Skip to main content
         

Vol. 04 Issue 2, Spring 1999

Research Commentary: Mammographic Densities and Breast Cancer Risk
The Ribbon 

Norman F. Boyd, Gina A. Lockwood, Jeff W. Byng, David L. Tritchler and Martin J. Yaffe. Cancer Epidemiology, Biomarkers & Prevention, Volume 7, 1133-1144, December 1998.

Mammography may provide women with more information than simply detecting potential/suspicious growths and tumors. The overall x-ray image of the female breast obtained in a mammogram is different for each individual woman due to differences in the relative amounts of fat, connective and epithelial tissue. These different radiological patterns created by variations in the relative amounts of these tissues, referred to as the parenchymal patterns of the breast, have been known to be associated with risk of breast cancer. Early studies that observed this association relied on a subjective, qualitatively based classification of breast parenchyma in terms of four distinct patterns from a radiologically lucent pattern (fat) that appears dark on a mammogram, to a more radiologically dense pattern (connective and epithelial tissue) which appears light on a mammogram. Subsequently, researchers developed quantitative methods to assess the proportion of the breast occupied by radiologically dense tissue based on visual examination, planimetry and digitized images with computer-assisted methods to measure breast density, considerably improving on measurement techniques and methods.

In this manuscript, the authors present a comprehensive and complete review and summary of research conducted between 1976 and 1997 on the relationship between breast density and the risk for breast cancer. Eight out of the nine studies (five conventional case-control studies and four nested case-control studies) conducted to date using quantitative assessment of breast density reported a dose-response relationship or statistically significant trend of increasing risk of breast cancer across categories of density analyzed in each study. Radiologically dense breast tissue is not only reported to be associated with a large increase in the relative risk (RR) of breast cancer but also appears to be present in a substantial proportion of subjects with the disease. The authors estimate that 28-33% of breast cancer cases may be attributable to dense tissue in >50% of the breast. They further propose that dense breast tissue indicates proliferation of the breast epithelium and stroma in response to growth factors induced by circulating levels of sex hormones.

There are however, several caveats regarding this potentially important finding. The first plausible argument that calls into question the association of breast density and risk is the "masking" effect of breast density in the detection of tumors. Breast cancer is easiest to detect by mammography in breast tissue with radiolucent parenchyma and most difficult to detect in breast tissue with dense parenchyma. Therefore it appears likely that more cancers may not be detected at first examination in subjects with dense breast tissue and will be detected subsequently. However, very similar estimates of risk obtained in case-control studies (mammograms taken at diagnosis) and cohort studies (mammogram taken at baseline entry to cohort) together with the persistence of risk over extended follow-up in cohort studies, suggests that "masking" does not distort estimates of risk and may not pose a problem in these studies. Furthermore, research on a cohort of subjects regularly examined over an extended period of time, has shown that any effect of "masking" on risk estimates will be small and short lived because cancers missed on one examination will eventually be detected at a later examination.

Mammographic density has been shown to be associated with several other risk factors for breast cancer. Most of these relationships are consistent with findings from observational studies on these other risk factors such as age, body weight, family history, parity etc. However certain anomalies remain and further investigation of these inconsistencies is warranted. For example, there is an apparent paradox in that breast cancer incidence increases with age, being higher in postmenopausal versus premenopausal women, yet dense breast tissue is more common before rather than after menopause. Thus, the decline in density with increasing age suggests that it is density at a given age, rather than density per se, that is the relevant measure with respect to breast cancer. The authors correctly point out that studies of density as a risk factor must therefore compare women of the same age. Another important risk factor for breast cancer is body size. Body weight and body mass index have been repeatedly shown to be inversely associated with breast density. This is consistent with the inverse effect of body size with premenopausal breast cancer but is at odds with the observation that obesity is a risk factor for postmenopausal breast cancer. Similar findings on family history and density are equivocal and do not indicate a clear relationship.

Very few studies have been conducted to examine the relationship between breast density and readily modifiable risk factors such as nutrition and exercise. The few studies that have been reported suggest that diet, particularly the intakes of total and saturated fat, may have a causal role to play in the etiology of breast density. The authors suggest that breast density is an independent risk factor for breast cancer as it remains associated with risk after adjusting for the effects of other risk factors. However, based on arguments presented in this review and current understanding of the etiology of breast cancer, breast density is better viewed as an intermediary biological marker for disease propensity. For the immediate future, radiological characteristics of the breast might be used to determine the length of the interval between mammographic screenings, and breast density can be effectively studied as an outcome measure or proxy for breast cancer risk. This report clearly underscores the need for further research to identify the dietary and hormonal factors that modulate and influence breast density and establish the biological and nutritional mechanisms that influence mammographic density.

Back to the top

Prepared by Banoo Parpia, Ph.D, Senior Research Associate,
Division of Nutritional Sciences, Cornell University

Last Update 06.23.00