Development of an In Vitro Drug Release Method to Enable In Vitro-In Vivo Correlation of Potassium Chloride Extended-Release Tablets.

An in vitro drug release test was developed to establish a level A in vitro-in vivo correlation (IVIVC) for predicting the in vivo performance of potassium chloride extended-release (ER) matrix tablets. Three ER formulations of potassium chloride with different in vitro release rates were designed using the USP dissolution test, and their urinary pharmacokinetic profiles were evaluated in healthy subjects. Due to the lack of IVIVC with the USP method, experiments were designed to investigate the effects of in vitro test conditions on drug release in order to match in vitro drug release with in vivo behaviors of different formulations. The evaluated in vitro variables included the type of USP apparatus, surfactant, and ionic strength of the dissolution medium. Based on the study findings and data analysis, a discriminatory drug release method was successfully developed that enabled the establishment and validation of a level A IVIVC model of the potassium chloride ER tablet using urinary pharmacokinetic data. This method uses USP apparatus I at 50 rpm in 900 mL of 150 mM NaCl solution containing 40 mM sodium dodecyl sulfate at 37 °C. The current study highlights the value of investigating test conditions in developing a predictive in vitro test method for establishing IVIVC.

Development of In Vitro-In Vivo Correlation for Potassium Chloride Extended Release Tablet Formulation Using Urinary Pharmacokinetic Data.

PURPOSE: To develop and validate a Level A in vitro-in vivo correlation (IVIVC) for potassium chloride extended-release (ER) formulations. METHODS: Three prototype ER formulations of potassium chloride with different in vitro release rates were developed and their urinary pharmacokinetic profiles were evaluated in healthy subjects. A mathematical model between in vitro dissolution and in vivo urinary excretion, a surrogate for measuring in vivo absorption, was developed using time-scale and time-shift parameters. The IVIVC model was then validated based on internal and external predictability. RESULTS: With the established IVIVC model, there was a good correlation between the observed fraction of dose excreted in urine and the time-scaled and time-shifted fraction of the drug dissolved, and between the in vitro dissolution time and the in vivo urinary excretion time for the ER formulations. The percent prediction error (%PE) on cumulative urinary excretion over the 24 h interval (Ae0-24h) and maximum urinary excretion rate (Rmax) was less than 15% for the individual formulations and less than 10% for the average of the two formulations used to develop the model. Further, the %PE values using external predictability were below 10%. CONCLUSIONS: A novel Level A IVIVC was successfully developed and validated for the new potassium chloride ER formulations using urinary pharmacokinetic data. This successful IVIVC may facilitate future development or manufacturing changes to the potassium chloride ER formulation.