Novel approaches for improving the management of gastric and duodenal neuroendocrine tumours.

Introduction: Gastroenteropancreatic neuroendocrine tumours (GEP-NENs), previously known as carcinoid tumours, arise within the gastrointestinal tract and pancreas. Gastric and duodenal NENs were previously thought to be rare, but recent evidence has suggested that they are increasing in both incidence and prevalence. Currently, most gastric and duodenal NENs are diagnosed as small, low grade, localised well-differentiated tumours. These usually have an excellent prognosis. Optimal management however remains controversial. Organoid models are complex 3D culture systems which arise from stem cells. They represent an emerging platform for developmental and regenerative experiments as well as disease modelling, precision medicine and pharmacological studies. Due to the rarity and number of subtypes of gastric NENs, large clinical studies and trials are not often feasible. Organoid models could provide a new culture system to bridge this gap and investigate gastric neuroendocrine differentiation and tumour development. Aims: The aims of this PhD were to review the current management of localised gastric and duodenal neuroendocrine tumours and to investigate the role of endoscopic surveillance or less aggressive tumour resection techniques in their management pathways. Furthermore, we aimed to develop a gastric organoid model system to study the biology of gastric ECL cells and the pathogenesis of neuroendocrine tumours in the stomach. Methods: A systematic literature review of the current management of low grade localised gastric and duodenal neuroendocrine tumours was initially performed. The outcomes of patients who were enrolled onto local endoscopic surveillance programmes for type I gastric NENs and non-functioning, non-ampullary duodenal NENs were interrogated. A retrospective multicentre study of localised grade 1 or 2 type III gastric NENs treated by endoscopic or local resection was also conducted. Finally, a gastric organoid model was developed and directed towards a neuroendocrine phenotype by manipulation of the composition of the organoid culture medium as well as the addition of gastrin and the small molecule ISX-9. Results: The quality of the data that currently informs management decisions in patients who have localised low grade gastric and duodenal NENs is very poor. Most patients who have type I gastric NENs have an excellent prognosis and are unlikely to develop distant metastases or die because of their disease. Initial endoscopic resection appears to be unnecessary in patients who have polyps ≤10mm. Periodic lifelong endoscopic surveillance should however still be undertaken in this group, as patients are still at risk of developing further NENs or gastric adenocarcinoma. Definitive treatment by endoscopic resection or limited surgical resection is feasible and safe in type III gastric NENs which demonstrate favourable low grade and well differentiated histology and small size (<10mm). Patients with ≤10mm sporadic, grade 1, non-functional, non-ampullary duodenal NENs, can also be safely managed by a ‘watch and wait’ strategy. Gastric organoids lose specialised epithelial cells whilst in culture. However, they can be directed towards ECL cell differentiation by manipulating the culture medium and by the addition of the small molecule ISX-9. Conclusions: Less aggressive treatment options should be considered for small, low grade gastric and duodenal NENs. Gastric organoid cultures offer a new approach to study neuroendocrine differentiation and tumour development in the stomach.