Chemotherapy treatment and survival in older women with estrogen receptor-negative metastatic breast cancer: a population-based analysis.

OBJECTIVES: To investigate the survival benefit associated with chemotherapy receipt in older women with estrogen receptor-negative (ER-) Stage IV breast cancer. DESIGN: Observational, retrospective cohort study using Cox proportional hazards regression to determine effect of chemotherapy on hazard of all-cause mortality. The two samples were an overall sample (n=1,519) and a propensity score-matched sample (n=580) to control for selection to treatment receipt. Hazard ratios (HRs) and 95% confidence intervals (CIs) were obtained for regression models. SETTING: U.S. women within the National Cancer Institute Surveillance, Epidemiology and End Results cancer registries (SEER) linked to Medicare enrollment and claims database. PARTICIPANTS: Female Medicare beneficiaries aged 66 and older with Stage IV ER- breast cancer diagnosed between 1999 and 2005. MEASUREMENTS: Outcome measure was all-cause death during the follow-up period. Survival was measured as time from breast cancer diagnosis until death or last follow-up date. Information on receipt of chemotherapy, defined as chemotherapy received within 6 months after diagnosis, was obtained from linked Medicare claims. RESULTS: One thousand five hundred nineteen ER- women diagnosed with metastatic breast cancer were identified; 494 (33%) received chemotherapy. Chemotherapy was associated with a statistically significant survival benefit (HR=0.61, 95% CI=0.54-0.70). Age did not modify the survival benefit of chemotherapy. CONCLUSION: Chemotherapy received within 6 months after diagnosis was associated with a 39% lower hazard of death within the time period for the study. These findings reflect chemotherapy use outside of the clinical trial setting and have important clinical and policy implications for the study of treatments in older women with advanced ER- breast cancer. © 2011, The American Geriatrics Society.