Last week, the U.S. Food and Drug Administration (FDA) issued a briefing document in advance of a joint advisory committee meeting held on Wednesday, September 14, 2016. See FDA Briefing Document. It was a joint meeting of the Psychopharmacologic Drugs Advisory Committee and the Drug Safety and Risk Management Advisory Committee and was held to discuss “Serious Neuropsychiatric Adverse Events with Drugs for Smoking Cessation.” Id. See title of report. Pfizer, the manufacturer of Chantix (varenicline), a product intended to help patients quit smoking, has asked FDA to remove a box warning of potential psychiatric risks to patients, including suicide and suicidal ideation. Pfizer provided data from a post-marketing study of 8,000 subjects at 140 sites around the world that was designed to assess the real-world risk of suicidal behavior associated with the use of smoking cessation drugs to support its request that the boxed warning be removed.
Chantix was launched in 2006 without the suicidality warning, and FDA required its addition in 2009 after reports of an increase in suicidal behaviors upon initiation of treatment. In the spring of this year, European regulators removed a similar warning on the drug’s labeling, sold in Europe under the brand name Champix, based on post-market clinical data from a different study conducted overseas.
FDA’s briefing document for the joint Advisory Committee meeting contains preliminary views from reviewers at the Center for Drug Evaluation and Research (CDER) at FDA where they specifically questioned numerous aspects of the data collection and interpretation of the post-market study submitted to FDA. Results from the study, which compared Chantix or Zyban (bupropion), a similar product marketed by GlaxoSmithKline, with a placebo or a nicotine patch in patients with and without a history of mental health problems, appeared to show that the drug did not significantly increase the incidence of suicide and suicidal thoughts.
However, FDA’s initial review suggested that the results of the study may be biased due to several problems associated with data collection and with the descriptions by investigators about the severity of some adverse events. Among other critiques, FDA suggested that money given to investigators from Pfizer may have influenced the study’s results. According to the briefing document, in sites where investigators received $25,000 or more in speaking and consulting fees from the company, only 1.8 percent of patients who had already been diagnosed with a psychiatric disorder were reported to have side effects, such as anxiety, agitation, hostility, or suicidal thoughts, during the trial. On the other hand, at locations where physician/investigators received less money or no payments at all from Pfizer, 6.4 percent of patients were reported to experience such side-effects.
It is noteworthy that FDA suggested publicly in its review of results of a study that, even in part, the influence of fees to physician/investigators may have biased results in a way that calls the validity of the data into question. The agency’s message in the briefing document was essentially clear—that the results of the study may not be trustworthy given how the data was collected and analyzed, and due to conflicts of interest related to the payments to the investigators. FDA routinely collects data on financial payments to investigators conducting studies under Investigational New Drug applications by pharmaceutical companies under 21 CFR Part 54, and routinely analyzes whether financial incentives create bias. However, the agency does not typically announce publicly its assessment of these interests and the potential impact on particular data. It will be interesting to see whether reviewers’ concerns about data integrity are influential such that the study results are not credited by the agency in its decision.
This briefing document is the first time we are aware of agency disclosure of concerns about bias from industry payments to clinical investigators. We will be watching closely for other indicators of concern by the agency regarding such payments. In this instance, the joint advisory committee voted in favor of removing the suicidality warning by a vote of 10 to 9 in favor of removal of the warning, with the ten in favor focused on the health benefits of quitting smoking v. the potential suicidality risks of the drug. Conversely, the nine panelists who voted either to change the current warning or retain it noted that they were concerned about setting a dangerous precedent for pharmaceutical companies seeking to remove warnings from drug products, and/or that the data integrity issues were troublesome. Given that reviewers have expressed concerns about the data used to support the removal, this regulatory decision will be particularly interesting and one that could influence conflicts of interest disclosure policy going forward.