文献速递胸部PETCT

基于PET/CT影像组学预测非小细胞肺癌PD-L1状态的重复研究

Replication study of PD-L1 status prediction in NSCLC using PET/CT radiomics.
PubMed 文献 · Eur J Radiol · 2024-11-17

本研究用新收集的254例非小细胞肺癌患者[18F]FDG PET/CT数据,重复验证Zhao等提出的影像组学机器学习模型预测PD-L1表达(≥1%)状态。Rad-score模型平均AUC为0.593(软组织窗)和0.573(肺组织窗),复杂模型为0.505和0.519,临床模型为0.555,均低于Zhao等报告的0.761、0.769和0.64。研究未能重复原 findings,提示模型性能不佳,对影像组学特征分类PD-L1表达的价值提出疑问。

要点

  • Rad-score模型平均AUC仅0.593(STW)和0.573(LTW),低于原研究0.761。
  • 复杂模型平均AUC为0.505(STW)和0.519(LTW),低于原研究0.769。
  • 临床模型平均AUC为0.555,低于原研究0.64,所有模型性能显著低于原报告。

核心结论

  • 本研究未能重复Zhao等的原始 findings,提示影像组学特征从PET/CT图像分类PD-L1表达的预测价值存疑,并强调需要严格验证。

研究设计与样本

  • 重复研究,独立队列254例非小细胞肺癌患者,行[18F]FDG PET/CT检查,原发肿瘤分割采用肺组织窗(LTW)和更保守的软组织窗(STW)两种方法,通过10折交叉验证和AUC分析评估Zhao等的影像组学模型(Rad-score和复杂模型)及临床分期模型,并进行比较不同机器学习模型流程的基准研究,临床病理数据从病历收集。

诊断标准与关键数值

  • PD-L1表达状态定义为≥1%;Rad-score模型平均AUC为0.593(STW)和0.573(LTW),原研究为0.761;复杂模型平均AUC为0.505(STW)和0.519(LTW),原研究为0.769;临床模型平均AUC为0.555,原研究为0.64;所有模型性能显著低于原研究。

局限性

  • 本研究未能重复原始 findings,提示模型性能不佳,对影像组学特征分类PD-L1表达的预测价值提出疑问。
  • 结果凸显重复影像组学机器学习模型的挑战,并强调需要严格验证。

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)。

英文原文(摘要)· 点击展开

This study investigates the predictive capability of radiomics in determining programmed cell death ligand 1 (PD-L1) expression (>=1%) status in non-small cell lung cancer (NSCLC) patients using a newly collected [18F]FDG PET/CT dataset. We aimed to replicate and validate the radiomics-based machine learning (ML) model proposed by Zhao et al. [1] predicting PD-L1 status from PET/CT-imaging. An independent cohort of 254 NSCLC patients underwent [18F]FDG PET/CT imaging, with primary tumor segmentation conducted using lung tissue window (LTW) and more conservative soft tissue window (STW) methods. Radiomics models ("Rad-score" and "complex model") and a clinical-stage model from Zhao et al. were evaluated via 10-fold cross-validation and AUC analysis, alongside a benchmark-study comparing different ML-model pipelines. Clinicopathological data were collected from medical records. On our data, the Rad-score model yielded mean AUCs of 0.593 (STW) and 0.573 (LTW), below Zhao et al.'s 0.761. The complex model achieved mean AUCs of 0.505 (STW) and 0.519 (LTW), lower than Zhao et al.'s 0.769. The clinical model showed a mean AUC of 0.555, below Zhao et al.'s 0.64. All models performed significantly lower than Zhao et al.'s findings. Our benchmark study on four ML pipelines revealed consistently low performance across all configurations. Our study failed to replicate original findings, suggesting poor model performance and questioning predictive value of radiomics features in classifying PD-L1 expression from PET/CT imaging. These results highlight challenges in replicating radiomics-based ML models and stress the need for rigorous validation.

来源:Eur J Radiol,查看原文。版权归原作者所有。

引用

  1. Stüber AT, Heimer MM, Ta J, et al. Replication study of PD-L1 status prediction in NSCLC using PET/CT radiomics. Eur J Radiol. 2024. PMID: 39657546. doi:10.1016/j.ejrad.2024.111825

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。