There is limited high-quality evidence to guide the optimal treatment of Mycobacterium kansasii pulmonary disease. We retrospectively collected clinical data from 33 patients with M. kansasii pulmonary disease to determine the time-to-sputum culture conversion (SCC) upon treatment with a standard combination regimen consist of isoniazid-rifampin-ethambutol. Next, MIC experiments with 20 clinical isolates were performed, followed by a dose-response study with the standard laboratory strain using the hollow-fiber system model of M. kansasii infection (HFS-Mkn). The inhibitory sigmoid maximum effect (Emax) model was used to describe the relationship between the bacterial burden and rifampin concentrations. Finally, in silico clinical trial simulations were performed to determine the clinical dose to achieve the optimal rifampin exposure in patients. The SCC rate in patients treated with combination regimen containing rifampin at 10 mg/kg of body weight/day was 73%, the mean time to SSC was 108 days, and the mean duration of therapy was 382 days. The MIC of the M. kansasii laboratory strain was 0.125 mg/L, whereas the MICs of the clinical isolates ranged between 0.5 and 4 mg/L. In the HFS-Mkn model, a maximum kill (Emax) of 7.82 log10 CFU/mL was recorded on study day 21. The effective concentration mediating 80% of the Emax (EC80) was calculated as the ratio of the maximum concentration of drug in serum for the free, unbound fraction (fCmax) to MIC of 34.22. The target attainment probability of the standard 10-mg/kg/day dose fell below 90% even at the MIC of 0.0625 mg/L. Despite the initial kill, there was M. kansasii regrowth with the standard rifampin dose in the HFS-Mkn model. Doses higher than 10 mg/kg/day, in combination with other drugs, need to be evaluated for better treatment outcome.
Keywords: hollow-fiber model system; nontuberculous mycobacteria; rifampin.