SAFETY TESTING OF COSMETICS 143 who have been sensitized in the patch test procedure. If these reactive indi- viduals do not respond to use applications, it is most unlikely that sensitization would be induced by normal use of the product. However, this reasoning cannot be applied to products used for whole-body applications. Whole-body exposures by sensitized panelists cannot be safely done, and limited area ap- plications may be misleading. In addition to the value of the repeated insult patch test as a predictor of sensitizing potential, it furnishes valuable information on the primary irritant characteristics and cumulative primary irritant properties of test materials. The latter property is descriptively referred to frequently as "skin fatigue." In summary of our consideration of the evaluation of sensitizing potential, guinea pig sensitization studies are valuable in the detection of strong sensi- tizers. The closed patch procedure of Buehler (3) is much more sensitive than the intradermal injection procedure. Products showing definite evidence of sensitization in guinea pigs should be dropped from further consideration un- less they can be employed in concentrations significantly below the minimal sensitizing level for guinea pigs. Irrespectivc of negative results obtained in tests on guinea pigs, test mate- rials with no history of tolerance by humans should be initially tested in the repeated insult patch test procedure in a pilot study on 10-12 panelists. Full- scale repeated insult patch tests should be carried out on all products prior to release to market. In approximately 20 years of experience the present authors have never en- countered a severe incidence of sensitizing potential in full-scale 60-200 pan- elist tests which was not predicted by the pilot study. A pilot repeated insult patch test on 10-12 human panelists is of much more value than guinea pig tests as a screening test for finished products. The total number of human panelists which should be employed in tests should be governed to some extent by the type and extent of exposure to the product which will be encountered in use. For materials which will be em- ployed in nonocclusive uses, such as perfumes, hand and face lotions, etc., and will be applied to limited body areas, completely negative results in tests on 60 panelists have proved to be adequate assurance that such products will be essentially innocuous under use conditions. Products which are used under occlusive test conditions should be evaluated on larger numbers of panelists. It is also desirable to carry out controlled use studies of such products. These should be preceded by a repeated insult patch test on 60 panelists and should be designed so that one group of panelists use the product somewhat in excess of recommended use. Challenge patclaes of the test material should be applied at the end of the use period. The discussion of patch testing above has been restricted to tests designed primarily for the detection of sensitizing potential of formulations or ingre- dients. Many cosmetics produce no primary irritation when evaluated in the above procedures. In a continuous patch procedure presented by Lanman

144 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS (8), a total of as many as 8 samples can be applied to the back of each pan- elist. Six of these are test samples and two should be reference samples of known irritative potential. The test patch is a 1-in? Webril swatch moistened with 0.5 ml of liquid or 0.2 ml of semisolid samples. The swatch is covered with Blenderm tape. The patches are removed each day and the sites are scored for irritation. Fresh patches are applied immediately. This routine is followed each day until the first signs of erythema are seen or for a maximum of 21 days. As each sample produces erythema its application to the affected subject is discontinued. At the conclusion the IT o (number of days required to produce erythema in 50% of the subjects) is calculated for each sample. By this procedure minimal differences in irritant potential of essentially innocu- ous formulations can frequently be demonstrated. MICROBIAL STUDIES The importance of control of the microbial quality of formulations needs no emphasis as this is an area of which we all are acutely aware. In addition to the health hazard to consumers from contaminated products, the effect of mi- crobial growth on product quality is also an important factor. A comprehen- sive review of this problem was presented by Bruch (9) giving in some detail the procedures for evaluating microbial quality of cosmetics and the testing programs which should be employed to assure the manufacturers that their products are adequately preserved and that they are produced under GMP guidelines. Since it is possible that preservatives human skin, both animal toxicology and fore cosmetic products are released for testing program for cosmetic products, microbiological quality of raxv materials may act as sensitizers, or irritants to human tests should be conducted be- sale. In addition to a recommended it is highly desirable to monitor the and the cleanliness of the plant man- ufacturing facilities. This is done by the collection of swab samples from vari- ous areas of the plant, the processing equipment and containers in order to evaluate sanitizing procedures, and by sampling raw materials for microbio- logical content. It is of prime importance also to carefully inspect for other contamination possibfiities due to the general location and layout of the build- ing, and for air contaminants originating from plants in the nearby area. From information obtained by these procedures, proper action can be taken to cor- rect any factors causing contamination. The testing program we use is designed to determine the adequacy of pre- servatives present in new and current cosmetic products. Its objectives are: 1. To determine if products in current production are sufficiently biostatic to prevent the growth of microorganisms, and to determine for new products the levels of biostatic agents required to render such products suitably biostatic for extended periods of time. 2. To determine if the products in current production are free of microor- ganisms which are potentially human pathogens.