Then Kelsoe, a prominent psychiatric geneticist at the University of California, San Diego, did something provocative for the buttoned-down world of academic medical research: He began selling bipolar genetic tests straight to the public over the Internet last month for $399.
His company, La Jolla-based Psynomics, joins a legion of startups racing to exploit the boom in research connecting genetic variations to a host of health conditions. More than 1,000 at-home gene tests have burst onto the market in the past few years.
The proliferation of these tests troubles many public health officials, medical ethicists and doctors. The tests receive almost no government oversight, even though many of them are being sold as tools for making serious medical decisions.
Health experts worry that many of these products are built on thin data and are preying on individuals' deepest anxieties.
"People are always rushing to the market on the basis of one or two studies," said Dr. Muin Khoury, director of the National Office of Public Health Genomics at the Centers for Disease Control and Prevention. "We have very little evidence that telling people their genetic information is going to make any difference."
Tests have become available claiming to help predict and diagnose everything from serious illnesses like cancer and Alzheimer's to athletic ability and a person's ideal diet. Psynomics' offering is one of the first psychiatric gene tests on the market.
Kelsoe, 52, acknowledges that bipolar disorder probably results from a combination of genetic factors and life experiences, and that the presence of these gene variations does not at all mean that someone will, in fact, develop the disease. He admits, too, that his findings about the genetic basis of the illness are far from complete.
But he said his test is a vital starting point toward moving away from the notoriously tricky practice of diagnosing bipolar disorder based purely on a person's behavior.
"The goal of this is to try and help doctors make an accurate diagnosis more quickly so the patient can be treated appropriately," Kelsoe said. "Anything is going to help, even if it just helps a little bit."
Bipolar sufferers experience intense mood swings as they cycle between manic, sometimes delusional highs and depressive lows that can lead to suicide if untreated. The disease is often misdiagnosed as other forms of depression, which delays treatment and can result in the prescribing of antidepressants that make some patients' symptoms worse.
To take the test, patients receive by mail a plastic cup that they spit into, seal and send back to Psynomics. The company analyzes DNA in the saliva.
Psynomics will send patients' test results only to their doctors to avoid the risk of self-diagnosis.
The report that accompanies those results instructs doctors that a positive test means patients are two to three times more likely to have bipolar disorder. But the studies from which those figures come also show the gene variations themselves are rare even among those with bipolar.
The report also points out that for now, the test is valid only for whites of Northern European ancestry who show some behavioral symptoms and have at least one other bipolar family member.
Patients taking Psynomics' bipolar test may feel branded by a positive result, even if they are not ultimately diagnosed with the disorder, said Hank Greely, a professor of law and genetics with the Stanford Center for Biomedical Ethics. Or they may feel false hope from a negative result, despite the company's disclaimers.
Likewise, doctors have little training beyond what companies tell them when it comes to applying the test results. "They may make a foolish decision that backfires to put you on meds," Greely said. "Or they may make a decision that backfires not to put you on meds."
[to top of second column]
Unlike many tests for other conditions on the market, Psynomics does not claim its bipolar test can predict a person's risk of developing the disorder later in life. It is meant to be used as a purely diagnostic tool for patients already showing symptoms.
That is an important distinction that makes the Psynomics test more responsible than others that promise a glimpse into the genetic crystal ball, according to Dr. Greg Feero, head of genomic health care at the National Human Genome Research Institute.
"Now you're talking about an individual who has symptoms or signs that already put them in a very different risk category than someone who has no symptoms or signs," Feero said.
Among hundreds of families Kelsoe has studied, one of the gene variations in the Psynomics test showed up in 1 percent of those unaffected by the disorder versus 3 percent who are affected. The other variation appeared in 7 percent of those without bipolar compared to 15 percent who have the disease.
Many other genes interacting with a patient's environment contribute to the development of bipolar disorder, Kelsoe and other researchers believe, meaning no single genetic variation ultimately causes the disease. Researchers in Kelsoe's lab are working to track down more genes.
"Why are we starting before it's finished? You've got to start somewhere," Kelsoe said. "Even if we knew everything about the genes, which we certainly don't, it's never going to be 100 percent predictive."
Psynomics has sold only a few tests so far but is projecting sales of 1,800 tests in 2008 and 30,000 in the next five years.
In coming months, at least two other startups led by genetic researchers are set to release their own psychiatric genetic tests. One test claims to predict the risk of developing schizophrenia. The other is designed to forecast the likelihood that some medications for major depression could heighten suicidal thoughts in patients.
The American Psychiatric Association has yet to create an official policy on genetic testing. A fact sheet issued by the Federal Trade Commission advises consumers to be wary of assertions made by at-home genetic testing companies.
The Food and Drug Administration does not evaluate the tests for accuracy, though a panel is working on a set of standards for the growing industry.
For now, worry persists that with the proliferation of tests, there is too little understanding of what to do with the results, or what they mean.
"We just don't know how people will use the information," said Dr. Jinger Hoop, a professor of psychiatric genetics and medical ethics at the Medical College of Wisconsin in Milwaukee. "We don't know whether it will be helpful to them in the long run."
On the Net:
Press; By MARCUS WOHLSEN]
Copyright 2008 The Associated
Press. All rights reserved. This material may not be published,
broadcast, rewritten or redistributed.