Evaluation of bacterial kill when modelling the bronchopulmonary pharmacokinetic profile of moxifloxacin and levofloxacin against parC-containing isolates of Streptococcus pneumoniae

Objectives: The increasingly recognized prevalence of first-step parC mutants in Streptococcus pneumoniae and the development of de novo resistance while on fluoroquinolone therapy are of concern. Previous work by our group demonstrated the ability of moxifloxacin, but not levofloxacin, to eradicate parC mutants. The objective of this experiment was to determine whether these fluoroquinolone antibiotics provided equivalent bacterial kill when similar AUC/MICs were examined. Methods: An in vitro pharmacodynamic model was used to simulate the epithelial lining fluid (ELF) concentrations following oral administration of levofloxacin 500 mg once daily and moxifloxacin 400 mg once daily in older adults. In addition, a range of AUC/MICs were also modelled, including levofloxacin 750 mg once daily. Five different S. pneumoniae containing first-step parC mutations and one isolate without mutations were tested for 48 h and time-kill curves were constructed. Samples at 0, 24 and 48 h were collected for phenotypic and genotypic profiling. HPLC was used to verify that target exposures were achieved. Results: The isolate without a parC mutation displayed a 4 log reduction in cfu after treatment with levofloxacin 500 mg and did not select for resistance. In all five isolates containing first-step parC mutations, resistance emerged within 48 h with a >= 16-fold increase in MIC and the acquisition of a gyrA mutant. Increasing the exposure of levofloxacin to similar to 750 mg dose still led to >= 16-fold increase in MIC at 48 h in two of the four isolates containing parC mutations. On the other hand, moxifloxacin 400 mg sustained bacterial killing against the two isolates tested without the selection of resistant mutants. It appears that the critical AUC/MIC necessary to prevent the acquisition of resistance for levofloxacin is 200 and similar to 400 for moxifloxacin. Conclusions: Due to suboptimal exposures, once-daily oral regimens of levofloxacin at both 500 and 750 mg inconsistently led to bactericidal activity and the frequent acquisition of a second-step gyrA mutation in S. pneumoniae isolates already containing a first-step parC mutation. Conversely, once-daily moxifloxacin 400 mg provides exposures that vastly exceed the apparent efficacy breakpoint and did not select for second-step mutants until exposures were decreased 4-fold. As a result of these data and the emerging literature involving mutations in the pneumococcus, caution should be exercised when the respiratory fluoroquinolones are used to treat patients infected with S. pneumoniae suspected of having parC mutations.