Diagnosis of tuberculosis and monitoring of treatment via RNA liquid biopsy

Tuberculosis is caused by infection with Mycobacterium tuberculosis and is a major global health threat with nearly 10 million new cases and 1.7 million deaths every year. 40-60% of patients with TB are not initiated on effective treatment, in large part due to the lack of informative diagnostics. Whole blood transcriptomic signatures have been extensively investigated as biomarkers, but the most promising signatures when implemented in multiple countries, was not useful for targeting short-course TB preventive therapy, and has substantial overlap with transcripts induced by other viral respiratory infections. We will pursue the highly translational concept that plasma cell-free RNA (cfRNA) is a promising new class of blood borne biomarker for TB. cfRNA measures cell death of circulating cell types and peripheral tissues and provides a fundamentally different window into disease dynamics relative to whole blood RNA (wbRNA). Our hypothesis is supported by a case-control study (see pilot studies) of 251 adult patients from three cohorts in Uganda, Vietnam and the Philippines in which plasma cell-free RNA signatures differentiated patients with active TB disease and symptomatic TB-negative patients with high sensitivity (97.1%) and specificity (85.2%) and correlated with measures of disease burden including CXR score. This test performance exceeds the minimum accuracy criteria for a TB triage test and outperforms best-in-class wbRNA biosignatures. We will couple well-characterized cohorts of TB patients from multiple countries, with advanced molecular assays and computational approaches i) to validate plasma cfRNA signatures in a large patient cohort of 1,378 adult participants recruited at clinical sites in five countries (samples available, Aim 1), ii) to compare the diagnostic accuracy of cfRNA against other host response readouts (data available), including Xpert TB Host Response (Xpert HR), Chest X-ray (CXR) and C-reactive protein (CRP) (Aim 1), iii) to translate the most promising cfRNA signatures to a prototype PCR assay (Aim 1), iv) to test the possibility to predict severity of disease on CXR, a key feature of current TB treatment stratification algorithms (Aim 2), and iii) to assess the utility of cfRNA signatures as a test of cure, and to monitor for relapse (Aim 3). Our specific aims are: Successful implementation of this study will provide a comprehensive evaluation of plasma cell-free RNA as an analyte to identify active TB disease in patients living with and without HIV and with and without diabetes, and to monitor therapeutic interventions. These studies have great potential for human health impact: there is an acute need for more informative biomarkers of TB, and cfRNA in plasma represents a fundamentally new class of biomarker. Successful implementation of this study will inspire future studies to translate promising cfRNA signatures into point-of-care assays and to develop cfRNA signatures of TB progression and incipient TB.