Cao Pingdong, Jia Xiao, Wang Xi, Fan Liyuan, Chen Zheng, Zhao Yuanyuan, Zhu Jian, Wen Qiang
Department of Radiation Oncology, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Shandong First Medical University, Jinan, China.
School of Control Science and Engineering, Shandong University, Jinan, China.
BMC Cancer. 2025 Mar 12;25(1):443. doi: 10.1186/s12885-025-13823-8.
Early and accurate identification of epidermal growth factor receptor (EGFR) mutation status in non-small cell lung cancer (NSCLC) patients with brain metastases is critical for guiding targeted therapy. This study aimed to develop a deep learning radiomics model utilizing multi-sequence magnetic resonance imaging (MRI) to differentiate between EGFR mutant type (MT) and wild type (WT).
In this retrospective study, 288 NSCLC patients with confirmed brain metastases were enrolled, including 106 with EGFR MT and 182 with EGFR WT. All patients were randomly divided into a training dataset (75%) and a validation dataset (25%). Radiomics and deep learning features were extracted from the brain metastatic lesions using contrast-enhanced T1-weighted (T1CE) and T2-weighted (T2W) MRI images. Features extraction and selection were performed using the least absolute shrinkage and selection operator (LASSO) and ResNet34. The predictive performance of the signatures for EGFR mutation status was assessed using receiver operating characteristic (ROC) curves and area under the curve (AUC) analyses.
No significant differences were found between the training and validation datasets. A four-feature radiomics signature (RS) demonstrated excellent predictive accuracy for EGFR MT, with α-binormal-based and empirical AUCs of 0.931 (95% CI: 0.880-0.940) and 0.926 (95% CI: 0.877-0.933), respectively. Incorporating deep learning signature (DLS) further enhanced the model's performance, achieving α-binormal-based and empirical AUCs of 0.943 (95% CI: 0.921-0.965) and 0.938 (95% CI: 0.914-0.962) in the training dataset. These findings were confirmed in the validation dataset, with AUCs of 0.936 (95% CI: 0.917-0.955) and 0.921 (95% CI: 0.901-0.941), demonstrating robust and consistent predictive performance.
The multi-sequence MRI-based deep learning radiomics model exhibited high efficacy in predicting EGFR mutation status in NSCLC patients with brain metastases. This approach, which integrates advanced radiological features with deep learning techniques, offers a non-invasive and accurate method for determining EGFR mutation status, potentially guiding personalized treatment decisions in clinical practice.
准确早期识别非小细胞肺癌（NSCLC）脑转移患者的表皮生长因子受体（EGFR）突变状态对于指导靶向治疗至关重要。本研究旨在开发一种利用多序列磁共振成像（MRI）的深度学习放射组学模型，以区分EGFR突变型（MT）和野生型（WT）。
在这项回顾性研究中，纳入了288例确诊为脑转移的NSCLC患者，其中106例为EGFR MT，182例为EGFR WT。所有患者被随机分为训练数据集（75%）和验证数据集（25%）。使用对比增强T1加权（T1CE）和T2加权（T2W）MRI图像从脑转移瘤中提取放射组学和深度学习特征。使用最小绝对收缩和选择算子（LASSO）和ResNet34进行特征提取和选择。使用受试者工作特征（ROC）曲线和曲线下面积（AUC）分析评估特征对EGFR突变状态的预测性能。
训练数据集和验证数据集之间未发现显著差异。一个四特征放射组学特征（RS）对EGFR MT显示出优异的预测准确性，基于α-双正态分布和经验的AUC分别为0.931（95%CI：0.880-0.940）和0.926（95%CI：0.877-0.933）。纳入深度学习特征（DLS）进一步提高了模型的性能，在训练数据集中基于α-双正态分布和经验的AUC分别达到0.943（95%CI：0.921-0.965）和0.938（95%CI：0.914-0.962）。这些结果在验证数据集中得到证实，AUC分别为0.936（95%CI：0.917-0.955）和0.921（95%CI：0.901-0.941），显示出强大且一致的预测性能。
基于多序列MRI的深度学习放射组学模型在预测NSCLC脑转移患者的EGFR突变状态方面表现出高效性。这种将先进的放射学特征与深度学习技术相结合的方法，为确定EGFR突变状态提供了一种非侵入性且准确的方法，可能在临床实践中指导个性化治疗决策。
