Research article

Gene expression profiles of prostate cancer reveal involvement of multiple molecular pathways in the metastatic process

Uma R Chandran¹ , Changqing Ma² , Rajiv Dhir² , Michelle Bisceglia² , Maureen Lyons-Weiler² , Wenjing Liang² , George Michalopoulos² , Michael Becich^1,2 and Federico A Monzon²

¹  Departmental of Biomedical Informatics, University of Pittsburgh, Parkvale Building M-183, 200 Meyran Ave, Pittsburgh, PA 15260, USA

²  Department of Pathology, University of Pittsburgh, S-417 BST, 200 Lothrop Street, Pittsburgh, PA 15261, USA

author email corresponding author email

BMC Cancer 2007, 7:64doi:10.1186/1471-2407-7-64

Published:	12 April 2007

Abstract

Background

Prostate cancer is characterized by heterogeneity in the clinical course that often does not correlate with morphologic features of the tumor. Metastasis reflects the most adverse outcome of prostate cancer, and to date there are no reliable morphologic features or serum biomarkers that can reliably predict which patients are at higher risk of developing metastatic disease. Understanding the differences in the biology of metastatic and organ confined primary tumors is essential for developing new prognostic markers and therapeutic targets.

Methods

Using Affymetrix oligonucleotide arrays, we analyzed gene expression profiles of 24 androgen-ablation resistant metastatic samples obtained from 4 patients and a previously published dataset of 64 primary prostate tumor samples. Differential gene expression was analyzed after removing potentially uninformative stromal genes, addressing the differences in cellular content between primary and metastatic tumors.

Results

The metastatic samples are highly heterogenous in expression; however, differential expression analysis shows that 415 genes are upregulated and 364 genes are downregulated at least 2 fold in every patient with metastasis. The expression profile of metastatic samples reveals changes in expression of a unique set of genes representing both the androgen ablation related pathways and other metastasis related gene networks such as cell adhesion, bone remodelling and cell cycle. The differentially expressed genes include metabolic enzymes, transcription factors such as Forkhead Box M1 (FoxM1) and cell adhesion molecules such as Osteopontin (SPP1).

Conclusion

We hypothesize that these genes have a role in the biology of metastatic disease and that they represent potential therapeutic targets for prostate cancer.