Skip to Main Content

Can we predict the risk of dementia decades before onset? In a study published in Science Translational Medicine on Wednesday, neuroscientists reported a startling clue.

The neuroscience researchers, who are from the National Institute on Aging and Johns Hopkins University School of Medicine, decided to peer into more than 4,000 proteins found in plasma, which is blood with the blood cells removed. They examined the link between the proteins and dementia risk in nearly 11,000 adults, aged between 45 to 65. After 25 years, they found that 32 proteins may be key to the early onset of dementia, suggesting that they could be used to predict the disease earlier in life.

advertisement

“The study is robust and impressive,” said Nicholas Seyfried, a professor of biochemistry and neurology at Emory University School of Medicine. “It’s really remarkable.” Seyfried, who was not involved in the research, said it offers significant insight into one of the biggest questions facing Alzheimer’s research — a field that’s complex and historically filled with challenges.

Alzheimer’s is a debilitating brain disease that robs patients of their memories and cognitive abilities. It is the most common cause of dementia, also known as loss of memory, in people who are 65 and older. In the U.S., nearly 7 million Americans are living with Alzheimer’s and the disease accounts for up to 80% of dementia cases. Although researchers have recorded scientific breakthroughs, including therapies, early brain changes that underlie dementia remain poorly understood.

The ability to predict dementia years in advance of cognitive decline could offer hope for early intervention against Alzheimer’s disease. “Imagine telling a 45- or 50-year-old that your plasma biomarker profile is suggestive that you would convert to dementia in the next 10 to 15 years — what a profound thing to say,” Seyfried told STAT.

advertisement

Keenan Walker, who co-led the study, said the proteins they identified can help researchers understand the changes that cause dementia. And, in turn, they could be used as biomarkers to predict whether someone can get the disease decades in advance before cognitive decline. Walker and his colleagues will need to do more work before the proteins can be validated and used as predictive markers for dementia in routine clinical testing. But the findings highlight growing interest in discovering the early biological signatures of Alzheimer’s disease — changes to the brain that start to occur even before symptoms appear.

There are already proteins, called amyloid beta and tau, that cause abnormal changes within the brain cells that help neurologists identify the symptoms of Alzheimer’s disease. While research shows that amyloid and tau work closely — like a spark that starts a forest fire and timber that fuels it — and causes healthy neurons to descend into disease state, many changes that happen overtime within the brain, leading to dementia, still remain unknown.

In their study, Walker and his colleagues found that 32 of the more than 4,000 proteins they analyzed were associated with dementia outcome. “We were able to see one by one which proteins predicted new onset dementia,” said Walker, who studies age-related cognitive disorders at the National Institute on Aging.

The study also revealed that the proteins played a role in changes occurring in key cellular functions such as immunity, synaptic function, proteostasis, and extracellular organization. This suggests possible dementia onset in people at risk, decades ahead of characteristic symptoms of Alzheimer’s disease such as neurodegeneration and loss of brain function, said Walker. “Alzheimer’s disease biology is really expressed early on,” he said.

The team further scrutinized the proteins to see whether they were associated with clinically known neurocognitive outcomes. They found 12 of the 32 proteins associated with dementia were  linked with cerebrospinal fluid biomarkers — which give researchers a window into the brain — of Alzheimer’s disease. “The relationship between these proteins and the known biomarkers for Alzheimer’s disease and the integration to the brain changes really strengthens the conclusions,” said Seyfried.

Eight of these candidate protein markers were abnormally expressed in human postmortem brain tissue from patients with Alzheimer’s. But some of the signature proteins most strongly associated with dementia risk, such as GDF15, were not detected in brain tissues. Seyfried said that this could suggest molecular changes below the neck in the peripheral nervous system that may influence people’s risk for developing dementia.

Instead of the use of sophisticated brain imaging techniques to detect the known proteins, amyloid and tau, the new study findings could allow researchers to use protein biomarkers in blood plasma to predict the risk of dementia.

Although the study is compelling, it has some limitations. The researchers recruited only Black and Caucasian participants from communities across the US. Thus, the results may not be representative of non-Black and non-white populations. Seyfried agrees that this is an issue and there’s a need for more research in an extensively ethnically diverse population.

“You can’t question the power of the study,” said Seyfried. “But we need to see if these markers generalize to populations.”

STAT encourages you to share your voice. We welcome your commentary, criticism, and expertise on our subscriber-only platform, STAT+ Connect

To submit a correction request, please visit our Contact Us page.