B

L(tstihrTmt

ocS2

a

Int. J. Radiation Oncology Biol. Phys., Vol. 63, No. 3, pp. 885–891, 2005Copyright © 2005 Elsevier Inc.

Printed in the USA. All rights reserved0360-3016/05/$–see front matter

doi:10.1016/j.ijrobp.2005.07.951

IOLOGY CONTRIBUTION

THE POLYMORPHISM AND HAPLOTYPES OF XRCC1 AND SURVIVALOF NON–SMALL-CELL LUNG CANCER AFTER RADIOTHERAPY

SANG MIN YOON, M.D.,* YUN-CHUL HONG, M.D., PH.D.,† HEON JOO PARK, M.D., PH.D.,‡

JONG-EUN LEE, PH.D.,§ SANG YOON KIM, M.D., PH.D.,� JONG HOON KIM, M.D., PH.D.,*SANG-WOOK LEE, M.D., PH.D.,* SO-YEON PARK, M.S.,† JUNG SHIN LEE, M.D., PH.D.,¶

AND EUN KYUNG CHOI, M.D., PH.D.*

Departments of *Radiation Oncology, �Otolaryngology, and ¶Internal Medicine, Asan Medical Center, College of Medicine,University of Ulsan, Seoul, South Korea; †Department of Preventive Medicine, College of Medicine, Seoul National University,

Seoul, South Korea; ‡Department of Microbiology, College of Medicine, Inha University, Incheon, South Korea; §DNA Link Inc.Seoul, South Korea

Purpose: The X-ray repair cross-complementing Group 1 (XRCC1) protein is involved mainly in the baseexcision repair of the DNA repair process. This study examined the association of 3 polymorphisms (codon 194,280, and 399) of XRCC1 and lung cancer in terms of whether or not these polymorphisms have an effect on thesurvival of lung cancer patients who have received radiotherapy.Methods and Materials: Between January 2000 and April 2004, 229 lung cancer patients with non–small-cell lungcancer in Stages I–III were enrolled. Genotyping was performed by single base primer extension assay using theSNP-IT Kit with genomic DNA samples from all patients. The haplotype of the XRCC1 polymorphisms wasestimated by PHASE version 2.1.Results: The patients consisted of 191 (83.4%) males and 38 (16.6%) females with a median age of 62 (range,26–88 years). Sixty percent of the patients were included in Stage I–IIIa. The median progression-free andoverall survival was 13 months and 16 months, respectively. The XRCC1 codon 194, histology, and stage wereshown to be significant predictors of the progression-free survival. The 6 haplotypes among the XRCC1polymorphisms (194, 280, and 399) were estimated by PHASE v.2.1. The patients with haplotype pairs other thanthe homozygous TGG haplotype pairs survived significantly longer (p � 0.04).Conclusions: Polymorphisms of XRCC1 have an effect on the survival of lung cancer patients treated with radio-therapy, and this effect seems to be more significant after the haplotype pairs are considered. © 2005 Elsevier Inc.

XRCC1 polymorphism, Lung cancer, Radiotherapy, Survival.

tttotoesrae

Rtaom

A

INTRODUCTION

ung cancer is the leading cause of cancer death worldwide1). Despite every effort to improve the treatment result inhese patients, the overall 5-year survival rates for non–mall-cell lung cancer (NSCLC) remain �20% (2, 3). Al-hough a number of clinical prognostic factors have annfluence on the overall survival in NSCLC, none of theseave been shown to have a precise relationship with theesponse and sensitivity to various treatment modalities.herefore, other prognostic factors for NSCLC, includingolecular markers that are based on an individual’s sensi-

ivity to the treatment regimen, are needed.

Reprint requests to: Eun Kyung Choi, M.D., Ph.D., Departmentf Radiation Oncology, Asan Medical Center, College of Medi-ine, University of Ulsan, 388-1 Pungnap-Dong, Songpa-Gu,eoul 138-736, South Korea. Tel: (�82) 2-3010-4432, Fax: (�82)-486-7258; E-mail: ekchoi@amc.seoul.krThe first two authors contributed equally to the present work.This work was supported by grants from the Ministry of Health

nd Welfare (Grant No. 02-PJI-PG10-20599-033), by the National

885

Currently, most studies have focused on the concept ofhe suboptimal DNA repair ability, because a complex sys-em of DNA repair enzymes is critical for the protection ofhe genome from carcinogenic exposure and the preventionf lung and other types of cancer (4–9). It is possible thathe DNA repair mechanism plays a key role in the responsef cancer cells to ionizing radiation, because the cytotoxicffect of radiation is mediated by the single- or double-trand breaks in the DNA (10, 11). There are 4 major DNAepair mechanisms for maintaining the genomic integritygainst damage caused by mutagens: mismatch repair, basexcision repair, nucleotide excision repair, and double-

and D for Cancer Control Program (Grant No. 0320320-2), byhe Ministry of Health and Welfare, by Korea Institute of Sciencend Technology Evaluation and Planning (KISTEP), the Ministryf Science and Technology (MOST), and by the Korean govern-ent through its BAERI Program Republic of Korea.Received May 8, 2005, and in revised form July 11, 2005.

ccepted for publication July 11, 2005.

sfia(rs

ttpDapmsp(thc(as

(Xm(spptha1ws

S

wcNdisasT

citdabt

R

fVdTi

G

A

E

W

S

H

T

S

A

s

E

E

E

886 I. J. Radiation Oncology ● Biology ● Physics Volume 63, Number 3, 2005

trand break repair. The base excision repair removes theragmented or nonbulky base adducts produced by the ox-dative stress, DNA methylating agents, and ionizing radi-tion (12). The X-ray repair cross-complementing Group 1XRCC1) plays an important role in the DNA base excisionepair pathway, giving the possibility that this enzyme hasome relationship with the response to radiotherapy.

The XRCC1 gene was identified by its ability to restorehe DNA repair activity in the Chinese hamster ovary mu-ant cell line EM9 (13) and interact with poly(ADP-ribose)olymerase and DNA ligase III in recognizing and rejoiningNA strand breaks, as well as with DNA polymerase � and

purinic/apyrimidinic endonuclease I (14–17). Polymor-hisms in XRCC1 that result in amino acid substitutionsight alter the efficiency of DNA repair and have functional

ignificance. Shen et al. (18) identified 3 coding polymor-hisms in the XRCC1 gene at codons 194 (Arg to Trp), 280Arg to His), and 399 (Arg to Gln). The functional effect ofhese polymorphisms is unclear, even though some studiesave revealed that amino acid changes at the evolutionaryonserved regions can alter its function (19). Lamerdin et al.20) reported that the XRCC1 gene at codon 194 occurred at

conserved residue in humans, indicating the functional

Table 1. Characteristics of the study population (N � 229)

Variables No. of patients (%)

enderMale 191 (83.4)Female 38 (16.6)

ge (years)�62 126 (55.0)�62 103 (45.0)

COG*0 68 (29.7)1 149 (65.1)2 12 (5.2)eight loss�5% 189 (82.5)5–10% 26 (11.4)�10% 14 (6.1)

tageI–IIIa 137 (60.0)IIIb 92 (40.0)

istopathologySquamous cell carcinoma 126 (55.3)Adenocarcinoma 74 (32.5)Large-cell carcinoma 5 (2.2)Unspecified 23 (10.0)

reatment aimDefinitive radiotherapy 116 (50.6)Postoperative radiotherapy 113 (49.4)

moking (pack-years)�40 114 (49.8)�40 115 (50.2)

lcohol consumption (g/week)�100 66 (49.7)�100 87 (50.3)

* Eastern Cooperative Oncology Group (ECOG) performancetatus.

ignificance of this evolutionary conservation. Lunn et al.

21) measured higher levels of aflatoxin B1 adducts in theRCC1 Arg399Gln polymorphism and suggested that thisight result in a deficient DNA repair capacity. Hu et al.

22) suggested that this polymorphism contributed to theusceptibility to ionizing radiation, as measured by therolonged cell cycle G2 delay. Theoretically, these polymor-hisms can favorably influence the response to radiotherapyhrough the inefficient removal of DNA adducts and the en-ancement of radiation sensitivity. Therefore, this study ex-mined the relationship between 3 polymorphisms (codon94, 280, and 399) of XRCC1 and lung cancer in terms ofhether or not these polymorphisms have an effect on the

urvival of lung cancer patients who received radiotherapy.

METHODS AND MATERIALS

tudy populationBetween January 2000 and April 2004, 229 patients with NSCLC

ere enrolled in this study at Asan Medical Center. The eligibilityriteria were a histologically or cytologically confirmed diagnosis ofSCLC, a surgically resected or unresectable Stage I–III tumor, asetermined by the American Joint Committee on Cancer TNM stag-ng system (23), Eastern Cooperative Oncology Group performancetatus of 0–2, and the use of radiotherapy as a curative or postoper-tive adjuvant. Patients who had received radiotherapy before theurgical resection or had other primary malignancies were excluded.he median follow-up period for all the patients was 16 months.All those participating provided informed consent. The patients

ompleted a self-administered questionnaire on their demographicnformation, including smoking history, level of alcohol consump-ion, other medical history, and their family history of majorisease. The Institutional Review Board at Asan Medical Centerpproved the study protocol. At the time of enrollment, wholelood was collected from the patients and delivered to the labora-ory for genotyping.

adiotherapyAll patients received radiotherapy with 6- or 15-MV X-rays

rom a linear accelerator (Varian Clinac 1800, 2100 C/D, 21EX;arian, Palo Alto, CA). Initially, 23 fractions, 1.8 Gy given onceaily 5 days per week, were delivered to a total dose of 41.4 Gy.he initial radiation field encompassed a volume that included the

psilateral hilum, bilateral mediastinal lymph nodal station with a

Table 2. Distribution of the XRCC1 genotypes

Variables No. of patients (%)

xon 6, codon 194Arg/Arg 93 (42.7)Arg/Trp 104 (47.7)Trp/Trp 21 (9.6)

xon 9, codon 280Arg/Arg 166 (77.9)Arg/His 45 (21.1)His/His 2 (1.0)

xon 10, codon 399Arg/Arg 126 (58.9)Arg/Gln 74 (34.6)

Gln/Gln 14 (6.5)

1mdtntCAtpaawod

G

u

Upwo(BpudSPCCGaAf5

G

A

E

W

S

H

S

A

E

E

E

887Polymorphisms in XRCC1 and survival of NSCLC ● S. M. YOON et al.

.5-cm margin, the ipsilateral supraclavicular fossa, and the pri-ary mass with a 1.5–2-cm margin in patients who received

efinitive radiotherapy. If the postoperative T stage was T1 or T2,he radiation field in patients with postoperative radiotherapy didot encompass the primary tumor bed. During the initial course ofhe 41.4 Gy of radiotherapy, CT simulation (AcQSim, Marconi,oventry, England) was performed to reduce the radiation field.fter the target volumes and critical normal organs were drawn,

he images were then transferred to a 3D planning system (Renderlan, Elekta, Stockholm, Sweden). Three- to seven-field coplanarnd/or noncoplanar beams were used to deliver the booster radi-tion dose. A total dose of 50.4 Gy was delivered to the patientsho had received postoperative adjuvant radiotherapy, and a dosef 65–70 Gy was delivered to the patients who had receivedefinitive radiotherapy.

enotyping assayThe genomic DNA was prepared from peripheral blood samples

sing Puregene blood DNA kit (Gentra Inc., Minneapolis, MN,

Table 3. Univariate analysis of the progression-free anusing Kaplan–M

VariablesNo. ofpatients

Progression-fre

1-year 3-y

enderMale 184 53.8 14Female 37 46.0 21

ge�62 110 51.8 21�62 111 56.7 9

COG0 66 54.6 151 146 53.4 152 9 55.5 22eight lossNo change 130 50.0 13Change 91 57.1 20

tageI–IIIa 132 62.9 18IIIb 89 41.6 11

istopathologySquamous cell carcinoma 121 58.0 18Adenocarcinoma 72 50.0 12Large-cell carcinoma 5 20.0 –Unspecified 22 45.5 9

moking�40 pack-years 108 51.9 13�40 pack-years 112 56.3 17

lcohol�100 g/week 82 47.6 17�100 g/week 85 64.7 21

xon 6, codon 194Arg/Arg 93 62.4 19Arg/Trp 104 45.2 15Trp/Trp 21 42.9 0

xon 9, codon 280Arg/Arg 166 53.0 17Arg/His 45 51.1 11His/His 2 50.0 0

xon 10, codon 399Arg/Arg 126 48.4 11Arg/Gln 74 56.8 20

Gln/Gln 14 71.4 28.6

SA) following the manufacturer’s protocol. SNP scoring waserformed by SNP-IT assays using the SNPstream 25K System,hich is customized to perform fully automated SNP genotypingf DNA samples in 384-well plates with a colorimetric readoutOrchid Biosciences, Princeton, NJ), as described before (24).riefly, a set of 3 primers was designed for each SNP, 2 PCRrimers were selected to amplify a 100–200 base pair productnder standard conditions, and a single-base extension primer wasesigned to be approximately 25 bp in length on one side of theNP site. The sequences for the forward and reverse primers forCR amplification were as follows: 5=-AGGATGAGAGCGC-AACT and 5=-TACTCACTCAGGACCCACGT for R194W, 5=-AGTGGTGCTAACCTAATCTACTCT and 5=-TTCTCCTCG-GGTTTGCC for R280H, and 5=-TAAGGAGTGGGTGCTGGA

nd 5=-ATAAGGAGCAGGGTTGGC for R399Q, respectively.lso, the sequences for the single-base extension primer were as

ollows: 5=-GAGGCCGGGGGCTCTCTTCTTCAGC for R194W,=-AGGGACTGGGGCTGTGGCTGGGGTA for R280H, and 5=-

all survival of lung cancer patients after radiotherapyrvival analysis

ival (%)No. ofpatients

Overall survival (%)

p value 1-year 3-year p value

181 85.4 66.30.641 37 87.6 72.7 0.626

110 85.3 71.30.033 108 87.4 61.4 0.505

66 85.8 71.0142 86.4 65.7

0.997 10 88.9 71.1 0.993

130 88.8 66.10.340 88 81.6 68.7 0.588

133 91.2 73.30.003 85 79.2 58.9 0.007

121 85.6 74.770 83.0 47.4

5 100.0 100.00.07 21 93.3 93.3 0.142

107 87.6 62.90.318 110 85.0 70.9 0.443

82 88.4 62.80.091 85 87.5 73.9 0.411

92 88.4 70.5101 86.6 67.1

0.030 21 73.2 54.3 0.203

166 83.1 66.942 92.7 68.3

0.554 2 100 100.0 0.698

125 85.9 65.772 89.3 72.1

d overeier su

e surv

ear

.1

.6

.8

.0

.2

.1

.2

.9

.9

.2

.2

.2

.5

.1

.0

.9

.1

.2

.4

.4

.5

.1

.9

.3

0.007 14 78.6 57.1 0.895

Atsr0Cec1wapcPaetamwSwtpTsccv

F

etpKtgw

fw

P

wmotpaacepttw

X

vrgsssa3ttshn

888 I. J. Radiation Oncology ● Biology ● Physics Volume 63, Number 3, 2005

CCGCATGCGTCGXCGXCTGCCCTCCC for R399Q, respec-ively. The automated handling of liquid by robotics was used toet up 5 �L PCR reactions in 384-well microtiter plates. Each PCReaction contained the following: 10.0 ng of DNA, 1� PCR buffer,.125 units of AmpliTaq Gold DNA polymerase (ABI, Foster City,A), 3.0 mM MgCl2, 0.25 mM of each dNTP, and 0.5 pmol ofach primer. Reactions were incubated at 95°C for 10 min, thenycled 35 times at 95°C for 30 s, at 55°C for 1 min, and 72°C formin, followed by 72°C for 5 min. The amplified PCR productsere digested with T7 exonuclease (0.45 U/�L) at room temper-

ture for 30 min. The 5= phosphothioates on one of the PCRrimers protected one strand of the PCR product from T7 exonu-lease digestion, resulting in the generation of a single-strandedCR product. The single-stranded PCR product was hybridized to384-well plate that contained covalently attached SNP-IT primer

xtension primer designed to hybridize immediately adjacent tohe SNP. After hybridization, the SNP-IT primer was extended for

single base with Klenow fragment of DNA polymerase I andixture of appropriate labeled terminating nucleotides, whichere labeled with either FITC or biotin and complementary to theNP. The identity of the incorporated nucleotide was determinedith serial colorimetric reactions with anti-FITC alkaline phospha-

ase and streptavidin horseradish peroxidase using p-nitrophenylhosphate and tetramethylbenzidine as a substrate, respectively.he results of yellow and/or blue color developments for eachample were analyzed with ELISA reader, and the final genotypealls were automatically assigned with QCReview program (Or-hid Biosciences). Automated genotype calls were corroborated byisual inspection of the data.

ollow-up and statistical analysisFollow-up examinations were carried out for all patients at least

very 3 months from the time of entering the study. The survivalime was determined from the first day of treatment, and therogression-free and overall survivals were calculated using theaplan–Meier method (25). The hazard ratio using Cox propor-

ional hazard regression was calculated after adjusting for age,ender, histology, stage, and alcohol consumption. These variablesere selected through the univariate analysis that had been per-

Table 4. Multivariate analysis of the progradiotherapy using Cox p

Predictors

Progressio

Hazard ratio

Exon 6, codon 194 1.61Exon 9, codon 280 0.56Exon 10, codon 399 1.07Age† 0.99Gender (male vs. female) 0.60Histology‡ 1.23Alcohol 0.62Stage§ 2.32

* The reference groups for hazard ratios owild-type allele (Arg/Arg for codon 194, 280,

† Age is a continuous variable.‡ Histology is divided into two groups, squ

carcinoma.§ Stage is divided into two groups, one includ

only Stage IIIb. The group from Stage I to IIIa was

ormed previously. The haplotype of the XRCC1 polymorphismsas estimated by PHASE version 2.1.

RESULTS

atient characteristicsTable 1 summarizes the patient characteristics. There

ere 191 males (83.4%) and 38 females (16.6%) with aedian age of 62 years (range, 26–88 years). The majority

f patients had good performance status, Eastern Coopera-ive Oncology Group 0–1. Of a total of 229 patients, 7% ofatients had Stage I, 29% had Stage II, 24% had Stage IIIa,nd 40% had Stage IIIb disease at the time of diagnosis orfter surgical resection. Fifty-one percent of patients receivedurative radiotherapy with/without chemotherapy, and the oth-rs received postoperative adjuvant radiotherapy after a com-lete resection of the tumor. Table 2 shows the distribution ofhe genotypes for each polymorphism. The prevalence ofhe homozygous variant genotypes for the polymorphismsas quite low (�10%).

RCC1 polymorphism and survivalThe median overall survival and progression-free sur-

ival time in all patients was 16 months and 13 months,espectively. Univariate analysis was performed on the pro-ression-free and overall survival data of all patients. Aignificant progression-free survival advantage was ob-erved in younger patients (�62 years), as well as a tumortage of I–IIIa, the wild-type XRCC1 Arg194Trp genotype,nd the variant-type XRCC1 Arg399Gln genotype (Table). However, no statistically significant prognostic fac-ors for overall survival, except tumor stage, were iden-ified in this univariate analysis. Multivariate analysishowed that an XRCC1 Arg194Trp polymorphism, theistologic subtype, and the stage were significant prog-ostic predictors for progression-free survival. The stage

n-free survival and overall survival afterional hazards regression

survival Overall survival

p value Hazard ratio p value

0.034 1.59 0.0700.160 0.67 0.3930.719 1.06 0.7930.810 1.03 0.1900.340 0.99 0.9900.035 0.89 0.3450.028 0.69 0.1500.003 1.92 0.048

enotypes were the groups with homozygous99).

cell carcinoma (reference group) and adeno-

m Stage I to Stage IIIa, and another including

ressioroport

n-free

*

f all gand 3

amous

ing fro

the reference group for hazard ratio.

whm3aWrtpTt(

gomrriaicanwa

neterwsXw

tppnvOiwysolptaf2

eocfaro

pcradr

X

889Polymorphisms in XRCC1 and survival of NSCLC ● S. M. YOON et al.

as significant only for overall survival (Table 4). The 6aplotypes from the XRCC1 polymorphisms were esti-ated by PHASE v.2.1, and the TGG (194Trp-280Arg-

99Arg) haplotype was found to be the most frequent typemong XRCC1 polymorphisms, as summarized in Table 5.hen the homozygous TGG haplotype was considered as a

eference, the other haplotype pairs had a significant pro-ective effect (Table 6). As shown in the survival curve,atients with haplotype pairs other than the homozygousGG haplotype pairs survived significantly longer than

hose with the homozygous TGG haplotype pairs (p � 0.04)Fig. 1).

DISCUSSION

The X-ray repair cross-complementing Group 1 (XRCC1)ene plays a role in DNA repair, particularly in the removalf nonbulky adducts produced by oxidative stress, DNAethylating agents, and ionizing radiation (11, 12). Ionizing

adiation causes DNA strand breaks. Therefore, the DNAepair mechanism may be considered a potential prognosticndicator for radiotherapy. Polymorphisms in XRCC1 canlter the efficiency of DNA repair and can affect lung cancern two contrasting ways. Decreased DNA repair might in-rease the risk of developing cancer, resulting in moreggressive tumors. On the other hand, it may simulta-eously improve the survival in patients already diagnosedith cancer (26). To date, many studies have examined the

ssociation between the XRCC1 polymorphisms and carci-

Table 6. Haplotype pairs of XRCC

Haplotype pairs*

Progressionhazar

(95

TGG† TGG 1.00 (reTGG Others§ 0.472 (0.Others Others 0.317 (0.

* Composed of three polymorphic sites: X(G/A).

† TGG: 194Trp-280Arg-399Arg.‡ Adjusted for age, gender, histology, stage

Table 5. The distribution of the haplotypes among theXRCC1 polymorphisms*

Haplotype† Frequency of haplotype

TGG 153CGG 136CGA 101CAG 39CAA 3TGA 2

* Estimated using PHASE v.2.1.† Composed of three polymorphic sites: XRCC1 194 (C/T),

RCC1 280 (G/A), XRCC1 399 (G/A).

§ Haplotype combinations other than TGG.

ogenesis in various types of cancer (4–9, 27, 28). How-ver, there are few reports concerning the relationship be-ween these polymorphisms and the treatment outcome,ven though some authors have reported that the DNAepair capacity affected the survival in NSCLC patients whoere treated with chemotherapy (29, 30). Therefore, this

tudy provides new knowledge about the polymorphisms ofRCC1 as a prognostic factor in NSCLC patients treatedith radiotherapy.This study showed that there was an association between

he XRCC1 Arg194Trp or Arg399Gln polymorphisms androgression-free survival. However, the two polymorphismsroduced opposite results. Patients with the wild-type ge-otypes (Arg/Arg) had a longer survival than those with theariant-type genotypes in codon 194 XRCC1 (p � 0.030).n the other hand, patients with the variant-type genotypes

n codon 399 polymorphism had a longer survival than thoseith the wild-type genotypes (p � 0.007). Multivariate anal-sis showed that an XRCC1 Arg194Trp polymorphism is aignificant predictor for progression-free survival. More-ver, patients with the TGG (194Trp-280Arg-399Arg) hap-otype pairs, the most predominant type among XRCC1olymorphism, had a lower overall survival compared withhose with the other haplotype pairs. From these results, wessume that the XRCC1 gene at codon 194 has some dif-erent functional effects than the XRCC1 gene at codons80 and 399.Theoretically, XRCC1 399Gln allele can favorably influ-

nce the response to radiotherapy via the inefficient removalf the DNA adducts and the prolonged delay of the G2 cellycle phase, which enhances radiation sensitivity (22). Theunctional effects of the XRCC1 Arg280His polymorphismfter radiotherapy in NSCLC are unclear. Therefore, furtheresearch will be needed to determine the functional effectsf XRCC1 Arg280His polymorphism.This study demonstrates an extension of the suggestion of

rimary prevention to tertiary prevention based on geneti-ally tailored treatment strategies. Therefore, identifyingadiosensitive lung cancer patients before treatment mightllow individual treatment to be tailored while normal tissueamage is minimized and the effectiveness of ionizingadiation is enhanced. This approach can potentially in-

) and an analysis of the survival

urvival‡

Overall survivalhazard ratio(95% CI)

1.00 (ref.).043) 0.255 (0.110–0.589).733) 0.199 (0.082–0.484)

194 (C/T), XRCC1 280 (G/A), XRCC1 399

lcohol consumption.

1(TGG

-free sd ratio% CI)

f.)213–1138–0

RCC1

, and a

cd

ta

890 I. J. Radiation Oncology ● Biology ● Physics Volume 63, Number 3, 2005

rease the response rates and survival outcomes while re-ucing the level of unnecessary treatment.

Fig. 1. Kaplan–Meier survival curve by the XRCC1 haplcurve.

In summary, polymorphisms of XRCC1 have an effect on a

REFEREN

DeVita VT, Hellmann S, Rosenberg SA, editors. Cancer prin-

he survival of NSCLC patients treated with radiotherapy,nd these effects seem more significant when the result is

: (a) progression-free survival curve, (b) overall survival

nalyzed with the haplotype pairs.

CES

1. Jemal A, Tiwari RC, Murray T, et al. Cancer statistics, 2004.CA Cancer J Clin 2004;54:8–29.

2. Pisani P, Parkin DM, Ferlay J. Estimates of the worldwidemortality from eighteen major cancer in 1985. Implications forprevention and projections of future burden. Int J Cancer1993;55:891–903.

3. Sekido Y, Fong KM, Minna JD. Cancer of the lung. In:

ciples and practice of oncology. 6th ed. Philadelphia: Lippin-cott-Raven; 2001. p. 917–983.

4. Harms C, Salama SA, Sierra-Torres CH, et al. Polymorphismsin DNA repair genes, chromosome aberrations, and lung can-cer. Environ Mol Mutagen 2004;44:74–82.

5. Yu HP, Zhang XY, Wang XL, et al. DNA repair gene XRCC1polymorphisms, smoking, and esophageal cancer risk. Cancer

otypes

Detect Prev 2004;28:194–199.

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

891Polymorphisms in XRCC1 and survival of NSCLC ● S. M. YOON et al.

6. Sanyal S, Festa F, Sakano S, et al. Polymorphisms in DNArepair and metabolic genes in bladder cancer. Carcinogenesis2004;25:729–734.

7. Ratnasinghe LD, Abnet C, Qiao YL, et al. Polymorphisms ofXRCC1 and risk of esophageal and gastric cardia cancer.Cancer Lett 2004;216:157–164.

8. Tae K, Lee HS, Park BJ, et al. Association of DNA repairgene XRCC1 polymorphisms with head and neck cancer inKorean population. Int J Cancer 2004;111:805–808.

9. Park JY, Lee SY, Jeon HS, et al. Polymorphism of the DNArepair gene XRCC1 and risk of primary lung cancer. CancerEpidemiol Biomarkers Prev 2002;11:23–27.

0. Price EA, Bourne SL, Radbourne R, et al. Rare microsatellitepolymorphisms in the DNA repair genes XRCC1, XRCC3,and XRCC5 associated with cancer in patients of varyingradiosensitivity. Somat Cell Mol Genet 1997;23:237–247.

1. Moullan N, Cox DG, Angele S, et al. Polymorphisms in theDNA repair gene XRCC1, breast cancer risk, and response toradiotherapy. Cancer Epidemiol Biomarkers Prev 2003;12:1168–1174.

2. Yu Z, Chen J, Ford BN, et al. Human DNA repair systems: Anoverview. Environ Mol Mutagen 1999;33:3–20.

3. Thompson LH, Brookman KW, Jones NJ, et al. Molecularcloning of the human XRCC1 gene, which corrects defectiveDNA strand break repair and sister chromatid exchange. MolCell Biol 1990;10:6160–6171.

4. Thompson LH, West MG. XRCC1 keeps DNA from gettingstranded. Mutat Res 2000;459:1–18.

5. Dianov GL, Prasad R, Wilson SH, et al. Role of DNA poly-merase beta in the excision step of long patch mammalian baseexcision repair. J Biol Chem 1999;274:13741–13743.

6. Caldecott KW, McKeown CK, Tucker JD, et al. An interac-tion between the mammalian DNA repair protein XRCC1 andDNA ligase III. Mol Cell Biol 1994;14:68–76.

7. Vidal AE, Boiteux S, Hickson ID, et al. XRCC1 coordinatesthe initial and late stages of DNA abasic site repair throughprotein-protein interactions. EMBO J 2001;20:6530–6539.

8. Shen MR, Jones IM, Mohrenweiser H. Nonconservative

amino acid substitution variants exist at polymorphic fre-

quency in DNA repair genes in healthy humans. Cancer Res1998;58:604–608.

9. Savas S, Kim DY, Ahmad MF, et al. Identifying functionalgenetic variants in DNA repair pathway using protein conser-vation analysis. Cancer Epidemiol Biomarkers Prev 2004;13:801–807.

0. Lamerdin JE, Montgomery MA, Stilwagen SA, et al. Genomicsequence comparison of the human and mouse XRCC1 DNArepair gene regions. Genomics 1995;25:547–554.

1. Lunn RM, Langlois RG, Hsieh LL, et al. XRCC1 polymor-phisms: Effects on aflatoxin B1-DNA adducts and glycoph-orin A variant frequency. Cancer Res 1999;59:2557–2561.

2. Hu JJ, Smith TR, Miller MS, et al. Amino acid substitutionvariants of APE1 and XRCC1 genes associated with ionizingradiation sensitivity. Carcinogenesis 2001;22:917–922.

3. Greene FL, Page DL, Fleming ID, et al. AJCC cancer stagingmanual. 6th ed. New York: Springer-Verlag Press; 2002. p.167–177.

4. Han W, Kang D, Park IA, et al. Associations between breastcancer susceptibility gene polymorphisms and clinicopatho-logical features. Clin Cancer Res 2004;10:124–130.

5. Kaplan EL, Meier P. Nonparametric estimation from incom-plete observations. J Am Stat Assoc 1958;53:457–481.

6. Wei Q, Frazier M, Levin B. DNA repair: A double-edgedsword. J Natl Cancer Inst 2000;92:440–441.

7. Abdel-Rahman SZ, Soliman AS, Bondy ML, et al. Inheritanceof the 194Trp and the 399Gln variant alleles of the DNArepair gene XRCC1 are associated with increased risk ofearly-onset colorectal carcinoma in Egypt. Cancer Lett 2000;159:79–86.

8. Duell EJ, Millikan RC, Pittman GS, et al. Polymorphisms inthe DNA repair gene XRCC1 and breast cancer. CancerEpidemiol Biomarkers Prev 2001;10:217–222.

9. Rosell R, Lord RV, Taron M, et al. DNA repair and cisplatinresistance in non-small-cell lung cancer. Lung Cancer 2002;38:217–227.

0. Gurubhagavatula S, Liu G, Park S, et al. XPD and XRCC1genetic polymorphisms are prognostic factors in advancednon-small-cell lung cancer patients treated with platinum che-

motherapy. J Clin Oncol 2004;22:2594–2601.