|
Activity Number:
|
134
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #308869 |
|
Title:
|
Prediction-Based Decision for Validation of (Bio)Analytical Methods Using Tolerance Intervals and Accuracy Profiles
|
|
Author(s):
|
Bruno Boulanger*+ and Walthère Dewe and Francois Moonen and Philippe Hubert
|
|
Companies:
|
Université de Liège and GlaxoSmithKline and Arlenda and University of Liege
|
|
Address:
|
Chemin du Foriest, Braine L Alleud, B1420, Belgium
|
|
Keywords:
|
validation assays ; tolerance interval ; accuracy profile ; fit-for-purpose ; prediction based decision
|
|
Abstract:
|
The purpose of validation of analytical methods is to demonstrate that it is suited for its intended purpose. For quantitative analytical method the objective is to quantify unknown samples with an acceptable accuracy. A decision about the suitability of the analytical method based on prediction is proposed at validation stage: a method is declared proper for routine use if most of the future results it will generated are likely to be accurate. This is achieved by using the "ß-expectation tolerance interval" and the corresponding "accuracy profile." A "fit-for-purpose" approach also proposed to select the most efficient standard curve. Four case studies will be reported showing that predictions made using the "ß-expectation tolerance intervals" can be trusted for decision making. This way the objectives of the analytical method, the objectives of the validation phase are conciliated.
|
