Noninvasive Preeclampsia Prediction Using Plasma Cell-Free RNA Signatures
Si Zhou,Jie Li,Wenzhi Yang,Penghao Xue,Yanning Yin,Yunfang Wang,Peirun Tian,Huanhuan Peng,Hui Jiang,Wenqiu Xu,Shang Huang,Rui Zhang,Fengxiang Wei,Hai-Xi Sun,Jianguo Zhang,Lijian Zhao
DOI: https://doi.org/10.1016/j.ajog.2023.05.015
IF: 9.8
2023-01-01
American Journal of Obstetrics and Gynecology
Abstract:BACKGROUND: Preeclampsia, especially preterm preeclampsia and early-onset preeclampsia, is a life-threating pregnancy disorder, and the heterogeneity and complexity of preeclampsia make it difficult to predict risk and to develop treatments. Plasma cell-free RNA carries unique information from human tissue and may be useful for noninvasive monitoring of maternal, placental, and fetal dynamics during pregnancy. OBJECTIVE: This study aimed to investigate various RNA biotypes associated with preeclampsia in plasma and to develop classifiers to predict preterm preeclampsia and early-onset preeclampsia before diagnosis. STUDY DESIGN: We performed a novel, cell-free RNA sequencing method termed polyadenylation ligation-mediated sequencing to investigate the cell-free RNA characteristics of 715 healthy pregnancies and 202 pregnancies affected by preeclampsia before symptom onset. We explored differences in the abundance of different RNA biotypes in plasma between healthy and preeclampsia samples and built preterm preeclampsia and early-onset preeclampsia prediction classifiers using machine learning methods. Furthermore, we validated the performance of the classifiers using the external and internal validation cohorts and assessed the area under the curve and positive predictive value. RESULTS: We detected 77 genes, including messenger RNA (44%) and microRNA (26%), that were differentially expressed in healthy mothers and mothers with preterm preeclampsia before symptom onset, which could separate participants with preterm preeclampsia from healthy samples and that played critical functional roles in preeclampsia physiology. We developed 2 classifiers for predicting preterm preeclampsia and early onset preeclampsia before diagnosis based on 13 cell-free RNA signatures and 2 clinical features (in vitro fertilization and mean arterial pressure), respectively. Notably, both classifiers showed enhanced performance when compared with the existing methods. The preterm preeclampsia prediction model achieved 81% area under the curve and 68% positive predictive value in an independent validation cohort (preterm, n=46; control, n=151); the early-onset preeclampsia prediction model had an area under the curve of 88% and a positive predictive value of 73% in an external validation cohort (early-onset preeclampsia, n=28; control, n=234). Furthermore, we demonstrated that downregulation of microRNAs may play vital roles in preeclampsia through the upregulation of preeclampsia-relevant target genes. CONCLUSION: In this cohort study, a comprehensive transcriptomic landscape of different RNA biotypes in preeclampsia was presented and 2 advanced classifiers with substantial clinical importance for preterm preeclampsia and early-onset preeclampsia prediction before symptom onset were developed. We demonstrated that messenger RNA, microRNA, and long noncoding RNA can simultaneously serve as potential biomarkers of preeclampsia, holding the promise of prevention of preeclampsia in the future. Abnormal cell-free messenger RNA, microRNA, and long non coding RNA molecular changes may help to elucidate the pathogenic determinants of preeclampsia and open new therapeutic windows to effectively reduce pregnancy complications and fetal morbidity.