Machine learning-driven identification of the gene-expression signature associated with a persistent multiple organ dysfunction trajectory in critical illness.

Critical illness Multiple organ dysfunction syndrome Precision medicine Sepsis Supervised machine learning

Journal

EBioMedicine
ISSN: 2352-3964
Titre abrégé: EBioMedicine
Pays: Netherlands
ID NLM: 101647039

Informations de publication

Date de publication:
23 Dec 2023
Historique:
received: 20 09 2023
revised: 08 12 2023
accepted: 12 12 2023
medline: 25 12 2023
pubmed: 25 12 2023
entrez: 24 12 2023
Statut: aheadofprint

Résumé

Multiple organ dysfunction syndrome (MODS) disproportionately drives morbidity and mortality among critically ill patients. However, we lack a comprehensive understanding of its pathobiology. Identification of genes associated with a persistent MODS trajectory may shed light on underlying biology and allow for accurate prediction of those at-risk. Secondary analyses of publicly available gene-expression datasets. Supervised machine learning (ML) was used to identify a parsimonious set of genes associated with a persistent MODS trajectory in a training set of pediatric septic shock. We optimized model parameters and tested risk-prediction capabilities in independent validation and test datasets, respectively. We compared model performance relative to an established gene-set predictive of sepsis mortality. Patients with a persistent MODS trajectory had 568 differentially expressed genes and characterized by a dysregulated innate immune response. Supervised ML identified 111 genes associated with the outcome of interest on repeated cross-validation, with an AUROC of 0.87 (95% CI: 0.85-0.88) in the training set. The optimized model, limited to 20 genes, achieved AUROCs ranging from 0.74 to 0.79 in the validation and test sets to predict those with persistent MODS, regardless of host age and cause of organ dysfunction. Our classifier demonstrated reproducibility in identifying those with persistent MODS in comparison with a published gene-set predictive of sepsis mortality. We demonstrate the utility of supervised ML driven identification of the genes associated with persistent MODS. Pending validation in enriched cohorts with a high burden of organ dysfunction, such an approach may inform targeted delivery of interventions among at-risk patients. H.R.W.'s NIHR35GM126943 award supported the work detailed in this manuscript. Upon his death, the award was transferred to M.N.A. M.R.A., N.S.P, and R.K were supported by NIHR21GM151703. R.K. was supported by R01GM139967.

Sections du résumé

BACKGROUND BACKGROUND
Multiple organ dysfunction syndrome (MODS) disproportionately drives morbidity and mortality among critically ill patients. However, we lack a comprehensive understanding of its pathobiology. Identification of genes associated with a persistent MODS trajectory may shed light on underlying biology and allow for accurate prediction of those at-risk.
METHODS METHODS
Secondary analyses of publicly available gene-expression datasets. Supervised machine learning (ML) was used to identify a parsimonious set of genes associated with a persistent MODS trajectory in a training set of pediatric septic shock. We optimized model parameters and tested risk-prediction capabilities in independent validation and test datasets, respectively. We compared model performance relative to an established gene-set predictive of sepsis mortality.
FINDINGS RESULTS
Patients with a persistent MODS trajectory had 568 differentially expressed genes and characterized by a dysregulated innate immune response. Supervised ML identified 111 genes associated with the outcome of interest on repeated cross-validation, with an AUROC of 0.87 (95% CI: 0.85-0.88) in the training set. The optimized model, limited to 20 genes, achieved AUROCs ranging from 0.74 to 0.79 in the validation and test sets to predict those with persistent MODS, regardless of host age and cause of organ dysfunction. Our classifier demonstrated reproducibility in identifying those with persistent MODS in comparison with a published gene-set predictive of sepsis mortality.
INTERPRETATION CONCLUSIONS
We demonstrate the utility of supervised ML driven identification of the genes associated with persistent MODS. Pending validation in enriched cohorts with a high burden of organ dysfunction, such an approach may inform targeted delivery of interventions among at-risk patients.
FUNDING BACKGROUND
H.R.W.'s NIHR35GM126943 award supported the work detailed in this manuscript. Upon his death, the award was transferred to M.N.A. M.R.A., N.S.P, and R.K were supported by NIHR21GM151703. R.K. was supported by R01GM139967.

Identifiants

DOI: 10.1016/j.ebiom.2023.104938 PMID: 38142638
pubmed: 38142638
pii: S2352-3964(23)00504-2
doi: 10.1016/j.ebiom.2023.104938
pii:
doi:

Types de publication

Journal Article

Langues

eng

Sous-ensembles de citation

IM

Pagination

104938

Investigateurs

Natalie Z Cvijanovich (NZ)
Julie C Fitzgerald (JC)
Scott L Weiss (SL)
Michael T Bigham (MT)
Parag N Jain (PN)
Adam J Schwarz (AJ)
Riad Lutfi (R)
Jeffrey Nowak (J)
Geoffrey L Allen (GL)
Neal J Thomas (NJ)
Jocelyn R Grunwell (JR)
Torrey Baines (T)
Michael Quasney (M)
Bereketeab Haileselassie (B)
Chris J Lindsell (CJ)

Informations de copyright

Copyright © 2023 The Author(s). Published by Elsevier B.V. All rights reserved.

Déclaration de conflit d'intérêts

Declaration of interests Conflict of interest: M.R.A, S.B, R.K, and Cincinnati Children's Hospital Medical Center hold a provisional patent for the work detailed in this manuscript. M.W.H received funding through the NIH, received royalties and honoraria, served on DSMB, and received study drug for clinical trials, all unrelated to the current manuscript. N.S.P has stocks in Celldom, Saccharo and Allyx, and has received grants from NIH. The remaining authors have no conflict of interests to disclose.

Auteurs

Mihir R Atreya (MR)

Division of Critical Care Medicine, Cincinnati Children's Hospital Medical Center and Cincinnati Children's Research Foundation, Cincinnati, 45229, OH, USA; Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, 45267, USA. Electronic address: Mihir.Atreya@cchmc.org.

Shayantan Banerjee (S)

Department of Biotechnology, Bhupat and Jyoti Mehta School of Biosciences, Indian Institute of Technology Madras, Chennai, 600 036, India.

Andrew J Lautz (AJ)

Division of Critical Care Medicine, Cincinnati Children's Hospital Medical Center and Cincinnati Children's Research Foundation, Cincinnati, 45229, OH, USA; Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, 45267, USA.

Matthew N Alder (MN)

Division of Critical Care Medicine, Cincinnati Children's Hospital Medical Center and Cincinnati Children's Research Foundation, Cincinnati, 45229, OH, USA; Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, 45267, USA.

Brian M Varisco (BM)

Division of Critical Care Medicine, Cincinnati Children's Hospital Medical Center and Cincinnati Children's Research Foundation, Cincinnati, 45229, OH, USA; Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, 45267, USA.

Hector R Wong (HR)

Division of Critical Care Medicine, Cincinnati Children's Hospital Medical Center and Cincinnati Children's Research Foundation, Cincinnati, 45229, OH, USA; Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, 45267, USA.

Jennifer A Muszynski (JA)

Division of Critical Care Medicine, Nationwide Children's Hospital, Columbus, 43205, OH, USA; Department of Pediatrics, Ohio State University, Columbus, 43205, OH, USA.

Mark W Hall (MW)

Division of Critical Care Medicine, Nationwide Children's Hospital, Columbus, 43205, OH, USA; Department of Pediatrics, Ohio State University, Columbus, 43205, OH, USA.

L Nelson Sanchez-Pinto (LN)

Department of Pediatrics, Northwestern University Feinberg School of Medicine, Chicago, 60611, IL, USA; Department of Health and Biomedical Informatics, Northwestern University Feinberg School of Medicine, Chicago, 60611, IL, USA.

Rishikesan Kamaleswaran (R)

Department of Biomedical Informatics, Emory University School of Medicine, Atlanta, 30322, GA, United States; Department of Biomedical Engineering, Georgia Institute of Technology, Atlanta, 30322, GA, United States.

Classifications MeSH