More genetic clues to Alzheimer's found

Monday April 4 2011

"Scientists have discovered five gene variants that raise the risk of Alzheimer's disease," reported The Guardian and many other newspapers today.

The news report is based on several large and important “genome-wide association” studies looking for associations between Alzheimer's disease and particular genetic variants. Together they have identified five new genetic variants that increase risk, and confirm some other associations found in previous studies.

Alzheimer's is a complex disease and there are likely to be more variants that contribute to risk. These well-conducted studies are essential first steps in identifying genetic components of disease risk. It will be several more years – an estimated 15 years by one of these researchers – before a test or treatment for Alzheimer’s might be available.

Dr Susanne Sorensen of the Alzheimer’s Society put these findings in context:

“These two robust studies mark an exciting development for scientists hoping to identify a cause and find a cure for Alzheimer's disease. Although these studies will not bring us any closer to being able to predict who might be at greater risk of developing Alzheimer's, they will give scientists clues as to how Alzheimer's might develop. Most importantly, their identification could also lead to the development of new drug treatments in the longer term.”

Where did the story come from?

The news stories are based on two companion studies that were carried out by several international research consortia, including hundreds of researchers from around the world. The first study – a genome-wide association study – was carried out by a consortia called The Alzheimer’s Disease Genetics Consortium (ADGC). The studies were funded by a variety of sources, including the National Institutes of Health in the US, the Alzheimer’s Association, the Wellcome Trust, and the Medical Research Council in the UK.

The studies were published in the peer-reviewed scientific journal Nature Genetics .

The research and findings have been covered well by the media, although some newspapers could have emphasised more that this is preliminary research and that interventions to prevent the disease will be some way off.

What kind of studies are these?

The two studies had similar aims: to identify new genetic variants associated with late-onset Alzheimer’s disease. Late-onset Alzheimer’s defines all those cases which arise in people over the age of 65, for whom there is no known cause, i.e. most cases of Alzheimer’s disease. Early-onset disease is much rarer, and affects younger people. As would be expected, early-onset Alzheimer’s has a different clinical course and different causes from what would be recognised as ‘normal’ – late-onset – Alzheimer’s disease.

Both studies investigated whether there were associations between particular genetic variations and late-onset Alzheimer’s disease. They then validated the findings from these initial studies in more independent samples of people.

Genome-wide association studies are commonly used to determine whether there are genetic differences between people with and without a condition. They involve a large number of genetic sites across the DNA being compared between these groups. This type of study usually involves a number of steps in separate independent samples, in which the comparisons are repeated in an effort to replicate and validate the findings from the initial comparison.

Study one

The first study, conducted by The Alzheimer Disease Genetics Consortium (ADGC), compared the genetic make-up of 8,309 people with late-onset Alzheimer’s disease with 7,366 older people who had been defined as ‘cognitively normal’. The researchers obtained the genetic and clinical data for this large sample of participants by combining the participants from nine separate cohort studies. The researchers then validated their findings by replicating the studies in separate samples. The first sample included 3,531 cases and 3,565 controls. The second sample included 6,992 cases and 24,666 controls.

Study two

The second study was set up to identify new genetic variants that may be associated with Alzheimer’s disease and also to test (in a series of samples) whether the associations found in the ADGC study could be validated. These researchers combined the results from four previous genome-wide association studies, including a total of 6,688 affected individuals (cases) and 13,685 controls. They then tested their findings in a second independent sample of 4,896 cases and 4,903 controls, and in a third comprising of 8,286 cases and 21,258 controls.

Genome-wide association studies are a form of case-control study, comparing the differences between people with a condition and those without it. The results present the likelihood of people with the disease having particular genetic variants.

What were the basic results of the studies?

Study one

The first study confirmed previously known associations on genes called CR1, CLU, BIN1 and PICALM. It also identified four new genetic variants that were more common in people with Alzheimer’s disease. These were variants on genes called MS4A4/MS4A6E, CD2AP, CD33 and EPHA1. These genetic variants were all significantly more common in people with late-onset Alzheimer’s disease.

Study two

The second study confirmed that four of the genetic variants identified in study one were associated with Alzheimer’s disease. They also identified a fifth genetic variant on gene ABCA7. The researchers go on to discuss the possible function of the genes in which these variants are located, highlighting some biologically plausible reasons why their dysfunction may be associated with Alzheimer’s.

How did the researchers interpret the results?

In general, both sets of researchers say that their findings are important and that there are plausible biological reasons why some of the genetic variants they have found may be associated with Alzheimer’s disease. The findings ‘provide new impetus for focused studies aimed at understanding’ Alzheimer’s disease and the way the disease starts and develops.


Genome-wide association studies like these identify genetic variants that contribute to the risk of disease. Diseases such as Alzheimer’s are complex and are likely to have many different causes, both genetic and environmental.

Each of the variants identified here carries a small increase in risk of the disease, and bring the total number of genetic variants associated with Alzheimer’s disease to 10. The more of these variants a person has, the greater their risk of Alzheimer’s disease. The BBC reports one lead researcher as saying, “if the effects of all 10 could be eliminated, the risk of developing the disease would be cut by 60%”.

There are likely to be other contributing genes, which will affect risk to varying degrees. The researchers in the first study say that because of the complexity of the disease, these undiscovered variants may contribute only a little to the risk and that larger studies will be needed to find them.

Overall, these two studies increase what we know about the associations between certain sites in our DNA and Alzheimer’s disease. The findings should be interpreted carefully because these particular genetic variants are not necessarily the cause of Alzheimer’s disease. Instead, they may be lying close to functioning genes that have negative effects. More research is now needed to identify the functional genes themselves.

It is also difficult to anticipate what these results mean for healthy people or for people who already have Alzheimer’s disease. A possible application may be to screen people for the disease, i.e. to profile a healthy person’s DNA to see whether they have these variants and are therefore at greater risk of disease. However, such screening would need careful consideration because:

  • having the variants does not mean a person will definitely have the disease. More work needs to go into determining what thresholds represent high, medium and low risk, particularly as there are now several different variants associated with the disease
  • there may be risks associated with testing people for the presence of these variants
  • there is currently nothing to offer people who are identified as being at potentially high risk to reduce their likelihood of getting the disease. Studies such as these are important initial steps in developing genetic treatments, but these are likely to be a long way off.

In general, these are important findings that will be of interest to the genetic research community. They will be added to the existing bank of knowledge about this disease and will undoubtedly direct further research into the mechanisms of this complex disease.

The Alzheimer’s Society, which helped fund the research, put these findings in context. Head of Research, Dr Susanne Sorensen said:

“These two robust studies mark an exciting development for scientists hoping to identify a cause and find a cure for Alzheimer's disease. Although these studies will not bring us any closer to being able to predict who might be at greater risk of developing Alzheimer's, they will give scientists clues as to how Alzheimer's might develop. Most importantly, their identification could also lead to the development of new drug treatments in the longer term.”

Analysis by Bazian
Edited by NHS Choices