Researchers move closer to a blood test for Alzheimer’s, replacing costly PET scans

Researchers are inching closer to a blood-based diagnostic test to detect Alzheimer’s disease before symptoms arise, according to a new exploratory study published Tuesday. And it may impact the future utility of positron emission tomography. 

A team led by scientists at the Washington University School of Medicine in St. Louis took this big step after comparing blood samples and brain scans from a small group of Alzheimer’s study participants. They found that specific protein levels in the blood rise as amyloid plaques begin to form in the brain.

And this could be a game-changer for identifying the sticky amyloid buildup that begins to develop up to two decades before symptoms arise, senior author Randall J. Bateman, MD, noted July 28.

“The finding of a unique tau species that is closely linked to changes caused by amyloid plaques will help to identify and predict people who have or will likely develop Alzheimer's disease," Bateman, a distinguished professor of neurology at Wash U Medicine, said in a statement. "This will greatly accelerate research studies, including finding new treatments, as well as improving diagnosis in the clinic with a simple blood test."

As it stands, PET scans are typically used to spot patients with amyloid plaques, but the modality is too costly and time-consuming for widespread screening strategies, the authors noted.

Researchers have been working for decades to develop an easy and affordable alternative that would spot people in the preclinical stages of the disease. The idea being, once drugs are available, these patients would never develop any symptoms of the memory-destroying affliction. 

Based on their discovery that tau fragments (p-tau-217) accumulate in the cerebrospinal fluid of Alzheimer’s patients before symptoms, the researchers hypothesized that this may also be the case in patients’ blood.

Bateman and Nicolas Barthélemy, PhD, who is also with Wash U Medicine’s neurology department, compared blood samples and brain scans from 34 participants. Of that total, 19 had no brain amyloid, five had buildup with no symptoms, and 10 had both. Using mass spectrometry, they found phosphorylated tau 217—a form of tau—levels correlated with amyloid plaques in the brain.

In fact, those with amyloid had two-to-three-times more of the protein in their blood samples compared to those without. This was true even in those with no cognitive symptoms.

Wanting to verify their results, the team copied the study in another group of 92 patients: 42 with no amyloid, 20 asymptomatic with amyloid, and 30 with each.

Their findings confirmed the original investigation: tau blood levels correlated to amyloid brain deposition in more than 90% of cases. And the method was 86% accurate in distinguishing people with no cognitive symptoms, but in the early stage of the disease from healthy patients. 

“This is just an exploratory study, but we think phosphorylated tau 217 is a promising target for an early diagnostic test," Barthélemy said Tuesday. "There was a large difference between the amyloid-positive and amyloid-negative groups, even amongst people who were cognitively normal. Once we improve the way we are preparing and concentrating the sample, we will be a step closer to developing a tau-based blood test that can identify people at risk of developing Alzheimer's dementia before symptoms arise."

Read more about the project published in the Journal of Experimental Medicine here.