Zhaolun Cai1, Yuan Yin2, Chaoyong Shen3, Jian Wang4, Xiaonan Yin5, Zhixin Chen6, Ye Zhou7, Bo Zhang8. 1. Department of Gastrointestinal Surgery, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan, China. Electronic address: caizhaolun@foxmail.com. 2. Department of Gastrointestinal Surgery, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan, China. Electronic address: Yinyuan10@gmail.com. 3. Department of Gastrointestinal Surgery, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan, China. Electronic address: scyshenchaoyong@163.com. 4. Department of Gastrointestinal Surgery, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan, China. Electronic address: wj66220@126.com. 5. Department of Gastrointestinal Surgery, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan, China. Electronic address: yxnyinxiaonan@163.com. 6. Department of Gastrointestinal Surgery, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan, China. Electronic address: chenzhixin@medmail.com.cn. 7. Department of Gastric Surgery, Fudan University Shanghai Cancer Center, Shanghai, 200032, China. Electronic address: zhouye79@yahoo.com. 8. Department of Gastrointestinal Surgery, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan, China. Electronic address: hxwcwk@126.com.
Abstract
BACKGROUND: Different preoperative, postoperative or perioperative treatment strategies, including chemotherapy or chemoradiotherapy, are available for patients with gastric cancer, but conventional meta-analyses that assess two alternative treatments are unable to compare differences in overall survival. Thus, we performed a network meta-analysis to identify the best treatment strategy. METHODS: We systematically searched and assessed studies for eligibility and extracted data. We then pooled the data and conducted a Bayesian network meta-analysis to combine direct comparisons with indirect evidence. The node-splitting method was used to assess the inconsistency. Rank probabilities were assessed by the probability of treatment rankings. RESULTS: Thirty-three eligible randomized controlled trials were included in the network meta-analysis. Four treatments that had significantly improved prognoses when compared with surgery only were postoperative chemotherapy [HR = 0.80 with 95% CrI: (0.73, 0.88)], postoperative chemoradiotherapy [HR = 0.73 with 95% CrI: (0.61, 0.87)], preoperative chemoradiotherapy [HR = 0.77 with 95% CrI: (0.62, 0.98)] and perioperative chemotherapy [HR = 0.69 with 95% CrI: (0.55, 0.84)]. Preoperative chemotherapy, however, did not significantly improve survival when compared with surgery alone [HR = 0.94 with 95% CrI: (0.71, 1.2)]. There was no statistically significant difference between postoperative chemotherapy, postoperative chemoradiotherapy, preoperative chemoradiotherapy and perioperative chemotherapy in terms of overall survival. Chemoradiotherapy after D2 lymphadenectomy did not significantly improve OS when compared with postoperative chemotherapy [HR = 0.95 with 95% CrI: (0.73, 1.3)]. CONCLUSION: Among patients with operable gastric cancer, perioperative chemotherapy had the highest probability of being the best treatment. Further clinical resources may be required to assess the efficacy and safety of perioperative chemotherapy for patients with gastric cancer.
BACKGROUND: Different preoperative, postoperative or perioperative treatment strategies, including chemotherapy or chemoradiotherapy, are available for patients with gastric cancer, but conventional meta-analyses that assess two alternative treatments are unable to compare differences in overall survival. Thus, we performed a network meta-analysis to identify the best treatment strategy. METHODS: We systematically searched and assessed studies for eligibility and extracted data. We then pooled the data and conducted a Bayesian network meta-analysis to combine direct comparisons with indirect evidence. The node-splitting method was used to assess the inconsistency. Rank probabilities were assessed by the probability of treatment rankings. RESULTS: Thirty-three eligible randomized controlled trials were included in the network meta-analysis. Four treatments that had significantly improved prognoses when compared with surgery only were postoperative chemotherapy [HR = 0.80 with 95% CrI: (0.73, 0.88)], postoperative chemoradiotherapy [HR = 0.73 with 95% CrI: (0.61, 0.87)], preoperative chemoradiotherapy [HR = 0.77 with 95% CrI: (0.62, 0.98)] and perioperative chemotherapy [HR = 0.69 with 95% CrI: (0.55, 0.84)]. Preoperative chemotherapy, however, did not significantly improve survival when compared with surgery alone [HR = 0.94 with 95% CrI: (0.71, 1.2)]. There was no statistically significant difference between postoperative chemotherapy, postoperative chemoradiotherapy, preoperative chemoradiotherapy and perioperative chemotherapy in terms of overall survival. Chemoradiotherapy after D2 lymphadenectomy did not significantly improve OS when compared with postoperative chemotherapy [HR = 0.95 with 95% CrI: (0.73, 1.3)]. CONCLUSION: Among patients with operable gastric cancer, perioperative chemotherapy had the highest probability of being the best treatment. Further clinical resources may be required to assess the efficacy and safety of perioperative chemotherapy for patients with gastric cancer.