Clinical outcomes and mechanisms of mesenteric fibrosis in small bowel neuroendocrine tumours

Small intestinal neuroendocrine tumours (SI NETs) are relatively rare, indolent tumours arising from the enterochromaffin cells of the small bowel. They can be associated with the development of fibrosis, most commonly in the heart (termed ‘carcinoid heart disease’) or the mesentery around a metastatic deposit (also known as mesenteric desmoplasia). A concise summary of our current understanding of the pathophysiology of neuroendocrine fibrogenesis is provided in Chapter 1. The development of mesenteric fibrosis can be asymptomatic, but in some patients, it can lead to significant complications, such as bowel obstruction or mesenteric ischaemia. However, the epidemiology and pathophysiology of mesenteric fibrosis have not been sufficiently explored. In addition, there is a lack of clinically useful biomarkers for the detection of image-negative mesenteric desmoplasia. The first aim of our study was to explore different clinical aspects of mesenteric desmoplasia and more specifically the epidemiological characteristics and clinical assessment of mesenteric fibrosis. An additional aim was to delineate further the pathogenesis of this process and evaluate a non-invasive, blood-based biomarker for the early detection of fibrosis. To investigate the clinical and epidemiological aspects of mesenteric desmoplasia, two large-scale retrospective studies were performed which are described in Chapter 2. The first study was a survival analysis of approximately 150 fibrotic SI NETs and the second was a survival analysis of almost 400 metastatic SI NETs investigating the effect of mesenteric fibrosis on prognosis in addition to other epidemiological data. We demonstrated that in a large cohort of almost 400 patients with metastatic small bowel neuroendocrine tumours the prevalence of mesenteric fibrosis was about 35% and that 50% of patients with mesenteric metastases had radiological evidence of desmoplasia. The presence of mesenteric fibrosis was associated with a worse overall survival compared to patients without mesenteric lymphadenopathy by both univariate and multivariate analysis. However, in a retrospective analysis of almost 150 patients with fibrotic small bowel neuroendocrine tumours we did not find a significant difference in overall survival or complications (bowel obstruction, mesenteric ischaemia) among patients with mild, moderate and severe mesenteric desmoplasia (patients were grouped using a radiological scoring system to grade the severity of desmoplasia). We determined that in midgut neuroendocrine tumours with radiological evidence of mesenteric fibrosis advanced age (>65) and elevated urinary 5-hydroxyindoleacetic acid (a product of serotonin metabolism measured in the urine) were independent predictors of survival. In addition, to explore the pathophysiology of mesenteric fibrogenesis, we performed a set of co-culture experiments using the small intestinal NET cell lines KRJ-I and P-STS and the stromal cell line HEK293 and these experiments are described in Chapter 3. This study evaluated the crosstalk between cancer and stromal cells in terms of effects on cell proliferation, metabolism, gene expression (RT2 PCR Profilers and RNA sequencing) and protein/cytokine secretion. A reduction in cell metabolic activity was observed in both KRJ-I and P-STS cells treated with HEK293 conditioned media, with no significant changes in cell proliferation. Furthermore, no significant changes were observed in HEK293 cell proliferation or metabolic activity upon culturing with conditioned media of cancer cells. The RT2 PCR arrays revealed that several genes with key regulatory functions in cancer biology and fibrogenesis were significantly up- or down-regulated in KRJ-I, P-STS and HEK293 cells (fold-change≥2, p<0.05). In addition, RNA sequencing and Ingenuity Pathway Analysis identified multiple pathways that were significantly activated or inhibited (p<0.05, z score ≥ |2|). These signalling pathways have a wide range of biological functions, such as involvement in protein synthesis (e.g. EIF2 pathway) or regulation of cell metabolism and energy production. Several changes in chemokine and growth factor secretion were also observed. A specific pathway that was activated in KRJ-I cells as a result of the paracrine effect - the integrin pathway - was further investigated in human tissue using qPCR, immunohistochemistry and Western blotting. This set of experiments is described in Chapter 4. A total of 34 patients were recruited into this prospective study and the gene/protein expression was evaluated in patients with graded severity of mesenteric fibrosis. The fibrosis grading was established using a complex and thorough assessment of the mesenteric desmoplasia which included a triangulation of different methodologies incorporating surgical, radiological and histological criteria. Using a variety of techniques, we provided evidence that the integrin pathway is activated in the fibrotic mesenteric mass of SI NETs. The detailed analysis of mesenteric desmoplasia (incorporating surgical, radiological and histological parameters), that was used to provide the grading of fibrosis severity, also revealed several cases of image-negative mesenteric fibrosis. This is a new concept, as traditionally the diagnosis of mesenteric fibrosis has been based on imaging studies. Finally, in Chapter 5, we evaluated a novel biomarker – the Fibrosome - that included 5 circulating transcripts from the NETest, which collectively exhibited high performance metrics for the detection of mesenteric fibrosis, even in the case of image-negative mesenteric desmoplasia. In conclusion, this study has provided useful epidemiological data about mesenteric fibrosis and more specifically regarding its prevalence, effect on overall survival and clinical outcomes, and has also provided insight into the assessment of the presence and severity of mesenteric fibrosis, demonstrating for the first time that a radiological evaluation alone is not sufficient. In addition, this is the first study to provide a comprehensive understanding of the bidirectional communication of cancer and stromal cells and we have validated the activation of the integrin pathway in the fibrotic microenvironment of SI NETs (human tissue). Finally, a novel circulating biomarker was developed with promising preliminary results, that warrant further validation in additional independent patient cohorts.