Activity of Imipenem with Relebactam against Gram-Negative Pathogens from New York City

Abstract

Imipenem with relebactam was active against Escherichia coli, Klebsiella pneumoniae, and Enterobacter spp., including K. pneumoniae carbapenemase (KPC)-producing isolates. Loss of OmpK36 in KPC-producing K. pneumoniae isolates affected the susceptibility of this combination. Enhanced activity was evident against Pseudomonas aeruginosa, including isolates with depressed oprD and increased ampC expression. However, the addition of relebactam to imipenem did not provide added benefit against Acinetobacter baumannii. The combination of imipenem with relebactam demonstrated activity against KPC-producing Enterobacteriaceae and multidrug-resistant P. aeruginosa.

TEXT

The spread of carbapenemases in Enterobacteriaceae, Pseudomonas aeruginosa, and Acinetobacter baumannii has created therapeutic dilemmas for clinicians. In particular, the acquisition of metallo-β-lactamases and the class A enzyme Klebsiella pneumoniae carbapenemase (KPC) affords protection against virtually all β-lactam therapeutic agents (1). The worldwide dissemination of bacteria possessing bla_KPC has been especially striking (2). First reported in the northeastern United States in the 1990s, pathogens harboring this β-lactamase are now endemic in countries in Asia, South America, and Europe (2).

Novel β-lactamase inhibitors are being developed to restore the utility of β-lactam antibiotics against carbapenemase-producing pathogens (3, 4). Avibactam and relebactam are diazabicyclooctane inhibitors with activity against a wide spectrum of β-lactamases, including class A (extended-spectrum β-lactamases [ESBLs] and KPC) and class C (AmpC) enzymes (3, 4). In this study, we determined the activity of imipenem with relebactam against Gram-negative pathogens from medical centers in New York City.

Between November 2013 and January 2014, single patient isolates of Escherichia coli, K. pneumoniae, Enterobacter spp., P. aeruginosa, and A. baumannii were collected from 11 hospitals in Brooklyn and Queens, New York. Susceptibility tests were performed in a central research laboratory using the agar dilution method, and results were interpreted according to CLSI guidelines (5). Isolates of E. coli and K. pneumoniae were presumed to harbor ESBLs if they were not susceptible to ceftazidime and/or ceftriaxone and did not have bla_KPC. Imipenem was tested both with and without the presence of relebactam (fixed concentration of 4 μg/ml). For the purposes of this study, imipenem breakpoints were used to interpret susceptibility to imipenem plus relebactam. Cephalosporin-resistant isolates were tested by PCR for the presence of bla_KPC, bla_OXA-23-type, bla_OXA-24-type, bla_VIM, bla_IMP, and bla_NDM using previously described primers (6–9). Isolates of K. pneumoniae with bla_KPC and isolates of A. baumannii with bla_OXA23-type underwent genetic fingerprinting by the repetitive element palindromic PCR (rep-PCR) method with the ERIC-2 primer, as described previously (6).

Susceptibility testing of imipenem with and without relebactam was also performed with a collection of previously characterized isolates of K. pneumoniae, P. aeruginosa, and A. baumannii (10–13). The expression of genes encoding β-lactamases, efflux pumps, and porins was correlated with the MICs for imipenem with relebactam.

Surveillance study results.

A total of 2,778 isolates of E. coli were gathered during the 3-month surveillance study. Susceptibilities are presented in Table 1. Of the bla_KPC-negative isolates, 383 were considered to have ESBLs and all were susceptible to imipenem. The imipenem MIC₅₀ and MIC₉₀ values for the ESBL-producing isolates were 0.25 and 0.5 μg/ml, respectively; with the addition of relebactam, the corresponding values were 0.25 and 0.25 μg/ml. Five isolates harbored bla_KPC. For these 5 isolates, the imipenem MICs ranged from 0.5 to >32 μg/ml. With the addition of relebactam, the MICs decreased to 0.12 to 0.5 μg/ml.

TABLE 1.

Susceptibility results for Enterobacteriaceae, P. aeruginosa, and A. baumannii isolates collected in surveillance study

Species and drug(s)	MIC50 (μg/ml)	MIC90 (μg/ml)	MIC range (μg/ml)	% susceptible
E. coli (n = 2,778)
Ertapenem	0.008	0.03	≤0.002 to >32	99.6
Imipenem	0.25	0.25	≤0.03 to >32	99.9
Imipenem + relebactam	0.25/4	0.25/4	≤0.03/4 to 1/4	100
K. pneumoniae (n = 891)
Ertapenem	≤0.125	8	≤0.125 to >8	86
Imipenem	0.25	4	0.06 to >16	88
Imipenem + relebactam	0.25/4	0.25/4	0.06/4 to 2/4	99.3
blaKPC-possessing K. pneumoniae (n = 111)
Ertapenem	>8	>8	0.5 to >8	2
Imipenem	16	>16	0.5 to >16	9
Imipenem + relebactam	0.25/4	1/4	0.12/4 to 2/4	97
Enterobacter spp. (n = 211)
Ertapenem	≤0.125	0.25	≤0.125 to >8	93
Imipenem	0.5	1	≤0.03 to >16	90
Imipenem + relebactam	0.25/4	0.5/4	≤0.03/4 to 2/4	99
P. aeruginosa (n = 490)
Imipenem	2	16	≤0.03 to >16	70
Imipenem + relebactam	0.5/4	2/4	≤0.03/4 to >16/4	98
Imipenem-resistant P. aeruginosa (n = 144)
Imipenem	8	>16	4 to >16	0
Imipenem + relebactam	1/4	2/4	0.25/4 to >16/4	92
A. baumannii (n = 158)
Imipenem	4	>16	≤0.03 to >16	49
Imipenem + relebactam	2/4	>16/4	≤0.03/4 to >16/4	51
blaOXA-23-possessing A. baumannii (n = 58)
Imipenem	>16	>16	≤0.03 to >16	12
Imipenem + relebactam	>16/4	>16/4	≤0.03/4 to >16/4	12

A total of 891 isolates of K. pneumoniae were collected (Table 1). Of the bla_KPC-negative isolates, 185 were considered ESBL producers. All of the ESBL producers were susceptible to imipenem, with MIC₅₀ and MIC₉₀ values of 0.25 and 0.5 μg/ml, respectively. With the addition of relebactam, the imipenem MIC₅₀ and MIC₉₀ values were 0.25 and 0.5 μg/ml, respectively. For 111 isolates that harbored bla_KPC, the imipenem MIC₅₀ and MIC₉₀ values were 16 and >16 μg/ml, respectively. With the addition of relebactam, the MIC₅₀ and MIC₉₀ values decreased to 0.25 and 1 μg/ml, respectively; three isolates had MICs of 2 μg/ml (intermediate resistance to imipenem). Twenty-three isolates with bla_KPC, from 11 different hospitals, underwent fingerprinting. Eight rep-PCR types were identified, including 12 isolates that belonged to one clone (data not shown).

There were 211 Enterobacter isolates (Table 1), including 90 Enterobacter aerogenes isolates and 120 Enterobacter cloacae isolates. Three E. aerogenes isolates and four E. cloacae isolates harbored bla_KPC. Of these seven isolates, six were not susceptible to imipenem; the MICs ranged from 0.5 to >16 μg/ml. With the addition of relebactam, the MICs ranged from 0.12 to 2 μg/ml, with six isolates being susceptible to imipenem.

Among 490 isolates of P. aeruginosa, the imipenem MIC₅₀ and MIC₉₀ values were 2 and 16 μg/ml, respectively (Table 1). These values decreased to 0.5 and 2 μg/ml, respectively, with the addition of relebactam. Among the 144 isolates that were not susceptible to imipenem, the addition of relebactam resulted in MIC₅₀ and MIC₉₀ values of 1 and 2 μg/ml, respectively.

Imipenem MICs with and without relebactam were similar among 158 isolates of A. baumannii (Table 1). Fifty-eight isolates were found to have bla_OXA-23-like, two carried bla_OXA-24-like, and one harbored bla_KPC. Eighteen isolates with bla_OXA-23-like, from eight hospitals, underwent fingerprinting. A total of 10 rep-PCR types were identified, including six belonging to a single clone (data not shown). For the isolates harboring bla_OXA-23-like, the imipenem MIC₅₀ and MIC₉₀ values were >16 and >16 μg/ml, respectively, with or without the addition of relebactam. Similarly, the imipenem MICs for the two isolates with bla_OXA-24-like did not change with the addition of relebactam. For a single isolate with bla_KPC, the imipenem MIC decreased from >16 μg/ml to 4 μg/ml with the addition of relebactam.

Results with previously characterized isolates.

Fourteen previously characterized isolates of KPC-producing K. pneumoniae were examined (10, 11). In the presence of relebactam, imipenem MICs did not correlate with the expression of ramA, acrB, or bla_KPC. Similarly, there was no correlation among 8 isolates without frameshift mutations in ompK35. Ten isolates had expression of ompK36 greater than the control levels, with imipenem MICs ranging from 2 to >16 μg/ml. With the addition of relebactam, all of the imipenem MICs were 0.25 to 0.5 μg/ml. Four isolates had reduced expression of ompK36; the imipenem MICs for those isolates were 4, >16, >16, and >16 μg/ml. With the addition of relebactam, the imipenem MICs decreased to 0.5, 2, 2, and 8 μg/ml, respectively. The latter isolate did not have amplifiable ompK36.

Thirty previously characterized isolates of P. aeruginosa were analyzed (12); none possessed carbapenemases. Six isolates were wild type regarding ampC and oprD expression (similar to control). Imipenem MICs ranged from 2 to 4 μg/ml for this group, and all of the isolates had imipenem MICs of 1 μg/ml with the addition of relebactam. Fourteen isolates had reduced oprD expression with wild-type ampC expression. For these isolates, the imipenem MICs ranged from 1 to >16 μg/ml. With the addition of relebactam, the MICs decreased to 0.25 to 8 μg/ml (average, 1.8 ± 1.9 μg/ml). Ten isolates had reduced oprD expression and upregulated ampC expression. The imipenem MICs for these isolates ranged from 2 to >16 μg/ml. With the addition of relebactam, the MICs ranged from 1 to 8 μg/ml (average, 4.6 ± 2.9 μg/ml).

Twenty-eight previously characterized isolates of A. baumannii were also included (13). In general, imipenem MICs were unchanged with the addition of relebactam. There was no clear relationship between the expression of ampC, bla_oxa-51, adeB, and abeM and the MICs for imipenem with relebactam.

The global spread of carbapenemases in pathogens that are already resistant to other classes of antibiotics has posed a serious therapeutic challenge for clinicians. RPX7009, avibactam, and relebactam are novel β-lactamase inhibitors with activity against primarily class A and class C β-lactamases (3, 4). When combined with imipenem, relebactam has demonstrated dose-dependent synergy against a small number of Enterobacteriaceae strains harboring bla_KPC (14, 15). In this report, relebactam (at a fixed concentration of 4 μg/ml) restored imipenem susceptibility to 97% of K. pneumoniae isolates with bla_KPC. When a group of well-characterized isolates were tested, downregulation of ompK36 appeared to partially offset the protective effect of relebactam. Restoration of imipenem susceptibility was also found for a small number of bla_KPC-containing isolates of E. coli and Enterobacter spp.

In addition, relebactam with imipenem has demonstrated activity against P. aeruginosa, including isolates with depressed oprD and increased ampC expression (14, 15). In our study, the addition of relebactam resulted in approximately 4-fold decreases in the imipenem MIC₅₀ and MIC₉₀ values, and imipenem susceptibility rates increased from 70% to 98% when relebactam was added. Restoration of imipenem activity was noted for isolates with depressed oprD expression, with or without increased ampC expression, although the MICs did continue to be higher than those for the wild-type isolates.

The addition of relebactam did not improve the activity of imipenem against A. baumannii, however. MICs were unchanged for isolates with overexpression of ampC and/or bla_OXA-51, suggesting a lack of relebactam activity against these enzymes. Diminished inhibitor activity has been observed against K. pneumoniae strains with OXA-48 and absent activity against pathogens harboring metallo-β-lactamases (15). Further development of new antimicrobial agents directed against pathogens harboring these β-lactamases is sorely needed.