Objective: To screen potential biomarkers and therapeutic agents for tamoxifen-resistant breast cancer based on gene expression profile and connectivity map. Methods: GSE67916 and GSE26459 which contain gene expression profiles of tamoxifen-resistant breast cancer cell were downloaded from the Gene Expression Omnibus (GEO) database. Differentially expressed genes (DEGs) were identified using the Bioconductor packages in the R language. Metascape was used to perform functional annotation for DEGs. Cytoscape was used to construct the protein-protein interaction network map of DEGs-coding proteins and then to identify the hub gene. GSE9893 was downloaded to validate the expression difference of hub gene between tamoxifen-sensitive patients and resistant patients and then to investigate the correlation between hub gene expression and relapse-free survival. Finally, potential therapeutic agents were screened by mapping the DEGs to connectivity map database and validated by cell proliferation assay. Results: A total of 462 genes were filtered as DEGs and they were mainly involved in androgen response, glycolysis, cholesterol homeostasis and so on. ATP-citrate lyase (ACLY) was identified as the hub gene related to tamoxifen resistance. The expression level of ACLY in tamoxifen-resistant patients was significantly higher than that in tamoxifen-sensitive patients (P<0.05). The relapse-free survival in patients with low expression levels of ACLY was significantly better than that in patients with high expression levels (P<0.05). Ten candidate therapeutic agents were screened for tamoxifen resistance, such as pimozide, LY-294002 and so on. LY-294002 was confirmed to inhibit the proliferation of tamoxifen-resistant breast cancer cells.
Conclusion: ACLY may be used as a biomarker for tamoxifen resistant breast cancer. Connectivity map can be used to rapidly discover potential therapeutic agents and lay the foundation for further studies.