|Home | About | Journals | Submit | Contact Us | Français|
There is a need to identify medications to aid in smoking cessation. Reducing withdrawal-related cognitive deficits represents a pharmacological target for new pharmacotherapies. Endogenous acetylcholine levels, which are modulated by acetylcholinesterase inhibitors (AChEIs), play an important role in smoking behavior and cognition. This pilot feasibility study tested whether an AChEI, donepezil, enhanced cognitive performance among healthy smokers.
Eighteen non-treatment seeking daily smokers (6 female) received either donepezil (5mg q.d) or placebo (double-blind; 2:1 allocation ratio) for four weeks. Smoking rate, side effects, and neurocognitive measures of working memory (Letter-N-back) and sustained attention (Penn Continuous Performance Task) were assessed weekly.
For the working memory task, there was a significant group × load × time interaction (p=0.03) indicating that the donepezil group demonstrated an increase in true positives from baseline to week 4 at the highest working memory load (3-back). The placebo group showed no change in accuracy. For the sustained attention task, there was a marginal effect in the same direction for discriminability, or d', p=0.08. There were no significant effects on reaction time during either task. There was also a reduction in cigarettes per day in the placebo group, but not the donepezil group.
AChEIs, such as donepezil, may have pro-cognitive effects among healthy smokers while they continue to smoke as usual. Given the association between cognitive deficits and relapse, AChEIs should be explored as potential therapeutics for smoking cessation.
Cigarette smoking remains the most significant public health problem globally (World Health Organization, 2008). Despite availability of FDA-approved medications for smoking cessation, only 1 in 4 smokers in the United States can maintain abstinence (Schnoll and Lerman, 2006). Indeed, most relapse within the first days following a quit attempt (Hughes et al., 2004). During this period, deficits in attention and working memory and alterations in task-induced neural activation are commonly observed (Addicott et al., 2012; Ashare and Hawk, 2012; Loughead et al., 2010; Myers et al., 2008); these deficits, in turn, predict relapse (Patterson et al., 2010; Powell et al., 2010). Thus, withdrawal-related cognitive deficits represent a novel target for nicotine dependence treatment development.
Several lines of research provide a rationale for examining acetylcholinesterase inhibitors (AChEIs) as potential therapeutics. First, nicotine administration enhances levels of acetylcholine (ACh) and the choline acetyltransferase (ChAT) enzyme, which is involved in the biosynthesis of ACh; whereas nicotine withdrawal in animals decreases ChAT enzyme activity (Arnold et al., 2003; Hernandez and Terry, 2005; Slotkin et al., 2008). AChEIs, which increase ACh in the synapse through inhibition of the catabolic enzyme, acetylcholinesterase (AChE), may substitute for the effects of nicotine. Further, genetic variation in the choline acetyltransferase (ChAT) gene, which regulates endogenous ACh levels, is associated with smoking cessation and nicotine dependence in independent cohorts (Ray et al., 2010; Wei et al., 2010). Lastly, AChEIs enhance cognition in patients with Alzheimer’s disease (Birks, 2006) and some healthy populations (Repantis et al., 2010). These studies support the biological plausibility of targeting endogenous acetylcholine levels to alleviate withdrawal-related cognitive deficits.
AChEIs are FDA-approved for treating the cognitive symptoms of Alzheimer’s disease (Terry and Buccafusco, 2003). Donepezil is the most commonly prescribed AChEI, and may have neuroprotective effects which are important for its pro-cognitive properties (Delrieu et al., 2011). Chronic treatment with donepezil increases functional up-regulation of the α7 and α4β2 nicotinic receptors (nAChRs), which are important for cognition (Takada-Takatori et al., 2009). Donepezil’s pro-cognitive effects in healthy populations are mixed (Repantis et al., 2010) and studies have begun to explore AChEI effects on smoking behavior (De la Garza and Yoon, 2011; Diehl et al., 2006; Kelly et al., 2008). However, no study to our knowledge has focused on the general population of smokers.
This pilot feasibility study examined: (1) tolerability and medication adherence, and (2) the effects of donepezil versus placebo on smoking behavior and cognitive performance in non-treatment seeking smokers. We predicted that four weeks of donepezil would improve working memory at the highest task difficulty level and sustained attention. Because participants in this study were not trying to quit, change in smoking behavior was a secondary outcome.
Eligible smokers were 18 to 50 years old, smoked at least 10 cigarettes per day for the previous 6 months, and had no plans to quit smoking in the next two months. Exclusion criteria included: pregnancy, lactation, or planning pregnancy; heart attack/stroke in previous 6 months; peptic ulcer disease; current diagnosis or history of DSM-IV Axis I disorders (except nicotine dependence); and current use of smoking cessation treatment, psychotropic medications, or contraindicated medications (e.g., anti-seizure medications).
This is a double-blind, between-subjects human laboratory study comparing four weeks of treatment with donepezil (5mg q.d.) to placebo. To gain more information about responses to donepezil, while minimizing loss of power, we employed a 2:1 allocation ratio (Friedman et al., 2010). Placebo and active study medication were packaged in identical capsules by Investigational Drug Services at the University of Pennsylvania. Following an intake to confirm eligibility, participants completed a baseline session, four testing days (at the end of each week during the 4-week treatment period), and four observation days (one in-between each testing day). During all visits, participants completed measures of smoking rate, pill count, and side effects and provided carbon monoxide breath samples. This manuscript focuses on the neurocognitive tasks (described below) assessed during baseline and testing days. The University of Pennsylvania Institutional Review Board approved all procedures and all participants provided written informed consent.
Sex, age, race, education, nicotine dependence (FTND; Heatherton et al., 1991), and cigarettes per day were assessed at baseline. At each visit, smoking rate was assessed using standard Timeline Followback methods (Brown et al., 1998). Weekly averages were computed to assess group differences in changes in smoking behavior from baseline to week 4.
Medication adherence was calculated as the percentage of pills consumed (out of 28). Side effect severity was rated on a 4-point scale (0=not present, 1=mild, 2=moderate, 3=severe) using a 38-item self-report measure based on common side effects of donepezil (e.g., nausea). The side effect summary score at week 4 (when medication reached steady state) was the dependent measure.
Neurocognitive task performance was assessed during baseline and each testing day (Day 7, 14, 21, and 28) using computerized tasks. Working memory was assessed with the Letter-N-back task (Ragland et al., 2002) and the primary outcomes were true positives and median correct reaction time. Sustained attention was assessed with the Penn Continuous Performance Task (P-CPT; Kurtz et al., 2001) and the primary outcomes were discriminability (d'), a signal detection measure calculated from hit rate (i.e., true positives) and false alarm rate (i.e., errors of commission) and the median correct reaction time. These tasks have been validated in healthy volunteers and patient populations (Gur et al., 2010; Kurtz et al., 2001; Ragland et al., 2002) and are sensitive to medication effects on cognition during abstinence and predict relapse (Patterson et al., 2010; 2009) (For detailed task descriptions see Patterson et al., 2009).
ANOVA models were used to examine medication adherence, side effects (controlling for baseline side effects), and smoking rate. Repeated-measures ANCOVAs tested the effects of donepezil on working memory (e.g., true positives, reaction time) and sustained attention (e.g., d', reaction time). For all models, group (donepezil vs. placebo) was a between-subjects factor and time (baseline, week 1, 2, 3, 4) was a within-subjects factor. For working memory, n-back load (0-, 1-, 2-, 3-back) was an additional within-subjects factor. Based on evidence that the 3-back condition is most sensitive to abstinence (Loughead et al., 2009) and medication effects (Loughead et al., 2010) and that performance during the 3-back condition predicts relapse (Patterson et al., 2010), post-hoc tests focused on the 3-back condition. The time by group interaction was tested, and sex, age, and baseline cigarettes per day were covariates in all models (results were comparable when FTND was a covariate).
Of the 89 participants eligible at phone screen, 55 scheduled an intake, and 22 were randomized (7 ineligible, 26 missed/refused). Eighteen (6 female) participants completed all study measures (4 withdrew). Using a 2:1 allocation ratio, 12 participants were randomized to donepezil and 6 to placebo. Demographic and smoking characteristics are presented in Table 1. Except for a marginal difference in nicotine dependence, F(1,17)=3.5, p=0.08, the groups did not differ on any demographic characteristics, all ps>0.18.
Medication adherence was excellent (96%) and did not differ between groups, F<1. Donepezil was well-tolerated and there were no reports of severe side effects. The side effect summary scores at week 4 were: donepezil (mean=1.1, SE=0.5), placebo (mean=2.3, SE=0.7), F(1,17)=1.6, p=0.22. In the donepezil group, all side effects were reported as mild and the most common were: abnormal dreams (n=3), dizziness (n=2), and cough/sore throat (n=2).
For changes in smoking behavior from baseline to week 4, the placebo group reported a greater reduction (mean difference=3.3 cigarettes, SE=1.4) compared to the donepezil group (mean difference=1.9 cigarettes, SE=1.0), group×time, F(1,12)=5.5, p=0.04. Change in cigarettes per day from baseline to week 4 was a covariate in cognitive task analyses.
For true positives, there was a significant group × load (0-, 1-, 2-, 3-back) × time interaction, F(1,11)=6.3, p=0.03; group×time, F(1,11)=6.4, p=0.03. Post-hoc comparisons revealed that there was a significant increase in true positives from baseline to Week 4 at the 3-back level in the donepezil group (mean difference=3.4, SE=0.88, p=0.001), but not the placebo group (mean difference=0.59, F<1), group × time p=0.037, d=1.4.
For reaction time, there were no significant group differences across time or load, Fs<1. There was an overall load×time interaction, F(1,11)=8.5, p=0.014, suggesting that from baseline to week 4, 3-back reaction time decreased (mean difference=134ms, SE=47, p=0.02). There were no significant changes at other n-back levels, all ps>0.20.
For discriminability (d'), there was a marginal group×time interaction, F(1,11)=3.8, p=0.08, d=0.66. In the donepezil group, d' tended to increase from baseline to week 4 (mean[SE]=4.5[0.29] vs 5.0[0.17], p=0.02), but the placebo group showed no change (mean[SE]=4.8[0.41] vs 4.7[0.24], p=0.90). There were no significant effects on P-CPT reaction time, all Fs<1.
Withdrawal-related cognitive deficits predict smoking relapse (Patterson et al., 2010) suggesting that pro-cognitive pharmacotherapies may be useful cessation aids. In the current study, donepezil, an FDA-approved treatment for the cognitive symptoms of Alzheimer’s disease, increased working memory accuracy among non-treatment seeking smokers. Consistent with our hypotheses, this effect was evident at the highest working memory load (3-back). There was a similar, but marginal, improvement in performance during the sustained attention task. Although the pro-cognitive effects of AChEIs in healthy populations and smokers with comorbid conditions are mixed (Kalechstein et al., 2011; Repantis et al., 2010; Sofuoglu et al., 2011), this is the first study to suggest that donepezil may have pro-cognitive effects among healthy smokers.
We did not observe benefits of donepezil for smoking behavior. This is consistent with previous research with galantamine among smokers with schizophrenia (Kelly et al., 2008; Sacco et al., 2008) and rivastigmine among methamphetamine-dependent smokers (De la Garza and Yoon, 2011). However, two studies of alcohol-dependent smokers suggested that galantamine and rivistagmine had beneficial effects on smoking rate (Diehl et al., 2006; 2009). Importantly, smokers in the current study were not actively seeking treatment and were asked to “smoke as usual” throughout the study period. Additionally, donepezil takes 15 to 21 days to reach steady state (Rogers et al., 1998), whereas galantamine takes 2 to 7 days and also enhances the affinity of α4β2 and α7 nAChRs for ACh , which may affect smoking behavior (Ago et al., 2011; Pandya and Yakel, 2011; Zhao et al., 2005). Therefore, extending donepezil’s treatment duration may be necessary to detect effects on smoking behavior. Notably, after controlling for the reduction in smoking rate observed in the placebo group, donepezil’s effects on neurocognitive performance remained significant.
These preliminary findings have some limitations. First, only two cognitive domains were assessed in a small group of smokers. Based on recruitment strategies previously used (Binder et al., 2001; Lee and Feng, 2005), a 2:1 allocation ratio was employed to maximize the amount of information gained from the donepezil group (Altman et al., 2001; Friedman et al., 2010). However, this resulted in a small placebo group and we should be cautious in our interpretations. Second, we used a lower dose of donepezil (5mg) than is typically prescribed for treating cognitive deficits (10mg). The 5mg dose was well-tolerated, but may have limited our ability to detect effects on reaction time and smoking behavior. These findings may not generalize to abstinence effects since cognitive performance was assessed while smokers were ‘smoking as usual.’ Future studies should examine abstinence effects of AChEIs in larger samples using a variety of cognitive tasks.
The cognitive benefits of donepezil were observed at the highest memory load. This is consistent with evidence that withdrawal-related cognitive deficits are most likely to emerge at the highest working memory load (3-back) (Loughead et al., 2010; 2009) and that deficits at the 3-back level predict relapse (Patterson et al., 2010). Therefore, treatments that alleviate these deficits may enhance abstinence rates (Sofuoglu, 2010). In summary, four weeks of treatment with the AChEI, donepezil was well-tolerated among healthy smokers. Importantly, AChEIs may improve neurocognitive performance among healthy smokers and future studies should examine whether AChEIs reduce withdrawal-related cognitive deficits and promote abstinence.
Role of Funding Source
This research was supported by grant #IRG-78-002-30 from the American Cancer Society and by grants from the National Cancer Institute and the National Institutes on Drug Abuse at the National Institutes of Health (P50 CA143187, R01 CA120594, and R01 CA130961). The ACS and NIH had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication.
The authors would also like to thank James Loughead, Robert Baron, Gretchen Evans, Rebecca Johnson, and Allison Gold for their contributions towards this study.
ContributorsDrs. Strasser, Ray, and Lerman designed the study and developed and implemented the protocol. Dr. Ashare conducted data analyses and wrote the initial draft of the manuscript. Drs. Strasser, Ray, and Lerman contributed to the statistical analysis and revision of subsequent drafts. All authors were involved in writing and revising the manuscript and all have approved the final manuscript.
Conflict of Interest
Dr. Ray has also received grant funding for an independent study from Pfizer, which markets Aricept. Pfizer has no contributions towards funding, design, or analysis from this study. Dr. Ray is no longer at the University of Pennsylvania; he is now employed by GSK Biologicals. Dr. Lerman has served as a consultant to Pfizer on pharmacogenetic testing for smoking cessation treatment and has received research funding from and consulted for AstraZeneca, Targacept, Pfizer, and GlaxoSmithKline, for work unrelated to this manuscript. Dr. Lerman also reports receiving compensation for expert testimony. Dr. Strasser has previously received grant funding for an independent study from Pfizer through the GRAND programme.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.