Modeling the Human Kinetic Adjustment Factor for Inhaled Volatile Organic Chemicals: Whole Population Approach versus Distinct Subpopulation Approach

Abstract

The objective of this study was to evaluate the impact of whole- and sub-population-related variabilities on the determination of the human kinetic adjustment factor (HKAF) used in risk assessment of inhaled volatile organic chemicals (VOCs). Monte Carlo simulations were applied to a steady-state algorithm to generate population distributions for blood concentrations (CAss) and rates of metabolism (RAMs) for inhalation exposures to benzene (BZ) and 1,4-dioxane (1,4-D). The simulated population consisted of various proportions of adults, elderly, children, neonates and pregnant women as per the Canadian demography. Subgroup-specific input parameters were obtained from the literature and P3M software. Under the “whole population” approach, the HKAF was computed as the ratio of the entire population's upper percentile value (99th, 95th) of dose metrics to the median value in either the entire population or the adult population. Under the “distinct subpopulation” approach, the upper percentile values in each subpopulation were considered, and the greatest resulting HKAF was retained. CAss-based HKAFs that considered the Canadian demography varied between 1.2 (BZ) and 2.8 (1,4-D). The “distinct subpopulation” CAss-based HKAF varied between 1.6 (BZ) and 8.5 (1,4-D). RAM-based HKAFs always remained below 1.6. Overall, this study evaluated for the first time the impact of underlying assumptions with respect to the interindividual variability considered (whole population or each subpopulation taken separately) when determining the HKAF.

Characterization of the human kinetic adjustment factor for the health risk assessment of environmental contaminants

“…< 1 year old) are . Impact of the conceptual model considered on HKAF calculation (Valcke et al, 2012). For both benzene and 1,4-dioxane's arterial blood concentration (CA ss ) and liver volume-adjusted rate of metabolism (RAM), the 95th or 99th percentile-based HKAF determined using the "distinct subpopulation approach" for adults (Ad), neonates (Neo), infants (Inf) and pregnant women (PW), were compared to the HKAF determined using the "whole population" approach (WP).…”

“…Approach followed by Valcke et al . () to reconstruct a distribution of internal dose metrics (here, arterial blood concentration (CA ss ) of 1,4‐dioxane) using steady‐state equations for a population of 100 000 individuals as per the Canadian demography (Statistics Canada, 2010). The number of iterations used in the Monte Carlo simulations for each subpopulation corresponded to the targeted number of individuals.…”

“…It is interesting to note that Valcke et al . () concluded that considering a different demographic profile did not modify their conclusions. These results confirmed that relying on the “distinct subpopulation” approach to compute the HKAF would be more health‐protective, as this approach is more inclusive of the sensitive individuals, particularly if they consist of a small proportion of the targeted population.…”

“…However, considering each subpopulation separately and comparing to the median individual of the entire population, the HKAFs varied between 2.2 and 2.5. More recently, Valcke et al (2012) used physiologically based steady-state algorithm to generate distributions of CA ss and RAM in a theoretical population of 100 000 individuals exposed chronically to low environmental levels of benzene and 1,4-dioxane by inhalation. The population distributions were reconstructed based on proportions of adults, elderly, children, neonates and pregnant women specified based on Canadian demography.…”

“…The consideration of the 99th or the 95th percentile value neither significantly affected the HKAF, although a greater difference was obtained in case of CA ss for 1,4-dioxane, the dose metric that exhibited the greatest overall population variability. However, the Valcke et al (2012) to reconstruct a distribution of internal dose metrics (here, arterial blood concentration (CA ss ) of 1,4-dioxane) using steady-state equations for a population of 100 000 individuals as per the Canadian demography (Statistics Canada, 2010). The number of iterations used in the Monte Carlo simulations for each subpopulation corresponded to the targeted number of individuals.…”