Speaker
Description
Abstract:
Background: Accurate prediction of metabolic drug-drug interactions (DDIs) is crucial for clinical medication. However, current predominant prediction models are limited. This study was designed to address these limitations by developing a new predictive model that is both accurate and readily applicable.
Methods: A simplified predictive model for metabolic DDIs was developed based on enzyme kinetics, incorporating inhibitor concentration, substrate concentration, and enzyme abundance, with effects quantified by the significant metabolic inhibition duration (ti). The model was validated using an integrated rat system (in vitro microsomes, in situ liver perfusion, in vivo CCl4-induced liver injury) and then applied to predict and compare DDI risk from midazolam-clarithromycin co-administration between elderly and young adults.
Results: The model revealed a substrate concentration threshold (0.1 Km), above which tᵢ rose sharply with concentration and below which impact was negligible; enzyme abundance robustly regulated tᵢ across all conditions. Consistently, in vitro studies confirmed that increasing enzyme content could attenuate or abolish DDIs, exerting a stronger effect than substrate concentration. Furthermore, in situ perfusion experiments showed that DDI prolongation occurred only with high substrate concentrations, and in vivo studies in enzyme-deficient (CCl4-treated) rats confirmed heightened DDI sensitivity. Finally, the model accurately predicted the clinically observed DDI risk for midazolam-clarithromycin co-administration across age groups, validating its reliability.
Conclusion: A simplified enzyme kinetics model was developed and validated for predicting metabolic DDIs. A key mechanistic insight is that DDI risk is co-determined by substrate exposure, inhibitor exposure, and the patient's intrinsic enzyme capacity.
Acknowledgments:
Supported by the National Natural Science Foundation of China, grant No. 8227131503.
