Skip to main content

A blood test for Alzheimer's remains a distant possibility, study suggests

Tuesday 22 January 2019

Researchers have developed a blood test that can predict your chances of getting Alzheimer's, claim several media reports.

But any suggestion of a major breakthrough is a little hasty. The research only involved people with a rare hereditary form of Alzheimer's disease caused by genetic mutations (Dominantly Inherited Alzheimer's Disease).

In these people, researchers could detect raised levels of a specific protein called neurofilament light chain (NfC) in their blood that is created when the structure of previously healthy nerve cells are damaged or destroyed.

Tests suggest that raised levels of NfCs could be detected as much as 16 years before symptoms of Alzheimer's start in people with this genetic form of the disease.

However, we have no idea whether these changes would be seen in the vast majority of people who develop Alzheimer's and who do not have these mutated genes.

There is currently no preventative treatment or cure for Alzheimer's so it is not clear what action you could take if the blood test suggested you're likely to develop the disease. This raises the question of how useful it is to cause anxiety by telling people when they're likely develop Alzheimer's if you can't do anything about it.

Where does the story come from?

The study was conducted by researchers from the German Center for Neurodegenerative Diseases, Tübingen, Germany and various other institutions in Europe, the US and Australia. Funding was provided by the National Institute on Aging and the German Center for Neurodegenerative Diseases. The study was published in the peer-reviewed scientific journal, Nature Medicine.

While the UK media's general reporting of the study was accurate, none of the headlines made clear that the results may only be applicable to people with a rare form of "genetic Alzheimer's".

What kind of research was this?

This was a laboratory study investigating whether it might be possible to detect signs of brain damage in a blood test, and so identify people who could be at risk of developing neurodegenerative conditions like Alzheimer's.

The research is built on the idea that brain changes appear many years before symptoms develop. In the case of Alzheimer's these brain changes involved the build-up of amyloid protein plaques, tau protein tangles in the nerve fibres, and overall thinning of the brain matter.

It's known that amyloid and tau proteins can be detected in the cerebrospinal fluid (CSF) that surrounds the brain and spinal cord but taking invasive lumbar punctures (where CSF is extracted from the base of the spine) from everyone is not practical due to reasons of cost and time, or arguably ethics. However, levels of NfC can be detected in the blood. NfC comes from damaged nerve fibres, and changes in the levels have been linked with brain damage in neurodegenerative diseases like Alzheimer's and others.

So researchers wanted to see whether NfC could be a blood marker for neurodegenerative changes.

What did the research involve?

The researchers made use of biological specimens collected by the Dominantly Inherited Alzheimer Network (DIAN). This network includes data from families who are at strong hereditary risk of Alzheimer's because they carry mutations of the APP or PSEN genes; 2 mutations known to trigger Alzheimer's.

These mutations are associated with onset of Alzheimer's systems at a fairly consistent age of between 30 to 50.

Therefore the researchers should be able to look at NfC levels in the blood or CSF and estimate roughly how far off the person would be from developing symptoms of Alzheimer's.

The researchers had CSF and blood samples for 243 people with APP or PSEN mutations, and 162 controls who did not carry these mutations.

What were the basic results?

The researchers found that NfC levels in the CSF significantly increased in people with the mutations (compared to controls) around 6.8 years before they'd expect symptoms to start.

They confirmed that there was a close link between NfC levels in the blood and CSF, so they focused on the blood samples.

Roughly half of the participants had a few blood samples taken, several years apart. Comparing these samples over time indicated that changes in the NfC were (observed compared to controls) 16.2 years before symptoms start – a whole decade earlier than they'd found with the one-off CSF samples.

Blood levels of NfCs seemed to peak when people were just at the point of starting to develop Alzheimer's symptoms with no change thereafter. They also found that NfC levels were linked with thinning of the brain matter, and with poorer performance on tests designed to assess cognitive function and ability.

How did the researchers interpret the results?

The researchers conclude: "NfCs dynamics in [blood] predict disease progression and brain neurodegeneration at the early pre-symptomatic stages of [hereditary] Alzheimer's disease, which supports its potential utility as a clinically useful biomarker.”

Conclusions

This is an interesting development which furthers our understanding of the development of Alzheimer's disease. It shows how increased levels of NfCs from nerve fibres can be detected years before hereditary Alzheimer's develops, and can be linked with other characteristic disease features.

Whether such a blood test would ever have a place in mainstream clinical practice is a completely different matter.

The main limitation is applicability to the general population. The test looked at how NfCs correlate with disease development in the rare individuals who have a strong predisposition to Alzheimer's due to mutations of the APP or PSEN genes. This accounts for the smallest fraction of people. The Alzheimer's Society estimates that 99 out of 100 cases of Alzheimer's are not inherited. This study can't tell us whether NfC levels precede symptom development in the vast majority of people who will develop Alzheimer's but had no genetic risk factors.

Even if the test was found to apply to the wider population, if there's no preventative treatment that can be given to stop Alzheimer's, then how useful is it? Would it be helpful to tell people how many years they may be from developing Alzheimer's when this could likely just cause psychological harm?

The findings are undoubtedly of interest to understanding the disease development of Alzheimer's. It will also be interesting to explore, as the researchers suggest, whether blood tests might be able to indicate the early development of other neurodegenerative conditions such as Parkinson's. But there currently remains no test available to predict Alzheimer's.

Analysis by Bazian
Edited by NHS Website