文献速递肿瘤 PET/CT甲状腺核医学

[18F]FDG PET/CT影像组学联合临床特征对不同临床分期肺腺癌EGFR突变状态的预测价值

The predictive value of [18F]FDG PET/CT radiomics combined with clinical features for EGFR mutation status in different clinical staging of lung adenocarcinoma.
PubMed 文献 · EJNMMI Res · 2023-04-04

回顾性纳入515例肺腺癌患者,按检查时间分为训练集404例和独立测试集111例,基于[18F]FDG PET/CT提取影像组学特征,用逻辑回归、随机森林和支持向量机构建9个模型。随机森林Rad-score在CT、PET和PET/CT三种模态中表现最佳,联合临床参数可提升PET/CT影像组学模型的预测性能,尤其对晚期肺腺癌患者。

要点

  • 随机森林Rad-score在CT、PET和PET/CT三种影像组学模型中表现最佳。
  • PET/CT联合模型在三个联合模型中表现最佳,训练集与测试集AUC分别为0.760和0.730。
  • 分层分析显示CT_RF对I-II期病变预测最佳,PET/CT联合模型对III-IV期病变预测最佳。

核心结论

  • 联合临床参数可提升PET/CT影像组学模型对肺腺癌EGFR突变状态的预测性能,尤其对晚期肺腺癌患者。

研究设计与样本

  • 回顾性收集515例肺腺癌患者,按检查时间分为训练集404例和独立测试集111例。
  • 对PET/CT图像半自动分割后提取影像组学特征,筛选CT、PET和PET/CT模态的最佳特征集。
  • 采用逻辑回归、随机森林和支持向量机构建9个影像组学模型,保留三种模态中测试集表现最佳的模型并计算Rad-score。
  • 结合有价值的临床参数(性别、吸烟史、结节类型、CEA、SCC-Ag)构建联合影像组学模型。

主要发现

  • 与逻辑回归和支持向量机相比,随机森林Rad-score在CT、PET和PET/CT三种影像组学模型中表现最佳,训练集和测试集AUC分别为0.688、0.666、0.698和0.726、0.678、0.704。
  • 三个联合模型中PET/CT联合模型表现最佳,训练集和测试集AUC分别为0.760和0.730。
  • 分层分析发现CT_RF对I-II期病变预测效果最佳,训练集和测试集AUC分别为0.791和0.797。
  • PET/CT联合模型对III-IV期病变预测效果最佳,训练集和测试集AUC分别为0.722和0.723。

关键数值与推荐

  • 随机森林Rad-score在CT、PET和PET/CT三种影像组学模型中训练集AUC分别为0.688、0.666和0.698,测试集AUC分别为0.726、0.678和0.704。
  • PET/CT联合模型训练集AUC为0.760,测试集AUC为0.730。
  • CT_RF对I-II期病变训练集AUC为0.791,测试集AUC为0.797。
  • PET/CT联合模型对III-IV期病变训练集AUC为0.722,测试集AUC为0.723。

临床怎么用

  • 原文未据此给出推荐。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: This study aims to construct radiomics models based on [18F]FDG PET/CT using multiple machine learning methods to predict the EGFR mutation status of lung adenocarcinoma and evaluate whether incorporating clinical parameters can improve the performance of radiomics models. METHODS: A total of 515 patients were retrospectively collected and divided into a training set (n = 404) and an independent testing set (n = 111) according to their examination time. After semi-automatic segmentation of PET/CT images, the radiomics features were extracted, and the best feature sets of CT, PET, and PET/CT modalities were screened out. Nine radiomics models were constructed using logistic regression (LR), random forest (RF), and support vector machine (SVM) methods. According to the performance in the testing set, the best model of the three modalities was kept, and its radiomics score (Rad-score) was calculated. Furthermore, combined with the valuable clinical parameters (gender, smoking history, nodule type, CEA, SCC-Ag), a joint radiomics model was built. RESULTS: Compared with LR and SVM, the RF Rad-score showed the best performance among the three radiomics models of CT, PET, and PET/CT (training and testing sets AUC: 0.688, 0.666, and 0.698 vs. 0.726, 0.678, and 0.704). Among the three joint models, the PET/CT joint model performed the best (training and testing sets AUC: 0.760 vs. 0.730). The further stratified analysis found that CT_RF had the best prediction effect for stage I-II lesions (training set and testing set AUC: 0.791 vs. 0.797), while PET/CT joint model had the best prediction effect for stage III-IV lesions (training and testing sets AUC: 0.722 vs. 0.723). CONCLUSIONS: Combining with clinical parameters can improve the predictive performance of PET/CT radiomics model, especially for patients with advanced lung adenocarcinoma.

来源:EJNMMI Res,查看原文。版权归原作者所有。

引用

  1. Gao J, Niu R, Shi Y, et al. The predictive value of [18F]FDG PET/CT radiomics combined with clinical features for EGFR mutation status in different clinical staging of lung adenocarcinoma. EJNMMI Res. 2023. PMID: 37014500. doi:10.1186/s13550-023-00977-4

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。