Search

Background and Objectives: Ganciclovir (GCV) and valganciclovir (VGCV) show large interindividual pharmacokinetic variability, particularly in children. The objectives of this study were (1) to develop machine learning (ML) algorithms trained on simulated pharmacokinetics profiles obtained by Monte Carlo simulations to estimate the best ganciclovir or valganciclovir starting dose in children and (2) to compare its performances on real-world profiles to previously published equation derived from literature population pharmacokinetic (POPPK) models achieving about 20% of profiles within the target. Materials and Methods: The pharmacokinetic parameters of four literature POPPK models in addition to the World Health Organization (WHO) growth curve for children were used in the mrgsolve R package to simulate 10,800 pharmacokinetic profiles. ML algorithms were developed and benchmarked to predict the probability to reach the steady-state, area-under-the-curve target (AUC0–24 within 40–60 mg × h/L) based on demographic characteristics only. The best ML algorithm was then used to calculate the starting dose maximizing the target attainment. Performances were evaluated for ML and literature formula in a test set and in an external set of 32 and 31 actual patients (GCV and VGCV, respectively). Results: A combination of Xgboost, neural network, and random forest algorithms yielded the best performances and highest target attainment in the test set (36.8% for GCV and 35.3% for the VGCV). In actual patients, the best GCV ML starting dose yielded the highest target attainment rate (25.8%) and performed equally for VGCV with the Franck model formula (35.3% for both). Conclusion: The ML algorithms exhibit good performances in comparison with previously validated models and should be evaluated prospectively. [ABSTRACT FROM AUTHOR]

Background and Objective: Several drugs on the market are substrates for P-glycoprotein (P-gp), an efflux transporter highly expressed in barrier tissues such as the intestine. Body weight, weight loss, and a Roux-en-Y gastric bypass (RYGB) may influence P-gp expression and activity, leading to variability in the drug response. The objective of this study was therefore to investigate digoxin pharmacokinetics as a measure of the P-gp phenotype in patients with obesity before and after weight loss induced by an RYGB or a strict diet and in normal weight individuals. Methods: This study included patients with severe obesity preparing for an RYGB (n = 40) or diet-induced weight loss (n = 40) and mainly normal weight individuals scheduled for a cholecystectomy (n = 18). Both weight loss groups underwent a 3-week low-energy diet (<1200 kcal/day) followed by an additional 6 weeks of <800 kcal/day induced by an RYGB (performed at week 3) or a very-low-energy diet. Follow-up time was 2 years, with four digoxin pharmacokinetic investigations at weeks 0, 3, and 9, and year 2. Hepatic and jejunal P-gp levels were determined in biopsies obtained from the patients undergoing surgery. Results: The RYGB group and the diet group had a comparable weight loss in the first 9 weeks (13 ± 2.3% and 11 ± 3.6%, respectively). During this period, we observed a minor increase (16%) in the digoxin area under the concentration–time curve from zero to infinity in both groups: RYGB: 2.7 µg h/L [95% confidence interval (CI) 0.67, 4.7], diet: 2.5 µg h/L [95% CI 0.49, 4.4]. In the RYGB group, we also observed that the time to reach maximum concentration decreased after surgery: from 1.0 ± 0.33 hours at week 3 to 0.77 ± 0.08 hours at week 9 (−0.26 hours [95% CI −0.47, −0.05]), corresponding to a 25% reduction. Area under the concentration–time curve from zero to infinity did not change long term (week 0 to year 2) in either the RYGB (1.1 µg h/L [−0.94, 3.2]) or the diet group (0.94 µg h/L [−1.2, 3.0]), despite a considerable difference in weight loss from baseline (RYGB: 30 ± 7%, diet: 3 ± 6%). At baseline, the area under the concentration–time curve from zero to infinity was −5.5 µg h/L [95% CI −8.5, −2.5] (−26%) lower in patients with obesity (RYGB plus diet) than in normal weight individuals scheduled for a cholecystectomy. Further, patients undergoing an RYGB had a 0.05 fmol/µg [95% CI 0.00, 0.10] (29%) higher hepatic P-gp level than the normal weight individuals. Conclusions: Changes in digoxin pharmacokinetics following weight loss induced by a pre-operative low-energy diet and an RYGB or a strict diet (a low-energy diet plus a very-low-energy diet) were minor and unlikely to be clinically relevant. The lower systemic exposure of digoxin in patients with obesity suggests that these patients may have increased biliary excretion of digoxin possibly owing to a higher expression of P-gp in the liver. [ABSTRACT FROM AUTHOR]

Background and Objective: Midazolam is the preferred clinical probe drug for assessing CYP3A activity. We have previously shown substantial intraindividual variability in midazolam absolute bioavailability and clearance in patients with obesity before and after weight loss induced by gastric bypass or a strict diet. The objective was to describe intraindividual variability in absolute bioavailability and clearance of midazolam in healthy individuals without obesity. Methods: This study included 33 healthy volunteers [28 ± 8 years, 21% males, body mass index (BMI) 23 ± 2.5 kg/m2] subjected to four pharmacokinetic investigations over a 2-month period (weeks 0, 2, 4, and 8). Semi-simultaneous oral (0 h) and intravenous (2 h later) midazolam dosing was used to assess absolute bioavailability and clearance of midazolam. Results: At baseline, mean absolute bioavailability and clearance were 46 ± 18% and 31 ± 10 L/h, respectively. The mean coefficient of variation (CV, %) for absolute bioavailability and clearance of midazolam was 26 ± 15% and 20 ± 10%, respectively. Approximately one-third had a CV > 30% for absolute bioavailability, while 13% had a CV > 30% for clearance. Conclusions: On average, intraindividual variability in absolute bioavailability and clearance of midazolam was low to moderate; however, especially absolute bioavailability showed considerable variability in a relatively large proportion of the individuals. [ABSTRACT FROM AUTHOR]

Introduction: Rosuvastatin pharmacokinetics is mainly dependent on the activity of hepatic uptake transporter OATP1B1. In this study, we aimed to investigate and disentangle the effect of Roux-en-Y gastric bypass (RYGB) and weight loss on oral clearance (CL/F) of rosuvastatin as a measure of OATP1B1-activity. Methods: Patients with severe obesity preparing for RYGB (n = 40) or diet-induced weight loss (n = 40) were included and followed for 2 years, with four 24-hour pharmacokinetic investigations. Both groups underwent a 3-week low-energy diet (LED; < 1200 kcal/day), followed by RYGB or a 6-week very-low-energy diet (VLED; < 800 kcal/day). Results: A total of 80 patients were included in the RYGB group (40 patients) and diet-group (40 patients). The weight loss was similar between the groups following LED and RYGB. The LED induced a similar (mean [95% CI]) decrease in CL/F in both intervention groups (RYGB: 16% [0, 31], diet: 23% [8, 38]), but neither induced VLED resulted in any further changes in CL/F. At Year 2, CL/F had increased by 21% from baseline in the RYGB group, while it was unaltered in the diet group. Patients expressing the reduced function SLCO1B1 variants (c.521TC/CC) showed similar changes in CL/F over time compared with patients expressing the wild-type variant. Conclusions: Neither body weight, weight loss nor RYGB per se seem to affect OATP1B1 activity to a clinically relevant degree. Overall, the observed changes in rosuvastatin pharmacokinetics were minor, and unlikely to be of clinical relevance. [ABSTRACT FROM AUTHOR]

Background: Maximum a posteriori Bayesian estimation (MAP-BE) based on a limited sampling strategy and a population pharmacokinetic model is frequently used to estimate pharmacokinetic parameters in individuals, however with some uncertainty (bias). Recent works have shown that the performance in individual estimation or pharmacokinetic parameters can be improved by combining population pharmacokinetic and machine learning algorithms. Objective: The objective of this work was to investigate the use of a hybrid machine learning/population pharmacokinetic approach to improve individual iohexol clearance estimation. Methods: The reference iohexol clearance values were derived from 500 simulated profiles (samples collected between 0.1 and 24.7 h) using a population pharmacokinetic model we recently developed in Monolix and obtained using all the concentration timepoints available. Xgboost and glmnet algorithms able to predict the error of MAP-BE clearance estimates based on a limited sampling strategy (0.1 h, 1 h, and 9 h) versus reference values were developed in a training subset (75%) and were evaluated in a testing subset (25%) and in 36 real patients. Results: The MAP-BE limited sampling strategy estimated clearance was corrected by the machine learning predicted error leading to a decrease in root mean squared error by 29% and 24%, and in the percentage of profiles with the mean prediction error out of the ± 20% bias by 60% and 40% in the external validation dataset for the glmnet and Xgboost machine learning algorithms, respectively. These results were attributable to a decrease in the eta-shrinkage (shrinkage for a MAP-BE limited sampling strategy = 32.4%, glmnet = 18.2%, and Xgboost = 19.4% in the external dataset). Conclusions: In conclusion, this hybrid algorithm represents a significant improvement in comparison to MAP-BE estimation alone. [ABSTRACT FROM AUTHOR]

Background and Objective: Given a high pharmacokinetic inter-individual variability and a low exposure target achievement, ganciclovir (GCV) therapeutic drug monitoring is sometimes used in children. We aimed to develop and validate Bayesian estimators based on limited sampling strategies for the estimation of GCV area under the concentration–time curve from 0 to 24 h in pediatric transplant recipients treated with valganciclovir (VGCV) or GCV. Methods: Solid organ transplant or stem-cell transplant recipients who received GCV or VGCV and had available GCV concentrations per standard of care were retrospectively included in this study for pharmacokinetic modeling and development of Bayesian estimators using the iterative two-stage Bayesian method. Validation datasets included additional child recipients of a solid organ transplant or stem-cell transplant, and child recipients of a kidney or liver transplant enrolled in a previous study. Various combinations of three or two sampling times, applicable in clinical practice, were assessed based on the relative mean bias, standard deviation, and the root mean square error in a development dataset and three independent validation datasets. Results: In the development dataset, the mean bias/standard deviation/root mean square error for the 1 h/2 h/3 h and 1 h/3 h limited sampling strategies were − 1.4%/9.3%/9.1% and − 3.5%/12.2%/12.3%, respectively for GCV, while for VGCV, the mean bias/standard deviation/root mean square error for the 1 h/2 h/6 h and 1 h/6 h limited sampling strategies were 0.7%/13.5%/13.3% and − 0.1%/12.1%/11.8%, respectively. In the independent validation datasets, seven (13%) and five (14%) children would have had misclassifications of their exposure using these Bayesian estimators and limited sampling strategies for VGCV and GCV, respectively. Conclusions: Three plasma samples collected at 1 h/2 h/3 h and 1 h/2 h/6 h post-dose for GCV and VGCV respectively, are sufficient to accurately determine GCV area under the concentration–time curve from 0 to 24 h for pharmacokinetic-enhanced therapeutic drug monitoring. [ABSTRACT FROM AUTHOR]

Background and Objective: Intracellular exposure of everolimus may be a better marker of therapeutic effect than trough whole blood concentrations. We aimed to develop pharmacokinetic population models and Bayesian estimators based on a limited sampling strategy for estimation of dose interval exposures of everolimus in whole blood and peripheral blood mononuclear cells (PBMCs) in renal transplant recipients.Methods: Full whole blood and PBMC concentration-time profiles of everolimus were obtained from 12 stable renal transplant recipients on two different occasions, 4 weeks apart. The dataset was treated as 24 individual profiles and split into a development dataset (n = 20) and a validation dataset (n = 4). The pharmacokinetic model was developed using non-parametric modeling and its performances and those of the derived Bayesian estimator were evaluated in the validation set.Results: A structural two-compartment model with first-order elimination and two absorption phases described by a sum of two gamma distributions were developed. None of the tested covariates (age, sex, albumin, hematocrit, fat-free mass and genetic variants such as CYP3A5*1, ABCB1 haplotype, PPARA*42, PPARA*48, and POR*28) were retained in the final model. A limited sampling schedule of two whole blood samples at 0 and 1.5 h and one PBMC sample at 1.5 h post dose provided accurate estimates of the area under the plasma concentration-time curve (AUC) in comparison with the trapezoidal reference AUC (relative bias ± standard deviation = - 3.9 ± 10.6 and 4.1 ± 12.3% for whole blood and PBMC concentrations, respectively).Conclusion: The developed model allows simultaneous and accurate prediction of everolimus exposure in whole blood and PBMCs, and supplies a base for a feasible exploration of the relationships between intracellular exposure and therapeutic effects in prospective trials. [ABSTRACT FROM AUTHOR]

Background and Objective: The pharmacokinetic disposition of ciclosporin shows great intra- and interpatient variability, and that combined with a narrow therapeutic window makes therapeutic drug monitoring of ciclosporin necessary. The nonlinear mixed-effects population pharmacokinetic program NONMEM® predicts individual pharmacokinetic parameters based not only on individual patient observations but also on population characteristics and the patient's covariates. The aim of this model development is to potentially use it in the clinical setting to optimize ciclosporin dosing in renal transplant recipients. Methods: A population pharmacokinetic model of ciclosporin has been developed with NONMEM® using full 12-hour pharmacokinetic profiles from 29 renal transplant recipients, 3 months of daily follow-up data from an additional 11 recipients, and both 3 months of follow-up data and full 12-hour pharmacokinetic profiles from nine patients. The internal validation of the model was based on data splitting and jack-knife methods. In addition, the model was validated for its clinical applicability on standard trough and 2-hour post-dose concentration data from 12 additional patients with 3 months of follow-up. Results: The model that best described the ciclosporin data was a two-compartment model with first-order absorption process with lagged time. The population pharmacokinetic parameters were oral clearance (CL/F) = 26.9L/h; central volume of distribution after oral administration (V1/F) = 24.4 L; absorption rate constant (ka) = 0.544 h-1; lag time = 0.460 h; peripheral volume of distribution = 1119L and intercompartmental clearance after oral administration (Q/F) = 19.6L/h. Three covariates had significant effect on a total of six pharmacokinetic parameters. These were bodyweight onV1/Fand ka, time after transplantation on ka, and age on CL/F, ka and V1/F. Cytochrome P450 3A5 genotype was also a significant covariate but was not included in the final model since such information is not available in clinical practice. The external validation showed that the model was able to predict ciclosporin concentrations in the 12 new patients with an average predictive error of 17.4 ± 14% when the standard sample concentrations from the previous week were given. Conclusion: A NONMEM® pharmacokinetic model for ciclosporin in renal transplant recipients was successfully developed and validated for the first 3 months post-transplantation. The model showed good predictability in a new patient cohort. After further clinical validation, the model may be applicable as a clinical tool for optimizing ciclosporin dosing in renal transplant recipients in the early post-transplant period. [ABSTRACT FROM AUTHOR]