Harvard Mental Health Letter

Psychiatric genetic tests raise concerns

Home kits are now available, but they may not be testing for the right genes.

It is now possible to pay for a commercial genetic test that the company claims will assess the risk of developing bipolar disorder. An individual can register on the company's Web site and pay the required $399 fee, and the company will ship a testing kit to the person's home. The individual requesting the test provides a saliva sample and ships the kit back to the company for analysis. The results are sent to whatever physician a person has designated, who is then expected to communicate the results.

And this is only the beginning. Genetic tests for major depression and schizophrenia are also expected to reach the market. Meanwhile, other companies are racing to develop technology to make sequencing an individual's genome affordable enough that people can learn whether they have gene variants that have been linked to particular diseases, including psychiatric disorders.

Is this the brave new world of genetics and patient empowerment — or a New Age version of snake oil?

At this point, probably the latter. Although scientists have been searching since the early 1990s, and the genetic suspects are legion, the genes responsible for most brain disorders remain unknown. (A notable exception is a variant of the APOE gene that increases risk for Alzheimer's.) Indeed, when it comes to commercial genetic tests, the age-old advice applies: Let the buyer beware.

One problem is that the field of genetics is advancing so rapidly that it's hard to keep up with new developments, never mind figure out which ones may be clinically relevant. One study, for example, found that fewer than 25% of psychiatrists felt competent to talk about genetic information with patients and families, and yet 83% thought it was their job to do so.

Multiple genes, small impact

The first disease genes ever identified were those for conditions like Huntington's disease, which results from mutations in a single gene. Anyone who inherits that gene develops the disease.

But most experts believe that psychiatric disorders develop because of the interplay between multiple genes, each exerting relatively small effects. That makes finding the responsible genes harder, because there are multiple small targets rather a single relatively large one.

Further complicating matters, research has revealed that many healthy relatives of people with psychiatric disorders have so-called risk genes — and yet do not develop the disorders themselves. Whether or not a person develops an illness depends on unknown ways the risk genes interact with other genes and environmental factors.

False positives are common

Scientists have now identified hundreds, perhaps thousands, of candidate genes that may contribute to schizophrenia, bipolar disorder, and other psychiatric conditions. But experts continue to debate which genes are actually involved.

One problem is that most candidate genes identified fail to hold up — meaning that the association between the gene and a given illness disappears when scientists try to replicate the results from an initial study in a different group of people. One analysis estimated that 70% to 80% of candidate genes are just such false positives. Thus, one criticism of the psychiatric genetic tests that will first reach the market is that they test for genes that are not proven to be involved in disorders such as depression or schizophrenia.

Where the science is headed

Scientists are optimistic that a genome-wide approach will help better identify the genes involved in psychiatric disorders. This approach identified a cluster of genes that affect the risk of developing coronary artery disease and diabetes — which, like psychiatric disorders, arise from the interplay of multiple genes and environmental influences. But a major challenge is that researchers must recruit thousands, and perhaps tens of thousands, of participants in order to find reliable genetic patterns that indicate risk.

The effort has already begun. One example is the Autism Genetic Resource Exchange, a joint effort by researchers collaborating in the Autism Consortium. They reported in 2008 that they had identified a 25-gene area on chromosome 16 that, when deleted or duplicated, increases susceptibility to developing autism. A genetic microarray test that examines the 25 genes at once is now being used at several academic medical centers as part of the diagnostic process to evaluate children suspected of having autism.

Someday it may indeed be possible to reliably assess risk for psychiatric disorders. But at this point, the technology — and the science — is still evolving.

Sanders AR, et al. "No Significant Association of 14 Candidate Genes with Schizophrenia in a Large European Ancestry Sample: Implications for Psychiatric Genetics," American Journal of Psychiatry (Jan. 15, 2008), electronic publication ahead of print.

Smoller JW, et al, eds. Psychiatric Genetics: Applications in Clinical Practice (Washington, D.C.: American Psychiatric Association, 2008).

For more references, please see www.health.harvard.edu/mentalextra.