University of Cambridge
Primary biliary cholangitis: genetic risk, phenotypic stratification and disease pathogenesis
Abstract
dc:description.abstractPrimary biliary cholangitis (PBC) is a rare disease which occurs in fewer than 1 in 2,000 people, mainly women over the age of 40. It is an autoimmune condition and is phenotypically associated with other immune-mediated inflammatory disorders (IMIDs). It is characterised by progressive destruction of the small, intra-hepatic bile ducts, causing chronic cholestasis, progressive fibrosis and leading in many cases to cirrhosis and end stage liver disease. Treatment options are limited and the disease spectrum encompasses those patients with an adequate response to treatment and a transplant-free survival comparable to the general population, to those with an inadequate response and an increased risk of liver-related outcomes. As such, PBC remains a major cause of liver related mortality and morbidity. Our current understanding of PBC disease pathogenesis outlines a complex disease which is likely driven by environmental causes in the context of a genetically susceptible individual to the development of autoimmunity. To advance our understanding of PBC and to improve patient outcomes, we need to delineate the genetic architecture of this disease, to provide insight into its aetiology, the basis behind high and low risk disease and more specifically identify potential targets for pharmacological intervention. Several genome wide association (GWA) studies have therefore been undertaken, which have successfully identified more than sixty risk loci for the disease. Candidate variants and genes have been proposed at these risk loci. However, there are limitations in interpreting these candidate variants and genes. This impedes efforts to translate genetic observations into novel therapies. Many of these risk loci for PBC are also risk loci for other traits. This is an example of pleiotropy. In this thesis to exploit pleiotropy and identify causal variants and genes for PBC, I harnessed power across GWA studies and present statistical colocalisation, a Bayesian approach that probabilistically determines whether a given locus is associated with two or more traits; whether a single, pleiotropic variant accounts for the association with each trait; and which variant across the locus best explains that shared association. I applied this approach to summary statistics from a genome wide meta-analysis (GWMA) of PBC, other IMIDs and GWA studies of DNA methylation, histone marks, gene expression, and plasma proteins respectively, to identify risk loci, causal variants and genes associated with PBC risk. I also present a computational modelling method for phenotypic stratification of PBC patients. Phenotypic stratification of PBC patients is necessary to inform further studies aimed at understanding the difference between groups with higher or lower risk of end stage liver disease (ESLD). It can also influence treatment and long term monitoring of patients. There are many different phenotypes of PBC patients these include: those at higher or lower risk of ESLD, portal hypertensive and autoimmune phenotypes. Phenotypic stratification of PBC patients is also of value clinically as it can influence patient treatments and also monitoring strategies in clinical settings. This approach aimed to identify homogenous subgroups with distinct disease trajectories. I then proceed to present an application for this computational modelling approach, and characterise two extreme subgroups of PBC patients using transcriptional profiling to review the immunology of treatment response. Using this initial work as a proof of principle, I present further transcriptional profiling work from a larger PBC cohort to characterise disease biology, disease activity and the mechanisms involved in response to therapy and low risk compared to low risk PBC disease. The results of this thesis show that, firstly, statistical colocalisation can be used to further validate known risk loci, confirm suspected risk loci, and identify causal variants and genes associated with PBC. Secondly, that genes prioritised by statistical colocalisation can be used to complete in silico drug efficacy screening and identify new therapeutic agents for use in PBC, for example, Ustekinumab an anti-IL-12/23 monoclonal antibody used for Crohn’s disease and psoriasis or anti-TNF-α monoclonal antibodies such as, adalimumab, used in the treatment of rheumatoid arthritis, Crohn’s disease and ulcerative colitis. Thirdly, that computational modelling can be used for phenotypic stratification in PBC to identify four novel subgroups each with distinct disease trajectories and long-term outcomes. Finally, that transcriptomic analysis can provide fresh insight into PBC disease biology, activity and progression. Identifying networks of genes associated with treatment response and/or disease status in peripheral blood mononuclear cell subsets. In this thesis I therefore present novel insight into PBC disease risk, biology and future therapeutic targets.
Degree
thesis:*- Name dc:type.qualificationname
- Doctor of Philosophy (PhD)
- Level dc:type.qualificationlevel
- Doctoral
- Grantor dc:publisher.institution
- University of Cambridge
- Year dc:date.issued
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Mulcahy, Victoria
- Advisors dc:contributor.advisor
-
- Mells, George
- Sandford, Richard
Subjects
dc:subject × 3Rights
dc:rightsIdentifiers
dc:identifier.*- DOI dc:identifier.doi
- https://doi.org/10.17863/CAM.116985
- OAI identifier oai:identifier
- oai:www.repository.cam.ac.uk:1810/381998