Users.ecs.soton.ac.uk

For reprint orders, please contact:[email protected] Effects of dopamine D2 receptor gene polymorphisms on smoking cessation: abstinence and withdrawal symptoms Marcus R Munafò &
Evaluation of: Robinson JD, Lam CY, Minnix JA et al.: The DRD2 TaqI-B
Elaine C Johnstone
polymorphism and its relationship to smoking abstinence and withdrawal
symptoms. Pharmacogenomics J. (2006) (Epub ahead of print). An increasing
1University of Bristol, Department of Experimental number of studies have investigated the role of genetic factors in smoking cessation, and in response to smoking cessation pharmacotherapy. Robinson and colleagues report data on the effects of the dopamine D2 receptor gene Taq1B polymorphism on smoking Tel.: +44 117 954 6841;Fax: +44 117 928 8588; cessation and withdrawal. While the results add to the growing literature on smoking cessation pharmacogenetics, they also illustrate that minor differences in phenotype definition may influence the results of pharmacogenetic analyses. Existing studies also lack 2University of Oxford, Department of Clinical direct comparability, which makes it difficult to judge evidence for replication. Several future directions for smoking cessation pharmacogenetics research are discussed.
Infirmary, Oxford OX2 6HA, UKTel.: +44 186 522 4338; The pharmacogenetic analysis reported by Rob- allele responding to the drug with a substantial inson and colleagues [1] of clinical trial data from reduction in negative affect, whereas those carry- a Phase II placebo-controlled trial of venlafaxine ing the A1 allele (A1/A1 and A1/A2) showed no for smoking cessation, given in conjunction with significant reduction in negative mood [6].
behavioral support and nicotine replacement Participants were smokers recruited for a transdermal patch, adds to the small but rapidly double-blind, placebo-controlled clinical trial of growing literature on the pharmacogenetics of the antidepressant venlafaxine for smoking smoking cessation. Cigarette smoking is the larg- cessation [7] who were subsequently genotyped est preventable cause of death and morbidity in for the DRD2 Taq1A and Taq1B polymorphisms the USA, Europe and the developed world [2], but to investigate possible pharmacogenetic effects.
despite decades of research to develop effective Participants were required to smoke at least ten smoking cessation treatments, a large proportion cigarettes per day at baseline, and be aged of smokers who attempt to quit are doomed to between 18 and 75 years. Both the Taq1A1 and fail [3]. Therefore, the public health impact of cig- Taq1B1 polymorphisms have been reported to arette smoking remains substantial, which means be associated with a decrease in D2 dopamine that any improvement in the currently modest receptor density in the striatum [8,9] (i.e., reduced success rates achieved by smokers attempting to function), although there are conflicting quit would have the potential to afford a consid- results [10]. It would be interesting to know erable improvement in public health. This has whether these markers are in linkage disequilib- prompted a growing interest in pharmacogenetics rium in this study, as has been reported else- research to determine whether the efficacy of dif- where [11], in particular given the different results ferent pharmacotherapies for the treatment of for these markers reported by Robinson and nicotine dependence is influenced by inherited variation in drug-metabolizing enzymes and drug Participants recorded their daily cigarette intake each night in a diary prior to going to bed The findings reported by Robinson and col- for 56 consecutive days (14 days prequit and leagues extend on this work by investigating var- 42 days postquit), recording cigarette use over iation in a relatively rarely investigated the past 24 h, counting even a single puff as a cigarette. Daily abstinence was calculated based Keywords: dopamine, DRD2,
(DRD2) gene, the Taq1B variant, as well as the on this smoking diary, with zero cigarettes more commonly studied Taq1A variant. This defined as abstinent and all other responses, follows a previous report by the same group on including missing, defined as nonabstinent.
the same sample indicating a DRD2 Taq1A gen- Withdrawal symptoms were measured using two otype × treatment interaction effect on negative standardized, validated instruments – the Wis- affect, with smokers homozygous for the A2 10.2217/14622416.8.5.513 2007 Future Medicine Ltd ISSN 1462-2416 Pharmacogenomics (2007) 8(5), 513–517
PRIORITY PAPER EVALUATION – Munafò & Johnstone which consists of seven subscales (anger, anxiety, among studies of genetic and pharmacogenetic concentration, craving, hunger, sadness and effects on smoking cessation, where very few of sleep) and the Minnesota Withdrawal Symptom those studies published to date have been explicitly Checklist (MWSC), which is a nine-item list of designed to investigate these effects.
tobacco withdrawal symptoms summed to gen-erate an overall withdrawal scale. Both question- naires were completed on a daily basis for The study by Robinson and colleagues adds to the 56 days, along with the smoking diary.
growing but disparate literature on the role of Although the original trial from which these genetic effects on smoking cessation. Case–con- data were drawn randomized participants, strati- trol studies of smoking status phenotypes (e.g., fied by sex and history of depression, to receive comparing current smokers with ex-smokers) either venlafaxine or placebo, no DRD2 have generally reported an increased likelihood of genotype × treatment interactions effects were being a smoker among those carrying one or more copies of the reduced function Taq1A or Taq1B colleagues for either the A1 or A2 variants.
variants. Nevertheless, these findings have by no Instead, the results of interest relate to main means been consistent, and findings across studies effects of genotype on short-term daily absti- remain highly heterogeneous, with the overall evi- nence and withdrawal symptoms. Despite the dence for association remaining modest [12].
authors predicting that the reduced function Pharmacogenetic phenotypes have been argued to offer advantages in the study of smoking cessation decreased daily abstinence rates and increased treatment compared with traditional case–control withdrawal symptoms, no effects of this poly- phenotypes, such as greater refinement (e.g., pro- morphism were observed on either short-term spectively assessed abstinence symptoms, side- effects and smoking cessation under different However, the reduced function Taq1B1 vari- treatment conditions) and the use of experimental ant was observed to be associated with signifi- designs that control the dosing and timing of ther- cantly reduced daily abstinence as the study apy [4,5]. The use of daily diary data by Robinson and colleagues illustrates well the potential genotype × time interaction effect (p < 0.05).
strength of using rich time-course data to specify homozygous for the Taq1B2 allele were absti- nent, compared with 60% of those carrying one findings to date. For example, reduced function or more copies of the Taq1B1 allele. By 25 days alleles (i.e., those associated with, for example, post quit this difference had increased, with reduced mRNA expression) for polymorphisms approximately 40% of smokers homozygous for in the DRD2 gene (e.g., Taq1A1) have generally the Taq1B2 allele remaining abstinent, com- been shown to predict better response to nico- pared with less than 20% of those carrying one tine replacement therapy (NRT) [13–15], while or more copies of the Taq1B1 allele.
increased function alleles (e.g., Taq1A2) have In addition, significant DRD2 Taq1B × time been reported to predict better response to interaction effects were observed for several bupropion [14,16,17]. However, this overall picture WSWS scales, including anger (p < 0.02), anxi- is complicated by evidence that the widely inves- ety (p < 0.004), concentration (p < 0.05), sadness tigated DRD2 Taq1A polymorphism alters an (p < 0.05) and sleep (p < 0.02). A similar interac- amino acid in a previously undescribed protein tion effect was observed for the MWSC summary kinase gene (ANKK1) near the DRD2 locus [18].
score (p < 0.003). In all cases, participants with Nevertheless, this does not rule out an associa- the B2B2 genotype reported fewer withdrawal tion with the DRD2 gene, since data from the symptoms over time compared with participants HapMap project reveal that the Taq1A variant is with one or more copies of the B1 allele.
in linkage disequilibrium with other variants in These findings suggest, albeit tentatively given the DRD2 gene, but not with variants in the the marginal statistical significance of some of the ANKK1 gene [101]. Nevertheless, there have also findings and the large number of statistical tests, been failures to demonstrate an effect of DRD2 that smokers may exhibit distinct withdrawal tra- jectories on the basis of DRD2 genotype.
Another difficulty is that differences in study Although the results are based on a subgroup anal- design, and in the specific variants examined, limit ysis of genetic effects, this is not uncommon the comparability of the existing trials. With the Pharmacogenomics (2007) 8(5)
Effects of DRD2 polymorphisms on smoking cessation – PRIORITY PAPER EVALUATION exception of the two randomized controlled trials necessary, implementing genome-wide analyses of bupropion versus placebo [14,16,20,21], these stud- rather than focusing on single or a small number ies are not directly comparable, comprising both of variants [23]. Those studies that focus on a sin- randomized controlled trials and open-label effec- gle gene should investigate multiple functional tiveness trials. In particular, it is difficult to com- variants to maximally capture variation across pare genotype × treatment interaction effects the gene, and, ideally, single-gene studies should observed in placebo-controlled trials of single med- be supplanted by large-scale studies of multiple ications (e.g., NRT only), with main effects of genes operating within a single pathway (e.g., genotype observed in trials where participants are multiple dopamine receptor subtype genes, as offered combination pharmacotherapy (e.g., venla- well as genes coding for the dopamine trans- faxine plus NRT). Moreover, as mentioned previ- porter and those coding for dopamine metabo- ously, most of these studies were not explicitly lizing enzymes such as dopamine β-hydroxylase designed to address pharmacogenetic questions.
and catechol-O-methyl transferase, in the case of The results of the study by Robinson and col- the dopamine pathway). Clearly this will require leagues, while interesting, add yet another study studies at least an order of magnitude larger than design to those already employed in pharmaco- those currently available to achieve sufficient genetic studies of smoking cessation. It is notable that in a previous analysis of the same data by the same group, a significant effect of DRD2 Taq1A medications approved for other indications (e.g., genotype on smoking cessation was observed rimonabant), as well as current second-line medi- when assessing abstinence at regular intervals up cations for smoking cessation (e.g., nortriptyl- to 52 weeks post quit [6], as opposed to daily diary ine), will also require investigation, both in the assessment as in the current study. This highlights context of placebo-controlled trials and effective- that relatively subtle differences in study design, ness trials comparing multiple active medica- such as the operational definition of abstinence tions. It is possible, for example, that the employed, may afford quite different results.
apparent lack of efficacy of certain medicationsmay mask the existence of subgroups of individu- als for whom these medications may be effective, The study of genetic and phamacogenetic effects and who may be identifiable on the basis of geno- on smoking cessation remains in its infancy, but type. In the case of the current study, as the the number of studies in this field is increasing authors themselves note, candidate genes related rapidly. As in all genetic association studies, robust to serotonin neurotransmission may have been replication is the ultimate goal of these studies.
Therefore, future studies should strive for compa- genotype × treatment interaction effects, given rability with existing studies in order for the the mechanism of action of venlafaxine on serot- extent to which early findings are replicated to be onin reuptake. Retrospective genotyping of clini- adequately assessed. The cost and complexity of cal trial data will allow this possibility to be trials of smoking cessation pharmacotherapies, investigated, although any promising findings which typically offer treatment in combination will require confirmation in larger, explicitly with behavioral support and, as in the case of the study by Robinson and colleagues, in combina- Genetic associations with tolerability and side- tion with other pharmacotherapies, presents an effects should also be examined, as well as with obstacle to this. While meta-analysis has afforded withdrawal symptoms, as Robinson and col- an (imperfect) solution to the perennial problem leagues have done, given that there may exist fur- of lack of statistical power in case–control genetic ther subgroups of individuals who respond well association studies [22], where comparability across to certain medications that are typically not well studies tends to be greater than in pharmaco- tolerated, without the frequently concomitant genetic studies, it is not clear that this will offer a side-effect profiles associated with these medica- viable solution when the heterogeneity between tions. For example, an association between varia- tion in the DRD2 gene and poor adherence to As well as improving comparability across bupropion medication due to side-effects has studies, if we are to accurately and reliably iden- been reported [17], which illustrates the potential tify genetic variants associated with treatment to predict likely tolerance on the basis of geno- response, very large studies explicitly designed to type, even in the case of pharmacotherapies that PRIORITY PAPER EVALUATION – Munafò & Johnstone well as offer the potential to translate these find- emerge, our knowledge regarding the role of ings into practice and offer individualized treat- genetic variation in smoking cessation and ment tailored on the basis of genotype. Only response to smoking cessation pharmacotherapy then will we know whether studies such as that will increase substantially. This will provide by Robinson and colleagues will fulfil their important insights into the mechanisms of promise of allowing genotype-specific treatment nicotine addiction and smoking cessation, as Executive summary
Study design
• A pharmacogenetic analysis of a Phase II randomized, double-blind, placebo-controlled clinical trial of the antidepressant venlafaxine, given in conjunction with behavioral support and nicotine replacement therapy for smoking cessation, was performed. A total of 166 smokers were recruited, of which 147 were randomized and 128 followed-up to 12 months post cessation.
Comparison groups
• All participants were genotyped for the dopamine D2 receptor (DRD2) Taq1A and Taq1B polymorphisms, with one or more copies of the reduced activity (‘at risk’) allele (A1A1 or A1A2, and B1B1 or B1B2, respectively) compared with those homozygous for the wild-type allele (A2A2 and B2B2, respectively).
Eligibility
• Participants were required to smoke at least ten cigarettes per day at baseline, and be aged between 18 and 75 years old. Individuals were excluded if they were participating in other smoking cessation treatments or taking psychoactive medication, or had uncontrolled systemic illness, contraindications for venlafaxine or nicotine patches, current substance abuse or other psychiatric disorders.
End points
• The end points for the study were daily abstinence up to 42 days post quit, based on smoking diaries completed by participants and verified by expired air carbon monoxide (≤10 ppm) at 1, 3 and 6 weeks post quit. Withdrawal symptoms were measured by the Wisconsin Smoking Withdrawal Scale and the Minnesota Withdrawal Symptom Checklist questionnaires.
Results
• Smokers with an ‘at-risk’ B1 (B1B1 or B1B2) genotype were found to be less likely to report abstinence over the course of the 6-week post quit period as time progressed.
• Smokers with an ‘at-risk’ B1 (B1B1 or B1B2) genotype were found to be likely to report more withdrawal symptoms as • No significant DRD2 Taq1B genotype × treatment (venlafaxfine vs placebo) interactions, and no effects of DRD2 Taq1A genotype, Conclusion
• Smokers carrying the DRD2 Taq1-B1 allele were less likely to abstain from smoking, and reported significantly more symptoms of daily smoking withdrawal over time compared with smokers homozygous for the B2 allele. These findings may have implications for the development of genotype-specific treatment strategies.
Doll R, Peto R, Boreham J, Sutherland I: Comprehensive review of
Papers of special note have been highlighted as Mortality in relation to smoking: 50 years’ pharmacogenetics of nicotine addiction
either of interest (•) or of considerable interest (••) treatment, including a discussion of the
Br. Med. J. 328, 1519 (2004).
social and ethical implications of tailoring
Robinson JD, Lam CY, Minnix JA et al.: Hughes JR: Motivating and helping smokers treatment on the basis of genotype.
The DRD2 TaqI-B polymorphism and its to stop smoking. J. Gen. Intern. Med. 18, Cinciripini P, Wetter D, Tomlinson G et al.: The effects of the DRD2 polymorphism on withdrawal symptoms. Pharmacogenomics J. evidence for a pharmacogenetic effect on Illustration of the potential of
mood. Nicotine Tob. Res. 6, 229–239 pharmacogenetics to offer insight into
Curr. Opin. Mol. Ther. 7, 202–208 (2005).
individual differences in smoking cessation
Cinciripini PM, Tsoh JY, Wetter DW et al.: trajectory (e.g., abstinence and withdrawal
Combined effects of venlafaxine, nicotine symptoms) among abstinent smokers.
Pharmacogenomics 6, 211–223 (2005).
Pharmacogenomics (2007) 8(5)
Effects of DRD2 polymorphisms on smoking cessation – PRIORITY PAPER EVALUATION and short-term effectiveness of the nicotine Berlin I, Covey LS, Jiang H, Hamer D: Lack Psychopharmacol. 13, 282–292 (2005).
patch. Pharmacogenetics 14, 83–90 (2004).
of effect of D2 dopamine receptor TaqI A Lerman C, Jepson C, Wileyto EP et al.: Role et al.: Polymorphisms in the dopamine D2 Nicotine Tob. Res. 7, 725–728 (2005).
receptor gene and their relationships to dopamine D2 receptor (DRD2) in response Berrettini WH, Wileyto EP, Epstein L et al.: striatal dopamine receptor density of healthy Catechol-O-methyltransferase (COMT) volunteers. Mol. Psychiatry 4, 290–296 therapy for tobacco dependence: results of gene variants predict response to bupropion therapy for tobacco dependence. Biol. Neuropsychopharmacology 31, 231–242 Psychiatry 61(1), 111–118 (2007).
Lerman C, Shields PG, Wileyto EP et al.: Yudkin P, Munafò M, Hey K et al.: gene with receptor-binding characteristics in Effectiveness of nicotine patches in relation alcoholism. Arch. Gen. Psychiatry 48, cessation in a bupropion clinical trial. Health randomised controlled trial. Br. Med. J. 328, Psychol. 22, 541–548 (2003).
receptors binding potential is not affected by genetic association studies. Trends Genet. 20, et al.: Pharmacogenetic clinical trial of gene. Mol. Psychiatry 3, 261–265 (1998).
Liu QR, Drgon T, Walther D et al.: Pooled Ritchie T, Noble EP: Association of seven cessation. Nicotine Tob. Res. (In Press) association genome scanning: validation and use to identify addiction vulnerability loci in receptor gene with brain receptor-binding Swan GE, Valdes AM, Ring HZ et al.: two samples. Proc. Natl Acad. Sci. USA 102, characteristics. Neurochem. Res. 28, 73–82 Dopamine receptor DRD2 genotype and Murphy M, Walton R: The genetic basis for Pharmacogenomics J. 5, 21–29 (2005).
Pharmacogenomics 7, 1137–1140 (2006).
smoking behavior: a systematic review and meta-analysis. Nicotine Tob. Res. 6, 583–597 ANKK1: a novel kinase gene closely linked Johnstone EC, Yudkin PL, Hey K et al.: to DRD2 on chromosome band 11q23.1. Genetic variation in dopaminergic pathways Hum. Mutat. 23, 540–545 (2004).

Source: http://users.ecs.soton.ac.uk/mrm/Publications/2007%20Pharmacogenomics.pdf