Background:
Tuberculosis (TB) continues to cause significant morbidity and mortality globally, yet current diagnostic tools like interferon-gamma release assays (IGRAs) are unable to adequately predict which infected individuals will progress to active disease. Identifying precise biomarkers is essential to enhance predictive power and allow for targeted interventions to control transmission. This study aimed to identify biomarkers in IGRA supernatants that predict progression to TB disease among close contacts in Brazil and India.
Abstract:
- Methods: A nested case-control study (1:2) was performed within the RePORT-Brazil cohort, matching 20 QuantiFERON-positive progressors with 40 nonprogressors. Twenty-nine cytokines were measured using a Luminex assay in baseline QuantiFERON-TB Gold Plus supernatants and evaluated using machine learning (random forest) for TB prediction. Findings were validated in an independent Indian cohort (8 progressors and 12 nonprogressors).
- Results: At baseline, progressors exhibited higher levels of IFN-$\gamma$, IL-13, and IL-2, while CCL3 levels were lower. A combined immune signature of IL-8, IL-10, and CCL3 predicted incident TB within two years with an area under the curve (AUC) of 0.75 in Brazil and 0.80 in India. This model achieved sensitivity and specificity above 80% in both cohorts, meeting World Health Organization target product profile goals.
- Conclusion: A baseline 3-cytokine immune signature (IL-8, IL-10, and CCL3) derived from IGRA supernatants can accurately predict TB disease progression in close contacts across distinct geographic settings.
Keywords: tuberculosis; cytokine; biomarker prediction; QuantiFERON; multicentric study; Brazil; India.
Clique aqui
- Data de Publicação: 07/01/2025
- Autores: Betânia M. F. Nogueira,1,2,3,4, Francys Rangel,3,5,6, Alice M. S. Andrade,3,5,6 Evangeline Ann Daniel,7,8, Marina C. Figueiredo,9 Cody Staats,9, Valeria C. Rolla,10 Afrânio L. Kritski,11 Marcelo Cordeiro-Santos,12,13,14, Amita Gupta,15, Luke Elizabeth Hanna,7 Timothy R. Sterling,9 Mariana Araújo-Pereira,3,5,6,a, and Bruno B. Andrade1,3,4,5,6,a, ; for the RePORT Brazil and RePORT India Consortia
