Metastasis is the main cause of cancer mortality. One of the initiating events of cancer metastasis of epithelial tumors is epithelial-to-mesenchymal transition (EMT), during which cells dedifferentiate from a relatively rigid cell structure/morphology to a flexible and changeable structure/morphology often associated with mesenchymal cells. The presence of EMT in human epithelial tumors is reflected by the increased expression of genes and levels of proteins that are preferentially present in mesenchymal cells. The combined presence of these genes forms the basis of mesenchymal gene signatures, which are the foundation for classifying a mesenchymal subtype of tumors. Indeed, tumor classification schemes that use clustering analysis of large genomic characterizations, like The Cancer Genome Atlas (TCGA), have defined mesenchymal subtype in a number of cancer types, such as high-grade serous ovarian cancer and glioblastoma. However, recent analyses have shown that gene expression-based classifications of mesenchymal subtypes often do not associate with poor survival. This "paradox" can be ameliorated using integrated analysis that combines multiple data types. We recently found that integrating mRNA and microRNA (miRNA) data revealed an integrated mesenchymal subtype that is consistently associated with poor survival in multiple cohorts of patients with serous ovarian cancer. This network consists of 8 major miRNAs and 214 mRNAs. Among the 8 miRNAs, 4 are known to be regulators of EMT. This review provides a summary of these 8 miRNAs, which were associated with the integrated mesenchymal subtype of serous ovarian cancer.
Cystadenocarcinoma, Serous ; genetics ; pathology ; Epithelial-Mesenchymal Transition ; Female ; Humans ; MicroRNAs ; physiology ; Ovarian Neoplasms ; genetics ; pathology

Objective To compare the differentiating effect of Berg Balance Scale (BBS), Mini-Balance Evaluation Systems Test (Mini-BESTest) and Brief-BESTest on the risk of falls in chronic obstructive pulmonary disease (COPD) patients. Methods From September, 2016 to March, 2017, 22 COPD patients with history of falls and other age-matched 66 COPD patients without history of falls were assessed with BBS, Mini-BESTest and Brief-BESTest. The data were compared with paired Logistic regression, Log-likelihood estimate, and the skewness coefficient were calculated. Results BBS, Mini-BESTest and Brief-BESTest all were valuable for differentiating the risk of falls in COPD patients. Log-likelihood estimate value was the least in Brif-BESTest (5.372), followed with Mini-BESTest (12.918) and BBS (17.644). There was not a ceiling effect in Brief-BESTest and Mini-BESTest, but there was in BBS. Conclusion All the BBS, Mini-BESTest and Brief-BESTest can predict the risk of falls in patients with COPD, and Brief-BESTest is the most predictive. There is a ceiling effect in BBS for COPD patients, but not in Brief-BESTest and Mini-BESTest.