Editor's Note: As carbapenem-resistant Gram-negative bacteria (CRO) have become increasingly prevalent, the development and application of novel β-lactam/β-lactamase inhibitor (BL/BLI) combinations have gained significant attention from researchers worldwide. A recent review published in Clinical Infectious Diseases delves into the advancements of these novel BL/BLIs in addressing non-KPC resistance mechanisms, providing crucial insights for clinical treatment. This article extracts the essential points of the review related to Enterobacterales to benefit readers.Background
The rise of carbapenem-resistant Gram-negative organisms (CRO) poses a significant public health challenge, prompting researchers to focus on developing new antimicrobial agents and enhancing infection control and antimicrobial stewardship strategies. CRO encompass various Gram-negative pathogens that exhibit diverse resistance mechanisms. Identifying these resistance mechanisms through methods such as the modified carbapenem inactivation method (mCIM) or EDTA carbapenem inactivation method (eCIM), as well as genotypic testing, including molecular or whole-genome sequencing, is critical for devising effective antimicrobial therapy.
Over the past decade, most newly developed antimicrobials have targeted Klebsiella pneumoniae carbapenemase (KPC) in Enterobacterales. These drugs include BL/BLI combinations like ceftazidime/avibactam, meropenem/vaborbactam, imipenem/cilastatin/relebactam, as well as other classes of antibiotics like cefiderocol, plazomicin, eravacycline, and omadacycline, primarily used for treating Gram-positive infections. However, BL/BLI combinations remain the cornerstone of therapy for KPC-producing Gram-negative bacterial infections.
While many novel BL/BLI combinations have been specifically developed to combat KPC, they have proven effective in neutralizing KPC enzymes and successfully treating CRO infections. Additionally, these new BLIs exhibit unique antibacterial activity against non-KPC carbapenemase-producing organisms (non-KPC CPO) and non-carbapenemase-producing CRO (non-CP CRO), an aspect that has not yet garnered widespread attention. In many countries, the resistance mechanisms of non-KPC CPO are an escalating concern that remains underappreciated.
This review focuses on the enzyme-mediated activity of these novel BL/BLI combinations, consolidating preclinical and clinical data to differentiate the performance of these antibiotics and provide guidance for treating non-KPC CPO and non-CP CRO infections.
Common β-Lactamases
The Ambler classification system, based on molecular homology, classifies β-lactamases into four classes, initially identifying classes A and B. Class A enzymes are characterized by a serine active site, whereas class B enzymes, known as metallo-β-lactamases (MBLs), require zinc to hydrolyze the β-lactam ring. Subsequent discoveries identified classes C and D, which also utilize serine as their active site but are structurally distinct from class A enzymes. These are referred to as AmpC cephalosporinases and oxacillinases (OXAs), respectively.
Class A enzymes include most extended-spectrum β-lactamases (ESBLs) and carbapenemases, such as KPC, Serratia marcescens enzyme (SME), imipenem-hydrolyzing β-lactamase (IMI), and Guyana extended-spectrum β-lactamase (GES).
Class B metallo-β-lactamases include plasmid-encoded New Delhi metallo-β-lactamase (NDM), Verona integron-encoded metallo-β-lactamase (VIM), and imipenemase (IMP). Stenotrophomonas maltophilia produces a metallo-β-lactamase called L1.
Class C AmpC β-lactamases, which are not carbapenemases, are usually chromosomally mediated but can also be plasmid-borne. They are inducible by β-lactam antibiotics.
Class D enzymes include a variety of carbapenemases, such as OXA-48-like enzymes found in Enterobacterales, which have weak hydrolytic activity against carbapenems and cephalosporins, and oxacillinases produced by Acinetobacter baumannii (such as OXA-23, OXA-58).
Chemical Properties of β-Lactamase Inhibitors
The combination of β-lactam antibiotics with β-lactamase inhibitors (BLIs) has been a strategy for decades, with commonly used combinations such as amoxicillin/clavulanate, ampicillin/sulbactam, piperacillin/tazobactam, and cefoperazone/sulbactam. These combinations involve early BLIs: clavulanic acid, sulbactam, and tazobactam, which possess a β-lactam ring capable of irreversibly forming a covalent bond with the serine nucleophile in the β-lactamase active site. By occupying the enzyme’s active site, BLIs prevent the hydrolysis of the β-lactam component of the combination. However, these early BLIs primarily target narrow-spectrum class A β-lactamases, with varying success in restoring β-lactam activity against ESBL-producing bacteria and limited efficacy against carbapenemases.
Newer generation BLIs, such as avibactam, relebactam, and vaborbactam, can effectively inhibit a broader spectrum of enzymes, attributable to their non-β-lactam structure. Avibactam and relebactam are diazabicyclooctane (DBO) compounds that form a reversible covalent bond with the serine nucleophile in the β-lactamase active site. The carbamoyl-enzyme intermediate effectively occupies the active site, preventing the hydrolysis of the β-lactam component of the combination. Compared to earlier BLIs, the structure of DBOs, particularly their sulfate groups, allows for interaction with conserved polar residues in the active site, creating a more stable interaction with certain enzymes. Additionally, DBOs are resistant to hydrolysis, meaning fewer molecules are required to inhibit β-lactamase activity. Vaborbactam, on the other hand, is a boronic acid inhibitor, with the boron atom reversibly covalently bonding with the serine nucleophile of β-lactamase, preferentially targeting serine β-lactamases over mammalian serine hydrolases. These newer BLIs possess little or no intrinsic antimicrobial activity and are ineffective against MBLs.
Enterobacterales
While β-lactamase production is the primary resistance mechanism of Enterobacterales against β-lactam antibiotics, some exceptions exist. Notably, the Morganellaceae family within Enterobacterales (including Proteus, Morganella, and Providencia species) exhibit intrinsic resistance to imipenem, not due to β-lactamase production, but rather owing to their weak affinity for penicillin-binding proteins and porin loss. As a result, combination therapy with relebactam for these bacteria is unnecessary. Similarly, not all carbapenem-resistant Enterobacterales (CRE) produce carbapenemases, referred to as non-CP CRE. In the CRACKLE-2 study, which analyzed 1,040 CRE isolates from U.S. hospitalized patients, 19% (194/1,040) did not possess carbapenemase genes based on whole-genome sequencing. Notably, non-CP CRE can exhibit ertapenem resistance through porin mutations and non-carbapenemase β-lactamase production (such as ESBLs or AmpC) while remaining susceptible to imipenem and meropenem.
In early in vitro studies, class A and C β-lactamases produced by Enterobacterales were found to be inhibited by avibactam, vaborbactam, and relebactam. These studies showed that these BLIs significantly reduced the minimum inhibitory concentrations (MICs) of β-lactam antibiotics against KPC-producing bacteria, emphasizing their efficacy against ESBL and AmpC enzymes. Further studies demonstrated that adding avibactam to ceftazidime reduced the MIC for bacteria producing TEM, SHV, and CTX-M-15 enzymes by 128-fold and for AmpC-producing bacteria by 8-fold. Consequently, ceftazidime/avibactam is a superior antibiotic compared to ceftazidime monotherapy for treating ESBL or AmpC-producing Enterobacterales infections. Similarly, vaborbactam and relebactam have also been shown to reduce MICs for ESBL-producing bacteria. However, clinical data supporting the use of meropenem/vaborbactam or imipenem/cilastatin/relebactam instead of their respective backbone β-lactam antibiotics for treating ESBL- and AmpC-producing Enterobacterales infections remain limited.
OXA and OXA-48-like Enzymes
In vitro studies show that avibactam can inhibit class D OXA carbapenemases produced by Enterobacterales, such as the common OXA-48-like enzymes. However, the inhibitory effect of avibactam on OXA enzymes is enzyme-specific. Although the kinetic constants of avibactam binding to OXA enzymes are generally favorable, certain enzymes, such as OXA-10, bind more slowly, leading to reduced inhibition. Nonetheless, in clinical practice, the limitations of this inhibition are not overly concerning, as most OXA enzymes (including OXA-10 and OXA-48-like enzymes) do not strongly hydrolyze ceftazidime. However, some OXA variants, such as OXA-11, can hydrolyze ceftazidime and may benefit from the addition of avibactam. In most cases, the MIC of ceftazidime/avibactam is significantly lower than that of ceftazidime alone, likely due to avibactam inhibiting ESBLs co-expressed with OXA enzymes rather than directly inhibiting OXA-mediated ceftazidime hydrolysis. This scenario is not uncommon, as carbapenemases often spread through plasmids that also carry ESBLs. Given the substrate specificity of OXA enzymes, combining avibactam with carbapenems may be particularly effective in inhibiting bacteria carrying these enzymes. For example, studies show that combining imipenem with avibactam can reduce the MIC of OXA-48-producing Klebsiella pneumoniae by 500-fold. However, no BL/BLI combination containing both avibactam and a carbapenem is currently available.
Compared to avibactam, relebactam shows weaker inhibition of OXA enzymes. In one in vitro study, 8 mg/L of relebactam restored the sensitivity of three strains (reducing the MIC from 4 mg/L to 1 mg/L or lower), but even at 32 mg/L, relebactam was only able to reduce the MIC to 16 mg/L for two imipenem-resistant isolates (MIC >64 mg/L). Vaborbactam, on the other hand, shows no inhibition of OXA enzymes. In a global in vitro study, the MIC50 and MIC90 of meropenem and meropenem/vaborbactam against 25 OXA-48-like carbapenemase-producing isolates were similar, at 16 mg/L and >32 mg/L, respectively.
Several in vivo studies in mouse models further confirm these in vitro findings. One study demonstrated that meropenem and meropenem/vaborbactam achieved similar efficacy against OXA-48-producing bacteria. Another study evaluated the efficacy of three β-lactam antibiotics combined with their respective BLIs, showing that human exposure to ceftazidime/avibactam significantly reduced bacterial density, with 96% (49/51) of isolates achieving at least a 1-log reduction, while imipenem/relebactam and meropenem/vaborbactam had relatively weaker in vivo activity. Although vaborbactam did not enhance the effect of meropenem against OXA-48-producing bacteria, the Clinical and Laboratory Standards Institute (CLSI) breakpoint for meropenem/vaborbactam in the U.S. is relatively high, which may cause bacteria to appear resistant to meropenem (MIC >1 mg/L) but susceptible to meropenem/vaborbactam (≤4/8 mg/L). If using the current European Committee on Antimicrobial Susceptibility Testing (EUCAST) breakpoint for meropenem/vaborbactam (≤8/8 mg/L), this issue could be more pronounced. Due to these inconsistent breakpoint standards, more than 50% of OXA-48-producing Enterobacterales may be mistakenly classified as susceptible to meropenem/vaborbactam. Therefore, clinicians should be aware that current breakpoints may lead to discrepancies between in vitro results and in vivo reality; molecular identification of carbapenemases should be conducted prior to using meropenem/vaborbactam in regions where OXA-48 prevalence exceeds 10%.
Several observational clinical studies emphasize that ceftazidime/avibactam is the preferred novel BL/BLI for treating infections caused by D-class OXA carbapenemase-producing bacteria. One study analyzed data from patients infected with OXA-48-producing Enterobacterales who were treated with ceftazidime/avibactam for at least 48 hours during an outbreak of OXA-48 Klebsiella pneumoniae in Spain from 2016 to 2017. Among the 57 patients included, 46 (80%) received ceftazidime/avibactam monotherapy, while 11 (20%) received combination therapy with other drugs. The 14-day all-cause mortality rate was 15% for the monotherapy group and 9% for the combination therapy group, with clinical cure rates of 80% and 64%, respectively. Another study followed 24 patients with OXA-48 CPE infections treated with ceftazidime/avibactam for at least 72 hours between 2014 and 2016, reporting 30-day and 90-day mortality rates of 8.3% and 20.8%, respectively, with a clinical cure rate of 62.5%. In a comparative study, patients with CPE bacteremia from three centers in Spain and one in Israel were included for analysis from 2012 to 2016. Of the 31 patients, 8 received ceftazidime/avibactam, while the remaining 23 received other treatments. In both the ceftazidime/avibactam and control groups, 62.5% and 60.8% of patients carried OXA-48-type carbapenemases, respectively. Although there was no significant difference in 30-day crude mortality between the two groups, the clinical cure rate was significantly higher in the ceftazidime/avibactam group than in the control group (75% vs. 35%, P = 0.031). The CAVICOR study, the largest cohort of OXA-48 CPE patients to date, compared ceftazidime/avibactam with the best available therapy (BAT). Of the 339 patients included, 255 (75%) had OXA-48 CPE isolated from blood, intra-abdominal, urinary, or pulmonary samples. The ceftazidime/avibactam-containing regimen was protective, with an adjusted odds ratio (OR) for 30-day mortality of 0.41 (95% CI: 0.20–0.80) after propensity score adjustment.
Currently, no clinical data support the use of imipenem/cilastatin/relebactam or meropenem/vaborbactam for treating OXA-producing Enterobacterales infections, and in vitro and animal studies suggest these drugs are not superior to carbapenem monotherapy. These findings align with the Infectious Diseases Society of America (IDSA) guidelines for treating CRE, which recommend prioritizing ceftazidime/avibactam over other BL/BLIs when OXA-48-like enzymes are detected or suspected.
Metallo-β-lactamases
Currently, no BL/BLIs are active against class B MBLs, representing a significant gap in antimicrobial development. The monocyclic β-lactam aztreonam is not hydrolyzed by MBLs; however, like other serine-based carbapenemases, MBLs often coexist with ESBLs, other carbapenemases, and chromosomal AmpC, leading to resistance to aztreonam in most MBL-producing isolates. Studies have shown that when aztreonam is combined with novel BL/BLIs, such as avibactam- or vaborbactam-containing combinations, the inhibitors can restore aztreonam’s antibacterial activity by inhibiting the co-produced enzymes, regardless of MBL presence. In vitro studies, case reports, and retrospective analyses of aztreonam (2 g q8h) combined with ceftazidime/avibactam (2.5 g q8h) have demonstrated good efficacy against Enterobacterales producing NDM and VIM enzymes. One study found that when combined with ceftazidime/avibactam, aztreonam’s MIC could be reduced from 256 mg/L to 0.125 mg/L. In a single-center, observational, retrospective study conducted from 2018 to 2021, 24 adult patients diagnosed with VIM-producing Gram-negative infections received at least 48 hours of ceftazidime/avibactam and aztreonam treatment. Of this cohort, 92% (22/24) were infected with Enterobacterales, and in vitro results showed synergy for both drugs against 85% (15/18) of Enterobacterales isolates. Two patients experienced clinical failure by day 14, with a 30-day mortality rate of 18% (4/22).
In clinical practice, research on the in vitro efficacy of ceftazidime/avibactam combined with aztreonam against MBL-producing bacteria is ongoing. While broth microdilution (BMD) remains the gold standard for antimicrobial susceptibility testing (AST), this method is impractical for most clinical microbiology laboratories, and automated AST systems cannot test this specific drug combination. To address this gap, innovative methods have been proposed to assess drug synergy and “hope zones,” including using gradient diffusion strips, overlapping gradient diffusion strips, or disc diffusion methods. Recently, a gradient diffusion strip containing aztreonam and a fixed concentration of 4 mg/L avibactam was introduced to provide MIC values.
The success of aztreonam combined with ceftazidime/avibactam in treating MBL-producing Enterobacterales infections suggests that other novel BL/BLIs may have similar efficacy. In vitro studies comparing ceftazidime/avibactam/aztreonam with meropenem/vaborbactam/aztreonam found similar reductions in MIC. A recent time-kill analysis evaluated the combination of imipenem/relebactam/aztreonam against Enterobacterales co-producing MBLs and serine β-lactamases. Adding relebactam to the aztreonam/imipenem combination showed synergistic effects against all 11 aztreonam-resistant isolates. Although clinical data on vaborbactam and relebactam remain limited, considering their similar inhibition of plasmid-mediated ESBLs and AmpC enzymes, they are expected to achieve similar therapeutic outcomes as avibactam.
Less Common Carbapenemases
While KPC, OXA-48, and MBL are the most clinically prevalent carbapenemases, clinicians should also be aware of several rarer class A carbapenemases in Enterobacterales and Serratia species, including SME, NmcA, IMI, and FRI-like enzymes. In a study of 19 clinical isolates (12 Enterobacterales and 7 Serratia marcescens) producing various carbapenemases, BMD testing revealed that all isolates were susceptible to ceftazidime monotherapy but resistant to imipenem and meropenem monotherapy. Regarding BL/BLI applications, adding vaborbactam restored meropenem susceptibility in all isolates, while relebactam restored imipenem susceptibility in 11/19 (58%) isolates. In vitro evaluation of eight SME-producing Serratia marcescens isolates yielded similar results. Clinicians may struggle to identify these uncommon carbapenemases using current diagnostic methods; however, these in vitro data suggest that ceftazidime/avibactam or meropenem/vaborbactam may be viable options for treating infections caused by bacteria harboring these rare carbapenemases.
Conclusion
In treating Enterobacterales infections, the differentiation between the three novel BL/BLIs—ceftazidime/avibactam, meropenem/vaborbactam, and imipenem/relebactam—is primarily based on their activity against class D OXA enzymes and the interaction between the BLIs and β-lactam backbone antibiotics. For Enterobacterales carrying OXA-48-like enzymes, ceftazidime/avibactam should be preferred over other available BL/BLIs.
While these novel BL/BLIs lack activity against MBL-producing bacteria, studies indicate that combining aztreonam with ceftazidime/avibactam provides effective treatment for Enterobacterales and certain Pseudomonas aeruginosa strains. Further research is needed to elucidate whether meropenem/vaborbactam or imipenem/relebactam can also be effectively combined with aztreonam as an acceptable treatment option.
Key:
++: In vitro activity and clinical efficacy
+: In vitro activity only
-: Limited in vitro activity
