Abstract
Carbapenem-resistant gram-negative organisms are a rising global concern, and novel therapeutic options are urgently needed. Cefepime/enmetazobactam (FEP/META), approved for the treatment of complicated urinary tract infections, could be an important asset in clinical care. This study aimed to determine the activity of FEP/META in a collection of genetically characterized carbapenem-resistant Enterobacterales and Pseudomonas aeruginosa. FEP/META susceptibility was tested in 104 carbapenem-resistant clinical isolates (Escherichia coli, 14; Klebsiella pneumoniae, 30; other Enterobacterales [oEs]), 29; P. aeruginosa, 31) as tested by agar disk diffusion according to the European Committee on Antimicrobial Susceptibility Testing protocols. A carbapenemase gene was identified in 75% (78/104) of all isolates. A cefepime-resistant phenotype was detected in 89% (93/104) of all isolates. FEP/META was active against 29% (30/104) of all isolates tested, including 2 E. coli, 5 K. pneumoniae, 11 oEs, and 12 P. aeruginosa isolates. Twenty-two out of 78 (28%) of all carbapenemase-positive and 8 out of 26 (31%) of all carbapenemase-negative isolates were tested susceptible to FEP/META, respectively. META restored FEP susceptibility in 22/93 (24%) of cefepime-resistant isolates (16 carbapenemase-positive, 6 carbapenemase-negative), indicating a direct META effect to restore cefepime susceptibility. In conclusion, this study provides evidence for activity of FEP/META in selected carbapenem-resistant Enterobacterales and P. aeruginosa. Susceptibility varied across resistance genotypes and included isolates carrying resistance determinants formally not targeted by META. FEP/META could be an option for carbapenem-resistant organisms if other recommended therapies fail.
Introduction
The rising prevalence of multidrug-resistant gram-negative pathogens represents a growing global public health threat and alarming raising trends across Europe. 1 Increasing antimicrobial resistance (AMR) trends are driven by the misuse and overuse of antibiotics in humans and animals, poor infection prevention and sanitation, environmental contamination, global travel, and gaps in regulatory oversight limiting effective antibiotic stewardship. Resistance to broad-spectrum β-lactam antibiotics—primarily due to the production of extended-spectrum β-lactamases (ESBLs), AmpC enzymes, and carbapenemases—has significantly reduced the effectiveness of conventional antimicrobial therapies. In fact, recent reports by the European Centre for Disease Prevention and Control and the Robert Koch Institute indicate that infections associated with ESBL-producing organisms now account for a substantial proportion of invasive infections, particularly in nosocomial settings.2,3 Given the global spread of ESBL-positive organisms, the emergence of carbapenemases on a global, national, and regional scale, 1 including Germany, 4 becomes even more worrisome. The World Health Organization has accordingly classified carbapenem-resistant Enterobacterales, Pseudomonas aeruginosa, and Acinetobacter baumannii as critical priority pathogens for the development of new antibiotics. 5
Cefepime/enmetazobactam (FEP/META) is a novel β-lactam/β-lactamase inhibitor combination of cefepime (FEP), a fourth-generation cephalosporin, and enmetazobactam (META), a zwitterionic penicillanic acid sulfone β-lactamase inhibitor. META has potent inhibitory activity against a broad spectrum of β-lactamases, including the most clinically relevant ESBL families (e.g., TEM, SHV, CTX-M). Indeed, in vitro analysis using broth microdilution assays has demonstrated potent FEP/META in vitro activity against Enterobacterales, including ESBL- and AmpC-producing isolates. Moreover, FEP/META showed in vivo efficacy against Klebsiella pneumoniae in mouse pneumonia models.6,7 FEP/META is currently approved for the treatment of adult patients with complicated urinary tract infections, including pyelonephritis, hospital-acquired pneumonia (including ventilator-associated pneumonia), bloodstream infections associated with these conditions, and infections due to aerobic gram-negative organisms in patients with limited treatment options.8,9
Given the global surge in AMR, the introduction of FEP/META represents a promising addition to the antibiotic armamentarium, particularly for the targeted, personalized treatment of severe infections caused by multidrug-resistant gram-negative bacteria. Interestingly, reports are indicating that FEP/META might exhibit inhibitory activity not only in ESBL- or AmpC-producing isolates but also in the presence of carbapenemases, specifically blaOXA-48-like.10,11 To further substantiate available evidence, we tested the activity of FEP/META in a collection of genetically characterized Enterobacterales and P. aeruginosa, exhibiting carbapenem-resistant phenotypes.
Methods
Bacterial isolates
Isolates to be included in this study were selected from a contemporary collection of isolates from clinical (wound and intraoperative swabs, urine, respiratory material, blood cultures, and catheter-related samples) and surveillance specimens (i.e., rectal swabs) between January 2015 and June 2024. Isolates were selected based on a carbapenem-resistant phenotype (defined as combined or isolated resistance against meropenem and imipenem, as determined on a Vitek 2 instrument using an AST-N428 card (bioMérieux, Marcy l′Étoile, France) and by agar diffusion. In total, 14 E. coli, 30 K. pneumoniae, 29 other Enterobacterales (oEs), and 31 P. aeruginosa fulfilled selection criteria and were available for analysis. The group of oEs included Citrobacter freundii complex (n = 6), Citrobacter braakii (n = 2), Enterobacter cloacae complex (n = 17), Serratia marcescens (n = 3), and Providencia stuartii (n = 1). All isolates were differentiated to the species level by mass spectrometry on a Microflex instrument (Bruker Daltonics, Bremen, Germany).
Susceptibility testing and interpretation
Susceptibility testing was carried out using agar disk diffusion according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) clinical breakpoints manual version 14.0 (1 January 1, 2024, to December 31, 2024) 12 and the addendum of breakpoints for FEP/META (May 2024). For P. aeruginosa, cefepime breakpoints were employed to infer potential FEP/META susceptibility. In brief, colonies from fresh overnight cultures were used to prepare a McFarland 0.5 bacterial suspension in 0.9% NaCl using a Densicheck instrument (bioMérieux). Suspensions were spread onto Mueller–Hinton agar (OXOID), and antibiotic-containing discs (cefepime [30 µg; MAST Group, Reinfeld, Germany], cefepime/enmetazobactam [30 µg/20 µg; Liofilmchen, Roseto Degli Abruzzi, Abruzzo, Italy], meropenem [10 µg; MAST Group], and imipenem [10 µg; MAST Group]) were added. Agar plates were incubated at 37°C for 16–20 hours, and inhibition zone diameters were determined. SIR categorization refers to EUCAST breakpoints 12 with modifications to support more convenient data analysis and readability: (1) isolates exhibiting a cefepime susceptible (increased exposure) phenotype are here referred to as susceptible; (2) isolates exhibiting FEP/META inhibition zone diameters falling within the ATU (21–22 mm) were regarded as susceptible if the zone diameter was 22 mm and resistant if the zone diameter was 21 mm; (3) no FEP/META breakpoints are currently available for P. aeruginosa, thus the cefepime breakpoints were employed as a proxy to infer potential FEP/META resistance (inhibition zone diameter <21 mm) or potential susceptibility (inhibition zone diameter ≥21 mm). It is acknowledged that for P. aeruginosa, the term “FEP/META-susceptible” as employed in this analysis does not meet the formal criteria of the EUCAST categorization of a clinical susceptibility. Full susceptibility profiles per isolate are provided in Supplementary Table S1.
Detection of carbapenemase-encoding genes
Carbapenemase-encoding genes were detected either by a lab-developed test targeting the five most common carbapenemase-encoding genes in Germany 4 or by whole genome sequencing. Isolates that tested positive for carbapenemase-encoding genes by either of the assays are referred to as carbapenemase-positive, while all other isolates are referred to as carbapenemase-negative.
Carbapenemase-specific PCR
Bacteria from pure cultures were diluted in 200 µL H2O and heated for 5 minutes at 95°C. DNA was purified from cleared supernatants on MagNAPure96 (Roche, Basel, Schweiz) or QIASymphony (Qiagen, Hilden, Germany) instrument. Purified DNA was used as a template for the detection of blaKPC, blaNDM, blaVIM, blaIMP, and blaOXA48-like. Primers and reaction conditions have been described elsewhere. 13
Detection of carbapenemases by whole genome sequencing
DNA was isolated using the DNeasy UltraClean Microbial Kit (Qiagen, Germany) according to the manufacturer’s instructions. DNA concentration was assessed using the Qubit 1X dsDNA BR Assay Kit (Thermo Fisher Scientific, USA), and for subsequent long-read sequencing (Oxford Nanopore Technologies [Oxford Nanopore Technologies], Oxford, UK), 40 ng/µL of DNA was sent to Microsynth Seqlab GmbH. Genomes were generated from Oxford Nanopore long-read sequencing data and assembled using Trycycler to obtain a consensus long-read assembly for each isolate. 14 AMR determinants were identified from these assemblies using NCBI AMRFinderPlus v3.12.8 with the AMRFinderPlus reference database v2024-07-22.1.15,16 AMRFinderPlus outputs were merged across isolates and filtered to retain β-lactam/β-lactamase-associated hits, defined as AMR entries with subclass containing “beta-lactam.”
Results
Characterization of isolate collection
In total, 104 clinical carbapenem-resistant isolates (E. coli, 14; K. pneumoniae, 30; oEs, 29; P. aeruginosa, 31) were included in the study. All isolates were tested for the presence of a carbapenemase either by PCR (n = 54) or PCR in combination with whole genome sequencing (n = 50). Seventy-five percent (78/104) of the isolates carried at least one carbapenemase-encoding gene (Supplementary Table S1). All additional 26 isolates were tested negative for a carbapenemase-encoding gene and therefore, alternative, yet here uncharacterized mechanisms account for the observed carbapenem-resistant phenotype (Fig. 1A). In 78 carbapenemase-positive isolates, Ambler class B carbapenemases (blaVIM n = 22, blaNDM n = 19, or blaIMP n = 1) showed the highest abundance (n = 42; 53.8%), followed by blaOXA-48-like (n = 18; 23.1%) and blaKPC (n = 14; 17.9%). Four isolates were positive for two independent carbapenemases (5.1%; blaNDM in combination with blaOXA-48-like) (Fig. 1B).

Study population and distribution of carbapenemase classes.
Overall FEP/META susceptibility
Using agar disk diffusion, a FEP/META susceptible phenotype was identified in 30/104 (28.8%) of all isolates (Table 1), including 2 E. coli, 5 K. pneumoniae, 11 oEs, and 12 P. aeruginosa isolates (Table 1). Stratified for species, 14.3% of all E. coli, 16.7% of all K. pneumoniae, 37.9% of all oEs, and 38.7% of all P. aeruginosa were FEP/META susceptible, respectively (Table 1). Notably, seven Enterobacterales isolates exhibited FEP/META inhibition zone diameters falling within the ATU, but were here counted as FEP/META susceptible.
Susceptibility of Cefepime/Enmetazobactam in Study Population
No EUCAST FEP/META breakpoints are available for Pseudomonas aeruginosa. P. aeruginosa breakpoints for cefepime were employed to infer potential FEP/META activity. The designation “FEP/META susceptible” thus does not formally indicate clinical usefulness according to EUCAST SIR definitions.
EUCAST, European Committee on Antimicrobial Susceptibility Testing; FEP/META, cefepime/enmetazobactam.
FEP/META activity in carbapenemase-positive isolates
Next, we hypothesized that FEP/META susceptibility could be associated with a specific species and/or carbapenemase. In total, 22/78 (28.2%) of all carbapenemase-positive isolates were tested FEP/META susceptible. FEP/META susceptibility was identified in all groups of bacteria (Fig. 2). Specifically, with regard to carbapenemase-positive isolates, 2/14 (14%) E. coli, 5/25 (20%) K. pneumoniae , and 6/19 (26.3%) of all oEs were FEP/META susceptible (Fig. 2). In P. aeruginosa, FEP/META exhibited activity against 9/20 (45%) of carbapenemase-positive isolates. Importantly, FEP/META susceptibility was identified across all types of carbapenemases, that is, 6/22 (27.3%) FEP/META susceptible, carbapenemase producing isolates carried class A (blaKPC), 11/22 (50%) class B (blaNDM/VIM), 5/22 (22.7%) class D (blaOXA-48-like) β-lactamases (Fig. 2). Compared to the overall distribution of carbapenemases, there was no evident overrepresentation of a specific carbapenemase class.

Activity of FEP/META in carbapenemase-positive isolates. Columns represent the proportion of susceptible isolates in relation to the total number of carbapenemase-positive isolates per species. Escherichia coli, n = 14; Klebsiella pneumoniae, n = 25; oEs, n = 19; Pseudomonas aeruginosa, n = 20. The detection of carbapenemase-types is indicated by colors, and numbers indicate the number of isolates positive for a specific carbapenemase. Note: No EUCAST FEP/META breakpoints are available for P. aeruginosa. P. aeruginosa breakpoints for cefepime were employed to infer potential FEP/META activity. The designation “FEP/META susceptible” thus does not formally indicate clinical usefulness according to EUCAST SIR definitions. EUCAST, European Committee on Antimicrobial Susceptibility Testing; FEP/META, cefepime/enmetazobactam.
Activity of FEP/META in cefepime-resistant isolates
Ninety-three out of 104 isolates exhibited a carbapenem- and cefepime-resistant phenotype (i.e., 70/78 [89.7%] of all carbapenemase-positive and 23/26 [88.5%] of all carbapenemase-negative isolates). Sixteen out of 70 (22.9%) carbapenemase-positive, and 6/23 (26.1%) carbapenemase-negative, cefepime-resistant isolates were tested FEP/META susceptible (Fig. 3A, B), demonstrating that β-lactamase inhibition by enmetazobactam supported a FEP/META susceptible phenotype in these isolates.

Activity of FEP/META in FEP-resistant isolates. Columns represent the proportion of FEP/META susceptible isolates in relation to the total number of FEP-resistant isolates per species.
Genomic analysis of carbapenemase-negative isolates identified the presence of class A, C, and D β-lactamases (blaTEM, blaSHV12, blaACT, blaDHA1, blaCFE, blaCMY, blaOXA-488, and blaOXA-848), alone or in combination. As these are formally susceptible to META-inhibition, this finding might explain the observed activity of FEP in the presence of META.
Of note, in the carbapenemase-positive group, six isolates carried a class B enzyme (i.e., blaVIM), for which enmetazobactam has not formally been shown to possess inhibitory activity (Fig. 3A). FEP/META has been demonstrated to have activity in isolates carrying class A (blaKPC) or D (blaOXA-48-like) carbapenemases. In line with this, the presence of META rescued cefepime susceptibility in 6/14 (13.6%) blaKPC and 4/18 (22.2%) blaOXA-48-like positive isolates, respectively.
Discussion
In this study, we evaluated the susceptibility of FEP/META in a collection of carbapenem-resistant Enterobacterales and P. aeruginosa isolates grown from surveillance and clinical specimens. Isolates were selected on the basis of their resistance phenotype (i.e., resistance against meropenem and/or imipenem). All isolates were carbapenem-resistant, and the majority carried at least one carbapenemase-encoding gene, which resembles the current epidemiological situation in Germany. 4
FEP/META demonstrated in vitro activity in approximately 30% of all carbapenem-resistant isolates, and this activity was observed across different species and carbapenem-resistance mechanisms. Carbapenem resistance in Enterobacterales is predominantly mediated by the production of carbapenemases. However, resistance can also emerge via carbapenemase-independent mechanisms—most notably through the overexpression of β-lactamases such as ESBLs or AmpC, in combination with porin loss and/or upregulation of efflux pumps. These alterations, when acting in concert, may result in a carbapenem-resistant phenotype in the absence of carbapenemase genes. 17 Thus, given proven inhibitory activity of enmetazobactam against Ambler class A but also class C β-lactamases,18–20 it appears possible that cefepime susceptibility could be restored in isolates in which blaKPC or carbapenemase-independent mechanisms are responsible for resistance. Indeed, recent evidence indicates that FEP/META is active in some Enterobacterales isolates that produce KPC, and this activity is likely linked to a direct inhibition of the carbapenemase. 10 Consistent with this, 6/14 blaKPC-positive isolates exhibited FEP/META susceptibility in our collection.
Conversely, resistance to FEP/META in the remaining eight blaKPC-positive isolates may be explained by the co-expression of additional cefepime resistance mechanisms that are not inhibited by enmetazobactam or by the presence of KPC variants with reduced affinity for META. Importantly, current evidence suggests that FEP/META activity is not predictably associated with the presence of blaKPC-2 or blaKPC-3, highlighting the role of still poorly characterized resistance determinants. 10
A recent report highlighted the activity of FEP/META in blaOXA-48-like-positive isolates, with a susceptibility rate of 96.7%, particularly among strains co-expressing ESBL enzymes that are effectively inhibited by META [10]. Given the limited alternatives to ceftazidime-avibactam for treating infections caused by OXA-48-producing organisms, FEP/META may offer a valuable therapeutic option. 11 In contrast, our data showed a markedly lower FEP/META susceptibility rate of 22.2% among blaOXA-48-like-positive isolates. This discrepancy may be attributable to the presence of additional, uncharacterized resistance mechanisms and underscores the importance of detailed phenotypic susceptibility testing rather than sole reliance on molecular markers for the prediction of resistance profiles. 11 The need for phenotypic resistance analysis is particularly relevant in isolates that exhibit a carbapenem-resistant phenotype without producing carbapenemases. In this subgroup, FEP/META susceptibility was observed in 8/26 isolates, 2 of which were also susceptible to FEP without META. Accordingly, a specific contribution of enmetazobactam to restoring cefepime activity could be demonstrated in only a small subset of 6/26 isolates (with four of them falling into the FEP/META ATU zone), clearly making an off-hand prediction of susceptibility in the absence of a carbapenemase impossible.
Perhaps most unexpectedly, we observed FEP/META activity in isolates harboring Ambler class B metallo-β-lactamases (MBLs), despite established knowledge that enmetazobactam lacks inhibitory activity against MBLs. 10 Although the presence of META-susceptible VIM or NDM isoforms cannot be entirely excluded, it is more likely that MBL expression levels were low and that resistance in these cases was driven by alternative mechanisms susceptible to inhibition by META.
Currently, no FEP/META EUCAST clinical breakpoints are available for P. aeruginosa. The activity of FEP/META in general and in particular against VIM-positive P. aeruginosa was therefore deduced by applying available cefepime breakpoints. Therefore, it needs to be stressed that the observed activity of FEP/META against P. aeruginosa does not necessarily also translate into clinical effectiveness.
This study has several important limitations. First, susceptibility to FEP/META was assessed exclusively using agar disk diffusion, without determination of minimum inhibitory concentrations (MICs) by broth microdilution. As a result, the susceptibility rates reported here are not directly comparable to studies utilizing MIC-based methods to evaluate FEP/META activity in carbapenem-resistant isolates. Nevertheless, agar disk diffusion is a EUCAST-recommended reference method and is widely used in routine diagnostics. Moreover, a recent study confirmed the reliability of this method for assessing FEP/META susceptibility, 21 suggesting that our findings represent a valid approximation of susceptibility patterns in this setting. Additional limitations include the relatively small sample size (n = 104), which restricts the generalizability of our results, and the preselection of isolates based on phenotype and availability from our institutional isolate collection, which may have introduced selection bias. Although PCR-based detection of the most clinically relevant carbapenemase genes was performed, whole genome sequencing was performed only on a subset of 50 isolates. In addition, carbapenemase production was not formally evaluated through the utilization of a phenotypic assay. 22 Thus, the presence of rare or atypical β-lactamases cannot be excluded in all isolates. Furthermore, resistance mechanisms unrelated to carbapenemase production were not characterized. Last, the in vitro findings presented here should not be directly extrapolated to clinical decision making without further validation in clinical studies.
In conclusion, FEP/META showed in vitro activity against a subset of carbapenem-resistant Enterobacterales and P. aeruginosa isolates, including some with blaKPC and blaOXA-48-like, and even a few MBL-carrying isolates. Susceptibility varied widely across different species and resistance mechanisms, reflecting the importance of diverse factors to shape specific FEP/META susceptibility profiles. These findings highlight the potential role of FEP/META in selected cases (e.g., resistance against first-line recommended antibiotics 23 ) while emphasizing the need for detailed phenotypic testing to guide therapy. It should also be noted that its effective use requires dosing optimization based on pharmacokinetic/pharmacodynamic principles to ensure adequate exposure and minimize the risk of resistance development.
Authors’ Contributions
P.B.: Data curation, funding acquisition, formal analysis, investigation, supervision, validation, visualization, and writing—original draft. S.W., L.B., M.S., J.F., and P.H.: Formal analysis, investigation, supervision, validation, and writing—review and editing. M.A.: Resources, supervision, and writing—review and editing. H.R.: Formal analysis, funding acquisition, project administration, resources, supervision, validation, and writing—original draft.
Footnotes
Acknowledgments
The authors would like to thank the technical staff at the Institute of Medical Microbiology, Virology and Hygiene for their technical support. They would also like to express their gratitude to Benjamin Berinson and Nicole Degel-Brossmann for their work, as well as to Jiabin Huang for his help with the bioinformatic analysis.
Author Disclosure Statement
The authors declare no conflicts of interest.
Funding Information
The study was supported by an unrestricted grant from Advanz Pharma (given to H.R.). The company had no influence on the study design, data acquisition, data interpretation, or article writing. P.B. was supported by the iDfellows: Hamburg Clinician Scientist Programme in Infectious Diseases (DFG funding code 493 624 519).
Supplemental Material
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
