AICc for Calibration Model Selection in Assay Validation. (2nd October 2017)

• Record Type: Journal Article
• Authors: Zhang, Lanju; Tan, Charles; LeBlond, David
• Is Part Of: Statistics in biopharmaceutical research. Volume 9:Number 4(2017)
• Page Start: 379
• Page End: 383
• Publication Date: 2017-10-02
• Subjects: AICc -- Akaike weight -- Assay procedure -- Bayesian -- Calibration model -- Linearity test -- Power -- Type I error rate; Pharmacy -- Statistical methods -- Periodicals; Pharmaceutical biotechnology -- Statistical methods -- Periodicals; Biopharmaceutics -- Periodicals; Biometry -- Periodicals; Pharmacy -- Statistical methods; Periodicals; 615.190727
• DOI: 10.1080/19466315.2017.1388835
• Languages: English
• ISSNs: 1946-6315

Abstract

ABSTRACT: During assay development, one of the key steps in designing a quantitative assay procedure is to choose an appropriate calibration model for the standard curve. A recent draft USP general chapter <1210> discussed drawbacks of the traditional hypothesis testing approach to selecting a calibration model and proposed two alternative approaches, AICc approach and two one-sided equivalence testing approach. However, the operating characteristics of the AICc approach have not been discussed in this context. In this paper, we uncover a connection between the AICc approach and traditional hypothesis testing approach and provide the operating characteristics of the former. These results can help us to clarify relationships between different calibration model selection methods currently discussed in the industry and are valuable for a revision and finalization of USP Chapter <1210>.