An analysis program MULTI(ELS) based on extended nonlinear least squares method for microcomputers.

Area of Science:

• Pharmacokinetics
• Computational Biology
• Statistical Modeling

Background:

• Population pharmacokinetics (PopPK) analysis is crucial for understanding drug behavior in diverse patient groups.
• Existing methods often require specialized software and significant computational resources.
• Accurate estimation of population parameters and variability is essential for effective drug development and dosing strategies.

Purpose of the Study:

• To develop and validate a microcomputer-based program, MULTI(ELS), for population pharmacokinetic analysis.
• To implement the extended least squares (ELS) method for simultaneous estimation of pharmacokinetic parameters and variability.
• To compare the performance of MULTI(ELS) against a well-established PopPK software, NONMEM.

Main Methods:

• Development of the MULTI(ELS) program using Microsoft minimum BASIC.
• Implementation of two least squares algorithms: quasi-Newton and simplex methods.
• Validation through comparison with NONMEM (Version I, Level 3) using multiple time course data sets.
• Calculation of Akaike's Information Criterion (AIC) for model evaluation.

Main Results:

• MULTI(ELS) successfully performed population pharmacokinetic analysis on a microcomputer.
• The program accurately estimated population pharmacokinetic parameters, inter-individual variabilities, and intra-individual variabilities.
• Results from MULTI(ELS) were consistent with those obtained from NONMEM.
• Akaike's Information Criterion (AIC) proved effective for comparing population characteristics and verifying model structures.

Conclusions:

• MULTI(ELS) provides a viable, accessible tool for population pharmacokinetic analysis on microcomputers.
• The program's ability to estimate variability and its consistency with NONMEM support its utility in research and clinical settings.
• AIC is a valuable metric for model selection and structure verification in population pharmacokinetic studies using MULTI(ELS).