Human urine and plasma concentrations of bisphenol A extrapolated from pharmacokinetics established in in vivo experiments with chimeric mice with humanized liver and semi-physiological pharmacokinetic modeling.

The aim of this study was to extrapolate to humans the pharmacokinetics of estrogen analog bisphenol A determined in chimeric mice transplanted with human hepatocytes. Higher plasma concentrations and urinary excretions of bisphenol A glucuronide (a primary metabolite of bisphenol A) were observed in chimeric mice than in control mice after oral administrations, presumably because of enterohepatic circulation of bisphenol A glucuronide in control mice. Bisphenol A glucuronidation was faster in mouse liver microsomes than in human liver microsomes. These findings suggest a predominantly urinary excretion route of bisphenol A glucuronide in chimeric mice with humanized liver. Reported human plasma and urine data for bisphenol A glucuronide after single oral administration of 0.1mg/kg bisphenol A were reasonably estimated using the current semi-physiological pharmacokinetic model extrapolated from humanized mice data using algometric scaling. The reported geometric mean urinary bisphenol A concentration in the U.S. population of 2.64µg/L underwent reverse dosimetry modeling with the current human semi-physiological pharmacokinetic model. This yielded an estimated exposure of 0.024µg/kg/day, which was less than the daily tolerable intake of bisphenol A (50µg/kg/day), implying little risk to humans. Semi-physiological pharmacokinetic modeling will likely prove useful for determining the species-dependent toxicological risk of bisphenol A.