Standard Migraine Drugs No Better Than Placebo in Kids

Pauline Anderson

October 27, 2016

A study of amitriptyline and topiramate in children with migraine has been halted early owing to futility.

The analysis of data from the Childhood and Adolescent Migraine Prevention (CHAMP) trial showed that neither of these preventive medications was more effective than placebo in reducing headache frequency or headache-related frequency, and both were associated with higher rates of adverse events (AEs).

Given the "null outcome" and the adverse events reported, "the data do not show a favorable risk-benefit profile for the use of these therapies in pediatric migraine prevention, at least over the 24-week duration of the trial," the authors write.

While amitriptyline and topiramate may be effective in treating headaches in adults, the study results "put doctors in a little bit of a conundrum" when it comes to children, commented lead author Scott W. Powers, PhD, a pediatric psychologist at the Department of Pediatrics, University of Cincinnati College of Medicine, Ohio.

Dr Scott W. Powers

The findings were published online October 27 in The New England Journal of Medicine, to coincide with their presentation at the Child Neurology Society Annual Meeting in Vancouver, British Columbia, Canada.

Commonly Used

There are currently no US Food and Drug Administration (FDA)–approved migraine prevention medications for patients younger than 12 years, although a survey of pediatric headache specialists showed that amitriptyline and topiramate are the most commonly used preventive medications in children.

The multicenter, double-blind, phase 3 CHAMP trial enrolled youngsters aged 8 to 17 years with episodic or chronic migraine. They were randomly assigned to receive oral amitriptyline, topiramate, or placebo. The target dose was 1 mg/kg per day for amitriptyline and 2 mg/kg per day for topiramate.

After a 24-week treatment period, there was a 2-week weaning period and a 4-week follow-up period.

The primary outcome was a relative reduction of 50% or more in the number of headache days. Headache information came from headache diaries that patients completed daily.

Secondary outcomes included headache disability, as measured by absolute changes in the score on the PedMIDAS scale, which assesses the effect of migraines on school, home, play, and social activities. Other secondary outcomes included the absolute reduction in the number of headache days, number of trial completers, and serious adverse events.

The trial randomly assigned 361 children, mean age 14 years, who were mostly female (68%) and white (70%) and had a mean of 11 headache days per month. The study compared amitriptyline with placebo, topiramate with placebo, and the two drugs against each other.

"Our purpose was to create a 'real world' study that would enroll the type of patients that practitioners see every day," said Dr Powers. "Our hypothesis was that we would find one of these medicines to be the most effective with the least amount of side effects so that pediatricians and family medicine doctors would have sort of a first-line prevention approach for such a chronic, common illness as migraine."

In November 2014, it was determined that criteria for the a priori futility cutoff — below 20% compared with placebo — were met and that adding participants was unlikely to change the outcome. The conditional power at the time of the interim analysis was 16 percentage points for the comparison between amitriptyline and placebo and 14 percentage points for the comparison between topiramate and placebo. Study organizers decided to close down the study.

At that time, researchers had complete data on 225 children and adolescents, with another 103 participants in the midst of finishing the trial, for a total 328 patients to analyze for the primary outcome (132 in the amitriptyline group, 130 in the topiramate group, and 66 in the placebo group).

In the intention-to-treat analysis of the 328 patients, 52% in the amitriptyline group, 55% in the topiramate group, and 61% in the placebo group met the primary outcome (reduction of 50% or more in the number of headache days).

Placebo "Not Inert"

Dr Powers noted that placebos are "not inert" and "actually change the brain." He and his colleagues hypothesized that half of the kids would get better on placebo and that 70% of them would improve on the medicines.

"The findings were kind of upside down; it was 61% of placebo and around 50% to 55% on the drugs who got better. It wasn't statistically different, but it was in the opposite numerical direction than what you would have predicted," he said.

The adjusted odds ratio (OR) for the primary outcome in the study was 0.71 (98.3% confidence interval [CI], 0.34 - 1.48; P = .26) for amitriptyline vs placebo and 0.81 (98.3% CI, 0.39 - 1.68; P = .48) for topiramate vs placebo.

There was no significant difference in effect when the two active drugs were compared with each other (OR for amitriptyline vs topiramate, 0.88; 98.3% CI, 0.49 - 1.59; P = .49).

If there's any "positive news," it's that kids do get a lot better on these therapies, said Dr Powers. "There's a 50% reduction in migraine frequency, and headache disability is pretty much down to little to none — but it's not because of the chemical in the medicine that they have been given."

There was no difference in response between younger kids (age 8 to 12 years) and older ones (age 13 to 17 years). "As a covariate or correlate, age did not have any relation to how well you got or didn't get," said Dr Powers.

As well, he said, response did not differ between patients with chronic and those with episodic migraine.

What was surprising to the investigators, said Dr Powers, was the fact that "it was so clear so early" that the active drugs were no better than placebo.

The findings were clear not just for the primary outcome but for disability outcomes on PedMIDAS and other secondary outcomes, including study completers, said Dr Powers.

"So it was so consistently not going to be effective, and to all of us investigators, once we had seen the information, it made perfect sense that closing the study was what we needed to do."

There were also side effects to consider. A total of 852 AEs occurred (301 with amitriptyline, 419 with topiramate, and 132 with placebo) in 272 patients. There were no deaths.

AEs that occurred significantly more often in the amitriptyline than in the placebo group were fatigue (30% vs 14%; P = .01) and dry mouth (25% vs 12%; P = .03).

AEs that occurred significantly more often in the topiramate group than in the placebo group were paresthesia (31% vs 8%; P < .001) and decreased weight (8% vs 0%; P = .02).

Why would amitriptyline and topiramate work in adult migraine but not in pediatric migraine? Dr Powers pointed out that adult studies typically don't use the 50% reduction endpoint and disability score that his study did, but instead use an absolute headache day reduction endpoint.

"So many of these drugs are approved if the difference between placebo and drug is 1 to 2 headache days a month."

During the trial, the FDA approved topiramate for the treatment of episodic migraine in patients aged 12 to 17 years, but it's Dr Powers' understanding that this was on the basis of one study rather than the standard two good trials from different labs.

"I would think that the FDA is going to have to look at this; maybe topiramate doesn't have the evidence that they thought it had."

Boost to CBT?

The study results might boost use of cognitive-behavioral therapy (CBT) and other nonpharmacologic approaches. Dr Powers cited the study he and his colleagues published a few years ago in JAMA that compared two groups of kids with chronic migraine taking amitriptyline, with one group also getting CBT and the other also getting headache education.

"The group that got CBT got better and stayed better," said Dr Powers. "We need to make CBT more accessible and more of a first-line treatment."

Children tend to respond better than adults to CBT — and to relaxation therapy — partly because they have fewer distractions, are more open-minded, and are used to learning and doing homework. "To get good at CBT and relaxation, you need to practice them," said Dr Powers.

Researchers are also starting to look at brain changes on imaging in patients receiving CBT compared with placebo, he said. "The hypothesis is that there will be changes, but you will see different changes. The pain literature suggests that CBT affects both the emotion amygdala center of the brain and the frontal executive part of the brain, and that placebo in pain typically only affects the executive frontal part."

Commenting for Medscape Medical News, Kenneth J. Mack, MD, PhD, professor of neurology and of pediatrics, Mayo Clinic, Rochester, Minnesota, agreed that nonpharmacologic approaches might be the next step in light of this new study.

The "big question" now is "where do we go from here?" said Dr Mack. "Most likely, one response will be to use more behavioral approaches for the treatment of headache pain."

Dr Mack described the study as "well designed and of the highest quality."

"It did not show the expected results, but that is why we have to do studies like this."

The CHAMP trial was funded by the National Institutes of Health. Dr Powers has disclosed no relevant financial relationships.

N Engl J Med. Published online October 27, 2016. Abstract

For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.