Pharmacogenomics in psychiatry

(This article was first printed in the January 2004 issue of the Harvard Mental Health Letter. For more information or to order, please go to http://www.health.harvard.edu/mental.)

The mapping of the human genome is may give new meaning to the popular conception of a designer drug. Today, it often refers to a chemical created in an underground laboratory to avoid the narcotics laws. Tomorrow it may mean a medication designed to fit the individual genetic pattern of the patient who uses it. A new technology could eventually help to transform not only psychiatric drug treatment but psychiatry itself.

Even today, genetic variation is sometimes considered in prescribing drugs. That’s what physicians are doing if they ask a patient about a family history of adverse drug reactions. Now the vast growth in the quantity of genetic information and the power to use it has given rise to the infant science of pharmacogenomics, which applies genomics — the study of the functions and interactions of all our genes — to drug treatment and the discovery of new drugs.

Expressing genes

The device at the basis of genomics is the DNA chip, or microarray, a small piece of glass containing hundreds of thousands of DNA sequences. A sample of blood or saliva is tagged with a fluorescent substance and dropped onto the glass. A given strand of active genetic material in this sample will bind only to the portion of the microarray that contains the right code. The spots that light up indicate which genes are now expressed, or turned on. It’s a way to identify many active genes at once rather than have to hunt them down one at a time. This technique can be used to map small individual variations in the pattern of a gene, called single nucleotide polymorphisms (SNPs), which can then be correlated with responses to a drug.

Psychiatric researchers concentrate on genes that regulate brain functions and the activity of psychotropic drugs. Some of these genes control pharmacokinetics — how the body absorbs and disposes of the drug. Others control pharmacodynamics — how the drug affects the body, or in this case the brain.

Today, the chief, practical psychiatric use of genetic technology is in pharmacokinetics. Many antidepressant and antipsychotic drugs, as well as nicotine and others, are metabolized in the liver by a group of enzymes known as the cytochrome P450 system. The drugs are broken down at different rates depending on genetic idiosyncrasies in the production of these enzymes — a major reason for individual differences in drug responses and side effects.

By competing for the attention of the cytochrome P450 enzymes, drugs can also modify one another’s activity. For example, selective serotonin reuptake inhibitors (SSRIs) may increase the blood level or prolong the action of a variety of other drugs. When a patient responds poorly to a drug, physicians can sometimes use their knowledge of these reactions and interactions to find alternatives. Today they rarely do genetic tests on individual patients, but that may change. Microarrays are being developed specifically to identify SNPs that affect the activity of P450 enzymes.

Genes, drugs, and neurotransmitters

The pharmacodynamics of psychiatric drugs is more complicated and less well understood. Many genes acting in concert are involved, and scientists have little detailed knowledge of how the drugs work. But they know that the targets are neurotransmitters and nerve receptors in the brain, and they have generated a few experimental results. For example, the effectiveness of SSRIs depends partly on variations in genes that control the synthesis of the serotonin transporter, which removes the neurotransmitter from the space between neurons. A variant of a serotonin nerve receptor gene may determine whether depressed patients treated with the SSRI paroxetine (Paxil) would improve. A variant of the opiate receptor gene may affect the likelihood that the opiate antagonist naloxone will help prevent relapse in alcoholics.

Genetic variations involving another neurotransmitter, dopamine, seem to affect the likelihood that a patient taking an antipsychotic drug will gain weight or suffer other side effects. One study found that a dopamine receptor gene variant predicted the therapeutic response to the novel antipsychotic drug clozapine. The dopamine transporter gene may influence responses to stimulants in children with attention deficit disorder.

According to another study, valproic acid (Depakote), one of the drugs used to treat bipolar disorder, neutralizes the effect of a single gene variant that contributes to the disorder. In a study of patients with Alzheimer’s disease, researchers found that those who carried the E4 variant of the apolipoprotein E gene are much less likely to respond to the anticholinesterase drug tacrine.

At present there are no tests that are clinically practical — that is, inexpensive, easy to administer, and easy to interpret — for a large number of genes and drugs. Interactions among genes are complicated, so the effects of one variation may be modified or neutralized elsewhere in the genome. Still, some scientific prophets are imagining a day not far in the future when physicians will prescribe drugs for their patients with the help of a DNA profile.

Some of the advantages are obvious. Drugs and doses will no longer have to be chosen by lengthy and frustrating trial and error. It will be easier to get the correct dose of the correct drug to the patients who need it when they need it. And drug treatment will become safer. Today, adverse reactions to drugs cause an estimated 100,000 hospitalizations and many deaths each year. Although lack of genetic identification is not the only cause of the problem, even a little improvement in that figure would make a big difference.

There’s more. Neglected drugs may be revived if they are found to be useful for people with a certain genetic constitution. Pharmacogenomics could also take some of the guesswork out of developing new drugs. Clinical trials could be quicker and less expensive if they require fewer participants because many people have been screened out genetically at an early stage. That could lead to faster approval of new drugs. Companies that develop new drugs are already storing blood samples from patients in their clinical trials for possible later genetic analysis.

Transforming psychiatry?

Some psychiatrists are saying that their specialty has even more to gain than other fields of medicine. Different combinations of genes interacting in different ways with one another and various features of the environment may cause the same apparent symptoms. That’s true of many complex conditions, but in psychiatry there are no objective measures like blood pressure or cholesterol. Diagnoses are based on patient reports and the symptoms and behavior observed by the professionals who treat them. A psychiatric diagnosis may not reflect the same underlying genetically distinct disorder or even group of disorders in different individuals.

This inadequacy in the classification of mental illnesses is probably the most important reason why individuals respond so differently to psychiatric drugs. So pharmacogenomics could contribute to a major change in the psychiatric diagnostic system. Genetic descriptions of individual drug responses may provide insight into biological mechanisms that can’t be directly observed. It could be a starting point for reclassifying and redefining mental illnesses — a revolution in psychiatry.