First-in-Class Insomnia Drug Gets Nod From FDA Committee

Pauline Anderson

May 23, 2013

A first-in-class insomnia drug (suvorexant; Merck Sharp & Dohme Corp.) has moved a step closer to market after day-long discussions by the US Food and Drug Administration's (FDA's) Peripheral and Central Nervous Systems Drugs Advisory Committee.

Committee members agreed the drug is effective and safe, although many shared concerns about effects at higher doses, and they concurred on dosing ranges suggested by the company.

However, despite the FDA's own warning that the lowest dose developed (15 mg) may not be low enough for safe use, the committee members voted against having the sponsor perform additional efficacy studies on a 10-mg dose before approval.

Suvorexant is a selective antagonist of orexin receptors OX1R and OX2R and blocks binding of orexin A and B to these receptors, presumably inhibiting activation of neurons of the arousal system. The drug has a 12-hour half-life, which is longer than that of other insomnia drugs.

Extensive Data

The sponsor provided what most agreed was extensive study data. A phase 2B double-blind, crossover trial investigated 10, 20, 40, and 80 mg of the drug and placebo in nonelderly patients (under age 65 years), including 243 receiving suvorexant. Three phase 3 trials (1 long-term safety and 2 confirmatory efficacy trials) that included 1784 patients taking suvorexant examined 15-mg and 30-mg doses in the elderly and 20-mg and 40-mg doses in the nonelderly.

Company representatives provided evidence that suvorexant is effective in sleep onset (latency) and maintenance. According to David Michelson, MD, vice president, clinical research, at Merck, the effect was sustained over 1 year, and patients perceived the efficacy as meaningful.

Subjective as well as objective data were presented indicating that, for the most part, the efficacy and safety of the drug are dose related. For example, the 7-point Insomnia Severity Index (ISI), which measures patient satisfaction and quality of life, showed that the drug was given scores twice that of placebo.

Committee members also heard that the drug is safe. According to Wm. Joseph Herring, MD, PhD, executive director, clinical research, Merck, there was "no withdrawal or clinically meaningful rebound upon stopping the drug," which has a "low potential for abuse." Serious adverse events were uncommon and similar between the drug and placebo, said Dr. Herring, adding that no differences were seen across age and sex.

While the company found that that 10-mg dose was the least effective in phase 2 studies, the FDA had a different take after conducting its own post hoc analysis of the data. Ronald Farkas, MD, PhD, clinical team leader, said he has "a high degree of confidence" that the efficacy of the 10-mg dose is similar to that of the higher doses. Serious adverse events are not well understood at this low dose, he said, because phase 2 studies that included this dose had too few patients.

"It may be worthwhile to explore strategies to maximize the efficacy of the 10-mg dose, or even lower doses," said Dr. Farkas.

New Options Needed

Many committee members agreed that current treatments for insomnia don't serve all patients well. Many of the shorter-acting benzodiazepines and "z drugs" don't maintain sleep as effectively, and some of the older drugs carry risks such as falls.

"New options are necessary for the treatment of insomnia," said Justin A. Zivin, MD, professor emeritus, University of California, San Diego. He echoed the sentiments of others when he added that all patients with insomnia should have their doses titrated and that they should always be started on the lowest effective dose.

But committee members expressed concern about several possible adverse effects linked to suvorexant, including dose-related next-day somnolence.

According to comments from Russell G. Katz, MD, director, Division of Neurology Products in the FDA's Center for Drug Evaluation and Research, there was an 8-fold increase in the incidence of somnolence between the low and high doses in the 3-momth controlled trial data, and the incidence of somnolence continued to increase for at least 60 days. There was, however, evidence that elderly patients had a lower incidence of somnolence at both the low and the high dose than did the nonelderly, he said.

There was a concern that somnolence contributed to impaired driving the morning after dosing. In a study of nonelderly adults, both the 20-mg and 40-mg doses were associated with an increase in those who had excessive deviation in lane position (weaving). This effect persisted after 8 days of continuous dosing of 40 mg. Five participants had to discontinue the testing because of excessive somnolence.

Dr. Farkas also expressed concerns over analysis showing that 20% of those under age 65 years could be impaired after a 20-mg dose. He emphasized that blood levels, not just dose, is correlated with somnolence.

Another issue was suicidal ideation, assessed through questionnaires. Among those taking placebo, 0.1% had suicide ideation compared with 0.2% in the low-dose and 0.7% in the high-dose group. Suicide ideation is an important concern as 23% of women aged 40 to 59 years take an antidepressant and depression and insomnia coexist, said Dr. Farkas.

Possible Cataplexy

Other concerns were excessive daytime sleepiness, sleep paralysis, and hypnagogic/hypnopompic hallucinations. Because orexin neurons are lost in the sleep disorder narcolepsy, cataplexy — which often coexists with narcolepsy — is another legitimate concern. According to the FDA, the presented research suggests at least 1 case of possible cataplexy; this patient had "weak knees when laughing." Mild cataplexy can be "easily overlooked," said Dr. Farkas.

Because drug clearance is slower in women and the obese, it's possible that obese women have a clearance that is up to 3 times lower than men with a normal body mass index, the committee heard.

Other concerns of the FDA as outlined by Dr. Farkas are that the selected patients in the studies were generally healthy and that there were few data on concomitant diseases or use with drugs commonly used in actual clinical practice.

As Robert R. Clancy, MD, professor, neurology and pediatrics, University of Pennsylvania School of Medicine, Philadelphia, pointed out, most elderly patients are taking 5 or more different drugs, and other committee members noted that older patients tend to self-medicate with alcohol and antihistamines during the day.

Some members emphasized that care should be taken because this is a new class of drugs and that sleeping aids are mostly prescribed by family practitioners and internists, some of whom may not investigate for all possible causes of insomnia.

Diana Zuckerman, PhD, president, National Research for Women & Families, wondered whether the risks of the drug outweigh the benefits. The benefits are statistically significant but are often modest — falling asleep 5 minutes sooner, for example, or having 20 more minutes of uninterrupted sleep.

Despite all the data, she said, "it's clear that there are serious and substantial risks at higher doses," and more data are needed for smaller doses. "Having 60 people in a study is not good enough," Dr. Zuckerman said.

She said drug labeling is not sufficient to warn about possible adverse effects. Others agreed that labels that advise people to "use with caution" or "don't operate machinery" are meaningless.

Robert B. Voas, PhD, senior research scientist, Pacific Institute for Research and Evaluation, said consideration should be given to a warning about using the drug with alcohol, in addition to the standard warning about not operating machinery.

Lisa M. Schwartz, MD, professor, medicine and community and family medicine, Geisel School of Medicine at Dartmouth, White River Junction, Vermont, said labeling should be "very explicit" about possible harm and "clear" about why patients could hurt themselves or someone else.

Lack of safety data was a concern for Sammy Almashat, MD, Public Citizen's Health Research Group. He warned that the drug should be discontinued 2 to 3 days before driving, but asked, "How likely is this to happen?"

The low-dose drug doubled somnolence and the high-dose tripled somnolence, he said. He also noted that the drug causes a dose-dependent increase in cholesterol.

Committee Votes

In a 12 to 4 vote (1 abstention), the committee decided that the proposed doses (20 and 40 mg for nonelderly; 15 and 30 mg for elderly) are effective for sleep onset, and in a 16 to 0 vote (1 abstention), they determined that these doses are effective for sleep maintenance.

Addressing the question of whether the sponsor should be required to perform additional efficacy studies on the 10-mg dose, the committee voted 5 in favor and 11 against (1 abstention).

Many members felt another study wouldn't add any valuable new information, but a few thought that such a study is needed because this is a new class of drugs, and that while the 10-mg dose is likely to be effective, there's currently not enough evidence.

Some members noted a drawback of "double dosing" with this low dose. Patients might take a second dose in the middle of the night when the first dose doesn't work, said Christian Guilleminault, MD, professor, Sleep Medicine Division, Stanford University School of Medicine, California.

Committee members determined, in a vote of 13 to 3 (1 abstention) that the starting doses of 15 mg and 20 mg are safe. However, in a vote of 7 to 8 (2 abstentions), they did not agree that the safety of the higher doses (30 mg and 40 mg), given if there's no response to the lower doses, is acceptable.

The biggest concern for many committee members was the significant jump in somnolence at the higher doses. Such evidence sends a "scary signal" with potential for harm that is "substantial," said Dr. Schwartz.


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.