Perspectives on Pooling as Described in the ICH Q1E Guidance

• Activity Number: 411
• Type: Topic Contributed
• Date/Time: Tuesday, August 6, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Stan Altan*+ and Jyh-Ming Shoung and Yan Shen and Areti Manola
• Companies: and Janssen R&D and Janssen R&D and Janssen R&D
• Keywords: Stability modeling ; ICH Q1E ; Mixed Modeling

Abstract

The June 2004 ICH Q1E guidance lays out a pooling strategy applicable to stability studies carried out according to the principles given in the Q1A(R2) guidance. The strategy assumes a completely fixed model to describe multiple batches in which hypothesis testing for poolability of slopes is followed by poolability of intercepts, with the criterion for poolability set at p=0.25. The approach is extended when pooling across fixed factors, such as containers or strengths. These recommendations have been in place for decades and tilt toward statistical considerations with little regard to chemical or process engineering considerations. The guidance does not acknowledge the complexities of a typical stability study in development. Moreover, current technology allows the routine use of mixed modeling and Bayesian hierarchical modeling which represent an advanced approach to characterizing batch manufacturing processes. This presentation proposes a mixed modeling framework and reshapes the pooling recommendations given in the Q1E guidance based on scientific and empirical claims.