Implications of dose-dependent target tissue absorption for linear and non-linear/threshold approaches in development of a cancer-based oral toxicity factor for hexavalent chromium

- Mouse target tissue absorption per unit dose decreases appreciably with dose.
- It is about 10-fold lower at the dose at the MCL than at the draft SFo POD dose.
- This toxicokinetic phenomenon is inconsistent with linear low-dose extrapolation.
- The dose-response curve accounting for this would clearly be non-linear.
- A non-linear approach is most defensible for a cancer-based oral toxicity factor.

Dose-dependent changes in target tissue absorption have important implications for determining the most defensible approach for developing a cancer-based oral toxicity factor for hexavalent chromium (CrVI). For example, mouse target tissue absorption per unit dose is an estimated 10-fold lower at the CrVI dose corresponding to the federal maximum contaminant level (MCL) than at the USEPA draft oral slope factor (SFo) point of departure dose. This decreasing target tissue absorption as doses decrease to lower, more environmentally-relevant doses is inconsistent with linear low-dose extrapolation. The shape of the dose-response curve accounting for this toxicokinetic phenomenon would clearly be non-linear. Furthermore, these dose-dependent differences in absorption indicate that the magnitude of risk overestimation by a linear low-dose extrapolation approach (e.g., SFo) increases and is likely to span one or perhaps more orders of magnitude as it is used to predict risk at progressively lower, more environmentally-relevant doses. An additional apparent implication is that no single SFo can reliably predict risk across potential environmental doses (e.g., doses corresponding to water concentrations ⩽ the federal MCL). A non-linear approach, consistent with available mode of action data, is most scientifically defensible for derivation of an oral toxicity factor for CrVI-induced carcinogenesis.