Tomorrow, there will be an FDA panel meeting where a committee of drug safety and pharmacology experts will decide whether Chantix’s black box warning for neuropsychiatric events should be removed from its label. This warning has been on the label since 2008, when the FDA found in its Adverse Event Reporting System (FAERS) a number of suicide-related, mood-related and aggression related events that were large enough in scope to cause the FDA to require this change. In my opinion, this black box warning should not be removed at this time.

Why am I saying this? Is it because I hate Pfizer and the pharmaceutical industry? No, I think Pfizer is an incredible company that has brought the world a number of life-saving and life-improving medicines for decades. I could do without their latest Viagra commercial which should be on Cinemax instead of ESPN, but I digress.

The reason that the black box warning should remain on the label is that the FAERS data continue to back up the elevated risk for neuropsychiatric events. Using the AdverseEvents Explorer platform, our analyst team put together a Drug Safety Monitor which we released to our subscribers earlier today. In a nutshell, it looked at all of the events that the FDA considers a Neuropsychiatric Event and looked at the case numbers and disproportionality figure for each of those events. We found that out of the 58 events analyzed, Chantix showed an active safety signal for 52 out of 58 of those events, including suicide attempts, suicidal behavior, aggression, depression and many more.

The Chantix story on suicide related behavior is a little more cloudy since there have been a number of lawyer generated reports for suicide and suicidal behavior that have enhanced those particular safety signals. Many people assume that every plaintiff lawyer is dishonest so these cases must not be real. I actually believe that this assumption is not true, partly because I’m a lawyer and I know what a judge will do to you if you are completely making up your plaintiff’s injuries. However, even if you assume that every lawyer was lying and take out the suicide factors altogether, there are still 46 active safety signals out the 50 that the FDA considers an important Neuropsychiatric event.

When you look at some of the nicotine replacement therapies like Habitrol, Nicotrol, Nicorette gum, and Commit, there have been less than 5 cases of suicide attempts for each of these therapies. Why aren’t the lawyers making up suicide attempt cases for these other smoking cessation remedies if they are making them up for Chantix? My only point is that lawyer cases mean something. They may not be as reliable as a doctor report, but they still have some meaning that we should not ignore.


Related Read: How Can You Tell When a Lawyer is Lying?


In Pfizer’s brief which was released yesterday, they argue that FAERS data is unreliable because it “includes stimulated reporting, under-reporting, and lacks a true denominator (number of patients exposed). On the issue of stimulated reporting, my colleague Dr. Keith Hoffman published a study last month that analyzed this issue and found that it really does not exist. On this issue of under-reporting, we are currently working on another study that will quantify the extent that FAERS cases are under-reported. Some older studies have estimated that FAERS only captures 10-15% of the actual cases that occur. To make up for this under-reporting issue, we use disproportionality analysis to find safety risks. Using this analysis we found that Chantix has active safety signals for 52 out of the 58 Neuropsychiatric events that the FDA follows. To counter the third issue of “lacking a true denominator”, we just launched incidence rate calculations which look at each reported issue as a percentage of the number of patients exposed during that year. For instance, in 2011, we found that 0.59% of all patients exposed to Chantix experienced depression (roughly 1 in 170). We believe that we are the first company to combine real patient population statistics with case report data to its clients.


Related Read: FAERS: Myths vs. Realities


To attempt to counteract the FAERS data which the FDA has analyzed and we have analyzed, Pfizer included in their brief 4 Patient Based population studies, 8 Double-Blind Randomized Clinical Trials that Pfizer conducted and meta-analyses of 18 additional Randomized Clinical Trials that others conducted. All of those studies found that Chantix had similar rates of neuropsychiatric events to the placebo or a comparator (with the exception of sleep disorders which Pfizer acknowledges is known adverse event for Chantix).

Can I say that all 30 of these studies are flawed? No, but I can tell you that like all clinical trials, they are limited in patient numbers/populations, they are generally short in duration, or they are fairly restrictive in what they count as an neuropsychiatric event. For instance, in the patient-based population studies, the patient had to be admitted to the hospital or visit the emergency room with a neuropsychiatric event within 30 days of initial treatment with Chantix or the comparator. What about cases that take longer than 30 days to develop or just involve a primary physician visit? Those are the type of cases that fall through the cracks in these types of studies.

Overall, I feel that the black box warning should remain on the label. Does Chantix still help people stop smoking and should it still be prescribed? Yes, but the warning of for these neuropsychiatric events should remain.

To see our full Chantix DSM report or more reports from our analyst team like "Halaven Drug Label Update: A Comparative Safety Analysis", sign up for a free trial now

Robert Kyle

Robert Kyle

Chief Product Officer

Robert Kyle Linkedin

Topics: Drug Safety, Drug / Indication Information

Bob Kyle

Written by Bob Kyle