Skip to main content

Early Alzheimer's brain changes studied

Thursday 23 December 2010

Scientists “may have found a way to check for Alzheimer's years before symptoms appear”, BBC News has reported.

The news is based on research that performed brain scans and protein tests on a group of elderly patients with no cognitive impairments, examining whether the results were linked to later brain changes. The research specifically looked at levels of amyloid beta in the patients’ cerebral spinal fluid (CSF). Amyloid beta is implicated in causing brain cell death in Alzheimer’s disease. Participants with lower levels of amyloid beta in their CSF tended to show greater brain cell loss over one year than those with higher levels of this protein.

The study authors point out that this research is preliminary, and that they did not follow people in the long term. They now want to determine whether people with lower levels of CSF amyloid beta have a higher risk of going on to develop Alzheimer’s. Devising methods to detect the disease early would be hugely beneficial in developing drugs that might halt the disease before it progresses. Importantly, though, the researchers believe that this test is not ready or appropriate for use as a screening test, in part because of the lack of appropriate treatments at this time.

Where did the story come from?

The study was carried out by researchers from University College London and was funded by the US National Institutes of Health, National Institute on Aging and National Institute of Biochemical Imaging and Bioengineering. It also received contributions from various pharmaceutical companies. The study was published in the peer-reviewed medical journal Annals of Neurology.

The research was covered well by the Daily Mail and the BBC News.

What kind of research was this?

This was a cross-sectional study that looked at brain magnetic resonance imaging (MRI) scans and proteins in the cerebral spinal fluid (CSF) of participants without dementia to see whether it was possible to detect who was at risk of developing Alzheimer’s disease.

The researchers were specifically interested in the levels of a protein called amyloid beta in the CSF. Amyloid beta is a protein routinely produced by the brain, but in people with Alzheimer’s disease it tends to accumulate within the brain. These accumulations of amyloid beta are thought to cause the brain cell death seen in Alzheimer’s disease. Accumulations of amyloid beta in the brain are thought to result in lower concentrations of amyloid beta circulating in CSF.

Usually, Alzheimer’s is diagnosed when individuals have noticeable cognitive impairment. However, the disease process may be quite advanced before people notice such impairments, consult their doctor and get diagnosed. By the time people are diagnosed many brain cells may already have died. In order to develop treatments that slow or halt the disease, researchers have been trying to find ways to diagnose Alzheimer’s very early on, before extensive death of brain cells and before people have clinical symptoms of the disease.

The researchers in this study wanted to see whether levels of amyloid beta in the CSF were linked to early brain scan changes associated with Alzheimer’s disease. They compared brain size using MRI (as brain cells die, the brain gets smaller) in non-demented patients who had either low or high levels of amyloid beta in their CSF.

What did the research involve?

The participants in this study were part of the Alzheimer’s Disease Neuroimaging Initiative, which is a longitudinal study that follows up people with Alzheimer’s disease and people with normal cognition by conducting serial MRI brain scans.

The researchers selected controls (people without Alzheimer’s or other dementias) who had given a CSF sample and had a brain scan taken at baseline (the start of the study), plus a follow-up brain scan one year later. They selected 105 participants. The participants had been assessed at baseline using a number of dementia tests and had been shown to have normal cognition.

The amount of amyloid beta in their CSF was assessed alongside the concentration of other proteins implicated in Alzheimer’s disease. The participants were grouped into people with low levels of amyloid beta in their CSF (the NC-low group) and people with high levels (the NC-high group). The average age of the 40 people in the NC-low group was 76 years, the average age of the 65 people in the NC-high group was 75 years.

The researchers also had information on the participants’ genetics. They were interested in what variant of the gene APOE the participants possessed, as different variants of this gene have been shown to be associated with a higher risk of developing Alzheimer’s. 

They used two brain images taken using MRI to calculate how the volume of the brain had changed in size between baseline and the follow-up scan a year later.

What were the basic results?

At the start of the study there were no differences between the NC-low group and the NC-high group in terms of brain volume or cognitive scores in all but one test.

The NC-low group had higher concentrations of the protein tau in their CSF (p=0.005). The NC-low group were also more likely to have the APOE4 variant form of the APOE gene, which is thought to be associated with a higher risk of developing Alzheimer’s (p<0.001). However, the variant of APOE that a person had was not correlated with the amount of brain loss.

The NC-low group displayed greater brain loss than the NC-high group. Over one year they lost 9.3ml of brain compared with 4.4ml of brain lost by the NC-high group (p<0.001). The amount of amyloid beta in the CSF at baseline in the NC-low group was strongly correlated with the rate of brain loss, so that lower amyloid beta was associated with greater brain loss over one year.

Subsequent follow-up of the control participants over three years revealed that, to date, five people have developed a mild cognitive impairment (which is a diagnosis to describe cognitive impairments that are worse than expected for an individual’s age or education, but mild enough not to interfere with the person’s ability to do their daily activities). One person has developed Alzheimer’s disease. Of these people, four had been classified as NC-low and one had borderline amyloid beta levels between the high and low groups.

How did the researchers interpret the results?

The researchers said that CSF amyloid beta levels in the NC-low group were similar to post-mortem samples that were taken from people who had died with Alzheimer’s disease. They said that “the control group with amyloid beta levels within the Alzheimer’s disease range had significantly higher rates of whole brain [loss] over the following year compared to those with higher CSF amyloid beta levels”.

They said that their data were consistent with the hypothesis that cognitively normal individuals with low CSF amyloid may not only be at higher risk of developing Alzheimer’s disease, but may “already be some way down the pathogenic pathway” (meaning the disease process has already started).


This is a useful study that has demonstrated that CSF amyloid beta could be associated with brain cell loss in the healthy elderly. However, as the researchers point out, they cannot say that these measurements can be used to determine whether someone will go on to develop Alzheimer’s or not.

Further research following a larger group of people over time would be needed to determine whether this is the case. Also:

  • The accuracy of the test (its sensitivity, specificity and predictive value) was not determined in this research.
  • The researchers also pointed out that the preliminary nature of this study means it did not include a large group of controls.
  • Those controls it did include had a relatively high level of amyloid protein, and this may not reflect the proportion of people who have such amyloid levels in the whole population.
  • To assess amyloid levels the researchers tested the participants’ spinal fluid, which had been extracted using lumbar punctures. This procedure is invasive, and unlikely to feature in routine screening tests due to the risks involved.

Importantly, the researchers are not recommending that this test is ready or appropriate for use as a screening test yet. Studies of diagnostic accuracy and better treatments for the condition are needed before it could be widely used to screen for Alzheimer’s.

Analysis by Bazian
Edited by NHS Website

Links to the headlines

New test combination could spot Alzheimer's early

The Daily Telegraph, 22 December 2010

Scientists discover new test that detects Alzheimer's before it takes hold

Daily Mail, 22 December 2010

Early Alzheimer's test 'possible'

BBC News, 22 December 2010

Links to the science

Schott JM, Bartlett JW, Fox NC et al.

Increased brain atrophy rates in cognitively normal older adults with low cerebrospinal fluid Aβ1-42

Annals of Neurology, December 22 2010 [First published online]