Alan Cassels is a pharmaceutical policy researcher at the University of Victoria, British Columbia, a journalist, and author of the The Cochrane Collaboration: Medicine’s Best-Kept Secret.
Almost every drug that has been removed from the U.S. market over the past 20 years has been preceded by the dreaded “black box warning,” which appears on a prescription drug’s label and “is designed to call attention to serious or life-threatening risks.”
But in mid December, in an unprecedented move, the FDA removed the black box warning from Pfizer’s smoking cessation drug Chantix (varenicline). While this didn’t come as much of a surprise to some, it was a newsworthy event, receiving coverage by Reuters, The Wall Street Journal, STAT, Pharma Times, Philadelphia Business Journal, and Medscape.
The decision to remove the black box warning had its roots in a study known as the EAGLES trial, which was mandated by the FDA and designed to establish the neuropsychiatric safety and efficacy of varenicline, buproprion and nicotine patches in smokers with or without psychiatric disorders. Pfizer maintained that this trial confirmed Chantix’s superiority and safety, yet the trial was highly criticized by some observers. Those objections received little attention from some news outlets that covered the label update.
As a randomized placebo-controlled trial sitting at the pinnacle of the evidence pyramid, EAGLES might be expected to settle the question of Chantix’s safety. Not so, according to Thomas Moore, a senior scientist with the Institute for Safe Medication Practices, an independent nonprofit group that studies drug safety. ISMP released numerous reports detailing Chantix adverse effects, including one in 2014 that analyzed Chantiz-related adverse events from 2007 through most of 2013. It found that Chantix had more cases of suicidal thoughts, self-harm, and homicidal thoughts than any other drug, by a more than three-fold margin. Moore believes that the Chantix warnings needed to be strengthened, not removed.
About the EAGLES trial, he told STAT: “With eight different treatment arms, the number of patients in each may not be enough to capture the severe psychiatric side effects for which the drug is known.”
The ISMP went further in describing EAGLES’s design problems in a statement urging the FDA not to remove the black box warning: “The study was powered to detect a moderate or severe drug event in at least 4% of the patients in any treatment group, a very large serious adverse drug
effect of any kind, for any drug, in any setting.” Suicidal thoughts and violent behaviors are thought to occur in much fewer than 4% of medication users, the ISMP statement noted. “Given an expected event rate measured in few cases per 1,000 it was a certain recipe for failure to design a safety trial capable of detecting an effect frequency of 40 per 1,000 cases (or 4%), and relying on an unvalidated measurement scale never used as a clinical trial endpoint. This trial was underpowered by an order of magnitude.”
Those concerns were echoed by Kim Witczak who sits on the the FDA Psychopharmacologic Drugs Advisory Committee as a consumer representative. As a long-time drug safety advocate, she intended to participate at the committee hearings on the safety of Chantix but was recused from this meeting because of a previous lawsuit against Pfizer. Witczak attributes her husband’s suicide in 2003 to the Pfizer antidepressant Zoloft.
Over the phone from Los Angeles, she told me she attended the Washington meeting on her own dime and expressed her concerns over what the EAGLES trial did and did not show.
She told the committee she looked at a small sample of 100 patients involved in the study, and in that sample there were two overdose cases and a patient who died in a car accident. None of these was coded as suicide, one of the potential adverse effects related to Chantix. Were those deaths appropriately coded? In Witczaks’s opinion, like that of Thomas Moore, the EAGLES study was funded by manufacturers and required an in-depth, independent review of all cases of adverse events in order to be certain. But this didn’t happen.
Witczak views FDA committee hearings as highly political events. In the case of Chantix, she says, patient groups dedicated to quitting smoking gave testimony begging for removal of the black box even though those groups are funded by the manufacturers of smoking cessation drugs.
In her mind, the removal of the warning “will be spun as safe, that the risk of smoking is greater than the risk of the drug.” But this characterization ignores the experiences of 2,700 people who took legal action against Pfizer, Witczak says. Many of those who attempted suicide–which they blamed on the drug–settled in a legal action against Pfizer and the details of those cases remain sealed (though details would have been made available to the FDA). For Witczak, “the 2,700 victims before the black box warning are all silenced.”
Many stories on the FDA decision still describe serious concerns about the drug. The Philadelphia Business Journal noted: “Chantix’s updated label will still carry a warning of potential adverse events such as psychosis, hallucinations, paranoia and anxiety and other neuropsychiatric problems. The difference is those events will not be highlighted in a ‘black box’ warning, the FDA’s sternest warning for an approved medicine.”
However, few stories gave adequate weight to the voices of expert, independent groups like U.S. Public Citizen, which was so concerned about Chantix’s dangers it launched a petition in 2014 asking the FDA for stronger warnings on Chantix. Other groups, including Consumer Reports, the National Center for Health Research and the National Physicians Alliance, also reviewed the evidence of Chantix and asked for additional side effects be added to the black-box warning including “risks of psychotic symptoms, blackouts, convulsions and impaired vision as well as unexplained hostility, anger and aggression toward others.”
In addition, few of the most recent stories noted the contentious nature of the September 2016 FDA advisory committee meeting, nor discussed the concerns of dissenting panel members. Ten members of the 19-member panel voted in favor of removing the black box, four recommended a language change and five recommended the strengthening of the label. While those details were missing, many stories quoted the Chief Medical Officer of Pfizer, who said in a news release that “The totality of available scientific evidence, including the outcomes of EAGLES, supports the safety and efficacy of CHANTIX.”
Critics fear that this is just a first step toward additional black box rollbacks supported by industry based on potentially biased evidence. For example, antidepressants have carried a black box warning on increased risks of suicide in young people since 2004. Studies questioning that warning, some of which have been conducted by researchers tied to the pharmaceutical industry, have been appearing ever since. FDA officials have been quick to point out the many flaws in these analyses, but one wonders how consistent that position will remain in the face of a changing U.S. administration with possibly radically different regulatory priorities.