Integrative metabolomic and proteomic signatures define clinical outcomes in severe COVID-19

Area of Science:

• Biochemistry
• Immunology
• Systems Biology

Background:

• The COVID-19 pandemic caused unprecedented global health challenges.
• Understanding the molecular underpinnings of COVID-19 is crucial for predicting patient outcomes.

Purpose of the Study:

• To uncover pathogenic complexities of COVID-19 using multi-omics analyses.
• To identify molecular signatures that predict clinical outcomes in COVID-19 patients.
• To develop a predictive model for COVID-19 disease severity.

Main Methods:

• Large-scale integrative multi-omics analyses of serum from 330 COVID-19 patients and 97 controls.
• Targeted metabolomic and proteomic profiling to assemble protein-metabolite interaction networks.
• Development and validation of a composite outcome measure based on metabolomics data.

Main Results:

• Identified distinct protein-metabolite crosstalk related to immune modulation, metabolism, vascular homeostasis, and collagen catabolism.
• Linked specific proteins and metabolites to clinical indices of long-term mortality and morbidity.
• Developed a novel metabolomics-based model predicting severe COVID-19 with high accuracy (0.83-0.93 in validation datasets).

Conclusions:

• Integrative multi-omics analysis reveals intricate molecular pathways in COVID-19.
• Specific molecular signatures can predict disease severity and long-term outcomes.
• The developed metabolomics model offers a promising tool for assessing COVID-19 severity.