Nicotine Metabolism-Related Genetic Polymorphisms Associated with Smoking Cessation in Korean Men: A Candidate Gene Association Study in a Korean Cohort

Cigarette smoking is a major risk factor for several human diseases. Among men in Korea, smoking accounted for 19.5% of all causes of mortality from 2011 to 2015 [1], and smoking accounted for 41.1% of all cancers and 33.4% of all cardiovascular diseases in 2012 [2]. In Korea, the smoking rate among men was 34.0% in 2020, which is notably higher than that in other developed countries [3]. Therefore, smoking cessation is crucial for reducing smoking-related diseases, lowering healthcare costs, and improving population health. However, despite widespread efforts, the success rate of smoking cessation varies among individuals.

The rate of nicotine metabolism, which is mediated by several enzymes, plays a key role in successful smoking cessation (Fig. 1) [4, 5]. Single nucleotide polymorphisms (SNPs) in nicotine metabolism-related genes influence nicotine metabolism across different ethnic groups [6, 7]. For instance, CYP2A6 SNPs, such as rs56113850 and rs113288603, are associated with nicotine metabolism in Finnish cohorts [8]. In African American smokers, CYP2A6 SNP rs12459249 significantly affects nicotine metabolism rates, explaining the differences between African American and European American smokers [9]. The frequency of CYP2A6 variants varies by ethnicity; slow-metabolizing alleles are more prevalent in Asian populations than in European and African populations, influencing smoking behaviors and nicotine dependence [10]. However, few studies have explored the association between genetic variants involved in nicotine metabolism and smoking cessation in Koreans. We hypothesized that specific polymorphisms in nicotine metabolism-related genes are associated with smoking cessation. To test this hypothesis, we examined a set of candidate genes associated with nicotine metabolism to identify relevant genetic polymorphisms associated with smoking cessation in men from a large Korean population-based cohort.

Data from the Korean Genome and Epidemiology Study (KoGES) Health Examination were used in this study. The cohort comprised community dwellers recruited from the National Health Examinee Registry (aged 40–79 years at baseline). Eligible participants were asked to volunteer, and data on their medical history and self-reported smoking status were collected. Participants were categorized as never smokers (individuals who had never smoked or smoked fewer than 100 cigarettes in their lifetime), former smokers (individuals who had smoked at least 100 cigarettes in their lifetime but had quit smoking at the time of the interview), and current smokers (individuals who had smoked at least 100 cigarettes in their lifetime and were actively smoking at the time of the interview), following a previous study [11].

The KoGES cohort comprised 58,701 individuals, all of Korean ethnicity, with available genome-wide SNP genotype data. Of these, female participants (n = 38,404) were excluded because of the relatively low smoking rate among women in Korea [3]. Participants with missing data (n = 59) and those with comorbidities (n = 15,874) were also excluded to minimize the confounding effect of diseases on smoking cessation. This approach enabled the selection of apparently healthy individuals, focusing the analysis on genetic rather than environmental factors.

Body mass index (BMI) was calculated as weight (kg) divided by height squared (m²). Regular exercisers were identified using the question, “do you engage in regular exercise to the extent of sweating?” with responses recorded as “yes” or “no.” Marital status was categorized based on the participants’ current marital status.

Genotype data, generated using the Korea Biobank Array (Affymetrix, Santa Clara, CA, USA), were provided by the Center for Genome Science, Korea National Institute of Health. The obtained data were filtered using the following criteria: call rate >97%, minor allele frequency >1%, and Hardy-Weinberg equilibrium p < 1 × 10⁻⁵. The genotypes were phased using ShapeIT v2 and IMPUTE v2 was used for imputation analyses of the genotype data, with 1,000 Genomes Phase 3 data as the reference panel. A total of 7,975,321 SNPs on chromosomes 1–22 were identified. From these SNPs, 12 gene regions associated with the nicotine metabolism pathway were selected (Fig. 1; online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000543543), encompassing 1,644 SNPs, with 57–395 SNPs per region.

The characteristics of the study population are presented as mean ± standard deviation or percentage. Continuous variables were evaluated using analysis of variance test to compare never, former, and current smokers, while categorical variables were assessed using chi-squared tests. To test the association between nicotine metabolism-related genetic polymorphisms and smoking cessation, SNPs in 12 gene regions (online suppl. Table 1) were analyzed. Genotype effects were determined by analyzing 1,644 SNPs in nicotine metabolism-related genes. Logistic regression analysis was conducted using PLINK v1.9 software by dividing the participants into three comparison groups: former compared to current smokers, never compared to current smokers, and never compared to former smokers. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated, adjusting for three sets of covariates across three models: model 1, adjusted for age; model 2, adjusted for age, drinking status, and exercise status; and model 3, adjusted for age, drinking status, exercise status, BMI, marital status, and monthly income. Statistical significance was set at p < 0.05.

A comprehensive list of SNPs analyzed in this study, along with their corresponding results, is provided in the online supplementary material. To calculate MAF and R², PLINK v1.9 software (https://www.cog-genomics.org/plink/1.9/) was used. SNPs with an MAF >1% were included to ensure robust statistical power and avoid rare variant bias. Independent SNPs were selected using a linkage disequilibrium threshold of R² <0.2 based on allele frequencies derived from the 1,000 Genomes East Asian reference panel.

For haplotype-based regression analysis, weighted linear regression was performed with smoking status as the dependent variable and the number of effect alleles as the independent variable. The analysis was performed in Python (version 3.12.4) using the generalized linear models from the statsmodels.api package (version 0.14.2, https://www.statsmodels.org/stable/glm.html). Covariates were included to adjust the models.

The demographic and lifestyle characteristics of the participants are shown in Table 1. The final analysis included 4,364 men (1,326 never; 1,684 former; and 1,354 current smokers). The average age of participants in different groups differed significantly, with current smokers being younger (49.6 ± 7.9 years) than former (53.1 ± 8.1 years) and never smokers (52.7 ± 8.5 years). Alcohol intake was significantly higher in current smokers than that in former and never smokers. In addition, a lower percentage of current smokers engaged in regular exercise than former and never smokers. Marital status and BMI also differed across the groups, whereas monthly income showed no significant variation.

The SNPs associated with smoking cessation in Korean men are presented in online supplementary Table 2, among which six SNPs (rs2431412, rs45625338, rs41297431, rs118063322, rs144769946, and rs2715904) showed significant associations after adjusting for age. Details of the associations of these SNPs across the three smoking status comparisons are shown in Table 2. The OR and p values for each SNP in the three models are shown in Table 2. In model 3, after adjusting for all covariates, four SNPs in the former compared to current smoker comparison and two SNPs in the never compared to current smoker comparison showed significant associations, whereas no significant SNP was identified in the never compared to former smoker comparison.

As shown in Table 3, a higher GRS in the former compared to current smokers and never compared to current smokers comparisons was associated with an increased likelihood of being a current smoker (OR = 1.18, 95% CI: 1.12–1.25, p = 4.53E−09 and OR = 1.13, 95% CI: 1.07–1.2, p = 3.33E−05, respectively). In contrast, no significant association was observed in the never compared to former smokers comparison (OR = 0.96, 95% CI: 0.92–1.02, p = 1.79E−01). These associations remained consistent across all three models.

Online supplementary Table 3 presents the distribution of effect allele numbers across the three smoking groups. Individuals with more effect alleles, linked to greater difficulty in smoking cessation, were more prevalent among the current smokers. An increase in effect allele numbers was associated with a higher proportion of current smokers, particularly in the higher effect allele category.

Online supplementary Table 4 presents the linear regression analysis linking the number of effect alleles with smoking status. An increased number of effect alleles was consistently associated with a higher likelihood of smoking in all models (OR = 1.04, 95% CI: 1.02–1.06, p < 0.0001 for each model).

Uhl et al. [14] demonstrated that successful smoking cessation is influenced by polygenic factors. Nicotine is primarily metabolized into cotinine [15], and the nicotine metabolic rate can help predict the success of transdermal nicotine therapy [4]. Kaufmann et al. [4] also showed that faster nicotine metabolism significantly reduces the likelihood of quitting smoking. To our knowledge, this is the first large population-based study to examine the genetic polymorphisms associated with smoking cessation in men, focusing on nicotine metabolism-related genes in Korea.

Our findings offer new insights into the association between genetic polymorphisms involved in nicotine metabolism and smoking cessation among Koreans, adding to research on ethnic variability. The significant correlation between a higher GRS and current smoking status aligns with studies showing genetic predisposition to nicotine dependence in other populations, such as African Americans and Europeans. For instance, the CYP2A6 SNP rs12459249 has been identified as a significant variant affecting nicotine metabolism rates in African American smokers, contributing to the observed ethnic differences in nicotine metabolism between African American and European American populations [9]. In our study, we identified several SNPs in nicotine metabolism-related genes associated with smoking cessation in Koreans, indicating that genetic variants influencing nicotine dependence may vary across populations. While SNPs like rs56113850 and rs113288603, which are linked to nicotine metabolic ratios in Finnish cohorts [8], were not prominent in our Korean population, this highlights population-specific genetic influences on smoking behavior.

This study identifies novel SNPs associated with smoking cessation, emphasizing the need for population-specific genetic research. These SNPs, selected based on their functional significance within nicotine metabolism pathways, have not been previously investigated in smoking cessation contexts. Although not previously linked to smoking cessation, their presence in key nicotine-metabolizing genes suggests that genetic variants within the same pathway may influence smoking behavior differently across populations. This study provides initial insights into the associations of these SNPs with smoking behaviors in a Korean cohort, expanding the understanding of genetic factors in nicotine dependence.

The GRS analysis revealed significant correlations between cumulative genetic risk and smoking behavior. A higher GRS was significantly associated with an increased likelihood of being a current smoker, across all adjusted models accounting for confounders such as age, drinking status, exercise, BMI, marital status, and income, underscoring the robustness of this relationship. The absence of a significant association between never and former smokers indicates that the GRS specifically distinguishes current smokers, making it a potential predictor of ongoing nicotine dependence. This finding aligns with those of previous studies showing that genetic predispositions play a crucial role in sustaining nicotine dependence, thereby affecting smoking cessation outcomes [16]. By aggregating the effects of multiple SNPs involved in nicotine metabolism, our GRS analysis provides a more comprehensive understanding of the genetic influences on smoking behavior than single SNP analyses. These findings highlight the importance of genetic studies across diverse populations to identify unique genetic factors in smoking behavior and cessation and to guide personalized interventions. In the Korean population, the association between higher GRS and current smoking status suggests that this model could identify individuals at higher risk of persistent smoking, informing more targeted and effective cessation strategies.

This study has some limitations. It is a candidate gene association study, not a genome-wide association study; therefore, we did not use a genome-wide association study p value significance threshold of 5 × 10⁻⁸. Our selection of specific genes was guided by prior knowledge of nicotine metabolic processes. The candidate gene approach allowed us to concentrate on known genetic factors in metabolic pathways, thus facilitating a more focused investigation. A genome-wide approach may not adequately explain these underlying mechanisms. We chose the candidate gene method to explore the specific functional roles of genes in nicotine metabolism. Furthermore, we did not apply a Bonferroni correction to the 1,644 individual SNP tests due to its highly stringent threshold. Instead, we conducted a GRS analysis to address multiple comparisons and SNP correlations. This method, which assesses the cumulative impact of multiple SNPs, achieved statistical significance even with stringent correction. The significant association in the GRS analysis underscores the collective influence of genetic variants on smoking cessation, offering insights into the complex genetic predisposition to smoking behavior and cessation challenges.

This study provides new insights into the link between genetic polymorphisms related to nicotine metabolism and smoking cessation in Korean men. Using a targeted candidate gene approach, specific SNPs within genes involved in nicotine metabolism were identified as influencing smoking status. Our findings highlight the potential of the GRS to predict smoking behaviors and emphasize the importance of population-specific genetic studies for customized smoking cessation interventions. These findings add to the growing understanding of genetic factors in nicotine dependence and support future research on personalized smoking cessation.

This study was conducted using bioresources provided by the National Biobank of Korea, Korea Disease Control and Prevention Agency, Republic of Korea (2019-059). We thank the participants and survey staff of the KoGES for their contribution to this study.

The study protocol was reviewed and approved by the Institutional Review Board of Theragen Bio (Approval No. 700062-20190819-GP-006-02). Written informed consent was obtained from all participants.

The authors have no conflicts of interest to declare.

This study was not supported by any sponsor or funding.

J.-M.P.: conceptualization, data curation, investigation, methodology, validation, visualization, and writing – original draft; J.-E.C.: data curation, formal analysis, resources, software, visualization, and writing – original draft; J.-W.L.: conceptualization, investigation, methodology, project administration, supervision, validation, and writing – review and editing; and K.-W.H.: formal analysis, project administration, resources, software, supervision, and writing – review and editing.

Jae-Min Park and Ja-Eun Choi contributed equally to this work and shared first authorship.

The data (KoGES Data) are publicly available at the following site: https://biobank.nih.go.kr/cmm/main/mainPage.do and can be obtained from the Korea Disease Control and Prevention Agency. Please contact the corresponding authors if you have any questions regarding the data. We will review your requests and assist you in accessing the data. To ensure transparency and provide full access to the data supporting our findings, a comprehensive list of all the SNPs analyzed in this study, along with their results, is available in the online supplementary material. This supplementary table includes detailed information on each SNP, allowing for a thorough examination of the genetic associations explored in our analysis.