12:00 AM
Jul 14, 2008
 |  BioCentury  |  Regulation

Presumption of guilt

After a meta-analysis of more than 43,000 subjects in placebo-controlled trials of 11 antiepileptic drugs showed an increase in suicidality, FDAwas considering adding a black box warning to all epilepsy drugs. A joint FDA advisory committee last week voted 14-4 with three abstentions against the black box. But it also voted in favor of an unboxed warning, even for drugs with no suicidality data.

The members of the Peripheral and Central Nervous System Drugs Advisory Committee (PCNS) and Psychopharmacologic Drugs Advisory Committee (PDAC) were concerned a boxed warning would scare patients away from essential medications, a lesson learned from adding black box warnings to antidepressants for adolescents in 2004.

Last week’s panel hoped that an unboxed warning and Medication Guides would encourage physicians and patients to discuss the risks.

Whichever type of warning the agency chooses, the panel voted 15-5 with one abstention that it should apply to all approved chronically administered antiepileptic drugs (AEDs), including three drugs that did not show an increased risk of suicidality in the meta-analysis and at least 14 that were not included in the meta-analysis.

Although the “no” voters expressed concern about extrapolating data to drugs that were not studied, the majority agreed that limiting the warning to the 11 drugs included in the analysis could drive physicians to prescribe older, less-studied drugs.

The panel agreed there’s no plausible mechanism to explain the safety signal across the drugs in the meta-analysis, which have at least three different MOAs. Rather than concluding that lack of scientific evidence should drive the agency to consider warnings on a drug-by-drug (or MOA-by-MOA) basis, the joint committee concluded that an increase in suicidality would likely be seen in all AEDs, regardless of MOA.

FDA’s concerns that a warning may scare patients away from drugs may turn out to be overblown, as epileptics have few alternatives for treatment. But some AEDs could see sales fall because they are used primarily in other indications for which patients do have alternatives.

Seeking signals

In early 2005, FDA decided to investigate the question of suicidality for AEDs after a manufacturer presented data that the company believed showed an increase in suicidality in controlled trials of its drug compared to placebo. FDA could not provide the name of the company.

The agency then requested data from all randomized, parallel-arm, placebo-controlled trials of AEDs with at least 20 patients in all treatment arms. In its request, FDA excluded trials that lasted less than one week, had subjects under five years old and/or had randomized withdrawal designs.

A total of 199 trials of 11 AEDs met the inclusion criteria. The trials involved 43,892 subjects, including 27,863 on active treatment and 16,029 on placebo.

Sponsors were asked to identify potential suicidality events, for which narrative descriptions were constructed and classified by a blinded reviewer according to the Columbia Classification Algorithm of Suicide Assessment (C-CASA).

C-CASA includes three classifications that are considered suicidal behavior: completed suicide, suicide attempt, and preparatory acts toward imminent suicidal behavior. Suicidal behavior plus a fourth classification of suicidal ideation are considered suicidality, which was the primary endpoint...

Read the full 2545 word article

User Sign in

Trial Subscription

Get a 4-week free trial subscription to BioCentury

Article Purchase

$150 USD
More Info >