Transcript: Lance Armstrong on Surviving Cancer

Prostate cancer is a disease that runs in families—if your grandfather had it and your father had it, you're at a higher risk than average of getting it, too—so it's always been clear that flaws in DNA play an important role. But for over a decade scientists have struggled to find genes that contribute to the disease. Instead, they've mostly found false alarms, candidates that have been implicated in one study and just as quickly discarded in the next. "We've known there was a genetic component," says National Cancer Institute researcher Stephen Chanock, "but we've had no robust, strong finding that everyone could agree on."

According to three new studies published Sunday in Nature Genetics, though, those days are over. Three separate groups of scientists have pinpointed seven variations in DNA that definitely increase a person's risk of prostate cancer. All of the variants are found on the same chromosome. But don't call them "prostate cancer genes"—the reason scientists couldn't find those before, it seems, is that the culprits turned out not to be genes at all. Instead, they are found in so-called "junk DNA," portions of the genome that don't make proteins. "What these variants are doing inside the cell is still a big question," says Brian Henderson, dean of the school of medicine at the University of Southern California. "But whatever we've found, it's the same finding by three different groups who didn't find anything else. After 15 years of looking, that's very exciting, believe me."

One of the studies, led by Henderson, David Reich of Harvard Medical School and others, lays out where the seven genetic variations are. All appear in seemingly barren stretches of chromosome 8 with virtually no genes. The other two studies confirm the location of the prostate-cancer risk factors and show independently that they do influence a person's risk for prostate cancer. In the uncertain world of prostate cancer, that confirmation is crucial, says William Catalona, a urologist and surgeon at Northwestern University who coauthored the second paper with the Icelandic firm deCODE Genetics. "One of the problems that has plagued prostate cancer is that nobody can ever confirm anybody else's work," Catalona says. "You'll have a really good research group saying, 'We have a signal here,' and then everyone else will try to reproduce that signal and they can't. So people get skeptical and think that it's a false positive signal, and in genetics, false positive signals occur all the time."

The scientists' frustration has been amplified by their certainty that genetics play a large role in who gets prostate cancer and who doesn't. Unlike breast and lung cancer, which are clearly influenced by factors such as obesity and smoking, prostate cancer doesn't have as strong as a link to lifestyle choices. A family history of the disease, which strikes one in every six men, is by no means a sure indicator that someone will get it, but it is a major warning sign. Race also seems to be involved—African-Americans' risk of developing prostate cancer is about 1.6 times higher than any other group's. The new studies begin to explain why. The seven genetic variants appear "across all ethnic groups," says Reich, but "all are more common in African-American than in European-American families."

The newly discovered variants on chromosome 8 are present in "a lot of prostate-cancer cases, the majority of them," says Henderson. But other portions of the genome also certainly contribute to the disease. It's hard to estimate how many more troublemaking variants there may be, he says, much less what chromosomes they're hiding on. But for the moment, he adds, the seven known variants "will certainly keep us busy for a while."

That's because scientists don't yet know what role the variants play in the body. It will be "a formidable challenge to unravel," says Chanock, a coauthor of the third paper. Increasingly, though, research has been revealing that "junk DNA" is a misnomer—it seems instead to play a key role in regulating the amount of proteins made by certain genes. The variants linked to prostate cancer may be involved in this type of regulation. They could influence, for instance, the activity of a gene called MYC, which controls cell division and has been linked to many different kinds of cancer. It, like the variants, is found on chromosome 8. "Everyone has a high level of suspicion about MYC. Could what we're finding be related to that gene?" wonders Henderson. "Maybe there's some long, distant effect it's having that we don't understand and don't have any prior knowledge of."

Alternately, the seven variants might be linked to an unusual property of chromosome 8, one that becomes clear only when a cell is cancerous. Tumor cells have genomes that look different from healthy cells; they develop mutations that enable them to thrive at the expense of the rest of the body. In almost all types of tumor tissue, including prostate tumors, says Chanock, "All hell breaks loose in this region where the variants are found." Healthy cells have two copies of chromosome 8; some tumor cells may have as many as 10 copies of this region. Perhaps, says Reich, "these genetic variants could be increasing the propensity of the DNA in this region to copy itself."

For scientists, then, task No. 1 is to explore the new biology they've found. "These discoveries may provide us with new markers and blood tests for prostate cancer susceptibility and aggressiveness," says Catalona, "as well as possible new targets for treatment and even prevention." However, it is far too early, adds Chanock, for clinicians to get involved. "Right now, we're a long way away from testing people for these variants and judging their risk by it. It would be challenging to determine exactly what you would say to counsel someone before and after the results of such a test," he says. But further down the road, he says, doctors will indeed be able to test men for these seven genetic variants and others, determining who's really at risk.