Abstract C43: Is the influence of high mammographic breast density on breast cancer incidence the same for non-Latina black and white women?

Abstract

Abstract Background: Non-Latina Black (black) women, despite their lower BC incidence, are more likely to die from breast cancer compared to their nL White (white) counterparts. While differential access to care, comorbidities and BC aggressiveness are potential contributors to this difference, the potential impact of HMD on this disparity has rarely been explored. For example, nL black women are disproportionately diagnosed with more aggressive tumors that lack estrogen and progesterone receptors (ER and PR) and that are high grade. Therefore, if the association of HMD with breast cancer incidence varied by tumor subtype this could have implications for disparities in breast cancer aggressiveness and subsequent outcomes. Most of the conclusions establishing high mammographic breast density (HMD) as a risk factor for increased breast cancer incidence are derived from studies conducted primarily in whites, which jeopardizes generalizability. We sought to estimate the association between HMD and breast cancer incidence separately for white and black women and by tumor subtype, using data from a single, large healthcare organization. Methods: We used data from screening mammograms performed in women ages 18-100, between 2001 and 2010, probabilistically linked to incident BC cases recorded in the Illinois State Cancer Registry (ISCR) during 2001-2011. Each screening mammogram received a breast density score from the interpreting radiologist using the American College of Radiology Breast Imaging Reporting and Data System (BIRADS 1-4), defined as fatty (BIRADS 1), scattered fibroglandular (BIRADS 2), heterogeneously dense (BIRADS 3) or extremely dense (BIRADS 4). A mean BIRADS breast density score was calculated using available scores from both breasts at each exam. The mean was dichotomized as high (mean BIRADS score > 2.5) versus low (mean BIRADS score ≤2.5) and the dichotomous variable was modeled in logistic regression with generalized estimating equations (to account for multiple screens per woman) to estimate the association between breast density and the probability of a breast cancer diagnosis within 12 months of the screen. Models included age, race, family history, parity and exam year as covariates, and separate models were estimated for black and white women. Odds ratios were interpreted as rate ratios (RR) due to the rarity of the outcome (<1%). Model-based standardization (predictive margins) was use to estimate adjusted rate differences from the logistic regression models. Results: Included in this analysis were 616,466 screens on 201,348 white or black women during 2001-2010, and 4,104 BC in 3,706 women during 2001-2011 (overall rate of 6.7 per 1000 screens, 6.7 for whites and 6.5 for blacks, p=0.21). There were disparities in the distribution of tumor subtypes: breast cancers diagnosed in black women were more likely to be ER negative (23% vs. 15%, p<0.0001) and high grade (38% vs. 30%, p<0.0001) when compared to white women. Upon adjustment for age, family history, parity and exam year, HMD was associated with increased breast cancer incidence overall (RR=1.24, 95% CI: 1.16, 1.32, equivalent to an adjusted Rate Difference of 1.4 per 1000 screens). The association between HMD and breast cancer incidence was similar for whites (RR=1.26, 95% CI: 1.16, 1.36) and blacks (RR=1.19, 95% CI: 1.05, 1.34) and similar for ER positive (RR=1.23, 95% CI: 1.15, 1.31) and ER negative breast cancer (RR=1.29, 95% CI: 1.16, 1.45). HMD was modestly associated with incidence of low grade (RR=1.26, 95% CI: 1.12, 1.41), and moderate grade (RR=1.18, 95% CI: 1.08, 1.29) tumors, but more strongly associated with incidence of high grade tumors (RR=1.45, 95% CI: 1.30, 1.61). These RRs translated into adjusted Rate Differences of 0.4, 0.4 and 0.8 breast cancers per 1000 screens, respectively. Discussion: Our findings reiterate high breast density as a risk factor for breast cancer incidence. In addition, we found that the association was of roughly equal magnitude among white and black women, and of roughly equal magnitude for ER negative and ER positive tumors. The stronger association of HMD with high grade tumors, however, implies that HMD may be a more detrimental risk factor for black women who are more likely to experience higher grade tumors than white women. Citation Format: Katherine Y. Tossas-Milligan, Firas Dabbous, Garth Rauscher. Is the influence of high mammographic breast density on breast cancer incidence the same for non-Latina black and white women? [abstract]. In: Proceedings of the Eighth AACR Conference on The Science of Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; Nov 13-16, 2015; Atlanta, GA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2016;25(3 Suppl):Abstract nr C43.