Citation
Use of Microdialysis as a Tool to Determine Tissue Distribution of Lipophilic and High Molecular Weight Compounds

Material Information

Title:
Use of Microdialysis as a Tool to Determine Tissue Distribution of Lipophilic and High Molecular Weight Compounds
Creator:
SCHUCK, VIRNA JOSIANE AURELIO ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Blood ( jstor )
Dialysis solutions ( jstor )
Dosage ( jstor )
In vitro fertilization ( jstor )
Muscles ( jstor )
Pharmacokinetics ( jstor )
Phosphates ( jstor )
Plasmas ( jstor )
Rats ( jstor )
Skin ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Virna Josiane Aurelio Schuck. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
12/31/2005
Resource Identifier:
71825705 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

USE OF MICRODIALYSIS AS A TOOL TO DETERMINE TISSUE DISTRIBUTION OF LIPOPHILIC AND HIGH MOLE CULAR WEIGHT COMPOUNDS By VIRNA JOSIANE AURELIO SCHUCK A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

This document is dedicated to my mother and to the loving memory of my father.

PAGE 3

iii ACKNOWLEDGMENTS I would like to thank Dr. Hartmut Derendor f for all the guidance and for showing me the importance of pharmacokinetics in therapeutics. I especi ally thank him for trusting me and for giving me the oppor tunity of being part of his group. I would like to thank the members of my committee Dr . Gnther Hochhaus, Dr. Jeffrey Hughes, and Dr. Maria Grant, for the guidance and availability. I also thank Dr. Maria Grant for her help with the clinical study and Dr. Gnther Hochhaus for the help with the assay development. I would like to extent my thanks to Dr. Markus Mller for the opportunity of working with him in the clinical study of dexamethasone and for teaching me the process of microdialysis. I thank CAPES, a Brazilian Organization, fo r granting me the scholarship and for sponsoring my studies at the University of Florida. I thank Mr. James Ketcham for the help with my application to the graduate school and throughout my staying here. I would pr obably not have started graduate school on time if it were not for his help. I thank Pat, Vada, and Andrea, for the help with the paperwork and for the technical support. I thank all my lab mates, especially Sriram, Ping, Ariya, Qi, Victor, Immo, and the post docs Nelamangala, Atul, and Rajanikanth, for the healthy discussions and for their friendship.

PAGE 4

iv I also thank Dr Hochhaus’ students for sharing the lab and all the good moments. I thank the exchange students Ronald, Ire ne, Ralph, and Maria B eatriz for the help with the lab work. I also thank Yufei fo r the help with the development of the radioimmunoassay and LC-MS/MS methods. I finally thank a very special person, my husband Edgar, not only for being the best friend, classmate, and lab mate I could have had in the last four years, but especially for giving me all his support and love throughout this journey.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES.............................................................................................................ix LIST OF FIGURES.............................................................................................................x ABSTRACT......................................................................................................................x ii CHAPTER 1 INTRODUCTION........................................................................................................1 Hypothesis....................................................................................................................2 Specific Aims................................................................................................................2 Specific Aim 1: In Vitro Microdialysis of Docetaxel...........................................2 Specific Aim 2: In Vitro Microd ialysis of Corticosteroids...................................2 Specific Aim 3: Pre-clinical Mi crodialysis of Corticosteroid...............................2 Specific Aim 4: Clinical Microdi alysis of Corticosteroids...................................3 2 REVIEW ON MICRODIALYSIS................................................................................4 Introduction................................................................................................................... 4 Microdialysis Principles...............................................................................................4 Recovery....................................................................................................................... 6 Extrapolation to Zero Flow Method......................................................................8 Slow Perfusion Method.........................................................................................9 Extraction Efficiency.............................................................................................9 Retrodialysis........................................................................................................10 Internal Standard Method....................................................................................11 No-Net-Flux........................................................................................................12 Factors Affecting Recovery.................................................................................13 Membrane characteristics.............................................................................13 Temperature.................................................................................................15 Perfusate.......................................................................................................15 Flow rate.......................................................................................................16 Physico-chemical characteristics of the analyte...........................................16 Tissue physiology.........................................................................................16 Analytical Considerations...........................................................................................19

PAGE 6

vi Microdialysis in Pharmacokinetics.............................................................................20 Applications.........................................................................................................23 Large molecules...........................................................................................39 Other Applications...............................................................................................41 3 IN VITRO MICRODIALYSIS OF DOCETAXEL...................................................43 Introduction.................................................................................................................43 Material and Methods.................................................................................................44 Chemicals............................................................................................................44 Standard Solutions...............................................................................................44 Plasma Preparation..............................................................................................45 Solid Phase Extraction (SPE)..............................................................................45 HPLC System......................................................................................................46 Microdialysis System..........................................................................................46 Microdialysis Experiments..................................................................................47 Extraction efficiency method.......................................................................47 Retrodialysis method....................................................................................47 No-net-flux method......................................................................................48 Binding................................................................................................................49 Results........................................................................................................................ .49 Analytical Method...............................................................................................49 Extraction Efficiency Method.............................................................................51 Retrodialysis Method...........................................................................................51 No-Net-Flux Method...........................................................................................53 Binding Assay.....................................................................................................55 Discussion...................................................................................................................55 4 IN VITRO MICRODIALYSIS OF CORICOSTEROIDS.........................................61 Introduction.................................................................................................................61 Methods and Materials...............................................................................................63 Materials..............................................................................................................63 Standard Solutions...............................................................................................63 Sample Preparation..............................................................................................64 Chromatographic System....................................................................................65 Microdialysis Experiments..................................................................................65 Extraction efficiency method.......................................................................65 Retrodialysis.................................................................................................66 24h Retrodialysis..........................................................................................66 Results........................................................................................................................ .67 Standard Curve Calibration.................................................................................67 Extraction Efficiency Method.............................................................................67 Correlation...........................................................................................................68 Retrodialysis........................................................................................................69 24h Retrodialysis.................................................................................................70 Discussion...................................................................................................................72

PAGE 7

vii 5 PRE-CLINICAL MICRODIALYS IS OF DEXAMETHASONE..............................77 Introduction.................................................................................................................77 Material and Methods.................................................................................................78 Materials..............................................................................................................78 Analytical Method...............................................................................................78 Standard preparation....................................................................................78 Plasma extraction.........................................................................................79 HPLC system................................................................................................80 Animal Procedure................................................................................................80 Blood sampling............................................................................................81 Muscle microdialysis....................................................................................81 Probe calibration..........................................................................................82 Experiment Design..............................................................................................82 Tissue Levels.......................................................................................................83 Pharmacokinetic Analysis...................................................................................83 Results........................................................................................................................ .85 Assay Validation.................................................................................................85 Microdialysis.......................................................................................................86 Non-Compartmental Analysis.............................................................................87 Compartmental Analysis.....................................................................................88 Discussion...................................................................................................................90 6 CLINICAL MICRODIALYSIS OF CORTICOSTEROIDS......................................96 Introduction.................................................................................................................96 Materials.....................................................................................................................9 7 Methods......................................................................................................................98 Human Pilot Study..............................................................................................98 Subjects........................................................................................................98 Study design.................................................................................................98 Microdialysis................................................................................................98 Analytical Method...............................................................................................99 Pharmacokinetic Analysis.................................................................................100 Results.......................................................................................................................1 01 Human Pilot Study............................................................................................101 Discussion.................................................................................................................103 7 CONCLUSION.........................................................................................................106 APPENDIX A DOCETAXEL INDIVIDUAL NO-NET-FLUX PLOTS.........................................108 B INDIVIDUAL PHARMA COKINETICS PARAMETER S OBTAINED IN THE NON-COMPARTMENTAL ANALYSIS OF TH EPLASMA AND MUSCLE PROFILE..................................................................................................................109

PAGE 8

viii C INDIVIDUAL PHARMA COKINETICS PARAMETER S OBTAINED IN THE COMPARTMENTAL ANALYSIS OF TH EPLASMA AND MUSCLE PROFILE..................................................................................................................111 D INDIVIDUAL PLASMA AND FREE MUSCLE PROFILES................................112 LIST OF REFERENCES.................................................................................................114 BIOGRAPHICAL SKETCH...........................................................................................126

PAGE 9

ix LIST OF TABLES Table page 3-1 Average drug concentration measured before and after each experiment and the respective recovery calculated for each experiment.................................................53 3-2 Average perfusate concentration in each no-net-flux experiment............................53 3-3 Concentration of the plasma sample s obtained from the NNF experiments............54 4-1 logP values and molecular weights of th e corticosteroids used in the experiments.69 4-2 Average perfusate concentration, measur ed before and after each retrodialysis experiment, and the respective recovery calculated for each experiment day.........70 4-3 Results obtained in the 2 4h-retrodialysis experiment..............................................71 5-1 Pharmacokinetics parameters obtained in the non-compartmental analysis of the plasma and muscle profiles......................................................................................88 5-2 Average pharmacokinetics (PK) parameters obtained after the fitting of the plasma and microdialysis profiles using the one compartment body model with first order conversion rate.........................................................................................................89 6-1 Dexamethasone pharmacokinetic parameters calculated from the plasma concentration profile obtained after oral administration of a single oral dose of 8 mg...........................................................................................................................10 2 B-1 Pharmacokinetics parameters obtained in the non-compartmental analysis of the individual plasma and musc le profiles obtained after the administration of a 50 mg/kg dose.............................................................................................................109 B-2 Pharmacokinetics parameters obtained in the non-compartmental analysis of the individual plasma and musc le profiles obtained after the administration of a 100 mg/kg dose.............................................................................................................110 C-1 Average pharmacokinetics parameters obt ained after the fittin g of the individual plasma and microdialysis profiles us ing the one compartment body model..........111

PAGE 10

x LIST OF FIGURES Figure page 2-1 Representation of the microdialysis probes available for PK studies........................5 2-2 Scheme of a microdialysis probe inserted in a tissue.................................................6 2-3 The effect of flow rate on the relative (concentration) and abso lute (mass) recovery of dopamine................................................................................................................7 2-4 Effect of membrane length on the re covery of acetaminophen, vanillic acid, and caffeine.....................................................................................................................14 2-5 Theoretical model used to descri be blood and brain concentrations.......................27 2-6 Methotrexate dialysate concentrations in blood and central and peripheral regions of osteosarcom xenographs in rats obta ined after intravenous infusion of a 37.5 mg/kg/3h dose..........................................................................................................32 2-7 Illustration of microdialysis proced ure used for the determination of the transdermal delivery of drug....................................................................................34 2-8 Application of microdialys is in transdermal studies................................................36 2-9 Propanolol concentration in the dermis after iontophoretic ad ministration of two doses, obtained by microdialysis..............................................................................37 2-10 Schematic representation of conventiona l microdialysis and the microdialysis with antibody coated beads adde d to the perfusate..........................................................40 3-1 Docetaxel chemical structure...................................................................................44 3-2 Average dialysate concentration (A) obt ained in each experiment day for each concentration and its respective calculated recovery%R (B) when using extraction efficiency as probe calibration method....................................................................51 3-3 Average dialysate concentration (A) obt ained in each experiment day for each concentration and its respective calcula ted recovery %R (B) when using retrodialysis as probe calibration method.................................................................52 3-4 Average recovery obtained for each c oncentration when two different methods were tested, extraction effi ciency and retrodialysis.................................................52

PAGE 11

xi 3-5 Plot of the net change in the concentr ation between perfusate and dialysate versus the perfusate concentration.......................................................................................54 3-6 Binding to the inlet and outlet t ubing of the microdialysis probe............................55 4-1 Corticosteroids structures.........................................................................................62 4-2 Average recovery obtained for each corticosteroid concentration tested.................68 4-3 Dexamethasone recovery obtained us ing two different probes, CMA/20 and CMA/60....................................................................................................................68 4-4 The correlation obtained when the overa ll averaged recovery obtained for each compound was compared to its logP (r2= 0.0042) or MW (r2= 0.0249)..................69 4-5 Recovery (R%) obtained for all three de xamethasone concentrations tested, when the extraction efficiency (EE) and retr odialysis (RD) met hods are applied.............70 4-6 Dialysate concentration obtained in the retrodialysis experiment of dexamethasone. The bars represent the standard deviation................................................................71 5-1 One compartmental model used to describe dexamethasone total plasma and free tissue profiles obtained after i.v. admi nistration of dexamethasone disodium phosphate..................................................................................................................85 5-2 Dialysate concentration obtained after the iv administration of 50 and 100 mg/kg dose of dexamethasone dissodium phosphate..........................................................86 5-3 Plasma and muscle profiles obtained after i.v. administrati on of 50mg/kg and 100 mg/kg doses to male Wistar rats..............................................................................87 5-4 Average plasma and microdialysis profile s obtained after the i.v. administration of 100 and 50 mg/kg dose of dexamethasone disodium phosphate.............................89 6-1 Plasma profile obtained after oral admi nistration of 8 mg dose of dexamethasone.102 6-2 Dialysate concentration profile obtai ned after the oral of 8 mg dose of dexamethasone, dialysate samples collected from the probe in the muscle, dialysate samples collected from the probe in the skin.........................................................102 A-1 Plot of the net change in the concentr ation between perfusate and dialysate versus the perfusate concentration.....................................................................................108 D-1 Average plasma and free muscle profiles ob tained after the i.v. administration of 50 mg/kg dose of dexamethasone disodium phosphate..............................................112 D-2 Average plasma and free muscle profiles obtained after the i .v. administration of 100 mg/kg dose of dexamethasone disodium phosphate.......................................113

PAGE 12

xii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy USE OF MICRODIALYSIS AS A TOOL TO DETERMINE TISSUE DISTRIBUTION OF LIPOPHILIC AND HIGH MOLECULAR WEIGHT COMPOUNDS By Virna Josiane Aurelio Schuck December, 2004 Chair: Hartmut Derendorf Major Department: Pharmaceutics Microdialysis is a sampli ng technique currently used for a wide range of compounds. However some drugs represent a speci al challenge to successfully apply this technique. It was the objective of this wo rk to evaluate the feasibility of using microdialysis to study tissue distribution of lipophilic and high molecular weight compounds, such as corticosteroids and docetaxel. Initially, the recovery of seven differe nt corticosteroids and docetaxel was determined in vitro. Despite the corticoste roids’ lipophilicity a nd molecular weight, recoveries ranging from 33.3 to 62.2% were obtained. Higher recove ries were observed for flunisolide, while budesonide gave lowe r recoveries. No signi ficant correlation was obtained between the average recovery a nd the compound’s lipophilicity and molecular weight. Docetaxel recovery by gain and by loss ranged from 34.7 to 49.3% and from 53.4 to 64.2%, respectively. The average recovery obtained by no-net-flux was 68.7%.

PAGE 13

xiii The application of microdialys is in the pre-clinical st udy allowed us to determine dexamethasone tissue penetration into the mu scle by comparing the AUC of free muscle and free plasma profiles. The AUC ratio was 1.13+/0.25 and 0.97 +/0.19 for the 50 and 100 mg/kg dose, respectively, which indicates that the drug freely distributes into the muscle. The clinical study wa s carried out to determine dexamethasone absorption and tissue distribution after oral and iontophoretic administrati on. Dexamethasone could be detected in the samples from both probes afte r oral administration; however no drug was detected after topical administration. The results obtained show that it is feasible to apply microdialysis in pre-clinical as well as clinical s ituations to study the pharmacokinetics of dexamethasone.

PAGE 14

1 CHAPTER 1 INTRODUCTION The effectiveness of therapeutics depends on how much drug reaches the site of action. Nowadays, mathematical models are used to predict if drug levels attained in different organs will be inside the therapeutic range or not. However, a better assessment would be to directly measure the drug leve l at the effect site. There are different techniques used to measure drug levels at sp ecific sites, but they do not always provide full information on the degree of drug pe netration. A more informative technique available is microdialysis. Mi crodialysis has been used fo r a wide range of compounds; however, some drugs represent a special challeng e to successfully appl y this technique. Microdialysis (MD) is an established t echnique to study the pharmacokinetics of hydrophilic compounds. However, sampli ng of lipophilic and high protein bound compounds by MD presents spec ial problems. The recovery of compounds such as estradiol and bethamethasone 17-valerate was investigated by MD after topical administration. The estradiol study showed some variability in the re sults, and it was not possible to detect the compound in 8 out of 10 subjects, with probe s positioned at a depth of 1.5 to 10 mm from the skin surface.1 In the bethamethasone study the detection of this compound in the dialysate was only possible af ter the administration of high doses or after maximizing treatment by prepar ing the drug solution in ethanol.2 The lack of detection of both drugs in the dialysates was attributed to the compound’s molecular weight and lipophilicity. On th e other hand, successful results in a microdialysis study of triamcinolone acetonide have been reported fo r both in vitro and in vivo experiments.3,4

PAGE 15

2 The controversial results reported in the literature regardi ng microdialysis of corticosteroids led us to investigate the feas ibility of performing microdialysis for this class of compounds and to investigate if f actors like lipophilicity and molecular weight do interfere in a significant manner with th e microdialysis process. Furthermore, the possibility of using this tec hnique in pre-clinical and c linical experiments was also studied, since the ultimate goal of applying mi crodialysis would be to investigate the compound’s pharmacokinetics at the biophase. Hypothesis This work will test the hypothesis th at the distribution profile and the pharmacokinetics of compounds with high mol ecular weight or lipophilic compounds can be evaluated by the use of microdialysis. Specific Aims Specific Aim 1: In Vitro Microdialysis of Docetaxel Specific aim 1 is to assess the feasibility of doing microdialysis of docetaxel and to determine the in vitro recove ry of this compound by three different methods: extraction efficiency, retrodia lysis and no-net-flux. Specific Aim 2: In Vitro Microdialysis of Corticosteroids Specific aim 2 is to determine the in v itro microdialysis profile of different corticosteroids and to evaluate the effect of lipophilicity and mo lecular weight on the recovery of these compounds. It is also the aim to determine dexamethasone in vitro recovery when different methods and different probe materials are sued. Specific Aim 3: Pre-clinical Mi crodialysis of Corticosteroid Specific aim 3 is to determine the dexame thasone tissue distribution in muscle of male Wistar rats by microdialysis after in travenous administration of dexamethasone

PAGE 16

3 dissodium phosphate, and to compare the free muscle concentration to the free plasma concentration. Specific Aim 4: Clinical Microd ialysis of Corticosteroids Specific aim 4 is to determine dexamethasone tissue distribution in muscle and skin after oral administration and the penetration efficiency af ter topical administration of dexamethasone dissodium phosphate by iontophoresis.

PAGE 17

4 CHAPTER 2 REVIEW ON MICRODIALYSIS Introduction Microdialysis (MD) is a technique used to determ ine the concentration of exogenous and endogenous substances in the extracellular fluid (ECF) of different tissues. It was first mentioned in the literat ure in the early 1970’s, but the analytical methods available at that time made its use difficult. Microdialysis was initially applied to small molecules and hydrophilic compounds, since the use of this technique with lipophilic compounds showed to be more challenging. However, with a better understanding of the factors affecting the proc ess of microdialys is and with the improvement of the analytical methods av ailable, it is possibl e to manipulate the experiment conditions in order to perf orm MD of a wider class of compounds. The following review aims to describe th e microdialysis technique, principles and features, as well as give some examples on how MD can be applied to elucidate a drug’s pharmacokinetics and pharmacodynamics. Microdialysis Principles Microdialysis is performed by introducing a tubular probe in to the tissue of interest. A typical microdialysis probe consists of ri gid concentric tubing with a semipermeable region (membrane), which is permeable to wate r and small molecules. There are different types of probes available, with different fo rmats and membranes materials. The choice of either type is made based on the tissue and compound under investigation (Figure 2-1).

PAGE 18

5 (A) (B) (C) Figure 2-1: Representation of the microdialysis probes available for PK studies. (A) Concentric probe; (B) Lin ear probe; (C) Shunt probe. 5 A physiological buffer solution is continuous ly perfused through the MD probe at a constant flow rate. The fluid entering the pr obe is called perfusate and, after getting in contact with the probe membrane and interact ing with the tissue, the fluid coming out the outlet of the probe is called dialysate. On ce introduced in the tissue, the MD probe function can be compared to that of an arti ficial blood vessel (Figur e 2-2). The perfusate flows through the probe with no net delivery or removal of the fluid from the tissue. Unbound substances present in the tissue surr ounding the MD probe will diffuse through the probe membrane into the perfusate a nd will be carried out of the probe by the constant flow.6,7 The dialysate concentration will repr esent the concentration attained in the tissue under investigation. Any changes in the tissue levels over time will be reflected in the dialysate concentration.8 Microdialysis employs the principle of passive diffusion for drug sampling. Substances with restricted size surrounding the semipermeable region of the probe will passively diffuse down a concentration gradient into the probe lumen, according to Fick’s law. The flux through the membrane is aff ected by the concentration gradient, the molecule’s diffusion coefficient, and the area available for the diffusion to occur.9-11 However, because of the continuous fluid flow inside the probe, equilibration between extracellular fluid (ECF) and the perf usate is incomplete during microdialysis.

PAGE 19

6 Therefore, the dialysate concentration will represent only a fraction of the free concentration observed in the tissue. In orde r to overcome this difference and to obtain the true tissue concentration, the probe recover y, or the fraction of free levels that can be sampled, needs to be determined. Interstitium Capillary Cell Perfusate Dialysate Figure 2-2: Scheme of a microdialys is probe inserted in a tissue.7 Recovery A precise estimation of the recovery is important if one wants to have a good prediction of the true tissue levels. However, the determination of the recovery can be affected by different factors related to the substance, the microdialysis instrumentation, and the physiology of the tissue under investig ation. How each factor can affect recovery will be discussed later. The recovery can be expressed as relativ e or absolute rec overy. The relative recovery is based on the dialys ate concentration while the abso lute recovery is expressed in terms of mass per time. The relative recovery is inversel y proportional to the flow rate, decreasing with higher flow ra te. The absolute recovery, on the other hand, increases with flow rate. The higher the flow rate, the more material is removed per unit of time, the higher the absolute recovery (Figure 2-3).12 The relative recovery of a compound reaches almost 100% at flow rates near to zero, decreasing exponentially with increasing flow

PAGE 20

7 rates.11-16 The relative recovery is used more because it allows the estimation of the tissue concentration, since the outlet concentration will always be a percentage of the tissue concentration.16 Figure 2-3: The effect of flow rate on the re lative (concentration) and absolute (mass) recovery of dopamine.17 Usually, high recoveries are preferred due to the high dialysate concentration that can be obtained. On the other hand, in order to obtain high recoveries, either low flow rates or large collection intervals are required. Low flow rates will result in samples with small volumes, which make samples analysis more difficult since a more sensitive method is required. Long collection interval s will cause a loss in time resolution.5 A good perfusion flow rate is a compromise betw een acceptable recovery, dialysate volume, adequate resolution time and assay sensitivity.12,14 In order to predict the tis sue concentration in a PK study, the recovery should preferably be determined in vivo. The rec overy by a MD probe depends on the diffusion through three different regions: the medium (or tissue), the probe membrane, and the probe lumen.18,19 The slower diffusion region will be the limiting step for the drug recovery. In in vitro experiments or in syst ems were the medium is a flowing fluid such as blood, the limiting step is the drug diffusion through the probe membrane. However,

PAGE 21

8 when performing MD in tissue, the drug diffusion through the ti ssue structures and extracellular fluid will dictate the drug recovery by MD probe.5 Therefore, differences in the recovery can be observed for many compounds depending on whether it was determined in vitro or in vivo.14 The in vitro recovery is usef ul to estimate the MD probe behavior for a specific compound before performing an animal or human experiment. It also provides the maximum recovery that may be expected for a specific compound.16 The recovery can be determined by different methods such as the extrapolation to zero flow method; slow flow method; extraction efficiency ; retrodialysis; internal standard; and the point no -net-flux or zero net flux. Extrapolation to Zero Flow Method The extrapolation to zero flow rate s method requires that the dialysate concentration be determined at different flow rates. Afterwards, the dialysate concentration obtained at each rate is plot ted against the flow rate and by linear regression it is possible to determine the con centration when the flow is zero. At zero flow rate, the dialysate is considered to be in equilibrium to the external medium, therefore, representing the real concentra tion in the probe surrounding. The equation used to describe the relationship betw een dialysate and flow rate is21 F rA dialC C/ 0exp (Equation 2-1) where Cdial is the dialysate concentration at the specific flow rate; C0 is the concentration outside the MD probe (in the so lution or tissue); r is the mass transport coefficient; A is the surface area of the MD membra ne; and F is the flow rate.21 The disadvantage of this method is th e low temporal resolution, since large collection times are required for low flow ra tes, in order to increase the dialysate

PAGE 22

9 volume.12 Another disadvantage of this method is related to outliers a nd variability of the dialysate concentration. Data w ith high variability make it diff icult to precisely estimate the C0.21 Slow Perfusion Method The slow perfusion method is based on the fact that recovery is dependent on the flow rate. At very slow flow rates, the contact time between perf usate and extracellular fluid increases and the exchange between MD and tissue comes very close to the equilibrium, resulting in a dial ysate with concentrations similar to the free concentration in the tissue.16 At very low flow rates (<50nl/min) it is assumed that the drug extraction is higher than 95% for small molecules.5,21 The problem with this method is the lo w volume of the samples, which makes analysis more complicated, and evaporation during the sample colle ction, which becomes a significant problem with very low dialys ate volume. The dead volume in the probe tubing can also affect recover y, since it might cause a signifi cant dilution of the dialysate during sampling.21 Extraction Efficiency The extraction efficiency (EE) method or recovery by gain consists of sampling compound from a matrix with know n analyte concentration. The in vitro set up of this method consists in placing the MD probe in to a tube containing a solution with known concentration of the analyte. The probe is then perfused with a blank physiologic solution. Drug molecules surrounding the membrane will diffuse from the solution into the probe lumen, and will be collected in the probe outlet. The relationship between the drug concentration in the dialys ate and in the matrix is considered the recovery or the

PAGE 23

10 fraction of drug extracted from the solution.14 This method is very simple and represents the in vivo situation when the drug is present in the tissue and diffuses into the probe. Since it is important to know the exact c oncentration in the solution, this method is only used to determine the in vitro recover y. Initially, it was assumed that the membrane resistance was the major step affecting recove ry and not the tissue resistance. Therefore, in vitro determination of the recovery woul d be enough to calibrate the probe. However, it is known now that the tissue tortuosity is th e rate limiting step in the in vivo recovery, and that the in vitro and in vivo r ecovery might not correlate well. On the other hand, the in vitro recovery is useful to determine probe performance, especially for self-made probes, and to es timate the recovery of a specific compound before performing any in vivo experiment. Retrodialysis In the retrodialysis method (RD) or calibra tion by delivery the substance of interest is added to the perfusate medium and its lo ss to the matrix is determined. The ratio between the substance loss to the medium , determined by calculating the difference between its concentration in th e perfusate and dialysate, a nd the perfusate concentration is used to calculate the recovery. This me thod is frequently used to determine the recovery in vivo, because it is not time cons uming and it can be performed for the same animal and probe, either before or after the experiment. In either way, it is assumed that the recovery remains constant throughout the experiment a nd that no tissue changes occur in this period.16 The RD relies in the reversibility of the passive diffusion process across the membrane. RD represents the opposite process of drug sampling by the probe. In order to use the recovery determined by RD to calcu late the true tissue concentration, it is

PAGE 24

11 assumed that the drug diffusion from either sides of the membrane is the same. The reversibility of the process can be verified in vitro and it is true for a series of compounds.22 Another assumption of this method is that the recovery remains constant throughout the experiment. Recovery changes th at occur during the experiment will not be detected by this method.5 Internal Standard Method The term retrodialysis has also been used in delivery experiments where a compound other than the analyte is added to th e perfusate (the internal standard method). The rate of delivery of the internal standard (IS) to the tissue is used to determine the recovery.5 In this method, first the in vitro delivery of the internal standard as well as the gain of the analyte are concomitantly determined and used to calculate the ratio of both in vitro recoveries. Afterwards, the internal standa rd is added to the pe rfusate and its in vivo recovery is determined. The in vivo recovery as well as the ratio determined in vitro are then used to calculate the analyte concen tration in the tissue by the relationship23 k R Rinvitro A invitro IS, , (Equation 2-2) invivo IS dial A tissue AR k C C, , , (Equation 2-3) where RIS,invitro is the in vitro recovery of the internal standard ; RIS,invivo is the in vivo recovery of the in ternal standard; RA,invitro is the in vitro recove ry of the analyte; CA,tissue is the analyte concentration in the tissue; CA,dial is the analyte concentration in the dialysate; and k is the ratio of the recoveries obtained in vitro for both compounds.

PAGE 25

12 Assumption are made that the recovery of the analyte has a constant known relationship to the delivery of the internal st andard and this relationship is the same under in vitro and in vivo conditions. This method requires a calibrant w ith physical-chemical and biological characteristics similar to the analyte. On the other hand, the IS cannot interfere with the experiment or with the tissue characteristics, and it should be eliminated from the tissue in the same rate as the analyte. The similar properties can be achieved by using either a radiolabeled drug or a compound with similar structure to the analyte.16 No-Net-Flux The no-net-flux method (NNF) is based on determining the mass transport of the analyte across the dialysis membrane as a f unction of perfusate conc entration. In the NNF method, it is assumed that the matrix surroundi ng the probe has a constant concentration of the analyte (steady state). Different drug concentrations are added to the perfusate, above and below the drug concentration in the matrix, and are pumped through the probe at different times. When the perfusate con centration is below the concentration in the fluid around the probe, there will be a gain of drug towards the probe lumen. On the other hand, when the perfusate concentration is high er than outside the pr obe, there will be a loss of drug towards the external media, resu lting in a lower dialys ate concentration. The observed net change in the dialysate concen tration is plotted ve rsus the perfusate concentration. The slope of th is plot, calculated by regressio n, is the probe recovery and the x-axis intercept of the curve is the point of no-net-flux (where th e concentration in the matrix and perfusate are identical).16 This method is considered more accurate than the other methods because it estimates the recovery by lost and by ga in, depending on the concentration gradient

PAGE 26

13 across the probe membrane.12,24 However, it has the disadvantage of being time consuming, requiring the calibration to be perf ormed in different days with different animals. Therefore, it is not commonly used to calibrate the probe in vivo especially in pharmacokinetics studies.12 The precision of this method also is highly dependent on the individual concentration determined and on the number of perfusion concentrations used.5 Factors Affecting Recovery A concern with microdialysis is the eff ect of variable membrane permeability on substance recovery. The recove ry should be independent of drug concentration and it is expected to be constant dur ing experiments. However, th e percentage recovery might change and it should be ordinarily determin ed for each individual pr obe and substance of interest.8 The recovery is influenced by many factor s such as (1) membrane characteristics (length, shape, thickness, diameter, and chemical composition); (2) temperature; (3) perfusate pH and ionic strength; (4) fl ow rate; (5) substance physico-chemical characteristics; and (6) tissue physiology.5,8,25 Membrane characteristics Interactions of the compound to the dialysis membrane have been reported to affect the mass transport through the membrane.15 There are probe membranes made of different material such as polycarbona te, regenerated cellu lose (cuprophan), polyethersulfone, and polyacrylonitrile. While all are hydrophilic, th e polyacrylonitrile probes have a much thicker membrane and are also negatively charged, which causes the recovery of anionic compounds to be sm aller if compared to other materials.12,22

PAGE 27

14 The membrane composition can affect more si gnificantly the in vitro recovery than the in vivo one. Hsiao and coworkers (1990) test ed the in vitro and in vivo recovery of four neurochemical compounds using three di fferent probe membra nes: polycarbonate, polyacrylonitrile, and cuprophan. They s howed that the in vitro recovery/mm2 was different among the different probe membranes. However, the in vivo recovery obtained for all probe membranes was similar fo r three out of four compounds tested.18 The probe recovery is proportional to th e membrane surface area available for drug diffusion. The surface area of a cylindrical membrane is determined by its length and radius. Longer membranes will have higher surface areas for the diffusion to occur, resulting in higher recoveries. The recovery will increase up to a limit, where increasing the membrane length will not result in an increase in the recovery (Figure 2-4).12,15,16,22,26 However, the tissue under investigation will dictate the membrane length, since shorter membranes are preferable for small tissues such as brain. Figure 2-4: Effect of membrane length on the recovery of aceta minophen, vanillic acid, and caffeine.22 The geometry of the probe is another charac teristic that could affect the recovery. Zhao and co-workers (1995) studi ed the effect of linear and flexible (concentric) probes

PAGE 28

15 on the recovery of a series of compounds. As a result, the recovery of linear probes was consistently higher for all compounds tested. The author concluded that the linear probe has a higher inner volume and lower linear velocity of the perfusate compared to the flexible probe, which resulted in an increased recovery.22 Temperature The temperature has an effect on the di ffusion coefficient of compounds. Because the microdialysis process follows Fick’s la w, changes in the drug diffusion coefficient will result in differences in the recovery. For each increase in C, a 1-2% increase in the diffusion coefficient was observed, resulting in higher recoveries.15 A higher recovery was observed at 37C than at room te mperature for compounds with different lipophilicities.3,15,26,27 Therefore, the in vitro recovery should be carried out at 37C in order to get a better predic tion of the in vivo recovery.16,28 Perfusate The perfusate composition can affect recovery by changing the compound chemical characteristics and tissue physiology. The perfus ate pH can change th e ionization state of the compound, resulting in the presence of a mo re or less ionized form of the compound. The non-ionized or neutral form will have a higher recovery than the charged form.22 The perfusate ionic strength can change the rec overy by changing the tissue ionic composition and osmolarity. These tissue changes can result in changes in the ti ssue tortuosity, either increase or decrease, which winds up cha nging the recovery. De Lange and coworkers (2000) observed a significant difference in the AUC of acetaminophen brain dialysate when isotonic and hypotonic solutions were used, with lower AUC for the hypotonic perfusate.29

PAGE 29

16 Flow rate The flow rate has an indirect exponential re lationship to the r ecovery. The recovery decreases exponentially as the flow rate increases.14,22,26 At low flow rates, the recovery is higher because there is more time for th e equilibration between the perfusate and the tissue to occur. However, lower flow rates result in dialysate with lower volume, which might be a problem for the analytical method. Th e change in the flow rate will affect the recovery by delivery and by gain in the same manner for a series of compounds.14,22 Physico-chemical characte ristics of the analyte The uptake into the MD probe is determin ed by physicochemical properties of the substance such as molecular weight, lipophi licity, charge, and shape (configuration). Since the diffusion coefficient is inversely proportional to the molecule size, smaller substances with spherical shape will diffu se more easily through the probe membrane than high molecular weight compounds.12,28 The recovery seems to be dependent also on the compound’s hydrophilicity and lipophilici ty. The probe membranes are generally hydrophilic, which results in a bette r permeability to hydrophilic compounds.26 Tissue physiology In an in vitro experiment, the major factor affecting th e recovery is the membrane resistance, which is determined by the memb rane length and radius. On the other hand, the in vivo recovery is affected by th e tissue resistance to the drug diffusion.18 The geometry of the extracellular space, blood fl ow, metabolic rate, and the properties of the layers through which the subs tances diffuse interfere with the diffusion process through the tissue. Diffusion of hydrophilic substances, for example, will be impeded by impermeable structures and cell membrane (tissue tortuosity).5,9,13 Changes in the blood flow are believed to affect recovery by ch anging the tissue resistance to drug diffusion.

PAGE 30

17 Depending on the capillary permeability of th e drug, increase in the blood flow could increase recovery.16 Song and Lunte (1999) showed th at the recovery of caffeine and acetaminophen in muscle decreased when the animal was submitted to a cardiac arrest. The authors showed that in tissues with a ra pid extracellular fluid to plasma exchange, such as muscle, the recovery will depe nd more significantly on the tissue blood flow. The transport through the tissue will be th e rate-limiting step determining the drug extraction by the MD probe. Biological factors that can affect the drug recovery are tissue tortuosity and volume fraction av ailable for drug diffusion. Tissu e tortuosity refers to the increased path length that a molecule has to pass through in order to reach the MD probe. The volume fraction refers to the restriction of the analyte to the extracellular fluid, which comprises 20% of the to tal volume. The combination of the increased path length and reduced volume fraction tends to lead to a reduced recovery.21 The tissue tortuosity and extr acellular volume can be aff ected by the ionic strength of the perfusate. Usually, the perfusate should have the same ionic composition of the extracellular fluid in the tissue of interest in order to avoid osmotic pressure effects and fluid loss either from the probe or from the ti ssue. However, it is not always possible to simulate the extracellular fl uid. The effects of hypo and hyperosmotic solution in the recovery can be related to differences in the tissue osmotic pressure and extracellular volume fraction. A decrease in the extracellula r volume will result in an increase in the tissue resistance and a decrease in the diffusion coefficient of the analyte in the tissue, which ends up affecting the drug recovery by the probe.16 This change in recovery occurs because the amount of analyt e entering the probe is a f unction of the analyte mass

PAGE 31

18 transport through the matrix (tissue stru cture or fluid), pr obe membrane, and perfusate.16,19 Fortunately, under normal conditions, most of the factors affecting drug diffusion are believed to remain constant throughout the experiment. Thus, the net flux of the compound through the MD membrane can be assumed to be constant and determined only by the concentration gradient in both sides of the probe membrane.14 The biggest assumption here is that the recovery will not change during the experiment period despite any tissue reaction to the MD probe implantation.16 The MD probe insertion can cause tissue edema, trauma, and a series of tissue responses such as increase in the tissue blood flow and gluc ose metabolism, release of inflammatory substances and cells, and rupt ure of membranes, for example, the blood brain barrier. Edema increases the extrace llular fluid, resulting in decreased tissue resistance to the diffusion of molecule s and, therefore, affecting the recovery.16 All these tissue alterations can change th e drug concentration in the tis sue and the recovery as well, leading to unreliable results. The trauma observed after probe in sertion will depend on the target tissue location and on the local r eaction to the insertion. The trauma caused by the probe implantati on in the skin was investigated by Ault and co-workers (1994).30 In this study skin histology was performed at different time points after probe implantation. No alteration in the skin was observed immediately after probe implantation, suggesting no physical damage to the skin. However, an inflammatory response was first observed 6h after probe insertion.30 The trauma caused by the MD probe insertion in the liver was investigated by Davies and Lunte (1995).31 No inflammatory response was observed after impl antation of a simple linear probe in the

PAGE 32

19 liver up to 4h. After this period, a mild inf iltration of inflammatory substances was observed with necrotic tissue being formed 8h after probe implantation. The rigid cannula probe showed more damage to the liver afte r probe implantation than did the other probe types tested.31 Nevertheless, minimal histological change s of the brain have been observed two days after probe implantation. The probe im plantation in the brain caused immediate changes in blood flow and glucose metabolis m nearby the probe site. Formation of antiinflammatory substances was observed in th e first hour after probe implantation. These brain changes were reversed to normal leve ls in an animal allowed to recover for 24h.13,15 Tumor tissue showed no changes after MD probe insertion, while muscle showed some infiltration by inflammatory cells after 6h.32 More sensitive organs such as the brain will show a more severe reaction, and the time required to have the tissue back to its normal levels might be longer than for muscle. In general, a period of 30 to 60 minutes might be enough to decrease the trauma in skin and muscle while in brain it might take 24h before starting the experiment.13,15,33 Analytical Considerations The samples obtained from microdialysis usually have a small volume and low concentration of the analyte, wh ich makes its analysis challenging.12 In some microdialysis experiments, as in brain microdi alysis, a very slow flow rate as well as a short collection interval is required in order to have a good temporal resolution and be able to detect biological ch anges. The temporal resolution will depend on the sensitivity of the analytical method available. The sample concentration can be increased by decreasing the flow rate; however this will re sult in smaller sample volumes. Increasing the collection interval can increase the sample volume, but this will result in loss of

PAGE 33

20 temporal resolution. Therefore, the trade off among perfusate flow rate, recovery, sample volume and concentration, and detection limit of the assay will dictate the temporal resolution that can be achieved in the experiment.5 Several analytical methods have been a pplied to analyze microdialysis samples, such as radioimmunoassay;34 capillary electrophoresis;35-37 and liquid chromatography connected to different detectors such as UV,38,39 fluorescence,40,41 electrochemical,42,43 and mass spectroscopy.44-46 The advantage and disadvantage of each method as well as its application in microdialysis b een discussed in the literature.47 Microdialysis in Pharmacokinetics Microdialysis (MD) was fi rst applied to study the pha rmacokinetics (PK) and drug penetration of compounds in the brain. However, this technique has opened new possibilities to study pharmacokinetic and pharmacodynamic processes in different tissues. There are many MD features that make this technique interesting to be applied in PK studies. MD provides a clean sample, which is pr otein free. Because of the membrane porosity and cut-off, large molecules such as proteins cannot get through the membrane, which results in a process of purification of the samples. Generally, microdialysis provides samples that can be directly anal yzed, with no need of processing it before injecting into the analysis system. The perfusate used in the MD process usually mimics the ECF fluid composition. Because of this similarity, no fluid exchange between the tissue and probe occurs during sampling. Besides, the constant flow inside the MD probe allows continuously sampling from the extracellular fluid (ECF). As a resu lt, there is no limitation in the number of samples that can be taken by MD. The lack of fluid loss during the MD process makes it

PAGE 34

21 possible to sample from different organs at the same time by using multiple probes, which provides the ability of simultaneously monitoring pharmacokinetics of compounds at different sites. The slow flow inside the probe and the similarity of the perfusate to the ECF results in minimal perturbation of the environmen t surrounding the probe membrane. Therefore, sampling is can occur unde r physiological conditions.12,13,25,48 MD can be performed in awake animals, th erefore each animal can be used as its own control in a cross over design, decreasi ng inter-animal variability and reducing the number of animal required to conduct a study. Sampling by microdialysis reduces the stress to the animal and also promotes fewer changes in physiological conditions of the animal, which contributes to the decrease in th e data variability. This design allows also to determine pharmacokinetics following ch ronic drug administration, to study dose dependence of PK parameters, and to dete rmine drug interactions, all in the same animal.12,13,48,49 One big feature of MD is th at it allows determining the free drug concentration in the tissue, which is very important if one c onsiders that only the free drug concentration is pharmacologically active. Because of the me mbrane cut off, only the free drug present in the tissue will be able to cross the pr obe membrane. Free tissue levels can be determined by microdialysis by dividing the co ncentration measured in the dialysate by the recovery obtained duri ng the probe calibration. Microdialysate samples are co llected over a time interval in a continuous process. The dialysate concentration obtained represents the concentration at the mid-point of the collection interval, assuming that the c oncentration in the probe surrounding is

PAGE 35

22 homogeneous. The MD profile should never be represented on time-axis by the end or the starting point of the sampling interval, since the profile will be distorted. For noncompartmental PK analysis, correlating the co llection mid point to the concentration is considered a valid approach if the sampling interval is shorter than the drug half life. Calculation of the slopes and half life is th en similar to standard methods. However, adjustments are necessary when the absorption or elimination half life is shorter than the collection time.29,48,50,51 The calculation of the area under the curve (AUC) for a microdialysis profile may be more accurate than for plasma profile. Th e AUC, usually calculated by the trapezoidal rule for plasma samples, is calculated by simply multiplying the measured microdialysate concentration by the collecti on interval (Equation 2-3).50 The same approach should be used to calculate the area under the first moment curve (AUMC). The other parameters calculated in a non-compartmenta l approach such as cleara nce, volume of distribution, and elimination half life can be ca lculated with standard equations.13,29,52 i i tt C AUC 0 (Equation 2-3) where AUC0 t is the area under the concentration versus time curve; Ci is the concentration of the extracellu lar fluid at the ith sample; ti is the collection interval for the ith sample. Other PK parameters that can be aff ected by microdialysis are the maximum concentration obtained (Cmax) and the time to reach this concen tration (tmax). The determination of both parameters in blood is dependent on sampling frequency. Because MD is a continuous process, the determination of these pa rameters is dependent on the

PAGE 36

23 collection interval. Long collection interval will result in loss of temporal resolution and, therefore, it can mask the real Cmax and tmax.13,29 Applications Microdialysis has been used to study drug fr ee levels in different tissues, such as muscle, adipose tissue, lung, ey e, skin, brain, bile, and blood.53,54 MD is an established technique used to study physiological, pharm acological, and pathological changes of wide low molecular weight substances in th e extracellular space of different tissues, in pre clinical and clinical studies. Amongst the exogenous drugs studied by MD are antibiotics, anti-inflammato ry, anti-tumor, and psycho active compounds. Extensive reviews in MD in PK can be found in the literature.5,13,49,52,54-58 Initially MD was used to investigate drug penetration into the brain, since it allows one to determine drug levels in both sides of a membrane, in this case the blood brain barrier (BBB). Because of the BBB, many drugs are unable to penetrate into the brain. Drug transport through the BBB is depende nt on the drug’s p hysico-chemical characteristics such as lipophilicity, ioni zation and pH. Penetration across the BBB under physiological and disease conditions can be assessed by the use of microdialysis. The microdialysis probe is implanted into a specific zone in the brain by the use of stereotaxic equipment and a guide cannula, which is used to keep the probe in place. By measuring the free drug in the brain and the total or free drug in the blood it is possible to determine the extent of drug penetration into the brai n as well as the mechanisms involved in the restricted drug penetrat ion into this organ.59,60 Brain ECF and blood profiles determined by MD have shown to have the same half life and a similar tmax for many drugs. It is expected that hydrophilic drugs would have a longer equilibration time between brain and blood than lipophilic drugs, which would

PAGE 37

24 result in a higher tmax and longer half life for these compounds. The rate of equilibration is determined by the perfusion rate and by the distribution of the drug into the brain. Despite the equilibration time, many drugs were shown to have a lower free level in the brain when compared to free levels in th e blood. However, some lipophilic drugs showed similar free levels in both tissues.61. The brain distribution of codeine in rats was studied by Xie and co-workers (1998) in a cross over design. In th is study, microdialysis was used to study the free drug levels in blood and brain after intravenous infusi on of two codeine doses. They found that codeine freely distributes into the brain, reaching e qual free levels in blood and brain for both doses studied.60 MD has also been used to study dr ug penetration in to brain tumor.62 Tumors are distinct tissues with specific physiological characteris tic. Brain tumors are difficult to treat not only because of the tumor physio logy but also because of the BBB, which represents a real barrier to drug penetration. Dukic and co -workers (1999) have studied the PK profile of methotraxate in the extracel lular fluid of brain tumors in rats by using MD. C6 glioma cells were used to induce br ain tumor. ECF levels in the tumor were measured by MD and compared to total and free blood levels. The results showed that methotrexate rapidly equilibrates into the tumor, but the free levels were only 1-2% of those measured in blood. The difference was at tributed to the influx or efflux clearances of the compound from the brain.63 On the other hand, methotrexa te free levels in healthy peripheral tissues such as muscle and liver show ed to be similar to the free plasma levels when at steady state.64

PAGE 38

25 The influx and efflux of compounds from br ain can also be determined by the use of microdialysis. The kinetics of 6-mercaptopurine into the brain was studied by Deguchi and co-workers (2000).59 MD was used to measure 6-mercaptopurine free levels in brain after intravenous infusion of a 0.4 mg/kg/min dose. The drug free con centrations in brain obtained by MD were 0.4 times lower than the free concentration in plasma, showing the restricted distribution of this compound in the brain. In order to determine if the difference was caused by the influx or efflux pu mps, the drug clearan ce in and out of the brain was determined. Pharmacokinetic analys is of the data using a two compartment single membrane model revealed that the clea rance out of the brain was almost 20 times higher than the clearance into the brain. The authors concluded that the distribution of 6mercaptopurine in brain is in fact restricted by efflux clearance. In order to further understand the efflux transport of 6-mercaptopurine across BBB, probenecid (organic anion channel inhi bitor) and benzoate (monocarboxylic acid transport inhibitor) were administered by the MD probe, concomitantly with an intravenous administration of 6-mercaptopur ine. The inhibition of these two efflux pumps increased the dialysate levels of 6-mercaptopurine, indicating an accumulation of the compound into the brain. In conclusion, it seems that these two transport systems are involved in the clearance of 6mercaptopurine from the brain.59 Hammarlund-Udenaes and co-workers ( 1997) derived a theoretical model to describe the brain profile and the equilibrati on between blood and brain in terms of the influx and efflux transport of drugs in the brain and used the data obtained by MD to validate the model. The model consisted of a one compartment body compartment and one brain compartment (Figure 2-5). The m odel assumptions were that the volume of

PAGE 39

26 brain is 1% of central comp artment volume; blood concentration is the driving force for brain concentration; and the uptake or elimin ation from the brain did not influenced the blood profile. The equations describing the model are shown below (Equation 2-5 and 26). brain u out blood u in brain u brainC Cl C Cl dt dC V, , , (Equation 2.-5) blood u blood uC Cl R dt dC V, inf , (Equation 2-6) where Vbrain and V are the apparent volumes of dist ribution of the free drug in the brain and body, respectively; Cu,blood is the free concentrations in the blood; Cu,brain is the free concentrations in the brain; CLin and CLout are the diffusional cleara nces in and out of the brain; CL is the clearan ce out of the body; and Rinf is the infusion rate constant. The clearance in (Clin) and clearance out (Clout) of the brain were simulated considering three situat ions: passive influx and efflux from brain; active transport into the brain; and active transport out of the brain. The active tran sport was characterized using Michaelis-Menten equation. With this approa ch, the authors were able to determine situations where the drug concentration and half life in the brain were dependent on the intake or elimination from the brain. In c onclusion, the ability of the drug to cross the BBB is dependent on Clin, but the half life in brain is determined by Clout.61 The model was then applied to explain brai n profiles of different drugs obtained by MD that are reported in the literature. The simulations combined with MD data showed that an active transport acro ss the BBB is present for most of the drugs. As many drugs show a half life in brain sim ilar to the blood half life, it seems that the elimination rate constant from the brain is much bigger than the elimination rate constant from the blood.

PAGE 40

27 For all drug tested, MD showed that equ ilibration between brain and blood occurs rapidly, which makes blood levels a st rong predictor of brain levels. Figure 2-5: Theoretical mode l used to describe blood and brain concentrations (Cu,blood: free concentrations in the blood; Cu,brain: free concentrations in the brain; CLin and CLout: diffusional clearances in and out of the brain; CL: clearance out of the body).61 The study of pharmacodynamic effects of dr ug in the brain is also possible by applying MD. The effect of a systemically administered compound on extracellular levels of neurochemicals such as dopamine, acetylcholine and glutamate, can be determined by MD.65 Brain microdialysis has been helpful in studying specific neurochemical reactions in the brain. The release of catecholamines a nd amino acids has also been studied during brain injury and trauma, ischemia, se izures, and encephalopathy, by applying microdialysis.15,66 Extensive reviews on brain micr odialysis can be found in the literature.65,67,68 Besides the extensive use of MD in brain studies, it has also been used to investigate drug levels in peripheral tissues.54 One class of compounds that have been extensively studied by MD is antibiotics. The selection of an antimicrobial agent and dose regimen to treat an infection is usually based on the minimum inhibitory concentrations (MIC) and on serum levels. Howe ver, this approach may only be valid for cases where the central compartment is th e main infection site. Nevertheless, for infections in peripheral compartments, the PK profile in the tissue determines the clinical outcome of antimicrobial therapy. The antibioti c levels in the tissue have been estimated

PAGE 41

28 by extrapolation from algorithms based on plasma levels. However, the algorithm assumes that the free levels in the central and peripheral compartments are equal. The direct measurements of free levels in inters titial space of peripheral tissues could provide a more appropriate approach to treat infec tions, since it would allow to determine the compound’s distribution kineti cs and could also reveal if free levels in both compartments are identical or not. The comparison of values obtained by MD with classical blood sampling provides a mean to ev aluate whether or not plasma values would be sufficient to predict tissue levels. MD is a technique that allows determini ng free levels at the site of action and, therefore, can help to establish a more e fficacious therapeutics. By using MD, a free profile in the infection site can be obtaine d. If the AUC of the pr ofile obtained in the tissue (AUCtissue,free) is compared to the free profile in the central compartment or plasma (AUCplasma,free) , the compound’s penetration into the site can be estimated. MD has been used to study the penetration of antibiotics such as gentamicin,69 cephalosporin,55,70,71 macrolides,55,71 penicillin,6,55,71,72 and quinolones55,71,73-75 into various tissues. It was observed that the free concentra tion of some antibiotics in a healthy tissue is less than the free concentration in plasma, as for fleroxacin, dirithromycin,71 cefaclor,76 and cefpodoxime.70 Therefore, an overestimation of th e free concentrations in the tissue occurs when plasma levels are used to es timate the concentration of these antibiotics, which could lead to the development of bacterial resistance. On the other hand, piperacillin have shown to have similar fr ee levels in muscle and plasma in healthy animals.77

PAGE 42

29 However, the antibiotic distribution in ill pa tients, as in the case of surgery, septic shock, and inflammation, can be affected by tissue alterations th at occur in these situations, such as changes in the hemodyna mics, capillary permeab ility and extracellular space volume. The distribution of antibiotics in ill patients has been studied by Joukhadar and co-workers (2001). When the concentratio ns of piperacillin in muscle and adipose interstitial space of healthy subj ects and patients with septic shock were compared, a 5 to 10 fold lower penetration was observed for both tissues in patients with septic shock. The tissue concentrations observed were below the MIC observed for an infective bacterium such as Staphylococcus aureus .6 The authors attributed this difference to a decreased in the blood flow and peripheral tissue perfusion observed in patients with septic shock. Despite the lower penetration (AUCtissue,free/AUCplasma,free), the piperacillin half life in patients was longer for all tissues studies. Th e longer half life coul d cause the antibiotic to stay longer at the infecti on site, resulting in longer times above the MIC. Since for beta-lactams the time above the MIC is the predictive parameters for clinical outcome, the higher half life could compensate for the observed low target concentrations. Another study performed with post operative and intensive care patients showed a similar result. Piperacillin penetration in mu scle and subcutaneous tissues was lower in patients than in healthy volunteers. The con centrations obtained for patient were below the MIC of strains with MIC50>20 mg/l. However, the half life in both tissues was longer for patients than for healthy subjects.72 The tissue PK parameters obtained in this study were used to simulate the antimicrobial effect of piperacillin in time kill curves experiments. The in vitro simulation showed that effective killing was reached when the PK parameters of the patients were use d, despite the observed low concentrations.78

PAGE 43

30 Other studies using MD to st udy antibiotic distribution in healthy and ill subjects can be found in the literature.73-75 MD is also useful in studying the dis position of anticancer agents and their metabolites in tumors.56,58,62 Tumors represent an especially important therapeutic target and one for which blood sampling may not provid e particularly suitab le information with respect to target tissue concentration. The pathophysiological charac teristics of a tumor represent a barrier for the delivery and eff ectiveness of cytotoxic therapy, which might lead to subtherapeutic levels a nd, consequently, treatment failure.56 Physiological barriers that develop during tumor formation comp romise local vessels, altering blood flow, vessel organization (dimension and path length), resistance, and permeability. The interstitial space composition, volume, and pressu re are also different in a tumor, which consequently affects molecule diffusion and convection through the ex tracellular space of a tumor.79 Therefore, a better predictor of an ticancer effects would be the direct measurement of drug levels and exposure on tumor tissue. Information regarding the disposition and metabolism of anticancer drugs in solid tumors may in part explain the variability in tumor response and toxicity. The use of MD in tumor tissues would allow not only investigating the drug PK and metabolism, but also studying the eff ect of multidrug resistance protein and angiogenesis inhibitors on an anticancer dr ug. The drug transfer process from the blood to the interstitial space of a tumor is consider ed a critical step in clinical resistance of tumors. The interstitial space represents th e compartment of immediate contact with tumor cells, and drug concentration in this space is related to the anticancer effect. MD is

PAGE 44

31 a technique that allows sampling from the inte rstitial space, and, th erefore, would allow the estimation of tumor exposure to anticancer drugs. The PK of anticancer drugs such as cisplatin and carboplatin;80 5-fluoracil; 81 and methotrexate63,82 in the tumor interstitial fluid wa s investigated by MD. For all these compounds, a difference in the free serum con centration and the free tumor concentration obtained by MD was observed. The results obta ined in the platinum study showed a high variability in the drug penetra tion into the induced subcutan eous breast tumor, with an average AUCtumor/AUCplasma ratio of 0.62 and 0.82 for cisplatin and carboplatin, respectively.80 The variability in the results is at tributed to the tumor heterogeneity, including differences in vasc ularity, density, and tortuosit y, which end up affecting drug distribution into the infected tissue. The methrotexate study showed higher free levels in the femur fibrous histiocytoma than in th e blood, which was attributed to the tumor physiology such as vascularization and necrosis.82 The methotrexate distribution inside a tu mor was also investigated by MD in an animal model.83 Ekstrom and co-workers (1997) implanted two MD probes in a subcutaneous osteosarcom xenograph, one in th e periphery and anothe r in the center of the tumor, and measure methotrexate levels at both sites simultaneously after its intravenous infusion. Methotrexate concentratio ns at the central part were significantly lower than the concentration measured at the periphery of the tumor. After the infusion was stopped, the situation invert ed, with higher levels observed in the central part of the tumor (Figure 2-6). The difference in methotre xate levels were attributed to a higher degree of vascularity in the pe ripheral region of a tumor, wh ich allows drug distribution and elimination to occur more rapidly, and to a higher protein binding in the central

PAGE 45

32 region of the tumor. The study not only show s the difference in distribution inside the tumor, but also demonstrates that the dr ug efflux from a tumor can be influenced by angiogenesis and necrosis thr ough the tumor. However, the AUC0-6 calculated for the central and peripheral region showed no significant differe nce, which might indicate a similar exposure level of both areas to methotrexate. Figure 2-6: Methotrexate dial ysate concentrations in bl ood and central and peripheral regions of osteosarcom xenographs in ra ts obtained after intravenous infusion of a 37.5 mg/kg/3h dose.83 MD has also been used to investigate an ticancer distribution into the tumor when it is administered in combination with angioge nesis inhibitors. Angi ogenesis is a process wherein endothelial cells divide and migrate to form new bl ood vessels. This process is required for solid tumors to grow and metast asize. Combination of anticancer drug with angiogenesis inhibitors represents a new ther apeutic strategy in cancer, which target tumor cell and endothelial cells or the proce ss of angiogenesis. However, angiogenesis inhibitors can also cause a decrease in capil lary permeability and, therefore, decreasing the anticancer agent distribution into the tumor.62

PAGE 46

33 The effect of the angiogenesis inhibito r TNP-470 on the tumor distribution of the anticancer drug temozolomide was investig ated by Devineni and co-workers (1996).84 Rats implanted with C6 glioma cells rece ived TNP-470 for 14 days previous to the temozolomide administration. Following the tr eatment, a linear MD probe was inserted into the tumor and temozolomide (40 mg/ kg) was administered intraarterially. The temozolomide AUC obtained by MD showed a 25% decrease for the treated group in comparison to the control group, which did not receive the angioge nesis inhibitor. Another area of application of microdialys is is in transdermal studies. Transdermal delivery systems have many advantages over other dosage forms. Basically, it allows controlling the drug delivery for a longer peri od of time; stopping administration at any time; and improving the dose administrated by modifying the properties of the biological barrier. The topical administration of drugs shows decreased fluctuation in the plasma concentration, which is important for drugs w ith a narrow therapeutic window. It is also an alternative to improve patient compliance to drugs with a short elimination half-life.85 Furthermore, when systemic levels are related to side effects and a local or topical effect is desired, the administration through the skin allows obtai ning a high concentration at the site of action, but still a low systemic concentration avoiding side effects. To establish the validity of transderma l drug therapy, it is required convincing documentation and results that show that transdermal absorption occurs in amounts sufficient to achieve therapeutics levels in the target tissue. Information on transdermal delivery and skin concentrations was obtained fr om in vitro delivery studies or in vivo by biopsies, suction blisters, or skin stripping methods.28 However, these in vivo methods cannot provide a full profile and are too trau matic to be used in disease skin. The

PAGE 47

34 measurement of plasma levels after transder mal application is acceptable to determine the efficacy of transdermal systems when a systemic effect is desired. Nevertheless, when a local effect is the objective, it would be id eal to measure drug concentrations in the stratum corneum or dermis, because blood concentrations could be undetectable or not relevant.86,87 MD allows sampling from a very superfic ial layer such as the stratum corneum (SC) and other layers of the skin. One or more MD probes can be inserted in the skin at different depths and the transdermal delivery system can be applied right above the probe window (Figure 2-7). As the drug is released fr om the system, it gets in contact with the skin and starts to move down the skin laye rs till it reaches the probe membrane and is sampled by the MD system. The advantage of us ing this technique in evaluation of skin penetration is the possibility of obtaining a full concentration drug profile in the SC under the same area of drug application. So, variations in penetration due to differences in skin characteristics are eliminated. In addition, this method is suitable for measuring SC concentrations at the same site after multiple dose administration, and it is suitable to measure penetration depth of a compound beyond the SC.88,89 Figure 2-7: Illustration of microdialysis pr ocedure used for the determination of the transdermal delivery of drug.1

PAGE 48

35 Because topical products deliver the drug directly to or near the intended site of action, measurement of the drug uptake into a nd its elimination from the SC by applying MD can provide a way of assessing tissue e xposure to topical drug products. Even when the target site is not the SC, the drug still has to pass through the SC, the rate limiting step in transdermal absorption, to reach the biophase and have action. Therefore, SC concentration profiles are assumed to be di rectly related to the concentration-time profiles of the active substance in the epidermis and dermis.87 Mller and co-workers (1997) showed the role of the SC as a ra te-limiting step in transdermal absorption by using MD. In this study, two MD probes were inserted in the vol unteer thigh, one under the corium and another right below the first pr obe, in the deep subc utaneous layer. After topical administration of diclofenac, no penetr ation into the deeper tissues was observed if the drug concentration in the more superficial probe was undetectable.90 The position of the MD probe in the skin or the probe depth should be determined when performing a transdermal MD study. If th e position is not reproducible, it could be the reason for some variabil ity observed in the result.89 The probe depth in transdermal studies has been reported to be in a range of 0.3 11.7 mm.89,90 Another variability of the results could be caused by the position of th e drug delivery device or the application area relative to the probe window. A small di splacement between the MD probe and the application area could cause changes in the recovery.89 Determination of the probe depth not only co ntributes to justify variability in the results, but also introduces another dime nsion in transdermal microdialysis. The penetration depth of nicotine after topical administration was investigated by applying MD. A MD probe was inserted in the sk in of volunteers and the probe depth was

PAGE 49

36 measured by high frequency ultrasound. Nico tine was administered as a transdermal system (21 mg/24 hr) and MD samples were collected at specific time points. The nicotine profiles obtained by MD were groupe d according to the pr obe depth (Figure 28). It could be observed that deeper laye rs were less exposed to nicotine. A positive correlation between the AUC in the sk in and the probe depth was obtained.1 MD not only permitted to determine the penetration dept h of nicotine through the skin, but also allowed to determine when steady sate flux wa s attained in the different skin layers. (A) (B) Figure 2-8: Application of microdialysis in transdermal studies. (A) Nicotine skin penetration after topical administration of a transdermal delivery system of 21mg/24h. (B) Correlation between the AUC obtained for different skin layers and the probe depth.1 One big feature of MD in transdermal st udies is the possibi lity of studying drug absorption through the skin when different a pplication devices such as iontophoresis are used to deliver the compound. I ontophoresis is a delivery syst em that uses current to facilitate the introduction of charged substa nces into the skin. It also enhances the delivery of uncharged solutes by the process of electro-osmosis, where the electrically driven water molecules act as a carrier fo r neutral molecules. Iontophoresis has the advantages of a transdermal system and, in addition, facilitates the delivery of compounds that cannot be delivered passively.91 It has been successfu lly used to deliver drug molecules such as propanolol89 and proteins91.

PAGE 50

37 The effective delivery of propanolol through the skin by iontophoresis was investigated by MD. Two linear microdialysis probes were in serted in parallel in the forearm skin (1mm deep) of volunteers, one probe received the drug by iontophoresis and another probe was used as a control site, wh ere no current was applied. The iontophoretic chamber containing the drug was placed right above the probe window and a 2mA current was applied for two periods of 1 hour . Despite the observed variability in the data, the average profile obtained by MD clear ly shows two peaks co rrespondent to the two doses applied (Figure 2-9).89 Figure 2-9: Propanolol concentr ation in the dermis after io ntophoretic administration of two doses, obtained by microdialysis.89 The process of iontophoresis per se can cause alterations in the skin structure such as blood supply, enzyme activity, and pH. It also causes some tingling sensation, irritation, and erythema.91 These alterations could affect the process of microdialysis, affecting drug recovery throughout the experiment. The effect of changes in blood flow after iontophoresis on the recove ry of sodium fluorescein wa s investigated by Stagni and co-workers.10 The iontophoresis current increased in more than 500% the skin blood flow measured by laser Doppler flowmetry. Howeve r, the recovery of sodium fluorescein determined by retrodialysis was the same with or without current. This study shows that

PAGE 51

38 the electrical current applied has no effect on the recovery of this compound despite the observed changes in blood flow.10 The effect of other skin perturbations su ch as irritation and delipidization in the absorption of drugs applied topically can also be investigated by MD. The skin permeability changes as the skin is exposed to the compound and formulation, either due to the healing process of the di sease or due to changes in skin characteristics, as water and lipid content and thickness. Removal of skin layers by the popular tape stripping method or the dermatitis promoted by applying sodium lauryl sulfate caused an increase in the absorption of salycili c acid of more than 80 fold when compared to normal skin.92,93 The study not only shows how changes in the skin due to diseases or drug formulation can affect drug absorption th rough the skin, but also demonstrates how feasible MD is to study transdermal de livery of compounds under disease conditions. The bioequivalence (BE) of topical compounds is another area of application of MD.94 The onset, duration, and magnitude of action of topical drug products depend on the release of the drug from the dosage form, its penetration through the skin barrier, and the generation of the desired pharmacological effect. Because topical products deliver the drug directly to or near the intended site of action, measurement of the drug uptake into and its elimination from the SC can provide a way of assessing BE of topical drug products. Two topical formulations that produc e comparable SC concentration profiles may be equivalent, just as tw o oral formulations are judged bioequivalent if they produce comparable plasma concentration profiles, esp ecially if one considers that the SC is the rate limiting step for drug absorption.95 Formulation with different levels of penetration

PAGE 52

39 enhancers showed an increase in the transdermal delivery of cyclosporine96 and valproate,97 as observed by a transdermal MD study. Transdermal MD has also been used to determine the transdermal penetration of other compounds such as diclofenac afte r single and multiple administration;88,90,98 methotrexate;99 of valproate;97 ciclosporin; 96 5-fluorouracil;30 ibuprofen;100 and ketonazole and miconazole.101 Large molecules In order to apply MD to study larg e molecular weight compounds, some modifications in the techni que are necessary. Large or lipophilic compounds show a lower diffusion coefficient due to a greater mass transport resistance, which will result in a decreased recovery for these type of co mpounds. In order to improve recovery, the probe membrane with higher cut off should be used to facilitate the diffusion of larger molecules through it. However, a recovery of only 1-5% has been observed for 10kDa or larger proteins when a membrane with a cut off of 100kDa was used. Membranes with larger cut off such as 300 kDa gave a recovery of 30% for interleukins.102 On the other hand, higher membrane cut off would allow othe r interferences present in the matrix to pass through the membrane, resulting in a dirty sample, which would require some preparation before analysis. Another way of improving drug recovery w ould be using a perfusate with better affinity for the compound. The drug affinity or solubility for the perfusate can be improved by adding albumin,103 cyclodextrines104 or any other compound that has some binding affinity for the drug. Ao and co-wor kers (2004) enhanced the recovery of cytokines (tumor necrosis factor and interleukins) by adding a antibody coated microspheres to the perfusate.102 The moment the cytokines diffuses through the probe

PAGE 53

40 membrane, they will get in contact to the antibodies present in the perfusate and will rapidly bind to it. Because of the high affinity of this binding, the cytokines will be trapped inside the probe lumen and will be carried away by the flow (Figure 2-10). The presence of these antibodies caused the cytoki ne concentration inside the probe to be almost zero, increasing the diffusive drivi ng force across the membrane (concentration gradient), which resulted in a four to twelve fold increase in the recovery. Figure 2-10: Schematic representation of conventional microdialysis and the microdialysis with antibody coated beads added to the perfusate.102 However, the modifications mentioned a bove, increased membrane pore size and a modified perfusion fluid, permit fluid loss fr om the probe, which ends up affecting the tissue physiology and drug recovery. In order to overcome the fluid loss, the osmotic pressure between perfusate and ECF can be compensated by adding serum albumin or dextran-70 to the perfusate.102,103 Another factor interfering w ith the microdialysis of large molecules is the binding of these compounds to the probe tubing a nd membranes. This phenomenon has been observed for the neuropeptides -endorphin, neurotensin, a nd neuropeptide Y (NPY). The recovery of NPY showed to be very sma ll, below 2%, due to binding of the peptide

PAGE 54

41 to the membrane.105 However, by adding 0.5% human seru m albumin to the perfusate, it was possible to increase its in vitro recovery up to 13 to 16%.106 MD has been used to study the re lease of neuropeptides in muscle;106 skin;107 and brain.108,109 The effect of typical (haloperidol) and atypical (risperidone) antipsychotic drugs on the release of NPY in the brain was investigated by MD. NPY is a 36-amino acid peptide widely distributed in the centr al and peripheral nervous system, which is believed to play a role in schizophrenia. The chronic treatment with haloperidol and risperidone resulted in a decreased NPY levels in the striatum. Hal operidol also increased NPY levels in the hypothalamus and occipital cortex. The results show the dopaminergic regulation of NPY and its region dependency.109 The use of MD to study other proteins and peptides can be found in the literature.110-112 Other Applications Besides the well established use in pharm acokinetics (PK), MD has also been used in pharmacodynamic (PD) and PK/PD studies.113 The drug effect can be determined by MD as well by evaluating drug i nduced versus basal concentr ations of a substance or biomarker. The effect of theophylline and milrinone on phosphodiesterase activity in the muscle, determined by measuring cAMP le vels, was investigated by applying MD. A MD probe was used to simultaneously sample drug and cAMP from the muscle. The free drug concentration obtained was directly correlated to the effect. Milrinone showed to increase cAMP level, or decrease the phos phodiesterase activity, in a concentration dependent manner. On the other hand, the theophy lline free levels attained in the muscle did not show any effect on the enzyme.114

PAGE 55

42 Morphine brain levels obtained by MD and blood levels were correlated to its nocceptive effect in rats. The concentrationeffect curve showed a hysteresis with an effect delay of 5 and 32 min for brain a nd blood levels, respectively. The study shows that the effect delay observed for morphine is significantly affected by the drug transport across the blood brain barrier.115 The concentrations measured by MD at the effect site were used to simulate the pharmacodynamic effect in vitro. PK/in vitr o PD correlations were performed for anticancer drugs such as methotrexate and 5-FU.56 The same approach has been used to antibiotics, such as piperacillin, where con centrations of the extr acellular space obtained by MD were used to simulate the antibiotic effect by time kill curve.78 A better correlation between concentration and effect can be found when levels at the effect site are used to simulate the pharmacodynamic eff ect. Therefore, the success and failure in therapy may be explained in the variability in the interstitial concentrations. In conclusion, microdialysis is a versatile technique that can be used to study drug delivery to the target tissue. The data obtai ned by MD can be used to differentiate between insufficient target site concen trations or pharmacodynamic unresponsiveness and to help establishing a more effective therapeutic.

PAGE 56

43 CHAPTER 3 IN VITRO MICRODIALYSIS OF DOCETAXEL Introduction The feasibility of doing microdialysis of different compounds depends significantly on the physical chemical characteristics of the compound. Lipophilic as well as high molecular weight compounds are less likely to diffuse through the probe membrane and, therefore, may not be feasible for microdialysis.116 Since the diffusion through the microdialysis membrane follows Fick’s law, factors as partition coefficient and surface area of the compound will affect the dr ug permeability through the membrane.11,116 Molecules with high molecular weight tend to have a lowe r diffusion coefficient through the dialysis membrane, which re sults in a decreased recovery.116 Low recoveries observed for lipophilic compounds ar e also attributed to the so lubility of the compound in the hydrophilic perfusate medi a, nonspecific binding to the probe, and high protein binding.116,117 Docetaxel (Figure 3-1) is a semi synthetic analog of paclitaxel. Its chemical structure is composed of a bulky, exte nded fused ring with several hydrophobic substitutes that provide its lipophilicit y and poor water solubility (logP= 3, MW= 807.9).118,119 It is an anticancer drug indicated in the treatment of breast, ovarian, and non-small-cell lung. The possibility of usi ng microdialysis to study the distribution profile of this compound would allow one to i nvestigate the drug dist ribution into a tumor tissue, which has different physio logical characteristics, and, therefore, make predictions about its effect.

PAGE 57

44 Figure 3-1: Docetaxel chemical structure. Therefore, it was the objective of this st udy to investigate the feasibility of doing microdialysis of docetaxel, a very lipoph ilic compound, by determining the in vitro recovery of this compound by the microdialysis probe. Material and Methods Chemicals The docetaxel reference standard was obtained from Aventis. Paclitaxel was purchased from Sigma-Aldrich. Both compounds we re stored at 4C in amber container. All the solvents were of HPLC grade. L actated Ringer’s solution USP were purchased from Abbott and used in the microdialysis experiments. Standard Solutions A stock solution of 100.0 g/ml of docetaxel in methanol (stock solution A) was used to prepare the standard curve and quality controls (QC) in l actated Ringer’s solution for analysis of the microdialysis samples. The stock solution A was divided into small tubes and stored at C when not in use. The standards for the calibration curve were prepared by diluting the stock solution A with Lactated Ri nger’s solution to obtain the concentration of 10.0 g/ml. This solution wa s further diluted with lactated Ringer’s solution in order to achieve the followi ng final concentrations: 0.5, 1.0, 3.0, 5.0, and 7.0 g/ml. The QC standards were prepared by diluting stock solution A with lactated

PAGE 58

45 Ringer’s solution in order to obtain the final concentrations of 0.8 (QC1), 4.0 (QC2), and 9.0 g/ml (QC3). Plasma Preparation A stock solution of 1 mg/ml of docetaxel in methanol (stock solution B) was used to spike blank human plasma and prepare the standard curve and QC s in plasma for the analysis of the plasma samples from the no-net-flux experiment. The stock solution B was stored at C when not in use. Th e standards for the calibration curve were prepared by diluting the stock solution B with plasma to get the concentration of 100.0 g/ml (stock solution B1). This plasma so lution was further diluted with plasma to achieve the following final concentrations: 10, 25, 50, and 75 g/ml. The QC standards were prepared by making another dilution of the stock solution B with plasma to get the concentration of 100.0 g/ml (stock solution B2). The stock solution B2 was further diluted with plasma in order to obtain the final concentrations of 15.0, 35.0, and 60.0 g/ml (QC1, QC2, and QC3, respectively). The plasma used in the microdialysis e xperiment (no-net-flux) was prepared by spiking plasma with the docetax el stock solution B to obtain the final total concentration of 62.5 g/ml. Solid Phase Extraction (SPE) The plasma standards were extracted by solid phase extraction (SPE). A 50-l aliquot of internal standard (IS) paclitaxel (50 mg /l of methanol) was added to 200 l of human plasma spiked with docetaxel stan dard as described above. The plasma was diluted with 1 ml of aceton itrile:water (30:70) solution. The solution was vortex and centrifuged at 3500 rpm for 15 min. 1 ml of th e supernatant was used in the extraction.

PAGE 59

46 The SPE column was conditioned with tw o 1.5-ml aliquot of acetonitrile followed by two 1.5-ml aliquot of water. The plasma was then added to the SPE column. Afterward, the columns were washed with tw o 1.5-ml aliquots of water. Docetaxel and the IS were eluted twice from the column with 2 ml of aceton itrile. The acetonitrile solution eluted from the column was eva porated to dryness by vacuum. The dried residues were reconstituted w ith 200 l of mobile phase and a volume of 25 l was injected into the HPLC system described. HPLC System The standard solutions and QCs prepared in lactated Ringer’s solution and the microdialysis samples were analyzed by a HP LC system consisted of a ConstaMetric IIIG LDC pump, a spectromonitor LDC an alytical 3200 set to 225 nm, a HP 3396 integrator, and a Perkin Elmer Serie 200 autosa mpler. A 25l sample was injected into Inertisil ODS-2 column (150X4.6 mm, 5m), co nnected to a guard column filled with Pellicular C18 material (3040 m), at a flow rate of 1ml/min. The mobile phase consisted of 0.3%phosphoric ac id: methanol (32.5:67.5). The same HPLC system was used for the plasma analysis; however, the column used for the analysis was a Discovery C18 reversed phase column (250 X 4.6 mm, 5 m) from Supelco. Microdialysis System Microdialysis probes were purchased fr om CMA microdialysis, Stockholm. The CMA/20 probe, with a membrane length of 10 mm and molecular cutoff of 20 kDa, was used in this study. The probe was connected to a 1000l Gastight syringe by a catheter connector (BBraun). A microinfusion pump (H arvard apparatus, model 22, South Natick, MA) was used to keep the fl ow constant through the probe.

PAGE 60

47 Microdialysis Experiments The in vitro recovery of docetaxel wa s determined by three different methods: extraction efficiency (EE), retrodialysis (R D), and no-net-flux (NNF). All methods were carried out at 37C. Each procedure is described in the following sections. Extraction efficiency method In the extraction efficiency method (EE), blank lactated Ringer’s solution was pumped through the microdialysis probe, which was placed into a glass tube filled with approximately 4 ml of drug solution. Three diffe rent docetaxel concen trations were used in this experiment, 2.5, 5, and 9 g/ml, all prepared in lactated Ringer’s solution. The flow through the membrane was initially set to 5 l/min for 10 minutes, and afterwards changed to 1.5 l/min for 1.5 hour (equilibration period). Subsequently, dialysate samples were collected every 25 minutes. A to tal of 5 samples were collected for each experiment, and a total of three to four experiments were performed for each concentration. The docetaxel concentration in the dialysates and in the tube before and after the experiments was determined by the HPLC/UV method described above. The probe recovery determined by the extraction efficiency was calculated by the equation: 100 % sol outC C R (Equation 3-1) where R% is the recovery in percentage; Cout is the concentration in the dialysate; and Csol is the average drug concentration in th e tube before and after the experiment. Retrodialysis method In the retrodialysis (RD) experiment the probe was placed in blank lactated Ringer’s solution and drug solutions of differe nt concentrations were pumped through the probe. The same equilibration period as in the EE method was followed and, after it, a

PAGE 61

48 total of 10 dialysate samples were collected every 25 minutes. A total of 3 experiments were performed for each concentration. The drug concentration in the microdialysis samples as well as in the syringe before a nd after the experiment were determined by HPLC/UV. The recovery in this experiment was calculated by the equation: 100 % in out inC C C R (Equation 3-2) where R% is the recovery in percentage; Cin is the average concentration in the perfusate before and after the experiment; and Cout is the concentration in the dilaysate. No-net-flux method In the no-net-flux method (NNF) experiment the microdialysis probe was placed in plasma solution containing 62.5 g/ml of docetaxel. The microdialysis probe was perfused at different times with three docet axel concentrations (2.5, 5, and 10 g/ml), all prepared in lactated Ringer’s solution. An equilibration time of 10 minutes at 5 l/min followed by 3h at 1.5 l/min was allowed before the first dialysate sample was collected for the initial concentration of 2.5 g/ml. For every change in the perfusate concentration the probe was allowed to equilibrate with the new concentration for 10 minutes at 5 l/min followed by 1.5 h at 1.5 l/min. A total of 3 dialysate samples were collected for each concentration. The drug concentration in th e syringe before and after the experiment was also determined. A total of five experiments were performed. Plasma samples from the tube were al so collected before starting the NNF experiment, at the end, and in every perfusat e concentration change. The plasma samples were extracted by SPE and analyzed by HPLC/UV method described above.

PAGE 62

49 The recovery was determined by plotti ng the net change in the docetaxel concentration versus the perf usate concentration. The slope of the curve represents the recovery and the intercept of the curve repres ents the point of no-net-flux, which is equal to the free docetaxel concentration in the plasma solution. Binding The binding to the inlet and outlet tu bing was determined by pumping drug solutions of different concentrations through the inlet and outlet t ubing of the probe. The drug solutions were prepared by diluting the stock solution of 100 g/ml with lactated Ringer’s solution in order to obtain the fina l concentrations of 2.5, 10, 20, and 30 g/ml. The tubing of an old probe were cut out a nd connected to a syringe filled with drug solution. The solution was pumped through the t ubing at a flow rate of 1.5 l/min, and samples were collected every 25 min. A total of 4 to 6 samples were collected for each experiment. The concentration in the samples, as well as the concentration in the syringe before and after the experiment, were meas ured by HPLC/UV method described before. The concentration obtained in the sa mples was compared to the average concentration in the syringe, obtained before and after the experiment. The concentration in the samples was considered as the percenta ge recovered at the end of the tubing, after pumping drug solution through it. Results Analytical Method The analytical method developed for the plas ma samples showed to be linear at the range 10 to 75 mg/l (r2=0.999). The mean regression curve was y=0.0202x-0.0243 (x= docetaxel concentration, y= sta ndard to internal standard pe ak height ratio). The lowest concentration of the standard curve showed a coefficient of varia tion (CV%) within day

PAGE 63

50 below 1.5% and was accepted as the lowest limit of quantification of the method. The recovery of the extract ion method was determined by compar ing the peak heights of four standard concentrations prepar ed in mobile phase with t hose obtained for the standards prepared in plasma and extracted by SPE. The SPE method showed an average recovery of 91.7 +/9% (n=6 for each concentration tested). The in tra and inter day precision was determined by the CV% obtained after inj ecting three times four concentrations representing the entire range of the standard curve. The in tra day precision ranged from 0.2 to 1.8% and the inter day precision range d from 8.3 to 12%. The assay accuracy was determined by comparing the nominal QC conc entration to the concentration measured using the standard curve. The accuracy of the method ranged from 80.9 to 106.5% for all three concentrations tested. The analytical method developed for the microdialysis samples was validated as described for the plasma samples. The method showed to be linear in the concentration range of 0.5 to 10 g/ml (r2=0.998). The mean regression curve was y=6802x-1529 (x= docetaxel concentration, y= peak height). The lower concentration of the curve was accepted as the lower quantification limit sin ce it showed a CV% within 15%. The intra day precision ranged from 0.6 and 12.1% and th e inter day precision ranged from 3.6 to 12.3%, with higher variability observed for th e lower concentration of 0.5 g/ml. The accuracy of the method ranged between 83.6 and 112.7%, with lower accuracy observed for the lower QC (0.8 g/ml). The two methods applied for the analysis of docetaxel showed to be within the required range of 15% variability for the pr ecision and accuracy of the lower limit of quantification.120

PAGE 64

51 Extraction Efficiency Method The dialysate concentration obtained fo r each extraction experiment and the calculated recovered are showed in Figure 3-2. Table 3-1 presents the averaged drug concentration and the respective recovery calc ulated for each experiment day, as well as the average recovery for each concentration. Th e SD represents the standard deviation of the concentrations in the tube measured be fore and after each experiment. The average recovery obtained for all three con centrations tested was 44.2 +/7.3%. A) Average dialysate concentration Extraction (2.5 mg/l)Experiment 1234 Dialysate (mg/l) 0 1 2 3 4 5 Extraction (5 mg/l)Experiment 123 Dyalisate (mg/l) 0 1 2 3 4 5 Extraction (9mg/l)Experiment 012345 Dialysate (mg/l) 0 1 2 3 4 5 B) Average recovery (R%) Extraction (2.5 mg/l)Experiment 012345 R% 0 20 40 60 80 100 Extraction (5 mg/l)Experiment 123 R% 0 20 40 60 80 100 Extraction(9 mg/l)Experiment 012345 R% 0 20 40 60 80 100 Figure 3-2: Average dialysate concentration (A) obtained in each experiment day for each concentration and its respective calcu lated recovery%R (B) when using extraction efficiency as probe calibration method. Retrodialysis Method Figure 3-3 shows the average dialysate conc entration obtained and the respective recovery (R%) calculated for each dialysate. The average perfusate concentrations before and after the experiment are depicted in Tabl e 3-1 for all three concentrations tested. The recovery, calculated for each concentration by applying Equation 3-2, is also described in

PAGE 65

52 Table 3-2. The average recovery calculated fo r all three concentrati ons tested is 60.3 +/6.8%. A) Average dialysate concentration Retrodialysis (2.5 mg/l)Sample 01234567891011 Dialysate (mg/l) 0 1 2 3 4 5 Retrodialysis (5 mg/l)Sample 01234567891011 Dialysate (mg/l) 0 1 2 3 4 5 Retrodialysis (9 mg/l)Sample 01234567891011 Dialysate (mg/l) 0 1 2 3 4 5 B) Average recovery (R%) Retrodialysis (2.5 mg/l)Sample 01234567891011 R% 0 20 40 60 80 100 Retrodialysis (5 mg/l)Sample 01234567891011 R% 0 20 40 60 80 100 Retrodialysis (9 mg/l)Sample 01234567891011 R% 0 20 40 60 80 100 Figure 3-3: Average dialysate concentration (A) obtained in each experiment day for each concentration and its respective calcula ted recovery %R (B) when using retrodialysis as probe calibration method. Figure 3-4 shows the average recovery obtained by extraction efficiency in comparison to the average recovery obtaine d by retrodialysis for each concentration. Concentration (mg/l) 2.55.08.09.010.0 R% 0 20 40 60 80 100 EE RD Figure 3-4: Average recovery obtained fo r each concentration when two different methods were tested, extraction efficiency and retrodialysis.

PAGE 66

53 Table 3-1: Average drug concentration measur ed before and after each experiment and the respective recovery calculated fo r each experiment. The averaged %R represents the mean recove ry for each concentration. Experiment Concentration ( g/ml) +/SD* Recovery (R%) +/SD* Average R% +/SD* 2.0 +/0.3 43.5 +/4.9 2.3 +/0.2 57.8 +/2.5 2.0 +/0.3 44.3 +/5.9 2.1 +/0.3 51.5 +/5.3 49.3 +/6.7 4.4 +/0.6 47.3 +/2.7 4.2 +/0.2 39.6 +/4.9 4.0 +/0.1 48.2 +/5.9 45.03 +/4.7 8.8 +/0.7 47.7 +/5.9 7.8 +/1.1 34.6 +/4.0 8.0 +/0.5 34.4 +/7.0 Extraction Efficiency (EE) 7.8 +/0.2 37.4 +/7.2 38.5 +/1.3 1.9 +/0.1 46.5 +/1.7 2.2 +/0.1 57.7 +/6.4 2.1 +/0.3 61.9 +/4.6 53.4 +/7.9 4.8 +/0.1 52.6 +/1.8 4.4 +/0.1 66.2 +/3.8 4.5 +/0.0 65.4 +/3.7 61.4 +/7.6 8.5 +/0.1 62.1 +/3.6 8.3 +/0.2 65.2 +/2.0 Retrodialysis (RD) 9.5 +/0.1 65.4 +/1.2 64.2 +/1.9 *SD: standard deviation No-Net-Flux Method The average perfusate concentrations obtai ned for each experiment are depicted in Table 3-2. Table 3-2: Average perfusate concentrat ion in each no-net-flux experiment. Nominal Concentration ( g/ml) 2.5 5 10 NNFA 2.3 +/0.1 4.5 +/0.1 9.8 +/0.0 NNFB 2.4 +/0.0 4.3 +/0.0 9.2 +/0.2 NNFC 2.4 +/0.0 4.8 +/0.1 9.1 +/0.7 NNFD 2.4 +/0.1 4.6 +/0.0 10.0 +/0.0 Measured Concentration ( g/ml) +/SD NNFD 1.9 +/0.1 3.9 +/0.0 7.7 +/0.0

PAGE 67

54 The net change in the dialysate and perfus ate concentration was plotted against the perfusate concentration. The slope of the regr ession line represents the drug recovery by the probe. The recovery obtained by this method ranged from 54.5 to 79.6%. The average curve obtained in all five NNF experiment s performed is show in Figure 3-5. The individual NNF plots are shown in Appendix A. The average recovery obtained by this method was 68.7 +/9.6% (n=5). The point where the line crosses the x-axis represents the free plasma concentration inside the vi al. The average free plasma concentration, calculated by the regression line obtained in each NNF experiment, was 4.7 1.1 g/ml. The measured plasma concentrations of the samples from the no-net-flux experiment are depicted in Table 3-3. -5 -4 -3 -2 -1 0 1 2 3 02.557.510 CperfusateCdialysate-Cperfusate Figure 3-5: Plot of the net ch ange in the concentration be tween perfusate and dialysate versus the perfusate concentration. Th e slope of the curve represents the recovery and the intercept with the x-axis represents the free plasma concentration in the vial (y=-0.679x+3.187; r2=0.9999). Table 3-3: Concentration of the plasma samples obtained from the no-net-flux experiments. Samples NNFA NNFB NNFC NNFD NNFE Initially 54.7 54.4 62.1 52.2 52.1 2.5 – 5 mg/l 54.4 57.3 46.2 48.2 52.8 5 – 10 mg/l 55.0 49.5 41.3 45.8 44.3 End 39.0 55.8 45.5 43.3 41.2

PAGE 68

55 Binding Assay The percentage recovered in the sample s, after pumping drug solution through the tubing, is shown in Figure 3-6. The dark bars represent the percenta ge of drug recovered at the end of the outlet tubi ng, while the clear bars repres ents the percentage of drug recovered at the end of the inlet tubing. The recovery for the inle t tubing ranged from 78 to 91%, while higher recoveries were obs erved for the outlet tubing, 87 to 103%. At concentration of 30 g/ml, no drug was lost to the outlet tubing, but only a 91% recovery was observed for the inlet tubing. BindingConcentration (mg/l) 0510152025303540 % Recovered 0 20 40 60 80 100 120 Probe inlet Probe outlet Figure 3-6: Binding to the inlet and outle t tubing of the microdialysis probe. Discussion The principle of microdialysis is base d on the diffusion of compounds through the probe membrane, which is permeable to small compounds. Because the probe is constantly being perfused by a physiologi cal solution, equilibrium between the drug concentration in the probe lumen and in the probe surrounding is never reached. Therefore, the concentration in the dialysate will always represent a fraction of the real concentration in the tissue.55 Therefore, calibration of the probe is very important to determine how much drug can be recovere d by the microdialysis probe. There are

PAGE 69

56 different ways one can determine the probe recovery, but the most frequently used methods are extraction efficienc y, retrodialysis, and no-net-flux. The recovery obtained by EE method (or re covery by gain) represents the fraction of the total amount of drug into solution that can be extracted by the probe. This method mimics in vitro the situation in vivo, when microdialysis is used to sample drug from a specific tissue. In this case, the drug is present in the tissue and diffuses through the membrane into the probe. Because the drug co ncentration in the tissue is expected to change overtime, it is important to verify if the drug recovery by the probe remains constant over different concentrations. In this study, three different concentrations were testes: 2.5, 5, and 9 g/ml. The concentra tions were selected based on the assay sensitivity and on the drug solubility in lact ated Ringer’s solution. C oncentrations higher than 9 g/ml were initially tested, however the drug solubility was a major problem for concentration above 10 g/ml. The recovery by gain showed to be similar for all three concentrations studied with an average r ecovery of 44.2 +/7.3%. It can be observed a tendency on the recovery to dr op at increasing concentratio ns, as showed by the lower recovery obtained for the concentration of 9 g/ml. However, the ANOVA analysis of the recoveries showed that the difference obser ved is not significantly different (p<0.1). The recovery obtained by retrodialysi s (RD) showed a higher range when compared to the recovery obtained by EE. The average recovery obtained for all three concentrations by this method was 61.8 +/4.2%. An ANOVA analysis of the results again did not reveal any differe nce in the recoveries obtained for all three concentrations tested for this method. On the other hand, a comparison between methods for each concentration should also be performed, since a major assumption for using the

PAGE 70

57 retrodialysis method when calib rating the probe in vivo is that the drug diffusion between both sides of the membrane (or the recovery) should be the same. When the recoveries for each concentration where compared, a significant difference was observed for the concentrations of 5 and 9 g/ml. In the retrodialysis experiment, samples we re collected over a l onger period of time in order to determine how long it would take to reach equilibrium, when the diffusion of the drug through the membrane is constant . When performing microdialysis, the concentration in the first samples tends to be lower than the last samples. This difference is observed due to the dilution of the drug w ith the blank solution present in the tubing and due to steep concentration gradient th rough the membrane observed at the beginning of the MD procedure, when the system is not in equilibrium.116 Before starting collecting samples when performing the probe calibration, it is important to wait until the system is in equilibrium, which means that the dr ug diffusion through the probe membrane is constant.14 The dialysate concentration in the firs t two samples obtained by RD is still increasing, even though an equi libration period of 1.5h at 1.5 l/min was allowed before collecting the first sample. Similar situati on was observed in the EE method, where the first dialysate sample showed lower concentrat ion than the other samples. The RD results also show that equilibrium is reached only af ter the third sample, which indicates that the best equilibration period could be as l ong as 3 hours at a flow of 1.5 l/min. The recovery measured by RD and NNF me thods was higher than the recovery obtained by EE, which may indicate that there are some other factors interfering with docetaxel recovery. The drug also may bind to plastic materials, as it is observed for taxol, and the binding affinity depends on the partition of the compound to plastic and to

PAGE 71

58 the solvent used as perfusate.121 This factor can affect the recovery of the compound, since it could bind to the pr obe tubing and membrane.9 The binding effect could explain why the recovery observed by RD is higher than the EE method. Since for RD the drug needs to pass by the inlet, membrane, and ou tlet of the probe, gett ing in contact with longer plastic tubing, while for EE the drug get in contact only with the probe membrane and outlet. The binding experiment showed that at c oncentrations below 20 g/ml, 78 to 91% of drug is lost to the plastic tubing used in the microdialysis probes. It is probably due to binding to the plastic material used in the manufacturing of the tubing (polyurethane) and due to the low drug solubility in lactated Ringer’s solution. The lost showed to be bigger for the inlet than the outlet, which may be rela ted to the fact that the inlet tubing is colorcoded. However, if concentrations are high enough, the loss becomes less significant for the inlet and almost no drug is lost to th e outlet tubing. It is known that docetaxel is unstable when stored in PVC bags at room temperature,122 however it is stable at room temper ature in glass containers for 4 weeks and in plasma for 15 to 24h.122,123 The drug concentration, eith er in the syringe or in solution, before and after the experiment, show ed to be similar as it can be observed by the low standard deviation obtained when the concentrations where compared. The similar concentrations obtained indicate that the compound was fairly stable in lactated Ringer’s solution throughout the experiment period. On the other hand, the plasma samples obtained in the NNF experiment showed a decrease in the concentration towards the end of the experiment, which may be an indication of drug instability at 37C.

PAGE 72

59 In the NNF, the drug may diffuse in both directions across the membrane, depending on the difference between concentrati on in the perfusate and in the tube. When the concentration of the analyt e in the perfusate is higher th an the concentration in the tube, some analyte will diffuse from the probe into the tube, resulting in a decreased dialysate concentration. On the other hand, when perfusate concentration is smaller than the solution concentration, drug will diffuse fr om the tube into the probe, resulting in increased dialysate concentration. At th e point where the perfusate and solution concentrations are identical, there will be no-net-flux across the membrane. At this point the curve crosses the x-axis and the free concen tration in the plasma solution can then be determined. Because of this characteristic of having drug diffusion through both sides of the membrane, the NNF is considered a more precise method to determine the recovery. The recovery obtained for docetaxel by NNF was similar to the recovery obtained by retrodialysis (67.9% a nd 61.8%, respectively). The initial drug concentrati on in the plasma was chos en to be around 62.5 mg/l, which would give a free concentration of 5 g/ml, according to the reported protein binding of docetaxel.118 This free concentration was c onsidered adequate for the NNF experiment since it is seated ri ght at in mid point among all concentrations tested and it is within the assay linear range. The free concen tration in the plasma solution obtained by this method was 4.7 +/1.1 g/ml (n=5). If th e free concentration is compared with the initial plasma concentration, the fraction bound to proteins can be estimated. The free fraction estimated in this case was 8.6 +/ 2.3%, or 91.4% bound, which is in excellent agreement to the protein binding reported in the literature for this compound of 92%.118

PAGE 73

60 The good recoveries obtained by all methods applied in this study show that the compound docetaxel has a good diffusion throug h the microdialysis membrane used in this study. However, due to the difference in the recoveries obtain ed for the extraction efficiency and retrodialysis methods, the pr obe calibration in vivo might be a challenge and some other alternative should be used for in vivo studies. In conclusion, the compound docetaxel seems to be very suitable for microdialysis despite its lipophilicity and high molecular weight.

PAGE 74

61 CHAPTER 4 IN VITRO MICRODIALYSIS OF CORICOSTEROIDS Introduction Microdialysis is a dynamic technique used to sample or deliver compounds to specific tissue. Because of the continuous pe rfusion of the probe, microdialysis takes place under non-equilibrium conditions and calib ration of the probe is previous any experiment. Therefore, the determination of th e recovery of the drug (the fraction of the free drug at the site of sampling which dialyses into the probe) is essential to estimate the free drug levels in the tissue.3 The literature shows controversial resu lts regarding the microdialysis of corticosteroids. While some authors have reported being difficult to dialysate corticosteroids,2 others reported no problems in pe rforming microdialysis of these compounds.3,4 Corticosteroids are fairly lipophili c compounds and have high molecular weight. The structures of a few corticoste roids are shown in Figure 4-1. Due to their structure and size, some authors believed that they cannot freely cross the microdialysis membrane. Furthermore, because the diffusi on of drugs through the microdialysis probe follows Fick’s law, factors as molecular size, shape, and lipophilicity will affect the drug diffusion through the probe and, therefore, the recovery.116 Another factor that makes one believes th at corticosteroids are not suitable for microdialysis, is their high protein bindi ng. Corticosteroids have a protein binding ranging from 60% to 90% in human plasma.124-126 Because microdialysis can only sample the free drug, compounds that have a hi gh protein binding will give very low

PAGE 75

62 concentration in the dialysate, which makes analysis of this type of samples very difficult.127 O H H OH OH O CH3CH3H HO H O H O H O C H3C H3H O H O F C H3 H OH O CH3CH3H O HO F O O C CH3CH3 Cortisol Dexamethasone Triamcinolone Acetonide H C O O O C O HO CH 2 OH CH 3 CH 3 C 3 H 7O H H H H HO F F O CH 3 CH 3 CH 3 CSCH 2 F OCOC 2 H 5 C O O O C O F HO CH 2 OH CH 3 CH 3 Budesonide Fluticasone Propionate Flunisolide C H O O C H 2 O H O C O C 2 C H 3 C H 3 O C O C 2 C H 3 H Beclomethasone propionate Figure 4-1: Corticos teroids structure. Therefore it is the objective of this chap ter to determine the in vitro recovery by gain of different corticosteroids and to co rrelate the recovery value obtained for each steroid to its logP and molecular weight valu es. Besides, in order to prepare for the in vivo experiments, the recovery of dexameth asone for two different probes, CMA/20 and CMA/60, was determined by extraction effici ency. The influence of concentration changes in the dexamethasone recovery was also assessed by retrodialysis.

PAGE 76

63 Methods and Materials Materials Microdialysis probes were purchased from CMA/Microdialysis (Stockholm, Sweden). The CMA/20 probe, with a membra ne length of 10 mm and a molecular cutoff of 20 kDa, and CMA/60, with a membrane le ngth of 16 mm and a molecular cutoff of 20 kDa, were used for the experiments. The probe was connected to a 1000 l Gastight syringe by a Perifix screw connector (B. Braun). A microinfusion pump (Harvard Apparatus Syringe Pump Model ) was used to keep the flow through the probe constant. All the solvents used in the HPLC analysis were of HPLC grade. Standard Solutions Stock solutions of 100.0 g/ml of each steroid (flunisolide, triamcinolone acetonide, dexamethasone, hydrocortisone, b eclomethasone dipropionate, budesonide, fluticasone propionate) were prepared in methanol and stored at C for future preparation of the calibration curves and quality controls . The fluticasone propionate stock solution was further dilute d with methanol in order to achieve the concentration of 10 g/ml. The stock solutions of flunisolide, triamcinolone acetonide, dexamethasone, hydrocortisone, and budesonide were diluted with lactated Ringer’s so lution in order to achieve the following final concentra tions: 0.5, 1.0, 2.0, 4.0, 8.0 and 16.0 g/ml. The quality control (QC) st andards were prepared by diluting the stock solution with lactated Ringer’s solution in order to ob tain the final concentr ations of 1.5 (QC1), 6.0 (QC2) and 12.0 g/ml (QC3). Beclomethasone dipropionate standard solutions of 0.5,

PAGE 77

64 1.0, 2.0, 4.0, 8.0 g/ml and QCs of 1.5, 3, and 6 g/ml were also prepared in lactated Ringer’s solution. The fluticasone propionate standard solu tions were prepared by diluting the 10 g/ml stock solution with lactated Ringer’s so lution in order to obtai n the concentrations of 0.0375, 0.05, 0.125, 0.25, 0.50, 1.0, 2.50, and 5.0 ng/ml. The QCs used in the fluticasone propionate analysis were prepared by diluti ng the 10 g/ml stock solution with lactated Ringer’s solution in order to ob tain the final concentr ations of 0.045 (QC1), 0.75 (QC2) and 4 ng/ml (QC3). The solutions used in the microdialysis experiments were prepared by diluting the stock solution with lactated Ringer’s solution in order to obtain the final concentrations of 1.5, 5.0 and 12.0 g/ml. Beclomethasone dipr opionate was tested for the concentration of 5.0 g/ml. Lower concentrations of fluticasone pr opionate were tested, due to the low solubility of the compound in lactated Ringer’s solution. The concentrations tested in the microdialysis experiment were 0.1, 0.33, and 0.84 g/ml. Sample Preparation The fluticasone propionate samples were extracted by a liquid-liquid method before analysis by the LC-MS/MS method. Briefly, 200 l samples and standards received ethyl acetate, in a 1:3 proportion, and 13C-fluticasone propionate (i nternal standard). The solution was vortex for 10 minutes, and the supe rnatant was transfer to a glass tube and evaporated to dryness under vacuum. The resi due was reconstituted in mobile phase and injected into the LC-MS/MS system.

PAGE 78

65 Chromatographic System The standard solutions, QCs and samples of all steroids but fl uticasone propionate were analyzed by a HPLC system consisti ng of a ConstaMetric IIIG LDC pump, a spectromonitor LDC analytical 3200 set to 254 nm, a HP 3396 integrator and a Perkin Elmer Serie 200 autosampler. 30 l samples were injected into a C 18 column (50x2.1 mm, 5m), connected to a guard column filled with Pelli cular C18 material (30-40 m ), at a flow rate of 0.4 ml/min. The mobile phase for flunisolide, triamcinolone acetonide, and dexamethasone consisted of acetonitrile:water:phosphoric ac id (28:72:0.15), pH of 2.3. The mobile phase for hydrocortisone and budesonide consis ted of acetonitrile:w ater (20:80) and acetonitrile:water (40:60), respectively. For be clomethasone dipropionate, a mobile phase of acetonitrile:water (50:50) was used. The fluticasone propionate c oncentration in solutions a nd dialysate were measured by a validated LC-MS/MS method.128 Microdialysis Experiments Extraction efficiency method The in vitro recovery of th e steroids was determined by the extraction efficiency method (EE). In this method, blank Ringe r’s solution was pumped through the microdialysis (MD) probe (CMA/20 or CMA/ 60), which was placed into a testing tube filled with approximately 4 ml of drug solu tion. Three different concentrations (1.5, 5.0 and 12.0 l/ml) were tested for each steroid. After placing the probe in the drug solu tion, it was allowed to equilibrate for 10 minutes at 5 l/min followed by 60 minutes at 2 l/min. Subsequently the equilibration period, dialysate samples were collected ev ery 20 minutes. A total of 5 samples were

PAGE 79

66 collected for each experiment, in a total of five experiments for each steroid concentration. The steroid concentration in the dialysat e and in the tube before and after the microdialysis experiment was determined by the HPLC method described above. The probe recovery determined by the extraction efficiency was calculated by using Equation 3-1. Retrodialysis In the retrodialysis experiment the MD probe (CMA/20) was pl aced in 4 ml blank Ringer’s solution and dexamethasone soluti on was pumped through it. An equilibration period similar to the EE experiment was allowed before the first sample was collected. Three different concentrations were tested: 1.5, 5, and 12 mg/l. The recovery was calculated by dividing the difference between perfusate and dialysate concentrations over the average pe rfusate concentration measured before and after the experiment (Equation 3-2). 24h Retrodialysis The MD probe (CMA/20) was placed in an Eppendorf tube fill ed with 500 l of lactated Ringer’s solution. The probe was connected to a 5-ml glass syringe filled with the dexamethasone solution of 30 g/ml and flushed. Thereafter, an equilibration period of 5 minutes at 5 l/min was allowed. The dialysate during the equilibration period was collected and its concentration was measured as well. After the equilibration period, the flow rate was changed to 2 l/min and dial ysate samples were collected every 30 minutes for the first 8 hours, every 2 hours for the next 4 hours, and at 24 h. The recovery was calculated as described for retrodialysis.

PAGE 80

67 Results Standard Curve Calibration The standard curves of flunisolide, tr iamcinolone acetonide, dexamethasone, budesonide and hydrocortisone in lactated Ringer’s solution s howed to be linear in the range of 0.5 to 16.0 g/ml. Beclomethasone dopropionate standard curve was linear in the range of 0.5 to 8.0 g/ml. The accuracy, determined by comparing the measured concentration of the QCs to their nomi nal concentration, ranged between 88.0 to 118.8%. The interday precision ranged from 0.1 to 13.3%. Extraction Efficiency Method The average recovery obtained for each co rticosteroid concentration tested is shown in Figure 4-2. The overall averaged recovery obtained for hydrocortisone was 53.5 +/9.6%; for flunisolide was 52.7 +/12.6%, for triamcinolone acetonide was 57.1+/10.3%, for dexamethasone was 51.5 +/9.2% and for budesonide was 41.7+/10.0%. Beclomethasone dipropionate showed not to be very soluble and stable in lactated Ringer’s solution and no drug could be detected in the dialysate when a concentration of 5 g/ml was tested. Fluticasone propionate was not detected in the dialysate of all concentrations tested for this compound. Dexamethasone recovery by EE was also determined using a CMA/60 probe. The recovery obtained with this probe range d from 66.9 to 88.7%. An ANOVA analysis showed no statistically significant difference am ong the recoveries obtained for all three concentrations tested using CMA/60. Figure 4-3 shows the averaged recovery obtained for each concentration tested using the CMA/ 60 in comparison to the recovery obtained when CMA/20 was used.

PAGE 81

68 Concentration (mg/l) 0246121416 R% 0 20 40 60 80 100 HC FLU DEXA TA BUD Figure 4-2: Average recovery obtained for each corticosteroid concentration tested (HC: hydrocortisone; FLU: flunisolide, DEXA: dexamethasone; TA: triamcinolone acetonide; BUD: budesonide). The bars represent the standard deviation. DexamethasoneConcentration (mg/l) 0235610111213141516 R% 0 20 40 60 80 100 CMA/60 CMA/20 Figure 4-3: Dexamethasone r ecovery obtained using two di fferent probes, CMA/20 and CMA/60. Correlation The average recovery of each steroid was co mpared to two factor s that are reported as having some influence in the recovery of lipophilic compounds: the oil/water partition coefficient (logP) and the molecular weight (MW). The logP values of the different corticos teroids are given in Table 4-1. These are calculated values which were esti mated using CLOGP version 3.54 and 3.55.129

PAGE 82

69 The recovery versus logP as well as rec overy versus MW plots are shown in Figure 4-4. The correlation coefficient obta ined for each factor was 0.0042 and 0.0249, respectively. Table 4-1: logP values and molecular wei ghts of the corticosteroids used in the experiments. CORTICOSTEROIDS logP Molecular Weight Hydrocortisone 1.86 362 Flunisolide 1.7 434 Dexamethasone 2.2 392 Triamcinolone acetonide 2.7 430 Budesonide 2.31 430 Becolmethasone dipropionate 4.2 521 Fluicasone propionate 3.9 501 129 MWMW 340360380400420440 R% 0 20 40 60 80 100 logPlogP 1.61.82.02.22.42.62.8 R% 0 20 40 60 80 100 Figure 4-4: The correlation obt ained when the overall averaged recovery obtained for each compound was compared to its logP (r2= 0.0042) or MW (r2= 0.0249). Retrodialysis Table 4-2 shows the average perfusat e concentration obtained in each RD experiment performed, as well as the cal culated recovery. The average recovery calculated for all three con centrations tested is 53.6 +/7.1%. Figure 4-5 shows the average recovery obtained for each concentra tion in comparison to the recovery obtained by the EE method for the same concentrations.

PAGE 83

70 Table 4-2: Average perfusate concentration, m easured before and after each retrodialysis experiment, and the respective recovery calculated for each experiment day. The averaged %R represents the me an interday recovery for each concentration and SD represen ts the standard deviation. Nominal Concentration ( g/ml) Measured Concentration ( g/ml) +/SD Recovery (R%) +/SD Average R% +/SD 1.5 +/0.0 43.2 +/5.3 1.6 +/0.1 47.4 +/5.3 1.5 1.6 +/0.0 46.9 +/13.3 45.8 +/2.3 4.7 +/0.1 53.4 +/-4.3 4.8 +/0.1 59.4 +/4.1 5 4.9 +/0.1 66.5 +/5.6 59.8 +/6.6 11.2 +/0.5 65.9 +/8.0 11.6 +/0.2 52.0 +/4.7 12 11.4 +/0.2 47.8 +/4.4 55.2 +/9.5 DexamethasoneConcentration (mg/l) 024610121416 R% 0 20 40 60 80 100 RD EE Figure 4-5: Recovery (R%) obtained for all th ree dexamethasone concentrations tested, when the extraction efficiency (EE) and retrodialysis (RD) methods are applied. 24h Retrodialysis Figure 4-6 shows the dialysate concentra tion changes during the experiment. As it can be observed, the concentration increases slightly throughout the experiment till it reaches a plateau, which is similar to the perfusate concentration (27.5 +/2.7 g/ml). Table 4-3 depicts the concentration measured in the tube and perfusate, as well as the concentration of the last dialysate sample co llected. The tube concentration was measured at the end of the experiment, and its aver age concentration obtained was 24.7 +/2.7

PAGE 84

71 g/ml. The average intraday recovery for each experiment, calculated in according to Equation 3-2, was 20.1 +/10.5%; 20.8 +/ 9.5%; and 26.7 +/7.5%. The average recovery, calculated for all three experi ments, was 22.5 +/3.7%. However, it was observed that the recovery decreased over ti me. The recovery calculated for the first dialysate sample was 39.9, 31.3, and 35.4%, fo r the RD1, RD2, and RD3, respectively. On the other hand, the recovery calculated for the last sample was 12.3, 21.3, and 22.4%, respectively. Sample 02468101214161820 Concentration (mg/l) 0 10 20 30 40 Dialysate Perfusate Tube Figure 4-6: Dialysate concen tration obtained in the retr odialysis experiment of dexamethasone. The bars repres ent the standard deviation. Table 4-3: Results obtained in the 24h-retrodialysis experiment. RETRODIALYSIS RD1 RD2 RD3 R% 20.1 +/10.5 26.7 +/7.5 20.8 +/9.5 R% (sample 3-7) 25.3 33.2 30.2 P [g/ml] 29.80 28.13 24.49 D [g/ml] 28.45 24.99 25.50 T [g/ml] 27.78 22.63 23.75 R%: recovery in percentage P: perfusate concentration T: concentration of the solution in the tube D: dialysate concentration

PAGE 85

72 Discussion The determination of the probe recovery is a very important step when applying microdialysis for a new compound. Before pe rforming any in vivo experiment, it is important to verify the feasibility of perf orming microdialysis for the specific compound by determining its in vitro recovery. The r ecovery can be measure by different methods. However, the comparison of the recovery obtained by extraction efficiency and by retrodialysis methods is more relevant in order to predict if the drug diffusion through both sides of the membrane is the same or not. The recovery obtained by the extraction e fficiency method represents the relative recovery of the compound. Although the in v itro and in vivo rec overies can not be compared, the in vitro experiments can tell if the compound is feasible for microdialysis or not. The average recovery obtained by EE for all corticosteroids, using the CMA/20 probe, ranged from 33.3 to 62.2%, with lower recoveries observed for budesonide. Lower recovery was also observed for almost all co mpounds at higher concentration (12 g/ml). The overall recovery can be considered a good recovery if one considers the lipophilicity and size of the compounds tested and the fact that the in vivo recovery will be almost always lower than the in vitro one. On the other hand, some variabil ity could be observed within the results for each corticosteroid. The coefficient of variability (CV%) of the averaged recovery obtained for each ster oid ranged from 18% for hydrocortisone to 24.7% for budesonide. The membrane size and composition are two important factors affecting the recovery of compounds through the microdi alysis probe membrane. The two probes tested here for dexamethasone, CMA/20 and CM A/60, are reported in th e literature to be used for animal and human studies, respec tively. These probes differ in size and

PAGE 86

73 membrane material. The CMA/20 has a shorter membrane (10mm) made of polycarbonate, while the CMA/60 has a longer membrane (16mm) made of polyacrilamide. When the recovery obtained for these two probes where compared, it is clear that the CMA/60 gave a higher averaged recovery (77.2 +/7.7%) than the CMA/20 (51.5 +/9.2%). The higher recovery is in ag reement to the fact that longer membranes will have higher recoveries due to the bigge r surface area for the drug exchange between perfusate and solution.9 The recovery obtained with the CMA/60 also showed to be more linear among concentration, and no decrease in the recovery was observed at higher concentrations. It has been reported in th e literature that the reco very of compounds through the probe membrane is a factor of the size (molecular weight) and lipophilicity of the compound.25,116 However, the results presented here show that these two factors do not have a significant effect on the recovery of corticosteroids, at least in the MW and logP range studied here. The in vivo calibration of the MD probe is performed by retrodialysis in most in vivo experiments. The recovery obtained by retr odialysis is assumed to be similar to the relative recovery of the com pound, which is true if the drug diffusion from both sides of the membrane is the same.5 Therefore, in order to determine if the drug diffusion through both sides of the membrane is similar for dexamethasone, the MD probe was also calibrated by retrodialysis. The retrodia lysis method was performed keeping all the variables from the EE constant, the only difference being that the drug solution was present in the perfusate a nd not in the tube. The recovery obtained by RD for dexamethasone ranged from 45.8 to 59.8%. Despite the observed difference on the

PAGE 87

74 recovery obtained for each concentration te sted, an ANOVA analysis did not show any statistically significant difference. However, when the recovery obtained by the two methods, extraction efficiency and retrodialysis, were compared for each c oncentration, a slightly significant difference was observed for the concentration of 1.5 g /ml. The compound characteristics might also be having an impact in the recove ry obtained these two different methods. A difference in the recovery obtained by bot h methods has also been reported for betamethasone.26 The ANOVA analysis of the dexameth asone recovery obtained by extraction efficiency for all concentrations tested, showed a statistically significant difference among the recoveries. However, this differen ce is observed between the concentration of 1.5 and 12 g/ml. No differences in the r ecovery were observed when the recovery obtained for the concentrations of 1.5 and 5 g/ml and 5 and 12 g/ml were compared. In the 24 h-retrodialysis experiment, the MD probe was placed in tube filled with only 500 L of the blank lactated Ringer’s solution. In that way, we could follow the concentration build up in the dialysate and inside the tube and assess how changes in concentration can interfere with the recovery. As it can be seen in Figure 4-4, up to the eighth dialysate sample the recovery is fairly constant. However, after the eighth sample the dialysate concentration starts to incr ease again until it reaches almost the same concentration of the perfusate. The change s in the dialysate concentration could be explained by the increase in the concentra tion inside the tube, which cause to the concentration gradient to d ecrease and, therefore, less drug would diffuse through the probe membrane.

PAGE 88

75 The concentration in the last dialysate sample for all experiments performed was slightly lower than the measured perfusate concentration. If microdialysis is performed for a long period of time, in a system where the sink condition is not maintained, it is expected that in a certain point the concentra tions in the perfusate, dialysate and inside the tube would be similar. The non-sink condi tion is true for this experiment, however, all three concentrations (perfusate, dialysat e, and tube) at 24h showed to be slightly different. It is possible that a longer period of time would be required in order for the system to reach equal concentrations since a lower concentration gr adient is expected when the concentration difference is smalle r. On the other hand, the difference observed in all three concentrations might not be si gnificant if the averaged concentrations are compared and the variability on the results considered. The recovery obtained in th e 24h-retrodialysis experime nt ranged from 20.1 to 26.7%. However, a higher recove ry was calculated for the initi al dialysate samples. This result is consistent with the increase in the dialysate samples concentration. As the dexamethasone concentration inside the t ube increased, the concentration gradient between the perfusate and solution decrea sed, resulting in a decrease on the drug diffusion through the membrane. The results presented here show that f actor such as lipophilicity and molecular weight do not interfere significantly with the recovery of corticoste roids. On the other hand, membrane size and concentration are pa rameters that need to be taken into consideration when perfor ming MD of dexamethasone. Furthermore, the average recovery obtained for dexamethasone by EE and RD showed to be quite similar. Despite the controversial results ar e reported in the literature regarding the recovery of

PAGE 89

76 corticosteroids,2-4 the results here show that cor ticosteroids are suitable compounds for microdialysis.

PAGE 90

77 CHAPTER 5 PRE-CLINICAL MICRODIALYSIS OF DEXAMETHASONE Introduction The dosing regimen is usually intended to produce therapeutic leve ls at the disease site throughout the dosing in terval. Frequently, dosing de cisions are based on plasma concentrations, which are easily accessible. Ho wever, levels reached in the biophase are the determinant for the therapeutic outcome. Measurement of plasma levels is valid if the free concentrations obtained in plasma will resu lt in similar free levels at the affected tissue sites. This assumption is often true for drugs that distribute into the body by diffusion, in which the only factor affecting distribution is the c oncentration gradient between the free plasma and tissue levels. Ho wever, sometimes other mechanisms, such as active transport at poor permeable barrie rs, are involved in drug distribution and can modify drug distribution in the tissue. Other factors that can affect drug distribution are tissues with altered physiologi cal characteristic, when the blood flow and th e interstitial matrix are altered by a disease.56 By measuring the drug concentr ation at the site of action and in disease tissue it is possible to determ ine to what extent the physiological changes that occur in a tumor or infect ed site affect drug distribution. It also allows determining if therapeutic levels are attained by giving a dose calculated based on plasma levels. One way of determining drug level in the tissue is by microdialysis. Microdialysis is technique that allows sampling drug from the extracellular fl uid and, therefore, determining the free drug concentration in the tissue. The MD probe, which has a semipermeable membrane, is constantly pe rfused by a physiological solution. The free

PAGE 91

78 drug molecules that are in the tissue fluids are filtered across the membrane into the probe capillary and are carried out of the probe by a constant flow.55 The drug concentration at the probe outlet reflects the fr ee levels in the tissue. The determination of free levels is important since only the fr ee drug concentration is pharmacologically active. Therefore, it is the objective of this study to determine the dexamethasone tissue distribution in muscle of male Wistar rats by microdialysis, after intravenous administration of dexamethasone phosphate, and to compare the free muscle concentration to the plasma concentration. Material and Methods Materials All surgical instruments used in the surger y were autoclaved before use. The sterile lactated Ringer’s solution was purchased from Abbot. Sodium chloride was purchased from Fisher. All the solvents used in th e HPLC analysis were of HPLC grade. Microdialysis probes were purchased fr om CMA/microdialysis, Stockholm. The CMA/20 probe used in the in vitro experime nt has a membrane length of 10 mm and the same molecular cut off (20kDa). The probe wa s connected to a 5000l BD plastic syringe by a catheter connector (BBraun). A microinfus ion pump (Harvard apparatus, model 22, South Natick, MA) was used to keep th e flow constant through the probe. Analytical Method Standard preparation A stock solution of 100.0 g/ml of dexamethasone was prepared in methanol and stored at C for future pr eparation of the calibration curv es and quality controls used in the microdialysis sample analysis. The stock solution was diluted with lactated

PAGE 92

79 Ringer’s solution in order to achieve the following final concentrations: 0.25, 0.5, 1.0, 2.0, 4.0, and 8.0 g/ml. The quality control (QC) standards were prep ared by diluting the stock solution with lactated Ringer’s solution in order to obtain the final concentrations of 0.35 (QC1), 1.5 (QC2), and 6.0 g/ml (QC3). A dexamethasone stock solution of 1 mg/ml of methanol was us ed to prepare the standards and QC in plasma. This solution wa s stored at C when not in use. The 1 mg/ml stock solution was initially diluted in pl asma in order to get the concentration of 100 g/ml (solution A1). This plasma solution was then diluted in order to get the concentrations of 0.5, 1, 2.5, 5, 10, 25, 50, and 75 g/ml used in the calibration curve. The QC standards were prepared by maki ng another dilution of the stock solution with plasma to get the con centration of 100.0 g/ml (solution A2). The plasma solution was further diluted with plasma in order to obtain the final concen trations of 2, 7, 40, 70, and 90.0 g/ml (QC1, QC2, QC3, QC 4, and QC5 respectively). The standard curve was prepared in rat and human plasma for the assay validation. Plasma extraction The plasma standards were extracted by liquidliquid extraction. The internal standard used in this analysis was triamcinolone acetonide (TA) 50 g/ml. A 150-l aliquot of internal standard (IS ) triamcinolone acetonide (TA) (50 g/ml of methanol) was added to 150 l of plasma spiked with dexamethasone standard as described above. The plasma was extracted with 500 l of ethyl acetate by shaking it for 10 minutes. Afterwards, the mixture was centr ifuged and the supernat ant was transferred to a glass tube. The samples were evapor ated under vacuum and the residue was reconstituted with 150 l of mobile phase.

PAGE 93

80 For the plasma samples analys is, 75 l of IS was added to 75 l aliquot of the samples obtained from the animal experiment. The samples were extracted with 250 l of ethyl acetate as described above. HPLC system The standard solutions, QCs, and samp les were analyzed by a HPLC system consisted of a ConstaMetric IIIG LDC pu mp, a spectromonitor LDC analytical 3200 set to 254 nm, a HP 3396 integrator, and a Perkin Elmer Serie 200 autosampler. A 30l microdialysis sample was injected into Di scovery C18 (50X2.1mm, 5m), connected to a guard column filled with Pe llicular C18 material (30-40 m), at a flow rate of 0.4 ml/min. The mobile phase consisted of acetonitrile:water (25:75). The same HPLC system was used in the an alysis of the plasma samples. However, 25 l sample was injected into a Discovery C18 column (150 X 4.6 mm, 5 m) connected to a Pellicular C 18 (30-40m) guard column. The mobile phase consisted of water:acetonitrile:phophoric acid 85% (28:72:0.15), pH= 2.3. Animal Procedure In this study male Wistar rats were anesthetized by administration of ethylcarbamate (urethane) in traperitoneally (1.2-1.5 g/kg dose). Anesthesia was confirmed by the absence of reflexes after pinching the rat’s footpads. After anesthesia, the animal was immobilized in a supine position on a dissecti ng board. An electric heating pad was used to keep animal normo thermic and body temperature and respiratory rate were monitored throughout the expe riment. The rat was intubated through a tracheotomy procedure and artificially vent ilated with room air by using a rodent respirator (Harvard Apparatus, model 683). Th e ventilator was set at a frequency of 6266 min-1 and a volume of 2 ml.

PAGE 94

81 Blood sampling Blood samples were collected by a polyethy lene catheter (inner diameter of 0.3 mm and outer diameter of 0.7 mm) introduced into the carotid. After the catheter insertion and after every blood sample collection, the catheter was irrigated with a heparin solution in saline (25 UI of sodium hepa rin/mL of 0.9% saline). Blood sample (300-400 l) were collected in heparinized tubes before the dose was administered (time zero) and at 10, 30, 60, 90, 120, 180, 240, and 300 minutes after drug administration. Right after the blood collecti on, the samples were placed on ice until the plasma separation. The samples were centr ifuged for 10 min at 5000 rpm. The plasma was transferred to another tube and kept at C until analyzed. Muscle microdialysis The skin area in the left hind leg muscle after skin area was shaved and cut open with a scalp and a microdialysis CMA/20 pr obe was inserted according to manufacturer instructions. Briefly, a 22 GA needle was used to insert a splitable guide (plastic tubing) into the rat thigh muscle. Af ter the insertion, the needle was withdrawn without removing the guide from the muscle. The MD probe wa s inserted through the tubing, which was removed by pulling it upwards and outwards leav ing the probe into the tissue. The probe was taped to the skin in order to keep it in place and the wound was closed with super glue. After the insertion, the probe was conn ected to a 5000l syringe by a catheter connector (BBraun) and flushed with lact ated Ringer’s solution (NaCl 137 mM, KCl 1.0 mM, CaCl2 0.9 mM, NaHCO 1.2 mM). The syringe was then connected to a microinfusion pump (Harvard apparatus, m odel 22, South Natick, MA) used to keep the

PAGE 95

82 microflow constant through the probe. A flow of 2 l/min was set for 30 minutes before the probe calibration. Probe calibration The probe was calibrated by retrodialysis. In this method, the syringe with lactated Ringer’s solution was replaced by a syringe wi th known dexamethasone concentration (2 g/ml). The drug solution was pumped at a fl ow of 2 l/min. The probe was allowed to equilibrate for 30 minutes at the same flow rate before retrodialysis samples were collected. The samples for the calibration pro cedure were collected every 20 minutes and a total of two samples were collected. The samples were frozen at -20C until analyzed. The drug concentration in the microdialysis sa mples as well as in the syringe before and after the experiment was determined by HPLC/UV method de scribed above. After the probe calibration, the drug perfusion was stopped and the probe was perfused with lactated Ringer’s solution for 30 minutes (wash out period) before the drug was administered. The in vivo recovery was cal culated by the Equation 5-1: 100 % perf dial perfC C C R (Equation 5-1) where Cperf is the concentration in the perfusate; Cdial is the concentration in the dialysate; and R% is the percentage recovered. Experiment Design Fourteen Wistar male rats were used in this study to investigate the pharmacokinetics and tissue distribution of dexamethasone. The study was first approved by the Institutional Animal Care and Use Comm itteeIACUCof University of Florida.

PAGE 96

83 The rats received either 100 mg/kg (n=8) or 50 mg/kg (n=6) of dexamethasone disodium phosphate salt by i.v.. After the surgery and probe calibration by retrodialysis, a wash out period of 30 minutes was allowed before the drug admini stration. Blood and microdialysis samples were drawn at specific time intervals be fore and after the drug administration. Microdialysis samples were co llected every 20 minutes thr oughout the experiment with the help of a microfraction collector (CMA /142) for dual probes sampling. The samples were frozen before analyses by a validated HPLC/UV method. Tissue Levels The drug concentration in the muscle wa s calculated by consid ering the recovery obtained in animal each experiment. The drug concentration in the tissue will be calculated using the Equation 5-2. R C Cdil free t, (Equaiton 5-2) where Ct, free is the muscle free concentration, Cdial is the concentration in the dialysate, and R is the recovery obtaine d in each animal experiment. Pharmacokinetic Analysis The blood profile was analyzed according to a non-compartmental and a compartmental approach. In the non-compartm ental approach, the peak concentration (Cmax) and the time to reach the peak (tma x) were extracted from the concentration versus time profiles in plasma and musc le of each animal. The non-compartmental analysis of the plasma profile was perfor med using the program WinNonlin. The model 200 was used in this analysis.

PAGE 97

84 The AUC for the microdialysis samples wa s calculated by a modified trapezoidal rule. Because the concentration obtained in the microdialysis samples is an averaged concentration of the collection time, the AUC was calculated as the sum of the dialysate concentration times the collection interval . The elimination rate constant, ke, was obtained from the terminal slope of the log linear concentration versus time plot. The half-life (t1/2) was calculated by the Equation 5-3. ke t 693 . 02 1 (Equation 5-3) The clearance (Cl) for the muscle profile was calculated by dividing the dose over the AUC up to the last time point. The volum e of distribution (Vz) was calculated by dividing the clearance by the elim ination rate constant (ke). The tissue penetration was calculated by dividing the AUCtissue, free by the AUCplasma, free. The compartmental analysis of plasma a nd muscle profiles was performed using program Scientist (MicroMath). An adapted one compartmental model was used to fit the individual plasma and tissue data. The model is shown in Figure 5-1. In this model, dexamethasone dissodium phosphate was consid ered a depot in the central compartment for dexamethasone, the active form. The conve rsion of the pro-drug into dexamethasone was assumed to be a first order process. The equations that describe the model and used to fit the total plasma and the free tissue profiles are: t ka t ke total plasmaV ke ka D ka Fr C ' ,exp exp ' ' (Equation 5-4)

PAGE 98

85 ', ,F fu C Cplasma total free tissue (Equation 5-5) where Cplasma,total is the total plasma concentration; Ctissue,free is the free tissue concentration; ka’ is the conve rsion rate constant; ke is th e elimination rate constant; D is the dexamethasone phosphate dose administer ed; V is the volume of distribution; fu is the fraction unbound; Fr is the fraction of dexamethasone phosphate converted to dexamethasone; and F’ is the penetration into the muscle factor. Figure 5-1: One compartmental model used to describe dexamethasone total plasma and free tissue profiles obtained after i. v. administration of dexamethasone disodium phosphate (DP= dexameth asone dissodium phosphate; DX= dexamethasone; Xc= central compartment; ke= elimination rate constant; ka’= conversion rate constant; Fr= fr action of DP converted to DX). Results Assay Validation The HPLC assay for the analysis of the plasma samples was validated according the requirements of accuracy, precision, sensitivity and linearity.120,130 The standard curve was initially prepared in rat plasma and extracted and analyzed by the HPLC system described above. The assa y showed to be linear in the concentration range of 0.5 to 100 g.ml. The interday prec ision obtained ranged from 2.6 to 12.5% and Xc DX DOSE DP (depot) ke Fr.Ka ’

PAGE 99

86 the intraday precision ranged from 0.1 to 5%. The accuracy obtained by this method ranged from 90.3 to 116.7%. The standard curve and quality controls in human plasma were prepared as described above. The assay showed to be linea r in the same concentration ranged tested for rat plasma. The intraday precision ranged fr om 0.1 to 4.7% and th e interday precision ranged from 0.7 to 11.7%. The accuracy of the method ranged from 89 to 115.6%. The assay using rat plasma wa s cross validated to the assay using human plasma as matrix by comparing the QCs obtained with each matrix. Microdialysis The dialysate concentration is shown in Figure 5-2. A higher variability in the dialysate levels is observed for the 100 mg /kg dose. However, this variability can be attributed to the variability in the reco very. The average in vivo recovery for dexamethasone, determined by retrodialysis, was 37.4 +/9.1% and 34.3 +/5.8% for the 50 and 100 mg/kg doses, respectively. The recovery obtained for each animal was used to calculate the free level in the muscle. 50 mg/kgTime (min) 050100150200250300350 Dialysate (mg/l) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 R1 R2 R3 R4 R5 R6 100 mg/kgTime (min) 050100150200250300350 Dialysate (mg/kg) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 R7 R8 R9 R10 R11 R12 R13 R14 Figure 5-2: Dialysate concentr ation obtained after the i.v. administration of 50 and 100 mg/kg dose of dexamethasone dissodium phosphate to male Wistar rats.

PAGE 100

87 Non-Compartmental Analysis The plasma and muscle profiles obtained after iv administration of both doses is shown in Figure 5-3. The free plasma concen tration was calculated based on the plasma profile measured and a protein binding of 84.7%.126 The tissue penetration factor was calculated as 1.13 +/0.25 for the dose of 50 mg/kg and 0.98 +/0.19 for the 100 mg/kg dose. The AUCtissue, free/AUCplasma, total ratio was 0.17 +/0.04 and 0.15 +/0.03 for 50 and 100 mg/kg doses, respectively. The non-compartmental analysis of the plasma profile was performed using program WinNonlin. The results obtained from one animal dosed with 100 mg/kg were not included in the analysis due to the sm all collection interval (120 minutes), which caused the elimination rate constant calculated to be much smaller than the one obtained for the rest of the animals. Besides, the AUC0 inf calculated for this specific animal represented more than 90% of the AUC0 t. The non-compartmental analysis of the muscle concentrations was performed us ing the Excel program. The averaged pharmacokinetic parameters obtained in the non -compartmental analysis are depicted in Table 5-1. Appendix B shows the individual pharmacokinetic parameters obtained in the non-compartmental analysis. 50 mg/kgTime (min) 050100150200250300350 Concentration (mg/l) 0.1 1 10 100 Free plasma Total plasma Free tissue 100 mg/kgTime (min) 050100150200250300350 Concentration (mg/l) 0.1 1 10 100 Total plasma Free plasma Free tissue Figure 5-3: Plasma and muscle profiles obtained after i.v. administration of 50mg/kg and 100 mg/kg doses to male Wistar rats.

PAGE 101

88 Table 5-1: Pharmacokinetics parameters obtain ed in the non-compart mental analysis of the plasma and muscle profiles. Plasma Muscle Dose 100 mg/kg 50 mg/kg 100 mg/kg 50 mg/kg Cmax (mg/l) 37.3 +/6.1 21.4 +/3.7 5.4 +/1.2 3.3 +/0.4 tmax (min) 34 +/11 35 +/12 73 +/27 57 +/16 ke (min-1) 0.0038 +/0.0030.0032 +/0.0020.0037 +/0.001 0.0052 +/0.002 t1/2 (min) 269 +/143 256 +/94 200 +/59 147 +/48 Vz/Fr (l/kg) 2.4 +/0.6 2.5 +/0.6 16.5 +/5.3 11.3 +/1.6 Cl/Fr (l/h/kg) 0.48 +/0.25 0.44 +/0.11 3.61 +/1.21 3.39 +/0.77 AUC0 inf (g/l/min) 15.16 +/6.08 7.19 +/1.92 1.83 +/0.60 0.92 +/0.21 AUMC0 inf (g/l/min) 7110.99 +/5312.87 2926.39 +/1601.87 608.07 +/322.86 233.51 +/109.51 MRT (min) 406 +/196 382 +/127 316 +/83 243 +/61 Compartmental Analysis The compartmental analysis was performed simultaneously for the plasma and microdialysis profiles using the model describe d before. Each animal profile was fitted individually. The average pharmacokinetics parameters obtained in the analyses are depicted in Table 5-2. Appendix C shows the individual pharmacokinetic parameters obtained in this analysis. The individual fits can be obser ved in Appendix D. Figure 5-4 shows the average plasma and microdialysis profiles simulated using the average pharmacokinetics parameter obtained in the an alysis. Appendix D shows the individual profiles fitted according to the model described above. The elimination half-life was 2.7 and 3.2h for the dose of 50 and 100 mg/kg, respectively. The conversion rate obtained for both doses was 0.12 min-1. The conversion

PAGE 102

89 rate constant can be considered as the elim ination rate constant for the dexamethasone phosphate ester through the degradation path. In this case, the elimination half-life for this compound through the degradation route is 5.8 min, which is in good agreement with the degradation half-life for the ester in humans.131 The fraction unbound in the plasma was obtained from the literature a nd fixed in 0.153 during analysis. 05010015020025030 0 Time (min) 10-210-1100101102Concentration (mg/l) 50 mg/kg C vs T C_CALC vs T CTF vs T CTF_CALC vs T C vs T C_CALC vs T CTF vs T CTF_CALC vs T 05010015020025030 0 Time (min) 10-210-1100101102Concentration (mg/l) 100mg/kg Figure 5-4: Average plasma and microdialysis profiles obtained after the i.v. administration of 100 and 50 mg/kg dose of dexamethasone disodium phosphate. The symbols represent the meas ured concentrations and the line represents the simulated profile using the average pharmacokinetics parameters obtained in the individual anal ysis. The bars represent the standard deviation. Table 5-2: Average pharmacokinetics (PK) parameters obtained after the fitting of the plasma and microdialysis profiles using the one compartment body model with first order conversion rate. DOSE PK PARAMETERS 100 mg/kg 50 mg/kg ka’ (min-1) 0.1204 +/0.0545 0.1205 +/0.0407 ke (min-1) 0.0035 +/0.0013 0.0039 +/0.0013 t1/2 (min) 229 +/95 192 +/59 V/Fr (l/kg) 2.5 +/0.4 2.2 +/0.4 F’ 0.91 +/0.19 1.03 +/0.21 MSCa 3.82 +/0.85 4.01 +/0.49 a: model selection criteria

PAGE 103

90 Discussion The tissue penetration of compounds into di fferent organs can be investigated by a number of in vivo techniques, such as biops y, tissue assay, blister fluid, and imaging methods. However, these tec hniques show some limitations, as the number of samples that can be obtained per subject; no time resolution; overestimation of the free levels or no differentiation of bound and free leve ls; besides being labor and animal intensive.49,67,132 Microdialysis is one technique that can be used to determine tissue levels. It has the advantage of allowing multiple sampling from the extracellular fluid from the same spot in the same animal. Therefore, the number of animals required for a distribution study is decreased, as well as the variability of the results obtained. In this study, microdialysis was used to investigate dexamethasone penetration into the muscle of male Wistar rats. It has b een shown already that dexamethasone can be recovered by microdialysis probe in in vitro conditions (Chapt er 4). However, the in vivo recovery can be smaller than the in vitro r ecovery. This difference is due to the tissue tortuosity, which plays an important role in the drug diffusion through the tissue affecting, therefore, th e amount of drug that reaches the probe membrane.9,13,14 The average in vivo recovery obtained for dexa methasone was 35.6 +/7.9% for both doses, while the averaged in vitro recovery was 53.5 +/3.7%, confirming the tendency of a lower in vivo recovery. Despite the lower recovery, it was still possible to measure dexamethasone in the dialysate since a high dose was giving to the animals. Due to the low water solubility of dexame thasone and the high dose required in the experiments, the phosphate ester was ad ministered to the animals instead. Dexamethasone phosphate is known to hydr olyze in vivo by alkaline phosphatases releasing dexamethasone.133,134 This process could also be observed in in vitro

PAGE 104

91 experiments performed with blank plasma.135 However, the process in vitro is longer (longer half-life) than in huma ns, indicating that the hydrolysis occurs not only in plasma, but also in some other organ as liver and kidney.133 The phosphate ester hydrolysis is known to be a temperature dependent process and freezing the samples right after collection and centrifugati on reduces the hydrolysis rate to less than 10%.135 Therefore, in order to avoid the in vitro hydrolysis of dexamethasone phosphate after blood sampling, the samples were placed on ice and centrifuged thereafter. The addition of a phosphatase inhibitor was also tested, in order to completely stop any post collection conversion of the ester pro-drug into dexamethasone alcohol. From all the phosphatase inhibitors tested, sodium arse nate showed to be more effective. By adding 0.5 M sodium arsenate in the propor tion if 1:10 to the blood samples, it was possible to stop the hydrolysis in vitro. Therefore, the dexamethasone measured in the samples would be originated only from th e in vivo enzymatic hydrolysis. However, sodium aresenate also caused hemolysis of the blood samples and could not be used in the study. A test was performed adding sodium arsenate to blood samples obtained from one animal. It allowed detecting the pres ence of dexamethasone phosphate in the 10minute sample from this animal. No dexa methasone was detected in the 30-minute sample. Because at 10 minutes the presence of dexamethasone disodium phosphate was still significant, the 10-minute samples were not included in the pharmacokinetics analysis in order to avoid interference from the in vitr o hydrolysis. It has al so been reported that dexamethasone phosphate is not present anym ore in human plasma 30 minutes after iv administration of 8 mg dose.136

PAGE 105

92 The non-compartmental analysis of the pl asma data was performed using program WinNonlin, model 200. The maximum dexamethasone concentration was found at the first sample time, around 30 minutes, for most of the animals from both doses. The tmax in the muscle happened later than in pl asma, around 1.2h and 0.95h for 100 and 50 mg/kg doses, respectively. The mean maximum con centration attained fo r the 100 mg/kg dose was slightly lower than twice the Cmax obtai ned after administration of 50 mg/kg dose in plasma and muscle. The difference may be due to the drug administration or to the fact that the 30-minute sample might not represen t the real peak, since a more frequent sampling was not possible at the beginning. The elimination rate constant, volume and clearance in plasma were similar for both doses. The AUC0 inf of the higher dose was twice as much the lower dose, showing the proportionality of dexamethasone pharmacokinetics in the dose range studied. The tissue penetration was determined by the ration AUCtissue,free/AUCplasma,free. This ratio has been used to determine the ti ssue penetration of other drugs such as antibiotics.73-75,137 Dexamethasone tissue penetration showed a mean value around 1.03, which indicates that dexamethasone freely distributes into the muscle. The half-life of the free concentration in the muscle seems to be shorter than in plasma. However, by looking at the plots, the curves in pl asma and muscle look parallel for each dose. A statistic analysis did not re veal any significant difference between the muscle and plasma half-lives of each dose and for the muscle half-life between doses (p<0.05). The difference in the half-life could be explained by the number of points and by the time range used to calculate the elimin ation rate constant. Microdialysis does not

PAGE 106

93 remove fluids from the animal body and a ri ch profile with more time points describing the elimination phase could be obtained (at most 15 points). Blood sa mpling can not be as frequent as the microdialysis sampling, and the plasma profile obtained had at most 8 points. Therefore, more time points could be used to calculate the elimination rate constant in muscle than in plasma. Dexamethasone clearance from the muscle was not significantly different between doses; however it was higher than the plasma clearance. The muscle clearance represents the clearance of the free fraction and was calculated based on the AUC obtained for the free fraction. If the tissue binding were sim ilar to the protein binding in plasma, the clearance would go down to 0.50 and 0.49 l/h/kg for 100 and 50 mg/kg doses, respectively. The volume of distribution for the 100 mg /kg dose showed to be significantly higher than the volume at 50 mg/kg. The volum e of distribution was calculated by Cl/ke. The Cl remained the same for both doses stud ied, however, the mean ke showed to be slightly higher for the 50 mg/kg dose. Th is difference is causing the volume to be different for both doses, despite the non-signi ficant difference observed for ke between doses. Different models have been used in the literature to model dexamethasone profile after the administration of dexamethasone phosphate. While some authors modeled dexamethasone profile assuming a simple intr avenous input after i.v. administration of the pro-drug,136 other authors modeled the profile a ssuming a first order conversion rate of dexamethasone phosphate into dexamethas one alcohol. This model also assumed a rapid conversion of the ester pro-drug into dexamethasone.131,133,138

PAGE 107

94 The model used in the pharmacokinetics analysis was a one compartmental model with a first order conversion rate constant of dexamethas one phosphate. In this model, dexamethasone phosphate was considered as a depot for dexamethasone in the blood. The pro-drug conversion was assumed to follow a first order process, which occurs very rapidly in blood, with half-lives ranging from 5.4 to 9.8 min.131,133 Although dexamethasone phosphate was ad ministered i.v., only the free dexamethasone was measured in the samples. Because the hydrolysis process could not be stopped after sampling, no samples were taken before 30 minutes. Therefore, the conversion phase is not characte rized in the plots, which cau se the determination of ka’ by the model to have a high variability. The dexamethasone phosphate half-life calculated based on the average ka’ is 5.8 mi nutes, which is similar to dexamethasone phosphate half-life in humans reported in the literature.131 In this model, the elimination rate consta nt for the total plasma and free tissue was assumed to be the same. By looking at the non -compartmental analysis, the ke values for the free tissue and total plasma were similar, except to the 50 mg/kg dose. However, this difference showed not to be statistically di fferent. The ke obtained in the compartmental analysis is similar to the one obtained in th e non-compartmental analysis, as well as the volume of distribution. The tissue penetration factor (F’) and fr action unbound (fu) were included in the equation to calculate the free tissue levels. The fraction unbound was fixed in 0.153 (value obtained from the literature).126 Although the fraction unbound was taken from the literature, this value compares to the AUCtisse,free/AUCplasma,total ratio calculated in this study of 0.17 and 0.15 for 50 and 100 mg/kg dos e, respectively. The F’ was included in

PAGE 108

95 the model as a parameter. The F’ calculated in the compartmental analysis is also very similar to the value obtained in the non-compartmental analysis. The compartmental analysis generated similar PK parameters for both doses. The model fitted the data nicely, with a model selection criterion (MSC) of 3.82 and 4.01 for the dose of 100 and 50 mg/kg, respectively. In conclusion, microdialysis showed to be a suitable technique to study dexamethasone penetration into the musc le. Despite the drug lipophilicity and high molecular weight, the in vivo probe recove ry was still within a good range, allowing obtaining concentrations still within the assay limits. By applying microdialysis, it was possible to show that free levels in muscle are similar to the free levels in the plasma, which indicates that dexamethasone freel y distributes in the body. Therefore, dexamethasone plasma levels could be used to predict free tissue levels in healthy subjects.

PAGE 109

96 CHAPTER 6 CLINICAL MICRODIALYSIS OF CORTICOSTEROIDS Introduction Microdialysis is a technique that has been used to study drug distribution and penetration into peripheral tissues such as brain, heart, muscle, and lung.54 Microdialysis can also be used to determin e the penetration profile of drugs applied topically.139 The advantage of using this technique in evaluation of skin penetration are the possibility of obtaining a full concentrati on drug profile in the stratum corneum (SC), under the same area of drug application.90 Therefore, variations due to differences in skin characteristics will be eliminated. Also, th is method is suitable for measuring SC concentrations after multiple dose administrati on at the same site, and suitable to measure penetration depth of a compound beyond the SC.1,88,89 Amongst the drugs available for topical administration are corticosteroids.140 This class of compounds has been used in the treatme nt of skin diseases as well as to treat local inflammations such as rheumatoid arthritis.141-143 Dexamethasone phosphate has been used in the treatment of rheumatoid ar thritis by intraarticular injection, with an average duration of action of 6 days. Corticoste roids placed in the join are systemically absorbed. Therefore, the side effects usually observed after system ic administration can also be observed after intr aarticular administration.142 Oral and intravenous corticosteroids ar e well known to be effective for the treatment of musculoskeletal inflammatory disorders. However, the efficiency of the transdermal route for the treatment of localized rheumatoid disorders remains

PAGE 110

97 questionable.142 One important piece of informati on lacking is the assessment whether transdermal absorption leads to target tissue concentrations equivale nt to those achieved by systemic administration. Therefore, it was the objective of this study to evaluate the use of microdialysis to determine dexamethasone transdermal absorp tion by performing a pilot study in humans. In this study, dexamethasone concentrations in target tissues (corium layer and muscle) and in plasma were compared after or al and iontopheretic administration of therapeutically relevant dexamethasone doses. Materials Microdialysis probes (CMA/60) were purchased from CMA/microdialysis, Stockholm. The CMA/60 probe has a membrane length of 16 mm and molecular cutoff of 20 kDa. The probe was connected to a 1000l BD plastic syringe by a catheter connector (BBraun). A microinf usion pump (Harvard apparatus, model 22, South Natick, MA) was used to keep the fl ow constant through the probe. The iontophoresis system was supplied by IOMED Inc. (Salt Lake City, UT). The system consisted of a hydrat able gelsponge electrode used for the iontophoretic drug delivery (IOGEL), a dispersive pad, and an iontophoretic power supply device (Phoresor PM900). The scintillation liquid used in the radioimmunoassay was purchased from ICN (Costa Mesa, CA). The rabbit anti-dexamethasone serum was purchased from IgG Corporation (Nashville, TN).

PAGE 111

98 Methods Human Pilot Study Subjects One healthy male volunteer was used in this pilot study. Th e selection of the volunteer was based on the results from the lab screening, history, and physical examination. The study was approved by the In stitutional Review Board (IRB-01) of the University of Florida. The volunteer signed th e Informed Consent Form before the study. Study design The pilot study followed a randomized, noncontrolled, open, crossover trial. According to the experiment design, the s ubject received one of the following three treatments: (1) a single 8 mg dose of dexamethasone ad ministered orally as two 4-mg tablets (Decadron) or (2) 2.5 ml of dexamethasone sodium phosphate 0.4% injection solution administered epicutaneously on the thigh by iontophoresis, at a dose of 40 mA-minutes applied at 2 mA for 20 minutes. After drug administration, mi crodialysis and blood samples were collected every 20 minutes up to 1 hour and every hour after that up to 6 hours. The wash out period between treatments wa s at least one week. On each treatment day, microdialysis was performed to measur e dexamethasone local tissue concentrations. Microdialysis Two microdialysis probes (CMA /60) were inserted in the volunteer thigh according to the manufacturer instructions. Briefly, th e probe was introduced in the tissue using a guide needle, which is open in the top. After needle insertion, the probe is hold in place

PAGE 112

99 by its wing and the needle is slide out of the tissue, leaving the probe in place. One probe was introduced into the subcutaneous tissu e and another into the skeletal muscle. Thereafter, the probe was washed for 30 minutes before the drug administration. Dexamethasone dose was administered either orally or topically as described above. Microdialysis samples were collected every 20 minutes for the first two hours and every hour after it for the next 6 hours. Concentrations versus time profiles in the tissue layers were followed for a total of 8 hours. Analytical Method The standard solutions used in the sample analysis were prepared either in plasma or in lactated Ringer’s solution. A stock so lution of 100 g/ml of dexamethasone (stock solution A) was prepared in methanol for th e analysis of the microdialysis samples. A second stock solution of 500 g/ml of dexameth asone (stock solution B) was prepared in methanol and used to prepare the standards for the analysis of the plasma samples. The standard curve used in the analysis of the microdialysis samples was prepared by diluting the stock solution A in lactated Ringer’s solution in order to obtain the concentration of 10 g/ml. This solution was further diluted to get the concentrations of 0.1 to 500 ng/ml, used in the RIA assay. The standard curve used in the plasma analysis was prepared by diluting stock solution B with lactated Ringer’s solution in order to get the concentration of 100 g/ml. This solution was further diluted with plasma to get the concentra tions ranging from 0.1 to 500 ng/ml. The samples were analyzed by a modified radioimmunoassay (RIA) described elsewhere.144,145 In brief, 50 l of the standard so lution, 100 l of the antibody solution, 100 l of labeled dexamethasone, and 450 l of buffer solution (pH= 7.4) were mixed in

PAGE 113

100 an Eppendorf tube and incuba ted at 4C for 24h. The labe led and cold drug were separated by incubation with 200 l of 1% dextran-coat ed charcoal for 5 minutes followed by a centrifugation at 4000 rpm for 15 minutes, at temperat ure of 4C. The radioactivity (CPM) was measured in a 560 l aliquot of the supernatant. The measured radioactivity was plotted agai nst the concentration and the linear part of the curve was used to determine the drug concentration in the samples. Pharmacokinetic Analysis The plasma concentration profile obtained after oral administration was analyzed according to a non-compartmental and compartmental approach. In the non-compartmental approach, the peak to plasma concentration (Cmax) and the time to reach the peak (tmax) were extrac ted from the concentration versus time plot. The elimination rate constant, ke, was obtained from the terminal slope of the log linear concentration versus time plot. The area unde r the concentration-time curve (AUC) up to the last time point was calculated according the trapezoidal rule. The AUC to infinity (AUC0 inf) was calculated by Equation 6-1. ke C AUC AUCn t 0 inf 0 (Equation 6-1) where AUC0 t is the area under the concen tration curve to time t; Cn is the last concentration measured and ke is the elimination rate constant. The AUC of the muscle and corium c oncentration profile was calculated by a modified trapezoidal rule. Because the c oncentration obtained in the microdialysis samples is considered the concentration at th e middle point of the co llection interval, the AUC was calculated as the sum of the dial ysate concentration times the collection interval.

PAGE 114

101 The compartmental analysis was performed using the program Scientist (Micromath). The plasma concentration pr ofile was fitted to a one compartment body model according to Equation 6-2. lag lagt t ka t t keVc ke ka ka D C exp exp (Equation 6-2) where C is the concentration, D is the dose ad ministered, Vc is the volume in the central compartment, ka is the absorption rate consta nt, ke is the elimination rate constant, t is time, and tlag is the lag time. Results Human Pilot Study The plasma profile obtained after the oral administration of 8 mg oral dose to the volunteer is shown in Figure 6-1. The data was fitted to a one compartment body model. The pharmacokinetic parameters obtai ned after the non-compartmental and compartmental data analysis are depicted in Table 6-1. No dexamethasone was detected in the plasma samples obtained after iont ophoretic administration of dexamethasone phosphate. Figure 6-2 shows the dialysate concentrati on profile obtained by microdialysis after the oral administration of dexamethasone. Th e AUC calculated by the trapezoidal rule for each dialysate profile was 16.9 and 21.3 ng/ml.h for the probe in the muscle and skin, respectively. A peak concentration of 4.2 ng/ ml was reached at 1.8h in the muscle, while a peak concentration of 5.3 ng/ml in the skin was attained at 3.5h.

PAGE 115

102 02468Time (hours) 100101 Time (hours) 02468 02468Time (hours) 100101 1028 mg po Figure 6-1: Plasma profile obtained after oral admini stration of 8 mg dose of dexamethasone. The solid squares repres ent the measured concentration and the line represents the fitted data according to a one compartment body model. Table 6-1: Dexamethasone pharmacokinetic parameters calculated from the plasma concentration profile obtaine d after oral administration of a single oral dose of 8 mg. NON COMPARTMENTAL ANALYSI S COMPARTMENTAL ANALYSIS tAUC 0 (ng/ml/h) 455.2 ka (h-1) 2.01 0AUMC (ng/ml/h) 976.1 k (h-1) 0.1939 MRT (h) 2.14 Vc/F (L) 110 Cmax (ng/ml) 52.6 tlag (h) 0.67 tmax (h) 1.67 t1/2 (h) 3.6 ke (h-1) 0.0952 MuscleTime (h) 0123456 Concentration (ng/ml) 0 2 4 6 8 10 SkinTime (h) 0123456 Concentration (ng/ml) 0 2 4 6 8 10 Figure 6-2: Muscle and skin di alysate concentration profile ob tained after the oral of 8 mg dose of dexamethasone. Concentration (ng/ml)

PAGE 116

103 Discussion Dexamethasone is a drug that has been used in the treatment of many inflammatory disorders, either orally or topically. The topical administ ration of dexamethasone would provide the advantage of having a high lo cal concentration, but a low systemic concentration, causing fewer side effects. However, the efficacy of the topical administration of dexamethasone and ot her anti-inflammatory drugs remains controversial.2,90 In this study, it was possible to assess th e tissue distribution of dexamethasone after oral administration by applying microdialysis . After oral administration a peak drug concentration of 52.6 ng/ml was achieved at 1.7h. The plasma concentration profile was fitted according to a one-compartment body model using the program Scientist (Micromath). The elimination half-live calcu lated for this compound was 3.6h, which is similar to what has been reported in the literature for humans.131 Despite the low plasma levels we were st ill able to measure dexamethasone in the MD samples after oral administration. Su ch low microdialysis concentration was expected, since dexamethasone has a protein binding of 77% in humans,126 and only the free drug could diffuse through the probe membrane. Therefore, the free peak concentration in plasma, considering the pr otein binding, would be 12.1 ng/ml. Besides protein binding, another factor affecting the drug concentrat ion in the dialysate is the recovery. Since microdialysis is a dynamic technique, only a fraction of the free levels can be sampled by the probe. The in vitr o dexamethasone rec overy obtained using CMA/60 ranged from 66.9 to 88.7% (Chapter 4). However, it is known that the in vivo recovery is usually less than the recovery obtained in vitro, whic h leads to dialysate concentrations lower than the observed free fraction in blood.

PAGE 117

104 The dialysate concentration profiles obtained from both probes after oral administration are shown in Figure 6-3. The probe placed in the skin shows a different dialysate profile from the probe placed in th e muscle. It seems that the drug takes longer to distribute in the skin, showing a longer tmax than the tmax obtained in muscle. It also remains in the skin for a longer period of tim e, since the concentra tion reaches a plateau and no elimination phase is observed in the sk in profile. On the other hand, similar AUC were calculated for both profiles, which indica tes that despite the different profile the tissue exposure to the drug is similar. After dermal administration by iontophores is, no dexamethasone was detected in the microdialysis samples. There are ma ny factors that could be causing to the dexamethasone levels in the microdialysate samples to be very low, even under the detection limit of the assay. One factor could be the conversion of dexamethasone phosphate into free dexamethasone. Dexameth asone was administered by iontophoresis as a pro-drug, and the RIA is a method selective to the free dexamethasone. If the conversion of the pro-drug into the free dexame thasone in the skin does not happen, or if it happens in a very small scale, the result w ould be very low dexamethasone levels in the microdialysis samples. It is also possible th at only dexamethasone phosphate is present in the samples, which is not detected by RIA. An attempt was made in order to cause the hydrolysis of the phosphate form by adding al kaline phosphatase (P-3895, from Sigma) in the samples. However, no free dexamethasone was measured in the hydrolyzed samples (data not shown), which could be an indication that there is no dexamethasone phosphate in the samples either.

PAGE 118

105 If that is the case, then dexamethas one levels attained after iontophoretic administration are below the ther apeutic levels to claim an anti-inflammatory effect in the subcutaneous and muscle tissues. However, on the day of experiment, a good blanching effect was observed on the skin, which is an indication that dexamethasone was released from the patch. Some in vitro studies usi ng hairless mouse skin have also shown that dexamethasone can be delivery from the patch by iontophoresis in vitro.146 Therefore, it is possible that the drug can not penetrate into deeper layers of the skin, resulting in a more superficial effect. In conclusion, the results here show that microdialysis can be performed to study dexamethasone after oral administration, despite the low levels observed in the microdialysis samples. On the other hand, no dexamethasone was detected in the dialysate after iontophoretic administration of dexamethasone phosphate, which is an indication that the drug can not get through the skin layers in a concentration high enough in order to achieve pharmacological effect.

PAGE 119

106 CHAPTER 7 CONCLUSION The development of a microdialysis met hod for new compounds requires that in vitro experiments be performed first. Th e recovery obtained by in vitro methods represents the maximum recovery that can be obtained for a specific compound. If the in vitro recovery shows low va lues, the compound might not be suitable for in vivo microdialysis. In this study, the in vitro re sults showed good recoveri es for all steroids tested, with exception of beclomethasone di propionate and fluticasone propionate. When the lipophilicity and molecular weight were co mpared to the average recovery of each steroid tested, no correlation was found. It s eems that these two factors do not have a great impact on the recovery, at least for the range studied. The probe membrane is another important factor in microdialysis. The dexamethasone recovery obtained for two different probes, CMA/20 and CMA/60, showed that longer membranes result in higher recovery, which would be expected due to th e increase in the surface area available for diffusion. Lower recoveries were obtained for dexa methasone in the in vivo experiment. The in vivo recovery is generally lower than the in vitro one due to tissue tortuosity. However, the dialysate levels were still within the assay quantification range. The in vivo recovery was assumed to be constant thr oughout the experiment. Mi crodialysis allowed determining the free dexamethasone concentrat ion in the muscle after the administration of a single dexamethasone phosphate dose. The free muscle and free plasma profiles were similar up to the last time point, which dem onstrate that dexamethasone freely distributes

PAGE 120

107 into the muscle. It also confirms the assumption that the recovery was constant throughout the experiment. Dexamethasone muscle levels obtained in animal experiment were above the therapeutic range due to the high dose administer ed to the animals. In the clinical study, the dexamethasone dose administered to the volunteer was the same dose indicated for therapeutic effect. Lower plasma levels were obtained due to the low dose given, which resulted in much lower dialysate concentrati on. The assay sensitivity in this study was the determinant factor in assessing the feasibility of using microdialysis in a clinical setting. The radioimmunoassay was found to be very sens itive, with a lower de tection limit of 0.5 ng/ml; however it also showed to have a hi gh variability. Despite the low plasma levels, it was still possible to detect dexamethasone in the dialysate obtained after oral administration of dexamethasone tablets, whic h shows that microdialysis can still be used to measure free levels in muscle at th erapeutics levels. On the other hand, no dexamethasone was detected after topical administration of dexamethasone phosphate by iontophoresis, which could pr obably be due to no transdermal absorption of the compound. In summary, microdialysis is a useful t echnique to study the drug penetration into different tissues, such as muscle and skin , in pre-clinical and clinical settings.

PAGE 121

108 APPENDIX A DOCETAXEL INDIVIDUAL NO-NET-FLUX PLOTS y = -0.6835x + 2.4911 R2 = 0.9977-5.00 -4.00 -3.00 -2.00 -1.00 0.00 1.00 2.00 3.00 4.00 0.002.004.006.008.0010.00CPerfusateCDialysate-CPerfusate(A) y = -0.6601x + 2.4751 R2 = 0.9969 -4.00 -3.00 -2.00 -1.00 0.00 1.00 2.00 3.00 4.00 0.002.004.006.008.0010.00 CPerfusateCDialysate-CPerfusate (B) y = -0.748x + 3.4746 R2 = 0.9999-4.00 -3.00 -2.00 -1.00 0.00 1.00 2.00 3.00 4.00 0.002.004.006.008.0010.00CPerfusateCDialysate CPerfusate(C) y = -0.545x + 3.4326 R2 = 0.9828-4.00 -3.00 -2.00 -1.00 0.00 1.00 2.00 3.00 4.00 0.002.004.006.008.0010.00CPerfusateCDialysate CPerfusate(D) y = -0.7962x + 4.137 R2 = 0.991-4.00 -3.00 -2.00 -1.00 0.00 1.00 2.00 3.00 4.00 0.002.004.006.008.0010.00CPerfusateCDialysate CPerfusate(E) Figure A-1: Plot of the net ch ange in the concentration be tween perfusate and dialysate versus the perfusate concentration. Th e slope of the curve represents the recovery and the intercept with the x-axis represents the free plasma concentration in the vial

PAGE 122

109 APPENDIX B INDIVIDUAL PHARMACOKI NETICS PARAMETERS OB TAINED IN THE NONCOMPARTMENTAL ANALYSIS OF TH EPLASMA AND MUSCLE PROFILE Table B-1: Pharmacokinetics parameters obtain ed in the non-compart mental analysis of the individual plasma and muscle prof iles obtained after the administration of a 50 mg/kg dose. 50 mg/kg Rat Cmax (mg/l) tmax (min) ke (min-1) t1/2 (min) Vz/Fr (l/kg) Cl/Fr (l/h/kg) AUC0 inf (g/l/min) AUMC0 inf (g/l/min) MRT (min) 1 16.7 60 0.0036 192.1 2.54 0.5490 5.462 1625.692 297.6 2 19.2 30 0.0027 252.8 2.86 0.4698 6.384 2383086 373.3 3 21.5 30 0.0020 356.0 2.52 0.2940 10.202 5325.695 522.0 4 25.7 30 0.0024 288.4 2.53 0.3648 8.230 3455.470 419.9 5 25.7 30 0.0064 107.6 1.52 0.5886 5.098 931.898 182.8 PLASMA 6 19.4 30 0.0020 342.2 3.19 0.3876 7.740 3836.554 495.7 1 3.3 70 0.0042 165.8 10.75 2.70 1.112 309.015 277.7 2 3.1 50 0.0031 222.2 13.27 2.48 1.207 409.047 338.7 3 3.9 70 0.0052 133.8 11.13 3.46 0.867 176.239 203.2 4 3.0 30 0.0043 162.4 12.96 3.32 0.904 238.266 263.6 5 2.7 70 0.0072 96.2 10.65 4.59 0.653 119.284 182.6 MUSCLE 6 3.5 50 0.007 99.2 9.08 3.81 0.788 149.231 189.4

PAGE 123

110 Table B-2: Pharmacokinetics parameters obtain ed in the non-compart mental analysis of the individual plasma and muscle prof iles obtained after the administration of a 100 mg/kg dose.. 100 mg/kg Rat Cmax (mg/l) tmax (min) ke (min-1) t1/2 (min) Vz/Fr (l/kg) Cl/Fr (l/h/kg) AUC0 inf (g/l/min) AUMC0 inf (g/l/min) MRT (min) 7 31.8 30 0.0004 1579 3.79 0.0996 60117.9 137528.28 2287.6 8 31.7 120 0.0104 66.8 1.51 0.9402 6381.8 905.387 141.9 9 36.8 30 0.0014 497.2 3.28 0.2748 21851.1 15930.036 729.0 10 34.1 30 0.0050 138.3 2.39 0.7194 8339.2 1812.630 217.4 11 34.8 30 0.0023 296.3 2.73 0.3828 15661.9 6976.883 445.5 12 38.1 30 0.0024 292.4 2.63 0.3744 16029.1 7131.117 444.9 13 49.5 30 0.0019 365.6 2.36 0.2682 22390.8 11755.958 525.0 PLASMA 14 41.9 60 0.0031 224.6 2.09 0.387 15506.5 5264.973 339.5 7 6.2 110 0.0052 132.9 15.18 4.75 1.264 301.171 238.4 8 6.3 90 0.004 174.2 11.62 2.77 2.164 646.425 298.7 9 4.9 50 0.0052 133.1 16.79 5.24 1.144 239.405 209.3 10 3.1 70 0.0033 207.2 26.6 5.34 1.124 354.515 315.4 11 5.8 30 0.0053 131.9 13.94 4.39 1.366 274.476 200.9 12 4.4 90 0.0026 266.3 20.67 3.23 1.859 746.483 401.5 13 6.4 50 0.0033 207.3 12.55 2.52 2.383 769.561 322.9 MUSCLE 14 5.4 70 0.0025 282.5 15.23 2.24 2.677 1163.910 434.8

PAGE 124

111 APPENDIX C INDIVIDUAL PHARMACO KINETICS PARAMETERS OBTAINED IN THE COMPARTMENTAL ANALYSIS OF TH EPLASMA AND MUSCLE PROFILE Table C-1: Average pharmacokinetics parame ters obtained after the fitting of the individual plasma and microdialysis profiles using the one compartment body model. DOSE Rat ka’ (min-1) ke (min-1)t1/2 (min) V/Fr (l/kg) F’ MSCa 1 0.0573 0.0039 178 2.4 1.31 4.78 2 0.1344 0.0033 210 2.5 1.14 4.17 3 0.1411 0.0025 273 2.3 1.03 4.11 4 0.1354 0.0054 128 1.7 0.82 3.31 5 0.0864 0.0055 126 1.7 0.76 3.92 50 mg/kg 6 0.1682 0.0029 236 2.6 1.11 3.76 7 0.2002 0.0025 281 2.0 0.64 4.07 8 0.0665 0.0033 213 2.0 0.91 5.12 9 0.1506 0.0036 191 3.1 0.97 2.61 10 0.0421 0.0056 125 2.3 0.65 2.74 11 0.1461 0.0045 153 2.6 1.07 3.45 12 0.0699 0.0046 150 2.4 0.85 4.41 13 0.1359 0.0018 394 2.9 1.17 4.17 100 mg/kg 14 0.1521 0.0022 322 2.7 1.03 4.01 a model selection criteria

PAGE 125

112 APPENDIX D INDIVIDUAL PLASMA AND FREE MUSCLE PROFILES Rat #1 050100150200250300Time (min) 10-1100101102 Rat #2 10-1100101102 050100150200250300Time (min) Rat #3 10-1100101102 Time (min) 050100150200250300 Rat # 4 Time (min) 050100150200250300 10-1100101102 Rat #5 10-1100101102 Time (min) 050100150200250300 Rat #6 10-1100101102 050100150200250300Time (min) Figure D-1: Average plasma and free muscle profiles obtained after the i.v. administration of 50 mg/kg dose of de xamethasone disodium phosphate. The symbols represent the measured concentr ations and the line represents the fitted profile according to the model described in Chapter 5. Concentration ( m g /l ) Concentration ( m g /l ) Concentration ( m g /l ) Concentration ( m g /l ) Concentration ( m g /l ) Concentration ( m g /l )

PAGE 126

113 Rat #7 050100150200250300Time (min) 10-1100101102 Rat #8 10-1100101102 050100150200250300Time (min) 050100150200250300 10-1100101102Rat #9Time (min) 10-1100101102Rat #10 Time (min) 050100150200250300 050100150200250300 10-1100101102Rat #11Time (min) Rat #12 Time (min) 050100150200250300 10-1100101102 Rat #13 050100150200250300Time (min) 10-1100101102 Rat #14 050100150200250300Time (min) 10-1100101102 Figure D-2: Average plasma and free muscle profiles obtained after the i.v. administration of 100 mg/kg dose of de xamethasone disodium phosphate. The symbols represent the measured concentra tions and the line represents the line represents the fitted profile according to the model described in Chapter 5. Concentration ( m g /l ) Concentration ( m g /l ) Concentration ( m g /l ) Concentration ( m g /l ) Concentration ( m g /l ) Concentration ( m g /l ) Concentration (mg/l) Concentration ( m g /l )

PAGE 127

114 LIST OF REFERENCES 1. Mueller M, Schmid R, Wagner O, Osten Bv, Shayganfar H, Eichler HG 1995. In vivo characterization of transdermal dr ug transport by microdialysis. J Control Release 37:49-57. 2. Benfeldt E, Groth L 1998. Feasibility of measuring lipophi lic or protein-bound drugs in the dermis by in vivo microdial ysis after topical or systemic drug administration. Acta Derm Venereol 78(4):274-278. 3. Rojas C, Nagaraja NV, Derendorf H 2000. In vitro recovery of triamcinolone acetonide in microdialysis. Pharmazie 55(9):659-662. 4. Rojas C, Nagaraja NV, Webb AI , Derendorf H 2003. Microdialysis of triamcinolone acetonide in rat muscle. J Pharm Sci 92(2):394-397. 5. Hansen DK, Davies MI, Lunte SM , Lunte CE 1999. Pharmacokinetic and metabolism studies using microdialysis sampling. J Pharm Sci 88(1):14-27. 6. Joukhadar C, Frossard M, Mayer BX, Br unner M, Klein N, Siostrzonek P, Eichler HG, Muller M 2001. Impaired target site penetration of beta-lactams may account for therapeutic failure in patients with septic shock. Crit Care Med 29(2):385-391. 7. Muller M 2002. Microdia lysis. BMJ 324(7337):588-591. 8. Connely CA 1999. Microdialysis update : Optimizing the advantages. Journal of Physiology 512(2):303. 9. Lindefors N, Amberg G, Ungerstedt U 1989. Intercerebral microdialysis: I. Experimental studies of diffusion kinetics. J Pharmacol Met 22:141-156. 10. Stagni G, O'Donnel D, Liu YJ, Kello g DL, Shepherd AMM 1999. Iontophoretic current and intradermal microdialysis r ecovery in humans. J Pharmacol Toxicol 41:49-54. 11. Kehr J 1993. A survey on quantitative microdialysis: theoretical models and practical implications. J Ne urosci Methods 48(3):251-261. 12. Chaurasia CS 1999. In vivo microdialysis sampling: theory and applications. Biomed Chromatogr 13(5):317-332.

PAGE 128

115 13. Elmquist WF, Sawchuk RJ 1997. Applica tion of microdialysis in pharmacokinetic studies. Pharm Res 14(3):267-288. 14. Sasongko L, Williams KM, Ramzan I, McLachlan AJ 2000. Assessment of in vitro and in vivo recovery of gallamine using microdialysis. J Pharmacol Toxicol Methods 44(3):519-525. 15. Benveniste H 1989. Brain microdialys is. J of Neurochem 52(6):1667-1679. 16. Stenken JA 1999. Methods and issues in microdialysis calibration. Anal Chim Acta 379:337-358. 17. Eliasson A. 1991. Microdialysis: prin ciples of recovery, Stockholm: CMA Microdialysis. 18. Hsiao JK, Ball BA, Morrison PF, Mefford IN, Bungay PM 1990. Effects of different semipermeable membranes on in vitro and in vivo performance of microdialysis probes. J Neurochem 54(4):1449-1452. 19. Bungay PM, Morrison PF, Dedrick RL 1990. Steady-state theory for quantitative microdialysis of solutes and water in vivo and in vitro. Life Sci 46:105-119. 20. Jacobson I, Sandberg M, Hamberger A 1985. Mass transfer in brain dialysis devicesA new method for the estima tion of extracellular amino acids concentration. J Neurosci Met 15:263-268. 21. Menacherry S, Hubert W, Justice JB, Jr. 1992. In vivo calibration of microdialysis probes for exogenous compounds. Anal Chem 64(6):577-583. 22. Zhao Y, Laing X, Lunte C 1995. Comparis on of recovery and delivery in vitro for calibration of microdialysis probe s. Anal Chimica Acta 316:403-410. 23. Larsson CI 1991. The use of an "internal standard" for control of the recovery in microdialysis. Life Sci 49(13):PL73-78. 24. Le Quellec A, Dupin S, Genissel P, Saivin S, Marchand B, Houin G 1995. Microdialysis probes ca libration: Gradient and tissu e dependent changes in no net flux and reverse dialysis methods. J Ph armacol Toxicol Methods 33(1):11-16. 25. Ungerstedt U 1991. Microdi alysisPrinciples and app lications for studies in animals and man. J Intern Med 230(4):365-373. 26. Groth L, Jorgensen A 1997. In vitro microdialysis of hydrophilic and lipophilic compounds. Anal Chimica Acta 355:75-83. 27. Deguchi Y, Terasaki T, Kawasaki S, Tsuji A 1991. Muscle microdialysis as a model study to relate the drug concentr ation in tissue inte rstitial fluid and dialysate. J Pharmacobiodyn 14(8):483-492.

PAGE 129

116 28. Benfeldt E 1999. In vivo microdialysis for the investig ation of drug levels in the dermis and the effect of barrier pertuba tion on cutaneous drug penetration. Acta Derm Venereol 206:1-55. 29. de Lange EC, de Boer AG, Brei mer DD 2000. Methodological issues in microdialysis sampling for pharmacokine tic studies. Adv Drug Deliv Rev 45(23):125-148. 30. Ault JM, Riley CM, Meltzer NM, Lunte CE 1994. Dermal microdialysis sampling in vivo . Pharm Res 11(11):1631-1639. 31. Davies MI, Lunte CE 1995. Microdia lysis sampling for hepatic metabolism studies. Impact of microdialysis probe design and implantation technique on liver tissue. Drug Metab Dispos 23(10):1072-1079. 32. Palsmeier RK, Lunte CE 1994. Microdi alysis sampling in tumor and muscle: study of the disposition of 3-amino-1, 2,4-benzotriazine-1,4-di-N-oxide (SR 4233). Life Sci 55(10):815-825. 33. Anderson C, Andersson T, Wardell K 1994. Changes in skin circulation after insertion of a microdialysis probe visua lized by laser Doppler perfusion imaging. J Invest Dermatol 102(5):807-811. 34. Miller MA, Geary RS 1991. RIA-linked mi crodialysis sampling in the awake rat: application to free-drug pharmacokinetic s of hydrocortisone. J Pharm Biomed Anal 9(10-12):901-910. 35. Parrot S, Bert L, Mouly-Badina L, Sauvinet V, Colussi-Mas J, Lambas-Senas L, Robert F, Bouilloux JP, Suaud-Chagny MF, Denoroy L, Renaud B 2003. Microdialysis monitoring of catecholamines and excitatory amino acids in the rat and mouse brain: recent developments based on capillary electrophoresis with laser-induced fluorescence detectionA mini-review. Cell Mol Neurobiol 23(45):793-804. 36. Qian J, Wu Y, Yang H, Michael AC 1999. An integrated decoupler for capillary electrophoresis with electrochemical detec tion: Application to analysis of brain microdialysate. Anal Chem 71:4486-4492. 37. Dawson LA 1997. Capillary electrop horesis and microdialysis: Current technology and applications. J Chromat B 697:89-99. 38. Mayer BX, Namiranian K, Dehghanyar P, Stroh R, Mascher H, Muller M 2003. Comparison of UV and tandem mass sp ectrometric detection for the highperformance liquid chromatographic determ ination of diclofenac in microdialysis samples. J Pharm Biomed Anal 33(4):745-754.

PAGE 130

117 39. Nakashima K, Yamamoto K, Al-Dir bashi OY, Nakashima MN 2002. Disposition of triazolam in the rat by brain microdialysis and semi-micro column highperformance liquid chromatography with UV absorbance detection. Biomed Chromatogr 16(3):219-223. 40. Yoshitake T, Fujino K, Kehr J, Ishida J, Nohta H, Yamaguchi M 2003. Simultaneous determination of norepine phrine, serotonin, and 5-hydroxyindole-3acetic acid in microdialysis samples fr om rat brain by microbore column liquid chromatography with fluorescence detec tion following derivatization with benzylamine. Anal Biochem 312(2):125-133. 41. Tsunoda M, Mitsuhashi K, Masuda M, Imai K 2002. Simultaneous determination of 3,4-dihydroxyphenylacetic aci d and homovanillic acid using high performance liquid chromatography-fluor escence detection and app lication to rat kidney microdialysate. Anal Biochem 307(1):153-158. 42. Zhang W, Cao X, Xie Y, Ai S, Jin L, Jin J 2003. Simultaneous determination of the monoamine neurotransmitters and gl ucose in rat brain by microdialysis sampling coupled with liquid chromatogr aphy-dual electrochemical detector. J Chromatogr B Analyt Technol Biomed Life Sci 785(2):327-336. 43. Chaurasia CS, Chen CE, Ashby Jr CR 1999. In vivo on-line HPLC-microdialysis: Simultaneous detection of monoamines a nd their metabolites in awake freelymoving rats. J Pharm Biom Anal 19:413-422. 44. Bergstrom SK, Markides KE 2002. On-lin e coupling of microdialysis to packed capillary column liquid chromatographytandem mass spectrometry demonstrated by measurement of free concentrations of ropivacaine and metabolite from spiked plasma samples. J Chromatogr B Analyt Technol Biomed Life Sci 775(1):79-87. 45. Kerns E, Volk KJ, Klohr SE, Lee MS 1999. Monitoring in vitro experiments using microdialysis sampling on-line with mass spectrometry. J Pharm Biomed Anal 20:115-128. 46. Hows M, Organ A, Murray S, Dawson L, Foxton R, Heidbreder C, Hughes Z, Lacroix L, Shah A 2002. High-performa nce liquid chromatography/tandem mass spectrometry assay for the rapid high sensitivity measurement of basal acetylcholine from microdialysat es. J Neurosci Met 121:33-39. 47. Davies MI, Cooper JD, Desmond SS, Lunte CE, Lunte SM 2000. Analytical considerations for microdialysis samp ling. Adv Drug Deliv Rev 45(2-3):169-188. 48. Fettweis G, Borlak J 1996. Topics in xenobiochemistryApplication of microdialysis techniques in pharmacokine tic studies. Xenobiotica 26(5):473-485. 49. Garrison KE, Pasas SA, Cooper JD, Davi es MI 2002. A review of membrane sampling from biological tissues with applications in pharmacokinetics, metabolism and pharmacodynamics. Eur J Pharm Sci 17(1-2):1-12.

PAGE 131

118 50. Stahle L 1992. Pharmacokinetic estimati ons from microdialysis data. Eur J Clin Pharmacol 43(3):289-294. 51. Stahle L 1993. Microdialysis in pharmacokinetics. Eur J Drug Metab Pharmacokinet 18(1):89-96. 52. Johansen MJ, Newman RA, Madden T 1997. The use of microdialysis in pharmacokinetics and pharmacodynamics. Pharmacotherapy 17(3):464-481. 53. Tsai TH 2003. Assaying protein unbound dr ugs using microdialysis techniques. J Chromatogr B Analyt Technol Biomed Life Sci 797(1-2):161-173. 54. de la Pena A, Liu P, Derendorf H 2000. Microdialysis in pe ripheral tissues. Adv Drug Deliv Rev 45(2-3):189-216. 55. Joukhadar C, Derendorf H, Muller M 2001. Microdialysis. A novel tool for clinical studies of anti-infective agen ts. Eur J Clin Pharmacol 57(3):211-219. 56. Brunner M, Muller M 2002. Microdialysi s: An in vivo approach for measuring drug delivery in oncology. Eur J Clin Pharmacol 58(4):227-234. 57. Davies MI 1999. A review of mi crodialysis sampling for pharmacokinetics applications. Analytica Chimica Acta 379:227-249. 58. Muller M 2000. Microdialysis in clini cal drug delivery studies. Adv Drug Deliv Rev 45(2-3):255-269. 59. Deguchi Y, Yokoyama Y, Sakamoto T, Hayashi H, Naito T, Yamada S, Kimura R 2000. Brain distribution of 6-mercaptopur ine is regulated by the efflux transport system in the blood-brain barrier. Life Sci 66(7):649-662. 60. Xie R, Hammarlund-Udenaes M 1998. Blood-brain barrier equilibration of codeine in rats studied with mi crodialysis. Pharm Res 15(4):570-575. 61. Hammarlund-Udenaes M, Paalzow LK, de Lange EC 1997. Drug equilibration across the blood-brain barrier--pharmacoki netic considerations based on the microdialysis method. Pharm Res 14(2):128-134. 62. Chu J, Gallo JM 2000. Application of microdialysis to characterize drug disposition in tumors. Adv Drug Deliv Rev 45(2-3):243-253. 63. Dukic S, Heurtaux T, Kaltenbach ML, Hoizey G, Lallemand A, Gourdier B, Vistelle R 1999. Pharmacokinetics of me thotrexate in the ex tracellular fluid of brain C6-glioma after intravenous infu sion in rats. Pharm Res 16(8):1219-1225. 64. Ekstrom PO, Andersen A, Warre n DJ, Giercksky KE, Slordal L 1996. Determination of extracellular methotrexate tissue levels by microdialysis in a rat model. Cancer Chemother Pharmacol 37(5):394-400.

PAGE 132

119 65. Bourne JA 2003. Intracerebral microd ialysis: 30 years as a tool for the neuroscientist. Clin Exp Pharmacol Physiol 30(1-2):16-24. 66. Siddiqui MM, Shuaib A 2001. Intracereb ral microdialysis and its clinical application: A review . Methods 23(1):83-94. 67. de Lange EC, Danhof M, de Boer AG, Breimer DD 1997. Methodological considerations of intracerebral microdial ysis in pharmacokinetic studies on drug transport across the blood-brain barrier. Brain Res Brain Res Rev 25(1):27-49. 68. Sawchuk RJ, Elmquist WF 2000. Microdialy sis in the study of drug transporters in the CNS. Adv Drug Deliv Rev 45(2-3):295-307. 69. Muller M, Schmid R, Georgopoulos A, Buxbaum A, Wasicek C, Eichler HG 1995. Application of microdialysis to clin ical pharmacokinetics in humans. Clin Pharmacol Ther 57(4):371-380. 70. Liu P, Muller M, Grant M, Webb AI, Obermann B, Derendorf H 2002. Interstitial tissue concentrations of cefpodoxime. J Antimicrob Chemother 50 Suppl:19-22. 71. Muller M, Haag O, Burgdorff T, Georgopoul os A, Weninger W, Jansen B, Stanek G, Pehamberger H, Agneter E, Eichler HG 1996. Characterization of peripheralcompartment kinetics of antibiotics by in vivo microdialysis in humans. Antimicrob Agents Chemother 40(12):2703-2709. 72. Brunner M, Pernerstorfer T, Mayer BX , Eichler HG, Muller M 2000. Surgery and intensive care procedures affect the targ et site distribution of piperacillin. Crit Care Med 28(6):1754-1759. 73. Joukhadar C, Stass H, Muller-Zellenberg U, Lackner E, Kovar F, Minar E, Muller M 2003. Penetration of moxifloxacin into healthy and inflamed subcutaneous adipose tissues in humans. Antimicr ob Agents Chemother 47(10):3099-3103. 74. Zeitlinger MA, Dehghanyar P, Mayer BX, Schenk BS, Neckel U, Heinz G, Georgopoulos A, Muller M, Joukhadar C 2003. Relevance of soft-tissue penetration by levofloxacin for target site bacterial killing in patients with sepsis. Antimicrob Agents Chemother 47(11):3548-3553. 75. Muller M, Brunner M, Hollenstein U, J oukhadar C, Schmid R, Minar E, Ehringer H, Eichler HG 1999. Penetration of cipr ofloxacin into the interstitial space of inflamed foot lesions in non-insulindependent diabetes mellitus patients. Antimicrob Agents Chemother 43(8):2056-2058. 76. de la Pea A, Dalla Costa T, Talton JD , Rehak E, Gross J, Thyroff-Friesinger U, Webb AI, Mueller M, Derendorf H 2001. Penetration of cef aclor into the interstitial space fluid of skeletal musc le and lung tissue in rats. Pharm Res 18:1310-1314.

PAGE 133

120 77. Nolting A, Dalla Costa T, Vist elle R, Rand KH, Derendorf H 1996. Determination of free extracellular concentrations of piperacillin by microdialysis. J Pharm Sci 85(4):369-372. 78. Sauermann R, Zeitlinger M, Erovic BM, Marsik C, Georgopoulos A, Muller M, Brunner M, Joukhadar C 2003. Pharmacodynamics of piperacillin in severely ill patients evaluated by using a PK/PD mode l. Int J Antimicrob Agents 22(6):574578. 79. Jain R 1998. Delivery of molecular a nd cellular medicine to solid tumors. J Control Release 53:49-67. 80. Johansen MJ, Thapar N, Newman RA , Madden T 2002. Use of microdialysis to study platinum anticancer ag ent pharmacokinetics in preclinical models. J Exp Ther Oncol 2(3):163-173. 81. Muller M, Mader RM, Steiner B, Steger GG, Jansen B, Gnant M, Helbich T, Jakesz R, Eichler HG, Blochl-Daum B 1997. 5-fluorouracil kinetics in the interstitial tumor space: C linical response in breast cancer patients. Cancer Res 57(13):2598-2601. 82. Ekstrom PO, Andersen A, Saeter G, Giercksky KE, Slor dal L 1997. Continuous intratumoral microdialysis during high-dose methotrexa te therapy in a patient with malignant fibrous histiocytoma of the femur: a case report. Cancer Chemother Pharmacol 39(3):267-272. 83. Ekstrom PO, Giercksky KE, Anders en A, Bruland OS, Slordal L 1997. Intratumoral differences in methotrexate levels within human osteosarcoma xenografts studied by microdialys is. Life Sci 61(19):PL275-280. 84. Devineni D, Klein-Szanto A, Gallo JM 1996. Uptake of temozolomide in a rat glioma model in the presence and absen ce of the angiogenesis inhibitor TNP-470. Cancer Res 56(9):1983-1987. 85. Berner B, John VA 1994. Pharmacokine tic characterisation of transdermal delivery systems. Clin Pharmacokinet 26(2):121-134. 86. Shah VP, Maibach HI. 1993. Topical drug bioavailability, bioequivalence, and penetration. ed., New York: Plenum Press. 87. Shah VP, Flynn GL, Yacobi A, Maibach HI, Bon C, Fleischer NM, Franz TJ, Kaplan SA, Kawamoto J, Lesko LJ, Mart y JP, Pershing LK, Schaefer H, Sequeira JA, Shrivastava SP, Wilkin J, Williams RL 1998. Bioequivalence of topical dermatological dosage formsMethods of evaluation of bioequivalence. Pharm Res 15(2):167-171.

PAGE 134

121 88. Mueller M, Rastelli C, Ferri P, Ja nsen B, Breiteneder H, Eichler HG 1998. Transdermal penetration of diclofenac after multiple epicutaneous administration. J Rheumatol 25(9):1833-1836. 89. Stagni G, O'Donnel D, Liu YJ, Lejjog DL, Morgan T, Shepherd AMM 2000. Intradermal microdialysis: kinetics of i ontophoretically delivery propanolol in forearm dermis. J Control Release 63:331-339. 90. Mueller M, Mascher H, Kikutra C, Sch aefer S, Brunner M, Dorner G, Eichler HG 1997. Diclofenac concentrations in de fined tissue layers after topical administration. Clin Pharmacol Ther 62(3):293-299. 91. Singh P, Maibach HI 1994. Transdermal iontophoresis. Pharmacokinetic considerations. Clin Phar macokinet 26(5):327-334. 92. Benfeldt E, Serup J 1999. Effect of barrier perturbation on cutaneous penetration of salicylic acid in hairless rats: in vi vo pharmacokinetics using microdialysis and non-invasive quantification of barrier function. Arch Dermatol Res 291(9):517526. 93. Benfeldt E, Serup J, Menne T 1999. E ffect of barrier perturbation on cutaneous salicylic acid penetration in human skin: in vivo pharmacokinetics using microdialysis and noninvasive quantifica tion of barrier function. Br J Dermatol 140(4):739-748. 94. Kreilgaard M 2002. Assessment of cuta neous drug delivery using microdialysis. Adv Drug Deliv Rev 54 Suppl 1:S99-S121. 95. FDA. 1998. Guidance for industry. T opical dermatologi cal drug product NDAs and ANDAs In vivo bioavailability, bioe quivalence, in v itro release, and associated studies., ed.: FDA. 96. Nakashima M, Zhao MF, Ohya H, Sakurai M, Sasaki H, Matsuyama K, Ichikawa M 1996. Evaluation of in vivo transdermal absorption of cyclosporin with absorption enhancer using intradermal mi crodialysis in rats. J Pharm Pharmacol 48:1143-1146. 97. Matsuyama K, Nakashima M, Ichikawa M, Yano T, Satoh S, Goto S 1994. In vivo microdialysis for the transdermal absorption of valproate in rats. Biol Pharm Bull 17(10):1395-1398. 98. Fang JY, Sung KC, Lin HH, Fang CL 1999. Transdermal iontophoretic delivery of diclofenac sodium from various poly mer formulations: In vitro and in vivo studies. Int J Pharm 178(1):83-92. 99. Matsuyama K, Nakashima M, Nakaboh Y, Ichikawa M, Yano T, Satoh S 1994. Application of in vivo microd ialysis to transdermal absorption of methotrexate in rats. Pharm Res 11(5):684-686.

PAGE 135

122 100. Tegeder I, Muth-Selbach U, Lotsch J, Rusing G, Oelkers R, Brune K, Meller S, Kelm GR, Sorgel F, Geisslinger G 1999. Application of microdialysis for the determination of muscle and subcutaneous tissue concentrations after oral and topical ibuprofen administration. Clin Pharmacol Ther 65(4):357-368. 101. Pershing LK, Corlett J, Jorgensen C 1994. In vivo pharmacokinetics and pharmacodynamics of topical ketoconazole and miconazole in human stratum corneum. Antimicrob Agents Chemother 38(1):90-95. 102. Ao X, Sellati TJ, Stenken JA 2004. Enha nced microdialysis re lative recovery of inflammatory cytokines using antibodycoated microspheres analyzed by flow cytometry. Anal Chem 76(13):3777-3784. 103. Trickler WJ, Miller DW 2003. Use of os motic agents in microdialysis studies to improve the recovery of macromolecules. J Pharm Sci 92(7):1419-1427. 104. Khramov AN, Stenken JA 1999. Enhan ced microdialysis recovery of some tricyclic antidepressants and structurally related drugs by cyclodextrin-mediated transport. Analyst 124(7):1027-1033. 105. Kendrick KM 1990. Microdialysis measurem ent of in vivo neur opeptide release. J Neurosci Methods 34(1-3):35-46. 106. Ernberg MM, Alstergren PJ 2004. Micr odialysis of neuropeptide Y in human muscle tissue. J Neurosci Methods 132(2):185-190. 107. Schmelz M, Luz O, Averbeck B, Bickel A 1997. Plasma extravasation and neuropeptide release in human skin as measured by intradermal microdialysis. Neurosci Lett 230(2):117-120. 108. Hastings JA, McClure-Sharp JM, Morris MJ 1998. In vitro studies of endogenous noradrenaline and NPY overflow from th e rat hypothalamus during maturation and ageing. Naunyn Schmiedebergs Arch Pharmacol 357(3):218-224. 109. Gruber SH, Mathe AA 2000. Effects of typical and atypica l antipsychotics on neuropeptide Y in rat brain tissue and microdialysates from ventral striatum. J Neurosci Res 61(4):458-463. 110. Cirrito JR, May PC, O'Dell MA, Taylor JW, Parsadanian M, Cramer JW, Audia JE, Nissen JS, Bales KR, Paul SM, DeMattos RB, Holtzman DM 2003. In vivo assessment of brain interstitial fluid with microdialysis reveals plaque-associated changes in amyloid-beta metabolism and half-life. J Neurosci 23(26):8844-8853. 111. Cook CJ 2001. Measuring of extracellu lar cortisol and cor ticotropin-releasing hormone in the amygdala using immunosensor coupled microdialysis. J Neurosci Methods 110(1-2):95-101.

PAGE 136

123 112. Weidner C, Klede M, Rukwied R, Lische tzki G, Neisius U, Skov PS, Petersen LJ, Schmelz M 2000. Acute effects of substan ce P and calcitonin ge ne-related peptide in human skinA microdialysis study. J Invest Dermatol 115(6):1015-1020. 113. Stahl M, Bouw R, Jackson A, Pay V 2002. Human microdialysis. Curr Pharm Biotechnol 3(2):165-178. 114. Muller M, Burgdorff T, Jansen B, Singer EA, Agneter E, Dorner G, Brunner M, Eichler HG 1997. In vivo drug-response measurements in target tissues by microdialysis. Clin Pharmacol Ther 62(2):165-170. 115. Bouw MR, Gardmark M, Hammar lund-Udenaes M 2000. Pharmacokineticpharmacodynamic modelling of morphine tr ansport across the blood-brain barrier as a cause of the antinociceptive effect delay in ratsA microdialysis study. Pharm Res 17(10):1220-1227. 116. Benveniste H, Huttemeier PC 1990. Microd ialysisTheory an d application. Prog Neurobiol 35(3):195-215. 117. Carneheim C, Stahle L 1991. Mi crodialysis of lipophilic compounds: A methodological study. Pharmacol Toxicol 69(5):378-380. 118. Clarke SJ, Rivory LP 1999. Clinical pharmacokinetics of docetaxel. Clin Pharmacokinet 36(2):99-114. 119. Nuijen B, Bouma M, Schellens JH, Beijn en JH 2001. Progress in the development of alternative pharmaceutical formulations of taxanes. Invest New Drugs 19(2):143-153. 120. Shah VP, Midha KK, Findlay JW, Hill HM, Hulse JD, McGilveray IJ, McKay G, Miller KJ, Patnaik RN, Powell ML, Tone lli A, Viswanathan CT, Yacobi A 2000. Bioanalytical method validationA revis it with a decade of progress. Pharm Res 17(12):1551-1557. 121. Song D, Hsu LF, Au JL 1996. Binding of ta xol to plastic and glass containers and protein under in vitro conditions. J Pharm Sci 85(1):29-31. 122. Thiesen J, Kramer I 1999. Physico-chemi cal stability of docet axel premix solution and docetaxel infusion solutions in PVC bags and polyolefine containers. Pharm World Sci 21(3):137-141. 123. Rosing H, Lustig V, Koopman FP, ten Bokkel Huinink WW, Beijnen JH 1997. Bio-analysis of docetaxel and hydroxylat ed metabolites in human plasma by highperformance liquid chromatography and automated solid-phase extraction. J Chromatogr B Biomed Sci Appl 696(1):89-98. 124. Derendorf H 1997. Pharmacokinetic and pharmacodynamic properties of inhaled corticosteroids in relation to efficacy and safety. Respir Med 91 Suppl A:22-28.

PAGE 137

124 125. Jusko WJ, Ludwig EA. 1992. Corticoste roids. In Evans WE, Schentag JJ, Jusko WJ, editors. Applied pharmacokinetics : Principles of therapeutic drug monitoring, Vancouver: Applie d Therapeutic. p 21-34. 126. Peets EA, Staub M, Symchowicz S 1969. Plasma binding of betamethasone-3 H, dexamethasone-3 H, and cortisol-14 C-A comparitive study. Biochem Pharmacol 18:1655-1663. 127. Sun L, Stenken JA 2003. Improving mi crodialysis extraction efficiency of lipophilic eicosanoids. J Pharm Biomed Anal 33(5):1059-1071. 128. Krishnaswami S, Mollmann H, Derendor f H, Hochhaus G 2000. A sensitive LCMS/MS method for the quantification of fl uticasone propionate in human plasma. J Pharm Biomed Anal 22(1):123-129. 129. Hansch C, Sammes PG, Taylor, JB 1990. Cumulative subject index and drug compendium. 1st ed., Oxford: Pergamon Press. 130. Shah VP, Midha KK, Dighe S, McGilver ay IJ, Skelly JP, Yacobi A, Layoff T, Viswanathan CT, Cook CE, McDowall RD, Pittman KA, Spector S 1992. Analytical methods validation: Bi oavailability, bioequivalence, and pharmcokinetics studies. J Pharm Sci 81(3):309-312. 131. Rohdewald P, Mollmann H, Barth J, Rehder J, Derendorf H 1987. Pharmacokinetics of dexamethasone a nd its phosphate ester. Biopharm Drug Dispos 8(3):205-212. 132. Muller M, Brunner M, Schmid R, Putz EM, Schmiedberger A, Wallner I, Eichler HG 1998. Comparison of three different e xperimental methods for the assessment of peripheral compartment pharmacokinetics in humans. Life Sci 62(15):PL227234. 133. Hare LE, Yeh KC, Ditzler CA, McMahon FG, Duggan DE 1975. Bioavailability of dexamethasone. II. Dexamethasone phosphate. Clin Pharmacol Ther 18(3):330-337. 134. Kroin JS, Schaefer RB, Penn RD 2000. Chronic intrathecal administration of dexamethasone sodium phosphate: Pharm acokinetics and neurotoxicity in an animal model. Neurosurgery 46(1):178-182; discussion 182-173. 135. Derendorf H, rohdewald P, Hoc hhaus G, Moellmann H 1986. HPLC determination of glucocorticoid alcoho ls, their phosphates and hydrocortisone in aqueous solutions and biological fluids . J Pharm Biomed Anal 4(2):197-206. 136. Tsuei SE, Moore RG, Ashley JJ, McBr ide WG 1979. Disposition of synthetic glucocorticoids.1. Pharmacokinetics of dexamethasone in healthy adults. J Pharmacok Biopharm 7(3):249-264.

PAGE 138

125 137. Legat FJ, Maier A, Dittrich P, Zenahlik P, Kern T, Nuhsbaumer S, Frossard M, Salmhofer W, Kerl H, Muller M 2003. Penetration of fosfomycin into inflammatory lesions in patients with cellulitis or diabetic foot syndrome. Antimicrob Agents Chemother 47(1):371-374. 138. Dietzel K, Estes kS, Brewster ME, Bodor NS, Derendorf H 1990. The use of 2hydroxypropyl-b-cyclodextrin as a vehicl e for intravenous administration of dexamethasone in dogs. Int J Pharm 59:225-230. 139. Schnetz E, Fartasch M 2001. Microdia lysis for the evalua tion of penetration through the human skin barrier A promissi ng tool for the futu re research? Eur J Pharm Sci 12:165-174. 140. Brazzini B, Pimpinelli N 2002. New and established topical corticosteroids in dermatology: Clinical pharmacology and th erapeutic use. Am J Clin Dermatol 3(1):47-58. 141. Schaefer-Korting M, Gysler A. 1997. Topical glucocorticoids with improved benefit/risk ratio. In Korting H, Schaef er-Korting M, editors . The benefit/risk ratio: A handbook for the rational use of potentially hazardous drugs, Boca Raton: CRC Press. 142. Caldwell JR 1996. Intra-articular cor ticosteroids. Guide to selection and indications for use. Drugs 52(4):507-514. 143. Hochhaus G, Moellmann H, Barth J 1990. Glucocorticoids for the intraarticular administration: Pharmacodynamic considerations. Akt Rheumatol 15:1-12. 144. Hochhaus G, Hochhaus R, Werb er G, Derendorf H, Mollmann H 1992. A selective HPLC/RIA for dexamethasone and its prodrug dexamethasone-21sulphobenzoate sodium in biological fluids. Biomed Chromatogr 6(6):283-286. 145. Hochhaus G, Froehlich P, Hochhaus R, Mollmann A, Derendorf H, Mollmann HW 1998. A selective HPLC/R IA for the determination of budesonide. J Pharm Biomed Anal 17(8):1235-1242. 146. Petelenz TJ, Buttke JA, Bonds C, Lloys LB, Beck JE, Stephen RL, Jacobsen SC, Rodriguez P 1992. Iontophoresis of dexa methasone: Laboratory studies. J Control Release 210:55-66.

PAGE 139

126 BIOGRAPHICAL SKETCH Virna J. A. Schuck started school at the Fe deral University of Rio Grande do SulBrazil, in 1994. During pharmacy school, she started doing research under the supervision of Dr. Elfrides E. S. Schapoval. At that time she had a chance of working in different projects performing pharmacological experiments with medicinal plant extracts and doing bioequivalence and bioavailability st udies. She graduated in 1994 and started a specialization course about pharmaceutical indus try. In 1997 she started graduate school at the Federal University of Rio Grande do SulBrazil, pursuing a master’s degree. During graduate school, she pe rformed a clinical study of pharmacokinetics/pharmacodynamics of ranitidin e. In 2000 she received her master’s title and moved to Florida to join Dr. Hart mut Derendrof’s group at the University of Florida. During graduate school at the Un iversity of Florida she worked with microdialysis and its application in pharmacokinetics, graduating in 2004.