KPC-31 expressed in a ceftazidime/avibactam-resistant Klebsiella pneumoniae is associated with relevant detection issues

Sir, Infections mediated by carbapenemase-producing Enterobacterales (CPE) are spreading worldwide and represent one of the major challenges for antimicrobial chemotherapy due to the dearth of available therapeutic options.1 CPE strains can produce different carbapenemases. KPC-, OXA-48-, VIM-, IMP- and NDM-type variants are those most prevalent worldwide.2 Surveillance of colonization by CPE is becoming increasingly important for infection control measures and selective chromogenic media are commonly used for rectal swab screening,3 while rapid confirmation of carbapenemase production by suspect isolates can be carried out with simple colorimetric tests such as Carba NP.4 Identification of the carbapenemase(s) produced by CPE isolates is important not only for surveillance purposes but also for antimicrobial stewardship, since the new β-lactamase inhibitor combinations (BLICs) exhibit differential activity against CPE producing different types of carbapenemases.5 Identification of the carbapenemase type can be achieved by detection of the carbapenemase genes by genotypic methods (e.g. real-time PCR) or by detection of the enzymes by lateral flow immune assays (LFIAs).6 The latter methods are becoming increasingly popular, being very rapid and relatively inexpensive.7 Ceftazidime/avibactam is a novel BLIC effective against CPE strains producing serine carbapenemases, including KPC- and OXA-48-type enzymes.5 Resistance to ceftazidime/avibactam, however, has already been described and is due to several mechanisms, including mutations of KPC, alterations affecting Ompk35 and Ompk36 porins and/or overexpression of KPC.8–11 In this report we describe the characterization of a ceftazidime/avibactam-resistant KPC-producing Klebsiella pneumoniae strain selected after treatment with the drug and producing KPC-31, a D179Y variant of KPC-3, already shown to be responsible for ceftazidime/avibactam resistance.9,11 We also report on the failure to detect this mutant enzyme by rapid LFIAs for detection of KPC carbapenemase and by the Carba NP test, and the failure of this strain to grow on a selective chromogenic medium. K. pneumoniae KP-PG-0618 was isolated from a screening sample (rectal swab) from an inpatient at Perugia University Hospital (central Italy) at the beginning of June 2018. The patient had previously experienced (April 2018) a surgical site infection caused by a ceftazidime/avibactam-susceptible KPC-producing K. pneumoniae and a carbapenem-resistant Acinetobacter baumannii, which had been treated with ceftazidime/avibactam plus ampicillin/sulbactam for 2 weeks. Identification of KP-PG-0618 at the species level was carried out using MALDI-TOF MS (MALDI Biotyper®, Bruker Daltonik GmbH, Bremen, Germany). Antimicrobial susceptibility testing, carried out with broth microdilution using lyophilized custom plates [Sensititre (Thermo Fisher Scientific, Basingstoke, UK) and MICRONAUT (Merlin, Bornheim-Hesel, Germany), ISO 20776-1 standard inoculum] and interpreted according to the EUCAST clinical breakpoints (http://www.eucast.org/), revealed that KP-PG-0618 was resistant to expanded-spectrum cephalosporins (ceftazidime, cefotaxime and cefepime), BLICs (piperacillin/tazobactam, ceftazidime/avibactam and ceftolozane/tazobactam) and aztreonam, but retained susceptibility to carbapenems (with the exception of ertapenem). The strain was also resistant to ciprofloxacin and trimethoprim/sulfamethoxazole, while being susceptible to aminoglycosides and colistin. See Table 1. Antimicrobial susceptibility of K. pneumoniae KP-PG-0618, determined using the broth microdilution method I, susceptible increased exposure; R, resistant; S, susceptible standard dosing regimen. Antimicrobial susceptibility of K. pneumoniae KP-PG-0618, determined using the broth microdilution method I, susceptible increased exposure; R, resistant; S, susceptible standard dosing regimen. Interestingly, KP-PG-0618 had been isolated on ChromID® ESBL medium (bioMérieux, Marcy-l’Étoile, France), but not on ChromID® Carba selective medium for CPE (bioMérieux), which yielded a negative screening result, and tested negative for carbapenemase production with the RAPIDEC® CARBA NP test (bioMérieux). Molecular analysis by real-time PCR12 revealed that KP-PG-0618 was positive for a blaKPC gene. Surprisingly, the strain tested negative with LFIA methods for carbapenemase detection, including KPC K-SET (Coris BioConcept, Gembloux, Belgium) and NG-Test Carba 5 (NG Biotech, Guipry, France). To further investigate the mechanisms of resistance, the K. pneumoniae KP-PG-0618 strain was subjected to WGS using an Illumina MiSeq platform (Illumina Inc.) with a paired-end approach (2 × 300 bp). De novo assembly of WGS data was performed using the SPAdes software (http://cab.spbu.ru/software/spades/). WGS analysis revealed that KP-PG-0618 belonged to ST512 and carried a complex resistome, including the resident blaSHV-11 and two acquired β-lactamase genes (blaKPC-31 and blaTEM-1), and resistance determinants for aminoglycosides [aph(3')-Ia, aac(6′)-Ib′ and ANT(3″)-Ia], chloramphenicol (catA1), sulphonamides (sul1), trimethoprim (dfrA12) and macrolides [mph(A)]. Additionally, the strain presented a non-functional Ompk35 (stop codon at amino acid position 89), previously associated with reduced ceftazidime/avibactam susceptibility,9 and a Gly134-Asp135 duplication in OmpK36, associated with reduced carbapenem susceptibility.13 Of note, KPC-31 (accession number MAPH01000113) is a single amino acid variant of KPC-3 (D179Y) that was previously reported in Italy9 and in the USA,11 and shown to confer a ceftazidime/avibactam-resistant phenotype associated with susceptibility or decreased resistance to carbapenems, piperacillin/tazobactam, cefepime, ceftriaxone and aztreonam, compared with KPC-3.11 The mechanism of resistance to ceftazidime/avibactam of KPC-31 has not been kinetically investigated, but the same D179Y mutation in a KPC-2 context was shown to increase the enzyme’s affinity to ceftazidime and decrease inhibition by avibactam.14 To the best of our knowledge, this is the second ceftazidime/avibactam-resistant KPC-producing K. pneumoniae isolate reported in Italy.9 The present findings confirmed that ceftazidime/avibactam-resistant mutants producing KPC-31 can be selected following a prolonged exposure to ceftazidime/avibactam,9 and underscored that similar mutants have relevant diagnostic implications, since they might be unable to grow on selective chromogenic media for CPE, while the detection of carbapenemase production might be missed by colorimetric tests such as the RAPIDEC® CARBA NP and LFIA tests. Under these circumstances, genotypic testing in combination with the resistance profile remains the only reliable option for suspecting the resistance mechanism, which should then be confirmed by sequencing of the allelic variant. Clinical microbiology laboratories should be aware of these diagnostic implications, which could impair and delay detection of this potentially transferable resistance mechanism to ceftazidime/avibactam. We thank Professor Thierry Naas, Professor Pierre Bogaerts and Dr Romano Mattei for providing NG-Test Carba 5, KPC K-SET and the RAPIDEC® CARBA NP test, respectively. This work was supported by internal funding. None to declare.