, new research shows. “There appears to be no justification for the FDA to label every new antiseizure medication with a warning that it may increase risk of suicidality,” said study investigator Michael R. Sperling, MD, professor of neurology, Thomas Jefferson University, Philadelphia.
“How many patients are afraid of their medication and do not take it because of the warning – and are consequently at risk because of that? We do not know, but have anecdotal experience that this is certainly an issue,” Dr. Sperling, who is director of the Jefferson Comprehensive Epilepsy Center, added.
The study was published online August 2 in JAMA Neurology.
In 2008, the FDA issued an alert stating that antiseizure medications increase suicidality. The alert was based on pooled data from placebo-controlled clinical trials that included 11 antiseizure medications – carbamazepine, felbamate, gabapentin, lamotrigine, levetiracetam, oxcarbazepine, pregabalin, tiagabine, topiramate, valproate, and zonisamide.
The meta-analytic review showed that, compared with placebo, antiseizure medications nearly doubled suicide risk among patients treated for epilepsy, psychiatric disorders, and other diseases. As a result of the FDA study, all antiseizure medications that have been approved since 2008 carry a warning for suicidality.
However, subsequent analyses did not show the same results, Dr. Sperling and colleagues noted.
“Pivotal” antiseizure medication epilepsy trials since 2008 have evaluated suicidality prospectively. Since 2011, trials have included the validated Columbia Suicidality Severity Rating Scale, they noted.
Meta analysis showed no increased risk
Dr. Sperling and colleagues conducted a meta-analysis of 17 randomized placebo-controlled epilepsy trials of five antiseizure medications approved since 2008. These antiseizure medications were eslicarbazepine, perampanel, brivaracetam, cannabidiol, and cenobamate. The trials involved 5,996 patients, including 4,000 who were treated with antiseizure medications and 1,996 who were treated with placebo.
Confining the analysis to epilepsy trials avoids potential confounders, such as possible differences in suicidality risks between different diseases, the researchers noted.
They found no evidence of increased risk for suicidal ideation (overall risk ratio, antiseizure medications vs. placebo: 0.75; 95% confidence interval: 0.35-1.60) or suicide attempt (risk ratio, 0.75; 95% CI: 0.30-1.87) overall or for any individual antiseizure medication.
Suicidal ideation occurred in 12 of 4,000 patients treated with antiseizure medications (0.30%), versus 7 of 1,996 patients treated with placebo (0.35%) (P = .74). Three patients who were treated with antiseizure medications attempted suicide; no patients who were treated with placebo attempted suicide (P = .22). There were no completed suicides.
“There is no current evidence that the five antiseizure medications evaluated in this study increase suicidality in epilepsy and merit a suicidality class warning,” the investigators wrote. When prescribed for epilepsy, “evidence does not support the FDA’s labeling practice of a blanket assumption of increased suicidality,” said Dr. Sperling.
“Our findings indicate the nonspecific suicide warning for all epilepsy drugs is simply not justifiable,” he said. “The results are not surprising. Different drugs affect cells in different ways. So there’s no reason to expect that every drug would increase suicide risk for every patient,” Dr. Sperling said in a statement.
“It’s important to recognize that epilepsy has many causes – perinatal injury, stroke, tumor, head trauma, developmental malformations, genetic causes, and others – and these underlying etiologies may well contribute to the presence of depression and suicidality in this population,” he said in an interview. “Psychodynamic influences also may occur as a consequence of having seizures. This is a complicated area, and drugs are simply one piece of the puzzle,” he added.
Dr. Sperling said the FDA has accomplished “one useful thing with its warning – it highlighted that physicians and other health care providers must pay attention to their patients’ psychological state, ask questions, and treat accordingly.”
The study had no specific funding. Dr. Sperling has received grants from Eisai, Medtronic, Neurelis, SK Life Science, Sunovion, Takeda, Xenon, Cerevel Therapeutics, UCB Pharma, and Engage Pharma; personal fees from Neurelis, Medscape, Neurology Live, International Medical Press, UCB Pharma, Eisai, Oxford University Press, and Projects in Knowledge. He has also consulted for Medtronic outside the submitted work; payments went to Thomas Jefferson University. A complete list of authors’ disclosures is available with the original article.
A version of this article first appeared on.