Human pharmacokinetic prediction of UDP-glucuronosyltransferase substrates with an animal scale-up approach.

The aim of the current study was to evaluate the accuracy of allometric scaling methods for drugs metabolized by UDP-glucuronosyltransferases (UGTs), such as ketoprofen, imipramine, lorazepam, levofloxacin, zidovudine, diclofenac, furosemide, raloxifene, gemfibrozil, mycophenolic acid, indomethacin, and telmisartan. Human plasma clearance (CL) predictions were conducted from preclinical in vivo data by using multiple-species allometry with the rule of exponents and single-species allometric scaling (SSS) of mice, rats, monkeys, or dogs. Distribution volume at a steady state (V(ss)) was predicted by multiple-species allometry or SSS of V(ss). Oral plasma clearance (CL(po)) was calculated under the assumption that F(a) × F(g) was equivalent across species. Each of the results was compared with the observed parameter calculated from the clinical data after intravenous or oral administration. Multiple-species allometry and SSS of mice, rats, and dogs resulted in a similar accuracy of CL and CL(po) predictions. Monkeys tended to provide the most accurate predictions of human CL and CL(po). The ability to predict the half-life, which was determined from CL and V(ss) predictions, was more accurate in SSS of rats and monkeys. The in vivo fraction metabolized by glucuronidation (f(m,UGT)) in bile duct-cannulated monkeys was relatively similar to that of humans compared with other animal species, which likely contributed to the highest accuracy of SSS prediction of monkeys. On the basis of the current results, monkeys would be more reliable than other animal species in predicting human pharmacokinetics and f(m,UGT) for drugs metabolized by UGTs.