Breast Density classifications according to the BI-RADS lexicon.
Breast density is thought to be a factor of increased risk for breast cancer. The idea was originally met with skepticism but consensus now is that it is indeed a true risk element, and of interest to screening mammography. Essentially, breast density is a comparison of the relative amounts of fat versus fibro-glandular tissues in the breast. Women with 'dense' breasts have a higher percentage of fibrous tissue and less fat tissue.
The BI-RADS classification system identifies four levels of breast density in keeping with relative increases in the amount of levels of fibro-glandular tissue. These are:
Breast Density: BI-RADS type 1.
The first 'type' classification of breast density is of almost entirely fat. Glandular tissue is less than 25%.
Breast Density: BI-RADS type 2.
In type 2 breast density, there are scattered fibroglandular tissues, ranging from 25%-50% of the breast.
Breast Density: BI-RADS type 3.
The breast tissue in type 3 may be termed 'heterogeneously dense', ranging 51%-75% of the breast tissue. 'Heterogeneous' means something contains many different items and has many different variations. With respect to breast density it implies that the fibrous tissue is prevalent throughout the breast, but not clustered together.
Breast Density: BI-RADS type 4.
The highest category of breast density, 'type 4', means that the breast contains greater than 75% glandular and fibrous tissue. At this level the sensitivity of the mammogram might even be reduced.
What exactly is 'breast density' measuring?
There are many different types of tissues in the breast. With breast cancer development we often speak of malignant cells originating in 'epithelial' tissue, and that generally refers to linings and walls between different anatomical elements. There are 'glandular tissues', and these are collections of cells involved in the secretion of fluids. Within the breast the main glandular tissues are, obviously, those that produce milk, and also those that secrete sweat. "Fibrous" tissues are essentially collections of cells that give strength and resilience. Connecting and joining tissues between different elements are also 'fibrous'. "Stroma" is another term we often hear associated with breast cancer and breast tissues, and it basically refers to fibrous tissues, but specfically those which 'function' to support organs and glands, which 'hang' from them more-or-less.
So, high breast density refers to larger-than-average amounts of fibro-glandular cells and tissue, more than would normally be required. Amounts of fibrous and glandular tissues also tend to be quite proportionate, like a 'tree' with fibrous' branches' and glandular 'leaves'. High breast density is like a tree with many branches and lots of bushy leaves, as compared to a 'low density' tree with fewer branches and fewer leaves.
Breast density is really looking at the 'parenchymal' tissues. It is an old Greek term which means 'poured-in', which gives a clue as to what we are specifically addressing. The 'parenchymal' tissues of the breast are those essential for it's functioning, as distinct from those structures which encompass it (lining tissues) or support it (stroma). In the case of a breast, parenchymal tissues would include both glandular and fibrous tissues, but an overabundance; more than are normally found to support breast function in all women.
"Odds ratio" for high breast density.
The four categories of increasing breast density were adopted and modified by the American College of Radiology as part of an 'integrated' risk assessment and staging platform (BI-RADS), which considers much more than just breast density. It is difficult to isolate breast density in terms of an odds or risk ratio because, in order to be reliable, so many other variables need to be controlled or eliminated. However, even an inconclusive study can give general trends that can still help guage cancer threat and plan for follow-up and treatment.
When comparing the least-dense breast tissue categories with the most-dense, studies suggest the chances of developing breast cancer ranges from an odds ratio of 1.8 all the way up to 6.0. That is quite a bit of variability, but the majority of research still shows an odds ratio of 4 or higher. ( An 'odds ratio' of 4.0 means that a women with dense breasts is four times more likely to develop breast cancer than a women without dense breasts.)
There are a couple of plausible explanations for the increased tendency towards breast cancer. Firstly, the development of premalignant lesions such as atypical ductal hyperplasia is more likely in an 'elevated growth' environment. Secondly, a woman may have naturally overactive aromatase, which increases estrogen production within the breast, and may stimulate the growth of latent tumor cells.
The effect of hormone levels on breast density
It is a known fact that hormone levels can effect breast density in women. Predicting the influence of both natural hormones and supplemental hormone therapies is always a bit of a guessing game. After menopause, women's breasts become fattier, in other words, 'less dense'. In terms of the breast cancer risk associated with breast density, this ought to be positive thing. Some hormone treatments work against this, however.
Women who take estrogen to counter the effects of menopause run the risk of both increased breast density, and faster breast cancer growth if a tumor is already developing. Aromatase is an enzyme that synthesizes estrogen, and the biosynthesis of aromatase can stimulate the growth of fibro-glandular cells in the breast. To counter these processes, 'aromatase inhibitors' are sometimes prescribed to block the synthesis of estrogen.
Is 'breast density' a modifiable risk factor?
Breast density, along with 'obesity', seems to be one of those conditions observed in a certain percentage of women with breast cancer, which brings about speculations as to whether the cancer could have been prevented. Unlike obesity, breast density has been proven to actually increase the chance of developing cancer, whereas obesity is in almost all cases a concurrent and unrelated observation. The question remains, however, is high breast density something a woman can influence through her behavior, or is it simply genetic?
Although results are still inconclusive, there is some evidence to support a genetic predisposition towards dense breasts, and some researchers even claim to have found one or two specific genes responsible. It is far too early to say what percentage of women may be genetically predisposed, but it could be over 50%.
Obesity would seem to have an inverse effect on breast density, as women with a high body-mass-index tend to have an overabundance of fat breast tissue (hence a very low breast density) So, conclusions as to the genetic effects on high breast density would have to be correlated somehow with genetic predisposition towards obesity. The one cancels the other out, so to speak.
In terms of a women's voluntary behavior, aside from taking hormones as discussed above, there is some speculation that diet and alcohol intake can influence breast density. While there may be minimal evidence to support these suppositions, the differences in density between those restrict their diet or alcohol intake versus those who do not is so very small, it is not statisically relevant.
There has also been some speculation that women who live in urban areas may, in general, have a higher breast density than women who live in rural areas. Again, the tendency towards an increased breast density is genetically pre-determined. It may be the case that women living in urban centres are generally more inclined to take hormone supplementation than women in rural areas and this could possibly lead to the suggestion of average increased breast density in some studies.
It is worth noting, however, that the elevated risk of breast cancer risk remains high for at least 510 years after the mammogram first showed increased density. In order to measure any presumed effects of 'voluntary lowering' of breast density, one would have to follow the study over a 5-10 year period.
Breast Density in Laymans terms
After a mammogram, a woman may have her breast density reported to her in plain english rather than the BI-RADS types. If a report describes a breast as 'very dense', they are probably talking about the type 4, 75-100% dense. If it uses words like 'moderately' or 'somewhat' dense, it most likely refers to category III, or 50%-75% density. A type 2 breast density might be desribed as 'moderately' or 'somewhat' fatty, with extra parenchymal tissue in the 25-50% range. If the report uses words like 'somewhat' or 'partially' fatty, it probably referst to type 1 or the 0-25% category. The radiologist might even say 'entirely fatty', meaning a density of 0%.
Note, however, that radiologists are not required to describe mammographic density in their reports. Breast density is often not even mentioned unless the breast is dense enough to interfere with mammograms.
Are there any other ways to determine breast density, aside from the mammogram?
With the advent of computer processing, more specific ways of calculating and classifying breast density are constantly being developed. In addition to the basic mammogram X-ray, experiments are underway which calculate density based on the overall two-dimensional area of fibro-grandular tissue in the image, and also 'textural' models based on the statistical distributions of parenchymal tissues. Essentially, these more-detailed analyses create many sub-categories within the four basic 'types' of increasing breast density. It is not immediately clear as to what purpose is served by attempting to sub-classify, or more precicely, to 'quantify' mammogram densities, as the four types are already very useful and only part of an overall risk assessment. However, over time, following the outcomes of highly specific breast density measures may lead to statistically meaningful trends, and to earlier, more focused breast cancer treatments.
- American College of Radiology. Breast Imaging and Reporting Data System. Fifth edition. Reston V A, American College of Radiology 2004
- Obenauer S, Hermann KP, Grabbe E. Applications and Literature Review of the BI-RADS Classification. Eur Radiol (2005) 15: 1027-1036
- Halls, Steven. B., Mammographic density of mammograms; categories affecting risk. http://www.halls.md/breast/density.htm
- Petroudi, S., Kadir, T., Brady, M., Automatic Classification of Mammographic Parenchymal Patterns: A Statistical Approach.
- Proc. International Conference IEEE Engineering in Medicine and Biology Society, volume 1, pages 798-801, 2003
- K. Bovis, K. and Singh. S., Classification of mammographic breast density using a combined classifier paradigm.; Proc. Medical Image Understanding and Analysis, pages 177-180, 2002
- Wolfe, J.N., Risk for breast cancer development determined by mammographic parenchymal pattern. Cancer, 37:24862492, 1976.
- Harvey, J., and Bovbjerb, V., Quantitative Assessment of Mammographic Breast Density: Relationship with Breast Cancer Risk.January 2004 Radiology, 230, 29-41.
- Pankow JS, Vachon CM, Kuni CC, et al. Genetic analysis of mammographic breast density in adult women: evidence of a gene effect. J Natl Cancer Inst 1997; 89:549-556.
- Mokbel K (2002). "The evolving role of aromatase inhibitors in breast cancer". Int J Clin Oncol 7 (5): 27983
- Maskarinec, G., Takata, Y., Pagano, I., Lurie, G., Wilkens, L.R., Kolonel, L.N.; Epidemiology: Alcohol consumption and mammographic density in a multiethnic population. International Journal of Cancer (Dec. 2005) Volume 118 Issue 10, Pages 2579 - 2583
- Perry, NM. City-Dwelling Women at Greater Risk for Breast Cancer. 93rd Scientific Assembly and Annual Meeting November 25 - 30, 2007 Chicago.
Copyright Steven B. Halls, MD Last edited 05-December-2010