Has Pharmacogenomic Testing Reached a Tipping Point?

A growing number of studies show that drug/gene interactions can influence how patients respond to specific medications. But there are barriers that prevent the full implementation of pharmacogenomic testing into routine clinical practice.

By John Halamka, M.D., president, Mayo Clinic Platform, and Paul Cerrato, senior research analyst and communications specialist, Mayo Clinic Platform

We have written often about the benefits of pharmacogenomic (PGx) testing in our blogs and books. The field explores the ways in which a person’s genetic makeup influences their response to specific medications. A mutation in the liver cytochrome P-450 enzyme system, for instance, can alter the way a person metabolizes a variety of drugs, which in turn can cause the drug to be toxic, or ineffective, depending on the specific genetic variant.

A new clinical trial published in JAMA lends further support to the view that these services may need to be made more widely available to patients taking psychotropic medication.

A close look at the literature demonstrates the need for better medication management in this patient population. Less than a third of patients with major depressive disorder (MDD), bipolar disease, and schizophrenia continue to take their medication and reach full remission, and between 30% and 50% of patients with MDD respond to the first antidepressant they are prescribed. Additionally, 25,000 U.S. patients go to the emergency department each year with antidepressant side effects. One reason for the poor compliance and adverse effects is the variability among individuals in the way they metabolize psychotropic drugs. Most are processed by three liver enzymes: CYP2D6, CYP2C19, and CYP3A4. There are several genetic variants of these enzymes and there is no easy way to tell which patients have which variants.

With these concerns in mind, David Oslin, MD, with the Department of Psychiatry, University of Pennsylvania, and his colleagues conducted a randomized clinical trial that included nearly 2,000 patients taking antidepressant medication for MDD. Pharmacogenomic (PGx) testing was performed on half the group while the second half received their usual care. Oslin et al found that patients who had undergone PGx testing were significantly less likely to be prescribed a medication with a potential drug-gene interaction.  However, the impact of PGx testing on symptom relief was less impressive. During the 24 weeks of the trial, remission rate for patients who had been tested was about 28% better when compared to patients not tested, but that difference was not significantly better at week 24.

Based on these results, critics will no doubt question the need for routine PGx testing among patients on prescription medications, but the trial suggests one logical reason for the results:  PGx testing can only uncover potential drug/gene interactions. If an individual patient does not have the genetic variant that is known to interact with the drug in question, it’s unlikely that switching medications will have an impact of their clinical course. In other words, the only patients likely to benefit from PGx testing are those who have a detectable drug gene metabolic interaction.  And one of the best way to detect such interactions is to routinely perform whole genomic sequencing to find the variants that put patients at risk. 

Although Oslin et al had mixed results during their trial, others have seen more encouraging results.  One trial found a “significant increase in responders after 8 weeks of antidepressant therapy when genetic information on CYP2D6, CYP2C19, CYP1A2, SLC6A4, and HTR2A was used to dose patients (n =114) as compared to standard treatment (n = 113).”  A meta-analysis of 5 RCTs that looked at the value of PGx testing confirmed these benefits, concluding that pharmacogenetic-guided therapy was about 70% more effective at achieving symptom remission, compared to standard therapy.

PGx testing has been an active area of research at Mayo Clinic as well. Eric Matey, a pharmacist at Mayo Clinic in Rochester, MN, and his colleagues recently performed genotyping in 82 healthy adult volunteers and found 93% were CYP1A2 rapid metabolizers while 88% were CYP3A5 poor metabolizers. The pre-emptive PGx testing enabled clinicians to provide better medication management to over half the participants. The investigators also suggested that adding PGx test results to a patient’s EHR and developing clinical decision support rules to alert providers to these genetic variants would likely improve patient outcomes.

On the other hand, Paul Croarkin, D.O., M.S., with the Mayo Clinic Department of Child Psychiatry, pointed out that many pragmatic studies have been conducted in this field that found no differences in clinical outcomes among depressed patients when drug/gene interactions were taken into consideration. “These tests have value but not always in the way they are marketed by the laboratories that offer them. With good intentions, they lump numerous drug/gene interactions together in one test panel, which makes the results difficult to defend scientifically. Even when these panels are used to study large patient cohorts, the results are very often underpowered statistically for any one drug/gene interaction. A better approach would be to study a single drug/gene interaction in psychiatric patients and also include several other variables in the study, including age, gender, family history and so on.”