Abstract IA19: Elucidating the origins of aggressive breast cancer in African American women

Abstract

Abstract Although American women of European ancestry (EA) have higher incidence of breast cancer, African American (AA) women are more frequently diagnosed with tumors with more aggressive characteristics, particularly lack of expression of the estrogen receptor (ER), triple negative status (negative for ER, PR, HER2), and basal-like subtype. Until recently, the reasons for these disparities have been unknown. However, with more attention turned to this pressing question, better understanding of the origins of aggressive breast cancer in AA women is beginning to unfold. We and others have been examining risk factors for breast cancer subtypes in AA women and, in 2011 we formed the AMBER Consortium, pooling data from several large studies of breast cancer in AA women. There had been accumulating data from smaller studies suggesting that, while parity reduces risk of ER positive breast cancer, it actually increases risk of ER negative and triple negative breast cancer. Importantly, breast feeding appears to greatly reduce this parity-associated breast cancer risk. We recently confirmed these associations in the AMBER Consortium, including data from almost 4,000 AA women with breast cancer and 14,000 controls, finding that, compared to women who had one birth and breastfed, ER negative cancer risk increased with each additional birth in women who had not breastfed, with an OR of 1.68 (95% CI = 1.15 to 2.44) for 4 or more births. We also examined associations between ER negative breast cancer and other epidemiological risk factors. Early age at menarche has long been considered a risk factor for breast cancer and, when we evaluated by ER status, we found that was true for ER negative disease, with risk increased with early menarche among both parous and nulliparous women. However, for ER positive breast cancer, early menarche only increased risk among women who had children, and the most important factor was the length of time between menarche and first full-term birth. These findings indicate that menarche, in and of itself, may be a critical event in the development of ER negative breast cancer whereas for ER positive disease, the time span between reproductive events is most important, suggesting differing mechanisms in the development of this heterogeneous disease. AA women are more likely to have early menarche, to have larger families, and not to breastfeed than EAs; these factors could contribute to their higher rates of aggressive breast cancer. This line of reasoning also led us to approach the etiology of aggressive breast cancer in AA women at the molecular level. Our driving question was: is ER negative breast cancer the same in AA and EA women, but the risk factors are more common among AA women?; or, is ER negative disease different between groups? We began to address this by performing a two-stage genome-wide methylation study using the Illumina Infinium Human methylation 450K platform (450K), first in a smaller study using fresh frozen samples from AA and EA women (n=138), and then in FFPE tissues from 733 EA and AA patients. In both studies, we found that hierarchical clustering separated groups by ER status, but not by race. We also found that there were nearly double the number of differentially methylated loci between EAs and AAs in ER negative breast tumors compared to ER positive. We also performed RNASeq to determine relationships between gene expression, DNA methylation, and ER status. Examination of the top differentially methylated loci showed that FOXA1, which affects the expression of a multitude of genes important for maintaining the luminal phenotype of mammary epithelial cells, was methylated and significantly downregulated in ER negative tumors from AA women compared to those from EA breast cancer patients, which could facilitate a more aggressive basal-like state. MetaCore pathway analysis of genes associated with differentially methylated loci by race revealed that, in ER negative breast cancer, many of the enriched pathways were in those involving immune response, particularly inflammation. Most of these genes tended to be hypo-methylated in ER negative tumors from AA women compared to ER negative tumors in EAs. When we employed modified mixture modeling (RPMM) to explore the relationships between patterns of DNA methylation and other factors, DNA methylation class membership was significantly affected by race and parity in ER negative tumors, with no associations for ER positive disease. Taken together, these findings indicate that there are, indeed, inherent differences in ER negative tumors between AA and EA women. Additional research is underway to better elucidate the underlying reasons for race differences in the etiology of ER negative tumors, with a goal to ultimately prevent aggressive breast cancer in AA women. Citation Format: Christine B. Ambrosone. Elucidating the origins of aggressive breast cancer in African American women. [abstract]. In: Proceedings of the Eighth AACR Conference on The Science of Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; Nov 13-16, 2015; Atlanta, GA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2016;25(3 Suppl):Abstract nr IA19.