Non-compartment model to compartment model pharmacokinetics transformation meta-analysis--a multivariate nonlinear...

Area of Science:

• Pharmacometrics
• Drug Development
• Biostatistics

Background:

• Model-based drug development relies on establishing pharmacokinetic (PK) models from literature.
• Pharmacokinetic models are central to drug development.
• Transforming PK parameters between non-compartment and compartment models is essential for data integration.

Purpose of the Study:

• To develop and validate a method for transforming published non-compartment model PK parameters into compartment model PK parameters.
• To facilitate the integration of diverse PK data for robust model-based drug development.
• To establish a meta-analysis approach for PK parameter conversion.

Main Methods:

• A meta-analysis was conducted using a multivariate nonlinear mixed model.
• A conditional first-order linearization approach was employed for statistical estimation and inference.
• The method was demonstrated using midazolam (MDZ) PK data from 10 publications.

Main Results:

• Successfully transformed 6 non-compartment PK parameters into 5 compartment PK parameters for MDZ.
• Multivariate nonlinear mixed model showed minimal relative bias (<1%) in simulations when all PK parameters were available.
• Small relative bias (<3%) and high coverage probabilities (>85%) were observed even with missing PK parameters.

Conclusions:

• The developed meta-analysis approach for PK parameter transformation offers valid statistical inference.
• This method can be routinely applied to enhance model-based drug development.
• It enables the effective utilization of published PK data for building comprehensive models.