Pediatric Severe Sepsis Proteome Associates With Inborn Error of Immunity-Related Variant Genotype

Monday, February 24, 2025

2:30pm – 3:30pm Eastern Time

Location: Connections Central Theater 1

Kate F. Kernan, MD

Children's Hospital of Pittsburgh of UPMC

Disclosure(s): No relevant financial relationship(s) to disclose.
JV

Jerry Vockley, MD PhD

Professor
Children's Hospital of PIttsburgh

Disclosure information not submitted.
SC

Scott Canna, MD

Associate Professor
Children's Hospital of Philadelphia

Disclosure information not submitted.
Hyun Jung Park, PhD

Assistant Professor
University of Pittsburgh

Disclosure information not submitted.
DN

Daniel Notterman, MD

Professor
Princeton University

Disclosure information not submitted.
JC

Joseph Carcillo, MD

Professor
Children's Hospital of Pittsburgh of UPMC

Disclosure information not submitted.

Disclosure(s):

Kate F. Kernan, MD: No relevant financial relationship(s) to disclose.

Introduction: We previously demonstrated a 60% prevalence of variants in inborn error of immunity related genes (IEI-RV) in children with severe pediatric sepsis. While these IEI-RV associate with clinical phenotypes, it is unclear if these associations are causally related or incidental correlations. To investigate this further, we sought to identify relationships between IEI-RV and molecular phenotypes at the protein level in children with severe pediatric sepsis.

Methods: We measured 1500 plasma protein analytes related to complement, coagulation, inflammation, immune function and specific organ dysfunction In 114 children with severe pediatric sepsis using SomaLogic’s DNA-based aptamer testing. We then used k-means clustering with PCA analysis to identify unique subgroups of based upon their protein signatures. We then compared the frequency of IEI-RV in each protein signature subgroup.

Results: K-means clustering identified four distinct proteomic signatures among participants. Pathway enrichment analysis showed that these groups were related to differences in cytokine-cytokine receptor interaction (p-value = 2.1x10-21), JAK-STAT signaling (p-value = 2.5x10-11), primary Immunodeficiency (p-value = 1.3x10-8), B cell receptor signaling pathway (p-value = 2.5x10-20), and T cell receptor signaling (p-value = 7.6x10-20). Additionally, the immunogenotype differed significantly between groups, with clusters 3 and 4 having the highest frequency of children with IEI-RV (Cluster 1: 64%, Cluster 2:74%, Cluster 3: 92% and Cluster 4: 100%, p = 0.039).

Conclusions: In this study, we demonstrate a correlation between IEI-RV and the plasma proteome based on a 1500 analyte profile of markers of coagulation, inflammation, immunity and organ injury. K-means clustering demonstrating the presence of four distinct subgroups that differ in levels of plasma proteins involved in immunologic function immunogenotype.