Pharmacogenomic (PGx) Testing in Psychiatry: What It Can and Cannot Tell You
Short answer
Pharmacogenomic (PGx) tests look at genes, such as CYP2D6 and CYP2C19, that affect how your body processes some medicines. The FDA lists certain psychiatric drugs for which these genes can change blood levels or side-effect risk. The FDA has also warned that claims that a test can show which antidepressant will work are not supported. Results can inform a prescriber's decision, but they do not choose the medicine for you.
In crisis or thinking about suicide? Call or text 988 now (in New York City it reaches NYC 988). If you are in immediate danger, call 911.
This guide is not enough if
- You are thinking about suicide or harming yourself. Call or text 988, or call 911 if you are in danger now.
- You want to change or stop a medicine because of a test result. Talk with your prescriber first.
- You want to know whether your insurance covers a test. Ask your plan and the laboratory directly.
- You are looking for a diagnosis. Pharmacogenomic tests look at drug processing, not at whether you have a mental health condition.
Why people ask about genetic testing
Maybe you have tried two or three medicines and each one caused side effects or did not help. Someone mentions a cheek-swab test that is supposed to show which drugs suit your body. It is a fair thing to ask about, and it helps to know what these tests can and cannot do before you spend time or money on one.
MedlinePlus, from the National Library of Medicine, defines pharmacogenomics as the study of how genes affect a person's response to drugs. The site already has reasons some people consider genetic testing and an overview of which lab tests are used in integrative psychiatry. This guide focuses on what the FDA says, so you can weigh any test against an official yardstick.
What the tests look at
For psychiatric medicines, the genes that appear most often in the FDA's gene-drug table are CYP2D6 and CYP2C19, which code for enzymes in the cytochrome P450 (CYP) family. MedlinePlus says the CYP2C19 enzyme helps process at least 10 percent of commonly prescribed drugs, and it names antidepressants among the drug groups affected by differences in that gene.
Results usually sort people into groups by how active an enzyme is likely to be. The FDA's table uses terms such as poor, intermediate and ultrarapid metabolizers. For several psychiatric drugs, the table says poor metabolizers end up with higher drug levels in the body; for others, it says results in either direction may alter levels.
A different kind of result involves HLA genes. The FDA's table lists the HLA-B*15:02 variant with a higher risk of severe skin reactions to carbamazepine, oxcarbazepine and lamotrigine.
Genes are only part of the picture. Even for a drug with a well-known gene link, MedlinePlus notes that sex, age, weight, diet and other medications also play a role in how people respond.
What FDA's table says about psychiatric medicines
The FDA keeps a Table of Pharmacogenetic Associations. For the gene-drug pairs it lists, the FDA says there is sufficient scientific evidence that people with certain gene variants are likely to process the drug differently, and in some cases to have different effects or side-effect risks.
Psychiatric examples in the section where data support specific prescribing recommendations include citalopram (CYP2C19), and aripiprazole, brexpiprazole, atomoxetine, vortioxetine, venlafaxine and amphetamine (CYP2D6), each for poor metabolizers. Other drugs, such as escitalopram, paroxetine and several tricyclic antidepressants, appear in a section where the data show a possible effect on drug levels only.
Three caveats come straight from the FDA. Being listed does not necessarily mean the FDA advocates testing before prescribing, unless the test is a companion diagnostic. Most associations have not been evaluated for whether testing improves outcomes. And the table is meant mainly for prescribers: patients should not adjust their medications without consulting their prescriber.
| Type of result | What it can tell you | What it cannot tell you |
|---|---|---|
| Poor metabolizer for a drug in FDA's first section (for example, citalopram and CYP2C19) | The FDA label may call for a dose change or another drug | Whether that medicine will help your symptoms |
| A changed result for a drug in FDA's drug-levels-only section (for example, escitalopram) | Blood levels may differ from average | What, if anything, to change; the FDA lists a possible effect on levels only |
| HLA-B*15:02 positive | Higher risk of severe skin reactions with carbamazepine, oxcarbazepine or lamotrigine | That a reaction will happen; the FDA says genotyping is not a substitute for clinical vigilance |
| A ranking of antidepressants as good or poor matches | Mostly how genes may affect drug levels | Which antidepressant will work; the FDA says that link has never been established |
| No variants found | None of the tested variants were detected | That you will have no side effects; age, diet and other medicines still matter |
What FDA has warned about
On October 31, 2018, the FDA issued a safety communication warning against many genetic tests with unapproved claims to predict how a patient will respond to specific medications. It said the relationship between DNA variations and the effectiveness of antidepressant medication has never been established, and it told patients not to change or stop any medicine based on a report from a genetic test they took on their own.
The same communication told clinicians that for most medications, the relationship between DNA variations and the medication's effects has not been established, and to check the FDA-approved drug label for whether genetic information should guide treatment. It noted a limited number of cases where at least some evidence links a gene variant to drug levels in the body. When a label does describe how to use genetic information, the FDA said, that guidance appears in sections such as the Boxed Warning, Warnings and Precautions, or Dosage and Administration.
In April 2019, the FDA sent a warning letter to a genomics laboratory whose tests claimed to predict responses to specific named drugs, including antidepressants. The FDA said selecting or changing drug treatment in response to such results could lead to serious health consequences. Its page on direct-to-consumer tests also says it has not authorized any direct-to-consumer pharmacogenetic test that predicts whether a person is likely to respond to, or have side effects from, a specific drug.
Questions to ask before you test
A test is most useful when it is tied to a real decision, such as choosing between drugs that appear in the FDA's table or making sense of unexpected side effects. If you and your prescriber are considering a test, these questions keep the decision grounded:
- Which medicine decision would this result inform, and is that drug in the FDA's table?
- What would we do differently if the result came back poor, normal or ultrarapid?
- Does the report make claims about which drug will work, beyond drug levels and side-effect risk?
- How will the result be kept in my record and shared with future prescribers?
- What will it cost me? Ask your insurer and the laboratory before the sample is taken.
When testing is not the right next step
If you are thinking about suicide or harming yourself, do not wait for a test result. Call or text 988, or call 911 if you are in immediate danger; crisis and hotline resources lists more options.
Do not stop or skip a current medicine while you wait for results, and do not act on a home test alone. The FDA's 2019 warning letter noted that stopping antidepressants abruptly could lead to withdrawal symptoms or relapse. If you are just starting treatment, questions to ask before starting psychiatric medication and antidepressant classes compared may be more useful first steps. For help with medicine choices, see medication management or integrative psychiatry. If you are weighing other add-ons, read supplements and mental health.
Frequently asked questions
Can a genetic test tell me which antidepressant will work?
Not according to the FDA. In its 2018 safety communication, the FDA said the relationship between DNA variations and the effectiveness of antidepressant medication has never been established. A test can show whether you may process certain drugs faster or slower than average, which can help a prescriber think about dose and side-effect risk. Choosing a medicine still depends on your symptoms, history and response.
What does it mean to be a poor metabolizer?
It means the test suggests a particular enzyme, such as CYP2D6 or CYP2C19, is likely to have low activity in your body. For some drugs the enzyme breaks down, that can lead to higher blood levels. The FDA's table, for example, says CYP2D6 poor metabolizers taking aripiprazole have higher drug levels and higher side-effect risk, and that a dose adjustment is recommended. Your prescriber decides what to do with that.
Is a home DNA test enough to change my medicine?
No. The FDA told patients not to change or stop any medicine based on a report from a genetic test they took on their own, and to discuss results with a health care provider. The FDA also says it has not authorized any direct-to-consumer pharmacogenetic test that predicts response to a specific drug. Share any home report with your prescriber and let them interpret it.
Will my results still matter for future prescriptions?
They may. A result can come up again if you are later prescribed another drug in the FDA's table, so keep a copy and share it with new prescribers. The FDA also updates the table, most recently on September 10, 2026, so ask a prescriber to check an older report against current labeling rather than relying on the report's original color codes or rankings.
Next: compare your options
How the main antidepressant groups differ on FDA labels: likely mechanism, common side effects and key safety rules.
Sources
- FDA: Table of Pharmacogenetic Associations
- FDA: Direct-to-Consumer Tests
- FDA: FDA issues warning letter to genomics lab for illegally marketing genetic test that claims to predict patients' responses to specific medications (April 4, 2019)
- FDA: Warning letter to Inova Genomics Laboratory (April 4, 2019)
- MedlinePlus Genetics: What is pharmacogenomics?
- MedlinePlus Genetics: CYP2C19 gene
- FDA: Laboratory Developed Tests
Recent changes
- September 10, 2026: The FDA updated its Table of Pharmacogenetic Associations. Additions to the section with prescribing recommendations include bupropion and dextromethorphan (CYP2D6), doxepin (CYP2C19 and CYP2D6), lamotrigine (HLA-B*15:02) and milsaperidone (CYP2D6). FDA: Table of Pharmacogenetic Associations
- September 19, 2025: After a federal district court vacated the FDA's May 2024 final rule on laboratory developed tests on March 31, 2025, the FDA issued a final rule reverting the regulation's text to its earlier wording. FDA: Laboratory Developed Tests
Reviewed by Dr. Beata Lewis, MD, Founder and Clinical Director, board-certified in adult psychiatry, child and adolescent psychiatry, and integrative medicine on November 8, 2026. Facts checked against the sources above on November 8, 2026. This guide is general education, not medical advice, a diagnosis or a treatment plan, and reading it does not make you a patient of Mind Body Seven. For advice about your situation, talk with a licensed clinician.