Although alterations in several specific genes have been implicated in breast cancer progression, greater understanding of the molecular basis of the disease may result from the evaluation of global gene expression patterns using microarray technology. To identify patterns of gene expression of prognostic importance in axillary node–negative breast cancer, we are studying a large cohort of patients with this type of cancer. Before using the limited amount of RNA from these specimens, we performed pilot studies designed to evaluate the feasibility of applying the technology on a larger scale. We studied gene expression in four different breast cancer cell lines (T47D, MDA231, SKBR3 and BT474) using complementary DNA microarrays containing 1,700 and 19,000 sequence-verified human cDNAs produced by the microarray facility at the Ontario Cancer Institute, Toronto (http://www.oci.utoronto.ca/services/microarray). Each hybridization compared Cy5-labeled complementary DNA from one of the cell lines with Cy3-labeled cDNA from a reference sample (MCF12A, a normal breast cell line). We also performed reciprocal labeling with subsequent hybridization to demonstrate the consistency and reproducibilty of the technology. In addition, different amounts of RNA (50 μg, 25 μg and 10 μg) from the same cell lines were labeled to determine the sensitivity of the system. We have been able to devise a system that can now be applied to primary breast tumors. After expression analysis, we will use biostatistical modeling to detect clusters of genes that are coordinately expressed, repressed or both. These clusters are likely to represent common pathways of genes involved in breast carcinogenesis.