High-performance PBPK model for predicting CYP3A4 induction-mediated drug interactions: a refined and validated approach.

Introduction: The cytochrome P450 enzyme 3A4 (CYP3A4) mediates numerous drug-drug interactions (DDIs) by inducing the metabolism of co-administered drugs, which can result in reduced therapeutic efficacy or increased toxicity. This study developed and validated a Physiologically Based Pharmacokinetic (PBPK) model to predict CYP3A4 induction-mediated DDIs, focusing on the early stages of clinical drug development. Methods: The PBPK model for rifampicin, a potent CYP3A4 inducer, was developed and validated using human pharmacokinetic data. Subsequently, PBPK models for 'victim' drugs were constructed and validated. The PBPK-DDI model's predictive performance was assessed by comparing predicted area under the curve (AUC) and maximum concentration (C Results: The rifampicin PBPK model accurately simulated human pharmacokinetic profiles. The PBPK-DDI model demonstrated high predictive accuracy for AUC ratios, with 89% of predictions within the 0.5 to 2-fold criterion and 79% meeting the Guest criteria. For Cmax ratios, an impressive 93% of predictions were within the acceptable range. The model significantly outperformed the static model, particularly in estimating DDI risks associated with CYP3A4 induction. Discussion: The PBPK-DDI model is a reliable tool for predicting CYP3A4 induction-mediated DDIs. Its high predictive accuracy, confirmed by adherence to evaluation standards, affirms its reliability for drug development and clinical pharmacology. Future refinements may further enhance its predictive value.