Determination of Acitretin in the Skin, in the Suction Blister, and in Plasma of Human Volunteers after Multiple Oral Dosing

JEAN-PHILIPPE LAUGIER', CHRISTIAN SUREER~~, HOT BUN', JEAN-MARIE GEIGERS, KLAUS-PETER WILHELM~~, ALAIN DURAND', AND HOWARD 1. MAIEACH~~

Received August 7, 1992, from the 'Laboratoire Hospitalo-Universitaire de Pharmcocinetique et Toxicocinetique, Marseille, France, the #Dermatologische Universitatsklinik, Petersgraben 4, CH-403 I , Basel, Switzerland, the §Roche International Clinical Research Center, Strasbourg, France, and the 11 University of California, San Francisco, CA. publication November 22, 1993".

Accepted for

Abstract Several HPLC methods for quantification of acitretin and its 134s isomer in biological fluids have been described. Only limited data are available on determination of this drug in skin samples. Our objective was to improve the sensitivity and selectivity of existing methods to measure drug in small skin samples from humans treated with acitretin. With a new optimized mobile phase [methanol: acetonitrile (7:3, v/v), purified water with 1.5 'YO (v/v) acetic acid, mixed in a 85:15 ratio (vlv)] and a new internal standard (arotinoid ethyl sulfone), a limit of quantification of 1 ng/g tissue was reached. Nine male volunteers were given an oral daily dose of 50 mg acitretin for up to 28 days. Blood and skin samples (punch and shave biopsies, suction blister skin, and fluid) were taken at various time points during and after treatment. Drug concentration and metabolism in plasma and skin samples appeared to be linked in that the trans-isomer concentration was always higher than the cis-isomer concentration during dosing and 3 h after the last dose. However, 7 and 14 days after the last dose in plasma and in all tissue samples (except the shave biopsy), the all-transacitretin concentration rapidly decreased and approached the detection limit. In the shave biopsy, the all- trans-acitretin concentration remained higher than the 13-cis-acitretin concentration. Furthermore, the elimination of two isomers from the shave biopsy was delayed. Our HPLC method has provided a suitable tool for pharmacokinetlc and drug monitoring studies of all-trans-acitretin and 13-cis-acitretin that can be performed by any laboratory with a darkroom and a basic isocratic HPLC system.

The oral administration of acitretin provides a successful therapeutic approach in various cutaneous diseases such as psoriasis, Darrier's disease, lichen planus, and other conditions characterized by disturbed keratinizati0n.I-3 Acitretin [Neoti- gason; Soriatane; Ro 10-1670; all- trans-9-(4-methoxy-2,3,6- trimethylphenyl)-3,7-dimethyl-2,4,6,8-nonatetraenoic acid] is a retinoid that elicits biological responses similar to those of retinol

CH,O &k All-Trans-Acitretin

(vitamin A) and retinoic acid. It is the primary metabolite of its carboxylic acid ester analogue, etretinate, which is currently marketed for the treatment of severe psoriasis and various disorders of keratinization. The use of the carboxylic acid ester analogue, etretinate, has been hampered by an extremely long elimination half-life of up to 120 days after multiple dosing.4 Due to the high lipophilicity of etretinate, the drug is retained in human adipose tissue,5 resulting in extensive accumulation during chronic administration.6 Furthermore serious adverse effects, such as teratogenicity? have reduced the clinical applicability and have severely complicated the management of patients using this drug. Acitretin is less lipophilic and has the

Abstract published in Advance ACS Abstracts, February 1, 1994.

same spectrum of therapeutic effects as etretinate. As a consequence, this provides the therapeutic advantage of being more rapidly eliminated (terminal half-life, 33-96 h),8 than the parent compound (prodrug). Plasma pharmacokinetics of etretinate, acitretin, and its metabolites, as well as concentrations of etretinate during long-term therapy in skin and other organs have been investigated ex ten~ive ly .~ .~~~ In contrast, acitretin concentration determinations and acitretin metabolism in the skin after multiple dosing as well as after cessation of the treatment are infrequently reported.9-11

Appropriate dosage regimens are often based on the relations of time, concentration at the active site (biophase), and drug response in humans. For reasons of feasibility, correlation of drug effects and drug concentration in blood are most often looked for and satisfactory correlations are often found. To improve our understanding of drugs acting in the skin, efforts have been taken to access the biophase or target organ directly. Ethical considerations in the conduct of human pharmacokinetic studies, such as site of sampling, sample size, and frequency of sampling, render it necessary to improve existing analytical assays.12

Our objectives were to improve both sensitivity and selectivity of existing HPLC methods to measure acitretin concentrations in small skin samples from humans and to determine acitretin concentrations in the skin after oral administration by different skin sampling techniques and to compare these concentrations with those of plasma.

Experimental Section Human Volunteers-Seven healthy volunteers (all male; mean age,

30.3 years; range, 26-33 years; mean weight, 71.4 kg; range, 61-81 kg; non-smokers) took part in the study. A medical history was elicited from each, and all underwent a complete physical and laboratory examination before, twice during, and once after the study. All volunteers totally free of any pre-existing dermatologic disease had to have normal blood values and had not to be previously treated with any natural or synthetic retinoids. After the briefing, they were asked to sign a consent form. The trial was approved by the local Ethics Committee. The trial was performed at UCSF, Medical Center, Drug Monitoring Unit, San Francisco, CA.

Dosing-Capsules containing 25 mg of all-trans-acitretin were supplied by Roche Dermatologics, Nutley, NJ. The dose was 50 mg/day during 21 days in group I (n = 2; initials, CS, HO) and 28 days in group I1 (n = 5, initials; DZ, TO, SW, SR, DY). The study was performed on an out-patient basis. The drug was given with a standard breakfast once daily. Drug and breakfast intake was strictly controlled. Due to a change in the personal schedule of two volunteers, CS and HO were only in the study for 21 days. There was no medical reason for the cessation of the treatment of CS and HO after 21 days. Steady-state plasma concentrations are achieved after 10 d a y ~ . ~ J ~

Sample Collection-Blood and skin samples were collected at different time points8J1 during and for up to 2 weeks after the treatment (see also table IV, V). Blood (10 mL) as collected by venipuncture into oxalated Vacutainers. Skin samples were taken from the upper outer quadrant of the buttock after infiltrating 0.5 mL of 2% lidocaine:

@ 1994. American Chemical Society and 0022-3549/94/ 1200-623$04.50/0 Journal of Pharmaceutical Sciences / 623 American Pharmaceutical Association VOI. a3, NO. 5, M ~ Y 1994

epinephrine (1:100 000) subcutaneously. Shave biopsies were obtained by pressing a thin (0.5 mm) stainless steel template with a 7-mm hole to the skin so that the underlying tissue popped through the hole. The tissue was removed with a lateral sawing motion of a halved sterile stainless steel blade. Punch biopsies were obtained with a 4-mm punch. A suction blister apparatus (Dermovac; Medko Medical, Espoo, Fin- land)" was used to form the blisters. The suction cup was applied to the skin and a pressure of 40 Pa was sustained for -2 h to create three blisters with a diameter of 4 mm. Once the blisters were formed, the suction blister fluid was aspirated with a syringe. The three blister skin specimens were removed with a scalpel blade. Solid samples were weighed on a Mettler AT 20 balance (precision 10 pg).

Photo Protection-Because of photo degradation and photo isomer- ization of the retinoids, all manipulations with the samples were performed under minimal light exposure (e.g., yellow light, total light protection). Sampling from thevolunteerswas performedunder subdued light. All samples were stored in light-protected tubes.

Reagents and Standard Solutions-All solvents were of analytical grade (Car10 Erba RS HPLC, France). HPLC grade water was prepared from deionized water by purification with a Milli-0 Reagent Water System (Millipore Corp., France). All-trans-acitretin, 13-cis-acitretin,

Internal Standard aroi~nod ELYI sulfonr. Ro lS-lS70

and internal standard (arotinoid ethyl sulfone) were supplied by Hoffmann-La Roche Ltd., Basel, Switzerland.

Stock methanolic solutions (100 pg/mL) of all-trans-acitretin, 13- cis-acitretin, and the internal standard were prepared every 2 months, stored at 4 "C in amber glass volumetric flasks, and screened from light. Working solutions of these compounds were freshly prepared every week by successive dilutions of the stock solutions in methanol.15J6

Purity controls of reagents and standards and standard solutions were performed throughout the study. All procedures (storage, sample treatment, assay procedures) were validated for a 90 day period. Under these study conditions no isomerization or degradation occurred.

Extraction Procedure-Plasma Extraction-The procedure has been extensivelyvalidated."J6 Briefly, after addition of a suitable volume of internal standard and pH 7 buffer (100 pL, Titrisol), extraction from 0.3-1 mL of plasma (depending on expected concentration8*" was performedwith 2mL ofdiethylether/ethylacetate (l:l,v/v). Theorganic solvent layer was separated and evaporated under a stream of nitrogen (25 "C). The residue was then dissolved in 30-50 pL of methanol and transferred into an injection vial for HPLC analysis. This procedure provided a maximal extraction recovery ranging from 81 to 97 % .

SkinExtraction: After additionof internalstandard, the skin sample with 6 mL of ethyl acetate:diethyl ether (l:l, v/v) wm homogenized for 2 min with an Ultra Turrax T25 (Ika Verk Corp.) with a 8-mm cutter at 24 OOO rpm (temperature <21 "C). After centrifugation at 2000xg for 10 min, the organic layer was transferred to a 10-mL tapered tube and evaporated to dryness under a stream of nitrogen (temperature, <21 OC). The residue was then dissolved in 30-50 pL of methanol for subsequent HPLC analysis.

This homogenization procedure has been extensively validated and provided a complete (>98%) disruption of the skin tissue (confirmed by light microscopy). In contrast to other tissue homogenization procedures (e.g., pottering, dissolution with strong acids or bases, enzymatic degradation, and others17, the method utilized has the advantage that the drug is not isomerized during the tissue homoge- nization procedure.

Chromatography-All analyses were performed with a Kontron liquid chromatograph equipped with an autosampler (model 460) and a variable-wavelength UV detector (Uvikon, model 430). A 250 X 4.2- mm (inner diameter) reversed-phase column with Nucleosil C18,5 pm was used at ambient temperature (20-21 "C). The following mobile

phase was prepared the organic phase, methanoLacetonitrile (7:3, v/v), was filtered through a 0.2-pm membrane; the aqueous phase, purified water with 1.5% (v/v) acetate acid, was filtered through a 0.45-pm membrane. The elution solvent obtained by mixing the organic and aqueous phases (8515, v/v) was deaerated with helium. The flow rate was 1.2 mL/min, and the detection wavelength was 350 nm. Under these conditions, the retention times were 6, 6.9, and 7.8 rnin for 13- cis-acitretin, all-trans-acitretin, and internal standard, respectively (Figure 1). Peak-height ratios were computed witha Kontron data system D450, and calibration curves for plasma were obtained from least squares linear regression established daily from at least four calibration points.

Extraction Recovery-Fresh skin samples from cosmetic surgery were spiked with a methanolic solution of the compound of interest and incubated for 6 h to allow maximal drug partitioning into the skin sample.1SJ9 Subsequently, the above skin extraction procedure was applied.

There is a general problem with the determination of the extraction recovery in solid tissues (e.g., skin). The in vivo drug distribution into skin after systemic drug application may be different from the in uitro experiment where excised skin tissue is spiked with the drug. Ethical considerations do not allow determination in vivo of the extraction recovery of radiolabeled compounds in humans. The in uitro experiment with human skin allows the closest approximation to the in uiuo situation in humans. An additional in uiuo experiment was performed to investigate the efficiency of the extraction procedure. In a study with orally treated rats, we performed two successive extractions of the same skin sample. Less than 5% of the drug found in the first extract could be measured after a second extraction."

Results Apart from dry lips and dry mouth, no other clinical signs of

retinoid-related adverse effects were noted. Laboratory pa- rameters of all subjects were within the normal range during the after treatment.

Analytical Results-Chromatograms-Figure 1 shows typical chromatograms obtained under the described analytical conditions for plasma and the various skin samples. The chromatograms show an excellent selectivity for all-trans- acitretin (peak 2; retention time, 6.9 min), cis-acitretin (peak 1; retention time, 6.0 min), and internal standard (peak 3; retention time, 7.8 min). Endogenous plasma and skin components are well separated from the peaks of interest.

Precision and Accuracy-the precision and accuracy20 were estimated on spiked fresh human skin samples (200 mg) containing 1-50 ng of each compound (Table 1 and 2). Inter- ferences between the peaks of interest and the peaks of endogenous compounds are related to the weight of the skin sample. With increasing sample weight, the peaks of endogenous compounds may interfere with the peaks of interest. Therefore, precision and accuracy were estimated on skin samples that were heavier than those actually collected from the human volunteers. Intraday assay coefficients of variation were <12% for both compounds. The bias is <13 !% for both compounds at the limit of quantification, and <lo% over 5 ng/mL. Interday repro- ducibility results obtained for 1-20 ng per sample were satis- factory.

Limit of Quantification-The limit of quantification was defined as the lowest concentration that can routinely be determined with respect to acceptable precision and accuracy.20 The limit of quantification was estimated to be 1 ng/sample for both all-trans-acitretin and 13-cis-acitretin (Tables 1 and 2).

Limit of Detection-Limit of detection was defined by a signal-to-noise ratio of ~ 3 . ~ 7

Linearity-In the range 1-20 ng/sample, linearity results were good with correlation coefficients of >0.999 (Table 3). Intercepts of the calibration curves were not significantly different from zero.

Drug Concentration in Skin and Plasma-The data on drug concentration in skin and plasma are presented in Tables 4 and 5. The limited number of skin specimens was dictated by

824 /Journal of Pharmaceutical Sciences Vol. 83, No. 5, May 1994

5

i I I I I

4 n

2

I 3

12 16 In in

5

I

I I I 4 8 12 rnin

5

4

2 1

5 -1 E

:I 0

3

2 !

4 I 1 ~ I--

8 12

Figure 1-Typical chromatograms obtained under the described analytical conditions. The chromatograms show a satisfactory selectivity for all- trans-acitretin (peak 2), 13-cis-acitretin (peak 1). and the internal standard (peak 3). Sample weight or volume were 10-50 mg or 200 pL, respectively. Due to a disk error, the time scale in chromatogram A is 16 min, but in all other chromatograms (B-F), the time scale is 12 min. (A, top left) Suction blister fluid (blank). (B, top middle) Punch biopsy (blank). (C, top right) Suction blister skin (SW) 3 h after last dose. (D, bottom left) Suction blister fluid (SW) 3 h after last dose. (E, bottom middle) Punch biopsy (HO) 3 h after last dose. (F, bottom right) Shave biopsy (HO) 3 h after last dose.

Journal of Pharmaceutical Sciences / 625 Vol. 83, No. 5, May 1994

Table 1-Intraday Precision and Accuracy for All-fransacltretln and 13-c&Acltretln In Sklna

Suction Blister Skin-Fourteen days after the last dose, no drug was detected.

~~

All-trans-acitretin 13-cis-Acitretin Amount

Added (X,), CV, Bias, CV, Bias, ng Xmb S," % d %" x, s, % %

1 1.02 0.10 9.8 2.0 1.13 0.05 4.4 13.0 10 10.24 0.43 4.2 2.4 9.98 0.81 8.1 -2.0 20 20.43 2.43 11.9 2.1 19.76 0.81 4.1 -1.2 50 49.49 0.83 1.7 -1.0 49.18 1.15 2.3 -1.6

a n = 6. Mean value. Standard deviation (n - 1). Coefficient of Variation (100 X SJX,). "(X, - X,) X lOOlX,.

Table 2-Interday Preclslon and Accuracy for All-transAcitretln and 13-clsAcltretln In Skina

All-transacitretin 13-cis-Acitretin Amount Added CV, Bias, CV, Bias, (XJ, ng Xmb Sxc % d %" X, S, % %

~~~~

1 1.03 0.17 16.5 3.0 0.98 0.13 13.3 2.0 2 2.04 0.06 2.9 2.0 2.08 0.10 4.8 4.0 5 5.05 0.48 9.5 1.0 4.73 0.20 4.2 -5.4

10 10.04 0.45 4.5 0.4 9.87 0.67 6.8 -1.3 20 19.93 0.26 1.3 -0.3 19.96 0.24 1.2 -0.2

a n = 6. Mean value. Standard deviation (n - 1). Coefficient of variation (100 X &/X,). "(X, - X,) X lOO/X,.

Table 3-Llnearlty Test for Acltretln In Skin

All-transacitretin 13-cis-Acitretin Quantity Added, nga a b P a b r

1,2,5.10,20 0.097 0.019 0.999 0.108 0.025 0.999

a y = ax i- b. Correlation coefficient.

ethical concerns. Therefore, the data of each volunteer are fully presented. Variability is high due to the small number of specimens and the size of the skin samples. To compare the relative change of the all-trans- and 13-cis-acitretin concentra- tions, an additional factor was introduced (D) . The parameter D is +1 when the all-trans-acitretin concentration is higher than the 13-cis-acitretin concentration and -1 when the all-truns- acitretin concentration is lower than the 13-cis-acitretin con- centration.

Plasma-In a separate study, the plasma pharmacokinetics were evaluated.13 Median half-lives (tip) of all-trans- and 13- cis-acitretin are 38 and 70 h, respectively. Three hours after the last dose, in all cases, all-trans-acitretin concentrations are higher than the 13-cis-acitretin concentration ( D is +l). Seven and 14 days after cessation of the treatment, the inverse is observed ( D is -1).

Shave Biopsy--D remains positive throughout the drug monitoring period. All-trans-acitretin concentrations were generally lower in shave biopsies than in plasma 3 h after dosing, but remained detectable in some samples in higher concentrations until 14 days after the last dose. The disappearance of 13-cis- acitretin in plasma and skin samples was similar in terms of time and quantities.

Punch Biopsy-Biopsies taken 3 h after the last dose showed higher all-trans-acitretin than 13-cis-acitretin concentrations.

Suction Blister Fluid-Generally, drug concentration in the suction blister fluid was lower than in plasma. The disappearance of the isomers and the translcis relationship in the suction blister fluid and in plasma were similar in terms of time and quantities.

Discussion

To improve our understanding of drugs acting in the skin, efforts have been taken to access the biophase or target organ directly. Ethical considerations in the conduct of human pharmacokinetic studies, such as site of sampling, sample size, and frequency of sampling, render it necessary to improved existing analytical assays. The sensitivity and selectivity was enhanced over previously proposed and comparable assays for all-trans-acitretin and its main metabolite 13-ci~-acitretinl~-~~ by a simple isocratic HPLC system. Further improvements are the use of arotinoid ethyl sulfone (synthetic retinoid) as internal standard and the use of a newly designed mobile phase that enhanced the separation of all-trans-acitretin, l&cis-acitretin, and the internal standard to - 1 min. Plasma pharmacokinetic data obtained are consistent with previously reported data.a Low acitretin concentrations and the small sample size have often created methodological problems as the amount of drug in the skin approached the detection limit of the assays. With the proposed method, a limit of quantification defined as the lowest concentration that can routinely be determined with acceptable precision and accuracy of 1 ng/sample was reached.20 Assuming that the skin is the target organ of the aromatic retinoids, we focused our investigation on the quantification of both all-trans- and 13-cis-acitretin in the skin.

Drug concentration and drug metabolism in plasma and all skin samples appeared to be linked in that the trans-isomer concentration was always higher than the cis-isomer concen- tration during dosing and 3 h after the last dose. However, 7 and 14 days after the last dose in plasma and in all tissue samples (except the shave biopsy), the all-trans-acitretin concentration rapidly decreased and approached the detection limit. The elimination of the 13-cis-isomer was slower. In contrast, in the shave biopsy, the all-trans-acitretin concentration remained higher than the 13-cis-acitretin concentration after cessation of the dosing. Furthermore, the elimination of both isomers from the shave biopsy was slower compared with the other tissue samples. This may suggest that the trans form has a higher affinity to a cellular retinoid receptor than its corresponding cis form.22 The development of well-defined suction blisters14923 located subepidermally allow the access to a biophase close to the assumed target in the skin. However, our data show that the trans-cis relationship in the suction blister fluid parallels the plasma data and is different from the shave biopsy data after cessation of the treatment. These results suggest that the drug concentration and drug metabolism in the suction blister fluid mirror the situation in plasma rather than skin tissue. Despite the fact that the data from the suction blister fluid is not sufficient to draw final conclusions, this intriguing obser- vation must be investigated further with other compounds. Contamination of the suction blister skin by blister fluid during tissue sampling may falsify the 'real' situation. An in vitro experiment confirmed that acitretin partitions from a fluid into stratum corneum/epidermis almost in~tantaneously.~g

In recent publications, Granhoj-Larsen and co-workers9 and Meyer and co-workers10 reported on the acitretin concentration in the skin of patients and volunteers. They studied epidermal and subcutaneous tissue and suction blister fluid and suction blister skin, respectively. Their values are of the same order of magnitude as ours. Differences can be explained by the use of another shave biopsy technique and by different extration recovery values. With our analytical approach, which involves mechanical tissue homogenization in the presence of the internal standard, we avoid the extensive artificial isomerization of the

626 /Journal of Pharmaceutical Sciences Vol. 83, No. 5. May 1994

Table 4-All-trans and 13-c/sAcItretin Concentrations In the Suction Bluster Fluid (ng/g), the Suctlon Blister Skln (ng/g), and Plasma (ng/mL)

3 h after the 7 Days after 14 Days after the Last Dose

1 Week before Cessetion and 3 h after Dosing Last Dose the Last Dose

Sample Subject trane cisb DC tranp cisb DC tram? cisb DC tranp cisb DC

SBFd DZ 115 63 +1 219 71 4-1 NDe 12 -1 ND 4 1 sw 105 31 +1 71 34 4-1 ND 4 -1 ND 4 -1

Plasma DZ 243 100 +1 337 110 +1 1 21 -1 <1 5 -1 sw 116 38 +1 114 37 +1 <1 5 -1 <1 5 -1

SBSmf DZ LB L 218 43 +1 ND 31 -1 ND ND n.a./ sw 247 105 +1 370 284 4-1 ND 23 -1 ND ND n.a.

-h

a All-transacitretin. 13-cisAcitretin. Dis 4-1 when trans> cisand -1 when trans < cis. Suction blister fluid. ND, Not detected. Suction blister skin. 8 L, Sample lost. -, Not available. 'n.a., Not applicable.

Table 5-All-trans and 13-c/s-Acltretln Concentratlons In the Punch Blopsles (ng/g), Shave Biopsies (ng/g), and Plasma (ng/mL)

1 Week before Cessetion 3 h after the 7 Days after 14 Days after and 3 h after Dosing Last Dose the Last Dose the Last Dose

Sample Subject tranSe cisb DC tranp cisb DC tranp cisb DC trane cisb DC

PBd SR 84 NDe +1 t' t -g t t - t t - DY 826 ND +1 t t - t t - t t - cs HO

SBh DZ TO s w SR DY cs HO

Plasma DZ TO s w SR DY cs HO

t t

62 88

141 49

179 t t

243 465 116 200 522

t t

t t

41 26 28 21 28 t t

100 145 38

150 153

t t

276 139

+1 t +1 108 +1 212 +1 117 +1 92

106 210

+1 337 +1 450 +1 114 +1 520 +1 375

276 50 1

- -

- -

- -

76 60 t

31 52 24 51 48 76

110 153 37

160 190 119 244

+1 t t +1 t t

9 9 +1 29 16 +1 41 ND +1 ND ND +1 90 59 4-1 376 57 4-1 215 86 +1 1 21 +1 2 14 +1 <1 5 +1 1 27 +1 3 19 +1 <1 27 +1 2 34

-

- -

n.a.' +1 +1

+1 +1 +1 -1 -1 -1 -1 -1 -1 -1

n.a.

t t 5

ND ND ND 43 26 64 <1 <1 <1 <1 <1 <1 <1

t t 4

ND ND ND ND 19 29 5

<1 5 6 3 5 9

- - +1

n.a. n.a. n.a.

+1 +1 +1 -1 -1 -1 -1 -1 -1 -1

*All-transacitretin. 13-cis-Acitretin. D is +1 when trans > cis and -1 when trans < cis. dPunch biopsy. eND, Not detected (<0.5 g/sample). ' t, Ethical considerations related to the frequency of tissue sampling did not allow us to collect samples at each time point.'*B--, Not available. Shave biopsy. n.a., Not applicable.

acidic retinoids. The latter is inevitable and difficult to control or quantitate when the tissue is hydrolyzed before drug extrac- ti0n.~4

The oral administration of acitretin has provided a successful therapeutic approach. Nevertheless the oral therapy is associated with serious systemic adverse e f f e ~ t s . ~ , ~ ~ , , ~ ~ These effects could be minimized by topical delivery. Assuming the skin is the target organ for acitretin and the therapeutic response is a function of drug concentration in the skin, topical delivery should, in theory, be effective. Therefore, we compared our data with those obtained after a single 24-h topical application of all-trans- acitretin27 in humans (Table 6). The drug concentrations in the skin after systemic application in a steady-state situation were comparable with the drug concentrations reached after a single 24-h topical application of a saturated acitretin/isopropyl- myristate formulation. The conditions in our study represent in both cases the maximum dose that can be administered in a therapeutic situation: 50 mg acitretin per day is nearly the maximum tolerated dose for the multiple administration, and a saturated solution of acitretin in isopropylmyristate provides the highest drug concentration that can be topically applied in this vehicle. The data show that the drug is detected in a significant amount in the skin after systemic and topical administration. However, no beneficial effects in psoriasis and

Table 6-Total Drug Concentration In the Skln after Oral and Topical Administrationa

Total Drug Concentration, nglg tissue

Sample Oral Application Topical Applicationb

Punch biopsy 275 Shave biopsy 160 Suction blister fluid 200 Suction blister skin 460

160 360 90

3800

a See text for details of sampling. Data from ref 27.

other disorders of keratinization were observed up to now by the topical administration of acitretin. One may postulate that drug concentration at a particular site within the skin following both routes of administration could be different due to the direction of the drug concentration gradient. This hypothesis has also been postulated and illustrated by Parry et a1.28 in their study comparing the clinical efficacy of topical and oral aciclovir. Model predictions and in vivo data agree that topical administration of aciclovir result in a much greater total epidermal aciclovir concentration than oral administration. However, mathematical modeling of the aciclovir concentration gradient through the epidermis revealed that the drug concentration in the target site

Journal of Pharmaceutical Sciences / 627 Vol. 83, No. 5, May 1994

of the herpes simplex infection, the basal epidermis, was 2-3 times less after topical administration than after oral admin- istration. Furthermore, one may postulate that drug metabolism could be different depending on the route of administration. However data supporting this hypothesis are still incomplete.

Conclusions Our HPLC method has provided a suitable tool for pharma-

cological, pharmacokinetic, and drug monitoring studies of all- trans-acitretin and 13-cis-acitretin that can be performed by any laboratory with a darkroom and a basic isocratic HPLC system.

References and Notes 1. Geiger, J.-M.; Ott, F.; Bollag, W. Curr. Ther. Res. 1984, 35, 735-

740. 2. Geiger, J.-M.; Czarnetzki, B. M. Dermatologica 1988, 176, 182-

190. 3. Pilkington, T.; Brogden, R. Drugs 1992,43, 597-627. 4. Massarella, J.; Vane, F.; Bugg6, C.; Rodriguez, L.; Cunningham, W.

J.; Franz, T.; Colburn, W. Clin. Pharmacol. 1985,37, 439-446. 5. Vahlquist, A.; Rollman, 0.; Pihl-Lundin, I. Acta Derm. Venereol.

6. Rollman, 0.; Vahlquist, A. Br. J. Dermatol. 1983, 109, 439-447. 7. Happle, R.; Traupe, H.; Bounameaux, Y.; Fisch, T. Dtsch. Med.

Wochenschr. 1984,109, 1476-1480. 8. Brindley, C. J. Dermatologica 1989, 178, 79-87. 9. G r ~ n h a j Larsen, F.; Vahlquist, C.; Anderson, E.; Torma, H.;

Kragballe, K.; Vahlquist, A. Acta Derm. Venereot. (Stockh). 1992,

10. Meyer, E.; Lambert, W.; De Leenheer, A.; De Bersaques, J.; Kint, A. Br. J. Clin. Pharmacol. 1992, 33, 187-189.

11. Laugier, J.-P.; Berbis, P.; Brindley, C.; Bun, H.; Geiger, J.-M.; Privat, Y.; Durand, A. Skin Pharmacol. 1989,2, 181-186.

(Stockh). 1986,66, 431-434.

72, 84-88.

14. 15.

16.

17. 18.

19.

20.

21.

22.

23.

24. 25.

26.

27.

28.

12. Svensson, C. K. Clin. Pharmacokinet. 1989,17,217-222. 13. Surber, C.; Laugier, J.-P.; Geiger, J.-M.; Bun, H.; Durand, A.;

Maibach, H. I. Pharm. Res. 1992,9, 1365-1369. . Kiistala, U. J. Invest. Dermatol. 1968. 50. 129-137.

Al-Mallah, N. R.; Bun, H.; Coassolo, P:; Aubert, C.; Cano, J. P. J. Chromatogr. 1987,421, 177-186. Al-Mallah, N.; Bun, H.; Durand, A. Analytical Letters 1988,21, 1603-1618. WYSS, R. J. Chromatom. 1990.531.481-508. Surber, C.; Wilhelm, K-P.; Maibach, H. I.; Hall, L. L.; Guy, R. H. Fund. Appl. Toxicol. 1990,15, 99-107. Surber, C.; Wilhelm, K.-P.; Hori, M.; Maibach, H. I.; Guy, R. H. Pharm. Res. 1990, 7,1320-1324. Shah, V. P.; Midha, K. K.; Dighe, S.; Mc Gilveray, I. J.; Skelly, J. P.; Yacobi, A.; Layloff, T.; Viswanathan, C. T.; Cook, C. E.; McDowall, R. D.; Pittman, K. A.; Spector, S. Pharm. Res. 1992,9, 588-592. .._

Jakobsen, P.; Gronhoj Lanen, F.; Gronhoj Larsen, C. J. Chromatogr. 1987.415.413-418. Siegenthker, G.; Saurat, J.-H. In Acne and related disorders; Marks, R.; Plewig, G., Eds., M. Dunitz: London, 1988; pp 169-174. Schiifer-Korting, M.; Korting, H.; Lukacs, A.; Heykants., J.; Behrendt, H. J. Am. Acad. Dermatol. 1990,22, 211-215. Vahlquist, A. J. Invest. Dermatol. 1982, 79, 89-93. Ellis, C. N.; Voorhees, J. J. J. Am. Acad. Dermatol. 1987,16,267- 291. Haliaoua, B.; Saurat, J.-H. Br. J. Dermatol. 1990,122 (suppl), 135- 150. Surber, C.; Wilhelm, K.-W.; Berman, D.; Maibach, H. I. Pharm. Res. 1993, 10, 1291-1294. Parry, G. E.; Dunn, P.; Shah, V. P.; Pershing, L. K. J. Invest. Dermatol. 1992,98, 856-863.

Acknowledgments We thank Drs. R. Armstrong, B. Czarnetzki, C. Behl, D. Hartmann,

N. Meltzer, and the reviewers for their helpful discussions and suggestions.

628 /Journal of Pharmaceutical Sciences Vol. 83, No. 5, May 1994