Thesis

The impact of single-point mutations and IgG subclass on the developability of high concentration monoclonal antibody formulations

Downloadable Content

Download PDF
Creator
Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2024
Thesis identifier
  • T17343
Person Identifier (Local)
  • 202151789
Qualification Level
Qualification Name
Department, School or Faculty
Abstract
  • The developability of therapeutic monoclonal antibodies (mAbs) is a growing field of research poised to increase the probability of successful clinical translation for early-phase mAb candidates. Developability assessments typically entail high-throughput, low volume biophysical assays with parallel in silico sequence and structure based predictions to scope manufacturing, safety and efficacy risks. These include colloidal and conformational stability, and solution viscosity, which impact formulation shelf-life, immunogenicity, and manufacturability risks. High viscosity presents challenges with increased filtration pressure and reduced recovery during processing steps, with implications for vial filling during manufacture, and injection failure during administration. The latter is a growing concern with the move to patient self-administration using subcutaneous devices for improved patient autonomy and adherence. The dose volume limitations in autoinjector device design and high dosing requirements for mAb potency further complicates viscosity associated risks in mAb solutions. In this thesis, the impact of single-point mutations introduced in IgG1 variable regions, and different mAb subclasses on viscosity and other biophysical developability properties was investigated. Chapter 2 investigated the use of computational molecular descriptors derived from homology constructs to engineer mutants and the role of solvent accessible surface potential in promoting mAb self-interactions was assessed. Mutations with significant reductions in hydrophobicity resulted in lower solution viscosity, and a lack of correlation with in silico descriptors demonstrate the need for case-by-case evaluation of mAbs. Whilst many studies explore the design of mutants to enhance viscosity, few demonstrate the impacts of these mutations on manufacturing process and critical quality attributes. Chapter 3 explored the manufacturability of the single-point Fv mutants, assessing upstream and downstream process observations as well as phase behaviour and process-related impurities. Significant modifications on mAb expression, required chromatography conditions, phase stability and post-translational modifications were observed and were mutation site-specific. Chapter 4 presented an insight into the reduced developability of an IgG3 relative to IgG1, with comparisons to in silico descriptors. This chapter targeted the knowledge gap in understanding the biophysical behaviour of the IgG3 subclass which holds unique therapeutic potential. The results in this chapter also demonstrate the impact of the constant domain sequence and structure on interactions governing viscosity. Overall IgG3 showed reduced developability, with increased viscosity, compared to the Fv-matched IgG1 ortholog assessed. Finally, the use of viscosity models in fitting and predicting formulation behaviour was evaluated in chapter 5. This chapter expanded upon interpretation limits of viscosity with regards to model-fit equation used and the concentration-dependence of viscosity, highlighting changes in contributing underlying mechanisms with increased molecular crowding. While low-concentration hydrodynamic parameters provided insights into such mechanisms, they poorly correlated with ultra-high concentration viscosity. Furthermore, the lack of generalisability of predictive models explored in this chapter highlights the necessity for machine-learning modelling to incorporate larger, diverse datasets for robust and accurate viscosity predictions of high concentration mAb formulations. Overall, this thesis has provided a framework for the combined computational and experimental assessment of the biophysical behaviour of mAbs in high concentration formulations. Mutants designed from targeting computed surface patches were ineffective in reducing viscosity in the dose relevant concentration regime, so future works include combining mutations, exploring mechanistic contributions to viscosity further and use of machine-learning models for directed mutagenesis.
Advisor / supervisor
  • Rattray, Zahra
  • Lewis, William
Resource Type
Note
  • Previously held under moratorium in Chemistry Department (GSK) from 22nd November 2024 to 27th May 2025.
DOI
Date Created
  • 2024
Funder

Relations

Items

Thumbnail Title Date Uploaded Visibility Actions
file details PDF of thesis T17343 2025-05-27 Public Download