Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.746395
Title: Antibody therapeutic prediction and design : immunogenicity and stability
Author: Northey, T. C.
ISNI:       0000 0004 7231 5175
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
One third of all drugs currently in development are monoclonal antibody therapeutics. Key to the efficacy and safety of therapeutic antibodies is the avoidance of immunogenic reaction upon administration. Immunogenicity describes a patient’s immune response to an antibody therapeutic that leads to poor drug efficacy and allergic reaction that can vary in severity. Immunogenicity is influenced both by the presence of T and B cell epitopes and by antibody stability. The aim this thesis was to develop in silico methods to aid the selection and design of antibody therapeutics by focusing on the prediction B-cell epitopes and biophysical stability. IntPred, an in-house high-performance general protein-protein interface predictor, was applied to the problem of B cell epitope prediction. Tested on set of antigen structures, IntPred was shown be unable to predict B cell epitopes. Thus, IntPred was retrained and amended to create the IntPred:Epi method. IntPred:Epi was able to outperform all methods tested, except SEPPA 2.0. However, further testing on larger test set shows that IntPred:Epi outperforms SEPPA 2.0. Next, the influence of ‘tolerated surfaces’ on the selection of epitopes was considered. Specifically, libraries of surfaces were created through analysis and generalization of human and mouse PDB structures. From these, descriptors were generated to describe the tolerance state of an antigen surface. These were then applied as a novel label in the B cell epitope prediction problem. Using tolerance labels as a simple filter on IntPred:Epi predictions, performance was improved on a test set of human antibody-bound antigens. The efficacy and immunogenicity of a therapeutic antibody is also affected by its stability. Natural VH-VL pair human Fabs were expressed and sequenced, and hydrophobicity and thermal stability data were generated. These data were then used to investigate the sequence determinants of the biophysical properties of antibodies. An exploratory analysis revealed promising correlations between certain sequence features and biophysical properties that can be applied in the future for the prediction of biophysical stability.
Supervisor: Martin, A. C. R. ; Shepherd, A. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.746395  DOI: Not available
Share: