Efficacy and safety comparison of chemotherapies for advanced gastric cancer: A network meta-analysis
Metrics: PDF 1500 views | HTML 1860 views | ?
Jinping Sun1,*, Zheng Ren1,*, Xinfang Sun1, Hongtao Hou1, Ke Li1, Quanxing Ge1
1Department of Digestive Internal Medicine, Huaihe Hospital of Henan University, Kaifeng, 475000, Henan, China
*These authors contributed equally to this work
Quanxing Ge, email: [email protected]
Keywords: advanced gastric cancer, chemotherapies, network meta-analysis, efficacy, safety
Received: February 07, 2017 Accepted: April 03, 2017 Published: May 11, 2017
Objective: Chemotherapy is one of the commonly used therapies for advanced gastric cancer. In this study, we performed a network meta-analysis on the efficacy and safety of common treatments to give evidences of their relative benefits.
Results: 32 trials with 8550 patients and 20 regimens were included in this study. According to the results of primary outcomes, 5-FU plus OXA, 5-FU plus DOC, CAP plus CIS, CAP plus OXA, S-1 plus OXA and S-1 plus PAC performed well in improving OS and ORR. As for the adverse events, S-1 had a safer effect than other treatments, conversely, 5-FU plus CIS ranked the last. However, there was no regimen with outstanding performances in both efficacy and safety.
Materials and Methods: Studies were searched from database and screened with criteria. The Bayesian framework based network meta-analysis was performed with software R and STATA. Overall survival (OS) and overall response rate (ORR) were considered as primary outcomes while adverse events as secondary outcomes. The outcomes were represented by hazard ratios or odd ratios with 95% corresponding credible intervals, respectively.
Conclusions: The network meta-analysis suggested that 5-FU plus OXA and 5-FU plus DOC were recommended when efficacy was stressed. S-1 was safest but poorly effective. A regimen, as an excellent combination of efficacy and safety, is still waiting to be discovered.
All site content, except where otherwise noted, is licensed under a Creative Commons Attribution 3.0 License.