ABSTRACT
The activity of the siderophore cephalosporin cefiderocol is targeted against carbapenem-resistant Gram-negative bacteria. In this study, the activity of cefiderocol against characterized carbapenem-resistant Acinetobacter baumannii complex, Stenotrophomonas maltophilia, Pseudomonas aeruginosa, and Enterobacteriaceae strains was determined by microdilution in iron-depleted Mueller-Hinton broth. The MIC90s against A. baumannii, S. maltophilia, and P. aeruginosa were 1, 0.25, and 0.5 mg/liter, respectively. Against Enterobacteriaceae, the MIC90 was 1 mg/liter for the group harboring OXA-48-like, 2 mg/liter for the group harboring KPC-3, and 8 mg/liter for the group harboring TEM/SHV ESBL, NDM, and KPC-2.
INTRODUCTION
The World Health Organization (WHO) and the Centers for Disease Control and Prevention (CDC) have designated that antibiotic resistance is a threat of enormous gravity to public health (1, 2). At least 2 million people in the United States acquire serious infections with bacteria that are resistant to the antibiotics that they are designed to treat, and more than 23,000 deaths from these infections occur each year. Novel agents and strategies are urgently required to combat this scourge.
This report is the first in a series of studies called ARGONAUT (Antibacterial Resistance Leadership Group [ARLG] Reference Group for the testing of Novel Therapeutics), supported by the ARLG (3). In this study, the in vitro activity of cefiderocol and comparator agents against reference collections of Gram-negative bacterial species (3–6), with an emphasis on carbapenem-resistant isolates, was determined to assess the spectrum of activity of this novel agent.
Cefiderocol is a siderophore cephalosporin under development and is targeted for activity against carbapenem and multidrug-resistant (MDR) Gram-negative species (7–9). This novel agent possesses a catechol moiety on the 3 position of the R2 side chain and binds primarily to PBP3 (10) (Fig. S1 in the supplemental material). The catechol moiety acts as a siderophore to form a chelating complex with ferric iron, which facilitates transport to the periplasmic space. Cefiderocol has been reported to be more stable against β-lactamases, such as KPC-3, VIM-2, L1 (the chromosomal metallo-type carbapenemase of Stenotrophomonas maltophilia), and NDM-1 carbapenemases than agents such as cefepime and meropenem (11). The human plasma protein binding of cefiderocol is 58% (12), and pharmacokinetic/pharmacodynamic (PK/PD) studies show that cefiderocol is an agent with time-dependent activity, with the fraction of the free drug concentration in plasma exceeding a MIC (fTMIC) target of 75% of the dosing interval (13, 14). At the proposed dosing regimen for cefiderocol of 2 g infused over 3 h every 8 h, PK/PD modeling showed that this 75% fTMIC target would be attained in >90% of patients for organisms with MICs of ≤4 mg/liter (15).
In this analysis, the MICs of cefiderocol and comparators were determined by broth microdilution according to the current Clinical and Laboratory Standards Institute (CLSI) guidelines (16, 17). Testing was performed using customized frozen 96-well trays provided by International Health Management Associates, Inc. (Schaumburg, IL). The range of concentrations of the agents tested and current MIC interpretative breakpoints are shown in Table S1 in the supplemental material. Cefiderocol was tested in iron-depleted cation-adjusted Mueller-Hinton (MH) broth, and MICs were read as the concentration in the first drug well in which the growth was significantly reduced (i.e., to a button of <1 mm or light/faint turbidity) relative to the growth observed in the growth control well containing the same medium; trailing endpoints were disregarded (16, 18). All other agents were tested in cation-adjusted MH broth. Appropriate quality control (QC) strains were tested on each day of testing using appropriate reference strains, including Escherichia coli ATCC 25922 and Pseudomonas aeruginosa ATCC 27853 for QC for cefiderocol (16).
A total of 1,086 Gram-negative clinical isolates were tested, including 834 Enterobacteriaceae isolates and 252 nonfermenters. The Enterobacteriaceae tested included Klebsiella pneumoniae (n = 794), E. coli (n = 35), Citrobacter freundii (n = 1), and Enterobacter species (n = 4); the resistance mechanisms present included blaKPC in 737 isolates, blaNDM in 28 isolates, blaOXA-48-like in 7 isolates, blaNDM and blaOXA-48-like in 1 isolate, and extended-spectrum β-lactamases (ESBLs) or AmpCs in 43 isolates; 18 isolates lacked ESBLs or carbapenemases. The Enterobacteriaceae group included 700 recent, carbapenem-resistant, clinical isolates of K. pneumoniae from the Great Lakes region (19). In addition, there were 116 clinical strains of various enterobacterial species selected from a reference collection with various β-lactamases and included the 18 isolates without ESBLs or carbapenemases. The nonfermenter isolates tested were clinical isolates from a worldwide collection of Acinetobacter baumannii complex isolates (n = 200; 101 were resistant to at least one carbapenem), S. maltophilia (n = 25 with blaL1), and P. aeruginosa (n = 27) isolates with blaVIM, blaPDC, and/or porin OprD deletions from hospitals in the Cleveland, OH, area (6, 20–22). All A. baumannii complex isolates and a subset of the K. pneumoniae isolates were sequenced by whole-genome sequencing (WGS) using paired-end NexteraXT libraries and an Illumina NextSeq sequencer (2× 150 bp) to ∼100-fold coverage. Reads were assembled using SPAdes software (23), annotated using NCBI’s PGAP pipeline (24), and deposited in the NCBI SRA and GenBank whole-genome sequencing (WGS) repositories (BioProject accession numbers PRJNA384060 and PRJNA384065). Resistance mechanisms in isolates whose whole genomes were not sequenced were characterized by PCR amplification and sequencing of the KPC β-lactamase genes, if present, as previously reported (4, 25). In addition, a modified carbapenemase multiplex PCR that detects blaIMP, blaNDM, blaOXA-48-like, and blaVIM was employed (26). The P. aeruginosa and S. maltophilia β-lactamase content was previously determined in other studies (20–22). In this manner, we identified the relevant β-lactamases for the entire collection of 1,086 Gram-negative clinical isolates. The results for each organism group are shown as MIC ranges, MIC50 and MIC90 values, and the percentage of isolates susceptible to each agent for cefiderocol and comparator agents in Tables 1 to 4.
Activity of antimicrobial agents tested against Enterobacteriaceaeb
Enterobacteriaceae.Cefiderocol MICs ranged from ≤0.03 to >64 mg/liter, with the MIC90 being 4 mg/liter (Table 1). The rates of susceptibility of the comparator agents ranged from 3.7% (ceftazidime; MIC90, >64 mg/liter) to 96.6% (ceftazidime-avibactam; MIC90, 2 mg/liter) and 98.4% (tigecycline; MIC90, 1 mg/liter). Analysis of these results based on β-lactam resistance mechanisms showed a cefiderocol MIC90 of ≤0.03 mg/liter for isolates without ESBLs or carbapenemases versus a cefiderocol MIC90 of from 1 to 8 mg/liter for isolates with ESBLs, carbapenemases, or AmpCs. There was no obvious association between the resistance mechanism and the MIC, although there was a bimodal MIC distribution, with peaks at 0.25 and 2 mg/liter (Table 2 and Fig. 1). In contrast, ceftazidime-avibactam susceptibility was related to β-lactam resistance mechanisms, with only 3/28 isolates (10.7%) with blaNDM being susceptible and 735/738 isolates (99.6%) with blaKPC being susceptible.
Activity of cefiderocol against Enterobacteriaceae with characterized ESBLs and carbapenemases
MIC distributions of meropenem (left) and cefiderocol (right) against Enterobacteriaceae. Arrows indicate the upper limit of susceptible and proposed susceptible MIC ranges for meropenem and cefiderocol, respectively. AmpC, CMY; OXA, OXA-48-like; TEM/SHV, TEM ESBL and/or SHV ESBL.
A. baumannii complex.Cefiderocol MICs ranged from ≤0.03 to >64 mg/liter, with the overall MIC90 being 1 mg/liter, and there was little difference between the carbapenem-susceptible and -resistant groups (Table 3; Fig. 2 and 3). Other agents active against both groups were colistin (to which 90.0% of isolates were susceptible) and tigecycline (to which 93% of isolates were susceptible). Agents more active against the carbapenem-susceptible group than against the carbapenem-resistant group included amikacin (94.9% versus 17.8%), ciprofloxacin (83.7% versus 0%), ceftazidime (86.9% versus 4.0%), ceftazidime-avibactam (64.6% versus 6.9%), and meropenem (100% versus 0%).
Activity of antimicrobial agents tested against carbapenem-susceptible and -resistant Acinetobacter baumannii complex isolatesa
MIC distributions of cefiderocol against carbapenem-susceptible and -resistant A. baumannii complex isolates.
Cumulative MICs of cefiderocol and comparator agents against A. baumannii complex isolates.
P. aeruginosa.Cefiderocol MICs ranged from ≤0.03 to 1 mg/liter, with the MIC90 being 0.5 mg/liter (Table 4 and Fig. 4). Notably, the MICs against isolates with blaVIM ranged from 0.06 to 1 mg/liter, with the MIC90 being 0.5 mg/liter. Colistin was the only comparator tested with good activity (96.3% of isolates were susceptible), while amikacin was active against 48.1% of isolates, aztreonam 22.2%, and ceftolozane-tazobactam 25.9%, and <10% of isolates were susceptible to ciprofloxacin, cefepime, ceftazidime, ceftazidime-avibactam, or meropenem.
Activity of antimicrobial agents tested against carbapenem-resistant P. aeruginosaa and S. maltophiliab
MIC distribution of cefiderocol against P. aeruginosa (left) and S. maltophilia (right).
S. maltophilia.Cefiderocol MICs ranged from ≤0.03 to 0.25 mg/liter, with the MIC90 being 0.25 mg/liter (Table 4 and Fig. 4). Colistin was active against 68.0% of isolates, while the MIC90 of tigecycline was 2 mg/liter. Amikacin and the other β-lactam agents tested had little activity, as expected, against this intrinsically aminoglycoside- and carbapenem-resistant species.
Several studies on the in vitro activity of cefiderocol against carbapenem-resistant species have recently been published, and our findings are in general agreement with the findings of these studies (10, 21–24, 28). However, overall, cefiderocol MIC90s against Enterobacteriaceae were lower (≤1 mg/liter) in two studies: (i) in the study of Kohira et al. against a 2009 to 2011 global collection of 617 Enterobacteriaceae isolates, although the MICs were up to 4 mg/liter against 226 of 233 strains (97.0%) with characterized β-lactamases, including 116 isolates with blaKPC, blaSME, or blaNDM (27), and (ii) in the study of Hackel et al. (29) against a U.S. and European collection of 6,087 Enterobacteriaceae isolates collected between 2014 and 2016, although the MIC90 was 4 mg/liter against 169 meropenem-nonsusceptible Enterobacteriaceae isolates.
Our findings on the in vitro activity of cefiderocol are in alignment with those in the publications discussed above for Enterobacteriaceae (MIC90, 4 mg/liter), P. aeruginosa (MIC90, 0.5 mg/liter), and S. maltophilia (MIC90, 0.25 mg/liter). Uniquely, our analysis also provides a correlation between the activity of cefiderocol and β-lactam resistance mechanisms; the activity against Enterobacteriaceae was affected by various β-lactamases, while the activity against nonfermenters was independent of the presence of β-lactamases, including carbapenemases. The wide distribution of MIC values for Enterobacteriaceae found here has been observed in other studies and is independent of the medium used, including iron-depleted-cation-adjusted (CA) MH broth (approved by CLSI), iron-depleted medium using the chelator ApoT, and non-iron-depleted CA MH broth (11, 12). This wide distribution is believed to be due to variations in iron transport channel expression, the primary mechanism for cell entry of siderophore antibiotic conjugates, which varies by species and within species (7, 11, 15).
For A. baumannii our MIC90 of 1 mg/liter was in agreement with the MIC90 values of 1 and 2 mg/liter, respectively, reported by Ito et al. (10) and Hackel et al. (29), while a later publication by Hackel et al. reported a higher MIC90 value of 8 mg/liter (30). The reason for this difference in activity against A. baumannii may be associated with regional differences in the resistance mechanisms of this species. Overall, our study showed the potent activity of cefiderocol against all isolates of P. aeruginosa and S. maltophilia tested and activity against most isolates of the Acinetobacter baumannii complex, with little difference between carbapenem-susceptible and -resistant isolates. Cefiderocol shows higher MICs against isolates with ESBLs (including the blaTEM ESBL, the blaSHV ESBL, and blaCTX-M) and carbapenemases (including blaKPC, blaNDM, and blaOXA-48-like) than against isolates of Enterobacteriaceae without ESBLs or carbapenemases.
Based on the proposed dosing regimen of cefiderocol and the PK/PD target attained at this dosing regimen, our findings support the in vivo activity of this agent against carbapenem-resistant Gram-negative nonfermenters and most carbapenem-resistant Enterobacteriaceae. Based on studies performed to date, cefiderocol may prove to be a particularly valuable addition to our limited armamentarium for combating infections caused by carbapenem-resistant Gram-negative nonfermenters.
Accession number(s).The reads obtained in this study have been deposited in the NCBI Sequence Read Archive and GenBank WGS repositories under BioProject accession numbers PRJNA384060 and PRJNA384065.
ACKNOWLEDGMENTS
Susceptibility test panels were provided by International Health Management Associates, Inc.
The research reported in this publication was supported by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health (NIH) under award number UM1AI104681. Additional NIH support to R.A.B. includes R01AI100560, R01AI063517, and R01AI072219. This study was also supported in part by funds and/or facilities provided by the Cleveland Department of Veterans Affairs, award number 1I01BX001974 to R.A.B. from the Biomedical Laboratory Research & Development Service of the VA Office of Research and Development, and the Geriatric Research Education and Clinical Center VISN 10. In addition, research support was also received from Shionogi.
The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the U.S. Department of Veterans Affairs.
FOOTNOTES
- Received 24 August 2018.
- Returned for modification 8 September 2018.
- Accepted 6 October 2018.
- Accepted manuscript posted online 15 October 2018.
Supplemental material for this article may be found at https://doi.org/10.1128/AAC.01801-18.
- Copyright © 2018 American Society for Microbiology.
