Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Cancer Epidemiol Biomarkers Prev. Author manuscript; available in PMC 2010 November 1.
Published in final edited form as:
PMCID: PMC2783392

Variation in the FGFR2 Gene and the Effects of Postmenopausal Hormone Therapy on Invasive Breast Cancer



Breast cancer concern is a major reason for the recent marked reduction in use of postmenopausal hormone therapy, though equally effective means of controlling menopausal symptoms are lacking. Single nucleotide polymorphisms (SNPs) in the fibroblast growth factor receptor two (FGFR2) gene are substantially associated with postmenopausal breast cancer risk, and could influence hormone therapy effects.

Participants and Methods

We interrogated eight SNPs in intron 2 of the FGFR2 gene for 2166 invasive breast cancer cases from the Women's Health Initiative clinical trial and one-to-one matched controls, to confirm an association with breast cancer risk. We used case-only analyses to examine the dependence of estrogen plus progestin and estrogen-alone odds ratios on SNP genotype.


Seven FGFR2 SNPs, including six in a single linkage disequilibrium region, were found to associate strongly (p<10−7) with breast cancer risk. SNP rs3750817 (minor allele T with frequency 0.37) had an estimated `per minor allele' odds ratio of 0.78, and was not in such strong linkage disequilibrium with the other SNPs. The genotype of this SNP related significantly (p<0.05) to hormone therapy odds ratios. For estrogen plus progestin, the odds ratios (95% confidence intervals) at 0, 1, and 2 minor SNP alleles were 1.52 (1.14, 2.02), 1.33 (1.01, 1.75), and 0.69 (0.41, 1.17), while corresponding values for estrogen-alone were 0.74 (0.51, 1.09), 0.99 (0.68, 1.44), and 0.34 (0.15, 0.76).


Postmenopausal women having TT genotype for SNP rs3750817 have a reduced breast cancer risk, and appear to experience comparatively favorable effects of postmenopausal hormone therapy.

Keywords: breast cancer, genotype, gene-treatment interaction, randomized controlled trial, postmenopausal hormone therapy


The Women's Health Initiative (WHI) randomized controlled trial included evaluation of the health risks and benefits of four preventive interventions in a partial factorial design (1, 2). A total of 68,132 postmenopausal women were enrolled during 1993–98 at 40 U.S. clinical centers.

An elevated breast cancer risk, along with health risks that were judged to exceed benefits (3, 4), led to the early stopping in 2002 of the estrogen plus progestin trial component, which compared 0.625 mg/day conjugated equine estrogen plus 2.5 mg/day medroxyprogesterone acetate to placebo among 16,608 women with uterus, over a 5.6-year average intervention period. The estrogen-alone trial component, which compared the same estrogen preparation to placebo among 10,739 women who were post-hysterectomy also stopped early, in 2004, in part because of a stroke increase of similar magnitude to that seen with estrogen plus progestin (5). Health risks and benefits with estrogen-alone were approximately balanced over an average 7.1-year intervention period, and included a non-significantly lower breast cancer risk (6).

The use of postmenopausal hormone therapy, especially combined estrogen plus progestin preparations, dropped sharply in the United States (7, 8) and elsewhere following our 2002 report (3), a change in medical practice that is hypothesized to be a major factor in the reversal from an increasing to a decreasing trend in national postmenopausal breast cancer incidence rates (9, 10). The plausibility of this explanation is enhanced by our recent report (11) from the post-intervention follow-up of the WHI estrogen plus progestin trial cohort, indicating that the elevated breast cancer risk dissipated quickly, mostly within 2–3 years, following cessation of hormone treatment.

However, vasomotor symptom control remains an important quality of life issue for millions of postmenopausal women, and it is important to explore whether there are subsets of women for whom hormone therapy has neutral or favorable effects on breast cancer risk. Rather comprehensive study of demographic and clinical variables in relation to breast cancer hazard ratios in the WHI trials failed to reveal any clear interacting variables for estrogen plus progestin (4). For estrogen-alone somewhat more favorable breast cancer hazard ratios were found (6) for women having low Gail model (12) breast cancer risk score. Also, for either hormone therapy preparation, breast cancer hazard ratios appear to be comparatively higher among women who first began hormone therapy at, or soon after, the menopause based on combined analysis of WHI clinical trial and observational study data (13, 14).

Findings from recent breast cancer genome-wide association studies provide an opportunity to seek genetic factors that may relate to hormone therapy effects on breast cancer. The strongest breast cancer association to emerge from these studies (15, 16) involves single nucleotide polymorphisms (SNPs) in intron 2 of the fibroblast growth factor receptor two gene (FGFR2). For example, SNP rs2981582, with a minor allele frequency of about 40%, has been found to convey a per minor-allele increase in breast cancer risk of about 30% (15, 16). FGFR2 has been shown to transform human mammary epithelial cells (17), and blocking FGFR2 signaling inhibits breast cancer cell proliferation (18). Intron 2 includes highly conserved regions and is dense in transcription factor binding sites, including sites thought to be relevant to the estrogen receptor (19). Also, a recent case-control study reported SNP rs1219648 in FGFR2 to interact with the effect of estrogen plus progestin therapy on breast cancer risk, among women of European ancestry (20).

We tested the two FGFR2 SNPs just mentioned, as well as six others, for association with breast cancer risk in the overall WHI clinical trial cohort, and for interaction with breast cancer hazard ratios for both estrogen plus progestin and estrogen-alone, in the WHI hormone therapy trial components. The FGFR2 SNPs were included in a larger study of 9039 SNPs that were considered to have potential for association with breast cancer risk. Results for SNPs, other than the FGFR2 SNPs, will be reported separately.


Study Population, Case and Control Selection

Clinical trial enrollees were postmenopausal, in the age range 50–79, without a history of invasive breast cancer, and with no suggestion of breast cancer on baseline mammogram and clinical breast exam. Clinical outcomes were self-reported at six-month intervals. Breast cancers were confirmed (21) by review of pathology reports by local physician adjudicators, followed by adjudication at the Clinical Coordinating Center that included coding of estrogen receptor status and progesterone receptor status (positive or negative per local pathology report), histology, and extent of disease using the National Cancer Institute's Surveillance, Epidemiology and End Results system. The trial was approved by human subjects review committees at each participating institution, and each study participant provided written informed consent.

All 2242 invasive breast cancer cases developing between randomization and the originally planned end of the intervention phase of the WHI clinical trial (March 31, 2005) were considered for inclusion. Of these 2166 (96.6%) had adequate quantity and quality of DNA. Each of these cases was matched to a corresponding control without a breast cancer diagnosis during this intervention phase, on baseline age, self-reported ethnicity, participation in each clinical trial component, years since randomization, and baseline hysterectomy status.

Laboratory Methods

As mentioned above, the eight SNPs from intron 2 of the FGFR2 gene were included in a larger project involving 9039 SNPs. These SNPs were selected from two genome-wide association studies (15, 16), and from a WHI-Perlegen Sciences pooled DNA genome-wide association study (22). Genotyping and quality control methods at Perlegen, where genotyping took place, have been described (23). The average call rate for SNPs meeting quality assurance criteria was 99.8% in this project, and the average concordance rate for duplicate samples (157 pairs in dataset) was also 99.8%.

Six of the eight FGFR2 SNPs selected (rs2981582, rs1219648, rs2912774, rs2981579, rs11200014, and rs2420946) were from a single linkage disequilibrium (LD) region where much of the interest from genome-wide breast cancer association studies has focused. These SNPs have a minimum squared pairwise correlation (r2) of 0.83 among Caucasian women. A seventh SNP, rs17102287, is in the same genomic region, but is in lower LD with the other six SNPs (maximum r2 of 0.32 among Caucasian women). The eighth SNP, rs3750817, is a little distant from the other seven and has a maximum pairwise r2 of 0.47 with the other FGFR2 SNPs among Caucasian women, but falls within a conserved region of the gene (15).

Statistical Methods

SNP associations with breast cancer risk were tested using logistic regression of case versus control status on SNP genotype and potential confounding factors. Potential confounding factors included log-transformed Gail model 5-year breast cancer risk score (12), previous hormone use (indicators for <5, 5–10, and ≥10 years for each of estrogen-alone and estrogen plus progestin), log-transformed body mass index, and the first four principal components from correlation analysis of project genotype data (24). This regression model also included the previously mentioned case-control matching variables. The principal component analysis involved 4680 SNPs in low linkage disequilibrium (r2<0.2). The top four principal components were sufficient to reconstruct self-reported ethnicity, and ethnicity explained no more than 1% of the variance in any higher order principal component.

Likelihood ratio procedures were used to test associations between SNP genotype and breast cancer incidence. Principal analyses used the number of SNP minor alleles in the log-odds ratio model, giving a per-minor-allele odds ratio estimate. Corresponding two degrees of freedom (nonparametric) tests were also performed by including separate indicator variables for one, and for two, minor SNP alleles in the log-odds ratio model.

Tests for SNP interactions with the hormone therapy odds ratio were carried out using case-only data analyses (2527). This approach, which here can be expected to yield essentially unbiased and highly efficient odds ratio estimates, involves logistic regression of intervention assignment (0 - placebo; 1 - intervention) on indicator variables for heterozygous and homozygous carriers of the minor SNP allele, with constant term (or `offset') given by log {q/(1-q)}, where q is the fraction of the trial cohort assigned to active intervention in the clinical trial component. Additional analyses of this type were restricted to cases among white women; to cases occurring prior to the cessation of intervention in the hormone therapy trials; and to cases who adhered to their assigned active or placebo intervention, by including only cases taking 80% or more of assigned pills up to six months prior to breast cancer diagnosis.

A brief elaboration of the basis for the case-only methods may be helpful, because these methods have not been widely used in epidemiologic research. A hazard rate (Cox) model that stratifies on the number, z, of minor SNP alleles, and includes intervention log-hazard ratios of β0, β1, and β2 at 0, 1, and 2 minor alleles, induces a logistic regression model for intervention assignment (0 - placebo; 1 - active) for a woman diagnosed with breast cancer at time t from randomization. The logit of intervention group assignment can be written


where I(•) denotes an indicator variable, and p(z, t) is the logit of intervention assignment for women under active follow-up and without a breast cancer diagnosis at time t. Because of the randomized treatment assignment, if the study disease is rare (here breast cancer is incident in <3% of the study cohort), and follow-up is equivalent for the randomized groups, then p(z, t) is independent of both t and z and takes value log{q/1-q)}, to an excellent approximation. We refer to the case-only estimates as odds ratios, since they derive from binary logistic regression, but hazard ratio terminology is also appropriate under the above equal follow-up and rare disease conditions.


The characteristics of the women participating in the WHI clinical trial have been published (4, 6, 28). Women were postmenopausal at enrollment, with an average age of 63 years, and about 20% were of minority race/ethnicity. About two-thirds were overweight or obese. All women were without a prior breast cancer diagnosis at enrollment.

Table 1 shows significance levels for association of the eight FGFR2 SNPs with breast cancer incidence. Seven SNPs show strong evidence of association (p<10−7) in this study population while rs17102287 provides weaker evidence (p=0.0035). The six SNPs in high LD, with minor allele frequencies of 0.41 – 0.44, each have a per-allele odds ratio of about 1.3. In contrast, SNP rs17102287 has a minor allele frequency of 0.19 and a per-minor-allele estimated odds ratio of 1.18; while SNP rs3750817 has a minor allele frequency of 0.37 and a per-minor-allele odds ratio estimate of 0.78.

Table 1
Association of FGFR2 single nucleotide polymorphisms (SNPs) with invasive breast cancer incidence, based on 2166 breast cancer cases and one-to-one matched controls from the Women's Health Initiative clinical trial

Table 2 shows breast cancer odds ratio estimates (95% confidence intervals) according to the number of minor SNP alleles, for both estrogen plus progestin and estrogen-alone. Significance levels for tests of independence of odds ratios from SNP genotype are also given. SNP rs3750817 shows evidence of interaction with both the estrogen plus progestin (p=0.033) and estrogen-alone (p=0.046) odds ratio, while SNP rs2981582 shows evidence (p=0.045) of interaction with the estrogen-alone odds ratio. With eight SNPs tested, the probability of a pair of p-values as extreme as those observed is less than 8(0.033)(0.046)=0.012 for rs3750817, whereas the corresponding number for rs2981582 is 0.12. Additional case-only analyses were carried out that allowed hormone therapy odds ratios to depend simultaneously on genotype of these two SNPs. There was no evidence of estrogen plus progestin odds ratio dependence on the number of minor alleles of rs2981582 (p=0.724) after allowing for rs3750817 genotype, whereas there was some evidence of odds ratio dependence on number of rs3750817 minor alleles remained (p=0.068) after allowing for rs2981582 genotype. The corresponding significance levels for the estrogen-alone odds ratio tests were 0.063 and 0.059 respectively.

Table 2
Breast cancer odds ratio estimates for postmenopausal hormone therapy, according to the number of minor alleles of FGFR2 single nucleotide polymorphisms (SNPs)

Table 3 shows odds ratio estimates for estrogen plus progestin and estrogen-alone according to the number of minor alleles of rs3750817, with breast tumors classified on hormone receptor status and histology. Odds ratio dependencies are similar for receptor positive and receptor negative tumors, but are somewhat more pronounced for the, more frequent, receptor positive tumors.

Table 3
Postmenopausal hormone therapy and breast cancer odds ratio estimates according to number of minor alleles of SNP rs3750817, by tumor receptor status and histology

Additional analyses restricted to white women gave similar results, with p=0.025 for a test of dependence of the estrogen plus progestin odds ratio on rs3750817 based on 415 cases; and p=0.042 for estrogen-alone, based on 196 cases. Analyses restricted to cases occurring during the active intervention periods for each hormone therapy trial, or to cases occurring among women who adhered to their assigned hormone therapy, gave results similar to those shown in Tables 2 and and33.


Our nested case-control study within the WHI clinical trial provides additional strong evidence for an association between SNPs in intron 2 of the FGFR2 gene and invasive breast cancer risk among postmenopausal women.

Our principal finding is that the odds ratios for both estrogen plus progestin and estrogen-alone in the WHI hormone therapy trials depend significantly (p<0.05) on genotype for SNP rs3750817. There is a high degree of sequence homology around rs3750817, and the region is likely to be regulatory based on a density of transcription factor binding sites. In fact, two different prediction algorithms [ and] predict that the minor allele (T) of this SNP introduces a YY1 transcription factor binding site into the human sequence. YY1 has been shown to mediate breast cancer cell migration (29), and to alter the expression of the ERBB2 oncogene in primary breast tumors (30). YY1 is also a transcriptional repressor of DR5 (TNFRSF10B), implicated in the regulation of TRAIL-induced apoptosis (31).

It is not currently known whether YY1 binding mediates the observed lower breast cancer risk, or more favorable hormone therapy effects on breast cancer, among women having the TT genotype for rs3750817. Neither is it known whether other genetic factors in high linkage disequilibrium with this SNP may contribute to these observed associations. However, these associations may have clinical relevance. Specifically, from Table 2, we see that the evidence for an elevated breast cancer risk with estrogen plus progestin in the WHI randomized controlled trial derives from women having zero or one minor allele of rs3750817, while the suggestive evidence of a reduced breast cancer risk with estrogen-alone appears to localize to women having two minor SNP alleles. The two hormone therapy trials, with their non-overlapping trial cohorts, provide mutual support for the observation of more favorable hormone therapy results among women having a TT genotype for this SNP.

Even though the two hormone therapy trials provide mutually supportive results, confirmation in other settings will be valuable. In the absence of other large randomized trials of these agents, it is natural to seek confirmation in observational cohorts. We carried out a preliminary study of this type in the WHI Observational Study, a prospective cohort study among 93,676 postmenopausal women with much commonality in eligibility criteria data collection, and outcome ascertainment to the WHI clinical trial (21, 32). A subset of the breast cancer cases of European ancestry and 1–1 matched controls from this cohort were included in the first replication phase of National Cancer Institute-sponsored C-GEMS program, which recently contributed some additional SNP-breast cancer associations (33, 34), based on assessment of about 30,000 SNPs. These included six of the eight FGFR2 SNPs discussed above (all except rs2912774 and rs11200014). We carried out logistic regression analyses of case versus control status on hormone therapy and SNP genotype, with restriction to women who were using the same estrogen plus progestin or estrogen-alone preparation studied in the WHI clinical trials at baseline or were not using any hormone therapy, and with restriction to women who had a mammogram within two years prior to enrollment. Table 4 shows odds ratio estimates both for estrogen plus progestin and estrogen-alone, as a function of the number of minor alleles for the six SNPs. Note that one sees odds ratio patterns as a function of the number of rs3750817 minor alleles that weakly support those from the clinical trial, but the variations are far from significant; and the odds ratio variations are similarly far from significant also for the other FGFR2 SNPs.

Table 4
Breast cancer odds ratios for postmenopausal hormone therapy, according to the number of minor alleles of FGFR2 single nucleotide polymorphisms (SNPs) in the WHI Observational Study

However, this attempt at confirmation cannot be regarded as definitive for at least two reasons: First, the WHI OS data primarily provide information on hormone therapy effects after five or more years of use (13, 14), whereas intervention in the estrogen plus progestin and estrogen-alone trials was stopped after 5.6 and 7.1 years respectively, substantially explaining the large odds ratios in Table 4 versus Table 1. Hence, genotype interactions that pertain primarily to the first few years of hormone therapy use would be unlikely to be identified in Table 4 analyses. Secondly, case-control analyses of the type shown in Table 4, for a given number of cases, can be expected to be considerably less powerful than corresponding case-only analyses as the latter, as noted in Methods, have power that is essentially the same as would arise if genotype data were available on the entire study cohort. Also, the Table 4 analyses lose power through the need to control for a list of potential confounding factors, contributing somewhat to the wide confidence intervals. These considerations suggest that larger numbers of cases and controls, perhaps using cohort study collaborations, will be needed to identify or confirm genotype by treatment interactions of this type in a purely observational mode.

Case-only analyses have received rather little usage in genetic epidemiologic research. The suitability of an assumption of independence between genotype and exposure (here hormone therapy) cannot be usefully tested empirically (e.g., from control group data) since odds ratio biases may be substantial even though an independence assumption cannot be rejected (35). Hence, a traditional case-control analysis will typically be needed for observational study of gene-environment interactions on the odds ratios based on observational study data, as in our Table 4.

We were unable to confirm an interaction between rs1219648 and the breast cancer odds ratio for estrogen plus progestin use (20), either in the estrogen plus progestin trial cohort (Table 2), or among white women in the WHI Observational Study cohort (Table 4). However, the power to detect moderate estrogen plus progestin odds ratio variations across the alleles of this SNP in either of our datasets may be limited.

Recent studies of SNPs in intron 2 of FGFR2 in relation to breast cancer risk have focused on the region of high linkage disequilibrium that contains six of the eight SNPs studied here (the first six in Table 1). For example, functional studies (36) and multi-ethnic fine mapping efforts (37) lead to an interest in rs2981578. This may raise questions about the plausibility of rs3750817 as a potential determinant of breast cancer risk, given its r2 values of 0.47 or less in relation to the SNPs in this other region. However, rs3750817 is itself strongly related to breast cancer risk in our clinical trial cohort (p<10−7), as is also the case in the WHI Observational Study, and the functional and fine mapping studies just mentioned have not included rs3750817 or other close-by SNPs. More fundamentally, the identification of causal genetic variants in non-coding regions is challenging indeed, and our analyses of rs3750817 in relation to breast cancer risk and hormone therapy influences, in conjunction with related earlier work (15, 16, 31, 36) may contribute to a further understanding of FGFR2 regulatory influences in relation to breast cancer risk.

The strengths of this study are the randomized controlled trial design for the study of interactions of hormone therapy effects on FGFR2 genotype. This study design implies precise orthogonality between genotype and hormone treatment, justifying the highly efficient case-only data analysis method employed. Other strengths include prediagnostic blood specimens, collected and stored according to a standardized protocol, and quality controlled SNP genotyping.

Study weaknesses include incomplete knowledge of the specific genetic variant(s) that may influence hormone therapy effects on breast cancer risk, and the mechanisms of such influences. Also, the average age at enrollment was 63 years in the WHI clinical trials, with a sizeable fraction of women beyond the age when hormone therapy decisions are typically made.

In summary, SNP rs3750817 in a highly conserved region of the FGFR2 gene has been shown to be strongly associated with breast cancer risk, and it appears to interact with the effects of both estrogen plus progestin and estrogen-alone therapy on breast cancer. Women homozygous for the minor allele (T) of this SNP could plausibly choose to downweight breast cancer risk in decision-making concerning the use of these products for menopausal symptom control, though further data will be needed to confirm these SNP-hormone therapy interactions.


Decisions concerning study design, data collection and analysis, interpretation of the results, the preparation of the manuscript, or the decision to submit the manuscript for publication resided with committees comprised of WHI investigators that included NHLBI representatives.

This work was supported by the National Heart, Lung, and Blood Institute, National Institutes of Health, U. S. Department of Health and Human Services [contracts HHSN268200764314C, N01WH22110, 24152, 32100-2, 32105-6, 32108-9, 32111-13, 32115, 32118-19, 32122, 42107-26, 42129-32, and 44221]. Clinical Trials Registration: identifier: NCT00000611. The work of Drs. Prentice and Huang was partially supported by grant CA53996 from the National Cancer Institute.

FINANCIAL DISCLOSURE Dr. Chlebowski reports receiving consulting fees from AstraZeneca, Novartis, Pfizer, Eli Lilly, and Wyeth Pharmaceuticals, lecture fees from AstraZeneca, Novartis, and Abraxis, and grant support from Amgen, Eli Lilly, and Organon.


Program Office: (National Heart, Lung, and Blood Institute, Bethesda, Maryland) Elizabeth Nabel, Jacques Rossouw, Shari Ludlam, Joan McGowan, Leslie Ford, and Nancy Geller.

Clinical Coordinating Center: (Fred Hutchinson Cancer Research Center, Seattle, WA) Ross Prentice, Garnet Anderson, Andrea LaCroix, Charles L. Kooperberg, Ruth E. Patterson, Anne McTiernan; (Medical Research Labs, Highland Heights, KY) Evan Stein; (University of California at San Francisco, San Francisco, CA) Steven Cummings.

Clinical Centers: (Albert Einstein College of Medicine, Bronx, NY) Sylvia Wassertheil-Smoller; (Baylor College of Medicine, Houston, TX) Aleksandar Rajkovic; (Brigham and Women's Hospital, Harvard Medical School, Boston, MA) JoAnn E. Manson; (Brown University, Providence, RI) Charles B. Eaton; (Emory University, Atlanta, GA) Lawrence Phillips; (Fred Hutchinson Cancer Research Center, Seattle, WA) Shirley Beresford; (George Washington University Medical Center, Washington, DC) Lisa Martin; (Los Angeles Biomedical Research Institute at Harbor- UCLA Medical Center, Torrance, CA) Rowan Chlebowski; (Kaiser Permanente Center for Health Research, Portland, OR) Yvonne Michael; (Kaiser Permanente Division of Research, Oakland, CA) Bette Caan; (Medical College of Wisconsin, Milwaukee, WI) Jane Morley Kotchen; (MedStar Research Institute/Howard University, Washington, DC) Barbara V. Howard; (Northwestern University, Chicago/Evanston, IL) Linda Van Horn; (Rush Medical Center, Chicago, IL) Henry Black; (Stanford Prevention Research Center, Stanford, CA) Marcia L. Stefanick; (State University of New York at Stony Brook, Stony Brook, NY) Dorothy Lane; (The Ohio State University, Columbus, OH) Rebecca Jackson; (University of Alabama at Birmingham, Birmingham, AL) Cora E. Lewis; (University of Arizona, Tucson/Phoenix, AZ) Cynthia A Thomson; (University at Buffalo, Buffalo, NY) Jean Wactawski-Wende; (University of California at Davis, Sacramento, CA) John Robbins; (University of California at Irvine, CA) F. Allan Hubbell; (University of California at Los Angeles, Los Angeles, CA) Lauren Nathan; (University of California at San Diego, LaJolla/Chula Vista, CA) Robert D. Langer; (University of Cincinnati, Cincinnati, OH) Margery Gass; (University of Florida, Gainesville/Jacksonville, FL) Marian Limacher; (University of Hawaii, Honolulu, HI) J. David Curb; (University of Iowa, Iowa City/Davenport, IA) Robert Wallace; (University of Massachusetts/Fallon Clinic, Worcester, MA) Judith Ockene; (University of Medicine and Dentistry of New Jersey, Newark, NJ) Norman Lasser; (University of Miami, Miami, FL) Mary Jo O'Sullivan; (University of Minnesota, Minneapolis, MN) Karen Margolis; (University of Nevada, Reno, NV) Robert Brunner; (University of North Carolina, Chapel Hill, NC) Gerardo Heiss; (University of Pittsburgh, Pittsburgh, PA) Lewis Kuller; (University of Tennessee Health Science Center, Memphis, TN) Karen C. Johnson; (University of Texas Health Science Center, San Antonio, TX) Robert Brzyski; (University of Wisconsin, Madison, WI) Gloria E. Sarto; (Wake Forest University School of Medicine, Winston-Salem, NC) Mara Vitolins; (Wayne State University School of Medicine/Hutzel Hospital, Detroit, MI) Michael Simon.

Women's Health Initiative Memory Study: (Wake Forest University School of Medicine, Winston-Salem, NC) Sally Shumaker.

No other potential conflict of interest relevant to this article was reported.


1. The Women's Health Initiative Study Group Design of the Women's Health Initiative clinical trial and observational study. Control Clin Trials. 1998;19:61–109. [PubMed]
2. Prentice RL, Anderson GL. The Women's Health Initiative: lessons learned. Ann Rev Public Health. 2007;29:131–50. [PubMed]
3. Rossouw JE, Anderson GL, Prentice RL, et al. the Writing Group for the Women's Health Initiative Investigators Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. JAMA. 2002;288:321–33. [PubMed]
4. Chlebowski RT, Hendrix SL, Langer RD, et al. the WHI Investigators Influence of estrogen plus progestin on breast cancer and mammography in healthy postmenopausal women: the Women's Health Initiative randomized trial. JAMA. 2003;289:3243–53. [PubMed]
5. Anderson GL, Limacher M, Assaf AR, et al. the Women's Health Initiative Steering Committee Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. JAMA. 2004;291:1701–12. [PubMed]
6. Stefanick ML, Anderson GL, Margolis KL, et al. the WHI Investigators Effects of conjugated equine estrogens on breast cancer and mammography screening in postmenopausal women with hysterectomy. JAMA. 2006;295:1647–57. [PubMed]
7. Hersh AL, Stefanick ML, Stafford RS. National use of menopausal hormone therapy: annual trends and response to recent evidence. JAMA. 2004;291:47–53. [PubMed]
8. Haas JS, Kaplan CP, Gerstenberger EP, Kerlikowske K. Changes in the use of postmenopausal hormone therapy with the publication of clinical trial results. Ann Intern Med. 2004;140:184–8. [PubMed]
9. Clarke CA, Glaser SL, Uratsu CS, Selby JV, Kushi LH, Herrinton IJ. Recent declines in hormone therapy utilization and breast cancer incidence: clinical and population-based evidence. J Clin Oncol. 2006;24:e49–e50. [PubMed]
10. Ravdin PM, Cronin K, Howlander N, et al. A sharp decrease in breast cancer incidence in 2003 in the United States. N Engl J Med. 2007;356:1670–4. [PubMed]
11. Chlebowski RT, Kuller LH, Prentice RL, et al. the WHI Investigators Breast cancer after use of estrogen plus progestin in postmenopausal women. N Engl J Med. 2009;360:573–87. [PubMed]
12. Gail MH, Constantino JP, Bryant J, et al. Weighing the risks and benefits of tamoxifen treatment for preventing breast cancer. J Natl Cancer Inst. 1999;91:1829–46. [PubMed]
13. Prentice RL, Chlebowski RT, Stefanick ML, et al. Conjugated equine estrogens and breast cancer risk in the Women's Health Initiative clinical trial and observational study. Am J Epidemiol. 2008;167:1407–15. [PMC free article] [PubMed]
14. Prentice RL, Chlebowski RT, Stefanick ML, et al. Estrogen plus progestin therapy and breast cancer in recently postmenopausal women. Am J Epidemiol. 2008;167:1207–16. [PMC free article] [PubMed]
15. Easton DF, Pooley KA, Dunning AM, et al. Genome-wide association study identifies novel breast cancer susceptibility loci. Nature. 2007;447:1087–93. [PMC free article] [PubMed]
16. Hunter DJ, Kraft P, Jacobs KB, et al. A genome-wide association study identifies alleles in FGFR2 associated with risk of sporadic postmenopausal breast cancer. Nat Genet. 2007;39:870–4. [PMC free article] [PubMed]
17. Moffa AB, Ethier SP. Differential signal transduction of alternatively spliced FGFR2 variants expressed in human mammary epithelial cells. J Cell Physiol. 2007;210:720–31. [PubMed]
18. Koziczak M, Holbro T, Hynes NE. Blocking of FGFR signaling inhibits breast cancer cell proliferation through downregulation of D-type cyclins. Oncogene. 2004;23:3501–8. [PubMed]
19. Carroll JS, Meyer CA, Song J, et al. Genome-wide analysis of estrogen receptor binding sites. Nat Genet. 2006;38:1289–97. [PubMed]
20. Rebbeck TR, DeMichele A, Train TV, et al. Hormone-dependent effects of FGFR2 and MAP3K1 in breast cancer susceptibility in a population-based sample of postmenopausal African-American and European-American women. Carcinogenesis. 2009;30:269–74. [PMC free article] [PubMed]
21. Curb JD, McTiernan A, Heckbert SR, et al. the WHI Morbidity and Mortality Committee Outcomes ascertainment and adjudication methods in the Women's Health Initiative. Ann Epidemiol. 2003;13:S122–8. [PubMed]
22. Prentice RL, Qi L. Aspects of the design and analysis of high-dimensional SNP studies for disease risk estimation. Biostatistics. 2006;7:339–54. [PubMed]
23. Saccone SS, Hinrich AL, Saccone NL, et al. Cholinergic nicotinic receptor genes implicated in a nicotine dependence association study targeting 348 candidate genes with 3713 SNPs. Hum Mol Genet. 2006;16:36–49. [PMC free article] [PubMed]
24. Price AL, Patterson NJ, Plenge RM, Weinblatt ME, Shadick NA, Reich D. Principal components analysis corrects for stratification in genome-wide association studies. Nat Genet. 2006;38:904–9. [PubMed]
25. Self SG, Longton G, Kopecky KJ, Liang KY. On estimating HLA/disease association with application to a study of aplastic anemia. Biometrics. 1991;47:53–61. [PubMed]
26. Piegorsch WW, Weinberg CR, Taylor JA. Non-hierarchical logistic models and case-only designs for assessing susceptibility in population-based case-control studies. Stat Med. 1994;13:153–62. [PubMed]
27. Vittinghoff E, Bauer DC. Case-only analysis of treatment-covariate interactions in clinical trials. Biometrics. 2006;62:769–76. [PubMed]
28. Prentice RL, Caan B, Chlebowski RT, et al. Low-fat dietary pattern and risk of invasive breast cancer: The Women's Health Initiative Randomized Controlled Dietary Modification Trial. JAMA. 2006;295:629–42. [PubMed]
29. Lee BC, Lee TH, Zagozdzon R, Avraham S, Usheva A, Avraham HK. Carboxyl-terminal Src kinase homologous kinase negatively regulates the chemokine receptor CXCR4 through YY1 and impairs CXCR4/CXCL12 (SDF-1alpha)-mediated breast cancer cell migration. Cancer Res. 2005;65:2840–5. [PubMed]
30. Allouche A, Nolens G, Tancredi A, et al. The combined immunodetection of AP-2alpha and YY1 transcription factors is associated with ERBB2 gene overexpression in primary breast tumors. Breast Cancer Res. 2008;10:R9. [PMC free article] [PubMed]
31. Baritaki S, Suzuki E, Umezawa K, et al. Inhibition of Yin Yang 1-dependent repressor activity of DR5 transcription and expression by the novel proteasome inhibitor NPI-0052 contributes to its TRAIL-enhanced apoptosis in cancer cells. J Immunol. 2008;180:6199–210. [PubMed]
32. Langer RD, White E, Lewis CE, Kotchen JM, Hendrix SL, Trevisan M. The Women's Health Initiative Observational Study. Baseline characteristics of participants and reliability of baseline measures. Ann Epidemiol. 2003;13:S107–21. [PubMed]
33. Thomas G, Jacobs KG, Kraft P, et al. A multistage genome-wide association study in breast cancer identifies two novel risk alleles at 1p11.2 and 14q24.1 (RAD51L1) Nat Genet. 2009 March 29; Epub ahead of print.
34. Ahmed S, Thomas G, Ghoussaini M, et al. Newly discovered breast cancer susceptibility loci on 3p24 and 17q23.2. Nat Genet. 2009;41:585–90. Epub ahead of print. [PMC free article] [PubMed]
35. Albert PS, Ratnasinghe D, Tangrea J, Wacholder S. Limitations of the case-only design for identifying gene-environment interactions. Am J Epidemiol. 2001;154:687–93. [PubMed]
36. Meyer KB, Maia A-T, O'Reilly M, et al. Allele-specific up-regulation of FGFR2 increases susceptibility to breast cancer. PLOS Biol. 2008;6:1098–1103. [PMC free article] [PubMed]
37. Udler MS, Meyer KB, Pooley KA, et al. FGFR2 variants and breast cancer risk: fine-scale mapping using African American studies and analysis of chromatin conformation. Hum Mol Genet. 2009;18:1692–1703. [PMC free article] [PubMed]