Toxicokinetics as a key to the integrated toxicity risk assessment based primarily on non-animal approaches

► Toxicokinetics (TK) is a key element to integrate the results from in silico, in vitro and in vivo studies. ► TK is needed to estimate target organ doses expected from realistic human external exposure scenarios. ► TK is necessary for quantitative in vitro–in vivo extrapolation (IVIVE). ► Physiologically based toxicokinetic modelling (PBTK) is the most adequate approach to simulate human TK. ► PBTK models are mechanism-based, but high-quality in vitro and in silico data is necessary for their success. Toxicokinetics (TK) is the endpoint that informs about the penetration into and fate within the body of a toxic substance, including the possible emergence of metabolites. Traditionally, the data needed to understand those phenomena have been obtained in vivo. Currently, with a drive towards non-animal testing approaches, TK has been identified as a key element to integrate the results from in silico, in vitro and already available in vivo studies. TK is needed to estimate the range of target organ doses that can be expected from realistic human external exposure scenarios. This information is crucial for determining the dose/concentration range that should be used for in vitro testing. Vice versa, TK is necessary to convert the in vitro results, generated at tissue/cell or sub-cellular level, into dose response or potency information relating to the entire target organism, i.e. the human body (in vitro–in vivo extrapolation, IVIVE). Physiologically based toxicokinetic modelling (PBTK) is currently regarded as the most adequate approach to simulate human TK and extrapolate between in vitro and in vivo contexts. The fact that PBTK models are mechanism-based which allows them to be ‘generic’ to a certain extent (various extrapolations possible) has been critical for their success so far. The need for high-quality in vitro and in silico data on absorption, distribution, metabolism as well as excretion (ADME) as input for PBTK models to predict human dose–response curves is currently a bottleneck for integrative risk assessment.