The noninvasive test was developed to detect the presence of the toxic protein amyloid beta, known to be present in people affected by the disease, and did so with 90% accuracy, according to the study, published Wednesday in the journal Nature
The team trialed the test on 121 patients from Japan and 252 from Australia with varying levels of health, ranging from healthy to mild cognitive impairments or Alzheimer's disease.
The idea behind the test is to predict the presence and levels of amyloid beta in a person's brain. Buildup of the abnormal protein in the brain is one sign of Alzheimer's disease, the most common form of dementia.
Other teams have developed similar tests to detect Alzheimer's, with one 2017 study
proving to have 86% sensitivity and specificity.
"This test is at least as good as current brain scan techniques and far surpasses existing blood tests," said Colin Masters, professor of dementia research at Melbourne's Florey Institute of Neuroscience and Mental Health, who led the study.
But the researchers cautioned that they were still far from practical clinical application.
The onset of Alzheimer's can come as early as 30 years before the patient experiences any symptoms, such as memory loss.
There is no concrete test to determine early onset of dementia, and patients with cognitive impairment are diagnosed based on a careful evaluation using brain scans and concise mental testing, according to the Alzheimer's Society UK.
The study used blood samples from patients to detect levels of the toxic protein in the blood and in turn predict how much of it is in the brain.
There is no treatment available for Alzheimer's disease, but early detection can speed access to therapy.
"Progress in developing new therapeutic strategies for Alzheimer's disease has been disappointingly slow. None of the three drugs currently on the market treat the underlying disease," Masters said in a statement.
The team highlights that more research needs to be conducted across a wider population to make the results more conclusive -- and applicable clinically.
"This is not a blood test for dementia that people who are worried about their memory and concentration should be asking their doctors about," said Rob Howard, professor of old age psychiatry at University College London, who was not involved in the study. "Not everyone with amyloid in their brains will turn out to have dementia, and not everyone who has dementia will be found to have amyloid in their brains."
Alzheimer's disease affects over 5 million people in the US
, where the estimated cost of caring for Alzheimer patients in 2017 was $259 billion
In the UK, there are more than 520,000 people affected by the disease, costing £26.3 billion, according to the Alzheimer's Society.
Though the new study provides a breakthrough in research, it is performed only in laboratories in Japan, where it will take up to a year to conform it to a level suitable to distribute to other laboratories and apply it to clinical trials.
"We need to see whether the test works in a larger population, but it has the potential both to speed up clinical trials and help people in the very early stages of Alzheimer's to access new treatments if and when they become available," said Doug Brown, policy and research officer at the Alzheimer's Society UK.
Carol Routledge, director of research at Alzheimer's Research UK, said that "further work with more people will need to build on this study to understand how well this approach could predict those who will go on to develop symptoms of Alzheimer's in the future."