Skip Navigation


Human Molecular Genetics Advance Access originally published online on December 12, 2006
Human Molecular Genetics 2007 16(2):154-164; doi:10.1093/hmg/ddl451
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrowOA All Versions of this Article:
16/2/154    most recent
ddl451v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (5)
Google Scholar
Right arrow Articles by Einarsdóttir, K.
Right arrow Articles by Wedrén, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Einarsdóttir, K.
Right arrow Articles by Wedrén, S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2006 The Author(s)
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Effect of ATM, CHEK2 and ERBB2 TAGSNPs and haplotypes on endometrial cancer risk

Kristjana Einarsdóttir1,*, Keith Humphreys1, Carine Bonnard2, Yuqing Li2, Yi Li3, Kee Seng Chia4, Edison T. Liu2, Per Hall1, Jianjun Liu2 and Sara Wedrén1

1 Department of Medical Epidemiology and Biostatistics, Karolinska Institute, 171 77 Stockholm, Sweden, 2 Population Genetics, 3 Information and Mathematical Sciences, Genome Institute of Singapore, Singapore 138672, Singapore and 4 Center for Molecular Epidemiology, Department of Community, Occupational and Family Medicine, National University of Singapore, Singapore 117597, Singapore

* To whom corresponding should be addressed at: Department of Medical Epidemiology and Biostatistics, Karolinska Institute, Box 281, Stockholm, Sweden. Tel: +46 8 524 86100; Fax: +46 8 314975; Email: Kristjana.Einarsdottir{at}ki.se

Received October 12, 2006; Accepted November 24, 2006


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Family history of endometrial cancer increases the risk of developing the disease, but it is still largely unknown which germ-line genetic factors are involved in the aetiology of endometrial cancer. In a Swedish population-based case–control study including 705 cases and 1565 controls, we examined common variation in the ATM, CHEK2 and ERBB2 genes in relation to endometrial cancer risk overall, restricted to tumours of certain characteristics or stratified by various endometrial cancer risk factors. We genotyped a large number of single-nucleotide polymorphisms (SNPs) in the genes and selected seven haplotype-tagging SNPs (tagSNPs) in ATM, six tagSNPs in CHEK2 and seven tagSNPs in ERBB2 that could predict common variants and haplotypes (frequency ≥0.03) in each gene with R2 ≥ 0.8. We included the tagSNPs or their haplotypes as explanatory variables in unconditional logistic regression models adjusted for age. Our results indicated an increased risk of developing endometroid endometrial cancer for homozygous carriers of the rare allele (AA) of a tagSNP (rs4987886) in CHEK2 (P = 0.005) when contrasted with GG carriers. We also found a decreased endometrial cancer risk among non-smoking carriers of a haplotype in ATM (P = 0.0007) and among carriers of a haplotype in CHEK2, who had experienced menopause below 49 years of age (P = 0.0009) compared with non-carriers of these haplotypes. We found no effect of genetic variation in ERBB2 on endometrial cancer risk. In conclusion, it is possible that common variants in the ATM and CHEK2 genes, in interaction with oestrogen-related exposures, are involved in endometrial cancer aetiology.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Endometrial cancer is the most common gynaecological cancer in the industrialized world. History of a first-degree relative with endometrial cancer has been related to a 2-fold increase in endometrial cancer risk (1), but as of yet, the germ-line genetic factors involved in the development of the disease are largely unknown. The main risk factor for endometrial cancer is unopposed exogenous or endogenous oestrogen exposure. Oestrogen metabolites have been reported to cause a number of DNA lesions (2), among which are double-strand DNA breaks (3). DNA double-strand breaks appear to be the predominant signal for the activation of pathways mediated by the ATM (ataxia-telangiectasia mutated) protein (4). Once activated, the ATM protein triggers phosphorylation of CHEK2 (checkpoint kinase 2) which in turn phosphorylates p53, Cdc25 and BRCA1, thereby promoting cell-cycle arrest and activation of DNA repair (5–11). Defects in the ATM (MIM 607585 [OMIM] ) and CHEK2 (MIM 604373 [OMIM] ) genes could thus be involved in endometrial cancer development via their role in DNA damage checkpoint regulation, especially in combination with increased oestrogen exposure. Variation in the ERBB2 (also named HER2, MIM 164870 [OMIM] ) gene might also be important in endometrial cancer aetiology, as the gene is often over-expressed and/or amplified in endometrial carcinomas (12–14), a phenomenon related to prognosis and survival of the disease (13–15). ERBB2 is a transmembrane glycoprotein with tyrosine kinase activity (16–20), which, if activated, results in a variety of cellular responses, including proliferation, cell differentiation, cell motility and survival (21–23).

Despite a plausible role for the ATM, CHEK2 and ERBB2 genes in the development of endometrial cancer, it has not—to our knowledge—been previously examined whether common variation in these genes is involved in endometrial cancer aetiology. We aimed to answer that question by genotyping a large number of single-nucleotide polymorphisms (SNPs) in the ATM, CHEK2 and ERBB2 genes in order to select haplotype-tagging SNPs (tagSNPs) that could predict at least 80% of the entire common variation in the genes. We then assessed the association between these SNPs or their haplotypes and endometrial cancer risk, overall and by tumour grade, in a large population-based study of postmenopausal Swedish women. Additionally, we explored effect modification by various endometrial cancer risk factors.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Study population
Participants in the present genetic study were selected from a large, population-based Swedish case–control study. Table 1 shows selected characteristics of the participants in the present genetic study compared with the parent questionnaire study. Proportions were similar between the two studies and case–control-differences reflected established associations. Endometroid tumours constituted 93% of all endometrial cancers in the present study and could be divided into Grade I tumours (39%), Grade II tumours (44%) and Grade III tumours (17%). In 29% of the endometrial cancers in the present study with information on myometrial invasion, the cancer had invaded at least 50% of the myometrium or the serosa.


View this table:
[in this window]
[in a new window]

  		Table 1. Selected characteristics of the endometrial cancer cases and controls in the present study compared with the parent study

 
Cases who participated via tissue sample donation were on average 2.1 years older than cases who participated by donating a blood sample (P = 0.002) and were more likely, though not significantly, to have poorly differentiated (Grade 3) tumours (P = 0.08). Importantly, tagSNP genotype frequencies were not different between cases who participated by donating a tissue sample and cases who donated a blood sample.

LD pattern and coverage estimation
Summary statistics on genotyping results and SNP coverage in the ATM, CHEK2 and ERBB2 genes are shown in Table 2. The SNPs successfully genotyped in 92 randomly selected controls are listed in Supplementary Material [Table S1 (ATM) and Table S2 (ERBB2)] and have been previously published by Einarsdóttir et al. (24). (CHEK2).


View this table:
[in this window]
[in a new window]

  		Table 2. Summary statistics on genotyping results and SNP coverage in ATM, CHEK2 and ERBB2

 
We included in our study 51 SNPs in ATM, 14 SNPs in CHEK2 and 13 SNPs in ERBB2 that were successfully genotyped in the 92 controls and found to be in Hardy–Weinberg equilibrium (HWE) (Table 2). Mean spacing between included SNPs was 2.9, 4.0 and 2.8  kb in ATM, CHEK2 and ERBB2, respectively (Table 2). We detected strong LD across all three genes [Supplementary Material, Figs S1 and S2 and Fig. 1 in Einarsdóttir et al. (24)] and by using the SNP dropping method (25), we found that the tagSNPs selected from the included SNPs could capture non-genotyped SNPs as efficiently as the included SNPs (Table 2).

TagSNP and haplotype analyses
From the included SNPs, we selected seven tagSNPs in ATM, six tagSNPs in CHEK2 and seven tagSNPs in ERBB2 that could predict the included SNPs and their haplotypes with an R2 of at least 0.8. The tagSNPs were genotyped in all cases and controls (Table 3), but five tagSNPs in ATM could not be genotyped in the cases who participated via tissue sample donation. All tagSNPs were in HWE among controls and none was associated with any of the endometrial cancer risk factors given in Table 1. Only one of the tagSNPs—TAG5 in ERBB2, also named I655V—conferred an amino acid change in the protein product.


View this table:
[in this window]
[in a new window]

  		Table 3. Characteristics of the tagSNPs genotyped in ATM, CHEK2 and ERBB2 and their association with endometrial cancer risk

 
When assessing the change in risk with each addition of the rare allele compared with non-carriers, we found TAG1 in CHEK2 to be associated with increased endometrial cancer risk (P = 0.01, Table 3), but multiple testing adjustment rendered the association non-significant (P = 0.23). Restricting the analysis to include only endometroid tumours yielded a stronger association with an odds ratio (OR) of 1.28 (95% CI 1.07–1.54, P = 0.007). This association disappeared when we restricted the analyses to the few non-endometroid tumours (OR 1.02, 95% CI 0.56–1.86). When we explored individual genotype risks for CHEK2 TAG1, the increased risk for developing endometrial cancers in general or only endometroid tumours appeared to be confined to homozygous carriers of the rare allele (AA) (Table 4). Compared with GG carriers, the risk in AA carriers was 2.11 (P = 0.012) for all tumours and 2.29 (P = 0.005) for the endometroid tumours (Table 4). Conditioning on the selection variables (menopausal hormone use and diabetes mellitus) or restricting the analyses to the randomly selected controls did not alter the results.


View this table:
[in this window]
[in a new window]

  		Table 4. Association of TAG1 in CHEK2 with endometrial cancer risk overall or restricted to endometroid tumours

 
The associations of the tagSNP haplotypes in ATM, CHEK2 and ERBB2 in relation to endometrial cancer risk are given in Table 5. The logistic regression models included the common haplotypes and the combined group of rare haplotypes for each gene, with the most common haplotype as reference. Haplotype 4 in both ATM and CHEK2 showed associations with endometrial cancer risk (P = 0.028 and P = 0.017, respectively) which did not, however, carry over to the global tests. These results were unaffected after conditioning on the selection variables (menopausal hormone use and diabetes mellitus) or restricting the analyses to the randomly selected controls.


View this table:
[in this window]
[in a new window]

  		Table 5. Common tagSNP haplotypes in ATM, CHEK2 and ERBB2 in relation to endometrial cancer risk

 
Table 6 shows global P-values for association between ATM, CHEK2 and ERBB2 haplotypes and endometrial cancer risk, restricted to certain tumour subtypes or stratified by endometrial cancer risk factors. We did not perform tests within the subgroups of medium potency oestrogen only (or in combination with progestin) since the low numbers might have affected the reliability of the global P-values. ATM haplotypes appeared to affect endometrial cancer risk among women who delivered their last child over 33 years of age (global P = 0.027, Table 6). Haplotype 5 showed a borderline significant association (P = 0.053) compared with haplotype 1 (Supplementary Material, Table S3) in this group of women, but the likelihood ratio test for interaction between age at last birth and haplotype 5 in ATM was not statistically significant (P = 0.08). A stronger association emerged between endometrial cancer risk and ATM haplotypes in non-smokers (global P = 0.009), but became non-significant after multiple testing adjustment (P = 0.32). This association was driven by haplotype 4 in ATM (P = 0.002), which decreased the risk of endometrial cancer (OR 0.50, 95% CI 0.32–0.77) compared with haplotype 1 (Supplementary Material, Table S3). When we compared carriers of haplotype 4 with non-carriers, the association was slightly stronger (OR 0.48, 95% CI 0.31–0.73, P = 0.0007), and the test of interaction indicated that the effect of haplotype 4 in ATM on endometrial cancer risk depended on smoking status (P = 0.0037).


View this table:
[in this window]
[in a new window]

  		Table 6. Global P-values for the association of ATM, CHEK2 and ERBB2 tagSNP haplotypes with endometrial cancer risk restricted to tumour subtypes and stratified by endometrial cancer risk factors

 
CHEK2 haplotypes were associated with endometrial cancer risk among women who had experienced menopause below 49 years of age (global P = 0.034, Table 6). In this group of women, all haplotypes in CHEK2 appeared to increase endometrial cancer risk when compared with haplotype 1 (Supplementary Material, Table S3). However, when we compared each haplotype with non-carriers of the respective haplotype, only haplotype 1 affected endometrial cancer risk among these women (OR 0.50, 95% CI 0.33–0.75, P = 0.0009). The risk related to haplotype 1 increased with increasing age at menopause (49–52 years: OR = 0.88, P = 0.30; >52 years: OR = 1.17, P = 0.31) and the test for interaction between age at menopause and haplotype 1 in CHEK2 was statistically significant (P = 0.007).

Rare variants
We also genotyped in our sample set the non-synonymous variants rs1800056 (2572 T -> C, F858L) and rs1800058 (4258 C -> T, L1420F) in the ATM gene and the 1100delC deletion in the CHEK2 gene. All variants were very rare in our population with minor allele frequencies of 1.6, 1.7 and 0.4% among the controls, respectively. We found no association between any of the three variants and endometrial cancer risk.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
We carefully assessed the association of common variation in the ATM, CHEK2 and ERBB2 genes with risk of endometrial cancer overall, restricted to certain tumour subtypes and stratified by endometrial cancer risk factors. We found homozygous carriers of the rare allele of TAG1 in CHEK2 to have increased risk of developing endometroid tumours. We also found decreased endometrial cancer risk among non-smoking carriers of haplotype 4 in ATM and among carriers of haplotype 1 in CHEK2 who had experienced menopause at an early age. However, these associations were no longer statistically significant following multiple testing adjustment. Hence, despite biological plausibility, we cannot exclude chance as an explanation for the findings.

This is a large and well-designed, population-based case–control study. Cases were ascertained from the nation-wide Swedish Cancer Registries with practically complete data on incident cancers (26). Furthermore, the same pathologist assessed histological specimens for all cases. Thus, it is doubtful that our findings can be attributed to misclassification of the outcome. Differential misclassification of the exposure is also unlikely to have accounted for our results. Genotyping was performed with reliable genotyping methods, DNA samples were randomly assigned to the genotyping plates and our genotyping personnel were blinded to case–control status. We also replicated genotype calls for a subset of samples, using a separate genotyping method with over 99.5% genotype concordance. Despite testing SNPs indirectly, we suffered minimal loss of power in our study. We were able to predict unobserved SNPs in the genes with average R2 of 0.92 for ATM, 0.93 for CHEK2 and 0.72 for ERBB2. For the ability of haplotypes to predict the allele count at a causal locus with minor allele frequency of 0.25—assuming {alpha} = 0.05—we had 88% power for ATM, 88% power for CHEK2 and 72% power for ERBB2 to detect an OR of 1.35.

Participation in the current genetic study was related to severe disease or death. We therefore sought to obtain tissue samples from deceased cases and from those who refused blood donation. We were able to acquire the majority of the requested tissue samples, but lack of tissue accessibility depended on the inability of the respective pathology department to retrieve the samples. As genotyping frequencies of the tagSNPs did not differ between tissue and blood samples, we believe that problems associated with this non-participation in our study are negligible.

At the initiation of this study, SNP data from the International Hapmap Project (27,28) were still sparse, so we decided to select SNPs from publicly available databases and characterize LD as well as choose tagSNPs, using our own study population. For the three genes included in our study, we genotyped 45 common SNPs (minor allele frequency ≥0.03) that were also genotyped in the European-American Centre d'Etude du Polymorphisme Humain (CEU) Hapmap population. For these SNPs, genotype frequencies were very similar between the CEU population and our Swedish population (R2 = 0.87), and the LD patterns we observed for the three genes resembled the Hapmap LD patterns. For each of the three genes, at least two of our tagSNPs were not genotyped in Hapmap.

We found homozygous carriers of the minor allele of the common tagSNP TAG1 in CHEK2 to be at increased risk of endometrial cancer. The rare allele of TAG1 was the only rare allele carried by haplotype 4 in CHEK2 and we consequently found an increased risk for haplotype 4 carriers in CHEK2. The effect of TAG1 in CHEK2 on endometrial cancer risk was stronger among endometroid tumours. Endometrial cancers can be divided into Type I endometroid tumours and Type II non-endometroid tumours (29–31), where endometroid tumours constitute the majority of endometrial cancers. The endometroid tumours appear to be the tumours that are mainly caused by oestrogen exposure (29–31). Oestrogen metabolites have been reported to cause a number of DNA lesions both directly and indirectly through redox cycling processes (2). Indirect damage includes single-strand DNA breaks, 8-hydroxylation of guanine bases and DNA adducts (2), whereas direct DNA damage caused by covalent binding of quinone intermediates of 4-hydroxyestrogens to DNA can result in the formation of mutagenic apurinic sites (32). The oestrogen metabolites 2- and 4-hydroxyestrogens have also been reported to cause double-strand breaks in vitro (3). DNA double-strand breaks seem to be the predominant signal for the activation of ATM-mediated pathways (4). The CHEK2 protein is activated by ATM and thus affects cell cycle arrest and DNA repair (5–8). Our results imply that a defect in the CHEK2 gene affecting the function or expression of the CHEK2 protein increases endometrial cancer risk mainly in combination with increased oestrogen exposure. This study was designed in such a way that the tagSNPs in each gene predicted common variation of over 3% in minor allele frequency with at least 80% probability. It is unlikely that TAG1 itself has a structural effect on the CHEK2 protein, as it is located in an intronic region, but it is still possible that it has a regulatory effect on the protein expression. Another likely scenario is that a common polymorphism in linkage disequilibrium with TAG1 might be responsible for this association.

Interestingly, we observed carriers of haplotype 4 in ATM to have decreased endometrial cancer risk if they had never smoked in their lifetime. Carriers of this haplotype also had decreased endometrial cancer risk overall, although it was not as pronounced as in non-smokers and did not carry over to the global test of significance. Haplotype 4 did not carry a rare allele from any of the tagSNPs (it carried only the tagSNP common alleles), which is in line with the observed lack of effect of the ATM tagSNPs on endometrial cancer risk. One plausible biological explanation for this finding is that non-smoking ATM haplotype 4 carriers are more efficient in repairing oestrogen-related DNA damage than non-carriers. Smoking has been suggested to have anti-oestrogenic effects (33), and women who smoke therefore are likely to be less exposed to oestrogen. These women may be able to adequately repair the lower levels of oestrogen-related DNA damage regardless of their ATM haplotype. In non-smokers, however, oestrogen levels have been found to be higher than in smokers (34–36). In this situation, the increased DNA damage may exceed the repair-capabilities of those women who do not possess ATM haplotype 4, whereas women with ATM haplotype 4 may be able to manage the excess levels of damage imposed by oestrogen.

Additionally, we found decreased endometrial cancer risk among carriers of haplotype 1 in CHEK2 who were younger than 49 years of age at time of menopause. The only rare allele carried by haplotype 1 was the C allele of TAG5 in CHEK2, but TAG5 itself did not appear to affect endometrial cancer risk. Women who are relatively young at time of menopause have experienced fewer ovulatory cycles and thus less exposure to oestrogen than women who experience menopause at an older age (37). It is possible that carriers of haplotype 1 are more capable of managing the low oestrogen-related DNA damage in women with early age at menopause than carriers of other haplotypes in CHEK2. However, in the presence of high oestrogen-related DNA damage (in women with higher oestrogen exposure), even haplotype 1 carriers appear to be unable to repair the large amount of DNA damage.

Neither the 1100delC deletion in CHEK2 nor the rs1800056 (2572 T -> C, F858L) and rs1800058 (4258 C -> T, L1420F) variants in ATM have been previously studied in relation to endometrial cancer risk, although they have been found to be associated with breast cancer risk in some studies (38–40). The 1100delC deletion leads to a premature termination of translation that abolishes CHEK2 kinase activity (41), and the rs1800056 and rs1800058 missense variants appear to cause chromosomal instability and abolish the radiation-induced kinase activity of ATM (42). We found no effect of these variants on endometrial cancer risk in our data, but since there were very few carriers of the rare alleles, our statistical power was low.

In conclusion, individuals carrying TAG1 in CHEK2, haplotype 4 in ATM or haplotype 1 in CHEK2 may harbour a variant that affects endometrial cancer risk in combination with oestrogen-related exposures. Our study was large, population-based and we had sufficient SNP coverage across all three genes. Our findings did not however withstand multiple testing corrections and thus await further corroboration.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
SUPPLEMENTARY MATERIAL
 REFERENCES
 
Parent study
The study base consisted of all Swedish-born women between 50 and 74 years of age who resided in Sweden between January 1994 and December 1995. During that period, we identified all endometrial cancer cases at diagnosis through the six regional cancer registries in Sweden, which provide virtually complete data on incident cancers in Sweden (26). We randomly selected controls that matched the cases in 5 year age strata from the Swedish Registry of Total Population. Of the eligible cases and controls, 802 (76%) cases and 3550 (84%) controls participated in this initial questionnaire-based study, providing detailed information about menopausal hormone use, reproductive history and other lifestyle factors. Histological specimens for the case women were retrieved from all 35 pathology departments in Sweden and reviewed and re-classified by the study pathologist. Results from the study have been published (43–50).

Present study
From the parent study, we selected all 802 endometrial cancer cases and randomly selected 802 age-frequency matched controls among the pre- or postmenopausal participants without any previous malignancy. With the intention of increasing statistical power in subgroup analyses, we further selected all 277 remaining controls who had used menopausal hormones (medium potency oestrogen alone or any combination of medium potency oestrogen and progestin) for at least 2 years and all 124 controls with self-reported diabetes mellitus. Additionally, 871 controls who were shared between the initial endometrial cancer study and a parallel breast cancer study with the same study base and inclusion criteria were added to our control sample. In total, we selected 802 cases and 2074 controls.

Following informed consent, participants donated whole blood. For deceased endometrial cancer cases (n = 96) and those cases who declined to donate blood but consented to our use of tissue (n = 83), we collected archived, paraffin-embedded, non-cancerous tissue samples. We acquired 65% of the requested tissue samples; the main reason for failure to retrieve the samples was unwillingness or lack of time at the respective pathology department to provide the tissue blocks. In total, we obtained blood samples and archived tissue samples for 603 and 116 endometrial cancer patients, respectively, and blood samples for 1574 controls. Reasons for non-participation included lack of interest in research, a negative attitude towards genetic research, old age and severe disease or death. Population-based participation rates (taking into account the proportion that did not participate in the questionnaire study) for cases and controls were 68 and 64%, respectively.

This study was approved by the Institutional Review Boards in Sweden and the National University of Singapore.

DNA isolation
The Swegene laboratories in Malmö (Sweden) extracted DNA from 4 ml of whole blood, using the QIAamp DNA Blood Maxi Kit (Qiagen), according to the manufacturer's instructions. From non-malignant cells in paraffin-embedded tissue, we extracted DNA using a standard phenol/chloroform/isoamyl alcohol protocol (51). We successfully isolated DNA from 600 (blood) and 116 (tissue) endometrial cancer patients and from 1567 controls.

SNP markers and genotyping
The ATM gene covers 146.3 kb of genomic sequence on chromosome 11, CHEK2 spans 54.1 kb on chromosome 22 and ERBB2 covers 33.7 kb on chromosome 17 (dbSNP build 125). We selected SNPs in the ATM, CHEK2 and ERBB2 genes and their 10 kb flanking sequences from dbSNP (124, http://www.ncbi.nlm.nih.gov/SNP/) and Celera databases, aiming for an initial marker density of at least one SNP per 5 kb. SNPs were genotyped using the Sequenom primer extension-based assay (San Diego, CA) and the BeadArray system from Illumina (San Diego, CA) following the manufacturers' instructions. All genotyping results were generated and checked by laboratory staff unaware of case–control status. Only SNPs in which more than 85% of the samples gave a genotype call were analysed further. As quality control, we genotyped 200 randomly selected SNPs in the 92 control samples, using both the Sequenom system and the BeadArray system. The genotype concordance was >99.5%, suggesting high genotyping accuracy.

TagSNP selection and coverage evaluation
We reconstructed haplotypes for all three genes, using the PLEM algorithm (52) implemented in the tagSNPs program (53), and selected tagSNPs on the basis of the R2 coefficient, which quantifies how well the tagSNP haplotypes predict the SNPs or the number of copies of haplotypes an individual carries. We chose tagSNPs so that common SNP genotypes (minor allele frequency ≥0.03) and common haplotypes (frequency ≥0.03) were predicted with R2 ≥ 0.8 (54). In order to evaluate our tagSNPs' performance in capturing unobserved SNPs within the genes and to assess whether we needed a denser set of markers, we performed a SNP-dropping analysis (25,55). In brief, each of the genotyped SNPs was dropped in turn and tagSNPs were selected from the remaining SNPs so that their haplotypes predicted the remaining SNPs with an R2 value of 0.85. We then estimated how well the tagSNP haplotypes of the remaining SNPs predicted the dropped SNP, an evaluation that can provide an unbiased and accurate estimate of tagSNP performance (25,55).

Tumour characteristics
Endometrial cancers can be divided into Type I endometroid tumours and Type II non-endometroid tumours (29–31), where endometroid tumours constitute the majority of endometrial cancers. Endometroid tumours can be further divided according to cell differentiation (Grade). We defined grades as follows: Grade I tumours were defined as well differentiated carcinomas, with maximum 5% solid areas; Grade II tumours as moderately differentiated, with 6–50% solid areas and Grade III tumours as poorly differentiated or entirely undifferentiated, with more than 50% solid areas. Myometrial invasion was classified as: No—none or less than 50% of the myometrial thickness; Yes—at least 50% of the myometrial thickness or through the serosa.

Statistical analyses
Our testing strategy was to fit a single model and to assess haplotype-trait association as a global likelihood ratio test within each stratum of risk factor subgroup. We accounted for the number of tests by using a permutation approach that controls the familywise error rate (probability of rejecting one or more true null hypotheses) and takes into account the dependence structure of the hypotheses (56). Only if the haplotype global test was significant in the stratified analysis did we proceed to explore the individual haplotype contrasts (57). To estimate power, we used a method described by Chapman et al. (58), which assumes co-dominant effects at an unobserved locus.

We first computed expected haplotype dosage, using the tagSNPs program (53), with haplotype frequencies estimated for cases and controls combined, assuming HWE of haplotypes. We then included the haplotype dosages or the tagSNPs as explanatory variables in our models. We applied unconditional logistic regression models adjusted for age (in 5 year age groups) to assess the association between ATM, CHEK2 and ERBB2 tagSNPs or haplotypes and risk of endometrial cancer overall. We also performed the analyses restricted to certain endometrial tumour subtypes or stratified by endometrial cancer risk factors. The appropriateness of these approaches is argued for by Stram et al. (53). That is, when R2 values are high, as is the case here, point and interval estimates obtained by this approach will be approximately accurate.

Confounding has been defined as the presence of a common cause to the exposure and the outcome (59). We believe that lifestyle and reproductive endometrial cancer risk factors are unlikely to cause genetic variation in the genes, but they could be intermediates in the causal pathway between the genes and endometrial cancer. For completeness, we assessed among the randomly selected controls—using Kruskal–Wallis and {chi}2 tests—whether the tagSNPs were associated with the endometrial cancer risk factors given in Table 1. Analyses were performed using the SAS system (release 9.1, SAS Institute Inc., Cary, NC, USA).


   SUPPLEMENTARY MATERIAL
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
REFERENCES
 
Supplementary Material is available at HMG Online.


   ACKNOWLEDGEMENTS
 
We are grateful to all the women who took the time and effort to participate in this study and to Anna Christensson and Boel Bissmarck who obtained consent and coordinated the collecting of samples from the study participants. We are also indebted to Christer Halldén at Swegene laboratories, Malmö, Sweden, for overseeing the DNA isolation from the blood samples; Meah Wee Yang and Ong Eng Hu Jason for genotyping; Lim Siew Lan and Irene Chen for isolating DNA from paraffin-embedded tissue; Marie Vivian Wong Tzu Yen for retrieving SNP information; Frans Verhoeff for processing the genotyping data; and Anthony S. Gunnell for critical review of the manuscript and intellectual input. The Wallenberg Consortium North and the Swedish Foundation for Strategic Research also deserve our gratitude for financially supporting K.H. This study was supported by funding from the Agency for Science, Technology and Research of Singapore (A*STAR).

Conflict of Interest statement. None declared.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 

  1. Hemminki K., Bermejo J.L., Granstrom C. (2005) Endometrial cancer: population attributable risks from reproductive, familial and socioeconomic factors. Eur. J. Cancer 41:2155–2159.[CrossRef][Web of Science][Medline]

  2. Liehr J.G. (2001) Genotoxicity of the steroidal oestrogens oestrone and oestradiol: possible mechanism of uterine and mammary cancer development. Hum. Reprod. Update 7:273–281.[Abstract/Free Full Text]

  3. Thibodeau P.A., Kachadourian R., Lemay R., Bisson M., Day B.J., Paquette B. (2002) In vitro pro- and antioxidant properties of estrogens. J. Steroid Biochem. Mol. Biol. 81:227–236.[CrossRef][Web of Science][Medline]

  4. Ismail I.H., Nystrom S., Nygren J., Hammarsten O. (2005) Activation of ataxia telangiectasia mutated by DNA strand break-inducing agents correlates closely with the number of DNA double strand breaks. J. Biol. Chem. 280:4649–4655.[Abstract/Free Full Text]

  5. Matsuoka S., Rotman G., Ogawa A., Shiloh Y., Tamai K., Elledge S.J. (2000) Ataxia telangiectasia-mutated phosphorylates Chk2 in vivo and in vitro. Proc. Natl Acad. Sci. USA 97:10389–10394.[Abstract/Free Full Text]

  6. Zeng Y., Forbes K.C., Wu Z., Moreno S., Piwnica-Worms H., Enoch T. (1998) Replication checkpoint requires phosphorylation of the phosphatase Cdc25 by Cds1 or Chk1. Nature 395:507–510.[CrossRef][Medline]

  7. Shieh S.Y., Ahn J., Tamai K., Taya Y., Prives C. (2000) The human homologs of checkpoint kinases Chk1 and Cds1 (Chk2) phosphorylate p53 at multiple DNA damage-inducible sites. Genes Dev. 14:289–300.[Abstract/Free Full Text]

  8. Lee J.S., Collins K.M., Brown A.L., Lee C.H., Chung J.H. (2000) hCds1-mediated phosphorylation of BRCA1 regulates the DNA damage response. Nature 404:201–204.[CrossRef][Medline]

  9. Savitsky K., Bar-Shira A., Gilad S., Rotman G., Ziv Y., Vanagaite L., Tagle D.A., Smith S., Uziel T., Sfez S., et al. (1995) A single ataxia telangiectasia gene with a product similar to PI-3 kinase. Science 268:1749–1753.[Abstract/Free Full Text]

  10. Cortez D., Wang Y., Qin J., Elledge S.J. (1999) Requirement of ATM-dependent phosphorylation of brca1 in the DNA damage response to double-strand breaks. Science 286:1162–1166.[Abstract/Free Full Text]

  11. Khosravi R., Maya R., Gottlieb T., Oren M., Shiloh Y., Shkedy D. (1999) Rapid ATM-dependent phosphorylation of MDM2 precedes p53 accumulation in response to DNA damage. Proc. Natl Acad. Sci. USA 96:14973–14977.[Abstract/Free Full Text]

  12. Riben M.W., Malfetano J.H., Nazeer T., Muraca P.J., Ambros R.A., Ross J.S. (1997) Identification of HER-2/neu oncogene amplification by fluorescence in situ hybridization in stage I endometrial carcinoma. Mod. Pathol. 10:823–831.[Web of Science][Medline]

  13. Nazeer T., Ballouk F., Malfetano J.H., Figge H., Ambros R.A. (1995) Multivariate survival analysis of clinicopathologic features in surgical stage I endometrioid carcinoma including analysis of HER-2/neu expression. Am. J. Obstet. Gynecol. 173:1829–1834.[CrossRef][Web of Science][Medline]

  14. Rolitsky C.D., Theil K.S., McGaughy V.R., Copeland L.J., Niemann T.H. (1999) HER-2/neu amplification and overexpression in endometrial carcinoma. Int. J. Gynecol. Pathol. 18:138–143.[Web of Science][Medline]

  15. Morrison C., Zanagnolo V., Ramirez N., Cohn D.E., Kelbick N., Copeland L., Maxwell L.G., Fowler J.M. (2006) HER-2 is an independent prognostic factor in endometrial cancer: association with outcome in a large cohort of surgically staged patients. J. Clin. Oncol. 24:2376–2385.[Abstract/Free Full Text]

  16. Akiyama T., Sudo C., Ogawara H., Toyoshima K., Yamamoto T. (1986) The product of the human c-erbB-2 gene: a 185-kilodalton glycoprotein with tyrosine kinase activity. Science 232:1644–1646.[Abstract/Free Full Text]

  17. Bargmann C.I., Hung M.C., Weinberg R.A. (1986) The neu oncogene encodes an epidermal growth factor receptor-related protein. Nature 319:226–230.[CrossRef][Medline]

  18. Coussens L., Yang-Feng T.L., Liao Y.C., Chen E., Gray A., McGrath J., Seeburg P.H., Libermann T.A., Schlessinger J., Francke U., et al. (1985) Tyrosine kinase receptor with extensive homology to EGF receptor shares chromosomal location with neu oncogene. Science 230:1132–1139.[Abstract/Free Full Text]

  19. King C.R., Kraus M.H., Aaronson S.A. (1985) Amplification of a novel v-erbB-related gene in a human mammary carcinoma. Science 229:974–976.[Abstract/Free Full Text]

  20. Schechter A.L., Stern D.F., Vaidyanathan L., Decker S.J., Drebin J.A., Greene M.I., Weinberg R.A. (1984) The neu oncogene: an erb-B-related gene encoding a 185,000-Mr tumour antigen. Nature 312:513–516.[CrossRef][Medline]

  21. Threadgill D.W., Dlugosz A.A., Hansen L.A., Tennenbaum T., Lichti U., Yee D., LaMantia C., Mourton T., Herrup K., Harris R.C., et al. (1995) Targeted disruption of mouse EGF receptor: effect of genetic background on mutant phenotype. Science 269:230–234.[Abstract/Free Full Text]

  22. Yarden Y. and Sliwkowski M.X. (2001) Untangling the ErbB signalling network. Nat. Rev. Mol. Cell Biol. 2:127–137.[CrossRef][Web of Science][Medline]

  23. Lee K.F., Simon H., Chen H., Bates B., Hung M.C., Hauser C. (1995) Requirement for neuregulin receptor erbB2 in neural and cardiac development. Nature 378:394–398.[CrossRef][Medline]

  24. Einarsdottir K., Humphreys K., Bonnard C., Palmgren J., Iles M.M., Sjolander A., Li Y., Chia K.S., Liu E.T., Hall P., et al. (2006) Linkage disequilibrium mapping of CHEK2: common variation and breast cancer risk. PLoS Med. 3:e168.[CrossRef][Medline]

  25. Weale M.E., Depondt C., Macdonald S.J., Smith A., Lai P.S., Shorvon S.D., Wood N.W., Goldstein D.B. (2003) Selection and evaluation of tagging SNPs in the neuronal-sodium-channel gene SCN1A: implications for linkage-disequilibrium gene mapping. Am. J. Hum. Genet. 73:551–565.[CrossRef][Web of Science][Medline]

  26. Cancer Incidence in Sweden 2003 (2005) The National Board of Health and Welfare, Centre for Epidemiology, Official Statistics of Sweden.

  27. The International Hapmap Consortium. (2003) The International HapMap Project. Nature 426:789–796.[CrossRef][Medline]

  28. International Hapmap Consortium. (2005) A haplotype map of the human genome. Nature 437:1299–1320.[CrossRef][Medline]

  29. Shiozawa T. and Konishi I. (2006) Early endometrial carcinoma: clinicopathology, hormonal aspects, molecular genetics, diagnosis, and treatment. Int. J. Clin. Oncol. 11:13–21.[CrossRef][Medline]

  30. Sherman M.E., Sturgeon S., Brinton L.A., Potischman N., Kurman R.J., Berman M.L., Mortel R., Twiggs L.B., Barrett R.J., Wilbanks G.D. (1997) Risk factors and hormone levels in patients with serous and endometrioid uterine carcinomas. Mod. Pathol. 10:963–968.[Web of Science][Medline]

  31. Bokhman J.V. (1983) Two pathogenetic types of endometrial carcinoma. Gynecol. Oncol. 15:10–17.[CrossRef][Web of Science][Medline]

  32. Cavalieri E.L., Rogan E.G., Chakravarti D. (2002) Initiation of cancer and other diseases by catechol ortho-quinones: a unifying mechanism. Cell Mol. Life Sci. 59:665–681.[CrossRef][Web of Science][Medline]

  33. Tanko L.B. and Christiansen C. (2004) An update on the antiestrogenic effect of smoking: a literature review with implications for researchers and practitioners. Menopause 11:104–109.[CrossRef][Web of Science][Medline]

  34. Westhoff C., Gentile G., Lee J., Zacur H., Helbig D. (1996) Predictors of ovarian steroid secretion in reproductive-age women. Am. J. Epidemiol 144:381–388.[Abstract/Free Full Text]

  35. Sterzik K., Strehler E., De Santo M., Trumpp N., Abt M., Rosenbusch B., Schneider A. (1996) Influence of smoking on fertility in women attending an in vitro fertilization program. Fertil. Steril. 65:810–814.[Web of Science][Medline]

  36. Van Voorhis B.J., Dawson J.D., Stovall D.W., Sparks A.E., Syrop C.H. (1996) The effects of smoking on ovarian function and fertility during assisted reproduction cycles. Obstet. Gynecol. 88:785–791.[CrossRef][Web of Science][Medline]

  37. Persson I. (2000) Estrogens in the causation of breast, endometrial and ovarian cancers—evidence and hypotheses from epidemiological findings. J. Steroid Biochem. Mol. Biol. 74:357–364.[CrossRef][Web of Science][Medline]

  38. Stredrick D.L., Garcia-Closas M., Pineda M.A., Bhatti P., Alexander B.H., Doody M.M., Lissowska J., Peplonska B., Brinton L.A., Chanock S.J., et al. (2006) The ATM missense mutation p.Ser49Cys (c.146C > G) and the risk of breast cancer. Hum. Mutat. 27:538–544.[CrossRef][Web of Science][Medline]

  39. Angele S., Romestaing P., Moullan N., Vuillaume M., Chapot B., Friesen M., Jongmans W., Cox D.G., Pisani P., Gerard J.P., et al. (2003) ATM haplotypes and cellular response to DNA damage: association with breast cancer risk and clinical radiosensitivity. Cancer Res. 63:8717–8725.[Abstract/Free Full Text]

  40. CHEK2 Breast Cancer Case-Control Consortium. (2004) CHEK2*1100delC and susceptibility to breast cancer: a collaborative analysis involving 10,860 breast cancer cases and 9,065 controls from 10 studies. Am. J. Hum. Genet. 74:1175–1182.[CrossRef][Web of Science][Medline]

  41. Wu X., Webster S.R., Chen J. (2001) Characterization of tumor-associated Chk2 mutations. J. Biol. Chem. 276:2971–2974.[Abstract/Free Full Text]

  42. Scott S.P., Bendix R., Chen P., Clark R., Dork T., Lavin M.F. (2002) Missense mutations but not allelic variants alter the function of ATM by dominant interference in patients with breast cancer. Proc. Natl Acad. Sci. USA 99:925–930.[Abstract/Free Full Text]

  43. Weiderpass E., Baron J.A., Adami H.O., Magnusson C., Lindgren A., Bergstrom R., Correia N., Persson I. (1999) Low-potency oestrogen and risk of endometrial cancer: a case–control study. Lancet 353:1824–1828.[CrossRef][Web of Science][Medline]

  44. Weiderpass E., Adami H.O., Baron J.A., Magnusson C., Lindgren A., Persson I. (1999) Use of oral contraceptives and endometrial cancer risk (Sweden). Cancer Causes Control 10:277–284.[CrossRef][Web of Science][Medline]

  45. Weiderpass E., Adami H.O., Baron J.A., Magnusson C., Bergstrom R., Lindgren A., Correia N., Persson I. (1999) Risk of endometrial cancer following estrogen replacement with and without progestins. J. Natl Cancer Inst. 91:1131–1137.[Abstract/Free Full Text]

  46. Weiderpass E., Persson I., Adami H.O., Magnusson C., Lindgren A., Baron J.A. (2000) Body size in different periods of life, diabetes mellitus, hypertension, and risk of postmenopausal endometrial cancer (Sweden). Cancer Causes Control 11:185–192.[CrossRef][Web of Science][Medline]

  47. Weiderpass E. and Baron J.A. (2001) Cigarette smoking, alcohol consumption, and endometrial cancer risk: a population-based study in Sweden. Cancer Causes Control 12:239–247.[CrossRef][Web of Science][Medline]

  48. Moradi T., Weiderpass E., Signorello L.B., Persson I., Nyren O., Adami H.O. (2000) Physical activity and postmenopausal endometrial cancer risk (Sweden). Cancer Causes Control 11:829–837.[CrossRef][Web of Science][Medline]

  49. Terry P., Wolk A., Vainio H., Weiderpass E. (2002) Fatty fish consumption lowers the risk of endometrial cancer: a nationwide case–control study in Sweden. Cancer Epidemiol. Biomarkers Prev. 11:143–145.[Abstract/Free Full Text]

  50. Terry P., Vainio H., Wolk A., Weiderpass E. (2002) Dietary factors in relation to endometrial cancer: a nationwide case–control study in Sweden. Nutr. Cancer 42:25–32.[CrossRef][Web of Science][Medline]

  51. Isola J., DeVries S., Chu L., Ghazvini S., Waldman F. (1994) Analysis of changes in DNA sequence copy number by comparative genomic hybridization in archival paraffin-embedded tumor samples. Am. J. Pathol. 145:1301–1308.[Abstract]

  52. Qin Z.S., Niu T., Liu J.S. (2002) Partition-ligation-expectation-maximization algorithm for haplotype inference with single-nucleotide polymorphisms. Am. J. Hum. Genet. 71:1242–1247.[CrossRef][Web of Science][Medline]

  53. Stram D.O., Haiman C.A., Hirschhorn J.N., Altshuler D., Kolonel L.N., Henderson B.E., Pike M.C. (2003) Choosing haplotype-tagging SNPS based on unphased genotype data using a preliminary sample of unrelated subjects with an example from the Multiethnic Cohort Study. Hum. Hered. 55:27–36.[CrossRef][Web of Science][Medline]

  54. Gabriel S.B., Schaffner S.F., Nguyen H., Moore J.M., Roy J., Blumenstiel B., Higgins J., DeFelice M., Lochner A., Faggart M., et al. (2002) The structure of haplotype blocks in the human genome. Science 296:2225–2229.[Abstract/Free Full Text]

  55. Iles M.M. (2006) Obtaining unbiased estimates of tagging SNP performance. Ann. Hum. Genet. 70:254–261.[CrossRef][Web of Science][Medline]

  56. Westfall P. and Young S. (1993) Resampling-Based Multiple Testing: Examples and Methods for P-Value Adjustment (John Wiley & Sons, Inc., New York).

  57. Schaid D.J., Rowland C.M., Tines D.E., Jacobson R.M., Poland G.A. (2002) Score tests for association between traits and haplotypes when linkage phase is ambiguous. Am J. Hum. Genet. 70:425–434.[CrossRef][Web of Science][Medline]

  58. Chapman J.M., Cooper J.D., Todd J.A., Clayton D.G. (2003) Detecting disease associations due to linkage disequilibrium using haplotype tags: a class of tests and the determinants of statistical power. Hum. Hered. 56:18–31.[CrossRef][Web of Science][Medline]

  59. Hernan M.A., Hernandez-Diaz S., Werler M.M., Mitchell A.A. (2002) Causal knowledge as a prerequisite for confounding evaluation: an application to birth defects epidemiology. Am. J. Epidemiol. 155:176–184.[Abstract/Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Cancer Res.Home page
S. Leng, C. A. Stidley, R. Willink, A. Bernauer, K. Do, M. A. Picchi, X. Sheng, M. A. Frasco, D. Van Den Berg, F. D. Gilliland, et al.
Double-Strand Break Damage and Associated DNA Repair Genes Predispose Smokers to Gene Methylation
Cancer Res., April 15, 2008; 68(8): 3049 - 3056.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrowOA All Versions of this Article:
16/2/154    most recent
ddl451v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (5)
Google Scholar
Right arrow Articles by Einarsdóttir, K.
Right arrow Articles by Wedrén, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Einarsdóttir, K.
Right arrow Articles by Wedrén, S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?