Caco-2 monolayers (in vitro), rat intestinal sheets mounted in modified Ussing Chambers (ex vivo), and in situ intestinal perfusion of rat ileum were used as models to determine and compare the absorption characteristics of the antiviral agent 9-[2-(phosphonomethoxy)ethyl]adenine (PMEA, adefovir) and its bis(pivaloyloxymethyl)-ester prodrug [bis(POM)-PMEA, adefovir dipivoxil]. Although metabolism of adefovir dipivoxil was more pronounced in the ex vivo and in situ models than in the Caco-2 system, the transport of 'total adefovir' [= adefovir dipivoxil and its metabolites mono(POM)-PMEA and adefovir] was comparable in the three models. Compared with transport of the parent compound (adefovir), use of adefovir dipivoxil resulted in a significant increase in transport of total adefovir in the in vitro ( approximately 100-fold) and the in situ ( approximately 10-fold) models; in contrast, the ex vivo method failed to demonstrate a remarkable transport enhancement when using the ester prodrug. Similar to the results obtained in the Caco-2 model, the inclusion of the P-glycoprotein inhibitor verapamil resulted in transport enhancement during in situ perfusion of rat ileum with adefovir dipivoxil; however, no effect of verapamil could be observed in the ex vivo model. The results of this study confirm the utility of both the in vitro and in situ models to assess intestinal transport and metabolism of adefovir dipivoxil. The ex vivo model appeared to be less appropriate because of its inability to discriminate transport following administration of adefovir or adefovir dipivoxil and because of the absence of an effect of verapamil on transport when using adefovir dipivoxil.
Copyright 2000 Wiley-Liss, Inc.