Varenicline for Smoking Cessation: Title and subTitle BreakDefinite Promise, But No Panacea

In this issue of JAMA, Gonzales and colleagues,¹ Jorenby and colleagues,² and Tonstad and colleagues³ report the results of 3 randomized trials on the efficacy of the nicotinic acetylcholine receptor partial agonist varenicline for achieving smoking cessation. For the clinician and the lay public, the major question from these trials is simply this: How effective and useful is varenicline for smoking cessation?

These 3 trials have several strengths that cogently argue that varenicline is efficacious for smoking cessation. Instead of the traditional 2-group active drug/placebo trial, 2 studies^{1 - 2} included a third group in which varenicline was compared with bupropion, a known efficacious first-line stop smoking agent.⁴ The trial by Jorenby et al² showed significantly better long-term (ie, at 52 weeks) cessation rates with use of varenicline compared with bupropion, and the trial by Gonzales et al¹ showed a trend in the same direction. In the trial by Tonstad et al,³ extended use of varenicline significantly reduced relapse at nearly a 1-year follow-up among persons who could successfully quit smoking while using the drug, based on continuous abstinence during weeks 11 to 12 of therapy.³

Another important strength is that varenicline represents an entirely new class of drug, providing clinicians and smokers with pharmacological alternatives. Until now, only 2 classes of smoking cessation aids were available, nicotine replacement with multiple delivery methods (eg, patch, inhaler, lozenge) and the antidepressant bupropion, which is thought to inhibit the reuptake of dopamine and norepinephrine.⁴ In contrast, varenicline is a nicotinic acetylcholine receptor partial agonist.^{1 - 3} With US Food and Drug Administration (FDA) approval of varenicline for smoking cessation, clinicians now have another class of drugs in their armamentarium.

Despite these strengths, it is also clear that these studies have several important limitations. Although varenicline was more efficacious than placebo (and in one study, more efficacious than bupropion as well), several issues temper some of the enthusiasm for this agent. First, the adverse effect profile of varenicline revealed that nearly 30% of participants reported nausea, a rate significantly higher than with either bupropion or placebo.^{1 - 2} Abnormal dreams were common and much more likely in the varenicline group.^{1 - 2} Another limitation involves the dropout rates that certainly were nontrivial, but similar to other pharmacological smoking cessation efficacy trials.⁴ However, the pattern of dropouts was concerning in that the highest rate of completion was in the varenicline group. An intention-to-treat analysis, in which all participants who dropped out of the study are assumed to be smoking, will favor the condition (in this case varenicline) with the highest completion rates and may have introduced bias.

In the study by Tonstad et al,³ varenicline was successful in preventing relapse at 24 weeks, but the conclusions are conditional on participants being abstinent from smoking at 12 weeks. By eliminating participants who failed to quit smoking in the open-label phase, the authors have eliminated approximately one third of individuals for whom this drug does not appear to be effective. Thus, the relapse prevention results reported are probably more optimistic than what would occur in a real-world situation. If patients who were not randomized (ie, were not abstinent at 12 weeks) are included as failures in this study, the continuous abstinence rates for weeks 11 to 52 for those taking the drug for 24 weeks are actually less than the continuous abstinence rates in the companion studies,^{1 - 2} in which patients received the drug for only 12 weeks.

In addition, the numerous exclusion and inclusion criteria in these studies may decrease the generalizability of varenicline results to the intended end users of this product. However, such strict criteria were arguably necessary safety parameters at this stage of research. Virtually all of the study sites were academic medical centers in which participants are likely to get the best of instructions and adherence monitoring. Moreover, it is well known that participants who enroll in and complete efficacy studies in academic medical centers tend to be better educated, more adherent, and healthier than those in other settings.^{5 - 7} The degree to which the findings from these studies can be generalized to real-world settings is unclear. However, many of these issues apply to most efficacy studies of pharmacological agents in the early stages of research on the product.

Another major concern is not necessarily related to the methods or results of the published articles but rather is the potential reaction to these reports by the public. Whenever a new stop smoking agent is approved by the FDA (and to some degree this occurs with any new drug), there is often unbridled enthusiasm regarding the potential to solve the problems associated with smoking. Starting with the release of nicotine gum, each product that the FDA approved for smoking cessation has received considerable media attention, which likely reflects the public health issues associated with smoking and how many people desperately need and want help to quit. Now, with the recent FDA approval of varenicline, the first new non–nicotine-containing agent for smoking cessation approved in almost a decade, the attention directed toward this drug is likely to be just as great.

It is important for clinicians to moderate some of the potential enthusiasm that is likely to occur as the result of the publication of these trials, FDA approval of the drug, and promotion by this manufacturer. On the one hand, these studies^{1 - 3} demonstrate that varenicline is associated with higher smoking cessation rates than placebo and may produce better cessation rates than bupropion, a first-line–approved smoking cessation drug. Importantly, varenicline represents a third class of drug with probably a different mechanism of action than either nicotine replacement therapy or bupropion. On the other hand, varenicline definitely is not a panacea for smoking cessation. Many participants in these trials experienced adverse events, stopped taking their study medication before they should have, and discontinued participation in the studies. Importantly, the majority of participants in these 3 studies did not quit smoking even with varenicline.

Clearly, quitting smoking, even with pharmacological and behavioral assistance, is extremely difficult. Patients currently cannot and probably never will simply be able to “take a pill” that will make them stop smoking. Smokers must want to stop smoking and must be willing to work hard to achieve the goal of smoking abstinence.

Although much research needs to be conducted to establish the effectiveness of varenicline, stop smoking researchers and clinicians, as well as smokers wanting to quit smoking, now have another product available that appears to help increase the probability of smoking cessation.