Novel biomarker more strongly associated with tuberculosis treatment outcome than conventional or genotypic methods

Introduction: Development of a surrogate biomarker for treatment outcome is a critical gap in TB clinical trials and improving efficiency of drug testing. We have developed an enhanced mycobacteriophage with fluorescent reporter (Φ 2 GFP10), which detects metabolically active MTB and has excellent potential as a prognostic biomarker in clinical sputum. Aims and objectives: To compare Φ 2 GFP10 assay to conventional TB culture and GeneXpert MTB/RIF in serial sputum samples from TB patients on treatment to predict 6-month outcome. Methods: Sputum samples were collected from prospectively enrolled patients presenting to a municipal chest clinic in Durban, South Africa at baseline and weeks 1,2,3,4, and 8. For each sample, liquid TB culture measured as time to positivity (TTP) (hrs.), GeneXpert quantified as cycle threshold (CT), and Φ 2 GFP10 assay enumerated by flow cytometry (9gated events9) was performed. Outcome was determined at 6-months by blinded clinician with standard WHO criteria. We compared inverse TTP and inverse CT to gated events to determine which was most associated with treatment outcome. Analysis was by t-test at baseline and 8 weeks. Further longitudinal analysis is planned. Results: We enrolled 20 patients at treatment start. Initial median gated events 3344 declined to 362 by week 8. Week 8 gated events in clinical failures was significantly higher than cure (p=.0067). While both TTP and CT declined during treatment, the difference was not significant in clinical failure and cure at week 8 (p=.515 and p=.366, respectively). Conclusions: The phage biomarker accurately predicts treatment outcome compared to conventional liquid culture and GeneXpert.