Background:Factors contributing to breast cancer survival disparities in underrepresented racial and ethnic groups and low-income populations are poorly understood as few clinical trials and population based studies have included these underserved populations. The Southern Community Cohort Study (SCCS), a prospective cohort of underserved, low-income adults with high representation of Black participants, provides a unique opportunity to evaluate such disparities in cancer outcomes. A previous study utilizing SCCS data found no evidence of increased breast cancer risk among women with diabetes in this population. The purpose of this study was to evaluate the association of diabetes with overall survival in women with breast cancer in the SCCS.

Methods:The SCCS enrolled approximately 86,000 participants aged 40-79 from 12 southeastern states between 2002-2009, 86% of whom were enrolled at Community Health Centers. This analysis includes women diagnosed with incident localized breast cancers identified through annual cohort linkage with 12 state cancer registries. Demographic data including participant age at breast cancer diagnosis, self-reported history of diabetes (patient answered yes to “has a doctor ever told you that you have diabetes”), body mass index (BMI), race, household income, and insurance coverage were obtained from baseline surveys, cancer type and stage data from state cancer registries, and survival data from death registries. Survival time was defined as the number of months between initial breast cancer diagnosis and death from any cause. Descriptive characteristics including mean (standard deviation) or number (%) were used to summarize demographics. We used Pearson Chi-squared analysis to examine the association between diabetes and overall survival. Multivariable Cox proportional hazards regression was used to evaluate overall survival and diabetes, adjusting for covariates including age (continuous), race, BMI (categorical by WHO classifications), household income (binary – annual income <$25,000, annual income >=$25,000), insurance coverage, cancer subtype, and cancer stage).

Results:We identified a total of 1,347 women diagnosed with breast cancer. Of these, 1,016 were diagnosed with localized disease (stage 1-3) and comprised our analytic sample. Difference in denominators reflects missing data. The women were predominantly Black (667/1,016, 65.6%), low income (719/1,016 annual income less than $25,000, 70.8%), and insured (Private insurance 220/763, 28.8%; Medicare 331/763, 43.4%; Medicaid 178/763, 23.3%). Average age at diagnosis was 60.7 years (SD 9.1, IQR 41-88). Approximately one quarter of the patients (258/994, 26.0%) self-reported diabetes and 59.6% (605/1,016) were obese (BMI >=30). The breast cancer immunohistochemistry subtypes in this cohort of women included HR+HER2- (392/564, 69.5%), HR+,HER2+ (55/564, 9.8%), HR-,HER2+ (31/564, 5.5%), and HR-HER2- (86/564, 15.3%). Women with diabetes had lower overall survival (174/258, 67.4%) than women without diabetes (587/746, 79.8%) (p<0.0001). In the adjusted multivariate Cox regression model, diabetes significantly decreased overall survival in women with breast cancer, hazard ratio 1.87, 95% Confidence Interval [CI] = 1.12-3.09.

Conclusion:

In a low-income, predominantly Black population with incident localized breast cancer, decreased overall survival was observed among women diabetes compared to those without diabetes. Future studies should explore additional biological, societal, and socio-economic factors affecting survival among women with breast cancer in medically underserved minority populations.

Citation Format: Lucy B. Spalluto, Sonya Reid, Diane Haddad, Tuya Pal, Ingrid A. Mayer, Xiao-ou Shu, Maureen Sanderson, Wei Zheng, William J. Blot, Loren Lipworth. Diabetes decreases overall survival in women with breast cancer in the southern community cohort study [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PD11-05.