David Altshuler, a geneticist at the Broad Institute, in Cambridge, MA, cautions that the findings need to be repeated, because the study groups and the control groups were drawn from two different populations, increasing risk for detecting genetic differences not linked to longevity. "The authors were very careful in their analysis to address these points, but it will nonetheless be important for independent investigators to confirm the results," he says.

Scientists haven't yet looked in detail at the genes implicated in the research; the microarrays used in these studies spotted genetic markers near genes, not the genes themselves. The next step will be to sequence some of the candidate genes in order to figure out what the longevity-linked variants do.

The study did highlight some genes previously associated with longevity, such as a protein involved in cholesterol metabolism and tied to relative risk of Alzheimer's, as well as genes linked to chromosomal instability and the insulin pathway. "A lot of the genes have no function associated with them as of yet," says Perls. "We will begin looking at genetic databases to figure out what pathways these genes point toward and maybe start to look at some animal models."

The researchers caution that the study was limited to people of European descent. "We have to do these investigations all over again for different ethnicities and maybe different environments as well," says Perls. "If you're up in Greenland, it probably takes a whole different set of genetic variations to survive in that environment than in Arizona." His team is already working with a group in Japan that is studying a group of Japanese centenarians.

The findings also raise the possibility of developing a genetic test to predict an individual's chances of living past 100. But the scientists caution against use of this type of test, at least in the near term. "I think from a social point of view, it's not ready for prime time. A lot more study has to be done in terms of what physicians can do for people with the results of this test," says Perls. "If someone has tons of longevity marks, I start to worry about what insurance companies and others would do."

However, people who have already had their genomes analyzed, through services such as 23andMe, will soon be able to predict their risk score through a free website that Perls' collaborator is developing. But Perls hopes to head off commercial efforts to market this kind of test. "We are concerned that the marketing [for such a test] will not mention the shortcomings of the test," says Perls.

"My hope has always been that we would learn much more about how to get lots of people to live to older ages in good health and delay the onset of disease to the end of life," says Perls. "I do not think this will lead to a treatment that will get people to become centenarians, but rather make a dent in diseases like Alzheimer's."