CEBP CTRC-AACR San Antonio Breast Cancer Symposium 2008 Conference on Cancer Prevention - Washington, D.C.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Warren, R.
Right arrow Articles by Easton, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Warren, R.
Right arrow Articles by Easton, D.
Cancer Epidemiology Biomarkers & Prevention Vol. 15, 1502-1508, August 2006
© 2006 American Association for Cancer Research

Associations among Mammographic Density, Circulating Sex Hormones, and Polymorphisms in Sex Hormone Metabolism Genes in Postmenopausal Women

Ruth Warren1, Jane Skinner5, Evis Sala1, Erika Denton6, Mitch Dowsett7, Elizabeth Folkerd7, Catherine S. Healey2, Alison Dunning2, Deborah Doody7, Bruce Ponder2, Robert N. Luben3, N.E. Day3 and Douglas Easton4

1 Department of Radiology, University of Cambridge, Addenbrooke's Hospital; 2 Cancer Research UK, Department of Oncology, University of Cambridge; 3 European Prospective Investigation into Cancer and Nutrition and 4 Cancer Research UK, Genetic Epidemiology Group, University of Cambridge, Strangeways Research Laboratory, Cambridge, United Kingdom; 5 Population Health Group, School of Medicine, Health Policy and Practice, University of East Anglia; 6 Department of Radiology, Norfolk and Norwich Hospital, Norwich, United Kingdom; and 7 Academic Department of Biochemistry, Royal Marsden Hospital, London, United Kingdom

Requests for reprints: Ruth Warren, Cambridge Breast Unit, Addenbrooke's Hospital, Cambridge CB2 2QQ, United Kingdom. Phone: 44-1223-586959; Fax: 44-1223-217886. E-mail: rmlw2{at}cam.ac.uk


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Statistical Methods
 Results
 Discussion
 Conclusion
 References
 
Mammographic density and serum sex hormone levels are important risk factors for breast cancer, but their associations with one another are unclear. We studied these phenotypes, together with single nucleotide polymorphisms (SNP) in genes related to sex hormone metabolism, in a cross-sectional study of 1,413 postmenopausal women from the European Prospective Investigation into Cancer and Nutrition-Norfolk. All women were >1 year postmenopausal and had not taken hormone replacement therapy for >3 months before sampling. Serum levels of 7 sex hormones [estradiol, testosterone, sex hormone-binding globulin (SHBG), androstenedione, 17-OH-progesterone, estrone, and estrone sulfate] and 15 SNPs in the CYP17, CYP19, EDH17B2, SHBG, COMT, and CYP1B1 genes were studied. Mammograms nearest in time to the blood sampling were identified through the national breast screening program and visually assessed by three radiologists using the Boyd six-category and Wolfe four-category scales. We found a weak positive association between mammographic density and SHBG levels (P = 0.09) but no association with any other hormones. None of the SNPs, including those shown previously to be associated with estradiol or SHBG, showed significant associations with density. We conclude that mammographic density is largely independent of postmenopausal steroid hormone levels, indicating that these risk factors have, to a large extent, an independent etiology and suggesting that they may be independent predictors of breast cancer risk. (Cancer Epidemiol Biomarkers Prev 2006;15(8):1502-08)


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Statistical Methods
 Results
 Discussion
 Conclusion
 References
 
The degree of density of the breast on a mammogram is a powerful marker of breast cancer risk. Epidemiologic studies have estimated the breast cancer risk in women with the densest breasts to be four to six greater than those with the most lucent (1). Mammographic density is a measure of the relative amount of stroma and glandular tissue (which appear white on the mammogram) to fat (which is translucent). The etiology of mammographic density remains poorly understood, but various evidence suggests that exposure to sex steroid hormones may be an important factor. Density declines with age over the period of the menopause, to a greater extent than at other ages (2). Moreover, density increases in women on hormone replacement therapy (HRT; ref. 3) and is reduced in women taking tamoxifen (4).

Exposure to endogenous estrogen is also an important breast cancer risk factor. A reanalysis of nine studies using data from 663 postmenopausal women who developed breast cancer and 1,765 who did not showed that women with the highest quintile of estrogen were associated with a relative risk [95% confidence interval (95% CI)] of 2.00 (1.47-2.71) when compared with the lowest quintile. Women with the highest quintile of free estradiol had a relative risk (95% CI) of 2.58 (1.76-3.78) when compared with the lowest quintile and the magnitudes of risk associated with the other estrogens and with the androgens were similar. Sex hormone-binding globulin (SHBG) was associated with a decreased breast cancer risk (Ptrend = 0.041; ref. 5).

Given the above observations, one might then expect mammographic density to be related to serum estradiol level. In a previous small study, Boyd et al. (6) found that total estradiol and progesterone levels were not related to percent density in either premenopausal or postmenopausal women. In postmenopausal women, free estradiol (negatively) and SHBG (positively) were significantly related to percent density. More recently other groups have published their findings as follows: Greendale et al. reported from the Postmenopausal Estrogen/Progestin Interventions study and concluded that there was a significant relationship between mammographic density and estrone, estradiol, and free estradiol (7); Aiello et al. concluded that there were no significant associations in never users of HRT (8); Noh et al. noted a relationship of borderline significance between progesterone and breast density after adjusting for body mass index (BMI; ref. 9); Tamimi et al. reported on 540 postmenopausal participants in the Nurses' Health Study and concluded that mammographic density is independent of sex hormone levels (10). Each of these studies uses a different list of hormones with a common core and some additional measures.

It is an associated question to seek a relationship between mammographic density and the genetic polymorphisms that determine sex hormone metabolism. For this question, there is a limited literature to date. Haiman et al. examined whether polymorphisms in genes involved in steroid hormone biosynthesis and metabolism were related to density (11). In 2005, the same group conducted a further study of women from a randomized controlled trial of estrogen therapy (12). Maskarinec et al. (13) examined premenopausal women for several polymorphisms. All the listed studies of hormones and polymorphisms used an interactive thresholding method of measuring density.

To evaluate these associations further, we have conducted a much larger study of serum hormones and mammographic density in postmenopausal women based on the European Prospective Investigation into Cancer and Nutrition (EPIC) study (14, 15). This series has been used previously to study associations between seven sex steroid hormone levels and single nucleotide polymorphisms (SNP) in genes involved in steroid hormone metabolism and synthesis (16). Highly significant associations between estradiol levels and two SNPs in CYP19 and between SHBG levels and SNPs in SHBG were identified, although the proportion of the variance in levels accounted for by these SNPs was small. If mammographic density is related to hormone exposure, these SNPs would be strong candidates to be associated with density, and we have also evaluated these associations.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Statistical Methods
 Results
 Discussion
 Conclusion
 References
 
The subjects included in this study were selected from the subcohort of the EPIC-Norfolk described previously by Dunning et al. (16). Briefly, from the 25,000 participants in the EPIC-Norfolk cohort, we selected women who were >55 years old, >1 year from their last menstrual period, and had not taken HRT for >3 months before blood sampling. Ethical approval was obtained from the Norwich Local Research Ethics Committee (LREC 98CN01). All study participants have given written, informed consent. Data on measured height and weight were obtained at the EPIC-Norfolk second health check. Alcohol intake (units/wk), HRT use, smoking status, number of children, and age at menopause, at menarche, and at birth of first child were derived from questionnaire data.

The measurements of circulating hormones and genotyping of SNPs on these subjects have been described previously (16). Briefly, all subjects with sufficient available plasma collected at the second health check had hormone measurements made in a sequential manner in the order: estradiol, testosterone, SHBG, androstenedione, 17-OH-progesterone, estrone, and estrone sulfate. Subjects with available DNA (extracted from an EDTA whole blood sample taken at second health check) had SNP genotypes determined using Taqman technology. DNA was arrayed in 96-well plates, which also contained empty wells (as negative controls) and duplicated samples (as controls for reproducibility). Fifteen SNPs in six hormone metabolism genes were assayed and passed the quality controls: COMT IVS2 g-a (rs6269), COMT M158V (rs4680), COMT exon3 t-c (rs4633), COMT exon4 c-g (rs4818), CYP17 5' untranslated region t-34c (rs743572), CYP17 c198t (rs6162), CYP17 g255t (rs6163), CYP19 3' untranslated region t-c (rs10046), CYP19 IVS4 [TCT]+/–, CYP1B1 A119S (rs1056827), CYP1B1 R48G (rs10012), CYP1B1 V432L (rs1056836), EDH17B2 S313G (rs8191245), SHBG 5' untranslated region g-a (rs1799941), and SHBG D356N (rs6269).

Mammograms and blood samples were available for 1,590 women who met the eligibility criteria. Of these, 16 were excluded, as they did not meet the criteria for postmenopausal women (defined as estradiol ≤150 pmol/L and testosterone ≤3.5 nmol/L). A further 92 of the samples assayed did not have sufficient plasma or serum to complete the estradiol or testosterone measurements; these were also excluded. Assays for the other hormones were done sequentially if sufficient plasma or serum remained, so that the sample size varies between the different hormone analyses. Data on parity, cigarette smoking status, or menopausal age, which were potential confounding factors, including in the regression analyses, were missing for 15 women and these women were therefore excluded from all analyses. Thirty-two women had reported HRT use in the 3 months before the mammogram was taken. A further 23 who had been diagnosed with breast cancer in the 5 years before the health check were also excluded, as they may have been taking tamoxifen or other antiestrogens. After these exclusions, there were 1,413 women who could be included in at least one of the analyses of mammographic density and hormone levels. No genotyping results were available for 127 of these women, and the genotype analyses are based on a maximum of 1,286 women.

Mammograms were extracted from records of the Norfolk and Norwich Breast Screening Service. Mammographic studies were undertaken as part of the UK National Health Service Breast Screening Program, in which women are screened every 3 years by one or two view mammography between ages 50 and 64 years. Density readings were made from film for each study available for each woman, in sequential date order, recording a score for right and left sides, which takes account of mediolateral oblique and craniocaudal views, where available, or mediolateral oblique alone when that was the full study. Over the relevant period, the UK National Health Service Breast Screening Program undertook two-view mammography for the prevalent screen and one-view mammography for incident screens. For the purpose of this analysis, mammograms taken closest in time to the date of the blood draw were used. Mammograms were digitized for future potential automated analysis of density.

Breast density was assessed by three experienced radiologists using both Wolfe four-category (17) and Boyd six-category (1) scales. Each breast was scored separately. Analyses presented here are based on the Boyd score. A summary density estimate for a given time point was then obtained as follows. First, the scores for each observer were represented by the midpoint of the relevant density category (0%, 5% for 0-10%, 17.5% for 10-25%, 37.5% for 25-50%, 62.5% for 50-75%, and 87.5% for 75-100%). A score for each breast was then obtained as the mean score over the three readers. Where some readers failed to score a particular image, the mean was taken over the available scores. Results for the two breasts were then averaged to give a summary density value. If results were available for only a single breast, the mean for this breast was used as the summary value.


   Statistical Methods
Top
 Abstract
 Introduction
 Materials and Methods
 Statistical Methods
Results
 Discussion
 Conclusion
 References
 
Data were analyzed using the Stata 8.2 software package (Stata Corp., College Station, TX). After the exclusions described above, sample sizes varied from 259 to 1,413 for the hormone analyses and from 616 to 1,286 for the SNP analyses. There were no associations between breast density and the presence or absence of measurements of any of the hormones or SNP genotypes.

Multiple linear regression was used to investigate the association between breast density and circulating hormone levels and SNP genotypes after adjustment for other covariates. Because the distribution of estimated percent dense breast tissue was positively skewed (skewness = 0.86), a square root transformation of percent breast density was used. The distribution of the residuals from the linear regression models was checked using various diagnostic plots and statistics and not found to be significantly nonnormal. The following covariates were included in the model: sextiles of BMI, parity (0, 1, 2, 3, 4+), cigarette smoking status (current, former, or never), age at menarche (9-11, 12-13, 14+) and years since menopause (<5, 5-9, 10-14, 15-19, 20+), all treated as categorical variables. When included in a multivariate model, these covariates were all associated with density at a 10% significance level. Age in years was also strongly associated with density. However, this was not included in the models as it was colinear with years since menopause, which was felt to be the more biologically relevant measure.

Density by quintiles of hormone levels and by SNP genotypes was summarized as unadjusted means and least square means adjusted for other covariates. For comparability with the regression analyses, estimates were based on the square root-transformed densities but squared to give values on the original density scale (i.e., they are 0.5th power means). For the analyses of the associations with SNPs, both 1 and 2 degrees of freedom (df) tests were carried out. The 1 df test was based on the trend in density with dose of the minor allele and was derived by fitting a single covariate for each SNP based on the dose of the rarer allele (0, 1, or 2). The 2 df test was based on treating genotype as a 3-level categorical variable and estimating variables for the heterozygotes and rare homozygotes, with common homozygotes being the baseline. In both cases, likelihood ratio tests were used to compare the fitted model with the null model.

Repeatability of assessments between each pair of readers was assessed using weighed {kappa} statistics using the standard weighting 1 – |i – j| / (k – 1), where i and j are the ratings of the two observers and k is the number of possible ratings. {kappa} varied between 0.59 and 0.71 for the Wolfe score and between 0.52 and 0.68 for the Boyd score.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Statistical Methods
 Results
Discussion
 Conclusion
 References
 
Table 1 summarizes the demographic characteristics of the study population and the mean and SD of each of the hormone levels. Table 2 shows the mean breast density by categories of potential confounding factors. As expected, density was negatively associated with BMI, age, parity, and years since menopause and positively associated with age at menarche. Age at birth of first child was associated with breast density in a univariate model but not in the adjusted model. Cigarette smoking status was not associated with breast density in a univariate model but showed some evidence of an association in the multivariate analysis. Alcohol consumption was not significantly associated with breast density but rose with increasing alcohol use in line with previous publications (18). Time since stopping HRT and family history of breast cancer were not significantly associated with density in either model, but there were few women who had ever taken HRT or with a family history of breast cancer.


View this table:
[in this window]
[in a new window]
  		Table 1. Characteristics of study subjects and circulating hormone levels

 

View this table:
[in this window]
[in a new window]
  		Table 2. Adjusted and unadjusted mean % breast density at screening nearest to second health check by potential confounding factors (n = 1,413)

 
Unadjusted and adjusted mean breast density levels by quintiles of hormone levels are given in Table 3 together with the corresponding variable estimates in the linear regression analysis. Density was inversely related to levels of estradiol, testosterone, estrone, and estrone sulfate. All these associations, however, were nonsignificant after adjustment for the potential confounders. There was weak evidence that density increased with increasing SHBG levels after adjustment for confounding factors (P = 0.09). The estimated effect would imply that a 50 nmol/L increase in SHBG levels (approximately the difference between 10th and 90th percentiles of SHBG levels) would correspond to an increase in breast density of ~2% (e.g., 20-22%).


View this table:
[in this window]
[in a new window]
  		Table 3. % Breast density by circulating hormone quintile and results of multiple regression of square root of % dense on circulating hormone levels

 
Table 4 summarizes the mean density by genotype for each SNP. Of the 16 SNPs, only CYP19A1 (rs10046) showed an association significant at the 5% level (P = 0.05 based on the 1 df test). None of the 2 df tests were significant.


View this table:
[in this window]
[in a new window]
  		Table 4. Mean % dense breast by SNP genotype

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Statistical Methods
 Results
 Discussion
Conclusion
 References
 
Our study failed to show any association between levels of most sex hormones and mammographic density. Among the hormones studied (estradiol, testosterone, SHBG, androstenedione, 17-OH-progesterone, estrone, and estrone sulfate), the only one for which there was a possible relationship was SHBG, which was associated with breast density at a 10% significance level. Even this association, however, was small in magnitude. In addition, when SHBG level was considered in quintiles, no monotonic relationship between density and SHBG levels was apparent. Furthermore, an extremely strong association between density and SHBG levels was apparent before adjustment that was largely eliminated by adjustment for confounding factors. This suggests that any association between SHBG and density may reflect some uncontrolled confounding. Based on our results, we estimate that variation in SHBG levels would explain <1% of the variance in breast density.

These findings can be compared with those of Boyd et al. (6), a study that included 189 postmenopausal women. He found that, after adjustment for other risk factors, serum levels of total estradiol and progesterone were unrelated to density but that free estradiol (negatively) and SHBG (positively) were significantly related to density in postmenopausal women (although not in premenopausal women). The magnitudes of effect in the two studies can only be compared crudely, because the studies used a different measure of density and different covariates and we used a square root transformation of density. However, the estimated regression coefficient for SHBG in the Boyd et al. study (0.02) was greater than in our study (~0.01), allowing for the transformation. This difference may simply reflect the play of chance or might reflect differences in the density measurement or confounding factors. The more recent articles on this subject are also of interest. The Greendale et al. analysis in 404 women from the Postmenopausal Estrogen/Progestin Interventions study show different findings in that there was a significant association of density with estrone, estradiol, and free estradiol, which persisted after adjustment. Only 197 of the women were never users of HRT, and the extent to which persistent effects from exogenous hormones may be implicated is hard to assess (7). By contrast, Tamimi et al. in 590 women from the Nurses' Health Study had findings similar to ours with mammographic density independent of sex hormone levels (10). Aiello et al. confined his study to a small number (88) of overweight postmenopausal women but found no associations in never users of HRT (8), and Noh et al. studied just 204 premenopausal women and found an association between progesterone and breast density borderline after adjusting. Of the studies of polymorphisms related to hormone metabolism, both the earlier study by Haiman et al. (11) and the later, larger one (for both number of subjects and number of SNPs; ref. 12) take the context of hormone therapy use and so are not comparable. Maskarinec et al. (13) found in 328 healthy women that the low-activity COMT and CYP1A2 variant alleles were weakly related to lower percent mammographic density after adjustment for age, ethnicity, BMI, and reproductive variables in premenopausal women only. They found no significant associations between breast density and the variant alleles for CYP1A1, CYP1B1, and CYP17.

Our study therefore is the largest of all studies in postmenopausal women and is of interest because it considers both hormones and SNPs related to sex hormone metabolism. It is substantially larger than the one by Boyd and is confined to women who are verified as postmenopausal and have not taken HRT for the previous 3 months. Because the mean age of our women is 65.2 years, we have only 341 women ages 55 to 59 years and only 73 women <5 years from the menopause, the mammographic screening program only serves women up to age 65 years, and the mean time between mammography and the blood draw is 2.2 years, it is unlikely that we have included any women who are premenopausal. HRT ever users represent only 18% of our cohort (252 of 1,413), with only 71 (5%) <3 years from cessation. Our study should therefore have greater power than all previous studies to show associations or to exclude chance associations.

Unlike all the studies quoted above, which use interactive thresholding methods of density determination, our study uses visual readings by six-category scale of breast density and has combined the readings of three experienced radiologists expert in this type of density assessment. The method of combining the readings results in a score, which can be used as a continuous variable or categorized. The {kappa} statistics comparing the readings gave a value of between 0.52 and 0.71 (moderate to substantial agreement) for the different readings. This can be compared with correlations of 0.94 in the Boyd et al. methodology article (1). All three readers scored all films (2-4 per woman); therefore, if the cut points between the six categories are being set differently between readers, this will not necessarily alter the rankings or affect the comparisons. These visual readings have been used in many other studies and have been shown to be effective predictors of breast cancer risk (1), and there is as yet no standard method of density estimation for mammography.

Because our study uses a prospective cohort from EPIC, the blood tests and mammograms do not necessarily coincide closely, and this makes a limitation. However, when the analyses were restricted to women whose blood tests were taken within 2 years of the mammogram, the results for all of the hormones were similar. The association between density and SHBG was not significant (ß coefficient = 0.0022; 95% CI, –0.0042 to 0.0087; n = 880). Restricting the analysis further to women with blood tests within 6 months, for example, of the mammogram was not possible, as the sample size would be too small (n = 287).

Women who had reported that they were on HRT in the 3 months before blood sampling or to the date of the mammogram were not included in this study. Restricting the analysis to women who had never taken HRT, the association between SHBG and breast density was slightly stronger (ß coefficient = 0.0061; 95% CI, 0.0005-0.0117; n = 1,157; P = 0.03).

Only one of the SNPs, CYP19 ex10, showed a significant association with density (P = 0.05, 1 df test), with the T allele being associated with higher density. Clearly, this association may simply be a type I error. However, it is interesting to note that this SNP also showed a significant association with both postmenopausal estradiol levels, with an increase of ~1 pmol/L per T allele, and estradiol/testosterone ratio. In a previous small study (396 women), Haiman et al. found no consistent relationship between SNPs in CYP17 (T27C), COMT (Val158Met), 17HSDB1 (Ser312Gly), and 3HSDB1 (Asn367Thr). Among current HRT users, Haiman et al. found greater breast density in the women homozygous for the COMT Met allele than in those with the Val/Val genotype (11). We cannot directly address this question because HRT users are largely excluded from our study. However, we found no evidence for an association between COMT V158M genotype and breast density overall or any association with estradiol levels.

Considering the known effects of exogenous hormones and antiestrogens on breast density and the observation that variations in density probably occur in relationship to the menopause, it is of interest that we have not shown that mammographic density is related to endogenous sex hormones in this cohort of postmenopausal women. This parallels the findings of Tamimi et al. (10) and Aiello et al. (8) and differs from the findings of Greendale et al. (7). HRT has been shown by many authors to affect breast density in some individuals and with some preparations (e.g., ref. 19). This association is most consistently found with combined estrogen and progesterone preparations (20, 21). By contrast, tamoxifen, which has antiestrogenic effects, causes reduction in breast density (4, 22). It is therefore essential that the population chosen for our study should be free from use of either of these groups of pharmaceuticals for the results to be trustworthy. Even after cessation of HRT, there may be residual effects from recent use, but all our women had ceased HRT for >3 months at least and only 71 (5%) had used it within the previous 3 years. We do not have details of the nature of the preparations or the preceding length of use in the ever users. Analysis using only never users of HRT made no difference to our findings. We excluded any woman who had been treated for breast cancer in the last 5 years but considered that the use of tamoxifen was unlikely after that interval, in line with practice in the United Kingdom.

Notwithstanding our results, it is clear that mammographic density is related to hormonal exposure, because density is increased on HRT use and is reduced by tamoxifen. The reasons for this discrepancy are unclear. It may be that the variation in postmenopausal estradiol levels is too small to have a detectable effect on mammographic density. In this respect, it may be that density is more strongly related to lifetime exposure and that postmenopausal levels have little effect. Moreover, although mammographic density is related to HRT use, there is little association with replacement therapy involving estrogen alone (20, 21, 23). In studies of postmenopausal women, variation in estrogen is being evaluated in the presence of very low levels of progesterone. Thus, one potential explanation is that the reduction in breast density in postmenopausal women is related more to absence of significant progesterone than to any relationship with estrogen levels.

A further consideration is the effect of high BMI, which is associated with low breast density (24) but high estrogen (24). The confounding effect of BMI on breast density may offset any tendency for high estrogen to give rise to high breast density.

The suggestion of an association with SHBG found in this study and also by Boyd et al. is also curious, because SHBG is negatively related to the levels of active free estradiol and negatively associated with breast cancer risk. It is possible that this reflects some complex interaction involving BMI. BMI shows a strong inverse association with breast density, and SHBG is also related to BMI (24).

There is strong evidence that mammographic density is largely inherited. A large twin study from North America and Australia showed that mammographic density has a major genetic component, with 63% (95% CI, 59-67) of the variance being attributable to additive genetic factors (25). No major genes underlying breast density have been identified. Although we found no evidence for associations between mammographic density and SNPs in hormone metabolism genes, we cannot rule out a contribution of these genes because we have not comprehensively assessed all common variants in these genes by tagging. Perhaps more likely, however, is that density is determined by genes in other pathways. If such genes could be found, they would improve our understanding of breast density and may also provide new genetic markers for breast cancer.


   Conclusion
Top
 Abstract
 Introduction
 Materials and Methods
 Statistical Methods
 Results
 Discussion
 Conclusion
References
 
The results of our study indicate that genes in the steroid hormone pathway are unlikely to be related to density and suggest that other genetic mechanisms may be important. The results also suggest that hormone levels and mammographic density are potentially independent predictors of breast cancer risk. This, however, needs evaluation in further prospective studies.


   Acknowledgments
 
We thank the women who participated in the EPIC study, thereby making their results available to us, and the staff of the Norfolk and Norwich Breast Screening Service for allowing us access to their mammographic archive.


   Footnotes
 
Grant support: Royal College of Radiologists; Medical Research Council UK, Cancer Research UK, European Union, Stroke Association, British Heart Foundation, Department of Health, Food Standards Agency, and Wellcome Trust (EPIC-Norfolk); Cancer Research UK (C.S. Healey, A. Dunning, D. Easton, and B. Ponder); and Institute of Cancer Research (M. Dowsett).

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.

Received 10/21/05; revised 5/ 7/06; accepted 5/23/06.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Statistical Methods
 Results
 Discussion
 Conclusion
 References
 

  1. Boyd NF, Byng JW, Jong RA, et al. Quantitative classification of mammographic densities and breast cancer risk: results from the Canadian National Breast Screening Study. J Natl Cancer Inst 1995;87:670–5.[Abstract/Free Full Text]
  2. Boyd N, Martin L, Stone J, Little L, Minkin S, Yaffe M. A longitudinal study of the effects of menopause on mammographic features. Cancer Epidemiol Biomarkers Prev 2002;11:1048–53.[Abstract/Free Full Text]
  3. Rutter CM, Mandelson MT, Laya MB, Seger DJ, Taplin S. Changes in breast density associated with initiation, discontinuation, and continuing use of hormone replacement therapy. JAMA 2001;285:171–6.[Abstract/Free Full Text]
  4. Cuzick J, Warwick J, Pinney E, Warren RM, Duffy SW. Tamoxifen and breast density in women at increased risk of breast cancer. J Natl Cancer Inst 2004;96:621–8.[Abstract/Free Full Text]
  5. Endogenous sex hormones and breast cancer in postmenopausal women: reanalysis of nine prospective studies. J Natl Cancer Inst 2002;94:606–16.[Abstract/Free Full Text]
  6. Boyd NF, Stone J, Martin LJ, et al. The association of breast mitogens with mammographic densities. Br J Cancer 2002;87:876–82.[CrossRef][Medline]
  7. Greendale G, Palla S, Ursin G, et al. The association of endogenous sex steroids and sex steroid binding proteins with mammographic density: results from the Postmenopausal Estrogen/Progestin Interventions Mammographic Density Study. Am J Epidemiol 2005;162:826–34.[Abstract/Free Full Text]
  8. Aiello E, Tworoger S, Yasui Y, et al. Associations among circulating sex hormones, insulin-like growth factor, lipids, and mammographic density in postmenopausal women. Cancer Epidemiol Biomarkers Prev 2005;14:1411–7.[Abstract/Free Full Text]
  9. Noh J, Maskarinec G, Pagano I, Cheung L, Stanczyk F. Mammographic densities and circulating hormones: a cross-sectional study in premenopausal women. Breast 2006;15:20–8.[CrossRef][Medline]
  10. Tamimi R, Hankinson S, Colditz G, Byrne C. Endogenous sex hormone levels and mammographic density among postmenopausal women. Cancer Epidemiol Biomarkers Prev 2005;14:2641–7.[Abstract/Free Full Text]
  11. Haiman CA, Bernstein L, Berg D, Ingles SA, Salane M, Ursin G. Genetic determinants of mammographic density. Breast Cancer Res 2002;4:R5.[CrossRef][Medline]
  12. Lord S, Mack W, Van Den Berg D, et al. Polymorphisms in genes involved in estrogen and progesterone metabolism and mammographic density changes in women randomized to postmenopausal hormone therapy: results from a pilot study. Breast Cancer Res 2005;7:336–44.
  13. Maskarinec G, Lurie G, Williams A, Le Marchand L. An investigation of mammographic density and gene variants in healthy women. Int J Cancer 2004;112:683–8.[CrossRef][Medline]
  14. Day N, Oakes S, Luben R, et al. EPIC-Norfolk: study design and characteristics of the cohort. European Prospective Investigation of Cancer. Br J Cancer 1999;80 Suppl 1:95–103.
  15. Slimani N, Kaaks R, Ferrari P, et al. European Prospective Investigation into Cancer and Nutrition (EPIC) calibration study: rationale, design and population characteristics. Public Health Nutr 2002;5:1125–45.[CrossRef][Medline]
  16. Dunning AM, Dowsett M, Healey CS, et al. Polymorphisms associated with circulating sex hormone levels in postmenopausal women. J Natl Cancer Inst 2004;96:936–45.[Abstract/Free Full Text]
  17. Wolfe JN. Breast patterns as an index of risk for developing breast cancer. Am J Roentgenol 1976;126:1130–7.[Abstract]
  18. Maskarinec G, Takata Y, Pagano I, Lurie G, Wilkens L, Kolonel L. Alcohol consumption and mammographic density in a multiethnic population. Int J Cancer 2006;118:2579–83.[CrossRef][Medline]
  19. Persson I, Thurfjell E, Bergstrom R, Holmberg L. Hormone replacement therapy and the risk of breast cancer. Nested case-control study in a cohort of Swedish women attending mammography screening. Int J Cancer 1997;72:758–61.[CrossRef][Medline]
  20. Greendale GA, Reboussin BA, Slone S, Wasilauskas C, Pike MC, Ursin G. Postmenopausal hormone therapy and change in mammographic density. J Natl Cancer Inst 2003;95:30–7.[Abstract/Free Full Text]
  21. Vachon CM, Sellers TA, Vierkant RA, Wu FF, Brandt KR. Case-control study of increased mammographic breast density response to hormone replacement therapy. Cancer Epidemiol Biomarkers Prev 2002;11:1382–8.[Abstract/Free Full Text]
  22. Brisson J, Brisson B, Cote G, Maunsell E, Berube S, Robert J. Tamoxifen and mammographic breast densities. Cancer Epidemiol Biomarkers Prev 2000;9:911–5.[Abstract/Free Full Text]
  23. Lundstrom E, Wilczek B, von Palffy Z, Soderqvist G, von Schoultz B. Mammographic breast density during hormone replacement therapy: effects of continuous combination, unopposed transdermal and low-potency estrogen regimens. Climacteric 2001;4:42–8.[Medline]
  24. Sala E, Warren R, McCann J, Duffy S, Luben R, Day N. High-risk mammographic parenchymal patterns and anthropometric measures: a case-control study. Br J Cancer 1999;81:1257–61.[CrossRef][Medline]
  25. Boyd NF, Dite GS, Stone J, et al. Heritability of mammographic density, a risk factor for breast cancer. N Engl J Med 2002;347:886–94.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Am J EpidemiolHome page
L. E. Kelemen, V. S. Pankratz, T. A. Sellers, K. R. Brandt, A. Wang, C. Janney, Z. S. Fredericksen, J. R. Cerhan, and C. M. Vachon
Age-specific Trends in Mammographic Density: The Minnesota Breast Cancer Family Study
Am. J. Epidemiol., May 1, 2008; 167(9): 1027 - 1036.
[Abstract] [Full Text] [PDF]

Home page
 Am J EpidemiolHome page
T. M. Peters, U. Ekelund, M. Leitzmann, D. Easton, R. Warren, R. Luben, S. Bingham, K.-T. Khaw, and N. J. Wareham
Physical Activity and Mammographic Breast Density in the EPIC-Norfolk Cohort Study
Am. J. Epidemiol., March 1, 2008; 167(5): 579 - 585.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
C. Nagata, T. Matsubara, T. Hara, H. Fujita, Y. Nagao, S. Yamamoto, C. Shibuya, Y. Kashiki, and H. Shimizu
Urinary 6-Sulfatoxymelatonin and Mammographic Density in Japanese Women
Cancer Epidemiol. Biomarkers Prev., November 1, 2007; 16(11): 2509 - 2512.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
C. M. Vachon, T. A. Sellers, E. E. Carlson, J. M. Cunningham, C. A. Hilker, R. L. Smalley, D. J. Schaid, L. E. Kelemen, F. J. Couch, and V. S. Pankratz
Strong Evidence of a Genetic Determinant for Mammographic Density, a Major Risk Factor for Breast Cancer
Cancer Res., September 1, 2007; 67(17): 8412 - 8418.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Warren, R.
Right arrow Articles by Easton, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Warren, R.
Right arrow Articles by Easton, D.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online