Background: Chronic low back pain (CLBP) is a highly prevalent condition that has substantial impact on patients, the healthcare system and society. Its aetiology is complex and the condition can be exacerbated by many psychological, physical and social factors. Pain management services (PMS), which aim to address the complex nature of back pain, are recommended in clinical practice guidelines to manage CLBP. Although the effectiveness of such services has been widely investigated in relation to CLBP, the quality of evidence underpinning the use of these services remains moderate. Given that these services are resource intensive, evidence is needed to determine their cost-effectiveness. Aim: This study aims to describe the patient characteristics, clinical outcomes and healthcare resource use of people with CLBP in a community-based pain management service (PMS) in Nottingham to derive an indicative estimate of the cost effectiveness of PMS compared with standard care (SC). Methods: The study followed the Medical Research Council (MRC) guidance for evaluating complex interventions. The MRC suggest conducting developmental and observational work before evaluating complex interventions on a larger scale. Therefore, this PhD research includes a service evaluation study, which was conducted in two community-based PMS in Nottingham. This was followed by a systematic review of the cost effectiveness of a PMS in CLBP. Finally, a decision analysis model was developed to assess the cost effectiveness of PMS compared with SC. In the service evaluation study, newly referred people with CLBP who provided written consent were included. Participants provided information on health status and healthcare resource use using postal questionnaires and diaries at baseline and then three and six months after recruitment. The outcome measures were the Brief Pain Inventory (BPI), the Roland Morris Disability Questionnaire (RMDQ) and the EuroQoL (EQ-5D-3L). In the systematic review, electronic searches were conducted in clinical and economic databases from their inception to August 2017. Full economic evaluations, undertaken from any perspective, conducted alongside randomised clinical trials (RCTs) or based on decision analysis models were included. The Cochrane Back Review Group (CBRG) risk assessment and the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) checklist were used to assess the methodological quality of eligible studies. These studies informed the development of an economic evaluation based on a decision analysis model to compare PMS with SC. Costs per extra quality-adjusted-life-year (QALY) were calculated from the UK National Health Service (NHS) England perspective using a lifetime horizon. Transition probability, utilities, healthcare resource use and costs were obtained from the service evaluation study and the published literature. Results: In the service evaluation study, 32 people were recruited over 10 months. The mean age was 58.8 years with an equal distribution of both genders. The dropout rates were 21% and 25% over three and six months respectively. Disability was measured using RMDQ (with a score range between 0-24), where higher scores mean more disability. The mean score at baseline was 13.5, after three and six months the means were 11.9 and 11.1 respectively. The BPI was used to measure pain intensity and interference. The scores ranged from 0 to 10, where higher scores mean more pain. The mean score for pain intensity was 5.9 at baseline, followed by a mean of 5.5 and 5.6 at three and six months respectively, whereas pain interference was 5.9 at baseline, followed by a mean of 5.2 at three and six months. As for the EQ5D-3L, higher scores in EQ5D means better health states. The mean score was 0.38 at baseline and, after three and six months, the scores were 0.46 and 0.40 respectively. The mean number of health visits per patient between baseline and three months was 4, whereas between four and six months the mean number of visits per patient was 2.9. In the systematic review, five studies fulfilled the eligibility criteria. The PMS varied significantly between studies in terms of the number of treatment modalities, intensity and the duration. The PMS was compared with either standard care, which varied according to the country and the setting or with two different surgical interventions. In this review, three out of five studies had a high risk of bias based on the design of the randomised controlled trials (RCTs). In addition, there were limitations in the statistical and sensitivity analyses in the economic evaluations. Therefore, the results from this systematic review need to be interpreted with caution. Finally, the indicative economic evaluation showed thatthe PMS was both more effective and more costly compared with SC. The mean incremental cost effectiveness ratio (ICER) was £761.43 and £706.98 per quality adjusted life years (QALY) gained in the deterministic and probabilistic analyses, respectively. At a ceiling willingness to pay (WTP) of £20,000, the PMS reaches a 51% probability of being cost-effective. This suggest that there is 51% probability that the PMS is both more effective and less costly. The incremental net monetary benefit generated a mean of £7884.07, indicating that the PMS is cost-effective compared with the SC at a WTP of £20,000. In the sensitivity analysis, the results were affected by changing the utility score for severe pain and increasing the initial cost of the intervention. In the scenario analysis, the incremental cost and QALYs were in favour of the PMS for people with severe pain at baseline. The probability of the PMS being cost effective for people with severe pain was 54%. In addition, the sustained reduction in treatment effect reduces the PMS probability of being cost effective to 48%. Finally, using the two months’ data to generate the transition probabilities demonstrated that PMS dominates the SC and the probability of the PMS being cost effective was 58% Conclusion: Community-based PMS have the potential to improve functional disability and pain interference for people with CLBP in primary care by providing well-timed access to a specialised pain management team, ensuring effective use of pain medicines and streamlining the treatment pathways based on the individual patient’s needs. A systematic literature review highlighted inconsistent evidence supporting PMS. The cost effectiveness studies included in this systematic review were alongside RCTs with a maximum follow up period of two years. As a result of the limitations of this type of economic evaluation, a decision analysis model was developed in order to assess the life time effectiveness of PMS compared with SC. The decision analysis model showed that PMS is more effective and costly compared with SC in the base case analysis; however, changing the source of transition probabilities from 12 months to two months demonstrated that PMS dominated SC, providing the potential for PMS to be cost-effective if high quality research is conducted to reduce the uncertainty around transition probabilities. The results were also sensitive to change in the utility score for severe pain, the initial cost of the PMS and using the PMS for people with severe pain at baseline. Therefore, further information is needed to assess the uncertainties in these parameters to provide a more robust estimate of the cost-effectiveness of PMS compared with SC.

Barrett’s oesophagus is a common condition in which the normal stratified squamous oesophageal epithelium is replaced by metaplastic reflux-induced glandular (“columnar”) mucosa (Jankowski, Barr, Wang et al. 2010;Playford 2005). Over the last three decades, the incidences of oesophageal adenocarcinoma (OA) and Barrett’s oesophagus have risen to the point that OA is now common in the United Kingdom, with Scotland having one of the highest rates in the world (Jankowski, Provenzale, & Moayyedi 2002). Unfortunately most cancers present at an advanced stage with five year survival less than 30% (Holmes and Vaughan 2007). Barrett’s oesophagus is associated with malignant progression via a recognised metaplasia-dysplasia-carcinoma sequence (Jankowski, Wright, Meltzer et al. 1999). The premalignant nature of Barrett’s oesophagus has powered intense clinical interest in the hope of eventually having an impact on the earlier diagnosis and treatment of dysplasia, and ultimately the prognosis of oesophageal adenocarcinoma. Despite years of research interest, Barrett’s oesophagus remains an enigmatic condition. The exact incidence is unknown, and it is recognised that not all patients with Barrett’s oesophagus will progress to adenocarcinoma. Current strategies aim to ascertain the presence of dysplasia, the current gold standard marker of malignant progression. However although Barrett’s mucosa is visible at endoscopy, the presence of dysplasia is difficult to diagnose as these areas tend to be focal and inconspicuous to the naked eye. Current systematic biopsy regimes are recommended, but can be fraught with sampling errors. Furthermore, the molecular mechanisms underlying Barrett’s metaplasia and progression to dysplasia remain unclear. Molecular risk biomarkers have been sought with modest success, and at present dysplasia remains the most reliable clinical marker. However dysplasia itself is not without limitations: focal dysplasia can be difficult to ascertain, with many biopsies sometimes necessary to detect it reliably (Abela, Going, Mackenzie et al. 2008). Inter-observer variability may cause over or under diagnosis, especially regarding LGD (Flejou 2005). Moreover, although patients with HGD are at elevated risk of progression to OA, few studies provide reliable data on rates of progression from HGD to OA, with estimates varying between 16-59% at five years (Reid, Blount, Feng et al. 2000;Schnell, Sontag, Chejfec et al. 2001;Shaheen and Richter 2009;Spechler SJ 2011). There is a real need, therefore, to be able to identify and treat those patients at greatest risk of malignant transformation, and reassure those at low risk. Without an improved molecular understanding of Barrett's metaplasia and progression to neoplasia, clinically useful prognostic biomarkers (allowing appropriate targeting of surveillance and therapy) will be delayed. The current challenges associated with Barrett’s oesophagus are 1) to accurately determine the rate of malignant progression of Barrett’s oesophagus and identify clinical risk factors, 2) to improve the endoscopic detection of dysplasia and early neoplasia allowing earlier diagnosis and treatment and, 3) to understand the molecular mechanisms involved in the initiation of Barrett’s metaplasia, and the pathways involved in disease progression. In an attempt to improve the care of patients with Barrett’s oesophagus within the West of Scotland, my thesis will address each of the main challenges associated with this puzzling condition at clinical, endoscopic and molecular levels. The hypotheses of my thesis are threefold - a) Patients with Barrett's oesophagus in the West of Scotland have high rates of progression to high grade dysplasia and oesophageal adenocarcinoma. b) The WavSTAT optical biopsy system will be able to correctly identify non-dysplastic and dysplastic Barrett's oesophagus. c) The Wnt signalling pathway is upregulated in Barrett's oesophagus and dysplasia. The aims of my thesis are as follows: 1)To present a general overview of the Barrett’s literature highlighting current clinical challenges 2)To examine the incidence of dysplasia and oesophageal adenocarcinoma in the West of Scotland by analysing a cohort of patients undergoing surveillance endoscopy 3)To review the current endoscopic imaging adjuncts for the diagnosis of Barrett’s oesophagus and dysplasia, and assess the role of optical biopsy forceps in determining the presence of dysplasia 4)To evaluate the role of Wnt signalling in Barrett’s oesophagus, from metaplasia to carcinoma in a mouse model, with complementary human studies Chapter 1 introduces the reader to Barrett’s oesophagus and highlights current areas of clinical challenge and debate. A universal definition of Barrett’s oesophagus does not exist and Chapter 2 explores the need for the presence of intestinal metaplasia in the diagnosis of Barrett’s oesophagus. Chapters 3 and 4 present original data from a West of Scotland Barrett’s oesophagus database, specifically analysing rates of dysplasia and adenocarcinoma and cause of death. This study suggests patients with Barrett’s oesophagus in the West of Scotland are at high risk of disease progression with almost 10% of patients dying from oesophageal adenocarcinoma. The results highlight the importance of a comprehensive surveillance in our “high risk” population - an ideal niche for future chemopreventative and molecular studies. In an attempt to improve the diagnosis of dysplasia in our West of Scotland population, Chapter 5 reviews current endoscopic imaging adjuncts used in research and clinical practice while Chapter 6 presents original data from a pilot study assessing the use of innovative optical biopsy forceps in the endoscopic diagnosis of dysplasia. While this technology is in its infancy and further changes in the algorithm are required, the optical forceps could be a promising tool for ongoing surveillance in high risk Barrett’s patients. Chapter 7 summarises the role of biomarkers in Barrett’s oesophagus, reviewing the literature and highlighting the lack of clinically useful markers of disease progression to date. The Wnt signalling pathway plays an important role in normal oesophageal (and intestinal) development, yet when aberrantly activated leads to carcinogenesis. To date, very little is known about the role of Wnt signalling in Barrett’s oesophagus. Chapter 8 presents the results of a mouse model of upregulated Wnt signalling and the interesting finding of dysplasia within the oesophageal mucosa. Chapter 9 therefore translates these results to the human population by assessing the role of Wnt signalling in Barrett’s metaplasia and dysplasia by immunohistochemical analysis of a panel of markers. The results suggest Wnt signalling is upregulated in Barrett’s dysplasia, particularly in high grade, and this may have a future role as a biomarker. Chapter 10 summarises the main findings of the thesis, and presents future directions.

Renal cancer is a malignancy which is not only increasing in incidence but there has also been an increase in mortality rates. There are various prognostic factors in renal cell cancer. We have demonstrated that some of these such as nuclear grading, tumour necrosis and systemic inflammatory response can be further refined to aid in prognosis but cannot be utilised at present to assess which would benefit from therapeutic agents when recurrence occurs. We investigated if SFK members are expressed in renal cancer. Eight SFK members were found to be expressed in renal cancer and were present to varying degrees. Furthermore, expression differed in organ confined disease and metastatic disease. Immunohistochemistry was employed to assess protein expression and activation of c-Src and SFK activity as well as the downstream marker FAK Y861. Analysis demonstrated that c-Src expression was associated with improved survival and expression of the downstream marker FAK Y861 was associated with poor survival and demonstrated a positive relationship with known prognostic factors. This would suggest that another SFK member was associated with poor survival. Dasatinib, a SFK inhibitor was utilised on renal cell lines, demonstrating a dose dependant reduction on cellular metabolic activity as well an increase in apoptotic rates. This would support that Dasatinib may be a useful therapeutic drug for RCC. Treatment with Dasatinib also demonstrated that expression of c-Src, SFK activity and FAK Y861 reduced in a dose dependant manner. It was necessary to further assess that another SFK member was responsible for poor prognosis and this was undertaken by silencing c-Src. Cellular metabolic activity rates increased following silencing c-Src and assessment of SFK activity (Src Y416) and FAK Y861 on cell pellets demonstrated no change suggesting that another SFK member is responsible for the phosphorylation of FAK Y861 and therefore responsible for poor survival. This would suggest that another SFK inhibitor and not c-Src inhibitors may play a role in the treatment of renal cell cancer and further work is required to ascertain which SFK member is responsible so that this can be targeted for treatment.

Colorectal cancer is the second commonest cause of cancer death in the western world. In addition to tumour factors such as depth of invasion, lymph node involvement and venous invasion it is increasingly recognised that host factors, are important determinants of survival. In particular the host local and systemic inflammatory responses are stage independent predictors of survival in operable disease. The present thesis further examines the prognostic importance of host and tumour factors in colorectal cancer, specifically: 1. An examination of the prognostic importance of venous invasion (detected using elastica stains) in colorectal cancer. 2. Detailed analysis of the determinants (including age, comorbidity and deprivation) of the systemic inflammatory response and their relationship with survival. 3. The application and validation of a prognostic score providing a measure of the local inflammatory response in colorectal cancer. 4. Detailed analysis of the determinants (including all white cells, lymphocytes and macrophages) of the local inflammatory response and their relationship with survival. 5. The inter-relationships between the local and systemic inflammatory responses in colorectal cancer specifically: early stage disease (node negative) and in patients receiving adjuvant chemotherapy. 6. Mediators (including immunological parameters and vitamin antioxidants) of the local and systemic inflammatory responses and their relationship with survival.

This project highlighted that Src and Src kinase family (SFK) members play a definitive role in breast cancer. Due to the paucity of translational studies, we investigated if SFK members are expressed in human breast tissue. Eight SFK members were present with distinct mRNA expression patterns in normal, non-malignant and malignant breast tissue. Immunohistochemistry was employed to investigate protein expression and activation of Src and SFK members. Survival analysis revealed that c-Src and activated Y419Src were associated with worse patient outcome, confirming current in vitro literature, whereas a different phosphorylation site of Src (Y215) and expression of Lck was associated with improved clinical outcome. Dasatinib was employed in different breast cancer cell lines to establish its effect on those phosphorylation sites. Decreased expression of c-Src and Y419Src was observed, whilst Y215Src expression stayed unchanged, providing a rationale for using this Src kinase inhibitor in clinical trials.

Colorectal cancer is the second most common cause of cancer death in the Western World. Although staging and prognosis is presently based on pathological assessment of primary tumour invasion and the presence of lymph node and distant metastases, it is increasingly recognised that other factors pertaining to both the tumour and host may similarly affect outcome. The local and systemic environment, encompassing host inflammatory responses and the tumour microenvironment, are examples of such. However, how such measures may compliment present TNM-based staging are not clear. Furthermore, tumour and host factors, both modifiable and non-modifiable, which may determine the local and systemic environment, remain to be fully determined. The present thesis examined the clinical and prognostic utility of assessment of the local and systemic environment, and potential tumour and host factors which may determine these responses. The following conclusions were drawn: Examining patients from the United Kingdom and Japan, Chapter 2 and 3 concluded that assessment of the systemic inflammatory response, utilising the modified Glasgow Prognostic Score, provides further prognostic stratification in addition to TNM stage. Although the proportion of patients exhibiting an elevated systemic inflammatory response differed between populations, the prognostic value was comparable. Chapter 4 validated assessment of the tumour stroma percentage as a prognostic factor independent of TNM stage and the local inflammatory cell infiltrate (cancer-specific survival HR 1.84, 95% CI 1.17-2.92, P=0.009). Chapter 7 further confirmed the prognostic value of a combined tumour microenvironment score, based on assessment of the generalised inflammatory cell infiltrate and tumour stroma percentage, in patients with primary operable colorectal cancer. This score, termed the Glasgow Microenvironment Score, was able to stratify patients into a good prognostic group, with five-year survival of 89%, an intermediate group with a two-fold increased risk of death and five-year survival of 75%, and a poor prognostic group, with a four-fold increased risk of death and five-year survival of 51%. Chapters 5 and 6 identified the presence of mismatch repair deficiency and activation of the JAK/STAT3 as two potential mechanisms which may determine host local and systemic inflammatory responses. However, the prognostic value of such candidate mechanisms was weak, suggesting that other pathways and tumour characteristics are implicated, and that molecular heterogeneity is likely to play an important role in determining not only the local and systemic environment, but also outcome. Chapter 9 concluded that the Immunoscore, an immunohistochemistry-based assessment of T-lymphocyte density within the tumour microenvironment, held greater prognostic value than assessment of the generalised inflammatory cell infiltrate using the Klintrup-Mäkinen grade. However, assessment of tumour stroma percentage provided additional prognostic value irrespective of the methodology employed to examine the local inflammatory cell infiltrate. Furthermore, the results of Chapters 7, 8 and 9 together suggested that loss of the local, anti-tumour immune infiltrate was the primary event which allows continued tumour growth, development of a tumour-supportive microenvironment and propagation of a systemic inflammatory response. Chapter 10 concluded that pre-diagnosis use of aspirin but not statins was associated with a lower modified Glasgow Prognostic Score, despite strong associations with comorbidity and BMI. This did not translate into an improvement in survival, potentially reflecting the underlying indication for use of these drugs primarily as cardiovascular secondary prevention medications. Finally, Chapter 11 examined the clinical utility of assessment of the tumour microenvironment using colonoscopic biopsy specimens, concluding that the use of biopsy-derived specimens was feasible. Furthermore, in addition to identifying patients who may benefit from therapies targeting the tumour microenvironment, assessment of a biopsy-derived Glasgow Microenvironment Score had comparable prognostic value to full section assessment of the tumour microenvironment.

Improving the survival of patients with pancreatic ductal adenocarcinoma (PDAC) remains an oncological and surgical challenge. PDAC pathogenesis is underlined by numerous molecular aberrations occurring at a genetic and epigenetic level, however their spectrum of occurrence and clinical impact has not yet been fully elucidated. The majority of patients present with locally advanced or metastatic disease and even the 15-20% of patients who undergo resection for cure have a median survival limited to 18-24 months. Surgical treatment carries a high morbidity and identification of patients expected to have a poor prognosis could assist in the decision making process. The present thesis examines the prognostic importance of pathological and molecular factors in PDAC, specifically: 1. An examination of the frequency, the prognostic impact of resection margin involvement, and furthermore the prognostic influence of tumour involvement at individual margins. 2. Determination of the prognostic impact of peripancreatic fat invasion following resection of PDAC. 3. The investigation of the relationship of candidate protein biomarker expression with overall survival in a large PDAC tissue microarray cohort using immunohistochemistry. 4. Determining gene expression profiles associated with pancreatic cancer compared to normal tissue using gene expression microarray analysis with subsequent development and validation of a prognostic gene signature. 5. microRNAs were identified that associated with pancreatic cancer clinicopathological factors including survival. 6. Copy number aberrations were identified using array comparative genomic hybridisation that correlated with clinicopathological factors following resection for PDAC. 7. Finally the identification of potentially important regulator genes contributing to pancreatic tumourigenesis, was made by integrating the gene expression, microRNA expression and copy number data from previous sections using a bioinformatic approach. In this work a combination of enhanced pathological staging criteria along with the correlation of molecular marker expression and genomic profiling signatures with clinical outcome data has yielded interesting results in patients undergoing resection for pancreatic cancer that allowed detailed disease characterisation and subsequent clinically relevant outcome stratification.

The postoperative period is characterized by massive shifts of fluid between body compartments and accumulation of fluid in the extracellular space, which may manifest clinically as central and or peripheral oedema. The incidence of oedema in patients undergoing routine major abdominal surgery (MAS) is unknown and there are no objective means of quantifying or monitoring its presence. Furthermore, the aetiological factors responsible for post-surgical oedema formation in patients with no overt signs of cardiovascular disturbance are poorly understood and the relationship between the development of oedema and clinical outcomes such as the recovery of gastrointestinal function, postoperative complications and duration of hospital stay is unclear. Observational studies were therefore conducted on patients undergoing MAS. The presence of oedema was related to changes whole-body impedance (Z), obtained at four frequencies (5, 50, 100 and 200 kilohertz (kHz)) using bioelectrical impedance analysis (BIA) and to clinical outcomes. The fluid intake and output and changes in plasma concentration of albumin, total protein, C-reactive protein (CRP) and reduced glutathione in whole blood (GSH) were compared before and after surgery in patients who subsequently developed oedema (OD group) and patients who consistently remained free of oedema (NOD group) Oedema occurred in 40 per cent of the patients observed prospectively and was significantly related to age (odds ratio 1.087 (95 per cent confidence interval (c.i), 1.016 -1.163; P =0.016). The preoperative ratio of Z at 200 kHz to 5 kHz (Z200/Z5) was higher in patients who subsequently developed oedema than those who did not (0.809 v 0.799; P = 0.015), suggesting that it may be possible to identify patients who are prone to abnormal fluid shifts preoperatively. The change in (Z) was greater in the oedematous than non-oedematous groups (at all frequencies (P < 0.001)), and more so at lower frequencies (5kHz) than higher frequencies (100 kHz) (P < 0.001). The impedance quotient (ht2/Z) in the whole group changed in a similar direction at each frequency but to a greater extent in the OD compared to NOD groups. The total volumes of administered fluids in both groups of patients were similar but the average urine output per kg body weight was significantly lower in the OD compared to NOD patients (29.4(2.3) versus 40.5(3.7) mls/kg, P = 0.023). There were no significant differences before and after surgery in the concentrations of albumin, total proteins and GSH in both patient groups. Preoperative CRP concentration in the OD and NOD patients were similar but the mean (s.d) CRP concentration over duration of observation in the OD compared to the NOD patients was significantly greater (148 (54.1) versus 89.6 (43.8) mg/L, P = 0.006). Oedema was associated with a significant delay in the recovery of gut function (median (range) 6(3-17) versus 5(1-13) days, P = 0.020) and prolonged hospital stay (17(8-59) versus 9(4-27) days, P = 0.001) and increased incidence of postoperative complications (65 versus 22%, P = 0.011). This study shows that the incidence of early postoperative oedema is high and preoperative identification and monitoring of surgical patients vulnerable to abnormal fluid shifts may be possible with non-invasive techniques. Age, impaired ability to excrete administered fluid load and an exaggerated inflammatory response to surgical trauma rather than hypoalbuminaemia and hypoproteinaemia were significant factors for oedema formation. Postoperative oedema was associated with a significant increase in postoperative morbidity.

Recent evidence suggests that stem cells are important for cancer metastasis and that the epithelial-to-mesenchymal transition also involves a transition toward stemness. Current thinking suggests that lgr5, a 7-transmembrane spanning G-protein, also marks a certain population of stem cells capable of regenerating an intestinal crypt and that the specialised immune secretory paneth cell, is important for maintenance of the stem cell niche. We implicate fascin in regulating the balance of lgr5 stem cells during acute intestinal inflammation and in regenerating intestinal tissue. Fascin is an actin bundling protein that drives the assembly of filopodia through the cross-linking of actin filaments into straight bundles. Conserved from amoebas to man, fascin was originally purified from extracts of sea urchin oocytes and coelomocytes and later found in Drosophila as the singed gene product. It is involved in the invasion and metastasis of multiple epithelial cancer types through stabilisation of actin in invadopodia, finger like protrusions used by cancer cells to invade into and degrade the extra-cellular matrix. Fascin, whilst normally low or absent from epithelia, localises to the leading edges of migratory cells and is over-expressed in many cancers of the same epithelial origin including lung, colorectal, pancreatic and liver. Fascin has also recently been shown to increase during inflammatory bowel disease (IBD) conditions such as diverticulitis, Crohn’s disease and ulcerative colitis. In this thesis I have investigated the role of fascin in murine models of IBD and have demonstrated that fascin is required for the haematopoietic production of leucocytes, in response to inflammation and that the loss of fascin, in the presence of high Wnt levels, results in enhanced proliferation of intestinal epithelial cells. One of the serious consequences of IBD is the increased lifetime risk of the patient developing an intestinal malignancy secondary to the disease. The exact mechanism underlying the increase in malignancies has not yet been fully established, however it is postulated that chronic inflammation and the effect this has on the major molecular pathways involved in carcinogenesis underlies the transformation from benign to malignant disease. Highest fascin expression has been shown in the dysplastic, pre-malignant cells in human IBD tissues indicating an important role of fascin in the transformation of benign to malignant cells. In this thesis, I have demonstrated that loss of fascin impairs tumour initiation in inflammatory driven and spontaneous intestinal tumourigenesis models, which is likely, in part, to be as a consequence of reduced leucocytes, in particular neutrophils, which may be CXCL2 mediated.

Introduction: Malignant Melanoma is the most lethal of the skin cancers and the UK incidence is rising faster than that of any other cancer. Breslow thickness remains the best predictor of metastasis and Sentinel Lymph Node Biopsy is the only method of detecting nodal spread in clinically node negative patients. Surgery is the only effective therapy. Angiogenesis – the growth of new vessels from pre-existing vasculature - is an absolute requirement for tumour survival and progression beyond a few hundred microns in diameter. Anti- angiogenic isoforms of VEGF have been demonstrated previously to be protective with regard to metastasis. The aims of this thesis were to determine whether VEGF expression within the tumour may allow prediction of the nodal status. Furthermore another aim was to determine whether via the “Seed and Soil” theory, by examination of angiogenic and lymphangiogenic profiles of the tumour and node can we determine that the tumour may control the microenvironment around the Sentinel Node? Finally, as a cohort of false negative patients emerged with a higher mortality rate than their true negative and true positive patient cohort counterparts, could any further patterns be established by performing the same experiments on these patients? Methods: Archived human tumour and corresponding Sentinel Node samples were used and immunohistochemistry was used to investigate the role of pro and anti angiogenic isoforms of VEGF, VEGF-C, LYVE-1 and CD31 within these patients. Results: VEGF-C expression was significantly increased in the intranodal component of positive Sentinel Lymph Nodes (p < 0.01 Bonferroni). This increased expression appeared to be independent of tumoural influences and no strong evidence for the “Seed and Soil” theory was proved. A significantly higher number of lymphatic vessel counts were identified within node negative patients (p < 0.05 ANOVA). No further significant findings were defined on examination of the false negative cohort of patients. Conclusions: This study has shown that positive Sentinel Lymph Nodes exhibit high levels of intranodal VEGF-C. This expression does not appear to be related to tumoural influences. It would therefore appear that VEGF-C expression within Sentinel Nodes warrants further investigation and may aid diagnosis of spread or represent a target to slow or even prevent the onset of metastasis.