New blood test could predict women at risk for preeclampsia

An experimental blood test uses plasma cell-free RNA (cfRNA) to accurately predict preeclampsia as early as 16 weeks

As the leading cause of pregnancy-related deaths, early detection of preeclampsia could significantly reduce global maternal morbidity and mortality. New research reported in the journal Nature shows promise for the Mirvie RNA test to do just that.

The RNA technology, developed by San Francisco-based biotech company Mirvie, uses machine learning to analyze cfRNA messages to track pregnancy progression at the placental, maternal, and fetal level, notes the study. With a single blood sample, this experimental test was proven to accurately track pregnancy progression independently of clinical factors, and reliably predict preeclampsia months before it develops.1

While earlier studies have suggested cfRNA could be used to predict preeclampsia, authors of the Nature study used one of the largest and most diverse datasets to date. To understand and characterize molecular changes in the maternal-fetal-placental unit over gestation, they looked at comprehensive transcriptome data in 2,539 blood samples from 1,840 women in the United States, Europe, and Africa. Preeclampsia data included 524 samples (72 cases and 452 noncases) from 2 independent cohorts, and was collected 14.5 weeks before delivery.

After identifying the patterns and comprehensive molecular profiles of normal pregnancy progression, researchers used machine learning to find disruptions in those patterns, which indicated preeclampsia risk. The results showed a robust prediction of preeclampsia, with a sensitivity of 75% and a positive predictive value of 32.3%, which is superior to the state-of-the-art method at 4.4%.

“The results demonstrate that we can predict preeclampsia early in pregnancy, before it becomes a threat to the mother and baby. This predictive ability is unprecedented in contemporary obstetrics,” said the study’s senior author Thomas McElrath, MD, PhD, an attending physician in maternal-fetal medicine, Brigham and Women’s Hospital, Boston, and professor of obstetrics, Harvard Medical School in a press release.2

Differentiating superimposed preeclampsia in women with preexisting hypertension and exacerbation of baseline chronic hypertension can be difficult. The study, however, found that the signal for preeclampsia is specific to hypertension driven by a placental disorder, and is independent of signals associated with chronic hypertension

These findings open a new window for early detection and may ultimately reduce maternal morbidity and mortality.


  1. Rasmussen M, Reddy M, Nolan R, et al. RNA profiles reveal signatures of future health and disease in pregnancy. Nature. Published online January 5, 2022. doi:10.1038/s41586-021-04249-w
  2. Study published in nature shows Mirvie can predict pregnancy complications months before they happen using new, proprietary RNA platform. Published January 5, 2022. Accessed January 6, 2022.