A pilot study of a new molecular test for the diagnosis and treatment monitoring of TB
Introduction: Tuberculosis (TB) remains a global health emergency, with 9 million new cases and 1.5 million deaths in 2013. New diagnostic tests which improve upon the low sensitivity and slow turnaround time of existing tests are urgently needed in order to identify the estimated 3 million new TB cases which go undiagnosed each year. RealTime MTB INH/RIF (Abbott Laboratories, Abbott Park IL, “RealTime MTB”) is a new molecular test with unknown clinical utility for TB detection as well as identification of drug resistance.
Methods: We propose a two-part study of RealTime MTB INH/RIF (RealTime MTB) for TB diagnosis and treatment monitoring within a high TB burden district of South Africa. We will estimate the diagnostic accuracy (sensitivity, specificity, positive and negative predictive values) of RealTime MTB relative to the reference standard of liquid TB culture among participants suspected of pulmonary TB, and compare to those of known tests, including Xpert MTB/RIF and smear microscopy. We will perform an empiric costing of all assays from the health system perspective in order to assess cost-effectiveness. We will then follow all RealTime MTB positive patients through TB treatment, to assess the ability of RealTime MTB to predict response to treatment, as measured by time to culture negativity.
Output: We will publish our findings in the scientific literature and utilize them as preliminary data to inform an R01 proposal for a pragmatic randomized trial to ascertain the diagnostic accuracy for TB diagnosis, detection of drug resistance, impact on patient important outcomes and treatment monitoring.
Neil Martinson, SOM.