Examination of time-kill curves of antibiotic-exposed bacteria using in vitro dynamic models allows pharmacokinetically related comparisons of antimicrobial effects but may or may not directly reflect the selective enrichment of resistant mutants. Bacterial resistance has been studied infrequently using these models. Limited observations reported from earlier timekill studies (3, 8, 21-23) precluded delineation of relationships of the area under the concentration-time curve (AUC)/MIC ratio with resistance because the ranges of the simulated AUCto-MIC ratios were too narrow. In fact, the first attempts to relate resistance to the AUC/MIC or peak concentration (C max )/MIC ratio were reported quite recently from studies that declared resistance analysis as a primary goal (1,7,17,18,20,(25)(26)(27)30,33,34; A. MacGowan and K. Bowker, Abstr. 41st Intersci. Conf. Antimicrob. Agents Chemother., poster A-440, 2001). Despite wide ranges of AUC/MIC ratios simulated in some recent studies (17)(18)(19)(20)27,33; MacGowan and Bowker, 41st ICAAC), reasonable relationships with resistance were not established. The relatively few studies of these relationships can be classified as those that directly attempt to relate resistance to the simulated pharmacokinetics but do not (17,20) and those that imply the existence of relationships with the AUC/MIC ratio measured within a 24-h dosing interval (AUC 24 /MIC) or with the C max /MIC ratio but do not actually report them (26,27,30). One study did report a complex effect of AUC 24 /MIC and duration of moxifloxacin treatment on bacterial resistance (MacGowan and Bowker, 41st ICAAC), but the three-dimensional plots masked rather than highlighted these links. For example, according to an analysis of these data (A. Firsov, S. Vostrov, I. Lubenko, S. Zinner, and Y. Portnoy, Abstr. 42nd Intersci. Conf. Antimicrob. Agents Chemother., abstr. A-1210Chemother., abstr. A- , p. 10, 2002, the reported 72-h area under the population analysis profile-time curve as an index of pneumococcal resistance did not correlate with simulated AUC 24 /MIC ratios (r 2 , 0.04).

These data suggest that AUC24/MICs >100 h may protect against the selection of resistant S. pneumoniae mutants. Since the usual 400 mg dose of moxifloxacin provides much higher AUC24/MIC (270 h), it is expected to prevent mutant selection at clinically achievable concentrations. Also, these data provide further support for the MSW hypothesis.

These findings support the MSW hypothesis and suggest comparable antistaphylococcal effects of clinically achievable AUC24/MIC90s of daptomycin and vancomycin but slightly better prevention against the selection of resistant S. aureus by daptomycin.

Overall, these findings predict greater efficacy of clinically achievable AUC/MIC (or AUC24/MIC) of ABT492 both in terms of the anti-staphylococcal effect and prevention of the selection of resistant mutants.

No abstract