Alzheimer’s disease affects an estimated 7.2 million Americans age 65 and older, according to the Alzheimer’s Association. Current diagnostic tests typically measure the levels of two proteins — amyloid beta (Aβ) and phosphorylated tau (p-tau) — in blood or spinal fluid. While these biomarkers are widely used, they may not fully reflect the earliest biological changes that occur as the disease develops.
Researchers at Scripps Research have now introduced a different type of blood test that focuses on how proteins are folded in the bloodstream rather than how much of them is present. Their findings, published in Nature Aging on February 27, 2026, show that structural differences in three plasma proteins are strongly linked to Alzheimer’s status. These changes allowed scientists to accurately distinguish cognitively normal individuals from those with Alzheimer’s and mild cognitive impairment (MCI). The method could eventually allow diagnosis and treatment to begin earlier.
“Many neurodegenerative diseases are driven by changes in protein structure,” says senior author John Yates, a professor at Scripps Research. “The question was, are there structural changes in specific proteins that might be useful as predictive markers?”
Protein Folding and the Breakdown of Proteostasis
For many years, Alzheimer’s disease has been closely associated with amyloid plaques and tau tangles that accumulate in the brain. However, scientists increasingly believe that the condition may involve a broader failure in proteostasis, the system responsible for keeping proteins properly folded and removing damaged ones.
As people age, this system becomes less effective. Proteins are then more likely to fold incorrectly during production or maintenance. Based on this idea, the researchers proposed that if proteostasis is disrupted in the brain, similar structural changes might also appear in proteins circulating through the blood.
Analyzing Structural Changes in Blood Proteins
To explore this possibility, the research team examined plasma samples from 520 participants divided into three groups: cognitively normal adults, individuals with mild cognitive impairment and patients diagnosed with Alzheimer’s.
The scientists used mass spectrometry to determine how exposed or buried certain locations within proteins were, which indicates changes in their structure. They then applied machine learning techniques to identify patterns connected to disease stage.
The results revealed a clear pattern across all groups. As Alzheimer’s progressed, some blood proteins became less structurally “open.” These structural changes proved to be more informative for identifying disease stage than simply measuring protein concentrations.
Three Proteins Linked to Alzheimer’s Progression
Among the many proteins analyzed, three showed the strongest association with disease status. These were C1QA, which plays a role in immune signaling; clusterin, which is involved in protein folding and amyloid removal; and apolipoprotein B, a protein that transports fats in the bloodstream and contributes to blood vessel health.
“The correlation was amazing,” says co author Casimir Bamberger, a senior scientist at Scripps Research. “It was very surprising to find three lysine sites on three different proteins that correlate so highly with disease state.”
Changes at specific sites within these proteins enabled researchers to classify participants as cognitively normal, MCI or Alzheimer’s with about 83% overall accuracy. When comparing two groups directly, such as healthy individuals versus those with MCI, accuracy rose above 93%.
Tracking Alzheimer’s Over Time
The three protein model remained reliable when tested in independent participant groups and when researchers analyzed blood samples collected months later.
In repeat tests taken months apart, the panel identified disease status with about 86% accuracy and reflected changes in diagnosis over time. The structural score also showed a strong relationship with cognitive test results and a more moderate association with MRI measurements of brain shrinkage.
Together, these findings suggest that analyzing protein structure in blood could complement existing amyloid and tau tests. Because this method focuses on structural changes connected to the underlying biology of the disease, it may help researchers identify disease stages, monitor progression and evaluate how well treatments are working.
Future Applications and Next Steps
“Detecting markers of Alzheimer’s early is absolutely critical to developing effective therapeutics,” says Yates. “If treatment can start before significant damage has been done, it may be possible to better preserve long-term memory.”
Before the blood test can be used in clinical settings, larger studies with longer follow up periods will be needed to confirm the results. Researchers are also exploring whether the same structural profiling method could be applied to other diseases, including Parkinson’s and cancer.
In addition to Yates and Bamberger, authors of the study “Structural signature of plasma proteins classifies the status of Alzheimer’s disease,” include Ahrum Son, Hyunsoo Kim and Jolene K. Diedrich of Scripps Research; Heather M. Wilkins, Jeffrey M. Burns, Jill K. Morris and Russell H. Swerdlow of the University of Kansas Medical Center; and Robert A. Rissman of the University of California San Diego.
Support for this study was provided by the National Institutes of Health (grants RF1AG061846-01, 5R01AG075862, P30AG072973 and P30-AG066530).