Philip R. Mayer

Unaltered etanercept pharmacokinetics with concurrent methotrexate in patients with rheumatoid arthritis

The purpose of this study was to evaluate the potential impact of concurrent weekly oral methotrexate administration on the pharmacokinetics of etanercept in patients with rheumatoid arthritis (RA) in a phase 3B trial. As part of a double‐blind randomized trial of 682 patients with rheumatoid arthritis who received etanercept (25 mg subcutaneously twice weekly), methotrexate (weekly oral dose, median weekly dose: 20 mg), or etanercept (25 mg subcutaneously twice weekly) plus methotrexate (weekly oral dose, median weekly dose: 20 mg), serum etanercept concentrations were measured in a subset of patients. Serum samples for 98 randomly selected patients (48 receiving etanercept‐alone treatment, 50 receiving etanercept plus methotrexate combination treatment) were analyzed to assess the pharmacokinetics of etanercept. A single blood sample was drawn from each patient at baseline and at the week 24 visit. Given the variable sampling time for patients in both groups, a population pharmacokinetic analysis using NONMEM was conducted for etanercept. A final covariate population pharmacokinetic model was constructed based on previously obtained etanercept data from both healthy subjects (n = 53) and patients with RA (n = 212) in 10 prior clinical trials. The predictive performance of the final model was assessed by both bootstrap and data‐splitting validation approaches. The final model was then used to estimate Bayesian pharmacokinetic parameters for the patients in both treatments in the current trial. The potential effect of the concurrent administration of methotrexate on the pharmacokinetics of etanercept was examined by comparing the clearance values between 2 treatments using statistical criteria. A population 2‐compartment model with first‐order elimination from the central compartment and with either zero‐order (intravenous administration) or first‐order (subcutaneous administration) input was selected based on the data from the prior 10 etanercept clinical studies. The following pharmacokinetic parameters (typical value ± standard error) were estimated: clearance (CL: 0.072 ± 0.005 L/h), volume of distribution in the central compartment (Vc: 5.97 ± 0.45 L), volume of distribution in the peripheral compartment (Vp: 2.05 ± 0.32 L), intercompartment clearance (Q: 0.0645 ± 0.0093 L/h), first‐order absorption rate constant (ka: 0.0282 ± 0.0039 1/h), and absolute bioavailability for subcutaneous administration (F: 0.626 ± 0.056). Interindividual variability of the pharmacokinetic parameters was quantified for CL (25.1%), Vc (41.7%), ka (53.1%), and F (24.2%). Residual variability consisted of combined additive (11.4 ng/mL) and proportional error (49.9%). Both age (< 17 years) and body weight (< 60 kg) were found to be important covariates on CL. The results of both validation tests indicated the adequate predictive performance of the population model. Based on the bioequivalence criteria, the Bayesian‐estimated clearance for patients receiving etanercept alone (mean: 0.070 L/h) was comparable to that for patients receiving a combination of etanercept and methotrexate (mean = 0.066 L/h). The pharmacokinetics of etanercept were not altered by the concurrent administration of methotrexate in patients with rheumatoid arthritis. Thus, no etanercept dose adjustment is needed for patients taking concurrent methotrexate.

Pharmacokinetics of gemtuzumab ozogamicin as a single-agent treatment of pediatric patients with refractory or relapsed acute myeloid leukemia.

Gemtuzumab ozogamicin is currently approved to treat CD33‐positive acute myeloid leukemia (AML) in first relapse in patients older than age 60 years. The objective of this study was to characterize the pharmacokinetics of gemtuzumab ozogamicin in pediatric patients with relapsed or refractory AML. The study population comprised 29 subjects younger than age 18 with AML in first relapse. Dosages of 6, 7.5, and 9 mg/m2 were administered during the study. Pharmacokinetic parameters were determined following each dose for hP67.6, total calicheamicin derivatives, and unconjugated calicheamicin derivatives. hP67.6 pharmacokinetic parameters had a consistent and statistically significant change between the first and second doses. Increases in AUC and decreases in both CL and Vss from the first dose to the second dose were consistent with those of the adult population. Changes between dose periods for total calicheamicin derivatives and unconjugated calicheamicin derivatives were consistent with those of hP67.6. Changes in pharmacokinetic parameters between dose periods are attributed to saturation of CD33 binding sites and diminished clearance resulting from a lower peripheral blast burden and antigen. Children receiving 9 mg/m2 had the following hP67.6 pharmacokinetic parameters: Cmax, 3.47±1.04 mg/L; AUC, 136 ±107 mg•h/L; CL, 0.12 ±0.15 L/h/m2; Vss, 6.5 ±5.5 L; and t1/2, 64±44 h after their first dose. Mean pharmacokinetic values are similar to values reported in adults. Individual children demonstrated large intersubject variability, similar to adults. The pharmacokinetics of gemtuzumab ozogamicin in pediatric patients closely follow the profile and variability of adult patients.

Pharmacokinetics of pantoprazole in patients with moderate and severe hepatic dysfunction

BACKGROUND Patients with impaired hepatic function usually require gastric acid-suppressant therapy but are at increased risk for drug interactions and may require dosage adjustments. The proton pump inhibitor pantoprazole is rapidly absorbed and eliminated, primarily by cytochrome P450 (CYP) 2C19 isozymes. OBJECTIVE This study sought to determine whether dosage adjustment of pantoprazole is required in patients with moderate or severe hepatic impairment by comparing the pharmacokinetic profile of pantoprazole in such patients with that in healthy slow metabolizers of pantoprazole, in whom no dosage adjustment is required. METHODS Patients with moderate (Child-Pugh class B) and severe (Child-Pugh class C) hepatic impairment received oral pantoprazole 40 mg once daily on days 1 through 4 and then on alternate days (days 6 and 8). Serial blood samples were collected on days 4 and 8 for analyses of plasma pantoprazole concentrations. Pharmacokinetic data were compared between the 2 groups with hepatic impairment and against historical data from 17 healthy subjects who were genetically slow CYP2C19 metabolizers of pantoprazole. RESULTS Twenty-two patients participated in the study, 13 in the Child-Pugh class B group and 9 in the Child-Pugh class C group. No clinically significant differences in pantoprazole pharmacokinetics were noted between the patients with hepatic impairment and the healthy slow metabolizers of pantoprazole on days 4 and 8. Pantoprazole was well tolerated. Four Child-Pugh class B patients and 3 Child-Pugh class C patients reported > or = 1 adverse event. Adverse events were generally mild or moderate, and were similar to those reported in healthy subjects. Two patients discontinued the study because of severe events related to their underlying disease. CONCLUSIONS The pharmacokinetics and tolerability of pantoprazole were similar in patients with moderate hepatic impairment, patients with severe hepatic impairment, and healthy slow metabolizers of pantoprazole, in whom no dosage adjustment is required. Thus, no dosage adjustment of pantoprazole is required in patients with hepatic impairment, regardless of its severity. However, caution should be exercised when giving pantoprazole to patients with severe hepatic impairment.

Pharmacodynamic Modeling of Pantoprazole's Irreversible Effect on Gastric Acid Secretion in Humans and Rats

The relationship between the pharmacokinetics of pantoprazole, an irreversible proton pump inhibitor, and its effect on gastric acid secretion was evaluated in humans and rats. Pantoprazole pharmacokinetics were studied in 6 rats (5 mg/kg, IV) and 22 healthy volunteers (10 to 80 mg, IV and oral). Gastric acid secretion under maximum pentagastrin stimulation was measured after IV administration of placebo or pantoprazole in 31 rats (0.12 to 1.15 mg/kg) for 4 hours and in 31 subjects (20 to 120 mg) for 24 hours. Pantoprazole has short half‐lives of 0.5 hours in rats and 0.8 hours in humans. After administration of the highest dose, acid secretion was fully inhibited within 1 hour and for the whole observation period in both species. An irreversible pharmacodynamic response model was successfully developed and validated. The apparent reaction rate constants of pantoprazole with the proton pumps were 0.691 L/mg/h in rats and 0.751 L/mg/h in humans, and the apparent recovery rates of the pumps were 0.053 h‐1 and 0.031 h‐1, respectively. The maximum inhibition and the overall effect of pantoprazole are related to exposure, and the onset is related to initial pantoprazole concentrations. It was concluded that this irreversible response model accurately describes the effect of IV and oral pantoprazole on gastric secretion and maybe used to predict effects under other dosage regimens.

Lack of Pharmacokinetic Interaction between Oral Pantoprazole and Cisapride in Healthy Adults

Pantoprazole, an irreversible proton pump inhibitor, may be administered with cisapride, a prokinetic agent. As increased cisapride concentrations may result in longer electrocardiogram (ECG) QTc intervals, a crossover study was conducted in healthy subjects to evaluate the oral pharmacokinetic interaction between cisapride (20 mg) and pantoprazole (40 mg). After dosing, serial blood samples and 12‐lead ECGs were collected, and cisapride plasma concentrations were quantitated. For cisapride alone, mean parameter values were the following: peak concentration (Cmax), 56 ng/mL; time to Cmax (tmax), 1.7 hours; area under the concentration‐time curve (AUC), 426 ng•h/mL; and terminal half‐life (t1/2), 5.8 hours. Pantoprazole coadministration did not alter cisapride AUC or other pharmacokinetic parameters except for a slight 17% decrease in Cmax, resulting in 90% confidence limits of 79% to 88%, which were marginally outside strict bioequivalence limits. In addition, cisapride did not affect ECG QTc intervals, with or without pantoprazole. Therefore, no dosage adjustment is needed when pantoprazole and cisapride are coadministered.

Dose-dependent pharmacokinetics of the aldose reductase inhibitor imirestat in man.

The pharmacokinetics of imirestat were studied in healthy volunteers following single and multiple oral doses. After single doses of 20 to 50 mg, imirestat plasma concentrations declined with an apparent elimination half-life of 50 to 70 hr over the 168 hr in which levels were measured. However, with lower doses (2 to 10 mg), an initial rapid decline in drug concentration was followed by a very slow terminal elimination phase with plasma concentrations decreasing little over the 1 week of sampling. This resulted in a decrease in apparent t1/2 with increasing dose, from 272 ± 138 hr at 2 mg to 66 ± 30 hr at 50 mg. During once-daily dosing of 2 to 20 nig/day for 4 weeks, mean steady-state imirestat concentrations appeared to be dose proportional, although the time required to achieve steady state decreased with increasing dose. The mean effective half-life for accumulation ranged from 54 to 98 hr, suggesting that the very slow elimination of drug at low concentrations did not produce disproportionate accumulation of drug at these doses. Mean oral clearance was independent of dose, ranging from 30 to 45 ml/min. At the 2-, 5-, and 20-mg doses, one subject in each group had steady-state concentrations two- to fourfold greater than any of the other five subjects at the same dose, although the reason for this was not apparent from these data. The overall kinetic profile of these data was suggestive of dose-dependent pharmacokinetics resulting from nonlinear tissue binding of imirestat. A two-compartment pharmacokinetic model incorporating saturable binding in the tissue compartment and elimination from the central compartment was developed and provided a good description of the plasma concentration data after both single and multiple dosing.

Saturable tissue binding and imirestat pharmacokinetics in rats

To investigate the hypothesis that the pharmacokinetics of imirestat, an aldose reductase inhibitor, are influenced by saturable binding to tissues, three experiments were done. (1) The nature of the dose dependence was characterized in rats. Two groups of nine adult male Sprague–Dawley rats received iv 14C-imirestat at doses of 2 or 8 mg/kg. Serial blood samples were obtained over 15 days. Volume of distribution at steady-state was significantly different between the high- and the low-dose groups (0.744 ± 0.103 1 and 1.10 ± 0.228 L, respectively). Clearance was independent of dose over this fourfold range (∼15 ml/hr). (2) The effect of either statil or AL3152, both aldose reductase inhibitors and potential competitors for aldose reductase binding, on the pharmacokinetics of a single 0.2-mg/kg iv dose of imirestat was assessed. A 2.4-mg/kg loading dose of statil was administered and a constant-rate infusion (56 µg/hr/kg) was begun 16 hr before imirestat. A 2-mg/kg loading dose of AL3152 and a constant-rate infusion (115 µg/kg/hr) were also administered 16 hr before imirestat. The infusions were maintained throughout the study. AL3152 administration decreased the imirestat steady-state volume of distribution by a mean of 63%. Statil administration decreased it by a mean of 39%. (3) The dosing regimen of the second study was repeated and, at two sampling times, nine tissues and plasma were obtained from four rats per sampling time for determination of imirestat tissue-to-plasma concentration ratio. The tissue/ plasma imirestat concentration ratio in the adrenals 24 hr after imirestat administration was 56.9 ± 20.0 in the imirestat group, 17.7 ± 1.27 in the statil-coadministered group, and 12.3 ± 2.59 in the AL3152-coadministered group. A similar trend of decrease in the ratios was observed in all tissues at both 24 and 168 hr. The results suggest that a saturable tissue binding phenomenon at least partially accounts for the nonlinear pharmacokinetics of imirestat.

Application of Statistical Moment Theory to Pharmacokinetics

A s with other selected pharmacokinetic concepts, statistical moment theory has been extracted from chemical engineering and related disciplines. The appeal of this methodology resides in its noncompartmental approach to pharmacokinetic data analysis. To date, statistical moments have been applied to pharmacokinetic and biopharmaceutic processes dealing primarily with drug absorption and distribution. These concepts were initially used in the biomedical literature by Perl and Samuel to analyze cholesterol input and output in man,1 and by Oppenheimer et al., in describing the disposition of iodothyronine.2 A more complete discussion of the potential utility of statistical moments in pharmacokinetic analysis was reported by Yamaoka et al.,3 and Cutler4 in 1978. In the past ten years, the role of statistical moments in pharmacokinetics has been extended and promoted as an adjunct to conventional data handling, although the approach is not without limitations. The theory of statistical moments rests essentially on a probability function. Whereas a derivation involving drug molecule probabilities within a normal distribution profile may seem rather complex on the surface,5 simpler definitions may provide a great deal of practical application. The zero and first moment of the drug concentration versus time curve are defined, respectively, as follows:

Comparison of the Pharmacokinetics and Pharmacodynamics of the Aldose Reductase Inhibitors, AL03152 (RS), AL03802 (R), and AL03803 (S)

The pharmacokinetics of AL03152 (RS) and its enantiomers, AL03802 (R) and AL03803 (S), were studied in the Sprague–Dawley rat following intravenous bolus administration. The enantiomers had differing pharmacokinetic profiles, while the racemic compound exhibited pharmacokinetic parameters approximating the mean values of the individual enantiomers. The total clearance (CLT) values of the two enantiomers were similar, but the intrinsic clearance (Clint) was much greater for the S-enantiomer than for the R-enantiomer. The volume of distribution (Vss) for AL03802 (R) was threefold greater than that for AL03803 (S). The stereoselectivity in Vss could not be totally accounted for by the slight difference in serum protein binding of the isomers and resulted in a difference in the half-lives of the enantiomers. Only the R-isomer exhibited a persistent terminal elimination phase, consistent with more extensive tissue binding than the S-isomer. AL03152 enantiomers were equivalent in potency assessed from in vitro IC50 values toward rat lens aldose reductase and rat kidney L-hexonate dehydrogenase and lens EC50 values in diabetic rats.

Pharmacokinetics and efficacy of structurally related spirohydantoin and spirosuccinimide aldose reductase inhibitors

1. Six potent aldose reductase inhibitors (ARI), three spirohydantoin (I to III) and three spirosuccinimide (IV to VI) compounds, showed similar IC50 activities in vitro for the inhibition of rat lens aldose reductase, but their ED50 values in diabetic rats varied as much as 20-fold in the lens and 50-fold in the sciatic nerve tissue. Pharmacokinetic studies were undertaken to investigate these findings. Structure-pharmacokinetic relationships were studied following i.v. administration to cynomolgus monkeys. 2. The clearance (CL) of each spirosuccinimide ARI was faster (greater than 5 times) than that of the corresponding spirohydantoin compound. In both series the CL values of the C(4) methyl and methoxy analogues were 4-fold greater than those for the unsubstituted compounds, although the CL values of the methoxy and methyl derivatives in the same series were not significantly different. 3. The volumes of distribution (Vss) of the spirohydantoins were about one-half those of the corresponding spirosuccinimides, and the Vss values of the parent compounds of both ARI series did not differ dramatically from those of their methyl and methoxy analogues. 4. All six compounds were eliminated from plasma in a biexponential fashion. The half-lives (lambda 1 and lambda 2) of the spirohydantoin compounds were much longer than those of the corresponding spirosuccinimide compounds, and the unsubstituted compounds had longer half-lives than their methyl and methoxy derivatives. The longest lambda 1 and lambda 2 half-lives were observed for imirestat, while two of the spirosuccinimides had the shortest half-lives. 5. These results indicate that the relationships observed between the in vitro and in vivo activities of the six ARI can be attributed to structurally dependent differences in metabolic clearance.