Menopausal Medicine: For clinicians who provide care for women

Mammographic breast density, endocrine therapies, and breast cancer risk

Assistant Professor of Medicine, Mayo Clinic College of Medicine, Rochester, Minnesota

Associate Professor of Epidemiology, Mayo Clinic College of Medicine, Rochester, Minnesota

MBD categories

Several classification schemes have been used to categorize MBD over the decades. The earliest categorization was Wolfe’s parenchymal pattern that classified the extent and type of density into 4 categories 2 :

• N1: Nondense, no ducts visible;

• P1: Prominent ductal pattern occupying less than one-fourth of the breast;

• P2: Prominent ductal pattern occupying more than one-fourth of the breast; and

• DY: Homogenous, plaque-like areas of density.

Currently, the most common clinical measure of density is the Breast Imaging Reporting and Data Systems (BI-RADS) density method, proposed by the American College of Radiology. 3 BI-RADS density is a subjective measure used by radiologists to classify a mammogram as follows:

• D1: Fatty;

• D2: Scattered density;

• D3: Heterogeneously dense; or

• D4: Extremely dense.

Computer-assisted methods are used to estimate quantitative measures of MBD, including percent density (percentage of the overall breast showing dense tissue; PD), absolute dense area, and nondense area. 4 , 5

MBD and breast cancer risk

Elevated MBD is considered one of the strongest risk factors for breast cancer regardless of whether it is assessed as a categorical or quantitative measure. 6 Women in the highest categories of breast density have a 4- to 6-fold increased breast cancer risk compared with women in the lowest categories. MBD has been shown to be a stronger risk factor for breast cancer than any others except age and genetic mutations. 6

The association between MBD and breast cancer has been seen both in older and younger women undergoing screening mammograms, as well as in Caucasian and non-Caucasian populations. 4 , 6 , 7

Hormonal influence on MBD

There is strong and consistent evidence that MBD is influenced by hormonal factors such as menopause, age at menarche, parity, age at first birth, and use of exogenous hormones. 8 , 9

The influence of exogenous hormones on MBD is best illustrated by positive associations of MBD with postmenopausal hormone (PMH) therapy and inverse associations with tamoxifen, as described below. These associations are important because they suggest interindividual variability in response to hormone therapies manifested in MBD changes, which may translate into differential breast cancer risk.

Influence of PMH

With estrogen depletion during menopause, the breast glandular tissue undergoes regression. 10 This is also reflected by a decrease in MBD during and after menopause.

What happens to breast tissue when this process is interrupted by PMH? When assessing the influence of conventional estrogen-plus-progesterone PMH use, studies evaluating MBD by quantitative measures suggest that these hormones are associated with an absolute increase of 3% to 6% in MBD. 11 , 12 The FIGURE illustrates one woman’s dramatic change in MBD after 1 year of PMH use.

In the Norwegian Breast Cancer Screening Program (n=1007), current users of PMH therapy had a significantly higher mean percent MBD than never users (MBD, 11.5 (9.7-13.7) for current users vs 7.2 (6.6-7.8) for never users; P for trend <.001). 13

In the Women’s Health Initiative (WHI) study, women were randomized to receive daily combined conjugated equine estrogens (CEE, 0.625 mg) plus medroxyprogesterone acetate (MPA, 2.5 mg) or placebo. Combined PMH (n=202) was associated with an absolute increase of 6.0% in MBD at 1 year compared with baseline, whereas the placebo group (n=211) had a decrease of 0.9% in density (a 6.9% difference between the treatment and placebo groups). 11 More than 75% of women on active PMH had an increase in MBD.

Similarly, the Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial, which examined the association between PMH and MBD, also saw increases with combined hormone therapy. 12 Women were randomized to 1 of 5 interventions: (1) placebo (n=112); (2) CEE, 0.625 mg/d (n=114);(3) CEE + cyclic MPA, 10 mg/d for 12 d/mo (n= 109); (4) CEE + daily MPA, 2.5 mg/d (n=121); or (5) CEE + micronized progesterone (MP), 200 mg/d for 12 d/mo (n=115). The mean changes in percent MBD over 12 months were statistically significantly greater for all combined therapy regimens relative to women in the placebo group (range, 3.1%-4.8%), although women in the estrogen-alone arm showed no significant change in MBD (1.17%, 95% confidence interval [CI], –0.28-2.62; P=.24). Moreover, there was no significant difference between the change in MBD with the type of progestin or pattern of progestin use for cyclic or continuous MPA (CEE + cyclic MPA, 4.8%; CEE + continuous MPA, 4.6%; and CEE + MP, 3.1%). 12

Taken together, these studies suggest that estrogen-and-progestin combined therapies at conventional doses are associated with increased breast density, regardless of the pattern of progestin administration. And, importantly, the associations of these therapies with MBD change parallel the associations with breast cancer risk, such that there are increases in breast cancer with combination therapy but not with estrogen alone. 14

Although lower-dose regimens are hoped to have less influence on breast density, a recent study found no difference between the associations of conventional and low-dose hormone therapy with MBD. 15 Importantly, Buist et al reported that 1 to 2 months of cessation of PMH use was associated with only small decreases in MBD, and stopping therapy for this short period did not affect mammogram recall rates. 16

Influence of tamoxifen, raloxifene, and aromatase inhibitors on MBD

In view of the positive association between PMH and MBD, the logical question is whether tamoxifen, raloxifene, or aromatase inhibitors reduce MBD.

Tamoxifen is a selective estrogen receptor modulator (SERM) that competitively binds to estrogen receptors and blocks estrogen synthesis. 17 The antiestrogenic influence on the breast resulted in its use as adjuvant endocrine therapy for women with estrogen receptor–positive postmenopausal breast cancer. Tamoxifen has also been used as a chemopreventive agent to reduce the risk of breast cancer in women at high risk. In the National Surgical Adjuvant Breast and Bowel Project (NSABP) randomized clinical trial of high-risk women, 5 years of tamoxifen therapy was shown to reduce invasive breast cancer risk by 49% and noninvasive breast cancer risk by 50% compared with placebo. 18

Several studies have examined the effect of tamoxifen on MBD 19 - 22 using different estimation methods for MBD and study populations (ie, women with known breast cancer on adjuvant tamoxifen or women at high risk for breast cancer on tamoxifen for chemoprevention). These studies have consistently demonstrated that women on tamoxifen do experience a statistically significant reduction in MBD. Some of these studies noted that decreases in percent MBD occurred at a greater frequency in premenopausal than in postmenopausal women. 19

Son et al 21 demonstrated that 87% of premenopausal women with breast cancer had a decrease in parenchymal area with tamoxifen use, whereas 29% of postmenopausal women experienced a decrease.

Similarly, in the Brisson study, 19 67% of high-risk women on tamoxifen who were under age 50 experienced a decrease in parenchymal pattern classification compared with 13% of postmenopausal women age 50 or older. Brisson et al also showed that the magnitude of the decrease in density with tamoxifen therapy was greater in younger women than in older women (–12.1% [SD, 11.0%] in women under age 50 compared with –5.7% [SD, 12.6%] in women age 50 and older). 19

In their study of premenopausal breast cancer cases, Ursin et al 4 showed that 80% of women experienced a decrease in MBD. However, not all studies saw differences between pre- and postmenopausal women. 22

Because a similar dose of tamoxifen (20 mg/d) was used across studies, it was not possible to evaluate dose response with MBD change. Common to all of these studies was the fact that not all women on tamoxifen experienced a reduction in breast density with therapy; the proportion of women experiencing a reduction in density ranged from 21% to 80%.

This was seen across studies with varying MBD estimation methods, populations (whether high-risk women or cancer cases), and as noted above, varying menopausal status. This suggests the hypothesis that decreases in MBD induced by tamoxifen in a portion of women have clinical significance, resulting in fewer incident cancers among high-risk women and fewer instances of breast cancer recurrence or contralateral events among cases. These women stand to benefit by remaining on tamoxifen. In contrast, those who see little reduction or even increases in MBD may be those who would benefit from alternative treatment approaches.

Raloxifene, another SERM agent, has been used for chemoprevention of breast cancer due to the demonstrated reduction in risk of invasive breast cancers after 5 years of therapy. 23 Some studies have shown similar decreases in MBD among women on raloxifene vs placebo, 24 whereas others have shown small changes in MBD with raloxifene use. 15

In a study of raloxifene and MBD assessed as volumetric breast density from full-field digital images, Eilertsen et al 15 showed a small reduction in volumetric MBD in the raloxifene group (median, –4.1%; 95% CI, –6.9%-2.1%), compared with an increase in MBD seen in the low-dose PMH group (median, 15.0%; 95% CI, 4.8%-28.6%; P < .0001).

Aromatase inhibitors (AIs) block local synthesis of estrogen in extracts of human breast tumors 25 and currently are the most efficacious endocrine therapy for estrogen receptor–positive postmenopausal breast cancer.

A few studies have examined the association between the AI, letrozole, and MBD, with mixed results. These include a study of 106 postmenopausal women who were randomized to either letrozole or placebo after 5 years of tamoxifen; this study found no difference between the 2 groups in change in MBD after 9 to 15 months. 26 Two additional studies examined the influence of letrozole on MBD among postmenopausal women taking PMH: one found no change in percent MBD among 42 high-risk women on either estrogen alone or combination PMH (estrogen and progestins) after taking 2.5 mg letrozole per day for 6 months, 27 whereas the other study found a reduction among women on low-dose combination therapy who were taking letrozole (2.5 mg) 3 times weekly for a median of 24 (range, 2-63) months (6.8% vs 1.4% reduction). 28

The inconsistent findings of studies of raloxifene and aromatase inhibitors with MBD may be related to the fact that these therapies are used only in postmenopausal women with lower baseline MBD, making small changes hard to detect. However, larger studies with well-calibrated density measures and follow-up for breast cancer are needed to determine whether MBD can be used as a biomarker for these and other endocrine therapies.

MBD as a clinical marker for breast cancer risk

How can this information be translated to the clinical care of our patients? Studies have recently incorporated the BI-RADs and quantitative MBD measure into breast cancer risk prediction models, which has shown some improvement in risk prediction (improvement c-statistic by 0.01 to 0.06). 29 An enhanced model with an MBD measure is, therefore, preferable to the currently existing gail model but remains poor for individualized risk. Also, it is important to recognize the clinical challenges associated with evaluating mammograms of women with increased MBD, including the need for repeat mammograms and breast biopsies and the difficulty of detecting clinically significant abnormalities. 30

Given the associations between exogenous hormones and a change in MBD, a natural question is whether a change in MBD is a potential marker for risk. In other words, if women experience increased MBD with PMH use, are they at greater risk for breast cancer than women who decrease or maintain MBD while on PMH? Or, if a woman on adjuvant tamoxifen decreases MBD, does this mean that she will have a reduced risk of recurrence compared with a woman who has no change in MBD while on tamoxifen?

Cuzick et al reported the first study to address this question, using women from the IBIS-1 trial, a clinical trial randomizing high-risk women to tamoxifen or placebo and following them over time for breast cancer events. 31 Mammograms were examined at study entry and at 12 to 18 months for 120 participants who developed breast cancer over the course of the trial and 945 matched controls who did not. MBD was assessed visually by an expert radiologist as a percent of the total breast area. Compared with the placebo arm, women on tamoxifen with ≥10% reduction in MBD had lower breast cancer risk (odds ratio [OR], 0.37; 95% CI, 0.2-0.69; P=.002), whereas women on tamoxifen with <10% reduction in breast density showed no decrease in risk (OR, 1.03; 95% CI, 0.66-1.61; P=.89).

These data suggest that change in MBD could be a biomarker for breast cancer risk reduction for women undergoing chemoprevention strategies and could allow for earlier identification of women who would not benefit from tamoxifen therapy. Studies are also under way to examine the influence of change in MBD with PMH use on breast cancer risk, and results are expected soon. At this time, it is not clear whether the resulting increases and decreases in MBD with endocrine therapies simply change the ability to detect new breast cancers or relate to the pathophysiology of the breast cancer. Additional work is needed to improve our understanding of agents that contribute to change in MBD and their association with breast cancer. Also, the development of reproducible and well-calibrated density measures that can accurately measure MBD are needed to allow for comparability of MBD change across studies and therapy types.

Summary

Understanding variability in the response to endocrine therapies is important so that the most effective therapy can be administered to patients in a timely manner. This includes the administration of exogenous estrogens to healthy women to reduce postmenopausal symptoms as well as breast cancer treatment for women with disease. MBD may contribute to our understanding of interindividual variability in response to therapy. This information can then translate to the clinical setting to facilitate personalized decision-making regarding options for breast cancer treatment and risk-reduction strategies.

Sexuality, Reproduction & Menopause ©2010 Quadrant HealthCom Inc.