Objective To evaluate the effectiveness and safety of lentinan on immune function in patients with advanced gastric cancer. Methods We searched MEDLINE (1969-2006), EMBASE (1984-2006), OVID (1969-2006), CENTRAL (Cochrane Central Register of Controlled Trials in The Cochrane Library) (Issue 4, 2006), the Chinese Biomedicine Database (1978-2006) and CNKI (1978-2006). We also handsearched relevant journals. Pharmaceutical companies were contacted to identify additional randomized controlled trials. We assessed the identified studies in order to include high quality studies. Results Ten studies (containing 786 patients) met the inclusion criteria. Six trials shown that lentinan+FAM had significant efficacy upon patients with advanced gastric cancer compared with FAM in overall response [Plt;0.01, RR1.70, 95%CI (1.39,2.09)]. In three trials, a significant effect of lentinan+FAM group compared with FAM group in quantity of CD3+ T, T4/T8, NK was found, but lower than FAM group in side- effect of digestive system [RR0.71, 95%CI (0.55,0.91)]. The other trail identified there were fewer side effects in lentinan+FAM group compared with FAM group, though did not discribe the overall response. In case the significant heterogeneity, meta-analysis could not be used for the other three trails included, since the components of chemotherapeutic agents (ATP+Co-A+Vc; DDP+ Epirubicin+5FU; 5FU+CF+VP16) were not the same. In the three trials, overall response was statistically significant better in the lentinan group than in the control group, and lentinan group could significantly increase the quantity of CD3+ T, T4/T8, NK compared with control group. Conclusions The present meta-analysis suggested that addition of lentinan to standard chemotherapy provided a significant advantage over chemotherapy alone in terms of efficacy for patients with advanced gastric cancer. However, most of trials included in the review were of low quality, therefore, it is of necessity to conduct multi-center randomized-controlled trials of high quality.