A Semi-physiologically Based Model for Methylphenidate Pharmacokinetics in Adult Humans

Abstract

To determine if a literature-sourced physiological model for immediate-release (IR) methylphenidate (MPH), with addition of parameters for extended-release (ER) absorption can be adapted for NONMEM analysis to describe extended-release MPH drug products (i.e., Concerta® ER 54 mg tablets and Ritalin-LA® 40 mg capsules) pharmacokinetics (PK) in adults. This semi-physiological model will provide a platform to allow more accurate determination of Cmax in the analysis of methylphenidate plasma data from formulations with complex absorption. Adult reference data for total MPH plasma levels (summation of d- and l- enantiomers) were generated using individual subject parameters from a published NONMEM model for ER MPH (individual subject parameter generated data was used because the true experimental data is proprietary with only the previously-published summary parameters available for public use). The IR physiological model required analysis of both d- and l-MPH enantiomers which were estimated at each time point for the data by calculating the d/l enantiomer ratio from total MPH plasma concentration levels, based upon literature ratio estimates. Absorption was characterized by a fast zero-order and a delayed slow first-order release. Consistent with the literature IR model, two duplicate physiological models were used to describe the d- and l-MPH enantiomers. The mean and variability ratios for the individual subject parameter-generated data/true experimental data were very close to 1.0. The predictive performance of the absorption-modified physiological model and its disposition parameters were demonstrated, as they described both the Concerta® and Ritalin LA® PK and Cmax values very well. The bias was less than 6% for Concerta® peaks while peak 1 for d-Ritalin LA® was 12.9% and peak 2, 7.5%. The IR MPH physiological model, adapted for NONMEM analyses of the ER MPH drug products Concerta® and Ritalin LA® described the individual parameter-generated reference data well for both formulations.