Association between default mode network functional connectivity and plasma p-Tau217 levels in patients with mild cognitive impairment
Shang Zhang, Haoyu Li, Junyao Wang, Duan Liu, Hongyu Zhang, Liang Gong
Study at a glance
AI-extracted from the abstract| Characteristics | Longitudinal |
|---|---|
| Topics | Default mode network |
Abstract
Background: Mild cognitive impairment (MCI) is a critical stage in the Alzheimer's disease continuum, yet accessible biomarkers reflecting both peripheral pathology and central network dysfunction are insufficient. This study investigated whether default mode network (DMN) functional connectivity correlates with plasma phosphorylated tau 217 (p-Tau217) in MCI patients.Methods: Thirty‑two MCI patients and 32 healthy controls underwent neuropsychological assessment, plasma p‑Tau217 measurement, and resting‑state fMRI. Seed‑based functional connectivity was analyzed using five DMN regions (medial prefrontal, bilateral posterior cingulate, bilateral precuneus). Group comparisons and partial Spearman correlations were performed with age, sex, and education as covariates. Support vector machine classification was exploratory.Results: MCI patients showed lower cognition and higher p‑Tau217 than controls. They exhibited increased DMN connectivity with dorsolateral prefrontal and inferior parietal regions, and decreased connectivity within posterior and medial DMN. In MCI, p‑Tau217 negatively correlated with left precuneus‑left dorsolateral prefrontal connectivity (rs = −0.38, p = 0.034) and left precuneus‑right dorsolateral prefrontal connectivity (rs = −0.37, p = 0.036). Higher p‑Tau217 also correlated with worse global cognition, executive function, and visuospatial performance. A combined model using DMN features and p‑Tau217 achieved better classification than either alone.Conclusions: MCI patients have elevated p‑Tau217 and altered DMN connectivity. The association between p‑Tau217 and precuneus‑prefrontal connectivity suggests a potential link between peripheral tau pathology and network dysfunction. Combined blood and neuroimaging biomarkers warrant further validation in larger longitudinal cohorts with confirmed AD pathology.