Ceftaroline fosamil in the treatment of experimental meningitis caused by methicillin-resistant Staphylococcus aureus


 Background Meningitis caused by methicillin resistant Staphylococcus aureus (MRSA) is rare and often fatal. The recommended treatment for MRSA meningitis, vancomycin, is limited by poor cerebrospinal fluid (CSF) penetration. Ceftaroline fosamil is a novel fifth-generation cephalosporin with potent antibacterial activity in vitro against MRSA. Several case reports in humans suggest that ceftaroline might be a potential treatment option for MRSA meningitis. Our primary objective was to com­pare the efficacy of ceftaroline and vancomycin for the treatment of MRSA meningitis using a rabbit meningitis model. We then characterized CSF drug concentrations over time.Methods Ninety rabbits received a direct intracisternal injection of 5 x 105 CFU S. aureus (MRSA 252). Following a 16-hour incubation period, approximately 0.5ml of CSF was withdrawn via intracisternal aspiration immediately prior to treatment for quantitative bacterial counts and CSF antibiotic concentrations. Rabbits then received (by 1:1:1 random assignment) either no treatment (control group), vancomycin 20 mg/kg at 0 and 12hrs, or ceftaroline 40mg/kg at 0 and 4 hrs via marginal ear vein. CSF collection was repeated every 12 hours for up to 40 hours. All animals were humanely euthanized at 40 hours post inoculation. The primary endpoint (difference in bacterial load [expressed as CFU/mL] for each animal between the initial (T1) and terminal (T3) taps were characterized and compared between groups using the Kruskal-Wallis test. CSF drug concentrations were determined using liquid chromatography with tandem mass spectrometry (LC/MS/MS) assay.ResultsAmong the forty-three evaluable animals with established MRSA meningitis, there was no statistical difference in median CFU observed between the groups in pretreatment CFU counts (p=0.16). Reductions in bacterial load from baseline were 88%, 79% and 75% in the control, ceftaroline and vancomycin-treated groups, respectively. There was no statistical difference observed between vancomycin and either control or ceftaroline at any time point. While animals treated with ceftaroline exhibited a significant increase in clearance rate (log ∆) between T1 and T2 when compared to control (p = 0.019), these differences were no longer statistically significant by T3 (p= 0.43) as CSF ceftaroline concentrations were undetectable by that time point.Conclusions While ceftaroline may be a reasonable alternative to vancomycin for MRSA meningitis, additional animal and clinical studies are required to determine optimal dosing to establish its effectiveness.


Abstract Background
Meningitis caused by methicillin resistant Staphylococcus aureus (MRSA) is rare and often fatal. The recommended treatment for MRSA meningitis, vancomycin, is limited by poor cerebrospinal uid (CSF) penetration. Ceftaroline fosamil is a novel fth-generation cephalosporin with potent antibacterial activity in vitro against MRSA. Several case reports in humans suggest that ceftaroline might be a potential treatment option for MRSA meningitis. Our primary objective was to com pare the e cacy of ceftaroline and vancomycin for the treatment of MRSA meningitis using a rabbit meningitis model. We then characterized CSF drug concentrations over time.

Methods
Ninety rabbits received a direct intracisternal injection of 5 x 10 5 CFU S. aureus (MRSA 252). Following a 16-hour incubation period, approximately 0.5ml of CSF was withdrawn via intracisternal aspiration immediately prior to treatment for quantitative bacterial counts and CSF antibiotic concentrations. Rabbits then received (by 1:1:1 random assignment) either no treatment (control group), vancomycin 20 mg/kg at 0 and 12hrs, or ceftaroline 40mg/kg at 0 and 4 hrs via marginal ear vein. CSF collection was repeated every 12 hours for up to 40 hours. All animals were humanely euthanized at 40 hours post inoculation. The primary endpoint (difference in bacterial load [expressed as CFU/mL] for each animal between the initial (T1) and terminal (T3) taps were characterized and compared between groups using the Kruskal-Wallis test. CSF drug concentrations were determined using liquid chromatography with tandem mass spectrometry (LC/MS/MS) assay.

Results
Among the forty-three evaluable animals with established MRSA meningitis, there was no statistical difference in median CFU observed between the groups in pretreatment CFU counts (p=0.16). Reductions in bacterial load from baseline were 88%, 79% and 75% in the control, ceftaroline and vancomycin-treated groups, respectively. There was no statistical difference observed between vancomycin and either control or ceftaroline at any time point. While animals treated with ceftaroline exhibited a signi cant increase in clearance rate (log ∆) between T1 and T2 when compared to control (p = 0.019), these differences were no longer statistically signi cant by T3 (p= 0.43) as CSF ceftaroline concentrations were undetectable by that time point.

Conclusions
While ceftaroline may be a reasonable alternative to vancomycin for MRSA meningitis, additional animal and clinical studies are required to determine optimal dosing to establish its effectiveness.

Page 3/11
Methicillin resistant Staphylococcus aureus (MRSA) is a common and often devastating cause of nosocomial bacterial meningitis. Currently, the standard treatment for MRSA meningitis is vancomycin, but its penetration into the cerebral spinal uid (CSF) is inconsistent and (at times) poor. Adjunctive treatments, such as intrathecal administration of vancomycin, are of unproven ine cacy and can confer signi cant risks to the patient. As a result, alternative treatments for MRSA meningitis are greatly needed.
Cephalosporins have a rich history in the optimal management of bacterial meningitis in humans.
Ceftaroline fosamil (Te aro®-Allergan, Madison, NJ, USA) is a fth-generation cephalosporin approved for treatment of community-acquired bacterial pneumonia and acute skin and soft tissue infections.
Recent animal studies suggest that ceftaroline can penetrate the blood-brain barrier in rabbits, and therefore might play a role in the treatment of MRSA meningitis in humans. While a few cases have reported some promise with ceftaroline for the treatment of S. aureus and Streptococcus pneumoniae meningitis, the e cacy of ceftaroline against MRSA meningitis and its e cacy relative to vancomycin has yet to be comparatively tested experimentally. Our purpose was to evaluate the e cacy of ceftaroline in the treatment of MRSA meningitis using a rabbit meningitis model. Our primary objective was to compare ceftaroline-and vancomycin-treated animals for the difference in bacterial load (expressed as CFU/mL) between the initial and terminal taps. Our secondary objective was to characterize CSF drug concentrations over time.

Ethics
All animal procedures were reviewed and approved by the Duke Institutional Animal Care and Use Committee prior to institution of any study-related procedures (Duke IACUC #A006-15-01).

Test organism
An isolated colony of S. aureus MRSA 252, a well characterized and widely used MRSA isolate, was incubated overnight in tryptic soy broth at 37 o C/220 RPM. The following day, a subculture was permitted to grow until the bacteria reached the log-phase of growth (approximately 2 hrs.). Cells were then washed twice in phosphate buffered saline (PBS) and re-suspended in PBS containing 20% glycerol at a concentration of ~ 10 6 cfu/µL. The suspension was divided into 1 mL glycerol stock aliquots, and stored at -80 o C. Bacterial inoculations were derived from frozen glycerol stock, and re-suspended at 5 × 10 5 cfu in 300 µL sterile PBS.

Animal model
The overall experimental design is illustrated in Fig. 1. Ninety 6-8-week-old male New Zealand white rabbits (~ 2.5 kg) were acquired from Robinson Services Inc. (Mocksville, NC, USA). Following a week of habituation, animals were anesthetized with an intramuscular injection of ketamine (38 mg/kg of body weight) and xylazine (5 mg/kg of body weight) prior to each intracisternal procedure. Animals were assigned to treatment groups 1:1:1 using a random number generator in Microsoft Excel (Microsoft Corp., Redmond, WA) at the beginning of the experiment. Each animal received an intracisternal inoculation of MRSA 252 at 0 hours, following the method of Perfect and Durack . At 16 hours post-inoculation (HPI), we collected the rst CSF sample followed immediately by antibiotic administration (Timepoint 1 [T1]).
We chose an inoculation dose of 5 × 10 5 cfu in order to attain su cient bacterial load without resulting in unacceptable mortality, which meant that most animals (regardless of treatment) cleared infection by the Agilent 1200 series LC system interfaced with Applied Biosystems/SCIEX API 5500 QTrap hybrid triple quadrupole-linear trap MS/MS spectrometer equipped with electrospray ionization (ESI) source was used for analysis. Analyst (version 1.6.2) software was used for mass tuning, data acquisition, and quanti cation. The assay compounds were individually infused as 100 nM solutions in 50%A/50%B at 10 µL/min ow rate and internal ion path parameters optimized to provide maximal ion count for "parent" and collision-produced ("daughter") MS/MS ions. Samples were prepared by adding pure standard of the measured compound to arti cial CSF. Lower limit of quanti cation (LLOQ) at 80% accuracy limit for both vancomycin and ceftaroline was 0.81 ng/mL from 20 µL CSF sample. The response of the peak area standard/internal standard. to nominal concentration was linear with r 2 = 0.998 or better.

Data analyses
Only individuals with CSF samples meeting the minimum criteria of 10 3 CFU/mL at T1 who successfully underwent CSF collection at T3 were included in the nal analyses. For the primary endpoint, the difference in bacterial load for each animal between T1 and T3 were characterized and compared between groups. Absolute bacterial loads as well as proportional changes between T1 and T2 and between T2 and T3 were examined. Descriptive statistics were performed for both bacterial loads, changes in these loads and drug concentrations. Differences in these measures by treatment group and speci c time points were assessed for statistical differences using the Kruskal-Wallis test. These tests were performed for all three treatment groups together and for pairwise comparisons of the treatment groups. Drug concentrations were characterized at two time points (T2 and T3) using descriptive statistics.

Results
Of the ninety animals used in this experiment, forty-three met our minimum criteria for inclusion, with fourteen animals analyzed in each of the control and ceftaroline groups and fteen in the vancomycin group. CSF bacterial loads at T1, T2 and T3 were comparable among all three treatment groups (T1: p = 0.16, T2: p = 0.17 and T3: p = 0.47; Table 1). Reductions in bacterial load from baseline were 88%, 79%, and 75% in the control, ceftaroline and vancomycin-treated groups, respectively (Table 1). No statistically signi cant differences occurred at any time point between controls or those receiving either vancomycin or ceftaroline (T1: p = 0.16, T2: p = 0.17 and T3: p = 0.47; Table 1). The rate of bacterial clearance among the three groups approached signi cance between T l and T 2 (p = 0.063), with ceftaroline clearing at a much faster rate than control (p = 0.019), nally reaching statistically not signi cant difference by T3 (p = 0.43). The bacterial clearance by ceftaroline was reversed by end of T 2 -T 3 , period with CSF bacterial counts increasing in the ceftaroline group, while the control group continued to clear bacteria (p = 0.0039). This result was partly explained by a notable drop in ceftaroline concentrations from T 2 to T 3 , while the vancomycin group maintained relatively high drug concentrations (Fig. 2).

Discussion
Nosocomial bacterial meningitis caused by Methicillin resistant Staphylococcus aureus (MRSA) is a devastating disease , and the standard treatment is vancomycin, . Vancomycin is associated with inconsistent and poor penetration into the cerebral spinal uid (CSF) is inconsistent and (at times) poor .
In this study, we attempted to explore alternative treatments for MRSA meningitis using ceftaroline and rabbit meningitis model. We compared CSF bacterial clearance over different time points among the rabbits treated with ceftaroline, vancomycin and control groups. In consistence with the previous studies, we demonstrated a very rapid bacterial clearance from the CSF in rabbits, suggesting that shorter time points would potentially help to properly assess bactericidal potential of newer antibacterial agents. We noticed a much faster bacterial clearance in ceftaroline treated rabbits compared to control between T l and T 2 (p = 0.019), however, it reached statistically not signi cant difference by T3 (p = 0.43). This result was partly explained by a notable drop in ceftaroline concentrations from T 2 to T 3 , while the vancomycin group maintained relatively high drug concentrations (Fig. 2). Others have demonstrated a similarly short half-life of ceftaroline in rabbit CSF compared to humans, suggesting that additional or higher doses of ceftaroline are necessary for persistent necessary drug exposure in the rabbit model.
This study has several limitations. First, as about fty percent of the rabbits included didn't meet the inclusion criteria, we didn't have substantially enough rabbits for a robust statistical comparison among the treatment groups. Second, the lack of prior information about the natural clearance of Staphylococcus aureus by rabbits complicated the study. Despite these limitations, using a very challenging rabbit meningitis model, this study is able to provide some useful information about the potential use of ceftaroline in selected MRSA meningitis patients unable to take vancomycin, or for patients in whom vancomycin has failed

Declarations
Ethics approval and consent to participate All animal procedures were reviewed and approved by the Duke Institutional Animal Care and Use Committee prior to institution of any study-related procedures (Duke IACUC #A006-15-01).

Consent for publication
We have provided the written informed consent for publication of this manuscript.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.  Rabbit Meningitis Model -Experimental Design.