The purpose of this study was to pool information in epithelial ovarian cancer by combining studies using Affymetrix expression microarray datasets made at different laboratories to identify novel biomarkers. Epithelial microarray expression information across laboratories was screened and combined after preprocessing raw microarray data, then ANOVA and unpaired T test statistical analysis was performed for identifying differentially expressed genes (DEGs), followed by clustering and pathway analysis for these DEGs. In this work, we performed a combination analysis on microarrays from three different laboratories using gene expression data on ovarian cancer and obtained a list of differential expression profiles identified as potential candidate in aggressiveness of ovarian cancer. The clustering and pathway analysis explored the different molecular basis of different ovarian cancer stages and potential important regulatory pathways in ovarian cancer development. Our results showed that combination of microarray data from different laboratories in the same platforms may overcome biases derived from probe design and technical features, thereby accelerating the identification of trustworthy DEGs, and demonstrating the advantage of integrative analysis in gene expression studies on epithelial ovarian cancer research.