Abstract 5057: Gene expression profile of chemoresponse of microdissected papillary serous tumors of the ovary identifies POLH and REV3L as potential therapeutic targets

Abstract

Abstract Treatment of advanced stage ovarian carcinoma often involves cytoreductive surgery followed by adjuvant chemotherapy with paclitaxel and either cisplatin or carboplatin. Most tumors have either inherent or develop acquired resistance to chemotherapy, resulting in patient demise from their cancer. The mechanism of drug nonresponse (refractory/resistance) remains unknown and currently there is no clinically useful molecular predictor of response. Here, we determined the gene expression profile of 52 microdissected advanced stage papillary serous ovarian cancer and identified gene expression signatures predictive for nonresponse to chemotherapy. To generate a predictive gene signature, class prediction algorithms were applied to genes differentially expressed between resistant and sensitive tumors (p<0.001) using leave-one-out cross-validation. We developed a predictive 96-gene signature for tumor refractory and 29-gene signature for tumor resistance to chemotherapy. Real-time quantitative RT-PCR validated the results of the cDNA microarray analysis. To further validate our array results and demonstrate therapeutic value of these targets, we pursued several novel genes to determine whether over expression will render ovarian cancer cell lines resistance to chemotherapeutic drugs. Overexpression of EDIL3, GNA13 and POLH genes imparted cisplatin resistance to the ovarian cancer cell lines. As DNA Polymerases contribute to cell survival and the emergence of resistant cells by allowing cells to replicate beyond cisplatin adducts to avoid replication arrest and cell death, we decided to evaluate the potential therapeutic value of the genes POLH (DNA polymerase eta) and REV3L (DNA polymerase zeta) using short interfering RNA (siRNA) approach. Reduced gene expression via gene specific siRNA molecules renders ovarian cancer cell lines (A2780-CP20 and UCI 107) sensitive to cisplatin in vitro. Further, we validated these results in an orthotopic mouse model of ovarian cancer using liposomal (DOPC) siRNA approach as these nanoscale liposomes can penetrate deeply into the tumor. A combination of siREV3L/siPOLH-DOPC with cisplatin resulted in a significant reduction in mean tumor weight compared to mice treated with nonspecific control siRNA (77% reduction; Mann Whitney test, P=0.0140), compared to mice treated by nonspecific control siRNA with cisplatin (72% reduction; Mann Whitney test, P=0.0108), and compared to those mice treated with combination siREV3L/siPOLH-DOPC (68% reduction; Mann Whitney test, P=0.0070). In conclusion, here we report gene profile signatures of the chemoresponse of ovarian cancer. POLH and REV3L identified as potential therapeutic targets which can serve as targets for specific molecular therapeutic molecules that can increase the sensitivity of ovarian cancer to standard chemotherapy. 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 5057. doi:1538-7445.AM2012-5057