Interrogation of ERG Gene Rearrangements in Prostate Cancer Identifies Novel Signatures Relative to Disease Progression and with Prognostic Implications
Mohammed Alshalalfa, Liang Hong Teng, Lars F Petersen, Ashraf Bakkar, Amal Al-Mami, Shuhong Liu, Chad Brenner, Michael Dolph, Felix Y Feng, Reda Alhajj, Tarek A Bismar. University of Calgary, Calgary, AB, Canada; Calgary Laboratory Services and Unversity of Calgary, Calgary, AB, Canada; University of Michigan, Ann Arbor, MI
Background: ERG gene rearrangements have been proposed to be reflective of specific molecular subtype of prostate cancer (PCA) with its prognostic implication. Herein, we investigated gene expression differences between those two classes of tumors to identify potential genetic targets and pathways.
Design: 6144 informative genes belonging to 46 castration resistant PCA samples were interrogated using the DASL platform. We used bioinformatic approach to identify significant dyregulated genes based on their ERG status (assessed by FISH).
Results: The employed method identified a group of 16 differentially expressed genes between ERG rearranged (19 samples) and ERG non-rearranged tumors (27 samples). We then were able to narrow down those genes to a 10 gene model which was the most accurate to predict prostate tissue sample classification relative to ERG status (77%). Several tumor suppressor genes as well as genes associated with cell growth, apoptosis and cancer metastasis were among the most significant deregulated genes between the two groups. Several of the genes described have never been characterized or have been poorly described in association with prostate cancer. Furthermore, we validated this model using Q-PCR and protein expression using progression TMA. The multi-gene model was also able to confirm significant prognostic value in prostate cancer beyond those of ERG alone when tested on public datasets.
Conclusions: ERG rearrangements tumors represent distinct subclass of PCA that is associated with downregulation of genes related to cell adhesion, and cell motility. This group as well shows downregulation of several tumor suppressor genes. As this represent a distinct subclass of tumors proposed to be associated with more aggressive behavior, analyzing and comparing the genes identified in this study with other genes associated with PCA could lead to better understanding to the molecular mechanism behind the aggressive behavior of ERG rearranged tumors. Furthermore, as the gene panel list identified showed significant association related to patients' prognosis in other tumor types, characterization of those genes could proof significant players in identifying pathways related to cancer progression in general and will help us to identify potential new targets for cancer therapy across several tumor types.
Category: Genitourinary (including renal tumors)
Monday, March 19, 2012 1:00 PM
Poster Session II # 160, Monday Afternoon