Pharmacogenetic Testing as a Tool in Bipolar Disorder Treatment

(Part 4 of 4)

In this video, Joseph F. Goldberg, MD, explains the utility and limitations of pharmacogenetics in the treatment of bipolar disorders. He also discusses the concept in his session titled "Bespoke Psychopharmacology: Tailoring Individualized Pharmacotherapy for Patients with Bipolar Disorder" being presented at the 2021 Psych Congress Regionals meeting series, which continues April 23 and 24 and May 21 and 22.

Dr. Goldberg, Clinical Professor of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, New York, is a Bipolar Disorders Section Editor for the Psychiatry and Behavioral Health Learning Network.

Pharmacogenetics, we can divide into really 2 main domains. One, we can call safety pharmacogenetics. The other, we can call efficacy pharmacogenetics. Safety bears on tolerability issues, and efficacy, as it implies, would tell us things about is a drug likely to work? Let me say a word or two about both.

We know a lot more about safety pharmacogenetics, in general, than efficacy pharmacogenetics in psychiatry. Safety pharmacogenetics will tell us things like if someone has the genes that code for an enzyme in phase I or phase II metabolism in the liver, that corresponds to someone either being a poor metabolizer or an ultra-rapid metabolizer of certain drug substrates.

That'll have implications about the likelihood of seeing certain side effects, or even seeing an end product. For example, in phase I­ metabolism, cytochrome P450 enzymes, about 6 to 10 percent of the population would count as a poor metabolizer of enzymes like 2D6 or 3A4.

If someone is in that population, if they're a poor metabolizer, say, of 2D6, and you give them risperidone, they're not going to be very efficient at metabolizing it to 9-hydroxyrisperidone, its active metabolite, and the parent compound's going to build up.

They may get more side effects. They also may not get much efficacy, if they can't make the parent into the metabolite compound. Aripiprazole is metabolized by 2D6 and 3A4. If someone was a poor metabolizer, again, of 2D6 or 3A4, there might be some buildup, a greater risk of side effects or tolerability issues.

We want to be thoughtful about someone who's complaining of a lot of side effects, looking back to see, well, are the medicines they're taking substrates for particular metabolic enzymes for which pharmacogenetic testing might affirm, indeed, this is someone who's a poor metabolizer?

Or if they're getting no effect, if there's a pathway involved, and we'd speculate about an ultra-rapid metabolizer, venlafaxine has to be converted to desvenlafaxine by 2D6. If I'm an ultra-rapid metabolizer of 2D6, I'm not going to make it to the metabolite compound.

That's one area in safety pharmacogenetics. A second has to do with specific genes that we know pose risk factors. Maybe the best one known is lamotrigine has a particular HLA haplotype, which if present in Southeast Asians, poses a significantly increased risk for developing Stevens-Johnson syndrome with carbamazepine.

There's another HLA gene that may pose a risk for Stevens-Johnson syndrome with lamotrigine. That, in some ways, is one of the best examples of safety pharmacogenetics. There's even a box warning. You cannot give carbamazepine to a South Asian individual without checking this HLA haplotype because of the risk for Stevens-Johnson syndrome.

Those are some examples of safety pharmacogenetics. Efficacy's not as well-established. That would be saying, if I have a particular candidate gene, can I anticipate whether the drug is going to work or not? We don't really have that yet.

Maybe one sort of direction there is in predicting the chance that an antidepressant might lead to a switch from depression to mania. There is a particular variation of the serotonin transporter gene that involves getting either a short or a long variation of the gene from each parent.

Some research says that, if you happen to inherit a short copy of that gene from each parent, you may have a higher chance of seeing an SSRI-induced mania or hypomania. These are all very preliminary studies. They're not quite yet ready to go out into the clinic, and let's genotype everybody before we give them an SSRI or before we make assumptions about drug effects like antidepressants, but these are some of the inroads that I think we're starting to make, hopefully to see greater inroads in the future, so we might be able to use pharmacogenetics to inform not just tolerability and safety, but efficacy as well.

More from the series:
Distinguishing Between Bipolar I Disorder and Bipolar II Disorder
Using Moderators and Mediators to Tailor Bipolar Disorder Treatment
Understanding Polarity-Proneness in Patients With Bipolar Disorder