A major obstacle in treatment of epithelial ovarian cancer is chemoresistance. The aim of this study was to determine whether distinct gene expression profiles are associated with chemoresistance in epithelial ovarian carcinoma. We performed global gene expression analysis in 13 primary epithelial ovarian cancer tissues including 5 primary chemosensitive tumors and 8 primary chemoresistant tumors using Affymetrix HGU133A microarray. The gene expression patterns of chemosensitive tumors were compared with those of chemoresistant tumors using fold change. Validity of microarray results was examined by semiquantitative RT-PCR. We identified over 320 genes differentially expressed in chemoresistant epithelial ovarian cancer (> or = twofold). Upregulated genes in chemoresistant tumors included cell cycle regulating genes (TOP2A, BCAT1, CDCA8, CCNA2, CENPE), and genes with previously known mechanisms in tumorigenesis (S100A9, APOA1, RNF125, IFI16). Downregulated genes in chemoresistant tumors included genes related to cell adhesion (MUC5B, CITED2), transcription regulating genes (FOXD1, MAD1L1, PAX2), genes involving signal transduction (SOSTDC1, SNX1, SFRP1, FOXA2, PLK2), and stress protein gene (TP53AP1). These data show that gene expression profiling can discriminate primary chemoresistant from primary chemosensitive ovarian cancers. This type of molecular profiling could provide a basis for additional functional studies.