Estimates of human interindividual differences in susceptibility to toxic insult may have their basis in pharmacokinetic (PK) and/or pharmacodynamic (PD) variations. Although the US EPA does consider separating uncertainty factors into PK and PD components, as done for perchlorate, there are no guidelines to inform such a process, and critical data are often absent. However, other agencies have separated the animal to human uncertainty factor (UFa) into PK and PD components. To quantify human interindividual differences and produce results which might be used to support data-derived uncertainty factors for human interindividual variability (UFh) with respect to the PK component, we have quantified and determined the distribution of several cytochrome P-450 (CYP) forms in a group of 141 human hepatic microsome samples. Mean and SD values (pmol CYP/mg microsomal protein) of CYP1A, CYP2E1 and CYP3A forms were 33.6 +/- 28.1, 59.4 +/- 18.2, and 141.9 +/- 104.5, respectively. Further, activity (pmol/min/pmol microsomal protein) towards substrates characteristic for a total of 7 CYP forms demonstrated standard deviations of 46 to 111% of mean values and normal distributions. Immunologically-detected CYP1A, CYP2E1 and CYP3A forms were also normally distributed. When data on enzyme selectivity for a given chemical exist, the specific activity of that enzyme can be combined with data describing the variability of enzyme expression to determine the extent of human interindividual variability in the intrinsic metabolism of the chemical. Variations in intrinsic metabolism can be included in physiologically based PK models as has been done for trichloroethylene (Toxicol Appl Pharmacol 152:376-387, 1998). This improves estimates of human PK variability by including physiologic constraints such as solubility in blood and delivery to the liver.
The Toxicologist. Society of Toxicology 39th Annual Meeting, March 19-23, 2000, Philadelphia, Pennsylvania