New blood tests boost detection of tuberculosis infection

Detecting latent Mycobacterium tuberculosis (Mtb) infection (LTBI) remains a major challenge for public health, especially among people who have been in close contact with tuberculosis patients and those with weakened immune systems. Traditional immune tests called interferon-γ release assays (IGRAs) vary in how well they spot infection in these groups. Researchers, with the corresponding author Alexandru Stoichiță, explored whether a different immune signal — the chemokine IP-10, which is produced at much higher levels than interferon-γ — could improve detection. The team carried out a cross-sectional diagnostic accuracy study in Bucharest, Romania, enrolling 99 adults: 49 people with culture-confirmed active pulmonary TB, 30 close TB contacts, and 20 individuals with autoimmune disease. They evaluated two IP-10 based tests described in the study — RIDA®QUICK TB (a lateral flow test) and RIDASCREEN® TB (an ELISA) — comparing their results with the established QuantiFERON-TB Gold Plus (QFT-Plus) assay and with bacteriological culture when available. The study focused on whether IP-10 assays could serve as reliable tools for identifying current or latent Mtb infection where IGRAs have known limitations.

Every participant in the study was tested with RIDA®QUICK TB, RIDASCREEN® TB and QuantiFERON-TB Gold Plus (QFT-Plus). When test results were indeterminate, the researchers reclassified them using a composite reference standard so that comparisons could be made consistently. Looking at the 49 people with culture-confirmed active pulmonary TB, RIDA®QUICK TB showed a sensitivity of 85.7% (95% CI: 72.8–94.1) and a positive predictive value (PPV) of 97.7%. RIDASCREEN® TB showed a sensitivity of 91.8% (95% CI: 80.4–97.7) and a PPV of 97.8%. Because the study population included very few true-negative individuals, specificity and negative predictive value (NPV) could not be reliably estimated from these data. Agreement between the IP-10 assays and QFT-Plus ranged from moderate to good (κ=0.47–0.93). When the team used QFT-Plus as a comparator in receiver operating characteristic analyses, the assays showed AUCs of 0.828 for RIDA®QUICK TB and 0.767 for RIDASCREEN® TB, indicating good immunological discrimination relative to QFT-Plus but not necessarily proving diagnostic accuracy against confirmed infection.

The findings indicate that IP-10 based assays performed well at identifying active, culture-confirmed TB in this study and had higher sensitivity than QFT-Plus, with very high positive predictive values against the bacteriological standard. Because IP-10 is produced at substantially higher concentrations than interferon-γ, these assays may be easier to read or more robust in settings where immune signals are weak or variable. The study supports using RIDA®QUICK TB and RIDASCREEN® TB as complementary tools to existing tests for detecting LTBI, particularly in contacts and people with immune-suppressing conditions who can be missed by IGRAs. However, the near-absence of true-negative individuals in this sample means that specificity and negative predictive value remain uncertain. The authors note that larger and more diverse studies are needed to define how these tests would perform in routine practice and how best to integrate them into TB screening programs alongside QuantiFERON-TB Gold Plus (QFT-Plus) and bacteriological methods.