prognostic significance of germ line polymorphisms of the ... · genes and clinical characteristics...

(CANCER RESEARCH 56. .1725-3730. Augusl 15. 1996]

Prognostic Significance of Germ Line Polymorphisms of the CYP1A1 andGlutathione S-Transferase Genes in Patients with Non-Small CellLung Cancer1

Isao Goto, Shuichi Yoneda, Mitsunobu Vaniamolo, and Kaname Kawajiri2

Pulmonary Medicine Clinic Â¡1.G.. S. Y.} and Thoracic Surgen' Clinic Â¡M.Y.I, Saitama Cancer Center Hospital, and Department of Biochemistry JK. K.J. Saitama Cancer Center

Research Institute. 8IN Ktimuro. Intt-machi. Kitaadachi-gitn. Saitama 362. Japan

ABSTRACT

CYP1A1 is responsible for the metabolic activation of benzo(a)pyrenein cigarette smoke, and high susceptibility to smoking-related lung cancer

has been associated with the Mspl polymorphism of the CYP1A1 gene.Individuals with a susceptible i'YI'l.M genotype have been found to be at

remarkably high risk when the genotype is combined with a deficientMu-class glutathione S-transferase (GSTM1) genotype. In this study, we

investigated the relationship between germ line polymorphisms of thesegenes and clinical characteristics or survival rates in 232 patients withnon-small cell lung cancer (NSCLC). Statistical analysis revealed a signif

icant association (P < 0.05) of the Msp\ polymorphism of the CYPIA1 genewith histolÃ³gica! type, performance status (general conditions of patients),and the extent of the primary tumor (T-factor). On the other hand, the

GSTM1 polymorphism was significantly associated with performance status, the extent of regional lymph node metastasis (N-factor), and the extentof distant metastasis i M-factor I. NSCLC patients with at least one sus

ceptible alÃeleof the Mspl polymorphism of the CYP1AI gene [heterozygous genotype B or a rare homozygous genotype C; n = 131; mediansurvival time iMS'l'i = 24.2 months) were associated with a shortened

survival compared with those with nonsusceptible homozygous alÃeles(genotype A; n = 101; MST = 65.2 months; P = 0.005 by log-rank test).Smokers with susceptible genotypes (n = 104; MST = 18.2 months) were

markedly associated with a shortened survival compared with those withgenotype A (n = 76; MST = 69.2 months; P = 0.024); such an associationwas not found among nonsmokers by genotypes. Genotype-dependent

survival was also observed in patients at an advanced stage of disease(/' = 0.010), but not in those at an early stage of disease (/' = 0.382).

Patients with the susceptible CYP1A1 genotype had remarkably shortenedsurvivals when the genotype was combined with a deficient genotypeGSTMl(-) (P = 0.017; degree of freedom = 3). Multivariate analysis by

the Cox proportional hazards model also revealed that the CYP1AI polymorphism was an independent prognostic factor in patients at a nonre-sectable advanced stage of NSCLC (P = 0.005; hazard ratio = 1.98; 95%confidence interval, 1.24-3.17).

INTRODUCTION

Human lung carcinogenesis usually requires exposure to the pro-

carcinogens that are contained mainly in cigarette smoke, and manyaromatic hydrocarbons, such as benzo(o)pyrene, first require metabolic activation by CYPIAI3 to their ultimate DNA-binding, muta-

genie, or carcinogenic forms (1). Activated metabolites of benzo-

(a)pyrene are subjected in part to metabolic detoxication by GSTM1(2). Thus, the expression and catalytic activity of these carcinogen-

Received 1/30/96; accepted 6/18/96.The costs of publication of this article were defrayed in part by the payment of page

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

1Supported in part by a special grant for advanced research on cancer from the

Ministry of Education. Science. Sports and Culture of Japan and a research grant from theMinistry of Health and Welfare of Japan for 2nd-Term Comprehensive 10-Year Strategy

for Cancer Control.2 To whom requests for reprints should be addressed. Fax: 81-48-722-1739.'The abbreviations used are: CYP1AI. human cytochrome P450 IAl: NSCLC.

non-small cell lung cancer; GSTM1. glutathione S-transferase Mu-l; MST. mediansurvival time: PS. performance status: df. degree of freedom; CI, confidence interval:T-factor. the extent of the primary tumor; N-factor. the extent of regional lymph nodemetastasis: M-factor. the extent of distant metastasis.

metabolizing enzymes in the lung and their metabolic balance may beimportant determinant host factors underlying lung cancer (3, 4).

We have shown previously that two mutually linked CYPIAI genepolymorphisms, the Mspl polymorphism located in the 3'-flanking

region of the gene and the Ile-Val polymorphism at amino acid residue

462 in the heme binding region of CYPIAI protein, are associatedwith high susceptibility to smoking-associated lung cancer (5-7).

Genotype C in the Msp\ polymorphism and genotype Val/Val in theIle-Val polymorphism among patients with squamous cell carcinoma

were more than twice as common relative to their frequency amongthe controls. Patients with genotype C or Val/Val contracted carcinoma after smoking fewer cigarettes than those with genotypes A orIle/Ile, respectively (8, 9). Relative risks were estimated at cigarettedose levels by case-control studies, revealing that individuals with

these genotypes were at an especially high risk of carcinoma at a lowdose level of cigarette smoking. Furthermore, individuals with thesusceptible CYPÃŒAIgenotype were shown to be at a remarkably highrisk when the genotype was combined with a deficient GSTMl genotype, GSTMI(-) (10). The functional significance of the polymor

phic human CYPIAI gene may be an association with inducibleexpression or difference in the catalytic activity of the CYPIAIenzyme (7, 11, 12). The GSTMI(-) genotype was associated com

pletely with deficient GSTMl enzyme activity (13).Lung cancers are characterized by multiple genetic changes, which

include the activation of proto-oncogenes and the inactivation of

tumor suppressor genes (14, 15). Among the genetic alterations identified in NSCLC. mutations in the p53 gene (16-18) and an allelicdeletion of chromosome 3p (19-21) seem to be the most frequent

targets in the process of lung carcinogenesis. Furthermore, it has beenreported recently that p53 (22, 23) or ras (24-26) mutations as well

as 3p deletions (23) in patients with NSCLC were found to beassociated not only with the genesis and progression of lung cancerbut also with shortened survival as predictors of poor prognosis,although this notion remains controversial (27). In this regard, it isinteresting to note that mutation frequencies of the two target genes inlung cancer, the p53 and Ki-Ras genes, were significantly affected by

the CYPIAI and GSTMl genotypes that are predisposing factors (28).In addition, Okada et al. (29) demonstrated that lymph node or distantmetastasis was more frequently observed among patients with squamous cell carcinoma with genotype C of the CYPIAI gene. Therefore,in this study, we investigated the relationship between various clinicalcharacteristics or survival rates in 232 patients with NSCLC and germline polymorphisms of the CYPIAI and GSTMl genes. We report herethat the patients with genotypes susceptible to lung cancer are associated with a shortened overall survival.

PATIENTS AND METHODS

Patients. We studied 232 Japanese patients (166 males and 66 females)with histologically confirmed primary NSCLC who received treatment between 1988 and 1993 at the Saitama Cancer Center Hospital. None of thepatients had received previous treatment by chemotherapy, radiotherapy, orsurgery. HistolÃ³gica! classification of tumors was conducted according to the

3725

Research. on November 28, 2020. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

C}'fi Al POLYMORPHISMS AM) PR(XÃŒN()SIS IN NSCLC

guidelines proposed by the WHO (30). According to the criteria, the casesconsisted of 124 adenocarcinomas. 93 squamous cell carcinomas, and 15 largecell carcinomas. PS was documented according to classification by the EasternCooperative Oncology Group (31). We used the American Joint CommitteeStaging system to stage the disease or to classify the tumors-nodes-metastasisfactors (32). Stages I-II were considered to be early stage, and stages Illa-IV

were considered to be advanced stage of the disease. We found that 65 patientshad stage I disease. 20 had stage II. 46 had stage Ilia. 42 had stage Illb. and59 had stage IV disease. A systematical review of pathological or clinicaldiagnoses was conducted. Of the group. 129 patients had been treated bysurgery as a routine therapeutic operation, whereas the remaining 103 patientshad been treated by chemotherapy and/or radiotherapy. Smoking habits forcases were reviewed troni medical records, in which routine questions wereasked of all outpatients at the time of the first visit to the hospital. The studywas approved by the medical ethical review committee in this hospital (Institutional Review Board).

Identification of the CYP1A1 and GSTMI Genotypes in Patients withNSCLC. Blood samples (5-10 ml) were obtained from the 232 patients with

NSCLC from whom we could obtain appropriate informed consent, andgenomic DNAs were isolated from the peripheral lymphocytes of the samplesusing standard procedures with proteinase K digestion and phenol-chloroform

extraction. The CYPIAI genotypes ascribed to the presence (CCGG) or absence (CTGG) of the Mspl site at the 264th base from the polyadenylateadditional signal in the 3'-flanking region were identified by the PCR-restric-

tion enzyme digest genotyping method as described previously (6). GSTMIgenotyping was conducted by the presence or absence of a PCR productaccording to the methods of Comstock el al. (33) and Groppi et al. (34).

Statistical Analysis. The Pearson jf test was used to compare the association between the CYPIAI or GSTMI genotype and several clinical andpathological parameters. The Kaplan-Meier method (35) was used to estimate

the probability of survival as a function of time, and survival differences wereanalyzed using the log-rank test (36). The Cox proportional hazards modeling

technique was used to identify which independent factors had a jointly significant influence on overall survival (37).

RESULTS

Overall Survival Rate in Patients with NSCLC in Relation tothe CYPIAI or GSTMI Polymorphisms. The clinical characteristicsof the 232 NSCLC patients examined are summarized in Table 1 inrelation to the genotypes of the CYPIAI or GSTMI gene. The Msp\polymorphism of the CYPIAI gene was associated with histolÃ³gica!type, PS, T-factor. and the ratio of subjects who had surgery:subjects

who did not have surgery with statistical significance (P < 0.05) andwas associated with N- and M-factors with borderline significance

(0.05 < P < 0.1). On the other hand, the GSTMI polymorphism wasassociated with PS, N- and M-factors. clinical stages, and the rate of

operation (P < 0.05). Namely, patients with susceptible genotypesB + C or GSTMI(-) had characteristics of more clinically advanced

disease compared to those with genotype A or GSTMl( + > whendiagnosed with NSCLC. It is suggested from these observations thatthe CYPIAI and GSTMI genotypes, which are predisposing factors tosmoking-related lung cancer, may be associated with survival rates in

patients with NSCLC.We first analy/.ed the association between CYPIAI genotypes and

overall survival in all patients with NSCLC (Table 2). The MST of24.2 months among patients with genotypes B + C (n = 131) wassignificantly shorter than that of 65.2 months for patients with geno-

Table 1 Characteristics of patients with NSCLC in relatinn In polymorphism! of the CYPIAI anil GSTMI genes

Variables No. of patients

CYPIAI polymorphismA/.v/il genotypes

A B + C

GSTMI polymorphismGSTMI genotypes

<+) f-l

CaseAge(mean Â±SD)<6565

SSexMaleFemaleSmoking

statusSmokersNever

smokedHistologyAdenocarcinomaSquamous

cellcarcinomaLargecellcarcinomaPS0-12-4T-factor1234N-

factor0123Clinical

Mage1IIlib1MbIVOperationPerformedNot

performed"P was calculated hy x~ lest.232122IKI1666618051124931518645399]3660%266347652046425912910310161.

6Â±9.55645673476246234587132447171350121')2033102216206437'

P calculated for adenocarcinoma trr.vM.ysquamous cell carcinoma and largeeel'P calculated lor Ml) (stage I-llIb)versus Ml (stage IV).13162.8

Â±9.3666599321042762591099321550194746144427321024263965661

carcinoma.37260.4440.1220.5380.033''0.030<O.OOI0.0570.3080.084'0.03710161.3

Â±10.3524975268318583858813IS441623381829162514232019643713163.1Â±8.570619140973366551098322\532037588343140623224065660.7680.4230.1690.286''0.0250.8220.0240.0320.042'0.037



CYPIAI POLYMORPHISMS AND PROGNOSIS IN NSCLC

Table 2 Overall survival in patients with NSCLC in relation to CYPIAI und GSTMI polymorphisms

Ms/*/ genotypesCasesNSCLCSmoking

statusSmokersNever

smokedHistologyAdenocarcinomaSquamous

cellcarcinomaClinical

slageEarlystages(I-II)Advanced

stages (IIIa-lV)AB+CAB+CAB+CAB+CAB+CAB+CAB+C1011317610424276262345943425889CYPIAI

polymorphismMST (mo) 3-yr survival rate(65.224.269.218.265.229.044.618.0NR''34.2NR''108.523.610.459.840.356.039.470.843.354.636.569.845.788.380.738.520.2%)P" GSTMI genotypesCases(

+10.005(-1(

+i0.024(-)(+)0.102

(-1(

+10.033(-)<+)0.025

(-)<

+)0.382(-)(+>0.010

(-)101131839718335866385539466285GSTMI

polymorphismMST (mo) 3-yr survival rale (%)P"50.129.042.625.364.438.138.423.6NR''41.8NR''137.821.712.454.844.252.341.666.750.950.940.760.450.784.684.235.422.20.0920.1160.4490.1430.3950.7610.061

' P is calculated by log-rank lest.' NR. median survival not reached.

1.0-

0.8-

I(0â€¢¿�0̂.6-

35

I"co~5 0.4-

E

O

0.2-

0.0

2 3

YearsFig. 1. Kaplan-Meier survival curve tor all patients with NSCLC. based on the

combined genotypes of the CYPIAI and GSTMI genes. Genotypes A and GSTM!t + )(O).n = 45; genotypes A and GSTMIt -)(â€¢).Â» = 56: genotypes ÃŸ+ C and GSTMK + I (O).ii = 56; genotypes ÃŸ+ C and GSTMI(-) (â€¢).n = 75. P = 0.017; df = 3.

type A (n = 101; P = 0.005 by log-rank test). When the patients weredivided into 180 smokers and 51 who never smoked, genotype-

dependent survival was clearly observed among smokers (Table 2).The MST of 18.2 months among smokers with genotypes B + C wassignificantly shorter than that of 69.2 months among those withgenotype A (P = 0.024). Furthermore, we investigated this relation

ship in more detail by taking into account either the histological typesof NSCLC or the stages of the disease. In patients with adenocarci-noma (;i = 124), the MST of 18.0 months for genotypes B + C(n = 62) was significantly shorter than that of 44.6 months for

genotype A (n = 62; P = 0.033; Table 2). Similarly, the log-rank testrevealed that patients with genotypes B + C (n = 59) in squamous

cell carcinoma had a significantly poorer prognosis of MST than didpatients with genotype A (n = 34; P = 0.025). The 3-year survival

rate for genotypes B + C or genotype A in all patients with NSCLCwas 40.3 and 59.8%, respectively; for adenocarcinoma. the figureswere 36.5 and 54.6%. respectively; and for squamous cell carcinoma.45.7 and 69.8%. respectively: indicating significant differences ingenotypes irrespective of histological types.

Next, we divided all the NSCLC patients into two groups: earlystage or advanced stage of the disease (Table 2). When a comparisonwas made within the group with early stage disease, the effect of theCYPIAI genotypes on survival was not statistically significant(P = 0.382). This could be due to the longer overall survival amongpatients with early stage disease compared to those with advanced-stage disease, and our follow-up study may not have been long enough

to find a survival difference in terms of genotype. By contrast, whena comparison was made within the group with advanced-stage disease,

the difference in survival times between patients with genotypesB + C (n = 89; MST = 10.4 months) and those with genotype A(n = 58; MST = 23.6 months) was highly significant (P = 0.010).

The differences in survival time between patients with genotypesB + C and those with genotype A with advanced-stage adenocarci

noma or squamous cell carcinoma were also statistically significant,but were not statistically significant in patients with early stagedisease (data not shown).

We also analyzed the association between the GSTMI polymorphism and overall survival. The survival in NSCLC patients withgenotype GSTMi(-) (MST = 29.0 months; n = 131) was shorterthan that in patients with genotype GSTMIt + ) (MST = 50.1 months;n = 101), with borderline significance (P = 0.092 in Table 2). When

we compared this relationship by taking the stage or histological typeof the disease into account, we found a poorer prognosis amongpatients with GSTMI(-) (MST = 12.4; n = 85) than among those

Table 3 Overall sun'ival Â¡nputtenls with NSCLC bused Â«i

CYPIAI and GSTMI nenescinnhined ^eniitvpes of the

CombinedgenolypesCYPIAIAAB

+CB+ CNo.

ofGSTMIpatients(

+ >45<->561

+ )56<-)75No.

ofdeaths21313251MST"(mo)100.041.822.824.23-yrsurvivalrale

(%)66.454.645.736.6

" P = 0.017 by log-rank test; df = 3.

3727



CYPIAl POLYMORPHISMS AND PROGNOSIS IN NSCLC

0.012 16

Months

20

Fig. 2. Kaplan-Meier survival curve for all patients with nonresectable advanced-stageNSCLC (stage IHa-IV) with respect to the Mspl polymorphism of the CYPIAl gene.Genotype A (G), n = 34; genotypes B + C (O), n = 64; P = 0.003.

with GSTMl( + > (MST = 21.7; n = 62) at the advanced stage ofNSCLC (P = 0.061 by log-rank test; Table 2). In addition, we

examined the effect of combined genotyping of genes for the carcinogen-metabolizing enzymes on prognosis. The Kaplan-Meier survival

curve for patients with NSCLC in relation to the combined genotypesof CYPIAl and GSTMI genes is shown in Fig. 1. As shown in Table3, the shortest survival time was observed in patients with the combination of B + C and GSTMl(-) (n = 75; MST = 24.2 months;

3-year survival rate was 36.6%), whereas the longest survival was

obtained in patients with combination genotypes of A and GSTMI(+)

(n = 45; MST = 100 months; 3-year survival rate was 66.4%). The

difference in overall survival among the four groups of patients withdifferent combined genotypes was found at the statistically significantlevel of P = 0.017 (log-rank test; df = 3). A similar survival profile

was obtained when a comparison was made within the group withadvanced-stage NSCLC in relation to the combined genotypes(P = 0.013; df= 3; data not shown).

CYPIAl Genotypes as an Independent Prognostic Factor forPatients at a Nonresectable Advanced Stage of NSCLC. As mentioned, the Mspl polymorphism of the CYPÃŒAIgene was significantlyassociated with several potential prognostic factors and overall survival rates in patients with NSCLC. To investigate the prognosticsignificance of the germ line polymorphism of the CYPIAl gene, weanalyzed the association between CYPIAl genotypes and MST inpatients with nonresectable advanced NSCLC (n = 98) in terms ofseveral prognostic factors. The Kaplan-Meier survival curves for allpatients with nonresectable advanced-stage (IIIa-IV) NSCLC in rela

tion to the CYPIAl genotypes are shown in Fig. 2. As shown in Table4, the MST of patients with genotypes B + C was shorter than that ofpatients with genotype A at the level of significance or borderlinesigificance when a comparison was made within each category (including age. sex, smoking status, histological type, PS, clinical stages,and GSTMI genotypes). In addition, the MST of 8.0 months forgenotypes B + C (n = 40) was significantly shorter than that of 21.7months for genotype A (n = 12) in patients with T4 (P = 0.008).Similarly, the log-rank test revealed that patients with genotypesB + C (n = 52) with N2-N3 had a significantly poorer prognosis ofMST than patients with genotype A (n = 24; P = 0.014).

Table 5 shows the result of univariate or multivariate analyses usingCYPÃŒAIgenotypes and covariables including age, sex, smoking status, histology, PS, clinical stages, and GSTMI genotypes. The Msplpolymorphism of the CYPIAl gene was shown to be an independentprognostic indicator in patients with nonresectable advanced-stage

NSCLC. Specifically, patients with genotypes B + C of the CYPÃŒAIgene survived for a significantly shorter period of time than those withgenotype A (hazard ratio from 1-1.98; 95% CI, 1.24-3.17;P = 0.005). A similar result was obtained when multivariate analysis

Table 4 Characteristics of patients with nonresectable advanced-stage NSCLC in relation to polymorphisms of the CYPIAl genes

VariablesOverallAge<6565SSexMaleFemaleSmoking

statusSmokersNever

smokedHistologySquamous

cellcarcinomaNon-sq'PS0-12-4Clinical

stageIliainbIVGSTMI(

+1<-)All9847517325791832665641II)30583662No.

of patientsA34151426829473721]2312191222B+C6432324717501425393529718392440P"0.5790.7430.2420.0630.3980.7540.829A17.521.914.715.617.520.010.0NR'17.518.88.932.621.713.921.914.7MST

(mo)ÃŸ+C7.97.97.57.98.67.97.26.49.410.16.68.010.17.46.67.9P"0.0030.0220.0440.0300.0260.0010.4870.0120.0490.0270.0710.0400.0850.0700.0320.042MST

(mo)All10.010.010.19.510.49.510.07.210.812.46.612.015.08.611.68.9P10.2210.7840.7660.786â€¢C0.0010.0780.233

a P was calculated by \~ lest fÂ°rgenotype A vs. B + C.

P was calculated by log-rank test for genotype A vs. B + C.P was calculated by log-rank test for each category.

' NR, median survival not reached.' Non-sq. adenocarcinoma and large cell carcinoma.

3728



CYPIAl POLYMORPHISMS AND PROGNOSIS IN NSCLC

Table 5 Cox proportional hazard regression analvsis of 98 patients with nonresectabieadvanced-stage NSCLC

VariablesUnivariate

analysisAge(S65 W.65<)Sex(female vs.mule)Smoking

(smokers vs. neversmoked)Histology(Sq. vs.non-sq.)uPS

(0-1 vs.2-4)Stage

(Ilia vs. Hlb vi.IV)GSTMl(/ + ) vs.(-)]CYPIAl

(A vs. B +C)MultivariateanalysisPS

(0-1 vs.2-4)CYPIAl(A vs. B + QHazard

ratio1.301.071.080.942.151.341.312.022.181.9895%CI0.85-1.990.66-1.720.636-1.850.59-1.491.39-3.340.96-1.870.840-2.041.27-3.211

.40-3.391.24-3.17P0.2230.7840.7670.786<O.OOI0.0860.2350.0030.0010.005

" Sq.. squamous: non-sq.. adenocarcinoma and large cell carcinoma.

was carried out among all cases with NSCLC (hazard ratio from1-1.78; 95% CI, 1.18-2.46; P = 0.004; data not shown).

DISCUSSION

In this paper, we have shown that a germ line Mspl polymorphismof the CYPIA] gene is associated with various clinical parametersresponsible for the poorer prognosis in patients with NSCLC; it is alsoan independent factor indicative of prognostic significance at thenonresectabie advanced stage of the disease. Although there is nodirect evidence clarifying how this predisposing factor for lung cancerdetermines overall survival in patients with NSCLC, our recent findings may provide a clue to the mechanism underlying this relationship. We recently examined the frequency of p53 mutations in genotypes of the CYPIAl gene using surgical specimens of NSCLCpatients who were smokers, and we found that patients with thesusceptible genotype C of the CYP1AÃŒgene had a 4.5-fold(P < 0.005; odds ratio, 4.48; 95% CI, 1.64-12.26) higher risk of

having a mutation of the p53 gene than those with the nonsusceptiblegenotype A (28). The aberration frequency of the p53 gene amongpatients with genotype B (heterozygous) was intermediate betweenthose with genotypes A and C. The relative mutation frequency witha combination of genotype C of the CYPÃŒAIgene and GSTMl(-)

remarkably increased 9-fold for p53 gene mutations compared with

the combination of genotypes A and GSTMI(+). These results showthat germ line polymorphisms of these two genes, which are associated with genetic predisposition for lung cancer, are related to cigarette smoking-associated p53 mutations. Similar results were alsoobtained between Ki-Ras mutations and the genotypes of these two

genes (28).Mutations of the p53 gene, which encodes a nucleoprotein func

tioning as a transcription factor that regulates cell cycle-related genes,

are one of the most common genetic abnormalities found in all typesof lung tumors (16-18). A crucial role of the p53 gene in lungcarcinogenesis is also underscored by the finding that wild-type p53

suppresses the growth of human lung cancer cell lines bearing othermultiple genetic lesions (38). Somatic genetic alteration of the p53gene seems to be a candidate in the precancerous genetic event foundin both metaplastic and dysplastic lesions of the lung (39, 40). Accordingly, it is likely that p53 gene mutation may facilitate furthergenetic alterations that are involved in tumor progression and/ormÃ©tastases,resulting in a poor prognosis. Indeed, p53 gene mutationhas been reported to increase the incidence of 3p deletion (23) ormetastatic potential (41), and the presence of p53 gene mutations inNSCLC patients may be independently associated with a shortenedoverall survival (22, 23). We are now investigating the associationamong CYPIA 1 genotypes, p53 mutations, and survival rate in termsof cigarette smoking dose.

The longer overall survival among patients with early stage diseasecompared to those with advanced-stage disease was dependent mainly

on whether or not the patients had received surgical treatment. Of thepatients with early stage NSCLC, 80 of 85 (94%) had been treatedwith curative surgery, whereas only 49 of 147 (33%) with advanced-

stage disease had been treated surgically. When a comparison wasmade in the group without surgical treatment (n = 103), the MST of8.0 months among patients with genotypes B + C (n = 66) was

significantly shorter than that of 18.8 months among patients withgenotype A (n = 37; /> = 0.0027 by log-rank test). On the other hand,

no significant survival difference was observed among surgicallytreated patients in relation to CYPIAl genotypes (P = 0.740). In this

regard, it is interesting to mention that the proportion of surgicallytreated cases among patients with NSCLC differed in accordance withthe genotypes of the CYPIAl gene. Among 101 NSCLC patients withgenotype A, 64 and 37 individuals had been treated with or withoutcurative surgery, respectively. By contrast, among 131 patients withgenotypes B + C, 65 and 66 individuals had been treated with orwithout surgery, respectively. The difference can be explained by theassociation of CYPÃŒA1genotypes with T-. N-, and M-factors

(Table 1).The CYP1A1 and GSTMl polymorphisms, which are heritable

predisposing factors to smoking-related lung cancer, are also related

to the prognosis for lung cancer. Individuals with combination genotypes with a high risk of lung cancer may have an increased probability of mutations in target genes, resulting in a shortened overallsurvival. In clinical trials among patients with nonresectabie advancedNSCLC, easy and precise identification of CYPÃŒAÃŒgenotypes usinglymphocyte DNA may have to be considered to evaluate the effect oftreatment on overall survival.

REFERENCES

1. Shimada, T.. Yun. C-H.. Yamazaki. H.. Gautier. J-C.. Beaune. P. H.. and Guengerich,F. P. Characterization of human lung microsomal cytochrome P450 1A1 and its rolein oxidation of chemical carcinogens. Mol. Pharmacol., 41: 856-864, 1992.

2. Mannervik. B.. and Danielson. U. H. Glutathione transferase: structure and catalyticactivity. Crit. Rev. Biochem. Mol. Biol.. 23: 281-337, 1988.

3. Nebert. D. W. Role of genetics and drug metabolism in human cancer risk. Mutai.Res., 247: 259-266, 1991.

4. Kawajiri, K.. and Fujii-Kuriyama. Y. P450 and human cancer. Jpn. J. Cancer Res., 82:1325-1335, 1991.

5. 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 P450IA1 gene. FEBS Lett., 263: 131-133, 1990.

6. Hayashi. S-I.. Watanabe, J.. Nakachi. K.. and Kawajiri. K. Genetic linkage of lungcancer-associated Mspl polymorphisms with amino acid replacement in the heme

binding region of the human cytochrome P450 IA1 gene. J. Biochem. (Tokyo), 110:407-411. 1991.

7. Kawajiri. K.. Nakachi. K.. Imai. K., Watanabe. J.. and Hayashi. S-I. The CYPIAlgene and cancer susceptibility. Crit. Rev. Oncol. Hematol.. 14: 77-87. 1993.

8. Nakachi. K., Imai. K.. Hayashi, S-I.. Walanabe. J.. and Kawajiri. K. Genetic susceptibility to squamous cell carcinoma of the lung in relation to cigarette smoking dose.Cancer Res., 51: 5177-5180. 1991.

9. Nakachi, K.. Imai, K., Hayashi, S-I.. and Kawajiri. K. Polymorphisms of the CYPIA Iand glutathione 5-transferase genes associated with susceptibility to lung cancer inrelation to cigarette dose in a Japanese population. Cancer Res.. 53: 2994-2999,1993.

10. Hayashi. S-I.. Watanabe. J.. and Kawajiri. K. High susceptibility to lung canceranalyzed in terms of combined genotypes of P450IAI and Mu-class glutathioneS-transferase genes. Jpn. J. Cancer Res.. 83: 866-870. 1992.

11. Petersen. D. D.. McKinney. C. E., Ikeya. K., Smith. H. H., Bale, A. E., McBride.O. W., and Nebert. D. W. Human CYPIAl gene: cosegregation of the enzymeinducibility phenotype and RFLP. Am. J. Hum. Genet.. 48: 720-725, 1991.

12. Crofts, F., Taioli, E., Trachman. J., Cosma. G. N.. Currie. D.. Toniolo, P., and Garte,S. J. Functional significance of different human CYPIAl genotypes. Carcinogenesis(Lond.). 15: 2961-2963. 1994.

13. Seidegard, J.. Vorachek, W. R., Pero, W. R., and Pearson. W. R. Hereditary differences in the expression of the human glutathione transferase active on /ra/i.v-stilbeneoxide are due to a gene deletion. Proc. Nati. Acad. Sci. USA, 85: 7293-7297. 1988.

14. Yokota, J., and Sugimura, T. Multiple steps in carcinogenesis involving alterations ofmultiple tumor suppressor genes. FASEB J., 7: 920-925, 1993.

15. Weinberg. R. A. Tumor suppressor genes. Science (Washington DC). 254: 1138-

1145. 1991.

3729



CYFIAI POLYMORPHISMS AND PROGNOSIS IN NSCLC

16. Takahashi, T.. Ã‘au. M. M.. Chiba. I., Birrer. M. J.. Rosenberg. R. K.. Vinocour. M..Levitt. M.. Pass. H.. Gazdar. A. F., and Minna. J. D. p53: a frequent target for geneticabnormalities in lung cancer. Science (Washington DC}. 246: 491-494. 1989.

17. Kishimoto. Y.. Murakami. Y., Shiraishi. M.. Hayashi. K.. and Sekiya. T. Aberrationsof p53 tumor suppressor gene in human non-small cell carcinomas of the lung. CancerRes., 52: 4799-4804. 1992.

18. Chiba, I.. Takahashi. T.. Ã‘au, M. N.. D'Amico, D.. Curie!. D. T., Mitsudomi, T..

Buchhagen, D. L., Carbone, D., Piantudosi, S.. Koga. H., Reissman. P. T.. Slamon.D. J., Holmes, E. C., and Minna. J. D. Mutations in the p53 gene are frequent inprimary, resected non-small cell lung cancer. Oncogene. 5: 1603-1610. 1990.

19. Brauch. H., Johnson, B.. Hovis, J.. Yano, T., Gazdar, A., Pettengill. O.. Graziano. S..Sorenson. G., Poiesz, B., Minna, J.. Linehan. M., and Zbar. B. Molecular analysis ofthe short arm of chromosome 3 in small-cell and non-small cell carcinoma of the lung.N. Eng. J. Med.. 317: 1109-1113. 1987.

20. Kok. K.. Osinga. J.. Carritt. B.. Davis, M., van der Hout, A., van der Veen, A.,Landsvater. R., de Leij, L.. Berendsen. H.. Postmus. P.. Poppema. S., and Buys. C.Deletion of a DNA sequence at the chromosomal region 3p21 in all major types oflung cancer. Nature (Lond.), 330: 578-581. 1987.

21. Yokota. A.. Wada. M.. Shimosato. Y.. Terada. M.. and Sugimura T. Loss of het-erozygosity on chromosome 3. 13, and 17 in small cell carcinoma and chromosome3 in adenocarcinoma of the lung. Proc. Nati. Acad. Sci. USA. 84: 9252-9256. 1987.

22. Mitsudomi. T.. Oyama. T.. Kusano. T.. Osaki. T.. Nakanishi. R.. and Shirakusa. T.Mutations of the pS3 gene as a predictor of poor prognosis in patients with non-smallcell lung cancer. J. Nati. Cancer Inst.. 85: 2018-2023, 1993.

23. Horio. Y.. Takahashi. T.. Kuroishi. T., Hibi. K.. Suyama. M., Numi. T.. Shimokata.K., Yamakawa, K., Nakamura, Y., Ueda. R., and Takahashi. T. Prognostic significance of p53 mutations and 3p deletions in primary resected non-small cell lungcancer. Cancer Res., S3: 1-4, 1993.

24. Mitsudomi. T.. Steinberg, S. M., Oie. H. K.. Mulshine. J. L.. Phelps. R.. Viallet. J..Pass, H., Minna. J. D.. and Gazdar, A. F. ras gene mutations in non-small cell lungcancers are associated with shortened survival irrespective of treatment intent. CancerRes.. 51: 4999-5002. 1991.

25. Slebos. R. J. C.. Kibbelaar, R. E.. Dalesio. O., Kooistra, A.. Stam. J.. Meijer. C. J. L.M., Wagenaar. S. S., Vanderschueren. R. G. J. R. A., van Zandwijk, N.. Mooi, W. J..Bos. J. L., and Rodenhuis. S. K-ros oncogene activation as a prognostic marker inadenocarcinoma of the lung. N. Eng. J. Med.. 323: 561-565, 1990.

26. Sugio. K.. Ishida, T.. Yokoyama, H.. Inoue. T.. Sugimachi. K., and Sasazuki, T. rasgene mutations as a prognostic marker in adenocarcinoma of the human lung withoutlymph node metastasis. Cancer Res.. 52: 2903-2906. 1992.

27. McLaren. R.. Kuzu. I.. Dunnill. M.. Harris. A.. Lane. D.. and Gatter. K. C. Therelationship of p53 immunostaining to survival in carcinoma of the lung. Br. J.Cancer. 66: 735-738. 1992.

28. Kawajiri. K.. Eguchi. H.. Nakachi. K.. Sekiya. T.. and Yamamoto, M. Association ofCYP1AI germ line polymorphisms with mutations of the p53 gene in lung cancer.Cancer Res.. 56: 72-76, 1996.

29. Okada, T.. Kawashima. K.. Fukushi. S., Minakuchi. T.. and Nishimura, S. Associationbetween a cytochrome CYPIAl genotype and incidence of lung cancer. Pharmaco-genetics, 4: 333-340. 1994.

30. WHO. Histological typing of lung tumors. Am. J. Clin. Pathol.. 77: 123-136. 1982.

31. Oken. M. M.. Creech. R. H.. Tormey, D. C.. Horion. J.. Davis, T. E.. McFadden.E. T.. and Carbone. P. P. Toxicity and response criteria of the Eastern CooperativeOncology Group. Am. J. Clin. Oncol., 5: 649-655. 1982.

32. Moutain. C. F. A new international staging system for lung cancer. Chest. 89 (Suppl.):225S-233S, 1986.

33. Comstock, K. E.. Sanderson. J. S., Clafin. G., and Henner, W. D. GSTl gene deletiondetermined by PCR. Nucleic Acids Res.. 18: 3670, 1990.

34. Groppi. A.. Coutelle. C.. Fleury. B.. Iron. A.. Begueret. J.. and Couzigous. P.Glutathione S-transferase class in French alcoholic cirrholic patients. Hum. Genet.,87.-628-630, 1991.

35. Kaplan. E. L.. and Meier. P. Nonparametric estimation from incomplete observation.J. Am. Stat. Assoc., 53: 457-481. 1958.

36. Mantel, N. Evaluation of survival data and two new rank order statistics arising in itsconsideration. Cancer Chemother. Rep.. 50: 163-170. 1966.

37. Cox, D. R. Regression models and life tables. JR. Stat. Soc.. 34: 187-220, 1972.

38. Takahashi. T.. Carbone. D., Takahashi. T., Ã‘au. M. M.. Hida, T.. Linnoila. !.. Ueda,R.. and Minna, J. Wild-type but not mutant p53 suppresses the growth of human lungcancer cells bearing multiple genetic lesions. Cancer Res., 52; 2340-2343, 1992.

39. Klein. N., Vignaud, J. M.. Sadmi. M.. Plenat. F.. Borelly. J.. Duprez. A., Martinet, Y.,and Martinet. N. Squamous metaplasia expression of proto-oncogenes and p53 in lungcancer patients. Lab. Invest.. 68: 26-32, 1993.

40. Sozzi. G., Miozzo. M.. Donghi. R.. Pillotti. S.. Cariani. C. T.. Pastorino. U.. Porta.G. P.. and Pierotti, M. Deletion of 17p and/>5.Ã¯mutations in preneoplastic lesions ofthe lung. Cancer Res.. 52. 6079-6082. 1992.

41. Marchetti. A.. Buttitta. F.. Merlo. G.. Diella. F.. Pellegrini. S.. Pepe. S.. Macchiarmi.P.. Chella. A.. Angeletti. A.. Callahan. R.. Bistocchi. M., and Squartini. F. p53alterations in non-small cell lung cancers correlate with metaslastic involvement ofhilar and mediaslinal lymph nodes. Cancer Res.. 53: 2846-2851. 1993.

3730



1996;56:3725-3730. Cancer Res Isao Goto, Shuichi Yoneda, Mitsunobu Yamamoto, et al. Non-Small Cell Lung Cancer

-Transferase Genes in Patients withS and Glutathione CYP1A1Prognostic Significance of Germ Line Polymorphisms of the

Updated version

http://cancerres.aacrjournals.org/content/56/16/3725

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

[email protected] 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/56/16/3725To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



prognostic significance of germ line polymorphisms of the ... · genes and clinical characteristics...

Documents