Estrogen receptor positivity improves breast cancer survival prognosis for all underserved ethnic groups except African Americans

Abstract

9591 Background: California Cancer Registry data from 1997–2001 indicate age-adjusted breast cancer mortality rates (per 100,000) for San Francisco County are highest for African Americans (AA; 43) compared to non-Hispanic Whites (W; 32), Hispanics (H; 16) and Asians/Pacific Islanders (A; 12). However, these data cut across heterogeneous socioeconomic strata and do not account for unequal access to care and diverse treatment practices. The San Francisco General Hospital (SFGH) Breast Clinic staffed by the same breast surgeon and medical oncologist for the past 12 years provides diagnostic and treatment services to the city’s underserved (low SES) and multiethnic population, enabling a breast cancer outcome comparison among low SES patients of different ethnicities uniformly offered treatment. Methods: Chart and registry review of 1014 invasive breast cancer cases diagnosed between 1992–2004, grouped by patient ethnicity (H, AA, W, A) and typed by tumor hormone receptor status (ER, PR). Results: Among ethnic groups there were no significant differences in median age at diagnosis(53y), age distribution(35–37% >60y), or tumor stage (20–30% Stage III/IV). Number lost to follow up equivalent in all groups. (see table) Conclusions: SFGH breast cancer survival while worse than SF County as a whole shows the same trend with respect to ethnic differences. AA women are least likely to be alive and disease free while A women are most likely. Analyses based on ER subtype suggest that the survival advantage of ER positivity is smaller for AA women than for other ethnic subgroups. This suggests that overall survival differences among ethnic groups cannot be fully explained by differences in SES or delivery of care. The comparative lack of benefit among ER+ AA cases cannot be accounted for by an excess proportion of ER+/PR- cases suggesting that ER+ breast cancers arising in AA possess other unique high-risk biological features. No significant financial relationships to disclose.