Blood-based biomarkers across the Alzheimer's disease continuum in older Korean population

Abstract

Background: Plasma phosphorylated tau 217 (pTau217) is a robust biomarker of Alzheimer's disease (AD) pathology. However, whether a single biomarker can adequately capture both clinical diagnosis and disease progression remains unclear. We investigated the complementary roles of plasma AD biomarkers and routinely measured metabolic markers across different stages of AD. Methods: Using the Gwangju Alzheimer's & Related Dementia (GARD) cohort, we analyzed cross-sectional and longitudinal data, including amyloid-PET, MRI, neuropsychological examinations, and APOE genotypes. Plasma pTau217, GFAP, and NfL levels were measured alongside routine biochemical markers to evaluate their associations with pathology and neurodegeneration. Results: Plasma pTau217 best identified amyloid-PET positivity (AUC > 0.91). For AD diagnosis, adding cholesterol and glucose to pTau217 improved the AUC from 0.781 to 0.873. To predict progression in amyloid-positive individuals, glucose alone outperformed pTau217 (AUC 0.840 vs. 0.747), whereas adding GFAP yielded the strongest prognostic performance (AUC 0.867). Discussion: Plasma pTau217 primarily reflects amyloid pathology, whereas metabolic markers provide complementary information related to the clinical expression and progression of the disease. Integrating biomarkers across distinct biological domains may improve risk stratification and prognostic assessment in AD.