What is the prognostic role of γẟ T cells in colorectal cancer?

Hayman, Hannah-Louise Jessica (2022) What is the prognostic role of γẟ T cells in colorectal cancer? PhD thesis, University of Glasgow.

Abstract

Colorectal cancer (CRC) is the third most common cancer in men and second most common in women worldwide, rising from fourth and third most common in 2002, respectively [1-3]. The Tumour Node Metastasis (TNM) staging system was devised to stratify CRC patients by their disease attributes; the penetration of the tumour into surrounding tissue and spread to lymph nodes and distant organs; but this method warrants significant improvement [4, 5]. Attempts to resolve this issue have produced a genomic and transcriptomic method from bulk tumour sequencing (consensus molecular subtypes) which classifies patients as CMS1 through to CMS4 [6]. Due to its origin from bulk sequencing the CMS is hindered by a lack of tumour specificity, and so the cell intrinsic subtypes were developed to utilise characteristics of the tumour cells only, producing CRIS-A through to CRIS-E [7]. Given the significant contribution that the immune landscape plays in CRC, there is significant interest in utilising the immune landscape to further hone CRC stratification, resulting in the Immunoscore® which utilises counts of CD3+ and CD8+ lymphocytes [8, 9]. However, the immune landscape includes a variety of immune cells with may hold additional information for patient stratification, including specifically unconventional γδ T cells in CRC due to their relative enrichment, and so this thesis seeks to examine the representation of γδ T cells within the primary tumour and adjacent normal tissue and determine their prognostic value, comparing this to the well characterised CD8 T cells.

To accomplish this goal, immunostaining was employed to measure the density of γδ T cells and CD8 T cells within the primary tumour and adjacent normal tissue of patients from three geographically distinct CRC cohorts representing Scotland, Norway, and Thailand. It was found that both lymphocyte populations were reduced in the primary tumour compared to the adjacent normal tissue and in the epithelium compared to the stroma/lamina propria. These data were used to histologically classify cases as low or high for γδ or CD8 T cells and these groups used to conduct survival analysis, determining that patients with a high density of the conventional CD8 T cells had a favourable prognosis, as expected from the literature [8]. However, contrary to previous studies which utilised a flawed transcriptional methodology [10, 11], γδ T cells were found to be an unfavourable prognostic factor. Furthermore, these results were transcriptionally validated by reclassifying cases based on the transcription of genes encoding γδ TCR and CD8 coreceptor chains. This not only validated the histological findings, but also highlighted that supervised analysis using these genes may be an avenue to improving transcriptional methods of lymphocyte deconvolution.

Following on from this, analysis of genomic and transcriptomic sequencing data was conducted to elucidate the genomic and transcriptomic landscapes underlying cases that are histologically classed as low or high for γδ T cells or CD8 T cells. This analysis revealed that cases with a greater density of γδ T cells or CD8 T cells are more likely to carry mutations in genes which are often mutated in the microsatellite regions of MSI cases (ASTE1, TTK), and are associated with a greater immune infiltrate. However, MMR deficiency, which is responsible for the development of MSI, was not sufficient to result in the greater immune infiltrate. Thus, this study hypothesises that MMR deficiency and subsequent MSI does not result in greater immune infiltration, but rather lays a foundation for which genetic aberrations can activate a route to immune infiltration. Unfortunately, transcriptional analysis was not informative.

This thesis has demonstrated a clear difference in the prognostic role of γδ T cells and CD8 T cells, highlighting the need to extend our measure of the immune landscape to include less well characterised immune cells. However, doing so may require a multiplex approach to better determine the role of these cells relative to each other, as this analysis produces two singleplex layers of information without details of their relationships. In addition, it not fully understood how these cells are functioning to exert such contrasting prognostic effects, and so further studies to characterise their relative function roles would be valuable. In conclusion, this thesis adds a histological component to the growing data showing that the immune landscape is a prime candidate to improve disease stratification for CRC patients, and ultimately further our ability to give them the best outcome.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Subjects: R Medicine > RC Internal medicine > RC0254 Neoplasms. Tumors. Oncology (including Cancer)
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Cancer Sciences
• Supervisor's Name: Coffelt, Dr. Seth, Edwards, Professor Joanne and Roseweir, Dr. Antonia
• Date of Award: 2022
• Depositing User: Theses Team
• Unique ID: glathesis:2022-83387
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 31 Jan 2023 10:26
• Last Modified: 31 Jan 2023 10:27
• Thesis DOI: 10.5525/gla.thesis.83387
• URI: https://theses.gla.ac.uk/id/eprint/83387