Conference Coverage

Routine pharmacogenetic testing in psychiatry not indicated



Someday, pharmacogenomics will advance precision psychiatry, but in the opinion of Erika L. Nurmi, MD, PhD, routine use of genetic testing to guide clinical treatment decisions is not indicated.

“It’s misleading to rely on results of genetic tests to drive clinical treatment,” Dr. Nurmi, a child and adolescent psychiatrist in the department of psychiatry and biobehavioral sciences at the University of California, Los Angeles, said during an annual psychopharmacology update held by the Nevada Psychiatric Association. “There’s a lot of hope and promise there. But currently, we only know the tip of the iceberg about how drugs work and the genetics influencing these effects. Current testing is probably a very poor reflection of the complexity of drug effects.”

Dr. Erika L. Nurmi, child and adolescent psychiatrist in the department of psychiatry and biobehavioral sciences at the University of California, Los Angeles

Dr. Erika L. Nurmi

According to Dr. Nurmi, there are at least 165 Food and Drug Administration–approved drugs with pharmacogenetic information on 64 different biomarkers – 37% with CYP p450 notations. Of these, 32 psychiatric drugs have pharmacogenetic information, and most of them are dosing recommendations based on whether a patient has the variant. However, there is wide public acceptance of genetic testing in preventing the wrong drug from being used, in selecting the best drug dose, and avoiding side effects (Pharmacogenomics 2012;12[3]:197-204). “Most people have a lot of hope [for genetic testing in psychiatry],” Dr. Nurmi said. “But is the science really there? It doesn’t matter, because these companies are doing it, and you are being shown these reports from patients. Whether or not the science supports it, we’re going to have to interpret these reports and explain them to our patients – even if we don’t order them.”

Currently, she continued, clinicians practice trial and error prescribing where they might try one treatment in a class that they think that will work based on previous literature. If nothing works, they try another one. If that’s intolerable, they try a third treatment, and so on. “When we finally find the right treatment, it can take some time to get the dosing right,” Dr. Nurmi said. “So, it can take many months to get a child on the right medication. Precision treatment, on the other hand, would start off by taking a saliva or blood sample to get a printout that lets physicians know which drugs might be used with caution because they might lack efficacy at standard doses, which ones would likely have adverse effects at standard doses, and which are the best choices and what are the dosing recommendations for those choices. If we could get all the information to guide us, that would be a useful product, but right now, we don’t know enough to be able to make these determinations.”

Current evidence-based genetic testing supports a limited role for CYP2D6 and CYP2C19 genotyping because most psychiatric drugs are metabolized by those two enzymes. Poor metabolizers have two dysfunctional copies of the enzyme-encoding gene. This results in increased drug plasma levels with a potentially increased rate of adverse effects.

“Intermediate and extensive metabolizers usually have a normal phenotype, but you can also have ultrarapid metabolizers who have duplications or other enhancing mutations of the CYP gene,” Dr. Nurmi said. “This can result in lower bioavailability and possibly efficacy. Psychiatrists treat poor metabolizers and ultrarapid metabolizers all the time, because the variants are very common.” An estimated 10% of White people are poor metabolizers at the CYP2D6 gene while about 7% are ultrarapid metabolizers. At the same time, an estimated 20% of Asians, Africans, and Whites are poor metabolizers at the CYP2C19 gene. “So, you’re seeing a lot of this in your practice, and you’re probably changing dosing based on genetic differences in metabolism,” she said.

The only FDA pharmacodynamic treatment guideline is for the risk of Stevens-Johnson syndrome (SJS) with the use of carbamazepine. In a study of 44 patients with SJS, all were positive for the HLA-B*1502 variant, compared with 3% of carbamazepine-tolerant patients (Nature 2004;428[6982]:486). The frequency of carrying this variant is an estimated 1:10,000 among Whites and 1:1,000 among Asians. In 2007, the FDA recommended that patients of Asian ancestry should be screened for HLA-B*1502 prior to starting carbamazepine.

Genetic variation also predicts clinical outcome with atomoxetine use. “Most child psychiatrists I know think atomoxetine doesn’t work as a second-line nonstimulant medication for ADHD,” Dr. Nurmi said. “I’d like to convince you that why you think it doesn’t work is because of the genetics.” In a study published in 2019, Dr. Nurmi and colleagues reviewed medical literature and provided therapeutic recommendations for atomoxetine therapy based on CYP2D6 genotype (Clin Pharmacol Ther 2019 Jul;106[1]:94-102). They observed 10- to 30-fold plasma differences in drug exposure between normal metabolizers and poor metabolizers.

“Poor metabolizers therefore get more benefit, but they are also going to get more side effects,” she said. “FDA recommended doses are inadequate for normal metabolizers, so they had to make guidelines based on poor metabolizers because there would be too much risk for them at higher doses. One-third of individuals require doses above the FDA limit to achieve a therapeutic drug level.”

Dr. Nurmi warned that the existing evidence base for using these genetic tests in children “is really poor. There is no data in adults with any diagnosis other than depression, and even those studies are plagued by concerns. When you’re implementing decision support tools in your practice, the key factors are patient presentation, history and symptoms, your clinical skills, the evidence base, FDA recommendations, and patient autonomy. Appropriate incorporation of genetic data should include avoiding a medication with high toxicity (like SJS), titration planning (dose and titration speed adjustments), and choosing between medications in the same class with an indication or evidence base for the target disorder.” She added that while the benefit of current genetic testing is limited, it may help some patients feel more comfortable tolerating a medication. “For example, being able to tell someone with anxiety that their genetics suggests that they will not have side effects could be very powerful,” she said.

In a 2018 safety communication, the FDA warned the public about its concerns with companies making claims about how to use genetic test results to manage medication treatments that are not supported by recommendations in the FDA-approved drug labeling or other scientific evidence. The American Academy of Child and Adolescent Psychiatry also published a guide for patients and families.

Dr. Nurmi disclosed that she is an unpaid advisory board member for Myriad Genetics and the Tourette Association of America, a paid adviser for Teva Pharmaceuticals, and a recipient of research support from Emalex Pharmaceuticals. She has received research funding from the National Institutes Health, the International OCD Foundation, the Tourette Association of America, and the Brain & Behavior Research Foundation.

Recommended Reading

Pandemic-stressed youths call runaway hotline
MDedge Psychiatry
Tips for managing youth with substance use disorders
MDedge Psychiatry
‘Robust’ increase in tics during the pandemic explained?
MDedge Psychiatry
Debate heats up on how best to treat gender-questioning kids
MDedge Psychiatry
What is the psychological impact of type 1 diabetes?
MDedge Psychiatry