A Blood Test to Help Predict and Prevent Early Preterm Birth

A Plain-Language Guide to the PRIME Trial (Iriye et al., 2026) on Biomarker Screening for Preterm Birth Risk

Why Predicting Preterm Birth Matters, and How Families Tried Before This Study

Being born too early — before 37 weeks of pregnancy — is one of the biggest threats to a newborn's health, and most families get no warning it's coming. A major new study tested whether a simple blood test partway through pregnancy could flag higher-risk pregnancies early enough to help, and found that acting on the results reduced serious newborn health problems and time in the NICU (neonatal intensive care unit, the hospital unit for sick or premature newborns).

That's the problem a large research study set out to solve. The Prematurity Risk Assessment Combined with Clinical Interventions for Improving Neonatal OutcoMEs (PRIME) trial [1] was carried out at 19 medical centers across the United States, including Yale University, the University of Virginia, and the Icahn School of Medicine at Mount Sinai. Researchers wanted to know whether a blood test could spot pregnancies at higher risk of early preterm birth even when nothing else about the pregnancy looked unusual, and whether doing something about that risk early would actually help the baby.

Long before this study, doctors and researchers tried several ways to predict and prevent preterm birth, with mixed success. Around the world, preterm birth has remained a leading cause of serious newborn health problems for years [2]. One older approach measures the cervix (the neck of the womb) by ultrasound, since a shorter-than-normal cervix in mid-pregnancy is linked to higher preterm birth risk; another, called fetal fibronectin testing, checks for a protein that can signal early labor is starting soon [3]. Both tests work reasonably well for people who already have a warning sign, like a short cervix or a previous premature birth, but they are much less useful for the majority of pregnant people who have no such sign to begin with. That gap led researchers to look for a blood test that could work regardless of someone's pregnancy history. An early version of such a test was developed and shown to predict preterm birth risk in a general group of pregnant people with no obvious risk factors [4], and a later, larger study confirmed the test worked similarly well in an independent group of patients [5]. Meanwhile, two treatments already had solid evidence behind them for people known to be at higher risk: vaginal progesterone (a hormone medication) for people with a short cervix [6], and low-dose aspirin for people having their first baby [7]. What no one had tested at a large scale was whether combining the blood test with these already-available treatments, offered specifically to the people the test flagged as higher-risk, could actually improve outcomes for babies. A smaller earlier study took a first step in that direction [8], and the same blood test ratio was separately confirmed in an international study population [9] — but PRIME [1] was designed to be the large, definitive test of the whole strategy.

Inside the PRIME Study: How Researchers Tested a New Blood Test

To test this, researchers enrolled 5,018 pregnant people who were between about 18 and 21 weeks pregnant, carrying a single baby (not twins or multiples), and who did not already have a known risk factor for preterm birth, like a previous premature birth or a cervix that measured too short on ultrasound. Everyone in the study had blood drawn for the biomarker test, which measures the ratio of two proteins: IGFBP4 (insulin-like growth factor binding protein 4) and SHBG (sex hormone binding globulin). A result predicting a 15% or higher chance of early preterm birth — roughly double the average risk for a single pregnancy in the US — counted as "higher risk."

Participants were randomly assigned, essentially by coin flip, to one of two groups. In the "screen-guided care" group, both the patient and her doctor were told the test result, and the roughly 1 in 4 participants (23.5%, or 594 out of 2,526) who screened higher-risk were offered a bundle of extra care: daily vaginal progesterone, daily low-dose aspirin, weekly check-in phone calls from a specialized nurse, two extra ultrasounds to re-check the cervix, and, if needed, a minor procedure called cerclage (a stitch to help support the cervix) for anyone whose cervix became very short. About 6 in 10 of those offered the bundle (57.7%, 343 of 594) actually started it before 24 weeks of pregnancy. In the "routine care" group, the same blood test was done, but the result was hidden from everyone — the patient, her doctor, and even the research team analyzing outcomes — so this group received only the usual pregnancy care regardless of their result. This kind of hidden-result design helped make sure any difference between the groups came from acting on the test, not from something else. In the end, researchers compared outcomes for 4,806 babies (2,416 in the screen-guided group and 2,390 in the routine-care group).

What the Study Found, in Real Numbers

The results favored the group that acted on the blood test. Newborns in the screen-guided care group had fewer serious health problems overall: using a 0-4 point scale for newborn health complications (things like breathing problems, bleeding in the brain, or infections), 90.1% of these babies had a score of zero (no complications), compared with 87.8% in the routine-care group — meaning about 16 fewer babies were affected by a significant health problem in the screen-guided group (56 affected babies versus 72). Babies in the screen-guided group also spent less time in the hospital overall (an average of 2.92 days versus 3.07 days) and were less likely to be admitted to the NICU at all: 10.2% (245 of 2,392) versus 12.8% (303 of 2,370) — a difference researchers describe using a "number needed to screen" of about 38, meaning that for roughly every 38 pregnant people who were tested and, if flagged, treated, one fewer baby ended up needing NICU care. Across the whole study population, babies in the screen-guided group spent 545 fewer total days in the NICU combined.

Very early preterm birth was also less common in the screen-guided group: births before 35 weeks occurred in 1.7% of that group versus 2.5% in the routine-care group, and births before 32 weeks occurred in 0.4% versus 0.9%. Interestingly, the average timing of delivery barely changed between the two groups (about 38 weeks and 6 days in both) — the benefit seems to come specifically from preventing the small number of earliest, most dangerous preterm births, rather than shifting everyone's due date later. Two babies died during the study, both in the routine-care group, and none in the screen-guided group. The extra treatments were generally well tolerated: mild skin itching or discharge related to the progesterone occurred in fewer than 1% of participants, one participant had a temporary liver enzyme change that went away after stopping the medication, and about 84% of participants who accepted the treatment bundle stuck with it consistently after 24 weeks.

What This Means for Families

For families, the headline takeaway is encouraging but comes with important context. If a pregnant person's blood test shows a higher risk of early preterm birth, this study suggests that starting a bundle of already-available, relatively low-cost treatments — vaginal progesterone, low-dose aspirin, and extra nurse support — soon after can meaningfully lower the chances of a serious newborn complication or a NICU stay, even for someone who otherwise looked like a low-risk pregnancy. That's genuinely useful, because most families who go on to have an early preterm birth have no advance warning under older screening methods.

At the same time, it's important to be honest about what the numbers mean at an individual level: most higher-risk participants who took the treatment bundle did not personally experience a prevented complication — the benefit shows up clearly at the population level, across thousands of pregnancies, more than as a guarantee for any one family. Families should also know that this test and trial were funded by Sera Prognostics, the company that makes and sells this exact blood test (marketed as PreTRM®). That doesn't mean the results are wrong — the study was carefully designed with independent, outside verification of the data, and the people analyzing results didn't know who got which treatment — but it's a real financial interest that families and doctors should feel comfortable discussing openly when this test comes up in prenatal care. Families should also know that the study only included people in the United States carrying a single baby with no major prior risk factors, so it's not yet clear how well the same approach would work for twins, for people with other risk factors, or in other countries. A related study looking specifically at first-time mothers has also been published, though its detailed results were not available for this summary [10], and the original study has a formal published correction that readers can look up directly for the most precise numbers [11].

What Researchers Are Working on Next

This study is unlikely to be the final word. Researchers and outside commentators are already discussing how a screen-guided strategy like this might be adapted for use in other countries and health systems, since PRIME [1] was carried out entirely in the United States [12]. Independent researchers not affiliated with the test's manufacturer will likely want to repeat this kind of study to confirm the findings hold up outside of an industry-funded trial. Cost is another open question: nobody has yet published a clear answer on whether widespread blood-test screening plus nurse support is affordable enough for typical health systems and insurance plans to adopt broadly. Scientists are also continuing to study exactly how accurate this specific blood test is and how to interpret its results most usefully [13], and researchers are looking more closely at a surprising finding in the study — slightly more cases of a condition called fetal growth restriction (where a baby grows more slowly than expected) among full-term babies in the screen-guided group — to understand whether this reflects babies being carried a bit longer and more safely overall, or something that needs closer attention on its own. An editorial published alongside the trial raised similar questions about whether the field is ready to put this kind of screen-guided care into routine practice [14]. For now, families can take away a hopeful but measured message: a promising new tool exists, it has been tested carefully in a large trial, and researchers are actively working through the next set of questions before it becomes standard care everywhere.

References

  1. Iriye BK, O'Brien JM, Ennen CS, et al. Neonatal impact of maternal biomarker screening for risk of preterm birth with targeted interventions (PRIME): A multicenter, randomized, controlled trial. Pregnancy. 2026;2(1):e70202. doi:10.1002/pmf2.70202 ↩
  2. Ohuma EO, Moller AB, Bradley E, et al. National, regional, and global estimates of preterm birth in 2020, with trends from 2010: a systematic analysis. Lancet. 2023;402(10409):1261-1271. doi:10.1016/S0140-6736(23)00878-400878-4) ↩
  3. Farnsworth C, Schuler EE, Woodworth A, Straseski J, Pschirrer ER, Nerenz RD. AACC Guidance Document on Laboratory Testing for the Assessment of Preterm Delivery. J Appl Lab Med. 2021;6(4):1032-1044. doi:10.1093/jalm/jfab039 ↩
  4. Saade GR, Boggess KA, Sullivan SA, et al. Development and validation of a spontaneous preterm delivery predictor in asymptomatic women. Am J Obstet Gynecol. 2016;214(5):633.e1-633.e24. doi:10.1016/j.ajog.2016.02.001 ↩
  5. Markenson GR, Saade GR, Laurent LC, et al. Performance of a proteomic preterm delivery predictor in a large independent prospective cohort. Am J Obstet Gynecol MFM. 2020;2(3):100140. doi:10.1016/j.ajogmf.2020.100140 ↩
  6. Romero R, Conde-Agudelo A, Da Fonseca E, et al. Vaginal progesterone for preventing preterm birth and adverse perinatal outcomes in singleton gestations with a short cervix: a meta-analysis of individual patient data. Am J Obstet Gynecol. 2018;218(2):161-180. doi:10.1016/j.ajog.2017.11.576 ↩
  7. Hoffman MK, Goudar SS, Kodkany BS, et al. Low-dose aspirin for the prevention of preterm delivery in nulliparous women with a singleton pregnancy (ASPIRIN): a randomised, double-blind, placebo-controlled trial. Lancet. 2020;395(10220):285-293. doi:10.1016/S0140-6736(19)32973-332973-3) ↩
  8. Hoffman MK, Kitto C, Zhang Z, Shi J, Walker MG, Shahbaba B, Ruhstaller K. Neonatal Outcomes after Maternal Biomarker-Guided Preterm Birth Intervention: The AVERT PRETERM Trial. Diagnostics (Basel). 2024;14(14):1462. doi:10.3390/diagnostics14141462 ↩
  9. Hirst JE, Boniface JJ, Le DP, et al. Validating the ratio of insulin like growth factor binding protein 4 to sex hormone binding globulin as a prognostic predictor of preterm birth in Viet Nam: a case-cohort study. J Matern Fetal Neonatal Med. 2024;37(1):2333923. doi:10.1080/14767058.2024.2333923 ↩
  10. Sciscione AC, Gyamfi-Bannerman C, Walker M, Shahbaba B, Godt J, Iriye BK. Biomarker screen-guided care for preterm birth risk in nulliparous pregnancies: a subgroup analysis of the PRIME randomized controlled trial. J Matern Fetal Neonatal Med. 2026;39(1):2689850. doi:10.1080/14767058.2026.2689850 ↩
  11. Iriye BK, et al. Correction to "Neonatal impact of maternal biomarker screening for risk of preterm birth with targeted interventions (PRIME): A multicenter, randomized, controlled trial." Pregnancy. 2026;2(5):e70437. doi:10.1002/pmf2.70437 ↩
  12. Di Renzo GC, Simeoni U, Anumba DOC. Commentary: Translating the PRIME Trial into European obstetric care. J Matern Fetal Neonatal Med. 2026;39(1):2642428. doi:10.1080/14767058.2026.2642428 ↩
  13. Garite TJ. Toward a Clearer Understanding of Accuracy and Proper Application of the IBP4/SHBG Ratio for Predicting Preterm Delivery. J Appl Lab Med. 2022;7(4):1004-1006. doi:10.1093/jalm/jfac019 ↩
  14. Ecker JL. Progress in reducing the burden of preterm birth: Are we ready for PRIME time? Pregnancy. 2026;2(1):e70210. doi:10.1002/pmf2.70210 ↩