NIH

Researchers develop new approach for predicting preeclampsia

March 14, 2025

To address the challenge of predicting preeclampsia early, an NIH-funded research team led by Drs. Raj Shree and Gavin Ha investigated the use of cell-free DNA in the bloodstream. Their study, published in Nature Medicine, analyzed prenatal cell-free DNA samples from over 1,800 women to identify epigenetic patterns that could indicate preeclampsia risk. These patterns were associated with abnormal blood vessel activity in the placenta.

The researchers developed a machine learning model called PEARL (preeclampsia early assessment of risk from liquid biopsy). By training the model on 450 samples and incorporating BP and BMI data, they achieved a prediction accuracy of 81% for preeclampsia cases leading to preterm birth, with a specificity of 80%. This approach could enable early identification of preeclampsia risk using existing prenatal screening samples, potentially allowing for life-saving interventions.

Source:

(2025, March 11). NIH. Predicting preeclampsia. https://www.nih.gov/news-events/nih-research-matters/predicting-preeclampsia