OBJECTIVE: To evaluate the clinical characteristics and survival factors of patients with duodenal gastrointestinal stromal tumors (GIST). METHODS: The clinical data of 41 patients with duodenal GIST were analyzed retrospectively at Cancer Hospital and Institute, Chinese Academy of Medical Sciences from June 1996 to August 2011. Kaplan-Meier method was used to calculate the recurrence-free survival rate and the Cox proportional hazard regression model employed for the recurrence-free survival analysis. RESULTS: The lesions of duodenal GIST were predominantly located in the descending (n = 26, 63.4%) and transverse portions (n = 10, 24.4%). Most duodenal GIST presented commonly with upper gastrointestinal bleeding (n = 18, 43.9%) and 12 cases (29.3%) were incidentally detected by physical examinations. Eight patients underwent pancreatoduodenectomy and 27 limited resection. The tumor size varied from 0.6 cm to 30.0 cm (mean: 8.4 cm). The recurrence-free survival rates analyzed by Kaplan-Meier method at 1, 2 and 5-year were 94.1%, 77.5% and 65.0% respectively. The results of Cox proportional hazards regression model indicated that the patients with >10/50 HP mitotic count showed a worse recurrence-free survival than those with ≤ 10/50 HP (HR = 3.7, 95%CI 1.0 - 13.7, P = 0.049). After adjusting other confounding factors, mitotic activity was one significant prognostic factor of recurrence (P = 0.024). There was no significant association between the risk of recurrence and other prognostic factors, including diagnostic age, tumor size, type of operation and the risk of aggressive behaviors (all P > 0.05). CONCLUSIONS: Mitotic activity is one prognostic factor of duodenal GIST. And R(0) resection should be regarded as an optional treatment for duodenal GIST.
OBJECTIVE: To evaluate the clinical characteristics and survival factors of patients with duodenal gastrointestinal stromal tumors (GIST). METHODS: The clinical data of 41 patients with duodenal GIST were analyzed retrospectively at Cancer Hospital and Institute, Chinese Academy of Medical Sciences from June 1996 to August 2011. Kaplan-Meier method was used to calculate the recurrence-free survival rate and the Cox proportional hazard regression model employed for the recurrence-free survival analysis. RESULTS: The lesions of duodenal GIST were predominantly located in the descending (n = 26, 63.4%) and transverse portions (n = 10, 24.4%). Most duodenal GIST presented commonly with upper gastrointestinal bleeding (n = 18, 43.9%) and 12 cases (29.3%) were incidentally detected by physical examinations. Eight patients underwent pancreatoduodenectomy and 27 limited resection. The tumor size varied from 0.6 cm to 30.0 cm (mean: 8.4 cm). The recurrence-free survival rates analyzed by Kaplan-Meier method at 1, 2 and 5-year were 94.1%, 77.5% and 65.0% respectively. The results of Cox proportional hazards regression model indicated that the patients with >10/50 HP mitotic count showed a worse recurrence-free survival than those with ≤ 10/50 HP (HR = 3.7, 95%CI 1.0 - 13.7, P = 0.049). After adjusting other confounding factors, mitotic activity was one significant prognostic factor of recurrence (P = 0.024). There was no significant association between the risk of recurrence and other prognostic factors, including diagnostic age, tumor size, type of operation and the risk of aggressive behaviors (all P > 0.05). CONCLUSIONS: Mitotic activity is one prognostic factor of duodenal GIST. And R(0) resection should be regarded as an optional treatment for duodenal GIST.