TESTING DRUGS FOR DERMAL TOXICITY 387 branes when there is a danger of contamination or when it is intended to apply a medicament for a therapeutic purpose. Tests should be carried out on the specific mucosa to which the compound is to be applied, but this may not always be the best for quantitative comparisons. The eye of the rabbit is most commonly used. 0.1 ml of a solution of the compound is instilled into the cupped lower lid of one eye and held for 30 seconds. The other eye is left untreated as the control. The solution may be washed out after a certain period of time, but it is more usual to leave it in contact unless it is causing severe damage. The effects on the cornea and conjunctiva are read at intervals over 7 days and scored according to an arbitrary scale, such as the one described by Draize (1). Substances which do not irritate the eye are unlikely to be irritant to the skin, but eye irritation does not preclude the use of a chemical on the skin. In addition to the eye, tests may also be carried out on the penile mucosa of the rabbit, and in the vagina of the rabbit, guineapig, rat, monkey or dog. SENSITIZATION STUDIES Sensitization reactions in the skin represent an allergic reaction to the chemical, and require prior exposure for development. They are prone to occur in humans, but are difficult to produce in animals. Little reliance can be placed on animal tests, and trials need to be made in human volunteers. DraJze (1) does describe a test in guineapigs which will detect severe sensitizers. A solution, or suspension, of the compound in normal saline is injected intradermally into the skin of guineapigs every other day up to a total of 10 injections over an area of 3 to 4 cm •. The first injection is 0.05 ml, and the remaining nine, 0.1 mi. Two weeks after the last injection a retest injection of 0.05 ml is made in an area just below the region of the sensitizing injections. 24 hours later readings are made of the diameter, height and colour of the reaction, and compared with the average of those observed for the original injections. Sensitization studies may include photosensitization, and this may be determined in rabbits submitted to ultraviolet light. SYSTEMIC I)ERMAL TOXICITY When drugs are applied to the skin for a local therapeutic effect it is possible for systemic absorption to occur, leading to distant toxic effects. The dermis forms a barrier to absorption, and if it is damaged considerable absorption may occur. Absorption is much greater from mucous membranes. For systemic effects, the drug must be absorbed into the tissues and general circulation, and the damage may occur in distant organs such as

388 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS the liver and kidneys. Acute effects follow a single application, chronic effects during repeated application over a prolonged period. An indication of the risk is obtained from a pharmacological and toxi- cological study of the drug given orally or parenterally in the normal type of toxicity test. This will allow a comparison between the amount of drug likely to be applied locally, and the dose likely to produce systemic effects. If no systemic effect could reasonably be expected, even if absorption was complete, further study of systemic absorption would not appear to be necessary. If the drug is toxic then the risk of systemic effects may be determined from a measure of the amount of percutaneous absorption from the prepara- tion applied. This may be assessed by established methods in which the amount absorbed is measured in the blood, urine, faeces or some particular organ (2). Radioactive tracer studies increase the sensitivity of such methods. When the compound has a characteristic pharmacological action this may be used for determining the amount of absorption. Thus, in the cat, the uptake of atropine from a suppository can be assessed by measuring the pupil dilatation. Oestrogens can be detected by their effect on the vaginal epithelium or the nipple, in the guineapig. An alternative approach is to measure the amount of the compound which disappears from the site of the application. To actually determine systemic toxicity following dermal application, the compound is applied to the skin and the animal observed for untoward effects. In general, the methods used resemble those utilized in the normal type of toxicity test by the oral and parenteral routes. It is the mode of administration which creates difficulties, making control of dosage and avoidance of oral contamination, a problem. For determination of acute effects, the substance is held in contact with the skin for periods up to 24 hours by means of a rubber sleeve. The pro- cedure and the dimensions of sleeves for different species of animals have been described by Draize (1). The doses applied are calculated on a body- weight basis, and they are introduced under the sleeve which covers approxi- mately 10 per cent of the body surface. By using groups of animals at ascending dose levels it is possible, with toxic substances, to determine the lethal dose (LD5o). Surviving animals are observed for ill effects over a subsequent period of at least two weeks, and this includes observations of their general health, growth and food consumption. The blood and urine are examined for abnormalities and animals showing ill effects are killed for pathological study. Tests for chronic effects are made by applying the substance daily to the same area of skin over a prolonged period of time. Observations are made