Abstract 2787: Identifying predictive markers of endocrine response in high-grade serous ovarian cancer using RNA sequencing

Abstract

Abstract Epithelial ovarian cancer is a complex disease and patients vary considerably in response to treatment. A subset of patients responds well to endocrine therapy, however hormone receptor positivity does not predict response and there are no biomarkers in clinical use to help select patients that would benefit from endocrine agents. We aimed to identify markers of endocrine response in high-grade serous ovarian cancer (HGSOC) by analysis of genes that are differentially expressed in tumors from patients that are endocrine responders compared with non-responders, combined with analysis of estrogen (E2) regulated genes that differ between E2 sensitive and E2 insensitive HGSOC cell lines. Methods: Among women recruited to the Australian Ovarian Cancer Study, we identified 10 HGSOC patients treated with endocrine therapy (5 responders and 5 non-responders, based on GCIG CA125 response criteria) and performed RNAseq on cryopreserved tumor tissue. We also utilized paired E2 receptor (ER) positive HGSOC cell lines, derived from the same patient, with differential sensitivity to the growth effect of E2, PEO1 (E2 insensitive) and PEO4 (E2 sensitive). RNA was extracted following treatment with 0.1 nM E2 (or vehicle) for 24 hrs and gene expression changes in response to E2 were determined using RNAseq. Expression of selected ER responsive genes was validated using PCR array. Differential gene expression was determined using EdgeR. Pathway and Gene Ontology enrichment analysis of differentially expressed genes were performed using Metacore. Results: Between endocrine responders and non-responders, 27 genes were significantly differentially expressed (False Discovery Rate (FDR)<0.05, absolute log2 Fold-Change (FC)≥1). In the paired HGSOC cell lines, significantly more genes were regulated by E2 in E2-sensitive PEO4, compared with E2-insensitive PEO1 (896 genes compared to 56 genes, respectively, FDR<0.05, absolute log2FC≥1). In response to E2, up-regulation of NRP1 in PEO1, up-regulation of LTBP1, MYC, PDZK1 and down-regulation of S100A6 and CYP1A1 in PEO4 were validated by PCR array profiling of ER-related genes. Genes regulated by E2 in cell lines and also differentially expressed between endocrine responders and non-responders included anoctamin 1 (ANO1) and NOTCH-Regulated Ankyrin Repeat Protein (NRARP). Both were up-regulated in endocrine responders and also up-regulated by E2 in PEO4. Conclusion: The use of RNAseq in both tumors and cell line models was able to identify plausible sets of genes with prior evidence of association with endocrine sensitivity. In breast cancer, ANO1 has been associated with good prognosis following tamoxifen treatment, while NOTCH signaling, implicated in endocrine resistance, is negatively regulated by NRARP. Further validation is required to test the utility of these genes as predictive biomarkers that may aid in identifying HGSOC patients likely to respond to endocrine treatment. Citation Format: Cristina Mapagu, Sian Fereday, Australian Ovarian Cancer Study Group, David D. Bowtell, Paul R. Harnett, Anna deFazio. Identifying predictive markers of endocrine response in high-grade serous ovarian cancer using RNA sequencing [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 2787. doi:10.1158/1538-7445.AM2017-2787