An investigation into the relationship between mode of presentation, clinicopathological factors and outcomes in colon cancer

Golder, Allan Matthew (2023) An investigation into the relationship between mode of presentation, clinicopathological factors and outcomes in colon cancer. PhD thesis, University of Glasgow.

Abstract

Colorectal cancer is the 4th most common cancer in the United Kingdom and the second most common cause of cancer related death after lung cancer. Resectional surgery remains the cornerstone of treatment with curative intent however, despite this, a large proportion of patients eventually succumb due to recurrent or metastatic disease.

Despite the widespread introduction of bowel cancer screening programmes, a significant proportion of cases of colorectal cancer continues to require investigation and treatment on an emergency basis. Emergency presentations have been reported to have significantly worse short-term and long-term outcomes than elective presentations even after adjustment for disease stage. It seems likely that as opposed to emergency presentations per se being associated with adverse outcomes in colorectal cancer, clinicopathological factors – tumour, host and other factors – are likely to be associated with emergency presentation and that it is these factors that are associated with adverse oncological outcomes.

The work presented in this thesis examines the impact of emergency presentation on short-term and long-term outcomes of patients with colorectal cancer. It examines, in detail, the association between mode of presentation and tumour and host factors in patients undergoing treatment with curative intent for colon cancer and subsequently the association between these factors and long-term oncological outcomes.

Chapter 1 provides an overview of colorectal cancer including epidemiology, risk factors, routes to presentation, presenting symptoms and signs and the investigation and management of patients with colorectal cancer.

Chapter 2 examines 30 years of published literature in a systematic review and meta-analysis and summarises the existing literature regarding the association between mode of presentation and tumour and host factors in patients with colorectal cancer. The results conclude that there are multiple differences in tumour and host factors between elective and emergency presentations of colorectal cancer. However, the studies identified were heterogenous, and it was not possible to carry out a review of the effect of these factors on short-term and long-term outcomes.

Chapter 3 examines the association between mode of presentation and basic clinicopathological factors within a regional cohort of patients presenting with colon or rectal cancer in the West of Scotland regardless of disease stage or treatment received. The results show that patients with colon cancer are more likely to undergo investigation and definitive treatment on an emergency basis in comparison to rectal cancer. Patients presenting emergently with colorectal cancer were more likely to have advanced disease at diagnosis. Furthermore, in a subgroup analysis of patients undergoing curative resectional surgery for TNM Stage I-III colon cancer, emergency presentation was associated with adverse short-term and long-term outcomes even after adjustment for disease stage.

Chapter 4 examines the association between basic clinicopathological factors (tumour and host factors identified within Chapter 2), mode of presentation and short-term and long-term survival within a regional cohort of patients undergoing resectional surgery with curative intent for TNM I-III colon cancer.

Younger age, increased comorbidity (as measured by ASA classification), lower BMI, more advanced T stage and extramural venous invasion were associated with both emergency presentation and with adverse oncological outcomes. However, emergency presentation remained independently associated with both adverse short-term survival and long-term oncological outcomes despite adjustment for these factors. Increased co-morbidity as measured by the Charlson Co-morbidity index was not associated with emergency presentation. When the association between mode of presentation and individual components of the Charlson Index was examined, only Diabetes Mellitus was associated with mode of presentation and was protective against emergency presentation. Within a subgroup analysis of patients with Diabetes Mellitus, no clear association between diabetic factors (Type 1 vs Type 2 Diabetes, type of diabetic control, metformin/sulfonylurea/insulin use) and mode of presentation was identified.

Chapter 5 examines the association between the systemic inflammatory response, mode of presentation and short-term and long-term survival in a regional cohort of patients undergoing resectional surgery with curative intent for TNM I-III colon cancer. Both the neutrophil-lymphocyte ratio and the modified Glasgow Prognostic Score were independently prognostic and combined into a Systemic Inflammatory Grade. This Systemic Inflammatory Grade was independently associated with emergency presentation. When the association between clinicopathological factors, including mode of presentation and Systemic Inflammatory Grade, and short-term and long-term outcomes were analysed, Systemic Inflammatory Grade remained independently associated with short-term and long-term survival. Mode of presentation remained associated with short-term but not long-term survival.

Chapter 6 examines the association between mode of presentation and CTderived body composition. High subcutaneous fat index and low skeletal muscle index were independently associated with emergency presentation and were associated with Systemic Inflammatory Grade even after adjustment for TNM Stage.

Chapter 7 examines the prior interaction with the bowel screening programme of a regional cohort of patients diagnosed with colorectal cancer. Only 19% of patients were diagnosed through screening. Screening diagnosis was associated with significantly improved long-term outcomes. The most common reasons for failure to diagnosis through screening were non-invitation to screening (either above or below routine screening age), non-return of screening test (associated with male sex, increased socio-economic deprivation, increased comorbid status and current smokers) and negative screening test (associated with female sex, preoperative anaemia, less comorbid status, right-sided tumours and screening with gFOBT testing).

Chapter 8 examines the association between tumour mutational status, mode of presentation and long-term outcomes in patients undergoing resectional surgery with curative intent for TNM I-III colon cancer. The results show that on unadjusted analysis, APC wild-type, KRAS mutant and BRAF wild-type colon cancer were associated with improved long-term outcomes. There may be an association between KRAS mutant status and an elevated systemic inflammatory response. On adjusted analysis, KRAS mutational status was independently associated with adverse long-term outcomes after adjustment for other clinicopathological factors. In this study, no statistically significant associations were seen between mutational status and mode of presentation however there were trends between P53 wild-type, KRAS mutant and PIK3CA mutant status and emergency presentation.

Chapter 9 examines the association between the preoperative systemic inflammatory response, emergency presentation and short-term and long-term outcomes in patients undergoing resectional surgery with curative intent for TNM Stage II colon cancer when controlled for the established high-risk factors of TNM Stage II disease. The results show that after adjustment for these factors, emergency presentation was not independently associated with either shortterm or long-term outcomes however a significant association was seen between the preoperative systemic inflammatory response and outcomes.

Chapter 10 presents the results from a national survey with regards to attitudes towards and the use of perioperative steroids in patients undergoing resectional surgery with colorectal cancer. The results show that perioperative steroids are widely used at the discretion of the anaesthetist with the primary aim of preventing postoperative nausea and vomiting. The results show that there is sufficient equipoise to carry out a randomised controlled trial examining the impact of single dose corticosteroid administration at induction of anaesthesia on the postoperative systemic inflammatory response and outcomes following colorectal resection.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Subjects: R Medicine > RC Internal medicine > RC0254 Neoplasms. Tumors. Oncology (including Cancer); R Medicine > RD Surgery
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Medicine, Dentistry & Nursing
• Supervisor's Name: McMillan, Professor Donald, Horgan, Professor Paul and Roxburgh, Professor Campbell
• Date of Award: 2023
• Depositing User: Theses Team
• Unique ID: glathesis:2023-83808
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 07 Sep 2023 11:45
• Last Modified: 11 Sep 2023 07:30
• Thesis DOI: 10.5525/gla.thesis.83808
• URI: https://theses.gla.ac.uk/id/eprint/83808

Related URLs

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