Abstract 2176: Deletion of 3p14 is linked to ERG-fusion positive prostate cancers and harbors the FOXP1 tumor suppressor gene

Abstract

Genetic rearrangements involving 3p14 are frequent alterations in prostate cancer, affecting approximately 15-20% of tumors. Both the clinical significance and the tumor suppressor at 3p14 locus are currently unknown. We performed SNP array analysis of 72 primary prostate cancers and show a commonly deleted region at 3p14 in 14 (19%) of the tumors. Our data indicate the existence of a small, recurrent deletion in a significant fraction of prostate cancers harboring 8 genes only. Ten of 14 tumors with 3p14 deletions detected in our study included the FOXP1 locus. To estimate the prevalence and clinical significance of FOXP1 deletions, we analyzed a tissue microarray containing 4,699 prostate cancers with histopathological and clinical follow-up data. Heterozygous FOXP1 deletions were found in 15% of tumors, and were strongly associated with ERG-fusion positive cancers (p<0.0001). No homozygous deletions were detected. FOXP1 deletions were unrelated to tumor stage or Gleason grade. FOXP1 deletions were linked to early PSA recurrence in ERG-positive cancers (p=0.0576) but not in ERG negative tumors (p=0.5425). To better understand the association between ERG fusion and FOXP1 expression, we performed FOXP1 immunohistochemistry on our TMA. FOXP1 expression was found in virtually all (95%) prostate cancers, and was particularly strong in ERG-fusion positive tumors. This finding argues for an important role of FOXP1 in prostate cancer in general and for ERG fusion positive tumors in particular. Since FOXP1 is a transcriptional repressor, we hypothesized that upregulation might be a consequence of ERG activation in order to control transcription of ERG target genes. However, neither expression of ERG in LNCaP cells, or depletion of ERG in VCaP cells had an effect on FOXP1 mRNA levels, strongly arguing against a direct link between ERG and FOXP1 at least in the tested cell lines. In addition neither the transcription nor the steady state levels of FOXP1 protein were modulated by activation of AR signaling. Finally, colony formation assays in both benign and malignant prostate cell lines revealed that FOXP1 may act as a tumor suppressor. In summary, our study demonstrates that FOXP1 functions as one of the tumor suppressor genes in prostate cancer that is targeted by genomic deletions at 3p14. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 2176. doi:1538-7445.AM2012-2176