No Suicide, Psychosis Risk With Varenicline?

Megan Brooks

June 05, 2015

A large Swedish population-based study finds no evidence of a causal link between the smoking cessation drug varenicline (Chantix, Pfizer Inc) and suicidal behavior, psychoses, criminal acts, or traffic accidents or violations. However, at least one expert remains unconvinced.

The study did show an increased risk for mood and anxiety conditions during periods of varenicline treatment in patients with preexisting psychiatric disorders, a fact that needs to be confirmed using other study designs, the researchers note.

Led by Seena Fazel, MD, of the Department of Psychiatry, University of Oxford, in the United Kingdom, the study was published online June 2 in the BMJ.

Black Box Warning

Since it was approved as a smoking cessation aid, varenicline has been linked to serious neuropsychiatric symptoms, including suicidal thoughts and behavior, depression, psychoses, and violence, prompting regulatory agencies in Europe and the United States to issue warnings.

In 2009, the US Food and Drug Administration (FDA) placed a black box warning on the product to highlight the risk. In 2014, an FDA panel revisited the issue and voted to maintain the black box warning, pending completion of an ongoing postmarketing randomized controlled trial (RCT) designed to capture serious neuropsychiatric adverse events.

Varenicline was also reported to increase the risk for traffic accidents and has been restricted or prohibited in several professions in the transportation industry, including those of pilots, air traffic controllers, and truck and bus drivers, as well as certain occupations in the military.

But Dr Fazel and colleagues note that their study does not support an association of varenicline use with risk for suicidal behavior or traffic accidents, and they suggest that prior reports of a link may not have taken full account of underlying risk factors.

"Links between varenicline and certain outcomes, such as traffic accidents and violence, that have been previously reported are mostly based on case reports and postmarketing surveillance. These are prone to selective reporting and various biases, including recall bias and other confounds," Dr Fazel told Medscape Medical News.

"There are two new aspects to the study," Dr Fazel explained. "First is that we examined multiple adverse risks in a total population of individuals being prescribed varenicline. Second, we report new observational data relating to crime outcomes, transport-related accidents, and traffic offenses, which have not been subject to meta-analyses of the trial evidence for various reasons, including feasibility."

The findings are based on roughly 7.9 million Swedish people aged 15 and older, of whom 69,757 (43,861 women) were prescribed varenicline between 2006 and 2009. In the whole population, 337,393 new psychiatric conditions were diagnosed during the study period. In addition, 507,823 suspected crimes and 338,608 criminal convictions, 40,595 suicidal events, 124,445 transport accidents, and 99,895 suspected traffic crimes and 57,068 traffic crime convictions were recorded.

In "between person" analyses, adjusted for age and sex, people taking varenicline had increased hazard ratios (HRs) of the adverse events investigated, the researchers note.

However, in their "principal analytical approach" comparing periods of treatment with periods of nontreatment within the same individual to control for confounding by indication, varenicline was not associated with significant risk for suicidal behavior, criminal offense, transport accidents, traffic violations, or psychoses.

"Our findings in relation to risks of psychosis and suicidality are consistent with syntheses of RCT evidence, which strengthens reports of the safety of varenicline for such outcomes," said Dr Fazel.

However, varenicline was associated with an increase in the risk for anxiety conditions (HR, 1.23; 95% confidence interval [CI], 1.01 - 1.51) and mood conditions (HR, 1.31; 95% CI, 1.06 - 1.63), which was only seen in people with preexisting psychiatric disorders.

Continue to Follow FDA Guidance

Commenting on the findings for Medscape Medical News, Sonal Singh, MD, MPH, of Johns Hopkins University School of Medicine, in Baltimore, Maryland, who was not involved in the analysis, noted that the "statistically significant increase in new-onset anxiety and depression are concerning, given varenicline's mechanism of action."

"The onset in anxiety and depression may be confounded by smoking cessation itself, and there was some support for this when we compared risks in individuals taking bupropion; when we compared risks of these conditions in individuals prescribed varenicline to those prescribed bupropion, there was a lower risk of mood disorders and no difference in risk of anxiety disorders. In addition, there is the possibility that individuals with psychiatric conditions who start taking varenicline become nonadherent to their other medications, but this needs examination in future research studies," Dr Fazel told Medscape Medical News.

Dr Singh also pointed out that, "given the low power of their study, as much a 37% increase in suicidal behavior and 84% increase in suspected traffic offense cannot be ruled out."

"It should be noted, in relation to study power, that the most recent meta-analysis of RCT evidence published by Thomas et al in the BMJ in 2015 reported on 15 suicidal events in the treatment arm of all the trials, compared with 1077 events in our study," Dr Fazel commented.

As reported by Medscape Medical News, this meta-analysis found no evidence of an increased risk for neuropsychiatric adverse events, including suicide, with varenicline compared with placebo.

But when the meta-analysis was published in March 2015, Dr Singh told Medscape Medical News that he believes it is flawed and that he will continue to warn patients about varenicline's potential for neuropsychiatric side effects, including suicide, as the FDA's warning advises physicians to do.

The study was funded by Wellcome Trust. The authors and Dr Singh report no relevant financial relationships.

BMJ. Published online June 2, 2015. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: