Studies of In Vitro Skin Permeation and Retention of a Leukotriene Antagonist from Topical Vehicles with a Hairless Guinea Pig Model

SARAN KUMAR'', A. WASEEM MALICK', NOEL M. MELEER*, J. D. MOUSKOUNTAKIS*, AND

Received March 18, 1991, from 'Pharmacmica! Research and Development, Hoffinann-La Roche, Inc., Nutley, NJ 071 10, and *Formulath Research and Development, Roche Dermatobgics, Division of Hoffmann-La Roche, Inc., Nutley, NJ 071 10.

CHARAN R. BEHL'

Accepted for publication Auguet 21, 1991.

Ab8tmcl d A leukotriene antagonist [Ro 23-3544; 6-acelyi-7-[5-(4- acetyl-3-hydroxy-2-propylphenoxy)pentyloxy]-3,4-dihydro-2H1 -ben- zopyran9carboxylic acid; 11 was studied in vitro for its permeation through and retention in hairless guinea pig skin from various topical vehicles. Both the free acid and the sodium saH forms of the drug were used. The vehicles evaluated were polyethylene glycol 400, propylene glycol, dimethyl sulfoxide (DMSO), C,2-C,b alcohol lactates, dimethyl isosorbide, butyrolactone, methylpyrrolidone, hexyl laurate, isopropyl myristate, and ceprylldcapric triglyceride (Neobee M5). For the salt form of the drug, the hlghest permeability coefficient and retention were obtained from OMSO and methylpyrrolidone, respectively. For the acid form, however, the highest permeability coefficient and retention were obtained from hexyl laurate and DMSO, respecthrely. The hlghest permeation and retention values were not obtained from the same vehicle for either the salt or the acid form of the drug. This observation questions the validity of using permeation (flux) measurements to screen topical drugs and formulations. Although the precise reasons for this lack of correlation between permeation and retention are not known at this time, this study has shown that the solubility parameters of the drug and the vehicles used may play an important role. It seems logical to conduct skin retention studies rather than flux measurements in evaluating drug delivery from dermatological products.

The benzopyran derivative Ro 23-3644 [6-acetyl-7-[5-(4- acetyl-3-hydroxy-2-propylphenoxy)pentyloxyl-3,4-di hydro- W-l-benzopyran-2-carboxylic acid; 11 i s a leukotriene antag onist that may be effective in dermatological conditions such as psoriasis.1 Many researchers have used flux (permeation) measurements o f drugs from dermatological preparations using different sk in membranes to evaluate, develop, and optimize topical producta.2-zO The objective o f topical derma- tological therapy i s effective delivery o f drug to the sk in and not necessarily systemic drug delivery. An assumption in using flux studies to optimize and select topical formulations was that flux properties are predictive and indicative o f drug retention in the skin. Current research has shown tha t th is latter assumption i s no t necessarily the caee, and therefore, some researchers have started conducting skin retention studies.21-m The present study wae undertaken to evaluate the in vitro sk in permeation and retention o f 1 from a number o f vehicles. From the pseudo-steady-state flux data, the

'-1 23-3544

permeability coefficients were computed and correlated with the drug retention in the sk in and the theoretical part i t ion coefficients. The part i t ion coefficients were calculated by using the solubi l i ty parametera of the drug, the vehicles, and the skin. These parametera have been used to predict drug flux through sk in and to show vehicl-kin interactions.

Experimental Section Animals and Skin Membrane--Female hairlees guinea pigs

(Charles River Laboratories, Inc., Wilmington, MA) were used in these studies. The animals are designated Crl:IAF/HA-HO and were -8 weeks old. Animals were sacrificed by intraperitoneal iqjection of a solution of embutramide, mebezonium iodide, tetracaine hydro- chloride, and dimethylformamide (T-61, Hoechst Rouesel, Somer- ville, NJ). The abdominal skin was carefully excised and mounted on the permeation cells. Four replicates were used for each experiment, with skin specimens being obtained from four animals. Hairless guinea pig skin was used in these studies, because we had previously determined the nvalue (a measure of the degree to which a particular skin can differentiate among compounds of increasing lipophilicities) to be 0.26. This value is very close to that obtained for the human skin.=

MaterialeThe free acid and the sodium salt forma of 14C-labeled 1 (["Cll) and 1 were obtained from Hoffmann-La Roche, Inc. The isotope was at the 6-(acetyl-C-l) position for both forma of the drug. Hexyl laurate and isopropyl myristate were obtained from Henkel Corporation, Chemical Specialities Division (Ambler, PA); C 1 A l 6 alcohol lactates (Ceraphyl-rll) were obtained from Van Dyk 6 Company (Belleville, NJ); dimethyl sulfoxide (DMSO) was obtained from Fisher Scientific (Fair Lawn, NJ); caprylidcapric triglyceride (Neobee M5) was from Stepan Company (NortMeld, IL); methylpyr- rolidone and butyrolactone were from GAF Chemicals Corporation (Wayne, NJ); dimethyl isosorbide was from ICI Americas Inc. (Wil- mington, DE); propylene glycol was from BASF Corporation, Per- sonal Care Products (Parsippany, NJ); and polyethylene glycol 400 (PEG 400) was from Union Carbide Corporation, Industrial Chemi- cals Division (Danbury, CT). All chemicals were used as received.

Experimental Conditions-A summary of the experimental con- ditions used is shown in Table I.

Drug Formulations-The drug formulations were prepared fresh 1 day prior to conducting the experimente. These formulations were spiked with the radiolabeled drug in a proportion such that the specific activity was 1 &i/2S &. ["Cll (acid) was received as a solution in methano1:toluene (1:l) with a specific activity of 115 &iJmg, v d ["Cll (sodium salt) was received as a powder with a specific activity of 109 &i/mg. To spike the various formulations, the radiolabeled acid or sodium salt (reconstituted in ethanol) was evaporated to drynees under nitrogen, and a su5cient volume of solution formulations was added to provide the required proportion of the radioactivity. The solution was subsequently eonicated to ensure complete dissolution and homogeneity of radioactivity. The concen- trations of the drug in the various vehicles used are shown in Table II for the acid and in Table III for the salt forms.

Solubility Determination-The solubility of the drug (acid or sodium salt) was determined by placing an excessive amount of the

0022-3549/92/0700-063 1 $02.50/0 @ 1992, American Pharmaceutical Associstion

Journal of PharmaceuHcal Sdences I 691 Vol. 81, No. 7, July 1992

Table CExperlmental Condltlona

Variable Permeation cell Animal skin

Skin area, cm2 Temperature, "C Receiver medium Receiver volume, mL Donor medium Formulation applied, A Receiver sample size, mL Assay

Number of replicates

Description or Value Modified vertical Franz cell Hairless guinea pig; 8 weeks;

whole skin; abdominal 1.77 Receiver, 37; donor, 32 pH 8 USP phosphate buffer 10-15 Solution formulation 25 1 Liquid scintillation counting with

4 skins from 4 animals Ecoscint

drug in screw-cap vials with the vehicle and allowing the mixture to equilibrate for >17 h on a vibrating mixer in a water bath maintained at 25 "C. The saturated solution was then centrifuged, and 2.0 mL of the supernatant liquid was filtered through a Millex HV 0.45-pn filter. The solution was assayed by high-performance liquid chroma- tography with a UV absorption detector. The high-performance liquid chromatographic system consisted of a n automated sample injector (WISP model 712), a reciprocating piston pump (Waters model 600E), a programmable multiwavelength detector (Waters model 4401, and a Novapak C,, reversed-phase column. The mobile phase was meth- anol:0.1% phosphoric acid (80:20). The drug was detected at a wavelength of 280 nm and had a retention time of -6 min at a flow rate of 1.0 mumin at room temperature.

Permeation ApparatueModified vertical water-jacketed Franz cells (Crown Glass Company, Somenrille, NJ) were used. The mod- ification involved is noted below. The apparatus consisted of receiver and donor compartments. The receiver side was jacketed to maintain a desired temperature, and the contents were stirred with a magnetic stirrer. The receiver compartment was also equipped with a side arm to withdraw samples. The donor compartment was an open-face cylindrical chamber that was placed on the flange of the receiver chamber and held in place with a clamp. The exposed skin had a surface area of 1.77 cm2. The modification of the Franz cell was such that no " 0 ring waa used between the two compartments because the cell half surface was ground to provide a leak-tight system. The assembled cell waa placed on a drive unit that provided circulation of constant-temperature water through the jacket on the receiver compartment and stirring of its contents.

Permeation Procedure-The liw3hly excised skin membrane was sandwiched between the two compartments of the permeation cell. The two cell halves were aeeembled together with a clamp and placed on a drive unit. The receiver compartment was filled with a pH 8.0 phosphate bufFer solution and maintained at 37 "c with constant stirring at 600 rpm. A 25-pL aliquot of the solution was evenly applied over the skin. The concentration of drug in the receiver compartment was monitored at 0, 0.5,2,16,18,21, and 24 h. A constant volume was always maintained in the receiver compartment.

Skin Retention Procedure-At the end of each permeation exper- iment, the epidermal surface of the skin was washed three times with 0.6-mL aliquots of ethanol with gentle stirring to dissolve unabsorbed drug and residual formulation. The skin was then digested in 10 mL of 0.3 N NaOH overnight in an oven at 70 "C. The digest was assayed for radioactivity to determine the drug content. Assay Procedure-Both permeation and retention samples were

assayed with a Beckman LS 9800 scintillation counter with Ecoscint as the scintillation cocktail. The digested skin samples were assayed by taking a 1-mL aliquot of the digested sample and neutralizing it with 1 mL of 0.3 N HC1 in the scintillation cocktail prior to determining the radioactivity.

Results The permeability coefficients (P) for both the free acid and

the salt forms of the drug (Tables I1 and 111, respectively) were determined with eq 1:

In eq 1, F is the flux of a drug from a vehicle (mg/h), P is the permeability coefficient (cm/h), A is the skin area exposed to the drug solution (crn'), and AC is the concentration gradient across the skin. The value of AC can be approximated by using the concentration of drug in the donor compartment because this value is much larger than the concentration of drug in the receptor compartment. The amount that permeated through the skin was plotted as a function of time. From the slopes of steady-state portions of the plots, the values of P were computed. The retention values reported in the last columns of Tables I1 and I11 were determined by extrapolating the observed retention values to the solubility of drug in the vehicle. This extrapolation is based on our finding that skin retention increases linearly with concentration of drug.*' Because of the wide range of drug solubility in the vehicles, the extrapolation provides a means for ranking the maximum retention that can be achieved from these vehicles. The ranking of P and retention values are also provided in Tables I1 and III.

The method for calculating solubility parameters,=.2Q us- ing the free acid form of 1 as an example, is shown in Table IV. The calculated solubility parameters and theoretical partition coefficients ( K ) for the drug for the various vehicles used are given in Table V. Figure 1 shows plots of both the logarithm of the theoretical K values and the logarithm of the experimental K values versus the solubility parameter of the vehicles used. The rank orders of the permeation and reten- tion data for the acid and the salt forms of the drug were used to determine the acid-to-salt ratios for both the P values and skin retention values (Table VI).

Dlscussion One objective of the preclinical development of a new

topical drug candidate is the selection of a formulation that maximizes the therapeutic activity of the drug. In the past, the selection of a topical formulation has often been based on the optimization of the flux of the drug through the skin. Such studies have been carried out with the assumption that membrane saturation prevails during steady-state conditions of F, based on Fickian diffusion kinetics.2 In this study, drug retention in the skin was determined at the end of experi- ments on permeation from various vehicles, for both the acid (Table 11) and the salt forms of the drug (Table 111). No apparent relationship between permeability and skin reten- tion was observed for either form of the drug from the various vehicles.

The steady-state F of a drug from a vehicle across the skin is given by the passive diffusion equation derived from Fick's law (eq I). The value of P can be described by eq 2:

P = KDIh (2)

In eq 2, K is the partition coefficient for the distribution of a drug between skin and a particular vehicle, D is the diffusion coefficient for the permeation of drug in the skin, and h is the thickness of the barrier layer of the skin. The value of Dlh is constant as long as the value of D for a drug in a particular type of membrane is independent of the vehicle used. Thus, the experimentally determined P value should be directly proportional to K:

P = (constant)K (3)

On the basis of the preceding theoretical background, Sloan et a1.m used eq 4 for the theoretical determination of K that usea solubility parametem of the drug (41, vehicle (&I, and skin (4): F = PAAC

632 I Journal of Pharmaceutical Sciences Vol. 81, No. 7, July 1992

Tabk ICSkln Retentlon and Permeation Parametera for AcM Form of [l4C]1

Vehicle ROmC Retention, ~O,," pg/cm2b

Applied Concentration, Solubility, P x 103, mg/mL mg/mL Cm/h

Hexyl laurate 0.81 8 1.19 21.3 1 10.1 10 lsopropyl myristate 0.860 1.18 11.5 2 11.6 9 Ceraphyl-41 4.61 8.89 9.02 3 15.8 6 DMSO 4.68 549 6.71 4 5508 1 Neobee M5 1.82 3.12 4.14 5 15.2 7 Methylpyrrolidone 31 3 626 3.42 6 3306 2 Dimethyl isosorbide 154 309 1.53 7 21 7 4 Propylene glycol 3.06 5.56 0.958 8 28.5 5 Butyrolactone 136 273 0.140 9 31 9 3 PEG 400 40.1 79.6 0.0384 10 12.9 8

retention. Rank order of permeability. The observed retention values were extrapolated to the saturation solubility of drug in vehicle. Rank order of

Table IlI-Skln Retentlon and Permeetlon Parametem for Salt Form of [14C]1

RO,' Retention, wcm2b RO-" Applied Concentration, Solubility, P X 103,

mg/mL mg/mL cm/h Vehicle

DMSO Meth y lpyrrolidone Ceraphyl-41 Dimethyl isosorbide Propylene glycol Butyrolactone PEG 400

2.68

3.29 1.63 4.87 1.06

192

43.5

4.18 48.5 1 40.2 3 382 31 .O 2 1787 1 12.6 26.3 3 27.7 4 1.64 13.9 4 3.21 7 9.17 9.1 1 5 54.2 2 1.53 3.94 6 4.55 6

86.3 0.187 7 11.8 5

a Rank order of permeability. *The observed retention values were extrapolated to the saturation solubility of drug in vehicle. Rank order of retention.

Table IV-Group Contrlbutlon Method for Calculating the Solublllty Parameter of the Acld Form of 1

Table V-Calculated Solublllty Parametera of the Vehlcba and log Partltlon Codflclentr (SklnNehlcle) for the Acld Form of 1

Number Atom or of Group Groups

Aq, cal/mol" A V , cm3/mol

3 3 X 1125= 3375 3 X 33.5= 100.5 9 9 x 1180= 10620 9 x 16.1 = 144.9

CH OH 2 2x5220=10440 2 x 13= 26.0 c=o 3 3 X 4150 = 12450 3 x 10.8 = 32.4 -c= 8 8x1030= 8240 8 X -5.5=-44.0 0 3 3 X 800= 2400 3 X 3.8= 11.4 Doublebond 6 6 x 400 = 2400 6 x -2.2= -13.2 Ring closure 3 3 x 250 = 750 3 x 16 = 48.0

ZAV = 301.0

CH3 CH2

5 5 x 820= 4100 5 X -1.O= -5.0

ZAq = 54 775

S, = (ZAqZAW'" 7 (54 775/301)'12 = 13.5 (caVm3)'", where q is the energy of vaporization.

In eq 4, V, is the molar volume of the drug, R is the gas constant, and T is absolute temperature. Good correlation between theoretical log K and experimental log P to describe the flux of theophylline from various vehicles has been observed for vehicles that exhibit S, values in the range 12-18 (~a l /cm~)"~. For vehicles or mixtures of vehicles with S, values in the range 8-12 (cal /~rn~)"~, no such correlation between theoretical log K and experimental log P was ob- served. This lack of correlation was attributed to "vehicle effects" on the skin.30 The values of S; for 1 and 8,. for the vehicles used in the present study were determined by using the group contribution method of Fedors28.29 (the method is shown in Table IV for 1, and the S, values for all vehicles are presented in Table V). The S, values fall between 8.5 and 10.6, except for propylene glycol, with S, = 13.5. Although, a plot of the theoretically determined P values versus S, values

Vehicle Solubility

Parameter, log (caVcm3)'" a

~ ~ _ _ _ _ _ _

lsopropyl myristate Hexyl laurate Dimethyl isosorbide DMSO Ceraphyl-41 Butyrolactone Methylpyrrolidone PEG 400 PEG

8.50 8.62 8.73 8.81 9.50 9.91

10.1 10.6 13.5

2.76 2.51 2.28 2.1 1 0.81 3 0.138

-0.150 -0.832 -2.65

a See refs 27 and 28. Determined from eq 4, with T= 305 K; S, = 13.5 (caVcm3)'" (from Table IV); S, from above; y = 301 cm3/mol (from Table IV); R = 1.98 cal/deg - mol; and S, = 10.0 (caI/~m~)''~ (see ref 29).

follows the expected parobolic relationship (Figure l), as predicted from eq 4, the experimentally determined P values did not show the same pattern. Because the S, values of the vehicles selected in this study fall in the &12 range, these vehicles are likely to affect the skin. Therefore, these vehicles show positive or negative deviation from the expected trend observed by Sloan et al.30 Similar observations of a vehicle effect on skin permeability of vidarabine (ara-A) were shown with DMSO.31

In this study, the permeability and the retention of both the acid and the salt forms of 1 were determined from the same vehicles. To evaluate better the effect of the drug form on permeability and skin retention, the ratios of P and skin retention were calculated (Table VI). In general, the perme- ability of the acid form was lower than that of the salt form of the drug. However, the acid form usually provided higher skin retention than the salt form, except with propylene glycol as the vehicle, in which case higher skin retention was observed with the salt form. These results do not conform to the classical

Journal of Pharmaceutical Sciences / 633 Vol. 81, No. 7, July 1992

I

8 a 10 11 12 13 14 Solubility Porometer of Vehicle8

Flgure 1-Theoretical log K (0) or experimental log P (0) of [14C]1 versus S, values of various vehicles.

Table VCAcld-to-Salt Ratloa of the Permeablllty CoeMcknt and Retentlon Values of I from the Vehlclea Studled In the In Vttro Halrless Gulnea Plg Model

~~ _ _ ~

AcidlSalt Acid/Salt Rank Permeability Rank Retention Vehicle

Ceraphyl-41 1 0.343 6 0.570 PEG 400 2 0.205 4 67.6 DMSO 3 0.138 1 137

Dlrnethyl isosorbide 5 0.110 3 67.6 Propylene glycol 6 0.105 7 0.526 Butyrolactone 7 0.0355 2 70.1

Meth ylp yrrolidone 4 0.110 5 1.85

pH-partition hypothesis that states better absorption is ex- pected from un-ionized species. Although the pH-partition hypothesis is based on dissociation in an aqueous medium, the formulations used in this study did contain either ionized (salt) or un-ionized (acid) fonts of the drug. The nonaqueous nature of the vehicles used in this study may be responsible for the deviation from the pH-partition hypothesis.

Because most of the pharmaceutical solvents and vehicles have S, values in the range 8-l2,32 these vehicles may be assumed to interact with the skin and, therefore, alter the values of P for the permeation of drugs from these vehicles into the skin. This assumption is supported by the large differences observed for the F values of 1 from the vehicles used in this study. However, the skin retention of 1 is not consistent with what might be expected from the measure- ments of F. Similar observations have been made by other researchers.3- The value of 6 may be one of the factors affecting the correlation between F and retention. Other factors may lead to deviations of the correlation between skin F and retention values. For example, skin is an active site of drug metabolism for some drugs that are applied topically.39 In addition, the stratum corneum is a reservoir for lipophilic drugs.40 On the basis of these deviations, programs to develop topical dermatological products should emphasize identifica- tion of formulations based on their ability to provide maxi- mum drug delivery to the target organ skin. However, even in vitro skin retention may not correctly predict in vivo skin retention. In vivo animal models could be used to determine such information; one such model is the in vivo hairless guinea pig model.41 Selection of a formulation based solely on F could be erroneous and misleading.

References and Notes 1. Welton, A. F. U S . Patent 4 885 309, 1989.

2. Barry, B. W. Dermatological Formulations: Percutaneous Ab-

3. Franz, T. J. J. Invest. Dermatol. 1975, 64, 190. 4. Poulsen, B. J. In Drug Design; Ariens, E., Ed.; Academic: New

5. Mckenzie, A. W. Arch. Dermatol. 1962,86, 611. 6. Wester, R. C.; Maibach, H. I. J. Invest. Dermatol. 1976,67, 518. 7. Bronaugh, R. L.; Stewart, R. F.; Congdon, E. R.; Giles, A. L.

8. Idson, B. Drug Metab. Rev. 1983,14, 207. 9. Bronaugh, R. L.; Congdon, E. R.; Schuplein, R. J. J. Invest.

10. Dempski, R. E.; Portnoff, J. B.; Wase, A. W. J. P h r m . Sci. 1969,

11. Komatsu, H.; Suzuki, M. J. Phurm. Sci. 1979,68,596. 12. Scheuplein, R. J.; Blank, I. H.; Brauner, G. J.; Macfarlane, D. J.

13. Roberta, M. S.; Anderaon, R. A,; Swarbrick, J.; Moore, D. E. J.

14. Stolar, M. E.; Rossi, G. V.; Barr, M. J. Am. Pharm. Assoc. 1960,

15. Roberts, M. S.; Horlock, E. J. Pharm. Sci. 1978,67, 1685. 16. Maruta, H.; Muami, K.; Yagamata, T.; Noda, K. Kurume Med. J.

17. Shahi, V.; Zatz, J. L. J. Pharm. Sci. 1978, 67, 789. 18. Shrewsbury, R. P.; Foster, T. S.; Dittert, L. W.; Quigley, J. W.;

Leavell, U. W. Curr. Ther. Res. Clin. Exp. 1980,28, 1002. 19. Franz, J. M.; Gaillard, A.; Maibach, H. I.; Schweitzer, A. Arch.

Dermatol. Res. 1981,271, 275. 20. Cooper, E. R.; Patel, D. C. In Topical Dr Delivery Formula-

tions; Osborne, D. W.; Amman, A. H., Eds.Sarce1 Dekker: New York, 1990; pp 1-12.

21. Wohlrab, W. Acta Dermatol. Venerol. (Stockholm) 1984,64,233. 22. Wohlrab, W.; Lasch, J. Dermatologica 1987, 174, 18. 23. Tauber, U. Anneim.-Forsch. 1987,37(1), 461. 24. Schaefer. H.: Zesch. A.: Stuttaen. G. Arch. Derm. Res. 1977.258.241.

sorption; Marcel Dekker: New York, 1983; pp 234,295.

York, 1973; Vol. 41, pp 149-192.

Taricol. Appl. Pharmacol. 1982, 62, 474.

Dermatol. 1981, 76, 94.

58, 579.

J . Invest. Dermatol. 1969,52, 63.

Pharm. Pharmmol. 1978,30,486.

49,144.

1977,24, 131.

25. Schaefer, H:; Stutt en, G.; Csch , A,; Schalla, W.; Gazith, J. Curr. Probl. Dermatol. 1578, 7, 80.

26. Behl, C. R.; Kumar, &;.Malick, A. W.; Patel, S. B.; Char, H.; Piemontese, D. In Methods For Skin Absorption; Kemppainen, B. W.; Reifenrath, W. G., Eds.; CRC: Boca Raton, FL, 1990; pp 1-21.

27. Mehta, D. M.; Kumar, S.; Malick, A. W.; Behl, C. R., unpublished resulte.

28. Fedora, R. E. Polym. Eng. Sci. 1974,16,472. 29. Fedora, R. E. Polym. Eng. Sci. 1974,14, 147. 30. Sloan, K. B.; Koch, S. A. M.; Siver, K. G.; Flowers, F. P. J. Invest.

Dermatol. 1986.87. 244. 31. Kurihara-Bergstrom, T.; Flynn, G. L.; Higuchi, W. I. J. Invest.

Dermatol. 1987, 89, 274. 32. Weaat, R. C.; Aitle; M. J.; Beyer, W. H. Handbook of Chemist

andPhysics, 68th ed.; CRC: Boca Raton, FL, 1988; pp C676, C67r 33. Shah, V. P.; Behl, C. R.; Flynn, G. L.; Higuchi, W. I.; Schaefer, H.

J . Pharm. Sci., in press. n, G. L.; Higuchi, W. I.; Schaefer, H.;

Maibach, d. I. In Dermato&al TherapeuticProducts Workshop ZZ. Princi les in the Establishment of Bioavarlability and BioequivaLnce of Dermatological Products; workshop report in

34. Shah, V. P: Behl, C. R.; Fl

- . . preparation.

35. Behl, C.R. In Proceedings of Topical Thempeutic Products Workshop I . Principles and Criterur for the Development and Optimization of Topical Thempeutic Products; in preparation.

36. Higuchi, W. I. In Proceedi s of Topical Thempeutic Products Workshop I . Prrnciples anycriterur. for the Development and Optimization of Topical Thempeutic Products; in preparation.

37. Flynn, G. L. In Proceedings of Topical Therapeutic Products Workshop I . Principles and Criterur for the Development and Optimization of Topical Therapeutic Products; in preparation.

38. Behl, C. R. Presented at the Eastern Regional Meeting of the AAPS, New Brunswick, NJ, June, 1990; paper PDD-1.

39. Noonan, P. K.; Wester, R. C. In Percutaneous Absorption: Mech- anisms, Methodolo y , and Drug Delive Bronau h, R. L.; Mai- bach, H. I., Ede.; darcel Dekker: Newyork, 198l; pp 53-75.

40. Menael, E.; Touitou, E. In Percutaneous Absorption: Mecha- nisms, Methodolo and Drug Delive Bronau h, R. L.; Mai- bach, H. I., Eds., &cel Dekker: Newqork, 1988; pp 121-133.

41. Kumar, S.; Char, H.; Patel, S. B.; Piemontese, D.; Malick, A. W.; Behl, C.R. In Proceedings of Topical Thempeutic Products Workshop 1. Principles and Critena for the Development and Optimization of Topical Thempeutic Products; in preparation.

634 I Journal of Pharmaceutical Sciences Vol. 81, No. 7, July 7992