When you’re expecting a baby, there’s a lot to think about. But for millions of parents around the world, their planning timeline gets unexpectedly cut short. Preterm birth—defined as delivery before 37 weeks—isn’t just an early arrival. It’s the leading cause of death in newborns and can lead to long-term health issues like breathing problems, developmental delays, and learning disabilities.
So what if there were a way to know, months in advance, if a pregnancy was at risk for delivering early?
A new study published in PLOS Medicine in April 2025 says that might now be possible. Researchers have discovered a way to analyze DNA fragments in a pregnant person’s blood—collected during routine non-invasive prenatal testing (NIPT)—to predict preterm birth with surprising accuracy.
Preterm birth (PTB) affects about 1 in 9 pregnancies globally. That’s around 11% of all births. And while advances in neonatal care have improved outcomes for many premature babies, early delivery still carries significant risks. Babies born too soon may face:
- Respiratory distress
- Low birth weight
- Feeding difficulties
- Vision or hearing problems
- Developmental delays
It’s not just the babies who are at risk—preterm birth is also linked to health complications for the birthing parent and is responsible for more than a third of all pregnancy-related deaths worldwide.
What makes preterm birth particularly challenging is that it’s incredibly difficult to predict. While factors like age, prior preterm delivery, or carrying multiples can increase risk, most parents who deliver early don’t have obvious warning signs.
What Is Cell-Free DNA?
If you’ve been pregnant recently, you might have heard of (or even taken) a non-invasive prenatal test (NIPT). These tests look at fragments of DNA that float freely in the blood—known as cell-free DNA (cfDNA)—to screen for conditions like Down syndrome.
What many people don’t realize is that this DNA includes genetic material not just from the baby (actually from the placenta), but also from the pregnant person’s own tissues. Researchers have recently discovered that this “scrap” DNA carries more information than we once thought.
In this study, scientists explored whether analyzing the way this DNA is packaged—specifically in the regions around gene “promoters” (the on/off switches for genes)—could provide early signs of trouble.
Researchers from Southern Medical University and partnering hospitals in China analyzed blood samples from 2,590 pregnant women. Of these, 518 pregnancies ended in spontaneous preterm birth, while 2,072 went full term.
Using a technique called “promoter profiling,” they looked at how tightly DNA was wrapped around nucleosomes in the samples. Think of it like analyzing not just which books are on a shelf, but how easy or hard it is to open them.
Their theory was simple: in pregnancies destined for preterm birth, certain genes—especially those related to placental health and immune function—might already be behaving differently, months before symptoms appear. And those differences would show up in the cfDNA collected during routine bloodwork.
To test this, they used machine learning to analyze patterns in 228 genes and developed a tool they named PTerm (Promoter profiling classifier for preterm prediction).
The Results: A Test That Gets It Right Most of the Time
PTerm was able to predict preterm birth with strong accuracy. In statistical terms, its “AUC” score was 0.849 across all cohorts—this is a way of measuring how well a test separates two outcomes (in this case, preterm vs full-term birth). A perfect test would score 1.0, while random guessing scores 0.5.
PTerm could identify pregnancies at risk of spontaneous preterm birth using blood samples taken between 12 and 28 weeks of pregnancy.
The prediction was particularly strong for very early preterm births (before 35 weeks), with an accuracy of 86.6%.
Even when researchers tried combining PTerm with other known risk factors like body mass index (BMI) or fetal DNA fraction, the PTerm classifier alone remained the most powerful predictor.
What Makes This Different from Current Preterm Screening?
Right now, there’s no standard blood test to predict spontaneous preterm birth. Doctors may monitor cervical length with ultrasound or consider a patient’s history, but these tools are limited and usually come into play later in pregnancy.
What’s revolutionary about this study is that it uses a blood sample many pregnant people are already giving anyway for routine genetic screening. No new blood draws. No added cost. Just more information from the same vial of blood.
If validated in future studies and approved for clinical use, PTerm could one day be added behind the scenes of standard NIPT—offering a heads-up to parents and care providers that extra monitoring or early intervention might be needed.
During pregnancy, DNA in the blood comes mostly from placental cells and immune system cells. These cells are constantly turning over, and as they die off, they shed DNA into the bloodstream.
This DNA isn’t floating around loose—it’s packaged in structures called nucleosomes. Depending on which genes are turned “on” or “off,” nucleosomes are arranged differently. For example, a gene that’s active will have more open space around its promoter region, while a gene that’s silenced is more tightly packed.
By sequencing cfDNA and examining how densely it clusters around these promoter regions, scientists can infer which genes are active in the placenta or immune system—and spot problems before they show up on an ultrasound.
In pregnancies that ended early, researchers noticed specific patterns: genes involved in inflammation, hormone signaling (like oxytocin), and placental function had distinct promoter profiles.
What Are the Potential Benefits of This Test?
If this test becomes part of standard prenatal care, it could offer:
- Early intervention: More time to implement strategies like progesterone treatment, cervical cerclage, or closer monitoring.
- Personalized care: Women with no previous risk factors could still be identified and supported.
- Better outcomes: With earlier detection comes the possibility of improving neonatal outcomes through preventive measures or planning for delivery in high-level care centers.
The study is a huge step forward, but it’s not without limitations.
- Population: The research was conducted entirely in China. More studies are needed to see how well PTerm performs across different ethnicities and healthcare systems.
- Timing: The test window is between 12 and 28 weeks, but the researchers didn’t pinpoint the best time to screen. Earlier detection might be possible with further study.
- Biological complexity: Preterm birth isn’t caused by just one thing. It can result from infection, inflammation, placental problems, and more. While PTerm captures patterns common to many cases, it may not catch all causes.
Still, the potential is enormous—and future versions of PTerm could be even more accurate by combining promoter data with other “omics” data like RNA or protein levels.
This study offers hope that preterm birth, one of the most unpredictable and dangerous complications of pregnancy, could become more predictable—thanks to information already hiding in a standard blood test.
More Pregnancy News:
- Can Avocados During Pregnancy Help Prevent Food Allergies in Babies?
- Treating Opioid Use Disorder in Pregnancy Cuts Risk of Complications by 30%
- Cracking the Placenta’s Genetic Code: How Tiny RNAs Help Shape a Healthy Pregnancy
Pregnancy Starter Pack
This post includes affiliate links where we will receive a small percentage if you purchase through the link.
