Avaliação da assinatura transcricional de 3 genes do hospedeiro como ferramenta de diagnóstico e prognóstico para tuberculose ativa em populações de alto risco

Abstract

Introduction: The World Health Organization (WHO) has identified the need for development of a non-sputum-free triage test for tuberculosis (TB) among high-risk populations. The aim of this thesis was to investigate the performance of the host 3-gene transcriptional signature as a blood biomarker of diagnostic and prognostic ability for active TB in high-risk populations. Methods: Two diagnostic study within a prospective were carried out. Study I: We validated the 3 genes signature (GBP5 [Guanylate Binding Protein 5], DUSP3 [Dual Specificity Protein Phosphatase 3] and KLF2 [Kruppel-like factor 2]) in three independent cohorts: (cohort 1) adolescents from a community rural South Africa with infection latent Mycobacterium tuberculosis (ILTB); (cohort 2) systemic screening for active TB in persons deprived of liberty from two prisons in Brazil; and (cohort 3) monitoring response to treatment (1, 4, and 24 weeks) for adults with positive culture for TB seen in primary health care clinics in South Africa. All three cohorts have provided whole blood samples to assess the expression of messenger RNA (mRNA) of the 3 genes by reverse transcription quantitative polymerase chain reaction (RTqPCR). Sputum samples for culture and confirmation of the disease were requested for all participants. Study II: We investigated the accuracy of blood-based screening using a cartridgebased rapid 3 genes expression prototype test Xpert-MTB-Host-Response (Xpert-MTB-HR) and a point-of-care C-reactive protein (CRP) for mass screening for active TB in persons deprived of liberty. All incarcerated individuals were screened by sputum collection and confirmed by Xpert MTB/RIF and solid medium culture. Of these, 100 confirmed TB cases and 200 controls (without TB) were selected and compared the performance of Xpert-MTB-HR and CRP. We evaluated the receiver operating characteristic area under curve (AUC) and assessed at 90% sensitivity and 70% specificity, consistent with WHO target product profile (TPP) benchmarks. Results: Study I: In the cohort 1, the 3 gene score identified progression from ILTB to active TB 6 months earlier, with a sensitivity of 86% and specificity of 84% (AUC, 0.86; 95% CI: 0.77-0.96). In the cohort 2, with a sensitivity of 90.91%, the 3 gene score achieved 68.75% specificity for active cases finding (AUC, 0.87; 95% CI: 0.79-0.94). In the cohort 3, with a sensitivity of 90.11%, the 3 genes score achieved 89.19% specificity and 99.42% for distinguishing active TB from healthy controls and patients with other lung diseases (AUROC, 0, 94; 95% CI: 0.88-0.99). Further, the 3 genes score also distinguished patients who had failed treatment (AUC, 0.93; 95% CI: 0.83-1.00). Collectively, in all cohorts (1-3), the 3- gene score identified patients with active TB with a sensitivity of 90%, specificity of 70%, and a negative predictive value above 95% with prevalence of 4% to 25%. Study II: With 90% sensitivity, Xpert-MTB-HR had significantly higher specificity (53.0%, 95% CI: 45.0-69.0%) than CRP (28.1%, 95% CI: 20.2-41.8%), but none met the TPP reference standards of 70% specificity. Among individuals with the highest bacillary load in sputum, the sensitivity was very high (97%) for the Xpert-MTB-HR with a specificity of 70%. Conclusion: Study I: It is concluded that across the 3 independent prospective cohorts, the 3 genes TB score approaches the TPP reference standards for non-sputum-based triage test with high negative predictive value. Study II: However, when evaluating the diagnostic potential for active cases finding the CRP and the Xpert-MTB-HR did not meet the TPP benchmarks for a triage test. Although, Xpert-MTB-HR was highly sensitive in detecting individuals with medium or high sputum bacillary burden.