EAGLES - Efficacy & neuropsychiatric safety of smoking cessation products (varenicline, bupropion, nicotine patch)

Anthenelli RM, 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;387:2507-20.

Clinical Question

In people, with or without a psychiatric disorder, who smoke, is there a difference in neuropsychiatric events or smoking abstinence rates between varenicline, bupropion, nicotine patch or placebo?


Bottom Line

In people who smoke who either have no psychiatric disorder or clinically stable psychiatric disorder, bupropion, nicotine patch and varenicline do not increase the risk of serious or overall neuropsychiatric events. All 3 of these drugs effectively increase the likelihood of smoking abstinence at 6 months compared to placebo, with patients receiving varenicline having the greatest likelihood of abstaining.



Allocation-concealed RCT with all (patients, clinicians, investigators) blinded, loss-to-follow-up of ~6%, analyzed using the intention-to-treat population.


  • Randomization stratified based presence/absence of psychiatric disorder & disorder type (mood, anxiety, psychotic, personality), as well as geography
  • 21% and 26% of non-psychiatric and psychiatric groups, respectively, dropped out
    • Overall similar dropout rates between treatment groups


Patients and Setting

  • 16 countries, 140 centers
  • November 30 2011 to January 13 2015
  • Inclusion criteria:
    1. Men or women
    2. 18-75 y
    3. Smoking average 10+ cigarettes/day during previous year
    4. Exhaled carbon monoxide concentration >10 ppm at screening
    5. Motivated to stop smoking
    6. +/- current or past psychiatric disorder, according to DSM-IV-TR definition
      • Stable, defined as
        • No exacerbation in previous 6 months, on stable treatment (drug & dose) x3+ months, no treatment changes anticipated during study, plus
        • Considered not to be at high risk of self-injury or suicidal behavior, based on responses on the Suicide Behaviors Questionnaire-Revised or Columbia-Suicide Severity Rating Scale
      • Mood disorder (including major depressive disorder, bipolar disoder)
      • Anxiety disorder (panic disorder +/- agoraphobia, PTSD, OCD, social phobia, GAD)
      • Psychotic disorder (schizophrenia, schizoaffective disorders)
      • Borderline personality disorder
  • Key exclusion criteria:
    • Suicidal behavior or suicidal ideation associated with intent/plan in the past year
    • Seizure disorder or risk of seizure (bupropion contraindication)
    • Substance abuse within previous 12 months (including ETOH & other substances)
    • Delirium, demention & other cognitive disorders
    • Substance-induced, factitious, dissociative & impulse control disorders
  • 11,186 screened -> 8,144 randomized (73%)
  • Average patient:
    • 46.5 y/o
    • 56% female
    • Race: White ~80%, Black ~15%
    • Region: USA 52%, Western Europe & others (including Canada) 30%, Eastern Europe 10%, South/Middle America 8%
    • Wt 80 kg
    • Smoking hx
      • Duration of smoking ~28 y
      • ~21 cigs/d (1 ppd)
      • 3-4 previous quit attempts
    • Psychiatric hx: Yes (50%)
      • Mood 70%, anxiety 20%. psychotic 10%, personality <1%
      • Those with psychiatric disorder: Hx suicidal ideation 34%, suicidal behavior 13%
    • Psychotropic medications in subgroup with psychiatric disorder: Any 50%
      • Antidepressant ~1/3, anxiolytic ~15%, antipsychotic ~15%, mood stabilizer 2%


Interventions and Control

  • Intervention 1: Bupropion x12 weeks
    • 1 week before quit date: Start 150 mg PO daily
    • 4 days before quit date: Increase to 150 mg PO BID (target dose) x11.5 weeks
  • Intervention 2: Nicotine patch x11 weeks
    • On quit date: Start 21 mg patch x7 weeks, then decrease to 14 mg patch x 2 weeks, then 7 mg patch x2 weeks, then stop
    • Note: Not stated whether patch removed at bedtime, but likely that it was kept on 24h/day given reported higher risk of abnormal dreams versus placebo (see outcomes below)
  • Intervention 3: Varenicline x12 weeks
    • 1 week before quit date: Start 0.5 mg PO daily
    • 3 days before quit date: Increase to 0.5 mg PO BID
    • Quit date: Increase to 1 mg PO BID (target dose) x11 weeks
  • Control: Placebo
  • Co-interventions common to all groups:
    • 12-week treatment phase, then 12-week follow-up phase
      • Up to 15 in-person visits + 11 telephone visits
        • Patch + pill counts at each visit to measure adherence (overall 80%)
        • Tobacco/nicotine product use evaluated at each in-person/telephone visit via questionnaire
        • Expired air carbon monoxide measurement performed at every in-person visit
    • Received identical placebo for other interventions (e.g. varenicline group received inert pill that looked like bupropion & inert patch that looked like a nicotine patch)



EAGLES trial outcomes in patients WITH psychiatric disorder

EAGLES trial outcomes in patients withOUT psychiatric disorder

NNT for abstinence from smoking in EAGLES trial

  • Efficacy:
    • All 3 drugs were more effective than placebo at achieving continuous abstinence from weeks 9 to 24 in both patients with and without psychiatric disorder
      • Varenicline was more effective than bupropion or nicotine patch in both patients with and without psychiatric disorder
    • Abstinence rates were slightly higher in patients without psychiatric disorder, but drugs had similar efficacy regardless of presence of psychiatric disorder
  • Safety: 
    • No statistically significant increase in overall or serious neuropsychiatric events with any smoking cessation agent versus placebo in both patients with and without previous psychiatric disorder
      • The 1 completed suicide occurred in the non-psychiatric placebo group
    • Low rates (<1%) of serious adverse events in all groups
    • Predictable drug-specific adverse events consistent with previous studies
      • Bupropion: Insomnia, dry mouth
      • Nicotine patch: Abnormal dreams and insomnia (likely from keeping patch on at night), patch-site irritation and itching
      • Varenicline: Abnormal dreams and insomnia, nausea (the latter occurs in 1/4 patients who take varenicline)



Patients: Represent those without psychiatric disorder and those with current very stable mood, anxiety or psychotic disorder

  • The results do not apply to those patients with recent suicide attempt, at otherwise high risk of suicide or patients whose psychotropic medications are undergoing tinkering
  • Neuropsychiatric safety of various smoking cessation products has not been demonstrated, and perhaps smoking cessation attempts should be deferred for these patients until clinical stability has been achieved, and patient conviction to quit is high.

Interventions: All 1st-line agents for smoking cessation with regimens generally reflecting clinical practice, with exception of the nicotine replacement therapy (NRT)

  • Though it's unclear if this was done in the study, the nicotine patch should be removed at bedtime to avoid abnormal dreams and insomnia
    • Unless the patient has a significant history of nocturnal nicotine cravings 
  • Patients were instructed to universally wean their nicotine patch dose after only 7 weeks, which may led to breakthrough withdrawal and nicotine cravings, negatively impacting these patients' quit attempt
  • Ideally, patients would combine the nicotine patch with short-acting nicotine products (e.g. gum, lozenges), as this has previously demonstrated greater abstinence rates
  • Previous RCTs have demonstrated greater abstinence rates with longer user of smoking cessation products. Patients should thus be encouraged to continue their smoking cessation product beyond 12 weeks if there is risk of relapse with discontinuing these at the 12-week mark.

Outcomes: Abstinence only measured until 6 months after start of smoking cessation. Some patients will restart smoking after 6 months & will need further quit attempts to successfully quit.

Summary author: Ricky Turgeon BSc(Pharm), ACPR, PharmD
Summary date: 22 June 2016