Commentary on: diagnostic accuracy and predictive value of the QuantiFERON-TB Gold Plus assay for tuberculosis in immunocompromised individuals: a prospective TBnet study

Mycobacterium tuberculosis is an ancient pathogen that continues to pose a major public health threat, with >10 million cases of active TB leading to an estimated 1.25 million deaths worldwide in 2023.¹ The interferon gamma release assay (IGRA) represented a major advance in diagnostics. The QuantiFERON-TB Gold Plus (QFT+), detecting CD8 and CD4 T-cell responses to specific antigens present in M. tuberculosis but absent from the Bacillus Calmette-Guérin vaccine, offers improved diagnostic performance compared to earlier iterations.² Sester and colleagues' comprehensive European study provides new real-world evidence on QFT+ performance in low TB-incidence settings.³ This prospective study enrolled 2663 participants across 21 centres in 11 European countries, with two components: cross-sectional analysis assessing diagnostic accuracy for active TB (336 with confirmed TB, 2313 without), and longitudinal follow-up.

The diagnostic accuracy results are a stark illustration of the dangers of mission creep. When employed to diagnose active TB, a use for which it was never designed, QFT+ detected only 70% of cases in immunocompromised patients and 81% in immunocompetent individuals. These figures fall far short of what should be expected from a diagnostic test for active disease, and the authors rightly conclude that QFT+ does not meet WHO-suggested minimal requirements for tuberculosis diagnosis. These data should not surprise us and serve as real-world support for guidelines that discourage the use of IGRAs to diagnose active TB.⁴ IGRAs detect immunological memory of previous exposure; they cannot distinguish between latent infection and active disease because both generate common cellular responses.⁵ Using an IGRA to diagnose active TB is rather like using a pregnancy test to determine gestational age: technically related, but the wrong tool for the purpose.

The longitudinal findings address a second, and perhaps more interesting question: what does a positive QFT+ result mean in low TB-incidence settings? In low TB-incidence countries, how likely is a positive QFT+ result to represent true, latent infection? The data suggest that our instincts may be overestimating the positive predictive value. Among 932 non-HIV immunocompromised patients followed for 2457 person-years (including 83 with positive QFT+ who didn't receive latent TB treatment) exactly zero developed active TB. This isn't test failure; it is a statistical expectation. When underlying prevalence is low, even a specific test will generate more false positives than true ones.

The incident tuberculosis cases validate this interpretation. All active TB cases among immunocompromised patients occurred in people living with HIV with detectable viral loads and low CD4 counts. Among this group with positive QFT+, progression rates reached 4.1 per 100 person-years, a 44-fold relative increase. Additionally, these cases were born in medium- or high-incidence TB settings. When prior probability is genuinely elevated, QFT+ offers meaningful predictive value.

These findings suggest re-evaluating TB prevention strategies in low-incidence settings. Current WHO guidelines recommend preventative therapy for people with positive IGRA living with HIV, receiving dialysis, initiating immunosuppression, or preparing for transplantation.⁶ This study suggests these groups have substantial differences in progression risk, requiring more nuanced approaches.

Crucially, the investigators recorded real-world decision-making around prescribing TB preventative therapy, providing vital insight into diagnostic testing's impact on clinical practice. Perhaps a telling observation is that only 15% of QFT+ positive immunocompromised patients received recommended preventive therapy. This contrasts with guidelines, but is this guideline indifference, or intelligent decision making? European physicians may intuitively understand what the data confirm: in low-incidence settings, many positive QFT+ results don't represent actionable latent tuberculosis. This likely reflects that immunoreactivity may persist even once infection has been cleared.⁷ The 100% protective effect of preventive therapy observed shouldn't discourage evidence re-assessment. Some people who would never have developed active TB may be undergoing treatment, representing not just inefficient resource use, but unnecessary drug exposure and potential harm.

Important caveats remain. This study was inevitably underpowered for rare outcomes in subgroup analyses and may not generalise across all European regions or different immunocompromised populations. Nevertheless, the core message remains robust: test interpretation cannot be divorced from epidemiological context.

IGRA remains an excellent tool for its intended purpose, but its utility depends on understanding its limitations and populations being tested. In the precision medicine era, our approach to TB prevention must be equally precise, guided by genuine risk rather than reflexive adherence to one-size-fits-all guidelines. Given that most active TB in low-incidence settings results from latent TB reactivation,⁵ improved detection of people who would derive maximal benefit from preventive therapy would be a major advance.⁸ With migration from higher TB-incidence settings, the medical community must address this challenge to deliver high-quality care to diverse populations.

Contributors

Both authors contributed equally to this commentary.

Declaration of interests

None declared.