Objective  To screen and mine the genes related to multidrug resistance (MDR) in ovarian cancer(OC) and their biological information.

Methods  Based on four different microarray expression profiles(GSE41499, GSE33482, GSE15372 and GSE28739) between resistant samples and sensitive samples relatedto OC, we performed a comprehensive bioinformatics analysis through gene expression analysis, geneticpathway enrichment analysis and text mining to predict the pathways and their genes related to MDR inOC.

Results  Eleven significantly upregulated pathways were found frequently among four OC microarraydatasets, including MAPK signaling pathway, ubiquitin-mediated proteolysis, axon guidance, focal adhesion,neurotrophin signaling pathway, pathways in cancer, renal cell carcinoma, citrate cycle, terpenoid backbonebiosynthesis, mismatch repair and Huntington’s disease(P<0.05); and seven significantly downregulatedpathways were found frequently, including glycerolipid metabolism, pentose phosphate pathway, fructoseand mannose metabolism, glutathione metabolism, proteasome, p53 signaling pathway and lysosome(P<0.05). By further text mining methods, we found five significantly upregulated genes, including ACO1,BDNF, CXCR4, HMGCR and NRP1(P<0.05), as well as three significantly downregulated genes, includingCDKN2C, FAS and SKP2(P<0.05), might be associated with MDR in OC.

Conclusion  OC MDR mightbe involved in various pathways and genes. ACO1, BDNF, CXCR4, HMGCR, NRP1, CDKN2C, FAS andSKP2 might play crucial roles in those pathways. Follow-up study would validate the roles of those genes inthe experiments and clinical practice.