Validation of a novel non-sputum-based host protein biosignature for diagnosis of childhood tuberculosis

With an estimated 1.25 million annual cases and 214,000 deaths, childhood tuberculosis (TB) remains a serious threat to global child health. More than 70% of the global childhood TB estimates occur in the World Health Organisation (WHO) Africa and southeast Asia region where childhood cases remain underreported due to difficulties with diagnosis. Currently, the gold standard for diagnosis of pulmonary TB involves culture of Mycobacterium tuberculosis (M.tb), the causative organism for TB disease, in sputum or other respiratory samples. However, sputum samples are difficult to obtain in children, particularly in young children. Even when sputum or other respiratory samples are successfully obtained, microbiological tests for TB such as M.tb culture or GeneXpert Ultra are less likely to have positive results in children; this is because children are more likely to have TB caused by a smaller number of M.tb. Therefore, diagnosis of childhood TB is often made without laboratory confirmation and instead based on combination of non-specific clinical and radiological features, with high likelihood of missed or clinically-significant delay in diagnosis. Hence, TB is becoming increasingly responsible for childhood morbidity and mortality caused by a treatable infectious disease. The WHO has identified the development and validation of non-sputum biomarker-based tests for detecting TB in children as a major research priority. Such tests should enable fast and accurate distinction between TB and other respiratory diseases (ORD) and perform in all children equally. Our research to date has reported the discovery of a novel 3-marker host protein biosignature of childhood TB (IL-1ra, IL-7 and IP-10) from utilising blood samples collected from HIV-uninfected children in The Gambia. The biosignature was identified in supernatants generated from small blood samples after overnight incubation and without antigenic stimulation. This biosignature reliably distinguished children with TB disease, including those missed by microbiological tests, from those with ORD, with its performance independent of age and nutritional status of the children. We are now applying for funds to validate our 3-marker host protein biosignature in the laboratory, and to refine the biosignature and optimise its performance further, using a unique combination of bio-banked and well-characterised samples obtained from independent paediatric cohorts recruited in Mali, Nigeria, Tanzania and The Gambia. Our specific objectives are to: Estimate the ability of the 3-marker biosignature to classify TB disease from ORD in HIV-infected and HIV-uninfected children from Mali, Nigeria, Tanzania and The Gambia. Compare the predictive accuracy of the 3-marker biosignature between unstimulated QuantiFERON supernatants and serum samples of children from Mali, Nigeria, Tanzania and The Gambia. Estimate the ability of the specific gene expression of the 3-marker protein biosignature (i.e. the biosignature-specific mRNA expression) to classify TB disease from ORD in HIV-infected and HIV-uninfected children from Mali, Nigeria, Tanzania and The Gambia. Determine the added value of combining the biosignature-specific mRNA expression with the protein biosignature to distinguish TB disease from ORD in HIV-infected and HIV-uninfected children from Mali, Nigeria, Tanzania and The Gambia. Our research team members have expertise in paediatrics, childhood TB and biomarker discovery. This proposal provides a unique platform to validate our promising biosignature and demonstrate the utility of serum as a non-sputum sample for biomarker-based diagnosis of childhood TB. Following the research project, it will be possible to develop the protein biosignature into a biomarker-based immunoassay test that uses, for example, micro-ELISA platforms based on microfluidic technology.