Background: Imatinib was approved in 2002 for unresectable and metastatic gastrointestinal stromal tumors. Our objective was to determine if the introduction of imatinib coincided with improved survival from metastatic gastrointestinal stromal tumor in the U.S. population and in specific socioeconomic groups. Methods: Query of the Surveillance, Epidemiology, and End Results registry identified 552 patients with metastatic gastrointestinal stromal tumor between 1995 and 2004. Year of diagnosis was categorized into two periods, 1995 to 2000 and 2001 to 2004, to account for the effect of imatinib. Kaplan-Meier and multivariate Cox regression analyses were used to examine differences in survival between periods and among socioeconomic groups. Results: Median survival increased from 12 to 33 months from 1995 to 2000 to 2001 to 2004 (P < 0.001); survival at 47 months increased from 21% to 41%, respectively (P < 0.001). Median survival times for White, Black, Hispanic, and Asian or Pacific Islander, and for low-, middle-, and high-income groups increased significantly in the era of imatinib (all P < 0.05). On multivariate analysis, Black race [hazard ratio, 1.96; 95% confidence interval (95% CI), 1.15-3.32; P = 0.013], Hispanic race (hazard ratio, 2.11; 95% CI, 1.14-3.88; P = 0.017), and low income (hazard ratio, 1.81; 95% CI, 1.13-2.89; P = 0.014) were associated with the poorest survival during the 1995 to 2000 period. During 2001 to 2004, these disparities in survival were no longer statistically apparent. Conclusions: Survival from metastatic gastrointestinal stromal tumor has improved significantly in the era of imatinib. This improvement has been uniform across all socioeconomic groups, with concomitant elimination of socioeconomic survival disparities potentially due to an assistance program intended to provide universal access to imatinib therapy. Copyright © 2008 American Association for Cancer Research.