About half of all depressed people who take standard antidepressant drugs fail to improve. Some suffer unpleasant side effects and abandon the medicines, while others simply don’t feel better. Commercial tests claim to predict, by a genetic analysis, how well individual patients will fare on different antidepressants, but a panel convened by the Centers for Disease Control and Prevention (CDC) in Atlanta now says that the tests don’t work as advertised.
The panel “discourages” use of such tests until further studies clarify their value, according to a statement the group published in the December Genetics in Medicine.
“That isn’t to say that eventually there won’t be a role for these tests. We just don’t know what that role is yet,” says panel member Joan Scott of the Genetics and Public Policy Center in Washington, D.C.
The tests scan a person’s DNA for variations in genes for two key liver enzymes. These enzymes break down selective serotonin reuptake inhibitors (SSRIs), a standard class of antidepressants that includes fluoxetine (Prozac) and nearly a dozen other drugs. Variations in the two enzymes affect how quickly different people clear SSRIs from their blood, which in turn influences the drugs’ effectiveness.
People classed as rapid or ultrarapid metabolizers, for instance, might clear the drugs before they can work in the brain. Slow metabolizers, in contrast, might maintain too much of the drug, leading to unpleasant side effects such as jitteriness or loss of libido.
Finding a happy medium is the goal of the genetic tests, which physicians can order from laboratories and consumers can mail order from Roche Molecular Diagnostics in Pleasanton, Calif., for $300 to $400. The Food and Drug Administration approved the Roche test in 2004, concluding that it accurately measured genetic variations in the two enzymes. However, the FDA never asked Roche to prove that its test led to better patient outcomes.
So the CDC panel reviewed 16 studies and found that enzyme variations do not neatly correspond to blood concentrations of SSRIs. The panel also found no evidence that the tests led to improved patient outcomes.
“There was no consistency in the data,” says Scott. Some individuals classified as ultrarapid metabolizers—who in theory should maintain low blood concentrations of the drugs—had higher drug levels than some so-called slow metabolizers.
That’s because the two enzymes are “just one piece of a bigger picture” of how the body breaks down SSRIs, says Scott. Diet, other drugs the patient may be taking, and genetic variations the tests don’t account for can also affect SSRI metabolism, she says.
Robert Bernstein, executive director of the Bazelon Center for Mental Health Law in Washington, D.C., says that “everybody would benefit from better tailoring of treatment for depression. The idea that gene testing might [lead to] better treatment is promising.”
However, he agrees with the CDC panel that much more work needs to be done to reach that goal.