Abstract 3004: Biomarker identification through integrative bioinformatics analysis of serous epithelial ovarian cancer tumor samples

Abstract

Abstract Ovarian cancer is the leading cause of death from gynecological malignancies and the fifth major cancer in women in the world. Once diagnosed, ovarian cancer is usually treated by cytoreductive surgery followed by platinum and taxane-based chemotherapeutic drugs. However, resistance to chemotherapy is a major impediment in management of serous epithelial ovarian cancer (SEOC). We hypothesize that a multifaceted view of the alterations taking place at multiple cellular levels using molecular profiling technologies will offer insight into the mechanisms which play key roles in drug resistant ovarian carcinomas. Also, the application of appropriate bioinformatic and statistical data processing and analysis is of utmost importance in identification of key drug resistance pathways. Current study is performed on 25 high-grade serous epithelial ovarian tumor tissue samples from patients that demonstrated favorable, or unfavorable response to chemotherapy treatment. Four different microarray platforms were used for molecular profiling of the full sample cohort at different molecular levels, namely: Single Nucleotide Polymorphisms (SNP), mRNA expression, miRNA expression and promotor tiling arrays (methylation)., Integrative and systematic analyses using up-to-date statistical approaches, such as empirical Bayes, AUC, SAM, permuted t-test and lassoed PCA, among others, have been employed on these large datasets obtained through the various high-throughput platforms. Preliminary mRNA expression analysis identified an enrichment of upregulated genes involved in cellular growth and proliferation, cellular development as well as differential gene expression changes in the TGFB1, TNF, PI3K, IFNG networks, between the chemotherapy responsive and unresponsive groups. The major molecular and cellular functions associated were cell-to-cell signaling, molecular transport and cellular movement. Differences were also seen in the, CTNNB1, LH and FSH networks as analysed by Ingenuity Pathway Analysis. Additionally, genes involved in activation of NFκB pathway showed differential expression in the two groups. Furthermore, our ongoing development of a streamlined database in which the multiple data types obtained from our statistical analyses are stored, will allow for localized or genome wide querying across the multiple levels of biological data. These approaches and software will potentially elucidate the synergistic roles that the various biological levels play in the deregulation of pathways involved in primary chemoresistance. Our research findings will lead to the determination of putative candidates for diagnostic and prognostic biomarkers that can be targeted for development of treatment regimens for the treatment of SEOC. 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 3004. doi:1538-7445.AM2012-3004