Ambive~ of

Arch Dermatol Res (1986) 278:465-469 9 Springer-Verlag 1986

Regional variation in percutaneous absorption in man: measurement by the stripping method A. R o u g i e r 1, D. D u p u i s 1, C. Lotte ~, R. R o g u e t r, R. C. Wester 2, a n d H, I. M a i b a c h 2 1 Dtpartement de Biologie, Laboratoires de Recherche Fondamentale de l'Oreal, 1 Avenue de Saint Germain, 93601 Aulnay sous Bois, France 2 Department of Dermatology, University of California, San Francisco, CA 94143, USA

Summary. The influence of anatomic site on the relationship between total penetration of a molecule and its quantities present in the stratum corneum (SC) 30 min after application was quantified in an in vivo study. For each site, six male volunteers received two symmetrical applications of 1,000 nmol benzoic acid 14C to an area of 1 cm 2 for 30 rain. The first application permitted measurement of total absorption of benzoic acid within 4 days (urinary excretion method), while the second enabled determination of the quantity of benzoic acid in the SC at the end of the application time. Total penetration according to site is: back < arm < chest < thigh < abdomen < forehead, (with the forehead being three times more permeable than the back). Whatever the sites and the origin of the differences observed, the results show that the single measurement of the amounts of a compound present in the SC at 30 min postapplication appears sufficient to predict its total penetration, these two parameters being linearly correlated (r = 0.97, P < 0.001). Key words: P e r c u t a n e o u s a b s o r p t i o n A n a t o m i c site - Stripping m e t h o d

-

Man

-

physicochemical nature of the molecule studied [t9], the type o f animal [8], the dose [8], the time o f application [20], a n d the vehicle [9]. A m o n g other factors likely to m o d i f y p e r c u t a n e o u s a b s o r p t i o n , it is clear that, whether in vitro [22] or in vivo [5, 10, 15, 26, 27], a n a t o m i c site is o f m a j o r i m p o r t a n c e , even t h o u g h the relationship between the differences observed, the skin structure, a n d the physicochemical nature o f the molecule administered r e m a i n obscure. This study ascertains whether, by use of the stripping m e t h o d in m a n , it is possible to predict the rate o f a b s o r p t i o n o f a molecule, w h a t e v e r the a n a t o m i c area dosed.

Materials and methods Benzoic acid penetration was measured in six anatomic areas (Fig. 1). For each anatomic area, a group of six male volunteers aged 38 + 2 years were used. Two identical applications were made to the skin of each volunteer at an interval :of 48 h. The first application, designed to measure the total penetration of benzoic acid, was made on the right-hand side. The second, in the contralateral site, was used to determine the amount of benzoic acid present in the SC after washing at 30 rain. Application conditions

Introduction

One thousand nanomoles of benzoic acid (ring 14C) (New England Nuclear), with a specific activity of 10 ~3 gCi/nmol, was applied to a surface area of I cm 2 in 20 Ixl of a vehicle consisting of ethylene glycol, to which 10% Triton x 100 had been added as a surface active agent. For each anatomic site, the surface treated was delimited by an open circular cell fixed by silicone glue to minimize chemical spread. After 30 min of contact, excessive benzoic acid in the dosed area was rapidly removed by two successive washes (2 • 300 lal) with an ethanol-water mixture (95: 5), followed by two rinses (2 • 300 p.1) with distilled water, and lightly cleaned with a cotton-wool ball.

Previous experiments on rats [19] and m a n [8] have s h o w n that by quantifying the a m o u n t s o f chemical present in the s t r a t u m c o r n e u m (SC), m e a s u r e d b y stripping the treated area 30 rain postapplication, it is possible to predict the total a m o u n t o f m o l e c u l a r p e n e t r a t i o n over 4 days with a linear-type correlation existing between these two p a r a m e t e r s . T h e predictive aspect o f this new m e t h o d o f measuring p e r c u t a n e o u s a b s o r p t i o n is i n d e p e n d e n t o f the

Percutaneous absorption measurements

Offprint requests to: A. Rougier (address see above)

Urinary Excretion. Utilizing the existing data relating to the urinary excretion of benzoic acid administered to different ani-

466

A. Rougier et al.: Variation in human percutaneous absorption

reals by intravenous, oral [4, 6], or percutaneous routes [8, 1I, 19], the total amounts penetrating during the following 4 days were determined by liquid scintillation counting of the amounts found in the 24-h urine; this represents about 75% of the total amounts absorbed.

Results

Tape stripping. Following washing, at the end of the second application (30 rain), 15 adhesive tape strippings (3 M Invisible tape) were performed on the dosed area. The radioactivity in each tape was measured after complete digestion of the keratinous material in Soluene 350 (Packard Instruments), addition of Dimilume 30 (Packard Instruments), and liquid scintillation counting. The total amounts of benzoic acid in the SC at the end of the application were then determined by adding together the amounts found on each tape.

S-

For each anatomic site studied, Table 1 gives the amounts of benzoic acid found in the urine over 24 h and the total amounts having penetrated over a period o f 4 days (calculated). There were appreciable differences between sites. To facilitate comparison, Fig. 2 shows the permeability value of each area expressed in relation to that of the arm. Our result show that total benzoic acid penetration in relation to the anatomic sites treated is: back < arm < chest < thigh < abdomen < forehead. It is of note that the penetration levels of benzoic acid obtained on the chest and the thigh were statistically different from those measured on the back and the arm (P < 0.005). With exception o f the thigh, permeability o f the abdomen was different from those o f all the other sites studied (P < 0.001). Moreover, extending previous work [10, 15], it appears that the forehead is three times more permeable to benzoic acid than the back (P < 0.001) and twice as permeable as the abdomen (P < 0.001). For the different areas studied, Fig. 3 shows the linear-type relationship between the total penetration values of the benzoic acid and the amounts found in the SC at the end of the application (30 min) (Table 1). By knowing the chemical a m o u n t in the SC at the end of application (x), the linear correlation y - 1.83 x - 0.52 (r-= 0.97, P < 0.00I) enabIes calculation o f the total amounts likely to penetrate over a period of 4 days @). As shown in Table 1, there is a close correspondance between the predicted and the measured penetration values. Discussion

Fig. 1. Anatomic sites tested: 1, upper back; 2, upper, outer arm; 3, chest; 4, anterior thigh; 5, abdomen; 6, forehead

Although most investigators assume the importance of the anatomic site in percutaneous absorption, there

Table 1. Percutaneous absorption of benzoic acid and anatomic site Anatomic site

Back Arm Arm c Chest Thigh Abdomen Forehead

Urinary Excretion 0-24 h (1) 6.41" (0.99)b 7.06 (0,77) 8.78 (0.98) 9.38 (t.07) 11.44 (1.44) 20.74 (2.71)

Total penetration in 4 days (2)

Amounts in stratum corneum 30 rain after application

Predicted penetration in 4 days

Skin temperature ~

g.55 (1.32) 9.41 (1.02) 8.90 (2.63) 11.70 (1.31) 12.50 (1.43) 15.26 (1.93) 27.65 (3.6J)

6.19 (1.27) 5.92 (0.62) 5.01 (1.06) 8.07 (1.19) 6.80 (0.72) 8.20 (1.24) 11.91 (228)

10.80 (2.32) 10.31 (1.14) 8.65 (1.42) 14.25 (2.18) 11.92 (i.3l) 14.48 (2.18) 21.24 (4.16)

31.6 (0.4) 31.3 (0.8)

(2) Values calculated from (1), i.e., (2) = (1)/0.75 Expressed in nmol. cm- z application area b SD According to [8]

31.4 (0.3) 28.7 (0.3) 31.2 (0.5) 33.0 (0.2)

A. Rougier et al.: Variation in human percutaneous absorption RELATIVE PERMEABILITY, A R M = 1

[

1

1

2

3

1

I

I

RACK ARM

I

-I-CHEST

[

-4--

I

--f--

THIGH

[

I'

ARDOMEN FOREHEAD

I

--~----

Fig. 2. Total penetration of benzoic acid accordingto anatomic site

30

6t

/A

E

~. 25, 0~

r = 0,97

0

E 0 e-

" 20r

t~ zI 1 5 Q U.I

1 BACK 2 ARM

re"

3 CHEST 4 THIGH 5 ABDOMEN

lal 13.

z O

6 FOREHEAD

5

!

I

I

i

5

10

15

20

AMOUNTS IN THE STRATUM CORNEUM 30 min. AFTER APPLICATION ( nanomoles, cm -2) ( STRIPPING )

Fig. 3.

Correlation between the level of penetration of benzoic acid within 4 days and its level in the SC after 30 min of application according to anatomic site. Curve A: y = 1.83 x--0.52, obtained in man in 1984 [8]

are limited data documenting this important phenomenon. Furthermore, general reviews [1, 13, 22] refer to experiments, the results of which lead to often contradictory explanations for the differences of permeability between anatomic areas. Although it is easy to imagine that a molecule may enter the body by follicular and transcorneal routes simultaneously,

467

it is difficult to assess the real importance of each route. Our results show (Table 1, Fig. 2) that total benzoic acid penetration in relation to the anatomic sites treated is: back < arm < chest < thigh: abdomen < forehead. The laws of diffusion expressed in mathematical form emphasize the thickness of the membrane to be passed through. General laws of this type, even if largely true, may nevertheless be at fault when applied to a discontinuous barrier of great physicochemical complexity such as the SC. For example, it has been established that by increasing SC hydration, not only is its thickness increased but also its permeability [7, 16, 17]. In the same way, it has been demonstrated that the palm and the forearm have similar levels of permeability to hydrocortisone [10], parathion , and malathion [15], although the thickness of the SC of these two areas differs greatly. Furthermore, the thickness and the number of cellular layers of the SC of the anatomic sites involved in this study are roughly comparable (12 gm for 18 layers, on average), whereas their permeability to benzoic acid is quite different [12, 23]. These examples appear to at least partially disprove the inverse relationship said to exist between permeability and membrane thickness. This relationship, dictated by pure mathematical logic, is nevertheless insufficient. The thickness of the SC alone cannot explain the differences observed in the absorption of benzoic acid. Although it is true that sebaceous glands cover only 0.1% - 1% of the cutaneous surface, it may be imprudent to conclude that they play only a minor role in absorption. Benfenati and Brillanti [2] measured in man the number of sebaceous glands in relation to the anatomic site. Per surface unit, the forehead possessed 50-100 times more sebaceous glands than the arm, the chest, the thigh, or the abdomen. Furthermore, below the surface, these appendices do not form a fully developed SC. Therefore, it could be conceivable to partially explain the higher penetration in sites where sebaceous glands are more numerous by an increase of the transfollicular absorption. However, how do we reconcile the great disproportion between, for example, the number of sebaceous glands on the abdomen and the forehead (factor 100) and the relatively small difference between the rates of benzoic acid penetration in these two areas (factor 1.8). As a consequence, we agree with others [3, 14, 21, 24, 25] that, although it is probable that transfollicular passage exists, it may have been overestimated. As shown by Plewig and Marples [18], in adults the surface of the corneum cells in the forehead region

468

A. Rougier et al.: Variation in human percutaneous absorption

Table 2. Ratios of total percutaneous absorption

In Fig. 3 it can be seen that in man this relationship is again confirmed when different anatomic sites are treated. In this way, by simply measuring the amounts (x) of benzoic acid found in the SC at the end of application (30 rain), it is possible to predict the total amount (y) penetrating over 4 days. It is of note that the experimental points obtained in this study are very close to the correlation curve A (Fig. 3) established in man in 1984 [8]. As a consequence, with the aim of preserving the homogeneity of the method proposed, the predicted penetration has been calculated by means of the equation y = 1.83 x - 0.52. As can be seen (Table 1) for all sites studied, whatever factors are involved in their differences of permeability, there is excellent concordance between the measured values of benzoic acid penetration and those predicted afer 30rain using the stripping method. It should be noted that after an interval of 2 years, both the predicted and measured benzoic acid penetration values on the same site (upper arm), under identical experimental conditions (age of subjects, dose, vehicle, etc.), are similar to those obtained in 1984 [8] on different subjects and in another country (Austria). For reasons of technical convenience, this study was performed using a radioactive molecule. However, given the large amounts of chemical found in SC after washing, the stripping method should enable percutaneous absorption measurments to be performed on human beings, either by using appropriate analytic techniques or by considerably reducing the levels of radioactivity applied. We do not care to overinterpret the significance of our present data. Examination of other molecules of varying physicochemical properties and additional anatomic sites should be examined before overgeneralizations are made.

Compounds

Benzoic acid Hydrocortisonea Parathion b Malathion b

Forehead Back

Abdomen

3.30 3.50

1.80 2.00 2.40

a Adapted from Feldmann and Maibach [10] b Adapted from Maibach et al. [15]

is 30% smaller than that of the cells from other areas, such as the arm, the abdomen, or the thigh. It would be interesting to know what contribution is made by this structural peculiarity in the differences in perme~ ability observed. In each case, the cutaneous temperature of the anatomic areas treated were taken, as well as that of the forehead. The results show (Table 1) that on average the forehead temperature is 2~ higher than that of all the other areas studied (33 ~C compared with 31~ This difference, although small, is statistically significant (P < 0.001). Although it is accepted that, in vitro [17], temperature has a major influence on transcutaneous absorption, the small but nevertheless real differences of temperature in vivo between one site and another may mean that this parameter has, perhaps mistakenly, been accorded too little attention. Each site has its own peculiarities and only few have been mentioned here. These known and unknown pecularities combine, but to varying degrees, to create the differences observed during experiments. Although it is possible to consider that the different means of penetration may simultaneously play a role, it is also possible that the relative importance of each of them may vary according to the physicochemical nature of the compound administered. It should nevertheless be noted that (Table 2), in the case of such chemically different molecules as benzoic acid, hydrocortisone [10], parathion, and malathion [15], the penetration ratios between the anatomic areas common to these different experiments are comparable. Previously [19], it has been demonstrated that the total amount of chemical likely to penetrate over 4 days could be predicted by simply measuring the amount which resides in SC 30 rain postapplication. Furthermore, the strict correlation between these two parameters, observed on such different subjects as man and hairless rats [8], proved unrelated to the physicochemical nature of the molecule studied [19, 20], the dose [8], the period of application [20], and the vehicle [9].

Acknowledgement. The authors thank Jim McMaster for his excellent technical assistance.

References

1. Barry BW (1983) Dermatological Formulations: percutaneous absorption, In: Swarbrick J (ed) Drugs and the pharmaceutical sciences,vol 18. M. Dekker, New York Basel 2. Benfenati A, Brillanti F (1939) Sulla distribuzione delle ghiandole sebacee nella cute del corpo umano. Arch Ital Dermato/15 : 33 -42 3. BIank IH, Scheuplein RJ (1969) Transport into and within the skin. Br J Dermatol 81:4-- l0 4. Bridges 3W, French MR, Smith RL, Williams RT (1970) The fate of benzoic acid in various species. Biochemistry 118:47-51 5. Britz MB, Maibach HI, Anjo DM (1980) Human percutaneous penetration of hydrocortisone: the vulva. Arch Dermatol Res 267: 313 - 316

A. Rougier et al.: Variation in human percutaneous absorption

469

6. Bronaugh RL, Stewart RF, Congdon ER, Giles AL (1982) Methods for in vitro percutaneous absorption studies. I. Comparison with in vivo results. Toxicol Appl Pharmacol 62: 474 - 480 7. Cronin E, Stoughton RB (1962) Percutaneous absorption: regional variations and the effect of hydration and epidermal stripping. Br J Dermatol 74: 265 - 272 8. Dupuis D, Rougier A, Roguet R, Lotte C, Kalopissis G (1984) In vivo relationship between horny layer reservoir effect and percutaneous absorption in human and rat. J Invest Dermatol 82: 353 - 356 9. Dupuis D, Rougier A, Roguet R, Lotte C (1986) The measurement of the stratum corueum reservoir: a simple method to predict the influence of vehicles on in vivo percutaneous absorption. Br J Dermatol (in press) 10. Feldmann R J, Maibach HI (1967) Regional variations in percutaneous penetration of 14C cortisol in man. J Invest Dermatol 48:181 - 183 11. Feldmann RJ, Maibach HI (1970) Absorption of some organic compounds through the skin in man. J Invest Dermatol 54: 399 - 404 12. Holbrook KA, Odland GF (1974) Regional differences in the thickness (cell layers) of the human stratum corneum : an ultrastructural analysis. J Invest Dermatol 62: 415 - 422 13. Idson B (1975) Percutaneous absorption. J Pharm Sci 64: 901 - 924 14. Ktigman AM (1983) A biological brief on percutaneous absorption. Drug Dev Ind Pharm 9:521 - 5 6 0 15. Maibach HI, Feldmann RJ, Milby T, Serat W (1971) Regional variation in percutaneous penetration in man. Arch Environ Health 23 : 208 - 211 16. Malkinson FD (1962) Permeability of the stratum corneum. In: Montagna W, Lobitz WC Jr (eds) The epidermis. Academic Press, New York, pp 435-448 17. McKenzie AW, Stoughton RB (1982) Methods for comparing percutaneous absorption in steroids. Arch Dermatol 86: 6 0 8 - 610

18. Plewig G, Marples RR (1970) Regional differences of cell sizes in human stratum corneum. J Invest Dermato154:13 18 19. Rougier A, Dupuis D, Lotte C, Roguet R, Schaefer H (1983) In vivo correlation between stratum corneum reservoir function and percutaneous absorption. J Invest Dermatol 81:275-278 20. Rougier A, Dupuis D, Lotte C, Roguet R (1985) The measurement of the stratum corneum reservoir. A predictive method for in vivo percutaneons absorption studies: influence of application time. J Invest Dermatol 84:66-68 21. Scheuplein RJ (1967) Mechanism of percutaneous absorption. II. Transient diffusion and relative importance of various routes of skin penetration. J Invest Dermatol 4 8 : 7 9 - 88. 22. Scheuplein RJ (1979) Site variations in diffusion and permeability. In: Jarrett A (ed) The physiology and pathology of the skin, vol 5, Press, New York, pp 1731 -1752 23. Seheuplein RJ, Blank HI, Kligrnan AM (1974) In: Fitzpatrick TB (ed) Sunlight and man, University Tokyo Press, p 741 24. Tregear RT (1961) Relative permeability of hair follicles and epidermis. J Physiol (Paris) 156:707-713 25. Treherne JE (1956) Permeability of skin to some nonelectro. lytes. J Physiol (Paris) 133 : 171 - 180 26. Wester RC, Maibach HI (1983) Cutaneous pharmaeokinetics: 10 steps to percutaneous absorption. Drug Metab Rev 14:169-205 27. Wester RC, Maibach HI, Bucks DA, Aufrere MB (1984) In vivo percutaneous absorption of paraquat from hand, leg and forearm of humans. J Toxicol Environ Health 14: 759 762

Received December 24, 1985