New dementia predictors are over 80% accurate but is screening a good idea?

5 minute read


Biomarkers have been identified which can forecast trouble ahead. But Australian experts are divided.


Dementia can be accurately predicted nearly 15 years before diagnosis, UK researchers say.

And while Australian experts agree it is promising research, they are divided on the role it should play in screening and early detection.

According to the researchers, who published their findings in Nature Ageing, there are four proteins present in unusually high levels in people who later go on to be diagnosed with all-cause dementia, vascular dementia, or Alzheimer’s disease (AD).

Using UK biobank data research, they examined blood samples from more than 52,000 adults without dementia that were collected and frozen between 2007 and 2010. These samples were analysed in 2021 and 2022 and included samples from the 1400 participants who had gone on to develop dementia.

The researchers are hoping to use their findings to develop a screening kit for use in the NHS which can identify high levels of these four proteins (GFAP, NEFL, GDF15 and LTBP2) to predict the development of dementia.

This protein profile, along with risk factors such as age, sex, education, and genetic susceptibility, allowed predictions to be made with an estimated 90% accuracy, they wrote.

Professor Ashley Bush, from the Department of Neuroscience and Psychiatry at the University of Melbourne, told TMR that disease-modifying interventions were most impactful when delivered at the first sign of dementia.

“[This] study gives us valuable information about plasma proteins and highlight in particular the value of GFAP elevation as a predictor of dementia,” said Professor Bush who is also Co-Director for Biomarker Development of The Australian Imaging, Biomarkers and Lifestyle Study of Ageing (AIBL).

“GFAP begins to elevate 10 years before the onset of the diagnosis, so this could be a very valuable blood test for detecting people whose brains are in trouble and might benefit from an early intervention.”

Acknowledging the fact that there was currently no cure for dementia, Professor Bush explained that the value of these tests was in secondary prevention.

“Secondary prevention is very valuable with such a costly set of diseases as the dementias. By the time a dementia is diagnosed, a lot of the brain damage is already irreversible,” he said.

Professor Bush also said that the identification of these four proteins was a rich resource for further research.

“GFAP elevation tells us that one of the protective cells in the brain, the astrocytes, are being activated early in the disease process. Beyond that, quite a few hypotheses can be set up to test based on the data provided,” he told TMR.

However, Professor Permidner Sachdev, clinical director of the Neuropsychiatric Institute and Prince of Wales Hospital, disagreed.

Professor Sachdev, co-director of the Centre for Healthy Brain Ageing, told TMR that while the figures sound impressive, at this stage he would not recommend the use of these markers to predict dementia risk.

Knowing that you have a higher risk of developing dementia through this screening is not helpful as there is no data to support earlier treatment, he said.

“I don’t think this is something to be translated into clinical practice at this stage. It’s promising research but the issue is what to do with this knowledge. This is unclear at this point,” he said.

“Where the field is at is that people with mild cognitive complaints or early deficits should present for clinical assessment of risk factors and we can advise from there.”

Professor Sachdev explained that the new drugs that were approved in the US and were currently being assessed for approval in Australia were intended for people with mild symptoms and who had undergone CSF screening and amyloid scans.

These treatments have not been tested on people who were asymptomatic and were not for use in people years prior to the development of amyloid. They also carried the risk of serious side effects, he said.

“And at what point are you going to start giving them anti-amyloid treatment?” he said.

“The threshold for treatment now is to be amyloid positive – what is the threshold using these genetic markers?”

He also raised issues of how many people would be screened, how often, and who would pay for it.

“We also don’t know at this point when these proteins start rising and what other factors affect this,” said Professor Sachdev.

“For example, what is the role of kidney disease or cardiac disease and how do they influence these markers? How does ethnicity affect this? This may work in a European population, but does it work in a Chinese population as well?

“If it were me, I would wait for some subjective decline in cognition or memory, especially if others are noticing it. Dementia doesn’t happen overnight and there is plenty of time to act, to be assessed, to put affairs in order and to take other measures in terms of your health and lifestyle.

“I don’t think it helps, it just causes extra anxiety and there’s no way forward. Prevention is for everybody, whether you have these genetic markers or not.”

Nature Aging, online February 12

End of content

No more pages to load

Log In Register ×