|Home | About | Journals | Submit | Contact Us | Français|
CXA-101, previously designated FR264205, is a new antipseudomonal cephalosporin. We evaluated the activity of CXA-101 against a highly challenging collection of β-lactam-resistant Pseudomonas aeruginosa mutants selected in vitro and after antipseudomonal treatment of intensive care unit (ICU) patients. The in vitro mutants investigated included strains with multiple combinations of mutations leading to several degrees of AmpC overexpression (ampD, ampDh2, ampDh3, and dacB [PBP4]) and porin loss (oprD). CXA-101 remained active against even the AmpD-PBP4 double mutant (MIC = 2 μg/ml), which shows extremely high levels of AmpC expression. Indeed, this mutant showed high-level resistance to all tested β-lactams, except carbapenems, including piperacillin-tazobactam (PTZ), aztreonam (ATM), ceftazidime (CAZ), and cefepime (FEP), a cephalosporin considered to be relatively stable against hydrolysis by AmpC. Moreover, CXA-101 was the only β-lactam tested (including the carbapenems imipenem [IMP] and meropenem [MER]) that remained fully active against the OprD-AmpD and OprD-PBP4 double mutants (MIC = 0.5 μg/ml). Additionally, we tested a collection of 50 sequential isolates that were susceptible or resistant to penicillicins, cephalosporins, carbapenems, or fluoroquinolones that emerged during treatment of ICU patients. All of the mutants resistant to CAZ, FEP, PTZ, IMP, MER, or ciprofloxacin showed relatively low CXA-101 MICs (range, 0.12 to 4 μg/ml; mean, 1 to 2 μg/ml). CXA-101 MICs of pan-β-lactam-resistant strains ranged from 1 to 4 μg/ml (mean, 2.5 μg/ml). As described for the in vitro mutants, CXA-101 retained activity against the natural AmpD-PBP4 double mutants, even when these exhibited additional overexpression of the MexAB-OprM efflux pump. Therefore, clinical trials are needed to evaluate the usefulness of CXA-101 for the treatment of P. aeruginosa nosocomial infections, particularly those caused by multidrug-resistant isolates that emerge during antipseudomonal treatments.
The growing threat of Pseudomonas aeruginosa antimicrobial resistance results from, on the one hand, the extraordinary capacity of this microorganism for developing resistance to almost every available antibiotic by the selection of mutations in chromosomal genes and, on the other hand, the increasing prevalence of transferable resistance determinants, particularly those encoding class B carbapenemases (or metallo-β-lactamases [MBLs]) or extended-spectrum β-lactamases (ESBLs), frequently cotransferred with genes encoding aminoglycoside-modifying enzymes (16, 19).
Particularly noteworthy among the mutation-mediated resistance mechanisms are those leading to the repression or inactivation of the porin OprD, conferring resistance to carbapenems (5, 7, 24, 26), or those leading to the hyperproduction of the chromosomal cephalosporinase AmpC, such as AmpD or PBP4 inactivation (11, 20), causing resistance to penicillins and cephalosporins. In addition, mutations leading to the upregulation of one of several efflux pumps encoded in the P. aeruginosa genome may confer resistance or reduced susceptibility to multiple agents, including almost all β-lactams, fluoroquinolones, and aminoglycosides (3, 18, 25). Furthermore, the accumulation of these chromosomal mutations can lead to the emergence of multidrug-resistant (MDR) (or even pan-antibiotic-resistant) strains which eventually may be responsible for outbreaks in the hospital setting (4). Indeed, sequential development of resistance to almost all available antipseudomonal agents is a not uncommon outcome of the treatment of severe P. aeruginosa infections, frequently occurring in intensive care unit (ICU) patients or patients with hematological disease (2, 10).
Unfortunately, over the last 2 decades there has been very limited progress in developing novel antipseudomonal agents which can overcome MDR in P. aeruginosa (19). CXA-101 (previously designated FR264205), a new cephalosporin with promising characteristics for the treatment of P. aeruginosa infections, appears stable against the most common resistance mechanisms driven by mutation in this species (27, 28).
The objective of this study was to investigate the activity of CXA-101 against β-lactam-resistant P. aeruginosa mutants selected in vitro and after antipseudomonal treatment of ICU patients. The in vitro mutants investigated included a highly challenging collection of strains with multiple combinations of mutations leading to various levels of AmpC overexpression (ampD, ampDh2, ampDh3, and dacB [PBP4]) and porin loss (oprD). Additionally, a well-characterized collection of 50 sequential isolates susceptible or resistant to penicillicins, cephalosporins, carbapenems, or fluoroquinolones (first-line antipseudomonal agents) that emerged during treatment of ICU patients was tested.
PAO1 oprD (PAOD) mutants were obtained by overnight plating of Luria-Bertani cultures in 4 μg/ml imipenem (IMP)-Mueller-Hinton agar. Three independent mutants were confirmed, after passage in antibiotic-free Mueller-Hinton agar, by oprD amplification and sequencing using previously described primers and conditions (7). Single and multiple PAO1 mutants with mutations in genes involved in the hyperproduction of the chromosomal β-lactamase AmpC (ampD [PAΔD], ampDh2 [PAΔDh2], ampDh3 [PAΔDh3], and the recently characterized dacB [PAΔdB], encoding PBP4) were obtained from previous studies (12, 20). OprD-AmpD and OprD-PBP4 double mutants were constructed in this study (from each of the three oprD mutants generated) using the cre-lox system for gene deletion in P. aeruginosa, following previously described procedures (12, 20). ampC expression in these strains was evaluated by real-time reverse transcription-PCR under basal and induced (50 μg/ml cefoxitin) conditions, using previously described primers and protocols (12).
In a previous study, development of resistance to β-lactams and/or fluoroquinolones during antimicrobial therapy was documented for 21 of 102 (20.6%) patients with P. aeruginosa infections admitted to a Spanish ICU between September 2002 and November 2003 (10). Those strains were fully characterized using various molecular methods. In all cases, the sequential resistant isolates were found through pulsed-field gel electrophoresis (PFGE) to be clonally related to the preceding susceptible isolates, indicating that resistance was a consequence of the selection of mutations and not of strain replacement (10). The studied patients had been treated with antipseudomonal agents for 19.0 ± 20.0 days. PFGE additionally revealed that a different P. aeruginosa clone infected each of the 21 patients, excluding interpatient spread as the origin of resistant strains (10). In isolates from 10 of these patients, the mechanisms of resistance to penicillins and cephalosporins were characterized at the molecular level. In all cases, AmpC hyperproduction (expression increased 12- to 657-fold) was demonstrated and was found to have been produced by classical AmpD inactivation (four cases) or by the recently described PBP4 mutation (four cases) or by both mechanisms together (two cases) (11, 20). In this work, the sequential susceptible-resistant isolates (50) from these 21 ICU patients were evaluated.
The MICs of CXA-101, ceftazidime (CAZ), cefepime (FEP), piperacillin (PIP), PIP-tazobactam (PTZ), IMP, meropenem (MER), ciprofloxacin (CIP), and tobramycin (TOB) were determined by standard CSLI broth microdilution (M100-S18). CXA-101 (lot 3F02) was provided by Calixa Therapeutic Inc. Other antibiotics were purchased from commercial sources. The concentrations tested ranged from 0.12 μg/ml to 256 μg/ml. The breakpoints defined by the CLSI (M100-S18) were used for all comparator antibiotics. Experiments using P. aeruginosa reference strains PAO1 and ATCC 27853 were performed in parallel for quality control. MDR strains were defined as those shown to be nonsusceptible to at least three of the following four antibiotics: CAZ, IMP, CIP, and TOB.
Sequencing of the three putative OprD mutants revealed mutations in oprD in all cases, and each of the mutants had a different type of mutation: PAOD1 had a nonsense mutation (G194A [W65X]), PAOD2 had a frameshift mutation [1-bp A insertion in nucleotide 335], and PAOD3 had a missense mutation (G3A [M1I]) affecting the initiation codon.
The MICs of CXA-101, CAZ, FEP, PIP, PTZ, aztreonam (ATM), IMP, and MER for the collection of mutants studied are shown in Table Table1.1. Basal and cefoxitin-induced ampC expression data are also shown. CXA-101 conserved activity against all single and combined mutations, leading to AmpC hyperproduction. CXA-101 MICs were not modified by ampD inactivation (MIC, 0.5 μg/ml), and the inactivation of multiple (two or three) ampD genes or dacB (PBP4) produced only a twofold increase in MIC (MIC, 1 μg/ml). The double AmpD-PBP4 mutant, showing extremely high ampC expression, had an MIC of 2 μg/ml (fourfold increase in MIC). In contrast, all other tested penicillins (PIP and the PTZ combination), cephalosporins (CAZ and FEP), and monobactams (ATM) were notably affected by the several mutations, leading to AmpC hyperexpression: ampD inactivation alone increased MICs two- to eightfold, although the MICs were still within the high end of susceptible ranges. PBP4 inactivation alone produced clinical resistance to CAZ, FEP, and ATM (and PIP and PTZ as well, when EUCAST [www.eucast.org] breakpoints [R > 16 μg/ml] are considered). Finally, the double AmpD-PBP4 mutant showed high-level resistance to PIP, PTZ, CAZ, FEP, and ATM.
As expected, oprD inactivation resulted in resistance to IMP and increased the MER MIC fourfold, although it was still in the susceptible range. No major differences were observed in the results for the three OprD mutants tested. MER MICs were notably increased further for the OprD-PBP4 and OprD-AmpD double mutants, surpassing the breakpoints for nonsusceptibility. Therefore, CXA-101 was the only β-lactam tested that showed conserved activity against all of these mutants (Table (Table11).
The MICs of CXA-101, CAZ, FEP, PTZ, IMP, MER, CIP, and TOB for the complete collection of 50 sequential susceptible-resistant isolates from 21 ICU patients are shown in Table Table2.2. Table Table22 also contains additional information regarding the source of infection, the antibiotics received before the emergence of resistance, and the clonal relatedness (PFGE) of the tested isolates. Development of resistance (according to CLSI breakpoints) to CAZ, FEP, IMP, MER, and PTZ was documented for 81, 67, 57, 52, and 43%, respectively, of the 21 patients. Resistance to PTZ increased to 76% when the EUCAST breakpoints were used. Pan-β-lactam (CAZ, FEP, PTZ, IMP, and MER) resistance development (according to CLSI breakpoints) was documented for 29% of the patients. Emergence of CIP resistance was documented for 57% of the patients. MDR development was observed for 38%. In contrast to the findings obtained with β-lactams and fluoroquinoles, mutational resistance development (according to CLSI breakpoints) to TOB during antipseudomonal treatment did not occur significantly (Table (Table22).
Table Table33 shows the CXA-101 MIC50, MIC90, MIC range, and geometric mean MICs for the overall collection of isolates tested, as well as for different subsets of isolates (sequential isolates resistant to each of the antibiotics tested, MDR isolates, pan-β-lactam-resistant isolates, and AmpC-hyperproducing isolates). CXA-101 showed MICs ≤ 4 μg/ml against the complete collection of sequential resistant isolates, with an MIC50/MIC90 of 0.5/2 μg/ml, a range of 0.12 to 4 μg/ml, and an MIC geometric mean of 0.72 μg/ml. These values remained within a twofold dilution for the different subsets of isolates resistant to each of the antibiotics tested (Table (Table3).3). Similarly, CXA-101 retained activity against the 10 AmpD- and/or PBP4 AmpC-hyperproducing mutants (MIC range, 0.25 to 4 μg/ml; mean, 0.71 μg/ml) (Table (Table3).3). CXA-101 MICs were still relative low for the two high-level-β-lactam-resistant AmpD-PBP4 double mutants included in the collection (21): the isolate from patient 16 had a CXA-101 MIC of 2 μg/ml, and the isolate from patient 8, additionally overexpressing MexAB-OprM as reported previously (13), had an MIC of 4 μg/ml (Table (Table2).2). Finally, CXA-101 retained certain activity against all pan-β-lactam-resistant isolates (MIC range, 1 to 4 μg/ml; mean, 2.5 μg/ml) and MDR isolates (MIC range, 0.5 to 4 μg/ml; mean, 1.83 μg/ml), although there was an increase in the absolute MICs. Clearly, those naturally occurring mutational resistance mechanisms have some impact on overall susceptibility of CXA-101 but to a significantly lesser degree compared to those for all other β-lactams tested.
The overall prevalence of nosocomial infections caused by MDR P. aeruginosa strains is increasing and is associated with significant morbidity and mortality (1, 14, 22). Moreover, sequential development of resistance to almost all antipseudomonal agents is a not uncommon outcome of the treatment of severe P. aeruginosa infections (2, 10).
Although colistin has reemerged as a reasonable option for the treatment of infections caused by MDR Gram-negative bacteria (15), there is still concern that this agent might not be as efficacious or safe as the first-line antibiotics (such as β-lactams or fluoroquinoles) used to treat susceptible microorganisms. Among the recently introduced antibiotics, only doripenem (carbapenem) shows slightly enhanced antipseudomonal activity (19). Nevertheless, cross-resistance to doripenem with MER has been clearly documented and involves the inactivation of OprD together with the hyperexpression of AmpC and/or the upregulation of the MexAB-OprM efflux pump (21).
In early studies, Takeda et al. (28) tested 193 P. aeruginosa isolates from Japan, documenting an MIC50/MIC90 of 0.5/1 μg/ml for FR264205 (CXA-101), although the collection included a very limited number of cephalosporin (CAZ)-resistant isolates (13 of 193). These initial studies also showed conserved activity against mutants overexpressing different efflux pumps, as well as against simple AmpD mutants, showing only moderately increased AmpC expression and β-lactam resistance (27).
In this work, we evaluated the activity of CXA-101 against a highly challenging collection of β-lactam-resistant P. aeruginosa mutants selected in vitro or after antipseudomonal treatment of ICU patients. CXA-101 was the only β-lactam tested that remained fully active against the Opr-AmpD and OprD-PBP4 double mutants. Additionally, we tested a well-characterized collection of 50 sequential isolates that were susceptible-resistant to penicillicins, cephalosporins, carbapenems, or fluoroquinolones that emerged during treatment of ICU patients. Although a certain increase in MIC was documented for isolates from some of the patients (up to eightfold in 4 of 21), all CAZ-, FEP-, PTZ-, IMP-, MER-, or CIP-resistant mutants, as well as pan-β-lactam-resistant strains, showed CXA-101 MICs ≤ 4 μg/ml. Additionally, as found for the in vitro mutants, CXA-101 retained activity (MICs ≤ 4 μg/ml) against the natural AmpD-PBP4 double mutants, even when additionally overexpressing the MexAB-OprM efflux pump. Whether CXA-101 is significantly affected by overexpression of other efflux pumps, such as MexXY-OprM, which is a frequent mechanism of resistance to particular cephalosporins such as FEP (8, 9), needs to be further explored. Nevertheless, we have tested CXA-101 against isolates from a recent large outbreak by a FEP-resistant (MICs ≥ 16 μg/ml) P. aeruginosa clone that overexpressed MexXY-OprM (23) and obtained low MICs (1 μg/ml).
Livermore et al. (17) recently tested a highly challenging panel of 56 MDR P. aeruginosa isolates from cystic fibrosis patients, finding that CXA-101 was the most active drug, although up to 36% of the isolates showed an MIC of >8 μg/ml. We have also recently evaluated a collection of carbapenem-resistant and MDR P. aeruginosa isolates obtained from a multicenter study in Spain (13). CXA-101 showed an MIC50/MIC90 of 1/4 μg/ml; interestingly, almost all of the isolates showing CXA-101 MICs > 8 μg/ml (5%) produced a horizontally acquired MBL or ESBL, which is consistent with recent findings showing that CXA-101 does not overcome these horizontally acquired resistance mechanisms (6, 17). Although the prevalence of MBLs and ESBLs is still low in most hospitals worldwide, this is an important aspect to be considered, specially in certain geographic areas.
In summary, the data presented on CXA-101 activity against in vitro and in vivo β-lactam-resistant mutants are encouraging, but susceptibility-resistance breakpoints still need to be defined by official agencies (CLSI, EUCAST, etc.) Similarly, the potential usefulness of CXA-101 for the treatment of P. aeruginosa nosocomial infections, particularly those caused by MDR isolates that emerge during antipseudomonal treatments, needs to be evaluated in pertinent clinical trials.
This work was supported by a grant from Calixa Therapeutics Inc. and by the Ministerio de Ciencia e Innovación, Instituto de Salud Carlos III, through the Spanish Network for the Research in Infectious Diseases (REIPI C03/14 and RD06/0008).
Published ahead of print on 19 January 2010.