Physiologically‐based pharmacokinetic modelling of uridine 5′‐diphosphoglucorosultransferase (UGT) substrate drugs in pregnant women

AIMS: While pregnancy-related changes in phase I enzyme activity are well-documented, less is known about the impact on phase II enzymes. This study aimed to test the hypothesis that changes in the pharmacokinetics (PK) of uridine 5'-diphosphoglucuronosyltransferase (UGT) substrates during pregnancy result from altered enzyme expression or activity. Physiologically based PK (PBPK) modelling was used to evaluate these changes in pregnant populations. METHODS: Induction of UGT1A1 and UGT1A4 in pregnancy was modelled using hormone-based equations, which account for the effects of 17β-oestradiol and progesterone. UGT1A3, UGT2B7 and UGT2B15 were assumed unchanged due to a lack of hormone-induced expression. The predicted-to-observed ratio is commonly used to assess the accuracy of PBPK models in predicting clinical data. PBPK model predictions are typically considered acceptable when they fall within a 1.5-fold error, defined as the ratio of the predicted to the observed PK parameter being within a 1.5-fold range. RESULTS: Overall, 82% of predicted to observed ratio of pregnant population were within 1.5-fold error. P:NP ratios were as follows: UGT1A1: 88%, UGT1A4: 50%, UGT1A9: 100%, UGT2B7: 100%, and UGT2B15: 25%. The P:NP ratio indicates that the PK changes in the pregnant population were successfully predicted for UGT1A1, UGT1A4, UGT1A9 and UGT2B7 substrates. CONCLUSIONS: p-PBPK models that incorporated pregnancy-induced changes in UGT1A1, UGT1A4 and UGT1A9 expression/activity, or assumed no change for UGT2B7, accurately predicted the pharmacokinetics of most UGT-metabolized drugs. These findings enhance our understanding of phase II metabolism during pregnancy and may support dose adjustment for those drugs in the future.