Immune Biomarker Profiling Across Clinical States and Development of a SOFA-Integrated Model for 28-Day Mortality Prediction in Sepsis: A Prospective Cohort Study.

Authors

Ming Hu, Tingting Huang, Runfeng Sun, Huiyi Wu, Jiaping Wang, Xuan Huang

Published in

Infection and drug resistance. Volume 19. Pages 627303. Epub Sep 01, 2026.

Abstract

Sepsis is characterized by complex immune dysregulation involving innate immune activation, myeloid stress, and adaptive immune exhaustion. Conventional severity scores mainly reflect organ dysfunction and may not fully capture immune heterogeneity. This study aimed to develop and internally validate an immune-based model for predicting 28-day mortality in sepsis.
This prospective cohort study included 695 participants, including 263 patients with sepsis, 160 infection non-sepsis patients, 152 non-infection critically ill patients, and 120 healthy controls. Immune biomarkers, including neutrophil CD64, heparin-binding protein (HBP), membrane-bound neutrophil alkaline phosphatase (mNAP), and PD-1⁺CD4⁺ T cells, were measured within 6 hours of ICU admission. A mortality prediction model was developed exclusively in the sepsis cohort using LASSO logistic regression and multivariable logistic regression. The model was internally validated using bootstrap resampling, Model performance was assessed using discrimination, calibration, decision curve analysis, and bootstrap internal validation.
Sepsis patients exhibited significantly higher levels of CD64, HBP, mNAP, and PD-1⁺CD4⁺ T cells than infection non-sepsis patients, non-infection critically ill patients, and healthy controls. Among 263 sepsis patients, 56 died within 28 days. LASSO identified five predictors: CD64, HBP, mNAP, PD-1⁺CD4⁺ T cells, and SOFA score. The combined SOFA-integrated immune model achieved an apparent AUC of 0.930 and a bootstrap-corrected AUC of 0.918, with good calibration.
The proposed model provides an exploratory, internally validated framework for 28-day mortality risk stratification in sepsis and requires external validation before clinical application.

PMID:
42701808
Bibliographic data and abstract were imported from PubMed on 06 Sep 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.