J. Cosmet. Sci., 60, 587–598 (November/December 2009) 587 Addressing technical challenges associated with the FDA’s proposed rules for the UVA in vitro testing procedure OLGA V. DUEVA-KOGANOV, COLLEEN ROCAFORT, STEVEN OROFINO, ULI OSTERWALDER, and JUAN BRITO, Ciba Corporation, Tarrytown, NY (O.V.D.-K., C.R., S.O., J.B.), and Basel, Switzerland (U.O). Accepted for publication June 15, 2009. Presented at the Annual Scientifi c Seminar of the Society of Cosmetic Chemists, New York, December 11, 2008. Synopsis The proposed rules of the Food and Drug Administration (FDA) for the UVA in vitro testing procedure were applied to the evaluation of various sunscreen formulations and the following technical challenges were ob- served: when proposed roughened quartz substrates were used, the required coeffi cient-of-variation criteria were not met, and the dynamic ranges of the available transmittance analyzers were exceeded for sunscreens with high SPF values. In the proposed rules, the FDA requested comment regarding the suitability of other possible substrates. In this research, two modifi cations to the FDA’s proposed rules were evaluated: (a) the use of an alternative substrate, Vitro Skin® N-19 (IMS, Inc.) instead of roughened quartz substrate and (b) an increase in application time from 10 seconds to 30 seconds to ensure a uniform distribution of sunscreen product over the application area of the substrate. These two modifi cations allowed meeting the required coeffi cient-of-variation criteria without exceeding the dynamic ranges of the available transmittance analyz- ers. The modifi ed test conditions were utilized for the evaluation of six commercial sunscreens, which fulfi lled criteria of “medium” or “high” categories—based on their UVAI/UV ratios. These fi ndings were in agree- ment with the statement in the proposed rules that the FDA is aware of the diffi culty for current sunscreen formulations to meet the “highest” category and believes that allowing such a category will foster additional research and development in this area. To determine if it was possible to achieve a UVA rating greater than 0.95, two experimental sunscreen prototypes with bisoctrizole (USAN), bemotrizinol (USAN), avobenzone, and octocrylene were tested under the modifi ed test conditions and attained the “highest” category. It should be noted that bisoctrizole and bemotrizinol are being evaluated by the FDA under TEA and are not permit- ted in the US at this time, but they are approved for use in the rest of the world. INTRODUCTION On August 27, 2007 the FDA released the long-awaited proposed rules on UVA protection (1), which offer a comprehensive evaluation of sunscreen product effi cacy in vivo (SPF and UVA-PF) and in vitro (UVAI/UV ratio). The FDA requires using both in vivo and in vitro UVA radiation testing methods to ensure that the magnitude and breadth of UVA protection is determined. The FDA asks US sunscreen manufacturers to declare, in addition to the SPF, the level of UVA protection, expressed in fi ve categories: “No UVA protection,” “Low,”

JOURNAL OF COSMETIC SCIENCE 588 “Medium,” “High,” and “Highest.” The FDA believes that increasing protection from UVA radiation is benefi cial for consumers’ health because scientifi c data demonstrates that UVB and UVA radiation protection is equally important for the skin. The FDA’s vision is to re- ward balanced and photostable UVB/UVA protection. The purposes of this research were: (a) to utilize the FDA’s proposed rules for the UVA in vitro testing procedure in the evaluation of various sunscreen formulations and to address the discovered technical challenges by mod- ifying this procedure (b) to evaluate various sunscreens according to the modifi ed procedure and (3) to determine if it was possible to attain a UVAI/UV ratio greater than 0.95. METHOD AND MATERIALS PRE-IRRADIATION DOSE The FDA proposed to specify a pre-irradiation dose (PID) in terms of “erythemal effective dose” because all solar simulators used by the industry may not have the exact fi lter com- bination and spectral transmittance of fi lters can vary. This dose was calculated by weigh- ing the output spectrum of the solar simulator with the reference action spectrum for erythema as defi ned by CIE according to the formula: 2 PID (J/m -eff) = SPF * 1 MED * 2/3 where 1 MED = 200 J/m2-eff. Pre-irradiation criteria were fulfi lled utilizing a Weather- O-Meter Ci65A with a Right LightTM inner/quartz outer fi lter combination recently in- troduced by Atlas (2), which provides an excellent match to natural sunlight, especially in the UV region (Figure 1). When the Ci65A spectral power distributions are normalized to 1.57 W/m2 at 420 nm, the 200 J/m2-erythemal effective dose (1 MED) is reached within ten minutes. Alterna- tively, this parameter can be normalized to 1.1 W/m2 at 420 nm in order to reach 1 MED within 14 minutes. Figure 1 shows that when the irradiation fl ux is normalized to 1.1 W/m2 Figure 1. Spectral power distributions of Ci65A with Right LightTM inner/quartz outer fi lters normalized to 1.57 W/m2 and 1.1 W/m2 at 420 nm. COLIPA Sun and Miami Peak Sun 26 S Direct.