Expert Panel Debates Benefits, Harms of Pharmacogenetic Testing


At the start of the APA Expert Debate, “Should I Be Using Genetic Testing to Guide Prescribing?,” on Monday an in-house survey revealed that 80 percent of the audience believed that genetic testing was not an effective prescribing guide.

James Kennedy, M.D., a professor and co-director of the Brain and Therapeutics Division in the Department of Psychiatry at the University of Toronto, hoped to make a case in defense of those other 20 percent.

“Genetic testing is far from perfect,” he agreed, “but I think that testing is useful in psychiatric practice.”

He qualified that statement to note that it applied to pharmacogenetics, that is, the genetic information that can tell you how fast an individual might metabolize a drug or how well an individual’s receptors can bind it. Testing for disease risk, he admitted, was not ready for primetime. But when adverse drug reactions account for around 5 percent to 7 percent of all hospital admissions in the United States, he said, an opportunity to reduce that even marginally would help many patients and health systems as a whole.

Over the course of the Learning Lab session, Kennedy and his colleagues on the panel—Chadi Calarge, M.D., Erika Nurmi, M.D., Ph.D., James McCracken, M.D., and Michele Pato, M.D.—fielded a range of questions from the audience on issues not just scientific, but also ethical, judicial, and financial.

For example, one attendee wondered whether the typically modest increase in probability that a drug won’t cause a bad reaction was worth the average $2,000 price tag for a genetic test. McCracken, the director of child and adolescent psychiatry at the University of California, Los Angeles (UCLA), did point out that one night at UCLA’s Resnick Neuropsychiatric Hospital also runs around $2,000, so the upfront costs of gaining some more patient information might balance out in the long run.

Another clinician wondered how much value a genetic test for drug metabolism really has at the individual level, since it provides one static piece of information that doesn’t factor in gender effects, subtle differences in generic versus brand-name formulations, patient lifestyle, or potential drug-drug interactions for polypharmacy patients.

Kennedy agreed those were all valid concerns. He noted that genetic tests are not going to signal the start of some automated computer prescribing. “It is important to stress that a genetic test is not a be-all-end-all answer; it is just one small part of a physician’s decision-making process.”

As the session drew to a close—with perhaps a few participants shifting sides—Nurmi, also a professor at UCLA, pointed out that pharmacogenetics is not necessarily some brave new world. “We have been doing this intuitively for years,” she said.

“Maybe a patient who is not responding has low plasma levels of the medication even though the patient is taking it regularly, so we up the dose because it’s likely they are metabolizing it quickly. Or we decide to lower a dose as a patient gets older because we know their liver is probably slowing down,” she said.

The only difference now is clinicians get the news a little ahead of schedule.

(Image: kentoh/Shutterstock)