Vol. 1, 505-51 1, S(’ptemher/Oc(oher 1992 Cancer Epidemiology, Biomarkers & Prevention 505

Review

Host Factors in Lung Cancer Risk: A Review ofInterdisciplinary Studies

Christopher I. Amos,’ Neil E. Caporaso, and AinsleyWeston

Division of Biostatisti 5 Research and Informatics, International Agencyfor Research on Cancer, Lyon, France [C. I. A.], and Genetic Studies

Section, National Institute 01 Arthritis, Musculoskeletal and SkinDiseases [C. I. A.], Environnienlal Epidemiology Branch [N. E. C.], andLaboratory of Human Carcinogenesis [A. W.], National Cancer Institute,

Bethesda, Maryland 20892

Abstract

Host-specific factors influence risk for lung cancer. Afew case-control and family studies of lung cancersusceptibility allowed for known lung cancercarcinogens and showed strong familial clustering withsome evidence for a codominantly acting major gene.Cytochrome P-450 enzymes (e.g., CYP1A 1) activatemany carcinogens in tobacco smoke but have showninconsistent associations with risk for lung cancer.Case-control studies that assess the effects of CYPIID6on lung cancer risk have consistently shown a mildlydecreased risk for lung cancer among poormetabolizers. Cell surface markers have shown littlerelation to risk for lung cancer. Studies involving DNAor hemoglobin adducts, sister chromatid exchange, oroncogene activation only indirectly measure host-specific risk, and these assays have suffered from poorreproducibility and high cost. We describeepidemiological designs to assess specific geneticfactors that may alter lung cancer risk.

Introduction

Lung carcinogenesis in humans usually requires exposureto environmental agents including inhalation of tobaccosmoke (1), radioactive ores, metals, asbestos, and petro-chemicals (2). Studies of host-environment interactionscan identify groups of individuals who are at the greatestrisk and may help in identifying oncogenic mechanisms.Knowledge of mechanisms that lead to lung cancer mayhelp to identify early stages of carcinogenesis and allowthe possibility of better-targeted intervention strategies.

Family studies and case-control designs have beenused to assess host factors in lung cancer risk. Familystudies collect the same information on all individualswithin a specified sampling scheme, according to whichfamilies of various sizes and structures may be selected.Case-control studies, on the other hand, are typically

Ree’ived 9/18/9 1.

‘To whom requests for reprints should be addressed, at Genetic StudiesSection, Laboratory of Skin Biology, National Institute’ of Arthritis, Mus-

cuboskebetal and Skin Diseases, NIH, Building 6, Room 429, Bethesda,MD 20892.

designed to collect extensive information from a bargeset of cases and controls and generally collect only lim-ted information on relatives of the cases and controls. A

few case-control studies followed systematic rules forcollecting family information and therefore representboth case-control and family studies. The purpose of thisreview is to consolidate findings from these two usuallydistinct approaches (family and epiderniobogical studies)to facilitate the future design of studies. The design offuture studies will depend upon attributes of the studiedbiornarken as well as characteristics of the assay for thebiornarker. We first discuss studies to assess the familial-ity of lung cancer and familial associations of lung cancerwith other diseases; then we discuss associations of spe-cific biomarkers with lung cancer risk.

Familial Clustering of Lung Cancer

Familiality of Lung Cancer. Familiality of lung cancer hasbeen examined in case-control studies, genealogy-basedsearches for familiality, through segregation analysis, andthrough twin studies. In establishing farniliality, measuresof smoking and environmental exposures must be in-cluded because these exposures tend to be correlatedamong family members (3, 4). Failure to allow for come-bated environmental exposures may falsely lead to evi-dence for famibiality. For this reason, results from geneab-ogy-based studies of familiality (5, 6) for lung cancer willnot be considered here. Studies of farniliality have notassessed the effects of passive smoking, although caseswere found to have a 1 .4-fold increased risk for maternalexposure after allowance for personal smoking measures(7).

Familiality of lung cancer has been examined in case-control studies by comparing the recurrence risk (i.e., theprobability that a relative is affected by the same cancer)among first-degree relatives in case families to that ob-served in control families. Tokuhata and Lilbienfeld (3)demonstrated excess lung cancer mortality among therelatives of 270 lung cancer patients compared with 270age-, race-, sex-, and location-matched controls and theirrelatives. Smoking was a more important determinant ofrisk for men, while family history of lung cancer was themore important risk factor in women. Smoking and afamily history had synergistic effects. Smoking relativesof cases had a relative risk of 2-2.5 for mortality fromlung cancer compared with smoking relatives of controls.Nonsmoking relatives of lung cancer cases were also athigher risk for lung cancer than nonsmoking relatives ofcontrols; and, compared with controls, a significant ex-cess of noncancer respiratory illnesses was observedamong case relatives, but this excess was not correlatedwith smoking status in the relatives.

on November 22, 2020. © 1992 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

506 Review: Lung Cancer Ecogenetics

More recent studies (8, 9) have used retrospectivecase-control designs to assess effects due to smoking andto familiality of misk for lung cancer. Ooi et a!. (8) studiedfirst-degree relatives of 336 deceased lung cancer casesand first-degree relatives of 307 of their spouses in Lou-isiana. Quantitative measures of smoking behavior andoccupational exposures were obtained. After making al-bowance for smoking and occupational exposures, thecase relatives had a relative risk of 2.4 for lung cancerwhen compared with the control relatives. Familial ef-fects were more strongly observed among the femalerelatives than among the male relatives. In a similar butsmaller study, also conducted in southern Louisiana,Sellers et a!. (9) reported a relative risk of 2.5 for lungcancer among siblings of lung cancer cases compared tosiblings of the spouse controls. Shaw et al. (10) studiedfamilies from south Texas and observed a younger age atonset among probands with relatives who had lungcancer.

Segregation analyses (1 1) of data collected by Ooiet a!. (8) were consistent with Mendelian segregation ofa single codominant locus when a model was fitted thatallowed for a genetic component affecting the age ofonset of lung cancer. Among lung cancer cases youngerthan age 50, over 70% were predicted to have somegenetic predisposition, while over 70% of cases 70 orolder were predicted to be attributable solely to smoking.This study did not model specific effects attributable topassive smoking, and effects from occupational expo-sumes were not modeled. Results from this study reflecta rather restrictive set of genetic hypotheses that couldbe considered, e.g. , oligogenic (processes affected byseveral genes) influences were not specifically consid-ered. The number of lung cancer events observed in twinstudies has been too small to draw conclusions regardingthe familiality of lung cancer (12). However, Fisher’shypothesis that a common genetic cause underlies therisk for both smoking and lung cancer was disproved byevaluation of monozygotic twins discordant for smokingbehaviom(13, 14).

Diseases with Altered Risk for Lung Cancer. Familieswith excess lung cancer of diverse histological types havebeen reported (1 5), with alveolar cell carcinoma showingconsistent familial aggregation (15-17). Lung cancer maybe more common in pedigrees having an excess of breastand ovarian cancers (18), and survivors of familial reti-noblastoma may be at increased risk for small-cell lungcancer (19).

COPD2 is associated with an increased risk for lungcancers (20-22). A causal relationship of either diseasewith the etiology of the other is unlikely because of theirsimilar age-at-onset distributions. However, oxidatingagents in cigarette smoke may initiate COPD throughinflammatory processes and may initiate lung cancer by

genotoxic effects mediated by P-450 enzymes (23). Bothcross-sectional (24) and longitudinal observations (25)suggest that delayed clearance of inhaled carcinogens

1 The abbreviations used are: COPD, chronic obstructive pulmonary

disease; PAH, polycyclic aromatic hydrocarbon; BP, benzo]a]pyrene;IIRAS-1, Harvey ras protooncogene; AHH, aromatic hydrocarbon hy-

droxylase; Ah, aromatic hydrocarbon; RFLP, restriction fragment lengthpolymorphism; ECDE, 7-ethoxycoumarin 0-deethylase; CYP, structural

genes for cytochrome P-450 enzymes; SCE, sister chromatid exchange.

among COPD-affected individuals increases the risk forprogression to lung cancer. In further support of thishypothesis, increased risk for lung cancer is observed insarcoidosis and sclemodemma (26, 27), which have de-creased clearance of inhaled carcinogens. Several studies(3, 24) have noted familial clustering of COPD and otherrespiratory illnesses and lung cancer. Cohen (24) showedthat similar smoking and socioeconomic factors amongfamily members were not sufficient to account for thecorrelation of COPD and lung cancer in families, sug-gesting an underlying familial component of risk for both

diseases.Mortality from lung cancer has been reported to be

lower among schizophrenic patients (28, 29) in spite ofelevated levels of smoking (30). Neumoleptics such asphenothiazines have been reported to inhibit tumorgrowth in animals (31). Inhibition of P-450 enzyme me-tabolism might account for the apparent deficit of lungcancer among schizophrenia patients (32).

Genetically Influenced Traits Associated withVariability in Risk for Lung Cancer

Chemical Carcinogenesis. Activation of PAHs and amy-lamines by cytochmome P-450s (e.g., CYP1A1, CYPJA2,and CYP3A4) leads to the formation of reactive chemicalspecies that can bind covalently to DNA to form carcin-ogen-DNA adducts (33). Fig. 1 depicts the metabolism ofPAHs by cytochmome P-450 enzymes. PAHs such as BP

are first oxidized by AHH or CYP1A1 (the structural genefor P-4501A1), resulting in the formation of an areneoxide. This metabolite can be further activated to form adihydmodiol by the action of an epoxide hydmatase (34).Dihydmodiol can then be further metabolized across theolefinic double bond by both cytochmome P-450(CYP3A4) and other oxidation systems (35), thus forminga diol-epoxide. Diol-epoxides are unstable and rearrangeto form highly reactive carbocations. For example,benzo[a]pymene-7,8-diol-9,1 0-epoxide forms covalentbonds primarily with the exocyclic amino group of guan-osine. This reaction has been shown to be responsiblefor activating mutations in the HRAS- 1 protooncogene inseveral experimental systems (36, 37).

Alternative and competing routes of metabolism canlead to inactivation of carcinogens (e.g., PAHs) throughthe formation of conjugates (glutathiones, glucumonides,and sulfate esters) or phenols and tetrahydrotetmols whichfacilitate excretion. The balance between metabolic ac-tivation and metabolic detoxification, as well as the effi-ciency of DNA repair mechanisms, affects cancer risk inan individual exposed to PAHs.

Carcinogenic aromatic amines, typified by 4-amino-biphenyl, are related to bladder cancer; and the acety-lation phenotype, mediated by the polymorphic enzymeN-acetyltmansferase, has been related to susceptibility forthis tumor. Aromatic amines can be found in tobaccosmoke (38, 39), and aromatic amine-DNA adducts havebeen detected in the lung and bladder (40-42). Noassociation has been noted, however, between the ace-tylator phenotype and lung cancer (43-44). Analysis of4-aminobiphenyl hemoglobin adducts from lung cancercases and controls showed a correlation with tobacco

consumption but not with diagnosis of lung cancer (45),arguing against any role for these compounds in lungcarci nogenesis.

on November 22, 2020. © 1992 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

7 OH BP(NtH Shift)

SulfateEsters

Initetion� Repair

Mutation

Cancer Epidemiology, Biomarkers & Prevention 507

Mixed Giutathione Glucuronide MixedFunction Conjugaoon Conjugates Function

� 0x�daxe f Epoxide f Oiodase

� � �450iA1t Hydrolase

� � BP-7,8-epoxide � BP-7,8.dihydrodiol .

cDReductase

BenzotalpyreneIBPI

P450

PhenolsQuinonesEpoxides

MecromalecularAdducts

tBP-7,8-diol 9,10-epoxide

Tetrol GlutathioneTrial

fig. 1. Metabolism and activation of benzo[a]pyrene. The parent hydrocarbon is metabolized by cytochrome P-450 enzymes (CYP1A1) and the

metabolite formed is a substrate for epoxide hydrolase. Epoxide hydrolase catalyzes the formation of benzo[a]pyrene-7,8-dihydrodiol, a proximate

carcinogen. Benzo[a]pyrene-7,8-dihydrodiol is further activated at the oleuinic double bond to the ultimate carcinogenic metabolite, benzo[a]pyrene-7,8-diol.9,10-epoxide by cytochrome P-450. Benzo[a]pyrene-7,8-diol-9,10-epoxide binds primarily to the exocyclic amino group of guanosine.

Cytochrome P-450 Enzymes. A variety of P-450 enzymeshave been proposed as having a role in lung cancerthrough the demonstrated role of these enzymes in theactivation of carcinogens in tobacco. CYP1A1 catalyzesthe oxygenation of PAHs such as BP more efficiently thanCYP1A2, while substrates for the CYP1A2 enzyme addi-tionalby include amybamines and heterocyclic amines (foodmutagens) (46). The CYP1A genes are located on chro-mosome 15q22-qtem (47). Levels of both CYP1A genesare inducible and are influenced by the binding of theAh receptor. Induction follows exposure to aromatichydrocarbons, but the inducibility varies among individ-uabs. Several other gene products are also inducible bythe binding of the Ah receptor to a substrate, includingUDP glucuronosyltransferase (48), NAD(P)H:menadioneoxidoreductase (49), and glutathione transfemase (50).

Regulation of levels of CYP1A 1 occurs via a numberof different mechanisms (51). Following the binding ofPAHs with the Ah receptor, the complex is transportedto the nucleus, where it binds to nuclear chromatin in atemperature-dependent step and induces transcription.An upstream regulatory sequence, an MSP1 site (52, 53)downstream of the coding region, and a product fromchromosome 2 (54, 55) may all influence expression inhumans. Interaction of the Ah receptor:compound corn-plex with regulatory regions of the CYP1A2 gene doesnot appear to be as important as it is for regulation ofCYPJAJ levels (46), but a barge variability in the consti-tutive mRNA expression of CYPJA2 has been observedin human liver samples (46).

Studies of in vivo bevels of CYP1A enzymes as a riskfactor for lung and other cancers are confounded by theinducibibity of these enzymes. Because individuals oftenchange their smoking around the time of lung cancerdiagnosis, case-control studies designed to study the riskfor lung cancer associated with the expression of CYP1Aenzymes must make some allowance for the state ofinduction of the study subjects. Petruzelbi et al. (56)compared AHH, ECDE, epoxide hydmoxylase, glutathi-one-S-tmansferase, U D P-glucu ronosyltransfemase, gluta-thione, and mabondiabdehyde contents from nontumo-mous pamenchymab samples from 54 lung cancer patientsand 20 patients undergoing resection for nonneoplasticdiseases. All enzymatic levels (AHH, ECDE, UDP-glucu-monosyltransfemase, glutathione S-tmansferase) were signif-icantly correlated. When smoking was ignored, therewere no differences between cancer and noncancer pa-tients. However, lung cancer patients who had smoked

within 30 days of surgery had significantly higher AHHlevels and ECDE activities than other current smokers.Negative correlations (significant for all but ECDE) wereobserved between all of the enzyme bevels and thenumber of days that had elapsed since the patients hadgiven up smoking. No significant differences were ob-served between nonsmoking lung cancer cases and non-smoking controls. These two observations suggest thatthe presence of lung cancer itself does not lead to aninduction of these enzymes. Quantitative studies ofNorthern blots of mRNA for CYPJAJ expression wereperformed on lung tissue resected during surgery forprimary lung cancer (57). CYP1A1 expression was absentin nonsmokers (0 of 20) but was detected in 3 of 7smokers who had abstained less than 1 month and waspresent in 1 7 of 19 current smokers.

High levels ofAHH inducibility have been shown tobe a risk factor for lung cancer in several case-controlstudies (58); earlier results (59) which indicated that AHHinducibility was predictive of case status, however, couldnot be duplicated in many studies (60-62). Paigen et a!.(60) could not replicate earlier work despite implement-ing a careful design including cases as well as theirhealthy offspring and spouses.

Measurement error may have obscured results fromsome of the earlier studies. Kouni et al. (63) suggestedthat the ratio of AH H inducibility to cytochnome c activitywas a more reliable measure than AHH inducibility alone.Trell et a!. (64), studying 20 healthy subjects, found thatthe intraindividual variability in this ratio following ex-posure to 3-methylcholantnene was 0.016, while theintenindividual variability that can be estimated from theirdata is 1.63. Among 1589 smokers, 304 ex-smokems and218 never-smoking individuals, tnimodal distributionswere observed for each category of smoking status similarto those reported by Kellerman et al. (59), consistent withthe observed codominant expression (65).

bsoenzyrnes of glutathione S-tnansferase have beenidentified (66), and the presence of the mu isoenzyme isdominantly inherited. Results from two case-control se-nies conducted by Siedegard (67, 68) and others in Swe-den and New York City show that consistently fewerlung cancer patients have the mu isoenzyme than smok-ing-matched controls. The mu isoenzyrne was beast fre-quent among small cell, large cell, and adenocarcinomapatients, but the first two categories included few pa-tients. The odds ratio comparing the presence of the muisoenzyme in controls to cases is 2.4 for both studies

on November 22, 2020. © 1992 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

508 Review: Lung Cancer Ecogenetics

Table 1 Summary of findings from cas e-control studies of lung ca ncer and debri soquine phenotypes

Frequency of Frequency of

Investigator control case Oddsratio

95% Cl” Crite’rionto I)e’ ,I PM

EMs� PMs EMs PMs

Law et a/. 95 9 102 2 4.83 1.14 20.28 MR > 12.6Ayeshetal. 213 21 241 4 5.94 2.10 16.79 MR> 12.6Caporaso et a). blacks) 41 1 48 1 1.17 0.12 1 1.56 MR > 20.8Caporaso et al. whites) 39 10 47 0 NE 4.77 -NE MR > 20.8Rootseta/. 240 30 251 19 1.65 0.91 -3.00 MR> 12.0Ducheetal. 234 20 143 10 1.22 0.57-2.64 MR> 12.6

Benitezetal. 133 10 80 4’ 1.50 0.48-4.68 MR> 12.6

All studies 995 101 912 40 278d 1.60- 1.40

EMs, extensive metabolizers; PM, poor metabolizers; MR, metabolic ratio; NE, not estimable from the data (infinite).b Confidence intervals calculated using Cornfield’s method.

( Includes three adenocarcinomas.d Odds ratio calculated with a Mantel-Haenszel statistic, stratifying on study.

combined. Neither of these studies assessed ethnic var-iation, and in both studies cases and controls were notobtained from the same hospital populations. A subse-quent study (69) conducted on a homogeneous popula-tion found low levels of glutathione S-transfemase to benonsignificantly protective of lung and other smoking-rebated cancers, but the power to detect a significantdifference at least as large as that found by Siedegamd(68) was only 70%.

CYP2D6 (Debrisoquine) Phenotype. The ability to metab-obize the antihypertensive drug debmisoquine appears tobe a risk factor for lung cancer in smokers. Metabolismof debnisoquine along with other drugs including anti-depressants, neumoleptics, beta blockers, and codeinedepends upon the activity of the CYP2D6 enzyme. Apossible mechanism is suggested by the report (70) thatCYP2D6 activates the carcinogen 4-(methylnitmosamino)-1 -(3-pyridyl)-1 -butanone, a nicotine-derived nitroso ke-

tone from tobacco. Unlike CYP1A enzymes, levels ofCYP2D6 are not inducible, but constitutive expressionvaries widely among individuals. Sample storage (71),cincadian (72), and intraindividual variation account forminor degrees of variability in phenotyping, while certainmedications, notably quinidine, predictably invalidatephenotyping.

From family and population studies (71, 73-75), thepoor metabobizer phenotype can be well differentiatedfrom extensive or intermediate metabolizem phenotypes.

Intermediate metabolizers show incomplete dominance,with mean values being that of the extensive metabo-bizems plus 30% of the difference between that of theextensive and poor metabobizers. Among Caucasians ofEuropean descent, the pooled estimate of the poor me-tabobizer genotype frequency is 6.4% based upon phen-otyping of 3280 unrelated, healthy individuals (71), whilethis frequency is estimated at 2. 1% among West Africans.

The gene coding for CYP2D6 has been cloned (76),and polymemase chain reaction-based assays now distin-guish most poor metabolizems from extensive or inter-mediate metabolizers (77). A summary (78) of mutantCYP2D6 genes describes three significant mutations. TheCYP2D6(B) mutation accounts for about 75% of poormetabolizem alleles and results from a splice site defectin exon 4. CYP2D6(A) accounts for a little more than 5%of defective alleles and results from a deletion in exon 5.The CYP2D6(D) mutation represents a total deletion of

the gene and accounts for 10-15% of poor metabolizeralleles. About 10% of poom metabolizems currently remainunrecognized by molecular techniques.

Metabolic status for debmisoquine has been found to

be a predictor of risk for lung cancer (Table 1 ). In a case-control study of 245 cigarette smokers with bmonchogeniccamcinoma, poor metabolizems were underrepresented(79, 80). Results from subsequent studies (43, 81-84)support these earlier results, although findings from sev-emab studies (43, 83, 84) were equivocal. Benitez et ,iI.(83) observed a significant effect only when adenocamci-nomas are excluded, a finding which reflects the earlierobservation that adenocarcinomas did not share elevatedrisk based on the debmisoquine metabolic ratio (80).Speims et a!. (85) did not find evidence for an effect ofdebrisoquine metabolic ratio on risk for lung cancer, butthe cases-only approach used precludes compilation orcomparison with other studies. Using a Mantel-Haenszelstatistic, the odds ratio comparing the relative odds ofbeing a poor metabolizer versus being an extensive orintermediate metabolizem among lung cancer cases versuscontrols is 2.3 (95% confidence interval = 1.6-3.4; P <

0.001). Mild variability among the strata is observed (P =

0.01), indicating some dissimilarity among the resultsfrom the various studies.

Blood Groups and Serum Markers. Type 0 blood grouphas been associated with decreased risk for lung cancer,but cases and controls in this study were not selected bythe same mechanism (43). Although case-control studiesto assess the effects of HLA antigens on lung cancer riskhave been performed, consistent effects have not l)eendocumented, although to date few studies (86) haveincluded Dr or other class II loci. Serum a antitmypsingenotypes are not related to lung cancer risk (87) or toabnormal sputum cytobogies. However, cs antitrypsinlevels (88) and trypsin inhibitory capacity (87) were sig-nificantly higher in lung cancer patients than in controlswith normal or abnormal sputum cytologies.

Oncogene Polymorphisms. A variable tandem repeatDNA sequence region tightly flanked by Mspl restrictionsites, which is located 3’ with respect to the cellularHRAS- 1 structural protooncogene (chromosomal region11p15.5), accounts for a DNA RFLP. Rare alleles at theHRAS-1 protooncogene locus were found more com-monly in lung cancer (89) cases than in controls. Noassociation was found between either risk of lung cancer

on November 22, 2020. © 1992 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

Cancer Epidemiology, Biomarkers & Prevention 509

or risk of metastasis and a RFLP defined by an EcoRl sitein the second intron of the L-myc protooncogene (90).Both the HRAS-1 and L-myc RFLPs were found to varysignificantly with mace (89, 90).

Biomarkers Associated with Carcinogenesis andProgressionBiomarkers associated with carcinogenesis and tumorprogression include oncogene expression, DNA and redblood cell adducts, sister chromatid exchange, and intumor tissue, presence of microsatelbites, alteration inHLA expression, and expression of abnormal proteinssuch as a-fetoprotein. Generally, the utility of many ofthese markers in assessing individual risk for tumomigen-esis is hampered by the high measurement error of theassays, the expense, technical parameters (i.e., coblec-tion, storage, and processing of biological materials undervery stringent conditions), and the lack of specificity ofthe assays. Certain markers, specifically adducts, havethe potential to assess the internal dose of specific cam-cinogens, while others (i.e., SCE) have limited utility inassessing the host-specific risk for lung cancer.

Adduct Formation and Sister Chromatid Exchange. The-ometicably, the study of DNA adduct formation allows ameasure of the internal dose of carcinogen and thereforeprovides an advantage over traditional exposure instru-rnents such as questionnaires. At least one study hasdemonstrated a relationship between adduct formationand both a carcinogenic exposure and a genetic suscep-tibility factor (91). DNA adduct levels in monocytes ex-posed to BP from 26 lung cancer patients younger thanage 46 were more than twice (P < 0.0025) that ofmatched healthy controls (92). Significantly higher ad-duct levels were observed in monocytes from all smokinglung cancer patients versus all smoking controls and fromall nonsmoking cases versus all healthy nonsmokers.Studying 86 first-degree relatives in 15 families, Nowaket a!. (93) observed greater interfamilial than intrafarnibialvariability in DNA adduct formation among monocytesexposed to BP, suggesting that host-specific factors affecteither adduct formation or the rate of DNA repair.

SCE following exposure of lymphocytes to BP oc-curmed significantly more frequently in lymphocytes fromthe healthy offspring of lung cancer cases who had fouror more relatives with cancer than in lymphocytes fromhealthy offspring of lung cancer cases who did not havea family history of other cancers (94). SCE was found tooccur more frequently in cigarette smokers than amongnonsmokers, however (95, 96). Lung cancer patients whowere ex-smokers had the same rates of SCE as nonsmok-ing controls, implying that lung cancer does not alter SCEmates, but lung cancer patients who smoked had moreSCEs than age- and cigarette smoking-matched controls(95, 96).

Cytogenetic and Molecular Genetic Studies. Cytogeneticstudies are useful in identifying chromosomab regionsthat are likely to be involved in tumonigenesis. Regionsthat are typically deleted may contain antioncogene ortumor suppressor boci, while areas that are replicatedcontain growth factors. Deletion of the short arm ofchromosome 3 has been observed in most karyotypesfrom fresh small cell lung cancer specimens (97-104) andin studies of other major histological types (103-105).Loci at l7pl3 also showed a loss of hetenozygosity in 16

of 16 tumor samples heterozygous for RFLPs in this region(102). Loss of heterozygosity for markers on chromosome17p was found in 8 of 9 squamous cell cancers but inonly 2 of 11 adenocarcinomas (105). Similarly, 9 of 21non-small cell lung cancer tissue samples had loss ofheterozygosity for loci in the region 13q12-33 (105).Localization of von Hippel Lindau (106) and renal cellcarcinoma susceptibility loci to 3p26 (107, 108), netino-blastoma to 13q (109), and the p53 gene to chromosome17p (1 10) provides candidate regions for further evalua-tion of any families that may appear to segregate lungcancer.

Prospects for Future Studies

Designs for Studies. Case-control studies may be a usefultool for identifying the environmental and some geneticrisk factors important in lung cancer etiology but providepoor estimators of the specific genetic and environmentalcontributions of the total risk. Despite difficulties in rn-plementing case-control studies which include the studyof relatives, the power of this approach to resolve epi-derniological versus genetic contributions to disease etiol-ogy is high, and this design is often worth the effort.Implementation of designs which include case and con-trol families is hampered by the high mortality from lungcancer, so that affected relatives are likely to be de-ceased. However, if case definition is restricted to thosewho developed cancer at an early age, the likelihood isgreater that more relatives (and more affected relatives)will be alive and available for study. Generalization ofresults from this design requires the assumption thatindividuals affected by early-onset cancer are represent-

ative of the general population affected by lung cancer.Cohort studies, especially if focused on high-risk

subjects, can evaluate multiple markers, including P-450phenotypes, and may also avoid sources of bias fromchanging behavior patterns. In addition, these studiescan more readily assign the proper causal structure un-derlying disease etiology. Unfortunately, large enough

cohorts with sufficient questionnaire data and biologicalsamples are not readily available.

Linkage studies are an extremely powerful tool foridentifying the genetic boci involved in disease etiology,although problems encountered in conducting case-con-trob studies that include relatives also impede the imple-mentation of linkage studies. A linkage study is moreeasily accomplished if multiple living affected personsare available in a family. Often, other affected relativesare already deceased once the lung cancer proband hasbeen identified, so that appropriate families are difficultto obtain. In addition, the inheritance pattern for lungcancer does not fit a simple Medebian pattern, furtherincreasing the sample size necessary to identify any spe-cific locus predisposing for this disease.

New methods which address both ofthese problemsare being developed. Genetic linkage studies can beaccomplished despite deceased family members whenthe genotypes of these unavailable persons can be de-duced from that oftheir living relatives. In addition, tissueblocks saved from surgical procedures provide sufficientgenetic material for analysis, although nontumorous tis-sue is necessary. Advances have also been made inmodeling inheritance patterns for non-Mendelian phe-notypes (111) and in constructing robust linkage tests

on November 22, 2020. © 1992 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

.5 1(3 Review: Lung Cancer Ecogenetics

(1 12, 113). The latter assess evidence for linkage byevaluating the association between affection status andallele sharing for sib and other relative pairs and detectgenetic linkage without requiring exact knowledge of thedisease process.

Methodological Considerations for Some Specific Bio-markers. The back of reliable and inexpensive measuresof metabolic capacity for exogenous carcinogens hashampered the implementation of family studies. How-ever, continued progress may soon reduce the cost andincrease the efficacy of these studies. Caffeine metabo-bites act as an excellent surrogate for 4-aminobiphenylmetabolic capacity and could be used to assay reliablyfor CYP1A2 activity (114, 115). As yet, however, therelationship between hepatic and liver CYPJAJ enzy-matic activity and differential effects of CYPJA1 andCYP1A2 enzymes on caffeine metabolism areunavailable.

Assays for DNA adduct formation are theoreticallyuseful nonspecific measures of susceptibility for the ni-tiation of lung cancer, but some obstacles must be over-come before these assays can be of use in family studies.First, the relationship between exposure and subsequentadduct formation in lung tissues and blood or urine mustbe quantified. Second, measures of adduct formationremain prohibitively expensive for barge studies, andreproducibility of any inexpensive assays must be ex-plomed prior to implementation. Measures of SCE for-mation might similarly be studied, but these provide avery nonspecific measure of susceptibility to lung cancer.

Methods for assessing exposures to cigarette smokeand other carcinogens rebated to lung cancer need furtherrefinement. Passive smoking was shown to elevate lungcancer risk slightly (7, 1 16-1 18). In a large study, how-ever, the elevation of risk from passive smoke was con-ferred primarily by spousal and occupational exposuresand did not stem from childhood contacts, althoughthese may be subject to recall errors (1 18). Passive smok-ing is thus not likely to provide an adequate explanationfor increased familial risk since the familial environmentalexposures occur primarily during childhood, although aninstrument to measure this exposure must be includedin further family studies. Metabobites of nicotine (me-covered in blood or urine) reflect recent exposure totobacco products and can be used to assess the validityofquestionnaime data concerning recent exposures (119).However, cotinine measures are affected by intemindivid-ual variation in nicotine metabolism, dietary exposures,and other factors (e.g. , medications) and do not providea measure of tobacco exposures during periods of initi-ation and early promotion of lung cancer. Methods forcollecting proxy information on affected deceased mdi-viduals require further standardization.

Studies of genetic factors for lung cancer risk seemdifficult to implement because a barge number of expo-sumes that increase the risk for lung cancer have beenwell characterized. However, because the requisite oc-cupational and environmental exposures are well known,delineation of risk into genetic, environmental, and in-teractions between genetic and environmental factorsshould be possible.

ConclusionsA general schema for incorporating measures of geneticand environmental exposures in lung cancer studies in-

cludes both case-control studies as well as family studies.For instance, the case-control studies should be used toevaluate the relationship between measures of metabo-lism in the lungs and easily available samples (e.g., urine,blood) and to estimate sources of intraindividual variation(such as measurement error and diurnal variation) andintemindividual variation (including ethnicity). For nonin-

ducibbe traits or traits not affected by the disease process,case-control studies can identify biomarkems for lungcancer. Ethnicity must be collected as a confounder inthese studies (120). For inducible biomarkems, adequatestatistical adjustment for changing behavior may be im-possible. Family studies of unaffected relatives of casesand controls would then be used to verify the evidencefor an effect of the host factors on the risk for lungcancer.

Finally, evidence that specific boci have a majorimpact on susceptibility for disease can be assessed bygenetic linkage studies. These methods, however, will beexpensive to implement because blood samples frommany affected individuals will not be available, requiringthat many oftheir offspring be studied. These approacheshave nevertheless been useful in studying other diseaseswith a late age at onset and rapid course (121, 122).

Several boci are likely to be involved in the diseaseprocess, possibly reducing the power of the studies.

Assays of the metabolic status of particular alleles definedby DNA polymorphisms are becoming possible for a fewboci such as CYP2D6. As more candidate loci are studied,methods for identifying allelic variation will become sim-pier and more reliable. Genetic linkage strategies willremain useful because they address the more generalhypothesis that a genetic locus situated in a particularregion of the genome affects susceptibility to lung cancer.

References

1. International Agency for Research on Cancer. IARC Monographs on

the Evaluation of the Carcinogenic Risk of Cheniicals to Humans: Toba� 0

Smoking, Vol. 38, pp. 83-126, Lyon: International Agency for Re’searchonCaner, 1986.

2. International Agency for Research on Cancer. IARC Monographs on

the Evaluation C)t the’ Carcinoge’nic Risks Ii) Humans: Oc upational Es-posures in Petroleum Refining: Crude Oil and Major Petrole’um Fuels,

Vol. 45. Lyon: International Agency for Research on Cance’r, 1989.

3. Tokuhata, C. K., and Lillien(e’ld, A. M. Familial aggregation of lungcancer in humans. J. NatI. Caner Inst., 30: 289-312, 1963.

4. Tokuhata, C. K., and Lillienfeld, A. M. Familial aggre’gatiim of lungcancer among hospital patients. Public Health Rep., 78: 277- 283, 1963.

5. Bishop, D. T., and Skolnick, M. A. Genetic epidemiology of cancer inUtah pedigree’s: a prelude to the niolecular genetics of common Cane ers.

J. Cell Physiol., ./)Suppl(: 63-77, 1984.

6. Weiss, K. M., Chakraborty, R., Smouse, P. F., Buchanan, A. V., and

Strong, L. C. Familial aggregation of cancer in Laredo, Texas: a generally

low-risk Mexican-American population. Cenet. Epidemiol., 1: 121 -143,1986.

7. Correa, P., Pickle, L. W., Fontham, E., Lin, Y., and Haenszel, W.Passive smoking and lung cancer. Lanet, 2: 595-597, 1983.

8. Ooi, W. L., Elston, R. C., Chen, V. W., Bailey-Wilson, I. E., and

Rothschild, H. Increased familial risk for lung Cancer. I. NatI. Cance’r Inst.,76: 217-222, 1986.

9. Sellers, T. A., Ebston, R. C., Stewart, C., and Rothschild, H. Familial risk

of cancer among randomly selected cane-er prc)bands. Cenet. Epideniiol.,5: 381-392, 1988.

10. Shaw, C. L., Falk, R. T., Pickle, L. W., Mason, T. I.. and Buffler, P. A.

Lung cancer risk associated with cancer in relatives. I. Clin. Epideniiol.,

44:429-437, 1991.

1 1. Sellers, T. A., Bailey-Wilson, I. F., Elston, R. C., Wilson, A. E., Elston,

R. C., and Rothschild, H. Evidence’ for Mendelian inheritance in thepathogenesisoflungcancer. I. NatI. Cancerlnst., 82: 1272-1279, 1990.

on November 22, 2020. © 1992 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

Cancer Epidemiology, Biomarkers & Prevention 5 1 1

12. Hrubec, Z., and Neel, J. V. Contribution of familial factors to theoccurrence of cancer before old age in twin veterans. Am. J. Hum.Genet., 34: 658-671, 1982.

1 3. Floderus, B., Cederlof, R., and Friberg, L. Smoking and mortality: a21-year follow-up based on the Swedish twin registry. tnt. J. Epidemiol.,17: 332-340, 1988.

14. Kaprio, J., and Koskenvuo, M. Twins, smoking and mortality: a 12-

year prospective study of smoking-discordant twin pairs. Soc. Sci. Med.,

29: 1083-1089, 1989.

15. Goffman, T. E., Hassinger, D. D., and Mulvihill, J. J. Familial respira-tory tract cancer, opportunities for research and prevention. JAMA, 247:

1020-1023, 1982.

16. Joishy, S. K., Cooper, R. A., and Rowley, P. T. Alveolar carcinoma inidentical twins with similarity in time of onset, histochemistry and site ofmetastasis. Ann. Int. Med., 87: 447-450, 1977.

17. Paul, S. M., Bacharach, B., and Goepp, C. A genetic influence onalveolar cell carcinoma. J. Surg. Oncol., 36: 249-252, 1987.

18. Lynch, H. T., and Krush, A. J. Carcinoma of the breast and ovary inthree families. Surg. Gynecol. Obstet., 133: 644-648, 1971.

19. Leonard, R. C. F., Mackay, T., Brown, A., Gregory, A., Crompton, G.K., and Smyth, J. F. Small-cell lung cancer after retinoblastoma [letter].Lancet, 2: 1503, 1988.

20. Stoloff, I. L., Kanofsky, P., and Magilner, L. The risk of lung cancer inmales with bullous disease of the lung. Arch. Environ. Health, 22: 163-167, 1971.

21. Cohen, B. H., Diamond, E. L., Graves, C. G., Kreiss, P., Levy, D. A.,Menkes, H. A., Permutt, S., Quaskey, S., and Tockman, M. S. A commonfamilial component in lung cancer and chronic obstructive pulmonary

disease. Lancet, 2: 523-526, 1977.

22. Skillrud, D. M., Offord, K. P., and Miller, R. D. Higher risk of lungcancer in chronic obstructive pulmonary disease: a prospective matched

controlled study. Ann. Int. Med., 105: 503-507, 1986.

23. Bartsch, H., Petruzelli, S., De Flora, S., Hietanen, E., Camus, A. M.,

Castegnaro, M., Geneste, 0., Camoirano, A., Saracci, R., and Giuntini, C.

Carcinogen metabolism and DNA adducts in human lung tissues as

affected by tobacco smoking or metabolic phenotype: a case-control

study on lungcancer patients. Mutat. Res., 250: 103-114, 1991.

24. Cohen, B. H. Is pulmonary dysfunction the common denominator

forthe multiple effects ofcigarette smoking? Lancet, 2: 1024-1027, 1978.

25. Tockman, M. S., Anthonisen, N. R., Wright, E. C., Donithan, M. G.,

et al. Airways obstruction and the risk for lung cancer. Ann. Int. Med.,

106: 512-518, 1987.

26. Petitjean, R., Burghard. G., Stampfler, G., and Fievez, M. Une nou-

velle observation d’association sclerodermie et cancer alveolaire du pou-mon. l. Fr. Med. Chir. Thorac., 22: 43-52, 1968.

27. Mulvihill, J. J. Genetic factors in pulmonary neoplasms. Birth Defects

Orig. Artic. Ser., 1 2: 99- 1 1 1 , 1976.

28. Achterberg, I., Collerain, I., and Craig, P. A possible relationship

between cancer, mental retardation, and mental disorders. Social Sci.Med., 12(A): 135-139, 1978.

29. Mortensen, P. B. The incidence of cancer in schizophrenic patients.

I. Epidemiol. Community Health, 43: 43-47, 1989.

30. O’Farrell, T. J., O’Connors, G. J., and Upper, D. Addictive behaviors

among hospitalized psychiatric patients. Addict. Behav., 8: 329-333,1983.

31. Driscoll, I. S., Melnich, N. R., Quinn, F. R., et al. Psychotropic drugs

as potential antitumor agents. A selective screening study. Cancer Treat.Rev., 62: 45-74, 1978.

32. Syv#{227}lahti, E. K. G., Lindberg, R., KaIlio, J., and de Vocht, M. Inhibitoryeffects on neuroleptics on debrisoquine oxidation in man. Br. J. Clin.

Pharmacol., 22: 89-92, 1986.

33. Oesch, E., Aulmann, W., Platt, K. L., and Doerjer, G. Individualdifferences in DNA repair capacities in man. Arch. Toxicol., 10 (Suppl.):172-179, 1987.

34. Gonzalez, F. J., Crespi, C. L., and Gelboin, H. V. DNA-expressedhuman cytochrome P450s: a new age of molecular toxicology and humanrisk assessment. Mutat. Res., 247: 113-127, 1991.

35. Benford, D. J., and Bridges, J. W. xenobiotic metabolism in lung. In:

I. W. Bridges and L. F. Chasseaud (eds.), Progress in Drug Metabolism,pp. 53-94. London: Taylor and Francis, Ltd., 1986.

36. Vousden, K. H., Bos, J. L., Marshall, C. J., and Phillips, D. H. Mutationsactivating human c-Ha ras proto-oncogene (HRAS1) induced by chemicalcarcinogens and depurination. Proc. NatI. Acad. Sci. USA, 83: 1222-1226,

1986.

37. Marshall, W., Vousden, K. H., and Phillips, D. H. Activation of c-Ha-ras-1 proto-oncogene by in vitro modification with a chemical carcinogen,benzo(a)pyrene diol-epoxide. Nature (Lond.), 310: 586-589, 1984.

38. Hem, D. W. Acetylator genotype and arylamine-induced carcinogen-esis. Biochim. Biophys. Acta, 948: 37-66, 1988.

39. Vineis, P., Caporaso, N., Tannenbaum, S. R., Skipper, P. L., Gb-gowski, J., Bartsch, H., Coda, M., Talaska, G., and Kadlubar, F. Acetylationphenotype, carcinogen-hemoglobin adducts, and cigarette smoking. Can-cer Res., 50: 3002-3004, 1990.

40. Weston, A., and Harris, C. C. Chemical carcinogenesis. In: J. F.Holland and E. Frei (eds.), Cancer Medicine, Ed. 3. Philadelphia: Lea and

Febiger, 1992.

41. Nebert, D. W., Petersen, D. D., and Puga, A. Human AH locuspolymorphism and cancer: inducibility of CYP1A1 and other genes bycombustion products and dioxin. Pharmacogenetics, 1: 68-78, 1991.

42. Talaska, G., Schamer, M., Skipper, P., Tannenbaum, S., Caporaso,N., Unruh, L., Kadlubar, F. F., Bartsch, H., Malaveille, C., and Vineis, P.Detection of carcinogen-DNA adducts in exfoliated urothelial cells ofcigarette smokers: association with smoking, hemoglobin adducts andurinary mutagenicity. Cancer Epidemiol., Biomarkers & Prey., 1: 61-66,1991.

43. Roots, I., Drakoulis, N., Pboch, M., Heinmeyer, G., Loddenkemper,R., Minks, T., Nitz, M., Otte, F., and Kocj, M. Debrisoquine hydroxylation

phenotype, acetylation phenotype, and ABO blood groups as genetichost factors of lung cancer risk. KIm. Wochenschr., 66 (Suppl. 1 1): 87-97, 1988.

44. Philip, P. A., Fitzgerald, D. L., Cartwright, R. A., Peake, M. D., and

Rogers, H. J. Polymorphic N-acetylation capacity in lung cancer. Carci-nogenesis (Lond.), 9: 491-493, 1988.

45. Weston, A., Caporaso, N. E., Taghizadeh, K., Hoover, R. N. Tannen-baum, S. R., Skipper, P. L., Resau, J. H., Trump, B. F., and Harris, C. C.Measurement of 4-aminobiphenyl-hemogbobin adducts in lung cancercases and controls. Cancer Res., 51: 5129-5223, 1991.

46. bkeya, K., Jaiswal, A. K., Owens, R. A., Jones, J. E., Nebert, D. W.,and Kimura, S. Human CYP1A2. Sequence, gene structure, comparisonwith the mouse and rat orthologous gene, and differences in liver 1A2mRNA expression. Mob. Endocrinol., 3: 1399-1408, 1990.

47. Jaiswal, A. K., Gonzalez, F. J., and Nebert, D. W. Human P-450 genesequence and correlation of mRNA with genetic differences inbenzo(a)pyrene metabolism. Nucleic Acids Res., 13: 4503-4520, 1985.

48. Owens, I. Genetic regulation of UDP-glucuronosyltransferase induc-tion by polycylic aromatic compounds in mice. Co-segregation with arylhydrocarbon (benzo(a)pyrene)hydroxylase induction. J. Biol. Chem., 252:

2827-2833, 1977.

49. Kumaki, K., Jensen, N. M., Shire, J. G. M., and Nebert, D. W. Geneticdifferences in induction of cytosol reduced NAD(p): menadione oxido-reductase and microsomal aryl hydrocarbon hydroxylase in the mouse.J. Biol. Chem., 252: 157-165, 1977.

50. Felton, J. S., Ketley, J. N., Jakoby, W. B., Aitio, A., Bend, J. R., et al.

Hepatic glutathione transferase activity induced by polycylic aromaticcompounds. Lack of correlation with murine Ah locus. Mol. Pharmacol.,18: 559-564, 1980.

51. Nebert, D. W., and Jones, J. E. Regulation of the mammalian cyto-chrome P1-450 (CYP1A1) gene. Int. J. Biochem., 21: 243-252, 1989.

52. Kawajiri, K., Nakachi, K., bmai, K., Yoshii, A., Shinoda, N., andWatanabe, J. Identification of genetically high risk individuals to lungcancer by DNA polymorphisms of the cytochrome P4501A1 gene. FEBSLett., 263: 131-133, 1990.

53. Petersen, D. D., McKinney, C. F., lkeya, K., Smith, H. H., Bale, A. E.,McBride, 0. W., and Nebert, D. W. Human CYP1A1 gene: cosegregationof the enzyme inducibility phenotype and an RFLP. Am. J. Hum. Genet.,48: 720-725, 1991.

54. Wiebel, F. J., Hlavica, P., and Grzeschik, K. H. Expression of aromaticpolycyclic hydrocarbon-induce mono-oxygenase (aryl hydrocarbon hy-droxylase) in man X mouse hybrids is associated with human chromo-some 2. Hum. Genet., 59: 277-280, 1989.

55. Ocraft, K. P., Muskett, J. M., and Brown, S. Localisation of the arylhydrocarbon hydroxylase gene to the 2q31 -2pter of chromosome 2. Ann.Hum. Genet., 49: 237-239, 1985.

56. Petruzelli, S., Camus, A. M., Carrozzi, L., Ghelarducci, L., Rindi, M.,Gianfranco, M., Angeletti, C. A., Ahotupa, M., Hietanen, E., Aitio, A.,Saracci, R., Bartsch, H., and Giuntini, C. Long-lasting effects of tobaccosmoking on pulmonary drug and metabolizing enzymes: a case controlstudy on lung cancer patients. Cancer Res., 48: 4695-4700, 1988.

57. McLemore, T. L., Adelberg, S., Liu, M. C., McMahon, N. A., Yu, S. J.,Hubbard, W. C., Czerwinski, M., Wood, T. G., Storeng, R., Lubet, R. A.,

on November 22, 2020. © 1992 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

5 12 Review:Lung Cancer Ecogenetics

Eggleston, I. C., Boyd, M. R., and Hines, R. N. Expression of CYP1A 1 genein patients with lung cancer: evidence for cigarette smoke-induced gene

expression in normal lung tissue and for altered gene regulation in primarypulmonary carcinomas. I. NatI. Cancer Inst., 82: 1333-1339, 1990.

58. Law, M. R. Genetic predisposition to lung cancer. Br. I. Cancer, 61:

195-206, 1990.

59. Kellermann, C., Shaw, C. R., and Luyten-Kellerman, M. AryI hydro-

carbon hydroxylase inducibility and bronchogenic carcinoma. N. EngI. I.Med., 289: 934-937, 1973.

60. Paigen, B., Gurtoo, H. L., Minowada, J., Houten, L., Bincent, R.,

Paigen, K., Parker, N. B., Ward, E., and Hayner, N. T. Questionablerelation of aryl hydrocarbon hydroxylase to lung-cancer risk. N EngI. I.

Med., 297: 346-350, 1977.

61. Ward, E., Paigen, B., Steenlannd, K., Vincent, R., Minowada, I..Gurtoo, H. L., Sartori, P., and Havens, M. B. Aryl hydrocarbon hydroxylase

in persons with Icing or laryngeal cancer. Int. I. Cancer, 22: 384-389.

1978.

62. ott, I. R., Branuni, F. I., Fontana, E. F., Taylor, W. F., an(I Moses, El.t_. Maromolecular I)inditlg of ] ‘F-I ]I)en/o),l)pyre’n(’ metabolites atid lym-

�)ho(’yte transformation in pati(’nts with Icing cancer ,ind in smoking andnonsmoking control subjects. Am. Rev. Respir. Dis., 120: 369-375, 1979.

63. Kouri, R. E., McKinney, C. E., Slomiay, D. J., Snodgrass, D. R., Wray,

N. P., and McLemore, T. L. Positive correlation between high aryl hydro-

carbon hydroxylase activity and primary lung cancer as analyzed in

cryopreserved lymphocytes. Cancer Res., 42: 5030-5037, 1982.

64. TreIl, L., Korsgaard, R., lanzon, L., and Trell, E. Distribution andreproducibility of aryl hydrocarbon hydroxylase mnducibility in a prospec-

tive population of middle-aged male smokers and nonsmokers. Cancer

(Phila.), 56: 1988-1994, 1985.

65. Kellermann, C., Luyten-Kellerman, M., and Shaw, C. R. Cenetic

variation of arylhydrocarbon hydroxylase in human lymphocytes. Am. I.Hum. Genet., 25: 327-331, 1973.

66. Siedeg�rd, I., and Pero, R. W. The hereditary transmission of high

glutathione transferase activity towards trans-stilbene oxide in human

mononuclear leukocytes. Hum. Cenet., 69: 66-68, 1985.

67. Siedegi�rd, J., Pero, R. W., Miller, D. C., and Beattie, E. I. A gbutathione

transferase in human leukocytes as a marker for the susceptibility to lung

cancer. Carcinogenesis (Lond.), 7: 751-753, 1986.

68. Siedeg�rd, J., Pero, R., Markowitz, M. M., Roush, C., Miller, D. C.,and Beattie, E. J. lsoenzyme(s) of glutathione transferase (class Mu) as a

marker for the susceptibility to lung cancer: a follow up study. Carcino-genesis (Lond.), 1 1: 33-36, 1990.

69. Heckbert, S. R., Weiss, N. S., Hornung, S. K., Eaton, D. L., andMotulsky, A. C. Clutathione S-transferase and epoxide hydrolase activityin human leukocytes in relation to risk of lung cancer and other smoking-

related cancers. I. NatI. Cancer. Inst., 84: 414-422, 1992.

70. Crespi, C. L., Penman, B. W., Celboin, H. V., and Gonzalez, F. I. Atobacco smoke-derived nitrosamine, 4-(methylnitrosamino). 1 -(3-pyridyll-

1 -butanone, is activated by multiple human cytochrome P450s includingthe polymorphic human cytochrome P4502D6. Carcinogenesis )Lond.),

12: 1197-1201, 1991.

71. Steiner, E., Iselius, L., Alv#{224}n,C., Lindsten, j., and SjOqvist, F. A familystudy of genetic and environmental factors determining polymorphic

hydroxylation ofdebrisoqumn. Clin. Pharmacol. Ther., 38: 394-401, 1985.

72. Shaw, C. L., Falk, R. T., Caporaso, N. E., Issaq, H. I., Kase, R. C., Fox,S. D., and Tucker, M. A. Effect of diurnal variation on debrisoquine

metabolic phenotyping. J. NatI. Cancer. Inst., 82: 1573-1575, 1991.

73. Henthorn, T. K., Benitez, J., Avram, M. J., Martinez, C., Llerena, A.,

Cobaleda, J., Krejcie, T. C., and Gibbons, R. D. Assessment of thedebrisoquin and dextromethorphan phenotyping tests by gaussian mix-lure distributions analysis. Clin. Pharmacol. Ther., 45: 328-333, 1988.

74. Evans, D. A. P., Mahgoub, A., Sloan, T. P., Idle, j. R., and Smith, R.L. A family and population study of the genetic polymorphism of debri-

soquine oxidation in a white British population. J. Med. Genet., 17: 102-

105, 1980.

75. Caporaso, N., Pickle, L. W., Bale, S., Ayesh, R., Hetzel, M., and Idle,

I. The distribution of debrisoquine metabolic phenotypes and implica-

lions for the suggested association with lung cancer risk. Genet. Epide-

miol., 6: 517-524, 1989.

76. Kimura, S., Umeno, M., Skoda, R. C., Meyer, U. A., and Conzalez,

F. I. The human debrisoquine 4.hydroxylase locus: sequence and iden-

tification of the polymorphic CYP2D6 gene, a related gene and a pseu-dogene. Am. I. Hum. Genet., 45: 889-904, 1989.

77. Sugimura, H., Caporaso. N. F., Shaw, C. L., Modali, R. V., Gonzalez,

F. l.� Hoover, R. N., Resau, J. H., Trump, B. F., Weston, A., and Harris, C.

C. Human debrisoqumne hydroxylase gene polymorphisms in cancer

patients and controls. Carcinogenesis (Lond.), 1 1: 1527-1530, 1990.

78. Conzalez, F. I., and Meyer, U. A. Molecular genetics of the debriso-

quine-sparteine polymorphism. Clin. Pharmacol. The’r., 50: 233 --238,1991.

79. Ayesh, R., Idle, I. R., Ritchie, I. C., Crothers, M. I.. and Hetzel, M. R.Metabolic oxidative phenotypes as markers for susceptil)ility to lung

cancer. Nature (Lond.), 112: 169-170, 1984.

80. Caporaso, N., Hayes, R. B., Dosemeci, M., Hoover, R., Ayesh, R.,

Hetzel, M., and Idle, I. Lung cancer risk, occupational exposure and thedebnsoquine metabolic phenotype. Cancer Res., 49: 3675 -3679, 1989.

81. Law, M. R., Hetzel, M. R., and Idle, I. R. Debrisoquine metabolism

and genetic predisposition to lung caner. Br. I. Cancer, 59: 686-687,1989.

82. Caporaso, N. F., Tucker, M. A., Hoover, R. N., Hayes, R. B., Pickle’,

L. W., lssaq. H. I.. Muschik, C. M., Creen-Callo, L., Buivy, D., Aisner, S.,Resau, I . 1 1., Trump, B. F .,Weston, A., 10(1 1 larris, (#{149}. ( . I ung at� er .10(1the (lebrisoquine’ nie’talxtlie )henotyp’. I . N.itl. ( ,lfl( (‘r Inst., 82: 12641272, t9�)0.

8 1. Benitei, I., 1.a(1(’r(i, I . M., lar,i, ( � ( arilbo, I . A., (ohak’iI.t, I ., I 1’r’na,A., Vargas, E., and Munoz, I. I. Polymorphic oxidation of de’brisoquine’ in

lungcancer patients. Eur. I. Cancer., 27: 158-161, 1991.

84. Duche, [-C., Joanne, C., Barre, I., de Cremoux, H., Dalphin, I. C.,Depierre, A., Brochard, P., Tillement, I. P., and Behtel, P. Lack of arelationship between the polymorphism of debrisoquine oxidation andlungcancer. Br. I. Clin. Pharmacol., 31: 533-536, 1991.

85. Speirs, C. I., Murray, S., Davies, D. S., Mabadeje, A. F. B., and Boobis,

A. R. Debrisoquine oxidation phenotype and susceptibility to lung cancer.Br. I. Clin. Pharmacol., 29: 101-109, 1990.

86. Romano, P. J., Bartholomew, M., Kloszewski, F., Stryker, I.. Houck,

I., and Vesell, F. S. HLA antigens influe’nce resistance to lung carcinoma.Hum. Immunol., 31: 236-240, 1991.

87. Harris, C. C., Cohen, M. H., Connor, R., Primack, A., Sacconianno,C., and Talamo, R. C. Serum a, antitrypsin in patients with lung Can((’r

orabnormal sputum cytology. Cancer (Phila.), 38: 1655-1657, 1976.

88. Harris, C. C., Priack, A., and Cohe’n, M. H. Elevated n antitrypsin

serum levels in lung cancer patients. Cancer (Phila.(, 34: 280-281, 1974.

89. Sugimura, H., Caporaso, N. F., Hoover, R. N., Modali, R., Resau, I..Trump, B. F., Lonergan, I. A., Krontiris, T. C., Mann, D. L., Weston, A.,

Harris, C. C. Association of rare alleles of the Harvc’y ras protoon(ogene’locus with lung cancer. Cancer Res., 50: 1857-1862, 1990.

90. Tamai, S., Sugimura, H., Caporaso, N. E., Resau, I. H., Trump, B. F.,Weston, A., and Harris, C. C. Restriction fragment length polymorphismanalysis of the L-myc gene locus in a case-control study of lung ( ancer.

Intl. Cancer, 46:411-415, 1990.

91. Vineis, P., Caporaso, N., Tanne’nhaum, S., Skipper, P., Clogowski, I..

Bartsch, H., Coda, C., Talaska, C., and Kadlubar, F. Acetylation pheno-type, carcinogen-hemoglobin adducts, and cigarette smoking. Cancer

Res., 50: 3002-3004, 1990.

92. Rudiger, H. W., Nowak, D., Hartmann, K., and Cerutti, P. Enhance’dformation of benzo)a)pyrene: DNA adducts in monocytes of patients

with a presumed predisposition tc lung cancer. Cancer Res., 45: 5890-

5894, 1985.

93. Nowak, D., Schmidt-Preuss, U., borres, R., Liebke, F., and Rudige’r,H. W. Formation C)f DNA adducts and water-soluble metabolite’s of

benzo[a]pyrene in human nionocytes is genetically controled. hit. I.Cancer,41: 169-173, 1988.

94. Dosaka, H., Abe, S., Sasaki, M., Miyamoto, H., and Kawakami, Y.Sister chromatid induction by 1)e’nzo(a)pyrene in cultured PeriPheral

blood lymphocytes of lung cancer patients and healthy individuals withC)� without a familial history of neoplasms. tnt. I. Cancer, 39: 329-332,

1987.

95. Perera, F., Maye’r, I., laretzki, A., Hearne, S., Brenner, D., Young, T.L., Fischman, H. K., Crimes, M., Crantham, S., Tang, M. X., Tsai, W-Y.,

and Santella, R. M. Comparison C)t DNA adducts and sister chromatidexchange in lung cancer cases and controls. Cancer Res., 49: 4446 --4451,

1989.

96. Hopkin, I. M., and Evans, H. I. Cigarette smoke-induced DNA dam-

age and lung cancer risks. Nature’ (Lond.), 28 3: 388- 390, 1980.

97. Whang-Peng, I., Kao-Shan, C. S., Lee, E. C., Bunn, P. A., Carney, D.N., Cadzar, A. F., and Minna, I. D. Specific chromosome defect associated

with human srnall-e11 lung cance’r: deletion 3p(14-23(. ScienCe, 215:

181-182, 1982.

98. Morstyn, C., Brown, I., Novak, U., Cardner, I., Bishop, I.. and Carson,M. Heterogeneous cytogenetic abnormalities in small cell lung ( ancer

cell lines. Cancer Res., 47: 3322- 3327, 1987.

on November 22, 2020. © 1992 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

Cancer Epidemiology, Biomarkers & Prevention 513

99. Naylor, S. L., Johnson, B. E., Minna, J. D., and Sakaguchi, A. Y. Lossof heterozygosity of chromosome 3p markers in small-cell lung cancer.Nature bLond.), 329: 451 -454, 1987.

100. Johnson, B. E., Sakaguchi, A. Y., Gazdar, A. F., Minna, J. D., Burch,D., Marshall, A., and Naylor, S. L. Restriction fragment length polymor-

phism studies show consistent loss of chromosome 3p alleles in small cell

lung cancer patients’ tumors. J. Clin. Invest., 82: 502-507, 1988.

101. Drabkin, H., Kao, F. T., Hartz, J., Gart, I., Garder, A., Weinberger,

C., Evans, R., and Gerber, M. Localization of human ERBA2 to the 3p22-3p24.l region of chromosome 3 and variable deletion in small cell lung

cancer. Proc. NatI. Acad. Sci. USA, 85: 9258-9262, 1988.

102. Mori, N., Yokota, J., Oshimura, M., Cavenee, W. K., Mizoguchi, H.,Noguchi, M., Shimosato, Y., Sugimura, T., and Terada, M. Concordantdeletions of chromosome 3p and loss of heterozygosity for chromosomes

13 and 17 in small cell lung carcinoma. Cancer Res., 49: 5130-5135,

1989.

103. Kok, K., Osinga, I., Carritt, B., Davis, M. B., van der Hout, A. H.,

van der Veen, A. Y., Landsvater, R. M., de Leij, L. F. M. H., Berendsen,

H. H., Postmus, P. E., Poppema, S., Buys, C. H. C. M. Deletion of a DNA

sequence at the chromosomal region 3p21 in all maior types of lung

cancer. Nature (Lond.), 330: 578-582, 1987.

104. Yokota, J., Wada, M., Shimosato, Y., Terada, M., and Sugimura, T.Loss of heterozygosity on chromosomes 3, 13, and 17 in small-cell

carcinoma and on chromosome 3 in adenocarcinoma of the lung. Proc.NatI. Acad. Sci. USA. 84: 9252-9256, 1987.

105. Weston, A., Willey, I. C., Modali, R., Sugimura, H., McDowell, E.

M., Resau, J., Light, B., Haugen, A., Mann, D. L., Trump, B. F., et al.Differential DNA sequence deletions from chromosomes 3, 1 1 , 1 3, and1 7 in squamous-cell carcinoma, large-cell carcinoma, and adenocarci-

noma of the human lung. Proc. NatI. Acad. Sci. USA 86: 5099-5 103,1989.

106. Hosoe, S., Brauch, H., Latif, F., Glenn, G., Daniel, L., Bale, S.,

Choyke, P., Gorin, M., Oldfield, E., Berman, A., Goodman, J., Orcult, M.L., Hampsch, K., Delisio, H., Modi, W., McBride, W., Anglard, P., Weiss,

G., Walther, M. M., Linehan, W. M., Lerman, M. I., and Zbar, B. Local-ization of the von Hippel-Lindau disease gene to a small region of

chromosome 3. Genomics, 8: 634-640, 1990.

107. Zbar, B., Brauch, H., Talmadge, C., and Linehan, M. Loss of alleles

on the short arm of chromosome 3 in renal cell carcinoma. Nature bLond.),

327: 721-724, 1987.

108. Kovacs, G., Erlandsson, R., Boldog, F., Ingvarsson, S., Muller-Brech-

bin, R., Klein, G., and Sumegli, J. Consistent chromosome 3p deletion andloss of heterozygosity in renal cell carcinoma. Proc. NatI. Acad. Sci. USA,

85: 1571-1575, 1988.

109. Cavenee, W. K., Dryja, T. P., Phillips, R. A., Benedict, W. F.,Godbout, R., Gallie, B. L., Murphree, A. L., Strong, L. C., and White, R.

L. Expression of recessive alleles by chromosomal mechanisms in reti-noblastoma. Nature (Lond.), 305: 779-784, 1984.

1 10. Benchimol, S., Lamb, P., Crawford, L. V., Sheer, D., Shows, T. B.,

Bruns, C. A. P., and Peacock, J. Transformation associated pS3 protein isencoded by a gene on human chromosome 1 7. Somatic Cell Mol. Genet.,

1 1: 505-509, 1985.

1 1 1 . Bonney, G. E. Interactions of genes, environment, and life-style in

lung cancer development. I. Natl. Cancer Inst., 82: 1236-1237, 1990.

112. Dawson, D. V., Kaplan, E. B., and Elston, R. C. Extensions to sib-

pair linkage tests applicable to disorders characterized by delayed onset.

Genet. Epidemiol., 7: 453-466, 1990.

1 1 3. Flanders, W. D., and Khoury, M. J. Extensions to methods of sib.

pair linkage analyses. Genet. Epidemiol., 8: 399-408, 1991.

114. Butler, M. A., Iwasaki, M., Guengerich, P., and Kadlubar, F. F.

Human cytochrome P-450 PA (P450IA2), the phenacetin-deethylase, isprimarily responsible for the hepatic 3-demethylation of caffeine and the

N-oxidation of carcinogenic arylamines. Proc. NatI. Acad. Sci. USA, 86:7696-7700, 1986.

1 1 5. Grant, D. M., Tang, B. K., and Kalow, W. Variability in caffeine

metabolism. Clin. Pharmacol. Ther., 33: 591-602, 1983.

116. Blot, W. J, and Fraumeni, j. F., Jr. Passive smoking and lung cancer.

J. NatI. Cancer Inst., 77: 993-1000, 1986.1 1 7. Janerich, D. T., Thompson, W. D., Varela, L. R., Greenwald, P.,Chorost, S., Tucci, C., Zaman, M. B., Melamed, M., Kiely, M., and

McKneally, M. F. Lung cancer and exposure to tobacco smoke in the

household. N. EngI. J. Med., 323: 632-636, 1990.

1 18. Fontham, E. T. H., Correa, P., Wu-Williams, A., Reynolds, P., Green-

berg, R. S., Buffler, P. A., Chen, V. W., Boyd, P., Alterman, T., Austin, D.

F., Liff, I., and Greenberg, S. D. Lung cancer in nonsmoking women: amulticenter case-control study. Cancer Epidemiol., Biomarkers & Prey.,

1: 35-43, 1991.

119. Riboli, E., Preston-Martin, S., Saracci, R., Haley, N. J., Trichopoulos,

D., Becker, H., Burch, J. D., Fontham, E. T. H., Gao, Y. T., jindal, S. K.,Koo, L. C., Le Marchand, L., Segnan, N., Shimizu, H., Stanta, G., Wu-Williams, A. H., and Zatonski, W. Exposure of nonsmoking women to

environmental tobacco smoke: a ten-country collaborative study. Cancercauses Control., 1: 243-252, 1990.

120. Sterling, T. D., and Weinkam, I. J. Comparison of smoking-related

risk factors among black and white males. Am. J. nd. Med., 15: 3 19-333,

1989.

121. St. George Hysbop, P. H., Tanzi, R. E., Polinsky, R. J., Haines, J. L.,

Nee, L., Watkins, P. C., Meyers, R. H., et al. The genetic defect causing

familial Alzheimer’s disease maps on chromosome 2 1 . Science (Washing.

ton DC), 235: 885-890, 1987.

122. Hall, J. M., Lee, M. K., Newman, B., Morrow, J. E., Anderson, L. A.,Huey, B., and King. M. C. Linkage of early-onset familial breast cancer to

chromosome 17q21. Science (Washington DC). 250: 1684-1689, 1990.

on November 22, 2020. © 1992 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

1992;1:505-513. Cancer Epidemiol Biomarkers Prev   C I Amos, N E Caporaso and A Weston  studies.Host factors in lung cancer risk: a review of interdisciplinary

  Updated version

  http://cebp.aacrjournals.org/content/1/6/505

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://cebp.aacrjournals.org/content/1/6/505To request permission to re-use all or part of this article, use this link

on November 22, 2020. © 1992 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from