A chronic oral reference dose for hexavalent chromium‐induced intestinal cancer

Abstract

High concentrations of hexavalent chromium [Cr(VI)] in drinking water induce villous cytotoxicity and compensatory crypt hyperplasia in the small intestines of mice (but not rats). Lifetime exposure to such cytotoxic concentrations increases intestinal neoplasms in mice, suggesting that the mode of action for Cr(VI)-induced intestinal tumors involves chronic wounding and compensatory cell proliferation of the intestine. Therefore, we developed a chronic oral reference dose (RfD) designed to be protective of intestinal damage and thus intestinal cancer. A physiologically based pharmacokinetic model for chromium in mice was used to estimate the amount of Cr(VI) entering each intestinal tissue section (duodenum, jejunum and ileum) from the lumen per day (normalized to intestinal tissue weight). These internal dose metrics, together with corresponding incidences for diffuse hyperplasia, were used to derive points of departure using benchmark dose modeling and constrained nonlinear regression. Both modeling techniques resulted in similar points of departure, which were subsequently converted to human equivalent doses using a human physiologically based pharmacokinetic model. Applying appropriate uncertainty factors, an RfD of 0.006 mg kg–1 day–1 was derived for diffuse hyperplasia—an effect that precedes tumor formation. This RfD is protective of both noncancer and cancer effects in the small intestine and corresponds to a safe drinking water equivalent level of 210 µg l–1. This concentration is higher than the current federal maximum contaminant level for total Cr (100 µg l–1) and well above levels of Cr(VI) in US drinking water supplies (typically ≤ 5 µg l–1). © 2013 The Authors. Journal of Applied Toxicology published by John Wiley & Sons, Ltd.

Integration of mechanistic and pharmacokinetic information to derive oral reference dose and margin‐of‐exposure values for hexavalent chromium

“…Including these data would shift the dose–response curve rightward, thereby resulting in a less conservative BMDL (see Thompson et al, 2014). Figure 4(A) shows the best fitting model of the combined male and female duodenum and ileum data.…”

“…The main effect of sex and the dose × sex interaction effect was assessed for relevant sections. Although this was done in our previous assessment (Thompson et al, 2014), changes to the internal dose metric and the change in diffuse epithelial hyperplasia incidence (see Section 2.2.3) warrants re‐examination.…”

“…Hyperplasia data from the jejunum were omitted for dose–response modeling, for reasons described previously (Thompson et al, 2014). In brief, hyperplasia incidence in the NTP (2008) bioassay was assessed microscopically via a single 5 μm biopsy taken at the approximate midpoint of each intestinal segment.…”

“…For all endpoints, pyloric flux of Cr(VI) was used to support interspecies extrapolations (i.e., determination of HEDs), because pyloric flux can be estimated in humans with much higher confidence than is possible for the other internal dose measures, as discussed in Thompson et al (2014). The PBPK model was used to support the following extrapolations to facilitate HED predictions for the point of departure (POD) values expressed in terms of SI sectional flux or portal flux (dose measures in brackets reflect those used to support interspecies extrapolation):

Mouse SI sectional flux → mouse SI total flux → [mouse pyloric flux = human pyloric flux] → HED

Mouse/rat portal flux → [mouse/rat pyloric flux = human pyloric flux] → HED

…”

“…We previously developed an RfD for Cr(VI) that focused specifically on intestinal effects (Thompson et al, 2014). Since that original publication, additional MOA information has been published that further supports threshold mechanisms for the tumors observed in mice and rats.…”