Researchers at The Scripps Research Institute have developed a fundamentally new type of blood test for Alzheimer's disease that measures structural changes in proteins rather than their quantity. The study, published in Nature Aging on February 27, analyzed blood samples from 520 individuals and used machine learning to identify a diagnostic panel of three proteins — C1QA, CLUS, and ApoB — whose shape alterations accurately distinguish between healthy individuals, those with mild cognitive impairment, and Alzheimer's patients.

Led by Dr. John Yates with NIH funding through the National Institute on Aging, the team used mass spectrometry to detect subtle folding differences invisible to standard blood tests. The approach not only diagnoses disease status but tracks progression over time. Notably, the study also uncovered sex-specific differences in protein structural patterns linked to neuropsychiatric symptoms, suggesting biological differences between males and females in how Alzheimer's manifests.

The findings establish an entirely new class of biomarkers that could enable earlier, cheaper diagnosis through a routine blood draw — a major step toward catching the disease before significant cognitive decline.