A first-trimester blood marker tracked preterm birth risk but predicted it weakly

Preterm birth is a leading cause of newborn death and of long-term complications in the children who survive it, and obstetrics has long wanted a way to flag the pregnancies at highest risk early enough to act. Placental growth factor, or PlGF, is a protein central to how the placenta builds its blood supply, and it is already measured later in pregnancy to help assess pre-eclampsia. Whether a first-trimester reading can forecast preterm birth is less settled. A prospective cohort study published in the International Journal of Gynecology & Obstetrics on 5 October tested the idea directly [s1].

The study

Researchers at Peking University International Hospital enrolled 1,448 women with singleton pregnancies between August 2023 and October 2025, measuring PlGF in a first-trimester blood sample drawn at 11 to 13 weeks plus six days of gestation [s1]. Raw concentrations were converted into multiples of the median using the Fetal Medicine Foundation's online calculator, a standard adjustment that accounts for gestational age and other factors [s1]. The primary outcome was preterm birth, defined as delivery between 28 and fewer than 37 weeks [s1].

Among the 1,448 participants, 80 (5.5%) delivered preterm [s1]. The analysis used multivariable log-binomial regression, restricted cubic splines and Cox proportional hazards models to relate the marker to that outcome [s1].

An association, in the expected direction

Lower first-trimester PlGF was associated with a higher risk of preterm birth, with an adjusted risk ratio of 0.303 (95% CI 0.14–0.61) after accounting for maternal characteristics [s1]. Because the ratio is expressed per unit of the marker, a value below 1 means that as PlGF rises, preterm birth risk falls — so women with lower readings were at greater risk.

The relationship was not a straight line. Restricted cubic spline analysis showed a significant nonlinear association (p for nonlinearity <0.001), with the excess risk concentrated among women at the low end of the PlGF range [s1]. A time-to-event analysis pointed the same way: higher PlGF was associated with a lower cumulative incidence of preterm birth, with an adjusted hazard ratio of 0.56 (95% CI 0.35–0.88) [s1].

Taken together, these are internally consistent signals that first-trimester PlGF carries real information about which pregnancies will end early.

Carrying information is not the same as predicting

The gap between association and prediction is where this study is most honest, and most useful to read carefully. When the researchers added PlGF to a clinical prediction model, the area under the curve — a standard measure of how well a model separates those who will have the outcome from those who will not — rose only from 0.590 to 0.684 (P=0.01) [s1]. The improvement was statistically significant, but the authors describe the model's predictive performance as remaining limited [s1].

An area under the curve of 0.68 is modest. A value of 0.5 is no better than a coin flip and 1.0 is perfect; 0.68 is the kind of figure that can shift population-level risk estimates but is not accurate enough to tell an individual woman whether she will deliver early. That distinction — a marker that is associated with an outcome across a group yet predicts poorly for any one person — is one of the most commonly misread results in clinical research, and the paper does not oversell it.

Limits

The study was conducted at a single centre, which constrains how far its numbers transfer to other populations and testing platforms [s1]. The authors themselves conclude that first-trimester PlGF might add information for early identification of higher-risk women, but that its clinical utility requires validation in larger and more diverse populations [s1]. There is also the structural caution that applies to any single biomarker for preterm birth: the condition has many causes — infection, cervical factors, placental disease, multiple others — and no one placental protein is likely to capture all of them.

What to watch

The useful next step is external validation in a different and larger population, ideally combining PlGF with other first-trimester measures rather than relying on it alone, since the modest standalone accuracy suggests its value, if any, lies in a composite. For now, the finding is a well-conducted confirmation that early PlGF tracks preterm birth risk, paired with a clear-eyed acknowledgement that tracking is not yet predicting.

This article describes research findings and is not medical advice.

Sources

• [s1] Association between first-trimester placental growth factor and preterm birth: A prospective cohort study. International Journal of Gynecology & Obstetrics, 5 October 2026. https://doi.org/10.1002/ijgo.71370