Variability in outcome of breast carcinoma among different racial groups has been identified between Japanese and Caucasians and between Caucasians and Blacks living in the United States. These differences are not fully explained by population differences of the known prognostic variables nodal involvement and tumor hormonal receptor status. Further elucidation of the differences in outcome should include a careful examination of other prognostic variables. These include tumor size, nuclear and histologic grade, and indicators of cell proliferation (labelling index and flow cytometric measures of S phase and DNA ploidy). More recent studies indicate that growth factor regulation, oncogene amplification (HER-2/neu) and expression, and cathepsin D levels may help to further identify prognostic subgroups. A review of the literature does not provide an answer to the question of whether there are population differences in response to standard treatments. Differences in drug distribution, elimination, and metabolism which could be related to genetic or dietary factors are intriguing hypotheses to explain why differences may exist. Careful, well controlled studies to answer these questions are urgently needed.