Leave-one-out procedure in the validation of elimination rate constant analysis

Area of Science:

• Pharmacokinetics and Drug Metabolism
• Bioequivalence Studies
• Regulatory Science

Background:

• Standardized methods for calculating the elimination rate constant (kel) exist, but lack validation for concentration-time (C-T) point selection.
• Accurate kel calculation is crucial for pharmacokinetic analysis, particularly in bioequivalence studies.

Purpose of the Study:

• To determine if kel analysis can utilize a condensed validation procedure.
• To establish acceptance criteria for validating C-T point selection in kel calculations.
• To assess the reliability of C-T point selection for kel determination across different drug lipophilicity.

Main Methods:

• Utilized pharmacokinetic data from bioequivalence studies of four drugs (itraconazole, atorvastatin, trimetazidine, perindopril).
• Performed pharmacokinetic calculations using WinNonlin Professional v 5.3.
• Implemented internal validation via leave-one-out cross-validation for kel analysis.

Main Results:

• The selection of C-T points for kel calculations cannot be automated, as some sequences failed validation criteria.
• Proposed validation criteria: Q² ≥ 0.6, R² ≥ 0.85, and Q² - R² < 0.3.
• The validation procedure confirmed the accuracy of kel calculations, even with varied C-T point sequences.

Conclusions:

• A robust, condensed validation procedure is necessary for selecting C-T data points in kel analysis.
• The proposed criteria ensure the optimal and accurate selection of data for reliable kel determination.
• This approach enhances the rigor of pharmacokinetic assessments in drug development and regulatory submissions.