Patient disposition. Adverse events (AEs) were laboratory abnormalities considered AEs and treatment-emergent and non–treatment-emergent AEs. Other indicates protocol violation, subject failed to meet entry criteria, noncompliance, and personal reasons.
Carbon monoxide–confirmed continuous quit rates for the all-subjects population. *P≤.05. †P≤.01. ‡P≤.001.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Nides M, Oncken C, Gonzales D, Rennard S, Watsky EJ, Anziano R, Reeves KR, . Smoking Cessation With Varenicline, a Selective α4β2 Nicotinic Receptor Partial AgonistResults From a 7-Week, Randomized, Placebo- and Bupropion-Controlled Trial With 1-Year Follow-up. Arch Intern Med. 2006;166(15):1561–1568. doi:10.1001/archinte.166.15.1561
Copyright 2006 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2006
Currently available smoking cessation therapies have limited success rates. Varenicline tartrate is a novel, selective nicotinic receptor partial agonist developed specifically for smoking cessation. This study evaluated the efficacy, tolerability, and safety of 3 varenicline doses for smoking cessation. Bupropion hydrochloride was included as an active control.
A phase 2, multicenter, randomized, double-blind, placebo-controlled study of healthy smokers (18-65 years old). Subjects were randomized to varenicline tartrate, 0.3 mg once daily (n = 128), 1.0 mg once daily (n = 128), or 1.0 mg twice daily (n = 127), for 6 weeks plus placebo for 1 week; to 150-mg sustained-release bupropion hydrochloride twice daily (n = 128) for 7 weeks; or to placebo (n = 127) for 7 weeks.
During the treatment phase, the continuous quit rates for any 4 weeks were significantly higher for varenicline tartrate, 1.0 mg twice daily (48.0%; P<.001) and 1.0 mg once daily (37.3%; P<.001), than for placebo (17.1%). The bupropion rate was 33.3% (P = .002 vs placebo). The carbon monoxide–confirmed continuous quit rates from week 4 to week 52 were significantly higher in the varenicline tartrate, 1.0 mg twice daily, group compared with the placebo group (14.4% vs 4.9%; P = .002). The bupropion rate was 6.3% (P = .60 vs placebo). Discontinuation owing to treatment-emergent adverse events was 15.9% for bupropion, 11.2% to 14.3% for varenicline, and 9.8% for placebo. No dose-related increases occurred in adverse events for varenicline.
Varenicline tartrate demonstrated both short-term (1 mg twice daily and 1 mg once daily) and long-term efficacy (1 mg twice daily) vs placebo. Varenicline was well tolerated and may provide a novel therapy to aid smoking cessation.
Cigarette smoking remains the world's leading cause of preventable death,1 contributing to 5 million premature deaths in 2000,2 which is estimated to increase to 10 million by 2020.1 Surveys show that most smokers want to quit,3 but most attempts are unaided, with success rates of only 3% to 5% at 1 year.3 Current pharmacotherapies, such as nicotine replacement therapy (NRT), bupropion hydrochloride, and nortriptyline hydrochloride, have shown moderate success, typically doubling short-term quit rates vs placebo,4- 7 with success at 1 year averaging approximately 7% to 30%, depending on the level of adjunctive behavioral counseling.8,9 Consequently, additional, more efficacious smoking cessation medications are needed.
Varenicline tartrate is a novel, nonnicotine agent developed expressly for smoking cessation. It is a selective nicotinic acetylcholine receptor partial agonist that binds specifically at the α4β2 nicotinic receptor subtype.10 The α4β2 receptor is thought to mediate the rewarding properties of nicotine by modulating the release of dopamine in the nucleus accumbens.11- 13 Cytisine, a plant-derived α4β2 partial agonist used for many years as a smoking cessation aid in eastern Europe,14 provided a structural starting point for the development of the higher-affinity varenicline. The agonist effect of oral varenicline on dopamine release is 35% to 60% of that observed with nicotine,10 theoretically sufficient to attenuate craving and withdrawal without producing its own dependence syndrome. The slower release of dopamine with varenicline compared with smoking would also reduce any potential for abuse.10 Varenicline also has a competitive antagonist effect on nicotine due to a substantially higher affinity for the α4β2 receptor.10 Starting therapy 1 week before the target quit day could potentially lead to at least partial extinction of smoking behavior by blocking the rewarding effects of smoked nicotine.15,16 In addition, the blockade of reward could reduce the chance that a “slip” while still undergoing treatment would lead to a full-blown relapse.
The current study was part of a phase 2 program conducted to select the optimal dose for larger-scale, phase 3 studies. The primary objectives were to assess the efficacy, tolerability, and safety of 3 doses of varenicline administered for 6 weeks. A bupropion arm was included as an active control.
This randomized, multicenter, double-blind, parallel-group, placebo- and active-controlled phase 2 clinical trial was conducted at 7 US sites from February 21, 2000, to January 3, 2003. Before the start of the study, a randomization list was computer generated using a method of randomly permuted blocks and a pseudo-random number generator. Investigators assigned medication to subjects in numerical order of acceptance into the study. Randomized subjects received 1 of 3 varenicline tartrate dose regimens (0.3 mg once daily, 1.0 mg once daily, or 1.0 mg twice daily), sustained-release bupropion hydrochloride (150 mg twice daily), or matched placebo. Varenicline doses were selected on the basis of tolerability data from phase 1 studies, and subjects were dosed for 6 weeks, receiving blinded placebo during week 7 to preserve treatment blinding. Bupropion, the primary, non–nicotine-based treatment currently prescribed for smoking cessation, was included as an active control. In accordance with US labeling recommendations, bupropion hydrochloride was dosed for 7 weeks, with titration from 150 mg once daily (days 1-3) to 150 mg twice daily through week 7. All subjects took study medication for 1 week before attempting to quit smoking on day 8 of the study.
During the 7-week treatment phase, subjects visited the study site weekly for efficacy and safety evaluations and up to 10 minutes of standardized, individual smoking cessation counseling from trained staff. Subjects were also given the Clearing the Air: How to Quit Smoking . . . and Quit for Keeps17 smoking-cessation booklet at the baseline visit.
After completing the 7-week treatment phase, subjects had the option to participate in the non–drug treatment phase, which continued through week 52. Continuing subjects had clinic visits at weeks 12, 24, and 52, where vital signs and smoking status were assessed, along with additional brief smoking cessation and relapse prevention counseling. Subjects were also contacted by telephone every 4 weeks beginning with week 16 and assessed for their use of cigarettes, other forms of tobacco, or any other smoking cessation products since the previous study contact.
Subjects were male and female smokers between 18 and 65 years old who were in general good health as determined by a detailed medical history, limited physical examination, electrocardiogram (ECG), and clinical laboratory tests. Subjects were required to have smoked an average of 10 cigarettes per day during the previous year, without a period of abstinence of more than 3 months. Exclusion criteria were major depression requiring treatment within the past year; history of panic disorder, psychosis, or bipolar disorder; history of anorexia nervosa or bulimia; treatment with bupropion within the past year; history of seizures or cardiovascular disease; uncontrolled hypertension; history of clinically significant allergic, hematologic, renal, endocrine, pulmonary, hepatic, gastrointestinal, or neurologic disease; alcohol or other drug abuse within the past year; or use of NRT within the past 3 months. Subjects who discontinued use of study medication prematurely were allowed to remain in the study.
This study was conducted in compliance with the Declaration of Helsinki. The study protocol and amendments were approved by the institutional review board for each site, and before study entry, all subjects signed informed consent forms approved by the sponsor and the site institutional review board.
Subjects kept daily diaries of the number of cigarettes smoked from baseline through week 7. Exhaled carbon monoxide (CO)levels were measured at each clinic visit through week 52, using a breath CO monitor (Bedfont EC50 Micro III Smokerlyzer, Bedfont USA, Medford, NJ). At each clinic and telephone visit beginning with week 1, subjects were asked whether they had smoked in the previous 7 days and since the previous visit.
The primary efficacy measure was the continuous quit rate (CQR) for any 4 weeks, defined as abstinence for any consecutive 28-day period during the treatment phase (determined by diary data). This measure was chosen to give the best possibility of detecting an efficacy signal in this early phase 2 study. Secondary efficacy measures included the CO-confirmed (≤10 ppm) 4-week CQR for weeks 4 to 7, as well as CQRs from week 4 to weeks 12, 24, and 52. Subjects who dropped out for any reason were considered to be smokers at all subsequent time points. Craving was assessed with the urge to smoke item of the Minnesota Nicotine Withdrawal Scale (MNWS)18 and the 10-item Brief Questionnaire of Smoking Urges (QSU-Brief).19 Withdrawal was evaluated using the remaining 8 items of the MNWS. The MNWS and QSU-Brief data were collected daily for the first 2 weeks and at each weekly visit through week 7.
The Modified Cigarette Evaluation Questionnaire (mCEQ) assesses the reinforcing effects of smoking through 12 questions that collectively make up 5 subscales: smoking satisfaction, psychological reward, enjoyment of respiratory tract sensations, craving relief, and aversion. Subjects completed the mCEQ daily through week 1 and at each weekly visit through the week 7 visit if they had smoked since the previous visit.15,20
Body weight was evaluated at each weekly visit during the treatment phase and summarized separately for smokers and “cessators” (subjects who did not smoke any cigarettes from the target quit date to the day of measurement, based on the daily smoking diary). Inferential analyses were not performed.
Assessments of adverse events (AEs), clinical laboratory measures, vital signs, a 12-lead ECG, and a physical examination were conducted. The AEs were recorded during each weekly visit. Serious AEs were reported from randomization through 30 days after the last dose of study medication. Those AEs that occurred after 30 days were reported if the investigator considered them related to the study medication. Samples for clinical laboratory evaluation were collected at screening, baseline, and weeks 1, 2, 4, 6, and 7. Limited physical examinations were conducted at screening or baseline and at the week 7 visit.
Sample size (approximately 125 per treatment group) was determined by detecting a clinically meaningful difference in response rates for active treatment vs placebo (assuming 38% vs 20%, respectively) on the primary efficacy variable with 80% power (α = .05, 2-tailed). Analyses are reported here for the all subjects population (those who reported taking ≥1 dose of study medication) for each treatment group vs placebo. The study was not powered for statistical analyses comparing varenicline with bupropion. All significance tests were 2-tailed using an overall level of significance of α = .05. For the primary end point, the Dunnett adjustment for multiple comparisons was used to preserve the family-wise type I error rate at α = .05. No adjustments for multiple comparisons were made for secondary end points. Four-week CQRs and other binary response rates were analyzed using a logistic regression model, including treatment and center, and testing was performed with the likelihood ratio χ2 test. Odds ratios (ORs) and 95% confidence intervals (CIs) reported for each active treatment group vs placebo are the least squares mean estimates from the logistic regression model. For continuous end points (MNWS, QSU-Brief, and mCEQ), inferential analyses were performed using an analysis of variance model, including baseline value of the end point as a covariate and the fixed effects of treatment and center.
Subject disposition for the treatment phase is shown in Figure 1. A total of 638 subjects were randomized to treatment. Twelve subjects did not take any study medication (2 in each active treatment group, 4 in the placebo group); 626 subjects were therefore included in the all subjects population and evaluated for safety and efficacy. The percentage of subjects who completed 7 weeks of study medication was similar for each group. The most frequent reasons for study discontinuation during the treatment phase were AEs and subject default (ie, withdrew consent, lost to follow-up). Of the subjects treated, 75.6% across treatment groups entered the nontreatment phase. Of those who were continuously quit from weeks 4 to 7, only 1 subject (in the placebo group) did not continue onto the nontreatment phase. Of those in the treatment group, 56.4% completed the week 52 visit.
Demographic and baseline characteristics for the all subjects population at screening are given in Table 1. Smoking history and dependence were similar across treatment groups; subjects represented a population of smokers with a mean consumption of approximately 20 cigarettes per day for an average of 24 years. Forty-four percent had previously used transdermal NRT. The frequency of previous bupropion use ranged from 13.0% to 20.6% across treatment groups.
The 4-week CQRs were significantly higher for varenicline tartrate, 1.0 mg twice daily (48.0%; OR, 4.71; 95% CI, 2.60-8.53; P<.001) and 1.0 mg once daily (37.3%; OR, 2.97; 95% CI, 1.63-5.40; P<.001), vs placebo (17.1%) and for bupropion (33.3%; OR, 2.53; 95% CI, 1.38-4.63; P = .002) vs placebo. The response rate increased with increasing dose of varenicline. Although the response rate for varenicline tartrate, 0.3 mg once daily, was numerically higher than placebo (28.6%; OR, 1.97; 95% CI, 1.07-3.65; P = .03), it did not reach statistical significance after applying the Dunnett adjustment for multiple comparisons (the Dunnett correction for 4 contrasts vs control requires P<.015 for significance at α = .05).
Similar results were seen for the CO-confirmed CQRs from weeks 4 to 7, with quit rates increasing with increasing varenicline dose. Quit rates for all 3 varenicline doses and for bupropion were statistically superior to that of placebo (Figure 2). Response rates were 3 times greater for varenicline tartrate, 1.0 mg twice daily, vs placebo compared with a bupropion response rate of approximately twice that of placebo.
Figure 2 also shows the rates of CO-confirmed CQRs from week 4 to weeks 12, 24, and 52. The CQRs for the varenicline tartrate, 1.0 mg twice daily, group were significantly higher than that for the placebo group at each time point. By week 52, the response rate had more than tripled. The rate for bupropion was significantly higher than that for placebo only at week 12.
Craving, as assessed with both MNWS item 1 and the QSU-Brief total score, was consistently reduced with varenicline tartrate, 1.0 mg twice daily, compared with placebo, reaching statistical significance at all weekly time points (Table 2). Bupropion reduced craving compared with placebo, although the differences reached statistical significance at fewer weekly time points across the 2 instruments. In general, withdrawal symptom scores were mild in all treatment groups, as measured by the composite score for MNWS items 2 to 9, with no clear treatment effect on composite score change from baseline.
Table 3 gives the mCEQ scores by treatment group at week 1, the day before quit day. This time point was chosen because it shows the potential effects of 1 week of receiving treatment while virtually the entire sample was still smoking. The varenicline tartrate, 1.0 mg twice daily, group demonstrated statistically significant differences from placebo on 3 subscales: smoking satisfaction, enjoyment of respiratory tract sensations, and aversion. The other treatment groups did not show any significant differences on any of the subscales. Interpretation of the aversion subscale may be confounded for varenicline because 1 of the 2 items measures nausea, an AE reported more frequently for patients taking varenicline compared with those taking placebo.
Mean weight gain from baseline to week 7 was generally greater among cessators than smokers. Among cessators, mean weight gain was numerically higher in the placebo group (+4.00 kg, n = 10) compared with the active treatment groups: varenicline tartrate, 0.3 mg once daily (+2.47 kg, n = 15), 1.0 mg once daily (+2.14 kg, n = 14), and 1.0 mg twice daily (+1.96 kg, n = 24), and bupropion (+1.68 kg, n = 22).
Table 4 gives the treatment-emergent AEs that occurred in 10% or more of subjects in any of the active treatment groups. Varenicline was safe and well tolerated at all 3 doses. The frequency of discontinuations related to treatment-emergent AEs was lowest among the placebo-treated subjects (9.8%) and highest in the bupropion group (15.9%). In the varenicline treatment groups, the rate of discontinuation due to treatment-emergent AEs does not appear to be dose related. The AEs that occurred most frequently among varenicline-treated subjects were nausea, insomnia, headache, abnormal dreams, and taste perversion. The incidence of these AEs increased with increasing dose, except for headache. Nausea was mild to moderate in severity and typically transitory (median duration ≤12 days), with most episodes beginning within the first week of treatment across all varenicline groups. Discontinuation owing to nausea was low: for varenicline tartrate, 1.6% in the 0.3 mg once daily group, 0.8% in the 1.0 mg once daily group, and 4.0% in the 1.0 mg twice-daily group; for bupropion, 0.8%; and for placebo, 0.0%. Depression was not observed as an AE with varenicline treatment. No deaths occurred during the study.During the treatment phase, only 1 patient in the varenicline tartrate, 1.0 mg twice daily, group experienced a serious AE (transient ischemic attacks in a subject with mild stenosis of the ipsilateral common carotid artery), whereas 4 subjects in the bupropion group experienced serious AEs (persistent intermittent bloody diarrhea, syncope, and convulsion [2 subjects]). All serious AEs were considered by the investigator to be possibly related to the study drug.
Results of clinical laboratory tests, ECGs, and vital signs demonstrated no safety issues of concern. The frequency of clinically significant laboratory test abnormalities was low and similar across all treatment groups.
In this study, varenicline, in combination with brief behavioral counseling, was highly efficacious for short- and long-term smoking cessation compared with placebo. Efficacy improved as the dose increased, with varenicline tartrate, 1.0 mg twice daily, providing the highest rates of continuous abstinence across all treatment groups, including bupropion. Moreover, varenicline tartrate, 1.0 mg twice daily, significantly reduced craving and several aspects of smoking reinforcement compared with placebo, supporting the hypothesized agonist and antagonist qualities of this selective α4β2 nicotinic receptor partial agonist. Varenicline exhibited a good safety and tolerability profile across all doses.
The approximate tripling of response rates observed for varenicline tartrate, 1.0 mg twice daily, compares favorably with previously reported studies for NRT and bupropion where brief behavioral counseling was also included. Meta-analyses of NRT typically find a doubling of success rates vs placebo at the end of treatment and at 1 year.5 Although bupropion exhibited the previously reported doubling of continuous abstinence from weeks 4 to 7,21- 23 why it did not demonstrate significant separation from placebo after week 12 is unclear. One potential limitation was the inclusion of subjects who had been previously exposed to bupropion. Two phase 3 studies24,25 that included only bupropion-naïve smokers and were appropriately powered to compare varenicline with bupropion have been completed.
Future studies will investigate whether a longer treatment period with varenicline would increase quit rates. In this study, the rate of relapse (smoking ≥1 puff) from week 7 to week 12 for continuous quitters from weeks 4 to 7 was similar for varenicline tartrate, 1.0 mg twice daily (29.3%), compared with bupropion (27.4%) and somewhat greater than placebo (21.4%). Relapse prevention remains a continuing challenge for any smoking cessation treatment.
The patient-reported data from the MNWS, QSU-Brief, and mCEQ support the hypothesis that varenicline's partial agonist and antagonist mode of action would also reduce the craving and the reinforcing effects of smoking. Varenicline tartrate, 1.0 mg twice daily, demonstrated a consistent and significant reduction in craving vs placebo over 2 separate patient-reported assessments. Moreover, although subjects were still smoking during the first week of treatment, only varenicline tartrate, 1.0 mg twice daily, was effective in reducing the reinforcing effects of smoking as measured by the smoking satisfaction and enjoyment of respiratory tract sensations subscales of the mCEQ. These reward blockade effects may promote at least partial extinction of smoking behavior during the prequit period and prevent a postquit slip from providing enough reward to trigger a full-blown relapse. Other novel treatments that combine agonist and antagonist effects are being investigated. Rose et al15,26,27 have demonstrated that a combination of nicotine and mecamylamine hydrochloride, a nicotine receptor antagonist, can increase smoking abstinence rates. Buprenorphine hydrochloride, an opioid receptor partial agonist, has also demonstrated effectiveness for treating opioid dependence.28
Varenicline was safe and well tolerated at all 3 doses. Discontinuation rates for each dose were similar to those seen in the placebo and bupropion groups. Nausea was a frequent AE but was mainly transient and mild to moderate in severity and infrequently led to discontinuation of study medication.
In summary, varenicline is a novel nonnicotine agent designed specifically for smoking cessation. In this study, varenicline tartrate, 1.0 mg twice daily, effectively helped subjects quit smoking, with response rates 3 times higher than those for placebo while demonstrating a good tolerability profile in this population of smokers who on average had smoked approximately 20 cigarettes per day for approximately 24 years. Efficacy was maintained in the non–drug treatment phase through week 52. The significant reductions in craving and in some of the rewarding effects of smoking seen with varenicline tartrate, 1.0 mg twice daily, may assist in promoting abstinence and preventing relapse.
Correspondence: Mitchell Nides, PhD, Los Angeles Clinical Trials, Suite 308, 2990 S Sepulveda Blvd, Los Angeles, CA 90064 (firstname.lastname@example.org).
Accepted for Publication: March 30, 2006.
Group Members: Mitchell Nides, PhD, Los Angeles Clinical Trials, Los Angeles; Alexander Glassman, MD, New York State Psychiatric Institute, New York; Cheryl Oncken, MD, MPH, University of Connecticut Health Center, Farmington; David Gonzales, PhD, Oregon Health & Science University, Portland; Stephen Rennard, MD, University of Nebraska Medical Center, Omaha; Elbert D. Glover, PhD, Department of Public & Community Health, University of Maryland, College Park; and Sharon Allen, MD, PhD, University of Minnesota, Minneapolis.
Financial Disclosure: Dr Nides has received research grants, consulting fees, and honoraria from Pfizer, Sanofi-Aventis, and GlaxoSmithKline. Dr Oncken has received research grants, consulting fees, and honoraria from Pfizer; received, at no cost, nicotine replacement and placebo products from GlaxoSmithKline for smoking cessation studies; and received honoraria from Pri-Med. Dr Gonzales reports having received research contracts from Pfizer, Sanofi-Aventis, GlaxoSmithKline and Nabi Biopharmaceuticals; consulting fees and honoraria from Pfizer, Sanofi-Aventis, and GlaxoSmithKline; and owning 5 shares of Pfizer stock. Dr Rennard has had or currently has a number of relationships with companies that provide product and/or services relevant to outpatient management of chronic obstructive pulmonary disease. These relationships include serving as a consultant (Adams, Almirall, Altana, Array Biopharma, AstraZeneca, Aventis, Biolipox, Centocor, Dey, Critical Therapeutics, GlaxoSmithKline, Johnson & Johnson, Merck, Novartis, Ono Pharma, Otsuka, RJ Reynolds, Roche, Sankyo, Schering-Plough, Scios, and Wyeth); advising regarding clinical trials (Altana, AstraZeneca, Aventis, Centocor, GlaxoSmithKline, Novartis, Pfizer, and Philip Morris); speaking at continuing medical education programs; and performing funded research at both basic and clinical levels (Altana, AstraZeneca, Boehringer Ingelheim, GlaxoSmithKline, and Novartis). He owns no stock in any pharmaceutical companies. Drs Watsky and Reeves and Mr Anziano are employees of Pfizer and own Pfizer stock or have stock options.
Role of the Sponsor: As the sponsor, Pfizer provided funding and was involved in all elements of the study, including, but not limited to, the study design and monitoring.
Acknowledgment: We gratefully acknowledge the contribution of fellow Varenicline Study Group colleagues Sharon Allen, MD, Elbert Glover, PhD, Alexander Glassman, MD, Clare B. Billing, MS, Jotham Coe, PhD, and Kelly Stein-Marcus, PhD.