Method of detection and breast cancer survival disparities in Hispanic women.

Abstract

Hispanic women in New Mexico (NM) are more likely than non-Hispanic women to die of breast cancer-related causes. We determined whether survival differences between Hispanic and non-Hispanic women might be attributable to the method of detection, an independent breast cancer prognostic factor in previous studies. White women diagnosed with invasive breast cancer from 1995 through 2004 were identified from NM Surveillance Epidemiology End Results (SEER) files (n = 5,067) and matched to NM Mammography Project records. Method of cancer detection was categorized as "symptomatic" or "screen-detected." The proportion of Hispanic survival disparity accounted for by included variables was assessed using Cox models. In the median follow-up of 87 months, 490 breast cancer deaths occurred. Symptomatic versus screen-detection was classifiable for 3,891 women (76.8%), and was independently related to breast cancer-specific survival [hazard ratio (HR), 1.6; 95% confidence interval (95% CI), 1.3-2.0]. Hispanic women had a 1.5-fold increased risk of breast cancer-related death, relative to non-Hispanic women (95% CI, 1.2-1.8). After adjustment for detection method, the Hispanic HR declined from 1.50 to 1.45 (10%), but after inclusion of other prognostic indicators the Hispanic HR equaled 1.23 (95% CI, 1.01-1.48). Although the Hispanic HR declined 50% after adjustment, the decrease was largely due to adverse tumor prognostic characteristics. Reduction of disparate survival in Hispanic women may rely not only on increased detection of tumors when asymptomatic but on the development of greater understanding of biological factors that predispose to poor prognosis tumors.