According to a briefing document for today's joint meeting of the FDA's Psychopharmacologic Drugs Advisory Committee and Drug Safety and Risk Management Advisory Committee to discuss the black box warning for Chantix (varenicline), Pfizer's study of the potential serious psychiatric side effects of Chantix, which reported no significant increased risk of adverse events, was marred by financial conflicts of interests of many of the study investigators and by under-reporting of adverse events, including potential episodes of suicidal ideation and depression that could have been drug-related.
According to the document, investigators at 32 of the 139 study sites had a financial conflict of interest by virtue of having received more than $25,000 in the past five years in speaking fees related to the promotion of Chantix or other Pfizer products. Moreover, at the sites with financially conflicted investigators, the reporting of adverse drug events was substantially lower than at non-conflicted sites. For example, while the rate of adverse neuropsychiatric events was 6.4% at non-conflicted sites, the rate was only 1.8% at sites at which an investigator had a conflict of interest.
In some cases, the under-reporting of adverse events seems quite blatant. For example, the document cites one patient, who was coded as not having an adverse neuropsychiatric event, as having reported: "I think I am having a major depression. I am worried. I cry easily. I have apathy, I have no desire to do things, insomnia, increased appetite, guilt, I have death thoughts (without suicidal ideation)."
The Rest of the Story
This story illustrates why investigators with personal financial interests in a drug company should not be conducting clinical trials on drugs manufactured by that company. The conflict of interest taints the research and could easily be seen as having influenced the conduct, analysis, or reporting of the research.
In this investigation by the FDA, not only did the agency find financial conflicts of interest of a large number of study researchers, but it also found significant under-reporting of adverse events which was highly correlated with whether a site had a conflicted researcher. This certainly creates the appearance that personal financial conflicts of interest with Pfizer created a bias in the conduct of the study and that the results may therefore be tainted and not trustworthy.
This story also explains my concern about an unreported financial conflict of interest in a newspaper article summarizing the results of an in vitro study in which the conflicted investigator improperly extrapolated the results to have clinical significance in humans and made what I see as a negligent medical recommendation to the public. In that case, the study showed that e-cigarette aerosol caused damage to respiratory cells in culture yet the investigator concluded that vaping can cause fatal lung disease and advised vapers to get vaccinated for bacterial lung infections, even though he failed to make the same recommendation for smokers.
In my opinion, investigators with personal financial conflicts of interest should not be conducting clinical research on products that are manufactured by the relevant company or its competitors. This is not just to protect the integrity of the research, but to protect the investigators and their institutions.