When, where and how does Alzheimer's disease start?

This article is a plain English adaptation of an article that appeared in The Lancet Neurology journal, written by Professor Nick Fox.

Professor Nick Fox is a researcher at the Institute of Psychiatry, University College London, and is co-chair of Alzheimer’s Society’s expert Research Advisory Committee. His previous research has included pioneering work in the development of magnetic resonance imaging (MRI) brain scanning for the diagnosis of dementia.

Biomarkers, either in the blood, cerebral spinal fluid (CSF) or related to brain imaging, could be important for our understanding of Alzheimer's disease, when it begins and how it progresses. They have the potential to tell us the order in which changes occur, and how changes can cause others. This is especially relevant now, as many trials are pointing towards earlier intervention with new drugs as a way to achieve better outcomes with new Alzheimer’s disease drugs.

A recent study, published in Lancet Neurology, looked at young adults aged 18–26 from a large family in Columbia that carry a mutation on the PSEN1 gene. This mutation results in the development of early-onset Alzheimer’s disease, usually in the person’s 40s, and is dominantly inherited – one copy will cause the disease, so the child of a parent with the mutation has a 50 per cent chance of inheriting the mutated gene and so developing Alzheimer’s disease themselves.

The study looked at the difference between those young adults with and without this mutation, comparing findings from blood plasma and CSF.

The amyloid-beta peptide that clumps together to form amyloid plaques was found at a higher concentration in the plasma and CSF of the gene carriers than the non-carriers, even this long before symptoms are likely to appear.

Amyloid-beta levels have been found to decrease in CSF of people with Alzheimer’s disease as amyloid plaques form, which may be a helpful diagnostic marker; it is thought that the amyloid-beta levels decrease as they begin to form plaques, leaving less amyloid-beta 'free' to enter the CSF. But why did the people with the genetic mutations have increased levels, many years before symptoms developed?

This could be related to the type of amyloid-beta produced – the type that was elevated in this study, and is reduced in people with amyloid plaques in their brain, is 42 amino acids long – and so is called amyloid-beta 42 (Aβ1-42).

Laboratory studies have shown that this type of amyloid-beta is increased relative to amyloid-beta 40, which is less prone to forming amyloid plaques. The increased concentrations of amyloid-beta 42 in the gene carriers, compared with the non-gene carriers, suggest that the production of amyloid-beta 42 is greater in those with the gene.

A question that this study did not answer is whether the amyloid-beta 42 levels have always been higher in these individuals, or if there was a point when they increased. As amyloid-beta 42 CSF levels are decreased in people with genetic mutations who have plaques in their brain, the levels of amyloid-beta 42 in these participants will fall – but this study does not tell us when.

A larger study with young adults, of the same age and from the same extended family, looked at MRI scans of individuals both with and without this genetic mutation. It found that those with the gene had signs of reduced grey matter volume, and were showing signs of dysfunction in the synapses that connect neurons.

However, as the earlier study showed, the amyloid-beta 42 CSF levels in similar family members had not yet fallen, and other studies have suggested that there are no plaques present within the brain at this stage.

If these changes are not related to the brain’s development in these individuals, this questions what scientists have previously thought about the development of Alzheimer’s disease.

These results suggest that neurodegenerative changes occur a long time before amyloid plaques are formed; the present hypothesis is that the amyloid plaques occur before, and are a cause of, neurodegeneration. In addition, the results suggest that neurodegeneration occurs a lot longer before symptoms occur than any previous studies have shown.

As with any single, relatively small study, these results should be treated with caution; they are a reason to investigate further, and may not be found to be generalisable to people with sporadic, rather than inherited, Alzheimer’s disease.

If, however, changes in the structure of the brain can be seen this far in advance of symptoms, this will have great implications for future investigations of Alzheimer’s disease and the way that we investigate those changes and potential treatments. For those who believe that treatment should be started before amyloid plaques begin to develop in order to have the best chance of success, this will be made incredibly difficult by the findings that these changes are already present in young adults.

It may be that findings such as these open up the potential for longer-term monitoring and measurements of people during a long phase of the development of Alzheimer’s disease before symptoms appear. The balance for future presymptomatic studies, which will be challenging in any case, is between the potential benefits of treating people very, very early in the disease process – i.e. many years before symptoms – and the greater feasibility of studying people nearer to when their symptoms begin. This balance has great implications for future trials of treatments for Alzheimer’s disease.  

The original article reference is: Fox, N. The Lancet Neurology: 11(12); 1017-1018 (PMID: 23137950)