Background: Chromium from natural and anthropogenic sources is present in drinking water in trivalent [Cr(III)] and hexavalent [Cr(VI)] forms. The current U.S. maximum contaminant level (MCL) is 0.1 ppm total chromium, which is based on the absence of toxicity in Sprague-Dawley rats exposed to 25 ppm Cr(VI) in drinking water for one year (Mackenzie et al., 1958). The MCL is currently being updated by the U.S. EPA to incorporate more recently released data, including those produced by the National Toxicology Program (NTP). These data include two-year cancer bioassays reporting that chronic exposure to ≥30 ppm Cr(VI) increased small bowel tumors (primarily in the duodenum) in B6C3F1 mice, and that exposure to 180 ppm increased oral mucosa tumors in F344 rats (NTP, 2008). The mode of action (MOA) underlying these tumorigenic responses is therefore important to elucidate and incorporate into risk assessment applications.

Objectives: Review in vivo toxicity and MOA data related to the oral tumors in F344 rats, and determine whether non-linear risk assessment approaches are scientifically justifiable if these tumors were selected as the basis for setting drinking water standards for Cr(VI).