Once your patient takes the medication that you prescribed, you hope that
it will produce the desired therapeutic outcome with few or no unwanted side
Prescribing psychotropic medications may soon become a more precise
science, however, as the exciting new field of psychiatric pharmacogenomics
moves into gear.
Specifically, variants in genes that code for drug-metabolizing enzymes in
the liver are being found to influence those enzymes' handling of psychotropic
medications, and therefore tests for such variants may have the potential to
predict patients' responses to such medications before they receive them.
The field of psychiatric pharmacogenomics got under way about five to 10
years ago. Since then, scientists in the United States, Canada, England,
Germany, Spain, Sweden, and elsewhere have become enthusiastic about it.
Moreover, the field has been given a big push by some pharmaceutical
GlaxoSmithKline is studying the relationship between patients' genetic
variability and their response to medications, with the aim of using the
information in drug development and in helping clinicians prescribe
medications and doses for patients that are most likely to benefit them. In
2001 Allen Roses, M.D., senior vice president of genetics research at
GlaxoSmithKline, predicted that before long, psychiatric pharmacogenomics
would become a reality (Psychiatric News, April 20, 2001).
Roche Molecular Systems has developed a technique for screening gene
variants of two drug-metabolizing enzymes involved in psychotropic drug
metabolism. This AmpliChip CYP450 Genotyping test was approved by the U.S.
Food and Drug Administration in January to help clinicians make more informed
Such keen interest and technological advances have in turn spurred some
provocative discoveries, the bulk having to do with the drug-metabolizing
enzyme cytochrome P450 2D6 (CYP2D6).
In the middle of the last century, scientists discovered that liver enzymes
are involved in drug metabolism. Then in the 1970s they found that one liver
enzyme—CYP2D6—is important for metabolizing psychotropic drugs.
Since then, they have learned that CYP2D6 is involved in breaking down some 75
percent of all psychotropic medications. And thanks to the new field of
psychiatric pharmacogenomics, they now know that there are more than 100
variations in the gene that codes for CYP2D6; that whereas most of these
variants decrease enzyme activity, some can increase it; and that the CYP2D6
gene has the peculiar propensity to accumulate in multiple copies on the
chromosome that contains it (chromosome 22).
So persons who have multiple copies of the gene metabolize psychotropic
medications very rapidly.
Such findings have led to testing for such gene variants. Depending on
which variant or variants they possess, people are usually categorized as
being ultra-rapid, extensive, intermediate, or poor CYP2D6 metabolizers. This
information can give psychiatrists some idea of how patients are going to
metabolize psychotropic medications before patients receive them.
Most such testing has been done on a research basis. For example, Adrian
Llerena, M.D., Ph.D., an associate professor of pharmacology at the University
of Extremadura Medical School in Spain, is testing subjects for variants in
genes that code for the CYP2D6 enzyme, as well as for some other
Jose de Leon, M.D., poses with lab technician Elaina Murray-Carmichael
in his psychiatric pharmacogenomics lab. De Leon is conducting a study to
determine whether testing for variants in the genes that code for the CYP2D6
and CYP2C19 enzymes is cost-effective.
Drug companies are also starting to use such testing to stratify subjects
before clinical trials into good drug metabolizers and poor drug metabolizers,
Mario Masellis, M.D., a postdoctoral fellow in psychiatry and neurology at the
University of Toronto, told Psychiatric News.
Some testing for gene variants is beginning to be applied clinically as
Mark Linder, an associate professor of pathology at the University of
Louisville, and coworkers operate a lab where psychiatrists can have patients
tested for variants in genes coding for both the CYP2D6 enzyme and the
cytochrome P450 2C19 enzyme (CYP2C19), he told Psychiatric News. The
CYP2C19 enzyme is known to break down some psychotropic medications, but not
nearly as many as the CYP2D6 enzyme does.
"We perform CYP2D6 and CYP2C19 gene analysis on psychiatric patients
both as a routine service for psychiatrists and on a research basis,"
Marja-Liisa Dahl, M.D., Ph.D., a professor of clinical pharmacology at Uppsala
University in Sweden, told Psychiatric News. "The genotype
analysis is also performed by at least two other clinical pharmacology
laboratories in Sweden—the Karolinska University Hospital in Stockholm
and the Linkoping University Hospital."
For several years the Mayo Clinic has had a lab in which patients can be
tested for variants in genes that code for both the CYP2D6 and CYP2C19
enzymes. David Mrazek, M.D., chair of psychiatry at Mayo, avails himself of
these tests before putting patients on antidepressants or antipsychotics. For
instance, he said, "The antidepressant paroxetine is primarily
metabolized by the CYP2D6 enzyme. So if you discover that a patient is a poor
CYP2D6 metabolizer, you would be careful about prescribing paroxetine for him
or her and perhaps prescribe something else instead."
"Any doctor in the country, or in the world, can send a small sample
of blood from a patient to our laboratory," Mrazek continued."
Every day lots of specimens come in, and we send results back. So Mayo
does a very good job, but there are lots of other choices as well, especially
now that any hospital lab can buy the AmpliChip."
Even with the ever-widening availability of labs where psychiatrists can
have patients tested for such gene variants, challenges still need to be met
before such gene-variant testing is widely embraced by psychiatrists.
For one, such testing must have practical value in a real-life clinical
setting, and some evidence is building in this regard. For example, Jose de
Leon, M.D., an associate professor of psychiatry at the University of
Kentucky, along with fellow researchers, had the DNA of some 500 subjects
taking risperidone tested using the AmpliChip assay. The investigators then
looked to see whether subjects who turned out to be poor CYP2D6 enzyme
metabolizers were the same subjects who had experienced adverse reactions to
risperidone. The answer was generally yes, they reported in the January
Journal of Clinical Psychiatry.
For patients to be open to gene-variant testing, they must see it as being
in their best financial interests. Testing patients for variants in the genes
that code for the CYP2D6 and CYP2C19 enzymes costs between $200 and $500."
But you only have to do this test once in patients' lifetimes since
their genes do not change," Mrazek stressed.
For psychiatric hospitals to use such testing, it must make financial
sense. De Leon is conducting a study on some 4,000 inpatients in Kentucky
state psychiatric hospitals to determine whether testing for variants in the
genes that code for the CYP2D6 and CYP2C19 enzymes is cost-effective.
Scientists in the field are optimistic that such challenges can be met.
Masellis anticipates that such testing will become clinically widespread
within the next five years. Mrazek, in fact, believes that "the field is
about ready to explode."
"Right now," he said, "most doctors just get a history
from patients, then think about the symptoms and think about the diagnosis,
and give a drug they are familiar with a try. Ninety-five percent of the
doctors do that. I think 10 years from now probably no one will do that! They
will have a small sample of blood from the patient tested and will then
conclude, `Oh, this patient would most likely respond well to drug X, but not
to drug Y, so I will prescribe drug X.' In other words, it's just an extra
window of information that most doctors don't have today, but that you will
have five to 10 years from now." ▪