by Laura Hercher

Only a few years ago, genetics was an obscure little corner of medicine focusing on a laundry list of rare diseases that ran in families according to the simple patterns familiar to anyone who had studied Gregor Mendel and his pea plants in high school biology. Working with technology that was destined to go from cutting edge to outdated in the blink of an eye, researchers in the 1970’s and 80’s painstakingly established the location of the genes responsible for disorders like cystic fibrosis and sickle cell anemia. In the 21st Century, geneticists have waded into the mainstream, seeking to unravel the genetic basis of common disorders like diabetes, cardiovascular disease and mental illness.

After years of hype with little to show for it in the way of clinical applications, there are tantalizing reports of breakthroughs in research – news that may raise the hopes of individuals with psychiatric disease and their families that genetics will soon have something to offer them. However, despite all those headlines saying “Researchers Discover Gene for Schizophrenia
,” identifying genes for illnesses like schizophrenia, bipolar disease, depression or autism is not the same as finding genes for Mendelian diseases, where a defect in a single gene causes the disease.

All of the major mental illnesses fall into a category geneticists define as “complex disease,” which means that they have multiple genetic as well as environmental risk factors, and that the contributing genes interact with one another as well as interacting with the environment. It is a very complicated puzzle with lots of interlocking pieces, but the bottom line is this: genes for complex diseases are important, but not determinative. How well do genes predict the risk of disease in mental illness? Consider this: if one identical twin has schizophrenia, his or her sibling will share that diagnosis about 60% of the time. So having a genetic predisposition is obviously a big part of the picture – but it is not the whole picture.

Mental illness, a major cause of sickness and disability all over the world, has been a primary target of the initial attempts to use information from the Human Genome Project to identify genes of interest in large scale research efforts called genome-wide association studies (GWAS). Genes are often described loosely as “causing” an illness but, as the name suggests, it is more accurate to say that the genes identified in these studies are associated with a disease. Disease causing genes are rare; nature, as a rule, is not very tolerant of genes that produce inevitable bad outcomes.

Many of the genes that have emerged from the early GWAS as likely candidates to play a role in schizophrenia or bipolar disease are common – that is, the variant associated with the disease is carried by a big percentage of the population. What does this tell us? It suggests two things. First, that the amount that any one gene increases a person’s risk is small. For example, a gene variant that was shown to increase the risk of schizophrenia by a factor of two would be considered a great success by researchers – but for the average individual, doubling the risk means only that a 1% risk of schizophrenia becomes a 2% risk of schizophrenia. Second, calling something “common” may not be a compliment in certain circles, but in genetics, commonness is a measure of success. A variant that has worked its way up to common must be doing something right. Clearly, not everybody carrying this gene variant is going to get sick, so for those who stay healthy, it may well be that this same variant confers some sort of advantage.

Paradigm Shift
So, in the paradigm shift from studying rare variants of large effect to studying common variants of small effect, we have produced lots of information but not so many answers. There are many potential applications for genetics in the practice of psychiatry. It could be used for risk assessment, to confirm a diagnosis, to predict the course of a disease. Pharmacogenetics – the use of genetic testing to predict drug response – has tremendous potential to help patients with depression, schizophrenia, bipolar disease and other mental illnesses where medication regimens are often established by a trial and error process that is expensive, slow and frequently painful for patients and families. Pharmocogenetics may also help establish who is at risk for certain side effects, reducing some of the dangers associated with treating mental illness. Neuromark, a biotechnology company based in Colorado, is hoping to market a test called Mark-C, currently under clinical review, that is designed to predict the risk of suicidal thoughts in response to treatment with the anti-depressant citalopram. In 2007, the government reviewed another test meant to be used by physicians to predict a patient’s response to anti-depressants; the working group applauded the concept of testing but indicated that in its current form the test did not supply enough validated information to be of use to clinicians.

Despite the fact that most of the experts protest that this information is not ready for prime time, a number of companies have moved forward with efforts to market tests for genes associated with mental illness. Products like the available bi-polar predisposition test from the start-up company Psynomics is one example of a new phenomenon: genetic tests offered directly to the consumer over the Internet. Advocates for on-line testing suggest that it gives consumers a way to get information that is private and anonymous; critics point out that the privacy guarantees are untested and that divorcing testing from medical settings means that patients and families have no independent source of information to assess the validity of the tests, or to counsel them on how to put the results in context.

But perhaps the most insidious effect of marketing genetic tests is that it perpetuates the myth of genetic determinism. The flood of new information on genetic determinants of mental illness has only proven one thing: no simple explanation will ever adequately address the question of what makes any given individual get ill. Ironically, constant media reports of researchers “unlocking the secrets” of autism, schizophrenia or bipolar disease suggest just the opposite. Tests for predisposition to these diseases send the same message: it’s all in your genes. Affected individuals and concerned family members will have to decide for themselves if they want to buy the test, but no one should buy the hype. We have heard a lot about genetics and we are going to hear a lot more, but no matter how much we learn from our genes, it will only ever be a part of the story.

*Laura Hercher is on the faculty of the Joan H. Marks Program in Human Genetics at Sarah Lawrence College where she teaches and writes about the legal, ethical and social implications of clinical genetics practice with a focus on issues surrounding the integrations into practice of predictive testing for complex heritable diseases such as schizophrenia. Her latest article, a consideration of direct-to-consumer marketing of genetic tests over the internet, was published in Scientific American in December 2007.