398 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS compounds may penetrate the skin. Without considering the different interpretations about mechanisms acting on percutaneous absorption, it is however well established that the main barrier is constituted by the stratum corneum (3,8,9), which also acts as a reservoir for molecules topically applied (10, l l). Moreover, it is likely that, at the early step of the absorption process, the interaction between the physicochemical properties of the drug, the vehicle, and the horny layer plays an important role in total absorption. In Part I of the present study, we hypothesize that the amount of chemical present in the stratum corneum at the end of application may represent the stratum corneum vehicle partitioning and could also reflect the rate of penetration of the chemical. In Part II, we ascertained that this hypothesis is independent of the main factors likely to modify the absorption level of a compound, i.e., contact time, dose applied, vehicle used, anatomic site involved, and animal species chosen. PART I: IN VIVO RELATIONSHIP BETWEEN STRATUM CORNEUM CONCENTRATION AND PERCUTANEOUS ABSORPTION MATERIALS AND METHODS Ten radiolabeled ingredients (Table I), having very different physicochemical proper- ties and belonging to different chemical classes, were tested on the hairless rat. For each compound, a group of 12 female hairless Sprague-Dawley rats, aged 12 weeks, weighing 200 _ 20 g, was used. The compounds were dissolved in ethanol-water mixtures (chosen according to the solu- bility of the chemical), and applied on 1 cm 2 of dorsal skin delimited by an open circular cell fixed by silicone glue in order to prevent any loss of chemical. The applica- tion time was 30 min and the standard dose was 200 nmol.cm -2. At the end of appli- cation, the excess product was removed rapidly by washing twice with ethanol-water (95-5) (vol/vol), followed by two rinsings with distilled water and drying lightly with cotton wool. The twelve animals were then divided into two groups (Figure 1). The animals of group l, wearing collars to prevent licking, were placed individually in metabolism cages for four days. Table I Characteristics of the Test Molecules Compound Specific activity Purity (Ring-•4C) benzoic acid 55 mCi/mmol 98% (Carboxyl-•4C) acetylsalicyclic acid 45 mCi/mmol 97.5% (1,2-3H) dehydroepiandros terone 95 Ci/mmol 97 % (Carboxyl-•4C) sodium salicylate 40 mCi/mmol 98.5% (•4C) testosterone 44 mCi/mmol 97.5% (8-3H) caffeine 140 Ci/mmol 98% (•4C) thiourea 65 mCi/mmol 97% D-(1-•4C) mannitol 54 mCi/mmol 98.5% (1,2,6,7-3H) hydrocortisone 120 Ci/mmol 97.5 % (1,2-3H) dexamethasone 115 Ci/mmol 98 %

PREDICTING PERCUTANEOUS ABSORPTION 399 I RATS AT THE TERM OF AP•t. ICATION I='RC•ED4JI• GROUP UI•NA RY AND FECAL E XCI•TION DURING 4 DAYS TOTAL I:•N•TR ATED AMOUNTS WITHIN 4 DAYS SACRIFICE: Ak•0UNTS IN THE WHOLE • (S.C. EXCEPTED GROUP •) 6 SUCCESSIVE STFUPPtNGS ON EACH RAT TREATED AREA TOTAL AMOUNTS PRESENT IN THE S.C. RESERVOIR AT 30 MIN APPLICATION Figure 1. Procedures for determining total percutaneous absorption within four days and stratum corneum reservoir at the end of application. Urinary excretion was established by daily sampling of the urine and liquid scintillation counting (Packard Instruments 460 C, USA). The feces were collected daily, pooled, and counted by liquid scintillation after lyophilization, homogenization, and combus- tion of the samples with an Oxidizer 306 (Packard Instruments, USA). After four days, the animals were sacrificed and a series of six strippings was carried out on the dosed area to determine the amount of product (not penetrated) within 96 h. The remaining skin of the treated area (epidermis and dermis) was sampled and counted by liquid scintillation after digestion in Soluene 350 (United Technology Packard, USA). The carcasses were lyophilized, homogenized, and samples were counted by liquid scin- tillation after combustion. The total amount of chemical penetrating within four days was then determined by adding the amounts found in the excreta (urine + feces), in the epidermis and dermis of the application area, and in the whole animal body. At the end of application and washing, the stratum corneum of the treated area of the animals from the second group was removed by six strippings, using 3M adhesive tape. The radioactivity on each strip was measured after complete digestion of the keratinic material in Soluene 350 (United Technology Packard, USA), addition of Dimilume 30 (United Technology Packard, USA), and liquid scintillation counting. In our experi- mental conditions, the capacity of the stratum corneum reservoir for each compound has been defined as the sum of the amounts found in the first six strippings. RESULTS AND DISCUSSION The percutaneous absorption results (Table II, Figure 2) show that after 96 h, large differences exist in the amounts of substances that have penetrated through the skin. Thus, one can observe that the species that penetrates most, benzoic acid, penetrates 50 times more than dexamethasone.