Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, September 2005, p. 3789-3793, Vol. 49, No. 9
0066-4804/05/$08.00+0 doi:10.1128/AAC.49.9.3789-3793.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Service of Infectious Diseases, University Hospitals of Geneva, CH-1211 Geneva 14, Switzerland
Received 16 February 2005/ Returned for modification 2 April 2005/ Accepted 18 June 2005
| ABSTRACT |
|---|
|
|
|---|
75% of each dosing interval. After 7 days of therapy with ceftobiprole medocaril or vancomycin, average counts of MRGR3 decreased significantly (P < 0.02) by 0.68 ± 0.28 (n = 29) and 0.88 ± 0.22 (n = 28) log CFU/ml of tissue cage fluid, respectively, compared with cages of untreated animals, but were not significantly different from each other. No resistant mutants were detected on ceftobiprole-supplemented agar following therapy with this cephalosporin. The in vivo activity of ceftobiprole medocaril against chronic MRSA foreign-body infections was equivalent to that of vancomycin and did not lead to the emergence of resistant subpopulations. | INTRODUCTION |
|---|
|
|
|---|
Recent development of ß-lactams targeting penicillin-binding protein 2' (PBP2'), which is chiefly responsible for methicillin resistance in S. aureus, represents an important advance in the field of antimicrobial chemotherapy. Ceftobiprole (formerly known as BAL9141) is a pyrrolidinone cephalosporin with high affinities for penicillin-binding proteins including PBP2'. Additional assets of ceftobiprole are its refractoriness to hydrolysis by class A and class C ß-lactamases (22), and recalcitrance towards development of endogenous resistance (9, 18, 23, 27; T. Bogdanovich, B. Bozdogan, and P. C. Appelbaum, Abstr. Intersci. Conf. Antimicrob. Agents Chemother., abstr. E-2018, 2004). Ceftobiprole exhibits excellent in vitro activity towards a wide array of gram-negative (22, 25, 26) and gram-positive (15, 18, 22, 26) pathogens, including MRSA and its vancomycin-intermediate or vancomycin-resistant derivatives (5, 9, 14, 15, 18, 22, 26). Both animal (3, 9, 18, 22) and human studies (37, 38; M. Heep, S. Querner, M. Harsch, and W. O'Riordan, Abstr. 44th Intersci. Conf. Antimicrob. Agents Chemother., abstr. L-361, 2004) point towards ceftobiprole as a promising candidate for treatment of MRSA infections.
We previously showed the usefulness of a rat tissue cage model of S. aureus chronic foreign-body infections for evaluating various categories of antimicrobial agents such as vancomycin (32), teicoplanin (36), imipenem (35), daptomycin (43), and several fluoroquinolones including fleroxacin (32), sparfloxacin, temafloxacin, and ciprofloxacin (7) and levofloxacin and trovafloxacin (44).
The present study aimed at evaluating the therapeutic potential of ceftobiprole and its propensity for emergence of resistance compared to the reference anti-MRSA agent vancomycin, in the rat tissue cage model of MRSA chronic foreign-body infection.
(This study was presented in part at the 44th Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, D.C., October 2004.)
| MATERIALS AND METHODS |
|---|
|
|
|---|
Antimicrobial agents. Ceftobiprole (weight purity, 90%) and its water-soluble prodrug ceftobiprole medocaril (formerly known as BAL5788; weight purity 89.2%) were supplied by Basilea Pharmaceutica AG (Basel, Switzerland).
For in vitro studies, a 1 mg/ml stock solution of ceftobiprole was prepared at room temperature by dissolving 2.2 mg of ceftobiprole in 198 µl of AR grade dimethyl sulfoxide admixed with 20 µl of AR grade glacial acetic acid; after vigorous vortexing the ceftobiprole solution was diluted with 1,782 µl of glass-distilled water.
For in vivo studies, fresh stock solutions of prodrug ceftobiprole medocaril were prepared within 30 min of use by dissolving either 300 mg (low-dose) or 900 mg (high-dose) of the prodrug in 10 ml of citrate buffer, pH 5.0. Solubilization of ceftobiprole medocaril was facilitated by mild sonication for a few minutes in a water bath. For low-dose regimens the prodrug solution contained 30 mg/ml of ceftobiprole medocaril, targeted to release 50 mg ceftobiprole/kg body weight; for high-dose regimens the prodrug solution contained 90 mg/ml of the ceftobiprole medocaril, targeted to release 150 mg ceftobiprole/kg body weight.
Vancomycin for intravenous infusion (Lilly, Giessen, Germany) was dissolved, stored at 4°C, and used within 48 h, as recommended by the manufacturer.
MICs and MBCs. The MICs and MBCs of ceftobiprole for MRSA strain MRGR3 or MSSA strain 25923 were determined by broth macrodilution according to National Committee for Clinical Laboratory Standards guidelines (33). Concentrations of ceftobiprole in cation-adjusted Mueller-Hinton broth (CAMHB, Difco, Detroit, Mich.) ranged from 8 down to 0.125 µg/ml and the bacterial inoculum (ca. 106 CFU/ml) was obtained from 100-fold dilutions of 3-h log-phase cultures of either strain MRGR3 or control strain ATCC 25923 grown in CAMHB.
To minimize possible carryover effects of ceftobiprole during MBC determinations, 100-µl portions from all tubes with no visible growth were subcultured, either undiluted or serially diluted 10-fold in saline, on Mueller-Hinton agar (MHA) for 36 to 48 h at 37°C. The MBC was defined as the lowest concentration that killed 99.9% of the original inoculum.
Killing kinetic studies. Sterile plastic tubes containing 1 ml of CAMHB supplemented with either 4 or 8 µg/ml of ceftobiprole were incubated at 37°C with ca. 106 CFU of MRSA strain MRGR3 or MSSA strain 25923 (obtained from 3-h log-phase cultures) either in static conditions or (when indicated) a shaking water bath. The number of viable organisms was determined by subculturing 100 µl of 10-fold serially diluted portions of broth on MHA after 0, 2, 4, 6, and 24 h of incubation. Colonies were counted using an automated colony counter (Countermat flash, IUL Instruments) after 48 h of incubation at 37°C. The detection limit was 2 log10 CFU/ml.
To evaluate the impact of tissue cage fluid proteins on the bactericidal activities of either 4 or 8 µg/ml of ceftobiprole, the elimination rate of strain MRGR3 was also recorded in tubes containing 1 ml of a mixture of CAMHB and sterile tissue cage fluid (pooled from 10 different cages of uninfected animals) in a 1:1 ratio.
Single-step resistance mutation frequency. A freshly prepared stock solution of ceftobiprole (1 mg/m) was admixed with melting MHA at 45°C. Aliquots (100 µl) of serial dilutions of washed overnight cultures of MRSA strain MRGR3 (108 to 105 CFU) were plated in duplicate onto antibiotic-free MHA and MHA containing 4 µg/ml or 8 µg/ml ceftobiprole. After 48 h at 37°C plates were inspected for staphylococcal colonies.
Pharmacokinetic properties of ceftobiprole in rat tissue cage fluids. A preliminary pharmacokinetic evaluation was performed to determine a dosing regimen for prodrug ceftobiprole medocaril yielding ceftobiprole levels in tissue cage fluids of at least 4 µg/ml (corresponding to the MBC of ceftobiprole for MRSA strain MRGR3) for at least 60% of the dosing interval. A low-dose regimen, delivering 50 mg-eq. of ceftobiprole per kg of body weight, and a high-dose regimen, delivering 150 mg-eq. of ceftobiprole per kg of body weight, were administered intraperitoneally every 12 hours over 1 week to six uninfected rats. Ceftobiprole levels in tissue cage fluids for both regimens were determined at 0, 2, 4, 6, and 12 h on day 4 (for a mid-therapy assessment of tissue cage fluid levels) and on day 7 (to assess possible accumulation of ceftobiprole in tissue cage fluids).
Aliquots (ca. 100 µl) of tissue cage fluid were dispensed into prechilled Eppendorf tubes coated with Na2EDTA and citric acid immediately after collection, mixed with the anticoagulants by gentle inversion (6 to 10 times), and stored on crushed ice while avoiding exposure to sunlight. Samples were clarified by centrifugation at ca. 1,500 x g for 15 min at 4°C and the supernatants immediately decanted and stored at 80°C pending quantification of ceftobiprole (3).
Antibiotic treatment of MRSA chronic tissue cage infections.
The experimental design was approved by the Ethics Committee of the Faculty of Medicine, University of Geneva, and by the Veterinary Office of the State of Geneva. Four tissue cages per animal were implanted subcutaneously (32) in Wistar rats that had been anesthetized by intraperitoneal injection with a mixture of ketamine (90 mg/kg) and xylazine (5 mg/kg). At 3 weeks postimplantation, tissue cage fluid was aspirated and checked for sterility. To establish a chronic local infection by MRSA, tissue cages were inoculated with 0.1 ml of saline containing 0.5 x 106 to 2.0 x 106 CFU of a log-phase washed culture of strain MRGR3. Two weeks later, MRSA titers in fluids from all tissue cages in all rats were quantified, and tissue cages containing
105 CFU/ml of fluid were included in the therapeutic protocols.
Rats were randomized to receive (intraperitoneal, 7 days) twice-daily doses of ceftobiprole medocaril (150 mg-eq. of ceftobiprole/kg), vancomycin (50 mg/kg), or no antibiotic treatment (control group). The outcome of antibiotic therapy on MRSA tissue cage infections was evaluated by quantitative cultures of 10-fold serially diluted tissue cage fluids, performed on MHA at day 1 (before treatment) and day 8 (12 h after the last injection of either ceftobiprole medocaril or vancomycin) of therapy. To optimize the yield of viable bacteria, tissue cage fluids were sonicated briefly (60 W, 1 min) to disrupt the biofilm and phagocytic cells before the serial dilutions and plating. The detection limit of bacterial counts was 2 log10 CFU/ml of tissue cage fluid. The differences in CFU counts between days 1 and 8 were determined and expressed as delta log10 CFU/ml. For each treatment group, results were expressed as means ± standard error of the mean. Comparison of bacterial counts in the different groups was performed by the nonparametric Kruskall-Wallis test and the Dunn procedure for comparison of specific groups (34). Data were considered significant when P was <0.05 using two-tailed significance levels.
Screening for in vivo emergence of resistance to ceftobiprole. Bacteria recovered from tissue cage fluids on day 8 of therapy were screened for emergence of resistance to ceftobiprole by plating 100-µl samples of 10-fold-diluted cage fluids onto MHA containing 4 µg/ml of ceftobiprole, equivalent to fourfold its MIC and onefold its MBC for strain MRGR3. Plates were incubated for 48 h at 37°C. The detection limit was 2 log10 CFU/ml of tissue cage fluid.
| RESULTS |
|---|
|
|
|---|
Time-kill studies showed equivalent reductions in the viable counts of strain MRGR3 by 4 and 8 µg/ml of either ceftobiprole or vancomycin at all time points, averaging 1.5 and >3 log10 CFU/ml at 6 and 24 h, respectively (data not shown). Additional controls comparing MRSA strain MRGR3 with MSSA strain 25923 (22) showed similar elimination rates of 4 and 8 µg/ml of ceftobiprole, and no difference was observed for cultures incubated in static or shaken conditions (data not shown). Reductions in viable counts of strain MRGR3 by 4 and 8 µg/ml of ceftobiprole in CAMHB supplemented with 50% tissue cage fluid were similar to those obtained in CAMHB alone (data not shown), indicating that the in vitro elimination rate of strain MRGR3 by ceftobiprole is unaffected by tissue cage fluid components.
Plating 105 to 108 CFU of MRSA strain MRGR3 onto MHA containing 4 or 8 µg/ml ceftobiprole completely suppressed bacterial growth during 48 h. Under these conditions, therefore, the single-step mutation frequency of strain MRGR3 for ceftobiprole is <108, even at a concentration as low as the MBC (4 µg/ml) for the test strain.
Pharmacokinetics of ceftobiprole in rat tissue cages.
To determine the adequate dosing regimen of ceftobiprole medocaril yielding local ceftobiprole concentrations
4 µg/ml (which is the MBC of ceftobiprole for MRSA strain MRGR3) in tissue cage fluids for at least 60% of the dosing interval (for twice a day, 7 h), two different regimens of the prodrug (designed to deliver 50 or 150 mg-eq./kg of ceftobiprole) were compared, referred to as low-dose and high-dose regimens of ceftobiprole medocaril, respectively.
The mean peak and trough levels of ceftobiprole in tissue cage fluids (n = 6) of animals treated every 12 hours with low-dose ceftobiprole/kg were 4.11 and 0.79 µg/ml, respectively, on day 4 and 5.81 and 0.47 µg/ml, respectively, on day 7 (Fig. 1). These results indicated that the low-dose regimen was inadequate for therapy, since tissue cage fluid levels of ceftobiprole exceeded the MBC (4 µg/ml) for <2 h (<17%) of the dosing interval. In contrast, mean peak and trough levels of ceftobiprole in tissue cage fluids (n = 6) of animals treated every 12 hours with 150 mg-eq. of ceftobiprole/kg were 16.14 and 2.40 µg/ml, respectively, on day 4 and 22.60 and 1.67 µg/ml, respectively, on day 7 (Fig. 1). Analysis of the clearance data for ceftobiprole in tissue cage fluids for the high-dose regimen indicated bactericidal levels (
4 µg/ml) towards MRSA strain MRGR3 for at least 9 h (75%) of each dosing interval. Thus, the high-dose regimen of ceftobiprole medocaril was chosen for treatment studies of tissue cages infected with MRSA strain MRGR3.
|
Treatment of chronic tissue cage infections. At the onset of therapy, average bacterial counts for cages infected with strain MRGR3 were 6.88 ± 0.18 log10 CFU/ml for controls (n = 30), 6.56 ± 0.18 log10 CFU/ml for animals receiving ceftobiprole medocaril (n = 29), and 7.07 ± 0.18 log10 CFU/ml for animals receiving vancomycin (n = 28). These three sets of bacterial counts did not differ significantly from one other (P = 0.15).
At the end of the 7-day treatment period, bacterial counts in tissue cages of control animals showed a slight albeit nonsignificant increase of 0.18 ± 0.20 log10 CFU/ml (n = 30). Compared to the control group, antibiotic treatment led to significant (P < 0.02) reductions in bacterial counts in tissue cage fluids of 0.68 ± 0.28 log10 CFU/ml (n = 29) for ceftobiprole and 0.88 ± 0.22 log10 CFU/ml (n = 28) for vancomycin (Fig. 2). The slightly greater CFU reduction for vancomycin-treated rats was not significantly different (P = 0.59) from that for ceftobiprole-treated animals.
|
| DISCUSSION |
|---|
|
|
|---|
2 µg/ml) and MIC90 (
4 µg/ml) values for ceftobiprole with diverse panels of MRSA clinical isolates (2, 14, 15, 18, 22, 26; L. M. Ednie and P. C. Appelbaum, Abstr. 44th Intersci. Conf. Antimicrob. Agents Chemother., abstr. E-2021, 2004; R. Reynolds, D. M. Livermore, and the BSAC Working Party on Bacteraemia Resistance Surveillance, 44th Intersci. Conf. Antimicrob. Agents Chemother., abstr. E-2035 and E-2036, 2004). Over the past decade, studies using the tissue cage model have confirmed the limited efficacy of antibiotic regimens for treatment of S. aureus foreign-body infections (7, 12, 13, 32, 35, 36, 43, 44). These studies indicated that bactericidal levels of anti-staphylococcal agents needed to be reached in tissue cage fluids to reduce viable counts of S. aureus in chronically infected tissue cages. A pharmacokinetic study of two dosing regimens of ceftobiprole medocaril identified 150 mg-eq. mg/kg of ceftobiprole/kg every 12 hours as promoting bactericidal levels of ceftobiprole in tissue cage fluids for at least 75% of the dosing interval.
At the selected doses ceftobiprole and vancomycin each achieved comparable peak and trough levels in tissue cage fluids. The two drugs also have comparable protein binding values of 38% and 30%, respectively (Basilea Pharmaceutica AG, data on file). We observed no significant difference in CFU reductions between ceftobiprole-treated and vancomycin-treated animals, though there was a significant reduction in viable counts of MRSA strain MRGR3 for antibiotic-treated relative to control animals. The modest reductions in viable counts observed after 7 days of treatment with high doses of ceftobiprole or vancomycin attests to the intractability of this infection model (7, 12, 32, 35, 36, 43, 44).
While comparison between experimental therapy in animals and clinical use in humans should be done with caution, the following arguments indicate that the serum/plasma levels required to reach bactericidal levels of ceftobiprole in tissue cage fluids were not exceeding those reasonably attainable with human dosing. First, previous experimental studies have shown the efficacy of ceftobiprole to be directly related to the percentage of the dosing interval during which the concentration of ceftobiprole in plasma was above the minimum MIC of the target pathogen (T>MIC) (D. R. Andes and W. Craig, Abstr. 40th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 1079, 2000). For MRSA infection, a conservative target of 4 µg/ml was retained. Second, the dosing regimen in the therapeutic phase 2 study in complicated skin and skin structure infection was 750 mg twice a day given as a 30-min infusion. At this dose the plasma levels of ceftobiprole in patients remained above the target value for 8 h (A. H. SchmittHoffmann, M. Harsch, M. Heep, M. Schleimer, T. Brown, A. Man, and W. O'Riordan, Abstr. 14th ECCMID, abstract P-1031, 2004). Third, ceftobiprole in animals and humans is restricted to the extracellular compartment, with no species differences in the extent of plasma protein binding. When ceftobiprole was administered to rats at a dose of 150 mg/kg twice a day, the predicted plasma concentrations of the cephalosporin were well under 4 µg/ml at 6 h after administration, corresponding to a T>MIC of less than 50%. Since the dose given to rats was somewhat inferior to the human therapeutic dose in terms of T>MIC, the efficacy of this suboptimal dose in this stringent model should be considered predictive for the efficacy in human at therapeutic doses.
Of six different fluoroquinolones tested using regimens producing bactericidal levels in tissue cage fluids (7, 12, 32, 44), only temafloxacin (withdrawn by Abbott laboratories in 1002, four months after Food and Drug Administration approval, for reasons of toxicity) and levofloxacin led to significantly higher reductions in viable counts than vancomycin (7, 44), whereas high-dose regimens of ciprofloxacin or teicoplanin failed to reduce significantly viable counts of MRSA strain MRGR3 (7, 36). Not surprisingly, the in vivo activity of ceftobiprole was much better than that of imipenem (35), a carbapenem once considered as a possible treatment for MRSA infections (10).
An important outcome of this work was the demonstration of a complete lack of emergence of MRSA strain MRGR3 derivates with diminished susceptibility to ceftobiprole, following 7 days of intensive therapy with this cephalosporin. This result is remarkable because the selective medium used for screening resistant subclones contained 4 µg/ml of ceftobiprole, which is just the MBC of this cephalosporin for MRSA strain MRGR3. Such a low concentration of ceftobiprole should facilitate detection of colonies exhibiting marginally decreased susceptibility to this antimicrobial agent. Historical comparison with previous studies in our tissue cage model indicates that emergence of resistance or phenotypic tolerance occurred with a number of antimicrobial agents. For example, microcolonial subpopulations of MRGR3 highly resistant to imipenem were detected on posttherapy isolates on imipenem-supplemented agar plates after 72 h of growth at 37°C (35). In other studies from our group, the local environment of chronic MRSA tissue cage infections promoted the in vivo emergence of phenotypic tolerance (41) or resistance (36, 42) to glycopeptides, especially teicoplanin, in strain MRGR3. Finally, the absence of resistance development to ceftobiprole in MRSA strain MRGR3 also contrasts with the emergence of resistant subpopulations during daptomycin therapy of S. aureus-infected tissue cages (43).
Collectively, these data strongly suggest that the tissue cage model is adequate for detecting emergence of ceftobiprole-resistant subpopulations of S. aureus, in particular because cage fluid is a protein-rich (ca. 20 g/liter) exudate enriched in nutrients presumably released from lysed host cells (e.g., red blood cells, leukocytes, others), which supports bacterial growth much better than serum (49). Our results also confirm previous reports that staphylococci (9, 18, 22, 23; T. Bogdanovich, B. Bozdogan, and P. C. Appelbaum, Abstr. Intersci. Conf. Antimicrob. Agents Chemother., abstr. E-2018, 2004) and other pathogens (27) are refractory to development of endogenous resistance to ceftobiprole. This likely represents a significant advantage of ceftobiprole over glycopeptide or lipopeptide therapy of foreign-body infections.
The favorable pharmacodynamic and therapeutic properties of ceftobiprole for the treatment of severe infections due to MRSA or other multiresistant gram-positive bacteria warrant further studies of this cephalosporin in clinical settings.
| ACKNOWLEDGMENTS |
|---|
We thank A. Schmitt-Hoffman and S. Shapiro for constructive criticisms of the manuscript and J. Spikerman and colleagues for quantification of ceftobiprole in tissue cage fluids.
| FOOTNOTES |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Clin. Vaccine Immunol. | Clin. Microbiol. Rev. |
|---|---|
| J. Clin. Microbiol. | ALL ASM JOURNALS |