[CANCER RESEARCH 58. 4858-4863. November I. 1998]

Associations of CYP1A1, GSTMI, and CYP2E1 Polymorphisms with Lung CancerSuggest Cell Type Specificities to Tobacco Carcinogens1

LoïcLe Marchand,2 Lakshmi Sivaraman,3 Lisa Pierce, Ann Seifried, Annette Lum, Lynne R. Wilkens, and

Alan F. LauEtiology Program. Cancer Research Center of Hawaii. University tif Hawaii. Honolulu, Hawaii V6HI3

ABSTRACT

The dramatic shirt in the pathological presentation of lung cancer [theproportional decrease in squamous cell carcinoma (SCO and increase inadenocarcinoma (AC)] observed in the United States after the 1950s mayhave taken place as the result of the reduction in polycyclic aromatichydrocarbons (PAHs) and the increase in iV-nitrosamines in inhaledsmoke from filtered low-yield cigarettes. The predominant mutation pat

terns of these tumors also suggest differences in their etiology. We testedthe hypothesis that genetic susceptibility to PAHs, as determined bypolymorphisms in ()/'/ I / and GSTMI, predominantly causes lungSCCs,

and susceptibility to nitrosamines, as determined by polymorphisms inCYP2KÌ,predominantly causes lung ACs. CYP1A1 and GSTMI play amajor role in the metabolic activation and detoxification of PAHs, respectively, and CYP2EI plays a major role in the metabolic activation ofnitrosamines. We conducted a population-based case-control study among

341 incident lung cancer cases and 456 controls of Caucasian, Japanese, orHawaiian origin. In-person interviews collected detailed information on

lifestyle risk factors, and DNA extracted from peripheral leukocytes wasused in PCR-based genotyping assays. Logistic regression analyses were

used to compute odds ratios and 95% confidence intervals (CIs) for eachcell type, adjusting for smoking and dietary variables. The presence of atleast one copy of the CYPIAI Mspl variant alíelewas found to be associated with a 2.4-fold (95% CT, 1.2-4.7) increase in the risk of SCC whenthis gene was considered singly and a 3.1-fold (95% CI, 1.2-7.9) increase

in the risk of SCC when combined with a GSTMI deletion. No significantassociation was found between Mspl and all lung cancers or other celltypes or with the CYPIAI exon 7 polymorphism. In contrast, the CYP2EÃŒKsal and Dral polymorphisms were not clearly related to SCC risk, butthese homozygous variant genotypes were associated with a 10-fold (95%CI, 0.0-0.5) decrease in the risk of overall lung cancer i/f.wil variant) andAC (Ural variant) compared to the homozygous wild-type genotypes.

Inverse associations with these two closely linked CYP2EI polymorphismswere also suggested for small cell carcinoma. In agreement with pastexperimental and epidemiological data, the associations found in thisstudy between CYPIAI and lung SCC and between CYP2E1 and lung ACsuggest a certain specificity of tobacco smoke PAHs for lung SCC andtobacco-specific nitrosamines for lung ACs.

INTRODUCTION

A dramatic change in the histological presentation of lung cancerhas taken place behind the shroud of the 20th century lung cancerepidemic. In the past 50 years, AC,4 a rare tumor type at the turn of

the century, has replaced SCC as the most frequent lung cancerhistological type in many developed countries (1). Indeed, for some

Received 5/19/98: accepted 9/2/98.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore he hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1Supported in part hy Grant ROI-CA-55874 and Contract NOI-CN-05223 from theUnited Stales National Cancer Institute and by Grant EDT-78 from the American Cancer

Society.- To whom requests for reprints should be addressed, at Etiology Program. Cancer

Research Center of Hawaii. University of Hawaii. 1236 Lauhala Street. Suite 407.Honolulu. HI 96813. Phone: (808) 586-2988; Fax: (808) 586-2982: E-mail: loic@crch.

hawaii.edu.1Present address: Department of Cell Biology. Baylor College of Medicine. Houston.

TX 77030.4 The abbreviations used are: AC. adenocarcinoma: SCC. squamous cell carcinoma;

PAH. polycyclic aromatic hydrocarbon: CI. confidence interval; OR. odds ratio.

subgroups of the United States population, such as white men inConnecticut, the incidence of lung SCC has now peaked, whereas therate for lung AC is still increasing (2). It has been proposed thatchanges in cigarette composition and smoking behavior may havecaused this histological shift (2, 3). Until the 1950s, cigarettes werepredominantly unfiltered and high tar and produced smoke too irritating to permit deep inhalation. As a result, most smoke particulateswere deposited on the epithelium at the branches of central bronchi,where SCCs predominantly occur (3, 4). Smokers of more recentlymanufactured cigarettes have compensated for the lower nicotineyield by inhaling more deeply and smoking more intensely, resultingin an increased exposure of the peripheral lung, the site of origin ofmost ACs, to tobacco carcinogens (3, 4). To achieve lower tar,tobacco ribs and stems and more burley varieties were incorporatedinto the tobacco blend of American cigarettes (3, 4). This reformulation has led to a reduction in PAHs and an increase in nitrogen oxidesand /V-nitrosamines in the inhaled smoke (3, 4). Thus, it is possible

that the proportional decrease in lung SCCs may be related to areduction in the PAH exposure of the central bronchi, and the dramatic rise in lung ACs may be related to an increased exposure of theperipheral lung to tobacco-specific nitrosamines.

A great deal of experimental data also suggest etiological differences between the two main lung cancer cell types. Mutations in thep53 tumor suppressor gene are twice as common in SCCs (65%) as inACs (33%), with a greater proportion of these mutations being G to Ttransversions in SCCs (45%) than in ACs (23%; Ref. 5). Benzo-

(ii)pyrene the most potent of the PAHs, has been shown to preferentially form adducts at major p53 hot spots for G to T transversions inlung tumors (6), providing a direct mechanistic link between a PAHin tobacco smoke and a mutation that occurs most frequently in SCCs.Tobacco-specific nitrosamines are known to induce lung ACs in

rodents when injected systemically (7). Such tumors contain activatedK-rt/.s proto-oncogenes with a mutation in codon 12 (7). Similar DNA

alterations are commonly found in the lungs of smokers and in lungACs (7).

Both PAHs and nitrosamines require metabolic activation by cyto-

chrome P450 enzymes to exert their genotoxic effects. CYPIAI andCYP2E1 are of critical importance for the activation of PAHs andnitrosamines, respectively (8). GSTMI plays a major role in thedetoxification of PAH-activated intermediates (9). These enzymes

exhibit wide interindividual variability in their activity. Genetic polymorphisms thought to be linked to functional changes in these enzymes have been associated with lung cancer in the past, but withgreat inconsistency across studies (10-25). Because past studies wereoften small, and all used convenience samples or hospital-based

designs, these inconsistent findings may have been due to insufficientpower and/or selection bias. We also postulated that they may haveresulted from some underlying cell type specificity in these associations. We conducted a large population-based case-control study of

genetic susceptibility to lung cancer to test the hypothesis that susceptibility to PAHs, as determined by polymorphisms in CYPIAI andGSTMI, predominantly causes lung SCCs, and susceptibility to nitrosamines, as determined by polymorphisms in CYP2EI, predominantlycauses lung ACs.

4858

on June 20, 2018. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ASSOCIATIONS OF POLYMORPHISMS WITH LUNC, CANTER

MATERIALS AND METHODS

The human subjects protocol for this study was reviewed and approved bythe Committee on Human Studies of the University of Hawaii (Honolulu. HI)and by the institutional review board of each participating hospital. We alsoobtained written informed consent from all subjects.

Lung cancer patients were identified by the rapid-reporting system of the

Hawaii Tumor Registry, a member of the Surveillance. Epidemiology, and EndResults Program of the National Cancer Institute. Eligible cases were allpatients with histologically confirmed primary lung cancer who were diagnosed between January 1, 1992 and March 31. 1997 in all main medical centerson the island of Oahu, Hawaii. Other eligibility criteria included age between26 and 79 years. Oahu residency, no previous history of lung cancer, andappropriate ethnicity (>75% Japanese. >75% Caucasian, any Hawaiian/part-

Hawaiian heritage). Histological information was directly abstracted from eachpatient's pathology report. An interview was completed for 64% of the eligible

cases. The main reasons for nonparticipation were patient refusal (17%).physician refusal (2%). and death with the absence of a suitable surrogate forinterview (17%).

Controls were selected randomly from a list of Oahu residents interviewedby the State of Hawaii Department of Health as part of a health survey of a 2%random sample of state households. This source was supplemented withcontrols from Health Care Financing Administration participants on Oahu. Onecontrol was matched to each case on sex. ethnicity, and age (±2years). Theoverall participation rate for the controls was 62%. Reasons for nonparticipation included refusal (25%), inability to locate (10%). serious illness (1%), anddeath (2%). The analysis presented here was conducted with 341 cases (76%of interviewed cases) and 456 population controls (80% of interviewed controls) who donated a blood specimen for the study.

In-person interviews were conducted at the subjects' homes by trained

interviewers. On average, cases were interviewed within 4 months of diagnosis. The questionnaire included detailed demographic information, includingthe ethnic origin of each grandparent, a lifetime history of tobacco and alcoholuse. a quantitative food frequency questionnaire, and a personal history ofsecond-hand smoke, various relevant medical conditions and occupational

exposures, and a family history of lung diseases. Information was collected onthe types (nonfiltered cigarettes, filtered cigarettes, cigars, and pipes) oftobacco product ever smoked daily for at least 6 months and. for each tobaccoproduct, the usual amount per day. the age at which the subject startedsmoking, the overall duration of use. and. for ex-smokers, the age at which the

subject stopped smoking. We also inquired about any periods of smokingcessation for each tobacco product during the subject's life. Smokers were

considered current smokers if they smoked up to 1 year before the date ofdiagnosis for cases or up to the date of the interview for controls. For thisanalysis, cigarettes, pipes, and cigars were treated equivalently.

The food frequency questionnaire used in this study (26) has been previously validated in our population (27). Frequencies and the amounts consumedwere sought for 242 food items or categories. The reference period for thedietary questionnaire was the year before diagnosis for cases or the year beforethe interview for controls. Colored photographs of most food items showingthree different portion si/es, as well as measuring cups and spoons, were usedin the interview to facilitate the quantification of intakes. Subjects were alsoquestioned about the brand and dose of any vitamin supplements (includingmultivitamin pills) taken for a minimum of 3 months during the referenceperiod. The food composition data were based primarily on the United StatesDepartment of Agriculture nutrient database (28) and supplemented with datafrom other research and commercial publications.

Laboratory personnel were blinded to the case-control status of the subjects.

DNA was purified from peripheral blood lymphocytes by SDS/proteinase Ktreatment and phenol/chloroform extraction (29). The first of two CYPIAIpolymorphisms studied is a T to C transition 264 bp downstream from thepolyadenylate signal that creates a Mspl restriction site. Genotyping for thispolymorphism was carried out by PCR amplification using primers 5'-TAG-GAGTCTTGTCTCATGCCT-3' and 5'-CAGTGAAGAGGTGTAGCCGCT-3', followed by digestion with Mspl (11). We also studied an A to G transition

in exon 7 of the CYPIAI gene that results in the substitution of valine forisoleucine at residue 462. We assessed this polymorphism by the allele-specificPCR method described by Hirvonen et al. (13), in which primers 5'-AAGAC-CTCCCAGCGGGCAAT-3' and 5'-AAGACCTCCCAGCGGGCAAC-3' are

used with the opposite strand primer 5'-GAAAGGCTGGGTCCACCCTCT-3'

in subsequent PCR reactions. The specific conditions of these reactions weredescribed in Ref. 30.

We modified the assay developed by Kim et cil. (31) to investigate thepolymorphism in intron 6 of CYP2EÃŒdescribed by Uematsu et al. 119) with therestriction endonuclease Oral. Primers 5'-TCGTCAGTTCCTGAAAG-CAGG-3' and 5'-GAGCTCTGATGCAAGTATCGCA-3' were used in a PCR

reaction consisting of an initial denaturation step at 94°Cfor 2 min followedby 35 cycles of denaturation at 94°Cfor 1 min. annealing at 60°Cfor 1 min.and extension at 72"C for 2 min. with final annealing and extension steps at60°Cfor 1 min and 72°Cfor 10 min. After digestion with Oral, the PCR

product was resolved on an agarose gel.CYP2EI also exhibits several polymorphisms in the 5' flanking region of

the gene. One of them includes two distinct base substitutions that are ingenetic disequilibrium with each other and creates Rsal and Pst\ restrictionsites (32). We used primers 5'-TTCATTCTGTCTTCTAACTGG-3' and 5'-CCAGTCGAGTCTACATTGTCA-3' to amplify a region containing the two

distinct base substitutions as described in Ref. 30. To detect the deletion of theGSTMI gene locus, we amplified exons 6-7 of the gene using primers5'-GAACTCCCTGAAAAGCTAAAGC-3' and 5'-GTTGGGGTCAAATAT-ACGGTGG-3' (33). We coamplified a 268-bp fragment of the ß-globingeneas an internal standard using primers 5'-CAACTTCATCCACGTTCACC-3'and 5'-GAAGAGCCAAGGACAGGTAC-3'.

In the statistical analysis, we used )f statistics for homogeneity to test forcase-control differences in the distributions of the genotypes or other param

eters under study. Correlations between categorical variables were measuredby the <f>coefficient (34). Unconditional logistic regression (35) was used tocompute the ORs and 95% CIs. with adjustment for several covariates foundto be associated with risk (age, sex. and race, indicator variables; saturated fatand total vegetable intakes, continuous variables). The matched pair designwas not followed in this analysis, because often only one member donated ablood sample. Nutrient intakes were adjusted for caloric intake using themethod of residuals (36). Other dietary variables, such as the intake of specificcarotenoids. vitamin C. vitamin E. and fiber, were considered but were notretained in the final model because they were not as strongly associated withrisk as that of total vegetables. Several ways of modeling the smoking effectwere explored, including separate categorization for duration and amount, useof a pack-years and age-started term, logarithmic transformations of thevariables, and the addition of higher polynomial terms. The best-fitting model

was one that included an indicator variable for smoking status (ever, neversmoked) and separate continuous terms for duration, amount and (duration)2.

The log-likelihood ratio test was used to test the statistical significance of

modeled effects. We also used this test to determine the interaction amongcertain variables with respect to lung cancer risk. The test compared a maineffects, no interaction model with a fully parameterized model containing allpossible interaction terms for the variables of interest. Genetic trends weremodeled by assigning a value of 1. 2. or 3 to the genotype variable accordingto the subject's number of variant alíeles(zero, one, and two rare alíeles,

respectively).

RESULTS

Table 1 presents relevant characteristics of the lung cancer casesand population controls. No significant differences were found in theage, sex, and ethnic distributions of cases and controls. As expected,smoking was strongly associated with lung cancer risk. In addition,and also in agreement with past studies in this population (37. 38),cases consumed more saturated fat and less vegetables than didcontrols.

Among population controls, the frequency of the CYPIAI Msplvariant alíelewas 36.8% in Japanese, 8.9% in Caucasians, and 42.4%in Hawaiians. The frequency of the CYPIAI Ile-Val polymorphism

was 24.4% for Japanese, 3.4% for Caucasians, and 16.3% for Hawaiians. The frequency of the CYP2E1 Oral variant alíelewas 33.3% inJapanese, 6.6% in Caucasians, and 15.4% in Hawaiians. The frequency of the Rsal rare alíelewas 25.6% in Japanese, 2.0% inCaucasians, and 16.2% in Hawaiians. The frequency of the GSTMI

4859

on June 20, 2018. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ASSOCIATIONS OF POLYMORPHISMS WITH LUNG CANCER

Table I Characteristics of lung cancer cases and controls

Mean age(yr)SexMaleFemaleBlhnicilyCaucasianJapaneseHawaiianSmoking

statusNeverPaslCurrentMean

duration of smoking(yr)Meanamount(unils/day)'Meansaturated fat(g/day)'Mean

total vegetables (g/day)Cases

(n =341)64.5208(61.0)"133(39.0)141

(41.3)112(32.8)88

(25.8)32

(9.4)142(41.6)167(49.0)37.925.426.0276.9Controls

(n =456)65.1297

(65. 1)159(34.9)176(38.6)175

(38.4)105(23.0)168(36.8)223

(48.9)65(14.3)18.514.622.5338.5

" Values in parentheses are percentages.'' The amount includes cigarettes, cigars, and pipes.' Values were adjusted for calories by the method of residuals.

deletion was 54.6% ¡nJapanese, 60.0% in Caucasians, and 67.0% inHawaiians. The genotype distributions within each ethnic group wereall consistent with the Hardy-Weinberg equilibrium for Mendelian

inheritance.Tables 2 and 3 show the overall distribution of cases and controls

by genotypes for the CYPÌAÌand CYP2EI polymorphisms, respectively. The frequencies of the CYPIAl Mspl or exon 7 variant alíelesfor all cases, ACs, small cell carcinomas, and other cell types weresimilar to those for population controls. However, the Mspl variantalíelewas significantly more common in cases with SCC than incontrols (P = 0.005). The frequency of the exon 7 variant alíelewas

also elevated for SCC, but not significantly so. The Msp\ and exon 7polymorphisms were closely linked in this study (<£= 0.7;P = 0.001), as reported earlier for these ethnic groups (11, 13). In

contrast, the variant alíelesfor the CYP2E1 Oral and Rsal polymorphisms and, in particular, the homozygote variant genotypes were lesscommon in lung cancer overall. ACs, and. to a lesser extent, small cellcarcinomas, as compared to controls. As previously found for theseethnic groups ( 19, 32), the Oral and Rxal polymorphisms were closelylinked in our data (<i>= 0.9: P = 0.001).

Table 4 presents adjusted ORs and 95% CIs for all lung cancer andspecific cell types by CYPIAl or CYP2E1 genotype. The presence ofat least one copy of the CYPIAl Mxpl variant alíelewas associated

with an OR for SCC of 2.4 (95% CI, 1.2-4.7), with the risk increasingwith the number of rare alíeles(P = 0.02). This association was

observed in each ethnic group and was particularly strong in Hawaiians (Table 5), although these ethnic-specific results were based on

small numbers of cases. There was no association between CYPIAland overall lung cancer or AC. In contrast, the homozygous variantgenotypes for CYP2E1 Oral and Rsal were associated with a 5- and10-fold decrease in the overall risk of lung cancer, respectively,compared to the corresponding homozygous wild-type genotype (Ta

ble 4). The association with Oral was strongest for AC, and theassociation with Rsal was strongest for small cell carcinoma. Nosignificant association was observed between CYP2E1 and SCC. Thenumber of cases was too small to conduct an analysis stratified on raceforCYP2El.

The joint effects of CYPIAl, Mspl, and GSTMl were assessedseparately for two cell types (Table 6). Compared to individuals withthe homozygous wild-type CYPIAl Mspl genotype and at least one

GSTMl alíele,those with at least one Mspl rare alíeleand at least onefunctional GSTMl alíelehad a 1.9-fold increase risk of lung SCC.Individuals with both polymorphisms had a 3-fold increase in the risk

of SCC. The P for interaction was 0.49. No clear pattern was found inthe joint effects for AC risk. No association was found between SCCor AC and any other genotype combinations with GSTMl (data notshown). Overall, deletion of GSTMl alone was not associated withlung cancer (OR = 0.9; 95% CI, 0.6-1.2) or with any cell types (SCC,OR = 1.3 and 95% CI = 0.7-2.4; AC, OR = 0.8 and 95% CI = 0.5-1.2; small cell, OR = 1.6 and 95% CI = 0.8-3.2: other, OR = 0.9 and95% CI = 0.5-1.7).

DISCUSSION

In this population-based case-control study of genetic susceptibility

to lung cancer, the presence of at least one copy of the CYPIAl Msplvariant alíelewas found to be associated with a 2.4-fold increase in therisk of SCC if this gene was considered singly and a 3-fold increase

in the risk of SCC if it was combined with a GSTMl deletion. Nosignificant association was found with the closely linked CYPIAlexon 7 polymorphism, perhaps because of a lack of statistical powerdue to the low frequency of the Val alíele.In contrast, the CYP2EIOral and Rsa\ polymorphisms were not clearly related to SCC risk,but these homozygous variant genotypes were associated with a

Table 2 Distribution of cuses und controls by CYPIAl genotypes

CYPIAlMsplControlsLung

cancerSCCACSmall

cellcarcinomaOtherml

/ml250(55.1)183(54.0)28(37.8)99(61.1)28(54.9)28

(53.8)ml/m2160(35.2)121

(35.7)31(41.9)52(32.1)17(33.3)21

(40.4)m2/m244

(9.7)35(10.3)15(20.3)1

1(6.8)6(11.8)3

(5.8)P"0.940.0050.330.880.57A/A335

(74.0)263(78.0)52

(70.3)126(78.7)42(82.3)43

(82.7)CYPIAl

Ile-ValA/G105(23.2)68

(20.2)21(28.4)31

(19.4)8(15.7)8(15.4)G/G13(2.9)6(1.8)1(1.3)3(1.9)1

(2.0)1( 1.9)P0.340.500.450.180.39

' P for x~ 'est for comparison with controls.

Table 3 Distribution of cuses and controls by CYP2EÃŒgenotypes

CYP2EIOralControls

Lung cancerSCCACSmall cell carcinomaOtherDD306

(67.7)240(71.0)

51 (68.9)112(69.6)41 (80.4)36 (69.2)DC121

(26.8)93 (27.5)20 (27.0)48 (29.8)10(19.6)15(28.9)CC25

(5.5)5(1.5)3 (4.0)1 (0.6)01 (1.9)f0.010.87

0.030.160.53cl/cl338

(74.4)269 (79.8)

56 (75.7)127(79.4)45 (88.2)41 (78.9)CYP2EI

Rsa\cl/c2102(22.5)

66(19.6)17(23.0)33 (20.6)

5 (9.8)11 (21.1)c2/c214(3.1)

2 (0.6)1 (1.3)01 (2.0)0P0.02

0.710.060.090.41

' P for x~ tesl f°rcomparison with controls

4860

on June 20, 2018. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ASSOCIATIONS OF POLYMORPHISMSWITH LUNG CANCER

Table 4 ORs" (and 95% CIs) for lung cancer by CYPIAl and CYP2E1 genotypes

CYPAIMsplml/mlml/m2m2/m2CYPIAl

Ile-ValA/AA/GG/GCYP2EÌ

OralDDDCCCCYP2EI

Rsalcl/clcl/c2c2/c2All

(341/456)*1.013(0.9-1.9)1.2(0.6-2.2)P

=0.34'1.01.0(0.6-1.5)0.7

(0.2-2.3)P=0.731.01.0(0.7-1.6)0.2(0.1-0.7)/>

=0.121.00.8(0.5-1.3)O.I

(0.0-0.5)P= 0.03SCC

(74/456)1.02.3(1.1^.7)2.6(1.0-6.7)P

=0.021.0\

1.2(0.6-2.5)J1.01.2(0.6-2.4)0.7

(0.2-3.0)P=0.921.0)

0.7(0.3-1.5)JAC

(162/456)1.01.0(0.6-1.7)0.7(0.3-1.6)P

=0.601.00.9(0.5-1.4)1.01.1

(0.7-1.8)0.1(0.0-0.5)P

=0.101.00.8(0.4-1.3)Small

cell carcinoma(51/456)1.01.2(0.5-2.7)1.1

(0.4-3.5)P=0.751.00.7(0.3-1.8)1.0}

0.6(0.3-1.4)1.00.3(0.1-0.9)Other

(52/456)1.01.5(0.7-3.2)0.6

(0.2-2.5)P=0.941.00.7(0.3-1.6)1.01.0(0.5-2.1)0.3

(0.0-2.5)P=0.431.00.6(0.3-1.4)

" ORs were adjusted for age, sex, ethnicity, smoking status, years of smoking, (years of smoking)", number of cigarettes smoked/day, and saturated fat and total vegetable intakes.' Number of cases/number of controls.' P for the genetic trend.

10-fold decrease in the risk of overall lung cancer {Rsal variant) andAC (Oral variant) compared to the homozygous wild-type genotypes.

Inverse associations with these two closely linked CYP2EI polymorphisms were also suggested for small cell carcinoma.

Cigarette smoke contains hundreds of constituents that may play arole in carcinogenesis (4). Compounds that have received the mostinterest include PAHs, Af-nitrosamines, active oxygen species, alde

hydes, and metal. The exact contribution of these agents to lungcarcinogenesis in humans is unknown. However, mutational spectrumanalysis of lung tumors implicates certain patterns of p53 mutationwith specific cell types. For example, many more lung SCCs presenta G:C to T:A transversion in the p53 gene compared to ACs (5). Suchmutations are hallmarks of mutagenesis involving certain PAHs, suchas benzo(i/)pyrene. It has recently been shown that benzo-

(a)pyrenediolepoxide, the ultimate carcinogen from benzo(a)pyrene.preferentially forms DNA adducts at guanine positions in codons 157,248, and 273 of p53 (6). These same positions are the major p53mutational hot spots in lung cancers. This study provided a directetiological link between a PAH and a specific mutation that occurspreferentially in lung SCC. It was also previously noticed that when

benzo(a)pyrene and other PAHs were administered in the lungs ofrats, the tumors induced were almost exclusively SCCs (39).

Cytochrome P450 1A1 and glutathione S-transferase p. are, respec

tively, phase I and phase II enzymes involved in the metabolicactivation and detoxification of PAHs found in tobacco smoke. Theactivity of cytochrome P450 1A1 has been shown to vary greatly inthe lung tissue of different individuals (40, 41) as an inheritable trait(42) and as the result of induction by tobacco smoke (43). Higherlevels of aryl hydrocarbon hydroxylase activity (which is mediated byCYP1AÌ)have been reported in the lymphocytes of lung cancerpatients than in those of controls (44, 45), but this finding has beendifficult to replicate (46). Consequently, interest has focused oncommon genetic polymorphisms recently identified in the CYPIAland GSTM1 genes. The GSTMI polymorphism is a deletion of thegene and results in a loss of enzymatic activity. The two linkedCYPIAl polymorphisms (Mspl and Ile-Val) have been associatedwith an increase in enzymatic activity as well as inducibility (47—50).

The association between CYPÃŒAIand lung cancer risk has beenexamined in a number of case-control studies. However, these studies

have all used convenience samples or a hospital series and rarely have

Table 5 ORs" for luna SCC by CYPIAl Mspl genotype anil race

Genotypeml

/mlml/m2m2/m2„*9/70

9/805/24JapaneseOR

(95%CI)'1.01.4(0.4-4.3)

2.1 (0.5-8.4)P = 0.29''Caucasiann16/146

. 8 /29OR

(95%Cl)1.0

2.5 (0.8-7.7)Hawaiiann3/34 15/549/18OR

(95%CI)1.05.3(1.0-26.6)

8.8(1.3-60.7)P = 0.02

" ORs were adjusted for age, sex, ethnicity, smoking status, years of smoking, (years of smoking)2, number of cigarettes/day, and saturated fat and total vegetable intakes.' Number of cases/number of controls.' OR and 95% Cl.'' P for the genetic trend.

Table 6 ORs" far the combined effect of GSTMI and CYPIAl Msp\ on cell types of lung cancer

SCC AC

CYPIAl MsplGSTMIhml/ml

+ml /mlml/m2, m2/m2 +ml/m2. m2/m2 -«'•11/101

17/14716/8130/121OR

(95%CI)1.0I.I

(0.4-2.7)1.9(0.7-5.1)3.1 (1.2-7.9)n39/101

59/14734/8129/121OR

(95%CI)1.00.9(0.5-1.6)

1.2(0.6-2.3)0.7(0.4-1.4)

" ORs were adjusted for age. sex. ethnicity, smoking status, years of smoking, (years of smoking)", number of cigarettes/day, and saturated fat and total vegetable intakes.* +, at least one functional alíele;-. no functional GSTMI alíele.

Number of cases/number of controls.

4861

on June 20, 2018. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ASSÕX'IATIONS OF POLYMORPHISMS WITH LUNG CANl'HK

carefully adjusted risk estimates for smoking variables. Overrepresenta-

tion of the variant Mspl and Rsal alíelesin lung cancer patients was firstreported in Japan, a population with a high frequency of these alíeles(10,11). This observation has since been confirmed in a second Japanesestudy (12). In the Japanese data, the association was clearly stronger forSCC than for AC (12, 51, 52). In contrast, studies of CYP1A1 and lungcancer conducted in Caucasians have been mostly inconsistent, withsome early studies finding no association (13-15), and more recent ones

reporting an increased risk with the variant alíeles(16, 17). These discrepancies have been attributed to a lack of statistical power in some ofthese studies due to the rarity of the CYPIAI polymorphisms in Caucasians. Interestingly, studies that have assessed polymorphisms in bothphase I and phase II pathways have been more consistent, including thosein Caucasians, in finding elevated ORs for lung cancer in individuals withboth CYPIAI and GSTM1 polymorphisms (53-57). Again, the evidence

for such an association was strongest for SCC. Similarly, the evidence foran association between GSTMI (alone) and lung cancer has been mostsuggestive for lung SCC, particularly at moderate levels of smoking,compared with any other cell types (58). Thus, overall, the findings fromthe present study, which is the first using a population-based design, are

consistent with those of past studies in implicating the linked Mspl andIle-Val polymorphisms in CYPIAI. especially when combined with a

GSTMI deletion, in determining the genetic predisposition to lung SCC.Tobacco-specific nitrosamines are known to be strong inducers of lung

ACs in rodents when administered systemically (7). They have limitedeffect when applied topically, because they need to be activated in theliver or in the target organ. Consequently, exposure to the ultimatecarcinogens is expected to be greatest in highly vascularized areas, suchas the bronchioles and alveoli, compared with the less vascularized largebronchi (3). The same DNA and hemoglobin adducts produced in rodentshave also been detected at higher levels in smokers than in nonsmokers(59). These adducts have been associated with the development of lungAC in rodents, with frequent codon 12 mutations in the K-mx protooncogene (60). In humans, the same K-mx mutations are found in 30%

of lung ACs and are rarely found in SCCs (61).Cytochrome P450 2E1 plays an important role in the metabolic

activation of various nitrosamines. including several potent tobacco-specific procarcinogens (8). In addition, it effectively reduces dioxy-

gen to radical species, thus contributing to lipid peroxidation andoxidative stress (62). Although the Rsal polymorphism in the CYP2EIgene has been associated with increased transcription of a reportergene in vitro (32), phenotyping studies using the drug chlorzoxazoneas a metabolic probe have shown that individuals with the variant Rsalalíelehave a lower basal CYP2E1 activity, and that enzyme activityis less inducible by ethanol (63-64).5 Uematsu et al. (19) originally

reported a proportionally lower frequency of the Oral variant alíeleinJapanese lung cancer patients than in controls. Three studies conducted in Caucasians failed to reproduce this observation (20-22).

However, in these three studies, lung cancer cases also had a lowerfrequency of Rsal rare alíeles,although not significantly so comparedto that of controls (40). More in agreement with the Japanese results,a recent study found a strong increase in lung cancer risk associatedwith the wild-type Rsal genotype in Mexican Americans and a moremoderate risk increase for the wild-type Oral genotype in both Mex

ican and African Americans (23, 24). However, none of these studieswere population based or had sufficient power to distinguish the riskby cell type. A hospital-based study conducted in Japan failed to find

any association between the Rsal polymorphism and lung cancer,even though histológica! types of lung cancer and cigarette smoking

5 L. Le Marchand, G. R. Wilkinson, and L. R. Wilkcns. Genetic and dietary predictors

of CYP2E1 activity: a phenotyping study in Japanese Americans using chlor/oxazone.suhmiltcd l'or publication.

were taken into account (25). Our study, in contrast, found a stronginverse association between lung cancer and both CYP2EÃŒOral andRsal polymorphisms. This association was not observed with SCC.

A certain specificity of PAHs for lung SCC and of nitrosamines forlung ACs is not only consistent with our results and those of pastexperimental and epidemiological studies, it also consistent with thehistological shift that took place after the introduction of filtered andlow-yield cigarettes, which probably decreased smokers' exposure to

PAHs and increased their exposure to nitrosamines. The associationsobserved between genotypes and lung cancer in this study may alsocontribute to the lower lung cancer risk of Japanese smokers and theincreased lung cancer risk of Hawaiian smokers, compared to Caucasians, that we have documented in a previous case-control study in

Hawaii (65). However, a detailed examination of these relationshipswould require a much greater sample size.

Recently, additional polymorphisms have been described in theCYPIAI and CYP2EÃŒgenes (66, 67) that, if associated with a functional change of the gene or protein, may have introduced somemisclassification because they were not considered in this study.However, the effects of these polymorphisms are likely to be smallbecause they are rare (<2%), and their functional significance isunclear. As expected in a case-control study of such a lethal disease,

the participation rate in this study was somewhat low. If survival wasincreased in patients with the CYPIAI wild-type genotype and/or the

variant CYP2E1 genotype, our results might have been biased. Thus,it will be important to confirm these findings in a prospective study.

In summary, in agreement with past experimental and epidemiological data, the associations found in this study between CYPIAI andlung SCC and between CYP2EI and lung AC provide additionalevidence for a certain specificity of tobacco smoke PAHs for lungSCCs and of tobacco-specific nitrosamines for lung ACs.

ACKNOWLEDGMENTS

We thank the Hawaii Tumor Registry. Castle Medical Center, Kaiser-Permanente Medical Center. Kuakini Medical Center, Queen's Medical Center, Sträub

Clinic and Hospital. St. Francis Medical Center, Tripler Medical Center, andWahiawa General Hospital for collaboration. We also thank Ronette Hunt. Barbara Burden. Geraldine Kaneshiro. and Yun Oh Jung for technical assistance.

REFERENCES

1. Travis, W. D., Travis. L. B.. and Devesa. S. S. Lung cancer (published erratumappears in Cancer (Phila.). 75: 2979. 1995]. Cancer (Phila.). 74: 191-205. 1995.

2. Thun. M. J.. Lally. C. A.. Flannery. J. T.. Calle. E. E.. Flanders. W. D., and Heath.C. W.. Jr. Cigarette smoking and changes in the histopathology of lung eancer. J. Nail.Cancer Inst.. 89: 1580-1586. 1997.

3. Wynder. E. L.. and Hoffmann. D. Smoking and lung cancer: scientific challenges andopportunities. Cancer Res.. 54: 5284-5295. 1997.

4. United States Department of Health and Human Services. Reducing the HealthConsequences of Smoking: 25 Years of Progress. A Report of the Surgeon Generalpp. 89-8411. Rockville, MD: United States Department of Health and Human

Services. 1989.5. Greenblatt. M. S.. Bennett. W. P.. Hollstein. M., and Harris. C. C. Mutations in the

p53 tumor suppressor gene: clues to cancer etiology and molecular pathogenesis.Cancer Res.. 54: 4855-4878. 1994.

6. Denissenko. M. F.. Pao. A.. Tang. M-S.. and Pfeifer. G. P. Preferential formation of

benzo[<i|pyrene adducts at lung cancer mutational hotspots in p5J. Science (Washington DC). 274: 430-432. 1996.

7. Hoffmann. D.. Rivenson. A.. Murphy. S. E.. Chung. F-L.. Amin. S., and Hecht. S. S.Cigarette smoking and adenocarcinoma of the lung: the relevance of nicotine-derived/V-nitrosamines. J. Smoking Rei. Disord.. 4: 165-190. 1993.

8. Kadlubar. F. F.. and Hammons. G. J. Role of cytochrome P-450 in metabolism ofchemical carcinogens, in: F. P. Guengerich (ed.). Mammalian Cytochromes P-450.Vol. 2. pp. 81-130. Boca Raton FL: CRC Press, 1987.

9. Ketterer. B.. Harris, J. M.. Talaska. G.. Meyer. D. J.. Pénible.S. E., Taylor, J. B..Lang. N. P.. and Kadlubar. F. F. The human glutathione 5-transfera.se supergenefamily, its polymorphism, and its effects on susceptibility to lung cancer. Environ.Health Perspect., 98: 87-94. 1992.

10. Kawajiri. K.. Nakachi. K.. Imai. K., Yoshii. A.. Shinoda. N.. and Watanabe. J.Identification of genetically high risk individuals to lung cancer by DNA polymorphisms of the cytochrome P450IAI gene. FEES Lett., 263: 131-133. 1990.

4862

on June 20, 2018. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ASSIX'IAÕÕONS OF POLYMORPHISMS WITH LUNG CANCliR

11. Hayashi. S.. Watanabc. J.. Nakachi, K.. and Kawajiri. K. Genetic linkage of lungcancer-associatedMv/il polymorphisms with amino acid replacementin the hemehindingregion of the human cylochrome P450IAI gene.J. Bioehem.. 110: 407-411. 1991.

12. Okada. T.. Kawashirna, K.. Fukushi. S.. Minakuchi. T.. and Nishimura, S. Associationbetween a cytochrome P450 CYPIA1 genotype and incidence of lung cancer. Phar-macogenetics. 4: 333-340. 1994.

13. Hirvonen. A.. Husgafvcl-Pursiainen. K.. Karjalainen. A.. Andila. S.. and Vainio. H.Poim-mulalinnal Mspi and 1/e-Vnl polymorphisms closely linked in the CYPIAIgene: lack of association with susceptibility to lung cancer in a Finnish studypopulation. Cancer Epidemici. Biomark. Prev.. /.- 485-489, 1992.

14. Tefre. T.. Ryberg. D., Haugen. A.. Neben. D. W., Skaug, V., Brogger. A., andBorresen. A-L. Human CYPIAÌ(cytochrome P,450) gene: lack of association between the Afv/il restriction fragment length polymorphism and incidence of lungcancer in a Norwegian population. Pharmaeogeneties. /: 20-25. 1991.

15. Shields. P. G.. Caporaso. N. E.. Falk. R. T.. Sugimura. H.. Trivers. G. E.. Trump.B. F.. Hoover, R. N., Weston. A., and Harris. C. C. Lung cancer, race, and a CYPIAIgenetic polymorphism. Cancer Epidemiol. Biomark. Prev.. 2: 481-485, 1993.

16. Drakoulis. N.. Cascorbi. I.. Brockmoller. J.. Gross. C. R.. and Roots. I. Polymorphisms in the human CYPIAI gene as susceptibility factors for lung cancer: exon 7mutation (4889 A to G) and a T to C mutation in the .V-flanking region. Clin.Investig.. 72: 240-248. 1994.

17. Xu. X.. Kelsey. K. T.. Wiencke, J. K.. Wain. J. C.. and Christiani. D. C. CytochromeP450 CYPIAI M.v/)Ipolymorphism and lung cancer susceptibility. Cancer Epidemiol.Biomark. Prev.. 5; 687-692. 1996.

18. Ishibe. N.. Wiencke. J. K.. Zuo. Z-F.. McMillan. A.. Spitz. M.. and Kelsey. K. T.Susceptibility to lung cancer in light smokers associated with CYPIAI polymorphisms in Mexican- and African-Americans. Cancer Epidemiol. Biomark. Prev.. 6:1075-1080. 1997.

19. Uematsu. F.. Kikuchi. H.. Motomiya, M.. Abe. T., Sagami. I.. Ohmachi. T.. Wakui.A., Kanamaru. R.. and Watanabe. M. Association between restriction fragment lengthpolymorphism of the human cytochrome P45QI1EI gene and susceptibility to lungcancer. Jpn. J. Cancer Res.. 82: 254-256, 1991.

20. Hirvonen, A.. Husgafvel-Pursiainen.K.. Anttila. S.. Karjalaincn. A., and Vainio. H. Thehuman CYP2E1 gene and lung cancer: Oml and Rsu\ restriction fragment length polymorphisms in a Finnish study population. Carcinogenesis(Lond. ). 14: 85-88. 1993.

21. Persson. I.. Johansson. !.. Bergung. H., Dahl. M-L.. Seidegard. J.. Rylander. R..Rannug. A.. Hogberg, J.. and Sundberg. M. I. Genetic polymorphism of cytochromeP4502EI in a Swedish population: relationship to incidence of lung cancer. FEBSLett.. 319: 207-211. 1993.

22. Kato. S.. Shields. P. G.. Caporaso. N. E.. Sugimura. H.. Trivers. G. E.. Tucker. M. A..Trump. B. F.. Weston. A-, and Harris. C. C. Analysis of cytochrome P450 2EI geneticpolymorphisms in relation to human lune cancer. Cancer Epidemiol. Biomark. Prev..3: 515-518. 1994.

23. Wu. X.. Shi, H., Jiang. H.. Kemp. B.. Hong. W. K.. Delclos. G. L. and Spit/. M. R.Associations between cytochrome P4502E1 genotype, mutagen sensitivity, cigarettesmoking and susceptibility to lung cancer.Carcinogenesis(Lond.). 18: 967-973. 1997.

24. Wu. X. F.. Amos. C. I.. Kemp. B. L.. Shi. H. H.. Jiang, H.. Wan. Y.. and Spitz. M. R.Cytochrome P450 2E1 Dru\ polymorphism in lung cancer in minority populations.Cancer Epidemiol. Biomark. Prev.. 7: 13-18. 1998.

25. Watanabe. J.. Yang. J-P.. Eguchi. H.. Hayashi. S.. Imai. K.. Nakachi. K., andKawajiri. K. An Rsal polymorphism in the CYP1EI gene does not affect lung cancerrisk in a Japanesepopulation. Jpn. J. Cancer Res.. <V6:245-248. 1995.

26. Hankin. J. H. A diet history method for research, clinical and community use. 23rdLenna Frances Cooper Memorial Lecture. J. Am. Diet. Assoc.. 86: 868-872. 1986.

27. Hankin. J. H.. Wilkens. L. R.. Kolonel. L. N.. and Yoshizawa. C. N. Validation of aquantitative diet history method in Hawaii. Am. J. Epidemiol.. 133: 616-628. 1991.

28. United States Department of Agriculture. Nutrient Database for Standard Reference.Release 10. Bethesda. MD: Human Nutrition Information Service. United StatesDepartment of Agriculture. 1992.

29. Maniatis. T.. Frisch. E. F.. and Sambrook. K. J. Molecular Cloning: A LaboratoryManual, pp. 458-459. Cold Spring Harbor. NY: Cold Spring Harbor Laboratory. 1982.

30. Lum. A., and Le Marchand. L. A simple mouthwash method for obtaining genomicDNA in molecular epidemiological studies. Cancer Epidemiol. Biomark. Prev.. 7:719-724. 1998.

31. Kim. R. B.. O'Shea. D.. and Wilkinson. G. R. Relationship in healthy subjects

between CYP2E1 genetic polymorphisms and the 6-hydroxylation of chlorzoxasone:a putative measure of CYP2EI activity. Pharmacogenelics. 4: 162-165. 1994.

32. Hayashi. S.. Watanabe. J.. and Kawajiri, K. Genetic polymorphisms in the 5'-flanking

region change transcriptional regulation of the human cytochrome P450IIEI gene.J. Biochem.. 110: 559-565. 1991.

33. Bell. D. A.. Thompson. C. L.. Taylor. J.. Miller. C. R.. Perera. F.. Hsieh. L. L., andLucier. G. W. Genetic monitoring of human polymorphic cancer susceptibility genesby polymerase chain reaction: application to glutathione transferase p. Environ.Health Perspcct.. !W: 113-117. 1992.

34. Fleiss. J. L. Statistical Methods for Ratesand Proportions. New York: John Willey &Sons, 1981.

35. Breslow. N. E.. and Day. N. E. Statistical Methods for Cancer Epidemiology. TheAnalysis of Case-Control Studies. Vol. 1. Lyon. France: IARC. 1980.

36. Willett. W. C., and Stampfer, M. J. Total energy intake: implications for epidemiologie analyses. Am J. Epidemiol.. 124: 17-27, 1986.

37. Le Marchand, L.. Yoshizawa. C. N.. Kolonel. L. N.. Hankin. J. H., and Goodman.M. T. Vegetable consumption and lung cancer risk: a population-based case-controlstudy in Hawaii. J. Nail. Cancer Insl.. SI: 1158-1164. 1989.

38 Goodman. M. T.. Hankin. J. H.. Wilkens. L. R.. and Kolonel. L. N. High-fat foods andthe risk of lung cancer. Epidemiology. 3: 288-299. 1992.

39. Deutsch-Wenzel. R.. Brune. H.. Grimmer. G.. Dettharn. G.. and Misfeld. J. Experimental studies in rat lungs on the carcinogenicity and dose-responserelationships ofeight frequently occurring environmental polycyclic aromatic hydrocarbons. J. Nati.Cancer Inst.. 71: 539-544. 1983.

40. Rannug. A.. Alexandrie. A-K., Persson, I., and Ingelman-Sundberg. M. Geneticpolymorphism of cytochromes P450 IA1. 2D6 and 2EI : regulation and lexicologicalsignificance. J. Occup, Environ, Med.. 37: 25-36. 1995.

41. Sabadie. N.. Richter-Reichhelm. H. B.. Saracci. R.. Mohr. U.. and Bartsch. H. Inter-individual differences in oxidalive benzoinIpyrcnc metabolism by normal and tumoroussurgical lung specimensfrom 105lung cancerpatients.Int. J. Cancer.27:417-425. 1981.

42. Nowak. D.. Schmidt-Preuss. U.. Jorres. R.. Liebke. F.. and Rudiger. H. W. Formationof DNA adducls and water-soluble metabolites of benzodopyrcne in human rnono-cyles is genetically controlled. Int. J. Cancer. 41: 169-173. 1988.

43. McLemore. T. L.. Adelberg. A.. Liu. M. C.. McMahon. N. A.. Yu. S. J.. Hubbard,W. C..Czerwinski. M.. Wood. T. G.. Storeng. R.. Lubek. R. A.. Eggleson.J. C.. Boyd. M. R..and Hiñes.R. N. Expressionof CYPIAI gene in patientswith lung cancer: evidence lorcigarette smoke-induced gene expression in normal lung tissue and for altered generegulation in priman pulmonary carcinomas.J. Nail. Cancer Inst.. 82: 1333-1339. 1990.

44. Kouri. R. E.. McKinney. C. E.. Slomiany. D. J.. Snodgrass. D. R.. Wray. N. P.. andMcLemore. T. L. Positive correlation between high aryl hydrocarbon hydroxylaseactivity and primary lung cancer as analyzed in cryoprcscrved lymphocytes. CancerRes.. 42: 5030-5037. 1982.

45. Hawke. L. J.. and Farrell. G. C. Increased binding of benzo(</)pyrenemetabolites tolymphocytes from patients with lung cancer. Cancer Leu.. 30: 289-297. 1986.

46. Law. M. R. Genetic predisposition to lung cancer. Br. J. Cancer. 61: 195-206,1990.

47. Pelersen. D. D.. McKinney. C. E., Ikeya. K.. Smith. H. H.. Bale. A. E.. McBridc.O. W.. and Neben. D. W. Human CYPIAI gene: eosegregation of the en/.ymeinducihility phenolype and an RFLP. Am. J. Hum. Genet.. 48: 720-725. 1991.

48. Landi. M. T.. Benazzi. P. A.. Shields. P. G.. Clark. G.. Lueier. G. W.. Gane. S. J..Cosma. G.. and Caporaso. N. E. Association between CYPIAI genotype. mRNAexpression and enzymatic activity in humans. Pharmacogenelics. 4: 242-246. 1994.

49. Crofts. F.. Taioli. E.. Trachman. ¡..Cosma. G. N.. Currie. D.. Toniolo. P.. and Garte.S. J. Functional significance of different human CYPIAI genotypes. Carcinogenesis(Lond.). IS: 2961-2963. 1994.

50. Kiyohara. C.. Hirohata. T.. and Inutsuka. S. The relationship between aryl hydrocarbon hydroxylase and polymorphisms of the CYPIAI gene. Jpn. J. Cancer Res.. 87:18-24. 1996.

51. Nakachi. K.. Imai. K.. Hayashi. S.. Watanabe. J.. and Kawajiri. K. Genetic susceptibility to squamouscell carcinoma of the lung in relation to cigarette smoking dose.Cancer Res.. 5/: 5177-5180. 1991.

52. Nakachi. K.. Hayashi. S.. Kawajiri. K.. and Imai. K. Association of cigarette smokingand CYPIAI polymorphisms with adenocarcinomas of the lung by grades of differentiation. Carcinogenesis (Lond.). 16: 2209-2213. 1995.

53. Hayashi. S.. Watanahe. J.. and Kawajiri. K. High susceptibility to lung canceranalyzed in terms of combined genotypes P450IAI and (¿-classglutathione S-transferase genes. Jpn. J. Cancer Res.. X3: 866-870. 1992.

54. Nakachi. K.. Imai. K., Hayashi. S.. and Kawajiri. K. Polymorphisms of the CYPIAI andglutathione .S-transferasegenesassociatedwith susceptibility to lung cancer in relation tocigarette dose in a Japanesepopulation. Cancer Res..53: 2994-2999. 1993.

55. Kihara. M.. Kihara. M.. and Noda, K. Risk of smoking for squamous and small cellcarcinomas of the lung modulated by combinations of CYPIAI and GSTMI genepolymorphisms in a Japanesepopulation. Carcinogenesis(Lond.). 16: 2331-2336. 1995.

56. Alexandrie, A. K.. Sundberg, M. I., Seidegard. J.. Tomling. G.. and Rannug. A.Genetic susceptibility lo lung cancer with special emphasis on CYPIAI and GSTMI:a study on host factors in relation to age at onset, gender and histológica!cancer types.Carcinogenesis (Lond.). 15: 1785-1790. 1994.

57. Garcia-Closas. M.. Kelsey. K. T.. Wiencke. J. K.. Xu. X.. Wain. J. C.. and Christiani.D. C. A case-control study of cylochrome P4501AI. glulathione 5-transferase Ml.cigarette smoking and lung cancer susceptibility (Massachusetts. United Stales).Cancer CausesControl. 8: 544-553. 1997.

58. London. S. J.. Daly. A. K.. Cooper. J.. Navidi. W. C.. Carpenter. C. L.. and Idle. J. R.Polymorphism of glutathione 5-transferase M l and lung cancer risk in African-Americans and Caucasians in Los Angeles County. California. J. Nati. Cancer Inst..87: 1246-1253. 1995.

59. Cannella. S. G.. Kagan. S. S.. Kagan. M.. Foiles. P. G.. Palladino. G.. Quart. A. M..Quart. E.. and Hecht. S. S. Mass spectromelric analyses of tobacco-specific nilrosa-mine hemoglobin adducts in snuff dippers, smokers and nonsmokers. Cancer Res..50: 5438-5445. 1990.

60. Ronai. Z. A.. Gradia. S.. Peterson, L. A., and Hechl, S. S. G to A transition and G toT transversions in codon 12 of the Ki-ra.v oncogene isolated from mouse lung tumorsinduced by 4-(methylnitrosamino)-l-(3-pyridyl)-l-butanonc (NNK) and related DNAmethylating and pyridyloxobutylating agents. Carcinogenesis (Lond.). 14: 2419-

2422. 1993.61. Rodenhuis. S.. and Slesbo.s.R. J. Clinical significance of nix oncogene activation in

human lung cancer. Cancer Res.. 52: 2665s-2669s. 1992.62. Ekstrom. G.. and Ingclman-Sundberg. M. Rat liver microsomal NADPH-supported

oxidase activity and lipid pcroxidation dependent on cthanol-inducible cytochromeP450 (P-450 I1E1). Biochem. Pharmacol.. 38: 1313-1319, 1989.

63. Lucas.D.. Menez.C.. Girre. C.. Berthou. F.. Bodcncz.P..Joannei.!.. Hispard. E.. Bardou.L-G-. and Menez,J-F. Cytochrome P4502EI genotypeandchlorzoxazonemetabolism inhealth) and alcoholic Caucasiansubjects.Pharmacoeenetics,5: 298-304, 1995.

64. Kim. R. B.. Yamazaki. H.. Chiba. K.. O'Shea. D.. Mimura. M.. Guengerich. F. P.,

Ishizaki. T.. Shimada. T.. and Wilkinson. G. R. In rm* and in vitro characterizationofCYP2E1 activity in JapaneseandCaucasians.J. Pharmacol.Exp. Ther.. 279:4-11. 1996.

65. Le Marchand. L.. Wilkens. L. R., and Kolonel. L. N. Ethnic differences in the lung cancerrisk associatedwith smoking. Cancer Epidemiol. Biomark. Prev.. /: 103-107. 1992.

66. Cascorbi. I.. Brockmoller. J.. and Roots. I. A C4887A polymorphism in exon 7 ofhuman CYPIAI: population frequency, mutation linkages, and impact on lung cancersusceptibility. Cancer Res.. 56: 4965-4969. 1996.

67. Hu. Y.. Oscarson, M.. Johansson.L. Yue. Q-Y.. Dahl. M-L.. Tabone. M., Arinco. S..Albano. E.. and Ingelman-Sundberg. M. Genetic polymorphism of human CYP2EI:characterization of two variant alíeles.Mol. Pharmacol.. 51: 370-376. 1997.

4863

on June 20, 2018. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1998;58:4858-4863. Cancer Res   Loïc Le Marchand, Lakshmi Sivaraman, Lisa Pierce, et al.   Carcinogenswith Lung Cancer Suggest Cell Type Specificities to Tobacco

PolymorphismsCYP2E1, and CYP1A1, GSTM1Associations of

  Updated version

  http://cancerres.aacrjournals.org/content/58/21/4858

Access the most recent version of this article at:

   

   

   

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

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

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

  Permissions

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

.http://cancerres.aacrjournals.org/content/58/21/4858To request permission to re-use all or part of this article, use this link

on June 20, 2018. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from