Large Placebo Effect Scuppers Pediatric Headache Trials

Need for Better-Designed Studies, More Research Into Novel Treatments for Kids

Caroline Cassels

January 29, 2013

High placebo response rates, coupled with a paucity of placebo-controlled trials, appear to be major obstacles in determining the efficacy of agents to successfully prevent and treat pediatric headache, 2 new studies suggest.

In a systematic review, investigators at the US Food and Drug Administration (FDA), Silver Spring, Maryland, found the high placebo response rate across all trials "may represent the principal challenge in pediatric trials of drugs for abortive treatment of migraine."

Similarly, in a second meta-analysis of pediatric headache prophylaxis, researchers from the Medical College of Wisconsin in Milwaukee also concluded that because of "significant placebo response, future trials need to include placebo controls."

Both studies are published online January 28 in JAMA Pediatrics.

Common and Disabling

A common complaint, headache is estimated to affect up to 49% of children in any given month, and 4.2% of children report having headaches on 10 or more days per month. Migraine affects nearly 8% of children. In younger children aged 5 to 9 years, migraine is more common among boys; in adolescence, it is more prevalent in girls.

Although some drugs trials have reported efficacy of various triptans in treating migraine in children and adolescents, only 2, almotriptan malate and rizatriptan benzoate, have been approved as safe and effective in this patient population by the FDA.

To identify possible causes for the failure of pediatric trials of triptans for the treatment of migraine, the FDA investigators, led by Haihao Sun, MD, PhD, reviewed all pediatric efficacy and pharmacokinetic trial data of drugs approved for treatment of adult migraine and submitted to the FDA from January 1, 1999, to December 31, 2011.

The investigators compared the design, participants' baseline characteristics, inclusion and exclusion criteria, efficacy endpoints, and pharmacokinetic profiles across 7 randomized, double-blind, placebo-controlled, parallel-group triptan trials.

The analysis revealed high response rates for placebo in all trials with pain relief at 2 hours ranging from 53% to 57.5%. In one 2011 study of rizatriptan, the researchers excluded patients with an early placebo response and then went on to randomly assign the remaining participants to active drug or placebo. This reduced the placebo response rate by 6% for headache freedom at the 2-hour post-treatment endpoint.

"The high placebo response rate has been recognized as a significant obstacle in triptan trials for treatment of adolescent migraine. Consistent with previous reports, a high placebo response rate was observed in our analysis of pediatric trials of abortive therapeutics for migraine," the authors write.

A Different Entity in Kids?

In the second study, investigators led by Khalil El-Chammas, MD, assessed data from efficacy trials of various agents, including anticonvulsants, antidepressants, and β-blockers, primarily for episodic migraine ( < 15 headaches per month). One trial assessed the efficacy of the antidepressant fluoxetine for the prevention of chronic daily headaches (≥15 headaches per month).

This analysis included 21 randomized trials of headache treatment among children and adolescents. In all, there were 13 placebo-controlled trials and 10 active comparator trials, 2 of which also included placebo. The investigators found that the anticonvulsant topiramate and the antidepressant trazodone were more effective than placebo for episodic migraines.

Ineffective agents included clonidine, flunarizine, pizotifen, propranolol, and valproate, the investigators report. Similarly, fluoxetine was found to be ineffective as prophylaxis against chronic daily headache.

Like the FDA group, Dr. El-Chammas and colleagues found a strong placebo response, noting that "patients given placebo experienced a significant decline in headaches."

Although the reasons for the high placebo response rate are unclear, the FDA investigators speculate that it may be due to the shorter duration of migraine attacks in adolescents and suggest that this population may have different types of migraines than adults. They add that "until we have better mechanisms to identify the causes of the different expressions of the disease, these patients may not require pharmacologic interventions."

Dr. El-Chammas and colleagues conclude that although the data from their study suggest trazodone and topiramate may be offered as migraine prophylaxis in young patients, the limited evidence makes drawing firm conclusions "impossible."

"More studies of pediatric headaches need to be conducted. Because there was a significant placebo response, future trials need to include placebo controls."

Few Studies Meet International Headache Society Standard

Commenting on the findings of the 2 reviews for Medscape Medical News, Jennifer L. Bickel, MD, pediatric neurologist at Children's Mercy Hospitals and Clinics in Kansas City, Missouri, said the findings from the FDA study further highlight the need to change pediatric migraine trial designs in order to decrease the placebo response as much as possible.

She noted that the paper by Dr. El-Chammas and colleagues provided an "excellent review" of pediatric headache prevention studies and again highlight the need for better-designed trials.

"The International Headache Society recommends that well-designed, placebo-controlled trials are necessary to establish efficacy in preventing headaches. The authors found very few studies that used this standard," said Dr. Bickel.

"Physicians often assume that migraine medications routinely used in adults will also benefit pediatric migraines. However, this is not necessarily the case. We need more information with better-designed trials. These studies show the need for more funding of pediatric headache research so we can develop more effective treatments for this highly disabling condition," she added.

Dr. Bickel said that in her experience as a headache specialist who has treated both adult and pediatric refractory headaches, medications are most effective when combined with teaching pain coping skills, addressing stressors, and increasing school functioning.

"It is very possible that what is considered 'placebo' in these studies is the result of education by the doctors and nurses in proper headache management. For example, though the adolescent may be receiving the nonactive drug, the participant likely received reassurance of the benign nature of their headaches which appears to sometimes decrease frequency and severity of headaches by itself."

Need for "Original" Meds

She added that if parents do not want to treat their child with medications, "we physicians lack good evidence to convince them otherwise. Children with short lasting headaches may be better served by a break in a dark quiet room rather than medications."

Nevertheless, said Dr. Bickel, medication can still be a useful treatment option for migraine management.

This research does not mean that there are no good treatments, she said. Rather, it implies that "we cannot assume that medications in adults are also helpful in children."

"This research should not only show the need for better-designed trials but should also show the lack of original medications for headaches. The triptans are a specific class of drugs used to abort headaches. However, no recent medication has been developed primarily for migraine prevention. Instead, the medications that were evaluated in this study are antidepressants, antihypertensives, and antiepileptics. We need more research at a basic science level to formulate medications for migraine prevention in adults and children."

Does It Matter?

In an accompanying editorial, Marco A. Arruda, MD, PhD, from the Department of Pediatric Neurology at the Glia Institute, Sao Paulo, Brazil, like Dr. Bickell, points out that these studies "force us to remember that 'no evidence of efficacy does not mean the evidence of no efficacy'."

Many of the medications for migraine prophylaxis are routinely prescribed off-label, Dr. Arruda points out, "and they seem to be efficient and safe. Are they effective or are clinicians observing the placebo effect?

Although placebo is the enemy of great clinical trials, it is likely the best friend of good clinicians. Dr. Marco A. Arruda

"This is impossible to answer other than by proposing another question: Does it matter?" he asks. "Although placebo is the enemy of great clinical trials, it is likely the best friend of good clinicians."

He is not advocating that clinicians ignore evidence or not demand it, he adds; the acute treatment of migraine in children "can be solidly conducted within label" because drugs such as ibuprofen, acetaminophen, almotriptan, and rizatriptan are all FDA approved for use in children.

"I am just acknowledging that, in many circumstances mainly involving the preventive treatment of migraine, pediatricians need to make best use of the little available evidence," Dr. Arruda writes.

Second, he notes, advances in clinical trial design have not kept pace with advances in the last decade in diagnosis, epidemiology, and the effect of migraine on children and adolescents. This "mismatch" becomes more frustrating, he says, in the face of recent evidence showing the effect of migraine on the quality of life, mental health, and school attendance and performance of affected children and their families.

"Therefore, I advocate that lack of evidence should not be translated into paralysis, with clinicians passively ignoring children's suffering," Dr. Arruda writes.

He concludes with some direction for clinicians. "(1) Keep in mind the incredible burden of migraine to affected children; (2) when possible, treat within label by giving preference to approved drugs; and (3) when no evidence is available, or when first-line therapies have failed, select a drug based on plausibility, proved efficacy in studies with adults, and proved safety of the drug in children.

"Lack of evidence associated with lack of good reasoning to navigate the lack of evidence is likely an ominous combination to children with migraine."

The studies' authors, Dr. Bickel, and Dr. Arruda have disclosed no relevant financial relationships.

JAMA Pediatr. Published online January 28, 2013. Abstract   Abstract    Editorial