INTRODUCTION: The incidence of gastrointestinal neuroendocrine tumors (GI-NETs) is increasing in recent times reflecting the widespread use of endoscopy, and an increased awareness among clinicians. Duodenal carcinoids are rare and accounting for 1%-3% of primary duodenal tumors. Duodenal carcinoids < 10 mm in size and limited to the submucosal layer are amenable for endoscopic resection. Despite the increase in incidence, the natural history of duodenal carcinoid tumors has not yet been well defined. The aim of our study was to evaluate the recurrence rate following endoscopic resection of well differentiated duodenal carcinoid tumors from 2006-2016 at our institution. METHODS: We performed a retrospective analysis on 39 patients who underwent endoscopic resection of duodenal carcinoids from 2006-2016 at the Moffitt Cancer Institute. The indications for endoscopic resection were lesions ≤ 10 mm in diameter, confined to the submucosal layer, and without lymph node or distant metastasis. Endoscopic resection was accomplished using endoscopic mucosal resection (EMR) and biopsy forceps. The patients were followed for a mean of 3 years. RESULTS: Among the 39 patients meeting criteria for endoscopic resection, 28 patients underwent surveillance endoscopies post EMR and were included in final analysis . EMR was performed in 12 tumors while biopsy forceps resection in 11 patients. Pathological complete resection(R0) was achieved in 20 patients (71%), without evidence of recurrence in 9 (45%) patients followed for at least 3 years and in the 11(55%) patients followed for lesser than 3 years. On the contrary, in the 8 patients (28%) had recurrence in the 3 year follow up period of which that was recurrence,6 (75%) had either positive lateral or deep margins on initial resection (Table 1). There were no post procedural complications in any of the patients in the follow up period. CONCLUSION: Endoscopic resection appears to be a safe and effective treatment for duodenal carcinoid tumors measuring ≤ 10 mm in diameter and confined to the submucosal layer. This retrospective study showed that the main predictor of recurrence is the presence of positive tumor margins at initial resection. Patients with negative and margins had no recurrences in the 3 year follow up period, thus alleviating the need for long term surveillance in this patient population. Larger multicenter studies are required to determine the optimal surveillance stragey in patients undergoing endoscopic resection of duodenal carcinoid tumors.