Mammographic Density, HRT and Hormonal Activity Genes

Institution: University of Southern California
Investigator(s): Thomas Mack, M.D., M.P.H. -
Award Cycle: 2000 (Cycle VI) Grant #: 6PB-0052 Award: $708,225
Award Type: Request for Applications
Research Priorities
Etiology and Prevention>Prevention and Risk Reduction: ending the danger of breast cancer



Initial Award Abstract (2000)
Currently available risk factors for breast cancer are not very useful for identifying high-risk women on a population basis. Several large epidemiologic studies have suggested that the level of mammographic density is highly associated with breast cancer risk. Evidence suggests that hormones taken as menopausal replacement may influence breast density, however details are lacking. It is also likely that breast density is partly inherited, but again details are lacking. We are now in the process of analyzing data from 1200 pairs of twins to confirm this observation.

While the genes controlling mammographic density have not yet been identified, it is likely that they have something to do with control of estrogen levels. Preliminary evidence suggests 4 genes that might be involved in estrogen metabolism and therefore mammographic density. We are in a unique position to examine these issues because we are completing a study in which we have collected mammograms from 1200 pairs of twins. We now propose to re-contact these twins to determine the exact dose and timing of their hormone replacement therapy (by contacting their physicians), and to collect DNA to assess the relationship between specific genes and mammographic density. We also propose to collect additional mammograms from these individuals to assess the effect of hormone replacement therapy by examining mammograms before and after HRT use.

Twins offer special advantages for several types of comparisons because they share the same genes and the same early environment. We will compare mammographic density between twins who are taking various kinds of HRT, and their co-twins who are not. Since identical twins share the same genes, we can attribute any differences in mammographic density to the effect of hormones (after adjusting for other pertinent characteristics). We will also compare mammographic densities in identical and fraternal twin pairs with different estrogen metabolism genes, to determine the effect of these genes on density. By understanding the determinants of mammographic density, we can sharpen our ability to identify women at especially high risk and hopefully better design studies helping us to prevent the occurrence of breast cancer in more women.


Final Report (2004)
The overall density of an ordinary mammogram, quite apart from the appearance of any dense area suggestive of malignancy, has been shown to be a strong predictor of future breast cancer. The level of density is known to vary with menopausal status and sequential alteration of exogenous hormone use. We have previously shown that mammographic density is also a strongly heritable trait.

The current grant has enabled us to gather DNA from a large number of identical and fraternal female twins on whom we have already characterized mammographic density. We can now select, prioritize, and test those genes that might be responsible for mammographic density and therefore might play a role in the cause of breast cancer. To do so we will compare the genetic characteristics of identical twins with similarly dense mammographic images to the genetic characteristics of identical twins with similarly clear mammograms. We have also gathered a history of hormone usage from these same women. Whereas women can reliably report the use of estrogen and/or progestin, they often cannot remember the precise periods of usage and the particular formulation. The second goal of this grant was to contact the physicians of the twin women who had given us information about their past hormone use in relation to the time of mammograms, in order to verify the details of that history. We have accomplished that goal by receiving the details of the prescriptions from the physicians of the twin subjects.

Although both the DNA and the prescription information from physicians has been gathered, the actual genetic analyses are not yet complete. We have verified that the genetic determinants of mammographic density seem to largely override the history of hormone usage. We will now carry out the genetic analysis as described, and compare the genetic effects with the effects of hormone usage. By doing so we expect to learn whether the appearance of the mammogram is a better guide to the genetic causes of breast cancer or to the hormonal causes. We hope that in the long run these characteristics can be used to identify those women at higher risk of breast cancer, and advise them how to minimize their risk.