Molecular evolution of aggressive early stage colorectal cancer

Abstract

Colorectal cancer (CRC) is a heterogenous disease and many groups have taken advantage of inter-patient heterogeneity to stratify CRC tumours, using transcriptional or histological data, into subtypes/subgroups with similar molecular and pathological features. A recurring poor prognostic feature in each of these studies is the enrichment of a stromal components within the tumour microenvironment, which is a defining characteristic of the consensus molecular subtype 4 (CMS4). In addition to being the worst prognostic subtype relative to the other CMS’s, CMS4 tumours derive no benefit from current standard of care therapies meaning that there is an urgent need to identify efficacious therapeutic interventions for this subtype. To address this issue, this project focused on identifying the difference between stroma-rich high fibroblast (HiFi) tumours that relapse compared to those that don’t relapse after surgery.

Using an in silico analysis comparing the non-relapse and relapse HiFi samples, a n=7 gene list was created called the HiFi-specific prognostic signature (HPS), which was able to stratify the HiFi tumours into good prognostic (high HPS) and poor prognostic (low HPS) subgroups. To characterise the biology underpinning the HPS, differences within the multiple compartments of a tumour ecosystem were investigated, including the stroma and epithelial cells (chapter 1) and immune cells (chapter 2). Tumours with high HPS signalling were enriched for a double-stranded RNA/viral response upstream of an interferon signalling cascade, represented both by gene set analysis and transcription factor activity. This biology was coupled with the activation of an immune response and an upregulation of antigen processing and presentation. The potential value of using the synthetic dsRNA compound Polyinosinic-polycytidylic acid (poly(I:C)) to induce these HPS-related biological phenotypes was identified and validated across multiple cell lines. The therapeutic efficacy of poly(I:C) in a stroma-rich setting was also investigated within an in vivo experiment, where the treatment resulted in significantly lower metastatic burden than the control group (chapter 3). The application of poly(I:C)-induced biology along with the understanding of HPS biology then enabled CRC samples to be aligned across an immune landscape, supporting potential therapeutic positioning of poly(I:C).

In summary, this work supports the use of poly(I:C) as a new therapeutic option for a subgroup of CRC tumours that have the worst prognosis along with no current therapeutic intervention.

Thesis is embargoed until 31 July 2028.

Date of Award	Jul 2024
Original language	English
Awarding Institution	Queen's University Belfast
Sponsors	Northern Ireland Department for the Economy
Supervisor	Mark Lawler (Supervisor) & Philip Dunne (Supervisor)