The Washington Post reported this morning that a San Diego company called Pathway Genomics later this week will begin selling a home testing kit that for $179 lets you know if you have any of 20 or more genes that makes you vulnerable to contracting certain diseases. The Food and Drug Administration may crack down on the firm because it is making medical claims for a test that hasn't been submitted to the agency for approval.

The test (you send them a vial of spit and they send you back the results) has bioethicists in an uproar since there is not much you can do with the results. Most people are not sophisticated enough to understand the difference between having a statistical propensity to develop a disease and have a sword of Damocles hanging over one's head.

Moreover, negative findings could wind up giving people a false sense of security, the story warned, leading them to abandon lifestyles that ward off disease or foregoing routine screening that might save their lives. Many if not most people develop cancer or Alzheimer's disease through bad luck or environmental exposures, not because they have a bad set of genes.

Scientists make the link between particular diseases and genes by conducting what are called genome-wide association studies. They sequence the genomes of large groups of people and look for patterns. If they can identify people who have particular mutations of a particular gene and develop a disease in greater numbers than the general population, they assume the genetic marker has validity.

And it can be true. That's how Mary Claire King discovered that women in families with a history of breast cancer had common mutations of the BRCA1 and BRCA2 genes. But it's important to remember that you can have the relevant mutations and not develop the disease. Only women with a family history AND the mutation are at much higher risk of getting breast cancer. Women with the mutations but no family history are not. Clearly, something else is going on and to quote the Buffalo Springfield song, what it is ain't exactly clear.

A new study on genes and Alzheimer's Disease (AD) in the latest edition of the Journal of the American Medical Association highlights the problem with genome-wide association studies. While not everyone who gets AD has a family history of the disease, those who have a family history have a high propensity to develop dementia - estimated at over 60 percent. Previous genome studies found familial AD is clearly associated with the presence of a particular mutation of the APOE gene, and many people are now tested for that mutation. Searching for other genes associated with this complex disease clearly was worthy of further research.

But the study, conducted by a small army of researchers led by scientists at Boston University, found that while a number of gene mutations were mildly associated with a propensity to develop AD, none of the information was clinically useful. At best, they concluded, the gene mutations might be used by researchers trying to develop drug to combat the disease.

An accompanying editorial by Nancy L. Pedersen of the Karolinska Institutet of Sweden offered some sound advice:

Lessons from increasing numbers of epidemiological studies with prospective information indicate that changes in midlife behavior, particularly those that are also conducive to cardiovascular health, can reduce risk of dementia or at least postpone onset. (The) findings (of this new study) reinforce the futility of using individual genetic risk profiling for AD beyond collecting information on age, sex, family history, and APOE status. The challenge for the clinician today is to ensure that individuals in midlife engage in the well-established, personally advantageous preventive behaviors already associated with benefit.