Accurate prediction of variability in CLint and Fm via 3A4 is only obtained by assessing a series of individual cryopreserved human hepatocyte batches

• Multiple in-vitro and in-vivo methods are currently assessed and under discussion to predict human clearance and pharmacokinetics from preclinical studies. A combination of high fm with a high fraction metabolism via a single pathway, e.g. via CYP3A4 (fmCYP3A4), has been recognized as a high risk factor for Drug Drug Interactions (DDI) in the clinical setting[1],[2],[3],[4] . Thus, an early predictive tool to allow for appropriate modeling of this potential risk for DDI is highly warranted. Hepatocytes, capable of both phase I and phase II reactions, are an attractive system to study fraction metabolized (fm) via a single pathway. In the present study, intrinsic clearance (CLint) was determined in cryopreserved human hepatocytes in suspension for a set of five compounds with known and variable fm via CYP3A4 (amitriptyline, loratadine, methylprednisolone, midazolam, and tacrolimus) in the absence or presence of ketoconazole. In order to get an insight into the influence of inter-individual variability, twelve batches of cryopreserved human hepatocytes with either high, moderate or low CYP3A4-dependent activity towards midazolam (MDZ) were chosen. Clint values were determined as substrate depletion under shaking conditions (900rpm) using an elliptic shaker as previously reported[5]. For all compounds, the mean CLint for individual donors in absence of ketoconazole correlated very well with literature data on the mean of individual donors1,2,3, and/or pools of donors5. Average fmCYP3A4 for midazolam was 83%, tacrolimus 64%, methylprednisolone 55%, amitriptyline 28%, and loratadine 19% are also well within the literature data2,3,4. Interestingly, the results obtained for a homogenous subpopulation regarding MDZ CLint and percent inhibition by ketoconazole, were not directly related to the ketoconazole sensitive CLint for the other CYP3A4 substrates tested. The variability in CYP3A4 contribution for compounds having multiple metabolic pathways cannot be predicted by the fm3A4 for MDZ. This suggests that an overall prediction of CLint or fm via CYP3A4 for compounds partially metabolized by this enzyme is not possible. Thus, the individual differences in CLint for a given compound and fmCYPi can only be well covered by assessing a series of individual cryopreserved human hepatocyte batches.[1] Lu, C., Miwa, G. T., Prakash, S. R., Gan, L-S. and Balani, S. K. (2007), Drug Metabolism And Disposition, 35: 1, 79–85[2] Lu, C., Hatsis,P., Berg,C., Lee, F. W. and Balani, S. K. (2008), Drug Metabolism And Disposition, 36: 7, 1255–1260[3] Lu, C., Hatsis,P., Berg,C., Lee, F. W. and Balani, S. K. (2008), Drug Metabolism And Disposition, 36: 7, 1261–1266[4] Emoto, C., Murase, S. and Iwasaki, K.(2006), Xenobiotica, 36: 8, 671 — 683[5] Simon ,S., Blanchard, N., Alexandre, E., Hewitt, N. J., Bachellier, P., Heyd, B., Coassolo, P., Schuler, F., Richert, L., (2009) ‘MV-HUF Copenhagen’

Publication and Content Type

ref (subject category)

art (subject category)