Abstract LB-124: Distinct gene expression signatures for endometrial cancer predict for subtype, stage, and prognosis

Abstract

Abstract Introduction: Endometrial cancer (EC) is the most common gynecologic malignancy in the United States. It is generally divided into one of two categories: Type I, which present as low grade, early stage tumors with good prognosis and Type II which are high grade and usually advanced stage with poor outcomes, despite chemotherapy, even in early stage disease. Despite the well-documented clinico-pathologic differences between the two types, little is known about the genetic etiology of the discrepancies in behavior and outcomes between the two subtypes. The goal of this study was to identify distinctly expressed genes (DEG) in early stages vs. late stages in each of type I and II tumors as well as DEGs in ‘good’ vs. ‘poor’ tumor outcomes in each of the two types. Materials and Methods: A total of 24 EC tissue samples were collected (type I =12, type II =12). Six cases of each type were early stages and six were late stage. The cases were further categorized as having either ‘good’ prognosis, defined as the patient being alive with no evidence of disease at follow-up or ‘poor’ prognosis, with the patient having either disease progression, persistence, recurrence or death from disease at follow-up. RNA was extracted and human genome wide illumine bead microarrays carrying 48,000 genes was performed. Statistical analysis using the limma program in the R-based bioconductor package was used to calculate the level of gene differential expression of each comparison conducted. Results: Four separate comparisons based on patient characteristics were performed including: type I (late vs. early stage), type II (late vs. early), type I prognosis (‘good’ vs. ‘poor’), and type II (‘good’ vs. ‘poor’). Using the expression value of the identified DEGs we were able to accurately cluster patients into their respective clino-pathologic groups (late vs. early stages and ‘good’ vs. ‘poor’ prognosis). For type I tumors (late vs. early), 111 DEGs were identified (p<0.01). For type II tumors (late vs. early) 274 DEGs were identified (p<0.01). For prognosis (good vs. poor) there were 112 DEGs for type I and 135 for type II tumors (p<0.01 each). Furthermore, there was little overlap in the DEG sets between stages in type I and type II tumors (only 4 shared DEGs among the 111 and 274 DEGs, respectively). Furthermore, there was an insignificant overlap in DEG between good and poor prognosis in type I and type II (only 1 shared DEG among the 112 and 135, respectively). Remarkably, there was no overlap between the stage-derived DEGs and the prognosis-derived DEGs for each of type I and type II tumors. Conclusion: These data suggest that type I and type II tumors have very distinct gene expression signatures. We have shown the involvement of very distinct groups of genes in tumor progression (early vs. late stage) and that they are different from those involved in tumor outcome in type I and type II tumors. These genes might be important in tumor staging and predicting outcome. The evaluation of mRNA expression of some of these genes by RT-PCR affymetrix technique is in progress in our laboratory to identify which ones have value in determining pathogenesis and progression. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr LB-124.