Evaluation of in vitro activity of ceftolozane-tazobactam in combination with other classes of antibacterial agents against Enterobacterales and Pseudomonas aeruginosa-the EM200 study.

Ceftolozane-tazobactam is a cephalosporin/β-lactamase inhibitor combination developed for use against some β-lactam- and multidrug-resistant Gram-negative organisms. This study aimed to evaluate the in vitro activity of ceftolozane-tazobactam against clinical bacterial isolates at the University Hospital of Marrakech. This is a descriptive and analytical prospective study. A total of 143 Enterobacterales and 48 Pseudomonas aeruginosa isolates were collected from January 2018 to December 2018 from patients with respiratory, urinary and intra-abdominal infections. The identification was made by Phoenix automated system (BioMérieux). MIC50/90 were tested by broth microdilution for ceftolozane-tazobactam, and other drugs using dried panels. Antimicrobial susceptibility results were interpreted according to CLSI guidelines. Ceftolozane-tazobactam inhibited 98% of Escherichia coli (MIC50/90; 0.25/0.5 μg/mL). The susceptibility rate of Klebsiella pneumoniae to ceftolozane-tazobactam was 68.8% (MIC50/90, 0.5/>32 μg/mL); other Enterobacterales have shown susceptibility rates of 80.4% (MIC50/90; 0.5/8 μg/mL). In carbapenemase-producing K. pneumoniae, the bla OXA-48 mutation was found in two isolates. Susceptibility of P. aeruginosa to ceftolozane-tazobactam was 91.7% (MIC50/90, 0.5/>32 μg/mL). In non-carbapenemase-producing P. aeruginosa, AmpC mutations were found in all isolates. Ceftolozane-tazobactam was satisfactorily active against a wide range of tested isolates and offers clinicians a potential therapeutic option even against resistant strains in patients with intra-abdominal infections, urinary tract infections and nosocomial pneumonia.

Introduction

Enterobacterales are responsible for severe respiratory infection, urinary tract infection and intra-abdominal infection due to antibiotic resistance [1]. They are represented mainly by Escherichia coli, Klebsiella pneumoniae, Enterobacter cloacae and Proteus mirabilis [1]. Pseudomonas aeruginosa is a pathogen that has expanded in hospitals, and causes fatal infections [2]. Infectious diseases pose a serious threat to global health [1,2].

Antibiotics have saved millions of lives every year, but today many bacteria are multidrug-resistant. The excessive and irrational use of antibiotics leads to the adaptation of bacteria to develop resistance to antibiotics [3]. The European Center for the Control of Infectious Diseases evaluates 33 000 cases of deaths per year due to bacterial resistance [4]; the problem has become critical in hospitals, because of the therapeutic problems observed mainly in Gram-negative bacteria such as Enterobacterales and P. aeruginosa [3,5].

WHO has issued a global action plan to develop new antibiotics [6], like ceftolozane-tazobactam (Zerbaxa™), which is a combination of ceftolozane, a third-generation cephalosporin, and tazobactam, an inhibitor of several β-lactamases. It is a new antibiotic, with activity against multi-resistant Gram-negative bacteria. Zerbaxa™ was approved by the US Food and Drug Administration and by the European Medicines Agency for complicated intra-abdominal infections, serious kidney infections (acute pyelonephritis), complicated infections of the urinary tract and hospital-acquired pneumonia, including ventilator-associated pneumonia [7,8]. In vitro studies have shown previously that the potential value of ceftolozane-tazobactam lies in its activity against Enterobacterales producing extended-spectrum β-lactamases (ESBL) [9], as part of a carbapenem saving strategy [10], and against P. aeruginosa combining several mechanisms of antibiotic resistance (including efflux pumps and overexpression of AmpC) [11,12].

The aim of this study was to evaluate the in vitro activity of ceftolozane-tazobactam against Enterobacterales and P. aeruginosa collected from different infection sites in the Marrakech University Hospital.

Materials and methods

Type of study method

This is a descriptive and analytical prospective study, extended over 1 year from January 2018 to December 2018, in two university hospital centres in Morocco: Marrakech and Rabat.

Bacterial isolates

Gram-negative aerobic bacteria, comprising E. coli, Klebsiella spp., Enterobacter spp. and Proteus spp., and Gram-negative anaerobic bacteria, comprising only P. aeruginosa, were isolated from different infection sites including respiratory infections, urinary infections and intra-abdominal infections. They were not collected sequentially (isolated strains were stored at –80°C or −20°C). Bacteria were identified using the Phoenix® automated Microbiology Identification System (Becton Dickinson, Franklin Lakes, NJ, USA).

Antimicrobial susceptibility testing

Antimicrobial susceptibility testing was performed at International Health Management Associates (IHMA) report 2018 following the 2017 Clinical and Laboratory Standards Institute (CLSI) guidelines [13]. Minimum inhibitory concentrations (MIC) were interpreted according to CLSI MIC values for ceftolozane-tazobactam, ceftazidime, meropenem, cefepime, piperacillin-tazobactam, amikacin, ciprofloxacin, ertapenem, imipenem, ceftriaxone, levofloxacin and colistin, and were determined using broth microdilution method panels manufactured by TREK Diagnostic Systems (East Grinstead, UK) according to CLSI guidelines for antimicrobials. A suspension using colonies and normal saline, equivalent to 0.5 McFarland standard concentrations, was incubated at 35°C for 16–20 hours.

For isolates of Enterobacterales, ceftolozane-tazobactam MIC were considered susceptible at ≤2 μg/mL. For isolates of P. aeruginosa, ceftolozane-tazobactam MIC was considered susceptible at ≤4 μg/mL.

Whole-genome sequencing of non-susceptible P. aeruginosa and K. pneumoniae to ceftolozane-tazobactam

Strains with ceftolozane-tazobactam MIC values ≥ 8 μg/mL were examined.

Genomic DNA was extracted using Qiagen DNeasy UltraClean kits (Qiagen, Valencia, CA, USA) and quantified using the Nanodrop™ ND-1000 spectrophotometer (ThermoFisher Scientific, Waltham, MA, USA).

For whole-genome sequencing, DNA sequences were obtained on the Illumina HiSeq sequencing instrument (Illumina, San Diego, CA, USA) with 2 × 150 bp pair-end reads with a target coverage depth of approximately 150×. All analyses were carried out using Qiagen's CLCBio Genomics Workbench version 11.

For β-lactamase resistance, genes were identified by Illumina whole-genome sequencing. β-lactamase gene inclusion was 72% and 80% for minimum nucleotide sequence identity and minimum sequence length, respectively.

For porin gene identification, ompk35 and ompK36 in K. pneumoniae and oprD in P. aeruginosa were searched by TBLASTN; for ftsI (encoding PBP3) gene analysis, searching was on a species-specific basis in de novo assemblies of each genome.

The appropriate multilocus sequence typing scheme and allelic profile of each of the guided assemblies was determined computationally (using the ‘find best match using k-mer spectra’ tool in CLC genomics).

Statistical analysis

Statistical analysis for comparison of susceptibilities of different isolates to ceftolozane-tazobactam was performed using SPSS 18 software (SPSS Inc., Chicago, IL, USA).

Results

The isolates comprised 142 Enterobacterales (49 E. coli, 48 K. pneumoniae, 16 Enterobacter spp. and 16 Proteus spp., 8 Klebsiella oxytoca, 3 Enterobacter aerogenes, 1 Klebsiella variicola and 1 Proteus vulgaris), and 48 P. aeruginosa. The clinical isolates were collected from intensive care units (57.4%, n = 109) and from non-intensive care units (46.6%, n = 81). The 190 Gram-negative isolates included 113 isolates from urinary tract infections, 56 from intra-abdominal infections and 21 from lower respiratory tract infections. Table 1 shows the antibiotics assessed, and the range and MIC values of each drug. Table 2 shows the susceptibility profile of different organisms to several antibiotics and their MIC50/90.

Table 1

MIC values distribution for all antimicrobials tested against Enterobacterales and Pseudomonas aeruginosa isolates

Organisms	Antimicrobial agent	MICs (μg/mL)
		0.06	0.12	0.25	0.5	1	2	4	8	16	32	64
Escherichia coli (n = 49)	Ceftolozane-tazobactam		7	26	15	0	0	1	0	0	0
	Piperacillin-tazobactam						30	5	6	4	1	3
	Cefoxitin						4	16	23	6
	Cefotaxime					36	1	0	0	3	9
	Ceftriaxone					36	0	0	0	2	11
	Ceftazidime					38	1	2	2	2	4
	Cefepime					37	2	0	4	0	6
	Meropenem	49	0	0	0	0	0	0	0
	Imipenem				48	1	0	0	0	0	0
	Ertapenem	48	0	1
	Aztreonam					37	1	2	1	8
	Amikacin							32	13	4
	Ciprofloxacin			32	2	15
	Levofloxacin					33	0	16
	Colistin		49	0	0
Klebsiella pneumoniae (n = 48)	Ceftolozane-tazobactam		1	7	16	6	3	1	7	0	7
	Piperacillin-tazobactam						6	8	6	5	9	14
	Cefoxitine						5	26	8	9
	Cefotaxime					13	2	1	0	0	32
	Ceftriaxone					15	0	0	1	0	32
	Ceftazidime					13	1	1	3	8	22
	Cefepime					16	0	0	3	11	18
	Ertapenem	27	3	6	1	1	1	9
	Imipenem				31	7	4	3	1	0	2
	Meropenem	36	1	1	0	3	4	1	2
	Aztreonam					13	2	1	1	31
	Amikacin							45	2	1
	Ciprofloxacin			20	3	25
	Levofloxacin					30	2	16
	Colistin					42	0	6
Enterobacter cloacae (n = 16)	Ceftolozane-tazobactam			4	3	0	0	3	3	1	2
	Piperacillin-tazobactam						6	1	0	0	1	8
	Cefoxitine							1	0	15
	Cefotaxime					6	1	0	0	0	9
	Ceftriaxone					7	0	0	0	0	9
	Ceftazidime					7	0	0	0	0	9
	Cefepime					8	5	1	0	0	2
	Ertapenem	7	0	0	1	4	2	2
	Imipenem				14	0	0	2	0	0	0
	Meropenem		13	1	0	0	0	1	1	0
	Aztreonam					7	0	0	0	9
	Amikacin							16	0	0	0
	Ciprofloxacin			6	0	10
	Levofloxacin					9	1	6
	Colistin					16	0	0
Proteus mirabilis (n = 16)	Ceftolozane-tazobactam			6	9	1	0	0	0	0	0
	Piperacillin-tazobactam					16	0	0	0	0	0
	Cefoxitine						9	6	0	1
	Cefotaxime					16	0	0	0	0	0
	Ceftriaxone					16	0	0	0	0	0
	Ceftazidime					16	0	0	0	0	0
	Cefepime					16	0	0	0	0	0
	Ertapenem	16	0	0	0	0	0	0	0
	Imipenem				3	10	3	0	0	0	0
	Meropenem	16	0	0	0	0	0	0	0
	Aztreonam					16	0	0	0	0	0
	Amikacin							14	2	0	0
	Ciprofloxacin				1	0	0
	Levofloxacin					1	0	0
Pseudomonas aeruginosa (n = 48)	Ceftolozane-tazobactam				24	11	1	6	0	0	6
	Piperacillin-tazobactam							12	19	4	1	12
	Cefepime					3	19	8	4	4	10
	Ceftazidime					0	18	15	2	5	8
	Meropenem		10	10	0	7	1	2	5	3
	Imipenem			1	4	26	7	1	1	4	5
	Aztreonam						1	26	8	13
	Amikacin							39	4	1	4
	Ciprofloxacin			35	4	0	9
	Levofloxacin					35	5	8
	Colistin					47	1	0
Other Enterobacterales (n = 14)	Ceftolozane-tazobactam			11	3
	Piperacillin-tazobactam						9	5
	Cefoxitin						3	7	1	3
	Cefotaxime					14
	Ceftazidime					14
	Ceftriaxone					14
	Cefepime					14
	Ertapenem	14
	Imipenem				11	3
	Meropenem		14
	Aztreonam					14
	Amikacin							14
	Ciprofloxacin			14
	Levofloxacin					14
	Colistin					13		1

Table 2

MIC50/90 and percentage susceptible of antimicrobials tested against 190 isolates

Organism	Antimicrobial agent	%S CLSIa	MIC50b	MIC90b)	Range
Pseudomonas aeruginosa (n = 48)	Ceftolozane-tazobactam	87.5	0.5	>32	0.5 to >32
	Amikacin	91.7	≤4	16	≤4 to >32
	Aztreonam	72.9	4	>16	2 to >16
	Cefepime	70.8	4	32	≤1 to >32
	Ceftazidime	72.9	4	>32	2 to >32
	Ciprofloxacin	81.3	≤0.25	>2	≤0.25 to >2
	Colistin	100	≤1	≤1	≤1 to 2
	Imipenem	77.1	1	32	≤0.5 to >32
	Levofloxacin	83.3	≤1	>4	≤1 to >4
	Meropenem	79.2	0.5	8	≤0.12 to >16
	Piperacillin-tazobactam	72.9	8	>64	4 to >64
Escherichia coli (n = 49)	Ceftolozane-tazobactam	98.0	0.25	0.5	0.12–4
	Amikacin	100	≤4	8	≤4 to 16
	Aztreonam	81.6	≤1	>16	≤1 to >16
	Cefepime	79.6	≤1	>32	≤1 to >32
	Cefotaxime	73.5	≤1	>32	≤1 to >32
	Cefoxitin	87.8	8	16	≤2 to >16
	Ceftazidime	83.7	≤1	16	≤1 to >32
	Ceftriaxone	73.5	≤1	>32	≤1 to >32
	Ciprofloxacin	67.4	≤0.25	>2	≤0.25 to >2
	Colistin	100	≤1	≤1	≤1 to ≤1
	Ertapenem	100	≤0.06	≤0.06	≤0.06 to 0.25
	Imipenem	100	≤0.5	≤0.5	≤0.5 to 1
	Levofloxacin	67.4	≤1	>4	≤1 to >4
	Meropenem	100	≤0.12	≤0.12	≤0.12 to ≤0.12
	Piperacillin-tazobactam	91.8	≤2	16	≤2 to 64
Klebsiella pneumoniae (n = 48)	Ceftolozane-tazobactam	68.8	0.5	>32	0.12 to >32
	Amikacin	100	≤4	≤4	≤4 to 16
	Aztreonam	33.3	>16	>16	≤1 to >16
	Cefepime	33.3	16	>32	≤1 to >32
	Cefotaxime	27.1	>32	>32	≤1 to >32
	Cefoxitin	81.3	4	>16	≤2 to >16
	Ceftazidime	31.3	16	>32	≤1 to >32
	Ceftriaxone	31.3	>32	>32	≤1 to >32
	Ciprofloxacin	47.9	2	>2	≤0.25 to >2
	Colistin	87.5	≤1	>4	≤1 to >4
	Ertapenem	77.1	≤0.06	>4	≤0.06 to >4
	Imipenem	79.2	≤0.5	4	≤0.5 to >32
	Levofloxacin	66.7	≤1	>4	≤1 to >4
	Meropenem	79.2	≤0.12	4	≤0.12 to >16
	Piperacillin Tazobactam	52.1	16	>64	≤2 to >64
Other 22Enterobacterales (n = 46)	Ceftolozane-tazobactam	80.4	0.5	8	0.25 to >32
	Amikacin	100	≤4	≤4	≤4 to 8
	Aztreonam	80.4	≤1	>16	≤1 to >16
	Cefepime	93.5	≤1	2	≤1 to 32
	Cefotaxime	76.1	≤1	>32	≤1 to >32
	Cefoxitin	58.7	4	>16	≤2 to >16
	Ceftazidime	80.4	≤1	>32	≤1 to >32
	Ceftriaxone	78.3	≤1	>32	≤1 to >32
	Ciprofloxacin	71.7	≤0.25	>2	≤0.25 to >2
	Colistin	63.0	≤1	>4	≤1 to >4
	Ertapenem	82.6	≤0.06	1	≤0.06 to >4
	Imipenem	89.1	≤0.5	2	≤0.5 to 4
	Levofloxacin	80.4	≤1	4	≤1 to >4
	Meropenem	95.7	≤0.12	≤0.12	≤0.12 to 8
	Piperacillin Tazobactam	80.4	≤2	64	≤2 to >64

%S, represents the percent susceptible by CLSI 2017 guidelines (EUCAST guidelines for colistin were applied for Enterobacterales).

MIC50, MIC90, and range in μg/mL; no intermediate breakpoint.

Ceftolozane-tazobactam activity against Enterobacterales

The activity of ceftolozane-tazobactam against E. coli was high (MIC50/90; 0.25/0.5 μg/mL). Ceftolozane-tazobactam susceptibility rate against E. coli was 98%. Susceptibility rates to other antibiotics were 100% to carbapenem; 100% to amikacin and 100% to colistin (Table 3).

Table 3

Whole-genome sequencing data for Klebsiella pneumoniae examined

MIC C/T (μg/mL)	OmpK 35	OmpK 36	PBP3-ftsIa	β-lactamase summary (72% ident, 80% coverage)b	MLST
8	No lesion	Lesion	WT	CTX-M-15; OXA-1; SHV-1; TEM-1B	628
>32	No lesion	No lesion	WT	CTX-M-15; OXA-48; SHV-1-like	Novel-2
>32	Lesion	No lesion	WT	CTX-M-15; OXA-1; OXA-48; SHV-11	37
8	No lesion	Lesion	WT	CTX-M-15; OXA-1; SHV-11-like; TEM-1B-like	392

Abbreviations: MLST, multilocus sequence typing; WT, wild-type.

PBP3-ftsl: Peniclin Binding Porin 3- FtsI gene.

Threshold for β-lactamase gene inclusion was 72% and 80% for minimum nucleotide sequence identity and minimum sequence length, respectively.

Susceptibility rate of K. pneumoniae to ceftolozane-tazobactam was 68.8%, with medium activity (MIC50/90, 0.5/>32 μg/mL), with higher susceptibility to amikacin (100%), colistin (87.5%), imipenem (79.2%), meropenem (79.2%) and ertapenem (77.1%) (Table 2).

Ceftolozane-tazobactam showed 80.4% inhibition (MIC50/90; 0.5/8 μg/mL) against Enterobacterales (Enterobacter cloacae, Proteus mirabilis, K. oxytoca, K. variicola, Enterobacter aerogenes, Proteus vulgaris) compared with susceptibilities to colistin of 63%, to meropenem of 89.1% and to amikacin of 100%.

From 15 ceftolozane-tazobactam non-susceptible K. pneumoniae isolates, molecular analysis of the resistance support was detected in 4/15 (27%). The carbapenemase gene blaOXA-48 was found in two isolates. ESBL production was confirmed in all isolates—blaCTX-M-15 (four of four), blaSHV-1 (two of four), blaOXA-1 (two of four) and blaTEM-1B (two of four). Mutations of ompK35 and ompK36 were detected in three of the isolates tested (Table 3).

Ceftolozane-tazobactam activity against P. aeruginosa

The susceptibility rate to ceftolozane-tazobactam against P. aeruginosa was 91.7%, with (MIC50/90 values of 0.5/>32 μg/mL) (Table 3). Colistin was the most active drug with 100% of isolates susceptible (MIC50/90, ≤1/≤1 μg/mL); susceptibility to amikacin-cefepime was 91.7%.

All six ceftolozane-tazobactam-resistant P. aeruginosa isolates produced PDC (Ampc mutation) and blaOXA-50; four had the β-lactamases genes blaPER-1 and blaVIM-2, and two had blaOXA-4. Mutation of oprD was detected in three isolates; all isolates were wild-type for ftsI (Table 4).

Table 4

Whole-genome sequencing data for Pseudomonas aeruginosa isolates examined

C/T MIC (μg/mL)	oprD	PBP3(ftsI)	WGS β-lactamase summarya	Class C (intrinsic)	MLSTb
>32	Lesion	WT	PDC-252-like; OXA-50-like; PER-1; VIM-2; OXA-4	PDC-252-like	233
>32	No lesion	WT	PDC-40-like; VIM-2; OXA-50-like	PDC-40-like	270
>32	No lesion	WT	PDC-72-like; OXA-50-like; PER-1	PDC-72-like	277
>32	Lesion	WT	PDC-119-like; VIM-2; PER-1; OXA-4; OXA-50-like	PDC-119-like	233
>32	No lesion	WT	PDC-183-like; OXA-50-like; VIM-2	PDC-183-like	769
>32	Lesion	WT	PDC-252-like; OXA-50-like; PER-1	PDC-252-like	233

Abbreviations: MLSTb, multilocus sequence typing; WGSa, whole-genome sequencing; WT, wild-type.

Discussion

Ceftolozane-tazobactam could be an important treatment option, including against multidrug-resistant strains. This type of study is interesting to evaluate the activity of ceftolozane-tazobactam against a selection of P. aeruginosa and Enterobacterales isolates. Our study is the first in Morocco.

In the current study, ceftolozane-tazobactam was active against 82.4% of Enterobacterales: 98% of E. coli, 68.8% of K. pneumonia and 80.4% of other Enterobacterales, which is similar to other studies assessing the activity of ceftolozane-tazobactam against Gram-negative isolates. Karlowsky et al. [14] reported a susceptibility of 89.7% of Enterobacterales, Kuo et al. [15] found 81.9% of K. pneumoniae and 91.9% of E. coli, Sader et al. [16] reported 98.5% of E. coli and 89.6% of K. pneumoniae, and Shortridge et al. [17] reported 95.5% of Enterobacterales. The variation in susceptibility profiles to ceftolozane-tazobactam can be explained by intrinsic and extrinsic mechanisms.

In our study, the susceptibility of K. pneumoniae to ceftolozane-tazobactam was the weakest. This was due to the incidence of carbapenemase and ESBL in this microorganism, most of them having a high rate of ceftolozane hydrolysis, such as blaOXA-48, blaCTX-M15, SHV and TEM. This is in accord with the results of Tuon et al. [18]. Carbapenem use results in heightened rates of carbapenem-resistant infections, limiting treatment options and growing mortality. Carbapenem resistance and ESBL detected in Enterobacterales, usually due to plasmid β-lactamases enzymes, have also become important issues [19].

In this study, the susceptibility of ceftolozane-tazobactam was 87.5% against P. aeruginosa, similar to reported susceptibility rates above 80% [[15], [16], [17],20,21]. Ceftolozane-tazobactam was the third most active of the β-lactam agents tested against P. aeruginosa after colistin and amikacin-meropenem. The same result was reported by Garcia-Fernandez et al. [22], who found a susceptibility rate of 91.3%, which was the third most active, after colistin (95.0%) and amikacin (93.8%). The activity of ceftolozane-tazobactam is specifically more important in the context of intensive care units, where this microorganism is an extreme worry in the management of nosocomial infections because of its resistance to different antibiotics.

Our results show the presence of intrinsic mechanisms in ceftolozane-tazobactam-resistant P. aeruginosa isolates by producing AmpC overexpression by single point mutations in the blaPDC gene (AmpC gene), and of extrinsic mechanisms by the presence of metallo-β-lactamase (VIM), associated with mutation of porin (OprD). This result is consistent with four studies demonstrating the presence of a carbapenemase in ceftolozane-tazobactam-resistant P. aeruginosa [[23], [24], [25], [26]]. Ceftolozane-tazobactam has a better safety profile compared with colistin, which has high frequencies of nephrotoxicity, neurotoxicity, and allergic and topical reactions, as do aminoglycosides [27].

The main strength of this study was the genome sequencing of ceftolozane-tazobactam-resistant P. aeruginosa and K. pneumonie and the isolates collected were tested using a broth microdilution MIC method semi-quantitatively, which remains the reference method. It is also highly accurate in particular for these antimicrobials: colistin, cefepime and more recently, ceftolozane-tazobactam [[28], [29], [30], [31], [32], [33]].

2A study limitation was the low number of Enterobacterales that were ceftolozane-tazobactam-resistant with genome sequencing. Although the study provided new information about the activity of ceftolozane-tazobactam against Enterobacterales and P. aeruginosa isolates. The study does not provide data about the incidence of infections in a given region.

Conclusions

Ceftolozane-tazobactam demonstrated relevant activity against most of the Enterobacterales and P. aeruginosa isolates. Ceftolozane-tazobactam could be considered a therapeutic alternative for the treatment of complicated urinary infections, complicated intra-abdominal infections and nosocomial pneumonia.

Funding

This study was part of EM200 surveillance programme funded by , a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA, which included funding for services related to preparing this manuscript.

Conflicts of interest

The authors have stated that there are no conflicts of interest.