Doi:10.1016/j.ejpb.2006.11.00

European Journal of Pharmaceutics and Biopharmaceutics 66 (2007) 296–301 In vitro and in vivo evaluation of the transdermal iontophoretic Sonal R. Patel a,1, Hui Zhong a, Ashutosh Sharma a, Yogeshvar N. Kalia b,c,* b School of Pharmaceutical Sciences, University of Geneva and University of Lausanne, Geneva, Switzerland c Centre Interuniversitaire de Recherche et d’Enseignement, ‘‘Pharmapeptides’’, Site d’Archamps, Archamps, France Received 18 July 2006; accepted in revised form 6 November 2006 The objective was to evaluate the transdermal delivery of the 5-HT1B/1D agonist, sumatriptan from an iontophoretic patch system, in vivo. Initial in vitro experiments were conducted to optimize formulation parameters prior to iontophoretic delivery in Yorkshireswine. It was found in vitro that increasing drug load in the patch from 9.7 to 39 mg had no statistically significant effect on cumulativedelivery (cf. 305.6 ± 172.4 vs. 389.4 ± 80.4 lg cmÀ2, respectively). However, for a given drug load (39 mg) increasing formulation pHfrom pH 4.7 to 6.8 significantly increased the cumulative amount of sumatriptan delivered across the skin (389.4 ± 80.4 vs.
652.4 ± 94.2 lg cmÀ2). A biphasic current profile comprising intensities of 1.8 mA from t = 0 to t = 180 min and 0.8 mA fromt = 181 min to t = 360 min was used for the in vivo experiments. Drug levels in the blood were 13.7 ± 4.5 and 53.6 ± 10.2 ng mlÀ1 atthe 30 and 60 min time-points, rising to 90–100 ng mlÀ1 during the 90–180 min time-period. The in vivo results show that the pharma-cokinetics following transdermal iontophoretic delivery are comparable to those after oral, nasal or rectal administration, but do notmatch those upon subcutaneous injection.
Ó 2006 Elsevier B.V. All rights reserved.
Keywords: Transdermal; Iontophoresis; Migraine; Sumatriptan; 5HT1B/1D agonist; Non-invasive; In vivo cy, slow onset and incomplete prevention of recurrence asmajor shortcomings of current therapies There is a con- Sumatriptan is a selective serotonin 5-HT agonist at the sensus that subcutaneous injection of sumatriptan provides 5-HT1B and 5-HT1D receptors () used in the treat- the most rapid response at the 30 min time-point (63%) and ment of acute migraine episodes It was the first of the complete relief at the 2 h time-point (67%) (conventional so-called ‘‘triptan’’ drugs which have had a significant endpoint for evaluating treatment efficacy) Yet, from impact on the treatment of acute attacks and it is available the patient’s perspective, this is the least desirable modality in several dosage forms including products for oral, nasal of sumatriptan administration. In addition to the reluc- and rectal delivery However, patients cite limited effica- tance for self-injection, there are also reports of skin sitereactions in more than 50% of patients Transdermal delivery offers a convenient alternative, par- * Corresponding author. Laboratory of Medicinal Chemistry, School of ticularly where nausea prevents administration of an oral Pharmaceutical Sciences, University of Geneva, 30 Quai Ernest Ansermet, dosage form. In addition, sumatriptan has a relatively poor 1211 Geneva 4, Switzerland. Tel.: +33 450 31 50 24; fax: +33 450 95 28 32.
oral bioavailability (only 14%) and a relatively short half-life Present address: Forest Laboratories Inc., Harborside Financial 1/2 $ 2 h) However, based on the molecular properties Center Plaza V, Jersey City, NJ 07311, USA.
of the weakly lipophilic sumatriptan base (log Ko/w = 0.93 0939-6411/$ - see front matter Ó 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.ejpb.2006.11.001 S.R. Patel et al. / European Journal of Pharmaceutics and Biopharmaceutics 66 (2007) 296–301 Sumatriptan succinate was custom synthesized (Natco Pharma Limited, Hyderabad, India); ketamine, xylazine and propofol were obtained from Henry Schein Inc. (Mel- ville, NY, USA). Ammonium acetate (ACS reagent) was obtained from Sigma–Aldrich (St. Louis, MO, USA); ace-tonitrile, acetic acid and methanol (all HPLC grade) were Fig. 1. Structure of sumatriptan (MW = 295.5 Da; pKa = 9.63; log Ko/ obtained from (G.J. Chemical, Newark, NJ, USA).
; log DpH 7.4 = À1.3) it is unlikely that passive diffusion Porcine skin was obtained from Thomas D. Morris, Inc.
across the skin could deliver therapeutic amounts of drug (Reistertown, MD, USA). The excised skin was derma- from reasonably sized patches. Indeed, Femenı´a-Font tomed ($500 lm) on the same day and stored at À20 °C et al. showed that the cumulative permeation of sumatriptan for a maximum period of up to 1 week. A proprietary across porcine skin after 6 h from an aqueous solution was two-compartment iontophoretic patch system was used only $1–2 lg cmÀ2 and although skin pre-treatment with during the studies (The electrode compartment R-(+)-limonene produced a >20-fold increase, cumulative comprised an Ag-mesh anode, a small amount of sodium delivery at the 6 h time-point was still only $40 lg cmÀ2 chloride (0.06%) and an ion exchange resin (AMBER- . Subsequent studies using bioadhesive films showed LITEä IRP-69, Rohm & Haas, Perth Amboy, NJ, USA) similar cumulative delivery of approximately 20 lg cmÀ2 that trapped Ag+ ions preventing them from competing at the 6 h time-point Based on the existing dosage forms with drug ions to carry current. The electrode compartment (e.g., subcutaneous injection of 6 mg) and the known phar- was separated from the sumatriptan succinate contained in macokinetics in man, it is probably necessary to deliver at the drug reservoir, made of polyvinylpyrrolidone (PVP, least $3–5 mg of drug across the skin . Thus, passive 15%; K-90F, BASF, Florham Park, NJ, USA), by a size- delivery would neither deliver sumatriptan sufficiently rap- selective membrane (MW cut-off 100 Da, SpectroPor; idly nor in sufficient amounts to treat acute migraine Rancho Dominguez, CA, USA). The active surface area of the anodal patch in contact with the skin was 4 cm2. A In contrast, transdermal iontophoresis is particularly vertical diffusion set-up was employed wherein the patch suited to delivering polar and charged molecules across was placed on and directly in contact with the skin, which the skin In addition to the benefits of passive transder- was placed on a polymeric support. A flow through system, mal administration, iontophoresis improves drug input built in-house, ensured that the drug did not accumulate in kinetics and enables rapid ‘‘bolus’’ drug inputs in response the receiver phase, which was replenished at a rate of to patient need An investigation into the anodal ion- 0.1 ml minÀ1. A constant current of 0.25 mA cmÀ2 was used tophoretic transport of sumatriptan from buffered aqueous in all of the experiments; this is within the limits generally solutions (75 mM NaCl, 20 mM HEPES) across porcine accepted for use in humans . An AgCl electrode was skin in vitro showed that cumulative permeation at the employed as the cathode. Unless indicated otherwise suma- 6 h time-point was $270 and $700 lg cmÀ2 at 0.25 and triptan succinate was dissolved in water at the appropriate 0.5 mA cmÀ2, respectively Furthermore, decreasing concentration required for the desired patch loading and competition between charge carriers by lowering the NaCl $400 ll of the drug solution was introduced into the anodal concentration in the anodal compartment formulation drug reservoir A passive ‘‘no-current’’ control to 25 mM produced a further increase in transdermal confirmed that there was negligible sumatriptan transport in the absence of an iontophoretic current.
These experiments demonstrate that significant amounts of sumatriptan can be delivered across the skin fromsolution formulations by transdermal iontophoresis. The To controller
Ag mesh anode in an
aim of the present study was to investigate sumatriptan electrode compartment
electrotransport from an iontophoretic patch system and containing ion exchange
resin
to determine whether therapeutically relevant delivery ratescould be achieved under these conditions . After an Size-selective membrane
initial investigation of formulation parameters and theireffect on sumatriptan transport across porcine skin in vitro, PVP drug reservoir
an in vivo feasibility study was conducted using an Fig. 2. Schematic representation of the anodal compartment from the iontophoretic patch system in Yorkshire pigs.
S.R. Patel et al. / European Journal of Pharmaceutics and Biopharmaceutics 66 (2007) 296–301 The samples were collected into chilled 3 ml glass vacutain- Composition of the formulations used in the in vitro and in vivo er tubes containing ethylenediaminetetraacetic acid tripo- tassium salt (K3EDTA) (BD, Franklin Lakes, NJ, USA).
The tubes were immediately placed on ice and centrifuged at 4 °C (1600g for 15 min). The contents were then split into two samples and stored in Nalgene cryopreserve vials (VWR, Westchester PA, USA). The plasma samples were Samples obtained from the in vitro experiments were assayed using reverse phase HPLC. The HPLC system comprised a 600 E Controller pump, an AutosamplerInjector 717-plus, and a 486 tunable UV Detector (Waters, The in vivo experimental protocol was approved by the Milford, MA, USA) and was equipped with a Zorbax RX local Ethics Committee. Three 18–19 kg (7–9 weeks) prepu- c18 column with guard and prefilter (4.6 mm internal diam- bescent female pigs were used in the study. The weight of eter, 25 cm in length and with a 5 lm particle size) (Agilent the animals was measured and recorded before the start Technologies, Palo Alto, CA, USA). The mobile phase of each experiment. Animal hair at the site of patch appli- comprising 15% acetonitrile and 85% 0.5 M ammonium cation was clipped the night before the experiment. Before acetate buffered pH 4.9 solution with 1% trifluoroacetic applying the patches, the skin was gently wiped with warm acid (TFA) was delivered at a flow rate of 1 ml minÀ1.
water followed by an alcohol swab and patted dry. The ani- The injection volume was 50 ll. Sumatriptan was detected mals were placed on a surgical table under general anaes- at 282 nm; the limit of detection was 1 lg mlÀ1.
thesia and jugular, ear vein and arterial catheters wereplaced either percutaneously or surgically. Anaesthesia was induced by intramuscular administration of ketamine (i) Extraction. The drug was extracted by protein precip- (11 mg kgÀ1) and xylazine (2 mg kgÀ1); it was maintained itation. The plasma samples were first allowed to thaw at by continuous infusion of propofol (12–20 mg kgÀ1 hÀ1).
room temperature. After vortexing, 100 ll of sample was Arterial blood pressure, end tidal CO2 volume, rectal tem- transferred into 2 ml Eppendorf tubes. Then, 10 ll of perature and ECG measurements were recorded during all MeOH–H2O (1:1 mixture) was added to the plasma sam- procedures. Respiratory rate and quality was monitored ples containing sumatriptan. After addition of 300 ll of visually. Body temperature was maintained by (i) placing acetonitrile and vortexing for a few seconds, the mixture a circulating water heating pad under the animal and (ii) was centrifuged at 1200g for 10 min. Then, 300 ll of the a thermal blanket over the pig to retain body heat.
resulting supernatant was transferred to 16 · 100 mm clean As with the in vitro studies, the two-compartment culture tubes and evaporated to dryness under nitrogen at iontophoretic patch system (with an active area of 4 cm2 35 °C (this took approximately 20 min). The samples were and where the PVP drug reservoir contained 37 mg of suma- then reconstituted with 100 ll of mobile phase and vor- triptan at pH 7.0), coupled to a programmable power texed before being transferred to injection vials and source, was used to apply the current ). Two anodal patches were applied to each animal, that is, the total active (ii) Assay. This was adapted from a published method surface area in contact with the skin was 8 cm2. The cathode Briefly, the LC system comprised a LC-10 AP pump consisted of another two patches again with a total surface and SCL-10M controller (Shimadzu Corporation, MD, area of 8 cm2. All three animals received the iontophoretic USA); autoinjector (Waters 717plus autosampler, Waters treatment involving application of a biphasic current Corporation, MA, USA) and was equipped with an Inertsil protocol. In step 1, from t = 0 to t = 180 min, the current ODS2 column (4.6 mm internal diameter, and 15 cm in intensity was 1.8 mA (0.45 mA cmÀ2); in step 2, from length with 5 lm particle size) (Keystone Scientific, Inc.
t = 181 to t = 360 min, a lower current intensity of PA, USA). Perkin-Elmer API 365 and API 3000 detectors 0.8 mA (0.2 mA cmÀ2) was applied. From t = 361 min to were used to detect sumatriptan. The mobile phase (20% t = 480 min, no current was applied although the patches methanol and 80% 10 mM ammonium acetate buffered were left in contact with the skin to investigate elimination pH 4.0 solution) was delivered at a flow rate of 1 ml minÀ1.
of the drug from the bloodstream. At the end of the studies, The injection volume was 10 ll. With respect to the MS conditions, the spectrometer employed a heated ion Blood samples (2 ml) were drawn at 15 min intervals nebulizer at 475 °C. The product ion had a molecular from t = À15 min to t = 240 min and at 30 min intervals weight of 251.1 Da. The limit of quantification was from t = 240 min to t = 420 min and again at t = 480 min.
S.R. Patel et al. / European Journal of Pharmaceutics and Biopharmaceutics 66 (2007) 296–301 shows that a 4-fold increase in patch load (9.7–39 mg) produced no statistically significant difference (t-test, a = 0.05) in the cumulative amounts of sumatriptan delivered after current application for 6 h (305.6 ± 172.4 In the next in vitro study, the pH of the drug formula- tion was increased from pH 4.7 to pH 6.8. The former pH is close to the isoelectric point of the skin and hence there is only a limited contribution of electroosmosis to iontophoretic transport a shows that increasing the formulation pH by two units produced a statistically significant (t-test, a = 0.05) increase of approximately 60% in cumulative sumatriptan delivery (389.4 ± 80.4 vs.
652.4 ± 94.2 lg cmÀ2) . It should be noted that the pH of the drug reservoir remained fairly constant during current application (b presents the ionto- phoretic flux observed under these conditions. The flux at 6 h was 109.8 ± 14.4 and 153 ± 25.2 lg cmÀ2 hÀ1 at pH 4.7 and pH 6.8, respectively. At the higher pH, the steady state flux corresponds to transport number of 0.056, mean- ing that $5.6% of the charge transferred during iontopho- resis is carried by the sumatriptan cation. Since there is minimal competition from cationic species, it is evident that the predominant charge carrier in the system is the chloride ion from the receptor compartment towards the The results from the in vitro delivery experiments were Fig. 4. Effect of increasing formulation pH from 4.7 to 6.8 on (a) the used to decide the conditions for the in vivo study. cumulative amount of sumatriptan delivered across porcine skin in vitro shows sumatriptan plasma concentrations during ionto- and (b) the corresponding drug flux, with a 6 h iontophoretic current phoretic current application and following subcutaneous application (0.25 mA cmÀ2) from a patch system with a PVP gel drugreservoir containing 39 mg of drug. Filled and hollow circles represent injection in Yorkshire swine. A biphasic current profile formulation pH of 4.7 and 6.8, respectively (mean ± SD; n = 4).
was employed wherein a higher current of 1.8 mA(0.45 mA cmÀ2) was applied for 3 h followed by 3 h at0.8 mA (0.2 mA cmÀ2). Blood levels of sumatriptan rose 13.7 ± 4.5 and 53.6 ± 10.2 ng mlÀ1 at the 15, 30 and gradually upon current application, achieving 3.4 ± 3.1, 60 min time-points and achieved fairly constant levels, ofbetween 90 and 100 ng mlÀ1, during the 90–180 min time-period. The current intensity was then decreased to0.8 mA (0.2 mA cmÀ2), during the 180–360 min period, and there was a concomitant decrease in drug levels in the blood. Current application was terminated at t = 360 min, at which point, sumatriptan levels fell progressive- ly as the drug was eliminated from the bloodstream, illustrating the control afforded by iontophoresis over drug Visual inspection of the skin at the patch application sites after sumatriptan iontophoresis (and comparison of photographs of the sites before and after iontophoresis) did not reveal any significant erythema.
Fig. 3. Effect of a 4-fold increase in drug load on the cumulative amount The data showed that use of the two-compartment of sumatriptan delivered across porcine skin in vitro with a 6 h iontopho- system resulted in sumatriptan transport rates that were retic current application (0.25 mA cmÀ2) from a patch system with a PVPgel drug reservoir. Filled and hollow circles represent patch loadings of 9.7 independent of patch load ); this could be significant and 39 mg, respectively (mean ± SD; n = 4).
for costly therapeutics such as peptides. It has been shown S.R. Patel et al. / European Journal of Pharmaceutics and Biopharmaceutics 66 (2007) 296–301 were in reasonable agreement with the corresponding val- ues in humans Using the above clearance and assum- SS to be $100 ng mlÀ1 (average value between t = 90 $1.0 mg hÀ1. Eq. enables calculation of the sumatriptan delivery efficiency in vivo, as measured by its transport Thus, upon insertion of the appropriate values, calculation Fig. 5. Plasma concentration profiles of sumatriptan as a function of time shows that tin vivo was $0.05, similar to that seen in vitro. As during subcutaneous injection (6 mg; hollow circles) and anodal ionto- noted above, the sumatriptan iontophoretic flux in vitro (at phoresis (filled circles) in Yorkshire swine using an iontophoretic patch pH 6.7) was $0.15 mg cmÀ2 hÀ1; therefore for an 8 cm2 system with an active area of 4 cm2 and where the PVP drug reservoircontained 37 mg of sumatriptan at pH 7. Two patches were applied to patch, the estimated in vitro delivery rate would be each animal (total area = 8 cm2). A biphasic current profile was applied $1.2 mg hÀ1 (cf. 1.0 mg hÀ1 in vivo).
(dashed line, secondary y-axis), in phase 1, 1.8 mA (0.45 mA cmÀ2) for the Inspection of the patch application site did not reveal t = 0–180 min time-period, then in phase 2, 0.8 mA (0.2 mA cmÀ2) during any erythema; this is notable since one of the principal side the t = 181–360 min time-period. (Cmax, Tmax and AUC values were effects after subcutaneous injection of sumatriptan is local $100 ng mlÀ1, 105 min and 27,600 ng mlÀ1 min, respectively, for ionto-phoretic administration; cf. 194 ng mlÀ1, 5 min and 8480 ng mlÀ1 min for irritation In contrast, during a study into the iontopho- the subcutaneous injection) (mean ± SD; n = 3).
retic delivery of alniditan, another 5-HT1D agonist,(0.2 mA cmÀ2, formulation pH 9.5) in human volunteers,investigators noted the presence of local erythema at the that, in the absence of competing cations and with only a anode for up to 48 h, perhaps due to the elevated pH or single monovalent anion in the receptor, the iontophoretic a drug–skin interaction at the application site .
flux of a cationic drug depends on the respective mobilities Upon oral administration in humans (tablets containing of the drug and the anion and is independent of drug con- 25 and 100 mg sumatriptan), Cmax was reported to be 16 centration in the formulation . In these published and 54 ng mlÀ1, respectively; Tmax was 1.5 h for both doses studies, the drugs were hydrochloride salts with good aque- Since it is known that the total blood volume in York- ous solubility and did not require the addition of NaCl to shire swine is of the order of 2–2.5 l, it is possible to extrap- provide chloride ions necessary for anodal electrochemis- olate the results obtained in this study to the human try. In the current study, sumatriptan was supplied as the scenario to estimate (to a first approximation) whether they succinate salt. Hence, we used a patch design wherein the are therapeutically relevant. Peak drug levels achieved here electrode compartment contained a cation exchange resin (Cmax $ 90–100 ng mlÀ1) would be comparable to those and was separated from the drug-containing gel reservoir seen in humans following oral delivery (assuming human by a low molecular weight cut-off size-selective membrane blood volume of 5 l and hence using a scaling factor of (100 Da) to reduce the effect of competing cations.
$0.4); furthermore, Tmax is $1.75 h, again similar to that The iontophoretic flux of sumatriptan across porcine ear observed in man. However, sumatriptan transport kinetics skin in vitro was 2.6 ± 0.4 lg cmÀ2 minÀ1 (This have recently been reported as being $2-fold higher across was approximately 2.5 times less than that observed porcine skin than human skin in vitro . If this is the case (6.3 ± 0.4 lg cmÀ2 minÀ1) using the same current density in vivo, then the Cmax observed in this study would proba- and an aqueous formulation containing 14.5 mM suma- bly be closer to the lower range of peak values seen in triptan succinate (and 25 mM NaCl) at pH 6.5 . In humans. In contrast to subcutaneous (and oral) adminis- other studies with low molecular weight cations, we have tration, the iontophoretic patch enables drug levels to be found that transport rates from patch systems were some- maintained (if necessary) and also provides the possibility times lower (by up to $50%) than those observed using of administering a second bolus dose if required. Neverthe- aqueous formulations (unpublished results).
less, using the conditions employed in this study, sumatrip- tan iontophoretic delivery kinetics did not provide the rapid Tmax ($10 min) observed following subcutaneousinjection, which would be a key factor in determining use- fulness as a therapeutic system. The time required to attaintherapeutic levels following iontophoretic administration where CL is the clearance (ml minÀ1) and CSS is the depends on both the physicochemical properties of the concentration at steady state (mg mlÀ1). The half-life drug, which determine mobility through the skin, and the and clearance (T1/2 $ 120 min and CL $172 ml minÀ1, pharmacological potency. Sumatriptan is less potent than respectively, determined using a one-compartment model) the other triptans, which are effective at lower doses; thus, S.R. Patel et al. / European Journal of Pharmaceutics and Biopharmaceutics 66 (2007) 296–301 they may be better candidates for transdermal iontophore- [8] A. Femenı´a-Font, C. Padula, F. Marra, C. Balaguer-Ferna´ndez, V.
sis even if they possess similar electric mobility.
Merino, A. Lo´pez-Castellano, S. Nicoli, P. Santi, Bioadhesivemonolayer film for the in vitro transdermal delivery of sumatriptan,J. Pharm. Sci. 95 (2006) 1561–1569.
[9] Y.N. Kalia, A. Naik, J. Garrison, R.H. Guy, Iontophoretic drug delivery, Adv. Drug Deliv. Rev. 56 (2004) 619–658.
The in vivo data confirm that transdermal delivery of [10] E.R. Viscusi, L. Reynolds, F. Chung, L.E. Atkinson, S. Khanna, Patient-controlled transdermal fentanyl hydrochloride vs. intrave- therapeutic amounts of sumatriptan is feasible using an ion- nous morphine pump for postoperative pain: a randomized controlled tophoretic patch system. The transport kinetics observed using the Yorkshire swine model suggest that transdermal [11] A. Femenı´a-Font, C. Balaguer-Ferna´ndez, V. Merino, A. Lo´pez- delivery in humans, using an iontophoretic patch system, Castellano, Iontophoretic transdermal delivery of sumatriptan: could result in blood levels similar to those seen after oral, effect of current density and ionic strength, J. Pharm. Sci. 94 (2005)2183–2186.
nasal and rectal delivery . Moreover, no irritation was [12] S.R. Patel, H. Zhong, A. Sharma, Y.N. Kalia, Transdermal ionto- observed at the patch application site. However, iontopho- phoretic delivery of sumatriptan in vivo, Cephalgia 24 (2004) retic delivery kinetics were not comparable to those seen after subcutaneous injection (as indicated by the respective [13] P.W. Ledger, Skin biological issues in electrically enhanced transder- mal delivery, Adv. Drug Deliv. Rev. 9 (1992) 289–307.
max and Tmax). In the next phase of our studies we will [14] K.N. Cheng, M.J. Redrup, A. Barrow, P.N. Williams, Validation of a investigate the iontophoretic transport of more potent liquid chromatographic tandem mass spectrometric method for the antimigraine therapeutic agents that require delivery of determination of sumatriptan in human biological fluids, J. Pharm.
smaller amounts of drug to achieve a pharmacologic effect.
[15] D. Marro, R.H. Guy, M.B. Delgado-Charro, Characterization of the iontophoretic permselectivity properties of human and pig skin, J.
Control. Release 70 (2001) 213–217.
[16] M.J. Pikal, The role of electroosmotic flow in transdermal iontopho- [1] P.P.A. Humphrey, W. Feniuk, Mode of action of the anti-migraine resis, Adv. Drug Del. Rev. 9 (1992) 201–237.
drug sumatriptan, Trends Pharmacol. Sci. 12 (1991) 444–446.
[17] G.B. Kasting, J.C. Keister, Application of electrodiffusion theory for [2] P. Tfelt-Hansen, P. De Vries, P.R. Saxena, Triptans in migraine: a a homogeneous membrane to iontophoretic transport through skin, J.
comparative review of pharmacology, pharmacokinetics and efficacy, Control. Release 8 (1989) 195–210.
[18] R.V. Padmanabhan, J.B. Phipps, G.A. Lattin, R.J. Sawchuk, In vitro [3] R.E. Ryan Jr., Patient treatment preferences and the 5-HT1B/1D and in vivo evaluation of transdermal iontophoretic delivery of agonists, Arch. Intern. Med. 161 (2001) 2545–2553.
hydromorphone, J. Control. Release 11 (1990) 123–135.
[4] C. Duquesnoy, J.P. Mamet, D. Sumner, E. Fuseau, Comparative [19] P. Singh, S. Boniello, P. Liu, S. Dinh, Transdermal iontophoretic clinical pharmacokinetics of single doses of sumatriptan following delivery of methylphenidate HCl in vitro, Int. J. Pharm. 178 (1999) subcutaneous, oral, rectal and intranasal administration, Eur.
[20] D. Marro, Y.N. Kalia, M.B. Delgado-Charro, R.H. Guy, Contribu- [5] K.M.A. Welch, N.T. Mathew, P. Stone, W. Rosamund, J. Saiers, D.
tions of electromigration and electroosmosis to iontophoretic drug Gutterman, Tolerability of sumatriptan: clinical trials and post- delivery, Pharm. Res. 18 (2001) 1701–1708.
marketing experience, Cephalgia 20 (2000) 687–695.
[21] A. Jadoul, J. Mesens, W. Caers, F. de Beukelaar, R. Crabbe´, V. Pre´at, [6] M. Adlard, G. Okafo, E. Meenan, P. Camilleri, Rapid estimation of Transdermal permeation of alniditan by iontophoresis: in vitro octanol–water partition coefficients using deoxycholate micelles in optimization and human pharmacokinetics data, Pharm. Res. 13 capillary electrophoresis, Chem. Commun. 21 (1995) 2241–2243.
[7] A. Femenı´a-Font, C. Balaguer-Ferna´ndez, V. Merino, V.RodillaA.
[22] A. Femenı´a-Font, C. Balaguer-Ferna´ndez, V. Merino, A. Lo´pez- Lo´pez-Castellano, Effect of chemical enhancers on the in vitro Castellano, Combination strategies for enhancing transdermal percutaneous absorption of sumatriptan succinate: effect of current absorption of sumatriptan through skin, Int. J. Pharm. 323 (2006) density and ionic strength, Eur. J. Pharm. Biopharm. 61 (2005) 50–55.

Source: http://huizhong1969.files.wordpress.com/2010/10/hui_zhongs_paper_in_2007.pdf