Variability of biomarkers in volunteer studies: The biological component

Biological monitoring has become one of the methods to measure exposure, with the advantage that it gives information about the concentration of a substance that actually enters the body and reflects the inter-individual differences in uptake and metabolic variation. However, limited information is available on inter- and intra-individual variability of biomarkers. The aim of this study was to gather information about the biological component of inter-individual variation in biomarkers using results from volunteer studies. Open literature and other (internal) sources were searched to find human volunteer studies utilizing biological monitoring. Ultimately 41 studies were included in our analysis, with a total of 6747 observations for one or more biomarkers from 223 volunteers. The data from these studies were grouped on the basis of study, substance under investigation, exposure route, biological matrix, exposure duration, dose and number of exposure events to obtain 278 homogeneous groups (strata) for statistical analysis. Variability was assessed in two ways. Firstly, estimates of biomarker half-life were calculated for each individual, thereby allowing the estimation of inter-individual variability in half-lives within the homogeneous groups. Secondly, variation in biomarker concentrations at a given time point was estimated. For estimated half-lives the GSDs ranged from 1.0 to 6.8. The variability in estimated half-lives did not differ much for the different types of substances. For concentrations at a given time point the average GSDs within strata ranged from 1.0 to 5.6. Again, variability did not differ much for different groups (e.g., type of substance). The median variability component was 0.11 (range 0-3.0). In conclusion, volunteer studies enable the estimation of both variation in half-lives and variation in biomarker levels in the well-defined homogeneous groups. Comparison of our results with other studies indicates that variation due to biological differences within and between people is quite substantial in homogeneous exposure groups. The relative contribution of this biological component to the total variation will be smaller when variance components are estimated in less homogeneous groups, such as those in occupational and environmental settings.