Low concordance between QIAreach QuantiFERON-TB, a novel interferon-gamma release assay, and QuantiFERON-TB Gold Plus, in a population-based survey in Blantyre, Malawi

Abstract

Urgent improvements in the diagnosis and management of Mycobacterium tuberculosis infection are required to reach End TB goals. Conventional interferon-gamma release assays (IGRAs), such as QuantiFERON-TB Gold Plus (QFT-Plus), require substantial laboratory infrastructure and large blood volumes, limiting use in high-burden settings. The QIAreach QuantiFERON-TB (QIAreach QFT) was developed to overcome these challenges but has not previously been evaluated in field conditions in a low-income, high-burden country, or at scale in children. We performed a diagnostic evaluation of QIAreach QFT against QFT-Plus, in a cross-sectional IGRA survey in Blantyre, Malawi. We recruited a population-representative sample of children aged 1-4 years and adolescents and adults aged 10-40 years, from households and primary care. We calculated sensitivity, specificity, and Cohen's kappa for QIAreach QFT against QFT-Plus, and constructed Bayesian hurdle-categorical models to compare quantitative test results. A total of 1,049 participants were recruited (64%: 1-4 years; 13%: 10-19 years; and 23%: 20-40 years). More participants had a positive QIAreach QFT result (32%) compared to QFT-Plus (15%). Over half of positive QIAreach QFT results had time-to-positivity of exactly 20 min, the assay cutoff. There was minimal agreement between QFT-Plus and QIAreach QFT results (κ = 0.26), which was lowest in children aged 1-4 years (κ = 0.13). Sensitivity and specificity of QIAreach QFT relative to QFT-Plus were 62% and 74%, respectively, with poor correlation between quantitative results. The suboptimal performance of QIAreach QFT, particularly in young children, suggests that it cannot currently be recommended for wider use and that the urgent need for an accessible test of Mtb infection remains unmet.

Importance: Almost a quarter of the world's population has evidence of Mycobacterium tuberculosis (Mtb) infection. Monitoring and addressing this substantial burden of so-called "latent" tuberculosis (TB) infection will be critical to reach End TB targets. However, current interferon-gamma release assays (IGRAs) for Mtb infection are costly, and require a large volume of venous blood and significant laboratory processing, which are major barriers to their wider use in low-income countries. The novel QIAreach QuantiFERON-TB (QIAreach) assay has been designed as a more accessible alternative. We sought to evaluate it against a reference standard of QuantiFERON-TB Gold Plus, in a large cross-sectional survey in Blantyre, Malawi. To our knowledge, this is the first diagnostic evaluation of QIAreach QFT to be performed in a population-based survey in a low-income high-incidence setting, and to specifically focus on young children (a priority group for interventions targeting Mtb infection). In contrast to previous studies in other settings, we observed poor performance of QIAreach QFT, particularly in young children where there was little correlation between the novel test and the reference standard. This leads us to conclude that this test cannot be widely recommended for use in its current form; indeed manufacture is currently suspended. We believe our findings are of urgent importance to policymakers, clinicians, and researchers and underscore the importance of careful evaluation of new diagnostics in the contexts where they are intended to be used.

Keywords: IGRA; Latent TB; Mtb infection; Mycobacterium tuberculosis; QIAreach; tuberculosis.

Fig 1

The relationship between QFT-Plus TB2 interferon-γ levels (not adjusted for nil values) and QIAreach QFT time-to-positivity (TTP) in participants with a positive QIAreach QFT, by age category. Locally estimated scatterplot smoothing (LOESS) functions were used to visualize the relationship, and are plotted in gray. In panel a, the raw QIAreach QFT TTP; in panel b, this is converted to a rate (1/TTP).

Fig 2

Posterior predictions from a hurdle categorical model of QIAreach QFT result and TTP at different levels of QFT-Plus TB2 IFN-γ, by age category. At each level of TB2, the y-axis designates the percentage of participants predicted to have negative (red) or positive (blue) QIAreach QFT result, with the intensity of blue indicating TTP, from the assay cutoff at exactly 20 mins (light blue) to lower TTP indicating more strongly positive results (dark blue). TTP, time to positivity.