Abstract
Placental dysfunction, including preeclampsia and fetal growth restriction, is a leading contributor to maternal and fetal morbidity and mortality. Development of effective prevention and treatment strategies has been limited by lack of identification of at-risk pregnancies. We performed small RNA sequencing of maternal serum microRNAs (miRNAs) to discover and independently verify biomarkers differentially expressed in patients who later developed preeclampsia. In this study, maternal serum was collected between 17-28 weeks gestation in two cohorts: The high-risk UCSD Placenta Study (19 cases, 29 controls); and the Sera Prognostics PAPR Study (54 cases, 110 controls). Samples were divided into Discovery (2/3) and Verification sets (1/3). Cases and controls were matched for gestational age (GA) at blood draw. miRNAs were isolated and sequenced to yield miRNA counts. Individual miRNAs and pairwise ratios (reversals) were tested in blinded verification. Biomarker discovery and verification were performed on the entire GA range and for 3 GA windows: 170/7 - 215/7 (Early); 195/7 - 244/7 (Middle); and 222/7 - 280/7 (Late) weeks. In Discovery, best performing miRNAs and reversals were selected for each GA window. Performance was then confirmed in blinded Verification (Table 2). 7/50 Early GA reversals had significant AUCs (mean Discovery AUC 0.78, mean Verification AUC 0.79). 21/50 Late GA reversals had significant AUCs (mean Discovery AUC 0.71, mean Verification AUC 0.68). Of biological importance, verified reversals contained a high proportion of miRNAs that were strongly expressed in placenta and/or encoded in the chr14q32 and chr19q13 miRNA clusters. miR-155-5p, present in 21 verified reversals, has been reported to mediate repression of eNOS by TNFα, which is a regulatory interaction associated with preeclampsia. We identified and verified differentially abundant miRNAs in maternal serum at 17-28 weeks GA in pregnancies that later developed signs and symptoms of preeclampsia compared to pregnancies with uncomplicated outcomes. Our findings suggest that maternal serum miRNAs may be useful as biomarkers for prediction of placental dysfunction and warrant validation in large independent cohorts.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.