Novel deconvolution method for linear pharmacokinetic systems with polyexponential impulse response

Area of Science:

• Pharmacokinetics and Pharmacodynamics
• Computational Biology
• Drug Metabolism and Disposition

Background:

• Accurate estimation of drug input into the systemic circulation is crucial for pharmacokinetic modeling.
• Existing deconvolution methods may have limitations in handling noise and complex input functions.
• Understanding drug input kinetics informs dosing strategies and therapeutic efficacy.

Purpose of the Study:

• To introduce a novel least-squares deconvolution method for quantifying drug input rate and extent.
• To evaluate the performance of the new method against existing deconvolution techniques.
• To demonstrate the method's applicability using real-world pharmacokinetic data.

Main Methods:

• The novel method employs a polyexponential approximation of the impulse response and a polynomial approximation of the input rate.
• Implementation utilizes standard multiple linear regression with a zero-intercept option.
• Performance was assessed using simulated data with varying levels of random noise and compared to two other deconvolution methods.

Main Results:

• The proposed least-squares deconvolution method demonstrated significant advantages, particularly in the presence of random noise.
• The method was successfully applied to analyze plasma pentobarbital levels following oral and intravenous administration.
• Comparative analysis indicated improved accuracy and robustness compared to existing deconvolution techniques.

Conclusions:

• The novel least-squares deconvolution method provides a robust and efficient tool for analyzing drug input kinetics.
• Its computational simplicity and advantages in noisy data make it a valuable addition to pharmacokinetic analysis.
• Further investigation into the assumptions and limitations of deconvolution in drug input analysis is warranted.