Advertisement
Mayo Clinic Proceedings Home

Frequently Reported Adverse Events With Smoking Cessation Medications: Post Hoc Analysis of a Randomized Trial

Open AccessPublished:June 08, 2021DOI:https://doi.org/10.1016/j.mayocp.2020.10.046

      Abstract

      Objective

      To compare the incidence, severity, and clinical course of frequently reported adverse events (AEs) after treatment with smoking cessation pharmacotherapies.

      Methods

      This was a multinational, multicenter, post hoc analysis of frequently reported treatment-emergent AEs from a large, phase 4, double-blind, randomized, triple-dummy, placebo-controlled trial (EAGLES), conducted between November 30, 2011, and January 13, 2015, that included smokers with and without psychiatric disorders (N=8144). Treatments were varenicline 1 mg twice daily, bupropion sustained-release 150 mg twice daily, and nicotine patch 21 mg once daily with tapering (12-week treatment, 12-week nontreatment follow-up), with incidence, time to onset, and duration of frequently reported AEs (≥5% of participants in any treatment group) measured. Risk differences for AEs for varenicline and bupropion vs nicotine patch were compared.

      Results

      Across frequently reported AEs, nausea, insomnia, abnormal dreams, anxiety, irritability, dry mouth, fatigue, and application site pruritus differed significantly in active treatment vs placebo groups. Risk differences were as follows: for nausea with varenicline vs nicotine patch, 15.50% (95% CI, 13.20% to 17.80%); for insomnia with bupropion vs nicotine patch, 2.58% (CI, 0.65% to 4.51%); and for abnormal dreams with varenicline and bupropion vs nicotine patch, –2.49% (CI, –4.35% to –0.64%) and –5.60% (CI, –7.27% to –3.93%), respectively. Frequently reported AEs of severe intensity and treatment discontinuation were experienced by less than 1.5% and less than 3% of participants across all groups, respectively.

      Conclusion

      Active treatments were well tolerated with comparable AE profiles. Most AEs are not clinically important, and prescribers can reassure patients that those experienced will be manageable.

      Trial Registration

      Clinicaltrials.gov Identifier: NCT01456936.

      Abbreviations and Acronyms:

      AE (adverse event), EAGLES (Evaluating Adverse Events in a Global Smoking Cessation Study), FDA (US Food and Drug Administration), HCP (health care professional), NRT (nicotine replacement therapy)
      Global deaths due to tobacco-related diseases exceed 7 million per year.
      World Health Organization
      Tobacco.
      The use of US Food and Drug Administration (FDA)–approved smoking cessation pharmacotherapies can markedly increase the chance of a successful quit attempt
      • Cahill K.
      • Lindson-Hawley N.
      • Thomas K.H.
      • Fanshawe T.R.
      • Lancaster T.
      Nicotine receptor partial agonists for smoking cessation.
      ; however, most quit attempts do not involve an approved treatment.
      • Caraballo R.S.
      • Shafer P.R.
      • Patel D.
      • Davis K.C.
      • McAfee T.A.
      Quit methods used by US adult cigarette smokers, 2014-2016.
      Approved therapies include nicotine replacement therapies (NRTs), such as gum, lozenge, patch, inhalers, and nasal spray, as well as the non-nicotine medications varenicline and bupropion.
      US Food & Drug Administration
      Want to quit smoking? FDA-approved products can help.
      Despite their efficacy, smoking relapse rates remain high, and opportunities exist to improve patient utilization and adherence that correlate with increased abstinence.
      • Jamal A.
      • Dube S.R.
      • Malarcher A.M.
      • Shaw L.
      • Engstrom M.C.
      Centers for Disease Control and Prevention (CDC)
      Tobacco use screening and counseling during physician office visits among adults—National Ambulatory Medical Care Survey and National Health Interview Survey, United States, 2005-2009.
      • Kaplan R.M.
      • Fang Z.
      • Morgan G.
      Providers' advice concerning smoking cessation: evidence from the Medical Expenditures Panel Survey.
      • Pacek L.R.
      • McClernon F.J.
      • Bosworth H.B.
      Adherence to pharmacological smoking cessation interventions: a literature review and synthesis of correlates and barriers.
      • Hays J.T.
      • Leischow S.J.
      • Lawrence D.
      • Lee T.C.
      Adherence to treatment for tobacco dependence: association with smoking abstinence and predictors of adherence.
      Smokers may be reluctant to use pharmacotherapies stemming from general misperceptions and safety concerns,
      • Vogt F.
      • Hall S.
      • Marteau T.M.
      Understanding why smokers do not want to use nicotine dependence medications to stop smoking: qualitative and quantitative studies.
      • Mills E.J.
      • Wu P.
      • Lockhart I.
      • Wilson K.
      • Ebbert J.O.
      Adverse events associated with nicotine replacement therapy (NRT) for smoking cessation. A systematic review and meta-analysis of one hundred and twenty studies involving 177,390 individuals.
      • Smith A.L.
      • Carter S.M.
      • Chapman S.
      • Dunlop S.M.
      • Freeman B.
      Why do smokers try to quit without medication or counselling? A qualitative study with ex-smokers.
      including neuropsychiatric safety concurrent with varenicline or bupropion use
      • Buggy Y.
      • Cornelius V.
      • Fogg C.
      • Kasliwal R.
      • Layton D.
      • Shakir S.A.
      Neuropsychiatric events with varenicline: a modified prescription-event monitoring study in general practice in England.
      ,
      • Pasternak B.
      • Svanstrom H.
      • Hviid A.
      Use of varenicline versus bupropion and risk of psychiatric adverse events.
      and cardiovascular safety in smokers taking varenicline or NRT.
      US Food & Drug Administration
      FDA Drug Safety Communication: Chantix (varenicline) may increase the risk of certain cardiovascular adverse events in patients with cardiovascular disease.
      ,
      • Sobieraj D.M.
      • White W.B.
      • Baker W.L.
      Cardiovascular effects of pharmacologic therapies for smoking cessation.
      Adverse events (AEs) associated with FDA-approved smoking cessation pharmacotherapies have been well documented.
      • Cahill K.
      • Lindson-Hawley N.
      • Thomas K.H.
      • Fanshawe T.R.
      • Lancaster T.
      Nicotine receptor partial agonists for smoking cessation.
      ,
      • Mills E.J.
      • Wu P.
      • Lockhart I.
      • Wilson K.
      • Ebbert J.O.
      Adverse events associated with nicotine replacement therapy (NRT) for smoking cessation. A systematic review and meta-analysis of one hundred and twenty studies involving 177,390 individuals.
      ,
      • Hughes J.R.
      • Stead L.F.
      • Hartmann-Boyce J.
      • Cahill K.
      • Lancaster T.
      Antidepressants for smoking cessation.
      A large, phase 4, multinational, randomized, controlled, smoking cessation clinical trial (Evaluating Adverse Events in a Global Smoking Cessation Study [EAGLES]) was conducted to investigate the safety and efficacy of varenicline and bupropion vs active control (NRT patch) and placebo in smokers with or without an established diagnosis of psychiatric disorder.
      • Anthenelli R.M.
      • Benowitz N.L.
      • West R.
      • et al.
      Neuropsychiatric safety and efficacy of varenicline, bupropion, and nicotine patch in smokers with and without psychiatric disorders (EAGLES): a double-blind, randomised, placebo-controlled clinical trial.
      The results of EAGLES and its extension trial offered definitive evidence and reassurance to health care professionals (HCPs) and patients that use of these therapies is not associated with an increased risk of clinically important neuropsychiatric or major cardiovascular AEs.
      • Anthenelli R.M.
      • Benowitz N.L.
      • West R.
      • et al.
      Neuropsychiatric safety and efficacy of varenicline, bupropion, and nicotine patch in smokers with and without psychiatric disorders (EAGLES): a double-blind, randomised, placebo-controlled clinical trial.
      ,
      • Benowitz N.L.
      • Pipe A.
      • West R.
      • et al.
      Cardiovascular safety of varenicline, bupropion, and nicotine patch in smokers: a randomized clinical trial.
      EAGLES captured detailed individual participant data related to all volunteered, observed, and solicited AEs. Analysis of frequently reported AEs within such a comprehensive and unique data set may further assist HCPs in their clinical decision-making and treatment choices, help expand openness to treatment, alleviate patients’ concerns, and manage treatment expectations.
      The objective of this post hoc analysis of EAGLES safety data was to compare the incidence, severity, and clinical course of frequently reported AEs after treatment with the approved smoking cessation pharmacotherapies varenicline, bupropion, and NRT patch.

      Methods

       Study Design and Population

      EAGLES was a 24-week (12-week treatment, 12-week nontreatment follow-up), double-blind, active- (NRT; nicotine patch) and placebo-controlled, multicenter, parallel-group, triple-dummy study designed to assess the safety and efficacy of varenicline and bupropion for smoking cessation in smokers with and without psychiatric disorders. A detailed description of the EAGLES study has been published previously.
      • Anthenelli R.M.
      • Benowitz N.L.
      • West R.
      • et al.
      Neuropsychiatric safety and efficacy of varenicline, bupropion, and nicotine patch in smokers with and without psychiatric disorders (EAGLES): a double-blind, randomised, placebo-controlled clinical trial.
      Participants received varenicline 1 mg twice daily, bupropion 150 mg twice daily, NRT 21 mg once-daily (24-hour) patch with tapering, or placebo. The study included a 12-week treatment phase and a 12-week nontreatment follow-up.
      EAGLES captured all volunteered, observed, and solicited AEs, regardless of whether the study site physician assessed them to be causally related to study medications. AE reporting was conducted at up to 15 face-to-face clinic visits and 11 telephone visits during the 24-week trial. This post hoc analysis reports treatment-emergent AEs, which included all AEs reported from the date of first dosing of study drug until the last date of dosing of study drug plus 30 days. The EAGLES population evaluated in this analysis consisted of as-treated participants, that is, those who received at least 1 partial dose of study drug (N=8058).

       Statistical Analyses

      A summary of all AEs in the EAGLES as-treated population as well as of AEs stratified by nonpsychiatric/psychiatric cohort was performed. All analyses were performed using SAS version 9.4 (SAS Institute). Incidence tables were prepared for all reported AEs and for the most frequently reported AEs, defined as those reported by 5% or more of participants in any treatment group, sorted by Medical Dictionary for Regulatory Activities version 21.0 preferred terms. Counts were tabulated of participants reporting the first occurrence of each frequently reported AE at defined levels of intensity: mild, no interference with daily functioning; moderate, some interference with daily functioning; or severe, significant interference with daily functioning. The timing of all AEs on a given day was not recorded; therefore, if a participant had experienced the same AE (ie, as defined by the preferred term) multiple times with the same onset dates, the occurrence with the worst intensity was counted as the first occurrence. If all had the same intensity, the occurrence with the latest end date was considered to be the first occurrence.
      To focus on AEs with greatest clinical relevance, additional statistical testing was performed on each frequently reported AE to determine whether an event rate in any active treatment group differed significantly from placebo. Pairwise comparisons of active treatment vs placebo were constructed. By use of a χ2 test in conjunction with Bonferroni adjustment, an AE would be selected for further analyses if 1 or more adjusted P values were less than .05.
      For each frequently reported AE, estimated risk differences were calculated to compare incidence in varenicline or bupropion vs NRT treatment groups. Statistical comparison of the active treatments (varenicline and bupropion) vs the EAGLES active control (NRT patch), rather than vs placebo, was considered to provide greatest clinical relevance because these approved pharmacotherapies would typically be compared in general practice. Risk differences were calculated on the basis of a general linear model with terms for treatment, psychiatric/nonpsychiatric cohort, region (United States or non–United States), and treatment by psychiatric/nonpsychiatric cohort interaction. For any impact of the psychiatric cohort on the risk differences of the combined population to be taken into consideration, separate cohort analyses were also conducted for comparison.
      Analyses of time to onset and duration of first occurrence of frequently reported AEs were presented using schematic box and whisker plots and included all as-treated patients.

       Ethics

      EAGLES was reviewed and approved by each site’s Institutional Review Board or ethics committee and was conducted in accordance with the Declaration of Helsinki and in compliance with all International Council for Harmonisation Good Clinical Practice guidelines. All participants provided written, informed consent.

      Results

       Study Population

      The EAGLES study randomized 8144 smokers between November 30, 2011, and January 13, 2015, of whom 8058 participants received a study drug as follows: varenicline, n=2016; bupropion, n=2006; NRT, n=2022; and placebo, n=2014 (Figure). Baseline characteristics of the EAGLES population were consistent when stratified by treatment group, including previous use of study treatments (Table 1). Detailed baseline characteristics of the participants for the psychiatric (n=4074) and nonpsychiatric (n=3984) cohorts have been published previously.
      • Anthenelli R.M.
      • Benowitz N.L.
      • West R.
      • et al.
      Neuropsychiatric safety and efficacy of varenicline, bupropion, and nicotine patch in smokers with and without psychiatric disorders (EAGLES): a double-blind, randomised, placebo-controlled clinical trial.
      Figure thumbnail gr1
      FigureEAGLES Consolidated Standards of Reporting Trials diagram. NRT, nicotine replacement therapy.
      Table 1Baseline Characteristics: As-Treated Participants
      C-SSRS, Columbia-Suicide Severity Rating Scale; FTCD, Fagerström Test for Cigarette Dependence; HADS, Hospital Anxiety and Depression Scale; NRT, nicotine replacement therapy.
      ,
      Categorical variables are presented as number (%). Continuous variables are presented as mean (SD).
      Total (N=8058)Varenicline (n=2016)Bupropion (n=2006)NRT (n=2022)Placebo (n=2014)
      Demographic characteristics
       Male
      The sex for 6 participants randomized to treatment was inaccurately recorded.
      sex
      3549 (44.0)902 (44.7)890 (44.4)881 (43.6)876 (43.5)
       Age (y)46.5 (12.34)46.5 (12.38)46.3 (12.62)46.9 (12.19)46.4 (12.14)
       Race
      White6584 (81.7)1668 (82.7)1636 (81.6)1641 (81.2)1639 (81.4)
      Black1145 (14.2)280 (13.9)281 (14.0)303 (15.0)281 (14.0)
      Asian95 (1.2)19 (0.9)26 (1.3)24 (1.2)26 (1.3)
      Other233 (2.9)49 (2.4)63 (3.1)54 (2.7)67 (3.3)
      Unspecified1 (0.0)0 (0.0)0 (0.0)0 (0.0)1 (0.0)
       Region
      United States4207 (52.2)1054 (52.3)1052 (52.4)1051 (52.0)1050 (52.1)
      Non–United States3851 (47.8)962 (47.7)954 (47.6)971 (48.0)964 (47.9)
      Smoking characteristics
       FTCD score5.8 (1.99)5.8 (1.98)5.8 (1.99)5.8 (1.96)5.7 (2.02)
       Duration of smoking (y)28.3 (12.37)28.3 (12.32)28.2 (12.67)28.6 (12.38)28.2 (12.11)
       Cigarettes smoked per day in past month20.7 (8.22)20.7 (8.16)20.6 (8.01)20.8 (8.64)20.6 (8.05)
       ≥1 lifetime serious quit attempts
      Quit attempt ≥24 hours.
      6647 (82.5)1664 (82.5)1651 (82.3)1683 (83.2)1649 (81.9)
      Prior use of study treatments
       Varenicline1236 (15.3)281 (13.9)338 (16.8)320 (15.8)297 (14.7)
       Bupropion785 (9.7)194 (9.6)205 (10.2)195 (9.6)191 (9.5)
       NRT2602 (32.3)645 (32.0)633 (31.6)681 (33.7)643 (31.9)
      Psychiatric history
       Cohort
      Psychiatric4074 (50.6)1026 (50.9)1017 (50.7)1016 (50.2)1015 (50.4)
      Nonpsychiatric3984 (49.4)990 (49.1)989 (49.3)1006 (49.8)999 (49.6)
       C-SSRS
      Positive C-SSRS response for suicidal behavior or ideation.
      lifetime
      1607 (19.9)402 (19.9)407 (20.3)391 (19.3)407 (20.2)
       HADS score6.4 (5.90)6.3 (5.89)6.4 (6.16)6.3 (5.87)6.4 (5.68)
      a C-SSRS, Columbia-Suicide Severity Rating Scale; FTCD, Fagerström Test for Cigarette Dependence; HADS, Hospital Anxiety and Depression Scale; NRT, nicotine replacement therapy.
      b Categorical variables are presented as number (%). Continuous variables are presented as mean (SD).
      c The sex for 6 participants randomized to treatment was inaccurately recorded.
      d Quit attempt ≥24 hours.
      e Positive C-SSRS response for suicidal behavior or ideation.

       Frequently Reported AEs

       Incidence

      A summary of all and frequently reported AEs and treatment discontinuations is presented in Table 2. Briefly, for all AEs, 5729 of 8058 (71.1%) participants reported 1 or more AE of any intensity, whereas the incidence of participants reporting 1 or more AE rated severe in intensity was lower (406/8058 [5.0%]).
      Table 2Summary of AEs and Treatment Discontinuations
      AE, adverse event; NRT, nicotine replacement therapy.
      ,
      Values are reported as number (%).
      Total (N=8058)Varenicline (n=2016)Bupropion (n=2006)NRT (n=2022)Placebo (n=2014)
      AE
      AEs reported during treatment and ≤30 days after last dose.
      (all reported)
       No. of participants
      Counts a participant once only.
      with AEs
      5729 (71.1)1503 (74.6)1446 (72.1)1435 (71.0)1345 (66.8)
       No. of participants with AEs of severe
      AE interferes significantly with participant’s usual function.
      intensity
      406 (5.0)106 (5.3)99 (4.9)112 (5.5)89 (4.4)
       No. of participants discontinued because of AEs626 (7.8)166 (8.2)176 (8.8)162 (8.0)122 (6.1)
      AE
      AEs reported during treatment and ≤30 days after last dose.
      (frequently reported
      AE reported by ≥5% of participants in any treatment group.
      )
       No. of participants with AEs3805 (47.2)1094 (54.3)951 (47.4)949 (46.9)811 (40.3)
       No. of participants
      Counts a participant once only.
      with AEs of severe
      AE interferes significantly with participant’s usual function.
      intensity
      84 (1.0)27 (1.3)15 (0.7)25 (1.2)17 (0.8)
       No. of participants discontinued because of AEs169 (2.1)56 (2.8)34 (1.7)53 (2.6)26 (1.3)
      a AE, adverse event; NRT, nicotine replacement therapy.
      b Values are reported as number (%).
      c AEs reported during treatment and ≤30 days after last dose.
      d Counts a participant once only.
      e AE interferes significantly with participant’s usual function.
      f AE reported by ≥5% of participants in any treatment group.
      Eleven frequently reported AEs were identified (nausea, headache, insomnia, abnormal dreams, nasopharyngitis, anxiety, upper respiratory tract infection, irritability, dry mouth, fatigue, and application site pruritus). Pairwise comparisons of active treatment vs placebo were constructed. By use of a χ2 test in conjunction with Bonferroni adjustment, nasopharyngitis, upper respiratory tract infection, and headache were eliminated from further analyses because of less noteworthy difference in incidence between treatment groups.
      At least 1 frequently reported AE was experienced by 3805 of 8058 (47.2%) participants (Table 2). However, the number of participants reporting 1 or more frequently reported AE of severe intensity was low (84/8058 [1.0%]), and the overall incidence was similar across the active treatment groups (range, 0.7%-1.3%). Less than 3% of participants discontinued study treatment because of treatment-emergent, frequently reported AEs.
      The proportion of participants and the intensity of the first occurrence of the remaining 8 frequently reported AEs, stratified by treatment, are presented in Table 3. The most frequently reported AEs were nausea in the varenicline group, insomnia with bupropion, and abnormal dreams with NRT. For each AE, most first occurrences were reported as mild, and AEs of severe intensity were uncommon (<5% of events in any treatment group, for any AE).
      Table 3Counts of Participants in Each Level of Intensity for First Occurrence of Most Frequently Reported AEs
      AE, adverse event; MedDRA, Medical Dictionary for Regulatory Activities; NRT, nicotine replacement therapy.
      ,
      Values are reported as number (%).
      ,
      AE reported by ≥5% of participants in any treatment group.
      ,
      Raw AE incidence.
      Intensity of AE (MedDRA version 21.0 preferred terms)Total (N=8058)Varenicline (n=2016)Bupropion (n=2006)NRT (n=2022)Placebo (n=2014)
      Nausea1048 (13.0)511 (25.3)201 (10.0)199 (9.8)137 (6.8)
       Severe13 (0.2)8 (0.4)3 (0.1)2 (0.1)0 (0.0)
       Moderate225 (2.8)112 (5.6)40 (2.0)53 (2.6)20 (1.0)
       Mild810 (10.1)391 (19.4)158 (7.9)144 (7.1)117 (5.8)
      Insomnia768 (9.5)189 (9.4)245 (12.2)195 (9.6)139 (6.9)
       Severe9 (0.1)1 (<0.1)2 (0.1)3 (0.1)3 (0.1)
       Moderate221 (2.7)42 (2.1)72 (3.6)57 (2.8)50 (2.5)
       Mild538 (6.7)146 (7.2)171 (8.5)135 (6.7)86 (4.3)
      Abnormal dreams662 (8.2)198 (9.8)129 (6.4)245 (12.1)90 (4.5)
       Severe1 (<0.1)1 (<0.1)0 (0.0)0 (0.0)0 (0.0)
       Moderate81 (1.0)20 (1.0)13 (0.6)31 (1.5)17 (0.8)
       Mild580 (7.2)177 (8.8)116 (5.8)214 (10.6)73 (3.6)
      Anxiety559 (6.9)132 (6.5)169 (8.4)138 (6.8)120 (6.0)
       Severe16 (0.2)4 (0.2)4 (0.2)4 (0.2)4 (0.2)
       Moderate171 (2.1)43 (2.1)46 (2.3)44 (2.2)38 (1.9)
       Mild372 (4.6)85 (4.2)119 (5.9)90 (4.5)78 (3.9)
      Irritability365 (4.5)82 (4.1)71 (3.5)108 (5.3)104 (5.2)
       Severe1 (<0.1)0 (0.0)0 (0.0)0 (0.0)1 (<0.1)
       Moderate66 (0.8)22 (1.1)13 (0.6)11 (0.5)20 (1.0)
       Mild298 (3.7)60 (3.0)58 (2.9)97 (4.8)83 (4.1)
      Dry mouth335 (4.2)66 (3.3)146 (7.3)59 (2.9)64 (3.2)
       Severe4 (<0.1)0 (0.0)3 (0.1)0 (0.0)1 (<0.01)
       Moderate44 (0.5)3 (0.1)31 (1.5)6 (0.3)4 (0.2)
       Mild287 (3.6)63 (3.1)112 (5.6)53 (2.6)59 (2.9)
      Fatigue339 (4.2)124 (6.2)57 (2.8)75 (3.7)83 (4.1)
       Severe2 (<0.1)1 (<0.1)0 (0.0)0 (0.0)1 (<0.1)
       Moderate93 (1.2)37 (1.8)13 (0.6)21 (1.0)22 (1.1)
       Mild244 (3.0)86 (4.3)44 (2.2)54 (2.7)60 (3.0)
      Application site pruritus159 (2.0)22 (1.1)12 (0.6)109 (5.4)16 (0.8)
       Severe5 (0.1)0 (0.0)0 (0.0)5 (0.2)0 (0.0)
       Moderate14 (0.2)2 (0.1)1 (<0.1)10 (0.5)1 (<0.1)
       Mild140 (1.7)20 (1.0)11 (0.5)94 (4.6)15 (0.7)
      a AE, adverse event; MedDRA, Medical Dictionary for Regulatory Activities; NRT, nicotine replacement therapy.
      b Values are reported as number (%).
      c AE reported by ≥5% of participants in any treatment group.
      d Raw AE incidence.

       Varenicline and Bupropion vs NRT

      Estimation of risk difference of frequently reported AEs for varenicline or bupropion vs NRT is presented in Table 4. Participants who received varenicline reported a significantly higher incidence of nausea compared with those who received NRT (25.3% vs 9.8%; P<.001; Tables 3 and 4). Of 511 first occurrences in the varenicline-treated group, 8 (1.6%) were rated severe (Table 3). Incidence of nausea in the bupropion-treated group was not significantly different from that in the NRT group (Table 4). Participants who received varenicline also reported a higher incidence of fatigue compared with those who received NRT (6.2% vs 3.7%; P<.001; Tables 3 and 4).
      Table 4Estimation of Risk Difference of Frequently Reported AEs: Varenicline or Bupropion vs NRT (As-Treated Participants)
      AE, adverse event; NRT, nicotine replacement therapy.
      ,
      AE reported by ≥5% of participants in any treatment group.
      AE and treatmentEstimated AE incidence
      Incidence after adjusting other factors in the model.
      (95% CI); P value
      Treatment comparisonEstimated risk difference
      Estimated risk difference was based on a general linear model with terms for treatment, cohort, region (United States and non–United States), and treatment by cohort interaction.
      (95% CI); P value
      Nausea
       Varenicline25.3% (23.42%-27.22%); <.001Varenicline vs NRT15.50% (13.20%-17.80%); <.001
       Bupropion10.0% (8.68%-11.31%); <.001Bupropion vs NRT0.17% (–1.67% to 2.02%); .85
       NRT9.8% (8.52%-11.12%); <.001
      Insomnia
       Varenicline9.4% (8.10%-10.65%); <.001Varenicline vs NRT–0.26% (–2.08% to 1.55%); .78
       Bupropion12.2% (10.79%-13.65%); <.001Bupropion vs NRT2.58% (0.65%-4.51%); .009
       NRT9.6% (8.35%-10.93%); <.001
      Abnormal dreams
       Varenicline9.5% (8.27%-10.73%); <.001Varenicline vs NRT–2.49% (–4.35% to –0.64%); .008
       Bupropion6.4% (5.44%-7.35%); <.001Bupropion vs NRT–5.60% (–7.27% to –3.93%); <.001
       NRT12.0% (10.60%-13.39%); <.001
      Anxiety
       Varenicline6.4% (5.32%-7.44%); <.001Varenicline vs NRT–0.35% (–1.86% to 1.17%); .65
       Bupropion8.4% (7.18%-9.58%); <.001Bupropion vs NRT1.65% (0.03%-3.27%); .05
       NRT6.7% (5.65%-7.81%); <.001
      Irritability
       Varenicline4.1% (3.24%-4.96%); <.001Varenicline vs NRT–1.24% (–2.53% to 0.06%); .06
       Bupropion3.5% (2.74%-4.35%); <.001Bupropion vs NRT–1.79% (–3.05% to –0.53%); .005
       NRT5.3% (4.36%-6.31%); <.001
      Dry mouth
       Varenicline3.3% (2.49%-4.05%); <.001Varenicline vs NRT0.36% (–0.71% to 1.44%); .51
       Bupropion7.3% (6.14%-8.41%); <.001Bupropion vs NRT4.36% (3.01%-5.72%); <.001
       NRT2.9% (2.17%-3.65%); <.001
      Fatigue
       Varenicline6.1% (5.06%-7.14%); <.001Varenicline vs NRT2.39% (1.07%-3.72%); <.001
       Bupropion2.9% (2.16%-3.63%); <.001Bupropion vs NRT–0.82% (–1.91% to 0.28%); .14
       NRT3.7% (2.89%-4.53%); <.001
      Application site pruritus
       Varenicline1.1% (0.64%-1.55%); <.001Varenicline vs NRT–4.30% (–5.38% to –3.22%); <.001
       Bupropion0.6% (0.25%-0.91%); <.001Bupropion vs NRT–4.82% (–5.85% to –3.78%); <.001
       NRT5.4% (4.41%-6.38%); <.001
      a AE, adverse event; NRT, nicotine replacement therapy.
      b AE reported by ≥5% of participants in any treatment group.
      c Incidence after adjusting other factors in the model.
      d Estimated risk difference was based on a general linear model with terms for treatment, cohort, region (United States and non–United States), and treatment by cohort interaction.
      Participants who received bupropion experienced a significantly higher incidence of insomnia compared with those who received NRT (12.2% vs 9.6%; P=.009); no significant difference in incidence of insomnia was observed between the varenicline and NRT groups (Tables 3 and 4). The overall incidence of dry mouth was reported most frequently in participants who received bupropion, and incidence was significantly higher than for those who received NRT (7.3% vs 2.9%; P<.001; Tables 3 and 4).
      The incidence of abnormal dreams was significantly lower with varenicline compared with NRT (9.8% vs 12.1%; P=.008) or bupropion compared with NRT (6.4% vs 12.1%; P<.001; Tables 3 and 4). The incidence of application site pruritus was significantly higher in participants who received active NRT patch than in those who received placebo patch with varenicline (5.4% vs 1.1%; P<.001) or bupropion (5.4% vs 0.6%; P<.001; Tables 3 and 4).
      The neuropsychiatric AEs anxiety and irritability were among those frequently reported. Overall, incidence of anxiety between varenicline-treated and NRT-treated participants did not differ significantly, but incidence was higher in the bupropion-treated vs the NRT-treated group (8.4% vs 6.8%; P=.05; Tables 3 and 4). When analyses were stratified by psychiatric/nonpsychiatric cohort, the overall incidence of anxiety was higher in the psychiatric compared with the nonpsychiatric cohort (8.5% vs 5.3%).
      • Anthenelli R.M.
      • Benowitz N.L.
      • West R.
      • et al.
      Neuropsychiatric safety and efficacy of varenicline, bupropion, and nicotine patch in smokers with and without psychiatric disorders (EAGLES): a double-blind, randomised, placebo-controlled clinical trial.
      There was no significant difference in incidence of anxiety for varenicline vs NRT in either cohort (Supplemental Table 1, available online at http://www.mayoclinicproceedings.org). In the bupropion-treated group, anxiety was significantly higher vs the NRT group in the nonpsychiatric but not in the psychiatric cohort (Supplemental Table 1).
      The incidence of irritability did not differ between the varenicline and NRT groups overall or stratified by nonpsychiatric/psychiatric cohort (Table 4 and Supplemental Table 1). Overall, participants who received NRT reported a higher incidence of irritability than those who received bupropion; similar results were observed in the nonpsychiatric cohort (Table 4 and Supplemental Table 1).

       Time to Onset and Duration of First Occurrence

      The time to onset and duration of first occurrence of the 8 frequently reported AEs are shown in Supplemental Figures 1 and 2 and Supplemental Table 2 (available online at http://www.mayoclinicproceedings.org). Whereas median time to onset of first occurrence of nausea was similar across active treatment groups and placebo (range, 8-10 days), varenicline was associated with the longest median duration of nausea (13 days) compared with bupropion, NRT, and placebo (6 or 7 days). Participants reported abnormal dreams approximately 9 days after starting any active treatment or placebo, which continued for the longest duration in those who received varenicline (median, 41 days). The median time to onset of first occurrence of insomnia was shortest in bupropion-treated participants (6 days) compared with 12 days for varenicline and NRT and 17 days for participants who received placebo; however, duration of first occurrence of insomnia was similar across all treatments and placebo. Participants who reported first occurrence of application site pruritus with NRT reported a median duration of 40 days.
      The median time to onset of first occurrence of anxiety was shortest for bupropion-treated participants (12 days), 25 days for varenicline and NRT, and 23 days for placebo; however, median duration of first occurrence of anxiety was similar across all active treatments and placebo (range, 11-15 days). Likewise, time to onset and duration of first occurrence of irritability were similar across all active treatments and placebo.

      Discussion

      The results of this post hoc analysis of EAGLES safety data indicate that most frequently reported AEs were of mild intensity, and the overall incidence of AEs of severe intensity was less than 1.5% across all active treatments and placebo. AEs associated with FDA-approved smoking cessation pharmacotherapies have been documented. The results of the current analyses with regard to the most frequently reported AEs for each pharmacotherapy and their perceived intensities are consistent with published findings for varenicline, NRT, and bupropion.
      • Cahill K.
      • Lindson-Hawley N.
      • Thomas K.H.
      • Fanshawe T.R.
      • Lancaster T.
      Nicotine receptor partial agonists for smoking cessation.
      ,
      • Mills E.J.
      • Wu P.
      • Lockhart I.
      • Wilson K.
      • Ebbert J.O.
      Adverse events associated with nicotine replacement therapy (NRT) for smoking cessation. A systematic review and meta-analysis of one hundred and twenty studies involving 177,390 individuals.
      ,
      • Hughes J.R.
      • Stead L.F.
      • Hartmann-Boyce J.
      • Cahill K.
      • Lancaster T.
      Antidepressants for smoking cessation.
      Of clinical importance is that the incidence of first occurrence of most frequently reported AEs of severe intensity was comparable across treatment groups. Overall, the median time to onset for the most frequently reported AEs was 5 days or more after starting therapy, whereas median duration was 41 days or less.
      We observed that nausea was the most frequently reported AE in participants who received varenicline, with median onset of about 1 to 2 weeks after the start of treatment and median duration of about 2 weeks. Insomnia was the most frequently reported AE in participants who received bupropion, with median onset about 6 days after start of treatment; median duration was approximately 2 to 3 weeks. Participants who received NRT most frequently reported abnormal dreams, with median onset 1 to 2 weeks after start of treatment and median duration of about 4 weeks.
      Quitting smoking without the aid of smoking cessation pharmacotherapies is associated with nicotine withdrawal symptoms, onset of which commonly occurs within 4 to 24 hours after nicotine withdrawal, with greatest intensity during the first week.
      • McLaughlin I.
      • Dani J.A.
      • De Biasi M.
      Nicotine withdrawal.
      The target quit date for all patients was day 8 in this study. The similar incidence of AEs experienced by participants who received placebo vs active treatments suggests that it remains challenging to disentangle pharmacotherapy adverse effects from tobacco withdrawal symptoms. However, pharmacotherapy has been demonstrated to reduce symptoms of nicotine withdrawal.
      • Cahill K.
      • Lindson-Hawley N.
      • Thomas K.H.
      • Fanshawe T.R.
      • Lancaster T.
      Nicotine receptor partial agonists for smoking cessation.
      ,
      • Brandon T.H.
      • Drobes D.J.
      • Unrod M.
      • et al.
      Varenicline effects on craving, cue reactivity, and smoking reward.
      Our findings provide information that will assist HCPs in discussions around the risk differences in adverse effects with different pharmacotherapies. These discussions can reduce patient barriers to treatment and promote medication adherence.
      Efforts to tackle misperceptions around the safety of smoking cessation pharmacotherapies are critical. Unassisted quit attempts remain the most common quit method but success rates are generally low, whereas extended use of pharmacotherapies is associated with a significantly higher probability of cessation.
      • Siahpush M.
      • Shaikh R.A.
      • McCarthy M.
      • Sikora Kessler A.
      • Tibbits M.
      • Singh G.K.
      Association between duration of use of pharmacotherapy and smoking cessation: findings from a national survey.
      EAGLES represents the largest randomized, double-blind, smoking cessation study to date and included smokers with a history of psychiatric disease and those with baseline cardiovascular disease risk factors. Therefore, a strength of this analysis is that the participant population is likely to provide an accurate representation of smokers seen in general medical practice, and it confirms that AEs associated with each of the FDA-approved pharmacotherapies are generally of mild intensity and tolerable.
      All 3 classes of smoking cessation pharmacotherapies result in a similar AE experience and do not appreciably differ in rates of severe AEs. Notably, of the 3 smoking cessation therapies included in EAGLES, only NRT patch is currently available as an over-the-counter therapy. Given the same limited indication for NRT and varenicline (ie, smoking cessation) and in the absence of major differences in AEs reported after treatment with varenicline or NRT, these analyses along with previously reported efficacy data suggest that varenicline may be a suitable candidate for over-the-counter designation.
      • Leischow S.J.
      Increasing smoking cessation in the United States: expanding the availability of over-the-counter medications.
      Among the most frequently reported AEs, 2 were categorized as neuropsychiatric: anxiety and irritability. This post hoc analysis demonstrated that the incidence of these neuropsychiatric AEs did not differ between varenicline and NRT overall or when results were stratified by nonpsychiatric and psychiatric cohort. As many smokers have a psychiatric diagnosis,
      • Peckham E.
      • Brabyn S.
      • Cook L.
      • Tew G.
      • Gilbody S.
      Smoking cessation in severe mental ill health: what works? An updated systematic review and meta-analysis.
      this information could be important for HCPs to consider in encouraging initiation and motivating adherence among smokers with and without a psychiatric history.
      Our study has several limitations. First, the data derive from a clinical trial setting in which participants received weekly follow-up and brief but frequent behavioral counseling. This level of support is likely to be greater than that provided in a real-world setting. Although it is reasonable to hypothesize that this may augment treatment continuation, more intense support has not consistently been observed to be associated with greater adherence as pharmacotherapy may seem less important to patients when they are provided with intensive behavioral counseling than when they are provided with limited behavioral support.
      • van Rossem C.
      • Spigt M.
      • Viechtbauer W.
      • Lucas A.E.M.
      • van Schayck O.C.P.
      • Kotz D.
      Effectiveness of intensive practice nurse counselling versus brief general practitioner advice, both combined with varenicline, for smoking cessation: a randomized pragmatic trial in primary care.
      However, the low rates of treatment discontinuations due to AEs observed in participants who received support in this trial may demonstrate the value of providing information and education about treatment expectations as well as modest levels of ongoing treatment support in the real-world setting.
      • Herbec A.
      • Tombor I.
      • Shahab L.
      • West R.
      "If I'd known ..."—a theory-informed systematic analysis of missed opportunities in optimising use of nicotine replacement therapy and accessing relevant support: a qualitative study.
      Second, the majority of participants identified as White (81.7%); therefore, results might not generalize to minority ethnic populations. Third, time to onset and duration data demonstrated first occurrence of frequently reported AEs to be transient and self-limited. Analyses included all as-treated participants and did not exclude those who discontinued treatment. Whereas this presents a limitation, less than 3% of participants discontinued their assigned study treatment because of treatment-emergent, frequently reported AEs; therefore, the impact of not excluding these data was considered to be low.
      This analysis reports on important characteristics of frequently occurring AEs in EAGLES, the largest randomized, controlled, smoking cessation pharmacotherapy trial conducted to date. Along with the previous prospective EAGLES analyses, which extensively evaluated clinically important neuropsychiatric
      • Anthenelli R.M.
      • Benowitz N.L.
      • West R.
      • et al.
      Neuropsychiatric safety and efficacy of varenicline, bupropion, and nicotine patch in smokers with and without psychiatric disorders (EAGLES): a double-blind, randomised, placebo-controlled clinical trial.
      and cardiovascular
      • Benowitz N.L.
      • Pipe A.
      • West R.
      • et al.
      Cardiovascular safety of varenicline, bupropion, and nicotine patch in smokers: a randomized clinical trial.
      AEs, these data provide substantial information about the general safety and tolerability of these medications.

      Conclusion

      Our findings demonstrate the tolerability of varenicline, bupropion, and NRT patch in a large population of smokers. Consistent with previous recommendations,
      Clinical Practice Guideline Treating Tobacco Use and Dependence 2008 Update Panel, Liaisons, and Staff
      A clinical practice guideline for treating tobacco use and dependence: 2008 update. A U.S. Public Health Service report.
      HCPs can confidently recommend approved smoking cessation pharmacotherapies to smokers with and without a psychiatric history. Our findings provide HCPs with information about which AEs to anticipate, how soon they might occur, and how long they could last if patients persist with pharmacotherapy. Such information can potentially improve treatment adherence and smoking abstinence outcomes.

      Acknowledgments

      Medical writing support was provided by Katy Beck, PhD, of Engage Scientific Solutions (Horsham, UK) and funded by Pfizer .

      Supplemental Online Material

      References

        • World Health Organization
        Tobacco.
        (Published 2018. Updated February 4, 2019)
        • Cahill K.
        • Lindson-Hawley N.
        • Thomas K.H.
        • Fanshawe T.R.
        • Lancaster T.
        Nicotine receptor partial agonists for smoking cessation.
        Cochrane Database Syst Rev. 2016; 2016: CD006103
        • Caraballo R.S.
        • Shafer P.R.
        • Patel D.
        • Davis K.C.
        • McAfee T.A.
        Quit methods used by US adult cigarette smokers, 2014-2016.
        Prev Chronic Dis. 2017; 14: E32
        • US Food & Drug Administration
        Want to quit smoking? FDA-approved products can help.
        (Published 2017. Updated February 4, 2019)
        • Jamal A.
        • Dube S.R.
        • Malarcher A.M.
        • Shaw L.
        • Engstrom M.C.
        • Centers for Disease Control and Prevention (CDC)
        Tobacco use screening and counseling during physician office visits among adults—National Ambulatory Medical Care Survey and National Health Interview Survey, United States, 2005-2009.
        MMWR Suppl. 2012; 61: 38-45
        • Kaplan R.M.
        • Fang Z.
        • Morgan G.
        Providers' advice concerning smoking cessation: evidence from the Medical Expenditures Panel Survey.
        Prev Med. 2016; 91: 32-36
        • Pacek L.R.
        • McClernon F.J.
        • Bosworth H.B.
        Adherence to pharmacological smoking cessation interventions: a literature review and synthesis of correlates and barriers.
        Nicotine Tob Res. 2018; 20: 1163-1172
        • Hays J.T.
        • Leischow S.J.
        • Lawrence D.
        • Lee T.C.
        Adherence to treatment for tobacco dependence: association with smoking abstinence and predictors of adherence.
        Nicotine Tob Res. 2010; 12: 574-581
        • Vogt F.
        • Hall S.
        • Marteau T.M.
        Understanding why smokers do not want to use nicotine dependence medications to stop smoking: qualitative and quantitative studies.
        Nicotine Tob Res. 2008; 10: 1405-1413
        • Mills E.J.
        • Wu P.
        • Lockhart I.
        • Wilson K.
        • Ebbert J.O.
        Adverse events associated with nicotine replacement therapy (NRT) for smoking cessation. A systematic review and meta-analysis of one hundred and twenty studies involving 177,390 individuals.
        Tob Induc Dis. 2010; 8: 8
        • Smith A.L.
        • Carter S.M.
        • Chapman S.
        • Dunlop S.M.
        • Freeman B.
        Why do smokers try to quit without medication or counselling? A qualitative study with ex-smokers.
        BMJ Open. 2015; 5: e007301
        • Buggy Y.
        • Cornelius V.
        • Fogg C.
        • Kasliwal R.
        • Layton D.
        • Shakir S.A.
        Neuropsychiatric events with varenicline: a modified prescription-event monitoring study in general practice in England.
        Drug Saf. 2013; 36: 521-531
        • Pasternak B.
        • Svanstrom H.
        • Hviid A.
        Use of varenicline versus bupropion and risk of psychiatric adverse events.
        Addiction. 2013; 108: 1336-1343
        • US Food & Drug Administration
        FDA Drug Safety Communication: Chantix (varenicline) may increase the risk of certain cardiovascular adverse events in patients with cardiovascular disease.
        (Published 2011. Updated February 5, 2019)
        • Sobieraj D.M.
        • White W.B.
        • Baker W.L.
        Cardiovascular effects of pharmacologic therapies for smoking cessation.
        J Am Soc Hypertens. 2013; 7: 61-67
        • Hughes J.R.
        • Stead L.F.
        • Hartmann-Boyce J.
        • Cahill K.
        • Lancaster T.
        Antidepressants for smoking cessation.
        Cochrane Database Syst Rev. 2014; 2014: CD000031
        • Anthenelli R.M.
        • Benowitz N.L.
        • West R.
        • et al.
        Neuropsychiatric safety and efficacy of varenicline, bupropion, and nicotine patch in smokers with and without psychiatric disorders (EAGLES): a double-blind, randomised, placebo-controlled clinical trial.
        Lancet. 2016; 387: 2507-2520
        • Benowitz N.L.
        • Pipe A.
        • West R.
        • et al.
        Cardiovascular safety of varenicline, bupropion, and nicotine patch in smokers: a randomized clinical trial.
        JAMA Intern Med. 2018; 178: 622-631
        • McLaughlin I.
        • Dani J.A.
        • De Biasi M.
        Nicotine withdrawal.
        Curr Top Behav Neurosci. 2015; 24: 99-123
        • Brandon T.H.
        • Drobes D.J.
        • Unrod M.
        • et al.
        Varenicline effects on craving, cue reactivity, and smoking reward.
        Psychopharmacology (Berl). 2011; 218: 391-403
        • Siahpush M.
        • Shaikh R.A.
        • McCarthy M.
        • Sikora Kessler A.
        • Tibbits M.
        • Singh G.K.
        Association between duration of use of pharmacotherapy and smoking cessation: findings from a national survey.
        BMJ Open. 2015; 5: e006229
        • Leischow S.J.
        Increasing smoking cessation in the United States: expanding the availability of over-the-counter medications.
        JAMA. 2019; 321: 541-542
        • Peckham E.
        • Brabyn S.
        • Cook L.
        • Tew G.
        • Gilbody S.
        Smoking cessation in severe mental ill health: what works? An updated systematic review and meta-analysis.
        BMC Psychiatry. 2017; 17: 252
        • van Rossem C.
        • Spigt M.
        • Viechtbauer W.
        • Lucas A.E.M.
        • van Schayck O.C.P.
        • Kotz D.
        Effectiveness of intensive practice nurse counselling versus brief general practitioner advice, both combined with varenicline, for smoking cessation: a randomized pragmatic trial in primary care.
        Addiction. 2017; 112: 2237-2247
        • Herbec A.
        • Tombor I.
        • Shahab L.
        • West R.
        "If I'd known ..."—a theory-informed systematic analysis of missed opportunities in optimising use of nicotine replacement therapy and accessing relevant support: a qualitative study.
        Int J Behav Med. 2018; 25: 579-591
        • Clinical Practice Guideline Treating Tobacco Use and Dependence 2008 Update Panel, Liaisons, and Staff
        A clinical practice guideline for treating tobacco use and dependence: 2008 update. A U.S. Public Health Service report.
        Am J Prev Med. 2008; 35: 158-176