AUDIE CORNISH, HOST:
From NPR News, this is ALL THINGS CONSIDERED. I'm Audie Cornish.
ROBERT SIEGEL, HOST:
And I'm Robert Siegel.
In California, a huge scientific project is beginning to show how genes, health habits, and the environment can interact to cause diseases. The goal is to find new ways to identify people at risk before they develop problems like heart disease, cancer and diabetes.
NPR's Jon Hamilton reports that the project began with 100,000 members of a health plan who agreed to donate a little saliva in the name of science.
JON HAMILTON, BYLINE: A few years ago, Kaiser Permanente decided to learn more about the genes of people in one of its health plans. Cathy Schaefer is an epidemiologist who helps run the project.
CATHY SCHAEFER: We sent a letter to all our adult members in Kaiser Permanente in Northern California, which is about two million people.
HAMILTON: The letter asked if they would fill out a survey and take part in a study. Schaefer says a lot of them did.
SCHAEFER: We then sent the informed consent and a little plastic kit that looks sort of like a large contact lens case, and in that people collected their saliva.
HAMILTON: And mailed it in. Before long, Kaiser had spit from 100,000 people. But researchers were still pondering the best way to extract genetic information from the saliva.
Neil Risch, a geneticist from the University of California, San Francisco, says the answer came to him during a talk by scientists from a company called Affymetrix. They were describing a new system that could obtain genetic information really quickly.
NEIL RISCH: And while they're doing this talk, my mind started racing like, OK, how much is this going to cost to do 100,000 people. And, you know, how long is it going to take and how many of these systems would we need?
HAMILTON: The National Institutes of Health provided $25 million, which allowed Affymetrix and the Kaiser team to process the saliva samples in just 15 months. [POST-BROADCAST CORRECTION: The processing was done by Kaiser and UCSF, not Affymetrix.]
HAMILTON: Then came another daunting task: matching each participant's genetic information with health information in Kaiser's electronic records. Cathy Schaefer says those records, which go back to 1995, include a lot of detail.
SCHAEFER: Diagnoses of disease, any procedures, all the medications they've been prescribed, imaging, x-rays, MRIs, CT scans, pathology results...
HAMILTON: Kaiser also had information on smoking and drinking habits, body mass measurements, and even geographic information that could be used to estimate exposure to certain chemicals in the air or water. It will take years to fully explore the data. But researchers presented some early results this month at the American Society of Human Genetics meeting in San Francisco. Risch says one result has to do with genes that affect so-called bad cholesterol.
RISCH: I think we discovered something like 45 different genetic variations that influence that measure, some of which were known before but a number of which were not previously known.
HAMILTON: Another early result involves cholesterol drugs known as statins. Risch says Kaiser has genetic information on nearly 28,000 members who started taking statins in the past few years.
RISCH: So we've already started looking at the difference between their cholesterol measures before and after they went on the statins. So we have a direct assay of drug response and we've already shown that there are genetic factors that are predictive of drug response.
HAMILTON: In other words, which people are going to benefit most from statins? The research team is also looking at genes that can interact with factors like weight to cause diabetes and high blood pressure. One person who is very excited about the project is Nobel laureate Elizabeth Blackburn at U.C. San Francisco.
ELIZABETH BLACKBURN: It's just this playground of incredibly rich data that we're just scratching the surface.
HAMILTON: Blackburn won her Nobel for work on telomeres. A telomere is a sort of cap on the end of a chromosome and it tends to get shorter with age. Activities like smoking shorten telomeres even more. Blackburn got Kaiser to include telomere measurements along with the other genetic information extracted from the 100,000 saliva samples. And she's been using those measurements to help understand differences in how men and women age.
BLACKBURN: For many years, people have known that women and men have different average length telomeres.
HAMILTON: On average, women have longer telomeres and, on average, they live longer than men. The Kaiser project, though, found that as young adults, men and women have telomeres that are about the same length.
BLACKBURN: But then we found there's a big split above age 50 and it gets bigger and bigger with age until about, you know, 75 or so. So a marvelous new finding nobody had ever seen before.
HAMILTON: The data also show that people with very short telomeres are more likely to die. Blackburn says someday, doctors will probably look at telomeres and lots of other genetic information when assessing a person's overall health.
BLACKBURN: Most of the time, we stand around and then, you know, the blow falls and we say oh, oh, we've got a disease. Well, it's not an event. It was a process that took years to happen inside our bodies.
HAMILTON: And Blackburn says the Kaiser project should help identify ways to interrupt that process. Jon Hamilton, NPR News.