98 - A Multitargeting Aptamer-based Identification of Plasma Biomarkers for Necrotizing Enterocolitis in Premature Infants

Necrotizing enterocolitis (NEC) is a devastating intestinal disease with high morbidity and mortality in infants, specifically very low birth weight and preterm infants. Predictive plasma/serum biomarkers for NEC are desperately needed.

Objective:

Identify diagnostic plasma/serum biomarkers specific for NEC.

Design/Methods:

After IRB approval and parental consent, plasma samples were obtained from premature infants diagnosed with NEC (n=12) and compared to respective age- or self-matched controls (n=6 per group). Proteins present in the plasma at a differential abundance were detected using an aptamer based SOMAscan 1.3K serum/plasma microarray assay. Differentially abundant proteins were selected using median fold change of aptamer relative fluorescence unit (RFU) signals and significance, with log2 -1.2 to 1.2 median fold change and p< 0.05 cut-offs, respectively. Statistically significant proteins were analyzed across self- and age-matched controls using GraphPad Prism 9.3.1.

Results:

A total of 16 proteins were determined to be differentially abundant (12 decreased and 4 increased). Relative abundance of selected proteins in the self- and age-matched pairs were compared between samples and proteins using Pearson’s correlation and z-score heat-maps which showed similar trends for biomarkers relative to each other and among both self- and age-matched samples. Individual biomarkers were evaluated using receiver operating characteristics (ROC). Differentially abundant proteins were evaluated, and literature was reviewed to determine whether they play a role in NEC, intestinal development, or immune response. Select proteins with the greatest differential abundances were used to generate a multitarget linear regression model to predict NEC, where we found increased abundance of CCL16, IGHA1-IGHA2, and decreased abundance of COLEC12, IL17B, MMP13, MICA, and FTCD. Two of the highest differentially abundant proteins, CCL16, an inflammatory chemokine triggered by monocytes, IL10, and lipopolysaccharide, as well the mucosal-specific secretory IgA (heterodimer IGHA1-IGHA2), can increase during NEC likely due to microbial dysbiosis. Based on median fold change, cases had positive slopes (n=10/11 (90.1%), S= 0.084), while control samples had negative slopes (n= 9/11 (81.8%), S=-0.091).

Conclusion(s):

Potential protein biomarkers for a NEC diagnosis were identified through differential abundance in self- and age-matched groups. Biomarker efficacy was evaluated individually or combined in a multitargeting linear regression model. Future studies will be required to evaluate biomarker efficacy in a larger cohort.