Diagnostic Accuracy of the Three‐Gene Host‐Response Test in Blood for Tuberculosis: A Systematic Review
International Journal of Microbiology · 2026-01 · 2026 (1)
DOI: 10.1155/ijm/3402245 ↗Abstract
Purpose Molecular diagnostic tests and sputum‐based assays are common techniques for the initial evaluation of tuberculosis (TB). However, given their limitations in certain clinical contexts, efforts have been made to develop other diagnostic markers. This review evaluated the diagnostic accuracy of a three‐gene host‐response test (TB score) for the diagnosis of TB. Methods We performed systematic searches in seven databases until July 2026. We included observational studies evaluating the diagnostic accuracy of the TB score to diagnosis of TB across different clinical contexts. Risk of bias was assessed using QUADAS‐2. Meta‐analysis was performed using the logit function, and GRADE was employed to assess certainty of evidence (CoE). Results Twenty‐eight studies were included. For distinguishing pulmonary TB from other respiratory diseases, the TB score showed a pooled sensitivity and specificity of 0.80 (95% CI: 0.73–0.84) and 0.72 (95% CI: 0.64–0.78), respectively, with an AUC of 0.82. When comparing pulmonary TB to healthy controls, sensitivity was 0.75 and specificity was 0.86, with an AUC of 0.87. Performance was lower for meningeal tuberculosis (sensitivity 0.58; specificity 0.67) and for distinguishing active from latent TB (sensitivity 0.73; specificity 0.76). Risk of bias was generally low. The CoE for most outcomes was very low. Conclusion The TB score shows moderate diagnostic accuracy for pulmonary TB, with pooled sensitivity below the WHO TPP minimum for a standalone triage test. High heterogeneity and very low CoE limit confidence in these findings. Its potential role as a complementary tool remains investigational and requires validation in prospective studies.
EUVIMED