乌鸦传媒

While use of pharmacogenomic testing reduced antidepressant prescriptions with known drug-gene interactions in patients with major depressive disorder (MDD), the effects on remission did not persist over time, according to a randomized trial.

Among over 1,900 patients, those in the pharmacogenomic-testing group were significantly less likely to be prescribed antidepressants with moderate (30% vs 54.6% in usual care) or substantial (10.7% vs 19.7%) drug-gene interactions, reported David Oslin, MD, of the Corporal Michael J. Crescenz VA Medical Center at the University of Pennsylvania in Philadelphia, and colleagues in .

Remission rates across 24 weeks were also higher in the testing group (OR 1.28, 95% CI, 1.05-1.57, P=0.02). But while differences between the testing and usual care groups, respectively, were greatest at 12 weeks, no significant difference was observed at 24 weeks:

• 12 weeks: 16.5% vs 11.2%, P=0.001
• 24 weeks: 17.2% vs 16%, P=0.45

Oslin told 乌鸦传媒 that despite the benefits seen in prescribing practices, these results are not a complete endorsement for pharmacogenomic testing.

"This study certainly doesn't tell you not to test. It tells you that there is some benefit for some patients, and that if either the provider or the patient are interested in testing, then you should test. This does not mean the testing should be standard in practice, though," he noted. "These tests aren't diagnostic tests, they're really about how people metabolize medications. They're not telling you that this antidepressant is going to work better for you, it's telling you that you don't know how to dose this antidepressant very well."

Oslin explained that the goal of the study was to determine the efficacy of the clinical use of pharmacogenomic testing, which has shown commercial success in recent years.

One major takeaway from the study was that only 19.7% of patients in the usual care group were prescribed an antidepressant with a substantial drug-gene interaction, which could mean that pharmacogenomic testing is only meaningfully beneficial for about 20% of patients overall.

"Of course, you don't know who to test before you test them, so you have to test a lot of people to have a benefit," Oslin said. "But I think it's important for those patients and providers to understand this test isn't useful in every patient."

In an , Dan V. Iosifescu, MD, MSc, of New York University School of Medicine, noted that the study should be considered "an important addition to the literature on pharmacogenomic testing in patients with MDD," pointing out that it was "significantly larger" than previous studies on the subject and that it was "one of the few investigations not funded by a manufacturer of pharmacogenomic tests."

It also boasted a longer follow-up period than previous studies, though ultimately the study showed the limited clinical relevance of this testing, he added.

"Clinicians will have to accept the limited value of the pharmacogenomic testing for most of the patients with MDD, including the majority of patients with treatment-resistant depression," Iosifescu wrote.

For this pragmatic study, the researchers included 1,944 patients (mean age 48 years, 25% women) from 22 VA medical centers, who were enrolled from July 2017 to February 2021, with follow-up until November 2021. All patients had to have a previous diagnosis of MDD and a plan to initiate or switch treatment using a single antidepressant. Patients with an active substance use disorder, mania, psychosis, or concurrent treatment with a specified list of medications were excluded.

Of the patients, 966 were randomized to receive their pharmacogenomic test result and 978 were randomized to usual care with access to pharmacogenomic results after 24 weeks. The researchers measured remission of depressive symptoms using the Patient Health Questionnaire-9, with remission defined as ≤5.

Compared with usual care patients, those in the test group were more likely to receive an antidepressant medication without or with lower drug-gene interactions:

• No interaction vs moderate/substantial interaction: OR 4.32, 95% CI 3.47-5.39, P<0.001
• No/moderate interaction vs substantial interaction: OR 2.08, 95% CI 1.52-2.84, P=0.005

Oslin and colleagues acknowledged that the effects in the testing group could have been due to a "placebo-type effect," since the clinicians and patients were not blinded. Furthermore, the study was not powered to evaluate outcomes such as the effect of changes in dosing in the testing group among those with predicted drug-gene interactions, the presence of adverse reactions, the effect of medication adherence by patients, or the effect of switching antidepressants after randomization.

Comment