Comment on “Steady State Solutions to PBPK Models and Their Applications to Risk Assessment I: Route to Route Extrapolation of Volatile Chemicals,” by Chiu and White in Risk Analysis, 26(3), 769–780

Abstract

Steady-state analyses of generic PBPK models for volatile organic chemical (VOC) exposure and risk assessment have been undertaken and applied for nearly two decades now. Chiu and White's paper on this subject adds little new to this earlier work. Their dismissive claim that ''Similar analyses have been done for specific chemicals and for inhalation'' is misleading, because some of this earlier work did indeed focus on ''generic'' PBPK models generally applicable to VOC exposure by multiple routes. In particular, the earliest of these previous studies developed steady-state solutions for generic PBPK models including respiratory and 1-compartment oral routes of exposure, and further specified how to add injection and dermal exposure routes. Chiu and White included a 2-compartment oral pathway and a lung compartment in an otherwise identical generic PBPK model, but did not consider other exposure pathways such as dermal uptake. Each of the earlier studies first presented a steady-state solution to a generic, multiroute PBPK model, and only then applied the generic solution to a problem or illustration involving a specific compound--i.e., the same approach used later by Chiu and White. For example, the earlier study included a simple, intuitive expression for low-dose metabolized fraction f*{sub m} of anymore » applied multiroute dose, allowing route-to-route extrapolation regardless of compound in low-dose contexts that typically are of interest in environmental VOC risk assessment. Section 2.2 of Chiu and White's paper (''Generalization to Time-Varying Exposures'') concludes that, under conditions of virtually linear metabolism, PBPK system ''solutions to steady-state exposures are directly applicable to intermittent exposures''--i.e., under such conditions, all steady-state system solutions (or output states) become valid when each dynamic input is replaced by its corresponding time-weighted average value. This conclusion, a well known axiom of linear systems theory, was stated explicitly to apply to f*{sub m} in an earlier study. A subsequent study addressed how generic steady-state PBPK solutions can be modified to estimate transient peak target-tissue concentrations at dynamic equilibrium, for dynamic exposure scenarios that involve exposure to a regular (e.g., daily) series of brief inputs by multiple pathways--an issue that may be importance for endpoints that have a cytotoxic mechanism of action.« less