Laleye, F.; Jackson, R.; Mi, J. D.; Shennan, A. H.; Hezelgrave - Elliott, N.; Vigilante, A.; Camunas-Soler, J.; Rasmussen, M.; Tribe, R. M.

Background Spontaneous preterm birth (sPTB) is a significant adverse outcome of pregnancy. Being able to identify and improve the management of those who may be at risk requires robust screening methods. The use of circulating molecular markers provides a promising and non-invasive solution to this problem to allow necessary and successful intervention. The role of inflammation has been continuously demonstrated to play a key role in the onset of sPTB with intrauterine inflammation being a key driver. Here we sought out to explore the inflammatory proteome using a nested case-control approach using samples from pregnant participants in the INSIGHT cohort. Objectives To explore the maternal blood proteome in the second trimester using the Olink Explore panel to identify inflammatory proteins associated with sPTB and assess their predictive value, both independently and in combination with cell-free RNA (cfRNA). Study Design We conducted a nested case-control study to investigate inflammatory protein profiles during the second trimester of pregnancy. A total of 138 maternal blood plasma samples were analyzed using a targeted proteomic assay quantifying 384 inflammation-related proteins. Differential expression analysis and a LASSO-logistic regression model with Leave-One-Out Cross-Validation (LOOCV) were applied to evaluate the association between inflammatory biomarkers and spontaneous preterm birth (sPTB) outcomes. Results Using predictive modelling of the maternal blood proteome, 16 inflammation-related proteins were identified as key discriminators of sPTB risk, with proteins such as PGF, COL9A1, CST7, CXCL6 and GALNT3 emerging as major contributors for predicting sPTB risk. Using inflammation-related maternal proteins alone to predict sPTB (<35 weeks) achieved an area under the receiver operating characteristic curve (AUC-ROC) of 0.76 (95% CI: 0.66-0.84). The incorporation of both cfRNA and proteomic data into an integrated model, improved the area under the curve to 0.85 (95% CI: 0.78-0.92). The integrated model highlighted inflammatory biomarkers that are not only implicated in preterm birth but also in essential physiological mechanisms such as placental function, tissue remodelling, and extracellular matrix composition, which are critical to maintaining pregnancy and preventing premature labour. Conclusions: These findings demonstrate that an integrated approach using both cfRNA and proteomic signatures of the second trimester maternal blood plasma yields a more comprehensive biomarker profile for predicting preterm birth risk. This multimodal strategy not only enhances the predictive accuracy but also captures a broader array of biological signals across multiple organ systems. Compared to relying solely on inflammatory proteome markers, this multiomic method offers a deeper molecular characterisation of preterm birth risk in the maternal blood plasma.