• HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
Thursday, November 19, 2026
BIOENGINEER.ORG
No Result
View All Result
  • Login
  • HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
        • Lecturer
        • PhD Studentship
        • Postdoc
        • Research Assistant
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
  • HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
        • Lecturer
        • PhD Studentship
        • Postdoc
        • Research Assistant
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
No Result
View All Result
Bioengineer.org
No Result
View All Result
Home NEWS Science News Health

Four-Gene Panel Identifies Sepsis Immune Subgroups to Guide Immunotherapy Choices

by
November 7, 2026
in Health
Reading Time: 5 mins read
0
Four-Gene Panel Identifies Sepsis Immune Subgroups to Guide Immunotherapy Choices

Four-Gene Panel Identifies Sepsis Immune Subgroups to Guide Immunotherapy Choices

Share on FacebookShare on TwitterShare on LinkedinShare on RedditShare on Telegram

A research team from Zhongda Hospital at Southeast University has identified a four-gene panel that assesses the immune status of sepsis patients and identifies subgroups with differing survival outcomes when treated with immunomodulatory therapies. Sepsis is characterized by a dysregulated immune response to infection, involving a delicate and dynamic balance between hyperinflammation and immunosuppression. Because this biological interplay varies widely between individuals, identifying a patient’s specific immune status in real time is a critical step toward administering appropriate immunotherapies. In a study published in Annals of Intensive Care, the authors demonstrated that profiling the expression of TBX21, GNLY, PRF1, and IL2RB can categorize patients into distinct immunological endotypes that correlate with differential responses to adjuvant drugs such as hydrocortisone and thymosin.

The study sought to address a persistent clinical bottleneck in sepsis management, where reliance on single biomarkers has proven insufficient to guide targeted treatments. Single indicators currently used in clinical immunomonitoring, such as monocyte human leukocyte antigen-DR isoform, reflect only isolated pathological processes. Because sepsis involves the concurrent activation and suppression of multiple biological pathways, a multi-gene panel offers a more comprehensive representation of overall immune status. Prior stratification efforts frequently targeted general prognostic risk or disease severity rather than directly evaluating underlying immune phenotypes, or they depended on high-throughput RNA sequencing panels containing dozens of genes that are challenging to deploy rapidly in critical care settings.

To screen candidate markers, the investigators utilized public transcriptomic datasets from the Gene Expression Omnibus, specifically GSE65682 and GSE154918. Single-sample gene set enrichment analysis was performed across 760 sepsis samples to measure the enrichment levels of 22 immune signatures, dividing the cohort into high and low immune activity subtypes based on the median score. The relative proportions of immune cell subsets were evaluated using the CIBERSORT deconvolution algorithm. Weighted gene co-expression network analysis was then conducted in R to build signed co-expression networks from the top 10,000 genes by median absolute deviation, selecting a soft-thresholding power of 20 with a scale-free fit of approximately 0.92. This analysis revealed that the dark-red module exhibited the strongest association with sepsis immune status, showing a Pearson correlation coefficient of 0.54 with a significance level of P = 6e-59.

Within the immune-associated dark-red module, genes were primarily related to T-cell and natural killer cell immunity. The researchers applied a high module membership threshold greater than 0.85 and gene significance greater than 0.50 to identify seven core genes: EOMES, PRF1, FGFBP2, TBX21, IL2RB, RUNX3, and GNLY. Concurrently, a protein-protein interaction network was constructed in the STRING database using experimental and database channels with a high-confidence combined score threshold of at least 0.70. From this network, hub nodes were selected based on degree and betweenness centralities, identifying six genes: TBX21, IL2RB, GNLY, PRF1, GZMB, and GZMH. The intersection between the stringent co-expression hub genes and the protein-protein interaction hubs yielded a definitive streamlined panel of four genes: TBX21, GNLY, PRF1, and IL2RB.

The discriminatory efficacy of the four-gene panel was evaluated across four machine learning architectures, comprising generalized linear models, support vector machines, XGBoost, and random forests. Training and hyperparameter optimization were executed on a stratified 70 percent split of the GSE65682 discovery dataset using repeated fivefold cross-validation with receiver operating characteristic area under the curve metrics. When evaluated on the independent external public dataset GSE154918, consisting of 39 adult patients with sepsis or septic shock, all four algorithms maintained high discriminative power. The observed area under the curve values with DeLong 95 percent confidence intervals were 0.891 (0.764 to 0.991) for random forest, 0.905 (0.812 to 0.997) for support vector machine, 0.900 (0.810 to 0.990) for XGBoost, and 0.909 (0.809 to 1.000) for the generalized linear model. When applying pre-specified classification probability cutoffs chosen by maximizing Youden’s J on the internal test set, external specificity reached 1.000 across models, while sensitivities ranged from 0.432 to 0.818 and accuracies ranged from 0.490 to 0.837.

To investigate the clinical utility and translational feasibility of the panel, the researchers established an assay based on real-time polymerase chain reaction and evaluated an independent clinical cohort of 99 randomized, double-blind sepsis patients admitted to the Department of Critical Care Medicine at Zhongda Hospital. Consensus clustering based on the four-gene expression profiles stratified these patients into two distinct immune status endotypes, termed ISS1 and ISS2. ISS1 was characterized by low expression of the four immune-related genes, representing an immunosuppressive endotype, whereas ISS2 showed high expression of the panel. Between the two subgroups, the median age was higher in ISS2 compared to ISS1 at 76 years versus 68 years (P = 0.0358), while baseline Sequential Organ Failure Assessment scores exhibited a non-significant trend toward higher values in ISS2 at 11 versus 9 (P = 0.0517), with other demographic and baseline clinical features remaining broadly comparable.

Within the high-expression ISS2 endotype, clinical outcomes differed markedly depending on the administration of adjuvant immunotherapies. Patients within ISS2 who received hydrocortisone demonstrated greater baseline clinical severity than untreated patients, including higher body temperature (37.8 versus 37.4 degrees Celsius, P = 0.0333), higher APACHE II scores (24 versus 18, P = 0.0168), and a higher prevalence of septic shock (95.2 percent versus 69.2 percent, P = 0.0305). In frequentist logistic regression models, hydrocortisone use in the ISS2 group was associated with higher 28-day mortality (52.4 percent versus 11.5 percent; odds ratio 8.43, 95 percent confidence interval 2.12 to 43.72, P = 0.0047) and higher 90-day mortality (61.9 percent versus 11.5 percent; odds ratio 12.46, 95 percent confidence interval 3.11 to 65.72, P = 0.0009). A similar elevation in 90-day mortality was observed for ISS2 patients receiving thymosin therapy (57.1 percent versus 24.2 percent; odds ratio 4.17, 95 percent confidence interval 1.13 to 16.51, P = 0.0347), whereas these adverse survival associations were not observed within the ISS1 endotype.

Because observational imbalances in baseline severity could confound treatment comparisons within the ISS2 subgroup, the investigators conducted several sensitivity and verification analyses. In Bayesian logistic regression models applying weakly informative Student-t priors to address small-sample uncertainty, posterior estimates remained directionally consistent with the frequentist findings. Propensity score matching within the ISS2 stratum yielded six matched pairs comparing hydrocortisone-treated and untreated patients; in this matched sample of 12 patients, Firth-penalized logistic regression demonstrated an odds ratio of 47.7 (95 percent confidence interval 2.94 to 7830.01, P = 0.0041) for 90-day mortality and an odds ratio of 23.4 (95 percent confidence interval 1.59 to 3579.50, P = 0.0188) for 28-day mortality. The authors emphasized that because of the restricted sample size and potential statistical separation, the wide confidence intervals necessitate cautious interpretation rather than causal inference.

In formal Cox proportional-hazards modeling incorporating a treatment-by-endotype interaction term, hydrocortisone exhibited an ISS2-specific hazard ratio of 7.44 compared to 2.02 in ISS1, though the interaction term was underpowered and did not achieve statistical significance (Wald P = 0.109). Similarly, thymosin exhibited a hazard ratio of 2.52 in ISS2 versus 1.37 in ISS1, with a non-significant interaction term (Wald P = 0.454). The investigators also attempted to apply the four-gene panel, in combination with patient age and sex, to predict the development of multiorgan failure three days after intensive care unit admission. Although the model achieved an area under the curve exceeding 0.7, its discriminative performance for multiorgan failure was lower than for immune status, reflecting the multifactorial physiological determinants of organ failure.

The authors outlined important limitations to the investigation, noting that clinical validation was conducted in a single-center cohort of 99 patients, which limited statistical power for prespecified interaction testing and generalizability beyond similar critical care populations. Furthermore, baseline immune cell fractions were inferred through computational deconvolution rather than direct cytometric enumeration. The authors concluded that while their findings demonstrate the feasibility of using a four-gene real-time polymerase chain reaction panel to stratify sepsis immune status within a few hours, adequately powered multicenter trials with endotype-stratified randomization and prospectively designed interaction hypotheses are necessary to validate the panel before it can guide routine immunotherapy decisions in clinical practice.

Subject of Research: Sepsis immune status biomarker panel and immunotherapy stratification

Article Title: Development of a gene panel for immune status assessment in sepsis

Article References: Gao, C., Lu, X., Jiang, Y., Tang, Y., Ni, Y., Chen, H., Wu, X., Zhou, X., Yang, Y., Liu, L., Chao, J., Xie, J., & Qiu, H. (2025). Development of a gene panel for immune status assessment in sepsis. Annals of Intensive Care, 15(1), Article 170. https://doi.org/10.1186/s13613-025-01594-1

Image Credits: AI Generated

DOI: 10.1186/s13613-025-01594-1

Keywords: Sepsis, Immunotherapy, Gene Expression Panel, Endotypes, Intensive Care, Machine Learning, Biomarkers, Hydrocortisone, Development, gene, panel, immune

News Source: Juliet Wilcox. (November 7, 2026). Four-Gene Panel Identifies Sepsis Immune Subgroups to Guide Immunotherapy Choices. Scienmag.

Tags: biomarkersDevelopmentendotypesgeneGene Expression Panelhydrocortisoneimmuneimmunotherapyintensive careMachine LearningPanelsepsis
Share12Tweet7Share2ShareShareShare1

Related Posts

BLST Protein May Help Protect Spinal Discs From Age-Related Damage

BLST Protein May Help Protect Spinal Discs From Age-Related Damage

November 19, 2026
Experts Define Transdisciplinary Competency Framework for Long-Term Care Case Management

Experts Define Transdisciplinary Competency Framework for Long-Term Care Case Management

November 19, 2026

Thymic Restoration Strategy Targets Age-Related Immune System Decline

November 19, 2026

Hospitalization Rates Rise Under China Case-Based Payment Reform

November 19, 2026

POPULAR NEWS

  • Corncrake Flagship Status Does Not Guarantee Local Recognition in Rural Romania

    Corncrake Flagship Status Does Not Guarantee Local Recognition in Rural Romania

    29 shares
    Share 12 Tweet 7
  • Pre-Rollout Survey Finds Eighteen Percent HPV Vaccine Uptake Among Chinese College Women

    29 shares
    Share 12 Tweet 7
  • Military Physicians Issue Fewer Medical Exemptions as Decision Fatigue Mounts Across Workdays

    29 shares
    Share 12 Tweet 7
  • Counseling and Mobile App Increase Postpartum Modern Contraceptive Uptake in Jordan

    29 shares
    Share 12 Tweet 7

About

We bring you the latest biotechnology news from best research centers and universities around the world. Check our website.

Follow us

Recent News

BLST Protein May Help Protect Spinal Discs From Age-Related Damage

Experts Define Transdisciplinary Competency Framework for Long-Term Care Case Management

Thymic Restoration Strategy Targets Age-Related Immune System Decline

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 88 other subscribers
  • Contact Us

Bioengineer.org © Copyright 2023 All Rights Reserved.

Welcome Back!

Login to your account below

Forgotten Password?

Retrieve your password

Please enter your username or email address to reset your password.

Log In
No Result
View All Result
  • Homepages
    • Home Page 1
    • Home Page 2
  • News
  • National
  • Business
  • Health
  • Lifestyle
  • Science

Bioengineer.org © Copyright 2023 All Rights Reserved.