The breasts of healthy women differ considerably in the proportion of stromal and epithelial tissues to fat tissue that they contain. Since fat is less dense than the other tissues, this proportion is reflected in the overall density of the breast tissue. Dense breast tissue is recognised to be a risk factor for developing breast cancer. Most of the variation in breast density is controlled by genetic factors rather than by age, parity, BMI or hormone levels, but the genetic loci involved are so far unknown.

A large, international group of researchers led by Rulla Tamimi of Harvard University, Boston, USA, has now conducted a meta-analysis of five genome-wide association studies of percent mammographic density within the Marker of Density (MODE) consortium [1]. All studies measured breast density from mammograms using the Cumulus software and adjusted the readings for age and BMI. A total of 4877 women were involved altogether, all of self-described European origin and 89% post-menopausal at the time of the mammogram.

Although differences in design between the five studies complicated the interpretation of the results, combining P values yielded six SNPs within the same linkage disequilibrium block that were significantly associated with differences in mammographic density. These lay within the region of the gene ZNF365, which is located on the long arm of chromosome 10 and encodes a zinc finger protein. The most significant result was for the SNP rs10995195, in which the C allele was associated with lower breast density.

A previous genome-wide association study had shown another SNP in this genetic region, rs10995190, to be associated with the risk of developing breast cancer [2]. The A allele of this SNP is in high linkage disequilibrium with the C allele of rs10995195, and Tamimi and her co-workers ranked it third in their list of variants associated with breast density. They investigated the link between rs10995190 and mammographic density further using samples from 1,690 breast cancer patients and 1,145 women without breast cancer from the Mayo Clinic Breast Cancer Study and the Sisters in Breast Screeing Study respectively. This study replicated the association between this SNP and breast density, giving a combined P value of 9.6x10¹⁰.  Adjusting for breast cancer status or excluding cases of breast cancer made little difference to the results. The researchers then estimated the association between rs10995190 and breast cancer before and after adjusting for breast density, and concluded that variations in this gene might influence breast cancer risk through their effect on the density of breast tissue.

The researchers acknowledged that the significance of their results was limited by measurement errors in estimating breast density from mammograms, and by differences in measurement techniques between the studies pooled in the meta-analysis. Nevertheless, they were able to conclude that an association between rs10995190, a SNP already associated with breast cancer risk, and breast density was highly statistically significant. They recommended larger genome-wide association studies to identify further genetic variants associated with breast density and thus breast cancer risk.

References

[1] Lindström S, Vachon CM, Li J and 32 others (2011). Common variants in ZNF365 are associated with both mammographic density and breast cancer risk. Nature Genetics, published online ahead of print 31 January 2011. doi:10.1038/ng.760

[2] Turnbull C, Ahmed S, Morrison J and others (2010). Genome-wide association study identifies five new breast cancer susceptibility loci. Nature Genetics 42(6): 504-7.