EVALUATION OF 10-PERCENT MINIMUM ELICITATION THRESHOLD FOR CR(VI)-INDUCED ALLERGIC CONTACT-DERMATITIS USING BENCHMARK DOSE METHODS

Historical patch test data have been used to propose health-based soil cleanup levels for Cr(VI) that are protective of eliciting allergic c ontact dermatitis (ACD) among previously sensitized individuals. Short comings regarding the use of these historical studies in the risk asse ssment of Cr(VI) have been identified and include the use of concentra tion as the dosimetric for ACD elicitation rather than the mass per su rface area. Information on the surface areas of the patches used by th e authors of three of the historical studies have been made available recently, and their dose levels have been converted from units of conc entration to mass per surface area. For this study, benchmark dose met hods were used to estimate the 10% minimum elicitation threshold (MET) based on the converted patch test data from these historical studies and from the data presented in a more recent patch test study by Nethe rcott et al. (1994). A truncated lognormal model was fitted to the his torical data from each individual historical patch test study, and to the data from the Nethercott et al. (1994) study using maximum likelih ood methods. The 10% MET from the Nethercott et al. (1994) study is se ven times lower than those from the historical studies. There are two primary reasons for this result First, Nethercott et al. used a 0.25% potassium dichromate patch to screen study participants, whereas the h istorical studies used patches with up to 0.5%. Hence, individuals who were less senstive and those who had irritant, rather than allergic r eactions at the high doses, were excluded. Second, Nethercott et al. u sed a TRUE-Test patch that is a more efficient and reliable allergen d elivery device than those used in the historical studies. Assuming 100 % bioavailability, the 10% MET from Nethercott et al. (1994) produces an ACD-based soil standard of 445 mg/kg compared with the ACD-based so il standards of 2,750 to 62,500 mg/kg calculated using the historical studies. The most recent patch study of Nethercott et al. (1994), whic h is based on modern patch testing methods and standardized diagnostic criteria, is the most scientifically appropriate for use in the risk assessment of Cr(VI) and produces the most conservative estimate of th e 10% MET.