Differences between ante mortem Alzheimer's disease biomarkers in predicting neuropathology at autopsy

Authors:
  • FIFor The Alzheimer'S Disease Neuroimaging Initiative
    For The Alzheimer'S Disease Neuroimaging Initiative
Publisher website
Google Scholar
Abstract
Introduction: This study aimed to assess whether biomarkers related to amyloid, tau, and neurodegeneration can accurately predict Alzheimer's disease (AD) neuropathology at autopsy in early and late clinical stages.Methods: We included 100 participants who had ante mortem biomarker measurements and underwent post mortem neuropathological examination. Based on ante mortem clinical diagnosis, participants were divided into non‐dementia and dementia, as early or late clinical stages.Results: Amyloid positron emission tomography (PET) and cerebrospinal fluid (CSF) amyloid beta (Aβ)42/phosphorylated tau (p‐tau)181 showed excellent performance in differentiating autopsy‐confirmed AD and predicting the risk of neuropathological changes in early and late clinical stages. However, CSF Aβ42 performed better in the early clinical stage, while CSF p‐tau181, CSF t‐tau, and plasma p‐tau181 performed better in the late clinical stage.Discussion: Our findings provide important clinical information that, if using PET, CSF, and plasma biomarkers to detect AD pathology, researchers must consider their differential performances at different clinical stages of AD.Highlights: Amyloid PET and CSF Aβ42/p‐tau181 were the most promising candidate biomarkers for predicting AD pathology. CSF Aβ42 can serve as a candidate predictive biomarker in the early clinical stage of AD. CSF p‐tau181, CSF t‐tau, and plasma p‐tau181 can serve as candidate predictive biomarkers in the late clinical stage of AD. Combining APOE ε4 genotypes can significantly improve the predictive accuracy of AD‐related biomarkers for AD pathology.
Funding Information
  • National Natural Science Foundation of China (82071201, 81971032)
  • Huashan Hospital (2022QD002)
  • Fudan University (3030277001)

This publication has 48 references indexed in Scilit: