ObjectiveTo investigate the value of conbercept in experimental corneal nevascularization (CNV) and lymphangiogenesis in rabbit model.

MethodsForty-four adult New Zealand rabbits were randomly divided into a conbercept injected group (9 rabbits), a ranibizumab injected group (9 rabbits), a normal saline control group (9 rabbits), a model control group (9 rabbits) and a normal control group (8 rabbits) by random number table method, take the left eyes as the experimental eye.Whatman filter papers (8 mm in diameter) were soaked in 1 mol/L NaOH solution and then applied on the middle surface of the cornea for 30 seconds.On day 1st after alkali burning, the eyes in the conbercept injected group were treated with 0.1 ml/1 mg of conbercept, the eyes in the ranibizumab injected group were treated with 0.1 ml/1 mg of ranibizumab, the eyes in the normal saline control group were treated with 0.1 ml 0.9% NaCl, the eyes in the model control group just received alkali burning, and the eyes in the normal control group received neither alkali burns nor subconjunctival injection any drugs.All rabbits were monitored daily after alkali burning.Digital pictures were taken on day 4th, 7th, 14th and 21th after modeling.Image analysis was performed on the area of neovascularization, a ascertain number of rabbits were killed respectively.Aqueous humor was collected for the concentration of vascular endothelial growth factor (VEGF) assay.Corneal specimens were analyzed by histopathologically and immunohistochemical staining of lymphatic endothelial cells hyaluronic acid receptor 1(LYVE-1). The use and care of the animals complied with the Statement of the Association for Research in Vision and Ophthalmology (ARVO).

ResultsOn the 4th day after alkali burning, the neovascularization buds grew into the edge of the cornea in the conbercept injected group, the ranibizumab injected group, the normal saline control group and the model control group, and the corneal edema decreased.On day 7th, the neovascularization of the conbercept injected group and the ranibizumab injected group was less than that of the normal saline control group and the model control group; On day 4th after alkali burning, corneal epithelial cells were increased, vacuoles were found in the epithelium, a large number of inflammatory cells were found in the matrix, and small vascular lumens were seen below the epithelium.On day 7th after modeling, neovascularization infiltrated the shallow matrix with a large number of inflammatory cells.Surface areas of induced CNV in conbercept injected group were (15.20±9.16)mm ², which were significantly less than those in ranibizumab injected group ([28.21±5.17]mm ²) on day 14th ( P<0.05). The concentration of VEGF in the conbercept injected group was (7.75±6.56)pg/ml, which was significantly lower than that in the ranibizumab injected group ([16.98±2.17]pg/ml on day 14th ( P<0.05). The normal control group had no lymphatic growth in the corneal tissue and no LYVE-1 positive particles.On day 4th after the alkil burning, corneal lymphatic vessels appeared in the corneal tissue of the conbercept injected group, the ranibizumab injected group, the normal saline control group and the model control group, which grew in parallel with the neovascularization.Lymphatic vessels counting in the conbercept injected group and ranibizumab injected group were 4.33±0.58 and 4.67±0.58 on day 7th and 14th, which were reduced significantly compared with normal saline control group (10.67±0.58) and the model control group (12.33±0.58) (all at P<0.05).

ConclusionsEarly subconjunctival administration of conbercept may successfully inhibit alkali-induced corneal neovascularization and corneal lymphangiogenesis in alkali burning animal model.The inhibit effect is related with the reduces of VEGF levels.