Pilot Award

Between 2000 and 2009, the estimated global incidence of multidrug-resistant tuberculosis (MDR TB) more than doubled, and extensively drug-resistant tuberculosis (XDR TB) was newly verified in 57 countries. The lack of rapid drug susceptibility testing (DST) has been a major impediment to effective prevention, treatment and epidemiologic research of drug resistant TB. In most settings, patients with drug resistant TB are routinely identified only after multiple failed treatment attempts, leading to mortality exceeding 90% among HIV co-infected patients and the potential for ongoing community transmission. The proposed study seeks to reduce the burden of drug resistant TB by improving early identification of patients with MDR TB and by examining the diagnostic accuracy and cost-effectiveness of non-commercial phenotypic assays. We will capitalize on an extensive pre-existing research infrastructure in Zimbabwe centered on a Division of AIDS-funded Clinical Trials Unit to enroll 300 active TB patients with risk factors for drug resistance. Direct inoculation of patient specimens to drug-free and drug-containing media will be performed using the Microscopic Observation Drug Susceptibility (MODS) assay and two novel non-commercial DST techniques, Thin Layer Agar (TLA) and the Nitrate Reductase Assay (NRA), utilizing solid media. We will calculate the sensitivity, specificity, and positive and negative predictive values of these assays for MDR-TB diagnosis and compare these measures of diagnostic accuracy to those obtained by a molecular reference assay. This work has the potential to (1) validate or improve upon current international clinical care standards for patients at risk for drug resistant TB, and (2) validate the use of rapid, low-cost phenotypic methods in a high HIV prevalence setting.