urinary tobacco smoke constituent biomarkers for assessing ... › content › canres › ... ·...

Review

Urinary Tobacco Smoke–Constituent Biomarkers forAssessing Risk of Lung Cancer

Jian-Min Yuan1,2, Lesley M. Butler1,2, Irina Stepanov3, and Stephen S. Hecht3

AbstractTobacco-constituent biomarkers are metabolites of specific compounds present in tobacco or tobacco smoke.

Highly reliable analytic methods, based mainly on mass spectrometry, have been developed for quantitation ofthese biomarkers in both urine and blood specimens. There is substantial interindividual variation in smoking-related lung cancer risk that is determined in part by individual variability in the uptake and metabolism oftobacco smoke carcinogens. Thus, by incorporating these biomarkers in epidemiologic studies, we can potentiallyobtain a more valid and precise measure of in vivo carcinogen dose than by using self-reported smoking history,ultimately improving the estimation of smoking-related lung cancer risk. Indeed, we have demonstrated this byusing a prospective study design comparing biomarker levels in urine samples collected from smokersmany yearsbefore their development of cancer versus those in their smoking counterparts without a cancer diagnosis.The following urinary metabolites were associated with lung cancer risk, independent of smoking intensity andduration: cotinine plus its glucuronide, a biomarker of nicotine uptake; 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol and its glucuronides (total NNAL), a biomarker of the tobacco carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK); and r-1-,t-2,3,c-4-tetrahydroxy-1,2,3,4-tetrahydrophenanthrene (PheT), a biomarkerof polycyclic aromatic hydrocarbons (PAH). These results provide several possible new directions for usingtobacco smoke–constituent biomarkers in lung cancer prevention, including improved lung cancer riskassessment, intermediate outcome determination in prevention trials, and regulation of tobacco products.Cancer Res; 74(2); 401–11. �2014 AACR.

IntroductionThe tobacco epidemic is responsible for 12% of all deaths

worldwide among adults over 30 years of age and is consideredthe world's leading cause of preventable premature death (1).Although tobacco consumption has decreased in the UnitedStates since 2000 (2), worldwide tobacco consumption hasincreased by 75% over the past 30 years to 7.4 million metrictons (3). It is estimated that there are approximately 1.4 billionsmokers worldwide (4). The highly addictive nature of nicotinecombined with effective exposure of susceptible target tissuesto carcinogens in tobacco smoke distinguish tobacco products

from all other commodities as the single greatest cause ofcancer-related deaths.

Lung cancer is one of the most commonly diagnosedcancer and is the leading cause of cancer-related deaths inthe United States and worldwide (5, 6). Cigarette smokingis the most important causal factor for lung cancer; anestimated 90% of all lung cancer–related deaths are attrib-utable to cigarette smoking (7). An estimated 11% of femalesmokers and 24% of male smokers may die from lung cancerover their lifetime, assuming no competing cause of death(8). Cigarette smoke contains 73 established animal carci-nogens, 16 of which are rated as carcinogenic to humans (9).The interindividual variation in smoking-related lung cancerrisk may be determined in part by variability in the uptakeand metabolism of tobacco smoke carcinogens, such as4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), N0-nitrosonornicotine (NNN), and polycyclic aromatic hydro-carbons (PAH; ref. 10). Although there is no doubt thatexposure to these compounds is a major cause of smoking-related lung cancer, we have little ability to predict whoamong the more than 1 billion smokers in the world willactually get lung cancer.

This review will discuss the possible future application oftobacco smoke carcinogen and toxicant biomarkers in lungcancer prevention, including improved lung cancer risk assess-ment, intermediate outcome determination in preventiontrials, and regulation of tobacco products.

Authors' Affiliations: 1Division of Cancer Control and PopulationSciences, University of Pittsburgh Cancer Institute; 2Department of Epi-demiology, Graduate School of Public Health, University of Pittsburgh,Pittsburgh, Pennsylvania; and 3Masonic Cancer Center, University ofMinnesota, Minneapolis, Minnesota

This article is being published as part of the AACR's commemorationof the 50th Anniversary of the Surgeon General's Report on SmokingandHealth. Youareencouraged to visit http://www.aacr.org/surgeon-general for information on additional AACR publications and activitiesrelated to the recognition of this important anniversary.

Corresponding Author: Jian-Min Yuan, University of Pittsburgh CancerInstitute, UPMC Cancer Pavilion, Suite 4C, 5150 Centre Avenue, Pitts-burgh, PA 15232. Phone: 412-864-7889; Fax: 412-623-3303; E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-13-3178

�2014 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 401

on June 15, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst January 9, 2014; DOI: 10.1158/0008-5472.CAN-13-3178

http://cancerres.aacrjournals.org/

Biomarkers Associated with Tobacco SmokingOur group has developed and applied laboratorymethods to

quantify urinary tobacco constituents and their metabolites inhumans. The quantified constituents include metabolites ofnicotine, a highly addictive, noncarcinogenic compound, aswell as established carcinogens, such as tobacco-specific nitro-samines, PAH, and volatile organic compounds (VOC). Beloware brief descriptions of the specific biomarkers illustratedin Fig. 1.

Nicotine and its metabolitesNicotine's stimulant effect is a major contributing factor to

the dependence-forming properties of tobacco smoking. How-ever, nicotine and its major metabolites are noncarcinogenic.Nicotine is extensively metabolized, primarily in the liver, andits major proximate metabolite is cotinine; on an average, 75%of nicotine is converted to cotinine, primarily by the liverenzyme cytochrome P450 2A6 (CYP2A6; ref. 11). Cotinine andcotinine-N-glucuronide are metabolites of nicotine; their sumis "total cotinine." Total cotinine is a reliablemeasure of uptakeof nicotine, and widely used as a biomarker of exposure toactive and passive tobacco smoking. Cotinine is furthermetab-

olized to trans-30-hydroxycotinine (3-HC) by the same liverenzyme CYP2A6. The in vivo half-life (t1/2) of cotinine is longer(16 hours) than that of nicotine (2 hours; ref. 12). Urinarycotinine concentrations average 4- to 5-fold higher than thosein plasma (13), making urine a more sensitive medium for thedetection of low-level exposure to tobacco smoke and use ofother nicotine-containing products. Alternatively, urinary totalnicotine equivalent, that is, the sum of total nicotine, totalcotinine, and total 3-HC, represents the uptake and metabo-lism of total nicotine, and presumably a better biomarker forthe consumption of cigarettes and other nicotine-containingproducts.

Tobacco-specific nitrosaminesThe tobacco-specific nitrosamines comprise one of the

major groups of carcinogenic chemicals in tobacco and ciga-rette smoke. The formation of tobacco-specific nitrosaminesoccurs primarily during tobacco curing. Among the seventobacco-specific nitrosamines found in tobacco, NNK andNNN are considered the most carcinogenic (10). Both NNKand NNN are classified as human carcinogens (group I) by theInternational Agency for Research on Cancer (IARC; ref. 14).

© 2014 American Association for Cancer Research

Cancer Research Surgeon General’s Report

Tobacco smoke

Nicotine

Nicotine

Cotinine

3-HC

HPMA∗

∗

SPMA∗ HEMA∗HBMA∗MHBMA∗PheTNNAL

PAH VOC’sTSNA

NNN NNK Ethyleneoxide

Croton-aldehyde

1, 3–ButadieneBenzeneAcroleinPhenanthrene

CH3

CH3

N

N

CH3N

N

N O

N O

OO O

O

O

CH3N

N O

N N

N

CH3

CH2NHCOCH3

N OOH

HOHO

OH

OH

OHRS

SR

OH

OH

SR

SR

R =

+

RS

OH

OOOH

N

NOH

SRHO

H

HN

Figure1. Overviewof tobaccosmokeconstituents and theirmetabolites asbiomarkersof toxicant andcarcinogenexposure in smokers.HBMA, 4-hydroxybut-2-yl mercapturic acid; 3-HC, trans-30-hydroxycotinine; HEMA, 2-hydroxyethyl mercapturic acid; HPMA, 3-hydroxypropyl mercapturic acid; MHBMA,monohydroxybutyl mercapturic acid; SPMA, S-phenyl mercapturic acid; TSNA, tobacco-specific nitrosamines; VOC, volatile organic compounds.

Yuan et al.

Cancer Res; 74(2) January 15, 2014 Cancer Research402




NNK is a strong lung carcinogen capable of inducing lungtumors in rodents independent of its route of administration(15). In the rat, the lowest total dose of NNK shown to inducelung tumors was 1.8 mg/kg, with a significant dose-responsetrend (16). This lowest total dose in rats is close to theestimated average uptake of 1.1 mg/kg of NNK for a heavysmoker with 40 years of smoking (15). NNK itself is notdetectable in human urine because of its rapid and extensivemetabolism to 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol(NNAL) and other products (15, 17).NNAL is a major metabolite of NNK. Like NNK, NNAL is an

established powerful lung carcinogen in rats and mice,although somewhat less carcinogenic than NNK. Urinary totalNNAL, that is, the sum of free NNAL and its glucuronides, is awell-established biomarker of NNK uptake (10). Thus, NNALdetection in urine signifies exposure to, and uptake of, the lungcarcinogen NNK. Given that NNK is found only in tobaccoproducts, a major advantage of the total NNAL biomarker is itsspecificity to tobacco smoke exposure. Another advantage ofurinary total NNAL over other tobacco constituents for epi-demiologic studies is its relatively long half-life (i.e., 10 days to 3weeks; ref. 10).Similar to NNK, NNN is produced during curing, aging, and

burning of tobacco. Human exposure to NNN can bemeasuredvia quantitation of unchanged NNN (also called free NNN) andits detoxification product NNN-pyridine-N-glucuronide (NNN-N-Gluc) in urine (18). The sum of free NNN and NNN-N-Gluc inurine is referred to as total NNN.Virtually all unburned commercial tobacco products con-

tain NNN and NNK, and they always occur together (14). NNNis also present in smoke from cigars and cigarettes, in the salivaof peoplewho chew betel quidwith tobacco, and in the saliva oforal-snuff users (19). There is great variation in levels of NNNandNNK inmainstream smoke of cigarettes. This ismainly dueto the differences in tobacco types used, agricultural practices,curing methods, and manufacturing processes (14). Levels ofNNN range from 20 to 58,000 ng/cigarette and NNK from 19 to10,745 ng/cigarette in tobacco fromcommercial cigarettes soldin different parts of the world. In mainstream smoke, theranges of NNN and NNK were reported from 4 to 2,830ng/cigarette and 3 to 1,749 ng/cigarette, respectively (14).A collaborative study between the World Health Organiza-

tion and the U.S. Centers for Disease Control and Preventionfound that the levels of NNK plus NNN in the mainstreamsmoke of Marlboro cigarettes purchased in 10 different coun-tries were significantly higher than local top-selling brands ofcigarettes from the same countries (20). For example, themedian levels of NNK plus NNN in Marlboro cigarettes in theUnited States were very similar to the Marlboro-brand cigar-ettes sold in China (i.e., 216.3 and 263.1 ng/cigarette, respec-tively), whereas considerably lower median levels werereported in the local popular brands of cigarettes in China(Hongtashan: 5.8 ng/cigarette) and India (Gold Flake: 12.4 ng/cigarette; ref. 20). These data suggest that differences in cancerrisk among smokers, after taking into account the number ofcigarettes consumed, may be due in part to differences in NNKand NNN uptake among smokers consuming different brandsof cigarettes.

PAHPAH as a group include hundreds of chemicals that com-

monly occur as mixtures in the environment. PAH are presentin cigarette smoke as well as in the general environmentresulting from incomplete combustion of organic matter.Extensive investigations have demonstrated that PAH-enriched cigarette smoke condensate fractions are carcino-genic to mouse skin and rat lung (21, 22). Fourteen individualPAH compounds, including the widely studied PAH ben-zo[a]pyrene (BaP), have been rated as having sufficient evi-dence for carcinogenicity in laboratory animals, and BaP isconsidered carcinogenic to humans (23, 24).

We have developed and validated several biomarkers forPAH uptake and metabolism (25–27). r-1,t-2,3,c-4-Tetrahy-droxy-1,2,3,4-tetrahydrophenanthrene (PheT) is a metaboliteof phenanthrene, the simplest PAH with a bay region, a featurethat is closely associated with the carcinogenicity. 1-Hydro-xyprene is a metabolite of the noncarcinogenic pyrene that isalways a component of PAH mixtures. We applied thesebiomarkers in smokers and nonsmokers in epidemiologicstudies (28–30).

VOCsBesides tobacco-specific nitrosamines and PAH, tobacco

smoke contains a myriad of VOCs that are toxic and somemaybe carcinogenic to humans. Levels inmainstream smoke of theroutinely quantified VOCs acrolein, benzene, 1,3-butadiene,and crotonaldehyde are 100 to 1,000 times greater than those oftypical PAH and NNK (28). Urinary mercapturic acids are well-established and validated biomarkers of uptake of these com-pounds (shown in Fig. 1), and all are found at higher levels inthe urine of smokers than in nonsmokers (28, 31). Benzene, 1,3-butadiene, and ethylene oxide are considered carcinogenic tohumans by the IARC, based mainly on occupational studiesand mechanistic data related to hematopoietic malignancies(32, 33). Acrolein is a highly toxic but marginally carcinogeniccompound, whereas crotonaldehyde is a relatively weak hepa-tocarcinogen (34, 35).

Biomarkers of Tobacco Smoke Constituents inRelation to Risk of Lung Cancer

Only an estimated 11% to 24% of lifelong smokers may diefrom lung cancer over their lifetimes (8). This variability insusceptibility to lung cancer among smokers is due in part tothe fact that not all smokers of a given frequency of cigaretteuse are exposed to the same levels of tobacco constituents.Sources of variability of the in vivo dose of tobacco smokecarcinogens include preference for certain brands of cigarettesthat contain different levels of carcinogens, as well as differ-ences in smoking behaviors, such as intensity of smoking,number of puffs per cigarette, and depth of inhalation. Abiomarker approach would more directly and closely measurethe internal dose to tobacco smoke constituents, thereforepotentially providing a better estimate of the risk of developinglung cancer.

Using the Shanghai Cohort Study, we conducted a seriesof nested case–control studies to examine the association

Tobacco Smoke Biomarkers and Lung Cancer

www.aacrjournals.org Cancer Res; 74(2) January 15, 2014 403




between urinary biomarkers of cigarette smoke constituentsand the risk of developing lung cancer among current smokers.The Shanghai Cohort Study enrolled 18,244 men betweenJanuary 1, 1986 and September 30, 1989 (36, 37). At enrollment,all subjectswere between 45 and 64 years of age and lived in oneof four geographically defined communities in the city ofShanghai, China. In-person interviews were conducted and aurine samplewas obtained fromeach subject upon enrollment.Cases of lung cancer were identified annually by in-person re-interviews of all surviving cohort members and routine reviewof reports from the Shanghai Cancer Registry and the ShanghaiMunicipal Vital Statistics Office. For the nested case–controlstudy, we included patients with lung cancer who smokedcigarettes at enrollment (i.e., at the time of urine collection).For each case, we randomly selected one control subject fromall cohort members who were current smokers at enrollment,without a history of cancer, and alive at the time of cancerdiagnosis of the index case. Controlswerematched to the indexcase by age at enrollment (�2 years), year and month of urinecollection (�1 month), and the same neighborhood of resi-dence at recruitment. Urine samples were retrieved from thebiorepository of the cohort study and analyzed for the follow-ing tobacco smoke carcinogen and toxicant biomarkers: totalcotinine, total NNAL, total NNN, PheT, and mercapturic acidmetabolites of acrolein, benzene, 1,3-butadiene, crotonalde-hyde, and ethylene oxide. In this article, we review thefindings from the Shanghai Cohort Study and other pro-spective studies for associations between urinary metabo-lites of nicotine, tobacco-specific nitrosamines, PAH, andVOCs and lung cancer risk. Table 1 summarizes the mainfindings on these biomarkers in relation to lung cancer fromthe Shanghai Cohort Study and other selected studies with asimilar design.

Nicotine metabolitesSeveral epidemiologic studies, mainly using a cross-sec-

tional or retrospective study design, have been carried out toexamine the association between plasma or urinary cotinine,a major metabolite of nicotine, and risk of lung cancer. Theearliest reported findings based on prospective epidemio-logic data were from a nested case–control study of 92 lungcancer cases and 305 controls from more than 26,000 womenof ages 40 to 64 years living in the Netherlands (38). Thestudy reported an odds ratio (OR) of lung cancer for thehighest tertile level of urinary total nicotine metabolites tobe 18.8 [95% confidence interval (CI), 8.2–42.9], comparedwith those with negative measurement of urinary nicotineand its pyridyl-containing metabolites (i.e., nonsmokers). Atrend was observed with increasing lung cancer risk amongsmokers by level of urinary total nicotine metabolites. Sim-ilar findings have since been replicated in a few prospectiveepidemiologic studies (39, 40).

On the basis of the Shanghai Cohort Study, we conducteda nested case–control study of lung cancer among men whowere self-reported current smokers and whose urinarycotinine levels were 35 ng/mL or higher. Our initial reportamong 155 lung cancer cases and 152 controls showed thatcurrent smokers with the highest tertile of urinary total

cotinine (�2,615 ng/mg creatinine) had an OR of 3.76 (95%CI, 1.75–8.06) relative to smokers with the lowest tertile oftotal cotinine (35–�1,196 ng/mg creatinine) after adjust-ment for number of cigarettes per day, number of years ofsmoking, and urinary total NNAL (Ptrend ¼ 0.002; ref. 39). Weexpanded this initial study to include 476 lung cancer casesand 476 controls from the same Shanghai Cohort Study andreported a similar association between urinary total cotin-ine and lung cancer risk; ORs (95%CI) for the second andthird tertile of total cotinine were 2.18 (1.43–3.32) and 3.52(2.30–5.41), respectively, compared with the lowest tertile(Ptrend < 0.001), also with the adjustment for smokingintensity (cigarettes/day) and duration (number of yearsof smoking), and urinary total NNAL, in addition to urinaryPheT (30).

In addition to the Shanghai Cohort Study, two nested case–control studies of lung cancer were also conducted amongcurrent smokers. Using the Singapore Chinese Health Studydataset, Yuan and colleagues conducted a nested case–controlstudy of lung cancer (91 cases and 93 controls) and reported a2-fold increased risk of lung cancer for the highest relative tothe lowest tertile of urinary total cotinine among currentsmokers, and the elevated risk diminished slightly after adjust-ment for urinary total NNAL (39). In contrast, a positiveassociation was not observed between serum cotinine leveland lung cancer risk in a nested case–control study of lungcancer among current smokers (100 cases and 100 matchedcontrols) that was conducted by Church and colleagues byusing the dataset of the Prostate, Lung, Colorectal, andOvarianCancer Screening Trial (PLCO; ref. 41).

The following two nested case–control studies were con-ducted among both nonsmokers and smokers. On the basisof the European Prospective Investigation into Cancer andNutrition (EPIC) cohort, an ongoing prospective study ofmore than 520,000 men and women between 1992 and 2000from 10 European countries, Timofeeva and colleaguesconducted a nested case–control study of lung cancerincluding current and former smokers as well as neversmokers (894 cases and 1,805 matched controls). Lungcancer risk increased monotonically with increasing serumcotinine levels; OR of lung cancer was 12.4 (95% CI, 7.1–21.9)for subjects in the highest decile (serum cotinine >1,800nmol/L or >316.8 ng/mL) relative to nonsmokers (serumcotinine <75 nmol/L or <13.2 ng/mL) after adjustment fornumber of cigarettes per day (42). In a similar analysisinvolving 1,741 lung cancer cases and the same number ofmatched controls of smokers and nonsmokers in Norway,Boffetta and colleagues reported a statistically significantpositive association between serum cotinine levels and lungcancer risk in a dose-dependent manner (40). Among thosewith smoking information in the analytic sample, there were16% never, 17% former, and 67% current smokers. Comparedwith individuals who had serum cotinine levels �5 ng/mL,which included never smokers, smokers with 24.8 to 114.7ng/mL of serum cotinine (i.e., the third highest group ofcotinine) had an OR of 2.80 (95% CI, 1.55–5.05), and smokerswith >378.8 ng/mL of serum cotinine (i.e., the eighth orhighest group of cotinine levels in the study) had an OR of

Yuan et al.





Table 1. Summary of findings from prospective cohort studies for tobacco-constituent biomarkers andlung cancer risk among current smokers

Tobaccoconstituent

Metabolite/biomarker Study population

OR (95% CI) forhighest vs. lowestbiomarker level Comments Reference

Nicotine Total cotinine Men, Shanghai (476 casesand 476 controls)

3.52 (2.30–5.41) Specimen type: urine. ORwas adjusted for age, dateof specimen collection,neighborhood ofresidence, number ofcigarettesperday, numberof years of smoking, andurinary total NNAL andPheT.

Yuan andcolleagues (30)

Men and women,Singapore (91 casesand 93 controls)

1.65 (0.69–3.97) Specimen type: urine. ORwas adjusted for age, sex,date of specimencollection, dialect group,number of cigarettes perday, number of years ofsmoking, and urinary totalNNAL.


Men and women, UnitedStates (100 cases and100 controls)

0.85 (0.59–1.23) Specimen type: serum. ORis associated with a unitSD increase adjusted forsex, age, number of yearsof smoking, and urinaryNNAL and PheT

Church andcolleagues (41)

TSNA Total NNN Men, Shanghai (93 casesand 93 controls)

1.02 (0.39–2.89) Specimen type: urine. ORwas adjusted for numberof cigarettes per day,number of years ofsmoking, and urinary totalcotinine and PheT

Stepanov andcolleagues (46)

Total NNAL Men, Shanghai (476 casesand 476 controls)

1.93 (1.28–2.90) Specimen type: urine. ORwas adjusted for numberof cigarettes per day,number of years ofsmoking, and urinary totalcotinine and PheT


Men and women,Singapore (91 casesand 93 controls)

2.64 (1.10–6.34) Specimen type: urine. ORwas adjusted for age, sex,date of specimencollection, dialect group,number of cigarettes perday, number of years ofsmoking, and urinary totalcotinine.



1.57 (1.08–2.28) Specimen type: serum. ORis associated with a unitSD increase adjusted forsex, age, number of yearsof smoking, and urinarycotinine and PheT


(Continued on the following page)






55.1 (95% CI, 35.7–85.0) for lung cancer. These results werenot adjusted for smoking history (e.g., status, intensity, andduration) or other biomarkers of tobacco carcinogens. It isworth pointing out that although there is no definitive cutoffpoint for cotinine to separate smokers from nonsmokers(43), the most widely used cutoff point is 14 ng/mL in serumor 50 ng/mL in urine (44).

A stronger association between cotinine levels and lungcancer risk observed in the studies by Timofeeva andcolleagues and Boffetta and colleagues was due to theinclusion of never smokers and former smokers who sup-posedly had very low or undetectable levels of cotinine intheir body fluids given a relatively short half-life of cotinine.On the other hand, the studies by Yuan and colleagues(30, 39) and Church and colleagues (41) included onlycurrent smokers with certain elevated levels of cotinine inurine or serum, which resulted in a narrow range of cotinineand a flattened dose–response relationship for cotinine andlung cancer risk. In addition, the adjustment for smokingintensity and duration and additional biomarkers of tobac-co carcinogens (e.g., NNAL and PheT) could further atten-uate the association between cotinine and lung cancer risk.Nonetheless, a statistically significant association betweenurinary total cotinine and lung cancer risk even afteradjustment for these tobacco carcinogen biomarkers (i.e.,NNAL and PheT) in the Shanghai Cohort Study (30) suggeststhat cotinine, while not carcinogenic, is acting as a surro-gate for other important compounds in cigarette smoke

that have yet to be characterized in terms of lung cancerrisk in humans.

Total NNAL and total NNNA limited number of studies have examined levels of total

NNAL and risk of lung cancer in smokers. Using the nestedcase–control study of lung cancer within the Shanghai CohortStudy described above, we found that smokers who developedlung cancer had statistically significantly higher levels of totalNNAL in baseline urine samples than smokers who did notdevelop lung cancer (P< 0.001). The risk of lung cancer doubledfor smokers with the highest tertile of urinary total NNALrelative to the lowest tertile after adjustment for smokingintensity and duration and urinary total cotinine (OR, 2.04;95% CI, 1.02–4.05; Ptrend ¼ 0.04; ref. 39). In the follow-up studywith a larger sample size (476 cases and 476 controls) withinthe same Shanghai Cohort Study, we found a similar relativerisk for lung cancer associated with elevated levels of urinarytotal NNAL (OR, 1.93; 95% CI, 1.28–2.90) after adjustment forsmoking intensity and duration, and urinary total cotinine andPheT (30). We also conducted a case–control study nestedwithin the Singapore Chinese Health Study and found a similarresult; the multivariate-adjusted OR for lung cancer was 2.64(95% CI, 1.10–6.34) for the highest tertile versus the lowesttertile of total NNAL (39).

In an United States–based case–control study nested withinthe PLCO, total NNALwas associatedwith a 1.57 times increasein lung cancer risk per SD or 40 fmol/mL (P ¼ 0.018; ref. 41).

Table 1. Summary of findings from prospective cohort studies for tobacco-constituent biomarkers andlung cancer risk among current smokers (Cont'd )

Tobaccoconstituent

Metabolite/biomarker Study population

OR (95% CI) forhighest vs. lowestbiomarker level Comments Reference

PAH PheT Men, Shanghai (476 casesand 476 controls)

2.34 (1.33–4.11) Specimen type: urine. ORwas adjusted for numberof cigarettes per day,number of years ofsmoking, and urinary totalcotinine and total NNAL



1.23 (0.88–1.72) Specimen type: serum. ORis associated with a unitSD increase adjusted forsex, age, number of yearsof smoking, and urinarycotinine and total NNAL


VOC HPMA Men, Shanghai (343 casesand 392 controls)

1.06 (0.62–1.80) Specimen type: urine. ORwas adjusted for numberof cigarettes per day,number of years ofsmoking, and urinarycotinine

Yuan andcolleagues (48)SPMA 1.20 (0.74–1.96)

MHBMA 1.08 (0.66–1.75)HBMA 0.97 (0.56–1.66)HEMA 1.12 (0.66–1.89)

Abbreviations: HBMA, 4-hydroxybut-2-yl mercapturic acid; HEMA, 2-hydroxyethyl mercapturic acid; HPMA, 3-hydroxypropylmercapturic acid; MHBMA, monohydroxybutyl mercapturic acid; SPMA, S-phenyl mercapturic acid; TSNA, tobacco-specificnitrosamines.

Yuan et al.





In this study, total NNAL, PheT, and cotinine were measuredin the serum collected before lung cancer diagnosis among100 cases and 100 matched controls; all of them were currentsmokers consuming 10 or more cigarettes per day. The sta-tistically significant association for NNAL remained afteradjustment for number of years of smoking, serum cotinine,and PheT.Like NNK, NNN is produced by the nitrosation of alkaloids

specific to tobacco, therefore, NNN and NNK always occurtogether (14). Experimental studies have demonstrated thatNNN is a potent carcinogen to the oral cavity and esophagus,but a relatively weak carcinogen to the lung of rats (15, 45). Byusing the Shanghai Cohort Study database, we observed noassociation between urinary total NNN and lung cancer risk;the ORs (95% CIs) for the second and third tertiles of total NNNwere 0.82 (0.36–1.88) and 1.02 (0.39–2.89), respectively (Ptrend¼0.959), after adjustment for smoking intensity and duration,and urinary total cotinine (46). In the same cohort, we havepreviously observed and reported a strong, dose-dependentrelationship between urinary NNN and esophageal cancer riskin current smokers (47). The ORs (95% CIs) of esophagealcancer for the second and third tertiles of urinary total NNNwere 3.99 (1.25–12.7) and 17.0 (3.99–72.8), respectively, com-pared with the first tertile after adjustment for smokingintensity and duration, and urinary total cotinine and totalNNAL (Ptrend < 0.001). These results are strikingly coherentwith the findings of studies in F-344 rats demonstrating thatNNK is a carcinogen selective for the lung while NNN affectsthe esophagus and oral cavity.

PheTAs described above in the nested case–control study of lung

cancer in current smokers within the Shanghai Cohort Study,we quantified urinary levels of PheT in 476 lung cancer casesand the same number ofmatched controls to examine whetherPheT levels are associated with risk of lung cancer (30).Compared with the lowest quintile, ORs (95% CIs) for lungcancer in the second, third, fourth, and fifth quintiles of urinaryPheT were 1.70 (1.00–2.88), 1.07 (0.62–1.84), 1.48 (0.86–2.53),and 2.34 (1.33–4.11), respectively (Ptrend ¼ 0.023) after adjust-ment for number of cigarettes smoked per day, number of yearsof smoking, and urinary total cotinine and total NNAL. In thesame cohort, we also conducted a nested case–control of lungcancer in lifelong nonsmokers and found a statistically signif-icant positive association between urinary biomarkers of PAH,including PheT, and lung cancer risk in a dose-dependentmanner (29). These results also demonstrate a role of PAH inthe development of lung cancer independent of tobaccocarcinogens.Results from the case–control study nestedwithin the PLCO,

however, do not support the role of PAH in the development oflung cancer in smokers (41). Mean serum PheT levels werehigher in cases than controls, but the difference was notstatistically significant (P ¼ 0.204). Results from the multivar-iable logistic regression model also indicated a small, butstatistically nonsignificant association after adjustment forsmoking duration, serum total cotinine, and total NNAL (OR,1.23; 95% CI, 0.88–1.72, per one SD increase in PheT). The small

size of the study (100 case–control pairs), as well as the generaloccurrence of PAH in the diet and the environment (asopposed to tobacco-specific compounds such as nicotine,NNK, and NNN), may have contributed to the null results (41).

VOCsWe quantified urinary mercapturic acid metabolites of

acrolein, benzene, 1,3-butadiene, crotonaldehyde, and ethyleneoxide in addition to urinary biomarkers of PAH, NNK, andnicotine in 343 lung cancer cases and 392 matched controlswithin the Shanghai Cohort Study described above (48). Com-pared with the lowest quartiles, highest quartiles of all mea-sured mercapturic acids were associated with statisticallysignificant approximately 2-fold increased risk for lung cancer(all Ptrend < 0.01) after adjustment for smoking intensity andduration. These positive associations were completelyexplained by urinary total cotinine in addition to smokingintensity and duration. Therefore, mercapturic acid metabo-lites of these VOCs are not independent risk predictors of lungcancer among male smokers in Shanghai, China.

Multivariable model of lung cancer risk among smokersWe examined the joint effect of three urinary biomarkers—

total cotinine, total NNAL, and PheT on risk of lung cancer incurrent smokers. For example, smokers in the highest tertilesof urinary total NNAL and total cotinine exhibited an 8.5-fold(95% CI, 3.7–19.5) increased risk for lung cancer relative tosmokers with comparable smoking intensity and duration butpossessing the lowest tertiles of urinary total NNAL and totalcotinine (39). When we simultaneously examined the numberof cigarettes per day, number of years of smoking, urinary totalcotinine, PheT, and total NNAL in current smokers, all fivevariables were statistically significantly associated with anincreased risk of lung cancer (all Ptrend < 0.05; ref. 30). ORs(95% CIs) for lung cancer were 1.93 (1.27–2.93) for number ofpacks of cigarettes per day (20 cigarettes/pack) and 1.94 (1.52–2.47) for every 10 years of smoking (30). Among the threeurinary biomarkers, the multivariable-adjusted ORs (95% CIs)for lung cancer for one unit in natural logarithmic value (anequivalent of 2.7-fold increment) of total cotinine, total NNAL,and PheT was 1.64 (1.34–2.01), 1.28 (1.02–1.59), and 1.41 (1.02–1.93), respectively (30). In summary, these three biomarkerswere positively associated with lung cancer risk among currentsmokers, independent of smoking history.

DiscussionThe worldwide burden of lung cancer is expected to con-

tinue growing given that worldwide consumption of tobaccocontinues to increase and cigarette smoking will remain thesingle most important causal factor for lung cancer. However,only one quarter of deaths among male smokers are attributedto lung cancers over their lifetime (8). Interview-based assess-ment of smoking intensity and duration has limitation incapturing interindividual variability in uptake andmetabolismof tobacco carcinogens. In this review, we have demonstratedthat appropriately chosen biomarkers would have potential toimprove the assessment of lung cancer risk over amodel based






solely on self-reported history of smoking habits for currentsmokers.

A biomarker approach to evaluate the heterogeneity oflung cancer risk among smokers has advantages. The mea-surement of the biomarker is objective and provides a directlink between the parent compound and lung cancer risk. Thesebiomarkers, when validated for their direct relation withlung cancer, could be used as intermediate markers to assessthe effectiveness of cancer prevention strategies among smo-kers. Important among these could be the identification ofsusceptible smokers at a young age, when preventive inter-ventions are more likely to be effective.

The strengths and limitations of a specific biomarker arealso dependent on their characteristics of being tobacco-specific (e.g., without significant non-tobacco exposuresources), having a relatively long half-life (e.g., representativeof long-term exposure), and/or being a carcinogen or ametabolite of an established carcinogen (17). Among thefour biomarkers for which our data support being indepen-dent risk factors for cancer among smokers, total cotinine,NNAL, and NNN are excellent measures of nicotine, NNK,and NNN uptake, respectively. Given their specific source oftobacco, total cotinine and NNAL have been used to evaluateenvironmental tobacco smoke exposure in nonsmokers(43, 49, 50). On the other hand, the PAH biomarker, PheT,could be derived from multiple sources. PAH are ubiquitousenvironmental contaminants formed in all processes involv-ing incomplete combustion of organic matter (24, 51). Theubiquity of PAH in the environment and their lack oftobacco specificity could contribute to the relatively weakassociation between PheT and lung cancer risk in the PLCOstudy (41).

Of the four tobacco constituents, NNAL and its parentcompound NNK as well as NNN are carcinogenic (15). PheTis a metabolite of the noncarcinogenic PAH phenanthrene,but the metabolism of phenanthrene to PheT closely par-allels that of BaP, a strong PAH carcinogen capable ofinducing tumors of the lung and other tissues in rodents,and rated as carcinogenic to humans (24). Cotinine is themajor metabolite of nicotine, a noncarcinogenic constituentof tobacco smoke. However, each dose of nicotine deliveredfrom a cigarette is accompanied by at least 70 establishedcarcinogens. Thus, nicotine and its metabolites could be agood surrogate biomarker for the uptake of other carcino-gens in tobacco smoke for which biomarkers have yet to bedeveloped.

A biomarker with a longer half-life has advantages overshort-lived biomarkers in terms of being able to captureexposures that occurred in the more distant past. NNAL hasthe longest half-life (t1/2 ¼ 10 days to 3 weeks) of the fourbiomarkers. The positive association between total NNAL andlung cancer risk among the three study populations (30, 39, 41)support that notion. Although cotinine is a well-establishedbiomarker for nicotine due to its high and frequent uptake, itswell-known limitation is the inability to distinguish the sourceof nicotine from non-tobacco (e.g., nicotine patch) versustobacco products. Furthermore, the half-life of cotinine isshort as compared with that of NNAL.

In summary, no single biomarker of tobacco smoke con-stituents stands out as the best for stratifying smokers by risk oflung cancer. Our finding that urinary cotinine was positivelyassociated with lung cancer risk among smokers, after adjust-ment for urinary NNAL and PheT, as well as smoking durationand intensity, supports the hypothesis that other compoundsin tobacco smoke are also likely to play a role in the develop-ment of lung cancer in smokers. Additional research is requiredto develop and validate biomarkers that could be used topredict the risk of lung cancer for smokers. An effective setof noninvasive and predictivemarkers of lung cancer risk couldallow the identification of the relatively small fraction ofsmokers at very high risk for lung cancer. This would help totarget lung cancer screening and smoking cessation efforts tosmokers at extremely high risk. The current recommendationsfor lung cancer screening are for annual low-dose computedtomography (CT) among smokers 55 to 79 years of age andwith30 or more pack-years of smoking (52). Although these screen-ing efforts will play an important role in identifying lungtumors at their early stage when effective treatment optionsare available, the high false-positive detection rate hinders theapplication of this screening method to large populations ofsmokers (53). A metabolite panel in addition to smokinghistory information may be able to reduce the false-positiverate associated with the use of CT scanning alone to identifylung tumors among asymptomatic smokers.

A demonstrated utility of tobacco-constituent biomarkersthat has not been fully explored is their use as intermediateoutcomes in chemoprevention trials among smokers. Ourgroup and others have contributed to the body of laboratoryevidence that implicates NNK in lung carcinogenesis inhumans and 2-phenethyl isothiocyanate (PEITC) as a chemo-preventive agent against NNK-induced lung carcinogenesis(51, 54, 55). To elucidate the potentially causal relationshipbetween dietary isothiocyanates (ITC) and lung cancer pro-tection in humans, Hecht and colleagues assessed the changesin urinary total NNAL in 11 cigarette smokers followingquantified consumption of watercress, a rich source of PEITC.Similar to results in the rodent experiments (54), the con-sumption of watercress significantly enhanced urinary excre-tion of total NNAL, indicating the activation of a possible NNKdetoxification pathway (56). Therefore, the identified biomar-kers of lung cancer risk could be further developed as inter-mediate markers for the evaluation of cancer preventionstrategies that involve the detoxification of NNK in smokers.

Using a biomarker-based approach to measure metabolitesin human urine that reflect the in vivo dose to tobacco smokeconstituents and evaluate the relationship betweenmetabolitelevels and lung cancer risk has important implications inpublic health and in setting a policy that limits the contentsand levels of specific compounds in cigarettes, includingnicotine and tobacco carcinogens (e.g., tobacco-specific nitro-samines; refs. 9, 57). The technology is certainly available toreduce these carcinogens in tobacco products by selectingcertain types of tobacco leaves and by altering the agri-cultural, curing, and storage processes. For example, it hasbeen shown that levels of NNN and NNK in selected popu-lar brands of smokeless tobacco that were more recently

Yuan et al.





introduced into the market is about 6- and 8-fold lower thanthe popular traditional brands, respectively (58). Cigarettescontaining lower amounts of nicotine (i.e., 0.05 mg/cigarette)are associated with reduced carcinogen exposure and reducednicotine dependence (59). In addition to establishing regula-tion of cigarettes that set maximum allowable toxicant levels,additional studies are warranted to demonstrate the reductionof lung cancer risk in smokers who consume cigarettes or othertobacco products with reduced carcinogens and toxicants.A relatively new product, the electronic cigarette (e-ciga-

rette), is growing in popularity, particularly among youths. TheNational Youth Tobacco Survey by the Centers for DiseaseControl and Prevention found that the prevalence of e-ciga-rette use in high school students in the United States doubledin the recent 2 to 3 years from 4.7% in 2011 to 10% in 2012 (P <0.05; ref. 60). Currently e-cigarettes are unregulated, althoughthe U.S. Food and Drug Administration is being urged topropose regulations that will address the advertising, ingre-dients, and sale to minors of e-cigarettes. e-Cigarettes delivernicotine to the user in the form of a vapor. Nicotine, althoughnot carcinogenic per se, often contains nornicotine as animpurity, and nornicotine is easily converted to NNN endog-enously by reaction with salivary nitrite. Previous studies haveshown large increase of urinary total NNN in individuals afterthey used oral nicotine replacement therapy products such asnicotine gumor lozenge (61, 62). The increase of e-cigarette usein the United States, particularly among youths, and thepotential for endogenous formation of NNN from nicotinepresent in e-cigarette vapor suggests a need for future researchin which the values of urinary biomarker of tobacco-specificnitrosamines and nicotine metabolites could be used to quan-tify carcinogen exposure and possibly cancer risk associatedwith e-cigarette use.In summary, the data reviewed here demonstrate that

several tobacco smoke toxicant and carcinogen biomar-

kers—total cotinine, total NNAL, PheT, and total NNN—arenot only biomarkers of exposure but also are biomarkers ofcancer risk. On the basis of comparison of smokers whoeventually developed lung cancer versus their healthy counter-parts, total cotinine, total NNAL, and PheT were risk biomar-kers for lung cancer, whereas total NNN was a risk biomarkerfor esophageal cancer. These results provide several possiblenew directions by which tobacco smoke biomarkers could beused for lung cancer prevention. With future research devotedto identifying additional objective biomarkers of tobaccoconstituents, we hope to improve lung cancer risk assessment.Tobacco-constituent biomarkers can be explored as interme-diate endpoints in lung cancer prevention trials among smo-kers. Given the long latent period for cancer development,using changes in biomarker levels, for example, betweenintervention and control groups could provide results morequickly and at less cost than waiting for cancer outcomes.Finally, the use of these and newly discovered tobacco smoke–constituent biomarkers for the purpose of establishing targetsfor regulation of tobacco products is another direction thatdeserves emphasis in future research.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: J.-M. Yuan, S.S. HechtAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): J.-M. Yuan, L.M. Butler, I. StepanovWriting, review, and/or revision of the manuscript: J.-M. Yuan, L.M. Butler,I. Stepanov, S.S. Hecht

Grant SupportThese studies were supported by U.S. NIH grants R01 CA-129534, CA-144034,

CA-92025, and CA-81301.

Received November 4, 2013; revised December 9, 2013; accepted December 13,2013; published OnlineFirst January 9, 2014.

References1. WHO. WHO global report: mortality attributable to tobacco. Geneva,

Switzerland: World Health Organization; 2012. Available from: http://whqlibdoc.who.int/publications/2012/9789241564434_eng.pdf.

2. Centers for Disease Control and Prevention. Consumption of cigar-ettes and combustible tobacco—United States, 2000–2011. MMWRMorb Mortal Wkly Rep 2012;61:565–9.

3. Food and Agriculture Organization of the United Nations. Projectionsof tobacco production, consumption and trade to the year 2010.Rome, Italy: FAO; 2003. Available from: http://www.fao.org/docrep/006/y4956e/y4956e00.htm#Contents.

4. Guindon GE, Boisclair D. Past, current and future trends in tobaccouse. Economics of tobacco control paper no. 6. Health, Nutrition andPopulation (HNP) discussion paper. 2003 Feb. Available from: http://siteresources.worldbank.org/HEALTHNUTRITIONANDPOPULATION/Resources/281627-1095698140167/Guindon-PastCurrent-whole.pdf.

5. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA CancerJ Clin 2013;63:11–30.

6. JemalA,BrayF,CenterMM,Ferlay J,WardE, FormanD.Global cancerstatistics. CA Cancer J Clin 2011;61:69–90.

7. International Agency for Research on Cancer. Tobacco smoke andinvoluntary smoking. IARC monographs on the evaluation of carcino-genic risk to humans. Vol. 83. Lyon, France: IARC Scientific Publica-tions; 2004.

8. Thun MJ, Henley SJ, Calle EE. Tobacco use and cancer: an epidemi-ologic perspective for geneticists. Oncogene 2002;21:7307–25.

9. Hecht SS.Research opportunities related to establishing standards fortobacco products under the Family Smoking Prevention and TobaccoControl Act. Nicotine Tob Res 2012;14:18–28.

10. Hecht SS. Tobacco carcinogens, their biomarkers and tobacco-induced cancer. Nat Rev Cancer 2003;3:733–44.

11. Hukkanen J, Jacob P III, Benowitz NL. Metabolism and dispositionkinetics of nicotine. Pharmacol Rev 2005;57:79–115.

12. BenowitzNL,HukkanenJ, JacobP III. Nicotine chemistry,metabolism,kinetics and biomarkers. Handb Exp Pharmacol 2009:29–60.

13. Benowitz NL, Dains KM, Dempsey D, Herrera B, Yu L, Jacob P III.Urine nicotine metabolite concentrations in relation to plasmacotinine during low-level nicotine exposure. Nicotine Tob Res2009;11:954–60.

14. International Agency for Research onCancer. Smokeless tobacco andtobacco-specific nitrosamines. IARCmonographs on the evaluation ofcarcinogenic risks to humans. Vol. 89. Lyon, France: IARC ScientificPublications; 2007.

15. Hecht SS. Biochemistry, biology, and carcinogenicity of tobacco-specific N-nitrosamines. Chem Res Toxicol 1998;11:559–603.

16. Belinsky SA, Foley JF,White CM, AndersonMW,Maronpot RR. Dose–response relationship between O6-methylguanine formation in Clara






cells and induction of pulmonary neoplasia in the rat by 4-(methylni-trosamino)-1-(3-pyridyl)-1-butanone. Cancer Res 1990;50:3772–80.

17. Hecht SS. Human urinary carcinogen metabolites: biomarkers forinvestigating tobacco and cancer. Carcinogenesis 2002;23:907–22.

18. Stepanov I, Hecht SS. Tobacco-specific nitrosamines and their pyr-idine-N-glucuronides in the urine of smokers and smokeless tobaccousers. Cancer Epidemiol Biomarker Prev 2005;14:885–91.

19. Knezevich A, Muzic J, Hatsukami DK, Hecht SS, Stepanov I.Nornicotine nitrosation in saliva and its relation to endogenoussynthesis of N0-nitrosonornicotine in humans. Nicotine Tob Res2013;15:591–5.

20. Wu W, Zhang L, Jain RB, Ashley DL, Watson CH. Determination ofcarcinogenic tobacco-specific nitrosamines in mainstream smokefrom U.S.-brand and non-U.S.-brand cigarettes from 14 countries.Nicotine Tob Res 2005;7:443–51.

21. StantonMF,Miller E,WrenchC,Blackwell R. Experimental induction ofepidermoid carcinoma in the lungs of rats by cigarette smoke con-densate. J Natl Cancer Inst 1972;49:867–77.

22. HoffmannD,Wynder EL. A study of tobaccocarcinogenesis. XI. Tumorinitiators, tumor accelerators, and tumor promoting activity of con-densate fractions. Cancer 1971;27:848–64.

23. Rodgman A, Perfetti TA. The chemical components of tobacco andtobacco smoke. Boca Raton, FL: CRC Press; 2009.

24. International Agency for Research on Cancer. Some non-heterocyclicpolycyclic aromatic hydrocarbons and some related exposures. IARCmonographs on the evaluation of carcinogenic risks to humans. Vol.92. Lyon, France: IARC Scientific Publications; 2010.

25. Hecht SS, Carmella SG, Le KA,Murphy SE, Li YS, Le C, et al. Effects ofreduced cigarette smoking on levels of 1-hydroxypyrene in urine.Cancer Epidemiol Biomarkers Prev 2004;13:834–42.

26. Hecht SS, Chen M, Yagi H, Jerina DM, Carmella SG. r-1,t-2,3,c-4-Tetrahydroxy-1,2,3,4-tetrahydrophenanthrene in human urine: apotential biomarker for assessing polycyclic aromatic hydrocarbonmetabolic activation. Cancer Epidemiol Biomarker Prev 2003;12:1501–8.

27. Carmella SG, Chen M, Yagi H, Jerina DM, Hecht SS. Analysis ofphenanthrols in human urine by gas chromatography-mass spectrom-etry: potential use in carcinogen metabolite phenotyping. CancerEpidemiol Biomarkers Prev 2004;13:2167–74.

28. Carmella SG, ChenM, Han S, Briggs A, Jensen J, Hatsukami DK, et al.Effects of smoking cessation on eight urinary tobacco carcinogen andtoxicant biomarkers. Chem Res Toxicol 2009;22:734–41.

29. Yuan JM, Butler LM, Gao YT, Murphy SE, Carmella SG, Wang R, et al.Urinary metabolites of a polycyclic aromatic hydrocarbon and volatileorganic compounds in relation to lung cancer development in lifelongnever smokers in the Shanghai Cohort Study. Carcinogenesis. 2013Nov 25. [Epub ahead of print].

30. Yuan JM, Gao YT, Murphy SE, Carmella SG, Wang R, Zhong Y, et al.Urinary levels of cigarette smoke constituent metabolites are prospec-tively associated with lung cancer development in smokers. CancerRes 2011;71:6749–57.

31. Hecht SS, Yuan J-M, Hatsukami D. Applying tobacco carcinogen andtoxicant biomarkers in product regulation and cancer prevention.Chem Res Toxicol 2010;23:1001–8.

32. International Agency for Research on Cancer. 1,3-Butadiene, eth-ylene oxide and vinyl halides (vinyl fluoride, vinyl chloride and vinylbromide). IARC monographs on the evaluation of carcinogenic risksto humans. Vol. 97. Lyon, France: IARC Scientific Publications;2008.

33. International Agency for Research on Cancer. Some industrial che-micals. IARCmonographs on the evaluation of the carcinogenic risk ofchemicals to humans. Vol. 60. Lyon, France: IARC Scientific Publica-tions; 1994.

34. International Agency for Research on Cancer. Acrolein. IARC mono-graphs on the evaluation of carcinogenic risks to humans. Vol. 63.Lyon, France: IARC Scientific Publications; 1995.

35. International Agency for Research on Cancer. Dry cleaning, somechlorinated solvents andother industrial chemicals. IARCmonographson the evaluation of carcinogenic risks to humans. Vol. 63. Lyon,France: IARC Scientific Publications; 1995.

36. Ross RK, Yuan JM, Yu MC, Wogan GN, Qian GS, Tu JT, et al. Urinaryaflatoxin biomarkers and risk of hepatocellular carcinoma. Lancet1992;339:943–6.

37. Yuan JM, Ross RK, Wang XL, Gao YT, Henderson BE, Yu MC.Morbidity and mortality in relation to cigarette smoking in Shanghai,China. A prospective male cohort study. JAMA 1996;275:1646–50.

38. Ellard GA, de Waard F, Kemmeren JM. Urinary nicotine metaboliteexcretion and lung cancer risk in a female cohort. Br J Cancer 1995;72:788–91.

39. Yuan JM, Koh WP, Murphy SE, Fan Y, Wang R, Carmella SG, et al.Urinary levels of tobacco-specific nitrosamine metabolites in relationto lung cancer development in two prospective cohorts of cigarettesmokers. Cancer Res 2009;69:2990–5.

40. Boffetta P, Clark S, Shen M, Gislefoss R, Peto R, Andersen A. Serumcotinine level as predictor of lung cancer risk. Cancer EpidemiolBiomarker Prev 2006;15:1184–8.

41. Church TR, Anderson KE, Caporaso NE, Geisser MS, Le CT, Zhang Y,et al. A prospectively measured serum biomarker for a tobacco-specific carcinogen and lung cancer in smokers. Cancer EpidemiolBiomarker Prev 2009;18:260–6.

42. Timofeeva MN, McKay JD, Smith GD, Johansson M, Byrnes GB,Chabrier A, et al. Genetic polymorphisms in 15q25 and 19q13 loci,cotinine levels, and risk of lung cancer in EPIC. Cancer EpidemiolBiomarker Prev 2011;20:2250–61.

43. Avila-Tang E, Al-DelaimyWK, Ashley DL, Benowitz N, Bernert JT, KimS, et al. Assessing secondhand smoke using biological markers. TobControl 2013;22:164–71.

44. Jarvis MJ, Tunstall-Pedoe H, Feyerabend C, Vesey C, Saloojee Y.Comparison of tests used to distinguish smokers from nonsmokers.Am J Public Health 1987;77:1435–8.

45. Balbo S, James-Yi S, Johnson CS, O'Sullivan MG, Stepanov I, WangM, et al. (S)-N0-Nitrosonornicotine, a constituent of smokeless tobac-co, is a powerful oral cavity carcinogen in rats. Carcinogenesis2013;34:2178–83.

46. Stepanov I, Sebero E,Wang R, Gao YT, Hecht SS, Yuan JM. Tobacco-specific N-nitrosamine exposures and cancer risk in the ShanghaiCohort Study: remarkable coherencewith rat tumor sites. Int J Cancer.2013 Oct 31. [Epub ahead of print].

47. Yuan JM, Knezevich AD, Wang R, Gao YT, Hecht SS, Stepanov I.Urinary levels of the tobacco-specific carcinogen N0-nitrosonornico-tine and its glucuronide are strongly associated with esophagealcancer risk in smokers. Carcinogenesis 2011;32:1366–71.

48. Yuan JM, Gao YT, Wang R, Chen M, Carmella SG, Hecht SS. Urinarylevels of volatile organic carcinogenand toxicant biomarkers in relationto lung cancer development in smokers. Carcinogenesis 2012;33:804–9.

49. Hecht SS,Carmella SG, LeKA,MurphySE, Boettcher AJ, LeC, et al. 4-(Methylnitrosamino)-1-(3-pyridyl)-1-butanol and its glucuronides inthe urine of infants exposed to environmental tobacco smoke. CancerEpidemiol Biomarker Prev 2006;15:988–92.

50. Stepanov I,HechtSS,DucaG,Mardari I.Uptakeof the tobacco-specificlung carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone byMoldovan children. Cancer Epidemiol Biomarker Prev 2006;15:7–11.

51. Chung FL, Kelloff G, Steele V, Pittman B, Zang E, Jiao D, et al.Chemopreventive efficacy of arylalkyl isothiocyanates and N-acetyl-cysteine for lung tumorigenesis in Fischer rats. Cancer Res 1996;56:772–8.

52. Humphrey LL, Deffebach M, Pappas M, Baumann C, Artis K, MitchellJP, et al. Screening for lung cancer with low-dose computed tomog-raphy: a systematic review to update the U.S. preventive services taskforce recommendation. Ann Intern Med 2013;159:441–20

53. National Lung Screening Trial Research Team Aberle DR, Adams AM,Berg CD, Black WC, Clapp JD, et al. Reduced lung-cancer mortalitywith low-dose computed tomographic screening. N Engl J Med2011;365:395–409.

54. Hecht SS, Trushin N, Rigotty J, Carmella SG, Borukhova A, Akerkar S,et al. Complete inhibition of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone-induced rat lung tumorigenesis and favorable modificationof biomarkers by phenethyl isothiocyanate. Cancer Epidemiol Bio-marker Prev 1996;5:645–52.

Yuan et al.





55. Morse MA, Amin SG, Hecht SS, Chung FL. Effects of aromatic iso-thiocyanates on tumorigenicity, O6-methylguanine formation, andmetabolism of the tobacco-specific nitrosamine 4-(methylnitrosa-mino)-1-(3-pyridyl)-1-butanone in A/J mouse lung. Cancer Res1989;49:2894–7.

56. Hecht SS, Chung FL, Richie JP Jr, Akerkar SA, Borukhova A, Skow-ronski L, et al. Effects of watercress consumption on metabolism of atobacco-specific lung carcinogen in smokers. Cancer Epidemiol Bio-markers Prev 1995;4:877–84.

57. Burns DM, Dybing E, Gray N, Hecht S, Anderson C, Sanner T, et al.Mandated lowering of toxicants in cigarette smoke: a description of theWorld Health Organization Tobacco Regulation Proposal. Tob Control2008;17:132–41.

58. Hecht SS, Stepanov I, Hatsukami DK. Major tobacco companies havetechnology to reduce carcinogen levels but do not apply it to popularsmokeless tobacco products. Tob Control 2011;20:443.

59. Hatsukami DK, Kotlyar M, Hertsgaard LA, Zhang Y, Carmella SG,Jensen JA, et al. Reduced nicotine content cigarettes: effects ontoxicant exposure, dependence and cessation. Addiction 2010;105:343–55.

60. Morbidity and Mortality Weekly Report (MMWR). Notes from the field:electronic cigarette use among middle and high school students—United States, 2011–2012. MMWR Morb Mortal Wkly Rep 2013;62:729–30.

61. Stepanov I, Carmella SG, Briggs A, Hertsgaard L, Lindgren B, Hatsu-kami D, et al. Presence of the carcinogen N0-nitrosonornicotine in theurine of some users of oral nicotine replacement therapy products.Cancer Res 2009;69:8236–40.

62. Stepanov I, Carmella SG, Han S, Pinto A, Strasser AA, Lerman C,et al. Evidence for endogenous formation of N0-nitrosonornicotine insome long-term nicotine patch users. Nicotine Tob Res 2009;11:99–105.






2014;74:401-411. Published OnlineFirst January 9, 2014.Cancer Res Jian-Min Yuan, Lesley M. Butler, Irina Stepanov, et al. Risk of Lung Cancer

Constituent Biomarkers for Assessing−Urinary Tobacco Smoke

Updated version

10.1158/0008-5472.CAN-13-3178doi:

Access the most recent version of this article at:

Cited articles

http://cancerres.aacrjournals.org/content/74/2/401.full#ref-list-1

This article cites 48 articles, 21 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/74/2/401.full#related-urls

This article has been cited by 10 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/74/2/401To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-13-3178

http://cancerres.aacrjournals.org/content/74/2/401.full#ref-list-1

http://cancerres.aacrjournals.org/content/74/2/401.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/74/2/401


urinary tobacco smoke constituent biomarkers for assessing ... › content › canres › ... ·...

Documents