RESUMO
The discrimination of different subtypes of endometrial carcinoma (EC) is frequently problematic when using the current histomorphological classification; therefore, new markers for this differentiation are needed. Here, we examined differences in miRNA expression between well- and poorly-differentiated (grades 1 and 3) endometrioid endometrial carcinoma (EEC) and between EEC and serous endometrial carcinoma (SEC). The expression of 84 tumour-suppressor miRNAs was analysed by real-time polymerase chain reactions in 62 EC and 20 non-neoplastic endometrial specimens. The potential functions of the differentially expressed miRNAs were determined by bioinformatics analyses. The expression of let-7c-5p, miR-125b-5p, miR-23b-3p, and miR-99a-5p in grade 3 EEC was decreased compared to grade 1 EEC. To discriminate between EEC and SEC, let-7g-5p, miR-195-5p, miR-34a-5p, and miR-497-5p expression was significantly downregulated in SEC. In bioinformatic analyses, miRNAs that could discriminate grade 1 from grade 3 mainly targeted genes involved in PI3K-AKT signaling, whereas miRNAs that could discriminate EEC from SEC targeted genes involved in several signaling pathways, but mainly MAPK signaling. Taken collectively, our results indicate that the activation of certain signaling pathways can be useful in the molecular characterization of EEC and SEC.