Breast Density and Metabolic Risk Factors: A Cross-Sectional Analysis of a Phase II Study in Premenopausal Women with Elements of Metabolic Syndrome

Abstract

Background Breast density is an established breast cancer risk factor. Metabolic disturbances and high adiposity also increase breast cancer burden, but their relationships with breast density are not defined. This dissertation research aims to provide more evidence of the associations of metabolic risk factors with breast density measures in a cohort of premenopausal women with elements of metabolic syndrome. Experimental Design We performed this dissertation research within the context of a Phase II clinical trial conducted at the University of Arizona Cancer Center in overweight or obese premenopausal women with metabolic dysregulations. The primary aim of the trial was to evaluate if metformin intervention for 12 months can exert changes in breast density. A cross-sectional analysis was performed using the baseline data of the study cohort to evaluate the associations of breast density measurements with metabolic disturbances. The breast density measures, acquired by fat-water MRI, included absolute dense volume, percent density, non-dense volume, and total breast volume. The measures of metabolic disturbances included anthropometric measures, elements of metabolic syndrome, insulin/insulin-like growth factor (IGF) axis, and adipokines. Results Our findings indicate that each breast density measurement is non-normally distributed and comprised of distinct subpopulations with normal distributions. We correlated the breast density measurements with anthropometric measures. Those relationships were unaffected by ethnicity but were affected by age and reproductive factors. Total breast volume and non-dense volume were positively related to waist circumference, a measure of central adiposity, after adjustment for potential confounders (i.e., adiposity, ethnicity, age, reproductive factors). These two density measures were also positively correlated with serum leptin levels following adjustment of potential confounders, including waist circumference, but not with other measures of metabolic disturbance. Individuals with elevated fasting glucose had lower absolute dense volume. The insulin and HOMA-IR (Homeostatic Model Assessment for Insulin Resistance) were inversely related to the absolute dense volume after adjustment for potential confounders. In addition, we observed a positive correlation between the absolute dense volume and serum leptin levels. Individuals with elevated fasting glucose had a lower percent density. Individuals with elevated fasting glucose had lower percent density but showed no correlation with insulin and HOMA-IR in analysis adjusted for potential confounders. Furthermore, percent density was positively related to a measure of bioavailable IGF-1 (molar ratio of IGF-1/IGF binding protein 3). Conclusion The heterogeneity of breast density measures indicates that there is a wide range of values within our cohort. The observed associations between breast density measures with selected metabolic disturbances suggest the importance of considering these metabolic disturbances when evaluating breast density measures for risk assessment in women with elements of metabolic syndrome. Further investigations are needed to understand the mechanisms responsible for the observed associations.