Are the new pharmacogenomic assays of any clinical utility? If so, when would one be of greatest benefit in making a treatment decision?
Great question, complicated answer. NEI members can get a free test from Genomind to test drive this idea in one patient. Insurance is actually reimbursing some tests especially the cytochrome P450 enzymes. Testing is also available from other companies like AssureRx. A number of assays are now commercially available and emphasize P450 enzymes, and a number of neurotransmitter related genes, and are probably most advanced for treatment resistant depression application, and more for treatment than for diagnosis in general in psychiatry. The most common genes available for testing are:
CYP 2D6, 1A2, 2C9, 2C19, 3A4/5
SERT (the serotonin transporter), variation associated with poor efficacy and poor tolerability;
5HT2A receptor gene;
5HT2C receptor gene
CACNA1C, calcium channel gene
DRD2 dopamine receptor gene
ANK3 signal transduction cascade
COMT catechol O methyl transferase enzymeís gene
MTHFR Methylene tetrahydrofolate reductase
Not enough space here to explain these, but in brief, these are probably best established to help guide selection of treatment, not to guide diagnosis; and the best data are in treatment resistant depression where some of these gene variants predict poor efficacy or poor tolerability.
However, no gene tells you everything about a patient to dictate any specific treatment. This is about the balance of the evidence where the genomics help you make a decision and help make you more confident (and the patient more confident) about your decision.
Stephen M. Stahl, MD, PhD
Adjunct Professor, Department of Psychiatry,
University of California, San Diego School of Medicine
Honorary Visiting Senior Fellow, University of Cambridge, UK