Abstract Background Breast cancer is the most common female cancer in US and is the second leading cause of cancer related death in women. Metastases are the primary cause of cancer morbidity and mortality. Axillary lymph node (LN) status has long been used as a prognostic factor for breast cancer. The molecular mechanisms that control LN metastasis remains poorly understood. To better understand the various genes and regulatory pathways that drive breast cancer LN metastasis, we compared the gene expression profiles between breast tumors that have metastasized to the LNs and those which have not in pre- and postmenopausal women. Material and Methods: Tumor cells were isolated from the primary tumors (ER+) of postmenopausal node positive (PMNP; N=20), postmenopausal node negative (PMNN; N=19), premenopausal node positive (PRNP; N=18) and premenopausal node negative (PRNN; N=16) women using laser capture microdissection. RNA was isolated using the RNAqueous®-micro kit (Ambion, Austin, TX). Total RNA was converted to Biotin-labelled aRNA using two rounds of amplification with MessageAmp II aRNA amplification kit (Applied Biosystems, Foster City, CA). The aRNA concentration was determined by Nanodrop 1000 and the quality was assessed with a Bioanalyzer. The aRNA was fragmented and hybridized to Human Genome U133 Plus 2.0 GeneChip (Affymetrix, Santa Clara, CA). Microarray raw data were analyzed using a variety of R programming packages for probe density processing, background correction, normalization, quality control/quality assessment, and calculation of gene expression value, etc. To identify differentially expressed genes, Wilcoxon rank sum test with FDR (false discovery rate) control was performed for pair-wise comparison between different groups. Functional analyses were performed on the identified statistically significant differentially expressed genes to search for the functional categories and pathways in which they are involved and further understand their potential roles in breast cancer metastatic process. Results: Multivariate data mining (hierarchical clustering analysis and principal component analysis, etc) revealed that in postmenopausal women, the node positive and node negative women are well separated while this was not the case in premenopausal women. Further analysis of the PMNN and PMNP groups to identify differentially expressed genes (with at least a 1.5 fold difference) at FDR =0.1 showed that 232 genes were upregulated and 470 genes were downregulated in PMNP vs PMNN groups. Gene function analysis revealed that genes down regulated in the PMNP group compared to PMNN are related to extracellular matrix, cell adhesion, EGF-like pathway, cytoskeleton etc, while the over-expressed genes are related to cell cycle and cell division, chromosome condensation, etc. Discussion: The ability to differentiate lymph node positive cases from lymph node negative cases in ER+ breast cancer based on transcriptional profiling may have an impact on the clinical management of ER+ breast cancer cases. Having transcriptional profiles that identify ER+ tumors likely to have poor outcomes would suggest more aggressive treatment for such patients. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P3-06-06.