Evaluation of animal models for predicting skin penetration in man.

The human skin grafted athymic nude mouse, pig skin grafted athymic nude mouse, hairless dog, and weanling Yorkshire pig were evaluated as models for predicting skin penetration in man. Nine radiolabeled compounds previously tested on man were applied topically (4 micrograms/cm2) to each animal. These compounds included caffeine, benzoic acid, N,N-diethyl-m-toluamide, three steroids, and three insecticides. To correct for incomplete excretion of the label following topical absorption, percentage penetration was calculated by dividing the percentage of the topically applied radioactive dose recovered in the excreta by the corresponding percentage after parenteral administration and multiplication by 100. In the case of the grafted athymic nude mouse, calculated values of percentage penetration were confirmed because significant correlations (r = 0.78 for the human skin grafted athymic nude mouse and r = 0.97 for the pig skin grafted athymic nude mouse) were found between the calculated values and percentage penetration determined by summing radioactivity recovered in the urine, feces, tissues, and carcass. The results revealed a significant correlation between human skin grafted athymic nude mouse values and human values (r = 0.74, p = 0.05), and between weanling Yorkshire pig values and human values (r = 0.83, p = 0.05). In contrast, no significant correlation existed between human values and those of the hairless dog and pig skin grafted athymic nude mouse. The disposition of radioactivity following topical application of the radiolabeled nerve agent analog ( diisopropylfluorophosphonate ) and simulant (diethyl malonate) was determined in the weanling pig and the human skin grafted athymic nude mouse.(ABSTRACT TRUNCATED AT 250 WORDS)