Comparison of Carcinogen Biomarkers in Smokers of Menthol and Nonmenthol Cigarettes: The 2015–2016 National Health and Nutrition Examination Survey Special Sample

Although the prevalence of cigarette smoking decreased from 20.9% to 14.0% between 2005 and 2019 (1), cigarette smoking still remains the primary preventable cause of death in the United States and is responsible for more than 480,000 deaths annually (2). Menthol, a monocyclic terpene alcohol, is widely used by tobacco companies as a characterizing flavor in cigarettes (3, 4). Menthol acts directly on sensory nerves in the mouth, throat, nose, and lungs to cause sensations of cooling (5). The use of menthol as an additive in cigarettes has been criticized because it hides or masks the unpleasant harshness (irritation) and/or bitter taste of cigarette smoke and aversive physiologic effects of smoking (6, 7). Because of this masking function, menthol in cigarettes may facilitate smokers’ exposure to nicotine (8–10), and the likelihood of smoking initiation especially among Black youths (6, 11–16).

Mainstream cigarette smoke contain more than 7,000 chemical compounds in cigarette smoke (2). Some of them are well-known human carcinogens and others are toxic agents that contribute to smoking-related illness, such as heart disease and lung disease (2). A few studies have investigated the differences between menthol and nonmenthol cigarette smokers in biomarkers of tobacco smoke exposure, such as carbon monoxide (CO) and tobacco-specific nitrosamines (17–24). Some of them indicated menthol cigarette users had higher levels of cotinine and CO than nonmenthol smokers (17, 19, 22–24), whereas others concluded that smokers of both menthol and nonmenthol cigarettes exhibited similar levels of nicotine and nitrosamine biomarkers (20, 21, 25). Jones and colleagues found that menthol cigarette users presented significantly higher levels of blood cadmium (ratio of geometric mean; menthol/nonmenthol: 1.10 [95% confidence interval (CI), 1.04–1.16] compared with nonmenthol smokers, and similar urinary levels of the nicotine-derived nitrosamine ketone (NNK) metabolite 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL; ref. 26). That menthol may have higher levels of certain toxicants is a concern and may inform regulatory decisions.

Both polycyclic aromatic hydrocarbons (PAH) and volatile organic compounds (VOC) are included in the US FDA's list of harmful and potentially harmful constituents (HPHC; ref. 27). PAHs are usually generated by incomplete combustion of tobacco and other organic components during smoking (28), and benzo[a]pyrene (BaP) is classified as a Group 1 carcinogen, one of the most potent carcinogens (29, 30). Tobacco smoke is the main source of non-occupational exposure to harmful VOCs in the United States (31). Toxic and carcinogenic VOCs such as acrylonitrile and benzene (32) contribute to the risks of different tobacco-related cancers and to noncancer disease risk (33–35).

The National Health and Nutrition Examination Survey (NHANES) is a multidecade public health surveillance effort that aims to assess the health and nutrition status of adults and children in the United States. The NHANES generates publicly available data with a stratified, multistage, random probability sampling design conducted by the National Center for Health Statistics (NCHS) at the Center for Disease Control and Prevention (CDC) in the United States (36). The NHANES provides a nationally representative sample of the civilian, noninstitutionalized US population (36). The NHANES has consistently provided analytes of tobacco exposure biomarkers since its inception (37) and in the special sample, data on urinary biomarker concentrations became available in 2015 to 2016.

The health effects of menthol added to tobacco have been reviewed including a 2013 US FDA evaluation of the impact of menthol on public health (38, 39). Since 2017, forms of menthol bans have been implemented at the local level in several states, and in the European Union, Brazil, most of the Canadian Provinces, and other countries (40). On April 29, 2021, the US FDA announced its commitment to ban menthol as a characterizing flavor or additive in cigarettes and ban all characterizing flavors (including menthol) in cigars within the next year. Many aspects of tobacco harm and addiction were reviewed including smoke chemistry, targeted marketing, tobacco use initiation and progression, nicotine dependence, and disease risk. The proposed ban also considered that menthol cigarettes have historically been targeted towards Black communities and facilitates smoking initiation in youth. One area of tobacco harm that remains inconclusive is whether menthol affects exposure to tobacco smoke constituent biomarkers. Menthol has been studied in relation to the nicotine metabolite cotinine, but there is little information on exposure to tobacco carcinogens. PAHs and VOCs are among the major classes of tobacco carcinogens and toxicants. In this study, we analyzed levels of these biomarkers from the special sample of NHANES 2015–2016 by cigarette menthol flavor status.

The NHANES study protocol was approved by the National Center for Health Statistics Institutional Review Board, and all participants provided oral and written informed consent (36). NHANES 2015–2016 obtained data from both smokers and nonsmokers. Participants ages 18 years and older from one-third of overall NHANES sample were chosen for urine collection. Accordingly, the NHANES 2015–2016 special sample consists of participants ages 18 years or older within the one-third subsample of NHANES 2015–2016 and all adult smokers ages 18 years or older. Adult smokers were defined as participants who had smoked at least 100 cigarettes lifetime and now smoke cigarettes every day. There were 2,462 observations in the NHANES 2015–2016 special sample. To be eligible for our study, participants must have answered the cigarette use survey, have been classified as either a menthol cigarette or nonmenthol cigarette smoker, and must have reported menthol status (menthol or nonmenthol) information. We excluded smokers who had used any other tobacco products (pipes, hookah, cigars, e-cigarettes, smokeless tobacco, or snuff) within the last 5 days prior to the visit. Among NHANES 2015–2016 special sample, 70% of participants were nonsmokers, and 30% were smokers (n = 737). We excluded nonsmokers from our study. Among subpopulation of 737 smokers, 48% (351/737) of them were exclusive cigarette users with nonmissing information on menthol status, demographic, and creatinine measurements. Therefore, the final analytic sample consists of 351 exclusive cigarette smokers, of which 46.0% (162/351) smoked menthol cigarettes and 54.0% (189/351) smoked nonmenthol cigarettes.

Information on smoking status, cigarettes smoked per day, cigarette rod length, cigarette menthol indicator, and Federal Trade Commission (FTC) nicotine content was obtained from cigarette use and recent tobacco use surveys, and the Universal Product Code. Cigarette rod length was categorized as regular/king (68–72 mm, 79–88 mm) and long/ultra-long (94–101 mm, 110–121 mm; ref. 41). Sociodemographic variables such as age at screening (continuous), gender (categorical, male vs. female), race/ethnicity (categorical, Non-Hispanic White vs. Non-Hispanic Black vs. Mexican American/Hispanic vs. all others), education attainment (categorical, less than high school vs. high school or higher), weight (continuous, kg), height (continuous, cm), body mass index (BMI, kg/m²), and ratio of family income to poverty (continuous) were collected through surveys administered in the mobile examination centers (MEC).

Biomarkers available in the NHANES 2015–2016 special sample included urine cotinine and urine 3′-hydroxycotinine (3HC). Cotinine and 3HC are the two predominant nicotine metabolites in urine (42, 43). Using these two measures, we calculated the nicotine metabolite ratio (NMR), which is the ratio of urine 3HC and cotinine (44). Urine metabolites of PAHs, VOCs, and heavy metal levels were also available in the NHANES 2015–2016 special sample. Although the NHANES provided VOCs data in blood, urine biomarkers of VOCs have a longer half-life and are more stable during storage and handling (45). For urine metabolites of PAHs, our analysis included 1-hydroxynapthalene, 2-hydroxynapthalene, 3-hydroxyfluorene, 2-hydroxyfluorene, 1-hydroxyphenanthrene, 1-hydroxypyrene, 2-hydroxyphenanthrene, and 3-hydroxyphenanthrene. Supplementary Table S1 listed all 15 urine metabolites of VOCs analyzed in our study including their parent compound and common names. For heavy metals, we included cadmium, cobalt, lead, and uranium. All urine metabolites of PAHs, VOCs, and heavy metal concentrations (continuous) were corrected for dilution by creatinine and were reported per gram of creatinine (41).

For biomarker results below the lower limit of detection, an imputed fill value was placed in the analyte results filed by NHANES. This value is the lower limit of detection (LLOD) divided by the square root of 2 (LLOD/sqrt[2]; ref. 36). Urinary nicotine metabolites were measured by two separate isotope dilution high-performance liquid chromatography/tandem mass spectrometric (HPLC/MS-MS) methods (36). Urine specimens are processed, stored, and shipped to the Division of Laboratory Sciences, National Center for Environmental Health, and Centers for Disease Control and Prevention for analysis. Additional detailed description of the laboratory methods are available elsewhere (36).

The relationship between the categorical and continuous descriptive variables were analyzed using the Rao–Scott χ² tests and two-sample t tests. Multivariable linear regression models were used to obtain the covariate adjusted geometric means for urine metabolites of PAHs, VOCs, and heavy metal levels. All biomarkers were natural log-transformed to better fit the regression assumptions and analysis models of NHANES biomarkers (41). We estimated multivariable adjusted ratios of geometric means of biomarkers of tobacco exposure (PAHs, VOCs, and metals) comparing current smokers of menthol cigarettes with nonmenthol cigarette smokers. The ratios of the geometric means and their 95% CIs were obtained by exponentiating the estimates obtained from the linear regression models on log-transformed biomarker levels (26). In addition to log transformation, urine metabolites of PAHs, VOCs, and heavy metals were creatinine-corrected to obtain covariates adjusted geometric means from the regression models. We included following covariates to control for any potential confounding effects: gender, race/ethnicity, education, cigarette rod length (41), age at screening, ratio of family income poverty, FTC nicotine (25), BMI, urinary nicotine metabolite ratio (NMR), and cigarettes per day (CPD).

SAS statistical software version 9.4 (SAS Institute Inc.) was used for all statistical analyses with a two-sided significance level of 0.05. SAS SURVEY Procedures (PROC SURVEYMEANS, PROC SURVEYFREQ, and PROC SURVEYREG) were used to perform all statistical analyses with appropriate weights (from the NHANES 2015–2016 special sample), strata, and clustering variables to account for the complex sampling design of NHANES and to obtain nationally representative estimates (36). Unweighted frequencies and weighted percentages were reported for ease of interpretation of the study findings.

The data analyzed in this study were obtained from National Health and Nutrition Examination Survey (NHANES) Special Sample 2015–2016. The NHANES data are available at https://www.cdc.gov/nchs/nhanes/index.htm.

Table 1 summarizes the sample characteristics of the menthol and nonmenthol cigarette smokers. The gender distribution was similar among exclusive cigarette smokers. There were significant differences between menthol and nonmenthol cigarette smokers by race/ethnicity, education attainment, age at screening, FTC nicotine, and cigarettes per day. Menthol smokers were more likely to be younger (42 vs. 48), have less than a high school education (24% vs. 14%), be of non-Hispanic black (38% vs. 6%), smoke less cigarettes per day (11 vs. 14) than nonmenthol cigarette users.

Table 2 lists the adjusted geometric means of urine metabolites of PAHs biomarkers by menthol status. There were no statistically significant differences between menthol and nonmenthol cigarette users in the 6 of 7 creatinine-corrected metabolites of PAHs biomarkers. Menthol cigarette smokers had significantly lower 1-hydroxyphenanthrene of creatinine-corrected PAHs concentrations compared with nonmenthol cigarette smokers (167.2 μg/g creatinine vs. 198.1 μg/g creatinine; ratio of GMs = 0.8; 95% CI, 0.7–1.0; P = 0.03).

Table 3 lists the adjusted geometric means of urine metabolites of VOCs by menthol status. There were no statistically significant differences in most of (12/15) urine metabolites of VOCs between menthol and nonmenthol cigarette smokers. Menthol cigarette smokers presented significantly lower concentrations of acrylamide (115.6 μg/g creatinine vs.154.6 μg/g creatinine; ratio of GMs = 0.7; 95% CI, 0.6–0.9; P = 0.02), N,N-dimethylformamide (403.1 μg/g creatinine vs. 452.5 μg/g creatinine; ratio of GMs = 0.9; 95% CI, 0.8–0.9; P = 0.01), and acrylonitrile (120.9 μg/g creatinine vs. 143.6 μg/g creatinine; ratio of GMs = 0.8; 95% CI, 0.7–0.9; P = 0.04) than nonmenthol cigarette smokers.

Table 4 lists the adjusted geometric means of urine heavy metals by menthol status. There were no statistically significant differences in all urine heavy metals between menthol and nonmenthol cigarette smokers after adjustment of covariates.

Supplementary Table S2 shows the adjusted geometric means for tobacco smoke exposure biomarkers and NMR. After adjusting for gender, race/ethnicity, education, cigarette rod length, age, BMI, FTC nicotine, and cigarettes per day, there were no differences in urine cotinine (ratios of GMs: 1.0; 95% CI, 0.8–1.3; P > 0.1), and 3HC (ratios of GMs: 1.1; 95% CI, 0.9–1.6, P > 0.1) between menthol and nonmenthol cigarette smokers. Adjusted geometric means for urine NMR were 1.8 and 1.6 for menthol and nonmenthol cigarette smokers, respectively (Supplementary Table S2; ratio of GMs: 1.1; 95% CI, 0.8–1.6; P > 0.1).

Both PAHs and VOCs are listed by FDA as HPHC in tobacco (27). Because of the inherent toxicity and carcinogenicity of PAHs and VOCs, it is important to consider the differences in exposure to these substances among users of mentholated tobacco. We quantified urine metabolites of 7 PAHs, 15 VOCs, and 4 heavy metals collected from the cigarette smokers participating in the NHANES 2015–2016 special sample. We found that one PAH metabolite (1-hydroxyphenanthrene) and the urine metabolites of three VOCs (acrylamide, N,N-dimethylformamide, and acrylonitrile) were significantly lower among menthol cigarette smokers compared with nonmenthol cigarette smokers. No differences by menthol use were found for the 22 other markers. PAHs are a class of compounds composed of two or more fused benzenoid rings, which lead to mutagenic effect (46). As reviewed by the International Agency for Research on Cancer (IARC), several PAHs in tobacco smoke are known human carcinogens (30). Further, acrylonitrile, a Group 2B carcinogen as classified by IARC, is possibly carcinogenic to human (47). However, there are no differences between menthol and nonmenthol cigarette smokers in most of urine metabolites of PAHs and VOCs biomarkers. For instance, biomarkers of 1,3-butadiene, which is related to development of hematologic malignancies (48), and acrolein, a potent cardiopulmonary toxicant which contributes to increased risk of cardiovascular disease (49) and potentially lung cancer (50), did not differ between menthol and nonmenthol cigarette smokers. In addition to PAHs and VOCs, it was previously reported in studies of NHANES that exposure to the tobacco-specific nitrosamine 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), another IARC classified lung carcinogen (51–53), does not appear to differ by menthol status as determined by its urinary biomarker 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol NNAL (20, 21). These findings appear consistent with studies that showed the rates and risk of lung cancer were similar in menthol and nonmenthol smokers (54–59). PAHs in cigarette smoke have also been linked to cardiovascular disease (60), and the current findings are also consistent with similar risks of CVD between menthol and nonmenthol smokers (61).

Tobacco smoke contains over thousands of chemicals, including carcinogenic and toxicants that lead to malignancy and disease, regardless of the cigarette flavor being smoked (32). The tobacco industry has historically marketed different tobacco products, especially menthol cigarettes, to Black Americans in urban communities (39). Because of aggressive tobacco marketing, more than 85% of Black smokers use menthol cigarettes (62). Black Americans constituted only 12% of the total US population, but menthol is responsible for 1.5 million extra smokers, more than 150,000 smoking-related premature deaths and 1.5 million excess life-years lost during 1980 to 2018 (62). Although menthol cigarette users experienced slightly but significantly lower concentrations of four urine metabolites of PAHs and VOCs in this study, there were no differences in most of biomarkers of toxicant exposure examined.

This study also examined the relationship between menthol and NMR, which is positively associated with cigarette consumption (63). The NHANES started to provide 3′-hydroxycotinine and cotinine measurements in 2013 and the data on special smoker sample with urinary biomarker concentrations became available for the first time with 2015–2016 wave of the survey. In-vitro studies suggest that menthol may impact metabolism of nicotine through inhibition of CYP2A6 activity (64) and clearance of nicotine (65), where higher NMR is sometimes associated with greater cigarette consumption. Using the nationally representative sample of smokers, we found that urine NMR did not differ between menthol and nonmenthol cigarette smokers (1.8 menthol users vs. 1.6 nonmenthol users; P-value > 0.1; Supplementary Table S2).

The results of our study should be considered within the context of several limitations. Cigarette brands are classified as menthol and nonmenthol by their manufacturers, but most cigarettes contain low levels of menthol even if they are classified and/or marketed as nonmenthol brands (25, 66). Also, there may be other cigarette design characteristics besides menthol flavor that can affect the exposure to tobacco smoke constituents (67). The major public health burden of added menthol is its role in facilitating smoking initiation among youth (6, 11–15). Separately, menthol may also facilitate increased tobacco smoke inhalation, but we did not find direct evidence of this effect in this study as shown by similar cotinine levels with nonmenthol smokers. Notably, the generalizability of our findings may be limited by the exclusion of dual/poly tobacco product users, such as dual users of cigarette and electronic cigarette (e-cigarette) given e-cigarette use has increased significantly worldwide among youth (68, 69). Although the NHANES 2015–2016 includes information on e-cigarette use, the sample size for dual users was considered too small to conduct an analysis by menthol use. Another limitation is that the NHANES 2015–2016 special sample captures data on adult smokers who smoke cigarettes every day, which hinders the generalizability of the current study results to non-daily smokers.

Overall, there were no significant differences in cotinine levels or in 22 of 26 HPHCs by menthol status, providing supporting evidence that menthol in cigarettes does not affect exposure to tobacco smoke constituents. There is considerable data summarized by the FDA that support a menthol ban including its effect on increased experimentation, initiation, and dependence in youths and young adults, particularly in black smokers. Menthol is also associated with poorer cessation rates. On April 28, 2022, FDA announced proposed product standards to prohibit menthol as a characterizing flavor in cigarettes. Pending public comments, the FDA will consider issuing a final product standard. These timely actions may benefit public health by significantly reducing disease and death attributed to traditional cigarette smoking.

J.E. Hayes reports personal fees from US FDA and Elsevier, and grants from US NIH [U01DC019573] outside the submitted work; also has a patent for Provisional Application 63/308,741 pending. No disclosures were reported by the other authors.

W. Lin: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. J. Zhu: Formal analysis, investigation, methodology, writing–review and editing. J.E. Hayes: Investigation, methodology, writing–review and editing. J.P. Richie: Investigation, methodology, writing–review and editing. J.E. Muscat: Conceptualization, investigation, methodology, writing–original draft, writing–review and editing.

J.P. Richie was supported by the NIH (under Award No. R01HL147344). J.E. Muscat was supported by the NIH (under Award No. P50 DA036107). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.