Nonlinear Mixed Effects Model Program Research Articles

Population pharmacokinetics of mizoribine in pediatric recipients of renal transplantation

An immunosuppressive agent, mizoribine, is excreted predominantly in the urine. The aim of this study was to investigate the pharmacokinetic variability of mizoribine in pediatric recipients of renal transplantation. Pharmacokinetic data for population analysis were collected from 51 recipients (32 males and 19 females) treated with oral administration of mizoribine (0.83-5.56 mg/day/kg). The population pharmacokinetic parameters of mizoribine were estimated using a nonlinear mixed effects model program. The pharmacokinetics of mizoribine in pediatric recipients of renal transplantation was well described by a one-compartment model with first-order absorption. The mean value of the absorption lag time (ALAG) and absorption rate constant (K (A)) was estimated to be 0.363 h and 0.554 h(-1), respectively. Apparent volume of distribution (V/F) was modeled as a function of body weight (WT), and the mean value was estimated to be 1.03 · WT L. Oral clearance (CL/F) was modeled as a function of creatinine clearance (CL(cr)), and the mean value was estimated to be 2.81 · CL(cr) · 60/1000 L/h. In addition, there was a positive correlation between CL(cr)-corrected CL/F and WT-corrected V/F in the pediatric recipients, indicating large interindividual variability in the bioavailability (F) of mizoribine. The present findings indicated that the rate of renal excretion and also the extent of intestinal absorption of mizoribine are responsible for the large interindividual pharmacokinetic variability of the drug.

Population pharmacokinetics of metformin in obese and non-obese patients with type 2 diabetes mellitus

The objective of this study was to develop a population pharmacokinetic model and investigate the effect of several demographic covariates on metformin pharmacokinetics in patients with type 2 diabetes mellitus, over a wide range of weights. A total of 105 patients received different metformin regimens, and pharmacokinetic sampling included a minimum of two concentrations per patient. Plasma determination of metformin was assayed by high performance liquid chromatography. Population pharmacokinetics was modelled using a nonlinear mixed effects model program (Monolix version 3.1s). An open one-compartment model adequately described metformin data. Lean body weight was a better size descriptor than actual body weight or ideal body weight for clearance (CL/F) and volume (V/F) parameters. CL/F was negatively related to age and serum creatinine (SCr). The estimation of specific coefficients for these effects gave better results than the use of renal function descriptors (Cockroft or MDRD). A dose effect in the relative bioavailability was demonstrated. The pharmacokinetics of metformin was influenced by lean body weight on an allometric basis and was related to markers of renal function, age, and serum creatinine in this population of 105 patients.

Population pharmacokinetics of mizoribine in adult recipients of renal transplantation

The aim of the present study was to estimate the population pharmacokinetic parameters of mizoribine in adult recipients of renal transplantation using a nonlinear mixed effects model (NONMEM) program. Pharmacokinetic data for population analysis were retrospectively collected from 114 recipients (66 males and 48 females) routinely treated with oral administration of mizoribine (25-450 mg/day). The range of creatinine clearance (CL(cr)) was 7.6-136.1 mL/min (mean 49.2 mL/min). The pharmacokinetics of mizoribine in adult recipients of renal transplantation was well described by a 1-compartment model with first-order absorption. The mean value of the absorption lag time (ALAG) and absorption rate constant (KA) was estimated to be 0.581 and 0.983 h(-1), respectively. Apparent volume of distribution (V/F) was modeled as a function of body weight (WT), and the mean value was estimated to be 0.858 × WT L. Oral clearance (CL/F) was modeled as a function of creatinine clearance (CL(cr)), and the mean value was estimated to be 1.80 × CL(cr) × 60/1000 L/h. In addition, there was a strong correlation between CL(cr)-corrected CL/F and WT-corrected V/F in the adult recipients, indicating large interindividual variability in bioavailability (F) of mizoribine. The present findings suggested that not only the rate of renal excretion but also the extent of intestinal absorption of mizoribine is responsible for the large interindividual pharmacokinetic variability of the drug.

Pharmacokinetic evaluation of the vinorelbine–lapatinib combination in the treatment of breast cancer patients

The objectives of this study were to investigate the pharmacokinetics of intra-venous vinorelbine combined with lapatinib as well as the effect of covariates in breast cancer patients. Women with HER2 + locally advanced or metastatic breast cancer progressing after ≤ 2 lines of trastuzumab-based treatment were treated with lapatinib per os starting 7 days (D) (D-7 to D0) before adding vinorelbine on a D1 & D8 every 3 weeks intravenous schedule. Lapatinib was given everyday. Dose levels [DL, lapatinib (mg)/vinorelbine (mg/m(2))] ranged from 750/20 to 1,250/25. A total of 29 patients, 37-76 years old, were treated with the combination of lapatinib + vinorelbine. For pharmacokinetic analysis, 7 time point samples were collected on D1 of cycle 1 for lapatinib and vinorelbine assays. For vinorelbine and lapatinib, respectively, whole blood and plasma concentrations were measured using ultra performance liquid chromatography with tandem mass spectrometry validated methods. Data analysis was performed using a non-linear mixed effect model program (Monolix version 3.1 s). A three-compartment open model adequately described vinorelbine pharmacokinetics. Body weight (BW) and platelet count significantly influenced blood vinorelbine clearance (CL). BW significantly influenced volume (V) and CL terms. Platelet count influenced vinorelbine elimination CL. The final parameter estimates were as follows: CL = 24.9 L/h, V1 = 8.48 L, Q2 = 50.7 L/h, V2 = 1,320 L, Q3 = 66.1 L/h, and V3 = 62.4 L (Qi and Vi denote inter-compartmental clearance and peripheral volume of distribution, respectively), normalized for a 70-kg patient according to BW allometric scaling (CL is normalized for a 250,000 platelet count). A one-compartment model with linear elimination adequately fitted the lapatinib plasma concentration-time data. The population pharmacokinetic parameters were CL = 27.7 L/h, V = 357 L, and the absorption constant, ka = 0.44 h(-1). The between-subject variabilities (BSV) could be well estimated for CL, V but not for ka. No covariate effect, including body surface area and vinorelbine dosage, could be identified for lapatinib. The pharmacokinetic modeling of vinorelbine and lapatinib was consistent with the results previously reported. BW and platelet count were confirmed as influencing blood CL of vinorelbine. A pharmacokinetic interaction occurred between vinorelbine and lapatinib probably due to lapatinib inhibition of CYP450-3A4. The combined lapatinib administration decreases statistically significant the vinorelbine CL. The maximal tolerated dose for the combination of lapatinib with vinorelbine on a q3w schedule is as follows: lapatinib 1,000 mg/day continuously and vinorelbine 22.5 mg/m(2) D1 & D8.

Developmental pharmacokinetics of etoposide in 67 children: lack of dexamethasone effect

A randomized clinical trial examined whether dexamethasone administration prior to ondansetron followed by etoposide and carboplatin infusions, and single-nucleotide polymorphisms (SNPs) of CYP3A4, CYP3A5 and MDR1 genes could modify etoposide pharmacokinetics in pediatric patients. Patients, 67 children, aged 14 weeks to 16.7 years, were treated for various malignancies and received either 3- or 5-day courses of etoposide and carboplatin: these two drugs were always administered after ondansetron infusion but combined or not with dexamethasone 5 mg/m²/day 30 min prior to etoposide infusion. Population pharmacokinetics was modeled using a non-linear mixed effect model program (Monolix version 31 s). Etoposide pharmacokinetics was ascribed to a 2-compartment model. The most significant covariate effect was bodyweight (BW), so the parameters were standardized to a 70-kg BW using the allometric ¾ or 1 power model for clearance (CL, Q) or volume terms (V), respectively. The population means for clearance and central volume of distribution were 2.05 l/h/70 kg and 9.21 l/70 kg with the corresponding between-subject variabilities, 0.26 and 0.28. Dexamethasone treatment had no effect on CL, either at the first or at the last administration occasion. CYP3A and MDR1 examined SNPs had no significant effect. Pharmacokinetics of etoposide was influenced by BW on an allometric basis in this pediatric population. Dexamethasone did not influence etoposide pharmacokinetics during these 3-5 days courses. These results should allow a better individualization of etoposide dosing in children.

Population Pharmacokinetics of R- and S-Carvedilol in Japanese Patients with Chronic Heart Failure

Carvedilol is a beta-adrenoceptor antagonist used for treating chronic heart failure (CHF). Two clinical studies were conducted to evaluate the population pharmacokinetics and pharmacodynamics of R- and S-carvedilol, and associated covariates, in patients with CHF. Fifty-eight patients (male=45, female=13) with New York Heart Association class I-IV CHF were enrolled in two clinical studies. R- and S-carvedilol concentrations were measured using HPLC at steady-state after oral administration of carvedilol at 1.25-20 mg o.d. or b.i.d. The data from both studies were used to estimate the population pharmacokinetic parameters and covariates using the nonlinear mixed effects model program. For 40 patients evaluated in one clinical study, the cytochrome P450 (CYP)2D6 *1, *10, and *5 genotypes were determined using allele-specific primer PCR, and individual patients' oral clearance (CL/F) of both enantiomers were estimated by the empirical Bayes method. A one-compartment model with a first-order absorption rate was established, in which body weight and alpha(1)-acid glycoprotein were significant covariates. Individual CL/F values for carvedilol were significantly lower in Japanese CHF patients with the CYP2D6 *1/*5, *5/*10 and *10/*10 genotypes. Estimation of the population pharmacokinetic parameters and their covariates for each enantiomer in Japanese patients with CHF showed that the CL/F values for R- and S-carvedilol were dependent on body weight, alpha(1)-acid glycoprotein, and CYP2D6 genotype. Prediction of exposure to free plasma carvedilol is important for dosage adjustment of beta-blocker therapy in patients with CHF.

Pharmacokinetics of sufentanil administered by target-controlled infusion in Chinese surgical patients

Background Target-controlled infusion (TCI) has been recently developed and successfully implemented in clinical practice. This study was conducted to determine the pharmacokinetics of TCI administered sufentanil in Chinese surgical patients. Methods The pharmacokinetics of sufentanil was investigated in 12 adult patients, aged 23–76 years, scheduled for prolonged surgery under general anesthesia. Anesthetic induction was carried out with propofol, rocuronium and TCI administered sufentanil aiming for target effect-site concentration of sufentanil 4 or 6 ng/ml. Sufentanil TCI lasted for 30 minutes. Frequent arterial blood samples (1.5 ml) were drawn during and up to 24 hours after sufentanil TCI. Plasma sufentanil concentrations were measured by liquid chromatography-tandem mass spectrometry; limit of sensitivity of mass spectrometry was 5 pg/ml. The data were analyzed with the nonlinear mixed-effect model program. Results The pharmacokinetics of TCI administered sufentanil were optimally described by a three-compartment model with the following parameters: the central volume of distribution (V1) = 5.4 L, the volume of distribution at steady-state (Vdss) = 195.4 L, systemic clearance (Cl1) = 1.10 L/min, and elimination half-life (t1/2 γ) = 271.8 minutes. Both age and gender affected the pharmacokinetic parameters. The rapid distribution clearance (Cl2) was negatively correlated with patient age, and the volume of slowly equilibrating compartment (V3) was positively correlated with age. The Cl2 and the volume of rapidly equilibrating compartment (V2) were influenced by gender with male patients showing higher values of Cl2 and V2 than female patients. There was no relationship of body weight, lean body mass, plasma albumin, or target effect-site concentration of sufentanil with any of the pharmacokinetic parameters studied. Conclusions The pharmacokinetics of TCI administered sufentanil in Chinese patients can be adequately described by a three-compartment model. Pharmacokinetics adjusted to the individual patient should improve the accuracy of TCI systems.

Population pharmacokinetics of oral levofloxacin 500mg once-daily dosage in community-acquired lower respiratory tract infections: results of a prospective multicenter study in China

This study aimed to explore the pharmacokinetic features of levofloxacin (LVFX) in Chinese patients with infections and to confirm oral LVFX 500 mg once daily as an optimal treatment regimen based on pharmacokinetic-pharmacodynamic (PK-PD) analysis. A total of 1052 plasma samples from 164 Chinese adult patients with community acquired lower respiratory tract infections (CALRTIs) and 18 healthy volunteers were used for population PK analysis. LVFX 500-mg tablets were given once daily. A nonlinear mixed effects model (NONMEM) program was used for population PK model-building and a two-compartment model with first-order absorption process was established. Creatinine clearance (CL(cr)) and body weight were identified as intrinsic factors which significantly affected oral clearance (CL(t)/F) and the apparent volume of distribution of the central compartment (V1/F), respectively. The final model is described as follows: CL(t)/F (l/h) = (8.97 + 0.917 x (CL(cr) (ml/min)-100.92) x 60/1000) x exp (eta(CLt/F)). V1/F (l) = (85.3 + 1.22 x (weight (kg)-60.75)) x exp (eta(V1/F)). Q/F (l/h) = 0.351. V2/F (l) = 6.81. k(a) (h(-1)) = 1.44 x exp(eta(ka)). Based on the population PK model, mean C(max) and AUC(0-24h) in CALRTI patients were estimated as 5.13 microg/ml and 58.98 microg.h/ml, respectively. A subgroup analysis showed that patients with mild renal dysfunction (50 ml/min < or = CL(cr) < 80 ml/min) had 34% higher AUC(0-24h) values compared to patients with normal renal function (CL(cr) > or = 80 ml/min). Postmodeling simulation using final population PK estimates also showed that C(max) and AUC(0-24h) increased markedly in patients with severe renal dysfunction. The results indicate that LVFX dosage adjustment should be individualized on the basis of the CL(cr), especially in those with CL(cr) less than 50 ml/min. None of the PK parameters had any correlation with the occurrence of adverse events. PK-PD analysis indicated that, in patients treated with LVFX 500 mg once daily, the AUC(0-24h)/MIC ratio exceeded the target for those major CALRTI pathogens isolated. In addition, the C(max)/MIC ratio reached 5 for Streptococcus pneumoniae, indicating that the emergence of LVFX-resistant S. pneumoniae could be prevented during the therapy with this dosage regimen. These results demonstrate that oral LVFX 500 mg once daily has favorable PK parameters and PK-PD features in patients with CALRTIs, and the results strongly support this dosage regimen for the treatment of CALRTI.

Population pharmacokinetics study of remifentanil in Chinese adult patients determined by an LC-MS/MS method

The aim of this study was to construct the population pharmacokinetic model in Chinese adult patients and to characterize the factors that affect the parameters of remifentanil pharmacokinetics. 15 patients were divided randomly into two groups: Index (modeling) group and validation group. 10 patients including 99 blood samples were in the Index group and 5 patients including 56 blood samples were in the Validation group. A LC-MS/MS method was developed to determine the whole blood concentration of remifentanil. Meanwhile, population modeling was performed using the NONMEM (nonlinear mixed-effect modeling) program with a 2-compartment pharmacokinetic model. The forward inclusion-backward elimination method was used to investigate the different covariates, including age, height, total bilirubin, etc. Liquid-liquid extraction was used to extract remifentanil from the blood. The calibration curve was linear from 0.5 to 50 ng/ml and the lowest limit of quantification (LOQ) was 0.5 ng/ml. The analytical method was rapid, selective and highly sensitive. The population pharmacokinetic results indicated that total bilirubin, age, alkaline phosphatase activity and gender significantly affected the parameters of remifentanil. The estimation of CL1, V1, CL2 and V2 were 1.85 l/min, 8.47 l, 1.28 l/min and 29.7 l, respectively. The population model was validated by another group of patients indicating that the model was effective and robust. The LC-MS/MS method for the quantification of remifentanil in human whole blood was rapid, selective and highly sensitive. The population model was acceptable and would be helpful for clinicians to assess the remifentanil pharmacokinetic parameters based on patient's specific demographic characteristics.

Population Pharmacokinetics of Lamotrigine with Data from Therapeutic Drug Monitoring in German and Spanish Patients with Epilepsy

This study develops a population pharmacokinetic model for lamotrigine (LTG) in Spanish and German patients diagnosed with epilepsy. LTG steady-state plasma concentration data from therapeutic drug monitoring were collected retrospectively from 600 patients, with a total of 1699 plasma drug concentrations. The data were analyzed according to a one-compartment model using the nonlinear mixed effect modelling program. The influences of origin (Germany or Spain), sex, age, total body weight, and comedication with valproic acid (VPA), levetiracetam, and enzyme-inducing antiepileptic drugs (phenobarbital [PB], phenytoin [PHT], primidone [PRM], and carbamazepine [CBZ]) were investigated using step-wise generalized additive modelling. The final regression model for LTG clearance (CL) was as follows: CL(L/h) = 0.028*total body weight*e(-0.713*VPA)*e0.663*PHT*e0.588*(PB or PRM)*e0.467*CBZ*e0.864*IND, where IND refers to two or more inducers added to LTG treatment; this factor as well as VPA, PHT, PB, PRM, and CBZ take a value of zero or one according to their absence or presence, respectively. The administration of inducers led to a significant increase in mean LTG CL (values of 0.045-0.070 L/h/kg vs. 0.028 L/h/kg being reached in monotherapy), whereas VPA led to a significant decrease in CL (0.014 L/h/kg). Thus, comedication with these analyzed drugs can partly explain the interindividual variability in population LTG CL, which decreased from the basic model by more than 40%. The proposed model may be very useful for clinicians in establishing initial LTG dosage guidelines. However, the interindividual variability remaining in the final model (clearance coefficient of variation close to 30%) make these a priori dosage predictions imprecise and justifies the need for LTG plasma level monitoring to optimize dosage regimens. Thus, this final model allows easy implementation in clinical pharmacokinetic software and its application in dosage individualization using the Bayesian approach.

Population Estimation Regarding the Effects of Cytochrome P450 2C19 and 3A5 Polymorphisms on Zonisamide Clearance

We aimed to evaluate the effects of cytochrome P450 (CYP) 2C19 and CYP3A5 polymorphisms on zonisamide (ZNS) clearance. The pharmacokinetics of the 282 ZNS concentrations at a steady state obtained from 99 Japanese epileptic patients was performed with a nonlinear mixed-effect modeling program, using a one-compartment open pharmacokinetic model with first-order elimination. The covariates screened included the total body weight, gender, ZNS daily dose, CYP2C19 and CYP3A5 genotypes, and the coadministered antiepileptic drugs. The final model of ZNS apparent clearance was as follows: CL = 1.22 x (BW/44)0.77 x DOSE(-0.17 x 0.84CYP2C19 hetero EM x 0.70CYP2C19 PM x 1.24CBZ x 1.28PHT x 1.29PB x eetaCL where CL is the apparent oral clearance of ZNS, DOSE is ZNS daily dose, and CYP2C19 heterozygous extensive metabolizer (EM) or CYP2C19 poor metabolizer (PM) is equal to 1 if one or two CYP2C19-defective alleles are carried, respectively; otherwise, it is 0. Carbamazepine (CBZ), phenytoin (PHT), or phenobarbital (PB) is equal to 1 if carbamazepine, phenytoin, or phenobarbital is coadministered, respectively; otherwise, it is 0. etaCL is the independent random error distributed normally with the mean zero and variance equal to omegaCL. The CL of ZNS was lower in the CYP2C19 heterozygous extensive metabolizers and poor metabolizers than in the homozygous extensive metabolizers by 16% and 30%, respectively (P < 0.001). An effect of CYP3A5 polymorphisms was not identified. The coadministration of carbamazepine, phenytoin, or phenobarbital increased the CL of ZNS by 24% to 29%. This report demonstrates that the CYP2C19 genotype affects the ZNS metabolism in Japanese epileptic subjects. The clinical relevance of these changes remains to be explored in future studies.

Population pharmacokinetics of cefepime in neonates with severe nosocomial infections

To define the pharmacokinetic behaviour of cefepime in neonates with severe nosocomial infections using a mixed effects model. Thirty-one newborn infants were included in the study; 10 additional infants participated in the validation of the pharmacokinetic model. Cefepime CL and V were determined using an open monocompartmental model with first-order elimination. The influence of demographic and clinical characteristics on the model was evaluated. The non-linear mixed effect model (nonmem) program was used to determine the pharmacokinetic population model. The mean corrected gestational age for infants participating in the construction and validation of the model were 35 and 33 weeks, respectively. Factors included in the final pharmacokinetic model were body surface area (BSA) and calculated CL(CR). The final population model was CL (L/h) = 0.457 BSA (m(2)) + 0.243 CL(CR) (L/h) and V(L) = 4.12 BSA (m(2)). This model explains 33.3% of the interindividual variability for CL and 12.8% for V. This model was validated in ten neonates with nosocomial infections by assessing the predictive capacity of plasma cefepime concentrations using a priori and Bayesian strategies. The predictive performance of this population model for cefepime plasma concentrations was adequate for clinical purposes and can be used for individualizing cefepime therapy in newborn infants with severe infections. Cefepime plasma concentrations can be predicted based on BSA and calculated CL(CR). Cefepime therapy using a 250 mg/m(2) dose administered every 12 h is adequate to achieve plasma concentrations greater than 8 mug/mL during more than 60% of the dosing interval and is expected to be effective in the treatment of bloodstream infections caused by most gram negative organisms in newborn infants. A dose of 550 mg/m(2) would be required for the treatment of infections caused by Pseudomonas sp.

Nonlinear Mixed Effects Model Analysis of the Pharmacokinetics of Aripiprazole in Healthy Japanese Males

The population pharmacokinetic parameters of aripiprazole in healthy Japanese males were estimated using a nonlinear mixed effects model (NONMEM) program. Pharmacokinetic data for population analysis were obtained from the single-dose (24 subjects), multiple-dose (15 subjects), and itraconazole-coadministration (27 subjects) trials. The time course of plasma aripiprazole concentration following oral administration was well described by a two-compartment model with first-order input. The mean values of the absorption lag time (ALAG) and absorption rate constant (KA) were estimated to be 0.805 h and 2.65 h(-1), respectively. The mean volume of the central (V(1)/F) and peripheral (V(2)/F) compartment was 3.84 and 1.54 l/kg, respectively, and the mean value of inter-compartment clearance (Q/F) was 0.168 l/h/kg. Oral clearance (CL/F) was estimated to be 0.0645 l/h/kg in the group with CYP2D6*1/*1, *1/*2 and *2/*2. The decrease in CL/F was estimated to be 0.0135 l/h/kg in the group with CYP2D6*1/*5, *1/*10, *2/*5, *2/*10, and *2/*41, and 0.0293 l/h/kg in the group with CYP2D6*5/*10, *10/*10, and *41/*41. The plasma concentration of aripiprazole was increased by coadministration of itraconazole, and the decrease in CL/F was estimated to be 0.0181 l/h/kg.

Population Analysis of Weight-, Age-, and Sex-Related Differences in the Pharmacokinetics of Lopinavir in Children from Birth to 18 Years

The pharmacokinetics of lopinavir were investigated by the use of a population approach performed with the nonlinear mixed effect modeling program NONMEM and 157 children ranging in age from 3 days to 18 years. The pharmacokinetics of lopinavir were well described by a one-compartment model in which the absorption and the elimination rate constants were equal. Typical population estimates of the apparent volume of distribution (V/F) and plasma clearance (CL/F) were 24.6 liters and 2.58 liters/h, respectively. The lopinavir V/F and CL/F were both related to body weight (BW), with an important increase in weight-normalized CL/F for the lowest BW. Combined treatment with lopinavir and nevirapine was found to increase the CL/F. The lopinavir CL/F was also age and sex related, as a 39% increase was observed after the age of 12 years for boys compared to the CL/F for girls. The consequences of these pharmacokinetic discrepancies and the necessity to modify the currently recommended dosage regimen should be further investigated.

GFR is better estimated by considering both serum cystatin C and creatinine levels

Serum cystatin C (cysC) is a potential marker of the glomerular filtration rate (GFR) that has generated conflicting reports in children. A prospective study was conducted to assess the benefit of considering cysC together with serum creatinine (SCr) and demographic and morphologic characteristics to better estimate the 51Cr-ethylenediaminetetraacetate (EDTA) clearance (CL), i.e., the GFR. Plasma 51Cr-EDTA data from 100 children or young adults (range: 1.4-22.8 years old) were analyzed according to the population pharmacokinetic approach by using the nonlinear mixed effects model (NONMEM) program. The actual CL was compared to the CL predicted according to different covariate equations. The best covariate equation (+/-95% confidence interval) was: GFR (ml/min)=63.2(+/-3.4) . [(SCr (microM)/96)(-0.35 (+/-0.20))] . [(cysC (mg/l)/1.2)(-0.56 (+/-0.19))] . [(body weight (kg)/45)(0.30 (+/-0.17))] . [age (years)/14)(0.40 (+/-0.16))]. This equation was associated with a less biased and more precise estimation than the Schwartz equation. CysC improves the estimation of the GFR in children if considered with other covariates within the mathematical formula.

Population pharmacokinetics of intravenous busulfan in patients undergoing hematopoietic stem cell transplantation

A population pharmacokinetic analysis was performed in 30 patients who received an intravenous busulfan and cyclophosphamide regimen before hematopoietic stem cell transplantation. Each patient received 0.8 mg/kg as a 2 h infusion every 6 h for 16 doses. A total of 690 concentration measurements were analyzed using the nonlinear mixed effect model (NONMEM) program. A one-compartment model with an additive error model as an intraindividual variability including an interoccasion variability (IOV) in clearance (CL) was sufficient to describe the concentration-time profile of busulfan. Actual body weight (ABW) was found to be the determinant for CL and the volume of distribution (V) according to NONMEM analysis. In this limited study, the age (range 7-53 years old; median, 30 years old) had no significant effect on busulfan pharmacokinetics. For a patient weighting 60 kg, the typical CL and V were estimated to be 8.87 l/h and 33.8 l, respectively. The interindividual variability of CL and V were 13.6 and 6.3%, respectively. The IOV (6.6%) in CL was estimated to be less than the intraindividual variability. These results indicate high interpatient and intrapatient consistency of busulfan pharmacokinetics after intravenous administration, which may eliminate the requirement for pharmacokinetic monitoring.

Nonlinear Mixed Effects Model Analysis of the Pharmacokinetics of Routinely Administered Bepridil in Japanese Patients with Arrhythmias

This study was performed to evaluate variability in the pharmacokinetics of bepridil in 38 Japanese patients with arrhythmias, and to investigate the effects of aprindine as well as CYP2D6 and CYP3A5 polymorphisms on the oral clearance of bepridil. We determined the polymorphic alleles of CYP2D6 and CYP3A5 in each subject. The plasma concentration of bepridil at steady-state following repetitive oral administration was measured with an HPLC-based method, and the oral clearance was estimated using the nonlinear mixed effects model (NONMEM) program. Mean oral clearance was significantly greater in the patients with the CYP2D6*10 allele than in those without it. On the other hand, no significant effect of the CYP3A5 polymorphism was observed on the pharmacokinetics of bepridil. In addition, aprindine seemed to reduce the oral clearance of bepridil in the patients with the CYP2D6*10 allele.

Population Pharmacokinetics of Higher-Dose Mizoribine in Healthy Male Volunteers

The population pharmacokinetic parameters of mizoribine in healthy subjects were estimated using a nonlinear mixed effects model (NONMEM) program. Pharmacokinetic data for population analysis were obtained in the previous study, in which 24 healthy Caucasian male subjects participated in a single-dose (3, 6, 9, 12 mg/kg) study, and 12 subjects participated in a multiple-dose (6, 12 mg/kg/d) study. The mean value of the absorption lag time, absorption rate constant (KA), and apparent distribution volume (V/F) was estimated to be 0.349 h, 0.869 h-1, and 0.834 l/kg, respectively. Oral clearance (CL/F) was modeled with creatinine clearance (CLcr), and the mean value was estimated to be 1.93.CLcr l/h. In addition, pharmacokinetic parameters in individual 36 subjects were obtained from population estimates according to Bayes' theorem. Pharmacokinetic parameters (KA, V/F, and CL/F) in the single-dose study were almost constant at a dose range of 3-12 mg/kg, and were similar to those in the multiple-dose study. These findings indicated that the pharmacokinetics of mizoribine is well described by a simple one-compartment model with first-order absorption.

Multiple Regression Analysis of Pharmacogenetic Variability of Carvedilol Disposition in 54 Healthy Japanese Volunteers

The aim of this study was to evaluate the pharmacogenetic variability in the disposition of carvedilol in the Japanese population. Five or 10 mg of carvedilol was orally administered to 54 healthy Japanese subjects (22-44 years old), and blood samples were taken at 2 and 6 h after dosing. We determined the polymorphic alleles of CYP2D6, CYP2C9, CYP2C19, CYP3A5, UGT2B7, and MDR1 in each subject. The whole blood concentration of R- and S-carvedilol was measured by an HPLC method. The pharmacokinetic parameters in individual subjects were estimated by the Bayesian method using the nonlinear mixed effects model (NONMEM) program. We then examined the effect of the genetic polymorphisms on the variability in the pharmacokinetics of carvedilol using a multiple regression analysis. The oral clearance (CL/F) and also apparent volume of distribution (V/F) of both enantiomers were significantly lower in the subjects with the CYP2D6*10 allele than those with the CYP2D6*1/*1, *1/*2, or *2/*2 genotype, confirming our previous finding that the bioavailability (F) and systemic clearance (CL) of R- and S-carvedilol in the liver is significantly altered in Japanese with the CYP2D6*10 allele. On the other hand, CYP2C9*3, CYP2C19*2, CYP2C19*3, CYP3A5*3, UGT2B7*2, and MDR1 C3435T did not significantly affect the pharmacokinetics of carvedilol in Japanese subjects.

Patient Characteristics Influencing Ciclosporin Pharmacokinetics and Accurate Bayesian Estimation of Ciclosporin Exposure in Heart, Lung and Kidney Transplant Patients

Population pharmacokinetic studies of ciclosporin microemulsion are needed to identify the individual factors influencing ciclosporin pharmacokinetic variability in transplant patients and to design efficient tools for the accurate estimation of ciclosporin overall exposure (area under the plasma concentration-time curve from 0 to 12 hours [AUC12]). In the present retrospective study, a large database of heart, lung (with or without cystic fibrosis) and kidney (both adult and paediatric) transplant patients receiving ciclosporin microemulsion was analysed with the aims of (i) building a population pharmacokinetic model and finding the main covariates linked with ciclosporin microemulsion pharmacokinetic parameters; and (ii) developing a maximum a posteriori probability Bayesian estimator (MAP-BE) to estimate ciclosporin microemulsion pharmacokinetic parameters using a limited-sampling strategy. 3,072 concentration data from 147 patients (i.e. 309 full pharmacokinetic profiles) were analysed using the nonlinear mixed-effects model program NONMEM. The influence of numerous covariates was tested, and the final model was validated by data splitting. For Bayesian estimation, the best limited-sampling strategy was determined based on the D-optimality criterion, and validation performed in an independent group of 60 patients. The pharmacokinetics of ciclosporin microemulsion were accurately described by a two-compartment model with Erlang distribution for the absorption process. The type of graft and post-transplantation period were identified as significant sources of variability of the absorption parameter. Both apparent volume of the central compartment after oral administration (V1/F) and apparent oral clearance (CL/F) increased with bodyweight. The best limited-sampling strategy for Bayesian estimation was 0 hour, 1 hour and 3 hour post-dose, providing accurate estimation of ciclosporin microemulsion AUC12 in all patients of the test group, with a mean bias of 2.0 +/- 10.5% (range: -19.1% to -21.4% and 95% CI -0.6, +4.7). Population pharmacokinetic analysis of ciclosporin microemulsion in allograft transplants resulted in the design of a new pharmacokinetic model for ciclosporin microemulsion, identification of significant covariates and the design of an accurate MAP-BE based on three blood concentrations and these covariates.