The test identifies 10 chemicals in the blood associated with the disease two to three years before symptoms start, but it might be able to predict Alzheimer’s decades earlier.
Globally, 35 million people are living with Alzheimer’s. It is characterised by a toxic build up of amyloid and tau proteins in the brain, which destroys the neurons.
Several blood tests can diagnose the disease, but until now, none has had the sensitivity to predict its onset.
Howard Federoff at Georgetown University in Washington DC and his colleagues studied 525 people aged 70 and over for five years. The group showed no signs of mental impairment at the start of the study.
Each year, the team performed a detailed cognitive examination and took blood samples from all the participants. During this time, 28 people developed Alzheimer’s or mild cognitive impairment, thought to be the earliest noticeable sign of dementia, including Alzheimer’s disease.
An analysis of the participants’ blood highlighted 10 metabolites that were depleted in those with mild cognitive impairment who went on to get Alzheimer’s compared with those who didn’t.
In subsequent trials, the team showed these chemicals could predict who would go on to get Alzheimer’s within the next three years with up to 96 percent accuracy.
The 10 metabolites play a key role in supporting cell membranes, maintaining neurons or sustaining energy processes.
“We think the decrease in these chemicals reflects the breakdown of neural populations in the brain,” says team member Mark Mapstone at the University of Rochester Medical Center in New York.
Once verified in a larger group, the test should provide a cheap and quick way of predicting Alzheimer’s. Mapstone says that it may even be able to predict the disease much earlier, because the brain changes associated with Alzheimer’s begin many years before symptoms occur.
“These metabolic changes might occur 10 or 20 years earlier – that would give us a real head start on predicting the disease,” he says.
The team is hoping to investigate this by looking back at other dementia studies in which blood has been taken over decades and seeing whether the chemical changes can be detected that early, says Federoff.
The group also analysed the full genome sequence of all of the participants in the study. That work has yet to be published, but Federoff says the changes in genes over the five years of the study are even more powerful than the metabolites at predicting who will develop dementia.
“The gene changes are linked to the metabolite changes, so we’re hoping to put all this together to provide a more complete description of the underlying pathology of the disease,” he says.
“What’s most exciting is that we know the function of all the affected genes so if we can intercept these changes, they might make good candidates for new drugs.”
But with no treatments available, would anyone want to take these tests?
Mapstone says yes. “In my experience, the majority of people are very interested to know whether they will get Alzheimer’s. They believe that knowledge is power – particularly when it comes to your own health. We may not have any therapy yet but there are things we can do – we can get our financial and legal affairs in order, plan for future care, and inform family members.”
If the test could predict the disease 20 years before symptoms appear, the implications are huge, he says. “Imagine what you would do in your early 40s to slow the onset of the disease. You could eat the right foods, avoid head trauma or do more exercise.”
“In the short term, I think some people would want to know and some wouldn’t,” says Tracy Young-Pearse, a neurologist at Harvard Medical School. However, if treatments are developed that are only effective before neurons have started dying in large numbers, then it will be an easy decision to choose to take the blood test, she says.
Meanwhile, the new test will be valuable for drug discovery efforts, she says. Years of failed drugs trials have shown that you have to catch the disease early to have any influence.
Three studies starting this year hope to do just that. One will test anti-amyloid drugs on healthy people with a rare mutation that gives them early onset Alzheimer’s by age 45 (see “Testing a drug for the memory curse”).
The second will take advantage of a chemical developed last year that can be injected into the body and which accumulates in tau tangles. It will allow researchers to track the progression of tau in the living brain.
A third trial will investigate whether anti-amyloid drugs can prevent Alzheimer’s in older people who don’t yet have memory problems but do have amyloid building up in their brain.
“If an even earlier pre-clinical population could be identified with this blood test, it could be game changing,” says Young-Pearse.