Development of a Dynamic Physiologically Based Mechanistic Kidney Model to Predict Renal Clearance

Area of Science:

• Pharmacokinetics
• Computational Biology
• Renal Physiology

Background:

• Current renal clearance prediction relies on empirical methods like QSPR and allometric scaling.
• Mechanistic in silico kidney models offer a promising alternative but often require empirical scaling factors.
• Accurate prediction of renal clearance is crucial for drug development and dosing.

Purpose of the Study:

• To develop a physiologically based mechanistic kidney model for predicting renal clearance.
• To simulate renal clearance, including filtration and pH-dependent passive reabsorption, using in vitro permeability data.
• To establish a robust in silico tool for quantitative prediction of drug renal clearance.

Main Methods:

• A 35-compartment physiologically based mechanistic kidney model was constructed.
• The model was validated using 46 diverse test compounds (neutrals, acids, bases, zwitterions).
• Active secretion and pH-dependent passive diffusion were incorporated and tested with specific compounds (e.g., PAH, cimetidine).

Main Results:

• The developed kidney model successfully simulated renal clearance based on in vitro permeability.
• Model predictions for renal clearance were within twofold of observed values for 87% of the test drugs.
• The feasibility of incorporating active secretion and pH-dependent passive diffusion was demonstrated.

Conclusions:

• The developed mechanistic kidney model provides a powerful tool for predicting renal clearance.
• This in silico approach reduces reliance on empirical factors and improves prediction accuracy.
• The model enables more reliable in vitro to in vivo extrapolation for drug development.