Genetics, Mammographic Density and Breast Cancer Risk

Hongjie Chen | 2020

Advisor: Sara Lindstroem

Research Area(s): Cancer Epidemiology, Clinical Epidemiology, Genetic Epidemiology

Full Text

Mammographic, or breast density, describing breast tissue composition, is one of the strongest risk factors for breast cancer, the most common cancer type among women in the US. In general, women in the highest breast density quartile are estimated to have a three to five-fold higher risk of developing breast cancer, compared to those in the lowest. Despite this strong association, the underlying biological mechanism remains unexplained. The association between mammographic density and breast cancer risk is not homogeneous in the general population. Previous studies suggest that menopausal status and postmenopausal hormone use may interact with high breast density to modify breast cancer risk. To our knowledge, only one previous study in postmenopausal women has focused on the interaction between breast density and other risk factors in relation to breast cancer risk. Further, it is not known whether established risk factors affect breast cancer risk differently by breast density among premenopausal women, or if certain risk factors interplay with breast density to alter the risk of specific breast cancer subtypes with worse prognosis.

In addition to being associated with non-genetic factors such as age, parity, body mass index (BMI) and menopausal status, mammographic density is highly heritable, with approximately 60% of the variation attributable to genetics. However, genetic variants identified by previous genome-wide association studies (GWAS) only explain up to 1% of the variation in density. The limited number of identified mammographic density loci is likely a consequence of limited sample sizes. Meanwhile, previous GWAS have focused on genome-wide significant signals, failing to further investigate the potential importance of genetic variants with comparatively weak effects. Estimating the heritability explained by common variants throughout the genome regardless of their marginal association statistics can provide a better understanding of the genetic basis of the traits.

In this dissertation, we attempted to extend our knowledge of mammographic density as an intermediate phenotype of breast cancer, from both environmental (Chapter 1) and genetic (Chapter 2) perspectives. The mammographic density traits studied in this work are the area of dense (epithelial, stromal and muscle) tissue, non-dense (adipose) tissue, and percentage of dense tissue, which have all been demonstrated to be significantly associated with breast cancer risk independently.

Specifically, we assessed multiplicative interactions between mammographic density and thirteen established breast cancer risk factors, in relation to risk of breast cancer overall and by molecular subtypes, using a case-control population nested under Nurses’ Health Studies I&II (Chapter 1). Although no interaction remained statistically significant after adjusting for number of comparisons, we did observe some noteworthy interactions with nominal significance at p<0.05 level.

Next, we performed a meta-analysis of GWAS for mammographic density measures, using genotype and phenotype data collected by the Marker of Density (MODE) consortium and the Breast Cancer Association Consortium (BCAC). We also conducted exploratory bioinformatics analyses, including genetic correlation analyses and a transcriptome-wide association study (TWAS) (Chapter 2). Using data on up to 27,900 women of European ancestry, we identified twenty-eight distinct genome-wide significant loci for mammographic density measures, including nine novel signals (5q11.2, 5q14.1, 5q31.1, 5q33.3, 5q35.1, 7p11.2, 8q24.13, 12p11.2, 16q12.2) that have not been reported before. Fourteen loci identified were also observed with significant association with the risk of overall, ER-positive and/or ER-negative breast cancer. Subsequent exploratory analyses improved our understanding of a shared genetic basis between mammographic density and breast cancer.

In summary, this work can help understand the underlying mechanisms driving the mammographic density breast cancer association as well as the genetic architecture of mammographic density. Findings discovered through this dissertation will help to better understand the determinants of breast density, which in turn can help identify strategies that reduce the risk of breast cancer, and ultimately mitigate the burden brought by the disease.