Skip to main content
NCBI home page
Nicotine & Tobacco Research logoLink to Nicotine & Tobacco Research
. 2022 Feb 16;24(10):1573–1580. doi: 10.1093/ntr/ntac043

The Promise of Polygenic Risk Prediction in Smoking Cessation: Evidence From Two Treatment Trials

Michael Bray 1,2,2, Yoonhoo Chang 3,2, Timothy B Baker 4, Douglas Jorenby 5, Robert M Carney 6, Louis Fox 7, Giang Pham 8, Faith Stoneking 9, Nina Smock 10,11, Christopher I Amos 12,13, Laura Bierut 14,15, Li-Shiun Chen 16,17,✉
PMCID: PMC9575976  PMID: 35170738

Abstract

Introduction

Tobacco use disorder is a complex behavior with a strong genetic component. Genome-wide association studies (GWAS) on smoking behaviors allow for the creation of polygenic risk scores (PRSs) to approximate genetic vulnerability. However, the utility of smoking-related PRSs in predicting smoking cessation in clinical trials remains unknown.

Aims and Methods

We evaluated the association between polygenic risk scores and bioverified smoking abstinence in a meta-analysis of two randomized, placebo-controlled smoking cessation trials. PRSs of smoking behaviors were created using the GWAS and Sequencing Consortium of Alcohol and Nicotine use (GSCAN) consortium summary statistics. We evaluated the utility of using individual PRS of specific smoking behavior versus a combined genetic risk that combines PRS of all four smoking behaviors. Study participants came from the Transdisciplinary Tobacco Use Research Centers (TTURCs) Study (1091 smokers of European descent), and the Genetically Informed Smoking Cessation Trial (GISC) Study (501 smokers of European descent).

Results

PRS of later age of smoking initiation (OR [95% CI]: 1.20, [1.04–1.37], p = .0097) was significantly associated with bioverified smoking abstinence at end of treatment. In addition, the combined PRS of smoking behaviors also significantly predicted bioverified smoking abstinence (OR [95% CI] 0.71 [0.51–0.99], p = .045).

Conclusions

PRS of later age at smoking initiation may be useful in predicting smoking cessation at the end of treatment. A combined PRS may be a useful predictor for smoking abstinence by capturing the genetic propensity for multiple smoking behaviors.

Implications

There is a potential for polygenic risk scores to inform future clinical medicine, and a great need for evidence on whether these scores predict clinically meaningful outcomes. Our meta-analysis provides early evidence for potential utility of using polygenic risk scores to predict smoking cessation amongst smokers undergoing quit attempts, informing further work to optimize the use of polygenic risk scores in clinical care.

Introduction

Tobacco use disorder is the leading cause of preventable death worldwide.1 In the United States, cigarette smoking contributes to approximately one in five deaths annually, which results in a 10-year reduced life expectancy.2 However, successful smoking cessation can improve long-term health outcomes. For example, individuals who quit smoking before the age of 40 reduce the risk of a smoking-related death by ~90%.2,3 Unfortunately, cigarette smoking is highly addictive and although 68% of adult smokers desire to quit, less than 10% of adult smokers successfully quit annually.4

The ability to successfully quit smoking is heritable, with heritability estimates from twin studies explaining up to 54% of variance in smoking cessation outcomes.5 However, the genetic influences of smoking cessation are complex.6 Large genome-wide association studies (GWAS) of smoking cessation and other smoking-related behaviors highlighted two well-known and documented genetic loci. The first genetic locus is CHRNA5 on chromosomal region 15q25 and is a nicotinic receptor.7 Within CHRNA5, one SNP, rs16969968, in particular drives this association.7–10 The rs16969968 risk allele (A) is associated with a lower likelihood of smoking cessation, and the high-risk genotype (AA) is associated with a four-year earlier median age of lung-cancer diagnosis.11–13 The second genetic loci is CYP2A6 on chromosomal region 19q136,7 and is the primary nicotine metabolizing gene.14CYP2A6 is highly polymorphic, and genetic variation within CYP2A6 is associated with changes with nicotine metabolism,15 which can affect rates of smoking cessation.16,17

Recent well-powered genome-wide association studies (GWAS) have identified many genetic variants associated with various smoking behaviors.7,18,19 Specifically, the GWAS and Sequencing Consortium of Alcohol and Nicotine use (GSCAN) discovered over 500 genetic variants associated with four key smoking behaviors: ever smoking, later age of smoking initiation, cigarettes smoked per day, and persistent smoking (failed smoking cessation), allowing for the opportunity to develop polygenic risk scores (PRSs).7 A PRS is made by combining the effects of many genetic signals from a GWAS into a single risk variable, which can be used to estimate an individual’s genetic propensity to develop a disease or trait. PRSs have been useful in predicting health outcomes including lung cancer, other cancers, and psychiatric disorders.20–24

Existing research has demonstrated the utility of PRS in predicting smoking behaviors, indicating the potential use of PRSs in clinical care.25,26 For example, Belsky and colleagues (2013)26 observed that individuals with high smoking-related polygenic risk scores were more likely to develop nicotine dependence if they smoked and were less likely to quit smoking.26 Other examples evaluate how PRS of nicotine metabolism markers can be used to predict smoking cessation.27,28 For example, research suggests the potential of using PRSs in predicting nicotine metabolism that could potentially inform treatment response. Although research is emerging with the use of PRSs in the prediction of smoking behaviors and health outcomes, there is still a gap of knowledge in whether PRSs of smoking behaviors derived from these large population studies predict clinical smoking cessation success among smokers making a quit attempt.

Previous research has shown that variation in individual genes such as CHRNA58,11,12 and CYP2A614,15 can predict smoking cessation in both population studies and clinical trials. For example, evidence from University of Wisconsin Transdisciplinary Tobacco Use Research Center (UW-TTURC) trial suggested that CHRNA5 genotypes may moderate the responses to nicotine replacement12 in individuals of European ancestry. More recent evidence from the Genetically Informed Smoking Cessation Trial (GISC) trial suggested that CHRNA5 genotypes may moderate the response to nicotine replacement and varenicline in individuals of African American ancestry.29 However, the results of such individual gene prediction studies have not always been consistent.30–32 Recent GWAS have identified PRSs that predict smoking behaviors in large cross-sectional population studies. To the extent that PRSs comprise multiple genetic variants which may enhance the prediction of cessation outcomes, their use should result in more accurate and consistent associations with such outcomes. We hypothesize that smokers with higher PRSs of problematic smoking behaviors (eg earlier onset, heavy smoking, or persistent smoking) are less likely to achieve smoking cessation in clinical trials. This hypothesis is based on the notion that characteristics, such as age of initiating regular smoking and smoking heaviness, have shown strong associations with dependence and smoking cessation failure in previous research.33–35 Using meta-analyses of two randomized placebo-controlled smoking cessation trials, we examine the utility of PRSs of four key smoking behaviors (ever smoking, later age of smoking initiation, cigarettes per day, and persistent smoking) in predicting smoking cessation among smokers attempting to quit.2,7

Methods

Study Samples

The Transdisciplinary Tobacco Use Research Centers (TTURCs) Study

The TTURC study is a randomized, placebo-controlled smoking cessation clinical trial at the University of Wisconsin Center for Tobacco Research and Intervention focusing on the genetic association of time to relapse after quitting.13 Each participant was randomly assigned to one of the six conditions: (1) placebo, (2) nicotine patch, (3) nicotine lozenge, (4) sustained-release bupropion, (5) nicotine patch and nicotine lozenge (combination nicotine replacement [C-NRT]), or (6) bupropion and nicotine lozenge. In addition, all participants received individual cessation counseling. For this study, 1091 smokers of European ancestry were included.

At the end of treatment (8 weeks), bioverified smoking abstinence was verified by expired-carbon monoxide level of less than 10 ppm, documenting abstinence at post treatment.

The Genetically Informed Smoking Cessation Trial (GISC) Study

The GISC is a prospective, randomized, placebo-controlled smoking cessation trial conducted at Washington University in St Louis.29 Each participant was randomly assigned into one of the three groups stratified by genotypes of rs16969968: nicotine patch and nicotine lozenge (C-NRT), varenicline tartrate, or placebo. All participants received cessation counseling. For this study, 501 smokers of European ancestry were included.

The primary outcome is 7-day point prevalence bioverified smoking abstinence at end of treatment (week 12). All participants self-reported smoking status for the primary endpoint, verified by an expired carbon monoxide level of less than 8 ppm.

Ethical Review Board

Both studies were approved by the appropriate institutional review boards and all participants provided informed consent.

Genetic Data

Genotyping

TTURC participants were genotyped at the Center for Inherited Disease Research at Johns Hopkins University using the Illumina Omni2.5 microarray. Gene-Environment Association Studies (GENEVA) Coordinating Center at the University of Washington led the data cleaning process. GISC participants were genotyped using Illumina Global Microarray.

Imputation

Using PLINK software,36 standard GWAS QC was performed to TTURC and GISC datasets. Single nucleotide polymorphisms (SNPs) were aligned to the 1000 Genomes Reference (+strand, build 37) and imputed on the University of Michigan Imputation server using 1000 Genome Reference (build 37, phase 5).9,37 Details are in Supplementary Text 1. Imputed SNPs that had an info score ≥0.9 and a minor allele frequency ≥1% were converted to hard calls. After imputation and QC, there were 47 109 470 SNPs in UW-TTURC and 47 109 465 SNPs in GISC.

Principal components (PCs) of genetic ancestry for GISC were created using EIGENTRAT.38 Information on how PCs were generated for TTURC can be found in the database of Genotypes and Phenotypes (dbGAP) repository (https://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?study_id=phs000404.v1.p1) under accession number (phs000404.v1.p1).

Polygenic Risk Scores

We generated PRSs of four smoking behavior phenotypes (ever smoking, later age of smoking initiation, cigarettes smoked per day, and persistent smoking [failed smoking cessation]) from GSCAN GWAS summary statistics using PRSice software.39 SNPs in each data set were pruned by PRSice (version 1.23) using p-value–informed linkage disequilibrium clumping: R2 <0.10 in a 500-kb window, collapsed to the most significant variant. For each of the four smoking-related summary statistics, we tested the predictability of PRSs that were generated from eight p-value thresholds (.5, .05, .005, .0005, 5 × 10−5, 5 × 10−6, 5 × 10−7, 5 × 10−8). In general, the inclusion of more SNPs led to more variance explained of the outcome and thus, all subsequent analyses involved PRSs generated using SNPs with p-value threshold of .5 or greater unless otherwise noted. For the GISC data set, there was a total of 136 874 SNPs for generating the later age of smoking initiation PRS, 136 689 SNPs for the cigarettes per day PRS, 135 124 SNPs for the ever smoking PRS, and 135 609 SNPs for the persistent smoking PRS. For the TTURC data set, there was a total of 183 998 SNPs for generating the later age of smoking initiation PRS, 183 342 SNPs for the cigarettes per day PRS, 181 341 SNPs for the ever smoking PRS, and 181 362 SNPs for the persistent smoking PRS. To ensure the interpretability across PRSs of different traits, all PRSs were standardized to Z-scores.

Combined Polygenic Risk Scores

In addition, we generated a combined PRS for risk smoking behaviors based on PRSs of four smoking behaviors (ever smoking, earlier initiation, higher cigarettes per day, persistent smoking) by taking the mean of all four z-transformed PRSs. We acknowledge the mean is an agnostic combination of these PRSs and have shown the correlation of these PRSs in the samples (Table S1).

Statistical Analysis

We modeled the association of each PRS to predict the outcome of bioverified smoking abstinence at end of treatment in both trials. Z-transformed PRSs were modeled as continuous variables. We examined the associations between PRSs and bioverified smoking abstinence at the end of the treatment in both trials using logistic regression models in R.40 We included the following covariates: age, sex, PC1, and PC2. Additional covariates included cigarettes smoked per day (CPD) and treatment by FDA-approved medication (GISC: combination nicotine replacement therapy (nicotine patch, nicotine lozenge), varenicline; TTURC: combination nicotine replacement therapy, bupropion). PRSs were converted to quartiles for the ease of interpretation. In addition, meta-analysis of both trials was performed in R (metafor package).40 We reported both fixed and random effects models. In addition, we generated receiver operating characteristic curve (ROC) and estimated the area under the curve (AUC) for the prediction of smoking cessation using clinical predictors and genetic predictors (SAS 9.4).

Results

Sample Characteristics

We examined whether polygenic risk scores of smoking behaviors were associated with bioverified smoking abstinence in both smoking cessation trials, TTURC and GISC. The sample characteristics for these two trials are shown in Table 1. The samples from both studies are of European descent. The mean ages were 46.6 (GISC) and 44.5 (TTURC). The ratio of males and females was similar between the two trials (55.9% female for GISC and 58.1% female for TTURC). The mean baseline CPD was 19.1 for GISC and 21.8 for TTURC. Bioverified smoking abstinence at the end of treatment was 20.6% for GISC (N = 103) and 47.5% for TTURC (N = 518) (Table 1).

Table 1.

Descriptive Statistics for the Samples from GISC and TTURC Studies Used for the Analysis

GISC TTURC
Total N 501 1091
Age (mean, SE) 46.6, 0.51 44.5, 0.34
Sex (n, %)
Male 221, 44.1 457, 41.9
Female 280, 55.9 634, 58.1
Baseline CPD (mean, SE) 19.1, 0.34 21.8, 0.28
Randomized to active pharmacotherapy (n, %) 329, 65.7 955, 87.5
History of anxiety/depression (n, %) 129, 25.7 226, 20.7
CO (mean, SE) 28.9, 0.59 26.5, 0.38
FTND (mean, SE) 4.9, 0.1 5.3, 0.07
Smoking age of initiation (mean, SE) 17.5, 0.19 17.3, 0.12
Smoking abstinence at EOT (n, %) 103, 20.6 518, 47.5
Open in a new tab

CPD = cigarettes per day; CO = carbon monoxide level; EOT = end of treatment; FTND = Fagerstrom Test for Nicotine Dependence; GISC = The Genetically Informed Smoking Cessation Trial Study; SE = standard error; TTURC2 = The Transdisciplinary Tobacco Use Research Centers Study.

Please note that TTURC is missing 2 CO, 14 FTND, 2 Smoking Age of initiation, 36 CPD values.

PRS of Specific Smoking Behaviors and Bioverified Smoking Abstinence

We evaluated the association between each PRS (ever smoking, later age of smoking initiation, cigarettes per day, and persistent smoking) and end of treatment bioverified smoking abstinence using meta-analyses from the results of the two trials. In random effect meta-analysis models (Figure 1), the PRS of later age of smoking initiation was significantly associated with bioverified smoking abstinence at end of treatment (OR [95% CI]: 1.20, [1.04–1.37], p = .0097). Specifically, smokers with PRS-later age of smoking initiation in the highest quartile compared with those with the lowest quartile were more likely to quit successfully (45.1% vs. 32.8%, Figure 1, a). The PRSs of the other risk smoking behaviors were associated with reduced abstinence but did not reach statistical significance.

Figure 1.

Figure 1.

Open in a new tab

Polygenetic risk scores (PRSs) of later age of smoking initiation, persistent smoking, ever smoking, and cigarettes per day and bioverified end of treatment smoking abstinence: Meta-analysis of two treatment trials (Random Effect Models). Meta-analysis random effect models a) Pheterogeneity = 0.24, b) Pheterogeneity = 0.18, c) Pheterogeneity = 0.015, and d) Pheterogeneity = 0.43. Logistic regression model for smoking abstinence with age, sex, PCs, study, and PRS. Q1 indicates the lowest quartile, while Q4 indicates the highest.

We also evaluated the fixed effect meta-analysis models that involved an assumption of a common effect size in these 2 trials in Supplementary Figure S1. The PRS of persistent smoking was significantly associated with bioverified smoking abstinence at end of treatment (OR [95% CI]: 0.89, [0.80–0.99], p = .033). Smokers in the highest quartile of PRS-persistent smoking compared with those in the lowest quartile were less likely to quit successfully (37.1% vs. 41.3% achieving bioverified smoking abstinence, Supplementary Figure S1, d).

Combined PRS of Smoking Behaviors and Bioverified Smoking Abstinence

We evaluated and meta-analyzed the association of the combined PRS of smoking behaviors and bioverified smoking abstinence. These four PRSs had low correlations (all ≤ 0.16) (Supplementary Table S1). Using an agnostic approach, we computed the combined PRS of smoking behaviors using the mean of these four PRSs. This combined PRS of smoking behavior was significantly associated with bioverified smoking abstinence (random effect model OR [95% CI]: 0.71 [0.51–0.99], p = .045; fixed effect OR [95% CI]: 0.67 [0.56–0.82], p = 8.2 × 10−5 (Figure 2 and Supplementary Figure S2). Smokers with the highest quartile compared with those in the lowest quartile of the combined PRS were less likely to quit smoking (33.0% vs. 48.5%, Figure 2).

Figure 2.

Figure 2.

Open in a new tab

Polygenetic risk scores (PRSs) of ever smoking and cigarettes per day and bioverified end of treatment smoking abstinence: Meta-analysis of two treatment trials (Random Effect Models). Meta-analysis random effect models a) Pheterogeneity = 0.015 and b) Pheterogeneity = 0.43. Logistic regression model for smoking abstinence with age, sex, PCs, study, and PRS. Q1 indicates the lowest quartile, while Q4 indicates the highest.

Prediction of Bioverified Smoking Abstinence

We evaluated whether the addition of basic clinical predictors (cigarettes per day and treatment), additional clinical predictors (baseline Fagerstrom Test for Nicotine Dependence [FTND], baseline carbon monoxide [CO], age of smoking initiation, history of depression/anxiety, and the addition of genetic predictors (PRSs of smoking behaviors) beyond use of the demographic predictors increases the prediction of bioverified smoking abstinence at end of treatment (Supplementary Figure S3). In evaluating the utility of adding genetic predictors to clinical predictors, we have compared model prediction with the area under the curve (AUC). We found that adding basic clinical, additional clinical, and genetic predictors significantly increased the AUC (basic clinical predictors—0.64 to 0.69, p < .0001; additional clinical predictors—0.69 to 0.70, p = .019; genetic predictors—0.70 to 0.71, p = .034).

Discussion

Our findings provide novel evidence for the utility of PRSs derived from smoking behaviors to predict smoking cessationin clinical trials. Meta-analysis of two trials revealed associations between PRSs and successful smoking cessation at the end of treatment. Specifically, the PRS of later age of initiation predicts an increase in successful smoking cessation, and the PRS of persistent smoking may predict a decrease in achieving successful smoking cessation. In addition, a combined PRS, which summarizes the PRSs of multiple aspects of risky smoking behaviors, could be useful in predicting smoking cessation among smokers making a quit attempt.

Most genetically informed research of smoking-related outcomes focus on single gene regions encompassing genes CHRNA58,11,12 and CYP2A6.14,15,17 For example, meta-analyses show strong evidence of the associations between CHRNA5 on chromosome 15q25 with cigarettes smoked per day, smoking cessation, and lung cancer.8 In addition, several studies have focused on the CYP2A6 region encompassing chromosome 19q13.6,14CYP2A6 encodes the enzyme that is the primary metabolizer of nicotine14 and is, therefore, a proxy for the nicotine metabolite ratio (NMR), a biomarker of nicotine metabolism. There is a substantial evidence that CYP2A6 is associated with smoking cessation. For instance, slow metabolism associated with this gene is associated with increased smoking cessation during adolescence.41 In addition, NMR has been implicated in smoking cessation in a previous study by Lerman and colleagues (2015).42 These authors observed that amongst individuals taking varenicline, normal nicotine metabolizers were more likely to quit smoking than slow metabolizers.42 Thus, there is evidence that both CHRNA5 and CYP2A6 provide information regarding the risk of smoking cessation success. Our findings, however, present evidence that PRSs based on many variants across the genome that are derived from crude smoking behaviors such as smoking initiation, age of smoking initiation, cigarettes per day, and persistent smoking hold the potential for predicting smoking cessation success. This suggests that risk for smoking cessation failure is highly heterogeneous and can also be assessed using broad polygenic approaches. Such heterogeneity is consistent with multifactorial assessments of cigarette dependence33 and with the evidence of heterogeneity from motivational mechanisms linked with persistent smoking.42–49

Our evidence suggests that a combined PRS, which captures the effects of different smoking behaviors, could be a more useful predictor than a PRS derived from a single trait. Another study has combined PRSs of lung imaging phenotypes and patterns of lung growth to test the power of a combined PRS to predict chronic obstructive pulmonary disease (COPD). This study has shown that the combined PRS was significantly associated with COPD.50 A combined PRS, or composite PRS, can capture genetic risk from multiple risk behaviors. In addition, a combined PRS may create a biologically relevant score for clinical trials and improve the relevance of genetic risk scores. There is emerging research on methods to combine individual PRSs to improve predictive power. Our study took an agnostic approach in combining individual PRSs based on the observed low correlation among the individual PRSs. Even though two smoking PRSs were not statistically significant, the effect sizes were in the predicted direction, suggesting the potential utility of these PRSs when the sample size increases. Future work to improve this methodology for combining individual PRSs has the potential to revolutionize how PRS are utilized.

Our study is based on two smoking cessation datasets investigating PRSs in clinical trials. Even so, some limitations exist. First, the power of this study was limited due to a modest sample size of 1592 individuals enrolled in trials. To counteract the sample size, we have analyzed two different datasets to reveal convergent results. For future studies, more replications are needed to confirm the associations between smoking behavior PRSs and smoking cessation. Future studies with larger sample sizes would be required for evaluating the interactive effects of polygenic risk scores and specific medications. Second, we are limiting our research to smokers with European ancestry. This is largely because the GWAS and Sequencing Consortium of Alcohol and Nicotine Use (GSCAN) summary statistics is based on 1.2 million smokers of European descent. Several studies prove that the risk scores derived from European population underperform in non-European populations.7 Third, even though 1KG reference panel is proven to provide valuable genomic resources that can augment the power of GWAS in groups with European ancestry,51 we acknowledge the limitation of using 1KG reference panel instead of more current reference panels (HRC or TopMed) for imputation. Finally, the participants in two studies received different medications for treatment, but we currently do not have enough power to evaluate how PRSs moderate response to specific medications. Furthermore, we observed variation in smoking cessation outcomes across the two trials due to heterogeneity in study design, outcome definitions, and time of the trial. Future studies involving increased number of trials and samples are needed to address this important question.

Ultimately, the genetic studies on smoking behaviors aim to enhance the prevention of smoking and aide in successful smoking cessation. The use of PRSs in clinical trials may be a helpful tool by incorporating multiple genetic signals to assess the inherent ability to quit smoking amongst smokers. This study is motivated by available GWAS results of smoking behaviors based on large general population studies, while GWAS for refined clinical outcomes such as smoking cessation in smoking cessation trials are not available due to limited sample sizes. Future research can explore different strategies for developing PRSs, including forming them on the basis of phenotypes that critically affect smoking cessation success such as withdrawal, craving, severity, and reward sensitivity. In addition, methods such as machine learning might suggest component weightings to use in future applications. Another future direction is using weighted genetic scores to approximate a biomarker or a mechanistic pathway. In a previous study by Buchwald and colleagues (2020), the authors observed that the CYP2A6 gene region explained up to ~36% of genetic variance of the nicotine metabolite ratio.50 Existing research, including our prior work, showed the potential utility of PRS in predicting nicotine metabolism.17,27 It is probable that if a selected gene region could explain a substantial proportion of genetic variance of the nicotine metabolite ratio, then a regional PRS would be predictive of smoking cessation at the end of treatment as well. Furthermore, PRSs should permit the study of interactions between non-biological factors and treatments. For example, there is a strong association between CHRNA5 nicotine receptor gene variant and partner smoking, which is largely an environmental factor, but also represents the genetic factors of partner selection.12 The genetic risk factor coupled with the environmental risk factor can result in a low rate of smoking reduction; PRSs should permit the exploration of other environmental factors that modulate smoking cessation success.

This study presents evidence on the potential utility of smoking behavior PRSs in predicting smoking cessation success amongst smokers in treatment trials. Our evidence based on the meta-analysis of two trials suggest that the PRS of later age of smoking initiation and the PRS of persistent smoking may be useful in predicting smoking cessation at the end of treatment. A combined PRS may be a useful predictor of smoking cessation by capturing the genetic propensity for multiple smoking behaviors. These findings help answer important clinical questions about the utility of polygenic risk scores in clinical smoking cessation outcomes.

Supplementary Material

A Contributorship Form detailing each author’s specific involvement with this content, as well as any supplementary data, are available online at https://academic.oup.com/ntr.

ntac043_suppl_Supplementary_Data
Click here for additional data file. (13.6KB, docx)
ntac043_suppl_Supplementary_Figures
Click here for additional data file. (323.8KB, pptx)
ntac043_suppl_Supplementary_Table_S1
Click here for additional data file. (14.6KB, docx)
ntac043_suppl_Supplementary_Taxonomy-form
Click here for additional data file. (728.5KB, pdf)

Contributor Information

Michael Bray, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO, USA; Department of Genetic Counseling, Bay Path University, Longmeadow, MA, USA.

Yoonhoo Chang, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO, USA.

Timothy B Baker, Department of Medicine, School of Medicine and Public Health, Center for Tobacco Research and Intervention, University of Wisconsin, Madison, WI, USA.

Douglas Jorenby, Department of Medicine, School of Medicine and Public Health, Center for Tobacco Research and Intervention, University of Wisconsin, Madison, WI, USA.

Robert M Carney, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO, USA.

Louis Fox, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO, USA.

Giang Pham, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO, USA.

Faith Stoneking, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO, USA.

Nina Smock, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO, USA; The Alvin J. Siteman Cancer Center, Washington University School of Medicine, St. Louis, MO, USA .

Christopher I Amos, Department of Medicine, Baylor College of Medicine, Institute for Clinical and Translational Research, Houston, TX, USA; Department of Biomedical Data Science, Geisel School of Medicine, Dartmouth College, Hanover, NH, USA.

Laura Bierut, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO, USA; The Alvin J. Siteman Cancer Center, Washington University School of Medicine, St. Louis, MO, USA .

Li-Shiun Chen, Department of Psychiatry, Washington University School of Medicine, St. Louis, MO, USA; The Alvin J. Siteman Cancer Center, Washington University School of Medicine, St. Louis, MO, USA .

Funding

M.B. was supported by the National Institutes of Health Training Grant 5T32MH014677. L.-S.C. was supported by the National Institute on Drug Abuse grant R01 DA038076, Siteman Cancer Center and NCI Cancer Center Support Grant P30 CA091842. T.B.B.’s involvement was supported in part by R01 HL109031. C.I.A. is a research scholar of the Cancer Prevention Research Institute of Texas and partially supported by RR170048. L.-S.C., L.J.B. and C.I.A. are partially supported by U19CA203654. L.J.B. was supported by National Center for Advancing Translation Sciences grant UL1TR002345 and by National Institute on Aging grant R56AG058726. The content is solely the responsibility of the authors and does not necessarily represent the official view of the National Institutes of Health.

Declaration of Interests

R.M.C. or a member of his family owns stock in Pfizer Inc. L.J.B. is listed as an inventor on Issued US Patent 8,080,371 “Markers for Addiction” covering the use of certain single nucleotide polymorphisms in determining the diagnosis, prognosis, and treatment of addiction, and served as a consultant for the pharmaceutical company Pfizer Inc. (New York City, NY) in 2008. All other authors declared no competing interests for this work.

Data Availability

Data is available from dbGaP and the NIDA Center for Genetic Studies: https://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?study_id=phs000404.v1.p1

https://nidagenetics.org/studies/study-35-genetically-informative-smoking-cessation-trial.

References

  • 1. World Health Organization. Tobacco; 2020. [cited 2020 8 October]. Available from: https://www.who.int/news-room/fact-sheets/detail/tobacco.
  • 2. CDC. Tobacco-Related Mortality: Centers for Disease Control and Prevention; 2018. [updated January 17, 2018. Available from: https://www.cdc.gov/tobacco/data_statistics/fact_sheets/health_effects/tobacco_related_mortality/index.htm.
  • 3. Jha P, Ramasundarahettige C, Landsman V, et al. 21st-Century hazards of smoking and benefits of cessation in the United States. N Engl J Med. 2013;368(4):341–350. [DOI] [PubMed] [Google Scholar]
  • 4. CDC. Smoking & Tobacco Use; 2020. Available from: https://www.cdc.gov/tobacco/data_statistics/fact_sheets/cessation/smoking-cessation-fast-facts/index.html.
  • 5. Xian H, Scherrer JF, Madden PA, et al. The heritability of failed smoking cessation and nicotine withdrawal in twins who smoked and attempted to quit. Nicotine Tob Res. 2003;5(2):245–254. [PubMed] [Google Scholar]
  • 6. Hancock DB, Markunas CA, Bierut LJ, Johnson EO. Human genetics of addiction: new insights and future directions. Curr Psychiatry Rep. 2018;20(2):8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Liu M, Jiang Y, Wedow R, et al. Association studies of up to 1.2 million individuals yield new insights into the genetic etiology of tobacco and alcohol use. Nat Genet. 2019;51(2):237–244. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Improgo MR, Scofield MD, Tapper AR, Gardner PD. The nicotinic acetylcholine receptor CHRNA5/A3/B4 gene cluster: Dual role in nicotine addiction and lung cancer. Prog Neurobiol. 2010;92(2):212–226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Bray MJ, Chen LS, Fox L, et al. Dissecting the genetic overlap of smoking behaviors, lung cancer, and chronic obstructive pulmonary disease: a focus on nicotinic receptors and nicotine metabolizing enzyme. Genet Epidemiol. 2020;44(7):748–758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Bierut LJ, Stitzel JA, Wang JC, et al. Variants in nicotinic receptors and risk for nicotine dependence. Am J Psychiatry. 2008;165(9):1163–1171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Chen LS, Hung RJ, Baker T, et al. CHRNA5 risk variant predicts delayed smoking cessation and earlier lung cancer diagnosis--a meta-analysis. J Natl Cancer Inst. 2015;107(5). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Chen LS, Baker TB, Piper ME, et al. Interplay of genetic risk factors (CHRNA5-CHRNA3-CHRNB4) and cessation treatments in smoking cessation success. Am J Psychiatry. 2012;169(7):735–742. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Piper ME, Smith SS, Schlam TR, et al. A randomized placebo-controlled clinical trial of 5 smoking cessation pharmacotherapies. Arch Gen Psychiatry. 2009;66(11):1253–1262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Xu C, Goodz S, Sellers EM, Tyndale RF. CYP2A6 genetic variation and potential consequences. Adv Drug Deliv Rev. 2002;54(10):1245–1256. [DOI] [PubMed] [Google Scholar]
  • 15. Tanner JA, Tyndale RF. Variation in CYP2A6 Activity and Personalized Medicine. J Pers Med. 2017;7(4):18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Taylor AE, Fluharty ME, Bjorngaard JH, et al. Investigating the possible causal association of smoking with depression and anxiety using Mendelian randomisation meta-analysis: The CARTA consortium. BMJ Open. 2014;4(10):e006141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Chen LS, Bloom AJ, Baker TB, et al. Pharmacotherapy effects on smoking cessation vary with nicotine metabolism gene (CYP2A6). Addiction. 2014;109(1):128–137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Furberg H, Kim Y, Dackor J, et al. Genome-wide meta-analyses identify multiple loci associated with smoking behavior. Nat Genet. 2010;42(5):441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Caporaso N, Gu F, Chatterjee N, et al. Genome-wide and candidate gene association study of cigarette smoking behaviors. PLoS One. 2009;4(2):e4653. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Dai J, Lv J, Zhu M, et al. Identification of risk loci and a polygenic risk score for lung cancer: a large-scale prospective cohort study in Chinese populations. Lancet Respir Med. 2019;7(10):881–891. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Schizophrenia Working GPGC. Biological insights from 108 schizophrenia-associated genetic loci. Nature. 2014;511(7510):421–427. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Maas P, Barrdahl M, Joshi AD, et al. Breast cancer risk from modifiable and nonmodifiable risk factors among white women in the United States. JAMA Oncol. 2016;2(10):1295–1302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Aly M, Wiklund F, Xu J, et al. Polygenic risk score improves prostate cancer risk prediction: results from the Stockholm-1 cohort study. Eur Urol. 2011;60(1):21–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Belsky DW, Moffitt TE, Houts R, et al. Polygenic risk, rapid childhood growth, and the development of obesity: evidence from a 4-decade longitudinal study. Arch Pediatr Adolesc Med. 2012;166(6):515–521. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Chen LS, Horton A, Bierut L. Pathways to precision medicine in smoking cessation treatments. Neurosci Lett. 2018;669:83–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Belsky DW, Moffitt TE, Baker TB, et al. Polygenic risk and the developmental progression to heavy, persistent smoking and nicotine dependence: evidence from a 4-decade longitudinal study. JAMA Psychiatry 2013;70(5):534–542. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Chen LS, Hartz SM, Baker TB, et al. Use of polygenic risk scores of nicotine metabolism in predicting smoking behaviors. Pharmacogenomics. 2018;19(18):1383–1394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. El-Boraie A, Taghavi T, Chenoweth MJ, et al. Evaluation of a weighted genetic risk score for the prediction of biomarkers of CYP2A6 activity. Addict Biol. 2020;25(1):e12741. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Chen L-S, Baker TB, Miller JP, et al. Genetic Variant in CHRNA5 and response to varenicline and combination nicotine replacement in a randomized placebo-controlled trial. Clin Pharmacol Ther. 2020;108(6):1315–1325. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Tyndale RF, Zhu AZ, George TP, et al. Lack of associations of CHRNA5-A3-B4 genetic variants with smoking cessation treatment outcomes in caucasian smokers despite associations with baseline smoking. PLoS One. 2015;10(5):e0128109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Sarginson JE, Killen JD, Lazzeroni LC, et al. Markers in the 15q24 nicotinic receptor subunit gene cluster (CHRNA5-A3-B4) predict severity of nicotine addiction and response to smoking cessation therapy. Am J Med Genet B Neuropsychiatr Genet. 2011;156(3):275–284. [DOI] [PubMed] [Google Scholar]
  • 32. Baker TB, Weiss RB, Bolt D, et al. Human neuronal acetylcholine receptor A5-A3-B4 haplotypes are associated with multiple nicotine dependence phenotypes. Nicotine Tob Res. 2009;11(7):785–796. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Piper ME, Bolt DM, Kim S-Y, et al. Refining the tobacco dependence phenotype using the Wisconsin Inventory of Smoking Dependence Motives. J Abnorm Psychol. 2008;117(4):747–761. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Cox LS, Wick JA, Nazir N, et al. Predictors of early versus late smoking abstinence within a 24-month disease management program. Nicotine Tob Res. 2011;13(3):215–220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Partos TR, Borland R, Yong H-H, Hyland A, Cummings KM. The quitting rollercoaster: how recent quitting history affects future cessation outcomes (data from the International Tobacco Control 4-country cohort study). Nicotine Tob Res. 2013;15(9):1578–1587. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Purcell S, Neale B, Todd-Brown K, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81(3):559–575. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Das S, Forer L, Schönherr S, et al. Next-generation genotype imputation service and methods. Nat Genet. 2016;48(10):1284–1287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Price AL, Patterson NJ, Plenge RM, et al. Principal components analysis corrects for stratification in genome-wide association studies. Nat Genet. 2006;38(8):904–909. [DOI] [PubMed] [Google Scholar]
  • 39. Euesden J, Lewis CM, O’Reilly PF. PRSice: polygenic risk score software. Bioinformatics. 2015;31(9):1466–1468. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. R Core Team. R: A Language and Environment for Statistical Computing. Vienna:R Foundation for Statistical Computing;2013. [Google Scholar]
  • 41. 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(4):232–235. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Lerman C, Schnoll RA, HawkLW, Jr, et al. Use of the nicotine metabolite ratio as a genetically informed biomarker of response to nicotine patch or varenicline for smoking cessation: A randomised, double-blind placebo-controlled trial. Lancet Respir Med. 2015;3(2):131–138. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Cook JW, Piper ME, Leventhal AM, et al. Anhedonia as a component of the tobacco withdrawal syndrome. J Abnorm Psychol. 2015;124(1):215–225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44. Bechara A, Berridge KC, Bickel WK, et al. A neurobehavioral approach to addiction: implications for the opioid epidemic and the psychology of addiction. Psychol Sci Public Interest. 2019;20(2):96–127. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Leventhal AM, Zvolensky MJ. Anxiety, depression, and cigarette smoking: a transdiagnostic vulnerability framework to understanding emotion-smoking comorbidity. Psychol Bull. 2015;141(1):176–212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Ditre JW, Zale EL, LaRowe LR. A reciprocal model of pain and substance use: transdiagnostic considerations, clinical implications, and future directions. Annu Rev Clin Psychol. 2019;15(1):503–528. [DOI] [PubMed] [Google Scholar]
  • 47. Cannon DS, Baker TB, Piper ME, et al. Associations between phenylthiocarbamide gene polymorphisms and cigarette smoking. Nicotine Tob Res. 2005;7(6):853–858. [DOI] [PubMed] [Google Scholar]
  • 48. Baker TB, Piper ME, Schlam TR, et al. Are tobacco dependence and withdrawal related amongst heavy smokers? Relevance to conceptualizations of dependence. J Abnorm Psychol. 2012;121(4):909–921. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Moll M, Sakornsakolpat P, Shrine N, et al. Chronic obstructive pulmonary disease and related phenotypes: polygenic risk scores in population-based and case-control cohorts. Lancet Respir Med. 2020;8(7):696–708. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Buchwald J, Chenoweth MJ, Palviainen T, et al. Genome-wide association meta-analysis of nicotine metabolism and cigarette consumption measures in smokers of European descent. Mol Psychiatry. 2021;26(6):2212–2223. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51. Belsare S, Levy-Sakin M, Mostovoy Y, et al. Evaluating the quality of the 1000 genomes project data. BMC Genomics. 2019;20(1):620. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ntac043_suppl_Supplementary_Data
Click here for additional data file. (13.6KB, docx)
ntac043_suppl_Supplementary_Figures
Click here for additional data file. (323.8KB, pptx)
ntac043_suppl_Supplementary_Table_S1
Click here for additional data file. (14.6KB, docx)
ntac043_suppl_Supplementary_Taxonomy-form
Click here for additional data file. (728.5KB, pdf)

Data Availability Statement

Data is available from dbGaP and the NIDA Center for Genetic Studies: https://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?study_id=phs000404.v1.p1

https://nidagenetics.org/studies/study-35-genetically-informative-smoking-cessation-trial.

Articles from Nicotine & Tobacco Research are provided here courtesy of Oxford University Press

RESOURCES