And so, the debate over Chantix continues. Two weeks after the FDA finished reviewing two epidemiological studies and decided that the benefits of the Pfizer pill continue to outweigh the risks, yet another study comes out and declares that the safety profile of the controversial smoking-cessation drug makes it unsuitable for first-line use.
The latest study found that Chantix was the primary suspect in 92 percent of suicides reported to the FDA for smoking cessation drugs over a period stretching nearly 13 years, and accounted for 10 times more reports of suicidal behavior and depression than the Zyban antidepressant, which is also used to treat smoking habits and is sold by GlaxoSmithKline. Another nugget: after adjusting for seven possible confounding factors, Chantix was 8.4 times more likely to result in a report to the FDA of suicidal behavior or depression than other nicotine replacement products.
The study, which was published yesterday in PLoS One, analyzed 3,249 case reports of serious injury in the FDA’s Adverse Event Reporting System from 1998 through September 2010 for self-injurious behavior or depression linked to Chantix, Zyban and nicotine replacement products. They found 2,925, or 90 percent, of the cases of suicidal behavior or depression were linked to Chantix, while 229, or 7 percent, were listed for Zyban. The remaining 3 percent were cited for nicotine replacement products (here is the study).
The findings, essentially, contradict the conclusion reached by the FDA just last month. The agency had sponsored two observational studies of neuropsychiatric adverse events with Chantix, and decided that neither one found a difference in risk of neuropsychiatric hospitalizations between Chantix and nicotine replacement therapy (back story).
The proclamation was the second such win for Pfizer, which has struggled to make a success of Chantix. The European Medicines Agency recently ruled that Chantix benefits outweigh any heart risks (see this and this), although psychiatric issues have generated the most controversy. And in addition to concerns about suicide, studies have also suggested Chantix can be responsible for violent behavior (look here and here). Meanwhile, Chantix is banned for use by pilots, air traffic controllers, military missile crews and pilots, and is also restricted for truck drivers.
As for the PLoS researchers, they are unapologetic. The "two newly released, but unpublished FDA studies of psychiatric hospitalizations provide little useful safety information, because the serious psychiatric side effects of smoking cessation treatment do not normally result in hospitalization. Notably suicide, aggression, depression, and assault would not normally result in hospitalization," study co-author Thomas Moore, a senior scientist with the Institute for Safe Medication Practices and who serves as a consulting expert in the civil litigation regarding Chantix, writes us in an e-mail. "In all, 82.2 percent of serious psychiatric side effects for (Chantix) that were reported to the FDA did not result in hospitalization and would have been overlooked in the two FDA studies.
We should note that Moore has co-authored previous studies examining Chantix links to violent behavior. Another co-author, Joseph Glenmullen of the Department of Psychiatry-Cambridge Hospital, Harvard Medical School, is also an expert witness for plaintiffs in Chanitx litigation. And yet another co-author, Sonal Singh of Johns Hopkins University, has released a meta-analysis concluding Chantix is linked to cardiovascular risk.
As for Pfizer, a spokesman sends us this: "Based upon the limited data we have seen, the conclusions of the authors are inconsistent with both the FDA’s statement of last week ...The analysis does not appear to contain any new information with respect to Chantix. These same authors persist in publishing analyses based on a review of spontaneous reports. It is important to remember that post-marketing reports do not establish a cause and effect relationship between a medicine and a reported adverse event.
"Also, it is not appropriate to draw conclusions based on comparisons between different drugs and reporting rates. Post-marketing reports can come from any source ranging from patients to healthcare providers, and from phone calls to internet postings and lawyers. Often these reports lack sufficient medical information to enable meaningful assessment. Because of the many limitations of post-marketing reports, the conclusions made by the authors are not supported by the data."
Pfizer, by the way, is conducting a large, double blind placebo controlled safety clinical trial to assess neuropsychiatric safety in patients with and without psychiatric disorders. However, results are not expected until 2017.
cig smoke thx to jo naylor on flickr
25 Comments
I've never before seen anyone attempt to analyze the safety of a drug using the method described in this paper. But if I had guessed someone would do so, I would have guessed that Dr. Singh would be involved somehow.
Of fourteen comments published in CMAJ on Dr. Singh's last Chantix paper, 12 were criticisms of his misuse of statistics.
We hope other readers will rely on more reliable scientific sources in judging the excellent analysis by my colleague Dr. Singh. His report on the cardiovascular effects of Chantix was confirmed by an independent FDA analysis of related data. It resulted in warnings in both the United States and Europe.
As we have reported in QuarterWatch his paper underestimates cardiovascular side effects by concentrating only the most serious events. We believe numerous other adverse effects such as hypertension, blurred vision, confusion, and loss of consciousness may be related to the drug's effects on blood pressure regulation.
I would not characterize the application of CV rates from patients with a history of CV data as "reliable". But of course readers can examine the criticisms of Dr. Singh's paper on the CMAJ website for themselves and form their own opinion.
The FDA did not endorse Dr. Singhs estimate of risk in patients without a history of CV disease. The EMEA rejected it as flawed.
The present study compares spontaneous reports of psychiatric AEs for a drug for which the existence of a possible issue had recieved substantial publicity to antibiotics, most of which are well known by physicians not to enter the CNS. The study is therefore hopelessly confounded by attribution bias on the part of reporting physicians. This is part of why AERS is considered to be useful for hypothesis generation, and not for providing measurements of risk.
Sorry, that first sentence should have read "I would not characterize the application of CV event rates from patients with a history of CV disease to patients lacking such a history as “reliable”"
sorry for the serial posts, but I would like to mention that I think it would have been appropriate for Mr. Moore to have disclosed that Curt Furberg, a co-author on both of the papers whose merits are being debated here, is on the board of Quarterwatch.
http://www.ismp.org/quarterwatch/2008q1.pdf
So much for the implied, independent endorsement.
So who is holding the fire to the feet of the FDA? The overdue report from the FDA was childlike with a deadly twist. They released a half baked study with an endpoint non-reflective of real life chantix users with reported side effects. The FDA admits the studies were flawed, but apparently continued on with their Pfizer like nonsense. Who ordered the flawed hospitalizations outcome? I have stated more than once that Pfizer must be held accountable for their continued role of chantix. The FDA sadly has embraced pharma at the expense of American’s lives.
John, where were you when the smoking cessation expert Jonathan Foulds posted his views here? He assumed it was not notable to tell the patients in his smoking cessation clinic that he was being paid by Pfizer to promote their drug, and no one cared. Get off your high-horse and let the truth set you free. As the Pfizer spokesperson stated, the study conducted by the PLoS researchers is not new info to them. Pfizer has known from the beginning that chantix is dangerous, and they prefer to hide the details.
To me there is nothing more admirable than a researcher washed in the truth. Thank you PLoS researchers.
I should add that while I am no researcher, I have been stating the dangers of chantix and cardio related events among other serious side effects for some time now. How's that for spontaneous?
John Q, I am neither an apologist for Chantix, nor an expert on its risk benefit ratio. My issue is bad science, and in the case of these particular researchers, bad science coupled with the aggressive use of press releases.
A previous paper by these same researchers includes data showing that reports of SAEs for estrogen rose over 5-fold after the results of the WHI came out. This in itself shows that AERS reporting rates have no quantitative significance. Nonetheless, in their anti-Chantix zeal, they chose to ignore their own observations, and directly compare AE reports for a drug which had extensive bad publicity to one which had not.
Bad science, whether undertaken with the goal of protecting the public, increasing sales, or building one's professional profile and obtaining tenure, is not in the public interest. In the case of these particular studies, the science is so bad that it is difficult to tell whether it is merely an issue of incompetence or deliberate disingenuity.
So I don't think I'm "on a high horse" here. Are you? Have you read the critiques of Singh's paper by statistics experts over at CMAJ? Or have you discounted them in advance because Singh's conclusions are aligned with your preconceptions?
I just follow the money, John. Experts? Surely you jest.
John, the other question I would ask you is how does one get an 8-fold increase in suicide attempts/self injury as posited in the PLoS paper, but no increase in psychiatric hospitalizations as seen in the more rigorous FDA cohort study?
Did you read the comments John, or are you just going to engage in ad hominem attacks on the commenters because they disagree with your preconceptions?
Following the data is why we stopped burning witches.
So you are suggesting there must be a hospitalization to document a true side effect? John this is far worse than burning witches. The amount of people injured with severe side effects from this rush to market novel drug surpass the witch count. And that excludes the ones that are dead due to Pfizer's non-money making scheme. But hey who cares? The dead ones can not be counted, right? No documented hospitalization.
John, I disagree with you, but that is not sufficient reason to imply that I am indifferent to suicide.
The number of suicide attempts, self injuries, deliberate overdoses, and suicidal behaviors in the PLoS study outnumbers completed suicides by almost 2:1. I don't know if an increase in the rate of successful suicides would have shown up in the hospitalization study done by FDA, but an increase in attempts would. If there are 8-fold as many of these patients, there should be an observed increase in hospitalizations. None was seen.
What? Your first sentence makes no sense. I never implied because you disagree with the truth that you are heartless. It does show your ignorance concerning the deaths due to your novel drug though. Who told you all the deaths were suicide? Stay focused.
Do the numbers reported in the PLos study exceed the number of side effects and adverse effects reported to the FDA by consumers? How can the PLoS report on chantix not be accurate when AE's reported to the FDA in little over a year alone resulted in the most reported side effects of any drug ever? Also remember Pfizer "forgot" to report even more deaths to the FDA. The excessive amounts of AE's reported to the FDA triggered initial concern by the ISMP.
The failed decision by Pfizer and FDA to acknowledge the collected reports to physicians and patients of not only psychosis and psychotic disorders but also severe skin adverse reactions, vision disturbance, high blood sugar and diabetes, rapid skin and tissue swelling, irregular heartbeat, convulsions, strokes, numerous reports relating to cardiac and other causes of death were unethical and even felonious.
What other Chantix report or study relating to its side effects and deaths show documentation of hospitalization within 30 days and only then equates a reportable AE? No wonder Pfizer’s initial studies do not reflect a true and accurate report of AE’s. They were following the hospitalizations rule.
Once again John, no need to be insulting.
The problem is that all of these AEs occur in people who are not taking Chantix. The question then becomes, how many of these reported events are due to the drug, and how many are unrelated?
The Fda study, for all its limitations, attempts to address this issue by comparing two cohorts, one treated with Chantix and one with nicotine replcement. It seems reasonable to believe that if there is an increase in suicide attempts in pts treated with Chantix, there should be an increase in hospitalizations. This was not seen.
Assuming a fixed rate of hospitalizations to suicide attempts, the PLoS paper predicts an eight fold increase in hospitalizations. Unless you can suggest a reason why Chantix induced suicide attempts should lead to a lower hospitalization rate than other suicide attempts, the inescapable conclusion is that the methodology used in the PLoS paper is deeply flawed, and all of its conclusions are suspect.
As I mentioned before, I am not an expert on Chantix's overall risk / benefit ratio. What i can say is that this particular paper is deeply flawed. I also believe that spontaneous reporting data needs to take a back seat to relative AE rates determined by more reliable methods, and that comments about relative spontaneous report rates needs to take into account the well established effects of negative publicity and relative sales volume. Chantix outsells Zyban 10 to 1, something that both Ed and the authors of the PLoS paper neglected to mention when stating that it was responsible for 90% on smoking cessation product psychiatric AE reports.
John, you might also find it informative to look at the AE rates for the placebo groups in some of the smoking cessation product trials. In the following JAMA paper, AEs experienced by the placebo group included nausea (8.4%), insomnia (13%), irritability (5.8%), sleep disorders/abnormal dreams (9%), headache (12%), and diziness (6%). All in all, it looks like about 50% of placebo-treated patients experienced an AE, and 28% or more experienced psychiatric AEs.
http://jama.ama-assn.org/content/296/1/47.full.pdf
"The problem is that all of these AEs occur in people who are not taking Chantix” You are on a bucking bronco rodeo ride. Where do you come up with these wild claims?
John, I came up with these "wild claims" by looking at the table of contents of a pre-Chantix general medicine textbook.
Diabetes, psychosis, irregular heartbeat, stroke, convulsions, and the other symptoms you listed were all well established medical syndromes before Chantix reached the market.
If we universally adopt the standard of evidence you are espousing, it would no longer be necessary for companies to perform clinical trials for second indications. Roche could get its B-cell depleting drug Rituxan approved for RA, and then just encourage oncologists using the drug to write the FDA to report any successes they had using it to treat lymphoma. Pfizer would never have paid a $430M fine for off-label promotion of Neurontin because pain specialists routinely try new anticonvulsants and some of them would have written the FDA with their positive results, leading to approval of Neurontin for this indication.
Clearly you would not support this? Does a single standard of evidence apply, or does the standard depend on whether a claim supports one's preconceptions?
Spontaneous reporting rates are a very marginal standard of evidence, whether one is discussing AEs or evidence of efficacy.
Lay the textbooks down and try to stay in reality. Why did you try to make this about Dr Singh? Clearly you are underdeveloped and are light years away. Your verbiage is laughable until reality of the thousands who have suffered at the hands of Pfizer and FDA pushing their novel drug. You can't even have a truthful dialogue concerning this drugs largest AE's ever reported to the FDA.
I wonder when the reports will be published concerning this novel drugs addictiveness.
John, uou accuse me of "not staying in reality", of being "underdeveloped", "laughable", and "untruthful". Please note that I tried to stay focused on laying out the facts and reasoning behind my position. I'd appreciate it if you would do the same.
Here are my points, which I do not think you have responded to. A simple yes or no would be sufficient.
1) Do patients who have not been treated with Chantix ever experience adverse events such as psychiatric symptoms?
2) Might the fact that each Chantix prescription comes with a Medication Guide, listing all of the AEs seen in the Chantix clinical trials, irrespective of whether they were seen with equal incidence in the placebo groups, along with an FDA AERS phone number, possibly have an impact on the AERS reporting rates for Chantix?
2) Does the demonstration of a 5-fold increase in AERS reports for hormone replacement therapy in the aftermath of the WHI report suggest that AERS reporting rates are affected by factors other than absolute incidence rates?
3) Is it possible to get an 8-fold increase in suicide attempts without seeing any increase in psychiatric hospitalizations?
4) Might the fact that Chantix outsells Zyban by a ratio of 10:1 possibly be relevant the the higher number of AERS reports for the former?
5) When calculating risk factors, is it appropriate to assume that patients who have never had an MI before have the same underlying risk as patients who have previously experienced myocardial infarction?
Again, no name calling is needed. A simple yes or no is all that is needed.
John I am through with you. I knew who/what you are when you first posted under the guise of correcting bad science. Go tout and push your nonsense to someone who gives a flip. I don't play little kid games. I also do not inject a might in science when lives are on the line. You know I was thinking you "might" be able to go dig up a few people killed by chantix and see if they are still breathing since they probably just had a placebo effect. Let me know how that works for you.
For thousands of years, everyone "knew" there were witches, that they ought to be burned, and that anyone denying this was evil. Science, and following the data with an open mind is why we don't do that anymore.
I've tried to engage you in a rational discussion, but you've simply responded with personal attacks.
How do you ever learn anything new if you do not approach the facts with an open mind?
Chantix is a very powerful, unpredictable, scary, scary drug. I was THRILLED when it came on the market. 'm a closet smoker 2 - 4 per day and wanted to put this habit to rest...almost put mySELF to rest instead.
You have to know people don't go to the trouble and expense of getting this script...w/hopes of having devastating side effects! Had I NOT taken this mySELF, I MIGHT have had a hard time believing some of the horror stories...but I DID and can ASSURE you, alot of these people really, truly DID have the negative side effects they claim.
I took two scripts back/back in '07. Endured two long hospital stays in '08. To Pfizer's credit...it's TRUE, it DID make smoking a chore. YAYYYYY!!! Is that IT??!! Am I done?? Did I just quit smoking??!!! Wow, THAT was easy!!! Then after a couple months the cravings came back. So, took aNOTHER regimen of Chantix. This time something went terribly, terribly wrong. I remember feeling beYOND exhausted, ran on empty for many months, dreams turned weird. VERY weird, learned later they could've been hallucinations.
Dreamt of deceased loved ones coming to my room at nite, begging me to come w/them. Said, "heaven is a blast." Also said, "your job is done here! Your girls are married w/children of their own! They don't need you! Your husband of 37 years could certainly find a young replacement and the same goes for your job you've loved for the past 23 years.
Finally listened. Please, listen to ME...this not a conspiracy...the whole world didn't get together and compare notes and try to come up w/believable stories. These stories are REAL. You can't MAKE this crap up!!
While on the right path...obviously Pfizer has some MAJOR KINKS to work out. Until that can happen THIS LITTLE BLUE PILL HAS GOT TO GO!!!!
I would like to know why the two experts did not speak to Terry's comment? Her story is far from the only one of its kind. What other research do we need? I just lost a friendto suicide that was taking Chantix. No history of depression no prior or current hospitilizations. He is dead now.
Take it off of the market already. Every smoker wants to quit but obviously not at.the cost this drug carries...