Eur Geriatr Med. 2026 Jul 28. doi: 10.1007/s41999-026-01562-3. Online ahead of print.
ABSTRACT
PURPOSE: Cortisol dysregulation has been implicated in Alzheimer’s disease (AD), but its associations with amyloid and tau-related outcomes remain unclear. This study examined whether baseline CSF cortisol was associated with cross-sectional and baseline-adjusted 24-month multimodal AD biomarkers and cognitive outcomes.
METHODS: A total of 764 participants were included, comprising cognitively normal (CN; n = 284), individuals with mild cognitive impairment (MCI; n = 356), and patients with AD dementia (n = 124). Associations between baseline CSF cortisol, multimodal AD biomarkers, and cognitive outcomes were assessed with FDR correction.
RESULTS: CSF cortisol increased modestly from CN to MCI and AD dementia, with a small but statistically significant overall group difference after FDR correction. Cross-sectionally, in the MCI group, higher CSF cortisol was associated with higher CSF total tau and p-tau181 and a lower Aβ42/total tau ratio. In baseline-adjusted 24-month analyses, in MCI groups, higher CSF was associated with higher CSF total tau and p-tau181, greater temporal tau-PET burden, lower hippocampal volume, greater ventricular volume, and poorer cognitive outcomes. In the full cohort and MCI group, CSF-defined amyloid positivity strengthened associations of higher baseline CSF cortisol with adverse 24-month tau-related outcomes. Exploratory analyses suggested that 24-month CSF total tau and hippocampal volume partly accounted for the association between higher baseline CSF cortisol and poorer 24-month cognition in MCI; however, these findings do not establish causal mediation.
CONCLUSION: Higher CSF cortisol may be linked to tau-related pathology and poorer cognition, supporting further evaluation of its prognostic value across the AD continuum.
PMID:42518151 | DOI:10.1007/s41999-026-01562-3
