Vol. 2, Issue 2, Part A (2025)

Background: Narrow therapeutic index (NTI) antiepileptic drugs require precise dosing to balance efficacy and safety, particularly in special populations where physiological variability alters pharmacokinetic behavior. Conventional approaches often fail to account for interindividual differences in drug metabolism and clearance, leading to suboptimal outcomes. Physiologically based pharmacokinetic (PBPK) modeling offers a mechanistic framework for predicting exposure across diverse patient groups and guiding dose individualization.
Objective: This study aimed to develop and validate a PBPK model for an NTI antiepileptic drug and to predict optimal dosing adjustments in elderly, pregnant, hepatic-impaired, and pediatric populations, ensuring plasma concentrations remain within the therapeutic window.
Methods: A validated PBPK model was constructed using physicochemical, biopharmaceutical, and physiological data derived from literature and regulatory databases. The adult model was calibrated against clinical pharmacokinetic data and extrapolated to special populations by modifying relevant physiological parameters, including hepatic enzyme activity, renal clearance, plasma protein binding, and organ blood flow. Model performance was evaluated by comparing predicted versus observed pharmacokinetic parameters (AUC, Cmax, Tmax), and statistical analyses, including ANOVA and sensitivity testing, were used to quantify variability.
Results: The adult model reproduced observed pharmacokinetic profiles within a ±25% prediction error margin, confirming model reliability. Simulation outcomes revealed increased systemic exposure in elderly (+28%) and hepatic-impaired (+54%) subjects, modest elevation in pregnancy (+18%), and decreased exposure in pediatrics (-32%). Model-informed dose adjustments reducing to 65% (hepatic impairment), 80% (elderly), 90% (pregnancy), and increasing to 130% (pediatric) normalized exposure within the NTI bioequivalence limits (0.8-1.25). Sensitivity analysis identified hepatic intrinsic clearance, plasma protein binding, and hepatic blood flow as primary determinants of interindividual variability.
Conclusion: PBPK modeling accurately predicted pharmacokinetic behavior and optimal dose modifications for an NTI antiepileptic across special populations, validating its utility in precision dosing. Model-informed dose adjustments can reduce adverse event risk and enhance therapeutic efficacy, providing a scientifically robust and ethically sound alternative to empirical dosing. Integrating PBPK-guided dosing with therapeutic drug monitoring may strengthen clinical decision-making, regulatory evaluation, and personalized pharmacotherapy in NTI drug management.