A Blood Test That Sees Alzheimer's Coming, Years Early

Keypoints:

• Researchers tracked healthy older adults instead of patients who already had memory problems
• A blood test measuring p-tau217 successfully identified people at higher risk of developing Alzheimer's later
• The biggest benefit may be earlier treatment

Most Alzheimer's research works backward. Scientists study people who already have symptoms and try to understand what went wrong. A study presented this month by researchers at Mass General Brigham in Boston, and published simultaneously in JAMA, took the opposite approach - it looked at people with no symptoms at all, and tried to predict who would develop them.

The tool is a blood test that measures p-tau217, a protein fragment tied to the buildup of amyloid plaque in the brain, one of Alzheimer's defining features. Researchers led by Rachel Buckley, a neuroscientist at the Mass General Brigham Neuroscience Institute, tracked older adults who showed no signs of cognitive decline at the start of the study. 

The finding is, people with higher p-tau217 levels were significantly more likely to develop measurable cognitive decline in the years that followed, especially when paired with elevated amyloid on a PET scan.

That combination matters significantly. A blood test alone is cheap and easy to administer in a regular doctor's office. A PET scan is expensive and not something most people get without a reason to suspect a problem already. Pairing the two gives doctors something closer to a cholesterol reading for the brain, a number that flags risk long before there's anything a patient would notice or report.

The practical use case isn't diagnosis. It's recruitment. Alzheimer's drug trials have long struggled with a basic problem: by the time someone qualifies for a study because they're showing symptoms, a meaningful amount of brain damage has often already happened. A test that flags high-risk, symptom-free adults years in advance means trials could start earlier, testing whether prevention works instead of only testing whether treatment slows a disease that's already underway.

This builds on a separate but related finding presented at the same conference, the Alzheimer's Association International Conference in London: in a real-world study of more than 1,300 patients, primary care doctors who had access to blood test results diagnosed Alzheimer's almost as accurately as specialists, 93 percent versus 94 percent. 

Before seeing the blood test, family doctors were right about 65 percent of the time. That gap, between what a family doctor can catch and what a specialist catches, is exactly the kind of access problem a cheap, scalable blood test is built to close.

None of this means a cure is closer. What it means is that the years-long guessing game around who's actually at risk may be getting shorter, and that matters for anyone whose family has watched Alzheimer's take longer to diagnose than it should have.