Diagnostic accuracy and predictive value of the QuantiFERON-TB gold plus assay for tuberculosis in immunocompromised individuals: a prospective TBnet study

Background: In low tuberculosis (TB)-endemic countries, tuberculosis preventive therapy (TPT) is recommended for immunocompromised individuals with a positive immunodiagnostic test. This study aimed to assess the performance of the QuantiFERON-TB Gold Plus (QFT+) assay and predictive power for future tuberculosis in immunocompromised individuals. Methods: In this prospective observational study, immunocompromised adults ≥18 years of age including people living with HIV (PLHIV), chronic renal failure, rheumatoid arthritis, solid-organ transplantation or stem-cell transplantation, and immunocompetent adults with and without TB-disease were recruited at 21 sites in 11 European countries and tested with the QFT+ assay. Individuals without TB-disease were followed up for the development of tuberculosis. TB incidence rates (IR) were calculated, stratified by QFT+ results and acceptance of TPT. This study is registered with Clinicaltrials.gov, NCT02639936. Findings: A total of 2663 individuals (1115 female, 1548 male) were enrolled from 03/11/2015 to 29/03/2019. Persons without tuberculosis were followed up for at least two years. Among 1758 immunocompromised individuals without active tuberculosis, 13.6% had positive QFT+ results. Sensitivity and specificity for TB-disease were 70.0% (52.1–83.3%) and 91.4% (89.6–92.9%), respectively, in immunocompromised, and 81.4% (76.6–85.3%) and 96.0% (92.5–97.9%), respectively, in immunocompetent individuals. During 2457 cumulative years of follow-up among 932 individuals with chronic renal failure, rheumatoid arthritis, solid-organ transplantation or stem-cell transplantation, including 83 persons with a positive QFT+ test without TPT, no-one developed active tuberculosis. In contrast, among 642 PLHIV without TPT, one with an indeterminate QFT+ and 3/30 individuals with a positive QFT+ developed active tuberculosis; all had detectable HIV-replication and low CD4 T-cell counts (incidence 4.1 (95% CI (1.3–12.4) per 100 person-years). No individuals receiving TPT developed active tuberculosis during 269 years of follow-up. Interpretation: In immunocompromised individuals in low TB-endemic countries, the 2-year-risk for active tuberculosis was highest among PLHIV with detectable HIV-replication and low CD4-counts. In this study, the QFT+ assay did not strongly predict progression to active tuberculosis, which emphasises the need to incorporate additional risk factors. Funding: None.