Research article

Gene expression meta-analysis identifies metastatic pathways and transcription factors in breast cancer

Mads Thomassen¹ , Qihua Tan^1,2 and Torben A Kruse¹

¹Department of Biochemistry, Pharmacology, and Genetics, Odense University Hospital and Human Microarray Centre (HUMAC), University of Southern Denmark, Odense, Denmark

²Institute of Public Health, University of Southern Denmark, Odense, Denmark

author email corresponding author email

BMC Cancer 2008, 8:394doi:10.1186/1471-2407-8-394

Published:	30 December 2008

Abstract

Background

Metastasis is believed to progress in several steps including different pathways but the determination and understanding of these mechanisms is still fragmentary. Microarray analysis of gene expression patterns in breast tumors has been used to predict outcome in recent studies. Besides classification of outcome, these global expression patterns may reflect biological mechanisms involved in metastasis of breast cancer. Our purpose has been to investigate pathways and transcription factors involved in metastasis by use of gene expression data sets.

Methods

We have analyzed 8 publicly available gene expression data sets. A global approach, "gene set enrichment analysis" as well as an approach focusing on a subset of significantly differently regulated genes, GenMAPP, has been applied to rank pathway gene sets according to differential regulation in metastasizing tumors compared to non-metastasizing tumors. Meta-analysis has been used to determine overrepresentation of pathways and transcription factors targets, concordant deregulated in metastasizing breast tumors, in several data sets.

Results

The major findings are up-regulation of cell cycle pathways and a metabolic shift towards glucose metabolism reflected in several pathways in metastasizing tumors. Growth factor pathways seem to play dual roles; EGF and PDGF pathways are decreased, while VEGF and sex-hormone pathways are increased in tumors that metastasize. Furthermore, migration, proteasome, immune system, angiogenesis, DNA repair and several signal transduction pathways are associated to metastasis. Finally several transcription factors e.g. E2F, NFY, and YY1 are identified as being involved in metastasis.

Conclusion

By pathway meta-analysis many biological mechanisms beyond major characteristics such as proliferation are identified. Transcription factor analysis identifies a number of key factors that support central pathways. Several previously proposed treatment targets are identified and several new pathways that may constitute new targets are identified.