Menopause-Linked Blood Proteins Track Later Brain Health

On September 22, 2026, Nature Medicine published “Blood proteomics of menopause map to brain aging and dementia risk”, a study involving UCSB researchers. The team examined whether blood-protein patterns associated with menopause also tracked with cognitive outcomes later in life.

The central finding

The researchers identified a menopause-associated protein pattern that was linked to less favorable cognitive outcomes across four cohorts totaling 11,925 older women. In two longitudinal cohorts, higher menopause proteomic scores were associated with steeper cognitive change over an average follow-up of about five years. In a third cohort, higher scores were associated with lower global cognitive performance. For healthy-aging research, the central finding is a connection between midlife reproductive biology and later cognitive health that gives researchers a framework for further investigation.

Evidence in people

Study type: Observational research in humans, not in vitro experiments or animal models.

The study began with a discovery sample of 80 women aged 43 to 58. According to the paper, the researchers identified 16 proteins elevated in postmenopausal compared with premenopausal women after adjusting for age. This comparison was the starting point for constructing the menopause-associated signature. The researchers then examined how that signature related to hormonal measures and how it performed in additional cohorts.

Age adjustment was central to the research question. The team was looking for a protein pattern associated with menopause rather than simply a pattern associated with being older. In the discovery cohort, the combined score tracked with menopause stage and hormonal measures more closely than with age, according to the paper’s abstract. That result gave the researchers a reason to investigate the pattern beyond its initial connection with menopause stage.

The 16-protein set covered several areas of biology. The paper grouped them as inflammatory, synaptic or neuronal, metabolic and Alzheimer’s-related proteins. The Alzheimer’s-related proteins included BACE1 and p-tau231. Taken together, the set gave the researchers a broader molecular pattern to study rather than requiring them to interpret each protein separately.

The menopause proteomic score summarized the joint pattern of those proteins. This distinction matters when reading the results: an association involving the combined score is a finding about that summary measure. It is different from reporting that every component had the same relationship with an outcome. Readers interested in that distinction can find further context in our guide to interpreting biological-age measurements.

The researchers also tested parts of the initial signal in a UK Biobank analysis of 2,814 women. The authors reported that nine of the 13 proteins measurable on both proteomic platforms replicated in that analysis. This step addressed a different question from the later cognitive analyses. It examined whether menopause-associated protein differences identified in the discovery sample also appeared in another group of women.

The older-women cohorts then provided a way to examine the score in relation to later-life outcomes. The study therefore connected two research questions: which blood-protein patterns were associated with menopause and whether a score built from those patterns was associated with cognition. Keeping those questions separate helps explain the study’s design. The discovery analysis established the signature used in the subsequent outcome analyses.

The cognitive findings also involved different kinds of measurements. The longitudinal analyses examined change over time, while the global cognitive performance result concerned overall performance. These are related questions rather than interchangeable findings. Reporting them separately preserves the distinction between an association with a cognitive trajectory and an association with a level of performance.

In UCSB’s account of the research, professor Emily Jacobs described menopause as “a whole-body biological transition.” That framing captures why the team examined a blood-based molecular pattern rather than treating menopause only as a reproductive event. Senior author Kaitlin Casaletto identified midlife as an important period for investigating opportunities to support brain resilience. The research places menopause within that broader effort to understand cognitive health across adulthood.

Where evidence stops

The paper’s limitations determine how the results should be interpreted:

• The discovery sample was small and cross-sectional. Its 80 participants provided a comparison between women at different menopause stages. The analysis did not track protein and hormone changes within the same women throughout the transition. It therefore could not establish the order in which those changes developed.
• Association does not establish cause. The findings do not show that menopause caused the protein differences or that those proteins caused dementia. Casaletto emphasized this distinction in UCSB’s reporting. A relationship between measurements provides a research lead rather than a demonstrated biological chain of events.
• Menopause symptoms were self-reported. The authors identified this as a limitation of the discovery analysis. Symptom reports were part of the evidence available to the researchers, so conclusions involving symptoms must be understood in the context of how that information was collected.
• Blood proteins can come from multiple tissues. The authors noted that a circulating protein may reflect activity in more than one part of the body. A blood-based association with cognition does not by itself locate the source of that signal in the brain. This matters when moving from a measurable pattern toward an explanation of the biology.
• Measurement platforms and cohorts differed. The study used different proteomic platforms across cohorts. The authors also noted differences in cohort composition that could affect comparisons. These differences matter when assessing how consistently a score or an individual protein behaves across study populations.
• Individual protein associations varied across cohorts. The authors reported that relationships between particular proteins and cognitive outcomes were not uniform. The combined score and its separate components therefore need to be evaluated on their own terms. A result for the score should not be assigned automatically to every protein it contains.
• The dementia findings were specific. The study did not find statistically significant associations between the score and all-cause dementia, vascular dementia or frontotemporal dementia. The authors described the Alzheimer’s dementia effect size as relatively small. The results should not be generalized into a claim that the signature predicts dementia broadly.
• Clinical prediction remains unestablished. Associations in older-women cohorts do not demonstrate that measuring this score in an individual midlife patient predicts her future diagnosis. UCSB described developing a blood test as a team goal. It also reported uncertainty about the clinical significance of individual Alzheimer’s-related protein changes detected in midlife.

For readers, the practical distinction is between following a research tool and making a personal medical decision. Menopause stage or symptoms should not be treated as an individual dementia prediction on the basis of this study. Personal concerns about symptoms or cognition belong in a discussion with a qualified health professional. Our guide to the evidence behind cognitive-health interventions offers a broader context for those questions.

The next test

Longitudinal research, including UCSB’s Longitudinal Menopause Project, should track how hormones and proteins change alongside brain health before prospective validation establishes whether this signature can contribute meaningfully to midlife risk assessment.

How AgeAmaze helps

Following menopause proteomics toward possible clinical use means assessing each new validation study, a step AgeAmaze supports through explanations of longevity research. Its coverage makes biomarkers and healthy-aging science more understandable for research-minded adults while keeping established findings distinct from speculative applications.

Read the research

Sources

1. Blood proteins reveal how menopause may shape brain health later life
2. Menopause-associated proteomic changes link to brain ... - Nature