Is the New Interferon-Gamma Releasing Assay Beneficial for the Diagnosis of Latent and Active Infections in Tertiary Care Setting?

Interferon-Gamma Release Assays (IGRAs) are widely used in the laboratory diagnosis of (MTB) infections, particularly in the latent form. We compared the performance of a newly developed IGRA, the Standard E TB-Feron ELISA (TBF) with the currently used QuantiFERON-TB Gold Plus assay (QFT-Plus) for the detection of latent tuberculosis infections (LTBIs) in tertiary care settings. We also investigated interferon-gamma (IFN-γ) released by T cell subsets via intracellular cytokine staining (ICS) and flow cytometry. A total of 335 subjects including 40 patients with active tuberculosis (ATB), 75 immunocompromised patients with LTBIs (P-LTBI), 70 health care workers with LTBIs (H-LTBI), and 150 healthy controls (HC) were studied. Overall, 168 subjects (50.1%) and 178 subjects (53.1%) displayed IGRA-positive results in the QFT-Plus and TBF, respectively. The overall concordance rate was 94.0%. The sensitivity and specificity of TBF were 88% and 95%, respectively, while the sensitivity and specificity of QFT-Plus were 90% and 100%, respectively. Twenty discordant results (6.0%) were observed in simultaneously performed QFT-Plus and TBF. Particularly, 13 LTBI subjects previously positive QFT-Plus showed negative results in QFT-Plus performed after enrollment. In TBF, six subjects showed positive results while five were negatively concordant with QFT-plus and two were indeterminate. The overall proportion of IFN-γ releasing CD8+ T lymphocytes was significantly higher in TBF compared to those of QFT-Plus TB1 and TB2 (0.21% vs. 0.01% and 0.02%; -value < 0.05). The recombinant protein antigens in the TBF stimulated TB-specific CD8+ T cells more efficiently. Therefore, TBF would be a useful alternative to current IGRAs such as the QFT-Plus, particularly in tertiary care settings where the immunocompromised patients are subjected to IGRA tests to differentiate MTB infection. Further strategies to analyze the implications of the discrepancies, particularly near the cutoff values between different IGRAs, are needed.