Developing and Validating an In-vivo In-vitro Correlation Using Formal Hypothesis-testing

• Activity Number: 221
• Type: Contributed
• Date/Time: Tuesday, August 10, 2004 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): James T. Kost*+ and Kalyan Ghosh
• Companies: Merck & Co., Inc. and Merck & Co., Inc.
• Keywords: bioequivalence ; dissolution ; IVIVC

Abstract

A model in drug development that relates the in-vitro dissolution profile of a formulation to its in-vivo plasma concentration profile is often referred to as an in-vitro-in-vivo correlation (IVIVC). The method proposed by the U.S. Food and Drug Administration for developing and evaluating potential IVIVC models possesses several well-known shortcomings, the foremost of which is its failure to incorporate any measure of variability into the validation process. No quantification of risk is avialable at the design stage and there is limited confidence in the ultimate decision to either accept or reject the potential model at study's end. A new method for developing an IVIVC model, with subsequent validation performed through formal testing, is proposed herein. This testing is similar to that usually performed in an in-vivo bioequivalence study, and is facilitated by the construction of a predicted plasma concentration curve for each in vitro dissolution run. The relative advantages and disadvantages of the proposed method are discussed and an example is given.