Luis Quintairos1, David Ortego1, Carlos Bertoncini1, Mariangeles Pérez de la Cruz1
1CHEMO Research, Insud Pharma
Introduction: Tapentadol is a mu-opioid receptor agonist that also inhibits norepinephrine reuptake, making it effective for managing various types of pain. The population pharmacokinetic approach is widely recognized for its advantages in assessing the influence of different conditions on pharmacokinetic outcomes. The novel model-integrated evidence (MIE) approach presents a compelling alternative to traditional bioequivalence (BE) methods in specific scenarios. In this poster, we explore the impact of food on the bioequivalence of a tapentadol formulation developed by Chemo Group (not yet commercialized in Spain), by simulating studies under fed conditions for dose strengths previously evaluated in fasting conditions. Objectives: •Adapt an existing population pharmacokinetic (PK) model of tapentadol to include the food effect as a covariate, using data from prior studies. •To simulate the bioequivalence of tapentadol at dose strengths of 100 mg, 150 mg, and 200 mg under fed conditions. Methods: Data from seven bioequivalence single-dose crossover studies were pooled, comprising five conducted under fasting conditions (at strengths of 50, 100, 150, 200, and 250 mg) and two in a fed state (at strengths of 50 and 250 mg), to adapt a previously published population pharmacokinetic (PK) model for tapentadol (1). This model was originally a one-compartment model with two first-order input functions to describe absorption and included a food effect by increasing the slope of the second input rate constant (Ka2). It accounted for inter-individual variability using an exponential error model, with age, body weight, and aspartate aminotransferase (AST) as covariates influencing clearance (CL/F) and relative bioavailability (F_a). A more accurate description of the model can be found elsewhere (1). The model required adaptation to fit our data, with the covariate effect of AST fixed to 1 due to insufficient recorded data, and the residual error model restructured since only phase I data was available. Treatment effects were assessed in absorption parameters, and interoccasion variability (IOV) was included in relative bioavailability to account for period effects. The selection of the final model was performed according to: i) minimum objective function value (MOFV), ii) plausibility of parameter estimates and precision (given a %relative standard error) iii) visual predictive check plots after 1000 simulations. Once accepted, the final model was used to simulate bioequivalence of dose strengths of 100, 150 and 200 mg in fed conditions following previously published MIE method (2). Model development was conducted using NONMEM (3), while R software version 4.3.3 (4) was employed for non-compartmental analysis (NCA) calculations and bioequivalence assessment. Finch Studio (Enhanced Pharmacodynamics LLC) was also utilized as a supportive tool. Results: A total of 7136 concentration-time observations from 322 healthy subjects were included in the analysis, with actual doses based on the certificate of analysis considered in the model. The previously published pharmacokinetic model was adapted to accurately fit our data. Residual error was best described by an additive plus proportional model, with distinct proportional error estimates for each study. The treatment effect (test or reference) was incorporated into the fast absorption constant (Ka1) and the relative bioavailability of the absorption compartments (F1/F2). Additionally, interoccasion variability (IOV) was included to account for period effects on relative bioavailability. The model was then used to simulate studies at dose strengths of 100 mg, 150 mg, and 200 mg under fed conditions. The 90% confidence intervals calculated from these simulations indicated that the formulations would be bioequivalent in fed conditions, with intervals ranging from 95.18 to 107.84 for Cmax and 95.38 to 106.13 for AUClast for the 100 mg dose strength, 92.12 to 107.96 for Cmax and 92.79 to 105.59 for AUClast for the 150 mg dose strength, and 92.68 to 108.02 for Cmax and 92.30 to 104.80 for AUClast for the 200 mg dose strength. Conclusions: MIE methods were successfully applied to simulate and establish the bioequivalence of tapentadol at dose strengths of 100 mg, 150 mg, and 200 mg in a fed state, utilizing data collected from previous studies conducted under both fasting and fed conditions. MIE represents a highly valuable tool in the development of generic drugs and could serve as a waiver method to reduce the number of studies required in abbreviated new drug applications.
1. Huntjens DR, Liefaard LC, Nandy P, Drenth HJ, Vermeulen A. Population Pharmacokinetic Modeling of Tapentadol Extended Release (ER) in Healthy Subjects and Patients with Moderate or Severe Chronic Pain. Clin Drug Investig [Internet]. 2016 Mar 21;36(3):213–23. Available from: http://link.springer.com/10.1007/s40261-015-0371-x 2. Chen X, Nyberg HB, Donnelly M, Zhao L, Fang L, Karlsson MO, et al. Development and comparison of model-integrated evidence approaches for bioequivalence studies with pharmacokinetic end points. CPT Pharmacometrics Syst Pharmacol [Internet]. 2024 Aug 23; Available from: https://ascpt.onlinelibrary.wiley.com/doi/10.1002/psp4.13216 3. Stuart Beal, Lewis Sheiner, Alison Boeckmann, Robert Bauer. NONMEM users guide: introduction to NONMEM 7.5.1. Ellicott City: Icon Development Solutions. 2017;128. 4. R Developement Core Team. R: A language and environment for statistical computing. [Internet]. 2024 [cited 2024 Mar 21]. Available from: http://www.r-project.org
Reference: PAGE 33 (2025) Abstr 11637 [www.page-meeting.org/?abstract=11637]
Poster: Drug/Disease Modelling - CNS