Abstract: Objective To evaluate the effect of topical KH902 on experimental corneal neovascularization (CNV) in rats． Methods Suture models of CNV were established in 60 adult healthy rats. The rats were randomly divided into 3 groups: Group A (20 rats) were treated with KH902 (KH902 30 mg/mL was injected subconjunctivally once every other day); Group B (20 rats) were treated with dexamethasone (0.1% dexamethasone was injected subconjunctivally once every other day); Group C (20 rats) served as blank control. CNV growth were observed by slitlamp microscopy 3, 7 and 14 d after suture. Medicines were administered 14 d after operations. CNV changes were observed and recorded 1, 7, 14, and 21 d after administration of medicines. VEGF expression in cornea was measured by immunohistochemistry. Results No significant differences were found in average areas of CNV between groups at 1 and 7 d after administration of medicines. At 14 and 21 d after administration of medicines, significant differences in average areas of CNV were found between Group A and Group C, and between Group B and Group C (P<0.01). Maximum expression of VEGF in corneal stroma was observed 14 d after corneal suture. The expression of VEGF decreased with medications, which was associated with neovascularization. Conclusion Topical use of KH902 inhibits corneal suture-induced CNV in rats without significant adverse corneal effect on eyes.