Accounting for inter-correlation between enzyme abundance: a simulation study to assess implications on global...

Area of Science:

• Pharmacology
• Biochemistry
• Computational Biology

Background:

• Physiologically based pharmacokinetic (PBPK) models often incorporate correlated parameters like organ volumes and blood flows.
• Recent proteomic advances reveal correlations between drug-metabolizing enzyme abundances in the liver.
• Population PBPK modeling increasingly focuses on extreme physiological cases, necessitating reliable parameter estimation.

Purpose of the Study:

• To assess the impact of enzyme abundance correlations on drug pharmacokinetics.
• To evaluate the consequences of including or omitting these correlations in PBPK models.
• To investigate the effect of enzyme correlations on global sensitivity analysis (GSA).

Main Methods:

• Simulation study using three semi-physiological PBPK models for different drug administration and metabolism scenarios.
• Consideration of correlations between two enzymes for drugs that are substrates of both.
• Application of variance-based GSA to a reduced PBPK model for repaglinide.

Main Results:

• Implementing enzyme abundance correlations can widen confidence intervals for pharmacokinetic parameters like AUC and bioavailability.
• Ignoring correlations may lead to implausible parameter combinations and inaccurate pharmacokinetic estimations.
• The inclusion of correlations significantly impacts GSA outcomes.

Conclusions:

• Known correlations between enzyme abundances should be consistently incorporated into population PBPK models.
• Failure to account for these correlations can compromise model validity and predictive accuracy.
• Accurate PBPK modeling relies on acknowledging and integrating known biological parameter relationships.