P M Sathe1, Y Tsong, V P Shah. 1. Office of Pharmaceutical Sciences, U.S. Food and Drug Administration, Rockville, Maryland 20855, USA.
Abstract
PURPOSE: To develop and propose a "model dependent' approach for the in-vitro dissolution profiles comparison. METHODS: Diltiazem hydrochloride tablet dissolution profiles were compared using a statistical approach based on a mathematical model. A similarity region (SR) was defined based on the intra- and inter-lot parameter variances of the final production size standard lots. Statistical distances between the test and reference lot parameter means were computed and normalized. A 90% confidence region (CR) was developed around the statistical distance. The confidence region was compared with the similarity region to assess the similarity or dis-similarity of the test and reference (REF) lot dissolution profiles. Two test lots, one with a "minor' modification (mm) the other with a "major' modification (MM), were evaluated. RESULTS: "Weibull' was selected as the "model' function. A comparison of the confidence regions around the statistical distance of "mm-REF' and "MM-REF' with the similarity region, suggested that the dissolution profiles of the "minor' modification lot were similar and that of "major' modification lot were dis-similar to the reference lot. CONCLUSIONS: A "model dependent' approach was shown to be useful for the inter-lot in-vitro dissolution profiles comparison.
PURPOSE: To develop and propose a "model dependent' approach for the in-vitro dissolution profiles comparison. METHODS:Diltiazem hydrochloride tablet dissolution profiles were compared using a statistical approach based on a mathematical model. A similarity region (SR) was defined based on the intra- and inter-lot parameter variances of the final production size standard lots. Statistical distances between the test and reference lot parameter means were computed and normalized. A 90% confidence region (CR) was developed around the statistical distance. The confidence region was compared with the similarity region to assess the similarity or dis-similarity of the test and reference (REF) lot dissolution profiles. Two test lots, one with a "minor' modification (mm) the other with a "major' modification (MM), were evaluated. RESULTS: "Weibull' was selected as the "model' function. A comparison of the confidence regions around the statistical distance of "mm-REF' and "MM-REF' with the similarity region, suggested that the dissolution profiles of the "minor' modification lot were similar and that of "major' modification lot were dis-similar to the reference lot. CONCLUSIONS: A "model dependent' approach was shown to be useful for the inter-lot in-vitro dissolution profiles comparison.
Authors: Dorys Argelia Diaz; Stephen T Colgan; Connie S Langer; Nagesh T Bandi; Michael D Likar; Leslie Van Alstine Journal: AAPS J Date: 2015-10-01 Impact factor: 4.009
Authors: Cynthia K Brown; Horst Dieter Friedel; Amy R Barker; Lucinda F Buhse; Susanne Keitel; Todd L Cecil; Johannes Kraemer; J Michael Morris; Christos Reppas; Mary P Stickelmeyer; Chikako Yomota; Vinod P Shah Journal: AAPS PharmSciTech Date: 2011-06-18 Impact factor: 3.246