Final Report of the Color Additive Scientific Review Panel

The Color Additives Scientific Review Panel considered whether there was information sufficient to perform a carcinogenic risk assessment on the colors D&C Red No. 19 (R‐19), D&C Red No. 37 (R‐37), D&C Orange No. 17 (O‐17), D&C Red No. 9 (R‐9), D&C Red No. 8 (R‐8) and FD&C Red No. 3 (R‐3) and to evaluate the assessments sent to FDA as part of the petitions for use of the colors for drug and external uses by the Cosmetic, Toiletry and Fragrance Association (CTFA). There is a lack of human data concerning the colors for making a human health assessment, so the assessments are based upon the extrapolation of animal data. The risk assessments are determined for exposure to single chemicals. Excluded from consideration are possible effects from exposure to multiple chemicals, such as cocarcinogenesis, promotion, synergism, antagonism, etc. In the light of recent efforts in establishing a consensus in risk assessment, the Panel has determined that the CTFA assessments for R‐10, O‐17, and R‐9 are consistent with present acceptable usages, although it questions some of the assumptions used in the assessments. The Panel identified a number of general assumptions made, and discusses their validity, their impact on total uncertainty, and the potential options to address the gaps in understanding that necessitate the assumption. The Panel also derived revised risk estimates using more “reasonable” assumptions than “worst‐case” situations, for 90th percentile and average exposure. For those assumptions that are easily quantifiable, the Panel's estimates are less than an order of magnitude lower than the CTFA risk estimates, indicating that the underestimates and overestimates of the CTFA risk estimates tend to balance each other. The impact of most of the assumptions is not quantifiable. The assessment for R‐3 is complicated by the fact that there is no good skin penetrance study for this color. It was assumed that the penetrance is similar to that of another water‐soluble xanthene color, R‐19. It is expected that the absorption of the color is not likely to exceed that of the smaller molecule, R‐19. Therefore, the risk estimates are similar to the CTFA estimates, but with different reasoning. The estimates for R‐8 and R‐37 are different from the others in that there is a lack of any exposure or toxicological information on these colors. The toxicological properties were considered to be similar to the chemically related colors R‐9 and R‐19, respectively. The risk calculated for these dyes is actually a unit risk (i.e., a risk based on particular unit exposure). The unit exposure used is the exposure estimated from the chemically related color, which, from the production values for the dyes, is likely to have much higher usage. Exposure to all of these colors from external and drug uses is quite low, especially as a result of exposure through skin. Therefore, the risk estimates are low. It is cautioned that usage information is based on a recent survey, and changes in usage will alter the risk estimate.