Login to Access Video or Poster Abstract: MP99-08
Sources of Funding: Prostate Cancer Foundation, National Cancer Institute, Urology Care Foundation

Introduction

Prostate cancer with ERG rearrangements constitute nearly 50% of cancers, while 10% of prostate cancers have recurrent mutations in SPOP. SPOP mutations and ERG rearrangements are mutually exclusive in prostate cancer, and are both thought to be early events in oncogenesis. Recent largely biochemical work has suggested that the predominant downstream regulator of SPOP mutation is stabilization of the ERG oncoprotein. However the oncologic relevance of this finding in murine and human cancers is unclear.

Methods

We utilized a mouse model in which the SPOP F133V mutation in combination with loss of PTEN results in the accelerated development of high grade prostate intraepithelial neoplasia (HgPIN) and invasive adenocarcinoma. SPOP mutation is a clear driver of carcinogenesis in these models, and the phenotype of these cancers phenocopies human SPOP mutant prostate cancer. We generated prostate organoids from these genetically altered mice and studied these organoids and mice using immunohistochemistry (IHC), immunoblotting, immunofluorescence (IF), and RNA sequencing. We also examined the evidence for SPOP stabilization of ERG being a major driver of prostate cancer in human tumors using multiple institutional and publicly available RNA and DNA sequencing data sources.

Results

We found that murine SPOP driven prostate adenocarcinoma and HgPIN did not express ERG by immunohistochemistry or western blotting. ERG expression was also not present in the presence of SPOP mutation in prostate organoids derived from these mice by IHC, IF, or western blot. Furthermore, mice engineered to coexpress F133V and ERG did not demonstrate evidence of ERG protein upregulation relative to those expressing ERG only. _x000D_ _x000D_ In our institutional human data, we found that none of 8 SPOP mutant cancers expressed ERG. We also found that ERG mRNA is expressed below levels that are generally considered adequate for expression in SPOP mutant cancers. Additionally, ERG rearrangement and SPOP mutation result in vastly distinct gene expression signatures. Organoids derived from SPOP mutant mice do not express ERG mutant gene signatures. Furthermore human SPOP mutant tumors cluster distinctly from ERG rearranged tumors relative to normal prostate tissue, and non-ERG rearranged non-SPOP mutant prostate cancers. Similarly, ERG and SPOP mutant cancers have discrepant methylation signatures, and develop different patterns of genomic rearrangements._x000D_

Conclusions

While SPOP may affect ERG stabilization in certain models, we find no evidence that SPOP stabilization of ERG is driver of mouse or human prostate cancer.

Funding

Prostate Cancer Foundation, National Cancer Institute, Urology Care Foundation