With the rapid development of endoscopic technology, endoscopic therapy (ET) has gradually become a new treatment choice for gastrointestinal stromal tumors (GISTs). However, due to the low incidence of duodenal GIST and the difficulty of ET, there is a lack of data to compare the long-term results of ET and surgical resection. Duodenal GIST patients from 2004 to 2015 were selected from the surveillance, epidemiology, and end result (SEER) database. We used the Kaplan-Meier method and log-rank test to describe the 5- and 10-year survival differences between the ET and the surgery groups. The multivariate Cox proportional hazard model was used for analyzing the risk factors influencing the prognosis of patients. We used a 1:1 propensity score-matched (PSM) to reduce confounding factors, and then we compared survival differences between the two groups again. A total of 294 patients with duodenal GIST were enrolled, including 41 (13.9%) patients with ET and 253 (86.1%) patients with surgical resection. Before PSM, the long-term survival of patients with duodenal GIST after ET and surgical resection was similar [5-year overall survival (OS) (79.7 vs. 79.3%, p = 0.876), 10-year OS (66.5 vs. 68.1%, p = 0.876)]. After adjusting the relevant variables using multivariate Cox analysis, we found that the ET and surgery groups were comparable in OS and cancer-specific survival (CSS). After PSM, there was also no significant difference between ET and surgical resection for long-term OS and CSS. Our study found no significant difference in long-term survival between ET and surgical resection in patients with duodenal GIST. However, to obtain high-quality evidence, more extensive sample size studies are needed in the future to evaluate the long-term effects of ET on patients. © 2022. The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.