RESUMO
The most relevant data for developing a reference dose (RfD) for perchlorate exposures comes from human epidemiology and clinical studies, supplemented with available and extensive information on experimental animals. Specifically, serum T4 decrease is the critical effect of perchlorate, based on a mode-of-action analysis and the evidence provided by the body of rodent studies on perchlorate. However, no T4 decreases have been observed in human populations following perchlorate exposure at non-therapeutic doses. An RfD of 0.002 mg/kg-day can be derived using an epidemiology study. A freestanding NOAEL of 0.006 mg/kg-day for T4 decrease was identified in children from the epidemiology study. The use of this NOAEL has the advantage of a being identified in a sensitive subgroup, neonates and children. Data are sufficient to estimate an overall uncertainty factor of 3-fold with this NOAEL based on expected differences in toxicokinetics and toxicodynamics between children, and pregnant women and their fetuses, the second identified sensitive subgroup for perchlorate, and concerns about the over-iodination of this population. This RfD is supported by a human clinical study using inhibition of iodine uptake in adults as a measurable surrogate for the critical effect of T4 decrease in humans. However, although this latter study has a well-established dose-response curve for inhibition of iodine uptake, even perchlorate doses that result in a 70% inhibition of iodine uptake have no apparent effect on human T4 levels. Thus, the use of this study as the primary basis of the RfD is problematic. Nevertheless, a benchmark dose of 0.01 mg/kg-day was identified in this clinical study, which supports a threshold value of 0.006 mg/kg-day identified by its authors and the RfD of 0.002 mg/kg-day estimated in this paper.