung Disease.
- A highly successful case report: a patient with intestinal hepatoid adenocarcinoma and liver metastases achieves excellent efficacy following immune-targeted therapy
- Third-generation EGFR-TKI plus pemetrexed-platinum chemotherapy versus TKI monotherapy as first-line treatment for EGFR-mutant lung adenocarcinoma with brain parenchymal metastases
BibTeX
@article{deng2026radiogenomic,
title = {{Radiogenomic modeling of EGFR mutation status in brain metastases from lung adenocarcinoma: a multicenter study with biological interpretability}},
author = {Fuxing Deng and Xianjing Chu and Wen Shi and Gang Xiao and Guilong Tanzhu and Lishui Niu and Zijian Zhang and Rongrong Zhou and Guang Yang},
journal = {NPJ Digital Medicine},
year = {2026},
month = jul,
doi = {10.1038/s41746-026-02931-9},
url = {https://doi.org/10.1038/s41746-026-02931-9},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Radiogenomic modeling of EGFR mutation status in brain metastases from lung adenocarcinoma: a multicenter study with biological interpretability
AU - Fuxing Deng
AU - Xianjing Chu
AU - Wen Shi
AU - Gang Xiao
AU - Guilong Tanzhu
AU - Lishui Niu
AU - Zijian Zhang
AU - Rongrong Zhou
AU - Guang Yang
PY - 2026
DA - 2026/07/06
JO - NPJ Digital Medicine
DO - 10.1038/s41746-026-02931-9
UR - https://doi.org/10.1038/s41746-026-02931-9
AB - Accurate prediction of epidermal growth factor receptor (EGFR) mutation status in lung adenocarcinoma (LUAD) with brain metastases (BMs) is crucial for guiding targeted therapy. However, noninvasive and biologically interpretable tools remain limited. In this multicenter radiogenomic study, we analyzed a total of 1303 BMs from 421 LUAD patients across three institutions. 3435 radiomic features were extracted from T1, T2, and contrast-enhanced T1 sequences. A four-task classification framework was developed to predict EGFR mutation status (EGFR+, 19Del, L858R, or sensitizing mutation) using an adaptive LightGBM-based modeling pipeline. The models achieved excellent performance in the internal cohort (AUCs up to 0.95) and were further validated in 94 lesions with pathologically confirmed EGFR status, reaching an accuracy of 83.0%, sensitivity of 84.7%, and specificity of 80.0%. SHAP and LIME analyses revealed that shape-based radiomic features, particularly original_shape_sphericity, were the most important predictors of EGFR mutational subtypes. Then, we conducted transcriptomic analysis on 38 matched surgical specimens. Radiogenomic correlation revealed that sphericity negatively correlated with RNF125 and SLC37A2. Downstream enrichment analysis identified EGFR-associated features linked to DNA replication, sister chromatid segregation, and ERBB signaling. The study demonstrates that radiogenomic modeling, grounded in interpretable biology, holds promise as a non-invasive, clinical strategy for precision stratification of LUAD BMs.
SN - 2398-6352
ER -