Liang Chen, Zheng Meiyu, Zou Han, Han Yu, Zhan Yingying, Xing Yu, Liu Chang, Zuo Chao, Zou Jinhai
Department of Nuclear Medicine, Cangzhou Central Hospital, Cangzhou, China.
The Second Clinical Medical College of Lanzhou University, Lanzhou, China.
Front Oncol. 2024 Sep 5;14:1402994. doi: 10.3389/fonc.2024.1402994. eCollection 2024.
There is still a lack of clinically validated biomarkers to screen lung cancer patients suitable for programmed dead cell-1 (PD-1)/programmed dead cell receptor-1 (PD-L1) immunotherapy. Detection of PD-L1 expression is invasively operated, and some PD-L1-negative patients can also benefit from immunotherapy; thus, the joint modeling of both deep learning images and clinical features was used to improve the prediction performance of PD-L1 expression in non-small cell lung cancer (NSCLC).
Retrospective collection of 101 patients diagnosed with pathology in our hospital who underwent 18F FDG PET/CT scans, with lung cancer tissue Tumor Propulsion Score (TPS) ≥1% as a positive expression. Lesions were extracted after preprocessing PET/CT images, and using deep learning 3D DenseNet121 to learn lesions in PET, CT, and PET/CT images, 1,024 fully connected features were extracted; clinical features (age, gender, smoking/no smoking history, lesion diameter, lesion volume, maximum standard uptake value of lesions [SUVmax], mean standard uptake value of lesions [SUVmean], total lesion glycolysis [TLG]) were combined for joint modeling based on the structured data Category Embedding Model.
Area under a receiver operating characteristic (ROC) curve (AUC) and accuracy of predicting PD-L1 positive for PET, CT, and PET/CT test groups were 0.814 ± 0.0152, 0.7212 ± 0.0861, and 0.90 ± 0.0605, 0.806 ± 0.023, 0.70 ± 0.074, and 0.950 ± 0.0250, respectively. After joint clinical feature modeling, the AUC and accuracy of predicting PD-L1 positive for PET/CT were 0.96 ± 0.00905 and 0.950 ± 0.0250, respectively.
This study combines the features of F-FDG PET/CT images with clinical features using deep learning to predict the expression of PD-L1 in NSCLC, suggesting that F-FDG PET/CT images can be conducted as biomarkers for PD-L1 expression.
目前仍缺乏经临床验证的生物标志物来筛选适合程序性死亡细胞1（PD-1）/程序性死亡细胞受体1（PD-L1）免疫治疗的肺癌患者。PD-L1表达检测具有侵入性，且一些PD-L1阴性患者也可从免疫治疗中获益；因此，采用深度学习图像与临床特征联合建模的方法来提高非小细胞肺癌（NSCLC）中PD-L1表达的预测性能。
回顾性收集我院101例经病理诊断且接受18F FDG PET/CT扫描的患者，以肺癌组织肿瘤阳性比例分数（TPS）≥1%作为阳性表达。对PET/CT图像进行预处理后提取病灶，使用深度学习3D DenseNet121分别学习PET、CT及PET/CT图像中的病灶，提取1024个全连接特征；将临床特征（年龄、性别、吸烟/非吸烟史、病灶直径、病灶体积、病灶最大标准摄取值[SUVmax]、病灶平均标准摄取值[SUVmean]、病灶总糖酵解量[TLG]）基于结构化数据类别嵌入模型进行联合建模。
PET、CT及PET/CT测试组预测PD-L1阳性的受试者工作特征（ROC）曲线下面积（AUC）及准确率分别为0.814±0.0152、0.7212±0.0861及0.90±0.0605，0.806±0.023、0.70±0.074及0.950±0.0250。联合临床特征建模后，PET/CT预测PD-L1阳性的AUC及准确率分别为0.96±0.00905及0.950±0.0250。
本研究利用深度学习将F-FDG PET/CT图像特征与临床特征相结合，以预测NSCLC中PD-L1的表达，提示F-FDG PET/CT图像可作为PD-L1表达的生物标志物。
