AMPA antagonist drug shows mettle in generalized tonic-clonic seizures.

Most patients with generalized tonic-clonic seizures (GTCSs) despite conventional therapy showed major improvement when perampanel (Fycompa) was added to their regimens in a phase III trial, researchers reported here.

Some 31% of patients achieved complete freedom from these extremely debilitating seizures after 17 weeks of treatment and about an equal number had at least a 50% reduction in GTCS frequency in the randomized, placebo-controlled trial, said Jacqueline French, MD, of New York University’s Comprehensive Epilepsy Center in New York City.

At a press briefing, French said the results were important because many drugs approved for less severe seizures are ineffective for GTCSs and may even worsen them. Perampanel, an antagonist of excitatory glutamate activity through so-called AMPA receptors, is currently approved for treating partial onset seizures.

For the current trial, French and colleagues enrolled 163 patients with idiopathic generalized epilepsy marked by GTCSs that did not respond to currently approved therapies. One patient in the placebo group was excluded from the efficacy analysis.

To be enrolled, patients had to have at least three GTCSs during an 8-week baseline observation period while taking from one to three approved therapies. Patients with vagal nerve stimulator implants were allowed into the study if the device had been used for at least 5 months prior to enrollment.

Individuals with progressive neurological disease, Lennox-Gastaut syndrome, history of recent status epilepticus, or use of rescue benzodiazepines more than twice in the 30 days prior to enrollment were excluded.

Perampanel was started at 2 mg/day and then increased in 2-mg increments to a target of 8 mg/day during a 4-week titration period. The drug was then continued for 13 weeks at either 8 mg/day or at the maximally tolerated dose if lower.

Primary endpoints were the responder rate and the reduction in seizure frequency from the level recorded during the initial 8-week baseline period.

As already noted, the responder analysis clearly favored perampanel. So did the other primary outcome: patients in the perampanel group showed a mean 76.5% reduction in GTCS frequency, compared with 38.4% with placebo (P<0.0001).

Complete seizure freedom was achieved by 30.9% of the perampanel group versus 12.3% of the placebo group (not significant).

A total of 23 patients did not complete the study, including 13 assigned to perampanel and 10 to placebo. Of these, 9 in the perampanel group and five of the placebo group stopped the assigned medication because of treatment-emergent adverse effects. There was no clear difference between treatment assignments in the types of adverse events leading to withdrawal, and rates of serious, severe, and fatal adverse effects were the same in the two arms.

Nonserious adverse events were somewhat more common with perampanel, particularly somnolence, dizziness, fatigue, and irritability.

Overall, French and colleagues indicated that perampanel’s side effect profile was similar to that seen in patients with partial-onset seizures.

Although the current trial tested the drug as an add-on, French said she expected that it would also be effective as monotherapy. However, the FDA is currently not approving drugs as monotherapies without trials specifically testing that use.

She said many in the epilepsy community believe that this policy is wasteful and unnecessary, because there is no reason to believe that, if a drug that reduces seizures in patients not responding to another approved therapy, it wouldn’t be effective as monotherapy.

She pointed out that taking patients off their current therapies — as could be required in a monotherapy trial — can be “catastrophic” for an epilepsy patient.

But without an approved monotherapy indication, neurologists and their patients face a dilemma when it is necessary to replace one drug with another. This situation commonly arises when a young epileptic woman becomes pregnant while taking a known teratogenic drug such as valproate. Switching this patient to a drug that is approved only as an add-on is thus, technically, off-label.

Consequently, she said, the field’s major professional societies — the AES and theInternational League Against Epilepsy — are currently drafting a white paper calling on the FDA to approve drugs as monotherapies on the basis of add-on studies.

In the meantime, perampanel’s manufacturer, Eisai, asked the FDA in August to approve an expanded indication covering adjunctive treatment of GTCSs, mainly on the basis of the current study.