An investigation of the tumour-derived Glasgow Microenvironment Score and survival in patients with operable colorectal cancer

Alexander, Peter G. (2023) An investigation of the tumour-derived Glasgow Microenvironment Score and survival in patients with operable colorectal cancer. MD thesis, University of Glasgow.

Abstract

Background: Colorectal cancer poses a significant disease burden worldwide, remaining the 2nd cause of cancer related death. The TNM staging system, whilst being constantly improved, is limited by its assessment of the tumour alone and not the host. Whereas, as the knowledge base grows regarding of the consensus molecular subtypes in colorectal cancer and the different microscopic phenotypes that these may produce, further disease biomarkers are required that reflect this understanding. The Glasgow Microenvironment Score (GMS) was developed by combining assessment of two phenotypic assessments of the colorectal cancer (CRC) microenvironment, both of which have been shown to have independent prognostic significance: the immune phenotype (assessed by Klintrup-Mäkinen grade (KM)) and the mesenchymal phenotype, assessed by tumour stroma percentage (TSP). However, further understanding of the pathological mechanisms underlying these phenotypic features is required.

Methods: The present thesis examines the prognostic utility of the GMS in several Scottish cohorts in order to validate the score in independent patient cohorts and also to assess its utility in detecting disease recurrence and understand its relevance in the context of current chemotherapy.

Results: In chapter 3, associations between markers of Epithelial-mesenchymal Transition (EMT) and the Glasgow Microenvironment Score were assessed. GMS 0 was associated with lower membrane Fascin and also lower membrane and nuclear B-catenin. GMS 1 was associated with high cytoplasmic Fascin, whereas GMS 2 was associated with higher nuclear B-catenin, a hallmark of EMT.

In Chapter 4, several cohorts were examined in order to validate the GMS in independent cohorts, including the patients from the Scot chemotherapy trial (TransScot cohort). The GMS was found to stratify survival in all of these cohorts. Furthermore, GMS 2 was found to be a risk factor for poor survival in an otherwise low-risk group.

In Chapter 5, the role of GMS in predicting disease recurrence patterns was assessed both for CRC as a whole and also in colon and rectal cancer individually. GMS independently predicted recurrence at any location for CRC and also for rectal cancers, although in colon cancers alone, this was not independent. GMS was also able to predict local recurrence, but not independently of T-stage and N-stage. GMS 2 had the highest risk for recurrence and therefore enhanced surveillance in this subgroup is recommended.

Associations between GMS and chemotherapy was assessed in Chapter 6. Standard chemotherapy did not appear to be particularly effective against GMS 2 tumours, although this data had its limitations and requires to be assessed in other cohorts. In the TransScot cohort, survival of patients with GMS 0 was better with FOLFOX compared with CAPOX.

Finally, in Chapter 7, the role of PDL1 (CD274) in prognosis of colorectal cancer and also in terms of response to immunotherapy in current trials. Whilst the expression of CD274 on immune cells was associated with good prognosis, expression on tumour tissue was equivocal in terms of survival outcomes. There is insufficient evidence regarding CD274 as a marker of response to immunotherapy in CRC and this needs to be addressed moving forward.

Conclusions: GMS has been validated in independent patient cohorts and shown to stratify survival as in the original cohort. GMS 2 has utility both in identifying patients at high-risk of disease recurrence, but also potentially in selecting patients for specific chemotherapy regimen.

• Item Type: Thesis (MD)
• Qualification Level: Doctoral
• Subjects: R Medicine > RC Internal medicine > RC0254 Neoplasms. Tumors. Oncology (including Cancer)
• Colleges/Schools: College of Medical Veterinary and Life Sciences
• Supervisor's Name: Park, Dr. James H. and McMillan, Professor Donald C.
• Date of Award: 2023
• Depositing User: Theses Team
• Unique ID: glathesis:2023-83653
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 20 Jun 2023 10:50
• Last Modified: 20 Jun 2023 10:51
• Thesis DOI: 10.5525/gla.thesis.83653
• URI: https://theses.gla.ac.uk/id/eprint/83653

Related URLs

• Article DOI
• Article DOI
• Article DOI
• Article DOI
• Article DOI