Don't believe the hereditary hype
Research into gene-disease links isn't yet a practical health tool

By Judy Peres | Tribune staff reporter
January 15, 2008

"Genetic marker linked to aggressive prostate cancer."

"Researchers ID new genes for breast cancer."

Such breathless headlines have become almost commonplace in recent months, but what is a lay person to do with this kind of information?


Not much, according to experts in genetics. It could take years for such a discovery to translate into something useful, such as a test that would tell you if you're at increased risk for a certain disease. And most of the time, the promise of the early finding fizzles out before it reaches that point.

Eight years after the human genome was decoded, the amount of practical information that has led to improvements in individualized medicine is minuscule. It turns out the genetic contributions to disease are much more complex than originally thought.

"Generally speaking, we believe all diseases -- diabetes, cancer, heart disease, stroke -- have underlying genetic components," said Scott Weissman, a genetic counselor at Evanston Northwestern Healthcare. "But those genetic components interact with lifestyle factors, environmental exposures and other genetic factors.

"They find markers all the time for Parkinson's disease, Lou Gehrig's disease, Alzheimer's disease. Only occasionally do they find a gene that causes a very high absolute risk of developing the disorder."

Commonly a genetic change, or "marker," is found to occur more frequently in people who have a certain disease than in those who don't, Weissman explained. But that doesn't mean someone who has the marker is likely to get the disease.

"A marker might increase the risk a little," Weissman said, "But you need something like an environmental exposure or another gene to make the disease come about."

That isn't stopping commercial firms from offering to scan your genes and tell you if they predispose you to certain diseases.

"The technology to look for variants in our genetic makeup is there," said Angela Trepanier, president of the National Society of Genetic Counselors. "But the knowledge to thoroughly assess what those variants means is lagging behind."

When a genetic marker confers a really high risk, it's generally in families where a lot of people going back several generations have had the disease.

A good example is the BRCA1 or BRCA2 gene. People who inherit certain mutations in those genes have a lifetime risk as high as 85 percent of getting breast cancer. Typically those people also have relatives with the disease, and they tend to get it at younger ages.

One way to tell if a gene discovery is useful, the experts say, is to look at how much it increases a person's risk of developing a certain disease.

"Lifetime risk" is always expressed in absolute terms. So an 85 percent lifetime risk means that, on average, 85 out of 100 people with a given risk factor will get the disease.

On the other hand, a study that made news headlines last summer was about the discovery of several genes that were found to increase the risk of breast cancer by about 50 percent. That's a relative risk increase.

"Women with breast cancer had the genes more often than women who didn't," Weissman said. "They raised the risk from 12 percent [the average woman's lifetime risk] to maybe 18 percent, a 6 percent absolute risk increase."

Another way to tell if a genetic marker is useful is to ask whether doctors would treat you differently if you had that marker.

In the case of a person with one of the new breast cancer genes, the answer is no. A 6 percent increase in the risk of getting breast cancer isn't nearly enough to start getting screening tests at age 25 or 30 or to have your ovaries surgically removed to reduce your risk.

Dr. James Doroshow of the National Cancer Institute explained at a recent medical conference (yes, even doctors don't understand this stuff) that the discovery of a genetic marker in a group of people with a certain disease is only the initial step in a very long process.

First, other researchers have to replicate the finding in other populations. ("If a study was done in Iceland, what does that mean for an African-American living in Los Angeles?" Trepanier asked.)

Next, they have to show that the marker identifies nearly all the people who have the disease and almost no one who doesn't. Otherwise, a test for the marker would be too inaccurate to base medical decisions on.

Then they have to test the marker prospectively in yet another population. That means screening a lot of people to see who has the marker and then following them for 10 or 20 years to see if they get the disease.

Few genetic tests are subjected to such rigorous scrutiny, said Trepanier.

----------

jperes@tribune.com
http://www.chicagotribune.com/featur...,2590652.story