Skip to main content

Could spinal fluid test give early Alzheimer’s warning?

Friday 21 March 2014

“A new Alzheimer’s detection test, which can diagnose the presence of the disease decades before symptoms appear, could be available to patients in just three years,” reports the Daily Express.

Sadly, this claim was not actually proven; what actually happened is that researchers developed a test that can detect low levels of an abnormal form of the protein amyloid beta. This protein accumulates in the brains of people with Alzheimer’s in the form of “plaques”.

Researchers tested 50 people diagnosed with probable Alzheimer’s disease, as well as 76 people with other brain conditions, for the abnormal protein.

They test was found to be effective. However, it is unclear how effective the test would be if used on people without dementia-like symptoms.

Due to the test’s invasive nature – which involved using a lumbar puncture, where a large needle is used to drain fluid out of your spine – it is highly unlikely that the results will lead to a screening programme for Alzheimer’s.

The researchers would like to use blood for their tests (as it is a simpler and more acceptable form of testing); however, it is unclear if this would work or be useful in medical practice.

Where did the story come from?

The study was carried out by researchers from the University of Texas Medical School and research centres in Italy. It was funded by the Alzheimer’s Association, CART Foundation, Mitchell Foundation, Italian Ministry of Health and MIUR.

One of the authors declared they were an inventor on several patents relating to the technique described in the study, as well as being the founder of Amprion Inc., a biotechnology company developing the technique for the diagnosis of neurodegenerative diseases.

The study was published in the peer-reviewed scientific journal Cell and has been made available on an open access basis, meaning it is free to read online.

The Daily Express’ claim that a new Alzheimer’s detection test that can “diagnose the presence of the disease decades before symptoms appear could be available to patients in just three years” is unsubstantiated. The test has only been used in people already given a probable diagnosis of Alzheimer’s.

While theoretically it may be possible to detect early changes before symptoms occur, this has not been tested or proven. The test also currently involves taking a sample of spinal fluid, and it is doubtful that  such a procedure would be willingly carried out on individuals who had no symptoms of Alzheimer’s. With such a sketchy evidence base, it is safe to assume that most would not want a large hollow needle inserted into the base of their spine for a test.

What kind of research was this?

This was a diagnostic study looking at developing a new method for diagnosing Alzheimer’s using cerebrospinal fluid (CSF). Spinal fluid surrounds and supports the brain and spinal cord.

Alzheimer’s disease is the most common cause of dementia. Detecting dementia currently requires a person to present with symptoms of the condition, after which the person will usually have a range of physical and mental assessments. If other physical causes for these symptoms are ruled out, a diagnosis of probable dementia may be given.

However, doctors can only give a definite diagnosis by examining a person’s brain after their death, to look for the characteristic effects of Alzheimer’s on brain tissue. 

These include the typical “amyloid plaques” which are made of deposits of the protein amyloid beta.

There is currently no cure for Alzheimer’s, and treatments may slow down the condition's progression, but cannot stop or reverse it.

One of the reasons may be that the brain has limited ability to repair itself, meaning that when symptoms are noticed, the damage cannot be reversed.

Researchers hope that if they find a way to detect the disease early, they could develop treatments to stop it progressing. The formation of abnormally large aggregates of amyloid beta in the brain is thought to start long before the onset of Alzheimer’s. If these formations could be detected, it could be an early sign of the illness.

What did the research involve?

The researchers wanted to see if they could develop a way of identifying people with Alzheimer’s by detecting abnormal (misfolded) amyloid beta in the spinal fluid.

The researchers were using a method they developed called protein misfolding cyclic amplification (PMCA). This method makes use of the fact that even small amounts of abnormal forms of proteins like the amyloid beta protein can speed up the aggregation (clumping together) of the protein. Amyloid beta is mainly found in the brain, but some of it moves into the fluid surrounding the brain and spinal cord (CSF).

They first tested it in the lab to make sure their method could detect low levels of misfolded amyloid beta. They then analysed CSF from 50 people with Alzheimer’s, 37 people who had other degenerative brain diseases (including other forms of dementia) and 39 people who were affected by non-degenerative brain diseases, but had normal cognitive function. They were investigating whether they could tell these groups of people apart based on the test results. The researchers testing the samples did not know which samples belonged to which people, ensuring the results could not be interpreted in a biased way.

What were the basic results?

The researchers found that their PCMA technique was effective at detecting low levels of abnormal amyloid beta protein in laboratory tests.

The researchers also found their test performed differently on CSF fluid samples from individuals with Alzheimer’s, other degenerative brain diseases and those with normal cognitive function. The abnormal amyloid beta in spinal fluid from people with Alzheimer’s sped up the aggregation of more amyloid beta protein during the testing process.

Based on their results for age-matched Alzheimer’s and all of the control samples (people with other kinds of brain diseases):

  • the test correctly picked up 90% of people with Alzheimer’s – meaning that 10% of those with the disease would be missed (false negatives)
  • the test correctly identified 92% of those without the disease – meaning that 8% of people who did not have Alzheimer’s would test positive (false positives)
  • of those who tested positive, 88% did have Alzheimer’s (so about 1 in 10 people who tested positive would not in fact have the illness)
  • of those who tested negative, 93% did not have Alzheimer’s (so under 1 in 10 of people who tested negative would actually have the illness)

They say these results are better than those others have achieved by testing a different set of markers in spinal fluid.

The researchers note that these results were all from samples collected in three centres. Their technique did not work with samples collected at a fourth centre (they were “not amenable to assay”). They suspected that an aspect of the sample collection method may have affected their test, and they are looking into this further.

How did the researchers interpret the results?

The researchers concluded that their findings provide proof of principle for developing a highly sensitive and specific test for Alzheimer’s diagnosis.


The current research suggests it may be possible to identify those with Alzheimer’s using a biochemical test carried out on a spinal fluid sample. However, it is important to note that:

  • the development of this test is at a very early stage, and the study design used here is not ideal for assessing the test’s diagnostic accuracy. Population-based testing may follow, allowing researchers to better assess the accuracy and, more importantly, how many people could be given an incorrect diagnosis of dementia based on this test. However, development still needs to be done before population-based testing could be considered
  • it included samples from a small number of people diagnosed with different neurological conditions. The people in the trial with Alzheimer’s had not had post-mortem brain examinations, so their diagnosis was confirmed using just clinical testing
  • the researchers could not get the test to work on one set of samples. To be used in real life medical practice, it would have to be shown to reproducibly get good results
  • the proportion of the positive and negative tests that are correct is affected by the number of people being tested who actually have the illness. The test performed well on a set of samples where around a third of participants had Alzheimer’s. These results would be different if fewer people being tested had the illness, and the proportion of positive tests for people without the disease would be higher

There are also other questions surrounding how the test might be used and how useful it would be in medical practice. This test currently uses spinal fluid. To get this requires an invasive procedure involving placing a needle into the spine and can have side effects. 

Doctors would not want to use a method like this unless they were fairly certain a person has Alzheimer’s, which essentially negates the usefulness of the current test.

This research should be regarded as a work in progress. It may lead to an effective blood test, which would be a lot more useful in screening people for Alzheimer’s disease; however, whether this becomes a reality is unclear.

Analysis by Bazian
Edited by NHS Website

Links to the headlines

Spine test to give an early warning of Alzheimer’s

Daily Express, 21 March 2014

Early detection of Alzheimer’s may be possible through spinal fluid test

Fox News, 20 March 2014

Links to the science

Salvadores N, Shahnawaz M, Scarpini E, et al.

Detection of Misfolded Aβ Oligomers for Sensitive Biochemical Diagnosis of Alzheimer’s Disease

Cell Reports. Published online March 20 2014