27P - Integrated system-level analyses of androgen receptor variant networks to identify novel prostate cancer-relevant genes that serve as prognostic bi...

Date 11 September 2017
Event ESMO 2017 Congress
Session Poster display session
Topics Basic Science
Genitourinary Cancers
Presenter Fiorella Magani
Citation Annals of Oncology (2017) 28 (suppl_5): v1-v21. 10.1093/annonc/mdx361
Authors F. Magani, E. Bray, S. Peacock, N. Zhao, K. Burnstein
  • Molecular And Cellular Pharmacology, University of Miami Miller School of Medicine, 33136 - Miami/US

Abstract

Background

Castration resistant prostate cancer (CRPC) is a rapidly progressing disease state for which there is no cure. The constitutively active androgen receptor (AR) splice variant AR-V7 represents a well-established mechanism of therapeutic resistance and disease progression. This variant lacks the AR C-terminal ligand binding domain and, as such, is not inhibited by androgen deprivation therapy. Designing high-affinity drugs to target the amino terminus of AR and AR-V7 is a major challenge due to the intrinsic disorganized structure of this region. Thus, there is an imperative need to identify novel AR-V7 hub genes in PC that may serve as novel therapeutic targets.

Methods

We performed a highly robust gene expression meta-analysis on PC patient samples. We defined gene modules correlated with PC progression using a Weighted Gene-Co-expression Network Analysis (WGCNA), a powerful systems biology approach. Further, we identified AR-V7 downstream target genes using gene expression profiling and mapped the AR-V7 functional interactome for the first time using a novel high-throughput synthetic genetic array screen in yeast. Finally, we combined the results from our three independent system-level analyses with experimental data to identify hub genes that were upregulated in PC patients, upregulated by AR-V7, and that also functionally interacted with AR-V7.

Results

The identified genes not only included select genes previously linked to PC, such as members of the topoisomerase and cyclin families, but also novel genes that had not been previously linked to PC progression. The identified gene-signature expression correlated with patients’ Gleason score and had a prognostic value that predicted disease free-survival at the time of patient biopsy in large independent cohorts.

Conclusions

In sum, we show here an unbiased integrated system-level analysis of AR-V7 networks, where we combined bioinformatic analysis of patient samples and cell-based approaches to identify new candidate genes in CRPC that may serve as novel prognostic markers and future targeted therapies.

Clinical trial identification

Legal entity responsible for the study

University of Miami Miller School of Medicine

Funding

Sylvester Comprehensive Cancer Center

Disclosure

All authors have declared no conflicts of interest.