| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Review |
1 From the Departments of Radiology (J.A.H.) and Health Evaluation Sciences (V.E.B.), University of Virginia, Box 800170, Charlottesville, VA 22908. Received July 18, 2002; revision requested August 27; revision received October 23; accepted January 8, 2003. Address correspondence to J.A.H. (e-mail: jah7w@virginia.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONCHANGES IN BREAST COMPOSITION...METHODS OF MEASURING BREAST...RELATIONSHIP OF BREAST DENSITY...ALTERNATIVE EXPLANATIONS FOR THE...RELATIONSHIP BETWEEN NONHORMONAL...RELATIONSHIP BETWEEN HORMONAL...HISTOLOGIC EVALUATION OF...PLAUSIBLE FACTORS THAT MAY...FUTURE DIRECTIONS SUMMARYREFERENCES |
|---|
© RSNA, 2003
Index terms: Breast, density, 00.92 • Breast neoplasms, diagnosis, 00.30 • Breast neoplasms, radiography, 00.11, 00.1215, 00.30 • Radiography, digital, 00.1215 • Review
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONCHANGES IN BREAST COMPOSITION...METHODS OF MEASURING BREAST...RELATIONSHIP OF BREAST DENSITY...ALTERNATIVE EXPLANATIONS FOR THE...RELATIONSHIP BETWEEN NONHORMONAL...RELATIONSHIP BETWEEN HORMONAL...HISTOLOGIC EVALUATION OF...PLAUSIBLE FACTORS THAT MAY...FUTURE DIRECTIONS SUMMARYREFERENCES |
|---|
Breast density is perhaps the most undervalued and underutilized risk factor in studies investigating the causes of breast cancer.
Celia Byrne (1)
In 1976, Wolfe published an article that demonstrated a relationship between breast density and breast cancer risk (2). However, investigators of subsequent studies either did not reproduce the association (3,4) or did not show the association to be as strong as that in the Wolfe original report (5–7). Reader inconsistency issues also caused doubt as to the usefulness of the Wolfe density classification scheme (6,8), which fell into disuse in this country in the 1980s, although it is still commonly used outside of the United States.
Beginning in the early 1980s, more precise quantitative methods of measuring breast density were developed. Since 1982, to our knowledge 12 studies (9–20) in which quantitative methods were used for assessing breast density have been performed; findings of all these studies showed a positive moderate association with breast cancer risk. The risk of breast cancer for women with increased breast density in most of these studies is four to six times that for women with less dense tissue, a relative risk greater than most traditional risk factors such as nulliparity and early menarche. Unlike most other breast cancer risk factors, breast density may be influenced by hormonal agents, alcohol use, and possibly other means.
In this article, we review normal changes in breast composition over time, methods of measuring breast density, the association of breast density measured with quantitative methods to breast cancer risk, alternative explanations for the association, how hormonal and other factors are associated with breast density, and plausible reasons that increased breast density increases breast cancer risk. Last, we examine potential future directions for use of quantitative assessment of breast density in relation to breast cancer risk.
|
CHANGES IN BREAST COMPOSITION OVER TIME: NORMAL PHYSIOLOGY |
|---|
|
TOPABSTRACTINTRODUCTIONCHANGES IN BREAST COMPOSITION...METHODS OF MEASURING BREAST...RELATIONSHIP OF BREAST DENSITY...ALTERNATIVE EXPLANATIONS FOR THE...RELATIONSHIP BETWEEN NONHORMONAL...RELATIONSHIP BETWEEN HORMONAL...HISTOLOGIC EVALUATION OF...PLAUSIBLE FACTORS THAT MAY...FUTURE DIRECTIONS SUMMARYREFERENCES |
|---|
As women near menopause, the menstrual cycle shortens. Specifically, the follicular phase shortens, with no significant change in the length of the luteal phase. This continues until the onset of oligomenorrhea. During the perimenopausal years, preovulatory estradiol levels tend to be higher than they are in younger women, while progesterone levels remain similar. Breast cysts are the most common form of breast lumps between ages 40 and 49 years, possibly due to the shortened follicular phase and elevated estrogen levels during this time. The breasts continue to become less dense, with about 50% of women in their 40s and about 65% of women in their 20s having 50% or greater breast density (26).
Menopause
With the reduction of estrogen and progesterone levels after menopause, the cyclic proliferative process becomes quiescent. Lobular tissue regresses, while the more proximal portions of the ductal system remain. The mammographic appearance of the breasts becomes increasingly radiolucent (26), with about 34% of women aged 75–79 years having fat-replaced breasts compared with only 11% of women aged 25–29 years. Likewise, only 30% of women aged 75–79 years have 50% or greater breast density (26).
|
METHODS OF MEASURING BREAST DENSITY |
|---|
|
TOPABSTRACTINTRODUCTIONCHANGES IN BREAST COMPOSITION...METHODS OF MEASURING BREAST...RELATIONSHIP OF BREAST DENSITY...ALTERNATIVE EXPLANATIONS FOR THE...RELATIONSHIP BETWEEN NONHORMONAL...RELATIONSHIP BETWEEN HORMONAL...HISTOLOGIC EVALUATION OF...PLAUSIBLE FACTORS THAT MAY...FUTURE DIRECTIONS SUMMARYREFERENCES |
|---|
The Wolfe classification has been applied inconsistently. In a 1993 review, Oza and Boyd (27) found that the Wolfe classification had an interobserver agreement of 52%–97% and an intraobserver agreement of 69%–97%. Authors of case-control and cohort studies have found an association between increased breast cancer risk and the Wolfe P2 and DY categories, although no association was as high as that in the original report by Wolfe (28). Findings of cross-sectional studies have shown a weak association or no association (27).
Attempts to develop a reproducible quantitative method of assessing breast density began in the early 1980s (9). Visual estimation of the percentage of the breast occupied by breast tissue has been used frequently, with the number of categories varying from five (9) to 20 (29), with the most frequently used method as proposed by Boyd et al (10) having six categories. Agreement for visual assessment was very good, with intrareader reliability estimation of 0.82 reported in one 20-category study (30) and 
value of 0.89 (31) and intraclass correlation coefficient of 0.94 (10) reported in the six-category studies.
In the United States, the Breast Imaging Reporting and Data System (BI-RADS) was developed to standardize mammography reporting terminology and the assessment and recommendation categories (32). BI-RADS density classification is included in the report to inform referring physicians of the decline in sensitivity of mammography with increasing breast density. It was not intended to serve as a method of measuring breast density. The four-category system (almost entirely fat; scattered fibroglandular densities; heterogeneously dense, may lower the sensitivity of mammography; extremely dense, which could obscure a lesion on mammography) is based on quantitative assessment, although the categories are not defined by the percentage of density. The lack of well-defined categories likely accounts for the only moderate interobserver agreement seen with the use of BI-RADS density categories, with an overall value of 0.43–0.59 (33,34). However, the advantage of the BI-RADS classification is that it is widely used in the United States and allows analysis of large study populations (18).
More consistent computer-assisted measurements of breast density have been developed. All methods use planimetry in some way. Computer-assisted planimetry was used as early as 1987 (11). With this method, an acetate overlay was placed over the mammogram, and the outline of the breast and breast tissue was made by using a wax pencil. A computerized compensating planimeter was then used to obtain the total area of the breast and the total area of the breast tissue (11–13). The percentage of density is then calculated by dividing the area of the breast tissue by the total area of the breast. Agreement with this method is 70%–94% (11,12). More recently, film mammograms have been digitized, and the area of the breast parenchyma and the total area of the breast are outlined by using either a mouse (25) or a digital segmentation of the mammogram (35,36). With the digital segmentation method, the histogram of the digitized mammogram is divided into fatty tissue and parenchyma by using interactive thresholding (35) (Fig 1). Agreement with this method is reported to be 90% or greater (35). Other investigators (37–40) are developing automated calculation of the percentage of density by using segmentation, with promising results, and most recent methods yielded 80%–90% agreement with radiologist assessment of density (38–40).
|
Another problem receiving increasing attention is that use of the percentage of mammographic density represents only a two-dimensional representation of a three-dimensional phenomenon. Women with large breasts may have a substantial amount of glandular breast tissue yet have low breast density percentages. On the other hand, women with small breasts and reduced amounts of glandular tissue may have a higher percentage of breast density. A number of investigators are trying to devise a means of determining the total glandular content from three-dimensional reconstruction of breast (Shephard J, oral communication, July 2002). It may be that this parameter would be a better predictor of breast cancer than quantification of density based on current two-dimensional techniques. Further studies are ongoing to address this issue in a meaningful way (Cummings S, oral communication, July 2002).
Other investigators are exploring the use of magnetic resonance imaging (41,42) and ultrasonography (43) for quantifying breast density, although these methods may not prove as useful for widespread use as density information already obtained on the mammogram. Shephard et al (44) are also investigating the use of dual x-ray absorptiometry for measuring breast density.
|
RELATIONSHIP OF BREAST DENSITY AND BREAST CANCER RISK |
|---|
|
TOPABSTRACTINTRODUCTIONCHANGES IN BREAST COMPOSITION...METHODS OF MEASURING BREAST...RELATIONSHIP OF BREAST DENSITY...ALTERNATIVE EXPLANATIONS FOR THE...RELATIONSHIP BETWEEN NONHORMONAL...RELATIONSHIP BETWEEN HORMONAL...HISTOLOGIC EVALUATION OF...PLAUSIBLE FACTORS THAT MAY...FUTURE DIRECTIONS SUMMARYREFERENCES |
|---|
Since 1982, authors of at least 12 studies have reported an association of breast density with breast cancer risk by using quantitative methods (Table) (9–20). Findings of all these studies have shown a moderate to strong positive association of increased breast cancer risk with higher levels of breast density. All were either population-based or clinic-based case-control studies, and several were nested within large cohort populations. All studies were substantial in size, with more than 150 cases, and half of the studies had two or more controls for each case. In all studies, radiologists were blinded to the case or control status.
|
Eight of these studies used both the Wolfe classification and quantitative assessment of breast density; in all but the Wolfe et al (11) and Kato et al (13) studies, the quantitative analysis showed a stronger association with breast cancer risk than did the Wolfe classification, which suggests that quantitative methods may be more useful in identifying women at increased risk for developing breast cancer.
Sala et al (45) report a positive association with higher histologic grade tumors and mammographic density assessed by using Wolfe criteria. In this issue of Radiology, Roubidoux et al (46) report a higher incidence of invasive breast cancers that are estrogen-receptor negative, are of higher histologic grade, and are of larger size in women with dense breast tissue that had negative clinical and mammographic screening findings within 17 months of diagnosis. However, authors of the study found that only tumor size was independent of age. The authors postulate that the association between larger tumor size and dense tissue at mammography may be due to either more rapid growth in glandular breasts or increased difficulty in detection due to the dense tissue.
Attributable Risk
Higher levels of breast density are fairly common, since 50% of women between the age of 40 and 49 years and 30% of women aged 70–79 years have breasts that are at least 50% dense (26). If breast density is a moderate risk factor and the risk factor is fairly common, breast cancers attributable to increased breast density could potentially account for an important percentage of total breast cancer cases. In two studies, attributable risk from increased breast density was 28%–30% for 50% or greater density and 40%–44% for any breast density (10,17). In comparison, less than 5% of breast cancers are attributable to breast cancer gene (BRCA1 and BRCA2) mutations.
Breast cancers attributable to dense breast tissue may become more prevalent in the future. Blane et al (47) have found that for women aged 50–59 years, breast density on screening mammograms obtained in the 1990s did not decline with age as rapidly as that seen in the 1980s, even in those women not undergoing hormone replacement therapy (HRT). Authors of that study did not control for such factors as parity, age at first birth, or weight; therefore, slower involution of the breast may be due to these or other causes. Regardless of the cause, findings of the study suggest that if breast density is a moderate risk factor for developing breast cancer, then the attributable risk due to density may become more significant in our population.
Dose Response
If high breast density is a risk factor for breast cancer compared to fat-replaced breasts, then increasing density should be associated monotonically with increasing risk. Boyd et al (48) reviewed the earliest nine of the 12 studies cited in the Table and found a positive trend in all but the Wolfe study (11). In four studies (10,12,13,17), statistical evaluation for trend was significant (P < .001 to P < .0001). This implies that the breast cancer risk increases in proportion to the degree of breast density.
In addition, if the relationship between breast density and risk is proportional, then changing density should change risk. Investigators of one study have found a trend of changes in risk with changes in density (19). In that study, women with 5%–25% density initially had an odds ratio of 5.7 (95% CI: 2.2, 15.2) for developing breast cancer if their density did not change over an 8-year interval. However, if their breast density decreased to less than 5% during the interval, then the odds ratio for developing breast cancer was only 1.9 (95% CI: 0.6, 6.1), while those with an increase in density to 25% or greater had an odds ratio of 6.9 (95% CI: 2.1, 22.9) (19). This was a small study with wide CIs, and the findings should be confirmed in larger studies.
|
ALTERNATIVE EXPLANATIONS FOR THE LINK BETWEEN BREAST DENSITY AND CANCER RISK |
|---|
|
TOPABSTRACTINTRODUCTIONCHANGES IN BREAST COMPOSITION...METHODS OF MEASURING BREAST...RELATIONSHIP OF BREAST DENSITY...ALTERNATIVE EXPLANATIONS FOR THE...RELATIONSHIP BETWEEN NONHORMONAL...RELATIONSHIP BETWEEN HORMONAL...HISTOLOGIC EVALUATION OF...PLAUSIBLE FACTORS THAT MAY...FUTURE DIRECTIONS SUMMARYREFERENCES |
|---|
"Masking" Explains the Apparent Association of Density to Increased Risk
Egan and Mosteller (49) proposed the masking hypothesis to explain the apparent relationship of breast density and breast cancer risk. Because cancers are more difficult to detect in dense breast tissue, prevalent cancers may be more likely to be missed at the first screening among women with dense tissue. When the tumors are detected later as apparent incident cancers, a spurious association may arise between the breast cancer risk and tissue density. Evaluation of a relationship between density and breast cancer risk from mammograms obtained in the 1st year or 2 years after initial screening may artificially suggest that density is a risk factor for breast cancer. However, if breast density was not a risk factor for breast cancer, then at some future point after the initial screening, cancer incidence in women with dense and in those with fatty breasts should become similar. In addition, findings in studies in which cancer diagnosis is evaluated at the time of the initial screening examination should show fewer cancers in women with dense breast tissue than in women who have fat-replaced breast tissue.
A small study (50) was performed examining the possible effect of masking on the relationship of breast density and cancer risk. Authors of the study found that increased density defined as more than 25% of the breast occupied by fibroglandular tissue was associated with a relative risk of 1.4 at the initial screening examination, 1.2 at 1–2 years, 3.3 at 3–4 years, and 1.2 at 5–6 years (50). These findings are consistent with the hypothesis that masking of cancers in dense breast tissue does occur, although the effect is small and peaks 3–4 years after the initial screening examination. Masking, however, does not explain the increased relative risk of breast cancer for women with dense breast tissue at study entry, since cancers that are masked within the dense tissue should result in a lower detection and therefore lower apparent relative risk. Of note, the study (50) used mammograms obtained in the 1970s, which were much lower in contrast than are those obtained by using more modern techniques.
Masking bias was less likely to occur in studies that had an initial period during which tumors were excluded from the study (10,17) or had a longer period between initial screening and case status ascertainment (12,19). Studies performed on the basis of cohorts with long-term follow-up that examine density and risk using mammograms obtained in a regularly screened population should also be less affected by masking (10–13,17,19). Findings of three nested case-control studies (10,12,17) have shown that risk remains elevated for at least 5–10 years of follow-up. Byrne et al (17) reported findings for women with 10 or more years between density assessment and case-control determination; among women with 0% density, there were 17 cases and 42 controls, while among women with density greater than 75%, there were 45 cases and 29 controls. Therefore, 45 (73%) of 62 cases had dense tissue and only 29 (40%) of 71 controls had dense tissue. For masking of prevalent cancers to account for this large difference at 10 years after initial screening, the effect would have to be substantial and long-lived. We estimate that about 75% of prevalent cancers in women with dense breasts would have been missed and subsequently diagnosed 10 or more years later to account for this difference, which is highly unlikely.
Masking has been shown to introduce some bias in the detection of breast cancer in dense tissue in one small study in which older mammograms were used (50). However, the association between breast cancer risk and increased breast density was consistent in every quantitative study reviewed, from those in which density and disease were ascertained nearly simultaneously (9,11,16) to those with up to a decade between density and disease ascertainment (17,19). Given the findings of these studies, masking bias does appear to occur but likely has only a small and short-lived effect.
Current but Not Past Density Is Associated with Increased Risk
In an editorial about breast density, White stated that "disagreement also exists over whether current or past breast density serves as the better risk predictor" (51). Since the effect of breast density on cancer risk persists for at least 5–10 years after assessment (10,12,17), past breast density appears to be important. On the other hand, findings of one study suggest that changes in density change risk (19). In that study, reduction of breast density over time lowered breast cancer risk but not to the extent as in women with comparable density at baseline. This implies that increased density has some persistent effect. Their data also suggest that during the 8 years of follow-up, striking changes in density were relatively uncommon in both cases and controls. The studies reviewed here have widely varying times between disease detection and density determination (Table), and all show associations between density and breast cancer risk. It appears likely that both past and current breast density are associated with risk.
|
RELATIONSHIP BETWEEN NONHORMONAL FACTORS AND BREAST DENSITY AND BREAST CANCER RISK |
|---|
|
TOPABSTRACTINTRODUCTIONCHANGES IN BREAST COMPOSITION...METHODS OF MEASURING BREAST...RELATIONSHIP OF BREAST DENSITY...ALTERNATIVE EXPLANATIONS FOR THE...RELATIONSHIP BETWEEN NONHORMONAL...RELATIONSHIP BETWEEN HORMONAL...HISTOLOGIC EVALUATION OF...PLAUSIBLE FACTORS THAT MAY...FUTURE DIRECTIONS SUMMARYREFERENCES |
|---|
The association of breast density with cancer risk seems counterintuitive given that breast cancer risk increases with advancing age at the same time that breast density decreases. However, other age-associated factors besides breast density are likely driving the age-associated increase in breast cancer risk. A parallel situation may be seen with smoking and the incidence of heart disease. As people age, they are more likely to quit smoking (52). However, an 80-year-old nonsmoker is more likely to die of heart disease than a 40-year-old smoker. This does not imply that smoking does not increase the risk of heart disease, only that other factors besides smoking increase the risk of heart disease. Likewise, other factors besides breast density contribute to breast cancer risk. The association between decreasing breast density and increasing breast cancer risk with aging does not negate the association between increased breast density and breast cancer risk.
Weight and Body Mass Index
Obesity is a known risk factor for breast cancer. Lam et al (18) specifically examined the relationship between weight and body mass index on the association of breast density and breast cancer risk. They found that women with higher body weight and body mass index were less likely to have dense breasts. The unadjusted odds ratio for developing breast cancer for women who weighed more than 81 kg compared with those weighing less than 63 kg was 1.7 (95% CI: 1.2, 2.6). However, after adjusting for density, the odds ratio increased to 2.1 (95% CI: 1.3, 3.2). This increase indicates that density is an independent risk factor and that women who are obese and have dense breasts are at higher risk. Likewise, the unadjusted odds ratio for women with extremely dense breasts was 3.2 compared with that in women with predominantly fatty breasts. However, when adjusted for weight and body mass index, the odds ratio increased to 4.5 (95% CI: 1.9, 10.6). Sala et al (53) found similar findings and noted that "this negative confounding of two positive risk factors means that the effect of parenchymal patterns on risk will tend to be underestimated when not adjusted for body mass index and waist hip ratio and vice versa."
Parity
Increased breast density is associated with nulliparity and late age at first birth (54). van Gils et al (55) examined the relationship of parity and breast density with breast cancer risk. In that study, nulliparous women with fatty breasts were not at increased risk compared with parous women. However, women who were nulliparous with greater than 25% density were 6.6 times more likely (95% CI: 2.6, 16.5) to develop breast cancer, while parous women with greater than 25% density were only 3.6 times (95% CI: 1.7, 7.7) more likely develop breast cancer.
Family History
Boyd et al (56) performed a subanalysis within the Canadian National Breast Screening Study by examining breast density and the influence of family history on breast cancer risk. For women with at least one affected first-degree relative, the relative risk of breast cancer was 11.1 (95% CI: 1.5, 80.4) for women with dense breasts compared with those with less than 10% density. The authors concluded that mammographic density may be strongly associated with the risk of breast cancer in women with a family history of the disease. Likewise, Ziv et al (57) have found that women with extremely dense breast tissue were more likely to have first-degree relatives with breast cancer.
Benign Breast Disease
Women with benign breast disease have a greater risk of breast cancer if they also have dense breast tissue than if they have less dense breast tissue. In patients participating in the Breast Cancer Detection Demonstration Project, an increased relative risk of breast cancer was seen with nonproliferative (relative risk, 5.8; 95% CI: 1.8, 18.6) and proliferative (relative risk, 3.2; 95% CI: 1.6, 6.6) benign breast disease in women with greater than 75% density than in women with less than 50% density (58). However, atypical hyperplasia had a higher relative risk in women with lower breast density (relative risk, 4.1; 95% CI: 2.1, 8.0) than in women with dense breasts (relative risk, 2.1; 95% CI: 0.6, 7.0). Therefore, the risk associated with benign breast disease is not explained by density alone and vice versa.
Race and Ethnicity
There are little data regarding the differences in breast density between different races and ethnic groups. Maskarinec and Meng (20) have found that the association between density and breast cancer risk for Asian women is similar to that of Caucasian women.
Diet
The relationship between diet and breast density is not well defined. Findings of two studies (16,59) have shown that high intake of saturated fat is associated with greater breast density, but findings of another study (60) showed lower density. High intakes of vitamins C and E have been associated with increased density (60), while high intake of vitamin A has been associated with lower density (16). Findings of a 2-year study (61) in which women were randomly assigned to no intervention or a low-fat high-carbohydrate diet did not show a change in density.
Intake of both soy protein and soy isoflavones has shown an inverse association with breast density in Chinese women in Singapore (62). Breast cancer rates are lower in Asian countries than they are in the United States and Western Europe (63), and dietary differences, including soy intake, have been hypothesized to contribute to a lowered risk. Soy products are high in phytoestrogens, which may displace endogenous estrogens at hormone-binding sites, decreasing the stimulatory effect of endogenous estrogens in the breast (62).
Alcohol
High alcohol intake is associated with increased breast density (60). However, while white wine intake was associated with increased density in postmenopausal women, red wine intake was associated with lower breast density (60). No difference was noted for premenopausal women and the type of alcoholic beverage.
Exercise
Only two studies, to our knowledge, have addressed the relationship between exercise and breast density. Authors of one study found no relationship (60), and authors of the other study found a weak inverse relationship between moderate exercise and breast density (64).
|
RELATIONSHIP BETWEEN HORMONAL FACTORS AND BREAST DENSITY AND BREAST CANCER RISK |
|---|
|
TOPABSTRACTINTRODUCTIONCHANGES IN BREAST COMPOSITION...METHODS OF MEASURING BREAST...RELATIONSHIP OF BREAST DENSITY...ALTERNATIVE EXPLANATIONS FOR THE...RELATIONSHIP BETWEEN NONHORMONAL...RELATIONSHIP BETWEEN HORMONAL...HISTOLOGIC EVALUATION OF...PLAUSIBLE FACTORS THAT MAY...FUTURE DIRECTIONS SUMMARYREFERENCES |
|---|
|
Breast density appears to be very responsive to HRT. Investigators of the PEPI trial have found that the greatest change in density occurs during the 1st year of use (73). The breast responds rapidly to hormonal manipulation, with a decrease in density seen after just 2 weeks after HRT cessation (Fig 3) (76). Colacurci et al (77) have found that women who stopped HRT for 3 weeks prior to their annual mammography had no significant change from baseline, whereas women who continued HRT had an increase in density.
|
At histologic examination, benign breast biopsies performed in women using estrogen plus progestin had significantly higher proliferation indices compared with biopsies performed in women either using estrogen alone or not undergoing HRT (82). In addition, the proliferation noted in women using estrogen with progestin in that study was localized to the terminal duct lobular unit, which is the site of the development of most breast cancers (82).
Selective Estrogen Receptor Modulators
Selective estrogen receptor modulators decrease mammographic density presumably due to the antiestrogen effect on the breast (Fig 4). In a breast cancer prevention trial, 44% of women using tamoxifen experienced a reduction in breast density compared with 15% of women randomly assigned to a placebo (83). Raloxifene is another selective estrogen receptor modulator that is approved by the U.S. Food and Drug Administration for treatment of osteoporosis. In the Multiple Outcomes of Raloxifene Evaluation Trial, raloxifene significantly reduced breast density compared with placebo and estrogen (36).
|
Other Hormonal Agents
Decrease in breast density has also been reported with use of a gonadotropin-releasing hormone agonist that was used in a clinical trial for birth control (87). The gonadotropin-releasing hormone agonist inhibits ovarian function and is hypothesized to reduce breast cancer risk (87). Tibolone is widely used in Europe for hormone replacement therapy but is not approved by the Food and Drug Administration in the United States. The precise mechanism of action of this agent is not known since it has multiple metabolites, each of which can exert differing effects. Its major effects are to inhibit sulfatase activity and thereby block estrogen synthesis from estrone sulfate and to exert androgenic effects (88). This compound increases breast density in a small number of women (2%–6%) compared with women who use estrogen and progesterone daily (46%–50%) (89). No data are available about the effect of oral contraceptives or androgens on breast density.
Changes in Breast Density with HRT and Selective Estrogen Receptor Modulator Mirror Breast Cancer Risk
Mammographic density appears to mirror the risk associated with hormonal agents (Fig 5). Since women with mammographically dense breasts have an increase in breast cancer risk, women who have an increase in mammographic density in response to HRT may be at higher risk for developing breast cancer than women who do not experience a change in density with use of HRT, although, to our knowledge, there are currently no studies in this area. Decrease in breast density in response to selective estrogen receptor modulators may also decrease breast cancer risk, and further studies to these points may be useful, particularly with the suggestion in one study (19) that changes in density change breast cancer risk.
|
|
HISTOLOGIC EVALUATION OF MAMMOGRAPHICALLY DENSE BREASTS |
|---|
|
TOPABSTRACTINTRODUCTIONCHANGES IN BREAST COMPOSITION...METHODS OF MEASURING BREAST...RELATIONSHIP OF BREAST DENSITY...ALTERNATIVE EXPLANATIONS FOR THE...RELATIONSHIP BETWEEN NONHORMONAL...RELATIONSHIP BETWEEN HORMONAL...HISTOLOGIC EVALUATION OF...PLAUSIBLE FACTORS THAT MAY...FUTURE DIRECTIONS SUMMARYREFERENCES |
|---|
As to epithelial changes, authors of several studies who correlated increased density with quantitative analysis have shown an increase in epithelial hyperplasia (58,91,95). Specifically, women participating in the Canadian National Breast Screening Study with extreme breast density have been shown to be 12.2 times more likely to develop hyperplasia without atypia (91). Of note, these women were also found to be 9.7 times more likely to develop atypical hyperplasia or carcinoma in situ (91). Authors of another study (96) compared nipple aspirate fluid cytologic results to breast density and found that women with mammographically dense breasts were four times more likely to have cytologic atypia. The mammographically dense breasts are therefore likely caused by an increase in both stromal and epithelial tissue. These changes may lead to development of the histologic precursors of breast cancer, as was seen with the increased incidence of atypia and carcinoma in situ in these studies (91,96).
|
PLAUSIBLE FACTORS THAT MAY EXPLAIN THE ASSOCIATION OF BREAST DENSITY WITH BREAST CANCER RISK |
|---|
|
TOPABSTRACTINTRODUCTIONCHANGES IN BREAST COMPOSITION...METHODS OF MEASURING BREAST...RELATIONSHIP OF BREAST DENSITY...ALTERNATIVE EXPLANATIONS FOR THE...RELATIONSHIP BETWEEN NONHORMONAL...RELATIONSHIP BETWEEN HORMONAL...HISTOLOGIC EVALUATION OF...PLAUSIBLE FACTORS THAT MAY...FUTURE DIRECTIONS SUMMARYREFERENCES |
|---|
Estrogen Exposure
Breast density is clearly hormonally responsive. Some investigators believe that breast density may reflect past exposure to estrogen (51). There is moderate evidence to support the role of endogenous estrogens in the cause of breast cancer. Experimental evidence suggests that the estrogen metabolites 4-hydroxyestrone and 4-hydroxyestradiol are further metabolized to estradiol quinones (99). These compounds can bind directly to adenine and guanine and pull these bases off segments of DNA. This is postulated to be a mechanism for direct mutagenic effects of estrogens. Others postulate that the 2-hydroxymetabolites of estradiol cause stable DNA adducts or that the 16-
estrogen metabolites can covalently bind to the estrogen receptor and constitutively activate it (99).
Estrogen production may occur within the breast itself through the enzymes aromatase and sulfatase. Aromatase converts androgens to estrogens, while sulfatase converts biologically inactive estrone sulfate into biologically active estrone. A surprising finding is that estradiol levels within breast cancers and the surrounding tissue are similar in pre- and postmenopausal women even though plasma concentrations are markedly higher in premenopausal women (100). This observation suggests that a major source of estradiol in the breast tissue in postmenopausal women who develop breast cancer is the breast itself due to overactive aromatase or sulfatase (96,101). Further work regarding the biologic cause of increased breast density will be helpful.
Serum Prolactin Levels
Elevated serum prolactin levels have been associated with higher breast density in postmenopausal women in a dose-response fashion (98). Findings of the Nurses Health Study showed a significant positive association between plasma level of prolactin and the postmenopausal breast cancer risk (highest vs lowest quartile; multivariate relative risk, 2.03; 95% CI: 1.24, 3.31) (102). Neither study findings showed an association of prolactin levels with breast density or breast cancer risk in premenopausal women. Thus, high mammographic breast density and breast cancer risk may be related to elevated prolactin levels only in postmenopausal women.
Genetics
In a cohort study of families with breast cancer, Pankow et al (30) evaluated primary relative relationships and correlation with breast density. The sister-sister correlation was statistically significant (r = 0.16–0.27, P < .05), while the mother-sister correlation was suggestive (r = 0.01–0.17). Paternal breast density obviously cannot be assessed. Segregation analysis suggests that a major autosomal gene influences breast density, probably with Mendelian transmission of a dominant gene. Further work with this cohort suggests that the gene may be on chromosome 6 (29).
Authors of a recent study correlated breast density between 218 pairs of monozygotic twins and 134 pairs of dizygotic twins in the United States and Australia (103). According to the study findings, heritability accounted for 63% of the variation in breast density in all twins studied (103). Also of interest, women with known BRCA1 or BRCA2 mutations have been shown to have denser breast tissue that is coarser and lower in contrast than women at low risk for developing breast cancer (104). Findings of these studies suggest that genetics plays a strong role in breast density.
|
FUTURE DIRECTIONS |
|---|
|
TOPABSTRACTINTRODUCTIONCHANGES IN BREAST COMPOSITION...METHODS OF MEASURING BREAST...RELATIONSHIP OF BREAST DENSITY...ALTERNATIVE EXPLANATIONS FOR THE...RELATIONSHIP BETWEEN NONHORMONAL...RELATIONSHIP BETWEEN HORMONAL...HISTOLOGIC EVALUATION OF...PLAUSIBLE FACTORS THAT MAY...FUTURE DIRECTIONS SUMMARYREFERENCES |
|---|
Use of Density as an Entry Criterion into Breast Cancer Prevention Trials
If density is a risk factor for breast cancer, use of high breast density as an inclusion criterion for breast cancer prevention trials may increase study power such that studies would require fewer patients and be directed at women at greater risk. For instance, a dietary intervention study recruiting women with 50% or greater density from 1982 to 1994 has noted a four to five times greater incidence of breast cancer than expected (105).
Use of Breast Density as a Surrogate Marker of Risk for Prevention Trials and HRT Regimens
Since changes in breast density appear to reflect breast cancer risk, some investigators have proposed using density as a preliminary surrogate marker in prevention trials. If a drug that is thought to reduce risk causes an increase in breast density, the risk may not be reduced. Further studies evaluating correlation of cancer risk and changes in density are necessary before this could be useful. Likewise, there is increasing interest from drug manufacturers regarding whether their particular HRT regimen is associated with an increase in density because of the possible implications regarding breast cancer risk. However, it should be stressed that currently there is only suggestive data that an increase in density is associated with an increase in risk (19), and further studies need to be performed to investigate this concept.
The Mammogram as a Physiologic Test
If future study findings do show that changes in density do correlate with changes in risk, mammography may someday be viewed as a physiologic test in addition to its current screening and diagnostic roles. Women may someday be given an estimate of their cancer risk based on risk models that included breast density at the time of their mammography. They could use this information to make decisions regarding HRT use, lifestyle changes, or chemoprevention. However, since breast cancer risk remains elevated for at least 5–10 years after the mammogram showed increased density, lowering density may not change cancer risk or may take 5–10 years after the intervention to affect breast cancer risk. This is an important area for investigation.
|
SUMMARY |
|---|
|
TOPABSTRACTINTRODUCTIONCHANGES IN BREAST COMPOSITION...METHODS OF MEASURING BREAST...RELATIONSHIP OF BREAST DENSITY...ALTERNATIVE EXPLANATIONS FOR THE...RELATIONSHIP BETWEEN NONHORMONAL...RELATIONSHIP BETWEEN HORMONAL...HISTOLOGIC EVALUATION OF...PLAUSIBLE FACTORS THAT MAY...FUTURE DIRECTIONS SUMMARYREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONCHANGES IN BREAST COMPOSITION...METHODS OF MEASURING BREAST...RELATIONSHIP OF BREAST DENSITY...ALTERNATIVE EXPLANATIONS FOR THE...RELATIONSHIP BETWEEN NONHORMONAL...RELATIONSHIP BETWEEN HORMONAL...HISTOLOGIC EVALUATION OF...PLAUSIBLE FACTORS THAT MAY...FUTURE DIRECTIONS SUMMARYREFERENCES |
|---|
