To investigate the effect of Gefitinib on insulin sensitivity in rats of type-2 diabetes and its mechanism. Normal Wistar rats were fed with high sucrose-fat diet for 1 month and then injected intraperitoneally with streptozotocin (STZ). The rats satisfying the definity of type-2 diabetes were divided into 3 groups and given control solution (0.9%NaCl), low-dose Gefitinib and high-dose Gefitinib for 8 days, respectively. Body weights, FBG and FINS were recorded before, just after and 7 days after administration, respectively. After the second round of administration executed the rats and got the livers from rats. Compared the expression of insulin receptor substrate-1 (IRS-1) and phosphatidylinositol-3-kinase (PI3K) genes in liver among the 3 groups with semi-quantitative Reverse Transcription-Polymerase Chain Reaction (PCR). Normal Wistar rats fed with high sucrose-fat diet for 1 month and then injected intraperitoneally with STZ could establish type-2 diabetes model. After administration, insulin sensitivity index (ISI) of Gefitinib groups were significantly increased compared with control group (2.96 +/- 1.38 vs 0.92 +/- 0.20, P < 0.05). There Were no significant difference between the two dose groups (2.95 +/- 1.51 vs 2.96 +/- 1.38, P > 0.05). 7 days after administration, the improved ISI did not change (2.03 +/- 0.72 vs 2.99 +/- 0.63, P > 0.05). Body weights of the 3 groups had no significant difference. The expression of IRS-1 gene significantly decreased in Gefitinib groups compared with control group [(3.3 +/- 1.5)% vs (6.3 +/- 2.4)% P < 0.05] and with no obvious difference between the two dose groups [(3.3 +/- 1.5)% vs (3.2 +/- 1.8)%, P > 0.05], while the expression of PI3K gene significantly increased in Gefitinib groups compared with control group [(1.27 +/- 0.73)% vs (0.41 +/- 0.24)%, P < 0.05] also with no obvious difference between the two dose groups [(1.27 +/- 0.73)% vs (1.43 +/- 0.71)%, P > 0.05]. Gefitinib can improve insulin sensitivity in rats of type-2 diabetes. 2 t(1/2) after administration, the effect still existed. The effect probably due to the increase of PI3K/Akt pathway of insulin signaling, and had no relationship with body weight in our study dose range.