Absorption and disposition of poorly water soluble weak bases: Application to delavirdin

The steady-state kinetics of delavirdine (DLV) and its N-desisopropyl metabolite (MET) were evaluated in HIV-positive patients after escalating oral doses and after repeated oral administrations at the same dose level. In addition, hepatic CYP3A activity was evaluated via the erythromycin breath test (ERMBT). The results showed that DLV displayed nonlinear kinetics as indicated by the decreasing oral clearance, fluctuation (C$sbsp{max}{rm ss}$/C$sbsp{min}{rm ss}$) and terminal rate constant ($lambdasb{rm z}$), and by the increasing half-life at higher doses; the ratio of MET formation clearance to MET elimination clearance was also reduced in the escalating-dose group. In the control (same-dose) group, the kinetics of DLV and MET were unchanged. Hepatic CYP3A activity was markedly reduced after acute and chronic exposure to all doses of delavirdine mesylate. DLV could maximally inhibit 70-75% of predose ERMBT values with an IC50 of about 0.9 $mu$M. There was also a significant correlation between ERMBT and the logarithm of MET/DLV plasma concentration ratios using pooled data (r = 0.7908; p $<$ 0.0001). In conclusion, DLV is a potent and reversible inhibitor of hepatic CYP3A as well as a substrate for this CYP450 isoform. It is likely that delavirdine will exhibit drug-drug interactions when coadministered with other CYP3A substrates.;The relationship between gastric pH, solubility, permeability, and extent of absorption for delavirdine was evaluated using an integrated absorption model. Based on simulation results, it was shown that the large bioavailability variability between patients is dependent on the difference in drug solubility which is, in part, due to bile salt concentrations. Elevation in gastric pH resulted in reduced fraction of dose absorbed, but the effect may be limited. Further, an increase in AUC can be intensified due to CYP3A4 inhibition and dose-dependent kinetics (CL$downarrow$, F$uparrow$). For delavirdine mesylate, precipitation had a minor effect on fraction of dose absorbed in the presence of sodium taurocholate at simulated fasted and fed states. As bile salt concentration approaches zero, the extent of precipitation should increase based on the calculated precipitate rate constant and simulation results. Consequently, the fraction of dose absorbed may reduce. This approach identified important in vivo factors for oral absorption and can be used to predict as well as explain clinical data for poorly water-soluble weak bases.