(228d) GSK Design Space Verification Lab Pilot Project

Design Space Verification Lab Pilot Project

This presentation describes the implementation of a Design Space Verification (DSV) lab for GSK drug substance process development and the key learning outcomes.  The DSV Lab is a pilot-scale project to develop and deploy reaction and crystallization platforms at lab and intermediate scale that enable faster, more efficient scale-up verification.  The DSV Lab includes proven monitoring and control technology.  This hardware is connected to an advanced informatics system that aggregates process, PAT and off-line data into a single comprehensive database.  Finally, a variety of modeling tools are provided that efficiently access the database to compare experiments run at lab and intermediate scale

QbD-based process development heavily relies on generating large amounts of data.  The time required to manage this data is typically greater than that required for analysis, making process development less efficient.  The DSV lab improved the efficiency of managing large data sets by collecting, integrating, and syncing data from multiple sources i.e. spectroscopic, process, images, soft sensors into a single repository using an open data format.  Similarly, advanced analytical and visualization tools were built into the DSV platform to improve the efficiency of data extraction and analysis by Chemists, Engineers, Particle Scientist and Modelers.    

Scalability Risk Assessment (SRA) is an integral part of process development with respect to late phase projects.  GSK is continually looking for better ways to assess scale-up risk.  Furthermore, the ability to extrapolate Design Spaces identified at lab scale to plant scale has been questioned by regulatory authorities in recent filings, sometimes leading to a need to do extensive verification protocols at manufacturing scale.  The DSV platform and the data sources were assembled across lab (1L) and intermediate-scale reactors (16 L) which can be utilized to mitigate risks identified in SRAs by verifying scaled-up process predictions.  Experiments were conducted at two scales to compare and predict the scale-up of empirical and mechanistic process models, utilizing statistical and chemometrics analysis to gauge the accuracy of these predictions. 

Ultimately, GSK aims at sharing the progresses made with respect to scale-up verifications and lab informatics amongst pharmaceutical manufacturers and regulators.  The DSV project is expected to further the discussions related to data standards (S88) and open data formats, as proposed by Allotrope.  The DSV lab is proposed as a scale-up verification tool to efficiently and accurately address regulatory scale-up concerns.