Background: Type II diabetes (T2D) has been found to modestly (10-20%) increase risk for breast cancer (BC) in mostly white populations. Some studies indicate that this association varies by BC receptor subtype, suggesting that T2D increases risk specifically for the more aggressive estrogen-receptor (ER)-negative and triple negative subtypes. While black women are more likely than white women to develop both T2D and ER-negative BC, no studies to date have explicitly focused on subtype-specific risk by race. Further, no studies have examined this association in women from medically underserved areas, where the prevalence and severity of T2D is greater than in areas with adequate resources.

Methods: Participants were women from the Southern Community Cohort Study (2002-2009), which prospectively recruited individuals age 40-79 years, primarily from community health centers in medically underserved areas across the Southeastern US. We identified 39,687 women who were cancer-free at baseline and self-reported black or white race. Baseline T2D status was defined as a self-reported diagnosis of medication-treated T2D at age > 30 years. BC incidence and receptor status were ascertained by state cancer registry linkages and pathology reports. Cox proportional hazards regression was used to estimate the hazard ratio (HR) and 95% confidence interval (CI) for BC (overall and by subtype) associated with T2D. Models were first adjusted for, and then stratified by race. Exploratory analyses modeled self-reported diabetes medication adherence (taking as directed all the time or most of the time vs. sometimes, rarely or never) as an exposure.

Results: Among 39,687 women (29.9% white and 70.1% black), 7501 (18.9%) reported having T2D. Over a median of 10.3 years of follow-up, 929 incident cases of invasive BC occurred; 224 (24.1%) were ER-negative, and among those with complete receptor data (n=808), 138 (17.1%) were triple-negative. Women with T2D were significantly more likely to develop BC (HR: 1.19, 95% CI: 1.02-1.40) compared to non-diabetics in models adjusted for age, race, BMI, mammography screening, menopausal status, alcohol use, and family history of BC. In subtype-specific analyses, the risk associated with T2D was greatest for the ER-negative (HR: 1.45, 95% CI: 1.01-2.08) and triple-negative (HR: 1.53, 95% CI: 1.01-2.31) subtypes. Race-stratified models showed similar elevations in risk among black (HR: 1.16, 95% CI: 0.97-1.39) and white women (HR: 1.13, 95% CI: 0.83-1.53) for BC overall, and for triple-negative BC, with a HR of 1.61 (1.02-2.53) in blacks and 1.94 (0.74-5.07) in whites. Across both races and by subtype, women reporting lack of adherence to diabetes medications had far greater risk for BC compared to non-diabetics, with an overall HR of 2.15 (1.27-3.66).

Conclusion: In this study, T2D was associated with a 19% increased risk of BC and 53% increased risk for the triple-negative subtype, with similar results for black and white women. The risk was notably greater in women who reported lack of T2D treatment adherence. Given these findings and the prevalence of T2D in medically underserved areas, interventions aimed at T2D prevention and adherence to medications in all races could help reduce the BC burden in this population.

Citation Format: Gross AL, Blot WJ, Visvanathan K. Type II diabetes and subtype-specific breast cancer risk in medically underserved black and white women [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P1-10-02.