Project 3 - Metabolic biomarkers of TB disease, treatment response and infectiousness Project Leader: Robin Wood Co-investigators: Kyu Rhee, Sara Suliman, Digby Warner, D. Branch Moody ABSTRACT Sputum-based tests are a mainstay of modern tuberculosis (TB) diagnostics that have historically proven invaluable. However, their utility has proven variable across clinical settings and patient populations where disease prevalence and mortality are high, including diagnosing TB in children and HIV co-infected patients. Existing diagnostics have further focused chiefly on disease detection. Yet, control of the TB pandemic ultimately also requires the ability to monitor treatment efficacy and disease transmissibility. Project 3 seeks to address these unmet diagnostic needs by developing a new panel of metabolite-based biomarkers present in human serum and urine, which are readily obtained from nearly all subjects. Our approach has the potential to enable real-time monitoring of treatment response, diagnose sputum- negative cases, and report on clinical infectiousness. Using new mass spectrometry platforms for broad and unbiased metabolite detection from serum and urine, we have discovered several host metabolites whose levels enabled non-invasive diagnosis and treatment monitoring of TB. Levels of one metabolite, diacetylspermine, were detectable with a clinical grade ELISA, and found to correlate with sputum bacterial load and treatment response over 14 days of therapy in independent discovery and validation cohorts. Going forward, we will validate these molecules further along the path to clinical development and use broad mass spectrometry profiling of human serum and urine to detect new host and bacterial metabolites associated with the TB disease state and treatment response. We will validate diacetylspermine, sphingomyelin and other existing lead candidate metabolites for their clinical utility in a prospective cohort of sputum-confirmed and sputum-negative TB patients during the initiation of chemotherapy. We will finally measure viable Mtb contained in exhaled bioaerosols of TB patients as the biological foundation of efforts to identify and develop diagnostic biomarkers of clinical infectiousness. We intend to validate new tests based on existing serum and urine biomarkers that are suitable for entry into the NIAID Feasibility of Novel Diagnostics for TB (FEND) program within the current TBRU term.