Maternal Plasma Proteins Associated with Birth Weight: A Longitudinal, Large Scale Proteomic Study

Area of Science:

• Proteomics
• Maternal-fetal medicine
• Biomarker discovery

Background:

• Infants born small (SGA) or large (LGA) for gestational age face increased health risks.
• Current fetal weight prediction via ultrasound is often imprecise.
• Reliable biomarkers are needed for optimal maternal and infant care.

Purpose of the Study:

• To investigate maternal plasma proteins as potential biomarkers for predicting fetal growth deviations.
• To develop predictive models for identifying large for gestational age (LGA) and small for gestational age (SGA) infants.

Main Methods:

• Maternal blood samples collected from 70 pregnant women across three trimesters.
• Quantified nearly 5000 proteins using SomaLogic platform.
• Applied machine learning (Random Forest) with cross-validation for predictive modeling.

Main Results:

• Machine learning models accurately predicted LGA infants (AUC > 0.8) using 20 proteins.
• Prediction of SGA infants was less successful.
• Identified 148 proteins with higher abundance in LGA vs. adequate for gestational age (AGA) pregnancies; only 4 proteins differed in SGA vs. AGA.

Conclusions:

• Maternal plasma proteome contains potential biomarkers for predicting LGA.
• Further research is warranted to validate these biomarkers for clinical application.