New Test Detects Alzheimer’s by Analyzing Altered Protein Shapes in Blood

Misfolded proteins gradually accumulate in the brain, disrupting normal cellular processes. Current diagnostic blood tests primarily measure the levels of certain proteins, but these markers may not fully capture early molecular changes associated with disease progression. Researchers have now developed a blood test that detects Alzheimer’s disease by analyzing the physical shapes of proteins circulating in the bloodstream.

In a collaborative study led by researchers at The Scripps Research Institute (La Jolla, CA, USA), the team used a chemical tagging technique combined with mass spectrometry to examine how proteins fold and interact in blood plasma samples. This method identifies which portions of a protein are exposed or hidden within its three-dimensional structure. When proteins misfold—a hallmark of neurodegenerative diseases—these structural features shift. By tracking these changes, researchers created a detailed molecular profile associated with different stages of cognitive decline.

The researchers analyzed blood samples from 520 volunteers, including healthy individuals, patients with mild cognitive impairment, and those diagnosed with Alzheimer’s disease. Using machine learning models, the team identified structural patterns in three proteins—C1QA, CLUS, and ApoB—that were strongly associated with disease progression. The deep learning model distinguished healthy aging, mild cognitive impairment, and Alzheimer’s disease with an accuracy of 83.44%. The structural approach also outperformed traditional models that measure only protein quantities. In a smaller follow-up group of 50 participants, the model correctly tracked disease progression 86% of the time.

The findings, published in Nature Aging, suggest that structural protein signatures may provide earlier and more detailed insight into Alzheimer’s disease than existing blood-based tests. Researchers also observed links between protein structure changes, genetic risk factors such as apolipoprotein E variants, and differences in neuropsychiatric symptoms between men and women. Further studies involving larger patient populations and longer follow-up periods will be needed to validate the test. If confirmed, this structural biomarker approach could help clinicians detect Alzheimer’s earlier, monitor disease progression, and evaluate responses to emerging therapies.

“With this work, we established a potential new biomarker panel that reveals structural disruptions in proteins linked to Alzheimer’s disease that are invisible to traditional approaches,” said John R. Yates III, a professor of Integrative Structural and Computational Biology at The Scripps Research Institute. “This approach accurately distinguishes stages of the disease, meaning that it could help enable earlier diagnosis.”

Related Links:
The Scripps Research Institute