Evaluation of the mRNA release assay for diagnosis of TB in HIV-infected individuals.

HIV-infected individuals are susceptible to () infection and are at high risk of developing active tuberculosis (TB). Interferon-gamma release assays (IGRAs) are auxiliary tools in the diagnosis of TB. However, the performance of IGRAs in HIV-infected individuals is suboptimal, which limits clinical application. Interferon-inducible protein 10 (IP-10) is an alternative biomarker for identifying infection due to its high expression after stimulation with antigens. However, whether mRNA constitutes a target for the diagnosis of TB in HIV-infected individuals is unknown. Thus, we prospectively enrolled HIV-infected patients with suspected active TB from five hospitals between May 2021 and May 2022, and performed the IGRA test (QFT-GIT) alongside the mRNA release assay on peripheral blood. Of the 216 participants, 152 TB patients and 48 non-TB patients with a conclusive diagnosis were included in the final analysis. The number of indeterminate results of mRNA release assay (13/200, 6.5%) was significantly lower than that of the QFT-GIT test (42/200, 21.0%) ( = 0.000026). mRNA release assay had a sensitivity of 65.3% (95%CI 55.9% - 73.8%) and a specificity of 74.2% (95%CI 55.4% - 88.1%), respectively; while the QFT-GIT test had a sensitivity of 43.2% (95%CI 34.1% - 52.7%) and a specificity of 87.1% (95%CI 70.2% - 96.4%), respectively. The sensitivity of the mRNA release assay was significantly higher than that of QFT-GIT test ( = 0.00062), while no significant difference was detected between the specificities of these two tests ( = 0.198). The mRNA release assay showed a lower dependence on CD4 T cells than that of QFT-GIT test. This was evidenced by the fact that the QFT-GIT test had a higher number of indeterminate results and a lower sensitivity when the CD4 T cells counts were decreased ( < 0.05), while no significant difference in the number of indeterminate results and sensitivity were observed for the mRNA release assay among HIV-infected individuals with varied CD4T cells counts ( > 0.05). Therefore, our study suggested that specific mRNA is a better biomarker for diagnosis of TB in HIV-infected individuals.