This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Garcia-Closas, M. Articles by Perou, C. M. Search for Related Content PubMed PubMed Citation Articles by Garcia-Closas, M. Articles by Perou, C. M.

Common Genetic Variation in GATA-Binding Protein 3 and Differential Susceptibility to Breast Cancer by Estrogen Receptor Tumor Status

¹ Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, Maryland; ² Departments of Genetics and Pathology, Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, North Carolina; ³ Core Genotyping Facility Advanced Technology Center, National Cancer Institute, Gaithersburg, Maryland; ⁴ Department of Genetics, Institute for Cancer Research, Rikshospitalet-Radiumhospitalet Medical Centre and ⁵ Faculty of Medicine, University of Oslo, Oslo, Norway; ⁶ Cancer Center and M. Sklodowska Curie Institute of Oncology, Warsaw, Poland; ⁷ Nofer Institute of Occupational Medicine, Lodz, Poland; ⁸ Office of Cancer Genomics and ⁹ Pediatrics Oncology Branch, Center for Cancer Research, National Cancer Institute, Bethesda, Maryland; and ¹⁰ Institute of Community Medicine, University of Tromsø, Tromsø, Norway

Requests for reprints: Montserrat Garcia-Closas, Hormonal and Reproductive Epidemiology Branch Division of Cancer Epidemiology and Genetics, National Cancer Institute, NIH, 6120 Executive Boulevard, Room 7076, MSC 7234, Rockville, MD 20852-7234. Phone: 301-435-3981; Fax: 301-402-0916. E-mail: montse{at}nih.gov

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
GATA-binding protein 3 (GATA3) is a transcription factor and a putative tumor suppressor that is highly expressed in normal breast luminal epithelium and estrogen receptor (ER)–positive breast tumors. We hypothesized that common genetic variation in GATA3 could influence breast carcinogenesis. Four tag single-nucleotide polymorphisms (SNP) in GATA3 and its 3' flanking gene FLJ4598 were genotyped in two case control studies in Norway and Poland (2,726 cases and 3,420 controls). Analyses of pooled data suggested a reduced risk of breast cancer associated with two intronic variants in GATA3 in linkage disequilibrium (rs3802604 in intron 3 and rs570613 in intron 4). Odds ratio (95% confidence interval) for rs570613 heterozygous and rare homozygous versus common homozygous were 0.85 (0.75-1.95) and 0.82 (0.62-0.96), respectively (P_trend = 0.004). Stronger associations were observed for subjects with ER-negative, than ER-positive, tumors (P_{heterogeneity} = 0.01 for rs3802604; P_{heterogeneity} = 0.09 for rs570613). Although no individual SNPs were associated with ER-positive tumors, two haplotypes (GGTC in 2% of controls and AATT in 7% of controls) showed significant and consistent associations with increased risk for these tumors when compared with the common haplotype (GATT in 46% of controls): 1.71 (1.27-2.32) and 1.26 (1.03-1.54), respectively. In summary, data from two independent study populations showed two intronic variants in GATA3 associated with overall decreases in breast cancer risk and suggested heterogeneity of these associations by ER status. These differential associations are consistent with markedly different levels of GATA3 protein by ER status. Additional epidemiologic studies are needed to clarify these intriguing relationships. (Cancer Epidemiol Biomarkers Prev 2007;16(11):2269–75)

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
GATA-binding protein 3 (GATA3) is a transcription factor that is highly expressed in normal breast luminal epithelium and the luminal A tumor subtype that has been defined in studies of gene expression patterns (1-5). GATA3 expression highly correlates with expression of estrogen receptor (ER; refs. 6, 7), as well as other genes thought to be important in breast luminal epithelial cell biology, including LIV1, RERG, and TFF3 (8). Luminal A tumors are associated with better prognosis than other tumor subtypes (luminal B, basal-like, and HER2+ subtypes; refs. 1, 2), the latter of which show little or no expression of GATA3. Building upon previous reports, including metaanalysis of microarray studies of GATA3 transcripts (7), Mehra et al. (9) have shown that GATA3 protein expression in tumors predicted prognosis. These authors also confirmed that GATA3 levels are lower in tumors that are ER-negative and in those with high histologic grade. Parikh et al. have also shown that among ER-positive tumors, GATA3 protein expression seems to predict tumor estrogen responsiveness (10).

The mechanisms underlying the role of GATA3 in estrogen response and breast cancer prognosis are not clear. For example, GATA3 mRNA is down-regulated in normal breast xenograft samples after estradiol treatment (11), but estradiol-treated MCF-7 cells do not show altered expression of GATA3 mRNA (6). The discovery of somatic mutations in GATA3 in some ER-positive breast cancers, coupled with the observation that GATA3-transduced cells have greatly decreased proliferation rates in vitro, suggest that GATA3 act as a tumor suppressor (12). Thus, we hypothesized that common genetic variation could alter expression and/or function of GATA3 and thus play an etiologic role in breast cancer. To address this hypothesis, we carried out a comprehensive evaluation of common variation in GATA3 and evaluated its relation with breast cancer risk in two independent study populations from Norway and Poland.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Study Populations
Norwegian Breast Cancer Study. Breast cancer patients (n = 731) were enrolled in accordance with local institutional review board guidelines and included four series of previously described case series.

(a) Breast cancer patients sequentially enrolled at Ullevål University Hospital from 1990 to 1994, representing the breast cancer population during this period (13, 14). The mean age was 64 years (range, 28-92 years). Blood samples were collected in 1994 to 1996 from 119 patients who were still alive (80%), living in the Oslo area, and consented to give blood (70% of eligible women). Time from diagnosis to blood collection was 0 to 6 years.

(b) Breast cancer patients admitted to the Norwegian Radium Hospital in 1972 to 1991 (15). The mean age at diagnosis was 57 years (range, 27-94 years). Blood was drawn between 1987 and 1991 (n = 224), either during follow-up visits, relapse, or at diagnosis.

(c) Breast cancer patients diagnosed at the Norwegian Radium Hospital and treated with radiotherapy in 1975 to 1986. The mean age at diagnosis was 59 years (range, 26-75 years). Blood samples were collected in 1996 from a subset of patients alive at that time (21%) who were part of a treatment evaluation and agreed (82%) to provide a blood sample (n = 263; ref. 16).

(d) Breast cancer patients with stage I and stage II disease enrolled in the Oslo micrometastases study between 1995 and 1998 who had blood samples collected at the time of diagnoses (n = 125; refs. 17, 18). The average age at diagnosis was 56 years (range, 29-82 years). Blood sampling was done just before primary surgery for their breast cancer from all patients included in the study.

The majority of control subjects (n = 1,015) were women with a negative mammogram from the Tromsø Mammography and Breast Cancer Study conducted in 2001 and 2002. About 70% of the women, 55 to 71 years of age, residing in the municipality of Tromsø in Norway and attending the Norwegian Breast Cancer Screening Program at the University Hospital of North Norway agreed to participate (19). The study was approved by the National Data Inspection Board and the Regional Committee for Medical Research Ethics. In addition, we included healthy woman (n = 109), 55 to 72 years of age, participating in the Norwegian Breast Cancer Screening Program in Bergen in 1999, with two negative mammogram during a 2-year period (20). From this latter group, women with p.o. hormone replacement therapy or history of diabetes or other endocrine disorders were excluded. The mean (range) ages for cases (n = 731) and controls (n = 1,124) in the Norwegian study were 56 (26-93) and 62 (55-72) years, respectively.

Polish Breast Cancer Study. A population-based case control study was conducted among women residing in two Polish cities, Warsaw and Lodz (21). Eligible cases were women ages 20 to 74 years who were newly diagnosed with either histologically or cytologically confirmed in situ or invasive breast cancer in 2000 to 2003. About 90% of cases were identified through a rapid identification system in participating hospitals, and the remainder through cancer registries, to ensure complete case ascertainment. Controls with no history of breast cancer were randomly selected from population lists during the case ascertainment period and were frequency matched to cases by city and age in 5-year categories. Institutional Review Board approval was obtained from all participating institutions, and signed informed consent was obtained for all respondents.

A total of 2,386 cases (79% of eligible cases) and 2,502 controls (69% of eligible controls) provided a personal interview on known and suspected risk factors. Blood samples for DNA extraction were obtained from 1,995 cases (84% of participating cases) and 2,296 controls (94% of participating controls). Most cases (94%) were diagnosed with invasive tumors. The mean (range) age was 56 (27-74) for cases and 56 (24-75) years for controls.

Genotyping. A resequence analysis of all exons, including the 5' and 3' untranlated regions and evolutionarily conserved intronic regions in GATA3 was done in 94 healthy Norwegian women and 102 individuals in the SNP500Cancer panel (23). Single-nucleotide polymorphisms (SNP) were selected using the haplotype-tagging program of Stram et al. 22 with r² of >0.80 and minor allele frequency of >0.05 and genotyped in the Norwegian and Polish studies. SNP selection included the closely positioned upstream neighboring gene FLJ45983 because of linkage disequilibrium observed with SNPs in GATA3, suggesting that variation in this gene might be involved in regulatory functions affecting GATA3. The function of this gene in unknown. Genotype analyses were done on blood DNA at the Core Genotyping Facility of the Division of Cancer Epidemiology and Genetics, National Cancer Institute for three SNPs in GATA3 (rs3802604, IVS4+1468G>A; rs570613, IVS4+401T>C; rs422628, IVS4-27C>T) and one SNP in the 3' flanking gene FLJ45983 (rs1149901, Ex1-425G>A). Description and methods for each genotype assay can be found at http://snp500cancer.nci.nih.gov (23). Duplicated DNA pairs from 95 subjects in the Polish study showed >99% concordance for all but one assay in intron 3 of GATA3 (rs3802604) with 98% concordance. Completion was 95% for all assays in the Polish study (5% of samples felt into the NTC cluster, whereas 1% were undetermined calls), and the percentage of completion was similar for cases and controls. Completion was 98% for all assays in the Norwegian study, except for rs570613 which had a completion of 92%. We did not observe significant departures from Hardy-Weinberg equilibrium for any of the SNPs evaluated in the Norwegian or Polish control populations.

Statistical Analyses. Odds ratios (OR) and their 95% confidence intervals (95% CI) were derived from unconditional logistic regression models adjusting for age in 5-year categories. The association between genotypes and breast cancer risk was tested using a trend test. Differences in mean age at diagnosis in cases across different genotypes were tested using a t test. Age-specific estimates of ORs (95% CI) for genotype-disease associations were obtained from logistic regression analyses with interaction terms between genotypes (assuming a linear trend) and age categories. The test for interaction between genotypes and age was done by including an interaction term in a logistic regression model, considering both variables as continuous. We evaluated heterogeneity in the genotype ORs by hormone receptor status in logistic regression models among cases with receptor status as the outcome variable and genotypes as explanatory variables adjusting for age and study. Case control analyses were also done to estimate associations between genotypes and different tumor types. Heterogeneity of estimated ORs by study was tested by introducing an interaction term for genotype and study. Estimates obtained using pooled data from both studies were adjusted by study.

Pairwise linkage disequilibrium was estimated between SNPs based on D' and r² values using Haploview.¹¹ Block structure was determined using genotype data from the control population, and the solid spline of linkage disequilibrium option (D'_threshold > 0.80). Haplotype frequencies within each block, ORs, and their 95% CIs were estimated using HaploStats¹² (version 1.2.1; ref. 24). A global score statistic, adjusted for the matching factors age (in 5-year categories) and study site (Lodz or Warsaw), was used to evaluate the overall difference in haplotype frequencies between cases and controls. Phylogenetic trees [neighbor-joining (ref. 25), nucleotide p distance] were constructed using MEGA 3.1¹³ (26) to assess nucleotide similarity of different haplotypes.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
We observed similar linkage disequilibrium patterns between SNPs in the control populations of the two study populations, although linkage disequilibrium between the SNP in the 3' neighboring gene FLJ45983 and SNPs in GATA3 was less in the Polish, than in the Norwegian, study (Fig. 1 ). There was low correlation between SNPs, with the exception of intron 3 (rs3802604) and intron 4 (rs570613) of GATA3 (r² = 0.67 and 0.65 in the Norwegian and Polish populations, respectively). The allele frequencies among controls in the two study populations were similar. rs1149901 allele frequency was 0.26 and 0.28 for the Norwegian and Polish populations, respectively, 0.40 and 0.37 for rs3802604, 0.45 and 0.40 for rs570613, and 0.26 and 0.24 for rs422628. None of these differences were statistically significant.

View larger version (13K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Patterns of linkage disequilibrium across the GATA3 and neighboring gene in the Norwegian and Polish control populations. Gray scale scheme is based on D' and LOD score values: white, D' < 1 and LOD < 2; shades of gray, D' < 1 and LOD 2. Numbers in squares, D' values the order of SNPs from left to right is FLJ45983 (rs1149901), GATA3 (rs3802604, rs570613, rs422628).

Pooled case control analyses stratified by ER status showed a reduced risk for ER-negative tumors associated with the intron 3 polymorphism [pooled OR (95% CI), 0.86 (0.72-1.04) for heterozygotes and 0.72 (0.54-0.96) for homozygote variants; P_trend = 0.02] and no association with ER-positive tumors (P_trend = 0.80), which was consistent between the two study populations (Table 2 ). Similar associations were observed for the intron 4 SNP (rs570613) in GATA3; however, differences by ER status were not statistically significant.

These analyses could not be done in the Norwegian study because controls were 55 years or older. Case-only analyses in both study populations showed that the SNPs evaluated were significantly associated with age at diagnosis in the Polish study; however, associations were weaker and not statistically significant in the Norwegian population (data not shown). The associations with age at diagnosis in the Polish study remained significant after adjusting for ER status of the tumors (data not shown).

We observed nine haplotypes with a frequency >1% in the control populations (Table 3 ; Supplementary Table S5 for analyses by study). The most common haplotype (GATT) carried common alleles for all SNPs and was present in 46% of the pooled control population. Analyses of pooled data showed four haplotypes carrying both variants in introns 3 and 4 (rs3802604 and rs570613) individually associated with decreased breast cancer risk (GGCT in 14% of controls, GGCC in 1% of controls, AGCC in 18% of controls, and AGCT in 2% of controls). Compared with the common haplotype GATT, only two of these haplotypes (GGCT and AGCC) were associated with reductions in the risk for ER-negative tumors [0.71 (0.58-0.88) and 0.85 (0.72-1.01), respectively; Table 3]. We observed two haplotypes carrying only one of the two variants (GACT in 6% of controls and GGTC in 2% of controls), and none of them were associated with significant reductions in risk of ER-negative tumors. Interestingly, the haplotype carrying the variant in intron 3 (GGTC) and a haplotype with a variant in exon 1 of FLJ45983 (AATT in 7% of controls) were associated with significant increases in risk for ER-positive tumors [1.71 (1.27-2.32) and 1.26 (1.03-1.54), respectively; Table 3]. This increase in risk was consistently found in the Norwegian and Polish populations (Supplementary Table S5).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

No previously published studies have evaluated the association between GATA3 polymorphisms and breast cancer risk. However, two tag SNPs in this report (rs570613 and rs3802604) were included in a genome wide-scan done in a total of 1,200 breast cancer cases and 1,200 controls from the Nurse's Health Study (NHS) under the Cancer Genetic Markers of Susceptibility project.¹⁴ The NHS data were consistent with a reduced risk of breast cancer associated with the homozygous variant genotype for rs570613 [OR (95% CI), 1.01 (0.86-1.22) for CT versus TT and 0.78 (0.60-1.00) for CC versus TT; calculated from genotype frequencies found at http://cgems.cancer.gov/data/], as found in our two study populations. However, the rs3802604 SNP was not associated with risk in the Nurse's Health Study (P_{2df adjusted} = 0.72). None of the other tag SNPs in GATA3 genotyped in the Nurse's Health Study/Cancer Genetic Markers of Susceptibility project showed significant associations with breast cancer risk (P_{2df adjusted} > 0.18).

The biology of GATA3 is complex and encompasses functions in stem cells during differentiation and in terminally differentiated cells (27). Kaufman et al. have shown by targeted disruption of GATA3 in mice that GATA3 is required for proper hair follicle stem cell function (28). In the immune system, GATA3 is required for multiple developmental decisions (29). Thus, if GATA3 signaling differs by developmental stage and cell type, then GATA3 SNPs might be expected to have different effects on the risk of different tumor subtypes that arise from different cell lineages. Heterogeneity in GATA3 signaling across different developmental lineages may underlie the etiologic heterogeneity with respect to ER status observed in this study.

A differential association of GATA3 variants by ER status is consistent with previous observations about different roles of GATA3 in ER-positive and ER-negative breast tumor subtypes. GATA3 protein expression varies according to ER status (1, 2), predicts hormone therapy responsiveness (10), and predicts outcomes in ER-positive patients (7, 9). In human breast tissue, GATA3 and ER expression are highly correlated with proteins expressed highly in cells that line the ducts (luminal epithelial cells). In addition, GATA3 and ER are known to transcriptionally regulate a number of genes in common (6, 30, 31), so ER may influence GATA3 and vice versa.

This study represents the first report of an association between GATA3 polymorphisms and breast cancer risk. The SNP in intron 3 (rs3802604) is found in a highly conserved region of GATA3. Conservation is often high in regions of DNA with important functional consequences, but functional data has not been reported for any of the GATA3 SNPs we investigated. The two intronic SNPs associated with risk were in linkage disequilibrium (D' 0.90, r² 0.65), and thus, it is unclear whether one or both SNPs are responsible for the observed associations. Neither of the two haplotypes with only one of these SNPs was significantly associated with ER-negative tumors, suggesting that both variants might be needed; however, this could also be due to low power to detect significant association for the less common haplotypes. It is also possible that one or both are in linkage disequilibrium with a protective allele not measured in our study. The observed associations between two haplotypes [one carrying the variant for the intron 3 (rs3802604) SNP] and increased risk of ER-positive tumors were unexpected based on individual SNP analyses and require confirmation in other study populations.

The agreement between two large and independent study populations in this report and in NHS data online suggests validity of findings because potential biases that could explain associations are unlikely to be the same in both populations. The rate of participation in the Polish study is among the highest for population-based studies with collection of biological specimens. Although we cannot rule out selection bias, associations with most established risk factors for breast cancer were of expected direction and magnitude (21), indicating that selection bias is unlikely to be important. We had less knowledge regarding participation rates and distribution of breast cancer risk factors among the breast cancer cases in the Norwegian study; however, the findings were generally consistent between the study populations, particularly after stratification by ER tumor status. Both study populations were of homogeneous ethnic background, thus reducing the possibility of bias due to population stratification. Blood samples from most cases in the Norwegian were collected sometime after diagnosis, which provides the possibility of survival biases if the genotypes were to be related to survival.

Results from this report, together with the evidence that GATA3 plays a role as a tumor suppressor (12), suggest that common variation in GATA3 differentially affects the risk for developing breast cancer depending on the ER status of the tumors. Further epidemiologic studies aimed at clarifying this relation are warranted.

	Acknowledgments

 
We thank Drs. Neonila Szeszenia-Dabrowska (Nofer Institute of Occupational Medicine), Witold Zatonski (M. Sklodowska-Curie Institute of Oncology and Cancer Center), and Aljcia Bardin-Mikolajczak (M. Sklodowska-Curie Institute of Oncology and Cancer Center) for their contribution to the design of the study and field work of the Polish Breast Cancer Study, Douglas Richesson (DCEG, National Cancer Institute) for his assistance on statistical analyses, Anita Soni (Westat) for her work on study management for the Polish Breast Cancer Study, Pei Chao (IMS) for her work on data and sample management, and physicians, nurses, interviewers, and study participants for their efforts during filed work.

	Footnotes

 
Grant support: NIH National Cancer Institute Breast Cancer Specialized Program of Research Excellence grants P50-CA58223 and R01-CA-101227-01 (C.M. Perou) and UNC Lineberger Cancer Control Education Program grant R25 CA57726 (M.A. Troester). The Polish Breast Cancer Study was supported by the Intramural Research Program of NIH National Cancer Institute Division of Cancer Epidemiology and Genetics and the Center for Cancer Research. The Tromsø Mammography and Breast Cancer Study was conducted in collaboration with Department of Clinical Research and the Department of Radiology, Center for Breast Imaging, University Hospital of North Norway; Norwegian Women and Cancer Study, University of Tromsø; and Cancer Registry of Norway. The study was supported by the Norwegian Cancer Society, Aakre Foundation, and Norwegian Women's Public Health Association. This work was also supported by the European Union's EU FP6 grant 502983, Research Council of Norway grant 155218/300, and Norwegian Cancer Society grant D 99061.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: Supplementary data for this article are available at Cancer Epidemiology Biomarkers and Prevention Online (http://cebp.aacrjournals.org/). M. Garcia-Closas and M.A. Troester contributed equally to the drafting of the manuscript.

¹¹ http://www.broad.mit.edu/mpg/haploview/index.php

¹² http://mayoresearch.mayo.edu/mayo/research/biostat/schaid.cfm

¹³ http://www.megasoftware.net/

¹⁴ http://cgems.cancer.gov/

Received 5/16/07; revised 8/ 3/07; accepted 8/16/07.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Perou CM, Sorlie T, Eisen MB, et al. Molecular portraits of human breast tumours. Nature 2000;406:747–52.[CrossRef][Medline]
2. Sorlie T, Perou CM, Tibshirani R, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A 2001;98:10869–74.[Abstract/Free Full Text]
3. Gruvberger S, Ringner M, Chen Y, et al. Estrogen receptor status in breast cancer is associated with remarkably distinct gene expression patterns. Cancer Res 2001;61:5979–84.[Abstract/Free Full Text]
4. West M, Blanchette C, Dressman H, et al. Predicting the clinical status of human breast cancer by using gene expression profiles. Proc Natl Acad Sci U S A 2001;98:11462–7.[Abstract/Free Full Text]
5. 't Veer LJ, Dai H, van de Vijver MJ, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002;415:530–6.[CrossRef][Medline]
6. Hoch RV, Thompson DA, Baker RJ, Weigel RJ. GATA-3 is expressed in association with estrogen receptor in breast cancer. Int J Cancer 1999;84:122–8.[CrossRef][Medline]
7. van de, RM, Perou CM, Tibshirani R, et al. Expression of cytokeratins 17 and 5 identifies a group of breast carcinomas with poor clinical outcome. Am J Pathol 2002;161:1991–6.[Abstract/Free Full Text]
8. Sorlie T, Tibshirani R, Parker J, et al. Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci U S A 2003;100:8418–23.[Abstract/Free Full Text]
9. Mehra R, Varambally S, Ding L, et al. Identification of GATA3 as a breast cancer prognostic marker by global gene expression meta-analysis. Cancer Res 2005;65:11259–64.[Abstract/Free Full Text]
10. Parikh P, Palazzo JP, Rose LJ, Daskalakis C, Weigel RJ. GATA-3 expression as a predictor of hormone response in breast cancer. J Am Coll Surg 2005;200:705–10.[CrossRef][Medline]
11. Wilson CL, Sims AH, Howell A, Miller CJ, Clarke RB. Effects of oestrogen on gene expression in epithelium and stroma of normal human breast tissue. Endocr Relat Cancer 2006;13:617–28.[Abstract/Free Full Text]
12. Usary J, Llaca V, Karaca G, et al. Mutation of GATA3 in human breast tumors. Oncogene 2004;23:7669–78.[CrossRef][Medline]
13. Bukholm IK, Nesland JM, Karesen R, Jacobsen U, Borresen AL. Relationship between abnormal p53 protein and failure to express p21 protein in human breast carcinomas. J Pathol 1997;181:140–5.[CrossRef][Medline]
14. Langerød AZHBØ, Nesland JM, Bukholm IK, et al. TP53 mutation status and gene expression profiles are powerful prognostic markers of breast cancer. Breast Cancer Res 2007.
15. Andersen TI, Holm R, Nesland JM, Heimdal KR, Ottestad L, Borresen AL. Prognostic significance of TP53 alterations in breast carcinoma. Br J Cancer 1993;68:540–8.[Medline]
16. Edvardsen H, Kristensen VN, Alnæs GIG, et al. Germline glutathione S-transferase variants in breast cancer: Relation to diagnosis and cuteaneous long-term adverse effects after two fractionation patterns of radiotherapy. Int J Radiation Oncology Biol Phys 2007.
17. Wiedswang G, Borgen E, Karesen R, et al. Detection of isolated tumor cells in bone marrow is an independent prognostic factor in breast cancer. J Clin Oncol 2003;21:3469–78.[Abstract/Free Full Text]
18. Naume B, Zhao H, Synnestvedt M, et al. Presence of micrometastasis in bone marrow is associated with different recurrence risk within molecular subtypes of breast cancer. Molecular Oncology 2007.
19. Gram IT, Bremnes Y, Ursin G, Maskarinec G, Bjurstam N, Lund E. Percentage density, Wolfe's and Tabar's mammographic patterns: agreement and association with risk factors for breast cancer. Breast Cancer Res 2005;7:R854–61.[CrossRef][Medline]
20. Helle SI, Ekse D, Holly JM, Lonning PE. The IGF-system in healthy pre- and postmenopausal women: relations to demographic variables and sex-steroids. J Steroid Biochem Mol Biol 2002;81:95–102.[CrossRef][Medline]
21. Garcia-Closas M, Brinton LA, Lissowska J, et al. Established breast cancer risk factors by clinically important tumour characteristics. Br J Cancer 2006;95:123–9.[CrossRef][Medline]
22. Stram DO, Leigh PC, Bretsky P, et al. Modeling and E-M estimation of haplotype-specific relative risks from genotype data for a case-control study of unrelated individuals. Hum Hered 2003;55:179–90.[CrossRef][Medline]
23. Packer BR, Yeager M, Burdett L, et al. SNP500Cancer: a public resource for sequence validation, assay development, and frequency analysis for genetic variation in candidate genes. Nucleic Acids Res 2006;34:D617–21.[Abstract/Free Full Text]
24. Schaid DJ. Evaluating associations of haplotypes with traits. Genet Epidemiol 2004;27:348–64.[CrossRef][Medline]
25. Saitou N, Nei M. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 1987;4:406–25.[Abstract]
26. Kumar S, Tamura K, Nei M. MEGA3: Integrated software for Molecular Evolutionary Genetics Analysis and sequence alignment. Brief Bioinform 2004;5:150–63.[Abstract/Free Full Text]
27. Zhong JF, Zhao Y, Sutton S, et al. Gene expression profile of murine long-term reconstituting vs. short-term reconstituting hematopoietic stem cells. Proc Natl Acad Sci U S A 2005;102:2448–53.[Abstract/Free Full Text]
28. Kaufman CK, Zhou P, Pasolli HA, et al. GATA-3: an unexpected regulator of cell lineage determination in skin. Genes Dev 2003;17:2108–22.[Abstract/Free Full Text]
29. Pai SY, Truitt ML, Ting CN, Leiden JM, Glimcher LH, Ho IC. Critical roles for transcription factor GATA-3 in thymocyte development. Immunity 2003;19:863–75.[CrossRef][Medline]
30. Finlin BS, Gau CL, Murphy GA, et al. RERG is a novel ras-related, estrogen-regulated and growth-inhibitory gene in breast cancer. J Biol Chem 2001;276:42259–67.[Abstract/Free Full Text]
31. Oh DS, Troester MA, Usary J, et al. Estrogen-regulated genes predict survival in hormone receptor-positive breast cancers. J Clin Oncol 2006;24:1656–64.[Abstract/Free Full Text]

This article has been cited by other articles:

				M.-L. Asselin-Labat, F. Vaillant, M. Shackleton, T. Bouras, G.J. Lindeman, and J.E. Visvader Delineating the Epithelial Hierarchy in the Mouse Mammary Gland Cold Spring Harb Symp Quant Biol, November 19, 2008; (2008) sqb.2008.73.020v2. [Abstract] [PDF]

				L. M. Dong, J. D. Potter, E. White, C. M. Ulrich, L. R. Cardon, and U. Peters Genetic Susceptibility to Cancer: The Role of Polymorphisms in Candidate Genes JAMA, May 28, 2008; 299(20): 2423 - 2436. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Garcia-Closas, M. Articles by Perou, C. M. Search for Related Content PubMed PubMed Citation Articles by Garcia-Closas, M. Articles by Perou, C. M.