Impact of fluoroquinolone resistance on bactericidal and sterilizing activity of a moxifloxacin-containing regimen in murine tuberculosis.

It has been shown previously that fluoroquinolone resistance (defined by resistance to at least 2 mg/liter ofloxacin) has a different impact on moxifloxacin monotherapy depending on the mutation in the sole fluoroquinolone target in Mycobacterium tuberculosis, i.e., DNA gyrase. Since tuberculosis treatment relies on multidrug therapy, we wished to determine the impact of fluoroquinolone resistance on the bactericidal and sterilizing activity of a second-line antituberculous regimen containing moxifloxacin. A total of 280 mice were inoculated with the wild-type Mycobacterium tuberculosis H37Rv strain or one of 3 isogenic fluoroquinolone-resistant mutant strains with increasing moxifloxacin resistance (the GyrB D500N, GyrA A90V, and GyrA D94G strains) and then treated for 6 months with a second-line regimen containing moxifloxacin, pyrazinamide, and ethionamide supplemented with amikacin during the first 2 months. Mice were sacrificed during treatment for measurement of bactericidal activity and 3 months after treatment completion for measurement of relapse rates (sterilizing activity). The CFU counts decreased faster in mice inoculated with the wild type than in mice inoculated with the mutant strains. The relapse rate after treatment completion was different among mice inoculated with mutant strains in relation to the drug resistance level: wild type, 0%; GyrB D500N strain, 33%; GyrA A90V strain, 50%; and GyrA D94G strain, 86%. The relapse rate observed with the GyrB D500N strain was the only one not statistically different from that observed with the wild-type strain. We demonstrated that the impact on sterilizing activity of the most active second-line drug regimen containing moxifloxacin depends on the MIC of moxifloxacin. We suggest that the precise level of moxifloxacin resistance be determined for all strains resistant to 2 mg/liter ofloxacin.