M. M. Goldbach1, D. I. Hoffman1, A. Malinovitch1, C. Huang1, M. Pomponio1, A. D. Williams1,3, S. M. Nazarian1, J. Tchou1,2 1Perelman School of Medicine at the University of Pennsylvania,Department Of Surgery,Philadelphia, PA, USA 2Perelman School of Medicine at the University of Pennsylvania,Abramson Cancer Center, Rena Rowan Breast Center,Philadelphia, PA, USA 3Lankenau Medical Center,Department Of Surgery,Wynnewood, PA, USA
Introduction: Mammographic breast density (BD) is an established risk factor for breast cancer. The relationship between BD and breast cancer receptor subtype is unclear. We therefore aimed to evaluate differences in BD across tumor receptor subtypes at a large, academic medical center. Several studies have shown that triple negative breast cancers (TNBC) often present as interval breast cancer, i.e in between normal screening mammograms. We hypothesized that women with TNBC have denser breasts compared to women diagnosed with other breast cancer subtypes as classified by their hormone receptor (HR) and human epidermal growth factor receptor 2 (HER2) expression status.
Methods: A retrospective database query from an institutional registry was performed to identify female patients diagnosed with primary breast cancer between 2009 – 2015 who also had a screening mammogram 6 ± 3 months prior to diagnosis at our institution. Patients were excluded if their breast cancers were noninvasive, metastatic at diagnosis, or had no receptor status information available. Patient demographic information, BD, and receptor status were collected. Categorical BD was measured using the Breast Imaging Reporting and Data System (BI-RADS) classification system (1 = almost entirely fatty, 2 = scattered fibroglandular densities, 3 = heterogeneously dense, 4 = extremely dense). A Chi-square test was performed to evaluate differences in BD across three tumor subtypes: HR+/HER2-, HER2+ (regardless of HR status), TNBC.
Results: BD was assessed in 488 patients with invasive breast cancer. The cohort had a median age of 62 years (range 38 – 92). 58.2% of patients (284/488) were Caucasian, 37.9% (185/488) were Black, and 3.1% (15/488) were Asian. 78.5% of patients (383/488) had ER+/HER2- cancer, while 11.9% (58/488) had TNBC and 9.6% (47/488) had HER2+ cancer. The distributions of categorical BD between tumor receptor subtypes were summarized in Table 1. Overall, approximately 90% of patients (448/488) had a BI-RADS score of either 2 or 3. BD did not differ significantly across tumor receptor subtype (X2 2.88, p=0.823).
Conclusion: In this pilot study, we report no significant difference in categorical breast density between breast cancer receptor subtypes. As categorical BD reporting is qualitative and has demonstrated inter- and intra-observer variability, we plan to quantify BD using our BD quantification software – Laboratory for Individualized Breast Radiodensity Assessment (LIBRA). We also plan to expand the cohort to increase our sample size. A multivariate logistic regression to control for other patient variables, including age, menopausal status, BMI, history of LCIS, and use of hormone replacement therapy is underway.
