'Genes behind long life found'

Friday July 2 2010

Note: The authors of “Genetic Signatures of Exceptional Longevity in Humans” fully retracted this paper from publication in July 2011.

“Genes that mean you will live to 100” have been discovered by scientists, said The Daily Telegraph.

The news is based on a study by US scientists who have built a genetic model that can predict exceptional longevity, defined as survival well beyond the average human lifespan, with 77% accuracy.

This is an interesting study that built and tested a genetic model for exceptional longevity in over 1,000 centenarians and 1,200 control subjects. It indicates that genetic factors play a critical and complex role in living an exceptionally long time. However, as the scientists note, their model is not perfect and further research on the variations of the human genome are needed to improve on this.

Whether this research is likely to be turned into a commercially available test is unclear, as is whether the test would be helpful for the individual. Our environment and lifestyle also clearly play a role in longevity, so it seems sensible to increase our chances of reaching a healthy old age through controlling these modifiable factors where possible, regardless of our genes.

Where did the story come from?

This study was carried out by researchers from Boston University in the US and the Istituto Di Ricovero e Cura a Carattere, in Milan, Italy. It was funded by the US National Institutes of Health and published in the peer-reviewed journal, Science.

The research was reported accurately by most newspapers. It is not clear whether The Independent ’s claims of a forthcoming commercial genetic test to predict longevity are likely, or how such a test might be useful.

What kind of research was this?

This research was a genome-wide association study of centenarians, who are seen to be a model of health ageing: the onset of disability in these individuals is generally delayed until they are well into their mid-90s. It was based on the hypothesis that exceptionally old people are carriers of multiple genetic variants that influence human lifespan.

The researchers point out that while there is evidence that environmental factors such as diet and exercise contribute to healthy ageing, other data indicates that genetic factors play an important contributory role in healthy ageing and especially in exceptional longevity.

What did the research involve?

The researchers compared the genetic makeup of a group of centenarians and non-centenarian controls to identify genetic variants that were more common in the centenarians and therefore might be contributing to their longevity. They used this information to build a model to identify whether an individual was a centenarian or not.

The scientists recruited 1055 people born between 1890 and 1910 (cases), already enrolled in two ongoing centenarian studies, and 1267 controls, most of them from a recognised genotyping control database, selected to match the genetic backgrounds of those in the centenarian studies. All of the participants were Caucasian, to avoid genetic differences due to differences in ethnicity from affecting results.

First the researchers took 801 of the centenarians and 926 controls, and looked at about 295,000 single letter variations across the code of their DNA, known as single-nucleotide polymorphisms (SNPs). Once they identified SNPs that were significantly more common in centenarians than controls, they then looked at these in a second sample of 254 centenarians and 341 controls (the replication sample) to confirm their results.

In the second part of their study, the researchers took the information from their analyses and built a genetic model based on the SNPs that showed the greatest differences between the centenarians and controls. This model was intended to predict whether a person was a centenarian or not. It was initially built using data from the first group of centenarians and controls, and then tested on the replication sample of centenarians and controls.

The researchers also looked at whether there were ‘clusters’ of people within the group of centenarians that had a similar genetic make-up and whether these clusters had similar health problems.

What were the basic results?

In the first part of their study, the researchers identified 70 SNPs that were significantly more common in centenarians than controls. Of these SNP’s, 33 were also significantly more common in the second sample of centenarians.

The researchers developed a model for predicting extreme longevity based on 150 SNPs. They found that their model correctly identified exceptional longevity in 77% of their replication set of centenarians. The model also correctly identified 77% of those who did not have exceptional longevity (controls).

Further computer analysis revealed that 90% of the centenarians could be grouped into 19 clusters of people who had similar genetic make-up, which they termed ‘genetic signatures’. These clusters had differences in the prevalence and age of onset of age-associated disease such as dementia, high blood pressure and cardiovascular disease.

The researchers also says that while large numbers of longevity-associated variants (LAVs) within the genes appear to be necessary for extreme survival, they did not observe a difference between centenarians and controls in the numbers of many known disease-associated genetic variants. They say this suggests that extreme longevity may be the result of an ‘enrichment’ of longevity genes that counter the effects of genetic variations that predispose towards disease.

How did the researchers interpret the results?

The researchers say they have identified the genetic data to predict extreme longevity without knowledge of other risk factors. They acknowledge that the prediction is not perfect, and its limitations confirm that environmental factors also contribute to the ability of humans to survive to very old age.

In a separate interview reported by the newspapers, one of the researchers predicted that this information, freely available in the public domain, could be used by biotechnology companies to develop a commercial test for extreme longevity, although he added, society was probably not ready for this.


This study has identified a genetic model that predicted with 77% accuracy whether an individual had lived well beyond the average lifespan. It provides some valuable information about the common genetic signature linked to extreme longevity. There are a number of points to note:

  • This research included Caucasian individuals only and the results may not apply to other ethnic groups.
  • The model successfully identified 77% of the centenarians and 77% of non-centenarian controls in the replication sample. This illustrates that the model does make some incorrect predictions. It is also worth noting that the model may perform differently when used to predict longevity in the wider population. Further tests of the model in the population would be needed to determine its performance.
  • It was not clear whether all of the control individuals had already died, or whether they were still living, but the latter seemed more likely. If they were still alive, it is possible that some of them will live to be centenarians themselves. This might affect the accuracy of the results obtained and therefore the reliability of the model.
  • The model was for predicting exceptional longevity - living to 100 and over. It was not aimed at predicting how long an individual would live.

It is possible that in future, similar methods could be used to identify or screen for genetic predispositions to long life, but whether or when this will become a reality is not yet clear.

At this stage, the study is of limited usefulness to the important issue of how to maintain good health in old age. In future, the findings may help researchers to understand aging better, and perhaps to develop ways to improve health as we age, but this is a long-term goal.

Although the researchers point out that centenarians are a model of healthy ageing, relatively few people live much beyond the average lifespan and it is debateable if many would wish to. Being able to predict who amongst us may live to 100 or more may be of  limited use to individuals, when the main priority for most people is to stay as healthy as possible as they get older rather than simply living for longer.

Analysis by Bazian
Edited by NHS Choices