Hazard of breast cancer-specific mortality among women with estrogen receptor-positive breast cancer after five years from diagnosis: implication for extended endocrine therapy.

Abstract

More than half of the patients with estrogen receptor (ER)-positive breast cancers will relapse and die from breast cancer at 5-10 yr after diagnosis despite 5-yr endocrine therapy. Subpopulations of ER-positive patients at high risk of breast cancer-specific mortality (BCSM) at 5-10 yr are undetermined. Using the Surveillance, Epidemiology, and End-Results program (1990-2003), we analyzed the relative hazard ratio (HR) and absolute HR of BCSM and the cumulative 10-yr breast cancer-specific survival (BCSS) in 111,993 breast cancer patients, stratified by ER, age, and lymph node (LN), and adjusted for other prognostic factors. At 5-10 yr after diagnosis, ER-positive patients had increased risk of BCSM [HR, 0.71; 95% confidence interval (CI), 0.66-0.76; ER-positive as reference] compared with ER-negative patients. Specifically, younger ER-positive patients (<40 yr) had a constant plateau of annual hazard rate, a higher hazard of BCSM (HR, 0.43; 95% CI, 0.35-0.52; ER-positive as reference), and poor 10-yr BCSS, despite LN status. Among ER-positive patients aged 40-60 yr having no obvious plateau of hazard rate, only those with LN-positive disease had a significantly increased hazard of BCSM and poor 10-yr BCSS. Elderly ER-positive patients aged 60-74 yr had a hazard of BCSM, similar to that of ER-negative patients, and those with LN-positive disease had poor 10-yr BCSS. Our findings help to define the ER-positive subpopulations at higher risk of BCSM at 5-10 yr after diagnosis and are useful in choosing candidates for clinical trials of extended endocrine therapy after 5-yr treatment and in guiding individualized treatment.