By Geoffrey S. Ginsburg, MD, PhD

Geoffrey S. Ginsburg, MD, PhDThe appeals of the leading companies that sell direct-to-consumer (DTC) genome scans are hard to resist, especially in this age of the ever-more informed and increasingly "take charge" patient.

With one small saliva sample and a fee, a buyer can learn whether he or she has genetic traits that are associated with the most common killers of our time, from heart disease to diabetes to cancers of all kinds.

It is fair to say that these firms provide accurate and scientifically valid genetic-testing results from the samples provided by their customers. But there is also evidence that the interpretation of the genetic results may differ from one firm to another.

Moreover, although it might be useful to know one’s possible genetic predispositions toward developing heart disease or other conditions, there is still much debate within the medical community and the scientific literature as to whether this information is ultimately beneficial.

Just this May, plans to sell the first over-the-counter DNA testing kits at Walgreens were put on hold after the Food and Drug Administration questioned their legality; the agency is also examining online sales of similar tests, which continue to be available as of this writing.

One side of the debate argues that information gleaned from DTC tests may be motivating to patients -- it may change their behavior and help them make more informed decisions about lifestyle and medications.

In fact, the REVEAL study recently reported that children of patients with Alzheimer's disease were not only interested in having their ApoE4 (Alzheimer’s susceptibility gene) genotype measured, but also had less stress for having done so. Many even made decisions about long-term care insurance based on the results.

The other side suggests that the data thus obtained provide only limited information about a patient’s true risk profile -- which makes them potentially misleading, possibly causing patients to make needless decisions or creating stress and anxiety. The tests also could create false security among patients who don't show genetic risk for some diseases that could still develop.

At this point in the evolution of the field of personalized medicine, it is clear that patients who wish to "take control" of their health should do so in collaboration with their health care provider.

And because physicians and other medical professionals can play an important role in counseling individuals who are contemplating DTC tests or have already taken them, it behooves the physician community to become aware of the different DTC tests on the market, their nuances, and their potential value for the individual.

Putting Genetic Risk into Context

One of the clinician’s primary goals when discussing these tests should be to help the patient see these test results as just one risk indicator in the universe of data that doctors maintain on their patients.

Genetic information is best placed in the context of other clinical information, with one of the most important being a thorough family history.

Patients should also know that most of the genetic data gained from current state-of-the-art DTC tests are incomplete. Many of the inheritable tendencies of consumer interest -- one's likelihood of getting diabetes, for example -- are based on numerous genes that work together in the development of, or in providing protection from, the condition in question. But current tests detect abnormalities in only a few of these genes. Tests do not yet exist for most of the others, nor does knowledge about how they all work together, or sometimes fail to.

This point may be confusing to patients who are familiar with the mendelian-genetics tests that already exist, such as those for cystic fibrosis or Huntington's, diseases that are the result of a single gene.

But given the genetics of complex diseases such as cancer, diabetes, and heart disease, the presence of one or even several abnormalities does not have a "no doubt about it" aspect and may actually be inconsequential. And if, for argument's sake, we did know all of the genes involved in a particular medical condition, and a test were available for each of them, and the telling mutations proved to be present, there would still be other factors, such as lifestyle and diet, that influence the outcome.

"Genes load the gun," it has often been said, "but environment pulls the trigger."

How best to interpret the results of genomic scans is also an area of significant uncertainty. A study in the October 8, 2009, issue of Nature, which compared two companies' results from the same individuals’ samples, found little difference in the accuracy of the genetic analyses done by the DTC testing outfits, but several differences in the reporting of risk.

This finding does not impugn the test providers so much as it illustrates that medicine is still as much an art as a science -- a reality not unique to the field of genomic and personalized medicine. Uniform industry-wide standards obviously need to be developed. But even under the best of circumstances, genetic test results and the interpretations that go with them should not be taken as gospel.

By the end of the next decade or so, it's possible that personal genomic testing could move from a boutique business serving the few who are interested and can afford it to a technology that is fully integrated into health care.

The benefits for both patients and providers could be enormous: imagine a health care system, for example, that could sequence virtually every person’s genome -- if possible, at birth -- thereby giving each person a lifelong tool for prevention, diagnosis, and treatment of the diseases he or she is most likely to develop.

Providers could work with their patients to formulate individually tailored health plans, be more strategic in their interventions, and use limited resources more effectively to improve people's health and reduce their risks. This is, in essence, what personalized medicine is all about.

At this point much work still needs to be done before we can fully explore such possibilities.

In the meantime, it is important for physicians to educate themselves in this exciting and uncertain new world. To do so, we need not all become experts in the details of genetics or even know which genes are being analyzed in any given case.

Clinicians should even consider simply having their own genomes scanned. This action would, more than any course or literature search, change the abstract concept of DTC genomic scans into a personal reality. It would also give clinicians a wider and richer perspective to bring to the patient-physician relationship, as physicians' and patients' understanding of genetic-testing innovations mature together.

Opinions expressed in Controversies in medicine are those of the author and do not necessarily represent Duke medicine as a whole.

Geoffrey S. Ginsburg, MD, PhD, is the founding director of the Center for Genomic Medicine in the Duke Institute for Genome Sciences & Policy. He is also a professor of medicine and in pathology at Duke University Medical Center.