A supervised, biologically informed Innate Immunity Score reflecting immune activation shows prognostic value and potential for treatment stratification in sepsis

Background Most sepsis subtyping strategies have been developed using unsupervised methods and were not specifically designed to quantify immune activation. Moreover, they often rely on categorical classifications, while a continuous representation may better capture sepsis immunobiology. Methods By integrating plasma cytokine response data with monocyte RNA sequencing profiles of 110 healthy volunteers who were intravenously challenged with 1 ng/kg lipopolysaccharide, we derived a continuous Innate Immunity Score (IIS), which we applied to five independent clinical cohorts of patients with sepsis ( n = 1267). We intersected the IIS with unsupervised established subtyping strategies (Mars, SRS, CTS, ARDS hyper/hypoinflammatory) and continuous sepsis scores (SRSq, lymphoid, and myeloid dysregulation scores). Furthermore, we performed an in-silico drug prediction analysis to explore potential therapeutics that reverse the transcriptomic signature associated with a high IIS, and explored heterogeneity of treatment effect (HTE) in sepsis for hydrocortisone using the IIS. A parsimonious classifier model was developed to support potential clinical applicability. Results The IIS was calculated based on the weighted expression of 46 genes that discriminated best between a high versus low plasma cytokine response to LPS administration in healthy volunteers. The IIS was associated with markers of immune activation, disease severity, organ dysfunction, and mortality in patients with sepsis. Several molecular pathways were uniquely upregulated (e.g. innate immune system) or downregulated (e.g. downstream Tcr signaling, generation of second messenger molecules, regulation of T-cell activation by CD28 family) in patients with a high IIS compared with established hyperinflammatory subtypes. A range of candidate drugs were predicted to reverse the transcriptional state associated with a high IIS, including corticosteroids, methotrexate, tyrosine kinase and mTOR inhibitors, and simvastatin. HTE was observed for hydrocortisone, suggesting harm for patients with a low IIS (≤ 0.66, 69% of patients) and benefit for those with a high IIS (> 0.66, 31%, p = 0.039). HTE could not be established using the SRSq ( p = 0.660), myeloid ( p = 0.648), or lymphoid ( p = 0.385) dysregulation scores. Finally, the three-gen​e ratio log 2 [(​​ GPR84 , PIM1) : SLC27A1 ​] accurately determined the IIS ( r = 0.91, p < 2.2 × 10 − 16 ; RMSE = 0.07). Conclusions A supervised, biologically informed continuous stratification approach reflecting immune activation in sepsis shows prognostic value and potential for treatment stratification. Therefore, it may provide a framework for exploring targeted immunomodulatory strategies. Trial registration CMO Arnhem-Nijmegen; reference no. NL68166.091.18 (registration date: 04-02-2019). Registered at Overview of Medical Research in the Netherlands (OMON) reference no. NL-OMON48466. Registered at ClinicalTrials.gov; reference no. NCT06801873.

Authors

Institutions

Publication Details

Journal
Critical Care
Published
2026-11-18
DOI
https://doi.org/10.1186/s13054-026-06467-y
Primary Topic
Sepsis Diagnosis and Treatment
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

A supervised, biologically informed Innate Immunity Score reflecting immune activation shows prognostic value and potential for treatment stratification in sepsis

Manu Shankar‐Hari, Jarne Koolen, Wouter A. van der Heijden, Lise J Estcourt et al.
4 links
Critical Care
Sepsis Diagnosis and Treatment
article

A supervised, biologically informed Innate Immunity Score reflecting immune activation shows prognostic value and potential for treatment stratification in sepsis

Manu Shankar‐Hari, Jarne Koolen, Wouter A. van der Heijden, Lise J Estcourt, Peter Pickkers, Rombout B. E. van Amstel, Evangelos J. Giamarellos-Bourboulis, Aron Jansen, Dirk P. T. van Lier, Jelle Gerretsen, Tom van der Poll, Mihai G. Netea, Niklas Bruse, Lonneke A. van Vught, Matthijs Kox, Nicole J. B. Waalders, Brendon P. Scicluna, Olaf L. Cremer, Renske Hoekstra, Bram Snoek, Wilson F. Käyser-Abdo
article en

Abstract

Background Most sepsis subtyping strategies have been developed using unsupervised methods and were not specifically designed to quantify immune activation. Moreover, they often rely on categorical classifications, while a continuous representation may better capture sepsis immunobiology. Methods By integrating plasma cytokine response data with monocyte RNA sequencing profiles of 110 healthy volunteers who were intravenously challenged with 1 ng/kg lipopolysaccharide, we derived a continuous Innate Immunity Score (IIS), which we applied to five independent clinical cohorts of patients with sepsis ( n = 1267). We intersected the IIS with unsupervised established subtyping strategies (Mars, SRS, CTS, ARDS hyper/hypoinflammatory) and continuous sepsis scores (SRSq, lymphoid, and myeloid dysregulation scores). Furthermore, we performed an in-silico drug prediction analysis to explore potential therapeutics that reverse the transcriptomic signature associated with a high IIS, and explored heterogeneity of treatment effect (HTE) in sepsis for hydrocortisone using the IIS. A parsimonious classifier model was developed to support potential clinical applicability. Results The IIS was calculated based on the weighted expression of 46 genes that discriminated best between a high versus low plasma cytokine response to LPS administration in healthy volunteers. The IIS was associated with markers of immune activation, disease severity, organ dysfunction, and mortality in patients with sepsis. Several molecular pathways were uniquely upregulated (e.g. innate immune system) or downregulated (e.g. downstream Tcr signaling, generation of second messenger molecules, regulation of T-cell activation by CD28 family) in patients with a high IIS compared with established hyperinflammatory subtypes. A range of candidate drugs were predicted to reverse the transcriptional state associated with a high IIS, including corticosteroids, methotrexate, tyrosine kinase and mTOR inhibitors, and simvastatin. HTE was observed for hydrocortisone, suggesting harm for patients with a low IIS (≤ 0.66, 69% of patients) and benefit for those with a high IIS (> 0.66, 31%, p = 0.039). HTE could not be established using the SRSq ( p = 0.660), myeloid ( p = 0.648), or lymphoid ( p = 0.385) dysregulation scores. Finally, the three-gen​e ratio log 2 [(​​ GPR84 , PIM1) : SLC27A1 ​] accurately determined the IIS ( r = 0.91, p < 2.2 × 10 − 16 ; RMSE = 0.07). Conclusions A supervised, biologically informed continuous stratification approach reflecting immune activation in sepsis shows prognostic value and potential for treatment stratification. Therefore, it may provide a framework for exploring targeted immunomodulatory strategies. Trial registration CMO Arnhem-Nijmegen; reference no. NL68166.091.18 (registration date: 04-02-2019). Registered at Overview of Medical Research in the Netherlands (OMON) reference no. NL-OMON48466. Registered at ClinicalTrials.gov; reference no. NCT06801873.

Critical Care
NHS Blood and Transplant (GB), NHS Lothian (GB), University of Bonn (DE), King's College London (GB), Utrecht University (NL), University of Malta (MT), National and Kapodistrian University of Athens (GR), Leiden University Medical Center (NL), Mater Dei Hospital (MT), Edinburgh Royal Infirmary (GB), Amsterdam UMC Location University of Amsterdam (NL), Radboud University Medical Center (NL), University Medical Center Utrecht (NL), University of Oxford (GB), Centre for Inflammation Research (GB), Amsterdam University Medical Centers (NL), University of Amsterdam (NL), University of Edinburgh (GB)
ZonMw
Good health and well-being
Sepsis Diagnosis and Treatment
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon