Xiaofei Wu, Xuanji Peng, Ranran Jia, Jiahua Ma, Hongyun Wang, Xiaohong Han
The K-PD modeling framework characterized the cross-species pharmacology of GalNAc-siRNAs, confirmed consistent translational patterns, and identified target-specific effects on PD kinetics. These findings facilitate early human pharmacological prediction, enabling a shift from empirical scaling toward model-informed translational decision-making.
BACKGROUND: Small interfering RNAs (siRNAs) conjugated with N-acetylgalactosamine (GalNAc) are liver-targeted therapeutics with sustained activity. However, their cross-species pharmacokinetic (PK) and pharmacodynamic (PD) translation remains poorly defined, particularly regarding target dependence and PD endpoint selection. This study aimed to develop a kinetic-pharmacodynamic (K-PD) modeling framework to characterize their long-term pharmacology, cross-species translational relationships, and target mRNA effects by quantifying biophase half-life, in vivo potency (IDK5 0), and PD turnover.
METHODS: Publicly available PD time-course data for 30 GalNAc-siRNAs across four species (mice, rats, monkeys, and humans) were compiled through a targeted search of the literature and other public sources. A unified K-PD modeling framework was applied to estimate key parameters and evaluate cross-species translational behavior and target-related variability.
RESULTS: Human biophase half-life exceeded that in preclinical species, whereas human IDK5 0 was lower. PD half-life in humans approximated that in nonhuman primates. Despite consistent cross-species trends overall, K-PD parameters varied across compounds and targets. PD turnover half-life showed greater within-target consistency than IDK5 0, suggesting differential sensitivity of these endpoints to target biology.
CONCLUSION: The K-PD modeling framework characterized the cross-species pharmacology of GalNAc-siRNAs, confirmed consistent translational patterns, and identified target-specific effects on PD kinetics. These findings facilitate early human pharmacological prediction, enabling a shift from empirical scaling toward model-informed translational decision-making.