NEW YORK – An Illumina-led team has identified a blood expression signature stemming from circulating RNAs that appears to coincide with increased risk of preeclampsia complications relatively early in pregnancy.
As they reported in a paper appearing in Science Translational Medicine on Wednesday, the researchers used transcriptome enrichment and sequencing to search for preeclampsia-related RNAs in blood samples collected over time from more than 100 pregnant women with or without early-onset preeclampsia — a condition linked to a rise in maternal and perinatal morbidity and mortality risk.
The initial search led to 30 placenta-, fetal development-, maternal immune-, and cardiovascular-related transcripts with potential ties to early-onset eclampsia in circulation, the team reported, while a subsequent machine learning analysis expanded this set to 49 preeclampsia-related blood transcripts that showed promise for finding severe early-onset preeclampsia in another group of 24 pregnant women.
The researchers believe that the application of circulating RNA "will ultimately provide comprehensive molecular monitoring of maternal and fetal health throughout the course of pregnancy," senior author Fiona Kaper, senior director of scientific research at Illumina, and her co-authors wrote.
They noted that the current study and others focusing on samples collected earlier in pregnancy "hold great promise for uncovering predictive biomarkers for early stratification of all at-risk pregnancies, informing prophylactic interventions, or more vigilant monitoring of pregnancy."
Because much of the RNA found in blood plasma represents ribosomal or globin RNA rather than transcripts, the researchers came up with a probe-based plasma RNA sequencing approach that enriched for transcripts from gene exons in circulating RNA libraries, taking between-individual differences in RNA concentrations in blood plasma into account.
The team first validated the approach in 152 samples collected over tine from the first trimester to third trimester in 41 women with healthy, uncomplicated pregnancies, and uncovering circulating RNAs with shifting expression over pregnancy.
After confirming that this workflow robustly detected pregnancy-related circulating RNA dynamics, the investigators then sought to identify changes in circulating RNA associated with pregnancy complications.
Across 113 blood samples collected from 40 women with severe, early-onset preeclampsia and 73 women with gestational age-matched pregnancies, they saw more than three-dozen suspicious transcripts with enhanced representation in blood plasma from the preeclampsia pregnancies, focusing in on 30 preeclampsia-related circulating transcripts with follow-up quantitative PCR and validation analyses.
Bringing in machine learning classifier clues, the team landed on a set of 49 circulating RNA transcripts that could pick up early-onset preeclampsia pregnancies with between 85 percent and 89 percent accuracy, on average, in another group of two dozen women with or without preeclampsia. The average accuracy of the circulating RNA-based classifier dipped to around 72 percent for women in the validation cohort experiencing late-onset preeclampsia.
"We detected molecular changes specific to the complex pathophysiology of early-onset severe [preeclampsia] at the time of diagnosis, supporting robust classification across cohorts," the authors reported, noting that the altered circulating RNA transcripts identified "represented contributions from maternal, placental, and fetal tissues, many of which would not be captured in studies focusing on placental tissues collected after delivery."