What You Should Know Before You Spit Into That Test Tube

By Rick Weiss
Sunday, July 20, 2008

Jeffrey Gulcher had no reason to think much about prostate cancer. He was just 48, and the disease typically strikes later in life. Even the most cautious medical groups agree that most men need not begin annual prostate screenings until age 50.

But Gulcher happens to be the chief scientific officer of deCODE Genetics -- one of several companies that, amid some controversy, have begun offering direct-to-consumer DNA tests that can help people predict which diseases they are likely to get. So in April, he spat into a test tube and, without giving the matter much thought, sent the sample in for analysis by his own company.

He was in for a shock. The test indicated that he carries a genetic variant that nearly doubles his lifetime risk of getting prostate cancer: While the average man has a 15 percent chance of being stricken, Gulcher had a 30 percent shot. That spurred his physician to order a standard blood test for prostate cancer. The result was toward the high end of the range considered normal, which, together with the DNA test, worried the doctor. He referred Gulcher to a urologist, who performed an exploratory biopsy -- and found that Gulcher's prostate gland was riddled with cancer, and a fairly aggressive version of it at that.

Gulcher is going in for surgery tomorrow, and not a moment too soon. Tests suggest that the disease has not yet spread to other parts of his body, a milestone that often portends death and that may well have been passed had he waited until he turned 50 to get a standard prostate-specific antigen (PSA) test.

Did genetic testing save Gulch er's life? I think it may have. His dramatic story seems to illustrate perfectly the claims, made by his company and others, that an open market of DNA tests is the 21st century's ticket to a healthier nation. But a closer look suggests that this fast-growing industry, with its snazzy Web-based come-ons, could benefit from some temperance and independent oversight.

The technology is undeniably impressive. For as little as $1,000, anybody who can drool into a mailing tube can now find out his or her genetic odds of getting any of 20 or more potentially debilitating diseases, including cancer, heart disease and diabetes. Most of these tests will not lead to a frank diagnosis, as happened with Gulcher. But discovering an inherited propensity toward a particular illness can motivate consumers -- or, as they used to be known, patients -- to get more frequent checkups, take preventive medicines or make lifestyle changes to try to ward off the specter of disease. At last, we seem to be on the cusp of the long-promised personalized-medicine revolution in which gene tests allow physicians to craft far more individualized and effective ways of keeping us well.

But tests that look into the fog of people's medical futures are freighted with tricky medical, economic and bioethical implications. For one thing, most genes are not determinative, so these tests can convey only odds, not destinies. Even with the doubled lifetime risk for cancer that's associated with Gulcher's prostate gene variant, two out of three men who receive a "positive" test for that gene will never get the disease. And many of those who do will get it so late in life and in such a benign form that no treatment would be justified. So that's at least two new members of the "worried well" who could be losing sleep and spending money on unnecessary follow-up tests for every person who would arguably be appropriately forewarned.

Moreover, the tests are still new and easily misinterpreted, even by professionals. Online results may be subject to security and privacy breaches. And some companies are using people's gene profiles to conduct independent research. That suggests to many ethicists and lawyers that these firms' paying clients ought to be informed that they are subjects in experiments, with full disclosure of potential risks and rights.

Most worrisome of all, at least a few companies seem to be peddling DNA-based versions of snake oil. Some firms claim to be able to identify inherited nutritional deficiencies that -- guess what? -- are treatable with pricey supplements that they just happen to sell. Some even promise to discern from your genes what kind of person you should marry to ensure a blissful sex life and healthier babies. Welcome to the Wild West of personalized genomics.

These problems are not insurmountable. But there is precious little oversight of this burgeoning new industry, in part because genetic analysis does not fit cleanly into any existing category of medical practice. And if the first wave of DNA-screening companies to hit the market gets its way, there won't be any more adult supervision in the foreseeable future. In a blatant effort to stave off regulation, top officials from all the major competing gene-test companies met early this month and quietly agreed to spend this summer hammering out a "best practices" document that they would promise to follow. This is a great idea, but it's not enough.

No state or federal agency can today assure consumers that the DNA tests they order will give accurate results -- or that the results, even if technically accurate, will have any practical value. The Food and Drug Administration says it has the authority to regulate all gene tests but has decided, at least for now, to ignore the vast majority of those developed so far. The Federal Trade Commission (FTC), which is supposed to protect consumers from fraudulent claims, has never taken an enforcement action against even the most transparently deceptive gene-test companies. And the Centers for Medicare and Medicaid Services, the division within the Department of Health and Human Services that oversees clinical laboratories, has so far opted to steer clear of the genetic-testing world, despite pleas from federal advisers to ensure a minimal standard of gene-test proficiency.

The companies say that what they do is different enough that they should not be shoehorned into the conventional medical-testing rules. "For the first year and a half of our existence, all we did was try to figure out how to fit into the regulatory environment," said Dietrich Stephan, co-founder of Navigenics, a leading California-based gene-test company, adding that the effort cost an estimated $10 million.

That's real money. Yet even with all that preparation, Navigenics and a dozen other testing companies recently received warnings from individual states accusing them of violating state rules for labs. Situations such as this cry out for the guiding hand of the federal government -- not necessarily through cumbersome regulations, which can be too rigid to keep up with quickly changing science, but through formal guidelines, at least, promulgated by HHS. These could set clear expectations about how accurate gene tests should be -- and what it means to be "accurate" in the brave new world of predictive health -- and what level of informed consent should be obtained from clients. The promulgation of such standards will take real effort from HHS Secretary Mike Leavitt, who has championed personalized medicine but who has thus far been largely AWOL on the gene-test issue and has little incentive to push hard in the final months of the Bush administration. The FTC also needs to show that it has teeth and can bite.

Genetic-testing companies need to ante up, too. The responsible ones could buy a lot of good will by offering the public easily accessible scientific details (online and elsewhere) about the specific genes or genetic markers they are testing for; citations for the published studies they use to justify their claims that those genes have real medical relevance; the privacy and security systems they have in place; and the protocols for any experiments that clients' specimens may be used in. The firms should also disclose any approvals they have sought, obtained or denied from independent scientific and ethical review boards.

I took heart that such a future is possible when, at an HHS meeting two weeks ago, I saw chiefs from the five major competing gene-test companies sitting next to one another, speaking cooperatively to federal advisers. If these executives move aggressively to do the right thing, and if federal officials help them with some smart but tough guidance, perhaps those corporate heads can avoid a future in which they are called upon to appear side by side again -- this time before Congress, looking more like those famously photographed tobacco CEOs, being asked tough questions about what exactly they have been selling, and at what cost to American health.


Rick Weiss, a former science reporter for The Post, is a senior fellow at the Center for American Progress.

© 2008 The Washington Post Company