Neutrophil Dysregulation Differentiates Pediatric Septic Shock Biomarker Based Mortality Risk Strata: Insights from Weighted Gene Co-expression Network and Transcriptomic Analyses

Leland Dunwoodie¹Min Huang²Andrew R. Moore^3,4Natalja L. Stanski^1,5Stephen W. Standage^1,5Jennifer Melissa Kaplan^1,5Basilia Zingarelli^1,5Kelli Harmon¹Julie Fitzgerald⁶Scott L Weiss⁷Michael T. Bigham⁸Adam J. Schwarz⁹Riad Lutfi¹⁰Neal James Thomas¹¹Bereketeab Haileselassie¹²Parag Jain¹³Timothy E. Sweeney¹⁴Rishikesan Kamaleswaran¹⁵Mihir R Atreya^1,5Andrew John Lautz^1,5*

• ¹Division of Critical Care Medicine, Cincinnati Children's Hospital Medical Center, Cincinnati, United States
• ²Department of Biomedical Informatics, Emory University School of Medicine, Atlanta, United States
• ³Stanford Institute for Immunity Transplantation and Infection, Stanford, United States
• ⁴Center for Biomedical Informatics Research, Stanford University School of Medicine, Stanford, United States
• ⁵Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, United States
• ⁶The Children's Hospital of Philadelphia, Philadelphia, United States
• ⁷Nemours Children's Hospital Delaware, Wilmington, United States
• ⁸Akron Children's Hospital, Akron, United States
• ⁹Children's Hospital of Orange County, Orange, United States
• ¹⁰Riley Hospital for Children at Indiana University Health, Indianapolis, United States
• ¹¹Penn State Health Milton S Hershey Medical Center, Hershey, United States
• ¹²Packard Children's Hospital Stanford, Palo Alto, United States
• ¹³The University of Texas Southwestern Medical Center, Dallas, United States
• ¹⁴Inflammatix Inc, Burlingame, United States
• ¹⁵Duke University School of Medicine, Durham, United States

ABSTRACT: Background: Pediatric sepsis is a leading cause of global mortality, particularly among children, with limited therapeutic options beyond antibiotics and organ support. The Pediatric Sepsis Biomarker Risk Model (PERSEVERE-II) stratifies mortality risk in pediatric septic shock, yet the molecular mechanisms underlying high mortality risk remain incompletely understood. Methods: We analyzed whole blood transcriptomes collected from 81 children with septic shock on day 1 of meeting study criteria. Patients were stratified into high-and low-mortality risk groups according to the PERSEVERE-II biomarker risk model. Using weighted gene co-expression network analysis (WGCNA) and differential gene expression analyses, we identified molecular pathways and transcription factors (TFs) associated with mortality risk. Cell type differences were inferred using CIBERSORTx and using a reference single-cell dataset inclusive of neutrophils and their subsets. Findings: We identified distinct molecular profiles with high-risk patients displaying significant overexpression of genes related to neutrophil degranulation and innate immunity, alongside suppressed adaptive immune responses. The predominance of developing neutrophils underscored a major role of emergency granulopoiesis. Key TFs identified, including LTF, FOXM1, KLF1, and CEBPB, were linked to high-risk gene expression signatures. Our findings indicate a pathological shift toward a dysregulated neutrophil-driven hyperinflammation and adaptive immune suppressive state, which together are associated with adverse outcomes. Interpretation: Our results suggest that neutrophil dysregulation underpins the high mortality risk conferred by the PERSEVERE-II model. The identified transcriptional regulators may provide potential targets to mitigate neutrophil dysregulation and improve outcomes among high-risk patients.