Surgical outcomes of gastrointestinal sarcoma including gastrointestinal stromal tumors: a population-based examination

Abstract

Introduction: The surgical approaches and outcomes for gastrointestinal sarcoma are determined largely from single institutional series.

Objective: We sought to determine patient outcomes after different surgical approaches for gastrointestinal sarcomas, including gastrointestinal stromal tumors (GIST), utilizing a large prospective cancer registry.

Material and methods: The Surveillance, Epidemiology, and End Results (SEER) database was queried from 1991 to 2002.

Results: Overall, 1873 gastrointestinal mesenchymal tumors were identified in the SEER data set, with 82% GIST and 18% smooth muscle neoplasms. Surgery was performed in 83% of the cohort. Median survival was 68 months for complete resection (CR), 51 months for partial resection (PR), and 10 months for no resection (NR) (P < 0.001 each category). Outcomes within the CR group were equivalent for wedge or total organ removal. Median survival rates for localized, regionally advanced, and metastatic disease were 97, 35, and 18 months, respectively, after CR, and in all cases significantly improved relative to patients not undergoing resection. Median survival rates in patients treated after 2000 have substantially improved in this cohort, possibly reflecting the impact of imatinib on overall population-based survival. Multivariate analysis identified organ, histologic grade, surgical resection, and date of surgery (pre-2000 or post-2000) as independent predictors of survival.

Conclusions: The outcomes after surgical therapy for gastrointestinal sarcomas, including GIST, support the operative goal of a complete resection. Improved outcomes after 2000 indicate the potential benefit of newer therapies, including imatinib.