We found no evidence of effects of imatinib in people with resectable or benign GIST, many of whom have an excellent prognosis following resection. In people with unresectable or metastatic GIST, prognosis is reported as poor. We found no studies directly comparing imatinib versus other or no treatment. Conclusions about relative effectiveness and safety depend on historical comparisons with untreated populations and, therefore, have limited reliability. The identified studies suggest that imatinib is a promising therapy in people with advanced GIST, particularly if it is the case that prognosis is uniformly poor with other management. However, it is not clear how people should be selected for treatment, what dose and duration should be used and, most importantly, whether clinical benefits will outweigh risks. Further trials are currently underway, which may help to address these issues.7,8 Until further results become available, imatinib continues to be an experimental treatment for GIST, which should be given in the context of research protocols.