Brad M. Keller PhD, Presenter: Nothing to Disclose

Anne Marie McCarthy, Abstract Co-Author: Nothing to Disclose

Jinbo Chen PhD, Abstract Co-Author: Nothing to Disclose

Nigel Sloan Bristol, Abstract Co-Author: Nothing to Disclose

Marie Synnestvedt, Abstract Co-Author: Nothing to Disclose

Emily F. Conant MD, Abstract Co-Author: Scientific Advisory Board, Hologic, Inc

Despina Kontos PhD, Abstract Co-Author: Nothing to Disclose

We investigate the association between measures of area and volumetric breast density and body mass index (BMI) in a large cohort of women screened with digital mammography. Understanding the magnitude and directionality of this association can aid in the assessment of confounding by BMI on the association between breast density and breast cancer risk.

We retrospectively collected and analyzed digital mammograms from a total of 3041 women (mean age 56.9y±10.9y) who had recorded BMI measures and raw (‘For Processing’) images available from a total of 10751 consecutive women screened for breast cancer over the course of one year at our institution (2010-11). A previously-validated fully-automated software algorithm developed at our institution was used to generate per-woman estimates of absolute dense area and area percent density. Volumetric estimates of absolute and percent dense tissue were obtained using FDA-cleared software (Quantra™, Hologic, Inc). The Pearson correlation (r) was used to assess the association between the breast density measures and BMI, and the coefficient of determination (R2) was used to determine the percentage of breast density variation explained by age and race-adjusted BMI.

BMI has a moderate, negative association with percent area (r=-0.34, p<0.001) and percent volume density (r=-0.31, p<0.001), a moderate positive association with absolute dense volume (r=0.47, p<0.001) and a weak association with absolute dense area (r=0.08, p<0.001). After adjustment for age and race, the strongest overall association is between BMI and absolute dense tissue volume (R 2 =0.26; p-value <0.001); whereas BMI is no longer associated with absolute dense area (R2=0.02; p-value=0.08).   

Breast density is significantly associated with BMI; however, only up to 26% of the observed differences in breast density between women can be attributed to differences in their BMI, age and race. Furthermore, given the negative association between percent density measures and BMI, yet their positive associations with risk, not considering BMI likely leads to an underestimation of the risk for breast cancer due to a woman’s breast percent density.

A woman’s BMI should be routinely collected and incorporated into risk assessment models, especially when also using breast density, in order to more accurately measure risk for breast cancer.

Keller, B, McCarthy, A, Chen, J, Bristol, N, Synnestvedt, M, Conant, E, Kontos, D, Associations between Body Mass Index and Breast Density: Implications for Breast Cancer Risk Assessment.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14045424.html