Depressive symptoms may signal early tau pathology in older adults

Clinical takeaway: New or worsening depressive symptoms in cognitively normal older adults may reflect emerging Alzheimer's pathology rather than only a mood disorder. Tracking mood trajectories could add early-detection value alongside biomarker and cognitive measures. 

When an older patient with intact cognition develops new depressive symptoms, the clinical meaning is ambiguous. The symptoms may represent a treatable mood disorder, a risk factor that raises the odds of future dementia, or an early manifestation of neurodegenerative disease already underway. Each reading points toward different monitoring and different conversations. Depressive symptoms affect upwards of a third of people with mild cognitive impairment or dementia, but their significance in patients whose cognition remains normal has been more difficult to pin down. 

Higher tau burden has repeatedly been correlated with the presence and severity of depressive symptoms, and a late-life depression diagnosis has been linked to increased risk of Alzheimer's dementia. What the correlational evidence could not resolve is order: whether mood changes are an early symptom of accumulating pathology or a contributor that increases vulnerability to it. A longitudinal analysis pairing repeated tau imaging with mood assessments now offers a temporal ordering. 

In cognitively normal older adults, tau came first. Three separate analyses pointed the same way. Participants with the fastest tau accumulation showed the steepest rises in depressive symptoms over time, measured by the Geriatric Depression Scale (GDS). But mood trajectories did not forecast tau accumulation. 

Modeling built to test directionality reached the same answer. Higher tau at one timepoint predicted higher GDS scores later. The effect was significant but carried a wide range of uncertainty. Depressive symptoms, in turn, did not predict later tau. 

A third analysis estimated when tau burden crossed into positivity for the subset of participants who became tau positive. The longer someone had been tau positive, the steeper their symptom rise. The pattern also did not depend on amyloid. Most of the cognitively normal group was amyloid negative, and adjusting for amyloid status left the results unchanged. 

None of this held in participants with mild cognitive impairment (MCI) or Alzheimer's disease (AD) dementia. The coupling between tau and mood appeared only before impairment, the early window in which the authors argue the signal may be most useful. 

The findings come from 390 participants in the Alzheimer's Disease Neuroimaging Initiative: 226 cognitively normal and 164 with MCI or AD dementia. None was clinically depressed at entry. Each completed at least two tau PET scans with matched GDS assessments, spanning roughly three years on average. 

Longer follow-up in the same cohort is ongoing as participants age, and some will cross into cognitive impairment or clinically significant depression. The authors point to scalable measurement tools, including momentary symptom assessment and wearables, as what has to develop before mood-trajectory monitoring becomes practical at scale. 

"Continuing to follow these trajectories, observing how early mood changes relate to later clinical outcomes, and establishing relationships with other methods tracking the development of Alzheimer's disease will provide more clarity on methods to determine which individuals may be at an increased risk," said Teodora Z. Markova of the Volen Center for Complex Systems, Brandeis University. 

Source: Markova TZ, et al. (2026 Aug 24) J Neurosci. Tracking the Continuous Time Dynamics of Tau and Depressive Symptoms Across the Alzheimer's Disease Continuum