https://orcid.org/0000-0003-3606-0761
Abstract
Patients infected with difficult-to-treat Pseudomonas aeruginosa are likely to receive meropenem (MEM) empirically before escalation to ceftolozane/tazobactam (C/T). We assessed whether pre-exposure to MEM affected C/T resistance development on C/T exposure.
Nine clinical P. aeruginosa isolates were exposed to MEM 16 mg/L for 72 h. Then, isolates were serially passaged in the presence of C/T (concentration of 10 mg/L) for 72 h as two groups: an MEM-exposed group inoculated with MEM pre-exposed isolates and a non-MEM control group. At 24 h intervals, samples were plated on drug-free and drug-containing agar (C/T concentration 16/8 mg/L) and incubated to quantify bacterial densities (log10 cfu/mL). Growth on C/T agar indicated resistance development, and resistant population was calculated by dividing the cfu/mL on C/T plates by the cfu/mL on drug-free agar.
At 72 h, resistant populations were detected in 6/9 isolates. In five isolates, MEM exposure significantly increased the prevalence of ceftolozane/tazobactam-resistance development; the percentages of resistance population were 100%, 100%, 53.5%, 31% and 3% for the MEM-exposed versus 0%, 0%, 2%, 0.35% and ≤0.0003% in the unexposed groups. One isolate had a similar resistant population at 72 h between the two groups. The remaining isolates showed no development of resistance, regardless of previous MEM exposure.
MEM exposure may pre-dispose to C/T resistance development and thus limit the therapeutic utility of this β-lactam/β-lactamase inhibitor. Resistance may be a result of stress exposure or molecular-level mutations conferring cross-resistance. Further in vivo studies are needed to assess clinical implications of these findings.
Introduction
Pseudomonas aeruginosa with a difficult-to-treat (i.e. DTR-P. aeruginosa) phenotype is a rising problem with limited therapeutic options.1,2 When DTR-P. aeruginosa causes invasive infections, it is associated with significant morbidity and mortality.1–3 The introduction of ceftolozane/tazobactam has improved clinical outcomes compared to traditional prescribed regimens as ceftolozane/tazobactam demonstrates in vitro potency against P. aeruginosa isolates resistant to all conventional β-lactams.2,4,5 However, resistance to ceftolozane/tazobactam has been identified,6–8 and is most commonly due to alterations in Pseudomonas-derived cephalosporinases (PDC).9,10
Optimizing therapy for P. aeruginosa infections remains a challenge as ∼20% of P. aeruginosa isolates in the USA are carbapenem-resistant.11,12 Patients infected with P. aeruginosa, even DTR-P. aeruginosa, are likely to receive meropenem as empiric therapy before escalating to agents such as ceftolozane/tazobactam in response to antimicrobial susceptibility testing results. In the current study, we assessed whether pre-exposure to meropenem affected the development of ceftolozane/tazobactam resistance on drug exposure using a 72 h in vitro serial passage model. We used nine isolates obtained from patients: eight of whom received meropenem treatment (range 2–19 days) before escalation to ceftolozane/tazobactam, and then subsequently developed resistance to ceftolozane/tazobactam during therapy (range 6–22 days of therapy).13
Materials and methods
Antimicrobial agents
Commercially available ceftolozane/tazobactam (Zerbaxa; lot SP1840) and analytical-grade meropenem (Sigma-Aldrich, St. Louis, MO, USA, lot LRAC5653) were used for in vitro studies. Cation-adjusted Mueller–Hinton broth (CAMHB) (Becton, Dickinson and Company, Sparks, MD, USA) was used for the in vitro studies.
Bacterial isolates
Nine clinical P. aeruginosa isolates were obtained from patients at The Johns Hopkins Hospital.10 The nine isolates were resistant to piperacillin/tazobactam, cefepime and meropenem.10 Eight of nine isolates were resistant to ceftazidime.10 All isolates were stored at −80°C in skim milk before being sub-cultured twice on Trypticase™ agar with 5% sheep blood (Becton Dickinson and Company, Sparks, MD, USA) for use in the in vitro studies. Pseudomonas aeruginosa INT-6-8 and P. aeruginosa 2149 were used as the quality control strain for ceftolozane/tazobactam agar plates.
Meropenem 72 h pre-exposure
Isolates were sub-cultured on two consecutive days onto Trypticase™ agar with 5% sheep blood. A 5 McFarland standard was prepared. From the suspension, 100 µL was added to the broth mixture containing 16 mg/L of meropenem (total volume 10 mL). Tubes were placed in a shaking water bath at 37°C for 72 h interrupted by a wash step every 24 h to avoid drug degradation and antibiotic carry-over. At the 72 h final wash step, isolates were re-suspended in saline to make a 5 McFarland standard to be used as meropenem-exposed isolates for the ceftolozane/tazobactam serial passages.
Ceftolozane/tazobactam in vitro serial passages
Each isolate was serially passed in the presence of ceftolozane/tazobactam as two groups: (i) meropenem-pre-exposed group and (ii) no-pretreatment control group. The serial passage was completed in a similar way to Werth and colleagues. Briefly, the meropenem-exposed group was inoculated with 100 µL of a 5 MacFarland standard of the isolate made from the meropenem pre-exposure. The meropenem-non-exposed replicate was inoculated with 100 µL of a 5 MacFarland standard after a free sub-culture on two consecutive days onto Trypticase™ agar with 5% sheep blood of each test isolate. A 5 McFarland standard was made, and 100 µL were added to 9.9 mL of CAMHB supplemented with ceftolozane/tazobactam at a final ceftolozane concentration of 10 mg/L to approximate the free Cmin of ceftolozane/tazobactam. Tubes were placed in a shaking water bath at 37°C for 24 h. After the first 24 h, serial dilutions were made and 10 µL were plated onto both drug-free and ceftolozane/tazobactam-containing Mueller–Hinton agar. Plates were incubated overnight at 37°C, and the colonies were counted to quantify bacterial densities (log10 cfu/mL). After each 24 h passage, a wash step was conducted by centrifuging the tubes for 20 min. Excess CAMHB was decanted off and the cell pellet was reconstituted in saline to make a 5 McFarland standard. From the new solution, 100 µL was added to a fresh ceftolozane/tazobactam CAMHB mixture. The second and third serial passages were done as described before.14,15 Figure 1 provides a summary of the meropenem exposure and serial passage work flow.
Meropenem exposure and in vitro serial passage workflow overview. MEM, meropenem; C/T, ceftolozane/tazobactam; cfu, colony forming unit.
Ceftolozane/tazobactam agar plates
Mueller–Hinton agar powder was suspended in distilled water, brought to boil to dissolve completely and sterilized by autoclaving. After cooling, ceftolozane/tazobactam was added for a final concentration of 16/8 mg/L ceftolozane/tazobactam. Then, 15 mL of the solution was added to each sterile petri dishes, set while covered and stored at 2–8°C.
Ceftolozane/tazobactam agar plates quality control
To evaluate the quality of the prepared plates, a sample of plates was tested every day over the course of the serial passage by adding a 10 µL of suspended isolates to the centre of agar plates. Two isolates were tested: one as a positive growth control, PSA INT-6-8 (ceftolozane/tazobactam MIC of 64 mg/L) and one negative growth control, PSA 2149 (ceftolozane/tazobactam MIC of 2 mg/L).
Analysis
The resistant population at each step were assessed by taking the cfu/mL on ceftolozane/tazobactam containing plates divided by the cfu/mL on drug-free plates from the same passage. The percentage of the ceftolozane/tazobactam-resistant population was compared between meropenem-exposed and non-exposed passages of each isolate in singlet for each pre-treatment group. Growth on the 16 mg/L ceftolozane/tazobactam agar plates was considered as an indication of resistance development, consistent with established breakpoints.16,17
Results
Ceftolozane/tazobactam agar plates quality control
At each of the 24 h serial passages, the quality control results for the ceftolozane/tazobactam agar plates were acceptable: i.e. there was no growth for the low MIC test isolate and obvious colonies indicating growth when using the high MIC test isolate. The results were consistent for all study days. No contamination of the plates was visualized.
Serial passage
Only one isolate, PSA 2161, showed a small percentage of resistance (1.66%) at 24 h in the meropenem-exposed group. This isolate was from a patient who clinically received meropenem for the longest duration of 19 days. All other isolate-exposure combination showed no growth at 24 h. At 72 h, resistant populations were detected in 6/9 isolates. In five isolates, meropenem exposure significantly increases the prevalence of resistance development against ceftolozane/tazobactam; the percentage of resistant population for the five isolates were 100%, 100%, 53.5%, 31% and 3% for the meropenem-exposed versus 0%, 0%, 2%, 0.35% and ≤0.0003% in the unexposed groups. One isolate had similar percent resistant population at 72 h between groups (≤0.0003 and ≤0.0005%). The remaining isolates showed no development of resistance in either group. Table 1 showed the cfu/mL at 72 h in drug-free plate versus cfu/mL in ceftolozane/tazobactam plates.
Mean cfu/mL of resistant and total population for meropenem versus non-meropenem-exposed isolates
| Isolate ID | C/T MIC (mg/L) | MEM exposure (yes/no) | Mean cfu/mL at 72 h in drug-free plate | Mean cfu/mL at 72 h in 16 mg/L C/T plate |
|---|---|---|---|---|
| PSA 2171 | 2 | Yes | 5.10E + 06 | 1.32E + 07 |
| No | 1.60E + 05 | NG | ||
| PSA 2161 | 0.5 | Yes | 9.06E + 07 | 1.44E + 08 |
| No | 3.10E + 06 | NG | ||
| PSA 2155 | 4 | Yes | 2.08E + 08 | 1.11E + 08 |
| No | 2.46E + 08 | 4.71E + 06 | ||
| PSA 2147 | 4 | Yes | 2.80E + 08 | 8.72E + 07 |
| No | 4.26E + 07 | 1.49E + 05 | ||
| PSA 2149 | 2 | Yes | 5.80E + 06 | 1.78E + 05 |
| No | 3.40E + 08 | ≤1.00E + 03 | ||
| PSA 2151 | 4 | Yes | 2.90E + 08 | ≤1.00E + 03 |
| No | 1.80E + 08 | ≤1.00E + 03 | ||
| PSA 2153 | 1 | Yes | 5.82E + 04 | NG |
| No | 7.25E + 04 | NG | ||
| PSA 2145 | 0.5 | Yes | 9.70E + 04 | NG |
| No | 9.70E + 06 | NG | ||
| PSA 2173 | 1 | Yes | 2.00E + 05 | NG |
| No | 3.12E + 05 | NG |
| Isolate ID | C/T MIC (mg/L) | MEM exposure (yes/no) | Mean cfu/mL at 72 h in drug-free plate | Mean cfu/mL at 72 h in 16 mg/L C/T plate |
|---|---|---|---|---|
| PSA 2171 | 2 | Yes | 5.10E + 06 | 1.32E + 07 |
| No | 1.60E + 05 | NG | ||
| PSA 2161 | 0.5 | Yes | 9.06E + 07 | 1.44E + 08 |
| No | 3.10E + 06 | NG | ||
| PSA 2155 | 4 | Yes | 2.08E + 08 | 1.11E + 08 |
| No | 2.46E + 08 | 4.71E + 06 | ||
| PSA 2147 | 4 | Yes | 2.80E + 08 | 8.72E + 07 |
| No | 4.26E + 07 | 1.49E + 05 | ||
| PSA 2149 | 2 | Yes | 5.80E + 06 | 1.78E + 05 |
| No | 3.40E + 08 | ≤1.00E + 03 | ||
| PSA 2151 | 4 | Yes | 2.90E + 08 | ≤1.00E + 03 |
| No | 1.80E + 08 | ≤1.00E + 03 | ||
| PSA 2153 | 1 | Yes | 5.82E + 04 | NG |
| No | 7.25E + 04 | NG | ||
| PSA 2145 | 0.5 | Yes | 9.70E + 04 | NG |
| No | 9.70E + 06 | NG | ||
| PSA 2173 | 1 | Yes | 2.00E + 05 | NG |
| No | 3.12E + 05 | NG |
C/T, ceftolozane/tazobactam; MEM, meropenem; NG, no growth.
Mean cfu/mL of resistant and total population for meropenem versus non-meropenem-exposed isolates
| Isolate ID | C/T MIC (mg/L) | MEM exposure (yes/no) | Mean cfu/mL at 72 h in drug-free plate | Mean cfu/mL at 72 h in 16 mg/L C/T plate |
|---|---|---|---|---|
| PSA 2171 | 2 | Yes | 5.10E + 06 | 1.32E + 07 |
| No | 1.60E + 05 | NG | ||
| PSA 2161 | 0.5 | Yes | 9.06E + 07 | 1.44E + 08 |
| No | 3.10E + 06 | NG | ||
| PSA 2155 | 4 | Yes | 2.08E + 08 | 1.11E + 08 |
| No | 2.46E + 08 | 4.71E + 06 | ||
| PSA 2147 | 4 | Yes | 2.80E + 08 | 8.72E + 07 |
| No | 4.26E + 07 | 1.49E + 05 | ||
| PSA 2149 | 2 | Yes | 5.80E + 06 | 1.78E + 05 |
| No | 3.40E + 08 | ≤1.00E + 03 | ||
| PSA 2151 | 4 | Yes | 2.90E + 08 | ≤1.00E + 03 |
| No | 1.80E + 08 | ≤1.00E + 03 | ||
| PSA 2153 | 1 | Yes | 5.82E + 04 | NG |
| No | 7.25E + 04 | NG | ||
| PSA 2145 | 0.5 | Yes | 9.70E + 04 | NG |
| No | 9.70E + 06 | NG | ||
| PSA 2173 | 1 | Yes | 2.00E + 05 | NG |
| No | 3.12E + 05 | NG |
| Isolate ID | C/T MIC (mg/L) | MEM exposure (yes/no) | Mean cfu/mL at 72 h in drug-free plate | Mean cfu/mL at 72 h in 16 mg/L C/T plate |
|---|---|---|---|---|
| PSA 2171 | 2 | Yes | 5.10E + 06 | 1.32E + 07 |
| No | 1.60E + 05 | NG | ||
| PSA 2161 | 0.5 | Yes | 9.06E + 07 | 1.44E + 08 |
| No | 3.10E + 06 | NG | ||
| PSA 2155 | 4 | Yes | 2.08E + 08 | 1.11E + 08 |
| No | 2.46E + 08 | 4.71E + 06 | ||
| PSA 2147 | 4 | Yes | 2.80E + 08 | 8.72E + 07 |
| No | 4.26E + 07 | 1.49E + 05 | ||
| PSA 2149 | 2 | Yes | 5.80E + 06 | 1.78E + 05 |
| No | 3.40E + 08 | ≤1.00E + 03 | ||
| PSA 2151 | 4 | Yes | 2.90E + 08 | ≤1.00E + 03 |
| No | 1.80E + 08 | ≤1.00E + 03 | ||
| PSA 2153 | 1 | Yes | 5.82E + 04 | NG |
| No | 7.25E + 04 | NG | ||
| PSA 2145 | 0.5 | Yes | 9.70E + 04 | NG |
| No | 9.70E + 06 | NG | ||
| PSA 2173 | 1 | Yes | 2.00E + 05 | NG |
| No | 3.12E + 05 | NG |
C/T, ceftolozane/tazobactam; MEM, meropenem; NG, no growth.
Discussion
In the current study, we assessed the impact of meropenem pre-treatment on ceftolozane/tazobactam-resistance development in nine DTR-P. aeruginosa isolates. All isolates were obtained from patients with pneumonia and/or bacteraemia and developed resistance to ceftolozane/tazobactam during therapy in the clinic.10 Our results suggest that meropenem pre-exposure increases the prevalence of in vitro resistance development against ceftolozane/tazobactam.
Eight of the nine isolates assessed in the current study were obtained from patients received meropenem treatment (range 2–19 days), before initiation of ceftolozane/tazobactam and then subsequently developed resistance to ceftolozane/tazobactam during therapy.13 Isolate PSA 2173 was from the only patient who did not receive meropenem. This isolate showed no resistance development for both the meropenem-exposed and non-exposed groups. Isolate PSA 2161 obtained from the patient who received 19 days of meropenem showed resistance after the first passage of ceftolozane/tazobactam. Similarly, at 72 h, this same isolate showed a high percentage of resistance in the meropenem-exposed group. Two isolates from patients who received 8 and 9 days of meropenem (PSA 2145, PSA 2153) did not develop resistance after ceftolozane/tazobactam serial passages for both the meropenem-exposed and non-exposed groups. Clinically, ceftolozane/tazobactam was not initiated until meropenem-resistance was known by conventional antimicrobial susceptibility testing per hospital policy.10 To simulate this clinical situation in the current study, we performed serial passage of ceftolozane/tazobactam to assess in vitro development of resistance for these isolates over 72 h. Our in vitro assessment found no growth on the ceftolozane/tazobactam containing agar plates after the first and second serial passages in all isolates except for one. These findings suggest that resistance developed after the ceftolozane/tazobactam exposure and not that isolates were resistant at baseline. The difference in the development of resistance in replicates of some isolate pre-treated with meropenem compared with the same isolates without pre-treatment may suggest that meropenem exposure pre-disposes isolates to the development of resistance to ceftolozane/tazobactam during treatment, as seen in clinical cases described elsewhere.13 A plausible influence of our finding regarding resistance development may be to consider earlier antimicrobial susceptibility testing, i.e. after 3 days if the patient has not improved and had a history of meropenem treatment, to detect changes in the resistance profile and ensure timely initiation of the proper antimicrobial agent.
DTR-P. aeruginosa is defined as being non-susceptible to piperacillin-tazobactam, ceftazidime, cefepime, aztreonam, meropenem, imipenem/cilastatin, ciprofloxacin and levofloxacin.18 Ceftolozane/tazobactam represents a valuable therapeutic option for challenging P. aeruginosa by maintaining high activity against DTR-P. aeruginosa isolates.19 However, soon after clinical introduction, resistance to ceftolozane/tazobactam has been detected during therapy.6–10 In general, resistance mechanisms of P. aeruginosa to ceftolozane/tazobactam are categorized in two main groups with the first being expression of exogenous extended-spectrum β-lactamases,20 including carbapenemases.13 The second group is characterized by the enhancement of intrinsic resistance mechanisms including the production of PDCs with improved β-lactamase hydrolytic activity against ceftolozane/tazobactam.21 We found that ceftolozane/tazobactam-resistance development was enhanced by meropenem exposure. There are two possible explanations for the development of resistance after exposure to meropenem. The first hypothesis is that meropenem may trigger a stress response, leading to an increase in resistance mechanisms that have overlap with other agents including ceftolozane/tazobactam.22 The second theory suggests that molecular-level mutations may occur, resulting in cross-resistance.21 The clinical isolates used in the current study and the corresponding subsequent isolates that developed resistance against ceftolozane/tazobactam were sequenced by Tamma et al. to investigate the mutations that may be responsible for the emergence of ceftolozane/tazobactam resistance.10 The authors found that resistance might be attributable to AmpC gene; including (G183D, E247K) and AmpR gene (D135G). Mutations in DNA polymerase subunits gamma and tau (dnaX gene) and PBP3 mutations were also detected between the ceftolozane/tazobactam susceptible isolates and the emergent resistant strains (Table S1, available as Supplementary data at JAC Online).10 These findings highlight a clinical challenge as delaying initiation of ceftolozane/tazobactam and continuing agents such as meropenem may adversely affect the future efficacy of ceftolozane/tazobactam.
Pre-exposure to one antibiotic agent contributing to resistance of another has been described for other bug-drug combinations. Indeed, in vitro exposure to fluoroquinolones has been associated with increased selection of meropenem-resistant isolates on meropenem exposure.23 The clinical implications of this in vitro phenomena have also been described. Fluoroquinolone prophylaxis for patients with haematologic malignancy was an independent predictor of infection with meropenem-non-susceptible P. aeruginosa necessitating clinicians to consider antibiotic treatment history when selecting empiric therapy.24 Exposure of one antimicrobial pre-disposing to resistance to another antimicrobial has also been described for other β-lactams. In vitro, imipenem exposed P. aeruginosa led to elevated MICs for piperacillin/tazobactam, indicating elevated piperacillin/tazobactam MICs were a result of the imipenem pre-exposure.25 Concerning meropenem’s impact on ceftolozane/tazobactam-resistance development, our findings provide foundational in vitro data of this phenomena. The implications of these findings on optimal treatment strategies warrant further clinical assessment.
In conclusion, the current study suggests that previous meropenem exposure may pre-dispose to ceftolozane/tazobactam-resistance development that may limit the therapeutic utility of ceftolozane/tazobactam. Resistance may be a result of stress response or molecular-level mutations. Further in vivo analysis using clinical exposures and clinical data are needed to ascertain the clinical implications of these findings, which would be crucial to optimize treatment of DTR-P. aeruginosa infections.
Acknowledgements
We would like to recognize the staff at the Center for Anti-infective Research & Development for their vital assistance in this study.
Funding
The study was internally funded by the Center for Anti-Infective Research and Development.
Transparency declarations
C.M.G. has received research funding from Cepheid, Shionogi, Everest Medicines and Entasis. D.P.N. served as a consultant, speaker’s bureau member or has received research funding from: Allergan, Cepheid, Merck, Pfizer, Wockhardt, Shionogi, Tetraphase and Venatorx. P.J.S. reports grants and personal fees from OpGen Inc, bioMérieux, Inc., Qiagen and BD Diagnostics, grants from Affinity Biosensors and T2 Biosystems; and personal fees from Shionogi, Inc., Entasis, Merck and GeneCapture, outside the submitted work. A.F., S.E.N. and P.D.T. report no disclosures.
Supplementary data
Table S1 is available as Supplementary data at JAC Online.
