Abstract A31: High-grade serous ovarian cancer subtyping identifies pathways for targeted therapy.

Abstract

Abstract Introduction: We recently identified three prognostic groups for high-grade serous ovarian cancers (HGS-OvCa) patients treated with platinum and taxane using transcriptomics and clinical data (Chen, Huhtinen, et al. Cancer Res. 2015). In this contribution we first use pathway analysis methods to identify transcript markers and pathways that drive resistance to platinum and experimentally inhibit of the key pathways to restore sensitivity to platinum. Two of the prognostic groups (Poor I and Poor II) were associated with poor survival but have distinct expression profiles. Poor II express high stromal response genes and confirms the C1 subtype characterized by Tothill et al. (2008), whereas Poor I is previously unidentified and characterized by genome-wide hypermethylation. Our objectives here are to: 1) identify key pathways for Poor I and Poor II based on transcriptomics data (data unpublished), 2) inhibit the herein identified pathways to suggest combinatorial therapeutics options for platinum resistance HGS-OvCa patients (data unpublished), 3) test the impact of hypermethylation drug Decitabine to platinum resistance in vitro (data partially unpublished). Methods: Expression levels were qRT-PCR validated using well described prospective patient cohort. The role of hypermethylation and specific pathways on cell viability were studied in vitro using commercial (CAOV4, NIHOVCAR3, OVCAR8, TYKnu) and spheroidal primary HGS-OvCa cell lines. Results: Poor I subtype presents with significantly higher methylation of CpG loci (p < 0.001) as compared with Good or Poor II subtypes. We observed significant up-regulation of three isoforms of DNA methyltransferase 3B (DNMT3B) in Poor I samples. The role of hypermethylation was further validated in vitro using patient derived cell lines with platinum resistance. Inhibition of methylation by Decitabine exhibited strong effect on cell viability specifically on Poor I characteristic HGS-OvCa cell lines. The data suggests that clinical trials testing the effect of Decitabine should be focused to patients with Poor I disease. To evaluate signaling pathways that are associated with platinum-taxane resistance we evaluated common pathways enriched in Poor I and Poor II using the transcriptomics data available at The Cancer Genome Atlas repository. The most prominent pathway is MAPK signaling pathway of which four genes (FAS, MAPK1, AKT and NR4A1) emerged from the analysis. Pathway analysis further suggested ERK5 and AKT as upstream regulators for NR4A1 in platinum-taxane resistance in HGS-OvCa. The impact of ERK5 and AKT pathways on platinum resistance was evaluated in vitro using commercial and primary HGS-OvCa cell lines with wild-type BRCA1/2. The combination of AKT-inhibitor API-2 and Cisplatin resulted in a marked decrease in cell viability over Cisplatin alone whereas AKT-inhibitor alone had only little effect. In contrast, ERK5 inhibition with XMD8-92 combined with Cisplatin did not have an effect on HGS-OvCa cell viability. However, XMD8-92 enhanced the effect of API-2 in subset of the cell lines. Thus, we tested whether combination of the three drugs results in better response. In general, the addition of XMD8-92 did not improve the effect of API-2 and Cisplatin. However, in two highly platinum resistant cell lines, CAOV-4 and M022i primary cell line, the combination of API-2, XMD8-92 and Cisplatin produced the highest reduction in cell viability. In conclusion, we have identified pathways that are likely driving drug resistance in HGS-OvCa and tested the impact of platinum combined with targeted AKT and ERK5 inhibition in vitro. We further show that the Poor I type HGS-OvCa cells are sensitive to Decitabine. Thus, Decitabine may be an effective (combinatorial) treatment option for patients with the Poor I subtype disease. Combination of AKT and ERK5 inhibitors with platinum can also be effective for a subset of platinum resistant ovarian cancer patients. Citation Format: Kaisa Huhtinen, Ping Chen, Katja Kaipio, Piia Mikkonen, Viljami Aittomäki, Rony Lindell, Johanna Hynninen, Annika Auranen, Seija Grenman, Rainer Lehtonen, Sampsa Hautaniemi, Olli Carpén. High-grade serous ovarian cancer subtyping identifies pathways for targeted therapy. [abstract]. In: Proceedings of the AACR Special Conference on Advances in Ovarian Cancer Research: Exploiting Vulnerabilities; Oct 17-20, 2015; Orlando, FL. Philadelphia (PA): AACR; Clin Cancer Res 2016;22(2 Suppl):Abstract nr A31.