RÉSUMÉ
The pharma industry designs increasingly less cytochrome P450 dependent and more metabolically stable drugs, and consequently UGT-metabolism becomes more frequently involved. This study compares 2 glucuronidation RAF-scaling approaches, product formation and substrate depletion, regarding their potential for prediction of in vivo DDI and the relative contribution of UGT-mediated phase II reactions in an industrial setting. RAFs were developed for UGT1A1, 1A3, 1A4, 1A6, 1A9, 2B7 and 2B15 recombinant UGT isoforms and a large 150-donor pooled human liver microsome batch. The RAF-values ranged from small values of 0.06 (UGT1A3), over 0.24 and 0.48 (UGT1A9 and UGT1A4), to values around 1 (1.11 for UGT2B7, 1.14 for UGT1A1), and high RAFs of 4.8 (UGT1A6) and 6.57 (UGT2B15). Both approaches identified the same primarily involved isoforms (≥75% relative contribution) of 5 clinical reference compounds (raloxifene, haloperidol, laropiprant, telmisartan and naloxone), in concordance with reported in vitro (R2 = 0.65) and clinical results for UGT1A1, 1A3, 1A4, 1A9, 2B7 and 2B15. This study is distinctive in that it is reporting the glucuronide formation in addition to substrate depletion. The product formation approach proved more sensitive and enables UGT phenotyping of slowly metabolized drugs, additionally it allows identification of structurally different glucuronides.