In vitro vs. canine data for assessing early exposure of doxazosin base and its mesylate salt

In this study, we evaluated the usefulness of biorelevant in vitro data and of canine data in forecasting early exposure after the administration of two phases of a BCS Class II compound, i.e., doxazosin base (DB) and its mesylate salt (DM). DB, DM, and doxazosin hydrochloride (DH) were prepared and characterized. In vitro data were collected in various media, including human aspirates. Solubilities of DB and DM in human gastric fluid were forecasted by data in fasted state simulating gastric fluid containing physiological components (FaSSGF-V2) but not by data in HCl(pH 1.8). Unlike data in FaSSGF-V2, dissolution of DB and DM tablets in HCl(pH 1.6) is rapid. Dissolution of DB tablet in FaSSGF-V2 is incomplete and conversion to DH seems to occur. Differences between DB and DM in dissolution in the small intestine are overestimated in the absence of physiological solubilizers. Using the in vitro data and previously described modeling procedures, the cumulative doxazosin profile in plasma was simulated and the 0-2h profile was used for evaluating early exposure. Individual cumulative doxazosin profiles in plasma, after single DM tablet administrations to 24 adults, were constructed from corresponding actual plasma profiles. Compared with in vitro DM data in pure aqueous buffers, DM data in biorelevant media led to better prediction of early exposure. Based on intersubject variability in early exposure after DM administration and simulated profiles, the administered phase, DB or DM, does not have a significant impact on early exposure. Partial AUCs were used for evaluating early exposure after DB and DM administration in 4 dogs. Early exposure was significantly higher after administration of DM to dogs. Dogs are not appropriate for evaluating differences in early exposure after DB and DM administrations.