Yan, Yuqian

62 Background: We have previously reported that mutations in SPOP promote bromodomain and extra-terminal motif (BET) protein upregulation and BET inhibitor resistance in prostate cancer (Zhang et al., Nature Medicine 23(9):1055-1062, 2017). Other studies indicate that histone acetyltransferases (HATs) such as P300 are also upregulated in SPOP-mutated prostate cancer (Blattner et al., Cancer Cell 31: 436-451, 2017). Methods: We examined the anti-cancer effect of NEO2734 (Epigene Therapeutics Inc), a novel dual inhibitor of both BET and CBP-P300, in prostate cancer using cell line, organoid and patient-derived xenografts (PDX) as working models. We examined the anti-cancer effect of NEO2734 (Epigene Therapeutics Inc), a novel dual inhibitor of both BET and CBP-P300, in prostate cancer using cell line, organoid and patient-derived xenografts (PDX) as working models. Results: We demonstrated that while prostate cancer cell lines expressing F133V, the most frequent SPOP mutant, are very resistant to both BET inhibitor JQ1 and CBP/P300 inhibitor CPI-637, they are very sensitive to simultaneous cotreatment with JQ1 and CPI-637. Similarly, we demonstrated that F133V mutant cells are also very sensitive to NEO2734. We established PDX and organoid models from a patient with prostate cancer expressing a novel SPOP mutation Q165P. Computer simulation analysis reveals that the Q165P mutation causes larger conformational fluctuation, indicating that the structure of Q165P is less stable than that of wild type SPOP. Accordingly, co-immunoprecipitation assay showed that Q165P mutation impaired the dimerization ability of SPOP. We further showed that the Q165P mutant organoid was very sensitive to NEO2734 in vitro and the Q165P mutant PDX was very sensitive to NEO2734 in mice. Conclusions: These PDX-based pre-clinical studies indicate that the inhibition of both BET and CBP-P300, either by co-administration of two agents, or by treatment with the dual inhibitor, NEO2734, is effective in SPOP-mutated prostate cancer. These data provide a strong rationale for the inclusion of patients with SPOP-mutated prostate cancer in clinical studies of NEO2734.

Abstract Purpose: Deletions or mutations in PTEN and TP53 tumor suppressor genes have been linked to lineage plasticity in therapy-resistant prostate cancer. Fusion-driven overexpression of the oncogenic transcription factor ERG is observed in approximately 50% of all prostate cancers, many of which also harbor PTEN and TP53 alterations. However, the role of ERG in lineage plasticity of PTEN/TP53–altered tumors is unclear. Understanding the collective effect of multiple mutations within one tumor is essential to combat plasticity-driven therapy resistance. Experimental Design: We generated a Pten-negative/Trp53-mutated/ERG-overexpressing mouse model of prostate cancer and integrated RNA-sequencing with ERG chromatin immunoprecipitation-sequencing (ChIP-seq) to identify pathways regulated by ERG in the context of Pten/Trp53 alteration. We investigated ERG-dependent sensitivity to the antiandrogen enzalutamide and cyclin-dependent kinase 4 and 6 (CDK4/6) inhibitor palbociclib in human prostate cancer cell lines, xenografts, and allografted mouse tumors. Trends were evaluated in TCGA, SU2C, and Beltran 2016 published patient cohorts and a human tissue microarray. Results: Transgenic ERG expression in mice blocked Pten/Trp53 alteration–induced decrease of AR expression and downstream luminal epithelial genes. ERG directly suppressed expression of cell cycle–related genes, which induced RB hypophosphorylation and repressed E2F1-mediated expression of mesenchymal lineage regulators, thereby restricting adenocarcinoma plasticity and maintaining antiandrogen sensitivity. In ERG-negative tumors, CDK4/6 inhibition delayed tumor growth. Conclusions: Our studies identify a previously undefined function of ERG to restrict lineage plasticity and maintain antiandrogen sensitivity in PTEN/TP53–altered prostate cancer. Our findings suggest ERG fusion as a biomarker to guide treatment of PTEN/TP53-altered, RB1-intact prostate cancer. Clin Cancer Res; 24(18); 4551–65. ©2018 AACR.

Abstract E26 transformation-specific transcription factor ERG is aberrantly overexpressed in approximately 50% of all human prostate cancer due to TMPRSS2-ERG gene rearrangements. However, mice with prostate-specific transgenic expression of prostate cancer–associated ERG alone fail to develop prostate cancer, highlighting that ERG requires other lesions to drive prostate tumorigenesis. Forkhead box (FOXO) transcription factor FOXO1 is a tumor suppressor that is frequently inactivated in human prostate cancer. Here, we demonstrate that FOXO1, but not other FOXO proteins (FOXO3 and FOXO4), binds and inhibits the transcriptional activity of prostate cancer–associated ERG independently of FOXO1 transcriptional activity. Knockdown of endogenous FOXO1 increased invasion of TMPRSS2–ERG fusion–positive VCaP cells, an effect completely abolished by ERG knockdown. Patient specimen analysis demonstrated that FOXO1 and ERG protein expression inversely correlated in a subset of human prostate cancer. Although human ERG transgene expression or homozygous deletion of Foxo1 alone in the mouse prostate failed to promote tumorigenesis, concomitant ERG transgene expression and Foxo1 deletion resulted in upregulation of ERG target genes, increased cell proliferation, and formation of high-grade prostatic intraepithelial neoplasia. Overall, we provide biochemical and genetic evidence that aberrantly activated ERG cooperates with FOXO1 deficiency to promote prostate tumorigenesis and cell invasion. Our findings enhance understanding of prostate cancer etiology and suggest that the FOXO1–ERG signaling axis can be a potential target for treatment of prostate cancer. Cancer Res; 77(23); 6524–37. ©2017 AACR.