A simple blood test could soon replace invasive Alzheimer’s disease screenings. Currently, the gold standard methods for detecting the signs of Alzheimer’s in the brain are PET scans, which create 3D images showing amyloid and tau accumulation, and tests of cerebrospinal fluid (CSF), the liquid surrounding the brain and spinal cord. However, PET scans are very expensive and involve radiation exposure, while obtaining CSF requires an invasive spinal tap procedure. Therefore, there has been an intense search for easier and more accessible ways to diagnose Alzheimer’s — a devastating form of dementia that robs people of their memories and cognitive abilities.
In recent years, scientists have developed blood tests for Alzheimer’s that measure specific forms of the tau protein that are altered in the disease. One such blood marker is called phosphorylated-tau217 (p-tau217). Previous studies found that blood levels of p-tau217 correlated with the amounts of amyloid and tau in the brain. But it wasn’t clear if a p-tau217 blood test could perform as well as established CSF tests.
To address this question, researchers from Sweden and the U.S. directly compared a high-precision p-tau217 blood test with CSF tests in over 1,700 individuals. The key advance was measuring the ratio of p-tau217 to non-phosphorylated tau, which helped account for individual differences unrelated to Alzheimer’s.
In the study published in the journal Nature Medicine, they found that the p-tau217 ratio blood test was just as accurate as FDA-approved CSF tests in identifying people with abnormal amyloid levels in the brain, which is a defining feature of Alzheimer’s. In people with mild cognitive impairment or mild dementia — the population likely to benefit from new Alzheimer’s drugs – the p-tau217 blood test had an accuracy around 90 percent, matching the CSF tests.
Even more impressively, the blood test outperformed the CSF tests in detecting people with elevated tau tangles in their brains. Tau tangles, rather than amyloid, are more closely tied to cognitive decline in Alzheimer’s.
“The accuracy of this blood test now enables us to diagnose the presence of Alzheimer’s disease pathology with a single blood sample,” says study co-senior author Dr. Randall J. Bateman, the Charles F. and Joanne Knight Distinguished Professor of Neurology at Washington University, in a media release. “This advance will increase accurate diagnoses for many patients.”
Tweaking the cutoff points for the blood test to a higher level to confirm Alzheimer’s, or a lower level to rule it out, boosted accuracy to around 95 percent for amyloid and tau detection. This two-cutoff approach left only about 1 in 5 people in an “uncertain” middle zone, comparable to CSF testing.
Importantly, the robust performance of the p-tau217 blood test was validated in separate cohorts from two countries, supporting its reliability and generalizability. While the test requires specialized mass spectrometry equipment, it could still greatly expand access to Alzheimer’s diagnostic testing.
“In the near future, this type of blood test will replace the need for costly and less accessible cerebrospinal fluid and PET imaging tests in specialist memory clinics,” explains study co-senior author Dr. Oskar Hansson, a professor of neurology at Lund University. “Next, we need to determine if the Alzheimer’s blood test also works in primary care. This is currently being investigated in Sweden.”
This landmark study demonstrates that a precise p-tau217 blood test can diagnose Alzheimer’s disease and predict brain pathology with similar accuracy as invasive CSF tests and expensive PET scans. While some unknowns remain, such as its performance in more diverse populations, this innovation represents a major step towards making early and definitive Alzheimer’s diagnosis more widely available.
“Imagine a person who is 55 or 60 and has a family history of Alzheimer’s or some high-risk genetic variants,” concludes study co-author Dr. Nicolas Barthélemy, an assistant professor of neurology at Washington University. “It would be really valuable to have an easy way to know whether they have amyloid pathology in their brains. If they do, they could come in, maybe once every two or three years, and get a therapy to clear the amyloid out and then never develop dementia at all. We’re still a few years away from such an approach, but I think that’s the future of Alzheimer’s care, and it depends on presymptomatic diagnosis and treatment.”
