Back

Plasma proteome profiling identifies predictive signatures for preterm birth risk

Kharbanda, N.; Biswas, A.; Thiruvengadam, R.; Bai, S.; Haldar, S. S.; Tiwari, A.; Saha, S.; Sharma, N.; Desiraju, B. K.; Kshetrapal, P.; Wadhwa, N.; Salunke, D. M.; Bhatnagar, S.; Maiti, T. K.; GARBH-Ini study group,

2025-09-29 obstetrics and gynecology
10.1101/2025.09.24.25336553 medRxiv
Show abstract

Preterm birth (PTB) is a major public health concern, and its associated complications account for 16.6% of deaths in children under five years. Over 50% of such deliveries are spontaneous, with unknown underlying causes. We identified significant changes in the plasma proteome across mid-trimester prior to the pre-term deliveries, indicating signals for identifying mothers at risk for PTB. Using quantitative LC-MS analysis and machine learning, we identified high plasma levels of Calcyophosin-2 (CAPS2) at 18-20 weeks of gestation in women who delivered preterm. Prediction model based on plasma CAPS2 level in a case-cohort (n=795) design accurately predicted high-risk sPTB with a detection rate of over 90% while reducing 40% false positivity and therefore avoiding subsequent unnecessary tests in them. Our findings strongly highlight CAPS2 as a novel candidate biomarker for prediction at early mid-trimester, accurately screening women at risk of preterm delivery, particularly in low-resource primary and secondary care settings where there are no existing screening programs for preterm birth. TeaserMid-trimester screening of mothers at risk of delivering preterm babies using a single marker test.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.