Genome-wide measures of gene expression have been used to classify breast tumors into clinically relevant subtypes, as well as provide a better means of risk assessment on an individual basis for lymph node-negative (LNN) breast cancer patients. We have applied Affymetrix GeneChips of 22,000 transcripts to analyze total RNA of frozen tumor samples from 286 LNN breast cancer patients in order to identify a gene signature for identification of patients at high risk for distant recurrence.