Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Observational Study
. 2025 May 8;25(1):677.
doi: 10.1186/s12879-025-10914-6.

Peripheral blood cytokine profiles predict the severity of SARS-CoV-2 infection: an EPIC3 study analysis

Xumin Li  1   2 , Vivek Pakanati  1 , Cindy Liu  1 , Tracy Wang  1 , Daniel Morelli  1 , Anna Korpak  1 , Aaron Baraff  1 , Stuart N Isaacs  3   4 , Amy Vittor  5   6 , Kyong-Mi Chang  3   4 , Elizabeth Le  7 , Nicholas L Smith  1   2 , Jennifer S Lee  7   8 , Jennifer M Ross  1   9 , Javeed A Shah  10   11 ; EPIC3 Investigators
Collaborators, Affiliations
Observational Study

Peripheral blood cytokine profiles predict the severity of SARS-CoV-2 infection: an EPIC3 study analysis

Xumin Li et al. BMC Infect Dis. .

Abstract

Background: Predicting which patients will develop severe COVID-19 complications could improve clinical care. Peripheral blood cytokine profiles may predict the severity of SARS-CoV-2 infection, but none have been identified in US Veterans.

Methods: We analyzed peripheral blood cytokine profiles from 202 participants in the EPIC3 study, a prospective observational cohort of US Veterans tested for SARS-CoV-2 across 15 VA medical centers. Illness severity was assessed based on the highest level documented during the first 60 days after recruitment. We correlated cytokine levels with illness severity using LASSO logistic regression, random forest, and XGBoost models on a 70% training set and calculated the AUC on a 30% test set.

Results: LASSO regression identified 6 cytokines as predictors of SARS-CoV-2 severity with 77.3% AUC in the test set. Random forest and XGBoost models achieved an AUC of 80.4% and 80.7% in the test set, respectively. All models assigned a feature importance to each cytokine, with IP-10, MCP-1, and HGF consistently identified as key markers.

Conclusions: Cytokine profiles are predictive of SARS-CoV-2 severity in US Veterans and may guide tailored interventions for improved patient management.

Keywords: COVID-19; Cytokines; Illness severity; LASSO regression; Predictive biomarkers; Random forest; SARS-CoV-2; Veterans; XGBoost.

PubMed Disclaimer

Conflict of interest statement

Declarations. Ethics approval and consent to participate: All study participants provided their individual written, informed consent to participate in the EPIC3 study. Our study is approved by the VA Central Institutional Review Board (reference 20–14). Consent for publication: The authors and participants provided their consent to have this manuscript and the data it contains published. Competing interests: The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
Heatmap of correlations between 10 cytokines with individual AUC 95% CI lower bounds exceeding 0.5
Fig. 2
Fig. 2
ROC curves for the LASSO, RF, and XGBoost models
Fig. 3
Fig. 3
Feature importance of each cytokine in the risk prediction models. A LASSO model. B RF model. C XGBoost model
See this image and copyright information in PMC

References

    1. (CDC) CfDCaP. COVID Data Tracker. Accessed Aug 31, 2024. https://covid.cdc.gov/covid-data-tracker/#datatracker-home
    1. Ioannou GN, Liang PS, Locke E, et al. Cirrhosis and Severe Acute Respiratory Syndrome Coronavirus 2 Infection in US Veterans: Risk of Infection, Hospitalization, Ventilation, and Mortality. Hepatology. 2021;74(1):322–35. 10.1002/hep.31649. - PMC - PubMed
    1. Fan VS, Dominitz JA, Eastment MC, et al. Risk Factors for Testing Positive for Severe Acute Respiratory Syndrome Coronavirus 2 in a National United States Healthcare System. Clin Infect Dis. 2021;73(9):e3085–94. 10.1093/cid/ciaa1624. - PMC - PubMed
    1. Ioannou GN, Locke E, Green P, et al. Risk Factors for Hospitalization, Mechanical Ventilation, or Death Among 10 131 US Veterans With SARS-CoV-2 Infection. JAMA Netw Open. Sep 1 2020;3(9):e2022310. 10.1001/jamanetworkopen.2020.22310 - PMC - PubMed
    1. Hu B, Guo H, Zhou P, Shi ZL. Characteristics of SARS-CoV-2 and COVID-19. Nat Rev Microbiol. 2021;19(3):141–54. 10.1038/s41579-020-00459-7. - PMC - PubMed

Publication types