To the Editor—Enterobacterales, such as Enterobacter spp, Serratia marcescens, Citrobacter freundii, Providencia spp and Morganella morganii, often referred to as the ESCPM group, may express high levels of chromosomal AmpC β-lactamases at high levels following exposure to β-lactams, mainly after third-generation cephalosporin therapy. ^{Reference Jacoby1} The induction or selection of derepressed isolates is a concern because they contribute to the isolation of organisms no longer susceptible to specific β-lactams and may lead to clinical failure, with scarce antimicrobial options. ^{Reference Alvarez, Tran, Chow and Jacoby2}

Ceftolozane with tazobactam (C/T) is a combination drug comprising a β-lactamase inhibitor (tazobactam) with a new cephalosporin (ceftolozane). Tazobactam inhibits class A extended-spectrum β-lactamases (EBSLs), and ceftolozane acts via a high affinity for some penicillin-binding-protein (PBPs). C/T is stable in the presence of AmpC β-lactamases and against OprD deficiency and efflux pumps. These characteristics make the C/T combination an important weapon in the treatment of infections due to extensively resistant Pseudomonas aeruginosa that are not carbapenemase producers. ^{Reference Craig and Andes3}

Despite the high efficacy described so far, emergence of resistance to C/T, mainly in P. aeruginosa isolates overexpressing AmpC-β-lactamase enzymes, have been reported. ^{Reference Castanheira, Mills, Farrell and Jones4} Although derepressed AmpC may occur in P. aeruginosa, the main target for C/T use, this resistance mechanism is more robust in Enterobacter cloacae complex isolates, with a higher ability than others from the ESCPM group to derepress AmpC-β-lactamase production, which has important clinical and therapeutic implications. ^{Reference Perez5}

The main objective of this study was to determine the C/T minimum inhibitory concentration (MIC) among E. cloacae complex isolates, producing or not derepressed AmpC-β-lactamases. Additionally, meropenem and ceftazidime/avibactam MICs were also determined.

A set of 123 E. cloacae complex isolates recovered from inpatients between August 2016 and December 2017, in southern Brazil, were included in this study. Bacterial identification was made using the Vitek 2 automated system (bioMérieux, Marcy I’Etoile, France) and matrix-assisted laser desorption/ionization time-of-flight mass spectroscopy (MALDI-TOF MS) for confirmation when necessary. Ertapenem susceptibility was determined using disc-diffusion testing. ⁶ The MICs of cefotolozane/tazobactam, meropenem, and ceftazidime/avibactam were determined using MIC test strips (MTS, Liofilchem, Italy) and were interpreted according to European Committee on Antimicrobial Susceptibility Testing (EUCAST) break points. ⁶ To attribute the resistance mechanism for the selected E. cloacae complex isolates, a synergistic test was applied using an enzymatic inhibition testing with clavulanic acid and cloxacillin and/or phenyl-boronic acid to detect ESBLs and AmpC enzymes, in that order, as reported elsewhere. ^{Reference Perez7} No isolate with carbapenemase production was included in this study, and for this study, all isolates were screened for a negative result using a blue-carba test to exclude class A and B carbapenemases and an OKN K-set immunochromatographic assay to exclude OXA-48-like production (ie, a carbapenemase with low hydrolysis activity for carbapenems and eventually resulting in a negative blue-carba test). ^{Reference Glupczynski, Jousset and Evrard8}

Among the 123 isolates, 39 (31.7%) and 84 (68.3%) were characterized as derepressed and not-derepressed AmpC-β-lactamase producers, respectively, according to phenotypic testing. For derepressed AmpC-β-lactamase producers, 36 of 39 isolates (92.3%) were resistant to ertapenem according to disc-diffusion testing. For all isolates, meropenem and ceftazidime/avibactam showed excellent activity: MIC₉₀ of 0.75 and 2.0 mg/L, respectively. No ceftazidime/avibactam resistance was observed among the isolates. However, 39 of 123 isolates (31.7%) were resistant to C/T and 37 of these 39 (94.9%) were derepressed AmpC-β-lactamase producers. Only 2 derepressed AmpC isolates were susceptible to C/T when EUCAST break points (≤1.0 and >1.0 mg/L, for susceptible and resistant, respectively) were considered (Fig. 1).

Fig. 1. Distribution of ceftolozane/tazobactam minimum inhibitory concentrations of 39 derepressed AmpC β-lactamase–producing organisms and 84 organisms not producing derepressed AmpC-β-lactamase. Black bars represent organisms expressing derepressed AmpC-β-lactamases based on positive results with cloxacillin synergy testing, and gray bars represent organisms not expressing derepressed AmpC-β-lactamases based on negative results with cloxacillin synergy testing.

Enterobacter spp, particularly Enterobacter cloacae complex, are the most problematic pathogens because of their potential for derepressing AmpC β-lactamase production, presenting an approximate rate that is 10-fold higher than that of other Enterobacterales. ^{Reference Pfaller, Jones and Marshall9}

Ceftolozane/tazobactam can overcome inactivation by ESBL β-lactamases, usually allowing maintenance of its activity against Enterobacterales producing the globally important ESBLs CTX-M-14 and CTX-M-15. On the other hand, C/T may be influenced by AmpC β-lactamase activity, based on the results presented here, despite the fact that this drug combination is considered to have improved steric hindrance to prevent AmpC β-lactamase-mediated hydrolysis.

High-level AmpC expression appears to confer a fitness cost to an organism because of the high metabolic energy required to express ampC regulation. Nevertheless, in the face of a persistent stimulus (eg, β-lactam exposure), this phenotype may be sustained, and for this reason, C/T (like many expanded-spectrum cephalosporins) may be discouraged for the treatment of infections caused by E. cloacae complex.

Some other points should also be considered. First, C/T was primarily designed to treat Pseudomonas infections, and because this bacterium is highly susceptible to this drug, ^{Reference Rodríguez-Núñez, Periañez-Parraga and Oliver10} it is reasonable to conclude that Pseudomonas usually does not derepress chromosomal AmpC enzymes, even those with an extensively drug-resistant phenotype. Second, C/T is an inappropriate drug for use in empirical therapeutic approaches; it should be preserved for the specific niche for Pseudomonas infection. Third, high inactivation of C/T by derepressed AmpC organisms justifies strict monitoring of its resistance level in both Enterobacterales and Pseudomonas, particularly those associated with poor outcomes when high MICs (>2 g/mL) are observed. ^{Reference Rodríguez-Núñez, Periañez-Parraga and Oliver10}

In conclusion, in this study, we identified a mechanism that seems to be predictive of high C/T resistance levels. More prominent among bacteria that are capable of derepressing AmpC β-lactamases, mainly the E. cloacae complex (but also Pseudomonas on a minor scale), this resistance to C/T should lead to stricter use and monitoring of this drug.