In the labyrinth of maternal health, few conditions are as insidious and life-threatening as preeclampsia. Often silent until it isn’t, this hypertensive disorder of pregnancy can turn the miracle of life into a medical emergency. But a groundbreaking study led by Mirvie Inc., published in Nature Communications, may have just cracked open a powerful new tool for prediction—one that uses the body’s own molecular whispers, long before the scream of clinical symptoms.
The study’s promise is bold: a simple blood test, conducted in mid-pregnancy, could forecast preeclampsia and related disorders months before they manifest, shifting obstetric care from reaction to prevention. For the millions of pregnant individuals walking an invisible tightrope, this discovery could be the net beneath their feet.
The Crisis of Hypertensive Disorders in Pregnancy
Hypertensive disorders of pregnancy (HDP), which include preeclampsia, gestational hypertension, and other related complications, now affect one in every six pregnancies in the United States. The incidence has nearly doubled between 2007 and 2019, making it one of the leading causes of maternal and neonatal morbidity and mortality.
Yet, despite its growing prevalence, our understanding of preeclampsia remains remarkably fragmented. Its manifestations vary wildly—from mild postpartum hypertension to severe early-onset cases requiring emergency preterm delivery. Researchers have proposed various triggers: viral infections, immune activation, and placental dysfunction chief among them. But without a clear and consistent biological roadmap, clinicians have long relied on crude predictors and reactive diagnostics.
Alarmingly, only 5% of pregnant individuals meet the established criteria for “high risk,” such as having chronic hypertension or a history of preeclampsia. That leaves a staggering 95% of pregnancies in an uncertain gray zone—monitored through demographic risk factors like race and income, which often offer little predictive power and can introduce unintended bias.
Until now, diagnosis typically arrives only after symptoms emerge—a dangerous delay in a condition that can escalate in days or even hours.
A Molecular Breakthrough: Listening to Cell-Free RNA
Mirvie’s study took a radically different approach. Instead of waiting for clinical signs, researchers tuned into molecular signals, specifically cell-free RNA (cfRNA)—tiny fragments of genetic material circulating in maternal blood that offer a window into both maternal and placental health.
The scale of the study was impressive: 10,745 pregnant participants enrolled across 11 clinical sites and through mobile phlebotomy services. Blood samples were collected between 17 weeks and 4 days to 22 weeks gestation, aligning with the standard anatomy scan window. A total of 9,102 samples were ultimately analyzed.
Using high-throughput sequencing, the team generated transcriptomes—a snapshot of which genes were being actively expressed at the time of collection. These gene expression profiles were then used to identify molecular subtypes of HDP.
Participants were split into a training set (5,399 samples) and a validation set (2,829 samples), allowing for robust model testing and confirmation across diverse patient populations and clinical severities.
Two Molecular Signatures, Two Distinct Risks
What emerged were two clear and distinct gene expression patterns, each correlating with different manifestations of hypertensive disorders.
1. The Placental Subtype: A Prelude to Prematurity
This group showed high expression of placental-specific genes, particularly PAPPA2, a gene associated with placental development and nutrient transfer. Individuals in this category were more likely to develop early-onset or severe preeclampsia, often requiring preterm delivery.
These cases weren’t just detected—they were detected early. The median lead time from blood draw to delivery was a full 108 days, giving clinicians a precious window for intervention. Importantly, PAPPA2 overexpression showed no link to spontaneous preterm birth, suggesting this signal is highly specific to hypertensive pathology.
Test-positive individuals in this group delivered, on average, 11.3 days earlier than their test-negative counterparts. Among newborns admitted to the NICU, test-positive cases stayed nearly three times longer—14 days versus just five.
2. The Immune Subtype: The Term-Time Threat
The second group displayed elevated expression of immune-related genes, such as cytokines and inflammatory markers. These individuals often developed term preeclampsia or gestational hypertension, conditions that may appear less severe on the surface but still carry significant risks for both mother and baby.
The median lead time to delivery for this group was 132 days, a substantial head start for a condition that has traditionally been unpredictable.
Outperforming Traditional Risk Models
The RNA-based classifier wasn’t just predictive—it was powerfully predictive, especially in those without traditional high-risk markers. Among individuals of advanced maternal age with no other risk factors, the classifier achieved an area under the curve (AUC) of 0.88—a metric that denotes excellent diagnostic performance.
Compared to the U.S. Preventive Services Task Force (USPSTF) guidelines, the positive likelihood ratio was 15 times higher, and the negative likelihood ratio dropped from 0.38 to a remarkable 0.12. The negative predictive value for placental-associated cases was 99.7%, meaning if the test says you’re in the clear, you almost certainly are.
Unlike existing screening methods that depend heavily on race, age, body mass index, and socioeconomic factors, this classifier was independent of demographic variables. That makes it not only more accurate but also more equitable.
Implications: A New Era for Maternal Precision Medicine
The implications of this discovery are profound. With just a single blood sample—drawn during routine mid-pregnancy care—clinicians could soon identify which pregnancies are quietly on the path to danger. This would enable:
- Personalized preventive care: Interventions like low-dose aspirin, closer monitoring, and early treatment could be directed only to those truly at risk.
- Better clinical trial design: Rather than recruiting all high-BMI or older pregnant individuals, trials could focus on molecularly high-risk populations.
- Reduced NICU burden: Earlier detection could mean fewer surprise emergencies, and potentially less need for intensive neonatal care.
- Improved maternal outcomes: Perhaps most importantly, it offers hope for lowering maternal mortality, which continues to rise in the U.S.
Looking Ahead: A Blood Test That Listens Before the Body Screams
For too long, hypertensive disorders of pregnancy have been a diagnosis of aftermath—a condition only visible in the mirror of emergency. This study represents a seismic shift in that paradigm. By listening to the genetic signals floating through maternal blood, we may finally be able to see preeclampsia coming long before it arrives.
It’s the difference between dodging a storm and walking straight into it.
As Mirvie’s classifier continues to undergo further validation and refinement, it points toward a future where pregnancy care isn’t guided by general risk categories or population-level statistics, but by the molecular story of each individual pregnancy.
In a field where seconds can save lives, the idea of buying months is nothing short of revolutionary.
Reference: Michal A. Elovitz et al, Molecular subtyping of hypertensive disorders of pregnancy, Nature Communications (2025). DOI: 10.1038/s41467-025-58157-y
Loved this? Help us spread the word and support independent science! Share now.