OL Oncology Letters 1792-1074 1792-1082 D.A. Spandidos 10.3892/ol.2011.463 ol-03-02-0351 Articles Single nucleotide polymorphisms in DNA repair genes and risk of cervical cancer: A case-control study ZHANGLIHUA1* RUANZHENCHAO2* HONGQINGYA3 GONGXIANGZHEN4 HUZHENGGUANG4 HUANGYAN2 XUAIDI5 Department of Radiation Oncology, Cancer Hospital, Fudan University, Shanghai 200032 State Key Laboratory of Genetic Engineering, Institute of Genetics, School of Life Sciences, Fudan University, Shanghai 200433 Hospital of Obstetrics and Gynecology, Attached Fudan University, Shanghai 200011 Hongkou Maternal and Child Health Care Institute, Shanghai 200437 Shanghai Hongkou District Health Bureau, Shanghai 200437, P.R. China Correspondence to: Dr Yan Huang, Institute of Genetics, School of Life Sciences, Fudan University, Room 604 Science Building, 220 Handan Road, Shanghai 200433, P.R. China, E-mail: huangyan@fudan.edu.cn. Professor Aidi Xu, Shanghai Hongkou District Health Bureau, 76 Feihong Rd., Shanghai 200086, P.R. China, E-mail: helenxu602003@yahoo.com.cn

Contributed equally

 2 2012 26 10 2011 3 2 351 362 27 5 2011 27 9 2011 Copyright © 2012, Spandidos Publications 2012

In this report, we describe a case control study in a Chinese population aimed at identifying possible associations between susceptibility to cervical cancer and single nucleotide polymorphisms in XRCC1 194C>T, XRCC1 280G>A, XRCC1 399G>A, ERCC2 751A>C, ERCC2 156C>A, ERCC1 118C>T, PARP1 762T>C, RAD51 135G>C and HER2 655A>G. The cases comprised 154 patients: 80 cervical squamous cell carcinomas (SCCs), 2 adenocarcinomas and 72 cervical intraepithelial neoplasias (CINs). A total of 177 healthy women were recruited as the controls. A significant association was found between ERCC1 118C>T and SCC in the additive genetic model [odds ratio (OR)=1.711; 95% confidence interval (CI), 1.089–2.880; p=0.021] and the dominant genetic model (OR=1.947; 95% CI, 1.056–3.590; p=0.033). Among women with a smoking family member, ERCC1 118C>T increased SCC risk in the additive model (OR=2.800; 95% CI, 1.314–5.968; p=0.008). For women who had first intercourse before 22 years of age, XRCC1 280G>A was found to act as a protective factor for SCC under the additive model (OR=0.228; 95% CI, 0.058–0.900; p=0.035), while RAD51 135G>C was a risk factor for CIN (OR=4.246; 95% CI, 1.335–13.502; p=0.014). For women who had first intercourse after 22 years of age, the additive genetic model showed RAD51 135G>C (OR=0.359; 95% CI, 0.138–0.934; p=0.036) and HER2 655A>G (OR=0.309; 95% CI, 0.098–0.972; p=0.045) to be protective factors for SCC. XRCC1 399G>A increased CIN risk among women who first gave birth before the age of 22 in the additive genetic model (OR=4.459; 95% CI, 1.139–17.453; p=0.032). For those who first gave birth after age 22, ERCC1 118C>T was found to be a risk factor for SCC in the additive genetic model (OR=1.884; 95% CI, 1.088–3.264; p=0.024). A significant interaction was observed between RAD51 135G>C and age at first intercourse (pinteraction=0.033 for SCC, pinteraction=0.021 for CIN), as well with sexual partner number (pinteraction=0.001 for SCC). The interaction between HER2 655A>G and age at first intercourse, ERCC2 156C>A and family smoking status and XRCC1 280G>A and alcohol consumption were significant, with pinteraction=0.023 for SCC, pinteraction=0.021 for CIN and pinteraction=0.025 for SCC, respectively.

 cervical cancer cervical intraepithelial neoplasia single nucleotide polymorphism association study squamous cell carcinoma Introduction

Cervical cancer ranks as the second most common female-related cancer worldwide, following breast cancer. Squamous cell carcinoma (SCC), adenocarcinoma (ADC), and adenosquamous cell carcinoma (ADSC) are the three most common histological subtypes of cervical cancer. Screening programs in developed countries have reduced the incidence and prevalence of cervical cancer in these areas, but developing countries often do not have the resources for these programs (1). The result is that 80% of cervical cancer cases now occur in developing countries (2).

In China, incidence rates of cervical cancer range from 2.4 to 4.6 per 100,000 women (3), and the mortality rates range from 2 to 4 per 100,000 women in urban areas. Various evidence has shown that certain types of the oncogenic virus human papillomavirus (HPV) are closely related to the occurrence of cervical cancer (4) and its precursor lesion cervical intraepithelial neoplasia (CIN) (5). However, only a small portion of women go on to develop cervical cancer following infection with HPV (6), and this suggests that other factors, including genetic susceptibility, may also contribute to cervical cancer.

An association between smoking and cervical cancer has been reported (7), as well as smoking-related DNA damage in the cervical epithelium (8). Defects in DNA repair pathways relate to a number of diseases including cancer, and the significance of DNA repair mechanisms in genetic stability maintenance is well accepted in the protection against cancer initiation (9).

There are a number of different types of DNA repair pathways; for single-strand breaks, these include the base excision repair (BER) and nucleotide excision repair (NER) systems and for double-strand DNA breaks, there are two principle mechanisms: homologous recombination (HR) and non-homologous end joining (NHEJ) (10). There are a number of genes known to be involved in these pathways. For example, it is known that the genes, ERCC1 and ERCC2, code for proteins involved in the NER system, removing bulky lesions from DNA caused by things such as toxic chemicals or ultraviolet light (11,12). Two other genes, XRCC1 and PARP1, are crucial to BER systems, which repair DNA damage due to causes such as ionizing radiation. PARP1 is also known to signal damage to other repair mechanisms (12,13). RAD51 functions in the DNA repair of double-strand breaks by HR mechanisms (10). It is also notable that these genes have been implicated in response (or non-response) to certain types of chemotherapeutic drugs (14).

A single nucleotide polymorphism (SNP) is a single nucleotide change in a DNA sequence between two individuals. Knowledge of the genes involved in these DNA repair mechanisms is enlightening investigators and enabling the study of the association between SNPs in these genes and the likelihood of developing cancer (15).

The associations of SNPs in DNA repair genes and various types of cancer and tumors have been extensively described. However, the evidence is frequently confusing, with some SNPs increasing the risk of certain types of cancer, but decreasing the risk of others. For example, previous studies have reported that XRCC1 Arg194Trp C>T (TT) increases the risk of esophageal (16) and bile duct cancer (17), but decreases the risk of gastric carcinoma (18). XRCC1 Arg280His G>A has been reported as a risk factor for breast cancer (19) and as a protective factor for bile duct cancer (17). Niwa et al (20) first reported that the XRCC1 Arg399Gln G>A polymorphism is related to the increased susceptibility to cervical cancer in a Japanese population.

We performed a case-control study, in a Chinese population, of eight SNPs from the DNA repair-related genes, ERCC1, ERCC2, XRCC1, PRAP1 and RAD51, as well as one SNP in HER2, which plays essential roles in stabilizing the active protein dimer.

 Materials and methods Study subjects

All subjects were genetically unrelated ethnic Han Chinese from Shanghai in eastern China. Patients diagnosed with histopathologically confirmed cervical cancer or CIN were consecutively recruited between June 2006 and May 2008 at the Margaret Willianson Hospital, Hongkou Maternal and Child Care Service Centre, and the Gynecology Department of Jiangwan Hospital, without restrictions on age and histology. The exclusion criteria included self-reported cancer history, previous radiotherapy and chemotherapy for unknown disease conditions and a family history of cancer. The controls were recruited during the same period from health examinees at the Hongkou Maternal and Child Care Service Centre. The controls were healthy women whose sexual history and age matched the cases, and who were histologically or cytologically diagnosed as having a normal cervix or chronic cervicitis, with no personal or familial history of cancer or other family genetic diseases.

 Questionnaire

All patients underwent a complete medical history interview at entry. Detailed information regarding demographic factors (including age at diagnosis, education level, number of sexual partners, age at first intercourse and age at first childbirth), occupational history, environmental exposure data (including tobacco smoking status, alcohol consumption and family smoking status) and family history of cancer were also included. Those who had smoked less than one cigarette per day for a period of one year or less were defined as non-smokers, and the rest were considered to have always smoked. Alcohol consumption was defined as ‘no’ if the participant drank alcohol less than once per week, but otherwise was regarded as ‘yes’. Family smoking status was ‘no’ for participants who had no family member who smoked in the home, but was otherwise regarded as ‘yes’. Family history of cancer was defined as any self-reported cancer in first-degree relatives. Following the interview, approximately 2 ml of venous blood was collected from each participant and maintained below −40˚C.

 Laboratory methods

Genetic DNA was extracted from peripheral blood by the standard phenol-chloroform procedure. SNP genotyping was conducted with the SNPware 12plex assay using the SNPstream™ system (Beckman Coulter, Inc., Brea, CA, USA). The amplifying and extension primers are shown in Table I. The nine SNPs were: XRCC1 194C>T (rs1799782), XRCC1 280G>A (rs25489), XRCC1 399G>A (rs25487), ERCC2 751A>C (rs13181), ERCC2 156C>A (rs238406), ERCC1 118C>T (rs11615), PARP1 762T>C (rs1136410), RAD51 135G>C (rs1801320) and HER2 655A>G (rs1801200).

Genotyping was performed according to the manufacturer's instructions. Briefly, multiple polymerase chain reaction (PCR) was performed with the following program: 94˚C for 1 min, then 94˚C for 30 sec, 55˚C for 30 sec and 72˚C for 1 min, for 39 cycles. The PCR product was used for SNP extension: 96˚C for 3 min, then 94˚C for 20 sec and 40˚C for 11 sec, for 45 cycles. A chip hybridization reaction was conducted at 42˚C for 2 h with hybridization solution and hybridization additive solution at the proportion of 18:1. Chip data were scanned and analyzed using the SNPstream machine (Beckman Coulter, Inc.).

 Statistical analysis

To examine the Hardy-Weinberg equilibrium, a χ2 goodness-of-fit test was performed using a web-based program (http://ihg2.helmholtz-muenchen.de/cgi-bin/hw/hwa1.pl). The frequency distribution of education, smoking status, alcohol consumption, family smoking status, number of sexual partners, age at first intercourse and age at first childbirth was compared between the controls and CIN/SCC cases using Fisher's exact χ2-test. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated by univariate logistic regression analyses and multivariate logistic regression analyses adjusted for age, family smoking status, age at first intercourse and age at first childbirth under the dominant genetic, recessive genetic and additive model. The subjects were stratified according to family smoking status, age at first intercourse, age at first childbirth and disease subtypes, in order to estimate specific ORs and CIs.

Interaction between the nine SNPs and demographic factors (family smoking status, age at first intercourse, age at first childbirth, sexual partner number and alcohol consumption) were estimated. All the regression analyses were performed using SPSS 16.0 software (SPSS Inc., Chicago, IL, USA). Quanto (version 1.2.4) software was used to calculate the statistical power. All p-values were two-sided, and p<0.05 was considered to be statistically significant.

 Results Characteristics of the study population

A total of 154 patients (82 carcinomas and 72 CINs) and 177 cancer-free controls were enrolled in the study. Among all carcinomas, two patients were diagnosed with ADC and the rest were diagnosed with SCC. Thus, our study focused on SCC and CIN. The characteristics of the study participants are shown in Table II. There was no significant difference in age between controls and patients with CIN/SCC. Similarly, education level, smoking status, alcohol consumption and the number of sexual partners were not significantly associated with CIN or SCC risk. However, significant differences between the cases and controls were found in family smoking status, age at first intercourse and age at first childbirth. These factors have previously been reported as risk factors for HPV infection, or co-factors for cervical cancer (3).

 Association between SNPs and risk of cervical SCC and CIN

The genotype distributions of controls were in Hardy-Weinberg equilibrium (data not shown). No SNP was found to be associated with the risk of SCC or CIN in all the subjects in the univariate logistic regression analyses (data not shown). Table III shows the adjusted association of the nine SNPs in the two case groups. In the dominant genetic model, a significantly increased risk of SCC was observed in ERCC1 118C>T (OR=1.947; 95% CI, 1.056–3.590; p=0.033) when adjusted for age, family smoking status, age at first intercourse and age at first childbirth. The effect was more clear when the additive model was applied (OR=1.771; 95% CI, 1.089–2.880; p=0.021).

The statistical powers of the dominant, recessive and additive model were evaluated for each SNP, and the additive model was found to have the highest statistical power in the study (data not shown). Thus, an additive model was used for all the following association analyses.

 Association analysis following stratification

Significant distribution differences between the controls and cases were found in family smoking status, age at first intercourse and age at first childbirth (Table II). Thus, stratification was performed by these three factors. Table IV shows the association using multivariate logistic regression analysis, stratified by family smoking status. Using the additive model where the family member smoking status was ‘yes’, the SNP ERCC1 118C>T had an increased risk of SCC (OR=2.800; 95% CI, 1.314–5.968; p=0.008). Meanwhile, ERCC2 156C>A tended to act as a risk factor for CIN under the additive model (OR=1.949; 95% CI, 0.951–3.944; p=0.068) if the family member smoking status was ‘yes’. However, ERCC2 156C>A tended to be a protective factor for CIN under the additive model (OR=0.616; 95% CI, 0.347–1.095; p=0.099) if the family member smoking status was ‘no’. The different effects of the ERCC2 156C>A polymorphism between CIN and SCC suggest that its function may depend on exposure to tobacco smoke.

Among women having their first intercourse after the age of 22, there was no significant association for CIN using multivariate logistic regression analysis (Table V). However, the additive model showed that RAD51 135G>C (OR=0.359; 95% CI, 0.138–0.934; p=0.036) and HER2 655A>G (OR=0.309; 95% CI, 0.098–0.972; p=0.045) may be protective factors for SCC. In the subgroup who had their first intercourse before age 22, XRCC1 280G>A showed a protective effect for SCC (OR=0.228; 95% CI, 0.058–0.900; p=0.035) under the additive model. Meanwhile, RAD51 135G>C was related to increased susceptibility to CIN under the additive model (OR=4.246; 95% CI, 1.335–13.502; p=0.014).

Using the additive model, XRCC1 399G>A increased susceptibility to CIN in the subgroup who first gave birth before age 22 (OR=4.459; 95% CI, 1.139–17.453; p=0.032), and ERCC1 118C>T was found to be a risk factor for SCC (OR=1.884; 95% CI, 1.088–3.264; p=0.024) among those who first gave birth after age 22.

 Interaction between SNPs and demographic factors

RAD51 135G>C decreased the CIN risk in combination with first intercourse at a later age (OR=0.217; 95% CI, 0.059–0.798; p=0.021), and increased CIN risk in combination with a greater number of sexual partners (OR=12.260; 95% CI, 2.874–52.305; p=0.001). ERCC2 156C>A increased the CIN risk in combination with family smoking status (OR=2.639; 95% CI, 1.157–6.020; p=0.021).

HER2 655A>G and RAD51 135G>C decreased the SCC risk when combined with first intercourse at a later age (OR=0.179; 95% CI, 0.041–0.790; p=0.023; and OR=0.250; 95% CI, 0.070–0.893; p=0.033; respectively). When combined with alcohol consumption, XRCC1 280G>A increased the SCC risk (OR=7.117; 95% CI, 1.274–39.754; p=0.025).

 Discussion

DNA repair genes have been frequently studied as significant determinants of cancer risk. Associations between SNPs in these genes and various types of cancer have been well documented but the picture remains complex, with often inconsistent data pointing to both increased and decreased risks (16–19). An association between the SNP XRCC1 Arg399Gln G>A and an increased susceptibility to cervical cancer has been shown in the Japanese population (20), in contrast to a decreased risk, but an increased persistence of HPV infection, in a population in Costa Rica (21). These data have been partially confirmed in this study where we observed an association between XRCC1 Arg399Gln G>A and an increased CIN risk among women who first gave birth before the age of 22. However, other changes in XRCC1, such as XRCC1 Arg194Trp C>T showed no significant association with CIN/SCC, and among women who had their first intercourse before 22 years of age, there was a significant association between XRCC1 Arg280His G>A and a decreased susceptibility to SCC. Although no previous study has reported an association between alcohol consumption and cervical cancer, we observed an interaction between XRCC1 280G>A and alcohol consumption (pinteraction<0.05), which may suggest that alcohol is involved in the development of cervical cancer.

The enzyme PARP1 (encoded by the gene of the same name) plays a role in repairing single-stranded DNA breaks. The PARP1 Val762Ala T>C polymorphism located in the catalytic domain has been shown to interact with XRCC1. In vitro, this polymorphism markedly reduces the enzymatic activity of PARP1, and has also been linked to cancer susceptibility (22). An increased risk of smoking-related lung cancer has also been reported (23). There are no previous reports evaluating the correlation of PARP1 Val762Ala T>C with cervical cancer as yet. Our analysis showed no significant association between the polymorphism and CIN/SCC risk.

The proteins, ERCC1 and ERCC2. are two of 16 proteins involved in NER systems, where they help to excise lesions from the DNA strand (11,24). The proteins are encoded by the genes ERCC1 and ERCC2, respectively. In ERCC1, a polymorphism at ERCC1 118C>T has been shown to decrease the rate of mRNA translation to the protein (25). It is possible that a decreased level of ERCC1 mRNA expression influences the repair efficiency of DNA damage, which then contributes to cancer susceptibility. Studies have also shown an association between the polymorphism and malignancies such as lung, ovarian and colorectal cancer (26–29), although no previous association with cervical cancer has been shown. Our study showed that ERCC1 118C>T may be a potential risk factor for SCC. Among the Chinese population, a homozygous CC polymorphism in ERCC2 at position 751A>C is related to a decreased DNA repair capacity in patients with lung cancer (30). In addition, there are data linking an ERCC2 polymorphism and glioblastoma (31). Our data indicate that ERCC2 751A>C and ERCC2 156C>A have no significant association in CIN/SCC. Furthermore, among women who first gave birth after 22 years of age, ERCC1 118C>T may be a risk factor for SCC.

The association between cervical cancer and smoking (and second-hand smoke) has already been established (7,32). The hypothesis is that the carcinogens in smoke initiate DNA damage, and then activate the DNA repair process; with polymorphisms in the DNA repair genes either increasing or decreasing the risk of cancer in the areas affected. In the case of cervical cancer, it is thought that smoke may have a directly carcinogenic effect by acting through the polycyclic aromatic hydrocarbon-DNA adduct in the cervix (33). When we looked at associations between family smoking status and SNPs, we found that ERCC1 118C>T, a potential risk factor for SCC, became more significantly associated with this outcome in women who live with family members who smoke. Although not statistically significant, ERCC2 156C>A tended to be a risk factor among patients who have smoking family members, but tended to play a protective role in women whose families did not smoke. Tsai et al (32) previously showed that non-smoking women exposed to second-hand cigarette smoke had a significantly greater risk of developing CIN than unexposed non-smokers. Our data indicate that ERCC2 156C>A may play a role in the association between second-hand smoke and CIN.

The gene RAD51 plays a significant role in DNA repair of double-strand breaks via HR (34). Previous studies, although not looking at functional evidence, have shown that the polymorphism RAD51 135G>C is related to breast cancer susceptibility (35). No RAD51 polymorphisms related to cervical cancer have been reported as yet. In this study, we observed that RAD51 135G>C may increase the risk of CIN in women who first had intercourse before the age of 22, but may be a protective factor for SCC in women who first had intercourse after the age of 22. Studies examining SNPs in HER2 have mostly focused on HER2 655A>G in breast cancer (36), and no previous association has been shown in cervical cancer. In our analysis, HER2 655A>G showed a decreased susceptibility to SCC in women who had first intercourse after the age of 22.

The interaction analysis in our study shows a consistent result with epidemiological data of cervical cancer. Women who have multiple sexual partners or a younger age at first sexual intercourse had a significantly higher risk of HPV infection (3) as well as an increased risk of cervical cancer (37). While our study indicates that polymorphisms in HER2 655A>G and RAD51 135G>C and the age of first intercourse are related, it is worth noting that the age we used as a cut-off (22 years) is an arbitrary age that is open to some debate. There is evidence that in certain countries as many as 50% of women have their first intercourse between the ages of 13 and 19 years (37). Altering the age cut-off in our interaction analyses may affect the outcome and correlations between cervical cancer and the selected SNPs, and these analyses will need to be performed before any more firm conclusions can be drawn.

It is also worth noting that the sample size of our study was relatively small, which limits the statistical power, and, hence, the conclusions that may be drawn. Other studies examining associations in other types of cancer, such as lung or breast cancer have recruited larger numbers of patients, and this larger number of data points has enhanced the robustness of the findings. Cervical cancer patients however, are fewer in number compared with other types of cancer, and therefore we were limited as to the number of patients we could recruit to our study. Our study of 154 patients (82 carcinomas and 72 CINs) compares favorably with other recently published case-control studies in the same geographical area (38–40), but clearly there is a need for further research with larger sample sizes across diverse populations.

In addition, the data presented in this study were gained using an additive genetic model. Although the additive and dominant models are often highly correlated, there is debate as regards which is the most appropriate, and there are also cases where these tests are not appropriate, particularly if testing traits that turn out to be recessive (41). Further exploration of different models is required, particularly if larger sample sizes are used.

In conclusion, this is the first association study between cervical cancer and SNPs in ERCC1, ERCC2, RAD51 and HER2. Interaction analysis suggests that sexual behavior, second-hand smoke and alcohol consumption are co-factors combined with SNPs, and further research is required to confirm these findings.

 Acknowledgements

This study was supported by the Shanghai Municipal Health Bureau Program 2006108 and the Shanghai Hongkou District Health Bureau Program 0601-03.

 References KitchenerHCCastlePECoxJTAchievements and limitations of cervical cytology screeningVaccine2463702006 ValdespinoVMValdespinoVECervical cancer screening: State of the artCurr Opin Obstet Gynecol1835402006 ShiJFQiaoYLSmithJSEpidemiology and Prevention of Human Papillomavirus and Cervical Cancer in China and MongoliaVaccine26M53M592008 WalboomersJMMJacobsMVManosMMHuman papillomavirus is a necessary cause of invasive cervical cancer worldwideJ Pathol18912191999 HausenHZViruses in Human CancersScience254116711731991 GiulianoARHarrisRSedjoRLIncidence, prevalence, and clearance of type-specific human papillomavirus infections: The young women's health studyJ Infect Dis1864624692002 WinkelsteinWSmoking and cervical cancer - current status: a reviewAm J Epidemiol1319459571990 SimonsAMPhillipsDHColemanDVDamage to DNA in cervical epithelium related to smoking tobaccoBMJ306144414481993 MaserRSDePinhoRAConnecting chromosomes, crisis, and cancerScience2975655692002 JacksonSPBartekJThe DNA-damage response in human biology and diseaseNature461107110782009 MuDParkCHMatsunagaTHsuDSReardonJTSancarAReconstitution of human DNA-repair excision nuclease in a highly defined systemJ Biol Chem270241524181995 SmithTRLevineEAFreimanisRIPolygenic model of DNA repair genetic polymorphisms in human breast cancer riskCarcinogenesis29213221382008 SchreiberVDantzerFAmeJCde MurciaGPoly(ADP-ribose): novel functions for an old moleculeNat Rev Mol Cell Biol75175282006 LiSXSjolundAHarrisLSweasyJBDNA repair and personalized breast cancer therapyEnviron Mol Mutagen518979082010 ManuguerraMSalettaFKaragasMRXRCC3 and XPD/ERCC2 single nucleotide polymorphisms and the risk of cancer: A HuGE reviewAm J Epidemiol1642973022006 XingDYQiJMiaoXPLuWFTanWLinDXPolymorphisms of DNA repair genes XRCC1 and XPD and their associations with risk of esophageal squamous cell carcinoma in a Chinese populationInt J Cancer1006006052002 HuangWYGaoYTRashidASelected base excision repair gene polymorphisms and susceptibility to biliary tract cancer and biliary stones: a population-based case-control study in ChinaCarcinogenesis291001052008 ShenHBXuYCQianYPolymorphisms of the DNA repair gene XRCC1 and risk of gastric cancer in a Chinese populationInt J Cancer886016062000 LoizidouMAMichaelTNeuhausenSLGenetic polymorphisms in the DNA repair genes XRCC1, XRCC2 and XRCC3 and risk of breast cancer in CyprusBreast Cancer Res Treat1125755792008 NiwaYMatsuoKItoHAssociation of XRCC1 Arg399Gln and OGG1 Ser326Cys polymorphisms with the risk of cervical cancer in Japanese subjectsGynecol Oncol9943492005 WangSSBrattiMCRodriguezACCommon variants in immune and DNA repair genes and risk for human papillomavirus persistence and progression to cervical cancerJ Infect Dis19920302009 WangXGWangZQTongWMShenYPARP1 Val762Ala polymorphism reduces enzymatic activityBiochem Biophys Res Commun3541221262007 ZhangXMMiaoXPLiangGPolymorphisms in DNA base excision repair genes ADPRT and XRCC1 and risk of lung cancerCancer Res657227262005 MuDHsuDSSancarAReaction mechanism of human DNA repair excision nucleaseJ Biol Chem271828582941996 YuJJMuCJLeeKBA nucleotide polymorphism in ERCC1 in human ovarian cancer cell lines and tumor tissuesMutat Res38213201997 KangSJuWKimJWAssociation between excision repair cross-complementation group 1 polymorphism and clinical outcome of platinum-based chemotherapy in patients with epithelial ovarian cancerExp Mol Med383203242006 ParkDJZhangWStoehlmacherJTsao-WeiDGroshenSGilJYunJSonesEMallikNLenzHJERCC1 gene polymorphism as a predictor for clinical outcome in advanced colorectal cancer patients treated with platinum-based chemotherapyClin Adv Hematol Oncol11621662003 SukRGurubhagavatulaSParkSPolymorphisms in ERCC1 and grade 3 or 4 toxicity in non-small cell lung cancer patientsClin Cancer Res11153415382005 ZhouWLiuGParkSGene-smoking interaction associations for the ERCC1 polymorphisms in the risk of lung cancerCancer Epidemiol Biomarkers Prev144914962005 YinJYVogelUMaYGGuoLWangHWQiRPolymorphism of the DNA repair gene ERCC2 Lys751Gln and risk of lung cancer in a northeastern Chinese populationCancer Genet Cytogenet16927322006 McKean-CowdinRBarnholtz-SloanJInskipPDAssociations between polymorphisms in DNA repair genes and glioblastomaCancer Epidemiol Biomarkers Prev18111811262009 TsaiHTTsaiYMYangSFLifetime cigarette smoke and second-hand smoke and cervical intraepithelial neoplasm–a community-based case-control studyGynecol Oncol1051811882007 HellbergDValentinJNilssonSSmoking as risk factor in cervical neoplasiaLancet2149714971983 GalkinVEEsashiFYuXYangSXWestSCEgelmanEHBRCA2 BRC motifs bind RAD51-DNA filamentsProc Natl Acad Sci USA102853785422005 JaraLAcevedoMLBlancoRRAD51 135G>C polymorphism and risk of familial breast cancer in a South American populationCancer Genet Cytogenet17865692007 XieDWShuXODengZLPopulation-based, case-control study of HER2 genetic polymorphism and breast cancer riskJ Natl Cancer Inst924124172000 LouieKSde SanjoseSDiazMEarly age at first sexual intercourse and early pregnancy are risk factors for cervical cancer in developing countriesBr J Cancer100119111972009 HanSSChoEYLeeTSInterleukin-12 p40 gene (IL12B) polymorphisms and the risk of cervical cancer in Korean womenEur J Obstet Gynecol Reprod Biol14071752008 KohaarIThakurNSalhanSTNF alpha-308G/A polymorphism as a risk factor for HPV associated cervical cancer in Indian populationCell Oncol292492562007 TongSYLeeJMSongESFunctional polymorphism in manganese superoxide dismutase and antioxidant status: Their interactions on the risk of cervical intraepithelial neoplasia and cervical cancerGynecol Oncol1152722762009 LettreGLangeCHirschhornJNGenetic model testing and statistical power in population-based association studies of quantitative traitsGenet Epidemiol313583622007

Primers for single nucleotide polymorphism detection.

rs1136410
 5′ primer TGAGCAGACTGTAGGCCAC
 3′ primer TCTGTCTCATTCACYATGATACCTA
 Ext primer CGACTGTAGGTGCGTAACTCGTCCAGCAGGTTGTCAAGCATTTCC
rs1801200
 5′ primer AAACTAGCCCTCAATCCCTG
 3′ primer AAAGACCACCCCCAAGAC
 Ext primer AGAGCGAGTGACGCATACTACGCCCCCAGCCCTCTGACGTCCRTC
rs1799782
 5′ primer TYAGGACCCAYGTTGTCC
 3′ primer ATGAGAGCGCCAACTCTCT
 Ext primer AGATAGAGTCGATGCCAGCTTCACCTGGRGATGTCTTGTTGATCC
rs11615
 5′ primer TTCGTCCCTCCCCAGAGG
 3′ primer ATGAGAGCGCCAACTCTCT
 Ext.primer GTGATTCTGTACGTGTCGCCTTACGTCGCCAAATTCCCAGGGCAC
rs13181
 5′ primer CTGTCCCTGCTCAGCCTG
 3′ primer AAGACTCAGGAGTCACCAGGA
 Ext primer AGGGTCTCTACGCTGACGATGCTAGAATCAGAGGAGACGCTG
rs25489
 5′ primer TTGACCCCCAGTGGTGCTA
 3′ primer TGTCACTGCCCCCTGTGC
 Ext primer GGCTATGATTCGCAATGCTTTCTTCTCCAGTGCCAGCTCCAACTC
rs1801320
 5′ primer AACTGCAACTCATCTGGGTT
 3′ primer TCCTCTCTCCAGCAGGCC
 Ext primer GCGGTAGGTTCCCGACATATGAGTAGAGAAGTGGAGCGTAAGCCA
rs25487
 5′ primer TAAGGAGTGGGTGCTGGA
 3′ primer ATTGCCCAGCACAGGATA
 Ext primer AGCGATCTGCGAGACCGTATCGCATGCGTCGGCGGCTGCCCTCCC
rs238406
 5′ primer TATGTGCGGGCGCAGTAC
 3′ primer TCCTGCCTCCCTCCCTCA
 Ext primer CGTGCCGCTCGTGATAGAATATGACACCAGCCTGCCCCACTGCCG

Ext, extension.

Distributions of demographic characters in the study population.

Controls CIN SCC
Variable (n=177) no. (%) (n=72) no. (%) p-valueb (n=80) no. (%) p-valueb
Age 43 (24–55)a 40 (24–59)a 44 (26–79)a
Education level
 ≤ Junior high school 64 (36.2) 28 (38.9) 0.772 29 (36.2) 1.000
 ≥ Senior high school 113 (63.8) 44 (61.1) 51 (63.8)
Smoking status
 No 167 (94.4) 72 (100) 0.067 76 (95.0) 1.000
 Yes 10 (5.6) 0 4 (5.0)
Alcohol consumption
 No 148 (83.6) 63 (87.5) 0.561 73 (91.2) 0.122
 Yes 29 (16.4) 9 (12.5) 7 (8.8)
Family smoking status
 No 61 (34.5) 47 (65.3) 0 46 (57.5) 0.001
 Yes 116 (65.5) 25 (34.7) 34 (42.5)
Number of sexual partners
 ≤1 144 (81.4) 50 (73.5) 0.218 62 (81.6) 1.000
 >1 33 (18.6) 18 (26.5) 14 (18.4)
Age at first intercourse
 ≤22 37 (20.9) 28 (41.8) 0.002 38 (51.4) 0
 >22 140 (79.1) 39 (58.2) 36 (48.6)
Age at first childbirth
 ≤22 24 (13.6) 13 (19.4) 0.317 19 (25.0) 0.030
 >22 153 (86.4) 54 (80.6) 57 (75.0)

Mean (minimum–maximum).

Fisher's exact test.

CIN, cervical intraepithelial neoplasia; SCC, squamous cell carcinoma.

Analysis of the association between the polymorphisms and risk of SCC/CIN with multivariate logistic regression analysis.

Control, n (%) CIN, n (%) AORa Pa SCC, n (%) AORa Pa
XRCC1
 rs1799782 CC 87 (49.2) 32 (44.4) 1.000 41 (51.2) 1.000
CT 71 (40.1) 33 (45.8) 1.293 (0.717–2.329) 0.393 31 (38.8) 0.932 (0.531–1.635) 0.805
TT 19 (10.7) 7 (9.7) 0.950 (0.360–2.510) 0.918 8 (10.0) 0.904 (0.365–2.239) 0.827
Additive 1.076 (0.711–1.628) 0.730 0.944 (0.634–1.404) 0.775
 rs25489 GG 142 (80.2) 61 (85.9) 1.000 68 (85.0) 1.000
GA 34 (19.2) 10 (14.1) 0.628 (0.288–1.373) 0.244 11 (13.8) 0.680 (0.325–1.426) 0.308
AA 1 (0.6) 0 - - 1 (1.2) 2.022 (0.124–32.955) 0.621
Additive 0.613 (0.284–1.323) 0.213 0.781 (0.400–1.525) 0.470
 rs25487 GG 109 (61.9) 38 (52.8) 1.000 43 (53.8) 1.000
GA 58 (32.8) 28 (38.9) 1.371 (0.758–2.481) 0.296 31 (38.8) 1.362 (0.776–2.388) 0.281
AA 10 (5.6) 6 (8.3) 1.668 (0.561–4.961) 0.358 6 (7.5) 1.526 (0.522–4.461) 0.440
Additive 1.326 (0.853–2.062) 0.210 1.293 (0.846–1.976) 0.236
ERCC1
 rs11615 CC 105 (60.0) 45 (62.5) 1.000 39 (48.8) 1.000
CT 61 (34.9) 22 (30.6) 0.870 (0.473–1.600) 0.654 34 (42.5) 1.501 (0.859–2.623) 0.153
TT 9 (5.1) 5 (6.9) 1.460 (0.450–4.735) 0.529 7 (8.8) 2.058 (0.714–5.932) 0.181
Additive 1.026 (0.642–1.638) 0.916 1.465 (0.958–2.241) 0.078
ERCC2
 rs13181 AA 148 (84.1) 57 (79.2) 1.000 68 (85.0) 1.000
CA 27 (15.3) 15 (20.8) 1.452 (0.712–2.958) 0.305 12 (15.0) 0.967 (0.462–2.024) 0.929
CC 1 (0.6) 0 - - 0 - -
Additive 1.359 (0.683–2.704) 0.383 0.896 (0.440–1.826) 0.762
 rs238406 CC 55 (31.6) 22 (30.6) 1.000 29 (36.2) 1.000
CA 83 (47.7) 35 (48.6) 1.042 (0.547–1.983) 0.900 36 (45.0) 0.823 (0.453–1.493) 0.521
AA 36 (20.7) 15 (20.8) 1.014 (0.459–2.237) 0.973 15 (18.8) 0.799 (0.376–1.697) 0.559
Additive 1.010 (0.683–1.493) 0.960 0.883 (0.609–1.281) 0.514
PARP1
 rs1136410 TT 54 (30.7) 28 (39.4) 1.000 25 (31.2) 1.000
CT 83 (47.2) 28 (39.4) 0.581 (0.305–1.107) 0.099 39 (48.8) 1.017 (0.553–1.868) 0.958
CC 39 (22.2) 15 (21.1) 0.775 (0.361–1.664) 0.513 16 (20.0) 0.883 (0.417–1.871) 0.745
Additive 0.836 (0.566–1.234) 0.367 0.947 (0.655–1.370) 0.772
RAD51
 rs1801320 GG 122 (69.7) 46 (64.8) 1.000 58 (72.5) 1.000
CG 50 (28.6) 24 (33.8) 1.316 (0.718–2.414) 0.374 20 (25.0) 0.846 (0.461–1.550) 0.587
CC 3 (1.7)) 1 (1.4) 0.782 (0.078–7.871) 0.835 2 (2.5) 1.420 (0.231–8.738) 0.705
Additive 1.207 (0.699–2.083) 0.499 0.929 (0.546–1.578) 0.784
HER2
 rs1801200 AA 131 (74.0) 52 (72.2) 1.000 62 (77.5) 1.000
GA 44 (24.9) 20 (27.8) 1.147 (0.611–2.152) 0.670 16 (20.0) 0.766 (0.401–1.463) 0.419
GG 2 (1.1) 0 - - 2 (2.5) 2.141 (0.294–15.581) 0.452
Additive 1.029 (0.565–1.876) 0.925 0.908 (0.517–1.596) 0.738

Adjusted for age - sample number equals to zero.

n, number; P, p-value; AOR, adjusted odds ratio.

Analysis of the association between the polymorphisms and risk of SCC/CIN stratified by family smoking status.

No Yes
CIN SCC CIN SCC
AORa Pa AORa Pa AORa Pa AORa Pa
XRCC1
 rs1799782 CC 1.000 1.000 1.000 1.000
CT 1.297 (0.561–2.997) 0.543 0.955 (0.418–2.184) 0.913 1.472 (0.579–3.745) 0.417 0.931 (0.405–2.138) 0.866
TT 0.645 (0.190–2.196) 0.483 0.360 (0.088–1.466) 0.154 1.156 (0.210–6.375) 0.868 1.928 (0.559–6.646) 0.299
Additive 0.915 (0.531–1.579) 0.751 0.710 (0.400–1.260) 0.241 1.222 (0.614–2.433) 0.569 1.223 (0.685–2.183) 0.496
 rs25489 GG 1.000 1.000 1.000 1.000
GA 1.017 (0.323–3.207) 0.977 1.071 (0.340–3.371) 0.907 0.516 (0.166–1.716) 0.281 0.578 (0.202–1.654) 0.307
AA - - 2.263e9 1.000 - - - -
Additive 1.017 (0.323–3.207) 0.977 1.398 (0.512–3.819) 0.513 0.508 (0.155–1.667) 0.264 0.558 (0.199–1.564) 0.267
 rs25487 GG 1.000 1.000 1.000 1.000
GA 1.579 (0.669–3.728) 0.298 2.463 (1.052–5.766) 0.038 1.515 (0.604–3.799) 0.376 0.783 (0.334–1.834) 0.573
AA 1.167 (0.316–4.303) 0.817 0.960 (0.214–4.305) 0.957 1.769 (0.172–18.239) 0.632 2.718 (0.554–13.348) 0.218
Additive 1.222 (0.689–2.167) 0.493 1.419 (0.777–2.592) 0.255 1.443 (0.666–3.127) 0.352 1.134 (0.591–2.177) 0.705
ERCC1
 rs11615 CC 1.000 1.000 1.000 1.000
CT 0.718 (0.305–1.687) 0.447 1.281 (0.564–2.907) 0.554 1.215 (0.470–3.143) 0.688 2.088 (0.910–4.788) 0.082
TT 1.343 (0.302–5.978) 0.698 1.009 (0.206–4.947) 0.992 0.954 (0.094–9.655) 0.968 4.913 (1.116–21.634) 0.035
Additive 0.945 (0.511–1.748) 0.857 1.128 (0.605–2.101) 0.705 1.111 (0.514–2.404) 0.788 2.161 (1.152–4.054) 0.016
ERCC2
 rs13181 AA 1.000 1.000 1.000 1.000
CA 1.500 (0.549–4.101) 0.429 0.569 (0.163–1.991) 0.378 1.494 (0.480–4.649) 0.488 1.730 (0.669–4.472) 0.258
CC - - - - - - - -
Additive 1.500 (0.549–4.101) 0.429 0.550 (0.158–1.913) 0.550 1.394 (0.466–4.169) 0.552 1.536 (0.623–3.785) 0.351
 rs238406 CC 1.000 1.000 1.000 1.000
CA 0.550 (0.220–1.376) 0.201 0.695 (0.281–1.723) 0.432 2.423 (0.758–7.745) 0.135 0.800 (0.333–1.926) 0.619
AA 0.360 (0.116–1.114) 0.076 0.423 (0.133–1.347) 0.145 3.461 (0.913–13.115) 0.068 1.298 (0.459–3.671) 0.623
Additive 0.595 (0.339–1.046) 0.071 0.655 (0.370–1.159) 0.146 1.843 (0.970–3.500) 0.062 1.093 (0.642–1.859) 0.744
PARP1
 rs1136410 TT 1.000 1.000 1.000 1.000
CT 0.431 (0.179–1.041) 0.062 0.786 (0.325–1.901) 0.593 0.948 (0.326–2.757) 0.923 1.157 (0.473–0.833) 0.749
CC 0.971 (0.313–3.007) 0.959 1.394 (0.448–4.331) 0.566 1.028 (0.301–3.510) 0.964 0.709 (0.229–2.192) 0.550
Additive 0.844 (0.486–1.466) 0.547 1.115 (0.638–1.950) 0.702 1.012 (0.545–1.877) 0.971 0.869 (0.510–1.482) 0.607
RAD51
 rs1801320 GG 1.000 1.000 1.000 1.000
CG 1.233 (0.551–2.760) 0.611 0.819 (0.349–1.921) 0.646 0.853 (0.283–2.573) 0.778 0.770 (0.301–1.969) 0.585
CC - - 2.729 (0.234–31.843) 0.423 1.875 (0.152–23.055) 0.624 - -
Additive 1.085 (0.504–2.335) 0.835 1.026 (0.496–2.119) 0.946 1.027 (0.424–2.490) 0.953 0.689 (0.284–1.671) 0.410
HER2
 rs1801200 AA 1.000 1.000 1.000 1.000
GA 0.970 (0.422–2.228) 0.943 0.544 (0.218–1.359) 0.192 0.974 (0.320–2.967) 0.963 1.041 (0.400–2.710) 0.935
GG - - - - - - 3.583 (0.471–27.226) 0.217
Additive 0.970 (0.422–2.228) 0.943 0.544 (0.218–1.359) 0.192 0.823 (0.294–2.305) 0.712 1.359 (0.654–2.824) 0.412

Adjusted for age - sample number equals to zero.

n, number; P, p-value; AOR, adjusted odds ratio. Bold numbers indicate statistical significance.

Analysis of the association between the polymorphisms and risk of SCC/CIN stratified by age at first intercourse.

≤22 >22
CIN SCC CIN SCC
AORa Pa AORa Pa AORa Pa AORa Pa
XRCC1
 rs1799782 CC 1.000 1.000 1.000 1.000
CT 2.040 (0.711–5.852) 0.185 1.131 (0.382–3.349) 0.825 0.991 (0.463–2.121) 0.982 0.996 (0.468–2.121) 0.992
TT 3.122 (0.438–22.263) 0.256 5.071 (0.795–32.331) 0.086 0.786 (0.232–2.658) 0.699 0.197 (0.024–1.591) 0.127
Additive 1.888 (0.850–4.194) 0.119 1.743 (0.824–3.690) 0.146 0.920 (0.540–1.568) 0.760 0.700 (0.390–1.258) 0.233
 rs25489 GG 1.000 1.000 1.000 1.000
GA 0.284 (0.071–1.142) 0.076 0.242 (0.063–0.930) 0.039 1.068 (0.414–2.757) 0.892 0.991 (0.367–2.677) 0.986
AA - - - - - - 4.508 (0.270–75.395) 0.295
Additive 0.284 (0.071–1.142) 0.076 0.242 (0.063–0.930) 0.039 1.002 (0.401–2.501) 0.997 1.239 (0.537–2.861) 0.615
 rs25487 GG 1.000 1.000 1.000 1.000
GA 1.945 (0.681–5.559) 0.214 0.975 (0.352–2.701) 0.961 1.052 (0.483–2.296) 0.898 1.737 (0.799–3.779) 0.164
AA 6.132 (0.524–71.825) 0.149 0.775 (0.044–13.594) 0.862 1.053 (0.258–4.301) 0.942 1.601 (0.389–6.597) 0.514
Additive 2.151 (0.901–5.133) 0.084 0.944 (0.392–2.276) 0.899 1.038 (0.586–1.837) 0.899 1.442 (0.817–2.547) 0.207
ERCC1
 rs11615 CC 1.000 1.000 1.000 1.000
CT 1.258 (0.411–3.851) 0.688 1.685 (0.599–4.736) 0.323 0.653 (0.294–1.450) 0.295 1.341 (0.611–2.942) 0.464
TT 4.608 (0.432–49.149) 0.206 2.476 (0.216–28.394) 0.466 0.814 (0.159–4.183) 0.806 2.478 (0.663–9.267) 0.178
Additive 1.643 (0.704–3.833) 0.251 1.640 (0.692–3.885) 0.261 0.755 (0.400–1.423) 0.384 1.478 (0.834–2.620) 0.181
ERCC2
 rs13181 AA 1.000 1.000 1.000 1.000
CA 2.066 (0.623–6.857) 0.236 1.082 (0.297–3.948) 0.905 1.252 (0.484–3.237) 0.643 0.917 (0.319–2.636) 0.872
CC - - - - - - - -
Additive 2.066 (0.623–6.857) 0.236 1.082 (0.297–3.948) 0.905 1.156 (0.464–2.877) 0.756 0.855 (0.309–2.365) 0.762
 rs238406 CC 1.000 1.000 1.000 1.000
CA 0.749 (0.241–2.324) 0.617 0.511 (0.164–1.590) 0.247 1.049 (0.458–2.403) 0.909 1.061 (0.466–2.415) 0.888
AA 1.404 (0.354–5.565) 0.629 1.494 (0.388–5.758) 0.560 0.908 (0.327–2.521) 0.852 0.596 (0.189–1.876) 0.377
Additive 1.125 (0.569–2.224) 0.735 1.112 (0.578–2.141) 0.751 0.961 (0.583–1.585) 0.877 0.819 (0.485–1.383) 0.455
PARP1
 rs1136410 TT 1.000 1.000 1.000 1.000
CT 0.585 (0.188–1.820) 0.355 0.493 (0.162–1.503) 0.214 0.552 (0.235–1.294) 0.172 1.202 (0.504–2.862) 0.678
CC 0.573 (0.140–2.347) 0.439 0.661 (0.176–2.481) 0.539 1.019 (0.399–2.601) 0.969 1.150 (0.406–3.257) 0.792
Additive 0.733 (0.363–1.478) 0.385 0.773 (0.401–1.490) 0.443 0.956 (0.583–1.566) 0.857 1.080 (0.648–1.799) 0.768
RAD51
 rs1801320 GG 1.000 1.000 1.000 1.000
CG 3.241 (1.050–10.005) 0.041 1.915 (0.603–6.082) 0.270 0.877 (0.393–1.955) 0.748 0.592 (0.248–1.416) 0.239
CC 3.408e9 NA 1.696e9 NA - - - -
Additive 3.524 (1.208–10.286) 0.021 2.347 (0.836–6.589) 0.105 0.764 (0.362–1.614) 0.481 0.540 (0.235–1.237) 0.145
HER2
 rs1801200 AA 1.000 1.000 1.000 1.000
GA 1.724 (0.538–5.523) 0.359 1.687 (0.529–5.375) 0.377 1.194 (0.544–2.623) 0.658 0.344 (0.114–1.042) 0.059

Adjusted for age - sample number equals to zero.

n, number; P, p-value; AOR, adjusted odds ratio. Bold numbers indicate statistical significance.

Analysis of the association between the polymorphisms and risk of SCC/CIN stratified by age at first giving birth.

≤22 >22
CIN SCC CIN SCC
AORa Pa AORa Pa AORa Pa AORa Pa
XRCC1
 rs1799782 CC 1.000 1.000 1.000 1.000
CT 1.260 (0.313–5.069) 0.745 0.516 (0.117–2.282) 0.383 1.326 (0.685–2.569) 0.402 1.018 (0.536–1.936) 0.956
TT - - 0.632 (0.085–4.723) 0.655 0.898 (0.294–2.736) 0.849 0.585 (0.178–1.923) 0.377
Additive 0.784 (0.247–2.494) 0.680 0.713 (0.279–1.820) 0.479 1.069 (0.671–1.703) 0.780 0.863 (0.538–1.384) 0.541
 rs25489 GG 1.000 1.000 1.000 1.000
GA 0.427 (0.067–2.739) 0.370 0.491 (0.088–2.725) 0.416 0.756 (0.318–1.796) 0.526 0.707 (0.299–1.670) 0.429
AA - - - - - - 3.165 (0.190–52.685) 0.422
Additive 0.427 (0.067–2.739) 0.370 0.491 (0.088–2.725) 0.416 0.727 (0.312–1.692) 0.459 0.872 (0.409–1.859) 0.722
 rs25487 GG 1.000 1.000 1.000 1.000
GA 8.522 (1.671–43.448) 0.010 2.588 (0.583–11.487) 0.211 1.016 (0.515–2.005) 0.963 1.158 (0.601–2.230) 0.662
AA 6.389 (0.296–137.767) 0.237 2.215 (0.118–41.529) 0.595 1.554 (0.479–5.040) 0.463 1.646 (0.507–5.340) 0.406
Additive 4.766 (1.266–17.943) 0.021 1.965 (0.622–6.201) 0.250 1.144 (0.697–1.876) 0.595 1.227 (0.755–1.993) 0.410
ERCC1
 rs11615 CC 1.000 1.000 1.000 1.000
CT 1.384 (0.290–6.611) 0.684 1.639 (0.393–6.835) 0.498 0.702 (0.350–1.408) 0.319 1.521 (0.798–2.900) 0.203
TT 2.540 (0.273–23.633) 0.413 1.144 (0.118–11.128) 0.908 1.182 (0.283–4.934) 0.818 2.682 (0.774–9.297) 0.120
Additive 1.531 (0.556–4.219) 0.410 1.242 (0.458–3.369) 0.671 0.853 (0.491–1.484) 0.574 1.581 (0.960–2.603) 0.072
ERCC2
 rs13181 AA 1.000 1.000 1.000 1.000
CA 1.788 (0.293–10.932) 0.529 3.559 (0.573–22.112) 0.173 1.423 (0.639–3.172) 0.388 0.905 (0.379–2.164) 0.823
CC - - - - - - - -
Additive 1.788 (0.293–10.932) 0.529 3.559 (0.573–22.112) 0.173 1.315 (0.608–2.844) 0.487 0.853 (0.367–1.986) 0.713
 rs238406 CC 1.000 1.000 1.000 1.000
CA 0.758 (0.142–4.044) 0.746 1.012 (0.243–4.218) 0.986 1.032 (0.504–2.113) 0.931 0.726 (0.360–1.464) 0.371
AA 2.364 (0.393–14.208) 0.347 0.488 (0.039–6.032) 0.576 0.856 (0.345–2.127) 0.738 0.913 (0.398–2.094) 0.829
Additive 1.508 (0.604–3.766) 0.379 0.803 (0.280–2.297) 0.682 0.937 (0.602–1.459) 0.775 0.929 (0.609–1.416) 0.732
PARP1
 rs1136410 TT 1.000 1.000 1.000 1.000
CT 0.615 (0.140–2.692) 0.518 0.862 (0.192–3.881) 0.847 0.607 (0.287–1.284) 0.192 0.919 (0.457–1.850) 0.814
CC - - 1.260 (0.191–8.324) 0.810 1.129 (0.498–2.559) 0.772 0.838 (0.351–2.001) 0.690
Additive 0.358 (0.109–1.179) 0.091 1.087 (0.426–2.768) 0.862 1.028 (0.670–1.575) 0.901 0.916 (0.597–1.406) 0.688
RAD51
 rs1801320 GG 1.000 1.000 1.000 1.000
CG 1.342 (0.243–7.423) 0.736 1.349 (0.204–8.936) 0.756 1.307 (0.668–2.560) 0.435 0.891 (0.451–1.761) 0.740
CC - - 1.829e9 N/A 0.902 (0.089–9.119) 0.930 0.802 (0.081–7.973) 0.851
Additive 1.342 (0.243–7.423) 0.736 1.991 (0.438–9.051) 0.372 1.207 (0.665–2.190) 0.537 0.892 (0.486–1.638) 0.713
HER2
 rs1801200 AA 1.000 1.000 1.000 1.000
GA 3.294 (0.580–18.718) 0.179 1.736 (0.295–10.213) 0.542 0.932 (0.457–1.901) 0.846 0.735 (0.352–1.535) 0.412
GG - - - - - - 1.259 (0.109–14.543) 0.853
Additive 3.294 (0.580–18.718) 0.179 1.726 (0.295–10.213) 0.542 0.843 (0.428–1.663) 0.623 0.807 (0.417–1.564) 0.526

Adjusted for age - sample number equals to zero.

n, number; P, p-value; AOR, adjusted odds ratio.