A four-gene blood transcriptomic panel for immune-status tracking and outcome risk in sepsis - 17/11/26

Highlights |
• | A four-gene panel tracked immune-status trajectories in sepsis. |
• | The panel remained informative at the infection–sepsis boundary. |
• | Persistent elevation was associated with longer-term outcome risk. |
• | Single-cell analysis linked the panel to innate immune programs. |
• | CLP experiments provided supportive in vivo biological context. |
Abstract |
Background |
Sepsis is characterized by heterogeneous immune trajectories during critical illness. Single-time-point clinical assessment cannot fully capture whether the host response is resolving or remains persistently dysregulated. We aimed to develop a compact blood transcriptomic panel for immune-status tracking in sepsis and to evaluate its diagnostic, longitudinal, and outcome-related relevance.
Methods |
Three public blood transcriptomic cohorts were integrated for discovery. Candidate genes were prioritized by differential expression, weighted gene co-expression network analysis, curated immune annotation, and persistent transcriptional dysregulation relative to healthy controls on days 1, 3, and 7 in the longitudinal peripheral blood mononuclear cell cohort. Hub genes were selected using least absolute shrinkage and selection operator (LASSO), random forest, and support vector machine recursive feature elimination. The four-gene panel was evaluated in comparisons of sepsis with healthy controls and with infection without sepsis, and in longitudinal, external course-reference, and molecular-state analyses. Single-cell and pathway analyses were used for biological interpretation. Exploratory cecal ligation and puncture experiments were used as supportive in vivo context.
Results |
ARG1, UPP1, PPARG, and GPR141 defined a four-gene blood transcriptomic panel with cohort-specific logistic-model areas under the curve of 0.982 and 0.954 for sepsis versus healthy controls and sepsis versus infection without sepsis, respectively. Across days 1, 3, and 7, ARG1 and GPR141 showed lower or declining expression in survivors and persistently elevated or increasing expression in non-survivors. Clustering of the four-gene expression profiles identified two host-response states associated with 90-day mortality in the institutional cohort; single-cell analyses showed myeloid expression alongside lymphocyte expression of UPP1 and GPR141. Exploratory cecal ligation and puncture experiments showed improved survival and reduced multiorgan injury after konjac glucomannan treatment.
Conclusions |
A four-gene blood transcriptomic panel captured diagnostic, longitudinal, and outcome-associated immune-status information in sepsis. These findings support further prospective validation of compact transcriptomic panels for immune monitoring in critically ill patients with sepsis.
Le texte complet de cet article est disponible en PDF.Abbreviations : APACHE II, ARRIVE, AUC, BH, CI, CLP, DEG, FDR, GEO, GM, GO, GSVA, H&E, HLA-DR, ICU, IL, IQR, JAK, KEGG, LASSO, LDL-C, mHLA-DR, MSigDB, NF-κB, PBMC, PCA, qPCR, RNA-seq, ROC, scRNA-seq, SD, SEM, SOFA, ssGSEA, STAT, SVM-RFE, TNFα, UMAP, WGCNA
Keywords : sepsis, immune monitoring, blood transcriptomics, biomarkers, critical care.
Plan
Bienvenue sur EM-consulte, la référence des professionnels de santé.
