MARKOV CHAIN MONTE CARLO SIMULATION OF BIOMONITORING IN HUMANS: APPLICATION TO BIOMARKERS OF CHRONIC EXPOSURE TO ALKYL BENZENES IN THE ENVIRONMENT

Thomas Peyret and Kannan Krishnan

International Conference on Health Sciences Simulation (ICHSS 2008)
Crowne Plaza Ottawa Hotel, Ottawa, Canada, April 14-17, 2008

---

Summary

Bayesian approaches are relevant for characterizing the population distribution of pharmacokinetic determinants as well as the exposure biomarkers of chemicals in the environment. The objective of this study was to conduct Bayesian analysis of the blood and alveolar air concentrations of alkyl benzenes (toluene, m-xylene and ethylbenzene) in humans chronically exposed to these chemicals in air. At steady-state, the blood and alveolar concentrations of alkyl benzenes are influenced by alveolar ventilation rate (QP), blood: air partition coefficient (PB), liver blood flow (QL) and intrinsic clearance (CLint). The prior information on these input parameters was obtained from the literature. The mean and variability of steady-state blood concentrations observed in a human volunteer study (n=4) was used as a basis to create a distribution (normal) from which samples (n = 16 and n = 50) were drawn using Monte Carlo approach. After Markov Chain Monte Carlo (MCMC) simulation with n = 16 (trial 1) and n = 50 (trail 2), posterior estimates of model parameters were obtained. The second updating of model parameters (trial 2) did not have an impact on the outcome. In general, the calculated steady-state biomarker concentrations compared well with the individual and population values. Overall, this study has demonstrated the feasibility of conducting MCMC simulations of human biomonitoring data, particularly during data-poor situations.

---

  

START Conference Manager (V2.54.5)