Skip to main content
NCBI home page
PMC Canada Author Manuscripts logoLink to PMC Canada Author Manuscripts
. Author manuscript; available in PMC: 2017 Jan 1.
Published in final edited form as: Drug Alcohol Depend. 2015 Nov 21;158:139–146. doi: 10.1016/j.drugalcdep.2015.11.017

Variation in CYP2A6 and tobacco dependence throughout adolescence and in young adult smokers

Meghan J Chenoweth a,b, Marie-Pierre Sylvestre c,d, Gisele Contreras c,d, Maria Novalen a,b, Jennifer O’Loughlin c,d, Rachel F Tyndale a,b,e
PMCID: PMC4698159  CAMSID: CAMS5389  PMID: 26644138

Abstract

Background

Smoking is influenced by genetic factors including variation in CYP2A6 and CYP2B6, which encode nicotine-metabolizing enzymes. In early adolescence, CYP2A6 slow nicotine metabolism was associated with higher dependence acquisition, but reduced cigarette consumption. Here we extend this work by examining associations of CYP2A6 and CYP2B6 with tobacco dependence acquisition in a larger sample of smokers followed throughout adolescence.

Methods

White participants from the Nicotine Dependence in Teens cohort that had ever inhaled (n=421) were followed frequently from age 12–18 years. Cox’s proportional hazards models compared the risk of ICD-10 tobacco dependence acquisition (score 3+) for CYP2A6 and CYP2B6 metabolism groups. Early smoking experiences, as well as amount smoked at end of follow-up, was also computed. At age 24 (N=162), we assessed concordance between self-reported cigarette consumption and salivary cotinine.

Results

In those who initiated inhalation during follow-up, CYP2A6 slow (vs. normal) metabolizers were at greater risk of dependence (hazards ratio (HR)=2.3; 95% CI=1.1, 4.8); CYP2B6 slow (vs. normal) metabolizers had non-significantly greater risk (HR=1.5; 95% CI=0.8, 2.6). Variation in CYP2A6 or CYP2B6 was not significantly associated with early smoking symptoms or cigarette consumption at end of follow-up. At age 24, neither gene was significantly associated with dependence status. Self-reported consumption was associated with salivary cotinine, a biomarker of tobacco exposure, acquired at age 24 (B=0.37; P<0.001).

Conclusions

Our findings extend previous work indicating that slow nicotine metabolism mediated by CYP2A6, and perhaps CYP2B6, increases risk for tobacco dependence throughout adolescence.

Keywords: nicotine, addiction, adolescent, smoking, genetics, CYP2A6, CYP2B6

1. Introduction

Approximately 90% of smokers begin smoking in adolescence (O’Loughlin et al., 2014b; U.S. Department of Health and Human Services, 2012). A substantial proportion (~40–75%) of smoking behaviour is influenced by genetics (Broms et al., 2006; Vink et al., 2005). CYP2A6 inactivates nicotine, the principle psychoactive compound in cigarette smoke, to cotinine (Nakajima et al., 1996). Genetic variation in CYP2A6 that reduces the rate of nicotine metabolism is associated with lower cigarette consumption (Malaiyandi et al., 2006; Wassenaar et al., 2011), dependence scores (Schnoll et al., 2014; Sofuoglu et al., 2012; Wassenaar et al., 2011), brain response to smoking cues (Tang et al., 2012), and greater cessation (Gu et al., 2000; Lerman et al., 2006; Schnoll et al., 2009), even in adolescence (Chenoweth et al., 2013). In adolescents, CYP2A6 slow nicotine metabolism was also associated with an increased risk of tobacco dependence acquisition at young ages (from age 12–16 years) (Al Koudsi et al., 2010 ; O’Loughlin et al., 2004), but slower escalation in nicotine dependence (Audrain-McGovern et al., 2007) and reduced cigarette consumption (O’Loughlin et al., 2004). In young adults, CYP2A6 slow (vs. normal) metabolizers were less likely to be smokers (Schoedel et al., 2004). Together these findings suggest that while CYP2A6 slow metabolism increases the risk of becoming a smoker in younger adolescence, slow metabolism also increases cessation, and reduces cigarette consumption in dependent smokers. However, it is not known whether CYP2A6 slow metabolism increases smoking acquisition in later adolescence, a period during which a substantial amount of smoking uptake occurs (O’Loughlin et al., 2014b).

A small proportion (~10%) of nicotine’s metabolism to cotinine occurs via a second enzyme, CYP2B6 (Al Koudsi and Tyndale, 2010). The CYP2B6*6 allele, a prevalent haplotype (~25% frequency in Whites (Rotger et al., 2007)) is associated with lower CYP2B6 hepatic protein levels (Al Koudsi and Tyndale, 2010) and slower CYP2B6-mediated metabolism of bupropion and efavirenz (reviewed in (Thorn et al., 2010)). In adult smokers, CYP2B6*6 was associated with lower abstinence rates in the placebo arm of a bupropion smoking cessation clinical trial; 15% of individuals with one or two copies of CYP2B6*6 achieved abstinence, compared to 32% of CYP2B6*1/*1 individuals (Lee et al., 2007a). In a separate study, the CYP2B6*6 allele was more frequent in nicotine dependent individuals compared to those that were not dependent (32% vs. 22%, respectively) (Riccardi et al., 2015). Whether CYP2B6*6 also influences the risk for acquiring nicotine dependence in adolescence is not known.

Here we examined associations for CYP2A6 and CYP2B6 with tobacco dependence acquisition in a larger (n>400) sample of adolescent smokers assessed four times each year across the entire adolescent period (age 12–18 years). We hypothesized that CYP2A6 slow (vs. CYP2A6 normal), and that CYP2B6 slow (i.e., individuals with one or two copies of CYP2B6*6) (vs. CYP2B6 normal) metabolizers would be at increased risk of acquiring dependence. We also hypothesized that a larger proportion of slow (vs. normal) metabolizers for each gene would report early smoking experiences, which are associated with the development of nicotine dependence (DiFranza et al., 2004). We also assessed cigarette consumption at the end of follow-up among dependent smokers, hypothesizing that CYP2A6 slow (vs. CYP2A6 normal) metabolizers would smoke fewer cigarettes; no association was expected between CYP2B6 genotype groups. For both CYP2A6 and CYP2B6, we further hypothesized that slow (vs. normal) metabolizers would be more likely to be dependent at end of follow-up. At age 24, we expected CYP2A6 slow (vs. normal) metabolizers to be at lower risk of dependence, as CYP2A6 slow metabolizers are less likely to be dependent smokers (vs. non-smokers) as adults (Schoedel et al., 2004). Finally, we hypothesized that CYP2B6 slow (vs. normal) metabolizers would be more likely to be dependent at age 24, consistent with the higher frequency of CYP2B6*6 in dependent (vs. non-dependent) adults (Riccardi et al., 2015).

Finally, an adjunct biochemical analysis to assess the validity of the self-reported cigarette consumption data was undertaken. We examined the construct-related validity of self-reported cigarette consumption against salivary cotinine, widely used as an objective biomarker of tobacco consumption (Connor Gorber et al., 2009), and also assessed its relationships with nicotine dependence and withdrawal scores.

2. Methods

2.1. Study population and data collection

As previously described (O’Loughlin et al., 2014a), 1294 adolescents from 10 secondary schools in Quebec were recruited in 1999 for the Nicotine Dependence in Teens (NDIT) cohort study (Montreal, Quebec, Canada). Self-report questionnaires were administered every three months during the 10-month school year over the five years of secondary school (grade 7–11), for a total of 20 survey cycles. Data from these 20 surveys for n=421 ever smoking Whites were included in the current analyses of tobacco dependence acquisition and cigarette consumption. Previous analyses in this population included data only up to survey cycle 16 (age 15–16 years) for only 281 smokers (O’Loughlin et al., 2004). Two additional surveys (survey cycles 21 and 22) were administered three and six years, respectively, after high school graduation (O’Loughlin et al., 2014a). Data from survey 22, completed when participants were aged 24 years on average, were used, along with salivary cotinine, for additional analyses. Parents or guardians provided written informed consent and participants provided assent at baseline. Participants (who had attained legal age) provided informed consent during post-high school survey cycles. The study was approved by McGill University (Quebec, Canada), the Centre de recherche du Centre hospitalier de l’Université de Montréal (Quebec, Canada), and the University of Toronto research ethics board (Toronto, Canada).

2.2. Determination of CYP2A6 and CYP2B6 genotype

DNA was extracted from saliva or blood samples, and participants were genotyped for four CYP2A6 alleles which occur at relatively high frequency (~1–8%) in Whites and have an established impact on reducing nicotine metabolism: CYP2A6*2, CYP2A6*4, CYP2A6*9, and CYP2A6*12 (Benowitz et al., 2006; Chenoweth et al., 2013). Participants were grouped into CYP2A6 normal, intermediate, or slow nicotine metabolism groups based on the predicted metabolic impact of each CYP2A6 variant allele (Chenoweth et al., 2013). Participants were also genotyped for the CYP2B6*6 allele, using a haplotyping method described previously (Lee et al., 2007a; Mwenifumbo et al., 2005); individuals with the CYP2B6*1/ *1 genotype were grouped as normal metabolizers, while those with one or two copies of the CYP2B6*6 alleles were grouped as slow metabolizers. For genetic analyses, the sample was restricted to White ever-smokers in order to minimize possible effects of population stratification. CYP2A6 genetic data were available for 421 White ever-smokers, while CYP2B6 genetic data were available for 391 White ever-smokers.

2.3. Study variables

Data on early smoking experiences (symptoms of nausea and dizziness), tobacco/nicotine dependence (measured by the International Classification of Diseases (ICD)-10 and the modified Fagerstrom Tolerance Questionnaire (mFTQ) were collected, as were data on nicotine withdrawal, other nicotine dependence symptoms, and self-medication scores (Chenoweth et al., 2013). Briefly, nicotine withdrawal, other nicotine dependence symptoms, and self-medication scores were measured in six, 14, and five individual items, respectively, and assessed symptoms of withdrawal including irritability, restlessness, anxiety, craving frequency, and endorsement of statements that smoking improves energy level, affect, and stress (Chenoweth et al., 2013). Scores were pro-rated if there were fewer than half the items missing for an individual score. If half or more of the items were missing, the participant was assigned a missing value.

Three measures of cigarette consumption were used. The mean number of cigarettes smoked/month in the 3-month interval preceding each survey cycle was assessed by multiplying the average number of cigarettes smoked/day by the average number of days smoked/month in each of the three months and calculating the average monthly consumption (O’Loughlin et al., 2014c). The number of cigarettes smoked in the past week was assessed at survey cycle 22 (approximately age 24) by summing the number of cigarettes smoked on each of the preceding seven days. Finally, the number of cigarettes smoked in the past 24 hours was assessed at survey cycle 22, indicated by the number of cigarettes smoked on the day prior to administration of the survey.

2.4. Determination of salivary cotinine level

Cotinine levels were determined from saliva samples collected from current smokers at survey cycle 22, when participants were 24 years of age on average. The level of cotinine was assessed using liquid chromatography-tandem mass spectrometry (Chenoweth et al., 2014; St Helen et al., 2012; Tanner et al., 2015). In total, cotinine levels were available for n=162 White self-reported current smokers with genetic data. The limit of quantification for cotinine was 1 ng/ml. Four participants had cotinine levels below the limit of quantification; as per convention, the cotinine value was replaced with the limit of quantification (1 ng/ml) divided by the square root of 2, yielding a value of 0.7 ng/ml (Kalkbrenner et al., 2010).

2.5. Statistical analyses

Analysis 1. Tobacco dependence acquisition, early smoking experiences, the level of smoking in adolescence, and dependence status at end of follow-up (Surveys 1–20)

Data from surveys 1–20 (age 12–18 years) were included in this analysis; surveys 21 and 22, both collected in young adulthood (ages 20 and 24 years, respectively), were not included in this analysis due to the much less frequent data collection interval. Using bootstrap-based multiple imputation to manage missing data, 10 imputed datasets were created as previously described (Chenoweth et al., 2013). Cox’s proportional hazards models were used to compare the risk of acquiring ICD-10 tobacco dependence (score of ≥3) in adolescence (age 12–18 years) between CYP2A6 normal, intermediate, and slow metabolizers who had ever inhaled on a cigarette, and between CYP2B6 normal and slow metabolizers who had ever inhaled on a cigarette. Models were stratified according to whether the participant had inhaled at cohort inception; separate models examined all inhalers (i.e., those who inhaled prior to cohort inception + those who inhaled during follow-up) and those who first inhaled during follow-up (i.e., incident inhalers), as data covering the entire smoking career were available in this latter group. Time was measured in days from time zero in all participants who ever inhaled on a cigarette, which corresponded to the day participants who had inhaled prior to cohort inception joined NDIT, or the day incident smokers first inhaled during NDIT follow-up. Participants were followed from time zero until they became dependent or were censored (i.e., were lost to follow-up or the follow-up period ended). Once smokers entered the analysis, all surveys were considered to be ‘at-risk’ periods. Early smoking experiences (nausea and dizziness) were assessed in the first survey completed for prevalent inhalers, and in the first survey completed following smoking initiation for incident inhalers. Past-month cigarette consumption at end of follow-up (i.e., each participant’s last available survey from surveys 1 to 20) was compared between CYP2A6 genotype groups and between CYP2B6 genotype groups, with and without covariates (sex, age, and duration of smoking) using linear regression analysis. Dependence status at end of follow-up was also assessed for each genotype group.

Analysis 2. Associations for CYP2A6 and CYP2B6 with tobacco dependence status in young adulthood (Survey 22)

Using the same study population from analysis 2, we performed chi-square tests to compare the frequency of ICD-10 tobacco dependence (score of ≥3) across genotype groups.

Analysis 3. Construct-related validity of self-reported smoking behaviour (cigarette consumption) against cotinine in young adults (Survey 22, age 24)

Cotinine was not normally distributed. Therefore, the strength of the association between self-reported cigarette consumption and cotinine level was assessed using Spearman’s rho and linear regression analysis. We also explored the relationships between cotinine and nicotine dependence scores (i.e., ICD-10, mFTQ dependence scores, withdrawal, other nicotine dependence symptoms, and self-medication) and smoking indicators using Spearman’s rho and by conducting a separate univariate linear regression analysis for each smoking or nicotine dependence indicator.

3. Results

Participant characteristics at baseline are shown in Table 1. In all participants (n=421), 78.9%, 14.5%, and 6.7% were CYP2A6 normal, CYP2A6 intermediate, and CYP2A6 slow metabolizers, respectively. In those for whom CYP2B6 genetic data was available (n=391), 58.3% were CYP2B6 normal (n=228), while 41.7% were CYP2B6 slow (n=163), metabolizers. The frequency of the CYP2B6*6 allele was 24%, consistent with a previous report in Caucasians (Rotger et al., 2007). Eleven individuals are expected to be both CYP2A6 and CYP2B6 slow metabolizers (6.7% x 41.7% x N=391); twelve were identified which does not deviate substantially from the expected value.

Table 1.

Selected baseline characteristics of genotyped White ever-smokers, according to whether they had inhaled at cohort inception. NDIT 1999–2012.

All inhalers (inhaled before cohort inception or during follow-up)
n=420–421a 		Incident inhalers (inhaled during follow-up)
n=214
Male, %a 36.2–36.3 35.5
Mean age in years (SD) 12.7 (0.5) 12.6 (0.4)
Francophone, %a 18.8–19.1 14.5
Single parent family, %a 9.3–9.5 5.6
Parent(s) university/college educated, %a 50.0–52.4 57.0–61.2
Parent(s) smoke, %a 42.9–43.3 28.5–29.0
Friend(s) smoke, %a 48.8–49.1 30.4–30.8
Open in a new tab
a

Ranges in values represent the values derived from the 10 imputed datasets

The median follow-up time was 1574 days. We first examined the risk of tobacco dependence acquisition according to CYP2A6 metabolism group in incident inhalers and in all inhalers. In the incident inhalers, CYP2A6 slow metabolizers were more likely to acquire tobacco dependence than CYP2A6 normal metabolizers, with a hazards ratio (HR) of 2.3 (95% confidence interval (CI) = 1.1, 4.8) (Table 2). Similar findings were observed in all inhalers (HR = 1.8, 95% CI = 1.0, 3.3 for tobacco dependence acquisition in CYP2A6 slow vs. normal metabolizers) (Table 2). In the incident inhalers, CYP2B6 slow metabolizers were no more likely than CYP2B6 normal metabolizers to acquire tobacco dependence (HR = 1.5, 95% CI = 0.8, 2.6). Similar findings were observed in all inhalers (HR = 1.2, 95% CI = 0.8, 1.7 for tobacco dependence acquisition in CYP2B6 slow vs. normal metabolizers) (Table 2). We were underpowered to test for potential interactions between CYP2A6 and CYP2B6 on tobacco dependence acquisition. No associations for CYP2A6 or CYP2B6 with the frequency of nausea or dizziness during early smoking were found (Table 3).

Table 2.

Hazard ratios (HR) and 95% confidence intervals (CI) for ICD-10 tobacco dependence in adolescent smokers according to CYP2A6 and CYP2B6 metabolism groups. NDIT 1999–2012.

Metabolism group HR (95% CI)
All inhalersa 		HR (95% CI)
Incident inhalers
CYP2A6
Normal (reference) 1.0; n=330–333 1.0; n=162–165
Intermediate 0.8 (0.4, 1.4); n=60–63 1.0 (0.4, 2.4); n=29–32
Slow 1.8 (1.0, 3.3); n=27–28 2.3 (1.1, 4.8); n=20–21
CYP2B6
Normal (reference) 1.0; n=228 1.0; n=121
Slow 1.2 (0.8, 1.7); n=163 1.5 (0.8, 2.6); n=82
Open in a new tab
a

Includes prevalent (i.e., those who had inhaled before cohort inception) and incident (i.e., those who first inhaled during follow-up) inhalers

Table 3.

Frequency of early smoking experiences among participants who inhaled, according to CYP2A6 and CYP2B6 metabolism group. NDIT 1999–2012.

Metabolism group All inhalersa Incident inhalers
% (n) Experiencing Nauseab % (n) Experiencing Dizzinessb % (n) Experiencing Nauseab % (n) Experiencing Dizzinessb
CYP2A6
Normal 34.5% (114) 13.7% (45) 33.7% (55) 13.6% (22)
Intermediate 37.7% (23) 18.0% (11) 36.7% (11) 13.3% (4)
Slow 40.7% (11) 14.8% (4) 40.0% (8) 15.0% (3)
P valuec 0.75 0.67 0.83 0.98
CYP2B6
Normal 33.6% (76) 14.2% (32) 33.3% (40) 13.3% (16)
Slow 33.7% (55) 16.0% (26) 34.1% (28) 14.8% (12)
P valuec 0.98 0.61 0.90 0.77
Open in a new tab
a

Includes prevalent (i.e., those who had inhaled before cohort inception) and incident (i.e., those who first inhaled during follow-up) inhalers

b

Nausea and dizziness was indicated by the response “a lot” or “a bit” (vs. “not at all”) when participants were asked about experiencing these symptoms the first time they smoked.

c

P values are derived from chi-square tests.

We next investigated the level of cigarette consumption in incident smokers at the end of follow-up in adolescence between CYP2A6 and CYP2B6 metabolism groups according to dependence status, with and without covariates. In the unadjusted and adjusted models, dependent CYP2A6 slow metabolizers smoked 108 and 74 fewer cigarettes/month, respectively, compared to dependent CYP2A6 normal metabolizers, however neither finding was significant (Table 4). In the unadjusted and adjusted models, dependent CYP2B6 slow metabolizers smoked 67 and 55 more cigarettes/month, respectively, compared to dependent CYP2B6 normal metabolizers, which was not significant (Table 4). At the end of follow-up in adolescence, a greater proportion of CYP2A6 slow (vs. normal) metabolizers were dependent (46% vs. 23%, respectively), but this did not reach significance (median P value from 10 imputed datasets = 0.09). CYP2B6 slow metabolizers were no more likely than CYP2B6 normal metabolizers to be dependent at end of follow-up (21% vs. 30%, respectively; P=0.15).

Table 4.

Linear regression analysis of past month cigarette consumption at end of follow-up among participants who initiated inhalation during follow-up, according to dependence status and CYP2A6 and CYP2B6 metabolism group. NDIT 1999–2012.

ICD-10 Tobacco Dependent
All Yes No
Coefficient (95% CI)a Coefficient (95% CI)a Coefficient (95% CI)a
Unadjusted Model
CYP2A6 Normalc 64.2 (39.6, 88.7); n=162–165 222.3 (146.4, 298.1); n=36–38 17.9 (2.2, 33.6); n=126–127
CYP2A6 Intermediate 1.9 (−66.1, 70.0); n=29–32 1.2 (−187.4, 189.7); n=6–8 2.0 (−54.9, 58.9); n=23–24
CYP2A6 Slow −13.1 (−100.3, 74.0); n=20–21 −108.4 (−303.8, 86.9); n=9–10 −17.6 (−72.3, 37.0); n=11–12
Adjusted Modelb
CYP2A6 Normalc 66.9 (43.1, 90.7); n=162–165 221.8 (149.1, 294.4); n=36–38 18.8 (3.2, 34.5); n=126–127
CYP2A6 Intermediate −5.8 (−72.5, 60.7) ; n=29–32 −41.4 (−228.5, 145.6); n=6–8 −4.6 (−61.9, 52.8); n=23–24
CYP2A6 Slow −30.5 (−115.5, 54.6); n=20–21 −74.0 (−269.4, 121.4); n=9–10 −17.4 (−71.6, 36.8); n=11–12
Unadjusted Model
CYP2B6 Normalc 46.3 (15,0, 77.6); n=121 178.8 (78.3, 279.3); n=25 11.8 (−8.7, 32.2); n=96
CYP2B6 Slow 47.2 (0.3, 94.2); n=82 66.6 (−66.1, 199.2); n=25 15.1 (−17.0, 47.3); n=57
Adjusted Modelb
CYP2B6 Normalc 48.7 (18.3, 79.2); n=121 184.5 (86.7, 282.3); n=25 12.6 (−8.4, 33.6); n=96
CYP2B6 Slow 41.2 (−4.5, 86.9); n=82 55.1 (−74.5, 184.8); n=25 13.1 (−20.5, 46.7); n=57
Open in a new tab

Abbreviations: CI, confidence interval

a

The regression coefficients and 95% confidence interval are reported, derived from 10 imputed datasets.

b

Covariates included: sex, age, time since first inhalation

c

Intercept in regression

Note: Coefficients for CYP2A6 intermediate and CYP2A6 slow metabolizers are expressed relative to CYP2A6 normal metabolizers, while coefficients for CYP2B6 slow metabolizers are expressed relative to CYP2B6 normal metabolizers. For example, in the unadjusted model in the dependent smokers, CYP2A6 slow metabolizers smoked 108 fewer cigarettes compared to CYP2A6 normal metabolizers.

At age 24, 29% of CYP2B6 normal metabolizers (CYP2B6*1/*1; n=82) were ICD-10 dependent compared to 45% of CYP2B6 slow metabolizers (CYP2B6*1/ *6 + *6/ *6; n=65), however this was not significant (P=0.05). There was no difference in dependence status between CYP2A6 genotype groups at age 24; 35% of CYP2A6 normal metabolizers (N=125) were dependent (score 3+), compared to 42% and 56% of intermediate (N=24) and slow (N=9) metabolizers, respectively (P=0.44). The low number of CYP2A6 slow metabolizers could have significantly limited statistical power for this cross-sectional analysis.

We also undertook an adjunct analysis to determine the construct-related validity of using self-reported smoking behaviour as an indication of tobacco dose. We first focused on the relationship between self-reported cigarette consumption (number of cigarettes/month), and salivary cotinine level (available for n=162 self-reported current smokers at age 24). The measures were strongly correlated (Rho=0.71, P<0.001) (Figure 1a). Using linear regression, self-reported cigarette consumption was associated with salivary cotinine level (B=0.37, P<0.001) (Figure 1a). We next examined the relationship between past-week and past-day self-reported cigarette consumption and salivary cotinine level, and these measures were also strongly correlated (Rho=0.75 and 0.73, respectively; P<0.001). Both past-week and past-day self-reported cigarette consumption were also associated with cotinine in linear regression analyses (B=1.8 and 11.2, respectively, P<0.001). We also observed correlations between cotinine and nicotine dependence scores (i.e., ICD-10 and mFTQ dependence scores, withdrawal, other nicotine dependence symptoms, and self-medication; Figure 1b–f; Rho=0.28–0.61, all P<0.001). In linear regression analyses, ICD-10 dependence (B=23.2, P=0.001), mFTQ dependence (B=37.6, P<0.001), withdrawal (B=7.9, P<0.001), other nicotine dependence symptoms (B=8.1, P<0.001), and self-medication scores (B=15.1, P=0.008) were all significantly associated with cotinine (Figure 1b–f).

Figure 1.

Figure 1

Open in a new tab

Self-reported smoking behaviours are strongly associated with salivary cotinine levels in young adults. Correlation between self-reported cigarettes/month and cotinine, demonstrating construct-related validity, is shown in (a). Correlations between cotinine and ICD-10 (b) and mFTQ (c) dependence scores, withdrawal (d), other nicotine dependence symptoms (e), and self-medication (f) scores are also shown in all White self-reported current smokers (n=162).

4. Discussion

In this longitudinal study in adolescent smokers, we extend previous findings (O’Loughlin et al., 2004) of an increased risk for tobacco dependence acquisition among CYP2A6 slow nicotine metabolizers relative to CYP2A6 normal metabolizers, by demonstrating that this elevated risk occurs throughout adolescence. Despite their increased risk of acquisition, CYP2A6 slow metabolizers are also more likely to quit smoking throughout adolescence and adulthood (Chenoweth et al., 2013; Gu et al., 2000); thus depending on the point in time during the transition from adolescence to adulthood, or in smoking dependence trajectory, slow metabolizers may be over- or under-represented in smokers versus non-smokers. This may contribute to the observed lower risk among slow (vs. normal) metabolizers for being a smoker versus non-smoker in young adulthood (Schoedel et al., 2004). While the precise mechanism(s) underlying the association between CYP2A6 slow metabolism and increased risk for tobacco dependence acquisition in adolescence is unknown, our data suggest that experiencing early nausea or dizziness may not contribute. The effect of slower nicotine metabolism on nicotine-mediated reward was recently characterized in nicotine-naïve adult mice (Bagdas et al., 2014). Slower nicotine metabolism, achieved through methoxsalen-mediated inhibition of CYP2A5 (the murine ortholog of human CYP2A6), caused animals to display a preference for nicotine in the conditioned place preference paradigm (Bagdas et al., 2014), suggestive of greater reward (Tzschentke, 2007). If these findings extend to novice smokers, those with slow nicotine metabolism may also experience greater reward during initial smoking experiences, potentially due to greater nicotine exposure, increasing their risk for developing nicotine dependence.

Despite their higher risk of acquiring tobacco dependence, once dependent, the level of cigarette consumption appeared to be lower in CYP2A6 slow (vs. normal) metabolizers, however this was not significant. Slow nicotine metabolism, measured by CYP2A6 genotype or NMR, is associated with lower cigarette consumption among dependent adults smoking a mean of 18+ cigarettes/day (Chenoweth et al., 2014; Malaiyandi et al., 2006; Wassenaar et al., 2011). Future investigations will clarify whether dependent adolescent smokers, even at comparatively lower levels of cigarette consumption versus adults, titrate their level of nicotine intake according to their rate of nicotine metabolism in order to maintain desirable levels (Ashton et al., 1979; Hill and Marquardt, 1980), with CYP2A6 slow metabolizers needing to smoke fewer cigarettes to achieve these levels. In contrast to CYP2A6, variation in CYP2B6 does not substantially alter the rate of peripheral nicotine metabolism or cigarette consumption (Lee et al., 2007b); we also showed a lack of significant association between CYP2B6 and adolescent cigarette consumption.

In a cross-sectional study of adolescent smokers (<5 cigarettes/day) aged 13 to 17 years (mean = 16 years), NMR was negatively associated with mFTQ dependence scores and cigarette consumption, where slower metabolizers displayed both higher dependence and higher consumption (Rubinstein et al., 2013). It is possible that this study captured an earlier time-point in smoking history than our study, where the CYP2A6 slow metabolizers were more likely to be dependent (and have higher dependence scores), but had not yet begun to titrate their nicotine intake according to their rate of nicotine metabolism (i.e., had not begun to smoke fewer cigarettes).

Many studies of adolescent smoking, including our own, utilize self-report questionnaires (Ary and Biglan, 1988; Rigotti et al., 2000). Misclassification in self-report data can occur in epidemiological studies, particularly in studies of substance use behaviour and in younger cohorts (Brener et al., 2003; Clarke et al., 2014; Fendrich et al., 2005; Morral et al., 2000). There may be issues stemming from recall bias, as well as the desire to conform with social norms (Brener et al., 2003). In order to help minimize recall bias, we collected data frequently (every three months) in adolescence (O’Loughlin et al., 2014a), and assessed tobacco dependence acquisition in all inhalers as well as in those who initiated inhalation during follow-up (i.e., incident inhalers). We also assessed the accuracy of self-reported cigarette consumption; there was strong agreement between self-reported cigarette consumption and salivary cotinine, supporting the construct validity of the questionnaire data. A limitation of our study is a lack of available biomarker data during adolescence. In addition, self-reported smoking is not always concordant with salivary cotinine levels in adolescents (Kandel et al., 2006). The associations between cotinine and the other smoking and dependence measures at age 24 suggest that these measures are also related to tobacco dose. Associations between dependence measures and salivary cotinine levels have been previously demonstrated in young adolescent smokers with an average age of 15 years (Rubinstein et al., 2007). Dependence measures including the mFTQ, nicotine dependence syndrome scale, timing of craving in the morning, and self-rated level of addiction were all correlated with cotinine level (Rubinstein et al., 2007). In these young adolescent smokers, the mean cotinine level and cigarette consumption were both lower than in our sample at age 24 (~44 vs. ~139 ng/ml, and ~4 vs. ~8 cigarettes/day, respectively). Together these findings suggest that dependence measures are related to tobacco dose across a range of exposures and ages (Rubinstein et al., 2007).

In a recent study of young smokers, the influence of CYP2A6 variation on smoking outcomes varied throughout the 6-year follow-up period (Cannon et al., 2015). At age 16, CYP2A6 intermediate metabolism (vs. normal and slow metabolism) was associated with the highest nicotine dependence syndrome scale (NDSS) score; in contrast, as young adults (age 22), CYP2A6 normal metabolizers had the highest NDSS score (Cannon et al., 2015). We did not replicate these findings in our sample at age 24. Differences between our findings and those from the longitudinal investigation (Cannon et al., 2015) may stem from differences in study design (cross-sectional vs. longitudinal), variation in assessment of CYP2A6 activity (categorical grouping of metabolism vs. use of a metabolism metric), and/or differences in smoking phenotypes examined (ICD-10 dependence status vs. NDSS score). Our small sample size for slow metabolizers (n=9) may have also affected statistical power. Future longitudinal investigations will help to clarify the factors, including the role of CYP2A6, associated with smoking in adolescence, and as youth transition to adulthood. A better understanding of these factors may lead to targeted interventions that help reduce the risk for lifelong smoking.

While there was some indication that CYP2B6 slow metabolizers were at greater risk of acquiring tobacco dependence in adolescence, this did not reach significance. When we assessed the participants in young adulthood (age 24), there was a trend toward higher dependence rates in CYP2B6 slow (vs. normal) metabolizers. These findings are consistent with a recent report in Italians where the frequency of the CYP2B6*1/*6 genotype was significantly higher in individuals with nicotine dependence (FTND score 4+) compared to those who were not dependent (36% vs. 21%, respectively), which was not seen for those with the normal CYP2B6*1/*1 genotype (20% vs. 33%, respectively) (Riccardi et al., 2015). Together with our findings, these data suggest that the CYP2B6*6 allele is associated with a higher risk of nicotine dependence in adulthood; over adolescence we speculate that the risk conferred by CYP2B6 slow metabolism increases. Data from an animal model suggests variable brain CYP2B activity may influence nicotine metabolism within the brain, in turn affecting nicotine-mediated behaviours (Garcia et al., 2015). In rats, the inhibition of brain CYP2B activity through intracerebroventricular injection of a selective CYP2B inhibitor led to higher rates of acquisition of nicotine self-administration behavior but no difference in level of responding among dependent animals; this was performed without altering peripheral nicotine levels or metabolism (Garcia et al., 2015). It is possible that CYP2B6 variation within human brain may lead to altered central metabolism of nicotine, which may account for the observed differences in nicotine dependence (Riccardi et al., 2015) and cessation outcomes (Lee et al., 2007a), while having no effect on the level of consumption as observed here.

Overall, we have shown that throughout adolescence, CYP2A6 slow nicotine metabolizers are at a higher risk of developing tobacco dependence. The role of genetic variation in CYP2B6 in smoking acquisition and dependence remains to be clarified, but may increase over time. The adjunct analysis validating our measure of smoking supports these conclusions. The findings highlight the role of genetic risk factors in nicotine addiction in adolescence, a development period in which the contribution of genetics to smoking behaviours has been studied only infrequently to date.

Footnotes

Conflict of interest: In the past three years, Dr. Tyndale has consulted for Apotex. The remaining authors declare no conflicts of interest.

References

  1. Al Koudsi N, O’Loughlin J, Rodriguez D, Audrain-McGovern J, Tyndale RF. The genetic aspects of nicotine metabolism and their impact on adolescent nicotine dependence. Journal of Pediatric Biochemistry. 2010;1:105–123. [Google Scholar]
  2. Al Koudsi N, Tyndale RF. Hepatic CYP2B6 is altered by genetic, physiologic, and environmental factors but plays little role in nicotine metabolism. Xenobiotica. 2010;40:381–392. doi: 10.3109/00498251003713958. [DOI] [PubMed] [Google Scholar]
  3. Ary DV, Biglan A. Longitudinal changes in adolescent cigarette smoking behavior: onset and cessation. J Behav Med. 1988;11:361–382. doi: 10.1007/BF00844936. [DOI] [PubMed] [Google Scholar]
  4. Ashton H, Stepney R, Thompson JW. Self-titration by cigarette smokers. Br Med J. 1979;2:357–360. doi: 10.1136/bmj.2.6186.357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Audrain-McGovern J, Al Koudsi N, Rodriguez D, Wileyto EP, Shields PG, Tyndale RF. The role of CYP2A6 in the emergence of nicotine dependence in adolescents. Pediatrics. 2007;119:e264–274. doi: 10.1542/peds.2006-1583. [DOI] [PubMed] [Google Scholar]
  6. Bagdas D, Muldoon PP, Zhu AZ, Tyndale RF, Damaj MI. Effects of methoxsalen, a CYP2A5/6 inhibitor, on nicotine dependence behaviors in mice. Neuropharmacology. 2014;85:67–72. doi: 10.1016/j.neuropharm.2014.05.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Benowitz NL, Swan GE, Jacob P, 3rd, Lessov-Schlaggar CN, Tyndale RF. CYP2A6 genotype and the metabolism and disposition kinetics of nicotine. Clin Pharmacol Ther. 2006;80:457–467. doi: 10.1016/j.clpt.2006.08.011. [DOI] [PubMed] [Google Scholar]
  8. Brener ND, Billy JO, Grady WR. Assessment of factors affecting the validity of self-reported health-risk behavior among adolescents: evidence from the scientific literature. J Adolesc Health. 2003;33:436–457. doi: 10.1016/s1054-139x(03)00052-1. [DOI] [PubMed] [Google Scholar]
  9. Broms U, Silventoinen K, Madden PA, Heath AC, Kaprio J. Genetic architecture of smoking behavior: a study of Finnish adult twins. Twin Res Hum Genet. 2006;9:64–72. doi: 10.1375/183242706776403046. [DOI] [PubMed] [Google Scholar]
  10. Cannon DS, Medina TR, Mermelstein RJ, Hedeker D, Bakian AV, Coon H, Cook EH, Hamil C, Weiss RB. CYP2A6 Longitudinal Effects in Young Smokers. Nicotine Tob Res. 2015 doi: 10.1093/ntr/ntv049. Epub ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chenoweth MJ, Novalen M, Hawk LW, Jr, Schnoll RA, George TP, Cinciripini PM, Lerman C, Tyndale RF. Known and Novel Sources of Variability in the Nicotine Metabolite Ratio in a Large Sample of Treatment-Seeking Smokers. Cancer Epidemiol Biomarkers Prev. 2014;23:1773–82. doi: 10.1158/1055-9965.EPI-14-0427. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Chenoweth MJ, O’Loughlin J, Sylvestre MP, Tyndale RF. CYP2A6 slow nicotine metabolism is associated with increased quitting by adolescent smokers. Pharmacogenet Genomics. 2013;23:232–235. doi: 10.1097/FPC.0b013e32835f834d. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Clarke P, Sastry N, Duffy D, Ailshire J. Accuracy of self-reported versus measured weight over adolescence and young adulthood: findings from the national longitudinal study of adolescent health, 1996–2008. Am J Epidemiol. 2014;180:153–159. doi: 10.1093/aje/kwu133. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Connor Gorber S, Schofield-Hurwitz S, Hardt J, Levasseur G, Tremblay M. The accuracy of self-reported smoking: a systematic review of the relationship between self-reported and cotinine-assessed smoking status. Nicotine Tob Res. 2009;11:12–24. doi: 10.1093/ntr/ntn010. [DOI] [PubMed] [Google Scholar]
  15. DiFranza JR, Savageau JA, Fletcher K, Ockene JK, Rigotti NA, McNeill AD, Coleman M, Wood C. Recollections and repercussions of the first inhaled cigarette. Addict Behav. 2004;29:261–272. doi: 10.1016/j.addbeh.2003.08.002. [DOI] [PubMed] [Google Scholar]
  16. Fendrich M, Mackesy-Amiti ME, Johnson TP, Hubbell A, Wislar JS. Tobacco-reporting validity in an epidemiological drug-use survey. Addict Behav. 2005;30:175–181. doi: 10.1016/j.addbeh.2004.04.009. [DOI] [PubMed] [Google Scholar]
  17. Garcia KL, Coen K, Miksys S, Le AD, Tyndale RF. Effect of Brain CYP2B Inhibition on Brain Nicotine Levels and Nicotine Self-Administration. Neuropsychopharmacology. 2015;40:1910–1918. doi: 10.1038/npp.2015.40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Gu DF, Hinks LJ, Morton NE, Day IN. The use of long PCR to confirm three common alleles at the CYP2A6 locus and the relationship between genotype and smoking habit. Ann Hum Genet. 2000;64:383–390. doi: 10.1046/j.1469-1809.2000.6450383.x. [DOI] [PubMed] [Google Scholar]
  19. Hill P, Marquardt H. Plasma and urine changes after smoking different brands of cigarettes. Clin Pharmacol Ther. 1980;27:652–658. doi: 10.1038/clpt.1980.92. [DOI] [PubMed] [Google Scholar]
  20. Kalkbrenner AE, Hornung RW, Bernert JT, Hammond SK, Braun JM, Lanphear BP. Determinants of serum cotinine and hair cotinine as biomarkers of childhood secondhand smoke exposure. J Expo Sci Environ Epidemiol. 2010;20:615–624. doi: 10.1038/jes.2010.7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kandel DB, Schaffran C, Griesler PC, Hu MC, Davies M, Benowitz N. Salivary cotinine concentration versus self-reported cigarette smoking: three patterns of inconsistency in adolescence. Nicotine Tob Res. 2006;8:525–537. doi: 10.1080/14622200600672732. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lee AM, Jepson C, Hoffmann E, Epstein L, Hawk LW, Lerman C, Tyndale RF. CYP2B6 genotype alters abstinence rates in a bupropion smoking cessation trial. Biol Psychiatry. 2007a;62:635–641. doi: 10.1016/j.biopsych.2006.10.005. [DOI] [PubMed] [Google Scholar]
  23. Lee AM, Jepson C, Shields PG, Benowitz N, Lerman C, Tyndale RF. CYP2B6 genotype does not alter nicotine metabolism, plasma levels, or abstinence with nicotine replacement therapy. Cancer Epidemiol Biomarkers Prev. 2007b;16:1312–1314. doi: 10.1158/1055-9965.EPI-07-0188. [DOI] [PubMed] [Google Scholar]
  24. Lerman C, Tyndale R, Patterson F, Wileyto EP, Shields PG, Pinto A, Benowitz N. Nicotine metabolite ratio predicts efficacy of transdermal nicotine for smoking cessation. Clin Pharmacol Ther. 2006;79:600–608. doi: 10.1016/j.clpt.2006.02.006. [DOI] [PubMed] [Google Scholar]
  25. Malaiyandi V, Lerman C, Benowitz NL, Jepson C, Patterson F, Tyndale RF. Impact of CYP2A6 genotype on pretreatment smoking behaviour and nicotine levels from and usage of nicotine replacement therapy. Mol Psychiatry. 2006;11:400–409. doi: 10.1038/sj.mp.4001794. [DOI] [PubMed] [Google Scholar]
  26. Morral AR, McCaffrey D, Iguchi MY. Hardcore drug users claim to be occasional users: drug use frequency underreporting. Drug Alcohol Depend. 2000;57:193–202. doi: 10.1016/s0376-8716(99)00048-4. [DOI] [PubMed] [Google Scholar]
  27. Mwenifumbo JC, Myers MG, Wall TL, Lin SK, Sellers EM, Tyndale RF. Ethnic variation in CYP2A6*7, CYP2A6*8 and CYP2A6*10 as assessed with a novel haplotyping method. Pharmacogenet Genomics. 2005;15:189–192. doi: 10.1097/01213011-200503000-00008. [DOI] [PubMed] [Google Scholar]
  28. Nakajima M, Yamamoto T, Nunoya K, Yokoi T, Nagashima K, Inoue K, Funae Y, Shimada N, Kamataki T, Kuroiwa Y. Role of human cytochrome P4502A6 in C-oxidation of nicotine. Drug Metab Dispos. 1996;24:1212–1217. [PubMed] [Google Scholar]
  29. O’Loughlin J, Dugas EN, Brunet J, DiFranza J, Engert JC, Gervais A, Gray-Donald K, Karp I, Low NC, Sabiston C, Sylvestre MP, Tyndale RF, Auger N, Mathieu B, Tracie B, Chaiton M, Chenoweth MJ, Constantin E, Contreras G, Kakinami L, Labbe A, Maximova K, McMillan E, O’Loughlin EK, Pabayo R, Roy-Gagnon MH, Tremblay M, Wellman RJ, Hulst A, Paradis G. Cohort Profile: The Nicotine Dependence in Teens (NDIT) Study. Int J Epidemiol. 2014a doi: 10.1093/ije/dyu135. Epub ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. O’Loughlin J, Paradis G, Kim W, DiFranza J, Meshefedjian G, McMillan-Davey E, Wong S, Hanley J, Tyndale RF. Genetically decreased CYP2A6 and the risk of tobacco dependence: a prospective study of novice smokers. Tob Control. 2004;13:422–428. doi: 10.1136/tc.2003.007070. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. O’Loughlin JL, Dugas EN, O’Loughlin EK, Karp I, Sylvestre MP. Incidence and determinants of cigarette smoking initiation in young adults. J Adolesc Health. 2014b;54:26–32. e24. doi: 10.1016/j.jadohealth.2013.07.009. [DOI] [PubMed] [Google Scholar]
  32. O’Loughlin JL, Sylvestre MP, Dugas EN, Karp I. Predictors of the occurrence of smoking discontinuation in novice adolescent smokers. Cancer Epidemiol Biomarkers Prev. 2014c;23:1090–1101. doi: 10.1158/1055-9965.EPI-13-0869. [DOI] [PubMed] [Google Scholar]
  33. Riccardi LN, Carano F, Bini C, Ceccardi S, Ferri G, Pelotti S. CYP2B6 gene single-nucleotide polymorphisms in an Italian population sample and relationship with nicotine dependence. Genet Test Mol Biomarkers. 2015;19:103–107. doi: 10.1089/gtmb.2014.0239. [DOI] [PubMed] [Google Scholar]
  34. Rigotti NA, Lee JE, Wechsler H. US college students’ use of tobacco products: results of a national survey. JAMA. 2000;284:699–705. doi: 10.1001/jama.284.6.699. [DOI] [PubMed] [Google Scholar]
  35. Rotger M, Tegude H, Colombo S, Cavassini M, Furrer H, Decosterd L, Blievernicht J, Saussele T, Gunthard HF, Schwab M, Eichelbaum M, Telenti A, Zanger UM. Predictive value of known and novel alleles of CYP2B6 for efavirenz plasma concentrations in HIV-infected individuals. Clin Pharmacol Ther. 2007;81:557–566. doi: 10.1038/sj.clpt.6100072. [DOI] [PubMed] [Google Scholar]
  36. Rubinstein ML, Shiffman S, Moscicki AB, Rait MA, Sen S, Benowitz NL. Nicotine metabolism and addiction among adolescent smokers. Addiction. 2013;108:406–412. doi: 10.1111/j.1360-0443.2012.04026.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Rubinstein ML, Thompson PJ, Benowitz NL, Shiffman S, Moscicki AB. Cotinine levels in relation to smoking behavior and addiction in young adolescent smokers. Nicotine Tob Res. 2007;9:129–135. doi: 10.1080/14622200601078517. [DOI] [PubMed] [Google Scholar]
  38. Schnoll RA, George TP, Hawk L, Cinciripini P, Wileyto P, Tyndale RF. The relationship between the nicotine metabolite ratio and three self-report measures of nicotine dependence across sex and race. Psychopharmacology (Berl) 2014;231:2515–2523. doi: 10.1007/s00213-013-3421-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Schnoll RA, Patterson F, Wileyto EP, Tyndale RF, Benowitz N, Lerman C. Nicotine metabolic rate predicts successful smoking cessation with transdermal nicotine: a validation study. Pharmacol Biochem Behav. 2009;92:6–11. doi: 10.1016/j.pbb.2008.10.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Schoedel KA, Hoffmann EB, Rao Y, Sellers EM, Tyndale RF. Ethnic variation in CYP2A6 and association of genetically slow nicotine metabolism and smoking in adult Caucasians. Pharmacogenetics. 2004;14:615–626. doi: 10.1097/00008571-200409000-00006. [DOI] [PubMed] [Google Scholar]
  41. Sofuoglu M, Herman AI, Nadim H, Jatlow P. Rapid nicotine clearance is associated with greater reward and heart rate increases from intravenous nicotine. Neuropsychopharmacology. 2012;37:1509–1516. doi: 10.1038/npp.2011.336. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. St Helen G, Novalen M, Heitjan DF, Dempsey D, Jacob P, 3rd, Aziziyeh A, Wing VC, George TP, Tyndale RF, Benowitz NL. Reproducibility of the nicotine metabolite ratio in cigarette smokers. Cancer Epidemiol Biomarkers Prev. 2012;21:1105–1114. doi: 10.1158/1055-9965.EPI-12-0236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Tang DW, Hello B, Mroziewicz M, Fellows LK, Tyndale RF, Dagher A. Genetic variation in CYP2A6 predicts neural reactivity to smoking cues as measured using fMRI. Neuroimage. 2012;60:2136–2143. doi: 10.1016/j.neuroimage.2012.01.119. [DOI] [PubMed] [Google Scholar]
  44. Tanner JA, Novalen M, Jatlow P, Huestis MA, Murphy SE, Kaprio J, Kankaanpaa A, Galanti L, Stefan C, George TP, Benowitz NL, Lerman C, Tyndale RF. Nicotine Metabolite Ratio (3-hydroxycotinine/cotinine) in Plasma and Urine by Different Analytical Methods and Laboratories: Implications for Clinical Implementation. Cancer Epidemiol Biomarkers Prev. 2015;24:1239–1246. doi: 10.1158/1055-9965.EPI-14-1381. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Thorn CF, Lamba JK, Lamba V, Klein TE, Altman RB. PharmGKB summary: very important pharmacogene information for CYP2B6. Pharmacogenet Genomics. 2010;20:520–523. doi: 10.1097/fpc.0b013e32833947c2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Tzschentke TM. Measuring reward with the conditioned place preference (CPP) paradigm: update of the last decade. Addict Biol. 2007;12:227–462. doi: 10.1111/j.1369-1600.2007.00070.x. [DOI] [PubMed] [Google Scholar]
  47. U.S. Department of Health and Human Services. Preventing tobacco use among youth and young adults: a report of the Surgeon General. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2012. [PubMed] [Google Scholar]
  48. Vink JM, Willemsen G, Boomsma DI. Heritability of smoking initiation and nicotine dependence. Behav Genet. 2005;35:397–406. doi: 10.1007/s10519-004-1327-8. [DOI] [PubMed] [Google Scholar]
  49. Wassenaar CA, Dong Q, Wei Q, Amos CI, Spitz MR, Tyndale RF. Relationship between CYP2A6 and CHRNA5-CHRNA3-CHRNB4 variation and smoking behaviors and lung cancer risk. J Natl Cancer Inst. 2011;103:1342–1346. doi: 10.1093/jnci/djr237. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES