Improved dose linearity of cyclosporine pharmacokinetics from a microemulsion formulation.

The pharmacokinetic dose proportionality and relative bioavailability of cyclosporine from a microemulsion formulation (Sandimmune Neoral) were compared to those of the commercial formulation (Sandimmune) over the dosage range 200 to 800 mg. Single oral administrations were given as soft gelatin capsules in an open randomized study with 48 healthy volunteers. Whole-blood cyclosporine concentrations were determined by a specific monoclonal radioimmunoassay. In comparison to Sandimmune, the absorption rate (maximum concentration) and systemic availability (area under the curve) of cyclosporine were greater for Sandimmune Neoral at all dose levels investigated. The area under the curve for Sandimmune increased in a less than proportional manner with respect to dose, whereas that for Sandimmune Neoral was consistent with linear pharmacokinetics. Because of this difference, no global assessment of relative bioavailability could be performed. The relative bioavailability of cyclosporine from Sandimmune Neoral ranged from 174 to 239% compared to Sandimmune, depending on the dose level. The improvements in oral bioavailability and dose linearity of cyclosporine exposure after administration as Sandimmune Neoral should facilitate more accurate dosage titration in the clinical setting.

Influence of a fat-rich meal on the pharmacokinetics of a new oral formulation of cyclosporine in a crossover comparison with the market formulation.

The influence of a fat-rich meal on the pharmacokinetics of cyclosporine from a new oral formulation (Sandimmune Neoral) was compared in a randomized, four-way crossover study to the currently marketed formulation (Sandimmune) in 24 healthy male volunteers. Single oral doses of 300 mg Sandimmune and 180 mg Sandimmune Neoral were each administered once under fasting conditions and once 30 min after starting a high-fat meal. Serial blood samples were obtained over a 48-hr period after each administration, and whole-blood cyclosporine concentrations were determined by a specific monoclonal radioimmunoassay method. Food had a marked effect on cyclosporine absorption from Sandimmune manifested by a nearly doubled time to reach the peak concentration and a 37% increase in the area under the curve. This was associated with significant elevations in subsequent whole-blood cyclosporine concentrations compared to the fasting administration. For Sandimmune Neoral the influence was less pronounced. The maximum concentration was decreased by 26%, without a relevant change in the time to reach the peak; the area under the curve showed a slight reduction of 15%. The relatively minor influence of a fat-rich meal on the absorption of cyclosporine from Sandimmune Neoral is advantageous when individualizing a dosage regimen under clinical and outpatient administration conditions.

Area-under-the-curve versus trough level monitoring of cyclosporine concentration: critical assessment of dosage adjustment practices and measurement of clinical outcome.

Cyclosporine (CsA) immunosuppressive therapy is confounded by a narrow therapeutic window and large intersubject variability. Trough-level monitoring cannot prevent the frequent occurrence of toxic side effects and graft rejection. Area-under-the-curve (AUC) monitoring affords better control of exposure of an individual patient to CsA, which is compensated for by a moderate increase in the number of blood samples. A linear pharmacokinetic model for CsA dosage adjustment using AUC measurements as feedback information decreased the incidence of delayed graft loss during the first month. This strategy failed, however, to lower the frequency of acute graft rejection despite a significant relationship between the probability of rejection and exposure to CsA (measured as average steady-state concentration). A nonlinear Michaelis-Menten model describes the relationship between oral dose rate and average steady-state concentration better than does a linear clearance model. Clinical utility of the nonlinear model remains to be proven.

Michaelis-Menten kinetics determine cyclosporine steady-state concentrations: a population analysis in kidney transplant patients.

Dosage adjustments of cyclosporine are confounded with an unexpected degree of variability, thus invalidating a direct proportionality between the oral dose rate and the steady-state concentration. In 1033 observations of dose rate and average steady-state concentration collected during therapeutic monitoring (area under the curve method) in 134 adult kidney transplant patients, a population pharmacokinetic analysis showed that a Michaelis-Menten model fitted the data better than a linear clearance model. It was further shown that the Michaelis-Menten constant (Km) parameter of the Michaelis-Menten model (the average steady-state concentration at half-maximal dose rate) increased during the first 4 months after transplantation whereas the maximal dose rate of the Michaelis-Menten model (Vmax) remained constant. The following parameters with interindividual variation in parenthesis were estimated: Vmax = 852 mg/24 hr (43%) and Km at 114 days after transplantation = 349 ng/ml (117%). An algorithm was derived from this population model that guides the clinician during the adjustment of oral cyclosporine dose rates.

Adrenochrome reaction implicates oxygen radicals in metabolism of cyclosporine A and FK-506 in rat and human liver microsomes.

The role of oxygen radicals in the metabolism of cyclosporine A (CyA), FR900506 (FK-506) and carbon tetrachloride (CCl4) catalyzed by the cytochrome P450 system was investigated in vitro in rat and human microsomal preparations. Varying concentrations of CyA, FK-506 and CCl4 (100 microM-1.0 mM) were added to microsomal preparations, and lipid peroxidation was measured by malondialdehyde (MDA) formation as detected by the thiobarbituric acid assay. The effects of oxygen radical scavengers [superoxide dismutase (SOD) and catalase (CAT)] and an antioxidant [glutathione (GLUT)] were tested on various incubations of CyA, FK-506 and CCl4 to assess the role of oxygen radicals in lipid peroxidation. CyA-dependent MDA formation was moderately inhibited by SOD in the rat model and increased by SOD in the human model. In both models, CAT slightly inhibited CyA-dependent MDA formation and GLUT significantly inhibited MDA formation. FK-506-dependent MDA formation, studied only in the rat model, paralleled CyA-induced MDA formation but showed greater inhibition with CAT and less inhibition with SOD or GLUT. In both models, CCl4-dependent MDA formation was significantly inhibited by GLUT and showed no sensitivity to SOD or CAT. In addition, the adrenochrome reaction, which measures the oxidation of epinephrine to adrenochrome, was used to measure the increased oxygen radical-flux resulting from the metabolism of CyA, FK-506 and CCl4. CyA with epinephrine showed the highest oxidative activity, followed by FK-506 and then CCl4, which showed the least formation of adrenochrome. These results indicated a role for oxygen radicals in CyA and FK-506 metabolism.

Generation of oxygen free radicals during the metabolism of cyclosporine A: a cause-effect relationship with metabolism inhibition.

A better understanding of the mechanism of lipid peroxidation during the metabolism of cyclosporine A (CsA) might help explain the toxicities of this immunosuppressive drug on various organs. Our in vitro work used microsomes prepared from livers of phenobarbital-induced male rats. The incubations (total volume 1ml) also contained a NADPH regenerating system and substrate (i.e., CsA, carbon tetrachloride, or aminopyrine) dissolved in ethanol. Lipid peroxidation was inferred from the presence of malondialdehyde (MDA) which was detected by the thiobarbituric acid assay. The formation of CsA hydroxylated metabolites (AM9 and AM1) was monitored by liquid chromatography. The activity of the microsomal incubation was confirmed by measurements of MDA and formaldehyde production caused by increasing concentrations of CsA, carbon tetrachloride, and aminopyrine. The occurrence of hydroxylated metabolites was not coupled to the production of MDA. Aminopyrine could inhibit MDA production by CsA, but CsA could not reduce the formation of formaldehyde by aminopyrine. Erythromycin, a competitor for the binding site of CsA on cytochrome P450, reduced MDA production by CsA, and CsA inhibited formaldehyde production by erythromycin. Interaction studies with SKF 525A, ketoconazole, superoxide dismutase, catalase, alpha-tocopherol, and reduced glutathione confirmed the role of cytochrome P450 and the presence of activated oxygen species as a source of microsomal peroxidation which in return may explain the inhibitory effect of CsA on cytochrome P450 itself.

Optimisation of immunosuppressive therapy using pharmacokinetic principles.

Clinical experience with immunosuppressive therapy is more extensive in the area of preventing the rejection of transplanted organs than in the treatment of autoimmune diseases. Among the many pharmacological agents presently in use, only prednisone (or methylprednisolone) and cyclosporin require dosage individualisation. Sources of interindividual variability in the pharmacokinetics of prednisone have been identified and are guiding the selection of individual dosage rates. As an alternative, a single timed concentration can determine an apparent value for prednisone clearance from which an individual dosage can be calculated. In contrast, numerous sources of inter- and intraindividual variability in cyclosporin pharmacokinetics prevent the easy selection of safe and effective starting dose rates. Indeed, test doses of cyclosporin followed by series of blood samples and the calculation of individual pharmacokinetic parameters are needed to assure successful immunosuppression right from the start. Furthermore, only continued monitoring sustains immunotherapy vis-à-vis intraindividual variability and a narrow therapeutic range of cyclosporin concentrations.

Pharmacokinetic strategies for cyclosporin therapy in organ transplantation.

Marked interindividual variations in cyclosporin (CsA) produce disparate clinical results in organ transplant recipients. In an attempt to eliminate marked deviations of insufficient or excessive CsA concentrations consequent to the administration of uniform drug doses, test dose pharmacokinetics were performed on each potential organ transplant candidate. An intravenous 3 mg/kg test dose delivered over 3 h proved to be readily performed, namely 53% perfect studies, and relatively reliable, namely 73% of observed concentrations within 10% of the predicted values. Furthermore, the use of CsA doses predicted by pretransplant studies reduces the incidence of delayed graft function, early rejection episodes and transplant loss. The oral test dose study predicted a suitable amount of CsA to achieve sufficient gastrointestinal absorption but was less accurate than the iv prediction method: namely, 40% of observed post-transplant concentrations were within 10% of the predicted target value. Furthermore, patients who received oral doses predicted by the test dose strategy showed no improvement in the incidence of acute rejection episodes between 7 and 60 days, and only modestly improved serum creatinine values. The lower accuracy of predictions from oral test dose studies may reflect the impact of non-linear oral (as opposed to iv) drug pharmacokinetics, of variable diet, and/or of altered postoperative gastrointestinal function.

The ability of pretransplant test-dose pharmacokinetic profiles to reduce early adverse events after renal transplantation.

Pretransplant test-dose pharmacokinetic profiles were used to determine individual cyclosporine drug bioavailability and clearance rates in renal transplant patients. Assuming a linear relation between dose and area under the concentration curve (AUC), starting i.v. and p.o. CsA doses were computed from the test-dose results. Target values were 400 ng/ml steady-state concentration (Css) during continuous intravenous infusion, and 500 ng/ml average drug concentration (Cavss = AUC/dosing interval) after oral administration, based upon measurements with the specific monoclonal antibody 3H-tracer radioimmunoassay. The outcomes after dose individualization with a 1-(n = 32), 2-(n = 38), or 3-(n = 41) hr i.v. infusion test dose and a p.o. test dose (n = 111) were compared with 228 historical control patients who received a uniform protocol of CsA i.v. at 2.5 mg/kg/day and p.o. at 14 mg/kg/day. The observed Css after i.v. CsA was within 10% of the target concentration in 73% of recipients tested with the 3-hr protocol, a significantly greater fraction than achieved with either the uniform dose (14%), or the 1-(34%) and 2-(25%) hr protocols. Patients in the 3-hr protocol group showed reduced incidences of delayed graft function, early graft loss, and rejection episodes, and a lower mean serum creatinine value, particularly at 7 but also at 30 days posttransplantation. Administration of the predicted oral dose produced a peak concentration of greater than or equal to 700 ng/ml drug absorption in 60% of recipients at 3 days, 90% at 5 days, and 98% at 7 days. The test-dose method less effectively predicted the appropriate oral CsA dose to produce target Cssav and failed to reduce the 90-day rejection incidence. Despite its limitations with the more-complicated p.o. route, the test-dose method successfully predicts i.v. CsA doses, thereby reducing the incidence of early adverse events.

Abbreviated kinetic profiles in area-under-the-curve monitoring of cyclosporine therapy.

Abbreviated kinetic profiles can reduce the number of phlebotomies and drug assays, and thereby the cost of area-under-the-curve (AUC) monitoring. In the present investigation, we used two independent data sets: group 1, 101 AUC profiles from 77 stable renal-transplant patients, which included a 5-h sample in addition to the usual 0-, 2-, 4-, 6-, 10-, 14-, and 24-h samples; and group 2, 100 profiles from 50 stable renal-transplant patients before and after a change in their daily oral dose of cyclosporine. Group I demonstrated a fair correlation between cyclosporine trough concentrations and the AUC calculated from a complete set of seven concentrations (r2 = 0.820 and 0.758 for the 24- and 0-h samples, respectively). Stepwise multiple linear-regression analysis revealed that the abbreviated set of three time points (2, 6, and 14 h) explained 96% of the variance in AUC values calculated from the full set of seven samples; additional time points increased the accuracy only slightly. For group 2, we examined the difference between the observed and the predicted concentrations by linear extrapolation; the error in the observed AUC value, compared with the predicted value calculated from seven time points (-13.2% to -1.2%), was similar to the error from just three time points (-11.5% to 4.5%). Abbreviated AUC profiles involving three time points used with a model equation seem to provide a reliable alternative to full seven-point profiles.

Area under the curve monitoring of cyclosporine therapy: the early posttransplant period.

The impact of a new monitoring strategy on whole blood concentrations of cyclosporine measured by a specific monoclonal radioimmunoassay was investigated in a group of 37 renal transplant patients. Before transplantation, the patients received a standard intravenous (i.v.) and oral (p.o.) test dose of cyclosporine to calculate their individual i.v. and p.o. starting dose rates to achieve a certain target steady-state cyclosporine concentration. After transplantation, the designated i.v. dose rate was continuously infused for 2 days, at which time the steady-state concentration was measured. Then, the designated oral dose for 24 h was administered while the infusion was continued at an unaltered rate. The oral absorption of cyclosporine was documented by blood samples over the following 8 h. If necessary, this overlap of i.v. and p.o. dosing was repeated until blood concentrations of cyclosporine rose at least 700 ng/ml over the steady-state concentration. By that time, the infusion was stopped and oral dosing continued. Individualized infusions led to steady-state concentrations within a range that did not exceed 1.1 times the median concentration of 472 ng/ml. Standard infusion rates in the past produced a much wider range of steady-state concentrations (9.6 times the median). Individualized infusions reached the target steady-state concentration with a significant positive bias of 17% (SEM = +/- 32%, range of -36 to +105%). Individualized oral doses reached the target average steady-state concentration (calculated by dividing the area under the concentration-time curve by the dosing interval) with an inferior precision (median = 2.6%, range of -54 to +130%) but without a positive or negative bias.(ABSTRACT TRUNCATED AT 250 WORDS)

Prediction of acute graft rejection in renal transplantation: the utility of cyclosporine blood concentrations.

While cyclosporine is recommended to be used only in conjunction with monitoring of its blood concentrations, the utility of these measurements in preventing treatment failure is not established. In a group of 52 patients trough levels and steady-state concentrations were monitored in serum and whole blood by specific (SP) and nonspecific (NS) assays (polyclonal radioimmunoassay, PR; fluorescence polarization immunoassay, FP; high-pressure liquid chromatography, HP). From as many as 10 determinations of trough level and steady state concentrations during the first 40 days after renal transplantation, the lowest measurement was selected. In the case of an acute rejection episode within that time period, only values until that event were considered. Trough level measurements in serum by PR/NS and by FP/NS and in whole blood by HP/SP were not significantly different between patients with and patients without rejection episodes. However, simultaneously measured steady-state values (serum/PR/NS and serum/FP/NS) were significantly lower in patients suffering from rejection (with rejection SS/serum/PR/NS mean = 127 ng/ml, SD = 41 ng/ml; without rejection mean = 163 ng/ml, SD = 60 ng/ml; P = 0.027, t test). This difference could not be demonstrated for steady state/whole blood/HP/SP measurements. A logistic regression analysis demonstrated that the probability of rejection can be decreased by up to 40% if steady state/serum/PR/NS or steady state/serum/FP/NS values never drop below 250 ng/ml early after renal transplantation.

Preemptive transplantation--an analysis of benefits and hazards in 85 cases.

The benefit of transplantation without prior dialysis might be contravened by the failure to develop possible immunologic disabilities associated with chronic uremia and dialysis. This study compares graft and patient outcome, cyclosporine toxicity, pharmacokinetics, rejection episodes, nutritional status, and social and vocational rehabilitation between a preemptive group of 85 patients transplanted without prior dialysis and a cohort of 84 demographically, temporally, and disease-matched recipients of renal transplants after a minimum of 6 months' chronic dialysis therapy. The groups were matched for donor type, gender, and age, as well as immunologic risk factors of HLA-mismatch and percent panel-reactive antibody. All patients received CsA and prednisone immunosuppression. There were only two differences between the cohorts. The preemptive group included more diabetic patients: 32 versus 15 (P less than 0.01). The control cohort included more recipients who had received any pretransplant transfusion: 55 versus 28 (P less than 0.001). Both of these factors (if having any impact) would be expected to reduce graft survival in the preemptive group. All patients in the study had a minimum follow-up of 1 year and over half of the recipients are beyond 40 months. The preemptive patients showed survival rates of 94, 93, and 91 percent at 1, 2, and 5 years. These rates were not significantly different from those of the control group, namely 96, 96, and 93 percent, respectively. The actuarial graft survival rates in the preemptive group of 83, 81, 76, 73, and 73 percent at 1, 2, 3, 4, and 5 years were not statistically different from the control group rates, namely 90, 81, 80, 77, and 76 percent. Preoperative blood transfusion or percent positive panel-reactive antibodies had no effect on postoperative outcome in either group. The incidence of CsA nephrotoxicity was 9.4 percent in the preemptive group, which was not statistically different from the 17.9 percent in the control group. The incidence of rejection episodes in the absence of patient noncompliance was comparable between the groups. Seven of the irreversible rejection episodes in the preemptive group were due to noncompliance, compared with none in the control group (P less than 0.001). Preemptive recipients were also more likely than control group patients to be employed fulltime both before transplantation (36 vs. 22, P less than 0.05) as well as after transplantation (38 vs. 20, P less than 0.01).(ABSTRACT TRUNCATED AT 400 WORDS)

Cyclosporine pharmacokinetics in liver transplant recipients: evaluation of results using both polyclonal radioimmunoassay and liquid chromatographic analysis.

Pharmacokinetic variables were derived from cyclosporine measurements using liquid chromatography (HPLC) and radioimmunoassay with a non-selective polyclonal antibody (PARIA) in 11 orthotopic liver transplant recipients studied in paired oral and intravenous studies both before and after permanent clamping of the biliary T-tube. After oral drug administration, mean areas under blood cyclosporine concentration versus time curves before clamping were around 5.2-fold greater by PARIA than HPLC but 2.9-fold greater after clamping and closer to comparable values after intravenous cyclosporine (2.5 and 2.3-fold, respectively). Cyclosporine clearance was smaller by PARIA than HPLC (mean 7.3 versus 3.3 ml.min-1.kg-1, respectively, before clamping). Both values decreased by 25% after clamping (to 5.5 and 2.4 ml.min-1.kg-1, respectively), although there was no significant change in distribution or elimination half-lives (around 0.5 and 8 h, respectively). The mean bioavailability of oral cyclosporine increased significantly after clamping in 9 patients (from 10.6% to 28.1% by HPLC and from 14.8 to 35.1% by PARIA) but in two patients who developed the vanishing bile duct syndrome values fell to less than 10% and the proportional overestimation of cyclosporine concentrations by PARIA increased. Clamping had no singificant effect on the mean apparent volumes of distribution but values of Vz were approximately twice those of Vss (around 2.6 and 1.31.kg-1 by PARIA and HPLC respectively). Mean half lives after clamping were shorter following oral than intravenous cyclosporine (t 1/2 lambda 2 around 15 h enterally versus 8 h parenterally).(ABSTRACT TRUNCATED AT 250 WORDS)

Consensus document: Hawk's Cay meeting on therapeutic drug monitoring of cyclosporine.

The optimal measurement method and clinical application of the therapeutic drug monitoring of cyclosporine remain uncertain. At a workshop held at Hawk's Cay, FL, from January 14 to January 17, 1990, 57 scientists presented their latest research findings, either in formal papers or as discussants. Lively debate and discussion followed presentation of extant and new methodologies for drug measurements as well as multicenter validation studies: applications of trough-concentration monitoring in renal, hepatic, and bone-marrow transplants as well as in autoimmune disease; and alternative pharmacokinetic approaches to guide cyclosporine therapy. The process of inducing and maintaining optimal immunosuppression to facilitate graft success is a complex and often challenging task, requiring the combined expertise of multiple disciplines. Thus, the assembly of four of the groups essential to the transplant process--clinicians, laboratory scientists, the pharmaceutical company, and the manufacturers of cyclosporine measurement kits--provided a unique opportunity to evaluate therapeutic drug monitoring issues facing the transplant field. Here we present the major conclusions reached at the meeting, brief discussions of the study data on which they are based, and a summary of unresolved problems that will require further rigorous investigations. The Consensus Document was reviewed by all the workshop participants before we submitted this final manuscript.