Concentration-guided strategies in drug development: experience with a cyclosporine analog in transplantation

Optimizing the absorption of valspodar, a P-glycoprotein modulator, Part I: Selecting an oral formulation and exploring its clinical pharmacokinetics/dynamics

Valspodar is a cyclosporine D analog used as a chemotherapy adjunct for modifying multidrug resistance. Two studies were sequentially performed to select an optimal oral formulation and to characterize selected aspects of its clinical pharmacokinetics/dynamics. An initial four-way crossover study with 20 volunteers compared the pharmacokinetics of single fasting administrations of 200 mg by intravenous infusion and 600 mg orally as a conventional oral solution, a microemulsion oral solution, and a microemulsion soft gelatin capsule. The two microemulsion dosage forms demonstrated significantly faster and less variable rates of absorption compared with the conventional oral solution. The microemulsion dosage forms were bioequivalent with absolute bioavailability nearly double that of the conventional oral solution. Based on these results, the microemulsion capsule was further investigated in a four-way randomized crossover study with 24 volunteers who received single fasting administrations of 200, 400, and 600 mg and a 400-mg administration after a fat-rich meal. Dose proportionality in area under the curve (AUC) was demonstrated over this dose range. Administration after a fat-rich meal caused a slight time lag until absorption began, a delay in time to reach the peak concentration, and a moderate increase of 24% in AUC. Serial determinations of total bilirubin were explored as a potential pharmacodynamic marker for P-glycoprotein inhibition. A similar magnitude of reversible hyperbilirubinemia was seen at all dose levels suggesting that P-glycoprotein inhibition in the biliary canaliculi was maximal even at the lowest dose tested. The microemulsion formulation (oral solution or soft gelatin capsule) represents an improved and less variable oral delivery form providing dose-proportional drug exposure over a clinically relevant dose range.

Optimizing the absorption of valspodar, a P-glycoprotein modulator, Part II: Quantifying its pharmacokinetic variability and refining the bioavailability estimate

A randomized, sequential, bioreplication study was performed with 24 healthy volunteers to assess the pharmacokinetic variability of oral and intravenous administrations of valspodar, a multidrug resistance modulator for use as a chemotherapy adjunct. Subjects received 200 mg as a 2-hour intravenous infusion and 400 mg orally as microemulsion soft gelatin capsules each given on two separate occasions. Following replicate intravenous administrations, reproducibility in peak concentration and area-under-the-curve was demonstrated by interoccasion equivalence testing. Low to moderate inter- and intra-subject pharmacokinetic variability were observed with coefficients of variation ranged from 9% to 19%. Absolute bioavailability from the microemulsion formulation was 60%. Replicate oral administrations demonstrated uniform and repeatable absorption with interoccasion equivalence in peak concentration, time to reach the peak concentration, and area-under-the-curve. Coefficients of inter- and intra-subject variation for these parameters ranged from 10% to 20% indicating low to moderate pharmacokinetic variability. When variabilities were compared between the two routes of administration, the absence of differences indicated that the rate and extent of drug absorption from the microemulsion formulation was as uniform and repeatable as by intravenous infusion.

Comparative pharmacokinetics and renal effects of cyclosporin A and cyclosporin G in renal allograft recipients

Cyclosporin G (CSG) has produced less nephrotoxicity than cyclosporin A (CSA) at equivalent doses in animal models. Conflicting results have been reported concerning differences in the pharmacokinetics of CSA and CSG in preclinical studies, and no data exist regarding the effect of steady-state oral administration of CSG on renal function in transplant patients or CSG-induced release of endothelin and nitric oxide (NO) in vivo. The objective of the study was to examine steady-state pharmacokinetic profiles of adult renal allograft recipients receiving CSA and CSG in relation to concentrations of endothelin-1 and NO2/NO3 in urine and plasma, creatinine clearance (Clcr), and urinary excretion of N-acetyl-beta-D-glucosaminidase (NAG) 9 months after transplantation. Concentrations of CSA and CSG were measured in whole blood over a 12-hour dose interval by both a monoclonal and polyclonal fluorescence polarization radioimmunoassay for CSA. A metabolite fraction was defined as the numerical difference between the levels obtained at each time point by both assays. Patient groups were defined as follows: group 1: initial CSA (n = 6); group 2: initial CSG (n = 7); group 3: five of the seven patients in group 2 taking CSG subsequently undergoing conversion to CSA; group 4: the same five patients in group 3 restudied 1 month after 1:1 dosage conversion to CSA; and group 5: CSA groups 1 and 4 combined (n = 11). In group 1, the metabolite fraction accounted for 32% to 54% of the total measurable drug concentration at each time point, whereas in group 2, the metabolite fraction accounted for at most 10% to 15% of the total drug levels measurable by polyclonal fluorescence polarization radioimmunoassay. Although there were no significant differences in any of the mean pharmacokinetic parameters between groups using monoclonal fluorescence polarization radioimmunoassay, the normalized area under the concentration-time curve (NAUC) value was less in four of five patients after conversion from CSG to CSA, with a more variable and delayed time to reach peak concentration (tmax) but equivalent apparent oral clearance (Clpa) values. Clcr was found to change significantly with time in groups 1 and 5 but not in group 2, with CSA producing a more profound and sustained decrease than CSG. Endothelin-1 and NO2/NO3 levels in plasma and urine remained relatively constant after administration of both CSA and CSG, and there were no significant differences between groups 3 and 4 regarding mean endothelin-1 and NO2/NO3 concentrations in plasma, urinary release of endothelin-1 and NO2/NO3, and mean AUC of endothelin-1 and AUC of NO2/NO3. However, monoclonal NAUC correlated significantly with total urinary endothelin-1 within CSA groups 1 and 5 but not within CSG group 2. Metabolite NAUC correlated significantly with total urinary NAG within CSA group 1. Although limited by the small number of patients, this study suggests that 1) CSG may produce less of a reduction in Clcr over time after oral administration at steady state than does CSA, and 2) this beneficial effect of CSG may be in part due to decreased intrarenal release of endothelin-1, as urinary excretion of endothelin-1 seemed to correlate better with CSA than with CSG exposure.