In Vitro Activity of the Siderophore Cephalosporin, Cefiderocol, against a Recent Collection of Clinically Relevant Gram-Negative Bacilli from North America and Europe, Including Carbapenem-Nonsusceptible Isolates (SIDERO-WT-2014 Study)

ABSTRACT Cefiderocol (formerly S-649266) is an investigational siderophore cephalosporin. Iron-depleted cation-adjusted Mueller-Hinton broth (ID-CAMHB) was prepared according to the Clinical and Laboratory Standards Institute (CLSI) protocol and used to perform broth microdilution testing of cefiderocol against a 2014-2015 collection of clinical isolates of Gram-negative bacilli from North America (n = 4,239) and Europe (n = 4,966). The concentrations of cefiderocol inhibiting 90% of isolates tested (MIC90s) were 0.5 μg/ml (North America; n = 3,007) and 1 μg/ml (Europe; n = 3,080) for all isolates of Enterobacteriaceae; 1 μg/ml (North America; n = 30) and 4 μg/ml (Europe; n = 139) for meropenem-nonsusceptible (MIC ≥ 2 μg/ml) isolates of Enterobacteriaceae; 0.5 μg/ml for both North American (n = 765) and European (n = 765) isolates of Pseudomonas aeruginosa; 0.5 μg/ml (North America; n = 151) and 1 μg/ml (Europe; n = 202) for meropenem-nonsusceptible (MIC ≥ 4 μg/ml) isolates of P. aeruginosa; 1 μg/ml for both North American (n = 309) and European (n = 839) isolates of all Acinetobacter baumannii strains as well as for both North American (n = 173) and European (n = 595) isolates of meropenem-nonsusceptible A. baumannii; and 0.5μg/ml (North America; n = 152) and 0.25 μg/ml (Europe; n = 276) for isolates of Stenotrophomonas maltophilia. MICs of cefiderocol were ≤4 μg/ml for 99.9% (6,078/6,087) of all Enterobacteriaceae, 97.0% (164/169) of meropenem-nonsusceptible Enterobacteriaceae, 99.9% (1,529/1,530) of all P. aeruginosa isolates, 100% (353/353) of meropenem-nonsusceptible P. aeruginosa isolates, 97.6% (1,120/1,148) of all A. baumannii isolates, 96.9% (744/768) of meropenem-nonsusceptible A. baumannii isolates, 100% of isolates of S. maltophilia (428/428) and 93.8% of isolates of Burkholderia cepecia (11/12). We conclude that cefiderocol demonstrated potent in vitro activity against a recent collection of clinical isolates of commonly encountered Gram-negative bacilli, including carbapenem-nonsusceptible isolates.

C arbapenems provide effective therapy for patients infected with extended-spectrum ␤-lactamase (ESBL)-producing Enterobacteriaceae and are considered agents of choice for many Gram-negative infections. Carbapenem-resistant Enterobacteriaceae have emerged or are in the process of emerging in many countries worldwide (1,2). Carbapenem resistance is also significant among nonfermenting pathogens such as Pseudomonas aeruginosa and Acinetobacter baumannii (3,4). Observed increases in carbapenem resistance among Enterobacteriaceae and nonfermenters are concerning, as there are few antimicrobial agents available that are safe and effective in the treatment of these infections and the development and distribution of new, more potent agents have not kept pace with increasing and diversifying resistance, particularly for commonly encountered pathogenic Gram-negative bacilli (5,6). Multidrugresistant phenotypes, including resistance to cephalosporins, aminoglycosides, and fluoroquinolones, are characteristic of most carbapenem-resistant isolates and present clinicians with difficult decisions to optimize therapy for patients (7). Carbapenem resistance in Gram-negative bacilli may arise by acquisition of class A (e.g., Klebsiella pneumoniae carbapenemase [KPC]), class B (e.g., NDM, IMP, VIM), or class D (e.g., OXA-48) ␤-lactamases (1,2,(7)(8)(9), by overproduction of chromosomal AmpC, or by acquisition of an ESBL in combination with a porin (e.g., OprD) deficiency or the presence of efflux pumps (9)(10)(11)(12).
Cefiderocol, formerly known as S-649226, is a novel siderophore cephalosporin for injection discovered by and currently in clinical development with Shionogi & Co., Ltd., to treat infections caused by carbapenem-resistant Gram-negative bacteria. Cefiderocol has a unique mechanism of cell entry. The cephalosporin moiety of cefiderocol binds primarily to bacterial penicillin binding protein 3 (PBP 3), similar to other cephalosporins, while the catechol moiety at the 3-position side chain of the cephalosporin contributes to forming a chelated complex with ferric iron that facilitates cefiderocol's crossing of the outer membrane of Gram-negative bacilli using the receptor-mediated bacterial iron transport system (13,14). Bacterial iron transport systems accelerate and increase the influx of cefiderocol to the periplasmic space of Gram-negative bacteria, where its cephalosporin moiety can inhibit cell wall synthesis, thereby enhancing its antimicrobial activity relative to carbapenems, ␤-lactam/␤-lactamase inhibitor combinations, and advanced-generation cephalosporins (14,15). Cefiderocol has been reported to be active against carbapenem-resistant Gram-negative bacilli harboring various carbapenemases and to be more stable than other ␤-lactam agents such as ceftazidime, cefepime, and meropenem against class A, B, and D carbapenemases such as KPC, VIM, IMP, NDM, and OXA (15)(16)(17). Cefiderocol is also active against ESBLproducing Escherichia coli and Klebsiella pneumoniae (15) as well as against meropenemresistant P. aeruginosa and A. baumannii (18).
Accurate in vitro susceptibility testing of cefiderocol by broth microdilution requires the use of iron-depleted conditions because such conditions induce the production of ferric iron transporters, which are strongly regulated by proximal iron concentrations. Iron-depleted conditions mimic the conditions faced by bacteria infecting human tissues and fluids (15,17,19). In January 2016, the Clinical and Laboratory Standards Institute (CLSI) Subcommittee on Antimicrobial Susceptibility Testing approved broth microdilution and disk diffusion methods and quality control MIC ranges for cefiderocol (20). Broth microdilution testing of cefiderocol requires iron-depleted cation-adjusted Mueller-Hinton broth (ID-CAMHB) (20). MICs determined using ID-CAMHB have been shown to be reproducible and to correlate well with in vivo efficacy in animal models (21)(22)(23). MIC assays for cefiderocol performed in CAMHB with iron concentrations of Ͼ0.03 g/ml show highly variable results and do not correlate with in vivo efficacy (22).
The current study, SIDERO-WT-2014, tested a 2014-2015 collection of 9,205 clinical isolates of Gram-negative bacilli from patients in North America and Europe against cefiderocol and comparators using CLSI broth microdilution methodology. This study generated the first in vitro surveillance testing data for cefiderocol using the recently approved CLSI broth microdilution MIC determination method (20).

RESULTS
The in vitro activities of cefiderocol and comparators are summarized in Table 1 for  the 4,239 isolates collected from North American medical center laboratories and in  Table 2 for the 4,966 isolates from European medical center laboratories. The concentrations of antimicrobial agent inhibiting 90% of isolates tested (MIC 90 s) for cefiderocol against Enterobacteriaceae were 0.5 g/ml (North America, n ϭ 3,007 isolates) and 1 g/ml (Europe, n ϭ 3,080 isolates). The MIC range for cefiderocol was Յ0.002 to 8 g/ml for both sets of isolates, and 99.9% (6,078/6,087) of all Enterobacteriaceae had cefiderocol MICs of Յ4 g/ml. One isolate of Serratia marcescens from North America had a MIC to cefiderocol of 8 g/ml, while six isolates of K. pneumoniae, one isolate of Enterobacter aerogenes, and one isolate of S. marcescens from Europe also had a MIC to cefiderocol of 8 g/ml. Five of the nine isolates with cefiderocol MICs of 8 g/ml were nonsusceptible to meropenem. Against meropenem-nonsusceptible (MIC Ն 2 g/ml) isolates from North America (Fig. 1) and Europe (Fig. 2), MIC 90 values were 1 g/ml (n ϭ 30) and 4 g/ml (n ϭ 139); 97.0% (164/169) of all meropenem-nonsusceptible Enterobacteriaceae had MICs to cefiderocol of Յ4 g/ml. In comparison, testing of recently approved antimicrobial agents ceftazidime-avibactam and ceftolozane-tazobactam against the same geographic sets of isolates of meropenem-nonsusceptible Enterobacteriaceae demonstrated MIC 90 values of 4 and Ͼ64 g/ml, respectively, for isolates from North America (Table 1) and Ͼ64 and Ͼ64 g/ml, respectively, for isolates from Europe (Table 2). Against meropenem-susceptible (MIC Յ1 g/ml) isolates from North America and Europe, MIC 90 values for cefiderocol were 0.5 g/ml (n ϭ 2,977) and 1 g/ml (n ϭ 2,941); 99.9% (5,914/5,918) of all meropenem-susceptible Enterobacteriaceae had MICs to cefiderocol of Յ4 g/ml. Figure 3 depicts the cumulative percentages of all meropenem-nonsusceptible isolates of Enterobacteriaceae that were susceptible to increasing concentrations of cefiderocol and its comparators.
The MIC 90 values for cefiderocol against P. aeruginosa were 0.5 g/ml (North America, n ϭ 765 isolates) and 0.5 g/ml (Europe, n ϭ 765 isolates); 99.9% (1,529/1,530) of all P. aeruginosa isolates had MICs of Յ4 g/ml. The one isolate with a MIC to cefiderocol of 8 g/ml was from North America. Against meropenem-nonsusceptible (MIC Ն 4 g/ml) isolates from North America (Fig. 1) and Europe (Fig. 2), MIC 90 values for cefiderocol were 0.5 g/ml (n ϭ 151) and 1 g/ml (n ϭ 202); all 353 isolates of P. aeruginosa that were meropenem nonsusceptible had MICs to cefiderocol of Յ4 g/ml. In comparison, the MIC 90 values for ceftazidime-avibactam and ceftolozane-tazobactam against isolates of meropenem-nonsusceptible P. aeruginosa from North America (Table  1) and Europe (Table 2) were 8 and 4 g/ml and 64 and Ͼ64 g/ml, respectively. Against meropenem-susceptible (MIC Յ 2 g/ml) isolates from North America and Europe, MIC 90 values for cefiderocol were 0.5 g/ml (n ϭ 614) and 0.5 g/ml (n ϭ 563); 99.9% (1,176/1,177) of all meropenem-susceptible isolates of P. aeruginosa had MICs to cefiderocol of Յ4 g/ml. Figure 4 depicts the cumulative percentages of all meropenem-nonsusceptible isolates of P. aeruginosa that were susceptible to increasing concentrations of cefiderocol and its comparators.
The MIC 90 values for cefiderocol against A. baumannii were 1 g/ml for both North American (n ϭ 309) and European (n ϭ 839) isolates; 97.6% (1,120/1,148) of all A. baumannii had MICs to cefiderocol of Յ4 g/ml. Of the 28 isolates with cefiderocol MIC values of Ͼ4 g/ml, 25 were from European medical laboratories (18 isolates from two sites in Russia; 6 isolates from one site in Turkey; and 1 isolate from Sweden) and 3 isolates were from two sites in the United States. Twenty-four of these 28 isolates (85.7%) were also nonsusceptible to meropenem. Against meropenem-nonsusceptible isolates of A. baumannii from North America ( Fig. 1; n ϭ 173) and Europe ( Fig. 2; n ϭ 595), MIC 90 values were 1 g/ml for both data sets; 96.9% (744/768) of meropenemnonsusceptible A. baumannii isolates had MICs to cefiderocol of Յ4 g/ml. Against meropenem-susceptible isolates from North America and Europe, MIC 90 values for cefiderocol were 0.25 g/ml (n ϭ 136) and 0.25 g/ml (n ϭ 244); 99.0% (376/380) of meropenem-nonsusceptible A. baumannii isolates had MICs to cefiderocol of Յ4   Figure 5 depicts the cumulative percentages of all meropenem-nonsusceptible isolates of A. baumannii that were susceptible to increasing concentrations of cefiderocol and its comparators. The MIC 90 values for cefiderocol against Stenotrophomonas maltophilia were 0.5 g/ml (North America, n ϭ 152 isolates) and 0.25 g/ml (Europe, n ϭ 276 isolates); all isolates of S. maltophilia had MICs to cefiderocol of Յ4 g/ml, while the MIC 90 s for cefepime, ceftazidime-avibactam, ceftolozane-tazobactam, and meropenem were Ն64 g/ml and for colistin and ciprofloxacin were Ն8 g/ml. There are no published CLSI breakpoints for S. maltophilia for any of the other antimicrobial agents tested in this study.

DISCUSSION
Carbapenem-resistant Enterobacteriaceae, P. aeruginosa, and A. baumannii are frequently multidrug resistant. Currently, there are very few antimicrobial agents available to clinicians to treat patients infected with carbapenem-resistant Gram-negative bacilli, and the few agents that are accessible to treat systemic infections are associated with   Activity of Cefiderocol against Gram-Negative Bacilli Antimicrobial Agents and Chemotherapy approved in several countries, but neither of these agents is active against isolates producing class B metallo-␤-lactamases (1,2). In the current study, meropenemnonsusceptible isolates of Enterobacteriaceae and P. aeruginosa challenged the in vitro activities of ceftazidime-avibactam and ceftolozane-tazobactam, particularly isolates from European medical center laboratories ( Table 2). The current study demonstrated that the in vitro activity of the novel siderophore cephalosporin, cefiderocol, was superior to that of comparators against recent clinical isolates of meropenem-nonsusceptible Enterobacteriaceae, P. aeruginosa, and A. baumannii from North America and Europe (Fig. 3 to 5), including isolates that were resistant to colistin and the ␤-lactam/␤-lactamase inhibitor combinations ceftazidimeavibactam and ceftolozane-tazobactam. Cefiderocol also demonstrated potent activity against S. maltophilia, while all six comparators were inactive. Cefiderocol exhibited MIC 90 s against P. aeruginosa and S. maltophilia that were 4 to Ͼ64 times lower than those of comparator agents. Against A. baumannii, cefiderocol (MIC 90 , 1 g/ml) was up to 64 times more potent than the comparator agents tested, with the exception of colistin, which also had an MIC 90 of 1 g/ml.
The mechanism(s) responsible for elevated MICs for cefiderocol in the limited number of isolates that have been observed are currently unknown. If one were to hypothesize a mechanism of resistance to cefiderocol, it would likely involve either a reduction in production of one or more components of the iron transport system or one or more mutations in the binding site for the iron transport system on the outer membrane of Gram-negative bacteria, as this has been reported previously for other siderophore ␤-lactams (25,26). The adaption-based resistance to other siderophoreconjugated antibacterial agents, such as MB-1, attributed to competition with native siderophores in P. aeruginosa (26), has not been observed for cefiderocol tested against isolates of P. aeruginosa, A. baumannii, or Enterobacteriaceae (21). The mechanism(s) of resistance to cefiderocol that appeared in the small subset of isolates with higher MICs to cefiderocol (Ն8 g/ml) in the current study requires future study.
Cefiderocol has been tested in several animal models and has shown in vivo efficacy against ESBL-producing, KPC-producing, and multidrug-resistant isolates of Gramnegative bacilli (21,23,27). In a neutropenic murine thigh infection model, cefiderocol treatment consistently generated CFU reductions of Ͼ1 log and sustained antibacterial effects against all isolates tested (23). In the same study, the effectiveness of cefiderocol was demonstrated to correlate well with the pharmacodynamics parameter % fTϾMIC (the percentage of a 24-hour period in which the unbound drug concentration exceeded the MIC) (23). Cefiderocol has also demonstrated efficacy against multidrugresistant and carbapenem-resistant isolates of P. aeruginosa, A. baumannii, and Enterobacteriaceae in a murine lung infection model (21) and against ESBL-producing or KPC-producing isolates of Enterobacteriaceae in various animal infection models (27). In pharmacokinetic modeling of human drug exposures using a dose of 2 g every 8 h infused over 3 h, cefiderocol achieves a 100% probability to target attainment for organisms with an MIC of 4 g/ml (28). Importantly, in a rat lung infection model that reproduced human pharmacokinetic and pharmacodynamics parameters, cefiderocol demonstrated bactericidal activity against carbapenem-resistant (KPC-and NDM-1positive) isolates of K. pneumoniae, with cefiderocol MICs as high as 4 g/ml (29). A clinical trial evaluating the efficacy and safety of cefiderocol versus imipenem or cilastatin in complicated urinary tract infections is currently ongoing (30) and will be followed by a clinical trial comparing cefiderocol to the best available therapies for the treatment of serious infections caused by carbapenem-resistant Gram-negative pathogens (31).
The current study tested cefiderocol, a promising, novel siderophore cephalosporin, using the CLSI-approved broth microdilution method with ID-CAMHB prepared by preincubation with Chelex 100 resin (20), and found it to demonstrate potent in vitro activity against Enterobacteriaceae, P. aeruginosa, A. baumannii, S. maltophila, and B. cepacia, including carbapenem-resistant isolates. A cefiderocol MIC of Յ4 g/ml was observed for 99.6% (9,166/9,205) of all isolates of Gram-negative bacilli tested in the current study. Previous studies that performed molecular characterization of isolates for ␤-lactamase genes have reported that cefiderocol is stable against a broad range of ␤-lactamases, including ESBL-, AmpC-, and carbapenemase-producing (class A, B, and D enzymes) isolates, including metallo-␤-lactamases (e.g., NDM-1, VIM, IMP), as well as against