Background & objective: Neoadjuvant chemotherapy has gained increasing attention as a treatment for gastric cancer since Wilke first reported its application to the treatment of gastric cancer in 1989. However, its value in treating gastric cancer remains controversial. This research was to assess the efficacy of neoadjuvant chemotherapy on gastric cancer through a Meta analysis of the randomized controlled trials published worldwide in both English and Chinese.
Methods: Cochrane systematic evaluation was used to search through Cochrane libraries of clinical comparative trials, PubMed, Embase, Chinese Biomedical Literature Database (CBM), Chinese Scientific Journal Full-text Database (CSJD) and Chinese Journal Full-text Database (CJFD), aided with manual retrieval and other retrievals. The quality of the assessment was independently evaluated and cross-checked by two evaluators, and the results of homogeneous studies were analyzed with RevMan4.2.10 software.
Results: Five randomized controlled trials involved a total of 838 patients were studied. Of the 5 trials, 2 were performed in Japan, 1 in the Netherlands, 1 in the United Kingdom, 1 in China. Of the 838 patients, 373 were treated with neoadjuvant chemotherapy and 465 were treated with surgery alone. Among the above 5 studies, one used blind method and one described random allocation concealment method. No statistical differences were found in the resection rate, cure rate, 1-and 5-year survival rates between neoadjuvant chemotherapy group and surgery group [odds ratio (OR)=1.09, 95% confidence interval (CI)=0.67-1.77 for resection rate; OR=1.25, 95% CI=0.85-1.84 for cure rate; OR=1.61, 95% CI=0.90-2.90 for 1-year survival rate; OR=1.13, 95% CI=0.83-1.53 for 5-year survival rate].
Conclusions: The efficacy of neoadjuvant chemotherapy on gastric cancer is not better than that of surgery alone. Therefore, neoadjuvant chemotherapy should not be recommended as a regular treatment for gastric cancer before obtaining evidences of its certain efficacy on gastric cancer.