Biomarkers for Asymptomatic Tuberculosis: A Potential Breakthrough in Diagnosis

Strategic Insights -

Researchers from Capital Medical University have made a significant breakthrough in the diagnosis of asymptomatic tuberculosis (TB) by identifying plasma biomarkers that can accurately discriminate between asymptomatic TB, latent TB infection, and healthy controls. The study, published in the Journal of Proteome Research, employed the Olink Proximity Extension Assay to analyze 92 inflammation-related proteins and integrated machine learning algorithms to identify the most discriminative biomarkers. The results showed that the combination of EN-RAGE and MCP-3 could accurately discriminate asymptomatic TB from healthy controls and latent TB infection, yielding an area under the curve (AUC) of 0.90 (95% CI: 0.85-0.95).

Key Takeaways:

  • The study identified EN-RAGE and MCP-3 as potential biomarkers for asymptomatic TB, which can accurately discriminate between asymptomatic TB, latent TB infection, and healthy controls.
  • The combination of EN-RAGE and MCP-3 yielded an AUC of 0.90 (95% CI: 0.85-0.95), indicating a high degree of accuracy in diagnosis.
  • ELISA validation performed in an independent cohort confirmed significant elevations of EN-RAGE and MCP-3 in asymptomatic TB compared to healthy controls and latent TB infection.
  • The study employed the Olink Proximity Extension Assay to analyze 92 inflammation-related proteins and integrated machine learning algorithms to identify the most discriminative biomarkers.
  • The researchers from Capital Medical University demonstrated the potential of EN-RAGE and MCP-3 as biomarkers for asymptomatic TB, which may facilitate early identification and improve patient outcomes.

Statistics:

  • The combination of EN-RAGE and MCP-3 yielded an AUC of 0.90 (95% CI: 0.85-0.95) in the discovery cohort and 0.837 (95% CI: 0.75-0.924) in the independent validation cohort.
  • The study analyzed 92 inflammation-related proteins using the Olink Proximity Extension Assay.
  • The researchers employed machine learning algorithms, including SVM, random forest, neural network, and XGBoost, to identify the most discriminative biomarkers.
  • The study included a total of 300 participants, consisting of asymptomatic TB cases, healthy controls, and latent TB infection.

Sources:

  • NewsRx. Researchers at Capital Medical University Describe Findings in Tuberculosis (Discovering Biomarkers for Asymptomatic Tuberculosis via Olink Proteomics and Machine Learning). TB & Outbreaks Week. October 21, 2025; p 3189.
  • American Chemical Society. Journal of Proteome Research. 2025.