In vitro activity of the novel β-lactamase inhibitor taniborbactam (VNRX-5133), in combination with cefepime or meropenem, against MDR Gram-negative bacterial isolates from China

Abstract

Objectives

To evaluate in vitro activity of the novel β-lactamase boronate inhibitor taniborbactam (VNRX-5133) combined with cefepime or meropenem against 500 urinary Gram-negative bacilli.

Methods

Cefepime/taniborbactam and 14 comparators were tested by broth microdilution or agar dilution methods. A total of 450 Enterobacteriaceae and 50 Pseudomonas aeruginosa were selected from 2017 to 2019 based on different β-lactamase-producing or resistance phenotypes. For carbapenem-non-susceptible isolates, the modified carbapenem inactivation method (mCIM), EDTA-CIM (eCIM) and amplification of carbapenemase genes were performed. For NDM-producing isolates and those with cefepime/taniborbactam MICs >8 mg/L, the MICs of meropenem/taniborbactam and/or mutations in PBP3 were investigated.

Results

Taniborbactam improved cefepime activity with the same efficiency as avibactam improved ceftazidime activity against 66 KPC-2 producers, 30 non-carbapenemase-producing carbapenem-non-susceptible Enterobacteriaceae and 28 meropenem-susceptible P. aeruginosa. However, cefepime/taniborbactam exhibited more potent activity than ceftazidime/avibactam against 56 ESBL-producing, 61 AmpC-producing, 32 ESBL and AmpC co-producing, 87 NDM-producing and 21 MBL-producing Enterobacteriaceae predicted by phenotypic mCIM and eCIM, 82 Enterobacteriaceae that were susceptible to all tested β-lactams and 22 carbapenem-non-susceptible P. aeruginosa. A four-amino acid ‘INYR’ or ‘YRIN’ insertion, with or without a one/two-amino acid mutation in PBP3, may have caused cefepime/taniborbactam MICs >8 mg/L among 96.6% (28/29) of the NDM-5-producing Escherichia coli, which accounted for the majority of isolates with cefepime/taniborbactam MICs >8 mg/L (76.1%, 35/46).

Conclusions

Taniborbactam’s superior breadth of activity, when paired with cefepime or meropenem, suggests these β-lactam/β-lactamase inhibitor combinations could be promising candidates for treating urinary tract infections caused by ESBL and/or AmpC, KPC or NDM-producing Enterobacteriaceae or P. aeruginosa.

Introduction

The increasing incidence of carbapenem-resistant Enterobacteriaceae (CRE) is jeopardizing global public health, since treatment options are limited to just a few last-resort antibiotics that are challenged by newly emerging resistance and adverse side effects.¹ In mainland China, KPC-2 is the major cause of carbapenem resistance in Klebsiella pneumoniae isolates, whereas NDM-5 and NDM-1 are the most common causes of carbapenem resistance in Escherichia coli and Enterobacter cloacae, respectively.² The broad-spectrum β-lactam enhancer avibactam (a diazabicyclooctane; DBO) is highly effective against serine β-lactamases [class A (including KPC), class C and OXA-48-like class D β-lactamases] when combined with the antipseudomonal cephalosporin ceftazidime and was approved for clinical use in mainland China in 2019, but has no inhibitory activity against MBL producers.^3,4 Moreover, avibactam is vulnerable to easily obtained KPC enzyme mutants and AmpC cephalosporinase mutants of Pseudomonas aeruginosa in vitro or in vivo and displays increased hydrolysis affinity for ceftazidime.^4–7 In E. coli, a four-amino acid ‘YRIK’, ‘YRIN’ or ‘TIPY’ insertion in PBP3, encoded by the gene ftsI and against which ceftazidime, aztreonam and cefepime have good potency and specific affinity, has been reported to decrease susceptibility to ceftazidime/avibactam and/or aztreonam/avibactam, but has no effect on carbapenem potency.^8,9 Furthermore, other candidate β-lactam enhancers, such as relebactam (a DBO) combined with imipenem and vaborbactam (boronic acid) combined with meropenem, are only active against class A and C serine β-lactamases.^3,4 Thus, there is an unmet medical need for new antimicrobial agents with activity against Enterobacteriaceae and P. aeruginosa expressing MBLs, particularly NDMs.

Taniborbactam (formerly VNRX-5133), a cyclic boronate broad-spectrum β-lactamase inhibitor, displays significant potency and a broad spectrum of inhibitory activity against both serine (class A, C and OXA-48-like enzymes) and metallo- (VIM and NDM, but not IMP) β-lactamases.¹⁰ When combined with cefepime, taniborbactam has the potential to address the serious unmet medical need for a safe and effective therapy to treat infections caused by MDR Enterobacteriaceae and P. aeruginosa.^3,4,11 Everest Medicines (NY, USA) is developing taniborbactam for use in combination with cefepime to treat complicated urinary tract infections (cUTIs), which have been associated with an alarming increase in the incidence of MDR pathogens.¹² This study aimed to evaluate the in vitro activities of cefepime combined with taniborbactam at a fixed concentration of 4 mg/L, as described previously,¹³ and 14 other comparators against 450 Enterobacteriaceae and 50 P. aeruginosa strains, particularly ESBL producers and CRE, isolated from UTIs. For isolates with cefepime/taniborbactam MICs >8 mg/L, the in vitro activities of meropenem/taniborbactam and meropenem were determined using the broth microdilution method.

Materials and methods

Bacterial isolates

A total of 500 non-repetitive urinary Gram-negative bacilli, including 150 E. coli, 150 K. pneumoniae, 50 E. cloacae, 50 Klebsiella oxytoca, 50 Citrobacter freundii and 50 P. aeruginosa isolates, were selected retrospectively from 2017 to 2019 to represent different important species, β-lactam susceptibility phenotypes and β-lactamase-producing backgrounds. The isolates were collected from 52 hospitals located in 21 provinces across the south, north, northwest, east and central districts of mainland China (Figure S1, available as Supplementary data at JAC Online). Of the 450 selected Enterobacteriaceae, 191 carbapenem-non-susceptible Enterobacteriaceae (CRE, meropenem or imipenem MIC ≥2 mg/L, ertapenem MIC ≥1 mg/L or carbapenemase-producing)¹⁴ were collected from the Chinese CRE network, while 259 isolates (28 CRE and 231 carbapenem-susceptible Enterobacteriaceae) were collected from other susceptibility surveillance programmes. Of the 231 carbapenem-susceptible isolates, most were selected for their phenotypic resistance to other β-lactams. Of the 50 P. aeruginosa isolates, 22 that were carbapenem-non-susceptible P. aeruginosa, (CRPA, meropenem and imipenem MICs ≥4 mg/L or carbapenemase-producing)¹⁴ were selected from other susceptibility surveillance programmes. All isolates were sent to the Clinical Microbiology Laboratory of Peking University People’s Hospital for re-identification by the Bruker Autoflex Speed MALDI-TOF mass spectrometer (Bruker Daltonics Inc., Bremen, Germany) and antimicrobial susceptibility testing.

Antimicrobial susceptibility testing

The MICs of cefepime alone, cefepime/taniborbactam (4 mg/L; VenatoRx Pharmaceuticals, Inc., PA, USA), ceftazidime alone and ceftazidime/avibactam were tested by broth microdilution. In addition, 11 other antibiotics (meropenem, imipenem, ertapenem, cefotaxime, cefotaxime/clavulanic acid, cefoxitin, ceftazidime/clavulanic acid, piperacillin/tazobactam, cefoperazone/sulbactam, amikacin and ciprofloxacin) were evaluated using the agar dilution method. For the NDM-producing isolates and non-bla_KPC/bla_NDM/bla_IMP-producing CRE with cefepime/taniborbactam MICs >8 mg/L, the MICs of meropenem/taniborbactam and meropenem were investigated using the broth microdilution method. The MICs of all drugs were determined according to the CLSI criteria in document M07 (11th edition, 2018).¹⁵ All drugs were purchased in powdered form from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China) or MedChemExpress (Monmouth Junction, NJ, USA). The MICs of cefepime and comparator agents were interpreted according to the breakpoints of CLSI guidelines M100 (29th edition, 2019).¹⁴ The breakpoints of cefoperazone/sulbactam for Enterobacteriaceae were taken from those for cefoperazone alone.

Phenotypic and molecular mechanism identification

The ESBL-producing isolates were identified phenotypically according to CLSI guidelines M100 (29th edition, 2019).¹⁴ Isolates that were non-susceptible to cefoxitin were identified as AmpC cephalosporinase-producing isolates, while those that were non-susceptible to cefoxitin and exhibited a clear synergy in cefepime and clavulanic acid double-disc synergy tests were identified as ESBL and AmpC cephalosporinase co-producing isolates, according to EUCAST guidelines.¹⁶

For carbapenem-non-susceptible Gram-negative bacilli, the modified carbapenem inactivation method (mCIM) and EDTA-CIM (eCIM; unsuitable for P. aeruginosa) were used to phenotypically discriminate between serine carbapenemase-, MBL- and non-carbapenemase-producing isolates. PCR was used to amplify carbapenemase genes (bla_KPC, bla_NDM, bla_IMP, bla_VIM and bla_OXA-48), as described previously.^17–19

Detection of FtsI (PBP3) mutations

Cefepime has potent and specific affinity for PBP3. In order to find the probable reason for isolates having cefepime/taniborbactam MICs >8 mg/L, PCR was used to detect mutations in ftsI among the 39 bla_NDM-5-producing E. coli;⁸bla_NDM-5 producers accounted for the majority of NDM producers with cefepime/taniborbactam MICs >8 mg/L. The amplification sequences were aligned to the WT nucleotide sequences of ftsI in E. coli MG1655 (GenBank no. NC_000913.3).

Ethics

This study was reviewed and granted ethics exemption by the medical ethics committee of Peking University People’s Hospital (No. 2017PHB224) and was approved by each participating hospital according to local requirements.

Results

Phenotypic and molecular classification of the tested isolates

Of the 450 selected Enterobacteriaceae, 219 were phenotypically classified as CRE, 56 as ESBL-producing Enterobacteriaceae, 61 as AmpC cephalosporinase-producing Enterobacteriaceae, 32 as ESBL and AmpC cephalosporinase-co-producing Enterobacteriaceae and 82 as Enterobacteriaceae susceptible to all tested β-lactams. Of the 50 P. aeruginosa isolates, 22 were phenotypically categorized as CRPA, while 28 were meropenem-susceptible P. aeruginosa. The dominant carbapenemase gene in the urinary CRE strains in this study was bla_KPC-2 (29.2%, 64/219), followed by bla_NDM-5 (23.3%, 51/219) and bla_NDM-1 (15.1%, 33/219). The dominant carbapenemase gene among carbapenem-non-susceptible K. pneumoniae in this study was bla_KPC-2 (67.0%, 61/91), whereas the dominant gene in carbapenem-non-susceptible E. coli was bla_NDM-5 (data not shown, 60.5%, 49/81). Six Enterobacteriaceae and three P. aeruginosa strains produced bla_IMP-like enzymes; all six Enterobacteriaceae isolates either co-produced bla_KPC and bla_NDM enzymes or co-produced bla_NDM and bla_IMP enzymes. Of the 54 CRE isolates for which no carbapenemase genes were detected (PCR negative for bla_KPC, bla_NDM, bla_IMP-like, bla_VIM-like and bla_OXA-48), 3 were phenotypically identified as serine carbapenemase-producing, 21 as MBL-producing and 30 as non-carbapenemase-producing (data not shown) by phenotypic mCIM and eCIM testing.

In vitro activity of cefepime/taniborbactam against isolates stratified by different phenotypes

The MIC distribution and the cumulative MIC distribution of cefepime/taniborbactam, ceftazidime/avibactam and other comparators are presented in Table 1 and Figures 1 and 2. For the 56 ESBL-producing isolates, taniborbactam reduced the cefepime MIC to ≤0.12/4 mg/L for 98.2% (55/56) of the isolates, whereas avibactam reduced the ceftazidime MIC to ≤0.12/4 mg/L for 32.1% (18/56) of the isolates. The susceptibility rates for meropenem and cefoxitin against these isolates were 100.0%, with 89.3% (50/56) of the ESBL-producing isolates susceptible to all tested β-lactam/β-lactamase inhibitors.

Figure 1.

Scatterplots of MICs of cefepime/taniborbactam versus cefepime tested by broth microdilution for isolates with different resistant phenotypes (a) or different enzyme-producing isolates (b) (drawn with R software version 3.4.4 and ggplot2 package version 2.2.1). The dots on the grey solid lines indicate the strains that shared identical MICs of cefepime/taniborbactam and cefepime alone. The dots on the blue dashed line indicate isolates for which the MIC of cefepime/taniborbactam was at least 2-fold greater than the MIC of cefepime. The dots on the orange dashed line indicate isolates where the MIC of cefepime was 2-fold greater than the MIC of cefepime/taniborbactam. Nine IMP producers included three bla_IMP-producing P. aeruginosa. KPC + NDM indicates isolates were co-harbouring bla_KPC and bla_NDM carbapenemase genes. NDM + IMP indicates isolates were co-harbouring bla_NDM and bla_IMP carbapenemase genes. SS, susceptible to all tested β-lactams. SS-P. aeruginosa, meropenem-susceptible P. aeruginosa. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC.

Figure 2.

The cumulative MIC distributions [(a) by phenotype and (b) for carbapenem-non-susceptible strains by gene background] of cefepime/taniborbactam, ceftazidime/avibactam and comparator meropenem for 500 tested isolates (drawn with R software version 3.4.4 and ggplot2 package version 2.2.1). The MICs of cefepime and cefepime/taniborbactam shown in the figure were tested by the broth microdilution method. The MICs of meropenem shown in the figure were tested by the agar dilution method. SS, susceptible to all tested β-lactams. SS-P. aeruginosa, meropenem-susceptible P. aeruginosa. Nine IMP producers included three bla_IMP-producing P. aeruginosa. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC.

Table 1.

The in vitro activity of cefepime/taniborbactam and its comparators against 438 Enterobacteriaceae and 50 P. aeruginosa

Phenotype or molecular type (total numbers of isolates)	%R	%I or %SDD	%S	MIC50/90 (mg/L)	Numbers of isolates at each MIC value (mg/L)
					≤0.008	0.016	0.03	0.06	0.12	0.25	0.5	1	2	4	8	16	32	64	128	256	>256
ESBL producer (56)
cefepime/taniborbactam				0.03/0.12	1	13	16	19	6		1
cefepime	44.6	32.1	23.2	8/32			1	1	1			5	5	6	12	15	6	2	1		1
ceftazidime	55.4	16.1	28.6	16/128						1		4	6	5	9	11	8	5	3	3	1
ceftazidime/avibactam	1.8		98.2	0.25/0.5			2	2	14	23	11	3									1
piperacillin/tazobactam	3.6	0	96.4	2/4							5	13	27	9						1	1
AmpC producer (61)
cefepime/taniborbactam				0.06/0.25		9	16	14	13	4	2	1	2
cefepime	3.3	6.6	90.2	0.12/2		4	11	10	7	10	3	6	4	2	2		1		1
ceftazidime	39.3	1.6	59	2/>256					1	2	13	13	4	3	1	2		3	6	6	7
ceftazidime/avibactam	3.3		96.7	0.5/1		1		1	7	20	12	15	3				1				1
piperacillin/tazobactam	6.6	13.1	80.3	4/64							1	11	14	5	8	10	3	5	2		2
ESBL and AmpC co-producer (32)								11	18	17	4	6	1		2		1
cefepime/taniborbactam				0.12/1		1	4	2	11	5	1	5	2	1
cefepime	71.9	28.1	0	32/256										3	6	6	5	6	2	1	3
ceftazidime	78.1	9.4	12.5	32/256										4	3	2	7	5	3	6	2
ceftazidime/avibactam	0		100	0.5/2			3		1	8	12	4	3	1
piperacillin/tazobactam	0	15.6	84.4	4/32							1	2	5	9	7	3	2	3
blaKPC-producer (66)
cefepime/taniborbactam				2/8			1	3	1	1	5	21	17	10	3	1	2	1
cefepime	95.5	4.5	0	>256/>256										1	2	3	1	4	5	5	45
ceftazidime	100	0	0	>256/>256													1	4	7	8	46
ceftazidime/avibactam	7.6		92.4	4/8							2	10	17	22	10	2	1	1			1
piperacillin/tazobactam	100	0	0	>256/>256															3	4	59
blaNDM-producer (87)
cefepime/taniborbactam				16/64					1	6	3	9	10	7	5	16	17	7	5
cefepime	100	0	0	>256/>256												3	1	6	12	5	60
meropenem/taniborbactam				0.25/4		1		11	20	18	7	12	7	5	2	2		1		1
meropenem	97.7	2.3	0	64/128									2	1	2	9	19	22	27	4	1
ceftazidime	100	0	0	>256/>256															1		86
ceftazidime/avibactam	98.9		1.1	>256/>256										1						1	85
piperacillin/tazobactam	100	0	0	>256/>256															10	16	61
Non-blaKPC/blaNDM/blaIMP-producing CRE (54)
cefepime/taniborbactam				4/32				1		5	7	7	6	10	1	5	7	3		2
cefepime	83.3	3.7	13	>256/>256					1	1		4	1	2			2	2	2	7	32
meropenem/taniborbactam				0.5/32					4	3	2	1	4	1			1	1
meropenem	88.2	5.9	5.9	64/128								1	1	1		1	3	3	6	1
ceftazidime	98.1	0	1.9	>256/>256										1		2	2	1	2	4	42
ceftazidime/avibactam	50		50	8/>256						1	5	4	9	4	4		1				26
piperacillin/tazobactam	70.4	13	16.7	256/>256								1			2	6	1	6	3	11	24
Susceptible phenotype (82)
cefepime/taniborbactam				0.016/0.06	3	40	25	13	1
cefepime	0	0	100	0.03/0.06	3	29	27	16	6				1
ceftazidime	0	0	100	0.25/0.5				3	30	33	12	4
ceftazidime/avibactam	0		100	0.12/0.25	1	1	3	15	36	21	3	2
piperacillin/tazobactam	2.4	0	97.6	2/4	2					1	8	25	36	7	1	2
CRPA (22)
cefepime/taniborbactam				8/32										8	7	3	2	1
cefepime	54.5	18.2	27.3	32/128										3	3	4	5	3	3		1
ceftazidime	68.2	4.5	27.3	64/>256										3	3	1		4	2	4	5
ceftazidime/avibactam	59.1		40.9	16/256										3	6	6	2		2	2	1
piperacillin/tazobactam	59.1	9.1	31.8	128/256										1	1	5		2	5	7	1
Meropenem-susceptible P. aeruginosa (28)
cefepime/taniborbactam				4/8							1	7	5	11	4
cefepime	0	7.1	92.9	4/8							1	5	7	11	2	2
ceftazidime	17.9	0	82.1	4/64								1	6	10	6		2	2			1
ceftazidime/avibactam	3.6		96.4	4/8							1	2	6	9	9				1
piperacillin/tazobactam	0	3.6	96.4	4/16								1	6	10	3	7	1

%R, percentage resistant; %I, percentage intermediate; %S, percentage susceptible; %SDD, percentage susceptible dose-dependent.

Susceptible phenotype, susceptible to all tested β-lactams. %SDD used for cefepime and %I used for ceftazidime and piperacillin/tazobactam.

Cefepime/taniborbactam, cefepime, ceftazidime and ceftazidime/avibactam were tested by the broth microdilution method; piperacillin/tazobactam was tested by the agar dilution method. For the 54 non-bla_KPC/bla_NDM/bla_IMP-producing CRE, meropenem/taniborbactam and meropenem were only tested against the 17 isolates with cefepime/taniborbactam MICs >8 mg/L by the broth microdilution method. This table does not include 3 Enterobacteriaceae co-harbouring bla_KPC and bla_NDM carbapenemase genes, 3 Enterobacteriaceae co-harbouring bla_NDM and bla_IMP genes and 6 Enterobacteriaceae producing bla_IMP-like enzyme, because MIC₅₀s and MIC₉₀s are not presented for groups of fewer than 10 isolates.

For the 61 AmpC cephalosporinase-producing isolates, taniborbactam reduced the cefepime MIC to ≤0.25/4 mg/L for 91.8% (56/61) of the isolates, whereas avibactam reduced the ceftazidime MIC to ≤0.25/4 mg/L for 47.5% (29/61) of the isolates. The susceptibility rates for meropenem, cefepime, cefoperazone/sulbactam, ceftazidime and piperacillin/tazobactam against these isolates were 100.0%, 90.2%, 83.6%, 59.0% and 80.3%, respectively.

For the 32 ESBL and AmpC cephalosporinase co-producers, taniborbactam reduced the cefepime MIC to ≤0.25/4 mg/L for 71.9% (23/32) of the isolates and avibactam reduced the ceftazidime MIC to ≤0.25/4 mg/L for 37.5% (12/32) of the isolates. The susceptibility rates for meropenem and piperacillin/tazobactam against these isolates were 100.0% and 84.4%, respectively.

For the Enterobacteriaceae susceptible to all tested β-lactams, taniborbactam reduced the cefepime MIC to ≤0.12/4 mg/L for 100.0% (82/82) of the isolates, whereas avibactam reduced the ceftazidime MIC to ≤0.12/4 mg/L for 68.3% (56/82) of the isolates.

Of the 22 CRPA strains, the MICs of cefepime/taniborbactam for 68.2% (15/22) isolates were ≤8 mg/L, while 40.9% (9/22) of isolates were susceptible to ceftazidime/avibactam (≤8 mg/L). Of the 28 meropenem-susceptible P. aeruginosa strains, 100.0% of isolates had cefepime/taniborbactam MICs ≤8 mg/L and 96.4% (27/28) were susceptible to ceftazidime/avibactam (≤8 mg/L).

In vitro activity of cefepime/taniborbactam against serine carbapenemase-producing strains

For the 66 bla_KPC-producing Enterobacteriaceae, taniborbactam protected the in vitro activity of cefepime, with a 128-/32-fold MIC₅₀/MIC₉₀ reduction, whereas avibactam protected the in vitro activity of ceftazidime, with a 64-/32-fold MIC₅₀/MIC₉₀ reduction. Moreover, taniborbactam reduced the cefepime MIC to ≤8/4 mg/L for 93.9% (62/66) of these isolates, whereas avibactam reduced the ceftazidime MIC to the susceptible range for 92.4% (61/66). The MICs of ceftazidime or cefepime for four of the bla_KPC-2-producing strains were not restored to the susceptible range (≤8/4 mg/L) by either avibactam or taniborbactam.

Three non-bla_KPC/bla_NDM/bla_IMP-producing CRE were phenotypically classified as serine-carbapenemase producers by phenotype mCIM and eCIM testing. Taniborbactam reduced the cefepime MIC to ≤8/4 mg/L for two of these three isolates, whereas avibactam reduced the ceftazidime MIC to ≤8/4 mg/L for only one of the three.

In vitro activity of cefepime/taniborbactam against MBL-producing strains

For the 87 bla_NDM-producing Enterobacteriaceae, taniborbactam protected the in vitro activity of cefepime, with a 16-/4-fold MIC₅₀/MIC₉₀ reduction, whereas avibactam did not protect the in vitro activity of ceftazidime as there was no MIC₅₀/MIC₉₀ reduction. Moreover, taniborbactam reduced the cefepime MIC to ≤8/4 mg/L for 47.1% (41/87) of these isolates, whereas avibactam reduced the ceftazidime MIC to the susceptible range for 1.1% of the strains (1/87, a bla_NDM-1-producing strain). Furthermore, taniborbactam was unable to restore the in vitro activity of cefepime to ≤8/4 mg/L for 46 of the bla_NDM-producing strains (data not shown; 35 E. coli, 1 K. pneumoniae and 1 E. cloacae producing bla_NDM-5; 2 K. pneumoniae, 2 C. freundii, 1 E. coli and 1 E. cloacae producing bla_NDM-1; 1 E. coli producing bla_NDM-3; 1 E. coli producing bla_NDM-7; and 1 E. coli producing bla_NDM-9).

For the nine bla_IMP-producing Enterobacteriaceae, taniborbactam reduced the cefepime MIC for two CRE and one CRPA to ≤8/4 mg/L, whereas avibactam did not restore the ceftazidime MIC for any of these strains.

A total of 21 non-bla_KPC/bla_NDM/bla_IMP-producing CRE were phenotypically classified as MBL-producing strains by phenotypic mCIM and eCIM testing. Taniborbactam reduced the cefepime MIC to ≤8/4 mg/L for 47.6% (10/21) of these isolates, whereas avibactam reduced the ceftazidime MIC to ≤8/4 mg/L for 4.8% (1/21) of the strains.

In vitro activity of cefepime/taniborbactam against carbapenemase co-producers and non-carbapenemase-producing CRE

For the three bla_KPC and bla_NDM co-producing CRE, the cefepime and avibactam MICs were not reduced to ≤8/4 mg/L when combined with taniborbactam or avibactam. The cefepime MIC was reduced to ≤8/4 mg/L when combined with taniborbactam for two of the three bla_NDM and bla_IMP co-producing CRE, whereas no reduction was observed in the ceftazidime MIC when combined with avibactam.

A total of 30 non-bla_KPC/bla_NDM/bla_IMP-producing CRE were found to be non-carbapenemase-producing strains by mCIM and eCIM testing. Taniborbactam reduced the cefepime MIC to ≤8/4 mg/L for 83.3% (25/30) of these isolates and avibactam also reduced the ceftazidime MIC to ≤8/4 mg/L for 83.3% (25/30) of the strains.

In vitro activity of meropenem/taniborbactam against isolates with cefepime/taniborbactam MICs >8  mg/L and PBP3 mutation detection in NDM-producing strains

For the 17 non-bla_KPC/bla_NDM/bla_IMP-producing CRE with cefepime/taniborbactam MICs >8 mg/L, taniborbactam protected the in vitro activity of meropenem, with a 128-/4-fold MIC₅₀/MIC₉₀ reduction from 64/128 mg/L to 0.5/32 mg/L (Table 1). Taniborbactam reduced the meropenem MIC to ≤1/4 mg/L for 58.8% (10/17) of these isolates.

In order to further confirm the inhibitory activity of taniborbactam against NDM enzymes, a supplementary experiment was conducted and showed that for 87 bla_NDM-producing Enterobacteriaceae, taniborbactam restored the in vitro activity of meropenem, with a 256-/32-fold MIC₅₀/MIC₉₀ reduction from 64/128 mg/L to 0.25/4 mg/L (Table 1). Taniborbactam reduced the meropenem MIC to ≤2/4 mg/L for 87.4% (76/87) NDM producers and to ≤8/4 mg/L for 95.4% (83/87) of the strains.

The bla_NDM-5-producing E. coli (76.1%, 35/46) accounted for the majority of the NDM-producing strains with cefepime/taniborbactam MICs >8 mg/L. For the 51 bla_NDM-5-producing strains, taniborbactam restored the in vitro activity of meropenem, with a 128-/64-fold MIC₅₀/MIC₉₀ reduction from 64/128 mg/L to 0.5/2 mg/L (data not shown). Among 39 bla_NDM-5-producing E. coli that had been examined for the PBP3 mutation, 28 of the 29 isolates with cefepime/taniborbactam MICs >8 mg/L possessed a four-amino acid INYR or YRIN insertion, with or without a one/two-amino acid mutation in PBP3, whereas 7 of the 10 isolates with cefepime/taniborbactam MICs ≤8 mg/L possessed WT PBP3 (Table 2).

Table 2.

The MIC distribution of meropenem/taniborbactam for 39 bla_NDM-5-producing E. coli with different amino acid changes in PBP3

PBP3 amino acid changes	No.	Cefepime/ taniborbactam (mg/L)	Meropenem/taniborbactam MIC (mg/L)
			0.008	0.016	0.03	0.06	0.12	0.25	0.5	1	2	4	8	16	32	64	128	256	>256
Isolates with cefepime/  taniborbactam MICs >8 mg/L	29
335aa insertion INYR,  E349K, I532L	6	16–128				1	2		1	1	1
337aa insertion YRIN	1	256						1
337aa insertion YRIN, A413V	1	16						1
337aa insertion YRIN, E349K	1	32								1
337aa insertion YRIN,  E349K, I532L	19	16–128				1	2	3	2	3	5	2		1
WT	1	128					1
Isolates with cefepime/  taniborbactam MICs ≤8 mg/L	10
337aa insertion YRIN, E349K	1	4					1
337aa insertion YRIN,  E349K, I532L	2	2–8		1		1
WT	7	0.25–4				2	2	2	1

aa, amino acid site.

The amplification sequences were aligned to the WT nucleotide sequences of ftsI in E. coli MG1655 (GenBank no. NC_000913.3).

Discussion

CRE infections have few treatment options and are associated with high rates of clinical failure;^20,21 thus, novel β-lactam/β-lactamase inhibitor combinations are urgently required.^3,22,23 This study investigated the in vitro activities of cefepime/taniborbactam and 14 comparators against 500 selected urinary Gram-negative bacilli.

In mainland China, E. coli is the most common causative organism of UTIs and ESBLs are the most prevalent β-lactam resistance-conferring enzymes.²⁴ In this study, taniborbactam restored cefepime activity in ESBL-producing isolates with potency equal to that of clavulanic acid for restoring ceftazidime or cefotaxime activity and to that of avibactam for protecting ceftazidime. KPC is the most prevalent carbapenemase both globally and in mainland China.² Taniborbactam protected and maintained the cefepime MIC at ≤8/4 mg/L in 93.9% (62/66) of the KPC-producing strains and 83.3% (25/30) of the non-carbapenemase-producing CRE isolates, similar to the potency of avibactam for restoring ceftazidime. Hence, cefepime/taniborbactam and ceftazidime/avibactam²⁵ could serve as good treatment options for reducing carbapenem consumption when treating ESBL-, KPC- and non-carbapenemase-producing CRE isolates.

In mainland China, NDM-5 accounts for the majority of the resistance mechanisms in carbapenem-resistant E. coli, with NDM-1 being the major cause of carbapenem resistance in E. cloacae.² However, novel β-lactam/β-lactamase inhibitors such as ceftazidime/avibactam, meropenem/vaborbactam and imipenem/relebactam cannot combat MBL-producing strains.¹¹ In this study, taniborbactam improved the cefepime MIC for 47.1% (41/87) of the NDM-producing strains and 47.6% (10/21) of the non-bla_NDM/bla_IMP/bla_VIM MBL-producing strains. Moreover, bla_NDM-5-producing E. coli (76.1%, 35/46) constituted the majority of the strains with cefepime/taniborbactam MICs >8 mg/L. Furthermore, 96.6% (28/29) of the NDM-5-producing E. coli with cefepime/taniborbactam MICs >8 mg/L harboured a four-amino acid INYR or YRIN insertion, with or without a one/two-amino acid change in PBP3, conferring good potency and specific affinity for cefepime but not meropenem. These findings suggest that taniborbactam has good activity for restoring cefepime activity against NDM-producing isolates. Moreover, PBP3 mutations may be the major reason for the cefepime/taniborbactam MICs >8 mg/L among NDM-producing E. coli and meropenem/taniborbactam was expected to combat such isolates.

For the isolates that produced AmpC cephalosporinase or co-produced ESBL and AmpC, all β-lactam-susceptible Enterobacteriaceae and CRPA, taniborbactam reduced the cefepime MIC to an even lower value than that to which avibactam reduced the ceftazidime MIC. For CRPA, the amikacin, cefepime/taniborbactam, ceftazidime/avibactam and ciprofloxacin susceptibility rates were 90.9%, 68.2%, 40.9% and 40.9%, respectively. Thus, cefepime/taniborbactam may be superior to ceftazidime/avibactam against AmpC-producing, ESBL and AmpC co-producing, β-lactam-susceptible Enterobacteriaceae and CRPA.

Our study encountered the following limitations. Firstly, this study used a challenge strain set that was highly enriched for β-lactam resistance, preventing comparison with the MIC statistics from prevalence-based surveillance approaches. Secondly, we did not characterize non-β-lactamase-mediated resistance mechanisms, such as porin expression/mutation and efflux pump expression, which affect the activity of cefepime, other cephalosporins and carbapenems.

In conclusion, when restoring the activity of cefepime or meropenem, taniborbactam displays superior effects to those of avibactam restoring ceftazidime, suggesting that these β-lactam/β-lactamase inhibitor combinations could be promising drug candidates for treating UTIs due to MDR Enterobacteriaceae and P. aeruginosa, including strains expressing serine β-lactamases (ESBLs and/or AmpC), serine carbapenemases (KPC) and metallo-β-lactamases (NDM). Further studies are needed to evaluate the promising roles of cefepime/taniborbactam and meropenem/taniborbactam in challenging the CRE treatment dilemma.

Acknowledgements

For providing strains, we would like to acknowledge Binghuai Lu, Chunlei Wang and Bin Cao (China-Japan Friendship Hospital), Li Gu and Chunxia Yang (Beijing Chao-Yang Hospital, Capital Medical University), Yunjian Hu (Beijing Hospital), Haiquan Kang and Bing Gu (Medical Technology Institute of Xuzhou Medical University, The Affiliated Hospital of Xuzhou Medical University), Zhusheng Guo (Dongguan Tungwah Hospital), Quan Fu (The Affiliated Hospital of Inner Mongolia Medical University), Xiaoqian Zhang (Henan Province Hospital of TCM), Jianrong Su and Liyan Ma (Beijing Friendship Hospital of Capital Medical University), Zhijie Zhang (Shengjing Hospital of China Medical University), Li Zhao and Yajun Zhang (Beijing Renhe Hospital), Liangyi Xie (Hunan Provincial People’s Hospital), Xiaobo Ma (The First Affiliated Hospital of Xiamen University), Zhongju Chen (Tongji Medical College of Huazhong University of Science & Technology), Kang Liao (The First Affiliated Hospital of Sun Yat-sen University), Jiajia Zheng (Peking University Third Hospital), Jinjing Tian (The Second People’s Hospital of Liaocheng), Yan Jin (Shandong Provincial Hospital), Yumei Zhang (People’s Hospital of Zunhua), Hainan Wen (Affiliated Hospital of Chengde Medical University), Liying Sun (Peking University First Hospital), Fengyan Pei (Jinan Central Hospital), Qiaozhen Cui (People’s Hospital of Shanxi Province), Xiaoying Li (Weifang Traditional Chinese Hospital), Junfang Pei and Ting Li (Yuncheng Central Hospital), Feng Zhao (Sir Run Run Shaw Hospital affiliated with the Zhejiang University) and Ji Zeng (Wuhan Puai Hospital).

Funding

This work was supported by Everest Medicines (US) Ltd (VNRX-5133-V-CN001, Beijing, China).

Transparency declarations

Chengcheng Zhou is an employee of Everest Medicines (US) Limited, Shanghai. All other authors: none to declare.