NEW BIOMARKERS IN SEPSIS: RECENT ADVANCES, CLINICAL READINESS, AND PERSONALIZED MEDICINE
Keywords:
Sepsis, Biological Biomarkers, Endotyping, Point-of-Care Testing, Presepsin, MR-proADM, Transcriptomics, Machine Learning, Immunoparalysis, Personalized MedicineAbstract
Background: Sepsis is a biologically complex and dynamic syndrome defined by a dysregulated host response to infection leading to organ dysfunction. Given its high clinical heterogeneity, conventional diagnostic approaches and static biomarkers often fail to capture real-time disease trajectories, leaving critical gaps in early identification, risk stratification, and targeted intervention.
Objective: To review emerging sepsis biomarkers across multiple biological domains, evaluate their dynamic prognostic and diagnostic capabilities, and assess their current clinical readiness within personalized care frameworks.
Key Domains & Findings:
- Established vs. Point-of-Care (POC) Biomarkers: Traditional markers (lactate, CRP, PCT) remain foundational for assessing hypoperfusion and guiding antibiotic de-escalation, though they lack specificity for sepsis endotypes [2, 8]. Novel POC and near-term markers (presepsin, MR-proADM, nCD64, calprotectin, PSP) offer rapid, high-accuracy signaling for early infection, severe endothelial stress, and acute clinical deterioration [4,5].
- Prognostic Kinetics & Immunophenotyping: Serial dynamics outperform static baseline measurements. Persistent elevations of cytokines such as IL-6 correlate with elevated 28-day mortality [1,9], whereas alterations in surface markers like mHLA-DR and sIL-7R identify distinct immunoparalysis phenotypes, laying the groundwork for phenotype-guided immunomodulatory therapies.
- Omics, Epigenetics & Digital Health: Next-generation tools—including organ-specific cfDNA methylation signatures for sepsis-associated AKI [10], non-coding RNAs like LncRNA HOTTIP [3], transcriptomic multigene panels [7], and targeted lipidomics—enable deep biological endotyping. When integrated with machine learning platforms, multi-omic profiles substantially improve risk prediction models compared to standard clinical scores [6].
Conclusion & Clinical Readiness: While routine markers remain the current standard, near-term biological tests (MR-proADM, presepsin, nCD64) demonstrate high diagnostic and prognostic value. High-dimensional omics and transcriptomic panels require further standardization of assays, prospective validation, and workflow integration before widespread adoption. Optimal clinical utility relies on combining rapid dynamic biomarker profiles, advanced multi-omics/AI modeling, and rigorous clinical assessment.
Downloads
References
1. Candelli, M., et al. (2025). The interleukin network in sepsis: from cytokine storm to clinical applications. Diagnostics, 15(22), 2927.
2. Dark, P., et al. (2025). Biomarker-guided antibiotic duration for hospitalized patients with suspected sepsis: the ADAPT-Sepsis randomized clinical trial. JAMA, 333(8), 682–693.
3. El-Sisi, M. G., et al. (2025). The long antisense non-coding RNA HOXA transcript at the distal tip (LncRNA HOTTIP) in health and disease. Naunyn-Schmiedeberg's Archives of Pharmacology, 398, 16537–16575.
4. Moura, E. L. B., et al. (2025). Presepsin as a diagnostic and prognostic biomarker of sepsis-associated acute kidney injury. Journal of Clinical Medicine, 14(19), 6970.
5. Piccioni, A., et al. (2025). Evaluation of presepsin for early diagnosis of sepsis in the emergency department. Journal of Clinical Medicine, 14(7), 2480.
6. Rahman, M. S., et al. (2024). Machine learning-based prognostic model for 30-day mortality prediction in Sepsis-3. BMC Medical Informatics and Decision Making, 24, 249.
7. Scicluna, B. P., et al. (2025). A consensus blood transcriptomic framework for sepsis. Nature Medicine, 31, 4119–4130.
8. Silvinato, A., et al. (2025). C-reactive protein in adult sepsis: systematic review and meta-analysis. Clinics, 81, 100848.
9. Varga, N. I., et al. (2025). IL-6 baseline values and dynamic changes in predicting sepsis mortality: a systematic review and meta-analysis. Biomolecules, 15(3), 407.
10. You, R., et al. (2024). A promising application of kidney-specific cell-free DNA methylation markers in real-time monitoring of sepsis-induced acute kidney injury. Epigenetics, 19, 2408146.
