Opportunities and Challenges Related to the Implementation of Model-Based Bioequivalence Criteria

Sample Size Determination and Study Design Impact on Dose-Scale Pharmacodynamic Bioequivalence: a Case Study Using Orlistat

Dose-scale pharmacodynamic bioequivalence is recommended for evaluating the consistency of generic and innovator formulations of certain locally acting drugs, such as orlistat. This study aimed to investigate the standard methodology for sample size determination and the impact of study design on dose-scale pharmacodynamic bioequivalence using orlistat as the model drug. A population pharmacodynamic model of orlistat was developed using NONMEM 7.5.1 and utilized for subsequent simulations. Three different study designs were evaluated across various predefined relative bioavailability ratios of test/reference (T/R) formulations. These designs included Study Design 1 (2×1 crossover with T1 60 mg, R1 60 mg, and R2 120 mg), Study Design 2 (2×1 crossover with T2 120 mg, R1 60 mg, and R2 120 mg), and Study Design 3 (2×2 crossover with T1 60 mg, T2 120 mg, R1 60 mg, and R2 120 mg). Sample sizes were determined using a stochastic simulation and estimation approach. Under the same T/R ratio and power, Study Design 3 required the minimum sample size for bioequivalence, followed by Study Design 1, while Study Design 2 performed the worst. For Study Designs 1 and 3, a larger sample size was needed on the T/R ratio < 1.0 side for the same power compared to that on the T/R ratio > 1.0 side. The opposite asymmetry was observed for Study Design 2. We demonstrated that Study Design 3 is most effective for reducing the sample size for orlistat bioequivalence studies, and the impact of T/R ratio on sample size shows asymmetry.

Model-Based Bioequivalence Analysis to Assess and Predict the Relative Bioavailability of Valproic Acid Formulations

BACKGROUND AND OBJECTIVE

Model-based bioequivalence (MBBE) encompasses the use of nonlinear mixed effect models supporting the estimation of pharmacokinetic endpoints to assess the relative bioavailability between multi-source drug products. This application emerges as a valuable alternative to the standard non-compartmental analysis (NCA) in bioequivalence (BE) studies in which dense sampling is not possible. In this work, we aimed to assess the application of MBBE compared to traditional methods in evaluating the relative bioavailability of two formulations with different drug release properties. Additionally, we sought to predict the performance of a modified-release formulation in a multiple-dose scenario, leveraging data from a single-dose study.

METHODS

MBBE analysis was implemented to estimate the BE endpoints (90% CI for the Test/Reference geometric mean ratio, T/R GMR) in area under the concentration-time curve (AUC) and maximum concentration (Cmax) using data from a single-dose, 2-period, 2-sequence BE study performed in 14 healthy subjects between a locally developed valproic acid extended-release formulation (Test) and the brand-name delayed-release formulation (Reference).

RESULTS

Results were compared with the standard approach, revealing that MBBE analysis achieved higher discrimination between formulations for Cmax, addressing limitations of the experimental sampling design and highlighting an advantage for this model-based analysis even when rich data are available. Additionally, the bioequivalence outcome under the multiple-dose scenario was predicted through a simulation-based study for both total and unbound valproic acid concentrations, considering the impact of valproic acid saturable binding on BE conclusions.

CONCLUSIONS

The MBBE analysis was superior to the NCA approach in detecting product-related differences, overcoming limitations in the study experimental design. Predictions for the multiple-dose scenario preclude that the extended-release properties of the Test formulation would persist at steady state, resulting in lower peak-to-trough fluctuation and bioequivalent performance in terms of the extent of drug absorption. Overall, these results should discourage unnecessary experimentation in healthy subjects.

Use of the Same Model or Modeling Strategy Across Multiple Submissions: Focus on Complex Drug Products

Evidence shows that there is an increasing use of modeling and simulation to support product development and approval for complex generic drug products in the USA, which includes the use of mechanistic modeling and model-integrated evidence (MIE). The potential for model reuse was the subject of a workshop session summarized in this review, where the session included presentations and a panel discussion from members of the U.S. Food and Drug Administration (FDA), academia, and the generic drug product industry. Concepts such as platform performance assessment and MIE standardization were introduced to provide potential frameworks for model reuse related to mechanistic models and MIE, respectively. The capability of models to capture formulation and product differences was explored, and challenges with model validation were addressed for drug product classes including topical, orally inhaled, ophthalmic, and long-acting injectable drug products. An emphasis was placed on the need for communication between FDA and the generic drug industry to continue to foster maturation of modeling and simulation that may support complex generic drug product development and approval, via meetings and published guidance from FDA. The workshop session provided a snapshot of the current state of modeling and simulation for complex generic drug products and offered opportunities to explore the use of such models across multiple drug products.

Evaluation of model-based bioequivalence approach for single sample pharmacokinetic studies

In a traditional pharmacokinetic (PK) bioequivalence (BE) study, a two-way crossover study is conducted, PK parameters (namely the area under the time-concentration curve [AUC] and the maximal concentration [C max ]) are obtained by noncompartmental analysis (NCA), and the BE analysis is performed using the two one-sided test (TOST) method. For ophthalmic drugs, however, only one sample of aqueous humor, in one eye, per eye can be obtained in each patient, which precludes the traditional BE analysis. To circumvent this issue, the U.S. Food and Drug Administration (FDA) has proposed an approach coupling NCA with either parametric or nonparametric bootstrap (NCA bootstrap). The model-based TOST (MB-TOST) has previously been proposed and evaluated successfully for various settings of sparse PK BE studies. In this paper, we evaluate, via simulations, MB-TOST in the specific setting of single sample PK BE study and compare its performance to NCA bootstrap. We performed BE study simulations using a published PK model and parameter values and evaluated multiple scenarios, including study design (parallel or crossover), sampling times (5 or 10 spread across the dosing interval), and geometric mean ratio (of 0.8, 0.9, 1, and 1.25). Using the simulated structural PK model, MB-TOST performed similarly to NCA bootstrap for AUC. ForC max , the latter tended to be conservative and less powerful. Our research suggests that MB-TOST may be considered as an alternative BE approach for single sample PK studies, provided that the PK model is correctly specified and the test drug has the same structural model as the reference drug.

Impact of model misspecification on model-based tests in PK studies with parallel design: real case and simulation studies

This article evaluates the performance of pharmacokinetic (PK) equivalence testing between two formulations of a drug through the Two-One Sided Tests (TOST) by a model-based approach (MB-TOST), as an alternative to the classical non-compartmental approach (NCA-TOST), for a sparse design with a few time points per subject. We focused on the impact of model misspecification and the relevance of model selection for the reference data. We first analysed PK data from phase I studies of gantenerumab, a monoclonal antibody for the treatment of Alzheimer’s disease. Using the original rich sample data, we compared MB-TOST to NCA-TOST for validation. Then, the analysis was repeated on a sparse subset of the original data with MB-TOST. This analysis inspired a simulation study with rich and sparse designs. With rich designs, we compared NCA-TOST and MB-TOST in terms of type I error and study power. With both designs, we explored the impact of misspecifying the model on the performance of MB-TOST and adding a model selection step. Using the observed data, the results of both approaches were in general concordance. MB-TOST results were robust with sparse designs when the underlying PK structural model was correctly specified. Using the simulated data with a rich design, the type I error of NCA-TOST was close to the nominal level. When using the simulated model, the type I error of MB-TOST was controlled on rich and sparse designs, but using a misspecified model led to inflated type I errors. Adding a model selection step on the reference data reduced the inflation. MB-TOST appears as a robust alternative to NCA-TOST, provided that the PK model is correctly specified and the test drug has the same PK structural model as the reference drug.

Population pharmacokinetics of mobocertinib in healthy volunteers and patients with non-small cell lung cancer

Mobocertinib is an oral tyrosine kinase inhibitor approved for treatment of patients with locally advanced or metastatic non‐small cell lung cancer (mNSCLC) with epidermal growth factor receptor gene (EGFR) exon 20 insertion mutations whose disease has progressed on or after platinum‐based chemotherapy. This population pharmacokinetic (PK) analysis describes the PK of mobocertinib and its active metabolites, AP32960, and AP32914, using data from two phase I studies in healthy volunteers (n = 110) and two phase I/II studies in patients with mNSCLC (n = 317), including the pivotal phase I/II study. The plasma PK of mobocertinib, AP32960, and AP32914 were well‐characterized by a joint semimechanistic model that included two compartments for mobocertinib with absorption via three transit compartments, two compartments for AP32960, and one compartment for AP32914. The observed time‐dependency in PK was described by an enzyme compartment with drug and metabolite concentration‐dependent stimulation of enzyme production, resulting in the enzyme increasing the apparent clearance of mobocertinib, AP32960, and AP32914. Effects of healthy volunteer status (vs. patients with mNSCLC) on apparent oral clearance of all three moieties and on apparent central volume of distribution for mobocertinib were included as structural covariates in the final model. No clinically meaningful differences in mobocertinib PK were observed based on age (18–86 years), race, sex, body weight (37.3–132 kg), mild‐to‐moderate renal impairment (estimated glomerular filtration rate 30–89 ml/min/1.73 m² by modification of diet in renal disease equation), or mild‐to‐moderate hepatic impairment, suggesting that no dose adjustment is required based on these covariates in patients with mNSCLC.

An experience on the model-based evaluation of pharmacokinetic drug-drug interaction for a long half-life drug

Fixed-dose combinations development requires pharmacokinetic drug-drug interaction (DDI) studies between active ingredients. For some drugs, pharmacokinetic properties such as long half-life or delayed distribution, make it difficult to conduct such clinical trials and to estimate the exact magnitude of DDI. In this study, the conventional (non-compartmental analysis and bioequivalence [BE]) and model-based analyses were compared for their performance to evaluate DDI using amlodipine as an example. Raw data without DDI or simulated data using pharmacokinetic models were compared to the data obtained after concomitant administration. Regardless of the methodology, all the results fell within the classical BE limit. It was shown that the model-based approach may be valid as the conventional approach and reduce the possibility of DDI overestimation. Several advantages (i.e., quantitative changes in parameters and precision of confidence interval) of the model-based approach were demonstrated, and possible application methods were proposed. Therefore, it is expected that the model-based analysis is appropriately utilized according to the situation and purpose.

Multivariate Assessment for Bioequivalence Based on the Correlation of Random Effect

Background and Objective

Bioequivalence tests are fundamental step in assessing the equivalence in bioavailability between a test and reference product. In practice, two separate linear mixed models (LMMs) with random subject effects, which have an area under the concentration-time curve (AUC) and the peak concentration (C_max) as the responses, have become the gold standard for evaluating bioequivalence. Recently, Lee et al developed a multivariate hierarchical generalized linear model (HGLM) for several responses that modeled correlations among multivariate responses via correlated random effects. The objective of this study was to apply this multivariate analysis to the bioequivalence test in practice and to compare the performance of multivariate HGLM and separate LMMs.

Methods

Three pharmacokinetic datasets, fixed-dose combination (naproxen and esomeprazole), tramadol and fimasartan data were analyzed. We compared the 90% confidence interval (CI) for the geometric mean ratio (GMR) of a test product to a reference product using the multivariate HGLM and two conventional separate LMMs.

Results

We found that the 90% CIs for the GMRs of both AUC and C_max from the multivariate HGLM were narrower than those from the separate LMMs: (0.843, 1.152) vs (0.825, 1.177) for C_max of esomeprazole in fixed-dose combination data; (0.805, 0.931) vs (0.797, 0.941) for C_max in tramadol data; (0.801, 1.501) vs (0.762, 1.578) for C_max and (1.163, 1.332) vs (1.009, 1.341) for AUC in fimasartan data, consistent with the random subject effects from two separate LMMs being highly correlated in the three datasets (correlation coefficient r = 0.883; r = 0.966; r = 0.832).

Conclusion

This multivariate HGLM had good performance in the bioequivalence test with multiple endpoints. This method would provide a more reasonable option to reduce the 90% CI by adding correlation parameters and thus an advantage especially in evaluating the bioequivalence of highly variable drugs with broad 90% CIs.

Antibiotics in Adult Cystic Fibrosis Patients: A Review of Population Pharmacokinetic Analyses

BACKGROUND

Lower respiratory tract infections are common in adult patients with cystic fibrosis (CF) and are frequently caused by Pseudomonas aeruginosa, resulting in chronic lung inflammation and fibrosis. The progression of multidrug-resistant strains of P. aeruginosa and alterations in the pharmacokinetics of many antibiotics in CF make optimal antimicrobial therapy a challenge, as reflected by high between- and inter-individual variability (IIV).

OBJECTIVES

This review provides a synthesis of population pharmacokinetic models for various antibiotics prescribed in adult CF patients, and aims at identifying the most reported structural models, covariates and sources of variability influencing the dose-concentration relationship.

METHODS

A literature search was conducted using the PubMed database, from inception to August 2020, and articles were retained if they met the inclusion/exclusion criteria.

RESULTS

A total of 19 articles were included in this review. One-, two- and three-compartment models were reported to best describe the pharmacokinetics of various antibiotics. The most common covariates were lean body mass and creatinine clearance. After covariate inclusion, the IIV (range) in total body clearance was 27.2% (10.40-59.7%) and 25.9% (18.0-33.9%) for β-lactams and aminoglycosides, respectively. IIV in total body clearance was estimated at 36.3% for linezolid and 22.4% for telavancin. The IIV (range) in volume of distribution was 29.4% (8.8-45.9%) and 15.2 (11.6-18.0%) for β-lactams and aminoglycosides, respectively, and 26.9% for telavancin. The median (range) of residual variability for all studies, using a combined (proportional and additive) model, was 12.7% (0.384-30.80%) and 0.126 mg/L (0.007-1.88 mg/L), respectively.

CONCLUSION

This is the first review that highlights key aspects of different population pharmacokinetic models of antibiotics prescribed in adult CF patients, effectively proposing relevant information for clinicians and researchers to optimize antibiotic therapy in CF.