Correlation and Interchangeability of Amyloid, Tau, and Glucose Metabolism PET in Mild Cognitive Impairment and Alzheimer: A Review

轻度认知障碍和阿尔茨海默病中淀粉样蛋白、Tau蛋白和葡萄糖代谢PET的相关性和互换性:综述

阅读:2

Abstract

Positron emission tomography (PET) allows for minimally invasive in vivo localization of amyloid and tau deposition, and visualization of glucose metabolism using amyloid PET, tau PET, and FDG PET. Clinically, these scans are used to determine A, T, and N (amyloid-β plaques, tau tangles, and neurodegeneration) status in Alzheimer's disease. In light of the recent anti-amyloid therapies, determination of the A, and the associated T and N status is increasingly important. This review explores the potential of a single PET scan to define multiple biomarkers. A literature search using the PubMed database and an additional citation search using Google Scholar identified 76 relevant publications up to 30 July 2025. Original work reporting amyloid, tau or FDG PET to determine two or more ATN-related biomarkers were included. Non-English, animal, and non-dementia related studies were excluded. For each study, quantitative outcomes such as correlations and ROC AUC scores were extracted and compared. Early phase amyloid and tau PET (n = 58) were consistently found to be good indicators of N status with a median (IQR) correlation of 0.82 (0.76-0.86). Limited research (n = 7) was performed for amyloid or tau PET to infer both A and T status, with tau-based studies having slightly higher ROC AUC scores (0.88-0.99) than amyloid-based studies (0.84-0.9). Initial results are promising (median ROC AUC scores of 0.88 (0.81-0.96)) but need to be validated. FDG PET was found to be less accurate for A or T status (n = 12) prediction (median ROC AUC scores of 0.83 (0.80-0.87)). Among the modalities, tau PET seems to be the most promising candidate for a single tracer approach to predict all three biomarkers.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。