Introduction

Pseudomonas aeruginosa is a common cause of respiratory infections in cystic fibrosis (CF) patients [1]. P. aeruginosa strains in CF commonly become mucoid, an exopolysaccharide alginate overproducing phenotype that initiates the chronic infection stage of the disease and a decline in pulmonary function [2,3,4].

As CF patients age, mechanisms of resistance to P. aeruginosa accumulate [5]. As a result, newer antibiotics are needed. Ceftolozane/tazobactam is a novel cephalosporin/β-lactamase inhibitor that maintains activity against many multidrug-resistant P. aeruginosa isolates. Resistance appears not to be driven by single-step mutations, which is promising for chronic respiratory infections in CF [6,7,8,9]. Although ceftolozane/tazobactam is not currently approved for treatment of pulmonary infections, it has been successfully used in acute pulmonary CF exacerbations [10, 11].

The purpose of this study was to compare the efficacy of ceftolozane/tazobactam against mucoid P. aeruginosa (MP) strains isolated from CF patients with that of standard-of-care (SOC) agents. This study also assessed resistance patterns of MP and specifically the ceftolozane/tazobactam minimum inhibitory concentrations (MICs).

Methods

Bacterial Strains

Ten clinical MP isolates from CF patients at an academic medical center were tested for susceptibility, and time–kill analyses were performed with ceftolozane/tazobactam and the following SOC antimicrobials: ceftazidime, cefepime, meropenem, ciprofloxacin, tobramycin, and polymyxin B. This study was approved by the Institutional Biosafety Committee. This article does not contain any new studies with human or animals performed by any of the authors. This study does not meet the definition of a clinical trial, and so was not registered at ClinicalTrials.gov.

Antimicrobials

Ceftolozane powder was obtained from Merck (Kenilworth, NJ, USA). Tazobactam (Sigma-Aldrich, St Louis, MO, USA), ceftazidime (Sagent Pharmaceuticals, Schaumburg, IL, USA), cefepime and meropenem (Sandoz, Princeton, NY, USA), ciprofloxacin and tobramycin (Hospira, Lake Forest, IL, USA), and polymyxin B (XGen Pharmaceuticals, Big Flats, NY, USA) were obtained commercially.

Susceptibility Testing

MICs were determined with Etest according to standard procedures and read by two individuals. If there was a difference between the readings, a third individual analyzed the result. If there was a difference in MIC of more than a half dilution, the Etest was repeated.

Time–Kill Experiments

Time–kill experiments were performed in duplicate in Mueller–Hinton broth (BBL; Becton, Dickinson, Sparks, MD, USA) in 24-well macrowell plates as previously described [12]. The plate was filled with 100 µL of antibiotic stock solution, 200 µL of a 1:10 dilution of a 1.75 McFarland-standard organism suspension for a target bacterial inoculum of 106 colony-forming units (CFU) per milliliter, and sufficient volume of Mueller–Hinton broth for a total volume of 2 mL. Sample aliquots of 100 µL were obtained from each well at 0, 4, 8, and 24 h, and were serially diluted in cold 0.9% sodium chloride. Bacterial counts were determined with a Whitely automatic spiral plater (Don Whitely Scientific, Shipley, UK). The lower limit of detection for time–kill studies was 101 CFU/mL. Plates were incubated at 37 °C for 24 h, at which time colony counts were performed with a ProtoCOL colony counter (Synoptics, Frederick MD, USA). All strains were tested against ceftolozane/tazobactam (1.5 and 3 g), ceftazidime (2 g), cefepime (2 g), ciprofloxacin (400 mg), meropenem (1 g), tobramycin (25 mg/L), and polymyxin B (1.25 mg/kg) with use of free physiologic peak concentrations. The physiologic free peak concentrations were 60/12.6, 120/25.2, 151, 112, 3.2, 110, 25, and 2.1 mg/L, respectively [13,14,15,16,17,18]. Time–kill curves were generated by plotting mean colony counts (log10 CFU/mL) versus time to compare 24-h killing effects (Fig. 1). Bactericidal activity was defined as a 3 log10 CFU/mL reduction or greater from the baseline.

Statistical Analysis

Differences in log10 CFU per milliliter were analyzed by analysis of variance with Tukey’s post hoc test. P < 0.05 was considered significant. All statistical analyses were performed with IBM SPSS Statistics (version 24.0; IBM, Armonk, NY, USA).

Fig. 1
figure 1

Time–kill evaluation results. CFU colony-forming units

Full size image

Results

Susceptibility Testing

All strains were susceptible to ceftolozane/tazobactam with MICs that ranged from 0.032 to 1.5 mg/L (susceptible if MIC is 4/4 mg/L or less) (Table 1) [19]. Meropenem and polymyxin B resulted in 100% susceptibility. Although cefepime and ciprofloxacin were the least effective agents, both were active against eight of the ten strains.

Table 1 Susceptibility of isolates
Full size table

,

Time–Kill Studies

The mean starting inoculum was 6.29 ± 0.22 log10 CFU/mL. The most potent activity was for polymyxin B, tobramycin, meropenem, and ceftolozane/tazobactam 3 g, demonstrated by reductions in starting inoculums of 5.07 ± 0.45, 4.58 ± 2.2, 4.76 ± 0.71, and 4.17 ± 0.94 log10 CFU/mL, respectively. Bactericidal activity was also observed with ceftolozane/tazobactam 1.5 g, cefepime, and ciprofloxacin, with average reductions of 3.74 ± 0.99, 3.42 ± 1.37, and 3.23 ± 1.98 log10 CFU/mL, respectively, from the starting inoculum.

Ceftolozane/tazobactam was bactericidal against 80% of evaluated strains. Despite lack of bactericidal activity against two strains, there were no differences between the inoculum reductions of ceftolozane/tazobactam and those of the other agents (P > 0.05), with the exception of tobramycin. No regrowth was noted at 24 h for any strain with either ceftolozane/tazobactam dose. Tobramycin displayed bactericidal activity rapidly (within 4 h) against seven of the ten strains and achieved killing to the limit of detection in nine strains by 24 h. However, strain HR21 displayed regrowth within 8 h after tobramycin exposure, although this could be explained by a tobramycin MIC greater than 1024 mg/L. When this tobramycin-resistant strain was excluded, tobramycin demonstrated the most potent killing effect, with an average reduction in starting inoculum of 5.26 ± 0.24 log10 CFU/mL.

The least active agents were ceftazidime and ciprofloxacin. Ceftazidime displayed bactericidal activity against only seven strains, with an average reduction in starting inoculum of 2.91 ± 2.22 log10 CFU/mL. Two of ten ceftazidime-treated strains displayed regrowth at 24 h. Additionally, five strains treated with ceftazidime displayed less potent reductions in starting inoculum than with the comparator agents (P < 0.05). Ciprofloxacin displayed bactericidal activity against six strains, with an average reduction in starting inoculum in these strains of 4.62 ± 0.36 log10 CFU/mL. Despite initial killing activity in all strains, 40% of ciprofloxacin-treated strains displayed regrowth at 24 h.

Discussion

Ceftolozane/tazobactam demonstrates bactericidal killing similar to that of SOC antimicrobials with regard to MP isolated from CF patients. The highest MIC with ceftolozane/tazobactam, 1.5 mg/L, corresponded to the strain that was resistant to the other cephalosporins tested. Ceftolozane/tazobactam MICs did not correlate with non-β-lactam MICs.

The MIC50/MIC90 of ceftolozane for P. aeruginosa isolates from adults was reported as 0.5/2 mg/L [9]. In that study, the ceftolozane MIC was 8 mg/L or less for 94% of MP isolates, with a mean MIC of 0.8 mg/L. Another study found that the ceftolozane MIC was greater than 8 mg/L for 36% of P. aeruginosa isolates from CF patients ; 57% were not susceptible to ceftazidime [8]. Another study evaluated ceftolozane/tazobactam against P. aeruginosa in CF: 48% of isolates were MP, and the MIC50/MIC90 was 2/8 mg/L, with 86% susceptibility [7]. However, there was a high frequency of multidrug-resistant isolates. In a trial evaluating the dynamics of resistance development using wild-type and mutator strains of P. aeruginosa, resistance development with ceftolozane/tazobactam was slower than with comparators, reaching eight times the MIC (basal MIC of 0.5 ug/mL) after 7 days of experiments (versus 64 times the MIC at 4 days with ceftazidime) in wild-type strains. In mutator strains, resistance development was much more rapid and significant with all compounds [20]. Our study, in comparison, showed MIC50/MIC90 of 0.25/0.5 mg/L, and all isolates were susceptible according to the breakpoint of 4/4 mg/L or less [19]. It is unlikely that our study included any mutator strains.

The activity of ceftolozane alone against P. aeruginosa isolates, including a wild-type reference, its alginate-hyperproducing mucoid mutant (mucA knockout), its mismatch-repair-deficient hypermutatable mutant (mutS knockout), and the double mucA–mucS mutant [6], has been studied. These phenotypes are fairly relevant in chronic conditions such as CF. Ceftolozane was found to have potent, concentration-independent biofilm bactericidal activity against both the mucoid strain and the hypermutatable strains. This study suggests that resistance to ceftolozane cannot be driven by single-step mutations in wild-type, mucoid, or hypermutatable strains, which is important in infections where P. aeruginosa eradication from the respiratory tract is not generally attainable after colonization has been established.

Although the only currently approved ceftolozane/tazobactam dosage is 1.5 g every 8 h, there is an ongoing clinical trial evaluating the safety and efficacy of ceftolozane/tazobactam at 3 g every 8 h for pneumonia [13, 21]. In addition, the 3-g dose has been used safely with success in patients with CF exacerbations [10, 11]. In one study, ceftolozane/tazobactam use resulted in a pharmacodynamic target attainment of more than 90% at MICs up to 8, 4, and 2 mg/L with 1.5 g every 8 h and 16, 8, and 4 mg/L with 3 g every 8 h with free time above the MIC target of 39%, 60%, and 100%, respectively [11]. There is no currently accepted free time above the MIC required for successfully treating acute CF exacerbations with ceftolozane/tazobactam. Until this is established, the authors of that study recommend the use of 3 g for CF patients with pulmonary exacerbations. Our study demonstrates that ceftolozane/tazobactam achieved bactericidal killing in MP strains from CF patients, but a more potent reduction of inoculum occurred with the 3-g dose as compared with the 1.5-g dose. Given our results, use of the higher-dose regimen seems appropriate for these patients.

There are some limitations to the study, as the data represent only in vitro efficacy. Only ten isolates were included in the study; therefore, we cannot extrapolate the conclusions of the time–kill analysis to other isolates, especially ones demonstrating higher ceftolozane/tazobactam MICs. In addition, although a deeper characterization of resistance mechanisms would have been interesting, we did not have the financial resources to determine β-lactam resistance mechanisms for ceftazidime and cefepime at this time.

Conclusion

As resistance in P. aeruginosa increases, newer antibiotics with expanded activity are needed. Ceftolozane/tazobactam (1.5 and 3 g) demonstrates activity against MP strains derived from CF patients comparable to that of SOC agents. However, further in vitro modeling and clinical trials are warranted.