Yesterday, Nature Genetics online released full reports of the two largest genomewide association studies of sporadic Alzheimer disease performed to date. Netting three new risk genes between them, these two studies stand out not only for their collaboration and data pooling—together they engaged several dozen institutions from 10 countries—but also for fingering three new genes robustly, without the lingering doubt of small sample size and missing replication that has accompanied most prior gene association results except ApoE.

This website reported the story from presentations at the International Conference on Alzheimer Disease held last July in Vienna, Austria (see ARF ICAD story). On September 4, Nature Genetics held a press conference to announce the data to the media, and today, readers can download the papers and supplemental material from the Nature Genetics website.

The same online issue of Nature Genetics carries results from an ALS GWAS, as well (see ARF related news story).—Gabrielle Strobel


  1. We carried out the most powerful genomewide study of Alzheimer disease (AD), involving over 16,000 people from eight countries, and identified two new genes that increase a person’s risk of developing the disease. These are CLU or clusterin and PICALM.

    We compared over half a million differences in the DNA of each of 4,000 people with AD with 8,000 people without the disease. In addition to the APOE gene—a known risk factor—CLU and PICALM showed overwhelming evidence for a relationship with AD. These findings were replicated in a further sample of over 2,000 Alzheimer’s sufferers and 2,000 controls.

    The findings are significant and conclusive.

    We also found strong evidence that other genes play a role in disease risk. Putting our data together with the results of Philippe Amouyel’s study, a third risk gene was confirmed; CR1, complement receptor 1 gene.

    This combination of discoveries forms an important breakthrough in the current impetus to discover the causes of AD.

    Three of the risk genes, APOE, CLU, and CR1, have roles in protecting the brain from damage. Perhaps the changes we see in these genes remove this protection or may even turn them into killers.

    Our results may highlight new targets for treatments. For example, clusterin has a role in dampening down inflammation in the brain. Up until now increased inflammation seen in the brains of Alzheimer’s sufferers had been viewed as a secondary effect of disease. Our results suggest the possibility that inflammation may be primary to disease development.

    If we were able to remove the detrimental effects of these genes through treatments, we could reduce the proportion of people developing Alzheimer’s by 20 percent. In the UK alone this would prevent just under 100,000 people developing the disease. So the significance of these results is truly meaningful.

  2. Two of the most important implications of this study are that it establishes that this approach works and suggests that larger studies are likely to identify other genes conferring similar risks to AD. For this reason we are planning an even larger study of 60,000 participants. This should allow us to identify other genes of relevance to AD.

    We are also undertaking more complex analyses of our current data—using new analytic approaches we have developed in Cardiff. These allow many genes to be analyzed at once and identify patterns in the data that implicate specific biological processes.

    The identification of multiple genetic risk factors should allow us to triangulate down onto specific biological processes. These might be related to β amyloid deposition, but they might also identify completely new disease mechanisms.

    The key thing about genetics is that it allows us to distinguish events of primary importance from those that are occurring as a consequence of the disease—downstream if you like—and these primary events are likely to be the best targets for new treatments.

    It’s also possible that in the future we might be able to use the results of genetic tests as part of a battery of indicators to identify those who might benefit from early intervention with new therapies. I should stress that the current genes on their own are not strong predictors of risk and are not suitable for risk testing.

    This study is one of the first major successes for the new MRC Centre in Cardiff which has been established this year to undertake genetic work in neurodegenerative diseases like AD and PD—as well as psychiatric disorders like schizophrenia and manic depression. This centre is allowing us to look not only at individual diseases, but also the overlap between them; we are currently looking to see whether the AD genes we have identified are involved in other forms of dementia such as Parkinson’s.

  3. By Jean-Charles Lambert and Philippe Amouyel

    We conducted a GWAS in 2,032 AD cases and 5,328 controls from the prospective population-based 3C study. We next replicated our most interesting hits in independent collections from Belgium, Finland, Italy and Spain totalling 3,978 AD cases and 3,297 controls. This replication work, the largest performed so far, strongly supports the association of CLU and CR1 with the risk of developing AD

    It is with enthusiasm that we learned the validation of our observations in the Harold et al. GWA study. The combination of these genetic results and the available pathophysiological data seem to strongly support the involvement of these genes in AD.

    However, it is important to keep in mind that in view of the large number of analyses performed, high-throughput approaches involve finding a balance between the risk of observing significant results by chance and the risk of rejecting biologically valid hypotheses on purely statistical grounds.
    Although no solution has been found to improve this dilemma, several approaches can be developed. It is possible to increase the statistical power associated with GWASs by performing meta-analyses. Other complementary approaches could also consist in better analyzing SNPs nominally associated with the risk of developing AD (p

  4. The paper by Lambert and colleagues is important and timely. For those of us in the field of neuroinflammation, their report adds fuel to the hypothesis that immune/inflammatory pathways are not epiphenomena, but are rather pathoetiologic in AD. Specifically, the authors have found that the complement receptor 1 gene, critically important for enabling the innate immune humeral response, is a bona fide genetic risk factor for AD. This represents a greater-than-incremental addition to a growing body of literature, both from the clinic and from transgenic AD mouse models, that strongly suggests these signaling pathways are intimately involved in the manifestation of AD pathology. Importantly, the authors’ GWAS approach in over 7,000 combined cases and controls provides a powerful message to remain focused on immune/inflammatory signaling pathways as possible therapeutic targets for the clinical syndrome.

    That said, I would like to comment on a statement made above by Julie Williams, that “our results suggest the possibility that inflammation may be primary to disease development.” As important as the current work of Lambert and colleagues is, it deserves mentioning that this suggestion actually arose in the 1990s due to epidemiologic studies (currently at least 25) that reported on the risk relationship between non-steroidal anti-inflammatory drugs (NSAIDs) and AD (see Szekely et al., 2007 for a review). Many of these early studies examined inflammatory conditions such as arthritis, for which NSAIDs are commonly indicated, and found an inverse association with AD. Based on these studies, it was suggested that history of arthritis was a surrogate for NSAID exposure, leading later investigations to focus specifically on NSAID use. There have been at least 12 such non-prospective studies to date, 10 of which concluded that AD patients were less likely to have been using these agents than were non-demented controls.

    Christine Szekely and Peter Zandi recently performed a systematic review of these studies (which met stringent inclusion criteria), with a resulting meta-analysis for eight non-prospective studies (1,833 cases and 13,780 controls) that showed a 53 percent AD risk reduction in those study participants who reported using non-aspirin NSAIDs compared with non-users (Szekely et al., 2004). They also performed a meta-analysis for five prospective studies (836 AD cases and 16,294 controls) and found a 29 percent risk reduction.

    Interestingly, AD risk reduction was more pronounced for longer (>2 years) duration of NSAID use, where a risk reduction of 58 percent was evident. Unfortunately, the one randomized controlled clinical trial of NSAIDs in non-demented elderly, ADAPT, was prematurely halted after two years of treatment and two years of follow-up due to possible cardiotoxicity of certain NSAIDs; post-hoc analysis of these data is still underway. [Editor's note: for recent news on ADAPT, see ARF related ICAD story.]

    In conclusion, the international genetics teams should be commended for a well-executed and powerful GWAS-based analysis of AD risk. The data of Lambert et al. stress the importance of keeping the AD therapeutic searchlight aimed at immune/inflammatory signaling pathways and providing an important target molecule: complement receptor 1.


    . NSAIDs for the chemoprevention of Alzheimer's disease. Subcell Biochem. 2007;42:229-48. PubMed.

    . Nonsteroidal anti-inflammatory drugs for the prevention of Alzheimer's disease: a systematic review. Neuroepidemiology. 2004 Jul-Aug;23(4):159-69. PubMed.

    View all comments by Terrence Town

Make a Comment

To make a comment you must login or register.


News Citations

  1. Vienna: In Genetics, Bigger Is Better—Data Sharing Nets Three New Hits
  2. Research Brief: Latest ALS GWAS Points to Loci on Chromosomes 9, 19

External Citations

  1. Nature Genetics website

Further Reading

Primary Papers

  1. . Genome-wide association study identifies variants at CLU and CR1 associated with Alzheimer's disease. Nat Genet. 2009 Oct;41(10):1094-9. PubMed.
  2. . Genome-wide association study identifies variants at CLU and PICALM associated with Alzheimer's disease. Nat Genet. 2009 Oct;41(10):1088-93. PubMed.