Fischer, Vivien; (2020) Rapid optimisation of preparative chromatography for monoclonal antibody purification using high-throughput methodology. Doctoral thesis (Eng.D), UCL (University College London).

Preview

Text EngD_Thesis_Vivien Fischer.pdf Download (11MB) | Preview

Abstract

A key challenge for early stage bioprocess development is the need to deliver effective operating conditions against aggressive timelines and stringent material constraints. For antibodies, the largest class of biopharmaceuticals to date, chromatography constitutes a key tool for product purification. The development of these steps typically involves multi-variate and multi-objective optimisation problems with various constraints and often complex trade-offs. This research investigates data-driven experimental optimisation strategies based on the Nelder-Mead simplex algorithm, and other more traditional approaches using DoE and fundamental chromatography understanding. Through three case studies, the polishing chromatography process was developed for a series of industrially-relevant binary and ternary mAb feeds. Bind-and-elute cation-exchange chromatography operations were optimised with regards to the loading and elution conditions, employing step and linear gradients. Experiments were conducted using conventional laboratory-scale chromatography. In the third case study, these were compared head-to-head with micro-scale chromatography methods, on a Tecan Fluent liquid handling station. Process insight was generated, and trends visualised and interpreted for key performance outputs such as product mass, recovery and productivity against pre-defined requirements for product purity. These trends were shown to be reproducible at micro-scale with comparable chromatographic resolution, employing residence time scaling. The proposed experimental optimisation strategies were shown to effectively deal with the high variability typical of antibodies, revealing interesting trade-offs for high loading conditions and in the vicinity of product breakthrough, leading to high-performing counter-intuitive operating regions. In these practical scenarios, the simplex was demonstrated to be an easy-to-use and dynamic tool that guides development efforts towards feasible and promising operating regions whilst also being interactive by accepting information from different sources to maximise the value gained from the available experimental data. Hence, this work has demonstrated how state-of-the-art experimental optimisation methods can be deployed for rapid and rational process development, enabling the identification of high-performing operating conditions that may not necessarily align with conventional assumptions and traditional expectations.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as CSVJSONXML

Downloads by country - last 12 months

Export as CSVJSONXML

Archive Staff Only

View Item