What Pharmacogenetic Testing Can, Can’t Tell You About Your Patient

A 15-year-old girl presents with depression and obsessive-compulsive disorder. You suggest fluoxetine to treat both conditions, given its extensive testing in young people with depression; however, her mom tells you her daughter can’t take fluoxetine “because of her genes.” She hands you the print-out of a commercial genetic test indicating that fluoxetine should only be used “with increased caution and more frequent monitoring.”

If you haven’t encountered a scenario like this yet, you may soon. Pharmacogenetics—the use of genetic testing to help predict efficacy and/or side effects of medications—is an advancing field that has garnered much publicity, assisted in part by former President Barack Obama’s Precision Medicine Initiative. Online commercial entities offering pharmacogenetic testing claim improved health outcomes, reduced risk of side effects, and financial savings from avoiding trials of ineffective medications.

Some psychiatrists with expertise in genetics and pharmacology have a counter message for clinicians and the public: Not so fast.

“I believe in the promise of pharmacogenetics, but I also believe we are a long way off from being able to use it in a way that will really aid clinicians and benefit patients,” said Erika L Nurmi, M.D., Ph.D., during a presentation at the American Academy of Child and Adolescent Psychiatry’s Pediatric Psychopharmacology Update last month in New York. “There is an evidence base for drug response based on genetic testing, but it is far behind public enthusiasm and the marketing claims of commercial companies.”

Nurmi is the associate medical director of the Child OCD, Anxiety, and Tic Disorders Program in the UCLA Department of Psychiatry and deputy chief of mental health for the Greater Los Angeles VA. She studies psychiatric genetics and pharmacogenetics in her laboratory at the Semel Institute for Neuroscience and Human Behavior at UCLA.

“People are clearly ready for pharmacogenetics,” she said, but the technology has not advanced enough for everyday clinical use. Important questions remain, including the following: Does the test accurately detect the genetic variant it claims in real-world settings? Does the test usefully change clinical management and improve health outcomes?

Nurmi said the association of a genetic variant with a specific phenotype affecting drug response is often oversimplified, ignoring the many factors that influence genetic expression. At the clinical level there are a range of factors—age, ethnicity, concomitant medication use, and history of substance use—that can affect drug response.

She said there has been a dearth of independent, randomized, controlled trials studying the value of prescribing informed by pharmacogenetic testing. Studies that exist tend to be supported and analyzed almost entirely by commercial entities, using proprietary algorithms for clinical decision-making and small sample sizes.

Existing guidelines on pharmacogenetic testing, such as those of the international Clinical Pharmacogenetics Implementation Consortium, are few and rely on theoretical interpretations of the limited data. A work group of the APA Council on Research is studying the use of pharmacogenetic testing in psychiatry.

Other experts on the subject corroborate Nurmi’s caution. “Pharmacogenetics is very au current,” Joseph Goldberg, M.D., a clinical professor of psychiatry at Icahn School of Medicine at Mount Sinai, New York, told Psychiatric News. “However, the present technology really just helps to affirm that someone may need a lower than usual medication dose to minimize certain side effects, or a higher than usual dose if they metabolize a drug too quickly. Mass media and test manufacturers seem to foster the misperception that pharmacogenetic testing will magically predict which drug works best for a given patient. The notion that current pharmacogenetic testing can tell patients which drug will work for them as if it were a crystal ball is a gross oversimplification of a complex medical decision-making process.”

Most Tests Assess Metabolization

The promise of pharmacogenetic testing and precision medicine as envisioned in Obama’s initiative is real. In oncology, where the science is most advanced, pharmacogenetic testing has vastly improved clinical outcomes. Such tests could one day yield enormous benefits in psychiatry, where response to psychopharmacology is highly variable and it can take months to determine whether the medication is helping a patient.

A recent report in AJP found that one genetic variant may have a role in predicting which patients will improve with antidepressants and which type of antidepressant may be most effective, adding to the foundational knowledge needed to advance a precision approach to personalized antidepressant choices.

What can pharmacogenetic testing today reliably tell you, and, more importantly, what can’t it tell you?

Nurmi explained there are two broad domains affecting drug response that could be influenced by an individual’s genetic profile: pharmacokinetics (what the body does to a drug, especially how quickly or slowly it metabolizes a medication) and pharmacodynamics (what the drug does to the body’s cells, tissues, and organs).

The bulk of available commercial testing is around pharmacokinetics. Genetic variations affecting the expression of cytochrome P450 (CYP) enzymes are known to affect whether the body metabolizes a particular drug quickly or slowly. There are a couple of clinical possibilities that might be indicated by testing CYP:

• A patient with a variation indicating unusually fast metabolization (an ultra-rapid metabolizer) may metabolize the drug in question so quickly that it does not reach therapeutic levels in the bloodstream at a normal dose.

• A patient with a variation indicating unusually slow metabolization (a poor metabolizer) would be likely to process the drug so slowly that it builds up in the bloodstream causing adverse effects at a normal dose.

Patients who order tests online may bring results into clinicians’ offices indicating, for instance, that a medication “should be used with caution and increased monitoring.” It is up to physicians to read test results more closely to determine what they indicate about metabolizer status.

Variations in expression of several CYP enzymes can cause poor or ultra-rapid metabolization of common psychiatric drugs. Two in particular—CYP2D6 and CYP2C19—affect the metabolization of common selective serotonin reuptake inhibitors (SSRIs), serotonin-norepinephrine reuptake inhibitors (SNRIs), tricyclic antidepressants, and atypical antipsychotics. For instance, it has been found that citalopram, which is metabolized by the CYP2C19 enzyme, may cause an increased risk of QT prolongation in poor metabolizers. For this reason, the Food and Drug Administration (FDA) has recommended that citalopram not be prescribed at doses above 20 mg/d in individuals identified as poor metabolizers. Electrocardiogram (ECG) monitoring and/or electrolyte monitoring should be performed if citalopram must be used in such patients.

With regard to pharmacodynamics, pharmacogenetic test results have only a handful of established correlates with clinical outcomes. Notably, a 2004 report in Nature found that the HLA-B1502 genetic mutation (common in individuals of Asian descent) is linked to risk for Stevens-Johnson Syndrome, a systemic reaction that involves a severe skin rash, when prescribed carbamazepine (sometimes used for seizures and aggression). For that reason, the FDA added a boxed warning requiring that people of Asian descent be screened for HLA-B1502 before starting carbamazepine.

Nurmi said research will invariably produce risk genes for mental illness that can be translated for pharmacogenetic testing in the clinic.

In the meantime, clinicians should educate themselves and be prepared to talk to patients and families about pharmacogenetics. “Prior to responsible implementation, the health care field needs to be able to understand, explain, and respond to genetic information provided to guide clinical care,” she said. ■