Subscribe Bookmark RSS Feed



Aug 19, 2014

Using JMP® Scripts to Automate Generation of Graphs and Analysis for Large Molecule Manufacturing Changes

Jessica  Behrle, Senior Principal Biostatistician, Janssen R&D   

Yinglei Li, PhD, Senior Biostatistician, Janssen R&D

Barry Hogan, PhD, Senior Scientist, Janssen R&D

Yonghui Wang, PhD, Senior Scientist, Janssen R&D

Michael Nedved, PhD, Associate Director, Janssen R&D

For process changes (site, scale, formulation, etc.) made to biopharmaceuticals (e.g., proteins, DNA, vaccines), comparability studies of pre-change and post-change materials are required to ensure quality at product release and during storage. Degradation studies at various temperatures are used to assess differences in stability profiles over time for pre- and post-change product using analytical methods for various product attributes. In addition to pre- and post-changes, these studies take into account factors such as dosage. Unfortunately, there is no point-and-click approach to do pairwise comparisons when there are three or more estimated slopes. With a powerful tool – a JMP script – all data is modeled and pairwise comparisons of the simple linear degradation slopes are performed. For each of the 20 or more measured analytical responses, the JMP scripts graph the observed data as a function of time, create the analysis, and automate all output into one JMP Journal. With the abundance of data from a study and the increasing frequency of such studies at Janssen R&D, automation of the statistical output via JMP performed by either a statistician or an analytical development scientist provides opportunities for increased efficiency.