Predictive performance of a physiologically based pharmacokinetic model of busulfan in children

Predicting busulfan exposure in patients undergoing hematopoietic stem cell transplantation using machine learning techniques

PURPOSE

This study aimed to establish an optimal model to predict the busulfan (BU) area under the curve at steady state (AUCss) by using machine learning (ML).

PATIENTS AND METHODS

Seventy-nine adult patients (age ≥18 years) who received BU intravenously and underwent therapeutic drug monitoring from 2013 to 2021 at Fujian Medical University Union Hospital were enrolled in this retrospective study. The whole dataset was divided into a training group and test group at the ratio of 8:2. BU AUCss were considered as the target variable. Nine different ML algorithms and one population pharmacokinetic (pop PK) model were developed and validated, and their predictive performance was compared.

RESULTS

All ML models were superior to the pop PK model (R2 = 0.751, MSE = 0.722, 14 and RMSE = 0.830) in model fitting and had better predictive accuracy. The ML model of BU AUCss established through support vector regression (SVR) and gradient boosted regression trees (GBRT) had the best predictive ability (R2 = 0.953 and 0.953, MSE = 0.323 and 0.326, and RMSE = 0.423 and 0.425).

CONCLUSION

All the ML models can potentially be used to estimate BU AUCss with the aim of facilitating rational use of BU on the individualized level, especially models built by SVR and GBRT algorithms.

Ontogeny of Drug-Metabolizing Enzymes

Almost 50% of prescription drugs lack age-appropriate dosing guidelines and therefore are used "off-label." Only ~10% drugs prescribed to neonates and infants have been studied for safety or efficacy. Immaturity of drug metabolism in children is often associated with drug toxicity. This chapter summarizes data on the ontogeny of major human metabolizing enzymes involved in oxidation, reduction, hydrolysis, and conjugation of drugs. The ontogeny data of individual drug-metabolizing enzymes are important for accurate prediction of drug pharmacokinetics and toxicity in children. This information is critical for designing clinical studies to appropriately test pharmacological hypotheses and develop safer pediatric drugs, and to replace the long-standing practice of body weight- or surface area-normalized drug dosing. The application of ontogeny data in physiologically based pharmacokinetic model and regulatory submission are discussed.

Improved outcomes using unmanipulated haploidentical hematopoietic stem cells combined with third-party umbilical cord blood transplantation for non-malignant diseases in children: The experience of a single center

BACKGROUND

Unmanipulated haploid HSCT for SAA has resulted in improved outcomes over recent years. However, studies related to unmanipulated haploid HSCs combined with tp-UCB transplantation for other types of NMD are rare. Accordingly, we present the outcomes of 109 pediatric patients with life-threatening NMD undergoing unmanipulated haploid HSCs combined with tp-UCB transplantation.

PROCEDURE

We retrospectively investigated 109 pediatric patients with life-threatening NMD treated with unmanipulated haploid HSCs combined with tp-UCB transplantation in a single center.

RESULTS

The median days of neutrophil and platelet engraftment were +13 and +22 days, respectively. None of the cases experienced PGF. The incidence rates for grade I-II, III-IV aGVHD and cGVHD were 44.9%, 24.8%, and 9.3%, respectively. The incidence rates of CMV and EBV viremia were 46.7% and 39.4%, respectively. The median follow-up duration was 997 days. In total, 106 patients survived, including 104 cases with FFS and 2 cases with SGF. Three patients died. The 5-year TRM, OS, and FFS were 2.8%, 97.2%, and 96.2%, respectively.

CONCLUSION

The results of unmanipulated haploid HSCs combined with tp-UCB in pediatric patients with life-threatening NMD were promising. However, further research is now needed to determine specific factors that might influence the engraftment of HSCs.

Drug Exposure to Establish Pharmacokinetic-Response Relationships in Oncology

In the oncology field, understanding the relationship between the dose administered and the exerted effect is particularly important because of the narrow therapeutic index associated with anti-cancer drugs and the high interpatient variability. Therefore, in this review, we provide a critical perspective of the different methods of characterising treatment exposure in the oncology setting. The increasing number of modelling applications in oncology reflects the applicability and the impact of pharmacometrics on all phases of the drug development process and patient management as well. Pharmacometric modelling is a worthy component within the current paradigm of model-based drug development, but pharmacometric modelling techniques are also accessible for the clinician in the optimisation of current oncology therapies. Consequently, the application of population models in a hospital setting by generating close collaborations between physicians and pharmacometricians is highly recommended, providing a systematic means of developing and assessing model-based metrics as 'drivers' for various responses to treatments, which can then be evaluated as predictors for treatment success. Characterising the key determinants of variability in exposure is of particular importance for anticancer agents, as efficacy and toxicity are associated with exposure. We present the different strategies to describe and predict drug exposure that can be applied depending on the data available, with the objective of obtaining the most useful information in the patients' favour throughout the full drug cycle. Therefore, the objective of the present article is to review the different approaches used to characterise a patient's exposure to oncology drugs, which will result in a better understanding of the time course of the response and the magnitude of interpatient variability.

Prediction of Drug Clearance from Enzyme and Transporter Kinetics

Accurate estimation of in vivo clearance in human is pivotal to determine the dose and dosing regimen for drug development. In vitro-in vivo extrapolation (IVIVE) has been performed to predict drug clearance using empirical and physiological scalars. Multiple in vitro systems and mathematical modeling techniques have been employed to estimate in vivo clearance. The models for predicting clearance have significantly improved and have evolved to become more complex by integrating multiple processes such as drug metabolism and transport as well as passive diffusion. This chapter covers the use of conventional as well as recently developed methods to predict metabolic and transporter-mediated clearance along with the advantages and disadvantages of using these methods and the associated experimental considerations. The general approaches to improve IVIVE by use of appropriate scalars, incorporation of extrahepatic metabolism and transport and application of physiologically based pharmacokinetic (PBPK) models with proteomics data are also discussed. The chapter also provides an overview of the advantages of using such dynamic mechanistic models over static models for clearance predictions to improve IVIVE.

Utilization of Physiologically Based Pharmacokinetic Modeling in Clinical Pharmacology and Therapeutics: an Overview

The purpose of this review was to assess the advancement of applications for physiologically based pharmacokinetic (PBPK) modeling in various therapeutic areas. We conducted a PubMed search, and 166 articles published between 2012 and 2018 on FDA-approved drug products were selected for further review. Qualifying publications were summarized according to therapeutic area, medication(s) studied, pharmacokinetic model type utilized, simulator program used, and the applications of that modeling. The results showed a 13-fold increase in the number of papers published from 2012 to 2018, with the largest proportion of articles dedicated to the areas of infectious diseases, oncology, and neurology, and application extensions including prediction of drug-drug interactions due to metabolism and/or transporter-mediated effects and understanding drug kinetics in special populations. In addition, we profiled several high-impact studies whose results were used to guide package insert information and formulate dose recommendations. These results show that while utilization of PBPK modeling has drastically increased over the past several years, regulatory support, lack of easy-to-use systems for clinicians, and challenges with model validation remain major challenges for the widespread adoption of this practice in institutional and ambulatory settings. However, PBPK modeling will continue to be a useful tool in the future to assess therapeutic drug monitoring and the growing field of personalized medicine.

Toward precision medicine in pediatric population using cytochrome P450 phenotyping approaches and physiologically based pharmacokinetic modeling

The activity of drug-metabolizing enzymes (DME) shows high inter- and intra-individual variability. Genetic polymorphisms, exposure to drugs, and environmental toxins are known to significantly alter DME expression. In addition, the activity of these enzymes is highly age-dependent due to maturation processes that occur during development. Currently, there is a vast choice of phenotyping methods in adults using exogenous probes to characterize the activity of these enzymes. However, this can hardly be applied to children since it requires the intake of non-therapeutic xenobiotics. In addition, sampling may be challenging in the pediatric population for a variety of reasons: limited volume (e.g., blood), inappropriate sampling methods for age (e.g., urine), and metric requiring invasive or multiple blood samples. This review covers the main existing methods that can be used in the pediatric population to determine DME activity, with a particular focus on cytochrome P450 enzymes. Less invasive tools are described, including phenotyping using endogenous probes. Finally, the potential of pediatric model-informed precision dosing using physiologically based pharmacokinetic modeling is discussed.

Implementation of a Physiologically Based Pharmacokinetic Modeling Approach to Guide Optimal Dosing Regimens for Imatinib and Potential Drug Interactions in Paediatrics

Long-term use of imatinib is effective and well-tolerated in children with chronic myeloid leukaemia (CML) yet defining an optimal dosing regimen for imatinib in younger patients is a challenge. The potential interactions between imatinib and coadministered drugs in this "special" population also remains largely unexplored. This study implements a physiologically based pharmacokinetic (PBPK) modeling approach to investigate optimal dosing regimens and potential drug interactions with imatinib in the paediatric population. A PBPK model for imatinib was developed in the Simcyp Simulator (version 17) utilizing in silico, in vitro drug metabolism, and in vivo pharmacokinetic data and verified using an independent set of published clinical pharmacokinetic data. The model was then extrapolated to children and adolescents (aged 2-18 years) by incorporating developmental changes in organ size and maturation of drug-metabolising enzymes and plasma protein responsible for imatinib disposition. The PBPK model described imatinib pharmacokinetics in adult and paediatric populations and predicted drug interaction with carbamazepine, a cytochrome P450 (CYP)3A4 and 2C8 inducer, with a good accuracy (evaluated by visual inspections of the simulation results and predicted pharmacokinetic parameters that were within 1.25-fold of the clinically observed values). The PBPK simulation suggests that the optimal dosing regimen range for imatinib is 230-340 mg/m2/d in paediatrics, which is supported by the recommended initial dose for treatment of childhood CML. The simulations also highlighted that children and adults being treated with imatinib have similar vulnerability to CYP modulations. A PBPK model for imatinib was successfully developed with an excellent performance in predicting imatinib pharmacokinetics across age groups. This PBPK model is beneficial to guide optimal dosing regimens for imatinib and predict drug interactions with CYP modulators in the paediatric population.

Physiologically based pharmacokinetic modelling of atomoxetine with regard to CYP2D6 genotypes

Atomoxetine is a norepinephrine reuptake inhibitor indicated in the treatment of attention-deficit/hyperactivity disorder. It is primarily metabolized by CYP2D6 to its equipotent metabolite, 4-hydroxyatomoxetine, which promptly undergoes further glucuronidation to an inactive 4-HAT-O-glucuronide. Clinical trials have shown that decreased CYP2D6 activity leads to substantially elevated atomoxetine exposure and increase in adverse reactions. The aim of this study was to to develop a pharmacologically based pharmacokinetic (PBPK) model of atomoxetine in different CYP2D6 genotypes. A single 20 mg dose of atomoxetine was given to 19 healthy Korean individuals with CYP2D6*wt/*wt (*wt = *1 or *2) or CYP2D6*10/*10 genotype. Based on the results of this pharmacokinetic study, a PBPK model for CYP2D6*wt/*wt individuals was developed. This model was scaled to those with CYP2D6*10/*10 genotype, as well as CYP2D6 poor metabolisers. We validated this model by comparing the predicted pharmacokinetic parameters with diverse results from the literature. The presented PBPK model describes the pharmacokinetics after single and repeated oral atomoxetine doses with regard to CYP2D6 genotype and phenotype. This model could be utilized for identification of appropriate dosages of atomoxetine in patients with reduced CYP2D6 activity to minimize the adverse events, and to enable personalised medicine.

Utility of physiologically based pharmacokinetic (PBPK) modeling in oncology drug development and its accuracy: a systematic review

PURPOSE

Physiologically based pharmacokinetic (PBPK) modeling, a mathematical modeling approach which uses a pharmacokinetic model to mimick human physiology to predict drug concentration-time profiles, has been used for the discover and development of drugs in various fields, including oncology, since 2000. There have been a few general review articles on the utilization of PBPK in the development of oncology drugs, but these do not include an evaluation of model prediction accuracy. We therefore conducted a systematic review to define the accuracy of PBPK model prediction and its utility throughout all the developmental phases of oncology drugs.

METHODS

A systematic search was performed in the PubMed, PubMed Central and Cochrane Library databases from 1980 to February 2017 for articles (1) written in English, (2) focused on oncology or antineoplastic or anticancer drugs, tumor or cancer or anticancer drugs listed in the U.S. National Institutes of Health and (3) involving a PBPK model. The absolute-average-folding-errors (AAFEs) of the area under the curve (AUC) between predicted and observed values in each article were calculated to assess model prediction accuracy.

RESULTS

Of the 2341 articles initially identified by our search of the databases, 40 were included in the review analysis. These articles reported on six types of studies, i.e. in vivo (n = 4), first-in-human (n = 5), phase II/III clinical trials (n = 9), organ impairment (n = 3), pediatrics (n = 4) and drug-drug interactions (n = 15). AAFEs of the predicted AUC for all groups of studies were within 1.3-fold of each other despite variations in experimental methodologies.

CONCLUSION

PBPK modeling is a potential tool which can be effectively applied throughout all phases of oncology drug development. The number of experimental animals and human participants enrolled in the studies can be reduced using PBPK modeling and PBPK-population-PK modeling. The limited number of publications of unsuccessful model application to date may contribute to bias toward the usefulness of modeling.

GSTA1 diplotypes affect busulfan clearance and toxicity in children undergoing allogeneic hematopoietic stem cell transplantation: a multicenter study

Busulfan (BU) dose adjustment following therapeutic drug monitoring contributes to better outcome of hematopoietic stem cell transplantation (HSCT). Further improvement could be achieved through genotype-guided BU dose adjustments. To investigate this aspect, polymorphism within glutathione S transferase genes were assessed. Particularly, promoter haplotypes of the glutathione S transferase A1 (GSTA1) were evaluated in vitro, with reporter gene assays and clinically, in a pediatric multi-center study (N =138) through association with BU pharmacokinetics (PK) and clinical outcomes. Promoter activity significantly differed between the GSTA1 haplotypes (p<0.001) supporting their importance in capturing PK variability. Four GSTA1 diplotype groups that significantly correlated with clearance (p=0.009) were distinguished. Diplotypes underlying fast and slow metabolizing capacity showed higher and lower BU clearance (ml/min/kg), respectively. GSTA1 diplotypes with slow metabolizing capacity were associated with higher incidence of sinusoidal obstruction syndrome, acute graft versus host disease and combined treatment-related toxicity (p<0.0005). Among other GST genes investigated, GSTP1 313GG correlated with acute graft versus host disease grade 1-4 (p=0.01) and GSTM1 non-null genotype was associated with hemorrhagic cystitis (p=0.003). This study further strengthens the hypothesis that GST diplotypes/genotypes could be incorporated into already existing population pharmacokinetic models for improving first BU dose prediction and HSCT outcomes. (N^o Clinicaltrials.gov identifier: NCT01257854. Registered 8 December 2010, retrospectively registered).

Physiologically Based Pharmacokinetic Modeling in Pediatric Oncology Drug Development

Childhood cancer represents more than 100 rare and ultra-rare diseases, with an estimated 12,400 new cases diagnosed each year in the United States. As such, this much smaller patient population has led to pediatric oncology drug development lagging behind that for adult cancers. Developing drugs for pediatric malignancies also brings with it a number of unique trial design considerations, including flexible enrollment approaches, age-appropriate formulation, acceptable sampling schedules, and balancing the need for age-stratified dosing regimens, given the smaller patient populations. The regulatory landscape for pediatric pharmacotherapy has evolved with U.S. Food and Drug Administration (FDA) legislation such as the 2012 FDA Safety and Innovation Act. In parallel, regulatory authorities have recommended the application of physiologically based pharmacokinetic (PBPK) modeling, for example, in the recently issued FDA Strategic Plan for Accelerating the Development of Therapies for Pediatric Rare Diseases. PBPK modeling provides a quantitative and systems-based framework that allows the effects of intrinsic and extrinsic factors on drug exposure to be modeled in a mechanistic fashion. The application of PBPK modeling in drug development for pediatric cancers is relatively nascent, with several retrospective analyses of cytotoxic therapies, and latterly for targeted agents such as obatoclax and imatinib. More recently, we have employed PBPK modeling in a prospective manner to inform the first pediatric trials of pinometostat and tazemetostat in genetically defined populations (mixed lineage leukemia-rearranged and integrase interactor-1-deficient sarcomas, respectively). In this review, we evaluate the application of PBPK modeling in pediatric cancer drug development and discuss the important challenges that lie ahead in this field.

Physiologically based pharmacokinetic and pharmacodynamic modeling in cancer drug development: status, potential and gaps

INTRODUCTION

Modeling and simulation have become important means of answering questions relevant to the development of a drug, making it possible to assess risks early and to reduce costs. Physiologically based pharmacokinetic and pharmacodynamic (PBPK/PD) models contribute to a comprehensive understanding of the drug, covering specific questions from early discovery through lifecycle management stages. As for other disease areas, in oncology, PBPK and PD models are important topics that remain to be addressed.

AREAS COVERED

This review describes current PBPK and PD approaches, their applicability in drug development in general and specifically in the area of oncology. It discusses the current status and then focuses on key challenges and the potential for future use. It provides cases in which modeling currently cannot answer the questions and assesses the requirements to close gaps for PBPK/PD in oncology.

EXPERT OPINION

PBPK/PD models have led to improvements in identifying risks and reducing costs during the drug development process. Nevertheless, there is a lot of potential, where more rigorous integration of biological knowledge and specific experimental design would result in a more comprehensive biological picture. Ideally, such approaches would reveal the extent to which preclinical work can be extrapolated to clinical settings, thus enabling reliable prediction and, ultimately, reducing failed trials in clinical oncology.

Modeling and simulation in pediatric drug therapy: Application of pharmacometrics to define the right dose for children

During the past decades significant progress has been made in our understanding of the importance of age-appropriate development of new drug therapies in children. Importantly, several regulatory initiatives in Europe and the US have provided a framework for a rationale. In the US, most notably the enactment of the Best Pharmaceuticals for Children Act (BPCA) and Product Research and Equity Act (PREA) has facilitated the studying of on-patent and off-patent drugs in children. The biggest challenge in pediatric studies is defining a safe and effective dose or dose range in a patient population that can span from premature neonates to adolescents. From a mechanism-based perspective, advances in the science of quantitative pharmacology and pharmacometrics have resulted in the development of model-based approaches to better describe and understand important age-related factors influencing drug disposition and response in pediatric patients. The application of modeling and simulation has been shown to result in better estimates of pediatric doses as evidenced by several studies, although the optimal approach is still being debated. The extrapolation of efficacy findings from adults to the pediatric population has streamlined the development process especially for studies in older children. However, a focus on developmental changes in neonates and infants as well as further developing a paradigm for conducting pharmacodynamic studies in neonates, infants, and children remain important unmet needs. In this overview we will review current approaches for age-appropriate dose selection and highlight ongoing efforts to define exposure-response and clinical outcome relationships across the pediatric age spectrum.