Brain health biomarkers: what sleep, steps and genetics can reveal

Brain health biomarkers are moving beyond a narrow focus on amyloid plaques. New research points to a wider network involving sleep architecture, brain immune activity, blood-brain barrier function, movement and cognitive change. Together, these signals may help explain why people with apparently similar Alzheimer’s risk can follow very different paths.

For women, that broader view matters. Women make up a substantial share of people living with Alzheimer’s disease, yet risk cannot be reduced to sex alone. Age, longevity, cardiovascular health, sleep, genetics and the hormonal transitions of midlife may all contribute. The latest studies do not settle how these factors interact in women, but they show why repeated, connected measurements are likely to be more informative than one isolated result.

Brain health biomarkers extend beyond amyloid

Amyloid remains a central Alzheimer’s biomarker, but plaque burden does not perfectly track symptoms or disease progression. A 2026 animal study in Alzheimer’s & Dementia examined sleep in APP/PS1 mice using EEG and EMG recordings alongside whole-brain measurements of amyloid and microglia, the brain’s resident immune cells. The researchers found that amyloid pathology was associated with reduced non-rapid eye movement sleep, cortical hyperexcitability, disrupted network coordination and microglial expansion into brain systems involved in sleep and wakefulness.

Temporarily depleting microglia restored more than two hours of sleep per day without reducing amyloid burden. That distinction is important. It suggests that sleep loss in this mouse model was not simply proportional to the amount of plaque. An immune response associated with the pathology appeared to help drive the functional disruption.

This was a mechanistic mouse experiment, not a human treatment trial. Removing microglia experimentally is not evidence that doing so would be safe or effective in people. The result instead identifies several candidate biomarkers that could be studied together, including EEG patterns, sleep duration, network excitability and inflammatory activity. It also reinforces the possibility that disturbed sleep can be an early expression of disease biology rather than merely a consequence of advanced dementia.

Sleep is useful, but not diagnostic

Sleep is appealing as a biomarker because it can be measured repeatedly and at relatively low burden. Yet total sleep time alone is blunt. The animal study distinguished NREM sleep, REM sleep, rebound after deprivation and electrical activity across brain networks. Consumer devices generally estimate sleep from movement and cardiovascular signals rather than directly measuring brain waves, so their outputs are not equivalent to laboratory EEG.

For women, sleep also varies with menstrual symptoms, pregnancy, caregiving demands, stress, medication use and perimenopause. Night sweats and changing hormone patterns can fragment sleep during the menopause transition. These influences do not make sleep tracking meaningless. They make context essential. A sustained change in sleep paired with changes in mood, cognition, energy or recovery carries more information than a single poor night or an unexplained wearable score.

The practical lesson is not that sleep can diagnose Alzheimer’s. It cannot. Rather, recurring sleep disruption may be one part of a longitudinal health record that helps a clinician interpret symptoms and decide whether further assessment is appropriate. How perimenopause tracking works can also help separate persistent trends from fluctuations associated with cycle stage or the menopause transition.

Brain health biomarkers may combine movement and cognition

A separate report linked daily walking with slower Alzheimer’s-related change. The news account described associations between physical activity, lower tau accumulation and less cognitive and functional decline, with apparent benefits beginning around 3,000 steps per day and strongest patterns between 5,000 and 7,500. However, the verified item is a 2025 Nature news article rather than the underlying research paper. Its findings have not been traced here to a published primary study, and the supplied material does not establish the design, sample size or representation of women.

That means the step thresholds should not be treated as prescriptions or causal cut-offs. People who walk more may differ in cardiovascular health, mobility, socioeconomic conditions or pre-existing brain health. Reverse causation is also possible, because very early cognitive or neurological changes could reduce activity before diagnosis.

Even with those limitations, steps are potentially valuable as a functional biomarker. A person’s usual activity level, and how it changes over months, may be more meaningful than comparison with a universal target. Declining movement could reflect pain, fatigue, depression, sleep disruption, illness or neurological change. Pairing steps with subjective energy and recovery can make the pattern easier to interpret; energy tracking over time provides a framework for connecting those signals rather than viewing activity in isolation.

Genetic risk does not determine outcome

A 2026 study in Nature Aging examined a different layer of Alzheimer’s biology: how APOE4 may contribute to dysfunction of the blood-brain barrier. The work identified fibronectin as a mediator in that pathway and was reported in the context of people who remain cognitively resilient despite high genetic risk. The supplied information does not report the study’s sample size or fully describe its human and experimental components, so its clinical reach should be interpreted cautiously.

The broader biomarker point is clearer than any immediate treatment implication. A risk-associated gene is not a forecast with a fixed outcome. Its effects operate through biological pathways, and protective variants or differences in those pathways may alter what happens downstream. Genetics, vascular integrity, protein accumulation, inflammation and cognition represent related but non-interchangeable layers of information.

This also matters when interpreting direct-to-consumer genetic results. APOE status cannot establish whether someone has Alzheimer’s, when symptoms might appear or whether they will appear at all. Genetic testing can carry consequences for family members and emotional wellbeing, so clinical guidance and appropriate counselling are more useful than reading a risk variant in isolation.

A pattern is more informative than a snapshot

These studies support a multidimensional view of brain ageing, but important gaps remain. The strongest mechanistic sleep finding comes from mice. The walking thresholds come through secondary reporting without the primary paper verified here. The genetics research advances understanding of a pathway but does not yet provide a routine test or proven prevention strategy. None of the supplied studies establishes how menopause timing, hormone therapy or cycle history modifies these specific biomarkers.

For now, sleep, movement, cognition and energy are best regarded as complementary signals rather than stand-alone tests. New or progressive memory difficulties still require clinical assessment, particularly when they interfere with everyday function. Potentially treatable contributors such as sleep disorders, mood symptoms, medication effects and nutritional deficiencies may also need evaluation.

Changes in sleep quality, daily movement, energy and cognitive symptoms only become readable when they are logged consistently over weeks. This is the kind of connected pattern Tulsy is built to surface.

Sources: ScienceDaily, Nature.

This content is for informational and educational purposes only. It is not intended to diagnose, treat, cure, or prevent any disease, and should not replace advice from a qualified healthcare professional.