In Vitro Activity of Meropenem-Vaborbactam against Clinical Isolates of KPC-Positive Enterobacteriaceae

ABSTRACT Vaborbactam (formerly RPX7009) is a novel inhibitor of serine β-lactamases, including Ambler class A carbapenemases, such as KPCs. The current study evaluated the in vitro activity of the combination agent meropenem-vaborbactam against a global collection of 991 isolates of KPC-positive Enterobacteriaceae collected in 2014 and 2015 using the Clinical and Laboratory Standards Institute (CLSI) standard broth microdilution method. The MIC90 of meropenem (when tested with a fixed concentration of 8 μg/ml of vaborbactam) for isolates of KPC-positive Enterobacteriaceae was 1 μg/ml, and MIC values ranged from ≤0.03 to >32 μg/ml; 99.0% (981/991) of isolates had meropenem-vaborbactam MICs of ≤4 μg/ml, the U.S. FDA-approved MIC breakpoint for susceptibility to meropenem-vaborbactam (Vabomere). Vaborbactam lowered the meropenem MIC50 from 32 to 0.06 μg/ml and the MIC90 from >32 to 1 μg/ml. There were no differences in the activity of meropenem-vaborbactam when the isolates were stratified by KPC variant type. We conclude that meropenem-vaborbactam demonstrates potent in vitro activity against a worldwide collection of clinical isolates of KPC-positive Enterobacteriaceae collected in 2014 and 2015.

C arbapenem resistance has emerged worldwide in clinical isolates of Enterobacteriaceae (1)(2)(3). The spread of carbapenem-resistant Enterobacteriaceae, facilitated by either the horizontal spread of carbapenemase genes or the clonal expansion of carbapenem-resistant isolates (e.g., Klebsiella pneumoniae sequence type 258), has been identified to be a global public health threat and is of particular concern for patients afflicted with health care-associated infections (4,5). Factors associated with escalating carbapenem resistance rates include the increased reliance on (and selective pressure from) carbapenems as treatment for the burgeoning number of infections caused by extended-spectrum ␤-lactamase (ESBL)-positive Enterobacteriaceae that have occurred worldwide over the last 2 decades as well as substandard infection control practices and the absence of antimicrobial stewardship programs in many hospitals (6,7). The spread of carbapenemase genes both to colonizing flora and to potential pathogens is of particular concern because carbapenemases frequently confer resistance to all ␤-lactams, the most widely prescribed class of antimicrobial agents (8). The majority of carbapenemase-producing Enterobacteriaceae are also multidrug resistant (1, 2), limiting the treatment options available for empirical and directed therapy (9). The list of antimicrobial agents currently available to treat patients infected with carbapenemresistant Gram-negative bacilli is short (aminoglycosides, tigecycline, colistin) and includes agents commonly associated with significant toxicities and increasing resistance or agents to which some species of Enterobacteriaceae show intrinsic resistance (6,7,10). New antimicrobial agents are urgently needed to address the increasing prevalence of carbapenem-resistant Enterobacteriaceae (4,5).
Carbapenemases are classified into three of the four Ambler (molecular) classes of ␤-lactamase enzymes: class A (e.g., KPC), class B (e.g., NDM, VIM, IMP), and class D (e.g., OXA-48) enzymes. Class A, class C (AmpC), and class D ␤-lactamase have serine-based active sites. Class B enzymes have zinc-based active sites and are known as metallo-␤-lactamases (MBLs). AmpC ␤-lactamases and ESBLs (a subset of class A ␤-lactamases) may also confer carbapenem resistance to isolates of Gram-negative bacilli when combined with porin mutations/loss, expression of efflux pumps, and/or alterations in penicillin-binding proteins (1-3, 6, 7, 9). KPCs have the greatest global distribution of all carbapenemases associated with Enterobacteriaceae (1-3, 7, 9). KPC-positive isolates are the most common carbapenemase-producing Enterobacteriaceae in the United States and have also been reported to be widespread in South and Central America, the Middle East, and China (7). In Europe, the highest incidences of KPC-positive Enterobacteriaceae are found in Italy and Greece (1-3, 7, 9).
A proven strategy to overcome ␤-lactamase-driven resistance is to restore the activity of an inactivated ␤-lactam agent by combining it with an inhibitor of the ␤-lactamase(s) responsible for the degradation of that ␤-lactam. Vaborbactam (formerly RPX7009) was developed specifically to inactivate KPC ␤-lactamases (11). It is a novel (first-in-class), non-␤-lactam, cyclic boronic acid pharmacophore that inhibits serine ␤-lactamases of class A and class C, including KPC, IMI, SME, NMC-A, BKC-1, and FR-1 carbapenemases (8,(11)(12)(13), with no inhibition of mammalian serine proteases (11). Vaborbactam was optimized to be a potent inhibitor of serine ␤-lactamases using in silico modeling of the active sites of key serine ␤-lactamases, principally, KPCs (11). Vaborbactam possesses no antibacterial activity alone (MIC, Ͼ64 g/ml) (12,14). Mechanistically, the affinity of boronates, such as vaborbactam, for serine-based active sites of ␤-lactamases is due to the formation of a covalent complex between the catalytic serine side chain and the boronate moiety, which mimics the tetrahedral transition state of the acylation or deacylation reaction complex (11). Vaborbactam is structurally distinct from other new ␤-lactamase inhibitors, such as avibactam and relebactam, which are diazabicyclooctane inhibitors and also inhibit KPCs (1,2). Avibactam is approved for use in combination with ceftazidime and is in clinical development in combination with other ␤-lactams, and relebactam is part of a combination with imipenem. KPCs are poorly inhibited by clavulanate, tazobactam, and sulbactam; and ␤-lactam-␤-lactamase inhibitor combinations including these three older ␤-lactamase inhibitors have no utility in the treatment of infections due to carbapenem-resistant Enterobacteriaceae.
Meropenem-vaborbactam (Vabomere) in a fixed-dose combination of meropenem and vaborbactam has recently been approved by the U.S. FDA for the treatment of complicated urinary tract infections and acute pyelonephritis. The New Drug Application (NDA) included a phase 3 clinical trial to evaluate its efficacy in the treatment of complicated urinary tract infection, including acute pyelonephritis, in comparison to that of piperacillin-tazobactam (TANGO I; ClinicalTrials.gov identifier NCT02166476). A second phase 3 clinical trial in which meropenem-vaborbactam is being assessed for its efficacy for the treatment of serious infections due to carbapenem-resistant Enterobacteriaceae, including hospital-acquired and ventilator-associated pneumonia, in comparison to that of the best available antimicrobial therapy (TANGO II; ClinicalTrials.gov identifier NCT02168946) was ongoing at the time of NDA submission and review.

RESULTS
The MIC range for meropenem-vaborbactam for 991 isolates of KPC-positive Enterobacteriaceae was Յ0.03 to Ͼ32 g/ml, with the MICs for only 2 isolates exceeding 16 g/ml (both were K. pneumoniae isolates with MIC values of Ͼ32 g/ml; 1 isolate was from Greece [and produced , and the other isolate was from Italy [and produced KPC-3]). Of the two isolates with meropenem-vaborbactam MIC values of Ͼ32 g/ml, one was resistant to ceftazidime-avibactam (MIC, 32 g/ml) and had a polymyxin MIC of 0.5 g/ml, whereas the other isolate was susceptible to ceftazidime-avibactam (MIC, 2 g/ml) and had a polymyxin B MIC of Ͼ16 g/ml.
The cumulative MIC distributions of meropenem-vaborbactam stratified by KPC variant are shown in Table 3. There were no appreciable differences in the activity of meropenem-vaborbactam against different KPC variants (KPC-2, KPC-3), suggesting  There was one isolate of Raoultella ornithinolytica in the study. It was included in the data set for all Enterobacteriaceae but not in a genus-specific subset of isolates. The MICs of meropenem and meropenemvaborbactam for this isolate were Ͼ32 and 0.25 g/ml, respectively.  Table 4. There were no appreciable differences in the activity of meropenem-vaborbactam against isolates which coproduced AmpC enzymes or ESBLs and KPC (MIC 90 values of 0.06 g/ml and 1 g/ml, respectively, compared to an MIC 90 of 1 g/ml for isolates which produced only KPC) ( Table 4).

DISCUSSION
In the current study, we observed that meropenem-vaborbactam inhibited 99.0% of KPC-positive isolates of Enterobacteriaceae at Յ4 g/ml, the U.S. FDA MIC breakpoint for susceptibility (15). In the current study, the in vitro activity of meropenem-vaborbactam was equivalent to that of ceftazidime-avibactam (to which 98.2% of isolates were susceptible) and tigecycline (to which 95.8% of isolates were susceptible) ( Table 1). Meropenem-vaborbactam was demonstrated to be a more potent antimicrobial agent in vitro than ceftazidime-avibactam, tigecycline, and all other antimicrobial agents tested against the recent worldwide collection of clinical isolates of KPC-positive Enterobacteriaceae tested (Table 1). On the basis of the MIC 90 s, meropenem-vaborbactam (MIC 90 , 1 g/ml) was four times more potent than ceftazidime-avibactam and Ͼ64 times more potent than meropenem alone.
Castanheira et al. evaluated the activity of meropenem-vaborbactam against 315 serine carbapenemase-producing Enterobacteriaceae isolates, including 308 KPC-positive isolates, using checkerboard-designed panels and reported a maximum potentiation for vaborbactam activity at a concentration of 8 g/ml (14). Castanheira et al. also reported that 93.7% of the 315 serine carbapenemase-producing isolates of Enterobacteriaceae were inhibited at a meropenem MIC of Յ1 g/ml (vaborbactam concentration, 8 g/ml) and that 96.5% of isolates were inhibited at a meropenem concentration of Յ2 g/ml (14). The MIC 50 and MIC 90 for the 315 isolates were Յ0.06 and 1 g/ml, respectively (14). These investigators identified seven isolates with meropenem-vaborbactam MICs of Ն16 g/ml. All seven isolates were K. pneumoniae, four of which coproduced an MBL (MIC, 16 to Ͼ64 g/ml); the other three isolates demonstrated decreased expression of ompK37 and/or elevated expression of the AcrAB-TolC efflux system (MIC, 16 g/ml) (14). Earlier, Livermore and Mushtaq also reported that an outer membrane porin deficiency combined with the presence of ␤-lactamases can diminish the effect of vaborbactam combined with biapenem, an observation that suggested that the utility of a carbapenem-␤-lactamase inhibitor combination against certain isolates may be limited (12). Livermore and Mushtaq tested vaborbactam in combination with biapenem against 300 Enterobacteriaceae isolates, including isolates carrying KPC-type enzymes or another class A serine ␤-lactamase alone or in combination with an ESBL, derepressed AmpC, or an intrinsic resistance mechanism (12). These investigators determined that vaborbactam potentiated the activity of biapenem against KPC-positive isolates; however, the activity of biapenemvaborbactam against isolates producing class B or D (OXA-48) ␤-lactamases was limited (12).
Previously, vaborbactam was also demonstrated to possess pharmacokinetics similar to those of ␤-lactam agents, including carbapenems, and displayed efficacy as a treatment for infections caused by KPC-positive isolates of Escherichia coli, Enterobacter cloacae, and K. pneumoniae in a neutropenic mouse thigh infection model (11,18,19). Meropenem-vaborbactam has also shown activity in an in vitro hollow-fiber model that simulated human exposure, where the data generated support for the use of meropenem-vaborbactam (vaborbactam concentration, 8 g/ml) for the treatment of infections caused by KPC-positive carbapenem-resistant Enterobacteriaceae isolates with meropenem MICs as high as 8 g/ml (20).
In the current study, the difference between the KPC variants associated with meropenem-vaborbactam MICs of Ն2 g/ml ( (21) and the emergence of resistance to ceftazidime-avibactam due to plasmid-borne KPC-3 mutations during treatment of carbapenem-resistant K. pneumoniae infections (22). The differences observed between the subsets of isolates resistant to meropenem-vaborbactam and ceftazidimeavibactam may also be related to the observation that the inhibition of KPC-2 by vaborbactam does not involve S130, a residue important for inhibition by avibactam (R. Tsivkovski, M. Totrov, and O. Lomovskaya, presented at Microbe 2016, Boston, MA).
The intent of the current study was to add to the limited amount of available in vitro data on the meropenem-vaborbactam MICs for KPC-positive Enterobacteriaceae isolates (14,16,17). Our data align with the findings of these previous studies and show that vaborbactam restores the in vitro activity of meropenem against the majority of isolates of Enterobacteriaceae carrying KPCs that would otherwise be nonsusceptible to carbapenems. On the basis of the results of our current study, meropenem-vaborbactam appears to be a promising, novel carbapenem-␤-lactamase inhibitor combination. Its continued clinical development may provide a valuable therapeutic option for treating infections caused by antimicrobial-resistant Gram-negative bacilli in the future.

MATERIALS AND METHODS
Bacterial isolates. The isolates of KPC-positive Enterobacteriaceae tested in this study (n ϭ 991) were randomly selected from isolates in the frozen stock culture collection of International Health Management Associates, Inc. (IHMA; Schaumburg, IL, USA), collected in 2014 and 2015 for a global clinical isolate surveillance study on the basis of their ␤-lactamase content and year of isolation. In total, 580 isolates were selected in 2014 and 411 isolates were selected in 2015. All 991 KPC-positive isolates were previously determined to be OXA-48 negative and MBL negative by a protocol that included screening of the isolates for the presence of genes encoding the following ␤-lactamases using published multiplex PCR assays as described previously (24) . The detected ␤-lactamase genes were amplified using flanking primers and sequenced. Enzyme subtypes were determined by comparison against the subtypes in the database maintained by the National Center for Biotechnology Information (www.ncbi.nlm.nih.gov). Isolates positive for OXA-48 or an MBL were excluded from the current study because previous publications have documented that vaborbactam does not inhibit these enzymes (13).