Polymorphisms in Nucleotide Excision Repair Genes and Endometrial Cancer Risk

Published OnlineFirst July 12, 2011; DOI: 10.1158/1055-9965.EPI-11-0119

Cancer Epidemiology, Research Article Biomarkers & Prevention

Jennifer A. Doherty1,2, Noel S. Weiss1,2, Sherianne Fish1, Wenhong Fan1, Melissa M. Loomis1, Lori C. Sakoda1,2, Mary Anne Rossing1,2, Lue Ping Zhao1, and Chu Chen1,2,3

Abstract Background: Exposure to estrogens increases the risk of endometrial cancer. Certain estrogen metabolites can form bulky DNA adducts, which are removed via nucleotide excision repair (NER), and the ability to carry out this repair might be related to endometrial cancer risk. Methods: We examined 64 tag and functional single-nucleotide polymorphisms (SNPs) in the NER genes ERCC1, ERCC2 (XPD), ERCC3 (XPB), ERCC4 (XPF), ERCC5 (XPG), LIG1, XPA, and XPC in a population-based case–control study in Washington state, with 783 endometrial cancer cases and 795 controls. Results: The presence of ERCC5 rs4150386 C, LIG1 rs3730865 C, XPA rs2808667 T, or XPC rs3731127 T alleles was associated with risk of endometrial cancer, with respective age-, county-, and reference year– adjusted per-allele ORs and 95% CIs of 0.68 (0.53–0.87, P ¼ 0.002), 1.46 (1.02–2.10, P ¼ 0.04), 0.71 (0.52–0.97, P ¼ 0.03), and 1.57 (1.13–2.17, P ¼ 0.007), respectively. Conclusions: Certain ERCC5, LIG1, XPA, and XPC genotypes might influence endometrial cancer risk. Impact: Because of multiple redundancies in DNA repair pathways (and therefore a low prior probability) and the large number of associations examined, false-positive findings are likely. Further characterization of the relation between variation in NER genes and endometrial cancer risk is warranted. Cancer Epidemiol Biomarkers Prev; 20(9); 1873–82. 2011 AACR.

Introduction The ability to repair DNA damage caused by estrogens could plausibly affect endometrial cancer risk. Nucleo- The increased risk of endometrial cancer associated tide excision repair (NER) identifies and excises bulky with estrogen exposure has been hypothesized to result, adducts formed by estrogen, as well as by other sub- at least in part, via 2 possible mechanisms: (i) increased stances such as components of cigarette smoke and well- cellular proliferation (1, 2) and (ii) through the production cooked meat (7). NER also corrects UV-induced pyrimi- of certain estrogen metabolites, which can result in DNA dine dimers and cross-links (8). As many as 30 proteins, damage by directly binding to DNA and forming bulky including those encoded by the xeroderma pigmentosum DNA adducts, and/or indirectly through the production (XP) genes, are involved in this process, in which a large of reactive oxygen species (3–5). Estradiol and estrone are multiprotein complex is assembled. First, the damaged metabolized primarily via 2-, 4-, and 16a-hydroxylation DNA is recognized by XPC, centrin 2, and HHRAD23B. (3, 6). The 2- and 4-hydroxylated metabolites can be fur- XPA and RPA (replication protein A) then bind to the ther oxidized to semiquinones and quinones, which can lesion, allowing the proper positioning of the endonu- form bulky DNA adducts and can undergo redox cycling, cleases ERCC1 (excision repair cross-complimenting 1)- producing reactive oxygen species that may cause oxida- ERCC4 (XPF) and ERCC5 (XPG; ref. 9). After the DNA in tive stress, lipid peroxidation, and DNA damage (4, 5). this region is unwound by the helicases ERCC3 (XPB) and ERCC2 (XPD), which are 2 subunits of the transcription factor IIH, ERCC1-ERCC4 and ERCC5 (XPG) excise a 0 0 Authors' Affiliations: 1Division of Public Health Sciences, Fred Hutch- 30-bp oligonucleotide fragment (at the 5 and 3 ends, inson Cancer Research Center; 2Department of Epidemiology, School of respectively) containing the bulky adduct. DNA poly- Public Health, and 3Department of Otolaryngology: Head and Neck Sur- gery, School of Medicine, University of Washington, Seattle, Washington merases and ligases use the intact strand as a template to fill the resulting gap (8). Note: Supplementary data for this article are available at Cancer Epide- miology, Biomarkers & Prevention Online (http://cebp.aacrjournals.org/). In an earlier study, we examined whether 11 candidate Corresponding Author: Chu Chen, Fred Hutchinson Cancer Research variants in NER genes were associated with endometrial Center, P.O. Box 19024, Mailstop M5-800, Seattle, WA 98109. Phone: cancer risk (371 cases and 420 controls; ref. 10). We have 206-667-6644; Fax: 206-667-2537; E-mail: [email protected] since nearly doubled our study population, and herein, we doi: 10.1158/1055-9965.EPI-11-0119 report findings from a more systematic evaluation of sin- 2011 American Association for Cancer Research. gle-nucleotide polymorphism (SNP) in the genes ERCC1,

www.aacrjournals.org 1873

Doherty et al.

ERCC2 (XPD), ERCC3 (XPB), ERCC4 (XPF), ERCC5 (XPG), For the most recent study (2003–2005; also described LIG1, XPA,andXPC in the combined studies. in Bodelon and colleagues; ref. 15), eligible case participants included all female residents of western Methods Washington state aged 50 to 74 years diagnosed with invasive endometrial cancer between July 1, 2003, and Study subjects November 31, 2005, in King, Pierce, and Snohomish We genotyped a total of 783 residents of western counties, identified through the Cancer Surveillance Washington who were diagnosed with endometrial can- System. Of 586 eligible cases, we were unable to locate cer in 1994–1995, 1997–1999, and 2003–2005 and 795 12, 34 were deceased before we could contact them, and population-based controls. The studies of women diag- 130 refused to participate (or their physician instructed nosed during the 1990s have been described previously us not to contact them). A total of 410 (70.0%) women (11). Briefly, eligible cases (n ¼ 582) comprised Caucasian were successfully interviewed. Of these, 401 provided a and African-American female residents of western blood or buccal sample. Washington state aged 50 to 69 years diagnosed with Control women were a subset of those enrolled in a invasive endometrial cancer between January 1, 1994, and population-based case–control study of ovarian cancer in December 31, 1995 (King county only), and between western Washington that included women aged 35 to 74 July 1, 1997, and December 31, 1999 (King, Pierce, and years with reference dates of 2002 to 2005, described in Snohomish counties). These women were identified detail in Rossing and colleagues (16). Controls were through the Cancer Surveillance System, a population- selected by RDD (13), and 82.0% of 17,768 residential based tumor registry affiliated with the Surveillance, telephone numbers were screened to determine whether Epidemiology and End Results Program of the National an eligible woman resided there. Of the 1,561 women Cancer Institute (12). A total of 472 (81.1%) women were identified as eligible controls, 1,313 were interviewed successfully interviewed and 383 of them (81.1%) pro- (84.1%); the remaining women refused (n ¼ 240) or were vided a blood sample. lost to follow-up (n ¼ 8). The overall control response Eligible control women included Caucasian and Afri- proportion (screening response interview response) can American female residents of the 3-county area dur- was 69.0%. From this set of population-based controls, ing the years the cases were diagnosed, with intact uteri we attempted to recruit women aged 50 to 74 years with and no prior history of endometrial cancer. Those intact uteri, who resided in King, Pierce, and Snohomish selected by random-digit dialing (RDD; ref. 13; women counties, and had reference dates comparable with the aged 50–65 years) and random selection from Health Care endometrial cancer cases. Of 365 eligible controls, 9 Financing Administration data files (women aged 66–69 women refused to participate and 356 women were years) were frequency matched to the cases by 5-year age interviewed. A total of 347 control women provided a group and county of residence. The overall RDD response blood or buccal sample (though 2 of these samples were [the screening response (91.3%) multiplied by the inter- collected after the other samples had been plated and view response (83.6%)] was 76.3%, resulting in 297 inter- thus were not genotyped). viewed women. Of the 175 eligible Health Care Financing After informed consent, all participants were adminis- Administration controls, 116 (66.3%) agreed to an inter- tered an in-person interview conducted according to a view. An additional source of population-based controls standard protocol. Each participant was asked only about was the CARE study, which was conducted during the events that occurred before her reference date, which is same period as the endometrial cancer case–control study the date of diagnosis for cases. Controls were assigned a using a similar questionnaire (14). The CARE study con- reference date on the basis of the distribution of diagnosis trols included Caucasian and African American women years for the cases. Data were collected on demographic aged 35 to 64 years ascertained through RDD in 5 metro- factors; height; weight at different ages; reproductive, politan areas of the United States, including King county, contraceptive, and menstrual history; family history of between 1994 and 1998. The screening and interview cancer; history of selected chronic conditions; and history response for King county were 83.6% and 88.3%, respec- of contraceptive and noncontraceptive hormone use. tively. We invited 132 King county CARE control women Color pictures of oral contraceptive and hormone repla- aged 50 to 64 years, with intact uteri, to provide a blood cement therapy pill packs were used to aid recall. The sample, and we successfully obtained a blood sample protocols of the studies were approved by the Institu- from 115. Overall, of the 929 eligible controls, 664 (71.5%) tional Review Board of the Fred Hutchinson Cancer were interviewed and 449 provided a blood sample Research Center. (67.7% of interviewed controls). The data from 2 controls in the earlier case–control study who were ascertained as Laboratory methods cases in the later case–control study, were included in We selected 64 tag and functional SNPs in ERCC1, both case and control groups; and 1 case was excluded ERCC2 (XPD), ERCC3 (XPB), ERCC4 (XPF), ERCC5 because of poor quality interview data. Thus, for the (XPG), LIG1, XPA, and XPC using the LDSelect algorithm earlier 2 studies, there were 382 cases and 449 controls on the National Institute of Environmental Health with blood samples available for genotyping. Sciences (NIEHS) Environmental Genome Project (17)

1874 Cancer Epidemiol Biomarkers Prev; 20(9) September 2011 Cancer Epidemiology, Biomarkers & Prevention

Nucleotide Excision Repair Genes and Endometrial Cancer

sequencing data from 23 European Americans (>5% allele regression model as the penetrance function, HPlus frequency; r2 > 0.64). directly assesses associations between haplotypes and Genotyping was conducted blinded to all characteris- outcome (e.g., endometrial cancer), after taking into tics of the study participants. In-house DNA of known account uncertainties of haplotype inference from genotypes was used for positive controls and negative unphased genotype data. controls were prepared identically but without template. Because endometrial cancer risk is so strongly influ- Cases and controls were arranged randomly on plates enced by obesity and menopausal use of high-risk estro- including 56 duplicate pairs of samples used to evaluate gen regimens, we explored whether the magnitude of the genotyping concordance. We used several genotyping associations with the SNPs varied by exposure to these methods including SNPlex, TaqMan Assay On Demand, factors. Because of the particularly high risk of endome- TaqMan Assay by Design, Snapshot (ABI), RFLP assays, trial cancer associated with a body mass index (BMI) 30 and fragment analyses. The assay method used for each kg/m2, we dichotomized BMI as <30 kg/m2 and 30 SNP and details about the assays (e.g., primers, probes, kg/m2. High-risk estrogen use was dichotomized as use restriction enzymes where applicable) are included in or nonuse of greater than 6 months of either unopposed Supplementary Table S1. estrogen or estrogen used with less than 10 days of progestin per month. We also examined associations Statistical methods stratified by parity (nulliparous vs. parous) and age Quality control procedures. For each polymorphism, (<55 years and 55 years). Tests for interaction between we calculated the proportion of samples that were suc- SNPs and these factors were conducted using a likelihood cessfully genotyped (the SNP genotyping success propor- ratio test comparing a model with the SNP of interest to a tion). For SNPs genotyped using the SNPlex platform, we model with an additional cross-product term for the SNP also assessed sample performance. The SNP genotyping and the exposure of interest. success proportion was calculated after excluding sam- For the set of SNPs that were associated with endo- ples that failed SNPlex (defined as those which were metrial cancer in our study, we evaluated whether successfully genotyped for <90% of the SNPlex pool). including them all in a single logistic regression model These samples were included in calculating genotyping affected the magnitude of the association between each of success for SNPs assayed with platforms other than the SNPs and endometrial cancer. We also evaluated SNPlex. For all of the SNPs, we evaluated genotype call whether carriage of a combination of the genotypes concordance between the 56 pairs of duplicate samples, was associated with endometrial cancer risk by first and we tested for deviation from Hardy–Weinberg equi- dichotomizing the data for each SNP, combining hetero- librium (HWE) among non-Hispanic white controls using zygotes and homozygotes for the minor allele. We Fisher’s exact test. SNPs that were successfully geno- defined "risk" genotypes as the following: if our observed typed in less than 95% of the samples or with duplicate OR for the minor allele genotypes was greater than 1, then concordance of less than 98% were excluded except as the "risk" genotype included heterozygotes and homo- noted in the Results section. SNPs that deviated from zygotes for the minor allele; and if our observed OR was HWE at P < 0.01 were also excluded. less than 1, then the homozygous major allele genotype Association analyses. Because we had too few was assigned as the "risk" genotype. women who were Hispanic or nonwhite (56 cases and Finally, we calculated false-positive report probabil- 59 controls) to be able to conduct meaningful analyses of ities (FPRP) for each SNP, using the Excel spreadsheet these subgroups, all analyses were restricted to non- provided by Wacholder and colleagues (23). The FPRP is Hispanic white women (727 cases and 736 controls). "the probability of no association given a statistically Per-minor allele ORs and 95% CIs for each SNP and significant finding." It is determined by the P value, endometrial cancer risk were calculated using uncondi- the prior probability for the association, and the statistical tional logistic regression and were adjusted for age, power (23). Prior probabilities are based on previous county, and reference year. We also calculated ORs knowledge of the relevance of the gene/SNP of interest and 95% CIs for homozygotes and heterozygotes com- to the outcome and evolve over time as more becomes pared with homozygotes for the common allele. These known. We therefore calculated the FPRP for a range of analyses were conducted using Stata SE 11 statistical prior probabilities [high (0.1), moderate (0.01), and software. The Max(T) permutation test implemented in low (0.001)]. Wacholder and colleagues (23) recom- PLINK v1.07 (18, 19), with 10,000 permutations, was used mend that for an initial study of an SNP–disease associa- to evaluate whether there were gene-level associations tion such as ours, the FPRP should be set at less than 0.5 to with endometrial cancer risk, after taking into account designate findings that are "noteworthy." linkage disequilibrium (LD) between SNPs and the number of SNPs in the gene. We used HPlus (20–22) to Results estimate gene-specific haplotype frequencies and calculated ORs and 95% CIs to estimate their associations with Compared with controls, endometrial cancer cases endometrial cancer risk, using the most common haplo- were slightly older, had higher BMI, had fewer births, type as the reference category. Utilizing the logistic were less likely to have used oral contraceptives, and

www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 20(9) September 2011 1875

Doherty et al.

ERCC2 Table 1. Characteristics of cases and controls otyping success for rs50872 was 92.8%, but because the duplicate concordance was 100%, we did Cases Controls not exclude this SNP. The duplicate concordance for (n ¼ 727) (n ¼ 736) ERCC5 rs4150375 was 96.4%, but genotyping success was 95%, so it was also included. The genotype dis- n % n % tribution for XPA rs3176689 deviated from that expected under HWE (P ¼ 0.001), but the genotype calls were Reference year unambiguous, so we did not exclude this SNP from – 1994 1999 371 51.0 424 57.6 analyses. – 2003 2005 356 49.0 312 42.4 Most of the genotypes that we investigated were not Age at reference associated with endometrial cancer risk (Table 2 and date, y Table 3). However, the presence of the rs4150386 C allele < 55 178 24.5 196 26.6 (ERCC5), the rs3730865 C allele (LIG1), the rs2808667 T < 55 to 60 194 26.7 192 26.1 allele (XPA), and the rs3731127 T allele (XPC) was asso- < 60 to 65 169 23.2 184 25.0 ciated with risk, with respective per-allele ORs, 95% CIs, < 65 to 70 150 20.6 133 18.1 and P values of 0.68 (0.53–0.87, P ¼ 0.002), 1.46 (1.02–2.10, < 70 to 75 36 5.0 31 4.2 P ¼ 0.04), 0.71 (0.52–0.97, P ¼ 0.03), and 1.57 (1.13–2.17, 2 BMI, kg/m P ¼ 0.007; Table 3). The empirical gene-level P values < 20 12 1.7 12 1.6 generated from gene set permutation testing (which takes < 20 to 25 226 31.3 402 54.7 into account the LD structure of the gene and the number < 25 to 30 184 25.5 187 25.4 of SNPs tested in the gene) were 0.007 for ERCC1, 0.12 for < 30 to 35 201 27.8 117 15.9 LIG1, 0.08 for XPA, and 0.04 for XPC. The association þ 35 99 13.7 17 2.3 between ERCC5 rs4150386 and endometrial cancer risk Missing 5 1 was considerably stronger among obese women (BMI Oral contraceptive 30 kg/m2: per-C-allele OR ¼ 0.41, 95% CI: 0.25–0.68) use, mo than women with BMI < 30 kg/m2 (per-C-allele OR ¼ Never user 245 34.0 180 24.6 0.85, 95% CI: 0.64–1.14, P ¼ 0.03). For both LIG1 – interaction 6 59 320 44.4 294 40.2 rs3730865 and XPA rs2808667, associations were stronger þ 60 156 21.6 257 35.2 among users of high-risk estrogen regimens than never Missing 6 5 users. For LIG1 rs3730865, per-C-allele ORs and 95% CIs Cigarette smoking were respectively: 2.98, 1.06–8.37 and 1.27, 0.86–1.89 Never 404 55.6 356 48.4 P ¼ XPA ( interaction 0.10), and for rs2808667, per-T-allele Former 235 32.3 246 33.4 ORs and 95% CIs were respectively: 0.39, 0.20–0.76 and Current (within last 5 y) 88 12.1 134 18.2 P ¼ 0.81, 0.57–1.16 ( interaction 0.08). The association with Number of term XPC rs3731127 did not vary by BMI or use of high-risk pregnancies estrogen regimens. For several SNPs in XPA and XPC 0 130 18.3 98 13.5 that were not associated with endometrial cancer risk 1 91 12.8 73 10.0 overall, suggestive associations were observed only 2 227 32.0 243 33.4 among women who had used high-risk estrogen regi- 3 156 22.0 184 25.3 mens (e.g., rs3176748, rs3176658, and rs1800975 in XPA þ 4 106 14.9 130 17.9 and rs3731151, rs2228000, and rs2733537 in XPC; Table 3). Missing 17 8 The magnitude of the observed associations did not vary Use of a high-risk by parity or age (data not shown). menopausal estrogen For LIG1 rs3730865, XPA rs2808667, and XPC regimen for 6mo rs3731127, associations were restricted to the single hap- Never 551 76.6 610 84.3 lotype in each respective gene that contained the variant Ever 168 23.4 114 15.8 allele (ORs and 95% CIs, respectively: 1.46, 0.99–2.15, P ¼ Missing 8 12 0.06; 0.70, 0.51–0.96, P ¼ 0.03; and 1.48, 1.04–2.11, P ¼ 0.03). The 2 haplotypes that contained the minor allele of ERCC5 rs4150386 were also associated with decreased were less likely to have ever smoked. Cases were also risk (0.75, 0.53–1.05, P ¼ 0.09; and 0.46, 0.22–0.97, P ¼ 0.04; more likely than controls to have taken a high-risk hor- Table 4). In a logistic model including all 4 of these SNPs, mone therapy regimen (Table 1). Distributions of risk results for ERCC5 rs4150386, LIG1 rs3730865, and XPC factors were similar for women who were and were not rs3731127 were unchanged. However, the magnitude of genotyped (data not shown). the association between the XPA rs2808667 T allele and Assays failed for 1 SNP in ERCC2 (rs3810366), 2 in endometrial cancer risk was closer to the null than the ERCC4 (rs1799797 and rs1799801), and 1 each in LIG1 unadjusted result (data not shown). We therefore (rs156641), XPA (rs2805834), and XPC (rs2227999). Gen- decided to exclude this SNP from the combined genotype

1876 Cancer Epidemiol Biomarkers Prev; 20(9) September 2011 Cancer Epidemiology, Biomarkers & Prevention

Downloaded from cebp.aacrjournals.org on September 30, 2021. © 2011 American Association for Cancer Research. www.aacrjournals.org Table 2. Distribution of genotypes for SNPs in ERCC1, ERCC2, ERCC3, and ERCC4 in women with endometrial cancer and controls, and per- Downloaded from allele ORs by BMI and use of a high-risk estrogen regimen

Genotype distributiona All BMI < 30 BMI 30 No use of Use of high-risk kg/m2 kg/m2 high-risk estrogen estrogen regimenb regimenb

cebp.aacrjournals.org AA Aa aa (727 cases, (422 cases, (303 cases, (551 cases, (168 cases,

736 controls) 601 controls) 135 controls) 610 controls) 114 controls) Published OnlineFirstJuly12,2011;DOI:10.1158/1055-9965.EPI-11-0119

Major Minor Controls Cases Controls Cases Controls Cases OR (95% CI) P OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) allele allele

ERCC1 rs32129866 C A 388 392 271 266 46 41 0.94 (0.79–1.12) 0.50 0.97 (0.79–1.20) 0.90 (0.64–1.27) 0.93 (0.76–1.13) 0.93 (0.61–1.43) rs3212985 G C 437 434 242 249 41 36 0.99 (0.83–1.18) 0.93 1.04 (0.85–1.29) 0.96 (0.67–1.37) 0.96 (0.78–1.17) 1.01 (0.66–1.53) on September 30, 2021. © 2011American Association for Cancer Research. rs3212961 C A 538 517 152 150 11 14 1.06 (0.85–1.33) 0.62 0.99 (0.75–1.29) 1.31 (0.81–2.10) 1.09 (0.85–1.41) 1.08 (0.63–1.85) rs11615 T C 275 283 337 338 92 103 1.04 (0.89–1.21) 0.64 1.05 (0.87–1.27) 0.96 (0.70–1.31) 1.03 (0.86–1.22) 1.12 (0.76–1.63) rs3212939 C T 626 613 40 34 0 1 0.88 (0.55–1.39) 0.58 0.87 (0.50–1.49) 1.45 (0.45–4.63) 1.10 (0.65–1.87) 0.35 (0.12–1.02) rs2298881 C A 575 563 127 133 11 6 0.97 (0.76–1.24) 0.83 0.88 (0.66–1.18) 1.39 (0.82–2.35) 1.08 (0.82–1.43) 0.77 (0.45–1.34) rs3212930 T C 449 429 219 225 34 27 1.02 (0.85–1.23) 0.82 1.10 (0.88–1.38) 0.89 (0.62–1.29) 0.94 (0.76–1.16) 1.27 (0.80–1.99) ERCC2 (XPD) rs3916898 T G 624 619 95 95 2 5 1.08 (0.81–1.44) 0.60 1.08 (0.77–1.53) 1.15 (0.63–2.08) 1.11 (0.80–1.52) 0.93 (0.44–1.94) rs13181 T G 282 269 333 347 99 87 0.99 (0.85–1.16) 0.90 0.97 (0.80–1.17) 0.97 (0.71–1.32) 0.93 (0.78–1.11) 1.23 (0.85–1.79) rs1052555 C T 314 299 313 310 74 67 0.99 (0.84–1.17) 0.92 0.95 (0.78–1.16) 0.96 (0.70–1.34) 0.96 (0.79–1.15) 1.09 (0.75–1.60) rs3916874 G C 358 365 288 273 56 43 0.89 (0.75–1.06) 0.19 0.93 (0.76–1.15) 0.77 (0.55–1.07) 0.89 (0.74–1.08) 0.92 (0.61–1.37) uloieEcso earGnsadEdmtilCancer Endometrial and Genes Repair Excision Nucleotide acrEieilBoakr rv 09 etme 2011 September 20(9) Prev; Biomarkers Epidemiol Cancer rs238415 C G 255 228 329 322 105 118 1.12 (0.95–1.30) 0.17 1.12 (0.93–1.36) 1.33 (0.96–1.83) 1.17 (0.98–1.40) 0.93 (0.64–1.36) rs3916839 A G 630 640 59 51 5 4 0.92 (0.65–1.30) 0.64 0.98 (0.66–1.46) 0.93 (0.44–1.98) 0.79 (0.53–1.17) 2.38 (0.84–6.75) rs50872 C T 396 366 248 256 44 43 1.07 (0.89–1.27) 0.48 1.12 (0.91–1.38) 1.08 (0.75–1.57) 1.01 (0.83–1.23) 1.33 (0.86–2.04) rs50871 T G 193 192 358 365 176 161 0.97 (0.84–1.13) 0.69 0.93 (0.78–1.11) 1.19 (0.88–1.61) 0.91 (0.77–1.08) 1.09 (0.78–1.52) rs3916823 AAAA Del. 513 514 182 181 21 13 0.92 (0.75–1.14) 0.45 0.95 (0.74–1.22) 0.85 (0.57–1.27) 0.93 (0.73–1.18) 0.78 (0.49–1.25) rs1799793 G A 318 291 313 350 90 75 1.03 (0.88–1.21) 0.68 0.99 (0.82–1.19) 1.07 (0.78–1.48) 1.04 (0.87–1.24) 0.99 (0.68–1.42) rs238406 G T 219 207 361 367 134 129 1.01 (0.87–1.18) 0.85 1.05 (0.87–1.26) 1.08 (0.79–1.49) 1.02 (0.86–1.21) 1.09 (0.75–1.58) ERCC3 (XPB) rs2276583 G A 279 277 332 306 91 98 0.99 (0.85–1.16) 0.94 1.02 (0.84–1.23) 0.89 (0.65–1.21) 0.93 (0.78–1.11) 1.25 (0.87–1.78) rs2134794 A C 436 411 228 238 37 32 1.05 (0.87–1.25) 0.63 1.05 (0.84–1.31) 1.04 (0.72–1.50) 1.08 (0.88–1.33) 0.90 (0.59–1.39) rs4150437 T C 616 606 72 88 3 5 1.20 (0.88–1.63) 0.26 1.37 (0.95–1.98) 0.89 (0.48–1.64) 1.03 (0.72–1.46) 1.93 (0.93–4.03) rs4150416 T G 296 304 320 293 86 84 0.96 (0.82–1.12) 0.58 0.92 (0.76–1.11) 1.05 (0.77–1.45) 1.01 (0.85–1.20) 0.83 (0.56–1.23) ERCC4 (XPF) rs3136064 C T 313 307 328 330 76 76 1.03 (0.88–1.21) 0.74 1.12 (0.92–1.36) 0.77 (0.56–1.06) 1.06 (0.89–1.27) 0.90 (0.61–1.33) rs1800067 G A 620 593 89 107 5 3 1.16 (0.87–1.55) 0.30 1.23 (0.87–1.75) 0.85 (0.51–1.43) 1.20 (0.87–1.64) 1.08 (0.52–2.25) rs3136215 T C 633 629 79 74 3 5 0.99 (0.73–1.35) 0.95 0.83 (0.57–1.23) 1.60 (0.82–3.12) 1.05 (0.74–1.48) 0.79 (0.37–1.70) rs1799801 T C 366 361 300 310 61 51 0.98 (0.83–1.15) 0.79 1.07 (0.88–1.31) 0.70 (0.51–0.98)c 0.98 (0.82–1.18) 1.00 (0.66–1.51)

aAmong all cases and controls; AA, homozygous major allele; Aa, heterozygous; aa, homozygous minor allele; numbers do not sum to total due to missing. b"No high-risk estrogen regimen use" was defined as no hormone use, or 6 months of unopposed estrogen, or estrogen plus progestogen <10 d/mo. "High-risk estrogen regimen use" was defined as greater than 6 months of use of unopposed estrogen, or estrogen plus progestogen <10 d/mo. cP < 0.05. 1877 1878

Table 3. Distribution of genotypes for SNPs in ERCC5, LIG1, XPA, and XPC in women with endometrial cancer and controls, and per-allele ORs Downloaded from acrEieilBoakr rv 09 etme 2011 September 20(9) Prev; Biomarkers Epidemiol Cancer by BMI and use of a high-risk estrogen regimen al. et Doherty

Genotype distributiona All BMI < BMI No use of Use of 30 kg/m2 30 kg/m2 high-risk high-risk estrogen estrogen regimenb regimenb

cebp.aacrjournals.org AA Aa aa (727 cases, 736 (422 cases, (303 cases, (551 cases, (168 cases,

controls) 601 controls) 135 controls) 610 controls) 114 controls) Published OnlineFirstJuly12,2011;DOI:10.1158/1055-9965.EPI-11-0119

Major allele Minor allele Controls Cases Controls Cases Controls Cases OR (95% CI) P OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI)

ERCC5 (XPG) rs2296147 T C 199 194 364 356 157 165 1.03 (0.89–1.20) 0.69 1.09 (0.91–1.31) 0.90 (0.67–1.20) 1.05 (0.89–1.24) 0.96 (0.68–1.34) rs4150261 G A 682 666 17 14 0 0 0.82 (0.40–1.69) 0.59 0.59 (0.22–1.56) 1.29 (0.32–5.21) 1.23 (0.55–2.76) 0.20 (0.02–1.82) rs4150276 A T 259 224 321 330 120 127 1.14 (0.98–1.32) 0.10 1.05 (0.87–1.26) 1.48 (1.09–2.01)c 1.17 (0.98–1.39) 1.00 (0.71–1.42) rs3818356 C T 451 407 222 243 29 29 1.18 (0.98–1.42) 0.09 1.11 (0.89–1.39) 1.64 (1.09–2.48)c 1.23 (0.99–1.52) 1.04 (0.68–1.58) on September 30, 2021. © 2011American Association for Cancer Research. rs4150351 A C 486 486 197 176 19 19 0.91 (0.74–1.12) 0.38 0.79 (0.61–1.02) 1.20 (0.80–1.81) 0.93 (0.74–1.17) 0.80 (0.50–1.28) rs4150355 C T 273 261 317 326 112 93 0.95 (0.81–1.11) 0.53 1.11 (0.92–1.33) 0.63 (0.46–0.87)d 0.92 (0.77–1.10) 1.13 (0.80–1.61) rs4150375 A G 600 575 118 138 8 9 1.14 (0.89–1.45) 0.31 1.19 (0.88–1.60) 0.96 (0.61–1.53) 1.10 (0.84–1.44) 1.41 (0.74–2.70) rs4150383 G A 504 460 179 198 19 22 1.19 (0.97–1.46) 0.09 1.16 (0.91–1.48) 1.47 (0.95–2.27) 1.23 (0.98–1.56) 1.10 (0.69–1.73) rs4150386 A C 480 516 177 124 7 6 0.68 (0.53–0.87) 0.002d 0.85 (0.64–1.14) 0.41 (0.25–0.68)d 0.70 (0.53–0.92)c 0.66 (0.36–1.20) rs17655 G C 408 418 248 254 47 42 1.03 (0.86–1.22) 0.78 0.97 (0.79–1.20) 1.21 (0.85–1.73) 1.05 (0.87–1.28) 0.92 (0.62–1.36) rs4150393 A G 552 543 145 128 5 10 0.95 (0.75–1.22) 0.70 0.84 (0.62–1.14) 1.16 (0.72–1.89) 1.02 (0.77–1.34) 0.69 (0.40–1.20) LIG1 rs3731043 T C 590 606 96 83 3 6 0.92 (0.69–1.23) 0.58 0.92 (0.65–1.30) 1.16 (0.62–2.15) 0.80 (0.58–1.12) 1.90 (0.87–4.12) rs2288881 G A 636 608 64 68 2 4 1.17 (0.84–1.64) 0.35 1.08 (0.71–1.65) 1.23 (0.64–2.35) 1.31 (0.89–1.92) 0.92 (0.43–1.97) rs20580 A C 161 163 365 341 173 176 1.03 (0.88–1.20) 0.70 1.05 (0.87–1.26) 0.86 (0.63–1.19) 1.03 (0.87–1.22) 0.96 (0.65–1.41) rs3730865 T C 652 608 46 70 3 3 1.46 (1.02–2.10) 0.04c 1.65 (1.09–2.50)c 1.18 (0.56–2.51) 1.27 (0.86–1.89) 2.98 (1.06–8.37)c rs20579 C T 534 513 150 153 18 15 1.02 (0.82–1.26) 0.88 0.94 (0.72–1.24) 1.07 (0.71–1.61) 1.15 (0.91–1.47) 0.70 (0.39–1.23) rs3730837 A G 581 574 136 127 4 6 0.95 (0.74–1.23) 0.72 0.96 (0.70–1.30) 0.89 (0.54–1.46) 0.91 (0.69–1.21) 1.36 (0.71–2.61) acrEieilg,Boakr Prevention & Biomarkers Epidemiology, Cancer XPA rs3176757 C T 468 468 207 190 27 23 0.91 (0.74–1.10) 0.32 0.91 (0.72–1.16) 1.01 (0.68–1.49) 0.94 (0.75–1.17) 0.71 (0.46–1.11) rs3176748 A G 333 313 290 305 79 63 0.98 (0.83–1.15) 0.79 0.90 (0.74–1.10) 1.17 (0.84–1.62) 0.90 (0.75–1.08) 1.53 (1.02–2.30)c rs2808667 C T 620 641 100 76 5 0 0.71 (0.52–0.97) 0.03c 0.77 (0.54–1.12) 0.57 (0.30–1.06) 0.81 (0.57–1.16) 0.39 (0.20–0.76)d rs3176683 T C 618 608 83 72 1 1 0.85 (0.61–1.19) 0.35 0.92 (0.62–1.34) 0.99 (0.47–2.11) 0.95 (0.65–1.39) 0.50 (0.24–1.03) rs3176658 C T 515 496 142 142 19 16 1.02 (0.82–1.27) 0.85 1.18 (0.91–1.53) 0.70 (0.45–1.10) 1.20 (0.93–1.54) 0.48 (0.29–0.81)d rs3176649 GCAC Del. 619 622 87 80 4 3 0.96 (0.71–1.30) 0.79 1.04 (0.72–1.49) 0.78 (0.42–1.43) 0.90 (0.63–1.28) 1.25 (0.65–2.41) rs3176646 C T 672 664 51 55 1 1 1.15 (0.78–1.69) 0.48 1.27 (0.82–1.98) 1.12 (0.46–2.70) 0.95 (0.60–1.50) 1.82 (0.78–4.27) rs1800975 G A 328 339 320 297 66 67 0.95 (0.81–1.12) 0.54 1.04 (0.85–1.26) 0.85 (0.62–1.17) 1.07 (0.89–1.29) 0.50 (0.34–0.74)d XPC rs1126547 C G 531 497 159 171 10 12 1.13 (0.90–1.41) 0.29 1.16 (0.88–1.51) 1.15 (0.72–1.85) 1.11 (0.86–1.43) 1.18 (0.71–1.96) rs2228001 A C 263 258 342 327 109 118 1.01 (0.87–1.17) 0.92 0.90 (0.75–1.08) 1.20 (0.87–1.64) 1.02 (0.86–1.21) 0.97 (0.68–1.38) rs3731151 G A 401 399 260 249 45 60 1.09 (0.92–1.29) 0.33 1.20 (0.98–1.47) 0.90 (0.65–1.25) 1.00 (0.82–1.21) 1.52 (1.02–2.26)c rs2228000 C T 384 411 278 257 61 49 0.88 (0.75–1.04) 0.15 0.90 (0.74–1.10) 0.87 (0.62–1.22) 0.95 (0.79–1.15) 0.64 (0.44–0.93)c rs3731127 C T 637 586 60 90 3 3 1.57 (1.13–2.17) 0.007d 1.55 (1.07–2.25)c 2.07 (0.94–4.55) 1.54 (1.07–2.20)c 2.54 (1.04–6.22)c rs2733537 A G 296 302 319 300 86 78 0.96 (0.82–1.12) 0.60 0.97 (0.80–1.18) 0.96 (0.69–1.34) 1.04 (0.87–1.25) 0.69 (0.48–0.99)c rs2607755 T C 179 177 353 341 170 163 1.00 (0.86–1.17) 0.97 1.08 (0.90–1.29) 0.90 (0.66–1.23) 0.98 (0.82–1.16) 1.08 (0.76–1.55)

Nucleotide Excision Repair Genes and Endometrial Cancer

Table 4. ERCC5, LIG1, XPA, and XPC haplotypes and endometrial cancer risk

Haplotypea,b Control freq. Case freq. ORc (95% CI) P

ERCC5 (rs2296147, rs4150261, rs4150276, rs3818356, rs4150351, rs4150355, rs4150375, rs4150383, rs4150386, rs17655, rs4150393) 10000100000 0.24 0.25 1.00 (Ref.) 00110001000 0.13 0.15 1.09 (0.84–1.43) 0.52 10000100100 0.12 0.09 0.75 (0.53–1.05) 0.09 00100000010 0.11 0.13 1.11 (0.83–1.47) 0.49 00100010010 0.06 0.05 0.83 (0.59–1.18) 0.31 10001000001 0.06 0.05 0.92 (0.63–1.35) 0.67 00001000000 0.05 0.05 0.81 (0.55–1.19) 0.27 00110000000 0.04 0.03 0.82 (0.52–1.30) 0.40 00001000001 0.04 0.03 0.79 (0.49–1.25) 0.31 00000000010 0.03 0.03 0.97 (0.60–1.56) 0.88 10110001000 0.02 0.02 1.08 (0.61–1.91) 0.79 00000000100 0.02 0.01 0.46 (0.22–0.97) 0.04 00001010001 0.01 0.02 1.41 (0.73–2.72) 0.31 01100000010 0.01 0.01 0.76 (0.34–1.67) 0.49 10110000000 0.01 0.01 1.22 (0.57–2.64) 0.61 10000110000 0.01 0.02 2.08 (1.01–4.29) 0.05 10100000010 0.01 0.01 0.78 (0.30–2.07) 0.62 00000100000 0.01 0.01 0.64 (0.21–1.94) 0.43 LIG1 (rs3731043, rs2288881, rs20580, rs3730865, rs20579, rs3730837) 000000 0.41 0.39 1.00 (Ref.) 001000 0.22 0.23 1.08 (0.88–1.33) 0.48 001010 0.11 0.11 1.07 (0.83–1.38) 0.61 001001 0.09 0.08 0.95 (0.70–1.28) 0.72 101000 0.07 0.07 0.98 (0.72–1.33) 0.88 000100 0.04 0.05 1.46 (0.99–2.15) 0.06 010000 0.02 0.03 1.20 (0.71–2.02) 0.50 010010 0.02 0.02 1.09 (0.63–1.88) 0.77 000001 0.01 0.02 2.00 (0.89–4.50) 0.10 XPA (rs3176757, rs3176748, rs2808667, rs3176683, rs3176658, rs3176649, rs3176646, rs1800975) 00000000 0.36 0.38 1.00 (Ref.) 01000000 0.25 0.25 0.95 (0.78–1.15) 0.59 10000001 0.08 0.08 0.90 (0.68–1.19) 0.46 00101001 0.07 0.05 0.70 (0.51–0.96) 0.03 01000100 0.07 0.06 0.94 (0.68–1.29) 0.69 10010001 0.06 0.05 0.84 (0.60–1.19) 0.33 00001001 0.06 0.08 1.26 (0.92–1.73) 0.15 10000011 0.04 0.04 1.09 (0.74–1.62) 0.67 XPC (rs1126547, rs2228001, rs3731151, rs2228000, rs3731127, rs2733537, rs2607755) 0100000 0.26 0.25 1.00 (Ref.) 0010001 0.25 0.26 1.10 (0.89–1.35) 0.39 0001011 0.24 0.22 0.99 (0.80–1.23) 0.94 1100000 0.13 0.14 1.13 (0.87–1.46) 0.35 0000110 0.05 0.07 1.48 (1.04–2.11) 0.03 0001010 0.04 0.03 0.71 (0.45–1.11) 0.13 0000010 0.03 0.01 0.58 (0.32–1.04) 0.07 0000000 0.01 0.01 1.21 (0.39–3.73) 0.75

a"1" indicates presence of the minor allele and "0" indicates presence of the major allele for each of the SNPs in order of their listing for each gene. bThe most common haplotype among controls is assigned as referent. cAdjusted for age, county of residence, and reference year.

www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 20(9) September 2011 1879

Doherty et al.

Table 5. Combined ERCC5, LIG1, and XPC genotypes and endometrial cancer risk by number of "risk" genotypes carried

Number of "risk" ERCC5 rs4150386 LIG1 rs3730865 XPC rs3731127 Cases Controls OR (95% CI) P genotypes carrieda

None 0b 0 0 103 161 1.00 (Ref.) 1c 418 421 1.60 (1.20–2.13) 0.001 1 0 0 394 401 1.57 (1.18–2.10) 0.002 0 1 0 15 10 2.61 (1.12–6.08) 0.03 0 0 1 9 10 1.62 (0.63–4.18) 0.32 2c 115 77 2.35 (1.59–3.46) 0.000 1 1 0 44 31 2.13 (1.25–3.62) 0.005 1 0 1 68 44 2.53 (1.60–4.03) 0.000 0 1 1 3 2 1.98 (0.31–12.5) 0.47 3c 1 1 1 8 2 6.54 (1.35–31.8) 0.02

a"Risk" genotypes are defined as: ERCC5 rs4150386 AA (vs. AC/CC); LIG1 rs3730865 CT/CC (vs. TT); XPC rs3731127 CT/TT (vs. CC). bFor each SNP, "1" indicates presence and "0" indicates absence of the "risk" genotype(s). cSummary categories for carriage of any 1, any 2, or all 3 risk genotypes, respectively.

analysis. In the combined genotype analysis, carriage of SNP were not noteworthy at a reasonable prior. There- increasing numbers of "risk" genotypes was associated fore, the results for the ERCC5 and XPC SNPs may be with increasing risk. Compared with carrying no "risk" slightly less likely that the others to be false positives. genotypes, ORs and 95% CIs for 1, 2, and all 3 "risk" DNA repair pathways are complicated, with many genotypes were respectively: 1.60 (1.20–2.13), 2.35 (1.59– redundancies and interactions between proteins. It is 3.46), and 6.54 (1.35–31.81; Table 5). biologically plausible that in a system with multiple redundancies, cumulative perturbations in component Discussion proteins might be necessary before a difference in risk is observed. It is therefore of particular interest to exam- We observed associations between some polymorph- ine combined genotypes. Because the SNPs that were isms in XPC, XPA, ERCC5, and LIG1 in the NER pathway associated with endometrial cancer risk in our study have and endometrial cancer risk. We previously reported that low minor allele frequencies, we combined heterozygous women who carried the minor alleles in XPC, both and homozygous minor allele genotype categories, and it rs2228000 and rs2228001, or XPA rs1800975 had a is possible that this model is not correct for all of the decreased risk of endometrial cancer (10). In our larger, SNPs. Nonetheless, we observed increasing risk asso- combined study population, we observed a borderline ciated with increasing numbers of "risk" genotypes car- decreased risk associated with carrying the XPC ried in XPC, ERCC5, and LIG1. Only 8 cases and 2 controls rs2228000 C allele and no association with the other 2 carried all 3 "risk" genotypes. While we observed a SNPs. The SNPs that were associated with risk in our particularly high risk of endometrial cancer among present study (XPC rs3731127 and XPA rs2808667) were women in this subgroup, we view this observation as not in LD with XPA rs1800975 or XPC rs2228000. The 4 no more than suggestive and it requires further explora- SNPs that were associated with endometrial cancer risk in tion in other studies. our study are intronic. On the basis of FastSNP bioinfor- With the exception of a single study that, like us, matics data (24), ERCC5 rs4150386 and LIG1 rs3730865 observed no association between ERCC1 rs11615 and may be intronic enhancers but XPA rs2808667 and XPC endometrial cancer risk (25), to our knowledge, there rs3731127 have no known function. are no other studies of NER genotypes and endometrial To evaluate the degree to which these findings were cancer risk to date. Many of the genes and SNPs included robust, we conducted gene set analyses for each gene and in our study have been examined in studies of cancers of calculated FPRP (23) for each of the 4 SNPs. The gene set the lung, head and neck, prostate, bladder, colon/rectum, analysis P values were lower than 0.05 only for ERCC5 breast, and esophagus, as well as glioma and melanoma and XPC, which lends support to the observed associa- (comprehensive reviews of most cancer types include tions with SNPs in those genes. With respect to FPRP, Goode and colleagues, ref. 26; Neumann and colleagues, associations for the ERCC5 and XPC SNPs were note- ref. 27; and Vineis and colleagues, ref. 28). While a worthy with a prior of 0.01 (defined as low) and above, recessive association between ERCC2 (XPD) rs13181 and the LIG1 SNP was noteworthy with a prior of 0.1 and lung cancer risk appears to be validated (28), no (defined as moderate) and above. The results for the XPA clear picture has emerged with respect to patterns of

1880 Cancer Epidemiol Biomarkers Prev; 20(9) September 2011 Cancer Epidemiology, Biomarkers & Prevention

Nucleotide Excision Repair Genes and Endometrial Cancer

associations with NER genes and SNPs across cancer number of possible associations examined in the NER types. pathways, there is the potential for false-positive find- Our study has several limitations. We attempted to ings, both overall and in subgroups. Broader exploration characterize the NER genes (while also including "func- of SNP associations in NER genes using data from exist- tional" SNPs), but we still had limited SNP coverage for ing (29) and pending genome-wide association studies some of the genes because we used an r2 value of 0.64 to may help to clarify whether genetic variation in NER is select tagSNPs. For example, if we had used an r2 cutoff associated with endometrial cancer risk. Conclusions point of 0.8 to select tagSNPs, we would have needed an about the influence of these genotypes on risk of endo- additional 10 SNPs in ERCC4,9inERCC1,7inERCC5, metrial cancer must await the results of additional studies and 5 in ERCC2, but for XPC and LIG1, only a single and pooled analyses. additional SNP would have been needed and for XPA, the coverage would have been identical. We selected our Disclosure of Potential Conflicts of Interest SNPs using data from sequenced individuals from the Environmental Genome Project (EGP) prior to the avail- No potential conflicts of interest were disclosed. ability of HapMap data. Because many of the EGP SNPs Acknowledgments were not genotyped in HapMap, it is difficult to compare SNP coverage with HapMap data. Also, while we were The authors thank Dr. Kathleen Malone (Fred Hutchinson Cancer not able to obtain a DNA sample from all women who Research Center) for facilitating the use of the CARE data (N01 HD 2 were interviewed for our studies, the distributions of 3166). They also thank the participants in our series of endometrial cancer studies, as well as the participants in the National Institute of Child Health characteristics (e.g., age, parity, oral contraceptive use, and Development CARE study. smoking, and hormone use) are similar between women who did and did not provide a blood sample. Because of Grant Support the small number of Hispanic and nonwhite women in the study (56 cases and 59 controls), we were not able to This work was supported by grants R01 CA 105212, R01 CA 87538, R01 CA 75977, R03 CA 80636, N01 HD 2 3166, R35 CA 39779, K05 CA 92002, and funds assess their risk associated with the genotypes. Finally, from the Fred Hutchinson Cancer Research Center. small numbers of women had ever taken a high-risk The costs of publication of this article were defrayed in part by the payment estrogen regimen, limiting our ability to see differences of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. by this exposure. While we observed suggestive associations between Received February 1, 2011; revised June 3, 2011; accepted July 1, 2011; SNPs in XPC, XPA, ERCC5, and LIG1, because of the large published OnlineFirst July 12, 2011.

References 1. Key TJ, Pike MC. The dose-effect relationship between ‘unop- 12. Hankey BF, Ries LA, Edwards BK. The surveillance, epidemiology, posed’ oestrogens and endometrial mitotic rate: its central role and end results program: a national resource. Cancer Epidemiol in explaining and predicting endometrial cancer risk. Br J Cancer Biomarkers Prev 1999;8:1117–21. 1988;57:205–12. 13. Waksberg J. Random digit dialing sampling for case-control studies. 2. Siiteri PK. Steroid hormones and endometrial cancer. Cancer Res In: Gail MH, Benichou J, editors. Encyclopedia of epidemiologic 1978;38:4360–6. methods. New York: John Wiley & Sons; 2000. p. 749–53. 3. Zhu BT, Conney AH. Functional role of estrogen metabolism in target 14. Marchbanks PA, McDonald JA, Wilson HG, Burnett NM, Daling JR, cells: review and perspectives. Carcinogenesis 1998;19:1–27. Bernstein L, et al. The NICHD Women's Contraceptive and Repro- 4. Roy D, Liehr JG. Estrogen, DNA damage and mutations. Mut Res ductive Experiences Study: methods and operational results. Ann 1999;424:107–15. Epidemiol 2002;12:213–21. 5. Cavalieri E, Frenkel K, Liehr JG, Rogan E, Roy D. Estrogens as 15. Bodelon C, Doherty JA, Chen C, Rossing MA, Weiss NS. Use of endogenous genotoxic agents–DNA adducts and mutations. J Natl nonsteroidal antiinflammatory drugs and risk of endometrial cancer. Cancer Inst Monogr 2000;75–93. Am J Epidemiol 2009;170:1512–7. 6. Yager JD, Liehr JG. Molecular mechanisms of estrogen carcinogen- 16. Rossing MA, Cushing-Haugen KL, Wicklund KG, Doherty JA, Weiss esis. Annu Rev Pharmacol Toxicol 1996;36:203–32. NS. Menopausal hormone therapy and risk of epithelial ovarian 7. Wogan GN, Hecht SS, Felton JS, Conney AH, Loeb LA. Environmental cancer. Cancer Epidemiol Biomarkers Prev 2007;16:2548–56. and chemical carcinogenesis. Semin Cancer Biol 2004;14:473–86. 17. NIEHS SNPs. NIEHS Environmental Genome Project, University of 8. Friedberg EC. How nucleotide excision repair protects against cancer. Washington. 2001. Available from: http://egp.gs.washington.edu/wel- Nat Rev Cancer 2001;1:22–33. come.html. 9. Volker M, Mone MJ, Karmakar P, van Hoffen A, Schul W, Vermeulen 18. Plink (v1.07). Center for Human Genetic Research, Massachusetts W, et al. Sequential assembly of the nucleotide excision repair factors General Hospital, and Broad Institute. 2009. Available from: http:// in vivo. Mol Cell 2001;8:213–24. pngu.mgh.harvard.edu/purcell/plink/. 10. Weiss JM, Weiss NS, Ulrich CM, Doherty JA, Voigt LF, Chen C. 19. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, Interindividual variation in nucleotide excision repair genes and risk et al. PLINK: a tool set for whole-genome association and population- of endometrial cancer. Cancer Epidemiol Biomarkers Prev based linkage analyses. Am J Hum Genet 2007;81:559–75. 2005;14:2524–30. 20. Li SS, Khalid N, Carlson C, Zhao LP. Estimating haplotype frequencies 11. Doherty JA, Weiss NS, Freeman RJ, Dightman DA, Thornton PJ, and standard errors for multiple single nucleotide polymorphisms. Houck JR, et al. Genetic factors in catechol estrogen metabolism Biostatistics 2003;4:513–22. in relation to the risk of endometrial cancer. Cancer Epidemiol Bio- 21. Zhao LP, Li S, Khalid N. A method for the assessment of disease markers Prev 2005;14:357–66. associations with single nucleotide polymorphism haplotypes and

www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 20(9) September 2011 1881

Doherty et al.

environmental variables in case-control studies. Am J Hum Genet ovarian and endometrial cancer in Korean women. A case control 2003;72:1231–50. study. Gynecol Obstet Invest 2007;64:84–8. 22. fhcrc.org. Creative Development Services. 2011. Available from: 26. Goode EL, Ulrich CM, Potter JD. Polymorphisms in DNA repair genes http://cdsweb01.fhcrc.org/HPlus/. and associations with cancer risk. Cancer Epidemiol Biomarkers Prev 23. Wacholder S, Chanock S, Garcia-Closas M, El Ghormli L, Rothman N. 2002;11:1513–30. Assessing the probability that a positive report is false: an approach 27. Neumann AS, Sturgis EM, Wei Q. Nucleotide excision repair as a for molecular epidemiology studies. J Natl Cancer Inst 2004;96:434– marker for susceptibility to tobacco-related cancers: a review of 42. molecular epidemiological studies. Mol Carcinog 2005;42:65–92. 24. Yuan H-Y, Chiou J-J, Tsang W-H, Liu C-H, Liu C-K, Lin Y-J, et al. 28. Vineis P, Manuguerra M, Kavvoura FK, Guarrera S, Allione A, Rosa F, FASTSNP: an always up-to-date and extendable service for et al. A field synopsis on low-penetrance variants in DNA repair genes SNP function analysis and prioritization. Nucleic Acids Res and cancer susceptibility. J Natl Cancer Inst 2009;101:24–36. 2006;34:W635–41. Available from: http://fastsnp.ibms.sinica. 29. Spurdle AB, Thompson DJ, Ahmed S, Ferguson K, Healey CS, edu.tw. O'Mara T, et al. Genome-wide association study identifies a com- 25. Jo H, Kang S, Kim SI, Kim JW, Park NH, Song YS, et al. The C19007T mon variant associated with risk of endometrial cancer. Nat Genet polymorphism of ERCC1 and its correlation with the risk of epithelial 2011;43:451–4.

1882 Cancer Epidemiol Biomarkers Prev; 20(9) September 2011 Cancer Epidemiology, Biomarkers & Prevention

Polymorphisms in Nucleotide Excision Repair Genes and Endometrial Cancer Risk

Jennifer A. Doherty, Noel S. Weiss, Sherianne Fish, et al.

Cancer Epidemiol Biomarkers Prev 2011;20:1873-1882. Published OnlineFirst July 12, 2011.

Updated version Access the most recent version of this article at: doi:10.1158/1055-9965.EPI-11-0119

Supplementary Access the most recent supplemental material at: Material http://cebp.aacrjournals.org/content/suppl/2011/07/12/1055-9965.EPI-11-0119.DC1

Cited articles This article cites 25 articles, 6 of which you can access for free at: http://cebp.aacrjournals.org/content/20/9/1873.full#ref-list-1

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department Subscriptions at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cebp.aacrjournals.org/content/20/9/1873. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.