Open Collections

Interrogating the TCR-pMHC complex in health and disease using immunogenomics methods

University of British Columbia

The adaptive immune system is a complex network of cells working towards a common goal: detection and elimination of foreign cells that can harm the host. In cancer, malignant cells acquire mutations which can appear foreign to the adaptive immune system. The immune cells most directly involved in destruction of cancer cells are CD8+ T cells, using their T cell receptor (TCR) to recognize mutated peptides presented on cancer cells in the context of class I Major Histocompatibility Complex (MHC) molecules (pMHC). Immunogenomics methods offer ways to interrogate this TCR-pMHC interaction using genomics data. The aim of this thesis is to adapt and apply novel and existing immunoinformatic methods to cancer datasets to identify relationships between the immune system and cancer in a pan-cancer context. This involves prediction of cancer neoantigens derived from single nucleotide variants (SNVs) from tumours, and correlation of this neoantigen burden with outcomes and markers of immune inhibition. It involves extraction of TCR sequences from RNA-seq datasets to gain value-added information from these existing datasets, with demonstrated utility in solid tumours and lymphomas. Finally, it defines and explores the size of the self-immunopeptidome to classify individuals based on their ability to present peptides on class I MHC molecules. I show that T cell infiltration of solid tumours correlates with improved outcomes, neoantigen load, but also markers of T cell inhibition, suggesting that these individuals would benefit from checkpoint blockade therapy. In established tumours, the T cell repertoire is not clonal, and among the most abundant T cells in the tumour are viral-specific T cells also found in the normal repertoire. This information is obtained directly from existing RNA-seq datasets of tumours. When applied to RNA-seq of sorted T cell populations, clonally expanded T cells are detectable by their TCR, and alpha-beta pairing can be inferred. The self-immunopeptidome can be used to predict neoantigen load and is used to infer signatures of neoantigen immunogenicity. This thesis contributes towards a better understanding of the interaction between T cells and cancer cells, which can inform future strategies to improve immunotherapies in cancer.

Text

2018-12-12

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/68056

Genome Science and Technology

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/