An assessment of the impact of multi-route co-exposures on human variability in toxicokinetics: A case study with binary and quaternary mixtures of volatile drinking water contaminants.

This study aimed to assess the impact of multi-route co-exposures to chemicals on interindividual variability in toxicokinetics. Probabilistic physiologically based pharmacokinetic multi-route interaction models were developed for adults and four younger subpopulations. Drinking water-mediated multi-route exposures were simulated for benzene alone or in co-exposure with toluene, ethylbenzene and m-xylene, for trichloroethylene or vinyl chloride (VC), alone and in mixture. These simulations were performed for "low" and "high" exposure scenarios, involving respectively the US EPA's short-term drinking water health advisories, and 10 times these advisory values. Distributions of relevant internal dose metrics for benzene, trichloroethylene and VC were obtained using Monte Carlo simulations. Intergroup variability indexes (VI) were computed for the "low" (VIL ) and "high" (VIH ) exposure scenarios, as the ratio between the 95th percentile in each subpopulation over the median in adults. Thus, for benzene, parent compound's area under the curve-based VIL for single exposures vs. co-exposures correspondingly varied between 1.7 (teenagers) and 2.8 (infants) vs. 1.9 and 3.1 respectively. VIH varied between 2.5 and 3.5 vs. 2.9 and 4.1. Inversely, VIL and VIH for the amount of benzene metabolized via CYP2E1 pathway decreased in co-exposure compared to single exposure. For VC and trichloroethylene, similar results were obtained for the "high" exposure, but "low" co-exposures did not impact the toxicokinetics of individual substances. In conclusion, multi-route co-exposures can have an impact on the toxicokinetics of individual substances, but to an extent, that does not seem to challenge the default values attributed to the factors deemed at reflecting interindividual or child/adult differences in toxicokinetics.