文献速递腹部泌尿生殖PETCT

双示踪PET/CT机器学习预测GEP-NET肝转移奥曲肽疗效

Predicting Response to Somatostatin Analog Therapy in GEP-NET Liver Metastases: A Machine Learning Approach with ⁶⁸Ga-DOTATATE and ¹⁸F-FDG PET/CT.
PubMed 文献 · Acad Radiol · 2026-09-18

回顾性纳入32例接受一线奥曲肽治疗的GEP-NET肝转移患者,从202个肝病灶提取PET/CT定量参数、双示踪剂比值及病灶大小。通过LASSO筛选特征并构建多个机器学习模型,XGBoost在患者级验证队列AUC为0.825,留一患者交叉验证AUC为0.856。SHAP分析显示⁶⁸Ga-DOTATATE TLR、原发灶大小和⁶⁸Ga-DOTATATE SUVmean为最重要预测因子。

要点

  • ⁶⁸Ga-DOTATATE TLR、原发灶大小及SUVmean是预测奥曲肽疗效的三个最重要指标。
  • 双示踪PET/CT定量参数可无创分层GEP-NET肝转移患者的奥曲肽治疗反应。
  • XGBoost模型在留一患者交叉验证中AUC达0.856,具有较好的稳健性。

核心结论

  • 整合双示踪PET/CT参数的机器学习模型可有效预测GEP-NET肝转移患者对奥曲肽治疗的反应。
  • 该模型可为治疗反应分层和个体化管理提供无创工具。

研究设计与样本

  • 回顾性研究,纳入32例接受一线奥曲肽治疗的GEP-NET肝转移患者。
  • 从202个肝病灶提取⁶⁸Ga-DOTATATE和¹⁸F-FDG PET/CT定量参数、双示踪剂比值及病灶大小。
  • 病变按RECIST 1.1尺寸变化阈值分为疾病控制组(SD/PR)和疾病进展组(PD)。
  • 在患者水平划分训练和验证队列,采用LASSO特征筛选并构建多个机器学习模型。
  • 使用留一患者交叉验证(LOPOCV)评估稳健性,SHAP分析解释模型。

关键影像表现/征象

  • ⁶⁸Ga-DOTATATE TLR、rTLR和原发灶大小在疾病控制组与疾病进展组之间有显著差异。
  • LASSO筛选出六个关键预测特征。

诊断标准与关键数值

  • XGBoost在患者级验证队列AUC为0.825,准确率为0.742。
  • XGBoost在留一患者交叉验证中AUC为0.856,准确率为0.787。
  • SHAP分析显示⁶⁸Ga-DOTATATE TLR、原发灶大小和⁶⁸Ga-DOTATATE SUVmean为最重要的三个预测因子。

临床应用与阅片要点

  • 在日常阅片中可重点提取⁶⁸Ga-DOTATATE TLR、原发灶大小和SUVmean用于评估奥曲肽疗效可能性。
  • 双示踪PET/CT参数联合机器学习模型有助于无创筛选适合奥曲肽治疗的GEP-NET肝转移患者。

局限性

  • 回顾性研究设计。
  • 样本量为32例患者,共202个肝病灶。

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

BACKGROUND: Predicting response to somatostatin analogs (SSAs) in gastroenteropancreatic neuroendocrine tumors (GEP-NETs) with liver metastases (LMs) remains challenging. This study aimed to develop and validate an interpretable machine learning (ML) model based on dual-tracer (68Ga-DOTATATE and 18F-FDG) PET/CT parameters to predict octreotide response and identify key imaging predictors using SHAP analysis.

METHODS: 32 GEP-NET patients with LMs who received first-line octreotide therapy were retrospectively enrolled. Quantitative PET/CT metrics, dual-tracer ratios, and lesion size were extracted from 202 hepatic lesions. Lesions were classified as disease-control (SD/PR) or disease-progression (PD) using size-change thresholds derived from RECIST 1.1. Patients were divided at the patient level into training and validation cohorts. LASSO regression was used for feature selection, followed by the construction of multiple ML models. Leave-one-patient-out cross-validation (LOPOCV) was performed to further assess model robustness, and SHAP analysis was applied for model interpretability.

RESULTS: ⁶⁸Ga-DOTATATE TLR, rTLR, and primary size differed significantly between disease-control and disease-progression groups. LASSO identified six key predictive features. The XGBoost model showed the best performance in the patient-level validation cohort, with an AUC of 0.825 and an accuracy of 0.742. In LOPOCV, XGBoost achieved an AUC of 0.856 and an accuracy of 0.787. SHAP analysis revealed that ⁶⁸Ga-DOTATATE TLR, primary size, and ⁶⁸Ga-DOTATATE SUVmean were the three most important predictors.

CONCLUSION: An interpretable ML model integrating dual-tracer PET/CT parameters can effectively predict octreotide response in GEP-NETs with LMs. This approach may provide a noninvasive tool for treatment-response stratification and individualized management.

来源:Acad Radiol,查看原文。版权归原作者所有。

引用

  1. Xu, Yu, Wang, et al. Predicting Response to Somatostatin Analog Therapy in GEP-NET Liver Metastases: A Machine Learning Approach with ⁶⁸Ga-DOTATATE and ¹⁸F-FDG PET/CT. Acad Radiol. 2026. PMID: 42760167. doi:10.1016/j.acra.2026.08.115

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。