Purpose: To evaluate the relationship between race/ethnicity and breast cancer specific survival and to investigate the mediating effects of tumor characteristics, treatment, anthropomorphic and sociodemographic factors on racial/ethnic disparities in survival.

Methods: Analysis included 19,480 women presenting to National Comprehensive Cancer Network centers with stage I-III breast cancer between January 2000 and December 2007 with National Death Index survival follow-up through December 2009. Multiple Cox proportional hazards regression models were used to compare breast cancer specific mortality by Non-Hispanic Asian-Pacific Islanders (Asian, n=634), Hispanics (Hispanic, n=1,291), Non-Hispanic Blacks (Black, n=1,500) as compared to Non-Hispanic Whites (White, n=16,055) respectively. Additionally models were analyzed overall and also stratified by tumor subtypes. Cox models were analyzed with control variables in steps: age adjusted, plus SES factors, plus tumor characteristics, plus treatment variables. Mediation analyses were performed to estimate the proportion of excess breast cancer mortality mediated through exposures.

Results: Median follow-up time was 6.9 years. Due to non-proportional hazards among Blacks, overall and within certain clinical subgroup models, analyses for total breast cancer, estrogen receptor positive and negative (ER+ and ER-) and basal tumors were performed in two time periods (0-3 years and 3 years to end of follow-up (EOF)). In multivariable fully adjusted models, Blacks had higher risk of breast cancer specific death overall (years 0-3: hazard ratio [HR] 1.48, 95% confidence interval [CI] 1.12-1.94; years 3 to EOF: HR 1.34, 95% CI 1.06-1.69), among ER+ tumors (years 0-3: HR 2.85, 95% CI 1.75-4.62; years 3 to EOF: HR 1.49, 95% CI 1.11-2.00), and for luminal B subtypes (HR 1.76, 95% CI 1.30-2.39) as well as for luminal A subtypes (HR 1.66, 95% CI 1.03-2.67) subtypes. After adjustment for age, SES factors, tumor characteristics and treatment variables there were no significant differences between Blacks and Whites for ER-, basal, or Her2 over expressed tumors. In fully adjusted models Asians were at significantly lower risk of death from breast cancer as compared to Whites (all cancers: HR 0.60, 95% CI 0.40-0.90; ER- tumors: HR 0.51, 95% CI 0.27-0.94; luminal A: HR 0.23, 95% CI 0.06-0.93; HER2 over expressed tumors: HR 0.25, 95% CI 0.07-0.92). There were no significant differences in breast cancer mortality between Hispanics and Whites. The estimated proportion of excess breast cancer mortality among Blacks that was mediated by tumor markers (estrogen, progesterone, and her2neu) and grade was 24.8% (p<0.0001). Other mediators included stage at diagnosis (18.2%, p=0.002), comorbidity score (13.8%, p=0.02), body mass index (BMI) (9.8%, p=0.04), and insurance type (9.5%, p=0.04). Among Asians, BMI (13.9%, p=0.06) was an important mediator.

Conclusions: Blacks are at higher risk of breast cancer death as compared to Whites, particularly in the first three years after diagnosis and predominantly among ER+, luminal A and luminal B tumor subtypes. This excess risk is mediated through differences in tumor characteristics, stage at diagnosis, comorbid conditions, BMI, and insurance type. Asian women have better breast cancer survival than Whites somewhat mediated through their lower BMI at diagnosis.

This abstract is also presented as Poster B64.

Citation Format: Erica T. Warner, Rulla M. Tamimi, Melissa E. Hughes, Rebecca A. Ottesen, Yu-Ning Wong, Stephen B. Edge, Richard L. Theriault, Douglas W. Blayney, Joyce C. Niland, Eric P. Winer, Jane C. Weeks, Ann H. Partridge. Racial/ethnic differences in breast cancer survival and mediating effects of tumor characteristics, sociodemographic, and treatment factors. [abstract]. In: Proceedings of the Fifth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2012 Oct 27-30; San Diego, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2012;21(10 Suppl):Abstract nr PR06.