Abstract: Despite sustained advances in high-throughput molecular profiling, early prediction of maternal–fetal disorders remains one of the most persistent challenges in perinatal medicine. Numerous biomarkers have been proposed, yet only a small fraction have translated into timely, clinically actionable tools. To clarify where progress has been made – and where it continues to stall – we conducted a Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020–compliant systematic review examining whether integrative multi-omics, particularly liquid biopsy-based approaches, can meaningfully advance early risk stratification in pregnancy. We systematically searched PubMed/MEDLINE, Embase, Web of Science, Scopus, the Cochrane Library, and relevant study registers from inception through January 2025, supplemented by manual reference screening. Eligible studies applied transcriptomic, proteomic, metabolomic, immunomic, or integrated multiomics approaches to predict preeclampsia, preterm birth, fetal growth restriction, or related adverse outcomes. Two reviewers independently screened records, extracted data, and assessed methodological robustness using design-appropriate tools. Thirty-five studies met the inclusion criteria. Across diverse cohorts and analytical platforms, maternal circulating biomarkers – most consistently cell-free RNA, plasma proteomic signatures, and metabolomic profiles – demonstrated predictive signals emerging weeks to months before clinical diagnosis, frequently within the first trimester. Models combining multiple parameters generally outperformed single-omic strategies, although apparent gains in discrimination were often tempered by overfitting risk, limited external validation, and reduced interpretability. Placental and spatial omics provided important mechanistic resolution but were largely confined to delivery-stage sampling, restricting their utility for early prediction. Recurrent barriers to clinical translation included assay complexity, cost, data governance constraints, and poor alignment with existing prenatal care workflows. Synthesizing these findings, we propose a translational readiness framework that distinguishes early predictive utility from downstream biological characterization. Our analysis suggests that meaningful clinical impact will depend less on identifying additional biomarkers than on integrating molecular parameters into interpretable, validated models that can be deployed within the narrow gestational window where prevention remains possible.
Andonotopo et al. (Thu,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: