Uso de biomarcadores para diagnóstico e predição de tuberculose ativa

Abstract

INTRODUCTION: Although tuberculosis (TB) is a preventable and curable disease, it remains a serious public health problem, with approximately 10.8 million new cases and 1.25 million deaths in 2023 (15.5 deaths per 100,000 population). It is estimated that one-quarter of the global population is infected with Mycobacterium tuberculosis, and between 5% and 15% of these individuals will develop active disease over their lifetime. The World Health Organization (WHO) aims to reduce TB cases by 80% by 2030, but this goal faces significant challenges, such as the lack of sensitive diagnostic tests and effective tools to predict the progression from latent infection to active disease. OBJECTIVES: To review current literature on biomarkers for the diagnosis of active TB and to evaluate the performance of biomarkers present in QuantiFERON® supernatants for predicting progression to active TB. Methods: This thesis is divided into two parts: (1) a literature review on diagnostic biomarkers for TB, their advances and limitations; and (2) a nested case-control study within the prospective RePORT-Brazil cohort, which followed 1,930 close contacts of TB cases for 24 months (2015–2021). Twenty IGRA-positive contacts who progressed to active TB were matched by sex, age, and exposure time with 40 IGRA-positive contacts who did not develop the disease. Cytokine levels were measured in QuantiFERON® supernatants, and descriptive and random forest analyses were conducted to identify predictive biomarkers. The performance of the identified signature was then evaluated in an independent contact cohort from India (8 cases and 12 controls). RESULTS: The literature review highlighted important advances in identifying diagnostic biomarkers for active TB but also emphasized challenges such as biological variability, clinical diversity, underrepresentation of key populations, and lack of external validation. In the Brazilian case-control study, we identified a signature composed of IL-8, IL-10, and CCL3 with good predictive performance. The ROC curve showed an AUC of 0.75 (95% CI: 0.61–0.90), sensitivity of 80% (95% CI: 0.62–0.88), and specificity of 85% (95% CI: 0.69–1.00), p=0.001. In the Indian cohort, the same signature showed an AUC of 0.804 (95% CI: 0.67–0.94), sensitivity of 79% (95% CI: 0.57–1.00), and specificity of 85% (95% CI: 0.69–1.00), p=0.01. CONCLUSION: This thesis identifies key barriers in the discovery of diagnostic biomarkers for TB and proposes that overcoming these challenges will require scientific collaboration and technological innovation. Additionally, it presents an externally validated predictive signature that meets WHO criteria and may serve as a promising tool to forecast the progression from latent TB infection to active disease among close contacts.