1296. Efficacy of Human-Simulated Exposures of Meropenem/Vaborbactam (MVB) and Meropenem (MEM) against OXA-48 β-lactamase-producing Enterobacterales in the Neutropenic Murine Thigh Infection Model

Abstract

BackgroundOXA-48 exhibits variable hydrolytic activity toward carbapenems, with imipenem and meropenem MICs, though increased, often reporting within the ‘susceptible’ or ‘intermediate’ ranges defined by CLSI and EUCAST. Although vaborbactam (VAB) does not enhance MEM activity against OXA-48, ~ 1/3 of OXA-48-producing Enterobacterales will test susceptible to MVB due to its higher breakpoint. Clinical implications of this discordance warrant investigation.Methods26 isolates harboring OXA-48 (n=24) and KPC (n=2) were evaluated in the neutropenic murine thigh model. MICs were determined in triplicate per CLSI. Human-simulated regimens of MVB (simulating doses of 2-2 g IV q 8 hours (h) over 3 h) and MEM (2 g IV q 8 h over 3 h) were administered resulting in similar f%T >MIC and fAUC as humans for MEM and VAB, respectively. Mice received MVB, MEM, or sham control for 24 h. Efficacy was assessed on the resulting overall change in mean± SD log10 CFU/thigh as well as the achievement of ≥ 1 log10 reduction as an established surrogate marker predictive of success for serious infections.ResultsMVB and MEM MICs ranged from 1- 64 and 2 - > 64 mg/L, respectively. Relative to 0 h control, the mean bacterial growth (mean ± SD, CFU/thigh) at 24 h in the untreated control mice was 2.69 ± 1.31. As anticipated for KPCs, MVB resulted in a mean bacterial reduction ≥ 1 log10 (-1.10 ± 0.26), whereas growth was observed on MEM (+1.45 ± 0.88). For all OXA-48 isolates, MVB resulted in variable bacterial densities ranging from -2.54 to +2.68, similarly MEM resulted in -2.18 to +2.66. Addition of vaborbactam did not enhance MEM activity for any isolate. For isolates with MVB MICs ≥ 16 (n=5), 8 (n=5, EUCAST breakpoint), 4 (n=9, CLSI breakpoint), and ≤ 2 (n=5) mg/L, 0%, 0%, 44%, and 60% of isolates treated with MVB or MEM achieved the target reduction of ≥ 1 log10 kill, respectively.ConclusionAcross the range of MICs evaluated, MVB and MEM humanized exposures in vivo resulted in similar reductions and growth in bacterial density for OXA-48-producing Enterobacterales. Moreover, these data highlight the poor efficacy of MVB for OXA-48 defined as susceptible using the current EUCAST and CLSI susceptibility criterion. Caution is therefore warranted when treating Enterobacterales testing susceptible to MVB without the genotypic profile.Disclosures David P. Nicolau, PharmD, Cepheid (Other Financial or Material Support, Consultant, speaker bureau member or has received research support.)Merck & Co., Inc. (Consultant, Grant/Research Support, Speaker’s Bureau)Wockhardt (Grant/Research Support)