Interplay of BMI and volumetric breast density measures and breast cancer risk for black and white women
Mattia A. Mahmoud; Stacey J. Winham; Christopher G. Scott; Sarah Ehsan; Aaron D. Norman; Matthew R. Jensen; Emily F. Conant; Karla M. Kerlikowske; Despina Kontos; Celine M. Vachon; Anne Marie McCarthy
Breast Cancer Research, 2025
Abstract
Background
Black women have lower Breast Imaging Reporting and Data System (BI-RADS) breast density than White women, likely due to body mass index (BMI) differences. No studies have directly compared BMI, race, and volumetric breast density in relation to breast cancer risk. This study examines the associations between BI-RADS density and Volpara-derived volumetric breast density measures and breast cancer risk in non-Hispanic Black and White women, focusing on BMI’s influence.
Methods
A nested case-control study was conducted with 3699 women (526 Black, 3173 White) from Mayo Clinic and the University of Pennsylvania. Invasive breast cancer cases (n = 1013) were matched with controls (n = 2686). Breast density was assessed using BI-RADS density categories and continuous density measures from Volpara, including dense volume and volumetric percent density. Conditional logistic regression was used to evaluate associations between density measures and breast cancer risk, adjusting for BMI and age.
Results
Black women had higher dense volume but lower volumetric percent density and BI-RADS density compared with White women. All density measures were significantly associated with breast cancer risk in both groups, with stronger associations after BMI adjustment for BI-RADS density and volumetric percent density. BI-RADS density showed a stronger association in Black women (OR 2.06, 95% CI 1.45–2.91) than in White women (OR 1.55, 95% CI 1.38–1.74) when adjusted for BMI (p-interaction = 0.04). Dense volume showed similar predictive value for both groups, regardless of BMI adjustment.
Conclusion
Using BI-RADS density categories to assess breast cancer risk requires adjustment for BMI for equitable comparison of predictive values across race. Associations of dense volume are not altered when BMI is included as an adjustment factor.