Body Mass Index, Breast Tissue, and the Epigenome

Purpose: Obesity is a known risk factor for breast cancer. Though the mechanisms underlying the relationship remain to be elucidated, biologic pathways involving hormones, inflammation, and insulin resistance have been implicated. The goal of this dissertation was to explore the relationships between body mass index (BMI), gene methylation, and gene expression in breast tissue to better understand how BMI influences breast tissue. Methods: Methylation values were calculated in two study populations. Patients from the Normal Breast Study (NBS) contributed non-diseased tissue, while patients from the Carolina Breast Cancer Study (CBCS) contributed tumor tissue. Multivariable linear regression was used to identify which probes were associated with BMI and, of those, which were correlated with gene expression. Pathway analyses were conducted to determine which biologic pathways were enriched among BMI-associated methylation sites. Results: Of the 431,568 sites analyzed for the non-diseased tissue sampled from participants of the NBS, BMI was associated with 2,573 probes (false discovery rate q-value <0.05). Inflammatory and insulin signaling pathways were enriched among the 2,573 probes. Of the 1,251 BMI-associated sites that had linked gene expression data, 226 were correlated with gene expression. Of the 935 probes analyzed for the tumor tissue sampled from participants of the CBCS, 2 were associated with BMI in the full dataset and 21 were associated with BMI among estrogen-receptor positive tumors (false discovery rate q-value