Expression profiling of 22 genes involved in the PI3K-AKT pathway identifies two subgroups of high-grade endometrial carcinomas with different molecular alterations

Previously, we showed that PIK3CA and p53 alterations in uterine endometrial carcinomas correlate with poor prognosis. However, the contribution of phosphatidylinositol 3-kinase (PI3K)-AKT deregulation to endometrial carcinogenesis is not completely understood. The purpose of this study was to analyze alterations of this pathway in endometrial carcinomas and correlate them with the most common genetic abnormalities. Expression profiling of 22 genes involved in PI3K-AKT signaling pathway was analyzed in 38 endometrial carcinomas using TaqMan low-density array (TLDA) analysis. The gene expression pattern was analyzed by hierarchical clustering analysis. Unsupervised clustering divided the high-grade endometrial carcinomas into two clusters. One cluster identified tumors with alterations in the PI3K-AKT signaling pathway (exon 20 PIK3CA mutations and/or PTEN mutations 9/15; 60%), and p16 protein overexpression (8/13; 62%). Almost all non-endometrioid adenocarcinomas (serous and clear cell adenocarcinomas) were segregated into this cluster. In contrast, the other cluster identified tumors with p53 alterations (6/6; 100%), p16 protein overexpression (5/5; 100%), and exon 9 PIK3CA mutations (2/6; 33%). Exon 20 PIK3CA and PTEN mutations were not found in this subgroup. Low-grade endometrial carcinomas clustered in a third subgroup characterized by high frequency of PTEN mutations (10/17; 59%) and microsatellite instability (6/17; 35%). Our results show that gene expression profile differences in the PI3K-AKT signaling pathway identify two subgroups of high-grade endometrial carcinomas with different molecular alterations (PI3K-AKT pathway vs p53 alterations) that may have distinct roles in endometrial carcinogenesis. Identification of these subgroups can provide insight into the biology of these tumors and may facilitate the development of future treatments. Modern Pathology (2010) 23, 694-702; doi: 10.1038/modpathol.2010.44; published online 19 February 2010