- Email: [email protected]
An integrated pharmacokinetic–pharmacodynamic approach to optimization of R-apomorphine delivery in Parkinson's disease
Advanced Drug Delivery Reviews 33 (1998) 253–263
L
An integrated pharmacokinetic–pharmacodynamic approach to optimization of R-apomorphine delivery in Parkinson’s disease M. Danhof a , *, R. Van der Geest a , T. Van Laar b , H.E. Bodde´ 1 ,a a
Leiden– Amsterdam Center for Drug Research, Divisions of Pharmacology and Pharmaceutical Technology, Sylvius Laboratory, P.O. Box 9503, 2300 RA Leiden, The Netherlands b Leiden University Medical Center, Department of Neurology, Rijnsburgerweg 10, P.O. Box 9600, 2300 RC Leiden, The Netherlands Received 14 January 1998; received in revised form 5 February 1998; accepted 13 February 1998
Abstract R-apomorphine is a mixed dopamine D 1 / D 2 receptor agonist which is potentially useful in the management of Parkinson’s disease. The delivery of R-apomorphine is complicated however by a number of pharmacokinetic and pharmacodynamic factors. This review describes the development of a transdermal iontophoretic delivery system for R-apomorphine on the basis of integrated pharmacokinetic–pharmacodynamic (PK / PD) investigations in patients with idiopathic Parkinson’s disease. The pharmacokinetics and metabolic pathways of R-apomorphine were determined following intravenous infusion of 30 mg kg 21 in 15 min in 10 patients. A stepwise infusion protocol was used to determine the therapeutic window. A wide interindividual variability in both pharmacokinetics and pharmacodynamics and a narrow therapeutic concentration range were observed. This shows the need for individualized and carefully controlled delivery of R-apomorphine in Parkinson’s disease. Transdermal iontophoretic transport was studied both in vitro in human stratum corneum and dermatomed full skin and in vivo in patients with Parkinson’s disease. These studies showed that the delivery of R-apomorphine by transdermal iontophoresis is feasible and furthermore that the rate of delivery can be carefully controlled by variation of the current density. It is concluded that the delivery of R-apomorphine by transdermal iontophoresis may be an attractive tool in future clinical pharmacological investigations in patients with Parkinson’s disease aiming at characterization of the influence of chronic treatment and disease progression on the pharmacokinetics and pharmacodynamics. Ultimately these studies may result in a system which is suitable for clinical application. 1998 Elsevier Science B.V. All rights reserved. Keywords: R-apomorphine; Pharmacokinetics; Pharmacodynamics; Pharmacokinetic–pharmacodynamic correlation; Controlled drug delivery; Transdermal iontophoresis; Parkinson’s disease
Contents 1. General introduction ................................................................................................................................................................ 1.1. Pharmacokinetic–pharmacodynamic modelling .................................................................................................................. 1.2. Controlled (transdermal) drug delivery .............................................................................................................................. 2. Pharmacokinetic–pharmacodynamic studies on R-apomorphine..................................................................................................
1
*Corresponding author. Tel.: 1 31-71-527-6211; fax 1 31-71-527-6292; e-mail: [email protected] Deceased September 8, 1996.
0169-409X / 98 / $ – see front matter 1998 Elsevier Science B.V. All rights reserved. PII: S0169-409X( 98 )00033-7
254 254 255 256
254
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
2.1. Pharmacokinetics, enantiomer interconversion and metabolic pathways ............................................................................... 2.2. Stepwise intravenous infusion to determine the therapeutic window..................................................................................... 3. Transdermal iontophoresis of R-apomorphine............................................................................................................................ 3.1. In vitro characterization of iontophoretic transport in human skin ........................................................................................ 3.2. Feasibility of transdermal iontophoresis in vivo .................................................................................................................. 4. Delivery of R-apomorphine by transdermal iontophoresis: conclusions and perspectives............................................................... References ..................................................................................................................................................................................
1. General introduction R-apomorphine is a potent D 1 / D 2 receptor agonist which has been used for a long time in the treatment of Parkinson’s disease [1,2]. The treatment with R-apomorphine however is not without complications. Upon oral administration, the drug is rapidly degraded in the gastrointestinal tract and subject to a high ‘first-pass’ effect, resulting in an oral bioavailability of 1.7% [3–5]. In addition, high oral doses of R-apomorphine may cause gastrointestinal complications and have been associated with nephrotoxicity. Therefore, oral administration is not considered a suitable delivery route. At present R-apomorphine is most commonly administered by repeated subcutaneous injections or by subcutaneous infusion [6]. This route of administration however invariably results in the appearance of subcutaneous nodules. More recently alternative routes of administration, such as the nasal, the sublingual and the rectal route have been studied [7–9]. So far however this has not resulted in a delivery system that is suitable for widespread clinical application. When designing an optimal drug delivery system, in general two fundamental questions should be taken into consideration [10,11]: 1. a clinical pharmacological one in terms of the optimal rate and timing at which the drug should be delivered; this requires profound knowledge of the factors that determine the time course of the pharmacological response in vivo in relation to the delivery profile (i.e. continuous versus pulsatile delivery); 2. a pharmaceutical technological one in terms of the most suitable system that can provide the required rate and time specifications; this requires knowledge on the capacity, flexibility, rate and time programming possibilities.
256 257 258 258 260 261 262
Ideally pharmacokinetic / pharmacodynamic modelling studies, with rate and time controlled input as important variables, should be conducted prior to the development of a suitable delivery system. In this contribution some general aspects of this approach will first be reviewed and discussed. Subsequently, the application of this approach will be illustrated for the development of a transdermal iontophoretic delivery system of R-apomorphine in Parkinson’s disease.
1.1. Pharmacokinetic–pharmacodynamic modelling At present pharmacokinetic considerations still play an important role in developing dosing regimens of new drugs. Important progress has been made however with a more comprehensive approach based on modelling of the relationship between pharmacokinetics and pharmacodynamics [12]. Such modelling allows for the characterization and the prediction of the time course of drug action rather than concentration and provides a scientific basis for development of the dosing regimen. This is shown schematically in Fig. 1. The development of pharmacokinetic–pharmacodynamic modelling has recently been reviewed and its relevance in drug development has been discussed [13]. Initial pharmacokinetic–pharmacodynamic models focused on the characterization of rapidly reversible direct drug effects. These models typically consist of three components: (1) a pharmacokinetic model characterizing the time course of drug (and metabolite) concentrations in blood or plasma; (2) a pharmacodynamic model, characterizing the relationship between drug concentration and pharmacological response intensity; and (3) often, a link model which serves to account for the delay of the effect relative to the plasma concentration [14]. More
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
Fig. 1. Schematic representation of the interrelationship between drug delivery and input (dose), pharmacokinetics (concentration), pharmacodynamics and clinical effects. Modelling of the relationship between pharmacokinetics and pharmacodynamics provides the scientific basis for the optimal delivery profile of a drug.
recently emphasis has been given to the development of models for drugs with an indirect mechanism of action, where the delay between drug concentration and effect is largely determined by the processes in the pharmacodynamic phase, rather than by slow distribution to the site of action [15,16]. The models for drugs with direct and indirect mechanisms of action are complementary, and have been applied successfully to the effects of a large number of drugs. There is no doubt that there will be further developments in pharmacokinetic–pharmacodynamic modelling. An important area, particularly in relation to drug delivery, is the development of models to characterize functional adaptation (i.e. tolerance) during long term drug treatment and under the influence of disease. Such modelling may provide critical information with regard to the most optimal delivery profile. Zero-order delivery is in many instances still considered ideal. In the meantime however there is clear evidence that for certain drugs and diseases, pulsatile or ‘on demand’ delivery may be optimal [17].
1.2. Controlled (transdermal) drug delivery For the controlled delivery of drugs to the systemic circulation various routes of administration are available, including oral and intravenous routes as well as a number of alternate routes like rectal,
255
buccal, sublingual, nasal and transdermal administration [18,19]. Each of these routes has specific advantages and disadvantages with regard to the opportunities for controlled drug delivery. Of the various routes of administration, the oral route shows the poorest control of both the rate and the extent of absorption, whereas the intravenous route provides optimal opportunities for control of the plasma concentration vs. time profile. The latter however is associated with many practical limitations. Other routes of administration may be considered as realistic alternatives. The choice of a route is thereby dependent on the type of concentration vs. time profile that is pursued. In situations where a constant plasma concentration vs. time profile is desirable, the rectal (using osmotic drug delivery devices) and the transdermal route of administration may be very suitable. On the other hand, if more fluctuating concentration vs. time profiles are preferred, buccal / sublingual and nasal administrations may be suitable alternatives [19]. In the management of Parkinson’s disease dopaminergic drugs are administered chronically over periods of several years. This makes transdermal delivery in theory an attractive approach. A major limitation of using the transdermal route for systemic drug delivery however is the high intrinsic barrier of the skin, which resides in the stratum corneum. Large and hydrophilic compounds cross this barrier with difficulty. In addition, the transport of highly lipophilic compounds can be hindered by poor partitioning from the stratum corneum to lower hydrophilic epidermal layers [20,21]. Another limitation of transdermal drug delivery is that, at least with the classical (passive) delivery systems, it is very difficult to control the concentration vs. time profile in blood. In general it is not possible to achieve significant concentrations in the blood rapidly. Furthermore, drug delivery does not always stop upon removal of the system, as a drug depot may have formed in the stratum corneum [22]. For a wide variety of drugs it has been shown in in vitro studies that by iontophoresis a significant increase of the transdermal transport rate as well as an accurate control of the input rate can be achieved [23]. Application of iontophoresis for the transdermal delivery of dopaminergic drugs in Parkinson’s disease may therefore be an attractive approach.
256
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
2. Pharmacokinetic–pharmacodynamic studies on R-apomorphine
2.1. Pharmacokinetics, enantiomer interconversion and metabolic pathways When studying the pharmacokinetic–pharmacodynamic correlation of a drug it is essential that various potentially complicating pharmacokinetic factors are taken into consideration [24]. The pharmacokinetics and metabolism of R-apomorphine are indeed quite complex and this may have important pharmacological and clinical implications. R-apomorphine is an optically active compound which is administered as the pure R-enantiomer (Fig. 2). In theory interconversion into the S-enantiomer may occur. Sapomorphine has been shown to act as an antagonist at both the D 1 and D 2 receptor subtypes [25]. Another factor is that the enzyme catechol-O-methyl transferase (COMT) may be involved in the metabolism of R-apomorphine, resulting in the formation of the metabolites apocodeine and iso-apocodeine. Apocodeine has been shown to act as a dopamine agonist, albeit that its potency appears to be relatively low [26]. Other metabolic pathways include
glucuronide and sulfate conjugation [27]. In order to fully understand the pharmacokinetic–pharmacodynamic correlation of R-apomorphine, it is essential to have detailed quantitative information on the various metabolic pathways and in particular the plasma concentration profiles of S-apomorphine and the metabolites apocodeine and isoapocodeine. Within the context of developing an integrated pharmacokinetic–pharmacodynamic model of Rapomorphine, its disposition was studied in 10 patients with idiopathic Parkinson’s disease following intravenous infusion of 30 mg.kg 21 in 15 min. The plasma concentrations of both R- and S-apomorphine as well as the concentrations of apocodeine and iso-apocodeine were determined using newly developed (enantioselective) HPLC assays [28]. In most patients the plasma concentration vs. time profile of R-apomorphine was best described by a biexponential function. The values of the relevant pharmacokinetic parameters were: clearance 40615 ml min 21 kg 21 , volume of distribution at steadystate 1.660.5 l kg 21 and terminal half-life 41613 min. Interestingly the value of the clearance exceeds hepatic blood flow indicating that metabolism must occur to a significant extent in extrahepatic tissues.
Fig. 2. Chemical structures of R-apomorphine and its potential metabolites. From a pharmacodynamic point of view in particular the interconversion into S-apomorphine and the formation of the methylated metabolites apocodeine and isoapocodeine are important. (From [29].)
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
This raises the very important question whether indeed R-apomorphine is eliminated to a significant extent by COMT as this enzyme is present in large quantities in extrahepatic tissues. Furthermore the considerable interindividual variability in the pharmacokinetic parameters with values of the clearance ranging between 24–69 ml min 21 kg 21 is an important observation. In plasma no measurable concentrations of Sapomorphine nor of the metabolites apocodeine and isoapocodeine were observed. The metabolism of R-apomorphine was further characterized on the basis of the excretion of unchanged R-apomorphine, S-apomorphine, apocodeine and iso-apocodeine and their respective glucuronide and sulfate conjugates in urine. The total excretion of unconjugated S-apomorphine, apocodeine and isoapocodeine was less than 0.1% of the total administered dose. The total excretion of unchanged R-apomorphine, apomorphine-sulfate and apomorphine glucuronide amounted to 0.360.4%, 3.861.0% and 6.062.2% of the administered dose, respectively [29]. It appears therefore that the formation of S-apomorphine and the formation of apocodeine and iso-apocodeine do not contribute to the elimination of R-apomorphine in man. Therefore these compounds do not seem to contribute to the pharmacological response observed upon the intravenous administration of a single dose of R-apomorphine. Unfortunately however, only 10% of the administered dose can be accounted for via the elimination pathways that have been studied so far. An important question remains therefore via which pathways the rest is eliminated. One possible pathway is non-enzymatic oxidation, which may indeed contribute to a significant extent to the elimination. The oxidation products of R-apomorphine are not likely to be pharmacologically active [30]. Another pathway which has not been studied so far and which may be important from a pharmacodynamic point of view is oxidative N-demethylation resulting in the formation of norapomorphine. It cannot be excluded that this metabolite exhibits pharmacological activity. To what extent the formation of this metabolite contributes to the elimination of R-apomorphine remains to be determined. It is unlikely to be a major elimination pathway however, since N-demethylation is confined primarily to the
257
liver and most of the elimination of R-apomorphine occurs in extrahepatic tissues.
2.2. Stepwise intravenous infusion to determine the therapeutic window Characterization of the concentration–effect relationship is a central issue in integrated pharmacokinetic–pharmacodynamic investigations. Typically in this kind of study, a single intravenous dose of a drug is administered and the time course of the pharmacological response is determined in conjunction with the time course of the concentration in plasma. By linking the effects to the concentration, concentration–effect relationships are obtained in individual patients. In the case of R-apomorphine however the application of this approach is complicated by its short elimination half-life. As a result the plasma concentrations following an intravenous bolus administration change rapidly, thereby not allowing sufficient time for a meaningful quantification of the pharmacological response intensity. In order to overcome this problem, a stepwise intravenous infusion protocol was developed, which results in multiple relatively stable plasma concentrations within each individual patient. According to this protocol the pharmacokinetic–pharmacodynamic correlation of R-apomorphine was studied in 10 patients with idiopathic Parkinson’s disease. In these studies intravenous infusion of R-apomorphine was started at a rate of 10 mg kg 21 h 21 , and increased every 20 min by 10 mg kg 21 h 21 , until the maximum tolerated infusion rate or the infusion rate of 100 mg kg 21 h 21 was reached. Thereafter the infusion rate was decreased every 20 min by 10 mg kg 21 h 21 until baseline. At the end of each infusion interval a blood sample was obtained for determination of the concentration of R-apomorphine and the pharmacological effect was measured. The primary effect parameters were tapping score and tremor assessment. To determine the tapping score, the patient had to tap two buttons, with a distance of 30 cm over 30 s, with the right hand as well as the left hand at maximal speed. Tremor was assessed on a 4-point scale (0 5 absent, 1 5 mild, 2 5 moderate, 3 5 severe). When the tremor score decreased with at least 2 points this was considered a positive clinical response. Side
258
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
effect was monitored on the basis of dyskinesia. For that purpose the same 4-point scale was used separately for the head, trunk and the four extremities, resulting in a maximal dyskinesia score of 18 points. Increase of the dyskinesia score with at least 4 points was considered clinically relevant. Adverse effects (systolic blood pressure below 90 mmHg; bradycardia: heart rate below 50 min; nausea; dizziness) were monitored continuously. Fig. 3 shows the time course of the plasma concentration of R-apomorphine and of the effect on the tapping score in a representative patient. When the plasma concentration increases gradually beyond the minimally effective concentration, a sharp increase in the effect to its maximum value is observed. This maximum effect is maintained until, with a decrease in the infusion rate, the concentration drops to subtherapeutic values and the effect returns to baseline. Within each individual patient the concentrations at onset and offset of effect were generally quite similar. Upon examination of the concentration–effect relationships it appeared that the effect was quantal rather than continuous. Therefore the pharmacodynamics were parameterized in terms of minimal effective concentration (MEC) at the onset of the beneficial effect (tremor decrease or 25% tap score increase), the minimal dyskinesia concentration (MDC) as the onset concentration of the increase in
dyskinesia and finally, the minimal toxic concentration (MTC) as the onset concentration of the adverse effects (nausea, vomiting, sleepiness, dizziness, hypotension). Between patients wide differences were observed with respect to the minimal concentrations for each of the effects. Of the 10 patients, eight responded to R-apomorphine and two were non-responders. In the responders the values of the minimally effective concentration (MEC) varied between 1.4–10.7 ng ml 21 , whereas the concentration at which dyskinesia was observed (MDC) varied between 2.7–20.0 ng ml 21 . The minimal toxic concentration varied between 8.5–24.5 ng ml 21 . In Fig. 4 the individual values of the minimal effective concentration (MEC) and the minimal dyskinesia score (MDC) are shown. Between patients there is a wide variability in the therapeutic window ( 5 the difference between the MEC on one hand and the MDC on the other) of R-apomorphine, which is typically quite narrow [31]. The wide interindividual variability in both pharmacokinetics and pharmacodynamics, in combination with the narrow therapeutic window in many patients underlines the need of individualized and carefully controlled delivery of R-apomorphine in the management of Parkinson’s disease.
3. Transdermal iontophoresis of R-apomorphine
3.1. In vitro characterization of iontophoretic transport in human skin
Fig. 3. The plasma concentration vs. time profile and the effect on the tapping score (number of taps / 30 s of both the left and the right arm) upon a stepwise intravenous infusion of R-apomorphine in an individual patient with Parkinson’s disease. The maximum infusion rate of R-apomorphine in this patient was 70 mg kg 21 h 21 . (From [30].)
Fundamental studies on the transdermal iontophoretic transport of drugs are typically conducted in in vitro test systems with excised human skin. In the studies on R-apomorphine transport a newly designed continuous flow-through transport cell was used, where the dimensions were optimized to obtain realistic estimations of the time course of the intrinsic flux upon the application of different electrical current profiles [32]. In initial studies R-apomorphine transport was studied in human stratum corneum, as the barrier to transport presumably resides in this layer. In all the experiments R-apomorphine was applied in the anodal compartment. The effect on the flux of the following parameters was studied: current density, pH, concentration, ionic strength, osmolari-
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
259
Fig. 4. Individual values of the minimal effective concentration (MEC 5 25% increase in tapping score) and the minimal dyskinesia concentration (MDC) in patients with idiopathic Parkinson’s disease (patients 1, 2, 5, 6 and 7). When no dyskinesia was noted the maximum concentration was plotted (patients 3 and 4). Note the wide inter-individual variability in the therapeutic window ( 5 MDC 2 MEC). (From [30].)
ty, buffer strength, temperature and skin type. The results of these investigations showed that the transdermal transport rate of R-apomorphine can be directly controlled by manipulation of the current density (Fig. 5). When no electrical current is applied, passive delivery is minimal and no depot effect is observed. Application of electrical current (500 mA cm 22 over 5 h at a temperature of 208C) resulted in enhancement of the delivery of Rapomorphine to 9066 nmol cm 22 h 21 . A steadystate flux was achieved in | 3 h, corresponding to a lag time of 40 min. The reversibility of the enhanced membrane permeability to the ‘normal’ state was tested by measuring the passive flux after switching off the current. In a period of | 3 h an exponential decay to a value of | 1 / 20th of the steady-state flux was observed. In the in vitro experiments various current densities in the range of 0–500 mA cm 22 were tested. Thereby a linear relationship (r 2 5 0.98) between the current density and the steady-state
Fig. 5. Time course of the flux of R-apomorphine across human stratum corneum in vitro at a donor concentration of 5 mM, pH 5 5 and temperature 5 208C (d, passive flux through nonpretreated skin; j, passive flux after 5 h of current pretreatment, 500 mA cm 22 without apomorphine; ♦, 5 h of iontophoretic delivery at 500 mA cm 22 followed by 3 h of passive delivery). (From [32].)
260
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
ratio between the transdermal transport rates in vivo and in vitro it can be calculated that therapeutically effective plasma concentrations can be reached with a transdermal iontophoretic patch of 20 cm 2 , which is realistic from a clinical point of view [33]. The findings of these in vitro studies show that in theory the delivery of R-apomorphine by transdermal iontophoresis is feasible and furthermore that modulation of the current density may be used to control (and thereby to individualize) the transdermal transport rate. Fig. 6. Relationship between current density and steady-state flux of R-apomorphine in human stratum corneum in vitro. (Experimental conditions: donor concentration 5 15 mM; pH 5 4; temperature 5 208C; current density range 5 100–500 mA cm 22 ). (From [32].)
transdermal flux was observed (Fig. 6), indicating that variation of the current density may be used to control the delivery of R-apomorphine across the skin. In general only minimal effects of changes in osmolarity, buffer strength, ionic strength and pH on the transdermal transport rate were observed. Temperature on the other hand was found to be a major determinant of both the maximum steady-state transport rate and the lag time. Upon increase of the temperature from 208C to 378C the steady-state flux was increased by a factor 2.3 and the lag time was decreased to 63 min. In the studies with stratum corneum, skin samples of different donors were tested. Only minimal interindividual variability in the maximal steady-state transport rate and the lag time were observed. The transdermal transport of Rapomorphine was subsequently studied in dermatomed full skin to determine the difference with stratum corneum. In dermatomed full skin the maximum steady-state flux was | 30% lower than in stratum corneum. An important question is whether the transdermal transport rates that have been observed in the in vitro experiments are sufficiently high to achieve therapeutically effective plasma concentrations. Thereby the pharmacokinetic properties and the therapeutic window of R-apomorphine that have been determined in previous in vivo investigations, must be taken into consideration. Assuming a one-to-one
3.2. Feasibility of transdermal iontophoresis in vivo In order to determine the feasibility of transdermal iontophoresis of R-apomorphine in vivo, a study was conducted in 10 patients with idiopathic Parkinson’s disease. According to a randomized cross-over design the patients received R-apomorphine by transdermal iontophoresis at a current density of either 250 or 375 mA cm 22 over 1 h on one occasion. At another occasion 30 mg kg 21 were administered intravenously over 15 min. In none of the patients was significant passive transdermal transport of R-apomorphine observed. Upon current application, increasing plasma concentrations were observed in all patients. The observed maximum concentrations were directly related to the applied current density: 1.360.6 ng ml 21 at 250 mA cm 22 and 2.560.7 ng ml 21 at 375 mA cm 22 . When the current was switched off all concentrations returned to baseline values in | 90 min (Fig. 7). These results show that also in the in vivo situation the transdermal transport rate of R-apomorphine is determined by the applied current density. By mathematical deconvolution it was demonstrated that steady-state transdermal flux values are reached within 1 h of current application. The values of the steady-state flux were 69630 nmol cm 22 h 21 at 250 mA cm 22 and 114634 nmol cm 22 h 21 at 375 mA cm 22 , indicating that variation in the applied current density may be used to control the delivery. The delivery of R-apomorphine at the aforementioned rates results in steady-state plasma concentrations of | 3–4.5 ng ml 21 which is at the lower end of the therapeutic concentration range. In this study also an assessment was made of the local skin irritation and
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
261
Table 1 Comparison of the transdermal transport characteristics of Rapomorphine in human skin in vitro and in vivo. The transport rates are normalized to a current density of 250 mA cm 22 . The donor concentration was 15 mM and the pH was 5 in both situations Experimental condition
Human skin, in vivo Stratum corneum, in vitro Dermatomed skin, in vitro Fig. 7. Average plasma concentration time profiles of R-apomorphine upon transdermal iontophoretic delivery in patients with idiopathic Parkinson’s disease at current densities of 250 mA cm 22 . s, n 5 4 and 375 mA cm 22 ; d, n 5 5, for 1 h starting at t 5 60 min. Steady-state plasma concentrations were predicted by fitting the data to a mathematical model describing the concentration vs. time profile upon zero-order delivery in a twocompartment pharmacokinetic model (from [33].)
toxicity. Local skin erythema was assessed visually and quantified by laser Doppler flowmetry (LDF). Upon removal of the transdermal iontophoretic patch mild transient erythema was observed which was reflected in elevated LDF values. No local adverse effects such as prolonged inflammation and green colouring (which is typically seen upon subcutaneous R-apomorphine administration) were observed. These findings indicate that the transdermal delivery of R-apomorphine is feasible at acceptable levels of skin irritation [34]. An interesting question is how the transdermal transport observed in the in vivo study compares to the transport observed in vitro. It appears that the steady-state transport rate in vivo at a current density of 250 mA cm 22 is remarkably similar to the value observed both in stratum corneum and in dermatomed full skin in vitro (Table 1). With respect to the time to reach steady-state a different situation is observed. The value of the time to reach steady-state of 10–20 min in vivo is in the same range as in stratum corneum in vitro. In the dermatomed full skin preparation however a much larger value is observed. Therefore in vitro studies in stratum corneum appears to have a better predictive value with regard to the in vivo transdermal transport
Steady-state transdermal flux (nmol cm 22 h 21 )
Time to reach steady-state (min)
69630
10–20
101613
15–30
65611
300
characteristics of R-apomorphine than studies in dermatomed full skin. The absence of dermal blood flow in the dermatomed full skin preparation may be an important factor in this respect.
4. Delivery of R-apomorphine by transdermal iontophoresis: conclusions and perspectives In this paper a new approach to optimization of delivery by transdermal iontophoresis of R-apomorphine in Parkinson’s disease is described which is based on a combination of in vivo and in vitro studies. First in the in vivo studies the pharmacokinetic–pharmacodynamic correlation was determined, which provides crucial information on the optimal delivery profile of R-apomorphine in patients with idiopathic Parkinson’s disease. Next the transdermal iontophoretic transport characteristics of Rapomorphine were determined in human skin in vitro. Finally, transdermal iontophoresis was applied in an in vivo study in patients with idiopathic Parkinson’s disease. The results of the various investigations show that precise control of the transdermal delivery of Rapomorphine is feasible, but that the concentrations that can be reached in this way are at the lower end of the therapeutic concentration range of 1.4 to 10.7 ng ml 21 . Many patients need concentrations that are significantly higher than is currently feasible. Furthermore, in clinical practice it may be necessary to make rapid dose adjustments. It is therefore important to explore the possibilities to further enhance
262
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
the transdermal transport rate. In this respect the application of neutral transport enhancers with small headgroups in combination with iontophoresis may be an interesting approach. An important question is whether by using this strategy it will still be possible to accurately control the delivery on the basis of modulation of the current density. Furthermore skin toxicity may become a critical issue. The results from the studies on the pharmacokinetic–pharmacodynamic correlation show that the effects can be directly related to the plasma concentration of R-apomorphine. A wide interindividual variation in the therapeutic concentration range was observed. An important question is which factors are the determinants of this interindividual variability. In this respect a particularly intriguing question is to what extent disease progression contributes to the variability in pharmacodynamics. In Parkinson’s disease, disease progression has been associated with different subtypes of patients which differ in their response to dopaminergic drugs. Both responders and non-responders can be distinguished. The responders can be further classified as ‘stable’, ‘wearing-off’ and ‘fluctuating’, depending on the severity of the disease. An important question is how the therapeutic concentration range of R-apomorphine changes with the progression of the disease. This requires studies in a large population of patients with Parkinson’s disease. Furthermore longitudinal studies in individual patients will be of considerable interest. Another important question is whether tolerance develops to the therapeutic and / or side effects of R-apomorphine. A fascinating question in this respect is to what extent the rate and extent of functional tolerance development is influenced by the delivery profile (i.e. zero-order versus pulsatile delivery). Delivery of R-apomorphine by transdermal iontophoresis may prove to be of great value in future clinical pharmacological investigations, addressing the aforementioned issues. In such studies transdermal iontophoresis may be used as a non-invasive alternative for stepwise intravenous infusion. Furthermore it may be of great value to provide zeroorder, pulsatile or ‘on demand’ delivery in studies which aim at characterization of functional tolerance development with chronic treatment. However the application of transdermal iontophoresis for these
indications requires that a system be available allowing higher transdermal transport rates than currently feasible. Delivery of dopaminergic drugs by transdermal iontophoresis is not limited to R-apomorphine, but can in principle also be applied to clinical pharmacological investigations with other drugs. Ultimately these developments may result in transdermal iontophoretic delivery systems that are useful for clinical application in Parkinson’s disease.
References [1] A.J. Lees, Apomorphine and Parkinson’s disease, in: J.P.W.F. Lakke, E.M. Delhaas, A.W.F. Rutgers (Eds.), Parenteral Drug Therapy in Spasticity and Parkinson’s Disease, The Parthenon Publishing Group, Carnforth, 1992, pp. 185–192. [2] J.L. Neumeyer, S. Law, R.J. Baldessarini, Historical highlights of the chemistry, pharmacology and early clinical uses of apomorphine, in: G.L. Gessa, G.U. Corsini (Eds.), Apomorphine and other Dopamimetics, Raven Press, New York, 1981, pp. 1–17. [3] R.J. Baldessarmni, G.W. Arana, N.S. Kula, A. Campbell, M. Harding, Preclinical studies of the pharmacology of aporphines, in: G.L. Gessa, G.U. Corsini (Eds.), Apomorphine and other Dopamimetics, Raven Press, New York, 1981, pp. 219–228. [4] G.C. Cotzias, P.S. Papavasiliou, E.S. Tolosa, J.S. Mendez, M. Bell-Midura, Treatment of Parkinson’s disease with apomorphines, N. Engl. J. Med. 294 (1976) 567–572. [5] S.T. Gancher, J.G. Nutt, W.R. Woodward, Absorption of apomorphine by various routes in parkinsonism, Mov. Disord. 6 (1991) 212–216. [6] C.M. Stibe, P.A. Kempster, A.J. Lees, G.M. Stern, Subcutaneous apomorphine in parkinsonian on-off oscillations, Lancet 1 (1988) 403–406. [7] T. Van Laar, E.N.H. Jansen, A.W.G. Essink, C. Neef, Intranasal apomorphine in parkinsonian on-off fluctuations, Arch. Neurol. 49 (1992) 482–484. [8] J.L. Montastruc, O. Rascol, J.M. Senard, Sublingual apomorphine in Parkinson’s disease: a clinical and pharmacokinetic study, Clin. Neuropharmacol. 14 (1991) 432–437. [9] T. Van Laar, E.N.H. Jansen, A.W.G. Essing, W.J. Rutten, C. Neef, Rectal apomorphine: a new treatment modality in Parkinson’s disease, J. Neurol. Neurosurg. Psychiatry 55 (1992) 737–738. [10] D.D. Breimer, Pharmacokinetic and pharmacodynamic basis for peptide drug delivery system design, J. Control. Release 21 (1992) 5–10. [11] D.D. Breimer, An integrated pharmacokinetic and pharmacodynamic approach to controlled drug delivery, J. Drug. Target. 3 (1996) 411–415. [12] M. Danhof, J.W. Mandema, Modeling of relationships be-
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
tween pharmacokinetics and pharmacodynamics, in: P.G. Welling, F.L.S. Tse (Eds.), Pharmacokinetics — Regulatory, Industrial, Academic Perspectives, Marcel Dekker, New York, 1995, pp. 139–174. D.D. Breimer, M. Danhof, Relevance of the application of pharmacokinetic–pharmacodynamic modelling concepts in drug development. The ‘Wooden Shoe’ paradigm, Clin. Pharmacokinet. 32 (1997) 259–267. N.H.G. Holford, L.B. Sheiner, Understanding the dose-effect relationship: clinical applications of pharmacokinetic–pharmacodynamic models, Clin. Pharmacokinet. 6 (1981) 429– 453. N.L. Dayneka, V. Garg, W.J. Jusko, Comparison of four basic models of indirect pharmacodynamic responses, J. Pharmacokinet. Biopharm. 21 (1993) 457–478. W.J. Jusko, H.C. Ko, Physiologic indirect response models characterize diverse types of pharmacodynamic effects, Clin. Pharmacol. Ther. 56 (1994) 406–419. D.D. Breimer, The need for rate and time programming in future drug delivery. Pharmacokinetic, pharmacodynamic and clinical considerations, in: R. Gurny, H.E. Junginger, N.A. Peppas (Eds.), Pulsatile Drug Delivery — Current Applications and Future Trends, Wissenschaftliche Verlagsgesellschaft mbH, Stuttgart, 1993, pp. 25–37. L.Z. Benet, Input factors as determinants of drug activity: route, dose, dosage regimen and the drug delivery system, in: F.G. McMahon (Ed.), Principles and Techniques of Human Research and Therapeutics, Futura, New York, 1974, pp. 9–23. M. Danhof, Parenteral neuropharmacological kinetics: pharmacokinetic aspects of controlled drug delivery to the brain, in: J.P.W.F. Lakke, E.M. Delhaas, A.W.F. Rutgers (Eds.), Parenteral Drug Therapy in Spasticity and Parkinson’s Disease, The Parthenon Publishing Group, Carnforth, 1992, pp. 163–176. J.E. Shaw, S.K. Chandrasekaran, Transdermal therapeutic systems, in: L.F. Prescott, W.S. Nimmo (Eds.), Drug Absorption, ADIS Press, Sydney, 1979, pp. 186–193. W.S. Nimmo, Recent clinical experience with transdermal systems, in: L.F. Prescott, W.S. Nimmo (Eds.), Novel Drug Delivery, Churchill Livingstone, Edinburgh, 1989, pp. 209– 221. B.H. Sage, Iontophoresis, in: E.W. Smith, H.I. Maibach (Eds.), Percutaneous Penetration Enhancers, CRC Press, Boca Raton, 1995, pp. 351–368. N.H. Yoshida, M.S. Roberts, Structure-transport relationships in transdermal iontophoresis, Adv. Drug. Del. Rev. 9 (1992) 239–264.
263
[24] M. Danhof, J.W. Mandema, A.M. Stijnen, Pharmacokinetic complexities of in vivo pharmacodynamic investigations, in: C.J. Van Boxtel, N.H.G. Holford, M. Danhof (Eds.), The In Vivo Study of Drug Action — Concepts and Applications of Kinetic-Dynamic Modelling, Elsevier, Amsterdam, 1992, pp. 31–60. [25] M.E. Goldman, J.W. Kebabian, Aporphine enantiomers: interactions with D 1 and D 2 dopamine receptors, Mol. Pharmacol. 25 (1984) 18–23. [26] J.G. Cannon, R.V. Smith, A. Modiri, S.P. Sood, R.J. Borgman, M.A. Aleem, J.P. Long, Centrally acting emetics: preparation and pharmacology of 10 hydroxy-11-methoxyapomorphine (isoapocodeine). In vitro enzymatic methylation of apomorphine, J. Med. Chem. 15 (1972) 273–276. [27] P.N. Kaul, E. Brochmann-Hansen, E.L. Way, Biological disposition of apomorphine. Urinary excretion and organ distribution of apomorphine, J. Pharm. Sci. 50 (1961) 244– 247. ¨ [28] R. Van der Geest, P.P. Kruger, J.M. Gubbens-Stibbe, T. Van ´ M. Danhof, Assay of R-apomorphine, Laar, H.E. Bodde, S-apomorphine, apocodeine, isoapocodeine and their glucuronide and sulfate conjugates in plasma and urine of patients with Parkinson’s disease, J. Chromatogr. Biomed. Appl. 702 (1997) 131–141. [29] R. Van der Geest, T. Van Laar, J.M. Gubbens-Stibbe, H.E. ´ R.A.C. Roos, M. Danhof, Pharmacokinetics, enantioBodde, mer interconversion and metabolism of R-apomorphine in patients with idiopathic Parkinson’s disease, Clin. Neuropharmacol. 21 (1998) 159–166. [30] K. Rehse, G. Piesker, R. Horowski, Neuropsychotrope ¨ und Toxicitat ¨ von Oxidationsprodukten des Aktivitat Apomorphins, Arch. Pharmacol. 311 (1977) 360–363. ´ P.H. [31] T. Van Laar, R. Van der Geest, M. Danhof, H.E. Bodde, Gossens, R.A.C. Roos, Stepwise intravenous infusion of R-apomorphine to determine the therapeutic window in patients with Parkinson’s disease, Clin. Neuropharmacol. 21 (1998) 152–158. ´ Validation and [32] R. Van der Geest, M. Danhof, H.E. Bodde, testing of a new iontophoretic continuous flow through transport cell, J. Control. Release 51 (1998) 85–91. ´ Iontophoretic [33] R. Van der Geest, M. Danhof, H.E. Bodde, delivery of apomorphine I. In vitro optimization and validation, Pharm. Res. 14 (1997) 1797–1802. [34] R. Van der Geest, T. Van Laar, J.M. Gubbens-Stibbe, H.E. ´ M. Danhof, Iontophoretic delivery of R-apomorphine Bodde, II: an in vivo study in patients with Parkinson’s disease, Pharm. Res. 14 (1997) 1803–1809.
L
An integrated pharmacokinetic–pharmacodynamic approach to optimization of R-apomorphine delivery in Parkinson’s disease M. Danhof a , *, R. Van der Geest a , T. Van Laar b , H.E. Bodde´ 1 ,a a
Leiden– Amsterdam Center for Drug Research, Divisions of Pharmacology and Pharmaceutical Technology, Sylvius Laboratory, P.O. Box 9503, 2300 RA Leiden, The Netherlands b Leiden University Medical Center, Department of Neurology, Rijnsburgerweg 10, P.O. Box 9600, 2300 RC Leiden, The Netherlands Received 14 January 1998; received in revised form 5 February 1998; accepted 13 February 1998
Abstract R-apomorphine is a mixed dopamine D 1 / D 2 receptor agonist which is potentially useful in the management of Parkinson’s disease. The delivery of R-apomorphine is complicated however by a number of pharmacokinetic and pharmacodynamic factors. This review describes the development of a transdermal iontophoretic delivery system for R-apomorphine on the basis of integrated pharmacokinetic–pharmacodynamic (PK / PD) investigations in patients with idiopathic Parkinson’s disease. The pharmacokinetics and metabolic pathways of R-apomorphine were determined following intravenous infusion of 30 mg kg 21 in 15 min in 10 patients. A stepwise infusion protocol was used to determine the therapeutic window. A wide interindividual variability in both pharmacokinetics and pharmacodynamics and a narrow therapeutic concentration range were observed. This shows the need for individualized and carefully controlled delivery of R-apomorphine in Parkinson’s disease. Transdermal iontophoretic transport was studied both in vitro in human stratum corneum and dermatomed full skin and in vivo in patients with Parkinson’s disease. These studies showed that the delivery of R-apomorphine by transdermal iontophoresis is feasible and furthermore that the rate of delivery can be carefully controlled by variation of the current density. It is concluded that the delivery of R-apomorphine by transdermal iontophoresis may be an attractive tool in future clinical pharmacological investigations in patients with Parkinson’s disease aiming at characterization of the influence of chronic treatment and disease progression on the pharmacokinetics and pharmacodynamics. Ultimately these studies may result in a system which is suitable for clinical application. 1998 Elsevier Science B.V. All rights reserved. Keywords: R-apomorphine; Pharmacokinetics; Pharmacodynamics; Pharmacokinetic–pharmacodynamic correlation; Controlled drug delivery; Transdermal iontophoresis; Parkinson’s disease
Contents 1. General introduction ................................................................................................................................................................ 1.1. Pharmacokinetic–pharmacodynamic modelling .................................................................................................................. 1.2. Controlled (transdermal) drug delivery .............................................................................................................................. 2. Pharmacokinetic–pharmacodynamic studies on R-apomorphine..................................................................................................
1
*Corresponding author. Tel.: 1 31-71-527-6211; fax 1 31-71-527-6292; e-mail: [email protected] Deceased September 8, 1996.
0169-409X / 98 / $ – see front matter 1998 Elsevier Science B.V. All rights reserved. PII: S0169-409X( 98 )00033-7
254 254 255 256
254
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
2.1. Pharmacokinetics, enantiomer interconversion and metabolic pathways ............................................................................... 2.2. Stepwise intravenous infusion to determine the therapeutic window..................................................................................... 3. Transdermal iontophoresis of R-apomorphine............................................................................................................................ 3.1. In vitro characterization of iontophoretic transport in human skin ........................................................................................ 3.2. Feasibility of transdermal iontophoresis in vivo .................................................................................................................. 4. Delivery of R-apomorphine by transdermal iontophoresis: conclusions and perspectives............................................................... References ..................................................................................................................................................................................
1. General introduction R-apomorphine is a potent D 1 / D 2 receptor agonist which has been used for a long time in the treatment of Parkinson’s disease [1,2]. The treatment with R-apomorphine however is not without complications. Upon oral administration, the drug is rapidly degraded in the gastrointestinal tract and subject to a high ‘first-pass’ effect, resulting in an oral bioavailability of 1.7% [3–5]. In addition, high oral doses of R-apomorphine may cause gastrointestinal complications and have been associated with nephrotoxicity. Therefore, oral administration is not considered a suitable delivery route. At present R-apomorphine is most commonly administered by repeated subcutaneous injections or by subcutaneous infusion [6]. This route of administration however invariably results in the appearance of subcutaneous nodules. More recently alternative routes of administration, such as the nasal, the sublingual and the rectal route have been studied [7–9]. So far however this has not resulted in a delivery system that is suitable for widespread clinical application. When designing an optimal drug delivery system, in general two fundamental questions should be taken into consideration [10,11]: 1. a clinical pharmacological one in terms of the optimal rate and timing at which the drug should be delivered; this requires profound knowledge of the factors that determine the time course of the pharmacological response in vivo in relation to the delivery profile (i.e. continuous versus pulsatile delivery); 2. a pharmaceutical technological one in terms of the most suitable system that can provide the required rate and time specifications; this requires knowledge on the capacity, flexibility, rate and time programming possibilities.
256 257 258 258 260 261 262
Ideally pharmacokinetic / pharmacodynamic modelling studies, with rate and time controlled input as important variables, should be conducted prior to the development of a suitable delivery system. In this contribution some general aspects of this approach will first be reviewed and discussed. Subsequently, the application of this approach will be illustrated for the development of a transdermal iontophoretic delivery system of R-apomorphine in Parkinson’s disease.
1.1. Pharmacokinetic–pharmacodynamic modelling At present pharmacokinetic considerations still play an important role in developing dosing regimens of new drugs. Important progress has been made however with a more comprehensive approach based on modelling of the relationship between pharmacokinetics and pharmacodynamics [12]. Such modelling allows for the characterization and the prediction of the time course of drug action rather than concentration and provides a scientific basis for development of the dosing regimen. This is shown schematically in Fig. 1. The development of pharmacokinetic–pharmacodynamic modelling has recently been reviewed and its relevance in drug development has been discussed [13]. Initial pharmacokinetic–pharmacodynamic models focused on the characterization of rapidly reversible direct drug effects. These models typically consist of three components: (1) a pharmacokinetic model characterizing the time course of drug (and metabolite) concentrations in blood or plasma; (2) a pharmacodynamic model, characterizing the relationship between drug concentration and pharmacological response intensity; and (3) often, a link model which serves to account for the delay of the effect relative to the plasma concentration [14]. More
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
Fig. 1. Schematic representation of the interrelationship between drug delivery and input (dose), pharmacokinetics (concentration), pharmacodynamics and clinical effects. Modelling of the relationship between pharmacokinetics and pharmacodynamics provides the scientific basis for the optimal delivery profile of a drug.
recently emphasis has been given to the development of models for drugs with an indirect mechanism of action, where the delay between drug concentration and effect is largely determined by the processes in the pharmacodynamic phase, rather than by slow distribution to the site of action [15,16]. The models for drugs with direct and indirect mechanisms of action are complementary, and have been applied successfully to the effects of a large number of drugs. There is no doubt that there will be further developments in pharmacokinetic–pharmacodynamic modelling. An important area, particularly in relation to drug delivery, is the development of models to characterize functional adaptation (i.e. tolerance) during long term drug treatment and under the influence of disease. Such modelling may provide critical information with regard to the most optimal delivery profile. Zero-order delivery is in many instances still considered ideal. In the meantime however there is clear evidence that for certain drugs and diseases, pulsatile or ‘on demand’ delivery may be optimal [17].
1.2. Controlled (transdermal) drug delivery For the controlled delivery of drugs to the systemic circulation various routes of administration are available, including oral and intravenous routes as well as a number of alternate routes like rectal,
255
buccal, sublingual, nasal and transdermal administration [18,19]. Each of these routes has specific advantages and disadvantages with regard to the opportunities for controlled drug delivery. Of the various routes of administration, the oral route shows the poorest control of both the rate and the extent of absorption, whereas the intravenous route provides optimal opportunities for control of the plasma concentration vs. time profile. The latter however is associated with many practical limitations. Other routes of administration may be considered as realistic alternatives. The choice of a route is thereby dependent on the type of concentration vs. time profile that is pursued. In situations where a constant plasma concentration vs. time profile is desirable, the rectal (using osmotic drug delivery devices) and the transdermal route of administration may be very suitable. On the other hand, if more fluctuating concentration vs. time profiles are preferred, buccal / sublingual and nasal administrations may be suitable alternatives [19]. In the management of Parkinson’s disease dopaminergic drugs are administered chronically over periods of several years. This makes transdermal delivery in theory an attractive approach. A major limitation of using the transdermal route for systemic drug delivery however is the high intrinsic barrier of the skin, which resides in the stratum corneum. Large and hydrophilic compounds cross this barrier with difficulty. In addition, the transport of highly lipophilic compounds can be hindered by poor partitioning from the stratum corneum to lower hydrophilic epidermal layers [20,21]. Another limitation of transdermal drug delivery is that, at least with the classical (passive) delivery systems, it is very difficult to control the concentration vs. time profile in blood. In general it is not possible to achieve significant concentrations in the blood rapidly. Furthermore, drug delivery does not always stop upon removal of the system, as a drug depot may have formed in the stratum corneum [22]. For a wide variety of drugs it has been shown in in vitro studies that by iontophoresis a significant increase of the transdermal transport rate as well as an accurate control of the input rate can be achieved [23]. Application of iontophoresis for the transdermal delivery of dopaminergic drugs in Parkinson’s disease may therefore be an attractive approach.
256
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
2. Pharmacokinetic–pharmacodynamic studies on R-apomorphine
2.1. Pharmacokinetics, enantiomer interconversion and metabolic pathways When studying the pharmacokinetic–pharmacodynamic correlation of a drug it is essential that various potentially complicating pharmacokinetic factors are taken into consideration [24]. The pharmacokinetics and metabolism of R-apomorphine are indeed quite complex and this may have important pharmacological and clinical implications. R-apomorphine is an optically active compound which is administered as the pure R-enantiomer (Fig. 2). In theory interconversion into the S-enantiomer may occur. Sapomorphine has been shown to act as an antagonist at both the D 1 and D 2 receptor subtypes [25]. Another factor is that the enzyme catechol-O-methyl transferase (COMT) may be involved in the metabolism of R-apomorphine, resulting in the formation of the metabolites apocodeine and iso-apocodeine. Apocodeine has been shown to act as a dopamine agonist, albeit that its potency appears to be relatively low [26]. Other metabolic pathways include
glucuronide and sulfate conjugation [27]. In order to fully understand the pharmacokinetic–pharmacodynamic correlation of R-apomorphine, it is essential to have detailed quantitative information on the various metabolic pathways and in particular the plasma concentration profiles of S-apomorphine and the metabolites apocodeine and isoapocodeine. Within the context of developing an integrated pharmacokinetic–pharmacodynamic model of Rapomorphine, its disposition was studied in 10 patients with idiopathic Parkinson’s disease following intravenous infusion of 30 mg.kg 21 in 15 min. The plasma concentrations of both R- and S-apomorphine as well as the concentrations of apocodeine and iso-apocodeine were determined using newly developed (enantioselective) HPLC assays [28]. In most patients the plasma concentration vs. time profile of R-apomorphine was best described by a biexponential function. The values of the relevant pharmacokinetic parameters were: clearance 40615 ml min 21 kg 21 , volume of distribution at steadystate 1.660.5 l kg 21 and terminal half-life 41613 min. Interestingly the value of the clearance exceeds hepatic blood flow indicating that metabolism must occur to a significant extent in extrahepatic tissues.
Fig. 2. Chemical structures of R-apomorphine and its potential metabolites. From a pharmacodynamic point of view in particular the interconversion into S-apomorphine and the formation of the methylated metabolites apocodeine and isoapocodeine are important. (From [29].)
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
This raises the very important question whether indeed R-apomorphine is eliminated to a significant extent by COMT as this enzyme is present in large quantities in extrahepatic tissues. Furthermore the considerable interindividual variability in the pharmacokinetic parameters with values of the clearance ranging between 24–69 ml min 21 kg 21 is an important observation. In plasma no measurable concentrations of Sapomorphine nor of the metabolites apocodeine and isoapocodeine were observed. The metabolism of R-apomorphine was further characterized on the basis of the excretion of unchanged R-apomorphine, S-apomorphine, apocodeine and iso-apocodeine and their respective glucuronide and sulfate conjugates in urine. The total excretion of unconjugated S-apomorphine, apocodeine and isoapocodeine was less than 0.1% of the total administered dose. The total excretion of unchanged R-apomorphine, apomorphine-sulfate and apomorphine glucuronide amounted to 0.360.4%, 3.861.0% and 6.062.2% of the administered dose, respectively [29]. It appears therefore that the formation of S-apomorphine and the formation of apocodeine and iso-apocodeine do not contribute to the elimination of R-apomorphine in man. Therefore these compounds do not seem to contribute to the pharmacological response observed upon the intravenous administration of a single dose of R-apomorphine. Unfortunately however, only 10% of the administered dose can be accounted for via the elimination pathways that have been studied so far. An important question remains therefore via which pathways the rest is eliminated. One possible pathway is non-enzymatic oxidation, which may indeed contribute to a significant extent to the elimination. The oxidation products of R-apomorphine are not likely to be pharmacologically active [30]. Another pathway which has not been studied so far and which may be important from a pharmacodynamic point of view is oxidative N-demethylation resulting in the formation of norapomorphine. It cannot be excluded that this metabolite exhibits pharmacological activity. To what extent the formation of this metabolite contributes to the elimination of R-apomorphine remains to be determined. It is unlikely to be a major elimination pathway however, since N-demethylation is confined primarily to the
257
liver and most of the elimination of R-apomorphine occurs in extrahepatic tissues.
2.2. Stepwise intravenous infusion to determine the therapeutic window Characterization of the concentration–effect relationship is a central issue in integrated pharmacokinetic–pharmacodynamic investigations. Typically in this kind of study, a single intravenous dose of a drug is administered and the time course of the pharmacological response is determined in conjunction with the time course of the concentration in plasma. By linking the effects to the concentration, concentration–effect relationships are obtained in individual patients. In the case of R-apomorphine however the application of this approach is complicated by its short elimination half-life. As a result the plasma concentrations following an intravenous bolus administration change rapidly, thereby not allowing sufficient time for a meaningful quantification of the pharmacological response intensity. In order to overcome this problem, a stepwise intravenous infusion protocol was developed, which results in multiple relatively stable plasma concentrations within each individual patient. According to this protocol the pharmacokinetic–pharmacodynamic correlation of R-apomorphine was studied in 10 patients with idiopathic Parkinson’s disease. In these studies intravenous infusion of R-apomorphine was started at a rate of 10 mg kg 21 h 21 , and increased every 20 min by 10 mg kg 21 h 21 , until the maximum tolerated infusion rate or the infusion rate of 100 mg kg 21 h 21 was reached. Thereafter the infusion rate was decreased every 20 min by 10 mg kg 21 h 21 until baseline. At the end of each infusion interval a blood sample was obtained for determination of the concentration of R-apomorphine and the pharmacological effect was measured. The primary effect parameters were tapping score and tremor assessment. To determine the tapping score, the patient had to tap two buttons, with a distance of 30 cm over 30 s, with the right hand as well as the left hand at maximal speed. Tremor was assessed on a 4-point scale (0 5 absent, 1 5 mild, 2 5 moderate, 3 5 severe). When the tremor score decreased with at least 2 points this was considered a positive clinical response. Side
258
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
effect was monitored on the basis of dyskinesia. For that purpose the same 4-point scale was used separately for the head, trunk and the four extremities, resulting in a maximal dyskinesia score of 18 points. Increase of the dyskinesia score with at least 4 points was considered clinically relevant. Adverse effects (systolic blood pressure below 90 mmHg; bradycardia: heart rate below 50 min; nausea; dizziness) were monitored continuously. Fig. 3 shows the time course of the plasma concentration of R-apomorphine and of the effect on the tapping score in a representative patient. When the plasma concentration increases gradually beyond the minimally effective concentration, a sharp increase in the effect to its maximum value is observed. This maximum effect is maintained until, with a decrease in the infusion rate, the concentration drops to subtherapeutic values and the effect returns to baseline. Within each individual patient the concentrations at onset and offset of effect were generally quite similar. Upon examination of the concentration–effect relationships it appeared that the effect was quantal rather than continuous. Therefore the pharmacodynamics were parameterized in terms of minimal effective concentration (MEC) at the onset of the beneficial effect (tremor decrease or 25% tap score increase), the minimal dyskinesia concentration (MDC) as the onset concentration of the increase in
dyskinesia and finally, the minimal toxic concentration (MTC) as the onset concentration of the adverse effects (nausea, vomiting, sleepiness, dizziness, hypotension). Between patients wide differences were observed with respect to the minimal concentrations for each of the effects. Of the 10 patients, eight responded to R-apomorphine and two were non-responders. In the responders the values of the minimally effective concentration (MEC) varied between 1.4–10.7 ng ml 21 , whereas the concentration at which dyskinesia was observed (MDC) varied between 2.7–20.0 ng ml 21 . The minimal toxic concentration varied between 8.5–24.5 ng ml 21 . In Fig. 4 the individual values of the minimal effective concentration (MEC) and the minimal dyskinesia score (MDC) are shown. Between patients there is a wide variability in the therapeutic window ( 5 the difference between the MEC on one hand and the MDC on the other) of R-apomorphine, which is typically quite narrow [31]. The wide interindividual variability in both pharmacokinetics and pharmacodynamics, in combination with the narrow therapeutic window in many patients underlines the need of individualized and carefully controlled delivery of R-apomorphine in the management of Parkinson’s disease.
3. Transdermal iontophoresis of R-apomorphine
3.1. In vitro characterization of iontophoretic transport in human skin
Fig. 3. The plasma concentration vs. time profile and the effect on the tapping score (number of taps / 30 s of both the left and the right arm) upon a stepwise intravenous infusion of R-apomorphine in an individual patient with Parkinson’s disease. The maximum infusion rate of R-apomorphine in this patient was 70 mg kg 21 h 21 . (From [30].)
Fundamental studies on the transdermal iontophoretic transport of drugs are typically conducted in in vitro test systems with excised human skin. In the studies on R-apomorphine transport a newly designed continuous flow-through transport cell was used, where the dimensions were optimized to obtain realistic estimations of the time course of the intrinsic flux upon the application of different electrical current profiles [32]. In initial studies R-apomorphine transport was studied in human stratum corneum, as the barrier to transport presumably resides in this layer. In all the experiments R-apomorphine was applied in the anodal compartment. The effect on the flux of the following parameters was studied: current density, pH, concentration, ionic strength, osmolari-
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
259
Fig. 4. Individual values of the minimal effective concentration (MEC 5 25% increase in tapping score) and the minimal dyskinesia concentration (MDC) in patients with idiopathic Parkinson’s disease (patients 1, 2, 5, 6 and 7). When no dyskinesia was noted the maximum concentration was plotted (patients 3 and 4). Note the wide inter-individual variability in the therapeutic window ( 5 MDC 2 MEC). (From [30].)
ty, buffer strength, temperature and skin type. The results of these investigations showed that the transdermal transport rate of R-apomorphine can be directly controlled by manipulation of the current density (Fig. 5). When no electrical current is applied, passive delivery is minimal and no depot effect is observed. Application of electrical current (500 mA cm 22 over 5 h at a temperature of 208C) resulted in enhancement of the delivery of Rapomorphine to 9066 nmol cm 22 h 21 . A steadystate flux was achieved in | 3 h, corresponding to a lag time of 40 min. The reversibility of the enhanced membrane permeability to the ‘normal’ state was tested by measuring the passive flux after switching off the current. In a period of | 3 h an exponential decay to a value of | 1 / 20th of the steady-state flux was observed. In the in vitro experiments various current densities in the range of 0–500 mA cm 22 were tested. Thereby a linear relationship (r 2 5 0.98) between the current density and the steady-state
Fig. 5. Time course of the flux of R-apomorphine across human stratum corneum in vitro at a donor concentration of 5 mM, pH 5 5 and temperature 5 208C (d, passive flux through nonpretreated skin; j, passive flux after 5 h of current pretreatment, 500 mA cm 22 without apomorphine; ♦, 5 h of iontophoretic delivery at 500 mA cm 22 followed by 3 h of passive delivery). (From [32].)
260
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
ratio between the transdermal transport rates in vivo and in vitro it can be calculated that therapeutically effective plasma concentrations can be reached with a transdermal iontophoretic patch of 20 cm 2 , which is realistic from a clinical point of view [33]. The findings of these in vitro studies show that in theory the delivery of R-apomorphine by transdermal iontophoresis is feasible and furthermore that modulation of the current density may be used to control (and thereby to individualize) the transdermal transport rate. Fig. 6. Relationship between current density and steady-state flux of R-apomorphine in human stratum corneum in vitro. (Experimental conditions: donor concentration 5 15 mM; pH 5 4; temperature 5 208C; current density range 5 100–500 mA cm 22 ). (From [32].)
transdermal flux was observed (Fig. 6), indicating that variation of the current density may be used to control the delivery of R-apomorphine across the skin. In general only minimal effects of changes in osmolarity, buffer strength, ionic strength and pH on the transdermal transport rate were observed. Temperature on the other hand was found to be a major determinant of both the maximum steady-state transport rate and the lag time. Upon increase of the temperature from 208C to 378C the steady-state flux was increased by a factor 2.3 and the lag time was decreased to 63 min. In the studies with stratum corneum, skin samples of different donors were tested. Only minimal interindividual variability in the maximal steady-state transport rate and the lag time were observed. The transdermal transport of Rapomorphine was subsequently studied in dermatomed full skin to determine the difference with stratum corneum. In dermatomed full skin the maximum steady-state flux was | 30% lower than in stratum corneum. An important question is whether the transdermal transport rates that have been observed in the in vitro experiments are sufficiently high to achieve therapeutically effective plasma concentrations. Thereby the pharmacokinetic properties and the therapeutic window of R-apomorphine that have been determined in previous in vivo investigations, must be taken into consideration. Assuming a one-to-one
3.2. Feasibility of transdermal iontophoresis in vivo In order to determine the feasibility of transdermal iontophoresis of R-apomorphine in vivo, a study was conducted in 10 patients with idiopathic Parkinson’s disease. According to a randomized cross-over design the patients received R-apomorphine by transdermal iontophoresis at a current density of either 250 or 375 mA cm 22 over 1 h on one occasion. At another occasion 30 mg kg 21 were administered intravenously over 15 min. In none of the patients was significant passive transdermal transport of R-apomorphine observed. Upon current application, increasing plasma concentrations were observed in all patients. The observed maximum concentrations were directly related to the applied current density: 1.360.6 ng ml 21 at 250 mA cm 22 and 2.560.7 ng ml 21 at 375 mA cm 22 . When the current was switched off all concentrations returned to baseline values in | 90 min (Fig. 7). These results show that also in the in vivo situation the transdermal transport rate of R-apomorphine is determined by the applied current density. By mathematical deconvolution it was demonstrated that steady-state transdermal flux values are reached within 1 h of current application. The values of the steady-state flux were 69630 nmol cm 22 h 21 at 250 mA cm 22 and 114634 nmol cm 22 h 21 at 375 mA cm 22 , indicating that variation in the applied current density may be used to control the delivery. The delivery of R-apomorphine at the aforementioned rates results in steady-state plasma concentrations of | 3–4.5 ng ml 21 which is at the lower end of the therapeutic concentration range. In this study also an assessment was made of the local skin irritation and
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
261
Table 1 Comparison of the transdermal transport characteristics of Rapomorphine in human skin in vitro and in vivo. The transport rates are normalized to a current density of 250 mA cm 22 . The donor concentration was 15 mM and the pH was 5 in both situations Experimental condition
Human skin, in vivo Stratum corneum, in vitro Dermatomed skin, in vitro Fig. 7. Average plasma concentration time profiles of R-apomorphine upon transdermal iontophoretic delivery in patients with idiopathic Parkinson’s disease at current densities of 250 mA cm 22 . s, n 5 4 and 375 mA cm 22 ; d, n 5 5, for 1 h starting at t 5 60 min. Steady-state plasma concentrations were predicted by fitting the data to a mathematical model describing the concentration vs. time profile upon zero-order delivery in a twocompartment pharmacokinetic model (from [33].)
toxicity. Local skin erythema was assessed visually and quantified by laser Doppler flowmetry (LDF). Upon removal of the transdermal iontophoretic patch mild transient erythema was observed which was reflected in elevated LDF values. No local adverse effects such as prolonged inflammation and green colouring (which is typically seen upon subcutaneous R-apomorphine administration) were observed. These findings indicate that the transdermal delivery of R-apomorphine is feasible at acceptable levels of skin irritation [34]. An interesting question is how the transdermal transport observed in the in vivo study compares to the transport observed in vitro. It appears that the steady-state transport rate in vivo at a current density of 250 mA cm 22 is remarkably similar to the value observed both in stratum corneum and in dermatomed full skin in vitro (Table 1). With respect to the time to reach steady-state a different situation is observed. The value of the time to reach steady-state of 10–20 min in vivo is in the same range as in stratum corneum in vitro. In the dermatomed full skin preparation however a much larger value is observed. Therefore in vitro studies in stratum corneum appears to have a better predictive value with regard to the in vivo transdermal transport
Steady-state transdermal flux (nmol cm 22 h 21 )
Time to reach steady-state (min)
69630
10–20
101613
15–30
65611
300
characteristics of R-apomorphine than studies in dermatomed full skin. The absence of dermal blood flow in the dermatomed full skin preparation may be an important factor in this respect.
4. Delivery of R-apomorphine by transdermal iontophoresis: conclusions and perspectives In this paper a new approach to optimization of delivery by transdermal iontophoresis of R-apomorphine in Parkinson’s disease is described which is based on a combination of in vivo and in vitro studies. First in the in vivo studies the pharmacokinetic–pharmacodynamic correlation was determined, which provides crucial information on the optimal delivery profile of R-apomorphine in patients with idiopathic Parkinson’s disease. Next the transdermal iontophoretic transport characteristics of Rapomorphine were determined in human skin in vitro. Finally, transdermal iontophoresis was applied in an in vivo study in patients with idiopathic Parkinson’s disease. The results of the various investigations show that precise control of the transdermal delivery of Rapomorphine is feasible, but that the concentrations that can be reached in this way are at the lower end of the therapeutic concentration range of 1.4 to 10.7 ng ml 21 . Many patients need concentrations that are significantly higher than is currently feasible. Furthermore, in clinical practice it may be necessary to make rapid dose adjustments. It is therefore important to explore the possibilities to further enhance
262
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
the transdermal transport rate. In this respect the application of neutral transport enhancers with small headgroups in combination with iontophoresis may be an interesting approach. An important question is whether by using this strategy it will still be possible to accurately control the delivery on the basis of modulation of the current density. Furthermore skin toxicity may become a critical issue. The results from the studies on the pharmacokinetic–pharmacodynamic correlation show that the effects can be directly related to the plasma concentration of R-apomorphine. A wide interindividual variation in the therapeutic concentration range was observed. An important question is which factors are the determinants of this interindividual variability. In this respect a particularly intriguing question is to what extent disease progression contributes to the variability in pharmacodynamics. In Parkinson’s disease, disease progression has been associated with different subtypes of patients which differ in their response to dopaminergic drugs. Both responders and non-responders can be distinguished. The responders can be further classified as ‘stable’, ‘wearing-off’ and ‘fluctuating’, depending on the severity of the disease. An important question is how the therapeutic concentration range of R-apomorphine changes with the progression of the disease. This requires studies in a large population of patients with Parkinson’s disease. Furthermore longitudinal studies in individual patients will be of considerable interest. Another important question is whether tolerance develops to the therapeutic and / or side effects of R-apomorphine. A fascinating question in this respect is to what extent the rate and extent of functional tolerance development is influenced by the delivery profile (i.e. zero-order versus pulsatile delivery). Delivery of R-apomorphine by transdermal iontophoresis may prove to be of great value in future clinical pharmacological investigations, addressing the aforementioned issues. In such studies transdermal iontophoresis may be used as a non-invasive alternative for stepwise intravenous infusion. Furthermore it may be of great value to provide zeroorder, pulsatile or ‘on demand’ delivery in studies which aim at characterization of functional tolerance development with chronic treatment. However the application of transdermal iontophoresis for these
indications requires that a system be available allowing higher transdermal transport rates than currently feasible. Delivery of dopaminergic drugs by transdermal iontophoresis is not limited to R-apomorphine, but can in principle also be applied to clinical pharmacological investigations with other drugs. Ultimately these developments may result in transdermal iontophoretic delivery systems that are useful for clinical application in Parkinson’s disease.
References [1] A.J. Lees, Apomorphine and Parkinson’s disease, in: J.P.W.F. Lakke, E.M. Delhaas, A.W.F. Rutgers (Eds.), Parenteral Drug Therapy in Spasticity and Parkinson’s Disease, The Parthenon Publishing Group, Carnforth, 1992, pp. 185–192. [2] J.L. Neumeyer, S. Law, R.J. Baldessarini, Historical highlights of the chemistry, pharmacology and early clinical uses of apomorphine, in: G.L. Gessa, G.U. Corsini (Eds.), Apomorphine and other Dopamimetics, Raven Press, New York, 1981, pp. 1–17. [3] R.J. Baldessarmni, G.W. Arana, N.S. Kula, A. Campbell, M. Harding, Preclinical studies of the pharmacology of aporphines, in: G.L. Gessa, G.U. Corsini (Eds.), Apomorphine and other Dopamimetics, Raven Press, New York, 1981, pp. 219–228. [4] G.C. Cotzias, P.S. Papavasiliou, E.S. Tolosa, J.S. Mendez, M. Bell-Midura, Treatment of Parkinson’s disease with apomorphines, N. Engl. J. Med. 294 (1976) 567–572. [5] S.T. Gancher, J.G. Nutt, W.R. Woodward, Absorption of apomorphine by various routes in parkinsonism, Mov. Disord. 6 (1991) 212–216. [6] C.M. Stibe, P.A. Kempster, A.J. Lees, G.M. Stern, Subcutaneous apomorphine in parkinsonian on-off oscillations, Lancet 1 (1988) 403–406. [7] T. Van Laar, E.N.H. Jansen, A.W.G. Essink, C. Neef, Intranasal apomorphine in parkinsonian on-off fluctuations, Arch. Neurol. 49 (1992) 482–484. [8] J.L. Montastruc, O. Rascol, J.M. Senard, Sublingual apomorphine in Parkinson’s disease: a clinical and pharmacokinetic study, Clin. Neuropharmacol. 14 (1991) 432–437. [9] T. Van Laar, E.N.H. Jansen, A.W.G. Essing, W.J. Rutten, C. Neef, Rectal apomorphine: a new treatment modality in Parkinson’s disease, J. Neurol. Neurosurg. Psychiatry 55 (1992) 737–738. [10] D.D. Breimer, Pharmacokinetic and pharmacodynamic basis for peptide drug delivery system design, J. Control. Release 21 (1992) 5–10. [11] D.D. Breimer, An integrated pharmacokinetic and pharmacodynamic approach to controlled drug delivery, J. Drug. Target. 3 (1996) 411–415. [12] M. Danhof, J.W. Mandema, Modeling of relationships be-
M. Danhof et al. / Advanced Drug Delivery Reviews 33 (1998) 253 – 263
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
tween pharmacokinetics and pharmacodynamics, in: P.G. Welling, F.L.S. Tse (Eds.), Pharmacokinetics — Regulatory, Industrial, Academic Perspectives, Marcel Dekker, New York, 1995, pp. 139–174. D.D. Breimer, M. Danhof, Relevance of the application of pharmacokinetic–pharmacodynamic modelling concepts in drug development. The ‘Wooden Shoe’ paradigm, Clin. Pharmacokinet. 32 (1997) 259–267. N.H.G. Holford, L.B. Sheiner, Understanding the dose-effect relationship: clinical applications of pharmacokinetic–pharmacodynamic models, Clin. Pharmacokinet. 6 (1981) 429– 453. N.L. Dayneka, V. Garg, W.J. Jusko, Comparison of four basic models of indirect pharmacodynamic responses, J. Pharmacokinet. Biopharm. 21 (1993) 457–478. W.J. Jusko, H.C. Ko, Physiologic indirect response models characterize diverse types of pharmacodynamic effects, Clin. Pharmacol. Ther. 56 (1994) 406–419. D.D. Breimer, The need for rate and time programming in future drug delivery. Pharmacokinetic, pharmacodynamic and clinical considerations, in: R. Gurny, H.E. Junginger, N.A. Peppas (Eds.), Pulsatile Drug Delivery — Current Applications and Future Trends, Wissenschaftliche Verlagsgesellschaft mbH, Stuttgart, 1993, pp. 25–37. L.Z. Benet, Input factors as determinants of drug activity: route, dose, dosage regimen and the drug delivery system, in: F.G. McMahon (Ed.), Principles and Techniques of Human Research and Therapeutics, Futura, New York, 1974, pp. 9–23. M. Danhof, Parenteral neuropharmacological kinetics: pharmacokinetic aspects of controlled drug delivery to the brain, in: J.P.W.F. Lakke, E.M. Delhaas, A.W.F. Rutgers (Eds.), Parenteral Drug Therapy in Spasticity and Parkinson’s Disease, The Parthenon Publishing Group, Carnforth, 1992, pp. 163–176. J.E. Shaw, S.K. Chandrasekaran, Transdermal therapeutic systems, in: L.F. Prescott, W.S. Nimmo (Eds.), Drug Absorption, ADIS Press, Sydney, 1979, pp. 186–193. W.S. Nimmo, Recent clinical experience with transdermal systems, in: L.F. Prescott, W.S. Nimmo (Eds.), Novel Drug Delivery, Churchill Livingstone, Edinburgh, 1989, pp. 209– 221. B.H. Sage, Iontophoresis, in: E.W. Smith, H.I. Maibach (Eds.), Percutaneous Penetration Enhancers, CRC Press, Boca Raton, 1995, pp. 351–368. N.H. Yoshida, M.S. Roberts, Structure-transport relationships in transdermal iontophoresis, Adv. Drug. Del. Rev. 9 (1992) 239–264.
263
[24] M. Danhof, J.W. Mandema, A.M. Stijnen, Pharmacokinetic complexities of in vivo pharmacodynamic investigations, in: C.J. Van Boxtel, N.H.G. Holford, M. Danhof (Eds.), The In Vivo Study of Drug Action — Concepts and Applications of Kinetic-Dynamic Modelling, Elsevier, Amsterdam, 1992, pp. 31–60. [25] M.E. Goldman, J.W. Kebabian, Aporphine enantiomers: interactions with D 1 and D 2 dopamine receptors, Mol. Pharmacol. 25 (1984) 18–23. [26] J.G. Cannon, R.V. Smith, A. Modiri, S.P. Sood, R.J. Borgman, M.A. Aleem, J.P. Long, Centrally acting emetics: preparation and pharmacology of 10 hydroxy-11-methoxyapomorphine (isoapocodeine). In vitro enzymatic methylation of apomorphine, J. Med. Chem. 15 (1972) 273–276. [27] P.N. Kaul, E. Brochmann-Hansen, E.L. Way, Biological disposition of apomorphine. Urinary excretion and organ distribution of apomorphine, J. Pharm. Sci. 50 (1961) 244– 247. ¨ [28] R. Van der Geest, P.P. Kruger, J.M. Gubbens-Stibbe, T. Van ´ M. Danhof, Assay of R-apomorphine, Laar, H.E. Bodde, S-apomorphine, apocodeine, isoapocodeine and their glucuronide and sulfate conjugates in plasma and urine of patients with Parkinson’s disease, J. Chromatogr. Biomed. Appl. 702 (1997) 131–141. [29] R. Van der Geest, T. Van Laar, J.M. Gubbens-Stibbe, H.E. ´ R.A.C. Roos, M. Danhof, Pharmacokinetics, enantioBodde, mer interconversion and metabolism of R-apomorphine in patients with idiopathic Parkinson’s disease, Clin. Neuropharmacol. 21 (1998) 159–166. [30] K. Rehse, G. Piesker, R. Horowski, Neuropsychotrope ¨ und Toxicitat ¨ von Oxidationsprodukten des Aktivitat Apomorphins, Arch. Pharmacol. 311 (1977) 360–363. ´ P.H. [31] T. Van Laar, R. Van der Geest, M. Danhof, H.E. Bodde, Gossens, R.A.C. Roos, Stepwise intravenous infusion of R-apomorphine to determine the therapeutic window in patients with Parkinson’s disease, Clin. Neuropharmacol. 21 (1998) 152–158. ´ Validation and [32] R. Van der Geest, M. Danhof, H.E. Bodde, testing of a new iontophoretic continuous flow through transport cell, J. Control. Release 51 (1998) 85–91. ´ Iontophoretic [33] R. Van der Geest, M. Danhof, H.E. Bodde, delivery of apomorphine I. In vitro optimization and validation, Pharm. Res. 14 (1997) 1797–1802. [34] R. Van der Geest, T. Van Laar, J.M. Gubbens-Stibbe, H.E. ´ M. Danhof, Iontophoretic delivery of R-apomorphine Bodde, II: an in vivo study in patients with Parkinson’s disease, Pharm. Res. 14 (1997) 1803–1809.