What Therapists Should Know About Pharmacogenomic Testing
A client comes to session holding a printout. Her psychiatric nurse practitioner swabbed her cheek two weeks ago, and the report sorts a long list of medications into colored groups. She wants to know whether this means her sertraline was "wrong for her genes" all along, and whether she should stop taking it.
Therapists, counselors, and social workers hear questions like this more often now, usually before the client has had a full conversation with the prescriber. You won't order the test or change a dose, but you are often the person the client trusts to help them make sense of it. That makes it worth knowing what these reports measure, what the research shows, and where the limits sit.
Image by vh-studio / Freepik
What the Test Measures
Most pharmacogenomic panels used in psychiatry look at genes that code for liver enzymes, mainly CYP2D6, CYP2C19, and CYP2B6. Variants in these genes change how quickly a person breaks down many antidepressants. Labs report the result as a metabolizer status, ranging from poor to ultrarapid. A poor metabolizer may build up higher drug levels on a standard dose and have more side effects, while an ultrarapid metabolizer may clear the drug before it has much effect.
Commercial panels marketed as genetic testing for antidepressants combine these results into a single report for the prescriber. GeneSight, from Myriad Genetics, is one example. According to the company, the test must be ordered by a clinician who can prescribe, uses a cheek swab collected in the office or at home, and usually returns results to the provider in about three days.
Some of these gene-drug links appear in FDA drug labeling. The agency's Table of Pharmacogenetic Associations notes, for example, that CYP2C19 poor metabolizers have higher citalopram concentrations and a greater risk of QT prolongation, with a maximum recommended dose of 20 mg. It lists a 10 mg maximum for vortioxetine in CYP2D6 poor metabolizers. The FDA also states that including a gene-drug pair in the table does not necessarily mean it advocates testing before prescribing.
What the Evidence Shows
The Clinical Pharmacogenetics Implementation Consortium (CPIC), an international group of clinicians and scientists that publishes peer-reviewed prescribing guidelines, updated its guidance on serotonin reuptake inhibitor antidepressants in 2023. It gives recommendations for using CYP2D6, CYP2C19, and CYP2B6 results to inform dosing and drug choice. It also reviewed two genes some panels include, SLC6A4 (the serotonin transporter) and HTR2A (a serotonin receptor), and concluded that the existing data do not support their clinical use in antidepressant prescribing.
Outcome trials show modest benefits. The GUIDED trial, published in 2019 in the Journal of Psychiatric Research, enrolled 1,167 outpatients with major depression who had an inadequate response to at least one antidepressant. At week 8, average symptom improvement in the test-guided group was not significantly different from usual care (27.2% versus 24.4%), but response rates (26.0% versus 19.9%) and remission rates (15.3% versus 10.1%) were significantly higher. The trial was industry-sponsored, and some authors worked for the company behind the test.
The PRIME Care trial in JAMA, published in 2022, was larger, with 1,944 patients at 22 Veterans Affairs medical centers who were starting or switching an antidepressant. Giving clinicians test results reduced prescriptions of medications with predicted drug-gene interactions. Remission rates over 24 weeks were somewhat higher in the tested group, but the difference was no longer significant at week 24. The authors described the effect on remission as small and nonpersistent.
Some clinicians who read these papers themselves now use AI tools to read mental health research faster as a first pass before checking the methods and limitations directly.
What a Result Cannot Tell You
A pharmacogenomic report does not diagnose anything, and it does not predict which antidepressant will work. It estimates how a person is likely to metabolize certain drugs, which is one factor among many. Diagnosis, symptom profile, prior response, other health conditions, cost, and client preference all shape the choice.
Other medications also matter. Some drugs inhibit the same liver enzymes, so a client with a "normal" genotype can process a medication more like a poor metabolizer when taking them together. The report reflects genes, not everything else the client is taking.
Color categories are easy to misread. A medication in a caution group is not forbidden, and a medication in the preferred group is not guaranteed to help. Clients who read the report on their own sometimes conclude that a drug they have taken for years was harming them, and that conclusion is worth talking through, with the prescriber involved, before they act on it.
Talking With Clients About Testing
Stay within scope. Your job is to help the client understand the general idea, prepare questions, and bring them back to the prescriber for medication decisions. A useful framing is that the test gives the prescriber information about how the body handles certain medications, and the prescriber weighs it alongside everything else.
A few questions can help clients get more from the appointment: What did my results show for the medication I take now? Would you change anything based on this? How will we know if a change is helping? Encourage clients never to stop or adjust a medication on their own after reading a report.
With a signed release, a short call to the prescriber can clear up confusion quickly and keep care coordinated. It also helps to know the broader treatment picture, and courses such as treating depression training cover how psychotherapy and pharmacotherapy fit together.
Clients may also ask about privacy. The federal Genetic Information Nondiscrimination Act limits how health insurers and employers can use genetic information, but it does not cover life, disability, or long-term care insurance, so it is reasonable for clients to ask the prescriber how results will be stored and shared.
Treat a pharmacogenomic report as one input to a medication conversation, keeping in mind that much of the research has focused on clients who already had a poor response to at least one antidepressant, and send dosing questions back to the person who writes the prescription.