Plasmatic circRNAs Panel to Predict the Risk of Macrosomia in Women with Gestational Diabetes Mellitus.

ABSTRACT

OBJECTIVES: Fetal macrosomia and its associated complications are the most frequent and serious morbidities for infants associated with gestational diabetes mellitus (GDM). In this study, we aimed to determine the expression of circulating circRNAs in humans, which may be promising biomarkers for the diagnosis of GDM or predicting the macrosomia in GDM patients. DESIGN: A multi-stage validation and risk score formula analysis was applied for validation. PARTICIPANTS/MATERIALS, SETTING, METHODS: A total of 26 circRNAs previously reported highly expressed in placenta tissues or umbilical cord blood of GDM patients during the pregnancy were enrolled. We recruited a total of 200 patients with GDM with or without macrosomia, 200 healthy pregnant woman, and 200 healthy volunteers. RESULTS: We discovered that four circRNAs including circRNA_1030, circRNA_23658, circRNA_0009049, and circRNA_32231 were upregulated in plasmatic samples of patients with GDM with or without macrosomia in training set and validation set compared with the healthy pregnant woman and healthy volunteers. Further receiver operating characteristic (ROC) curve analysis in risk score formula indicated a high diagnostic ability and area under ROC curve value (AUC) of 0.950 and 0.815 in training set and validation set for predicting GDM from controls group, for predicting macrosomia from GDM, the AUC was 0.975 and 0.820, respectively. The four circRNAs were further investigated with stable expression in human plasma samples. LIMITATIONS: The study was limited by larger scale of sample validation and the detailed mechanism investigation. CONCLUSION: The circRNA_1030, circRNA_23658, circRNA_0009049, and circRNA_32231 might be the potential biomarkers for predicting the GDM and macrosomia during the perinatal period.