Different for malignancy, in women toward the denser end

Different parenchymal density patterns were first described by Leborgne(3) in 1953 and were subsequently classified in 1976 by Wolfe (4) as possiblerisk factors for breast cancer. Since its inception, the Breast ImagingReporting and Data System (BI-RADS) lexicon has classified mammographicdensity into four categories, with the percentage of each tissue density in thegeneral screening population estimated as follows: 10% of women havebreasts that are almost entirely fatty, 40% have scattered areas offibroglandular density, 40% have heterogeneously dense breasts, and 10%have extremely dense breasts (5). The fourth edition of the BI-RADS lexiconcreated quartiles for each of the four density categories (<25% glandular,25%–50% glandular, 51%–75% dense, and >75% dense, respectively) in anattempt to distribute the population more equally among the categories;however, these quartiles proved to make very little difference and wereremoved from the BI-RADS fifth edition.Data from the Breast Cancer Surveillance Consortium (BCSC) that werecollected from registries around the United States show that on 934,098negative screening mammograms obtained from 1994 to 2008, thedistribution of tissue densities was 9.0%, 44.1%, 38.3%, and 8.6% for fattybreasts, breasts with scattered areas of fibroglandular density,heterogeneously dense breasts, and extremely dense breasts, respectively(6).When discussing breast density, researchers combine the classifications of”heterogeneously dense” and “extremely dense” and categorize them as”dense,” whereas the patterns of “fatty” and “scattered areas of fibroglandulardensity” are labeled as “nondense.” Some researchers report that asrelatively low as 31%–43% of the general screening population have densebreasts (7). However, data from the BCSC registry suggest that 46.9% of thescreening population have dense breasts (6). As noted in the BI-RADS fifthedition, there is very little difference between the tissue patterns, and likelythe risk for malignancy, in women toward the denser end of the “scatteredDISCUSSIONBreast density and its implications inmammographic screening by masking effect andan independent risk factor for breast cancer.Masking EffectIt is an established truth that with increasing density mammographicsensitivity decreases, due to superimposition of overlapping radiopaquedense breast tissue on an underlying radiopaque cancer when thethree-dimensional breast is imaged in a two-dimensional plane. Thelanguage of the BI-RADS lexicon accounts for this by wording the densetissue patterns as “the breasts are heterogeneously dense, which mayobscure small masses” and “the breasts are extremely dense, whichlowers the sensitivity of mammography” (5). In two separate BCSCreports of film-screen mammography, mammographic sensitivitydecreased from a level of 85.7%–88.8% in patients with almost entirelyfatty tissue to 62.2%–68.1% in patients with extremely dense breasttissue (8,9).The masking effect of breast density leads to an increased percentageof interval cancers (cancers that manifest within 1 year of a normalmammogram) in women with dense breasts..The results of a meta-analysis by McCormack and dos Santos (8)confirm the masking effect of breast density. It has been suggested thatbecause dense breasts may make a woman more likely to be diagnosedwith an interval cancer, women with dense breasts might benefit fromshorter screening intervals (8). In a recently published retrospectivereview of 335 breast cancers detected at screening US, 263 (78%) wereoccult at mammographic review because they were obscured byoverlapping dense breast tissue, a finding that confirms the maskingeffect of dense tissue on breast cancer detection (9).Density as an Independent Risk FactorBreast density itself is an independent risk factor for breast cancer.Density refers to the amount of epithelial and stromal elements of thebreast, and breast cancers most commonly arise in epithelial cells.morethe amount of epithelial tissue in the breast indicates more chance that!4cancer may arise in one of the epithelial cells. Some researcherspropose that breast density may increase the risk for breast cancer byup to six times, and breast density is often reported to cause a fourfoldincrease in the risk for breast cancer in women with dense breasts(8,9,10). In a meta-analysis of studies that evaluated breast density asan independent risk factor for breast cancer, the relative risk associatedwith dense breasts was 2.92 for breasts that were 50%–74% dense and4.64 for breasts that were 75% or more dense (8). A linear increasingtrend in the relative risk for breast cancer with respect to increasingtissue density has been noted when density is measured quantitatively(8). In a summary of studies that evaluated breast cancer risk withrespect to quantitatively measured tissue density, the odds ratio of therisk for breast cancer ranged from 3.6 to 6.0 (7).The relative risk of breast density is much smaller than that of othermajor risk factors for breast cancer, such as age, family history,reproductive history, and genetic mutations. However, becausemammographically dense breasts are relatively common (approximately50% of the screening population), some authors have proposed that therisk factor of density alone contributes far more cancer risk to thepopulation than other much stronger but less common risk factors, suchas a significant family history or known deleterious genetic mutationssuch as BRCA mutations (7,8).Instead, given that nearly half of the population has dense breasts, itmakes more sense to compare the risk for cancer in women with densebreasts to the risk in the average patient (who will have a tissue densityapproximately halfway between the two middle categories of scatteredareas of fibroglandular density and heterogeneously dense breasts).When such a comparison is made,”the relative risk for cancer in women with heterogeneously densebreasts compared with the average woman is approximately 1.2, andthe relative risk for cancer in women with extremely dense breastscompared with the average woman is approximately 2.1 (10)”.Furthermore, the findings of a randomized case-control study conductedas part of the International Breast Cancer Intervention Study (IBIS)suggest that women whose breast tissue density decreased whileundergoing tamoxifen therapy had a greater reduction in breast cancerrisk than women whose breast density did not decrease to the samedegree (11). Similarly, in postmenopausal women undergoing adjuvanttamoxifen therapy after a diagnosis of breast cancer, a decrease inmammographic density has been associated with a survival advantagecompared with women who did not have a breast density change (12).!5Therefore, although a masking effect is definitely present in women withbreast cancer, there is at least some element of breast density as a riskfactor for breast cancer that is independent of other risk factors. Inaddition, although breast density measurements are inherentlyinaccurate (especially in the majority of studies that have reported onareas of fibroglandular density” pattern and women at the less dense end of