Abstract 3864: Aneuploidy predicts aggressiveness and poor prognosis in endometrial cancer, and is reflected in a 9-gene signature

Abstract

Abstract Introduction: Ploidy is among the most frequently studied biomarkers in endometrial cancer (EC), but lacks validation in larger studies. Mechanisms leading to tumor aneuploidy are only partly understood, and few studies have examined the molecular distinctions between diploid and aneuploid tumors in EC. Loss of the cohesin subunit STAG2 has been proposed as a driver of aneuploidy in several cancer types, but has not yet been explored for EC. Aims: In this study, we wanted to investigate ploidy status in primary EC in relation to standard clinicopathologic variables including FIGO stage, grade, histologic type and age, as well as outcome. Further, we wanted to evaluate genes differentially expressed between aneuploid and diploid primary tumor samples, and the proposed aneuploidy marker STAG2 in EC. Materials and methods: DNA ploidy was determined by flow cytometry in fresh tumor tissue for 794 patients, enrolled from 1980 - 2013. For expression analysis, RNA was extracted from fresh frozen primary tumor tissue from 144 samples, and measured with Agilent DNA microarrays (cat. no G4 112F). Significance Analysis of Microarrays (SAM) was run to identify differentially expressed genes between diploid and aneuploid cases. In order to identify the signature genes reflecting the differences between the two groups with highest accuracy, support vector machine (SVM) with 10-fold cross validation was applied. An aneuploidy score was calculated by subtracting the sum of expression of down-regulated genes from the sum of expression of up-regulated genes in the signature. Immunohistochemistry (IHC) was performed on tissue microarrays from 526 patients, with STAG2 antibody SA-2 (J12): sc-81852 (Santa Cruz Biotechnology). TCGA EC data were explored, and used for external validation. Results: Aneuploidy was significantly associated with predictors of poor outcome including FIGO stage, grade, histologic subtype and age (p<0.001), and worse 5-year disease specific survival (DSS, p<0.001). SVM identified 9 genes (3 up- and 6 down-regulated) yielding a good accuracy for identifying aneuploid tumor status. The aneuploidy score was significantly associated with all standardly applied clinicopathologic surrogate markers for outcome (p<0.01) and DSS (p = 0.001). No association was found between STAG2 expression by IHC and ploidy status, however loss of STAG2 was associated with better DSS than intact STAG2 (p = 0.05). In TCGA data, 8.1% of EC patients had STAG2 mutations, with slightly better DSS compared to patients without mutation (p = 0.05). Conclusions: DNA ploidy estimated by flow cytometry adds prognostic information in EC. Loss of STAG2 expression is not associated with aneuploidy. A 9-gene signature reflects ploidy status and predicts aggressive tumor behavior. The role of the 9 genes will be further explored to unveil potential involvement in aneuploidy development in EC. Citation Format: Karen K. Mauland, Kanthida Kusonmano, Elisabeth Wik, Mari K. Halle, Jone Trovik, Hans K. Haugland, Anne M. Oyan, Helga B. Salvesen. Aneuploidy predicts aggressiveness and poor prognosis in endometrial cancer, and is reflected in a 9-gene signature. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 3864. doi:10.1158/1538-7445.AM2015-3864