Investigating CD8+ T-cell gene expression signatures as disease prognostic biomarkers in paediatric inflammatory bowel disease.

Background: Currently, it is not possible to predict disease behaviour for children with inflammatory bowel disease (IBD), which is a major obstacle in an era where we strive to deliver personalised, tailored therapy. Previous investigation of gene expression profiles from CD8+ T-cells in adult IBD cohorts identified promising signatures, including a T-cell exhaustion signature, to predict disease outcome in these patients. Hypothesis and aim: We hypothesised that adult CD8+ T-cell prognostic signature and T-cell exhaustion signature would also predict outcome in paediatric IBD. We also hypothesised that CD8+ methylation profiles would underpin changes in gene expression, hence providing an alternative potential predictor. The aim of this project was to test whether CD8+ T-cell gene expression and methylation signatures can predict disease outcome in children with IBD. Methods: Purified CD8+ T-cells from a prospective cohort of 112 children newly diagnosed (treatment naïve) with IBD were subjected to cellular genome-wide RNA and DNA profiling (affymetrix and epic methylation microarrays). Detailed clinical information from each patient was recorded in a clinical database (1.5 years follow-up). First, the adult CD8 prognostic signatures were applied to the paediatric data in order to test for their ability to differentiate children according to their disease behaviour. Subsequently, the paediatric data was analysed on its own through unsupervised clustering analysis and Weighted Gene Co-expression Network Analysis (WGCNA) to test for correlations between gene expression data and clinical outcome parameters. Survival analysis (kaplan meyer) was used to compare patient groups for disease outcomes, including number of treatment escalations, use of biologic treatments and surgical intervention. Results: Applying the adult CD8 prognostic signature and the T-cell exhaustion signature to the paediatric dataset did not generate groups with significant differences in disease outcomes. Furthermore, the clinical data collected from the paediatric cohort showed that two thirds of the children had at least two treatment escalations, compared to less than 40% of the adult patients from the previous study. The analysis of the paediatric data per se identified correlations with clinical outcomes including use of biologics in Crohn’s (WGCNA correlation index (CI) < 0.4) and surgical intervention in ulcerative colitis (top CI: +0.38 and – 0.59). Preliminary analysis of the CD8 methylation profile did not show any correlation with clinical outcomes in this paediatric cohort. Conclusion: The adult prognostic CD8 signatures did not prove to be effective in children with IBD. We speculate that this could be due to the paediatric IBD phenotype being homogeneously more severe. Our findings hint the hypothesis that absent T-cell exhaustion in paediatric CD8+ T-cell could underlie a more severe disease phenotype in children. Further understanding of the mechanism of T-cell exhaustion in children has the potential to open up to future target options in paediatric IBD.

Name dc:type.qualificationname

Doctor of Medicine (MD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2019

Author dc:creator

• Gasparetto, Marco

Advisor dc:contributor.advisor

• Zilbauer, Matthias

• paediatric inflammatory bowel disease
• prognostic biomarker
• CD8+ T-cells
• gene expression signature

Licence

• https://www.repository.cam.ac.uk/bitstreams/278f7e8b-dae5-48d4-b36f-cf97ef6bb200/download
• https://www.rioxx.net/licenses/all-rights-reserved/

Language dc:language

en

DOI dc:identifier.doi

https://doi.org/10.17863/CAM.37525

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/290297