Bayesian assurance and sample size determination in the process validation life-cycle. Issue 1 (2nd January 2017)

• Record Type: Journal Article
• Authors: Faya, Paul; Seaman, John W.; Stamey, James D.
• Is Part Of: Journal of biopharmaceutical statistics. Volume 27:Issue 1(2017)
• Page Start: 159
• Page End: 174
• Publication Date: 2017-01-02
• Subjects: Number of batches -- pharmaceutical manufacturing -- potency uniformity -- process capability -- qualification -- quality; Pharmacy -- Statistical methods -- Periodicals; Drugs -- Testing -- Statistical methods -- Periodicals; Biometry -- Periodicals; Biopharmaceutics -- Periodicals; Pharmacokinetics -- Periodicals; 615.19
• DOI: 10.1080/10543406.2016.1148717
• Languages: English
• ISSNs: 1054-3406

Abstract

ABSTRACT: Validation of pharmaceutical manufacturing processes is a regulatory requirement and plays a key role in the assurance of drug quality, safety, and efficacy. The FDA guidance on process validation recommends a life-cycle approach which involves process design, qualification, and verification. The European Medicines Agency makes similar recommendations. The main purpose of process validation is to establish scientific evidence that a process is capable of consistently delivering a quality product. A major challenge faced by manufacturers is the determination of the number of batches to be used for the qualification stage. In this article, we present a Bayesian assurance and sample size determination approach where prior process knowledge and data are used to determine the number of batches. An example is presented in which potency uniformity data is evaluated using a process capability metric. By using the posterior predictive distribution, we simulate qualification data and make a decision on the number of batches required for a desired level of assurance.