© 2009 april marie barbour - university of...
TRANSCRIPT
PHARMACOKINETIC/PHARMACODYNAMIC CHARACTERIZATION OF CEFTOBIPROLE
By
APRIL MARIE BARBOUR
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2009
1
© 2009 April Marie Barbour
2
To my father
3
ACKNOWLEDGEMENTS
I would like to express my great appreciation to Dr. Hartmut Derendorf for his support
and guidance throughout my graduate career. Working with Dr. Derendorf has been an
incredible learning experience and I have truly enjoyed my time at UF.
I would like to thank the members of my supervisory committee, Dr. Guenther Hochhaus,
Dr. Jeffrey Hughes, and Dr. Kenneth Rand, for their support and guidance.
I would like to thank the staff of the General Clinical Research Center and Shands
Hospital for their support during the microdialysis study.
I would also like to thank the administrative staff of the Department of Pharmaceutics,
especially Mrs. Khan, Mrs. Keirnan-Sanchez, and Mr. Rhoden.
Finally, I would like to thank all the graduate students and post-docs in the Department of
Pharmaceutics, especially Stephan, Sab, Navin, and Martina.
4
TABLE OF CONTENTS page
ACKNOWLEDGEMENTS .............................................................................................................4
LIST OF TABLES ...........................................................................................................................9
LIST OF FIGURES .......................................................................................................................11
ABSTRACT ...................................................................................................................................13
CHAPTER
1 CLASS-DEPENDENT RELEVANCE OF TISSUE DISTRIBUTION IN THE INTERPRETATION OF ANTIMICROBIAL PK/PD INDICES ..........................................14
Hypothesis and Specific Aims ................................................................................................14 Hypothesis ..............................................................................................................................14 Specific Aims ..........................................................................................................................14 Introduction .............................................................................................................................15 Beta-lactams ...........................................................................................................................17 Aminoglycosides ....................................................................................................................18 Fluoroquinolones ....................................................................................................................19 Oxazolidinones .......................................................................................................................20 Tetracyclines and Glycylcyclines ...........................................................................................21 Macrolides, Ketolides, Azalides .............................................................................................22 Glycopeptides .........................................................................................................................23 Conclusion ..............................................................................................................................24
2 APPLICATIONS OF ANTIMICROBIAL PHARMACOKINETICS/PHARMACODYNAMICS ..........................................................31
Characterization of Antimicrobial Pharmacokinetics/Pharmacodynamics ............................31 Pharmacodynamic Models ......................................................................................................33
One-Population Models ...................................................................................................33 Two-Population Models with Persistent Bacteria ...........................................................34 Two-Population Models with Resistant Bacteria ............................................................36 Adaptation Models ..........................................................................................................38
Summary .................................................................................................................................39
3 CEFTOBIPROLE: A NOVEL CEPHALOSPORIN WITH ACTIVITY AGAINST GRAM-POSITIVE AND GRAM-NEGATIVE PATHOGENS, INCLUDING MRSA ........46
Introduction .............................................................................................................................46 Pharmacokinetics ....................................................................................................................47
Plasma Concentrations ....................................................................................................47 Protein Binding ................................................................................................................48
5
Volume of Distribution ....................................................................................................48 Tissue Distribution ..........................................................................................................48 Clearance and Metabolism ..............................................................................................49 Half-life ...........................................................................................................................49 Pharmacokinetics in Patients with Complicated Skin and Skin Structure Infections .....50 Drug Administration ........................................................................................................50 Adverse Events ................................................................................................................51
Pharmacodynamics .................................................................................................................52 In Vitro Activity ..............................................................................................................52 Resistance Studies ...........................................................................................................53 In Vivo-Animal Studies ...................................................................................................53 In Vivo-Pivotal Studies ...................................................................................................55
Pharmacokinetics/Pharmacodynamics (PK/PD) ....................................................................56 PK/PD Characterization ..................................................................................................56 PK/PD for Dose Selection ...............................................................................................56
Summary .................................................................................................................................57
4 IN VITRO MICRODIALYSIS OF CEFTOBIPROLE ...........................................................65
Objective .................................................................................................................................65 Chemicals and Equipment ......................................................................................................65
Test Article ......................................................................................................................65 Reagents ..........................................................................................................................65 Equipment ........................................................................................................................65
Reagent Preparation ................................................................................................................66 Ceftobiprole Standards and Quality Control Solutions ...................................................66 HPLC Mobile Phase ........................................................................................................66 0.1% Formic Acid in DMSO ...........................................................................................67
Sample Preparation .................................................................................................................67 Calibration Solutions for Microdialysis ..........................................................................67 Dialysate Samples ...........................................................................................................67
Apparatus Setup ......................................................................................................................67 Sample Analysis .....................................................................................................................68
HPLC/UV Set Up ............................................................................................................68 HPLC Pump Conditions ..................................................................................................68 Autosampler Conditions ..................................................................................................68 Detector Conditions .........................................................................................................69 Analysis Procedure ..........................................................................................................69
Microdialysis ..........................................................................................................................69 Introduction .....................................................................................................................69 Extraction Efficiency Method (EE) .................................................................................70 Retrodialysis Method (RD) .............................................................................................71
Results .....................................................................................................................................71 Method Validation Results ..............................................................................................71 In vitro Microdialysis Results .........................................................................................72
Conclusions .............................................................................................................................72
6
5 HPLC/UV METHOD VALIDATION OF CEFTOBIPROLE IN LACTATED RINGER’S SOLUTION .........................................................................................................76
Objective .................................................................................................................................76 Validation Procedure ..............................................................................................................76 Results .....................................................................................................................................77
Reproducibility of the Calibration Curve Parameters .....................................................77 Lower Limit of Quantification ........................................................................................77 Intra-batch Variability for Quality Control Samples .......................................................77 Inter-batch Variability for Quality Control Samples .......................................................78 Freeze-thaw Stability .......................................................................................................78 Stability at Room Temperature .......................................................................................79 Refrigeration Stability of QCs .........................................................................................80 Freezer (long-term) Stability ...........................................................................................80 Robustness .......................................................................................................................80
Chemicals and Equipment ......................................................................................................81 Test Article ......................................................................................................................81 Reagents ..........................................................................................................................81 Equipment and Disposables ............................................................................................81
Reagent Preparation ................................................................................................................82 Ceftobiprole Stock Standard Solution .............................................................................82 Ceftobiprole Working Standard Solutions ......................................................................82 Ceftobiprole QC Samples ................................................................................................82 HPLC Mobile Phase ........................................................................................................83 0.1% Formic Acid in DMSO ...........................................................................................83 HPLC Wash Solution ......................................................................................................83
Sample Preparation .................................................................................................................83 Assay Procedure .....................................................................................................................84
HPLC/UV Set Up ............................................................................................................84 HPLC Pump Conditions ..................................................................................................84 Autosampler Conditions ..................................................................................................84 Detector Conditions .........................................................................................................84 Analysis Procedure ..........................................................................................................84 System Suitability ............................................................................................................85
Summary .................................................................................................................................85
6 SOFT TISSUE PENETRATION OF CEFTOBIPROLE IN HEALTHY VOLUNTEERS DETERMINED BY IN VIVO MICRODIALYSIS ....................................98
Introduction .............................................................................................................................98 Materials and Methods ...........................................................................................................99
Healthy Volunteers ..........................................................................................................99 Microdialysis .................................................................................................................100 Study Design .................................................................................................................100 Analysis Methods ..........................................................................................................101
Results ...................................................................................................................................103 Discussion .............................................................................................................................104
7
7 PHARMACODYNAMC PROPERTIES OF CEFTOBIPROLE AGAINST MRSA ..........109
Introduction ...........................................................................................................................109 Materials and Methods .........................................................................................................110
Study Conduct ...............................................................................................................110 Preparation of Ceftobiprole ...........................................................................................110 Preparation of Sterile Broth and Normal Saline ............................................................110 Inoculum Preparation ....................................................................................................111 MIC Determination .......................................................................................................111 In Vitro Infection Model with Constant Antibiotic Concentration ...............................111 Bacterial Quantification .................................................................................................112 Stability of Ceftobiprole in Mueller Hinton Broth (MHB) ...........................................112
Data Analysis ........................................................................................................................113 Results ...................................................................................................................................113
MIC Determination .......................................................................................................113 Time-Kill Curves ...........................................................................................................113 Stability of Ceftobiprole in MHB ..................................................................................113 Pharmacodynamic Model Development .......................................................................114 Pharmacodynamic Model Validation ............................................................................116
Conclusions ...........................................................................................................................116
8 POPULATION PHARMACOKINETICS AND MONTE CARLO SIMULATIONS FOR THE EVALUTION OF THE CURRENTLY RECOMMENDED DOSING REGIMEN OF CEFTOBIPROLE ........................................................................................124
Introduction ...........................................................................................................................124 Materials and Methods .........................................................................................................125
Population Pharmacokinetic Model Development ........................................................125 Probability of Target Attainment ...................................................................................125
Results ...................................................................................................................................126 Population Pharmacokinetic Model Development ........................................................126 Probability of Target Attainment ...................................................................................128
Conclusions ...........................................................................................................................129
9 CONCLUSIONS ..................................................................................................................145
LIST OF REFERENCES .............................................................................................................149
BIOGRAPHICAL SKETCH .......................................................................................................166
8
LIST OF TABLES
Table page 1-1 Volume of distribution of antimicrobials by class determined from clinical studies ........30
2-1 Definition of each symbol used throughout the chapter ....................................................45
3-1 In vitro activity of Ceftobiprole against common skin and soft tissue and respiratory pathogens ...........................................................................................................................59
3-2 Pharmacokinetics following single and multiple doses presented as mean .......................60
3-3 Most common adverse events in healthy volunteers and patients .....................................61
4-1 Standard preparation of BAL9141 .....................................................................................73
4-2 Method validation accuracy and precision ........................................................................74
4-3 In vitro microdialysis results ..............................................................................................75
5-1 Intra-batch variability of quality control samples ..............................................................86
5-2 Inter-batch variability of quality control samples ..............................................................87
5-3 Freeze/thaw stability of QCs ..............................................................................................88
5-4 Freeze/thaw stability of ceftobiprole stock standard solution ............................................89
5-5 Room temperature stability of ceftobiprole stock standard solution after 5.5 hours ........90
5-6 Room temperature stability of ceftobiprole QCs after 3 hour ...........................................91
5-7 Refrigeration (4ºC) stability of QC samples over 24 and 48 hours ...................................92
5-8 Freezer stability of QC samples over 32 days ...................................................................93
5-9 Interbatch variability of ceftobiprole using a different column to test robustness ............94
5-10 Standard preparation of ceftobiprole .................................................................................95
5-11 Quality control sample preparation of ceftobiprole ...........................................................96
6-1 Noncompartmental pharmacokinetic analysis .................................................................107
6-2 Soft tissue penetration of four cephalosporins determined by microdialysis ..................108
7-1 Pharmacodynamic parameters based on time-kill curves and CDD Results ...................117
9
8-1 Results from bootstrap of base model ..............................................................................140
8-2 Results from bootstrap of covariate model ......................................................................141
8-3 Probability target attainment for fT>MIC of >30% at steady state .....................................142
8-4 Probability target attainment fT>MIC of >40% at steady state ...........................................143
8-5 Probability of target attainment based on 1-log kill at 24 hours ......................................144
10
LIST OF FIGURES
Figure page 1-1 Time-kill curves with Pseudomonas aeruginosa (ATCC 27853) against tobromycin,
ciprofloxacin, and ticarcillin ..............................................................................................26
1-2 Time-kill curves for Streptococcus pyogenes against a glycopeptide, two beta-lactams, a fluoroquinolone, and a macrolide .....................................................................27
1-3 Both the time above MIC and AUC correlate well to the efficacy of tigecycline against Streptococcus pneumoniae 1199 in the neutropenic murine thigh infection model compared to the Cmax ..............................................................................................28
1-4 Increase in free azithromycin concentration once infection was induced as determined by in vivo microdialysis in rats .......................................................................29
2-1 Two hypothetical killing profiles resulting in identical microbial burden after 24 h of antimicrobial exposure. ......................................................................................................41
2-2 Curve fits for four bacterial strains with the various constant concentrations. A monophasic activity profile is displayed over 6 hours .......................................................42
2-3 Time-kill curves for Streptococcus pyogenes exposed to five antibiotics at concentrations ranging from 0 to 64 times the MIC. A biphasic activity profile is displayed over 24 hours with the exception of vancomycin. .............................................43
2-4 Effect of dose schedule on efficacy of ciprofloxacin against Bacillus anthracis ..............44
3-1 Mean concentration-time profile for a 500 mg two hour i.v. infusion in healthy volunteers ...........................................................................................................................62
3-2 Mean volume of distribution, clearance, dose normalized AUC0-∞, and dose normalized maximum concentration vs. dose. ...................................................................63
3-3 Formation of diacetly (2,3-butanedione) during the conversion of ceftobiprole medocaril to the active moiety ceftobiprole .......................................................................64
5-1 Typical chromatogram for ceftobiprole. ............................................................................97
6-1 Mean ceftobiprole concentration in plasma, skeletal muscle ISF, and s.c. adipose tissue ISF over twelve hours. ...........................................................................................106
7-1 Time-kill curves of Ceftobiprole against MRSA performed over 24 hours. ...................118
7-2 Degradation of Ceftobiprole in MHB with MRSA at 37ºC. ............................................119
7-3 Diagnostic plot of observed CFU/mL vs. population predicted. .....................................120
11
7-4 Diagnostic plot of the natural log observed CFU/mL vs. individual predicted. ..............121
7-5 Diagnostic plot of weighted residuals vs. individual predicted. ......................................122
7-6 Individual fits of the model population predicted and individual predicted curves for time-kill experiments .......................................................................................................123
8-1 Two-compartment body model with elimination from the central compartment and drug delivery via i.v. infusion. .........................................................................................131
8-2 Plots of interindividual variability vs. possible covariates. .............................................132
8-3 Dependent variables vs. population predicted values ......................................................133
8-4 Dependent variables vs. individual predicted values .......................................................134
8-5 Weighted residuals vs. population predicted values. .......................................................135
8-6 Individual fits of the covariate model population predicted and individual predicted curves for free plasma ......................................................................................................136
8-7 Individual fits of the model population predicted and individual predicted curves for skeletal muscle .................................................................................................................137
8-8 Individual fits of the model population predicted and individual predicted curves for s.c. adipose tissue .............................................................................................................138
8-9 Example of a success and a failure based on achieving 1-log kill at 24 hours ................139
12
Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
PHARMACOKINETIC/PHARMACODYNAMIC CHARACTERIZATION OF
CEFTOBIPROLE
By
April Marie Barbour
May 2009 Chair: Hartmut Derendorf Major: Pharmaceutical Sciences
Ceftobiprole is a novel-cephalosporin with broad-spectrum activity against both Gram-
positive and Gram-negative pathogens, including activity against numerous resistant species
such as methicillin-resistant Staphylococcus aureus (MRSA) and penicillin-resistant
Streptococcus pneumonia (PRSP). The lead indication of ceftobiprole is complicated skin and
skin structure infections. Therefore, to further characterize the pharmacokinetics, the
concentration in the interstitial space fluid of skeletal muscle and subcutaneous adipose tissue
after a single 500 mg 2 hour intravenous infusion was determined by microdialysis and
compared to plasma concentrations. To characterize the pharmacodynamics, a time-kill
experiment was performed with MRSA to evaluate the pharmacodynamic activity of ceftobiprole
over time. Pharmacokinetic and pharmacodynamic models were developed and simulations
were performed to evaluate the generally recommended dosing regimen of ceftobiprole, a 500
mg 2 hour i.v. infusion every 8 hours. It was found that this dosing regimen is efficacious at the
site of action based on two different pharmacodynamic endpoints, whether the time the free
concentration remained above the MIC for at least 40% of the dosing interval and whether at
least a 1-log bacterial kill was achieved at 24 hours.
13
CHAPTER 1 CLASS-DEPENDENT RELEVANCE OF TISSUE DISTRIBUTION IN THE
INTERPRETATION OF ANTIMICROBIAL PK/PD INDICES
Hypothesis and Specific Aims
Ceftobiprole is a novel, broad-spectrum cephalosporin with activity against both Gram-
negative and Gram-positive bacteria including some resistant pathogens such as MRSA and
PRSP. Currently, ceftobiprole medocaril, the water soluble prodrug of ceftobiprole, is under
FDA review for approval with the lead indication of complicated skin and skin structure
infections (cSSSI). The lead indication of cSSSI merits the exploration of not only plasma
concentrations but also of free concentrations with interstitial space fluid (ISF) of subcutaneous
tissues to ensure penetration of this intravenously administered compound to the site of action.
Once the tissue concentrations have been determined, it is important to explore the
corresponding pharmacodynamic result. The pharmacodynamic parameter typically used in
PK/PD evaluations is the minimum inhibitory concentration (MIC). However, this
pharmacodynamic measure has many drawbacks such as an inherent two-fold variability and the
lack of a relationship between the antimicrobial activity and time. Therefore, time-kill curve
experiments are a more rational approach for pharmacodynamic characterization of a compound.
Hypothesis
A currently recommended dosing regimen of ceftobiprole, 500mg administered as a 2 hour
intravenous (i.v.) infusion every eight hours, is efficacious at the site of action based on the
pharmacodynamic endpoints of time above the MIC and a 1-log bacterial kill.
Specific Aims
1. Perform in vitro microdialysis experiments to validate the feasibility of using microdialysis as a sampling technique to determine free ISF tissue concentrations of ceftobiprole.
2. Develop and validate an HPLC-UV method for the analysis of ceftobiprole in lactated Ringer’s solution according to GLP guidelines.
14
3. Perform an in vivo microdialysis clinical study to determine the free ISF concentrations in skeletal muscle and subcutaneous adipose tissue of ceftobiprole in healthy volunteers after a single 500mg dose given as a two-hour i.v. infusion.
4. Perform in vitro pharmacodynamic studies to characterize the antimicrobial activity of ceftobiprole against MRSA by performing time-kill experiments and develop a mathematical model to describe the antimicrobial activity.
5. Develop and validate a pharmacokinetic model to evaluate the efficacy of the current dosing regimen in plasma, skeletal muscle, and subcutaneous adipose tissue based on two different pharmacodynamic endpoints, the free time above MIC as a percentage of the dosing interval and a 1-log bacterial kill from the starting inoculum at 24 hours.
Introduction
Pharmacokinetic/pharmacodynamic (PK/PD) characterization of antimicrobial agents has
allowed a better understanding of why a particular dosing regimen achieves clinical success or
failure. Using techniques first pioneered by Eagle (62, 63) and further developed by Craig (6, 8,
44), dose selection has become a much more sophisticated process over previous empiric
methods. In vitro and in vivo experiments are used to define a relationship between drug
concentration (PK) and effect (PD) and allow for a clear target to be identified so that efficacy is
achieved in the clinical setting (2). The PK/PD indices typically used for antimicrobials include
the time the free drug concentration remains above the minimum inhibitory concentration (MIC)
expressed as a percent of the dosing interval (fT>MIC), the ratio of the maximum concentration
and MIC (Cmax/MIC), and the ratio of the twenty-four hour area under the concentration-time
curve and MIC (AUC0-24/MIC) (44). The in vivo PK parameters are usually determined from
serum or plasma drug profiles while in vitro PD parameters commonly use culture media drug
concentrations.
The index which best correlates with a particular antibiotic depends on several factors.
One of these factors is the pattern of microbial kill exhibited by the compound which is
frequently referred to as either time-dependent killing or concentration-dependent killing.
15
Antibiotics which display time-dependent killing typically reach a maximum effect at a
concentration approximately 4 times the MIC (47, 99). Once this maximum effect is reached
increasing the concentration does not increase the rate of bacterial death as shown for ticarcillin
in Figure 1-1 (47). Antibiotics which display concentration-dependent killing produce an
increasing effect as the concentration increases as shown for tobramycin and ciprofloxacin in
Figure 1-1. However, categorizing an agent’s antimicrobial activity as time- or a concentration-
dependent is not always obvious as shown in Figure 1-2 where profiles for concentrations greater
than 4 times the MIC appear similar for all compounds (136).
Concentration-dependent antibiotics usually correlate efficacy with exposure and the MIC,
i.e. the Cmax/MIC or the AUC0-24/MIC. The efficacy of time-dependent antimicrobials usually
depends on fT>MIC as a percentage of the dosing interval. However, time dependent antibiotics
which display a prolonged post-antibiotic effect (PAE), the persistence of activity after removal
of the drug or after concentrations drop below the MIC (29, 45, 190), often correlate well with
the AUC0-24/MIC. Although the mechanisms of the PAE have been speculated, the PAE seems
to depend on the drug, the pathogen, and the infection model (29). The term PAE is used
collectively to explain residual effects of complex pharmacokinetic and pharmacodynamic
processes that are not well defined.
Other factors which influence the significance of PK/PD indices for predicting clinical
efficacy that are often overlooked include protein binding and tissue distribution. Recently it has
been stressed that only free concentrations should be taken into account in these indices as only
free drug has the ability to exhibit a pharmacological effect (138, 160, 194). Similarly, it
follows that the most relevant concentrations for antibiotic efficacy would be the concentration
within the interstitial space fluid (ISF) of target tissues as most infections are located outside the
16
plasma or serum where drug concentrations are commonly monitored. The volume of
distribution, and subsequently the degree of tissue distribution, varies greatly between
antimicrobial classes as shown in Table 1-1 (30, 40, 89, 102, 115, 128, 167, 174, 184, 185, 196).
It is the goal of this paper to review the PK/PD relationship of various antimicrobial classes and
analyze this relationship by examining the mechanism of action, in vitro kill-kinetics, and free
tissue concentrations. Additionally, the PAE will be discussed, in some instances as it may be
used to account for additional uncharacterized PD effects.
Beta-lactams
Beta-lactam antibiotics, which include the penicillins, cephalosporins, and carbapenems,
work by inhibiting cell wall synthesis by binding to penicillin binding proteins (PBPs). The class
displays time-dependent killing that is consistent with this mechanism of action. In general, the
beta-lactams correlate well with fT>MIC . The PAE is virtually absent with gram-negative
bacteria, except the carbapenems (32, 45), and only a modest PAE is reported with gram-positive
bacteria, mainly staphylococci (46, 145, 192). The time needed for the concentration to remain
above the MIC is slightly different for each subclass but consistent within groups as long as only
free concentrations are considered due to fact that protein binding can vary greatly between
compounds, with 2% reported for meropenem and approximately 90% for ertapenem (11, 112,
160). The fT>MIC needed for bactericidal effects in gram-negatives is approximately 40-50% for
carbapenems, 50-70% for penicillins, and 70-80% for cephalosporins (17, 30). For gram-
positive bacteria the fT>MIC needed is approximately 25-45% for carbapenems, 35-50% for
penicillins, and 40-50% for cephalosporins (17). It can be seen that the carbapenems display
efficacy with a lower fT>MIC, mainly because this subclass displays the fastest rate of kill (44).
Also noteworthy, is that the fT>MIC needed for efficacy is generally less than 40-50% for
Staphylococcus spp. possibly due to the PAE (46, 192), which may be better defined by more
17
thoroughly characterizing PK and/or PD properties. It is interesting that while the beta-lactams
have a volume of distribution of ~18L (30) and they do not accumulate at the site of action, i.e.
the ISF where concentrations are typically equal to or less than free plasma concentrations (94,
101, 181), the time needed for free concentrations to remain above the MIC is less than the
dosing interval.
Aminoglycosides
The PK and PD of the aminoglycosides are fairly well characterized and appear to be
consistent throughout the class. The mechanism of action of this class involves binding to the
30S ribosomal subunit and preventing protein synthesis (65). Typically, antibiotics that inhibit
protein synthesis or replication display a concentration-dependent killing pattern and a PAE, as is
the case for the aminoglycosides (16, 99). With these pharmacodynamic characteristics, the
efficacy of aminoglycosides in clinical studies correlates to a Cmax/MIC ratio that exceeds 10
(95) or an AUC0-24/MIC ratio between 80 and100 for maximum effects in animal models of
infection (184, 191). One possibility that may explain a better the correlation between Cmax/MIC
compared to the AUC0-24/MIC is the emergence of resistant populations. Resistance occurs
much less when a Cmax/MIC ratio of 10 is achieved (20). Also, uptake of these compounds into
bacterial cells is energy dependent (27, 65) and decreases after an initial exposure (49).
Therefore, by allowing a larger, less frequent dose and subsequent larger Cmax, better efficacy
may be achieved due prevention of resistant mutants, mechanism of antibacterial uptake, and
other PAE mechanisms. In fact, clinical investigators advocate once daily dosing of
aminoglycosides (41, 135) but more data is needed to show a statistically better clinical outcome
with a once daily dosing regimen.
With the low protein binding and a volume of distribution approximately equal to
extracellular space fluid (184), the target AUC0-24/MIC of 80-100 for aminoglycoside clinical
18
efficacy indicates that the average concentration exceeds the MIC over a 24 hour period. A
target of approximately 24 would be expected from described PK and PD parameters. The
higher AUC0-24/MIC suggested target of 80-100 results from several factors. These factors
include, the target for clinical efficacy it is based on a maximum effect and not bacteriostasis,
the low protein binding may not truly be negligible, and use of plasma concentrations rather than
tissue ISF levels. Again, the most relevant concentrations are the free concentrations at the site
of action and it has been shown that antibiotics which distribute to the ISF, may have tissue
concentrations lower than plasma concentrations. For example, in subcutaneous adipose tissue
gentamicin had a Cmax and AUC0-360 that were approximately 39% and 60% of values
determined from plasma, respectively (108). Even when the protein binding is considered,
approximately 20% (132), free plasma concentrations do not equal free ISF concentrations.
Fluoroquinolones
The fluoroquinolones prevent DNA replication by inhibiting type II topoisomerases, also
called DNA gyrase, and topoisomerase IV (76, 81), and may also affect bacterial membranes
(176). Like the aminoglycosides, the fluoroquinolones display concentration dependent killing
(47) and both the AUC0-24/MIC and Cmax/MIC ratios are correlated with efficacy. However,
several clinical studies have revealed the AUC0-24/MIC ratio is the better index (61, 70, 71, 149).
A high target Cmax/MIC ratio may aid in the prevention of resistance development (149). In
gram-negative infections an AUC0-24/MIC ratio of 125, free ~75, was found to be sufficient for
efficacy of ciprofloxacin (71). Similarly, an AUC0-24/MIC of ≥87, free ~62, was found to be
efficacious for levofloxacin (61) and an AUC0-24/MIC of 75-175, free 37.5-87.5 for
grepafloxacin provided an 80% probability of cure for patients with chronic bronchitis (70). For
gram-positive pathogens, such as, Streptococcus pneumoniae, the breakpoints seem to be lower.
In patients with community-acquired pneumonia or acute exacerbation of chronic bronchitis that
19
received either levofloxacin or gatifloxacin, a free AUC0-24/MIC ratio above 33.7 provided
complete efficacy (3). A Cmax/MIC ratio of 12.2 was also found to predict efficacy in a clinical
study (149). This finding was supported by a dose fractionation study in the neutropenic mouse
model with lomefloxacin (60). However, in this animal model the AUC0-24/MIC ratio was more
highly correlated to efficacy at lower doses where the Cmax/MIC was less than10.
It is interesting that the magnitude of the plasma based PK/PD indices which correlate with
efficacy are in good agreement among this class. The magnitudes of these targets are also
similar to the aminoglycosides when comparing free AUC0-24/MIC. However, the
fluoroquinolones distribute more extensively into tissue with free tissue/free plasma ratios of
approximately 0.9->2 (88, 166, 195). Therefore, it is somewhat surprising that the magnitude of
these indices associated with efficacy is not lower and may be due to some unexplained
difference in the PD associated with different mechanisms of action between aminoglycosides
and fluoroquinolones.
Oxazolidinones
Linezolid is the only approved oxazolidinone on the market and has activity against several
resistant pathogens including MRSA, VRSA, VRE, PRSP (26, 35, 84). The mechanism of
action of this class is inhibition of protein synthesis at the initiation stage by binding to the 50S
subunit of the bacterial ribosome (197) and preventing the first amino acid, methionine, tRNA
complex from binding at the peptidyl site (21). The pharmacodynamics of linezolid were
described in a dose fractionation study using the neutropenic murine thigh infection model (8).
Although this antibiotic displays time-dependent kill kinetics (157, 198), it best correlates with
the AUC0-24/MIC ratio, not the time above MIC, for Streptococcus pneumoniae and correlates
well with both the AUC0-24/MIC and time above MIC for Staphylococcus aureus (8). The AUC0-
24/MIC ratio required for a bacteriostatic effect with linezolid was 48 for pneumococci and 83
20
staphylococci in the neutropenic murine thigh infection model (8). In a clinical study with
critically ill patients with bacteremia, an AUC0-24/MIC of ≥105 and a time above MIC of 82%
was associated with a faster time to bacteria eradication (152). The slightly higher target in the
clinical study is probably due to a more aggressive endpoint. Taking the reported protein
binding of 8.7-31% (53, 111) and tissue distribution into consideration, free tissue and free
plasma concentrations are approximately equal (53). These targets are in agreement with the
aminoglycosides and fluoroquinolones.
Tetracyclines and Glycylcyclines
The pharmacokinetics and pharmacodynamics of the tetracyclines and glycylcyclines,
especially tigecycline, are not fully understood (14). The mechanism of action of this antibiotic
class is inhibition of protein synthesis by binding to the 30S subunit of the prokaryotic ribosome
(196). The tetracyclines display a time-dependent killing pattern and exhibit considerable PAE
(44, 48, 142, 148, 189). A prolonged PAE, as observed with tigecycline (189), suggests that the
best PK/PD index is the AUC0-24/MIC. However, time above the MIC for free drug levels versus
effect also had a high correlation coefficient as shown in Figure 1-3 (189). In clinical trials with
patients with complicated skin and skin structure infections (cSSSI) or complicated intra-
abdominal infections (cIAI), the AUC/MIC ratio was used as the PK/PD index for target
development and ratios of 17.9 and 6.96, respectively, produced a better outcome (117, 146).
Although the PAE was used in target selection, it is important to remember that it is the
result of residual PK and PD effects that are not well characterized. The PAE of this class of
antibiotic may be the result of extensive tissue distribution resulting in higher active
concentrations at the site of action. The tetracyclines display a varied volume of distribution
with values ranging from 0.14L/kg to 1.6L/kg (196). Tigecycline, however, is the most
extensively distributed compound in this class with a volume of distribution of 7-10L/kg (40,
21
128). The high volume of distribution reflects that free concentrations in the ISF could be
significantly higher than plasma. This theory is supported by the low PK/PD target values
calculated for tigecycline from clinical trials. When protein binding is considered, 36-47% at a
pH of 7.3 (144), the targets are much lower than would be expected to achieve efficacy, a
minimum of 24 for the free AUC0-24/MIC. However, the suggested targets for the free drug are
<12 for cSSSI and <5 for cIAI.
Macrolides, Ketolides, Azalides
The macrolides are characterized by a 14-16 member macrocyclic lactone ring. The
mechanism of action involves reversible binding to the 50S ribosomal subunit and prevention of
protein synthesis (109). Although this type of mechanism usually results in a bactericidal agent,
the macrolides can display bacteriostatic or bactericidal killing depending on the bacterial
pathogen (69, 153). Overall, this class mainly displays time-dependent killing (5). As with the
tetracycline/glycylcycline class, a residual antimicrobial effect or PAE is observed for
macrolides (143, 151). However, no single PK/PD parameter appears to correlate with efficacy
for all members in this class and all three target parameters have been suggested (5, 55, 89, 137,
141, 169, 188). This likely due to several factors including differences in kill-kinetics and large
differences in the pharmacokinetic properties (89).
In a clinical study with telithromycin, it was found that a breakpoint of 3.38 (AUC/MIC)
was predictive of efficacy (106). This is much lower than most other antibiotic classes but
similar to tigecycline. Like tigecycline, telithromycin displays extensive tissue distribution with
a volume of distribution of 2.9L/kg (147, 169), suggesting that concentrations may be higher at
the site of action. This is also supported by a microdialysis study in healthy volunteers with
telithromycin which found AUC0-8 ratios of free tissue/free plasma of 2.1±1.6 and 1.5±0.9 for
subcutaneous adipose and muscle tissue, respectively (79). Another explanation for a low
22
PK/PD target index is that these antibiotics accumulate intracellularly, especially azithromycin
and telithromycin (24, 80, 133). In human polymorphonuclear leukocytes (PMNs) after 2 hours
of exposure with 10 µg/mL, azithromycin displayed an intracellular/extracellular ratio of 79
while erythromycin displayed and intracellular/extracellular ratio of 16 (80). When
telithromycin was given as a 600 mg oral daily dose for 10 days, the concentration in white
blood cells was approximately 100 times that in plasma (124, 133). Therefore, the macrolides
may actually be carried to the infection site by white blood cells resulting in a lower magnitude
of the appropriate PK/PD index for efficacy (4, 80, 159). This is supported by several studies,
such as a microdialysis study where the free concentration of azithromycin within infected tissue
was much higher that noninfected tissue (Figure 1-4). Also, a study using skin blister fluid found
the concentration of azithromycin to be higher in inflammatory blisters than either plasma or
non-inflammatory blisters (4, 72). These high concentrations are the result of ion-trapping of
azithromycin, a dibasic compound, within acidic lysosomes (4). Therefore, the low PK/PD
target is at least in part the result of higher concentrations at the site of action.
Glycopeptides
Like the beta-lactams, glycopeptides also inhibit cell wall synthesis. This antibiotic class
works by forming hydrogen bonds with bacterial cell wall intermediate peptides and thereby
inhibiting peptidoglycan synthesis (154). As characteristic for antibiotics that inhibit cell wall
synthesis, the glycopeptides mainly display time-dependent killing in vitro (19, 67, 100, 110,
156). However, oritavancin and dalbavancin have been shown to display bactericidal
concentration-dependent activity in vitro (19, 175). In terms of PK/PD, the neutropenic mouse
thigh infection model revealed the AUC/MIC to be predictive of efficacy with vancomycin
against Staphylococcus aureus (64). In the nonneutropenic mouse peritonitis model with
vancomycin against Staphylococcus aureus and Streptococcus pneumoniae, survival was
23
correlated to the Cmax/MIC (98). In a clinical study involving patients with Staphylococcus
aureus lower respiratory tract infection receiving vancomycin, an AUC0-24/MIC ratio of ≥400
resulted in a significantly better clinical outcome and bacteriological response (119). Although a
magnitude of 400 is high compared to other antibiotics, this value is in better agreement, when
free tissue concentrations are considered. This compound displays a variable protein binding,
~10-80% (1, 177), and penetration into the ISF may be less than free plasma concentrations
(171). Dalbavancin efficacy has been found to correlate to both AUC0-24/MIC and Cmax/MIC in a
neutropenic murine thigh infection model with and R2 of 77% and 57% respectively against
Staphylococcus aureus and 78% and 90% respectively against Streptococcus pneumoniae (6).
Recently, it has been shown that oritavancin best correlates with fT>MIC as a percent of the dosing
interval (18). In patients with Staphylococcus aureus bacteremia infections, it was found that a
fT>MIC of 22% had a 76% probability of microbiological cure and an 87% probability of clinical
success. This was similar to the findings of an in vivo neutropenic murine thigh infection model,
where a fT>MIC of ~17-20% was bacteriostatic (18, 25). However, in the murine model the free
Cmax/MIC was the most highly correlated parameter. The PK/PD parameter which best
correlates to efficacy seems to be dependent on drug, pathogen, and infection model, which is a
reflection of the difference in kill-kinetics and possibly other PK and PD properties among this
class. Therefore, each compound should be individually characterized both in vitro and in vivo
before a PK/PD target is selected.
Conclusion
Pharmacokinetic/pharmacodynamic characterization of antimicrobial agents allows for a
much more rational approach to antimicrobial dosing than traditional empiric methods. PK/PD
indices have been well established for antimicrobial agents but are not fully understood among
all antibiotics and/or antibiotic classes. The rationale behind which PK/PD index best predicts
24
efficacy and the needed magnitude is depended on several factors. As a general rule,
antimicrobial agents which have a mechanism of action of inhibition of either protein or bacterial
synthesis generally display concentration-dependent killing and correlate well with the AUC0-
24/MIC ratio or Cmax/MIC ratio. Antimicrobials which act on the cell wall display time-
dependent killing and correlate to the time the free concentration remains above the MIC.
However, a major exception is that antimicrobials which display a PAE correlate better with the
AUC0-24/MIC ratio. The PAE is identified as antimicrobial activity in the absence of antibiotics
or at concentrations below the MIC.
One area that is often overlooked when developing PK/PD targets is the tissue
distribution and in fact the in vivo PAE may be at least in part due higher free tissue
concentrations at the site of infection as these are concentrations are responsible for the desired
clinical effect (121, 179). Beta-lactams and aminoglycosides for example distribute well into
ISF but free concentrations in tissue may be lower than free concentrations in plasma.
Tigecycline and azithromycin, however, both extensively distribute into tissues and the active
tissue concentrations at the infection site may be higher than plasma concentrations. In
conclusion, when optimizing the dosing regimen the pharmacokinetic properties including
protein binding and tissue distribution, and the pharmacodynamics, both in vitro and in vivo,
should be well characterized.
25
Figure 1-1. Time-kill curves with Pseudomonas aeruginosa (ATCC 27853) displaying
tobromycin and ciprofloxacin as having concentration-dependent kill-kinetics and ticarcillin as having time-dependent kill-kinetics. Reprinted with permission from (47).
26
Figure 1-2. Time-kill curves for Streptococcus pyogenes showing that although vancomycin, a
glycopeptide, displays time-dependent killing, there is little difference between the other compounds which include two beta-lactams, a fluoroquinolone, and a macrolide. Reprinted with permission from (136).
27
Figure 1-3. Both the time above MIC (panel a) and AUC (panel b) correlate well to the efficacy
of tigecycline against Streptococcus pneumoniae 1199 in the neutropenic murine thigh infection model compared to the Cmax (panel c). Reprinted with permission from (189).
28
Figure 1-4. Increase in free azithromycin concentration once infection was induced as
determined by in vivo microdialysis in rats. Azithromycin was given 50mg/kg subcutaneously. (Figure from Scaglione, presented at ICAAC 2006, work unpublished)
29
Table 1-1. Volume of distribution of antimicrobials by class determined from clinical studies (30, 40, 89, 102, 115, 128, 167, 174, 184, 185, 196)
Antimicrobial class Volume of distribution Physiological distribution Beta-lactams ~0.26 L/kg* EF Aminoglycosides ~0.1-0.3 L/kg EF Oxazolidinones ~0.51-0.67 L/kg* TBW Fluoroquinolones 1.3-8.3 L/kg >TBW Tetracycline (Tigecycline) 0.14L/kg-1.6L/kg (7-10L/kg) <TBW->TBW Macrolides 0.64L/kg-23L/kg ~TBW-Much Greater than TBW Glycopeptides ~0.1-1L/kg EF-TBW *Based on an average weight of 70kg. EF=Extracellular Fluid. TBW=Total Body Water
30
CHAPTER 2 APPLICATIONS OF ANTIMICROBIAL PHARMACOKINETICS/PHARMACODYNAMICS
Characterization of Antimicrobial Pharmacokinetics/Pharmacodynamics
The learn/confirm paradigm for drug development, as discussed by Sheiner (168), has
provided a relatively new but more efficient method for development of an investigational
pharmaceutical. This model-based approach suggests the focus of drug development should be
an understanding of the science rather than using empirical evidence to make decisions. While
sequential drug development may be slightly more time consuming early in the process,
numerous examples in literature have been cited in which model-based drug development has
been used to make critical decisions including lead compound selection, trial design, and dose
selection (38, 118) . All of these examples have ultimately increased efficiency by saving time,
money, and resources. The FDA has also stated its support of model-based drug development
(186) and approved gabapentin for post herpetic neuralgia based partly on efficacy evidence
provided by PK/PD modeling (118).
The techniques mentioned above can also be applied to the development of antimicrobial
agents and post-approval for the evaluation and possible adjustment of recommended dosing
regimens. Unique to antimicrobials, the pharmacodynamics can be extensively and accurately
characterized in in vitro and in vivo animal models with a good correlation to the effect in
infected patients (2) due to the fact that the site if action is the same in all systems, the bacteria.
In order to predict efficacy and select the appropriate dosing regimen, antimicrobial PK/PD has
been traditionally related to one of three indices; the time the free concentration remains above
the minimum inhibitory concentration (MIC), the maximum free concentration to MIC ratio, and
the area under free twenty-four hour concentration-time curve to MIC ratio (44). Dose fraction
studies are typically performed to select the most appropriate index and magnitude of the index
31
for each agent (6, 7, 46, 189). Additionally, Monte Carlo simulations are performed to find the
most appropriate dosing regimen or evaluate a current regimen based on the pharmacokinetic
variability and pharmacodynamic variability, i.e. inter/intrapatient variability and the MIC
distribution (59, 104, 105, 134). Although these techniques have greatly improved antimicrobial
development over empiric methods, there is an opportunity and means to improve the processes
currently used.
Traditionally, plasma samples are used as the pharmacokinetic input in PK/PD models. It
has now been generally accepted that free concentrations should be used in these models as only
free drug is active (125, 138, 160). Also, it is important to consider the free concentration at the
site of action. Although plasma maybe the site of action, i.e. bacteriemia, it is often not. For
example, in regards to complicated skin and skin structure infections, it is more meaningful to
determine the concentration within the interstitial space fluid (ISF) of subcutaneous soft tissues,
e.g. skeletal muscle or adipose tissue. This can be accomplished by using the microdialysis
sampling technique, which has been proven suitable for the measurement of free ISF
antimicrobial concentrations in virtually any tissue in both healthy volunteers and patients (28,
75, 180, 182).
As previously mentioned, the pharmacodynamic parameter most often used with
antimicrobials is the MIC. However, there are many limitations if the MIC is used as the only
pharmacodynamic parameter. Mainly, the MIC does not characterize the antimicrobial activity
over time but is rather a static one-point in time measurement. It is possible for an agent to
display different kill-kinetics but the same bacterial load after a given incubation period (Figure
2-1). The MIC also does not indicate the degree of pharmacological effect, e.g. a 1-log kill or a
2-log kill over twenty-four hours. The MIC has a two-fold variability and is somewhat
32
subjective as it is based on visual detection of bacterial growth. Finally, this parameter does not
measure the presence or concentration of persistent/resistant bacteria that may be present but
may not have grown to a detectable limit by the unaided eye. Time-kill curves are an alternative
experimental technique to characterize the pharmacodynamic activity of an antimicrobial agent
and circumvent many of the limitations of using the MIC as the pharmacodynamic parameter.
These experiments measure the antimicrobial activity over time, determine the extent of
antimicrobial activity, and may detect persistent/resistant populations. The concentrations used in
these experiments maybe static or dynamic, i.e. changing to simulate the half-life. Dynamic
experiments also have the additional benefit of capturing the antimicrobial activity after
concentrations drop below the MIC.
Due to the numerous advantages of this technique over the MIC, kill-curve experiments
have gained in popularity. The mathematical models which have been developed and applied to
these experiments for pharmacodynamic characterization are compared below. Additionally, the
applications of these experiments and models are mentioned.
Pharmacodynamic Models
One-Population Models
The primary goal of time-kill curves is to characterize the pharmacodynamic activity of
an antimicrobial agent often with the intention of using this information to optimize the dosing
regimen. These experiments are performed by exposing bacteria to a range of concentrations,
generally as multiples of the MIC. The results are then typically fitted to an Emax model,
Emax= , to characterized the activity and calculate the pharmacodynamic parameters as a
function of the change in the number of bacteria over time (Equation1) (113, 123). This model
can be modified to account for a delay in growth and/or kill and saturation in the number of
33
bacteria in the in vitro system (Equation 2) (183). All symbol definitions are provided in Table
2-1.
(Equation 1)
1 1 1 (Equation 2)
As can be seen by these equations and the monophasic kill pattern (Figure 2-2), the
bacteria are all assumed to have the same susceptibility. However, if these experiments are
carried out over a long enough time-period a biphasic antibacterial effect is usually observed
(Figure 2-3). Therefore, models which account for differences in susceptibility have gained
much popularity as they better characterize the pharmacodynamics.
Two-Population Models with Persistent Bacteria
Many pharmacodynamic models have been developed which describe the bacterial
population as two distinct subpopulations, one which is susceptible to drug and one which is not.
In these models, the susceptible population is growing while the persistent/resistant population
may be dividing or in a state of hibernation. The presence of a persistent/resistant population has
been experimentally validated (13, 58, 97, 178) but it is difficult to say that either resistance or
persistence is the rationale behind the biphasic profile of antimicrobial activity as it could be a
combination of both. The difference between the two populations is that resistant bacteria have a
genetic mechanism of resistance and grow on agar supplemented with antibiotic, while persistent
bacteria remain sensitive to the antibiotic once regrown. It has been suggested that if a large
inoculum is used, then the biphasic profile is more likely to be attributed to resistance as the
mutation rate is 10-7 -10-8 (178). If a smaller inoculum is used, e.g. the standard of ~106
CFU/mL, then the presence of persisters may be more likely than a genetically acquired
34
resistance mechanism (136). Many authors have applied the persistent theory to model the kill
kinetics of antimicrobials from time-kill experiments.
A persistent model was developed during a comparison of the in vitro bactericidal
kinetics of S-4661, a new carbapenem, to meropenem, imipenem, cefpirome, and ceftazidime
using several bacterial strains which included, Escherichia coli, Pseudomonas aeruginosa, and
Staphylococcus aureus (193). This model places the bacteria in one of two states, susceptible
and dividing or persistent (Equ and 4)ations 3 (193).
(Equation 3)
(Equation 4)
A similar but slightly modified model was developed by using an additional term to
account for the natural death rate on both the persistent and susceptible bacteria (Equations 5-8)
(136). This model also takes into account the transition of susceptible cells into persisters once
the maximum bacterial concentration is reached in the system, where ksr is equal to a
proportionality constant times the total bacterial concentration in the system. The purpose of this
experiment was to identify a robust model that could be used for several drug/bug combinations
with the promise of eventually aiding in dose optimization of new antimicrobials. This model
was able to fit the kill-kinetics of several antimicrobials from different classes against
Streptococcus pyogenes including the penicillin benzylpenicillin, the cephalosporin cefuroxime,
the macrolide erythromycin, the fluoroquinolone moxifloxacin, and the glycopeptide
vancomycin. In this paper, several models were applied to the data which placed the
antimicrobial effect on the growth rate (Equation 5), as an additive effect on the natural death
rate (Equation 6), or as an effect on death separate from the natural death rate (Equation 7). The
equation for the persister population is the same during the comparison of the models (Equation
35
8). It should be noted that krs was fixed to zero and a mixture model was used so that the initial
inoculum were either in logarithmic growth or a mixture of logarithmically growing bacteria and
persisters. If the starting inoculum was a mix with both persisters and susceptible bacteria, then
it was estimated that 5% of the population were persisters.
1 (Equation 5)
(Equation 6)
1 (Equation 7)
(Equation 8)
In these equations Drug is the Emax model with a hill factor. Equations 5-7 presented above for
the antimicrobial effect fit the data equally well.
Two-Population Models with Resistant Bacteria
As previously mentioned, it is also possible to have two distinct populations within a
bacterial culture, one susceptible to drug and one resistant. These populations may in fact
display different profiles for growth and kill, presented below as a net effect model (Equations 9
and 10) (33). This model provided a better fit than a model which had similar growth and death
rate constants or one which allowed for adaptation and the appearance of resistance. The
purpose of model development was to characterize the antimicrobial effect of ciprofloxacin in
vitro on both susceptible and resistant populations after exposure to concentrations equivalent to
those observed in vivo a r roflo in dosin mens. fter va ious cip xac g regi
1 (Equation 9)
1 (Equation 10)
36
In a subsequent study, two-populations models were compared to examine which and
how the parameters are affected by the antibiotic (39). The models compared included the net
effect model (Equations 9 and 10), the growth inhibition model (Equations 11 and 12), the death
stimulation model (Equations 13 and 14), and the MIC based model (Equations 15 and 16) (116).
It was found that based on goodness-of-fit, bias of observed vs. predicted, precision of the
estimates, and predictive performance that the net effect model and growth inhibition model
were the superior models.
1 1 Equation (11)
1 1 Equation (12)
1 1 Equation (13)
1 1 Equation (14)
1 Equation (15)
1 Equation (16)
Theoretically, it is possible to have several populations of bacteria within a given
inoculum. In the above MIC model, originally proposed by Meagher et. al., three different
bacterial populations were assumed, i.e. three different values for SITm, the concentration needed
to produce the median effect for C/MIC (116). However, the pharmacodynamics were
adequately described with only two values, where SITm is either susceptible (SITmS) or
intermediately susceptible (SITmR), with all other parameters remaining the same between the
37
different populations (Equation 17) (116). Additionally, in this model, growth was a function of
the bacterial load in the system and not a constant. This model was originally proposed to
characterize the pharmacodynamics of ciprofloxacin and model the effects of different
ciprofloxacin dosing regimens and formulations, an extended release and an immediate release.
1 (Equation 17)
Adaptation Models
Thus far models have been discussed that assume two distinct populations. It is also
possible to model a bacterial population which undergoes adaptive resistance in the presence of
an antimicrobial agent. For example, to explain the biphasic kill profile, a model was developed
which assumed that adaptive resistance changed the rate of bacterial kill from an initial rapid kill,
ε1, to a slower permanent kill, ε2. This model was also developed for comparison of the
pharmacodynamics of two different dosing regimens of ciprofloxacin, a twice daily immediate
release and an once daily extended release, by simulating in vivo concentrations in an in vitro
model, i.e. concentrations that changed according to the half-life (Equations 18 and 19) (165).
1 (Equation 18)
(Equation 19)
In the above model, the adaptation was placed on the kill rate constant. However, it is
also possible to say that the adaptation is due to a change in the susceptibility and not the rate of
kill, i.e. a change in the EC50 (Equations 20 and 21) (178). This model was developed to
characterize the pharmacodynamics of meropenem against Pseudomonas aeruginosa. Although,
like many other models, it is robust and has the potential to be applied to other drug/bug
38
combinations (178). To ensure the presence of resistant population(s) at the start of the study, a
large initial inoculum,108 CFU/mL, was used.
1 (Equation 20)
1 1 (Equation 21)
Similar to the approaches used during model development in two-population models, the
possibility of adaptation due to a change in the growth rate was also examined (Equation 22)
(123). This model was developed to compare the pharmacodynamics of two different dosing
approaches wit , con fusio nd interm infusion. h ceftazidime tinuous in n a ittent
1 1 (Equation 22)
Summary
In vitro experiments to determine the antimicrobial activity of a particular agent often
correlate well with the in vivo situation due to the fact the site of action is the bacteria. Kill-
curve experiments are particularly useful as they provide a detailed pharmacodynamic profile, a
major advantage over the MIC approach. In fact, the MIC may be misleading as two different
dosing regimens may produce the same AUC/MIC ratio but different profiles of activity (Figure
2-4). Several examples have been given in literature using kill-curves to evaluate/select the
optimal dosing regimen. A few examples include comparing continuous and intermittent
infusions of ceftazidime (123), a comparison of an extended release and immediate release
formulation of ciprofloxacin (116, 165), dose selection of levofloxacin against Bacillus anthracis
(58), and optimizing the dosing regimen to prevent resistance development (33, 93). While most
examples use in vivo plasma data combined with time-kill curves, it seems more meaningful to
explore the PK/PD relationship based on antibiotic concentrations at the site of infection, e.g.
39
subcutaneous adipose tissue for complicated skin and skin structure infections, to make a dosing
recommendation (54). Therefore, there are still means to improve antimicrobial dose
optimization.
Many semi-mechanistic models have been proposed to model the pharmacodynamic
activity of antimicrobial agents from time-kill experiments. As can be seen from a comparison
between the models, they share many similarities and are not necessary entirely unique. More
research is needed to fully understand the mechanisms which lead to this biphasic
pharmacodynamic profile. Currently, the models presented within this chapter all seem justified
and serve the purpose of their development. However, a better understanding of the
pharmacodynamics may lead to improved dose optimization.
40
Figure 2-1. Two hypothetical killing profiles resulting in identical microbial burden after 24 h of
antimicrobial exposure. Reprinted with permission from (178).
41
Figure 2-2. Curve fits for four bacterial strains with the various constant concentrations (mg/L):
(A) azithromycin against Streptococcus pneumoniae ATCC6303; (B) azithromycin against Streptococcus pneumoniae ATCC 49619; (C) azithromycin against Moraxella catarrhalis ATCC 8176; (D) azithromycin against Haemophilus influenzae ATCC 10211. A monophasic activity profile is displayed over 6 hours. Reprinted with permission from (183).
42
Figure 2-3. Time-kill curves for Streptococcus pyogenes exposed to five antibiotics at
concentrations ranging from 0 to 64 times the MIC. A biphasic activity profile is displayed over 24 hours with the exception of vancomycin which displays a monophasic profile over 24 hours. Reprinted with permission from (136).
43
Figure 2-4. Effect of dose schedule on efficacy of ciprofloxacin against Bacillus anthracis. A
ciprofloxacin exposure of an AUC24 of 16 mg · h/liter (AUC24/MIC=256) was given as two equal doses at 12-h intervals or as a single dose at 24-h intervals. The twice-daily regimen prevented resistance emergence. The growth of an untreated control is also shown. Reprinted with permission from (58).
44
Table 2-1. Definition of each symbol used throughout the chapter Symbol Definition ε Maximum kill rate constant EC50 Concentration needed to produce half-the maximum effect C Concentration of antimicrobial H Hill factor N Number of bacteria λ Growth rate constant Nmax Maximum number of bacteria dg Delay in growth dk Delay in kill t Time NS Number of susceptible bacteria NR Number of persistent/resistant bacteria kSR Transfer rate constant of bacteria into the persistent/resistant stage kRS Transfer rate constant of bacteria into the susceptible stage Bmax Maximum number of viable bacteria in compartment 1 kdeath Natural bacterial death rate constant VGmax Maximum growth velocity MIC Minimum inhibitory concentration Nm Number of bacteria at which replication is half maximum SITm Median effect value of C/MIC (value at which the drug effect is half maximal) δ Maximum fraction increase of kdeath Cr Concentration which induces adaptive resistance IC50 Concentration of the adaptive resistance at which ε1 is half maximal z Delay factor ke Elimination rate constant tlag Initial lag time for adaptive resistance kecr Rate constant for the decrease in the concentration which induces adaptive resistance C0 Initial antibiotic concentration β Maximum adaptation factor Φ Adaptation factor
45
CHAPTER 3 CEFTOBIPROLE: A NOVEL CEPHALOSPORIN WITH ACTIVITY AGAINST GRAM-
POSITIVE AND GRAM-NEGATIVE PATHOGENS, INCLUDING MRSA1
Introduction
The development and spread of resistance to currently available antibiotics is a world-wide
concern. As resistance increases there is an overwhelmingly negative impact on mortality and
health care costs. For example, it is estimated that it costs $4-7 billion annually in the US to treat
resistant infections (187). Additionally, it has been shown that in intensive care patients with
blood stream infections, improper antibiotic selection results in a mortality rate approximately
double that of properly treated patients, 61.9% vs. 28.4% (87). Especially alarming is the high
percentage of methicillin-resistant Staphylococcus aureus (MRSA) isolates for a number of
reasons. One, this species is one of the most common pathogens in a wide-range of infections
including blood stream (91), pneumonia (77, 114), and skin and soft tissue (74). Two, resistance
has developed in this species to current last-defense antibiotics including vancomycin (170) and
linezolid (78). Three, MRSA has become a virulent pathogen in both the health-care setting and
the community. Therefore, there is definite medical need to continually develop new
antimicrobial agents, especially those with MRSA activity.
Ceftobiprole is a new broad-spectrum cephalosporin which displays a wide-range of
activity against both Gram-positive and Gram-negative pathogens including several resistant
species such as MRSA and penicillin-resistance Streptococcus pneumoniae (PRSP) (73, 82, 92,
155) (Table 3-1). It is approved in Canada under the trade name Zeftera with the indication of
1 This chapter is copyrighted as: Barbour A, et al. Ceftobiprole: a novel cephalosporin with activity against Gram-positive and Gram-negative pathogens, including methicillin-resistant Staphylococcus aureus (MRSA). In press. Int. J. Antimicrob. Agents (2009). Copyright © Elsevier.
46
complicated skin and skin structure infections (cSSSI) including non-limb threatening diabetic
foot infections without osteomyelitis (90). It is currently under review by the FDA in the United
States for approval.
Pharmacokinetics
Plasma Concentrations
The current generally recommended dosing regimen for ceftobiprole is a 500 mg dose
every eight hours as a two hour i.v. infusion. The concentration-time profile for a single dose
with this regimen is presented in Figure 3-1. A one hour infusion every twelve hours is
recommended in cases with Gram-positive infections without diabetic foot infections (90). The
major pharmacokinetic (PK) parameters for ceftobiprole are presented in Table 3-2 for the
recommended 500 mg dose and for the extensively studied 750 mg dose. The Cmax for
ceftobiprole with a 500 mg dose delivered intravenously (i.v.) over two hours is reported as 29.2
mg/L with an AUC0-8 of 90.0 mg*h/L (131). The AUC0-∞ for a 500 mg one hour infusion was
116 mg*hr/L. Ceftobiprole displays dose proportionality as demonstrated by the single rising
dose study as no significant difference was observed in Vss and CLs or dose-normalized AUC and
Cmax for doses between 125-1000 mg (164) (Figure 3-2). From a multiple dose study it was
concluded that accumulation is negligible as the accumulation on days 1 and 8 for a 500 mg and
750 mg dose were 1.06±0.18 and 1.04±0.09 with a twice daily dosing regimen (163). The
accumulation (Rss) for an eight hour dosing regimen should also be small based on a half-life of
3-4 hours and the formula, Rss=1/(1-e-ke*tau), where ke is the elimination rate constant and tau is
the dosing interval. For example, if the half-life were 4 hours the accumulation would be 1.33.
Finally, ceftobiprole displays a two-compartment body-model, which is demonstrated by a
biphasic log concentration-time profile (164).
47
Protein Binding
It has been shown for beta-lactams and other antimicrobials that only free drug is
responsible for the pharmacological activity (138, 194). Therefore, it is important to accurately
measure the plasma protein binding of each compound. In vitro experiments conducted in
human plasma with physiologically relevant concentrations of ceftobiprole, 0.5-100 mg/L, found
the protein binding to be linear and ~16% measured by ultrafiltration (129). This is half the in
vivo reported value of 38% (163). The protein binding was also determined in vivo in a
microdialysis tissue distribution study and ranged from 17-32% with an average of ~22% (15).
Volume of Distribution
As typical of most beta-lactams, ceftobiprole displays a volume of distribution
approximately equal extracellular fluid volume. This is due to the poor ability of beta-lactams to
cross membranes making them a good therapeutic option for extracellular infections. The Vdss
has been reported as 15.5-21.7 L for a 500 mg 2 hour i.v. infusion (129, 131). It is independent
of dose as shown by the single ascending dose and multiple dose studies where the volume of
distribution ranged from 16.1 L to 19.8 L for doses ranging from 125-1000 mg given as a 0.5 hr
i.v. infusion (163, 164).
Tissue Distribution
Microdialysis is a sampling technique which has proven suitable to determine free drug
concentrations within the interstitial space fluid (ISF) of the tissue of interest (31, 127, 172, 182).
The ability of ceftobiprole to penetrate into soft tissues was determined using microdialysis in
healthy volunteers after a single dose (15). In this study, two microdialysis probes were inserted
into a thigh of each of the twelve subjects, one into the subcutaneous (s.c.) adipose tissue and one
into the skeletal muscle. After each probe was calibrated by retrodialysis (173), each subject
received a single 2 hour i.v. infusion of 500 mg ceftobiprole. Ceftobiprole was found to
48
penetrate into the ISF both of subcutaneous adipose tissue and skeletal muscle as the ratios of the
area under the free concentration-time curve for tissue over plasma (fAUCtissue/fAUCplasma) was
0.49 and 0.69, respectively.
Using radiolabelled-ceftobiprole in mice and rats it was found that ceftobiprole distributes
rapidly into tissues with the highest tissue to plasma ratio being 1.3 for kidney (90). It was also
found it penetrates very little into the brain with a tissue to plasma ratio of 0.01 (90).
Clearance and Metabolism
In healthy volunteers administered 500 mg every 8 hours via 2 hr i.v. infusion, the amount
of drug eliminated unchanged in the urine was ≥83%, with <7% representing the open-ring
metabolite (131). At steady state the clearance of ceftobiprole, 4.98 L/hour for a 2 hour 500 mg
infusion, was approximately equal to the maximum glomerular filtration rate multiplied by the
fraction unbound suggesting clearance by glomerular filtration. In addition, the clearance of
ceftobiprole was highly correlated to creatinine clearance, also indicating that this drug is
eliminated by glomerular filtration (129). This was further supported by a probenecid study
which eliminated the possibility of active tubular secretion and subsequently the possibility of
drug-drug interactions through the pharmacokinetic changes in clearance (129). Since this drug
is predominately eliminated renally, drug interactions via the CYP enzymes are unlikely and
confirmed with in vitro experiments with human microsomes and hepatocytes with CYP
inducers and substrates (129).
Half-life
Ceftobiprole displays a half-life of ~3-4 hours, which is consistent across all doses tested
regardless of infusion time, with the exception of a reported half-life of 2.84 hours (P=0.055) for
a 250 mg dose administered over 0.5 hours (129, 131, 163, 164). However, as expected, the
half-life is significantly increased in patients with renal impairment with a reported half-life of
49
11 hours in severe renal impairment (90). Therefore, a dose adjustment is needed in patients
with moderate renal impairment, 500 mg every 12 hours as a 2 hour i.v. infusion, and severe
renal impairment, 250 mg every 12 hours as a 2 hour i.v. infusion (90).
Pharmacokinetics in Patients with Complicated Skin and Skin Structure Infections
It is also important to determine the PK in patients after the initial characterization in
healthy volunteers, as they are not necessarily the same. A phase II efficacy study was
conducted to compare PK parameters in patients with cSSSI caused by Gram-positive pathogens
and healthy volunteers each receiving a 30 minute i.v. infusion of 750 mg (129, 162). In
patients, the approximate steady state Cmax was 57.9±29.4 mg/L and the AUC0-∞ was
143±39.5mg*hr/L, which was comparable to the Cmax and AUC0-∞ in healthy volunteers of
60.6±9.99 mg/L and 165±12.8mg*hr/L, respectively.
Drug Administration
Ceftobiprole is administered as a 500 mg dose given via 2 hour i.v. infusion every 8 hours
or 1 hour i.v. infusion every 12 hours in cases with documented Gram-positive infections
excluding diabetic foot infection (90). The efficacy of these dosing regimens was demonstrated
in two large-scale trials (139, 140). It is delivered as the water soluble prodrug, ceftobiprole
medocaril, which is quickly and almost completely hydrolyzed in vivo to the active as can be
seen by negligible prodrug plasma concentrations almost immediately after administration. In
vitro experiments showed conversion to ceftobiprole is 38 seconds in human plasma (129). Type
A esterases are thought to be responsible for this conversion as shown by inhibition studies with
EDTA (129).
Ceftobiprole can be administered according to standard dosing recommendations to both
males and females and patients with hepatic impairment. In a study comparing gender
differences in pharmacokinetics, systematic exposure (AUC0–∞) was 15% higher in females than
50
males. However, after the dose was normalized on a mg/kg basis, there was no significant
difference between the exposure in males and females and no dose adjustment was needed (161).
In patients with hepatic impairment, no dose adjustment is needed since the vast majority of the
drug is eliminated renally. However, as previously mentioned patients with renal impairment
may need a dose adjustment depending on the severity.
Adverse Events
The most common adverse events (AEs) among healthy volunteers and patients are
presented in Table 3-3. The most common AE reported among healthy volunteers was
dysgeusia, or taste disturbance (163, 164). This is likely due to the formation of diacetyl as the
prodrug converts to the ceftobiprole (Figure 3-3) and is supported by the fact that it primarily
occurs during infusion as the rapid conversion is occurring. Additionally, it seems to be
concentration dependent as all healthy volunteers receiving multiple 750 mg doses experienced
this event. The following AEs not listed in the table occurred in one healthy volunteer; fatigue,
feeling hot, pharyngitis, somnolence, and abnormal urine. In healthy volunteers there were no
clinically significant changes from baseline in clinical labs.
In patients, the most common AE reported was nausea (139, 140). In this population
dysgeusia was only reported in 30 patients, however, it was not reported in >5% of patients in
the comparator study with ceftobiprole and vancomycin plus ceftazidime. This is somewhat
inconsistent as previous studies noted this as a common AE. In the two large-scale comparative
studies the incidence and severity of adverse events were similar in both arms, ceftobiprole vs.
vancomycin and ceftobiprole vs. vancomycin plus ceftazidime, except hypersensitivity which
occurred more in the vancomycin plus ceftazidime arm. Also a total of 3 deaths occurred
among the 932 patients in the ceftobiprole arms between the two studies but none were attributed
to the study medication.
51
Pharmacodynamics
In Vitro Activity
Ceftobiprole has demonstrated in vitro activity against a wide-range of Gram-positive and
Gram-negative pathogens. The MIC90 of common pathogens for cSSSI and pneumonia, a
possible second indication, are presented in Table 3-1. Especially noteworthy is the activity of
ceftobiprole against several resistant pathogens including MRSA, PRSP, and even
heteroresistant, vancomycin-intermediate Staphylococcus aureus (VISA), MICs ≤2mg/L (56).
The potency of ceftobiprole against these organisms lies within its inherent ability to bind to the
PBP’s which are associated with resistance in certain pathogens including PBP2x, PBP1a, and
PBP2b associated with resistance in Streptococcus pneumoniae (51) and PBP2a of
Staphylococcus aureus (50) and Staphylococcus epidermidis (82).
The broad spectrum of activity of ceftobiprole can also be attributed to the ability to
withstand hydrolysis by certain beta-lactamases, displaying an MIC of 0.25 mg/L against
Staphylococcus aureus with PC1 beta-lactamase, 0.25 mg/L and 0.5 mg/L against Klebseilla
pneumoniae with TEM-26 and SHV-1, respectively, and 0.5 mg/L against Escherichia coli with
TEM-1 (150). However, it should be noted that ceftobiprole is susceptible to hydrolysis by some
beta-lactameses; e.g. ESBL CTX-M-15, carbapenemases VIM-2 and KPC-2, and some AmpC of
Pseudomonas aeruginosa (150). Additionally, ceftobiprole does not bind sufficiently to PBP5 of
Enterococcus faecium (82) which explains its poor activity against this pathogen, MIC90 of >32
mg/L (92).
Also noteworthy, ceftobiprole is bacterialcidal as demonstrated by an MBC/MIC ratio ≤4
with a majority of resistant pathogens including MRSA, hVISA, and PRSP (56). It was also
found to be bactericidal against Haemophilus influenzae at concentration of 2XMIC, MIC≤2
mg/L, and several Gram-positive cocci at a concentration of 4 mg/L including Streptococcus
52
pneumoniae, viridans group streptococci, Staphylococcus aureus, coagulase-negative
staphylococci, and Enterococcus faecalis in time-kill studies (22, 57).
Resistance Studies
Once drugs hit the market and wide-spread use has occurred there is an opportunity for
resistance to develop. Therefore, it is important to evaluate the possibility of resistance
development in order to make a recommendation for drug usage. This has been done with
ceftobiprole using serial passage in subMIC concentrations. The potential for resistance
development was low as after 50 passages; the highest MIC increase was 4XMIC (22, 23). This
only occurred in one VISA vancomycin-intermediate Staphylococcus aureus strain of the ten
Staphylococcus aureus strains tested (23) and in only one Haemophilus infuenzae stain of the
eights strains tested (22). This was similar to previous findings using Staphylococcus aureus
strains, three methicillin-resistant and one susceptible, which found only a 2-fold increase in the
MIC after six passages (82). Additionally, no resistance was observed in vivo when lung
homogenates from pneumococcal mice treated with ceftobiprole after infection with PRSP were
plated for MIC comparison after treatment (12). The same was found with MRSA and VISA
when homogenized tissues where plated and MICs determined from the aortic valve, spleen, and
kidney after induced endocarditis (36).
In Vivo-Animal Studies
The in vivo pharmacodynamics of ceftobiprole have been characterized in several animal
models including mouse pneumococcal pneumonia, mouse Enterococcus faecalis peritonitis,
mouse septicemia, mouse subcutaneous (s.c.) abscesses, and rabbit Staphylococcus aureus aortic
valve endocarditis. In support of the lead indication of cSSSI, ceftobiprole demonstrated good
potency against MRSA and VISA in the mouse s.c. abscesses model (82). In this study, doses
were administered 1 and 3 hours after infection by s.c. injection of the inoculum under the loose
53
skin of the groin. The animals were killed three days post infection to determine the
CFU/abscess and ceftobiprole proved to be more potent than competitors with a 5.12 log
decrease in CFU compared to control animals versus a decrease of 3.42 and 0.80 log CFU for
vancomycin and linezolid against MRSA compared to controls. With VISA, ceftobiprole
displayed a decrease of >4.79 CFU compared to a decrease of 0.13 and 0.64 log CFU for
vancomycin and linezolid, respectively.
For the potential indication of pneumonia it is also important to evaluate the
pharmacodynamics of ceftobiprole in pneumococcal pneumonia (12). A study was conducted
with neutropenic Swiss albino mice infected with 107 CFU into the lower respiratory tract.
While there were no significant differences between ceftobiprole and ceftriaxone in this survival
study, the doses used for ceftobiprole, 51 and 75 mg/kg/day, were much lower than the doses
needed of ceftriaxone, 200 and 400 mg/kg/day, to produce a comparable survival rate against
P40984, a ceftriaxone-resistant PRSP strain. The doses needed for ceftobiprole were less with
P52181, a penicillin-cefotaxime-cetriaxone susceptible strain, 2.1-4.2 mg/kg/day, to produce a
similar survival rate compared to ceftriaxone, 10-20 mg/kg/day.
Additionally, ceftobiprole has shown good activity against a range of bacteria in a
septicemia model. In Swiss albino mice infected with intraperitoneal injection, doses were given
at 1 and 3 or 1, 3, and 5 hr after infection depending on the bacteria used. With MRSA,
ceftobiprole was more potent, ED50 2.4 mg/kg, than cefepime, ceftriaxone, meropenem, all with
ED50s >25mg/kg, and vancomycin, ED50 6.7mg/kg (82). Ceftobiprole was also effective against
several PRSP strains with ED50 0.6-1.1 mg/kg (82). Similarly, in mice infected with four strains
of Enterococcus faecalis, including two vancomycin-resistant strains, by intraperitoneal injection
the 50% protective dose was lower for ceftobiprole compared to ampicillin in all strains and was
54
significantly lower in the beta-lactamase producing strain HH22(TX0921), 2mg/kg vs.
>100mg/kg (10).
The in vivo pharmacodynamics for ceftobiprole have also been explored in the rabbit
endocarditis model with MRSA and VISA, showing it may have potential to treat deep-seated
infections. With MRSA-76, ceftobiprole and vancomycin showed a significant reduction of
CFU/g in the aortic valve vegetation, spleen, and kidney compared to control animals. In
endocarditis by VISA, animals treated with ceftobiprole demonstrated a significantly lower
CFU/g than both the control animals and vancomycin treated animals in aortic valve vegetation,
spleen, and kidney (36).
In Vivo-Pivotal Studies
Ceftobiprole was evaluated for the treatment of cSSSI caused by Gram-positive bacteria
and compared to vancomycin in a large-scale study (140). This trial involved 282 patients
receiving ceftobiprole 500 mg twice daily and 272 patients receiving vancomycin 1 g twice
daily. There was no significant difference between the cure rates for ceftobiprole, 93.3%,
compared to vancomycin, 93.5%. Interestingly, in this study it was found that in Panton-
Valentine leukocidin (PVL)-positive MRSA, ceftobiprole had a higher cure rate, 93.1%, than
vancomycin, 84.6%. This is clinically relevant because the PVL genotype is associated with the
more virulent community acquired-MRSA and maybe the cause behind increased ambulatory
visits (83).
The efficacy of ceftobiprole was also compared to vancomycin plus ceftazidime for
treatment of both Gram-positive and Gram-negative cSSSI (139). This study enrolled patients
2:1 into the ceftobiprole arm so that a total of 485 patients and 244 were clinically evaluable for
ceftobiprole and vancomycin plus ceftazidime, respectively. The clinical cure rates were 90.5%
for ceftobiprole and 90.2% for vancomycin plus ceftazidime. It was also noted in this study that
55
in the subgroup of diabetic patients with foot infection, ceftobiprole was just as efficacious as the
competitor, 86.2% and 81.8% clinical cure rate. Additionally, ceftobiprole was just as effective
in severe infections; severity was based on C-reactive protein, whether the infections involved
fascia or muscle, and PVL-MRSA infection.
Pharmacokinetics/Pharmacodynamics (PK/PD)
PK/PD Characterization
The PK/PD indices used for the characterization of antibiotics have been well established
and most antibiotics are frequently placed into one of two categories (42). In one category, the
time the free antibiotic concentration remains above a threshold value, the MIC, is usually
indicative of efficacy. In the other category, efficacy is usually correlated to a ratio of exposure,
Cmax or AUC, and the MIC. The efficacy of beta-lactams, including ceftobiprole, in usually
correlated to the fT>MIC and this relationship between the dosing regimen and effect is usually
found by performing dose fractionation studies. The PK/PD relationship of ceftobiprole was
characterized in the neutropenic mouse thigh infection model with several Gram-positive and
Gram-negative strains (9). Twenty-four hours post infection, the animals were euthanized and
the thighs were homogenized so that viable bacteria could be determined. For MSSA and
MRSA the fT>MIC needed for stasis was 14.4-27.8% of the dosing interval. This is slightly lower
than the recommended 40-50% to achieve bacterial stasis with gram-positive organisms (42).
For Gram-negative organisms in this experiment the fT>MIC was longer and ranged from 41.2-
58.6%, which is close to the usually recommended breakpoint for Gram-negatives, fT>MIC of
≥40% of the dosing interval (30).
PK/PD for Dose Selection
Once the PK/PD indices were established this information was used to optimize the dosing
regimen. Dose selection was done using Monte Carlo simulation in two studies. The first study
56
used data from 12 healthy volunteers, 6 receiving 500 mg and 6 receiving 750 mg doses (120).
The population PK parameters were calculated and based on a target of fT>MIC>40%, MCS were
performed and a conservative dosing recommendation was made of 750 mg twice daily. A
second study was conducted using a much larger and more varied population to find the PK
parameters, which included 150 subjects, 29 of which were from a phase II trial (107). In this
study 2 regimens were examined 500 mg every 8 hours (0.5, 1, 2 hour infusions) and 500 mg
every 12 hours (1 hour infusion). This study also examined the impact of renal function based
on a creatinine clearance distribution from previous studies which used a similar population.
The recommendation was made that in patients with cSSSI the dosing regimen should be 500 mg
every 12 hours as a 1 hour i.v. infusion. This dosing regimen was selected for a pivotal
comparative study with vancomycin against Gram-positive bacteria (140). The final dosing
recommendation for nosocomial pneumonia was 500 mg every 8 hours delivered as a 2 hour
infusion.
The appropriateness of this dose was examined in a microdialysis study which determined
that free drug concentrations in the ISF of s.c adipose tissue and skeletal muscle remained above
the MIC, for an MIC of 2 mg/L, for greater than 40% of the 8 hour dosing interval. For an MIC
of 4 mg/L the fT>MIC was approximately 72% in plasma, 54% in skeletal muscle, and 35% in s.c.
adipose tissue for an eight hour dosing regimen (15).
Summary
The pharmacokinetics of ceftobiprole are similar to other beta-lactams in that it displays a
two-compartment body-model, distributes into extracellular fluid, does not cross membranes
well, and is primarily cleared renally. It also displays a safety profile similar to other beta-
lactams, with the exception of dysgeusia being a common AE. Additionally, the
pharmacokinetics and clearance mechanisms leave the potential for drug-interactions low and
57
dose adjustment in special populations, i.e. renally impaired, noncomplex. The potential for
resistance development is low and therefore ceftobiprole is a good first-line therapeutic option
against a variety of infections. This compound is not active against all pathogens and is subject
to hydrolysis by some ESBLs and carbapenemases. However, as current therapeutic options
dwindle with resistance development, ceftobiprole is a promising new cephalosporin with a
wide-range of activity, including MRSA.
58
Table 3-1. In vitro activity of Ceftobiprole against common skin and soft tissue and respiratory pathogens (73, 82, 92, 155)
MIC90 MIC range Staphylococcus aureus Methicillin-Susceptible 0.5-1 ≤0.12-2 Methicillin-Resistant 1-4 0.12-4 Methicillin Susceptible Coagulase Negative 0.25 ≤0.012-1 Methicillin-Resistant Coagulase Negative 1-2 ≤0.012-8 Streptococcus.pneumoniae Penicillin-Susceptible ≤0.015 ≤0.012-0.3 Penicillin-Resistant 0.25-2 ≤0.012-4 Viridans group streptococci penicillin-resistant
0.25-1 ≤0.012-32
Beta-hemolytic streptococci ≤0.015-0.06 ≤0.012-0.25 Enterococcus faecalis 2-4 0.12->32 Enterococcus faecium >32 0.25->32 Haemophilus influenzae ampicillin-susceptible
≤0.06-1 ≤0.015-1
Moraxella catarrhalis 0.12-1 <0.008-1 Escherichia coli 0.06-0.12 ≤0.015-2 ESBL E. coli >32 0.03->32 Klebsiella pneumoniae 0.06->8 ≤0.015-16 ESBL K. pneumoniae >32 ≤0.015->32 Acinetobacter spp. >32 ≤0.015->32 P. aeruginosa ceftazidime susceptible 8-16 0.12-16
59
Table 3-2. Pharmacokinetics following single and multiple doses presented as mean (standard deviation) (129, 131, 162-164)
500 mg Dose SD 0.5hour infusion
SS q12 hr 0.5hr infusion
SD 1hr infusion
SD 2hr infusion
SS q8hr 2hr infusion
N=11 N=6 N=18 N=28 N=27 Cmax 38.3(7.1) 44.2(10.8) 34.2(6.1) 29.2(5.5) 33.0(4.8) AUC0-∞ 89.9(6.7) 108(22.2) 116(20.2) 104(13.9) NC t1/2 3.54(0.40) 4.04(0.31) 2.85(0.55) 3.10(0.30) 3.3(0.3) CLs 5.69(0.41) 5.05(0.95) 4.46(0.84) 4.89(0.69) 4.98(0.58) Vss 17.9(2.0) 16.7(3.6) 11.0(2.9) 21.7(3.3) 15.5(2.3) 750 mg Dose SD 0.5hr infusion SS q12hr 0.5hr
infusion SS q12hr 0.5hr patients
N=30 N=6 N=22-25 Cmax 58.8(10.0) 60.6(10.0) 57.9(29.4)* AUC0-∞ 151(20.9) 165(12.8) 165(39.5) T1/2 3.16(0.43) 4.11(0.41) 3.06(0.62) CLs 5.08(0.71) 4.84(0.34) 5.99(1.74) Vss 15.2(2.7) 16.1(2.2) 20.1(5.2) * The infusion time in patients ranged from 22-90 minutes. SD=single dose. SS=steady state.
60
Table 3-3. Most common adverse events in healthy volunteers and patients (139, 140, 163, 164) Healthy Volunteers Patients Total Number of Subjects 42 932 Total Number of Subjects with at least one AE
21 507
Total Number of Subjects with at least one serious AEs
0 63
Dysgeusia 17 30 Nausea 10 113 Headache 10 68 Abdominal Pain 2 - Vomiting 1 61 Diarrhea 1 62 Constipation 0 33 Hypersensitivity* 0 49 Infusion Site Reaction 1 48 *including rash and pruritus. – not specifically listed.
61
Figure 3-1. Mean concentration-time profile for a 500 mg two hour i.v. infusion in healthy
volunteers (n=12). Error bars represent standard deviation. Modified from (15).
62
Dose (mg)
0 200 400 600 800 1000 1200
Vss
(L)
5
10
15
20
25
30
Dose (mg)
0 200 400 600 800 1000 1200
CLs
(L/h
r)
0
2
4
6
8
10
Dose (mg)
0 200 400 600 800 1000 1200
Dos
e N
orm
aliz
ed A
UC
(mg*
hr/L
)
0
10
20
30
40
Dose (mg)
0 200 400 600 800 1000 1200
Dos
e N
orm
aliz
ed C
max
(mg/
L)
0
2
4
6
8
10
12
14
Figure 3-2. A-Mean volume of distribution at steady state (Vss) vs. dose. B-Mean systemic clearance (CLs) vs. dose. C-Mean dose normalized AUC0-∞ vs. dose. D-Mean maximum concentration (Cmax) vs. Dose. Dose normalized to the 125 mg dose. Error bars represent standard deviation (164).
63
Figure 3-3. Formation of diacetly (2,3-butanedione) during the conversion of ceftobiprole
medocaril (top) to the active moiety ceftobiprole (bottom) (164).
64
CHAPTER 4 IN VITRO MICRODIALYSIS OF CEFTOBIPROLE
Objective
The aim of this study was to determine the ability of ceftobiprole to cross the microdialysis
(MD) membrane using two dose ranging, in vitro microdialysis experiments. The two
experiments performed used the extraction efficiency (EE) and retrodialysis (RD) methods. In
both of these methods the recovery percent (R%) was determined. The R% had to be more than
10% if microdialysis was to be used for sampling ceftobiprole from soft tissues.
Chemicals and Equipment
Test Article
Ceftobiprole (BAL9141) was provided by Johnson and Johnson. The compound was
stored in the original shipping box and vial in the -70°C freezer.
Ceftobiprole-BAL9141: Provided by Johnson & Johnson Pharmaceutical Research and
Development
920 U.S. Route 202 P.O. Box 300 Raritan, NJ 08869
Reagents
All reagents were of HPLC grade unless otherwise stated. They were obtained from the
indicated sources or equivalent suppliers.
• Double distilled water Filtered in house by Corning AG-3 • Acetonitrile Fisher Scientific A998-4 • Lactated Ringer’s solution Baxter 2B2323 • DMSO Fisher Scientific BP231-1 • Formic Acid Fisher Scientific A118P-100
Equipment
The following pieces of equipment or equivalent substitutes were used.
65
• Balance Mettler AE240 • Vortex Kraft Apparatus Inc. Model PV-5 • Micropipettes Eppendorf Research • Pipette tips (1-200 µL) Fisherbrand 22-707-500 • Pipette tips (100-1000 µL) Fisherbrand 21-197-8F • Autosampler vials Sun Sri 200 046 • Aluminum seals Sun Sri 200 100 • Sonicator Fisher Scientific FS110H • 15mL centrifuge tubes Corning 430052 • HPLC Agilent 1100 Series • Column Supelco C18 Discovery 4.6mm, 15 cm, 5,6µm • Analytical software Agilent Technologies-Agilent ChemStation • Syringes Becton Dickinson 309603 • Microcentrifuge Tubes Fisherbrand 05-408-129 • Syringe Pump Harvard Apparatus Model 55-4150 • Heated Stir Plate Fisherbrand Isotemp • Thermometer Fisherbrand 76mm Immersion 14-997 • Guard Column C18 2cm 30-40 micron (made in house) • Microdialysis Probes CMA-60 P000002
Reagent Preparation
Ceftobiprole Standards and Quality Control Solutions
BAL9141 was provided as pure powder. 10 mg were weighed out and dissolved in 10 mL
of 0.1 % formic acid in DMSO. The 1000 µg/mL stock solution was diluted to a secondary
stock of 100 µg/mL with lactated Ringer’s solution. From here the working standards were
prepared by serial dilution with lactated Ringer’s solution. The standard curve ranged from 100-
0.049 µg/mL. Quality controls (QCs) of 0.098, 3.13, and 50 µg/mL were used for the low, mid,
and high QCs, respectively. The standards and quality controls were prepared using the same
dilution scheme (Table 4-1).
HPLC Mobile Phase
950 mL of double distilled water was mixed with 50 mL of acetonitrile and degassed by
sonication.
66
0.1% Formic Acid in DMSO
0.114 mL of 88% formic acid was added to a 100 mL volumetric flask and filled with
DMSO.
Sample Preparation
Calibration Solutions for Microdialysis
Six calibration solutions of BAL9141 (25, 12.5, 6.25, 3.13, 1.56, 0.78μg/mL) were
prepared. This was done by preparing a 1mg/mL stock solution of BAL9141 in 0.1% formic
acid in DMSO. Then a 100μg/mL solution was prepared using lactated Ringer’s solution. From
here serial dilutions were performed with lactated Ringer’s solution to make all other solutions.
The concentrations selected were chosen because the expected maximum concentration in soft
tissues was ~30μg/mL.
Dialysate Samples
During the sampling period approximately 30µL of sample were collected. The sample
was divided into two equal aliquots of 12µL and then each aliquot was diluted with 24µL of
lactated Ringer’s solution to obtain a sufficient sample volume for HPLC analysis. The samples
taken from the calibration tube and/or the syringe were also diluted 2:1 with lactated Ringer’s
solution. Therefore, to get the true concentrations the values obtained were multiplied by three.
Apparatus Setup
Before the experiments were started each MD probe was checked for functionality. To do
so, the inlet of the MD probe was connected to a syringe containing blank lactated Ringer’s
solution. The probe was flushed manually. The probe was functional and ready to use when no
liquid drops appeared on the MD probe membrane. The solution should only exit the probe from
the outlet tubing. If droplets appear on the membrane, the probe was not used.
67
To control temperature the sampling device was assembled on a heated stir plate and was
maintained at 37oC. During setup, a 5mL syringe was filled either with blank lactated Ringer’s
solution (EE method) or analyte solution (RD method) and the enclosed air was cleared from the
syringe. The syringe was put in place on the syringe pump and fastened. The pump was then
connected to the inlet of the probe and run at a flow rate of 1.5μL/min for 15min to allow for
equilibration. The probe itself was in a 15mL centrifuge tube containing either blank lactated
Ringer’s solution (RD method) or analyte solution (EE method). Attention was paid so that the
membrane of the probe is completely covered with fluid and that it did not touch the wall of the
tube. The dialysate was collected in a microcentrifuge tube covered with parafilm, which helped
to fix the outlet tubing from the microdialysis probe in place and avoid evaporation during
sampling.
Sample Analysis
HPLC/UV Set Up
The HPLC system was the Agilent 1100 series which included a DAD detector
(G1315B), an autosampler (G1329A), a column oven (G1316A), a degasser (G1379A), and a
quaternary pump (G1311A). The work station for data retention was an hp Compaq p4. The use
of a guard column was optional.
HPLC Pump Conditions
• Mobile Phase: 95% Distilled Water, 5% Acetonitrile • Flow rate: 1.0 mL/min • Retention Time: ~3.5 min
Autosampler Conditions
• Injection Volume: 20µl • Run time: 8 min • 1 injection per vial (blanks may be injected more than once) • 1 vial per sample
68
• Temperature: 4ºC
Detector Conditions
• Wavelength: 300nm
Analysis Procedure
• The test samples were run with a calibration curve and 3 sets of quality control standards at 3 levels (low, medium, and high) placed throughout the run.
• A blank injection of lactated Ringer’s solution and a blank injection of 0.1% formic acid in DMSO were included.
Microdialysis
Introduction
Microdialysis (MD) is an extremely useful tool for antibiotic sampling because the
samples are taken directly from soft tissues, the most common site of infection, and because only
the free, pharmacologically active drug is able to pass through the probe membrane. In this
technique, the microdialysis probe is placed into the tissue of interest and continuously perfused
with a physiological solution (perfusate). Based on diffusion, the drug passes from the tissue
into the probe and is collected (dialysate). Ideally, an absolute equilibrium between the tissue
and perfusate will be established. In reality, due to the fact that the MD probe is perfused at a
constant flow rate of 1.5µL/min, an absolute equilibrium will not be reached. The ability of the
drug to pass through the membrane and establish equilibrium at this flow rate will be established.
It is termed the recovery (R) and this value has to be known to back-calculate the actual
concentration at the sampling site, Ctissue, from the concentration in the dialysate, Cdialysate. This
is done by using the following equation:
1%100 −××= RCC dialysatetissue
In microdialysis, the sampling time is determined by the flow rate. A higher flow rate
would lead to a shorter sampling time since there is a minimum volume requirement. However,
69
the recovery is decreased because there is not enough time for equilibration between the solution
inside the probe and the surrounding media to occur. Therefore, a compromise has to be made
between sample volume and flow rate. Depending on the sensitivity of the assay, the sample
volume should not be smaller than 20μL. Therefore, at a flow rate of 1.5μL/min and a sampling
time of 20min, 30µL are collected.
The equilibration period is determined by the dead volume of the MD probe tubing which
can be calculated from the manufacturer’s specifications. Since this experiment will be
performed using six different concentrations, all of the tubing has to be completely flushed
before the sampling procedure can start. The equilibration time is the product of dead volume
multiplied by flow rate.
Extraction Efficiency Method (EE)
In the EE method, blank lactated Ringer’s solution was pumped through the MD probe at
a flow rate of 1.5μL/min. The MD probe was then placed into the calibration tube containing
analyte solution, starting with the lowest concentration. Drug diffused from the calibration tube
into the MD probe, and the dialysate was collected for analysis. Each sample was collected for
20min after the end of a 15min equilibration period. To ensure that the prepared solution was the
concentration expected within the calibration tube, two samples were taken from this tube, one
before the sampling period and one after. It was important to sample from the calibration tube
since it was critical to know the actual concentration in the tube to perform the calculations.
Additionally, the two samples can be compared to see if the concentration within the calibration
tube was consistent throughout the sampling period. The same procedure was done for the
remaining five samples. After the highest concentration was completed the probe was flushed for
one hour with blank lactated Ringer’s solution. All experiments were performed in triplicate.
The percent recovery, R%, for the EE method was calculated as follows:
70
perfusate
dialysate
CC
R ×= 100%
Retrodialysis Method (RD)
In the RD method, the syringe contained the analyte solution that was pumped through
the MD probe at a flow rate of 1.5μL/min. The MD probe was placed into a calibration tube that
was filled with approximately 10mL of blank lactated Ringer’s solution. The analyte diffused out
of the probe into the calibration tube. The loss of analyte through the membrane was then
determined from the Cdialysate. Samples were taken for 20min after the end of the 15min
equilibration period. To ensure that the prepared solution was the concentration expected within
the syringe, a sample was taken from the syringe after the sampling period. Additionally, two
samples were taken from the calibration tube, one before and one after the sampling period, to
see if the concentration within the tube was consistent throughout sampling. Again, the lowest
concentration was sampled first. In this method, the calibration tube, which contains a small
amount of analyte after the sampling period, had to be exchanged with a new tube containing
fresh blank lactated Ringer’s solution. The same procedure was done for the remaining five
samples. After the highest concentration was sampled the probe was flushed for one hour with
blank lactated Ringer’s solution. All experiments were performed in triplicate. The percent
recovery, R%, for the RD method was calculated as follows:
⎟⎟⎠
⎞⎜⎜⎝
⎛×−=
perfusate
dialysate
CC
R 100100%
Results
Method Validation Results
A calibration curve of 0.049-100 µg/mL was used. Quality controls of 0.098 µg/mL,
3.13 µg/mL, and 50 µg/mL for low, mid, and high respectively, in triplicate were incorporated.
71
This method provided reasonable accuracy and precision (Table 4-2). The analysis was
performed with weighted linear regression (1/X2). The slope for the standards was 15.39 and the
intercept was -0.03. Also, the R2 value was 0.999996. The retention time of ceftobiprole was
approximately 3.5 minutes.
In vitro Microdialysis Results
In this experiment two in vitro microdialysis methods were used, the extraction efficiency
(EE) and the retrodialysis (RD) methods. Both methods are acceptable to characterize how the
compound interacts with the MD membrane and if the compound can freely pass through the
membrane. These in vitro experiments were done as a preliminary study to an in vivo
experiment. There are some fundamental differences in the setup of both experiments which can
explain the difference in R%. The retrodialysis method is the same method which will be used in
the in vivo experiment for MD probe calibration. See Table 4-3 for results.
Conclusions
This experiment confirmed that ceftobiprole has the ability to freely cross the
microdialysis membrane. The R% was well over 10% and therefore, microdialysis may be used
as a sampling technique to obtain a PK profile.
72
Table 4-1. Standard preparation of BAL9141 Starting Concentration (µg/mL)
Volume (mL) Final Concentration (µg/mL)
Total Volume (mL)
1000 0.2 100 2 100 1 50 2 50 1 25 2 25 1 12.5 2 12.5 1 6.25 2 6.25 1 3.13 2 3.13 1 1.56 2 1.56 1 0.78 2 0.78 1 0.39 2 0.39 1 0.20 2 0.20 1 0.098 2 0.098 1 0.049 2
73
Table 4-2. Method validation accuracy and precision Quality Controls
Conc Observed Conc
Mean Observed
SD CV% Accuracy%
Mean Accuracy%
A 0.098 0.088 0.090 0.01 7.43 89.80 91.50 A 0.098 0.097 99.00 A 0.098 0.084 85.71 B 3.13 3.10 3.07 0.22 7.24 99.04 98.00 B 3.13 3.27 104.47 B 3.13 2.83 90.42 C 50 52.29 52.44 0.96 1.84 104.58 104.88 C 50 53.47 106.94 C 50 51.56 103.12 Conc=Concentration (µg/mL)
74
Table 4-3. In vitro microdialysis results Concentration (µg/mL)
EE% SD R% SD
0.78 91.08 23.72 72.72 28.76 1.56 86.38 15.25 79.62 5.37 3.13 86.44 14.75 83.00 5.90 6.25 89.06 8.29 81.25 2.70 12.5 86.50 12.16 80.70 2.39 25 89.21 7.57 77.06 5.44 All samples mean (SD)
88.25 (12.15) 79.06 (11.03)
75
CHAPTER 5 HPLC/UV METHOD VALIDATION OF CEFTOBIPROLE IN LACTATED RINGER’S
SOLUTION
Objective
The objective of this series of experiments was to validate an HPLC/UV method for the
determination of ceftobiprole in lactated Ringer’s solution using good laboratory practices
(GLP). The assay was designed to be used in an in vivo microdialysis study of ceftobiprole in
humans. The assay was validated after in vitro studies deemed ceftobiprole suitable for use with
the microdialysis sampling technique.
Validation Procedure
One standard curve and three of each QC concentration were assayed on each analytical
day for three days to find the lower limit of quantification (LOQ) and linear range using
inter/intra-batch accuracy and precision. The acceptance criteria were defined as accuracy
between 80-120% at the (LOQ) and 85-115% for all other standards and QCs. Precision was
defined as relative standard deviation and must be <20% at the LOQ and <15% at all other
concentrations. The accuracy% was calculated as follows:
100*][][%
ExpectedMeasuredAccuracy =
The following parameters were also investigated during the validation: freeze/thaw
stability of quality control samples and stock standard solution, refrigeration stability of quality
control samples, room temperature stability of quality control samples and stock standard
solution, robustness of the method, and assay selectivity. For a set of samples to be considered
acceptable two-thirds of samples must meet acceptance criteria.
76
Results
Reproducibility of the Calibration Curve Parameters
The 3 calibration curves generated during the determination of inter- and intra-day
precision and accuracy were linear with an average slope of 16.815 ± 0.6688. Consistently good
correlation coefficients (r2>0.9999) were obtained in these experiments. A 1/X2 weighting
scheme was used. This option provided a similar correlation coefficient to the linear model and a
y-intercept closer to 0.
Lower Limit of Quantification
The limit of quantification (LOQ) was identified as 0.1 µg/ml. It was defined as the lowest
concentration of quality controls used in this validation for which the accuracy% was between
80-120%.
Intra-batch Variability for Quality Control Samples
Intra-batch evaluation of the analyte was performed on three different days. Intra-batch
variability on day 1 was 3.78%, 2.57%, 0.345%, and 0.569% for the 0.1, 0.2, 20, and 40 µg/ml
quality control samples, respectively (n=3, Day 1). Accuracy for day 1 of all samples ranged
from 101.1% to 110.3% (Day 1, Table 5-1 for averages). Intra-batch variability on day 2 was
10.7%, 5.07%, 1.91% and 1.66% for the 0.1, 0.2, 20, and 40 µg/ml quality control samples,
respectively (n=3, Day 2). Accuracy for day 2 of all samples ranged from 90.94% to 114.4%
(Day 2, Table 5-1 for averages). Intra-batch variability on day 3 was 9.09%, 3.33%, 1.39%, and
0.894% for the 0.1, 0.2, 20, and 40 µg/ml quality control samples, respectively (n=3, Day 3).
Accuracy for day 3 of all samples ranged from 88.49% to 106.1% (Day 3, Table 5-1 for
averages). No QC sample fell out of range.
77
Inter-batch Variability for Quality Control Samples
Inter-batch variability was 8.06%, 7.07%, 2.89%, and 2.75% for the 0.1, 0.2, 20, and 40
µg/ml quality control samples, respectively (Table 5-2). Mean accuracy was between 101.7%
and 104.3% (Table 5-2). The accuracy range based on individual QCs is already stated above.
Freeze-thaw Stability
Three cycles of freeze/thawing were performed on the LOQ, low, mid, and high quality
controls (0.1, 0.2, 20, and 40 µg/ml, n=3) between 24May07 and 25May07. The QCs were also
injected with the standard curve before freezing to ensure correct preparation. The following
time table was followed:
• Freeze 1: 5:15 pm 24May07 • Thaw 1: 8:26 pm 24May07 • Freeze 2: 9:31 pm 24May07 • Thaw 2: 11:30 am 25May07 • Freeze 3: 12:37 pm 25May07 • Thaw 3: 2:47 pm 25May07
The samples were stable for two freeze/thaw cycles (Table 5-3). One of the 0.2 µg/ml QC
samples fell out of acceptance criteria after the third freeze/thaw cycle with an accuracy of
80.881%. The ranges for accuracy are 94.99%-111.0%, 86.51%-103.3%, and 80.32%-94.89%
for the 1st, 2nd, and 3rd, freeze/thaw cycles respectively. Therefore, it is concluded that the QCs
are stable for 2 freeze thaw cycles.
The freeze/thaw stability of the stock standard solution was also tested due to the fact that
ceftobiprole is diluted in a different matrix (0.1% formic acid in DMSO) than the QC samples. It
is important that ceftobiprole is stable in this DMSO solution because the stock standard cannot
be prepared fresh daily due to the limited amount of sample. To test its freeze/thaw stability,
standard curves were compared and analyzed as quality control samples. The standard curve that
was prepared fresh from the powdered compound on 21May07 is used as the true standard curve
78
and all subsequent standard samples are analyzed as QCs. The standard curves prepared on
24May07, 28May07, 29May07, 04Jun07, and 11Jun07, and 13Jun07 were analyzed.
Additionally, the stock standard solution was thawed on 08Jun07 but the samples were not
analyzed due to poor chromatography. This results in a total of 7 cycles over 23 days and shows
that the stock standard was stable over this time (Table 5-4). The inter-batch accuracy of all
samples individually ranged from 88.41%-118.4% (this sample was at the LOQ). All individual
samples were within acceptance criteria. Only accuracy was compared to validate the stock
standard stability. If degradation were occurring the accuracy% would be out of acceptance
criteria.
Stability at Room Temperature
The stability of ceftobiprole stock standard solution and QCs were tested at room
temperature. To do so a standard curve and set of QCs were prepared. Then after a defined
length of time a second set of QCs were prepared from the stock standard left out on the bench-
top, and injected onto the column. The concentrations of these QC samples were determined
using the standard curve injected previously. The stock standard was shown to be stable for 5.5
hours at room temperature (Table 5-5). The accuracy of each individual QC sample ranged from
89.86%-108.51%.
To test the stability of ceftobiprole in lactated Ringer’s solution a standard curve and a set
of QCs were prepared. Then an aliquot of each of the QCs was injected with the standard curve
while the remaining portion was left at room temperature. At a defined length of time, another
aliquot was taken from the portion at room temperature and loaded into the autosampler. It
should be mentioned that degradation at room temperature is much greater than degradation at
4°C, the temperature of the autosampler. The quality controls were shown to be stable for 3
79
hours at room temperature (Table 5-6). The accuracy of each individual QC sample ranged from
89.15%-104.2%.
Refrigeration Stability of QCs
To test the stability of QC samples at 4ºC the set of QC prepared on 21May07 was
reinjected 24 and 48 hours later. The standard curve from 21May07 was used to calculate the
concentration of the QCs after storage. The accuracy of each individual QC sample after 24
hours ranged from 91.57% to 104.31%. The accuracy of each individual QC sample after 48
hours ranged from 96.89% to 113.63% (Table 5-7).
Freezer (long-term) Stability
Long-term freezer stability has been demonstrated for 32 days of ceftobiprole in lactated
Ringer’s solution. An aliquot of each sample prepared on 27Sep07 was frozen at -70ºC and
analyzed on 29Oct07. The standard curve from 27Sep07 was used for analysis. The QC-2-0.2
µg/mL sample was originally out of range. A confirmation injection was done on 30Oct07 and
the sample was within acceptance criteria. This value was used in the calculations for the
summary table (Table 5-8).
Robustness
To test the ability of this method to endure changes a different column was used to test
accuracy and precision. Column ID 89216-02 was used instead of column ID 86248-02, which
was used for most other validation experiments. The intra-batch precision was 7.61%, 7.34%,
2.12%, and 1.19% for the 0.1, 0.2, 20, and 40 µg/ml quality control samples, respectively (n=3)
(Table 5-9). Accuracy ranged from 96.64%-113.1% for individual samples.
80
Chemicals and Equipment
Test Article
Ceftobiprole (BAL9141) was provided by Johnson and Johnson. The compound was
stored in the original shipping box and vial in the -70°C freezer.
Johnson & Johnson Pharmaceutical Research and Development
920 U.S. Route 202 P.O. Box 300 Raritan, NJ 08869 Reagents
All reagents were of HPLC grade unless otherwise stated. They were obtained from the
indicated sources or equivalent suppliers.
• Double distilled water Filtered in house by Corning AG-3 • Acetonitrile Fisher Scientific A998-4 • Lactated Ringer’s solution Baxter 2B2323 • DMSO (Dimethylsulfoxide) Fisher Scientific BP231-1 • Formic Acid Fisher Scientific A119P-10
Equipment and Disposables
• Balance Mettler AE240 • Vortex Kraft Apparatus Inc. Model PV-5 • Micropipettes Eppendorf Research • Pipette tips (1-200 µL) Fisherbrand 22-707-500 • Pipette tips (100-1000 µL) Fisherbrand 21-197-8F • Autosampler vials Sun Sri 200 046 • Aluminum seals Sun Sri 200 100 • Sonicator Fisher Scientific FS110H • Centrifuge tubes (15 mL) Corning 430052 • Microcentrifuge tubes Fisherbrand 05-408-129 • HPLC System Agilent 1100 Series • Work Station hp Compaq p4 • Column Supelco C18 Discovery 4.6mm, 15 cm, 5,6µm • Analytical software Agilent Technologies-Agilent ChemStation • Guard Column C18 2cm 30-40 micron (made in house) • Manual Crimpers Fisher 03-375-7 • Volumetric flask Pyrex 5640 • Microsyringes Tyco Healthcare 8881501400
81
• Plastic Cannula BD 303345 • Graduated Cylinder Pyrex 2982
Reagent Preparation
Ceftobiprole Stock Standard Solution
A 1 mg/mL stock standard solution of ceftobiprole was prepared by weighing out 10 mg of
ceftobiprole and dissolving it in 10 mL 0.1% formic acid in DMSO. This was done by adding
the ceftobiprole to a 10 mL volumetric flask and filling with the solvent. A secondary stock of
100 µg/mL was prepared using a microsyringe and adding 0.1 mL of stock standard solution to
0.9 mL of lactated Ringer’s solution. All standard solutions were prepared fresh daily, except
the stock standard solution. The 1mg/mL stock is stored at -70ºC until needed.
Ceftobiprole Working Standard Solutions
After the 100 µg/mL secondary stock solution was prepared, the working standards were
prepared with serial dilutions using lactated Ringer’s solution (Table 5-10). The highest
concentration in the standard calibration curve was 50 µg/mL.
Ceftobiprole QC Samples
Three sets of quality controls (QC) were prepared, each set contained a limit of
quantification, low, mid, and high QC. The high QC should be between 75-90% of the highest
standard. The mid QC should be 40-50% of the highest standard and the low QC should be no
more that 2x the limit of quantitation (LOQ). Additionally, there should be one QC at the limit
of quantification. From the stock solution of 1 mg/mL a solution of 100 µg/mL was prepared.
This was done using a microsyringe. From here dilutions were made to get the desired
concentrations for the QCs (Table 5-11). These dilutions were performed using a pneumatic
pipette. The starting volume and total volume could vary as long as the ratio remained the same
82
so that the same final concentrations are obtained. All quality control solutions should be
prepared fresh daily.
In this dilution scheme (Table 5-11) the concentrations of 40, 20, 0.2, and 0.1µg/mL were
used as the high, mid, low, and LOQ QCs respectively, as the standard curve concentration range
was 50-0.098 µg/mL.
HPLC Mobile Phase
950 mL of distilled water were mixed with 50 mL of acetonitrile. The mobile phase was
degassed. The final volume of mobile phase prepared could vary as long as the ratio of distilled
water to acetonitrile remained the same, 95:5.
0.1% Formic Acid in DMSO
0.1 mL of 90% formic acid was added to a 100 mL volumetric flask and filled with
DMSO.
HPLC Wash Solution
500 mL of double distilled water was mixed with 500 mL acetonitrile. The wash solution
was degassed. The final volume of mobile phase prepared could vary as long as the ratio of
distilled water to acetonitrile remained the same.
Sample Preparation
All samples that have lactated Ringer’s solution as the matrix were directly injected onto
the HPLC column. However, samples obtained from microdialysis, both in vivo and in vitro,
needed to be diluted with lactated Ringer’s solution to obtain a sufficient sample volume for
HPLC analysis. This was done by dividing the sample into two equal aliquots of 12 µL each and
adding 24 µL of lactated Ringer’s solution. By dividing the sample a back-up was available.
The sample results found from the analysis needed to be multiplied by three to get the true
concentration.
83
Assay Procedure
HPLC/UV Set Up
The HPLC system used was the Agilent 1100 series which included a DAD detector
(G1315B), an autosampler (G1329A), a column oven (G1316A), a degasser (G1379A), and a
quaternary pump (G1311A). The work station used for data retention was an hp Compaq p4.
See Figure 5-1 for a typical chromatogram for ceftobiprole.
HPLC Pump Conditions
• Mobile phase: 95% Distilled Water, 5% Acetonitrile • Flow rate: 1.0 mL/min • Retention time: ~3.5 minutes
Autosampler Conditions
• Injection Volume: 20 µL • Run time: 8 minutes • 1 injection per vial (blanks may be injected more than once) • 1 vial per sample • Temperature: 4ºC
Detector Conditions
• Wavelength: 300nm
Analysis Procedure
• Each batch of test samples was run with a calibration curve. Additionally, each run contained 3 sets of quality control standards at 4 levels (LOQ, low, medium, and high) placed randomly throughout the run.
• With each run a blank injection of lactated Ringer’s solution was made at the start to ensure system equilibration.
• Blank lactated Ringer’s solution injections were made periodically throughout run. • After every 20 injections the column underwent a brief wash. This involved a blank lactated
Ringer’s solution injection with the wash solution running for 15 minutes as the mobile phase. Next the column was allowed to equilibrate by making another blank lactated Ringer’s solution injection and running the original mobile phase for 15 minutes.
84
System Suitability
Three injections of the 50 µg/mL standard were made. The CV% for these injections was
less than 5%. The symmetry was ≥0.8 and ≤1.2. The retention time for ceftobiprole was 3.6min
± 0.4 min.
Summary
An HPLC-UV method for the analysis of ceftobiprole in lactated Ringer’s solution was
developed for the analysis of microdialysis samples. The calibration curve range, 0.1-50 mg/L,
is appropriate based on the expected in vivo concentrations. The compound is stable under the
conditions which will be encountered during an in vivo clinical study.
85
Table 5-1. Intra-batch variability of quality control samples (n=3 per day and per concentration) QC Concentration (µg/ml)
0.1 0.2 20 40
Day 1 (17May07) Mean 0.11 0.21 20.3 40.7 SD 0.00403 0.00548 0.0699 0.232 CV (%) 3.78 2.58 0.345 0.569 Accuracy (%) 106.6 106.3 101.4 101.8 Day 2 (18May07) Mean 0.10 0.22 21.5 43.0 SD 0.0107 0.0111 0.408 0.715 CV (%) 10.7 5.07 1.90 1.66 Accuracy (%) 100.2 109.1 107.3 107.4 Day 3 (21May07) Mean 0.10 0.19 20.5 41.4* SD 0.00892 0.00634 0.285 0.370* CV (%) 9.09 3.34 1.39 0.894 Accuracy (%) 98.14 95.03 102.4 104.5 Data was generated on 17May07, 18May07, and 21May07. *Based on n=2.
86
Table 5-2. Inter-batch variability of quality control samples (n=9 per concentration) Days 1-3 0.1 0.2 20 40* (17,18, 21 May 2007) Mean 0.10 0.21 20.7 41.73 SD 0.00820 0.0146 0.599 1.15 CV (%) 8.06 7.07 2.89 2.75 Accuracy (%) 101.7 103.5 103.7 104.3 Data was generated on 17May07, 18May07, and 21May07. *Based on n=8 due to sample oading error. l
87
Table 5-3. Freeze/thaw stability of QCs Concentration (µg/ml) 0.1 0.2 20 40 Cycle 1 Mean 0.10 0.20 19.3 39.0 SD 0.00688 0.00631 0.305 0.412 CV (%) 6.60 3.19 1.58 1.06 Accuracy (%) 104.3 98.97 96.70 97.45 Cycle 2 Mean 0.10 0.18 18.66 38.23 SD 0.00327 0.00378 0.677 0.653 CV (%) 3.30 2.14 3.63 1.71 Accuracy (%) 99.63 98.53 93.31 95.58 Cycle 3 Mean 0.09 0.17 17.71 37.14 SD 0.00461 0.00887 0.195 0.724 CV (%) 5.39 5.16 1.10 1.95 Accuracy (%) 85.47 85.83 88.56 92.85 D
ata was generated on 24May07 and 25May07.
88
Table 5-4. Freeze/thaw stability of ceftobiprole stock standard solution Expected 0.1 0.2 0.39 0.78 1.56 3.13 6.25 12.5 25 50 Mean 0.11 0.20 0.38 0.75 1.50 3.01 6.08 12.26 24.76 49.73 SD 0.0402 0.0787 0.156 0.310 0.621 1.24 2.48 4.98 9.84 1.58 CV (%) 37.7 39.1 41.2 41.4 41.4 41.1 40.9 40.6 39.7 3.18 Accuracy (%)
109.1 100.5 97.1 96.1 96.1 96.2 97.2 98.1 99.1 99.5
Data was generated on 24May07, 28May07, 29May07, 04Jun07, and 11Jun07, and 13Jun07.
89
Table 5-5. Room temperature stability of ceftobiprole stock standard solution after 5.5 hours (n=3)
Concentration (µg/mL) 0.1* 0.2 20 40 Mean 0.10 0.19 19.1 39.8 SD 0.0149 0.0111 0.993 2.03 CV (%) 15.3 5.96 5.20 5.11 Accuracy (%) 97.94 93.39 95.50 99.50 Data was generated on 04Jun07. *Based on n=2 due to poor chromatography of one QC sample.
90
Table 5-6. Room temperature stability of ceftobiprole QCs after 3 hour Concentration (µg/mL) 0.1* 0.2 20 40 Mean 0.10 0.19 19.00 39.81 SD 0.00364 0.00795 0.126 0.387 CV (%) 3.62 4.27 0.665 0.971 Accuracy (%) 100.6 93.06 95.00 99.53 Data was generated on 14Jun07. *Based on n=2 due to poor chromatography of corresponding QC sample which was injected with the standard curve
91
Table 5-7. Refrigeration (4ºC) stability of QC samples over 24 and 48 hours (n=3) Concentration (µg/ml) 0.1 0.2 20 40* 24 hours Mean 0.10 0.20 20.27 41.11 SD 0.0018 0.0127 0.379 0.213 CV (%) 1.78 6.51 1.87 0.562 Accuracy (%) 101.4 97.9 101.3 102.8 48 hours Mean 0.11 0.21 20.01 40.98 SD 0.0002 0.0168 0.371 0.653 CV (%) 0.185 7.98 1.85 1.14 Accuracy (%) 112.9 105.0 100.1 102.5 Data generated on 22May07 and 23May07 using the standard curve prepared on 21May07. *n=2 due to sample loading error
92
Table 5-8. Freezer stability of QC samples over 32 days (n=3) Concentration (µg/mL) 0.1 0.2 20 40 Mean 0.10 0.19 19.99 38.73 SD 0.015 0.0091 0.28 3.75 CV (%) 14.7 4.78 1.41 9.67 Accuracy (%) 100.4 95.50 99.96 96.84 Data generated on 29Oct07 using the standard curve from 27Sep07 for 32 day stability.
93
Table 5-9. Interbatch variability of ceftobiprole using a different column to test robustness (n=3) Concentration (µg/mL) 0.1 0.2 20 40 Mean 0.10 0.21 20.27 40.81 SD 0.00795 0.0154 0.429 0.486 CV (%) 7.61 7.33 2.12 1.19 Accuracy (%) 104.4 105.3 101.3 102.0 Data generated on 28May07.
94
Table 5-10. Standard preparation of ceftobiprole Starting Concentration (µg/mL)
Starting Volume (mL) Final Concentration (µg/mL)
Total Volume (mL)
100 1 50 2 50 1 25 2 25 1 12.5 2 12.5 1 6.25 2 6.25 1 3.13 2 3.13 1 1.56 2 1.56 1 0.78 2 0.78 1 0.39 2 0.39 1 0.20 2 0.20 1 0.098 2
95
Table 5-11. Quality control sample preparation of ceftobiprole Starting Concentration (µg/mL)
Starting Volume (mL) Final Concentration (µg/mL)
Total Volume (mL)
100 1.6 80 2 80 1 40 2 40 1 20 2 20 0.8 8 2 8 0.8 3.2 2 3.2 1 1.6 2 1.6 1 0.8 2 0.8 1 0.4 2 0.4 1 0.2 2 0.2 1 0.1 2 0.1 1 0.05 2
96
Figure 5-1. Typical chromatogram for ceftobiprole. Compounds showing visible absorption at 300 nm are shown. The X-axis is in minutes.
0 1 2 3 4 5 6 7
mAU
0
20
40
60
80
100
120
g
( )
1.73
11.
814
2.05
12.
279
2.37
1
3.48
8
5.64
75.
949
0 1 2 3 4 5 6 7
mAU
0
20
40
60
80
100
120
g ( )
1.73
11.
814
2.05
12.
279
2.37
1
3.48
8
5.64
75.
949
0 1 2 3 4 5 6 7
mAU
0
20
40
60
80
100
120
g ( )
1.73
11.
814
2.05
12.
279
2.37
1
3.48
8
5.64
75.
949
97
CHAPTER 6 SOFT TISSUE PENETRATION OF CEFTOBIPROLE IN HEALTHY VOLUNTEERS
DETERMINED BY IN VIVO MICRODIALYSIS2
Introduction
Ceftobiprole, the active compound in ceftobiprole medocaril, is a promising new
cephalosporin with activity against both Gram-negative and Gram-positive bacteria. This
includes good activity against methicillin-resistant Staphylococcus aureus (MRSA) with reported
MIC90 values of 2 mg/L (92) and 4 mg/L (82) and penicillin-resistant Streptococcus pneumoniae
(PRSP) with an MIC90 of 2 mg/L (82). Currently, ceftobiprole is under regulatory review and is
only approved in Canada and Switzerland. Two dosing regimens are recommended, 500 mg as a
2 hour i.v. infusion every 8 hours in cases with gram-positive and/or gram-negative infections
including diabetic foot infections and 500 mg as a 1 hour i.v. infusion every 12 hours in cases of
documented Gram-positive infections only excluding diabetic foot infections (90). This study
focuses on the more general dosing regimen, i.e. the 2 hour 500 mg infusion every 8 hours, and
references to the dosing interval refer to eight hours. This prolonged infusion time, 2 hours, was
chosen in an attempt to prolong the time the concentration remains above the MIC.
Traditionally, plasma samples have been taken to determine the pharmacokinetic (PK)
properties of a compound and make efficacy predications based on
pharmacokinetic/pharmacodynamic relationships. However, these concentrations are sometimes
presented as total concentrations while only the free drug is pharmacologically active (138, 194).
Also, free concentrations at the site of action/infection are much more relevant to determine
efficacy (121, 179) and exploring the concentration at the site of action has been recommended
2 This chapter is copyrighted as: Barbour A, et al. Soft tissue penetration of ceftobiprole in healthy volunteers determined by in vivo microdialysis. In press. Antimicrob Agents Chemother (2009). Copyright © American Society for Microbiology.
98
by regulatory agencies (34, 37, 66). One technique that has proven useful for the measurement of
free drug concentrations in the interstitial space fluid (ISF) of subcutaneous (s.c.) adipose tissue
and skeletal muscle is microdialysis (52, 53, 85, 103, 126, 158).
The aim of this study is to examine the penetration of ceftobiprole from plasma into the
ISF of s.c. adipose tissue and skeletal muscle using microdialysis and determine if efficacy
breakpoints of relevant pathogens are met.
Materials and Methods
This study was performed in accordance with the Declaration of Helsinki and Good
Clinical Practice guidelines. Prior to study initiation approval was received from the local ethics
committee, the Institutional Review Board (IRB) at UF/Shands. All volunteers were consented
verbally and in writing prior to participation in the study and written consent was obtained.
Healthy Volunteers
This study included 15 healthy volunteers (9 males and 6 females) between the ages of 20
and 34. Health was determined based on physical examination, medical history, vital signs, 12-
lead ECG, clinical laboratory tests (serum chemistry, hematology, and urinalysis), BMI, negative
hepatitis B surface antigen, negative hepatitis C antibodies, negative human immunodeficiency
antibodies, and normal renal function based on serum creatinine and the Cockcroft-Gault
equation. Subjects also had a negative urine drug test and alcohol breath test at screening and
admission and were nonsmokers. Females in the study had a negative pregnancy test at
screening and admission and were postmenopausal, sterile, abstinent, or practicing an effective
measure of birth control. Additionally, subjects did not use any other medication one week prior
to the study drug administration until after the last blood draw with the exceptions of
acetaminophen for pain, hormonal contraceptive medications, or hormonal replacement drugs.
99
Microdialysis
Microdialysis is a sampling technique which is based on simple diffusion of free analyte
through the semipermeable membrane at the tip of a microdialysis probe. This technique has
been explained in detail previously (172). Once in the ISF of the tissue of interest the flexible
probe is continuously perfused with a physiological solution at a low flow rate by use of a
syringe pump. The equilibrium between the probe and the tissue is incomplete and therefore the
probes must be calibrated once in place, typically by retrodialysis (173). In this technique, a low
concentration of the analyte is perfused through the microdialysis probe and the disappearance
into the tissue is measured from the dialysate. This recovery value is later used to calculate the
true tissue ISF concentration. The recovery value is calculated as recovery%=100-
(Cdialysate/Cperfusate *100); where Cperfusate is the analyte concentration flowing into the probe and
Cdialysate is the concentration of the analyte leaving the probe.
Study Design
Pilot Study
Pilot study subjects were admitted to the General Clinical Research Center (GCRC) at
Shands Hospital the morning of the study and underwent only the calibration procedures. After
the site of probe insertion was cleaned and disinfected, two microdialysis probes (CMA 60,
CMA Microdialysis AB, Solna, Sweden) were implanted without anesthesia into the same thigh
by a study physician, one into the skeletal muscle and one into the s.c. adipose tissue. The
probes were perfused with lactated Ringer’s solution for 30 minutes at a flow rate of 1.5 µL/min.
The pilot study subjects then received ceftobiprole via the microdialysis probes at a
concentration of 200 mg/L and a flow rate of 1.5 µL/min for 60 min. A sample was collected
during the last 30 min to determine the individual recovery% values. The probes were then
100
flushed with blank lactated Ringer’s solution for three hours and samples were collected every
30 min.
Main Study
Subjects in the main study (6 male, 6 female) were admitted to the GCRC the evening before
dosing. On the morning of dosing, two microdialysis probes were implanted and each was
calibrated via retrodialysis as described above. After a four hour washout period, subjects
received a single 666.7mg dose of ceftobiprole medocaril (Johnson & Johnson Pharmaceutical
Research and Development, Raritan, NJ, USA), corresponding to 500 mg ceftobiprole, as a two
hour i.v. infusion. The microdialysis probes were perfused with lactated Ringer’s solution at a
flow rate of 1.5 µl/min by a syringe pump from the start of the washout phase until after the 24 h
sample was collected. Dialysate samples were collected in 20 min intervals from pre-dose
through 12 hours after the start of the infusion and at 16 and 24 hours. Blood samples for PK
determinations were collected at pre-dose, 40 min, 1 h, 1 h 40 min, 2 h, 2 h 20 min, 3 h, 4 h, 6 h,
8 h, 12 h, 16 h, and 24 h. Blood samples for protein binding determination were collected at 2 h
and 12 h. Blood sampling occurred from the opposite arm that the drug was administered and
K2EDTA tubes were used for collection.
Analysis Methods
Sample Analysis-Microdialysis Samples
The dialysate samples were stored at -80ºC until analysis at the University of Florida,
Department of Pharmaceutics (Gainesville, FL) and analyzed using a validated HPLC-UV
method. The limit of quantification (LOQ) for this method was 0.1 µg/mL. An Agilent 1100
series HPLC with an UV detector was used with a reverse phase column (Supelco C18
Discovery). The mobile phase consisted of water and acetonitrile (95:5). The flow rate was 1
mL/min. The injection volume was 20 µL and the detection wavelength was 300 nm. The true
101
tissue ISF concentrations were calculated from the dialysate concentrations and adjusted with the
recovery value. The calculation was as follows: CISFtissue=100*Cdialysate*recovery%-1.
Sample Analysis-Plasma Samples
Plasma samples were stored at -80ºC until shipment with dry ice to SFBC Analytical
Laboratories, North Wales, PA. They were analyzed there using a validated LC/MS/MS method
for ceftobiprole with K2EDTA and citric acid. The LOQ was 0.050 μg/mL. A Perkin-Elmer 200
HPLC autosampler (4ºC) and pump were used. A gradient elution with two mobile phases (A-
1:5:95 formic acid/methanol/water and B-1:50:50 formic acid/methanol/water) on a reverse
phase column (Synergi 4μ Polar-RP, 50 x 2 mm) was used. The tandem mass spectrometer (PE
Sciex API 4000 Series) with a turbo ionspray interface was set to monitor ceftobiprole at m/z
535-308 and m/z 539-312 for the internal standard (ceftobiprole-d4). Plasma samples were
prepared by adding 50μL of sample to 50 μL of internal standard, mixing, then adding 200μL of
0.1% formic acid in acetonitrile and 300μL of 10% perchloric acid. The samples were then
vortexed for one minute and centrifuged at 3000rpm for 15 minutes. Protein binding was
determined by ultrafiltration.
Data Analysis
Total plasma concentrations were adjusted based on individual protein binding prior to
PK analysis to determine the free plasma concentrations. PK analysis was performed using
commercially available software (WinNonlin 5.2, Pharsight Corporation, Mountain View, CA,
USA) by noncompartmental analysis. The AUC0-last was calculated using the linear trapezoidal
rule. The AUC0-∞ was calculated as AUC0-last + Clast/λz. The elimination rate constant, λz, was
calculated using linear regression of the concentration-time data. The points chosen for
calculation of λz were based on the best fit of the terminal phase and visual inspection. The
102
AUCs were compared using Wilcoxon’s matched paired tests. A two-sided P<0.05 was
considered significant.
Results
The aim of this study was to assess the tissue penetration of ceftobiprole into the ISF of
skeletal muscle and subcutaneous adipose tissue. A pilot study was first conducted to test the
feasibility of using microdialysis with ceftobiprole and to determine the washout period needed
to ensure that no drug was remaining in the tissues and the microdialysis system at the time of
dosing.
The mean recovery values (±SD) in the pilot study were 64.1%±12.6 and 47.5%±3.3 for
s.c. adipose and muscle tissue respectively. In the pilot study, one probe malfunctioned after
insertion and therefore the mean recovery value for the muscle is calculated from two samples.
From the pilot study it was determined that the recovery was high enough to continue to the main
study and a four-hour washout period should be allotted after calibration, prior to dosing.
The mean protein binding between all subjects was 21.7%±6.6. The mean recovery
values in muscle and s.c. adipose tissue for the main study were 58.3%±5.1 and 59.4%±5.4,
respectively, and the measured concentrations were adjusted accordingly. The mean
concentration-time profiles for plasma, free plasma, free skeletal muscle ISF, and free s.c.
adipose tissue ISF are presented in Figure 5-1. The pharmacokinetic parameters are summarized
in Table 5-1. The mean fAUC0-∞ (±SD) ratios of tissue ISF compared to plasma were 0.69±0.13
and 0.49±0.28 for skeletal muscle and s.c. adipose tissue, respectively. The results show that
there is a significant difference between the fAUC in plasma and the AUCs of both soft tissues.
Additionally, the AUC of skeletal muscle is significantly higher than the AUC of s.c. adipose
tissue.
103
Discussion
This study shows that ceftobiprole distributes into the interstitial space fluid (ISF) of s.c.
adipose tissue, fAUCs.c.adipose/fAUCplasma 0.49±0.28, and skeletal muscle, fAUCmuscle/fAUCplasma
0.69±0.13. The degree of tissue penetration with ceftobiprole correlates well other
cephalosporins such as cepodoxime (103), cefixime (103), cefpirome (85, 158), and cefaclor (52)
(Table 6-2). There was a significant difference between the fAUC of plasma and the fAUC of
both tissues and between the fAUC of muscle and the fAUC of s.c. adipose tissue. The
differences in penetration ratios can be due to several factors including the perfusion of the
particular tissue, local capillary density (52), the degree of tissue binding, the possibility of
active transporters (53), loss of drug from the peripheral compartments (94), and physiochemical
properties of the compound (53), such as lipophilicity. Therefore, it is important to measure the
free, active drug in each tissue and not make the assumption that free plasma levels equal free
tissue ISF levels, even in well perfused tissues.
The pharmacokinetics determined in this study are in good agreement with previously
summarized results (129). The half-life, clearance, Cmax, and AUC of 2.61 hr, 5.15 L/hr, 25.8
mg/L, and 98.0 mg/L*hr, respectively, were all within one standard deviation of the previously
reported parameters with the same dosing regimen (129, 130). The Vss of 14.6 L is lower in this
study than the reported value of 21.7 L for a single dose. However, it is in agreement with the
multiple dose Vss of 15.5 L. The volume of distribution suggests that this compound distributes
to the ISF which is common with this antibiotic class. This property is advantageous because the
ISF is often the location of infectious pathogens.
The major advantage to the microdialysis technique is the ability to measure the free drug
at the site of action, usually the ISF of soft tissues in regards to skin and skin structure infections.
It is this concentration that should be used to measure if efficacy breakpoints are met. For β-
104
lactams the time the free concentration remains above the minimum inhibitory concentration
(fT>MIC) is thought to predict efficacy. It has been shown in neutropenic animals that efficacy is
established if the concentration remains above the MIC for at least 40-50% of the dosing interval
(42, 43, 46), i.e. 3.2-4 h of an 8 h dosing interval. The MIC90 for ceftobiprole against MRSA (92)
and PRSP (82) has been reported as 2 mg/L. The concentrations in the ISF of both skeletal
muscle and s.c. adipose tissue remained above 2 mg/L for at least 50% of the dosing interval and,
therefore, this dosing regimen should be efficacious with these subcutaneous soft tissue
pathogens. Also, sufficient concentrations are achieved to meet the efficacy breakpoint in
organisms with an MIC90 of 4 mg/L in skeletal muscle, 53.6%±9.62 %fT>MIC. In s.c. adipose
tissue, the time the concentration remains above the MIC is close to 40% of an 8 h dosing
interval, 35.1%±22.2 %fT>MIC, and it has been suggested that the time free concentration needs
to remain above the MIC is less than 40% for Staphylococcus aureus and Streptococcus
pneumoniae (46). It is important to remember that further studies should be conducted in
patients to see the relationship between the host response, free ceftobiprole concentration at the
site of action, and clinical outcome. Additionally, ceftobiprole has demonstrated similar clinical
cure rates compared to vancomycin in Gram-positive complicated skin and skin structure
infections and vancomycin plus ceftazidime in both Gram-positive and Gram-negative
complicated skin and skin structure infections in large-scale pivotal studies (139, 140). This
study demonstrates that ceftobiprole distributes into the ISF of soft tissues in healthy volunteers.
This finding and ceftobiprole’s wide-range of activity make it a promising new single agent for
the treatment of complicated skin and skin structure infections.
105
Figure 6-1. Mean ceftobiprole concentration in plasma (circles), skeletal muscle ISF (squares), and s.c. adipose tissue ISF (triangles) over twelve hours. Free plasma concentration (dashed line) was calculated based on the plasma protein binding of each individual patient.
106
Table 6-1. Noncompartmental pharmacokinetic analysis PK parameter Plasma (total) Plasma (free) S.C. Adipose Muscle Cmax (mg/L) 25.8±2.96 20.2±2.63 9.61±4.74 14.0±3.22 Tmax (mg/L) 1.92±0.15 ND 2.25±0.21 2.25±0.14 t1/2 (hr) 2.61±0.33 ND 2.56±0.39 2.61±0.52 AUC0-last (hr*mg/L) 97.1±10.3 76.0±8.81 34.3±19.0 50.6±10.9 AUC0-∞ (hr*mg/L) 98.0±10.5 76.9±8.94 36.5±19.4 53.2±11.5 CL (L/hr) 5.15±0.53 ND ND ND Vz (L) 19.4±3.61 ND ND ND Vss (L) 14.6±2.17 ND ND ND AUCISF/fAUCplasma - - 0.49±0.28 0.69±0.13 ND: Not Determined
107
Table 6-2. Soft tissue penetration of four cephalosporins determined by microdialysis Cefpodoximea Cefiximea Cefpiromeb Cefpiromec Cefpiromed Cefaclore Cefaclorf Cefaclorg
AUCplasma 22.4 25.6 - - - 17.6 13.7 22.1
fAUCplasma 17.7 9.0 275.0 230.7 175.0 13.2 10.3 16.6
fAUCmuscle 15.4 7.3 - 130.0 80.0 9.49 7.02 11.53
fAUCs.c.adipose - - 218.5 117.0 87.2 - - -
fAUCmuscle/ fAUCplasma
0.89 0.84 - 0.56 0.46 0.73 0.67 0.70
fAUCs.c.adipose/ fAUCplasma
- - 0.79 0.51 0.50 - - -
All AUCs are in mg/L*hr. Data presented as mean values from each study. a. Dosed orally 400 mg. AUC0-∞ presented. fAUC calculated based on protein binding (103). b. Dosed i.v. over 15 minutes 2 g. AUC0-4 presented. Data from healthy volunteers only (158). c. Dosed i.v. over 10 minutes 2g. AUC0-8 presented (85). d. Dosed i.v. over 12 hours 2g (only 10 hours observed). AUC0-8 presented (85). e. Dosed orally 500mg IR. AUC0-∞ presented. fAUC calculated based on protein binding (52). f. Dosed orally 500mg MR. AUC0-∞ presented. fAUC calculated based on protein binding (52). g. Dosed orally 750mg MR. AUC0-∞ presented. fAUC calculated based on protein binding (52).
108
CHAPTER 7 PHARMACODYNAMC PROPERTIES OF CEFTOBIPROLE AGAINST MRSA
Introduction
Antimicrobial pharmacodynamics are typically defined by the minimum inhibitory
concentration (MIC). The MIC is the lowest concentration that completely inhibits visible
growth of the organism after 18–24 h of incubation at 35°C with a starting inoculum of
approximately 5x105 colony forming units (CFU)/mL. In turn, the clinical target for
antimicrobial agents are normally evaluated on the basis of one of the following PK/PD indices;
the time free drug concentration in plasma is above the MIC (fT>MIC) expressed as a percentage
of the dosing interval, the ratio of maximum concentration to MIC (Cmax/MIC), and the ratio of
the area under the twenty-four hour concentration-time curve to MIC (AUC0-24/MIC). There are
some major limitations to these MIC based PK/PD indices. For example, the MIC does not show
the time course of the antibiotic’s effect and it has a two-fold variability. Another approach that
has been successfully used to study the PD of antibiotics is time-kill analysis. Time-kill curves
display the change in the number of bacteria over time by using a range of static or dynamic
concentrations. Once this experiment is performed a mathematical model can be developed to
determine the PD parameters. Beta-lactams typically fit an Emax model (5). The PD parameters
derived from time-kill experiments with this model can then be combined with in vivo PK data
in an integrated PK/PD model that describes the antibiotic’s activity as a function of time and
concentration.
Ceftobiprole is currently under FDA review with the primary indication of complicated
skin and skin structure infections. This compound is delivered as a water-soluble prodrug,
ceftobiprole medocaril. Ceftobiprole displays a wide-range of activity against both Gram-
109
positive and Gram-negative pathogens including several resistant species such as MRSA and
penicillin-resistance Streptococcus pneumoniae (PRSP) (73, 82, 92, 155).
It is the goal of this series of experiments to perform time-kill experiments to characterize
the activity of ceftobiprole against methicillin-resistant Staphylococcus aureus (ATCC33591). A
modified Emax model will be developed and validated to determine the pharmacodynamic
parameters.
Materials and Methods
Study Conduct
The MICs for this study were determined in compliance to the Clinical and Laboratory
Standards Institute (CLSI) approved standard methods using the macrodilution study design.
Time-kill curves also followed CLIS methods but the complete design of these experiments is
not yet standardized. The methods used in this experiment followed those of a previous
publication and are described below (183).
Preparation of Ceftobiprole
A ceftobiprole stock solution, 1.28 mg/mL was prepared by weighing 12.8 mg of
ceftobiprole (Johnson and Johnson Pharmaceutical Research and Development) and dissolving it
in 10 mL of 0.1% formic acid in DMSO. The stability of ceftobiprole in this stock 1 solution has
been shown to be at least 23 days at -70ºC with at least 7 freeze/thaw cycles. From this stock 1
solution, further dilutions for the MIC determination and time-kill curves were made using
Mueller-Hinton Broth (MHB).
Preparation of Sterile Broth and Normal Saline
Mueller-Hinton broth (Becton Dickinson BBL) was prepared according to the
manufacturer’s instructions and autoclaved prior to use for 10 minutes at 121°C. Normal saline
110
was prepared by weighing out 9g of sodium chloride and dissolving it in 1L double distilled
water. It was then autoclaved for 20 minutes at 121 °C.
Inoculum Preparation
The inoculum was prepared from colonies incubated overnight on sheep’s blood agar
plates (Remel). The microorganisms were suspended in sterile normal saline to a concentration
equivalent to a 0.5 value in the McFarland Equivalence Turbidity Standard (Remel Microbiology
Products) as measured with a turbidimeter which equals 1x108 CFU/mL. For the MIC 0.01 mL
of the initial inoculum were added to each well containing 2mL of solution for a final initial
inoculum of 5x105 CFU/mL. For time-kill curves, 200 μL of the suspension were inoculated
into the in vitro model to achieve a final initial inoculum of approximately 1X106 CFU/mL. The
culture flasks in the time-kill experiments were incubated for 2 hours to allow the bacteria to
reach the logarithmic growth phase prior to the addition of ceftobiprole.
MIC Determination
The MIC was determined using two-fold dilutions in 24-well plates with an inoculum of
approximately 5X105 CFU/mL and antibiotic concentrations started at 0.06 mg/L and ranging to
32 mg/L. The plates were read after 20 hours of incubation at 37ºC, and the MIC was defined as
the lowest concentration of the antibiotic allowing no visible growth. Determinations of the MIC
were performed six times (3 plates with 2 readings per plate). Additionally, each MIC
determination contained a positive control (only bacteria, no antibiotic) and a negative control
(no bacteria or antibiotic) to validate the results.
In Vitro Infection Model with Constant Antibiotic Concentration
This model was used to investigate the effect of constant concentrations of ceftobiprole
against MRSA as a function of time. The in vitro infection model consisted of a 50-mL canted
neck, vented cap, tissue culture flask containing 20 mL of the MHB. The bacteria were exposed
111
to a series of concentrations of ceftobiprole based on the MIC (0.25-16XMIC) for 24 hours.
Samples were taken at predefined time points as defined below.
Bacterial Quantification
Samples (20 μl) were collected from the in vitro model at the following time points: zero,
0.5, 1, 1.5, 2, 3, 4, 6, 8, 12, 16, 20, and 24 hours. Bacterial counts were determined by plating
serial dilutions of the samples on appropriate agar plates. The dilutions (10X) are made in sterile
normal saline. Aliquots (50 μl in five 10 µL drops) of each dilution were plated in duplicate and
an average was taken. Sheep’s blood agar plates were used (Remel Blood Agar (TSA w/Sheep
Blood) plates). The plates were incubated at 37°C overnight before reading. The procedure was
repeated at least 3 times per concentration. Positive controls (with bacteria, no drug) are run
simultaneously in order to assess the method. Following incubation, colonies are counted on all
readable plates that showed up to approximately 250 colonies/50μL sample, 50 colonies/10μL
droplet. The LOQ of this method was deemed as 3000 CFU/mL, or 15 CFU/50μL sample. This
LOQ was chosen so that at least one 10X dilution was made to avoid residual antimicrobial
activity after sampling. The data was entered into Microsoft Excel spreadsheets. Data points
below the LOQ were excluded.
Stability of Ceftobiprole in Mueller Hinton Broth (MHB)
The stability of ceftobiprole in MHB was determined by sampling at time 0, 8, 16, and 24
hours after the addition of antibiotic to the time-kill curve experiment. One hundred microliters
was obtained from each concentration and frozen until analysis. Prior to analysis, 200 μL of
acetonitrile was added to precipitate any proteins. The samples were then centrifuged at 3000g
for 15 minutes. The supernatant was analyzed using the analytical conditions described in
Chapter 5 with a standard curve ranging from 2mg/L to 64mg/L and quality controls of 6mg/L,
112
24mg/L, and 48mg/L in triplicate. Standards and quality controls were prepared in MHB from
the stock 1 solution and underwent the protein precipitation procedures. The degradation was
assumed to be first order.
Data Analysis
The measured kill curves were analyzed using an appropriate model with nonlinear
regression in NONMEM VI using the ADVAN6 subroutine. Pharmacodynamic parameters were
determined (growth rate, maximum kill rate, EC50, and others depending on the model).
Results
MIC Determination
The MIC was determined to be 2mg/L in all six replicates. No growth of bacteria was
observed in the negative control and growth was observed in the positive control, validating the
experiments. These results are in agreement with the commonly reported MRSA90 of
ceftobiprole, 1-4 mg/L (73, 82, 92, 155).
Time-Kill Curves
The results from the time-kill curve experiments are presented in Figure 7-1.
Ceftobiprole induced rapid bacterial killing with concentrations equal to or above the MIC.
Ceftobiprole concentrations above the MIC allowed for microbial growth, although at a
concentration of 1 mg/L the growth was slightly inhibited for a period of time.
Stability of Ceftobiprole in MHB
Ceftobiprole was found to be degrading throughout the time-kill experiment. Samples
from the flasks containing concentrations ranging from 2-32 mg/L were analyzed. The lower
concentrations (0.5 and 1 mg/L) were below the LOQ of the analytic method after dilution with
acetonitrile for protein precipitation and, therefore, were not analyzed. The results are presented
113
in Figure 7-2. A first-order degradation was assumed and an average degradation rate from all
concentrations was used in the modeling step, i.e. 0.0149 hr-1.
Pharmacodynamic Model Development
Figure 7-1 displays that during the time-kill experiment a maximum is reached in growth
and the kill-kinetics display a biphasic profile. During growth a maximum is reached at a certain
point due to environmental factors, i.e. limited space and nutrients, and the bacteria can no
longer increase in concentration. One common theory used to explain the biphasic profile of
antimicrobial activity is the two-subpopulation model (39, 96, 136). In this model, there are two
bacterial populations present, a drug susceptible population and resistant/persistent population.
This idea was used to model the time-kill curves of ceftobiprole with MRSA. The kill of
susceptible bacteria due to the antibiotic is typically described by an Emax model. In this model
the effect is represented by the following equation:
where Kmax is the maximum effect, C is the concentration of ceftobiprole, and the EC50 is the
concentration needed to produce half the maximum effect. An additional term maybe added, and
was in this instance, to account for the steepness of the slope termed the hill factor (H):
Often the antimicrobial effect is not displayed immediately once drug is added to the in vitro
system. This delay in kill (DEK) can be characterized by adding an additional parameter (dk)
and multiplying the drug effect by the following factor (183):
1
114
Additionally, due to the fact that the antibiotic was degrading throughout the experiment the
following first-order equation was used to account for the changing ceftobiprole concentrations
throughout the experiment:
C=Starting Concentration * e(-degradation rate*time)
The differential equations for the change in the number of susceptible (s) and resistant/persistent
(r) population ith respect to re as follows: s w time a
1
Where 1-(S+R/Nmax) is used to describe the saturation in bacterial concentration due to growth
and Ksr and Krs are the transfer rate constants of the bacteria between the susceptible and
resistant stages.
The above two-subpopulation model was able to describe the data well and the parameters
are presented in Table 7-1. Interexperimental variability was accounted for by using an
exponential error model on the Kmax and EC50 parameters. Additionally, an additive residual
error model was used on the log-transformed data, i.e. a proportional model on the
nontransformed data. Data from two kill curves, out of the 48, were deemed outliers and
excluded from the final data set. This was due to analyst error of omitting to change the pipette
tips at the forth dilution step on one experimental day resulting in bacterial carry-over. The
bacterial concentrations for this particular day were higher than the other five replicates. The
excluded kill-curves were the growth control and the 0.25XMIC concentration. The
concentrations equal to or greater than the MIC were not effected as the number of dilutions
115
steps was always less than or equal to four. The fit of the model was assessed by examination of
the diagnostic plots and individual fits (Figures 7-3,7-4,7-5,7-6).
Pharmacodynamic Model Validation
This model was validated using case deletion diagnostic in which the experimental data
from one day of experiments was removed from the data set and the parameters were
recalculated. The parameters from the validation were compared to those calculated form the
full data set (Table 7-1). While the parameters are fairly consistent between the modified data
sets and the original data set, minimization was only successful in one instance out of the six
attempts and parameters were not able to be obtained from one NONMEM run. This resulted in
only five sets of parameters available for comparison with the parameters obtained from the
original data set. However, the parameters that were obtained were comparable to the full data
set and the model was considered validated.
Conclusions
Ceftobiprole displays antimicrobial activity against MRSA. Time-kill curves are an
appropriate technique to assess the pharmacodynamic profile of ceftobiprole as the antimicrobial
effect over time is assessed. Additionally, a two-subpopulation model, which included a
susceptible population and a resistant/persistent population, was used to fit the time-kill curve
data. This model and the calculated pharmacodynamic parameters may now be used to assess a
specific dosing regimen when combined with pharmacokinetic data.
116
Table 7-1. Pharmacodynamic parameters based on time-kill curves and CDD Results Parameter Full Data
Set Curve 1 Deleted
Curve 2 Deleted
Curve 3 Deleted
Curve 4 Deleted
Curve 5 Deleted
KS (hr-1) 0.913 0.871 0.92 0.823 0.876 0.774 Nmax (CFU/mL)
1.13*1010 1.03*1010 1.22*1010 1.33*1010 9.70*109 5.43*109
Kmax (hr-1) 1.71 1.68 1.67 1.76 1.05 1.26 EC50 (mg/L) 1.17 1.22 1.17 2 1.52 1.94 H 5.13 9.34 4.88 12 7.34 6.2 DK 0.515 0.496 0.551 0.361 0.522 0.856 Ksr (hr-1) 0.122 0.0996 0.13 0.15 0.0895 0.0028 Krs (hr-1) 0.0359 0.0349 0.0337 0.0203 0.0403 0.0417
117
Bac
teria
(CFU
/mL)
Time (hour)
Figure 7-1. Time-kill curves of Ceftobiprole against MRSA performed over 24 hours.
118
Figure 7-2. Degradation of Ceftobiprole in MHB with MRSA at 37ºC.
0
20
40
60
80
100
120
0 8 16 24
%Starting
Con
centration
Time (hr)
2mg/L
4mg/L
8mg/L
16mg/L
32mg/L
119
Figure 7-3. Diagnostic plot of observed CFU/mL vs. population predicted.
120
Figure 7-4. Diagnostic plot of the natural log observed CFU/mL vs. individual predicted.
121
Figure 7-5. Diagnostic plot of weighted residuals vs. individual predicted.
122
Figure 7-6. Individual fits of the model population predicted and individual predicted curves for time-kill experiments. Dependent variables (circles), population predicted (dashed line), individual predicted (solid line).
123
CHAPTER 8 POPULATION PHARMACOKINETICS AND MONTE CARLO SIMULATIONS FOR THE
EVALUTION OF THE CURRENTLY RECOMMENDED DOSING REGIMEN OF CEFTOBIPROLE
Introduction
The FDA has recognized that model-based drug development provides an opportunity to
streamline the drug development process (68). There have been numerous cases in which
model-based development has saved money and time by aiding in dose selection, clinical trial
design, or support of a given dosing regimen (118). In antimicrobial drug development, and drug
development in general, pharmacokinetic models are based on plasma concentrations. However,
it would be better to base these models on drug concentrations at the site of action. Additionally,
to truly characterize the antimicrobial effect over time, time-kill experiments should be
performed. Therefore, the most rationale approach for dose evaluation would be to use free,
active concentrations at the site of action for PK model development and data from time-kill
curve experiments for PD model development. An integrated PK/PD model could then be used
to calculate the predicted target attainment rate of a given dosing regimen.
It is the goal of this chapter to use modeling and simulation techniques to evaluate the
one of the currently recommended dosing regimens of ceftobiprole, 500mg administered as a
two-hour i.v. infusion every eight hours. This will be done by developing a pharmacokinetic
model to simultaneously simulate free concentration-time curves in three different tissues, i.e.
plasma, skeletal muscle ISF, and s.c. adipose tissue ISF, based on plasma and microdialysis data.
This information will then be used to predict the target attainment rate using two different
endpoints, the traditional fT>MIC and the endpoint of 1-log kill at 24 hours.
124
Materials and Methods
Population Pharmacokinetic Model Development
The details of the microdialysis study are presented in chapter 6. Plasma data from the
microdialysis experiment was adjusted for individual protein binding and compiled with the
microdialysis data, which also represents free concentrations, to make the full data set for use. A
model was developed using NONMEM VI with the ADVAN6 subroutine. This subroutine
allowed the differential equations to be user supplied. Interindividual variability was included in
the model using an exponential error model if deemed appropriate, i.e. if incorporation of an
additional parameter revealed a significant drop in the objective function. A drop of 3.84 was
used representing a 0.05 level of significance. A proportional plus additive model was used to
describe the residual error. Model fit was assessed by visual inspection of the diagnostic plots.
The model was validated by performing a bootstrap analysis using 1000 runs with the aid of
wings for NONMEM. The possibility of covariates was explored by using the GAM analysis in
Xpose4 and visual inspection of the correlation matrix between ETAs (interindividual
variability) and possible covariates.
Probability of Target Attainment
Time above MIC approach: Once the population PK model was finalized and validated,
NONMEM was used to simultaneously simulate the pharmacokinetic profiles for free plasma,
free skeletal muscle ISF, and free s.c. adipose tissue ISF for 1000 subjects, 500 females and 500
males. Then the fT>MIC was calculated at steady state by simulating three consecutive doses,
500mg every 8 hours delivered as two-hour i.v. infusions, and using the last dosing interval for
the calculation of the fT>MIC. To calculate a clinically relevant target attainment probability, a
MRSA MIC distribution obtained from patients with bactereamia was used (86). The target
attainment rate for the fT>MIC of 30% and 40% of the dosing interval were both calculated.
125
Time-kill curves: Using the PD model developed in chapter 7, the 1000 simulated
concentration-time profiles of all three tissues were inserted into Emax model. Then, 1000
simulations were performed to yield the antimicrobial effect over 24 hours. The probability of
target attainment was assessed based on a 1-log kill at 24 hours from the starting inoculum. The
starting inoculum used in the simulation was an average of the starting inoculums from the six
separate time-kill experiments, i.e. 48 starting inoculums total.
In order to assess the target attainment rate using a population of bacteria, not just the
single experimental strain from the time-kill curve experiments, the MIC distribution was
converted to an EC50 distribution. Again, the target attainment rate was assessed based on a 1-log
kill at 24 hours. The MIC was converted to the EC ing the following equation (122): 50 us
10
10
/
In this equation, time is set at 20 hours, Nt is the concentration at which growth is visible (107
CFU/mL), N0 is the starting inoculum (5*105 CFU/mL), and the remaining parameters are fixed
from the time-kill curve experiments. This method makes the major assumption that although
the EC50 changes, the remaining PD parameters do not change. This may not necessarily be true.
After calculating the new EC50 values, the simulations were performed again to find the
probability of target attainment based on a 1-log kill at twenty-four hours.
Results
Population Pharmacokinetic Model Development
A simple two-compartment body model with elimination from the central compartment
was able to accurately fit the data (Figure 8-1). This model was developed using free
concentrations in all tissues. The deferential equations for this model are as follows:
126
112 1 21 2 10 1 0
2
12 1 21 2
In the above equations, K0 represents the infusion rate and was included in the model during
dosing only. To obtain concentrations, instead of amounts, a scaling factor was used. For the
central compartment the scaling factor is the volume of distribution of the central compartment
(Vc). However, for the microdialysis data, the scaling factor would be the volume of distribution
of the peripheral compartment (Vp). Due to the fact that at steady state K12*Vc=K21*Vp, Vp
was replaced by K12*Vc/K21. Also, since free concentrations in plasma do not equal free
concentrations in tissue ISF an additional parameter was added to the model to account for this,
termed the distribution factor for muscle or adipose (DFM or DFA). Therefore, the equations for
the free concentration in plasma, skeletal muscle, and s.c. adipose tissue respectively are:
1
2 21
12
. . 2 21
12
Interindividual variability was accounted for by using an exponential error model on the
following parameters; K12, Vc, and DFA. Once the base model was established, a bootstrap
analysis was performed for model validation. However, from the bootstrap, it was seen that K12
displays a high degree of fluctuation, with a 90% confidence interval of 12.3-110, and the single
run parameter estimate was approximately half the mean from the 1000 runs. Therefore, this
parameter was fixed to the single run estimate, 21.6 hr-1. Justification for fixing this parameter
was based on calculating the terminal phase of the average concentration time-curve (Beta) using
127
the microconstants and comparing that value to the value obtained for the elimination rate
constant (λz) from the noncompartmental analysis. These values should be approximately equal.
Additionally, the volume of distribution (Vss) of beta-lactams is approximately equal to total
body water. Therefore, using the equation Vss=Vc*(1+K12/K21) and substituting the calculated
parameters, a volume of distribution approximately equal to total body water was obtained as
expected. After K12 was fixed the bootstrap analysis was performed again (Table 8-1).
After a stable final base model was established, age, weight, height, and sex were
examined as possible covariates. Typically, weight is observed as a covariate on the volume of
distribution. However, due the small number of subjects, 12, and the fact that only healthy
volunteers were selected for the microdialysis study, the only significant covariate revealed from
the GAM analysis and visual inspection of the plots of interindividual variability (etas) vs.
possible covariates was sex on the distribution factor for s.c. adipose tissue (Figure 8-2). This
factor was added to the model by the following equation, where in the data set males are
represented by 0 and females by 1: DFA=THETA(6)+(1-SEX)*THETA(7). The addition of the
parameter led to a decrease of the objective function of 16, 210→194.
The fit of this final covariate model was examined by visual inspection of the diagnostic
plots (Figures 8-3, 8-4, 8-5, 8-6, 8-7, 8-8). Again a bootstrap analysis was performed to validate
the final covariate model and the parameters from the single run were used for simulations
(Table 8-2).
Probability of Target Attainment
Time above MIC: The time above MIC was calculated at steady state from 1000
simulated subjects. The probability of target attainment in each tissue at steady state for each
MIC and overall probability of target attainment for each tissue is presented in Tables 8-3 and 8-
4 for a target of fT>MIC of 30% and 40% of the dosing interval, respectively. The two slightly
128
different targets were examined since 40% is the typical target for beta-lactams (42); however, in
vivo animal studies have shown that with ceftobiprole against MRSA a lower target maybe
acceptable (46).
Time-kill curves: The probability of target attainment was calculated by combining the
pharmacokinetic data from the simulation of 1000 subjects with the pharmacodynamic model
and again performing 1000 simulations. The probability of target attainment was based on the
endpoint of 1-log kill at 24 hours. The probability of target attainment in each tissue was as
follows: 99.90% in plasma, 99.90% in muscle, and 88.50% in adipose tissue. An example of a
clinical success and failure is presented in Figure 8-9.
In order to obtain a bacterial susceptibility distribution encountered clinically, the MIC
values were converted to EC50 values. At each EC50 value 1000 simulations were performed
using the same PK input (Table 8-5). From Table 8-5 it can be seen that this dosing regimen is
efficacious against highly susceptible organisms. A different dosing regimen may be needed
against pathogens with a low susceptibility to ceftobiprole, especially to obtain the target in s.c.
adipose tissue. The results obtained using this technique should be interpreted cautiously,
however, since using the MIC to calculate an EC50 results in step-wise increases in the EC50,
which is not a true representation of the population susceptibility. Additionally, the simulations
assume that while the EC50 is different the other pharmacodynamic parameters remain the same.
Conclusions
The free concentration of ceftobiprole was determined in three tissues, plasma, skeletal
muscle ISF, and s.c. adipose tissue ISF. The concentration in these tissues is important due to
the fact that they represent the relevant site of action. A model was then developed to predict the
concentration of all tissues simultaneously. To evaluate efficacy two different pharmacodynamic
endpoints were used, fT>MIC and 1-log kill. Based on the data obtained from the pharmacokinetic
129
and pharmacodynamic studies performed both endpoints supported the generally recommended
dosing regimen of 500mg every eight hours administers as a two-hour i.v. infusion. More time-
kill curve studies are needed to obtain a susceptibility distribution for ceftobiprole against
MRSA. However, converting an MIC distribution to an EC50 distribution and calculating the
target attainment rate of achieving a 1-log kill at 24 hours suggests that different dosing regimen
may be needed with organisms that have a lower susceptibility. These results should be
interpreted cautiously as some major assumptions are made.
130
Figure 8-1. Two-compartment body model with elimination from the central compartment and
drug delivery via i.v. infusion.
131
ETA1
-0.1 0.2 0.0 0.6 60 70 80
-0.3
0.1
-0.1
0.2
ETA2
ETA3
-1.0
0.0
0.0
0.6
SEX
HT
160
175
6070
80
WT
-0.3 0.1 -1.0 0.0 160 175 20 2620
26AGE
Figure 8-2. Plots of interindividual variability vs. possible covariates.
132
Figure 8-3. Dependent variables vs. population predicted values (mg/L).
133
Figure 8-4. Dependent variables vs. individual predicted values (mg/L).
134
Figure 8-5. Weighted residuals vs. population predicted values.
135
Figure 8-6. Individual fits of the covariate model population predicted and individual predicted
curves for free plasma. Dependent variables (circles), population predicted (dashed line), individual predicted (solid line).
136
Figure 8-7. Individual fits of the model population predicted and individual predicted curves for
skeletal muscle. Dependent variables (circles), population predicted (dashed line), individual predicted (solid line).
137
Figure 8-8. Individual fits of the model population predicted and individual predicted curves for
s.c. adipose tissue. Dependent variables (circles), population predicted (dashed line), individual predicted (solid line).
138
Figure 8-9. Example of a success and a failure based on achieving 1-log kill at 24 hours. Data
obtained from the simulation of antibiotic effect over 24 hours in adipose tissue using the experimentally obtained EC50 value.
1.0E+03
1.0E+04
1.0E+05
1.0E+06
1.0E+07
1.0E+08
0.00 5.00 10.00 15.00 20.00 25.00 30.00
Bacteria (C
FU/m
L)
Time (hour)
Failure
Success
139
Table 8-1. Results from bootstrap of base model (1000 runs) PK Parameter Single Run Bootstrap 90%CI
K12 (hr-1) 21.6 21.6 N/A
K21 (hr-1) 2.98 3.04 2.44-3.83
K10 (hr-1) 3.17 3.17 2.57-3.88
Vc (L) 1.91 1.93 1.65-2.25
DFM 0.667 0.671 0.61-0.761
DFA 0.46 0.465 0.355-0.579
Interindividual Variability
K12 0.0271 0.0249 0.028-0.0519
Vc 0.0171 0.0138 0.00296-0.0317
DFA 0.36 0.333 0.15-0.551
Residual Variability
Sigma 1 0.0324 0.0312 0.024-0.039
Sigma 2 0.0579 0.0548 0.0264-0.0902
140
Table 8-2. Results from bootstrap of covariate model (1000 runs) PK Parameter Single Run Bootstrap 90%CI
K12 (hr-1) 21.6 21.6 N/A
K21 (hr-1) 3.04 3.10 2.46-3.96
K10 (hr-1) 3.14 3.14 2.53-3.87
Vc (L) 1.93 1.95 1.64-2.31
DFM 0.668 0.672 0.601-0.762
DFA 0.362 0.350 0.197-0.538
Sex on DFA 0.193 0.231 0.0124-0.436
Interindividual Variability
K12 0.027 0.0248 0.00453-0.0523
Vc 0.0172 0.0139 0.00305-0.0318
DFA 0.207 0.150 0.0208-0.387
Residual Variability
Sigma 1 0.0315 0.0305 0.024-0.0378
Sigma 2 0.0605 0.0548 0.0259-0.0919
141
Table 8-3. Probability target attainment for fT>MIC of >30% at steady state MIC (mg/L) 0.25 0.5 1 2 4 Fraction MIC Distribution
0.24 2.20 26.10 70.73 0.73
Plasma 100 100 100 100 100 Muscle 100 100 100 100 100 S.C. Adipose 100 100 100 98.50 80.70 Overall Attainment
Plasma 100 100
98.80 Muscle S.C. Adipose All values expect the MIC are in percentages.
142
Table 8-4. Probability target attainment fT>MIC of >40% at steady state MIC(mg/L) 0.25 0.5 1 2 4 Fraction MIC Distribution
0.24 2.20 26.10 70.73 0.73
Plasma 100 100 100 100 100 Skeletal Muscle
100 100 100 100 100
S.C. Adipose 100 100 99.90 97.00 73.10 Overall Attainment
Plasma 100 100
97.66 Muscle S.C. Adipose All values except the MIC are in percentages.
143
Table 8-5. Probability of target attainment based on 1-log kill at 24 hours (3 doses) MIC (mg/L) 0.25 0.5 1 2 4 EC50 (mg/L) 0.26 0.52 1.04 2.09 4.17 Fraction MIC Distribution
0.24 2.20 26.10 70.73 0.73
Plasma 99.90 99.90 99.90 99.70 72.30 Muscle 99.90 99.90 99.90 95.80 24.20 S.C. Adipose 99.90 99.60 91.90 57.80 13.70 Overall Attainment
Plasma 99.56 96.45 67.40
Muscle S.C. Adipose All values except the MIC and EC50 are in percentages.
144
CHAPTER 9 CONCLUSIONS
Ceftobiprole is promising new cephalosporin with activity against a wide-range of both
Gram-positive and Gram-negative pathogens, including several resistance species such as MRSA
and PRSP. The generally recommended dosing regimen of ceftobiprole is 500mg administered
as a two hour i.v. infusion every eight hours. However, in cases of documented Gram-positive
infections, excluding diabetic foot infections, 500mg administered as a one hour i.v. infusion
every twelve hours is recommended. It was the goal of this project to evaluate the efficacy of the
more empiric, general dosing regimen of 500mg administered as a two-hour i.v. infusion every
eight hours.
Traditionally, plasma samples have been used as the PK input in PK/PD indices for the
evaluation of efficacy with antimicrobial agents. However, with antimicrobials such as
ceftobiprole, with the lead indication of cSSSI, it is more rationale to use free, pharmacologically
active concentrations at the site of action for efficacy evaluation, i.e. the ISF of subcutaneous
(s.c.) soft tissues. Therefore, a microdialysis study was conducted in healthy volunteers to
determine the free concentration-time profile in plasma, the ISF of s.c. adipose tissue, and the
ISF of skeletal muscle. The data from this study was used to develop a pharmacokinetic model
so that the concentration in all three tissues could be predicted simultaneously. It was
determined that a two-compartment body-model with elimination from the central compartment
fit the data accurately. Additionally, due to the fact that free concentrations in the ISF were not
equal to free plasma concentrations, a distribution factor for each soft tissue was included in this
model.
The major PD parameter used for antimicrobials is the MIC. This parameter, however, has
many limitations. Mainly, the MIC does not show the antimicrobial activity over time.
145
Additionally, the MIC does not show the extent of kill, it has a two-fold variability, and it does
not show the presence of resistant/persistent bacteria, which may be present at a concentration
below visual detection. Therefore, it seems more rationale to perform time-kill curve
experiments to characterize the antimicrobial activity. It these experiments an antimicrobial
profile is obtained over time and a mathematical model, typically an Emax model, is developed so
that the pharmacodynamic parameters can be calculated. Time-kill curves were performed with
MRSA ATCC 33591 and ceftobiprole. A two-subpopulation PD model, which included a drug
susceptible population and a drug resistant/persistent population, was developed and the PD
parameters were calculated. In this model additional terms were included to account for a
saturation in growth and a time delay in antimicrobial effect.
Once the PK and PD models were developed and validated, simulations were performed to
predict the efficacy of the specified dosing regimen using two-different pharmacodynamic
endpoints. The first method involved using the traditional PD parameter, the MIC, and a target
of fT>MIC of 30% or 40% of the dosing interval. Simulations were performed to create a data set
with the PK profiles of 1000 subjects from all three tissues simultaneously over 24 hours, i.e.
three subsequent doses. A MIC distribution obtained from literature determined that this dosing
regimen meets the PK/PD target in all three tissues at steady state. In plasma and skeletal muscle
it was predicted that with 1000 subjects the target would be reached 100% of the time. In s.c.
adipose tissue it was determined that the target would be reached approximately 98% of the time
using the target of fT>MIC of 40%.
The second method involved using the PD parameters obtained from the time-kill
experiment and modeling to evaluate the efficacy of the specified dosing regimen. The same
simulation that created the PK profiles for 1000 subjects was used as the PK input into the PD
146
model. Then 1000 PD profiles were simulated and efficacy was evaluated based on the target of
at least a 1-log kill at 24 hours. The probability of target attainment in each tissue was as
follows: 99.90% in plasma, 99.90% in muscle, and 88.50% in adipose tissue. Based on this data
the dosing regimen was deemed efficacious. However, this is based on the susceptibility of only
one strain of MRSA. In order to get an accurate representation of a bacterial population, a
distribution of susceptibilities is needed. This was accomplished by using the MIC distribution
to create an EC50 distribution by calculating the EC50 based on the MIC. Again simulations were
performed 1000 times based on each new EC50 using the same 1000 simulated PK profiles as
input into the PD model. It was found that this dosing regimen is efficacious in all three tissues
with pathogens which are highly susceptible to ceftobiprole. However, with pathogens which
are less susceptible a different dosing regimen may need to be considered. The overall target
attainment based on a target of 1-log kill at 24hours was as follows: 99.56% for free plasma,
96.45% for skeletal muscle, and 67.40% for s.c. adipose tissue. It is important to interpret these
results cautiously for several reasons. One reason is that using the MIC to create an EC50
distribution is not entirely accurate due to the fact that the MIC is only available in two-fold
increments and the EC50 can be any number inbetween two MIC values. For example, the EC50
does not have to be 1.04 or 2.09 mg/L but it can be 1.17 mg/L as it was found to be in the
calculation of the PD parameters from the time-kill experiments presented in Chapter 7. Also,
by using the MIC to calculate the EC50 and then performing simulations based on the PD model
the major assumption is made that although the EC50 value changes, all other PD parameters
remain the same. This may or may not be true. Finally, the PK determined from the
microdialysis study was done in healthy volunteers. The PK in patients and the distribution of
ceftobiprole into the ISF of infected tissues may be different.
147
148
In conclusion, based on the experiments performed, the currently recommended general
dosing regimen of 500mg administered as a two-hour i.v. infusion every eight hours is
efficacious based two pharmacodynamic endpoints, a fT>MIC of 40% of the dosing interval at
steady state and a 1-log bacterial kill at 24 hours following three subsequent doses.
LIST OF REFERENCES
1. Ackerman, B. H., E. H. Taylor, K. M. Olsen, W. Abdel-Malak, and A. A. Pappas. 1988. Vancomycin serum protein binding determination by ultrafiltration. Drug Intell Clin Pharm 22:300-3.
2. Ambrose, P. G., S. M. Bhavnani, C. M. Rubino, A. Louie, T. Gumbo, A. Forrest, and G. L. Drusano. 2007. Pharmacokinetics-pharmacodynamics of antimicrobial therapy: it's not just for mice anymore. Clin Infect Dis 44:79-86.
3. Ambrose, P. G., D. M. Grasela, T. H. Grasela, J. Passarell, H. B. Mayer, and P. F. Pierce. 2001. Pharmacodynamics of fluoroquinolones against Streptococcus pneumoniae in patients with community-acquired respiratory tract infections. Antimicrob Agents Chemother 45:2793-7.
4. Amsden, G. W. 2001. Advanced-generation macrolides: tissue-directed antibiotics. Int J Antimicrob Agents 18 Suppl 1:S11-5.
5. Andes, D., J. Anon, M. R. Jacobs, and W. A. Craig. 2004. Application of pharmacokinetics and pharmacodynamics to antimicrobial therapy of respiratory tract infections. Clin Lab Med 24:477-502.
6. Andes, D., and W. A. Craig. 2007. In vivo pharmacodynamic activity of the glycopeptide dalbavancin. Antimicrob Agents Chemother 51:1633-42.
7. Andes, D., and W. A. Craig. 2006. Pharmacodynamics of a new cephalosporin, PPI-0903 (TAK-599), active against methicillin-resistant Staphylococcus aureus in murine thigh and lung infection models: identification of an in vivo pharmacokinetic-pharmacodynamic target. Antimicrob Agents Chemother 50:1376-83.
8. Andes, D., M. L. van Ogtrop, J. Peng, and W. A. Craig. 2002. In vivo pharmacodynamics of a new oxazolidinone (linezolid). Antimicrob Agents Chemother 46:3484-9.
9. Andes D., W. Craig. 2000. Presented at the 40th Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, Sep 17-20.
10. Arias, C. A., K. V. Singh, D. Panesso, and B. E. Murray. 2007. Evaluation of ceftobiprole medocaril against Enterococcus faecalis in a mouse peritonitis model. J Antimicrob Chemother 60:594-8.
11. AstraZeneca 2007, posting date. Merrem IV Prescribing Information. [Online.]
12. Azoulay-Dupuis, E., J. P. Bedos, J. Mohler, A. Schmitt-Hoffmann, M. Schleimer, and S. Shapiro. 2004. Efficacy of BAL5788, a prodrug of cephalosporin BAL9141, in a mouse model of acute pneumococcal pneumonia. Antimicrob Agents Chemother 48:1105-11.
149
13. Balaban, N. Q., J. Merrin, R. Chait, L. Kowalik, and S. Leibler. 2004. Bacterial persistence as a phenotypic switch. Science 305:1622-5.
14. Barbour, A., S. Schmidt, B. Ma, L. Schiefelbein, K.H. Rand, O. Burkhardt, H. Derendorf. 2009. Clinical Pharmacokinetics and Pharmacodynamics of Tigecycline. Accepted in Clinical Pharmacokinetics.
15. Barbour A., B. Murthy, S. Schmidt, S. Sabarinath, D. Skee, H. Tian, D. Desai-Kreiger, D. Balis, M. Grant, C. Seubert, and H. Derendorf. 2008. Presented at the 48th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, DC, October 25-28.
16. Begg, E. J., B. A. Peddie, S. T. Chambers, and D. R. Boswell. 1992. Comparison of gentamicin dosing regimens using an in-vitro model. J Antimicrob Chemother 29:427-33.
17. Bhavnani, S. 2008. Presented at the International Society of Anti-Infective Pharmacology-William C. Craig Symposium, Madison, WI.
18. Bhavnani, S. M., J. A. Passarell, J. S. Owen, J. S. Loutit, S. B. Porter, and P. G. Ambrose. 2006. Pharmacokinetic-pharmacodynamic relationships describing the efficacy of oritavancin in patients with Staphylococcus aureus bacteremia. Antimicrob Agents Chemother 50:994-1000.
19. Biavasco, F., C. Vignaroli, R. Lupidi, E. Manso, B. Facinelli, and P. E. Varaldo. 1997. In vitro antibacterial activity of LY333328, a new semisynthetic glycopeptide. Antimicrob Agents Chemother 41:2165-72.
20. Blaser, J., B. B. Stone, M. C. Groner, and S. H. Zinner. 1987. Comparative study with enoxacin and netilmicin in a pharmacodynamic model to determine importance of ratio of antibiotic peak concentration to MIC for bactericidal activity and emergence of resistance. Antimicrob Agents Chemother 31:1054-60.
21. Bobkova, E. V., Y. P. Yan, D. B. Jordan, M. G. Kurilla, and D. L. Pompliano. 2003. Catalytic properties of mutant 23 S ribosomes resistant to oxazolidinones. J Biol Chem 278:9802-7.
22. Bogdanovich, T., C. Clark, L. Ednie, G. Lin, K. Smith, S. Shapiro, and P. C. Appelbaum. 2006. Activities of ceftobiprole, a novel broad-spectrum cephalosporin, against Haemophilus influenzae and Moraxella catarrhalis. Antimicrob Agents Chemother 50:2050-7.
23. Bogdanovich, T., L. M. Ednie, S. Shapiro, and P. C. Appelbaum. 2005. Antistaphylococcal activity of ceftobiprole, a new broad-spectrum cephalosporin. Antimicrob Agents Chemother 49:4210-9.
24. Bosnar, M., Z. Kelneric, V. Munic, V. Erakovic, and M. J. Parnham. 2005. Cellular uptake and efflux of azithromycin, erythromycin, clarithromycin, telithromycin, and cethromycin. Antimicrob Agents Chemother 49:2372-7.
150
25. Boylan, C. J., K. Campanale, P. W. Iversen, D. L. Phillips, M. L. Zeckel, and T. R. Parr, Jr. 2003. Pharmacodynamics of oritavancin (LY333328) in a neutropenic-mouse thigh model of Staphylococcus aureus infection. Antimicrob Agents Chemother 47:1700-6.
26. Bozdogan, B., D. Esel, C. Whitener, F. A. Browne, and P. C. Appelbaum. 2003. Antibacterial susceptibility of a vancomycin-resistant Staphylococcus aureus strain isolated at the Hershey Medical Center. J Antimicrob Chemother 52:864-8.
27. Bryan, L. E., and H. M. Van Den Elzen. 1977. Effects of membrane-energy mutations and cations on streptomycin and gentamicin accumulation by bacteria: a model for entry of streptomycin and gentamicin in susceptible and resistant bacteria. Antimicrob Agents Chemother 12:163-77.
28. Buerger, C., N. Plock, P. Dehghanyar, C. Joukhadar, and C. Kloft. 2006. Pharmacokinetics of unbound linezolid in plasma and tissue interstitium of critically ill patients after multiple dosing using microdialysis. Antimicrob Agents Chemother 50:2455-63.
29. Bundtzen, R. W., A. U. Gerber, D. L. Cohn, and W. A. Craig. 1981. Postantibiotic suppression of bacterial growth. Rev Infect Dis 3:28-37.
30. Burgess, D. S., and C. R. Frei. 2005. Comparison of beta-lactam regimens for the treatment of gram-negative pulmonary infections in the intensive care unit based on pharmacokinetics/pharmacodynamics. J Antimicrob Chemother 56:893-8.
31. Burkhardt, O., M. Brunner, S. Schmidt, M. Grant, Y. Tang, and H. Derendorf. 2006. Penetration of ertapenem into skeletal muscle and subcutaneous adipose tissue in healthy volunteers measured by in vivo microdialysis. J Antimicrob Chemother 58:632-6.
32. Bustamante, C. I., G. L. Drusano, B. A. Tatem, and H. C. Standiford. 1984. Postantibiotic effect of imipenem on Pseudomonas aeruginosa. Antimicrob Agents Chemother 26:678-82.
33. Campion, J. J., P. J. McNamara, and M. E. Evans. 2005. Pharmacodynamic modeling of ciprofloxacin resistance in Staphylococcus aureus. Antimicrob Agents Chemother 49:209-19.
34. CDER 2003, posting date. Guidance for Industry: Bioavailability and Bioequivalence Studies for Orally Administered Drug Products - General Considerations. [Online.]
35. Cercenado, E., F. Garcia-Garrote, and E. Bouza. 2001. In vitro activity of linezolid against multiply resistant Gram-positive clinical isolates. J Antimicrob Chemother 47:77-81.
36. Chambers, H. F. 2005. Evaluation of ceftobiprole in a rabbit model of aortic valve endocarditis due to methicillin-resistant and vancomycin-intermediate Staphylococcus aureus. Antimicrob Agents Chemother 49:884-8.
151
37. Chaurasia, C. S., M. Muller, E. D. Bashaw, E. Benfeldt, J. Bolinder, R. Bullock, P. M. Bungay, E. C. DeLange, H. Derendorf, W. F. Elmquist, M. Hammarlund-Udenaes, C. Joukhadar, D. L. Kellogg, Jr., C. E. Lunte, C. H. Nordstrom, H. Rollema, R. J. Sawchuk, B. W. Cheung, V. P. Shah, L. Stahle, U. Ungerstedt, D. F. Welty, and H. Yeo. 2007. AAPS-FDA workshop white paper: microdialysis principles, application and regulatory perspectives. Pharm Res 24:1014-25.
38. Chien, J. Y., S. Friedrich, M. A. Heathman, D. P. de Alwis, and V. Sinha. 2005. Pharmacokinetics/Pharmacodynamics and the stages of drug development: role of modeling and simulation. AAPS J 7:E544-59.
39. Chung, P., P. J. McNamara, J. J. Campion, and M. E. Evans. 2006. Mechanism-based pharmacodynamic models of fluoroquinolone resistance in Staphylococcus aureus. Antimicrob Agents Chemother 50:2957-65.
40. Conte, J. E., Jr., J. A. Golden, M. G. Kelly, and E. Zurlinden. 2005. Steady-state serum and intrapulmonary pharmacokinetics and pharmacodynamics of tigecycline. Int J Antimicrob Agents 25:523-9.
41. Contopoulos-Ioannidis, D. G., N. D. Giotis, D. V. Baliatsa, and J. P. Ioannidis. 2004. Extended-interval aminoglycoside administration for children: a meta-analysis. Pediatrics 114:e111-8.
42. Craig, W. A. 2001. Does the dose matter? Clin Infect Dis 33 Suppl 3:S233-7.
43. Craig, W. A. 2002. Pharmacodynamics of Antimicrobials: General Concepts and Applications, p. 1-22, Antimicrobial Pharmacodynamics in Theory and Clinical Practice. Marcel Dekker, Inc., New York.
44. Craig, W. A. 1998. Pharmacokinetic/pharmacodynamic parameters: rationale for antibacterial dosing of mice and men. Clin Infect Dis 26:1-10; quiz 11-2.
45. Craig, W. A. 1993. Post-antibiotic effects in experimental infection models: relationship to in-vitro phenomena and to treatment of infections in man. J Antimicrob Chemother 31 Suppl D:149-58.
46. Craig, W. A., and D. R. Andes. 2008. In vivo pharmacodynamics of ceftobiprole against multiple bacterial pathogens in murine thigh and lung infection models. Antimicrob Agents Chemother 52:3492-6.
47. Craig, W. A., and S. C. Ebert. 1990. Killing and regrowth of bacteria in vitro: a review. Scand J Infect Dis Suppl 74:63-70.
48. Cunha, B. A., P. Domenico, and C. B. Cunha. 2000. Pharmacodynamics of doxycycline. Clin Microbiol Infect 6:270-3.
152
49. Daikos, G. L., V. T. Lolans, and G. G. Jackson. 1991. First-exposure adaptive resistance to aminoglycoside antibiotics in vivo with meaning for optimal clinical use. Antimicrob Agents Chemother 35:117-23.
50. Davies, T. A., M. G. Page, W. Shang, T. Andrew, M. Kania, and K. Bush. 2007. Binding of ceftobiprole and comparators to the penicillin-binding proteins of Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus, and Streptococcus pneumoniae. Antimicrob Agents Chemother 51:2621-4.
51. Davies, T. A., W. Shang, and K. Bush. 2006. Activities of ceftobiprole and other beta-lactams against Streptococcus pneumoniae clinical isolates from the United States with defined substitutions in penicillin-binding proteins PBP 1a, PBP 2b, and PBP 2x. Antimicrob Agents Chemother 50:2530-2.
52. De La Pena, P. A., M. Brunner, H. G. Eichler, E. Rehak, J. Gross, U. Thyroff-Friesinger, M. Muller, and H. Derendorf. 2002. Comparative target site pharmacokinetics of immediate- and modified-release formulations of cefaclor in humans. J Clin Pharmacol 42:403-11.
53. Dehghanyar, P., C. Burger, M. Zeitlinger, F. Islinger, F. Kovar, M. Muller, C. Kloft, and C. Joukhadar. 2005. Penetration of linezolid into soft tissues of healthy volunteers after single and multiple doses. Antimicrob Agents Chemother 49:2367-71.
54. Delacher, S., H. Derendorf, U. Hollenstein, M. Brunner, C. Joukhadar, S. Hofmann, A. Georgopoulos, H. G. Eichler, and M. Muller. 2000. A combined in vivo pharmacokinetic-in vitro pharmacodynamic approach to simulate target site pharmacodynamics of antibiotics in humans. J Antimicrob Chemother 46:733-9.
55. den Hollander, J. G., J. D. Knudsen, J. W. Mouton, K. Fuursted, N. Frimodt-Moller, H. A. Verbrugh, and F. Espersen. 1998. Comparison of pharmacodynamics of azithromycin and erythromycin in vitro and in vivo. Antimicrob Agents Chemother 42:377-82.
56. Deshpande, L., P. R. Rhomberg, T. R. Fritsche, H. S. Sader, and R. N. Jones. 2004. Bactericidal activity of BAL9141, a novel parenteral cephalosporin against contemporary Gram-positive and Gram-negative isolates. Diagn Microbiol Infect Dis 50:73-5.
57. Deshpande, L. M., and R. N. Jones. 2003. Bactericidal activity and synergy studies of BAL9141, a novel pyrrolidinone-3-ylidenemethyl cephem, tested against streptococci, enterococci and methicillin-resistant staphylococci. Clin Microbiol Infect 9:1120-4.
58. Deziel, M. R., H. Heine, A. Louie, M. Kao, W. R. Byrne, J. Basset, L. Miller, K. Bush, M. Kelly, and G. L. Drusano. 2005. Effective antimicrobial regimens for use in humans for therapy of Bacillus anthracis infections and postexposure prophylaxis. Antimicrob Agents Chemother 49:5099-106.
153
59. Dowell, J. A., B. P. Goldstein, M. Buckwalter, M. Stogniew, and B. Damle. 2008. Pharmacokinetic-pharmacodynamic modeling of dalbavancin, a novel glycopeptide antibiotic. J Clin Pharmacol 48:1063-8.
60. Drusano, G. L., D. E. Johnson, M. Rosen, and H. C. Standiford. 1993. Pharmacodynamics of a fluoroquinolone antimicrobial agent in a neutropenic rat model of Pseudomonas sepsis. Antimicrob Agents Chemother 37:483-90.
61. Drusano, G. L., S. L. Preston, C. Fowler, M. Corrado, B. Weisinger, and J. Kahn. 2004. Relationship between fluoroquinolone area under the curve: minimum inhibitory concentration ratio and the probability of eradication of the infecting pathogen, in patients with nosocomial pneumonia. J Infect Dis 189:1590-7.
62. Eagle, H., R. Fleischman, and M. Levy. 1953. "Continuous" vs. "discontinuous" therapy with penicillin; the effect of the interval between injections on therapeutic efficacy. N Engl J Med 248:481-8.
63. Eagle, H., R. Fleischman, and A. D. Musselman. 1950. The bactericidal action of penicillin in vivo: the participation of the host, and the slow recovery of the surviving organisms. Ann Intern Med 33:544-71.
64. Ebert, S. 1987. Presented at the 27th Interscience Conference on Antimicrobial Agents and Chemotherapy, New York.
65. Edson, R. S., and C. L. Terrell. 1991. The aminoglycosides. Mayo Clin Proc 66:1158-64.
66. EMEA/CPMP 2000, posting date. Points to consider om pharmacokinetics and pharmacodynamics in the development of antibacterial medicinal products. [Online.]
67. Fasola, E., S. K. Spangler, L. M. Ednie, M. R. Jacobs, S. Bajaksouzian, and P. C. Appelbaum. 1996. Comparative activities of LY 333328, a new glycopeptide, against penicillin-susceptible and -resistant pneumococci. Antimicrob Agents Chemother 40:2661-3.
68. FDA. 2004. Innovation or Stagnation: Challenge and Opportunity on the Critical Path to New Medical Products.
69. Fernandes, P. B., R. Bailer, R. Swanson, C. W. Hanson, E. McDonald, N. Ramer, D. Hardy, N. Shipkowitz, R. R. Bower, and E. Gade. 1986. In vitro and in vivo evaluation of A-56268 (TE-031), a new macrolide. Antimicrob Agents Chemother 30:865-73.
70. Forrest, A., S. Chodosh, M. A. Amantea, D. A. Collins, and J. J. Schentag. 1997. Pharmacokinetics and pharmacodynamics of oral grepafloxacin in patients with acute bacterial exacerbations of chronic bronchitis. J Antimicrob Chemother 40 Suppl A:45-57.
154
71. Forrest, A., D. E. Nix, C. H. Ballow, T. F. Goss, M. C. Birmingham, and J. J. Schentag. 1993. Pharmacodynamics of intravenous ciprofloxacin in seriously ill patients. Antimicrob Agents Chemother 37:1073-81.
72. Freeman, C. D., C. H. Nightingale, D. P. Nicolau, P. P. Belliveau, M. A. Banevicius, and R. Quintiliani. 1994. Intracellular and extracellular penetration of azithromycin into inflammatory and noninflammatory blister fluid. Antimicrob Agents Chemother 38:2449-51.
73. Fritsche, T. R., H. S. Sader, and R. N. Jones. 2008. Antimicrobial activity of ceftobiprole, a novel anti-methicillin-resistant Staphylococcus aureus cephalosporin, tested against contemporary pathogens: results from the SENTRY Antimicrobial Surveillance Program (2005-2006). Diagn Microbiol Infect Dis 61:86-95.
74. Fritsche, T. R., H. S. Sader, and R. N. Jones. 2007. Potency and spectrum of garenoxacin tested against an international collection of skin and soft tissue infection pathogens: report from the SENTRY antimicrobial surveillance program (1999-2004). Diagn Microbiol Infect Dis 58:19-26.
75. Frossard, M., C. Joukhadar, B. M. Erovic, P. Dittrich, P. E. Mrass, M. Van Houte, H. Burgmann, A. Georgopoulos, and M. Muller. 2000. Distribution and antimicrobial activity of fosfomycin in the interstitial fluid of human soft tissues. Antimicrob Agents Chemother 44:2728-32.
76. Fukuda, H., and K. Hiramatsu. 1999. Primary targets of fluoroquinolones in Streptococcus pneumoniae. Antimicrob Agents Chemother 43:410-2.
77. Gales, A. C., H. H. Sader, and R. N. Jones. 2002. Respiratory tract pathogens isolated from patients hospitalized with suspected pneumonia in Latin America: frequency of occurrence and antimicrobial susceptibility profile: results from the SENTRY Antimicrobial Surveillance Program (1997-2000). Diagn Microbiol Infect Dis 44:301-11.
78. Gales, A. C., H. S. Sader, S. S. Andrade, L. Lutz, A. Machado, and A. L. Barth. 2006. Emergence of linezolid-resistant Staphylococcus aureus during treatment of pulmonary infection in a patient with cystic fibrosis. Int J Antimicrob Agents 27:300-2.
79. Gattringer, R., E. Urbauer, F. Traunmuller, M. Zeitlinger, P. Dehghanyar, P. Zeleny, W. Graninger, M. Muller, and C. Joukhadar. 2004. Pharmacokinetics of telithromycin in plasma and soft tissues after single-dose administration to healthy volunteers. Antimicrob Agents Chemother 48:4650-3.
80. Gladue, R. P., G. M. Bright, R. E. Isaacson, and M. F. Newborg. 1989. In vitro and in vivo uptake of azithromycin (CP-62,993) by phagocytic cells: possible mechanism of delivery and release at sites of infection. Antimicrob Agents Chemother 33:277-82.
81. Gootz, T. D., R. Zaniewski, S. Haskell, B. Schmieder, J. Tankovic, D. Girard, P. Courvalin, and R. J. Polzer. 1996. Activity of the new fluoroquinolone trovafloxacin
155
(CP-99,219) against DNA gyrase and topoisomerase IV mutants of Streptococcus pneumoniae selected in vitro. Antimicrob Agents Chemother 40:2691-7.
82. Hebeisen, P., I. Heinze-Krauss, P. Angehrn, P. Hohl, M. G. Page, and R. L. Then. 2001. In vitro and in vivo properties of Ro 63-9141, a novel broad-spectrum cephalosporin with activity against methicillin-resistant staphylococci. Antimicrob Agents Chemother 45:825-36.
83. Hersh, A. L., H. F. Chambers, J. H. Maselli, and R. Gonzales. 2008. National trends in ambulatory visits and antibiotic prescribing for skin and soft-tissue infections. Arch Intern Med 168:1585-91.
84. Hoellman, D. B., G. Lin, L. M. Ednie, A. Rattan, M. R. Jacobs, and P. C. Appelbaum. 2003. Antipneumococcal and antistaphylococcal activities of ranbezolid (RBX 7644), a new oxazolidinone, compared to those of other agents. Antimicrob Agents Chemother 47:1148-50.
85. Hollenstein, U., M. Brunner, B. X. Mayer, S. Delacher, B. Erovic, H. G. Eichler, and M. Muller. 2000. Target site concentrations after continuous infusion and bolus injection of cefpirome to healthy volunteers. Clin Pharmacol Ther 67:229-36.
86. Hope, R., D. M. Livermore, G. Brick, M. Lillie, and R. Reynolds. 2008. Non-susceptibility trends among staphylococci from bacteraemias in the UK and Ireland, 2001-06. J Antimicrob Chemother 62 Suppl 2:ii65-74.
87. Ibrahim, E. H., G. Sherman, S. Ward, V. J. Fraser, and M. H. Kollef. 2000. The influence of inadequate antimicrobial treatment of bloodstream infections on patient outcomes in the ICU setting. Chest 118:146-55.
88. Islinger, F., R. Bouw, M. Stahl, E. Lackner, P. Zeleny, M. Brunner, M. Muller, H. G. Eichler, and C. Joukhadar. 2004. Concentrations of gemifloxacin at the target site in healthy volunteers after a single oral dose. Antimicrob Agents Chemother 48:4246-9.
89. Jain, R., and L. H. Danziger. 2004. The macrolide antibiotics: a pharmacokinetic and pharmacodynamic overview. Curr Pharm Des 10:3045-53.
90. Janssen-Ortho Inc June 26, 2008, posting date. Zeftera Product Monograph. [Online.]
91. Jones, R. N. 2003. Global epidemiology of antimicrobial resistance among community-acquired and nosocomial pathogens: a five-year summary from the SENTRY Antimicrobial Surveillance Program (1997-2001). Semin Respir Crit Care Med 24:121-34.
92. Jones, R. N., L. M. Deshpande, A. H. Mutnick, and D. J. Biedenbach. 2002. In vitro evaluation of BAL9141, a novel parenteral cephalosporin active against oxacillin-resistant staphylococci. J Antimicrob Chemother 50:915-32.
156
93. Jumbe, N., A. Louie, R. Leary, W. Liu, M. R. Deziel, V. H. Tam, R. Bachhawat, C. Freeman, J. B. Kahn, K. Bush, M. N. Dudley, M. H. Miller, and G. L. Drusano. 2003. Application of a mathematical model to prevent in vivo amplification of antibiotic-resistant bacterial populations during therapy. J Clin Invest 112:275-85.
94. Karjagin, J., S. Lefeuvre, K. Oselin, K. Kipper, S. Marchand, A. Tikkerberi, J. Starkopf, W. Couet, and R. J. Sawchuk. 2008. Pharmacokinetics of meropenem determined by microdialysis in the peritoneal fluid of patients with severe peritonitis associated with septic shock. Clin Pharmacol Ther 83:452-9.
95. Kashuba, A. D., A. N. Nafziger, G. L. Drusano, and J. S. Bertino, Jr. 1999. Optimizing aminoglycoside therapy for nosocomial pneumonia caused by gram-negative bacteria. Antimicrob Agents Chemother 43:623-9.
96. Katsube, T., Y. Yamano, and Y. Yano. 2008. Pharmacokinetic-pharmacodynamic modeling and simulation for in vivo bactericidal effect in murine infection model. J Pharm Sci 97:1606-14.
97. Keren, I., D. Shah, A. Spoering, N. Kaldalu, and K. Lewis. 2004. Specialized persister cells and the mechanism of multidrug tolerance in Escherichia coli. J Bacteriol 186:8172-80.
98. Knudsen, J. D., K. Fuursted, S. Raber, F. Espersen, and N. Frimodt-Moller. 2000. Pharmacodynamics of glycopeptides in the mouse peritonitis model of Streptococcus pneumoniae or Staphylococcus aureus infection. Antimicrob Agents Chemother 44:1247-54.
99. Lacy, M. K., D. P. Nicolau, C. H. Nightingale, and R. Quintiliani. 1998. The pharmacodynamics of aminoglycosides. Clin Infect Dis 27:23-7.
100. Larsson, A. J., K. J. Walker, J. K. Raddatz, and J. C. Rotschafer. 1996. The concentration-independent effect of monoexponential and biexponential decay in vancomycin concentrations on the killing of Staphylococcus aureus under aerobic and anaerobic conditions. J Antimicrob Chemother 38:589-97.
101. Legat, F. J., R. Krause, P. Zenahlik, C. Hoffmann, S. Scholz, W. Salmhofer, J. Tscherpel, T. Tscherpel, H. Kerl, and P. Dittrich. 2005. Penetration of piperacillin and tazobactam into inflamed soft tissue of patients with diabetic foot infection. Antimicrob Agents Chemother 49:4368-71.
102. Leighton, A., A. B. Gottlieb, M. B. Dorr, D. Jabes, G. Mosconi, C. VanSaders, E. J. Mroszczak, K. C. Campbell, and E. Kelly. 2004. Tolerability, pharmacokinetics, and serum bactericidal activity of intravenous dalbavancin in healthy volunteers. Antimicrob Agents Chemother 48:940-5.
103. Liu, P., M. Muller, M. Grant, B. Obermann, and H. Derendorf. 2005. Tissue penetration of cefpodoxime and cefixime in healthy subjects. J Clin Pharmacol 45:564-9.
157
104. Lodise, T. P., Jr., R. Pypstra, J. B. Kahn, B. P. Murthy, H. C. Kimko, K. Bush, G. J. Noel, and G. L. Drusano. 2007. Probability of target attainment for ceftobiprole as derived from a population pharmacokinetic analysis of 150 subjects. Antimicrob Agents Chemother 51:2378-87.
105. Lodise, T. P., R. Nau, M. Kinzig, G. L. Drusano, R. N. Jones, and F. Sorgel. 2007. Pharmacodynamics of ceftazidime and meropenem in cerebrospinal fluid: results of population pharmacokinetic modelling and Monte Carlo simulation. J Antimicrob Chemother 60:1038-44.
106. Lodise, T. P., S. Preston, V. Bhargava, A. Bryskier, R. Nusrat, S. Chapel, M. Rangaraju, and G. L. Drusano. 2005. Pharmacodynamics of an 800-mg dose of telithromycin in patients with community-acquired pneumonia caused by extracellular pathogens. Diagn Microbiol Infect Dis 52:45-52.
107. Lodise, T. P. J., R. Pypstra, J. B. Kahn, B. P. Murthy, H. C. Kimko, K. Bush, G. J. Noel, and G. L. Drusano. 2007. Probability of target attainment for ceftobiprole as derived from a population pharmacokinetic analysis of 150 subjects. Antimicrob Agents Chemother 51:2378-87.
108. Lorentzen, H., F. Kallehave, H. J. Kolmos, U. Knigge, J. Bulow, and F. Gottrup. 1996. Gentamicin concentrations in human subcutaneous tissue. Antimicrob Agents Chemother 40:1785-9.
109. Lovmar, M., T. Tenson, and M. Ehrenberg. 2004. Kinetics of macrolide action: the josamycin and erythromycin cases. J Biol Chem 279:53506-15.
110. Lowdin, E., I. Odenholt, and O. Cars. 1998. In vitro studies of pharmacodynamic properties of vancomycin against Staphylococcus aureus and Staphylococcus epidermidis. Antimicrob Agents Chemother 42:2739-44.
111. MacGowan, A. P. 2003. Pharmacokinetic and pharmacodynamic profile of linezolid in healthy volunteers and patients with Gram-positive infections. J Antimicrob Chemother 51 Suppl 2:ii17-25.
112. Majumdar, A. K., D. G. Musson, K. L. Birk, C. J. Kitchen, S. Holland, J. McCrea, G. Mistry, M. Hesney, L. Xi, S. X. Li, R. Haesen, R. A. Blum, R. L. Lins, H. Greenberg, S. Waldman, P. Deutsch, and J. D. Rogers. 2002. Pharmacokinetics of ertapenem in healthy young volunteers. Antimicrob Agents Chemother 46:3506-11.
113. Marie P, B. X., Schumitzky A, and R.W. Jelliffe 1994. Clinical computations of bacterial growth and kill dynamics-implications for therapy., p. 146. In A. S. f. MIcrobiology (ed.), 34th Interscience Conferene on Antimicrobial Agents and Chemotherapy. , Washington DC.
114. Mathai, D., M. T. Lewis, K. C. Kugler, M. A. Pfaller, and R. N. Jones. 2001. Antibacterial activity of 41 antimicrobials tested against over 2773 bacterial isolates from
158
hospitalized patients with pneumonia: I--results from the SENTRY Antimicrobial Surveillance Program (North America, 1998). Diagn Microbiol Infect Dis 39:105-16.
115. Matzke, G. R., G. G. Zhanel, and D. R. Guay. 1986. Clinical pharmacokinetics of vancomycin. Clin Pharmacokinet 11:257-82.
116. Meagher, A. K., A. Forrest, A. Dalhoff, H. Stass, and J. J. Schentag. 2004. Novel pharmacokinetic-pharmacodynamic model for prediction of outcomes with an extended-release formulation of ciprofloxacin. Antimicrob Agents Chemother 48:2061-8.
117. Meagher, A. K., J. A. Passarell, B. B. Cirincione, S. A. Van Wart, K. Liolios, T. Babinchak, E. J. Ellis-Grosse, and P. G. Ambrose. 2007. Exposure-response analyses of tigecycline efficacy in patients with complicated skin and skin-structure infections. Antimicrob Agents Chemother 51:1939-45.
118. Miller, R., W. Ewy, B. W. Corrigan, D. Ouellet, D. Hermann, K. G. Kowalski, P. Lockwood, J. R. Koup, S. Donevan, A. El-Kattan, C. S. Li, J. L. Werth, D. E. Feltner, and R. L. Lalonde. 2005. How modeling and simulation have enhanced decision making in new drug development. J Pharmacokinet Pharmacodyn 32:185-97.
119. Moise-Broder, P. A., A. Forrest, M. C. Birmingham, and J. J. Schentag. 2004. Pharmacodynamics of vancomycin and other antimicrobials in patients with Staphylococcus aureus lower respiratory tract infections. Clin Pharmacokinet 43:925-42.
120. Mouton, J. W., A. Schmitt-Hoffmann, S. Shapiro, N. Nashed, and N. C. Punt. 2004. Use of Monte Carlo simulations to select therapeutic doses and provisional breakpoints of BAL9141. Antimicrob Agents Chemother 48:1713-8.
121. Mouton, J. W., U. Theuretzbacher, W. A. Craig, P. M. Tulkens, H. Derendorf, and O. Cars. 2008. Tissue concentrations: do we ever learn? J Antimicrob Chemother 61:235-7.
122. Mouton, J. W., and A. A. Vinks. 2005. Pharmacokinetic/pharmacodynamic modelling of antibacterials in vitro and in vivo using bacterial growth and kill kinetics: the minimum inhibitory concentration versus stationary concentration. Clin Pharmacokinet 44:201-10.
123. Mouton, J. W., A. A. Vinks, and N. C. Punt. 1997. Pharmacokinetic-pharmacodynamic modeling of activity of ceftazidime during continuous and intermittent infusion. Antimicrob Agents Chemother 41:733-8.
124. Muller-Serieys, C., J. Andrews, F. Vacheron, and C. Cantalloube. 2004. Tissue kinetics of telithromycin, the first ketolide antibacterial. J Antimicrob Chemother 53:149-57.
125. Muller, F. U., D. H. Hunneman, R. Kahles, and G. Hellige. 1993. Investigation of cardiac metabolism using stable isotopes and mass spectrometry. Basic Res Cardiol 88:272-81.
159
126. Muller, M. 2000. Microdialysis in clinical drug delivery studies. Adv Drug Deliv Rev 45:255-69.
127. Muller, M., H. Stass, M. Brunner, J. G. Moller, E. Lackner, and H. G. Eichler. 1999. Penetration of moxifloxacin into peripheral compartments in humans. Antimicrob Agents Chemother 43:2345-9.
128. Muralidharan, G., M. Micalizzi, J. Speth, D. Raible, and S. Troy. 2005. Pharmacokinetics of tigecycline after single and multiple doses in healthy subjects. Antimicrob Agents Chemother 49:220-9.
129. Murthy, B., and A. Schmitt-Hoffmann. 2008. Pharmacokinetics and pharmacodynamics of ceftobiprole, an anti-MRSA cephalosporin with broad-spectrum activity. Clin Pharmacokinet 47:21-33.
130. Murthy B., D. Skee, D. Wexler, D. Balis, I. Chang, D. Desai-Kreiger, and G. Noel. 2007. Presented at the 17th European Conference of Clinical Microbiology and Infectious Disease, Munich, Germany, Mar 31-Apr 3.
131. Murthy B., D. Skee, Wexler D., Balis D., Chang I., Desai-Kreiger D., Noel G. 2007. Presented at the 17th European Congress of Clinical Microbiology and Infectious Diseases Munich, Germany, 31 Mar- 04 Apr.
132. Myers, D. R., J. DeFehr, W. M. Bennet, G. A. Porter, and G. D. Olsen. 1978. Gentamicin binding to serum and plasma proteins. Clin Pharmacol Ther 23:356-60.
133. Nguyen, M., and E. P. Chung. 2005. Telithromycin: the first ketolide antimicrobial. Clin Ther 27:1144-63.
134. Nicolau, D. P., and P. G. Ambrose. 2001. Pharmacodynamic profiling of levofloxacin and gatifloxacin using Monte Carlo simulation for community-acquired isolates of Streptococcus pneumoniae. Am J Med 111 Suppl 9A:13S-18S; discussion 36S-38S.
135. Nicolau, D. P., C. D. Freeman, P. P. Belliveau, C. H. Nightingale, J. W. Ross, and R. Quintiliani. 1995. Experience with a once-daily aminoglycoside program administered to 2,184 adult patients. Antimicrob Agents Chemother 39:650-5.
136. Nielsen, E. I., A. Viberg, E. Lowdin, O. Cars, M. O. Karlsson, and M. Sandstrom. 2007. Semimechanistic pharmacokinetic/pharmacodynamic model for assessment of activity of antibacterial agents from time-kill curve experiments. Antimicrob Agents Chemother 51:128-36.
137. Nightingale, C. H. 1997. Pharmacokinetics and pharmacodynamics of newer macrolides. Pediatr Infect Dis J 16:438-43.
138. Nix, D. E., K. R. Matthias, and E. C. Ferguson. 2004. Effect of ertapenem protein binding on killing of bacteria. Antimicrob Agents Chemother 48:3419-24.
160
139. Noel, G. J., K. Bush, P. Bagchi, J. Ianus, and R. S. Strauss. 2008. A randomized, double-blind trial comparing ceftobiprole medocaril with vancomycin plus ceftazidime for the treatment of patients with complicated skin and skin-structure infections. Clin Infect Dis 46:647-55.
140. Noel, G. J., R. S. Strauss, K. Amsler, M. Heep, R. Pypstra, and J. S. Solomkin. 2008. Results of a double-blind, randomized trial of ceftobiprole treatment of complicated skin and skin structure infections caused by gram-positive bacteria. Antimicrob Agents Chemother 52:37-44.
141. Novelli, A., S. Fallani, M. I. Cassetta, S. Arrigucci, and T. Mazzei. 2002. In vivo pharmacodynamic evaluation of clarithromycin in comparison to erythromycin. J Chemother 14:584-90.
142. Noviello, S., F. Ianniello, S. Leone, M. Fiore, and S. Esposito. 2008. In vitro activity of tigecycline: MICs, MBCs, time-kill curves and post-antibiotic effect. J Chemother 20:577-80.
143. Odenholt-Tornqvist, I., E. Lowdin, and O. Cars. 1995. Postantibiotic effects and postantibiotic sub-MIC effects of roxithromycin, clarithromycin, and azithromycin on respiratory tract pathogens. Antimicrob Agents Chemother 39:221-6.
144. Odenholt I., E. Lowdin, O. Cars. 2008. Presented at the 48th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC) and the Infectious Diseases Society of America (IDSA) 46th Annual Meeting, Washington DC, October 25-28.
145. Parker, R. F., and S. Luse. 1948. The Action of Penicillin on Staphylococcus: Further Observations on the Effect of a Short Exposure. J Bacteriol 56:75-81.
146. Passarell, J. A., A. K. Meagher, K. Liolios, B. B. Cirincione, S. A. Van Wart, T. Babinchak, E. J. Ellis-Grosse, and P. G. Ambrose. 2008. Exposure-response analyses of tigecycline efficacy in patients with complicated intra-abdominal infections. Antimicrob Agents Chemother 52:204-10.
147. Perret, C., B. Lenfant, E. Weinling, D. H. Wessels, H. E. Scholtz, G. Montay, and E. Sultan. 2002. Pharmacokinetics and absolute oral bioavailability of an 800-mg oral dose of telithromycin in healthy young and elderly volunteers. Chemotherapy 48:217-23.
148. Petersen, P. J., C. H. Jones, and P. A. Bradford. 2007. In vitro antibacterial activities of tigecycline and comparative agents by time-kill kinetic studies in fresh Mueller-Hinton broth. Diagn Microbiol Infect Dis 59:347-9.
149. Preston, S. L., G. L. Drusano, A. L. Berman, C. L. Fowler, A. T. Chow, B. Dornseif, V. Reichl, J. Natarajan, and M. Corrado. 1998. Pharmacodynamics of levofloxacin: a new paradigm for early clinical trials. JAMA 279:125-9.
161
150. Queenan, A. M., W. Shang, M. Kania, M. G. Page, and K. Bush. 2007. Interactions of ceftobiprole with beta-lactamases from molecular classes A to D. Antimicrob Agents Chemother 51:3089-95.
151. Ramadan, M. A., A. F. Tawfik, A. M. Shibl, and C. G. Gemmell. 1995. Post-antibiotic effect of azithromycin and erythromycin on streptococcal susceptibility to phagocytosis. J Med Microbiol 42:362-6.
152. Rayner, C. R., A. Forrest, A. K. Meagher, M. C. Birmingham, and J. J. Schentag. 2003. Clinical pharmacodynamics of linezolid in seriously ill patients treated in a compassionate use programme. Clin Pharmacokinet 42:1411-23.
153. Retsema, J., A. Girard, W. Schelkly, M. Manousos, M. Anderson, G. Bright, R. Borovoy, L. Brennan, and R. Mason. 1987. Spectrum and mode of action of azithromycin (CP-62,993), a new 15-membered-ring macrolide with improved potency against gram-negative organisms. Antimicrob Agents Chemother 31:1939-47.
154. Reynolds, P. E. 1989. Structure, biochemistry and mechanism of action of glycopeptide antibiotics. Eur J Clin Microbiol Infect Dis 8:943-50.
155. Rouse, M. S., J. M. Steckelberg, and R. Patel. 2007. In vitro activity of ceftobiprole, daptomycin, linezolid, and vancomycin against methicillin-resistant staphylococci associated with endocarditis and bone and joint infection. Diagn Microbiol Infect Dis 58:363-5.
156. Rybak, M. J. 2006. The pharmacokinetic and pharmacodynamic properties of vancomycin. Clin Infect Dis 42 Suppl 1:S35-9.
157. Rybak, M. J., D. M. Cappelletty, T. Moldovan, J. R. Aeschlimann, and G. W. Kaatz. 1998. Comparative in vitro activities and postantibiotic effects of the oxazolidinone compounds eperezolid (PNU-100592) and linezolid (PNU-100766) versus vancomycin against Staphylococcus aureus, coagulase-negative staphylococci, Enterococcus faecalis, and Enterococcus faecium. Antimicrob Agents Chemother 42:721-4.
158. Sauermann, R., G. Delle-Karth, C. Marsik, I. Steiner, M. Zeitlinger, B. X. Mayer-Helm, A. Georgopoulos, M. Muller, and C. Joukhadar. 2005. Pharmacokinetics and pharmacodynamics of cefpirome in subcutaneous adipose tissue of septic patients. Antimicrob Agents Chemother 49:650-5.
159. Schentag, J. J., and C. H. Ballow. 1991. Tissue-directed pharmacokinetics. Am J Med 91:5S-11S.
160. Schmidt, S., K. Rock, M. Sahre, O. Burkhardt, M. Brunner, M. T. Lobmeyer, and H. Derendorf. 2008. Effect of protein binding on the pharmacological activity of highly bound antibiotics. Antimicrob Agents Chemother 52:3994-4000.
161. Schmitt-Hoffmann, A. 2004. Presented at the 14th European Congress of Clinical Microbiology and Infectious Diseases, Prague, Czech Republic, May 1-4.
162
162. Schmitt-Hoffmann, A., M Harsch, M Heep. 2004. Presented at the 14th European Congress of Clinical Microbiology and Infectious Diseases, Prague, Czech Republic, May 1-4.
163. Schmitt-Hoffmann, A., L. Nyman, B. Roos, M. Schleimer, J. Sauer, N. Nashed, T. Brown, A. Man, and E. Weidekamm. 2004. Multiple-dose pharmacokinetics and safety of a novel broad-spectrum cephalosporin (BAL5788) in healthy volunteers. Antimicrob Agents Chemother 48:2576-80.
164. Schmitt-Hoffmann, A., B. Roos, M. Schleimer, J. Sauer, A. Man, N. Nashed, T. Brown, A. Perez, E. Weidekamm, and P. Kovacs. 2004. Single-dose pharmacokinetics and safety of a novel broad-spectrum cephalosporin (BAL5788) in healthy volunteers. Antimicrob Agents Chemother 48:2570-5.
165. Schuck, E. L., A. Dalhoff, H. Stass, and H. Derendorf. 2005. Pharmacokinetic/pharmacodynamic (PK/PD) evaluation of a once-daily treatment using ciprofloxacin in an extended-release dosage form. Infection 33 Suppl 2:22-8.
166. Schuck, E. L., M. Grant, and H. Derendorf. 2005. Effect of simulated microgravity on the disposition and tissue penetration of ciprofloxacin in healthy volunteers. J Clin Pharmacol 45:822-31.
167. Shaw, J. P., J. Seroogy, K. Kaniga, D. L. Higgins, M. Kitt, and S. Barriere. 2005. Pharmacokinetics, serum inhibitory and bactericidal activity, and safety of telavancin in healthy subjects. Antimicrob Agents Chemother 49:195-201.
168. Sheiner, L. B. 1997. Learning versus confirming in clinical drug development. Clin Pharmacol Ther 61:275-91.
169. Shi, J., G. Montay, and V. O. Bhargava. 2005. Clinical pharmacokinetics of telithromycin, the first ketolide antibacterial. Clin Pharmacokinet 44:915-34.
170. Sievert, D. M., J. T. Rudrik, J. B. Patel, L. C. McDonald, M. J. Wilkins, and J. C. Hageman. 2008. Vancomycin-resistant Staphylococcus aureus in the United States, 2002-2006. Clin Infect Dis 46:668-74.
171. Skhirtladze, K., D. Hutschala, T. Fleck, F. Thalhammer, M. Ehrlich, T. Vukovich, M. Muller, and E. M. Tschernko. 2006. Impaired target site penetration of vancomycin in diabetic patients following cardiac surgery. Antimicrob Agents Chemother 50:1372-5.
172. Stahl, M., R. Bouw, A. Jackson, and V. Pay. 2002. Human microdialysis. Curr Pharm Biotechnol 3:165-78.
173. Stahle, L., P. Arner, and U. Ungerstedt. 1991. Drug distribution studies with microdialysis. III: Extracellular concentration of caffeine in adipose tissue in man. Life Sci 49:1853-8.
163
174. Stalker, D. J., G. L. Jungbluth, N. K. Hopkins, and D. H. Batts. 2003. Pharmacokinetics and tolerance of single- and multiple-dose oral or intravenous linezolid, an oxazolidinone antibiotic, in healthy volunteers. J Antimicrob Chemother 51:1239-46.
175. Steiert, M., and F. J. Schmitz. 2002. Dalbavancin (Biosearch Italia/Versicor). Curr Opin Investig Drugs 3:229-33.
176. Suerbaum, S., H. Leying, H. P. Kroll, J. Gmeiner, and W. Opferkuch. 1987. Influence of beta-lactam antibiotics and ciprofloxacin on cell envelope of Escherichia coli. Antimicrob Agents Chemother 31:1106-10.
177. Sun, H., E. G. Maderazo, and A. R. Krusell. 1993. Serum protein-binding characteristics of vancomycin. Antimicrob Agents Chemother 37:1132-6.
178. Tam, V. H., A. N. Schilling, and M. Nikolaou. 2005. Modelling time-kill studies to discern the pharmacodynamics of meropenem. J Antimicrob Chemother 55:699-706.
179. Theuretzbacher, U. 2007. Tissue penetration of antibacterial agents: how should this be incorporated into pharmacodynamic analyses? Curr Opin Pharmacol 7:498-504.
180. Tomaselli, F., P. Dittrich, A. Maier, M. Woltsche, V. Matzi, J. Pinter, S. Nuhsbaumer, H. Pinter, J. Smolle, and F. M. Smolle-Juttner. 2003. Penetration of piperacillin and tazobactam into pneumonic human lung tissue measured by in vivo microdialysis. Br J Clin Pharmacol 55:620-4.
181. Tomaselli, F., A. Maier, V. Matzi, F. M. Smolle-Juttner, and P. Dittrich. 2004. Penetration of meropenem into pneumonic human lung tissue as measured by in vivo microdialysis. Antimicrob Agents Chemother 48:2228-32.
182. Traunmuller, F., M. Zeitlinger, P. Zeleny, M. Muller, and C. Joukhadar. 2007. Pharmacokinetics of single- and multiple-dose oral clarithromycin in soft tissues determined by microdialysis. Antimicrob Agents Chemother 51:3185-9.
183. Treyaprasert, W., S. Schmidt, K. H. Rand, U. Suvanakoot, and H. Derendorf. 2007. Pharmacokinetic/pharmacodynamic modeling of in vitro activity of azithromycin against four different bacterial strains. Int J Antimicrob Agents 29:263-70.
184. Turnidge, J. 2003. Pharmacodynamics and dosing of aminoglycosides. Infect Dis Clin North Am 17:503-28, v.
185. Turnidge, J. 1999. Pharmacokinetics and pharmacodynamics of fluoroquinolones. Drugs 58 Suppl 2:29-36.
186. U.S. Department of Health and Human Services, FDA. 2004. Challenge and Opportunity on the Critical Path to New Medical Products.
164
165
187. US Congress Office of Technology Assessment. 1995 Sep. Impacts of Antibiotic-Resistant Bacteria OTA-H-629.
188. Van Bambeke, F., and P. M. Tulkens. 2001. Macrolides: pharmacokinetics and pharmacodynamics. Int J Antimicrob Agents 18 Suppl 1:S17-23.
189. van Ogtrop, M. L., D. Andes, T. J. Stamstad, B. Conklin, W. J. Weiss, W. A. Craig, and O. Vesga. 2000. In vivo pharmacodynamic activities of two glycylcyclines (GAR-936 and WAY 152,288) against various gram-positive and gram-negative bacteria. Antimicrob Agents Chemother 44:943-9.
190. Vogelman, B., and W. A. Craig. 1986. Kinetics of antimicrobial activity. J Pediatr 108:835-40.
191. Vogelman, B., S. Gudmundsson, J. Leggett, J. Turnidge, S. Ebert, and W. A. Craig. 1988. Correlation of antimicrobial pharmacokinetic parameters with therapeutic efficacy in an animal model. J Infect Dis 158:831-47.
192. Vogelman, B., S. Gudmundsson, J. Turnidge, J. Leggett, and W. A. Craig. 1988. In vivo postantibiotic effect in a thigh infection in neutropenic mice. J Infect Dis 157:287-98.
193. Yano, Y., T. Oguma, H. Nagata, and S. Sasaki. 1998. Application of logistic growth model to pharmacodynamic analysis of in vitro bactericidal kinetics. J Pharm Sci 87:1177-83.
194. Zeitlinger, M., R. Sauermann, M. Fille, J. Hausdorfer, I. Leitner, and M. Muller. 2008. Plasma protein binding of fluoroquinolones affects antimicrobial activity. J Antimicrob Chemother 61:561-7.
195. Zeitlinger, M. A., F. Traunmuller, A. Abrahim, M. R. Muller, Z. Erdogan, M. Muller, and C. Joukhadar. 2007. A pilot study testing whether concentrations of levofloxacin in interstitial space fluid of soft tissues may serve as a surrogate for predicting its pharmacokinetics in lung. Int J Antimicrob Agents 29:44-50.
196. Zhanel, G. G., K. Homenuik, K. Nichol, A. Noreddin, L. Vercaigne, J. Embil, A. Gin, J. A. Karlowsky, and D. J. Hoban. 2004. The glycylcyclines: a comparative review with the tetracyclines. Drugs 64:63-88.
197. Zhou, C. C., S. M. Swaney, D. L. Shinabarger, and B. J. Stockman. 2002. 1H nuclear magnetic resonance study of oxazolidinone binding to bacterial ribosomes. Antimicrob Agents Chemother 46:625-9.
198. Zurenko, G. E., B. H. Yagi, R. D. Schaadt, J. W. Allison, J. O. Kilburn, S. E. Glickman, D. K. Hutchinson, M. R. Barbachyn, and S. J. Brickner. 1996. In vitro activities of U-100592 and U-100766, novel oxazolidinone antibacterial agents. Antimicrob Agents Chemother 40:839-45.
BIOGRAPHICAL SKETCH
April Marie Barbour was born March 2, 1982, in Dearborn, MI, to Raymond and Lori
Barbour. She has one sibling, Christina June Barbour, who was born March 15, 1983. April
grew up in Southgate, MI, and graduated high school from Southgate Anderson High School in
2000. She attended Wayne State University in Detroit, MI, from the fall of 2000 through the
summer of 2002 and transferred to the State University of New York at Plattsburgh in the fall of
2002. She graduated from the State University of New York at Plattsburgh in May 2005 with a
bachelor of science in biochemistry. April started graduate school in August 2005 at the
University of Florida, College of Pharmacy, Department of Pharmaceutics. Her advisor
throughout her graduate work was Dr. Hartmut Derendorf. She graduated in May 2009 with a
doctorate of philosophy in pharmaceutical sciences.
166