SILVER SPRING, MD. – The boxed warning about the risk of serious neuropsychiatric effects associated with the smoking cessation drug varenicline should remain on the drug’s label, and the need for this warning can be reevaluated when the results of a postmarketing safety study become available next year, according to the majority of a Food and Drug Administration advisory panel.
On Oct. 16, at a joint meeting of the FDA’s Psychopharmacologic Drugs and the Drug Safety and Risk Management Advisory committees, 11 of the 18 panelists voted to retain the boxed warning about neuropsychiatric symptoms and suicidality, agreeing that a decision about whether to retain the warning should not be made until the results of the study are available.
Pfizer, which manufactures varenicline, a nicotinic receptor partial agonist in a tablet formulation, as Chantix, is conducting the prospective, randomized, double-blind study comparing neuropsychiatric events in 8,000 smokers, with and without a psychiatric history, treated with varenicline, nicotine replacement therapy, bupropion, or placebo. Although the results are expected in 2015, the company maintained that the boxed warning is no longer justified and can be put in the warnings and precautions section, based on analyses of observational studies and randomized clinical trials. The company submitted those analyses to the FDA earlier this year.
The boxed warning in the drug’s prescribing information states that the serious events reported in patients taking the drug include but “are not limited to” depression, suicidal ideation, suicide attempts, and completed suicides; and that some cases “may have been complicated by the symptoms of nicotine withdrawal in patients who stopped smoking.” The reason the boxed warning is used for this drug is that there is a “serious adverse reaction that can be prevented or reduced in frequency or severity,” the FDA said.
Because the precedent for removing a boxed warning from a drug label is limited and the FDA’s view that the decision should consider the final results of the postmarketing trial, the agency convened the joint meeting to review the analyses submitted by the company. Overall, the panelists were not convinced by the analyses. They cited various weaknesses in the observational studies and the meta-analysis, including questions about whether the correct outcomes were being captured.
Six panelists voted to modify (but not weaken) the language in the boxed warning, and most recommended that the last few lines about the health benefits of quitting smoking be removed and argued that the wording had a promotional tone and did not belong in a boxed warning. Several panelists recommended adding sleep disruption and disorders, and recognized adverse events of the drugs to the boxed warning.
Dr. Jeanmarie Perrone
Among those who voted to retain the boxed warning, Dr. Jeanmarie Perrone, director of the medical toxicology division in the emergency medicine department at the University of Pennsylvania, Philadelphia, said there was biologic plausibility for these events. Her biggest concern, she said, was that removing the boxed warning might be interpreted as “an endorsement of safety and that has not been demonstrated.” In the emergency department, she said, she sees many patients who have been prescribed varenicline “by a myriad of clinicians,” and the presence of a boxed warning helps keep this safety issue in mind.
Dr. Rajiv N. Rimal, professor and chair of the department of prevention and community health, George Washington University, Washington, who voted to retain the warning with modified wording, said: “I heard enough compelling evidence to suggest that even some of the more rare events were severe enough” that revisiting the labeling change would be necessary once the other data became available.
Varenicline, approved by the FDA in 2006 as an aid to smoking-cessation treatment, provides a “low level” of dopamine release as people try to quit smoking, according to Pfizer. The boxed warning was added to the label in July 2009, based on the FDA’s review of spontaneous adverse event reports, conducted after an alert about suicidality reports associated with the drug from European regulators about 1 year after FDA approval. About the same time, there was a widely publicized case of a Texas musician who was taking varenicline and was shot and killed by a neighbor while exhibiting uncharacteristic aggressive behavior in the fall of 2007.
The reports provided by the company at that time included 102 suicide-related cases, and 525 reports of aggressive/irrational behavior. A review of the reports in the FDA’s adverse events database included many cases with “hallmarks of drug relatedness,” patients reporting unusual symptoms with “similar distinctive language,” and positive dechallenges and rechallenges, said Dr. Celia Winchell, the medical team leader of addiction products in the FDA’s division of anesthesia, analgesia, and addiction products. Not all cases involved suicidality, depression, or aggression. Postmarketing reports have provided a “rich narrative with detail” about patient experiences, including descriptions that are relatively easy to describe or classify, such as suicide and aggression, while others “defy coding,” such as one person who reported feeling “like a zombie,” she told the panel.