Background/aims: Published data on the dose of imatinib to treat gastrointestinal stromal tumors seemed inconclusive. To derive a more precise estimation of dose of imatinib to treat gastrointestinal stromal tumors, a meta-analysis was performed.
Material and methods: Studies have been identified by searching PubMed and Embase. Inclusion criteria were patients with incurable or advanced gastrointestinal stromal tumors, received chemotherapy of imatinib of 400 mg once a day (low-dose group) or 300 mg/400 mg twice a day (high-dose group) for controlled trial, and efficacy evaluation were cumulative response based on SWOG criteria, including complete response and partial response. The safety evaluation of imatinib included toxic effect and dose reduction. The odds ratio for cumulative response, toxic effect and dose reduction in patients with low-dose of imatinib over those with high-dose of imatinib with 95% confidence interval was calculated for each study as an estimation of the better effect of imatinib.
Results: A total of 3 studies including 1,787 patients were involved in this meta-analysis. The overall odds ratio for cumulative response, toxic effect and dose reduction were 0.91 (95% confidence interval, 0.75-1.09; p=0.31, Pheterogeneity=0.80), 0.55 (95% confidence interval, 0.42-0.72; p<0.0001, Pheterogeneity=0.01), respectively.
Conclusion: The meta-analysis indicated that cumulative response for patients with low-dose imatinib showed no significant difference compared with that with high-dose imatinib. However, high-dose imatinib brought patients more toxic effect and more dosage adjustment for down regulation was made for these patients with severe toxic effect. With long follow-up of the patients, result of neither progression-free survival nor overall survival reached statistical significance between low-dose arm and high-dose arm.